Middle to Late Holocene near-shore foraging strategies at Caution Bay, Papua New Guinea

Highlights

• Caution Bay, PNG, provides evidence for c.5000 years of marine resource use.

• Biometric analyses are used to investigate mollusc harvesting at the site of Tanamu 1.

• Size/maturity varies between the pre-Lapita, Lapita and post-Lapita occupation phases.

• Results indicate a complex interaction of socio-ecological processes through time.

Abstract

Caution Bay, on the South Coast of Papua New Guinea, offers a unique opportunity to assess the possible impacts of predation by pre-Lapita, Lapita, and post-Lapita peoples on local mollusc resources from at least 5000 years ago. Using biometric analysis of the bivalve Anadara antiquata and gastropod Conomurex luhuanus from the site of Tanamu 1, we examine trends in size and maturity variability through time. Results indicate a reduction in valve size of A. antiquata from c. 5000–2800 cal BP (the pre-Lapita period) to c. 2800–2750 cal BP (falling during the Lapita period), while C. luhuanus undergoes a change in maturity categories between the Lapita period and c. 700–100 cal BP (post-Lapita), with the latter containing lower proportions of both immature and mature individuals. Considering that these two mollusc taxa have the capacity to resist high predation pressures through their reproductive strategies and growth rates, in combination with low discard rates throughout Tanamu 1, it is unlikely that the observed trends are solely related to human predation. Rather, set against a context of significant environmental variability and shifting habitats through time, the pre-Lapita, Lapita, and post-Lapita phases represent significant socio-economic changes, whereby there is a shift from mobile foraging to an increasing reliance upon agriculture. It is therefore likely that there were a range of environmental and socio-economic factors influencing mollusc harvesting and the foraging economy more broadly through time.

Introduction

Mollusc shells are a common component of Lapita sites across the Western Pacific, representing an important subsistence resource and raw material for artefact manufacture. Research on Lapita invertebrate assemblages has focused on broader subsistence patterns, and manufacture and trade of shell valuables (e.g. Cochrane et al., 2011, Connaughton et al., 2010, Gosden et al., 1989, Kirch, 1988, Nunn et al., 2004, Szabó, 2009, Szabó, 2010, Szabó et al., 2020), but studies focused on the potential impact of human predation have been more limited (Spennemann, 1987, Spennemann, 1989).

Since islands of Remote Oceania were uninhabited prior to Lapita colonisation, and therefore lack comparative pre-Lapita shell assemblages, the effects of Lapita arrivals on local mollusc populations have been difficult to assess. In contrast, Caution Bay, situated on the south coast of mainland Papua New Guinea (PNG) west of Port Moresby, had resident human populations prior to the arrival of Lapita peoples (e.g. David et al., 2011, David et al., 2012, McNiven et al., 2011, McNiven et al., 2012a, McNiven et al., 2012b), offering an opportunity to investigate the possible impacts of Lapita people on local resources by comparing the mollusc shell assemblages from multiple periods. In 2009–2010, archaeological excavations at 122 sites were undertaken at Caution Bay. Nine of these sites are known to contain Lapita pottery, with other sites awaiting analysis (Thakar et al., 2017, McNiven et al., 2011). Pre-Lapita (i.e., for this region: pre-ceramic) assemblages at Caution Bay began to form by at least 5000 cal BP, with dense Lapita-period assemblages thought to represent village sites suddenly commencing at 2900 cal BP. This > 2100 year-long pre-Lapita archaeological record provides a baseline from which to evaluate whether the arrival of Lapita peoples and the establishment of apparently long-lived Lapita settlements impacted the structure of local mollusc resources via increasing levels of exploitation.

A growing number of international studies have implemented biometric analyses of molluscs in order to understand past people-environment interactions, trends in resource use, and the impacts of human predation and/or environmental change on local molluscan populations (e.g. Barker, 2004, Faulkner, 2009, Faulkner, 2010, Faulkner, 2013, Giovas et al., 2010, Jerardino, 1997, Jerardino and Navarro, 2008, Jerardino et al., 2008, Peacock and Mistak, 2008, Spennemann, 1987, Thangavelu et al., 2011, Whitaker, 2008, Ulm et al., 2019). In some cases, human impacts have been argued to be the major causal factor in mollusc population shifts (be they increasing or decreasing in size or abundance), while in others environmental/climatic change has been invoked, often in combination with human behavioural factors (e.g. Claassen, 1998, Faulkner, 2013, Giovas et al., 2013, Hausmann et al., 2020, Jerardino, 1997, Thangavelu et al., 2011). In several studies, predation pressures exerted by human foraging were seen to have led to a reduction in mollusc size, whereby the predator–prey relationship becomes unbalanced as the level of exploitation exceeds growth, reproduction, and recruitment rates (see Claassen, 1998, Faulkner, 2013, Thangavelu et al., 2011). This, however, should not always be the expected pattern, as in several cases shell size increased or remained stable in the face of intensive human exploitation (e.g. Giovas et al., 2010, 2013; Ulm et al., 2019).

Whole, intact shells have commonly been targeted for metric analyses (e.g. Antczak et al., 2008, Baez and Jackson, 2008, Bailey and Milner, 2008, Bailey et al., 2008, Barker, 2004, Faulkner, 2009, Giovas et al., 2010, Giovas et al., 2013; Ulm, 2006; Whitaker, 2008), although highly fragmented archaeological assemblages tend to decrease the measurable sample size, creating size bias due to differential preservation (Faulkner, 2010, Jerardino and Navarro, 2008, Faulkner, 2010, Jerardino and Navarro, 2008). Traditional biometric methods based on linear regression analyses have been used in archaeomalacological research to overcome issues of sample size and/or size bias linked to increasing fragmentation. Linear morphometric analyses have been successfully implemented to predict original shell size, with these data often fed into explorations of resource depression or resilience stimulated by human predation or environmental change (e.g. Ash et al., 2013, Campbell and Braje, 2015, Faulkner, 2010, Glassow et al., 2016, Singh and McKechnie, 2015, Thangavelu et al., 2011, Yamazaki and Oda, 2009). Importantly, a number of researchers have highlighted the need to consider the complex interaction of natural and cultural processes that can stimulate shifts in size and/or age structure in mollusc assemblages beyond human exploitation alone (e.g. Campbell, 2008, Faulkner, 2013, Flores Fernandez, 2017, Giovas et al., 2010, Giovas et al., 2013, Thakar et al., 2017). Given the potential effects of environmental change on molluscs, palaeoenvironmental records also need to be considered in the interpretation of archaeological assemblages.