Landscape, Rock Art 

Recording, Narrative:  

A Biography of  

The Harald Pager 

Archive. 

 
Lemishka Moodley 

 

 

 

 

 

 

 

 

 

  

 

A Research Report submitted to the Faculty of Humanities, University of the Witwatersrand, Johannesburg, 
in partial fulfilment of the requirements for a degree of Masters in History of Art. 
Johannesburg, March 2019. 

COVER IMAGE 



1 
 

Declaration 
 

I declare that this Research Report is my own unaided work. It is submitted in partial fulfilment for the degree 

of Masters in History of Art at the University of the Witwatersrand, Johannesburg. It has not been submitted 

before for any other degree or examination at any other university. 

 

 

Lemishka Moodley 

15th day of March 2019  



2 
 

Dedication 
 

 

For my mother, sisters and support system. 

  



3 
 

Acknowledgements 
 

My biggest thank you goes to Dr Justine Wintjes, for her incredible supervision and support. I am so unbelievably 

grateful for her incisive yet insightful comments, generous sharing of knowledge and expertise. Without her 

guidance, tremendous amount of patience and confidence in me, I would not have been able to produce this 

work.  

I also really appreciate:  

All the interlocutors who agreed to have a conversation with me about Harald Pager’s life and work. Their 

contributions added richness and depth to my project. Aron Mazel and Paul Stopforth took the time to have e-

conversations with me. John Wright, Leon (Jakes) Jacobson, Neil Lee and Shirley Woodhouse (the wife of the 

late Bert Woodhouse) engaged in a group conversation about Pager’s life and work. Finally, David Lewis-

Williams graciously agreed to have a discussion with me as well. Without their ‘personal histories’ contribution, 

my report would not have been this rich.  

David Pearce of the Rock Art Research Institute, who facilitated access to the Harald Pager Archive.  

And I thank my family, who supported me throughout this process and “crossed the finish line” with me.  



4 
 

Table of Contents 
 

Declaration 1 

Dedication 2 

Acknowledgements 3 

Preface 5 

Foundations 8 
An archival focus 8 
Theoretical toolkit 9 

Archive 9 
Landscape 12 
Narrative 13 

Methodology 13 
Importance of this research 15 
Chapter structure 16 

CHAPTER 1: The Primary Archive 16 
CHAPTER 2: The Secondary Archive 16 
CHAPTER 3: The Tertiary Archive 16 
APPENDICES 17 

Chapter 1: Primary Archive 18 
uKhahlamba-Drakensberg Mountains 19 
Didima Gorge 22 
Sorcerer’s Rock 24 

Chapter 2: Secondary Archive 26 
History of rock art recording and interpretation in the Drakensberg 27 
The Pager Archive through a focused study of Sorcerer’s Rock 30 
Pager’s recording method 34 
Visual essay 36 

Chapter 3: Tertiary Archive 43 
Story of the physical archive 44 
Reflection of the digital archive 46 
Oral histories and archival reflections 46 
Conversations 48 

Reflective Summary 50 

Appendices 52 
Appendix 1: A creative visual essay 52 

List of Figures 55 

References 56 
 



5 
 

Preface 
 

In the early days of the discipline of archaeology, the archaeologist Sir Charles Thomas Newton stated, in a 

lecture at Oxford University in June 1880, that “the subject-matter of Archaeology is threefold – the Oral, the 

Written and the Monumental” (Newton 1880, p.3–7). By ‘oral’, he meant expressions of spoken language as a 

form of patterned communication passed down from the past, which he considered to be as significant as 

written texts and traces of the built environment. Newton’s statement also resonates with ongoing attempts 

to make sense of the fragmentary remains of the past by including living sources with links to those remains, 

often referred to as ‘oral traditions’. I argue that the domain of ‘oral’ could be extended in the contemporary 

context to refer to the realm of the ‘spoken word’.  The spoken word is mobilised in the telling of personal 

histories of researchers engaged in making sense of the past, consistent with a move towards the study of the 

production of archaeological knowledge, and the broader context of the history of science. The past, and the 

investigation of that past, can be easily lost or erased with time, unless it is documented in some form. 

For as long as I can remember, I’ve had a keen interest in knowing about the world that preceded me. So 

naturally, movies like Raiders of the Lost Ark (1981), The Mummy (1999), Lara Croft: Tomb Raider (2001), 10 000 

BC (2008), The Prince of Egypt (1998), Brother Bear (2003) and many others, fascinated and triggered my 

overactive imagination. These points of inspiration even prompted many crazy shenanigans and adventures 

that took place in my back yard. Needless to say, my mother was never impressed because her garden was in 

constant danger of being destroyed through micro-excavations (as I had to find the treasure I buried the 

previous evening!). However, it was a Pixar animated film — The Croods (2013) — that recently re-ignited my 

interest in deeper human history. Despite its status as a relatively mainstream commercial animated film, The 

Croods sparked my curiosity to find ways to visit the past and bring yesterday’s stories into our lives today.  

 

 

 

FIGURE 1 

Figure 1: Screenshot of Grug drawing on a cave wall. Captured by Lemishka Moodley from The Croods (2013). Animated Film. 
Directed by C. Sanders. U.S.A. Pixar. 



6 
 

After their cave collapses, the first Crood family has to undertake a journey to find a new home. The father, 

provider and leader of his pack, recounts multiple stories that urge and caution his children to follow the rules 

described on the cave wall and to not try anything “new”, or generally be curious for that matter. He sees 

newness as dangerous, with potentially serious repercussions like death. He prefers for them to dwell in the 

“safe” compounds of the cave in which they live. He develops “cautionary” narratives through visual means by 

drawing pictures on the cave walls using pigment and dirt he scoops off the ground. The animated film highlights 

the ways in which a pictorial expression in the form of rock markings could be enlivened in its original context 

of performance and story-telling, and, also, the ways in which history is forged through personal experience.  

The film also prompted me to reconsider the academic realm in which rock art is primarily situated in southern 

Africa, which is archaeology, and the possible perspectives that other disciplines could bring to bear on these 

materials, such as art history. My understanding of the reasons why rock art (and its copies) can and should also 

be considered and studied as an art form, ultimately stems from my own experience in art education. The rock 

art works made by the San people are encoded with “the history and culture of a society” that is “thousands of 

years old” and “a testament of the displaced ancient African culture and the San presence in the world” 

(Solomon 2005). By studying these traces as artworks, researchers, historians and archaeologists are reminded 

to look at the visual features, but also beyond the physical aspects of the work. They begin to consider the 

processes that contributed to its making, and the various interpretations and meanings that the work had in its 

context of production, as well as in its subsequent readings. An examination of the process of knowledge 

production not only draws attention to the development of artmaking and the manner in which different 

materials were used to create artworks, but also demonstrates the precarious nature of the meaning of 

rupestrian imagery. An example of this instability of meaning is evidenced by a body of oral histories relevant 

to the San context, starting with Joseph Orpen’s documentation (1874) of a mountain Bushman called Qing’s 

account of the meaning behind particular rock art panels. Thereafter, in 1911, Wilhelm Bleek and Lucy Lloyd 

produced Specimens of Bushmen Folklore, a book of 87 recorded legends, myths and other traditional stories 

of the |Xam Bushmen in their now-extinct language (Solomon 2005). These resources integrated the “spoken 

word” with Bushman ethnographic research and laid the foundation for how we read and write about the art 

today, and they also provide numerous different ways into the question of interpretation.  

I have utilised my previous first-hand experience as an intern at the Origins Centre, where Simone Opperman 

and myself worked under the guidance of Steven Sack and Lara Mallen. In working towards the exhibition titled, 

The Origins of Walter Battiss: “Another Curious Palimpsest”, we worked closely with the Rock Art Research 

Institute (RARI), and my interest in rock art and its archive grew immensely. Here, I learned of the Harald Pager 

archive, which is physically located at RARI. After speaking to my supervisor, Dr Justine Wintjes, and listening to 

the way in which she briefly recounted the story of Harald Pager, who was an active rock art recorder and 

researcher during the 1960s and 1970s, I wanted to learn more about the man who seemingly died for his craft. 

Wintjes mentioned that Pager recorded San rock art in the Drakensberg area along with his wife for many years. 

The Pagers relied heavily on their personal resources in order to keep the recording and documentation process 

going, and ended up incurring financial debt. Harald Pager was passionate about archaeology and sought to 

understand the rock paintings he discovered in his travels across the region, which also included what was then 

South West Africa (now Namibia). In one of the conversations held in the course of my Masters research, Neil 

Lee explained that Pager dwelled for months on end in a shelter in order to finish the copies of rock paintings 

he sought to record. Pager’s adamance and determination, and even his belief that he could change his 

metabolism, apparently led to ill health while recording rock art in the Brandberg in 1985. He died a short time 

later in Windhoek (Neil Lee 2018, pers. comm.).  



7 
 

Given the scale and meticulousness of Pager’s archive, and in light of the conversations I had about him, it 

seemed strange that his work and story are engaged with so seldom within academia. Nevertheless, Pager’s 

vast archive lives on in the storerooms of RARI, and now also in the digital world. His recordings were recently 

digitised during the course of the South African Rock Art Digital Archive (SARADA) project, which sought to scan 

and make more accessible the content related to southern African rock art at RARI and other institutional and 

private archives. With the rapid rate at which technology is developing and progressing, physical archives are 

being turned into digital databases, while also being supplemented by newly created digital materials. However, 

like all archives, digital archives are not ‘permanent’. They have their own kind of media-specific fragility. The 

digital archive also has the potential to become obsolete and demonstrates a different kind of limited lifespan. 

Although digital files are also not physically present in the same way that a physical copy or photograph is, and 

“lose” a particular kind of materiality and stability, they gain a virtual presence and potentially greater 

accessibility. As with the ‘original’ archive of rock art that exists in the landscape, the traditional paper archive 

and the digital archive represent different kinds of longevity and fragility, and have a complex relationship with 

each other. While being subject to changes and deteriorations of various kinds, the archive is important as a 

record of archaeological materials in the landscape, but is also a site where knowledge is produced, and the 

practice of science is conducted. 

Thus, I set out to devise a project that would address some of these interests and problems, to find connections 

between the present and the past, and to focus on ‘process’ across the rock art archive in the broad sense of 

the term. I also wanted my project to address absences and silences within the archive, embedded within the 

documents and copies as well as the rock art itself. I approached the archive as a layered domain that extends 

beyond the strictly ‘documentary’ archive, and adopted a ‘personal histories’ approach as a way of forging 

connections across that archive. I explain this working strategy in more detail in the ‘Methodology’ section 

below, but first I explore some of the conceptual elements that frame the project. 

 

 

 

 

  



8 
 

Foundations 
 

An archival focus 

 

When I began this project, I explored the sites that Pager recorded at Didima Gorge. Leopard Cave attracted my 

attention because of my own particular interest in the single feline representation among the numerous painted 

figures at the site. Across the regional corpus of San rock painting, felines are a relatively rare motif and have 

received some attention as a topic of a focused study or part of a larger discussion (Pager 1971, 1993, 2000; 

Lewis-Williams 1981a, 1983; Herrera Maldonado 2009). Nevertheless, feline representations are perhaps not 

as well researched as other animals depicted in the art and a deepened study would be beneficial. The feline 

figure thus seemed to potentially open up a viable area of new research. However, a focus on this particular 

motif was not necessarily compatible with my central interest in the making of the archive, or the forging of 

Pager’s own focus areas, and so seemed to pull towards a different kind of study, one with an iconographic and 

interpretive focus on the rock art itself through engagement with the ethnography, which could constitute a 

thesis on its own. Considering the scope of a Masters research report, I decided rather to adopt a biographical 

approach, not towards Pager himself, but towards the archive. I was particularly interested in the making of the 

archive, and the personal reasons that drove Pager to immerse himself so intensely in the solitary process of 

recording rock art through photography, drawing and painting. I initiated conversations with various 

interlocutors, including people who knew or worked with him in various contexts, who provided me with as 

much information as they could possibly remember. Much of what they told me was based on their own 

immediate personal encounters with Pager, and the details of his earlier life, before he became involved in rock 

art recording, remain vague and speculative. All the interlocutors concurred that Pager was an introvert and 

never really shared much about where he grew up and the reasons for his affinity with rock art.  Geoff Blundell 

suggests that in some ways, Pager’s interest in African prehistory may have had some similarity to Leni 

Riefenstahl’s photographic work in Africa after the end of Nazi Germany, in the context of a post-World War II 

search for an Eden “untouched” by corruption of civilisation (pers. comm 2019). I hoped to uncover more of 

this kind of subtle detail surrounding his deeper biography and motivations through the ‘personal histories’ 

approach I adopted. This proved challenging, as there aren’t many people left who knew Pager well.   

Because of the scale of Pager’s archive, I nonetheless found it helpful to introduce a narrower site focus as a 

particular node within the bigger archive, and a practical set of connections to attempt to trace back through 

the layers. Sorcerer’s Rock was not a particular topic of any of the conversations I held with the various 

interlocutors, but it functions as a sub-set within the larger archive through which to explore the biography of 

the archive as a whole. Pager’s archive, like most archives, is a vast repository of documents comprising traces 

of many lived experiences and cannot be unpacked all at once. After considering several of the other Didima 

sites, I chose to focus on Sorcerer’s Rock and its main rock art panel, because of the manageable scale of the 

site, and the apparent ‘narrative’ coherence of the main panel, in order to explore Pager’s approach to 

recording and his contributions to rock art interpretation. By choosing this one site, I could address, in specific 

terms, the thematic threads presented in this paper. Although researchers have previously examined the 

paintings at this shelter, they have primarily focused on the central and most elaborate ‘sorcerer’ figure on the 

panel, and there has not yet been a study of the entire panel or site at Sorcerer’s Rock. While my study of this 

site doesn’t purport to be a fully-fledged interpretation of the site in line with current practices of 

ethnographically based rock art interpretation, my study goes some way to contribute to this site-specific body 

of research, and I discuss the shelter and its rock art archive in more detail later in this report. 



9 
 

I acknowledge that the scope of the study I set out to do was ambitious. I hoped to address previously 

undocumented aspects of Pager’s life and work, while also delving into several well-developed theoretical areas 

of scholarship (landscape, archives and rock art interpretation), within the time and length constraints of a 

Masters research report. I see the resulting work, not so much as a comprehensive achievement in any of the 

aforementioned areas, but rather as a hybrid piece that makes a methodological contribution in terms of a 

structure for approaching the unarchived (or even unarchivable), within a research-historical framework. 

 

Theoretical toolkit 

My theoretical toolkit is a set of interlinked concepts that I have devised to explore Pager’s fragmented 

biography and the archive he created. I have structured the archive into three tiers, which are the primary, 

secondary and tertiary archives (defined and explored in more detail below). The toolkit, in its expanded 

dimension, presents three themes that run like threads through these layers and are embedded throughout 

this document. These themes are ‘archive’, ‘landscape’ and ‘narrative’. In terms of the structure of this report, 

I think of the three-tiered archive as a horizontal scaffold for this thesis, which is interwoven with the three 

interlinked themes as ‘vertical’ threads that cut across these horizontal layers.  

 

Archive 
 
Traditionally, archives are viewed as a window into the past and retain particular documentary by-products of 

human activity for their long-term value (International Council on Archives 2016). The examples of physical, 

paper-dense repository and digital database I have been working with both contain certain landscape and 

narrative thematic threads. In my definition, the archive extends beyond the institutionalised documentary 

domain to the physical locale of the various rock art sites recorded by Pager, and beyond this to the wider world, 

towards the unarchived and potentially archivable. Thus, the archive is positioned as the foundational topic 

throughout this report. 

 
After my experience at RARI, I began to reflect on the contents of the archive as well as the kinds of materials 

that tend to be included in formal institutional archives (and by extension excluded). The archive at RARI houses 

a range of photographic negatives as well as prints of some of the rock shelters. There are several large-format 

recordings of the rock art sites that Pager collected at Didima Gorge, including several other huge, almost hyper-

realistic, roll-out panels. I describe them as hyper-realistic because having initially seen only the digital scans on 

the RARI website, the impressive and monumental 

materiality of the physical copies struck me forcefully 

once I had the opportunity to view them first-hand. 

Like so many copies, they translate the rock art into 

flat, distorted images, but they also embody a kind of 

enhanced reality, because of Pager’s attention to 

making the rock paintings themselves particularly 

visible and colourful against the backdrop of the rock 

canvas. 

Making these copies involved a labour-intensive 

method whereby Pager produced visually complex 

Figure 32: Harald Pager, n.d, H. Pager 33, oil painted eland on 
black-and-white photograph. The Rock Research Institute. 
Captured by Lemishka Moodley.  

 

Figure 42: Harald Pager, n.d, H. Pager 33, oil painted eland on 
black-and-white photograph. The Rock Research Institute. 
Captured by Lemishka Moodley.  

FIGURE 2 

 

Figure 2: Harald Pager, n.d, H. Pager 33, oil 

painted eland on black-and-white photograph. 

The Rock Research Institute. Captured by 

Lemishka Moodley.FIGURE 2 



10 
 

reproductions through a process of combining three types of media (photography, tracing and oil painting). To 

begin, he documented the details of each site photographically. The negatives were then developed in 

Johannesburg and printed as 1:1 black-and-white prints. Thereafter, he returned to the site with the 

photographic prints and traced all the figures that were not clearly visible in the pictures. Pager sought to 

replicate the rock art with a high degree of accuracy by painting directly onto the photographs in oil paint. The 

kind of accuracy he sought was an enhanced visibility of the painted figures whose clarity and colour he 

emphasised against the background of the rock canvas. He spent a considerable amount of time in the presence 

of the art in the landscape so that he could capture the exact texture of the paintings (the pronounced cracks 

and bumps in addition to the eroded and exfoliated parts of the rock surface). These artworks also embody 

Pager’s unique style. Understandably, no copy is perfectly accurate and comprehensive. However, Roland 

Barthes’ central argument in ‘The Death of the Author’ (1968) is that the viewer is given power over the 

interpretation of the artwork and the same idea would apply to the rock art recorder’s work. Interpretatively, 

the nature of the accuracy that Pager was able to achieve considers the overall visual aesthetic of the copies 

when compared to other rock art recordings by various copyists. His copies embody a kind of pictorial 

enhancement, comparable in some ways to digital enhancement in that it brings differential modification to 

the surface of the image in order to make the rock paintings stand out more from the rest of the image. This 

increases the saturation of the painted figures so that what is painted is clearly distinguishable from the rock, 

but at the same time is still presented in the geological context with all of its textures.  

The Pager archive housed at RARI also contains a copy of the book Ndedema (Pager 1971), along with a filing 

cabinet that contains smaller drawings from several other locations outside of Didima and Pager’s fieldwork 

tools. The physical archive feels somewhat overwhelming compared to the digitised versions of these 

documents that can be accessed via the SARADA website1 because of the number, scale and fragility of the 

copies. In addition, the recordings of the rock shelters, as spaces, constitute a large part of the Pager collection. 

For example, the Sorcerer’s Rock documents, which can be considered a kind of ‘sub-archive’ representative of 

the total archive, comprises copies of the main rock art panel, information about the site, and photographs that 

can also be viewed in the Ndedema book. Via SARADA’s digital platform, anyone can have access to view the 

‘condensed’ version of Sorcerer’s Rock. The platform houses more than 200 records for this site, including 

several of Pager’s hand-painted copies. There are also copyright restrictions, and for that reason, high-

resolution images can only be accessed after sending a formal request to SARADA. The digital archive does not 

show Pager’s tools or stationery. There is also a screen interface that stands between the researcher and the 

physical space of the archive. Although convenient and accessible, provided that one can connect to the 

internet, the website offers a different kind of experience when compared to sifting through documents with 

white gloves in the physical archive. Not all the senses are engaged in the same way.  

Michel Foucault described the archive as more than a collection of artefacts and the space that houses them, 

elucidating the rules that comprise the archive as a discursive formation and pointing to the inherent 

discontinuities (Foucault 1972, cited in Greenberg 2012, p. 161). Therefore, instead of simply reproducing 

information, the archive produces meaning as well; moreover, Foucault’s concept of the archive reminds us 

that the storage, organisation, and redistribution of information are never passive or innocent; they always 

inform political and historical discourse (Evans 2010). A second wave of intense interest in the archive as a 

significant discursive space was initiated by Jacques Derrida. His 1996 book titled Archive Fever: A Freudian 

Impression, explored various approaches to the archive, interrogated how it is constructed, and questioned 

what informs the archive’s representation and meaning (Evans 2010). Foucault and Derrida’s work continues 

                                                           
1 Website link: http://www.sarada.co.za/#/library/  

http://www.sarada.co.za/#/library/


11 
 

to be influential in drawing attention to the archive and the complex ways in which we construct and organise 

our histories (whether official, collective, and/or personal).  

This then led me to think of Sigmund Freud’s text, ‘A Note Upon the “Mystic Writing Pad”’ (1925), in which 

Freud offers the idea of a mnemonic system that describes the capacity which a substance or organism 

possesses for retaining the after-effects of an experience, or the stimulation undergone by itself or its 

progenitors (Freud 1925, p. 20). What this suggests in the context of rock art, and in relation to Pager’s copies, 

is that the copy becomes a place where a form of memory is deposited and can be evoked or written into history 

at any point in time. The question of the unarchived and unarchivable arises from the fact that what is archived 

is highly selective and subjective. There are only small traces that allude to the fullness of what actually 

happened in the past. The memory of the documented moment is additionally limited in the sense that one will 

only generally know of the documented information rather than the experiences that the artist endured for 

instance. The documented paintings on the cave wall then become a representation and mediated account 

within a specific moment in time which can be used as an excellent referential tool in an empirical context, for 

example, to understand the rate at which the deterioration of the ephemeral rock art occurs in the landscape.  

As a further consideration, a layering of ‘traces’ is consequently present in the landscape and can potentially be 

traced into personal memory. By this, one could argue that the expanded Pager archive encapsulates the 

relations among “the living and the dead, the roles of ordering, preservation, and registration, the rival 

patriarchal aspirations of archives and psychoanalysis to a place of ultimate authority, authenticity, 

originariness, and truth, and the differentiation of the natural (biological) from the artificial (technological or 

mediative)” (Derrida 1995, p. 91, in Greenberg 2012, p. 161). This idea has also contributed to my interest in 

investigating the realm of personal histories, which are forged through the recording of living memory.  

RARI and SARADA are jointly extending the life of the physical archive while simultaneously preserving it on a 

digital platform so that it can be accessible for years to come, but the recording of narratives personal histories 

in the form of narratives has the potential to bring this archive to life and add valuable layers of meaning. In 

relation to the Pager archive (and no doubt to many other archives), there exists an opportunity to learn about 

the entanglement between a rock art copyist and a recorded landscape. This is another aspect of investigating 

the archive beyond its sense of authoritative, institutionalised documentation, by exploring how it might 

encompass a more intangible shared memory of Pager, his life and his work. Thus, I look at an even broader 

definition of archive, one that, beyond the landscape, encompasses also the stories and narratives that have 

not yet been written down, over and above the traditional documentary archive that these theorists refer to. 

This suggests that the archive has a life beyond the dimensions of the physical archive that can be recalled 

through historical research as well as memory, what might be referred to as the ‘biography’ of the archive. I 

approach the archive’s biography by ‘exploding’ the archive into primary, secondary and tertiary tiers in order 

to gain insight into its layered history and its expanded dimensions. As discussed above, I define the primary 

archive as the site in the landscape (following Williams 2014), the secondary archive as Pager’s documentation, 

and the tertiary archive as belonging to the living2 context in which I am working, which includes the digital 

space, as well as the as-yet-unarchived. The tertiary archive is unavoidably less precise and more difficult to 

describe and contain, because it exists as a kind of ‘afterlife’ of the other tiers of the archive; however, it 

                                                           
2 It should be noted that the tertiary archive is fundamentally a projection of my cotemporary interest on to the past. I acknowledge 

that there is no evidence that the San artists considered the construction of the images as a form of archive, however, based on an 
interpretation of the rock art, it may as well be viewed as a documented account. Hence, I merely used the aforementioned archives 
theorist writing to inform my understanding. 



12 
 

provides me with a space for experimentation and reflection on my own process of exploring and adding to the 

archive.  

 

Landscape 

Landscape is an important theme that runs through all the 

tiers of my expanded definition of archive, as well as being a 

strong theme of the Pager archive itself. Since Pager sought 

to record aspects of the physical landscape as well as the 

archaeological artefacts that are embedded within it, 

landscape becomes a fundamental archival material, and 

also an organising principle that helps make sense of the 

archive, and that is also linked to time. I have found Tim 

Ingold’s 1993 text, ‘The temporality of the landscape’, useful 

because he aims to unite archaeological and anthropological 

perspectives through a focus on the landscape and its 

impermanent nature. Landscape can be viewed through a 

‘dwelling perspective’, whereby the “landscape is 

constituted as an enduring record of — and testimony to — 

the lives and works of past generations who dwelt within it, 

and in so doing, have left there something of themselves” 

(Ingold 1993, p. 152).  

Ideas of landscape can be traced through Pager’s work, and also traverse the fundamental understanding of 

San religious thought and art (which are essentially all embedded in components of landscape) (Lewis-Williams 

2010, p. 1). Christopher Chippindale and George Nash postulate the importance of understanding the reasons 

behind the creation of rock art and learning the identity of the artist, while also considering the role of the 

landscape (Chippindale & Nash 2004, p. 21, in Lewis-Williams 2010, p. 1). The rise of landscape approaches 

within rock art studies led to Benjamin Smith and Geoffrey Blundell’s proposal that an emphasis on landscape 

can lead researchers down the wrong track, and that it is essential that include ethnographic contextual 

methodologies to support their research (Smith & Blundell 2004, p. 259, in Lewis-Williams 2010, p. 1).  

And yet landscape in relation to rock art can be understood in different ways. Pager’s archive arguably embodies 

the notion of landscape through a kind of dwelling perspective reminiscent of Ingold’s ideas, and based on his 

interaction with and experience of the world. The archive, essentially, is a remnant of Pager’s existence and a 

testimony to his presence in the various sites that he documented. The archive shows how people’s active 

engagement with the world can further unravel and clarify the meaning of landscape, in contrast to the less 

embodied concepts of land, nature and space (Ingold 1993, p. 174). In fact, landscape raises a crucial point 

because, in addition to being a thematic thread that runs through all the tiers of the archive as defined in my 

study, the land as a place that houses a set of materials also forms a kind of archive in and of itself, which 

constitutes the ‘primary archive’ of my study. Although this research project has not involved any fieldwork, 

landscape nonetheless became a significant aspect in gaining an understanding of the Pager archive, in 

conjunction with my exploration of ‘archive’ and ‘narrative’. Even as a thematic thread throughout this research 

report, ‘landscape’ opens out from the present into the past and bears the residue of the social, cultural, 

Figure 3: Image of Didima Gorge from Pager (1971, p. 85). 

 

Figure 53: Image of Didima Gorge from Pager (1971, p. 
85). 

FIGURE 3 

 

Figure 3: Image of Didima Gorge from Pager 

(1971, p. 85).FIGURE 3 



13 
 

religious and spiritual exchanges of those whose lives are woven into it (Atha et al 2013, p. 1). However, the 

remnants within the landscape perish over time and constitute a different form of recording the past when 

compared with a tangible, documented archive. For that reason, I have employed narrative (as a final theme) 

to describe but also recount the land’s past experiences.  

 

Narrative 

Narrative, as a conceptual thread, has the potential to enliven and inform the landscape and the archive, and is 

also linked to the ‘personal histories’ component that I identify and explain in the methodology section below. 

Through Pamela Jane Smith’s ‘personal histories’ approach, narrative, as a theoretical tool, is also used as a tool 

to further unpack the biography of the Pager archive by engaging in conversations with several interlocutors 

about Pager, his life and work. By extension, this fosters a deeper comprehension and interpretation of the 

process of research (specifically through ‘oral’ exchanges), and the social context in which research takes place, 

which questions and challenges the ‘aura’ of the traditional archive and its contents. ‘Meaning cannot rest or 

stay the same’ every time one chooses to activate a repository of memory, and narrative influences how one 

understands and writes about it (Vinnicombe 2010).  

According to Sven Ouzman, rock art sites are best understood as places where a world order could be 

contemplated, questioned, supported or altered by means of ‘mindscaping’, which can be defined as the 

process of shaping the landscape culturally (Ouzman 2001, p. 238). My sense is that the frame of rock art 

discourse readily encourages the development of metaphors in the form of storytelling, which could enable the 

exploration of old topics in new ways through a ritualised human behaviour for example — which Pager himself 

arguably also performed in some form in order to document all the works at Didima. I understand Pager’s 

ritualised behaviour to be located in the ways in which he went about recording each site. His method of copying 

was systematic, meticulous, immersive, and involved several steps that he followed habitually. After reading 

David Lewis-Williams’s article, ‘The imagistic web of San myth, art and landscape’ (2010), I began making links 

to the notion of a dwelling perspective in conjunction with mindscaping, linkable also to the “parallels and 

interrelationships between certain myths, paintings and landscapes” (p. 1). Thus, the archive in relation to the 

landscape is able to be retold, reconstructed, reformed, and studied in the present through a network of lived 

experiences.  

 

Methodology 
 

My exploration of the physical and digital archive encouraged a deeper analysis of Pager’s archive through 

narrative, following a method adapted from Smith’s inspirational ‘Personal Histories Project’ (University of 

Cambridge 2015). The use of inter-linked theoretical concepts throughout this project helped me comprehend 

and interpret how the landscape and digital archives move beyond the documented archive into an expanded 

archival domain that can enable the extension and critical unpacking of the history of Pager’s archive. In order 

to extract, excavate and learn about the archive’s biography, I use one of the shelters Pager recorded in Didima 

Gorge as a node within the larger archive, as previously mentioned. My focus on the main rock art panel at 

Sorcerer’s Rock afforded me the opportunity to expand my investigation of the archive and contribute detail to 

the archive’s grand narrative.  

Most of the information obtained to write this research report came from the physical and textual archive at 

RARI, and the SARADA online database, in addition to excerpts extracted from the conversations had with the 

various interlocutors to form an extended ‘oral’ archive. In order to mobilise narrative as a mode and medium 



14 
 

of exploration, I used a 'personal histories' approach to create links between the different tiers of the archive 

in order to investigate the context of Pager’s work, and his contribution to the history of rock art studies in 

South Africa. The conceptual strands consist of archive, landscape and narrative, and are informed by a 

consultation with living sources, who assisted me in understanding the materials in both the physical and digital 

archive. Ultimately, the links between the respective archival domains extend beyond the archive’s materiality 

to include narrative aspects in the form of conversations about Pager, which point to the ‘personal histories’ 

component of this project. My study of the Pager archive explores aspects of how the archive was made and 

highlights the various personal experiences embodied in it. ‘Narrative’ creates synergy and meaningful links 

between the theoretical strands.  

Because Harald Pager died just over thirty-three years ago, and there was no way for me to interview him to 

answer the many questions that I had, I incorporated a ‘personal histories’ component into my methodology 

(Carpeneti 2012), where I interviewed various people who knew Pager in different ways and invited them to 

tell stories about his interest in rock art and his documenting practice. The recording of personal histories 

through conversations with interlocutors contributed to my understanding of Pager’s biography and his archive. 

This demonstrates that a ‘personal histories’ approach has the potential to enliven and add valuable layers of 

meaning to the existing archive. The oral historian Pamela Jane Smith (University of Cambridge 2015) points 

out that those personal histories have the potential to provide insight into disciplinary histories, as well as bring 

together people from different social classes and age groups, transport education from the institution to the 

world, and question the traditional boundaries of conducting research. Smith’s Personal Histories Project 

(University of Cambridge 2015) offers a way of using narrative in order to illuminate the development of 20th-

century archaeology (Carpeneti 2012). In wanting to understand how knowledge is produced, her approach 

considers narrative through the idea of knowledge and its production being recounted as a story told from first-

hand experience. Inspired by her project, I wanted to explore how the links created between the different layers 

of the archive allow me to investigate the context of Pager’s work and his contribution to the history of rock art 

studies in South Africa. Since no biography has been written on Pager, a narrative approach is a useful tool for 

exploring the primary, secondary and tertiary archives. This approach is guided by the personal narratives and 

mutual exchanges of ‘narrators’ or ‘interlocutors’ with whom I have held conversations, and whose 

contributions enable me to engage with and explore Pager’s biography and archive (Carpeneti 2012). Thus my 

research report incorporates a range of personal accounts, interactions and perspectives related to the sites in 

Didima Gorge, to Pager and his archive, and ultimately to the history of rock art studies and archaeology in 

southern Africa. In the process of unravelling something of the archive’s biography, I was able to gain a deeper 

understanding and interpretation of the process of research and the social context in which Pager’s research 

took place.  

I have also used a narrative style in the writing of my research report more generally, in the hope of creating a 

text that is engaging and reads like a story, by engaging with history-writing as a form of story-telling. 

Storytelling, as one kind of narrative method, has the power and capacity to enliven the past and firmly root 

the historical materials in a social context, while also exploring the potential of narrative as a research method 

and form of writing. The use of narrative allows me to creatively explore the researched content and potentially 

produce an alternative exploration of what Pager and other researchers have previously proposed about 

Sorcerer’s Rock, as one example. In Paul Thompson’s The Voice of the Past: Oral History, he states that the 

domain of oral history consists of the content of history as well as the processes of writing it, and that it breaks 

through boundaries between educational institutions and the outside world (Thompson 1988, p. 25). This 

explanation suggests that by shifting the focus from fact-finding research to new avenues of attaining 

knowledge, one can challenge the assumptions and judgements of historians by bringing recognition to diverse 



15 
 

groups of people with relevant experiences (Thompson 1988, p. 25). Mieke Bal elaborates on the notion of 

‘narrative’ as a mode rather than a genre, and as a major reservoir of cultural baggage that enables us to make 

meaning (2004, p. 10). Her notion of ‘narrative’ as a transdisciplinary concept — not confined to any one 

academic discipline — informs my whole project. The archive is constructed by a series of events that have 

taken place at, and in relation to, a specific place in the landscape, including the experiences of an earlier artist, 

past recorders of the site, and me, who is attempting to investigate the formation of the primary archive, 

including its extension into the secondary and tertiary archives through the recording process and its afterlife. 

In many senses, narrative also then guides the proposed use of a ‘personal histories’ approach which has the 

potential to enliven the archive of Sorcerer’s Rock, the Didima area, and beyond. 

  

Importance of this research 
 

David Coulson suggests that rock art in Africa represents some of the earliest remaining expressions of 

humankind’s visual communications and previously unrecognised artistic ability in the deep past (Coulson 2004, 

p. 8, cited in Deacon 2004, p. 8). Rock art is arguably the oldest, longest running and most extensive record of 

human thought in pictorial form. Rock art is a body of traces that we can consider ‘archival,’ and its study makes 

it possible for one to gain access to the past through a pictorial medium. But rock art archives embedded within, 

buried under or scattered over parts of the land are under major threat because they are a fragile and non-

renewable resource. Rock art is exposed to a range of ongoing processes of natural weathering, and was not 

necessarily designed to last any great length of time. The longevity of rock art is further endangered through 

vandalism and through less intentional damage caused by people who visit or camp in shelters that house rock 

art. These threats to the rock art reaffirm my focus on the archive, which functions as a mechanism for 

extending the presence and footprint of the rock art in the wider world. Nowadays, the accepted value of the 

markings made on cave walls is constructed within a robust, ethnographically informed theoretical academic 

framework — in addition to its widely recognised heritage and tourism value, as well as its aesthetic value, 

acknowledged also by many contemporary artists. Rock art and its extended archive are also presented to the 

public in many visual forms (such as museum exhibitions and displays, books, films, television shows, and the 

many other ways in which this knowledge can be disseminated). However, it must be noted that the living 

context of copying and researching is largely lost when the findings of rock art studies are presented to a wider 

audience. This research study could contribute to addressing this loss, but the translation or interpretation of 

the findings of rock art research for popular reception is not the primary focus of this paper.  

An additional point is that Pager’s recording work has not been engaged with in much depth in the South African 

context because his interpretations of what the art “means” have long been superseded. Also, Pager died in 

1985, so a ‘personal histories’ project in relation to his archive had to happen soon in order to gain access to 

and preserve something of the living context of its production. To some degree, Pager’s memory lives on in the 

texts written about him and the narratives told by those who knew him. Still, there are currently relatively few 

people alive who had direct experience of working with Pager. In the course of this study, I identified and 

attempted to contact several people who knew Pager in different capacities. The archaeologist Aron Mazel and 

artist Paul Stopforth took the time to have e-conversations with me. Later on, I was fortunate to be part of a 

group conversation with historian John Wright, archaeologist Leon (Jakes) Jacobson, rock art researcher Neil 

Lee and Shirley Woodhouse (the wife of the late Bert Woodhouse) about Pager’s life and work. Finally, I had a 

conversation with rock art researcher David Lewis-Williams who also knew Pager and was able to contribute to 

an understanding of how the Pager archive ended up at RARI, and related his experiences with Pager through 



16 
 

an explanation of the history of rock art interpretation. Unfortunately, my repeated attempts to contact Pager’s 

wife (Shirley-Anne Pager) and his daughter (Cara Pager) were unsuccessful, and I have still not received a 

response from either. I also attempted to contact John Kinahan and would have liked to have had additional 

conversations with archaeologists Tilman Lenssen-Erz, Peter Breunig, and Beatrice Sandelowsky.  

This research is important also because it creates an awareness of the different ways in which narrative can be 

used to make sense of and enliven an archive. Narrative as a research method (involving multiple readings and 

interpretations, and the creation of multiple meanings) has played a role in many other knowledge spaces over 

time (anthropology, ethnology, post-colonial research, etc.). In the domain of rock art research, for example, 

rock art researcher Patricia Vinnicombe recounts her experience in Western Australia in 1977 in her text, 

“Meaning cannot rest or stay the same” (2010). Through Vinnicombe’s conversations with local ‘indigenous’ 

peoples, she was able to understand and discover new aspects of the meaning of the Aboriginal rock art that 

dealt with a specific experience of the Aboriginal past. Vinnicombe broadened her understanding and thinking 

about a ‘moment’ in Australia’s past from the perspective of her understanding of San rock art, and the theories 

that contributed to that understanding (Vinnicombe 2010, p. 24). She was, moreover, able to recount the 

stories in relation to what she had viewed, and, in many ways, a narrative approach also informed her research 

because an understanding of how the Aboriginal culture functioned was also explained through the 

performative gesture of storytelling. 

Lastly, at a higher level, I am interested in connecting my research to some of the ways in which narrative 

features in current theoretical domains in humanities research. The ethnographical approach by Vinnicombe, 

Lewis-Williams and others, uses relevant accounts from archival sources to recall the past, make various 

interpretations through an inductive process, and thereby shed light on the meanings and interpretations of 

the rock art paintings. This, then, also forms part of a narrative mode.  

 

Chapter structure 

 

I have structured the report according to the primary, secondary and tertiary archives (as outlined above), with 

different thematic threads that weave across and hold together the various layers of the archive.  

 

CHAPTER 1: The Primary Archive  

This chapter defines the primary archive as rooted in the landscape and provides a descriptive physical, 

archaeological and historical overview of the Drakensberg landscape with a focus on Didima Gorge.  

CHAPTER 2: The Secondary Archive  

This chapter explores the secondary archive, defined as the body of historical recordings of rock art. It traces 

the history of recording in Didima Gorge prior to Pager, and shifts to a focus on Pager’s recording campaigns, 

his copying method and the archive it generated. I explore the archive related to the Sorcerer’s Rock site in 

greater detail, in the form of a ‘visual essay’, as a small sample that provides insight into the wider documentary 

archive.  

CHAPTER 3: The Tertiary Archive  

The tertiary archive is a less precise entity, but can be defined as the living context and ‘afterlife’ of Pager’s 

archive, and comprises the documentary archive in its current state, the digital archive, and the space in which 

I am working, all of which I explore in this chapter. I thread together the various narratives collected from my 



17 
 

conversations with various people about Pager, his life, his work and his experiences, particularly at Didima 

Gorge, as a contribution to the archive. I also offer a reflection on the ‘personal histories’ approach that I have 

used to ‘dig’ into the biography of Pager’s archive (my object of curiosity), as well as a reflection on the archival 

and digital space I have been working in.  

 

APPENDICES 

Two documents append the report: a creative visual essay (Appendix 1). A selective transcription of the 

conversations held during the course of this research was previously included, however, an ethical decision was 

made to omit the conversation transcripts. It became necessary to protect the privacy and confidentiality of the 

individuals involved in those conversations, especially since the discussions contained sensitive and personally 

identifiable information. Moreover, the omission ensures that they are not used or shared in ways that were 

not originally intended.  



18 
 

Chapter 1: Primary Archive 
 

 

Kent Williams, as a student of art history, describes rock 

art as a blurred, ‘written’ amalgamation of text and 

images (2014, p. 2), and argues that rock paintings within 

the landscape can be considered archival ‘documents’ 

that record aspects of San religious beliefs. Hence, this 

establishes the possibility for the landscape as a whole to 

be considered as a primary archival repository, where one 

should consider “the ‘primary archive’ to be the rock art 

out on the land, whose living context has vanished but 

that has been extensively studied by rock art researchers 

and anthropologists” (Williams 2014, p. 2). Taking 

inspiration from Williams, I define the landscape as a kind 

of primary archive. In other words, the primary archive is 

the foundational layer of the total archive, from which the 

secondary and tertiary layers are derived. The primary 

archive is the land that encompasses the infinite yet 

socially constructed landscape, on and through which 

human interactions and impressions were made, buried 

and rediscovered (Williams 2014, p. 2). Landscape, in this 

research process, is the site where the rock art was conceived; the place where the people who conceived the 

art lived; and a physical space that is constantly changing and adapting. Ingold considers landscape as “a kind 

of lowest common denominator of the phenomenal world” (Ingold 1993, p. 153). The layered, unbounded and 

all-encompassing land opens out into the past and enfolds the lives and times of predecessors, who, over many 

generations, have moved around it and have played a part in its formation (Ingold 1993, p. 152). Thus, the 

primary archive in which rock art is embedded is the landscape (Williams 2014, p. 2). 

Landscape, as a concept, opens up ways of thinking about the past and “becomes an insightful tool that shows 

how and from what the contemporary world has been constructed” (Williams 2014, p. 1). The land illustrates 

the active engagement and relationship between people and their environment, and how the residue of 

people’s engagement with the land challenges contemporary researchers to unpack the constructed narrative 

of the past (Godby 2010, p. 128). By interpreting the material evidence of human imprints in the archive, one 

can read the landscape like a text that is continuously shaped and re-shaped by a “collection of information 

amassed and redefined over centuries and millennia” (Bastian 2014, p. 47).  

In this chapter, I outline how the landscape of the uKhahlamba-Drakensberg mountains is not only made up of 

a collection of natural landforms (such as mountains, hills, plains, plateaus, streams, and soils such as sand or 

clay), natural vegetation, and other features of the natural landscape, but also includes a cultural landscape 

that people have shaped over time (Wright & Mazel 2007, p. vii). This reasserts the idea of a “dwelling 

perspective” and illustrates how the landscape becomes a space that testifies to the existence of a past 

community through the residual fragments they leave behind (Ingold 1993, p. 152). Essentially, I recognise how 

the landscape works against a perceived binary relationship between humans and nature (Meinig 1979, p. 2, 

Figure 4: Babylonian concept of the world’s cosmos in the 
form of a three-tiered structure. From De Gracia (2015).  

 

 

Figure 64: Babylonian concept of the world’s cosmos in the 
form of a three-tiered structure. From De Gracia (2015).  

 

FIGURE 4 

 

Figure 4: Babylonian concept of the world’s 

cosmos in the form of a three-tiered structure. 

From De Gracia (2015). FIGURE 4 



19 
 

cited in Ingold 1993, p. 154). A descriptive overview of the ‘Berg’ (as the uKhahlamba-Drakensberg mountain 

range is colloquially known in KwaZulu-Natal) and a detailed description of the Didima Gorge valley will form a 

basis for understanding how the ‘organically’ evolved landscape has manifested itself, and how the spiritual, 

economic, and cultural significance of the area is entrenched in and has developed alongside the landscape’s 

physical characteristics. The physical and archaeological landscape of the region is discussed in relation to the 

associated cultural landscape, and in relation to the physical evidence of historical human use at various sites 

(e.g. the rock paintings on cave walls, and the remnants and objects in cave deposits).  

 

uKhahlamba-Drakensberg Mountains 

 

Figure 5: Photograph of the Drakensberg Mountains. Captured by Moiseenko (2018).  

 

Since this project is purely archival, and I have not conducted any fieldwork, I familiarised myself with the 

broader landscape through consulting various sources related to the uKhahlamba-Drakensberg mountains. 

Video footage shows these mountains as a vast landscape with enormous rock formations that lie majestically 

immersed in lush flora. The mountains currently lie within the Maloti-Drakensberg Transfrontier Conservation 

and Development Area, a World Heritage Site that spans the Sehlabathebe National Park in Lesotho and the 

uKhahlamba-Drakensberg Park in South Africa (UNESCO n.d). This area is one of the largest protected sites in 

southern Africa because it is occupied by many threatened and endemic species, and has a prolific number of 

rock paintings made by the San3 people (UNESCO n.d.).  

                                                           
3 Website link: http://www.sarada.co.za/#/library/  

‘San’ or ‘Bushman’ are conventional terms, yet are also in some instances pejorative terms used as an umbrella for a broad grouping 
of hunter-gatherer people generally thought to be responsible for the figurative rock art across the entire region (Lewis-Williams 2003, 
p. 123). Prior to colonial intervention, the hunter-gatherer community were usually considered to be the San. However, the allocated 
label/name does not consider the more detailed questions of authorship of the art, of smaller-scale group identity in relation to 
authorship, and of the diversity in the art across space and time, and remains an externally constructed label, but one for which it is 
challenging to find alternatives for the archaeological time period. 

FIGURE 5 

 

 

http://www.sarada.co.za/#/library/


20 
 

 

Through Tracks In A Mountain Range: Exploring the History of the uKhahlamba-Drakensberg (2007) by John 

Wright and Aron Mazel, I learnt about the history of the Drakensberg and the people, animals and plants that 

have occupied the region. The mountain range spans 200 kilometres and bears different archaeological 

signatures, including the remnants of the San hunter-gatherers, who dwelled across the escarpment for 

thousands of years prior to the settlement of black farmers within the Drakensberg area (Wright & Mazel 2007, 

p. 1). What is most important, or rather most remarkable, about the uKhahlamba-Drakensberg is that it is 

considered to possess “the most extensive and impressive galleries of rock art to be recorded anywhere in the 

world”, as well as fragments of culture and subsistence “found on the floors and in the deposits of numerous 

rock shelters” that prove the existence of the long-term occupation of the San (Wright & Mazel 2007, p. 1).  

 

 
Figure 7: Lateral view of the uKhahlamba-Drakensberg showing the 
geological formations from Pager (1971, p. 5).  

 

 

 

 

 

 

 

 

At the geological scale, the oldest layer of rock found in the Drakensberg is estimated to have existed 250 million 

years ago (Wright & Mazel 2007, p. 2):  

FIGURE 6 

FIGURE 7 

Figure 6: Image of a map indicating the transboundary and geographical location of the uKhahlamba-Drakensberg heritage site 
(as circled above in red). From Pager (1971, p. 3). 



21 
 

 

The earliest deposits consisted of sandstones and mudstones which were laid down under semi-arid 

conditions in lakes, flood plains, and meandering river channels. These rocks which today lie at an 

altitude of about 1300 metres, belong to what geologists have named the Beaufort Group. Overlying 

them are layers of the Stormberg Group, which today make up most of the foothills. The lowest of these 

layers belong to the Molteno formation.  

 

Correspondingly, the archive, like the landscape, is also (literally) layered, and these layers also form a kind of 

archive relevant for the deeper geological time-scale. The mountain is, overall and primarily, a sandstone 

palimpsest phenomenon (Wright & Mazel 2007, p. 1–2). The layer above the Red Beds in relation to the 

categories mentioned above is one of the layers that form part of the Stormberg Group. Below the Red Bed 

layer is cream-and-white sandstone currently known as the Clarens Formation, and still often referred to as 

Cave Sandstone (Lewis-Williams 2003, p. 9; Fig. 5). The massive stratum formed by the long lines of ‘kranses’, 

or cliffs, runs just under the summit of the ridges of the Little Berg and represents its most distinctive feature. 

It is in this layer, particularly, that hundreds of rock shelters and overhangs are to be found (Lewis-Williams 

2003, p. 9). The Clarens Formation consists of many strata and varies in permeability and hardness, and rock 

shelters formed from the softer layers that eroded more rapidly than the overlying harder layers (Wright & 

Mazel 2007, p. 2–3):  

 

In some instances, large blocks of sandstone were dislodged from the cliffs and settled on the slopes 

below in ways that form overhangs, thus assisting in the creation of habitable rock shelters. In these 

shelters and overhangs San hunter-gatherers made their homes for thousands of years, and here the 

great majority of rock paintings are to be found. 

 

The landscape is mostly covered in grasslands, with scattered and varied plant communities and other 

associated fauna governed by the climate, geology, soil, drainage and effects of fire (Wright & Mazel 2007, p. 

4). In the past, the Drakensberg area accommodated “a wide variety of mammals, dominated by antelope and 

smaller carnivores”, whereas today, only 11 types of antelope and the odd carnivore are found there (Wright & 

Mazel 2007, p. 5).  

Wright and Mazel (2007) debate how far back in time the ‘San’ as a group identity in this region can be 

ascertained. Micro communities of hunter-gatherers for which we have no specific ethnonyms have lived across 

the uKhahlamba-Drakensberg area (2007, p. 6 & 88), but the chronological details of these hunter-gatherer 

occupancies are vague and cannot be pin-pointed exactly, hence the ongoing topic of archaeological research. 

Wright and Mazel suggest, however, that hunter-gatherers have lived in the area from before 25 000 years ago, 

but their occupation of the area dates primarily to the last 3 000 years, i.e. the Later Stone Age (Wright & Mazel 

2007, p. 23). Alex Willcox stated that Middle Stone Age hunter-gatherers would have been unable to gain access 

to the mountains due to the thick forest and snow that covered the landscape at that time (Wright & Mazel 

2007, p. 23); however, this hypothesis proved to be invalid when Mazel identified Middle Stone Age artefacts 

in the river gravels in the north and south of the Drakensberg, as well as on the floors of five rock shelters in 

the north (Wright & Mazel 2007, p. 23). Wells’s expedition in the 1930s had already discovered Middle Stone 

Age artefacts in the rock art shelter deposits of the Cathedral Peak and Cathkin Peak areas (Wright & Mazel p. 

24–25). Even today, archaeologists are unable to establish clearly when the land was first occupied (Wright & 

Mazel 2007, p. 24). But because the deposits are scarce and are far apart, one can argue that the occupation 

by hunter-gatherers was not extensive before 25 000 years ago (Wright & Mazel 2007, p. 23–25).  



22 
 

 

From recent history there are very few written records pertaining to the history of the San occupation in the 

mountains prior to the British establishment of the rudiments of a functioning bureaucracy (Wright & Mazel 

2007, p. 89). Written historical accounts of the past are therefore rooted in the colonial experiences of the 

white man, rather than in those of the people they ruled (Wright & Mazel 2007, p. 10). However, the rock 

paintings and related documentation provide a major source of evidence of a hunter-gatherer presence in the 

Drakensberg (Wright & Mazel 2007, p. 28). By “drawing on evidence from archaeology and from the 19th 

century documentary records, […] stories about the hunter-gatherers as a part of […] ‘the history of the 

Drakensberg’” are told even though nothing is known about how they perceived their history (Wright & Mazel 

2007, p. 7). Rock art combined with a culturally informed history by San ‘intellectuals’, became (primary) 

testaments/evidence of the San’s ways of thinking — especially about what we could call spiritual matters — 

but at present we have no way of interpreting these sources as clues to the specific histories told in their 

communities (Wright & Mazel 2007, p. 7). It would be worthwhile to engage more closely with the unwritten 

stories rooted in the pre-colonial materials that have survived in the memories of oral historians among the 

people who dwell in the Cathedral Peak area, and in other nearby communities, and indeed the question of 

whether the rock art might be considered something akin to an ‘archive’ from a San perspective, but this goes 

beyond the scope of this research report as well as Wright and Mazel’s study (Wright & Mazel 2007, p. 8). I 

contend that it would be possible to deepen a consideration of this landscape as an ‘archive’ with further stories 

and histories to uncover, because layered records demonstrate a relationship between the land and its 

occupiers, who continuously shape and re-shape “a collection of information amassed and redefined over 

centuries and millennia” (Bastian 2014, p. 47).  

 

Didima Gorge 

Figure 8: Image of Didima Gorge. From Night Jar Travel (n.d.). 

FIGURE 8 



23 
 

 

Didima Gorge is a valley situated in the Cathedral Peak area, in the northern part of the uKhahlamba-

Drakensberg Park (Mazel 2011, p. 284). Didima Gorge is the area where Pager recorded 3 909 ‘paintings’4 and 

documented seventeen shelters over a period of four years (1965–1969). His study covered Sebaaieni Cave, 

Elephant Shelter, Water Shelter, Botha’s Shelter, Rock-Fall Shelter, Shirley’s Shelter, Nuttall’s Shelter, Sorcerer’s 

Rock, River Rock, Poacher’s Shelter, Leopard Cave, Nänni’s Rock, Silt Rock, Sugar Loaf Rock, Asbestos Shelter, 

Junction Shelter and Exhaustion Shelter (Pager 1973, p. 6; Fig. 7). Pager’s recordings from the valley cover a 

mere 10% of the c.40 000 paintings in the Gorge (Mazel 2011, p. 286). Mazel (2011, p. 286) suggests that the 

valley 

 

is not only distinguished by the sheer quantity of paintings, but it contains a higher proportion of figures 

which Pager (1971) has referred to as “ritual” than the surrounding Cathedral Peak and Cathkin Park 

areas and the uKhahlamba Drakensberg as a whole. 

 

The “ritual” aspect here connotes a trance experience that the San undergo for secular and sacred purposes 

(Lewis-Williams 2010, p. 6). Herein lies an anthropological framework through which an understanding of the 

painted figures can be established. Although rock art research has moved on since Pager’s time, he was already 

becoming aware of the fundamental theme of transformation, and of affinities imagined between animals and 

humans in the art. For Pager, the painted figures shed light on transformative experiences through which the 

San became mythical creatures such as therianthropes, ceremonial figures, bristle bulls, mythical baboons and 

other fantasy animals (Pager 1971, p. 338–344).  

                                                           
4 Pager and many rock art researchers refer to the individual figures as ‘paintings’, but approaching the panel as a whole ‘painting’ 

makes it easier to describe and understand the entire composition. This points to the different perspectives that might be applied by 
art historians and archaeologists respectively.  

Figure 9: This image is a combination of two 
images (a photograph and a line drawing) 
published in Harald Pager’s Ndedema (Pager 
1971, p. 86–87). Composite image by Caroline 
Thompson (2014, p.29). It shows the location of 
the seventeen painted rock shelters recorded in 
the gorge. Sorcerer’s Rock is number 7. 

Shelters: 

1. eSibayeni 
2. Elephant Shelter 
3. Water Shelter 
4. Botha’s Shelter 
5. Rock-Fall Shelter 
6. River Rock 
7. Sorcerer’s Rock 
8. Nuttall’s Shelter 
9. Shirley’s Shelter 
10. Poacher’s Shelter 
11. Leopard Cave 
12. Nanni’s Rock 
13. Silt Rock 
14. Sugar-Loaf Rock 
15. Asbestos Shelter 
16. Junction Shelter 
17. Exhaustion Shelter 

FIGURE 9 



24 
 

In addition to the visual references in the art, Mazel pays attention to the acoustic properties of the landscape. 

The word “Didima” may allude to the echoing of rumbling thunder during a heavy rain storm, when the sound 

travels “through the huge kranses of the narrow gorge” and produces a clattering vibration (Mazel 2011, p. 

291). Mazel’s acoustic research reveals another interesting insight about hunter-gatherer history in the 

uKhahlamba-Drakensberg and how it may have influenced the painted representations (2011, p. 291). Mazel 

explores Pager’s comment that “The name of the gorge is a Zulu one, meaning ‘The Reverberating One’, 

implying shaking and upheaval” (Pager 1971, p. 5, cited in Mazel 2011, p. 291). However, based on a personal 

communication with Peter Raper in 2010, Mazel proposes that “the word ‘Didima’ may be an adaptation of the 

|Xam word ‘!gum’ meaning ‘to roar’” (Mazel 2011, p. 292).  

I now narrow my focus on the Didima Gorge valley to examine more closely a single rock shelter that acts as a 

microcosm of the wider landscape and archive.  

 

Sorcerer’s Rock  

Figure 10: Image of Elevations of Sorcerer’s Rock. From Pager (1971, p. 66).  

Sorcerer’s Rock is a rock art shelter situated in Didima Gorge, in the KwaZulu-Natal Drakensberg. The shelter is 

located at 28°59’50” S and 29°16’42” E and is 1 580m a.s.l. (Pager 1971, p. 66). According to Miss L. N. Peirson 

of Richmond, Natal, a group of Natal and Transvaal Mountaineers “discovered” Sorcerer’s Rock in 1948, and in 

July 1958, Pager and A. P. Botha rediscovered and recorded the shelter (Pager 1971, p. 66). Though there are 

earlier references to Sorcerer’s Rock, Pager was the first to publish a detailed account of the site.  

 

Pager noted that the vantage point of the site allowed one to see the entire yellowwood forest (Pager 1971, p. 

66). He recorded access to the site from river level as being up a steep grassy slope to gain entrance to the open 

shelter formed by an enormous fallen rock “that had come to rest half way between the Lower Main Krans and 

the river” (Pager 1971, p. 66). Pager recorded various aspects of the physical properties of the site and its 

immediate landscape. The shelter is small and the tiny flat area of floor below the painted ceiling provides just 

enough space to accommodate two people. During wet and rainy weather, one’s feet get wet. The naturally 

formed shelter has shallow banks on either side of the overhang that protect its occupants from harsh gusts of 

wind. The area toward the deeper interior of the shelter floor slopes, and becomes so narrow that the roof 

meets the floor. A sectioned side-elevation view of the site presents the shape as an acute triangle (Fig. 7). 

Pager suggests that, due to the limited space, many ceremonial rituals could have taken place at the river, in 

FIGURE 10 



25 
 

addition to day-to-day activities such as washing, cooking, and collecting water for the arid shelter. Most of the 

area immediately surrounding the site is covered by grass and ferns, and only supports a few trees. Other than 

that, Pager observed a number of artefacts, including an end scraper, a concave scraper and a grindstone with 

two potsherds (Pager 1971, p. 66).  

 

 

  



26 
 

Chapter 2: Secondary Archive 
 

In this chapter, the primary archive coalesces into the secondary archive, with a continued focus on Sorcerer’s 

Rock and the motifs that the main rock art panel features. This chapter first explores the history of rock art 

recording and traces the history of recording in Didima Gorge prior to Pager. This history shows something of 

the ways in which people have interpreted and understood rock art over time. Thereafter, my focus shifts 

toward Pager’s recording campaigns, his copying method and the archive it generated. The detailed copies 

Pager produced afford him the title of the greatest recorder of southern African rock art (Blundell, 2018, pers. 

comm.). The archive pertaining to Sorcerer’s Rock is explored in greater detail, in the form of a ‘visual essay’ 

that provides insight into the wider documentary archive. I provide a denotative and connotative reading of 

each figure based on insights attained from academic scholarship. Although I pay homage to Pager, his life and 

his work through the archive, I am critical of all my findings and am cognisant that there are many errors in his 

copies (Blundell, 2017, pers. comm.). 

Pager and his wife visited the shelter several times between February and March 1968 and spent a total of 30 

hours copying and recording the rock art. Some of the rock paintings were executed on the low ceiling while 

others appear on the fallen rock lying in the middle of the shelter. It is in the middle of the site that the lowest 

figures are painted, approximately 20cm above floor level. Of the 43 painted images at the site, the depictions 

on the ceiling are well preserved, and are enlivened during the day by reflected spots of sunlight cast on the 

shelter floor. The name Sorcerer’s Rock was chosen, apparently by Pager (although he doesn’t say so explicitly), 

because of one unusual painting of a crouching, white human figure depicted frontally with elaborate reddish-

brown painted patterns on its entire body, who is seemingly engaged in esoteric activities. However, the figure 

does not necessarily depict a sorcerer or medicine man, even 

though several early researchers ascribed this identification to this 

type of figure at other sites (Frobenius 1931, p. 22 and Goodall 

1959 cited in Pager 1971, p. 66). Based on similar depictions 

elsewhere, the current interpretation would not necessarily see it 

as depicting a sorcerer in any literal sense. In the chapter that 

follows, I explore the current interpretations of this figure further 

through a review of previous scholarship. Apart from the sorcerer 

figure that has attracted some attention, there are several other 

painted images and elements on the rock panel. I could have 

begun my visual analysis of the rock art from any given point on 

the panel, but chose to begin with the peculiar figure on the left-

hand side that is displayed amid and is encompassed by varied 

lines in different directions. I read the panel as a narrative through 

the ways in which the brown contour lines connected each figure 

to create a ‘story’ about the panel. I finish my visual analysis with 

the interestingly shaped group of figures who appear to be 

returning from a painted red daub over a black crack in the rock 

surface towards the right edge of the panel.  

 

In a short obituary, Ione Rudner noted that Pager (1923–1985) was 

born in Czechoslovakia and went on to live and study in Austria to 

become a graphic artist, and later became a world-renowned archaeologist through his work on rock paintings 

located in South Africa and South West Africa (now Namibia) (Rudner 1985). After emigrating to South Africa in 

Figure 11: A scanned and cropped image of the 
crouching white figure/ Sorcerer. From Pager 
(1971, p. 164). 

FIGURE 11 



27 
 

1955, he initially began working as an industrial designer and soon thereafter became interested in rock art. 

Prior to 1960, he visited and recorded briefly rock art sites in Tanzania and Zimbabwe. During the latter part of 

the 1960s, Pager — along with his wife, Shirley-Anne — researched and recorded 17 rock shelters in the 

Ndedema (now known as Didima) Gorge in the Natal (now KwaZulu-Natal) Drakensberg. For four years, the 

Pagers lived and worked in rock shelters, spending an estimated 2 168 hours recording rock art in the Didima 

Gorge. They recorded every painted motif and compiled a significant amount of data on the area in relation to 

its history and topography, including almost everything that lived and lurked there, its flora and fauna, and its 

archaeology (Rudner 1985). Of the 17 recorded shelters, however, four did not have any art (Mazel 2011, p. 

284).  

I chose to focus on Sorcerer’s Rock because of the manageable scale and the apparent ‘narrative’ coherence of 

the shelter’s main panel — the compositional coherence created through the denotative and connotative 

analysis of each figure on the main rock area panel that tells a story in itself. Although researchers have 

examined the paintings at Sorcerer’s Rock to some extent (e.g. Frobenius 1931, Goodall 1959, Solomon 1994 

and Thorp 2015), they have considered the central and most elaborate figure on the panel more or less in 

isolation instead of in the context of the entire composition, or the site as a whole. Here, I discuss the Sorcerer’s 

Rock archive in greater detail. I provide a summary of what previous rock art researchers have proposed 

regarding the central figure and go on to discuss the entire panel in the form of a ‘visual essay’, as a small sample 

that could provide insight into the wider documented archive. I first present an overview of rock art recording 

in the Drakensberg to establish a context for the consideration of Pager’s work, and go on to reflect on his 

recording campaign and methods.  

 

History of rock art recording and interpretation in the Drakensberg 

It is challenging to write such a piece, because it is a synthesis of a great deal of work that draws from archival 

sources that extend beyond the scope of this report. However, my intention in the following paragraphs is to 

establish a context within which to situate the work that Pager did. He initiated a method that was quite 

different from those used by his predecessors, involving a more comprehensive, layered colour copying 

technique that embodied observations gained from long hours in the presence of the paintings. Furthermore, 

Pager’s copying technique has not really been replicated since. 

The recording of rock art in the Drakensberg began with Sir Henry Bulwer in 1863. After Bulwer came across 

several studies about the culture of the Bushmen who lived in the Cape by Wilhelm Bleek and Lucy Lloyd (1911), 

and in what was then Basutoland by Joseph Orpen (1874), his interest in rock art was sparked, and he requested 

a father and son, Mark and Graham Hutchinson, with artistic abilities to copy the paintings at Giants Castle 

(Wright & Mazel 2007, p. 51). Most of Pager’s subsequent predecessors, like Helen Tongue, George Stow, Henry 

Breuil, Clarence Van Riet Lowe and Walter Battiss (to name just a few of those who recorded rock art in the 

wider region), also preferred traditional modes of copying in the form of drawing, tracings and paintings. For 

example, Walter Battiss, who later became a celebrated South African artist, produced direct tracings and 

colour copies in the 1940s (Lewis-Williams 2000, p. 124), and wrote a series of books on South African rock art 

in which he analysed its form, style and sequence (Smith 2016, p. 133).  

Rock paintings, after Bulwer, were viewed as representations of scenes from everyday life, and interpretations 

of the illustrations focused on the hunting practices of the San community and the conflict they experienced 

with other communities (Wright & Mazel 2007, p. 52). Wright and Mazel suggest that the first sustained effort 

to record art in the Drakensberg was in the year 1910, when the increasing damage to the paintings through 

human interaction and through natural causes such as weathering prompted concern on the part of the 



28 
 

government of the time. Trooper Whyte was instructed to locate all the paintings within the area (Whyte 1910). 

The result of his research produced the first quantitative record of paintings over 37 sites and 1  041 images. 

Archaeology and art enthusiasts visited these rock art sites and debated the stylistic attributes, age, and 

aesthetic qualities of the paintings, as well as the identity of unusual figures. Their discussions were not, 

however, grounded in thorough research.  

Between 1910 and 1960, the recording of rock art slowly started to gain momentum again (Wright & Mazel 

2007, p. 52). Most scholars proposed that the rock art had been created purely for aesthetic reasons, in line 

with the philosophy of “art for art’s sake”. However, a handful of scholars recognised that the art made by the 

San hunter-gatherers could be imbued with a deeper meaning. After York Mason participated in Lawrence 

Herbert Wells’s 1932 expedition to Cathkin Peak and Cathedral Peak, he proposed that “the phase in which 

polychrome art was produced was a period of religious fervour” (Wright & Mazel 2007, p. 51 – 52). The 

formulation of Mason’s hypothesis was based on the subject matter of the paintings — hybrid animals, figures 

apparently wearing masks and various ritual scenes. During the 1950s, Alex Willcox and Patricia Vinnicombe 

published separately on the paintings in the Drakensberg (Wright & Mazel 2007, p. 52), and publications by 

Woodhouse, Lee, Pager and Lewis-Williams followed in the 1960s (Wright & Mazel 2007, p. 53).  

During the 1970s, tension between professional academics and people outside of the academy grew during 

what is sometimes referred to as a ‘revolution’ in the history of rock art interpretation. In a conversation with 

Lewis-Williams, he recounted what the climate of the period was like, with the conflict between the 

‘professionals’ and the ‘amateurs’. It was an era of new archaeology — a scientific archaeology and analytical 

archaeology (Lewis-Williams, 2018, pers. comm.): 

Archaeologists of this period opted for testing and hypothetical-deductive methods to analyse rock art. 

Up until then, rock art was, not entirely, but overwhelmingly the field of the amateur. The amateurs 

were the individuals who did not come from an academic background. Consequently, a new legislation 

was introduced about getting permits or a license to excavate, so unless you were a trained 

archaeologist, you could not take part in the process. Amateurs could carry buckets but could not 

excavate and that caused a lot of bitterness that led to a huge debate. Rock art people, being largely 

amateurs without any sort of academic training or background, were defaulted into the amateur 

category. Being grouped as someone who was not qualified, and incompetent did not sit well with most 

amateurs like Pager. This categorisation upset him enormously because he said he was a graphic 

designer prior to being a rock art recorder and what he was doing when recording was graphic design, 

so one could not typify him as an amateur. 

Lewis-Williams notes how the methodology of rock art recording is a fundamental issue worldwide, and 

questions the ways in which researchers should approach the rupestrian images of other cultures that were 

created in the distant past (1996, p. 41). His suggestion is to consider how ethnography, neuropsychology, and 

the rock art itself are intertwined, in order to formulate persuasive explanations of specific rock art images 

through meaning making, interpretation and understanding the context of “the social trajectories in which the 

image intervened” (Lewis-Williams 1996, p. 41). Lewis-Williams also mentions that stereotypes of the San 

played a role in the formation of colonial conceptions about southern Africa’s past (1996, p. 39). Thus, “southern 

African rock art and much rock art research have been influential in the construction and reproduction of 

stereotypes of the San” and remain largely misunderstood (Lewis-Williams 1996, p. 39). These perceptions 

based on stereotypes extend into the public realm and are further complicated by viewers who observe rock 



29 
 

paintings or copies of them. Essentially, when actively engaging with the work viewers tend to challenge and 

interpret what they see according to their existing cultural baggage and prejudice (Lewis-Williams 1996, p. 39).  

It was Vinnicombe, however, who first began to open up a more convincing approach in the interpretation of 

rock art in the Drakensberg with her 1976 book, People of the Eland, in which she “pioneered the systematic 

use of Bushman ethnography to explain southern African rock art” (Huffman 1983, p. 49). Thereafter in 1981, 

drawing from a wider range of ethnographic sources, Lewis-Williams published a book titled Believing and 

Seeing: Symbolic Meanings in Southern San Rock Paintings (Lewis-Williams 1981a),5 where he argued how each 

painting present on the shelter walls could be seen as a product of shamanistic belief systems, rituals and 

experiences (Wright & Mazel 2007, p. 53–54). He established a three-tiered “shamanistic cosmology” that 

focused on shamanistic beliefs, rituals and experiences (Wright & Mazel 2007, p. 54). Vinnicombe’s and Lewis-

Williams’s work drew on the beliefs of the hunter-gatherers who had dwelled in the southern Drakensberg and 

Giants Castle areas, respectively (Wright & Mazel 2007, p. 53).6 Studies done in these areas by Vinnicombe and 

Lewis-Williams — and by many other researchers subsequently — made use of Pager’s meticulous recordings 

of rock art at Cathedral Peak (Wright & Mazel 2007, p. 53). The role of the animal in San religious beliefs became 

a pivotal area of research (Wright & Mazel 2007, p. 53). From the 1970s onwards, Lewis-Williams explored the 

ritual and symbolic forms of the painted subjects in rock art from different parts of southern Africa. His 

exploration of and insight into rupestrian imagery derived from documented oral accounts of various 

indigenous people (Wright & Mazel 2007, p. 54).  

In spite of considerable theoretical and methodological developments over the nearly four decades since, the 

“overstated” role of the shaman, according to Dowson (2007, p. 49), has largely retained its dominance. This 

dominance was questioned by Anne Solomon, who proposed that the rock paintings were primarily concerned 

with mythology, and were only marginally concerned with the shamanistic rituals and allied religious beliefs 

that were suggested by the ethnographic historical approach (Wright & Mazel 2007, p. 54). In a sense, 

Solomon’s notions might be seen to productively dilute and complicate the approach Lewis-Williams had 

pioneered.  

There has also been some work on San painting technology. Wright and Mazel note that the pigment that 

hunter-gatherers dug out of the basalt mountains to make paint with, was believed to possess supernatural 

powers (2007, p. 56–57). Often, paintings would begin with a sacramental hunt (Wright & Mazel 2007, p. 57). 

San women prepared the pigment during a full moon, when they heated the basalt until red hot and thereafter 

ground it into a fine powder. Afterwards, eland blood was either mixed with the pigment or used in its raw state 

to paint with, as it was believed that the animal’s blood was a ‘storehouse’ of potency. The San used a variety 

of tools to apply the paint to the rock surface. These tools came in the form of small brushes made from 

feathers, reeds, quills, bones and wooden sticks with a sharp point (Wright & Mazel 2007, p. 57).  

 

                                                           
5 From my conversation with Lewis-Williams, I learnt that his book was in fact a published version of his PhD submission in 1977 

(Lewis-Williams, 2018, pers. comm.). 
6 According to Blundell, Vinnicombe and Lewis-Williams used two different archives, which led to two different emphases. Vinnicombe 

focused on the Natal archives, which were largely historical in nature, while Lewis-Williams focused on the Bleek and Lloyd material 
which was largely anthropological (2019, pers. comm.).  



30 
 

The Pager Archive through a focused study of Sorcerer’s Rock  

Figure 12: Lemishka Moodley, Merged images of 3-sections of Pager’s Archive, 2018, photograph. 

Pager’s multi-stage, mixed-media copying method, with its emphasis on completing as much of the work on 

site as possible, produced images that are uniquely recognisable among rock art copies. RARI houses a 

substantial number of Pager’s large-format copies that are hung up in storage. Some of these have already been 

digitised through high-definition scanning or photography, and are available upon request through the SARADA 

website, but some of the larger ones still await digitisation because of the technical challenges posed by their 

size. The digitisation of copies adds yet another layer, and creates a further remove from the primary archive.  

After Pager’s death in 1985, Woodhouse took the Didima collection and exhibited it in various places under the 

auspices of Murray & Roberts, the construction company for whom he worked. In the process of exhibition, the 

copies were damaged. Eventually they came to RARI at Wits. In the late 1990s, Murray & Roberts were 

approached by RARI to fund the restoration of the collection. This took a long time and the collection had to be 

sent to the restorers twice. A lamentable decision was made to cut the panels for preservation purposes. 

Murray & Roberts also funded an archive room in the old Archaeology building at Wits to house the collection. 

Corporate involvement here was thus essential to the creation of the archive (Lewis-Williams 2018, pers. 

comm). This story about the archive is told in greater detail in the following chapter. 

From the results of the fieldwork conducted at Didima Gorge, Pager published a book titled Ndedema: A 

documentation of the rock art paintings of the Ndedema Gorge (1971). According to Rudner (1985), his book 

was produced with integrity, and was enormously significant, as it presented the most comprehensive study of 

rock art at the time. In spite of suffering personal financial difficulties because of his devotion to the Ndedema 

project, Pager persevered and devoted all his time to rock art. He went on to write another book titled Stone 

age myth and magic (1975) and published numerous papers and articles (‘The Antelope Cult of Prehistoric 

Hunters of South Africa’ (1975); ‘Quantitative Analyses Elucidate the Motives of South African Painters’ (1976); 

‘The Rating of Superimposed Rock Paintings’ (1976); ‘The Paintings of the Upper Brandberg’ Part I–IV (1989, 

1993, 1995 & 2006 respectively); ‘San Trance Performances Documented in Ethnological Reports and Rock 

Paintings of Southern Africa’ (1994)), which afforded him the opportunity to deliver many lectures in South 

Africa and abroad (Rudner 1985).  

FIGURE 12 



31 
 

The purpose of this section is to provide an overview of the scholarship that has dealt with the rock art at 

Sorcerer’s Rock in addition to some focused analysis of the single figure that the site was named after. I do not 

purport to provide a comprehensive interpretation of the panel, which is a substantial project in and of itself 

needing to draw from dozens of additional sources, rather this overview tracks something of the history of 

copying and interpretation, from Pager through to the present day.  

FIGURE 13 

Figure 13: Image of mythological painted figure. From Willcox & Pager (1967).  



32 
 

More recent scholarship goes into closer and more substantiated detail in identifying and interpreting the 

‘sorcerer’ figure. Pager’s interpretation of the ‘mythic being’ is unknown in the formal sense of research and 

writing. In ‘“Mythic Women”: A Study in Variability in San Rock Art and Narrative’ (1994) followed by ‘Rock Art 

Incorporated: An Archaeological and Interdisciplinary Study of certain Human Figures in San Art’ (1995), Anne 

Solomon claims that the unusual human figure depicted can be categorised as a mythic being or, more 

specifically, a mythic woman (1996, p. 33). She considers five attributes specific to this motif type for the figure 

to be categorised accordingly: “raised arms, an object or objects brandished aloft (bow, stick(s), crescent), 

splayed legs, an ‘emission’ or genital emphasis” and varied exaggerated or swollen torsos (Solomon 1996, p. 

33). A. J. B. Humphreys challenged Solomon in his 1996 paper titled, ‘Mother Goddesses and Mythic Women: 

An Alternative View’, where he suggests that the figure is male (Humphreys 1996, p. 32). Solomon (1996, p. 33) 

had actually pointed out that the ascription of gender based on body attributes is ambiguous in rock art research 

and has, consequently, been considered by association with tools the figures are surrounded by or carry. Such 

observations are based on San ethnographic accounts which describe how, for example, women were not 

allowed to touch men’s hunting equipment because they would contaminate it (e.g. Silberbauer 1981, in 

Solomon 1996, p. 33). Nevertheless, the possibility of women with hunting equipment should not be 

disregarded, and some ethnographical records of rituals do involve women using hunting equipment (Solomon 

1996, p. 33). Solomon also concurs with Humphreys’ proposition that the figure could also be male because it 

has an infibulated penis (Solomon 1996, p. 34). In a sense, Humphreys’ view encouraged Solomon to nuance 

and complicate her perception and interpretation about a clearly gendered ‘mythic woman’ based, among 

other sources, on a |Xam account about San ethnographic practices and a sorceress shooting invisible arrows 

translated by Bleek and Lloyd7 (1911, n.p., in Solomon 1996, p. 34). Moreover, during ritual practices the 

biological sex of the individual who is performing or is a part of the ritual may be temporarily transformed or 

transcended (Lewis-Williams 1981b, p. 72, in Solomon 1996, p. 34). A label of ‘sorcerer’ is therefore not an 

entirely inappropriate descriptor for this figure in that it could possibly be ‘gynandromorphic’ (hermaphrodite) 

(Stevenson 1995, p. 109–10, 17, in Thorp 2015, p. 182).  

Carolyn Thorp, on the other hand, in ‘Rain’s things and girls’ rain: marriage, potency and frog symbolism in |Xam 

and Ju ’hoan ethnography’, suggests that the figure at Sorcerer’s Rock could be a therianthrope with frog-like 

features (2015). The platanna (African clawed frog; Xenopus laevis) has similar genitalia when compared to the 

‘sorcerer’ (Thorp 2015, p. 182). This important anatomical characteristic is a clue that points to the symbolism 

of frog-like features, and is not only reminiscent of female genitalia but also looks like the large red and round 

protruding bulges between the ‘sorcerer’s’ thighs (Thorp 2015, p. 182). A link between the frog imagery and 

that of the figure at Sorcerer’s Rock hints at the idea of reproduction (Thorp 2015, p. 182). Thorp agrees with 

Solomon’s identification of features as being relatable to the theme of ‘mythic women’ who emit fluid from 

their genitals (referred to as menstrual blood) and who carry links to female initiation rituals (Solomon 1996, p. 

34). Thorp builds further on Solomon’s interpretation by suggesting that the depicted paintings “refer to 

potency […] and the necessity to control potency through shamanistic activity” (Stevenson 1995, p. 114, 116, 

in Thorp 2015, p. 183). Ultimately, a connection between the features of the painted figure and Bushman beliefs 

around blood and potency can determine aspects of the sex/gender through an examination of the genitalia 

(Thorp 2015, p. 183). Yet, Solomon’s use of Lewis-Williams’s view puts it best: “while some images may have 

been iconic, they are better understood as ‘icons’ fulfilling a symbolic function” (Lewis-Williams 1981b, p .6b, 

in Solomon 1996, p. 35). Hence, there is often a disjuncture between the literal appearance of an image and its 

potential meaning or symbolism.  

As this history of interpretation shows, the ‘sorcerer’ can be argued to be many different characters based on 

the intertwined narratives of San historiographical, ethnographical, and mythological accounts. The importance 

                                                           
7 For more information see The Digital Bleek & Lloyd. ONLINE. URL: http://lloydbleekcollection.cs.uct.ac.za/stories/717/index.html. 
Date Accessed: 17/03/2018.  

http://lloydbleekcollection.cs.uct.ac.za/stories/717/index.html


33 
 

of Pager’s work arguably prompted detailed inquiry into and consideration of the sorcerer figure within a wider 

anthropological context, if not so much its visual context. As evident above, even though Pager himself did not 

have much to say about the sorcerer figure, he drew attention to it, and more recent interpretations have built 

on aspects of his interpretations and recordings. Furthermore, he was clearly interested in the wider pictorial 

and landscape context of this figure. Although he may not have expressed this much in words, this becomes 

clear in an examination of the copies he made and in the detailed documents about the specifics of each site. 

In the context of this research report particularly, Pager’s archive also became a point of departure to extend 

my understanding of theoretical threads of archive, landscape and narrative that I chose to explore to explain 

and illustrate some of the figures in the main rock art panel. I expand my visual analysis in the form of a visual 

essay that points towards the potential richness of an analysis of the other figures in the panel. 

 

FIGURE 14 

Figure 14: ARACHNE website 



34 
 

Pager’s recording method 

 

The history of interpretation reflects the history of copying in relation to understandings around the function 

or meaning of the art that was generated in its first context. Prior to Pager, rock art recording was viewed as a 

process of recording and documenting visual works of art and their surrounding environment through 

photographs and illustrations.  

Acknowledging the difficulties that many recorders before him had experienced when trying to record rock 

paintings, Pager opted to develop a more labour-intensive copying method from his own particular skills and 

interests arising from his training as a graphic designer. He described in some detail what this method entailed 

in his book Ndedema (1971, pp. 81–82). The painted rock faces were photographed from right-angle camera 

positions on a 6 x 6cm or 6 x 9cm black-and-white film. A tape measure, attached to the rock, was included in 

each frame. Where the walls of the shelters were even, areas of c. 1m² each were photographed in a way that 

adjoining panels overlapped with each other by a few centimetres. Where rock faces were curved, or consisted 

of small panels facing various directions, a series of photographs had to be taken in order to avoid perspectival 

distortions (Pager 1971, p. 81). 

He employed photography as a starting point, while also thinking carefully about the strengths and weakness 

of the photographic medium. Although from one angle photography provides a realistic representation of a 

captured moment, Pager viewed it as significantly less accurate than what could be seen by the naked human 

eye. Photography still leads to some confusion between the texture of the rock and the grain of the photograph, 

as well as the colour of the paint versus the natural colorations of the rock — subt