Préhistoire et géomorphologie : quelques réflexions d'ordre méthodologique - article ; n°1 ; vol.9, pg 121-135

Jacques Besançon

Description

Collection de la Maison de l'Orient méditerranéen. Série géographique et préhistorique - Année 1979 - Volume 9 - Numéro 1 - Pages 121-135
The region of the lower Nahr el Kebir has much to offer of interest to the prehistorian. The geomorphologist can also be greatly assisted in his efforts to date Quaternary deposits and relief forms, whenever the artifacts found are diagnostic of a certain phase. However, a surface-collection cannot, by itself, be taken to represent an industry. It was only by the combination of geomorphological criteria and the detailed analyses of the artifacts that we were able to extract from the latter any clear and meaningful order. The following is a brief review of the methods we used, which we considered indispensable if the full potential of the lithic evidence was to be realised.
1. Sites in stratigraphie context
All the artifacts found in geological context were transported, then dropped, at the same time and by the same processes which produced the alluvial fills in which they occur. However, the resulting assemblages present a misleading picture of the lithic industry from which they were originally derived ; pieces made of soft or fragile stone are altered in shape or have disappeared ; selective water-sorting is carried out by the river ; the material is usually contaminated by the inclusion of earlier fill alluvium (and its artifacts, if any).
a. Fluviatile contexts : It can clearly be seen that the most recent deposits have the most typological mixture. There are even rolled potsherds in the Holocene formation (QO) which forms the present flood-plain of the Nahr el Kebir. Of the three Würmian fills, the last (Jraimaqiye Formation) is identified by the presence in it of post-Levalloiso-Mousterian elements. In the QIa and QIb (Ech Chir Formation) the artifacts are relatively more numerous ; typologically one can see no evolution, and these two substantial fills can only date back to the first part of the Würm. The Rissian deposits are more homogenous and at the same time (considering their thickness) poorer. None of the pieces recovered from within Rissian stratigraphic fills in section could give any unequivocal chronological information.
The pre-Riss deposits did not produce sections. However, from a study of the artifacts found on the surface, useful results were eventually obtained.
b. Fluvio-marine contexts : We were not able to collect any artifacts from within any purely marine deposits, but there is a fluvio-marine platform at Hennadi, near the village of Khellale, with a fluviatile stratum at the base, which con tains large numbers of Middle Acheulean artifacts, far more weathered than a few pieces of Levailois debitage which also occur. An older lens of fluvio-marine deposits also exists at Jabal Idriss.
c. Conclusions : At the Nahr el Kebir, the artifacts in stratigraphie context within the gravels could, by themselves, only rarely furnish enough data for dating purposes. Würmian and Holocene deposits provided the clearest evidence. For the others, given the limited number of typologically diagnostic pieces, we were obliged to base our hypotheses mainly on geological criteria. It is the extraordinary erosive force which Mediterranean streams can exert when in spate which is responsible for the wide dispersal of artifacts, usually into the body of the fill, but an additional factor is the long-lasting nature of the silicious material from which the artifacts were made, which favours their survival until the next climatic cycle, when they could be included into later fills together with younger material. The erroneous back-dating of the terraces is a real danger (cf the group of Late Acheulean pieces found inside the Würmian deposits at Jbarioun).
2. Surface sites
On the present-day surfaces, sites with several hundred flint artifacts were often encountered.
a. Processes leading to concentration of artifacts : These rich sites are the resuit largely of two geodynamic agents, namely, selective erosion and soil-formation. Whenever there is good drainage, the soils produced in Mediterranean regions are red-brown to red, clayey, and decarbonised, so that the only elements able to remain in place are the insoluble minerals, that is, residual clay, pebbles of volcanic rock, and flint. The surface horizon, freed of carbonates which once formed the major portion of the fill, receives from the original bedrock minerals, which were up till now thinly dispersed in it. Simultaneously, part of the colloids, silts and sands are removed by slope-wash and sheet-flood, leaving behind a paving of coarse pebbles. At the end of these processes it may happen that ail that remains of a terrace are a few dozen large boulders (Jabal Aayoub). In short, a combination of erosive and soil-forming agencies produces a stoney cover. In it, the percentage of limestone pebbles decreases accord ing to age. None are left in the pre-Günz raised beach of Mchairfet es Samouk.
b. Typological ambiguities : The concentration of the coarse fraction at the surface level enables us to apply statistical methods to the typology of the numerous flint artifacts which are included, but mixture of industries is the rule. This could have resulted from (1) mixture during the deposition phase (see above) or (2) from sub-aerial erosion or weathering, which removes some artifacts while possibly adding others from extraneous locations ; finally (3), successive human populations could have dropped their artifacts into the same surface (in fact we found a fairly general 'pollution' by Neolithic material). It is therefore incumbent on us to separate the industries.
c. Criteria : To do this we were obliged to combine the results of analyses on :
1) the degree of conservation of the morphology of each piece ;
2) the amount of chemical alteration the blank had suf- fered, and
3) the degree to which it corresponded to the known typology of like age from stratified cave or shelter sites.
Only in the case of artifacts from marine alluvia can one discard pieces which are rolled ; a very pronounced usage could indicate that it was 'borrowed' from an older beach. Long exposure at the surface combines with the effects of demineralization and remineralization, and can finally result in a piece of flint losing a substantial amount of weight. We paid particular attention to the readily visible geochemical changes and modifications, such as the depth and thickness of the patina : it was found that sorting the pieces based on their patina cross-cut any sorting based on the degree of rolling. It also enabled us to eliminate later, man-made additions during the Neolithic or Recent times.
d. Correlation with geomorphological data : It is easy to distinguish fluviatile from marine deposits. In the Levant, the former correspond to the pluvials (glacials of higher latitudes), while the latter refer to interpluvials (interglacials). For the rest, the establishment of correlations between Quaternary alluvia and topographic layers is made difficult due to the remarkable efficiency of the processes of erosion in the Mediterranean regions ; the top of two neighbouring lenses of alluvium, both of different epochs, could today be found at the same height (e. g. Roudo Superieur and Jabal Berzine). Conversely, contemporary terrace remnants are often found at very different heights (e. g. Dahr el Ayani and Dahr Ouadi Hassane).
The usual sedimentological criteria are not of much assistance, because successive formations often present strong similarities, even in their variation of facies. Without being able to use the detailed techniques of morphometry and morphoscopy of sand, minerais, clays, pollens etc., we utilised the nature and state of the soils, more especially of the least mobile fraction (gravels, pebbles, boulders) in applying the principles of geochemical analysis already employed in prehistoric typological analysis.
None of the Holocene or Würmian terraces have even slightly evolved soils. On the latter, the flints show no trace of chemical alteration, but the basalt pebbles have already been eliminated. In the paving of Rissian terraces, the limestone pebbles, still intact, form the majority, while the flint shows a light patina. The Hennadi platform (Mindel-Riss) retains, as an exception, a fairly thick clayey layer of red soil, locally completely decarbonised. A portion of the limestone pebbles are truncated. At the summit of Jabal Berzine (Mindel) one cannot find any limestone pebbles, the greenstone rocks are rare, and besides the heavily patinated artifacts, one can observe ferruginised and light-weight flints.
3. Conclusions
The artifacts at present on the surface are not usually considered useful for the establishment of a geomorphological chronology. In the Levant, it appears to us that they can be made to divulge information, on condition that account is taken of their typology, state of conservation, chemical alterations, etc. At the same time, geochemical obliterations, and especially the patinas, as well as the nature of the soils, must also be taken into consideration.
15 pages
Source : Persée ; Ministère de la jeunesse, de l’éducation nationale et de la recherche, Direction de l’enseignement supérieur, Sous-direction des bibliothèques et de la documentation.

Informations

• Publié par: DIALOGUES_D-HISTOIRE_ANCIENNE
• Publié le: 01 janvier 1979
• Nombre de lectures: 26
• Langue: Français
• Poids de l'ouvrage: 3 Mo

Extrait