Sample-specific sex estimation in archaeological contexts with commingled human remains: a case study from the Middle Neolithic cave of Bom Santo in Portugal

Estimating sex on large assemblages of commingled skeletal human remains is challenging because it prevents the systemic observation of the skeleton and thus reduces the reliability of sex-ratio estimation. In order to tackle this problem, the applicability of sample-specific odontometric methods was assessed on the human skeletal remains from the Middle Neolithic cave necropolis of Bom Santo in Portugal. We present an approach that confirms some of the assumptions – the normal distribution of the data and the 1:1.5 sex ratio – indicated by Albanese et al. (2005) for the application of sample-specific methods. These assumptions are often difficult to assess in archaeological samples and thus prevent the use of sample-specific methods. The mean bucco-lingual diameter of 51 lower right canine teeth was used as a cut-off point to discriminate between sexes within a sample from Bom Santo. Before that, Shapiro–Wilk statistics was used to confirm that the distribution of the data in a sample of 51 lower canine teeth was normal. In addition, the range and central tendency of the data were compared to other samples for which the sex of the individuals was known in order to confirm that those parameters were consistent with those of a sample presenting a balanced sex ratio. The canine sex estimations were then compared with the sex estimation obtained from mandibles where canine teeth were still in situ (n = 8). No clear disagreement between the two methods was found thus demonstrating good potential of this method for sex estimation and for the sex ratio estimation in commingled human skeletal remains. Results indicated that sex ratios in Room A and Room B at Bom Santo were quite different. This indicates that the two locations may have been used in a different way according to sex