Getting from Sunda to Sahul

Abstract

Introduction The level of intentionality in the behaviour of early modern humans is a continuing debate that we have had with Atholl over the years. For our part, we argue that the cognitive abilities of early modern humans, reflected in the patterns of the late Pleistocene archaeological records, are qualitatively different from their predecessors and not vastly different from our own. This being the case, we feel able to assume purpose in past human behaviour that Atholl does not. Presented with evidence that is often meagre, indirect or ambiguous, Atholl counters with explanations that require little or no intent on the part of these humans. Nowhere has our debate crystallised as clearly as in the case of early watercraft and their use in colonising Sahul, a milestone modern-human achievement for which no direct evidence is or is likely to become available. We all accept without question that it happened, that watercraft were involved and that this water crossing was only achieved by modern humans. Despite some continuing debate, it is increasingly likely that first colonisation occurred about 45,000 years ago. Beyond this, we do not know where people stepped off the Asian continent, what routes to Sahul may have been taken, what length of time elapsed between humans leaving Sunda and arriving in Sahul (Figure 1), what viable population sizes might have been needed to form successful colonisations, where first Sahul landfall was, whether the first successful entry was preceded by unsuccessful ones, whether there were multiple successful entries at different locations, and so on. But one can engage in constructive, theoretically driven, potentially testable speculation on all of these points. This paper was first presented at an ANU seminar in 2005 to draw attention to these questions, largely untouched since Birdsell's (1977) seminal paper. It provoked useful discussion between Atholl and us and was a focus for subsequent work (Bulbeck 2007; O'Connell et al. 2007; O'Connor 2007), and we offer it here to acknowledge our intellectual debt to Atholl Anderson. Since direct evidence to answer questions concerning the initial colonisation of Sahul is 32 Islands of Inquiry / Jim Allen and James F. O'Connell terra australis 29 unlikely to be forthcoming in the archaeological record, alternative approaches are limited. For us, inductive approaches are currently unsatisfactory, relying as they do on arguable interpretations of sketchy data that are frequently date-driven. For example, O'Connor (2007) predicates her entire argument on a human arrival date in Sahul of 60,000 BP. This date, mooted 18 years ago (Roberts et al. 1990), has received no compelling support in the intervening years and today only a tiny minority defend it. The incorrect argument of Chappell et al. (1996) and Fifield et al. (2001) that our appeals to a younger chronology were the result of blind adherence to radiocarbon has been continuously weakened by the increasing use of luminescence and improved radiocarbon pre-treatment (ABOX-SC). To date, neither technique has produced an archaeologically associated date approaching 50,000 BP, outside the two original Northern Territory claims of Nauwalabila and Malakunanja. Regardless of what calibration is applied to available radiocarbon dates, a current 'best-estimate' date for human arrival is c. 45,000 years ago, or a little earlier (Allen and O'Connell 2003; O'Connell and Allen 2004; O'Connell et al. 2007). A better alternative is to develop models for this colonisation that will throw up testable hypotheses to approach these questions. This paper offers one such model that assumes that the first successful human colonisation of Sahul was the consequence of many small but deliberate decisions that involved conscious and continuing risk assessment of behaviours intended to maximise reproductive fitness. People crossed from Sunda to Sahul as a consequence of these behaviours, rather than with conscious intent, like the chicken, to get to the other side. Even so, the colonisation of Sahul was not accidental.