Levallois techniques have been recognized as one of the most distinctive hallmarks of Middle Palaeolithic technology since the original definition of the Mousterian by Gabriel de Mortillet in the late nineteenth century (de Mortillet 1883: 240, 255; see also Commont 1909: 122). The central and diagnostic feature of these techniques has always been seen as the attempt to control or predetermine the overall shape and size of the intended flakes by means of careful preparation of one face of the parent core. The classic definition of the Levallois concept is embodied in the widely quoted definition of Levallois flakes proposed by François Bordes: i.e. "un éclat à forme prédéterminée par une préparation spéciale du nucléus avant enlèvement de cet éclat" (Bordes 1961a: 14; 1980: 45). Later confusions have arisen mainly from the wide variety of forms of flakes and associated core types which can be accommodated within this definition. As Bordes himself was at pains to emphasize (e.g. 1950a, 1961a, 1980 etc.), Levallois techniques - defined by his criteria - could be used to produce a remarkable variety of different flake forms ranging from "classic" forms of broad, oval flakes showing distinctively converging "centripetal" patterns of core preparation (Figs 3.3, 3.4) through to more elongated, tapering forms which could be variously described as either "Levallois points" or "Levallois blades" depending on the particular morphological or metrical criteria employed (Figs 3.7, 3.8, 3.18). With this range of diversity, it is hardly surprising that the literature has been plagued by debates on the exact implications and definitions of Levallois flaking. (See for example, Boeda (1993b) for a general review of these debates and Copeland (1983) for a discussion of some of the confusions over the use of the term Levallois in the literature on Middle Eastern industries.)
Recently, several attempts have been made to gain new insight into the essential character of Levallois flaking techniques, partly to resolve some of these long-standing terminological confusions and partly to clarify the technological strategies which underlay these procedures (e.g. Bradley 1977; Shche-linskii 1974, 1983 (in Plisson 1988); Geneste 1985; Perpere 1989; Van Peer 1991, 1992; Delagnes 1990, 1992; Schlanger 1994). The most detailed studies which have been applied to European industries so far are those carried out by Eric Boeda in the course of his doctoral research at the University of Paris (Boeda 1982, 1984, 1986, 1988a,b,c, 1993a,b; Boeda et al. 1990; Beyries & Boeda 1983). Boeda has identified what he refers to as a basic "Levallois concept", which to him represents the unifying element behind all flaking techniques to which the term Levallois can properly be applied. This is defined by a basic division in the initial stages of preparation and shaping of Levallois cores into two main components: first, the preparation of a continuous striking platform extending around most of the perimeter of the selected nodule - normally produced by successive blows delivered more or less vertically on the upper face of the core and extending over a substantial part of the lower face; and second, by the systematic shaping of the upper surface of the core by blows delivered from various points around the perimeter of this prepared striking platform
(Boeda 1988a; Boeda et al 1990). As he points out, this defines effectively, in both an operational and a conceptual sense, two major components of the core form and at the same time effectively restricts all the subsequent primary flake removals to a prescribed and delimited area of the core surface (Figs 3.5, 3.9). Boeda emphasizes that this also defines the potential productivity of the core in a 'volumetric' sense by restricting the effective volume of flake production to the uppermost part of the core surface.
As Boeda points out, however, this still leaves scope for many different flaking strategies within the general heading of his Levallois concept and allows a surprising latitude in the form and character of the flakes produced by different techniques. Boëda (1988a, 1993b; Boëda et al 1990) divides these different strategies into two main groups, which he refers to respectively as 'lineal' and 'recurrent' Levallois techniques. Within each a number of more specific strategies of core preparation led to a variety of flake products. Briefly, the main features of these different strategies can be summarized as follows:
Figure 3.4 'Classic Levallois corefrom the site of La Borde (Lot). After Jaubert et al. 1990.
Was this article helpful?