The island of Rotuma is relatively remote, located four hundred sixty-five kilometres north of the northernmost island in the Fiji group. Rotuma has been politically affiliated with Fiji for more than a century, first as a British colony and since 1970 as part of the independent nation. Rotuma's people are, however, culturally and linguistically distinct, having strong historic relationships with Tonga, Samoa, and other Polynesian islands. A fertile volcanic island of forty-three square kilometres surrounded by a fringing coral reef and a number of offshore islets, Rotuma is divided into seven districts, each headed by a titled chief.
During the late nineteenth century, colonial officials and Rotumans alike expressed concern about Rotuma's depopulation,  but their apprehensions have not been realised. Figure 1 presents the growth of the Rotuman population from the first census of 1881 to the census of 1986. 
It is clear from this figure that Rotumans have experienced explosive population growth since the 1920s, growing from a total population of 2,112 in 1921 to 8,652 in 1986. However, the number of Rotumans on Rotuma was nearly the same in 1986 as it had been 50 years earlier (2,554 compared to 2,543), with migration draining off the net population increase (Table 1).
Fiji census reports over the past several decades document a dramatic shift in the distribution of Rotumans, with an ever-increasing proportion recorded away from their home island. According to the 1986 census, seventy percent of Rotumans lived elsewhere in Fiji. Although official counts in other countries do not enumerate Rotumans separately, data collected from Rotuma residents in 1989 and from Rotuman migrants to Australia, New Zealand and the United States in 1994 suggest that between 500 and 1,000 Rotumans live abroad. 
The effect of this outmigration on Rotuma's age and sex structure can be seen in Figure 2, which shows the shape of the island's population structure in relation to the total Rotuman population within Fiji. In 1956 the age structure of the total Rotuman population reflected increases in numbers resulting from a decrease in death rates and continued high birth rates. Thus it had a broad base of children and tapered toward a peak at old age. The age structure of Rotuma paralleled that of the total Rotuman population, suggesting that migration to that point had included both sexes and all age groups in approximately equal proportions (with a slightly higher number of males in the 20-29 age group leaving, and fewer of the elderly).
By 1966 one sees the effects of increased outmigration among young men; there is a distinct indentation in the 20-29 age group for males on Rotuma. The overall population continued to show the effects of rapid increase, with even more broadening of the base. By 1976 outmigration has affected a broader segment of the population. A pronounced reduction in the proportion of women on Rotuma in the 20-29 age group then appeared alongside that of men, and the structure of the island's population above the age of 20 began to look more like a column than a pyramid. In contrast, the structure for the total population resembled an even more sloped pyramid than before, although a reduction in the size of the 0-4 age group suggests a decline in the birth rate. (As can be seen from the 1986 figure, however, the decline was only temporary.) By 1986 the difference between the structure of the overall Rotuman population and that of the island was more dramatic still. Although the overall structure retained the shape of a broad-based pyramid, the population on Rotuma resembled the shape of an hourglass, with smaller proportions of young children than previously, an indentation in the middle age groups, and relatively high proportions in the older age categories. This suggests that outmigration increasingly has involved young couples who either migrated with their children, or left Rotuma single, married in Fiji and had their children there.
Outmigration has also had an effect on household size. According to census reports the number of persons per household decreased from 7.1 to 5.9 persons on Rotuma between 1966 and 1976. This corresponds to the period of maximum outmigration, when the population of Rotumans on the island dropped by sixteen percent.
Meanwhile, the proportion of children under 10 on the island declined with each census, from 34.2 percent (1,024 of 2,993 total population) in 1956 to 27 percent (699 of 2,554) in 1986. Although the percentage of children age 10-14 fluctuated, there has been an overall decrease, such that the total proportion of children under 15 dropped by nearly 10 percent (from 48.4 to 38.8 percent, or 1,449 to 1,004 in absolute numbers) over the thirty year period. This change may be attributed at least in part to changing migration patterns. An examination of dependency ratios over time is instructive.
The 1976 census shows a high dependency ratio for Rotuma (118 dependents to 100 adults of 'working age,' that is, ages 15-59) (Bryant 1990:140). But by 1986 the dependency ratio for Rotuma had dropped to 96/100. Furthermore, a comparison of the Fiji censes from 1956-1986 shows a steady increase in the dependency ratio for Rotumans in Suva, from 58/100 in 1956 to 67/100 in 1986. Whereas previously the Suva population of Rotumans included a higher proportion of pioneers, without spouses and children, because they now establish families-and keep their children with them-the population profile approaches that for the overall Rotuman population. The dependency ratios for Rotumans on Rotuma and in Suva both appear to be converging toward the overall dependency ratio for Rotumans, which in 1986 was 76 dependents per 100 of working age (Figure 3).
Although the dependency ratio is dropping on Rotuma, the percentage of the population 60 and over has more than doubled, from 4.3 percent (129 persons) in 1956 to 10 percent (258 persons) in 1986. By 1986, 50 percent of Rotumans 60 and older (258 of 519) were on Rotuma, whereas only 30 percent of the total Rotuman population lived there. This may be due in part to the fact that, more so than urban Fiji, Rotuma provides an environment in which older people are valued for their knowledge, wisdom and other contributions to their households and communities.
The high rate of emigration for Rotumans of working age is understandable. Fiji's diversified economy provides a broad base of employment whereas Rotuma's does not. Rotumans in Fiji are employed not only by the government but by private organisations; according to the 1976 Fiji census, 583 Rotumans worked for the government while 1,042 held positions in the private sector. After young Rotumans leave the island in search of further education and employment, many opt to stay away, to marry and establish families and residences of their own. Some choose to go back to Rotuma, for shorter or longer periods, to visit, take a job, find a spouse, or resettle. Whether or not they return, many Rotuman migrants actively maintain connections with their home island.
Table 2 shows that only 29.5 percent of all Rotumans were living on Rotuma in 1986, while 28.3 percent were recorded as living in Rewa district, 19.7 percent were recorded as living in Naitasiri and 14.6 percent were recorded as living in Ba. Comparatively, while indigenous Fijians were also heavily concentrated in these areas, they are far more evenly distributed across the country. The Index of Difference  for Rotumans and Fijians suggests that almost 30 percent of Rotumans would have to relocate to approximate the population distribution of indigenous Fijians.
Off-island Rotumans are heavily concentrated in the major urban centers of Fiji, with the largest concentration in the Suva/Lami area (64.1 percent). An additional 9.5 percent were in Lautoka and 6.1 percent in Vatukoula. Fully 87.8 percent of Rotumans living off-island were classified as 'urban' in the 1986 census, representing 61.9 percent of the total Rotuman population.
The 1986 Census shows Rotumans reporting high rates of educational attainment, with 58 percent completing Form One or higher and over 4 percent reporting at least some post-secondary education. In contrast, Fijians reported 47 percent having completed Form One or higher, and only 1.5 percent some post-secondary education. The Index of Difference indicates that Fijian educational patterns would have to shift 17 percent to approximate the somewhat higher rates of educational attainment seen among Rotumans. While some of this difference may be attributed to differences in age structure, the effect is minimal as both populations are relatively young.
Table 3 presents differences in occupational activity for 1986. The majority of Rotuman men were engaged in cash-based employment (57.1 percent) while Rotuman females were more evenly split between cash employment and the role of homemaker (27.5 percent versus 40.1 percent).  After cash employment, the most likely activity for Rotuman males was that of student (9.6 percent). A similar, though slightly less marked, relationship is seen for Fijians with 55.3 percent of Fijian males engaged in cash employment, while 6.7 percent are students. The somewhat higher percentage of total Rotumans engaged in cash work is accounted for mostly by more women in the work force (27.5 percent of Rotuman women versus 17.8 percent of Fijian women). Correspondingly, a smaller percentage of Rotuman women are categorised as homemakers.
The projected populations for Rotumans in 1996 and 2006 are presented in Table 4.  These numbers are presented graphically in Figure 4. Even with moderate fertility decline the greatest growth is among the youngest Rotumans across the projection period. It should also be noted, however, that due to an anticipated mortality decline, the elderly population grows as well. This group, while small, requires serious policy considerations since a growing population of elderly requires the development of specialised medical and social support infrastructures.
Figure 4 presents population pyramids for Rotumans across the projection period, with the 1986 population representing the lowest bar of each age group, the 1996 population representing the middle bar, and the 2006 population representing the top bar of each age group. This figure shows that while the current population of young Rotumans is large, the proportion of middle-aged Rotumans is also increasing. If fertility continues to decline, aging Rotumans may have fewer children to provide them with care and support when they can no longer engage in productive labor force activity.
The overall picture of Rotuman population growth is one of moderate increase in the foreseeable future. This will contribute to Rotumans' maintaining a relatively young population into the twenty-first century. Continued fertility decline will reverse this trend eventually, but a shift is unlikely to occur in the near future. Rotumans therefore have sufficient time-from forty to sixty years-to plan for a transition from a youthful to an aging population.
While the Rotuman population comprises only a small segment of the total population of Fiji, it has grown substantially over the past century. We expect this trend to continue for the foreseeable future, though it is highly unlikely that Rotumans will become a numeric force in Fijian population dynamics. Nonetheless, contributions to a society cannot be measured in mere numbers. In this regard, Rotuman demographics present a picture of a well-educated, economically productive people who should be viewed as an asset to the nation of Fiji as a whole.
At the time of discovery by Europeans in 1791 Rotuma was divided into seven districts, each relatively autonomous and headed by a gagaj 'es itu'u 'district chief.' There were also three positions that were pan-Rotuman in scope: the fakpure, sau and mua. The early literature primarily refers to the fakpure in two capacities: as convener and presiding officer of the council of district chiefs, and as the person responsible for appointing the sau and ensuring that he was cared for properly. The fakpure was gagaj 'es itu'u of one of the districts, presumably the one who headed an alliance that was victorious in the latest war. The sau was an object of veneration whose basic role was to take part in the ritual cycle, oriented towards ensuring the island's prosperity. The role of mua received less commentary in the early literature than that of fakpure and sau, but most of what was written refers to the mua's activities in the ritual cycle. A French priest, Fr. Trouillet, wrote ca. 1873 that the sau appeared to be an appendage of the fakpure, while the mua appeared to be more associated with spiritual power (Histoire de la Station...).
Most early accounts focus on the office of sau (generally mis-translated into English as 'king'). Early observers generally agree about some aspects of the sau's office. Most reported that the sau was appointed by the fakpure and that he was chosen from different districts in turn, although no order of rotation is specified. However, Dillon (1829:95), who visited the island in 1827, wrote that the district chiefs 'meet in congress every six months, when they elect a president (sau) and deliberate upon state affairs,' and Bennett (1831:473) wrote that 'The royal office is held for six months, but by the consent of the other chiefs, it may be retained by the same chief for two or three years.'
The consensus among the early reports is that the sau exercised no secular power and that his main tasks were to eat gluttonously, drink kava and take part in the six-month ritual cycle. There is less agreement on several other points, however. For example, it is unclear who was eligible to be selected as sau, although those who commented generally agreed that eligibility was limited to individuals of chiefly lineage. Whether a person was actually supposed to hold a title in order to be eligible is unstated. The length of the sau's reign is also unclear. Gardiner (1898:461) states that although the term of office was for six months (one Rotuman ritual cycle), an incumbent sau could continue in office as long as he could accumulate the great masses of food that he was required to provide. Since he did not provide food by working, this may mean either that he was allowed to remain in office as long as the island prospered, or that his reign was extended as long as the people in the district where he stayed were prepared to bear the burden of providing the surplus food needed to maintain feasting at an appropriate level. Historical data summarised by Howard (1985:70) indicates that the average length of a sau's reign diminished from 2.5 years (5 ritual cycles) between 1797 and 1820, to 1.0 years between 1820 and 1850, and then to .6 years between 1850 and 1870. One might hypothesise that this decline resulted from acculturative factors that increased the burden of caring for the sau at the same time that it was becoming increasingly difficult for fakpure to exert secular power to enforce compliance. A second possibility is that the diseases and other misfortunes brought by Europeans, which resulted in depopulation, led Rotumans to question more intensely the effectiveness of individuals who occupied the office of sau.
A further puzzle concerns the rules of residence for sau. Allen (1895) reported that the district whose turn it was to select a sau would go to a neighbouring district, choose someone, and bring him to their own district to live, and in one narrative recorded by Churchward (1938:356), the story-teller stated that if it was one district's turn to provide the sau, it would be another's turn to look after him. Indeed, Trouillet's oral history records numerous movements of the sau from one district to another although no regularities appear. Perhaps all that can be said is that Rotumans characterised sauship in terms of interdistrict residence, possibly as a way of emphasising that the role was pan-Rotuman in scope.
The position of mua was not extensively documented by the early observers but has generally been referred to as the head priest of the island (Histoire de la Station...). The main task of the mua was to bring prosperity to the island by incorporating the power of supernatural beings (Churchward 1938:356). One of the mua's duties was to preside over a ritual procession for the relief of drought or famine. The mua also held special prayers for a good harvest three times a year (Macgregor 1932).
Allardyce (1895-1896:142) was told that the position of mua rotated between districts. He reports that the fakpure appointed the mua for indefinite periods, although it was customary to resign after a year (most likely a Rotuman year, or six lunar months). Wood (1875) claimed that the mua was elected by capture and therefore always resided in the most powerful district of the island. Consultants to Macgregor (1932) indicated that the mua and sau were rarely housed in the same district as the expense was too great for a single district to bear.
The positions of sau and mua were abandoned in the early 1870s, largely as a result of missionary pressure; the missionaries perceived the rituals associated with these offices as pagan rites. 
Land tenure: an historical overview
According to Gardiner (1898), the pre-contact land-holding unit was the ho'aga, a kinship community under the direction of a titled subchief, the fa 'es ho'aga, who acted as steward (pure) of the land. It was his prerogative to divide it among ho'aga households for planting purposes. He was also responsible for settling disputes within the ho'aga. District chiefs (gagaj 'es itu'u) did not exercise control over ho'aga lands although they were given first fruits and were called on to settle disputes between ho'aga.
This arrangement was altered by three historical factors following European contact: a population decline, the development of a commercial economy, and the establishment of Christian missions. As adjacent ho'aga joined together in order to maintain adequate manpower in the face of depopulation, the kinship unity of the group was diminished, and ho'aga evolved into localised work units without regard to land holdings. At the same time, the growth of the copra trade gave households a longer-term interest in specific plots of land than previously, and with the encouragement of missionaries and traders, concepts of individual ownership (the right of the pure to allocate and transact land) emerged. In time, the right of the fa 'es ho'aga to distribute land gave way to the rights of household heads and their descendants. This process of fragmentation was furthered by land sales for money, pigs, and other goods.
After Rotuma was ceded to Britain in 1881, the colonial administration established the resident commissioner as magistrate and interpreter of custom. A land tax was implemented, the payment of which served to legitimise land claims. During the first two decades following cession the commissioners followed a policy of resolving land disputes by dividing holdings-a policy that intensified fragmentation. As the population rebounded, a point of diminishing returns was reached and a trend toward declaring joint rights developed in cases where neither plaintiff nor defendant held a judicial advantage. This shift, along with an expanding population, led to a dramatic increase in communal land holdings.
Kinship and land tenure
The basic concept behind Rotuman kinship is the word kainaga, which in its broadest sense means 'kind, sort, variety, species, class' (Churchward 1940:235), in other words, 'members of the same category.' Since kinship on Rotuma is traced bilineally (through both mother and father), a person's kainaga consists of all their 'blood' relatives. The term kainaga is also used in the more restricted sense of common descent from an ancestor who resided at, and held rights in, a given house site (fuag ri 'foundation'). Each person is said to have rights in the fuag ri of their eight great-grandparents, although they may selectively exercise claims in only three or four. House sites are named and people describe their affiliation by referring to these names (e.g., 'I belong to the Rirou kainaga.') Associated with each fuag ri are sections of bush land, and to claim rights in a fuag ri is to claim rights in these lands. Ordinarily, the senior member who lives on the fuag ri controls the land in the name of the kainaga. As steward of the land (pure) he or she is obligated to grant use privileges to other members of the kainaga. If a pure is unreasonable or overly stingy, the kainaga have a right to hold a meeting and depose him or her in favor of another person. If the pure dies or otherwise leaves the ancestral fuag ri, the kainaga are supposed to meet to choose a successor. Theoretically succession goes from elder brother to younger brother, to eldest son of eldest brother, to youngest son of eldest brother, to youngest son of younger brother, etc. Women succeed to pureship only if there are no eligible males.
This traditional model of kinship and land tenure has been undergoing change as a result of several interrelated factors, most importantly extensive migration, housing changes, and increased commercialisation. Outmigration has resulted in whole families moving to Fiji or abroad, with the prospect of leaving land rights in the hands of distant relatives who may be reluctant to relinquish the land should migrants return. As a result, some families designate one member to stay on Rotuma to occupy their land. Siblings may take turns over a period of years assuming this responsibility. In other instances, migrants may simply allow their kainaga rights to go dormant. With so many relatives away, succession is now more an informal process with minimal consultation.
Changes in house construction have also affected land tenure and concepts of ownership. Previously houses were made of thatch or limestone, local materials that were transformed into houses through communal labor. Houses could be erected in short periods of time at minimal expense. If a new pure was selected the move was not unduly burdensome. With modern housing, however, built with imported, purchased materials, individual families have much greater capital investment in their homes. As a result they are resistant to giving up their rights in houses or the land on which they are built. Extensive capital investments in housing are now recognised as sufficient justification for a family group to remain on a fuag ri. This has the added implication of strengthening the claims of immediate descendants not only to the site but to associated garden lands and titles, if any (see Rensel 1991; 1994:257-287).
Some fuag ri carry with them chiefly titles to which kainaga members can lay claim. Titleholders are known as as togi 'successor to the name' and are entitled to the privileges, and burdened by the responsibilities, that go with the title. The assumption of a title requires a ceremony (hül 'umefe) in which the symbol of chiefly status, a short-legged eating table ('umefe) is turned upright, kava is partaken, the candidate is anointed with oil by another titleholder, and a feast is eaten.
In 1951 H.S. Evans identified 121 titled men in a population of 2780; he estimated that seventeen percent of adult males held a title (Evans 1951).  At any given time only a portion of all known titles are taken; some remain dormant for indefinite periods. Men  who hold titles may or may not assume secular leadership positions, of which there are two levels: gagaj 'es itu'u 'district chiefs' and fa 'es ho'aga 'hamlet or village chiefs.' While district headmen are always titled, ho'aga leaders may or may not be. This suggests a conceptual separation between ceremonial and secular leadership roles.
It is the responsibility of fa 'es ho'aga to organise the labor of households under their direction on ceremonial occasions and when district work needs to be done. They are also responsible for organising collections of food and valuables for distribution when called for by the district chief, or when required by events involving their group.
At ceremonies titled men have special rights and responsibilities not afforded untitled men. They are honored in kava ceremonies at which their titles are called out in rank order, and are expected to give speeches on behalf of their group. Titled men are also expected to be more generous when presentations of food and valuables are required.
Titles 'belong' to the descendants of previous title holders. In most districts three or four kainaga claim rights to a title suitable for the district chief. Collectively these groups are referred to as mosega 'a bed,' the implication being that the claimants are descendant from the same original source. Eligible kainaga are thus related to each other through (often unspecified) ancestral siblings. Ideally, the gagaj 'es itu'u should be chosen successively from each branch of the mosega in turn, but in practice the process is highly politicised. The second ranking title in each district is that of faufisi, whose holder serves as the district chief's 'right hand.' He customarily acts as head of the district when the gagaj 'es itu'u is away. Lesser titles are bestowed on those occupying other special roles (such as head fisherman and messenger), ho'aga headmen, and title holders who play no role in district administration.
Titles are ordinarily held throughout one's lifetime, but if a man is particularly remiss in his role or otherwise earns the enmity of his fellow kainaga members, he may be pressured to give up his title. Rotuman custom requires a man who spends even one night in jail to forfeit his title. Whether the mosega has a right to take back titles once given is currently a matter of debate, although such a prerogative seems consistent with other aspects of Rotuman custom.
Colonial administration and the Rotuma district council
The overall impact of European contact on chiefly powers prior to British administration is difficult to estimate, since some changes increased the chiefs' authority while others reduced it. Thus commercialisation of the economy initially enhanced the status of chiefs, for they acted as intermediaries between their people and ships' captains; but commercialisation also contributed to individual control of land,  diminishing chiefly authority in that area. Likewise, while the missionaries worked through the chiefs and strengthened their hands in some nontraditional ways, they also undermined chiefly authority by institutionalising a new religious order over which the chiefs had little control (Howard 1966:63-78).
The colonial administration, having successfully instituted a system of indirect rule in Fiji, expected to do the same in Rotuma. They failed to take into consideration the differences in chiefly systems, however. In Fiji, where patrilineal primogeniture reinforced a hierarchical system of chiefly authority, obedience was institutionalised. In Rotuma, with its bilateral kinship emphasis, the contenders for a title were often numerous, with any ancestral link to a previous chief making a man eligible. The number of male children who might eventually succeed to chieftainship was therefore likely to be extensive, and no one was apt to receive the special privileges normally given a Fijian chief's elder sons. As a result respect for chiefly authority was far more conditional in Rotuma.
Under colonial rule the Rotuman chiefs apparently expected to be granted privileges commensurate with those enjoyed by their Fijian counterparts (Howard 1966:69), but the new administration resisted increasing their powers beyond what was allowed by Rotuman custom. The people objected neither to the imposition of English law nor to the authority invested in the resident commissioner. They accepted English laws and officials as a price for the material benefits they foresaw, but they were unwilling to enhance the power of their chiefs. A letter from one of the first resident commissioners to the Governor of Fiji, shortly after Cession, reports: 'I have repeatedly heard the people say we do not wish our chiefs to be placed in authority over us. We will obey the regulations made by the government but not the rules made by the chiefs.' 
A Council composed of the seven district chiefs was set up to advise the resident commissioners, but it had no policy-making or administrative powers of its own. Politically chiefs became little more than messengers between the resident commissioner and the people in the districts. The chiefs were criticised by their constituents for making unpopular demands on behalf of the commissioners, and by the commissioners for failing to gain the compliance of their subjects. As a consequence, the traditional rules governing succession, flexible as they were, gave way to a lax toleration allowing almost any adult male to fill a vacancy. It became commonplace for the people in a district to nominate several candidates and permit the commissioner to make the final selection.  Not only did the commissioners participate actively in choosing chiefs, they showed little hesitation in deposing men who failed to meet their expectations.
The problem for the resident commissioners, it seems, was that they saw Rotuman political institutions as neither fish nor fowl. The gagaj 'es itu'u did not have the kind of authority they associated with chiefdoms such as Fiji, but the system also lacked elements crucial to their understanding of democracy. They were determined to resolve the issue one way or the other. Whereas some opted to reinforce the status of chiefs (without, of course, giving up any real power themselves), others instituted moves toward democratic representation on the Council. In 1939, with the approval of the Governor of Fiji, Commissioner Cornish introduced a reform whereby chiefs would be elected for a period of three years, after which the kainaga who had elected him would elect a new chief, or re-elect the old one. The first chief appointed under this procedure failed to get re-elected. He complained to the Government against his dismissal on the grounds that the new procedures were not in accordance with Rotuman custom. By this time Cornish had died, and following an investigation the traditional custom was re-instated (Sykes 1948).
A few years later, J.W. Sykes, who was sent to Rotuma for the purpose of investigating the administration of the island, proposed that the Council of Chiefs be abolished and replaced by an elected council (Sykes 1948). Sykes' recommendation was not implemented, in large measure because it was opposed by H.S. Evans, the district officer  appointed to Rotuma the year after the report was issued.
However, in 1958 a compromise was reached and the Council was reconstituted to include one representative from each district, elected by secret ballot, in addition to the chiefs.  Its name was changed from the Rotuma Council of Chiefs to the Council of Rotuma. Its role, to advise the district officer and communicate his rulings to the people in the districts, remained the same.
The post-colonial period
Soon after Fiji gained independence in 1970 a confrontation took place between the district officer and the Rotuma Council. Under the colonial administration the D.O. had been the gagaj pure-the boss. His authority had come from the Governor, whom he represented, and ultimately from the British Crown. With independence the basis of his authority became ambiguous. The district officer at the time of independence was an educated Rotuman who had strong ideas about how Rotuma should be governed. According to reports, he intruded into the process of chiefly selection on several occasions and picked the person he favored, without regard to custom. When the chiefs complained to the newly formed government about his high-handedness, they met with immediate success. The Prime Minister himself came to the island and personally ordered the district officer's removal, replacing him with an experienced clerk. 
This action on the government's part completely reversed previous responses to Rotuman requests that district officers (or district commissioners before them) be disciplined or removed. It signaled the beginning of an entirely different relationship between district officer and Council. Whereas prior to Fijian independence the Council had been merely an advisory body, it was now empowered as a policy-making organization. The district officer was relegated to the role of advisor and administrative assistant to the Council. This meant that Council members-chiefs and district representatives alike-were in a position to exercise real power for the first time since Cession. However, ambiguity remains concerning the respective rights, responsibilities and prerogatives of the Council and district officer. If conflict is to be avoided in future these must be more clearly defined.
The status of chiefs today
One result of these changes is that Rotuma has become a much more political community than it was in the past. During the colonial period people rarely discussed political issues, and were reluctant to express viewpoints concerning the directions future change should take. Dissatisfaction with the district officer's policies was usually expressed by grumbling and passive resistance. Now many people not only hold a definite point of view but are prepared to speak openly, to debate issues, and to criticise those in authority directly.
Rotumans today are also far more committed to progress and development than they were in the past; they evaluate leaders more by what they accomplish (or do not accomplish) than by what they say or how they act. People want well-constructed modern houses, refrigerators, modern appliances, cars and stereos. But while economic development has progressed slowly on the island, Rotumans elsewhere have continued to make their mark, not only in Fiji but abroad as well. Many Rotumans have risen to positions of responsibility and leadership in government, the military, and private industry. They have not only demonstrated an ability to lead, but have accumulated political power far beyond that in the hands of the chiefs. They also enjoy a standard of living to which people on the home island only aspire.
These circumstances have created a dilemma for chiefs on Rotuma. They are expected to formulate policy for development, to take fiscal responsibility for managing the budget, and to administer programs. But they are neither well educated nor trained for these tasks of modern government, and from the people's standpoint, continually botch the job. Furthermore, they are finding that the real power to do good for Rotuma lies not with them, but with Rotumans who have powerful positions in Fiji and abroad. In order to get things done they have to maneuver through bureaucratic channels they do not understand, and they become irritated with Rotumans in Fiji who try to educate them about the realities of modern government and industry. They express resentment when their kinsmen in Fiji do not bow to their authority and respond to their beck-and-call. This has sometimes led to rather strained relationships between the chiefs and Rotuman leaders in Fiji.
No one currently presumes that chiefs from the home island can exercise authority over Rotumans in Fiji or elsewhere. In fact, the chiefs formally abrogated that possibility in 1946, when they refused a request from Rotumans in Fiji that the Council of Chiefs appoint someone to be their 'headman.' The Fiji Rotumans at that time expressed the view that someone appointed by the Council of Chiefs would be more respected, but the chiefs opted out and suggested that the people choose their own headman.  As the financial and political power of Rotumans in Fiji has grown, they have exercised increasing influence on their home island, a circumstance that arouses apprehension among the chiefs and some others who express concern that control of the island's destiny is passing, or has passed, outside the local community. The chiefs seem to feel, with good reason, that they are losing sovereignty over Rotuma itself.
The chiefs' moral authority has been undermined in the eyes of many as a result of their handling of money. They are often criticised for using the limited monies available to the Council for doubtful purposes such as trips to Fiji. Some chiefs have been accused of skimming funds from development projects in their districts, from ships' landing fees, and from cooperative and church accounts. As a result people are often reluctant to support local projects, and may refuse to give either money or labor to communal efforts managed by a chief.
That the chiefs should be tempted to use public monies for their own benefit should not be surprising. In the past, chiefs were expected to live in a manner befitting their status, and to represent the dignity of the their district. A chief's house was used for receiving guests to the district, and was expected to be imposing. But now chiefs see people around them without titles, and from ordinary families, building expensive, elaborate residences, buying cars and videos, and enjoying a standard of living they cannot match. Furthermore, this comes at a time when people are less willing to provide support in labor and materials to maintain chiefly prerogatives. It is therefore understandable that the chiefs feeling insecure, for feeling that they are entitled to use the community's monies for enhancing their status.
Leadership on Rotuma today is therefore in a state of crisis. The chiefs are at a great disadvantage. As members of the Rotuman Council they are supposed to formulate policies and guide the development of the island, but they are not well-equipped to do so. They lack the education and experience required to manage an expanding economy and to make informed choices concerning development opportunities. They are uncomfortable with bureaucratic procedures and with bureaucrats who control resources. Internally, they are perceived by most Rotumans as self-interested and ineffective, lacking in moral authority.
But for the people on Rotuma, and for many Rotumans abroad, chieftainship remains an institution of vital importance. Indeed, it is considered to be at the heart of Rotuman custom, and hence Rotuman identity. Chiefly titles provide continuity with the past; embedded in them are family and district histories. Without chiefs ceremonies of all kinds-births, marriages, welcomings, village and district fetes-would lose their significance, for it is the presence of chiefs that lends dignity and cultural meaning to such occasions. Virtually all formal ritual at ceremonies involves chiefs; without them nearly everything that is distinctly Rotuman would disappear.
So despite the heavily criticised behaviour of the present chiefs, the idea of chieftainship is something few Rotumans are prepared to abandon. While they freely complain about chiefs, singly and collectively, most people remain committed to the institution as a whole. A common suggestion is that chiefs be removed from positions of public administration, and that the Council of Rotuma be reconstituted to exclude chiefs. That way, it is argued, the chiefs could concentrate on Rotuman custom and would be freed from involvement in secular politics and economic management. Such matters should be in the hands of Rotumans who have been educated and trained to deal with them, the argument goes. When chiefs take on such responsibilities, especially if they are inept, their moral authority is undermined, subverting the dignity of Rotuman custom.
For some Rotumans chieftainship is central to their sense of identity. They see themselves as special because of their chiefly affiliations, either as descendants of prior chiefs, as close kinsmen of contemporary titleholders or as titleholders themselves. They see titles as embodying the Rotuman notion of ideal personhood, and feel themselves to be elevated as a result. Thus it appears that chieftainship in modern Rotuma will continue to play a role for some time to come. But whether it will be transformed into an institution of secular power, or will be perpetuated as an institution of solely ritual significance, remains to be determined.
Unquestionably, the most important institutions on Rotuma aside from government are the churches. The Catholic and Methodist Churches, and more recently Seventh Day Adventists and Jehovah's Witnesses, organise many important activities and provide the basis for group (though not necessarily community) cohesion. A series of church-related events punctuates life on Rotuma, ranging from island-wide celebrations like the annual Methodist Conference to daily prayer meetings.
In the past the rift between Catholics and Methodists was profound, stemming from wars fought in the 1870s. Neither side would participate in the events of other, and those who intermarried or converted risked permanent rupture with their previous family members, kinsmen and community mates. Fortunately, in the 1960s and 70s a new generation of ministers and priests took positive steps to encourage interfaith cooperation. Nowadays Catholics freely attend Methodist Conference fundraisers (and donate money as well). Perhaps even more significant was the fact that Pepjei, a predominantly Catholic district, hosted the Conference in 1989. Likewise, Methodists now donate labor, money and goods to events sponsored by the Catholic Church.
The churches are in a position to play an important role in the development of Rotuma, economically as well as socially and morally. Whether they have lived up to their promise is debatable, however. We have heard many complaints about lack of leadership within the church hierarchies, and splits within congregations over various issues seem to have increased in recent years.
A variety of other organisations organise activities on a sporadic basis, ranging from sports clubs to women's interest groups. Some of these have played important roles in mobilizing talent for various durations, and have contributed socially and/or economically to the communities they serve.
Business organizations like the Rotuma Cooperative Association and Raho have played a crucial role in recent history, but describing and assessing their significance is beyond the scope of this essay. 
To give perspective to constitutional issues it is important to understand Rotuman attitudes toward authority. In principle, most Rotumans are committed to Polynesian notions of social hierarchy-that certain people, by virtue of lineage or outstanding accomplishment, are fit to be leaders. However, Rotumans temper notions of aristocracy with demands that leaders represent and serve their people's rather than their own interests. In this respect, a strong strain of egalitarianism runs through the Rotuman cultural vein.
Rotuman legends are quite clear with regard to the basic constitution of authority. It requires potency, or mana, which traditionally derived from the gods and ancestral spirits. To be effective, and legitimate, potency must be tempered by domestication. Collectively the stories reveal the pitfalls of either extreme. Those chiefs whose ambitions are unconstrained by concern for the populace bring hardship and misfortune; their vitality is misdirected. But selflessness without potency is also a formula for disaster. A proper chief is one whose mana is potent but sufficiently under control to serve the welfare of his people. He is entitled to first fruits and a reasonable portion of the produce of the land, but he cannot demand too much. A chief who takes too much is the conceptual equivalent of a cannibal-he ravages his people by consuming their crops and labor. In the context of Rotuman cultural logic, such excesses justify rebellion. 
These notions, along with the fact that, given the bilineal nature of descent, virtually anyone can claim chiefly lineage, means that authority is very difficult to establish and sustain in contemporary circumstances. Furthermore, Rotumans value autonomy to a degree that is unusual in Polynesian societies. Even by western standards Rotuman children are granted an astonishing degree of autonomy. Parents rarely compel children to do things they do not want to do. The overriding principle is that it is undesirable to force people, children included, to do things against their will. One expression of this emphasis on autonomy is the frequently heard phrase, 'Puer se 'äea/irisa' 'It's up to you/them,' when people are asked about expected behaviour, contributions, etc.
As one result of this socialisation pattern, Rotumans are socially sensitive and ready to react defensively when their sense of autonomy is threatened. In defense of their autonomy people are prepared to stand up for what they perceive to be their rights, against their own chiefs if necessary.  It should be noted that although Rotumans may 'talk a good fight' on occasion-verbal skills are encouraged and rewarded-talk rarely translates into violent action.
The principle of autonomy operates throughout the social structure. Not only do individuals exercise autonomy within their households and communities, but villages are autonomous vis-a-vis one another, and districts are essentially independent political units. The emphasis on autonomy has often been expressed in Rotuman history in the form of self-reliance. From the beginning of colonial rule Rotumans have sought to control their own destiny: by buying vessels and attempting to operate shipping themselves (always unsuccessfully), by forming the Rotuma Co-operative Association and forcing the commercial firms off the island, even by refusing relief supplies following Hurricane Bebe. The rhetoric of the late Senator Wilson Inia was saturated with references to the importance of self-reliance.
We might add that concerns for autonomy and self-reliance apply as much to women as to men. In general, women enjoy high status on Rotuma and while they do not serve as chiefs (although the idea is not inconceivable), they freely express opinions and exert considerable influence. By custom women participate in meetings at every level and they can be expected to play an active role in any processes that may affect Rotuma's future political status.
The issue of Rotuma's status vis-a-vis Fiji has divided the Rotuman community in recent years. Shortly after the first coup in Fiji, on 14 May 1987, the Rotuma Council held an emergency session to discuss the coup's implications for Rotuma. The members of the Council resolved to pledge their support to the new government and remain part of Fiji. In response, Henry Gibson, who claims the title of sau, wrote to the Council expressing his fears that the position of the Rotuman people would deteriorate under the new arrangement. Subsequently Gibson addressed the Council and said he would not follow its decision for Rotuma to remain with Fiji.
The importance of Gibson's opposition to the Council ruling lies in the fact that he has an extremely vocal following in Rotuma and among Rotumans in Australia and New Zealand. Although he is only part-Rotuman, and lives in New Zealand, his claims to genealogical connections with previous Rotuman sau are accepted by many of his kinsmen, who recognise the title (Gagaj Sau Lagfatmaro) that he claims for himself. A karate master by training, Gibson also captured the imagination of some of the other Rotumans who are dissatisfied with the Rotuma Council. His pronouncements thus served to galvanise discontent.
Despite Gibson's objections, in July 1987 the Rotuma Council sent representatives to attend a meeting of the Great Council of Chiefs in Fiji to express Rotuma's desire to remain part of Fiji. Following the return of this delegation, meetings were held in each of Rotuma's seven districts to ascertain the views of the Rotuman people. According to the deposition of the D.O. at the time, Viki Epeli, 'It was the overwhelming view of the majority of the Rotumans who attended these meetings, that Rotuma should remain part of Fiji even if Fiji were to become a Republic.' 
Following the second coup in Fiji, on 25 September 1987, and the declaration of the Republic by Rabuka, the Rotuma Council again met and resolved that Rotuma would remain part of Fiji. A copy of the resolution was sent to the President of the newly formed Republic with a copy to the Prime Minister. The following month, in New Zealand, Henry Gibson declared Rotuma independent of Fiji and wrote to Queen Elizabeth asking for recognition. His argument was that the Rotuman chiefs had originally ceded the island to Great Britain, not to Fiji, and that Council members serve only by sufferance of the Queen. By renouncing their affiliation with the Commonwealth the Council lost their legitimacy, because according to Gibson only the Queen could sever the Commonwealth tie.
Acting on the basis of Gibson's pronouncement, his followers held a meeting in April 1988 and selected new headmen for each district.  In May the dissident 'chiefs' were arrested and charged with sedition, and after a hearing at a special sitting of the Magistrate's Court on Rotuma the case was sent to the High Court of Fiji. This action was taken in response to the argument of Tevita Fa, lawyer for the defendants, that given the nature of the dispute, Fiji's right to try his clients was a matter of contention. The following month Chief Justice Sir Timoci Tuivaga issued a judgment ruling that 'for legal and other purposes Rotuma continues to be a part of the independent sovereign State of Fiji.' 
We found opinions among Rotumans to be quite diverse regarding the issues involved. Among Henry Gibson's staunch followers the view is openly expressed that Rotuma will suffer in the long run if it remains politically integrated with Fiji. They are doubtful about Fiji's commitment to Rotuma's development and believe Rotuma would do better on its own. In response to questions concerning an independent Rotuma's economic viability, the program they favored included four major sources of income: (1) the leasing of fishing rights in the 200-mile zone around Rotuma; (2) foreign aid from developed countries; (3) tourism; and (4) the marketing of agricultural produce, and particularly Rotuman oranges.
Those opposed to Gibson regard him as a political menace and a threat to Rotuma's security. They resent the disruption he has created within the Rotuman community and are apprehensive about the effects the secessionist movement he initiated will have on Rotuma's relations with Fiji. They see Gibson's claim to sauship as self-glorifying and without legitimacy.
Many Rotumans we have talked with are ambivalent. They can see some merit in considering the possibility of an independent Rotuma, and regard Gibson as a basically good man who has gone about things in the wrong way. They feel the issue of Rotuma's status vis-a-vis Fiji is legitimate grounds for discussion and debate, and have not reached firm conclusions. We were somewhat surprised to discover that a good many Rotumans in Fiji hold such a viewpoint, since Rotuma's independence might jeopardise the possibilities they now have for moving readily back and forth between Fiji and Rotuma. Some, however, suggest that a status for Rotuma short of full independence-some version of free association-might resolve this problem.
A diversity of opinion currently exists within the Rotuman community concerning Rotuma's relationship with Fiji. Those favouring a continuing allegiance to the Republic can point to many benefits Rotumans have received from national union. Foremost has been the opportunity to freely move back and forth, giving individuals ready access to education, training and job opportunities. Rotumans have done very well in Fiji, disproportionately attaining positions at the upper ends of the occupational ladders. They have made solid contributions to Fiji as a whole through their work in government and business. Their successes have also made it possible for them to provide people back home with remittances and supplies that helped raise the standard of living there. Rotumans in influential positions in Fiji have also done much to channel services, grants and facilities to the island.
The fact that more than two-thirds of the current population of Rotumans now reside in Fiji, and that a substantial proportion of this population was born and reared there, suggests that Fiji may be replacing Rotuma as the 'homebase' of the Rotuman population. The implications of this population shift for the social, economic, political and cultural life of the Rotuman people is not yet clear and needs to be discussed.
On the negative side, Rotumans have some valid complaints about governmental neglect. Perhaps most important has been the erratic nature of shipping to the island, resulting in recurrent periods of deprivation and hardship. The lack of reliable shipping not only results in basic supplies being cut off from time to time, it also makes it impossible for Rotumans to market their agricultural produce on a sustained basis. This has been a long-standing problem and is as acute today as in the past. It is a problem Rotumans have complained about over and over again without satisfactory redress.  The psychological impact on the island has been profound, and has contributed to a sense of alienation from Fiji. The substantially higher cost of goods on Rotuma adds to dissatisfaction and raises questions, rightly or wrongly, about exploitation and profiteering.
For those advocating independence, the prospect of Rotuma being re-opened as a port of entry is seen as a resolution to these problems. Rotuma was closed as a port of entry shortly after cession, forcing all commerce to be funneled through Fiji. If Rotuma were to be re-opened as a port of entry, the argument goes, Rotuma's economic woes would be alleviated by allowing international vessels to bring supplies from New Zealand or Australia and take produce directly to places like Samoa, Tuvalu and Tokelau. Such an arrangement could occur in union with Fiji, or as a result of Rotuma's independence.
Also fueling sentiment for independence is the frustration many Rotumans on the island experience with the bureaucratic red tape they encounter with the Fiji Government. Visiting officials (some of whom are Rotumans) come to the island and make promises that fail to materialise, projects get lost in government offices, requests and protests are met with form letters. As a result many Rotumans feel disillusioned and powerless. They see independence as a way of assuming more direct control of their destiny.
Even those Rotumans who favour remaining part of Fiji see a need for improvements in these areas. The building of a wharf in 1973, the inauguration of the airstrip in 1981, and the recent installation of satellite communication has helped to relieve the sense of isolation that prevailed throughout the colonial period. But if Rotuma is to be truly integrated into the nation of Fiji more has to be done. Improvements to the airport allowing larger planes with cheaper fares, more reliable shipping, improvements to the island's infrastructure including roads and facilities, and more functional bureaucratic channels will all be required. The proposal for a special ministry for Rotuman Affairs can be seen as a plea by Rotumans for creating conditions that will alleviate frustrations that have generated doubts about Rotuma's affiliation with Fiji. It is a proposal that should be given serious consideration by the Review Commission.
Politically, Rotumans express discontent on two levels. At the national level they have felt slighted by what they consider under-representation in the Legislature. The fact that they were given no seats in the lower house of Parliament in the original constitution was distressing, and the current demand for two seats, one representing the constituency on the island, the other Rotumans in Fiji, is generally seen as a necessary correction. From a socio-political as well as a demographic point of view this seems justified since the two constituencies face somewhat different circumstances; their interests only partially overlap.
At the local level, on Rotuma, there is a good deal of discontent with the current arrangement. Complaints about lack of communication between the Rotuma Council and the people they serve, about the Council's ineffectiveness in formulating and carrying out development policies, and about the alleged self-serving behaviour of Council members, are widespread. Many of the people we have talked to propose reconstituting the Council so that it is more representative and its members more accountable. Whatever support the independence movement has on Rotuma, and particularly that provided by the Mölmahao group  probably derives as much or more from discontent with the Council as from dissatisfaction with Fiji. The attempt by the dissidents to replace the chiefs with 'ministers' from each district is testimony to their anger and sense of alienation. Quite apart from the independence issue it would seem to be worthwhile for Rotumans to consider ways to reconstitute a governing body that would enjoy popular support as well as authoritative legitimacy.
To date the voices that have received the most attention have often been the shrillest and the most polarised. Both sides make unrealistic claims regarding the degree of support they enjoy among the Rotuman people. We believe that before any final decisions are made Rotumans need to participate in a full discussion of the issues and to consider a range of practical, workable solutions. Hopefully saner, more reasonable, voices will prevail. At some point, after the options have been thoroughly debated and narrowed down to those that are practicable, a referendum might be held so that an accurate assessment can be made concerning Rotuman opinion.
The situation of Rotumans in the nation of Fiji is unique insofar as they are there as a result of an accident of colonial history. Rotumans nevertheless contributed willingly to Fiji's development during the colonial and post-colonial periods. However, the termination of colonial rule, the coups, and the subsequent withdrawal of Fiji from the British Commonwealth, raise legitimate questions concerning the legacy of union. Rotumans have good reasons for wanting to preserve their unique cultural heritage. Many are apprehensive about being dominated, about being nothing more than a neglected minority in a multi-cultural state. If Fiji wants to retain and strengthen Rotuman loyalty it will have to address these concerns and work together with Rotumans to find satisfactory solutions. For their part, Rotumans need to formulate for themselves more clearly than has yet been done, through discussion and possibly referenda, just what it is they want, so that negotiations can proceed on a firm footing.
[ 1] See Gardiner (1898:497); also Eason (1951:88, 122-123) [back to text]
[ 2] This paper was prepared from published materials, unpublished archival sources and ongoing anthropological field work among Rotumans by Howard and Rensel. James McNally and Sela Panapasa analyzed data from the 1986 Fiji census and contributed to their analysis. Portions of the text were drawn from previously published papers, including: A. Howard, Land Tenure and Social Change in Rotuma, Journal of the Polynesian Society 73:26-52 (l964); The Rotuman District Chief: A Study in Changing Patterns of Authority, Journal of Pacific History l:63-78 (1966); History, Myth and Polynesian Chieftainship: The Case of Rotuman Kings, in A. Hooper and J. Huntsman (eds.), Transformations of Polynesian Culture, Auckland: Polynesian Society (1985); The Resurgence of Rivalry: Politics in Post-Colonial Rotuma, Dialectical Anthropology 14:145-158 (1989); A. Howard and J. Rensel, Rotuma in the 1990s: From Hinterland to Neighbourhood, Journal of the Polynesian Society 103:227-254 (1994); and J. Rensel, The Fiji Connection: Migrant Involvement in the Economy of Rotuma, Pacific Viewpoint 34(2):215-240 (1993). Additional material was extracted from a chapter by A. Howard and J. Rensel entitled 'Ritual Status and Power Politics in Modern Rotuma,' from a book on Chiefs in Modern Oceania, edited by G. White and L. Lindstrom, to be published by Stanford University Press. [back to text]
[ 3] This estimate would be increased significantly if the offspring of marriages to non-Rotumans were included. [back to text]
[ 4] The Index of Difference is a commonly employed demographic measure which is defined as being one half of the absolute difference of two distributions. It provides a value free measure of how much one of the two distributions would have to change in order for both distributions to be identical. For example, an Index of Difference of twenty-five percent between two age distributions indicates that one of the two would have to change by twenty-five percent to identically match the other. An Index of Difference of zero percent indicates that the two groups are identical while an Index of Difference of 100 percent indicates that the two groups are completely different from each other. [back to text]
[ 5] If only off-island Rotumans are considered, the percentage of wage employment goes up considerably, with 70.3 percent of males and 34.4 percent of females so engaged. [back to text]
[ 6] We make the assumption that Rotuman fertility will decline across the projection period, but that this reduction in population growth will be moderated by a continued decline in mortality as well. These assumptions result in a series of projections where the Rotuman population continues to experience moderate growth in line with observed intercensal percentage change. This is considerably below their potential rate of increase, but we have no reason to anticipate that Rotuman fertility will increase significantly during the projection period. [back to text]
[ 7] Some commentary on the claims of Henry
Gibson to the sauship and the title Lagfatmaro
may be in order here. According to published accounts and unpublished
reports by early visitors to Rotuma, family groups (or 'clans' as they
are called in Rotuman-English vernacular) do not have the right to
bestow the office of sau on individuals.
This right has either been attributed to the dominant district chief
(fakpure) or to a council of district
chiefs. Furthermore, the position of sau was
not attached to a single individual or to any one lineage, but rather
was an office occupied by a series of men from different districts
for limited periods of time for the purpose of conducting specific
[ 8] Today it much more difficult to determine how many titleholders there are or what their ratio is to adult males since the major portion of Rotumans reside off island in Fiji. Many titleholders leave the island, either for extended periods or permanently. Others make regular sojourns abroad, resulting in significant variations from month to month. [back to text]
[ 9] Although in recorded history all titleholders have been male, it is not inconceivable for a woman to hold a title and assume a chiefly role. A few years back a highly regarded retired female schoolteacher was considered seriously for the position of district chief. Presumably she would have taken a title if installed. [back to text]
[10 ] See Howard 1961 for an account of this process. [back to text]
[ 11] Dispatch from C. Mitchell to Governor of Fiji, 12 October 1881, Outward Letters of Rotuma District Office. [back to text]
[12 ] For examples see Minutes of the Rotuma Council of Chiefs, 1 September 1910, and dispatch from A.E. Cornish to Colonial Secretary, 30 January 1939, Outward Letters of Rotuma District Office. [back to text]
[ 13] Following a reorganization of administration in the Colony of Fiji in the 1930s the appointed official in charge of Rotuma was known as the District Officer. [back to text]
[ 14] Previously each district sent a representative, but the latter was chosen by the chief and acted more or less as his assistant. [back to text]
[ 15] As pointed out earlier, the chiefs who ceded Rotuma to Great Britain expected to be granted the same prerogatives as Fijian chiefs only to be thwarted by the colonial administration. The post-independence government, however, is based in Fijian chieftainship and appears to support chiefly privilege in Rotuma based on the Fijian model. [back to text]
[ 16] Minutes of the Rotuma Council, 10 October 1946. [back to text]
[17 ] For recent work on Rotuma's economic history see Rensel 1993, 1994. [back to text]
[ 18] For an analysis of authority in Rotuman legends see Howard 1986. [back to text]
[ 19] This makes chieftainship in Rotuma a somewhat different phenomenon than in societies such as Fiji, where chiefs have real power over their subjects. In Rotuma chiefs are honored on special occasions, when they are 'in role,' but most of the time they are 'out of role' and are treated respectfully but without reverence. In turn, chiefs, like parents, respect their subjects' autonomy. They are generally unable to force people to do things against their will (for more information on the nature of chieftainship in Rotuma, see Howard 1963, 1966). [back to text]
[ 20] Fiji Times, 10 June 1988, p. 13. [back to text]
 There is some ambiguity about the titles these new 'officials' were supposed to hold. In the press, and among most Rotumans I spoke with, the English term 'chiefs' was used in reference to their claims. Some of Gibson's followers, however, insisted that these new officials were not meant to replace the present gagaj 'es itu'u, and referred to them as 'ministers' (as in 'governmental ministers'). [back to text]
[ 22] Fiji Times, 10 June 1988, p. 41. He concluded, therefore, that Fiji had jurisdiction over the case. [back to text]
[ 23] For instance, see extracts from the Senate speeches of Wilson Inia on the topic of inadequate shipping to Rotuma, in Howard 1994:132-141. [back to text]
 Mölmahao is the name of the fuag ri 'house foundation' with which Henry Gibson's followers associate the title Lagfatmaro. [back to text]
Allardyce, W. L. (1885-6). Rotooma and the Rotoomans. Proceedings of the Queensland Branch of the Geographical Society of Australasia 1:130-144.
Allen, W. (1895). Rotuma. Report of Australasian Association for Advancement of Science. VI: 569-579.
Bennett, G. (1831). A Recent Visit to Several of the Polynesian Islands. United Services Journal 33:198-202, 473-482.
Bryant, J. (1990). Rotuman Migration and Development: A Response to Uneven Development. Migration and Development in the South Pacific. Canberra: National Centre for Development Studies.
Churchward, C. M. (1938-9). Rotuman Legends. Oceania 9(1,2,3,4): 109-126; 217-231; 326-339; 462-473.
Churchward, C. M. (1940). Rotuman Grammar and Dictionary. Sydney: Australasian Medical Publishing Co.
Dillon, P. (1829). Narrative...of a Voyage in the South Seas. London: Hurst Chance.
Eason, W. J. E. (1951). A Short History of Rotuma. Suva: Government Printing Department.
Evans, H. S. (1951). Notes on Rotuma. Fiji National Archives, Suva.
Gardiner, J. S. (1898). Natives of Rotuma. Journal of the Royal Anthropological Institute 27:396-435, 457-524.
Histoire de la Station Notre Dame de Victoires, Sumi, Rotuma 1868-1881, Notebook II, 2, Catholic Diocesan Office, Suva, Fiji. Pacific Manuscripts Bureau reel 159.
Hocart, A.M. (1913). Field Notes from Rotuma. Manuscript in Turnbull Library, Wellington, New Zealand.
Howard, A. (1961). Rotuma as a Hinterland Community. Journal of the Polynesian Society 70:272-299.
Howard, A. (1963). Land, Activity Systems and Decision-Making Models in Rotuma.Ethnology 2:407-440.
Howard, A. (l964). Land Tenure and Social Change in Rotuma.Journal of the Polynesian Society 73:26-52.
Howard, A. (1966). The Rotuman District Chief: A Study in Changing Patterns of Authority.Journal of Pacific History. l:63-78.
Howard, A. (1985). History, Myth and Polynesian Chieftainship: The Case of Rotuman Kings. In Transformations of Polynesian Culture, edited by A. Hooper and J. Huntsman. Auckland: Polynesian Society.
Howard, A. (1986). Cannibal Chiefs and the Charter for Rebellion in Rotuman Myth.Pacific Studies 10:1-27.
Howard, A. (1989). The Resurgence of Rivalry: Politics in Post-Colonial Rotuma.Dialectical Anthropology 14:145-158.
Howard, A. (1994). Hef Ran Ta (The Morning Star): A Biography of Wilson Inia. Suva:Institute of Pacific Studies, University of the South Pacific.
Howard. A. and J. Rensel (1994). Rotuma in the 1990s: From Hinterland to Neighbourhood.Journal of the Polynesian Society 103:227-254.
Howard. A. and J. Rensel (1997). Ritual Status and Power Politics in Modern Rotuma. In Chiefs in Modern Oceania, edited by Geoffrey White and Lamont Lindstrom. Stanford: Stanford University Press.
Macgregor, G. (1932). Field Notes on Rotuma. Manuscript in the Bishop Museum, Honolulu.
Minutes of the Rotuma Council. Fiji National Archives, Suva.
Outward Letters, Rotuma District Office. Fiji National Archives, Suva.
Rensel, J. (1991). Housing and Social Relationships on Rotuma. In Rotuma: Hanue Pumua, Fatiaki et al., Suva: Institute for Pacific Studies, University of the South Pacific.
Rensel, J. (1993). The Fiji Connection: Migrant Involvement in the Economy of Rotuma.Pacific Viewpoint 34(2):215-240 (1993).
Rensel, J. (1994). For Love or Money? Interhousehold Exchange and the Economy of Rotuma. Ph.D. Thesis, University of Hawaii Department of Anthropology.
Sykes, J. W. (1948). Confidential Report on Rotuma. Fiji National Archives, Suva.
Wood, C. F. (1875). A Yachting Cruise in the South Seas. London: Henry King.