Volume 105 1996 > Volume 105, No. 2 > The soapberry, a neglected clue to Polynesia's prehistoric past, by Robert Langdon, p 185-200
                                                                                             Previous | Next   

- 185

In the long-running debate on whether American Indians preceded Polynesians in the settlement of Easter Island and other Eastern Polynesian islands, one seemingly significant ethnobotanical clue has been almost entirely neglected. This is the soapberry, Sapindus saponaria, a small tree native to the American tropics that was found on Easter Island at the time of European contact. The tree is also known in other islands of Eastern Polynesia but not Western Polynesia. Its remains have been found in Peruvian archaeological sites dating back 2,500 to 3,000 years. The ancient Peruvians cultivated it and used its berries as a source of soap and its seeds for beads (Towle 1961:62-3; Ugent, Pozorski and Pozorski 1986:92). This therefore raises the question of whether American Indians planted the tree on Easter Island or elsewhere or whether it got there by other means.

S. saponaria is one of 12 or 13 species of the genus Sapindus in the order Sapindaceae. All are native to either the Asian or the American tropics. All contain saponin, a glucoside, in large cavities of the flesh of their fruits. When shaken with water, this forms a frothy detergent lather like soap. Hence the name Sapindus, from Latin sapo ‘soap’ and indicus ‘Indian’. In America, S. saponaria grows to a height of from four to ten metres. Its leaves consist of about four or five pairs of oblong, elliptical or lanceolate leaflets. It has small, white flowers that produce round, brown berries with orange-brown translucent flesh. Each berry contains a small globular black seed about 1.2 cm or almost half an inch in diameter. Saponin is found to some extent in the whole plant. To the Quechua-speaking people of modern Peru, the plant is known as sullucu, cholloco and choloque (Towle 1961:62-3).

The soapberry was first reported on Easter Island in 1773 when Captain Cook, then on his second Pacific voyage, sent a party ashore to examine the countryside. The party included the astronomer William Wales and the naturalist J. R. Forster. Cook incorporated Wales' report into his own journal. In a ‘long excursion’, Wales said, the party saw no trees or wood of any kind, except at the island's southwestern corner. There they found ‘a small shrub whose leaf was not much unlike that of an Ash’. Its white and brittle wood also resembled that of the ash ‘in some measure’ (Beaglehole 1961:347,827). Forster did not mention this discovery in his own journal, but he collected a specimen of the plant which he described in an Easter Island - 186 vocabulary as ‘Fraxinifolia, a small shrub with ash leaves’. He gave its local name as mareecooroo = marikuru (Schuhmacher 1978:5). Forster's son George later referred to the Easter Island plant as Sapindus saponaria (Forster 1786:29).

The elder Forster's specimen was eventually acquired by the British Museum. The botanist Berthold Seemann, who found it there in the 1860s, devoted a footnote to it in his flora of Fiji:

Sapindus saponaria, Linn. I have examined Forster's original specimen of this plant from Easter Island, preserved at the British Museum, as there was some doubt expressed of its belonging to this species, and as far as it goes I have found it absolutely identical with the American specimens (Seemann 1865-73:47).

The Easter Island soapberry was not mentioned in print again until the 1890s following the visit to the island of the USS Mohican in 1886. Both the ship's paymaster, William J. Thomson (1891:547), and the surgeon, George H. Cooke (1899:722), had learned that marikuru was its local name - or, at least, for what they termed ashwood. A generation later, Francisco Fuentes (1913:326,334), a Chilean botanist, described S. saponaria in his pioneering survey of the island's flora of 1911. In his view, it was like the banana: it had been on Easter Island since ‘a remote epoch’. He found specimens of it in a semi-naturalised state at Tahai, north of Hangaroa village, and at Hotu-iti at the eastern end of the island. The islanders, he said, used it as the source of an astringent medicine and for soap while its seeds were used for personal adornment and rosaries. Fuentes added (author's translation):

This plant…belongs to tropical America. Its common name in Cumaná [Vene-zuela] is parapara. The Easter Islanders use this name to describe the bark of mahute [the paper mulberry tree] and also paper. This identicality of American and Polynesian words is frequent among the common names of plants and suggests former links between the people of one region and the other.

After a long stay on Easter Island in 1934-35, the French anthropologist Alfred Métraux (1940:17) summarised the published accounts of the soapberry of Cook, George Forster and Fuentes, but did not mention Fuentes' claim that the islanders used it as the source of an astringent and soap. He said that the leaves and branches of a specimen from Rano Kao had been given to the Bishop Museum herbarium in Honolulu; that it had been identified as S. Saponaria; a nd that the islanders extracted a ‘white color substance’ from the tree's rotten stumps and branches. He did not say what they used it for.

Father Sebastian Englert, who lived on Easter Island for almost 35 years until his death in 1969, scarcely mentioned the soapberry in his - 187 writings. In his first Rapanui-Spanish dictionary (Englert 1938:80), he recorded marikuru merely as the name of a ‘certain tree’ and of a white clay. In a posthumous revised version (Englert 1978:192), the plant was identified as S. saponaria. Very few specimens were said to remain on the island.

An agronomist, Fusa Sudzuki (1979:5) stated that he had not encountered a single specimen during a survey of the island's flora in 1971. But the geographer John Flenley (1993:11-2) has since claimed that he saw a small specimen near Rano Kao in 1977. Flenley said in the same context that if the plants introduced to Easter Island before its European discovery in 1722 could be correctly distinguished from indigenous species and those introduced in post-European times, then ‘some indication’ might be gleaned about the place of origin of the island's aboriginal settlers. However, Flenley's presentation of evidence on controversial ethnobotanical issues scarcely allowed his readers to make a balanced assessment. Thus although he quoted Fuentes and Métraux as saying that the fruits of the soapberry were used for necklaces and that a white-coloured substance was extracted from its rotten wood, he said ‘the more well-known use of the fruits as a substitute for soap’ had ‘not commonly been recorded’ and this made its introduction to Easter Island by man ‘seem rather doubtful’. The pollen evidence, he added, was ‘too exiguous to be conclusive’.

The first account of S. saponaria elsewhere in Eastern Polynesia is from the Marquesas Islands where it is called koku'u. This account is in a manuscript of the Rev. Samuel Greatheed, of the London Missionary Society, which was compiled from information supplied by William Pascoe Crook, the first missionary to live in the group. Crook spent 19 months on Tahuata and Nukuhiva in 1797-98. The Greatheed account of the soapberry reads:

The Kogu [sic] abounds so much that some of the Vallies are filled almost entirely with it, either in large timber or underwood. It has a curious leaf, apparently consisting of several smaller leaves, some of which serve as stems to others. It produces a hard black round berry, resembling a marble, which is bored and strung upon the strings by which their [the islanders'] gourds are suspended. It is also thought to be used in dy[e]ing as a colour (Sheahan 1952:lxxxi).

The whaling surgeon Frederick Debell Bennett (1840:II:348), who visited the Marquesas in 1835, reported the soapberry only on the island of Tahuata. It was a lofty tree, he said, and so abundant that the islanders supplied it to ships as firewood. They themselves used it as ‘an ingredient in the turmeric cosmetic’ that they smeared over their bodies. It caused the dye to adhere to their skins.

The French naturalist Edelstan Jardin (1857:307) dubbed the Marquesan - 188 soapberry Sapindus microcarpa in his survey of the group's flora. He said its yellow wood was very hard, but was inclined to split and dry out. The Marquesans did not use it for anything, but knew that its sap could be used to bleach cloth.

In another such survey three generations later, the American botanist Forest B. H. Brown (1935:160) referred to the Marquesan soapberry as ‘Sapindus saponaria Linnaeus variety Jardiniana F. Brown’. It was endemic in the group, he said, being found in open, more or less exposed situations at from 10 to 400 metres altitude on Eiao, Nukuhiva, Uahuka, Uapou and Hivaoa. It grew to heights of 10 to 15 metres. The islanders used its wood for making tapa beaters and pegs for husking coconuts. The undiluted juice from the leaves and fruit was applied externally in the treatment of skin diseases. It had also been used in the ancient embalming process. Brown gave its Marquesan name as kokuu. The correct form, koku'u, contains a glottal stop which reflects the r in Mangarevan kokuru and Rapanui marikuru.

Dictionaries compiled by 19th century missionaries to Mangareva, some 1300 km SSE of the Marquesas, record kokuru as ‘the name of a (small) tree’ (Tregear 1899:35; Janeau 1908:48). But no botanist ever determined whether the tree was S. saponaria or not. If it was, as seems likely, then the name was probably applied to it by Mangareva's first settlers, who are thought to have come from the southeastern Marquesas in about the 12th century A.D. (Green 1966:21-2). This, of course, would have been before the r in most Marquesan words had been lost.

The botanist Harold St John (1988:18), who made a census of Mangareva's flora in 1934, found specimens of S. saponaria on the slopes of Mt Duff. His local informant called it tutae kiore, literally ‘rat faeces’, a term that also appears as ‘the name of a tree’ in one of the 19th century dictionaries (Tregear 1899:114). According to St John (p.3), immigrants from the Tuamotu Archipelago had largely replaced the original population by his time. So if kokuru originally meant ‘soapberry’, it may have fallen out of use by then.

S. saponaria occurs on Pitcairn Island, 250 km eastward of Mangareva and about 2,000 km west of Easter Island (Smith 1985:590). However, as Pitcairn was uninhabited when the Bounty mutineers settled there in 1791, we cannot now know what its earlier inhabitants called it or what, if anything, they used it for.

The soapberry does not grow in the Tuamotu Archipelago. But Mangarevans or/and Marquesans who settled there before and after the r of their language was lost evidently applied the terms kokuru and koku'u to another small, shrub-like tree, Suriana maritima. The term kokuru is used on the atolls of Fangatau, Fakahina, Hao, Reao, Tatakoto and Vahitahi; kokū(= kokuu) is used on Anaa and Takaroa. Tahitian 'o'uru, the equivalent of - 189 Tuamotuan kokuru, also signifies S. maritima (Emory 1947:271). By contrast, the Mangarevan term for S. maritima is gegie, cf. Tongan ngingie (Buck 1938b:8; Yuncker 1959:154).

In the Society Islands, S. saponaria is called a'ea'e. It was evidently of little importance in early historical times as none of the European explorers mentioned it (Oliver 1974). The missionary lexicographer John Davies (1851), who arrived in Tahiti in 1801, gives aeae (sic) as ‘the name of a tree used only for fuel’, while Edouard Butteaud (1891), author of a study of Tahiti's flora, lists aeae as the local name for Ratonia stipitata (matayba), a member of a separate genus of Sapindus. The first botanical description of S. saponaria in the Society Islands relates to specimens collected by the French expedition of Captain Dupetit-Thouars in 1838 (Drake del Castillo 1893:143).

In the Cook Islands, S. saponaria was unknown to botanists until an American, W. Arthur Whistler (1990a:355), collected specimens on Mauke a few years ago. Its wood there is used as timber and for axe handles. Its local name is akeake. More than half a century earlier, G. P. Wilder (1931:70,76), author of a flora of Rarotonga, recorded the name akeake as that of Xylosma gracile, a small tree of the order Bixales, and ake as the local name for Dodonaea viscosa, a species of Sapindus. D. viscosa is also called ake or akeake in Niuean and New Zealand Maori (Sykes 1970:186; Poole and Adams 1963:136). But in Tongan, ake is the name of Fagara pinnata, a member of the Rutaceae family (Yuncker 1959:152).

In the Hawaiian Islands, a'e (or a'ea'e) is the name for S. saponaria and all species of Fagara (Pukui and Elbert 1977:3-4). But S. saponaria is also called mānele. According to the authors of the most recent flora of the group, the tree is found only on the large island of Hawai'i where the islanders use its seeds to make lei and necklaces (Wagner, Herbst and Sohmer 1991:1228-9). The Hawaiian tree is said to be ‘quite distinct’ from that of the Marquesas but to compare closely with that of the Society Islands (Brown 1935:161).

In the Galapagos archipelago, 1,000 km west of mainland Ecuador, S. saponaria is the largest forest tree. Dense stands of it occur on Isabela Island near Villamil. The authors of the standard flora of the archipelago consider it to be indigenous, but claim that it was probably introduced to the Pacific Islands ‘through cultivation and naturalization’ (Wiggins and Porter 1971:752-4). The Galapagos Islands are about 4,000 km northeast of Easter Island and are the nearest place from which the tree could have reached Eastern Polynesia prehistorically.

The botanists H. René Papy (1954:288) and Otto Degener (1945:202) both favoured sea flotation as the means by which S. saponaria arrived in Polynesia. However, while Papy thought its presence in the Society Islands - 190 could be accounted for in this way, he suggested that its vegetative characters had been modified there. According to Degener, the flesh of the soapberry shrinks from the seed when the fruit dries, forming a cavity filled with air. In this state, every seed can float. When removed from the fruit, however, about half of the seeds sink in fresh water and one fourth in sea water. ‘Whether the plant owes its wide distribution to the floating ability of all of its fruits and at least half of its seeds is not definitely known’, Degener said.

Other specialists have suggested that birds were responsible for the soapberry's distribution. The biologist Sherwin Carlquist (1967:157) claimed that internal transport of seeds and fruits by birds was the most important mode of transport to the high islands of the Pacific for most of their indigenous plants. There was ‘considerable circumstantial evidence’, he said, that certain shore birds which migrated throughout Polynesia were mixed feeders and often carried seeds internally. He did not discuss the dispersal of S. saponaria specifically, but claimed that internal carriage by birds was the most likely mode of transport for representatives of the genus. However, the naturalist H. B. Guppy (1906:325) saw ‘several difficulties’ in trying to account for the dispersal of Sapindus in this way. While the fleshy mesocarp of its fruit might attract birds, he said, it was ‘not easy to perceive’ how birds could carry its large seeds over long stretches of ocean. Nevertheless, the same difficulty existed with a few other genera such as Osmathus and Sideroxylon.

As the soapberry was an important economic plant in prehistoric Peru, human agency is a possible third way of accounting for its presence in Polynesia. Guppy and Degener, at least, would not have considered this possibility because the idea of prehistoric human contact between America and Polynesia was not given credence at the time they wrote. The sweet potato, Ipomoea batatas, the first major pointer to such an idea, was still not a contentious subject. Although an American origin for it had long been suspected, ‘much mystery’, as Guppy himself said (p.415), still surrounded its homeland.

By the mid-1920s, botanists were agreed that the sweet potato did originate in America. So that raised the question of how it came to be present in Polynesia under names such kumara which were remarkably similar to kumar in Ecuadorean Quechua. A German scholar, Georg Friederici (1929), proposed that the Spanish explorer Alvaro de Mendaña had introduced the plant and its Quechua name to the Marquesas Islands in 1595 and that it had been dispersed to other Polynesian islands from there in the seven generations before Cook. But the American ethnologist Roland B. Dixon (1932) soon demolished that idea. The plant's presence in Polynesia, he said, could be accounted for in one of only two ways: either Polynesian voyagers had - 191 sailed to America and had brought it back with them, or Peruvian or other American Indians had carried it to Polynesia. Much of the debate on the sweet potato that followed is relevant to the question of whether the soapberry could have reached Polynesia with such voyagers.

One of the first to make a positive claim on the question of the sweet potato was the part-Maori scholar Peter Buck (1938a:313-6), otherwise Te Rangi Hiroa. He asserted that some time before the 13th century some Polynesian voyagers had set out from the Marquesas Islands in search of new lands and, ‘because of the empty eastern sea’, they were forced to go on to South America. Their contact with that continent was ‘too short to make any lasting exchange in religious or social ideas’, although they possibly passed on the seeds of the gourd (Lagenaria siceraria) to the South American people. Certainly, Buck said, they had collected a supply of sweet potato tubers and had sailed for, and duly reached, their homeland. There the sweet potato was planted and thrived. Later, it was carried eastward to Mangareva and Easter Island and westward to the Society Islands.

Buck advanced no evidence to support his hypothesis. On the contrary, he seemed to have little appreciation of such practical matters as sailing against trade winds and contrary currents for he claimed that, at seven miles an hour, a Marquesan canoe could have covered the 4,000 miles to northern Peru in ‘a little over three weeks’. In fact, the only way to make continual easting in the eastern South Pacific is to get into the zone of the westerlies in from 35° to 40° S. latitude, some 1,500 miles south of the Marquesas. Had Buck's Marquesans done that, they would have had to cover a minimum of 11,000 miles (17,600 km) to get the sweet potato first to their homeland and then to Easter Island -a greater distance than from Southampton to Singapore via Capetown (Lewis and Campbell 1963:12-13). The approximate trajectories involved are: Nukuhiva-Pitcairn Island (latitude 25° 04' S), 1,118 miles; Pitcairn-Valparaiso, 3,077; Valparaiso-Guayaquil, 1,981; Guayaquil-Nukuhiva, 3,659; Nukuhiva-Pitcairn, 1,118; Pitcairn-Easter Island, 1,165: total, 11,000 miles (Day 1953).

The ‘partisanship’ of such scholars as Buck prompted James Hornell (1945), a specialist on indigenous sailing craft and co-author of Canoes of Oceania, to consider the practicalities of prehistoric voyaging between South America and Oceania in an article in the Journal of the Polynesian Society. He concluded that the sweet potato was more likely to have arrived on a one-way drift voyage from Peru than on a two-way Polynesian voyage. ‘Far too much emphasis’, Hornell said, had been placed on the alleged lack of maritime enterprise among the ancient Peruvians and on the notion that the logs of their balsa rafts did not remain buoyant for long periods. In fact, the Spaniards of the conquest had been considerably - 192 surprised at the great size of the Peruvian and Ecuadorian rafts, while two early 18th century chroniclers had stated that these vessels made voyages with cargo from Guayaquil to Puna, Tumbez and even Paita in northern Peru. ‘Quite long voyages’ would therefore have been possible, even to the islands of Polynesia. Ample supplies of food and water could easily have been carried. If a balsa raft had been dismasted, the set of the current was likely to have carried it to the vicinity of the Marquesas Islands.

In 1947, the Norwegian scholar Thor Heyerdahl (1950) and five companions demonstrated the validity of Hornell's ideas when they sailed the Kon-Tiki, a replica of a balsa raft of Spanish times, from Callao, Peru, to Raroia Atoll, Tuamotu Archipelago, a distance of 4,300 miles. The Kon-Tiki was towed out of Callao to begin its voyage because its crew did not then know how to use its centreboards or guaras to steer it. Raroia was reached in 101 days. Specimens of several plants, including the sweet potato, survived the voyage.

Later, Heyerdahl (1952:443-5) pointed out that since Buck had suggested that two-way Polynesian voyagers might have taken the seeds of the gourd to South America, thousands of gourd shell pieces had been found in archaeological sites at Huaca Prieta, northern Peru. They had been dated to between 3,000 and 5,000 years earlier. It was therefore obvious that American Indians had had the gourd long before the Polynesian era began. Hence, any Polynesian sailings to America would have been far too late to account for its presence in America. On the other hand, Peruvian voyagers of A.D. 500 or thereabouts could well have carried both it and the sweet potato to ‘any of their settlements in Polynesia’, Heyerdahl added. Since then the botanist Whistler (1990b) has shown that Lagenaria siceraria did not exist prehistorically in Western Polynesia. So, like the sweet potato, it could only have reached Eastern Polynesia from South America.

In the 26 years after the Kon-Tiki's voyage, a dozen other American-style rafts made voyages from Peru and Ecuador to Polynesia. Several got as far as Australia (Heyerdahl 1978:51). Meanwhile, a French adventurer, Eric de Bisschop (1959), made the only realistic attempt to prove the feasibility of a Polynesian voyage to South America and back. With a crew of four, de Bisschop left Tahiti in November 1956 in a bamboo raft complete with centreboards which, in his view, was what the ancient Polynesians would have used. Having sailed southward to well beyond Ra'ivavae, in the Austral Islands, de Bisschop picked up winds that conveyed the raft towards South America in about latitude 35° S. After about seven months, the crew had to radio for help when the raft began to sink in a heavy storm south of the Juan Fernandez Islands. They reached Valparaiso, some 400 miles away, by courtesy of the Chilean Navy.

- 193

A new raft was built at Constitución and sailed the 1,500 miles from there to Callao to attempt a return voyage to Polynesia (Danielsson 1960). This began in mid-April 1958. More than 100 days later, near Starbuck Island, de Bisschop's second argosy also began to sink. The crew built a smaller raft from the sinking one and transferred to it, and did the same again a few days later. Finally, 138 days out of Callao, the voyage ended on the reef of Rakahanga Atoll, northern Cook Islands, where de Bisschop was drowned.

De Bisschop's exploits undoubtedly demonstrated that, with much more luck than he had, a party of ancient Polynesians could, conceivably, have sailed from Polynesia to South America and back, bringing the sweet potato, gourd and soapberry with them. On the other hand, those exploits also showed that, without instruments, any Polynesian voyaging party which reached South America from the Marquesas Islands or anywhere else would have been most unlikely to get back to its starting point. However, neither de Bisschop's exploits nor those of Heyerdahl and his successors threw any positive light on where in Eastern Polynesia the sweet potato, gourd and soapberry had been introduced.

In 1953, Heyerdahl and two archaeologists uncovered the first evidence to indicate that prehistoric voyagers had made frequent voyages to the Galapagos Islands from Ecuador and Peru. On an expedition to those islands, they recovered more than 2,000 potsherds from four archaeological sites on three different islands. The sherds were identified as belonging to several different cultures and periods of prehistoric Ecuador and Peru. The earliest was that of coastal Tiahuanaco of the 10th century A.D. Because of the shallowness of the sites and the long time-span the sherds represented, the discoveries were interpreted as indicating periodical landings and short encampments by fishing expeditions from the mainland rather than permanent habitations (Heyerdahl and Skjölsvold 1956).

The Galapagos discoveries lent substance to Hornell's suggestion that a well-stocked raft could have drifted to Polynesia in prehistoric times. Such a voyage could have been facilitated by a strong current that apparently flows southwestward from the Galapagos towards Easter Island during the onset of the El Niño phenomenon. In 1838, a British naval vessel, HMS Sulphur, was evidently caught up in it while trying to reach Callao from the Galapagos. The current carried the ship to within 400 miles of Easter Island before it could head for its intended destination (Belcher 1843:I:191). The same current also seems to have enabled several early 19th century American whalers to sail from near the Galapagos to Easter Island in only nine to sixteen days (Langdon 1984:25). 1

The possible implications of Heyerdahl's Galapagos discoveries seem to have had little impact on the thinking of most Polynesianists concerned with - 194 the prehistory of Eastern Polynesia. Such scholars have been much more influenced by the findings of a study by the linguist Samuel H. Elbert (1953) made at much the same time. The study sought to discover ‘something about the interrelationship’ of 20 Polynesian languages by ‘counting likenesses’ in their basic vocabularies and computing the relative nearness to and farness from a reconstructed proto-language. Elbert's basic vocabularies contained 202 words and included ‘sweet potato’. His main conclusions on the languages of Eastern Polynesia (i.e. those spoken east of Samoa other than New Zealand Maori) were:

  • 1. All were descended from a single proto-language, Proto-Eastern Polynesian, which had probably developed ‘in the centrally located Society Islands’.
  • 2. The ‘Eastern people’ had lived together long enough to ‘simplify their structure’ and to achieve ‘a decided lexical homogeneity’ - 46 words found in at least two Eastern languages being unknown in those of Western Polynesia.
  • 3. Proto-Easter Islanders were probably the first to leave ‘the Eastern dispersal point’- in about 300 to 400 A.D. They were followed by proto-Marquesans who settled the Marquesas and Mangareva. Thereafter Hawaii, New Zealand, the Tuamotu Archipelago and Rarotonga were settled.

Several years after Elbert's study was published, linguists discovered that the Easter Island language, Rapanui, preserved the Proto-Polynesian glottal stop. This made it unique among the Eastern Polynesian languages. However, the fact that Elbert had made his study without taking this significant phonological feature into account did not undermine the validity of his findings in the eyes of his contemporaries. On the contrary, the prehistorian Roger Green (1966:17) claimed that the retention of the glottal stop in Rapanui ‘lent weight to Elbert's position’. Moreover, Green thought Elbert's date of 300 to 400 A.D. for the departure of the Proto-Easter Islanders from their ‘Eastern dispersal point’ was ‘reasonably well in line with the meagre existing archaeological estimates for settlement of [their] island’. But, like Elbert, Green overlooked one crucial fact: the Proto-Eastern Polynesians and Proto-Easter Islanders could not have had a word for sweet potato in 300-400 A.D. unless (a) Polynesian voyagers had already carried the plant from South America to the proto-Eastern Polynesian homeland, or (b) American Indians had done it. So there was, in fact, a strong possibility that some of Elbert's conclusions about the Eastern Polynesian languages were wrong.

Nowadays, the assumed presence of the gourd in the Proto-Eastern Polynesian homeland must also be accounted for. This plant is called hue in all languages of the region other than Rarotongan, which has the cognate 'ue. - 195 Scholars (e.g. Green 1988:51) acknowledge that the Eastern Polynesian term is a reflex of Proto-Polynesian *fue ‘a creeping plant’, whose reconstruction is based on the occurrence of fue with that meaning in several Western Polynesian languages such as Tongan, Samoan and Futunan. This therefore means that early Western Polynesian immigrants to Eastern Polynesia must have found the gourd on their arrival there; that they gave it their name for creeping plant because of some fancied resemblance; and that they and their descendants thereafter used this name in other islands, whether they carried the gourd there themselves or not. If their new homeland was, indeed, in the ‘centrally located Society Islands’, as seems likely, then the gourd was obviously brought there by someone else. Hence a scenario is called for that would have enabled early Western Polynesian immigrants to Eastern Polynesia to adopt the South America-derived gourd under a name of their own and the American sweet potato under a borrowed American name. The following four-point scenario is suggested:

  • 1. Early in the Christian era, a party of male and female American Indians drift to Easter Island in attempting to reach the Galapagos Islands. The supplies they have on their raft include sweet potatoes and gourds and enable them to establish a settlement on that island.
  • 2. Over the ensuing centuries, descendants of Easter Island's American Indian settlers are drifted with specimens of their sweet potatoes and gourds to some of the islands to leeward. Directly or indirectly, they reach the Society Islands and are well established there when the first immigrants from Western Polynesia arrive.
  • 3. The Western Polynesians intermarry with the American Indians. A new language develops that is basically Polynesian but simpler in its structure than those in the west. The language has many new words such as kumara, while some old ones, such as fue, are given new meanings. Sometimes they acquire new pronunciations, too, because the American Indians cannot cope with f, for example, in certain circumstances (see Biggs 1978:709).
  • 4. Over time, the proto-Eastern Polynesians and their descendants (a mixed-race people) carry their new language with or without the sweet potato and gourd, to other islands. The last place they reach - probably without any gourds or sweet potatoes - is Easter Island. 2 There they find a long-established, long-isolated American Indian population that has been cultivating both plants for hundreds of years. The American Indians have no resistance to the alien diseases that the Polynesians bring with them and begin to die off. This process quickens and is completed when European ships arrive in the 18th century. Meanwhile, the Polynesian names for sweet potato and gourd, kumara and hue respectively, become
- 196
  • entrenched on Easter Island, and so give the superficial impression to European scholars of later times that both were in use when the island was first occupied.

The evidence relating to the soapberry is consistent with the foregoing scenario. Rapanui marikuru ‘soapberry’ is obviously cognate with Marquesan koku'u and so indicates prehistoric contact between Easter Island and the Marquesas Islands. But whereas the trade winds could have easily carried aboriginal Easter Islanders to the Marquesas, Marquesans could only have travelled in the opposite direction with enormous difficulty. Hence, there is little doubt that the Rapanui term is ancestral to the Marquesan, and that the seeds of the soapberry were carried to the Marquesas from Easter Island, not the reverse. Moreover, as the soapberry was abundant in that archipelago in the late 18th century, it is clear that its seeds must have arrived there a long time ago. This suggests, in turn, that the soapberry trees of Mangareva, Pitcairn Island, the Society, Hawaiian and Cook Islands all had the same origin, and that those of the Society and Hawaiian Islands had plenty of time to develop the distinctive characteristics that Brown and Papy noted.

Two other items of evidence support this scenario: (a) the people who applied the terms kokuru, kokuu and 'o'uru to Suriana maritima in the Tuamotu Archipelago and Society Islands obviously came from the east, and (b) those who called the soapberry ake(ake) and a'e(a'e) in the Cook, Society and Hawaiian Islands equally clearly came from the west. As the people from the west could have had no tradition of using the soapberry because of its non-existence in Western Polynesia, it is not surprising that the tree was of little consequence to the Polynesians of the Cook, Society and Hawaiian Islands at the time of European contact.

In summary, the evidence relating to the long-neglected soapberry supports the present writer's conclusions from other evidence that South American Indians played a significant role in the early prehistory of Eastern Polynesia (Langdon 1994,1995a; Langdon and Tryon 1983). An increasing number of scholars is beginning to share this view. For example, Edwin N. Ferdon Jr (1993:138), a retired archaeologist, says in a recent book that until firmer evidence is forthcoming as to the sources, times, and means of introduction into Polynesia of the sweet potato and other American Indian domesticates, ‘a more balanced approach’ to the question of American Indian influence in the region is justified than has been apparent in the past.

- 197
  • Beaglehole, J. C., (ed.), 1961. The Journals of Captain James Cook on His Voyages of Discovery: The Voyages of the Resolution and Discovery. Cambridge: Hakluyt Society.
  • Belcher, Edward, 1843. Narrative of a Voyage round the World performed in Her Majesty's Ship Sulphur… 1838-1842. 2 vols. London: H. Colburn.
  • Bennett, Frederick Debell, 1840. Narrative of a Whaling Voyage round the Globe from the Year 1833 to 1836. 2 vols. London: R. Bentley.
  • Biggs, Bruce, 1978. The history of Polynesian phonology, in S. A. Wurm and Lois Carrington (eds), Second International Conference on Austronesian Linguistics: Proceedings. Canberra: Department of Linguistis, RSPacS, Australian National University, pp. 691-714.
  • Bisschop, Eric de, 1959. Tahiti-Nui. London: Collins.
  • Brown, Forest B. H., 1935. Flora of Southeastern Polynesia: III. Dicotolydons. Bulletin 130. Honolulu: B. P. Bishop Museum.
  • Buck, Peter H.(Te Rangi Hïroa), 1938a. Vikings of the Sunrise. New York: Frederick A. Stokes Company.
  • —— 1938b. Ethnology of Mangareva. Bulletin 157. Honolulu: B. P. Bishop Museum.
  • Butteaud, Edouard, 1891. Flore tahitienne. Pape'ete: Imprimerie du Gouvernement.
  • Carlquist, Sherwin, 1967. The biota of long-distance dispersal.V. Plant dispersal to Pacific Islands. Bulletin of the Torrey Botanical Club, 94:129-62.
  • Cooke, George H., 1899. Te Pito te Henua, known as Rapa Nui, commonly called Easter Island. United States National Museum Annual Report, 1897, part 1:689-723.
  • Danielsson, Bengt, 1960. From Raft to Raft. London: George Allen and Unwin.
  • [Davies, John], 1851. A Tahitian and English Dictionary. Tahiti: London Missionary Society's Press.
  • Day, A., 1953. Admiralty Distance Tables, Volume 5. Pacific Ocean. 2nd edn. London: Hydrographic Department, Admiralty.
  • Degener, Otto, 1945. Plants of Nawaii National Park. Ann Arbor: Edward Brothers. (Reprint of 1930 edn).
  • Dixon, Roland B., 1932. The problem of the sweet potato in Polynesia. American Anthropologist, n.s.34:40-66.
  • Drake del Castillo, Emanuel, 1893. Flore de la Polynésie française. Paris: G. Masson.
  • Elbert, Samuel H., 1953. Internal relationships of Polynesian languages and dialects. Southwestern Journal of Anthropology, 9:147-73.
- 198
  • Emory, Kenneth P., 1947. Tuamotuan plant names. Journal of the Polynesian Society, 56:266-77.
  • Englert, Sebastian, 1938. Diccionario Rapanui-Español. Santiago de Chile: Prensas de la Universidad de Chile.
  • —— 1978. Idioma Rapanui: Gramática y Diccionario del Antiguo Idioma de la Isla de Pascua. Santiago de Chile: Ediciones de la Universidad de Chile.
  • Ferdon, Edwin N. Jr, 1993. Early Observations of Marquesan Culture, 1595-1813. Tucson: University of Arizona Press.
  • Flenley, John R., 1993. The present flora of Easter Island and its origins, in Steven Roger Fischer (ed.), Easter Island Studies: Contributions to the History of Rapanui in Memory of William T. Mulloy. Oxbow Monograph 32. Oxford: Oxbow Books, pp.7-15.
  • Forster, George, 1786. Florulae lnsularum Australium Prodromus. Göttingen: Johann Christian Dieterich.
  • Friederici, Georg, 1929. Zu den vorkolumbischen Verbindungen der Südsee mit Amerika. Anthropos, 24:441-87.
  • Fuentes, Francisco, 1913. Reseña botanica sobre la Isla de Pascua. Boletín del Museo Nacional de Chile, 5:320-37.
  • Green, Roger, 1966. Linguistic subgrouping within Polynesia: the implications for prehistoric settlement. Journal of the Polynesian Society, 75:6-38.
  • —— 1988. Subgrouping of the Rapanui language of Easter Island in Polynesian and its implications for East Polynesian prehistory, in Claudio Cristino and others (eds), First International Congress: Easter Island and East Polynesia. 1: Archaeology. Facultad de Arquitectura y Urbanismo, Instituto de Estudios Isla de Pascua, Santiago, Chile, pp.37-57.
  • Guppy, H. B., 1906. Observations of a Naturalist in the Pacific between 1896 and 1899. Vol. II. Plant-Dispersal. London: Macmillan.
  • Heyerdahl, Thor, 1950. The Kon-Tiki Expedition. London: George Allen and Unwin.
  • —— 1952. American Indians in the Pacific. London: George Allen and Unwin.
  • —— 1978. Early Man and the Ocean. London: George Allen and Unwin.
  • ——and Arne Skjölsvold, 1956. Archaeological evidence of pre-Spanish visits to the Galapagos Islands. Memoirs of the Society for American Archaeology 12.
  • Hornell, James, 1945. Was there pre-Columbian contact between the peoples of Oceania and South America? Journal of the Polynesian Society, 54:167-91.
  • [Janeau, Vincent Ferrier], 1908. Essai de Grammaire de la Langue des Iles Gambier ou Mangareva [and] Dictionnaire Mangarévien-Français. Braine-le-Comte: Zech et fils.
  • Jardin, Edelstan, 1857. Essai sur l'histoire naturelle de l'Archipel de Mendana, 2e Partie: Botanique. Mémoires de la Société Impériale des Sciences Naturelles et Mathématiques de Cherbourg, 5: 289-331.
  • Langdon, Robert, 1975. The Lost Caravel. Sydney: Pacific Publications.
  • ——(ed.), 1984. Where the Whalers Went: An Index to the Pacific Ports and Islands Visited by American Whalers (and Some Other Ships) in the 19th Century. Canberra: Pacific Manuscripts Bureau, Australian National University.
  • —— 1988. The Lost Caravel Re-explored. Canberra: Brolga Press.
- 199
  • —— 1994. Further thoughts on the terms hanau eepe and hanau momoko and why they should mean ‘long ears’ and ‘short ears’: reply to Emily Mulloy. Rapa Nui Journal, 8(3):75-8.
  • —— 1995a. New light on Easter Island prehistory in a ‘censored’ Spanish report of 1770. Journal of Pacific History, 30(1):112-20.
  • —— 1995b. The significance of Basque genes in the prehistory of Easter Island. Rapa Nui Journal, 9(1):9-15.
  • —— 1995c. Pakomio Maori: red-haired, blue-eyed key to Easter Island's prehistoric past. Rapa Nui Journal, 9(4): 109-17.
  • —— and Darrell Tryon, 1983. The Language of Easter Island: Its Development and Eastern Polynesian Relationships. Monograph 4. La'ie, Hawai'i: Institute for Polynesian Studies.
  • Lewis, Clinton, and J. D. Campbell, 1963. The Oxford Atlas. Oxford: Oxford University Press.
  • Métraux, Alfred, 1940. Ethnology of Easter Island. Bulletin 160. Honolulu: B. P. Bishop Museum.
  • Oliver, Douglas L., 1974. Ancient Tahitian Society. Canberra: Australian National University Press.
  • Papy, H. René, 1954. Tahiti et les Îles Voisines: La Végétation des Îles de la Société et de Makatea (Océanie Française). Toulouse: Les Artisans de l'lmprimerie Douladoure.
  • Poole, A. L. and N. M. Adams, 1963. Trees and Shrubs of New Zealand. Wellington: Government Printer.
  • Pukui, Mary Kawena and Samuel H. Elbert, 1977. Hawaiian Dictionary. Honolulu: University Press of Hawai'i.
  • St. John, Harold, 1988. Census of the Flora of the Gambier Islands, Polynesia. Honolulu: The Author.
  • Schuhmacher, W. W., 1978. Un vocabulario inédito de la Isla de Pascua compuesto por Johann Reinhold Forster durante el segundo viaje de Cook. Moana 1, No. 12 (Montevideo).
  • Seemann, Berthold, 1865-73. Flora Vitiensis. London: Reeve.
  • Sheahan, George M., 1952. Marquesan Source Materials [including Account of the Marquesas Islands compiled by Samuel Greatheed from information supplied by William Pascoe Crook. Original ms.in Mitchell Library, Sydney]. Quincy, Mass.: The Author.
  • Smith, A. C, 1985. Flora Vitiensis Nova: A New Flora of Fiji. Vol. 3. Lawai, Kaua'i, Hawai'i: Pacific Tropical Botanical Garden.
  • Sudzuki H., Fusa, 1979. Vegetación arborea y erosión en la Isla de Pascua. Boletín Técnica, Facultad Agronómica de la Universidad de Chile, 45:3-8.
  • Sykes, W. R., 1970. Contributions to the Flora of Niue. Bulletin 200. Christchurch: New Zealand Department of Scientific and Industrial Research.
  • Thomson, William J., 1891. Te Pito te Henua, or Easter Island. Annual Report of the Board of Regents of the Smithsonian Institution for the Year Ending June 30, 1889, pp. 447-552.
- 200
  • Towle, Margaret A., 1961. The Ethnobotany of pre-Columbian Peru. Chicago: Aldine Publishing Co.
  • Tregear, Edward, 1899. A Dictionary of Mangareva. Wellington: Government Printer.
  • Ugent, Donald, Shelia Pozorski and Thomas Pozorski, 1986. Archaeological manioc (Manihot) from coastal Peru. Economic Botany, 40:78-102.
  • Wagner, Warren L., Derral R. Herbst and S. H. Sohmer, 1991. Manual of the Flowering Plants of Hawai'i. Honolulu: University of Hawai'i Press.
  • Whistler, W. Arthur, 1990a. Ethnobotany of the Cook Islands. Allertonia, 5:347-424.
  • —— 1990b. The other Polynesian gourd. Pacific Science, 44:115-22.
  • Wiggins, Ira L. and Duncan M. Porter, 1971. Flora of the Galapagos Islands. Stanford, Calif.: Stanford University Press.
  • Wilder, Gerrit Parmile, 1931. Flora of Rarotonga. Bulletin 86. Honolulu: B. P. Bishop Museum.
  • Yuncker, T. G., 1959. The Plants of Tonga. Bulletin 220. Honolulu: B. P. Bishop Museum.
1   Nineteenth century American whalers which made voyages from near the Galapagos Islands to Easter Island in from nine to 16 days were: George and Susan, 1823; Lydia, 1827; Eagle, 1828; Aurora, 1832; Hector, 1833; Abigail, 1838; Navigator, 1842.
2   The present writer has argued elsewhere (e.g. Langdon 1975, 1988, 1995a, 1995b, 1995c) that Polynesian-speaking people did not reach Easter Island until the late 17th century and that they were of part-Spanish descent from the crew of the Spanish caravel San Lesmes, which was lost in the eastern South Pacific in 1526.