The first quantitative assessment of radiocarbon chronologies for initial pottery in Island Southeast Asia supports multi-directional Neolithic dispersal

Abstract

Neolithization, or the Holocene demographic expansion of farming populations, accounts for significant changes in human and animal biology, artifacts, languages, and cultures across the earth. For Island Southeast Asia, the orthodox Out of Taiwan hypothesis proposes that Neolithic expansion originated from Taiwan with populations moving south into Island Southeast Asia, while the Western Route Migration hypothesis suggests the earliest farming populations entered from Mainland Southeast Asia in the west. These hypotheses are also linked to competing explanations of the Austronesian expansion, one of the most significant population dispersals in the ancient world that influenced human and environmental diversity from Madagascar to Easter Island and Hawai‘i to New Zealand. The fundamental archaeological test of the Out of Taiwan and Western Route Migration hypotheses is the geographic and chronological distribution of initial pottery assemblages, but these data have never been quantitatively analyzed. Using radiocarbon determinations from 20 archaeological sites, we present a Bayesian chronological analysis of initial pottery deposition in Island Southeast Asia and western Near Oceania. Both site-scale and island-scale Bayesian models were produced in Oxcal using radiocarbon determinations that are most confidently associated with selected target events. Our results indicate multi-directional Neolithic dispersal in Island Southeast Asia, with the earliest pottery contemporaneously deposited in western Borneo and the northern Philippines. This work supports emerging research that identifies separate processes of biological, linguistic, and material culture change in Island Southeast Asia.

Mainland-coastal interactions in East Borneo: Inter-site comparison and Bayesian chronological models of two Late Pleistocene–Holocene sequences (Liang Abu and Kimanis rock shelters)

“…266-267). A recent study, based on a Bayesian chronological model that argues that neither the "Out of Taiwan" nor the "Western Route Migration" hypotheses are particularly supported by the results, dates the introduction of pottery in North Borneo to 4560-2460 BP (Cochrane et al 2021). This interval, although large and imprecise, corresponds to our results from Liang Abu and Kimanis.…”

“…Bayesian modeling is becoming an increasingly used tool in the context of Island South East Asian and Pacific archaeology. For example, Athens 2019 andCochrane et al 2021 recently used the Oxcal software on the chronology of the Late Holocene expansion in Oceania and the Neolithic dispersal in Island Southeast Asia, respectively. Our analysis of East Bornean sites is a case-study to use a different software, called ChronoModel 2.0.18 (Lanos and Dufresne 2019), presenting distinctive features when compared with Ox-Cal and Bcal (Lanos and Philippe 2017, Lanos and Philippe 2018):…”

Genome Analysis Traces Regional Dispersal of Rice in Taiwan and Southeast Asia

“…Human genetic studies suggest, for example, that indigenous peoples related to those in contemporary mainland SE Asia were already present in island SE Asia before the arrival of Austronesian-speaking groups ( Lipson et al 2018 , 2014 ; McColl et al 2018 ; Larena et al 2021 ). Early movement of peoples in the region is also supported by a recent reappraisal of ceramic assemblages from 20 archaeological sites in island SE Asia and western Oceania, which likewise indicate a more complex, multidirectional set of Neolithic dispersals over the last ∼5,500 years ( Cochrane et al 2021 ).…”

Bayesian analyses of direct radiocarbon dates reveal geographic variations in the rate of rice farming dispersal in prehistoric Japan

“…The application of Bayesian phase models for regional studies was originally introduced to study Late Glacial human occupation in Northwest Europe ( 40 ) and has since been used to study a variety of similar phenomena [e.g., ( 41 ) for a recent application]. The approach is effectively an adaptation of models typically designed to investigate the stratigraphic chronology of individual sites and has the benefit of providing estimates of arrival dates while taking into account the uncertainties associated with sampling and measurement errors.…”