Unpacking the Discard Equation: Simulating the Accumulation of Artifacts in the Archaeological Record

Quantifying discard to accurately estimate the duration of site occupation is critical middle-range research necessary for understanding assemblage diversity, the nature of settlement systems and mobility strategies, and population size, and for testing any anthropological theory that depends on the accurate measurement of these variables. We address this middle-range research by employing a computer simulation to explore assumptions inherent in the discard equation and to determine the accuracy with which cooking pot refuse measures the length of site occupation. The accumulation of discarded cooking pot sherds is simulated using a strong archaeological case: the Duckfoot site, a Pueblo I residential site located in the Mesa Verde region of southwestern Colorado. We argue that estimating the length of site occupation using data from a strong archaeological case is superior to using the discard equation and ethnographic data, but that the discard equation and ethnographic data-used judiciously-can provide reasonable estimates if a strong archaeological case is not available. Results indicate that the most variable and least accurate results are generated by short-term occupations of sites by small numbers of households. We further conclude that quantifying the accumulation of discarded cooking pot sherds has considerable promise as a means of estimating the length of site occupation.