chapter 16

0 downloads 0 Views 3MB Size Report
mer of a bi-convex transversal section, which becomes plan-convex after ...... la table-ronde internationale organisée à Caen (Basse-Normandie – France), 14- ...
CHAPTER 16

FROM THE MOUSTERIAN OF ACHEULIAN TRADITION TYPE A TO TYPE B: A CHANGE IN TECHNICAL TRADITION, RAW MATERIAL, TASK, OR SETTLEMENT DYNAMICS? MARIE SORESSI v Abstract. Since the Mousterian of Acheulian tradition (MTA) was first defined, virtually no approach other than typology has been used to try to understand the behavioral significance of the MTA and its two variants, type A and type B, nor has any consensus been reached within this limited approach. Discussed in this paper is the author's approach of applying technological analyses to several MTA reference assemblages in order to gain new insights into the significance of the MTA. Major results show that behavioral changes leading from MTA type A to MTA type B were not related to a different technical tradition, to raw material constraints, or to a different specialized toolkit. Within the MTA type A, emphasis on the use of bifaces, on long-term organization of knapping activity, and possibly on hafting, describe the main differences between the two variants of the MTA. MTA bifaces were a source of flakes, were resharpenable multifunctional tools, and were preferred among other tools for transportation across the regional settlement. The significance of these changes is related to a change within the settlement dynamics of the peoples utilizing the same technical tradition, involving more long-term planning within the MTA type A than within the MTA type B. Interestingly, this behavioral change does not reveal a tendency toward more complexity but instead, toward more opportunistic behavior. Résumé. Depuis la définition du Moustérien de tradition acheuléenne (MTA), la typologie a été pratiquement la seule approche utilisée pour tenter de discuter la signification de ce faciès et de ces deux variantes, le MTA de type et le MTA de type B. De plus, aucun consensus n’a jamais été atteint sur ce point. J’applique ici une approche technologique à plusieurs ensembles MTA de référence pour tenter d’apporter de nouveaux éléments de réponse à cette question. Les principaux résultats de cette analyse montrent que les changements comportementaux du MTA A au MTA B ne sont pas liés à une tradition technique différente, à de nouvelles contraintes de la matière première ou bien à un outillage plus spécialisé. L’intérêt porté à l’utilisation des bifaces, à l’organisation à long terme des activités de taille dans le territoire, et probablement à l’utilisation d’emmanchement, dans le MTA de type A, 343

MARIE SORESSI sont les principales différences ente les deux variantes du MTA. Les bifaces MTA étaient des sources d’éclats, potentiels supports d’outils retouchés, des outils réaffûtables et ont été les objets préférés lors des déplacements dans le territoire. Le signification de ces changements du MTA A au MTA B sont donc liés à un changement dans l’organisation de l’occupation du territoire, impliquant une plus grande organisation à long terme dans le MTA A que dans le MTA B. On remarque que, d’une manière qui pourrait apparaître surprenante, ce changement de comportement va vers la simplification du MTA A au MTA B.

INTRODUCTION

FROM THE MOUSTERIAN OF ACHEULIAN TRADITION TYPE A TO TYPE B: A CHANGE IN TECHNICAL TRADITION, RAW MATERIAL, TASK, OR SETTLEMENT DYNAMICS?

Table 1. List of the studied assemblages: excavator, taxonomy and radiometric dating. Name (author, date of first publication)

Taxonomy

Le Moustier H (Peyrony 1930)

MTA type B

Pech-de-l’Azé I layer 4 (Bordes 1954-55) Pech-de-l’Azé I layer 5 (Bordes 1954-55)

MTA type A MTA type A/B

Pech-de-l’Azé I layer 6 (Bordes 1954-55)

MTA type B

Pech-de-l’Azé I layer 7 (Bordes 1954-55)

MTA type B

La Rochette MTA type A (Hauser 1911) La Rochette layer 7 (Delporte 1963)

MTA type A MTA type B

La Grotte XVI layer C (Rigaud et al. 1995)

MTA type A

Le Moustier G (Peyrony 1930)

MTA type A

Radiometric dating (mean age)

TL on flint: 50,300 ± 5500, 55,800 ± 5000 ESR: 43,000 ± 2300, 47,000 ± 2500 (Valladas et al. 1987; Mellars and Grün 1991) TL on flint: 42,500 ± 2000, 46,300 ± 3000 ESR: 39,700 ± 2400, 41,000 ± 2600 (Valladas et al. 1987; Mellars and Grün 1991) 14 C: 42,230 ± 1340 BP (Soressi 1999; 2002: 30) ESR: 49,000 ± 6000 (EU), 51,000 ± 7000 (LU) (Jones 2001; Soressi 2002: 31-32) ESR: 39,000 ± 2000 (EU), 47,000 ± 4000 (LU) (Jones 2001; Soressi 2002: 31-32) ESR: 42,000 ± 8000 (EU), 49,000 ± 7000 (LU) (Jones 2001; Soressi 2002: 31-32 ) ≥ age of the above layer: 52,500 ± 3400 14 C AMS: 52,500 ± 3400 (Soressi 2002: 34, after Valladas, unpublished) TL on soil: 64,600 ± 3100 (mean age) (Guibert et al. 1999)

The Mousterian of Acheulian tradition (MTA) describes a group of industries characterized by small and finely retouched cordiform bifaces and by backed knives (Peyrony 1920). These industries are recognized by their two type-fossils virtually unknown within other Mousterian variants (Mellars 1996: 172-73). Different regional variants of the MTA have been distinguished according to the presence of bifaces with a peculiar morphology (MTA with triangular bifaces within northern France, see Tuffreau 1990 and Cliquet 2001; MTA with bout-coupé form unique to Britain, see White and Jacobi 2002; MTA with cleavers in French Brittany, see Molines et al. 2001). In this paper, I will only discuss the original variant of MTA with cordiform bifaces (Bordes 1984: 137-38). MTA assemblages with cordiform bifaces occur in a specific geographic location, centred in the south-west of France, and in a specific time period, from approximately 50,000 years BP (see Soressi 2002: 9 for a synthesis of available radiometric dating). Neanderthal remains have been found associated with this industry (Maureille and Soressi 2000). Two variants have been distinguished within the MTA. MTA Type A has been characterized by the production and use of mainly bifaces, while MTA type B has been defined by the production and use of mainly backed knives and elongated flakes (Bordes and Bourgon 1951). The two sub-facies, MTA type A and MTA type B, are probably two successive episodes. Their relative chronology is based on their relative stratigraphic position at key sites: Pech-de-l’Azé I and IV, Le Moustier and La Rochette (Bordes 1984: 149; Delporte 1970). The available radiometric dating has not yet yielded the exact duration of each stage and the resolution of dating is not sufficient to establish a high resolution chronology for the MTA (table 1). Interestingly, several authors have supported the idea that the MTA type B, after its evolution from MTA type A, eventually evolved to the Chatelperronian (Bordes 1972). This evolution would have occurred under the influence of the first anatomically modern humans in western Europe (Mellars 1973, 1989) or independently of any foreign influence (Pelegrin 1995: 260-65; d’Errico et al. 1998; Soressi 2002: 277-84). Soon after the industry was recognized, the MTA has been interpreted as a cultural unit by D. Peyrony (1930) and later by F. Bordes (1953). It is well known that L. Binford and S. Binford (1966) argued that Mousterian variants were functional, while P. Mellars placed emphasis on the chronology of some of these variants, the

MTA being more recent than the Quina, which is itself more recent than the Ferrassie (Mellars 1965, 1996: 183-190, 325-331). This last point is generally confirmed by radiometric dating (e.g., Valladas et al. 1986), even if examples of recent dating have indicated that some Quina Mousterian might be more recent or equally as recent as the MTA assemblages (e.g., Jaubert et al. 2001; Turq et al. 1999). The other theory used to interpret Mousterian variants from a more dynamic point of view, put forth by H. Dibble as the scraper reduction model, has not been applied to the MTA (Dibble 1995). Since the term was initially defined, virtually no approach other than typology has been used to try to understand the behavioral signification of the MTA and its two variants. Yet, several factor might have played a role within the behavioral changes from MTA type A to MTA type B, including, for instance, changes in the quality or availability of raw material. The goal of this paper is to apply a technological approach to these industries in order to evaluate more successfully how appropriate each of the above hypotheses is in explaining the change from MTA type A to type B. With this aim in mind, a technological analysis has several advantages. It allows us to recognize the following: • The skill and knowledge involved within the knapping activity which were transmitted from one generation to another. Recognizing this makes it possible to test the similarity of the technical traditions under which different lithic assemblages were produced.

344

345

MARIE SORESSI sont les principales différences ente les deux variantes du MTA. Les bifaces MTA étaient des sources d’éclats, potentiels supports d’outils retouchés, des outils réaffûtables et ont été les objets préférés lors des déplacements dans le territoire. Le signification de ces changements du MTA A au MTA B sont donc liés à un changement dans l’organisation de l’occupation du territoire, impliquant une plus grande organisation à long terme dans le MTA A que dans le MTA B. On remarque que, d’une manière qui pourrait apparaître surprenante, ce changement de comportement va vers la simplification du MTA A au MTA B.

INTRODUCTION

FROM THE MOUSTERIAN OF ACHEULIAN TRADITION TYPE A TO TYPE B: A CHANGE IN TECHNICAL TRADITION, RAW MATERIAL, TASK, OR SETTLEMENT DYNAMICS?

Table 1. List of the studied assemblages: excavator, taxonomy and radiometric dating. Name (author, date of first publication)

Taxonomy

Le Moustier H (Peyrony 1930)

MTA type B

Pech-de-l’Azé I layer 4 (Bordes 1954-55) Pech-de-l’Azé I layer 5 (Bordes 1954-55)

MTA type A MTA type A/B

Pech-de-l’Azé I layer 6 (Bordes 1954-55)

MTA type B

Pech-de-l’Azé I layer 7 (Bordes 1954-55)

MTA type B

La Rochette MTA type A (Hauser 1911) La Rochette layer 7 (Delporte 1963)

MTA type A MTA type B

La Grotte XVI layer C (Rigaud et al. 1995)

MTA type A

Le Moustier G (Peyrony 1930)

MTA type A

Radiometric dating (mean age)

TL on flint: 50,300 ± 5500, 55,800 ± 5000 ESR: 43,000 ± 2300, 47,000 ± 2500 (Valladas et al. 1987; Mellars and Grün 1991) TL on flint: 42,500 ± 2000, 46,300 ± 3000 ESR: 39,700 ± 2400, 41,000 ± 2600 (Valladas et al. 1987; Mellars and Grün 1991) 14 C: 42,230 ± 1340 BP (Soressi 1999; 2002: 30) ESR: 49,000 ± 6000 (EU), 51,000 ± 7000 (LU) (Jones 2001; Soressi 2002: 31-32) ESR: 39,000 ± 2000 (EU), 47,000 ± 4000 (LU) (Jones 2001; Soressi 2002: 31-32) ESR: 42,000 ± 8000 (EU), 49,000 ± 7000 (LU) (Jones 2001; Soressi 2002: 31-32 ) ≥ age of the above layer: 52,500 ± 3400 14 C AMS: 52,500 ± 3400 (Soressi 2002: 34, after Valladas, unpublished) TL on soil: 64,600 ± 3100 (mean age) (Guibert et al. 1999)

The Mousterian of Acheulian tradition (MTA) describes a group of industries characterized by small and finely retouched cordiform bifaces and by backed knives (Peyrony 1920). These industries are recognized by their two type-fossils virtually unknown within other Mousterian variants (Mellars 1996: 172-73). Different regional variants of the MTA have been distinguished according to the presence of bifaces with a peculiar morphology (MTA with triangular bifaces within northern France, see Tuffreau 1990 and Cliquet 2001; MTA with bout-coupé form unique to Britain, see White and Jacobi 2002; MTA with cleavers in French Brittany, see Molines et al. 2001). In this paper, I will only discuss the original variant of MTA with cordiform bifaces (Bordes 1984: 137-38). MTA assemblages with cordiform bifaces occur in a specific geographic location, centred in the south-west of France, and in a specific time period, from approximately 50,000 years BP (see Soressi 2002: 9 for a synthesis of available radiometric dating). Neanderthal remains have been found associated with this industry (Maureille and Soressi 2000). Two variants have been distinguished within the MTA. MTA Type A has been characterized by the production and use of mainly bifaces, while MTA type B has been defined by the production and use of mainly backed knives and elongated flakes (Bordes and Bourgon 1951). The two sub-facies, MTA type A and MTA type B, are probably two successive episodes. Their relative chronology is based on their relative stratigraphic position at key sites: Pech-de-l’Azé I and IV, Le Moustier and La Rochette (Bordes 1984: 149; Delporte 1970). The available radiometric dating has not yet yielded the exact duration of each stage and the resolution of dating is not sufficient to establish a high resolution chronology for the MTA (table 1). Interestingly, several authors have supported the idea that the MTA type B, after its evolution from MTA type A, eventually evolved to the Chatelperronian (Bordes 1972). This evolution would have occurred under the influence of the first anatomically modern humans in western Europe (Mellars 1973, 1989) or independently of any foreign influence (Pelegrin 1995: 260-65; d’Errico et al. 1998; Soressi 2002: 277-84). Soon after the industry was recognized, the MTA has been interpreted as a cultural unit by D. Peyrony (1930) and later by F. Bordes (1953). It is well known that L. Binford and S. Binford (1966) argued that Mousterian variants were functional, while P. Mellars placed emphasis on the chronology of some of these variants, the

MTA being more recent than the Quina, which is itself more recent than the Ferrassie (Mellars 1965, 1996: 183-190, 325-331). This last point is generally confirmed by radiometric dating (e.g., Valladas et al. 1986), even if examples of recent dating have indicated that some Quina Mousterian might be more recent or equally as recent as the MTA assemblages (e.g., Jaubert et al. 2001; Turq et al. 1999). The other theory used to interpret Mousterian variants from a more dynamic point of view, put forth by H. Dibble as the scraper reduction model, has not been applied to the MTA (Dibble 1995). Since the term was initially defined, virtually no approach other than typology has been used to try to understand the behavioral signification of the MTA and its two variants. Yet, several factor might have played a role within the behavioral changes from MTA type A to MTA type B, including, for instance, changes in the quality or availability of raw material. The goal of this paper is to apply a technological approach to these industries in order to evaluate more successfully how appropriate each of the above hypotheses is in explaining the change from MTA type A to type B. With this aim in mind, a technological analysis has several advantages. It allows us to recognize the following: • The skill and knowledge involved within the knapping activity which were transmitted from one generation to another. Recognizing this makes it possible to test the similarity of the technical traditions under which different lithic assemblages were produced.

344

345

MARIE SORESSI

• The quality and the location of raw material, which are factors related to the energy spent to gather raw material with appropriate knapping properties. • The ergonomics of blanks and retouched tools, constraining how artefacts were used. • The duration and location(s) of the knapping activity, including the resharpening of artefacts. This last point presents us with a better understanding of the economy of raw material and the organization of knapping activities within a settlement dynamic.

FROM THE MOUSTERIAN OF ACHEULIAN TRADITION TYPE A TO TYPE B: A CHANGE IN TECHNICAL TRADITION, RAW MATERIAL, TASK, OR SETTLEMENT DYNAMICS?

THE STUDIED ASSEMBLAGES

Technological analysis was applied to the MTA eponym assemblages (Le Moustier layer G and H, Pech-de-l’Azé I, layer 4 to 7), as well as to three others assemblages (La Rochette MTA type A and layer 7, La Grotte XVI layer C). The studied sites are all located in France within the Perigord region, and they were all dated by radiometric methods (fig. 1; table 1). Within the studied assemblages, not only does the proportion of bifaces and backed knives change from MTA type A to MTA type B, but also the proportion of scrapers (fig. 2). The relative proportion of flakes produced from bifaces (with a soft hammer) or from cores (with a hard hammer) also changes, consequently allowing more emphasis to be placed on producing thin blanks or thick blanks (fig. 2; Soressi 2002: 72, 162, 211-12).

TECHNICAL UNITY WITHIN THE MTA

The MTA method of shaping bifaces is characterized by the shaping with a soft hammer of a bi-convex transversal section, which becomes plan-convex after retouch (Soressi 2002: 113-14, 94). The removals that create this volume are generally struck from the lateral sides of the bifaces (Soressi 2002: 113). The MTA method of producing bifaces is also characterized by the retouch of the edges on both sides of the points. The retouch is not symmetric on both sides of the points, often one edge is longer than the other (at Pech-de-l’Azé I, layer 4, 69% of the retouched edges of bifaces are asymmetric, n=58 bifaces). The retouch is generally unifacial and creates a plan-plan edge or a slightly convex-plan edge (57% and 18% of the cases at Pechde-l’Azé I, layer 4, n=137 bifaces). Also, the proximal overhangs between retouch negatives are removed by very light flaking which creates the effect of an edge with a regular angulation and a regular delineation (figs. 3a and 3b; 78% of the longest retouched edges at Pech-de-l’Azé I, layer 4 are regularized, n=55). The MTA bifaces could be described as finely retouched convergent scrapers on bifacially shaped blanks. Why this configuration, as well as the dimensions of the MTA bifaces, might be considered efficient will be discussed below. These characteristics are observed on bifaces from the four MTA type A assemblages as well as on bifaces from the four MTA type B assemblages (fig. 3c) (see details in Soressi 2002: 87-99).

Several methods of producing flakes thicker than the ones produced when shaping the bifaces were employed within MTA assemblages. The Levallois method, the discoid method and a peculiar method producing elongated blanks around part of the periphery of the core are documented (Soressi 2002: 240; Soressi n.d.). These flakes are retouched mainly in notches and denticulates (from 58% to 70% of the total number of retouched tools on hard-hammer flakes), whereas the flakes produced during the shaping of the bifaces often exhibit more scraper retouch or raclette retouch

346

347

Fig. 1. Location of sites discussed in this chapter.

MARIE SORESSI

• The quality and the location of raw material, which are factors related to the energy spent to gather raw material with appropriate knapping properties. • The ergonomics of blanks and retouched tools, constraining how artefacts were used. • The duration and location(s) of the knapping activity, including the resharpening of artefacts. This last point presents us with a better understanding of the economy of raw material and the organization of knapping activities within a settlement dynamic.

FROM THE MOUSTERIAN OF ACHEULIAN TRADITION TYPE A TO TYPE B: A CHANGE IN TECHNICAL TRADITION, RAW MATERIAL, TASK, OR SETTLEMENT DYNAMICS?

THE STUDIED ASSEMBLAGES

Technological analysis was applied to the MTA eponym assemblages (Le Moustier layer G and H, Pech-de-l’Azé I, layer 4 to 7), as well as to three others assemblages (La Rochette MTA type A and layer 7, La Grotte XVI layer C). The studied sites are all located in France within the Perigord region, and they were all dated by radiometric methods (fig. 1; table 1). Within the studied assemblages, not only does the proportion of bifaces and backed knives change from MTA type A to MTA type B, but also the proportion of scrapers (fig. 2). The relative proportion of flakes produced from bifaces (with a soft hammer) or from cores (with a hard hammer) also changes, consequently allowing more emphasis to be placed on producing thin blanks or thick blanks (fig. 2; Soressi 2002: 72, 162, 211-12).

TECHNICAL UNITY WITHIN THE MTA

The MTA method of shaping bifaces is characterized by the shaping with a soft hammer of a bi-convex transversal section, which becomes plan-convex after retouch (Soressi 2002: 113-14, 94). The removals that create this volume are generally struck from the lateral sides of the bifaces (Soressi 2002: 113). The MTA method of producing bifaces is also characterized by the retouch of the edges on both sides of the points. The retouch is not symmetric on both sides of the points, often one edge is longer than the other (at Pech-de-l’Azé I, layer 4, 69% of the retouched edges of bifaces are asymmetric, n=58 bifaces). The retouch is generally unifacial and creates a plan-plan edge or a slightly convex-plan edge (57% and 18% of the cases at Pechde-l’Azé I, layer 4, n=137 bifaces). Also, the proximal overhangs between retouch negatives are removed by very light flaking which creates the effect of an edge with a regular angulation and a regular delineation (figs. 3a and 3b; 78% of the longest retouched edges at Pech-de-l’Azé I, layer 4 are regularized, n=55). The MTA bifaces could be described as finely retouched convergent scrapers on bifacially shaped blanks. Why this configuration, as well as the dimensions of the MTA bifaces, might be considered efficient will be discussed below. These characteristics are observed on bifaces from the four MTA type A assemblages as well as on bifaces from the four MTA type B assemblages (fig. 3c) (see details in Soressi 2002: 87-99).

Several methods of producing flakes thicker than the ones produced when shaping the bifaces were employed within MTA assemblages. The Levallois method, the discoid method and a peculiar method producing elongated blanks around part of the periphery of the core are documented (Soressi 2002: 240; Soressi n.d.). These flakes are retouched mainly in notches and denticulates (from 58% to 70% of the total number of retouched tools on hard-hammer flakes), whereas the flakes produced during the shaping of the bifaces often exhibit more scraper retouch or raclette retouch

346

347

Fig. 1. Location of sites discussed in this chapter.

MARIE SORESSI

FROM THE MOUSTERIAN OF ACHEULIAN TRADITION TYPE A TO TYPE B: A CHANGE IN TECHNICAL TRADITION, RAW MATERIAL, TASK, OR SETTLEMENT DYNAMICS?

backed knives scrapers

According to the definition of a cultural unit recently put forward by S. McBrearty and A. Brooks (2000; see also Sackett 1990; Wynn 1996; Barham 2001), the MTA would be a cultural unit only if its technical tradition is different from contemporaneous and traditions from other regions, and was employed to produce tools used to satisfy equivalent needs. Comparisons with behaviors described within the contemporaneous Late Micoquian of central Europe (or industries of the Keilmesser group) have shown that the MTA should be considered a distinct cultural unit (Soressi 2002: 218-58 and Soressi n.d.).

EQUIVALENT RAW MATERIAL CONSTRAINTS

bifaces cores

soft hammer flakes

hard hammer flakes

Fig. 2. Relative proportion of bifaces and cores, of soft-hammer and hard-hammer flakes, of backed knives and scrapers within some of the studied MTA type A and MTA type B assemblages. The two different techniques of percussion are linked to the two main methods of production: production of biface with soft-hammer and production of thick flakes from cores with hard-hammer (a conservative approach was used to determine the technique of percussion; see details on criteria in Soressi 2002: 53-54, 81).

The same raw material was used for producing MTA type A and MTA type B assemblages within the same sites (from 97% to 100% of local flint). These raw materials were mainly of high quality flint from the Senonian strata. Bifaces and cores were produced on the same flint collected at the same locations (primary sources, in situ erosion of primary sources or secondarily derived alluvial sources; see Soressi 2002: 74-80, 163). The availability of this raw material was the same during the MTA and even after the MTA. At Le Moustier and La Rochette, for instance, researchers uncovered three and four lithic assemblages which were produced after the MTA on the same raw material (Peyrony 1930; Delporte and David 1966). Behavioral changes between MTA type A and MTA type B, therefore, do not correspond to changes in raw material constraints, as the nature and the availability of the raw material remained the same during this time period.

EQUIVALENT MULTIFUNCTIONAL TOOLS

(about 87% at Pech-de-l’Azé I, layer 4); the proportion of retouched backed knives varies from less than 1% at Pech-de-l’Azé I, layer 4, to 33% at La Rochette, layer 7 (Soressi 2002: 82, 87, 169). The unique method of shaping bifaces and the varying methods of producing flakes from cores within the MTA type A and the MTA type B assemblages provide evidence for the use of the same technical knowledge and knapping skill. This technical tradition had been transmitted from generation to generation across OIS 3 in the south-west of France.

Now let us address how the methods and techniques used to produce artefacts determine the ergonomics of these artefacts, and the adaptiveness of their form for use in a variety of tasks or in only some specialized tasks such as cutting or scraping. Within MTA type A assemblages, flakes are thin because they were mainly produced during shaping of bifaces with a soft hammer. If their transversal section were perfectly triangular, the angle of their edges deduced from their width and thickness would be 20° ± 9 at Pech-de-l’Azé I, layer 4, according to the trigonometric formula. Measurements of edge angle were all done using a calliper, which is the more precise way of measuring in this case (cf. Dibble and Bernard 1980). Measurements were made at 4 mm inside the edge, with the edge angle = arctan (thickness of the edge 4 mm inside the edge/4)/¹*180). The value for flakes produced by hard-hammer is 36° ± 10; the difference with soft-hammer flakes is significant (n =330 and n=93, P