Prehistoric settlement and economy in a tropical small island environment : the Banks Islands, Insular Melanesia

Abstract

Searching for the Islas de Salomon and their fabulous wealth, the Great Southern Continent beloved of European cartographers and the multitudes of souls surely awaiting salvation there, Quiros the third of the Spanish voyages into the southerly latitudes of the Pacific, stumbled upon the groups of tiny oceanic islands at the intersection of the Solomons and New Hebrides archipelagos. Three or four millennia earlier these same island groups must have been significant links in the initial human colonization of the island Pacific. Their linking role might well have been invoked more than once because they are near the centre of an area bounded by Melanesia, Micronesia and Polynesia. The focus of this study are the Banks Islands, a group of small volcanic islands and atoll-like islets in Insular Melanesia where a variety of ecological situations are manifest, a major source of differentiation being the high-island/low-island dichotomy familiar elsewhere in the Pacific. The research has two major themes: The one concerns the role of the islands in the process of the settlement of the region and the other relates to the processes of economic adaptation of prehistoric communities to small tropical island environments and their changes over time. Various subsidiary themes occur, including the role of ethnographic, or ethnoarchaeological, and ethnohistorical accounts in the interpretation of archaeological data. The main method of enquiry, however, was the collection of archaeological data through survey and excavation. The Banks Islands occur at the north of the New Hebrides archipelago; they comprise a group of several large volcanic islands along with a smaller number of sand quays and raised coral islets. The high islands are characterized by towering peaks, roughly dissected interiors and lush tropical vegetation from high remote fastnesses to tropical strand complexes; rainfall is particularly heavy in the upper reaches and several rivers arise in the centres of the larger islands, depositing at lower attitudes sediments concentrating rich volcanic soils. The narrow coastal flats are bordered by black sand beaches and broken by high rocky headlands; in some places stretches of brilliant white coralline sand are found inside fringing reefs but the major expanses of reef deposits are associated with the atoll-like off-lying islets. These are generally consolidations of recent alluvia but there are also exposures of raised limestone reef formation. The soils of these islets are shallow and infertile but the islets themselves provide access to rich littoral resource areas. Our earliest knowledge of the Banks Islanders is provided by Munilla's brief account of the abduction by Quir6s in April 1606 of men from Gaua in the southern Banks group where they later were exchanged for a pig and other foods (quotation from Kelly above). Two and a half centuries later, at the height of a more intensive labour trade, the islet of Pakea at the entrance to Port Patterson on the northern island of Vanua Lava (a major harbour of the 'blackbirders') was effectively abandoned. Neither does it figure in the accounts of those other soulful traders, the Melanesian Missionaries, who made their headquarters on nearby Mota. Yet here, on Pakea, perhaps the most concentrated evidence of prehistoric human occupation in the group is found. The islet lacks the substantial attributes of the adjacent volcanic high islands, so that more specialized factors must have allowed it to sustain a significant community in the past. Archaeological research has suggested that settlement of this area of Insular Melanesia dates from perhaps three or four millennia ago. Its prehistory has generally been discussed in terms of the ceramics made and used by the prehistoric communities, the two major ones being that associated with the Lapita Complex and the Linear Incised and Applied Relief Ware. The results of the investigations of the Banks Islands indicate that the initial settlers - whose occupation dates from about 3000 BP - used a variety of the latter ceramics related to the Mangaasi Ware of the Central New Hebrides, made a variety of shell tools and ornaments from local resources, including adzes made from the lip as well as the hinge of the Tridacna clams, and that pigs as well as fishes and other marine fauna were staples of their diet. Later in the prehistoric sequence, the use of pottery diminished and probably was discontinued by about two millennia ago and there was a greater concentration of effort toward the production of a more limited range of shell artefacts; there developed the use of basalt for extensive monumental ceremonial centres and large areas of irrigated horticulture. At the same time, there is evident, at Pakea, a change in the proportion of the protein diet obtained from marine as opposed to terrestrial sources. It is suggested that the communities on Pakea developed a specialized role in the production of a limited range of valued commodities and that these were exchanged for other items not produced on the islet. Inferences are made as to the processes promoting this particular adaptation; several variables are explored, including ecological and locational factors as well as cultural criteria. While accounts of the Banks Islands date from early in the seventeenth century AD, few reliable written records exist from before the late 1800's; nevertheless, much of the contemporary economic behaviour of the Banks Island communities can be seen as being of a traditional type and providing a basis for the evaluation of the archaeological evidence of prehistoric adaptive strategies. Systematic study was made of some aspects of fishing practices; especially the patterns of exploitation of marine molluscs by women, and every opportunity was taken in the field to elicit the advice of men and women knowledgeable in traditional areas. An extensive archaeological and ecological survey was undertaken to sample most islands in the northern part of the group along with sections of the main island in the south, Gaua, during 1973. These confirmed and extended the findings of Groube, who had visited the group the previous year, and which had already egated the Shutlers' conclusions that there was no pottery in the area and that archaeological sites were few and insignificant. A major point of departure in the research was the work of Garanger in the Central Islands of the New Hebrides (some four hundred kilometres to the south) which provides a cultural historical framework for the region. However, a major lacuna in earlier studies in the area has been the lack of consideration given to subsistence strategies. These are emphasized in the present research. Because a major part of this emphasis devolves upon marine resources, the seminal work by Reinman on Oceanic fishing was a useful starting point for the discussion. As a result of the surveys, three proposals for continued research were outlined. The first focussed on the socio-ceremonial aspects of the grade organizations and their physical manifestation in the large stone structures associated with the proto and prehistoric settlements throughout the group; the second considered the opportunity to evaluate the chronology of the extensive irrigated gardens(both in use and abandoned)and their development over time both as a particular adaptation to local conditions and as one manifestation of a Pacific-wide phenomenon of which little was known in detail. The third focussed on the variety of coastal settlement sites throughout the group, many of which contained considerable evidence of intensive use of littoral resources in stratified contexts; it was clear that these provided excellent opportunities to explore the development of coastal-oriented economies over time. The third project best fitted the original concerns of the study; it was refined to concentrate the field research upon a series of stratified mounds on Pakea, an atoll-like islet lying off Vanua Lava in the north of the group, and a further series of minor studies, both archaeological and ethnographic, on other islands of the north. An intensive survey was made of Pakea in mid 1974; excavations were completed there and on Rah, Aro and Vanua Lava in April of the following year. Systematic environmental and ethnographic studies were conducted of the Rah littoral and its contemporary pattern of exploitation. While the results of the various studies made elsewhere are incorporated where appropriate, the present discussion focuses on the problems illuminated by the research on Pakea. The first part of the thesis places in archaeological context the two main themes: The prehistory of settlement of the region and the processes of economic adaptation to the small tropical island environments (Chapters I and II). The second and third parts focus on the place of Pakea Islet in the Banks Islands (Chapter III)and describes the results of the archaeological and environmental surveys and detailed mapping and test excavations carried out there, concluding with a discussion of the major excavations of Area A of the mound site, BN-PK-1 (Chapters IV and V). The fourth part covers the various field and laboratory analyses of the major categories of excavated materials, drawing conclusions concerning the chronology of the settlement of the islet from the excavation of the mounds and its secondary position in the sequence of human colonization of the region (Chapter VI), the character and longevity of the ceramic sequence (Chapter VII), the uses made of artefacts fabricated from stone (Chapter VIII) and those widely utilized shells of marine molluscs (Chapters IX and X), concluding with detailed analyses of the contribution to the prehistoric diet of various sources of foodstuffs which were manifest by the archaeological record (Chapter XI). The fifth part draws together the results of the various detailed analyses in a summary fashion and discusses their co

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 24 citations exist upstream.