But whence comes the religious character of the totemic beliefs and practices? For the fact that a man considers himself an animal of a certain species does not explain why he attributes marvellous powers to this species, and especially why he renders a cult to the images symbolizing it.—To this question Lang gives the same response as Frazer: he denies that totemism is a religion. "I find in Australia," he says, "no example of religious practices such as praying to, nourishing or burying the totem."[607] It was only at a later epoch, when it was already established, that totemism was drawn into and surrounded by a system of conceptions properly called religious. According to a remark of Howitt,[608] when the natives undertake the explanation of the totemic institutions, they do not attribute them to the totems themselves nor to a man, but to some supernatural being such as Bunjil or Baiame. "Accepting this evidence," says Lang, "one source of the 'religious' character of totemism is at once revealed. The totemist obeys the decree of Bunjil, or Baiame, as the Cretans obeyed the divine decrees given by Zeus to Minos." Now according to Lang the idea of these great divinities arose outside of the totemic system; so this is not a religion in itself; it has merely been given a religious colouring by contact with a genuine religion.
But these very myths contradict Lang's conception of totemism. If the Australians had regarded totemism as something human and profane, it would never have occurred to them to make a divine institution out of it. If, on the other hand, they have felt the need of connecting it with a divinity, it is because they have seen a sacred character in it. So these mythological interpretations prove the religious nature of totemism, but do not explain it.
Moreover, Lang himself recognizes that this solution is not sufficient. He realizes that totemic things are treated with a religious respect;[609] that especially the blood of an animal, as well as that of a man, is the object of numerous interdictions, or, as he says, taboos which this comparatively late mythology cannot explain.[610] Then where do they come from? Here are the words with which Lang answers this question: "As soon as the animal-named groups evolved the universally diffused beliefs about the wakan or mana, or mystically sacred quality of the blood as the life, they would also develop the various taboos."[611] The words wakan and mana, as we shall see in the following chapter, involve the very idea of sacredness itself; the one is taken from the language of the Sioux, the other from that of the Melanesian peoples. To explain the sacred character of totemic things by postulating this characteristic, is to answer the question by the question. What we must find out is whence this idea of wakan comes and how it comes to be applied to the totem and all that is derived from it. As long as these two questions remain unanswered, nothing is explained.
V
We have now passed in review all the principal explanations which have been given for totemic beliefs,[612] leaving to each of them its own individuality. But now that this examination is finished, we may state one criticism which addresses itself to all these systems alike.
If we stick to the letter of the formulæ, it seems that these may be arranged in two groups. Some (Frazer, Lang) deny the religious character of totemism; in reality, that amounts to denying the facts. Others recognize this, but think that they can explain it by deriving it from an anterior religion out of which totemism developed. But as a matter of fact, this distinction is only apparent: the first group is contained within the second. Neither Frazer nor Lang have been able to maintain their principle systematically and explain totemism as if it were not a religion. By the very force of facts, they have been compelled to slip ideas of a religious nature into their explanations. We have just seen how Lang calls in the idea of sacredness, which is the cardinal idea of all religion. Frazer, on his side, in each of the theories which he has successively proposed, appeals openly to the idea of souls or spirits; for according to him, totemism came from the fact that men thought they could deposit their souls in safety in some external object, or else that they attributed conception to a sort of spiritual fecundation of which a spirit was the agent. Now a soul, and still more, a spirit, are sacred things and the object of rites; so the ideas expressing them are essentially religious and it is therefore in vain that Frazer makes totemism a mere system of magic, for he succeeds in explaining it only in the terms of another religion.
We have already pointed out the insufficiencies of animism and naturism; so one may not have recourse to them, as Tylor and Jevons do, without exposing himself to these same objections. Yet neither Frazer nor Lang seems to dream of the possibility of another hypothesis.[613] On the other hand, we know that totemism is tightly bound up with the most primitive social system which we know, and in all probability, of which we can conceive. To suppose that it has developed out of another religion, differing from it only in degree, is to leave the data of observation and enter into the domain of arbitrary and unverifiable conjectures. If we wish to remain in harmony with the results we have already obtained, it is necessary that while affirming the religious nature of totemism, we abstain from deriving it from another different religion. There can be no hope of assigning it non-religious ideas as its cause. But among the representations entering into the conditions from which it results, there may be some which directly suggest a religious nature of themselves. These are the ones we must look for.