- 171

This paper deals with the transformation of the title system, or chiefly system, in Western Samoa, especially its recent transformation, multiplication and role differentiation of titles. As described in the Constitution of Western Samoa, Samoan custom is one of the bases of Western Samoa as a nation. Probably every Samoan agrees that the matai system is one of the core institutions of fa'a-Sāmoa (Samoan custom). To Western Samoans, the matai system is the wisdom of their ancestors continuing from ancient times. But, nevertheless, historical research into the matai system has revealed that it underwent significant change after the contact with the West, although it never faced collapse, as in other Polynesian societies.

There are several important monographs on this topic, mainly from historical perspectives, such as Meleisea's Making of Modern Samoa (1987) and Powles' doctoral thesis entitled “The Persistence of Chiefly Power and its Implications for Law and Political Organisation in Western Polynesia” (1979). Both authors worked on detailed historical documents and show how colonial power and Government administration affected the transformation of the matai system. Their concern is mainly on the power (colonial or modern independent) on the matai system.

My perspective here is a little different from those of Meleisea and Powles in that my concern is not an examination of the changing matai system in the historical process but rather of the structural transformation of the system as a whole. I shall analyse cases of title-splitting in detail and try to identify the mechanism of the multiplication of titles. O'Meara (1991) represents another important study which is an ethnography of a rural village based on the author's socio-economic research on land tenure and the chiefly system under transformation. His major concern is the reason why Samoan planters do not produce more agricultural products. O'Meara demonstrates that Samoan conservatism is the surface appearance of their own economic rational choices. I agree with his discussion on so-called Samoan conservatism, though I am rather concerned with the tension between their rational choices and their value system, or, in other words, the process in which rational choices are disguised in the reinterpretation of values/norms. For this reason, - 172 I devote one section to describe in detail the ideal matai system which exists in Samoans' minds.

I emphasise two points here. Firstly, I focus on aspects of urbanisation, since they seem to me to have significant effects on the title system in Western Samoa, though there are various factors affecting its transformation. There is “hypertrophy” of the non-traditional capital town of Apia and a large labour migration (especially among young and middle generations) overseas; the rural population is decreasing. Secondly, there are elements inherent in the title system itself which are relevant to the change, responding to such worldwide phenomena.

Formally speaking, a Samoan titleholder must reside on the land of his kin group, whose resources and manpower he controls for the benefit of his whole group. Recently, however, as the number of titleholders increased as titles were split and vacant titles restored, more and more titles have been conferred on the emigrant members who live in the Apia urban area or overseas. Although there was much criticism of this phenomenon as a deviation from the traditional custom, a close examination reveals that the recent transformation follows the Samoan traditional custom, in a sense. Of concern here is to demonstrate that the Western Samoan chiefly system has been transformed to adjust itself to the socio-economic conditions of the world system.

The Samoan chiefly system is characterised by its relatively autonomous territorial organisation (M. Yamamoto 1987). This is one factor which kept Samoa from developing a centralised kingship such as in Hawai'i, Tahiti and Tonga, and which prevented social destruction by the colonial powers. An 'āiga (kin group) owns particular housing lots, farmland, and chiefly title names (suafa matai), each of which has a certain prestige, privilege and role, and which are explained in oral traditions. When a titleholder dies, a suitable member of the 'āiga is to take the title. The matai (titleholders) of a village have regular highly ritualised council meetings to maintain its autonomy. A Samoan chiefly title is, thus, based in a particular localised land-holding kin group in a territorial organisation. And it is not controlled by a superior in the ranking system of titles but by the kin group to which it belongs. Thus, the Samoan chiefly system is less centralised than the system operating in Tonga, in which titles are ranked in the pyramidal structure of authority according to rules based on kinship. 1

The two Samoas, Western and American, used to be one society sharing one culture and one language. They were divided in the colonisation process of the late- 19th century. Because of the different metropolitan powers and different colonial administration policies, each has achieved different degrees and different modes of acculturation.

- 173

Western Samoa, where my research was conducted, was colonised by Germany in 1900, became a trust mandate of New Zealand after the First World War, and became independent in 1962. In 1962 it established an election system in which only chiefly titleholders were able to vote and were eligible as candidates for Parliament; this was a compromise between the traditional Samoan chiefly system and the Western idea of democracy. However, in late 1990, universal suffrage was introduced. Compared with American Samoa, where the market economy has thoroughly penetrated because of the successful fish-canning industry operating since the late 1950s, rural Western Samoans still live in traditional larger families in a state of semi-subsistence, cultivating taro, ta'amū, banana, etc. They are often described as conservative, continuing to use traditional formal speech, honorifics and manners peculiar to Samoan culture. However, at the same time, modernisation in Western Samoa has been accelerated, especially since Independence. Proud of conserving their traditional culture, Western Samoans scorn American Samoans as having lost most of their own culture, but they admire the material culture of industrialised countries and are eager to earn cash in order to buy manufactured goods from overseas. Overall, they have an ambivalent attitude towards Western things and customs (fa'a-Pālagi).

Christianity was introduced by missionaries to Samoa some 160 years ago, and has become an integral part of Samoan life. Some Samoans do not go to church, but almost all Samoans claim to be Christian.

The matai system is also an important social institution for every Samoan. Every extended family is represented by its matai, who supervises the household as a trustee, while a village or a district is governed by its council, who comprise the matai in its locality. To Western Samoans, the matai system embodies the wisdom of their ancestors continuing from ancient times. It seems to them that this traditional system is vital to Samoan society and has never changed.

Although Western Samoans embrace this image of a never-changing matai system, it, nevertheless, has been transformed through contact with the West, starting before the mid- 19th century. For example, in 1934 in his Modern Samoa, Felix Keesing wrote that:

The higher forms of pule (authority) are falling into disuse or being disputed, and the local matai, head of his family and trustee of the family land and honor, is becoming the important unit. Overlordship titles may still receive ceremonial recognition, but are losing their working significance. The process of splitting - 174 higher titles is desiccating the authority formerly vested in single individuals, and even should the sum-total remain fairly constant the various incumbents have less and less importance as the process continues (1934:247).

By the end of the 19th century, modern currency and manufactured calico cloth had been introduced into the categories of exchange valuables (Stair 1897: 173; Krämer 1902:2:90-1; Mead 1969:74). And, before the First World War, kegs of salt beef had become indispensable to the ceremonial feast honouring a carpenter when he had built a church, and Samoans needed cash to obtain these kegs (Handy and Handy 1924:16-17). (Kegs of salt beef fall into the same category of valuables as pigs (M. Yamamoto 1990:89-90)). It seems that the cash economy started to penetrate into Samoan society as early as the descriptions above. In this sense, the matai system cannot be the static model system described by both Samoans and also several anthropologists — that is, it has not been a continuous institution since ancient times.

With such reservations, I proceed to describe the traditional 2 model matai system for the present-day Samoan. Its quintessential existence was observed probably around the mid-20th century. Although matai are said to have held the power of life and death in the old days, they no longer had such power even at the beginning of the 20th century (Mead 1943:216). While the cash economy was gradually penetrating into Samoan life, only a few 'āiga had wage labourers and there were no emigrants sending back remittances, so it was an important duty of a matai to control the whole labour force of his own āiga. Here I describe the matai system of those days, the model matai system which the people cherish in their minds. The matai system is, of course, relevant to the 'āiga or kinship system. Anyone who can trace ancestry to an 'āiga, whether on the father's or mother's side, may join the 'āiga somehow, though each claim on the 'āiga varies with the relationship to it. Thus, an 'āiga is a loosely organised kin group mostly composed of ambilineal resident members and many latent members living elsewhere who can trace their descent to the 'āiga and retain some claim to it. In fact, the term 'āiga denotes a primary family as well as a larger group up to the one which covers a whole district on a different level of organisation, though what I am referring to here as an 'āiga is a kin group which is based in a village, controlling land and titles.

By way of explanation, a title is a particular name which is passed from generation to generation in an 'āiga. A titleholder is a matai who manages the 'āiga land and the labour of the members, and who represents the 'āiga to the outside. A title name has its own duties and privileges, which are also passed from generation to generation; thus, the continuity of an 'āiga organisation is secured. Each title name has its own secret oral traditions which explain its origin which are transmitted through the 'āiga to prove its legitimacy; it also - 175 has a certain status in the ranking structure of titles in territorial organisations (M. Yamamoto 1987). An 'āiga, then, is 'āiga integrated into a centralised system but receiving a certain social control from outside. Additionally, titles have been controlled by the colonial Administrations through the Land and Titles Commission (later, the Land and Titles Court) 3 since 1903.

An 'āiga, for example, led by a matai whose title name is Loto, is called Sā Lotō: his name with the prefix of Sā. Sā Lotō might be classified as an 'au 'āiga, whose other titleholders are under Loto's supervision, or be a small-scale 'āiga whose only title name is Loto. In either case, each matai manages a household composed of those who share the common ancestor cognatically and their spouses and children. Today such a household contains some 20 members, whereas, in the old days, it might have been 50 (Turner 1884: 173). 4 A matai administers both the material (land) and non-material estate of the 'āiga, controlling the labour of those living there. All the 'āiga members, young or old, must contribute somehow to the 'āiga according to the instructions of the matai. An 'āiga member may never be driven out of an 'āiga to be deprived of food and shelter as long as he obeys to his matai. An 'āiga takes care of all its members, including the helpless young, the feeble old and the handicapped. In order to manage the communal life of a large household, a matai has a range of duties from making management plans to giving detailed instructions to the members within an 'āiga. Samoan youth are not trained to work voluntarily, but to follow detailed instructions given by the matai, since it is generally believed that untitled youths are unable to behave themselves without such instructions. Samoans often explain this Samoan idea by saying that an untitled person is like a child, and so a matai must also supervise moral behaviour of the youths. It is a matai's duty to look after these persons lest they should go moetotolo (stealing into another 'āiga for a woman at night) or steal the property of another 'āiga (M. Yamamoto 1986: 127-9). 5

In addition to the obligation to manage the affairs of his own 'āiga thus described, a matai has to represent the 'āiga in chiefly councils of his village and district. He may have privileged roles such as making a formal speech or distributing kava and food, or exchanging valuables which are assigned to his own title name.

Undertaking an appropriate ceremonial exchange in ritual contexts is another important role for a matai, at least at present, though it is difficult to discern how this role was filled in the first half of the 20th century. Having studied those ceremonial exchanges, I have assumed that those which were once practised only in ceremonies for high chiefs and their families have come to be practised by every matai as part of the process of equalisation among titles (Yamamoto and Yamamoto 1981: 134-7). The number of these ceremo- - 176 nial exchanges seems to have gradually increased through the first half of the 20th century. Nowadays, it is a major obligation for a matai to prepare pigs, cash and fine mats for a ceremonial exchange transaction a few times a month, or sometimes once or twice in a week.

How is such a matai recruited? Early ethnographers noticed that a Samoan chiefly title was not automatically transmitted as in other places in Polynesia where the primogeniture was respected. Appointment was made unanimously by the gathering of an 'āiga to which might come every person who could trace descent to the 'āiga, including descendants of members who married out. In those days it was a privilege of the tamatāne (literally, descendants of male members) to have the first access to a title, leaving only a slight chance for tamafafine (descendants of female members) to become a matai. In her research at Manu'a Islands in the 1920s, Mead reports that tamafafine held the veto power while tamatāne had the initiative (Mead 1969: 18). Schültz, who was the Imperial Justice of German Samoa, says, “tamafafine only have their turn when no real tamatāne is forthcoming” (Schültz 1911: 51).

Among tamatāne, however, there seems not to have been any strict priority. Turner says, “The son may succeed to the title which his father had, but it may be given to an uncle, or a cousin, and sometimes the son is passed over, and the title given, by common consent, to a perfect stranger…” (Turner 1884: 173). According to Schültz, the first son was not necessarily the heir apparent, and the title name of a deceased might be given to his brother and again be given to the nephew of the successor in the next generation. A dying matai could state his will (māvaega) as to who should succeed him, and this utterance had to be taken into account, but it was possible that the 'āiga might still make a different decision. Adopted sons also had a chance to succeed to a title (Turner pp.51-3). Flexibility of succession, which emphasises the leading ability of a successor, is one of the major characteristics of the matai system. It is not an exaggeration to say that it is only because of this character that such a “feudalistic institution” like the matai system could have survived in the worldwide trend of democratisation that has come with modernisation. When Western Samoa was preparing for Independence, Samoan representatives insisted that the matai system was part of their culture and their own way of democracy, and they were supported by scholars such as Davidson (1967) and the Keesings (1956). This opinion was also confirmed in interviews I had with some political leaders of Western Samoa about 20 years after Independence. Armed with this view — that the matai system is an alternative democratic system in which representatives are selected from families and these representatives in turn select from their ranks some titleholder to represent the village at the district level and so on — Western - 177 Samoans adapted their traditional system to the modern world.

The succession of titles at present has been transformed into a system more democratised than before, probably because of this view. Today, the major guideline in the Land and Titles Court of Western Samoa is Notes written by Marsack (1961:6), who was President of the Land and Titles Court for nearly 20 years shortly after the Second World War. According to this memorandum, the difference in claim between tamatāne and tamafafine is replaced by that between filifiliga (inner core of the family) and tāupulega (outer ranks of the family) (1961:6) 6 and the appointment of a title to a woman is also recognised by the Court. Actually, the Court does not recognise any difference between tamatāne and tamafafine today (according to a conversation which I had with the then Registrar, Mr Tuiletufuga Enelē, in 1978). Today the Court tries to reduce the difference between genders in claims under customary law.

What, then, are the criteria in selecting a matai?

• 1. Blood relationship is respected in the most flexible way. A husband of a female member, or an adopted son, might be appointed to a title, though, “when there is a disagreement in the family and an appointment to a vacant title must be made by the Court, the Court will never appoint a fai-avā (husband of a female member) to the title when there is any good candidate available from the 'āiga proper” (1961:11).
• 2. Various kinds of tautua (service) are counted. How much service was rendered to the previous matai? The service includes contributing to the matai by collecting food for village ceremonies, providing everyday proper food to him, donating money to the church, working in the plantation according to his instructions, etc. Service on behalf of the village or district is also considered. In traditional terms, such kinds of service are possible when a person lives on the 'āiga land.
• 3. General fitness for the title is also a consideration. It is necessary for a prospective matai to have a good knowledge of family history and traditions, and he must be well versed in Samoan customs such as formal speeches and ceremonies. He also “should be of such character that he is able to maintain that standing in the eyes of the district and in the Council of Chiefs and Orators” (1961:13).
• 4. The will of a deceased matai may greatly influence the 'āigadecision, but without any legal validity.

The principles above are carried into effect by the Court when a case is taken to the Court, though a decision can be made outside of these principles if members of an 'āiga can make a unanimous decision. A husband of a female member or an adopted son or even a total outsider may succeed to a title only if an 'āiga can reach a consensus. Recently, a title has been occasionally given to - 178 a foreigner either in order to recognise friendship between him and an 'āiga or a village or to thank him for his kindness (for example, for money or goods given by him). As the Land and Titles Court does not recognise a foreigner holding a title, such a person will be unable to register his title. Also, as the person has little possibility of using 'āiga land, this kind of title-giving has only nominal value. Traditionally, besides suafa titles, there is another category of titles called ao which are highly honoured but have no 'āiga land to administer. A title given to a foreigner seems to be similar to an ao title. I have found a case in which a title was given to a carpenter of no relation to thank him for building a church for a village; people said that it was because the villagers could not collect enough money to pay him. These fairly large variations in title recipients, which have been increasing recently, fall into a slightly expanded traditional category of candidates for titles.

Before discussing title-splitting and the large increase of non-resident matai, I shall present a general overview of recent urbanisation in Western Samoa. In the traditional context, there is a high ranked capital village as the political centre of a district, which is actually composed of several villages. For example, 'Upolu island is divided into three districts, each with a capital village: Atua district with Lufilufi village in the east, A'ana district with Leulumoega village in the west, and Tuamasaga district with two complementary villages of Malie and Afega in the centre. The present capital town of Apia is irrelevant to such a traditional ranking order among villages. It was built by Europeans as a trade centre in the mid-19th century for geographic reasons, being located in the middle of 'Upolu island next to a deep gulf on the edge of the only plain in Western Samoa. It prospered as a port town, as a base for those copra schooners which travelled as far as Tuvalu and the Cook Islands to buy copra in exchange for manufactured goods. Since the end of the 19th century, Apia has been not only a trade centre but also the seat of the colonial Government of Western Samoa. In short, Apia has been a forerunner of modernisation in Samoa, as opposed to traditional culture. Apia, thus, was the place where Europeans and their Samoan wives and children lived. For ordinary Samoans, excluding those who worked for the Government and for European households, Apia was the place to come to present themselves at Government departments and to buy manufactured goods. But gradually, Samoans, attracted by certain aspects of European culture, started to live in Apia, seeking a cash income and high standards of living and education. To live in Apia requires one to buy or lease a piece of land and build a house or to rent an apartment. It is not easy to produce and to live on subsistence crops like taro and banana, as land is limited. In most cases, Samoans in Apia are - 179 dependent on their cash income for their food. While city life burdens people with many difficulties, Samoans are attracted to its freedom and the stimulating and entertaining atmosphere. The population of the Apia urban area, which includes several adjacent traditional villages in Vaisigano and Faleata Districts, is continually increasing.

Table 1 shows the population dynamics of this area in comparison with the population of the whole of Western Samoa. Today, one fifth of the whole population is concentrated in this urban area. Owing to the recent development of transport, nearby villages, especially those located between Apia and Faleolo Airport, have become dormer villages for the people working in town. In this sense, urbanisation is still expanding, though the rate of population increase of the urban area has recently dropped.

In addition to urbanisation within Western Samoa, there is an international urbanisation process. It is only since the Second World War that large-scale emigration from Western Samoa to New Zealand has started.

Table 2 shows the census data on Samoans in New Zealand prepared by the New Zealand Government. There is a population of 157,000 in Western Samoa, and another Samoan population of 63,000 in New Zealand (1986 census). However, the urbanisation process has had much more serious effects in American Samoa because American Samoans, as American nationals, 7 have no barriers to emigrating to Hawai'i and to the mainland United States. According to the 1980 census data, some 42,000 Samoans live in the United States (Franco 1987:5), while the population of American Samoa is only 32,000. The accuracy of these data is, however, disputed; it is suggested by other data that 85,000 live in the United States (Douglas and Douglas 1989: 11). As wages are higher in American than in Western Samoa, a significant number of Western Samoans have migrated to work there. Of the population of 32,170 on Tutuila, 9,540 are Western Samoan born in 1985

- 180

(American Samoa Government, Economic Development and Planning Office 1988). Through their relatives of American Samoan origin, it is possible for Western Samoans to emigrate to the United States. Australia is another new destination for Western Samoan emigrants. Such a rapid increase of emigration has, of course, a serious impact on the society; I am concerned here with an influence on the transformation of the matai system. Before proceeding to my main topic, I shall present here data on remittances from emigrants, showing their great economic contribution to the country. Table 3 shows the trade balance by exports, imports, and private money transfers.

The amounts of money involved have been increased greatly because of the devaluation of currency against the US dollar since 1979, especially between 1980 and 1984, when the total devaluation amounted to 86% (Y. Yamamoto 1987:314). Nevertheless, total remittances exceeded exports in the late 1970s and have been increasing ever since. I often heard Samoans explain the situation, saying, “We export people because we have nothing else. Our parents export children and have money sent back.” There is, in fact, scarcely a household which has no brothers, sisters or children overseas and which does not receive any remittances. Families do not necessarily depend on them for their everyday life but do so at least for special expenses such as ceremonial exchanges and community obligations like special donations for church. There are also heavy voluntary remittances during Christmas and White Sunday. 8 When a matai plans to build a new house, he relies on his children overseas in most cases. Other forms of assistance by emigrants may not show up in statistical tables, such as air fares for relatives in Samoa to go overseas, and money brought back by these relatives.

- 181

I have already noted the possibility in the traditional Samoan chiefly system of recruiting matai from a very wide range of candidates. I shall now examine the custom of title-splitting, the mechanism for the multiplication of titles peculiar to the Western Samoan system. Until late 1990, when universal suffrage was finally introduced after long debate and turmoil, Western Samoa had maintained a unique system of representation in which only matai were able to vote. The older system was a sort of compromise between traditional customs and democratic ideas which are hegemonic in much of the present world. 9 But this system of representation came to be much abused when politicians started giving out as many titles as possible with the consent of their āiga in order to gain enough votes to win. This abuse was fatal to the election system as well as to the matai system itself. The most famous case is that of Vaisigano No.1 constituency in which only 108 votes were cast in 1964 while 1295 votes in 1967. This shows that many titles were given to various individuals for the purpose of winning votes (S. Tiffany 1975). 10 In order to increase the number of matai, the āiga often resorts to splitting title names as well as looking for vacant titles. 11 As the term suggests, title-splitting is a custom of giving the same title name to more than one person. It often happens when it is difficult for an āaiga to reach a unanimous decision on a single candidate. 12 Marsack, thus, describes this practice, which started even before Independence:

In the register of matai s published in July 1957, for example, there appear the names of 21 holders of the title Papali'i of Sapapali'i, 20 holders of the title Asiata of Satupa'itea, and 17 of the title Fuimaono of Salani. The practice is - 182 widespread and examples can be found in practically every village of Western Samoa.

Title-splitting, however, is a custom practised since the old days. Sixty years ago, F. Keesing noticed that it was becoming more frequent, and issued a warning about it (1934:245-6). 13

A foreign researcher in Western Samoa may often be surprised by the fact that even the highest chief (ali'i sili) of a village is not a single person. An 'āiga is often divided into several branches, each of which contains the same title name. These branches are called itūpaepae (or fuaifale), and seem to have existed long before the election system was introduced. For example, today in Fasito'outa village, where the two high chiefly titles of Leaupepe and Aiono reside, there are three branches of Sāa Aionō, headed respectively by Aiono Tuala, Aiono Su'amatai'a and Aiono Fina'i. According to the genealogical record by Krāmer, Aionolevave, who seems to have been the first titleholder of Aiono of the Sātuala family in A 'ana District, had three wives. His first marriage with Iliganoa, a daughter of Amituana'i, brought him two sons, Tuala and Tuaifaiva. Another marriage with Togitoto, a daughter of Alipia of Leulumoega village, brought him a son, Su'amatai'a. A third marriage with Filitua, a daughter of Ogatai of Leulumoega village, brought him a son, Aionofina'i (Krāmer 1902:1:177). Though it may be naīve to conclude that these branches were started in the second generation of the ancestor, as according to this record, it seems, nevertheless, that the branching of the 'āiga Sā Aionō is quite old.

Another example is the title of Leota, the highest chiefly title of Solosolo village in Atua District. The fa'alupega (formal greetings to the order of the titles of a territorial organisation) book published by the Congregation Christian Church (1978) confirms that there are four branches of the 'āiga headed by a Leota title: Leota Leulua 'iali'i, Leota Seiuli, Leota To'omata and Leota Lemusu (p.141). As fa'alupega seldom mention such branches, they seem to be of long standing.

I have discussed elsewhere the splitting of the Matai'a title of Faleata (M. Yamamoto 1987:228-30). As my detailed analysis there shows, it seems quite old, too, since the official residences of two Matai'a titles are, respectively, located in two different villages spatially separated by several other villages.

These cases of title-splitting are accompanied by the fission of 'āiga into branches each with its own power of control (pule) over the land and the - 183 members as an autonomous organisation. In this sense, title-splitting is a part of the reproductive processes of the 'āiga system. As the population of an 'āiga increases, it has more difficulty in its management and, as a result, undergoes fission into branches under different leaders. In such cases, title-splitting is itself a result of an 'āiga development. As an extension of this custom, another form of title-splitting has recently developed; in this new form, titles are split without any 'āiga fission. In other words, a title name is shared by several members within a single 'āiga organisation. In discussing this form of title-splitting, we need to consider the conspicuous increase of nonresident titleholders. Sugimoto, who carried out research on the matai population in two different villages of Sāmamea on 'Upolu and Fagafau on Savai'i in 1965 and 1980, confirms the phenomenon (Yabuuchi and Sugimoto 1967:22-3; Sugimoto 1982:161-5). Table 4 is based on his data and shows the increase in the number of titleholders due to the large increase in nonresident titleholders.

We must recognise that the prevailing custom of nonresident title-holding also originated from the traditional Samoan title system. As Marsack discusses, although a matai was required to reside in his own 'āiga to assume his chiefly duties, there have been exceptional cases (Marsack 1961:12).

There are many instances of course of matai holding two or more titles from different villages, in accordance with the wishes of the families concerned. An example is the holder of the title Tautī of Nofoali'i who is also Auali'itia of Leauva'a and La'ulu of Gataivai, Savai'i. It is, of course, physically impossible - 184 for Tautī to reside permanently in all three villages; but he is expected to keep in touch with all three, and to be present at family functions of La'ulu in Gataivai just as much as in family functions of Tautī at Nofoali'i. The Court would not in ordinary circumstances make such appointments but, as I have already stated, it recognises the power of the family to do so (Marsack p.12-3).

In such cases of two or more appointments between different villages, nonresident matai were expected to fulfil their duties in the villages in which they did not live. In other words, even if they did not reside in the same village as the 'āiga, their nonresidency was overlooked by the Court as long as they cared for the 'āiga. And it seems to have been not only the Court rule but also the customary way of maintaining two or more appointments. By a slight extension of this rule, it is certainly permissible for nonresident Samoans living in Apia, Pago Pago, or other overseas emigrant communities to hold titles as long as they look after their own 'āiga.

In the following section, I shall examine three cases of title-splitting and residency from my own data.

Village A is one of the four villages headed by the ao title M, which is given by unanimous consent of the orator group of the four villages. The title M is more of an honorary nature, without any land of its own, while the title N which originated from the title M, is the highest except for M, and may exercise more practical power. Now, the 'āiga Sā N had been divided into four branches within each of which much title-splitting had occurred before I began research in early 1979.

Those given the title N number 25 individuals, of whom only six persons lived in the village, including one who usually used another title name of his own. There may have been several reasons why the title N was shared by so many persons, one being that there are no subordinate titles under this title, in spite of the fact that it is the most important title in the village. Although I have no data on the genealogical relationships between these branches, it will be enough for my present purpose to show the relationship of each of the titleholders within each branch.

Branch 1. This branch had nine titleholders, matai whom I shall label 1a, 1b, 1c, 1d, 1e, 1f, 1g, 1h and 1i. Of these nine, 1a and 1b resided in the village (using bold lettering to show resident status), 1a was in his late 60s and the oldest of the five resident N holders, taking the role of giving speeches and drinking kava in the chiefly council of the village to represent the whole 'āiga. - 185 The other resident holder was 1b, who was in his 50s and was usually silent in the council, 1c and 1d lived in the next village, 1e, 1f, and 1g in the Apia area, 1h in New Zealand, and 1i in the United States. From genealogical data, I constructed Figure 1. This branch clustered around the six titleholders of near relations: 1a, 1c, 1d, 1e, 1f, and 1h, though the exact relation of 1h to the branch was not known. It was said that 1b had first tried to succeed to his deceased father's title P, but in vain, because the 'āiga could not come to a unanimous decision. So he had asked 1a, his distant relation, to give him the title N on the understanding that the appointment was nominal. When a church dedication ceremony 14 was held in the village, these near-related titleholders except 1h, who lived in New Zealand, gathered under the leadership of 1a to make a large ceremonial exchange, 1i and 1g, who did not join in the ceremonial exchange and were not seen by me during the research period, were said to be the most nominal among the nonresident titleholders.

Branch 2. Of the four titleholders of this relatively small branch — 2a, 2b, 2c, and 2d — the first two resided in the village. 2a's father, who was also a holder of the title name, was an adopted son of a holder of the title N. 2c and 2d were the husband of one of his sisters and the son of another of his sisters, respectively; both of them lived on the island of 'Upolu without much commitment to the 'āiga or to 2a, who was in poor health. 2b, a friend of 2a, who did not have an 'āiga to live in for various reasons, had been conferred a title by 2a. It seemed that 2a gave 2b a title in expectation of receiving service from him. This was an extension of the idea of conferring a title on a stranger because of his good service to the matai. A subordinate title would have been given, had there been any. 2b would remain to continue rendering his good services to 2a on occasions of ceremonial exchanges.

Figure 1. The 'āiga Sā N, Branch 1.

- 186

Branch 3. This branch had seven titleholders: 3a, 3b, 3c, 3d, 3e, 3f, and 3g. This group was characterised not only by its leaders but also by many of the members having received higher education and becoming pastors and teachers. Nevertheless, none of the titleholders lived permanently in the village. 3a had been educated overseas as a Government scholarship student and, in 1970 when he was elected as a Member of Parliament, 15 he left his teaching job in a school of tertiary education. After he lost his seat in the 1973 election, he continued to seek a political career, so he visited the village occasionally to distribute presents to the villagers and to make a large donation to his church there. 3c was a medical doctor working in Apia Hospital. 3f was a Government official living in Apia, who had built a house on his plantation land where nobody usually lived. It is a custom that a matai may cultivate any virgin land within the territory of his village and claim it as his own. Activating this customary law by employing workers, he had become the owner of a large tract of plantation of taro for export. 3e lived in the adjacent village, 3b and 3d in other villages on Savai'i, and 3g in American Samoa. On the occasion of the church dedication ceremony mentioned above, 3a, 3b, and a member who was the pastor 16 in the next village came to make large ceremonial exchanges in the name of each individualmatai, including those who did not come. The amount of the valuables given by this group to the collection for the payment to the carpenter who had constructed the church building was so large that the only valuables given in the name of 3a alone exceeded those given by everyone else except the pastor of the church. Figure 2 shows the genealogical relationships of the core titleholders of this branch.

Branch 4. This branch had five titleholders: 4a, 4b, 4c, 4d, and 4e. Although 4c and 4d were nonresident, they lived in the adjacent village. And both households of recently deceased matai 4f# — a Member of Parliament as well as a holder of an ao title of the adjacent village — and 4g# made

Figure 2. The 'āiga Sā N, Branch 3.

- 187

contributions to the chiefly council of the village in the name of the deceased, so that the two households might secure the matai positions with them, though it had not been decided who would succeed to the titles. Of the four branches, Branch 4 seemed to be the most attached to the village. Regrettably, the genealogical relationships of this branch are not yet available.

Table 5 shows for each branch the number of holders of the title N, in relation to their residence. Of all 26 holders of the title N, only six lived in the village. If another five who lived in nearby villages were included, still only 11 matai out of 26 were able to come often and to look after 'āiga. (It should be noted that, although there were 65 titleholders registered in the village, whose population was listed as 246 in the 1976 census, only 36 of them actually lived in the village.)

The title Q was one of the earliest titles to have been split for the purpose of elections. According to the Ala'ilima study on the title Q, there were only six titleholders in 1959, but 14 in 1964 (Ala'ilima and Ala'ilima 1966:242). Actually, two neighbouring villages have the title names of QX and QY as their respective highest chiefly titles. The fact that each title name has a different legend to explain the origin and that the name is split into two villages indicates the title Q was split into QX and QY a long time ago. There are four branches for the 'āiga of the title QX, while five branches for QY. Here, I shall describe of one of the branches of Sā QY. The following information was provided when I visited the village in 1981.

The branch of Sā QY which is the focus of this section is here called Branch 1. This branch had six holders of the title QY: QY1a, QY1b, QY1c, QY1d, QY1e, and QY1f. QY1a and QY1b were an old man and his son who made up the core of Branch 1. QY1a, who received his title in 1940, was over 70 years old and had survived his wife. Sick and supported by his son, he was - 188 respected by all the villagers. QY1b, who received his title in 1974, was the actual manager of the household. QY1c, who received his title in 1971, was a Government official living in the suburb of Apia. Both QY1d, who had received his title in 1974, together with QY1b, and QY1e, who had received his title only recently, worked and lived in Apia. Because QY1d, a store manager in Apia, tried to employ the youth of the whole 'āigaand to help them in finding jobs, he was respected by all matai of the 'āiga. QY1f owned a house on the portion of the coastal land on the main island which belonged to Village B itself, whose traditional locality was on a small nearby island, and he made a living cultivating his plantation on the main island. His house was a base for children of the branch living on the small island to attend the secondary school nearby. The reason why QY1a and QY1b lived on such a small island while it would have been much more convenient to live on the main island was that the land on the small island where they resided was the most prestigious housing lot for the whole 'āiga Sā QY; they did not intend to abandon the lot.

Figure 3 shows the genealogical relationships of holders of the title QY. As the table shows, both QY1e and QY1f are adopted sons of an already adopted line of this 'āiga. There were probably some benefits for the branch to take them into the 'āaiga. Their deceased foster father, who had the same title name and was promoted to a high rank in the police force, seemed to have had no

Figure 3. The 'āiga Sā QY, Branch 1.

- 189

chance to live on his 'āiga land. As for QY1e, he often had come to contribute for ceremonial exchanges before he asked for a title name. Whenever titleholders of Branch 1 gathered for decision-making, nonresident matai respected the will of QY1a and QY1b.

The 'āiga of R, which is located in a suburban village of Apia, has been split into two itū paepae (branches), each of which has again been split into two fuaifale (a branch less formal than the itū paepae). Each of the four branches is led by a holder of the leading title R. Branch 1, one of these branches, was an 'au 'āiga led by R1, who died of old age. As a result, this branch had a dispute about who was going to be the successor to the title, and took the matter to the Land and Titles Court. The Court eventually appointed R1' to be the next leading matai of the branch. R1' succeeded to the title at the end of 1979. Early the next year, R1' gave titles to eight members under him, giving a total of 14 subordinate matai. The genealogical relations are shown in Figure 4. There used to be seven subordinate matai under six different title names. Branch 1 activated five different title names which had been vacant for many years or never used before, and distributed them among eight members of the group. The consent of the other fuaifale is required in order to give the title name R, while R1' may decide to give subordinate titles by himself, though he must consult the matai of his own branch.

Branch 1 had previously given titles only to resident members in order to protect their livelihood (though one of the matai had migrated after living with them for many years). Now, four matai of this branch were nonresident, two of whom lived in Hawai'i and New Zealand, respectively. It is interesting to note that these two nonresident matai living in other villages received other title names in their own villages at later dates. In the course of time, they

Figure 4. The 'āiga Sā R, Branch 1.

- 190

dropped out from regular contributions for ceremonial exchanges of Branch 1 except for large exchange occasions.

The other branch of the same itū paepae had 13 matai, seven of whom were nonresident: the highest matai, R2, who belongs to an 'afakasi (“half caste'”, i.e., part-European Samoan) family, lived in New Zealand; another was in New Zealand, two were in the Apia urban area, and three in other villages.

The whole 'āiga Sā R had split the title R into only four, one each for a branch. Nonetheless, it multiplied the number of titleholders by giving subordinate titles, sometimes splitting them, to those who needed titles. For example, a title name G, which originated from the personal name of the first holder of the title R, was split into eight: six in Branch 1 and two in Branch 2. In this case, we find that the complementarity between branches might be another reason for multiplication of titles. It is sometimes very difficult for a leading matai to refrain from conferring titles on nonresident members in order to maintain the dignity of the 'āiga, once another complementary branch has started the mass conferring of titles. It seldom ends without a claim for a title from a subbranch group which does not have one, even if it is for a title for a nonresident. And, once a leading matai confers a title on a nonresident member of a subbranch, he must do the same for other claims from other subbranches. Thus, the more the 'āiga tries to equalise each sub-branch, the more it ends up giving titles to nonresident members since most male Samoans want to be matai at some point in their lives. I surmise that the mass title-giving which happened in Branch 1 early in 1980 had something to do with the fact that Branch 2, the other branch in complementary relation to it, already had many nonresident titleholders. 17

The famous “Tamasese Report”, which made a recommendation to Parliament to adopt universal suffrage, is rather moderate in saying, “The Committee believed that the elections may be one of the reasons why these titles have been slitted…”. But it seems that the Committee thought the election system was the major reason, since it not only failed to give any other reason for title-splitting but also concluded that, “The Committee is therefore firm on its belief that this problem of multiplying the number of matai will be solved if suffrage is granted” (Western Samoa Parliament 1979:8). Additionally, the matai suffrage was the major target of accusations made by Samoan opinion leaders for title-splitting. 18

However, in the examination of the three cases of title multiplication above, we have found that the recent abrupt increase in titles is not caused by - 191 the election system alone. Each case shows there were various reasons for conferring titles on nonresident members. Ordinary cases are those in which titles are given to nonresidents who live in Apia or overseas because of their job and who have no plan to return but who maintain their relationships with their parents and siblings resident in the 'āiga. There are also cases in which nonresident members of the second generation, with less attachment to their 'āiga, either ask for titles and receive them, or are offered titles by resident matai as if they were presents. And there are several cases in which nonresidents living in other 'āiga, possibly in other villages, ask for titles because, for some reason or another, they are unable to have ones in their resident 'āiga.

There is considerable variation among the relationships which those nonresident titleholders maintain with their 'āiga. On the one hand, there are those who have stopped making any contributions, keeping only the title names. On the other hand, there are those who keep visiting often and make large contributions frequently not only to their 'āiga but also to the whole village and to each of the villagers, possibly in order to promote some political ambition. Between these extremes are those who come back when their 'āiga has ceremonial exchanges or who give some contribution when asked even if they are unable to attend personally. I shall now consider from another perspective the reason why the numbers of nonresident titleholders are increasing.

From the standpoint of the traditional model matai system, it is almost impossible to take care of an 'āiga completely without living on 'āiga land. Why, then, do those resident members who protect and live on 'āiga land give titles to nonresident members who do not need access to the land? (Actually, if any members of an 'āiga should bring to the notice of the Court the fact that a matai has not lived in, and has neglected, the 'āiga, the Registrar of the Court is authorised to remove his title, but scarcely anybody makes such a complaint.) Since resident members, who are quite dependent on the 'āiga system, have much say about who may succeed to a title, why should they allow 'āiga to distribute titles to nonresident members? Here, I shall discuss the benefits for resident members of conferring titles on nonresident members and, as well, the benefits for nonresident members of receiving titles.

First, I shall examine the standpoint of resident members, especially of resident matai. A matai has to assume two kinds of roles: a manager or caretaker of the land estate and the members under him; and a ritual performer of the 'āiga. In order to assume the performer's role, he has to possess detailed knowledge of ceremonial protocols and speech-making, including fa'alupega - 192 and knowledge of genealogies and legends, plus the ability to collect enough valuables for ceremonial exchange.

The exchange valuables used to be tōga (female valuables) of female production—fine mats and bark cloths — and 'oloa (male valuables) of male production — food such as pigs and taro, weapons and tools. In the old days, people raised more pigs by themselves, and it took a long time to prepare special large pigs for funerals. In examining the transformation of the matai system, it is important to note that these valuables have undergone a significant transformation. Nowadays, Samoans use many kegs of salt beef and cartons of canned fish instead of pigs, and they use cash as a category of valuables in a certain context instead of food. This transformation of valuables has also introduced inflation into the exchange system as it is relatively easy to collect valuables by collecting money. And 'āiga have become all the more competitive because of this inflation. The exchange practices have been more and more escalated. 19 In effect, people have come to need progressively more large amounts of money whenever they have a ceremonial exchange. This transformation has, in fact, brought a crisis to the matai system: resident 'āiga members, who have enough resources to maintain its subsistence economy, cannot also produce what the 'āiga needs in order to make the appropriate exchanges to maintain its prestige as an 'āiga. This situation never happened in the old days when every 'āiga produced enough pigs and taro, as well as fine mats. People need cash to pay school fees, church donations, buy cloth, sugar, flour, etc., these days, but these expenses are fairly small and can be planned; they need a large amount of cash for ceremonial exchange and one cannot plan ahead for some things, such as funerals (Yamamoto and Yamamoto 1981:97-138; M. Yamamoto 1990).

Thus, an 'āiga has to obtain money from outside and this is largely responsible for the significant increase of nonresident matai. Nonresident members make their own living without using any resources of their 'āiga. Nonetheless, they have an obligation to send money to their 'āiga on occasion if they are children of resident members of the 'āiga. The matai may ask nonresident siblings, or sometimes nephews and nieces, to contribute for important exchanges. It is very difficult to ask someone other than these relatives to give these contributions and, therefore, it may become an advantageous strategy for resident matai to confer a title on a nonresident member in order to incorporate his contribution. Someone who works in Apia or in an overseas country can easily be such a target. Whatever his genealogical relationship to the 'āiga may be, he has an obligation to maintain 'āiga prestige once he becomes a matai of the 'āiga, so that the 'āiga can expect certain contributions from him. Though an urban working nonresident titleholder has his own latent claim to the resources of the 'āiga, it is less likely - 193 that he would exercise such a claim to undermine the distribution of resources among resident members. Thus, giving a title to a nonresident member is a potentially profitable possibility for a resident matai, as he may expect contributions without giving out any resources. And, if a nonresident member who has received a title should not contribute anything as expected, the resident members can refuse to give another title to anyone in the next generation of such a member's line on the grounds that he has not contributed.

A title may often be given to a nonresident member without such a strategic intention. If a nonresident member had helped on several other occasions and were willing to give a contribution when asked by a resident matai who was in a bind, unable to collect enough valuables for a sudden ceremonial exchange occasion, it would not be at all strange that the matai should give him a title in recognition of his contributions to the 'āiga, since helping matai who are in trouble is considered a worthy service (tautua). And, when such a generous nonresident member asks for a title, the resident matai will probably not reject the request, as in the case of 'āiga QY detailed above. Even though the 'āiga might be ambivalent about admitting him, such a nonresident matai would probably continue to give his contributions.

There is another advantage in giving a title to a nonresident member besides the economic ones described above. An 'āiga often gives titles to nonresidents such as Government officials, well-educated or well-known members who have never lived in the 'āiga. Such title-conferring is made to seem like a kind of reward to recognise their accomplishments which are another form of tautua (of indirect benefit to the 'āiga, since its fame is increased), and, at the same time, it is a strategy to make the 'āiga well-known by linking it to the name of such conspicuous persons. It is not rare for such an elite person to be offered titles by several 'āiga related to him at the same time. As these prestigious title recipients have good cash income in most cases, they are likely to contribute readily for ceremonial exchanges.

For a long time, 'āiga have often given titles to 'afakasi (part-European Samoans) living in Apia, 20 rather than to members living in the 'āiga and serving the matai. Keesing warned about this trend more than 50 years ago, saying it was contrary to custom and would be fatal to the matai system (Keesing 1934:250). Nevertheless, part-European Samoans were rather well off, as they specialised in trading while living in Apia, so that an 'āiga could expect contributions to bring cash into the 'āiga. It is not a recent phenomenon when an 'āiga looks to its outside members for assistance.

I shall now examine this matter from the standpoint of nonresident members. Why do nonresident members take titles even though they are thus burdened with the obligation of ceremonial exchanges without any prospect of exercising their latent claim on 'āiga resources? This is a delicate question, - 194 as everybody talks about it in terms of “alofa (‘love’) to the 'āiga”, but at least two answers are possible.

First, the respect given to a matai is very important in Samoan society. A matai is addressed by his special title name and addressed and spoken of with honorifics, not only when visiting the village but also in public circles in Apia and emigrant communities overseas. He may join a matai group chatting. If his title name is high-ranking, he can not only join circles of matai but also become acknowledged even by other matai. Considering the ratio of titleholders to the total population, which has been roughly more than 10% in Western Samoa, it might easily be assumed that becoming a matai in one's middle age is a normal life stage for an average male, even if he lives in Apia. Conversely, it might bring a kind of unhappiness to lack a title in one's middle or old age. Or it may not necessarily be an obsession for a title but a feeling of something missing that leads an untitled man to take a title in his middle age. In interviews I carried out in 1979, young people who had graduated from universities often told me that they were not interested in receiving titles, as they were just tokens like medals and useless for their lives. However, many of them have taken titles since then. And as for Samoans overseas, it appears that titles are less useful if they live overseas, but titleholders are still respected in overseas Samoan communities.

For a nonresident matai living in another village in Western Samoa, a title really has a great meaning. An untitled person in middle age living in a village needs one, as he often has set up his own household, being obligated to render service to his matai under whose auspices his household live. If he has a title, even though it is from another village, he no longer needs to render service to the matai. He does not need to give tributary food (sua) nor produce as many valuables as he otherwise would, though he is obligated occasionally to contribute at ceremonial exchanges of the 'āiga where he holds a title. And, if he makes a formal food presentation to the council of matai of the village where he lives, he may join in the council, though his status in the council would be as low as a common matai irrespective of his title name. If an untitled person finds it difficult to receive a title in the 'āiga where he is a resident member, he often asks for one from 'āiga to which he is connected. It is interesting to note that his taking a nonresident matai title often stimulates the unwilling 'āiga of his residence finally to give him one of their titles, because, once an untitled resident becomes a matai of a different 'āiga, the 'āiga of his own residence loses a large part of his contributions in many ways which otherwise it would have certainly obtained. This was probably the case with J and K of the 'āiga Sā R.

The other answer to the question is that the title recipient secures the latent - 195 claim in an 'āiga. By receiving a title, a nonresident secures the possibility of coming back from any place to his 'āiga and claiming his share of the 'āiga land in case he loses his job. Of course, theoretically, he may come back whether he is a matai or not, but he does not want to come back begging the mercy of resident matai to live under them. And, by receiving a title, he may also secure the latent claim on behalf of the whole nonresident branch of his near kinsmen such as sons and daughters, grandsons and granddaughters, and nephews and nieces. This is because a Samoan 'āiga often recognises a claim to membership according to whether one has a direct or near kinsman, alive or deceased, who was a matai of the 'āiga. The nearer one's relationship to such a kin, the better founded the claim. A title was once offered to a successful Samoan living in New Zealand who had no prospect of coming home. He told me that he was rather reluctant for some time before he finally made up his mind to accept the title, believing that it might be useful for his descendants some time in the future.

The above analysis shows that both resident and nonresident members have an interest in creating nonresident titleholders. Resident matai's strategy of conferring titles on nonresident members for the purpose of integrating them into the whole 'āiga is successful, since both sides are interested in maintaining the system.

The foregoing discussion has outlined the process of transformation of the Samoan chiefly system under the influence of social urbanisation. A matai has a wide range of roles, including: to live on the 'āiga land, to manage 'āiga estate, to control 'āiga manpower, to provide moral leadership to 'āiga members, to represent 'āiga in the council of the village to which the 'āiga belongs, and to participate in ceremonial exchanges. An ideal matai should be not only a good manager, but also a traditional intellectual well versed in ceremonial protocol and oral traditions, as well as a moral leader of the whole 'āiga and a person with a good salary so as to spend a great deal of money in ceremonial exchanges. Is such a leader possible in modern Samoa? Even if a matai were able, would any untitled youths work for him? Most 'āiga are short of manpower these days as many of their able youths have emigrated overseas. A matai, who is described by Samoans as having sat in the house all day giving commands to untitled members of the 'āiga, nowadays actually has to work himself, planting root crops for subsistence, taking care of cash crops such as banana, cacao, copra and taro, if he is a resident matai. So the traditional roles of matai have to be differentiated among (a) those resident matai who oversee 'āiga land without enough manpower left in the villages which are generally losing their young population and (b) those nonresident - 196 matai who have emigrated and taken up modern jobs and send or bring back money to 'āiga for ceremonial exchanges. The tendency of the role differentiation has been intensified by the inflation of ceremonial exchanges, and, in the dialectic process, the role differentiation has, in its turn, intensified the inflation. It may not be an exaggeration to say that the modern 'āiga system is possible only because of this role differentiation between resident and nonresident matai, for almost no 'āiga is self-sufficient without any assistance from nonresident matai or even nonresident members living in Apia or overseas. 21 Overseas members are likely to become nonresident matai eventually, because of their contributions to the 'āiga. Nowadays, the incorporation of nonresident matai is vital to the matai system.

It would be interesting to study how long and to what extent the matai system will keep this strong power of cultural integration. At present, Samoan urbanisation and modernisation connect villages to Apia and to overseas countries and, while the population moves slowly along this vector, the major Samoan value orientation still focuses on the shrinking traditional villages. Whenever two or more titled persons meet, whether in Apia or overseas, they address each other using fa'alupega (formal greetings), confirming their relative positions in the whole system of traditional social organisation. Each individual is related to every other in terms of the relationships of titles each of which has its own specific place in the total universe of the Samoan chiefly system. When a holder of a high title or a member of an important 'āiga dies overseas, the body will often be brought back by plane to Samoa for the funeral and, of course, it will be buried on 'āiga land.

Wherever a Samoan lives and whatever he does, he belongs to the place where he should be: to the land of 'āiga Sā X of Village Y in the Samoan Islands of the South Pacific. His title is proof of his invisible tie with his 'āiga. A title given to a nonresident is a medium which unites him with his special spot on the earth. By these media, money required to maintain the chiefly system flows from overseas to Apia and to villages, and from Apia to villages. We have seen that title multiplication is deeply rooted in the Samoan chiefly system itself. The election system of matai suffrage was abolished recently because of the dominant opinion that it has caused much title-splitting and thus degraded matai prestige. But in the present analysis, it becomes clear that the election system was not the sole reason for title multiplication. It is true that title multiplication has degraded matai prestige and may cause the collapse of the chiefly system at some time in the future but, at the same time, it allows the chiefly system to adapt to the socio-economic conditions of the world system. 22

- 197

The initial research on which this article is based was conducted in June 1978 - Sept 1979. Since then, there have been several short-term visits. I appreciate the financial support for theseresearch projects given by the East-West Center, Hoso Bunka Foundation, Japanese Ministry of Education, and Toyota Foundation. This article is based on my earlier Japanese paper “A chiefly system urbanized: Role differentiation among the title holders in Western Samoa” (Toshi-ka no nakano shuchō shisutemu — Nishi Samoa ni okeru shuchō shōgō hojisha kanno yakuwari bunka —), which appeared in the Bulletin of the National Museum of Ethnology, special issue 6:301-29., 1989. Although the basic discussion is not changed, I have rewritten substantially and updated the data. Earlier versions of this article were presented at a symposium of 17th Pacific Science Congress, on O'ahu, May 1991, and at a session of the Association for Social Anthropology in Oceania, in New Orleans, February, 1992. Among many of my friends and colleagues who helped and encouraged me, I am especially grateful to Nelson Graburn and Bonnie Burns. I appreciate the kindness of C. Guy Powles, who sent me a copy of his thesis. But I am most thankful to the Samoan people who kindly helped and supported my research in various ways.

- 198 Page of endnotes

- 199 Page of endnotes

- 200

• Ala'ilima, Fay C. and Vaiao J. Ala'ilima, 1966. Consensus and Plurality in a Western Samoan Election Campaign. Human Organization, 49:240-55.
• American Samoa Government, Economic Development and Planning Office, 1988. American Samoa Statistical Digest 1988. Pago Pago: American Samoa Government, Economic Development and Planning Office.

• Congregation Christian Church of Samoa (London Missionary Society), 1978. O le Tusi Fa'alupega o Samoa. Apia: Malua Printing Press. (First Edition in 1940s)

• Davidson, James W., 1967. Samoa mo Samoa. Melbourne: Oxford University Press.
• Douglas, Norman and Ngaire Douglas, 1989. Pacific Island Yearbook. 16th edition. New South Wales: Angus and Robertson.

• Franco, Robert W., 1987. Samoans in Hawaii. A Demographic Profile. Honolulu: East-West Center.

• Handy, E.S.C. and W.C. Handy, 1924. Samoan Housebuilding, Cooking and Tattooing. Bulletin 15. Honolulu: B. P. Bishop Museum.

• Keesing, Felix M., 1934. Modern Samoa. Its Government and Changing Life. London: Allen and Unwin.
• —— and Marie M. Keesing, 1956. Elite Communication in Samoa. A Study of Leadership. Stanford: Stanford University Press.
• Krämer, Augustin F., 1902. Die Samoa-lnseln: Entwurf einer Monographie mit Besonderer Berü;cksichtigung Deutsch-Samoas. 2 vols. Stuttgart: N. Naegele.

• Marcus, George E., 1980. The Nobility and the Chiefly Tradition in the Modern Kingdom of Tonga. Wellington: Polynesian Society.
• Marsack, C. C., 1961. Notes on the Practice of the Court and the Principles Adopted in the Hearing of Cases Affecting (1) Samoan matai Titles and (2) Land Held According to Customs and Usages of Western Samoa. (revised edit.) Apia: Government Printer.
• Mead, Margaret, 1943. Coming of Age in Samoa. Harmondsworth, Middlesex: Penguin Books, (first published in 1928).
• —— 1969. The Social Organization of Manua. Bulletin 76. Honolulu: B. P. Bishop Museum, (first published in 1930).
• Meleisea, Malama, 1987. The Making of Modern Samoa. Suva: Institute of Pacific Studies, University of the South Pacific.

• New Zealand Government, 1973. Official Yearbook.. Wellington: Department of Statistics.
• —— 1985. Official Yearbook. Wellington: Department of Statistics.
• —— Department of Statistics, 1988. Birthplaces and Ethnic Origin. New Zealand Census of Population and Dwellings, Series C, Report 6. Wellington: Department of Statistics.

- 201

• O'Meara, Tim, 1990. Samoan Planters: Tradition and Economic Development in Polynesia. Fort Worth: Holt, Reinhart and Winston.

• Pitt, David, 1970. Tradition and Economic Progress in Samoa. Oxford: Clarendon Press.
• —— and Clunie Macpherson, 1974. Emerging Pluralism. The Samoan Community in New Zealand. Auckland: Longman Paul.
• Powles, Charles Guy, 1979. The Persistence of Chiefly Power and Its Implications for Law and Political Organisation in Western Polynesia. Unpublished Ph.D. thesis, Australian National University.

• Samoa Observer, 1990. Letter to the Editor, Friday 20th July, 1990.
• Samoa Times, 1988. 19th Feb.
• Schültz, E., 1911. The Most Important Principles of Samoan Family Law, and the Laws of Inheritance. Journal of the Polynesian Society, 20:43-53.
• Stair, J.B., 1897. Old Samoa. London: Religious Tract Society.
• Sugimoto, Hisatsugu, 1982. A Japanese Chief in Western Samoa—A Research Diary 1965-1980 (in Japanese). Kyoto: Kokin Shoten.
• Suttles, Wayne, 1991. Streams of Property, Armor of Wealth: the Traditional Kwakiutl Potlatch, in Aldona Jonaitis (ed.), Chiefly Feasts: The Enduring Kwakiutl Potlatch. Seattle: University of Washington Press.

• Tiffany, Sharon, 1975. Entrepreneurship and Political Participation in Western Samoa: a Case Study. Oceania, 46:85-106.
• Tiffany, Walter W., 1975. High Court Adjudication of Chiefly Title Succession Disputes in American Samoa. Journal of the Polynesian Society, 84: 67-92.
• Turner, George, 1884. Samoa, a Hundred Years Ago and Long Before. London: Macmillan.

• Western Samoa Government, Department of Economic Development, 1975. Third Five Year Development Plan 1975-1979. Apia:Department of Economic Development.
• —— 1984. Western Samoa's Fifth development Plan 1985-1987. Apia: Department of Economic Development.
• ——, Department of Statistics, 1967. Statistical Yearbook. Apia: Department of Statistics.
• —— 1978. Annual Statistical Abstract. Apia: Department of Statistics.
• —— 1980. Annual Statistical Abstract. Apia: Department of Statistics.
• —— 1982. Annual Statistical Abstract. Apia: Department of Statistics.
• —— 1989. Annual Statistical Abstract. Apia: Department of Statistics.
• Western Samoa Parliament, 1979. Parliamentary Paper 1979, No.13. Apia: Legislative Department.

• Yabuuchi, Yoshihiko and Hisatsugu Sugimnoto, 1967. Polynesia—Family, Land, Household [in Japanese]. Tokyo: Taimeido.
• Yamamoto, Matori, 1986. Samoan Family [in Japanese] in Hiroko Hara (ed.), Ethnography of Families. Tokyo: Kobundo, pp.117-36.
• —— 1987. Territorial Organization of Faleata: a Case Study of the Title System in Samoan Society, in I. Ushijima and K. Sudo (eds), Cultural Uniformity and

- 202

• Diversity in Micronesia. Senri Ethnological Studies 21. Osaka: National Museum of Ethnology, pp.205-37.
• —— 1989. A Chiefly System Urbanized: Role Differentiation Among the Title Holders in Western Samoa [in Japanese]. Bulletin of the National Museum of Ethnology, Special Issue 6:301-29.
• —— 1990. Transformation of Exchange Valuables in Samoa. Man and Culture in Oceania, 6: 81-98.
• Yamamoto, Yasushi, 1987. Comparative Sociology of Contextualist Society [in Japanese], in Munesuke Mita and Takashi Miyajima (eds), Culture and Society, 291-324. Tokyo: University of Tokyo Press, pp.205-37.
• —— and Matori Yamamoto, 1981. Prohibition of Consumption/ Prohibition of Sexuality (1) Structure of the Exchange System in Samoan Society [in Japanese]. Bulletin of the Institute of Journalism and Communication Studies, 29: 67-186.