SciELO - Scientific Electronic Library Online

vol.104 issue11-12What became of the SAJSScience councils in South Africa author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand



Related links

  • On index processCited by Google
  • On index processSimilars in Google


South African Journal of Science

On-line version ISSN 1996-7489
Print version ISSN 0038-2353

S. Afr. j. sci. vol.104 n.11-12 Pretoria Nov./Dec. 2008




From generalities to specifics in San rock art



J.D. Lewis-Williams‡; D.G. Pearce

Rock Art Research Institute, University of the Witwatersrand, Private Bag 3, WITS 2050, South Africa



Today, virtually all researchers agree that San rock art is neither the product of idle hours nor merely the result of an innate human desire to make beautiful things. Research conducted over the last half-century has shown beyond reasonable doubt that the images were in some sense religious: overwhelmingly, they have to do with experiences, beliefs and practices that, though an integral part of daily life, dealt with the affairs of a notional spirit realm and the ways in which people interacted with that realm.

Although this understanding of the art came principally through a study of specific painted themes and the ways in which nineteenth-, twentieth- and twenty-first-century San ethnographic records illuminate them, many people still tend to think in terms of 'the art', a vague and highly generalized concept. They do not think of individual panels of images. An analogy: we are not satisfied with the general statement that much Renaissance art is Christian. Rather, we examine specific paintings to see how individual artists deployed Christian iconography, what they were thereby saying, in what contexts they placed their artistic creations, and numerous other lines of enquiry. In the long run, a comparable approach to San rock art will entail looking at a great many panels to identify commonalities. By publishing and discussing individual panels, we can tackle new questions.

This article therefore focuses on a specific panel of San images. (For similar studies, see, for example, refs 1–3.) In doing so, we narrow the field still further by selecting a single issue for detailed discussion: what the act of painting meant for the San. A subsidiary but nevertheless intimately related purpose is to emphasize the rock face with its facets, angles and inequalities so that we can see how the painters responded to the immutable features of their complex 'canvases'. We seek the conventions of San image-making.

Because San ethnography is silent on points like this, we have to start from what we already know about the images, examine individual panels like the one that is at the centre of our present discussion, and then move from there into the unknown.


Site ANG1

The site with which we deal is in the Maclear District of the Eastern Cape province (Fig. 1). Rock art on the farm was first noted in 1939 by Clarence van Riet Lowe in the course of compiling his list of rock art sites in South Africa. Since then, a number of researchers have visited the site, but it has not until now been published. ANG1 is a large rock shelter, approximately 50 m long, at the head of a small watercourse. The panel on which we focus (Fig. 2) is in a small recess to the south of the main painted shelter. Like most panels, it has distinct and less well-preserved images.





The rock face with its facets and inequalities is a significant component of the panel. We indicate the major features of the rock with a heavy dashed line in Fig. 2. Facets 1 and 2 are at a sharp, obtuse angle to each other. Facets 2 and 3 are on the same plane, but facet 3 is a raised surface of rock.



What has already been discovered about certain types of San images situates the ANG1 panel in the complex of San religion. An especially diagnostic feature is that two of the figures, one on either side of the lower part of the panel, are clapping (facets 1 and 2). Their fingers are individually drawn. The act of clapping points to the ritual known as the medicine, or trance, dance. During this dance the participants activate a supernatural potency, frequently associated with powerful animals, such as the eland, in order to heal, make rain and go on out-of-body journeys (ref. 4; ref. 5, pp. 63–90; ref. 6, pp. 81–100). It is the most inclusive and frequent San ritual. As is the case at many other sites, a full, 'realistic' dance is not depicted at ANG1. Nevertheless, the clapping figures [what have been called 'fragments of the dance' (ref. 6, pp. 99–100)] suggest that the overall significance of the panel has something to do with the medicine dance. This inference is borne out by other images.

All the human figures, except possibly the ill-preserved one on the lower part of facet 2, are kaross-clad. While it is true that the San wore both short and long karosses, those of the enveloping kind shown here are especially associated with figures that have zoomorphic features, the so-called therianthropes. In this panel, the heads of the figures that have not been damaged are clearly of antelope. The one left of centre (facet 1) has a well-preserved antelope head with horns, the forward curvature of which suggests mountain reedbuck (Redunca fulvorufula). The uppermost figure (facet 3), seen de face, has horns, though the rest of the head is not preserved. Three of the figures have strange pointed 'feet'. The 'hands' of the uppermost figure are similarly depicted.

The therianthropic figure with mountain reedbuck horns has six lines, or tassels, falling from its neck. They are a fairly common feature of the art, but they do not appear to depict anything realistic. Vinnicombe interpreted such lines as representing 'thinking strings'. She suggested that the San notion of 'thinking strings' situated in the throat possibly originated 'from an awareness of the jugular pulse during moments of stress, tension or intense concentration' (ref. 7, p. 260). She linked this idea to the 'string' that was said to vibrate inside a rain-maker (ref. 7, p. 344).

Four bags are depicted. Three of them (facets 1 and 2) have the leather tassels that are frequently still part of San animal- skin bags. In the paintings, these tassels recall the lines of 'thinking strings' and are comparable to those that fall from the seated figure on the lower-left of facet 1. Three of the bags are detached, while one appears to be hanging from the large therianthropic figure. The shape of the bags suggests that they are the quadrilateral, general carrying bags used by both men and women; men's hunting bags are conical. The small bag hanging from the figure with mountain reedbuck horns is especially interesting. Although the relative sizes of images are not always realistic in San rock art, this bag appears to be smaller than the detached ones. It may be one of the small bags in which San !gi:ten (shamans) store and carry small pieces of substances that they believe contain supernatural potency. But why did the San paint bags? An answer to that question can be found in San mythology. In one myth, the trickster-deity /Kaggen, himself the first shaman, gets into a bag to hide and to change himself into a flying creature (ref. 6, pp. 120, 126). Bags are thus obliquely associated with the dance, visits to the spirit realm and therianthropic transformations such as we see in this panel. In at least one other panel, two bags are shown transforming into eland (ref. 7, fig. 107; ref. 6, fig. 6.6).

An insufficiently studied feature of San rock paintings is the repetition of grouped images. Of significance at ANG1 is the combination of seated figures with adjacent bags (facet 1). Two such pairings are shown. Figure 3 shows a comparable pair from Main Caves, Giant's Castle, about 230 km to the northeast of ANG1. The Main Caves pair is clearly associated with a flying creature that has a head identical to that of the seated human figure. Known as flying or trance buck, flying creatures like this are implicated in the transformations that take place in the spirit realm.



On the left of facet 1, the upper seated figure with a bag has what is, as far as we know, a unique feature. A line that is too long and thin for an arm emerges from its shoulder and ends in a finely painted antelope head. The head is directed towards the mountain reedbuck therianthropic figure. There is thus some indication that seated figures with adjacent bags refer to the activation of animal potency and transformation.

Just right of the centre of facet 2 is what appears to be a pair of snakes or eels. Then, to the right of them, there is a long bi-coloured line and another pair of lines that is associated with a kaross-clad figure. The tapering form near the snakes (or eels) is reminiscent of much better-preserved eastern Free State figures that have stylized humanoid heads.2 Snakes enter the ground through holes, while eels live underwater. In San thought, these creatures were associated with crossing over into underground or subaquatic realms (ref. 6, pp. 51–55, 131–132, 144).

In the light of these observations on specific images, we can now consider what is the most interesting feature of the panel. It is the sinuous line that at first glance appears to be under the mountain reedbuck therianthrope. In fact, it was painted subsequently, in sections, to give the appearance of being under the therianthrope. This trompe l'oeil is an instance of what has been called 'factitious superpositioning'.8 The lowest section of the line leads from (or to) a prominent crack in the rock. It then loops 'behind' the therianthrope before it swings down towards the major angle in the rock face. Although it has no fringe of white dots, the line is reminiscent of the 'threads of light' that San !gi:ten say lead them to the spirit realm and that are frequently painted in the southern Drakensberg.9


The rock face and image-making

The technique of factitious superpositioning, though not commonly employed, has important implications for how the art was made and how we should view it today. The technique points to the three-dimensional nature of San rock art. An image (here the line) could be 'slipped under' an existing painting as if there were a space between the earlier image and the rock face. It was important for the painter that the line was perceived as being between the therianthrope and the rock – in an intermediary space. Factitious superpositioning depends on this concept: images 'float' on and, at the same time, engage with the rock face. They thereby suggest a dimension of depth stretching back from the daily world into a spirit realm that only !gi:ten can explore.

Some of the ANG1 images depict people and objects that can be seen in the material world (for example, seated people with their bags) and some that cannot (for instance, beings with antelope heads). It is clear that San painters were not merely making pictures. They were performing a ritual that blended the material world of daily life with the spirit realm. The rock face was not a meaningless support. Rather, it contextualized, and gave meaning to, images. It was a permeable 'veil' between this world and the spirit realm.10 We should therefore attend to the various ways in which painters exploited the cosmological interface that was their 'canvas'. Indeed, many instances have been noted in which San painters took advantage of steps, cracks, rough surfaces, and water salt trails. They used these natural features in the making of images that appear to penetrate the rock.

Believing that the spirit realm lay behind the rock face, the San painter(s) at ANG1 exploited the major, almost vertical, angle between facets that is shown in Fig. 1. In the centre of the panel, the remains of branching lines seem to emerge from the angle. It is possible that these lines may be part of a poorly preserved bag, but it is impossible to say with any certainty. The important point is that the lines were carefully positioned so that they extend right into the angle. Another image, one that may represent an antelope head, again as seen from the front, similarly emerges lower down.

Viewed overall, the panel seems to focus on the central angle in the rock face: most of the images face towards it, while lines and an antelope head emerge from it. We argue that the angle was not merely a compositional device. The painter (or painters) used it and its associated facets to produce a three-dimensional statement of supernatural relationships. It seems probable that the human figures on either side of it are, by means of their clapping and the powerful 'medicine songs' that the clapping implies, activating potency to transform what was an 'ordinary' inequality in the rock face into an entrance into the spirit realm.

We therefore conclude that the very act of painting was a means of penetrating the veil: people were 'painting their way' into another dimension. The whole complex process of making a painting was a religious ritual (ref. 6, pp. 100–106, 181). Images highlighted, or themselves opened up, entrances into the spirit world in ways that deserve close study. We may go so far as to say that the creation of paintings was in this way comparable to the medicine dance itself.

Geoffrey Blundell first suggested that we should work on the site. Sheila Bell-Cross, Helen Lechmere- Oertel and Adele Moore helped us in the field. We thank Winnie Mokokwe and Justine Olofsson who assisted in tracing the images. Wendy Voorvelt prepared the final copy. The Rock Art Research Institute is funded by the National Research Foundation and the University of the Witwatersrand.


1. Hollmann J.C. (2001). 'Big pictures': insights into southern African San rock paintings of ostriches. S. Afr. Archaeol. Bull. 56, 62–75.        [ Links ]

2. Lewis-Williams J.D. and Pearce D.G. (2004). Southern African San rock painting as social intervention: a study of rain-control images. Afr. Archaeol. Rev. 21, 199–228.        [ Links ]

3. Hollmann J.C. and Lewis-Williams J.D. (2006). Species and supernatural potency: an unusual rock painting from the Motheo District, Free State Province, South Africa. S. Afr. J. Sci. 102, 509–512.        [ Links ]

4. Katz R. (1982). Boiling Energy: Community healing among the Kalahari Kung. Harvard University Press, Cambridge, MA.        [ Links ]

5. Marshall L. 1999. Nyae Nyae !Kung: Beliefs and rites. Peabody Museum, Cambridge, MA.        [ Links ]

6. Lewis-Williams J.D. and Pearce D.G. (2004). San Spirituality: Roots, expressions, and social consequences. Double Storey, Cape Town; Alta Mira Press, Walnut Creek, CA.        [ Links ]

7. Vinnicombe P. (1976). People of the Eland: Rock paintings of the Drakensberg Bushmen as a reflection of their life and thought. University of Natal Press, Pietermaritzburg.        [ Links ]

8. Lewis-Williams J.D. (1987). A dream of eland: an unexplored component of San shamanism and rock art. World Archaeology 19, 165–177.        [ Links ]

9. Lewis-Williams J.D., Blundell G., Challis W. and Hampson J. (2000). Threads of light: re-examining a motif in southern African San rock art. S. Afr. Archaeol. Bull. 55, 123–136.        [ Links ]

10. Lewis-Williams J.D. and Dowson T.A. (1990). Through the veil: San rock paintings and the rock face. S. Afr. Archaeol. Bull. 45, 5–16.        [ Links ]



‡ Author for correspondence. E-mail:

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License