§ 3 The Conceptional Hypothesis
Sir J. G. Frazer, after very candidly relinquishing two early suggestions—by no means fanciful—concerning the origin of Totemism, as unsupported by sufficient evidence, has put forward a third—“conceptional Totemism”—which occurred to him upon the discovery by Messrs. Spencer and Gillen of the doctrine of totemic descent prevalent amongst the Arunta and allied tribes. These tribes are said not to be aware of the connexion between sexual intercourse and pregnancy. They say that a child is the result of the entry into a woman of the spirit of some pre-existing tribesman of this or that Totem; which, therefore, will also be the child’s. What Totem it is depends upon the place where the woman first becomes aware of quickening; for there are certain places known to the natives where the people of such or such a Totem “went into the ground” [perhaps were buried]; and in passing such a place any woman is liable to be impregnated by one of the discarnate spirits; who are always on the look-out for an opportunity to re-enter the mortal state. This is the real cause of pregnancy, for which marriage is merely a preparation.520 A young woman who does not desire the dignity of a matron, when passing such haunted ground, runs crouching by, and cries out that she is an old woman; for facility in being deceived is a saving grace of spirits. In modified forms this doctrine of “no paternity and re-incarnation” is professed by tribes throughout the centre and north of the continent, in Queensland and in parts of West Australia.521
The Arunta doctrine of conception by animal or plant spirits cannot, however, be the origin of Totemism; because, as Sir James Frazer points out, the impregnating spirits are already totemic. One must, therefore, suppose (he says) an earlier state of things, in which a woman, ignorant of the true causes of childbirth, imagined at the first symptoms of pregnancy (by which the quickening seems to be meant) that she had been entered by some object, or by the spirit of some object, which had been engaging her attention at the time, or which she may have been eating—a wallaby, emu, plum or grass-seed—and later believed that the child she bare must be, or be an incarnation of, that object or spirit, and in fact nothing else than a wallaby, emu, plum or grass-seed with the appearance of a human being. If other women had similar experiences in connexion with other animals or plants, and if the descendants of their children remembered the stories, and considered themselves to be wallabies, grass-seed and so forth, the hypothesis would fully explain that identification of groups of men with groups of things which is characteristic of Totemism; and the other characters naturally follow. Such is the conceptional theory of Totemism, deriving that institution from “the sick fancies of pregnant women.”522
Circumstances, which Sir James Frazer regards as very similar to those which he imagined as having prevailed at some former time amongst the Arunta, have been discovered by Dr. Rivers at Mota and Motlar in the Banks Islands. There many people are “by the custom of the island” not permitted to eat certain animals or fruits or even to touch certain trees; because they are believed to be those animals or plants—their mothers having suffered some influence from such animals or plants at conception or at some subsequent period of pregnancy. “The course of events is usually as follows: a woman, sitting down in her garden, or in the bush, or on the shore, finds an animal or fruit in her loin-cloth. She takes it up and carries it to the village, where she asks the meaning of the appearance. The people say that she will give birth to a child who will have the characters of this animal, or even (it appeared) would be himself or herself that animal.”523 She takes it to its proper home, tries to keep it, and feeds it; but after a time it will disappear, and is then believed to have entered into her. There is no belief in physical impregnation by the animal, nor of its invading the woman as a physical object; such an animal seems to be considered as “more or less supernatural, a spirit-animal, from the beginning.” “The belief is not accompanied by any ignorance of the physical rôle of the human father.” Apparently the prohibition against eating animals or plants thus connected with oneself rests on the idea that it would amount to eating oneself—a sort of cannibalism. One partakes of its physical and mental characters. But the resemblance to, and the taboo on eating, a certain animal are individual matters; there is no belief in their being passed on to one’s descendants. In this alone the belief at Mota falls short of Totemism. “Yet it occurs in a people whose social system has no totemic features at the present time, whatever it may have had in the past.”524 Possibly former Totem-clans have been merged in secret societies, but there is no clear evidence of it. In Melanesia, Totemism occurs in Fiji, Shortland Islands, Bismarck Archipelago (probably), Reef Islands, Santa Cruz, Vanikolo and in some regions of the Solomon Islands, but not in the Banks Islands nor in the Torres Islands to the south.
This hypothesis that Totemism is derived from the fancies of women concerning the causes of their own pregnancy suggests several adverse considerations. In the first place, it is to me incredible that the Arunta are really ignorant of paternity; young women and children perhaps may be, but not the seniors. My reasons for thinking so are set out at length in the J.R.A.I., 1819 (p. 146), namely, that the facts of childbirth are too interesting to be overlooked or misconstrued, and are well within the grasp of savage understanding; that those who think the Arunta capable of such stupidity have attended too exclusively to particular incidents of pregnancy, especially the quickening, and have not considered that there is a definite series of closely connected incidents; that there is testimony, not to be lightly put aside, that the old men know the truth in the case of human beings, and that knowledge of the parallel phenomena amongst animals is shared even by the Arunta children; that it is possible for knowledge to be repressed by dogma without being extinguished; that this is an old man’s dogma, and that we must not assume that in Australia, any more than in Europe, what people are accustomed to say is good evidence of what they really believe;525 that other tribes at the same level of culture, in South-East Australia, are so convinced of the importance of paternity that they say the child proceeds entirely from the father; that those who deny paternity show by their customs and myths that they are secretly aware of it; and, finally, that in some cases they have clear motives for maintaining the dogma and suppressing the truth.
Secondly, the observations of Dr. Rivers in the Banks Islands give not the slightest support to the conceptional hypothesis. For (a) the Banks Islands belief “is not accompanied by any ignorance of the physical rôle of the human father.” (b) The belief is that the child born to a woman who has been visited by a plant or animal is in some way “influenced” by the experience; not, as with the Arunta, that it is entirely due to it (except for some “preparation” by the father). (c) Before the Banks Islands belief can give rise to Totemism there must be a belief in the continuous inheritance of the plant or animal influence; and not only is this absent, but there is a strong tendency to prevent it. A woman having been influenced by an eel, her child is an eel; but if that child be a girl, there is no further transmission of eel-like qualities; the girl, on becoming pregnant, may be influenced by a yam, and then her child will be a yam. This hinders the formation of a group of human beings mysteriously allied to an animal or plant, such as Totemism implies. But (d) the conceptional hypothesis refers the origin of Totemism to the ignorant imaginings of women themselves, who think they have been entered by this or that; whereas in the Banks Islands a woman’s belief in the influence of an animal or plant upon her offspring is not prompted by her own fancy, but by a popular superstition. For when she carries the thing found in her loin-cloth to the village, “the people say” she will give birth to a child who will have the qualities of, or even will be that thing; and a child afterwards born may not eat or touch that thing “by the custom of the country.” Had the Arunta, then, of pre-totemic days such a superstition and custom for the guidance of mothers? If so, the history of that superstition and custom must be more obscure than Totemism itself. But in the Banks Islands their origin may admit of a wide solution; for although it is said that no manifest traces of Totemism are to be found there now; yet, seeing that the Australians, Papuans, Polynesians and the Melanesians to the north all have Totemism or plain vestiges of it, the improbability that Melanesians in the Banks Islands never had it is very great; and it is reasonable to suppose that the superstition concerning the influence of animals or plants upon pregnant women, and the taboo upon the eating of such an animal or plant by the offspring, represent not a possible origin of Totemism, but a survival of it.
These considerations raise a further question as to the employment of the Comparative Method. Natives of the Banks Islands are far in advance of the Arunta in all the arts of life, as well as in relation to Totemism which they seem to have outlived. Is it admissible to take some trait of their life as an example of what may have existed amongst the Arunta at a stage of culture which is generally assumed to have been inferior even to the present? When the antecedents of an institution have been lost amongst any given people, and we set out to supply them from parallel cases amongst other peoples, must we not require the same relative order of development? If a social phenomenon is found in the Banks Islands, its possibility is proved; but strictly it was possible only where it occurred: elsewhere we can expect only something similar, according to the similarity of other relevant circumstances. Comparing the Banks Islands and their people with Central Australia and its people, one discovers hardly anything common to them. And Arunta tradition (whatever worth) shows no sign of anything like the southern Melanesian culture: the Alcheringa ancestors are represented as having been already Totemists. To assume, however, that the former condition of the Arunta was even lower than the present, may not be justifiable: social degeneration is not uncommon. There would be something to guide our judgment if we knew how recently the desiccation of their country approached its present severity. But, in any case, can we suppose it to have been at all like the Melanesian?