Volume 105 1996 > Volume 105, No. 3 > Legend and history: Did the Vanuatu-Tonga kava trade cease in A.D. 1447?, by David Luders, p 287-310
                                                                                             Previous | Next   

- 287

This paper has two purposes. One is to propose that myths and legends were, sometimes at least, historical records clothed in layers of allegory. The other is to demonstrate this by interpretation of legends recorded by Firth on Tikopia in 1928 (Firth 1970:216-8, reproduced in the Appendix to this paper). These legends appear to record that there was a trade in kava between Vanuatu and Tonga which terminated in the mid-fifteenth century, perhaps in A.D. 1447, five years before the explosion of the island of Kuwae in A.D. 1452 (see Discussion below).

Kava is of Melanesian origin. Recent work places the most probable site of its domestication in northern Vanuatu (Crowley 1994:92; Lebot, Merlin and Lindstrom 1992:51-3).

There are many myths of the origin of kava. Several of them have common threads, suggesting common origins. Lebot, Merlin and Lindstrom (1992:121-9) discuss a collection of such myths and attempt some interpretations. These attempts seem to be based, despite a reference to history, on a broad assumption that the myths are no more than tales of how kava came to be. In a sense, they skirt the bolder possibility that they are fragments of codified history.

If, however, one supposes that the stories are not fables but legends recording history in an allegorical code, interpretation acquires a different context. It then becomes a matter of cracking the code. For this to be done it is necessary to compare the legend with known historical facts, but that is not all. In the case of Polynesian legends at least, one must give full allowance for the role of metaphor, allegory and multiple meaning which evidently was integral to the genre of legend-making.

In many cases the records of the legends which we have are rather bald, shorn by abbreviated recording of much of the detail, nuance and context which we can presume them once to have had. A notable exception is those recorded by Firth. His record captures a good deal of the style of the telling, with its repetitions subtly changed to suggest secondary meanings and interwoven contexts. Read sympathetically, it comes alive with meanings, not the least of which is a demonstration of the style and method of the original legend-maker. One can also hear Firth's informant speak, repeating - 288 words faithfully memorised; that skill was the other essential component of the art-form of the legend.

For all that, it is technical science which has provided clues crucial to the interpretation of Firth's record as being one of a codified history. These clues are two. The first is the technical information on the origin of kava, vital for any analysis of such legends. The second is the data on the volcanic explosion of the island of Kuwae in central Vanuatu in A.D. 1452. This event was, I contend, the basis of the complex of metaphor which is the climax of the legends and the history they portray. Other information, relating to the ethnology of Vanuatu, also contributes to the interpretation. To interpret the legends properly we must be acquainted with all this information.


It is to the agronomist Lebot and his various collaborators that we owe the elucidation of the origin of domesticated kava. Lebot's work in the early 1980s in Vanuatu produced strong circumstantial evidence for supposing that domestication of kava occurred in Vanuatu and subsequent publications, in particular Lebot, Merlin and Lindstrom (1992), bring the work to a conclusion. In the latter publication the evidence is more direct and powerful than that presented by Brunton (1989), who argues that kava might have been domesticated elsewhere in Melanesia and seems to favour Papua New Guinea in this respect.

From the beginning, Lebot showed that there was a much greater range in Vanuatu than anywhere else of kava varieties and that these varieties were usually more potent chemically than those from elsewhere (e.g., total kavalactone analyses of Vanuatu kavas were two to five times that of common Fijian varieties). The chemical analyses were also quite diverse. Equally significantly, the name for kava in local languages was as diverse as nigui (Hiw, Torres Is), maloku (Marino, Maewo), mele (Sa, South Pentecost), bir (Tur, Santo), hae (Malo), nimvulum (South West Bay, Malekula), nikawa (Kwamera, Tanna), kava (Aneityum) to select some (Lebot and Cabalion 1986:83-93). This suggests an origin more ancient than in Fiji or Polynesia. Only the last two of these names are cognate with the Polynesian kava. These occur in the south and it is quite likely that they were introduced there from Polynesia (Crowley 1994:95; Lebot, Merlin and Lindstrom 1992:52).

Lebot, Merlin and Lindstrom present the results of extensive analysis of many kavas using three means: morphotypes, chemotypes and zymotypes. Morphotypes are essentially varieties, distinguished by physical characteristics. Chemotypes are groupings made on the basis of the chemical analysis of the kavalactones, the active substances in kava. - 289 Zymotypes are distinctions made on fundamental genetic characteristics; this kind of analysis can be loosely termed genetic finger-printing.

Although it can be said that these studies were not exhaustive of every kava extant, they range so widely and so much further than any other that their results are compelling. They reach a conclusion almost impossible to deny.

Some two-thirds of the morphotypes studied are found in Vanuatu and nowhere else. Of the eight chemotypes, six are present in Vanuatu and in no other place are there more than three: in Samoa alone were three recorded; in the rest of Polynesia one or two, and in Fiji one. The distribution of zymotypes is still more persuasive. There are ten of them, of which seven are Piper wichmanii (“wild” kava) and three are Piper methysticum (“cultivated” kava). Five of the “wild” kavas were found only in Papua New Guinea. All but one of these are genetically remote from domesticated kava. The two “wild” kavas not found in Papua New Guinea are found in Vanuatu (and one is found in the Solomon Islands) and these are genetically the closest to “cultivated” kava. No “wild” kavas were found in Polynesia or Micronesia.

The distribution of the three “cultivated” zymotypes also points to domestication's having occurred in Vanuatu. One occurs only in northern Papua New Guinea and is of minor significance. The other two occur in Vanuatu. One of these appears also in southern Papua New Guinea. The other is the sole genetic type occurring in Fiji, Polynesia and Micronesia.

It is difficult to find any grounds for doubting the conclusion of Lebot, Merlin and Lindstrom that kava was domesticated in Vanuatu. The kava to be found in Fiji and Polynesia is the result of a thin trickle of varieties out of Vanuatu. This reflects the testimony of Firth's record. Just where in Vanuatu domestication occurred is still conjectural but the evidence points to northern Vanuatu, possibly Maewo island.


Until A.D. 1452, Kuwae was an island about 75 km long and about 15 km wide, in what is now central Vanuatu 1 (see Fig. 1). Then, twin gigantic jets of highly pressurised magma, each perhaps two kilometers in diameter, blew off the overlying mantle of rock and earth from a site east of Kuwae's centre, breaking the island into pieces (Monzier, Robin and Eissen 1994:207, fig. p. 209). Eissen, Monzier and Robin (1994) place the date at 1452 ± 1 on the basis of data by Delmas et al. (1992) and Hammer et al. (1980), these being separate analyses of ice cores, the former from the South Pole and the latter from the Greenland icecap. These analyses register dust particles and acidity with precision. It was

- 290
Fig. 1. Map showing Efate/Shepherds region, the approximated outline of the former Kuwae and the location in Vanuatu.
- 291

evidently on the conclusions of Delmas et al. (1992:344) that Eissen et al. felt certain that the date could be pinpointed. Delmas et al. date the unspecified event they recorded to A.D. 1452. Eissen et al. observe that the Greenland ice cores and other northern hemisphere data show that cold conditions occurred in 1453 (“un hiver rigoreux”) and, one might say conservatively, put the date at 1452 ± 1.

This was a colossal eruption. It was one of the eight greatest volcanic events in the past ten thousand years. At least thirty million cubic metres of rock, earth and magma were hurled into the atmosphere at an initial velocity of about 300 kilometres per hour and another vast volume slid into the sea. The dust-pall circled the globe, initially in the southern hemisphere and later in the northern hemisphere. The polar ice cores record that it persisted for at least three years (Delmas et al. 1992). The resultant blotting-out of the sun reduced temperatures so as to produce a minor ice-age with the result of widespread famine, evidently on a global scale.

Pang (1993) has collected from various sources evidence of climatic disruption which he considers attributable to the results of the explosion. The most dramatic of these concern the fall of Constantinople, besieged by the Ottoman Turks, in May 1453, which ended the Byzantine empire. The city's gardens produced little that year and in its last days exceptional hail, rain and fog evidently made it almost untenable; strange “lights” were seen and interpreted as omens. Other data for 1453 include western U.S. bristlecone pines showing frost damage, snow damaging Chinese wheat crops and causing thousands to freeze to death and abnormally narrow tree-rings in England. In the years immediately following, Swedish corn tithes fell to zero, German wine was “bad”, French and Finnish trees were “stunted”, Chinese cypresses had narrow rings. In 1454 “it snowed for forty days south of the Yangtze river … countless died of cold and famine”; the river Huai and Taihu (Great Lake) froze. Hassig (1981) also discusses a famine in central Mexico, described in Aztec records circa 1454, but it is not clear that this famine can be attributed directly to Kuwae's effects.

The site of the blast is now beneath the sea, between the islands of Epi and Tongoa (Fig. 1). It remains intermittently active. Before the explosion it was covered by the land which joined these two islands and three other outcrops.

Monzier (personal communication 1994) has described the local effects the event must have had. Those relevant to the interpretion of Firth's record of legends are as follows. Land areas of the erstwhile Kuwae would have been scorched at about 300°C and denuded, exposing red volcanic earth. Much of the ash which then fell on it would have been washed off in subsequent heavy rains. Revegetation occurred slowly. A large area of surrounding sea would have been heavy with floating pumice and ash. The - 292 blast would have been heard as far away as Australia (and so, probably, Tonga). The dust pall would have reddened the sky and by reflection the sea for at least three years following the event; this would probably have been most pronounced in early mornings and late afternoons. There was almost certainly a period of some years of increasing seismic shakings prior to the eruption as the covering mantle came under increasing pressure from the chamber of magma and gas below.


Ethnological data from the Efate-Shepherd Islands region of Vanuatu, of which Kuwae was part, are relevant to the interpretations of the legends which follow. They show imprints of Polynesian influence and it is quite possible that some of this influence aided the dispersal of kava into Polynesia. Aside from the recognised Polynesian outliers in this region (Emae, Ifira/Mele) there is rather firm evidence that its political structure derived from Polynesia, dating back as much as 1200 years. The most extensive published information available from the area is that of Guiart (1973) and Garanger (1972, 1982).

From the mass of detail which Guiart recorded in the region about 1960, it is clear that the chiefly system there is hereditary, primogeniturial and hierarchical. He records in a number of cases lengthy genealogies and aspects of the tributary relationships but his data are incomplete, partly because information was withheld. 2

The political form is best preserved on Tongoa, Emae and Tongariki. There, dominant titles are attached to land which is “rented” to nakainang (Namakura, Tongoa/Shepherds ‘subjects’) for tribute in produce. Titles are primogeniturial but male or female relatives may hold them as regents. There is a plethora of titles, some attached to land, some functional as part of a chiefly “court”, some in the nature of consular functions for a higher chief. Tributes are in more than one form, one of which is for land use. (Guiart records only nasaotonga (Namakura, Tongoa/Shepherds) tribute in pigs symbolising fealty and usually payable once in a lifetime.)

This pattern, distinctly different though it is from the graded Melanesian social system abutting it to the north and the less structured hereditary pattern to the south, is not the only indication of Polynesian infusion. Some chiefly titles suggest Polynesian or Fijian forms; Ti Nambua is cognate with Tui Nabua of Naqalimare, Nadroga, Fiji (Geraghty, personal communication 1994). Maraki Pule and Marikitapu might be contractions of old Polynesian forms, Ma Ariki Pule, Ma Ariki Tapu (Ma is a term of respect in the Shepherd Islands). Many others are prefixed Ti (cf. Tu'i) and/or suffixed Ariki, Mata, Muri, Pule, Rangi. Moreover, - 293 lengthy genealogies were preserved - a trait scarcely to be thought Melanesian - and a number of these are attached to histories giving origins from “elsewhere”. Although the latter are evidently adulterated, one cannot quite dismiss the Polynesian suggestions in them because they contrast markedly with neighbouring myths giving local origins.

Other indications of a Polynesian heritage are various and diffuse. Broad waist-mats are chiefly regalia; tapa-making was practised (though might possibly have been adopted); attire was the pubic mat (distinct from the male penis-wrapper or nambas to both the north and the south); young males, on Tongoa at least, were supercised in the manner described by Mariner in Tonga (Martin 1817, vol I, footnote to p. 340). Some words suggest infusions of Polynesian chiefly and navigational vocabulary, e.g., nakainanga (Nakanamanga, Tongoa), ‘subjects of a chief’, is cognate with the Tongan kainanga of similar meaning; tokelau (North Efate, Nakanamanga, Namakura) for ‘north-east wind’. Legends recorded strongly suggest layered allegorical chiefly histories (and that this manner was employed is confirmed by my informants); they include such forms as “crab” as a euphemism for a human sacrifice (Guiart 1973:189). Songs recording political events were often quite complex (e.g., one of six and a half minutes, sung by a Tongoan chief, which I recorded in 1994 and which tells of a conflict on Efate). Many such were constructed around key words, the knowledge of which was the proof of ownership of the “inner layer” of the history thus recorded.

These are examples of many suggestions of Polynesian vestiges in Efate-Shepherds culture. They add to the inferences which might be drawn from Garanger's discoveries.

Garanger excavated a remarkable burial on Retoka, a small island off Efate's west coast. A high personage was buried with some 30 retainers, evidently of both high and low rank. Garanger dated this burial to circa A.D. 1265 but the dating is now subject to a degree of doubt. Determinations made on bone collagen in the 1960s are now thought to have been inadequate in the removal of sources of contamination; checks are now being run (Spriggs personal communication 1996). The grave goods from the burial suggest a blending of Polynesian and Melanesian practices. They mix ornaments of whale's tooth with pig tusks, trochus shell bracelets and shell beads. The last were worn about the waist and ornamented pubic mats and upper arm bands.

Garanger also excavated a burial on Tongoa island believed to be that of Ti Tongoa Liseiriki, the hero of the recolonisation of Tongoa after the eruption of Kuwae. Supporting his interpretation of this excavation he reiterated Guiart's rendition of the legend of the explosion.

In 1994 I recorded substantially the same legend, with the verifying songs. Briefly, the legend is that the catastrophe was caused by the - 294 actions of a young man whose parents had been refugees from an erupting island to the north (Lopevi, then called Lumbarae). His mother had been reduced to prostitution and he was tricked into sexual congress with her. In his shame, he determined to wreak revenge by destroying the island, using magic from their home island. He held feasts on successive days, each time killing a pig and preserving the pig's bladder, dried and inflated. On the sixth day, at the final feast, he burst each bladder in turn, causing successively more violent earth movement. On his bursting the final bladder, the volcano erupted from beneath him. The second part of the legend deals with the adventures of a second youth who survived the explosion and in due course became Ti Tongoa Liseiriki.

This account provides a clue to a metaphor employed in both this and Firth's second legend: a day (or night) stood for a year. The 83-year-old aore/manuvasa (Namakura, Tongoa/Shepherds ‘songmaster/historian’) who gave me the legend took literally the daily feasts (as he had been trained to do) and was thus most reluctant to think that any could have had time to flee the island. This contrasted with the remarks of a kinsman of his, a “retired” holder of a higher title. This man, on listening to the recording, commented that it was accurate except that the span of time was not in days but in years. What makes this contention significant is that this informant then went on to give the history of his chiefly line, a history which includes the chief taking his people from Kuwae to Emau island prior to the eruption and taking with them their two treasured breadfruit varieties. These were then grown on Emau and later brought back to Tongoa, the remnant of Kuwae which had been their former home. Five or six years of warnings make this plausible; five or six days are the legacy of an allegorical style of layered legend like that of Polynesia.

That chiefly history spanned 48 generations, of which 26 were said to precede the Kuwae explosion and 22 to follow it. Allowing for olioli (Namakura, Tongoa/Shepherds ‘regents’), the latter figure fits the period from A.D. 1452 quite well. The history recounts migrations, giving reasons for each move, from the eastern point of Efate to the final home on Kuwae/Tongoa. The detail of the history, in common with all in the region, is held confidentially. It was given to me in trust. Guiart was given a truncated version.

Guiart's records of genealogies except for one covered about fifteen generations. These date from a great conflict on Efate (which probably occurred circa A.D. 1600), when the chiefly system there seems to have been overthrown, with the result that the chiefs and their courts fled to the Shepherd Islands. The exception (Matariliu of Panita, Tongoa) claimed 36 generations, but without names. This suggests an arrival on east Efate circa - 295 A.D. 950. By the same reckoning, the 48-generation history might descend from an arrival circa A.D. 800. It is noteworthy that almost every history citing arrival from elsewhere gives the area near the eastern tip of Efate as the place from which the (local) history began. One might postulate that this was a landfall used for anything up to 800 years.

It is Garanger's excavation on Retoka which gives substance to the strands of information about the period before the Kuwae explosion. The excavation has remained enigmatic but it certainly suggests the developed expression of an extended period of melded cultures. If that were so, Tongans, as in Firth's legends, might have found conditions more conducive to trade than they would have on islands purely Melanesian.

The Polynesian outliers in the region are unlikely to have been the basis of these legacies. The chiefly pattern suggests otherwise. On Emae, the outlier community is in the minority. The history of its dominant chiefly title dates only from about A.D. 1600 and the songs recording chiefly history are not in its Polynesian language but in one of the languages of the Shepherd Islands. This community appears to have been absorbed into the local political framework well after the explosion of Kuwae. The Ifira/Mele community is even less likely to have exercised influence. Its chiefly and social system is like that of the rest of Efate - that is, a remnant one apparently dating from after the great conflict. This community appears to have been closely confined, politically and geographically, by the surrounding ones.


Before proceeding to an interpretation of the legends recorded by Firth (see Appendix), it is well to consider some aspects of their context and to examine some of the crucial metaphors employed.

It is fortunate that Firth records two versions, that they were given by the same man and that the second was given about a month after the first. The similarities and differences between the two and the delay in revealing the second suggest strongly that the first was one for a more general and the second for a more restricted audience. Pa Rangifuri presumably decided in the intervening month that the more confidential version should be divulged. The first gives a version of the widely-recorded elements of kava growing out of a buried body, a rat chewing the kava and so forth and then trails off to “the Tongans” taking (leaves of) kava to “their chief” to drink. The second dispenses with the preliminaries and launches into the detailed allegorical history. The two renditions overlap only in that the Tongans took the leaves of kava to “their chief”/“Ti Tonga” to drink and that he accepted it.

- 296

This duality reflects what can be seen elsewhere. One version is provided for the general populace, another is available to those closer to the seat of power. The latter is probably the second defence against revelation of the secret inner story - the simple tale of the real politics or history being recorded. Many governments do this, whether they be chiefly, monarchical or democratic. There is nothing odd about it. Three layers are common. In democracies, ministers make public utterances, senior officials know more than is revealed but are not privy to all that determines policy. In a society with oral records of history and politics, a multiple-layered allegory is a natural tool and that the “outer” layer should employ the supernatural is only to be expected.

Thus, Firth's two stories appear as a fairly full but possibly incomplete rendition of a layered historical account. It came from Tikopia and, as Firth is careful to note, was “probably traditional in Tikopia and not a modern borrowing from some visiting Tongan”. We may take him at his word but whether we do or not, the ultimately Tongan origin is unmistakable (see, text of Appendix and Discussion below). Tikopia was, after all, peopled partly from Tonga (Kirch and Yen 1982: 341,342). Traditions are often preserved on the fringes of a culture while they alter at its changing centre, and Firth's Tikopia was a cultural backwater.

But also we know of Tongan influence as far away as the Loyalty Islands (Guiart 1996:56, footnote 35) and it is quite improbable, given the practicalities of sailing, that Tongans did not also contact the intervening islands, Vanuatu. This is of some importance in considering the locale of Firth's stories. He observes cautiously that the first gave the origin of kava in Tonga and in the second his informant “gave due priority to Tikopia”. He reports kava “coming from this land” and “originating in the Female Deity, Pufafine”. Only once is Tikopia specifically mentioned. How much to place on this we do not know, but Firth was a painstaking recorder, and we have entered a world of metaphor. If Tongans ranged over these islands, they would probably have stitched them together into a conceptual entity — Firth's “this land” — much as Europeans did the “New World”, the “Spice Islands” or simply, the “East”. As we will see, Pufafine stood for Kuwae, which the Tongans might then have regarded as being as much part of “this land” as Tikopia. Thus the Tikopian legends, derived from Tonga, referred to the place of origin not specifically as Tikopia but more likely as this region of Melanesia; the one reference to Tikopia is an understandable lapse, if such it is, after 500 years.

In that context, Firth's two tales are easier to understand and interpret than in isolation, as Firth perforce tried to do. Yet they make no real sense without a comprehension of the critical metaphors. These are the keys of the “code”.

- 297

I am indebted to the linguist Michel Aufray for his opinions on the imagery employed in Pacific legends. In particular, the rat was symbolic of inland people and seabirds of coastal or voyaging people. (The metaphor is sometimes taken further and the rat's whiskers stand for the beards of the inland, Melanesian, people.) The analogy is immediately clear. In a Tongan legend, a man seeing a rat chew the kava plant is simply the discovery that Melanesians used kava (and perhaps also that they prepared it by chewing). His subsequent adoption of kava follows naturally. That rats do in fact chew growing kava — the stems and less frequently the bases — does not disavow this interpretation. Rather it attests to the artfulness of the makers of legends.

This metaphor, which may have been a standard one as Aufray implied to me, helps to anchor the legendary origin of kava in Melanesia where the technical data so firmly puts the actual origin. More potent though is the interpretation of the complex of metaphors relating, in the second of Firth's legends, to the Female Deity Pufafine and her relations with “the man”.

If we take this Female Deity to stand for the whole of the place, a conceived mystical integration of land, chief, people, ancestors, spirits, it is a small step to understanding her to be representing any one of them at various points of the story. When the kava grew “from her body”, it grew on the land, tended by the people, used by the chiefs to communicate with ancestors and spirits. When she is pregnant, she is the island threatening to erupt — pregnant with menace, in fact; when she gives birth, she becomes an erupting volcano. When she is speaking to “the man”, she is the chief and people negotiating with the Tongans and taxing them with greed and irreverence. When she “dwelt with the kava” she represented peaceful conditions restored.

It is the sequence concerning this Female Deity's giving birth which enables us to connect her to Kuwae. The birth is the explosion and the seemingly muddled metaphors make sense in that context. The revealing passage is

The woman…gave birth on the fifth night, and became unconscious [?died]. Then came the man. He looked on the lips of the woman, open wide, and gasping for air. He saw the ocean which was red, and the skies which were red, because her entrails had spread out and her tongue had fallen back. The man looked on the placenta of the woman which had become red in the lips of the woman, and was afraid.

The “lips of the woman, open wide” represent the surrounds of the “mouth” of the volcano — now the land of the coasts of Epi and Tongoa, two other small islands and an outcrop, encircling a stretch of sea some 10 km x - 298 5 km in extent (Fig. 1). “Gasping for air” may mean both that the mouth was full of water and that the people were in great distress, hard put to find the necessities of life after the catastrophe. The ocean was red and the skies were red, as we know they would have been for some years after the explosion. They were so “because her entrails had spread out” — pumice and ash, from her interior, had covered land and sea. “Her tongue had fallen back”: the great tongue of erupting magma had shot up, ceased, and seawater had then rushed into the “mouth”, which may have remained mildly active as, intermittently, it is still. The man “looked on the placenta of the woman which had become red in the lips of the woman”: the land around the site of the eruption, the “mouth” (and also the source of the placenta, birth being eruption), was red, raw, naked volcanic earth. (There may also have been a redness visible beneath the sea due to remnant activity. Tongoan eyewitnesses have described such a sight to me.) Well might the man be afraid!

The imagery seems plain enough. It is also noteworthy that three bodily metaphors — mouth and tongue, birth and placenta, entrails — are interwoven in a confusing way. This may be an example of the legend-maker's art of measured obscurity in poetic allegory.

In common with other stories of the origin of kava, from both Polynesia and Melanesia, Firth's first legend has kava growing from a buried body. Both the fact that it is widespread and that it was given to Firth in the more “accessible” version of the story suggests that this belongs to the “outer layer” of legend and is therefore not vital to the history being recorded — or at least to its confidentiality. Here again, Firth gives more detail than do most records, and a context and an interpretation offer themselves, though admittedly highly allegorical.

The relevant passage is the opening of Firth's first legend, up to the point where a rat is observed to go and eat of the kava and sugarcane: “A canoe was being built [in Tonga] …”. The “man” could not feed the canoe-builders, cooked his son as food, which they rejected and left the work. They buried the son and two shrubs grew on the grave.

If we suppose the child to stand for future generations, a canoe being built to represent efforts to extend voyaging and absence of food for the builders to mean that food was scarce (from overpopulation, drought or any other circumstance) we may perceive an allegory. The grave in which the boy was laid then becomes land sought for future generations to “rest” in. The interpretation would then be as follows.

There was insufficient land (in Tonga) and voyages had been made in all known places without success; all likely possibilities had been covered (“All was finished.”). But still, the spectre of hunger loomed. - 299 The leaders (Tu'i Tonga? chiefs?), at a loss, could say only that the future generations would have to live within the confines of the land they had and, by implication, population growth would have to be restricted. This was an unpalatable idea. Rebellion was in the air (“they put their adzes on their shoulders and left the work”). Perhaps the next passage means that the skilled classes of boatbuilders and navigators (“the expert builders”) took matters into their own hands and carried out more voyages.

It is tempting to infer that this passage is an allegory for considerable social disturbance, presenting turbulent events in the softest possible terms. This may be reading too much into the metaphor but it is nevertheless worth proposing. A man killing and cooking his own son is, after all, a rather drastic metaphor to choose; it may also have had other, concurrent, meanings.

Whatever the case, the “grave” produced kava and sugarcane. Melanesia was the source of these two crops. 3 We can take the grave to be the new lands discovered, but then the focus becomes kava. Hence, one may assume that the search for new land yielded nothing more than the new crop, discovered in the west whence sugarcane had come, “this land”.

One must also consider the striking references to kava being first prepared from the leaves. Firth was clearly faithful in recording this. He even records doubting his record in the first legend. The repeated references in the second allow little doubt that the Tongans first knew only of kava prepared from leaves and stems, and probably from wild kava in the first instance.

The stems and leaves of kava do contain kavalactones, but at a much lower level than does the root material. A drink prepared from them gives a mild version of the sense of transport that good kava imparts. Wild kava typically has higher concentrations of kavalactones than cultivated kava but its chemotype is unfavourable, having more of the kavalactones producing unpleasant side-effects. (Lebot, Merlin and Lindstrom 1992:77 present clearly the sequence of change in chemical analysis in the process of domestication.) A preparation from the leaves and stems of wild kava would thus be stronger but less pleasant than a similar one from cultivated kava. It would be something like good kava but overlaid with deleterious effects, probably mild nausea, headache and lassitude. (Brunton 1989:32-3 gives examples of the use of wild kava as a beverage.)

Consequently, the sequence begins to take on a clarity. In the first legend there was plainly some doubt as to whether the new drink might be poisonous, so others tested it before “the chief” drank of it. In the second, the Tongans did not at first know of good kava but “then came a fleet from Tonga … and plucked its leaves and its stem” and after tasting it, “Ti Tonga threw away the kava which his land used to prepare”.

- 300

At that point, presumably the Tongans had learnt of the existence of better, cultivated, kava and were determined to gain access to it, yet still used leaves and stems. The inference to be drawn is that the Melanesians still guarded their secret closely. They would presumably have taken care to see that their monopoly was not broken. A simple way of doing so would be to excise the growing-points from all material traded to outsiders. That would preserve the monopoly but also provoke the Tongans to seek to break it. There might have been, however, a more honourable and reverential basis of conduct, at least initially.

With these constructions on context and metaphors one can proceed to interpret the whole sequence portrayed in Firth's two stories. The myths become legends telling a history rather clear in outline but somewhat speculative in detail.


The first of the two legends recorded by Firth, the “outer” version, may be interpreted succinctly as follows.

In Tonga, efforts were made to seek new lands without success. It seemed that future generations would have to confine themselves to the land available but this was not a popular solution and new efforts were made, ranging further afield. The resulting westward voyages discovered new lands. These lands, whence sugarcane had come, were fully populated but one reward was the discovery of the mystical qualities of kava. The Melanesians however protected their sacred crop by offering the Tongans only the stems and leaves of kava and kept the secret of preparation of the beverage from the roots. The Tongans returned to Tonga with the stems and leaves to offer to the Tu'i Tonga. There, the new drink was, as a matter of caution, tasted by lesser chiefs before it was offered to the Tu'i Tonga.

The interpretation of the second legend is necessarily more complex. A sequence of simple metaphors is employed, as well as the major ones dealt with above, but also the sequence of telling moves about in a manner suggesting that it may actually be two legends run together. We may, in interpreting it, pass over the local Tikopian references.

In Tonga, kava was first prepared from the leaves and stems of wild kava. The Tongans then knew no better. But then they learned of a milder form and returned to find it. Still believing the stems and leaves to be the source of the drink, they took these, either by force or trade. (The “great fleet”, the “a hundred canoes”, could mean a great single force or many voyages or both.) The Tu'i Tonga “found its taste was good”: the Tongans were converted to and sought the cultivated varieties.

- 301

They did so with vigour and their thirst for it reduced stocks considerably (“…a crew … went to bring again the kava … they came again and did the same, plucking it and returning … to Tonga, and the kava was again mixed; Ti Tonga again drank of it … The Tongans told Ti Tonga there was only one kava plant standing … [he] drank-and drank of it … and said … the Tongans should go again.”). Then in the record there follows a sentence which seems to close abruptly that sequence: “They came, and plucked one plant, but also dug out its roots.” This presumably stands for the ultimate discovery of the use of the roots, but it may also mean that the Tongans made a clean sweep of stocks available, or again that the means of growing kava in Tonga was acquired. It may mean all three.

The account changes at this point, suggesting that Pa Rangifuri for some reason switched to another version, perhaps a different layer, of the legend. More evasive analogy is employed, “the Tongans” are replaced by “the man”, “plucking” and “digging” kava both figure in the following passages, and there is a shift in scene, to the domain of the Female Deity, which I take to be Kuwae-Efate.

The Tongans had been trading kava elsewhere than Kuwae-Efate, probably further to the north. That trade excluded the people of Kuwae-Efate, but excited their interest. (“The man who first seized the kava was treated as married by the Female Deity. Pufafine had conceived a desire for the man who was digging some way off.”) They made contact with the Tongans.

The Kuwae-Efate people then negotiated cannily, perhaps knowing that the Tongans had nearly exhausted supplies from elsewhere (see above). They held off (“he and the Female Deity did not copulate”) and when the Tongans had been sufficiently humbled in the negotiations, they offered trade on their terms (“but when he went to sleep the Female Deity went to him and had intercourse … with him”). The Tongans were relieved, and glad to accept the terms (“he was overcome”).

Trade began, but under close control. Negotiations were restricted, perhaps to the highest-ranking traders, perhaps to the first comers. (“The Female Deity did not appear to all the men from Tonga but only to the one man, the man in front.”) The Tongans were not - at first - permitted to stray inland and discover the location of the kava plantations. (“The mass of men all stayed in the canoe.”) The Tongans however managed to discover the secret and perhaps ultimately plundered the plantations. (“The one man went inland to pluck the kava. Then he returned to dig the kava. He dug out the base and its rootlets. Then there remained only one (kava plant), not a large one.”) Very little kava was then left, either from excessive trading or theft.

- 302

The seismic precursor to the explosion of Kuwae began and the Tongans were no longer welcome. (“And the woman had become pregnant, the Female Deity indeed was pregnant. Then the Female Deity told the Tongan that she was big in her belly and thath … he might go.”) Yet here there is a suggestion that the local chiefs were saying to the Tongans that matters might be mended and trade resumed when things had settled down again (“that while he might go, in five nights [i.e. five years] he should come again”). There is of course an element of hindsight in the legend, the “five nights”.

After five years of earthquakes Kuwae exploded and it was not until a further five years that the Tongans returned to reconnoitre the resumption of trade. (“He stayed ten nights in Tonga. The woman stayed and gave birth on the fifth night … Then came the man.”) In the intervening time, the volcano had become quiescent and/or conditions had been very difficult for the people of the Kuwae-Efate area (the woman “became unconscious [?died]”). The arriving Tongans beheld the physical and social devastation much beyond what they had suffered. (“He looked on the lips of the woman…”; the whole passage describing the aftermath of the explosion, as given above.)

Disembarking, the Tongans were met with hostility, with indignation that their greed for kava had caused this catastrophe, the fury of the gods. (“Then as he went up she said to him: ‘Who are you here to reject me?’”) In reply they protested that they had not been warned, perhaps of the sanctity of kava, though this was an excuse (“‘Who knew? You didn't tell me so that I could hear!’ - the lie that the man uttered because of his fear.”)

This was brushed aside and the Tongans were obliged to acknowledge their guilt. (“Then the man hid his head on his knees.”) They were admonished to change their ways and show due respect to the gods. (“The woman told him to turn his head and look up above.”) The trade was finished; the Tongans were told not to return, that if they did so it would mean war. (“Then he went and the woman said to him: ‘You are going, don't turn your head round to the rear; just look in front. If you look round, you'll be eaten!’”)

The Tongans returned home, never to return. (“The man went paddling off, looking only ahead … and arrived in Tonga.”) It was some time after they had returned that they felt sure there would be no reprisal. (“When he got inland, he craned his head round behind - and the sky and the ocean had become light-coloured.”) They perhaps felt guilty at having broken a tapu, with such resounding consequences. They may also have feared that the Kuwae-Efate people would discover that in the course of their plundering they had secured some planting material, breaking the monopoly, for now they had cultivated kava varieties. (“The kava had stayed with the man. It was planted in Tonga.”) Nevertheless, the best kavas were still secure in their place of origin and normal cultivation returned; the whole experience - 303 reinforced possessiveness for kava there. (“But the woman dwelt with the kava in this land; it stood, grew large, thence was called ‘the plant of the Female Deity’.”)


If one accepts that Pufafine's giving birth stood for volcanic eruption, one may well ask whether this was indeed that of Kuwae. It is most of all the references to the sky and ocean being red that restrict the possibilities to the great eruptions. Only they could produce the general stratospheric ash-pall which could produce such an effect. Of those major eruptions (Monzier, Robin and Eissen 1994: 216) three are in the Pacific: Taupo (New Zealand, A.D. 186), Kuwae, and Ambrym (some 60 km to the north of Kuwae, about 2000 B.P.). Taupo erupted before Polynesians found New Zealand. Ambrym, about half the size of the Kuwae event, erupted one thousand years or more before the known era of Tongan voyaging and from the configuration of the island, with the craters high and inland, could scarcely have produced the imagery given in Firth's records. It is most difficult to conceive that Tongans could have observed, in the terms of the legends, the after-effects of any but the Kuwae event.

That seems to be a sufficiently firm basis on which to conclude that a trade in kava between Vanuatu (in latter years at least, Kuwae-Efate) and Tonga terminated just prior to A.D. 1452, probably in A.D. 1447.

There are many parts of these interpretations which may be doubted, including some practicalities. Of moderate importance in the story is the matter of the Tongans first taking leaves and stems to prepare the kava beverage. Was it possible to carry these from Vanuatu to Tonga and have them arrive in satisfactory condition? It was. Wrapped in wet leaves and shaded, kava stems will remain fresh for two weeks or more and from the data of Lewis (1972), the voyage could have been made in that time. Even if leaves (or the stems) dried out, the resinous kavalactones would remain intact, as they do in the modern trade in dried kava. An agricultural and voyaging people could certainly have managed that transit for so prized a commodity.

Prized kava certainly was, especially in scarcity. Reserved for the highest classes, its qualities considered an avenue to the unseen world, it bordered on the mystical. It was probably more alluring to Polynesians than were spices, tea, coffee and opium to Europeans and Asians. If those commodities could produce the voyaging, rivalry and conflicts they did, kava could have drawn Tongans to Melanesia. Indeed, there can be few more likely conjunctions than Vanuatu's wealth of kava and sailors of “imperial” Tonga to explain the carriage of kava into Polynesia.

- 304

Interpreting these legends is fraught with potential error because of the nature of layered legends, their imperfect recording, deficiency of context and various other subtleties. It is difficult to know for example whether “he disregarded her” had a potent meaning, was a supportive statement or was a mere filler. Similarly, the “conversations” between Pufafine and “the man”, the Tongans, can be construed variously within the general context. Most of all, the opening sequence centring on the cooking of a boy is of uncertain meaning and the interpretation of it is offered here only as a possible one.

That sequence opens the account and thus refers to the circumstances which led to the beginning of the trade. The data on the Kuwae explosion peg the end of the trade but the beginning is problematic. We might well assume that it was in the heyday of Tongan voyaging and so put it no earlier than circa A.D. 1200. It could well be much later: more than two centuries seems a long time for enterprising Tongans to trade in a prized commodity without securing growing stock for themselves. The deficiency of our knowledge of the interface is most of all the state of things on Kuwae-Efate. Garanger's excavation shows that a developed, hierarchical political system existed on Efate, evidently in the thirteenth century. There is reason to believe that Garanger excavated only one of a sequence of burials. The histories predating the Kuwae explosion and the survival of such chiefly titles as Ti Evate, Ti Nambua Mata, Maraki Pule and Marikitapu suggest an old neo-Polynesian polity. It may have held over an extended span of time. Tongans may have encountered a locally powerful political structure. Firth's record suggests that they did.

It is perhaps a possibility that kava was not traded but taken as tribute. The data we have do not suggest so. The legends suggest trade and a spirited rejection of its continuation when matters turned sour. Nor is there evidence that any part of Vanuatu was part of the Tongan empire in any formal sense. Rather, the region might have been in the nature of a neighbouring state, a source of traded items.

However long the trade went on, the Kuwae explosion stopped it. One can scarcely overstate the effect this stupendous occurrence must have had. The famine induced by the cold conditions following it may well have led to political change in the Pacific as it appears to have done elsewhere. For example, famine might have been the trigger for the events in Tonga that led to the setting-up of the Tu'i Ha'a Takalaualine, which Gailey (1987:64, 69, 70) puts at around A.D. 1450. Rutherford (1977:34, 35) writes that Takalaua “made heavy exactions on the labour and time of his people, even in the yam-planting season” to build his sepulchre and that “This, at least, was the pretext for a conspiracy against him … and for his murder”. One might propose that the exactions were made heavy by famine conditions, leading to Takalaua's - 305 assassination and thence to the political reorganisation by his son - the division between the Tu'i Tonga and the new Tu'i Ha'a Takalaua.

However that may be, there can be little doubt that Tonga was the source of the first of Firth's stories. It bears a close resemblance to others from Tonga, as Firth (1970:218) observes. Lebot, Merlin and Lindstrom (1992:123, quoting Bott 1972) give another with much the same elements. Others from Polynesia which they quote have lesser degrees of the same theme or none at all. Frequently Tagaloa figures in these stories. Of them, one recorded by Lebot (op. cit.) in Samoa is the closest to the Tongan theme. There is an impression, then, that the “outer” version came from Tonga and pieces of it went with kava further afield, being progressively diluted or modified until at length the form bore no resemblance to the original.

Whether or not that impression is a true one, the legends of the origin of kava to be found in Vanuatu seem to belong to another stream. Some years ago I was given what was said to be the full story of the origin of kava. It came from Maewo in the north of Vanuatu, which is often favoured as the site of domestication of kava. It was claimed that only pieces of this story were to be found elsewhere in Vanuatu and the myths from Vanuatu reported by Lebot, Merlin and Lindstrom bear that out. This legend begins with a culture hero who might or might not be Tagaloa. His nephew had a love affair with his wife. His revenge was to kill his wife, bury her head down and with legs splayed and then to force his nephew to eat her rotting vulva. The nephew did not swallow the material but later spat it out at a place from which the kava plant then grew. In a dream the woman told the nephew to drink of the plant - with various sexual references - to remember their love. He did so. It was only when a rat chewed the kava that others learned his secret.

One may observe a few points in common between this and Firth's “outer” legend (e.g. the much-repeated buried body, the rat). One may also, on close examination, see how this story might have been the parent of some myths in Vanuatu in the same manner that the Tongan tale of a historical event may have been the parent of myths elsewhere in Polynesia. Close examination of the stories is not to be attempted here. It is merely propounded that many are myths, modified fragments of original legends which had meanings rooted in history and presented in “code”.

However speculative that suggestion may be, it would seem rather less so than attempting to find meanings in the modern remnants of the cultures from which the stories are presumed to have sprung. There is enough to indicate that subtle imagery was employed, often using bodily parts and functions as well as surrounding nature, in an allegorical style more akin to poetry than any other form of western art. Recorders were not necessarily of a poetical or allegorical turn of mind and meanings may well have gone quite over their - 306 heads. Study of the genre and thus of the mentality that produced it may well reveal more than interpretations cluttered by modern western conceptions.


John Lynch freely and kindly advised me on the tenor and text of this paper. It was mainly through the co-operation of George Pakoasongi that I was able to gather privy information from the Shepherd Islands. A number of others gave me assistance and encouragement. I acknowledge with thanks all this assistance.

Firth's Record of the Origin of Kava

Firth's record is Tikopian, which is to say that he collected it on Tikopia against a context of observing kava's use in Tikopia ritual. His remarks relate to that context. His text (Firth 1970:216-8) follows.


Stories about the origin of the kava focused on the plant and the drink, not on the ritual, and did not seem to be well known; they were rarely mentioned. There were two versions, one with the scene laid in Tonga, the other in Tikopia. I recorded both from Pa Rangifuri (later, Ariki Tafua) in 1928.

- 307

The first ran: ‘A canoe was being built [in Tonga]. As the end of the work drew near the man had no food with which to feed the expert builders. All was finished. He took then his small son, Tuipania, put him in the oven and covered him up. When he was cooked he opened the oven and sent the food to the builders. They looked - it was the flesh of man - they did not eat. They put their adzes on their shoulders and left the work. They took Tuipania and buried him. By and by two shrubs grew on the grave. The Tongans observed a rat which went to eat of the shrubs. It ate the root of one, the kava, and became unconscious. It revived and then ate of the other shrub - which was sugar cane - and frisked about. Then the Tongans knew that one shrub was bitter/stupefying [kona] and the other sweet [kara]. They took the leaves [sic in my notebook? root] of the kava and made them into a beverage. They proposed to give it to their chief to drink. Some man said: “No! Give it to the elders so that the chief may live.” This was done, but the elders found it good. The chief then drank of it also.’

This was the origin of kava in Tonga. It was embodied in a dance song known as Sakitasi:

The springing up of the kava Te fitianga o te kava
Which stands there from Tonga E tu ra mai Tonga
On the grave I te tano
Of Tuipania O Tuipania
Who was cooked in the oven E tao i te umu
Tuipania was cooked in the oven Tuipania ne tao i te umu

Here the cultural idiom was a Tikopia one, but it was the Tongans' kava that was being described. About a month later Pa Rangifuri told me another story, in which he gave due priority to Tikopia.

'Its origin was in this land; the kava of Tonga was afterwards; the first kava was the kava of this land. There is a story about it.

‘The kava of Tonga was made foolishly, was made with a wild plant: they did not know the kava. The head of the kava was the deity of us [Tafua], Rakiteua. He was the Worker of all the chiefs. He alone stands up and recites the formula of the Kava House [in Marae Lasi]’… (Here my informant went off into a description of the kava of Marae (Work of the Gods, 1967, pp. 294-8). He continued: 'The kava originated in the Female Deity, Pufafine; it grew from her body. Then came the fleet from Tonga, a hundred canoes, a great fleet, they came and plucked its leaves and its stem. They carried it off for Ti - 308 Tonga [the Tongan ruler] to drink. When the kava was mixed Ti Tonga drank of it and found its taste was good; he threw away the kava which his land used to prepare - the kava in Tonga was a wild kava [kavaatua].

'Then a crew prepared and went to bring again the kava from Tikopia; they came again and did the same, plucking it again and returning in their canoe. It went to Tonga, and the kava was again mixed; Ti Tonga again drank of it. The Tongans told Ti Tonga that there was only one kava plant standing. Ti Tonga drank and drank of it, and said one day that the Tongans should go again. They came, and plucked one plant, but also dug out its roots. The man who first seized the kava had been treated as married by the Female Deity. Pufafine had conceived a desire for the man who was digging some way off. He and the Female Deity did not copulate, but when he went to sleep the Female Deity went to him and had intercourse [spiritually] with him; he was overcome. [Cf. p. 101.] The Female Deity did not appear to all the men from Tonga but only to the one man, the man in front. The mass of men all stayed in the canoe; the one man went inland to pluck the kava. Then he returned to dig the kava. He dug out the base and its rootlets. Then there remained only one (kava plant), not a large one. And the woman had become pregnant, the Female Deity indeed was pregnant.

'Then the Female Deity told the Tongan that she was big in her belly, and that while he might go, in five nights he should come again. But the man went and did not return; he disregarded her. He stayed ten nights in Tonga. The woman stayed and gave birth on the fifth night, and became unconscious [?died]. Then came the man. He looked on the lips of the woman, open wide, and gasping for air. He saw the ocean which was red, and the skies which were red, because her entrails had spread out and her tongue had fallen back. The man looked on the placenta which had become red in the lips of the woman, and was afraid. Then as he went up she said to him: “Who are you here to reject me?” He replied: “Who knew? You didn't tell me so that I could hear!” - the lie that the man uttered because of his fear.

'Then the man hid his head on his knees. The woman told him to turn his head and look up above. Then he went and the woman said to him: “You are going, don't turn round to the rear; just look in front. If you look round, you'll be eaten!” Then the man went paddling off, looking only ahead.

‘He went on and on, and arrived at Tonga. When he got inland, he craned his head round behind - and the sky and the ocean had become light-coloured. The kava had stayed with the man. It was planted in Tonga. But the woman dwelt with the kava in this land; it stood, grew large, thence was called “the plant of the Female Deity”. The origin of the kava in Tonga came from this land.’

- 309
  • Brunton, Ron, 1989. The Abandoned Narcotic. Kava and Cultural Instability in Melanesia. Cambridge: Cambridge University Press.
  • Crowley, Terry, 1994. Proto Who Drank Kava? in A. K. Pawley and M. D. Ross (eds), Austronesian Terminologies: Continuities and Change. Pacific Linguistics C-127. Canberra: Australian National University, pp.87-100.
  • Daniels, John and Christian Daniels, 1993. Sugarcane in Prehistory. Archaeology in Oceania, 28:1:1-7.
  • Delmas, Robert J., Severine Kirchner, Julie M. Palais and Jean-Robert Petit, 1992. 1000 Years of Explosive Volcanism Recorded at the South Pole. Tellus, 44B:335-350.
  • Eissen, Jean-Philippe, Michel Monzier et Claude Robin, 1994. Kuwae, l'Éruption Volcanique Oubliée. La Recherche, 270 Nov. 1994, Volume 25:1200-2.
  • Firth, Raymond, 1970. Rank and Religion in Tikopia. A Study in Polynesian Paganism and Conversion to Christianity. London: George Allen and Unwin.
  • Gailey, Christine Ward, 1987. Kinship to Kingship; Gender Hierarchy and State Formation in the Tongan Islands. Austin, Texas: University of Texas Press.
  • Garanger, José, 1972. Archéologie des Nouvelles-Hébrides: Contribution à la connaissance des Îles du Centre. Paris: ORSTOM. (Publications de la Société des Océanistes, No. 30.)
  • —— 1982. Archaeology of the New Hebrides: Contribution to the Knowledge of the Central Islands. Sydney. Oceania Publications. (Oceania Monographs No. 24).
  • Guiart, Jean, 1973. Le dossier rassemblé, in J. J. Espirat et al., Système des Titres, Electifs ou héréditaires, dans les Nouvelles-Hébrides Centrales d'Efate aux Îles Shepherd. Paris: Institut d'Ethnologie, Musée de l'Homme. (Mémoires de l'Institut d'Ethnologie, 10)
  • —— 1996. Archéologie et Ethnologie, in Michèle Julien, Michel and Catherine Orliac (eds), Mémoire depierre, mémoire d'homme Tradition et archéologie en Océanie. Paris: Publications de la Sorbonne (Collection Homme et Société 23).
  • Hammer, C. U., H. B. Clausen and W. Dansgaard, 1980. Greenland ice sheet evidence of post-glacial volcanism and its climatic impact. Nature, 288 No. 5788:230-5.
  • Hassig, Ross, 1981. The Famine of One Rabbit: Ecological Causes and Social Consequences of a Pre-Columbian Calamity. Journal of Anthropological Research, 37:171-181.
  • Kirch, Patrick Vinton and D. E. Yen, 1982. Tikopia. The Prehistory and Ecology of a Polynesian Outlier. Honolulu: B. P. Bishop Museum Bulletin 238.
  • Lebot, Vincent and Pierre Cabalion, 1986. Les kavas de Vanuatu. Cultivars de Piper methysticum Forst. Paris: Éditions de l'ORSTOM, Collection Travaux
- 310
  • et Documents No. 205. Translation published in 1988 as Kavas of Vanuatu: Cultivars of Piper methysticum Forst. Noumèa: South Pacific Commission. Technical Paper No. 195.
  • Lebot, Vincent, Mark Merlin and Lamont Lindstrom, 1992. Kava. The Pacific Drug. New Haven: Yale University Press.
  • Lewis, David, 1972. We, The Navigators. Canberra: Australian National University Press.
  • Martin, John, 1817. An Account of the Natives of the Tonga Islands (as told by William Mariner). London.
  • Monzier, Michel, Claude Robin and Jean-Philippe Eissen, 1994. Kuwae (1452 A.D.): the forgotten caldera. Journal of Volcanology and Geothermal Research, 59:207-218.
  • Pang, Kevin D., 1993. Climatic Impact of the Mid-Fifteenth Century Kuwae Caldera Formation as Reconstructed from Historical and Proxy Data. Eos, Transactions of the American Geophysical Union, 74:43:106.
  • Rutherford, Noel (ed.), 1977. Friendly Islands. A History of Tonga. Melbourne: Oxford University Press.
1   The name Kuwae is recorded in legend and is still in fairly common use in the Shepherd Islands. The limits of its general dimensions are easily ascertained by examination of submarine contours.
2   Two of his informants gave me information withheld from him. These and other Tongoan chiefs assert or infer that information was withheld. This does not detract from Guiart's remarkable effort. Rather, it is surprising that he was given as much as he was. Chiefly histories were always held secretly. They were and remain the basis of landholding and chiefly power, and to divulge them fully would be to relinquish reserve powers and, in effect, offer them to rivals. When Guiart was collecting his data, Tongoa was in ferment on this very subject. Now, some chiefs are prepared to have some more material recorded in confidence.
3   For a review of the origin of sugarcane, see Daniels and Daniels (1993). While the ultimate origin may have been more remote, to Tongans Melanesia might have been the origin. Whether they would have retained such a folk memory is speculative, but of small import. In any case sugarcane may have been associated with kava in the legend for other reasons.