Rock Art Images on shelter walls have obviously not moved since they were painted: they are where they were intended to be and we may ask whether their place is as significant as their time and their form? This question is as important at the small scale (where is the image in the site?) as at a larger one (where is the site in the landscape?). If images have to be where they are, we may ask why? What is significant about that place? Does the image sometimes, or in some sense, reflect an occasion, an event or a circumstance with relevance to that place? What, we can ask, is the connection between image, place and event? Whilst difficult, attempts to understand these relationships are facilitated by improved levels of detail from enhanced imagery, digital mapping of sites along with their surroundings, and the development of virtual reality models. We now have the capacity to reconstruct landscapes as they were made by people fixing memories of place and occasion.
John Parkington, Stephen Wessels, Joe Alfers, Royden Yates, Andrew Paterson
“Places not only are, they happen” Casey 1996: 13.
Here we pose the question what is the significance of the placement of a rock art image, using as an initial example a particular painting along the middle reaches of the Brandewyn drainage in the Agter Pakhuis region? We recognize a long local and international interest in the issue of phenomenological analysis of rock art [see inter alia the work of Tim Ingold, Chris Tilley, Barbara Bender, Richard Bradley, Julian Thomas and Andrew Jones in Europe, Daniel Arsenault in Canada, Paul Tacon in Australia and Janette Deacon, David Morris, Anne Solomon, Pippa Skotnes and Sven Ouzman in South Africa].
As an initial aside, a conceptual framework for our work, we describe briefly developments in the late 1960s and early 1970s among wildlife biologists of East Africa after the introduction of new technologies such as aerial photography and sophisticated statistical modelling to manage elephant numbers. Culling seemed a cruel and impersonal way to manage such iconic animals. Traditionally elephants lived in ‘populations’ and were culled in response to numerical models of carrying capacity and growth in numbers. Iain Douglas-Hamilton introduced novel practices of long term, intimate monitoring of elephant ‘communities’ that transformed elephants from mere cyphers ‘standing for the notion elephant’ into individuals living as families, extended kin groups and inter-related and interacting social units. Generalisations were replaced by particular observations on specific life histories played out by named individuals (‘Emma’, ‘Elsa’, ‘Torn Ear’) and detailed interactions became the key to increased understanding of elephant ecology. ‘Elephanthropology’ legitimized anthropomorphism (Daston and Mitman 2005) and introduced multi-generational narratives (Moss, Croze and Lee 2011).
Use your mouse or finger to pan around the image. Interactive aerial 360 panorama showing the topographic landscape around the Brandewyn river, where the 'initiation scene' is painted. 360 degree panorama by Stephen Wessels
Perhaps ambitiously, but probably reflecting widespread stirrings within rock art thinking in southern Africa (Pearce 2021, in press), we are suggesting something similar to this transformation of elephant biology for rock art studies, focusing initially on those of the Western Cape, and specifically the Agter Pakhuis (Parkington and Paterson, 2021 in press). A painting of a line of human figures is not simply an exemplar of a ‘species’ of painting (lines of men, lines of women, single clapping woman, front-facing seated figure, elephant and calf, handprint, colonial figure with rifle) but an individual construct made in a moment (hour, day, week), placed deliberately, referring to some intersection of time, detail and location and conferring a permanence on this for contemporary and future viewers. Such acts were instrumental in constructing social landscapes through memories made material. Locations were surely not inconsequential, merely sets of coordinates for locating an image or convenient canvases? As we have said, paintings were placed, places painted and landscapes relatively permanently framed as memory sets. Such an entanglement of personal experiences, feelings about places and belonging would seem to imply specific relationships between image, place and occasion, rather than ‘symbolic’ generalities?
We begin with an image that has been photographed many times, traced once at least by Royden Yates (using a photographic slide projected onto a white wall) and referenced often, usually as an ‘initiation scene’. Its attraction has been that it is noticeably framed in the rock outcrop, does not appear to contain over-paintings and is usually taken to reflect a unified conceptual project. Simply put, it consists of two lines of human figures, both facing right but differently organised, including many details that relate to dress, equipment and posture and incorporating a conventional eland torso, also facing right, of a shape widely seen in Western Cape rock paintings. There are also paint patches that have conventionally been termed ‘palettes’ in colours that match those of the human and animal images. We argue here that this is a depiction of a specific event or occasion, held at this particular place and memorialized, perhaps as a part of the ceremony, into a permanent marker of the (spi)ritual relationship between people, identity and belonging.
Use your mouse or finger to move around and explore the 3D model. Interactive 3D model of the Initiation scene panel and surrounding context. 3D model by Stephen Wessels
Time: we have no chronological purchase on the age of the painting, except to note that there are no details that would be out of place in a hunter gatherer social and technological context, no evidence of colonial, or even pastoralist, objects or references and every indication in the residual nature of the paint traces on the rock surface that the painting is of ‘some antiquity’, feasibly a time of hunter gatherer occupation. Even harder to assess is the contemporaneity of all image components and the amount of time taken to produce the painting. As we argue below, the coherence and compositional unity persuade us that it reflects a single conceptual and creative ‘moment’ and one not of long duration. Although this is a ‘guesstimate’ we suggest the painting could well have been made in a few days.
Click the arrows to change the image. Initiation scene panel, with three D-Stretch enhancements. Tracing by Royden Yates. Photograph and enhancement by Stephen Wessels.
Form: because we have recently presented the argument in detail (Parkington and Paterson, in press, Azania 2021), we will assume that readers accept the broad outlines of our description of the painted forms here. Specifically, that the patches are ‘palettes’, the red shape in the lower centre is an eland torso, the lines consist of males wherever sex is discernible with no indisputable females, the upper line has cloaked figures in the rear and naked (uncloaked) figures to the front, the lower line features only naked figures with many details not found in the upper line, that hunting (cloaks, bows, bags) characterizes the upper line and dancing (sticks, headbands) the lower line, figures toward the front of both lines tend to bend or crouch and reach forward to touch the figure in front, and that the objects to the top right are best interpreted as hunting bags. Although Royden was not able to fully resolve the shapes at the time of tracing, enhancement of images makes it clear that two figures in the lower line have one or more elephants ‘on their heads’ and that there are as yet unidentified shapes (equipment or clothing?) on the bodies of the crouching figures. We also emphasise that, unlike in some Western Cape imagery, there is little stereotyping of figures, even those adjacent in lines, which we take to imply specificity or individuality, even a personalizing of representations. These do not ‘stand for’ men, they are individuals participating in two linear occasions, one likely a dance. Overall, we argue it is a complex image full of specificity, detailed differentiation and strong organization.
Enhancements of details in lines of human figures. Tracing by Royden Yates. Photograph enhancement by Joe Alfers.
Place: although this is obscured somewhat by recent building constructions and biological introductions, the panel of rock selected for the placement of this image is prominently visible and presents almost ‘stagelike’ in its verticality, directionality and physical detail. Paintings are framed by natural markings generated by plant growth and water percolation. There are, indeed, other paintings fairly close by, within a few tens of metres, but none of these is nearly as prominent in the local topography. This context, when scoured of its historic intrusions, is a broad opening up of the floodplain of the Brandewyn river at a confluence with a smaller tributary from the east and with a reliable water flow through flat rocky outcrops, pooling in places. From the river the painted panel, though not the details of imagery, is easily visible and faces almost due east into the rising sun and moon.
Social context: detailed ethnographic accounts of Agter Pakhuis hunter gatherers are not available because destructive encounters with colonizing and dispossessing settlers took place prior to the potential for such more professional interactions. We do know, of course, that local hunters and gatherers were San groups who enter the written records as Soaqua and who were broadly related (biologically, culturally, linguistically) to San (sometimes Bushman) groups from further north where they survived long enough to be encountered and described by more sensitive, later professional anthropological, outsiders. As is clear from ethnographic syntheses (Schapera 1930, Barnard 1992) and more specific field accounts such as those of Dorothea Bleek among the Naron (1928), recently reviewed by Viestad (2018), useful descriptions of hunter gatherer practices in Botswana, Namibia and Angola are available to help contextualise the imagery from the Western Cape (Parkington and Paterson 2021, in press, Azania). We access these descriptions not as plug-in solutions to our interest in interpretation, but as guiding ‘hypotheses’ for our research objectives (Hitchcock 2019).
From these accounts, it is clear that several distinct San communities (!Kung, Naron, G/wi, !Xo), prior to large-scale disruption, practiced lengthy, group male initiation events that involved older men taking groups of younger initiates into ‘the bush’ for some days or weeks characterized by isolation, transformational experiences, dancing, tattooing and life history instruction. Although these events were by no means identical across linguistic or cultural boundaries, the commonalities are suggestive of a formerly widespread set of practices. Initiates were chosen from a number of neighbouring bands, and locations were agreed on in advance, presumably based on ecological and social criteria. Our position is that the similarities of the painted details to the ethnographically described circumstances of what Richard Lee has called (1979: 365) ‘group male initiation camps’ are too close to ignore. Our hypothesis is that the Brandewyn imagery reflects such an event or occasion, not the notion in general, but one camp in particular.
If this is a hypothesis, can it be tested? Does it apply to all or just some paintings? We suggest that the persuasiveness of the idea depends on the documentation of detail and the sensitive mapping of form and place (time would help) at multiple scales, but it is clear that experimentation in the strict sense is not possible. Paintings are certainly where they are and nowhere else, and no two paintings, nor any two places are identical, but the significance of these is debatable. Archaeologists differ on the value of ethnographic support (see Hitchcock 2019, for example), but we doubt stone age archaeologists of Southern Africa will ignore a voluminous literature on technological, social and ecological practices still, or until recently, observable in context. Lorraine Daston (2005:51), referring to the comparable attraction of anthropomorphism among ethologists and comparative psychologists, writes: “they practiced anthropomorphism with a kind of desperation, aware of its risks and limits yet convinced that there was no other way to play the game and that the game was worth the candle”. Surely understanding rock art imagery is worth risking burnt fingers from a reading of Richard Lee and other Kalahari ethnographers?
Images showing perspectives of the rock art panel
In the case of the Brandewyn initiation camp hypothesis, we offer the following. Particularly with enhancement, the two lines of figures offer a high level of specificity, achieved through the assembling of much detail and generating undeniable differentiation between individuals. Why, we ask, seek out that level of detailed representation unless the model for the painting, the occasion envisaged, remembered or referenced was not a homogenized, averaged exemplar but a real, experienced event? (illustration of a range of social and life history occasions along the Brandewyn) Images of other local occasions, significant for men or for women, sometimes both, are located nearby to create a palimpsest of memory, a lived landscape. The alternative hypothesis, that place has no role in image creation, seems less likely than the one we prefer. If our argument is accepted, then why scramble real events across a painted landscape in a way that does not reflect the actual geography of historical occasions? Under such randomized practices, lived experiences become less meaningful, the bond between individual and belonging weakened. Does the complex detail of the form of the painting imply a specificity of associated time and place?
We anticipate some opposition to this view, perhaps accusing us of trivializing the underlying meaning and significance of rock art production by literalising or perhaps caricaturing the art as ‘mere’ recording. We want to make clear that specific or literal references in the paintings in no way diminish the significant ritual associations and implications. After all, we cannot consider the implications of a ‘man bleeding from the nose’ unless it is accepted that ‘man’ and ‘bleeding from the nose’ are literally justified interpretations of paint on rock. Nor, however, do these implications rule out a literal component. Quite the contrary, if the ritual events associated with a ‘group male initiation camp’ included, alongside dancing and singing, the painting and memorializing of the chosen place, then the paintings must have retained a significance and power appropriate to the events and beliefs underlying these various activities. Literal and (spi)ritual meanings co-exist and strengthen one another. Painting had the advantage over other components of creating permanence.
In future contributions we will explore the issue of handprinting, where image, place and event seem easier to associate; of scenes of violence, where antagonists may be identifiable; of elephant compositions, where behavioural detail may be a useful clue; and of images where gender- or life history-linked detail may imply specific, albeit repeated, occasions. Meanwhile, we welcome comments and criticisms.
Barnard, A. 1992. Hunters and Herders of Southern Africa: a Comparative Ethnography of the Khoisan Peoples. Cambridge. Cambridge University Press.
Bleek, D.F. 1928. The Naron: a Bushman tribe of the central Kalahari. Cambridge. Cambridge University Press.
Casey, E.S. 1996. ‘How to get from space to place in a fairly short stretch of time: Phenomenological prolegomena’ in S. Feld and K. H. Basso (eds) Senses of Place, pp 13-52. Santa Fe. School of American Research Press.
Daston, L and G Mitman (eds). 2005. Thinking with Animals: New Perspectives on Anthropomorphism. New York. Columbia University Press.
Hitchcock, R.K. 2019. Hunters and gatherers past and present: Perspectives on diversity, teaching and information transmission. Reviews in Anthropology. Taylor and Francis Group. Doi:10.1080/00938157.2019.1578025
Lee, R.B. 1979. The !Kung San: men, women and work in a foraging society. Cambridge. Cambridge University Press.
Moss, C.J., Croze, H. and P.C. Lee (eds). 2011. The Amboseli Elephants: a long-term perspective on a long-lived mammal. London. University of Chicago Press.
Parkington, J. and A. Paterson. 2021 in press. Cloaks and Torsos: Image recognition, ethnography and male initiation events in the rock art of the Western Cape., accepted Azania.
Pearce 2021, in press
Schapera, I. The Khoisan Peoples of South Africa. 1930. London. Routledge and Kegan Paul.
Viestad, V.M. 2018. Dress as Social Relations: an Interpretation of Bushman Dress. Johannesburg. Wits University Press.