The children of concubitants who have neglected to intermarry do not, as in Mbau, become tabu, but are made to repair their parents' default by themselves becoming concubitants.

[Pageheader: CONCUBITANCY UNKNOWN IN POLYNESIA]

In Lau, Thakaundrove, and in the greater portion of Vanualevu, the offspring of a brother and sister respectively do not become concubitant until the second generation. In the first generation they are called tabu, but marriage is not actually prohibited. The children of two brothers or of two sisters are, as in Mbau, strictly forbidden to intermarry.

Inquiries that have been made among the natives of Samoa, Futuna, Rotuma, Uea, and Malanta (Solomon Group), have satisfied me that the practice of concubitant marriage is unknown in those islands; indeed, in Samoa and Rotuma, not only is the marriage of cousins-german forbidden, but the descendants of a brother and sister respectively, who in Fiji would be expected to marry, are there regarded as being within the forbidden degrees as long as their common origin can be remembered. This rule is also recognized throughout the Gilbert Islands, with the exception of Apemama and Makin, and is there only violated by the high chiefs. In Tonga, it is true, a trace of the custom can be detected. The union of the grandchildren (and occasionally even of the children) of a brother and sister is there regarded as a fit and proper custom for the superior chiefs, but not for the common people. In Tonga, other things being equal, a sister's children rank above a brother's, and therefore the concubitant rights were vested in the sister's grandchild, more especially if a female. Her parents might send for her male cousin to be her _takaifala_ (_lit._, "bedmaker") or consort. The practice was never, however, sufficiently general to be called a national custom. So startling a variation from the practice of the other Polynesian races may be accounted for by the suggestion that the chiefs, more autocratic in Tonga than elsewhere, having founded their authority upon the fiction of their descent from the G.o.ds, were driven to keep it by intermarriage among themselves, lest in contaminating their blood by alliance with their subjects their divine rights should be impaired. A similar infringement of forbidden degrees by chiefs has been noted in Hawaii, where the chief of Mau'i was, for reasons of state, required to marry his half-sister. It is matter of common knowledge that for the same reason the Incas of Peru married their full-sister, and that the kings of Siam marry their half-sisters at the present day.

_Origin of the custom._--I venture to offer here three possible explanations of the origin of this custom, leaving it to the acknowledged authorities upon the history of marriage to point out what in their opinion is the true explanation:--

1. It may be a survival of an earlier custom of group-marriage and uterine descent such as is to be found in the Banks Islands, where the entire population is divided into two groups, which we will call X. and O. A man of the X. group must marry an O. woman, and _vice versa_. The children, following the mother, are O.'s, and are, therefore, kin to their mother's brother rather than to their own father. Their mother's brother, an O., marries an X. woman, whose children are X.'s, and are potential wives to their first cousins; although in the Banks group the blood relationship is not lost sight of, and close marriages are looked upon as improper, whilst in Fiji such a union would be obligatory.[75]

The children of two brothers of the X. group, following their mothers, would be O.'s, and therefore forbidden to marry; and so also would be the children of two sisters. Thus far the results of the two customs are the same; but in the Banks group consanguineous marriage is checked by public opinion, which in Fiji favours such marriages. Group-marriage on precisely the same lines has been noticed in Western Equatorial Africa[76] and among the Tinne Indians in North-West America.[77]

In Fiji, agnatic has generally taken the place of the uterine descent (although in some parts of Vanualevu traces of the custom still appear to linger), but the existing system of _vasu_, which gives a man extraordinary claims upon his maternal uncle, may be an indication that concubitant marriage is a survival of the more ancient custom. The _vasu_ system is found to some extent among all peoples who trace descent through the mother. Tacitus, speaking of the ancient Germans, says that the tie between the maternal uncle and his nephew was a more sacred bond than the relation of father and son.[78]

[Pageheader: ORIGIN OF CONCUBITANCY]

2. It is also possible that concubitant marriage is a relaxation of the stricter prohibition in force amongst the Polynesians. The origin of these prohibitions may, perhaps, be found in some such occurrence as that described in the "Murdu" legend of Australia, quoted by Messrs.

Fison and Howitt in _Kamilaroi and Kurnai_--

"After the Creation brothers and sisters and others of the closest kin intermarried promiscuously, until, the evil effects becoming manifest, a council of the chiefs was a.s.sembled to consider in what way they might be averted."

Some such crisis must have been reached in every group of islands that was peopled by the immigration of a single family, and the natural solution in every case would have been to prohibit the marriage of both cla.s.ses of cousins-german. But, little by little, the desire for alliances among chief families, for the restoration of the claims of _vasu_, and for the restoration of an equivalent of the tillage rights given in dowry, may have chafed against the prohibitions until these were so far relaxed as to allow the marriage of cousins in the degree most effective for promoting an interchange of property. For a similar reason Moses ordered the daughters of Zelophehad to marry men of their father's tribe, in order that their property should not pa.s.s out of the tribe, and "their inheritance remained in the tribe of the family of their father" (Numbers x.x.xvi. 12).

3. A third solution may be found in the transition from uterine to agnatic descent, a change that came about gradually as social development prompted the sons to seize on the inheritance of their father to the exclusion of the nephew (_vasu_). With the admission of the father's relationship to his son grew the idea that he was the life-giver and the mother the mere vehicle for the gestation of the child, and the child came to be regarded as related to his father instead of to his mother.[79] Thus Orestes,[80] arraigned for the murder of his mother, Clytemnestra, asks the Erinyes why they did not punish Clytemnestra for slaying her husband Agamemnon; and, upon their answer that she was not kin to the man she slew, he founds the plea that by the same rule they cannot touch him, for he is not kin to his mother.

The plea is admitted by the G.o.ds. By this rule, a man is not kin to his father's sister's daughter, she being kin to her father only; but her affinity to him would render their marriage convenient as regards the family possessions. From long usage a sense of obligation would be evolved, and such cousins come to be regarded as concubitant. The children of sisters would still be within the forbidden degrees, for, although not kin through their mothers, their fathers, being presumably the concubitant cousins of their mothers, would be near kin.

I incline to accept the first explanation--that the custom of concubitancy has been evolved from an earlier system of group-marriage and uterine descent. I think that it dates from the remote period when there was indiscriminate intercourse between the members of two exogamous marrying cla.s.ses, when it was impossible to say who was the actual father of the children born. Under such a system the reputed offspring of two brothers might in reality be the children of only one of them, and the children of two sisters might have a common father, and their union be incestuous. But the children of a brother and sister respectively could not possibly have a common parent, and their intercourse was therefore innocuous. For the same reason the children of concubitants who were not known to have cohabited were still held to be tabu to each other, for the male concubitant had a right of cohabitation with the female of which he might at any time have availed himself, and their offspring reputed to be by their other partners might in reality be half brother and sister without their knowledge.

[Pageheader: CENSUS OF CONSANGUINEOUS MARRIAGES]

Though the Fijian system of relationships is closely allied to those of the Tamils in India and the Two-mountain Iroquois, and the Wyandots in North America, none of these, except the Tamils, I believe, recognize the principle of concubitant cousinship. The custom must be regarded, I think, as being one of limited range, evolved from marriage laws of far wider application. It undoubtedly exercises upon the Fijians a marked influence in promoting consanguineous marriages--an influence from which the other races in the Pacific are comparatively free, if we except the inhabitants of the island of Tanna in the New Hebrides and possibly some other islands not yet systematically investigated.

_Concubitancy in practice._--The fact of a race of men habitually marrying their first cousins promised to exhibit such remarkable features in vital statistics that we did not stop short at investigating the theory alone. We caused a census to be taken of twelve villages, not selected from one province, but chosen only for convenience of enumeration in the widely separated provinces of Rewa, Colo East, Serua, and Ba. I am indebted to the late Mr. James Stewart, C.M.G., for the a.n.a.lysis of the returns which follows:--

In the twelve villages there were 448 families. The couples forming the heads of these families have had born to them as children of the marriage 1317 children, an average of 294 to each marriage. But of these 1317 children, only 679 remain alive, 638 being dead. The heads of these families therefore do not replace themselves by surviving children, for only 515 per cent. survive, while 485 are lost.

We divided the married couples into four cla.s.ses--

(1) Concubitant relations who have married together. These we found to be on inquiry in nearly every case actual first cousins. They formed 297 per cent. of the total number of families.

(2) Relations other than concubitant cousins who have intermarried.

Two-fifths of these are near relations, uncle and niece, and non-marriageable cousins-german, brother and sister according to the Fijian ideas. But the remaining three-fifths are more distantly related than are the concubitants. These form 123 per cent. of the total number of families.

(3) Fellow villagers--natives of the same village, but not otherwise related--who have married together. These form 321 of the total number of families.

(4) Natives of different villages, not being relations who have intermarried. These form 259 of the total number of families.

Thus it will be seen that the concubitant and other relations who have intermarried number over two-fifths of the people, while one-third of the married people have been brought up together in the same village, and only one-fourth, not being relations, have come from different villages.

When we examined the relative fecundity of these divisions the result was not a little startling--

133 concubitant couples have had 438 children, or 330 children per family.

55 families of relations have had 168 children, or 306 children per family.

144 families of fellow-villagers have had 390 children, or 271 children per family.

116 families of natives of different villages have had 321 children, or 277 children per family.

It will thus be seen that as regards fecundity, concubitant marriages are greatly superior to any of the other cla.s.ses.

But since fecundity does not necessarily mean vitality, the question is, how many of the children born to these respective divisions have survived? and we find the unexpected result that whereas the other cla.s.ses have changed places, the concubitants again show themselves to be superior.

Of 133 families of concubitants, there were 438 children, of whom 232 survive, and 206 are dead.

Of 55 families of relations, not concubitants, there were 168 children, of whom 72 survive, and 96 are dead.

Of 144 families of townspeople, there were 390 children, of whom 212 survive, and 178 are dead.

Of 116 families of natives of different villages, there were 321 children, of whom 163 survive, and 158 are dead.

[Pageheader: VITALITY OF INBRED CHILDREN]

The concubitants with an average surviving family of 174 show, therefore, not only a higher birth-rate, but far the highest vitality of offspring.

The relations other than concubitants show, it is true, the highest fecundity next to the concubitants, but their rate of vitality is the lowest of the four cla.s.ses, since more of their children have died than are now living.

Second in point of vitality come the fellow-villagers, but they are far behind the concubitants.

From our preconceived ideas of the advantages of out-breeding we should expect to find that the offspring of natives of different villages would have shown, if not the highest fecundity, at least the highest vitality, for this is the cla.s.s in which the parents are not related. On the contrary, we find that the families of these unrelated people are only third in point of vitality.

In view of the unfavourable position which the "relations other than concubitants" hold in this a.n.a.lysis, it is well to divide the group into two sub-cla.s.ses. Of the fifty-five families of "relations," thirty-three are stated to be _kawa vata_ (_i.e._ of the same stock, but not necessarily of the same family or generation). The remaining twenty-two families, on the other hand, consist of such unions (incestuous from the Fijian point of view) of _vei-nganeni_ or _vei-tathini_, that is to say, brother and sister, or cousins not concubitant; _vei-vungoni_, uncle and niece, or aunt and nephew; _vei-tamani_, father and daughter, or paternal uncle and niece; and _vei-luveni_ or _vei-tinani_, maternal aunt and nephew, or mother and son. We have therefore, for purposes of identification, divided the group into--first, relations distant; second, relations specified.

----------------------+------------+---------------------------------- | | Divisions. | Number of | Children of the Marriage.

| Families. +---------------------------------- | | Alive. | Dead. | Total.

----------------------+------------+-------------+---------+---------- Relations (distant) | 33 | 49 | 61 | 110 Average per family | -- | 148 | 185 | 333 | | | | Relations (specified) | 22 | 23 | 35 | 58 Average per family | -- | 105 | 159 | 264 ----------------------+------------+-------------+---------+---------- Total | 55 | 72 | 96 | 168 ----------------------+------------+-------------+---------+---------- Average per family | 131 | 175 | 36 ----------------------+------------+-------------+---------+----------

The fecundity of these distant relations thus appears to be much higher than that of the specified relations, and a little higher even than that of the concubitants--the highest of the four groups. The comparative figures are as follows--

------------------------------------+---------------------------------- | Average Family.

------------------------------------+-----------+---------+------------ | Alive. | Dead. | Total.

------------------------------------+-----------+---------+------------ _Vei-ndavolani_ (concubitants) | 174 | 156 | 330 Relations (distant) | 148 | 185 | 333 ------------------------------------+-----------+---------+------------

The vitality therefore is much less in the case of relations distant than among the children of the concubitants.

The fecundity of the division, "relations specified," is lower than that of any of the four groups, and the vitality of their progeny is greatly inferior to any of the other cla.s.ses.

For the last twenty years the Fijians have been either stationary, slightly increasing, or decreasing, according to the prevalence of foreign epidemics, the balance being in favour always of decrease. The different figures show that no cla.s.s of the population replaces itself by surviving children of the marriage. But the deficiency is made up by the children of former marriages, and illegitimate children, who form a large portion of the population, but whose case it was not necessary to consider for the purposes of this chapter. But we find the startling fact that the cla.s.s that most nearly succeeds in replacing itself is that of the concubitants, which, consisting of 133 families, or 266 individuals, have, out of a total number of children born to them of 438, a surviving progeny of 232. If we add the surviving step-children of these individuals, their total surviving progeny becomes 317, thus replacing the heads of existing families, and leaving 51 children to replace the parents of the step-children. In every respect the concubitants appear to be the most satisfactory marriage cla.s.s. They amount to only 297 per cent. of the population, but they bear 333 per cent. of the children born, and they rear 342 of the children reared; and, including step-children, they rear 347 of the children who survive.