Some remarks on Saharan terminology.

Pre-pastoral archaeology from the Libyan Sahara and Middle Nile Valley

Savino di Lernia and Elena A.A. Garcea*

 

 

ABSTRACT

1. TERMINOLOGY: A REAPPRAISAL?

2. HISTORICAL BACKGROUND

3. PRE-PASTORAL ARCHAEOLOGY: A BRIEF REVIEW

              3.1. The Tadrart Acacus

              3.2. The Nile Valley

4. FINAL REMARKS

REFERENCES

 

 

 

 

 

 

 

ABSTRACT

 

Terminology is a controversial topic for African prehistory, which has become increasingly problematic with the pro­gress of research. The importance of a solution to this matter has been already felt and discussed. The need to build a conventional terminology in archaeology is an epistemological duty and a substantial tool for the communication among scholars. Most of the terms used in African prehistory derived from European and Near Eastern contexts, as originated from those areas. In this paper we shall discuss this specific areas, namely the Tadrart Acacus region (South-western Libya) from the Middle Nile Valley. Particular emphasis will be drawn Holocene human groups characterised by an extractive economy.

 

1. TERMINOLOGY: A REAPPRAISAL?

 

African scholars (Mudimbe 1988; Irele 1991; Andah 1995) fought a particular struggle against the use of European-derived archaeological terms, in order to stress the autonomous cultural paths which characterise African prehistory: Early Stone Age, Middle Stone Age and Late Stone Age were used to define techno­cultural processes of Pleistocene archaeology, although several scholars continued to use terms like Acheulean or Mousterian. We could recall the repeated efforts made by the Pan-African Congresses, which even created the so-called Pan-African Congress of Prehistoric Commission on Nomenclature and Terminology Bulletin, in order to satisfy the need to create African sequences, instead of European ones.

 

Paths toward food-production and their terminology probably are the most debated issues of African prehistory, as the quantity of different terms in use seems to demonstrate. Epipalaeo­alaeolithic, Mesolithic, Keramikum or Ceramic Age, Saharo-Sudanese Neolithic and others, have all been used to define human groups which often shared the same cultural and economic context. In 1967, J. Sutton suggested to reject the term Neolithic with regards to Eastern Africa (Quoted in Sinclaire-Shaw-Andah 1993), and still today A.B. Smith speaks of Neolithic as a ("barbaric concept" for the Sahara (Smith 1996). The term Mesolithic was first proposed by A.J. Arkell (Arkell 1949), and it was accepted and rejected several times'.

 

In our opinion, communication among scholars should be filtered by linguistic and epis­temological analyses. As usual, controversies are mainly based on the diagnostic and substan­tial features to distinguish the main features of past communities. A basic difference can be traced between English-speaking and French-speaking scholars, the former stressing economic features and the latter emphasising technological elements, such as pottery.

 

The importance of finding a solution to this matter has been already felt and discussed (Safa Conference 1992). Also the recent 10^th Pan-African Congress for Prehistory and Related Studies, as well as the 13th International Congress of Prehistoric and Protohistoric Sciences, revealed the urgency to come to an agreement among different scholars. Such a terminological confusion probably not only hides different ethical standpoints (i.e. African terms versus European terms), but also a theo­retical reluctance to identify cultural phenomena and to define anthropological complexities.

 

2. HISTORICAL BACKGROUND

 

The paradox of archaeological excavations is that we know from where and when we start, but we never know where and when we are going to finish. While on the spot, we are not always aware of all the information we are collecting. Only after laboratory analyses and data processing can we come up with chronological sequences, vertical and horizontal distributions, and variations in the archaeological record. These data constitute the solid evidence on which we base our interpretations.

 

Changes result from data comparisons aiming at separating diagnostic features. V. Gordon Childe (Childe 1956: 123) defined 'culture' as a plurality of types that are repeatedly and exclusively associated with one another and correspond to a recognisable distribution pattern. However, the question of defining cultures through archaeological records originates from the need to distin­guish discontinuities within apparently homogeneous behaviours over an uninterrupted period of time (Otte 1985: 420).

 

K.M. Trinkaus considers that macro-differences can be more easily appreciated in the avail­able record, as archaeology can basically reveal long-term changes (Trinkaus 1987). An unmensurable amount of data gets lost first during their accumulation, then during post-depositional processes, and finally during and after excavations.

 

Recognised prehistoric 'cultures' in particular do not necessarily represent real different cultural identities. James R. Sackett considers that archaeology can only provide material out­puts and, therefore, archaeologists can only determine isochrestic variations, or stylistic vari­ations in artefacts (Sackett 1977; 1990). According to this perspective, formal variation is the "definer of style". Margaret W. Conkey extended the materialistic approach to a wider perspective where style does not tell us about cultures or groups, but rather "the contexts in which groups or other social/cultural phenomena are mobilised as processes"  (Conkey 1990: 15).

 

As a matter of fact, artefacts and their stylistic variations may be better understood if they were related to their social and cultural context. Furthermore, as Stephen J. Shennan puts it, «cultures» are also "a way of classifying spatial variations in the archaeological record" (Shennan 1989: 6). In our opinion, material products resulting from archaeological research should be referred to a wider approach relating stylistic variations to cultural, economic, chronological, and spatial variations. This brings us back to Marce Mauss, who considers that techniques make indus­tries and crafts, and techniques, industries, and crafts together make the technical systems of existence (Mauss 1947). That is to say that techniques should not be separated from systems or contexts that produced them.

 

All variations, whether stylistic, economic, cultural, or technical, occur in a continuum with no beginning and no end, which is impossible for archaeologists to trace. This is the real limit for the archaeological research. The only way to overcome it is to place conventional yard­sticks for recognising and measuring the differences that are actually perceived from the archaeological record. According to Trinkaus, our goal is to study changes and to set bounda­ries along the process of growth of such changes (Trinkaus 1987).

 

The archaeological record is but a small part of the evidence of a continuous process dur­ing which one context develops and merges into another. We are never dealing with a succes­sion of single, separate units, but rather with ever-changing forms of behaviour and adapta­tion of human cultures in the natural environment.

 

After all, the tendency to historicize and divide our past into periods is felt as a necessary, although difficult challenge. According to, Clifford Geertz, it is particularly troublesome as there are a lot of ways of doing it, and each of them seems arbitrary, almost completely related to questions of descriptive convenience (Geertz 1995).

 

At a more practical level, when stylistic differences are perceived and when they match with variations in the archaeozoological, palaeobotanical, geological, and/or radiometric evi­dence, it seems legitimate to identify and define different cultural and socio-economic hori­zons. Daniel Cahen noted that the first effort of prehistorians is to describe, classify, interpret, and arrange archaeological materials according to coherent and recurring entities, provided within a certain geographic and chronological stability (Cahen 1985: 39). Different clusters may be separated on the basis of comparisons of each element in the archaeological record.

 

Relevant progress in the theoretical archaeological framework was provided by New Archae­ology scholars, who emphasised processes more than events (e.g. Binford 1965; Flannery 1967). Such an approach was mostly based on the analysis of material cultures and their relationships with settlement patterns in a functional perspective. An «unformalized» post-processual approach to prehistory aims at overlooking the «people-based» archaeological reconstruction, emphasising the «individual-­based» analysis (e.g. Hodder 1985). However, it seems to link dangerously prehistory to idealistic thematics.

 

A basic contribution to archaeological modelling was certainly the ethnoarchaeological approach (e.g. Binford 1968;Lee-DeVore 1968; Gould 1980): even if it should be considered with many theoretical nuances, ethnoarchaeology probably provides the best tools for testing our methodological «hammers» (Moore-Keene 1983) and for checking our ability to decode past archaeological records by means of actual/living analysis. As for African ethnoarchaeology, there are some significant and authoritative studies (For an African review, (see Atherton 1983; Agorsah 1990), although it does not seem to be thoroughly exploited, since there are more -buckshot- contributions than theoretical «general models» for the anlysis of past societies (Sensu Yellen 1977: 2). According to Atherton, also the ethnohistorical approach has not been exhaustively implemented, in such a highly poten­tial area like Africa (Atherton 1983: 79).

 

Archaeology deals with artefacts and processes, and our purpose is to link the former to the latter in a convincing, and our purpose is to link the former to the latter in a convincing and previously developed anthropological framework: if such an  attempt succeeds, it should be possible to translate artefacts into 'cultures' and to give them names. This implies the attribution of conventional names and the development of a termi­nology, in order to simplify the identification of the different horizons.

 

What we propose here is the characterisation of different pre-pastoral horizons that may be identified in the Central Sahara and the Middle and Upper Nile Valley during the Early Holocene, based on the analysis of both field and literature data.

 

3. PRE-PASTORAL ARCHAEOLOGY: A BRIEF REVIEW

 

'Pre-pastoral' archaeology is the term generally used for the human groups of the Cen­tral Saharan Early Holocene: the absence of food-production is the lowest common denomi­nator, assuming that their cultural path goes towards pastoral economy. In the archaeologi­cal debate, the prefix 'pre' is often considered as something amorphous in relation to well-defined cultural traits, but we should not forget that such a feeling probably co-exists with the term 'pre-history itself...

 

In our view, 'pre-pastoral' archaeology, which has been largely used as a conventional term to overcome the impossibility to find a more specific common terminology (e.g. Garcea 1993), includes several and different features, which can define diverse anthropological contexts of Early Holocene groups (di Lernia 1996).

 

3.1. The Tadrart Acacus

 

A review of the pre-pastoral sites of the Acacus in Libyan Sahara (Mori 1965; Barich 1987c; Lupacciolu 1992; Cremaschi-di Lernia 1998) led to the recognition of a significant differentiation in the previously undifferentiated 'pre-pastoral' archaeological sequence of several sites  (di Lernia 1996), like Ti-n-Thora East and Ti-n-Thora Two Caves in the northern Acacus (Barich 1987a; Barich et al. 1984), and Uan Afuda (Cremaschi-di Lernia 1995; di Lernia 1996) and Uan Tabu (Garcea 1992; 1996b) in the central range (Fig. 1).

 

Starting from the recent research by the Italo-Libyan Joint Mission of the University of Rome 'La Sapienza', the Uan Afuda cave is the first site where such distinction was recog­nised (Cremaschi-di Lernia 1995), thanks to a well preserved 4 m-thick stratigraphic sequence: a Late Pleistocene human occupation is attested  by a few unweathered Middle Palaeolithic (or MSA) Artefacts. Groups of 'Epipalaeolithic' hunter-gatherers re-occupied the cave at the very beginning of the Holo­cene (C¹⁴ datings span from 9,765 to 9,260 years B.P ). They feature a multi-activity site organi­sation, selective Barbary sheep hunting (only two species attested, and Ammotragus represents the 80%), a diversified lithic industry with a microlithic tool-kit mostly of quartz and quartz­ite. No evidence of pottery, and scarce emphasis toward seed processing and plant exploita­tion are shown.

 

Conversely, large accumulations of straw and plant remains indicate the occupation by later groups which had been first labelled as 'Mesolithic': a large settlement area, specialised functional areas, the emergence of a diversified technological inventory, like a well-developed pottery production, high incidence of grinding equipment and macrolithic industry on silici­fied sandstone are the main features. This phase features a broad spectrum of animal resources: several species are attested, small and big mammals, fish and birds. A particular form of Barbary sheep management has been also recognised (di Lernia-Cremaschi 1996b). Radiocarbon datings span from around 8,900 to 8,000 years B.P.

 

A systematic survey in the surroundings of the Acacus carried out in the last field-seasons provided new elements for the reconstruction of the settlement pattern and the technological organisation of the Early Holocene human groups (Cremaschi - di Lernia 1995; Cremaschi - di Lernia 1996b; di Lernia1996). Several sites along the lake basins in the dune valleys were found in the erg Uan Kasa and the Edeyen of Ubari. They show small dimen­sions, specialised tool-kits and evidence for selective hunting (Barbary sheep). Stratigraphic indications from the lake sediments provide us with some ante quem dates around 8,500 years B.P. for these sites, which can be considered 'Epipalaeolithic'. First time, the systematic analy­sis on raw material exploitation during the Holocene shows that the intensive use of quartz, but also silcrete, is a typical feature of 'Epipalaeolithic' groups (di Lernia - Cremaschi - Notarpietro 1997). Only very few scattered findings related to 'Mesolithic' occupations were found around the lakes, giving further evi­dence of an increasing sedentism during the ninth millennium in the Acacus range (di Lernia 1996).

 

The 255 cm-thick deposit at the Uan Tabu rock shelter provides further support to the cul­tural and economic differentiation of the early ninth millennium B.P. Three different units could be separated on the basis of their sedimentological and organic features. The first two units include a greyish sediment with ashes and plant fibres and are 120 cm-thick. Unit III is more brownish and is 80 cm-thick. Plant remains are extremely rare. The earliest dates come from Unit III, being 8,880±100 and 8,870±100 B.P. Unit II is dated between 8,800±100 and 8,840±90 B.P., and Unit I covers a period from 8,600±90 and 8720±110 B.P. Four combustion structures appear in both Unit I and Unit II, whereas only two of them are present in Unit III. Pottery is never frequent, but is differently distributed: in Units I and II there are five sherds per unit, and only one sherd is from Unit III. As for decorations, packed dotted zigzags dominate in all units, although another motif, showing plain zigzags on the rim, appears in Units I and II (Garcea 1996b): this motif is also attested at Fozzigiaren and Uan Afuda (di Lernia - Manzi in press; di Lernia in press). Blades and bladelets, as well as microlithic backed tools, are particularly frequent in in Unit III. Interestingly, quartz occurs with 8.3% of the total raw materials, but 97.1% comes from Unit 111 (Garcea 1997). Grinding stones are more frequent in Unit III (46 pieces) than in Unit I (14 pieces) and II (26 pieces) together. All faunal remains from unit III belong to Ammotragus lervia, whereas Histrix cristata appears in Unit II. Considering all these elements, two major cultural entities, as already observed³⁹, come to light: an upper one, including Units I and II, and a lower one represented by Unit III. The former was exclusively specialised in Barbary sheep hunting, the latter developed a broader exploitation of resources: diversified hunting, plant processing, use of ceramic containers.

 

Some new excavations at Uan Muhuggiag were carried out in the wadi area, a few metres from the shelter; a 200 cm-thick archaeological deposit was brought to light. Five layers were identified and grouped as follows: an upper sequence (layers 1-2) shows typical Middle Pas­toral Neolithic archaeological features; domesticated and wild animals are present. The lower sequence (layers 3-5) is characterised by a small number of artifacts, and higher incidence of microliths, with wild animals and no pottery. A human burial was found at the base of the series. Radiometric indications from the lower sequence span from 7,400 to 7,800 years B.P.: age, archaeological features and a broad spectrum of hunting, are all clues of a 'Mesolithic' occupation (di Lernia - Manzi in press).

 

Further evidence for a `Mesolithic' occupation could be found at Fozzigiaren (Mori 1965; Barich 1987b: 115-116) on the basis of the material culture, site formation process and faunal assemblage (di Lernia - Manzi in press). Other 'Epipalaeo­lithic' (MS; Lancusi TR; UI7; 94/38; 94/83; 94/84; 04/96; 94/102; 94/4; 94/8; mt126; mt135; mt145a; mt145b) and 'Mesolithic' sites (th49; th61; th89; th123; th131; Tagzelt 5; 94/62; 94/40; 94/77; 94/102; mt127) were identified during systetematic surveys in the Central Acacus and surroundings: some diagnostic materials, faunal remains and chronological indications if available were used in order to detect the kind of 'cultural' phase (Cremaschi - di Lernia 1995; 1996a).

 

In the northern Acacus, systematic excavations were carried out during the 70's and 80's (Barich 1987c). At Ti-n-Thora East, layer Rinf seems characterised by a high incidence of microlithic tool-kits, with mostly backed tools: the organisation of technology also seems to point to the use of a few lithotypes, with a greater account for quartz, quartzite and flint, rather than silicified sand­stone, which increases from the level Ri f to the level C (Barich 1974: 115); no potsherds were found at the bottom part of the layer (Rinf base), dated to 9,080±70 B.P. (Barich 1974: 149; Barich 1987b). As regards the economic strat­egies, Ammotragus is the most exploited; mammal (>70%), whereas a few other species are present; the grinding equipment counts a large number of specimens. By contrast, layers C are characterised by stone structures, a well-defined pottery production, mainly decorated with rocker stamp technique. The technological organisation shows an increase of silicified sand­stone and grinding equipment. All these data provide a clue for an increasing sedentism, as Barich observed (Barich 1987a), and also for a different organisation in the settlement pattern, as the changes in economic strategies and mobility suggest (di Lernia 1996; Cremaschi - di Lernia 1996b).

 

We can recognise a similar trend at Ti­n-Thora Two Caves (Barich 1987b), with a general change between layer III and layers II-I. They have a similar chronological framework and are stratigraphic succession. Also in this case, we prefer to think for a distinction rather than an undifferentiated cultural process.

 

To summarise, two different groupings (Tab. 1) can be made on the basis of the above mentioned features, which can tentatively be related to different social, economic and cultural entities (Cremaschi - di - Lernia 1995; di Lernia 1996) :

 

1 - Groups with a specialised extractive economy, logistic settlement pattern (mountain and ergs), and related high mobility; sites show simple structures (stone concentrations; wind­breaks; mostly unstructured hearths); the economic basis consists of a narrow spectrum of resources, with specialised exploitation of mammals, particularly Ammotragus lervia (>70%), and low plant processing. The lithic production shows some recurrent features (incidence of of quartz; etc.), whereas pottery is extremely rare. The chronological framework range spans from 9,800 up to around 8,800 years B.P. In preliminary phase, they were named as 'Epipalaeolithic'.

 

2 - Groups with a diversified extrative economy, an almost exclusive mountain environment settlement pattern and semi-residential occupations. Site formation features thick plant accumulations and more complex structures. Different animals are hunted and a more diversified technological inventory is present (pottery, basketry); grinding stones show a high incidence, probably in relationship to the increase of wild cereals gathering. The lithic industry indicates a development of macro-flake debitage, mainly of silicified sandstone, and quartz becomes very rare. Ceramic production is characterised above all by rocker impressions, in some cases with dotted wavy line motifs. Datings go from the beginning of the ninth millennium B.P., as the term 'Mesolithic' had been used to distinguish them from the 'Epipalaeolithic' ones.

 

It should be stressed that such a distinction must be used as a schematic framework, which obviously implies chronological overlapping, as well as different nuances in the archaeological indicators. What is more important is that, where systematic excavations were carried out, these two different groups were found in a stratigraphic sequence.

 

The recognition of a differentiation in two different cultural horizons of the 'pre-pastoral' period is not new for the North-African literature (e.g. Caneva 1988) but it is quite original for the Acacus mountain range (di Lernia 1996). 

 

Tab. 1 - Main features of the 'pre-pastoral' groups of the Tadrart Acacus mountain range.

3.2. The Nile Valley

 

As for the Middle and Upper Nile Valley (Fig. 2), the existence of two different cultural horizons within the Early Khartoum horizon has been pointed out on several occasions (e.g. Arkell 1972; Caneva 1988; Caneva - Marks 1990; Garcea 1994; Garecea 1996a). The so-called 'Wavy Line Culture' refers to the earlier hunting-gathering horizon of the Nilotic Holocene. A later phase includes several peculiar features, the most traditional being the occurrence of unburnished dotted wavy line and the absence of wavy line incisions. The earlier unit covers a long period of time, comprised between the ninth and the first half of the seventh millennium B.P., and the later unit dates from the end of the seventh millennium B.P. Differences can be noted in the technological products, going from pottery decorations and pastes, lithic typologies and raw materials (Caneva - Garcea - Gautier - Van Neer 1993). Further changes occur in the organisation of subsistence strategies. At Abu Darbein, which is the earliest Early Khartoum site, fishing was practised from the river bank instead of the centre of water courses by means of boats or rafts. Consequently, the catch only included smaller fishes at the earliest site, whereas much bigger fishes could be later on caught in higher waters (Peters 1991). Moreover, riverine resources and fishing activities gradually lost their importance in favour of a different subsistence base, that would be fully adopted later on with animal domestication and pasto­ral economy. 

 

The two units also vary according to their settlement pattern. During the earlier phase, all sites are aligned along the terrace of the Nile Valley. Later on, there was a westward shift of the course of the Nile, which took place between , the end of the seventh and the midst of the sixth millennium B.P. (Caneva - Garcea - Gautier - Van Neer 1993). At that time, they appear to be less dependent on riverine resources and begin to explore other environments, such as the savannah.

 

4. FINAL REMARKS

 

The evidence from the studied areas proves that different cultural phases can be distin­guished in the Early Holocene hunting-gathering communities. Environmental, chronological, and cultural diversities do exist between the Central Sahara and the Nile Valley. These differ­ences should be referred to specific and independent patterns of development.

In our view, if borderlines are necessary, subsistence economy and the related extractive or productive features of human groups should not be ignored. In addition to this distinction between food extractors and producers, we wanted to point out the existence of further sub­division within extractive groups, on the basis of their cultural and economic organisation.

 

Considering the theoretical panorama conce mg the discipline in general and our study area in particular, several terms can no longer stay in use, as they have lost their informative function. It also seems senseless to reject some terms, like 'Epipalaeolithic' or 'Mesolithic', and to save others, like 'Neolithic'. However, if we reject all terms, new ones have to be replaced. Some features, such as subsistence strategies, settlement pattern and mobility, appear more relevant than others and were taken into account by one of us (Savino di Lernia), by using conventional labels to define the Early Holocene groups of the Acacus and surroundings.

It is no longer possible that a single term could represent the enormous complexity of the archaeological record on which the contemporary, prehistoric research is based. A single term cannot indicate chronology, subsistence economy, and technological organisation altogether. As an example, extractive groups, such as hunters, gatherers, or fishermen are not alike, any more than productive groups, such as cattle herders, pastoralists, or farmers. Therefore, we should abandon the effort to find a single term conveying all specificities. We should rather aim at recognising possible cultural differences existing in the archaeological record. If this is possible, we suggest to employ terms referring to local cultural horizons in order to avoid further controversies and confusion. As for the Egyptian desert, a similar attempt has been already successfully made with respect to the Masara cultural unit of Dakhleh Oasis (McDonald 1991).

 

Thus, at least as far as the Libyan Sahara is concerned, the first interesting element, which is not new in the North-African literature (see Caneva 1988; Garcea 1993), but is, original for the Tadrart Acacus, is the identification of two distinct cultural phases (Cremaschi - di Lernia 1995), previously presented at the l0^th Pan-African Congress in Zimbabwe as 'Epipalaeolithic' and 'Mesolithic' (di Lernia - Cremaschi 1996a).

 

The theoretical consideration herein discussed and the available archaeological evidence seem to support the use of specific terminology: we suggest the term of Early Acacus for the phase with selective economy, high mobility, a specific technological assemblage, and a chronological range between 10.000 and 8.800 B.P., previously and preliminary called 'Epipalaeolithic'. Moreover, the term Late Acacus refers to the phase with a broad spectrum of resources, reduced mobility, increasing sedentism, a different technological assemblage, and a chronological range between the beginning of the ninth and the middle of the eighth millennium B.P., preliminary called 'Mesolithic'. It must also be precautionarly pointed out that the definitions of Early and Late do not imply an ineluctable, evolutionary sequence.

 

As regards the Nile Valley, we can point out that the general anthropological features, of the Early Acacus phase, including high mobility and selective resource exploitation, do not occur in the Nile Valley. The Late Acacus phase shares a similar pattern with the earlier phase of Early Khartoum with concern to their settlement patterns and broad spectra of resource exploitation. Finally, the later phase of Early Khartoum is not attested in the Central Sahara, where animal domestication occurs much earlier than in the Nile Valley.

 

 

REFERENCES

Agorsah K.E. 1990, 'Ethnoarchaeology: the search for a self-corrective approach to the study of past human behaviour', African Archaeological Review 8, 1990: 189-208.

Andah B.W. 1995, 'Studying African Societies in Cultural Context', in P.R. Schmidt and T.C. Patterson (eds.), Making Alternative Histories: The Practice of Archaeology and History in Non-Western Settings, Santa Fe 1995: 149-181.

Arkell A.J. 1949,  Early Khartoum, London.

Arkell A.J. 1972, 'Dotted wavy-line pottery in African prehistory', Antiquity, 46 (183), 1972: 221-222.

Atherton J.H.  1983, 'Ethnoarchaeologry in Africa', African Archaeological Review, 1, 1983: 75-104.

Barich B.E. 1974, 'La serie stratigrafica dell'Uadi Ti-n-Torha (Acacus, Libia)', Origini, 8, 1974: 7-184.

Barich B.E. 1987a, 'The Wadi Ti-n-Thora Facies, in B.E. Barich (ed.), Archaeology and Environment in the Libyan Sahara.The Excavations in the Tadrart Acacus, 1978-1983 (Cambridge Monographs in African Archaeology, 23. BAR Int. Series, 368), Oxford 1987: 91-112.

Barich B.E. 1987b, 'Research in the Central Acacus Massif', in B.E. Barich (ed.), Archaeology and Environment in the Libyan Sahara. The Excavations in the Tadrart Acacus, 1978-1983 (Cambridge Monographs in African Archaeology, 23. BAR Int. Series, 368), Oxford, 1987: 115-122.

Barich B.E. 1987c (ed.), Archaeology and Environment in the Libyan Sahara. The Excavations in the Tadrart Acacus, 1978-1983 (Cambridge Monographs in!African Archaeology, 23. BAR Int. Series, 368), Oxford 1987. Barich B.E.,  G. Belluomini, F. Bonadonna, M. Alessio and L. Manfra 1984, 'Ecological and cultural relevance of the recent new radiocarbon dates from Libyan Sahara', in L. Krzyzaniak and M. Kobusiewicz (eds.), Origin and Early Development of Food-Producing Cultures in North-Eastern Africa, Poznan 1980, PoznAn 1984: 411-417. Binford L.R. 1965, 'Archaeological systematics and the study of culture process', in D.L. Clarice (ed.), Models in Archaeology, London 1965: 109-166.

Binford L.R. 1968, 'Methodological considerations of the archaeological use of ethnographic data', in R.B. Lee and I. Devore (eds.), Man the hunter, Chicago 1968: 268-273.

Cahen D. 1985, 'Fonction, industrie et culture', in M. Otte (ed.), La signification culturelle des industries lithiques, Oxford 1985: 39-56.

Caneva I. 1988 (ed.), El Geili. The History of a Middle Nile Environment. 7000 B. C -A.D. 1500 (Cambridge Monographs in African Archaeology, 29. BAR Int. Series (424), Oxford 1988.

Caneva I., E.A.A. Garcea, A. Gautier and W. van Neer 1993, 'Pre-pastoral Cultures along the Central Sudanese Nile', Quaternaria Nova, 3, 1993: 177-252.

Caneva I. and A.E. Marks 1990, 'More on the Shaqadud Pottery: Evidence for Saharo-Nilotic Connections during the 6th-4th millennium B.C.', Archéologie du Nil Moyen, 4, 1990: 11-35.

Childe V.G. 1956, Piecing together the Past, London 1956.

Conkey M.W. 1990, 'Experimenting with style in archaeology: some historical and theoretical issues', in M.W. Conkey and C.A. Hastorf (eds.), The uses of style in archaeology, Cambridge 1990: 5-17.

Cremaschi M. and S. di Lernia 1995, 'The transition between Late Pleistocene and Early Holocene in the Uan Afuda Cave (Tadrart Acacus, Libyan Sahara). Environmental changes and human occupation', Quaternaire,6 (3-4), 1995: 173-189.

Cremaschi M. and di Lernia 1996a, 'Climatic changes and adaptive strategies in the Central Saharan Massifs: the Tadrart Acacus and Messak Settafet perspective', in G. Pwiti and R. Soper (eds.), Aspects of African Archaeology. Papers from the 10^th Congress of the Pan-African Association for Prehistory and Related Studies, Harare 1996: 39-52.

Cremaschi M. and S. di Lernia 1996b, 'Current research on the prehistory of the Tadrart Acacus (Libyan Sahara). Survey and excavations 1991-1995', Nyame Akuma, 45, 1996: 50-59.

Cremaschi M. and S. di Lernia (eds.) 1998, Teshuinat. Palaeoenvironment and Prehistory in South-western Fezzan (Libya), Milano.

Di Lernia S. 1996, 'Changing adaptive strategies: a long-term process in the Central Saharan Massifs from Late Pleistocene to Early Holocene. The Tadrart Acacus perspective (Libyan Sahara)', in G. Aumassip, J.D. Clark and F. Mori (eds.), The Colloquia of the XIII International Congress of Prehistoric and Protohistoric Sciences, XV. The Prehistory of Africa, Forli 1996: 195-208.

Di Lernia S. in press, 'Early Holocene "pre-pastoral" cultures in the Uan Afuda Cave, Wadi Kessan (Tadrart Acacus, Libyan Sahara)', in M. Cremaschi and S. di Lernia (eds.), Teshuinat. Palaeoenvironment and Prehistory in South-western Fezzan (Libya), Milano, forthcoming.

Di Lernia S. and M. Cremaschi 1996a, 'Analysis Of the Pleistocene-Holocene transition in the Central Sahara: culture and environment in the Uan Afuda Cave (Tadrart Acacus, Libya)', in G. Pwiti and R. Soper (eds.), Aspects of African Archaeology. Papers from the, 10^th Congress of the Pan-African Association for Prehistory and Related Studies, Harare 1996: 221-234.

Di Lernia S. and M. Cremaschi 1996b, 'Taming Barbary sheep: Wild animal amenagement in the Early Holocene hunter-gatherers at Uan Afuda (Libyan Sahara)', Nyame Akuma, 46, 1996: 43-53.

Di Lernia S. and G. Manzi in press, 'Funerary practices and anthropological features at 7,800 - 6,000 BP. Some evidence from Central-southern Acacus', in M. Cremaschi and S. di Lernia (eds.), Teshuinat. Palaeoenvironment and Prehistory in South-western Fezzan (Libya), Milano, forthcoming.

Di Lernia S., M. Cremaschi and A. Notarpiero 1997, 'Procurement, exploitation and circulation of raw material: analysis of the Early and Middle Holocene lithic complexes from South­western Libya (Tadrart Acacus and Messak Setiafet)', in R. Schild and Z. Sulgostowska (eds.), Man and Flint. Proceedings of the VII^th International Flint Symposium, Warszawa 1997: 233-242.

Flannery K.V. 1967, 'Culture history v. cultural process: a debate in American archaeology', Scientific American, 217, 1967: 119-122.

Garcea E.A.A. 1992, 'Il problema dell'adattamento dell'uomo all ambiente sahariano nelle trasformazioni climatiche degli ultimi 20.000 anni', in M. Lupacciolu (ed.), Arte e culture del Sahara preistorico, Roma 1992: 57-62.

Garcea E.A.A. 1993, Cultural dynamics in the Sahato-Sudanese prehistory, Roma 1993.

Garcea E.A.A. 1994, 'Mesolithic sequence in the settlement network of the Khartoum Province', in Ch. Bonnet (ed.), Études Nubiennes. Conference de Genève. Actes du VIP" Congres international d'etudes nubiennes, Geneve 1990, Genève 1994: 83-87.

Garcea E.A.A. 1996a, 'La culture des pêcheurs d'"Early Khartoum" un exemple de la Vallge du Nil', Prehistoire Anthropologie MdditerranJennes, 5, 1996: 207-214.

Garcea E.A.A. 1996b, 'The long sequence at Uan Tdbu (Libyan Sahara): Aterian and Holocene deposit', in G. Aumassip, J.D. Clark and F. Mori (eds.), The Colloquia of the XIII International Congress of Prehistoric and Proto­historic Sciences, XV. The Prehistory of Africa, Forli 1996: II83-194.

Garcea E.A.A. 1997, 'L'exploitation du quartz dans 1'Holocene saharo-soudanais', Prehistoire Anthropologie Mediterraneennes, 6, 1997, in press.

Geertz C. 1995, After the Facts. Two Countries, Four Decades, One Anthropologist, Cambridge 1995.

Gould R. 1980, Living Archaeology, New York 1980.

Hodder I. 1985, 'Post-processual archaeology', in M. Schiffer (ed.), Advances in Archaeological Method and Theory, VIII, New York 1985: 1-26. 

Hugot H.J. 1980, 'Préhistoire du Sahara