INTRODUCTION

Amongst many uncertainties, rock art researchers have certainty about place: that painted and engraved rock art remain in their original context. This is unusual in archaeology and apart from a handful of studies that use landscape to further our understanding of rock art (Manhire and colleagues 1983; Bradley and colleagues 1994; Bradley 1997; Hartley & Wolley Vasser 1998; Whitley 1998; Nash & Chippindale 2002; Parkington 2003; Chippindale & Nash 2004), this paper suggests that this quality of rock art may not have been exploited to its full potential. A geographic information system (GIS) which is developed to analyse spatial relationships both vertically and horizontally is a tool with which to approach the analysis of rock art in place and space. This paper considers how GIS might be useful in rock art studies through two examples, the large-scale analysis of horizontal spatial patterning and the small-scale analysis of vertical spatial patterning in rock art.

In order to understand the meaning of rock art, its authors must be identified. The value of approaching rock art interpretation from an ethnographic context has been demonstrated by Lewis-Williams's seminal work in understanding the San rock art of southern Africa (Lewis-Williams 1981, 2006). However, the link between rock art and a relevant ethnography is often impossible to make due to our inability to place rock art in time (this has hampered rock art interpretation in the Sahara, for example). The use of GIS to analyse rock art's place in space may contribute to identifying authorship, which in turn may allow it to be placed in time.

In the last decade there has been a rise in the number of theoretical papers by archaeologists using GIS, a tool to store, manipulate and display geo-referenced data. The success of GIS has been won outside of archaeology. Besides its more obvious uses in environmental sciences, it is commonly used to understand the pushes and pulls of market economies amongst societies that are themselves familiar, well known and understood. One of the criticisms of the use of GIS in archaeology is that a gulf remains between the theory (which seeks to humanise GIS) and the method (which requires quantifiable variables) (Llobera 1996).