INTRODUCTION

As Taçon (2001: 118) has remarked, there is general agreement that good rock art research begins with formal analysis based on reliable descriptions, typologies and chronologies, but there is disagreement about how to proceed further. Some prefer setting up testable hypotheses while others argue for starting with social theory and relevant ethnography. This paper proposes a third dimension. It describes a reflexive recording method that in certain circumstances can provide a visual record of rock art that is deliberately constructed to merge with the ethnographic record and emphasise landscape features and beliefs that were of significance to the original artists. I have called it ‘narrative photography’ but it could be developed as a branch of visual anthropology.

Rock art is a visual and tangible medium for the recording of intangible cultural beliefs and practices and it deserves to be recorded visually, yet the standard methods made it impossible to summarise the intangible aspects and the impact of scale in standard landscape and close-up photographs. Distribution maps and tracings of selected images, too, seem to dispel the power of the engravings in their setting.

Rock art databases, by their very nature, are cryptic records. Recording forms the basis of empirical methodology, prefers numbers to words and, until the advent of digital imagery, rarely merged visual information seamlessly with data on site location. In South Africa, good examples of rock art site records, such as those at the Rock Art Research Institute in Johannesburg, the National Museum in Bloemfontein and the KwaZulu-Natal Museum in Pietermaritzburg, have been invaluable in demonstrating the geographical distribution of sites and describing and quantifying the content and variability of the images. The enormous value of the tracings and redrawings of rock paintings in the Drakensberg done 50 years ago by Patricia Vinnicombe has only been truly appreciated with hindsight. A new generation of data management has been initiated with opportunities created by digital scanning and the Rock Art Research Institute at the University of the Witwatersrand is currently scanning photographs, slides, tracings and documents and linking them to a map-based recording system.