Now, the forms of totemic exogamy unmistakably constitute a developmental series. In the simplest arrangement, there are no restrictions whatever upon marriage between members of one clan and those of another with which marriage relations exist. Such exogamy, however, is relatively rare in Australia, the land in which the developmental forms of exogamy are chiefly to be found. It seems to be limited to tribes that have merely a dual organization, in which event the clan coincides with one-half the tribe. Even in such cases we find transitions to the next form of exogamy. In this second system, exogamy is restricted to particular totems of the two clans of one and the same tribal division; and, just as in the first case, the children are, as a rule, born directly into the totem group of the mother, or, less commonly, into that of the father. Following this exogamy with direct maternal or paternal descent and undeniably proceeding out of it, we finally have, as the third main form, exogamy with indirect maternal or paternal descent. In this form of exogamy, as in the preceding ones, the children belong to the totem of the mother or to that of the father so far as birth is concerned; as respects their exogamous totem relation, however, they pass over into another totem of the same clan. Thus, birth-totem and marriage-totem are here distinct, and every member of a group belongs to two totems that differ in significance. Now, in the case of a marriage by capture in which the individuals belong to different clans, the question of the totem does not enter. When, therefore, this mode of marriage remains undisturbed by further conditions, we have exogamy of the first form. When a suitor seeks to win the favour of the clan by means of a gift presented to the parents or the kin, marriage by capture passes over directly and without further change into the simple marriage by purchase. The two more exclusive forms of exogamy, on the other hand, are obviously connected with the rise of totemism; they are the result both of the clan divisions which follow from tribal partition and of the accompanying separation into totem groups. The question, therefore, concerning the development of these forms of exogamy, dependent as they are both upon clan divisions and upon totem groups, is essentially bound up with the question concerning the temporal relation of the two important phenomena last mentioned. An unambiguous answer to the latter question, however, may be gathered precisely by a study of Australian conditions, at least so far as the development in these regions is concerned. If we recall our previous schema (p. 141), representing the tribal organization of the Kamilaroi, and here, as there, designate the totemic groups (emu, kangaroo, opossum, etc.) comprised within the clan by m n o p..., it is apparent that the totems must be at least as old as the division into the two tribal halves. Unless this were the case, we could not explain the fact that, with very minor exceptions, precisely the same totems exist in the two tribal divisions. The condition might be represented thus:—
It is also evident, however, that the totems could not have influenced this first division, otherwise their members would not have separated and passed over into the two tribal divisions, as they did in almost every case. Remembering that the totemic groups are also cult associations, we might express the matter thus: At the time of the first tribal division, the cult groups were not yet strong enough to offer resistance to the separation of the tribal divisions, or to determine the mode of division; therefore, members of totem m, for example, went here or there according as other external conditions determined. Conditions were quite different at the time of the second division, when the tribal half I separated into clans A and B, and II into C and D, according to the schema:—
These clans, as we see, separated strictly according to totems. The bond of cult association had now become so strong that all members of a particular totem regularly affiliated themselves with the same clan, though the grouping of the totem divisions within the clans of the two tribal halves proceeded along absolutely independent lines, as may be concluded from the fact that the totems composing the clans within the two tribal divisions are grouped differently. The formation of such cult or totem groups, thus, may already have begun in the primitive horde. At that time, however, these cult groups were probably loosely knit, so that when the horde split up, its members separated, each of the two tribal divisions, generally speaking, including individuals of all the various tribes. Not so in the case of those divisions of the tribe which originated later, after mankind had advanced further beyond the condition of the horde. By this time the totem unions must have become stronger, so that the members of a cult group no longer separated but, together with other similar groups, formed a clan. When the growth of the tribe, together with the conditions of food supply and the density of population, led to a separation of the tribe, certain totem groups invariably joined one division and others, the other, but the more firmly organized groups remained intact.
A further phenomenon of great importance for the development of exogamous marriage laws must here be mentioned—one that occurs throughout the entire realm of totemic culture but is particularly prominent among the Australian totem groups. This phenomenon consists in totem friendships. Certain totem groups regard themselves as in particularly close relations with certain other groups. Friendships similar to these, in a general way, are to be found even in connection with the highest forms of political organization. For modern States themselves enter into political alliances or friendships, and these, as is well known, are subject to change. Such alliances occur from the beginnings of totemism on up to the advanced plane of modern international culture. Though these affiliations eventually come to be determined primarily by the commercial relations of peoples, the determining factors at the outset were faith and cult. In both cases, however, the friendships are not of a personal nature, but are relations based on common interests. This common interest may consist, for example, in the fact that, as has been observed among some of the Australian totem alliances, the member of a totem may slay the totem animal in the hunt, but may not eat of it, though the member of the friendly totem may do so. Thus, the interest in cult becomes also a means for the satisfaction of wants, as well as a bond that unites more closely the particular totem groups.
These facts help to explain how the unlimited exogamy which first arises from marriage by capture comes to pass over into a 'limited exogamy,' as it does immediately upon the appearance of conditions that regulate the forceful capture and substitute for it the friendly exchange of women. These factors, however, always come into play whenever the intercourse between tribal members becomes closer, and particularly when the struggle with strange tribes keeps in check the strife between individuals of the same tribal association. In such cases, exchange, or, in later development, purchase, proves the means of putting an end to force. Thus, blood revenge, which persists into far later times, is displaced by the wergild which the murderer pays to the kin of his victim. This transition is precisely the same, in its own field, as that which occurs in the institution of marriage, for in the former case also the strife involves members of the same tribe. The passion, however, which causes the murder and which creates the demand for vengeance, sometimes prevented the introduction of peaceful means of settlement. In the case of marriage by capture, however, a marriage relationship, unrestricted and friendly in character, was doubtless first developed between the two clans, particularly wherever tribal division and clan were identical. And though marriage by capture was for a time still occasionally practised—since all changes of this sort are gradual—such marriages, nevertheless, more and more assumed a playful character. The actual capture everywhere finally gave way to exchange and later to the gift. When, however, the totem groups, and with them the cult associations that established a bond between clan and clan, gained the ascendancy, the totem groups naturally displaced the clan in respect to marriage arrangements; those totems who maintained close cult relations with one another, entered also into a marriage relationship. Thus, exogamy became limited; the members of a totem of clan A married only into the friendly totem of clan B, and this usage became an established norm whose violation might result in the death of the guilty person, unless he escaped this fate by flight. This transition of exogamy from clan to totem group, and from the unlimited to the limited form, came only gradually. This is clearly shown by the conditions among the Dieri. Certain of their totems have already entered upon the stage of limited marriage relationship, whereas others have not advanced beyond unlimited exogamy.
But even after the development had reached its final form and limited totemic exogamy was completely established, further changes ensued. For the basis of such exogamy, we may conjecture, is the fact that certain totem groups of associated clans enjoy particularly close relations with one another. Even on these primitive levels, however, the friendships of such groups are not absolutely permanent any more than are the political friendships of modern civilized states, though their degree of permanence is probably greater than that of the latter. Migrations, changes in hunting-grounds, and other conditions, were doubtless operative also in totemic culture, loosening the bonds between friendly totems and cementing others in their stead. This led to changes in the exogamous relations of totem groups. Instead of groups n and r of clans A and B, n and q might then come to have exogamous connections (see diagram III on p. 148). But the severance of the old connection did not immediately obliterate the tradition of the former relationship. The influence of the latter would naturally continue to be felt, not in connection with acts of a transitory nature, such as wooing and marriage, but in matters permanent in character and thus affecting the traditional organization of the tribe. Such a permanent relation, however, is totem affiliation. This explains how it happens that, even after the old totem connection gave way to the new, it nevertheless continued to exercise a claim on the totem membership of the children born under the new marriage conditions; hence also the recognition of the claim on the part of custom. In one respect, indeed, such recognition was impossible. More firmly established than any form of exogamy was the law that children belonged to the mother, or, in the case of paternal descent, to the father. This law could not be violated. Hence exogamous and parental tribal membership became differentiated. The latter ordained that children in every case belong to the totem of the parent who determines descent; the tradition of the former decreed that children belong, not to the parental totem, but to some other totem of the same clan. Such a condition of dual totem membership might, of course, arise from a great variety of conditions, just as may the similarly overlapping social relations within our own modern culture—such, for example, as the military and the so-called civil station of a man. The customary designation of the first two forms of limited exogamy as exogamy with direct maternal descent, and of the third as exogamy with indirect maternal descent, is plainly inappropriate and may easily give rise to misunderstandings. For it may suggest that the maternal totem disposes of its rights in respect to marriage arrangements to another totem group, and that eventually this even occurs in accordance with a definite agreement. But this is certainly not the case. For maternal descent or, speaking more generally, the fact that children belong to the parents, obtains invariably. It would be preferable, therefore, simply to distinguish the parental totem connection from the traditional exogamous connection, or one system in which the exogamous and the parental connections coincide, from a second in which they differ.