Throwing in the Middle and Upper Paleolithic: inferences from an analysis of humeral retroversion.

When in evolutionary history did long-range projectile weapons become an important component of hunting toolkits? The archeological evidence for the development of projectile weaponry is complex and generally indirect, and has led to different conclusions about the origin and spread of this technology. Lithic evidence from the Middle Stone Age (MSA) has led some researchers to suggest that true long- range projectile weaponry developed in Africa perhaps as early as 80,000 years ago, and was part of the subsistence toolkit carried by modern humans who expanded out of Africa after 50,000 years ago. Alternatively, temporal patterns in the morphology of pointed lithics has led others to posit an independent, convergent origin of projectile weaponry in Africa, the Near East, and Europe during the interval between 50,000-40,000 years ago. By either scenario, projectile weapons would not have been a component of the hunting arsenal of Neandertals, but may have been in use by European early modern humans and thus, projectile technology may have entered into the competitive dynamics that existed between these two groups. The origins of projectile weapons can be addressed, in part, through analyses of the skeletal remains of the prehistoric humans who made and used them. Habitual behavior patterns--including those related to the production and use of technology--can be imprinted on the skeleton through both genetic and epigenetic pathways. Recent studies in the field of sports medicine indicate that individuals who engage in habitual throwing have increased humeral retroversion angles in their throwing arms and a greater degree of bilateral asymmetry in retroversion angles than do non-throwers. This contribution investigates humeral torsion through analysis of the retroversion angle in samples of Eurasian Neandertals, European early modern humans of the middle and late Upper Paleolithic, and comparative samples of recent humans. This analysis was conducted under the assumption that if throwing-based projectile weaponry was used by early modern Europeans but not Neandertals, Upper Paleolithic samples should be similar to recent human groups engaged in habitual throwing in the degree of humeral retroversion in the dominant limb and in bilateral asymmetry in this feature. Neandertals on the other hand, would not be expected to show marked asymmetry in humeral retroversion. Consistent with other studies, Neandertals exhibit increased retroversion angles (decreased humeral torsion or a more posteriorly oriented humeral head) relative to most modern human samples, although this appears more likely related to body form and overall activity levels than to habitual throwing. Although Neandertals with bilaterally preserved humeri sufficient for measurement are rare (consisting of only two males and one female), levels of bilateral asymmetry in humeral retroversion are low, suggesting a lack of regular throwing. While patterning across fossil and comparative samples in levels of humeral retroversion was not clear cut, males of both the middle and late Upper Paleolithic demonstrate a high level of bilateral asymmetry, comparable to or in excess of that seen in samples of throwing athletes. This may indicate habitual use of throwing-based projectile weaponry by middle Upper Paleolithic times. Small sample sizes and relatively great variance in the fossil samples makes these results, however, suggestive rather than conclusive.

Adaptations to humeral torsion in medieval Britain.

Adaptations to the humeral torsion angle have been identified in the professional throwing athlete. This movement pattern increases the humeral torsion angle, and also increases the extent of external rotation movements in the dominant, throwing limb when compared with the nondominant limb. The purpose of this paper is to test the hypothesis that the humeral torsion angle is an adaptation to upper limb use. This project examines the humeral torsion angle in a number of medieval British populations, as well as a modern cadaver-based sample. The results identify significant differences in the humeral torsion angle both between and within male (P < 0.001, ANOVA) and female (P < 0.014, ANOVA) populations, although the results are not consistent with expected behavior patterns. Statistically significant differences between males and females within the same site were identified in 2 of the 5 samples examined. The mean level of bilateral asymmetry does not approach that reported for the professional throwing athlete. However, a number of individuals have high levels of asymmetry in excess of that identified in the professional throwing athlete. This analysis demonstrates the need for individual rather than population-based analyses, as the heterogeneity within population samples obscures individual variation in activity patterns. The diversity within British medieval society and a lack of specific known behaviors prevent further identification of the functional significance of the humeral torsion angle within the archaeological record examined here.

Activity-related skeletal change in medieval humeri: cross-sectional and architectural alterations.

This paper examines humeral cross-sectional properties in two different samples of later medieval date: a group of blade-injured males from the sites of Towton, North Yorkshire, and Fishergate in the City of York, England, and a comparative group of nonblade-injured males also from the site of Fishergate in York. CT image slices were taken of the humeral shaft at 20%, 35%, 50%, 65%, and 80% from the distal end to investigate population differences in levels and patterns of mechanical loading. Bilateral asymmetry is investigated and comparisons are made with different populations of varying activity levels. Architectural changes such as humeral torsion are also investigated to determine the relationship between architectural changes and biomechanical efficiency. Results show significant differences in diaphyseal robusticity between the Towton sample and the comparative population, as well as significant differences in diaphyseal shape both between limbs within the Towton sample and between blade-injured samples. Population differences were also identified in the level of bilateral asymmetry, further demonstrating the differences in movement and activity patterns both between and within samples. These variations may relate to distinctive, more strenuous weapon use and differences in strenuous movement patterns in the two groups.