ALTHOUGH NEAR OCEANIA has been populated for over 35,000 years, the settlement of the islands east of the main Solomon Island chain, known as Remote Oceania, occurred about 3000 years ago. The archaeological signature of these colonizers is a distinctive type of pottery called Lapita. It is found from Aitape on the north coast of New Guinea in the west to Samoa in the east in contexts dated between c. 3350 and 2350 B.P.
Archaeologists working in the region today are attempting to understand the nature of the societies that produced and used Lapita pottery and the nature of the interactions among these societies. Similarities in pottery decoration, in particular dentate stamped motifs in association with radiocarbon estimates, have been used by archaeologists to identify the spread of settlement over the Pacific and also the presence of inter-island interaction and exchange among geographically separated areas. On the basis of such stylistic similarities and to a lesser extent on the nature of inter-island exchange (Green 1978:3), these areas have been grouped into a number of Lapita provinces (see Fig. 1). Terms such as "stylistic provinces," "interaction spheres," and "exchange networks" are used, sometimes interchangeably, in the literature. For instance, the terms "Far Western Lapita" (or "Early Western Lapita" as Spriggs [1995:116] prefers), "Western Lapita," "Southern Lapita," and "Eastern Lapita" have been given to "geographical sub-regions of Lapita" or "provinces" (Kirch 1997:58, 71). Yet they are also seen as "sub-styles" of Lapita, which have both geographical and chronological significance (Spriggs 1995:116). Regions thus delimited have also been used to describe exchange networks or different interaction spheres.
[FIGURE 1 OMITTED]
As outlined below, these provinces are important heuristic devices in modeling the process of colonization and the nature of group interactions. It should be noted that in other regions, such as Europe or America, archaeologists have recognized that migrationist models do not adequately explain the spatial distribution of materials or the spread of agriculture. This has led them to examine other explanatory mechanisms. Yet in the Pacific, migrationist models have "explanatory power" (Spriggs 1989) as here we are dealing with a rapid colonization movement into unoccupied Remote Oceania. Thus in developing these models of colonization and interaction in Remote Oceania, language, material culture, and subsistence have been grouped together. Where this grouping is not so straightforward is in Near Oceania, which has been occupied for over 35,000 years.
A major problem in assessing the changing nature of Lapita interaction within Near Oceania is the lack of sites showing stylistic change over time within the Bismarck Archipelago. The definition of provinces was based on single-phase sites showing little change within an assemblage. An attempt is made here to redress this problem by presenting the results of a stylistic analysis on pottery assemblages with long sequences from West New Britain, and to critically assess whether they conform to the defined sequences already established for the region. A comparison is made with assemblages from Western Melanesia in an attempt to review the processes of interaction among Lapita communities based on ceramics.
The paper is structured into four parts. The first reviews the models for the Lapita colonization of the western Pacific, the definition of Lapita provinces, and their relationship to processes of interaction. It discusses how an analysis of longer pottery sequences in the Bismarck Archipelago can assess the insularity of Lapita provinces. The second part sets out the assemblages under study, while the third presents the results of the stylistic analysis. This is followed by a discussion of the results and a re-assessment of the regionalization of the western Pacific.
MODELS OF COLONIZATION AND INTERACTION
There are three main models for the Lapita colonization of Remote Oceania. …