The Roles of Domesticated Animals and Broomcorn Millet in the Diet and Subsistence of Bronze Age Peoples of Kazakhstan Investigated through Lipid Residue Analysis of Archaeological Pottery

  • Mengyao Zhang

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

The Bronze Age in Kazakhstan is characterised by ‘nomadic empires’ consisting of
highly specialized mobile pastoralists appearing from at least the third millennium BC
onwards. By the late 3rd or early 2nd millennium BC, increased interactions through the Inner
Asian Mountain Corridor connected populations across Eurasia. The transfers of crops from
China and West Asia converged in the mountain areas of Kazakhstan. The development of
agriculture during this period has been studied through archaeobotany and zooarchaeology and
stable carbon isotopic analysis of human and animal bone collagens. However, due to the
limited availability of such remains, important archaeological questions, such as those relating
to the adoption of broomcorn millet and the specialisation of animal exploitation, remain
unanswered.
Lipids can be well-preserved in unglazed archaeological ceramic vessels, offering a
direct reflection of the original foods processed within vessels, and indirectly reflecting
agricultural production activities. Miliacin, a specific biomarker of broomcorn millet (Panicum
miliaceum), provides an proxy to identify the processing of this important cereal. To gain a
comprehensive understanding of miliacin behaviour during vessel use and burial and to aid in
the interpretations of ‘millet residues’ identified in pottery vessels, cooking and degradation
experiments were conducted using replica pots under laboratory conditions. The GC, GC-MS
and GC-C-IRMS analysis of the resulting absorbed lipid extracts revealed that the fat from coprocessing millet with animal products enhanced miliacin absorption into the pot matrix, yet
that miliacin was much more stable than fatty acids during burial. Burial conditions, i.e. oxic
versus anoxic, did not appear to significantly influence on miliacin preservation. The presence
of the C16:0 and C18:0 fatty acids along with miliacin in lipid extracts of archaeological cooking
vessels indicated the processing of animal products and millet in the same vessels. Furthermore,
during degradation, the δ
13C values of the C16:0 and C18:0 fatty acids in pots used to process
millet and animal products remained unchanged.
Theoretical mixing models were developed to evaluate the potential impacts of
mixing/processing of animal products (ruminant adipose or dairy or non-ruminant adipose
products) with varying proportions of broomcorn millet grains. While the interpretation of the
original of the animal fat has the potential to be affected by mixing with millet lipids, in
practical culinary scenarios, the δ
13C16:0 and δ
13C18:0 values obtained from fatty acids were
shown to plot within the range of the fatty acids from the animal products. This is because
millet contains significantly lower concentrations of lipid compared to animal products, such
that the isotopic contribution from millet grain fatty acids will be overshadowed by those of
the animal products. Consequently, for archaeological residues containing miliacin the δ
13C
values of the C16:0 and C18:0 fatty acids represent the animal fat origin. The results of these
experimental studies were used to interpret the results of organic residue analyses of ‘cooking’
vessels from Bronze Age Kazakhstan.
Over 650 potsherds from Bronze Age sites across central, south and east Kazakhstan,
mainly from the Late and Final Bronze Age, were analysed to investigate the extent of millet
agriculture and regional subsistence practices. Ruminant carcass and dairy fats comprised the
majority of animal fats identified from pottery vessels. The results are consistent with the faunal
assemblages, supporting the theory that Bronze Age pastoralists mainly relied on animal
carcass and their secondary products. The detection of miliacin confirmed the processing of
broomcorn millet within pottery vessels in the south and east regions, and thus local cultivation,
while there was no evidence of broomcorn millet cultivation and processing among the Bronze
Age sites of southern Kazakhstan, which is consistent with the archaeobotanical analysis and
stable isotopic studies to date.
Date of Award4 Feb 2025
Original languageEnglish
Awarding Institution
  • University of Bristol
SupervisorRichard P Evershed (Supervisor) & Lucy J E Cramp (Supervisor)

Keywords

  • Organic residue analysis
  • Bronze Age
  • Broomcorn millet
  • miliacin
  • Kazakhstan
  • lipids

Cite this

'