[1171] Nor. Tr., pp. 294-296. It is curious that, on the contrary, the Anula regard the rainbow as productive of rain (ibid., p. 314).

[1172] The same process is employed among the Arunta (Strehlow, III, p. 132). Of course we may ask if this effusion of blood is not an oblation designed to win the powers which produce rain. However, Gason says distinctly that this is a way of imitating the water which falls.

[1173] Gason, The Dieri Tribe, in Curr, II, pp. 66-68. Howitt (Nat. Tr., pp. 798-800) mentions other rites of the Dieri for obtaining rain.

[1174] Ethnological Notes on the Western Australian Aborigines, in Internationales Archiv. f. Ethnographie, XVI, pp. 6-7. Cf. Withnal, Marriage Rites and Relationship in Man, 1903, p. 42.

[1175] We presume that sub-totems may have tarlow, for, according to Clement, certain clans have several totems.

[1176] Clement says a tribal family.

[1177] We shall explain below (p. 362) why this is incorrect.

[1178] On this classification, see Frazer, Lectures on the Early History of Kingship, pp. 37 ff.; Hubert and Mauss, Théorie générale de la Magie, pp. 61 ff.

[1179] We say nothing of what has been called the law of opposition, for, as MM. Hubert and Mauss have shown, a contrary produces its opposite only through the intermediacy of a similar (Théorie générale de la Magie, p. 70).

[1180] Lectures on the History of Kingship, p. 39.

[1181] It is applicable in the sense that there is really an association of the statue and the person encharmed. But it is true that this association is the simple product of an association of ideas by similarity. The true determining cause of the phenomenon is the contagiousness peculiar to religious forces, as we have shown.

[1182] For the causes determining this outward manifestation, see above, pp. 230 ff.

[1183] M. Lévy-Bruhl, Les Fonctions mentales dans les sociétés inférieures, pp. 61-68.

[1184] Golden Bough², I, pp. 69-75.

[1185] We do not wish to say that there was ever a time when religion existed without magic. Probably as religion took form, certain of its principles were extended to non-religious relations, and it was thus supplemented by a more or less developed magic. But if these two systems of ideas and practices do not correspond to distinct historical phases, they have a relation of definite derivation between them. This is all we have sought to establish.

[1186] Loc. cit., pp. 108 ff.

[1187] See above, pp. 203 f.

[1188] Of course animal societies do exist. However, the word does not have exactly the same sense when applied to men and to animals. The institution is a characteristic fact of human societies; but animals have no institutions.

[1189] The conception of cause is not the same for a scholar and for a man with no scientific culture. Also, many of our contemporaries understand the principle of causality differently, as they apply it to social facts and to physico-chemical facts. In the social order, men frequently exhibit a conception of causality singularly like that which was at the basis of magic for a long time. One might even ask if a physicist and a biologist represent the causal relation in the same fashion.

[1190] Of course these ceremonies are not followed by an alimentary communion. According to Strehlow, they have another name, at least when they concern non-edible plants: they are called, not mbatjalkatiuma, but knujilelama (Strehlow, III, p. 96).

[1191] Strehlow, III, p. 8.

[1192] The Warramunga are not the only ones among whom the Intichiuma takes the form of a dramatic representation. It is also found among the Tjingilli, the Umbaia, the Wulmala, the Walpari and even the Kaitish, though in certain of its features the ritual of these latter resembles that of the Arunta (Nor. Tr., p. 291, 309, 311, 317). If we take the Warramunga as a type, it is because they have been studied the best by Spencer and Gillen.

[1193] This is the case with the Intichiuma of the black cockatoo (see above, p. 353).

[1194] Nor. Tr., pp. 300 ff.

[1195] One of these two actors does not belong to the Black Snake clan, but to that of the Crow. This is because the Crow is supposed to be an "associate" of the Black Snake: in other words, it is a sub-totem.

[1196] Nor. Tr., p. 302.

[1197] Ibid., p. 305.

[1198] See Spencer and Gillen, Nat. Tr., p. 188; Strehlow, III, p. 5.

[1199] Strehlow himself recognizes this: "The totemic ancestor and his descendant, who represents him (der Darsteller) are presented as one in these sacred hymns." (III, p. 6). As this incontestable fact contradicts the theory according to which ancestral souls do not reincarnate themselves, Strehlow adds, it is true, in a note, that "in the course of the ceremony there is no real incarnation of the ancestor in the person who represents him." If Strehlow wishes to say that the incarnation does not take place on the occasion of the ceremony, then nothing is more certain. But if he means that there is no incarnation at all, we do not understand how the officiant and the ancestor can be confounded.

[1200] Perhaps this difference is partially due to the fact that among the Warramunga each clan is thought to be descended from one single ancestor about whom the legendary history of the clan centres. This is the ancestor whom the rite commemorates; now the officiant need not be descended from him. One might even ask if these mythical chiefs, who are sorts of demigods, are submitted to reincarnation.

[1201] In this Intichiuma, three assistants represent ancestors "of a considerable antiquity"; they play a real part (Nat. Tr., pp. 181-182). It is true that Spencer and Gillen add that these are ancestors posterior to the Alcheringa. Nevertheless, mythical personages are represented in the course of the rite.

[1202] Sacred rocks and water-holes are not mentioned. The centre of the ceremony is the image of an emu drawn on the ground, which can be made anywhere.

[1203] We do not mean to say that all the ceremonies of the Warramunga are of this type. The example of the white cockatoo, of which we spoke above, proves that there are exceptions.

[1204] Nor. Tr., pp. 226 ff. On this same subject, cf. certain passages of Eylmann which evidently refer to the same mythical being (Die Eingeborenen, etc., p. 185). Strehlow also mentions a mythical snake among the Arunta (Kulaia, water-snake) which may not differ greatly from the Wollunqua (Strehlow, I, p. 78; cf. II, p. 71, where the Kulaia is found in a list of totems).

[1205] We use the Arunta words, in order not to complicate our terminology; the Warramunga call this mythical period Wingara.

[1206] "It is not easy to express in words what is in reality rather a vague feeling amongst the natives, but after carefully watching the different series of ceremonies, we were impressed with the feeling that the Wollunqua represented to the native mind the idea of a dominant totem" (Nor. Tr., p. 248).

[1207] One of the most solemn of these ceremonies is the one which we have had occasion to describe above (p. 217), in the course of which an image of the Wollunqua is designed on a sort of hillock which is then torn to pieces in the midst of a general effervescence.

[1208] Nor. Tr., pp. 227, 248.

[1209] Here are the terms of Spencer and Gillen in the only passage in which they speak of a possible connection between the Wollunqua and rain. A few days after the rite about the hillock, "the old men say that they have heard Wollunqua speak, that he was satisfied with what had passed and that he was going to send rain. The reason for this prophecy was that they, as well as ourselves, had heard thunder rolling at a distance." To such a slight extent is the production of rain the immediate object of the ceremony that they did not attribute it to Wollunqua until several days later, and then after accidental circumstances. Another fact shows how vague the ideas of the natives are on this point. A few lines below, thunder is spoken of as a sign, not of the Wollunqua's satisfaction, but of its discontent. In spite of these prognostics, continue our authors, "the rain did not fall. But some days later, they heard the thunder rolling in the distance again. The old men said that the Wollunqua was grumbling because he was not contented" with the way in which the rite had been celebrated. Thus a single phenomenon, the noise of thunder, is sometimes interpreted as a sign of a favouring disposition, and sometimes as a mark of evil intentions.

However, there is one detail of the ritual which, if we accept the explanation of it proposed by Spencer and Gillen, is directly efficient. According to them, the destruction of the hillock was intended to frighten the Wollunqua and to prevent it, by magic constraint, from leaving its retreat. But this interpretation seems very doubtful to us. In fact, in the very case of which we were speaking, where it was announced that the Wollunqua was dissatisfied, this dissatisfaction was attributed to the fact that they had neglected to take away the debris of the hillock. So this removal is demanded by the Wollunqua itself, and in no way intended to intimidate it and exercise a coercive influence over it. This is probably merely one case of a more general rule which is in force among the Warramunga: the instruments of the cult must be destroyed after each ceremony. Thus the ritual ornamentations with which the officiants are decorated are violently torn off from them when the rite is terminated (Nor. Tr., p. 205).

[1210] Nor. Tr., pp. 207-208.

[1211] Ibid., p. 210.

[1212] See, in the list of totems drawn up by Strehlow, Nos. 432-442 (II, p. 72).

[1213] See Strehlow, III, p, 8. Among the Arunta there is also a totem Worra which greatly resembles the "laughing boy" totem of Warramunga (ibid., and III, p. 124). Worra means young men. The object of the ceremony is to make the young men take more pleasure in the game labara (for this game, see Strehlow, I, p. 55, n. 1).

[1214] See above, p. 373.

[1215] A case of this sort will be found in Nor. Tr., p. 204.

[1216] Nat. Tr., p. 118 and n. 2, pp. 618 ff.; Nor. Tr., pp. 716 ff. There are some sacred ceremonies from which women are not wholly excluded (see, for example, Nor. Tr., pp. 375 ff.); but this is exceptional.

[1217] See Nat. Tr., pp. 329 ff.; Nor. Tr., pp. 210 ff.

[1218] This is the case, for example, with the corrobbori of the Molonga among the Pitta-Pitta of Queensland and the neighbouring tribes (see Roth, Ethnog. Studies among the N.W. Central Queensland Aborigines, pp. 120 ff.).—References for the ordinary corrobbori will be found in Stirling, Rep. of the Horn Expedition to Central Australia, Part IV, p. 72, and in Roth, op. cit., pp. 117 ff.

[1219] On this question see the excellent work of Culin, Games of the North American Indians (XXIVth Rep. of the Bureau of Am. Ethnol.).

[1220] See above, p. 81.

[1221] Especially in sexual matters. In the ordinary corrobbori, sexual licence is frequent (see Spencer and Gillen, Nat. Tr., pp. 96-97, and Nor. Tr., pp. 136-137). On sexual licence in popular feasts in general, see Hagelstrange, Süddeutsches Bauernleben im Mittelalter, pp. 221 ff.

[1222] Thus the exogamic rules must be violated in the course of certain religious ceremonies (see above, p. 216, n. 1). A precise ritual meaning probably could not be found for these excesses. It is merely a mechanical consequence of the state of super-excitation provoked by the ceremony. It is an example of rites having no definite object themselves, but which are mere discharges of energy (see above, p. 381). The native does not assign them a definite end either; he merely says that if these licences are not committed, the rite will not produce its effects; the ceremony will fail.

[1223] Here are the very words used by Spencer and Gillen: "They (the ceremonies connected with the totems) are often, though by no means always, associated with the performance of the ceremonies attendant upon initiation of young men, or are connected with the Intichiuma" (Nor. Tr., p. 178).

[1224] We leave aside the question of what this character consists in. It is a problem which would lead us into a very long and technical development and which must therefore be treated by itself. Moreover, it does not concern the propositions established in this present work.

[1225] This is chapter vi, entitled Ceremonies Connected with the Totems.

[1226] Strehlow, III, pp. 1-2.

[1227] This explains the error of which Strehlow accuses Spencer and Gillen: that they applied to one form of the ceremony the term which is more appropriate for the other. But in these conditions, the error hardly seems to have the gravity attributed to it by Strehlow.

[1228] It cannot be otherwise. In fact, as the initiation is a tribal feast, novices of different totems are initiated at the same time. So the ceremonies which thus succeed one another in the same place have to do with several totems, and, therefore, they must take place away from the places with which they are connected by the myth.

[1229] It will now be understood why we have never studied the initiation rites by themselves: it is because they are not a ritual entity, but are formed by the conglomeration of rites of different sorts. There are interdictions, ascetic rites and representative ceremonies which cannot be distinguished from those celebrated at the time of the Intichiuma. So we had to dismember this composite system and treat each of the different rites composing it separately, classifying them with the similar rites to which they are to be related. We have also seen (pp. 285 ff.) that the initiation has served as the point of departure for a new religion which tends to surpass totemism. But it has been sufficient for us to show that totemism contained the germs of this religion; we have had no need of following out its development. The object of this book is to study the elementary beliefs and practices; so we must stop at the moment when they give birth to more complex forms.

[1230] Nat. Tr., p. 463. If the individual may choose between the ceremonies of his paternal and maternal totems, it is because, owing to reasons which we have set forth above (p. 183), he participates in both.

[1231] See below, ch. v, p. 395.

[1232] See Essai sur le Sacrifice, in Mélanges d'histoire des Religions, p. 83.

[1233] Piacularia auspicia appellabant quæ sacrificantibus tristia portendebant (Paul ex Fest., p. 244, ed. Müller). The word piaculum is even used as a synonym of misfortune. "Vetonica herba," says Pliny, "tantum gloriæ habet ut domus in qua sita sit tuta existimetur a piaculis omnibus" (XXV, 8, 46).

[1234] Nor. Tr., p. 526; Eylmann, p. 239. Cf. above, p. 305.

[1235] Brough Smyth, I, p. 106; Dawson, p. 64; Eylmann, p. 239.

[1236] Dawson, p. 66; Eylmann, p. 241.

[1237] Nat. Tr., p. 502; Dawson, p. 67.

[1238] Nor. Tr., pp. 516-517.

[1239] Ibid., pp. 520-521. The authors do not say whether these were tribal or blood relatives. The former hypothesis is the more probable one.

[1240] Nor. Tr., pp. 525 f. This interdiction against speaking, which is peculiar to women, though it consists in a simple abstention, has all the appearance of a piacular rite: it is a way of incommoding one's self. Therefore we mention it here. Also, fasting may be a piacular rite or an ascetic one, according to the circumstances. Everything depends upon the conditions in which it takes place and the end pursued (for the difference between these two sorts of rites, see below, p. 396).

[1241] A very expressive illustration showing this rite will be found in Nor. Tr., p. 525.

[1242] Ibid., p. 522.

[1243] For the principal forms of funeral rites, see Howitt, Nat. Tr., pp. 446-508, for the tribes of the South-East; Spencer and Gillen, Nor. Tr., p. 505, and Nat. Tr., pp. 497 ff., for those of the centre; Roth, Nor. Queensland Ethnog., Bull. 9, in Records of the Australian Museum, VI, No. 5, pp. 365 ff. (Burial Customs and Disposal of the Dead).

[1244] See, for example, Roth, loc. cit., p. 368; Eyre, Journals of Exped. into Central Aust., II, pp. 344 f.

[1245] Spencer and Gillen, Nat. Tr., p. 500; Nor. Tr., pp. 507, 508; Eylmann, p. 241; Parker, Euahlayi, pp. 83 ff.; Brough Smyth, I, p. 118.

[1246] Dawson, p. 66; Howitt, Nat. Tr., p. 466; Eylmann, pp. 239-240.

[1247] Brough Smyth, I, p. 113.

[1248] W. E. Stanbridge, Trans. Ethnological Society of London, N.S., Vol. I, p. 286.

[1249] Brough Smyth, I, p. 104.

[1250] Howitt, Nat. Tr., p. 459. Similar scenes will be found in Eyre, op. cit., II, p. 255, n., and p. 347; Roth, loc. cit., pp. 394, 395, for example; Grey, II, pp. 320 ff.

[1251] Brough Smyth, I, pp. 104, 112; Roth, loc. cit., p. 382.

[1252] Nor. Tr., pp. 511-512.

[1253] Dawson, p. 67; Roth, loc. cit., pp. 366-367.

[1254] Nat. Tr., pp. 508-510.

[1255] A little wooden vessel, of which we spoke above, p. 334.

[1256] Nat. Tr., pp. 508-510. The other final rite at which Spencer and Gillen assisted is described on pp. 503-508 of the same work. It does not differ essentially from the one we have analysed.

[1257] Nor. Tr., pp. 531-540.

[1258] Contrarily to what Jevons says, Introduction to the History of Religion, pp. 46 ff.

[1259] This makes Dawson say that the mourning is sincere (p. 66). But Eylmann assures us that he never knew a single case where there was a wound from sorrow really felt (op. cit., p. 113).

[1260] Nat. Tr., p. 510.

[1261] Eylmann, pp. 238-239.

[1262] Nor. Tr., p. 507; Nat. Tr., p. 498.

[1263] Nat. Tr., p. 500; Eylmann, p. 227.

[1264] Brough Smyth, I, p. 114.

[1265] Nat. Tr., p. 510.

[1266] Several examples of this belief are to be found in Howitt, Nat. Tr., p. 435. Cf. Strehlow, I, 15-16; II, p. 7.

[1267] It may be asked why repeated ceremonies are necessary to produce the relief which follows upon mourning. The funeral ceremonies are frequently very long; they include many operations which take place at intervals during many months. Thus they prolong and support the moral disturbance brought about by the death (cf. Hertz, La Representation collective de la mort, in Année Sociol., X, pp. 48 ff.). In a general way, a death marks a grave change of condition which has extended and enduring effects upon the group. It takes a long time to neutralize these effects.

[1268] In a case reported by Grey from the observations of Bussel, the rite has all the aspects of a sacrifice: the blood is sprinkled over the body itself (Grey, II, p. 330). In other cases, there is something like an offering of the beard: men in mourning cut off a part of their beards, which they throw on to the corpse (ibid., p. 335).

[1269] Nat. Tr., pp. 135-136.

[1270] Of course each churinga is believed to be connected with an ancestor. But it is not to appease the spirits of the ancestors that they mourn for the lost churinga. We have shown elsewhere (p. 123) that the idea of the ancestor only entered into the conception of the churinga secondarily and late.

[1271] Op. cit., p. 207; cf. p. 116.

[1272] Eylmann, p. 208.

[1273] Ibid., p. 211.

[1274] Howitt, The Dieri, in J.A.I., XX (1891), p. 93.

[1275] Howitt, Nat. Tr., p. 394.

[1276] Howitt, ibid., p. 396.

[1277] Communication of Gason in J.A.I., XXIV (1895), p. 175.

[1278] Nor. Tr., p. 286.

[1279] Gason, The Dieri Tribe, in Curr, II, p. 68.

[1280] Gason, The Dieri Tribe: Eylmann, p. 208.

[1281] Howitt, Nat. Tr., pp. 277 and 430.

[1282] Ibid., p. 195.

[1283] Gason, The Dieri Tribe, in Curr, II, p. 69. The same process is used to expiate a ridiculous act. Whenever anybody, by his awkwardness or otherwise, has caused the laughter of others, he asks one of them to beat him on the head until blood flows. Then things are all right again, and the one who was laughed at joins in the general gaiety (ibid., p. 70).

[1284] Eylmann, pp. 212 and 447.

[1285] See above, p. 385.

[1286] The Religion of the Semites, lect. XI.

[1287] This is the case in which the Dieri, according to Jason, invoke the Mura-mura of water during a drought.

[1288] Op. cit., p. 262.

[1289] It is also possible that the belief in the morally tempering virtues of suffering (see above, p. 312) has added something here. Since sorrow sanctifies and raises the religious level of the worshipper, it may also raise him up again when he falls lower than usual.

[1290] Cf. what we have said of expiation in our Division du travail social³, pp. 64 ff.

[1291] See above, p. 301.

[1292] Spencer and Gillen, Nat. Tr., p. 460; Nor. Tr., p. 601; Roth, North Queensland Ethnography, Bulletin No. 5, p. 24. It is useless to multiply references for so well-known a fact.

[1293] However, Spencer and Gillen cite one case where churinga are placed on the head of the dead man (Nat. Tr., p. 156). But they admit that the fact is unique and abnormal (ibid., p. 157), while Strehlow energetically denies it (II, p. 79).

[1294] Smith, Rel. of Semites, p. 153; cf. p. 446, the additional note, Holiness, Uncleanness and Taboo.

[1295] Howitt, Nat. Tr., pp. 448-450; Brough Smyth, I, pp. 118, 120; Dawson, p. 67; Eyre, II, p. 251; Roth, North Queensland Ethn., Bull. Mo. 9, in Rec. of the Austral. Museum, VI, No. 5, p. 367.

[1296] See above, p. 320.

[1297] Nor. Tr., p. 599; Nat. Tr., p. 464.

[1298] Among the Hebrews, for example, they sprinkled the altar with the blood of the expiatory victim (Lev. iv, 5 ff.); they burned the flesh and used products of this combustion to make water of purification (Numb. xix).

[1299] Taplin, The Narrinyeri, pp. 32-34. When two persons who have thus exchanged their umbilical cords belong to different tribes, they are used as inter-tribal messengers. In this case, the exchange of cords took place shortly after birth, through the intermediary of their respective parents.

[1300] It is true that Smith did not admit the reality of these substitutions and transformations. According to him, if the expiatory victim served to purify, it was because it had nothing impure in itself. At first, it was a holy thing; it was destined to re-establish, by means of a communion, the bonds of kinship uniting the worshipper to his god, when a ritual fault had strained or broken them. An exceptionally holy animal was chosen for this operation in order that the communion might be as efficacious as possible, and efface the effects of the fault as completely as possible. It was only when they no longer understood the meaning of the rite that the sacrosanct animal was considered impure (op. cit., pp. 347 ff.). But it is inadmissible that beliefs and practices as universal as these, which we find at the foundation of the expiatory sacrifice, should be the product of a mere error of interpretation. In fact, we cannot doubt that the expiatory victim was charged with the impurity of the sin. We have shown, moreover, that these transformations of the pure into the impure, or the contrary, are to be found in the most inferior societies which we know.

[1301] William James, The Varieties of Religious Experience.

[1302] Quoted by James, op. cit., p. 20.

[1303] See above, pp. 230 ff.

[1304] Only one form of social activity has not yet been expressly attached to religion: that is economic activity. Sometimes processes that are derived from magic have, by that fact alone, an origin that is indirectly religious. Also, economic value is a sort of power or efficacy, and we know the religious origins of the idea of power. Also, richness can confer mana; therefore it has it. Hence it is seen that the ideas of economic value and of religious value are not without connection. But the question of the nature of these connections has not yet been studied.

[1305] It is for this reason that Frazer and even Preuss set impersonal religious forces outside of, or at least on the threshold of religion, to attach them to magic.

[1306] Boutroux, Science et Religion, pp. 206-207.

[1307] See above, pp. 379 ff. On this same question, see also our article, "Représentations individuelles et représentations collectives," in the Revue de Métaphysique, May, 1898.

[1308] William James, Principles of Psychology, I, p. 464.

[1309] This universality of the concept should not be confused with its generality: they are very different things. What we mean by universality is the property which the concept has of being communicable to a number of minds, and in principle, to all minds; but this communicability is wholly independent of the degree of its extension. A concept which is applied to only one object, and whose extension is consequently at the minimum, can be the same for everybody: such is the case with the concept of a deity.

[1310] It may be objected that frequently, as the mere effect of repetition, ways of thinking and acting become fixed and crystallized in the individual, in the form of habits which resist change. But a habit is only a tendency to repeat an act or idea automatically every time that the same circumstances appear; it does not at all imply that the idea or act is in the form of an exemplary type, proposed to or imposed upon the mind or will. It is only when a type of this sort is set up, that is to say, when a rule or standard is established, that social action can and should be presumed.

[1311] Thus we see how far it is from being true that a conception lacks objective value merely because it has a social origin.

[1312] See also above, p. 208.

[1313] Lévy-Bruhl, Les fonctions mentales dans les sociétés inférieures, pp. 131-138.

[1314] Ibid., p. 446.

[1315] See above, p. 18.

[1316] William James, Principles of Psychology, I, p. 134.

[1317] Men frequently speak of space and time as if they were only concrete extent and duration, such as the individual consciousness can feel, but enfeebled by abstraction. In reality, they are representations of a wholly different sort, made out of other elements, according to a different plan, and with equally different ends in view.

[1318] At bottom, the concept of totality, that of society and that of divinity are very probably only different aspects of the same notion.

[1319] See our Classifications primitives, loc. cit., pp. 40 ff.