Academic journal article Southeastern Archaeology

The Problems with Visual Identification of Dover and Fort Payne Chert

Academic journal article Southeastern Archaeology

The Problems with Visual Identification of Dover and Fort Payne Chert

Article excerpt

The use of visual identification techniques to assign source or provenance to cultural materials manufactured from chert is common practice within archaeology despite studies that demonstrate high error identification rates (Boisvert and Williams 2014; Calogero 1991, 1992; Ferguson and Warren 1992 for a positive case; Hess 1996; Perry 1992; Price et al. 2012; Spielbauer 1984). High error identification rates are here defined as scores reported in blind tests having a 40-70 percent error rate. Furthermore, consistency between observers is difficult to achieve. The potential implication of visual identification methodology is that our explanations of human behavior, particularly in the realms of mobility and intergroup interactions, are potentially inaccurate. The purpose of the current study is not an argument to eliminate visual identification of chert type, but rather a cautionary tale that no matter what provenance method employed, whether visual or analytical testing, a researcher must standardize the process. The researcher must define the categories used for classification, characterize deposits at the desired level of spatial resolution (outcrop/formation), and match unknown artifacts within the range(s) of variation established. The Dover/Fort Payne case study presented here is not meant to provide a mechanism for differentiating these southeastern US chert types visually but rather to illustrate how commonly used macroscopic traits can have difficulty in clearly distinguishing material type even at the parent formation scale.

Replicability between observers, defining classes, and standardization are common issues confronting every classification system in science. In fact, the methodological pitfalls associated with the visual identification of chert are not a new revelation having been espoused by others (Andrefsky 2012; Butler 1984; Cooper 2012; Högberg and Olausson 2007; Luedtke 1976, 1978, 1979; Odell 1984; Parish 2013). However, the discussion of the reliability of this widely used methodology continues to be relevant (Andrefsky 2012). In this paper, we contextualize the problems of visual analysis by first placing the methodology in its historical roots. Next, we present qualitative data from samples obtained from a large regional survey of Dover and Fort Payne chert deposits as a case study. Finally, a basic outline is presented that may assist researchers who routinely conduct visual source identifications. Here, we humbly argue that our current methodology is flawed and a dynamic view of lithic resource availability is still needed, which incorporates a realization that macroscopic variability is potentially ubiquitous across a large geographic area. The inaccuracy of visual type identification in turn potentially impairs our ability to meaningfully use chert provenance data as a proxy for addressing even coarse-grained anthropologic questions.

Historical Context

The use of macroscopic traits in identifying lithic materials has its roots in late-nineteenth and early-twentieth century prehistoric quarry research (Smolla 1987:128). Both in the Old and New World, a focus on primary procurement sites for lithic materials was common in the archaeological literature (Babbit 1880; Blackman 1903, 1907; Dorsey 1900; Fowke 1892; Gilder 1908, 1909; Harrington 1925, 1926, 1929; Holmes 1919; Schild 1987; Wilson 1897). The research was driven by studies concerned with the origins for raw material used for particular classes of artifacts that were widely distributed in antiquity (Smolla 1987). The largest quarry sites were surveyed in great detail as researchers marveled at both the size and extent of prehistoric "industry" evidenced by the pits and debitage piles. The quarry surveys and observations made of artifacts in various sequenced forms at quarries helped establish the American "reduction sequence" by observing stone tool production at all stages of manufacture.

The identification and documentation of the most impressive procurement sites also established them as the primary sources for artifacts manufactured from look-alike material. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.