Home Page
Howard-Rensel Papers

12. History, myth and Polynesian chieftainship: the case of Rotuman kings

Alan Howard

[Published in Transformations of Polynesian Culture, edited by A. Hooper and J. Huntsman Auckland: Polynesian Society, 1985]

The relationship between myth and history has become a central issue for anthropologists interested in the study of traditional societies. It has been brought to the fore by the work of Levi-Strauss on myth and the sharp contrast he draws between these alternative modes of organising discourse about social phenomena. The privileged position Levi-Strauss grants to myth has led to impassioned critiques and counter-critiques involving Marxists, structuralists and a number of prominent European intellectuals. The issue is perhaps especially important for Polynesianists since so much of the early literature in the region focused on oral narratives, recounting the deeds of ancestors whose characteristics ranged from godlike to mundanely human. In large part this body of literature was spawned by European fascination with the problem of Polynesian origins and migrations. Informants were incessantly asked where their ancestors had migrated from, triggering founding myths, stories of epic voyages, and the like. However, it is also apparent that Polynesians found myth a congenial medium for communication, and seem to have felt that they were disclosing truly important information about themselves when relating myths.

In attempting to make sense out of Polynesian myths early scholars such as Abraham Fornander, S. Percy Smith, and Te Rangi Hiroa treated them as ethnohistory, correct in their main features though possibly incorrect in detail (see, for example, Smith 1910:19). They viewed the narratives as reflective of actual events, some of which, particularly those occurring in the distant past, were overlain with mythical rhetoric. This view naturally reflected their own preoccupation with historical problems and their eagerness to use the narratives, which for the most part were translated into an historical idiom, as evidence for their theories. Although such use of oral narratives was severely attacked by sceptical critics, the effort to place these materials in the service of history has not been readily abandoned. Roberton (1962), for example, has argued that the distortions which creep into legendary material can be compensated for by proper analytical techniques, and that such data cannot be dismissed as unworthy of serious consideration as history. Perhaps the most compelling use of myth as history is incorporated in the study of Tikopia's prehistory by Kirch and Yen. They evaluated the validity of oral traditions collected by Firth by checking them against archaeological data and found the correspondence much too close to dismiss their credibility as history. In their opinion the oral traditions and the "quasi-history" they represent

...provide a window to the past quite different--although complementary--to that offered by archaeology. This view is one of social process, as opposed to that of the material, technological, and environmental conditions of culture change more readily revealed by archaeological evidence (1982:364).

An entirely different perspective on the relationship between history and myth has recently been presented in a brilliant essay by Marshall Sahlins. In considering the dramatic events that followed Captain Cook's discovery of the Hawaiian Islands, Sahlins argues that "Hawaiian history often repeats itself, since only the second time is it an event. The first time it is myth" (1981a:9). For Polynesians, he maintains, myths present archetype situations in which the experiences of mythical protagonists are re-experienced by the living in analogous circumstances (1981a:14).

If Sahlins is correct, and I believe that he is, the study of myth in Polynesian societies can be viewed as an important means of organising and interpreting history rather than chronicling it. In this paper I address an historical problem from the island of Rotuma, which is now part of the Republic of Fiji [1] The problem concerns a curious form of kingship in which the position was held by representatives of different districts in rotation, for restricted periods of time. After providing historical documentation of this institution, I present two myths for interpretation. The interpretation forms a basis for reflection on the character of political institutions in Rotuma, as conceived by the Rotumans. An attempt is then made to relate these conceptions to political pragmatics, particularly as these were affected by historical processes following European intrusion. An explanation is offered for the institution that has implications for Polynesian chieftainship in general, and these are explored in the final section. It is my hope that the case will prove sufficiently compelling so that others will be encouraged to explore the value of myth for unravelling historical mysteries.

Early Accounts of the Rotuman Political System

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'. However, there were also three positions that were pan-Rotuman in scope: the fakpure, sau and mua. The fakpure was referred to primarily in two capacities in the early literature: 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. He was gagaj 'es itu'u of one of the districts, presumably the one who headed an alliance that was victorious in the last war. The sau's basic role was to take part in the ritual cycle, oriented towards ensuring prosperity, as an object of veneration. 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 c. 1873 that the sau appeared to be an appendage of the fakpure, while the mua appeared to be more associated with spiritual power (Sumi Mission Station, Rotuma manuscript).

Most early accounts focus on the office of sau, which generally was translated into English as 'king'. A curious aspect of this position is that it was held by district representatives in rotation, for restricted periods of time. Rotuman chieftainship at this level has been compared with that of Mangaia and Easter Island, two other Polynesian societies for which rotating chieftainship has been documented (Williamson 1924).

There is general agreement among informants about several aspects of the sau's office. All agree, for example, that the sau was appointed by the fakpure and that he was chosen from different districts in turn, although no one specifies an order to this rotation. It is also agreed that the sau exercised no secular power and that his main tasks were to eat rather gluttonously on a daily basis, drink kava and take part in the six-month ritual cycle. There is some confusion or disagreement on several important points, however. For example, it is unclear who was eligible to be selected as sau. Lesson reported following his visit in 1824 that Rotuma was divided into 24 districts, each governed by a chief who succeeded to the office in order of seniority. There is nothing known to correspond to these units, since there are only seven itu'u 'districts' at most and considerably more ho'aga, the next smallest unit over which a chief presides. Nevertheless, there does seem to be agreement among those who did comment that eligibility was limited to individuals of chiefly rank [2] Whether a person was actually supposed to hold a title in order to be eligible is nowhere stated. The length of the sau's reign is also unclear. Gardiner 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 (Gardiner 1898:461). 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. Lesson mentioned 20 months as the duration, which makes no sense in terms of the Rotuman ritual cycle, but may reflect his informant's estimate of an average reign (Lesson 1838-39, II:432). Allen, a Methodist missionary who served in Rotuma during the late l9th century, reported that the sau was generally "elected" for short periods of six to twelve months (Allen 1895), while one of Hocart's informants indicated that two cycles was usual (Hocart n.d.), and Dillon was told,

...it sometimes happens that the president does not wish to resign his post at the expiration of six months; when rather than quarrel, they allow him to exceed the time appointed by law: but should he persist in a further maintenance of his power, the other chiefs league together, and compel him by force of arms to retire (Dillon 1829:95).

A further puzzle concerns the rules of residence for sau. Allen 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 (Allen 1895), and in one narrative recorded by Churchward, 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 (Churchward 1938:356). 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 mua also seems to have been a rotational position. Thus Allardyce reported that the districts had the honour of mua "in a kind of turn", and that he was appointed by the fakpure for an indefinite period, though it was customary to resign after about a year (Allardyce 1885-6:142).

How are these early accounts to be interpreted? Just what do they reveal to us about the constitution of the Rotuman polity? And what else might we learn about Rotuman chieftainship by analysing the texts of oral narratives? These three questions motivated the analysis that follows.

In answer to the first question, it is quite clear that the descriptions were obtained verbally from informants, most likely in response to specific questions, rather than from direct observation. None of the accounts describes actual political or ritual events that were witnessed by the writer. At most, then, the descriptions appear to be based upon ideal statements concerning these roles rather than upon political enactment. If Rotumans were motivated to record history in the sense of providing an "accurate" account of actual events we might nevertheless be inclined to treat such descriptions as characteristic of actual practice, but there is sufficient evidence to suggest that Rotuman oral narratives do not represent an attempt to do that kind of work. Instead they appear to be motivated by an interest in exploring the permutations of key structural dilemmas, as I shall try to demonstrate below (see also Howard, in press). In essence, then, Rotuman ethnohistory and myth merge with one another, both being powerfully patterned by an underlying semiotic system of cultural logic. This is not to say Rotumans are incapable of reporting events accurately, they do so all the time. However, the statements recorded by early observers were not of specific events but of presumed usual practice. It is precisely here that the power of the semiotic codes is most in evidence. In one important respect this simplifies our task, for we can dismiss the problem of interpreting traditional political practice on the grounds that we have virtually no usable evidence. All of the data, however, including the mythical texts, are relevant for interpreting Rotuman conceptions of chieftainship and political structure.

For these reasons we must treat Williamson's conclusions concerning the Rotuman political system with extreme scepticism. He accepts Gardiner's speculation that originally the offices of the sau 'sacred ruler' and fakpure 'secular ruler' were united, but that in time they became distinct (Gardiner 1898:460). Concerning the rotation of sau Williamson offers the following speculative scenario:

The sacred king and his family, the trunk family of the group, would probably continue to occupy the ancestral demesne, and there would be a number of families of chiefs, branches of the original royal family, each occupying its own area. The office and over-riding jurisdiction, so far as retained, of the sacred king, would remain with the trunk family, in which the original godship and sanctity would be believed to be specially immanent, and each chief would be subject to that over-riding authority, such as it was, and to the authority of the secular king, retaining, however, some local jurisdiction over his own area. As time went on, the growth and development of the group would continue; the branch families of the chiefs would increase in numbers; and a powerful aristocracy would be evolved. There would be among them a competition for power and predominance, which would show itself in intrigue and inter-family fighting within the group; matrimonial connections between families, and inter-family military alliances would affect the powers of the respective families; and the tendency would be for them to group themselves into mutually hostile combined parties who would contend with each other for secular dominance, success first falling to one and then to the other. Thus would come into being the division of the people into two great camps--the conquerors and the conquered, the strong and the weak--as described by writers .

The position and authority of the sacred king himself might readily be affected, and perhaps undermined, by developments of this character. Thus, whilst in some islands, as in Mangaia, he continued to retain immense power, in others as in Tonga, his power, and even his sacred duties as a high priest, died out altogether, or nearly so; whilst in Rotuma his office became a matter of periodic election from one or other of the families of the island, its hereditary character being lost, and indeed the evidence suggests that he was subject at any time to deprivation of office and replacement as the result of conflicts among his subjects (Williamson 1924:427-8).

Noble as such an attempt might be to account for the constitution of the Rotuman polity, we must recognise that there is virtually no evidence, beyond its inherent plausibility, to support such a conclusion. The answer to the second question is therefore that we know very little about either the historical sequence leading to the political system as described or about the conduct of politics in traditional Rotuma. What we do have is some information about categories of actors and their associations with one another and with types of activities. But this is a reasonable start if we are to set our goal as comprehending the cultural logic of Rotuman politics, and it is in the interest of developing such comprehension that we approach the mythical texts. Before actually presenting the narrative material however, it is necessary to deal with some fundamental issues concerning the character of Rotuman myth and its relationship to the cultural order.

Myth in Rotuman Culture

The first record of Rotuman oral narratives was collected by Fr. Trouillet c. 1873. The account is in the form of a history of Rotuman chieftainship, beginning with the founding of the island by a chief from Samoa named Rao (Raho), [3] and ending in 1868 when the last sau held office. Stories and incidents are set in the reigns of various fakpure, mua and sau. The document is remarkable for its chronological ordering of fabled events, and sets a useful framework for the study of Rotuman mythology. [4]

In the years following Trouillet's initial account several other visitors to the island collected texts of Rotuman narratives, many of which provide alternative versions or elaborations of Trouillet's. These include accounts by Romilly (1893), Gardiner (1898), Hocart (n.d.), MacGregor (n.d.), Churchward (1937-9) and Russell (1942). Of the published accounts, only that of Churchward includes Rotuman texts; the rest appear only in English. [5]

We must recognise these texts for what they are--residues of living performances, recorded by individuals who had particular notions about what was worth recording. They provide virtually no information about the teller and the context of performance, let alone the way the stories were learned and transmitted. One suspects that they are responses to specific questions such as: "Where did the Rotumans come from?"- and "Do you know any other interesting stories about the old days?" In such a context they would likely be stripped of elaborations that would mark performance before indigenous audiences. Nor do we know how the story-tellers classified their tales. Rotumans distinguish between three types of narrative. The most mundane is covered by the word rogo, which as a verb translates as 'to be reported, heard of, generally known' and as a noun as 'report, news' (Churchward 1940:300). The word, sometimes used in its reduplicated form, refers essentially to events witnessed by the teller or reported by presumed reliable sources. A second category is hanuju 'tale, story, especially a fictitious one'. Churchward speculates that this is probably a corruption of hagnuju, from haga 'to feed' and nuju 'mouth' (Churchward 1940:216). The implication seems to be of frivolity, of stories meant merely to entertain. This contrasts with the third type, fäeag tupu'a 'legend, myth' (Churchward 1940:189). The first word, fäeag, [6] refers to speech, the second, tupu'a, is translated by Churchward as follows:

tupu'a, n., (in mythology) immortal man; rock or stone reputed to be such a person petrified: image, statue, idol, doll; star or constellation, esp. as a point for steering by; point of the compass, direction, bearings (Churchward 1940:337).

One of Hocart's informants specifically stated that, in contrast with hanuju, fäeag tupu'a are "true" stories. Whether this referred to a belief in the literal occurrence of events as described is problematic. I suspect that the reference was rather to a notion of structural truth--that these stories truly represent the nature of Rotuman collective experience, in much the same way as an icon represents true religious experience.

Yet another problem is that we know very little about the degree of variation and its correlates. There is some evidence from Hocart's notes that Rotumans of his day discussed specific myths in order to arrive at a consensual account, but where we do have multiple versions there are significant discrepancies. What makes the problem of interpretation somewhat more acute is that the collection of texts covers a period of some seven decades, and we cannot distinguish genuine cultural changes through time from individual and geographical variation within a given time period.

We are thus in a position somewhat akin to that of an archaeologist who is attempting to reconstruct a material order from a fragmented artefactual record, divorced from its performative context. The problem is one of making reasonable inferences, based upon certain assumptions about human behaviour, the utility of various artefacts, the consistency of particular patterns, and so forth. Just as archaeologists have found ethnographic observations (i.e., ethno-archaeology) and comparative analysis useful for grounding such assumptions, analysts of myth can benefit from like efforts. Based upon my own field work on Rotuma in 1960, supplemented by the observations of others, and a comparison of Rotuman cultural patterns with those of other Polynesian societies, I have arrived at the following set of working assumptions concerning Rotuman oral narratives:

1. Stories are constructed out of an extensive array of semiotic codes, which are transmitted in a variety of media. In addition to codes embedded in the string of words from which written texts are constructed, there are expressive codes embedded in speech and gesture, spatial and temporal codes, and a number of other performative codes available that lend meaning to oral narratives when they are told in vivo.

2. Both digital and analogic codes are used. Digital codes are constructed out of basic oppositions between such concepts as land/sea, male/female, person/spirit, east/west, raw/cooked and so on. As Lévi-Strauss has noted, digital codes generate mediating categories, such as beaches, mountains, birds, male-like females, demigods, etc. Digital codes are used in Rotuman myth (and perhaps all myth) to convey messages about basic categories of social construction, e.g., the components of chieftainship, the distinctions between men and women, between humans and gods, and so on. Analogic codes involve changes in degree of states, such as emotion, potency, acceptability and the like. They are used to convey messages about limits and boundaries, and about the implications of variation within categorical parameters. They are most conveniently embedded in expressive media such as gestures and voice intonation, and hence are more prone to being lost when stories are transformed into written texts. Perhaps this is one reason why digital codes have received so much more attention from armchair analysts. [7]

3. In Rotuma, the codes used to construct oral narratives are generative. That is to say that they are subject to a set of meta-rules, or story syntax, that allows for the production of a range of acceptable variation for any particular story. Generative possibilities are influenced by variations in social settings in which stories are told, time allotments, relations of teller to audience as well as the personal and social characteristics of the story-teller. This differs from some societies in which at least a significant segment of oral literature is ritualised, embedded in chant and otherwise restricted to singular versions. It means that attempts to find "correct", "official" or "consensus" versions of Rotuman myths are unwarranted, and raises some questions about the relationship between cultural codes and individual usage. This makes it imperative to examine the full range of available texts before attempting inferences about the meaning of any particular narrative.

4. The written texts recorded by visitors to Rotuma that provide the basis for our analysis are restricted to certain codes and therefore only represent partial semantic structures. Their full meaning has been lost, and it is possible that performative codes significantly altered, perhaps even inverted, some of the meanings in the written texts (as, for example, an ironic tone of voice inverts meaning in English). Corollary to this, the full meaning of key symbols, metaphors and metonyms cannot be recovered from such residual texts. At best they can be inferred from the contextualisation of their usage. For example, in Rotuman myths the opposition between chiefs and commoners is so consistently associated with geographical directions that I feel quite confident in making inferences based upon them. More problematic are inferences to be drawn from proper names of persons or places. In some instances the overt meaning is blatantly obvious, in some it is somewhat suggestive but thoroughly ambiguous, while in other instances there are no grounds whatever for making an inference. This underscores the importance of examining the full range of available texts for consistency of usage so that at least core features of semantic units can be inferred with some degree of conviction. [8] Fortunately, a considerable degree of redundancy occurs in the texts, between as well as within codes. It is reasonable to assume that the messages of greatest concern were the most redundant, and that they were the least likely to be nullified or drastically altered by unrecorded performative codes.

5. Although the texts of narratives are often written in the idiom of history, they do not appear intended to do the work of history in our usual sense, i.e., to accurately record significant events of the past in correct chronological sequence. Whether or not certain incidents related in the narratives are based on actual events, they are processed through such a powerful codification system that their validity as history must be regarded as extremely problematic. A more defensible view is that chronological sequencing provides a mechanism by which structural oscillations are explored in their various permutations (see Howard in press).

6. Rotuman myths appear to reflect a preoccupation with cultural dilemmas associated with relations between humans and supernatural beings, on the one hand, and between chiefs and commoners, on the other. Since gods and chiefs merge conceptually at certain levels of contrast, these two themes can be considered as derivative from a single overarching cultural problem, namely the problem of the genesis and control of mana 'potency'. [9]

At a more basic (often implicit) level the concern is with the continual regenesis of life--with the fertility of the land and the people. The fundamental issue is one of harnessing the mana of the gods in the service of this goal.

7. Fäeag tupu'a 'myths' seem to owe their sense of drama to the fact that they involve explorations of basic structural properties of the cultural system. In contrast with rogo or hanuju, which deal with variations within the received structure, myths explore the consequences of altering the parameters of structure. They thus probe structural properties, and examine the possibilities for structural transformations. Within myths the possible effects of adding, deleting or altering the value of categories can be played with, a process which has the effect of providing a visibility to key aspects of the cultural order that might not be elsewhere apparent (except, perhaps, in ritual). It must be added that the overall effect of myth in Rotuma appears to have been conservative in so far as it focuses on the negative consequences of breaching fundamental principles of structure, e.g., of violating the rules of the use of power (see Howard, in press).

The Mythical Origins of Authority

The mythical bases for political authority in Rotuma are contained in stories concerning two key characters, Raho (Rao) and Tokainiua (Tokaniua). Raho is described as a Samoan chief in most versions, although in some accounts his origins are ambiguously Samoan or Tongan. [10] He is credited with "founding" Rotuma, and is associated with a cultural category, hanuet, that can be glossed as 'indigenous inhabitants' or 'people of the land'. Tokainiua is generally considered to be of Tongan origin, although in Trouillet's account he is described as coming from Fiji. He follows Raho to Rotuma and challenges him for supremacy. Tokainiua is a prototypical usurping warrior and is associated with a cultural category that is in opposition to 'people of the land'; for want of a better gloss we can label this category as 'chief'. [11] It is in the feats of these two demigod figures, and the interplay between them, that the cultural logic of the Rotuman polity is encoded. Since Churchward's versions were the most carefully collected, and include both Rotuman text and English translation, I shall use them as the basis for my interpretation, citing other versions where they suggest elaborations or alternative possibilities. Two narratives, originally published in Oceania (Churchward 1937:109-16, 250-60) are presented below. My procedure is to present a segment of the English text, followed by interpretation, another segment of text, additional interpretation, and so forth. Let us begin with the story of "The Founding of Rotuma," or as Churchward points out in a footnote, more literally, "The Planting of this Rotuma".

1. A chief was living in Samoa, named Raho. He had three sisters. 2. The name of the eldest was Mamaere; the middle one, Mamahioväre; the youngest, Mamafiarere. 3. The youngest was the one that ruled over Savai'i, while the eldest was the one that ruled over the place where Raho lived.

4. Now it was the custom of the eldest sister, as soon as the sun had set [each evening], to go to the door of the house that faced the west, to sleep there; 5. and as soon as the sun rose, she would go along to sleep at the door that faced the east.

6. Now after a short time Mamaere became pregnant; 7. and everybody came to know that the woman was with child, 8. but they were afraid to tell Raho about it, since the woman had no husband.

9. But, as time went on, Raho discovered that his sister was approaching confinement. 10. Thereupon he gathered his people together and asked them who it was that was responsible for his sister's condition. 11. To which the people replied that none of them had been near the woman. 12. So Raho then told the people to start making preparations for the birth-feast.

13. By and by the woman's time arrived and her pains came on. 14. So Raho sent word round and his people gathered together. 15. But the birth-pains continued until night, 16. and it was not until [the next morning], just at sunrise. that the woman was delivered. The baby was a girl. 17. And then the baby rolled as far as the doorway that faced the east, and immediately sat up, 18. and called out to her father, "Raho!"

19. The man asked [what the child wanted], 20. to which [she] replied that she was hungry.

21 Raho then told his people to bring some food, 22. and they brought what had been cooked for the child, namely a hand of bananas and a pig. 23. So they got things ready and fed the child. 24. But her mother was still having pains. 25. And as soon as the child had finished eating, she got up and went out to play, saying to Raho, 26. "I am going, Raho; and note that my name is Nujmaga."

27. By and by, as the day wore on, and the sun was on the point of setting, the woman gave birth to another baby girl. 28. And the baby at once called out Raho's name, adding that she was hungry. 29. So Raho told those who were attending to the cooking to bring some food, and the people brought another hand of bananas and a pig wherewith to feed the child. 30. And no sooner had the child finished eating than she got up to go out to plan, saying to Raho, 31. "My name is Nujka'u."

32. The two children also gave orders to Raho to refrain from calling them. 33. If, however, a day should arrive when he should have a special task to be performed for him, then (but not 'till then) he was to call them.

34. Now Raho's second* sister (Mamahioväre) had no children.

Notes to text: [12]
  34. Lit., middle.

A number of important features of Rotuman cultural logic are foreshadowed in these opening paragraphs. To begin with, some basic elements of the digital code are introduced, including male/female, east/west, [13] sunrise/sunset and chief/people. The importance of bananas and pig as symbolic items is also established. It should be noted that all of the introductory characters, with the exception of Raho, are female, and that the eldest and youngest sisters are pure 'rulers' and thus surpass Raho in political power. The importance of women for political structuring is thus unequivocally set in these first few lines. Significantly, the eldest and youngest siblings are rulers. The middle sister is excluded from mention in this capacity and, in addition, her impotence is underlined by a specific statement (line 34) to the effect that she had no children. This would seem to mark eldest and youngest siblings (at least of the same sex) as of special cultural relevance.

Perhaps most fundamental is the concern that is introduced for fertility, and the involvement of spirits in it. This goes beyond the contrast between the fertile elder sister and the barren middle one; it is implicit in the types of food prepared for the birth-feast. Both pigs and bananas (particularly the red variety specified in the text) are sacrificial foods presented to the gods at ritual presentations. [14] They are foods used to feed the gods, in exchange for which the gods are expected to provide prosperity, including fertility of land and people. The names of the characters in the story are of interest in this regard insofar as they signal a preoccupation with food and eating. The word mama, which is the common root in each of Raho's sisters' names, as a verb translates as either 'to chew' or 'to cover a native oven with leaves'; as a noun it refers to chewed food' or 'the leaves used in covering a native oven' (Churchward 1940:258). [15] Likewise, the word nuju, which is the common root of the twins' names, translates as 'mouth'. [16]

The central message of this introductory segment, however, concerns the undifferentiated nature of Rotuma's beginnings, and is embedded in a code based on kinship relations. Thus, we start with only Raho and his sisters--a set of siblings, all of the same substance (i.e., parentage). [17] This undifferentiated condition is underscored by the circumstances surrounding Mamaere's pregnancy. On the one hand her conceiving is associated with the sun; on the other there is a strong hint of incest between Raho and his sister. [18] In either instance, there is no legitimate husband-wife relationship represented, and it is the husband-wife relationship that is the essence of persons of different substance uniting.

Incest is therefore, by cultural logic, prototypical of undifferentiated nature reproducing itself. The results of the union further dramatise this condition, for the offspring are twins of the same sex, the epitome of sameness in the idiom of kinship. They are even more like one another than their parents' sibling group. The paranormal circumstances surrounding their production is reinforced by the fact that twins are a culturally anomalous category to whom supernatural abilities are attributed. Indeed, female twins are central characters in Rotuman mythology, and are known as han lep he rua 'women' + 'sandy projection of land into the sea' + diminutive + 'two'. They often appear in the form of birds and perform guidance or locative functions that connect one place with another. As this and their name suggests, they are mediators. Their very births, in relation to one another, symbolically mediate between sunrise and sunset, between east and west, between night and day, between indoors and outdoors. As actors, as we shall see, the twins mediate between sky and earth, sea and land, or more generically, between spirits (who dwell beneath or beyond the sea or in the sky) and humans (who dwell on the land). In every respect, then, Nujmaga and Nujka'u represent the principle of merging, of unification. In the beginning, this segment of the myth decisively communicates, the world was a unity, constructed of the same substance.

To continue the narrative:

35. But Raho had a daughter whose name was Vamarasi, 36. who was married to a high chief in Samoa named Tu'toga. 37. Tu'toga, moreover, had a Samoan wife [as well], 38. And the Samoan wife became pregnant first, and was approaching the time of her confinement, before the fact that Vamarasi [also] was with child became noticeable. 39. And the Samoans started to make preparations for the feast that would be held in honour of the Samoan- woman's baby, without considering Vamarasi's baby. 40. Raho did not like this--the Samoans preparing a birth-feast for the baby of their own kinswoman, while neglecting the baby of Vamarasi.

41. And so Raho made ready a present, and then sent for his two children (Nujmaga and Nujka'u). 42. After a while these two girls came and asked Raho what it was that he wanted. 43. To which Raho replied that he wanted Vamarasi's baby to be born before the Samoan woman's. 44. "Unfortunately the woman is approaching the time of her confinement," said the two girls, "whereas the fact that Vamarasi is with child has only just become [apparent]."

45. But Raho still wanted Vamarasi to be delivered before the Samoan woman. 46. So [finally] the two girls said, "it is a prodigious thing that is about to happen here in Savai'i--this change that you are going to bring about."

47. So when the Samoan woman's birth-pains began, the* two girls went to her at once, and pressed on the feet of the [unborn] child, 48. so that the child turned round and the woman's birth-pains ceased. 49. The two girls then went immediately to Vamarasi, and pummelled her abdomen to bring on the birth, 50. keeping at it until the woman succeeded in giving birth [to the baby]. 51. The result was that the feast which the Samoans had got ready was given to the baby of Vamarasi, 52. and was made the feast of the first-born. 53. Now Vamarasi's baby was a girl, her name being Maiva.

54. When the feast was over, the Samoan woman's pains came on again. 55. And after a while she gave birth to a boy, 56. to whom they gave the name Fumaru.

Notes to text:
  39. Lit., their.
  47. Lit., these. And similarly in many other places.

In this segment, the processes of differentiation come into play that will result eventually in Rotuma's separation from Samoa. Note, however, that the purity of the initial generation is preserved--no wife is attributed to Raho, and his daughter is implicitly wholly of his substance. But this daughter, Vamarasi, has a husband, and so a new and different substance is introduced. The process of differentiation is given impetus in two precise ways. First, Vamarasi's husband's name, Tu'Tonga (Tu'i Tonga) translates as 'King of Tonga', [19] setting up a contrast between Tonga and Samoa, which is represented by her husband's other wife. A full examination of Rotuman myth strongly suggests that Tonga, as a concept in Rotuman thought, represents the male principles of potency and vitality, while Samoa represents the female principles of fecundity and domesticity. As we shall see, it is in the meshing of these complementary principles that the legitimation of authority resides. Second, Vamarasi's child, Maiva, has a half-sibling of the opposite sex. Paternal half-siblings are, par excellence within the idiom of kinship, representative of entities that are the same, but different. They both share substance and are of different substance. Socially they are traditionally portrayed as rivals, and are therefore ideal symbols for a pivotal point from which differentiation occurs. Raho's concern that his grandchild be born first reflects this competitive aspect, as well as the significance of the common Polynesian preoccupation with genealogical precedence. So we have in this segment a structure that sets the stage for differentiation--a symbolic shift from the unity of twins of the same sex to half-siblings of opposite sex.

The next segment describes the incident leading to Raho's decision to leave Samoa:

57. As time went on these two children grew up, 58. and one day they* went to the beach to play, 59. and began fishing for penus named Tua'nakvalu. 60. And as they continued fishing for penus, Maiva caught a red penu* named Tua'nakvalu, 61. which she thereupon took and put into a vessel of water. 62. By and by Fumaru came and found the penu in the vessel of water, 63. and picked it up and surreptitiously put into his mouth the penu that belonged to his sister. 64. Afterwards Maiva came back, and found that her penu had been taken. 65. So she went along and told Fumaru to drop her penu out of his mouth. 66. But the boy refused to do so. 67. So then Maiva went to her grandfather (Raho), crying, and telling [him] what her brother had done to her. 68. Raho then pleaded with his granddaughter, but she would not relent.

69. Raho then sent for the twins* again, 70. and they came, and Raho told them what had happened to his granddaughter; 71. and [he said that] he wanted to make a home for his granddaughter, which should be far away from Samoa.

Notes to text:
  58. Lit., this brother-and-sister.
  60. A very small variety of crab.
  69. Lit., the two sandy-point (lepi) girls (hani).

Both the beach and the penu 'a variety of crab' are intermediate categories, the beach between land and sea, the crab between land food and sea food. Fumaru, Maiva's half-brother, takes her catch and symbolically consumes it. In doing so he not only challenges the legitimacy of their birth order (older siblings have license to take objects from younger siblings, but not vice versa), he also deprives her of food that she has produced. This incident sets the stage for a theme of differentiation leading to the eventual branching off of the Rotuman polity. From half-siblings on a beach catching crabs (all intermediate, hence merged, categories) emerges a usurping male and a dispossessed female figure. Their alienation foreshadows a distinction between two sets of association that form an integral part of the digital code: male:chief:sea:: female:commoner:land.

The impetus towards the founding of Rotuma is here rooted in an issue of chiefly prerogative (represented by Fumaru, a male) versus commoner rights (represented by Maiva, a female), particularly with regard to food. [20] This is an issue that pervades Rotuman myth in one form or another (see, for example, Howard in press). Several variations of this legendary incident have been recorded in other texts, but they are structurally consistent with each other. The incident is always located on a beach and the disputed item is always a crab. Maiva is always initially partially merged with her antagonist (sometimes male, sometimes female) as a close relative or playmate. In some versions alienation is precipitated by her antagonist challenging the legitimacy of Maiva's heritage (for example, in one account her rival calls her a foreigner without claims on the family [Russell 1942:240]). In other versions Maiva's anger is piqued by insults directed at her deformed foot, [21] suggesting a stigmatised status. The movement from a relatively (but not entirely) undifferentiated to a differentiated state is therefore at the heart of the myth. This differentiation is symbolised strongly by the geographical separation that follows.

72. Thereupon the twins* filled two baskets with earth--a presentation basket and an ordinary basket. 73. The name of the presentation basket was Fuarei, while the name of the ordinary basket was Fua'a. 74. The twins then put these two baskets on board a canoe of aftea wood, 75. and they, together with Raho and his household, got into the canoe and came to found this island of Rotuma.

76. Now it is said that when Raho came to found this island many high chiefs in Tonga and Samoa heard about it. 77. And so, when Raho and his company left, a chief named Tokainiua (it is not known whether he was a Tongan or Samoan), accompanied by a number of others, went after Raho.

78. [By] and [by] Raho with his company came 79. and found in the midst of the ocean a rock of great size, 80. the two extremities of which were well above the water, while the middle was just awash. 81. So the twins emptied out the presentation basket of earth on to the rock, 82. [thus] forming an island.

83. This done, the twins left Raho and his company behind on the island, 84. and took the [other] basket of earth and flew off [with it] towards Futuna. 85. On and on the two girls flew till they got there, 86. and then they emptied out the basket of earth and formed the island known as 'Arofi.

Notes to text:
  72. Lit., these two girls.

The key symbols in this segment are the baskets, which affirm the differentiation of chiefs and commoners previously described. A la agai 'presentation' basket is used for presenting food to chiefs and is a common metonym for chieftainship. It also symbolises chiefly rights to food. A common basket stands in metonymic relationship to commoner status, and symbolises the people of the land (i.e., those who produce food from the land). In this version of the tale Rotuma is formed from the contents of the chiefly basket, and is differentiated as an entity from geographically remote 'Arofi (Alofi), which is presumably common. [22] A more compacted and structurally more significant version has the presentation basket tipped out in the district of Malhaha, the common basket tipped out in Faguta (see Churchward 1937:109). This conforms to a north-south opposition used to codify chief-commoner relations in other narratives (see footnote 12 and Howard in press).

There follows an incident that echoes Maiva's dispute with Fumaru, involving Raho and Tokainiua.

87. The twins then came back, and found that Raho and his people were still here, 88. and they suggested to Raho that he should mark the island as his, 89. in case another person should come later on and a dispute should arise. 90. And so Raho marked the* island as his by means of a green coconut-leaf tied round the fesi tree at Vakpäre, 91. requesting the twins to go to Tonga to bring him some kava.

92. But, as soon as the twins had departed, Tokainiua and his company sighted this island, and thereupon directed their [canoe] towards it. 93. They* landed at Oinafa. 34 By and by Tokainiua came to Malhaha, and discovered Raho's coconut-leaf tied round the fesi tree at Vakpäre, and [noticed that] it was still green. 95. Thereupon he resorted to a strategem: 96. he fetched a coconut-leaf that was already dry, and tied that round the tree to mark the island as his.*

97. After a while Raho came and found Tokainiua standing by the fesi tree, having marked the island as his by means of a dry coconut-leaf. 98. And so the two men began arguing. 99. Raho said it was his land, 100. while Tokainiua said it was his. 101. Raho said it was he that had formed the land; 102. but Tokainiua maintained that the land was his-- 103. his coconut-leaf had been fixed round the tree for a long time, while that of Raho [as shown by the fact that it was still green] had been put on quite recently.

Notes to text:
  90. Lit., this.
  93. Lit., the travellers.
  96. Lit., and made down as his fapui. A fapui is something set up to mark a   tree or a plantation etc. as forbidden to others. Usually it consists of a   coconut-leaf tied round the trunk of the tree.

Just as Maiva found the penu, then had it taken away from her by guile, Raho founds (literally, 'plants') Rotuma only to be tricked out of his rightful claim by Tokainiua. This completes the inversion of Raho's initial trickery, i.e., his use of the twins to alter the birth order of Maiva and Fumaru. Raho creates Maiva's precedence in Samoa through guile, then has his own precedence in Rotuma usurped in like manner. In Trouillet's account this inversion occurs in a stronger form inasmuch as the very twins who do Raho's bidding in the first instance turn against him in his dispute with Tokainiua. The Trouillet version also marks the association of Raho with the land and Tokainiua with the sea in a more direct manner. It reads as follows:

Tokaniua accosts Rao, saying to him: This country, to whom does it belong?--It is my country, answers Rao.--But where are your subjects? says Tokaniua.--They are in the interior, responds Rao.--But, says Rao in his turn, where are your subjects? --they are on the seashore, replies Tokaniua. Let us go see, says Rao, and together they go around Rotuma. Rao notices that indeed the country is inhabited and upon their return to Oinafa the quarrel becomes livelier.

Rao tries at first to embarrass Tokaniua. He goes down to the sea, brings back an immense basket of sand which he spreads on a mat and tells Tokaniua to count the grains. Tokaniua accepts the challenge and right then pulls from his breast two small serpents which he had brought with him: One of them sprawls in the sand and the other counts the grains; the one who counted the grains first then sprawls in his turn and the other counts the grains, and so it goes until the contents of the basket had been counted entirely. Tokaniua gives an account to Rao who had nothing to say. From that moment on the two Leprua women, displeased by Rao's conduct and by his lack of success, abandon him and even help Tokaniua to embarrass Rao; they advise him to tell Rao to count the waves of the sea which constantly come onto the rocks which are called Vos. Tokaniua follows this advice and Rao accepts; he therefore goes to the seashore, he counts one full day and one whole night, but the waves keep succeeding each other; at last he is tired out and in confusion he flees; his foot is caught in the serpent who is called Kine, he falls down, gets up and full of shame he escapes to Atana [Hatana] (Sumi Mission Station, Rotuma manuscript).

Trouillet's version of the encounter accentuates the importance of women, in the form of the twins, for effecting a transfer of power. The twins' infidelity to Raho and their complicity in Tokainiua's deception is instrumental in bringing about the success of the latter's attempted usurpation. The central role of women in rebellions against established authority is a pervasive theme in Rotuman myth. They appear as victim provocateurs, as mediators with the supernatural and as leaders of rebellion (see Howard in press) . It seems that, while chiefly authority is considered to be essentially a male prerogative in Rotuman thought, setting up the male:chief::female :commoner opposition in the digital code, females are instrumental for increasing or decreasing chiefly potency in the analogic code.

But the story does not end here, with Tokainiua's apparent triumph. From a logical standpoint the nature of legitimate authority remains problematic, for Tokainiua's putative claim to precedence is patently false. His true basis for assuming priority is his mana--in the fact of his triumph. This sets up the central problem of the myth, namely, how is legitimate authority to be constituted given the antagonistic claims of the people of the land (based on their priority as planters and food producers) and chiefs (based on genealogical priority and efficacy in contests)? The Rotuman solution is to treat these principles as complementary, and to emphasise constraint of both. This is the theme of the final segment of Churchward's narrative.

104. Upon this [Tokainiua's successful challenge] Raho became angry and struck Tokainiua. 105. But the Sa'aitu* came and held Raho back, and covered Tokainiua over at the foot of the fesi tree, 106. and Raho did not see him again.

107. Raho then took it into his head to go and break up the island, 108. so that Tokainiua should not have it. 109. So Raho went along to the western end of the island, 110. and took a digging-stick, 111. and drove it into the ground 112. and levered up the point, 113. and [lo and behold the small islands of] Uea, Hatana and Hafliua sprang into being.

114. But the woman who lived in the scrub, observing that the land was about to he spoiled by Raho, 115. came running towards him, 116. and bowed herself at his feet, 117. and besought him 118. not to be angry, and not to spoil the land, 119. for Tokainiua had told a lie, the land being really Raho's.

120. "That being so," replied Raho, "I will do as you request."*

121. With that, he pulled his digging-stick out of the ground, 122. put it on his shoulder, 123. and returned to Malhaha.*

124. [In doing this] Raho came [first] to Motusa. 125. He then followed the inland road, 126. going on until he reached the country behind the houses at Vai. 127. There he let down his digging-stick, 128. and dragged it towards the coast, 129. and the place where he dragged the stick along became a watercourse, 130. the name of which is the Watercourse of 'Alüstägtäge.

131. Raho then went down to the beach, but the kava was not there. 132. Now the twins had arrived at Tonga, 133. and had sent the kava plant 134. which had then come [over the water] alone. 135. But on arriving here, it had learned that Raho, in a fit of anger, had gone to spoil the island, 136. and so the kava plant had left Valta* and had gone inland, 137. and had made its way to the queen at Fag'uta.

138. After a while the twins came back, but Raho had not yet had any kava to drink. 139. So Raho sent them again, 140. and once more they returned to Tonga, 141. and brought some dry kava, 142. wrapped in palm-leaves. 143. Raho's kava was then prepared on top of the Kamea stone. 144. And the bowl-like hollow* is there even now on top of the rock, 145. and there [nearby] is the spring [that supplied the water] with which the kava was mixed.

146. And after drinking the kava,* then it was that Raho and his women-folk went to [live at] Hatana.*

Notes to text:
  105. A certain class of supernatural beings.
  120. Lit., and then Raho said, "Only that, and it-is-good."
  123. Lit., and then his return to Malhaha.
  136. A part of the village of Pa'olo, in the Oinafa district.
  144. Lit., the kava-bowl (tano'a).
  146. Lit., at the finishing of the kava.
  146. (end). See the next story.

Whereas the emphasis in the early part of the myth is upon differentiation of people from chiefs, in the final segment a reintegration, rather weak in form, takes place. Thus, Tokainiua, the usurping warrior from overseas, is covered with earth at the foot of a fesi tree (a metonym for chieftainship). He "takes root" in the land and so is constrained, and symbolically made indigenous. His powers are thus domesticated. On the other side of the coin, Raho, after his humiliation, reasserts his rights to the land through shaping its features. His rage is assuaged by hanit e ma'us (translated by Churchward as 'the woman who lived in the scrub', but more appropriately 'the wild-woman', or 'spirit-woman' of the bush). She is a well-known character in Rotuman myth, and combines female domesticity with supernatural danger, particularly as a succubus. Her assurance to Raho that the land is really his, despite the success of Tokainiua's challenge re-establishes his legitimacy, and with it, the rights of the hanuet 'people of the land'. Nevertheless, Raho leaves the main island and goes to dwell on the islet of Hatana, off the western end of Rotuma. His claim is thus constrained by the sea between Hatana and Rotuma, just as Tokainiua's claim is constrained by the land which covers him.

This incipient complementarity is signalled in another way in the text. Note that, whereas Tokainiua gains ascendancy through the symbolism of a dry, withered coconut frond from Rotuma, Raho gains symbolic sanctity through the medium of dried kava from Tonga. [23] In combination these codes serve to provide a mythical foundation for a complementary system of rights and obligations between chiefs and commoners. The conceptual separation of chief and commoner is retained, however, and even strengthened symbolically, by locating Tokainiua on the eastern end of the island (the chiefly side) while Raho is placed at the extreme western end of the Rotuman world (the commoner side).

We shall now consider a second myth related by Churchward, which he calls Sau Mumua 'e Rotuam 'i 'The First Kings in Rotuma'. Although the myth is presented as a separate story, Raho is located at Hatana and Tokainiua at the fesi tree in Malhaha, their positions at the end of the founding myth. In fact, the narrative amplifies the theme of the initial story and further develops the cultural logic that lies behind Rotuman conceptions of chieftainship. The story begins in the sky.

1. It is said that there was a country in the sky, 2. and that there was a king in that country named Tü'rotoma, the mua being named Tü'feua.* 3. And when these two men saw Rotuma down below, 4. they thought they would like to send somebody down to see whether it was a good land or a bad one.

5. And so the king chose a man from [the people of] his country, to go down to see what this country was like. 6. It is said that the name of the man was Titofo. 7. And so they lowered Titofo, 8. and he arrived down here below, 9. happening to alight at a place at Pephaua named Faufono, 10. where there was a tupu'a named Toväe.* 11. So Titofo took up his abode with this tupu'a at Faufono 12. while he went and looked at the various parts of this country, to see what sort [of a place] it was. 13. And as he continued his investigations, he found that this country was quite a good place, not a place to be afraid of.* 14. Finally, therefore, Titofo returned to the sky to tell the king that the country was a very good one.

15. When Titofo arrived at his destination, he said to the king, 16. "The country, sir, is a splendid country."

17. So the king then gave his son, Fagatriroa, 18. while the mua gave one of his daughters, Päreagsau by name, 19. that the two of them should come down here below 20. to take care of this country. 21. The king appointed also two men, to come down with these two young folk, 22. to remain and to look after them. 23. And so the four of them came down [from the sky], and dwelt at Pephaua. 24. Of these two men, the name of one was Moeauita, while that of the other was Orivai.

25. They remained for a long time, and then Päreagsau became pregnant by Fagatriroa, 26. And when the two men observed that the woman was with child, 27. they were angry, 28. and they returned to the sky, 29. leaving Päreagsau and Fagatriroa here below. 30. On arrival [in the sky], they told the king and the mua what their two children had done, 31. and [that] they did not approve of it 32. and [so] had left them down below. 33. But the king answered the two men, saying, 34. "Don't be angry, 35. for that is the very thing that we sent them down to do, 36. so that the country should be populated by their children."

Notes to text:
  2. In olden times the mua was a chief next in rank to the sau or king. Neither   office exists now.
 10. In Rotuman mythology, a tupu'a appears to be a kind of immortal man.   Nowadays its principal meaning is image, statue, doll, or idol. It also means   heavenly body, star, or constellation.
  13. Lit., a cruel place.

Perhaps the most striking aspect of the opening segment of this myth is the way in which it inverts significant aspects of the previous story. In this instance it is men rather than women who rule in the place of origin, and it is two men rather than two women who act as mediators. Instead of facilitating a birth they are made angry by its prospect. More central is the inversion of the differentiation theme. In this case differentiation is a feature of the polity from the beginning. Rule is divided in the sky between a sau and mua, Rotuma already exists as a distinct entity, and the first child is the offspring of an unrelated couple (of different substance). Also contained in this segment is a clue to the conceptual relationship between sau and mua. The sau provides a son and the mua a daughter in the interest of fertility. It appears, then, and other data strongly support such a conclusion, that the sau represents the male principle of vitality while the mua represents the female principle of fecundity. They thus represent a complementarity parallel to that symbolised by Tokainiua and Raho, and indeed it seems clear that Raho:Tokainiua::mua:sau. The plausibility of this interpretation is strengthened by the fact that mua means 'first', and Raho was the first to 'plant' Rotuma. The story continues:

37. And so the two men came back once again to the earth, 38. to look after Fagatriroa and his wife.* 39. And in the course of time the woman gave birth to her child: it was a boy. 40. and [they] called his name Muasio. 41. By and by the woman became pregnant again, 42. and gave birth to another boy, 43. whom [they] called Seamrefäega. 44. Now [the births of] these first two children were not reported to Raho at Hatana;* 45. but, as time went on, and the third child was born, 46. the two men proceeded to Hatana to tell Raho about it. 47. Upon arrival they said to Raho, 48. "Fagatriroa and Päreagsau have a baby boy--he has just been born--49. and so we have come to report the matter to you, 50. so that you may be kind enough to say what is to be done about it."

51. Raho's reply to the two men was: 52. "I know all about it: 53. there were two children born before, and you did not tell me. 54. However, go back, and name the child Tu'iterotuma;* 55. he is to be the king of the country, 56. and a courtyard is to be cleared for him* at Halafa so as to be near me. 57. And the name of the courtyard is to be Mariki. "

Notes to text:
  38. Lit., this married couple.
  44. Cf. "The founding of Rotuma" 146. Hatana is a small island off the   western end of Rotuma.
  54. Meaning the (-t) king (tu'i) in (e) Rotuma. In Rotuman the word tu'i   (king) seems to be used only in compounds.
  56. Lit., his courtyard is to be swept.

Again we have an inversion. Instead of three female siblings in the parental generation, we have here three male siblings in the generation of offspring. But we also have a parallel with the first myth insofar as here, too, Raho inverts the order of precedence, granting to the last conceived child priority over siblings conceived previously. The implication that I draw from this is that legitimate authority derives only secondarily from chiefly pedigree. Its primary source of legitimation rests upon the consent of the people, as symbolised by Raho. It is the prerogative of the people, ultimately, to decide upon which eligible contender is to be elevated. This message is underscored by Raho's locating the selected child at Halafa, which is on the extreme western end of Rotuma, opposite Hatana. A curious incident follows:

58. So, as soon as the two men had arrived back from Hatana, 59. they conveyed Raho's decision to Fagatriroa and his wife, 60. and immediately went [up] to the sky to give an account of how they were all getting on, 61. and to get [a] pig with which to prepare a feast of cooked food to take to Hatana. 62. And as soon as they arrived in the sky, they had a talk with the king, 63. and finally the king gave them a pig (64. a boar of no mean size* it is said to have been), 65. and they brought it down to prepare the feast.

66. But as the two men were carrying their pig, and had not yet arrived at their destination, 67. they met Seamrefäega. 68. And the man took the pig from them by force, killed it, and put it into the oven to roast. 69. And when the oven was opened up, 70. Seamrefäega cut the pig in halves across the centre, 71. and said to the two men, 72. "You are to take the fore part to Raho at Hatana, 73. but I am going to have this hinder part myself."

74. And so the two men proceeded to take the fore part [of the pig] to Raho at Hatana. 75. But when they arrived with it, 76. Raho said to them, 77. "Haven't I told you 78. that a partially eaten thing is never to be brought to me, 79. but that if [you] had a thing that had not been eaten at all you might bring it along? 80. Who was it that told you to bring this half-eaten thing?"

81. And then Raho, in anger, flung the half pig into the sea,-- 82. and that is the origin of the blow-hole which foams in the sea at Hatana at the present time.*

Notes to text:
  64. Lit., a boar not to-be-joked-about.
  82. There is a pun here on the word kou, which has two meanings--boar (or   other male quadruped) and blowhole.

Structurally this incident duplicates the encounter between Raho and Tokainiua in the founding myth. In this case Raho's precedence is symbolised by the gift of a pig from the sau in the sky (i.e., from the gods). The pig, however, is seized and spoiled--it is "half-eaten" by Seamrefäega, the second son of Päreagsau and Fagatriroa. Seamrefäega is associated with the sky, being a direct descendant of the sau and mua in the sky, but he is the conceptual equivalent of Tokainiua, who comes from overseas, since sea:land::sky:earth. Thus, again an outside usurper, using guile, blunts Raho's claim. But Raho is pointedly still given the forepart of the pig, signalling his ultimate priority in the now disputed claim. This conveys the same message as the assurance provided by hanit e ma'us 'the wild woman of the bush'--that the land really belongs to Raho. The final equivalence is more direct. As before, Raho expresses his anger by altering geographical features, again symbolically recreating the land. The fact that in both instances the features shaped by Raho include land and water suggests a conceptual reintegration of these complementary components of structure. A blowhole is perhaps the ideal symbol of the dynamics of relationship between these components. The sea rushes in, penetrates the land, spouts up towards the sky, washes down to overwhelm the land, then recedes only to repeat the process. And so it is with chiefs. They come into office with great vitality, ascend to heights of virility and sanctity, overwhelming the people of the land; they then decline and either die or are deposed by a vigorous new chief, and the process repeats. The sexual symbolism of penetration adds another dimension to the metaphor, since chiefs are the symbolic inseminators of the land, bringing fertility and prosperity to the people. Or so the idealised conception goes.

Next comes a segment involving the death of Tu'iterotuma, the first earthly sau.

83. So the two men returned from Hatana, 84. and then proceeded to carry out Raho's instructions regarding Tu'iterotutma's being made king. 85. And, gathering the people together, 86. they went to Halafa, 87. and cleared a courtyard for the king at Mariki, 88. and made that the king's place of abode. 89. And then the king was brought to Halafa to live, so as to be near Raho at Hatana.

90. A long period elapsed, and then the king was taken ill, 91. and before long he died. 92. Thereupon the two men went to Hatana, and told Raho that the king was dead. 93. Raho told them to go back, 94. and to have a bier made,* and to place the [dead] king thereon. 95. The people were then to support [the bier] on their shoulders, 96. and to carry it across country, 97. while he would send two birds to go in front of the bearers [to show the way].

98. So the two men returned to Halafa, 99. and the people made a bier, 100. and placed the dead king thereon, 101. and began carrying it across country, 102. when, lo and behold, the two birds that Raho had sent came flying along,-- 103. the name of the one being Manteifi, that of the other Manteafa. 104. So the two birds flew on ahead, 105. while the bearers walked along behind them. 106. On and on they went until they reached a spot inland from Lopta* in the region of Muasolo, 107. when [they noticed that] the two birds* acted as if they were about to alight. 108. the bearers then stopped and looked, 109. and [they saw that] the two birds did not actually settle, but just flew on. 110. Moeautia and Orivai thereupon told them to put the [dead] king down, 111. for that was what Raho had told them to look out for: 112. [he had said] that when the two birds acted as if they were about to settle in a certain spot, 113. that was the spot where Tu'iterotuma's grave was to be dug.

114. Accordingly, they put the corpse down, and there they dug the grave, 115. after which they buried their king [there] at Muasolo. 116. Raho had said, moreover, 117. "The place where Tu'iterotuma is to be buried, that is the place which will produce abundant supplies of food for this country." 118. And that was the first cemetery here in Rotuma, namely the cemetery in which the first person to be buried was Tu'iterotuma, at Muasolo, a little way inland* from Huo (Lopta).

119. After that one of Tu'iterotuma's two brothers became king 120. which of them [we] do not know.

Notes to text:
   94. Lit., that they (pl., not dual) should bind (fa'u) a bier.
  106. Lit., the-back-of-the-houses at Lopta.
  107. Lit., these two little (he) things (te).
  118. See note on 106.

A proper interpretation of this segment requires some background knowledge concerning the importance of cemeteries, and particularly the burial places of sau and mua, in traditional Rotuman culture. Gordon MacGregor, an anthropologist who visited Rotuma in 1932, made the following observations:

The Rotuman graves form the most fascinating side of the study of their culture. The island has been described not inaccurately as "one great cemetery." Certainly graves are to be found everywhere, under house sites, alongside most of the roadways, in great village cemeteries now preserved by European law, on top of little islands along the reefs, and throughout the bush. They are monuments to a remarkable industry and devotion to the dead, qualities now sadly failing among the present inhabitants. The dead were buried in double stone vaults of great size built up of thick slabs of conglomerate rock or coral cut from the reef. Important graves had top slabs cut from a quarry of basaltic rock in the western end of the island. All these were transported overland by groups of labourers while a priest stood on top and muttered incantations and prayers to make the burden lighter. Great crafts were built too, to carry these stones longer distances down the coast. One slab of coral found on the King's cemetery high in the bush measured 17 by 7 by one and half feet. The lower vault of the grave was made of six slabs of stone in box shape set in the ground. The body was wrapped in mats and the whole was buried in sand. On top of this vault the superstructure varied according to the importance of the dead in the estimation of the family. Chiefs and family vaults for later corpses were covered by a second vault which rested on the ground level. Some had merely a capstone or an upright monolith as markers (MacGregor n.d.).

The burial places of the sau and mua were the focus of ritual attention during the annual cycle. One of the important feasts in the cycle took place at Muasolo (in Oinafa), the burial place of the mua. Mua were interred in a special house built for them there. At the feast the burial house was rethatched and the old thatch was distributed, presumably to ensure the possessors a fruitful season. After this, kava was prepared and an entire bowl was poured out to the dead mua. A description of the event is provided by Gardiner:

A great quantity of food is then placed in the [burial] house, as this feast differed from all others in that no food could be carried away from it. The mua alone can enter the house, and so has to carry all the food in. The old people, both men and women, while he is doing so, walk in procession round the house, while a prayer for a fruitful season is chanted, each fruit being mentioned by name.

Te moiea naragosou, mua ...

Be fruitful, mighty spirit, mua.

E te moiea favorou' mua ...

Be fruitful to the fava tree, mua

Te moiea se, oh, oh, oh ...

Be fruitful to us, oh, oh, oh.

Moiea ifi ma moiea fava ...

A fruitful ifi and a fruitful fava

Te moiea se, oh, oh, oh, etc.,etc ...

Be fruitful to us, oh, oh, oh, etc., etc.

Se le mua le; sol, oh, oh, oh


(Gardiner 1898:465). [24]

Another major feast was held on the top of the hill at Sisilo (in Noatau) where the sau were buried. Kava was poured on the graves of the various sau, and the living sau would also drink kava, then eat of the different grasses on the hill. The sau's graves, rather than being in a house, were marked by stone tombs and were meticulously maintained. Lesson described the site as containing about 20 tombs in 1824. At the head of each tomb rose an eight-foot stone slab, with a four-foot stone at the foot and two long stones on either side. The area was surrounded with a grove of trees that had been planted with care (Lesson 1838-9, II:437).

It is evident from these accounts, and from a wealth of additional information, that the mua and sau not only were primary links to supernatural spirits who brought good or ill fortune to the island, but that they themselves were also transformed into powerful spirits at death. In fact, it was in death, as spirits, that they were most able to perform their major function of ensuring prosperity.

If we examine the foregoing segment of myth against this background several points stand out. One is that the dead "king" is carried on the shoulders of "the people" across Rotuma from the western to the eastern end of the island. The metaphor is straight-forward enough--it is the people who are responsible for elevating chiefs, for facilitating their sanctification. They are the very foundation of chieftainship. This underlying theme is nicely underscored in the phrasing used by Raho concerning the preparation of the funeral bier. He uses the word fa'u 'to bind'. The word is also used in reference to 'space or place further from the sea, further inland', and for the western end of Rotuma. As a verb it translates as 'to follow, to go behind or after; to back up, support' (Churchward 1940:198-9). The chief is bound to the people of the land, who are his followers and supporters.

The incident described in this segment resonates with two previous events involving Raho. On the one hand, it recalls the founding expedition, also guided by the twins. In the former case, however, the journey was from east to west, resulting in the formation of the land; in this case the voyage is from west to east, establishing the sanctity of chieftainship. On the other hand, it reinforces the message conveyed by Raho's selection of Tu'iterotuma as king. Thus, here again it is Raho who arranges for Tu'iterotuma's elevation. The message that it is the people who are the ultimate source of chiefly authority is redundantly communicated.

The statement that the cemetery is the place that will provide abundant supplies of food reflects Rotuman conceptions of the role of supernatural spirits in bringing prosperity. Their belief that in death chiefs are deified--that they become powerful spirits capable of bringing good or ill fortune to the land--is a presupposition to this statement. As a focus for the transformation from human being to powerful spirit the cemeteries of the mua and sau were the most sacred places for the Rotumans.

Although Tu'iterotuma is described in the text as sau, he is interred at Muasolo, burial place of the mua. This constitutes evidence for the equivalence of the two offices at some level of conceptualisation. Indeed, there are several parallels in the symbolism associated with the sau and mua, and Trouillet describes an historical sequence in which the position of mua is initially established by Raho, then is superseded several generations later when the position of sau emerges following a rebellion against the eighth mua. Overall the evidence suggests that the positions of sau and mua symbolise complementary aspects of sacred chieftainship, with the mua representing that component of authority which derives from first occupancy, traced back to Raho, and the sau representing that component of authority derived from conquest and usurpation, traced back to Tokainiua. Rotuman conceptions thus seem to be sequentially oriented, such that the initial position of mua is differentiated into mua and sau. There is linguistic support for such an interpretation inasmuch as mua means 'to be or go in front or before or first--either in place or in time' (Churchward 1940:268). In this case mua might be interpreted as 'the first sau, or 'the one who preceded the sau.' This notion of differentiation parallels a conception of chiefs emerging as persons of a different order from common Rotumans. They are of the people of the land, but are different from them (hence the common symbolism of their emerging from the sea). In the founding myth this sequence is symbolised by Tokainiua's successful challenge to Raho's sole claim. We can illustrate this structure in the following way:

            mua = sau
mua (:Raho)/sau (:Tokainiua)

 This indicates that the primary concept of mua incorporates the notion of sauship in Rotuman thought (i.e., that the mua and sau were initially one and the same, with mua the unmarked category), but that from this undifferentiated state emerged the positions of mua, associated with Raho and incorporating the principle of first occupancy, and sau, associated with Tokainiua and incorporating the principle of military vitality. The apparent anomaly of the sau being buried in the mua's cemetery thus seems to reflect the concepts in their undifferentiated state.

An interesting additional feature of this segment is the indifference as to which elder brother succeeds Tu'iterotuma. The proper order of succession (from eldest to youngest) had already been violated by the initial choice of the youngest sibling; the indifference to the birth order of his successor merely punctuates the underlying message--that approval of the people is the basic condition of legitimate authority.

We go on now to the next segment of the myth.

121. Not very long after this, Fagatriroa, Päreagsau's husband, died; 122. and they took and buried him at a spot in the interior of Malhaha, named Tagkoroa. 123. And that was the second cemetery to be opened on this island.

124. It was not so very long after this when a company of voyagers came from Samoa, 125. led by a man named Vilo. 126. And it is said that one of the men from this company went ashore and took up his abode with Raho at Hatana. 127. This man's name was Fuanofo. 128. After a while Fuanofo took a fancy to Päreagsau, the widow of Fagatriroa 129. Accordingly, Fuanofo and Päreagsau were married, that they might produce children who would be the first Samoan half-castes here in Rotuma.

130. So the marriage was properly celebrated, 131. and then, after a somewhat lengthy period, Päreagsau became pregnant. 132. And when Päreagsau's child was born--a boy-- 133. the name by which [they] called him was Takalhöl'aki. 134. Time went on, and this couple had another child, a boy [as before], 135. and [they] named him Tukmasui. 136. Then, later on, they had still another child, a boy [once again], 137. to whom they gave the name Muamea.

138. Now it is said that when, in the course of time, both of Tu'iterotuma's brothers died, 139. then Takalhöl'aki, the child of Päreagsau by Fuanofo, became king. 140. And when Takalhöl'aki died, his younger brother Tukmasui succeeded him. 141. And when Tukmasui died, then Muamea was [made] king in his stead.

142. It is said, further, that during the time when Tukmasui was king the people of Noa'tau * equipped an army for the purpose of going to kill the king. 143. The army then went off, and fought, 144. but the king's army gained the victory, 145. and the Noa'tau army returned home without killing the king 146. And it is said that that was the first war that ever took place on this island.

Notes to text:
  142. Lit., this Noa'tau: the story being related at Noa'tau.

Fagatriroa's death opens the way for a second set of brothers to assume the sauship, this time in the appropriate order of priority, from eldest to youngest. As father to the sau, Fagatriroa is also a sacred personage and his burial is of considerable importance. The theme relating cemeteries to material abundance here seems to be encoded in the name of the burial ground; tag 'to move convulsively' + koroa 'goods, wealth, possessions, riches', suggesting the gorging up of plenitude.

The identification of Päreagsau's second husband as Samoan, and his residential affiliation with Raho, exemplifies the paradox referred to above--that the chiefs are of the people, but are different from them. Fuanofo, like Raho, is from Samoa. In this sense he is identified with the people of the land; but he is also from overseas, hence not from the land. Of course, Raho also presents the same anomaly, and it is in this light that we can understand the symbolism involved in their residence on the offshore islet of Hatana, for an islet is both of an island (the land) and different from it.

The paradox is a central one that pervades Rotuman myth. It is the source of oscillations in the narratives between indigenously conceived chiefs and those from elsewhere. The designation of Fuanofo's children with Päreagsau, all of whom become sau, as half-caste Samoans, symbolises very effectively the anomalous position of chiefs.

It is noteworthy that the first war is described as an insurrection against the sau by the people of Noa'tau rather than as a challenge by a rival chief. In contrast with the mythology of other Polynesian cultures (e.g., Hawai'i), Rotuman narratives play upon the problematic relationship between people and chiefs much more than on chiefly rivalry. The failure of the insurrection amounts to an assertion of the legitimacy of the sau's authority, especially since no cause is given to justify rebellion.

To continue:

147. Later on, during the reign of Muamea, a man at Noa'tau named Moea went and married a woman at Malhaha named Panai. 148. And after a while, it is said, the king developed a liking for the woman, 149. and spoke to her, suggesting that she, Panai, should leave Moea and marry him [instead]. 150. Thereupon Panai left her husband and lived in adultery with Muamea. 151. And it is said that that was the first case of adultery here in Rotuma.

152. Moea then came to Noa'tau, 153. weeping, and telling his people* what had befallen him at Malhaha. 154. He was very sore over what Muamea had done, 155. and he loved his wife too, 156. but he would not be able to get her again, seeing that she preferred the king.

157. On hearing this, Hanfakiu his sister said, 158. "Don't cry! it's all right! stay where you are, 159. and I will accomplish what you desire. 160. You are a man, and yet you cry like a child."*

161. Now what the woman proposed was that Noa'tau should go to war with Malhaha, 162. with a view to killing the king (Muamea). 163. And so she* went to her house, and strangled herself, and so died. 164. Having died, she then proceeded to Malhaha, 165. the person whom she was going to see being an 'atua* at Malhaha named Penua. 166. On she went until she came in sight of Penua's home, where she found Penua sitting. 167. Penua at once turned round to see Hanfakiu approaching, and noticed what a sight she looked. 168-169. "Good gracious, Hanfakiu," she said, "how terrible you look! 170. your eyes are all bloodshot. 171. and your tongue is hanging out helplessly."

172. "I have come," said Hanfakiu, "to get something done. 173. And I want you to be kind enough to help me to carry it out."

174. Penua asked what it was that she wanted done; 175. to which Hanfakiu replied, 176. "The fact is that I want the king to be killed, to avenge my brother."

177. Penua then said, 178. "You go to the fesi tree at Vakpäre: 179. for Tokainiua has been struck by Raho, and the sa'aitu have covered him over,* and he is still lying [there]. 180. So you go and look closely, 181. and when you see one of his big toes, 182. make a grab at it, 183. grip it tightly, 184. and pull him up, with a sudden jerk, into a standing position. 185. If you succeed in doing this to the man, your desire will be fulfilled."

186. So the woman went straight to the spot indicated to her by Penua, to find the earth heaped up at the foot of the fesi tree. 187. She then looked narrowly at it until she spied one of his big toes, 188. whereupon she made a grab at it, 189. and grasped it tightly, 190. and gave a sudden jerk upwards, and Tokainiua stood up.

191. The woman then said to the man, 192. "Come with me to Noa'tau, and let us equip an army, that we may come and fight against Muamea and his people. 193. And if we are victorious, [the District of] Oinafa will be yours."

194. "Very well," replied the man, "let us go."

195. So the two of them came to Noa'tau, 196. and this District equipped its army, 197. and then proceeded to Malhaha to fight against Muamea and his men, 198. the leader of the [Noa'tau] army being Tokainiua .

199. When they arrived, the fighting began immediately; 200. and they fought on until the battle ended with the death of the king, Noa'tau gaining the victory. 201. Thus Noa'tau gained the right of choosing the king,* 202. and immediately on returning home they annointed Riamkau, at Sav'ea, as king. 203. Thus the kingship was now conferred on Riamkau,* 204. while Oinafa became Tokainiua's: 205. all the land from Remoa* to the stony ground between Huo and Malhaha was given to Tokainiua at the conclusion of this war.

206. And from that time onwards [it was the custom] that the kings of Rotuma should be chosen from each District in turn.

Notes to text:
  153. Lit., and he wept to his elders and made-known the thing which had    happened to him, etc.
  160. Lit., like little children.
  163. Lit., this Hanfakiu. And similarly in many other places.
  165. That is, a ghost or a dead person thought of as still living in a ghostly   (but not immaterial) form.
  179. See "Founding of Rotuma" 105.
   201. Lit., and so this Noa'tau brought the chiefship.
   203. This appears to be the sense required by the context, though it can    hardly be got from the Rotuman text, which, literally, means, "And so the fan    would now be opposed (or, matched) by (or, at) Riamkau's accession"
   205. The eastern extremity of the island.

In this instance the rebellion is given justification, and it is successful. The ostensible reason is the sau's usurpation of a wife from one of "the people", but encoded in the names is a deeper message. The man's name, Moea, means 'crops, harvest', while his wife's name, Panai, means 'of certain trees, about to fruit'. The suggestion is that the sau takes more than his share of the fruit of the land, justifying a rebellion. In another version of the myth the rebellion is provoked by a failure of the chief (in this case mua) to distribute food at a feast in an equitable manner (Sumi Mission Papers; see also Howard in press). The notion that the rebellion is a popular one is reinforced by the name of Moea's sister, Hanfakiu, which is composed of the roots for 'woman' 'man', and 'ten thousand'.

The success of the rebellion, which is facilitated by 'atua 'spirits' and the demigod Tokainiua, results in the sauship passing to a Rotuman, as opposed to a half-Samoan. In Trouillet's oral history, this is the beginning of a series of oscillations between indigenous and stranger chiefs, [25] as well as the first in a series of successful rebellions (Howard in press).

The solicitation of Tokainiua's assistance in this rebellion is ironic in so far as he is the prototype of the alien chief, but the incident merely reinforces the earlier development of complementarity between him and Raho, symbolised by his burial under the fesi tree and Raho's movement to Hatana. Here Tokainiua is further domesticated and indigenised, to the point of being given a district established for him--a symbolic way of transforming him into a founder in the mould of Raho. The differentiation dramatised in their initial encounter is thus markedly softened.

The connection between the success of the rebellion and the final statement in the narrative--that from that time onward it was the custom that the kings in Rotuma should be chosen from each district in turn--is not immediately apparent. In part, however, it seems to suggest a resolution to the basic paradox of chieftainship. The custom, it will be recalled, is for the sau to be selected from one district and to reside in another. He is therefore a stranger to the district of his residence, though a native to the whole of Rotuma, over which he presides. In this way both aspects of chieftainship--being indigenous yet a stranger--are expressed.

More than this, however, the statement is the culmination of a set of redundant messages that there can be no sustained legitimate hegemony in Rotuma, either of a kin-based aristocracy or of district pre-eminence. Thus, no one kinship line sustains dominance, nor does one district prevail. Rather it is the impermanence of authority that is underscored and justified.

Summary of Rotuman Political Conceptions

The two myths presented above reveal a conceptual paradigm that lies at the heart of Rotuman political thought. Of fundamental concern is the issue of prosperity--the prosperity of the island as manifest in human fertility and the productivity of the land. The central symbol is food; its abundance is indicative of a proper political order, its scarcity indicative of political malaise. The ultimate source of prosperity is the spirit world, but it is the primary responsibility of chiefs to act as intermediaries with the gods who dwell there (some of whom are presumed to be their ancestors) and so influence them to act benignly. Conceptually the distinction between gods and chiefs is somewhat blurred and chiefs, upon their death, are transformed into powerful spirits. The mythical prototypes of chiefs, Raho and Tokainiua, are best described as demigods, with characteristics of both men and spirits. This conceptualisation sets up the central paradox of the myths --that chiefs are at once persons and not persons. They come from the people but are different from them.

The paradox is expressed in the myths through explorations of themes involving differentiation and reintegration. Rotuma is differentiated from Samoa, the land is differentiated from the sea, and people are differentiated from chiefs; then, in various ways, reintegration takes place and constraints are placed upon the oppositions involved. Mediating categories such as islets and trees come to predominate over oppositions between sea and land, sky and earth. As part of this reintegration, the opposition between the people, represented by Raho, and the chiefs, represented by Tokainiua, is muted and constrained. The relationship between people and chiefs, is finally construed as one of complementarity, with the people producing food (and other goods and services) for the benefit of chiefs, who intercede with the gods, who provide abundance to the land. However, this conception renders the nature of chieftainship problematic, for where is the source from which legitimate chiefly authority derives? Is it from the gods, whose association with the chiefs provides them with supernatural potency, or is it from the people, who have elevated the chiefs and supported them with the products of their labour? Both, of course, are sources of legitimacy, but the degree of emphasis on one or the other has important implications. The problem is common to all Polynesian societies, and resolutions differ. Some of them, particularly the highly stratified ones like Fiji, Tonga, Hawai'i and Tahiti, emphasise the affiliation of chiefs and gods. The association is strengthened through lengthy genealogies tracing descent directly to ancestral deities, and the differentiation of chiefs from the people is clearly and sharply drawn. In those societies the mythology seems to reflect a preoccupation with chiefly rivalry, and in practice chiefs vied with one another for ascendance and manipulated their genealogies to legitimate their affiliation with the gods. In Rotuma the situation was different. While there is undeniable rivalry between chiefs reflected in the narratives (the contest between Raho and Tokainiua being a case in point), a more salient theme concerns relations between chiefs and the people. The relative lack of differentiation between them accentuates the underlying anomaly, and the resultant tension is expressed through numerous tales of insurrection and rebellion (Howard in press). The basic message appears to be that chiefs are expected to use their godly powers for the benefit of the people, and that if they do not, if they turn mean and selfish at the expense of the people, then rebellion is not only justified, it is likely to be supported by the gods.

The myths also help to clarify the positions of mua and sau in Rotuman political thought. Both apparently embody representation of the total Rotuman polity, the mua in its primal undifferentiated generic state of conception, and as representative of the people in its differentiated form; the sau as representative of chieftainship in its differentiated form. In its differentiated state the mua and sau represent the complementary principles of domestication and vitality that together are the essence of legitimate chieftainship.

Structurally the second myth inverts key aspects of the first one. Thus, the founding myth begins with an undifferentiated world and moves towards a differentiated polity characterised by a stabilised opposition between the people of the land (symbolised by Raho) and chiefs (symbolised by Tokainiua). The second myth begins with this stabilised opposition (symbolised by the mua and sau in the sky) and moves towards an instability requiring rotation of authority as a solution. By means of this rotation unity is re-established.

But while the myths encode the fundamental logic of Rotuman political thought, and thus provide a necessary background for interpreting political institutions, such narratives do not provide sufficient information for explaining their historical manifestations. The completion of the picture requires an examination of political pragmatics, to which I now turn.

It will be recalled that at the time of discovery by Europeans Rotuma was divided into seven districts headed by gagaj 'es itu'u 'district chiefs' and that the fakpure, who presumably appointed the sau and mua, was the head of one of these districts. The districts were divided into territorially distinct kinship communities known as ho'aga, each of which was headed by a titled male. Titles were ranked, and in theory district chiefs were chosen from a set of ho'aga tracing ancestry to a common chiefly source (moseaga 'from the same bed'). Ho'aga in the set were supposed to rotate the privilege of choosing a successor to district chieftainship, with kinship seniority heavily weighted as a criterion for selection. [26] If the man appointed to the position proved unsatisfactory for one reason or another he could be deposed by members of his ho'aga, who had the right to take away the title, hence authority, and allocate it to another.

In contrast with those stratified societies in which all major chiefs traced their ancestry directly to deified ancestors, Rotuman district chiefs thus drew their authority more directly from the people in their locality, and since the districts were autonomous political units, this posed a problem with regard to the relationship of the island as a whole to the gods. The problem was one of potency, for only truly powerful chiefs could exert influence upon the gods, who were perceived as capricious and willful. There was therefore a strong cultural push towards a chiefly hierarchy reflecting relative potency, or perhaps more accurately, demonstrating the great potency of the dominant chief. Since success in warfare was prima facie evidence of potency, a chief whose district was on the winning side of a battle was a candidate for paramouncy. All available evidence suggests that wars in Rotuma generally involved dichotomous alliances, and that the head of the prevailing alliance would assume a position of paramouncy, becoming fakpure.

This still left a problem, however. Since the fakpure was chief of one district among seven, and since he was engaged in secular politics, he was not a very suitable figure for symbolising the unity of Rotuma. The position of sau was a solution. The sau occupied a sacred post, divorced from secular politics. He personified the total polity, and represented it (along with the mua, who for these purposes was alter ego to the sau) to the gods. His suitability, measured by the net prosperity of the people (bounty minus labour and tribute), was a direct reflection of the suitability of the fakpure, whose secular power kept the sau in office. [27] The solution was elegant, but it entailed some practical problems associated with the selection of candidates and the burden of supporting the sau in an appropriate manner. In the system of ranked lineages which characterised the great Polynesian chiefdoms, selection did not pose the same order of problem, since rank was relatively unambiguous and primogeniture provided a definite rationale for choice. As a corollary, persons of lesser rank were obligated to provide support for their superiors by the extension of kinship rules. In Rotuma, however, where locality outweighed kinship as a political principle, ranking was far more problematic. Thus, there were multiple contenders for sauship making succession a recurrent issue of potential dispute. Warfare was one mechanism for resolving such status ambiguities; rotation, as Williamson pointed out, was another. Rotation appears as an early solution in Rotuman myth, but never to the exclusion of warfare. Indeed, Trouillet's narrative relates repetitive challenges to fakpure and sau suggesting that rotation between districts did not settle the issues involved.

A key issue seems to have been the appropriate length of a sau's reign. Rotation ingeniously involved selecting a person from one district and setting up his residence in another, thus symbolising both qualities--indigenous and foreign--which combine to constitute paramount chieftainship. It seems from the narratives, however, that the people of the host district bore the brunt of responsibility for supplying the gluttonous needs of the sau and for them the balance of benefits versus costs may have quickly shifted. Resentment of such burdensome demands is a prominent theme throughout the oral history of the island. There is evidence to suggest that over time the term of office for sauship shortened, and by the time the institution was terminated c. 1870 sau were serving for minimal periods. From Trouillet's documentation of sauship during historic times (1797-1870), three periods can be distinguished, as follows:


Average Reign
Rotuman Cycles

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, [28] led Rotumans to question more intensely the efficacy of individuals who occupied the office of sau. It may well have been, as Frazer pointed out many years ago in The Golden Bough, that as the public image of a chief approached impotence, the need to replace him increased. Rotumans seem to have simply amplified the institutionalised mechanism already available to them, and used installation as a repetitive means of revitalising the declining office.

* * *

I have suggested in the foregoing analysis that the problem in cultural logic confronting Rotumans in conceptualising their political system arose from a set of categorical paradoxes associated with chieftainship: that chiefs are gods, but are human; that they are of the people, but are different from them; that they represent the unity of the polity, but have parochial interests within it. While these paradoxes are posed in digital form within the myths, I believe they are constructed from an underlying set of analogical premises, common to all Polynesian systems. In short, I suspect Polynesians conceived of people as more or less godlike, with the paradoxical dilemmas emerging in relation to specific instances (the myths being, in this view, an exploration of such instances). Two principles were involved, rank and distance. Rank was conceived primarily in genealogical terms, traced back through first-born children of first-born parents to founding ancestors, and, ideally, to the gods of creation. In many Polynesian societies genealogies were truncated as a result of other contingencies, Rotuma being an extreme example. In terms of process, this principle was one of elevation, i.e., establishing correct genealogical links was a means of elevating one's social status. The principle of distance had both physical and social aspects. Physically, removal of a person from normal social arenas was a way of making him more remote; socially, distancing was achieved through ritual prohibitions and other means of differentiating the person's behaviour from normal patterns. At the extreme, such persons inverted social norms (e.g., committed incest, ate human flesh), thus emulating the behaviour of gods. From a processual standpoint distancing involved the principle of mystification, rendering the person less culturally human and more like the gods. The two principles can be portrayed diagrammatically as in Figure 1.

At the apex of rank and distance were the high gods of Polynesian mythology, at the base were slaves, persons utterly without rank or sanctity. As suggested by their positioning in the figure, local secular chiefs enjoyed some rank but were only slightly distanced, local gods somewhat greater rank and a moderate degree of distancing, while high chiefs were in the upper ranges of both dimensions, at least in the more stratified societies. However, to think of persons or supernatural beings as fixed in position is to miss the point, for two reasons. One is that positions were relative--a chief may have been godlike to a commoner, but just another man to a person of comparable status while a commoner may have been perceived as godlike by his children. In other words, space within the figure should be thought of as fluid and relational, except, perhaps, at its extreme parameters. A second reason that positions should not be thought of as fixed is that the Polynesian concept of mana, which might loosely be translated as 'potency', involved a notion of inherent instability, since it was in action that it was manifest and codified (see Firth 1940). Hence all statuses vis-à-vis one another were continuously waxing or waning.

The point I wish to make is that this underlying Polynesian cultural logic unfolded differently in different societies, depending upon historical circumstances. In those archipelagos containing large islands and substantial populations, where lineality was unrestrained by pragmatic circumstances favouring local autonomy, these principles were carried to their logical extremes. Genealogies were traced back to creator gods, and high chiefs were distanced from commoners both physically and socially to the point where their mystification approximated that of high gods. As a class they were so far removed from the realm of the people that their significant relationships were confined to each other and to the gods. The myths from these societies reflect this situation. In Rotuma, however, which is an isolated island of rather small size (43 square kilometers) and a medium-sized population, pragmatic constraints favoured local autonomy and set limits on the degree to which chiefs could be differentiated from the people. Genealogies were shallow and distancing was difficult both physically, because of the small size of the island, and socially, because the population was too small to facilitate a distinct breeding population of chiefs, keeping kinship distance within boundaries. As a result Rotuman chiefs were not in a strong position to be either elevated in rank or mystified to a level approximating gods. Conceptually they were much closer to the people, and this presented the problems that Rotuman myths focused upon.

Further comparative analysis should help to clarify the way in which cultural logic interacted with particular environments to produce the variety of political structures in the Polynesian culture area. In such an endeavour the analysis of myth can be expected to play a central role.


This paper was inspired by participation in a seminar on Polynesian chieftainship held at the University of Hawaii during the Spring of 1981. All participants in the seminar contributed to the interpretations presented in this paper, but I am especially indebted to Marshall Sahlins, whose work on Fijian and Hawaiian cultures provided the theoretical framework for discussion. He supplied a detailed critique of a first draft of this paper and I have revised accordingly. I am also grateful to Jacob Bilmes, David Hanlon, John Kirkpatrick and Bradd Shore, each of whom provided insightful criticism.

1 For background on Rotuman history and ethnography consult Gardiner 1898; Howard 1966, 1970. [back to text]

2 Even this agreement is called into question, however, by one of Hocart's informants who referred to a time when there was only one eligible person in Rotuma, the legendary Fonmon. He was supposed to have impregnated ambitious women from around the island, making their offspring eligible. The informant added, however, that sometimes and individual was appointed as a result of hard work (Hocart n.d.). That a person might conceivably be appointed sau for achievement is also hinted at in several myths. [back to text]

3 Where there are multiple versions of the same name used in different accounts I have included alternative representations in parentheses at strategic points. Throughout this paper I use Churchward's orthography except when quoting directly from another source. He offers the following guide to pronunciation, using English equivalents: a as in clam, but shorter, unless written a; a as in want; a as in cat; ä as in fan; e as in bet; f as in fish; g as ng in sing; h as in heart; i as in sit; j as tch in pitch; k as in rake; l as in laugh; m as in mask; n as in nine; o as in obey; ö pronounced in German, somewhat like er in her; p pronounced as in English, but blunted somewhat towards b; r pronounced with a slight trill; s between English s and sh; t pronounced strictly dental, the tip of the tongue being pressed against the back of the top teeth; u as in put; ü pronounced as in German (this sound may be approximated by endeavouring to pronounce ee in see, with the lips rounded); v as in vat; when v falls at the end of a word, particularly following an a, it is often imperfectly articulated and sounds like o; ' glottal stop (Churchward 1940, Part II). [back to text]

4 Trouillet's account, in French, was never published and his journals were transported to the Vatican archives just before my arrival in 1960. Fortunately, however, copies were made by Gordon MacGregor, an anthropologist who visited the island in 1932, and by H.S. Evan, and Englishman who served as District Officer on Rotuma from 1949-1952. [back to text]

5 Hocart collected texts in Rotuman but these were neither translated nor published; they remain in his collection of field notes at the Turnbull Library, Wellington, New Zealand. [back to text]

6 This is a metathesised form of the complete phase of the word, which is fäega (see Churchward 1940:189). [back to text]

7 This is not to deny that some analysts (e.g., Freudians) pay attention to analogic properties of codes when interpreting myth. [back to text]

8 My assumption is that proper names used in Rotuman myths condense a range of associated meanings in much the same way that dream symbolism does. Without the opportunity to elicit association we are unable to decipher all but the most obvious meanings. [back to text]

9 I am following the usage of Keesing (n.d.), who suggests we gloss the term mana as 'potency' rather than 'power' since the stative form fails to capture the dynamic nature of the concept. [back to text]

10 Neither the concept of "Samoa" not "Tonga" should be interpreted as simply a reference to the corresponding geographical entities. A full examination of their usage suggests a more complex semantic structure. "Savai'i" or Savaiki" are often substituted for "Samoa" in the myths, these being cognate forms for the generic Polynesian "homeland". I suspect that they represent traditional usage, and that only after European contact were they replaced by "Samoa" (the island of Savai'i being identified by Europeans as part of the Samoan archipelago). The term "Tonga" seems to be a generic term referring to a mythical, or quasi-mythical, source of supernatural potency. In some narratives "Tonga" is located beneath the earth or sea. The word is also used as a adjective in reference to the south-east trade wind. [back to text]

11 There are several terms in Rotuman that can be glossed as 'chief', none of which are clear equivalents. The problem is that 'chief' condenses various aspect of rank which Rotuman differentiates them. [back to text]

12 The notes following each segment of text are Churchward's, as is the numbering of sentences. Only those of Churchward's notations that pertain to translation are included; additional notes refer to points of grammar in the Rotuman text. [back to text]

13 Rotuman myth uses a geographical code based on the east-west distinction. East is associated with chieftainship, and particularly with conquering chiefs who come from abroad, while west is associated with the indigenous people. Within Rotuma the geographical code is based on a division of the island into three segments along the east-west axis, and a north-south division. The island is divided into two main parts, joined by an isthmus of sand, forming a configuration of about 13 kilometres long and at its widest nearly 5 kilometres across, with its lengthwise axis running almost due east and west. That portion of the land to the east of the isthmus is called Fa'u 'Back' and is strongly associated with the indigenous people. This contrasts with the remainder of the island, which is termed mua 'Front'. (The west end of the island is also referred to as sio 'down', the east end as se'e 'up'.) The eastern segment is further divided into an end and middle section. The end section includes Oinafa and Noatau, which, being at the extreme eastern part of the island, is most closely associated with stranger-chiefs. The mid-section includes Malhaha, Fag'uta and the portion of Itu'ti'u east of the isthmus. In the myths, contrasts between the extremities of the island (e.g., between Oinafa/Noatau and Fa'u) imply strong opposition; contrasts between either end and the mid-section a somewhat weaker form. Another opposition is between north and south, north being associated with chieftainship, south with common status. Whereas east is used to signify externally derived chieftainship, north is a marker for indigenously derived chiefs (see Howard in press). [back to text]

14 The equation of pig = human sacrificial cannibal victim is explicit in the myth of Möstoto (Churchward 1939:462-9). The red colour of the variety of bananas referred to (prmea) is apparently associated with blood, making this, too, a "blood sacrifice." [back to text]

15 If we assume, in a Freudian vein, that associated words add semantic value to a condensed symbol (see note 8), it is likely that the word mama (with a lengthened final vowel) is implied as well. It translates as 'without clothes, in a state of nakedness', or alternatively (in 'a mama) 'to eat meat or fish, etc., without vegetables or with very little vegetable food' (Churchward 1940:258). Both imply a god-like state. [back to text]

16 Further speculation with these names is possible but would involve skating on thin ice. For example, Nujmaga translates as 'Big mouth', Nujka'u as 'Little mouth' (more properly 'a mouth distorted by yaws') (Churchward 1939:330fn.). The latter might be a euphemism for 'anus', suggesting a contrastive set based on:


If one wanted to push the argument one could find supporting evidence in the names of Raho's sisters: the eldest = mama + e + re = 'maker of chewed food', the youngest = mama + fia'rere = 'chewed food + 'to squat' (hence to defecate?). The middle sister's name, Mamahioväre, implies uselessness or barrenness: hio = 'an ancient dance' + väre = 'worthless'. Tempting as such speculations may be, it would be extending interpretive licence a bit far to accept them at face value. In fact, alternative roots could be postulated with quite different results. [back to text]

17 In Rotuma, as in other Polynesian societies, consanguineal kinsmen as conceived as sharing the same substance both through common descent and through sharing food from the same ancestral lands. [back to text]

18 In some versions of the myth the sun is explicitly identified as progenitor. [back to text]

19 This should not be construed as a literal reference to the Tongan archipelago, but rather to a mythical source of supernatural potency (see note 10). Tü'toga thus implies enormous power. The concept is also used in reference to 'food grown on a strip of gardening land, going right across a number of adjacent gardens, and set aside as sacred to a high chief' (Churchward 1940:338). [back to text]

20 This interpretation is based on the notion that, although sisters have superior status over their brothers in the tradition of Western Polynesia (as manifest in the superior status of Raho's sisters), women as wives, i.e., childbearers, are of inferior, common status. Maiva, as a half-sibling, is in a somewhat anomalous position. The myth uses this anomaly as a basis for furthering the theme of differentiation (see Ortner 1981, Shore 1981 for an explication of the symbolic significance of woman as sister versus woman as childbearer). [back to text]

21 As in the Oedipus myth a deformed foot suggests association with the earth, with incomplete differentiation from it. Symbolically this "roots' Maiva to the land as opposed to the sky. [back to text]

22 Alofi is a small island adjacent to Futuna. As the smaller of a pair of islands it nicely symbolises the inferior half of an implied opposition. [back to text]

23 There is more to the kava incident than this, of course. The initial kava, which is given anthropomorphic characteristics, forsakes Raho for an unnamed female chief (sauhani = female sau). This seems to call into question Raho's primacy and points to the prior association of the female principle with the people of the land. Raho, being male, might only be a second order symbol. [back to text]

24 Concerning the interpretation of this chant Gardiner writes:

"The language is antique, and now nearly forgotten; I could get no translation to the last two lines. The third and fourth lines are repeated with the names for all the fruits substituted for the ifi and fava; uktrua is supposed to mean that it is finished. All carry during the ceremony a stick, the poki; it is held over the head with both hands and moved rhythmically to and fro with the singing. The garagsou was explained to me as the head of Limari, the abode of departed spirits, and also as the god of the winds, rain, and sun, but Marafu identified him as being the same as Tagaloa Siria [the highest god] (Gardiner 1898:465). [back to text]

25 The notion of "stranger chiefs", (i.e., individuals who either in fact or symbolically come from outside to reconstitute the social order) has been compellingly developed by Sahlins for Fiji (see Sahlins 1981b). [back to text]

26 For a more extensive account of succession see Howard 1964. [back to text]

27 Although I have presented the case in temporal terms, the issue is one of cultural logic rather than history. There is no imputation of chronology intended. [back to text]

28 The population declined from an estimated 3000-5000 at contact to fewer than 2000 in the early part of the 20th century (see Howard 1979). [back to text]


ALLARDYCE, W. L., 1885-6. "Rotooma and the Rotoomans." Proceedings of Queensland Branch of the Geographical Society of Australia, 1st sets:130-44.

ALLEN, W., 1895. "Rotuma." Report of Australian Association of the Advancement of Science:556-79.

CHURCHWARD, C. M., 1937-39. "Rotuman Legends." Oceania, 8:104-16, 247-60, 351-68; 9:326-39, 462-73.

------ 1940. Rotuman Grammar and Dictionary. Sydney, Australian Publishing Co. Ltd. for the Methodist Church of Australasia, Department of Overseas Missions.

DILLON, P., 1829. Voyage in the South Seas, vol. I. London, Hurst Chance.

FIRTH, Raymond, 1940. "The Analysis of Mana: An Empirical Approach." Journal of the Polynesian Society, 49:383-410.

GARDINER, J. S., 1898. "The Natives of Rotuma." Journal of the Royal Anthropological Institute, 27:396-435, 457-524.

HOCART, A. M., Ms. n.d. Field Notes, Rotuma. Wellington, Turnbull Library.

HOWARD, Alan, 1964, "Land Tenure and Social Change in Rotuma." Journal of the Polynesian Society, 73:26-52.

------ 1966. "The Rotuman District Chief: A Study in Changing Patterns of Authority." Journal of Pacific History, 1:63-78.

------ 1970. Learning to Be Rotuman. New York, Columbia Teachers' College.

------ 1979. "The Power to Heal in Colonial Rotuma." Journal of the Polynesian Society, 88:243-75.

------ in press. "The Charter for Rebellion in Rotuman Myth." Pacific Studies.

KEESING, R., Ms. n.d. Rethinking Mana.

KIRCH, Patrick V., and Douglas E. Yen, 1982. Tikopia: The Prehistory and Ecology of a Polynesian Outlier. Bernice P. Bishop Bulletin, 238. Honolulu.

LESSON, R. P., 1838-9. Voyage autour du monde...sur 'La Coquille'. Paris, Pourrat Freres.

MACGREGOR, Gordon, Ms. n.d. Field Notes, Rotuma. Bernice P. Bishop Library.

ORTNER, Sherry B., 1981. "Gender and Sexuality in Hierarchical Societies: The Case of Polynesia and some Comparative Implications", in S. B. Ortner and H. Whitehead (eds), Sexual Meanings: The Cultural Construction of Gender and Sexuality. Cambridge, Cambridge University Press.

ROBERTSON, J. B. W., 1962. "The Evaluation of Maori Tribal Traditions as History." Journal of the Polynesian Society, 71:293-309.

ROMILLY, H., 1893. Letters from the Western Pacific and Mashonaland 1878-1891. London, D. Nutt.

RUSSELL, W. E., 1942. "Rotuma: Its History, Traditions and Customs." Journal of the Polynesian Society, 51:229-55.

SAHLINS, Marshall, 1981a. Historical Metaphors and Mythical Realities. Ann Arbor, University of Michigan.

------ 1981b. "The Stranger-King." Journal of Pacific History, 16:107-32.

SHORE, Bradd, 1981. "Sexuality and Gender in Samoa: Conceptions and Missed Conceptions", in S. B. Ortner and H. Whitehead (eds), Sexual Meanings: The Cultural Construction of Gender and Sexuality. Cambridge, Cambridge University Press.

SMITH, S. Percy, 1910. Hawaiki: The Original Home of the Maori. London, Whitcombe and Tombs.

SUMI MISSION STATION, Ms. n.d. Histoire de Rotuma. Rotuma.

WILLIAMSON, R. W., 1924. The Social and Political Systems of Central Polynesia, 3 vols. Cambridge, Cambridge University Press.

Home Page
Howard-Rensel Papers