some elongated flakes) (fig. 5,). Otherwise, the arris and the core edges guide remov-als on the trapezoidal or the pyramidal surfaces. Each debitage face shape reflects, therefore, a specific story, according to the debitage choice. The use of the core edges and of flakes leads to a flat debitage surface. The exploitation of two surfaces, on the contrary, results in wider and wider angles or it keeps good angles, and therefore in a pyramidal abandoned core. The core types (more or less cortical patches, removal organisation) could be a voluntary variation in the debitage in order to produce the most numerous and the most different kinds of flakes and for the longest time pos-sible. From flint could be produced smaller and thinner flakes (good quality of the stone or men's needs?).
The size comparison of the cores shows that, regardless of the raw materials, the values are similar. A preference goes to the smallest pebbles (between 30-40 to 60 mm). The polyedric cores are smaller than the other ones. The cores with a cortical surface produce, in addition, the few largest ones. If we compare the core size and the removal size, we can see that men really wanted a lot of small flakes. The largest cores, which are present, result in a lot of small removals. The microlithic assemblage is, therefore, not only imposed by the small pebble size but also by a definite choice.
Because of its long sequence, Kûlna cave brings much more "Taubachian" or "mic-rolithic" features to light. The main difference between the upper Micoquian levels, dated to the OIS 4, and Taubachian level 11 is based on the artefact size. The cores show the same processing system as in the Taubachian level.
40However, the cores are technologically more diversified and the stone acquisition is really different as well, with a preference of flint as main raw materials for the Micoquian.
414 0 B O Ë D A 1995.; R I N K et al. 1996.; M O N C E L 2003.
4 1 V A L O C H 1988.
101
type of cores frequency striking platform
or debitage surface debitage surface A: periph-ery (partial edge): part of a pebble or a flake as a natu-ral striking platform step
use of the pebble faces (cubic or quadrangular pebbles) several debitage faces (2, 3 or more) more or less large cortical patches cortical striking platform or use of the previous removals short debitage
Table III: The different types of cores in Tata 102
The men's needs: pebble tools, rough or retouched debitage products, retouched cores In the three assemblages, several functional blanks were used by h u m a n s : micro peb-ble tools (15-60 mm), some large pebpeb-ble tools (100-240 m m for Kûlna for example), flaking products, tools on flaking products or cores and whole pebbles (hammers or raw material stock?). Most of t h e m are, however, rough flakes. M e n obtained what they needed, using different methods: by a shaping system, which however stays on a very small scale and not easy to distinguish from flaking (both tool and core along their life, fig, 6.), and by a main debitage system, which gives flakes and even cores like blanks.
Some specific types of flakes were produced by the tool-makers. However, they were less demanding. N u m e r o u s broken flakes were used. This low degree of d e m a n d probably depended on the kinds of stones used, like quartzite or q u a r t z which easily break themselves, but not always.
- The products are very small, in the majority of the cases less than 30 m m (more than 80% less than 10-15 mm), but there are also large flakes up to 80 m m . The fine-grained stone flakes seem to be mainly small (perhaps for its stone quality, for a micro debitage or their needs). However, Taubach artefacts are in general longer (20 to 50 mm) (fig. 7-8.).
- The flakes are, in general, similar for all the kinds of raw material, short, thick and wide. T h e shapes are various; even the rectangular and triangular morphologies (parallel edges) are more numerous for fine-grained stones than for quartzite a n d q u a r t z . Some laminar flakes are also present, especially in silicites in Kûlna and Tata (bladelets).
- Between 25 and 50% of the flakes have a back (cortical or not), sometimes even two opposite backs. A debitage break could have been perceived like a debitage back (these flakes were used or retouched like entire flakes). It is possible that the wide and thick platforms could also have been perceived like a back.
Retouched artefacts make up less than 10% of the assemblage (fig, 9.), M o s t of the tools are side-scrapers and partially retouched points. The equipment is then lim-ited in category. Retouches were made on flakes of all size. However, h u m a n s used the shape diversity of the flakes for the retouch. Triangular flakes or blanks with two convergent edges are used first for making points. The edges are either entirely retouched or partially retouched. Q u a r t z i t e and fine-grained stone flakes are more often retouched on a long part of the available sharp side. The longest edge is chosen first for the side-scrapers. The retouches are, in general, opposite to the back, when it exists, or even to the platform. A back could have been retouched or used, especially when they were two on a piece.
The retouch does not, therefore, really change the flake shape in most of the cases.
The retouch is simple and thin in general, more invasive on o r t h o q u a r t z i t e in Kûlna, It is often small and steep on silicites, radiolarites and flint. Some silicite tools seem to have been used for a long time or the work is tidy on this kind of stone. Several series of retouches are indeed visible (due perhaps to a long utilisation or a resharp-103
ening linked to the quality of the raw material). Bifacial retouches are less numer-ous, except in Tata. Sometimes, flat retouches are located at the b o t t o m of a point or on the inferior face of the flake, often on fine grained stone artefacts or long distance stones like porcelanite stone in Kûlna (precious stones, curiosity, tool collecting?).
The isotopic stage 5 microlithic assemblages
The technological analysis of the three collections shows a lot of common points. The exploitation of raw material is always conducted to gather local and different rock types. The geological studies and the presence of some large pebbles prove that the great number of small pebbles was a h u m a n choice and not imposed by the environ-ment. The various stones could be employed like complementary raw materials, each one having its proper function (hardness, ability, pebble shape).4 2 The flaking system is dominant, mainly using cores with two debitage surfaces and the pebble shape. The flake types are diverse. In contrary, the tools are in small n u m b e r and the tool types are limited to side-scrapers a n d points. T h e bifacial retouch is rare, except in T a t a for small points. Shaping and flaking were certainly successively practiced on some cores to obtain more blanks. W e are in a voluntary microlithic world and a specific techno-logical world, really different from the Micoquian behaviour observed in the upper levels in Kûlna ( O I S 4).
How to explain the microlithic patterns?: activities, kind of sites, traditions?
The environmental conditions could involve with the most frequent production of small artefacts. The forest context, the temperate climate, good places for living near water springs, (archeological layers in travertins), river beaches, or caves (Kûlna) could explain an original behaviour adapted to the special climatic conditions. W o o d work, easy in a forest context, could also be an aspect of the main activities of these people (small stone tools used to work numerous wooden tool). W o o d pieces remains are indeed present in travertine G e r m a n sites. In most of these sites, mammals are great size species. Sometimes there are only two species, Rhinoceros, Elephants, Bison stags, M a m m o t h (Tata), Deers, Horses.4 3 A specialized activity in butchery is consequently possible to explain the technical patterns. But, a specialized butchery activity cannot be the only explanation to the small size of the lithic industry. Small flakes could be, of course, as good as other kinds of pieces for all kinds of activities. In Germany, microwear studies on small flakes give evidence of a use on vegetal prod-ucts.4 4 W o o d e n tools could be also well adapted to a diversified exploitation of what
*2 S V O B O D A 1 9 9 4 . ; M O N C E L - S V O B O D A 1 9 9 9 .
*3 G Á B O R I - C S Á N K 1968.; P A T O U - M A T H I S 1993.
*4 R I C H T E R 2001.
104