The Growth of the Marriage Relation. 77 



Islanders, in which several brothers had their wives in 

 common, or several sisters had their husbands in common. 

 The Australian system, as now seen in practice, gives 

 every man and woman a wife or husband, and also one or 

 more " accessory spouses," each of whom may be husband 

 or wife to some other person. In all cases the restriction 

 on the intermarriage of kinsfolk is strictly enforced. 



The Australian system appears to me to furnish the key 

 to all other phases of the marriage relation. These depend 

 on the conditions under which they have been developed, 

 but they stand in a definite relation to group-marriage. 

 Thus, First, the group idea may be abolished on the male 

 side, in which case we have a group of women married to 

 a single man, (a) where the wives are kinswomen, as in the 

 sister polygyny of the American aborigines, and (b) where 

 the wives are strangers in blood, giving the harem or com- 

 mon polygyny of the East. Secondly, the group idea may be 

 abolished on the female side, in which case we have a 

 group of men married to a single female, (a) where the 

 husbands are kinsmen, as in Tibetan polyandry, and (b) 

 where the husbands are strangers in blood, as in Nair 

 polyandry, if this can be accepted as a true phase of the 

 marriage relation. Thirdly, with the group idea abolished 

 on both sides, we have the intermarriage of two individuals, 

 giving (a) individual marriage, with power for the husband, 

 under special circumstances, to marry a second wife, or to 

 take accessory wives the monandry of the Turanian 

 peoples and the so-called pairing-family arrangement of the 

 American aborigines ; and (b) the individual marriage in 

 which a man has but one wife, and a woman but one hus- 

 band, unless the marriage relation is terminated by divorce 

 or death, being the monogamy of modern civilization. 



It is perhaps impossible to determine whether the primi- 

 tive groui>marriage was developed first in the direction of 

 polygyny or polyandry, but as there is reason to believe 

 that men at first left their own family groups to live in the 

 groups to which their wives belonged, a practice which is 

 usually associated with kinship through females, and as 

 this is found to exist among the more primitive human 

 races, many of whom also practice polygyny, we may 

 assume that polyandry was of later development. Prob- 

 ably the marriage of a man to several sisters originated in a 

 state of society where families were comparatively isolated, 



