70 See Frazer, On Some Ceremonies of the Central Australian Tribes in Australian Association for the Advancement of Science, 1901, pp. 313 ff. This conception is also of an extreme generality. In India, the simple participation in the sacrificial act has the same effects; the sacrificer, by the mere act of entering within the circle of sacred things, changes his personality. (See, Hubert and Mauss, Essai sur le Sacrifice in the Année Sociologique, II, p. 101.)
71 See what was said of the initiation above, p. 39.
72 We shall point out below how, for example, certain species of sacred things exist, between which there is an incompatibility as all-exclusive as that between the sacred and the profane (Bk. III, ch. v, § 4).
73 This is the case with certain marriage and funeral rites, for example.
74 See Spencer and Gillen, Native Tribes of Central Australia, pp. 534 ff.; Northern Tribes of Central Australia, p. 463; Howitt, Native Tribes of S.E. Australia, pp. 359-361.
75 See Codrington, The Melanesians, ch. xii.
76 See Hubert, art. Magia in Dictionnaire des Antiquités.
77 For example, in Melanesia, the tindalo is a spirit, now religious, now magic (Codrington, pp. 125 ff., 194 ff.).
78 See Hubert and Mauss, Théorie Générale de la Magie, in Année Sociologique, vol. VII, pp. 83-84.
79 For example, the host is profaned in the black mass.
80 One turns his back to the altar, or goes around the altar commencing by the left instead of by the right.
81 Loc. cit., p. 19.
82 Undoubtedly it is rare that a ceremony does not have some director at the moment when it is celebrated; even in the most crudely organized societies, there are generally certain men whom the importance of their social position points out to exercise a directing influence over the religious life (for example, the chiefs of the local groups of certain Australian societies). But this attribution of functions is still very uncertain.
83 At Athens, the gods to whom the domestic cult was addressed were only specialized forms of the gods of the city (Ζεύς κτήσιος, Ζεύς ἑρκεῖος). In the same way, in the Middle Ages, the patrons of the guilds were saints of the calendar.
84 For the name Church is ordinarily applied only to a group whose common beliefs refer to a circle of more special affairs.
85 Hubert and Mauss, loc. cit., p. 18.
86 Robertson Smith has already pointed out that magic is opposed to religion, as the individual to the social (The Religion of the Semites, 2 edit., pp. 264-265). Also, in thus distinguishing magic from religion, we do not mean to establish a break of continuity between them. The frontiers between the two domains are frequently uncertain.
87 Codrington, Trans. and Proc. Roy. Soc. of Victoria, XVI, p. 136.
88 Negrioli, Dei Genii presso i Romani.
89 This is the conclusion reached by Spencer in his Ecclesiastical Institutions (ch. xvi), and by Sabatier in his Outlines of a Philosophy of Religion, based on Psychology and History (tr. by Seed), and by all the school to which he belongs.
90 Notably among numerous Indian tribes of North America.
91 This statement of fact does not touch the question whether exterior and public religion is not merely the development of an interior and personal religion which was the primitive fact, or whether, on the contrary, the second is not the projection of the first into individual consciences. The problem will be directly attacked below (Bk. II, ch. v, § 2, cf. the same book, ch. vi and vii, § 1). For the moment, we confine ourselves to remarking that the individual cult is presented to the observer as an element of, and something dependent upon, the collective cult.
92 It is by this that our present definition is connected to the one we have already proposed in the Année Sociologique. In this other work, we defined religious beliefs exclusively by their obligatory character; but, as we shall show, this obligation evidently comes from the fact that these beliefs are the possession of a group which imposes them upon its members. The two definitions are thus in a large part the same. If we have thought it best to propose a new one, it is because the first was too formal, and neglected the contents of the religious representations too much. It will be seen, in the discussions which follow, how important it is to put this characteristic into evidence at once. Moreover, if their imperative character is really a distinctive trait of religious beliefs, it allows of an infinite number of degrees; consequently there are even cases where it is not easily perceptible. Hence come difficulties and embarrassments which are avoided by substituting for this criterium the one we now employ.
CHAPTER II
LEADING CONCEPTIONS OF THE ELEMENTARY RELIGION
I. — Animism
Armed with this definition, we are now able to set out in search of this elementary religion which we propose to study.
Even the crudest religions with which history and ethnology make us acquainted are already of a complexity which corresponds badly with the idea sometimes held of primitive mentality. One finds there not only a confused system of beliefs and rites, but also such a plurality of different principles, and such a richness of essential notions, that it seems impossible to see in them anything but the late product of a rather long evolution. Hence it has been concluded that to discover the truly original form of the religious life, it is necessary to descend by analysis beyond these observable religions, to resolve them into their common and fundamental elements, and then