The Melanesians: Studies in their Anthropology and Folklore/Chapter 2



There will be no attempt made here to deal with the Ethnology of Melanesia. The origin of the Melanesian people, in their various seats and in their various divisions, may be taken to be unknown; as they themselves apparently have no traditions and no opinions about the matter, and in the stories which pass among them represent themselves to have been created where they are. The variety of their languages, and to a much less extent of their arts and customs, shews that they have not come in one body into the islands they now inhabit; an examination of their languages discovers a very considerable underlying sameness; and the present book may be taken perhaps as an evidence of a large general resemblance in the religious beliefs and practices, the customs and ways of life, which prevail in the islands which are here embraced in a common view. As knowledge extends and detailed information is brought in from all sides, a connexion will no doubt be traced with regions beyond Melanesia; the loom, for example, peculiar to Santa Cruz alone among the islands here treated of, may connect the people of that group with those of the Caroline Islands; many things in common between Fiji and Madagascar besides language may bring those countries and much that lies between them into whatever ethnographic province the latter is held to belong to; but to endeavour to trace such connexion is beyond the present purpose, which is confined to the exhibition of the Melanesian people as they now appear. There are not wanting some myths of origin, over and above the stories of creation told of Koevasi, Qat, or Tagar. It is said at Saa for example, in Mara Masiki, that men sprung spontaneously from a sugar-cane of a particular sort, tohu nunu: two knots began to shoot, and the cane below each shoot burst asunder; from one came out a man, and from the other a woman, the parents of mankind. It is of more consequence to observe the meaning of the words by which the people of the various islands describe themselves as men. It is said sometimes that people discovered in isolation from others call themselves merely 'men,' without a name for their race or nation, as if they thought themselves the only men in the world. In Melanesia, when natives were first asked who they were, they answered 'men,' meaning that they were not demons or ghosts, but living men; and they did so because they did not believe their visitors to be men, but ghosts themselves, or demons, or spirits belonging to the sea.

In the native view of mankind, almost everywhere in the islands which are here under consideration, nothing seems more fundamental than the division of the people into two or more classes, which are exogamous, and in which descent is counted through the mother. This seems to stand foremost as the native looks out upon his fellow men; the knowledge of it forms probably the first social conception which shapes itself in the mind of the young Melanesian of either sex, and it is not too much to say that this division is the foundation on which the fabric of native society is built up. There are no Tribes among the natives; if the word tribe is to be applied as it is to the Maori people of New Zealand, or as it is used in Fiji. No portion of territory, however small, can be said to belong to any one of these divisions; no single family of natives can fail to consist of members of more than one division; both divisions where there are two, and all the divisions where there are more than two, are intermixed inhabitation and in property; whatever political organization can be found can never be described as that of a tribe grouped round its hereditary or elective chief. It is probably true that in every account of Melanesian affairs given to the world tribes are spoken of; but a belief that every savage people is made up of tribes is part of the mental equipment of a civilized visitor; when one reads of the 'coast tribes' or the 'bush tribes,' nothing more is meant than the people who inhabit the coast or the inland part of some island.

There is, however, one very remarkable exception to this general rule of division in the Solomon Islands; it is not to be found in Ulawa, Ugi, and parts of San Cristoval, Malanta, and Guadalcanar, a district in which the languages also form a group by themselves, and in which a difference in the decorative art of the people, and in the appearance of the people themselves, thoroughly Melanesian as they are, can hardly escape notice. In this region, the boundaries of which are at present unknown, there is no division of the people into kindreds as elsewhere, and descent follows the father. This is so strange that to myself it seemed for a time incredible, and nothing but the repeated declarations of a native who is well acquainted with the division which prevails in other groups of islands, was sufficient to fix it with me as an ascertained fact. The particular or local causes which have brought about this exceptional state of things are unknown; the fact of the exception is a valuable one to note[1].

Speaking generally, it may be said that to a Melanesian man all women, of his own generation at least, are either sisters or wives, to the Melanesian woman all men are either brothers or husbands. An excellent illustration of this is given in the story of Taso from Aurora in the New Hebrides, in which Qatu discovers and brings to his wife twin boys, children of his dead sister: his wife asks, 'Are these my children or my husbands?' and Qatu answers, 'Your husbands to be sure, they are my sister's children.' In that island there are two divisions of the people; Qatu and his wife could not be of the same, Qatu and his sister and her children must be of the same; the boys therefore were possible husbands of Qatu's wife, but had they belonged to the other division their age would have made her count them her children rather than her brothers. It must not be understood that a Melanesian regards all women who are not of his own division as in fact his wives, or conceives himself to have rights which he may exercise in regard to those women of them who are unmarried; but the women who may be his wives by marriage, and those who cannot possibly be so, stand in a widely different relation to him; and it may be added that all women who may become wives in marriage and are not yet appropriated, are to a certain extent looked upon by those who may be their husbands as open to a more or less legitimate intercourse. In fact appropriation of particular women to their own husbands, though established by every sanction of native custom, has by no means so strong a hold in native society, nor in all probability anything like so deep a foundation in the history of the native people, as the severance of either sex by divisions which most strictly limit the intercourse of men and women to those of the section or sections to which they do not themselves belong. Two proofs or exemplifications of this are conspicuous, (1) There is probably no place in which the common opinion of Melanesians approves the intercourse of unmarried youths and girls as a thing good in itself, though it allows it as a thing to be expected and to be excused; but intercourse within the limit which restrains from marriage, where two members of the same division are concerned, is a crime, is incest. In Florida in old times the man would have been killed, and the woman made a harlot; now that the severity of ancient manners is relaxed, money and pigs can condone the offence, but much more than is exacted if a man is found sinning with one who might possibly have become his wife. In the Banks' Islands, where the divisions of the people are two, if it became known that two members of one of them had been guilty of this disgraceful crime, as they considered it, the people of the other division would come and destroy the gardens of those who belonged to that in which the offence had been committed, and these would make no resistance nor complaint. It was the same in Lepers' Island; where the offending man had also to make large payment to the near relatives of the woman with whom he had offended, so as to appease their anger, and 'fence against' the fault. Cases of incest of this kind were always rare in all the islands, so strong was the feeling against intercourse within the kin. (2) The feeling on the other hand that the intercourse of the sexes was natural where the man and woman belonged to different divisions, was shewn by that feature of native hospitality which provided a guest with a temporary wife. That this is done now or has lately been done is readily denied in the Solomon and Banks' Islands, but is not denied in the Northern New Hebrides; there can be little doubt that it was common everywhere. But the woman supplied to the guest was of necessity one who might have been his wife; the companionship of one of his own kin never could be allowed.

It will be convenient in the more particular treatment of this subject, to take examples first from the Banks' Islands and Northern New Hebrides, where the people are divided into two kins, and then from the Solomon Islands, where the divisions are more than two. The same two divisions run through the Banks' Islands, with the Torres Islands and the Northern New Hebrides. A Banks' islander wherever he goes in his own group knows his own kin, and if he passes to Aurora in the New Hebrides he finds the same. The Aurora men know well who are their kin in Pentecost and Lepers' Island; the Lepers' islanders know theirs in Espiritu Santo. Strange, therefore, as the language is to a Mota man in Pentecost, or to a Lepers' islander in Motalava, each is at home in a way which would be impossible to him in the Solomon Islands[2]. In neither the Banks' Islands nor the New Hebrides is there a name to distinguish the division or kindred; nor is there any badge or emblem belonging to either; in their small communities every neighbour is well known. Each of the divisions is in Mota called a veve, in Motlav vev, a word which in itself signifies division. Those who are of one veve are said to be tavala ima to the others, that is 'of the other side of the house.' A woman who marries does not come over to her husband's side of the house; she is said to be ape mateima, 'at the door,' the doors being at the ends of the native houses; nor does the husband go over to the wife's side; the children belong to mother's side. All of the same 'side of the house' are sogoi to one another. Hence a man's children are not his sogoi, his kindred; his nearest relations are his sister's children. There is no account seriously given of the origin of the two divisions in the Banks' Islands. Within the two veve there are certain families among the Banks' Island people, the members of which have a certain family pride, and endeavour to keep up by intermarriage the family connexion. The best known of these is the Lo Sepere family, from the place of that name in Vanua Lava, where Qat is believed to have lived.[3] Adoption is common, and has no particular significance. Childless parents naturally adopt a child of kin to the wife, so that the adopted child occupies the position of one born in the house; but if, as sometimes happens, an orphan child from the husband's kin is adopted out of pity, it is brought up as of kin to the wife, and care is taken to conceal the fact of adoption. When the child grows up and by some chance finds out that he has been brought up on the wrong 'side of the house,' he will leave his foster parents, and go and live with his own sogoi. Much grief and bitterness is caused by such a discovery.

In Aurora, Maewo, the nearest of the New Hebrides to the Banks' Islands, with one of which, Merlav, there is a good deal of communication, the members of the two divisions speak of one another as 'of the other side,' ta tavuluna; and they have a story that the first woman, a cowry shell that turned into a woman, called the men to her and divided them into her husbands and her brothers, fathers and maternal uncles, according to the present arrangements. The presence of families within the kin in this island is very remarkable. There are several in the northern part of the island, mostly named from the places where they are formed. There is one, however, named from the octopus, wirita, belonging originally to Bugita, a place upon the shore. The connexion between this family and the octopus is obscure; they have no notion of descent from the wirita, and eat it as freely as other natives; but if a man of another family desired to get wirita for food, he would take with him one of the wirita family to stand on the beach at Bugita, and cry out, 'So-and-so wants wirita'; then plenty would be taken. It seems rather as though the residence of this family where wirita are abundant, and where the beach would naturally be their preserve for fishing, had given rise to a belief in a connexion and to a name. Another family named 'At the Wotaga,' from their home near a certain fruit tree, would not bring up a light-coloured child; if such a one were among them they thought that they would die[4].

In Araga, Pentecost Island, though irregular intercourse between members of the same kin is punished by the destruction of the gardens of the offending side by the members of the other, yet marriages within the kin are not unknown. Those who contract them are despised, and even abhorred, but money and pigs having been given and received, the marriage stands. In Lepers' Island, Omba, the two divisions are called 'bunches of fruit,' wai vung, as if all the members hang on the same stalk. Their story is that when Tagar first made men he made two, both male, and then one of these took a tuber of qevu, a kind of yam, and threw it at the other, who at once turned into a woman, and cried with a loud voice that many men should die because of women. This woman had two daughters, who fell out; and from one of these sprang one waivung, and from the second the other. In case of the adoption of a child by a foster-mother who is of the other 'bunch,' the secret of the kindred is carefully kept; the true state of the case is never mentioned by those who know it, until the time for marriage comes. This is done out of consideration for the feelings of the adopting parents; but the repugnance to marriage within the kin is too great to allow of permanent concealment.

The system of the division of the people into strictly exogamous kins is no doubt best seen and considered where the division is simple and separates the whole population on the one 'side of the house' and the other. Two questions may here therefore be suitably raised; the first, whether in this division there are traces of a communal system of marriage; the second, whether the system is sufficient to prevent that which it seems intended or maintained in order to prevent, namely, the marriage of persons too closely allied in blood. In regard to the first question it must be said, on the one hand, that the people have no memory of a time when all the women of one side were in fact common wives to the men of the other side, and that there is no occasion on which the women become common to the men who are not of their kin. The license of a gathering at a feast is confessed to be great, but it is disorderly and illegitimate, and is not defended on the ground of prescription. If a great man making a feast gives it to be understood that he will not allow the harmony of the gathering to be spoilt by jealous quarrelling about women, it is taken as a festive concession; if he gives out that people are to behave well, they know that any one who takes liberties will have to answer for it, not only, as on ordinary occasions, to the injured husband, but to the powerful master of the feast. The stories also of the creation of mankind, and particularly of woman, represent individual marriage. When Qat wove Iro Lei with pliant rods and made her live, it was to be his own wife; his brothers tried to carry her off for themselves, one woman among eleven of them, but they are said to be stealing her, not claiming a right. When he made men, male and female, he assigned to each man his wife. On the other side is to be set the testimony, the strong testimony, of words. This is given by the plural form in which the terms for 'mother' and 'husband' or 'wife' are expressed. In the Mota language the form is very clear; ra is the plural prefix; the division, side, or kin, is the veve, and mother is ra veve; soai is a member, as of a body, or a component part of a house or of a tree, and ra soai is either husband or wife. To interpret ra as a prefix of dignity is forbidden by the full consciousness of the natives themselves that it expresses plurality. The kin is the veve, a child's mother is 'they of the kin,' his kindred. A man's kindred are not called his veve because they are his mother's people; she is called his veve, in the plural, his kindred, as if she were the representative of the kin; as if he were not the child of the particular woman who bore him, but of the whole kindred for whom she brought him into the world. By a parallel use to this a plural form is given to the Mota word for child, reremera, with a doubled plural sign; a single boy is called not 'child' but 'children,' as if his individuality were not distinguished from the common offspring of his veve. The same plural prefix is found in other Banks' Island words meaning mother; rave in Santa Maria, retne in Vanua Lava, reme in Torres Islands. The mother is called ratahi in Whitsuntide, and ratahigi in Lepers' Island, that is the sisters, the sisterhood, because she represents the sister members of the waivung who are the mothers generally of the children. Similarly the one word used for husband or wife has the plural form. In Mota a man does not call his wife a member of him, a component part of him, but his members, his component parts; and so a wife speaks of her husband. It is not that the man and his wife make up a composite body between them, but that the men on the one side and the women on the other make up a composite married body. The Mota people know that the word they use means this; it was owned to myself with a blush that it was so, with a Melanesian blush, and a protestation that the word did not represent a fact. The word used in Motlav, part of Saddle Island, gives hardly the less confirmation to this interpretation of the Mota word because it has not a plural form; in Motlav ignige has the same meaning with the Mota soai; a man says of his leg or his arm ignik, my member, one of my members, and he calls also his wife ignik, while she calls him the same.

As concerns the second point in question, it is apparent that the strict rule of exogamy as regards the kin leaves marriage open to those who are very near in blood; for a man is not of kin to his own children, and a man is not of kin to his brother's children. But although it is the intermarriage of sogoi, members of the same veve, that is strictly forbidden, and the descent is always counted by the mother, yet the blood connexion with the father and the father's near relations is never out of sight. Consequently the marriage of those who are near in blood, though they are not sogoi and may lawfully marry, is discountenanced. In Mota, for example, the children of a brother and sister are thought too near to marry. The brother and sister are both of one veve, A, as children of one mother; the children of the sister are of her veve, A; the brother's children are of the veve B, following their mother, who must needs be of the other side of the house. It appears then that the two cousins, children of a brother and sister, are not sogoi, one being A and the other B, and that they can marry. But they will not; the match will not be made; if they married they would be said to 'go wrong[5].' It will be seen that the succession to property shews the same tendency, perhaps a recent tendency, to the recognition of agnatic descent.

Florida, and the parts of the Solomon Islands adjacent to it, afford an example of the division of the people into more than two exogamous kindreds. In Florida these divisions are six, called kema, and each has its distinguishing name. These are the Nggaombata, the Manukama or Honggokama, the Honggokiki, the Kakau, the Himbo, and the Lahi. But these six kema no doubt represent a much simpler original division; for two of them have local names, of Nggaombata in Guadalcanar, and Himbo, the Simbo somewhat indefinitely placed among the islands to the west, from whence these two kema are known to have come. The Nggaombata and the Himbo, perhaps only as strangers, go together; and the Lahi, a small division, are said to be so closely connected with Himbo that the members cannot intermarry. Whether Honggokama and Manukama are names of one kema, or of two divisions into which the one is separating, is a question. The Honggokama and the Honggo-kiki, the great and the little, are plainly parts of one original. It is not the case in Florida that an originally double division has simply split and split again; but the settlement of foreigners has so complicated the arrangement that few natives profess to be able to follow it[6]. Yet the foreigners have undoubtedly brought with them a distinct sense of kinship with one or other of the local kema. The strict rule of exogamy is not a sufficient limit to the right of marriage; here also, as in the eastern islands, it is supplemented by a strong public opinion as to what is right. A remarkable instance of this occurred a few years ago, when Takua, a considerable chief, took to himself the daughter of one of his wives. The girl was not, of course, of his own kema, and so far he was within his right, but the sense of decency and propriety of the people was outraged, and the man's influence as a chief was much diminished. In Bugotu of Ysabel there are three vinahuhu: Dhonggokama, Vihuvunagi, and Posomogo, not one of which now corresponds exactly with either of the Florida kema. But the Dhonggokama, they say, is the same as the ancient kema which has split into the Honggokama and Honggokiki in Florida; and the other two may be well believed to be themselves the divided other member of the original pair. The meaning of the names of three of the Florida kema, besides the two that are local, are known; Honggo is cat's-cradle, Manukama is an eagle, Kakau is a crab. It is evident that when the divisions of a people multiple names must be given them; where there are two 'sides of the house' no name is needed for either, but when a man may have wives and children of three or four kindreds not his own, a name for each kin is necessary to maintain the matriarchal system of descent through the mother.

It adds very much to the distinction between these kema, that each has some one or more buto from which its members must keep clear, abstain from eating, approaching, or beholding it[7]. One of the very first lessons learnt by a Florida child is what is its buto, its abomination, to eat or touch or see which would be a dreadful thing. In one case, and in one case only, this buto is the living creature from which the kema takes its name; the Kakau kin may not eat the Kakau crab. The Nggaombata may not eat the giant clam; the Lahi may not eat of a white pig; the Manukama may not eat the pigeon; the Kakau, besides their eponymous crab, may not eat the parrot Trichoglossus Massena. The Manukama are at liberty to eat the bird from which they take their name. If the question be put to any member of these kema he will probably answer that his buto is his ancestor; a Manukama will say that the pigeon he does not eat is his ancestor; but an intelligent native, describing this native custom, writes:—'This is the explanation of the buto. We believe these tindalo (the object of worship in each kema) to have been once living men, and something that was with them, or with which they had to do, has become a thing forbidden, tambu, and abominable, buto, to those to whom the tindalo belongs.' He gives the example of the clam of the Nggaombata. The ghost, tindalo, of a famous ancient member of that kema, named Polika, haunted a beach opposite Mage, and a large snake, poli, was believed to represent him there. The Nggaombata could not approach that beach, Polika was their buto[8]. On another beach where they catch fish wherewith to sacrifice to Polika is a gima, a clam, which they call Polika, and used to believe to be in some way Polika; hence the gima in their buto.

There will occur at once the question whether in this we do not find totems. But it must be asked where are the totems? in the living creatures after which two of the divisions are named, or in those creatures which the members of the several divisions may not eat? It is true that the Kakau kindred may not eat the crab kakau; but the Manukama may eat the bird manukama. If there be a totem then it must be found in the buto; in the pigeon of the Manukama and the giant clam of the Nggaobata, which are said to be ancestors. But it must be observed that the thing which it is abominable to eat is never believed to be the ancestor, certainly never the eponymous ancestor, of the clan; it is said to represent some famous former member of the clan, one of a generation beyond that of the fathers of the present member of it, a kukua. The thing so far represents him that disrespect to it is disrespect to him. The most probable explanation of these buto may indeed throw light upon the origin of totems elsewhere, but can hardly give totems a home in the Solomon Islands. The buto of each kema is probably comparatively recent in Florida; it has been introduced at Bugotu within the memory of living men. It is in all probability a form of the custom which prevails in Ulawa, another of the Solomon Islands. It was observed with surprise when a Mission school was established in that island, that the people of the place would not eat bananas, and had ceased to plant the tree. It was found that the origin of this restraint was recent and well remembered; a man of much influence had at his death not long ago prohibited the eating of bananas after his decease, saying that he would be in the banana. The elder natives would still give his name and say, 'We cannot eat So-and-So.' When a few years had passed, if the restriction had held its ground, they would have said, 'We must not eat our ancestor.' This represents what is not uncommon also in Malanta near Ulawa, where, as in Florida also, a man will often declare that after death he will be seen as a shark.

These divisions, kema, are not political divisions[9]. It is not, as in the Banks' Islands where every house must needs contain members of both divisions, that every kema will be represented in every village, for one or two of the smaller may have no member there; but every man's wife, or wives, and all his children, must needs be of a kema different from his own, and every village must have its population mixed. The property of the members of each kema is intermixed with that of the others. In a considerable village the principal chief is the head of the kema which predominates there, and he exercises his authority over all, while the principal men of the less numerous kema are lesser chiefs. It is evident that the predominance of any kema cannot be permanent. A chief's sons are none of them of his own kin; and, as will be shewn, he passes on what he can of his property and authority to them. If then in a certain district one kindred is now most numerous, in the next generation it cannot be so, for the children of those now most numerous will be naturally many more in number, and will none of them be of kin to their fathers. Thus it was that twenty years ago the Nggaombata was the dominant kema in Florida, and to be a great chief it was said that a man must be Nggaombata; but now the Manukama are rising into the chief place, and supply the chiefs in many districts of the island.

The system by which the Melanesian people are thus divided into exogamous groups in which descent follows the mother, receives of course the name of a Matriarchal system; but it must be understood that the mother is in no way the head of the family. The house of the family is the father's, the garden is his, the rule and government are his; it is into the father's house that the young bridegroom takes his wife, if he has not one ready of his own. The closest relationship, however, according to native notions, is that which exists between the sister's son and the mother's brother, because the mother who transmits the kinship is not able to render the service which a man can give. A man's sons are not of his own kin, though he acts a father's part to them; but the tie between his sister's children and himself has the strength of the traditional bond of all native society, that of kinship through the mother. The youth as he begins to feel social wants, over and above the food and shelter that his father gives him, looks to his mother's brother as the male representative of his kin. It is well known that in Fiji the vasu, the sister's son, has extraordinary rights with his maternal uncle. The corresponding right is much less conspicuous and important than this in the Melanesian Islands west of Fiji; but it is a matter of course that the nephew should look to his mother's brother for help of every kind, and that the uncle should look upon his sister's son as his special care; the closeness of this relation is fundamental. The connexion of kinship through the mother with the great exogamous group, and that of blood through the father with his family, thus stand in clear recognition, and to a certain extent necessarily conflict one with another. The connexion caused by marriage between members of the groups and families is a third relation equally felt and expressed in words. The terms therefore in which the various degrees of relationship are conveyed fall into three classes; the first of the kinship through the mother, the second of the family generally on father's and mother's side, the third those following on marriage.

A complete view of the system of relationship with the terms that express it, in any one native field in Melanesia, cannot indeed be taken to shew what everywhere prevails, but as giving a representative example is very valuable; the Mota system, which may well stand for that of the Banks' group, can perhaps be shewn completely and exactly.

(1) It has been said that all the members of each of the two exogamous divisions of the people are sogoi, that is of kin, to one another; the only other relation belonging to this kinship is that between the maternal uncle and his sister's children, male and female, expressed in the terms maraui and vanangoi. The uncle is maraui to his sister's child, the nephew or niece is vanangoi to the mother's brother; but the nephew is also called maraui to his uncle. The relation passes on to the second generation; the children of a man's sister's daughters are his vanangoi, they are still of his kin; but his sister's son's children are of the other veve, the special tie of kindred is broken; they are called his children, being brought up to stand in the same generation with their parents. A man's sister's child, his vanangoi, stands as if in the same generation with himself.

(2) Putting aside connexion by marriage, and the special relation of the maraui and vanangoi, which follows upon the passing of kindred through the mother, relationship generally can be arranged in four successive stages of generation; the grandparents, the parents, the children, the grandchildren. Take the present generation, tarangiu, of young married men and women; they are brothers and sisters; the generation above them are their fathers and mothers; the generation below them are their children; the generation below that will be their grandchildren, to whom again all who come before their parents are grandparents and ancestors. The terms tamai and veve must be translated by father and mother, and are used generally to all of the same generation with the parents who are 'near' and belong to the family connexion. A child, son or daughter, is natui; grandparent and grandchild, ancestor and descendant, is tupui[10]. The terms equivalent to brother and sister are used on a different principle from that with which we are familiar, and according to which the sex of the person referred to determines the use of the word. In Melanesia, as elsewhere, one word describes the relationship of persons of the same sex, and the other word describes the relationship of persons of different sexes. Men are tasiu to men, and women tasiu to women; men are tutuai to women, and women tutuai to men. There is a further difference, the sex being the same, the elder man or woman is tugui to the younger, the younger man or woman is tasiu to the elder; but tasiu is the prevailing use. It may be observed in this system of terms of relationship that all of one generation, within the family connexion, are called fathers and mothers of all the children who form the generation below them; a man's brothers are called fathers of his children, a woman's sisters are called mothers of her children; a father's brothers call his children theirs, a mother's sisters call her children theirs. Upon this it has to be remarked that this wide use of the terms father and mother does not at all signify any looseness in the actual view of proper paternity and maternity; they are content with one word for father and uncle, for mother and aunt, when the special relation of the kinship of the mother's brother does not come in; but the one who speaks has no confusion as to paternity in his mind, and will correct a misconception with the explanation, 'my own child, tur natuk; his real father, tur tamana; tur tasina, his brother not his cousin[11].'

(3) A general term qaliga embraces all of the other side of the house who have been brought near by marriage, fathers-in-law, mothers-in-law, sons- and daughters-in-law, and all their brothers and sisters. A man and his wife's brother call one another wulus, and a woman and her husband's sister call one another walu; but the man is also called walu; and both terms are extended to the cousins of the husband or wife. A woman does not call her husband's brother her brother-in-law; she is nothing to him, though her children, being his brother's children, are called his. A man calls his daughter-in-law tawarig. There is, moreover, a term of marriage relation to which no equivalent exists in English; parents whose children have intermarried call one another gasala, which may be translated fellow-wayfarers.

A genealogical table or pedigree of a Mota family (see p. 38) will supply examples of the various relationships subsisting, and make clear the application of the various terms. The two veve, the two sides of the house, are distinguished by the letters A and B for males, a and b for females. All A and a, B and b, are sogoi respectively, as belonging to the same side of the house; and as besides they are 'near' to one another by blood, they will call one another tasiu and tutuai when the relationship strictly conveyed by those words is absent. The prefix Ro marks a feminine name. The points in the pedigree marked with asterisks require some explanation, but are almost entirely covered by the principle that a man's sister's son, his vanangoi, takes his place in the family on the same


(A) Taqale
Ro Sava (b)
(B) I Gene
(a) Ro Gene
(a) Ro Gene
(B) Leveveg
(B) Rivlava
(a) Ro Gene
(b) Ro Tapermaro
(A) Womele
(A) Siplaglano
(A) Arisqoe
(a) Ro Wawawol
(B) Matevagqoe
(a) Ro Talevui
(B) Sagroro
(A) Marostuwale
(B) Pantutun
(a) Ro Milerawe
(b) Ro Maututun
(A) Virsal
(A) Tavrowar
(A) Mowur
(A) John
(a) Agnes
(B) Dudley

(1) Taqale is tamai to 2, 3, 4, who are natui to him.

tupui to 10, 11, 12, 15, 16, 17, 20, 21, 22, 23, 24, who are tupui to him and his wife.
gasala to 6, who is gasala to him.
qaliga to 5, 7, 9, and to 8, all qaliga to him.

(2) Leveveg is tamai to 10,1 11, 12, 15, and 22, 23.*

rasoai to 9, who is rasoai to him.
tugui to 3, who is tasiu to him.
tutuai to 4, who is tutuai to him.
maraui to 16, 17, and 24,* who are vanangoi to him.
wulus to 7, 8, who are wulus to him.
tupui to 20, 21, who are tupui to him.
qaliga to 13, 14, who are qaliga to him.

(4) Ro Tapermaro is natui to 1.

rasoai to 7.
tutuai to 2, 3, and 13.*
veve to 16, 17, 10, 11, 12, and to sister's children.
tupui to 20, 21, 22, 23, 24.
qaliga to 6, who calls her tawarig.
nothing to 8.
walu to 5, 9.

(8) Sipaglano is tamai to 16 and 17.

(13) Matevagqoe is qaliga to 2, 9, 3, 5, 7.

tutuai to 4.*
wulus to 10.
nothing to 12, 14.

(16) Pantutun is tupui to 20, 21.*

vanangoi to 2 and 3.
tamai to 10, 15.*

(11) Ro Wawalol is natui to 2, 3, 4.

tasui to 12.
tutuai to 10.

(17) Ro Maututun is veve to 15.*

(24) Dudley is tamai to 22, 23, but calls them tasiu and tutuai.*

(22) John is tamai to 20, 21, but calls them tasiu.*

(15) Marostuwale is tasiu to 10.

tutuai to 11.
natui to 16, 17.

(20) Tavrowar is tasiu to 21. level with his uncle, maraui, as if in the place of his mother. Thus Leveveg is in fact great-uncle to John and Agnes, but counts as uncle only because they are grandchildren of his sister. The grandchildren of his brother are his grandchildren, tupui, that is his great-nephews and nieces. For the same reason Leveveg, who is in fact maternal great-uncle to Dudley, counts as his maternal uncle, maraui, Dudley ascending into his mother's place. So Pantutun is first cousin to the mothers of Tavrowar and Mowur, and, being of the generation above them, would be called father or uncle, tamai, and they his children, if it were not that he is cousin to their mothers through his mother, whose place therefore he takes on the second ascending step, and becomes tupui, great-uncle. Thus he is father, tamai, that is uncle, to his first cousins Arisqoe and Marostuwale; and his sister Maututun is their mother or aunt; because he ascends into his mother's place, who was their father's sister. The same rule makes Dudley father or uncle properly to his first cousins John and Agnes, though, as they are of the same generation and older than himself, he calls them improperly brother and sister: improperly, because they are not his sogoi, and he could in strictness, though not with public approval, marry Agnes. It is still more remarkable that John is properly father or uncle to his second cousins Tavrowar and Mowur, who are much older than himself; but his father Pantutun is their great-uncle, tupui, and he is therefore their uncle, tamai, or as it naturally sounds to us their father[12]. The case of Matevagqoe and Ro Tapermaro is distinct from this: he married her brother's daughter, and to do that must have been of her side of the house, her sogoi. If it had been her sister's daughter, she and her niece's husband would be qaliga; but that cannot be between sogoi, so they call themselves cousins, brother and sister.

The pedigree here exhibited does not shew the polygamy which existed in its early stages, and it may be asked whether the terms of relationship would not undergo some change in such a case; whether, for example, the sons of the same father by two mothers would not be distinguished from the sons of the same father and mother. The answer is that no difference is made. A man's wives, if he should have many, must all be sogoi, of the same side of the house, calling one another sisters, and calling each the other's children hers, whether they were married to the same man or had different husbands. This does not however shut out altogether the relationship of step-father and mother. A man who has a son by one of his wives who is dead, does not bring in a step-mother to the boy if he adds another to his living wives; the woman would come in as another mother, and the boy would take no notice. But if a woman with children loses her husband, and becomes the wife of a man who is not 'near' to her previous husband, being of course sogoi but with no recent blood relation, the man will come in as step-father, and the term usur, successor, is applied to him, the connexion being called usur-gae, bond of succession. A looser connexion than this is enough to make an usur as when a boy's father has had a wife, not the mother of the boy, who after becoming a widow marries another man; the boy will take liberties with the man as having come into his father's place; he will take yams from his garden. When a step-father sneezes the step-son will cry out, Matia revereve gam o sulate! a sneeze to draw out a worm for you! the notion being that the former husband has a certain grudge against his successor, and sends a worm from a point of land on which ghosts congregate.

Where, as in Florida and the neighbouring parts of the Solomon Islands, the divisions of the people are three, four, or six, and where a man may have a wife or wives from any one of them but his own, it would seem likely to be more difficult to keep accurate count of the various degrees of relationship in which people stand to one another; and it is probable that, though the native system is precise in following every step and connexion, the people do in fact content themselves commonly with general terms. The special relation of the sister's son to his mother's brother is of course conspicuous; each calls the other tumbu; and this term is applied also to the father's mother's brother by his grand-nephew, and by the great-uncle to his sister's grandchild. In a generation of members of the same kema all of them call one another hogo in the same sex, and, with more or less attention to nearness of blood, brothers and sisters; that is to say, an elder brother or sister is tuga to one of the same sex, and a younger brother or sister is tahi, while a brother or sister is vavine to one of the other sex. With the exception of the mother's brother, the blood relations of the ascending generation are all father and mother, tama and tina. In the generation above, with the exception of the father's mother's brother aforesaid, who is tumbu, all male and female are kukua. In descending a man's sons and daughters, and his brother's and cousin's children, are dale, distinguished as dale mane and dale vaivine, according to sex, a man's sister's child being tumbu and in the same way a woman's children and her sister's and female cousin's children and her husband's brother's and sister's children are all her children, dale mane male, dale vaivine female. Descending to the next generation, all are again kukua to their grandparents and great uncles and aunts, and all above them; except that, as aforesaid, the relation of tumbu subsists between a great-nephew and his father's mother's brother. Husband and wife are tau. A father- or mother-in-law, and son- or daughter-in-law, is vungo, the term being applied widely to persons connected by marriage who are not of the same generation. Brothers- and sisters-in-law, and generally persons of the same generation connected by marriage, are iva to one another[13].

It would seem that the absence of exogamous divisions of the population in that region of the Solomon Islands in which descent follows the father (namely, in Malanta, about Cape Zelee, in Ulawa, and in San Cristoval), must make the system of family relationship very different there from that which has been described as prevailing in the Banks' Islands and in Florida. To a very considerable extent no doubt this is so; but it is improbable that the peculiar closeness of relation between a man and his sister's son should entirely fail to appear. Of this I have little evidence to offer[14]; the families are formed upon the father, and the only restriction upon marriage is nearness in blood. To whatever extent, however, it may be that descent through the father removes that characteristic feature of the Melanesian family system which appears in the relation between the maternal uncle and his sister's child, it is certain that the main structure is the same as elsewhere; that is to say, that no terms corresponding to uncle and aunt, nephew and niece, or cousin exist. All on the same level are brothers and sisters, if children of brothers and sisters or of cousins; they look upon the children of brothers, sisters and cousins as their children, and the children call them all fathers and mothers; the ancestors above father and mother, and the descendants in the second and lower generations, are all united under one general term, which covers ancestry and posterity alike. At Wango in San Cristoval, where owing to immorality and infanticide the population has been kept up by the adoption of children from the bush, adopted children take the position in the family which would have been theirs if they had been born in it; although no blood relationship exists, they cannot marry those who are near through the adoptive father. These children appear to be by traders called slaves because they are bought; the people themselves call them their children.

The subject of marriage relations is incomplete without notice of the reserve so remarkably exercised towards the persons and names of those who have become connected by marriage. This is conspicuous in the Banks' Islands, and makes but little show in the Solomon Islands. In Lepers' Island, a singular reserve is strongly shewn, as it is in Fiji, by brothers and sisters, and also by mothers and sons; but this reserve, though its existence and its cause may well throw light upon that exercised between those connected by marriage, has no proper place here. In Florida, in the Solomon Islands, there is no difficulty about meeting, or mentioning the name of, father- or mother-in-law, or any of a wife's kindred, and no extraordinary marks of respect are shewn. It is the same at Saa. The extraordinary separation of the sexes in Santa Cruz and the neighbouring islands, however instructive to observe in this connexion, does not follow on relation by marriage. In the Banks' Islands the rules of avoidance and reserve are very strict and minute. As regards the avoidance of the person, a man will not come near his wife's mother; the avoidance is mutual; if the two chance to meet in a path, the woman will step out of it and stand with her back turned till he has gone by, or perhaps if it be more convenient he will move out of the way. At Vanua Lava, in Port Patteson, a man would not follow his mother-in-law along the beach, nor she him, until the tide had washed out the footsteps of the first traveller from the sand. At the same time a man and his mother-in-law will talk at a distance. A man does not avoid his father-in-law, nor a woman hers. A man does not avoid his wife's brother, but will not sleep with him; he does not avoid his son's wife, or his own wife's sister. Boys and girls who are engaged generally avoid one another, but through shyness, not by rule. Where the persons above mentioned do not avoid one another, they are careful to shew respect in not taking anything from above the head or stepping over the legs of a father-in-law or wife's brother. It is disrespectful at all times for a young man to take anything from above an elder man's head, for there is something naturally sacred, rongo, about the head, and no one will take the liberty of stepping over the legs of any but a brother or intimate friend. To avoid the mention of a name shews a lower degree of respect than to avoid a person. A man who sits and talks with his wife's father will not mention his name, much less his wife's mother's name; a man will not name his wife's brother, but he will name his wife's sister, she is nothing to him. A woman will not name, but does not avoid, her husband's father; she will on no account name her daughter's husband. Two people whose children have intermarried, who are gasala, will not name each other. The reserve with regard to the name extends to the use of it, or of any part of it, in common conversation. A man on one occasion spoke to me of his house as a shed, and when that was not understood, went and touched it with his hand to shew what he meant; a difficulty being still made, he looked round to be sure that no one was near and whispered, not the name of his son's wife, but the respectful substitute for her name, amen Mulegona, she who was with his son, and whose name was Tawurima, Hind-house[15]. Thus, referring to the Mota pedigree given on page 38, Leveveg could not use the common words mate, to die, or qoe, pig, because of his son-in-law Matevagqoe; Virsal could not use the common words panei, hand, or tutun, hot, because of his wife's brother's name, or even the numeral tuwale, one, because of his wife's cousin's name. To meet the difficulty caused by this limitation of vocabulary, a word may be used improperly like paito, shed, for ima, house; or a knife may be called a cutter and a bow a shooter; but there is a stock of words kept in use for this very purpose, to use which instead of the common words is called to un. Thus the un words used in the cases mentioned above would be karwae for qoe, saproro for mate, lima for panei, val for tuwale[16]. This avoidance of the person and of the name is ascribed by the natives themselves to a feeling of shyness and respect, a certain inward trembling which they say prevents their mentioning their own names also; to blurt out a name is to take a liberty, to avoid the use of it shews delicate respect, and one will extend this respect to more distant connexions rather than apply it too narrowly. A native when asked the name of some other, will often turn to some bystander who answers for him, and the explanation is given in the one word qaliga. Respect is also shewn in Mota by using a dual pronoun in addressing or speaking of a single person; 'Where are you two going?' is asked of a qaliga, as if both husband and wife were present.

In the New Hebrides the practice is much the same. In Lepers' Island a man speaks to his mother-in-law, and she to him, but they will not come near; when he speaks to her she turns away. A mother-in-law or father-in-law does not mind using the name of daughter's husband or son's wife in speaking of them to others, but cannot use it in addressing them. When a woman calls to her son-in-law she addresses him as mim, you in the plural; when she sends a message to him she says, using his name, 'They want Tanga to go to them', that is, 'I want Tanga to come to me.' A daughter-in-law does not avoid her husband's father, a man sends his wife with messages to his father. A man will not speak at all the name of his wife's brother; speaking of him he says, 'my brother-in-law,' speaking to him he says, 'you' in the plural; if he meets him in the path he turns aside, and asks 'Where are you (plural) going?' In this case only it appears that the name is never spoken; the reserve among connexions by marriage is much less marked than that between brother and sister. No one will step across the legs of another, or take anything from over his head, especially a brother's; that is thought a serious piece of disrespect. In the neighbouring island of Araga, Pentecost, the intercourse of fathers- and mothers-in-law with their daughter's husband or son's wife is very little restricted; the chief, if not the only, reserve in speaking is exercised by engaged couples before the giving of property for the girl is complete; this is called lalag.

  1. 'Descent is still uterine in some parts of Fiji; most of the tribes, however, have advanced to agnatic descent.'—Rev. L. Fison.
  2. A Lepers' Island youth staying at Mota was delivered from some little difficulty with the remark, O tanun we wia gai, gate tanun ta Qauro, He is a man of the right sort, not a Solomon islander.
  3. The Lo Sepere family of Vanua Lava is the same with the Tupueviga of Gaua and the Anamele of Mota. On the other side of the house the Tapulia of Gaua and Merlav are counted the same with three groups at Mota, viz. Alo Gapmaras of Takelvarea, the Wotawota of Maligo, and the Liwotuqe of Gatava. These family groups lie within the veve, but do not take in all the veve; neither side of the house is exhaustively divided into family groups.
  4. To these lesser divisions or family groups my informant (A. Arudulewari) gives the name of veve, as to the two great kindreds. For example, he and Walter Gao are of the Wirita family, Tarisuluana is of the Ta Wongi, a place now deserted; Vile is Ta Lau of the beach, Tilegi of Suwumea.
  5. As in the case of Dudley and Agnes in the Mota pedigree further on.
  6. This is illustrated by the case of Alfred Lombu, who, returning from Norfolk Island in search of a wife, proposed for a daughter of Takua, the chief of Mboli. The girl was not of the same kema in name with Lombu, and he maintained that he was not aware that his kema and hers were in fact the same; but Takua imposed upon him a heavy fine, seeing an opportunity for possessing himself of the money accumulated for the marriage, and professing great indignation at the outrage on propriety.
  7. Thus in 'Percy Pomo' a man is horrified at seeing blue trousers, the colour of some part of the inside of the shark, which was his buto.
  8. Na butodira Gaobata na tidalo eni, That ghost is the buto of the Nggaombata. The origin of the prohibition is respect for Polika; those of his kema would not intrude upon the beach he haunted, nor would they eat the clam, because the clam on the reef represented him. They have now looked in vain for the snake.
  9. When some outrage on white men has been committed the 'tribe' is supposed responsible; but any party of natives concerned is sure to be made up of members of both veve or several kema, and some of these probably do not belong to the place where the outrage is committed. Of the five natives who cut off the boat at Mandoliana in 1880, only two were of the same kema, and only one was at home at Gaeta.
  10. It may be observed that the principal terms of relationship are generally the same, not only in the Melanesian islands here in view, but throughout the languages with which the Melanesian languages are connected; mother being an exception. Common words however are not always used in the same application, as the Florida tubu is no doubt the Mota tupu.
  11. Before the native use is well understood it is certainly perplexing and misleading. As an example, a boy named Tarioda came from Araga to Norfolk Island. Remembering a youth of the same name from the same island, I enquired if he had anything to do with him; the boy answered that he was hi father, and that he had seen him and knew him, meaning that he was a cousin of his father's. Such an answer might well be the ground of a statement that paternity was very little thought of in the New Hebrides. English people probably had perfectly clear conceptions about family ties before they used the words uncle, aunt, cousin, nephew and niece.
  12. It sometimes happens that a boy is in this way 'father' to one old enough to be his natural father, or 'grandfather,' tupui, to one of his own age. When it is so the formal relationship is practically merged in the general tasiu, brotherhood.
  13. The word mavu, which is used for 'namesake,' is also used as a term of family relationship. Unfortunately the full list of Florida terms made by me many years ago lacks a key.
  14. From Rev. R. B. Comins I learn that at Wango and Fagani in San Cristoval the term for the relation between the maternal uncle and his sister's child is mau. The terms iha, ifa, hungo, fungo, are the Florida iva and vungo; 'ama, 'ina, 'asi, are tama, tina, tahi.
  15. The word amaia, with him, is used not only for a wife's name but in place of 'his wife'; nan amaia wa, then said his wife. In the case referred to, Tawurima, the name of the daughter-in-law, contains the word ima, house. The father of Tawurima, again, could not use the common word for to go, mule, because it is part of her husband's name, Mulegona.
  16. These un words are particularly valuable, because they often shew a connexion with other languages which does not appear in more common words. Words are not invented for this purpose; words are taken which lie comparatively unused in the language.