By Janette Deacon, 2002
in collaboration with members of the Southern African Rock Art Project (SARAP)


ROCK ART by hunter-gatherers, herders, and/or later farming communities occurs in almost all countries in Africa. There is, however, a distinctive set of Southern African traditions, with regional and temporal variation, in Tanzania, Malawi, Zambia, Zimbabwe, Angola, Namibia, Botswana, Mozambique, South Africa, Swaziland, and Lesotho. In essence, Southern African rock-art traditions record experiences related to the belief systems and rituals of the indigenous people of the region. Sites representative of these traditions are worthy of consideration for inclusion on the World Heritage List for the following reasons:

  • Many individual sites and images - both paintings and engravings - are masterpieces of human creative genius that illustrate a combination of sophisticated ideas and beliefs, exquisite and unusual detail, extraordinary imagination, and artistic mastery of the chosen media;
  • Collectively, over a period of nearly thirty millennia in the subcontinent, the artists recorded significant interchanges of human values - particularly with respect to religious and ritual practices - that cannot be recovered from stone artefacts and other inanimate remains;
  • There is excellent ethnographic information available from indigenous people in certain key areas that has assisted in the understanding and authentic interpretation of the meaning and motivation of the art - a feature that is missing in many other regions of the world;
  • Paintings and engravings of successive traditions have been done at selected places and areas over a long time period and the integrity of this relationship is still intact, incrementally adding to the tangible and intangible significance and power of these places and the landscape; and
  • The shamanistic inspiration for much of the art demonstrates the time depth and nature of the human quest for supernatural power in this part of the world.

Recognizing the importance of their rock art, cultural heritage conservation authorities in Southern African countries, coordinated informally through the Southern African Rock Art Project (SARAP), have collaborated since 1996 in identifying and nominating a representative sample of rock art in the region for the World Heritage List. As part of this co-operative programme, training courses and workshops have been held to assist the responsible heritage institutions to draft nominations, draw up management plans, and train the staff responsible for management of the sites.

This report is the result of consultation at SARAP meetings in South Africa, Tanzania, Zambia, and Zimbabwe with representatives from most of the Southern African countries. Of the countries that attended the meetings, Lesotho has not yet signed the World Heritage Convention. Swaziland was invited but did not send a delegate, and is in any case not yet a signatory to the Convention. Angola attended only one meeting, in 1998, and although invitations were sent to attend subsequent meetings, no reply was received.

There are sufficient contrasts between areas of high concentration of rock art to warrant the nomination of a series of sites for World Heritage listing. Initially it was hoped to present all the sites together as a single serial nomination to the World Heritage Committee. Practical problems have developed, however, and each State Party is responsible for submitting their nomination(s) when they are ready to do so. At the time of writing, the rock art of the uKhahlamba Drakensberg Park (South Africa) and of Tsodilo (Botswana) have been inscribed, and nominations by Malawi (Dedza-Chongoni), Tanzania (Kondoa- Irangi), Zimbabwe (the Matopos), Zambia (Kasama), and Namibia (Twyfelfontein) are in various stages of preparation.

Ultimately, the selected rock-art sites will represent the range of variation in the region and symbolically re- connect the artistic heritage of Southern Africans for the first time since the colonial era. We present here an overview of the rock art in the region and propose criteria for the evaluation of nominations of Southern African rock- art sites for the World Heritage List.


II.1 Distribution

A SUMMARY OF ROCK ART databases in Southern African countries indicates that there are at least 14 000 sites on record, but that many more exist than have been formally recorded (Deacon 1997). There are probably well in excess of 50,000 sites in the region as a whole, with a conservative estimate of more than two million individual images. With the exception of the uKhahlamba Drakensberg Park in South Africa, Tsodilo in Botswana, and the Brandberg in Namibia, few areas have been thoroughly searched and recorded. The densest known concentrations of rock art occur in parts of Lesotho, Namibia, Botswana, South Africa, Zambia, and Zimbabwe The lowest numbers of recorded sites are in Angola, Malawi, and Mozambique.

The region has both rock engravings (petroglyphs) and rock paintings (pictographs). There are no reliable records to indicate the relative percentage of paintings to engravings, but painting sites are probably in the majority. In general, both paintings and engravings have similar themes and images, but the engravings tend to include less detail and fewer human figures.

The distribution of the two techniques is largely governed by geology. Engravings occur out in the open and are usually, but not exclusively, associated with igneous rocks such as dolerite. Such rock formations tend not to form shelters or caves. Rock paintings, on the other hand, are most common in areas where there are caves or rock shelters in outcrops of granite and in sedimentary rocks formations of limestone, sandstone and quartzite. It is rare, but not unknown, to find both rock paintings and engravings together at the same site.

II.2 Traditions and styles

IN BROAD TERMS, there are three rock-art traditions in the region with distinctive styles and content that are largely the result of differences in the cosmology and beliefs of Stone Age hunter-gatherers, of Stone Age herders, and of Iron Age agriculturists. Within these traditions, and often cross-cutting them, there are further differences in the techniques used to paint and engrave (for a general overview see Lewis-Williams 1983; Coulson & Campbell 2001).

II.2.i Paintings
As elsewhere in the world, the most common pigment used for rock paintings is red ochre, with some paintings in maroon, yellow, black, and white. There is some ethnographic evidence that the pigment was mixed with a variety of binders such as blood, egg, fat and plant juices, but the exact recipes are not known (Lewis-Williams 1983). The techniques applied in the majority of paintings can be summarized as follows:

  1. Fine-line paintings, almost exclusively the work of hunter-gatherers, in red or yellow ochre, white clay, or black charcoal or manganese oxide, done with a brush or other fine instrument, using techniques such as the following:
    • Outline of the image with a single line (rare everywhere);
    • Outline of the image with the interior filled with lines of the same colour (characteristic of Tanzania (Anati 1986), with some examples elsewhere);
    • Monochrome image with the colour blocked in (most common almost everywhere);
    • Outline in one colour with the image filled in with another slightly different colour;
    • Bichrome, in which two blocks of colour are used in the same image;
    • Polychrome in which three or more colours are used in the same image (most common in Namibia, Zimbabwe, South Africa and Lesotho);
    • Shaded polychrome in which several colours blend into each other to create depth and shading (most common in Lesotho and the Drakensberg region of South Africa);
    • Handprints and finger-dots
  2. Paintings done with a finger or very broad brush or applicator, most often by herders and agriculturists, often bold and highly stylized designs that include domesticated animals, in:
    • Monochrome red, white, or black (yellow rare);
    • Bichrome (rare);
    • Handprints, both plain and decorated.

II.2.ii Engravings
Rock engravings in all traditions were made most commonly by removing the weathered outer surface of rocks such as dolerite to create a colour contrast with the underlying unweathered rock (Dowson 1992). This could be achieved by using another harder rock such as quartz or chalcedony. The weathered surface was either scraped away over the whole area of the image, or the image was outlined in a single line, or the weathered rock was pecked out in a chopping motion. Some engravings show extraordinary artistic skill with careful details of skin folds, eyes and posture delicately portrayed. Engravings are also found on rocks where there is little or no colour contrast but where a large flat expanse is exposed, for example on ancient glacial pavements in the Northern Cape Province in South Africa.

So-called cupules and grooves, sometimes in regular patterns, have been made in relatively soft rock types where the colour contrast between the surface and underlying rock has been less important than the granularity of the rock. Many of these were probably not 'art', although they may have had a ritual use.

II.3 Dating

THE CREATIVE SPIRIT that inspired the engravings and paintings on the rocks of Southern Africa dates back at least 27,500 years and persisted in some areas until the 20th century AD. The oldest date is the average calculated from fifteen radiocarbon dates on charcoal from an occupation layer in Apollo 11 cave in southern Namibia. Seven small painted slabs, on rock that was not derived from the cave wall or floor, were recovered during two excavations in 1969 and 1972 (Wendt 1976). The next oldest date is from the Matopos in Zimbabwe, where a spall that had flaked off the painted wall in the Cave of Bees was found incorporated in the deposit in the floor. Charcoal from the relevant layer was dated to about 10,500 BP, giving a minimum age for the original painting (Thackeray 1983).

The oldest dated rock engravings in the region come from Wonderwerk Cave in the Northern Cape, South Africa. Five small slabs with clearly identifiable fine-line rock engravings of animals and non-representational geometric patterns, and another six with lines that may have been part of engravings, were found in different layers. Associated charcoal dated the oldest to about 10,200 BP and the youngest to about 2000 BP (Thackeray 1983). Cation ratio dates on rock engravings from open sites in South Africa also gave preliminary results of between 10,000 and 2000 years (Whitley & Annegarn 1994). Relative dating of the weathered crust on engraved surfaces on a small sample of South African rock engravings suggests that the fine- line style is the oldest, and that the pecked and scraped engravings were done more recently.

Geometric zigzag patterns, combined to make diamond shapes and engraved on two small pieces of hard ochre, have been found in a cave at Blombos on the southern coast of South Africa; associated with Middle Stone Age artefacts. A layer of dune sand overlying the deposit has been dated at 77,000 years ago by optical luminescence. Although the find is claimed by some to be the earliest art (Whitfield 2002) the archaeologists who excavated the site argue only that people at that time were capable of abstract thought and design (Henshilwood et al 2001). Similar designs are found on bone artefacts and ostrich eggshell dating within the last 18 000 years (Deacon & Deacon 1999), but are not common in either rock paintings or rock engravings.

Most of the art was done by hunter-gatherers whose traditions persisted in south-eastern South Africa until the 19th century AD (Lewis-Williams 1990). In some countries, such as Tanzania (Anati 1986), Malawi, and Mozambique, where there are no reports of hunter-gatherer style paintings that include domesticated animals or images of the colonial era, the hunter-gatherer art is estimated to be older than AD 1000 (Smith 1997).

Within the last 2000 years, herders and Iron Age agriculturist people entered the region from the north and added to the corpus of rock art with different styles and content. The oldest art in these traditions is generally thought to be less than 1500 years old and the most recent paintings and engravings were done in the late 19th and early 20th centuries. In some areas in South Africa, Zambia, and Malawi, agriculturist art was still being practised for initiation rituals in the late twentieth century (Smith 1995).

II.4 Ethnography

THERE IS A WEALTH of ethnographic information from the 19th and 20th centuries that can and has been used successfully to interpret many of the metaphors and symbols in hunter- gatherer and later agriculturist art dating back hundreds and even thousands of years. This information gives an unparalleled insight into the meaning and context of rock art. An understanding of altered states of consciousness, and its role in shamanism and rock art as developed by researchers David Lewis-Williams and Thomas Dowson (Lewis- Williams 1990; Lewis-Williams & Dowson 1989), has in turn helped to identify the source of some of the metaphors found in the Palaeolithic rock art in western Europe (Clottes & Lewis-Williams 1996). With the exception of Australia and North America, very few rock-art regions elsewhere have such detailed sources for interpretation.

The ethnographic records provide considerable evidence in Southern Africa that hunter-gatherer rock paintings and rock engravings were part of religious practices for rain- making, healing, and other shamanistic activities such as out-of-body travel and the control of game animals. These practices involved altered states of consciousness that enabled medicine people or shamans to access supernatural power through certain animals or through ancestral spirits. The wide distribution of this rock-art tradition, from South Africa to Tanzania, provides evidence for a broad high-level similarity in the cosmology of Southern African hunter-gatherer peoples. There are nevertheless important regional differences in the way that shamanistic experiences were perceived and used and in the metaphors that were transferred to the art.

For example, the beliefs of the /Xam San that were recorded in the 1870s (Bleek & Lloyd 1911; Vinnicombe 1976; Lewis- Williams 1981, 1990) led to an understanding in the 1970s of the reason why the eland was the animal most commonly depicted in the rock art in the south-eastern region in South Africa. The work of anthropologists and psychologists amongst 20th century San in Botswana and Namibia enabled some of the /Xam records to be interpreted and better understood (Marshall 1976; Katz 1982; Lee 1984; Guenther 1986; Biesele 1993). By combining the information from these diverse sources, Vinnicombe (1976) concluded that the eland was the pivot around which the social organization and beliefs of the Drakensberg San revolved. Lewis-Williams (1981) described how the eland played a key role in boys' and girls' initiation, and its role in healing and rainmaking because it was believed that associating with the eland could bring the medicine-person or shaman closer to god and supernatural power. Shamans in trance would feel as though they were transformed into eland, as is clear in many paintings of therianthropes with human and eland body parts that are combined in one image. In other areas such as Zimbabwe and Namibia, however, the eland is less common than animals such as the kudu and the giraffe. This suggests regional variation in the ritual significance of particular animals.

There is relatively little ethnography that has been applicable to herder rock art, and indeed there is only circumstantial evidence that the schematic and highly stylized art with a wider range of geometric patterns, best represented at Tsodilo (Botswana), was done by early herders. This art is attributed to herders because it includes domesticated animals, and because in places where it occurs with the earlier tradition it tends to be superimposed on what is clearly hunter-gatherer art (Coulson & Campbell 2001). Herder art is also found above and below hunter-gatherer art. A good example is the Limpopo Valley, where there was a brief period of interaction between the two groups in the 1st millennium AD (Hall & Smith 2000).

Agriculturist rock-art displays some general similarities within the region and is quite distinct from the hunter-gatherer art in several respects. It tends to be bolder, less detailed, more schematic, and with a smaller range of colours and subject matter. In some areas the paintings are called 'late whites' because where superimposition occurs they are always on top and they are done in white paint with a finger rather than a brush. Local traditions and ethnographic records in Zambia and Malawi (Smith 1995, 1997) indicate some of this art was part of secret male and female initiation practices and of rituals such as rainmaking. The meaning of the designs is known only to the initiated.

II.5 Content

THE CONTENT of the rock art varies within the region, but there are several themes that are sufficiently widespread to indicate broad, high-level geographical and temporal continuity within the Southern African hunter-gatherer, herder, and agriculturist belief systems over the period in which rock paintings and rock engravings were done. 
The most significant similarities are the persistent occurrence of illustrations of and metaphors for altered states of consciousness or trance experience, particularly in the art of the hunter-gatherers. These experiences are evident in the postures in which people are illustrated, the consistent selection of certain animals in preference to others, and in the presence and incorporation of geometric patterns that depict entoptic phenomena 'seen' during trance (Lewis-Williams & Dowson 1989).

In hunter-gatherer art people are shown in postures such as bending from the waist, lying down horizontally, or with contorted limbs. Additional details such as bleeding from the nose or a red line that sometimes has white dots emanating from the back of the neck or the feet may be seen, as well as the transformation of people into animals and animals into people. Processions of dancing people or groups of clapping women record the dances that helped shamans go into trance. Healers, shown touching sick people to draw out the arrows of sickness, may be associated with arrows, they may have animal heads, or may be sweating or bleeding from the nose. Rainmakers may be shown in close contact with large rain animals such as elephant, hippo, or eland, or with huge animals that are not identifiable to species. People in deep trance may be shown collapsing as if dead, or swimming underwater with fish or fish tails, or flying, with or without wings.

Images of transformed medicine-people with animal heads and other features (therianthropes) are especially detailed in paintings in the Drakensberg (South Africa and Lesotho), in Zimbabwe, and in the Brandberg (Namibia). Animals may also be shown with human hind legs. Therianthropes occur in parts of Angola, Botswana, Zambia, Mozambique, Malawi, and Tanzania, but are not as common as they are in the other Southern African countries.

Differences in the content of the art can be seen in the posture and dress of the people who are illustrated. For example, paintings of people in Namibia, Zimbabwe, and Tanzania have dramatic hairstyles that are not seen so often in the art further south. In contrast, there are many more paintings in the south of people wearing cloaks, some of which are elaborately decorated.

As noted above, there are variations in the frequency of certain animals depicted in the rock paintings and engravings of the region. These variations are not a mirror of the distribution of fauna, but an indication of the animals that the artists and their society regarded as significant in their religious and ritual practices. The eland predominates in the south-east, the eland and elephant in the south-west, the rhino is prominent in some areas, while in Zimbabwe, Namibia, and Tanzania the kudu and giraffe reign supreme (Garlake 1987).

In some parts of South Africa, Zambia, and Malawi the rock art is dominated by agriculturist paintings depicting symbols significant during initiation ceremonies and ritual practices. A stylized image of the crocodile or lizard is most common in some areas where it is a symbol of power, while in others domesticated cattle are more prevalent. Agriculturist art also records activities associated with colonial times such as trains, motor vehicles, aircraft, wagons, horses, camels, and other forms of transport.

Schematic designs illustrating entoptic phenomena with dots, zigzags, grids, wavy lines, nested u-shapes, concentric circles, sunbursts, and vortices are widespread and occur in all rock-art traditions, not only in Southern Africa but elsewhere in the world as well. In Southern Africa they are more common in rock engravings than in paintings, but are nevertheless seen in paintings throughout the region. In the hunter-gatherer tradition they are subtly integrated into the fine-line paintings and engravings; in the herder art they are often more common than images of people or animals; and in the agriculturist art they may be combined with other stylized designs. The presence of these entoptic patterns in all three traditions emphasises their connection with altered states of consciousness and, therefore, the link between the art and trance experience that persisted in various forms over a long period of time and through major changes in economy and beliefs.

II.6 Landscape setting

IN SEVERAL CASES in hunter-gatherer, herder, and agriculturist traditions, there is ethnographic evidence that rock art has been used to enhance the power and significance of particular places in the landscape. The paintings or engravings were placed there because it was a rainmaking or initiation site, adding intangible value to the place. In a few cases, such as Kasama (Zambia) and the /Xam heartland in South Africa, there are ethnographic records that explain the significance of the place. Some of the rock-art site nominations to the World Heritage list will be proposed as cultural landscapes or as mixed cultural and natural sites so as to emphasize this relationship between culture and nature that has not been affected unduly by economic or agricultural development. The integrity of the art within the natural landscape has therefore been retained. Good examples are Tsodilo (Botswana), the Matopos (Zimbabwe), the Brandberg (Namibia), and the uKhahlamba Drakensberg Park and the /Xam Heartland (South Africa) (see Coulson & Campbell 2001 for illustrations).


III.1 General World Heritage criteria

This document does not wish to pre-empt the criteria as set out in paragraph 24.a of the Operational Guidelines for the Implementation of the World Heritage Convention (UNESCO 1999) that may be cited by individual Southern African countries in support of their nominations of rock-art sites for World Heritage listing. However, the following could be addressed.

Criterion i           Creative genius

This criterion should apply not only to the work of individual artists represented at the site, but also to the collective genius of the society that developed the sophisticated symbolism, metaphors, and multiple meanings displayed in the art. The site should display outstanding examples of the art of the tradition, as well as exceptional talent in the execution of the paintings and/or engravings. Sites using this criterion should have a variety of exceptionally well preserved rock art in which details of content and technique can be clearly seen.

Criterion ii           Important interchange of human values

This criterion would apply if more than one rock-art tradition is represented at the site. The historical process that contributed to the technological and artistic development at the site should be described. Dating and other details of the succession from hunter- gatherer to herder and/or agriculturist presence in the area should be supplied from archaeological research, ethnography, written records and/or oral history. Significance does not necessarily go hand-in-hand with 'the oldest' or 'the most recent' or 'the largest number' or 'highest density' of sites.

Criterion iii           Unique or exceptional testimony to a cultural tradition

In applying this criterion, well documented ethnographic or historical evidence would be required to interpret the rock art in relation to the cosmology and beliefs of the society that created it. The art should not be seen as the only testimony to the cultural tradition. It formed an integral part of the social fabric at the time it was created and some evidence of the activities that inspired the art should therefore be presented to place the art in its historical and social context. Few sites in the region have direct ethnographic evidence of this context, but sufficient research has been done, and could still be done, to make valid comparisons and connections between living traditions and those that have disappeared over the last century in areas within the region. For example, the presence of entoptic patterns and trance postures in the art is a clear indication that the artists experienced altered states of consciousness and that this had a close connection with their artistic tradition. To make this testimony unique or exceptional, however, there would have to be unusual techniques used in depicting these experiences, unusual features in the content of the art, and/or outstanding or site-specific ethnographic or historical evidence that connects the art with the cultural tradition of a particular people.

Criterion iv           Outstanding example of a landscape which illustrates a significant stage in human history

This criterion could be used in cases where the integrity of the rock art within the landscape has been retained and no developments or agricultural activities interfere with the original setting. Ideally, this should be combined with oral history or ethnographic records that add intangible values associated with the rock art and that particular landscape (ie an associative cultural landscape as described in paragraph 44(b)(iv) of the Operational Guidelines).

Criterion v           Outstanding example of a traditional human settlement or land-use representative of a culture vulnerable under the impact of irreversible change

As there are no known traditional hunter-gatherer or herder communities in Southern Africa who practise rock art, this criterion would apply primarily to agriculturists. It would be very important to inscribe such an example as it could represent the only living tradition of rock art in the region. The nomination would have to include ethnographic and anthropological evidence from the community that rock art is still an integral part of their cosmology and beliefs that are associated with their traditional human settlement or land-use pattern.

Criterion vi           Direct or tangible association with beliefs and artistic works of outstanding universal significance

This criterion can be used in conjunction with others to emphasize the universal significance of Southern African hunter-gatherer rock art and related ethnography that has contributed to an understanding of the relationship between rock art and altered states of consciousness, including entoptic phenomena.

Paragraph 24.b of the Operational Guidelines also sets out other requirements relating to nominated sites and monuments. The first of these is the test of authenticity in design, material, workmanship and setting. In the case of rock-art sites, survey by a rock-art specialist of all sites within the area to be nominated must verify the authenticity in design, material, and workmanship. It is important to ensure that the setting remains largely intact and has not been altered significantly since the art was done, since the setting is likely to have contributed to the original meaning of the site. The setting can be destroyed by the inappropriate placement of protective measures and information boards and facilities, as well as by insensitive local development not directly connected to the rock art.

The second requirement relates to adequate protection and management. Some form of management plan is obligatory for all nominated properties. The management plan must include a full inventory of all sites within the area proposed for World Heritage listing. The inventory must at least have a description and photographs of the art at each site, directions for locating the place, and distribution maps. Ideally, it should also include a condition report by a rock-art conservation specialist for each site.

In assessing the appropriateness and effectiveness of protective measures and management mechanisms at rock-art sites, it is important to ensure that at least one trained rock-art conservation specialist is on the permanent staff of the organization responsible for management of the site. As several sites to be nominated are within nature reserves or parks, the management personnel are trained only in nature conservation. This training is inadequate for the long-term monitoring, conservation, and management of rock art.

Interpretive displays and publications explaining the meaning and significance of the rock art and how to behave at the site must be available if the site is open to the public.

The nomination must confirm that descendant communities of the original artists have been consulted regarding the meaning and motivation behind the art. If such communities have been identified, they must be included in decision- making for, and management of, the site.

III.2 Specific criteria for Southern African rock-art sites

IN EVALUATING NOMINATIONS of rock-art sites in Southern Africa for World Heritage listing, the following criteria could be applied.

III.2.i Representivity
The site may:


  • be representative of a defined or definable tradition of rock paintings or rock engravings in the Southern African region, in that it includes many of the characteristics of that tradition;
  • have outstanding examples of a painting or engraving style, technique, or method; and/or
  • have the only (or some of the few) examples known of a rare tradition, style, technique, or method.

III.2.ii Quality of rock art
The rock art may:

  • include images of outstanding technical quality;
  • have superb examples of artistic skill and detail;
  • have excellent examples of graphic design and composition;
  • be exceptionally well preserved; and/or
  • have several panels of paintings or engravings that clearly depict unusual themes or images.

III.2.iii Setting, size of area, and density of rock-art sites
The area that is nominated should:

  • be large enough to display outstanding examples of rock art in an environmental setting that is as close as possible to that which existed at the time the paintings or engravings were done;
  • retain much of its original ambience;
  • be definable in terms of its outstanding environmental or cultural characteristics; and/or
  • include at least 50 rock-art sites, OR, if only a few sites are nominated, should include well over 100 individual images.

III.2.iv Association between rock art, ethnography, oral history, and/or ritual practices
A concerted effort should have been made to:

  • document interpretations of the rock art by conducting oral history surveys amongst indigenous local communities and/or ancestral communities and undertaking a thorough search of relevant ethnographic and anthropological sources;
  • identify recurrent images in the rock art and establish their significance in current, recent or ancient practices; and
  • liaise with local communities when developing the conservation management and tourist management plans.

III.2.v Quality of research and documentation
The nomination dossier must show that:

  • a purposeful survey of the rock art in the area in order to find and record as many sites as possible has been undertaken, preferably within the last ten years;
  • there is a record form for each site with location details, a description of the images, a condition report, as well as tracings and/or photographs;
  • a map (at least 1:50,000 and preferably 1:10,000) records the location of each site;
  • the records are kept in a safe place and copies have been made for use during monitoring; and
  • copies of publications and reports on rock art and related research in the area are kept in the institution responsible for site management. Dating
Statements about the dating of the rock art must be substantiated by good corroborating evidence. Age alone should not be the main criterion for nomination.

III.2.vii Management plans for conservation and tourism
It is essential that:

  • every nomination be accompanied by an up-to-date conservation management plan that is aimed specifically at retaining the significance of the rock art and its setting;
  • there is a written commitment to, and budget for, the training of personnel responsible for the day-to-day management of the site;
  • a visitor management plan is included; and
  • regular monitoring of the rock art by trained personnel is a stated priority.


A representative sample of excellent rock art sites has been identified by representatives of official heritage organizations in Southern Africa for nomination to the World Heritage List. More may be added later, but for the time being they serve to highlight the artistic quality, technical range, sophisticated cosmology and beliefs, detailed ethnography, landscape setting, and universal significance of the rock art of the region.

Criteria for the evaluation of Southern African rock-art sites that may be nominated for the World Heritage List have been suggested, based on the exceptional qualities of the art in the region.

From north to south, the rock-art sites identified by SARAP members as suitable for nomination to the World Heritage list are summarized in Table 1. Angola, Swaziland, and Lesotho have yet to select sites. As far as we are aware, all those in the Table have been placed on the tentative lists by the countries concerned. The list includes both rock engravings and paintings in the hunter-gatherer, herder, and agriculturist traditions, with a variety of techniques, content, and settings.

Table 1: Summary of rock-art sites in Southern Africa for possible nomination to the World Heritage List
Country Site Tradition Regional
(March 2002)
Tanzania Kondoa-Irangi District Hunter-gatherer Paintings with unusual technique and content Survey and management plan in preparation
Malawi Dedza-Chongoni District Agriculturist Paintings with good ethnographic detail Nomination in preparation
Mozambique Manica Province Hunter-gatherer Well preserved paintings Awaiting detailed survey
Zambia Kasama District Hunter-gatherer and agriculturist Well preserved paintings with good ethnographic detail Awaiting detailed survey and management plan
Zimbabwe Matopos National Park Hunter-gatherer High quality, well preserved, detailed paintings in natural landscape Management plan and nomination in preparation
Botswana Tsodilo Herder Large number of well preserved paintings in natural landscape Nomination submitted June 2000, inscribed December 2001
Namibia Brandberg Hunter-gatherer Large number of high quality well preserved rock paintings in natural landscape Excellent and detailed survey completed; management plan required
Twyfelfontein Hunter-gatherer Large number of high-quality well preserved rock engravings in natural landscape Partial survey; management plan required
South Africa uKhahlamba Drakensberg Park Hunter-gatherer Large number of high quality, detailed and well preserved rock paintings in natural landscape Nomination submitted June 1999; inscribed 2001
/Xam Heartland Hunter-gatherer Cultural landscape with excellent ethnographic records and high quality rock engravings Detailed survey and management plan required



Anati, E 1986 The rock art of Tanzania and the East African sequence. BCSP 23: 15-68.

Biesele, M 1993 Women like meat: the folklore and foraging ideology of the Kalahari-Ju/'hoan. Bloomington and Indianapolis: Indiana University Press.

Bleek, W H I, 1911 Specimens of Bushman folklore. London: George Allen. & Lloyd, L C

Clottes, J, & Lewis-Williams, J D 1998 The Shamans of Prehistory: trance and magic in the painted caves. New York: Abrams.

Coulson, D, & Campbell, A 2001 African rock art. New York: Abrams.

Deacon, H J & Deacon, J 1999 Human beginnings in South Africa. Cape Town. David Philip.

Deacon, J 1997 A regional management strategy for rock art in Southern Africa. Conservation and Management of Archaeological Sites 2: 29-32.

Dowson, T A 1992 Rock engravings of Southern Africa. Johannesburg: Witwatersrand University Press.

Dowson, T A. & Lewis-Williams, J D (eds) 1994 Contested images: diversity in Southern African rock art research: Johannesburg: Witwatersrand University Press.

Garlake, P 1987 The painted caves - an introduction to the prehistoric art of Zimbabwe. Harare: Modus.

Guenther, M 1986 The Nharo Bushmen of Botswana: tradition and change. Hamburg: Helmut Buske Verlag.

Hall, S, & Smith, B 2000 Empowering places: rock shelters and ritual control in farmer-forager interactions in the Northern Province. South African Archaeological Society Goodwin Series 8: 30-46.

Henshilwood, C S, et al. 2001 Emergence of modern human behaviour: Middle Stone Age engravings from South Africa. Science. (details not known at time of writing)

Katz, R 1982 Boiling energy: community healing among the Kalahari !Kung. Cambridge, Mass.: Harvard University Press.

Lee, R B 1984 The Dobe !Kung. New York: Holt, Rinehart & Winston.

Lewis-Williams, J D 1981. Believing and seeing. London: Academic Press.

Lewis-Williams, J D 1983 The rock art of Southern Africa. Cambridge: Cambridge University Press.

Lewis-Williams, J D 1990 Discovering Southern African rock art. Cape Town: David Philip.

Lewis-Williams, J D, & Dowson, T A 1989 Images of power: understanding Bushman rock art. Johannesburg: Southern Book Publishers.

Marshall, L 1976 The !Kung of Nyae Nyae. Cambridge, Mass.: Harvard University Press.

Smith, B W 1995 Rock art in south-central Africa: a study based on the pictographs of Dedza District, Malawi and Kasama District, Zambia. Unpublished PhD thesis: University of Cambridge.

Smith, B W 1997 Zambia's ancient rock art: the paintings of Kasama. Livingstone: National Heritage Conservation Commission of Zambia.

Thackeray, A I 1983 Dating the rock art of Southern Africa. South African Archaeological Society Goodwin Series 4: 21-26.

UNESCO 1999 Operational Guidelines for the Implementation of the World Heritage Convention.

Vinnicombe, P 1976 People of the Eland. Pietermaritzburg: Natal University Press.

Wendt, W E 1976. 'Art mobilier' from the Apollo 11 cave, South West Africa: Africa's oldest dated works of art. South African Archaeological Bulletin 31: 5-11.

Whitfield, J 2002 Engraved stones from South Africa could be the oldest works of art. Science Update, 11 January.

Whitley, D S, & Annegarn, H J, 1994 Cation-ratio dating of rock engravings from Klipfontein, Northern Cape. In: Dowson. & Lewis-Williams, 1994: 189-97.


Par Jean Clottes, Ancien Président du Comité International d'Art Rupestre (ICOMOS)

version PDF

 L'Art Rupestre : Une étude thématique et critères d'évaluation


LE TERME d'art rupestre qualifie les manifestations artistiques sur support rocheux. C'est la seule manifestation culturelle de l'humanité qui se soit poursuivie sans interruption pendant plus de trente millénaires pour parvenir jusqu'à nous sous ses formes multiples, inchangées depuis les origines. Les débuts de la création artistique ne sont pas le fruit d'une culture ou d'une ethnie particulière mais une composante essentielle de l'Homo sapiens sapiens : dès que celui-ci se répand dans le monde, les manifestations d'art rupestre apparaissent partout, de l'extrême nord de la Scandinavie à l'extrême sud de l'Afrique, de l'Espagne et du Portugal à la Sibérie, couvrant toute l'Asie et les Amériques, et répandu dans toute l'Océanie, particulièrement en Australie, mais jusqu'à l'Ile de Pâques.

Dans l'art rupestre, on constate une grande diversité et une extraordinaire complexité. Les études ethnologiques nous éclairent sur la signification des motifs et des symboles. Par exemple, dans le monde entier, cet art a été choisi comme vecteur des mythes sur la Création du Monde. L'art préhistorique et tribal, le plus souvent, est un art de pouvoir, dont les symboles sont conçus pour faciliter la vie, chercher de l'aide auprès de puissances extérieures, avoir une meilleure prise sur le réel comme sur le monde des esprits auquel il est inextricablement mêlé.

Il constitue donc un patrimoine unique et particulièrement précieux. D'abord, par son antiquité, puisqu'il représente une chaîne ininterrompue de plus de 35 000 ans. Par ses chef-d'oeuvres aussi, qu'il s'agisse des peintures des Magdaléniens dans les cavernes franco-cantabriques, des oeuvres des Aborigènes d'Australie ou de celles des Bushmen d'Afrique australe. Il témoigne de pratiques religieuses qui plongent dans la nuit des temps et en constitue la seule manifestation tangible. Il apporte des renseignements multiples sur les modes de vie, les vêtements, les armes et les outils, les cérémonies et sur tout ce qui constituait le monde matériel et spirituel de civilisations disparues.


SI NOMBREUX que soient les sites d'art rupestre, ils ne constituent qu'une infime partie de ce qui a existé. En effet, qu'il se trouve en grottes, sous abris ou sur des roches en plein air, l'art rupestre est soumis à des agressions de toutes sortes, de la part de la nature et des hommes. Chaque site est comparable à un musée qui se trouverait exposé aux éléments et aux actes de vandalisme. Du monde entier parviennent des échos alarmants sur sa dégradation, dont l'accélération est due à des causes multiples.

Jusqu'à la fin du XIXème siècle dans certaines régions de l'Afrique et des Amériques, ou même la première moitié de ce siècle en Australie, l'art rupestre se renouvelait, grâce à des traditions religieuses toujours vivantes pratiquées par des sociétés traditionnelles dont les modes de vie avaient peu varié. Cet état de choses n'existe plus, sauf à l'état de relique pour des groupes de plus en plus réduits. Dans son immense majorité, l'art rupestre est devenu un art fossile. Il faut donc protéger ce qui existe dans la forme où les oeuvres nous sont parvenues.

Cet art non renouvelable se dégrade vite et disparaît sous l'effet de phénomènes naturels. La nature de la roche joue un rôle crucial : certains grès, par exemple, se desquament régulièrement, de sorte que les peintures de la couche superficielle sont détruites. Les animaux concourent également aux détériorations : dans les abris, des nids d'oiseaux ou d'insectes recouvrent les peintures et les font disparaître peu à peu ; de grands mammifères se frottent aux parois ornées. Tous ces phénomènes, selon les lieux, provoquent une érosion progressive de l'art rupestre.

Les effets dévastateurs de l'action humaine sont plus nocifs encore. Ils prennent des formes diverses. Dans le nord de l'Europe, ce sont les pluies acides, dues aux pollutions industrielles de cette seconde moitié du XXème siècle, qui menacent d'une entière destruction une grande partie de l'art rupestre scandinave. Partout, dans le monde, le vandalisme sous ses multiples aspects intervient, dégrade et fait disparaître des sites entiers.

L'art rupestre, dans son contexte culturel d'origine, était protégé par le respect qui l'entourait, de même que l'environnement naturel où il s'inscrivait. Actuellement, des forces opposées le menacent. Dans la conscience du grand public, la vénération des anciens graphismes n'existe plus. De sorte que les travaux d'urbanisation, les carrières, l'ouverture de routes, la construction de barrages détruisent chaque année des milliers de représentations. Devant les nécessités du développement économique, les investissements et les emplois mis en jeu, la préservation de l'art rupestre passe trop souvent au second plan, qu'il s'agisse de pays industrialisés ou non.

Cet art ne présente pas le caractère monumental des constructions architecturales. Il est beaucoup plus vulnérable, car il peut disparaître avec la roche elle-même, dans le cas de grands travaux, mais également si la pellicule superficielle est attaquée d'une façon quelconque. À l'heure actuelle, sans noircir le tableau, il est certain qu'une grande partie de l'art rupestre mondial sera détruite au cours des prochaines décennies. Il s'agit donc d'un patrimoine en très grave péril.

La situation est très différente selon les pays, en fonction du nombre des sites et de leur isolement, du degré d'intérêt pour la culture, de la densité de la population et de bien d'autres paramètres. Parfois, des sites sont classés parmi les monuments d'intérêt historique et/ou culturel et bénéficient ainsi d'une protection légale. Dans leur très grande majorité, ils ne le sont pas. Rappelons que seuls treize ensembles de sites d'art rupestre figurent à ce titre sur la Liste du Patrimoine Mondial de l'Unesco : ils se trouvent en Algérie, en Australie, en Libye, au Brésil, au Mexique, en Argentine, en France, en Italie, au Portugal, en Espagne, en Suède et en Norvège. Aucun n'a été mis sur la Liste en Asie, en Afrique sud-saharienne, aux Etats-Unis ou au Canada.

La protection de l'art rupestre ne fera des progrès significatifs que dans la mesure où elle sera perçue comme une nécessité absolue, et non pas comme marginale et secondaire par rapport à des objectifs économiques ou touristiques. C'est un problème d'éducation et de promotion culturelle au sens le plus large.


LES SITES d'art rupestre se comptent dans le monde par dizaines, voire centaines, de milliers. Il faut donc faire des choix, bien que tous soient signifiants et précieux à tel ou tel titre. La liste provisoire et préliminaire ci-après ne peut qu'être indicative. L'intérêt relativement récent pour l'art rupestre dans certains pays et le manque de moyens fait que des ensembles ornés se découvrent tous les jours et certains sont de toute première importance. L'absence d'un site ou d'en ensemble de sites sur cette liste ne devrait donc pas empêcher, dans l'avenir, qu'une demande de mise sur la Liste du Patrimoine mondial soit examinée selon ses mérites propres. Tout ce que l'on peut dire à ce stade, c'est que tous les sites mentionnés ci-après méritent amplement de figurer sur la Liste, mais que leur inventaire reste ouvert.

Ils ont été pré- sélectionnés en fonction des critères suivants, après enquête auprès de plusieurs de nos représentants régionaux :

  • leur caractère exceptionnel sur le plan esthétique ;
  • sur le plan ethnologique ;
  • sur le plan archéologique et chronologique ;
  • sur le plan environnemental ;
  • le nombre des représentations sur une superficie bien délimitée (lieux sacrés) ;
  • la protection dont ils sont l'objet.

Les étoiles (*) marquent l'évaluation d'une importance qui ne peut être que subjective.

III.1 Sites rupestres du Tadrart Acacus (Jamahiriya Arabe Libyenne) © UNESCOuKhahlamba / Parc du Drakensberg (Afrique du Sud) © UNESCO - M.L. PearseAfrique
(actuellement 3 sites sur la Liste : Afrique du Sud (uKhahlamba/Drakensberg), Algérie (Tassili n'Ajjer) et Libye (Acacus).

En Afrique, le nombre des sites d'art rupestre dépasse les 150 000. C'est pourquoi, la plupart des sites retenus - à l'image de ceux qui figurent déjà sur la liste pour ce continent -, sont en fait desensembles de sites, couvrant de vastes régions. Des sites très importants existent aussi au Maroc, au Malawi, au Lesotho, au Botswana, en Zambie.

III.1.1         Afrique saharienne

 -  Libye
+*** Messak (milliers de sites avec des gravures splendides)
+** Mathendous (une dizaine d'oueds avec de nombreux sites prestigieux)
 -  Tchad
+** Ennedi (très nombreux sites gravés et peints)
+** Tibesti (très nombreux sites gravés et peints)

III.1.2         Afrique subsaharienne  *
 -  Afrique du Sud
+* Cedarberg (concentration d'abris peints)
 -  Namibie
+*** Brandberg (plus de mille sites peints dans un massif isolé)
+* Twyfelfontein (plusieurs milliers de gravures en un même lieu ; parc)
 -  Tanzanie
+** ensembles de Kondoa-Singida (près de 350 sites ornés de gravures et de peintures)
 -  Zimbabwe
+*** Matopos Hills (très important ensemble d'abris ornés)

* L'étude Southern African rock-art sites, rédigée par Janette Deacon en collaboration avec les membres du Southern African Rock Art Project (SARAP), a été éditée par l'ICOMOS en 2002.


III.2 Cueva de las Manos, Rio Pinturas (Argentine) © UNESCOAmériques

III.2.1         Amérique Centrale et Amérique du Sud
(actuellement 3 sites sur la Liste : Bolivie (Samaipata), Brésil (Serra da Capivara) et Argentine (Cueva de las Manos).

Les travaux - et les découvertes - se multiplient sur ce continent, dont l'art rupestre est encore mal connu.

 -  Argentine
+ Guachipas, Salta (une trentaine de grottes et abris ornés avec peintures et gravures)
 -  Brésil
+** Rio Peruaçu, Minas Gerais (plusieurs grands abris avec peintures dans un environnement exceptionnel)
 -  Chili
+ Tarapacà, Arica (géoglyphes)
+ Sierra de Arica (peintures)
+ Taira (peintures)
+ El Médano (peintures)
 -  Guatemala
+ Naj Tunich (grotte avec peintures et gravures Maya)
 -  Mexique
+ Chalcazingo, Morelos (gravures Olmèques)
+ Juxlahuaca, Oxtotitlan (grottes à peintures Olmèques polychromes)
+ Loltun et autres grottes de Oxkutzcab, Puuc, Yucatan
 -  Pérou
+** Toro Muerto, Arequipa (ensemble de plusieurs milliers de blocs gravés)
 -  Venezuela
+ Rio Orinoco (une cinquantaine d'abris et grottes ornés)

III.2.2         Amérique du Nord
(actuellement 1 seul site sur la Liste : Mexique (Baja California))

Les sites d'Amérique du Nord sont largement ignorés du public. On en compte cependant des milliers, qui sont encore parfois l'objet de pratiques cultuelles.

 -  Canada
+** Writing-on-Stone, Alberta (gravures)
+ Petersborough, Ontario gravures)
 -  Etats-Unis
+*** Great Gallery (plusieurs grands abris peints dans un environnement exceptionel)
+*** Coso Range, Californie (milliers de gravures rupestres sur roches à l'air libre, en plusieurs canyons)
+ Pleito Creek, Californie (plusieurs grottes avec peintures)
+** Sites Dinwoody/Bighorn Basin, Wyoming (gravures)
+** Sites de la Lower Pecos River, Texas (abris peints)

III.3 Asie
(actuellement aucun site sur la Liste)

Des dizaines de milliers de sites existent en Asie. Ils sont très peu connus.

 -  Chine
+** Helanshan, dans le nord (gravures)
+** Huashan, dans le sud (peintures)
 -  Inde
+*** nombreuses grottes peintes du Bhimbetka, Bhopal (Madhya Pradesh)
 -  Indonésie
+** Grottes ornées du Kalimantan, à Bornéo
 -  Russie
+* Tamgaly, Kazakhastan (gravures de plein air)
+** plusieurs ensembles de milliers de gravures dans les vallées de la Lena, du Ienissei, de l'Angara, de la Tomsk

III.4 Gravures rupestres de Tanum (Suède) © UNESCOSites d'art rupestre préhistorique de la vallée de Coa (Portugal) © UNESCOEurope
(actuellement 7 sites sur la Liste : Portugal (Foz Côa), France (sites de la Vallée de la Vézère), Espagne (Altamira et sites du Levant), Italie (Valcamonica), Suède (Tanum) et Norvège (Alta).

Les grottes et ensembles cités ci-après sont en majorité du Paléolithique et uniques au monde de ce point de vue.

 -  Espagne
+** Grottes ornées de la Corniche Cantabrique (Asturias, Cantabria, Pays-Basque) (plusieurs dizaines de grottes ornées paléolithiques)
 -  France
+*** Grotte Chauvet, Vallon-Pont-d'Arc (splendide caverne ornée, avec les dates les plus anciennes du monde)
+** Grottes paléolithiques pyrénéennes (Ariège, Haute-Garonne : 13 grottes ornées parmi les plus prestigieuses connues)
+*** Mont Bego, Alpes-Maritimes (environ 40 000 gravures rupestres chalcolithiques ; Parc du Mercantour)

III.5 Parc national de Kakadu (Australie) © UNESCO - I. HascovecOcéanie
(actuellement 1 site sur la Liste : le Parc de Kakadu en Australie).

L'Australie doit compter entre 80 000 et 100 000 sites. Les traditions sont souvent conservées. Bien d'autres ensembles seraient également dignes de figurer sur la Liste.

 -  Australie
+*** Région de Laura (Cape York)(plusieurs centaines d'abris peints ; gravures ; près de 30 000 ans de tradition d'art rupestre)
+*** Kimberley (Bradshaw sites ; des centaines d'abris peints avec une forme d'art très originale)
+** Pilbarra (gravures)

Additional information