Home Page
Howard-Rensel Papers

21. Rotuma in the 1990s: From Hinterland to Neighbourhood

Alan Howard and Jan Rensel

[Published in Journal of the Polynesian Society 103:227-254, 1994]

In 1961 Howard published a paper entitled "Rotuma as a Hinterland Community", describing the increasing flow of population, goods and services, and information and ideas between the island of Rotuma and Fiji. At the time, Rotuma was a part of the British Crown Colony of Fiji. When Fiji gained independence in 1970, the Rotuman people opted to remain with Fiji and, not without controversy, affirmed their decision following the military coups of 1987. This paper aims to update and reassess the findings reported in Howard's 1961 paper, which was based on two years of field work on Rotuma and among Rotuman enclaves in Fiji from mid-1959 through mid-1961.

Data for the 1961 paper were obtained from Government records, including census reports and ministry files, from an island-wide household survey conducted on Rotuma in 1960 (including residential histories from most adults), and from participant observation. More recent data have been collected during five annual field trips between 1987-91 by both authors. In 1989 we conducted another survey, of 85% of the households on the island, obtaining much information comparable to that obtained by Howard in 1960, as well as additional data reflecting on household economies. Also in 1989, as part of her dissertation research, Rensel conducted an intensive study of daily household interactions in one village (17 households) over a 13-week period. Findings from these studies are supplemented by census and ministry data, informal interviews, and in this day of fax machines, by a continual stream of communications from friends on the scene. [1]

Map of Rotuma

In contrast to some other Pacific island populations, Rotumans are entitled to unrestricted movement between their home island and main overseas destination, Fiji. There are no international boundaries to cross, no immigration rules hindering travel. Population movement has only been constrained by available modes of transportation and the costs of using them. As transportation has been made more readily accessible, and money become more plentiful, the flow has intensified. As our title suggests, the data show that, whereas in 1960 Rotuma was a hinterland in relation to Fiji's urban centers (i.e., it was a relatively remote, clearly distinguishable community, with special bonds to Suva), by the late 1980s it had become more like a neighbourhood--one of many neighbourhoods in Fiji--among which Rotumans freely circulate. This shift implicates a number of changes that we explore in this paper, including alterations in access to resources, productivity of island agriculture, standard of living, housing and land use, social relationships, values, political power and cultural identity. Before addressing these issues, we present a section describing significant developments during the past 30 years, then provide an update to data concerning the flows of population, goods and information.

Developments on Rotuma: 1961-1991

The opening of an airport on Rotuma in 1981 created new opportunities for intercourse with the outside world. Initially two flights a week were scheduled, but the airfare (F$95) proved prohibitive for most Rotumans, and low passenger loads resulted in a reduction to weekly flights in 1988. Ships have, therefore, remained the primary means of traveling to and from Rotuma, since fares are considerably cheaper and space is more readily available. A wharf was completed at Oinafa (see map) in 1973, easing the task of loading and unloading people and supplies. Previously ships had to anchor beyond the fringing reef, while launches transported people and goods back and forth to land. Two ships recently bought by Rotuman companies have helped to place shipping on a more regular basis.

Far more of the island is electrified now. In 1960 only the Government station and Catholic churches at Sumi and Upu had generators. Now several villages have generators that provide electricity for a few hours per day, primarily for lighting, and more than 30 individual households have their own generators. In addition, several solar power units are in use. Soon further changes may be in store: the European Community is committed to providing a generator for island-wide electrification within the next few years.

Another physical change during the interim was in housing. In 1972, Hurricane Bebe destroyed most Rotuman-style thatched houses. For the most part, they were replaced by concrete houses with corrugated iron roofs. The New Zealand army came on a relief mission after the devastation and, in about three weeks, engineered the building of more than 300 houses. A 1966 survey of house types categorised 240 (50.7%) as ri hafu (cement or stone), 60 (12.7%) as ri ai (wood), 84 (17.8%) as ri pota (corrugated iron), and 89 (18.8%) as ri fakrotuam (Rotuman style; thatched). [2] In a subsequent count, during 1981, 82.8% of the houses were categorised as ri hafu and the count for ri fakrotuam was zero [3] (see Rensel 1991 for a fuller discussion of housing changes on Rotuma).

The discovery of underground fresh water supplies in 1976 has also had important consequences for housing. Piped water is now available, so people no longer have to depend on rainwater storage tanks. Many homes now have indoor kitchens with sinks, whereas previously kitchens were separated from the main house. Also, thanks to additional assistance from New Zealand, water-seal toilets are commonplace, either inside or just outside the main building. These have replaced pit latrines in the near bush and outhouses on piers over the tidal flats.

Although the main road around the island remains rough and unpaved, bush roads have replaced footpaths into the interior. This permits people easier access to their remote gardens, since motor vehicles can now be used. In 1960 the only motor vehicles on the island were two trucks at the Government station, two others belonging to business firms, two cars operated as taxis by Indian shopkeepers, and a handful of motorbikes. In 1989, Government agencies and the Rotuma Co-operative Association had several trucks and cars at their disposal, and the Rotuma Council operated two buses. In addition we counted 21 privately owned cars or trucks and 150 motorbikes in working condition, or nearly one for every three households (there were perhaps 100 more in non-working condition). Thus mobility on the island has greatly increased. Whereas in 1960 a trip to the other side of the island was a major excursion, today it is routine.

The most significant economic change since 1960 was the closing of the two firms, Morris Hedstrom and Burns Philp, that dominated trade during the colonial era. From the late 1960s through the 1980s the Rotuma Co-operative Association (RCA) enjoyed a virtual monopoly over the island's business affairs. In recent years, however, a rival co-operative (Raho) has strongly challenged RCA's dominance, providing a renewed, invigorating competition for the first time since the firms folded.

Politically the most important change took place following Fiji's independence in 1970. During the colonial period, the Government-appointed District Officer (D.O.) was in full control of the island's administration--he was boss. The Rotuma Council, composed of paramount chiefs and elected representatives from each of the island's seven districts, the head medical officer, and headmaster of the high school, acted only in an advisory capacity. The D.O.'s authority came from the Governor, whom he represented, and ultimately from the British Crown. With independence, the basis of his authority became ambiguous.

Following a dispute between the Council and D.O. shortly after independence, their relative positions were reversed (see Howard 1989). The Council was given primary policy-making powers and the D.O. was made its advisor. One result of this change 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 D.O.'s policies was usually communicated by grumbling and passive resistance. Now many people seem to have a definite point of view and are prepared to speak out openly, to debate issues, and to criticise directly those in authority. Many other changes have taken place in the past thirty years, but to comprehend their significance it is necessary to examine the flows that are the main focus of this paper.

The Flow of Population

During 1960 it was clear that the flow of people from Rotuma to Fiji was accelerating and would have a major impact on the future of the island. At the time there were approximately 3000 persons on the island and about half that many Rotumans in Fiji. On Rotuma one could sense the pressure on land, manifest in an increasing frequency of disputes and much talk about land issues. Fuelling the problem was the fact that, although death rates on Rotuma had dropped dramatically in the 1950s, birth rates remained high, so the Rotuman population was increasing at a rapid rate. One could also sense, particularly among youths, the pull of Fiji's urban centers as sources of employment, education and a more modern lifestyle.

An examination of census data collected since then shows vividly that, although the Rotuman population has continued to grow, outmigration from Rotuma to Fiji has actually reduced the population of Rotuma. Using the 1956 census as a baseline, when the number of Rotumans on the island was reported as 2,993, we find an increase to 3,235 in 1966, followed by a sharp decrease to 2,707 in 1976 and 2,588 in 1986. During the same 30-year period, the number of Rotumans in Fiji (not including Rotuma) more than quadrupled, from 1,429 to 6,098 (see Table 1). [4]

TABLE 1 Rotumans on Rotuma and in Fiji, 1956-1986




























Data obtained from Fiji Census Reports, Government Press, Suva.

The effect of this emigration on Rotuma's age structure can be seen in Figure 1, 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 emigration 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 still more dramatic. 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 emigration increasingly has involved young couples who either migrated with their children, or left Rotuma single, married in Fiji and had their children there.

/Figure 1 about here/

The effects of this outmigration on Rotuma's social and economic life can be better appreciated in the light of changes in household size and structure. According to census reports the number of persons per household decreased from 7.1 to 5.9 persons on Rotuma between 1966 and 1976 (see Table 2). This corresponds to the period of maximum emigration, when the population of Rotumans on the island dropped by 16 percent. The decline is reflected in our survey data as well. In 1960 Howard recorded 417 households on Rotuma with a total of 2892 persons, or 6.9 persons per household. Our 1989 survey revealed an average of 5.3 persons per household. As can be seen in Figure 2, this drop in average household size can be accounted for mainly by a dramatic increase in small households, those with three or fewer persons, and, to a lesser extent by a decrease in large households, those with seven or more persons. In part, this may reflect the loss of individuals from existing households through emigration.

TABLE 2: Persons per Household on Rotuma, 1956-86
Persons per Household
Rotumans on Rotuma



* Approximate figures, based on total of "Other" households (non-Fijian, non-Indian) and thus including some "Part-Europeans" and "Other Pacific Islanders"

Data obtained from Fiji Census Reports, Governement Press, Suva, Fiji

But that cannot be the whole story, becausethere has been a substantial increase in the number of households as well (417 in 1960, 489 in 1989). To some extent, the increase in small households may represent return migration by individuals who have opted to establish their own households rather than join existing ones. It may also reflect investments by Rotumans abroad in maintaining active links to the island. By building or refurbishing a home and having it occupied by close kinsmen, emigrants insure that they, and/or their immediate family, will have a place to return to in Rotuma. A number of houses on Rotuma are in fact occupied on a caretaking basis for relatives who have sent remittances to have new homes built or old ones improved. In other words, the occupants of many small households may be in the position of protecting the resettlement rights of their close kinsmen abroad. [5]

A comparison of findings from our 1989 and 1960 surveys gives additional insight into changes in household structure (see Table 3). The major change over the years has been a substantial increase in households composed of single persons or married couples (11.2% in 1989 compared to only 3.3% in 1960). Considering the fact that single individuals are not viable production units for subsistence purposes, the data on household structure would appear to support the interpretation offered above--that a number of households are occupied by caretakers for kinsmen abroad.


Household Types in Rotuma, 1960 and 1989









Single (expanded)



Single (extended)



Single (expanded + extended)



Total Single





Married Couple



Married Couple (expanded)



Married Couple (extended)



Married Couplee (expanded + extended)



Total Married Couple








Nuclear (expanded)



Nuclear (extended)



Nuclear (expanded + extended)



Total Nuclear













SubNuclear (expanded)



SubNuclear (extended)



SubNuclear (expanded + extended)



Total SubNuclear





Total mimimal, all types





Total expanded, all types





Total extended, all types





Total expanded + extended, all types





Total, All types



The basic (minimal) type was determined by the presence or absence of primary kinsmen of the household head (spouse, children, parents, siblings). If the household head's spouse and children were present the household was classified as Nuclear; if only his/her spouse as Married Couple; if no spouse, but a parent or sibling, as SubNuclear; if his/her spouse + a sibling and the sibling's spouse as Joint; and if no primary kinsmen as Single. If grandchildren or great-grandchildren were present the household was additionally classified as Extended; if kinsmen other than those meeting the criteria for the basic type or for extension were present, the household was additionally classified as Expanded. Only individuals listed as regular members of the household who were resident at the time of the census interview were considered.

Sources: 1960 survey conducted by A. Howard; 1989 survey conducted by A. Howard and J. Rensel.

The Flow from Rotuma to Fiji

In 1960 Howard concluded that the main stimulus for people leaving Rotuma to settle in Fiji was population pressure, the island being small and limited in its capacity to absorb increasing numbers of people. An absence of commercial industries, requiring dependence on copra and subsistence agriculture, exacerbated the problem. Circumstances have changed. As indicated above, the population of Rotuma reached a peak in 1966 and has been dropping steadily since. By 1986 it had decreased to less than 2600, a decline of 21% from its 1966 peak. Whatever pressure may have been felt in 1960 has, therefore, been somewhat relieved. One indicator of lessened population pressure is a noticeable reduction in land disputes. Whereas in 1960, land disputes dominated social discourse, today they are much more in the background, although they remain a source of social tension and still motivate some people to move away from the island to escape resulting antagonisms (see Howard 1990).

Commercial activity has increased dramatically on the island since 1960, particularly in the volume of consumer goods purchased, but its effect on population flow appears to be minimal. Primary agricultural and livestock enterprises have for the most part not fared well; no secondary industries have yet emerged. Thus, job opportunities for those without overseas training remain severely limited, with the exception of low paying, unskilled Government jobs, positions with the RCA paying nominal wages (averaging $F35 per fortnight in 1989), and a few jobs with the Raho Cooperative. In effect, therefore, nothing much has been done in the past thirty years to hold ambitious people on Rotuma.

As before, the main channel for mobility remains education. In 1960 the Rotuma High School only went up to Form IV; to continue beyond this level students had to leave the island. In the intervening years Form V was added, and in 1991-2 a major fund-raising effort was mounted toward construction of science teaching facilities at the secondary school. But many parents still believe there are educational advantages to schooling in Fiji, so send their children there, often to stay with relatives.

The flow of population to Fiji has been greatly enhanced by short-term visitors who go for special events such as weddings, funerals, births, etc.; for specialised medical treatment; for workshops or training programmes; to participate in sporting events; to visit relatives; or just for fun. Such sojourns have become routine, as shown in Figure 3. Whereas, in 1960, 76% of the adults surveyed had been off-island only once or not at all, in 1989 this was true of only 25.4%. And while only 11.4% of the 1960 group reported three or more trips abroad, 52.6% of the 1989 sample had made at least three trips. Fifty-seven individuals in the 1989 group reported making 10 or more trips, and a few reported as many as 40. Indeed, we were aware of a number of people who travel back and forth several times a year.


We do not believe that the intensified flow of people from Rotuma to Fiji reflects changes in motivation as much as it reflects increased opportunities. As Howard pointed out in his 1961 paper, an eagerness to travel was documented by the earliest visitors to Rotuma (see also Howard 1991). The initiation of air travel to the island, greater shipping availability, and, especially, more money to bear the cost of travel, have all played a role. Perhaps even more important is the growing number of Rotumans residing elsewhere. Our 1989 survey included three questions concerning household membership: (1) who are members of this household currently present, (2) who are members of this household currently away, and (3) who staying here now is not a member of this household. [6] Question one generated a list of 2,199 individuals, question two 1,265 and question three 20 individuals. Of the household members away, 208 were staying in other households on the island, but 1,057, or nearly one-third of those people considered to be household members, were away from the island. Nearly 70% of the households surveyed in 1989 listed at least one absentee household member, and most of those were close relatives, such as children (59%) or siblings (19%) of the household head and his/her spouse.

Increasing numbers of Rotumans have left Fiji and established residence abroad. Substantial enclaves of Rotumans now exist in Australia (in Sydney more than 70 households support a Rotuman pastor) and New Zealand (where a Rotuman pastor compiled a list of 62 families with at least one Rotuman member). Our 1989 survey reflected this trend: Although 81% of household members listed as off-island were reported to be residing in Fiji (860), the remainder were distributed among other countries: Australia (54), New Zealand (22), other Pacific Islands (20), the United States (13), Europe (10), and Canada (6); in addition 10 men were sailors on ships and 9 others were with the Fiji military forces serving as United Nations peacekeepers in the Middle East. [7] In part, the establishment of households outside of Fiji is the result of intermarriage between Rotumans and foreigners who have brought their spouses to their home country. In response to these new opportunities, sojourns in foreign countries are increasing, and it is not uncommon for island residents to spend months visiting their kin abroad.

The Flow from Fiji to Rotuma

In 1961 Howard described the following reasons Rotumans in Fiji returned to Rotuma: (1) to reassert rights to land, (2) to assume Government positions, usually on a term basis, (3) to visit relatives or come for a special occasion, (4) to find a spouse, and (5) to find a haven following occupational or educational failures in Fiji. Each of these reasons still apply, although they have changed in form and/or intensity. In addition, a substantial number of individuals are choosing to return to Rotuma after a number of years away, including an increasing number of retirees.

As mentioned above, one reason for the decline in average household size is because of an increase in the number of caretaking units--households maintained primarily to keep active control of family lands. Thus, one sibling out of a large set may be designated to stay behind, or siblings may rotate the responsibility, taking one another's place in turn to manage the family homestead on Rotuma. However, a less radical solution to the problem of maintaining rights in land is to keep actively connected to the households of close kin through periodic visits. In any case, people who stay away too long may weaken their claims to family land and run the risk of being land-impoverished should they or their children eventually choose to return.

The number of Government workers--medical personnel, school teachers, postal workers, policemen, agricultural officers, etc.--posted to Rotuma has increased substantially since 1960 when there were about 30; in 1992 there were more than 100. [8] The majority of these are professional or clerical workers who have received advanced training in Fiji or abroad. Most are posted to Rotuma for limited periods, ranging from a few months to several years.

As in 1960, many Rotumans who reside in Fiji or abroad come to visit the island for brief visits, often lasting from one plane or boat to the next. Typically a visitor will spend a few days with each of a few different close relatives, reaffirming ties. Increasingly trips to Rotuma have been planned in advance by organised groups to coincide with special occasions, such as the 150th anniversary of the arrival of missionary John Williams at Rotuma in 1839. For many Rotumans born away from the island, a trip "home" has special significance, reinforcing their sense of ethnic identity and affiliation with the broader Rotuman community.

For one subgroup--those in the market for mates--the motivation to return to Rotuma has no doubt diminished. In 1960 the Rotuman population in Fiji was half that on the island, where the large majority of potential (Rotuman) mates was to be found. Today, however, Fiji constitutes a much larger pool. If there has been any change in the direction of this flow it has reversed, with persons of marriageable age having their best chances in Fiji. The stigma of marrying non-Rotumans has relaxed in the interim, resulting in an even greater pool of potential mates in Fiji and abroad.

Finally, there are the returnees, Rotumans who have lived abroad for extensive periods and choose to return to the island to settle. These can be divided into three subgroups. (1) There are those who have failed in Fiji or abroad, and return to Rotuma as a haven, just as their predecessors did in 1960. Young men and women may go to Fiji for further education or looking for jobs, only to find frustration. Especially in urban areas they are likely to be a burden on the families with whom they stay, and pressure mounts for them to return to Rotuma where they can be productive gardening, fishing and performing traditional roles. One indication of Rotuma's place as a haven is an evident increase in the number of disabled persons residing on the island (see Rensel and Howard 1992). (2) A second subgroup consists of people who have opted voluntarily for Rotuma's lifestyle, or choose to return for other reasons (e.g., to be close to certain relatives, to help support aging parents). (3) Finally there are the retirees,people who have reached retirement age and either want to return to Rotuma for sentimental reasons or find that their pensions are too small to allow them to live comfortably in Fiji. The potential pool of retired returnees grows each year, reflecting the increased curve of emigration to Fiji following World War II (see Figure 1). In our 1989 survey we identified 45 individuals with work histories in Fiji or abroad who had returned to Rotuma after ten or more years away. Overall, 26% of adults in our 1989 survey reported having stayed abroad for periods of five or more years. This, incidentally, is comparable to the 1960 figures (28.9%), suggesting the continuation of a pattern of return migration that was well established over thirty years ago. [9]

The Flow of Goods and the Money Economy

In 1961 Howard documented Rotuma's increasing economic dependence on Fiji's urban centers. At the time, business on the island was still dominated by Morris Hedstrom and Burns Philp, although the Rotuma Co-operative Association was rapidly gaining ground. Income then was almost entirely dependent upon copra exports, supplemented by a small number of wage positions. The intervening years have seen a dramatic increase in commercial activity, with copra now accounting for only a small portion of cash income.

In 1970, two years after the firms closed their shops, RCA reported a store turnover of F$319,044. By 1986 RCA's volume of sales had increased 237% and surpassed F$1,000,000 annually. Income from copra, however, which remains the island's chief export, only increased 49%. Whereas in 1970 store turnover exceeded copra income by only 40%, by 1986 the discrepancy was 217%. It is apparent, therefore, that now most of the money being spent in the shops is coming from somewhere other than copra sales.

A portion of additional revenue has come from entrepreneurial activities by people on the island, including yam, vanilla, bêche-de-mer and lobster exports, but these have been small-scale, family operations. Increased wage earnings, and the availability of bank loans from the Rotuma branch of the National Bank of Fiji in July 1988, have also contributed to purchasing power. Cash remittances have also grown considerably in recent years and now constitute a major income source. In 1976 the average monthly total sent to Rotuma by telegraphic money order (TMO) was between F$5000 and F$6000 (Plant 1991:210). In contrast, the amounts sent by TMO for the years 1982-8 averaged over F$10,000/month (Rotuma Post Office records). Money orders are but one means by which remittances reach Rotuma; cash and checks are also sent by mail or brought by visitors.

According to our 1989 study, just under half (49%) of Rotuman households reported receiving remittances; the number of individuals listed as contributing financial resources to a given household ranged from none to seven. Reported amounts ranged from F$10 to F$4000 at a time, with a median amount of F$100. Cash was sent primarily for general support, that is, to be spent on food and other household needs. Other remittances came as money gifts for special occasions--Mother's or Father's Days, birthdays, Christmas, funerals--or periodic needs such as school fees. Larger amounts were sent (often in response to a request) for church fund-raisers, for house construction or improvement projects.

Rotumans are putting more and more of their resources into modernising and improving their homes. A number of two-story homes have been built, and such features as verandahs, louvered windows, and rubber-tiled floors are now common. European-style furniture is also very much in evidence. Whereas two-thirds of the homes were without furniture in 1960, [10] almost every home now has tables and chairs; most have sofas and standing beds. Sewing machines, refrigerators, gas stoves and other household appliances have also become commonplace (see Rensel 1991:190-191 for details). Although remittances account for a major portion of these improvements, a number of wage earners recently have taken out substantial bank loans to improve their houses, or buy vehicles or obtain household furnishings.

Material goods flowing from Rotumans on the island to those in Fiji have also increased in volume over the intervening years. In large measure this is the result of the increased size of the Rotuman community in Fiji, which continues to value products from the island. These include: Rotuman handicrafts, especially fine white mats (vital for presentation on ceremonial occasions such as weddings and funerals); livestock such as pigs and goats; island produce such as taro, yams, coconuts, oranges; and prepared Rotuman delicacies, frequently sent by plane with a passenger. Any of these items may be sent to relatives for special occasions, to accompany a visiting family member, or just as te fakhanisi 'a gift', helping to express and maintain ties. Improvements in transportation between Rotuma and Fiji have facilitated the flow in both directions.

The Flow of Information and Ideas

The reduction of Rotuma's isolation over the years is also manifest in a substantial increase in the flow of information between the island and the outside world. In 1960, between ship arrivals, communication with the outside world was limited to a radio-telephone at the Government station and short-wave radios. Mail came and went with the ships, which often meant waiting 6-8 weeks. Nowadays mail comes and goes on the weekly plane (if it isn't off-loaded in favor of more lucrative cargo), which also brings copies of the Fiji Times for numerous subscribers. The old radio-telephone, noted for it erratic reception and transmission, was replaced in 1990 by a new, more powerful and reliable radiophone. Telephone lines were laid in the section of the island near the Government station, and a number of households now have telephone service. A switchboard, with trained operators in attendance has been installed makes it much easier for people in Fiji to contact their kin on Rotuma direct. And, while calls are still relatively expensive, the fact that people on the island have a lot more money at their disposal has greatly increased the volume of calls to Fiji (and abroad, since it is now possible to get international connections). The telephone is, therefore, rapidly becoming a major source of information exchange between Rotuma and the outside world, transmitted on a daily basis. It also provides a ready vehicle for making requests for money and assistance, a source of some concern for wage-earning Rotumans in Fiji and abroad.

Whereas only a few households on Rotuma owned radios in 1960, nearly all now have them. People regularly listen to news broadcasts in English and Fijian, as well as popular music. A weekly Rotuman language program on Radio Fiji provides news, discussions and announcements of events of interest to the entire Rotuman community. Live television programming has just begun in Fiji, but cannot be received on Rotuma for lack of a cable or satellite dish. Nevertheless, VCRs made their appearance a few years ago and a fresh supply of video cassettes are imported on a regular basis by a few individuals who have made a business out of showing recent movies on television monitors. This has completely replaced the ancient films formerly shown to large audiences once a week in a Quonset hut theater. Rotumans are therefore much better informed nowadays about modern fashions, at least as they are portrayed in the movies.

Perhaps most important of all has been the intensification of personal contacts between Rotumans who have extensive overseas experience and those who do not. Many young Rotumans have been raised and educated in urban settings, have traveled widely, and hold cosmopolitan views. When they return to Rotuma, either to take Government positions for a few years or for short visits, they inform, criticise and otherwise share their opinions. Although not always listened to with deference, their views circulate and gradually become part of mainstream Rotuman culture.

To summarise, information flows between Rotuma and the outside world have increased dramatically in volume since 1961, and have altered to include a much greater proportion of ideas and cultural notions that derive from mainstream cosmopolitan world culture.

Sociocultural Implications

Many important economic, social, political and cultural changes have accompanied the intensified flows of people, goods, money, information and ideas described above. While cause and effect are difficult to untangle, since many of the changes occurred concurrently, we describe in this section changes which, if not directly caused by augmented interchange with the outside world, were at least partially shaped by it. We discuss four areas of transformation: social differentiation, social cohesion, political control and cultural identity.

Social Differentiation

During the colonial period, social differentiation on Rotuma was based primarily on Polynesian notions of aristocracy. Europeans, including colonial administrators, were external to a system of chiefly titles and graded genealogies tracing ancestry to prior chiefs. There was no chiefly caste; bilineal kinship reckoning insured that most men were eligible for one title or another. Titles were ranked (although not fixedly) in order of prestige and political power. The paramount chiefs of Rotuma's seven districts were at the top of the hierarchy, and within each district were sub-chiefs with various rights and responsibilities. Titles often, but not always, included rights to blocks of land and hence to resources on the land, but since resources were largely limited to consumables, surpluses were generally redistributed and thus transformed into prestige. The exception was copra, which was sold for export and produced cash income. However, in the copra trade chiefs enjoyed only a marginal advantage, if any, over their subjects, and were under pressure to redistribute what wealth they might have accumulated in the form of lavish gifts on ceremonial occasions. Housing was perhaps the one area where chiefly advantage was materially manifest, and this was mainly limited to the paramount chiefs. Bcause it was expected that paramount chiefs would entertain important visitors to their district, their homes were generally more lavish than their subjects, but only by a small degree. In fact, in 1960 an outsider would have had difficulty in distinguishing chiefs outside of ceremonial contexts. Social differentiation on Rotuma, although very much on people's minds (in the sense that they were constantly concerned with issues of rank and prestige), therefore was muted by limited availability of wealth-producing resources.

Future changes were foreshadowed in 1960 by the remarkable success that the vanguard of Rotuman migrants were experiencing in Fiji. Many had received secondary and tertiary education and had gone into professions. Others had taken Government positions and were advancing up the bureaucratic ladders. At the Vatukoula gold mine Rotumans were disproportionately represented in positions of responsibility (Howard 1970). A number of successful individuals returned to Rotuma, mainly in Government posts, to form an "emergent elite" in counterpoint to the chiefs (Howard 1963).

The success of Rotuman migrants abroad has continued, and in Fiji they are well-represented among the upper ranks in most occupations. Many are now comparatively wealthy and/or wield a significant degree of political power at a national level. Few of them hold traditional titles, nor do they aspire to them. While they respect chiefs as symbolic figureheads who embody Rotuman tradition, titles are perceived as largely irrelevant to their main endeavours, particularly in Fiji. The chiefs themselves are, with one exception, without advanced education and devoid of significant income or wealth. Culturally, therefore, wealth has been dissociated from chieftainship, and the off-island urban elite now enjoy a recognisably privileged position relative to their kinsmen on Rotuma.

The advantages enjoyed by chiefs on Rotuma have been threatened by the increased number of wage-earners on the island and the rising tide of remittances that have flowed to Rotuma. Some households on the island are the recipients of sufficient funds to build elaborate houses, to furnish them well if not lavishly, to buy cars or trucks, and generally to sustain an elevated lifestyle. Funds available from remittances, loans and grants are invested by some in business ventures that further increase their access to material resources. Other households receive nothing and remain close to subsistence standards. [11] Whether a person is a chief or not is of little consequence in this developing economic differentiation.

Economic differentiation has been accompanied by perceptible shifts in values that underlie notions of social merit and prestige. Whereas, in 1960, the main measure of a man was his productivity as a farmer and fisherman, the measure of a woman her productivity as a mat-maker, today esteem (and often envy) is more often based on education, occupation and wealth in Western terms. In particular, competition has developed with regard to housing, which has come to symbolise a family's fortunes in a way comparable to the productivity of one's gardens in the past. What is perhaps most significant about this value change is that, whereas produce from the land and sea are generally shared rather widely in the form of ceremonial redistribution and interhousehold exchange, houses are almost entirely for the benefit of the immediate family. The basis for prestige is, therefore, shifting from communal to household welfare and, with regard to education and occupation, to individual accomplishments.

This is not to say that the old values are dead. Productivity and communal sharing are still highly valued, and fine white mats are still vital for any ceremonial event. However, there are clear indications that, despite increased access to plantations via improved bush roads and motorised transport, agricultural productivity has significantly declined over the past thirty years (see Figure 4). Fishing productivity also appears to have declined, although no precise figures are available to verify this, aside from the extensive turnover of tinned fish in the shops. The manufacture of fine white mats has also noticeably declined; few young women learn the skill, and what new mats are produced are woven by a shrinking cadre of older women. As a consequence of increased demand (due to an expanded population, including Rotumans in Fiji), fewer weavers and general inflation, fine mats that would have cost around F$10 in 1960 are now valued at over F$400. Other prestige items, like pigs, cows, and tinned corned beef (all of which are used in ceremonial prestations), have likewise inflated in value. This means that access to money is increasingly important if one is to meet ceremonial obligations. With the depressed price of copra on the world market, access to money through wages or remittances is more and more vital if one is to compete for prestige through such contributions.


Thus, an incipient class structure is developing within the Rotuman community. In Fiji, a Rotuman elite enjoy a comfortable lifestyle, even by cosmopolitan standards. Beneath them is a broadening middle-class of white-collar and skilled manual workers who own their own homes, have cars and are economically relatively secure. At the lower end of the spectrum are unskilled workers and the unemployed, although most of these are attached to households with wage earners. Long term unemployment often results in a return to Rotuma, to subsistence activities.

On Rotuma social differentiation based on wealth has not progressed as far, but differences in lifestyle are beginning to emerge. Households with ready access to cash have more elaborate houses, furniture, appliances; they own vehicles; they purchase and consume a much higher proportion of imported, high prestige, foods; they may even own a video and TV monitor. Little time may be spent on subsistence activities: gardening; tending to chickens, pigs, cows and goats; cutting copra; fishing on the reef. Cash-poor households must engage in all of these activities, which are increasingly seen as hard work for relatively low payoff when compared with wage employment. Titles are no longer the main avenue for social enhancement, and the prestige enjoyed by chiefs and titled men is significantly diminished to the extent that they must rely on subsistence activities to maintain their households.

Social Cohesion

Older Rotumans frequently mourn the loss of community spirit that they recall from the colonial period. Indeed, in 1960 the range and frequency of community activities were much greater than in recent years. Community members cooperated regularly in village cleanups, road maintenance, building and repairing communal buildings, and various other multi-household endeavors. What goes unsaid is that the District Officer used to demand that most of these chores be done by donated labour, and there were sanctions for shirking. Today, many of the same jobs are the responsibility of paid Government workers. Nevertheless, there were other indicators of community spirit present in 1960 that are less in evidence today, corroborating the old-timers' laments. Foremost among these was the sheer amount of time people spent together in public spaces as opposed to the confines of their homes. Whereas in the past people generally used their homes only for sleeping, cooking (in adjacent cookhouses) and eating, today they engage in a much wider range of home-based activities. More subjectively, commitments to one's village mates appear to have been on firmer footing then than now.

To the extent that there has been a demise in community spirit, several factors associated with migration have contributed. One is that the movement of people into and out of villages has increased dramatically, leading to a lower density of shared histories (both in the sense of a history of common participation in past activities and in the sense of less knowledge about the details of one another's lives). Limitations on motorised transport in the past meant that people were largely confined to their home communities, and that trips to other parts of the island were major expeditions. Today almost every household has access to a motorbike, car or truck, in addition to the school buses and transportation for hire. People now travel around the island routinely, and spend long periods away. In addition, increased opportunities for travel abroad, often for extended stays, contribute to a lower intensity of common experience (which in Rotuma as in most other small communities is anchored in shared participation in memorable events).

This differentiation of experience through increased mobility, along with greater exposure to people from other cultures and international media, has also led to a greater range of ideas being incorporated into the Rotuman cultural pool, and to a broader range of attitudes, beliefs and values. Perhaps indicative is the advent of two new religious groups that have taken root on Rotuma, the Seventh Day Adventists and Jehovah's Witnesses. Given the centrality of church activities in Rotuman social life, religious fragmentation is symptomatic of declining social cohesion, at least within the broader community. Furthermore, there is compelling evidence that religion is losing its integrative power, even at the level of the household. Whereas, in 1960, communities, with very few exceptions, were religiously homogeneous (either Catholic or Methodist), today heterogeneity is the rule and, even within households, members may attend different churches.

Values have also differentiated along lines associated with "progress and modernisation" versus cultural conservatism, a line of cleavage that infuses many political issues. A recent controversy concerning tourism is a case in point. On one side were "modernists" promoting the case for bringing tourists to Rotuma. They emphasised the economic benefits that would accrue from having visitors to the island. On the other side were the conservatives, led by the clergy, who warned of the threat tourism would pose to traditional Rotuman values. They expressed deep concern about the deleterious effects greed for money would have on social solidarity. Ironically, although public opinion shifted toward acceptance after the first few tourist vessels visited Rotuma, political dissension within the community over who should organise and control such visits has led to paralysis and a cessation of organised tourism.

The greater availability of money, through increased wages and remittances, means that Rotumans today are less materially dependent on their immediate kin and neighbours. Whereas, in the past, interdependence was fostered by the need to help one another building thatched houses, for example, today's homes require purchased materials (cement and wood) and skilled labor that must be cash compensated (see Rensel 1991). Having money at one's disposal is, thus, becoming more relevant for material well-being than a commitment to mutual support within the community. As a result, people are now investing more in relationships outside their villages, and particularly in relationships with relatives abroad who might send remittances, building supplies and other valuables. A concomitant shift is toward an emphasis on lineal and nuclear relations at the expense of maintaining a more extended network of kin, since children and siblings are more inclined to send money and share material resources. Reciprocal exchange patterns have therefore been shifting their locus from geographically confined to geographically dispersed networks, and to a longer time frame for reciprocation.

Political Control

With the dissolution of colonial status in 1970, when Fiji gained its independence from Great Britain, political authority on Rotuma shifted from the Government-appointed District Officer to the Council of Rotuma (Howard 1989). The chiefs, district representatives and other members of the Council are responsible for formulating policy and administering a budget composed of an annual subvention and special grants. In theory, the Council is in a strong position to run the affairs of the island and to direct the course of development. The reality is quite different, however, in large measure because Council members lack the skills and political resources of Rotumans in Fiji. As education is a prerequisite for neither chief nor district representative, most Council members are ill-equipped to manage what has become an increasingly complex political economy. They are unschooled as well in the manipulation of bureaucratic structures that so frequently frustrate even well-planned social and economic programs.

In contrast, the Rotuman elite in Fiji have accumulated a good deal of political and economic know-how. As pointed out above, many hold powerful positions in Government, banking, private industry and the professions. They control resources well beyond those available to the Council of Rotuma, and have established power-laden networks of relationships throughout Fiji and, in some instances internationally. They know how to navigate projects through the bureaucracies, and have the political clout to twist a few arms if necessary. As a result, the Council has become increasingly dependent on the good will of Rotumans in Fiji and abroad to see developmental projects through.

Not surprisingly, the elite in Fiji who have successfully championed Rotuman causes have on occasion pushed their own agendas for the island. Sometimes this has been to the dismay of Council members, who feel it is their prerogative to set policy. The chiefs, in particular, express the view that Rotumans everywhere should respect their authority. For the most part we have been impressed at the lengths educated Rotumans go to accommodate the chiefs, but there are limits. On several occasions chiefs have made unreasonable demands, expecting their (usually younger) kinsmen to abuse powers on their behalf. Rejections have led to disputes, that on one occasion, resulted in a new chief's being chosen in opposition to the reigning office-holder, splitting allegiances within the district (Howard and Rensel 1993).

To summarise, tension has been building between the emergent elite in Fiji, who control significantly more resources and have far more political clout nationally and internationally, and the hereditary and elected leaders on Rotuma, who are in a much weaker position but make strong claims for the right to decide the island's destiny by virtue of their being in residence. Although chieftainship as an institution is valued by almost all Rotumans, educated and uneducated alike--it is rightfully perceived as being at the heart of Rotuman custom and identity--the balance of influence appears to be shifting toward the elite. In large measure this is because the chiefs increasingly are seen as incapable of filling the leadership roles they have assumed. Calls for their relegation to strictly ceremonial positions, with leaders chosen from among the better educated segment of the community, have become more strident in recent years. Indicative of the shift in power to the migrant community is the fact that Rotuma's representative in Fiji's first elected post-coup Parliament is a Sandhurst (England) educated military-cum-business leader who has spent very little time on Rotuma since his youth.

Cultural Identity

In a paper published in 1977, Howard and Howard argued that migration to Fiji had resulted in the development of a pan-Rotuman identity that had hitherto been muted. Prior to the establishment of enclaves in Fiji, they argued, district and village loyalties predominated, so that oppositions at the level of Rotuman/Other were relatively rarely encountered. In Fiji, however, Rotumans came into daily first-hand contact with Fijians, Indians, Europeans and other Pacific Islanders, to whom they were ethnically categorised as "Rotuman". Inevitably, their consciousness at this level of cultural identity was raised and, in Fiji at least, given precedence. Continuous outmigration has resulted in new complications, however, and the past few years have seen questions raised about who have rights to claim Rotuman identity. One complication is language; another is inter-marriage between Rotumans and outsiders.

Since virtually all the early emigrants were raised on the island, they spoke Rotuman fluently. They might have learned some English in school, and possibly picked up some Fijian, but the Rotuman language was at the heart of their social existence. When in Fiji they generally spoke Rotuman at home, but at work and in the outside community they were required to use Fijian and English, and sometimes Hindi as well. More importantly, their children learned English and Fijian at school. If their parents spoke to them in Rotuman, and if they spent a good deal of time interacting with Rotuman-speaking friends and relatives, the children readily became trilingual. However, some families were more isolated, and the Rotuman language was relegated to lesser use, resulting in imperfect socialisation. In some instances (particularly when one spouse was not Rotuman), the language of the home was English or Fijian, and the children grew up knowing little, if any, Rotuman. As a result there now exists a subset of the Rotuman population who are non-speakers of their mother tongue, while for others Rotuman is a second or third language. Thus, one of the most distinctive markers of Rotuman cultural identity is being blurred.

Marriage with outsiders--the almost inevitable result of residing in a multi-ethnic environment--has also introduced complications. Like many other Pacific Islanders, Rotumans have historically been inclusive in responding to claims of Rotuman identity. Anyone with some Rotuman ancestry, or anyone who had been adopted into a Rotuman family and lived by appropriate cultural pre- and proscriptions, was likely to be accepted unproblematically. Inclusion was, therefore, not a problem for the offspring of marriages between Rotumans and outsiders. With the increased volume of such marriages, however, in conjunction with a sharpened competition for scarce resources, people have begun to question the grounds for entitlement to claims of Rotuman identity. A recent case involving the son of a Rotuman woman married to an Indian man brought the issue to a head. The boy had applied, successfully, for a scholarship reserved for Rotuman students. But he did not speak Rotuman and had never been to the island. He was virtually unknown in the Rotuman community. To make matters worse, he had an Indian name. People were quite disconcerted that a rare commodity like a scholarship, designed to benefit the Rotuman community, was allocated to someone so marginal.

Precisely because Rotuman cultural identity is becoming more problematic, emigrants are finding it increasingly important to make trips to the island periodically, to reinforce their ties with relatives there. Parents often take special pains to bring or send their children on return trips to validate their ethnic claims, since ultimately it is by maintaining active ties of kinship with core members that one can secure an unproblematic Rotuman identity. Thus, some emigrants' offspring are fully accepted even though they speak little Rotuman and have little knowledge of the culture, while others who are more knowledgeable remain marginal by virtue of their self-imposed isolation.

After a period of indifference, Rotumans are now concerned about preserving their language and heritage within the expanded community. They have gained acceptance in the school system for teaching Rotuman as an optional vernacular language, along with Fijian and Hindi. (When, just a couple of years ago, we asked a Fijian education specialist why Rotuman was not taught as a vernacular language he replied that it was not necessary, that the children just picked it up at home!) Teaching materials about Rotuman language and customs have been developed, and Rotuman teachers are now seeking resources about Rotuman history. Fijian history is taught in the schools, but there is virtually no mention of Rotuma, and they want something done about it. Migration has therefore not only led to the formation of a pan-Rotuman cultural identity; it has rendered that identity problematic and a cause for concern.


We may have stretched a metaphor a bit by suggesting that the island of Rotuma is becoming a neighbourhood within an extended Rotuman community. The point we wish to make by this exaggeration is that the classic model of migration, imaged upon one-way voyages to a destination community, is too impoverished to describe, let alone account for, contemporary patterns of population movement in the contemporary Pacific (see Chapman 1991). The multidirectional flows of population must be considered along with concomitant flows of money, goods and commodities, information and ideas. We are suggesting that, in the case of Rotuma, these flows are not significantly different from those that pertain between neighbourhoods within large cities, and that interactions between subsections of the Rotuman population--by telephone, letters, and occasional sojourns back and forth--resemble relationships maintained between kinsmen and friends in modern urban contexts. A trip to Rotuma is like going back to "the old neighbourhood" for a visit. The intensity of relationship varies, of course, but we should not let the fact of distance (and the intervening ocean) lead us to overemphasise differences from other dispersed communities. The important point to make is that the vast majority of contemporary Rotumans form a lively interactive community, and that all parts of it are evolving in concert. As technology improves, making transportation and communication even easier, the bonds between the various segments are likely to become even stronger, although, as in any community, some people will drift away, forsaking their ties to the old neighbourhood.


Support for the research on which the paper is based was received from several sources. The National Institute of Mental Health funded Howard's 1959-61 study. During the 1987-91 period he received funding from the Wenner-Gren Foundation, the University of Hawaii Office of Research Administration, and the University of Hawaii Program on Conflict Resolution (UHPCR). Rensel's research was sponsored in part by a Fulbright Grant administered by the Institute for International Education. We are indebted to the many Rotumans who have provided data, and especially the school teachers who conducted our 1989 household survey. The paper was originally presented at an Association for Social Anthropology in Oceania symposium on Pacific Migration, Kona, Hawai'i, March 24th-27th, 1993.


[1] We would especially like to thank John Bennett and Visanti Makrava for keeping us abreast of developments by fax communications. [back to text]

[2] Figures for one district, Malhaha, are excluded from the copy of the survey made available to us. [back to text]

[3] There is some evidence that native style houses are making a comeback, however. In 1989, 40 houses, or approximately 10% of houses on the island, had thatched roofs, and six of these had thatched walls as well. [back to text]

[4] In fact, the full extent of the increase in the population of Rotumans cannot be determined by the Fiji census alone, since a substantial number of Rotumans now reside abroad in New Zealand, Australia, England, the United States and elsewhere. [back to text]

[5] This is also a strategy for keeping active rights in productive bush land, especially if the house is built on a fuag ri "named house foundation" to which bush land is attached. [back to text]

[6] The individuals in this third category were counted for purposes of determining household size, but not household structure. [back to text]

[7] For the remaining 53 individuals no information was recorded for current residence. A rough estimate of Rotumans abroad can be obtained by multiplying by a factor of 7 (derived from dividing the 1986 Census figures for Rotumans in Fiji by the number of household members listed as Fiji residents in our 1989 survey). This is based on the admittedly dubious assumption that there are no biases in reporting household members away depending on their location. [back to text]

The extrapolation results in estimates of 378 Rotumans in Australia, 154 in New Zealand, 140 in other Pacific Islands, 91 in the U.S., 70 in Europe, and 42 in Canada. These figures obviously must be viewed with caution.

[8] The figure for 1960 derives from Howard's survey, during which 28 people were identified as Government employees. The 1992 data are from lists furnished by the Rotuma District Officer and Fiji Education Department. Of the 106 persons on the latter lists, there are 37 teachers, 16 medical staff, 14 Public Works employees, 13 individuals working for the District Administration, and 8 Water Supply workers; the remaining 18 were distributed among seven other Government departments (post & telecommunications, police, civil aviation, meteorology, agriculture, co-operatives, women & culture). [back to text]

[9] We have reason to believe that the 1989 figures are less accurate, in the direction of under-reportage, than the 1960 figures. The 1960 data are based on complete residential histories, while the 1989 data are based only on reports of longest period of residence abroad. [back to text]

[10] During Howard's 1960 survey his two Rotuman research assistants categorised each house as "Rotuman style" or "European style" depending on interior furnishings. Rotuman style was characterised as having minimal furniture, requiring families to eat, sleep and sit on floor mats; European style houses had furniture for these purposes. [back to text]

[11] Whereas some households both include a wage-earner and receive remittances, almost 27% of households in our 1989 study reported neither. It is likely that at least some of these made money from sources not addressed by our survey, e.g., periodic sales of produce or livestock on the island or as exports. Also, those with financial resources do redistribute their bounty to some extent, e.g., paying people for services, sharing food. This, along with Rotuma's natural bounty, helps to mitigate the development of an impoverished underclass. [back to text]


Chapman, Murray, 1991. Pacific Island Movement and Socioeconomic Change: Metaphors of Misunderstanding. Population and Development Review 17:263-292.

Howard, Alan, 1961. Rotuma as a Hinterland Community. Journal of the Polynesian Society 70:272-299.

------ 1963. Conservative and Non-Traditional Leadership in Rotuma. Journal of the Polynesian Society 72:65-77.

------ 1970. Learning to Be Rotuman. New York: Teachers College Press.

------ 1989. The Resurgence of Rivalry: Politics in Post-Colonial Rotuma. Dialectical Anthropology 14:145-158.

------ 1990. Dispute Management in Rotuma. Journal of Anthropological Research 46:263-292.

------ 1991. Seamanship in Rotuma. Paper presented at Symposium on Seafaring in Oceania at the annual meeting of the Association for Social Anthropology in Oceania, March 27-31, 1991, Victoria, British Columbia.

------ and Irwin Howard, 1977. Rotumans in Fiji: The Genesis of an Ethnic Group. In M. Lieber (ed.), Exiles and Migrants in Oceania. Honolulu: University of Hawaii Press.

------ and Jan Rensel, 1993. Ritual Status and Power Politics in Modern Rotuma. Paper presented at Symposium on Chiefs Today at the annual meeting of the Association for Social Anthropology in Oceania, March 24-27, 1993, Kona, Hawaii.

Plant, Chris, 1991. The Development Dilemma, in Anselmo Fatiaki et al., Rotuma: Hanue Pumue (Precious Land). Suva: Institute for Pacific Studies, University of the South Pacific

Rensel, Jan, 1991. Housing and Social Relationships on Rotuma, in Anselmo Fatiaki et al., Rotuma: Hanue Pumue (Precious Land). Suva: Institute for Pacific Studies, University of the South Pacific.

------ and Alan Howard, 1992. The Place of Disabled Persons in Rotuman Society. Paper presented at working session on Culture and Disability in the Pacific at the annual meeting of the Association for Social Anthropology in Oceania, February 19-23, 1992, New Orleans, Louisiana.

Home Page
Howard-Rensel Papers