From 1979 to 1983 the American Research Center in Egypt carried out an architectural, archaeological and geo-archaeological study of the Giza Sphinx. Photogrammetry and conventional surveying techniques were used to prepare detailed plans and front and side elevations of the monument. These have made it possible to construct a computer model of the current condition of the Sphinx, and its hypothesized condition in ancient times, both as originally carved in the 4th Dynasty, and as remodelled and renovated during the 18th Dynasty. Careful analysis of surviving detached fragments of the Sphinx allowed details of beard and uraeus to be included in the reconstruction. This process of creating a computer model of the Sphinx is akin to sculpting the statue again in computer memory.

This article addresses the question of human symbolic behaviour in the Lower and Middle Palaeolithic. Bednarik argues that the case against such early symbolic behaviour usually relies on untestable propositions about the stages of human cognitive development, and that too little attention has been paid to the full range of evidence for Lower and Middle Palaeolithic symbolism. He urges that the same criteria should be applied in assessing this evidence as in assessing the more widely accepted evidence for Upper Palaeolithic symbolism. In their following reply, Chase & Dibble observe that many quite reasonable hypotheses in the historical behavioural sciences cannot be refuted absolutely, and that where competing hypotheses are presented it may be difficult to decide which is most probable. Finally Davidson, also replying to Bednarik's criticisms, concludes that their different views of the evidence for early hominid symbolic behaviour arise from different objectives and conventions of understanding

In 1987, we published a review of archaeological data that had been interpreted by others as evidence for the making and use of symbols before the Upper Palaeolithic (Chase & Dibble 1987). It has become clear to us since then that our study raises a number of issues we did not address directly. The foregoing critique by Bednarik touches on a few of them, but there are more fundamental issues he does not confront. Also, some of his criticisms seem to stem primarily from a misunderstanding or misreading of our article. What we will do here, then, is to respond briefly to his criticisms (we will not discuss his comments on the work of other authors), and then go on to address some of the important issues that he has not considered.

The editor invited a comment on this ‘attack’ on Davidson and Noble (1989), but asked me ‘not merely toattack points of detail’. Any careful reading of Bednarik's article and comparison with the works it attacks would show that in almost every detail it lays itself open to such attack. Nevertheless I am grateful to the editor for the opportunity to set out some of the argument that Noble and I have been trying to develop, and that so clearly Bednarik has not understood. There are good reasons why it has not been understood: it is clear from every phrase of his article that our objectives are utterly different.

The Minoan mountain peak sanctuary of Atsipadhes Korakias (Rethymnon, Crete) was excavated by the author in 1989. This article presents the first published account of that excavation and addresses its contribution to broader issues, methodological and interpretative, of Cretan Bronze Age religion. The spatial distribution of over 7500 finds was recorded, enabling the detailed reconstruction of the use and function of the sanctuary. This reconstruction is compared with other approaches to the identification of prehistoric cult places in Greece. Of the other excavated peak sanctuaries, the best known are élite sanctuaries associated with palatial centres; Atsipadhes Korakias is a poor, rural sanctuary, and offers a counterbalanced perspective on Minoan peak sanctuaries.

Ethnographic data on the production of rock art in far western North America - the historic hunter-gatherer cultures of California and the Great Basin - are reviewed and analyzed to identify widespread patterns in the origin and, in certain cases, symbolism of the late prehistoric/historical parietal art of this region. These data, collected in the first few decades of this century by a variety of ethnographers, suggest only two origins for the art: production by shamans; and production by initiates in ritual cults. In both instances, the artists were apparently depicting the culturally-conditioned visions or hallucinations they experienced during altered states of consciousness. The symbolism of two sites, Tulare-19 and Ventura-195, is considered in more detail to demonstrate how beliefs about the supernatural world, and the shaman's relationship to this realm, were graphically portrayed.