The difficulty in defining past events based on role differentiation in society is part of the more general difficulty, for the archaeologist, of characterizing human agency. Because of the influence of natural sciences, typology and stratigraphy were for a long period the main focus of archaeological studies. Only later did the discipline approach lifestyle reconstruction. Furthermore, such a reconstruction, put forward by the program of the new archaeology, was meant as the reconstruction of processes in which the actors remained in the background. Indeed, the adaptive and systemic explanations that dominated the scene from the 1960s to the 1980s neglected putting emphasis on the individuals.
Nevertheless, in North African archaeology the processual study of food production and of the state also involved the study of concrete, social organizational phenomena. Let us consider how human groups planned to use the environment, as an example. Moreover, the broad-spectrum exploitation of resources, the evidence of sedentism, wealth accumulation, and emergence of ranking linked with resource control and redistribution, represented more mature specifications about the social context. Although such reconstructions remained in the realm of adaptive interpretations of observed phenomena, they featured in detail the type of relationship that was supposedly instituted between the social group and territorial space. Further, they implicitly introduced elements of choice and decision by the actors in the reconstruction drawn from cultural material data. The problem of developing methods to properly recognize elements connected with human actions from data often very scant and fragmentary was overcome, at that time, by making reference to the sphere of living societies. Ethnoarchaeology showed how to derive materials from ethnographic societies that could serve as reliable models against which archaeological data could be measured.
Recently, having shifted the latest archaeological orientations from the reconstruction of processes to the inquiry of contexts, a renewed interest in social structure asserted itself. For this effort, more attention is also given to agents that are not directly observable but that work within the documentation, such as gendered identities. Being part of the social trend of postprocessual archaeology, gender studies delve into differentiations in the social participation of sexes. At the same time, as noted by Kent (Chapter 3) and Gifford-Gonzalez (Chapter 7), such studies reveal how a certain dominating mentality led to obscuring and mystifying certain problems and roles. If it is true that postprocessual literature, either contextual or structural, has placed emphasis on meanings as opposed to functions, the recognition of roles, in the sense of “meaning” attributed to the sexes, is certainly one of the most promising fields of study. Nevertheless, there has been a unanimous call for a more effective methodology to record the engendered relations (Dobres 1995b; see also Parkington, Chapter 2). Actually, until now, gender archaeology has been manifested primarily at the theoretical level. Only rarely has it faced the problem of active research (Claassen 1992a). The question of how to develop a theory capable of understanding and recognizing human agency in past events has been neglected. Consequently, also overlooked has been the problem of how to characterize the role taken by the various social components.
We think that in these conditions the traditional relationships among paleoethnology, ethnology, and anthropology must be further reinforced. However, using the ethnographic analogy in a simple form does not help much. We agree with Gifford-Gonzalez (Chapter 7) that there is a need for reinterpreting analogy as a way of systematically transferring knowledge from a familiar to an unfamiliar context warranted by the existence of causal relations (Barich 1988). Contextual studies of contemporary forager societies have allowed for new interpretations (e.g., Testart 1982), in some way contrasting with the traditional views linked to hunter-gatherers. Nevertheless, it is necessary to take it one step further: such interpretations are an example of the reference basis on which a more profound analysis, directed toward attributing meaning to human activity, can be developed.
The situation we present here has been taken from an archaeological context—the Saharan Holocene—whose abundant and varied documentation could easily be reread in light of previous considerations. As a matter of fact, it seems to be a promising sample to investigate from a gender perspective. We refer to a rather wide context, in time and space: approximately from 10,000 to 7000 BP in the central Saharan belt from Mauritania to Sudan, and from lower Tunisia to Niger (Figure 6.1). The continuous and uniform element is given by the form of adaptation. It is a complex of similar situations, all strongly based on exploiting aquatic resources, that were defined as “aqualithic” (Sutton 1977). In Mali, the lacustrine phase between 8500 and 6900 BP was recognized through stratigraphic sections in the ancient lakes bordering the Erg (Petit-Maire and Riser 1983). In Algeria, research in the Hoggar and in the Tassili regions has shown a long cultural sequence documented in the ancient sites of Amekni (Camps 1969) and Ti-n-Hanakaten (Aumassip 1984). Other human occupation centers, pertaining to the early and mid-Holocene, were recognized in Chad and Tibesti by French and German archaeologists (Courtin 1969; Gabriel 1977). Lately, the most recent research in Niger, Libya, and the Egyptian Sahara has placed the first human presence at the very beginning of the Holocene, after the resuming of the monsoon season.
Figure 6.1. Map of North Africa and the Sahara.
In literature these similar adaptation phenomena are often referred to as the “Saharo-Sudanese Neolithic,” according to the definition coined by G. Camps (1974, 1982), which set off a debate that is still going on today. Actually, these situations can be better defined as transitional contexts, in which prototypes appeared for the first time, that were later destined for domestication, through repetitive gathering and selection processes.1 Until now these societies have been studied in a processual perspective, featuring the elements of change. Lately, attention has moved from consideration of cultural change to context—that is, to the consideration of horizontal, synchronic, interrelations in the cultural system. These reconstructions have been useful for specifying social functioning.
The Saharan region appears as a generalized ecosystem of fishing and hunting-gathering activities aimed at a broad-spectrum utilization of resources. Subsistence relies primarily on shellfish, fish, and small game, plus, overall, plant resources. Large game is a minor dietary component. For this reason, women could have played an important role in food procurement. Efforts to expand the exploitable base, using resources at a short distance from the home bases, brought about a change in settlement behavior. Ultimately, the limited nomadism was followed by semiresidential arrangements. We can list Ti-n-Torha East in Libya (around 8500 BP: Barich 1974) and site E-75-6 at Nabta Playa in Egypt (around 8200 BP: Wendorf and Schild 1980) as examples of a real village-type organization showing stone-lined huts for family units (Figure 6.2). Similarly, the family household is the basic economic unit related to such structures.
Figure 6.2. Plan of the Ti-n-Torha East settlement site in the Tadrart Acacus, Libya (modified after Barich 1974).
The year was divided into two parts, the dry and rainy seasons. In summer, at the return of the rains, the community temporarily left the main settlement and dispersed into the hinterland. The Ti-n-Torha group’s seasonal scheduling was primarily tied to crop maturation rhythms—those belonging to the millet group. During winter, the group migrated toward the bottom of the valley to collect and process plant food and to fish in the small permanent pool. They also hunted migratory birds. During summer, mostly ungulate hunting (barbary sheep, gazelle, hartebeest) was practiced on the plateau by task groups—probably adult males. Perhaps, in summer, tending the earliest proto-domestic items was also performed. The simultaneous appearance of large ceramic vessels and permanent dwellings allows us to affirm that in this way preliminary food storing was established, together with major sedentism, demographic growth, and role division between the sexes. Ethnographic observations indicate that women tended to be responsible for procuring wild and domestic plants (Ehrenberg 1989:84–85). The amount of wild gramineae species (Pennisetum, Brachiaria, Echinochloa, Urochloa, Setaria, Cenchrus, Panicum) found at Ti-n-Torha Two Caves (around 8500 BP) reveals the importance of wild plants in the diet (Barich 1992; Wasylikowa 1992). We can suppose, therefore, that in this context women, who probably also practiced both fishing and small land-mammal hunting, played a primary role in food procurement. The participation of women in acquiring food probably resulted in a gender equality that was not present in later societies. It should be remembered, in fact, that in environments subject to recurrent shortages, such as those under study, gathering, more than hunting, must have represented the principal form of sustenance. Furthermore—as suggested above—female involvement also in small mammal hunting has been hypothesized by various authors (Estioko-Griffin and Griffin 1981; see also Wadley, Chapter 4). It is interesting to note that this type of condition must have had a long tradition, since it had already existed in the Pleistocene.
The examples under study, which come from the contextual reading of data from a multidisciplinary range, find strong relational analogies with the pattern suggested by Testart (1982) in his study of complex gatherers. With this definition, the author indicated the more sedentary foragers of the Northwest Coast and California Indians who, following Binford’s typology, could also be defined as collectors (Binford 1980). The most distinctive element in respect to other hunter-gatherers (e.g., Bushmen/Basarwa) is the increased sedentism and seasonal food storage:
The usual residence of hunter-gatherers practicing storage is a village or a permanent camp built around food reserves from which seasonal expeditions requiring a certain mobility, such as hunting, are launched. (Testart 1982:524)
The increased sedentism in relation to food storage had a notable effect on the birth spacing in the women of the Northwest Coast. Births, in fact, do not need to be spaced at intervals as among the mobile hunter-gatherers. Testart observed (1982:525) that, while the higher population density among hunter-gatherers had been explained, up to now, with an increased environmental productivity, this, in reality, must be put in relation to food storing and consequent semisedentary customs. Food storage could also be the basis for the rise of social inequality (Testart 1982:525). Conversely Kent (Chapter 3) very relevantly emphasizes the frequent “invisibility” of foragers and their being undifferentiated at the gender level, since their group organization is without stratification. It is necessary to note that among the Saharan hunter-gatherers, broad-spectrum exploitation of a limited environment was not able to create real surplus products nor, therefore, any form of ranking.
Increased sedentism also produced an increased availability of goods. If we are looking at highly mobile groups, or groups living in transient camps, we find only very light equipment; while in base camps ornaments (necklace beads and pendants), amulets (animal bones, both pierced and decorated), and less portable implements are found. Among these we can also include grinding equipment (lower and upper grinding stones) and hoes, associated with the harvesting activities of the women who, most likely, were also their producers (compare Goodale 1971, for Tiwi women of Australia). Finally, with increased sedentism and availability of more goods, the production of a larger number of containers, either basketry or pottery, was encouraged.
As is well known, the Saharan epipaleolithic societies are among the most ancient pottery makers. The earliest examples were found in the central-western regions of Libya and Niger (Tagalagal and Temet in Niger: 9330 ± 130 and 9550 ± 100 BP; Ti-n-Torha in Libya: 9080 ± 70), while toward the east, on the Nile, pottery appears later, generally in the 9th millennium BP contexts. A question still unanswered today is why the early groups were inclined to produce ceramics. More than a necessity for holding liquids, for which groups could also have used ostrich egg containers, it was perhaps the need to store solid goods, made available by the intensification of harvesting activities, that called for ceramic production. As Ehrenberg (1989:87) notes, it was probably the women, the most directly involved in harvesting activities, who felt the need to innovate. Even now, data known from modern societies attribute an important role to women in the development of ceramic technology. In Mali, pottery is produced primarily by women for each of the numerous groups (Bozo, Somono, Peul, Bambara, Songhai, Dogon) that today populate the inner Delta of Niger (Gallay et al. 1996:29–33). The scarcity of fragments in the earliest settlements of the 10th millennium BP tells us that up to that moment vessels had not been destined to a domestic use, but rather had been reserved for particular purposes. At Nabta, compared with the large amount of small pits for cooking in settlement E-75-6, which contained charcoals and plant remains, the amount of ceramic sherds is very low. Close has suggested that this production could imply some social and symbolic function (Close 1995:28–29). It can be added that in this and other analogous sites particular spatial concentrations that could be studied in the way indicated by Kent (Chapter 3) and Casey (Chapter 5) were not recorded.
In these ancient Saharan sites there is proof that, from the beginning of the occupation, the economic model also included cattle breeding. These sites allow us, therefore, to investigate the transition from a gathering economy, integrated with herd tending, to a pastoral-based economy, whether exclusive or predominant. In this way we might be able to clarify the fundamental transformation undergone by the female role during the passage from one type of organization to the other and, eventually, her probable loss of meaning. Understanding this transformation represents a rather important goal, whose success is tied to the possibility of adequately evaluating the established relationships within the group. In fact, while during the initial phase of pastoral organization women could have added herd tending to their other activities (Ehrenberg 1989), later on, with the increase of the herds, pastoralism became a male prerogative, reducing women to a marginal role.
The early cattle evidence is known in the early Holocene Libyan Sahara and, above all, in the Egyptian Western Desert (Kiseiba, Dachla, Wadi Bakt). Wendorf and colleagues (Wendorf et al. 1990) built a mobility model for the Nabta and Kiseiba groups that foresees cyclic movements between the basin’s bottom in dry months and the plateaus in the rainy season. Such a model also features the role of task groups (males and/or females) that, periodically, on the return of the rains, left the base group to pasture the first herds (Wendorf et al. 1990). By the mid-Holocene the pastoral economy prevailed, and even increased in importance, becoming a dominant economic strategy in a phase of climatic deterioration. Many small hearth places started appearing on all the Saharan plains about that time, with a particular intensity around the oases. They are clear evidence of shepherds’ rest-stops (Gabriel 1984, 1987). The social meaning achieved by pastoral organization during the 7th millennium BP is also documented by funerary tumuli discovered in the Sahara. The structures uncovered right in the Nabta Playa region (megalithic alignments, calendar circles, and so on) have a great importance because they are a sign of pastoral-based ranking (Wendorf and Schild 1995). Some ritual cattle burial findings show the ideological value attached to this same resource.
Up to this point we have considered the material remains of the culture, to which data from tombs can be added. These are very rare throughout the Saharan area and consist of isolated tombs, or groups of two or three, within the sites themselves. A notable exception is represented by the necropolises of Mali (Taoudenni basin), which, instead, group a noteworthy number of individuals (Dutour 1989). Rituals are elementary and grave goods are exceptional. Nevertheless, various tombs of children and women (such as the Uan Muhuggiag and Amekni examples: Mori and Ascenzi 1959; Camps 1969) highlighted a social consideration of both classes.
These reconstructions, which widely use ethnographic comparison, were carried out according to the ecological-functional paradigm. But Saharan archaeology can rely on another formidable source of information, especially significant for our research since it is an expression of the symbolic world of society. As a matter of fact, an important contribution can come from the rock art field, provided that the study is focused on the artwork meanings through an analytical reading of them. This position can be traced back to Leroi-Gourhan (1968) and attributes meanings with reference to contexts and actions (Conkey 1989:152). By analytical procedure, nuclei of concepts specific to single prehistorical groups, as an expression of specific relationships established within the social group and with the outside world, could be identified. The degree to which our interpretations are valid could be measured against similar nuclei of thought—analogs—established through the analysis of ethnographic and oral sources. We can accept the association (Barich 1990; Sansoni 1994) of the multiresource environments that we have described up till now, with a rather significant corpus of paintings: the “round heads” art, so called for the way in which the human face is represented. Known in Tassili and Libya (with a few documented also in Tibesti), this is an extremely diversified art, which therefore corresponds well to a society that is just as diversified and changing. Obviously, direct reading can offer only a preliminary approach to these paintings, whose symbolic and ideological content could also have altered the relationships in reference to reality (Bednarick 1990/91; Hassan 1993; see also Parkington, Chapter 2). For example, the scarce characterization of the human figure can mean that there was a disinterest in the individuality of the portrayed person, who supposedly played a ceremonial and symbolic role in the scene (Sansoni 1994).
Figure 6.3. A family life scene from site Sefar XXI (Tassili n’ Ajjer, Algeria). A man and a woman together with two children, surrounded by containers probably holding vegetables (round-head style; modified after Sansoni 1994, Fig. 82, p. 138).
However, the copiousness of depicted scenes, on the whole, allows for reconstructing the surrounding environment: the rather luxuriant ambiance of the humid tropical savanna belt, which we learned about through excavations. Mufflons, gazelles, antelopes, and buffaloes were the most typical animal species. There are fish at Wadi Djerat, Tassili (Sansoni 1994: Fig. 25), and in the Acacus Mountains at Imha and In Taharin (Mori 1971:45–51) and also plants, even if they are few. As an example, one can quote a scene in the Tassili that represents a family ritual (Sefar XXI, Sansoni 1994: Fig. 82) (see Figure 6.3). From our point of view, it is interesting to note that in the artworks males and females performed distinct roles, and that the female figure (in contrast to the bubaline Maghrebian rock art, in which women are scarcely represented) occurs much more frequently. This fact could mean that the woman had gained higher value in the social organization (Sansoni 1994:208). This significance attributed to the female gender agrees with the organization that has been reconstructed and that emphasized the role carried out by women in gathering plants and manufacturing pottery.
The mastery of ceramic technology—transformation of material from an inanimate to an animate condition, from crude to cooked—eventually attributed a great status to women. The fact that women were so closely linked to ceramics, to pots, so much as to be identified with them, goes along with what appeared in the “round-head” art. It is not by chance that the female figures are sometimes richly decorated with motives typical of the ceramic repertoire (Figure 6.4). This testifies to the metaphorical meaning given to the woman, the main creator of the new vessels (see Aschwanden, quoted in Collett 1993:506–507).
Figure 6.4. Worship scene from the Sefar LXI site. The female figures all in a row bear body paint designs inspired by the ceramic repertoire (round-head style; modified after Sansoni 1994, Fig. 82, p. 138).
The large diffusion of the Bovidian art (starting from about 6000 BP) corresponds to the emergence of pastoral ideology. The copiousness of cattle scenes indicates that herds were the most highly valued social good, perhaps right at the time when the effective contribution of cattle to the diet and economy could have been rather limited (Holl 1989:348).
As noted by Gifford-Gonzalez (Chapter 7), an “androcentric” mentality reinforced the thesis of the pastoral world’s attributing a greater meaning to the male figure. Even the art would document this reversal of roles. Recently C. Dupuy has highlighted the underrepresentation of the female figure in the pastoral engravings (Dupuy 1995:205). Actually, as Ehrenberg showed (1989:102), when animal husbandry enters a mixed economy, the women’s role depends on the scale of herds. When herding is the prevalent activity, men tend to be more involved in herding, leaving women a marginal role such as processing milk. Based on the documentation known in the Sahara, throughout the 8th millennium BP there was a mixed economy, in which the gathering tradition was still strong; therefore we can suppose that a division of gender roles also continued. Maybe women primarily carried out gathering, together with tending the first herds and milking, while men were more involved in hunting. However, Lyn Wadley (Chapter 4) warns that we must define the women’s involvement in hunting better, and, as an example, she mentions the cooperative (man/woman) hunting among Mbuti. The Saharan model seems to have been progressively pushed to the east. Today it is reproduced by some Nilotic populations (see the Nuer, whose pastoral component, however, is much more relevant than in the Saharan example).
We are dealing with a crucial transformation, rather delicate for the balance of gender relations, which requires attentive study and making use of all possible information available. Even the study of the pottery inventory can indicate structural change. The decorative motifs of the early Holocene offer a wealth of themes that allow for individuating site-specific characteristics, both in syntax and techniques. We suppose that this phenomenon pertains to societies that are matrilocal and matrilinear, where the art of pottery is handed down directly from mother to daughter. On the other hand, the greater uniformity of mid-Holocene ceramic motifs could also be attributable both to women’s circulating within wider circuits following the patrilocal custom, prevalent among pastoral societies, and to the consequent style’s intermingling.
In the study of Saharan societies, ethnographic observation has encouraged contextual analysis, which, in turn, contributed to the specification of the social pattern. The early Holocene societies (from 10,000 to 6000 BP), in which gathering activities prevailed, gave an important significance to women for their contribution in food procurement. This position the women held probably was changed in the mid-Holocene as the organization also changed and the pastoral economy predominated. Essentially, these contexts can be a useful domain for investigating the progressive structuring of roles. It is necessary, however, to develop methods to reach a more sophisticated level of meaning, avoiding dangerous simplifications. Ethnology shows how to derive information about spatial arrangements, which can be useful for recognizing a differential access and utilization of resources, from the contextual study of settlements. Definitely, this is a new perspective even for whoever applies very systematic recording methods. Finally, rock art interpretation, if directed toward understanding the profound message expressed by the scenes the art bears, could also contribute to our better recognizing the true division of roles based on gender.
1. According to the orientation of European studies, such communities could be defined as “Mesolithic.” Nevertheless, for Africa it seems preferable to avoid definitions based on the dynamics of temperate regions.
