inconsistent with one another: the one aims at destroying the bones, and is observed in the case of sacred animals; the other at preserving them, and is observed in the case of game. Another savage parallel may be found in a belief already illustrated,1 namely, that the food of a divine king, such as the Mikado, or a superior chief, is fatal to his subjects or slaves. Much more, therefore, would the sacrificial animal of which a god had partaken be fatal, and great would be the need to save incautious, heedless persons from the danger of eating the remains which they might find lying about. Here we are approaching the true explanation; but, since we hope to show before the end of this chapter that the conception of the god's eating the victim only came relatively late, we cannot see in it the origin of the primitive custom in question, though we do see in it a powerful reinforcement thereof. Again, it is a savage belief that you can injure a man not merely by means of his nail-parings, hair-clippings, and other things associated with him, but also by the refuse of his food. In Victoria, the natives believe that "if an enemy gets possession of anything that has belonged to them, even such things as bones of animals they have eaten, broken weapons, feathers, portions of dress, pieces of skin, or refuse of any kind, he can employ it as a charm to produce illness. in the person to whom they belonged. They are therefore very careful to burn up all rubbish or uncleanness before leaving a camping-place ";" and "the practice of using a man's food to injure him is found in Polynesia generally, Tahiti, the Washington Islands, Fiji, Queensland, and amongst the Zulus and Kaffirs." 3 Now, this belief, coexisting as it does in Polynesia with the custom of burying the remnants of the sacrificial meal, cannot but strengthen the observance of that custom. But it is to be doubted whether it was the origin of the practice. The eagerness displayed by the Saracen worshippers to obtain a portion of the victim, and the dismay of Hakluyt's West Indians if they failed to get a piece, both show that originally, as in Peru, the victim was accounted 2 Dawson, Australian Aborigines, 54. 1 Supra, pp. 83, 84. * Folk-Lore, vi. 134, note 2. "very sacred indeed"; and that the emotion which swayed the worshippers, and their motive for devouring the whole of the victim, was not fear lest the remnants should be used against them, still less anxiety about what might happen to incautious strangers, but desire on the part of each to obtain for himself as much as possible of something that was in the highest degree desirable. Now, that the sacrificial animal should be accounted "very sacred indeed" is intelligible enough, if it was (in the savage times when the whole victim was consumed) the totem animal and god of the clan making the sacrifice. As for the eagerness of the worshippers, it need not be doubted; but of the savage's motives for that eagerness we ought to try and form for ourselves some clear idea. In the sacrificial rite itself, as an external act of worship, the essential feature is that the worshipper should partake of the offering; but it is only after a time that this central feature disengages itself from the repulsive accessories which were indeed inevitable concomitants of a savage feast, but were no part of the essence of the rite. We may therefore reasonably expect to find the rite on its inward side, i.e. as it presented itself to the worshipper, following a parallel line of development. That the idea of "communication and communion with spiritual beings," which, as we have seen, is the Chinese conception of sacrifice, is the aspect of the rite which has persisted longest, we will take for granted. Whether it was present dimly, and obscured or overlaid by other associations, but still implicitly present to the consciousness of savage man, is a question which depends for its answer on what view we take of that identity in difference which exists between civilised and uncivilised man, and makes the whole world kin. We may regard selfishness and the baser desires as alone "natural" and as constituting the sole identity; or, by the same question-begging epithet, we may credit the savage with the "natural" affections as well. The question has always divided philosophers, not merely in Europe, but in China, where Seun sides with Hobbes, and Han-yu anticipated the view of Butler that good instincts as well as bad are natural. If, therefore, here we take our stand, without hesitation, but without argument, on the side of the latter, it other view, but because We shall therefore, with is not that we wish to ignore the this is not the place to discuss it. the reader's leave, assume that the mere existence of the family and of the clan implies the existence of some measure of affection between parents and children and between bloodrelations. But if this be granted, the rest follows: where affection exists in one direction it may come to exist in others; and communion is sought only with those towards whom we have affection. Here, then, lay the germ in the conception of the clan-god as a permanently friendly power. As the leader of the clan in war, he claimed and received the affectionate loyalty of those on whom he conferred protection and victory; as the father of his worshippers, the filial affection of his children. It was not always or everywhere that the seed bore fruit: in the case of many savages still existing, e.g. most or all of the Australian aborigines, the conception of the totem-god as a protecting power has been lost, and they have lapsed almost into their original animism. But where it did germinate, its growth was accompanied by the intellectual and material development, by the movement towards civilisation, of the peoples amongst whom it flourished. But the desire for union with the spiritual being with whom the fate and fortunes of the tribe were identified, was necessarily in savage times enveloped and conditioned by savage modes of thought and savage views of nature and her processes. One of these views has been called in by some writers to explain in part the motive with which the sacrificial victim was originally eaten it is that with the flesh the qualities of the animal are absorbed and assimilated; and as a matter of fact some savages do eat tiger to give them courage, or deer to give them fleetness. But, it is important to note, it is not the characteristic quality of the totem animal that the savage, in his sacrificial meal, desires to appropriate many or most totems-turtle, snail, cockle, etc. -have, as mere animals, no obviously desirable qualities to recommend them. It is not the natural but the supernatural 66 fact 1 Professor Tylor (Academy, No. 1237, N.S. p. 49) regards it as a that savage families, with all their rough ways, are held together by a bond of unselfish kindness, which is one of the wonders of human nature.” qualities of the totem that the savage wishes to assimilate. It is as god, not as animal, that the totem furnishes the sacrificial meal. The savage seeks against the supernatural powers by which he is surrounded a supernatural ally; and it is in the confidence which the sacrificial rite affords him that he undertakes that forcible, physical expulsion of evil spirits which has already been mentioned.1 Hence, then, his eagerness to partake of the victim-an eagerness so great that none of the animal was left uneaten. It was the desire to fortify himself as completely as possible for the dangerous encounter for which it was the preparation. When, however, advancing civilisation made the complete consumption of the animal impossible, the remnants of the sacrificial feast were naturally treated with every precaution known to the savage, both to protect himself against his enemies, and to protect his friends against the danger of inadvertently eating food so highly taboo as was the flesh of a totem animal. Here, perhaps, the reader may feel it a difficulty that the totem animal should be tabooed food and yet should be eaten by his worshippers. The difficulty and its solution are exactly the same here as in connection with intruding on holy places. Such places are indeed forbidden ground, yet those who would seek the god must enter them, and so may enter them for that purpose and with due precautions. On the Loango Coast, the sanctuary of a certain god may be entered by those who seek his aid, but all others become his slaves for ever if they trespass on his precincts." Now, what is characteristic of the sacrificial meal all over the world is precisely the fact that it is distinguished from ordinary eating by restrictions and precautions which are the same everywhere and amongst all races: the meal must be eaten in a certain place, at a certain time, by certain persons, in a certain way, for a certain purpose. As we have seen, only clansmen may eat of it, and everyone of them must partake of it. They must consume it, wholly, in the sanctuary, there and then. It is not at all times that the rite is celebrated, but once a year that the feast is held and the conflict with evil spirits undertaken-and then only after due preparation by fasting, etc.; for, as those who have come into contact with 1 Supra, p. 105, 2 Supra, p. 63; Bastian, Loango Küste, 218, things taboo, e.g. mourners, have to fast, etc., so those who are about to enter into such contact have to observe the same rule. The "unclean" must not communicate their uncleanness to the community; much more, therefore, must those who are about to enter into relation with sacred things avoid carrying with them any uncleanness; and in both cases they are tabooed, i.e. isolated, for a time, that they may not, in the one instance, contract, or in the other, communicate, "uncleanness." From this point of view it is possible to explain another restriction, or rather precaution, namely, that which requires the sacrifice to be nocturnal. The fasting which is obligatory on mourners is only compulsory during the daylight; and the same remark applies to the fasting of those who are under a vow.2 The annual sacrifice and eating of the god could not, however, continue to be the only sacrifice: pestilence, which proved the presence of evil spirits and the necessity of expelling them; war, which involved an encounter not merely with the human foe but with his supernatural ally,3 came at irregular periods, and consequently the annual rite came to be supplemented by other sacrifices. Not only did the number of these supplementary sacrifices come to be increased, but the character of the rite was greatly changed in pastoral times. But, before going on to pastoral times, it will be well to ask how our argument stands exactly with regard to the pre-pastoral period, when man lived by hunting and fishing, and, in a word, was on the natural basis of subsistence. It stands thus: on the one hand, we find savages, who are still on the natural basis, treating their totem animals as gods, sometimes-not always, for we know totemism only in various stages of decay. On the other hand, we find in pastoral times, or later, animals sacrificed which once had been, and in Egypt even still were, totems. For instance, on the Gold Coast there is a god Brahfo, " antelopes are sacred to him, and no worshipper of Brahfo may molest one or eat of its flesh," yet once a year an antelope is killed and "the 2 Ibid. 1 Supra, pp. 77, 78. Hence it is that war is regarded by so many savages as a religious function, for which preparation must be made by various forms of abstinence and purification and other religious rites and ceremonies, e.g. those of the fetiales, 4 Ellis, Tshi-speaking Peoples, 64, |