The Project Gutenberg eBook of The Elementary Forms of the Religious Life, by Emile Durkheim.

[114] Jevons has made a similar remark. With Tylor, he admits that the idea of the soul comes from dreams, and that after it was created, men projected it into things. But, he adds, the fact that nature has been conceived as animated like men does not explain how it became the object of a cult. "The man who believes the bowing tree or the leaping flame to be a living thing like himself, does not therefore believe it to be a supernatural being—rather, so far as it is like himself, it, like himself, is not supernatural" (Introduction to the History of Religions, p. 55).

[115] See Spencer and Gillen, Nor. Tr., p. 506, and Nat. Tr., p. 512.

[116] This is the ritual and mythical theme which Frazer studies in his Golden Bough.

[117] The Melanesians, p. 119.

[118] Ibid., p. 125.

[119] There are sometimes, as it seems, even funeral offerings. (See Roth, Superstition, Magic and Medicine, in North Queensland Ethnog., Bulletin No. 5, § 69 c., and Burial Customs, in ibid., No. 10, in Records of the Australian Museum, Vol. VI, No. 5, p. 395). But these offerings are not periodical.

[120] Spencer and Gillen, Nat. Tr., pp. 538, 553, and Nor. Tr., pp. 463, 543, 547.

[121] See especially, Spencer and Gillen, Northern Tribes, ch. vi, vii, ix.

[122] The Religions of Primitive Peoples, pp. 47 ff.

[123] Myth, Ritual and Religions, p. 123.

[124] Les Religions des peuples non civilisés, II, Conclusion.

[125] The Religion of the Semites, 2 ed., pp. 126, 132.

[126] This is the reasoning of Westermarck (Origins of Human Marriage, p. 6).

[127] By sexual communism we do not mean a state of promiscuity where man knows no matrimonial rules: we believe that such a state has never existed. But it has frequently happened that groups of men have been regularly united to one or several women.

[128] See our Suicide, pp. 233 ff.

[129] Spencer, Principles of Sociology, I, pp. 129 f.

[130] The Melanesians, p. 123.

[131] Dorsey, A Study of Siouan Cults, in XIth Annual Report of the Bureau of Amer. Ethnology, pp. 431 ff., and passim.

[132] La religion des peuples non civilisés, I, p. 248.

[133] V. W. de Visser, De Graecorum diis non referentibus speciem humanam. Cf. P. Perdrizet, Bulletin de correspondance hellénique, 1899, p. 635.

[134] However, according to Spencer, there is a germ of truth in the belief in spirits: this is the idea that "the power which manifests itself inside the consciousness is a different form of power from that manifested outside the consciousness" (Ecclesiastical Institutions, § 659). Spencer understands by this that the notion of force in general is the sentiment of the force which we have extended to the entire universe; this is what animism admits implicitly when it peoples nature with spirits analogous to our own. But even if this hypothesis in regard to the way in which the idea of force is formed were true—and it requires important reservations which we shall make (Bk. III, ch. iii, § 3)—it has nothing religious about it; it belongs to no cult. It thus remains that the system of religious symbols and rites, the classification of things into sacred and profane, all that which is really religious in religion, corresponds to nothing in reality. Also, this germ of truth, of which he speaks, is still more a germ of error, for if it be true that the forces of nature and those of the mind are related, they are profoundly distinct, and one exposes himself to grave misconceptions in identifying them.

[135] This is undoubtedly what explains the sympathy which folk-lorists like Mannhardt have felt for animistic ideas. In popular religions as in inferior religions, these spiritual beings of a second order hold the first place.

[136] In the essay entitled Comparative Mythology (pp. 47 ff).

[137] Herabkunft des Feuers und Gōttertranks, Berlin, 1859 (a new edition was given by Ernst Kuhn in 1886). Cf. Der Schuss des Wilden Jägers auf den Sonnen-hirsch, Zeitschrift f. d. Phil., I, 1869, pp. 89-169. Entwickelungsstufen des Mythus, Abhandl. d. Berl. Akad., 1873.

[138] Der Ursprung der Mythologie, Berlin, 1860.

[139] In his book Hercule et Cacus. Étude de mythologie comparée. Max Müller's Comparative Mythology is there signalized as a work "which marks a new epoch in the history of Mythology" (p. 12).

[140] Die Griechischen Kulte und Mythen, I, p. 78.

[141] Among others who have adopted this conception may be cited Renan. See his Nouvelles études d'histoire religieuse, 1884, p. 31.

[142] Aside from the Comparative Mythology, the works where Max Müller has exposed his general theories on religion are: Hibbert Lectures (1878) under the title The Origin and Development of Religion; Natural Religion (1889); Physical Religion (1890); Anthropological Religion (1892); Theosophy, or Psychological Religion (1893); Contributions to the Science of Mythology (1897). Since his mythological theories are closely related to his philosophy of language, these works should be consulted in connection with the ones consecrated to language or logic, especially Lectures on the Science of Language, and The Science of Thought.

[143] Natural Religion, p. 114.

[144] Physical Religion, pp. 119-120.

[145] Ibid., p. 121; cf. p. 304.

[146] Natural Religion, pp. 121 ff., and 149-155.

[147] "The overwhelming pressure of the infinite" (ibid., p. 138).

[148] Ibid., pp. 195-196.

[149] Max Müller even goes so far as to say that until thought has passed this first stage, it has very few of the characteristics which we now attribute to religion (Physic. Rel., p. 120).

[150] Physic. Rel., p. 128.

[151] The Science of Thought, p. 30.

[152] Natural Religion, pp. 393 ff.

[153] Physic. Rel., p. 133; The Science of Thought, p. 219; Lectures on the Science of Language, II, pp. 1 ff.

[154] The Science of Thought, p. 272.

[155] The Science of Thought, I, p. 327; Physic. Rel., pp. 125 ff.

[156] Mélanges de mythologie et de linguistique, p. 8.

[157] Anthropological Religion, pp. 128-130.

[158] This explanation is not as good as that of Tylor. According to Max Müller, men could not admit that life stopped with death; therefore they concluded that there were two beings within them, one of which survived the body. But it is hard to see what made them think that life continued after the body was decomposed.

[159] For the details, see Anthrop. Rel., pp. 351 ff.

[160] Anthrop. Rel., p. 130.—This is what keeps Max Müller from considering Christianity the climax of all this development. The religion of ancestors, he says, supposes that there is something divine in man. Now is that idea not the one at the basis of the teaching of Christ? (ibid., pp. 378 ff.). It is useless to insist upon the strangeness of the conception which makes Christianity the latest of the cults of the dead.

[161] See the discussion of the hypothesis in Gruppe, Griechishen Kulte und Mythen, pp. 79-184.

[162] See Meillet, Introduction à l'étude comparative des langues indo-européennes, p. 119.

[163] Oldenberg, Die Religion des Vedas, pp. 59 ff.; Meillet, Le dieu Iranien Mythra, in Journal Asiatique, X, No. 1, July-August, 1907, pp. 143 ff.

[164] In this category are a large number of the maxims of popular wisdom.

[165] It is true that this argument does not touch those who see in religion a code (especially of hygiene) whose provisions, though placed under the sanction of imaginary beings, are nevertheless well founded. But we shall not delay to discuss a conception so insupportable, and which has, in fact, never been sustained in a systematic manner by persons somewhat informed upon the history of religions. It is difficult to see what good the terrible practices of the initiation bring to the health which they threaten; what good the dietetic restrictions, which generally deal with perfectly clean animals, have hygienically; how sacrifices, which take place far from a house, make it more solid, etc. Undoubtedly there are religious precepts which at the same time have a practical utility; but they are lost in the mass of others, and even the services which they render are frequently not without some drawbacks. If there is a religiously enforced cleanliness, there is also a religious filthiness which is derived from these same principles. The rule which orders a corpse to be carried away from the camp because it is the seat of a dreaded spirit is undoubtedly useful. But the same belief requires the relatives to anoint themselves with the liquids which issue from a corpse in putrefaction, because they are supposed to have exceptional virtues.—From this point of view, magic has served a great deal more than religion.

[166] Contributions to the Science of Mythology, I, pp. 68 f.

[167] Lectures on the Science of Language, II, p. 456 ff.; Physic. Rel., pp. 276 ff.—Also Bréal, Mélanges, p. 6, "To bring the necessary clarity into this question of the origin of mythology, it is necessary to distinguish carefully the gods, which are the immediate product of the human intelligence, from the fables, which are its indirect and involuntary product."

[168] Max Müller recognized this. See Physic. Rel., p. 132, and Comparative Mythology, p. 58. "The gods are nomina and not numina, names without being and not beings without name."

[169] It is true that Max Müller held that for the Greeks, "Zeus was, and remained, in spite of all mythological obscurations, the name of the Supreme Deity" (Science of Language, II, p. 478). We shall not dispute this assertion, though it is historically contestable; but in any case, this conception of Zeus could never have been more than a glimmer in the midst of all the other religious beliefs of the Greeks.

Besides this, in a later work, Max Müller went so far as to make even the notion of god in general the product of a wholly verbal process and thus of a mythological elaboration (Physic. Rel., p. 138).

[170] Undoubtedly outside the real myths there were always fables which were not believed, or at least were not believed in the same way and to the same degree, and hence had no religious character. The line of demarcation between fables and myths is certainly floating and hard to determine. But this is no reason for making all myths stories, any more than we should dream of making all stories myths. There is at least one characteristic which in a number of cases suffices to differentiate the religious myth: that is its relation to the cult.

[171] See above, p. 28.

[172] More than that, in the language of Max Müller, there is a veritable abuse of words. Sensuous experience, he says, implies, at least in certain cases, "beyond the known, something unknown, something which I claim the liberty to call infinite" (Natural Rel., p. 195; cf. p. 218). The unknown is not necessarily the infinite, any more than the infinite is necessarily the unknown if it is in all points the same, and consequently like the part which we know. It would be necessary to prove that the part of it which we perceive differs in nature from that which we do not perceive.

[173] Max Müller involuntarily recognizes this in certain passages. He confesses that he sees little difference between Agni, the god of fire, and the notion of ether, by which the modern physicist explains light and heat (Phys. Rel., pp. 126 f.). Also, he connects the notion of divinity to that of agency (p. 138) or of a causality which is not natural and profane. The fact that religion represents the causes thus imagined, under the form of personal agents, is not enough to explain how they got a sacred character. A personal agent can be profane, and also, many religious forces are essentially impersonal.

[174] We shall see below, in speaking of the efficacy of rites and faith, how these illusions are to be explained (Bk. III, ch. ii).

[175] Voyages and Travels of an Indian Interpreter.

[176] This idea was so common that even M. Réville continued to make America the classic land of totemism (Religions des peuples non civilisés, I, p. 242).

[177] Journals of Two Expeditions in North-West and Western Australia, II, p. 228.

[178] The Worship of Animals and Plants. Totems and Totemism (1869, 1870).

[179] This idea is found already very clearly expressed in a study by Gallatin entitled Synopsis of the Indian Tribes (Archæologia Americana, II, pp. 109 ff.), and in a notice by Morgan in the Cambrian Journal, 1860, p. 149.

[180] This work had been prepared for and preceded by two others by the same author: The League of the Iroquois (1851), and Systems of Consanguinity and Affinity of the Human Family (1871).

[181] Kamilaroi and Kurnai, 1880.

[182] In the very first volumes of the Annual Report of the Bureau of American Ethnology are found the study of Powell, Wyandot Government (I, p. 59), that of Cushing, Zuñi Fetiches (II, p. 9), Smith, Myths of the Iroquois (ibid., p. 77), and the important work of Dorsey, Omaha Sociology (III, p. 211), which are also contributions to the study of totemism.

[183] This first appeared, in an abridged form, in the Encyclopædia Britannica (9th ed.).

[184] In his Primitive Culture, Tylor had already attempted an explanation of totemism, to which we shall return presently, but which we shall not give here; for by making totemism only a particular case of the ancestor-cult, he completely misunderstood its importance. In this chapter we mention only those theories which have contributed to the progress of the study of totemism.

[185] Published at Cambridge, 1885.

[186] First edition, 1889. This is the arrangement of a course given at the University of Aberdeen in 1888. Cf. the article Sacrifice in the Encyclopædia Britannica (9th edition).

[187] London, 1890. A second edition in three volumes has since appeared (1900) and a third in five volumes is already in course of publication.

[188] In this connection must be mentioned the interesting work of Sidney Hartland, The Legend of Perseus, 3 vols., 1894-1896.

[189] We here confine ourselves to giving the names of the authors; their works will be indicated below, when we make use of them.

[190] If Spencer and Gillen have been the first to study these tribes in a scientific and thorough manner, they were not the first to talk about them. Howitt had already described the social organization of the Wuaramongo (Warramunga of Spencer and Gillen) in 1888 in his Further Notes on the Australian Classes in The Journal of the Anthropological Institute (hereafter, J.A.I.), pp. 44 f. The Arunta had already been briefly studied by Schulze (The Aborigines of the Upper and Middle Finke River, in Transactions of the Royal Society of South Australia, Vol. XIV, fasc. 2): the organization of the Chingalee (the Tjingilli of Spencer and Gillen), the Wombya, etc., by Mathews (Wombya Organization of the Australian Aborigines, in American Anthropologist, New Series, Vol. II, p. 494; Divisions of some West Australian Tribes, ibid., p. 185; Proceedings Amer. Philos. Soc., XXXVII, pp. 151-152, and Journal Roy. Soc. of N.S. Wales, XXXII, p. 71 and XXXIII, p. 111). The first results of the study made of the Arunta had also been published already in the Report on the Work of the Horn Scientific Expedition to Central Australia, Pt. IV (1896). The first part of this Report is by Stirling, the second by Gillen; the entire publication was placed under the direction of Baldwin Spencer.

[191] London, 1899. Hereafter, Native Tribes or Nat. Tr.

[192] London, 1904. Hereafter, Northern Tribes or Nor. Tr.

[193] We write the Arunta, the Anula, the Tjingilli, etc., without adding the characteristic s of the plural. It does not seem very logical to add to these words, which are not European, a grammatical sign which would have no meaning except in our languages. Exceptions to this rule will be made when the name of the tribe has obviously been Europeanized (the Hurons for example).

[194] Strehlow has been in Australia since 1892; at first he lived among the Dieri, and from them he went to the Arunta.

[195] Die Aranda- und Loritja-Stämme in Zentral Australien. Four fascicules have been published up to the present. The last appeared at the moment when the present book was finished, so it could not be used. The two first have to do with the myths and legends, and the third with the cult. It is only just to add to the name of Strehlow that of von Leonhardi, who has had a great deal to do with this publication. Not only has he charged himself with editing the manuscripts of Strehlow, but by his judicious questions he has led the latter to be more precise on more than one point. It would be useful also to consult an article which von Leonhardi gave the Globus, where numerous extracts from his correspondence with Strehlow will be found (Ueber einige religiöse und totemistische Vorstellungen der Aranda und Loritja in Zentral Australien, in Globus, XCI, p. 285). Cf. an article on the same subject by N. W. Thomas in Folk-lore, XVI, pp. 428 ff.

[196] Spencer and Gillen are not ignorant of it, but they are far from possessing it as thoroughly as Strehlow.

[197] Notably by Klaatsch, Schlussbericht über meine Reise nach Australien, in Zeitschrift f. Ethnologie, 1907, pp. 635 ff.

[198] The book of K. Langloh Parker, The Euahlayi Tribe, that of Eylmann, Die Eingeborenen der Kolonie Südaustralien; that of John Mathews, Two Representative Tribes of Queensland, and certain recent articles of Mathews all show the influence of Spencer and Gillen.

[199] A list of these publications will be found in the preface to his Nat. Tr., pp. 8-9.

[200] London, 1904. Hereafter we shall cite this work by the abbreviation Nat. Tr., but always mentioning the name of Howitt, to distinguish it from the first work of Spencer and Gillen, which we abbreviate in the same manner.

[201] Totemism and Exogamy, 4 vols., London, 1910. The work begins with a re-edition of Totemism, reproduced without any essential changes.

[202] It is true that at the end and at the beginning there are some general theories on totemism, which will be described and discussed below. But these theories are relatively independent of the collection of facts which accompanies them, for they had already been published in different articles in reviews, long before this work appeared. These articles are reproduced in the first volume (pp. 89-172).

[203] Totemism, p. 12.

[204] Ibid., p. 15.

[205] Ibid., p. 32.

[206] It should be noted that in this connection, the more recent work, Totemism and Exogamy, shows an important progress in the thought as well as the method of Frazer. Every time that he describes the religious or domestic institutions of a tribe, he sets himself to determine the geographic and social conditions in which this tribe is placed. Howsoever summary these analyses may be, they bear witness nevertheless to a rupture with the old methods of the anthropological school.

[207] Undoubtedly we also consider that the principal object of the science of religions is to find out what the religious nature of man really consists in. However, as we do not regard it as a part of his constitutional make-up, but rather as the product of social causes, we consider it impossible to find it, if we leave aside his social environment.

[208] We cannot repeat too frequently that the importance which we attach to totemism is absolutely independent of whether it was ever universal or not.

[209] This is the case with the phratries and matrimonial classes; on this point, see Spencer and Gillen, Northern Tribes, ch. iii; Howitt, Native Tribes, pp. 109 and 137-142; Thomas, Kinship and Marriage in Australia, ch. vi and vii.

[210] Division du Travail social, 3rd ed., p. 150.

[211] It is to be understood that this is not always the case. It frequently happens, as we have already said, that the simpler forms aid to a better understanding of the more complex. On this point, there is no rule of method which is applicable to every possible case.

[212] Thus the individual totemism of America will aid us in understanding the function and importance of that in Australia. As the latter is very rudimentary, it would probably have passed unobserved.

[213] Besides, there is not one unique type of totemism in America, but several different species which must be distinguished.

[214] We shall leave this field only very exceptionally, and when a particularly instructive comparison seems to us to impose itself.

[215] This is the definition given by Cicero: Gentiles sunt qui inter se eodem nomine sunt (Top. 6). (Those are of the same gens who have the same name among themselves.)

[216] It may be said in a general way that the clan is a family group, where kinship results solely from a common name; it is in this sense that the gens is a clan. But the totemic clan is a particular sort of the class thus constituted.

[217] In a certain sense, these bonds of solidarity extend even beyond the frontiers of the tribe. When individuals of different tribes have the same totem, they have peculiar duties towards each other. This fact is expressly stated for certain tribes of North America (see Frazer, Totemism and Exogamy, III, pp. 57, 81, 299, 356-357). The texts relative to Australia are less explicit. However, it is probable that the prohibition of marriage between members of a single totem is international.

[218] Morgan, Ancient Society, p. 165.

[219] In Australia the words employed differ with the tribes. In the regions observed by Grey, they said Kobong; the Dieri say Murdu (Howitt, Nat. Tr., p. 91); the Narrinyeri, Ngaitye (Talpin, in Curr, II, p. 244); the Warramunga, Mungái or Mungáii (Nor. Tr., p. 754), etc.

[220] Indian Tribes of the United States, IV, p. 86.

[221] This fortune of the word is the more regrettable since we do not even know exactly how it is written. Some write totam, others toodaim, or dodaim, or ododam (see Frazer, Totemism, p. 1). Nor is the meaning of the word determined exactly. According to the report of the first observer of the Ojibway, J. Long, the word totam designated the protecting genius, the individual totem, of which we shall speak below (Bk. II, ch. iv) and not the totem of the clan. But the accounts of other explorers say exactly the contrary (on this point, see Frazer, Totemism and Exogamy, III, pp. 49-52).