In support of his theory Mr. Hartland makes reference to the belief of some savages, that charms may be made from dead bodies against the surviving relatives of the deceased,[65] and to certain rites of healing in which, besides the patient himself, “other members of his tribe, presumably kinsmen,” take part.[66] But the former belief is a superstition connected with the wonder of death, from which no conclusion must be drawn as to relations between the living; and in the ceremonies of healing the medicine-man plays a much more prominent part than the other bystanders—whose relationship to the patient, besides, is so little marked that Mr. Hartland only presumes them to be kindred. He further observes that in the wide-spread custom of the Couvade we meet with the idea that the child, being a part of the father, is liable to be affected by various acts committed by him.[67] And from Sir J. G. Frazer’s ‘Golden Bough’ might be quoted many instances of a belief in some mysterious bond of sympathy knitting together absent friends and relations—especially at critical times of life—which has, in particular, led to rules regulating the conduct of persons left at home while a party of their friends is out fishing or hunting or on the war path.[68] But all these rules are taboo restrictions of a definite and altogether special kind, generally, it seems, referring to members of the same family, and frequently to wives in their husbands’ absence. In order to make his hypothesis acceptable, Mr. Hartland ought to have produced a fair number of facts proving that the members of the same clan really are believed to be connected with each other in such a manner, that whatever affects one of them at the same time in a mysterious way affects the rest. But we look in vain for a single well-established instance of such a belief.
[65] Ibid. ii. 437 sq.
[66] Ibid. ii. 432 sqq.
[67] Hartland, Legend of Perseus, ii. 406.
[68] Frazer, Golden Bough, i. 27 sqq. See also Haddon, Magic and Fetishism, p. 11 sq.
It seems that the importance which savages attach to a common blood has been much exaggerated. Clanship is based on a method of counting descent by means of names, either through the father or through the mother, but not through both at once. This, however, by no means implies that the other line is not recognised as a line of blood-relationship. The paternal system of descent is not necessarily associated with the idea that the mother has no share in parentage, nor is the maternal system necessarily associated with unconsciousness of the child’s relation to its father;[69] even the Couvade, which assumes the recognition of a most intimate relationship between the child and its father, has been found to prevail among some peoples who regard the child as a member of the mother’s clan.[70] Nay, there are instances in which the clan-bond is obviously not regarded as a blood-bond at all, in the strict sense of the word. Of some tribes in New South Wales Mr. Cameron tells us that, although a daughter belongs not to her father’s clan but to that of her mother’s brother, they believe that she emanates from her father solely, being only nurtured by her mother;[71] and the Arunta of Central Australia, who have the paternal system of descent, maintain that a child really descends neither from its father nor from its mother, but is the reincarnation of a mythical totem-ancestor.[72] Their theory is “that the child is not the direct result of intercourse, that it may come without this, which merely, as it were, prepares the mother for the reception and birth also of an already-formed spirit child who inhabits one of the local totem centres”;[73] and its totem-name, which is derived from the spot where it is supposed to have been conceived,[74] is different from its clan-name. It is useful to scrutinise Mr. Hartland’s theory in the light of this class of facts. They evidently prove that clanship and what we are used to call the system of counting “descent,” is not necessarily based on the notion of actual blood-relationship, but on kinship as a fact combined with a name; whereas Mr. Hartland’s hypothesis presupposes, not that the members of a clan really are, but that they consider themselves to be all of one blood.
[69] Mr. Swan informs me that the Waguha of West Tanganyika, among whom children are generally named after their father, recognise the part taken by both parents in generation; and Archdeacon Hodgson writes the same concerning certain other tribes of Eastern Central Africa, who trace descent through the mother.
[70] Ling Roth, ‘Signification of Couvade,’ in Jour. Anthr. Inst. xxii. 227, 238.
[71] Cameron, ‘Notes on some Tribes of New South Wales,’ in Jour. Anthr. Inst. xiv. 352.
[72] Spencer and Gillen, Native Tribes of Central Australia, ch. iv. especially pp. 121, 124.