Waiting for the snow to melt…
In an effort to assess the age of 5GN149, a lithic workshop in the Gunnison Basin, Colorado (Cooper, this volume), we investigated morphometric variation in the blades from this assemblage. We followed the analytical lead of Collins (1999), who, in his valuable study of Clovis blade technology, grappled with the question of how to recognize Clovis-age blades in the absence of independent age control or associated diagnostic artifacts. Collins’s solution was to plot on triangular graphs the ratios of length, width, and thickness to the sum of all three measures in blades of unknown affinity—notably, the Keven Davis cache—against comparable data from blades of known Clovis and non-Clovis affinity (Collins 1999:Figures 8.6– 8.10). Although Collins was careful to incorporate additional observations, it was primarily from his visual inspection of the scatter of points on the graphs that he concluded the Keven Davis specimens “fit well within the range of known Clovis blades” and were thus of Clovis age (Collins 1999:172). This approach is more intuitively satisfying than statistically informative, for it requires the analyst to assess visually how well defined a scatter of points from an assemblage might be, and the degree to which the scatter from an unknown sample overlaps the scatter of any known groups. Such an approach cannot give a meaningful measure of the distances between scatters; it also cannot help to determine whether such distances are statistically significant, or identify the variables most effective in sorting blades of different ages.
Items connected by shared entities, co-authorship, citations, or semantic similarity.