Abstract
This study focused on prehistoric pottery production patterns in the Lapita and plainware periods from the islands of ‘Upolu and Manono, Samoa. Incorporating a holistic approach to excavated pottery assemblages, stylistic, temper and clay analysis was undertaken to identify whether initial production technology matched a larger regional signature and to test how production strategies changed through the plainware phase.
Ceramics were sampled from Auckland War Memorial museum collections of five previously excavated sites. This encompassed the only known Lapita site of Mulifanua, as well as a range of temporally and geographically distinct plainware sites located along the north coast of Manono and ‘Upolu. A combination of stylistic and physico-chemical techniques were undertaken to determine the full range of production variation present in Samoan ceramics. Stylistic analysis is a common method on Lapita assemblages, providing insight into distinctive cultural markers and regional cultural suites. This technique is, however, limited on plainware assemblages due to restricted vessel forms and an almost complete absence of decoration. Therefore, chemical analysis was undertaken using an electron microprobe on the temper and clay components of 149 sherds to produce data on production patterns associated with the plainware and Lapita phases.
The results of the stylistic analysis confirm a lack of distinctive features on plainware pottery, and argue against the thin/thick ware division established for pottery assemblages in the archipelago. Two of the plainware sites, Falemoa and Jane’s Camp show strong similarities in forms of decoration to the only known Samoan Lapita site of Mulifanua. A red decorative slip is recorded from these three sites, matching similar descriptions from assemblages on other islands within the archipelago and further afield. The presence of a carinated vessel from Jane’s Camp suggests continuity in vessel forms between Lapita and early plainware sites. Early or transitional plainware sites might therefore be characterised by more diverse vessel forms then is currently established.
The results of the chemical analysis indicate that almost all pottery was produced locally, with the number of resource procurement zones declining over time and a change in production techniques. Initial production utilised a variety of sources, most centred on the coast. Through the plainware period the focus shifts towards inland sources, with pottery produced at the end of the sequence from Sasoa’a showing a marked change to local trachytic tempers. The homogeneous nature of ‘Upolu makes differentiating clay sources difficult, but they appear to match a pattern of local production.
This research shows how production patterns for initial ‘Upolu settlement were established, including the plainware period, an area currently understudied in Pacific archaeology. Initial colonisation by a Lapita people at Mulifanua was shown to be reflective of a larger regional colonising strategy, utilising the same production technologies and stylistic elements. There appears to be strong continuity in pottery production between the Lapita colonisers and the subsequent plainware settlements. Pottery production is local, with vessels becoming thicker and more heavily tempered over time, suggesting either changes to resource access or the exhaustion of quality clays. Overall production patterns for the two islands match previous work undertaken from other islands in the archipelago. This research provided a key quantifiable dataset and offers the opportunity to further expand prehistoric ceramic studies from Samoa. This thesis has shown continuity in pottery production between Lapita and plainware phases, and suggests the Samoan identity is descendant from the first people to colonise its islands.