Abstract

In this chapter we present a study of flake retouching on one level of the Combe Grenal, located in the Black Perigord of France. We use the results to reflect on existing explanations of Middle Paleolithic tool production and diversity. Our evidence indicates the nonstaged and multilinear character of implement production and the apparent importance of blank form in influencing the pattern of retouch distribution and intensity. This inference implies that models of the implement classes, as stages of reduction, are not a viable depiction of the retouching technology represented in Layer 21. Instead, our reconstruction of scraper retouching demonstrated that each of Bordes's implement types has multiple histories of retouching. Some implements received little retouch, whereas others were intensively retouched; retouch sometimes changed a specimen to such an extent that the type into which it was classified was altered, whereas other specimens remained typologically stable even though they received additional retouch. The possibility that different specimens belonging to a type had different histories is a reason that typological groups may make poor analytical units for many technological questions.

INTRODUCTION

Questions of artifact reduction have been central to a number of high-level debates in Paleolithic archaeology. In particular, there have been extensive discussions about whether the traditional practice of analyzing retouched flakes by classifying them into a number of normative categories, called implement types, is valid or problematic, and whether those types represent tools of distinctly different designs or arbitrary divisions in sets of morphologically variable objects.