they would make a practice of capturing women from other tribes; and, indeed, it is well known that marriage by capture has been assumed to be at the base of exogamy by both Lubbock and Spencer.  But the importance which has been attached to this form of marriage in the literature of sociology is due to the fact that these eminent writers have constructed theories on the assumption that marriage by capture was widespread and important, more than to anything else.  For, to say nothing of the fact that the theories of both these writers are too weak to stand even if capture were found to be very prevalent, the evidence from Australia shows that capture was comparatively little practiced there, although that country affords most of the examples referred to by writers on this subject.  Spencer and Gillen say in this connection: 

The method of capture which has so frequently been described as characteristic of Australian tribes, is the very rarest way in which the Central Australian secures a wife.  It does not often happen that a man forcibly takes a woman from someone else within his own group, but it does sometimes happen, and especially when the man from whom the woman is taken has not shown his respect for his actual or tribal Ikuntera (father-in-law) by cutting himself on the occasion of the death of one or the other of the latter’s relations.  In this case the aggressor will be aided by the members of his local group, but in other cases of capture he will have to fight for himself.  At times, however, a woman may be captured from another group, though this again is of rare occurrence, and is usually associated with an avenging party, the women captured by which, who are almost sure to be the wives of men killed, are allotted to certain members of the avenging party.[227]

Curr reports to the same effect: 

On rare occasions a wife is captured from a neighboring tribe and carried off....  At present, as the stealing of a woman from a neighboring tribe would involve the whole tribe in war for his sole benefit, and as the possession of the woman would lead to constant attacks, tribes set themselves generally against the practice.[228]

It is, of course, not to be denied that the sexual impulse of the male was sometimes strong enough to lead him to seize a woman wherever he found her, if he could not get a wife otherwise, but there is no evidence that capture ever formed a regular or important means of getting wives.[229]

On the contrary, the evidence points to the view that as soon as for any reason men ceased to marry with the women of their own blood and went outside of their immediate families for women, they ordinarily secured them in a social, not a hostile, way, and from a different branch of their own group, not, as a rule, from a strange group.  In fact, the regular means of securing a wife other than a woman of one’s own family seems to have been to exchange a woman of one’s family for a woman of a different family.