The Spatial Organization of Middle Palaeolithic Sites

Studies of the detailed spatial distribution of occupation residues on well defined living surfaces can be one of the most productive lines of research into the organization of Palaeolithic groups. Potentially, these studies can shed light on several features of crucial importance to understanding the behaviour and organization of the communities - the size of the social groups who occupied the sites, the duration of the individual episodes of occupation, the nature of the economic and technological activities carried out on the sites, and the ways in which all these activities were organized in relation to the total living space available within the sites. The kinds of patterns documented at such well known Upper Palaeolithic settlements as Pincevent, La Verberie and Etiolles in France, and Meer in Belgium, provide impressive illustrations of the behavioural insights which can be secured from detailed spatial studies of this kind (Leroi-Gourhan & Brezillon 1972, Leroi-Gourhan 1976, Audouze 1987,1988, Cahen et al 1979).

Interpretations from these studies must of course be treated with caution. As discussed earlier, it is rarely possible to be certain that the spatial distributions recorded in any occupation level reflect the residues of only a single, relatively brief episode of activity on the sites, and to this extent most of the recorded patterns must be seen as potential palimpsests of several separate episodes of occupation on the same location. There are also questions of variable patterns of survival of different kinds of occupation residues on the sites (especially in the case of faunal remains, as discussed in Chapter 7) and the possibility of various forms of post-deposi-tional disturbance of the original occupation patterns by either human or natural means. Whether these problems are as serious in practice as they may seem in theory is perhaps more debatable. As Meignen (1993: 161-3) and others have pointed out, there are reasons to think that in many sites (especially cave and rock shelters) the basic spatial constraints imposed by the size and character of the available occupation space within the sites led to at least broadly similar patterns of use during separate episodes of occupation, and accordingly to some relatively clear patterns in the resulting archaeological residues. Similarly, the relatively rapid rates of sedimentation and generally favourable conditions of preservation in many cave and rock-shelter sites seem to have led to remarkably complete survival of the original occupation residues.

The number of Middle Palaeolithic sites which have been subjected to this kind of detailed recording and analysis of occupation surfaces is unfortunately still fairly limited. The aim of the present chapter is to focus on a number of French sites where the most detailed recording of this kind has been car ried out, and to examine some of the patterns which emerge. The more general questions of the economic and social inferences which can be drawn from these studies will be discussed in the later sections of the chapter.

Grotte Vaufrey

The Grotte Vaufrey stands approximately 100 metres above the small valley of the

Ceou, only 2 km to the south of its confluence with the Dordogne valley (Rigaud 1988). The cave is cut into the local Coniacien limestone formation and faces almost due west, receiving full sunlight during the middle and later parts of the day. It has a high and impressive entrance but extends as a major chamber for a distance of only 15-20 metres from the cliff face (Fig. 9.1). The total area of the sheltered part of the cave is around 170 square metres.

Figure 9.1 Plan and section of the excavated zone in the Grotte Vaufrey. After Rigaud 1988.

Figure 9.1 Plan and section of the excavated zone in the Grotte Vaufrey. After Rigaud 1988.

In front of the cave the hillside drops away steeply towards the valley, requiring a strenuous climb to gain access from the valley floor. Access from the adjacent plateau above the cave is rather easier but still requires navigation of a steep and difficult slope. In view of these difficulties of access it is perhaps not surprising that the site seems to have been occupied by Palaeolithic groups on a relatively ephemeral, sporadic basis throughout the whole of the occupation sequence.

Excavations carried out by Jean-Philippe Rigaud between 1969 and 1982 explored 5 metres of deposits, apparently spanning the period from isotope stage 10 or 11 to the earlier part of the last glacial (Rigaud 1988). Most of the archaeological levels proved to be rather poor in artefacts and only level VIII provided a sufficient concentration of material within a single, well defined occupation horizon to allow the finds to be analysed in detailed spatial terms. Taxonomically, the industries from this and the immediately adjacent levels have been attributed by Rigaud to an early form of Typical Mous-terian, characterized by a moderate percentage of racloirs and notches/denticulates, and showing highly developed use of Leval-lois techniques. On various geological and faunal grounds the occupation of this level is attributed to isotope stage 7 or the earlier part of stage 6 and is dated on the basis of a uranium series of measurements of adjacent flowstone deposits to ca 170-200,000 BP.

The total area occupied by the deposits of layer VIII is estimated by Rigaud to be in the region of 160 square metres, of which some 90 square metres were explored and fully documented in the recent excavations (Fig. 9.1). It is not certain that this represents the whole of the occupied area in this level but the overall distribution of occupation material suggests strongly that the excavated area incorporates at least the major zone of human activity in the site. To summarize analyses provided by Rigaud & Geneste (1988), Simek (1988) and

Binford (1988), the most significant features of the distribution patterns are as follows:

1. Perhaps the most striking feature is the highly localized distribution of the archaeological material, concentrated mainly towards the southern wall of the cave. As shown in Fig. 9.2, over 90 percent of the artefacts were found here, occupying an area of only 30-40 square metres and mainly distributed between a scatter of large limestone blocks which were evidently present on the cave floor before the human occupation. As Rigaud and Geneste (1988) point out, there can be little doubt that the human groups who occupied the site were very small -possibly no more than three or four individuals during any episode of occupation.

2. The distribution of lithic flaking debitage in the site (particularly that represented by very small flakes, less than 1.5 cm in length) reveals three main concentrations, each measuring only 3-4 square metres (Fig. 9.2). The principal concentration is located along the south wall of the cave (between the scatter of large stone blocks noted above), while the others lie approximately 3-4 metres to the north and east. Refitting studies demonstrate that flaking was carried out in situ in these locations (Fig. 9.3) but have not so far produced clear links between the material in the different clusters. Whether or not the different episodes of flaking represented in the clusters were carried out during precisely the same phase of occupation on the site or during a succession of separate occupations must remain open.

3. The distribution of other categories of lithic artefacts (Levallois flakes, complete nodules of raw material and various forms of retouched tools) conforms broadly to the same distributional pattern and shows the same general concentration mainly towards the southern wall of the cave. Nevertheless, it is clear that these categories of material are rather less tightly concentrated in this partic

Figure 9.2 Overall distribution of lithie artefacts in layer VIII of the Grotte Vaufrey. The zones of highest density are enclosed by dotted lines. The outlines of the large stone blocks on the floor of the cave are also indicated. After Rigaud & Geneste 1988.

Figure 9.2 Overall distribution of lithie artefacts in layer VIII of the Grotte Vaufrey. The zones of highest density are enclosed by dotted lines. The outlines of the large stone blocks on the floor of the cave are also indicated. After Rigaud & Geneste 1988.

ular area and more widely distributed over the whole of the cave interior. Thus separate concentrations of complete Levallois flakes were detected towards the eastern end of the cave (Fig. 9.4), while the distribution of unworked or partially flaked flint nodules was similarly dispersed and patchy in several different areas of the site (Fig. 9.5). Unfortu nately, no detailed information is provided in the published report on the distribution of flint cores.

4. A detailed study of the distribution of different categories of retouched tools is unfortunately hampered by the small numbers of pieces recovered for most of the tool

Figure 9.3 Spatial distribution of two major groups of refitted artefacts in layer VIII of the Grotte Vaufrey (see also Fig. 3.2). After Rigaud & Geneste 1988.

Figure 9.3 Spatial distribution of two major groups of refitted artefacts in layer VIII of the Grotte Vaufrey (see also Fig. 3.2). After Rigaud & Geneste 1988.

types. However, there is an indication that retouched tools are generally more widely distributed on the site than are the products of primary flaking debitage, and there are some hints of variable patterns in the distribution of the two major retouched tool categories - racloirs and notches/denticu-lates. From the data provided by Rigaud and Geneste (1988: 608-9) it appears that notches and denticulates are more concentrated adjacent to the main debitage clusters (especially in the main area of debitage along the south

111+ lili 6-10

Figure 9.4 Distribution of Levallois flakes in layer VIII of the Grotte Vaufrey. After Rigaud & Geneste 1988.

111+ lili 6-10

Figure 9.4 Distribution of Levallois flakes in layer VIII of the Grotte Vaufrey. After Rigaud & Geneste 1988.

wall) while racloirs seem to be more widely distributed in the peripheral parts of the occupation zone (Figs 9.6, 9.7). In view of the small sample sizes of these tool types, however, it might be premature to attach too much weight to this pattern.

5. The most difficult and controversial patterns to interpret in the Vaufrey deposits are those relating to the distribution of animal remains (Fig. 9.8). Whilst there is general agreement that some component of the animal remains almost certainly derives from

Figure 9.5 Distribution of unworked river cobbles in layer VIII of the Grotte Vaufrey. After Rigaud & Geneste 1988.

Was this article helpful?

0 0

Post a comment