Patterns in glass : obsidian and economic specialisation in the Admiralty Islands

Abstract

This thesis considers the association between western Melanesian ethnographic economic specialisation and prehistoric systems of production and distribution. Contrasting theories for the development of historical specialisation are reviewed and the criticism made that these are chronologically limited to the late Holocene. The statement is made that to fully appreciate temporal change we must expand our view to encompass the preceramic period. Obsidian is one of the few archaeologically visible materials which was distributed in both preceramic and ceramic times. This material is chosen as a “measuring device” to map variation in production and distribution patterns in the Admiralty Islands, Papua New Guinea. A review of ethnographic and anthropological literature revealed that the Admiralty Islands were characterised by a high level of village or lineagebased economic specialisation. Obsidian was one of the materials produced and distributed within this system. A study was carried out on obsidian use at Pamwak Rocksheiter on Manus Island, and at a number of mid to late Holocene localities on Manus and Mouk. Characterisation analysis revealed that offshore obsidian, probably from the Pam Islands, began to be utilised in the terminal Pleistocene. Trends of increasing accessibility through time and a move to incorporate increasing quantities of Lou obsidian were revealed. A significant discovery is the possibility of a major increase in the use of Lou obsidian coincident with the appearance of Lapita. Retouched obsidian blades and microblades were found to be present in only post- Lapita contexts. Chronological change in blade production strategies was revealed on Lou. This involved the development of highly standardised triangular forms by approximately 1600-1300 BP, followed by a simplification of technology as reflected in the appearance of minimally modified tanged forms. This occurred within the last 1000 years and is interpreted as showing increased demand for weapon points. This demonstrates a move toward the form of spearpoint production recorded by nineteenth century ethnographers. The conclusion drawn is that none of the models yet advanced for the devlopment of economic specialisation in Melanesia is adequate for interpreting change in production and distribution in the Admiralty Islands. The roots of economic specialisation may lie further back in time than catered for by existing models.