Volume 88 1979 > Volume 88, No. 3 > Kindred and alliance on Anuta Island, by Richard Feinberg, p 327-348
                                                                                             Previous | Next   

- 327

Anuta is an isolated Polynesian outlier 75 miles north-east of Tikopia in the Solomon Islands. 1 Its population of approximately 200 people 2 is divided into 19 elementary domestic units or patongia. Ideally these patongia form patrilateral extended families and are the smallest property owning, producing, and consuming units on the island. 3 They operate as semi-independent entities in most day-to-day activities. Often, however, assistance and co-operation among patongia are necessary. There are several ways in which they may be brought together into larger groupings. One of these is through ties of “descent” from a common “ancestor”, as in the case of the kainanga ‘clan’; 4 a second is through bonds established among members of the various patongia on the basis of their relationship to a common ego. In this latter category is the kindred and affinal alliance system.

Kinship on Anuta is intimately bound up with aropa, a word denoting positive affect as manifested through material assistance and cooperation. 5 It is expressed most potently among close kin, particularly within the same patongia, where it is seen in common ownership of property. Between patongia it is expressed through pooling of resources or the exchange of goods and labour. 6

Owing to Anuta's isolation and small size, a high rate of island endogamy, 7 and mechanisms for incorporating immigrants into the kinship system, each Anutan has a kin relationship with every other. Consequently, every Anutan is potentially a member of the kano a paito ‘kindred’ of every other person on the island and may be mobilised to provide economic support in various situations. In one sense, interaction among kin is phrased in dyadic terms. However, since the patongia is the smallest unit of property ownership, any economic action that a person takes involves his entire domestic unit, of which he may be seen as a representative. And the closer the kin tie that two people share the more powerful will be the bonds of aropa (and, therefore, the more frequent and intense the level of assistance and exchange) both between them as individuals and between their patongia..

While an Anutan's kin include the total population of the island, the - 328 ‘kindred’ is subdivided into a number of discrete categories. In addition to the basic kin classes (e.g., tamana ‘father’, pae ‘mother’, etc.), Anutans distinguish between maori ‘near’ and pakaapaapa ‘distant’ kin. 8 The ‘kindred’ proper is frequently distinguished from the ‘kindred on the woman's side’ (this includes maternal relatives and affines). In many rites of passage all of the subject's ‘fathers’ (“classificatory” as well as “real”) act together as a group, the ‘father's sisters’ constitute a second group, the ‘mother's brothers’ constitute a third, and the ‘grandparents’ a fourth. 9 Each of these groupings has a focal member who acts as its primary representative in dealing with the initiate and his patongia, particularly in exchange of presents. Both the prestations made on the initiate's behalf by his patongia and the focal member of each section of the ritual kindred are known by the term inati. The inati are always in a different patongia from the initiate and are often only distantly related to him genealogically. By serving as the primary recipients of these ritual prestations, however, they and all members of their patongia are transformed into ‘close’ kin of the initiate and his domestic unit.

Since a man's affinal relatives are his child's maternal kin, kindred and alliance are structurally interdependent and may be treated as aspects of a single system. The Anutans have no positive injunction stipulating a specific class of people into which one should marry, but only a proscription forbidding marriage to close kin. Moreover, there is no clear cut-off point, but rather a gradation of propriety — except for ‘siblings-in-law of opposite sex’, who are mildly preferred as spouses, the closer the relationship between two people the less acceptable they are as marriage partners. Aside from persons to whom one is related closely in genealogical terms, kin too close to be considered as potential spouses include anyone who has been ‘adopted’ into ego's patongia, members of a patongia into which ego has been ‘adopted’, anyone with whom ego has established a ‘bond-friend’ relationship, 10 the primary recipients of the inati prestations during ego's life crisis rites, anyone whom ego has served as primary recipient of the inati prestations, and all members of the patongia involved in these transactions. Thus, such a large proportion of the populace falls within the prohibition that despite the absence of a positive regulation, the virtual equivalent of an explicitly defined marriage class is created.

Marriage transforms distant “consanguineal” relationships into close affinal ones. 11 A man's closest affines are his children's nearest consanguines ‘on the woman's side’, and the group which was, for him, the most appropriate from which to take a spouse becomes among the least appropriate for his children. In this way a cycle is produced which ties together all of the patongia through an intricately interwoven set of marriages, maximising solidarity by continually reinforcing those bonds which are the most tenuous. 12

- 329
Most commonly used Anutan kin terms and their usual application according to genealogical criteria.
- 330

The Anutan phrase, te kano a paito, may be used with any of several different referents. It may denote a patrilineal ancestor-focused unit of any degree of genealogical depth. It may refer to the aggregate of paternal kinsmen, and it may be used, with qualifications, to designate kinsmen on the mother's side as well.

Boundaries of the kano a paito shift with the context in which one is speaking. I have had informants tell me that any kinsman, however related and however distant, is included in his kano a paito, while at other times I have been told that the kano a paito refers only to the individual patongia. I have been informed that relatives by marriage and maternal kin are not members of one's kano a paito while other informants have insisted that they are, although with the reservation that te kano a paito e tai i te pai o te papine ‘the kano a paito is the same on the side of the woman’ (i.e., affinal and maternal kin are in ego's kano a paito, but on the woman's side).

The kano a paito, then, is not a group with constant borders. Anutans disagree on who should be included and who does not belong, and even the same informant may be inconsistent from one occasion to the next. When conceptualised in economic rather than genealogical terms, however, much of the confusion disappears. If one asks an Anutan why his patongia has joined forces with another to form a co-operative grouping known as te ngutuumu 13 he probably will answer, “Because they are my kano a paito,” and in describing a ceremonial feast or ritual exchange he is likely to explain that he was assisted by contributions of food, koroa (durable goods such as mats, fishhooks, paddles, or clothing), or both from members of his kano a paito. On different occasions a man or woman might be aided by different relatives. Sometimes only the closest kin may come to assist, although for more important rites the most distant kinsmen may help as well. Also, a popular and influential person will find it easier than someone who is less so to mobilise his distant kin.

Most important rites involve an exchange of food, and sometimes durables. A person, assisted by his paternal kin, usually comprises one side of the exchange; his maternal kin comprise the other. The paternal kin, therefore, express their aropa ‘love; solidarity’ for ego through a pooling of resources while he and his maternal kin express their ties by means of reciprocity. Pooling characterises ego's “own” group and differentiates it from the “others”, with whom he interacts on the basis of reciprocity. Yet, pooling and reciprocity are aspects of a single integral process. Ego's group (ego and his paternal kin) pool their goods while the “other” group pool theirs. The two groups then come together and exchange their goods in an act of reciprocity. This relationship is recapitulated on the terminological level by the counterposition of te - 331 kano a paito (which, unless qualified, is assumed to mean on the paternal side) on the one hand, and te kano a paito i te pai o te papine ‘the kano a paito on the woman's side’ on the other.

The kano a paito is formed by an intersection of genealogical and behavioural principles. Any kinsman is potentially a member of one's kano a paito, but he is only such in fact when he acts the part by cooperating with ego's patongia in the economic sphere, either through pooling resources or by participating in a reciprocal exchange. Both forms of economic action are taken as expressing aropa, and I was told explicitly, “If you say that someone is of a different kano a paito it means you do not aropa to him; if you say, ‘the kano a paito is the same’, it means you aropa to him.” It is no coincidence that the closest thing Anuta has to a generic term for “kinsman” is “my kano a paito.”

Firth (1963:213-7) describes the Tikopian kano a paito as a bilateral kindred. This designation also is apt for the Anutans as long as it is remembered that there is a fundamental cleavage into sections which are (in ritual contexts) inherently opposed. Moreover, one of those sections, te pai o te papine, is only included in the kano a paito with some major reservations.


All life crisis rites involve exchange of food, and often of durables, among the various sections of one's bilateral kindred. In most exchanges, however, one person in each section is singled out to represent his section, act as the primary centre of attention, and receive the major presents. Both these focal persons and the presents they receive are termed inati. The process is designated iki nga inati ‘carrying the inati’, a term which also may be used for the persons to whom the gifts are presented.

On most occasions the inati are three: te pai maatuaa ‘the side of the parents’ (which usually means the paternal side), te pai makitanga ‘the side of the ‘father's sister’’, and te pai tuatina ‘the side of the ‘mother's brother’’. 14 A crucial feature of this arrangement, however, is that the maatuaa, makitanga, and tuatina do not refer, in this context, to the “real” father, father's sister and mother's brother. Rather, the kindred is structured in such a way as to maximise the diversification of ties among patongia and reinforce ties within the kainanga ‘clan’, especially between the ‘clan’ leader's patongia and the other units of his ‘clan’. 15

Normally the pai maatuaa prestation is taken to the head of ego's ‘clan’, or rather, to the senior man in ego's father's generation who is in the patongia of his ‘clan’s' leader. If this man is absent from the island, or if he is deceased, his eldest available brother or son is substituted.

- 332

The inati cannot be carried to a member of one's own patongia, and therefore, if ego is a member of the chief's (or ‘clan’ leaders) patongia, the prestation must be made to another group within the ‘clan’. Neither may the inati be given to a member of a patongia with one or more of whose members ego has a very close genealogical link. Moreover, marriage to the child of one's own inati is a violation of rules regarding incest and exogamy. Yet, on occasion, it occurs and when it does, the prestation to be made on behalf of one's own children must be carried elsewhere to bring things back into line.

These considerations may be illustrated by the case of Pu Paone. Pu Paone is a member of the Kainanga i Mua, the senior ‘clan’. The appropriate person to receive the pai maatuaa gifts to be made on his behalf, then, was the father of the present senior chief. When Pu Paone reached adulthood he married the present chief's sister, and consequently his children's pai maatuaa prestations cannot be brought to the chief's patongia. Neither are they brought to the three units headed by Pu Paone's patrilateral parallel first cousins, since they all formed a single patongia until recently, making these units too closely related to serve as inati to one another. This leaves two remaining patongia in the ‘clan’. As far as I can tell, the decision as to which of these was to receive the pai maatuaa prestations given on behalf of Pu Paone's children was made purely on the basis of personal preference.

Personal likes and dislikes provide a final basis for altering the pai maatuaa prestation. If ego's parents are angry at a member of the patongia to whom the inati normally would be taken they may express their feelings by carrying their presents elsewhere. This is rare in the first two ‘clans’ as such behaviour would mean slighting a chief. In the nonchiefly Kainanga i Pangatau it is more frequent. The fourth-ranking Kainanga i Rotomua has only one patongia and its pai maatuaa prestations, for several generations, have been brought to one of the middleranking units in the Kainanga i Pangatau. Regardless of how the choice is made, however, the relationship is passed down in the paternal line. If A's pai maatuaa is carried to B, that of A's son is carried to the son of B, that of A's paternal grandson is carried to the paternal grandson of B, and so on. This relationship is maintained indefinitely, until such time as it is ended by an inappropriate marriage or personal feud, at which point a new and equally enduring relationship is formed.

The recipient of the pai makitanga prestation ordinarily is the eldest sister of the pai maatuaa, although if this man has no sister or close female cousin the present may go to his eldest daughter or the daughter of his brother or male cousin — whoever is the most senior female to have been born into his patongia. My data show a few exceptions to this rule, but they are rare and seem to be short-lived. The pai tuatina represents the mother's side and is normally the son of the man to whom - 333 her pai maatuaa was presented, although again the pattern may be interrupted by personal incompatibility or a poorly chosen marriage.

Before the birth of their first child the parents and their patongia have a degree of flexibility in determining where their children's inati prestations will be taken. Once the birth of the eldest child has been marked by a ritual known as the pai panaunga, however, the inati of all the couple's children are set for the remainder of their lives. At all future rites of passage the pai maatuaa, pai makitanga, and pai tuatina presentations will be carried to the same patongia, and, if possible, to the same persons as they were during the panaunga ceremony. This ritual is performed only for the first-born son and the first-born daughter, and in most other rites junior siblings are simply pakapipiki ‘stuck’; ‘attached’ on to a senior sibling or cousin, the recipient of the inati prestations depending on the principal subject of the ceremony. On occasions where a ritual relationship must be invoked for a junior sibling directly, as in the exchanges surrounding his marriage, he follows the patterns set initially at the panaunga of the first-born.

In addition to the pai maatuaa, pai makitanga, and pai tuatina, a fourth section of the inati sometimes is invoked. This is called the pai tupuna ‘the side of the grandparent’, and the presentation is made to the parents of the pai tuatina (the pai maatuaa of ego's mother). In this way a second gift is added to the mother's side to counterbalance the two on the father's, and an Anutan tendency to organise things in terms of paired oppositions is maintained. 16 That the pai tupuna's primary function is to maintain the consistency of a dualistic structure, and not to play a necessary pragmatic role, is suggested by its dispensability. It is not represented at the panaunga ceremony, and in later ones it is always optional. The ritual structure of the kindred, as reflected in the inati, is outlined in Figure 2.

The Ritual Structure of the Kindred
  Father's Side Mother's Side
Central pai maatuaa pai tuatina
Peripheral pai makitanga 17 pai tupuna*
- 334

The account thus far has been presented in highly schematic terms. In point of fact, in different rites different relationships may be important. At times focal relatives or other members of their patongia are crucial. 18 Sometimes all members of a particular kin class may play a collective role, while at others the inati are central. Sometimes exchanges are limited to the paternal side, occurring only between ego's patongia and those of his ‘father's sisters’, and on one occasion — as part of the rite surrounding a boy's circumcision — all the women of his kindred, ‘mothers’ and ‘father's sisters’ alike, work together in preparing food for presentation to the ‘mother's brothers’. In order to present the reader with an idea of the complexity and variation in actual patterns of ritual exchange I shall summarise the major rites of passage performed on Anuta.

Pai Panaunga

Shortly after the birth of a couple's first son (te urumatua tangata), and again after the birth of the first daughter (te urumatua papine), a ceremony called the pai panaunga is held to recognise the mother and her child.

On the first night of an infant's life its ‘father's sisters’ bring presents of durable goods for the parents of the child, and for the next several nights one or more ‘father's sisters’ stay at the baby's house to assist the mother while she regains her strength. When the panaunga ceremony begins, this procedure comes to a halt.

The precise timing of the ceremony depends upon the weather as large numbers of ocean fish are needed for the rite. When an adequate supply has been obtained the populace begins preparing food. The patongia on the father's side work together as a single unit, and those whose closest tie is with the mother do the same. Then both sides bring their prepared food to the child's house where it is distributed by the baby's father, assisted by his close kin. The 19 patongia, the two chiefs, the catechist, and his assistants, each receive a basket as in any normal distribution, but in addition the pai maatuaa, the pai makitanga, and pai tuatina are given at least one basket apiece. Then all the baby's ‘fathers’ gather in the house of the pai maatuaa, the ‘father's sisters’ congregate in the pai makitanga's house, and the ‘mother's brothers’ in the pai tuatina's for a ritual meal. 19 In the one panaunga that I witnessed, 21 baskets of food were presented by the child's paternal kindred and 10 by his maternal kin. The 10 from the mother's side plus nine from the father's were distributed among the 19 patongia. Two baskets for the chiefs, three for the catechist and his assistants, three for the crew of the canoe which provided the fish, three for the three branches of the inati, and one for the - 335 mother of the infant all came from the gardens of the child's paternal kin.

Pangai Ika

Burying the umbilical cord, baptism, and weaning are not marked by major feasts or ceremonial exchange. After the panaunga the next major ritual is the pangai ika, which occurs when the child is fed its first fish at about a year of age. This also is performed only for the first-born child of each sex.

On the first day of the ceremony the child's maternal grandfather (not the inati), or a member of the grandfather's patongia, takes the baby for its first look at the hilltop in a rite known as te pakamatamata ‘the showing around’. In the meantime all of the patongia prepare food which is presented on behalf of the child's father to the man who took the baby to the hill. That evening at least one representative from each patongia on the mother's side feast in the house of the maternal grandfather while the father's kin gather outside to eat. The following morning the child's maternal kin return to feast in the grandfather's house. Before leaving they are given mats and other durables which they reciprocate with equivalent or slightly larger presents.

About one week after these events the mother's kin present the father with food on the maternal grandfather's behalf. The father sends for his close “consanguines” to join him in his house and there is another feast. This time the child's paternal kin eat inside the father's house with others gathering outside. Unlike at the earlier feast, no durables are exchanged.


Sometime during childhood a rite called the angaa is performed. There are several types of angaa, the most important of which is that held for the eldest son and other boys who are ‘attached’ for the duration of the ceremony. This is te angaa pora koroa ‘the angaa of spreading goods’, said to be put on by the father to honour and to demonstrate affection for his son. In terms of time and expense this is the most important rite in the life of a child. 20

The father and his patongia prepare for this rite for many months, planting and cultivating large quantities of food. As the event approaches, members of the child's patongia, assisted by his kano a paito, gather food and bake it in their ovens. While the food is at the ovens the child's pai makitanga, pai tuatina, and pai tupuna send for all the youngster's ‘father's sisters’, ‘mother's brothers’, and ‘grandparents’, respectively, to come to the house where the ceremony is taking place, and eat. The pai tuatina takes the child to the houses of all the ‘mother's brothers’, and at each he is given presents of durables. He brings these back to his father's house and then begins a circuit of his - 336 ‘father's sisters’. They also present durables which he brings back home.

When this is done the youngster's kin all come together in the house of his parents or some classificatory ‘father’ in the boy's patongia, and begin the feast. The ‘mother's brothers' ’ food is piled in the foremost section of the house (te mataapare, the section of highest honour), and that is where they sit to eat. The ‘father's sisters’ gather round their food, which has been piled in the rear (tuaumu, the spot of lowest ritual esteem), while the ‘grandparents’ ' is at the (neutral) ends. The boy, along with any ‘sibling’ who has been ‘attached’, is seated in the centre of the house and shares the food of all three groups of kinsmen. Meanwhile, the boy's parents and classifactory ‘fathers’ spend the day preparing food which has been taken from the gardens of the father and his kano a paito.

For an entire week the feasting continues in the house of the angaa. On the last two days a dance takes place inside the house — an act which normally is tapu. On the final day a dance begins inside the house and then moves inland to the ovens where food has been prepared for the arrival of the party. There the relatives resume the feast. Before they leave they are given large quantities of uncooked food which they all take home, prepare, and in the evening bring back to the house of the angaa. The food presented by the pai makitanga, the pai tuatina, and the pai tupuna, plus that prepared by the father and his kano a paito, may mean that as many as 200 baskets are brought together. The people feast outside the house, and the child partakes with each section of his kindred. At this point the angaa is finished, but sometime later the pai makitanga, pai tuatina, and pai tupuna make return gifts of food to the parents of the child. This is called the aererepanga ki muri.

Two varieties of angaa are held for girls. Both are simpler than the ceremony for a boy. The more complex resembles the boys' angaa except that there is no prestation of durables, the distribution of raw food is not made, and there is no pai tupuna section. In the other version no dancing takes place, the pai tuatina is eliminated along with the pai tupuna, and participants are limited to the father's side — the ‘fathers’ and ‘father's sisters’ of the girl. This is a smaller, more modest affair and is most likely to occur when a scarcity of resources makes the more elaborate form impossible.

Vai Pa

The return of a boy from his first trip to Patutaka, an uninhabited island about 30 miles away, is marked by a rite called the vai pa. Upon his return the initiate is carried by one of his ‘father's sisters’ from the beach to her house where he is washed in a warm infusion of fragrant leaves and fed ritually esteemed foods. For the next two weeks the youngster takes his meals in the house of one of his ‘father's sisters’ (this - 337 may be, but is not necessarily the pai makitanga), until this rite is concluded by a feast called te pakatavanga, sponsored by the ‘father's sister's’ ‘clan’, to honour the initiate. During this two-week period the hostess's patongia is given presents of food from the boy's domestic unit.

Often a boy's angaa is timed to coincide with his first fishing trip in a canoe. If not, an abbreviated version of the vai pa is performed to recognise and sanction this event.


Somewhere between the ages of 11 and 15 years a boy undergoes a rite of circumcision. Like the angaa, this ritual is performed primarily for the eldest son while younger brothers are ‘attached’. Sometimes the angaa pora koroa is postponed until the time of circumcision, in which case the two rites are combined. Otherwise the ceremony consists of another type of angaa, known as te pakavao. 21

The timing of the circumcision rite is determined by the youngster's age and the condition of his patongia's gardens. Normally the operation is performed by the pai tuatina, but should the boy's parents question this man's competence a substitute may be requested. In such a case the substitute should be the most senior available competent man in the pai tuatina's patongia.

If the rite, known as te puru nga kere ‘cleansing the dirt’, is being held in conjunction with the angaa pora koroa, the operation is performed outdoors in the vicinity of the youngster's house; otherwise it is conducted in the bush. All the boy's ‘mother's brothers’ may be present, but no one else may see the operation. Only two men touch the youth, the tipunga ‘surgeon’ and one assistant. 22 The others stand around to watch and give encouragement.

While the operation is conducted the boy's parents prepare puddings. When the food is ready the ‘mother's brothers’ return to the youngster's house and eat. The parents give the ‘surgeon’ a new pandanus mat on which to sleep, and he stays with the parents and their son until the wounds have healed. 23 The boy's ‘mothers’ and ‘father's sisters’ (te pare pae) provide him with more mats and other goods.

During the period in which the youth is recovering from the operation the ‘surgeon’ remains at his side, and the parents supply the pair with food. After a month or so, when the wounds have healed, the ceremony is ended with a rite known as te panopano o nga nima o te tipunga ‘the washing of the surgeon's hands’. On this occasion the whole island contributes food and durables to the surgeon on behalf of the boy's parents, and the parents “wash” the surgeon's hands with turmeric pigment — a symbolic cleansing of ritual pollution from the hands which have handled a defiling organ.

- 338

As is the case with other rites of passage, the marriage ceremony involves exchange of goods. The kindreds of two people rather than just one, however, are involved in the exchange. Moreover, this rite differs from the others in that a human being transfers her primary affiliation from one domestic unit to another. Here I describe the mobilisation and alignment of kin and the exchange of goods in the marriage rites. The structural implications of the marriage system for establishing alliances through the exchange of persons will be discussed in the following section.

There is no formal period of courtship on Anuta. Men and women are not expected to pair off into couples before engagement, and premarital sexual relations are forbidden by the church. Boys and girls come to know one another as they play and work together during childhood and adolescence.

When a man decides that he would like to marry, he selects a woman and consults his parents on his choice. If his parents approve, he sends an emissary to the woman bearing presents of tobacco, bath soap, cloth, or other goods. Most commonly the emissary is the young man's mother or his sister, although the father, brother, or other close relative or friend may go. 24 If the woman is amenable she accepts the presents and may send return gifts to the man. This is called te pakakoroa. At this point the man and woman are formally “engaged”, and people refer to them as te tau toa ‘formal friends’.

The timing of the wedding rites is up to the parents of the couple. On the designated evening the parents of the man send word to members of their son's kano a paito on both sides. The bridegroom's kin bring him gifts of durables (e.g., mats, canoe paddles, fishhooks, etc.) from each of their patongia, to be presented to the bride's patongia. The bride's hair is cut short in the bridegroom's house by his pai makitanga, aided by the pai makitanga's own close female kin. The bride is painted with turmeric pigment and then dressed in a ceremonial bark cloth skirt, provided by the bridegroom's kin.

When these procedures are complete, all the group of kin assembled in the bridegroom's house go to the house of the bride's parents to present their gifts, known as te marae. They press noses with the members of the woman's kindred. After this the couple are called te rumatua ‘a married couple’. 25 Considerations determining the alignment of kin for a marriage rite may be illustrated by the case of Ta Nukurava “the Nukuravas”, outlined in Table 1.

Given Anuta's size and the rate of island endogamy, everyone will most likely be kin to both the bride and bridegroom. Choices must be made, therefore, as to the side with which a person and his patongia will affiliate for purposes of the marriage rite. When someone clearly is - 339 related more closely to one party than to the other there may be little difficulty, but the matter often is ambigious. If one is related closely to both sides, often he may go to one house and have another member of his patongia participate with the other. Alignment based on interpersonal preference may further complicate the picture.

The night of the marae prestation the couple sleep together in the house of the bridegroom. The next morning they are married in the church, and preparations begin for the wedding feast. If the sea is reasonably calm, one or two canoes go out that afternoon to provide fish; if the passage is too rough a communal fish drive is held on the reef. At the same time food is harvested and cooked to be donated to the feast. Each patongia on the island contributes something, but a disproportionate amount comes from the gardens of the bridegroom's patongia. The bridegroom's eldest sister's husband is known as te tokomatua, and he plays a special role, along with the catechist and perhaps the chiefs, in organising these activities.

Alignment of Kin for Marriage Rite of Ta Nukurava
I. Bridegroom's Side  
Kinsman 26 Relationship or Reason for Alignment
Pu Raropuko Pu Nukurava's father.
Nua Raropuko Pu Nukurava's mother.
Margaret Vaerua Pu Nukurava's sister.
Daisy Rarakena Pu Nukurava's sister.
Phyllistus Kirirua Pu Nukurava's sister.
Nau Ropanga Pu Nukurava's closest living relative in the grandparental generation.
Pu Atapu Husband of Pu Nukurava's eldest sister; went to represent his wife.
Nau Rongovaru Wife of Pu Atapu's brother; accompanied her brother-in-law at the instruction of her husband.
Nau Tepuko First cousin of Pu Nukurava's father, Pu Raropuko.
Pu Tepuko (junior chief) Husband of Nau Tepuko; accompanied his wife.
Pu Penuatai Son of Pu Tepuko; accompanied his father.
Nau Penuatai Wife of Pu Penuatai; came to accompany her husband.
Pu Tongotere Pu Nukurava's matrilateral parallel first cousin.
Nau Tongotere Pu Tongotere's wife; accompanied her husband.
Pu Maravai Pu Tongotere's half-brother; accompanied his half-sibling.
Lillian Takua Pu Nukurava's bilateral cross-cousin.
Pu Pareatai First patrilateral cross-cousin of Pu Raropuko, Pu Nukurava's father.
Pu Koroatu (senior chief) Chief of Pu Nukurava's ‘clan’; came first to the bridegroom's house to leave some presents before going on to the bride's people.
Pu Tokerau Went to represent the senior chief, his elder brother, in the house of a member of the chief's ‘clan’. The chief, himself, went to the house of the woman's closest Anutan kin, people who also were of his ‘clan’, after first making an apperance at the house of the bridegroom.
- 340
Nau Tokerau Wife of Pu Tokerau; accompanied her husband.
Pu Teaokena Younger brother of Pu Tokerau and the senior chief; accompanied Pu Tokerau and their elder brother.
Nau Teaokena Wife of Pu Teaokena; accompanied her husband.
Pu Teraupanga Nau Raropuko's half-brother and Pu Nukurava's ‘mother's brother’.
Nau Paone Pai makitanga of Pu Nukurava; therefore, charged with the responsibility of cutting the bride's hair while waiting in the husband's house.
Nau Nukutamaaroa Sister of Pu Paone; accompanied Nau Paone, her brother's wife, to assist in cutting the bride's hair.
Nau Rangateatu Pu Paone's patrilateral parallel first cousin; accompanied her ‘sister-in-law’, Nau Paone, to assist in cutting the bride's hair.
Nau Taramoa Pu Nukurava's first patrilateral cross-cousin.
Nau Nukumanaia Pu Raropuko's ‘adopted child’ and Pu Nukurava's ‘sister’.
Pu Nukumanaia Husband of Nau Nukumanaia; accompanied his wife.
Pu Penuakimoana Came to represent his patongia at the bridegroom's house while his elder brother, Pu Pourou, went to the house of the bride. I could discern no particularly close relationship between this patongia and that of the groom, but perhaps the members felt it necessary to be represented at both sides due to their position as leaders of one of the three lines of the Kainanga i Pangatau.
Nau Pourou Wife of Pu Pourou; accompanied her husband's brother, Pu Penuakimoana.
II. Bride's Side  
Pu Akonima Tauranga ‘formal friend’ of Nau Nukurava's parents, who were from Tikopia and had returned home. It was in the house of his brother. Pu Rotopenua, that the bride's ‘kindred’ gathered.
Nau Akonima Wife of Pu Akonima; Nau Nukurava's tauranga.
Pu Rotopenua Younger brother of Pu Akonima and member of the same patongia.
Nau Rotopenua Wife of Pu Rotopenua.
Arthur Pakameretangata Pu Akonima's patrilateral parallel first cousin and member of the same patongia.
Pu Notau Brother of Pu Akonima and Pu Rotopenua; tauranga of the bride's parents.
Nau Notau Wife of Pu Notau; accompanied her husband.
Pu Paone Pu Akonima's patrilateral parallel first cousin and tauranga of the bride's parents.
Pu Rongovaru Pu Akonima's matrilateral cross-cousin.
Nau Atapu Wife of Pu Atapu, Pu Rongovaru's brother. Her husband went to the bridegroom's house and sent her to the bride's house to represent him there. This, despite the fact that she is Pu Nukurava's full sister.
Pu Nukutamaaroa Husband of Pu Paone's sister, making him a close relative of Pu Akonima, et al.
- 341
Pu Pourou Pu Paone's matrilateral cross-cousin.
Judah Mataamako Pu Akonima's sister's son.
Anne Pakairipita Nau Akonima's sister.
Nau Tepae Nau Akonima's mother's sister. Pu Tongotere, her son, went to the bridegroom's house, and he asked his mother to represent their patongia at the bride's.
Arikitotoro Pu Tongotere's paternal half-brother; Pu Tongotere asked his half-brother to accompany his mother to the house of the bride's patongia.
Pu Nukumarere Friend and secondary tauranga of the bride's father.
Nau Nukumarere Wife of Pu Nukumarere; accompanied her husband.
Nau Raroipi Pu Akonima's ‘sister’; Ta Nukumarere's daughter-in-law, and mother of Richard Maraetanu.
Richard Maraetanu Nau Raroipi's son and Pu Akonima's ‘sister's son’, but not particularly close.
Pu Koroatu (senior chief) The bride's nearest Anutan kin were among the high chief's leading men; therefore, after visiting and leaving his brothers to represent him in the house of Pu Nukurava, he went to put in an appearance with Pu Akonima and Pu Rotopenua.
Nau Koroatu Wife of the senior chief; accompanied her husband. Also, she is the daughter of Ta Nukumarere, the bride's parents' tauranga.
Pu Parekope Patarilateral parallel first cousin and closest living ‘brother’ of Pu Tepuko, the junior chief, who was at the house of the bridegroom. As the chief could not be present at the bride's house, he delegated the responsibility to Pu Parekope as his representative.

The feast is held the next day, and each day for the next two to three weeks, different patongia brings presents of food to the newly married couple. In theory the man's relatives are supplying him with food and the woman's kin are supplying her. In point of fact, all food is accepted by the couple and shared with the man's patongia. This practice is called te uke rongi.

Approximately one year after the wedding, a rite known as te pakatatanga o nga ariki is performed in honour of the chiefs. Each of the island's patongia contributes food, but the bulk comes from the husband's gardens, with a slightly smaller quantity provided by the natal patongia of the wife. A basket of food is presented to each of the patongia and large baskets of especially prized foods are given to each of the two chiefs. The population gathers in the vicinity of the chiefs' houses where they take their meals for the next day or two until the ceremonial foods have been consumed.


Most post-marital involvement in life crisis rites is as a parent sponsor- - 342 ing a ceremony for his child, or as someone else's inati. Illness, of course, may occur at any point in one's life, and when it does the other patongia express their concern by sending food. Large parcels of the most esteemed foods are sent by the mother's brother who spends a great deal of his time with the victim giving what emotional support he can during the period of recovery. 27 When someone is about to leave the island for a lengthy period he takes a ceremonial meal and receives presents from each of the island's patongia, after which he goes through a period of ritual wailing with representatives from each domestic unit. And when he returns from a protracted stay abroad he takes a meal in turn with each of the patongia. 28 The major post-marital rite through which each Anutan must go, however, is the funeral.

As soon as someone on Anuta dies word is spread throughout the island. The entire population divides itself into several groups on an ad hoc basis, and each group begins to practise a different funeral dirge to be sung in the house of the deceased. One after another the groups file into the house and each spends about an hour wailing dirges before it is replaced by another group.

Each patongia sends at least one representative to wail over the body while other members prepare food which will be used to feed the mourners. After the various contingents have finished wailing they go from house to house partaking of the food that has been prepared.

When the mourners have eaten with each of the patongia they return to the house of the deceased. The corpse is painted with turmeric pigment, dressed in new clothes, and wrapped in several layers of pandanus mats and bark cloth sheets. It is carried to the church for a final service, and then on to the graveyard where the mother's brothers and their close kin have dug a grave in the sand, about six feet deep. The mourners gather around the grave, the body is laid in, some final prayers are said, fragrant leaves are tossed in as a gesture of farewell, and the grave is refilled with sand.

After the funeral, durables and sometimes even garden land are given as compensation by the patongia of the deceased to the men who dug the grave, in a prestation called te punepu. Awarding land to the mother's brother's unit serves to counterbalance, on the level of normative rules, the practice of giving gardens to the sister's son upon a woman's marriage. The actual frequency of including gardens in a funeral exchange is low, however, and I have only a few such cases in my records.

In its punepu payments the family of the deceased may be assisted by kin on both sides. The main recipient of the gifts is the mother's brother of the deceased rather than the inati. If the mother's brothers no longer are alive, or if they are too old to participate in digging the grave, the prestations will be made to the senior male in the mother's brother's patriline. The presents then may be distributed among the other gravediggers.

- 343

In a very real sense the marriage system is an aspect of the kindred. The marital bond creates a close relationship between the patongia of the husband and the wife. The marriage ritual involves exchange of food and durables and is followed by a shift in the allegiance of the woman from the patongia of her parents to that of her husband. Moreover, a man's affinal relatives become his child's maternal kin, an identity most clearly seen in the Anutans' classification of both maternal relatives and ‘inlaws’ as te pai o te papine.

Anuta does not have marriage classes, and the closest thing to a positive marriage rule is a mild preference for a man to take his taina papine ‘sister-in-law’ for his spouse. The closer the ‘sister-in-law’ the more appropriate the choice, and the best possible marriage partner from a kinship point of view is the sister of one's brother's wife. Such a marriage has the effect of reinforcing an alliance which has already been established, and it maintains the social and economic solidarity of sisters by bringing them into the same patongia. However, a man does not have a ‘sister-in-law’ until someone of his generation marries. Up to that time all female relatives in a male ego's generation are uniformly classed as kave ‘sibling of opposite sex’, and no positive preferences may be inferred, therefore, from the alliance pattern of the parents' generation.

More important for Anuta are proscriptions specifying whom one may not marry. The rule is that close kin must be shunned. Incest prohibitions and their derivative exogamic regulations are calculated on a cognatic basis. 29 Closeness, however, is relative, and consequently the rules of incest and exogamy are phrased in terms of degrees of impropriety rather than absolute prohibition. Marriage to a third cousin or anyone more distant usually does not arouse particular concern, although in the abstract, informants sometimes say that it should not be done. Marriage to a second cousin is looked upon with disfavour, marrying a first cousin is definitely tapu, 30 and marriage to a full biological sibling is deemed too absurd for serious consideration. 31

A close kinship bond established through behaviour manifesting aropa generates the same prohibition as a putative genetic tie. Thus, in addition to the relatives who are excluded on the grounds of genealogical closeness, one may not marry into the patongia of his inati or ‘bond-friend’. The child of one's step-parent is equally forbidden, and a prohibition against marrying into one's ‘adoptive’ patongia led to a case of murder in the oral traditions.

In structuralist terms the Anutans have a complex rather than an elementary marriage system. 32 Nonetheless, alliance between groups of affines provides an important means of reinforcing social integration, operating in conjunction with consanguineal and ritual kinship to draw together the various patongia into a set of working, co-operating, - 344 reciprocating, exchanging relationships. The proscription against marrying close relatives means that each patongia is constantly reaching out to create new bonds with groups to which it does not already have strong kinship ties. By tracing proximity bilaterally it is assured that an individual will reach out to distant kin and reinforce the bonds which normally are expected to be the weakest. And since a man must seek a woman in a patongia other than that from which his father took his wife, each generation must produce a set of alliances which is different from that of the preceding one.

The prohibition on marriage into the patongia of one's inati may be seen from the same perspective. Since the pai maatuaa is always presented to someone in ego's ‘clan’, and usually to the ‘clan's’s leading man, it forces a degree of exogamy upon a unit (the kainanga) which has no formal rule proscribing marriage among its members. In this way bonds among the island's ‘clans’ are reinforced through marriage, and the chief, who receives inati prestations from almost all his ‘clan's’ patongia, usually is forced to marry out.

All Anutans have strong ties with their inati's patongia, and these ties are emphasised repeatedly throughout the people's lives without requiring the reinforcement of a marital bond. Therefore, to marry someone in the unit of one's inati would have the effect of creating a “superfluous” relationship and “waste” a marriage which otherwise could be used to substitute a close affinal relationship for a distant “consanguineal” one. Since a man cannot take his wife from his own patongia, those of his close relatives on both his father's and his mother's side, his inati or anyone he serves as inati, his ‘adoptive parents’, or his ‘bond-friends’, all the groups with which he would have close ties without the necessity of marriage are precluded. Therefore, he is forced to marry someone from a group whose relationship to his own needs reinforcement, and the marriage system promotes the widest possible diffusion of powerful bonds among all the patongia on the island.

Finally, the bilateral nature of incest prohibitions, combined with the fact that a man's immediate affines are his children's close “consanguines” on their mother's side, makes the most appropriate source of spouses in one generation among the least appropriate in the next. As the generations go by this “source” gradually becomes more appropriate again until eventually another marriage is consummated with a member of that patongia and the cycle starts anew.


Each Anutan has a wide-ranging kindred which, on various occasions, may include the island's entire population. Kinship is characterised by positive affect as expressed through giving, pooling, or sharing of goods - 345 and services. Since the patongia is the elementary unit of property ownership a kinship tie between two people entails economic bonds between their units, and the closer the kin tie the higher will be the level of interdependence and solidarity between their patongia. Proximity, however, is not determined strictly according to genealogical criteria. A close relationship may be established through several forms of ‘adoption’, the creation of a ‘bond-friend’ relationship, or through the giving and receiving of inati prestations during rites of passage.

Marriage also establishes solidary relations between patongia through the exchange of goods and services, and even more importantly, through the exchange of persons. One should not marry a close kinsman or a member of any close kinsman's patongia. Thus, a man's distant kin become his closest affines and his children's close consanguines, and only over a period of several generations do the patongia become sufficiently remote that their members may marry once again.

  • FEINBERG, Richard, 1973. “Anutan Social Structure” in Anuta: A Polynesian Outlier in the Solomon Islands, edited by D. E. Yen and Janet Gordon. Pacific Anthropological Records, Number 21. Honolulu, Bernice P. Bishop Museum.
  • —— 1978. “Rank and Authority in Anuta Island” in Adaptation and Symbolism: Essays on Social Organization, edited by Karen Ann Watson-Gegeo and S. Lee Seaton. Honolulu, The University Press of Hawaii.
  • —— 1979. Anutan Concepts of Disease. A Polynesian Study. Institute for Polynesian Studies Monograph No. 3. Laie, Hawaii, The Institute for Polynesian Studies.
  • —— in press a. Social Structure of Anuta Island. Copenhagen and Laie, Danish National Museum in co-operation with the Institute for Polynesian Studies.
  • —— in press b. “The Meaning of ‘Sibling’ on Anuta Island” in Siblingship in Oceania: Studies in the Meaning of Kin Relations, edited by Mac Marshall. ASAO Monograph No. 8., Ann Arbor, The University of Michigan Press.
  • —— in press c. “Supernatural Sanctions and the Social Order on a Polynesian Outlier.” Anthropological Forum 4(3).
  • FIRTH, Raymond, 1954. “Anuta and Tikopia: Symbiotic Elements in Social Organization.” Journal of the Polynesian Society 63(1): 87-131.
  • —— 1963. We, the Tikopia: Kinship in Primitive Polynesia. Boston, Beacon Press.
  • —— 1967. “Bond-Friendship” in Tikopia Ritual and Belief by Raymond Firth. Boston, Beacon Press.
  • LÉVI-STRAUSS, Claude, 1965. “The Future of Kinship Studies” in Proceedings of the Royal Anthropological Institute 93: 1-11.
  • —— 1969. The Elementary Structures of Kinship. Boston, Beacon Press.
  • WALSH, D. S. and Bruce BIGGS, 1966. Proto-Polynesian Word List I. Te Reo Monographs. Auckland, The Linguistic Society of New Zealand.

- 346 Page of endnotes

- 347 Page of endnotes

- 348 Page of endnotes

1   Research on which this paper is based was conducted in the Solomon Islands over a 14-month period, from February, 1972, to April, 1973. Eleven months out of that period were spent on Anuta while a total of about three months was occupied in working with Anutans residing in other parts of the Solomons. The research was sponsored by a United States Public Health Service Training Grant, administered by The University of Chicago's Department of Anthropology, and to both those institutions I must express my deep gratitude.
2   Of these, approximately 160 are likely to be living on Anuta at any given time. The remainder reside predominantly in the Russell Islands, Guadalcanal, and Tikopia.
3   For a more detailed account of the patongia see Feinberg (1973:9-11; in press a: Ch. 4). The patongia is structurally analogous with the Tikopian paito (see Firth 1963: Ch. IX).
4   I put “descent” and “ancestor” in quotes to indicate that these relationships may be established on Anuta through adoption of appropriate behaviour as well as genealogical connection. For further discussion of this point see Feinberg (in press a). See Feinberg (1973: 13-17; 1978; in press a: Ch. 6) for an account of the Anutan kainanga.
5   See Feinberg (1978; in press a; in press b) for a discussion of the concept of aropa.
6   By “pooling” I mean to indicate that independent units contribute their own resources which then are used collectively by the entire grouping. Such action commonly takes place between patongia. Within a patongiaby contrast, all resources belong from the beginning to the unit as a whole. I refer to this as “sharing” or “coownership”.
7   At the time of my field study there were eight native Anutans married to people from overseas. All but one of these marriages were to Tikopians.
8   For further discussion of the distinction between ‘near’ and ‘distant’ kin see Feinberg (in press a: Ch. 3; in press b). Anutan kin terms and their common genealogical referents are indictated in Figure 1.
9   By “classificatory” in this context I mean to indicate someone who is not normally designated as “father” in English, but whom the Anutans call by the same term as the putative biological father, and similarly for other kin terms. My intention is not to get into a discussion of “classificatory” as opposed to “descriptive” kinship systems.
10   The Anutan terms which might be glossed as ‘formal friend’ or ‘bond-friend’ are tauranga and toa. Toa may refer either to another Anutan with whom one has established a close ‘sibling’ relationship by mutual agreement or someone from overseas who has been incorporated into the kinship system by being brought into one of the patongia. In the latter case the immigrant is said to have a toa relationship with all the members of the patongia into which he has been incorporated. In dealing with Tikopia each Anutan patongia has a special bond with one or more Tikopian paito. The members of such linked units are referred to as tauranga. For further discussion of this point see Firth (1954: 117: 1967), and Feinberg (in press a: Ch. 4).
11   I use “consanguineal” in quotes to indicate that while a kinsman of this type is nonaffinal he is not necessarily perceived to be related biologically. In dealing with kin who are perceived by the Anutans to be related biologically the quotation marks are omitted.
12   The absence of explicitly stated marriage classes makes this a “complex” rather than an “elementary” system in structuralist terms (see Lévi-Strauss 1969). The number of proscriptions, however, almost creates a de facto marriage class which changes every generation, and the cycle involving several groups bears some resemblances to an elementary system of generalised exchange except that it is completed only after a number of generations rather than in each one. The Anutan marriage patterns are most reminiscent of what have been termed “Crow-Omaha” systems (e.g., see Lévi-Strauss 1965; 1969: xxxv-xlii), although without the corresponding kinship nomenclature. Full discussion of this issue must await a later publication.
13   A ngutuumu is a grouping of two or more patongia which co-operate in planting, cultivating, harvesting, preparing, and consuming food. See Feinberg (1973: 11; in press: Ch. 5) for further details.
14   Huntsman (personal communication) has suggested to me that pai in this context may be the Anutan cognate of the Tokelauan fai ‘to make’, and that a better gloss might be ‘designated father’, ‘designated father's sister’, and ‘designated mother's brother’ for the three most important inati. This is not an issue I can address with certainty because the Anutan language has both pai ‘to make’; ‘to do’ and pai (probably cognate with the proto-Polynesian *fAsi ‘to split’ [Walsh and Biggs 1966: 9] and certainly the Tikopian fasi), meaning ‘side’. However, since the inati prestations do no make ‘fathers’, ‘father's sisters’, and ‘mother's brothers’ out of people who were not previously in those relationships, but simply transform a distant member of a kin class into a close member of that same class, and since the point of the prestations, it seems to me, is to distinguish sections (or ‘sides’) of the ‘kindred’ I tend to think it is in the latter sense that pai is used in this situation.
15   Anuta has four kainanga which, following Firth (1963: Ch. IX), may be glossed very approximately as ‘patrilineal clans’ (see also Feinberg 1973: 13-17; in press a). In order of ritual precedence they are the Kainanga i Mua, the Kainanga i Tepuko, the Kainanga i Pangatau, and the Kainanga i Rotomua. The first two are led by ariki ‘chiefs’ while the second two have non-chiefly leaders who are appointed by the senior chief. For further detail see Feinberg (1978 and in press a: Ch. 6).
16   I plan to discuss the details of Anuta's elaborately involuted system of dual organization in a later publication.
17   The pai makitanga is peripheral in the sense that she is not normally of the pai maatuaa's unit. The pai tupuna is peripheral in the sense that the presentation is optional, but when it is included it always is presented to the same unit as the pai tuatina.
18   By “focal relative” I mean to indicate that member of a particular kin class who is most closely related to ego from a geneaological point of view. In the class, tamana ‘father’, which includes the father, his brothers, and male cousins, for example, the “focal relative” would be ego's own father.
19   Actually, it is only necessary to have one representative from the patongia of each ‘father’, ‘father's sister’, and ‘mother's brother’, respectively. Thus, if Pu Tokerau attended a feast it would not be necessary for his brothers, Pu Teaokena and the senior chief, to attend as well. However, one of them, at least, should make an appearance at the panaunga of a child in the first descending generation.
20   No angaa was held while I was on Anuta. What follows is an abbreviated version of one informant's detailed account.
21   I did not see this rite performed, but it was described to me by several informants. The following account is a summary of their descriptions.
22   Tipunga is the normal word for ‘carpenter’. A tipunga is someone who is skilled at fashioning implements, canoes, or houses from wood. It also is the term by which the man who actually performs the circumcision is called, perhaps indicating that he is taking a mass of raw material and through his efforts is fashioning it into a finished product. In addition, it is cognate with the usual Polynesian word for ‘priest’, and this may serve to emphasise the ritual properties associated with his role.
23   In another version of this procedure I was told that the child sleeps in the tipunga's house while he is recovering from the operation. During this period the boy's parents contribute food to their son and the man who performed the operation.
24   The mother is preferred to the father as an emissary because it is felt that she has a greater interest in her son's settling down while the father will sympathise with the young man's countervailing wish for travel and adventure and is less likely to make a persuasive case.
25   In former days, informants said, Anutans practised a kind of “marriage by capture” similar to that described by Firth for Tikopia (1963: Ch. XIII), and the marae prestations were seen as atonement by the husband's group to that of the bride. Members of the man's group pressed their noses to the kness of members of the women's group as a token of abasement, and all of this was done amid mock fighting which would, not infrequently, get out of hand. This custom was abolished at the order of Pu Teukumarae grandfather of the present senior chief, who objected that a marriage ought to be a happy occasion, not one for fighting. He also objected to the view that one group must abase itself by pressing noses to the kness of members of the other, and it was his decision, that henceforth, the man's relatives would press their noses to the noses of the woman's kin.
26   Actual names are used here and throughout this article.
27   See Feinberg (in press c; 1979) for an account of illness and the ways in which illness is handled on Anuta.
28   For reasons of space the descriptions of these rituals, which do not particularly involve the subdivisions of the kindred, have been severely abbreviated.
29   This is demonstrated both by the abstract statements of my informants and by the fact that in the two first cousin marriages I encountered the man had married a woman on the father's side and in one case the couple were patrilateral parallel cousins—just the opposite of what one would expect to find in a society with a patrilineal orientation if that bias were carried over into rules of incest and exogamy.
30   In fact, I could find only two instances of first cousin marriage (see footnote 27 above). In one of these cases the couple was threatened with exile to the ocean for their offence. In this instance the fact that both the husband and wife came from the same patongia, however, probably was more important than their status as first cousins.
31   One informant, when queried on this score, responded with ten minutes of uncontrolled laughter followed by another five of chuckles as he fought to regain his composure. Then, for the next two weeks or so, every time we met he would mutter, “Marry your full sibling!” and another bout of laughter would ensue. It also is significant in this regard that kava ‘siblings of the opposite sex’ in the same patongia may sleep on the same mat, under the same covers, since the possibility of sexual attraction between them is considered so remote.
32   See footnote 12 above. The character of the Anutan marriage system as one of the complex variety is confirmed by the fact that neither side in an affinal alliance outranks the other. In fact, not only are wife gives and wife takers indistinguishable in honorific terms; they are not distinct at all. A man's sister-in-law is preferred as a prospective wife, but the direction of the marriage is not stipulated in the preference, which means that the marriage of a man and his sister to a woman and her brother is as acceptable and frequent as a pair of brothers marrying a pair of sisters. This is correlated with the absence of a fahu relationship of the Tongan or Fijian type as well.