The human colonisation of Oceania was one or the most remarkable acmevements in world prehistory, and the strategies employed during the settlement of the Pacific and subsequent diversification of island societies are key issues in Oceanic archaeology. Once the hundreds of islands across the vast span of Oceania had been settled by small founding groups, subsequent diversification of island societies - especially in the eastern Pacific - is generally thought to have proceeded in relative isolation. However, documenting the changing role and significance of inter-island contacts in sociocultural change has drawn increased attention (Kirch 1988a:4; Terrell 1986). Since isolation is a fundamental condition fostering divergence (Bellwood 1974:278; Cherry 1985:27; Clark and Terrell 1978:307; Green 1968:106; Irwin 1992:200; Kirch 1988b: 106; Kirch and Green 1987:440; Terrell 1986), understanding how variations in the degree of interaction between societies have influenced the evolution of human diversity in the Pacific now becomes a fundamental problem that must be examined for each island sequence. Although few would currently believe that once settled, island societies evolved in total isolation (Terrell et al. 1997), it is now the challenge of archaeologists to empirically assess conditions which promoted varying levels of interaction, for isolation cannot be "adopted as an assumption to guide research" (Schortman and Urban 1987:81; see also Dalton 1977:204).
Isolation has been considered a "circumstance of change" (Irwin 1992:195) and thus not a cause but a condition (Terrell 1986:122). Wide water gaps, as occur increasingly in the eastern Pacific, and rugged terrain (such as valley settings in the Marquesas Islands) do make regular interaction less likely. Additionally, Roger Green (1968:106) has suggested that the geographic barrier of open water between West and East Polynesia "served to maintain the cultural differentiation between them". A similar argument has been advanced for the isolating effect of the 850 to 1000 km water gap between Vanuatu and Fiji, with regard to differentiation between Western and Eastern Lapita (Green 1979, 1982). Kirch has suggested that this water gap "inhibited regular two-way voyaging contacts and the maintenance of exchange relationships" (1988b: 106). ...................