The Elementary Forms of the Religious Life Part 10

ORIGINS OF THESE BELIEFS

_Critical Examination of Preceding Theories_

The beliefs which we have just summarized are manifestly of a religious nature, since they imply a division of things into sacred and profane.

It is certain that there is no thought of spiritual beings, and in the course of our exposition we have not even had occasion to p.r.o.nounce the words, spirits, genii or divine personalities. But if certain writers, of whom we shall have something more to say presently, have, for this reason, refused to regard totemism as a religion, it is because they have an inexact notion of what religious phenomena are.

On the other hand, we are a.s.sured that this religion is the most primitive one that is now observable and even, in all probability, that has ever existed. In fact, it is inseparable from a social organization on a clan basis. Not only is it impossible, as we have already pointed out, to define it except in connection with the clan, but it even seems as though the clan could not exist, in the form it has taken in a great number of Australian societies, without the totem. For the members of a single clan are not united to each other either by a common habitat or by common blood, as they are not necessarily consanguineous and are frequently scattered over different parts of the tribal territory. Their unity comes solely from their having the same name and the same emblem, their believing that they have the same relations with the same categories of things, their practising the same rites, or, in a word, from their partic.i.p.ating in the same totemic cult. Thus totemism and the clan mutually imply each other, in so far, at least, as the latter is not confounded with the local group. Now the social organization on a clan basis is the simplest which we know. In fact, it exists in all its essential elements from the moment when the society includes two primary clans; consequently, we may say that there are none more rudimentary, as long as societies reduced to a single clan have not been discovered, and we believe that up to the present no traces of such have been found. A religion so closely connected to a social system surpa.s.sing all others in simplicity may well be regarded as the most elementary religion we can possibly know. If we succeed in discovering the origins of the beliefs which we have just a.n.a.lysed, we shall very probably discover at the same time the causes leading to the rise of the religious sentiment in humanity.

But before treating this question for ourselves, we must examine the most authorized solutions of it which have already been proposed.

I

In the first place, we find a group of scholars who believe that they can account for totemism by deriving it from some previous religion.

For Tylor[552] and Wilken,[553] totemism is a special form of the cult of the ancestors; it was the widespread doctrine of the transmigration of souls that served as a bridge between these two religious systems. A large number of peoples believe that after death, the soul does not remain disincarnate for ever, but presently animates another living body; on the other hand, "the lower psychology, drawing no definite line of demarcation between the souls of men and of beasts, can at least admit without difficulty the transmigration of human souls into the bodies of the lower animals."[554] Tylor cites a certain number of cases.[555] Under these circ.u.mstances, the religious respect inspired by the ancestor is quite naturally attached to the animal or plant with which he is presently confounded. The animal thus serving as a receptacle for a venerated being becomes a holy thing, the object of a cult, that is, a totem, for all the descendants of the ancestor, who form the clan descended from him.

Facts pointed out by Wilken among the societies of the Malay Archipelago would tend to prove that it really was in this manner that the totemic beliefs originated. In Java and Sumatra, crocodiles are especially honoured; they are regarded as benevolent protectors who must not be killed; offerings are made to them. Now the cult thus rendered to them is due to their being supposed to incarnate the souls of ancestors. The Malays of the Philippines consider the crocodile their grandfather; the tiger is treated in the same way for the same reasons. Similar beliefs have been observed among the Bantous.[556] In Melanesia it sometimes happens that an influential man, at the moment of death, announces his desire to reincarnate himself in a certain animal or plant; it is easily understood how the object thus chosen as his posthumous residence becomes sacred for his whole family.[557] So, far from being a primitive fact, totemism would seem to be the product of a more complex religion which preceded it.[558]

But the societies from which these facts were taken had already arrived at a rather advanced stage of culture; in any case, they had pa.s.sed the stage of pure totemism. They have families and not totemic clans.[559]

Even the majority of the animals to which religious honours are thus rendered are venerated, not by special groups of families, but by the tribes as a whole. So if these beliefs and practices do have some connection with ancient totemic cults, they now represent only altered forms of them[560] and are consequently not very well fitted for showing us their origins. It is not by studying an inst.i.tution at the moment when it is in full decadence that we can learn how it was formed. If we want to know how totemism originated, it is neither in Java nor Sumatra nor Melanesia that we must study it, but in Australia. Here we find neither a cult of the dead[561] nor the doctrine of transmigration. Of course they believe that the mythical heroes, the founders of the clan, reincarnate themselves periodically; _but this is in human bodies only_; each birth, as we shall see, is the product of one of these reincarnations. So if the animals of the totemic species are the object of rites, it is not because the ancestral souls are believed to reside in them. It is true that the first ancestors are frequently represented under the form of an animal, and this very common representation is an important fact for which we must account; but it was not the belief in metempsychosis which gave it birth, for this belief is unknown among Australian societies.

Moreover, far from being able to explain totemism, this belief takes for granted one of the fundamental principles upon which this rests; that is to say, it begs the question to be explained. It, just as much as totemism, implies that man is considered a close relative of the animal; for if these two kingdoms were clearly distinguished in the mind, men would never believe that a human soul could pa.s.s so easily from one into the other. It is even necessary that the body of the animal be considered its true home, for it is believed to go there as soon as it regains its liberty. Now while the doctrine of transmigration postulates this singular affinity, it offers no explanation of it. The only explanation offered by Tylor is that men sometimes resemble in certain traits the anatomy and physiology of the animal. "The half-human features and actions and characters of animals are watched with wondering sympathy by the savage, as by the child. The beast is the very incarnation of familiar qualities of man: and such names as lion, bear, fox, owl, parrot, viper, worm, when we apply them as epithets to men, condense into a word some leading features of a human life."[562] But even if these resemblances are met with, they are uncertain and exceptional; before all else, men resemble their relatives and companions, and not plants and animals. Such rare and questionable a.n.a.logies could not overcome such unanimous proofs, nor could they lead a man to think of himself and his forefathers in forms contradicted by daily experience. So this question remains untouched, and as long as it is not answered, we cannot say that totemism is explained.[563]

Finally, this whole theory rests upon a fundamental misunderstanding.

For Tylor as for Wundt, totemism is only a particular case of the cult of animals.[564] But we, on the contrary, know that it is something very different from a sort of animal-wors.h.i.+p.[565] The animal is never adored; the man is nearly its equal and sometimes even treats it as his possession, so far is he from being subordinate to it like a believer before his G.o.d. If the animals of the totemic species are really believed to incarnate the ancestors, the members of foreign clans would not be allowed to eat their flesh freely. In reality, it is not to the animal as such that the cult is addressed, but to the emblem and the image of the totem. Now between this religion of the emblem and the ancestor-cult, there is no connection whatsoever.

While Tylor derives totemism from the ancestor-cult, Jevons derives it from the nature-cult,[566] and here is how he does so.

When, under the impulse of the surprise occasioned by the irregularities observed in the course of phenomena, men had once peopled the world with supernatural beings,[567] they felt the need of making agreements with these redoubtable forces with which they had surrounded themselves. They understood that the best way to escape being overwhelmed by them was to ally themselves to some of them, and thus make sure of their aid. But at this period of history men knew no other form of alliance and a.s.sociation than the one resulting from kins.h.i.+p. All the members of a single clan aid each other mutually because they are kindred or, as amounts to the same thing, because they think they are; on the other hand, different clans treat each other as enemies because they are of different blood. So the only way of a.s.suring themselves of the support of these supernatural beings was to adopt them as kindred and to be adopted by them in the same quality: the well-known processes of the blood-covenant permitted them to attain this result quite easily. But since at this period, the individual did not yet have a real personality, and was regarded only as a part of his group, or clan, it was the clan as a whole, and not the individual, which collectively contracted this relations.h.i.+p. For the same reason, it was contracted, not with a particular object, but with the natural group or species of which this object was a part; for men think of the world as they think of themselves, and just as they could not conceive themselves apart from their clans, so they were unable to conceive of anything else as distinct from the species to which it belonged. Now a species of things united to a clan by a bond of kins.h.i.+p is, says Jevons, a totem.

In fact, it is certain that totemism implies the close a.s.sociation of a clan to a determined category of objects. But that this a.s.sociation was contracted with a deliberate design and in the full consciousness of an end sought after, as Jevons would have us believe, is a statement having but little harmony with what history teaches. Religions are too complex, and answer to needs that are too many and too obscure, to have their origin in a premeditated act of the will. And while it sins through over-simplicity, this hypothesis is also highly improbable. It says that men sought to a.s.sure themselves of the aid of the supernatural beings upon which things depend. Then they should preferably have addressed themselves to the most powerful of these, and to those whose protection promised to be the most beneficial.[568] But quite on the contrary, the beings with whom they have formed this mystic kins.h.i.+p are often among the most humble which exist. Also, if it were only a question of making allies and defenders, they would have tried to make as many as possible; for one cannot be defended too well. Yet as a matter of fact, each clan systematically contents itself with a single totem, that is to say, with one single protector, leaving the other clans to enjoy their own in perfect liberty. Each group confines itself within its own religious domain, never seeking to trespa.s.s upon that of its neighbours. This reserve and moderation are inexplicable according to the hypothesis under consideration.

II

Moreover, all these theories are wrong in omitting one question which dominates the whole subject. We have seen that there are two sorts of totemism: that of the individual and that of the clan. There is too evident a kins.h.i.+p between the two for them not to have some connection with each other. So we may well ask if one is not derived from the other, and, in the case of an affirmative answer, which is the more primitive; according to the solution accepted, the problem of the origins of totemism will be posed in different terms. This question becomes all the more necessary because of its general interest.

Individual totemism is an individual aspect of the totemic cult. Then if it was the primitive fact, we must say that religion is born in the consciousness of the individual, that before all else, it answers to individual aspirations, and that its collective form is merely secondary.

The desire for an undue simplicity, with which ethnologists and sociologists are too frequently inspired, has naturally led many scholars to explain, here as elsewhere, the complex by the simple, the totem of the group by that of the individual. Such, in fact, is the theory sustained by Frazer in his _Golden Bough_,[569] by Hill Tout,[570] by Miss Fletcher,[571] by Boas[572] and by Swanton.[573] It has the additional advantage of being in harmony with the conception of religion which is currently held; this is quite generally regarded as something intimate and personal. From this point of view, the totem of the clan can only be an individual totem which has become generalized.

Some eminent man, having found from experience the value of a totem he chose for himself by his own free will, transmitted it to his descendants; these latter, multiplying as time went on, finally formed the extended family known as a clan, and thus the totem became collective.

Hill Tout believes that he has found a proof supporting this theory in the way totemism has spread among certain societies of North-western America, especially among the Salish and certain Indians on the Thompson River. Individual totemism and the clan totemism are both found among these peoples; but they either do not co-exist in the same tribe, or else, when they do co-exist, they are not equally developed. They vary in an inverse proportion to each other; where the clan totem tends to become the general rule, the individual totem tends to disappear, and _vice versa_. Is that not as much as to say that the first is a more recent form of the second, which excludes it by replacing it?[574]

Mythology seems to confirm this interpretation. In these same societies, in fact, the ancestor of the clan is not a totemic animal; the founder of the group is generally represented in the form of a human being who, at a certain time, had entered into familiar relations with a fabulous animal from whom he received his totemic emblem. This emblem, together with the special powers which are attached to it, was then pa.s.sed on to the descendants of this mythical hero by right of heritage. So these people themselves seem to consider the collective totem as an individual one, perpetuated in the same family.[575] Moreover, it still happens to-day that a father transmits his own totem to his children. So if we imagine that the collective totem had, in a general way, this same origin, we are a.s.suming that the same thing took place in the past which is still observable to-day.[576]

It is still to be explained whence the individual totem comes. The reply given to this question varies with different authors.

Hill Tout considers it a particular case of fetis.h.i.+sm. Feeling himself surrounded on all sides by dreaded spirits, the individual experienced that sentiment which we have just seen Jevons attribute to the clan: in order that he might continue to exist, he sought some powerful protector in this mysterious world. Thus the use of a personal totem became established.[577] For Frazer, this same inst.i.tution was rather a subterfuge or trick of war, invented by men that they might escape from certain dangers. It is known that according to a belief which is very widespread in a large number of inferior societies, the human soul is able, without great inconvenience, to quit the body it inhabits for a while; howsoever far away it may be, it continues to animate this body by a sort of detached control. Then, in certain critical moments, when life is supposed to be particularly menaced, it may be desirable to withdraw the soul from the body and lead it to some place or into some object where it will be in greater security. In fact, there are a certain number of practices whose object is to withdraw the soul in order to protect it from some danger, either real or imaginary. For example, at the moment when men are going to enter a newly-built house, a magician removes their souls and puts them in a sack, to be saved and returned to their proprietors after the door-sill has been crossed. This is because the moment when one enters a new house is exceptionally critical; one may have disturbed, and consequently offended, the spirits who reside in the ground and especially under the sill, and if precautions are not taken, these could make a man pay dearly for his audacity. But when this danger is once pa.s.sed, and one has been able to antic.i.p.ate their anger and even to make sure of their favour through the accomplishment of certain rites, the souls may safely retake their accustomed place.[578] It is this same belief which gave birth to the personal totem. To protect themselves from sorcery, men thought it wise to hide their souls in the anonymous crowd of some species of animal or vegetable. But after these relations had once been established, each individual found himself closely united to the animal or plant where his own vital principle was believed to reside. Two beings so closely united were finally thought to be practically indistinguishable: men believed that each partic.i.p.ated in the nature of the other. When this belief had once been accepted, it facilitated and hastened the transformation of the personal totem into an hereditary, and consequently a collective, totem; for it seemed quite evident that this kins.h.i.+p of nature should be transmitted hereditarily from father to child.

We shall not stop to discuss these two explanations of the individual totem at length: they are ingenious fabrications of the mind, but they completely lack all positive proof. If we are going to reduce totemism to fetis.h.i.+sm, we must first establish that the latter is prior to the former; now, not merely is no fact brought forward to support this hypothesis, but it is even contradicted by everything that we know. The ill-determined group of rites going under the name of fetis.h.i.+sm seem to appear only among peoples who have already attained to a certain degree of civilization; but it is a species of cult unknown in Australia. It is true that some have described the churinga as a fetish;[579] but even supposing that this qualification were justified, it would not prove the priority which is postulated. Quite on the contrary, the churinga presupposes totemism, since it is essentially an instrument of the totemic cult and owes the virtues attributed to it to totemic beliefs alone.

As for the theory of Frazer, it presupposes a thoroughgoing idiocy on the part of the primitive which known facts do not allow us to attribute to him. He does have a logic, however strange this may at times appear; now unless he were completely deprived of it, he could never be guilty of the reasoning imputed to him. Nothing could be more natural than that he should believe it possible to a.s.sure the survival of his soul by hiding it in a secret and inaccessible place, as so many heroes of myths and legends are said to have done. But why should he think it safer in the body of an animal than in his own? Of course, if it were thus lost in s.p.a.ce, it might have a chance to escape the spells of a magician more readily, but at the same time it would be prepared for the blows of hunters. It is a strange way of sheltering it to place it in a material form exposing it to risks at every instant.[580] But above all, it is inconceivable that a whole people should allow themselves to be carried into such an aberration.[581] Finally, in a very large number of cases, the function of the individual totem is very different from that a.s.signed it by Frazer; before all else, it is a means of conferring extraordinary powers upon magicians, hunters or warriors.[582] As to the kins.h.i.+p of the man and the thing, with all the inconveniences it implies, it is accepted as a consequence of the rite; but it is not desired in its and for itself.

There is still less occasion for delaying over this controversy since it concerns no real problem. What we must know before everything else is whether or not the individual totem is really a primitive fact, from which the collective totem was derived; for, according to the reply given to this question, we must seek the home of the religious life in one or the other of two opposite directions.

Against the hypothesis of Hill Tout, Miss Fletcher, Boas and Frazer there is such an array of decisive facts that one is surprised that it has been so readily and so generally accepted.

In the first place, we know that a man frequently has the greatest interest not only in respecting, but also in making his companions respect the species serving him as personal totem; his own life is connected with it. Then if collective totemism were only a generalized form of individual totemism, it too should repose upon this same principle. Not only should the men of a clan abstain from killing and eating their totem-animal themselves, but they should also do all in their power to force this same abstention upon others. But as a matter of fact, far from imposing such a renunciation upon the whole tribe, each clan, by rites which we shall describe below, takes care that the plant or animal whose name it bears shall increase and prosper, so as to a.s.sure an abundant supply of food for the other clans. So we must at least admit that in becoming collective, individual totemism was transformed profoundly, and we must therefore account for this transformation.

In the second place, how is it possible to explain, from this point of view, the fact that except where totemism is in full decay, two clans of a single tribe always have different totems? It seems that nothing prevents two or several members of a single tribe, even when there is no kins.h.i.+p between them, from choosing their personal totem in the same animal species and pa.s.sing it on to their descendants. Does it not happen to-day that two distinct families have the same name? The carefully regulated way in which the totems and sub-totems are divided up, first between the two phratries and then among the various clans of the phratry, obviously presupposes a social agreement and a collective organization. This is as much as to say that totemism is something more than an individual practice spontaneously generalized.

Moreover, collective totemism cannot be deduced from individual totemism except by a misunderstanding of the differences separating the two. The one is acquired by the child at birth; it is a part of his civil status.

The other is acquired during the course of his life; it presupposes the accomplishment of a determined rite and a change of condition. Some seek to diminish this distance by inserting between the two, as a sort of middle term, the right of each possessor of a totem to transmit it to whomsoever he pleases. But wherever these transfers do take place, they are rare and relatively exceptional acts; they cannot be performed except by magicians or other personages invested with special powers;[583] in any case, they are possible only through ritual ceremonies which bring about the change. So it is necessary to explain how this prerogative of a few became the right of all; how that which at first implied a profound change in the religious and moral const.i.tution of the individual, was able to become an element of this const.i.tution; and finally, how a transmission which at first was the consequence of a rite was later believed to operate automatically from the nature of things and without the intervention of any human will.

In support of his interpretation, Hill Tout claims that certain myths give the totem of the clan an individual origin: they tell how the totemic emblem was acquired by some special individual, who then transmitted it to his descendants. But in the first place, it is to be remarked that these myths are all taken from the Indian tribes of North America, which are societies arrived at a rather high degree of culture.

How could a mythology so far removed from the origins of things aid in reconst.i.tuting the primitive form of an inst.i.tution with any degree of certainty? There are many chances for intermediate causes to have gravely disfigured the recollection which these people have been able to retain. Moreover, it is very easy to answer these myths with others, which seem much more primitive and whose signification is quite different. The totem is there represented as the very being from whom the clan is descended. So it must be that it const.i.tutes the substance of the clan; men have it within them from their birth; it is a part of their very flesh and blood, so far are they from having received it from without.[584] More than that, the very myths upon which Hill Tout relies contain an echo of this ancient conception. The founder who gave his name to the clan certainly had a human form; but he was a man who, after living among animals of a certain species, finally came to resemble them. This is undoubtedly because a time came when the mind was too cultivated to admit any longer, as it had formerly done, that men might have been born of animals; so the animal ancestor, now become inconceivable, is replaced by a human being; but the idea persists that this man had acquired certain characteristics of the animal either by imitation or by some other process. Thus even this late mythology bears the mark of a more remote epoch when the totem of the clan was never regarded as a sort of individual creation.

But this hypothesis does not merely raise grave logical difficulties; it is contradicted directly by the following facts.

If individual totemism were the initial fact, it should be more developed and apparent, the more primitive the societies are, and inversely, it should lose ground and disappear before the other among the more advanced peoples. Now it is the contrary which is true. The Australian tribes are far behind those of North America; yet Australia is the cla.s.sic land of collective totemism. _In the great majority of the tribes, it alone is found, while we do not know a single one where individual totemism alone is practised._[585] This latter is found in a characteristic form only in an infinitesimal number of tribes.[586] Even where it is met with it is generally in a rudimentary form. It is made up of individual and optional practices having no generality. Only magicians are acquainted with the art of creating mysterious relations.h.i.+ps with species of animals to which they are not related by nature. Ordinary people do not enjoy this privilege.[587] In America, on the contrary, the collective totem is in full decadence; in the societies of the North-west especially, its religious character is almost gone. Inversely, the individual totem plays a considerable role among these same peoples. A very great efficacy is attributed to it; it has become a real public inst.i.tution. This is because it is the sign of a higher civilization. This is undoubtedly the explanation of the inversion of these two forms of totemism, which Hill Tout believes he has observed among the Salish. If in those parts where collective totemism is the most fully developed the other form is almost lacking, it is not because the second has disappeared before the first, but rather, because the conditions necessary for its existence have not yet been fully realized.

But a fact which is still more conclusive is that individual totemism, far from having given birth to the totemism of the clan, presupposes this latter. It is within the frame of collective totemism that it is born and lives: it is an integral part of it. In fact, in those very societies where it is preponderating, the novices do not have the right of taking any animal as their individual totem; to each clan a certain definite number of species are a.s.signed, outside of which it may not choose. In return, those belonging to it thus are its exclusive property; members of other clans may not usurp them.[588] They are thought to have relations of close dependence upon the one serving as totem to the clan as a whole. There are even cases where it is quite possible to observe these relations: the individual aspect represents a part or a particular aspect of the collective totem.[589] Among the Wotjobaluk, each member of the clan considers the personal totems of his companions as being his own after a fas.h.i.+on;[590] so they are probably sub-totems. Now the sub-totem supposes the totem, as the species supposes the cla.s.s. Thus the first form of individual religion met with in history appears, not as the active principle of all public religion, but, on the contrary, as a simple aspect of this latter. The cult which the individual organizes for himself in his own inner conscience, far from being the germ of the collective cult, is only this latter adapted to the personal needs of the individual.

III

In a more recent study,[591] which the works of Spencer and Gillen suggested to him, Frazer has attempted to subst.i.tute a new explanation of totemism for the one he first proposed, and which we have just been discussing. It rests on the postulate that the totemism of the Arunta is the most primitive which we know; Frazer even goes so far as to say that it scarcely differs from the really and absolutely original type.[592]

The singular thing about it is that the totems are attached neither to persons nor to determined groups of persons, but to localities. In fact, each totem has its centre at some definite spot. It is there that the souls of the first ancestors, who founded the totemic group at the beginning of time, are believed to have their preferred residence. It is there that the sanctuary is located where the churinga are kept; there the cult is celebrated. It is also this geographical distribution of totems which determines the manner in which the clans are recruited. The child has neither the totem of his father nor that of his mother, but the one whose centre is at the spot where the mother believes that she felt the first symptoms of approaching maternity. For it is said that the Arunta is ignorant of the exact relation existing between generation and the s.e.xual act;[593] he thinks that every conception is due to a sort of mystic fecundation. According to him, it is due to the entrance of the soul of an ancestor into the body of a woman and its becoming the principle of a new life there. So at the moment when a woman feels the first tremblings of the child, she imagines that one of the souls whose princ.i.p.al residence is at the place where she happens to be, has just entered into her. As the child who is presently born is merely the reincarnation of this ancestor, he necessarily has the same totem; thus his totem is determined by the locality where he is believed to have been mysteriously conceived.

Now, it is this local totemism which represents the original form of totemism; at most, it is separated from this by a very short step. This is how Frazer explains its genesis.

At the exact moment when the woman realizes that she is pregnant, she must think that the spirit by which she feels herself possessed has come to her from the objects about her, and especially from one of those which attract her attention at the moment. So if she is engaged in plucking a plant, or watching an animal, she believes that the soul of this plant or animal has pa.s.sed into her. Among the things to which she will be particularly inclined to attribute her condition are, in the first place, the things she has just eaten. If she has recently eaten emu or yam, she will not doubt that an emu or yam has been born in her and is developing. Under these conditions, it is evident how the child, in his turn, will be considered a sort of yam or emu, how he regards himself as a relative of the plant or animal of the same species, how he has sympathy and regard for them, how he refuses to eat them, etc.[594]

From this moment, totemism exists in its essential traits: it is the native's theory of conception that gave rise to it, so Frazer calls this primitive totemism _conceptional_.

It is from this original type that all the other forms of totemism are derived. "When several women had, one after the other, felt the first premonitions of maternity at the same spot and under the same circ.u.mstances, the place would come to be regarded as haunted by spirits of a peculiar sort; and so the whole country might in time be dotted over with totem centres and distributed into totem districts."[595] This is how the local totemism of the Arunta originated. In order that the totems may subsequently be detached from their territorial base, it is sufficient to think that the ancestral souls, instead of remaining immutably fixed to a determined spot, are able to move freely over the surface of the territory and that in their voyages they follow the men and women of the same totem as themselves. In this way, a woman may be impregnated by her own totem or that of her husband, though residing in a different totemic district. According to whether it is believed that it is the ancestor of the husband or of the wife who thus follow the family about, seeking occasions to reincarnate themselves, the totem of the child will be that of his father or mother. In fact, it is in just this way that the Guanji and Umbaia on the one hand, and the Urabunna on the other, explain their systems of filiation.

But this theory, like that of Tylor, rests upon a begging of the question. If he is to imagine that human souls are the souls of animals or plants, one must believe beforehand that men take either from the animal or vegetable world whatever is most essential in them. Now this belief is one of those at the foundation of totemism. To state it as something evident is therefore to take for granted that which is to be explained.

Moreover, from this point of view, the religious character of the totem is entirely inexplicable, for the vague belief in an obscure kins.h.i.+p between the man and the animal is not enough to found a cult. This confusion of distinct kingdoms could never result in dividing the world into sacred and profane. It is true that, being consistent with himself, Frazer refuses to admit that totemism is a religion, under the pretext that he finds in it neither spiritual beings, nor prayers, nor invocations, nor offerings, etc. According to him, it is only a system of magic, by which he means a sort of crude and erroneous science, a first effort to discover the laws of things.[596] But we know how inexact this conception, both of magic and of religion, is. We have a religion as soon as the sacred is distinguished from the profane, and we have seen that totemism is a vast system of sacred things. If we are to explain it, we must therefore show how it happened that these things were stamped with this character.[597] But he does not even raise this problem.

But this system is completely overthrown by the fact that the postulate upon which it rests can no longer be sustained. The whole argument of Frazer supposes that the local totemism of the Arunta is the most primitive we know, and especially that it is clearly prior to hereditary totemism, either in the paternal or the maternal line. Now as soon as the facts contained in the first volume of Spencer and Gillen were at our disposal, we were able to conjecture that there had been a time in the history of the Arunta people when the totems, instead of being attached to localities, were transmitted hereditarily from mother to child.[598] This conjecture is definitely proved by the new facts discovered by Strehlow,[599] which only confirm the previous observations of Schulze.[600] In fact, both of these authors tell us that even now, in addition to his local totem, each Arunta has another which is completely independent of all geographical conditions, and which belongs to him as a birthright: it is his mother's. This second totem, just like the first, is considered a powerful friend and protector by the natives, which looks after their food, warns them of possible dangers, etc. They have the right of taking part in its cult.

When they are buried, the corpse is laid so that the face is turned towards the region of the maternal totemic centre. So after a fas.h.i.+on this centre is also that of the deceased. In fact it is given the name _tmara altjira_, which is translated: camp of the totem which is a.s.sociated with me. So it is certain that among the Arunta, hereditary totemism in the uterine line is not later than local totemism, but, on the contrary, must have preceded it. For to-day, the maternal totem has only an accessory and supplementary role; it is a second totem, which explains how it was able to escape observation as attentive and careful as that of Spencer and Gillen. But in order that it should be able to retain this secondary place, being employed along with the local totem, there must have been a time when it held the primary place in the religious life. It is, in part, a fallen totem, but one recalling an epoch when the totemic organization of the Arunta was very different from what it is to-day. So the whole superstructure of Frazer's system is undermined at its foundation.[601]

IV