One of the most exciting research frontiers in Australian prehistory over the past five or so years has been the emerging possibility of applying advanced absolute dating methods--in particular, accelerator mass spectrometry (AMS) radiocarbon dating--to measuring the antiquity of Aboriginal rock images (paintings and engravings). It is a field of international interest, with the leading research in recent years having been carried out in Australia and France. In both areas, credible direct dates from the art itself have been obtained extending back to some 25-30,000 years ago. These research results can capture the public imagination, as evidenced by newspaper reports (Dayton 1993; Hammond 1993; Leech 1998).
The grand issue under investigation is whether or not the human capacity for art developed independently at about the same time on both the northwest and southeast edges of the Old World. Rock images of considerably lesser antiquity have also been successfully dated in the southwest of the USA and in South Africa. The images are physical manifestations of cognitive processes representing non-verbal communications about myths, ritual activities, symbolic expressions, and so on. Expressed on rocks and on the walls of caves, they are a unique aspect of the archaeological record, in that they engage directly with the ways in which people once viewed themselves and their world.
Within the European Palaeolithic record, scholars were able, even by the end of the last century, to give a credible assessment of the age of the images, because examples had been carved on ivory and bone which were found in stratified cave deposits. Stylistic comparisons could be made between these three-dimensional objects and the motifs on the cave walls. Despite this, it is of historical interest that a deep antiquity for Franco-Cantabrian cave art was only generally accepted when Spencer and Gillen's accounts were published of the brilliant ritual images of the Aborigines of Central Australia (Cartailhac 1902). In the Franco-Cantabrian region, most rock image dating has been done on charcoal taken directly from motifs on the walls of deep limestone caves. The technical virtuosity of this work is extraordinary, because of the ethical and aesthetic need to take absolutely the minimum quantity of material from a painting. The AMS targets can be as small as 50-millionths of a gram, because the equipment is capable of measuring one [sup.14]C atom in a million trillion ([10.sup.-15]) [sup.12]C atoms.(1) Despite this, the main thrust of the results of AMS technology on Upper Palaeolithic cave paintings has been to confirm and in places to refine the established chronology (J Clottes 1995, pers. comm.).(2)
The situation in Australia is considerably more complex than in France, and it is probable that we know as little about the antiquity and cultural sequence of Australian rock imagery as was known in France by the beginning of this century. There are many reasons for this (Rosenfeld 1993; Watchman 1993a). Most rock images in Australia are in open galleries; usually, the bed rock is sandstone and quartzite and not underground in limestone caves. The climatic and environmental conditions are also different and, indeed, differ markedly across the continent itself. Despite these variabilities, we are making rapid progress in finding new ways to date Australian rock images.
In the early 1980s, it began to be realised, when paintings in the field were examined microscopically, that they were often covered with a series of mineral crusts and skins (or accretions, Watchman 1987). The mineralogy of these rock surface coatings has proved to be extremely complex, reflecting changes in past environmental and climatic regimes. The important aspect, from the point of view of potential dating of the imagery they cover, was that, in some situations, the mineralogical compositions of the skins themselves offered opportunities for absolute dating analyses. …