Reconstruction
At Nanzhuangtou, on the piedmont east of the Taihang Mountains in modern Hebei, people occupied a lowland landscape during the transition from the last ice age to the Holocene. Excavated deposits contain pottery sherds, grinding slabs and handstones, hearths, pits, ditches, bone and antler implements, worked wood, and animal remains. These finds do not describe an agricultural village. Instead, they reveal foragers combining older hunting and gathering practices with technologies that made collected foods easier to process, cook, and perhaps store.
The site’s chronology requires care. Early publications reported uncalibrated radiocarbon ages of roughly 10,500–9,700 BP, while broader syntheses have sometimes assigned a correspondingly long span. Researchers later excluded a potentially misleading organic-silt determination because of possible old-carbon effects. Five other samples place the principal occupation at approximately 11,500–11,000 calibrated years BP, or about 9550–9050 BCE. This shorter range is currently the more defensible estimate, although discussions of the site still employ differing chronologies.
Nanzhuangtou’s plant evidence includes water caltrop, wild grape, pondweed, unidentified wood and leaves, and microscopic starch residues. Starches recovered from cultural deposits and ground-stone tools include Paniceae grasses, the group containing wild and domesticated millets, as well as Triticeae grasses and underground plant organs. The evidence demonstrates collection and processing of starchy plants. It does not establish a fixed recipe, prove that the grasses were fully domesticated, or show that millet was already a staple. Grinding could have reduced small seeds or other plant tissues into meal suitable for mixing with water, but the final food remains a reconstruction.
Pottery added another possible stage to this processing sequence. Some Nanzhuangtou sherds bear smoke blackening consistent with exposure to fire, supporting—but not conclusively proving—the use of vessels in cooking. No securely identified Nanzhuangtou dish survives. Evidence from nearby early Holocene sites, especially Zhuannian and Donghulin, directly associates comparable pottery and tools with cooked or ground millet, acorns, and other plants. Those parallels make simmered plant foods at Nanzhuangtou reasonable to consider, but they cannot substitute for direct residue evidence from its vessels.
Animal bones show that plant processing belonged to a broader subsistence system that included deer, wild boar, aquatic resources, and other fauna; some bones carry burning or cut marks. Dogs were present, but claims concerning domesticated pigs or chickens remain less secure. Nanzhuangtou is significant because pottery and substantial grinding equipment appeared before unequivocal dependence on crop farming. Its inhabitants illustrate how repeated harvesting, grinding, heating, and sharing of wild resources could reorganize food labor before agriculture.
The site is therefore not the documented birthplace of millet cuisine. Its defensible modern connection is technological: grinding grain and cooking it with water remain basic operations behind millet meal, porridge, and other cereal foods, even though no continuous recipe tradition can be traced from Nanzhuangtou to kitchens today.
Historical context
The Younger Dryas cold phase was ending and the Holocene began around 9700 BCE. Northern China experienced shifting monsoon strength, vegetation, wetlands, and animal habitats rather than a single uniform environmental change. There were no cities, states, or fully agricultural landscapes in the region; communities were comparatively small and depended substantially on hunting, fishing, and gathered plants. Pottery was already known elsewhere in East Asia, including much earlier examples in southern China and Japan, so Nanzhuangtou represents a northern adoption or local development of an existing technological tradition rather than the invention of ceramics itself. More formally organized farming villages appeared in northern China only later.
Evidence
Archaeological
StrongExcavations directly documented pottery sherds, grinding slabs and handstones, pits, hearths, ditches, bone and antler artifacts, worked wood, and faunal remains.
Archaeobotanical
ModerateReported macroremains include water caltrop, wild grape, pondweed, wood, and leaves, but systematic recovery of plant macroremains was limited.
Residue chemical
ModerateMicroscopic starch analysis identified Paniceae, Triticeae, and underground plant sources in cultural deposits and on grinding tools. Starch identification supports plant processing but does not by itself establish domestication or a specific recipe.
Food identification
ModerateWild or managed millet-related grasses and other starch-rich plants were processed. Distinguishing wild from domesticated foxtail and broomcorn millet using starch morphology alone remains difficult.
Dating
StrongFive accepted radiocarbon samples cluster at approximately 11,500–11,000 calibrated BP. A broader chronology partly depends on determinations, including organic silt, that researchers have treated cautiously because of possible old-carbon effects.
Preparation method
ProbableGrinding is directly supported by tools and starch residues. Smoke-blackened pottery is consistent with heating, but a particular Nanzhuangtou plant food cooked in a vessel has not been directly identified.
Geographic attribution
StrongThe excavated site is securely located near Nanzhuangtou Village in Xushui, Hebei, at the western edge of the North China Plain.
Historical interpretation
InterpretiveCharacterizing the assemblage as intensified or broad-spectrum foraging is a scholarly interpretation based on the combined plant, animal, tool, and feature evidence, not a directly observed behavior.
Visual reconstruction
LimitedThe tools and sherds can be reconstructed from excavated objects, but vessel forms, prepared foods, clothing, structures, and activity scenes require substantial inference.
Sources
- 1.Xiaoyan Yang, Zhiwei Wan, Linda Perry, et al. (2012). Early millet use in northern China. Proceedings of the National Academy of Sciences 109(10): 3726–3730. doi:10.1073/pnas.1115430109Scientific literature
- 2.Xiaoyan Yang, Wenxiang Wu, Linda Perry, et al. (2018). Critical role of climate change in plant selection and millet domestication in North China. Scientific Reports 8: 7855. doi:10.1038/s41598-018-26218-6Scientific literature
- 3.Zhikun Ma, Chi Zhang, Quan Li, et al. (2017). Understanding the Possible Contamination of Ancient Starch Residues by Adjacent Sediments and Modern Plants in Northern China. Sustainability 9(5): 752. doi:10.3390/su9050752Scientific literature
- 4.David J. Cohen (2011). The Beginnings of Agriculture in China: A Multiregional View. Current Anthropology 52(S4): S273–S293. doi:10.1086/659965Modern synthesis
- 5.Gideon Shelach-Lavi and Dongdong Tu (2017). Food, pots and socio-economic transformation: The beginning and intensification of pottery production in North China. Archaeological Research in Asia 12: 1–10. doi:10.1016/j.ara.2017.10.001Modern synthesis
Limitations
- moderatedate start/date end
The supplied 10,500–8,500 BCE span reflects a broad and disputed chronology. A focused radiocarbon reassessment excluded a potentially old-carbon organic-silt result and placed the principal occupation at approximately 11,500–11,000 cal BP, about 9550–9050 BCE.
- minordescription
Plant use and grinding are directly supported, but 'intensified foraging' is a scholarly interpretation rather than a directly measured condition. The evidence also does not demonstrate fully domesticated millet or crop-dependent farming.
