Diagnostic flake analysis : a replication-based method for reconstructing reduction techniques, strategies, and technologies
Diagnostic flake analysis is based on the concept that distinct flintknapping techniques produce distinctive flakes. Seen in this light, the information potential of flakes is enormous. Unfortunately, this information is virtually lost in analyses based upon size-grading a flake assemblage or separating flakes statistically based on a few "key" attributes. The intent of this study is to provide and apply a well-integrated analytical approach based upon the diagnostic flake concept.In order to integrate the static lithic artifacts to the dynamic behavior that produced them, a generalized flow chart model of the knapping event is developed. The flow chart model emphasizes the debitage produced during knapping, rather than finished lithic tools. The flow chart model is described in detail, and the terms"technology", "strategy", and "techniques" are defined and contrasted.A total of 30 reduction experiments were conducted in the course of this study, producing an estimated 27,000 flakes and flake fragments. Based on this sample and previous work conducted by Flenniken (n. d. ) and others, nine diagnostic flake types and three significant flake attributes are defined.An ideal methodology for a lithic analysis is developed. This ideal methodology includes: 1) assessing the types of raw materials present on a site; 2) reconstructing the technology based on negative-flake scars on finished tools; 3) flake refitting; 4) classifying flakes into the diagnostic flake categories nonstatistically and polythetically, with special emphasis placed on recognizing previously unidentified diagnostic flake types; 5) developing a flow chart model of reconstructed prehistoric technology; and 6) summarizing the flow chart i n verbal form.The methodology is applied to the Middle and Late Woodland components of the stratified All Seasons site located in central Indiana. Analysis of the Middle Woodland assemblage results in the recognition and definition of conical core flake blanks.The methodology is applied to a blind test manufactured by Donald Cochran to assess bias that may have been introduced into the flake type definitions by using debitage produced only by the author. Cochran's behavior is accurately reconstructed.Finally, the results of the study are discussed, and the study's strengths and weaknesses are determined. Diagnostic flake analysis is found to be a powerful approach that derives an optimal amount of high-quality information from a chipped stone assemblage.