It is not well to draw too sharp a distinction between modern man and, more especially, the continental European and primitive man, when it comes to the question of marriage and the gratification of the sex impulse. There is no doubt that the two categories are usually distinct in savage society. Marriage is often based purely upon economic grounds. Man needs someone to work for him, to cook, and to labor in the field. But how different is this from the kind of union often found in peasant communities in Europe and country districts in America? We often fail to realize that the AngloSaxon family ideal is late in putting in an appearance and is nearly unique.

In the consideration of marriage as an institution, the discussion will be limited to the question of legal wedlock, leaving aside the subject of sex. As has already been observed, marriage and the gratification of the sex impulse may belong to two distinct categories, not only among savages but also among many peoples who do not come under this classification. From the point of view of sociology, the nature of sexual unions is of little consequence until they are recognized by custom or law.

Westermarck's latest definition of marriage is "A relation of one or more men to one or more women which is recognized by custom or law and involves certain rights and duties both in the case of the parties entering the union and in the case of the children born of it." It is needless to add that the rights and duties involved therein vary widely. It should be noted also that this definition does not say anything about the exclusive possession of one or more men by one woman, or of one or more women by one man. We shall see later that

certain forms of marriage in some parts of Australia, for example, are characterized by what has often been called group-marriage, or "sexual communism."

The historical approach brings out clearly the various theories that have been held on the origin of marriage and of the domestic group.

CLASSICAL IDEAS OF THE FAMILY

Among the first theorists regarding the family were Plato and Aristotle, both of whom use for illustrations of an early family group the Cyclopes of Homer, who "have neither assemblies for consultation nor themistes, but everyone exercises jurisdiction over his wives and his children." This was the patriarchal theory made more precise by Filmer's "Patriarchia," published in London in 1680. This idea of a patriarchate is, however, usually associated with Sir Henry Maine, formerly Regius Professor of Civil Law at Cambridge. In his Ancient Law, published in 1861, he considered the primordial cell of social development to be the patriarchal family. It is difficult to say "what society of men had not been originally based upon the paternal family." He found it among the Romans, Greeks, Hindus, Celts, and the Teutonic and Slavonic peoples. Maine really knew intimately only the Roman family, where the pater familias was the head of the household and had the power of life and death over his wife, children, and slaves. There is a considerable difficulty in determining how large a part the patriarchal family has played in the early histories of the other peoples enumerated above. Maine certainly showed a "lofty contempt" of the large mass of data upon savage peoples. The patriarchal family is by no means a simple group, and it is certainly

not the "primordial cell" out of which the family has grown.

EVOLUTIONARY IDEAS OF THE FAMILY

Bachofen's Das Mutterrecht appeared in the same year as Maine's book. This was a general attack upon the theory of the patriarchate as being the first form of the family. Maine's theory was first definitely challenged by Spencer. Post, McLennan, Morgan, Lubbock, and a host of other writers whom we may call the Evolutionary School also disagreed with Main. Although often differing among themselves in detail, these writers all believed in a definite series of steps in the evolution of the family, postulating promiscuity as the original stage of the sexual relations of the savage.

Dismissing for the moment the arguments advanced for and against a universal beginning in promiscuity, let us investigate the other main points of the Evolutionary School. The matriarchate would necessarily follow a promiscuous start, as fatherhood would be unknown, and the mother would naturally be the head of the family and society would be dominated by her. But a true matriarchate or woman-rule has never been found in any society. Female descent, however, is by no means uncommon, and the distinction should always be made between mother-rule, or the matriarchate, and mother-descent, or the matrilineal family. McLennan believed that female infanticide was a common, perhaps a general custom. The natural consequence of this was few women-hence the marriage of one woman with several men, polyandry, which is the earliest type of the family, according to this theory, which rests on marriage. Multiple husbands in time give way to mul

tiple wives, commonly called polygamy, but more properly termed polygyny. The couvade marks the transition between the two types, where emphasis is laid upon the father. With polygyny the patriarchal family follows as a matter of course, the male head ruling over his numerous wives and children. Finally monogamy is placed at the end of the series.

MODERN IDEAS OF THE FAMILY

The third stage in the history of the discussion of the evolution of the family and marriage centers around the names of Starcke and Westermarck, although others might well be added to the list of those who are critical of the Evolutionary School.1 With the possible exception of Rivers, practically all modern investigators of this question seem to agree that promiscuity never was a universal stage in the history of the sexual life of the savage. Rivers writes, "We have clear evidence that existing varieties of mankind practise sexual communism, and man must therefore have tendencies in that direction." But he adds, "We need more evidence before we should make up our minds concerning the existence of group-marriage as a regular feature of the history of human society. Even if its existence in the case of the clan can be proved, or if it can be shown to be, or to have been, a widespread practice, it need not follow that it has been universal among mankind and has formed a constant feature of the evolution of human society." 2

There are no definite steps starting with promiscuity and leading upward to monogamy at the apex of the pyramid. Some of the peoples with most rudimentary cultures, such as the Bushmen, the Andamanese, and the Veddahs are strictly monogamous. The pairing of

one man and one woman has a great "survival value," as it affords to the offspring the care of both parents.

The arguments seeming to favor a universal stage of promiscuity "flow from two sources." The first is composed of books of ancient authors, the tales of travellers, and the accounts of missionaries. From the time of Herodotus and Strabo down to the present, many writers have attempted to describe the customs of savages. The traveller's pen has been declared to be a "thinking organ." There is perhaps no phase of early life where more confusion and misinterpretation have followed from this "organ" in the hands of the traveller than in regard to sexual relations and marriage. Unlimited promiscuity has been reported from all parts of the world, and nowhere is it really found. The reports of missionaries often fail to interpret correctly marital relations differing from those of civilized man. There are important exceptions to this rule, however, and we have many excellent and accurate accounts from ministers of the Gospel regarding savage customs.

The second class of arguments which seem to favor promiscuity falls into the realm of folk-lore-survivals -comparable to the rudimentary organs of the body. These no longer function, but their mere presence shows that at some time in the past they had an important part to play. Much of value may often be learned from the study of certain customs which seem to be emblematic of the past. Marett likens this study of survivals in folk-lore to the process of casting out of the drawingroom the unfashionable bit of furniture and placing it in the man's room, in the children's playroom, or even in the attic. These things were formerly admired and useful; now they are scorned. There is always a danger

3