Excavated Remains
During the last decade in particular a number of studies have demonstrated that contact between Bushmen and Bantu-speaking farmers was much more complex than previously thought.24 In the Thukela Basin, for example, excavations have provided empirical evidence for interaction between Stone Age and Iron Age groups.25 Pieces of talc schist and soapstone, often used to make bowls and other vessels, have been found at hunter-gatherer sites in deposits post-dating 2000 BP; this material has not been recovered from deposits prior to this date, that is, before contact with the Iron Age farmers.26 These sites are in the upper reaches of the Thukela River and some distance from the sources of these rock types. Similar stone was recovered from sites occupied by farming communities, one of which also produced an assemblage of Stone Age or hunter-gatherer tools that archaeologist Tim Maggs believes are contemporary with the farming occupation of the site.27
Maggs and Aaron Mazel argue that the patterning of material culture in the Natal area shows that interaction between Bushmen and farmers, which must have started as soon as the farmers came into contact with Bushmen, was initially extensive and amicable. Mazel has demonstrated that these relationships, certainly at the outset, were on a relatively equal footing, unlike the kind of clientship the Bushmen entered into with the farmers as reported in the nineteenth-century records.28 The possibility that clientship became more substantial as a result of decimation of the Bushman people by European colonists should be investigated; I touch on it again below. Further, there is every indication that new complex social relations accompanied and formed the basis of the new economic relations.29
Bushman reactions to Bantu-speaking farmers and European colonists can no longer be seen in terms of weaker people meekly submitting to more sophisticated people with more advanced modes of subsistence. Maggs and Gavin Whitelaw believe that further studies in this field need to deal with understanding the more complex 'intermeshing processes' between the various economies.30 The key to these processes is, I argue, to be found in social and cognitive enquiries.31 Turning to cognitive issues is a recent and important trend in archaeology.32 In contrast, by relying heavily on early historical documents and emphasising economic concerns at the expense of cognitive issues, Kalahari revisionists have been able to argue that the Bushmen were merely an oppressed class within the overall farming society rather than a cultural group with its own values, religion and sense of identity.33
An examination of Bushman ritual and religion, however, shows that relations between the Bushmen and the farmers or pastoralists clearly did develop in some respects. Nevertheless, despite changes that took place within their society, the Bushman people did not lose their identity as a separate group. A social enquiry into the nature of change both between Bushmen and other groups and within Bushman society itself leads to the second part of the twofold 'Bushman record': rock art.
Rock Art
For decades Eurocentric attitudes towards Bushman art prevailed. It was not until researchers began to use authentic Bushman beliefs34 that a much deeper appreciation of the art began to emerge. The current trend began by recognising that much of the art in southern Africa reflected shamans' experiences during the trance ritual and beliefs about it.35
This research led to the investigation of the role of the shaman and rock art in Bushman society, before and after contact with farmers.36 David Lewis-Williams, examining what was then thought to be 'traditional Bushman art', proposed that it provided a permanent backdrop to the daily social relations that shamans maintained during their trance rituals. Colin Campbell, on the other hand, examined 'contact art', art that had elements such as cattle, shields, horses and wagons, and so clearly resulted from Bushman interaction with other cultural groups. Campbell argues that these new images were incorporated into the art to provide an appropriate backdrop to a new set of social relations that were, like the 'traditional' relations, still maintained by shamans. Both Lewis-Williams's and Campbell's analyses were, however, carried out with a misconception of what is and what is not 'contact art'; the division cannot be made on subject matter alone. Hence a new examination is required.
To begin with, I argue that the rock art is not just a reflection or depiction of beliefs and experiences associated with the trance ritual. The art is a material item that was always, both before and during the so-called contact period, actively implicated in the reproduction and transformation of social relations, specifically those relations involving shamans. Because of the detailed and abundant trance imagery in the paintings, it is highly likely that shamans were the principal if not only producers of the art. Most panels, if not all, have some depictions that can be unequivocally associated with trance belief and experience. The historical documents and Bushman art are thus strikingly similar. Both create, transform and reinforce dominant ideologies: the historical documents continue to negotiate white settler ideologies, whereas Bushman rock art negotiated shaman ideologies.
To be able to discern exactly how the art played this role we need to understand, first, how the depictions were produced. The cognitive structure of the art was socially produced in that meanings attached to specific combinations of formal attributes come out of day-to-day social practice. The art was thus intimately implicated in developing social relations and the reproduction and transformation of social forms. Generally, and very briefly, I demonstrate how these processes came together and how the art negotiated Bushman ideology, particularly shaman ideology.
The proximity of Bantu-speaking farmers generated a new set of social relations in which the Bushmen in general and the shamans in particular were implicated. Farmers recognised the Bushmen as the original inhabitants and custodians of the land, and it was natural for the farmers to turn to them. This relationship, posited essentially on land ownership, came to centre on rain-making. The farmers, more than the Bushmen themselves, were dependent on rain; even minor droughts and, perhaps more important, delayed rains, affected their crops and herds far more than they did the Bushmen's antelope and plant foods. The mediator thus turned out to be the shaman. Part of the shamans' symbolic work was rain-making.37 Even though the farmers had occupied the land, they were unable to farm successfully without rain. The farmers requested Bushman rain-makers to perform rituals and gave them cattle in return. It was the shaman who had (ideological) control over the farmers' economy.
Because the shamans were paid for their rain-making services with cattle, presumably among other things, they acquired a new status as procurers of meat, and no doubt achieved power through a newly developed right to distribute the meat. With the depletion of antelope herds by white hunters and the extermination of the Bushmen by white commandos, Bushman shamans were forced to become more dependent on the farmers: the shamans had to tighten their grip on the farmers. This resulted in Bushman shaman families going to live with black farmers.38 It could be that these Bushmen were acknowledging the farmers' control of the land, but, at the same time, trying to retain some power and status.
Within Bushman society, diminishing traditional resources and, at the same time, new sources of wealth resulting from new social and economic relations with the farmers engendered competition between shamans. People looked to them as their go-betweens with the farmers and, increasingly, as the most reliable procurers of food. Shamans thus began to compete with one another and with important non-shaman members of the group for positions of influence. These power struggles, as well as the stresses of cultural contact between farmers and hunter-gatherers, were manifested in the art. The art, produced by shamans, became active and instrumental in forging new social relations that developed out of these power struggles.
People negotiate personal and social identities by means of stylistic statements.39 Social identities become important during situations of intergroup competition and the need for co-operation to attain social, political or economic goals. Competition among individuals and an increase in options for individual enterprise result in strong personal identities. Contact between