A new method for identifying sherd refits: A case study from the Neolithic of Northumbria, UK

Here we present a flexible, quantitative methodology for refitting handmade ceramics. Rather than analyzing the physical joins between sherds, we assess the likelihood that non-adjoining sherds of similar appearance are from the same vessel. The customary disregard of non-adjoining sherds in refitting experiments prompts us to address this issue. Using the Neolithic ceramic assemblage from Thirlings (Northumberland, U.K.) as a case study, we test several scoring systems for assessing the probability of two sherds pertaining to the same vessel. The utility of this approach is supported by the comparison of a set of percentage scores statistically verified using Spearman's rank correlation coefficient. We conclude with a discussion of the contribution of the new method to post-excavation ceramic analyses, and its applicability to other assemblages.