395 Pre-LGM Sahul and the Archaeology of Early Modern Humans Chapter 32 Pre-LGM Sahul (Pleistocene Australia-New Guinea) and the Archaeology of Early Modern Humans ments on the implications of these indings for future research on modern human origins. Humans in pre-LGM Sahul: an overview Geographic seting Pleistocene Sahul was a large continent (Fig. 32.1). When sea levels were at their lowest (20–22 kya bp), it covered nearly 11 million square kilometres, roughly the same area as sub-Saharan Africa or Eurasia west of the Ural Mountains. Relief was moderate: more than 90 per cent of its surface was less than 500 metres above maximum low sea level. Signiicant uplands included only the New Guinea Highlands on the north and the Great Dividing Ranges on the east (maximum elevations ~5200 m and ~2400 m above maximum low sea level, respectively). Major biomes in modern Australia-New Guinea include tropical forest, sub-tropical savanna, and low temperate desert. Small but important patches of temperate forest are found in the southeastern and southwestern corners of mainland Australia and in Tasmania. During the last glacial cycle (75–10 kya bp), cooler, drier climates and lower CO 2 levels generally reduced the distribution of tree cover, increased the size of the arid zone, and favoured the development of glacial and peri-glacial habitats in both the northern highlands and far southeast. Arrival date Sahul was irst occupied 42–45 kya bp and widely colonized by 35 kya bp (Fig. 32.2; O’Connell & Allen 2004). Though claims have sometimes been made for landfalls >50 kya (e.g. Roberts et al. 1990; 1994; Thorne et al. 1999), their validity is undercut by uncertainties about associations between dates and evidence of human activity, about the dates themselves, or both (O’Connell & Allen 2004). At present, no colonization dates >45 kya bp are reliably supported, nor indeed as James F. O’Connell & Jim Allen Discussions of the origins of modern humans and their distinctive behavioural capabilities oten refer to material traits thought to mark those capabilities archaeologically. Traits most commonly nominated include complex lithic and organic technologies, evidence of broad-based subsistence economies (pos- sibly involving habitat management), indicators of symbolism in the form of art, ornament or ‘style’, and burial of the dead. Recent treatments typically focus on the dates at which these proposed markers irst appear in the material record, the implications for the rate at which modern human capabilities might have developed, and (less consistently) the processes that might have been involved (e.g. Klein 2000; McBrearty & Brooks 2000; Mellars 2005). An important problem with this approach lies in the assumption of a 1:1 relationship between be- havioural capabilities and material evidence of their presence. The archaeological record of Pleistocene Australia-New Guinea, the continent sometimes called Sahul, illustrates the point. Although it is generally agreed that this region was colonized by fully modern humans about 42–45 kya bp , the record they let behind includes few of the conventionally identiied markers of modern behaviour until af- ter peak of the Last Glacial Maximum (LGM, here pegged at 20 kya bp), more than 20 millennia later. Most do not appear widely until the mid-Holocene. This observation raises important questions about the determinants of those markers and the signiicance of their absence. Here we provide an overview of the pre-LGM Sahul record, with special atention to evidence of modern human behavioural capabilities and to the absence or limited presence of many conventionally deined markers thereof. We then briely consider the factors that might account for the paterning observed, especially those involving climate change and human population density. We conclude with some com-