Family, from Latin familia, which meant, primarily, the holding in slaves, and secondarily, the whole domestic property, of a paterfamilias; but which, as commonly used, denoted the body of people within a gens (a number of such bodies making up the gens) who traced their descent through males to a common ancestor. The family, in civilised countries, is the little group consisting of a man (its recognised head), his wife, and their children, which forms the inmost circle of relationship; and is connected more or less with other similar groups by relationship—relationship being reckoned by degrees, and counted through both father and mother, and determining rights of inheritance ab intestato and the law of incest. It was the universally received opinion until lately that this group existed substantially in the form known to us, with the father (sometimes with more than one wife) at its head, from the beginning of society; and that it was the germ from which all societies had been developed. It was thought there could be nothing more according to nature, and that there could have been nothing more primitive. Given such a family (so the reasoning went on), as the children and their descendants married, a number of similar groups would be formed round it, separate from one another, but all subject to their patriarch, in whose family they would be as long as he lived. They would probably separate from each other at his death, and expand and multiply each by itself; but by-and-by the family groups thus arising would find it convenient to go on living together, and thereafter they would become a set of separate tribes, many of which would be neighbours, and which might form in time the population of a district. Then the remembrance of their origin remaining, it would be easy for them to act together for common purposes; and this point arrived at, the descendants of one man would be well in train for constituting a people or nation. The history of Israel was thought to favour this account of the growth of tribes and states out of the family (known as the Patriarchal Theory); for it made each of the twelve tribes consist of descendants of a son or grandson of the patriarch Jacob, while the union of the tribes constituted the nation. In Genesis, too, the population of the world is represented as composed of tribes and nations descended from Noah's three sons, Shem, Ham, and Japheth. A modified form of this theory, put forward by Sir H. S. Maine, in his work on Ancient Law (1861), is still perhaps (at any rate, wherever English is spoken) the most popular account of the origin of societies.
Maine assumed that the family of ancient Rome gives the true suggestion of what the primitive family was; and this he accordingly described as having consisted of a paterfamilias, who had patria potestas—i.e. unlimited power over his household, with his wife, his children, persons adopted by him, and slaves. The primitive relationship was in his view merely the bond made by common subjection to the paterfamilias; and as he assumed that women when they married would become subject to a new paterfamilias—that there would be no marriages between persons of the same family—he deduced that a woman's descendants would be out of the relationship of the family of her birth at first, and that thus relationship would afterwards come (he did not show how) to be traced through males only—the relationship called at Rome agnation—and that the clans, tribes, and societies into which the family gradually developed would everywhere be agnatic. But it is certain that early societies have not been all agnatic; and, indeed, the Roman type of family, with patria potestas and agnation, seems not to occur in any system of old law except the Roman. Maine's primitive family, too, is too complex to belong to the beginning of society; and, the social tendency of men considered, his primitive father who acknowledged obligations to nobody is hardly to be reconciled with human nature. His theory thus appears to be liable to objections special to itself. There are others which apply to the patriarchal theory in both forms; and two of these should be mentioned.
First, there are numerous societies (not indeed of the most important) and various forms of the family of which the patriarchal theory does not even attempt to give any account. And next, in the societies upon contemplation of which it was founded, a most serious difficulty for it is presented by the tribes, which consist of several clans, each clan considered separate in blood from all the others. The patriarchal theory of course involves that the clans in a tribe are all of the same blood. Maine suggested that the clans in those cases united by the help of some fiction analogous to adoption; but the suggestion is too vague to be useful.
One form of the family which the above-described theories neglected, but which has prevailed very largely, is especially interesting because of the disclosure of it which is given us in the Book of Genesis. In beenah marriage (the name is taken from Ceylon) the man goes to live with his wife's family, usually paying for his footing in it by service; he is in general an unimportant person in the family; and the children are not his—they belong to the family and kindred of his wife. He is lost to his family and kindred so long as he remains a beenah husband; he may even have (as in New Zealand) to fight against his own relations. Beenah marriage, as we know it, is commonly practised concurrently with marriage by purchase, in which the husband takes away the wife and becomes entitled to her issue; but a few tribes are known which make no other marriages. Now Jacob made a beenah marriage into the family of Laban; and Genesis, xxiv. 1–8, shows that there was much more than a possibility that Isaac, as a condition of marrying into his father's kindred, might have had to do the same. It need scarcely be pointed out that Samson's marriage with a Canaanite woman was a marriage of this type. Moreover, the first reference to marriage in the Scriptures (Genesis, ii. 24) is scarcely intelligible except as a reference to beenah marriage; it cannot possibly refer to a marriage of the Roman type, and scarcely to any marriage under which the bride would go to live with her husband's kinsfolk. In Jacob's case, Laban claimed the children. Jacob, his wives concurring, had stolen away with them—and it is clear that they were not his. Isaac had married into the same family, and his children were his, because he had been allowed to purchase the mother; but Jacob had not purchased—he had merely won for himself a place in Laban's family as the husband of Laban's daughters, and to that family both he (while he chose to remain in it) and his children thus belonged. The marriage of Isaac shows that beenah marriage was not exclusively practised by the kindred of Laban—that, when they saw fit, they gave their daughters to be taken away. But, in connection with this, it is to be noticed that the bridal gifts (bride's price) for Rebekah were given not to her father, but to her mother and brother. Here the father was not the head of the family, but a secondary person in it; the mother was the head, and the daughter belonged to her and her kin.
Such might the family be among early Semites. And, of necessity, the kinship was not agnatic. There was admitted relationship between Laban and Jacob, his sister's son; and, from Judges, ix. 1–4, and many other passages, it is to be gathered that anciently a man's relatives on the mother's side considered him 'their bone and their flesh'—i.e. of their clan, or, at any rate, of their near kindred. Among the Semites of Arabia, beenah marriage was maintained for women of condition down to a comparatively late period; and at all times it was common for men who had been received into protection by a tribe to get a wife from it and to become incorporated with it through marriage.
Marriage by purchase ultimately supplanted the beenah marriage among the Hebrews, and became the prevailing marriage among the Arabs (and no doubt the same has happened in many cases). We proceed to describe the incidents which are found with it everywhere, while it is still in a direct or crude way a means of constituting family relations.
We find at this stage that the man often gets with his wife any children she has had already, and thus he gets children born of her while he has been her beenah husband; he gets, a fortiori, the child of which his wife is pregnant at the time of the marriage; he gets all children born to her thereafter whether they are his own or not (about which he is often found to be indifferent); and, in cases far from rare, he may even have children of the woman born after his death accounted his, and entitled to be his heirs, by means of what is called the Levirate. He may actually be the father of his wife's children, or of most of them; but very commonly he is ready to give her in loan, sometimes out of hospitality, sometimes for a price (in which latter case, the child, if any, may go to the borrower); he calls in the services of a friend who gives promise of being able to beget goodly offspring; and a child by his brother may appear to him to be in precisely the same case as a child by himself—in which we find an explanation of the Levirate. The foundation of the paternal relation in such a family is, in short, purchase, not paternity; the man has bought the woman and her issue, and (whether there be much paternal feeling or little) her children are counted his, as the increase of his cattle is. Purchase is here a means for taking children out of one kindred into another. Men desire children for the strengthening of their kindred, and no doubt because they have felt stirrings of paternal feeling. But paternity has not learned to be exacting, and it is of no avail to give any right to children. The purchase marriage of the early Arabs illustrates every point that has been mentioned, and so also does the family of the early Hindus.
The family founded upon purchase in this crude way is plainly a transitional form, and its future course is certainly upward. It takes a more familiar look, as paternal feeling strengthens, and the relationship of its members puts the contractual source of that relationship into the shade. And by-and-by purchase itself disappears or remains only in a symbol. In what may be taken to be an early stage, we very frequently find that at a man's death his brother succeeds him, inheriting his wife and children with his other property. Ultimately the sons succeed. And then we often find that (as in the Hindu family) the idea of family property is firmly established; so that the father, though having sole control, is regarded as only joint-owner with his sons of the family possessions—a state of things which must have originated in some form of joint-family, in which the father had been co-owner with his brothers or other kinsmen.
Now, to go back, with what does beenah marriage appear to be connected? We have seen its relation to marriage by purchase.
With beenah marriage, children belong to the family and kindred of their mother, and we find the same thing, and a reason for it, in the rudest family system which is known to us. This is what has been called the Nair family, because of the striking example of it found among the Nairs of Malabar. Among the Nairs bodies of the nearest relations form a joint-family, and hold their land in common, the control of the joint-interests being in the oldest male; but the family proper among them consists, when at the fullest, of a woman with her mother and brothers, and of her children. The uncles are the protectors of the children, who are their heirs; brother first succeeding to brother, and then the sister's children succeeding. The woman has no husband living with her; there is no father in this family; the woman has a number of husbands who visit her in turn; and, as a result of this polyandry and of the composition of the family, no Nair knows or thinks of his own father, and it is the uncles who fulfil the father's part. The children are of the kin of the mother only, because there is uncertainty of fatherhood, and no man in the father's place. And we may safely take it that in nothing less than this, or something fully equivalent, can the ignoring of the father and tracing of kinship through the mother only have anywhere originated.
A number of rude tribes are known which seem to have only the Nair family; but the Nair polyandry is more commonly found alongside of a system of polyandry which yields a family of a more advanced sort, in which the wife goes to live with her husbands, and the husbands are brothers (which, in the Nair system, they need not be, and generally are not). The eldest brother takes the wife and is head of the family, all the children being accounted his. This has been called Tibetan polyandry, because of its prevalence in Tibet; and it has been shown upon direct evidence to have prevailed very widely, both in ancient times and in modern. The Arabian polyandry described by Strabo is of this kind; so also and more distinctly the Hindu polyandry disclosed in the Mahābhārata; the polyandry of the Britons, described by Cæsar, is closely akin to it, but somewhat different. In families of the Tibetan type, the brother-husbands have got the woman into their mastery (be it by capture or by exchange or purchase); the children may be seen, as with monandry, to be of the blood of their kindred; and they might have been peacefully added to that kindred wherever exchange or purchase had been thought of. Only a sufficiency of women and of the means of living would then be wanting to change (let us say, among the Arabs) the Tibetan family into the monandrous purchase family already described. And the Nair would similarly be changed into the beenah family. It is male kinship that is now found with Tibetan polyandry, the brothers in order succeeding to the headship of the family, and then the eldest son of the brotherhood.
And where this succession law appears in the monandrous purchase family, brothers first succeeding, and the sons only after them, we may reasonably infer that there has been a movement from the Tibetan to the monandrous family. The same conclusion may also be drawn wherever, with succession of son to father established (a more advanced stage), it is held that a man may beget a son for his brother either in his lifetime or after his death (the latter case being the Levirate—familiar as occurring among the Hindus and the Hebrews); that provided a man's wife is mother, it suffices that his brother should be father, as it would have done in the Tibetan family. A prevalence of the Nair family, again, at some earlier period may be inferred, as we have seen, from our finding kinship counted through females only, although the family where this occurs may not be Nair.
We find this kinship with monandry, usually with the man head of the house—another form of the family, husband and children not being counted relatives—very widely, and among peoples in very different stages of advancement. Thus, it is very common in tribes like those of Australia and America, and it is established also among peoples comparatively so far advanced as the Ashantis. In the greater number of Australian tribes, though polygamy is practised largely, and a wife is little better than a slave, and is valued chiefly for her services (licentious practices which tend to make fatherhood uncertain being, however, exceedingly prevalent), children belong to the kindred of their mother, and, as the father is not allowed to marry women of his own kindred, not to that of their father; so that the children may be of as many kindreds as the man has wives, but are never of his own kindred. Among the Ashantis, amid distinctions of rank, with wealth accumulated and with women jealously watched over, we find children similarly counted of the kin not of their father, but of their mother, the sister's son being a man's heir in preference to his son; while among the Fantis, a neighbour people of the same race, failing a sister's son, the chief slave is heir, the son being excluded from succession in his father's family altogether. The family religions seem to have in this and similar cases preserved a kinship which is not now consistent with the domestic relations.
The same kinship has of course prevailed wherever the wife has been the head of the family, of which many examples, more or less trustworthy, have been collected. In some of these daughters only have been heirs (Lycia, Egypt), sons going into other families with or without a dowry. A former prevalence of this kinship (and therefore of the Nair family) may, moreover, be inferred from such facts as permission of marriage between persons closely related through the father (but not relations under female kinship)—e.g. brother and sister german, who might marry in a certain case in Attica (for which we also have in Athenæus a clear tradition of the Nair family—of a time when no man knew who was his father), and more generally, it would seem, among the Hebrews (Abraham and Sarah, Amnon and Tamar); as the feeling for the sister's son which Tacitus noted among the Germans—which was precisely that of the rudest female kinship tribes; and, a fortiori, from the succession of the sister's son, if only to the sovereignty, as among the Celts of Scotland, or of the mother's brother, as in the story of Meleager. By inference the evidence for the Nair family may thus be greatly enlarged, as we have seen that the evidence for the Tibetan family may also be.
Taken altogether, the evidence suggests strongly that these two systems occurred commonly together among a people, as beenah marriage and purchase marriage do—each having left its mark upon the same people's customs. The evidence shows also, if good for anything, that the Nair kinship was supplanted (traces of it long remaining) by the kinship which the purchase family brought in or established.
So much polyandry among early men would of course indicate that very generally their circumstances were like those of most polyandrous peoples now, in that the struggle for existence was severe, and they had learned to keep down the number of their women by infanticide to the lowest possible point. And geologists have shown us that their circumstances were generally far harder than those of any peoples now known to us. As men have had to reason out for themselves systems of kinship, it is now intelligible that at first and for long they should not get beyond the kinships which are most obvious, those which arise through the mother, and which sufficed to connect the children with her tribe (polyandry of the Tibetan type being indeed impossible until this kinship has made a good deal of progress); and that thus the early bands of men at first acknowledged only the Nair kinship.
The theory of the growth of societies which the various forms of the family taken together seem to point to can now be indicated, but here it can only be outlined very briefly.
Among tribes of the ruder sort, whether having female or male kinship, the true family (and so travellers constantly name it) consists of all those who are considered relatives—of all who are of the same Totem (q.v.)—all of them being prohibited from marrying each other (exogamy); being bound together by the regard they have for the totem; and also all liable equally to be struck at for each other's offences, and all bound equally to revenge each other's injuries (the blood-feud). The totem is usually an animal or plant after which they are named, and from which they believe themselves to be descended. The tribe among such peoples is not a body of kindred; it is made up of people of several different totems; and the same totems are usually spread over many neighbouring tribes—all those who are of the same totem in all the tribes constituting the totem family. Now exogamy and female kinship account for this interfusion of the totem families throughout the tribes. It is exactly what would arise if a number of neighbouring totem families, having the female or Nair kinship, became exogamous—i.e. ceased to take to wife their own women and took the women of their neighbours. And, as the whole system depends upon the totem, we must take it that the totem families existed separately before exogamy inter-fused them. This gives us, for a beginning, a totem group or family having Nair kinship and not exogamous; and, next, when exogamy (in whatever way) became established, an admixture of neighbouring totems in each of the groups; which admixture, if male kinship supervened, would necessarily remain—except in so far as the scattered portions of the totem families joined together and drew apart and became each family a separate clan. With male kinship (i.e. children of the totem or kindred of their father) exogamy would bring no totem from the outside into a group; and, therefore, we are forced to believe that wherever a tribe consists of several exogamous male kinship clans or families, these must at first have had the Nair kinship. Exogamy has been the prevailing marriage law among all races; female kinship, too, is known upon direct evidence to have prevailed widely; and, accordingly, this gives a good explanation of the constitution of all the tribes which have consisted of several clans, that difficulty for the patriarchal theory for which Maine proposed the hypothesis of a fiction.
As to the family proper, we have, on this view, to believe that (though the partnership of brothers in women may have commenced in the totem group while still separate, when women would probably be got by capture), the kinship was Nair, or predominantly Nair, until the interfusion of the totems had taken place. After that, peaceful marriages could be made within the tribe between persons of different totems; they might be either of the Nair type or the Tibetan—these giving way, as circumstances improved, to beenah marriage and the more familiar form of monandry; and by-and-by, when men came to desire it, marriage agreements (made upon the consideration of exchange or purchase) would allow of children being added to the totem of their father. This brings us to the competition of the Nair or beenah family and the purchase family, which naturally ended in the triumph of the latter, followed by a progress therefrom to the family in its more familiar forms. But a kinship established in the totem family could not be easily supplanted; so that female kinship might have a long career under favourable circumstances. Licentious practices which make fatherhood uncertain, and servility to custom, seem to have maintained it among the lower tribes; while the force of religion—the totem (animal or plant or other natural object) having become a god, and its worship regulating in many ways the lives of its devotees—accounts for its having lasted among the higher African peoples.
Interfusion of 'families' (i.e. 'kindreds'), such as is produced by exogamy and female kinship, is nowhere more strikingly illustrated than among the Ashantis and their neighbour peoples.
For facts and reasonings bearing upon the subject of this article, see Sir H. S. Maine's works, especially Ancient Law, and Early Law and Custom; J. D. Mayne's Hindu Law and Usage; Bachofen's Das Mutterrecht, and in connection therewith, Giraud-Teulon's La Mère chez certains Peuples de l'Antiquité; Studies in Ancient History, and the Patriarchal Theory, by J. F. McLennan; also, 'The Worship of Animals and Plants,' Fortnightly Review (1869-70); Kinship and Marriage in Early Arabia, by W. Robertson Smith; Herbert Spencer's Principles of Sociology, vol. i.; Sir J. Lubbock's Origin of Civilisation; and Lewis H. Morgan's Ancient Society.