Archaezooological perspectives on the Chinese Early and Late Paleolithic from the Ma’anshan site (Guizhou, South China)

Ma’anshan is a Paleolithic cave site in the Guizhou province of southern China. The total area of the excavations is ca. 48 m2, and the cultural deposit is divided into two layers representing the Chinese Late Paleolithic and the later part of the Early Paleolithic. The upper layer dates between 19,295 BP and 31,155 BP by the 14C AMS technique, and the lower layer is dated to around 53 ka BP by the U-Series technique. Thousands of bone fragments were recovered from the two layers, of which 4,358 pieces could be identified to macro-mammal taxon or mammal body size class and skeletal element. Bone surface modifications, species abundance, skeletal element representation, and mortality patterns were used to investigate patterns of prey selection and meat consumption by the hominins, including food transport and butchering practices. Interpretations of these data are strengthened by reference to experimental studies and ethnoarchaeological data. The archaeological data indicate that the later occupants of Ma’anshan Cave hunted mainly medium and small game, while the earlier occupants tended to prey upon larger animals. In the later period, hominins also made fuller use of the carcasses. The breadth of the meat diet also increases with time, due to the inclusion of small animals (bamboo rats and birds) in the diet of the later period. The differences between the two faunal assemblages are consistent with the chronologic boundary currently drawn between the Chinese Early and Late Paleolithic ca. 30 ka BP-27 ka BP based on technological and other evidence. Together with the evidence of chronological dating, palaeoenvironment, polished bone tools, the micromammal and bird bones from the two cultural layers, this study proposes three hypotheses (climatic, cultural and demographic) that could explain these differences. While none of the hypotheses can be refuted completely by the available evidence, a combination of climate-driven and demographic (predator pressure) effects is most likely.