A re‐evaluation of Kulubnarti: A synthesis of the osteoarchaeological evidence

Abstract

AbstractThe medieval Nubian cemeteries at Kulubnarti (R and S) have greatly impacted bioarchaeological knowledge. Recent dating of the cemeteries showed they are contemporaries, contrary to prior conclusions that the R cemetery was later in date. Four lines of evidence used to argue for socio‐economic status (SES) differences between the cemeteries in the literature are evaluated: (1 and 2) architecture of houses/churches, (3) age‐at‐death distributions and pathology prevalence, and (4) distribution/frequency of textile types. It is hypothesized that the S cemetery is probably a special use cemetery for subadults and/or there was a monastic presence at the site. First, the original monographs and articles were synthesized to describe the characteristics of Kulubnarti. Second, to verify that genetic differences from social isolation due SES disparities were not operating at Kulubnarti, cranial metrics and nonmetrics from the Kulubnarti cemeteries and other samples from medieval Lower Nubia (Semna South, el‐Hesa) were separately input into Mahalanobis D2 variations, principal coordinates analyses and Procrustes analyses to produce and plot biodistances. Space and time differences were also tested via Mantel tests. Third and fourth, age‐at‐death distributions from the Kulubnarti cemeteries were modelled with a binomial logistic regression, as were textile‐type distributions/frequencies. The literature demonstrates that the cemeteries and the settlements/churches attributed to them did not overlap in time. The biodistance plots show that there are only small differences between the Kulubnarti cemeteries, which are not suggestive of genetic differences due to social isolation from SES disparities. The age‐at‐death logistic regression indicates a more than sixfold increase in the odds of infants being interred in the S cemetery, as compared with adult females, and a nearly threefold increase of other subadults. Further, there were no significant differences in textile‐type distributions/frequencies between the cemeteries. Alternative models explaining who was buried in the S cemetery are better supported than differences in SES.