Early human occupation of northern Australia: archaeology and thermoluminescence dating of Jinmium rock-shelter, Northern Territory

Abstract

The nature and date of the human colonization of Australia remains a key issue in prehistory at the world scale, for a sufficiently early presence there indicates either Homo sapiens sapiens arriving precociously in a place remote from a supposed African origin, or a greater competence in sea-crossing than has been expected of archaic humans. Stratigraphic integrity, the new science of luminescent dating and the recognition of worked stone and of rock-engraving are immediate issues in this report from far northwestern Australia.

Humanity from African Naissance to Coming Millennia

“…A Holocene antiquity fits also with geomorphological indications of a relatively young age, and with the 14 C chronology obtained using elemental carbon extracted from large (>125 µm) pieces of charcoal (see below). It is also consistent with two reinterpretations of the published TL data for Jinmium (Roberts, 1997;Spooner, 1998), both of which concluded that the reported TL ages (Fullagar et al, 1996) were much too old.…”

“…A subsequent application to the dating of rock paintings in the northern Kimberley region of Western Australia foreshadowed the deployment of single-grain methods at the habitation and rock art site of Jinmium. The wall of this shelter is covered in pecked cupules, and the original excavators (Fullagar et al, 1996) reported that a fallen fragment of engraved sandstone was buried in sediments dated to between 50 000 ± 6100 and 75 300 ± 7000 years using TL. Fullagar et al (1996) also recovered stone artefacts from levels dated by TL to between 116 000 ± 12 000 and 176 000 ± 16 000 years, which would have made it the oldest known human occupation site on the continent, two to three times as ancient as Malakunanja II (Roberts et al, 1990) or Nauwalabila I (Roberts et al, 1994a).…”

“…The reason for this difference can be explained by the observation (Wintle, 1997;Aitken, 1998) that it takes only a few seconds or minutes to reset the OSL signal, whereas it takes many hours to reset the TL signal used by Fullagar et al (1996). Hence, while the TL signal had been reduced only slightly, the OSL signal of many grains (but not those derived from rubble) had been mostly erased before burial.…”

“…Underneath this dated unit was a fine powdery silt that contained flaked stone artefacts (S. O'Connor, personal communication), and efforts are continuing to obtain numerical-age estimates for this unit. To the east of Carpenter's Gap lies the Jinmium rock shelter, where human habitation stretching back more than 100 000 years was proposed by Fullagar et al (1996) on the basis of artefact-bearing sediments which were dated by thermoluminescence (TL). TL dating had previously been used to derive ages for archaeological sediments at Malakunanja II (Roberts et al, 1990), a rock shelter located in western Arnhem Land in the far north of the Northern Territory, a likely entry point for the first human arrivals into Australia.…”

Dating the First Australians

“…Jinmium. The sandstone rock shelter site of Jinmium in northwestern Australia came to notoriety with TL dates suggesting pecked cupules on buried rocks were older than 58 ka, ochre at 75-116 ka, with stone artifacts and first occupation at 116-176 ka (Fullagar et al 1996). These results were widely disbelieved, with criticism on dating and sedimentological grounds (e.g.…”