Philip E. L. Smith
Problems and possibilities of the prehistoric Rock Art of Northern Africa
African Historical Studies, I, 1 (1968): 1-39
The Geographical and Chronological Frameworks
The first discoveries of rock art in North Africa were made by officers of the French army in the Southern Oran Mountains of Algeria in 1847, when such unsuspected animals as elephants, rhinoceros, and lions were found drawn on the rocks. Since 1850, when Heinrich Barth made the first discovery of Saharan rock drawings in the Libyan Fezzan, similar finds have come to light in most of the areas where massifs or rock outcrops offer the opportunity for painings or engravings. The names of Frobenius, Breuil, Monod, Huard, Reygasse, Dalloni, Capot-Rey, Graziosi, Winkler, Vaufrey, Lhote, Mori, Dunbar, and Mauny are only the best known of the many individuals linked with the discoveries of the past century.
The richest zones are probably the Southern Oran range in Algeria, the Tassili-n-Ajjer in southeastern Algeria, and the Fezzan massif in western Libya. But many paintings and engravings are also found in southern Morocco and Rio de Oro and in the massifs of the Hoggar, Adrar des Iforas, Air, Tibesti, and Ennedi (see map). East of Ennedi they are known in Gebel Uweinat on the Libyan-Egyptian frontier, in the Darfur region of the Sudan, and discontinuously in the outcrops and oases leading to the Nile Valley. The prehistoric art in the Nile Valley itself, like that between the Nile and the Red Sea, is sometimes set apart by those discussing the art of the Sahara proper, though it is doubtful whether this distinction can be justified on any grounds but geographical convenience. Within this immense distribution there are not only considerable stylistic variations based on spatial and temporal factors but also clear indications of certain emphases or specializations based, apparently, on local traditions and accentuated by isolation and environmental factors. For instance, paintings are extremely rare in the Southern Oran area and in the Nile Valley and Eastern Desert, where nearly all the art is represented by engravings of various sorts. On the other hand paintings are very abundant in the massifs of the central eastern Sahara, where they are usually accompanied by engravings as well. On what may be regarded as the same time horizon or “period” there can be considerable differences in expression in spite of very similar techniques: for example sexual themes are very common in the earliest pre-pastoral engravings of the Tassili but far less so in Southern Oran, whereas the reverse is true of scenes showing man-animal associations. The extent to which these and other variations are to be regarded as determined by different ethnic or cultural groupings, by differences in religious or magical practices, by environmental influences, and by diffusion or the lack of it has given rise to a great deal of debate in the past few decades.
It may be convenient here to sketch the broad outlines of the chronological framework in which North African rock art is placed today. It hardly needs to be pointed out that there is still a good deal of disagreement on details, on the position of individual figures or sites, and especially on the absolute (chronometric) dating of periods, styles, and figures. A full discussion of these fine points is beyond the scope of this paper. Nevertheless, it is worth emphasizing that, by and large, there is a fair amount of agreement on the general developmental scheme for North African rock art. The basis for this concensus will be discussed later along with some of the assumptions involved. At this point a generalized description may be more helpful in providing a framework which the reader can use in evaluating the later discussion.
Leaving aside for the moment the question of the origins and absolute age of the earliest art in northern Africa, it is generally agreed today that the earliest art figured on large rock surfaces (art rupestre) seems to be represented by engravings only. This is the so-called Bubalus period or style or phase 5, which is found especially in Southern Oran and the Tassili-Fezzan area and is characterized by large naturalistic engravings of such animals as rams, cattle, rhinoceros, hippopotamus, equids, and, particularly, by the large extinct buffalo with wide sweeping horns, Bubalus antiquus 6. Humans are also shown, often in hunting or coitus scenes or with zoomorphic heads. The animals, especially rams, are sometimes drawn with “discs” between the horns, with collars, pendants, and festooned lines representing perhaps lassos leading to the heads. The deep grooves are sometimes polished, and the interiors of some figures may be polished or pecked as well. These engravings are today considered to be, in large part if not totally, the work of hunting groups. In the central Sahara there may also have been a “Hunter Period” style of engravings in which the Bubalus never appeared.
These types are followed, at least in Tassili, Fezzan, and Ennedi, by a peculiar style of paintings to which Breuil 7 gave the name Round-Heads. A number of techniques and sub-styles is known, but the characteristic form shows the famous “white Martians” — humans, often gigantic, with round and usually featureless heads, and sometimes horned and masked. Animals such as elephants, rhinoceros, giraffes, ostriches, and Bubalus represent a largely Ethiopian-type fauna. I shall come back later to the alleged racial and religious characteristics of the humans and their activities; for the moment we can say that some evidence suggests the beginning of the Round-Head style may be older than 6000 B.C.
Immediately after the Round-Head period, or perhaps overlapping it somewhat, is the principal and best known grouping of art in the Sahara, the Bovidian Pastoral style. The dominant theme here concerns cattle, by now unquestionably domisticated, and a great many polychrome scenes showing herding, milking, etc. have been found, as well as scenes of social life. Engravings are also present but are less well developed than the paintings and do not represent identical subjects or themes; there may be some continuity in style with the Bubalus engravings at the beginning. Cattle are the animals most frequently shown, often in huge herds, but apart from the absence of Bubalus there are few important changes in the wild fauna from the earlier period. An Ethiopiantype fauna and, for the most part, a very favorable environment, continue to be reflected; this grouping can very probably be related to a “Neolithic Wet Period” or to a climatic optimum documented in other parts of the Old World about this time. Nevertheless, hunting scenes are rare. Extremely important for our purposes is the fact that the human figure including the face is frequently shown with absolute realism. Pottery was used in the dwelling sites.
The Bovidian Pastoral phase lasted from at least 5500 B.C., judging from Mori's recent research in the Acacus 8, until at least the third millennium B.C. By the terminal phases the style had become simplified (“decadent” to some writers), and human figures are rather schematic. There is reason to believe that between 2500 and 1200 B.C. there were marked climatic and faunal changes in the Sahara leading to an impoverishment of the area and centrifugal movements to the peripheries. The C-Group movement into Nubia may reflect one such migration from the desert to the more hospitable regions about 2300 B.C. By about 1500 B.C. the art can be classed as belonging to the Horse Period, with the appearance of paintings and engravings showing at first horse-drawn carts or chariots and later cavalry. Some of the Ethiopian fauna (elephant, giraffe, rhinoceros) are still shown in hunting scenes, but mouflon and ostrich have apparently become the principal game. The final phase of the Horse Period, a few centuries B.C., sees the appearance of the Libyco-Berber script in some drawings and of warriors shown in a bitriangular style. Probably about 300 A.D. the Camel Period begins and continues into historic times as environmental conditions deteriorated even further, aided in part, at least according to some writers, by man and his domesticated animals.
This very simplified outline cannot pretend to do justice to the intricacies of the actual developments and sequence. It glides over some very basic problems and no doubt incorporates certain assumptions and errors which will themselves be the targets for criticism later on in this paper. Nevertheless, all our information at the present time indicates that it is, broadly speaking, representative of the actual sequences in the central parts of the Sahara, if somewhat less so in the peripheral zones. With this scheme in mind we can proceed to consider some of the basic problems and contributions of the data so far collected.
5. Also known as the Large Wild Fauna period (Mori). But Huard rejects both these expressions and prefers simply Hunter Period. P. Huard and J. M. Massip, "Gravures rupestres de Ye Lulu Loga (confins nigéro-tchadiens)," Bulletin de la Société de Préhistoire Française, c.r., 8 (1964), 192-197.
6. The exact zoological terminology seems to be in dispute and I have kept the traditional expression.
7. Henri Breuil, "Les roches peintes du Tassili-n-Ajjer," Actes du Congrès Panafricain de Préhistoire, II session, Alger, 1952 (Paris, 1955), 65-219.
8. Fabrizio Mori, "Contributions to the Study of the Prehistoric Pastoral Peoples of the Sahara. Chronological Data from the Excavations in the Acacus, " Miscelanea en Homenaje al abate Henry Breuil, 1877-1961, E. Ripoll Perelló, ed., 2 (Barcelona, 1965), 172-179.