Pointed forms

In conventional conceptions of Middle Palaeolithic technology various forms of points are generally seen, along with side scrapers, as amongst the most distinctive and characteristic retouched tool forms (e.g. de Mortillet 1883). The diagnostic features are two carefully and more or less symmetrically retouched edges, usually shaped by flat, invasive retouch, which converge towards the distal end of the flake to form a fairly sharply

Arrowhead Drawings
Figure 4.13 Examples of various pointed forms from French Middle Palaeolithic sites. After Bordes 1961a.

defined point (Fig. 4.13). The traditional notion was that these pieces functioned primarily as the hafted tips of hunting spears; as discussed below, however, this interpretation would not now be accepted for more than a small proportion of conventional pointed forms.

The classic taxonomy of these pointed forms was formulated in the typology of François Bordes (1961a) (Table 6.1). According to Bordes at least three main variants can be differentiated, representing two major functional categories: first, 'true' Mousterian points, subdivided into ordinary and elongated variants, which he felt probably (along with his category of retouched Levallois points) represented true, hafted missile heads; and second, a variety of forms of convergent and déjeté racloirs, which he believed were functionally separate from true points and represented specialized variants of his racloir category. The distinctions between these categories of true points and convergent or déjeté racloirs has always been one of the more controversial aspects of the Bordes typological system and the one most difficult to apply in practice. Bordes could give only general and subjective guidelines for distinguishing between the different forms. Thus true points were defined essentially by their overall symmetry, regularity and relative pointedness while convergent racloirs were generally thicker, less regular and less intuitively appropriate for use as missile tips (Bordes 1961a). The well known cartoon by Pierre Laurent (1965: 39) epitomizes the dilemma faced by many workers applying these distinctions to the wide spectrum of pointed forms documented in many Middle Palaeolithic industries. The one major variant in Bordes' taxonomy which can be identified in a fairly objective way is the category of déjeté racloirs, which is defined not so much by the overall form of the tools as by the distinctively angled orientation of the retouched edges in relation to the main flaking axis of the parent flake (Bordes 1961a).

The main problems posed by the analysis and interpretation of these pointed forms are similar to those involved in the interpretation of various racloir forms discussed in the preceding section:

1. To what extent do these represent intentional, deliberate forms imposed on the tools in the initial stages of manufacture, either in terms of specific design norms or intended function?

2. How far can we identify significant distinctions within the category of pointed forms - either in morphological terms or in terms of intended tool function?

Both issues have again been brought into focus by the recent tool reduction models of Dibble and others (e.g. Dibble 1984a,b, 1987a,b,c, 1988a,b, 1989, 1991a,b; Rolland & Dibble 1990; Dibble & Rolland 1992; Rolland 1990; Jelinek 1976; Holdaway 1989). As in the case of the various side-scraper forms discussed above, Dibble (1984a,b, 1987a,b,c) has argued that very few, if any, of the conventional forms of Mousterian points and convergent or déjeté racloirs were conceived and manufactured from the outset as such but represent heavily reduced versions of much simpler tool forms that were successively resharpened and remodelled in the course of use. Specifically, he envisages a sequence in which certain forms of flake blanks were transformed from simple, single-edged racloirs, to more heavily reduced double-edged forms and finally into forms in which the two retouched edges eventually converged to produce a point (Fig. 4.12). According to this model there is no significant distinction, either conceptually or functionally, between the various categories of pointed forms and most of the simpler racloir types. Nor, in his view, can we make any distinction between the heavily reduced forms of convergent and déjeté racloirs and pieces which Bordes and others have always regarded as 'true' points in the sense of func tionally specialized missile heads (Dibble 1989; also Holdaway 1989).

As in the analogous models of lateral/ transverse racloir reduction sequences, Dibble cites two specific features in support of these interpretations (e.g. 1984a, 1987a,b): first, the fact that the majority of pointed forms carry significantly heavier patterns of retouch than do the simpler, single and double-edged racloir types; and second, that these pieces tend to show smaller ratios of tool surface area in relation to the sizes of the striking platforms of the parent flakes (Fig. 4.9). Both observations, he argues, are consistent with the view that points have been subjected to much heavier resharpening and reduction than have the morphologically simpler racloir types.

My own impression is that while Dibble may well have identified a significant component of variation in some of the documented point and convergent racloir forms, this is not sufficient to dismiss the whole status of point types as an intentional tool form in the Middle Palaeolithic. The features I would emphasize in this context are as follows:

1\ It seems evident from a visual examination of the tools that not all of the forms of convergent racloirs and Mousterian points can be dismissed as simple end-products of racloir-reduction sequences of the kind envisaged by Dibble. As Bordes and others have emphasized, many of these pieces show an overall regularity and clear bilateral symmetry in form (as reflected for example in the length, shape and overall treatment of the retouched edges) which seems to argue strongly for some kind of design-norms or mental templates in the minds of the flint workers (Bordes 1961a; Mellars 1964; Callow 1986a). Arguably, the most significant feature of the tools, however, is the specific treatment applied to the tips and bases of the implements. In the case of Bordes' categories of true Mousterian points and retouched

Levallois points, for example, the retouch was often applied in a highly selective, discontinuous way which was clearly intended to improve the sharpness and regularity of the point of the tool rather than to modify, let alone resharpen, the lateral edges (see Fig. 4.13). Similar observations can be made on the treatment applied to the bases. Here the retouching often extends around the base in such a way that the original striking platform of the flake has been totally removed. In other cases there is evidence for extensive, invasive flaking on the ventral surface which usually has the effect of either reducing or obliterating the original bulb of percussion (see Figs 4.13, 4.14). It is difficult to see how any of these features can be attributed to an opportunistic ad hoc resharpening of the edges of simple side-scraper forms. As discussed further below, these features are more likely to reflect deliberate attempts to accommodate the bases of the tools to specific hafting procedures.

2. Recent research seems to reveal some clear patterns in the deliberate selection of certain specific forms of flake blanks for the production of particular forms of pointed tools. The best evidence has come from the studies of Eric Boeda (1988c) on the material from level IIA at the site of Biache-Saint-Vaast. Thus Boeda points out that virtually all the typical specimens of convergent racloir and Mousterian point forms represented in this industry seem to have been manufactured on one particular form of flake blank which was produced at one specific point in the overall core reduction sequence - namely, the third-order removals in his strategy of 'bipolar recurrent' Levallois techniques (see Chapter

3, Figs 3.10, 3.11). All the simpler forms of single- and double-edged racloirs, by contrast, show no obvious selectivity, and seem to have been manufactured fairly indiscriminately from a wide variety of flake blanks (Boeda 1988c: 211, Table 5). The obvious conclusion, as Boeda points out, is that these

Hombre Delauricocha Formas Vida

Figure 4.14 Pointed forms with basal trimming of the bulbar surfaces. From Bordes 1961a and other sources. It is most likely that the thinning of the bases of the tools was related to some form ofhafting procedures for these pieces (see Fig. 4.15).

Figure 4.14 Pointed forms with basal trimming of the bulbar surfaces. From Bordes 1961a and other sources. It is most likely that the thinning of the bases of the tools was related to some form ofhafting procedures for these pieces (see Fig. 4.15).

specific forms of flake blanks were either produced, or at least selected, deliberately with production of the convergent racloir forms in mind.

3. These conclusions are further reinforced by results of recent micro-wear studies. The clearest data come from Sylvie Beyries' studies of the convergent racloirs from Biache-Saint-Vaast, and Patricia Anderson-Gerfaud's similar studies of the material from the open-

Hombre Delauricocha Formas Vida

Figure 4.15 Convergent raeloir forms showing micro-wear traces of hafting, from the site of Biache-Saint-Vaast in northern France. After Beyries 1988b. The horizontal lines shown against each piece indicate the extent of the hafting traces along the length of the tools (see Beyries 1988a: Fig. 12.6).

Figure 4.15 Convergent raeloir forms showing micro-wear traces of hafting, from the site of Biache-Saint-Vaast in northern France. After Beyries 1988b. The horizontal lines shown against each piece indicate the extent of the hafting traces along the length of the tools (see Beyries 1988a: Fig. 12.6).

air site of Corbiac in southwestern France. At both sites it wTas found that virtually all specimens of typically convergent racloirs and Mousterian points showed evidence of deliberate hafting, in the form of distinctive patterns of abrasion and polishing confined exclusively to the lower, proximal parts of the tools and visible on both the dorsal and ventral surfaces (Fig. 4.15) (Beyries 1988a,b; Anderson-Gerfaud 1990). Anderson-Gerfaud argues that these seem to reflect hafting of tools in a relatively loose form which allowed some movement between the tool and associated haft during use. The functions documented showed more variation, ranging from the whittling of wood to the scraping of bone or skins, but it is significant that all these use-wear traces were normally confined to the distal ends of the tools which extended beyond the encasing hafts (Fig. 4.15) (Anderson-Gerfaud 1990). None of this evidence easily supports the notion that all the retouch applied to these tools was due to the resharpening of utilized edges, since in most cases the retouched edges extended well below the parts encased in the hafts. In any event it is significant that with the exception of these typically convergent forms very little evidence for deliberate hafting has so far been documented on any other forms of Mousterian tools (Beyries 1987, 1988a; Anderson-Gerfaud 1990). The evidence provides strong support for the idea that many examples of convergent raeloir forms were intended for a specific economic function involving the use of a haft, and that the overall form and design of the tools was dictated largely by this anticipated function.

Figure 4.16 Diagram to illustrate how the deliberate production of typical pointed forms (shown on the left) can result in identical patterns of edge retouch to those resulting from the progressive 'reduction' of different racloir forms, as envisaged in the tool-reduction models of Dibble (see Fig. 4.12).

Figure 4.16 Diagram to illustrate how the deliberate production of typical pointed forms (shown on the left) can result in identical patterns of edge retouch to those resulting from the progressive 'reduction' of different racloir forms, as envisaged in the tool-reduction models of Dibble (see Fig. 4.12).

4. It is difficult to see how the specific metrical features documented by Dibble (1984a, 1987a,b,c) for convergent scrapers and Mous-terian points can be used to argue for his racloir-reduction model rather than for the alternative interpretations suggested above (Fig. 4.9). As Dibble (1987a) points out, both the major features which he has documented (i.e. relatively heavy retouch on the worked edges and apparently reduced areas of tool surface in relation to the sizes of the associated striking platforms) can be seen essentially as a reflection of the amount of deliberate shaping (or in his terms reduction) applied to the tools. But both these features are equally consistent with the hypothesis that deliberate retouching and reduction of tools was designed not simply to resharpen damaged edges but to allow a specific overall form to be imposed on the finished tools. If convergent racloirs and points were produced to a large extent according to an explicit design, then this would inevitably involve in many cases a substantial degree of reduction of the original flake blanks, with consequently heavy retouching on certain parts of the tool edges, to achieve this overall, preconceived form (Fig. 4.16). Dibble's observations on these features are interesting and relevant but hardly allow one to choose objectively between the alternative interpretations discussed above.

The final question of whether there were significant morphological or functional distinctions within the broad category of pointed forms is by far the most difficult issue to resolve. The problems stem partly from the scarcity of detailed studies of the overall range of morphological variation of different pointed forms in particular industries and partly from the sparsity of detailed use-wear studies. The long-standing debate over the existence of spear heads in the Middle Palaeolithic seems impossible to resolve with complete certainty. Both Anderson-Ger-faud (1990) and Beyries (1987, 1988a) have stressed that in their recent studies of micro-wear patterns there is as yet no unambiguous evidence for the existence of missile points -although it is arguable that the use of Mous-terian points for this function might well leave few discernible traces in the micro-wear data (cf. Jelinek 1988b). John Shea (1989,

Was this article helpful?

0 0

Post a comment