Search

Two paths of investigation are open to researchers interested in the growth, development and life history of early African hominids. One approach relies on the ‘hard evidence’ of the fossil record and seeks to make inferences about how hominids grew and matured based on their fossil teeth and bones. The other approach is based on the ‘soft story’ of statistics and modelling and attempts to apply what is known about the ontogeny of extant primates to extinct hominids. The second approach, statistical modelling, has become increasingly important in efforts to elucidate life history. However, when current methods are scrutinised it emerges that there are a number of problems inherent in the methodology used to date. Recognising this problem, the thesis develops a new statistical method for modelling in palaeoanthropology. It for the first time advances multivariate techniques in the wider context of hominid life history. It also extends the theory of residualisation techniques in multivariate analyses as a way to circumvent some of the observed mathematical problems. The new method is then put to the test in two examples attempting to model molar eruption and longevity in Plio-Pleistocene hominids. It emerges as a result that the bonobo (Pan paniscus) is by and large the best model for the australopithecines, whereas early Homo spp. are intermediate between modern great apes and humans. This is yet more independent confirmation for the very recent evolution of the modern pattern of life history. Given that the modern human pattern is not of ancient origin, the question arises how, when and why history patterns changed over the coarse of 4 million years of hominid evolution from ape-like beginnings to the pattern observed today. This is discussed in the final chapter and various evolutionary-ecological scenarios are constructed to explain why modern humans are members of uniquely unique species: the most socially and intellectually complex, the longest-lived and most widespread primate ever.