Prominence was given to the association of totemism with exogamy, found throughout the world with few exceptions. Where totemism is not associated with exogamy it is not difficult to suppose that exogamy has been lost. With considerably less confidence it may be supposed that tribes, containing exogamous groups, have lost their totemism. Finally there are some peoples, e.g., the Andaman islanders, and the Eskimo, at a primitive level of culture, Who have neither exogamous groups nor totemism. From this distribution of totemism and exogamy it follows that any theory of totemism must either be a "psychological" theory, explaining both totemism and exogamy, or a "historical" theory, which regards totemism, wherever it occurs, as the result of a spread of cultural influence from some one centre where the totemic complex originated. By a "psychological" explanation is meant an explanation in terms of observed facts about the be haviour of groups, or by means of generalizations about the be haviour of groups deduced from known psychological principles. If, antecedently to the development of a given institution, it could be stated on psychological grounds that an institution of that sort would probably develop, then such an institution can be said to be explained psychologically. Early theories of totemism have not explained in this sense, for totemism and exogamy have only been shown to be probable, independently of the development of the one or the other, from which it follows that the distribu tion of totemism and exogamy, assumed above, would not occur; e.g., if, owing to the nature of man and his attainment of some very simple form of society, it were probable that he should develop totemic ideas and practices, and also that he should de velop exogamous groups, but for reasons other than those which lead to his developing totemism, then it would be indeed sur prising that most peoples who have the one institution have the other, while those who, although in a primitive state of society, do not have the one do not have the other. The difficulty can be got over by supposing that totemism and exogamy are necessarily developed by man in the course of his evolution; but it is difficult in this case to offer a plausible explanation of the disappearance of totemism and exogamy among some very primitive peoples, and its retention by others at a more advanced stage of civilization.
Two theorists may be mentioned for their attempts to provide a psychological explanation of both totemism and exogamy; viz., Durkheim and Freud. For Durkheim the totem results from the action of the group on the individual. The group is the nearest and most powerful coercive force to which the individual is subject, particularly when the group is in violent activity and the individual is conscious of this overpowering influence, when he loses himself completely in the group. Symbolic expression alone can be given to such a force; hence the totem is a sacred symbol, symbolizing that very real force, the clan. Hence totem ic symbols are sacred rather than the totem-species itself (and this, it must be admitted, is true, at least for Australia).
But why, it may be asked, should there be many clans and totems, and not one totemic group? Here Durkheim becomes obscure. If there are already several groups, then only in so far as the effective unity of each group is felt should there be more than one totemic group within the tribe. Passing over this diffi culty, we have still to enquire why the totemic group is exoga mous: the answer is not very convincing. It would take too long to follow the various steps which give unity and intelligibility to the scheme, but it is the tabu on the blood of the clan which is ultimately responsible for exogamy.
Freud attempted to give a single explanation of totemism and exogamy, and he appeared to have indicated the direction in which we are to seek an explanation, whatever the adequacy of his presentation.
Prior to the development of culture, man is supposed to have lived in "cyclopean" families, containing a principal male who monopolized the women, including his own daughters, and pre vented his sons from having intercourse within the family. What more natural than that the sons should combine and kill the father? But the attitude of the sons is ambivalent, combining both affection and hostility. The murder committed, the sons deny themselves the fruits of the deed, owing to the operation of a psychological process familiar to psycho-analysts, and the group becomes exogamous. The affective ambivalence towards the father has also provided the beginnings of totemism. By substituting an animal for the father, a process of symbolism which psycho analysts consider fundamental in human nature, the animal be comes a vehicle for that affective attitude towards the father which is for the moment repressed. With the killing of the father by the momentary release of this repressed Oedipus com plex, both totemism and exogamy are born. The group with one father, and therefore one totem, is exogamous, and the totem becomes the object of observances, which are really the ex pression of the Oedipus complex, which has now become more completely repressed by means of a social mechanism. The diffi culties in the way of this theory are considerable, and have been critically analysed by Profs. Kroeber and Malinowski. Doctor Freud's theory has been elaborated and modified in a compre hensive study of Australian totemism on psycho-analytic lines by Dr. Geza Roheim. While the search for a "psychological" ex planation of totemism does not reward us with any convincing theory of its origin, an "historical" explanation is equally elusive, though it has been maintained by one school that the distribution of totemism can only be explained on the assumption of diffusion from a single centre, a conclusion which, as we have already seen, follows if no psychological explanation of totemism com bined with exogamy can be found. No evidence for the existence of exogamy, and only the most slender evidence for totemism is provided in the Egyptian centre from which the totemic com plex is supposed to have spread; in fact, on this theory totemic exogamy is derivative in the various parts of the world to which totemism spread from a kind of dual organization with marriage injunctions. The only "historical" explanation of totemism which has been elaborated is, therefore, no more, if no less, convincing than the numerous "psychological" explanations which have been offered.
Dissatisfied with this impasse that the search for explanations leads to, both in the case of totemism and in that of many other primitive institutions, many anthropologists have renounced the search for psychological or historical explanations, and confine themselves to the search for the functions of institutions. Possibly this is the only kind of explanation of totemism within our grasp, and, although there is reason to suppose that totemism without function (occurring as a survival merely), occurs in some places, it must be admitted that, until totemism is understood in this functional sense in those cases where it appears to be a significant part of the culture of the group, it is premature to attempt to account for its origin.