透過來源分析討論，西南太平洋史前Lapita社會網絡：由新喀里多尼亞的遺址群談起

Reconstructing Prehistoric Lapita Social Networks through Source Analysis: A Case Study from New Caledonia Lapita Sites

本文綜論以新喀里多尼亞的大範圍陶器分析結果來舉例，說明如何以Lapita陶器岩象分析與化學成分分析結果，來檢討過去對於史前大洋洲的移民模式所形成的假說有何需要補充與改進的地方。根據現有的成果，可以看出忠誠群島很早就有人在當地定居，但是當地的陶器並未出現在新喀里多尼亞的主島或是南部外海的松島的遺址中。松島則是取得了絕大多數來自距離四百公里遠的主島西北部所生產的陶器，而這些陶器絕少出現在忠誠群島的遺址中。位於主島南端的Goro遺址，取得陶器的地區則大多數落於從西北到東北部的沿岸，與松島Vatcha遺址取得陶器的地點幾乎沒有重複。這種分布模式讓我們得以對當地陶器傳送模式的複雜性提出新的看法。本文也特別著重呈現不同分析方式所遇見的困難，以及對部分困難提出可行的解決之道。

This article summarizes the results of a wide range of analyses on pottery assemblages from New Caledonia, including both petrographic analysis and chemical composition analysis. This study illustrates how such techniques can be employed to review past hypotheses about the migration patterns of prehistoric Oceania, and how such studies can supplement and improve our current understanding of the social relationships between the populations involved. Despite populations settling in the Loyalty Islands early on, our current results indicate that pottery made from this area was not transferred to Grande Terre of New Caledonia, or to theÎle des Pins located to the south. Lapita communities located at Vatcha on theÎle des Pins acquired most of their pottery from the Diahot River valley located at the northwestern part of Grande Terre some 400 kilometers away, yet pottery made from this region rarely appears in the Lapita sites of the Loyalty Islands. The Goro community located at the southern end of Grande Terre received most of their pots from various areas located along the north coast, from the northwest to the northeast, yet like communities in the Loyalties, they rarely received any pots from the Diahot River valley. This distribution pattern allows us to put forward new perspectives on the complexity of the local pottery exchange patterns. This article also details difficulties that were encountered when using different analytical methods, and proposes solutions to overcome some of these issues.