When archaeological skeletons are assessed, the prevalence (and patterns of bone involvement) of trauma is important. The number and pattern of fractures can be used to gain insight into the occurrence of interpersonal violence, workload and living conditions. However, the question remains as to how these results should be interpreted—such as what constitutes high or low levels of trauma? The aim of this study was to investigate the occurrence of trauma in a population of modern Greeks living in Crete, as well as South African (SA) whites and blacks in the Pretoria Bone and Raymond Dart collections. The sample comprised mostly of older individuals (n = 90–100 within a sex-population group). Each skeleton was studied for healed trauma. For the vertebrae, only spondylolysis was assessed. In the Greek sample, it was found that 42% of the males and 46% of females had at least one fracture, with corresponding figures of 63 and 44% for SA whites and 83 and 69% for SA blacks. Radius, rib and femur fractures were most common in Greeks, with skull, radius and ribs most common in SA whites and skull, ulna and ribs in SA blacks. These prevalences of trauma are high, but the composition of the samples (mostly of lower socio-economic origin) should be kept in mind. It may also be questioned whether these individuals reflect the society as a whole. It seems that the fractures in Greeks are mostly related to old age due to falls and accidents (radius and hip fractures), while the SA black sample reflects high prevalences of interpersonal violence (such as cranial vault and ulna fractures). The SA white sample follows a comparatively moderate pattern of trauma. These comparative figures may be useful when assessing trauma in other skeletal populations. Copyright © 2009 John Wiley & Sons, Ltd.