Social Complexification and Pig (Sus scrofa) Husbandry in Ancient China: A Combined Geometric Morphometric and Isotopic Approach

Abstract

Pigs have played a major role in the economic, social and symbolic systems of China since the Early Neolithic more than 8,000 years ago. However, the interaction between the history of pig domestication and transformations in Chinese society since then, have not been fully explored. In this paper, we investigated the co-evolution from the earliest farming communities through to the new political and economic models of state-like societies, up to the Chinese Empire, using 5,000 years of archaeological records from the Xiawanggang (XWG) and Xinzhai (XZ) sites (Henan Province). To trace the changes of pig populations against husbandry practices, we combined the geometric morphometric analysis of dental traits with a study of the stable carbon and nitrogen isotope ratios from bone collagen. The domestication process intensified during the Neolithic Yangshao, prompted by greater selective pressure and/or better herd control against wild introgression. After that, pig farming, in XWG, relied on local livestock and a gradual change of husbandry practices overtime. This was characterized by a gentle increase in millet foddering and animal protein intake, until a complete change over to household management during the Han dynasty. The only rupture in this steady trend of husbandry occurred during the Longshan period, with the appearance of small sized and idiosyncratic pigs with specific feeding practices (relying on millet and household scraps). From three exploratory hypothesis, we explored the possibility of anti-elite pig production in XWG during the Longshan period, as a means to resist incorporation into a new economic model promoting intensified domestic production. This exploratory hypothesis is the most suitable to our dataset; however, numerous areas need to be explored further in order to adequately document the role of pigs in the rise of China's complex societies.

Fig 1. Location of the archaeological sites studied within their chrono-cultural context.

Fig 2. Notched boxplot of the M₂ Log transformed centroid size for all samples.

The box represents 50% of the variation around the median (horizontal black line); the brackets represent the minimum and maximum. Notches represent a 95% interval confidence of the median.

Fig 3. Molar shape differences between the two extent wild boar sub-species and the domestic pigs of China.

First canonical variates (CV) computed on size corrected shape variables. The molar shape divergence between the wild and domestic type along the CV1 is displayed by shape reconstruction on each axes extremity; the divergence between the two wild boar sub-species is displayed along the CV2. Confidence ellipses contain 90% of the data points with a 0.9 probability.

Fig 4. Dental shape relationships among modern and archaeological samples.

Phenogram showing the M₂ shape relationships between the geographic and chronological mean shape of modern and archaeological samples.

Fig 5. M₂ shape differentiation among samples from XZ and XWG.

First two axes of the CVA computed on size corrected shape variables. Confidence ellipses contain 90% of the data points with a 0.9 probability.

Fig 6. Scatter plot of the δ¹³C and δ¹⁵N values of bone collagen from the XWG site displayed per phase.

Fig 7. Relationship between ¹³C and ¹⁵N isotopic values and the M₃ length measurements across XWG phases.

Chronological phases of XWG are depicted by six different symbols. The Longshan specimens of phase 3 are clustered according to their M₃ length, either below or above the 37 mm threshold.