Swept under the rug: the problem of unacknowledged ambiguity in lithic residue identification

Microscopic analysis of organic residues on stone tools is used to interpret prehistoric stone tool functions. The morphology of some residues can be difficult to interpret, yet this ambiguity is rarely acknowledged in the literature. Our research seeks to understand the nature of this ambiguity by objectively identifying ambiguous residues in our reference collection. We trained four archaeologists in residue analysis using one part of our reference collection, then tested their ability to identify sixty-eight residues in another part of the same collection. Forty-eight of the residues in the test (70%) were correctly identified by three or all four subjects. We considered the remaining twenty residues, which were correctly identified by two or fewer of the subjects, to be ambiguous. These are most often in the hide-scraping, bone-scraping, and hardwood-scraping (macerated) categories, and tend to have an atypical morphology which falls in the range of variability of another residue category. Some of these residues also have optical properties which make them more difficult to image than others. We explore the potential for scanning electron microscopy (SEM) to improve residue identification in a second test. This test shows a modest improvement in identification success rates of ambiguous residues when SEM images are included. We conclude that while images from different types of microscopes can improve reliability of identification, some residues will always be ambiguous. Rather than being ignored, these ambiguities should be brought to light, closely examined, and published as such.