PDF compression, OCR, web-optimization with CVISION's ...

Southern African Humanities

Vol. 19

Pages 123–151

Pietermaritzburg

December, 2007

The ‘cutting edge’ of rock art: motifs and other markings on Driekuil Hill, North West Province, South Africa Jeremy C. Hollmann Natal Museum, P. Bag 9070, Pietermaritzburg, 3200 and University of KwaZulu-Natal, School of Anthropology, Gender & Historical Studies, P. Bag X01, Scottsville, 3209 South Africa; [email protected] ABSTRACT Driekuil Hill is one of at least twelve sites, all on outcrops of pyrophyllite, or ‘wonderstone’, that include Gestoptefontein Mountain and Gestoptefontein Hill, regarded collectively as the largest and most significant Khoe-San rock art sites in the former Western Transvaal (now North West Province). Two types of markings are identified on Driekuil Hill based on Josephine Flood’s distinction between ‘referential art’ (pecked and incised images of anthropomorphs, zoomorphs, aprons, skins and other objects) and ‘gestural art’ (grooves, pits, hammered areas, clusters of pecks and cut marks) that may have been made as part of ceremonies. It is suggested that the art is associated with Khoe-San girls’ puberty rituals and that the wonderstone outcrops on which the motifs and other markings occur were believed by the Khoe-San to be the body of a great Water Snake. KEY WORDS: engravings, Khoekhoe, Khoe-San, petroglyphs, puberty rites, pyrophyllite, rock art, Water Snake.

In comparison to the proliferation of interpretative work done on Bushman paintings from the south-eastern mountains, research on the meanings and significance of rock engravings is less numerous. Maria Wilman (1933), first director of the McGregor Museum in Kimberley, Northern Cape Province, recorded and interpreted engravings in Northern Cape Province and Botswana. Ernst-Rudolf Scherz recorded engravings in Namibia (Scherz 1975). Gerhard and Dora Fock began recording engravings in 1958 when Mr Fock began work as Museum Archaeologist at the McGregor Museum. They produced the most comprehensive survey of southern African rock engraving sites to date (Fock 1979; Fock & Fock 1984, 1989) but for the most part avoided interpretation. More recent work on rock engraving sites focuses on possible interpretations of the rock art (Morris 1988, 1990, 2002; Ouzman 1995, 1996, 1998, 2001). These researchers have emphasised the choice of place by the artists in selecting where in the landscape to make their rock markings (Deacon 1988, 1997; Morris 1988, 2002; Ouzman 1995, 1996, 1998, 2001). This paper developed out of an archaeological impact assessment and subsequent mitigation work on Driekuil Hill (Figs 1–3), a rock art site near Ottosdal in North West Province (Hollmann 2005; Hollmann & Huffman 2005). Driekuil Hill is an outcrop of pyrophyllite, or ‘wonderstone’, the form of pyrophyllite unique to the area (Astrup & Horn 1998: 601). It is an aluminasilicate, dense but easily marked (hardness equivalent to talc, i.e. 1.0–1.5 on the Mohs scale of mineral hardness). Wonderstone crucibles are used in the manufacture of industrial diamonds, owing to the stone’s ability to resist extreme thermal shock. The mining company Wonderstone Limited sought permission to destroy the site in order to assess the quality of the wonderstone outcrop on which the rock art is pecked and incised. Before permitting the destruction of the hill—which http://www.sahumanities.org.za

123

PDF compression, OCR, web-optimization with CVISION's PdfCompressor

124

SOUTHERN AFRICAN HUMANITIES, VOL. 19, 2007

Fig. 1. The location of the Gestoptefontein-Driekuil (G-D) complex of sites, North West Province, South Africa.

v

Fig. 2. Driekuil Hill viewed from the south. It is one of 12 sites in the area that occur on outcrops of wonderstone (pyrophyllite).

Fig. 3. Driekuil Hill after the top of the hill was scraped off to examine the wonderstone deposit beneath.


HOLLMANN: ROCK ART ON DRIEKUIL HILL

125

took place in June 2005—the South African Heritage Resources Agency (SAHRA) required the removal of more than a hundred stones bearing a representative sample of the various types of markings to Klerksdorp Museum and the University of the Witwatersrand, as well as the establishment of a museum on the mine (Hollmann 2005). The selection of motifs and other markings for removal were chosen with four criteria in mind: (i) the need to rescue a representative sample of the rock art; (ii) the degree of preservation of details of particular marked surfaces; (iii) ease of removal and reassembly of marked rock surfaces; and (iv) resources available for removal and relocation. Prior to removal, occurences on the hill were mapped and photographed in detail using digital and analogue photography. A database of the occurences of motifs and other markings was created and certain motifs and other markings were traced. Data are lodged at the Rock Art Research Institute, University of the Witwatersrand, and at SAHRA’s Head Office in Cape Town. I was granted a permit to collect samples for further study. As mitigation proceeded, I realised that the markings on Driekuil Hill and the other sites in the vicinity were of great research interest, hence my use of the phrase ‘cutting edge’ in the title. Literally, the phrase describes the predominance of incised motifs and rock markings at Driekuil Hill. Figuratively, the rock markings at Driekuil Hill are relevant to ‘cutting edge’ debates about the identification and interpretation of a distinct Khoekhoen rock art tradition (Anderson 1997; Eastwood 2003; Eastwood & Eastwood 2006; Eastwood & Smith 2005; Hollmann & Hykkerud 2004; Smith & Ouzman 2004). In addition to Driekuil Hill, there are eleven known sites on wonderstone outcrops and it is likely that more exist (Fig. 4). I refer to them as the Gestoptefontein–Driekuil (G-D) complex after the surveyed farm names on which the outcrops occur. The largest site—approximately 8 ha—was Gestoptefontein Mountain, but it has been destroyed completely by mining. It is not clear whether any rock art was ever recovered from the site. The destruction of this site, possibly without any documentation, is one of South Africa’s unpublicised archaeological tragedies. Gestoptefontein Mountain, together with the adjoining Gestoptefontein Hill, both on the Wonderstone-owned portion of Gestoptefontein 349, are the best known of the G-D sites (Anderson 1888; Battiss 1948; Hübner 1871; Van Riet Lowe 1937, 1945; Zelizko 1925) and were regarded by researchers Gerhard and Dora Fock (1984) as the largest and most significant KhoeSan rock art sites in the former Western Transvaal (now North West Province). ‘SUPER-NATURAL’ QUALITIES OF THE G-D COMPLEX

Each G-D site is located on a wonderstone outcrop within 100–2 500 m of its neighbour (Fig. 4); the outcrops total about 15 km in length (De Jager 2001). The wonderstone substrate of the G-D sites provided the Khoe-San artists with unique mark-making opportunities. The softness of the rock enabled them to mark it with relative ease by cutting and hammering. The variety of mark making on Driekuil Hill—from heavily hammered surfaces to finely incised patterns—shows that they exploited this quality. The result is a proliferation of finely incised motifs and other markings that boast an amount of detail and clarity that is distinctive and exclusive to the G-D sites. There are intriguing anecdotes about these hills that confirm their significance to past inhabitants of the area. The first white farmers in the area, who arrived in the 1840s, apparently found human skeletons on Skeleton Hill (De Jager 2001: 208) (Fig. 4) . The story goes that these were the skeletons of ‘Bushmen’ who had fled there


126


Fig. 4. Detailed map of the G-D complex. The star symbols denote those parts of the pyrophyllite outcrops on the farms Gestoptefontein 349 IO and Driekuil 280 IP known to bear markings and motifs. It is likely that there are yet more marked areas, although these are probably not as extensive as those already known. In most cases it was necessary to coin names for these sites. In each such case the name was chosen to reflect an easily identifiable characteristic. Scale in 0.5 km units. 11 Vorster’s Farm (Gestoptefontein 349 IO): motifs and other markings reported but not yet investigated. 12 Married Quarters (Gestoptefontein 349 IO): rough estimate of area = 0.01 ha. Not surveyed. 13 Gestoptefontein Mountain (Gestoptefontein 349 IO, portion owned by Wonderstone Ltd): rough estimate of area = 8 ha. Completely destroyed by mining. 14 Gestoptefontein Hill (Gestoptefontein 349 IO, portion owned by Wonderstone Ltd): 1.19 ha. Currently being surveyed. 15 Charlie Badenhorst’s Farm (Gestoptefontein 349 IO): rough estimate of area = 0.01 ha (100 m2). 16 Driekuil Hill (Driekuil 280 IP): estimated area 0.63 ha. Completely destroyed by mining. 17 The Mound (Driekuil 280 IP): a few, scattered, incised marks. No estimate of area. 18 Skeleton Hill (Driekuil 280 IP): rough estimate of area = 0.15 ha. 19 Driekuil Ridge North (Driekuil 280 IP): rough estimate of area = 0.15 ha. 10 Driekuil Ridge South (Driekuil 280 IP): rough estimate of area = 0.1 ha. 11 Boschpoort Road North (Driekuil 280 IP): rough estimate of area = 0.15 ha. 12 Boschpoort Road South (Driekuil 280 IP): rough estimate of area = 0.02 ha.



127

in the belief that they would become invisible and escape their pursuers. The hills apparently had special extra-ordinary qualities that they believed would afford them protection—there are no natural features on the hills such as caves or dense vegetation that would have provided cover or camouflage. Gestoptefontein Mountain and Gestoptefontein Hill have another oral tradition associated with them. Since the 1870s at least, when the explorer and collector Emil Holub visited Gestoptefontein, they have been known by the Tswana name Tlogo Pitsane (Zelizko 1925)— still used today in preference to ‘Gestoptefontein’. The name, according to Mr Josia Modise Rapoto, who now lives in the village of Mareetsane, North West Province, and worked at the mine for twenty years, means ‘horse’s head’ (pers. comm. June 2007). He explained that the name comes from an enormous snake that lived (and probably still does, says Rapoto) on these two outcrops. The snake basks in the sun on a large slab of wonderstone. The shadow cast by its head resembles that of a horse, hence the name Tlogo Pitsane. At night the snake roams around with a light on its head. Rapoto related that a colleague of his had seen the snake and told him that it could also swim and fly and walk, ‘like a likkewaan’ (monitor lizard). The snake is apparently undisturbed by the mining activities but, Rapoto added, should dynamite ever be used, the snake would become very angry—he mimed the snake writhing furiously—and displace the hills from the mine property. Rapoto’s explanation of the origins of the name ‘Tlogo Pitsane’ is deeply embedded in traditional southern African beliefs. The characteristics of the snake he describes— its size, the un-snakelike shadow cast by its head, shape-shifting abilities, the light on its head, nocturnal activities, and immense power—indicate that this is no ordinary snake. The qualities Rapoto describes are all features ascribed by Khoe-San and Bantu speakers to the Water Snake, a creature believed to control the flow of water and the appearance of rain (Hoff 1997; Van Vreeden 1955, 1959). The belief that the snake has a horse’s head, I discovered later, has wide currency. Mr Poem Mooney, a member of the Attaqua, a Khoekhoen group in the Western and Eastern Cape Provinces, observed that the Koranna believe that a Water Snake has a horse’s head (pers. comm. 10 August 2007). Both Rapoto and Mooney’s comments suggest that the name Tlogo Pitsane links Gestoptefontein Mountain and Gestoptefontein Hill to the presence of a Water Snake. These stories demonstrate that for at least a century people have attributed special qualities to the wonderstone outcrops. These beliefs could help explain why most wonderstone outcrops have been marked. The artists may have construed the texture of wonderstone, the restricted distribution of the outcrops, and their proximity to the Driekuil stream1 as evidence of its ‘super-natural’ qualities (see Tilley & Bennett’s (2001) interpretation of ‘natural’ features in the United Kingdom). David Morris (2002: 203, 217) suggests that the engraved glaciated pavement at Driekopseiland on the Riet River, Northern Cape Province, was seen by Khoe-San people as the embodiment of a Water Snake that was submerged in summer and exposed in winter. Perhaps at certain times in the history of the G-D sites, engravers held similar beliefs about the wonderstone 1

The existence of the Driekuil stream and the presence of several springs in the area over at least the past hundred years is borne out by the names of several farms in the area called Korannafontein (‘Koranna spring’) after the Koranna, a group of Khoekhoen; the town Ottosdal itself used to be called Korannafontein (De Jager 2001).


128


outcrops—parts of these distinctive formations ‘emerge’ from the surrounding countryside, others are ‘submerged’ below ground. Such features may have occasioned a feeling of ubiety—the word that Morris (2002: 201) uses to describe Khoe-San attitudes to the Driekopseiland engraving site. Janette Deacon’s (1988: 138) pioneering discussion of topophilia, the ‘human attachment to familiar places’, and Sven Ouzman’s (1998) work on ‘mindscapes’ is also important in this regard. These supernatural qualities of the wonderstone outcrops and the possibility that they represent a great Water Snake are central to my interpretation of the motifs and other markings on Driekuil Hill. BACKGROUND TO THE STUDY

In this paper, I build interpretations of the motifs and other markings on the recently destroyed site on Driekuil Hill (Hollmann 2005; Hollmann & Huffman 2005). Driekuil Hill is a promising subject for investigation: mitigation has provided a body of data for analysis and the presence there of ‘geometric’ motifs—linked by some rock art researchers to the presence or influence of Khoekhoen immigrants—provides the opportunity to investigate the ways in which a specific rock art site in an underresearched area may reflect the nature of this influx. In addition, the presence of apparently nonrepresentational markings such as fine incisions, grooves, and pecked areas whose proliferation is made possible by the softness of the substrate, forces one to challenge the distinction between utilitarian mark-making and ‘art’ and to appreciate the importance of local nuances in rock art studies. I use Leonhard Schultze’s (1928) term ‘Khoe-San’ (originally ‘Khoi-San’)—that includes Khoekhoen, Damara, Khoe-speaking Bushmen and non-Khoe-speaking Bushmen—to encompass the probability that over many centuries, even millennia, the G-D sites were significant to a variety of groups of click-speaking indigenes that existed as hunter-gatherers and herders, a distinction that correlates with the terms Bushman/ San and Khoekhoen. This division between ‘hunter-gatherer’ and ‘herder’ is not hard and fast, however (Barnard 1992: 17), nor is it reflected in deeply entrenched pan KhoeSan worldviews. Beliefs and ceremonies regarding girls’ puberty, for example, are similar and widespread and cut across differences in modes of production. Tony Humphreys (2003) has recently criticised what he (and others such as Mitchell (2005: 67–8)) see as the abuse of ‘Kalahari analogues’ (Humphreys 2003/2004: 36), that is, the over-reliance on !Kung Bushman ethnography, or what Mitchell (2005: 67) calls the ‘holy trinity’ of Ju|hoansi, G|wi and |Xam, to understand the Later Stone Age in South Africa. Another reason for using a term as general as Khoe-San is to avoid what Morris (2002) has criticised as the ‘reification’ of ethnicity. He points out that certain studies of the Northern Cape rock art site Driekopseiland were obsessed by the question of race and identity, as though by stereotyping the artists in this way one could understand their work and the significance of the place. The use of ethnographic analogy is inescapable and indispensable when delving into issues of archaeological and anthropological meaning and interpretation (Wylie 2002). But how does one apply ‘ethnography’ or ‘indigenous knowledge’ to a site like Driekuil Hill? The G-D sites fall within an ancient zone of ‘moving’ and ‘static’ frontiers (Alexander 1984; Humphreys 1988). San hunter-gatherers, Khoekhoen pastoralists and, more recently, Tswana groups, have shared the landscape. The challenge one faces


129


when trying to understand the motifs and other markings on Driekuil Hill is that it is a fixed and prominent point in a dynamic social and historical landscape. How can ethnography help to identify possible processes of fusion and change that may have taken place in specific, local contexts without falling prey to ‘ethnic essentialism’ and overarching generalisations? Obviously, this is not an easy question to answer. Not much is known about the local prehistory, nor are there any absolute dates associated with the rock art. There is a paradox here. While one wants to be sensitive to local historical developments and processes of identity formation, the probability that fluid social, political and economic conditions may have pertained over the past two millennia requires one to interpret the rock art by analogy with the broadest possible range of Khoe-San beliefs and practices. In the absence of Khoe-San ethnography about ‘mark-making behaviour’, I look for analogues in cultures that have produced similar-looking rock markings. The intention is to recognise concepts and practices in other cultures that may have parallels among Khoe-San. I use Josephine Flood’s (2004) work amongst indigenous Australians, as well as studies of Native American rock art (Callahan 2000; Loubser 2005; Taylor & Keyser 2003), in which the concept of supernatural potency is linked to properties of certain kinds of rock and to rock art. THE ROCK ART OF DRIEKUIL HILL

The documentation of the motifs and other markings on Driekuil Hill required the creation of categories. A typological approach enables one to distinguish quantities, forms and sequences within a body of rock markings (Table 1). However, a typology is only as good as its structuring principles—the old computer maxim, ‘garbage in, garbage out’ also applies to the construction of a typology. Out of more than 1 000 motifs and other markings on over 350 rocks, just over 1 % depicts zoomorphs and anthropomorphs—imagery one would usually associate with a hunter-gatherer worldview. Similarly, the pecked and painted ‘geometric’ motifs associated with Khoekhoen pastoralists (Anderson 1997; Eastwood & Eastwood 2006; Eastwood & Green 2007; Eastwood & Smith 2005; Smith & Ouzman 2004) comprise only 12 % of the total. The TABLE 1 Typology of rock markings at Driekuil Hill. Note that statistics are not available for Category 3 and 8. These two categories were only subsequently discerned. They are subsumed under Category 12 and 7, respectively. Category 1 2 3 4 5 6 7 8 9 10 11 12

anthro/zoomorphic incised patterns parallel cut marks incised grids and meshes pecked geometric motifs (aprons) pecked geometric motifs (not identified) clusters of peck marks percussion marks (groups of) deeply pecked pits abraded grooves clusters of incisions irregular incisions Total

Count

%

11 3 — 100 47 87 298 — 32 51 194 250 1 073

1.0 0.3 — 9.3 4.4 8.1 27.7 — 3.0 4.8 18.1 23.3 100


130


remainder consists of hundreds of rock markings—some very striking—that do not seem to depict ‘things’ or ‘objects’ at all: these are clusters of fine cuts, pits and grooves scattered all over the hill, alone and in association with motifs. They challenge the distinction between ‘utilitarian’ and ‘non-utilitarian’. I therefore found Josephine Flood’s distinction (1997: 148; see Ouzman 2001 for more South African examples) between ‘referential’ and ‘gestural art’ useful to distinguish two broad categories of mark-making on Driekuil Hill: pecked and incised engravings of objects (referential) and rock markings of various kinds that seem to be ‘by-products’ of activities (gestural). I do not imply that the mark makers themselves discerned such a dichotomy—merely that the distinction is useful for analytical purposes and then only to a limited extent. One cannot pigeonhole all kinds of markings into one or the other category; some markings may straddle these boundaries. In the absence of ethnography or for other reasons, one may allocate a particular marking incorrectly and end up drawing erroneous conclusions. Referential motifs and other markings I start with the the component of the rock art that appears to be ‘referential’, i.e. that depicts ‘objects’. 1. Peckings of anthropomorphs and zoomorphs Usually the mainstay of a ‘typical’ hunter-gatherer engraving site, zoomorphs and anthropomorphs (Fig. 5) comprise only about 1 % (11 recorded instances) of the total of motifs on Driekuil Hill. A single engraving represents the normally ubiquitous eland symbol. There are eight anthropomorphs and a single therianthrope (human–animal combination). People here had a different and visually distinct repertoire of markings. In order to understand these we must explore avenues of thought additional to interpretations of hunter-gatherer art based on motifs of anthropomorphs and zoomorphs. 2. Incised ornamental patterns Incised ornamental patterns comprise a few standardised and repeated forms, some of which are bounded by lines (Fig. 6). Their appearance is ornamental, similar to motifs used on Khoe-San-manufactured sticks, ostrich eggshell containers, headbands, pubic aprons and belts (e.g. Humphreys & Thackeray 1983; Luschan 1923; Valiente-Noailles 1993; Wiessner 1984: fig. 3e). Some of these patterns have symbolic significance and affective power in many Khoe-San cultures (Valente-Noailles 1993; Wiessner 1984). None of these ornamental patterns is involved in superpositioning—one cannot therefore place them within a relative chronology of motifs. They are, however, juxtaposed with the other categories and do not necessarily represent a discrete episode. The significance of the motifs may be similar to the principle that underlies the parallel cut marks and pecked geometrics that I discuss next. 3. Series of parallel cut marks Another category on Driekuil Hill comprises sets of parallel cut marks (Figs 7, 9, 10). Such incisions occur all over the site in configurations that vary in dimension and number. Each individual set of markings, however, is of approximately even length and equal spacing. Some of these arrangements seem reminiscent of tattoo marks, as Gerhard



131

Fig. 5. Pecked engraving of an eland, one of the very few zoomorphs on Driekuil Hill.

Fig. 6. Incised pattern of nested diamond shapes, about 170 mm wide × 20 mm high.

Fock has suggested of motifs elsewhere (1969: 126). Tattooing (Fig. 8) is widely practiced by Khoe-San in association with initiation, marriage and hunting rituals (Barnard 1992; Hoff 1990; Hollmann 2004; Marshall 1999; Schapera 1930; Silberbauer 1963; ValienteNoailles 1993; see Mitchell & Plug 1997 for an overview). Ashes of magical and medicinal roots are rubbed into the cuts to achieve a variety of ends. Sometimes the putative fat of the Water Snake is used to protect people from attack (Hoff 1997: 31). In the context of female initiation among the G|wi, these substances are used to ward off


132


Fig. 7.Two rows of parallel cut marks about 90 × 12 mm.

Fig. 8. Series of cut marks may depict tattoo marks similar to those on the face of this G|wi initiate. (Photograph of woman from Silberbauer 1965: 85. Original caption reads ‘G|wi girl after her puberty ceremony, showing tattoo marks and tonsure.’)



133

Fig. 9. Several sets of short, roughly parallel and evenly spaced cut marks on rock surface, 290 × 105 mm.

Fig. 10. A series of ten cuts in the rock, approximately 90 × 10 mm.


134


‘want and disaster’ (Silberbauer 1963: 20). People regard the markings as attractive to others, as well as to God and the spirits of the dead (Valiente-Noailles 1993: 163). In certain cases, the number of cuts made and their placement and arrangement on the body is very specific. Some of the configurations of cut marks on Driekuil Hill may be modelled on particular patterns. Carlos Valiente-Noailles (1993: 163) mentions an incident that may illustrate a principle behind people’s motivation to make tattoo marks on the rocks: when referring to decorations on an apron that her husband had made for her, a woman explained to us that it had the same design as the incisions that her mother had made on her face when she was a child.

Valiente-Noailles (1993: 163, original emphasis) also notes that people often compared the designs on objects to the markings on their faces and bodies as though the former [i.e. designs on objects] were derived from the latter [i.e. markings on face and body] and the original reasoning were ‘if these markings beautify the body they also embellish these objects’.

The act of transposing the markings from body to object may therefore have affected the transference of supernatural potency from person to possession. Similar processes of identification by the association and transference of supernatural potency using motifs occurs in another sphere of Khoe-San life—female initiation. Ansie Hoff (1997: 30) recounts how a Griqua initiate is associated with the Water Snake ‘by means of her make-up’: A spot on the forehead, linked eyebrows and beauty spots were painted on her face so that she resembled the Water Snake and therefore associated with him.

In this example, the adoption of the facial markings of the supernaturally potent Water Snake confers protection on an initiate. The rationale that Hoff describes is essentially similar to that underlying Vailente-Noailles’s example and may be extended to understanding the significance behind not only the creation of tattoo marks on the rock, but more generally to the making of other motifs on the rocks. The transfer of supernatural potency was probably a two-way process—recall that these places may have embodied the Water Snake itself. The association and transference of supernatural potency between people, objects and supernaturally potent places are fundamental to my understanding of the rock art at the G-D sites. 4. Incised grids and meshes Grid and mesh motifs range in size from about 10 × 10 mm to approximately 500 × 200 mm. A ‘grid’ is an arrangement of incisions that intersect at approximately 90˚ (Fig. 11), while the incisions that constitute a ‘mesh’ intersect at about 45˚ (Fig. 12). These designs make up about 9 % of images on Driekuil Hill (100 recorded instances). Analogous forms may occur in the Magaliesberg, some 200 km north-east of the G-D sites (e.g. Steel 1988: 33), elsewhere in North West Province (Morris 1988: 116), and further afield in Northern Cape Province (Morris 1988: figs 6g, h, j). Similar-looking motifs at Wonderwerk Cave date from between 1 800 to 12 500 years BP (Beaumont 1990; Thackeray et al. 1981). Francis Thackeray (2002) suggests that the Wonderwerk Cave motifs are entoptic in origin, but this identification alone does not explain their symbolism. In the past, many motifs that we now recognise as Khoekhoen designs, such as zigzags, concentric circles,



135

Fig. 11. Example of an incised mesh.

Fig. 12. Example of an incised grid.

and spirals—and whose entoptic provenance is uncertain—were labelled entoptic with no further attempt made to place them in context or unpack their symbolism (e.g. Dowson 1992: figs 32, 37, 42, passim). I do not discount an entoptic (visionary) origin for grids and meshes—certain decorative motifs do derive from entoptic imagery (ReichelDolmatoff 1978). Amongst the Lakota in the Black Mountains of North America, for example, women may have visions of quillwork or beadwork designs. Although not necessarily derived from direct experience of entoptics in an altered state of consciousness, these designs enhance the beauty and supernatural power of the item to which they are applied (Sundstrom 2004: 95–6).


136


Perhaps, as I argued for the tattoo motifs, grid and mesh motifs have strong symbolic and affective associations. Grid and mesh designs occur on Khoe-San bone artefacts (Heinz 1966: 123, fig. 20; Inskeep 1987: plates 14, 15) and ostrich eggshells (Humphreys & Thackeray 1983; Luschan 1923). The grids also recall the sieves that women used to sift ants’ eggs from the soil or to separate melon seeds from the ash after roasting. For the nineteenth-century |Xam such sieves had a mythological origin—they originate from an ancient time when animals were people. The sieve itself derived from the hindquarters of a ‘quagga-woman’ (Diä!kwain in Bleek & Lloyd 2005). On a large scale of analysis grids are very widespread and generalised forms with a variety of significances. On a more local and restricted level, however, they may constitute a specific genre. Further documentation of finely incised grids and meshes at the other G-D sites may yield patterns and associations that suggest possible interpretations for these designs. 5. Pecked geometric motifs Pecked geometric motifs account for about 12 % of all markings recorded at Driekuil Hill. They are sometimes superimposed on incised motifs (including grids and meshes). This sequence was also identified at Gestoptefontein Hill (Van Riet Lowe 1937: 255), although as Van Riet Lowe (1945: 333) remarks of superpositioning of engravings in the Magaliesberg, ‘we cannot say whether the difference in age is to be gauged in minutes or millennia’. Inter-site comparisons at the G-D sites may help to establish the time depth between these kinds of markings. Although pecked geometric motifs form a relatively small proportion of the total number of motifs and other markings, their significance is paramount. Following work by Ed and Cathelijne Eastwood (Eastwood 2003; Eastwood & Eastwood 2006) and Catherine Namono (Namono & Eastwood 2005) in Limpopo Province, it is evident that certain motifs on Driekuil Hill and the other G-D sites depict various kinds of aprons. Emil Holub, who removed at least 140 engravings from Gestoptefontein to European museums, reached this conclusion over a hundred years ago (Zelizko 1925). Three types of apron are depicted on Driekuil Hill. Their format seems to correlate with the apron types found amongst Khoe-San—fringed pubic aprons (Fig. 13), decorated pubic aprons, and rear aprons. One of the stones removed from Driekuil Hill is piece #63b, an elaborate engraving of a pubic apron (Fig. 14). This depiction is unusual in its large size and detail, which includes tassels around the edges of the apron and rows of triangular shapes within, probably representing apron decorations. It may represent the new and elaborately patterned pubic apron that women present to a female initiate. According to Lorna Marshall’s research among the Ju|’hoansi in Namibia and Botswana (1999: 147): The front apron that covers the women’s genitals is a symbol of female sexuality. A man should not touch the apron, and, in a strictly observed avoidance, he must not say aloud the word for front apron. If a man said the word, the meat of the animal he killed would taste very bad and rot quickly. If he touched the apron, he would become lazy and loll about in the camp instead of setting out to hunt—which is totally unacceptable social behavior for men in hunting age.

Thus amongst the Ju|’hoansi aprons are not simply items of clothing—they have powerful associations. Aprons are indicators of female status change in many Khoe-San societies (e.g. Engelbrecht 1936; Marshall 1999; Valiente-Noailles 1993). Amongst the G|wi, G||ana (Valiente-Noailles 1993: 163) and Koranna (Engelbrecht 1936: 104, fig. 6), for



137

Fig. 13. Pecked image of a fringed pubic apron, 110 mm wide and 75 mm high.

Fig. 14. Spectacular pecked front pubic apron, 365 mm wide and 200 mm high.


138


example, an uninitiated girl wears a fringed pubic apron made from plaited plant fibres or strips of leather. After she emerges from the mandatory seclusion required during her first menses, a ‘new’ woman receives a new, often highly decorated, pubic apron, as well as a larger rear apron (Valiente-Noailles 1993). The rear apron is often a single animal skin that wraps around the hips and reaches down to the knees. A portion of the legs of the animal skin is retained and is used to tie the skin around the waist. Some of the Kgalagadi and Batlhaping groups—who lived in close proximity to Khoe-San communities in what is now South Africa’s North West Province—wore similar apron types (e.g. Duggan-Cronin & Lestrade 1929: plates 3, 26). There is an additional motif on Driekuil Hill that could relate to Khoe-San female puberty rituals. This design resembles the engravings of fringed pubic aprons. However, instead of the usual five-stranded apron, these forms have seven or more strands (up to twelve) and are consistently wider than the aprons (Fig. 15). Valiente-Noailles (1993: 164) reports that the G|wi and G||ana in central Botswana make beautiful ornaments for the initiate, including headbands. Perhaps the broader motifs represent a headband given to a girl on her transition to womanhood. Headbands probably had a power and symbolism of their own (Wiessner 1984). Pecked motifs of aprons and headbands may thus be feminine symbols associated with status change of females from ‘girls’ to ‘women’, a time when their supernatural potency is exceedingly strong and must be managed (Engelbrecht 1936; Hoernlé 1985; Hoff 1995; Hollmann 2004; Marshall 1999; Silberbauer 1963; Valiente-Noailles 1993; Waldman 1989). Many Khoe-San women describe their emergence as fully-fledged adults as one of the happiest times of their lives (Marshall 1999: 199; Valiente-Noailles 1993: 164). The pecked clothing and other decorative motifs on Driekuil Hill may therefore have had great affective power based on such positive attitudes toward the transition to womanhood. Other pecked motifs at the G-D sites remain unidentified, but based on those designs already distinguished, it seems likely that some will turn out to be abstracted representations of other symbolically and affectively charged objects. 6. Clusters of peck marks Clusters of individual peck marks are one of the largest categories of markings, at around 28 % of the total recorded (Fig. 16). The individual peck marks that comprise the clusters vary in diameter and depth; some are very consistent in size and shape, suggesting that people used a punch to make them. The clusters themselves vary in their degree of ‘tightness’—the distance between the individual peck marks that make up the cluster. Clusters range from ‘tight’ (individual peck marks less than 10 mm apart) to ‘loose’ (peck marks up to 50 mm apart). Clusters of peck marks cover areas from 50 to 200 mm2. They occur on their own, together with other engravings and markings, and often, over clusters of cut marks. People seem not to have made peck marks over pecked apron motifs. It is not clear which clusters of peck marks are referential and which may be gestural. The clusters could relate to patterns of dots made on the initiate’s face and other body parts, or the body of the Water Snake. If so, they would be more referential than gestural in nature. Other clusters may be gestural—in this scenario, people struck the rock surface because the rhythm and, possibly, the sound and sensation of the blows contributed to achievement of states of mind desirable for the occasion (Ouzman 2001).



139

Fig. 15. Pecked image, perhaps of a headband.

‘Gestural’ art: residues of ritual activity? Perhaps the most striking feature of the Driekuil Hill site is the proliferation of apparently non-representational rock markings that are juxtaposed with incised and pecked forms that depict objects. Where superpositioning occurs, gestural markings are always under the pecked geometric motifs. Gestural markings are probably neither ‘images’ nor ‘utilitarian’ in nature but rather evidence of another sort of (stereotyped) marking behaviour. People were doing something, probably the same thing, repeatedly on different parts of the hill. I interpret the presence of such patterns as graphic evidence of conventional, patterned behaviour that, given the context of the wonderstone hills, may indicate the performance of ritualised activities. The marks are thus ‘residues of ritual activity’ (Lewis-Williams & Blundell 1997). Not that people necessarily regarded the marks as ‘mere’ by-products—the very fact that they structured their behaviour to include mark making suggests that they regarded the resultant patterns as important and significant marks in their own right. 7. Smoothed surfaces With few exceptions and regardless of whether they subsequently made ‘images’ or ‘non-representational’ markings, people first ‘prepared’ the rock surfaces they utilised by smoothing them (Fig. 17). There is no indication that any special tools were used— the rock is soft enough to remove the outer patina and create a smooth and lighter shade of rock by protracted rubbing, perhaps with animal-hide clothing. Smoothing rocks may have been an important step in preparing the rock surface for marking, but it could also have had supernatural or cosmological significance. In the context of Khoe-San cosmology, certain rock formations and surfaces were considered potent, permeable, and liminal pathways to the spirit realm (Lewis-Williams & Dowson 1990; Ouzman 1998, 2001). Ouzman (2001: 245) suggests that:


140


Fig. 16. A cluster of peck marks 110 mm wide × 60 mm high. Repetitive rubbing, possibly with a strong-smelling animal hide, and the sweat from one’s exertions . . . would have created a powerful and visceral bodily engagement with image, rock and the Spirit World beyond.

To this picture, one might add the accompaniment of singing and chanting. Similarly, smoothing of rock surfaces after marking them may have been part of ritual activities. 8. Percussion marks When I described and interpreted the referential motifs on Driekuil Hill, I drew on Khoe-San ethnography to identify them and to place them in an ethnographic context. Not much is known about the manufacture and significance of non-representational markings, however, beyond a snippet—the well-known nineteenth-century instance of a |Xam woman pounding the earth with a ‘digging-stick stone’ in order to ascertain from the ‘spirit people’ why her husband was unfortunate in hunting (Diä!kwain, ‘Luck in hunting’ in Hollmann 2004: 264). Here gestural activity (striking the ground) may have created impact markings on the digging-stick stone that were incidental to the activity itself. A similar principle may apply to those rocks on Driekuil Hill that have clearly been battered (Fig. 18)—the marks of the blows dealt to the stones may record people’s attempts to make contact with the supernatural realm (Ouzman 2001). 9. Deeply pecked ‘pits’ ‘Pits’ are bigger and deeper than peck marks and range from 10–50 mm in diameter and 5–35 mm deep (Fig. 19). People produced them by focusing blows on a small area of rock. Sometimes pits occur in isolation or in rows, but in other cases, people made them on top of other motifs, but not over pecked ‘geometric’ motifs. Pits make up almost 3 % of the recorded total.



141

Fig. 17. People have smoothed this section of rock on Driekuil Hill by protracted rubbing.

Fig. 18. A smoothed stone 600 mm long × 300 mm high, covered with percussion marks.


142


Fig. 19. A deeply pecked pit about 35 mm in diameter and 5 mm deep.

Flood (2004) reports on pits in the Musgrave Ranges in Central Australia at a so-called ‘increase site’. This is a large boulder, on the sides of which people have hollowed out many pits. During the 1940s, ethnographer C. P. Mountford observed the making of some of these pits and recorded their significance (Mountford 1976). The boulder is the body of Tukalili, a pink cockatoo woman, and the pits are wounds. The production of pits was part of an increase ritual for pink cockatoos, whose eggs are an important food source. The dust produced by hollowing out the pits was the kuranita (life essence) of cockatoos. This dust inseminates living female cockatoos and causes them to lay eggs. Two aspects of this ethnography are relevant to understanding the presence of the pits on Driekuil Hill. The production of dust for supernatural purposes is a transitory and invisible phenomenon— without the relevant ethnography, the reasons behind the production of pits are obscure or incomplete. Secondly, although one may argue (as does Flood), that the resultant pits have less significance than the production of dust, these pits have nonetheless been integrated into the ‘story’ of the site and are imbued with meaning. The Australian analogy underlines the importance of local ethnography for understanding specific examples of widespread and superficially similar markings (in this case pits) and demonstrates the importance of emic understandings of rock markings in which the process of production is at least as important as the product itself. In North America, Kevin Callahan (2000) cites ethnography collected by Merriam (1955) and Parkman (1995) that links cupmarks to concepts of fertility. Amongst the Pomo in California, for example, people treated infertility by eating rock dust obtained from cupmarks. The link between fertility and rock dust is emphasised by the term that the Pomo and Shasta use for cupmarks—they are ‘baby rocks’ (Heizer 1953; Merriam 1955; Parkman 1995: 8, cited in Callahan 2000). Callahan also cites Merriam (1955) to the effect that the production of cupmarks released supernatural potency and attracted game and rain. Pits and rock dust are sources of supernatural potency associated with



143

human fertility and, more generally, with the abundance of animals and precipitation— positive events that relate directly to notions of fertility. These ethnographic analogies from Australia and North America provide a background to understanding the pits on Driekuil Hill. How do Khoe-San notions of ‘fertility’ relate to the pits on Driekuil Hill? Khoe-San girl’s puberty rituals are indeed concerned with issues of fertility, abundance and ‘good’ and ‘bad luck’ (Hoff 1995: 30; see Eastwoood (2005) on rock paintings and female puberty rituals). Many observances are intended to ensure that the female initiate behaves appropriately so that her considerable liminal potency does not devastate the community, its livestock and the vegetation on which they depend. A Ju|’hoan girl may not touch the ground with any part of her body—were she to do so, she would become very thin and waste away (Marshall 1999: 191, 192). A Nama girl menstruating for the first time may not walk in the veld because the edible roots and berries would shrivel and die (Hoernlé 1985: 63). After her emergence from seclusion, a Khoe-San female is formally reintroduced to the community, livestock, the local water source, and to the land itself (Hoernlé 1985; Hoff 1995: 37–39). It is auspicious to be the first person presented to an initiate—Silberbauer (1963: 87) reports that amongst the G|wi to a man it brings specially good fortune in hunting and a woman is helped by this in her searches for food and in bearing and rearing her children.

Amongst the Nama, the newly-initiated young woman dusts the testicles of the young men with an aromatic powder, to protect them from sexual disease and to ensure fertility (Hoernlé 1985: 65). Thereafter she goes to the cattle kraal and strews the same powder onto all the male animals, as well as all the trees and bushes encountered on her way. Finally, she is taken to the local water source and is scattered with water to ‘introduce’ her to the Water Snake (Hoff 1995: 38; Waldman 1989: 38–43). All these beliefs and practices revolve around abundance and fecundity. How exactly may the pits on Driekuil Hill have fitted into this picture? Perhaps the wonderstone rock itself, with its wondrous properties, was the great Water Snake’s body. The hollowing out of pits may then have released the snake’s supernatural potency. People may have eaten the rock dust as a means of linking themselves to the Water Snake.2 The dust could have been an ingredient of supernaturally potent mixtures. Initiates may have been painted with a potent slurry of dust and water in the various designs that are part of this rite of transition (Hoernlé 1985; Hoff 1995; Waldman 1989). The pits on Driekuil Hill may therefore have been implicated in a web of (now invisible) ritual activities. 10. Abraded grooves Grooves range from 5 × 5 × 100 mm to 50 × 50 × 1 000 mm (Fig. 20). They comprise just under 5 % of the total of recorded motifs and other markings on the hill. Grooves are widespread in southern Africa—in Botswana (Van der Ryst et al. 2004), Limpopo (Eastwood & Eastwood 2006), Free State (Hollmann & Swart 1991), Eastern Cape (see slide RSA KRO2 1 in South African Rock Art Digital Archive http:// ringingrocks.wits.ac.za) and Western Cape (Sadr & Fauvelle-Aymar 2006). I do not assume that all instances of grooves on the subcontinent are conceptually similar and confine myself to a discussion of the grooves on Driekuil Hill. 2

I do not preclude the possibility that the powder may also have had other uses, some more utilitarian.


144


Fig. 20. A cluster of about 50 abraded grooves on the rock surface, 1.6 m × 0.45 m.

Grooves are often in clusters, arranged in roughly parallel ‘rows’. They may occur on extensively incised rock surfaces. Another noteworthy and repeated configuration is the arrangement of two grooves to form an ‘X’ shape. Two sets of rocks at Driekuil Hill feature spectacular concentrations of grooves (rocks #12 and 51). In both cases, and subsequent to groove manufacture, people have repeatedly struck the grooves and portions of the surrounding rock with a stone or other heavy object to create ‘pits’. There is a popular local belief that the grooved rocks on Driekuil and Gestoptefontein Hill are ‘sharpening stones’. The explanation offered is that Mzilikazi’s warriors left these marks early in the nineteenth century after using the rocks to sharpen their spears. There are, however, problems with this account. Grooved stones are not unique to Driekuil Hill: they occur far beyond the historical distribution of the Matabele, often at rock art sites (e.g. Eastwood & Eastwood 2006; Van der Ryst et al. 2004). Nor does the spear-sharpening hypothesis account for the pits and peck marks associated with grooves, or the apparent ‘development’ of fine incisions into grooves that I discuss under the next heading. It also seems unlikely that the grooves are residues of metal spear sharpening. The process of abrading the rock to form grooves will soon blunt an iron spear point or blade—to sharpen an iron blade one abrades the sides on a flat stone surface. The proposal that the grooves at Driekuil Hill are residues of Matabele spear sharpening therefore seems unlikely. Nonetheless, some sort of instrument—stone, bone, or metal—is associated with the grooves. Whatever it was, the rock surface and the abrading tool were both shaped by the encounter. Was the production of a round-tipped or spindle-shaped artefact the primary aim of groove manufacture? In the absence of Khoe-San



145

ethnography, I turn to accounts of abraded grooves in Australia and North America. Linea Sundstrom (2004) suggests that grooves in the rock art of the Lakota in the Black Mountains are part of girls’ puberty rites—they are the residues of bone awl manufacture. To make an awl people rub a suitably sized piece of bone on the rock—in the process, an awl and a groove are created. Amongst the Lakota the awl ‘symbolised femininity’ and at puberty a girl is presented with her own awl and awl case (Sundstrom 2004: 90). Moreover, Sundstrom suggests that the process of shaping the awl in the groove involved repetitive, rhythmic action and induced a meditative state. Thus ‘tool sharpening might not have been just tool sharpening— it might also have been an accompaniment to chatting, swaying, or other prayer activity’ (Sundstrom 2004: 94). The Wardaman in the Victoria region of Australia’s Northern Territory told Josephine Flood (1997: 150) that the grooves and percussion marks at certain sites ‘bring out the power in the rock’. These rocks embodied ancestral beings whose power was released through rubbing or cutting their ‘rock-flesh’. Given the significance the Khoe-San attach to rocks and rock formations as ‘super-natural’ entities (e.g. Deacon 1988) and the significance likely associated with the wonderstone hills, it may be that the people who marked the hills on Driekuil and Gestoptefontein had analogous beliefs and attitudes. Perhaps groove making released supernatural potency from the Water Snake’s body. The object used to make the grooves may also have been an item that was symbolic of women, perhaps a bone-tipped digging stick, or a stone object charged through use. As with pit manufacture, the resultant rock powder may have been eaten and used as an ingredient in paints and powders. 11. Clusters of parallel and overlapping incisions Clusters consist of a group of at least ten parallel and overlapping incisions, but in a few instances, the number of cuts exceeds a hundred. An incision is a single, relatively thin cut in the rock, 1 mm or less in breadth and depth. A typical cluster is approximately 50 × 150 mm (Fig. 21). They comprise 18 % of recorded markings and engravings. People made these cuts all over the northern, western, and southern slopes. In common with all the other categories of rock marking, they are absent from the eastern slopes, perhaps because of the lack of large enough surfaces. Clusters of parallel and overlapping incisions and pecked clusters often occur together—‘clouds’ of peck marks are commonly superimposed on the sets of incisions. There may be conceptual (and sequential) links between these two categories. In at least ten instances, people transformed an incision into an abraded groove by broadening and deepening the cut mark. This phenomenon may suggest symbolic links between these sorts of incisions and the manufacture of grooves. Examination of published rubbings of engraved surfaces in the Magaliesberg suggests that similar clusters of parallel and overlapping incisions occur there (e.g. Steel 1988: 26) and that they are associated with finely incised depictions of antelope and rhinoceros, as well as grid and mesh motifs. Given what I have already discussed about the supernatural potency of the rock surface, it seems reasonable to infer that the act of cutting into the rock surface relates to similar circumstances. More specifically, the clustering of cut marks could suggest that each cluster is the residue of a discrete sequence of cuts, testimony to an event that was


146


Fig. 21. A cluster of parallel and overlapping incisions, 100 × 45 mm.

repeated all over the hill. Groups of clustered cuts accumulated on a rock surface in the same way that cut marks collect on a kitchen chopping board. People were perhaps cutting up substances on the rock as part of a ceremony or purposely cutting into the rock itself. Future use-wear and residue analysis may establish what materials (i.e. stone or metal blades) produced the cuts and whether the clusters contain residues of plant, animal, or any other material. 12. Irregular incisions This category brings together those incisions that do not appear to form discrete sets (e.g. the parallel incisions I interpret as tattoo marks and the clusters of parallel and overlapping incisions). It is not, however, a rag-bag of unclassifiable markings. Apparently randomly arranged cut marks of varying lengths often occur on the rock surfaces, frequently underneath other types of markings and images. As with all the other forms of rock marking, irregular incisions probably relate to accessing the supernatural potency of the rocks—people may have made them in the context of ritual activities; they seem to be primarily gestural. AUTHORSHIP, CHANGE AND TRADITION

Up until now, I have concentrated on explaining the rock markings on Driekuil Hill as though they were exclusively Khoe-San in character and origin. This emphasis is intentional—one may readily understand the rock markings in terms of Khoe-San cosmology and ritual practices. I have deliberately blurred distinctions between the



147

many groups subsumed under the collective term Khoe-San. The difference between ‘hunters’ and ‘herders’ may be exaggerated (Barnard 1992: 15, 17) and the cosmologies and customs of Khoe speakers, particularly those concerning female initiation, seem to transcend this ‘divide’. Nonetheless, the motifs of aprons and other objects are strikingly different to engravings of anthro- and zoomorphs that typify hunter-gatherer sites in certain parts of the country. These distinctive motifs of clothing and other objects are moreover widely distributed—one can identify paintings and engravings of similar motifs at quite disparate sites, most of which are located close to permanent water and occur especially alongside major rivers in the arid interior (Dowson et al. 1992; Fock 1969: 103, 126; Hollmann & Hykkerud 2004; Morris 1988: 114; Smith & Ouzman 2004). Driekuil Hill is perhaps best understood in terms of ‘localism’—it is the focus of particular groupings of people with similar broad beliefs and motifs, but with distinct styles and techniques from other sites. Smith and Ouzman (2004) argue that these rough pecked geometrics are associated with Khoekhoen movements within the last 2 000 years. The Koranna, a Khoekhoen group that inhabited this part of the country in historical times (Cope 1977; Tobias 1955), may have used the G-D sites for ceremonial purposes—suggestively, ‘wonderstone’ was formerly known as ‘Korannastone’ (Nel et al. 1937). This association of the wonderstone outcrops with the Koranna, in addition to the association of the water sources with these people mentioned earlier, may be an indication that specific groups of Koranna people (however membership of this very large and variable group is defined) were the authors of the rock art. The G-D sites, though, may not have been the exclusive province of the Khoe-San. As Ouzman (1995: 56) suggests: All rock art produced within the last 2 000 years should be regarded as potentially reflecting and moulding elements of different belief systems: forager, farmer, pastoralist and so forth, however differential the expression may be.

Certainly, beliefs about a giant Water Snake are common to Tswana and Khoe-San and it is very likely that any groups moving through this area in the last 2 000 years would be aware of the G-D sites and their significance. Ouzman (1995: 60) has argued for ‘a remarkable formal correspondence between the rain-making beliefs of the San of the central interior and those of Tswana-speakers’. According to Ouzman, these similarities include the categorisation of the rain as an animal, the ‘hunting’ of the rain, the manufacture of rain medicine kept in horns, and the use of Khoe-San engraving sites by Tswana speakers. This utilisation of erstwhile Khoe-San sites, perhaps for rainmaking, may not necessarily have involved rock marking, although Morris (1990: 136) suggests that rock markings at sites in the Kuruman Valley and Ghaap Plateau, Northern Cape Province, reflect a succession of frontiers where hunter-gatherers and herders, Iron Age agropastoralists, and the sundry individuals and groups who were the vanguard of colonial penetration, have variously jostled and mingled over at least the past 2 000 years.

While Morris (1990) acknowledges the difficulties in attributing specific images to certain ethnic groups, he mentions snakes and meandering lines as imagery that could have been made by Tswana people. Writing about images at the nearby Thaba Sione complex of sites, Ouzman (1995: 56) notes that Tswana people may have been responsible for an engraving of a saurian motif. Images of crocodiles and lizard-like animals are symbols of fertility and rainmaking amongst Bantu-speakers (Prins & Hall 1994). They are drawn as if seen from above. Amongst the Northern Sotho, this motif is


148


called koma—a word meaning something like ‘initiation’ (Eastwood & Eastwood 2006). The koma plays an important part in Northern Sotho boys’ initiation rites and a variant motif, koma-ya-basadi (women’s koma) is prominent in the rock paintings at Northern Sotho girls’ initiation ceremonies (Eastwood & Eastwood 2006; Namono & Eastwood 2005). There is a single example of such a lizard-like motif on Driekuil Hill. Unlike all the other motifs, this image and a nearby motif of a horse and rider are not engraved, but drawn on the rock in a shiny, grey pigment. Fieldwork planned for the other G-D sites may reveal the presence of more motifs and other markings that could indicate a Tswana presence. I have argued that referential and gestural rock markings on Driekuil Hill relate mainly to Khoe-San girls’ puberty rites. This emphasis fits well with findings in Limpopo Province (Eastwood 2005, 2006), where rock paintings express concepts, beliefs and practices linked to Khoe initiation practices. Unlike the rock art of the south-eastern mountains where imagery apparently relates directly to the practices of shamans (e.g. Lewis-Williams 1998), people elsewhere had different ritual emphases (Eastwood 2006). Certain ‘principles’ such as the tiered cosmos and the significance of marking the rock appear to be ‘pan–Khoe-San’ while other practices and conventions—the focus on rites of passage and the prominence of gestural markings at Driekuil Hill—may be local. Further research on the ‘cutting edge’ may help to clarify the nature and extent of KhoeSan and Tswana rock art traditions. ACKNOWLEDGEMENTS

I thank Pieter de Jager of ‘Driekuil’ for his hospitality and for sharing his knowledge about the G-D study area. Riaan Rifkin helped with fieldwork. Wayne and Laura Glenny mapped the site. I thank Mr Modise of Mareetsane, North West Province, for providing information about the name ‘Tlogo Pitsane’; thanks to Erika and Otto Insam for filming the interview with Mr Modise. Several delegates at the Southern African Conference on Rock Art (SACRA) held in Kimberley, South Africa, in 2006, suggested ideas and references about gestural art—Joakim Goldhahn, Antti Lahelma, Jannie Loubser, Sven Ouzman and Andrzej Rozwadowski. I am grateful to them for their help in improving this paper. I also thank the organisers (especially Airi Yamawaki) and sponsors of SACRA for an exciting and successful meeting. I thank Benjamin Smith for early discussions about the scope of this paper, Ed Eastwood for acting as a sounding board, and the two reviewers for their comments. REFERENCES ALEXANDER, J. 1984. Early frontiers in southern Africa. In: Hall, M., Avery, G., Avery, D. M., Wilson, M. L. & Humphreys, A. J. B., eds, Frontiers: southern African archaeology today. Oxford: BAR International Series 207, pp. 12–23. ANDERSON, A. A. 1888. Twenty-five years in a wagon in the gold regions of Africa. 2nd edn. London: Chapman & Hall. ANDERSON, G. 1997. Fingers and finelines: paintings and gender identity in the south-western Cape. In: Wadley, L., ed., Our gendered past: archaeological studies of gender in southern Africa. Johannesburg: Witwatersrand University Press, pp. 13–69. ASTRUP, J. & HORN, G. F. J. 1998. Talc and pyrophyllite. In: Wilson, M. G. C. & Anhaeusser, C. R., eds, The mineral resources of South Africa. Handbook 16. Pretoria: Council for Geoscience, pp. 599– 603. BARNARD, A. 1992. Hunters and herders of southern Africa: a comparative ethnography of the Khoisan peoples. Cambridge: Cambridge University Press. BATTISS, W. 1948. The artists of the rocks. Pretoria: The Red Fawn Press.



149

BEAUMONT, P. B. 1990. Wonderwerk Cave. In: Beaumont, P. & Morris, D., eds, Guide to archaeological sites in the Northern Cape. Kimberley: McGregor Museum, pp. 101–34. B LEEK, W. H. I & L LOYD , L. 2005. Lloyd and Bleek Collection. Retrieved May 2006. http:// www.lloydbleekcollection.uct.ac.za/uct/originals/ef/89a38ac8ec68261c3a366085ad 79874eea999bac.jpg. CALLAHAN, K. L. 2000. Pica, geophagy, and rock art: ingestion of rock powder and clay by humans and its implications for the production of some rock art on a global basis. Paper read at the Philadelphia SAA Conference, 8 April 2000. Retrieved May 2006. http://www.geocities.com/athens/acropolis/ 5579/pica.html. COPE, R. L. 1977. The journals of the Rev. T. L. Hodgson: missionary to the Seleka-Rolong and the Griquas 1821–1831. Johannesburg: Witwatersrand University Press. DEACON, J. 1988. The power of a place in understanding Southern San rock engravings. World Archaeology 20: 129–40. ––––––1997. ‘My heart stands in the hill’: rock engravings in the northern Cape. Kronos: Journal of Cape History 24: 18–29. DE JAGER, P. 2001. Die Korannafonteiners: stories oor Ottosdal en sy mense. Ottosdal, North West Province, South Africa: Pieter de Jager. DOWSON, T. A. 1992. Rock engravings of southern Africa. Johannesburg: Witwatersrand University Press. DOWSON, T. A., BLUNDELL, G. & HALL, S. L. 1992. Finger paintings in the Harts River Valley, Northern Cape Province, South Africa. Southern African Field Archaeology 1: 27–32. DUGGIN-CRONIN, A. M. & LESTRADE, G. P. 1929. The Bantu tribes of South Africa. Vol. II, section I. Cambridge: Deighton Bell & Co., Ltd. EASTWOOD, E. B. 2003. A cross-cultural motif in San, Khoekhoe and Northern Sotho rock paintings of the central Limpopo basin, southern Africa. South African Archaeological Bulletin 58: 14–26. ––––––2005. From girls to women: female imagery in the San rock paintings of the Central Limpopo Basin, southern Africa. Before Farming 2005/3: article 2. ––––––2006. Animals behaving like people: San rock paintings of kudu in the central Limpopo basin, southern Africa. South African Archaeological Bulletin 61: 26–39. EASTWOOD, E. B. & EASTWOOD, C. 2006. Capturing the spoor: exploring the rock art of Limpopo. Cape Town: David Philip. EASTWOOD, E. B. & GREEN, D. 2007. Tsodilo: some ideas on authorship. The Digging Stick 24 (2): 9–12. EASTWOOD, E. B. & SMITH, B. W. 2005. Fingerprints of the Khoekhoen: geometric and handprinted rock art in the central Limpopo basin, southern Africa. South African Archaeological Society Goodwin Series 9: 63–76. ENGELBRECHT, J. A. 1936. The Korana: an account of their customs and their history. Cape Town: Maskew Miller. FLOOD, J. 1997. Rock art of the Dreamtime: images of ancient Australia. Sydney: Angus & Robinson. ––––––2004. Linkage between rock art and landscape in Aboriginal Australia. In: Chippindale, C. & Nash, G., eds, Pictures in place: the figured landscapes of rock art. Cambridge: Cambridge University Press, pp. 182–200. FOCK, G. J. 1969. Non-representational rock art in the Northern Cape. Annals of the Cape Provincial Museums (Natural History) 6 (11): 103–36. ––––––1979. Felsbilder in Südafrika. Teil I. Die Gravierungen auf Klipfontein, Kapprovinz. Köln: Böhlau Verlag. FOCK, G. J. & FOCK, D. 1984. Felsbilder in Südafrika. Teil II. Kinderdam und Kalahari. Köln: Böhlau Verlag. ––––––1989. Felsbilder in Südafrika. Teil III. Die Felsbilder im Vaal-Oranje-Becken. Köln: Böhlau Verlag. HEINZ, H. J. 1966. The social organization of the !Kõ Bushmen. MA dissertation, University of South Africa, Pretoria. HEIZER, R. F. 1953. Sacred rain rocks of northern California. Reports of the University of California Archaeological Survey 20: 33–8. HOERNLÉ, W. 1985. The social organization of the Nama and other essays. Carstens, P., ed. Johannesburg: Witwatersrand University Press. HOFF, A. 1990. Die tradisionele wêreldbeskouing van die Khoekhoen. PhD thesis, University of Pretoria. ––––––1995. Puberteitsrite van ’n Khoekhoemeisie. South African Journal of Ethnology 18: 29–41. ––––––1997. The Water Snake of the Khoekhoen and |Xam. South African Archaeological Bulletin 52: 21–37. HOLLMANN, J. C. 2004. Customs and beliefs of the /Xam Bushmen. Johannesburg: Witwatersrand University Press. ––––––2005. The documentation of engravings on Driekuil Hill, farm Driekuil 280 IP, Lichtenburg, NorthWest Province, South Africa: Phase 2 mitigation report commissioned by Wonderstone Limited. Rock Art Research Institute, University of the Witwatersrand, Johannesburg.


150


HOLLMANN, J. C. & HUFFMAN, T. N. 2005. An archaeological assessment of the Driekuil engraving site, Northwest Province: a Phase 1 report prepared for Wonderstone (Pty) Ltd. Rock Art Research Institute, University of the Witwatersrand, Johannesburg. HOLLMANN, J. C. & HYKKERUD, M. K. 2004. Khoekhoen herder paintings in the Karoo: exciting new finds from South Africa. International Newsletter on Rock Art (INORA) 40: 7–15. HOLLMANN, J. C. & SWART, T. N. 1991. Smithfield survey #1: 9 October–23 October. Rock Art Research Institute, University of the Witwatersrand, Johannesburg. HÜBNER, A. 1871. Eingrabungen von Thiergestalten in Schiefer auf ‘gestoppte Fontein’, Farm von van Zyl bei Hartebeestfontein in Trans Vaal. Zeitschrift für Ethnologie 2/3: 51–3. HUMPHREYS, A. J. B. 1988. A prehistoric frontier in the Northern Cape and western Orange Free State: archaeological evidence in interaction and ideological change. Kronos: Journal of Cape History 13: 3–13. ––––––2003/2004. ‘De-!kunging’ the Later Stone Age of the central interior of South Africa. South African Field Archaeology 13/14: 36–41. HUMPHREYS, A. J. B. & THACKERAY, A. I. 1983. Ghaap and Gariep: Later Stone Age studies in the northern Cape. Cape Town: The South African Archaeological Society. INSKEEP, R. R. 1987. Nelson Bay Cave, Cape Province, South Africa: the Holocene levels. Oxford: BAR International Series 357 (i–ii). LEWIS-WILLIAMS, J. D. 1998. Quanto? The issue of ‘many meanings’ in southern African San rock-art research. South African Archaeological Bulletin 53: 86–97. LEWIS-WILLIAMS, J. D. & BLUNDELL, G. 1997. New light on finger-dots in southern African rock art: synesthesia, transformation and technique. South African Journal of Science 93: 51–4. LEWIS-WILLIAMS, J. D. & DOWSON, T. A. 1990. Through the veil: San rock paintings and the rock face. South African Archaeological Bulletin 45: 5–16. LOUBSER, J. H. N. 2005. In small cupules forgotten: rock markings, archaeology, and ethnography in the Deep South. In: Loendorf, L., Chippindale, C. & Whitley, D., eds, Discovering North American rock art. Tucson: University of Arizona Press, pp. 131–60. LUSCHAN, F. VON. 1923. Buschmann-Einritzungen auf Strausseneiern. Zeitschrift für Ethnologie, Berlin 55: 31–40. MARSHALL, L. 1999. Nyae Nyae !Kung beliefs and rites. Cambridge: Peabody Museum of Archaeology and Ethnology, Harvard University. MERRIAM, C. H. 1955. Studies of California Indians. Berkeley: University of California Press. MITCHELL, P. J. 2005. Why hunter-gatherer archaeology matters: a personal perspective on renaissance and renewal in southern African later stone age research. South African Archaeological Bulletin 60: 64–71. MITCHELL, P. J. & PLUG, I. 1997. Ritual mutilation in southern Africa. In: Wadley, L., ed., Our gendered past: archaeological studies of gender in southern Africa. Johannesburg: Witwatersrand University Press, pp. 135–66. MORRIS, D. 1988. Engraved in place and time: a review of variability in the rock art of the Northern Cape and Karoo. South African Archaeological Bulletin 43: 109–21. ––––––1990. Water-snakes and wagons: rock engravings of the Ghaap Plateau and Kuruman River Valley. In: Beaumont, P. & Morris, D., eds, Guide to archaeological sites in the Northern Cape. Kimberley: McGregor Museum, pp. 135–42. ––––––2002. Driekopseiland and ‘the rain’s magic power’: history and landscape in a new interpretation of a Northern Cape rock engraving site. MA dissertation, University of the Western Cape, Cape Town. MOUNTFORD, C. P. 1976. Nomads of the Australian desert. Adelaide: Rigby. NAMONO, C. & EASTWOOD, E. B. 2005. Art, authorship and female issues in a Northern Sotho rock painting site. South African Archaeological Society Goodwin Series 9: 77–85. NEL, L. T., JACOBS, H., ALLEN, J. T. & BOZZOLI, G. R. 1937. Wonderstone. Pretoria: Geological Survey of South Africa, Bulletin 8. OUZMAN, S. 1995. Spiritual and political uses of a rock engraving site and its imagery by San and Tswanaspeakers. South African Archaeological Bulletin 50: 55–67. ––––––1996. Thaba Sione: Place of rhinoceroses and rain-making. African Studies 55: 31–59. ––––––1998. Toward a mindscape of landscape: rock-art as expression of world-understanding. In: Chippindale, C. & Taçon P. S. C., eds, The archaeology of rock-art. Cambridge: Cambridge University Press, pp. 30–41. ––––––2001. Seeing is deceiving: rock art and the non-visual. World Archaeology 33: 237–56. PARKMAN, E. B. 1995. California dreamin’: cupule petroglyph occurrences in the American west. In: Steinbring, J., ed., Rock art studies in the Americas. Oxford: Oxbow Books, pp. 1–12. PRINS, F. & HALL, S. 1994. Expressions of fertility in the rock art of Bantu-speaking agriculturalists. African Archaeological Review 12: 171–203.



151

REICHEL-DOLMATOFF, G. 1978. Drug-induced optical sensations and their relationship to applied art among some Colombian Indians. In: Greenhalgh, M. & Megaw, V., eds, Art in society: studies in style, culture and aesthetics. London: Duckworth, pp. 289–304. SADR, K. & FAUVELLE-AYMAR, F-X. 2006. Ellipsoid grinding hollows on the west coast of South Africa. Southern African Humanities 18 (2): 29–50. SCHAPERA, I. 1930. The Khoisan peoples of South Africa. London: Routledge & Kegan Paul (Ltd). SCHERZ, E. R. 1975. Felsbilder in Südwest-Afrika, Teil II: Die Gravierungen im Nordwesten Südwest-Afrikas. Cologne: Böhlau. SCHULTZE, L. 1928. Zur Kentnis des Körpers der Hottentotten und Buschmänner. Zoologische und Anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Südafrika 5 (3): 147–227. SILBERBAUER, G. B. 1963. Marriage and the girl’s puberty ceremony of the G|wi Bushmen. Africa 33: 12–24. SMITH, B. W. & OUZMAN, S. 2004. Taking stock: identifying Khoekhoen Herder rock art in southern Africa. Current Anthropology 45: 499–526. STEEL, R. 1988. Rock engravings of the Magaliesberg Valley. Johannesburg: Robbie Steel. SUNDSTROM, L. 2004. Storied stone: Indian rock art of the Black Hills country. Norman: University of Oklahoma Press. TAYLOR, M. W. & KEYSER, J. D. 2003. The Columbia Plateau scratched style: a preliminary investigation. American Indian Rock Art 29: 11–20. THACKERAY, F. 2002. Geometric imagery in South African rock art, with particular reference to Blombos and Wonderwerk Caves. Retrieved May 2006. http://www.nfi.org.za/palaeo/blombos.htm. THACKERAY, A. I., THACKERAY, J. F., BEAUMONT, P. B. & VOGEL, J. C. 1981. Dated rock engravings from Wonderwerk Cave, South Africa. Science 214: 64–7. TILLEY, C. & BENNETT, W. 2001. An archaeology of supernatural places: the case of West Penwith. Journal of the Royal Anthropological Institute (NS) 7: 335–62. TOBIAS, P. V. 1955. Taaibosch Koranas of Ottosdal: a contribution to the study of the Old Yellow South Africans. South African Journal of Science 51: 263–9. VALIENTE-NOAILLES, C. 1993. The Kua: Life and soul of the Central Kalahari Bushmen. Netherlands: A. A. Balkema. VAN DER RYST, M., LOMBARD, M. & BIEMOND, W. 2004. Rocks of potency: engravings and cupules from the Dovedale Ward, southern Tuli block, Botswana. South African Archaeological Bulletin 59: 1–11. VAN RIET LOWE, C. 1937. Prehistoric rock engravings in the Vaal River Basin. Transactions of the Royal Society of South Africa 24: 253–61. ––––––1945. Prehistoric rock engravings in the Krugersdorp–Rustenburg area of the Transvaal. South African Journal of Science 41 February: 329–344. VAN VREEDEN, B. F. 1955. Die Waterslang. Tydskrif vir Volkskunde en Volkstaal 12 (1): 1–10. ––––––1959. Die Waterslang. Tydskrif vir Volkskunde en Volkstaal 15 (3): 15–17. WALDMAN, P. L. 1989. Watersnakes and women: a study of ritual and ethnicity in Griquatown. Honours thesis, University of the Witwatersrand, Johannesburg. WILMAN, M. 1933. The rock-engravings of Griqualand West and Bechuanaland, South Africa. Cambridge: Deighton Bell & Co., Ltd. WIESSNER, P. 1984. Reconsidering the behavioural basis for style: a case study among the Kalahari San. Journal of Anthropological Archaeology 3: 190–234. WYLIE, A. 2002. Thinking from things: essays in the philosophy of archaeology. Berkeley: University of California Press. ZELIZKO, J. V. 1925. Felsgravierungen der Südafrikanischen Buschmänner. Leipzig: F. A. Brockhaus.

v


152

