Volume 98 1989 > Volume 98, No. 3 > Possible prehistoric contacts between Tonga and Anuta, by R. Feinberg, p 303-318
                                                                                             Previous | Next   

- 303

In recent decades, archaeologists, in collaboration with historical linguists, have made considerable strides in reconstructing Polynesian Outlier culture history. For the most part, these interpretations have been consistent with ethnographic evidence and oral traditions. I take issue, however, with certain aspects of what has become “common wisdom” regarding Anutan settlement and prehistoric interisland contacts.

The position I wish to dispute is illustrated in a recent summary of developments in the study of Polynesian Outlier prehistory (Kirch 1984). There, Kirch correctly stated that:

Anutan oral traditions speak of two periods of settlement, with an autochthonous population (the earth-sprung apukere) having been supplanted some 12 generations ago by immigrants from Uea, presumably Uvea (p.230). 1

This, however, was immediately followed by the observation that:

The first linguistic study of Anutan, by Green, indicated that the language was derived from Nuclear Polynesian, although a period of Tongan borrowing was suggested. More recently, Biggs has outlined a hypothesis for the inclusion of these Tongan loan words, suggesting that Anuta was colonized by Uveans not long after Uvea itself had been conquered by Tongans (in about the 16th century A.D.). Thus the Tongan loan words were introduced into Anutan speech via Uvea, and not as a result of a Tongan intrusion in Anuta (p.230, emphasis added).

In a comment in the following year, Kirch restated this position still more forcefully. There, he asserted:

recent linguistic studies of Green (1971) and Biggs (1980) . . . demonstrate conclusively that Anutan is a Nuclear Polynesian language, and strongly suggest that the island was settled directly from 'Uvea (Wallis Is.). According to Biggs, the few Tongan lexemes in the Anutan language had already been borrowed into 'Uvean, and therefore do not imply a direct Tongan origin (Kirch 1985:382).

- 304

In fact, according to Biggs, Kirch's “few Tongan lexemes in the Anutan language” constitute approximately 34 per cent of Anuta's lexicon (Biggs 1980:124). In addition, he seems to suggest that Anuta was settled primarily not from 'Uvea but from Tikopia, with subsequent borrowing of a number of Tongic linguistic features from 'Uvean (1980:124-5).

A problem common to both Kirch's and Biggs' hypotheses is the reference to Tongan contacts in Anutan oral traditions as well as those of other Outliers. Kirch addresses this by questioning the accuracy of such traditions and suggests instead that:

immigrants from west Polynesia brought with them stories and traditions relating to the great Tongan expansion of the 16th and 17th centuries, including the conquest of such islands as Uvea, Niuatoputapu, Niuafo'ou, and Rotuma. Thus the Outlier traditions may be an echo of political currents that swept the west Polynesian region in the final centuries before European contact (1984:238).

By contrast, I should like to argue that, in many cases, including those purporting to account for Anutan origins, oral traditions describing Tongan contacts are plausible. Such accounts are internally consistent and compatible with linguistic, archaeological, and ethnographic evidence. Before the historical accuracy of Anutan oral traditions may be evaluated, however, we must be clear as to just what they say.


Several indigenous versions of Anutan settlement claim that the population is descended from a group of Tongan immigrants. One such version is reported in Firth (1954) and repeated by Bayard (1966), Pawley (1967), and Green (1971). More authoritative statements, however, assert that the island was settled by two groups of immigrants, one from Tonga and one from Uvea, which arrived together about 15 generations ago. According to one version, they descended simultaneously as crews of two separate canoes (see Appendix A). A second variant has them arriving in a double-hulled canoe after having fought their way through several Tuvaluan atolls (see Appendix B). In either case, Anutans believe the Uvea in question to be, as Kirch and others have suggested, East 'Uvea or Wallis Island. 2 The Tongans were led by a man named Pu Kaurave; the Uvean leader was one Pu Taupare, also known as Pare.

After a brief dispute with the Tikopian chief Pu Ariki (also known as Pu Lasi or Pu Taumako Lasi), the latter permitted the immigrants to remain on - 305 Anuta while he returned to Tikopia. The first chief of the new population was Pu Kaurave, the Tongan. Pu Kaurave was succeeded by his son, Ruokimata. Ruokimata, however, had no son. Thus, he was succeeded by a man named Toroaki, son of Pu Taupare, the Uvean. Since that time, the chieftainship has been in the Uvean line. Still, the Tongans had made their mark culturally and continued to be represented in the population through the women. (Pu Kaurave and Pu Taupare were tau maa ‘brothers-in-law’, as a result of “sister exchange” — see Appendix B.)

In later generations, further contacts are alleged to have occurred with Tonga, Tuvalu, Rotuma, and Futuna to the east, as well as Tikopia and Taumako to the west.

The first such visit took place during Toroaki's reign, when a man named Pu Maatopa (Rotoiporau) arrived from Tonga. 3 Pu Maatopa's son was Pu Pourou (Manoamero or Manavamero); Pu Pourou's son in turn was Pu Pangatau, founder of the Kainanga i Pangatau, one of Anuta's four contemporary “clans” (see Feinberg 1981:129-34). Pu Pangatau's brothers died in a series of battles leading to the present “clan” system's establishment.

Two generations after the first immigrants, a man named Remonu is said to have arrived from Samoa. Remonu's son, Paovaka, became Anuta's military guardian (pau penua) and is credited with playing an important role in the successful resistance to two Tongan invasions. This line died out, however, in the following generation when Paovaka's five sons attempted to usurp Anuta's chieftainship. Toroaki's great-grandsons — Tearakura, Pu Tepuko, and Tauvakatai — with Pu Pangatau's assistance, defeated the “Samoan” brothers, thereby maintaining the Uvean line's hegemony.

Anutans claim two voyages from their island to the Polynesian Triangle. According to a well-known story, Toroaki's eldest son, Pu Ratu, sailed to Tonga and never returned. His exploits were made known by a Tongan calling himself Takaraua, who sailed to Anuta to report on Pu Ratu's activities. 4

The second story involves Pu Tingirau, Tearakura's father. Pu Tingirau sailed to Uvea to bring back a pearl-shell fishing lure for his children, but he stayed many years before returning home. By the time he did return, Tearakura was well established as chief. The rule of succession prohibited assumption of the chieftainship while the former chief was still alive. Therefore, as Pu Tingirau approached his destination, the spirits intervened, causing his canoe to swamp and the crew to be lost at sea. One man survived to swim ashore and tell the people on Anuta what had happened; he then returned to the ocean and was never seen again.

Several other visits from Tuvalu, Tonga, and Taumako, are said to have occurred during the first five generations of Anuta's current population. The - 306 Tuvaluan invasion was led by warriors named Porongai, Piikia, Poepoe, and Manuuri. This story has been independently confirmed by the Nanumean inlaws of an Anutan who married one of Porongai's descendants, and by anthropologist Keith Chambers on the basis of extensive field research in Nanumea (personal communication). The other invasions are less well documented. Still, there is evidence supporting Anutan identification of the interlopers' lands of origin. In particular, the invaders from the east are said to have called their double-hulled canoe te tongiaki. This is the Tongan term for their large voyaging canoes, while neither Haddon and Hornell (1975) nor Burrows (1937) cite a similar term for 'Uvea. The western voyagers called their canoes tepuke, a term still used by Polynesians from Taumako and the Reef Islands of the Santa Cruz group (Lewis 1972).

These visits apparently were brief and hostile, and appear to have made little lasting impact on the population. By contrast, many Tikopians and at least one Rotuman apparently stayed on Anuta, where they married and left progeny. Descendants of the Rotuman and one Tikopian are represented by distinct lines (pare ‘houses’) in the present population. Most of these relationships are shown in the detailed genealogies recorded in my book on Anuta's social structure (Feinberg 1981).


Anutan traditions are in firm agreement with Kirch's generalisation that:

Outlier culture histories are often as complex as those of the major southwest Pacific archipelagoes. . .(1984:236). Simple models of cultural development. . .are out of place in the complex world of the southwestern Pacific (p.238).

Thus, Kirch's tendency to describe Anutan culture history as if the only significant external influence were 'Uvean is perplexing. If he is to reject Anutan claims of an early Tongan presence, one would presume that the evidence against indigenous assertions must be overwhelming. What, then, is the nature of this evidence?

Social structural considerations show Anuta as more Tongic than Samoic (Kaeppler 1973). 'Uvea, however, despite having a language that is of the same highest order subgroup as Samoan, closely resembles Tonga in social structure (see Burrows 1937, Goldman 1970). Consequently, much of Anuta's apparently Tongic social structure could have been imported from 'Uvea. Still, a few features (e.g., succession of chiefly titles from father to eldest son rather than from elder brother to next eldest) are more similar to Tonga than 'Uvea, suggesting some direct Tongan influence. Archaeology - 307 tells us much about Anuta's past, but, on the question of possible Tongan ancestry, it is silent. This, then, leaves historical linguistics as the basis for rejecting the historical validity of Anutan oral narratives; and indeed, it is such evidence that Kirch (1982:253; 1984:230; 1985:382) cites for his conclusions. Let us, then, examine the linguistic evidence.

First, it seems odd that Kirch cites Green (1971) rather than Feinberg (1977) as the authoritative source on the Anutan language (ANU). Green was on the island for one day, does not speak Anutan, and worked through a Tikopian interpreter. At the time, the only other source on the Anutan language consisted of the few words reported in Firth (1954), and Green did not know that another anthropologist would soon be spending a year on the island. His efforts, therefore, were commendable. Still, his published wordlist includes many mistakes, and much of it is Tikopian rather than Anutan. By contrast, I lived on the island for a year, speak the language, and put together my word-list without the aid of an interpreter.

To say that my lexicon is more exhaustive and may have fewer mistakes than Green's, of course, is not to say that his conclusions are in error. Indeed, Biggs' important article (Biggs 1980), based on data in Feinberg (1977), corroborated Green's conclusion that Anutan is a Nuclear Polynesian language. At the same time, he confirmed my earlier suggestion (Feinberg 1977:4-5) that East 'Uvean (EUV) was a major contributor to the Anutan language, and that this ancestry accounts in large measure for both Anutan's Tongic (TO) and its Nuclear Polynesian (NP) elements. This conclusion, however, does not entail that local traditions are mistaken when they claim a Tongan presence among Anuta's early settlers and visitors.

While I shall not dispute Anutan's classification as an NP language, I believe that the extent of Tongic influence has been underemphasised in recent writings. Biggs (1980:124) seems more aware than Kirch (1985:382) of major Tongan (TON) input into ANU. Yet I suggest that even his assessment is an underestimate. In particular, Biggs seems unaware of several Tongic features present in Anutan. 5 Detailed correction of misapprehensions in Biggs' generally perceptive article, and thorough reanalysis of ANU's position among Polynesian languages, must await another occasion. At this time, I should simply like to indicate a few of the more important bits of evidence that Biggs seems to have missed, indicating Tongic influence in ANU.

Perhaps most importantly, Biggs states:

Six grammatical forms that Pawley says are innovations of Nuclear Polynesian, but none of those he considers to be innovations of Tongic, are found in ANU, - 308 thus providing strong evidence that the language is a member of the former subgroup (1980:121).

In a table labelled “Tongic Morphological Innovations in EUV and ANU,” Biggs lists mo as the Anutan reflex of PPN *ma ‘and’. This places ANU in line with PTO (*mo) and contemporary Tongan (mo). EUV shows both mo and ma. 6

The two other items in Biggs' table are PPN *eni ‘this’ and *ena ‘that’. Biggs lists (*)heni and (*)hena as the PTO and EUV reflexes, but he cites no ANU cognates. In fact, ANU uses eni and ena in very much the same way as heni and hena are used in TON and EUV. As Pawley notes:

the semantic distinction between “demonstrative pronoun” and “demonstrative locative” is marked syntactically rather than morphologically; i.e., the semantic distinction is marked not in the demonstrative forms themselves but by other features of the utterance: a positional or directional particle plus a demonstrative is interpreted as a locative, e.g., TON mei heni “from here, from this place”, ki heni “to here, to this place”; but following a nominal base or specifying particle, a demonstrative is given a pronominal interpretation as TON ko eni “this (one), these (ones)”, ko e fale eni “this is the house” (Pawley 1967:272-3).

ANU uses mai eni ‘from here’, i eni ‘here’, ki eni ‘to here’, ‘toward this place’, and ko eni ‘this’, ‘this one’. Ena is used in a similar manner. Although ANU does not employ the word order cited by Pawley for the sentence ‘This is the house’, the lexemes are essentially the same: Anutans say Ko eni ko te pare. By contrast, such common Nuclear Polynesian reflexes as teenei ‘this’ and kunei ‘here’ are not used on Anuta. ANU does, however, use NP the construction te—nei ‘this—’ and te—na ‘that—’.

In addition to eni and ena, Pawley (1967:271) cites PPN *koo ‘there (yonder place)’, reflected as koo in TON and EUV. I would cite koee ‘over there’ (Te tangata koee ne tu ki runga) as the probable ANU reflex. 7

A second set of diagnostic items involves vowel length. The demonstrative naa in Nuclear Polynesian languages typically employs the long vowel, aa. In ANU, my ear indicates a short vowel.

On a related point, Pawley (1967:274) notes that:

PNP *laa “distant from speaker and addressee”, a postposed marker substituting for *nei “near speaker” and *naa “near addressee” is reflected by all known NP languages. . . .

If so, ANU is not an NP language. ‘Distant’ may be expressed by such - 309 lexemes as na, koee, i ena, or ae, but not laa. 8

With respect to the important issue of personal pronouns, I have equivocated. At times I have represented maua, taua, matou, tatou, naua, and natou with short vowels, suggesting a possible Tongic affiliation, and sometimes as maaua, taaua, maatou, taatou, naaua, and naatou, with long vowels, indicating a Nuclear Polynesian affiliation. In truth, both my ear for ordinary discourse and informants in response to my questions seem to indicate that both forms are used, perhaps reflecting the presence in ANU of both TO and NP elements. 9

Phonemically, Biggs notes correctly that ANU has three reflexes of the PPN *s: s, t, and ø. S is not regarded by informants as proper Anutan. It appears only in words recently borrowed from languages possessing that phoneme—particularly TIK — and in those words it varies freely with t. In ANU, more basic vocabulary items reflect *s as loss of *s than as t; therefore, Biggs concludes that “Anutan lost *s in directly inherited words but borrowed words reflect *s as t” (1980:122). He then argues that t from *s must have been borrowed from an NP language, but says nothing of the possible origin of ø from *s. Since *s is reflected as h in TON and EUV, and since the TON h is reflected in ANU as ø (Biggs 1980:123), is it too far-fetched to suggest the “directly inherited” words in which PPN *s is reflected as ø may be Tongic in origin? 10

Ranby (1982), in his discussion of the dual Anutan reflexes of the PPN *s, shares my view that ANU ø was derived from a language which reflected *s as h—most likely TON or EUV—but argues that t is the directly inherited ANU reflex of both *t and *s while ø is the borrowed form. As Ranby acknowledges (1982:5), evidence for choosing one of these hypotheses over the other is “extremely slender”. For present purposes, it is sufficient to note that, whether ANU ø from TON/EUV h is borrowed or directly inherited, it is a major feature of Anutan. In either case, Tongic shows itself as an important source of ANU.

Biggs fails to mention a set of words (e.g., nima ‘arm’, ‘five’, naua/naaua ‘the two of them’, and natoulnaatou ‘all of them’) in which the usual NP l/r shifts to n in TON and ANU. He does mention the shift from a to o as a Tongic feature (p.121), but the only case he specifically cites is PPN *ma, which is reflected in ANU as mo (p.129). In addition, ANU ruo ‘hole’; puko, a term for several varieties of tree; and toko, indicating human number, come to mind as representing this group. The latter is cited by Pawley (1967:279) as particularly significant.

PNP *fua ‘just’, ‘only’, ‘merely’, ‘at random’, ‘without purpose’ is reflected as fua in Nanumean (NAN), Vaitupuan (VAI), East Futunan (EFU), - 310 and Samoan (SAM) and fuere in Tikopian (TIK) (Pawley 1967:274). 11 This concept is normally expressed as in both Tongan and Anutan. (‘A little bit’ in TON is si'isi'i pē; in ANU it is pakatiitii pē.)


Further illustrations could be cited. It is likely that such information would not alter the final verdict that ANU is predominantly an NP language. However, it does demonstrate a very substantial Tongic input — far more than one is likely to imagine from reading Green's, Kirch's, or even Biggs' analyses. When this is placed next to the relatively modest claims for Tongan contact made by indigenous traditions—certainly less than supposed by Firth (1954), Bayard (1966), Pawley (1967), Green (1971) or Blake et al. (1983)— the apparent contradiction between oral tradition and linguistic evidence begins to disappear.

Still, there remain two problems. The first is Biggs' observation that “virtually all of the identifiable Tongic borrowings in ANU are also to be found in EUV though the converse does not apply” (1980:123), strongly suggesting that Anuta's Tongic features were derived indirectly via EUV.

At first blush, Biggs' argument looks incontrovertible. It may be well to note, however, that the correspondence is imperfect. For example, one might cite PPN *ma, which is reflected in Tongan as mo, ANU as mo, and EUV as both mo and ma. Similarly, Ranby (1982:6, 10) has suggested that PPN *laho ‘testicle’ and *singano ‘kind of pandanus’ are rao and ingano in ANU, laho and hingano in TON, and have no EUV cognates. Ranby's assertion is called into question by data in Rensch's East 'Uvean dictionary (1984:156, 215). Be that as it may, when one adds to this the observation that TON and EUV share between 72 and 86 per cent of their basic vocabulary (Biggs 1980:125), it is hardly surprising that most of the Tongan features found in ANU also appear in EUV. Thus, in the end Biggs (1978:354) concedes, “While there is some indication that the Tongic borrowings in Anutan are from Uvean rather than directly from Tongan it is not, in my opinion, conclusive”.

The second problem may be even thornier. Despite the many points of similarity among ANU, EUV, and TON, ANU clearly resembles TIK more closely than it does any other language. Indeed, Biggs (personal communication) has suggested ANU might best be regarded as a dialect of TIK. Hence, he says:

Anuta was probably settled from Tikopia . . . and borrowed heavily from Tikopian. East 'Uvea was colonised by Tongans perhaps five hundred years ago[;] and subsequently, after their language had borrowed heavily from - 311 Tongan, 'Uvean speakers came to Anuta and, in turn, their Tongan-adulterated language became donor to the language of that tiny isle (1980:124-5).

However, the proposition that Anuta started as a Tikopian colony directly contradicts the version of the island's settlement presented by oral traditions. The solution, I believe, is partially supplied by Biggs two paragraphs earlier, where he observes:

The same linguistic data if applied to the question of settlement of the islands on which our languages are at present spoken, however, are inherently incapable of providing unique, unambiguous answers because any existing linguistic situation could have arisen from any number of different historical circumstances (1980:124).

But what alternative historical circumstances might account for the linguistic situation and still be compatible with traditional accounts?

It strikes me as plausible that a group of immigrants, including Tongans and 'Uveans, arrived from western Polynesia 15 generations ago, and, as oral traditions assert, the voyagers found the island uninhabited. Tikopia had already been inhabited for a long time, and Tikopians were accustomed to making canoe voyages to Anuta. Anutans also soon learned the star paths to Tikopia. If, as I have suggested (Feinberg 1988), voyages have averaged at least one every few years, this would have meant perhaps 100 visits in one direction or the other over the past 15 generations. Voyages were often made by as many as four canoes sailing together, which could mean a transfer of over two dozen persons at a time. Some of the visitors from each community married and settled in the other. Most stayed at least a year or two before returning home. When one considers that Anuta's population has been fewer than 150 people for most of its history, the number of Tikopian “contact hours” per Anutan is immense, and it is little wonder that, over a period of 15 generations, Tikopia has had a tremendous impact on Anutan speech patterns.

To this it should be added that the advent of Government shipping has dramatically increased Anuta/Tikopia contact and Tikopian influence on the Anutan language. In the single decade between my first and second visits to Anuta, the increase in Anutan use of what informants regard as Tikopian words and Tikopian pronunciation of Anutan words increased markedly.

Shipping itself has become more frequent and reliable in recent decades, but, to some degree, it has been available at least since Anuta's conversion to Christianity by the Anglican Church, allegedly in 1916. Thus, Anuta had experienced more than a half-century of intensified Tikopian contact, com- - 312 pliments of European shipping, by the time of Green's and my studies of the language. In precontact and early contact times, ANU may have shown a considerably lower proportion of TIK and correspondingly higher proportion of EUV/TON elements than recent studies indicate.


Anutan oral traditions claim both Tongan and 'Uvean ancestry. Common wisdom among Pacific archaeologists and historical linguists accepts claims of 'Uvean ancestry but seems to reject the idea of Tongan contacts. Since archaeology itself apparently is silent on the issue, and comparative ethnography shows striking similarities between Tongan and Anutan social structure, the conclusions of culture historians depend on the evidence of historical linguistics.

That evidence, indeed, appears to show Anutan as a Nuclear Polynesian language. On the other hand, it also indicates extensive Tongic influence. While it is possible that this influence has been effected solely through 'Uvean contact without any Tongans ever having been present on Anuta, however, that is not an inescapable conclusion. In short, I see Anutan oral traditions, in their major outlines, as tenable in terms of linguistic and archaeological data, and if there is no specific reason to disbelieve them, I submit that they should be at least provisionally accepted.

In a sense, I find it ironic to be arguing this point under these circumstances. Kirch, in collaboration with Douglas Yen (1982), has made a persuasive case that Tikopian oral traditions provide a remarkably accurate account of events on that island over the past several hundred years. They even demonstrated the plausibility of Tikopian legends claiming Tongan contact. But if, as Kirch and Yen suggest, we should take Tikopians' versions of their island's past as largely sound historical accounts, can we refuse to pay Anuta similar respect?


I am indebted to Niko Besnier, Jacob Love, and Karen Watson-Gegeo for helpful comments on an earlier version of this manuscript. My studies of Anuta over the past two decades have been supported by the U.S. Public Health Service and Kent State University's Research Council. Issues addressed in this article will be discussed at greater length in a forthcoming volume on Anutan oral traditions.

- 313

(This and Appendix B are transcriptions and translations of two versions of Anuta's initial settlement from the Polynesian Triangle. The first account is taken from a longer narrative dictated by Pu Nukumarere in 1972. The second is part of an account recorded on tape in 1983 by Moses Purianga. I have given a phonemic rather than phonetic transcription, omitting obvious mistakes which the narrators themselves corrected. I represent vowel length in accordance with what seems to be the preponderance of opinion as to correct pronunciation, but, for reasons cited above, it should not be taken as definitive. Tikopian terms used by my informants have been retained in the transcriptions, but they are italicised to differentiate them from terms regarded as proper Anutan.)

- 314

(This account begins with the story of the autochthonous apukere, who are said to have met their demise as a result of an altercation with the Tikopian chief Pu Ariki. The latter went back home to Tikopia; then returned to bury his defeated rivals. After sailing back again to Tikopia, he returned once more to Anuta.)

Pu Ariki ne au. Ne poki mai. Noporaki nga Tonga mo nga Uvea. Ko nga Tonga mo nga Uvea ne ipo, tau vaka rua. Take vaka nga Uvea; take vaka nga Tonga. Te ingoa o te vaka o nga Tonga, Kavakiteuta. Te ingoa o te vaka o nga Uvea, Pirikiuvea.

(After describing the encounter between Pu Ariki and the interlopers [see above], Pu Nukumarere continued as follows:)

Te vatia naatou ne o mai, kairo e ngokai i te penua nei. Ko naatou ngokai ne au mai, ku o mo. Ko naatou ne noporaki i te penua nei, oro rea o keri. Kairo ni mea ne to ki te kere paia kairo ni ngokai i te penua nei. Ne keri te ao nei. Apongipongi, oro o mamata ki ei, ki naatou kere. Ko nga taro ku mauri. Mo nga puti. Mo nga uupi. E toru nga ngokai. Nga taro mo nga puti mo nga uupi. Kairo e tai mea ne to i ei. Ne apu pero mai te kerekere.

Erua naatou atua. Nga Uvea, te ingoa o te atua o nga Uvea, Tokitaaitekere. Te ingoa o te atua o nga Tonga, Putiuraua. Te tino o te atua o nga Tonga, ko te toke. Te tino o te atua o nga Uvea, ko te moko.


Pu Ariki came. [He] returned. The Tongans and 'Uveans were dwelling [on Anuta]. The Tongans and the 'Uveans descended [in] two canoes. One canoe [was that of] the 'Uveans; the other [was that of] the Tongans. The name of the Tongans' canoe was Kavakiteuta. The name of the 'Uveans' canoe was Pirikiuvea.

When they arrived, there was no food on this island. The food they brought had spoiled. They dwelt on this island [and] and went to dig [their gardens]. Not a thing was planted in the ground because there was no food on this island. They dug today. Tomorrow they went to look at it—at their land. The taro had sprung to life. And the bananas. And the yams. The foods are three. The taro and the banana and the yam. No one planted it. [They] simply sprang up from the soil.

Their gods are two. As for the 'Uveans, the name of the 'Uveans' god was Tokitaaitekere. The name of the Tongans' god was Putiuraua. The body of the Tongans' god was the moray eel. The body of the 'Uveans' god was the lizard.


Ko te vaka, e ati na ingoa ko te pua rua. Te pua rua, pua rua a nga Tonga. Ne oro o tau aere i nga penua i te Atu Runga. Te vaka nei ne pakaoko ki te penua e ati na ingoa ko Nuui. Oro o tau ia rei i ei.

Toka rua raua e oro rau vaka nei. Take tangata ko Pare. Na toko rua, ko Pu Kaurave. Pu Kaurave ko te Tonga. Ko Pare te Uvea. Toko rua nei pe tau aavanga ki rau kave.

Toko rua ne aavanga ki rau kave ne pe tauiaki. Oro toko rua nei tau i ei. Ne rapirapi rei ke pakangaromia—ke pakarekutia—te kanopenua ko Nuui. Teaa. Kae ravaatia ko te tangata na ingoa ko Pu Kaurave.

Ravaatia. Patiia na raakau. Pokitakinaria na nopine pe ki te riuara. Kae rere mai te tau o te kanopenua. Karanga atu rea na nopine, ko ia ka rere o aru. Ka aru o aa? Naatou rea ka? Kae karanga atu rea Pu Kaurave ki na nopine, “Toku raakau ku mapati ee rea.”

“A ko tea te mea e tu i tou tua na? Tau raakau ko ena e tu i tou tua.”

Teaa, popoo ake te tangata, ko Pu Kaurave; popoo ake rea ki na raakau e tu i na tua. Teraa, - 315 poki rei poki o tau. Tau tau tau. Piripiri rei, rapirapi ke pakarekutia kanopenua. Kae muna rei te kanopenua i a tuku. E tuku te tau.

Teraa, karanga atu rea ko te paaoa, te kau vaka o Pu Kaurave, ke taui mai raatou ora. Taui mai rei ko te tipa. Te tipa kura.

Teraa, pare pare te paaoa nei. Pakaui rea. Ka oro ki te penua, e ati na ingoa ko te penua i te Atu Ellice. Ka oro poki tau aere i ei.

Motutia rei te apaa. Te apaa nei ne oko. Ne maatea rei. Muna atu rei na maa, te Uvea ae, “Te vaka papare ke tou oro. Te vaka ka repetia.”

Peepee te taumuri ki te matangi. Kae rere aa rei ki te murimatangi. Karanga atu rea ko na maa, te Uvea, ko Taupare, “Taatou vaka na e pakauu rea ki te petuu ae tu mai na, e ati na ingoa ko Manu.”

Teaa, ko te mataavaka ku pakauu rea te petuu e ati na ingoa ko Manu. Pena ipo te vaka nei. Rere rea te penua ko Anuta.


The canoe is named te pua rua ‘double-hulled canoe’. The pua rua was a Tongan pua rua. [Its crew] went on the warpath among the archipelagos of Polynesia. This canoe made landfall at the island known as Nui. They went to battle there.

The two of them proceeded in their canoe here. One of the men was Pare. His counterpart was Pu Kaurave. Pu Kaurave was a Tongan. Pare was an 'Uvean. These two were married to each other's sisters.

The two of them married each other's sisters, who were exchanged. They two went to make war there. [They] almost wiped out Nui's population. So it was. And then the man named Pu Kaurave was defeated.

Defeated. His club was broken. His wife just returned on the path as he ran from the battle of the island's population. His wife said to him that he was running away [and asked] what [he] was going to do. What will become of them? Then Pu Kaurave spoke out to his wife, “My club has utterly broken”.

“What is the thing standing at your back there? Your club is there behind you.”

Thus, the man, Pu Kaurave, reached up; reached up towards the club standing behind him. Thus, [he] returned again to battle. Battle battle battle. [He] almost, just about wiped out the island's population. Then the population said to cease. To abandon the fight.

Thus spoke out the people [constituting] Pu Kaurave's crew, [suggesting] that [the people of Nui] ransom their lives. [They should] pay a pearl-shell. A shiny pearl-shell.

Thus, these people stayed and stayed. Finally they cast off. They will go to the island known as an island in the Ellice Archipelago. They will also go on the warpath there.

A storm split everything apart. The storm arrived. It was tremendous. His brother-in-law, the 'Uvean, here said, “We must turn the canoe around and leave. [If we do not] the canoe will be broken to pieces.”

[They] flung the stern towards the wind. Then they ran downwind. His brother-in-law, the 'Uvean, Taupare said, “Let our canoe there enter the star that stands before us [and] is known by the name Manu ‘Bird’ [Sirius].”

Thus, the bow was directed into the star known by the name Manu. This canoe proceeded downward. The island of Anuta sped [upward].

  • Bayard, Donn, MS 1966. The cultural relationships of the Polynesian Outliers. Unpublished M.A. thesis, University of Hawai'i.
- 316
  • Biggs, Bruce, 1978. Review of The Anutan Language Reconsidered: Lexicon and Grammar of a Polynesian Outlier by Richard Feinberg. Journal of the Polynesian Society, 87:353-5.
  • ——1980. The position of East 'Uvean and Anutan in the Polynesian language family. Te Reo, 23:115-34.
  • Blake, N. M., B. R. Hawkins, R. L. Kirk, K. Bhatia, P. Brown, R. M. Garruto, and D. C. Gajdusek, 1983. A population genetic study of the Banks and Torres Island (Vanuatu) and of the Santa Cruz Islands and Polynesian Outliers (Solomon Islands). American Journal of Physical Anthropology, 62:343-61.
  • Burrows, Edwin G., 1937. The Ethnology of Uvea. Honolulu, Bernice P. Bishop Museum, Bulletin 145.
  • Feinberg, Richard, 1977. The Anutan Language Reconsidered: Lexicon and Grammar of a Polynesian Outlier. 2 Vols. New Haven, Connecticut, Human Relations Area Files Press. HRAFlex Books.
  • ——1981. Anuta: Social Structure of a Polynesian Island. Laie and Copenhagen, Institute for Polynesian Studies in Cooperation with the National Museum of Denmark.
  • ——1982. Some observations on a Polynesian naming system: Personal names and naming on Anuta. Journal of the Polynesian Society, 91:581-8.
  • ——1988. Polynesian Seafaring and Navigation: Ocean Travel in Anutan Culture and Society. Kent, Ohio, Kent State University Press.
  • Firth, Raymond, 1954. Anuta and Tikopia: Symbiotic elements in social organisation. Journal of the Polynesian Society, 63:87-131.
  • ——1985. Tikopia-English Dictionary: Taranga Fakatikopia ma Taranga Fakainglisi. Oxford and Auckland, Oxford University Press and Auckland University Press.
  • Goldman, Irving, 1970. Ancient Polynesian Society. Chicago, University of Chicago Press.
  • Green Roger C., 1971. Anuta's position in the subgrouping of Polynesian languages. Journal of the Polynesian Society, 83:427-42.
  • Haddon, A. C. and James Hornell, 1975. Canoes of Oceania. Honolulu, Bernice P. Bishop Museum Press. Reprint of 1936-8 edition.
  • Kaeppler, Adrienne, 1973. A comparative note on Anutan social organisation, in D. E. Yen and Janet Gordon (eds), Anuta: A Polynesian Outlier in the Solomon Islands. Pacific Anthropological Records, Number 21. Honolulu, Department of Anthropology, Bernice P. Bishop Museum. pp.21-4.
  • Kirch, Patrick V., 1982. A Revision of the Anuta Sequence. Journal of the Polynesian Society, 91:245-54.
  • ——1984. The Polynesian outliers: Continuity, change, and replacement. Journal of Pacific History, 19:224-38.
  • ——1985. On the genetic and cultural relationships of certain outlier populations. American Journal of Physical Anthropology, 66:381-2.
  • ——and D. E. Yen, 1982. Tikopia: The Prehistory and Ecology of a Polynesian Outlier. Honolulu, Bernice P. Bishop Museum, Bulletin 238.
- 317
  • Lewis, David, 1972. We, the Navigators. Honolulu, University of Hawai'i Press.
  • Pawley, Andrew K., 1966. Polynesian languages: A subgrouping based on shared innovations in morphology. Journal of the Polynesian Society, 75:39-64.
  • ——1967. The relationships of Polynesian outlier languages. Journal of the Polynesian Society, 75:259-96.
  • Ranby, Peter, 1982. The dual reflexes of Proto-Polynesian *s in Anuta. Te Reo, 25:3-11.
  • Rensch, Karl H., 1984. Wallisien-French Dictionary. Canberra, Department of Linguistics, Research School of Pacific Studies, Australian National University.
- 318
Pacific Studies

a journal devoted to the study of the Pacific—its islands and adjacent countries


Anthropology·Folklore·History·Sociolinguistics·Political Science·Sociology

Vol. 12, No. 3 (July 1989)

Char Miller, Rumors and Language of Social Change in 19th-Century Hawaii; A.H. Angelo, H. Kirifi & A. Fong Toy, Law and Tokelau; David Wetherell, “The Bridegroom Cometh”: Queensland Melanesians in New Guinea; L.J. Gorenflo & Michael Levin, Demographic Evolution of Ebeye; James Howe, Julia Hecht & Caroline Ralston, Book Review Forum: Making History: Pukapukan and Anthropological Constructions of Knowledge, by Robert Borofsky

Pacific Studies is published three times each year. Subscriptions are US $20.00 per volume; back issues are $7.50 each. Checks or money orders may be addressed to The Institute for Polynesian Studies, Brigham Young University-Hawaii, Box 1979, Laie, HI 96762 U.S.A.

1   Actually, my genealogies (see Feinberg 1981:138, 213-28) show 15 rather than 12 generations since initial settlement. With regard to the putative homeland, Anutan v is pronounced as w when following a u; thus Uvea would be pronounced Uwea. Therefore, I believe Uea to be an understandable misrendering of Uvea.
2   The name Uvea, in this article, refers to the legendary homeland cited in Anutan oral traditions. 'Uvea, written with the glottal stop, refers specifically to the island that we know as East 'Uvea or Wallis Island. While these two names may ultimately refer to the same place, I think it is important to keep them analytically distinct.
3   See Feinberg (1981:218; 1982) for a discussion of Anuta's naming system.
4   Takaraua is clearly a Tongan name (cf. one of Tonga's major chiefly lines, the Ha'a Takalaua).
5   Whether these omissions have resulted from lack of clarity in my writing or lack of care in Biggs' reading of my work need not be resolved for present purposes.
6   Besnier (personal communication) informs me that mo appears in at least a few NP languages, and therefore is not a definitive Tongic marker. Several other markers commonly assumed to be distinctive of TO, Besnier also asserts to be present in some NP languages. These include a short a in the demonstrative na and most personal pronouns, and the final o (as opposed to the NP a) in ruo ‘hole’, puko, the name for several types of tree, and toko, indicating human number. A proper critique of the criteria for distinguishing NP from TO languages, however, is beyond the present contribution's scope. Thus, while recognising that Pawley (1966, 1967) may not be the last word on the subject, for present purposes I rely upon the features he has cited for discriminating the highest order PN subgroups.
7   I also have suggested ei ‘there’ (Kau oru ki ei; Kau o mai ei) as a possible ANU reflex of PPN *hee ‘there (distant)’ which is reflected in TON and EUV hee ‘there (distant)’ and in Niuean (NIU) as ee ‘this’, ‘these’. Besnier (personal communication) has argued plausibly to the contrary, that ANU ei is an anaphoric pronoun, cognate with ai in both TO and NP languages.
8   In this respect, ANU differs from other NP languages with which I am familiar. For example, Tikopian uses raa and Nukumanu laa to express the same concept.
9   I should acknowledge here that a short a by itself may be insufficient to establish Tongic derivation for matou/maatou and tatou/taatou and natou/naatou (cf. TON kimautolu, kitautolu, and kinautolu, respectively). On the other hand, the n in naua/naaua and natou/naatou would appear to be Tongic. Also, ANU and TON share the identical first person singular personal pronoun: (ko) au. The SAM equivalent, a'u, would be aku in ANU.
10   If these words had been borrowed from a language reflecting PPN *s as s, a language like the Nuclear Polynesian TIK, it probably would have been borrowed as t. In fact, in 1983 I heard people say ato ‘day’ and then explain to me that they were using Tikopian (cf. ANU ao). Actual Tikopian pronunciation is aso.
11   Pawley renders TIK fuere as fuare (see Firth 1985:137).