all photos by the author, except where otherwise noted
Aporia: … the nonpassage, or rather … the experience of the nonpassage, the experience of what happens [se passe] and is fascinating [passionne] in this nonpassage (Derrida 1993:309).
A porias—from the Greek a + poros (“without” + “passage”)—are defined by Jacques Derrida (1993) as philosophical expressions of doubt and confusion; they are irresolvable logical disjunctions in a given line of enquiry. Stuart Murray (2009:11) describes aporias as being constituted by “an intrinsic undecidability … a contradiction, a puzzle or a paradox.” In this article, I argue that academic interpretation of San rock art has been marked by certain aporias—moments of inherent, irresolvable confusion or doubt—because of the methodological tendencies that contemporary researchers have employed. By viewing the interpretation of San paintings as a detective-like “deciphering”—that is, as a journey that must end at the “arrival” of a static meaning— researchers have created pockets of contradictions: San rock art paintings are part of a process of perceptual meaning-making through aesthetic expression, not merely clues for deciphering a simplified “bygone” past. In this view, I advance a phenomenological approach to interpreting these paintings. As Firnhaber explains, rock art is accessed and experienced with the entire body:
In order to examine rock art, we must travel into the land; we must walk, hike, climb, and move about. Rock art is not just a visual medium … (Firnhaber 2007:1).
Methodologically speaking, phenomenology—which takes the body as the “first coordinates” (Merleau-Ponty 1962:115)—allows contemporary researchers to work with the aporias of interpreting San rock art, as opposed to being confounded by them. Hans Penner contends that “to experience is to interpret” (2000:57); I argue that, by shifting the goal of analysis away from determining fixed meanings of individual painted elements and towards an understanding of the experiential capacity and work of those elements as they operated and operate holistically with their environments and their various audiences, academic rock art interpretation can move closer to understanding the work and purpose of these paintings. Phenomenology highlights the perspective of the subject being studied (Wilson and Washington 2007:63); it can therefore facilitate a more nuanced understanding of art works made in a context that is today wholly inaccessible.
The starting point for my analysis was the so-called motif of the elephant therianthrope—exemplified by the Elephant Man, from eBusingatha in the Drakensberg (Fig. 1).1 Therianthropes are understood as figures that have both human and animal features (cf. Jolly 2002); thus elephant therianthropes are figures that have both elephant and human parts, grafted to form a unified body. Therianthropes generally have been given an iconic status in academic interpretations of San rock art, a status that separates them from their contexts, both physical and conceptual. For this study, I chose two sites with elephant therianthropes in the Cederberg region in the Western Cape, South Africa. Both sites are located on the western border of the Cederberg mountains, at the very edge of the mountain range, the beginning of its incline. Monte Christo is 10 kilometers northeast of the town of Porterville, and Groothexrivier is 84 kilometers north of Monte Christo (Fig. 2). Both sites are also located in close proximity to water.2
My choice of sites was informed by a number of factors: First, in order to enact a phenomenological approach and explore the notion of embodiment, it was crucial that I visited the sites in person. This requirement therefore confined my choice to sites that had been interpreted in the literature previously, but would still be accessible to me in person. Two of the sites discussed in Tim Maggs and Judith Sealy's article “Elephants in Boxes” (1983) matched those criteria: Monte Christo and Groothexrivier. I spent ten days, on and off, in the Cederberg mountains in early July 2015. Before embarking on my trip to the Western Cape, I had no indication of the details of either of the sites—their topography, proximity to landmarks or water, inter alia—nor did I know the size or positioning of the individual paintings within the landscape. This perplexing lack of information, as I conducted preliminary research, first illuminated the aporias embedded in the methodologies of recording and interpreting rock art.
DESCRIPTION AS INTERPRETATION
Ezra Chitando (2005) proposes that a phenomenological approach benefits researchers who are attempting to interpret knowledge systems and ideas that are outside of their own cultural backgrounds. Drawing from religious phenomenologists such as James Cox, he offers that phenomenology enables a “non-judgmental analysis” that refrains from evaluating because it does not seek an absolute truth (Chitando 2005:300). He advances full and accurate description as one of the keys to enabling this non evaluative method (Chitando 2005:305). While the scope of this paper limits my application of this approach to only a few examples from each site, I do not assert this study as a comprehensive, definitive reconstruction of the meanings of these paintings; the nature of phenomenological analysis precludes such fixed, static interpretations. My aim is to demonstrate the efficacy of such an approach as a means through which San rock art may be interpreted as “continuous visual fields whose various parts cannot be isolated without changing the meaning of the whole” (Wintjes 2012:179); as part of a dynamic sociopolitical, cultural, personal, inter alia constellation and not merely as demonstrative ethnographic codes. Nonetheless, by providing as full a description of both sites as I am able, and as this medium allows, I am attempting to dehierarchize their various elements—painted and otherwise. This approach resists the empiricist tendency to evaluate what aspects are of more or less relative import, as will be discussed in the fifth section.
The painted panel at Monte Christo (Fig. 3) is located inside a jagged, rocky enclave—elevated, and approximately 25 meters away from a waterfall and stream on the western base of the mountain. Locating the panel on the farm involves an arduous trek through a tangled mass of fynbos3 and trees, following the stream, which has carved a pointed indent into the mountainside (Fig. 4). The setting of this panel is secluded and slightly hidden: The paintings are not easily visible from a great distance. The enclave itself is narrow—approximately 3.5 meters deep and no more than 1.5 meters wide at any point; it is a space in which only two or three people could comfortably fit. Uneven, protruding rocks form a “floor”—with foot-holds on which people can stand, lean, or crouch—but movement inside the enclave is severely restricted by a lack of space.
The panel itself comprises myriad images of varying sizes, in various shades of reddish ochre, superimposed upon one another. The panel measures approximately 2.5 meters in length and 3.5 meters in width from left to right; this measurement encompasses every painted surface on the rock panel. A prominent subject recurs across the width of the panel: Groups of elephants varying in size—with respect both to sizes of figures and numbers of figures within groups—are enclosed and encircled by lines of differing breadth, saturation, and design. In the top right corner of the panel, the area furthest inside the enclosure, a group of twelve or thirteen elephants—some large and others smaller, possibly adults and children—has been painted in a dark, almost maroon shade. For purposes of later analysis, I name this group Elephants A. The elephants face both left and right and, as a group, they are surrounded by two thick, dome-shaped lines (Fig. 5). Beneath and to the right of Elephants A, a variety of feline and human figures have been painted in the same dark maroon hue. Among these is a row of faded human figures, all facing left, which I name Procession A (Fig. 6).
Below this composition and to the left—towards the middle of the panel—a pair of right-facing elephants has been painted within lines, similar to the composition of Elephants A; I name this Elephants B (Fig. 7).4 The lines that encircle these elephants are distinguished on the panel: They encircle the elephants very tightly and are far more numerous than elsewhere on the panel; they are also more zig-zagged and more boldly painted. To the bottom right of Elephants B, as with Elephants A, a procession of left-facing figures has been painted: Procession B.5 Although the paint is smudged in parts, the figures still have visibly extended noses—like elephant trunks—as well as obvious protrusions bulging from their groins—likely indicating penises (Fig. 8).
To the left and slightly below Procession B is Elephants C (Fig. 9). This is an image that has been isolated from the panel at Monte Christo—in the manner of academic conventions for studying rock art that will be discussed briefly below, its “elephants in boxes” (Maggs and Sealy 1983) have come to represent the entire painted surface in most of the literature about this site. Elephants C is large and centrally located on the panel; it is also the most visually and perceptually complex of all the groups, and it is the most striking—whether because of age or paint composition. The composition follows the same theme as the two sets of elephants previously mentioned: a group of elephants enclosed in different styles of lines, with processions and other figures painted in its near vicinity.
The dominant figures in Elephants C are twelve elephants that have been grouped by the various lines. These elephants range in size from very large to very small—the most pronounced variation of elephant size amongst the groups on the panel (Fig. 10). Ten of these elephants have been clustered together and enclosed as a group by the lines; the remaining two elephants—corresponding with the middle size of the group, likely adolescents or young adults—have been painted to the right of the group of ten. These elephants face right and are simultaneously connected to and separated from the larger group by continuations of the joining lines. The separation of these two elephants, still connected by lines, is a distinguishing element of Elephants C.
Another unique feature of the composition is the inclusion of two small elephant therianthropes—each no more than 5 cm tall. They have been painted “inside” the group of ten—towards the top of the composition—between one of the elephants and the largest zigzag line. Below and a small distance from these therianthropes, a third figure of a similar size and style has been painted in the same hue of paint. Erosion of the image renders it difficult to discern whether it is or ever was a therianthrope; however, there is a noticeable separation of this figure from the two therianthropes.6 The therianthropes in Elephants C seem to carry objects on their backs or to be covered by karosses, and they have round heads with short, squat “trunks”—which are not distinctly elephant-like—and round torsos. They have not been painted in profile, as most figures tend to be: Instead they appear to be walking towards the viewer, their anthropomorphic faces turned at an angle. As is the case for the therianthropes in Procession B, these figures do not appear to have any other elephant features: It is only their elongated noses that suggest their hybridity.
Many small figures surround this composition—painted in the same lighter hue of red as the small therianthropes and third, possibly human figure inside Elephants C (Fig. 11). These are scattered mostly below the composition and include small feline figures, human figures of various shapes and sizes—in groupings or processions; elephants and distorted animals; and finally marks, smudges, lines, and patches—one of which is larger than my hand. Another short procession of left-facing human/therianthrope figures has been painted beneath Elephants C (Fig. 12). In this procession, which I name Procession C, the largest figure in the group is depicted with one of the smallest, least-developed therianthrope trunks seen on the panel, and appears to carry a bundle on its back that has a clearly human head.7 The figure immediately in front of the therianthrope in Procession C is without a visible trunk or other anthropomorphic features and therefore appears to be human. This figure holds a very small human figure in front of it, from an extended arm. To the right of these figures, another figure “floats”—located on a different horizontal plane to the others—and faces in the opposite direction (to the right). This figure, which appears to be human, carries a comparatively large bundle on its back, although the bundle has no head. Moreover, pointed bulges, which may reasonably be read as breasts (Parkington 2013), protrude from the figures torso.
Moving to the furthest top left corner of the enclave, its outermost part, a uniquely isolated, carefully detailed—and brilliantly preserved—elephant therianthrope has been painted (Fig. 13). This therianthrope, painted in profile in the same bright, deep-red hue as the elephants and lines in Elephants C, faces left out of the enclave. Like the other therianthropes on this panel it measures only 6 cm tall and is covered by a kaross, or carries an object on its back. There are, however, a number of features that distinguish this therianthrope from the others. One of these is its considerable distance, and therefore isolation, from any other image on the panel; another is a unique depiction amongst the therianthropes of a clear, exaggerated penis along with a clearly defined, comparatively long elephant “trunk”: the most fully elongated among the therianthropes on the panel. The exact detail with which this therianthrope has been rendered—it is the most meticulous on the panel—distinguishes it further.
As mentioned above, the therianthropes with elongated noses do not appear to have been depicted with any other elephant-like features—such as elephant ears, hands, feet, or tusks—although these may have eroded over time. This suggests that these apparently anthropomorphic figures may not be elephant-human hybrids. My reading of them as such is supported by three factors—the prominence of elephant figures on the panel, the similarity between the hybrid figures’ elongated noses and an elephant's trunk, and previous interpretations in the literature—which identify figures on the panel with this feature as elephant therianthropes (Maggs and Sealy 1983). Thus, if all figures at this site with elongated noses are to be read as therianthropes, then the variations in their depictions in other respects is notable. To begin with, all therianthropes share unique spatial relationships to the other painted figures, marks, smudges, and lines on the panel. While all therianthropes carry bundles of some kind, only some bundles have distinct human heads. In addition, the only elephant group composition on the panel in which therianthropes have been included also includes, uniquely, the depiction of a separation of two elephants from the group and the distinctive zigzag lines. The therianthropes with visible penises have indistinctly drawn features, irrespective of the effects of weathering. Finally, the isolated, detailed therianthrope in the furthest corner of the panel has the most fully developed feature that associates these figures with elephant imagery—the trunk—as well as an exaggerated, clearly defined penis (Fig. 14). It is also at a remove from human, elephant, feline, and other figures, as well as from the numerous lines and smudges on the panel. No other therianthrope on the panel has been depicted with such clearly defined features, and no other therianthrope on the panel grafts both the qualities of “elephantness”—the trunk—with those of maleness—the penis. These elements will be discussed in the analysis that follows.
The painted panels at this site face out on to the expanses of the Cederberg valley (Fig. 15). Like those at Monte Christo, these paintings are located within 50 meters of, and elevated above, flowing water: the Olifants River. However, in contrast to Monte Christo, the panel with elephant therianthropes at the Groothexrivier site is in an expansive, open area of land that is easily accessible without much arduous climbing or hiking. An unusual geological feature in the shape of an arch makes the site readily distinguishable in the landscape (Fig. 16); the site is therefore visible from a considerable distance, but the paintings are not. Sheltered by trees and other rocky outcrops, they are somewhat protected: the most apparent figures become visible from a distance of about 25 meters. The painted images at this site, including the elephant therianthropes, flank the arch feature. The lush grass and comfortable, secure space immediately in front of the panel provide an area seemingly large enough for numerous people to gather, whether sitting or standing.
Located on a smoothly undulating rock surface approximately 4 meters to the right of the alcove—from a point of view facing it with one's back to the Olifants River—is a procession of five elephant therianthropes that have been painted in light red (Fig. 17). The figures are approximately 35 cm high—many times larger than those at Monte Christo—and the procession covers a width of around 75 cm; the composition is at an elevation of approximately 1.2 meters from the ground. These figures are the only ones clearly visible on this particular rock surface; however, more faded paintings, which I discuss below, can be found elsewhere on the site.
Most noticeable about these therianthropes, particularly when contrasted with those at Monte Christo, is their considerable size and detail (Fig. 18). These large therianthropes have highly realistic elephant features: Their trunks are long and curved, rendered with delicate detail and fine, tipped ends, and they have been painted with clear, flat elephant feet. They have small elephant-like ears, and their hunched postures evoke the slow, swooping movement of an elephant. Contrasting with Monte Christo, these stylistic elements combine to communicate a likeness to elephants in these therianthropes that is beyond mere elongated noses. Further, no other figures or marks appear to have been painted—concurrently, or as a superimposition—on the rock surfaces surrounding these therianthropes. This pictorial isolation enhances the prominence accorded to these figures by their large size and intricate detail.
The elephant therianthropes at Groothexrivier resemble their counterparts at Monte Christo in that they also carry bundles on their backs.8 A further similarity is the range of sizes and processual depiction of the Groothexrivier therianthropes, yet at Groothexrivier this has been expressed differently: The therianthropes—isolated from other paintings—all face right, away from the alcove, descending in size from very large to very small. All these figures appear to have equally “developed” elephant hybridities, with identical bundles and other features. Thus, contrasting with the numerous variations in form and in spatial relationship discussed at Monte Christo, size is the only differentiating characteristic of the Groothexrivier therianthropes.
The painted surfaces at Groothexrivier are not as well protected from weathering elements as the site at Monte Christo and the paintings are therefore more faded, some of them barely visible. John Parkington (2003, 2013) has reproduced the panel seen in Fig. 19 in a number of publications. These tracings were made from the area in Fig. 20, which is to the right of the rock surface with the elephant therianthrope procession. Although the images that appear in various reproductions of the panel are no longer visible to the naked eye, reproductions depict a procession of male figures—most with clear penises—entering what Parkington (2013:9) suggests may be a maturity or death rite.
On the opposite side of the alcove—to its left, from a point of view facing away from the river—there is another collection of paintings (Fig. 21). These paintings depict human figures9 performing what appear to be domestic tasks: They face each other and lean over an object between them, in the way a person might, for example, lean over a grinding stone. Despite being more faded than the therianthropes, these are markedly more visible than the paintings reproduced in Fig. 20. They are at a similar elevation from the ground to the therianthropes and at a similar proximity to the alcove.10 For purposes of this study, it is notable that none of the images still visible at this site depict elephants. Instead, they appear to depict the demarcated tasks of male and female life: domestic rituals performed by women and what may be a transformative or maturation ritual performed by men.
EMPIRICAL INTERPRETATIONS AND THEIR APORIAS
As mentioned above, robust description combats evaluative research. Of the interpretations of elephant therianthropes established in the literature, Maggs and Sealy 1983 is the most comprehensive study of the Monte Christo site to date, and it includes a brief discussion of the elephant therianthropes at Groothexrivier. The article thus serves as a suitable beginning point for my own analysis, as well as being an example of the methodological tendencies that have informed previous interpretations at these sites and more generally. In summary, Maggs and Sealy (1983) posit that the elephant therianthropes at both Monte Christo and Groothexrivier are representations of the San shaman at the peak of deep trance and further interpret a selection of images that surround them as being in support of what is known colloquially as the “trance hypothesis.” The article does not, however, contextualize any of the paintings it discusses within their respective sites: The journey to the site, its topography, and the positions of painted images are excluded from description or analysis.11 Instead, analysis follows what Justine Wintjes (2012:8) terms “figural iconography”: a tendency to isolate figural motifs and interpret them as culturally informed symbols. Such iconographical analysis necessitates another tendency that informed Maggs and Sealy's study, namely the reconstruction of a “San context” which frames a symbolic, or semiotic, reading of the images. In the following section I will discuss the impact of these tendencies on constructing “an” interpretation of elephant therianthropes at these sites, arguing that their application of fixity and their failure to account for myriad crucial factors is what constitutes the aporias of their interpretations.
In their discussion of Monte Christo, Maggs and Sealy (1983) isolate the elephants enclosed in lines as the primary objects of analysis. Included in their analysis is a discussion of two of the therianthropes on the panel (belonging to Elephants B [Fig. 8]); these are, however, the only hybrid figures they identify in their textual analysis—and these are arguably the most questionably anthropomorphic figures of all those painted. Other painted elements on the same panel, such as the feline creatures, all the other therianthropes, the processions of humans, inter alia, are not mentioned. The smudges, marks, and lines (not related to the elephants) have similarly been excluded; moreover, Maggs and Sealy flatly state that, aside from those they mention, “there are no other paintings in this shelter” (1983:46). Disproportionately foregrounded and isolated as particularly iconographically relevant are the zigzag lines and their relationship to the therianthropes—which are read as symbols of trance potency. This conclusion draws on neurological research (Siegel 1977) that suggests not only that geometric and zigzag lines are a prevalent, if not universal, visual experience reported amongst individuals in trance states—drug-induced or “states of excitation and arousal of the central nervous system” (Maggs and Sealy 1983:47)—but also that, at the most intense moments of this altered state, individuals report experiencing a unity with their hallucinated imagery. The lines are therefore read as representations of universal entoptic sensation, brought about by trance hallucination, “caused by some intrinsic property of the human mind” (Maggs and Sealy 1983:47).
It is in this context that the other figures and painted marks on the panel—especially those interpreted by them as elephant therianthropes—are analyzed. If a person at the peak of trance-hallucination experiences both visualizations of zigzag lines and a feeling of communion with a hallucinated subject, Maggs and Sealy posit that the elephant therianthrope is a direct representation of “the San trancer” in the “peak” of a drug- and/or trance-induced hallucination (Maggs and Sealy 1983:48), in which the “medicine man”12 is unified with “the symbol of elephant-power” (Maggs and Sealy 1983:47). They argue that the artist/shaman “sees” himself in the moment of “becoming” the elephant—whose power he is harnessing in trance—and in that state, or in recalling that state, he records this experience and figures it as an elephant therianthrope. This interpretation fixes “the” meaning of the therianthrope as a “San ritual specialist” (Blundell 2004:141), by drawing hermeneutically from different sources and conceptual frameworks such as ethnography, neuropsychology, and mythology.
One of the limitations of this approach is the application of a fixed meaning—determined through the detective-like associations reviewed above—to all similar motifs across all times and places (Blundell 2004). The aporias brought about by this approach are illustrated through the example of Maggs and Sealy's analysis: They apply the same “shaman” interpretation of the elephant therianthropes at Monte Christo to those at Groothexrivier, with only the hybrid elephant-human figure as “evidence.” This is despite the absence of both visible elephants and zigzag lines at the latter site—the very empirical data on which they argued for a trance hypothesis at Monte Christo. This analysis thus not only fails to account for many of the other painted forms that could be read as figures or signs at both sites—including the various therianthropes and the animal and human figures—it also limits the analysis of the therianthropes at Groothexrivier to a mere supportive demonstration of the reading of the elephant therianthrope in the Cederberg as a shaman. In this formulation, figural iconography has occluded the close, individual study of paintings, fixing one meaning to isolated elements and reproducing that meaning regardless of the other elements on the given panel.
What further accounts for the aporias of this empiricist approach is its failure to account for painted elements that fall “outside” of figural, iconographic forms. The tendency to privilege “subject matter over form” (Wintjes 2012:17) and render painted images as “self-sufficient wholes” (Davis 1993:271) results in a conceptualization of San rock art as operating within a rigid system of signs. Visual semiotics—a paradigm imported from Western thought (see Bal 1996)—requires a minimum significant unit or “fundamental atom” (Bal and Bryson 1991:194), which means that it cannot account for subsemiotic elements.13 Semiotic analysis tends to focus primarily on “large-scale forms … or large-scale properties held by groups of marks” (Elkins 1995:822). James Elkins argues that semiotics is “closely wedded to the distinction between meaningless marks and meaningful signs” and tends to “gloss over marks in favor of the scenes they compose” (1995:822). In Maggs and Sealy's (1983) interpretations, elements that would be classified as subsemiotic have been wholly excluded from both description and analysis: The authors have wholly adopted a semiotic framework for interpretation and have thus failed to analyze prominent aspects of these sites.
Marks blur and fade into one another, and even the freshest drawing will have uncertain moments where the texture of the … [surface] confounds the sense of the mark, or a group of marks converge into a dark confusion, or a mark moves so lightly across [the surface] that it is not securely visible. No image is composed in any other way (Elkins 1995:834).
In San rock art interpretation especially, fixing what is or is not a significant and discrete sign is therefore not as simple as the process of figural isolation suggests. Acknowledging “‘only’ the figure or the represented thing” (Elkins 1995:834) reinforces the idea of a gap between marks and signs. Georges Didi-Huberman (1989:136) analyzes the semiotic preoccupation with the “detail” of painting as it is used to bolster positivist claims for readings of art. In line with the quote above, Didi-Huberman's (1989) study of an art historical fixation with “detail” demonstrates that tendencies to isolate iconographic and/or semiotic elements in a painting—dissecting that painting into a sum of constituent and totalizing parts—discount the impact of nonsemiotic marks and “make it possible to ignore the effects of uttering, of enunciation” and of the material effects of painting (Didi-Huberman 1989:137). Pertinently to rock art interpretation, he argues that “storia”—the indications of narrativity in a painting (1989:140)—are distinguished only through their placement with other elements on the painted surface; if storia are viewed “in detail”—that is, in close proximity and via a method of division and subsequent summation—they become mere marks: “semiotically unstable entities” (1989:141). It is therefore the viewer's perception that connects these elements to one another, forming a coherent image or representation and enabling a narrative or emotional interpretation. The “work” of meaning-making, of sign and/or mark interpretation, is done not by the painting, but by the viewer.
If this is indeed the case, then a phenomenological approach— which foregrounds personalized, embodied experience—appears most appropriate for understanding these paintings. Applying a Western paradigm for reading images—namely, semiotics— creates an aporia in interpretation, because semiotics cannot account for so many of the elements that comprise a rock art panel. There is, however, one further area in which this empiricist tendency has resulted in an aporia: A semiotic reading of these paintings necessitates the reconstruction of the “socio religious or mythico-ritual contexts” (Nettleton 1985:50) in which such “signs” may have been produced and circulated—in short, a San sign-system that fixes signifiers to signifieds. There are a number of consequences to this situation that are detrimental to an incisive interpretation of these paintings.
First, in a semiotic operation the painting and its context must necessarily be separated, so that the “explanandum,” the painting, now stands as evidence for the explanation, the context—and vice versa (Bal and Bryson 1991:179). The painting is thus rendered as a mere demonstration for a proposed context. The proposed context, in the case of the elephant therianthropes painted at Monte Christo and Groothexrivier, respectively, has until this point largely relied on contextual reconstructions that have been built around secondhand ethnographic material. Excellent scholarship has been undertaken on the validity of this ethnographic material, and the use of San “informants”14: In summary, the spatiotemporal distances between the San painters who made these images and the “informants” who provided information to researchers—as well as the cultural, language, and other differences between “informants” and researchers—render much of this material largely unreliable in terms of accuracy. For example, Solomon illustrates that many early San societies conceptualized time as “markedly spatialized” (1997:8); this influenced how they viewed death. Such drastic conceptual difference was a great hindrance to Western researchers who first undertook San ethnography.
Keith Moxey is among a number of scholars who argue that attempts by art historians to interpret any works of art are limited by their own “cultural equipment” (Moxey 2001:77). He argues that all images, whether symbolic or not, are encoded with various meanings, and that access to these codes is made possible by access to the culture in which those codes are produced and circulate. Further, an individual is only able to access all other images—both those within his or her own cultural context and those outside of it—through their own personalized cultural equipment. Interpretation, therefore, is mediated first and foremost by the interpreter or researcher's own background and experience, “filtered through the peculiar configuration of [one's] subjectivity” (Moxey 2001:76). Therefore, applying theories of San ethnography or mythology to the interpretation of rock art images produces an aporia: Contemporary researchers are without the necessary cultural equipment to decipher the specific codified meanings of San paintings. Further, their own cultural equipment impedes their ability to see clearly the signs of other groups without the imposition of their own coded systems. This is the case not only in visual analysis, for even the assumption that the San lived within dynamics that we have come to know as communities, or even families, is one that denies the possibility that the San had wholly different conceptualizations of social organization. The aporia arises because these concepts are irredeemably out of the reach of contemporary Western researchers.
The empiricist methodology—to isolate, assign the status of a sign, and then interpret within a semiotic operation—is therefore aporetic. The internal contradictions of these approaches are brought about by their inappropriateness for the subjects being studied. San rock paintings operate very differently to the paintings for which such methodologies were developed. In the following section, I will advance a phenomenological approach to San rock art interpretation, not as a means to combat the aporias, but rather to embrace them and use them as fruits for deeper analysis.
ADVANCING A PHENOMENOLOGICAL APPROACH
Phenomenology is a style or manner of thought rather than a set of doctrines, rules, or procedures … (Tilley 2004:1).
A phenomenological approach to San rock art interpretation takes the body as the primary vessel and instrument for meaning. That emphasis is useful to this study on two counts in particular: First, it enables robust analysis and interpretation of the embodied experience of accessing and viewing the elephant therianthropes at each respective site. Second, and more specifically from an art historical perspective, it brings the focus of investigation to the substance of the individual painted images themselves. In the instance of the elephant therianthrope, the body has been used as a site onto which complex ideas and notions have been mapped: Whatever it may be, the concept of the painting has been expressed through a (distorted) representation of the human body in an “in-between” state of becoming.
Merleau-Ponty (1962:112) describes the perceptual experience of the connection between body and consciousness as the “body schema,” which is the operation of the body as a unified experience, where the body is both subject and object. Merleau-Ponty (1962) illustrates that this unification occurs through “bodily purpose,” which is an action of consciousness that joins the various parts of the body into a “whole.”15 What this elucidates for interpretations of elephant therianthropes is that changing the parts of the unified body schema involves a deep immersion in the notions that the nonhuman forms represent. They have been grafted perceptually into the very fabric with which a human enables his or her consciousness in the world. Bryan Turner posits that the mapping of “major moral, personal, and political concerns” of a social group on to the “conduit of the body” (Turner 1996:6) is a means for that group to express and problematize these issues. The combination of elephant and human parts into a seamless elephant therianthrope body schema therefore strongly suggests that certain “qualities” of elephantness that reverberate with San life concerns were being incorporated into the body schema—the unified embodied experience.
At this point, a hermeneutic approach is a useful complement to a phenomenological reading: A greater breadth of understanding of elephant behavior elucidates possibilities for what these qualities of “elephantness” may have been.16 With a specific focus on the elephants depicted at Monte Christo, clearly demarcated into groups, Iain and Oria Douglas-Hamilton (1975) observe that kinship groups most often include thirty to eighty elephants, although some studies have recorded groups as large as two hundred. Phyllis Lee and Cynthia Moss (1986) observe that these kinship groups—closely bonded by enhanced communication— comprise smaller hierarchical family units of about ten elephants each, and that these family units are matriarchal. Adaptations in their feet and trunks for detecting seismic vibrations enable elephants to communicate nonverbally over very long ranges; Cynthia O'Connell-Rodwell (2007) remarks on instances in East Africa where entire elephant kinship groups gathered at a single, previously unknown, water source, all having traveled from hundreds of kilometers away. Further, Douglas-Hamilton and Douglas-Hamilton (1975) explain that elephant cows lactate and feed their calves for long periods, which requires long-term protection and parenting. They argue that female domination of family units—whereby female elephants form tight bonds which emphasize “alloparenting” or aunting—possibly results from this. Young males, once they reach maturity are forcibly excommunicated from their herds, where they “could linger on the edge of the family for perhaps several years” (Douglas-Hamilton and Douglas-Hamilton 1975:164). This is done in order to encourage them to mate away from the herd.
The elephants depicted at Monte Christo could therefore be read as a representation of a kinship group, with smaller family units of ten to twelve clustered into the groups discussed, enclosed by various lines. The size variations may reasonably correspond with age variations.17 The two separated elephants, in this same composition, could further be read as representations of male adolescent exiles, their lingering connection to the family unit reinforced by the extension of lines to encompass them into the group. The zigzag lines in Elephants C have been argued previously to represent entoptic images—representations of trance-induced hallucinations (Maggs and Sealy 1983). However, I argue that they can also be seen as representations of tension when considered phenomenologically Johanna Uher (1994) observes that, in many forms of art, representations of teeth take the form of zigzag lines: The sharpness of a row of teeth closely resembles a zigzag line. This bodily resemblance has enabled what she terms a “surrogate meaning” (Uher 1994:309): The hostility and antagonism of bared teeth is read in the zigzag of the line, and the line thus comes to stand for metaphorical tension. In her words, “… the zigzag is not a symbol with any single referent but rather a form of punctuation that marks events of tension, usually in an antagonistic context” (Uher 1994:307).
Considering the therianthropes in relation to elephant sociability and these lines of tension, two of the compositions on the panel contain both zigzag lines and therianthropes. In Elephants B the lines have been depicted with great intensity: boldly rendered and tightly surrounding the two “adolescent-sized” elephants. The therianthropes closest to Elephants B, in Procession B, have crudely rendered elephant hybridities and penises—the elements of both maleness and therianthropy are implied through marks, scraps, and smudges, as opposed to being meticulously illustrated. In Elephants C, two anthropomorphic figures have been singularly superimposed inside the enclosing lines of “tension,” within the “matriarchal” herd; they have also been painted as comparatively underdeveloped therianthropes—their trunks are especially short and squat in relation to the other hybrid figures on the panel.18 Therefore the depiction of “tension” as read in the lines seems to be connected to the depiction of male and/or elephant therianthropes. The tense event of male exile, as described by Douglas-Hamilton and Douglas-Hamilton (1975), may account for the painted isolation of the mid-sized elephants in both Elephants B and Elephants C from the rest of the elephants on the panel. The lines may therefore not be entoptic images, they may be phenomenological expressions and representations of the emotional or psychological action of male “becoming.”
As mentioned above, these figures do not have any further features which may designate them as elephants—such as the pachyderm-like feet and ears painted on the elephant therianthropes at Groothexrivier—and their small size, combined with their rough-hewn, squat trunks, makes a reading of them as therianthropes highly contingent. What most likely designates these figures as elephant-human hybrids is their placement with and around the prominent elephant figures, their elongated noses— the intimation of a trunk—and the presence of other, more “fully developed” therianthropes on the surrounding surface. At Groothexrivier the grafting of “elephantness” onto the body schema is far more explicit, and there are no visible paintings of elephants at the site. The Groothexrivier therianthropes have been rendered with clearly defined elephant features: They have long, detailed trunks, but also elephant-like ears and pachyderm-like feet. At Monte Christo, it is therefore an active engagement between marks and traces on the painted surface, and it is the viewer's own eye—her gaze—that informs the identity of the figures as elephant therianthropes. Conversely at Groothexrivier, the coherence of transformation has been mapped on to the body schema explicitly—elephant and human parts have been mapped realistically to create a unified body that is an unambiguous hybrid of human and elephant.
The notion of the gaze is central to art history—it is, in Merleau-Ponty's words, “how vision is brought into being” (1962:78). I am attempting, in this study, to understand vision—one of the primary senses through which the paintings are experienced—from a nonevaluative, phenomenological perspective. Merleau-Ponty's phenomenology conceptualizes vision as occurring on two simultaneous fields: the horizon, or wide field, and the concentrated visual field. He demonstrates that one's consciousness looks at objects by mentally or intellectually “inhabiting” them19 (Merleau-Ponty 1962:79). It is this kind of inhabitation that, he offers, splits the act of viewing. One's consciousness cannot inhabit both visual fields simultaneously: In order to focus on an object—to view it in the concentrated field—one must put the horizon into “abeyance,” “because to look at the object is to plunge oneself into it” (Merleau-Ponty 1962:78).
Yet, he continues, the horizon and the concentrated field are in a continuous process of informing each other: The horizon informs the identity of an individual object, and close inspection of an object adds richness to a perspective on the horizon. In this formulation the gaze, which is activated by the same grasping desire as the body schema, is instrumental to the process of meaning-making; the gaze roves over the horizon, informing the nature and degrees of concentrated or generalized viewing much like a “blind man's stick” (Merleau-Ponty 1962:177). Through this, the gaze maps “perceptual habit” (Merleau-Ponty 1962:176) onto the body schema; the mind learns to “see,” guided by the integrated actions of the body schema as the eye and other organs engage in the act of looking. Therefore, the ways in which a body—specifically the eye, as connected to vision—moves over a given visual field, alternating between concentrated and generalized viewing, are intrinsic to the processes of meaning-making. Close looking therefore means cutting off the horizon, but the ways in which that horizon has been accessed before, after, and around that focused looking inform the identity of the object of concentration and the viewer's attitude toward it.
Returning to Monte Christo, and with a specific focus on the elephant therianthropes, the small size of the figures—particularly the human and therianthrope figures and those that surround the elephant group compositions—necessitates an extremely close, focused viewing. Grouped compositions, with significant space between them, are scattered over the panel. Each composition comprises small figures and marks that are packed closely together, sometimes overlapping, requiring corporeal immersion: In order to view the details of these figures, I needed to bring my face within 20 cm of the painted surface. Blundell observes that this necessity ensures that “… a person's range of view is restricted so that eventually the images take up all of one's vision … indeed, one becomes immersed in the images very quickly simply by viewing them” (Blundell 2004:167).
Further, the narrow space of the enclave does not allow for much free movement, nor for many people to be in it at one time. This enhances the immersion with the figures, as both the visual and physical horizons—in terms of maneuverability—are severely limited. The notion of immersion is thus central to my analysis. If the act of looking involves inhabiting an object, the small, delicate elephant therianthropes at Monte Christo require an extensive immersion of the full body schema into the act of viewing, thus facilitating a fuller inhabitation. Moreover, all processions and all therianthropes on the panel at Monte Christo face left, towards the outside of the enclave; even the three-quarter therianthropes in Elephants C are left-facing. The “development” of elephant trunks on the therianthropes—i.e., the extent and delicacy of their elongation—appears to intensify if the viewer moves her gaze in a similarly leftward direction: The figures on the furthest right of the panel are barely visible, but appear to be human; the therianthropes in the center of the panel, including those inside the elephant lines, have short- to medium-length trunks, and the figure with the most fully developed trunk is isolated on the outermost lefthand corner of the enclave. This processual movement—of both the figures on the panel and the gaze as it moves across and over it—perceptually maps the transformative movement of the elephant-human hybridization to the viewer's body schema.
The leftmost therianthrope—at the end of this visual and perceptual procession—is distinguished from the other therianthropes by a number of features. First, its position as the last visible point in the procession is enhanced by its being the most “developed” hybrid of the therianthropes on the panel. It is also spatially separated from them, located towards the top of the panel and away from all other paintings. This produces a severe cutting off of the horizon; no other figure is so singularly visible. Further, this therianthrope has been rendered in the most delicate detail and has been painted with an exaggerated penis. The immersion that is called upon by this therianthrope is arguably the most profound of all those on the panel: Its positioning and small size necessitated that, in order to view this figure's details, I had to balance on the very edge of the jagged footholds, stretching my neck and squinting my eyes to see it clearly. Further, the isolation of this male therianthrope—in its placement and in the act of viewing it—resonates with that of the exiled adolescent-sized elephants in Elephants B and Elephants C. The notion of change and “becoming” further attaches to these paintings when they are read in this way.
At Groothexrivier, although a similar figural motif is depicted, the elephant therianthropes are accessed in a very different way to those at Monte Christo. As above, these figures are relatively large, realistically and clearly defined, and have been painted in a right-facing procession—towards a panel that, although no longer visible, is purported to have once contained a depiction of an exclusively male initiation or death rite (Parkington 2013). The procession faces and “moves” away from a painting depicting women performing domestic tasks, on the opposite side of the dividing rock archway. The therianthropes at Groothexrivier are identical in all respects except for size, with the smallest figure in the procession painted closest to the now-invisible panel. Harald Pager (1971) offers that processions of figures may indicate movement, which is in line with a phenomenological position, reading the body as moving through space, thus changing the viewers perspective. It is also possible that, as above, the size differentiation relates to differing ages or levels of experience.
Despite the range in size, even the smallest therianthrope in this procession is visible, in detail, from a distance of around 10 meters. The distinctive size of these therianthropes, when viewed in terms of Merleau-Ponty's (1962) phenomenological approach, means that they are incorporated into the horizon, or general visual field. These therianthropes are therefore accessible without immersion, and do not require a severing of the horizon. The wide, open area in front of this panel further extends the visual field in which these paintings sit, and the area of clear ground in front of the painted rock surfaces suggests an area in which many people may have gathered and viewed the paintings at once. Blundell writes of a San “informant” in the 1910s who described dances that were held in an area with a similar large floor area, in front of an SDF (Specially Differentiated Figure; see Blundell 2004) at Ncengane Shelter in Nomansland; he recounts her description of participants turning “with hands open toward the paintings to harness the supernatural potency that resided in the images” (Blundell 2004:166).
The isolation of the Groothexrivier therianthropes on the rock surface is another indication of the unobstructed access these paintings offer: They do not rely on surrounding figures or images to identify them; they are unified and coherent on one surface, just as they represent unified and coherent body schemas of human and elephant hybrids. The other paintings at this site, although no longer clearly visible, appear to represent male and female figures on opposite sides of the alcove; the therianthropes have been painted on the right side, next to the panel with male figures. This suggests a delineation of male and female tasks; however, the extent of weathering limits this analysis. Nonetheless, the large size and realistic conflation of forms at Groothexrivier communicates the mapping of human and elephant on to a unified body schema; as opposed to at Monte Christo, where surrounding paintings seem to be more directly implicated in the “storia” of the elephant therianthrope. At Groothexrivier it seems possible, however, that there is a meaningful correlation between the right-facing direction in which the procession of therianthropes—associated with males—moves, and the position of the male-only ritual painted on the adjacent rock surface.
The therianthropes at Groothexrivier are large, clear, and isolated paintings, easily visible from a distance of around 10 meters. In these paintings, the incorporation of “elephantness” into the human body schema is explicit and therefore relatively easily accessed. The therianthropes at Monte Christo, however, are painted in a small, confined enclave, amongst steep, jagged rocks. The figures on the panel are small and require close viewing. At Monte Christo, accessing the mapping of elephant and human into a unified body is not as easy as at Groothexrivier: The figures require immersion, both corporeal and cognitive, and display far more complex interactions between elements than at Groothexrivier. At both sites, a core connection between elephants and San humans is being expressed; however, the enunciation of this connection differs vastly. I argue that by reading the utterances and effects of these enunciations, particularly from the perspective of bodily experience (both of bodies represented and the bodies that view) the aporias of empirically based interpretations cease to be confounding puzzles. Rather, through this mode of analysis contemporary researchers are able to execute more deeply penetrating explorations of how, as opposed to what, these paintings may mean.
Fieldwork research for this article was made possible by funds from the Stein-Lessing Scholarship program at the University of the Witwatersrand. Part of the article was originally submitted as a thesis for degree qualification at Wits and has been further developed as part of a Writing Fellowship at Rhodes University The author would like to extend her gratitude and appreciation to Professor Emeritus Anitra Nettleton for her invaluable guidance and feedback, and for her unwavering support. Deepest thanks are also extended to Dr. Justine Wintjes (Wits), Prof. John Parkington (UCT), Mr. Andrew Patterson (non-academic), and to Prof. Heinz Ruther and his team at the Zamani Project (UCT), who were all instrumental teachers and without whom this paper would not have been possible. Special thanks are also extended to Prof. Ruth Simbao for her assistance in readying this article for publication, and for her generous personal support.
This image (Fig. 1) is visually and perceptually rich and is part of a frieze depicting bees, antelope, and other animals. However, in 1947 the Bureau of Archaeology, a government agency led by Clarence Van Riet Lowe, removed the panel under the auspices of protecting it from vandalism (Wintjes 2012). The removed panel, currently at the Natal Museum in Pietermaritzburg, has become an isolated, iconic figure of the rock art at Cinyati and many of the reproductions and analyses of Elephant Man distill it to a highly abstracted figural icon.
Geoffrey Blundell (2004:172) posits that “in San cosmology, both waterholes and rock art sites are portals to the supernatural world.” This is expanded further below.
A distinctive type of low-growing flora indigenous to the southern tip of Africa.
For this composition, the paint used is a brighter, lighter red than in Elephants A and Procession A. Anitra Nettleton (1985) is among scholars who argue that differences in paint hue, or type, on a single panel plausibly indicate superimposition: That painting is likely to have occurred over more than one session.
These paintings have suffered less erosion than Procession A and are thus more distinct.
These figures have been painted in a lighter, more orange hue than the dark red of the elephants and the lines in Elephants C, suggesting that they may have been superimposed.
In Vinnicombe's study of the Drakensberg, such figures were interpreted as being children, carried on their mothers’ backs (Vinnicombe 1976:246), whereas Blundell suggests that in such instances, particularly when the figures carrying the “human” bundles are therianthropes, the subject matter of paintings refers to the mentoring of young shamans (2004:142).
Similar to the Elephant Man at eBusingatha (Wintjes 2012)—amongst other examples of elephant therianthropes (Blundell 2004)—each bundle carried by a therianthrope has a long, paddle-shaped object extended from its top. At Monte Christo, where bundle extensions are present they depict small human heads.
Vinnicombe (1976) and others suggest that curved backs and pronounced buttocks may be features that, aside from breasts, indicate women.
I have found no references to these images in the literature, and in my extensive conversations with researchers in the field (John Parkington, Heinz Ruther, personal communications, 2015), I did not come across anyone who was aware of them.
All analyses in the article draw from (referring only to) a selection of black-and-white photographs and monochrome tracings of paintings; the omission of color in the pictorial reproductions is wholly unaccounted for in the text. Further, the reproductions show only a limited selection of the paintings themselves and are often zoomed in to a very tight image, which cuts the figures off not only from the landscape but also from the other painted images on the rock surface. This has the further effect of distorting the scale of the images, and dimensions are mentioned neither in the captions of these images nor in the accompanying text.
Ethnographic material asserts that the San shaman was always male; therefore in this account of the therianthrope-as-shaman, the therianthrope is seen to represent a male experience of transcendentalism. See Lewis-Williams and Clottes 1998.
These are elements such as stylistic variation, chiaroscuro, composition, marks, brushstrokes, paint thickness, inter alia (Elkins 1995:823), otherwise known as subatomic elements.
For a critique on the ethnographic studies of the San conducted by Wilhelm Bleek and Lucy Lloyd see, for example, Solomon 1997.
For example, one does not need to inform each body part consciously how it should move in order to retrieve a glass from a table: the choice, derived from desire, to obtain the glass unifies the requisite parts of the body to bring the operation to fruition.
John Parkington (personal communication, 2015) elucidates that by virtue of their reliance on natural resources, San groups in the Cederberg would have had extensive knowledge of the fauna and flora in the area; knowledge of the rhythms of plant and animal systems was essential to survival.
The depiction, for example, of a tiny elephant wedged beneath a bigger one in Elephants C is suggestive of the care of a baby by its mother.
Many factors, including different hands, may account for this; however, other differences, such as the absence or presence of accoutrements, support the view that an effort has been made to distinguish these figures from one another.
In order for something to be, for example, “on” a table from my perspective, I must either take up the position of the table or of the object.