By far the most common interpretation of meaning in rock art is the iconographic interpretation of motifs by the observer. We are told what the beholder of the art thinks it depicts. In many cases, the motif has such outstanding diagnostic features that these identifications do sound convincing, but in many other cases the picture is not at all clear-cut. Moreover, many researchers define various aspects of the motif in an entirely subjective fashion: they tell us that the subject is running, falling, swimming, pregnant, praying, dead or whatever else they happen to perceive in the art. Provided that all these interpretations are offered for the purpose of creating a folklore about the art, a new mythology, one could not possibly object to them. Indeed, such interpretations may even be useful to the scientist, because from them he can learn about the perception of the person interpreting the art. If the person speaks his language and is capable of analysing his own responses to the art (to tell, for instance, very precisely why he thinks an animal figure is of a dying individual), an analysable example of an ethnographic reaction to an alien art becomes available. The person whose perception is so analysed may be a ‘tribal’ person from Orissa or an ethnographer — the analysis of his perceptive processes is in both cases ethnographic work.Many theories about rock art are based entirely on the identification of motifs in the art. These identifications are essentially subjective and they cannot be confirmed or refuted by any form of scientifically acceptable procedure (Bednarik 1991a, 1991b). Such identifications are non-refutable hypotheses based solely on the self-proclaimed abilities of ‘interpreters’. In the only case the identifications of a researcher were able to be checked against those of the artist himself, the ‘scientific’ interpreter of the art reported that he had failed abysmally (Macintosh 1977). The iconographic decisions of observers who have no access to the iconographic conventions of the pre-Historic culture in question, who are simply making up interpretations in the knowledge that they cannot be refuted, are not relevant to rock art science. Perhaps their conviction that their own cognitive processes of identifying and classifying iconicity are identical to those of the ancient artists is valid, but how would one demonstrate this?Palaeoarts can be studied scientifically, but this is not served by striving to determine a quality that cannot be determined, such as meaning. We can either find out what else science can do, and can do properly — or we can abandon the rigour of science and take a shortcut to ‘meaning’, creating and projecting our own favoured interpretation of the art. It may involve shamans, metaphysics, arithmetic, lost tribes, trance visions, interstellar space travellers, blood-curdling rituals, virgins, head-hunting, aspects of religion, warfare, cannibalism, and a whole gamut of somewhat less entertaining interpretations. None of them can provide any insights about the rock art so interpreted, but they are always a revealing tool for exploring the psychology of the interpreter. Precisely the same applies to the iconographic interpretation of motifs: its only scientific function is to study the perception, cognition and cultural conditioning of the interpreter.Provided we indulge in interpretation without physical interference with the art it is a perfectly harmless pastime, and there can be no objection to it. Rock art interpretation is highly stimulating, it enriches our experience and it can enrich our own art, culture and existence. It can help us create more myths about the past, we can invent our own favoured story of what happened in that past (in the fashion of Frankel’s [1993] sculptor). Provided that in the process we do not belittle any other culture or inflict any damage on the rock art, there can be no objection to such quests — as long as we make no attempt of presenting them as science.

R. G. Bednarik, 2001