Primitively, doubtless, as between the classes, the genetic idea as regards sexual matters was (as still with savages in questions of food) to favour the seniors and defend their rights in defining each one's status. But actually, with the decay of incest, it would become what it is as among lower races, where nothing is more remarkable than the strict interdict upon any union between members of different generations.[9]

It is evident that hence complications might arise perplexing to the savage mind. For instance, we may expect to find cases where the niece is an adult, whilst the aunt is still an infant, and yet marriage between the former and the son of the latter is obligatory, as they are cousins of the same generation. Here, probably, we have a clue to one of the most bizarre facts in anthropology, where the universal rule as to sexual connection between generations seems to be wantonly disobeyed, although in reality the reverse may be seen to be the case on examination. It is recorded of the Keddies of Southern India that a very singular custom exists among them, a young woman of sixteen or twenty years of age may be married to a boy of five or six years. She, however, lives with some other adult male (perhaps a maternal uncle or cousin), but is also allowed to form a connection with the father's relatives, occasionally it may be the boy's father himself, i.e. the woman's father-in-law! Should there be children from these liaisons, they are fathered on the boy husband. When the boy grows, the wife is either old or past child-bearing, when he in turn takes some other boy's wife in a manner precisely similar to that in his own case and procreates children for the boy husband.

By the classificatory system, as each in fact is a member of the same generation, they are born husbands and wives. The enforced virginity of the wife, implied under such conditions, entailed a celibacy incompatible with all lower ideas. It is easy to imagine the compromise between his conscience and his desires which a savage would make in such a case when favoured (or forced) by circumstances of environment, for it is unknown elsewhere. The infant nephew goes through the ceremony of marriage, which, by a fiction, being thus legally consummated, the wife is left free to follow her desires. These, however, are by no means allowed to run riot. They are regulated in a fashion of which, although the peculiarity is noted by the authors of the extract, the full significance can only be appreciated in connection with our hypothesis. She formed indeed connections outside of her husband, but solely with those of the legally eligible totem. As I believe the Keddies have male descent, these would be sons of the father's sister, or sons of the mother's brother, or again with the latter himself, who was her father-in-law, whereas union with the sons of the father's brothers, or of the mother's sisters, as being of the same totem, would not take place—and this we find to be the actual fact, as evidence proves.

But still other complications will be found to arise as the effect of the original concept of the classificatory system when brought face to face with new and advanced social order, which will have closer relation to our present argument. The distinctive feature in the economy of the primitive group in its relation to all other groups was mutual hostility. The instinctive distrust of strangers would be accentuated by the habitual hostile capture of females, for such groups, except in the case of the incest between father and daughter, were yet purely exogamous. But such mutual hostility implies isolation of each community. Thus all law evolved, as we have said, would be purely with a view to regulation of the internal economy of a single consanguine group alone. Now in such a group, the division into generations of old male, male, and young male implied (although not as yet understood as between generations) the relationship of parents and children. Each generation is either child or parent to the other. As marriage is communal,[10] all the fathers in one generation are fathers to all the children in the next indiscriminately, and conversely these children recognise as fathers all the males of the senior generation. It follows that the relationship of all the members of a generation is purely fraternal, all are brothers and sisters to each other, and in this consanguine family they were really either actually so, or at least half brothers and half sisters.

Between these the primal law of celibacy between brother and sister as such embraced the whole generation. Now as long as the family was thus simply constituted, no friction would arise. The brothers, in common, captured and married in common some outside female,[11] and their children constituted solely the next generation. The sisters were either stolen or emigrated to other groups; but we have seen that a moment would come when this process ceased to be universal. The sister came to remain in her own group, and she was joined by some outside suitor; with the advent of their children, who are cousins to the others, would arise dire perplexities, in view of the old law.

We may now begin to see more distinctly, in the fact of the presence of the cousins, the resolution of the problem as to how a sister's son came to be also a brother's, and we will find that Mr. Morgan was not the first to be baffled by the problem. It was too intricate for primitive man at any rate. When first presented to him, we may surmise that he, in fact, refused to recognise it as a problem at all. Since the beginning of things in the group, as constituted by all tradition, the children of one generation were children of another simply, and nothing more. That as a result of the presence of the outside male, some intricate process of scission had occurred, and things were not as they seemed, was an idea far too abstract to be readily seized. All in a generation had been ever, to early man, brother and sister, and brother and sister they should continue.

We have seen in a past chapter that it was actually to the interest of senior male group-members, while incest reigned, that this condition of things should endure. It put at their sole disposal the daughters of their brothers-in-law, and in the primal law placed a ban on sexual intercourse between all the younger male and female members, as constituting them brothers and sisters. As a factor in this case, however, the effect of incest was more or less temporary. The real agent in the tardy or non-recognition of the cousinship thus created, was the conservative force of old habit and tradition. We must remember that, in so early a group, personal descent as such was in no way recognised. Mere local contiguity alone constituted the sense of relationship, exogamy for instance took the form of local exogamy, for as all within a locality were (locally) relations, so all outside were, as strangers, free in marriage. While then so strong a sense of the value of contiguity continued, and was in practice, the evolution of an idea of non-relationship of two individuals with a common habitat would be too complex. Again, a recognition in fact implies a vast modification of the whole organisation of the group, which thus contains in cousins the elements of marriage within itself. But this is the latest and highest type of group and constitutes the tribe. We can understand that such a step was not taken at once by early man. Even when recognised we know that definition lags behind the event.

Thus in such a case as cited, and at the stage we are studying, if we find two cousins in presence, who are yet unrecognised as cousins, then, if nomenclature has taken place, we should find exactly the terms employed in the Malayan table which misled Mr. Morgan. A sister's son would be termed the brother's son, simply because the individual was as yet ignored, although existent, as a cousin, as members of the same generation they were brother and sister. Classes by generations alone were recognised.

Now as regards the validity of our assumption that relativity in age served as a means to determine privilege as to wedlock, proof can be furnished by certain nomenclatory features, as between members of a class or generation, to be found in the Malayan table in Ancient Society and elsewhere. This will afford, incidentally, strong negative proof of our theory as to non-union between brother and sister. It will also incidentally furnish the strongest negative evidence that, so far from brother and sister living in incest, as Morgan holds, brother and sister were regarded as quite apart in the sense of any sexual relation between them. It will be seen that there is a profound distinction made in address between inter-marriageable people and those between whom celibacy is enjoined.