Pre-Lapita valuables in island Melanesia

Abstract

An unusual obsidian stemmed tool found by Jim Specht at Boku Hill, West New Britain, Papua New Guinea, provides evidence for the existence of valuables in the pre-Lapita period. The large amount of skill, care, and effort invested in the manufacture of this large artefact combined with its symmetry and fragility imply that, unlike the other stemmed tools found at the site, Specht’s find was not used in an utilitarian context. As in the case of axes in the Highlands of New Guinea, stemmed artefacts were therefore both useful tools and non-utilitarian objects which circulated in some form of ceremonial exchange system and for which control of the Kutau/Bao obsidian sources may have been important. The occurrence of valuables pre-dating Lapita pottery demonstrates that social systems based on ceremonial exchange and prestige goods were not imported, but developed in situ. TORRENCE, ROBIN, 2004. Pre-Lapita valuables in island Melanesia. In A Pacific Odyssey: Archaeology and Anthropology in the Western Pacific. Papers in Honour of Jim Specht, ed. Val Attenbrow and Richard Fullagar, pp. 163–172. Records of the Australian Museum, Supplement 29. Sydney: Australian Museum. Records of the Australian Museum, Supplement 29 (2004): 163–172 ISBN 0-9750476-3-9 www.amonline.net.au/pdf/publications/1413_complete.pdf It is well known among Melanesian archaeologists that Jim Specht has an uncanny ability to find important sites, often those with the early dates that are highly prized in this discipline. He excavated Misisil Cave, which for many years represented the earliest occupation of the Papua New Guinea islands (Specht et al., 1981; 1983); he began work in the Yombon area and found pottery (probably Lapita plainware) far inland of its normal distribution (Specht et al., 1981); he led Christina Pavlides to sites near Yombon mission where she found the earliest evidence for colonization of New Britain (Pavlides, 1999; Pavlides & Gosden, 1994); and he also began work at Kilu Cave on Buka, which later produced the earliest date for the Solomon Islands (Wickler & Spriggs, 1988). Although his achievements are perhaps not unexpected, since he has been exploring Melanesia for many years, I believe he has a gift. Not only did he, “by accident,” find obsidian stemmed tools in the base of a new latrine at Bitokara Mission, an event which led to the excavation of the type site for the region and set the basis for the use of volcanic stratigraphy within landscape archaeology (cf. Specht et al., 1988; Specht et al., 1991; Torrence et al., 1999a, 2000), but recently he also made a find with much significance for Melanesian prehistory, as I hope to demonstrate in this paper. A remarkable discovery During the 1999 field season, Jim and I visited the newly developed Garu Plantation, part of the Numundo Group owned by New Britain Palm Oil Ltd (Torrence et al., 1999a) with a number of team members. The manager, Kefu Boromana took us to the top of Boku Hill where he had observed obsidian artefacts in an area recently levelled by bulldozer for a new house (Fig. 1). The hill is approximately 80 m a.s.l., overlooks an extensive peat swamp and has views stretching to the sea on the west side of the Willaumez Peninsula (about 10 km away). A small raised area on the north west side had been pushed over the edge of a steep cliff to provide the required level surface. As a result, most of the archaeological material dating to at least the past 6,000 years had been removed. Obsidian flakes were thinly 164 Records of the Australian Museum, Supplement 29 (2004) scattered over the damaged area and careful survey also produced several pieces of Lapita pottery and a small stemmed tool (Fig. 2). Jim found a patch where the bulldozer had removed the distinctive W-K2 tephra dated to c. 3,600 B.P. (Torrence et al., 2000) revealing the red-brown clay beneath. He targeted this area for his search and before long had recovered most of a second and very distinctive stemmed tool with a fresh break. The remainder was found by Leigh Jago a week later in spoil about 100 m away. The artefact has since been restored by the Materials Conservation Division at the Australian Museum and returned to the National Museum and Art Gallery, PNG (Fig. 3). About a month later, one of the workmen sold a third stemmed tool, supposedly found at the same location, to a local collector (Fig. 4). All in all, then, three stemmed tools with widely ranging shapes and sizes have been found in the disturbed area of Boku Hill. Jim’s discovery of a highly worked stemmed tool on Boku Hill (FABN) on Garu Plantation, West New Britain (Figs. 1 and 3) has opened up new possibilities for understanding the nature and evolution of prehistoric societies in Melanesia. In this paper I argue that the three stemmed tools provide significant new information about this enigmatic class of artefact and that these new data raise important questions about the nature of society in West New Britain prior to Lapita pottery. On the basis of this new find combined with recent research, I conclude that the concept of “special artefacts” or “valuables” was present in West New Britain in the period before 3,600 cal. B.P. (cf. Araho et al., 2002). Fig. 1. Location of the Willaumez Peninsula and sites mentioned in the text. Obsidian sources are in italics. Shaded area shows distribution of obsidian outcrops. Boku Hill The Boku Hill site represents an unusual environmental setting for both Lapita pottery and stemmed tools. Along with the three stemmed tools, eight potsherds (five plain and one each dentate-stamped, incised rim and notched rim) were found in the bulldozed area. We can only provide a dated archaeological context for these surface finds by correlation with similar material recovered in the local region. Using this approach, summarized below, it is clear that the stemmed tools derive from a context which precedes that of the Lapita pottery which is clearly separated from it at this site by the distinctive W-K2 tephra. Three small test pits (a total of 2.5 sq m) were excavated on Boku Hill, but these produced disappointing results. Within the most successful test pit (XXV), a good sequence of Holocene tephras derived from the Witori volcano was preserved in situ, thereby providing an excellent source for relative dating using reference to the well-studied tephra stratigraphy of the region (e.g., Machida et al., 1996; Torrence et al., 1999a, 2000). Obsidian artefacts were recovered from soil underneath the W-K1 tephra as well as on soils developed on the following tephras: W-H4/5; WK4; W-K2; W-K1. The relative age of the artefacts can be determined with reference to the dates for the tephras listed in Table 1. No stemmed artefacts or pottery were recovered from the test pits. The data nevertheless demonstrate that the site had been used periodically through much of the Holocene with a significant break between the W-K3 and W-K4 tephras. Table 1. Approximate dates for Witori tephras (in calendar years) based on radiocarbon dates presented in Torrence et al. (2000) and Machida et al. (1996). tephra date (cal. B.P.)

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.