[814] Strehlow, II, p. 76.

[815] Strehlow, ibid.

[816] At bottom, the only real difference between Strehlow and Spencer and Gillen is the following one. For these latter, the soul of the individual, after death, returns to the nanja tree, where it is again confounded with the ancestor's soul (Nat. Tr., p. 513); for Strehlow, it goes to the isle of the dead, where it is finally annihilated. In neither myth does it survive individually. We are not going to seek the cause of this divergence. It is possible that there has been an error of observation on the part of Spencer and Gillen, who do not speak of the isle of the dead. It is also possible that the myth is not the same among the eastern Arunta, whom Spencer and Gillen observed particularly, as in the other parts of the tribe.

[817] Strehlow, II, p. 51.

[818] Ibid., II, p. 56.

[819] Ibid., I, pp. 3-4.

[820] Ibid., II, p. 61.

[821] See above, p. 183.

[822] Strehlow, II, p. 57; I, p. 2.

[823] Strehlow, II, p. 57.

[824] Roth, Superstition, Magic, etc., § 74.

[825] In other words, the totemic species is made up of the group of ancestors and the mythological species much more than of the regular animal or vegetable species.

[826] See above, p. 254.

[827] Strehlow, II, p. 76.

[828] Strehlow, ibid.

[829] Strehlow, II, pp. 57, 60, 61. Strehlow calls the list of totems the list of ratapa.

[830] Howitt, Nat. Tr., pp. 475 ff.

[831] The Manners and Customs of the Dieyerie Tribe of Australian Aborigines, in Curr, II, p. 47.

[832] Howitt, Nat. Tr., p. 482.

[833] Ibid., p. 487.

[834] Taplin, Folk-Lore, Customs, Manners, etc., of the South Australian Aborig., p. 88.

[835] The clan of each ancestor has its special camp underground; this camp is the miyur.

[836] Mathews, in Jour. of Roy. Soc. of N.S. Wales, XXXVIII, p. 293. He points out the same belief among other tribes of Victoria (ibid., p. 197).

[837] Mathews, ibid., p. 349.

[838] J. Bishop, Die Niol-Niol, in Anthropos, III, p. 35.

[839] Roth, Superstition, etc., § 68; cf. § 69a, gives a similar case from among the natives on the Proserpine River. To simplify the description, we have left aside the complications due to differences of sex. The souls of daughters are made out of the choi of their mother, though these share with their brothers the ngai of their father. This peculiarity, coming perhaps from two systems of filiation which have been in use successively, has nothing to do with the principle of the perpetuity of the soul.

[840] Ibid., p. 16.

[841] Die Tlinkit-Indianer, p. 282.

[842] Swanton, Contributions to the Ethnology of the Haida, pp. 117 ff.

[843] Boas, Sixth Rep. of the Comm. on the N.W. Tribes of Canada, p. 59.

[844] Lafitau, Mœurs des sauvages Amériquains, II, p. 434; Petitot, Monographie des Dénè-Dindjié, p. 59.

[845] See above, pp. 134 ff.

[846] See above, p. 137.

[847] Howitt, Nat. Tr., p. 147; cf. ibid., p. 769.

[848] Strehlow (I, p. 15, n. 2) and Schulze (loc. cit., p. 246) speak of the soul, as Howitt here speaks of the totem, as leaving the body to go to eat another soul. Likewise, as we have seen above, the altjira or maternal totem shows itself in dreams, just as a soul or spirit does.

[849] Fison and Howitt, Kurnai and Kamilaroi, p. 280.

[850] Globus, Vol. CXI, p. 289. In spite of the objections of Leonhardi, Strehlow maintains his affirmations on this point (see Strehlow, III, p. xi). Leonhardi finds a contradiction between this assertion and the theory according to which the ratapa emanate from trees, rocks or churinga. But the totemic animal incarnates the totem just as much as the nanja-tree or rock does, so they may fulfil the same function. The two things are mythological equivalents.

[851] Notes on the West Coastal Tribes of the Northern Territory of S. Australia, in Trans. of the Roy. Soc. of S. Aust., XXXI (1907), p. 4. Cf. Man, 1909, No. 86.

[852] Among the Wakelbura, where, according to Curr and Howitt, each matrimonial class has its own totems, the animal shows the class (see Curr, III, p. 28); among the Buandik, it reveals the clan (Mrs. James S. Smith, The Buandik Tribes of S. Australian Aborigines, p, 128). Cf. Howitt, On Some Australian Beliefs, in J.A.I., XIII, p. 191; XIV, p. 362; Thomas, An American View of Totemism, in Man, 1902, No. 85; Mathews, Journ. of the Roy. Soc. of N.S. Wales, XXXVIII, pp. 347-348; Brough Smyth, I, p. 110; Nor. Tr., p. 513.

[853] Roth, Superstition, etc., § 83. This is probably a form of sexual totemism.

[854] Prinz zu Wied, Reise in das innere Nord-Amerika, II, p. 190.

[855] K. von den Steinen, Unter den Naturvölkern Zentral-Bräsiliens, 1894, pp. 511, 512.

[856] See Frazer, Golden Bough², I, pp. 250, 253, 256, 257, 258.

[857] Third Rep., pp. 229, 233.

[858] Indian Tribes, IV, p. 86.

[859] For example, among the Batta of Sumatra (see Golden Bough², III, p. 420), in Melanesia (Codrington, The Melanesians, p. 178), in the Malay Archipelago (Tylor, Remarks on Totemism, in J.A.I., New Series, I, p. 147). It is to be remarked that the cases where the soul clearly presents itself after death in an animal form all come from the societies where totemism is more or less perverted. This is because the idea of the soul is necessarily ambiguous wherever the totemic beliefs are relatively pure, for totemism implies that it participate in the two kingdoms at the same time. So it cannot become either one or the other exclusively, but takes one aspect or the other, according to the circumstances. As totemism develops, this ambiguity becomes less necessary, while at the same time, spirits more actively demand attention. Then the marked affinities of the soul for the animal kingdom are manifested, especially after it is freed from the human body.

[860] See above, p. 170. On the generality of the doctrine of metempsychosis, see Tylor, II, pp. 8 ff.

[861] Even if we believe that religious and moral representations constitute the essential elements of the idea of the soul, still we do not mean to say that they are the only ones. Around this central nucleus are grouped other states of consciousness having this same character, though to a slighter degree. This is the case with all the superior forms of the intellectual life, owing to the special price and dignity attributed to them by society. When we devote our lives to science or art, we feel that we are moving in a circle of things that are above bodily sensations, as we shall have occasion to show more precisely in our conclusion. This is why the highest functions of the intelligence have always been considered specific manifestations of the soul. But they would probably not have been enough to establish the idea of it.

[862] F. Tregear, The Maori-Polynesian Comparative Dictionary, pp. 203-205.

[863] This is the thesis of Preuss in his articles in the Globus which we have cited several times. It seems that M. Lévy-Bruhl also tends towards this conception (see his Fonctions mentales, etc., pp. 92-93).

[864] On this point, see our Suicide, pp. 233 ff.

[865] It may be objected perhaps that unity is the characteristic of the personality, while the soul has always been conceived as multiple, and as capable of dividing and subdividing itself almost to infinity. But we know to-day that the unity of the person is also made up of parts and that it, too, is capable of dividing and decomposing. Yet the notion of personality does not vanish because of the fact that we no longer think of it as a metaphysical and indivisible atom. It is the same with the popular conceptions of personality which find their expression in the idea of the soul. These show that men have always felt that the human personality does not have that absolute unity attributed to it by certain metaphysicians.

[866] For all this, we do not deny the importance of the individual factor: this is explained from our point of view just as easily as its contrary. If the essential element of the personality is the social part of us, on the other hand there can be no social life unless distinct individuals are associated, and this is richer the more numerous and different from each other they are. So the individual factor is a condition of the impersonal factor. And the contrary is no less true, for society itself is an important source of individual differences (see our Division du travail social, 3rd. ed., pp. 267 ff.).

[867] Roth, Superstition, Magic, etc., §§ 65, 68; Spencer and Gillen, Nat. Tr., pp. 514, 516.

[868] Spencer and Gillen, Nat. Tr., pp. 521, 515; Dawson, Austral. Aborig., p. 58; Roth, op. cit., § 67.

[869] Spencer and Gillen, Nat. Tr., p. 517.

[870] Strehlow, II, p. 76 and n. 1; Spencer and Gillen, Nat. Tr., pp. 514, 516.

[871] Spencer and Gillen, Nat. Tr., p. 513.

[872] On this question, see Negrioli, Dei Genii presso i Romani; the articles Daimon and Genius in the Dict. of Antiq.; Preller, Romische Mythologie, II, pp. 195 ff.

[873] Negrioli, ibid., p. 4.

[874] Ibid., p. 8.

[875] Ibid., p. 7.

[876] Ibid., p. 11. Cf. Samter, Der Ursprung der Larencultus, in Archiv f. Religions-wissenschaft, 1907, pp. 368-393.

[877] Schulze, loc. cit., p. 237.

[878] Strehlow, I, p. 5. Cf. Spencer and Gillen, Nat. Tr., p. 133; Gason, in Curr, II, p. 69.

[879] See the case of a Mura-mura who is considered the spirit of certain hot springs, in Howitt, Nat. Tr., p. 482.

[880] Nor. Tr., pp. 313 f.; Mathews, Journ. of the Roy. Soc. of N.S. Wales, XXXVIII, p. 351. Among the Dieri there is also a Mura-mura whose function is to produce rain (Howitt, Nat. Tr., pp. 798 f.).

[881] Roth. Superstition, etc., § 67. Cf. Dawson, p. 59.

[882] Strehlow, I, pp. 2 ff.

[883] See above, p. 249.

[884] Nor. Tr., ch. vii.

[885] Spencer and Gillen, Nor. Tr., p. 277.

[886] Strehlow, I, p. 5.

[887] It is true that some nanja-trees and rocks are not situated around the ertnatulunga; they are scattered over different parts of the tribal territory. It is said that these are places where an isolated ancestor disappeared into the ground, lost a member, let some blood flow, or lost a churinga which was transformed into a tree or rock. But these totemic sites have only a secondary importance; Strehlow calls them kleinere Totemplätze (I, pp. 4-5). So it may be that they have taken this character only by analogy with the principal totemic centres. The trees and rocks which, for some reason or other, remind one of those found in the neighbourhood of an ertnatulunga, inspire analogous sentiments, so the myth which was formed in regard to the latter was extended to the former.

[888] Nat. Tr., p. 139.

[889] Parker, The Euahlayi, p. 21. The tree serving for this use is generally one of those figuring among the sub-totems of the individual. As a reason for this choice, they say that as it is of the same family as the individual, it should be better disposed to giving him aid (ibid., p. 29).

[890] Ibid., p. 36.

[891] Strehlow, II, p. 81.

[892] Parker, op. cit., p. 21.

[893] Codrington, The Melanesians, pp. 249-253.

[894] Turner, Samoa, p. 17.

[895] These are the very words used by Codrington (p. 251).

[896] This close connection between the soul, the guardian genius and the moral conscience of the individual is especially apparent among certain peoples of Indonesia. "One of the seven souls of the Tobabatak is buried with the placenta; though preferring to live in this place, it may leave it to warn the individual or to manifest its approbation when he does well. So in one sense, it plays the rôle of a moral conscience. However, its communications are not confined to the domain of moral facts. It is called the younger brother of the soul, as the placenta is called the younger brother of the child.... In war, it inspires the man with courage to march against the enemy" (Warneck, Der bataksche Ahnen und Geistercult, in Allg. Missionszeitschrift, Berlin, 1904. p. 10. Cf. Kruijt, Het Animisme in den indischen Archipel, p. 25).

[897] It still remains to be investigated how it comes that after a certain moment in evolution, this duplication of the soul was made in the form of an individual totem rather than of a protecting ancestor. Perhaps this question has an ethnological rather than a sociological interest. However, the manner in which this substitution was probably effected may be represented as follows.

The individual totem commenced by playing a merely complimentary rôle. Those individuals who wished to acquire powers superior to those possessed by everybody, did not and could not content themselves with the mere protection of the ancestor; so they began to look for another assistant of the same sort. Thus it comes about that among the Euahlayi, the magicians are the only ones who have or who can procure individual totems. As each one has a collective totem in addition, he finds himself having many souls. But there is nothing surprising in this plurality of souls: it is the condition of a superior power.

But when collective totemism once begins to lose ground, and when the conception of the protecting ancestor consequently begins to grow dim in the mind, another method must be found for representing the double nature of the soul, which is still felt. The resulting idea was that, outside of the individual soul, there was another, charged with watching over the first one. Since this protecting power was no longer demonstrated by the very fact of birth, men found it natural to employ, for its discovery, means analogous to those used by magicians to enter into communion with the forces of whose aid they thus assured themselves.

[898] For example, see Strehlow, II, p. 82.

[899] Wyatt, Adelaide and Encounter Bay Tribes, in Woods, p. 168.

[900] Taplin, The Narrinyeri, pp. 62 f.; Roth, Superstition, etc., § 116; Howitt, Nat. Tr., pp. 356, 358; Strehlow, pp. 11-12.

[901] Strehlow, I, pp. 13-14; Dawson, p. 49.

[902] Strehlow, I, pp. 11-14; Eylmann, pp. 182, 185; Spencer and Gillen, Nor. Tr., p. 211; Schürmann, The Aborig. Tr. of Port Lincoln, in Woods, p. 239.

[903] Eylmann, p. 182.

[904] Mathews, Journ. of the Roy. Soc. of N.S. Wales, XXXVIII, p. 345; Fison and Howitt, Kamilaroi and Kurnai, p. 467; Strehlow, I, p. 11.

[905] Nat. Tr., pp. 390-391. Strehlow calls these evil spirits Erintja; but this word is evidently equivalent to Oruncha. Yet there is a difference in the ways the two are presented to us. According to Spencer and Gillen, the Oruncha are malicious rather than evil; they even say (p. 328) that the Arunta know no necessarily evil spirits. On the contrary, the regular business of Strehlow's Erintja is to do evil. Judging from certain myths given by Spencer and Gillen (Nat. Tr., p. 390), they seem to have touched up the figures of the Oruncha a little: these were originally ogres (ibid., p. 331).

[906] Roth, Superstition, etc., § 115; Eylmann, p. 190.

[907] Nat. Tr., pp. 390 f.

[908] Ibid., p. 551.

[909] Ibid., pp. 326 f.

[910] Strehlow, I, p. 14. When there are twins, the first one is believed to have been conceived in this manner.

[911] Spencer and Gillen, Nat. Tr., p. 327.

[912] Howitt, Nat. Tr., pp. 358, 381, 385; Spencer and Gillen, Nat. Tr., p. 334; Nor. Tr., pp. 501, 530.

[913] As the magician can either cause or cure sickness, we sometimes find, besides these magical spirits whose function is to do evil, others who forestall or neutralize the evil influence of the former. Cases of this sort will be found in Nor. Tr., pp. 501-502. The fact that the latter are magic just as much as the former is well shown by the fact that the two have the same name, among the Arunta. So they are different aspects of a single magic power.

[914] Strehlow, I, p. 9. Putiaputia is not the only personage of this sort of whom the Arunta myths speak: certain portions of the tribe give a different name to the hero to whom the same invention is ascribed. We must not forget that the extent of the territory occupied by the Arunta prevents their mythology from being completely homogeneous.

[915] Spencer and Gillen, Nor. Tr., p. 493.

[916] Ibid., p. 498.

[917] Ibid., pp. 498 f.

[918] Howitt, Nat. Tr., p. 135.

[919] Ibid., pp. 476 ff.

[920] Strehlow, I, pp. 6-8. The work of Mangarkunjerkunja must be taken up again later among other heroes; for, according to a belief that is not confined to the Arunta, a time came when men forgot the teaching of their first initiators and became corrupt.

[921] This is the case, for example, of Atnatu (Spencer and Gillen, Nor. Tr., p. 153) and the Witurna (Nor. Tr., p. 498), If Tendun did not establish these rites, it is he who is charged with the direction of their celebration (Howitt, Nat. Tr., p. 670).

[922] Nor. Tr., p. 499.

[923] Howitt, Nat. Tr., p. 493; Kamilaroi and Kurnai, pp. 197 and 247; Spencer and Gillen, Nor. Tr., p. 492.

[924] For example, see Nor. Tr., p. 499.

[925] Nor. Tr., pp. 338, 347, 499.

[926] It is true that Spencer and Gillen maintain that these mythical beings play no moral rôle (Nor. Tr., p. 493); but this is because they give too narrow a meaning to the word. Religious duties are duties: so the fact of looking after the manner in which these are observed concerns morals, especially because all morals have a religious character at this period.

[927] The fact was observed as early as 1845 by Eyre, Journals, etc., II, p. 362, and, before Eyre, by Henderson, Observations on the Colonies of N.S. Wales and Van Diemen's Land, p. 147.

[928] Nat. Tr., pp. 488-508.

[929] Among the Kulin, Wotjobaluk and Woëworung (Victoria).

[930] Among the Yuin, Ngarrigo and Wolgal (New South Wales).

[931] Among the Kamilaroi and Euahlayi (northern part of New South Wales); and more to the centre, in the same province, among the Wonghibon and the Wiradjuri.

[932] Among the Wiimbaio and the tribes on the lower Murray (Ridley, Kamilaroi, p. 137; Brough Smyth, I, pp. 423, n., 431).

[933] Among the tribes on the Herbert River (Howitt, Nat. Tr., p. 498).

[934] Among the Kurnai.

[935] Taplin, p. 55; Eylmann, p. 182.

[936] It is undoubtedly to this supreme Mura-mura that Gason makes allusion in the passage already cited (Curr, II, p. 55).

[937] Nat. Tr., p. 246.

[938] Between Baiame, Bunjil and Daramulun on the one hand, and Altjira on the other, there is the difference that the latter is completely foreign to all that concerns humanity; he did not make man and does not concern himself with what they do. The Arunta have neither love nor fear for him. But when this conception is carefully observed and analysed, it is hard to admit that it is primitive; for if the Altjira plays no rôle, explains nothing, serves for nothing, what made the Arunta imagine him? Perhaps it is necessary to consider him as a sort of Baiame who has lost his former prestige, as an ancient god whose memory is fading away. Perhaps, also, Strehlow has badly interpreted the testimony he has gathered. According to Eylmann, who, it is to be admitted, is neither a very competent nor a very sure observer, Altjira made men (op. cit., p. 134). Moreover, among the Loritja, the corresponding personage, Tukura, is believed to celebrate the initiation ceremonies himself.

[939] For Bunjil, see Brough Smyth, I, p. 417; for Baiame, see Ridley, Kamilaroi, p. 136; for Daramulun, see Howitt, Nat. Tr., p. 495.

[940] On the composition of Bunjil's family, for example, see Howitt, Nat. Tr., pp. 128, 129, 489, 491; Brough Smyth, I, pp. 417, 423; for Baiame's, see L. Parker, The Euahlayi, pp. 7, 66, 103; Howitt, Nat. Tr., pp. 502, 585, 407; for Nurunderi's, Taplin, The Narrinyeri, pp. 57 f. Of course, there are all sorts of variations in the ways in which the families of these great gods are conceived. The personage who is a brother here, is a son there. The number and names of the wives vary with the locality.