reference:
Yoon, Soon-bong. 2025. The Millet Matrix: An Integrative Analysis of Genetic, Archaeological, and Linguistic Evidence for the Demic Diffusion of Neolithic Agriculture in North China. Study of YoonSoonBong.

https://yoonsb.com/2025/08/05/202589/

The Millet Matrix: An Integrative Analysis of Genetic, Archaeological, and Linguistic Evidence for the Demic Diffusion of Neolithic Agriculture in North China

  

Abstract 

This study demonstrates through multidisciplinary evidence from archaeology, genetics, and linguistics that the spread of millet agriculture in Neolithic China was a process of demic diffusion involving human migration, rather than a mere transmission of technology. Ancient DNA analysis, in particular, has clearly confirmed that significant, bidirectional population movements and gene flow occurred between the two major agricultural centers of North China—the West Liao River basin and the Yellow River basin—from the Late Neolithic to the Bronze Age.

This population movement was not a unidirectional expansion from a single center but a complex, reciprocal process of interaction between the two farming populations. This process fused the previously distinct northern and southern genetic ancestries, forming a new meta-population in North China. This admixed population became the demographic foundation for later Bronze Age civilizations such as the Shang and Zhou dynasties and established the genetic substratum of the modern northern Han Chinese. In conclusion, the spread of millet agriculture was a decisive event that fundamentally reshaped the genetic and cultural landscape of North China through human migration and admixture.

keywords: Millet Agriculture, Neolithic China, Demic Diffusion,  Ancient DNA (aDNA), West Liao River, Hongshan Culture.

 

Part 1: The Dawn of Millet – Establishing the Cradle of Agriculture 

The origins of agriculture in mainland China provide a fundamental context for understanding subsequent population movements and cultural developments. In particular, Neolithic millet agriculture in North China played a decisive role in shaping the demographic, genetic, and linguistic landscape of prehistoric East Asia. Elucidating the origins and early development patterns of this agriculture is an essential prerequisite for discussing how and by whom it was spread. This section defines the geographical and chronological parameters of millet agriculture based on archaeobotanical and archaeological evidence, exploring its origins by focusing on two key centers: the West Liao River basin and the Yellow River basin.

 

1.1 The Dualistic Structure of Chinese Agriculture 

The origin of agriculture in China is characterized by a globally unique dualistic structure. This means that two independent centers of crop domestication, adapted to different ecological environments, developed almost contemporaneously. In the north, centered on the Yellow River basin, a dryland farming system developed based on foxtail millet (Setaria italica) and broomcorn millet (Panicum miliaceum). In the south, centered on the middle and lower Yangtze River basin, a wetland farming system emerged based on rice (Oryza sativa) (Matuzevičiūtė and Liu 2021). According to archaeobotanical evidence, both agricultural systems were established by 8000 BCE at the latest (Matuzevičiūtė and Liu 2021).

These two systems developed largely in isolation for millennia, with significant interaction and integration accelerating only in the Bronze Age (Matuzevičiūtė and Liu 2021). In intermediate zones like the Huai River basin, mixed farming zones where both crops were cultivated together appeared from around 6,000 years ago (He et al. 2017). The boundaries of these agricultural systems were not fixed but shifted fluidly in response to climatic changes such as the East Asian summer monsoon (He et al. 2017).

This fundamental north-south agricultural divide profoundly influenced the population structure of prehistoric East Asia. The establishment of highly productive agricultural systems in different ecological environments (millet vs. rice) supported independent large-scale population growth in North and South China, respectively. This demographic separation, sustained over thousands of years, was very likely the main driver in forming the deep north-south genetic cline observed in East Asian populations through ancient DNA analysis (Bennett, Liu and Fu 2025; Wang et al. 2021). In other words, the difference in agricultural systems was not merely a difference in food production methods but a fundamental cause that shaped the macro-genetic structure of the East Asian population.

 

1.2 The West Liao River Basin – A Center of Early Cultivation 

The West Liao River basin in northeastern China is considered one of the important cradles of North China’s dryland agriculture (Ma et al. 2016). The Xinglongwa culture (c. 8000–7000 BP) of this region is a key archaeological culture that provides the earliest evidence of millet agriculture. Analysis of starch grains from grinding stones and pottery at the Xinglonggou site has clearly confirmed the presence of domesticated foxtail and broomcorn millet dating back to about 8,000 years ago (Ma et al. 2016).

Interestingly, analysis of microbotanical remains from the Xinglongwa and Zhaobaogou culture periods (c. 8000–6400 BP) shows that alongside millet remains exhibiting morphological traits of domestication, such as larger grain size, abundant wild plant resources like acorns, tubers, and legumes were also found (Liu et al. 2015). This suggests that the transition to agriculture was not a rapid replacement of the existing diverse and stable hunter-gatherer economy but a long-term process in which farming was gradually integrated into the existing economic system.

The emergence of this early agriculture is closely related to climate change. The period when millet domestication began in the West Liao River basin coincides with the Holocene Climatic Optimum, which featured a warmer and more humid climate than today (Jia et al. 2016a). These favorable climatic conditions increased the natural abundance of C4 plants, including the wild ancestors of millet, providing hunter-gatherers with a stable food source and creating a favorable ecological opportunity to experiment with domestication (Wang et al. 2021).

However, the relationship between climate and human society was not deterministic. When the climate deteriorated towards colder and drier conditions around 1500 BCE, the Lower Xiajiadian culture, which relied on intensive millet farming, declined. It was replaced by the Upper Xiajiadian culture, which had a more pastoral-oriented economy, as evidenced by an increased proportion of sheep and goats and a rise in artifacts associated with a mobile lifestyle (Jia et al. 2016b). This shows that while climate was a catalyst that enabled the start of agriculture, climate deterioration also acted as a pressure that forced strategic shifts in existing subsistence methods. In other words, human societies did not passively react to environmental changes but actively adapted their strategies in a dynamic relationship.

 

1.3 The Central Plains – The Rise of the Yellow River Agricultural Core 

Almost contemporaneously with the West Liao River basin, the Central Plains region, encompassing the middle and lower Yellow River basin, also developed as another core center of millet agriculture. Archaeobotanical evidence from various sites such as Cishan and Dadiwan pushes the history of millet cultivation in this region back to before 10,000 cal BP (Lu et al. 2009; Barton et al. 2009; Wang et al. 2019).

Notably, during the Peiligang-Cishan culture period around 8,000 years ago, a “trinity” dryland agricultural package combining foxtail millet, broomcorn millet, and domesticated pigs had already formed in a rudimentary state (Zhao and Fan 2016). This combination became the foundation of traditional North Chinese agriculture. By the Yangshao culture period (7,000–5,000 BP), this dryland farming system developed rapidly to become the dominant economic base throughout the Yellow River basin, signifying the completion of the “agricultural revolution” in this region (Zhao and Fan 2016). Stable isotope analysis of human bones from the Yangshao period Lingkou site in Henan Province clearly shows their diet was heavily reliant on the C4 plant millet, corroborating the archaeological evidence (Wang et al. 2025a).

Thus, in Neolithic North China, two core centers of millet agriculture—the West Liao River basin and the Yellow River basin—developed independently but almost simultaneously. Over the subsequent millennia, these two regions engaged in complex interactions, influencing each other and functioning as the two main axes that shaped the demographic and cultural landscape of all of North China.

Table 1: Major Neolithic and Bronze Age Cultures in North China

Culture	Main Region	Dates (BP)	Key Features	Main Subsistence Strategy
Xinglongwa	West Liao River Basin	8000–7000	First jades, incised pottery, early settled villages	Early millet cultivation + hunting-gathering
Hongshan	West Liao River Basin	6500–5000	Elaborate jades (dragons, birds), altars, temples, tombs; emergence of complex society	Intensified millet farming + hunting-gathering
Yangshao	Middle Yellow River Basin	7000–5000	Painted pottery, large villages, established agricultural economy	Intensive millet farming + pig husbandry
Longshan	Middle & Lower Yellow River Basin	5000–4000	Black pottery, walled towns, increased social stratification, inter-regional exchange	Intensive millet farming, crop diversification
Lower Xiajiadian	West Liao River Basin	4000–3500	Bronze metallurgy, intensive agriculture, dense population, defensive structures	Intensive millet farming + pig husbandry
Upper Xiajiadian	West Liao River Basin	3000–2600	Bronze daggers, increased pastoral-related artifacts, increased mobility	Agro-pastoral economy (increased pastoralism)

 

Part 2: The Grand Hypothesis and Its Dissenters – The Transeurasian Debate 

The question of how millet agriculture, originating in the West Liao River basin, influenced the linguistic and population distribution across East Asia and the Eurasian continent, beyond being a mere food production technology, is one of the hottest topics in recent academia. In particular, the ‘Farming/Language Dispersal’ hypothesis, published in Nature by Martine Robbeets et al. in 2021, caused a major stir by synthesizing evidence from linguistics, archaeology, and genetics to link the origin of the Transeurasian languages to the early millet farmers of the West Liao River basin. However, this grand hypothesis immediately faced strong criticism from scholars in various fields. This section delves into the core of this debate, illuminating the conflicting perspectives on the true nature of the West Liao farmers’ expansion.

 

2.1 In-depth Analysis of the ‘Farming/Language Dispersal’ Hypothesis (Robbeets et al. 2021) 

The core of the hypothesis proposed by Robbeets et al. is that the vast Transeurasian language family—including Japonic, Koreanic, Tungusic, Mongolic, and Turkic languages—originated from a single common ancestor, and its spread was driven by the migration of Neolithic farmers from the West Liao River basin, i.e., demic diffusion (Robbeets et al. 2021). They supported this claim by ‘triangulating’ evidence from three fields.

1. Linguistic Evidence: Through a phylogenetic analysis of 98 Transeurasian languages, they estimated that the common ancestor, Proto-Transeurasian, was spoken around 9,181 years ago in the early millet farming communities of the West Liao River basin (Robbeets et al. 2021). Furthermore, they argued that the reconstructed proto-language contained core vocabulary related to agriculture (e.g., ‘field’, ‘seed’, ‘millet’, ‘to cultivate’), suggesting that the early speakers were farmers (Robbeets et al. 2021).
2. Archaeological Evidence: They claimed that the diffusion path and timing of millet agriculture originating from the Xinglongwa culture in the West Liao River basin showed a clear correlation with the branching patterns of the language family tree (Robbeets et al. 2021). For example, the archaeological pattern of farming cultures spreading eastward from the West Liao River basin to the Liaodong Peninsula and the Korean Peninsula was said to match the timing and route of the divergence of the Tungusic and Koreanic-Japonic branches (Robbeets et al. 2021).
3. Genetic Evidence: Through the analysis of ancient DNA from human remains across Northeast Asia, they asserted that the spread of agriculture in this region was accomplished through the actual physical movement of farmers from the West Liao River basin (Robbeets et al. 2021). Specifically, they identified a genetic component called ‘Amur-like ancestry’ common to all modern speakers of Transeurasian languages and linked it to the early farmers of the West Liao River basin. The spread of this genetic heritage, they argued, proves the migration of farmers and the dispersal of their language (Robbeets et al. 2021).

The convergence of these three lines of evidence provided a powerful and integrated narrative: the millet farmers of the West Liao River basin spread their agricultural technology, along with their genes and language, across Northeast Asia.

 

2.2 Critical Reassessment and Alternative Models 

The hypothesis of Robbeets et al. was met with strong criticism from several scholars, including Tian et al. (2022), immediately after its publication. They argued that there were fundamental problems with each of the three lines of evidence used by the Robbeets team and that their conclusions were the result of an oversimplification of the data.

1. Linguistic Critique: Tian et al. (2022) countered that the lexical and grammatical evidence for a common Proto-Transeurasian ancestor presented by the Robbeets team was very weak, and that the few similarities could be better explained by borrowing and areal contact over a long period rather than a common origin. They pointed out that of the proposed list of over 3,000 cognates, only a tiny fraction satisfied strict sound correspondence rules, which was insufficient to prove a genetic relationship (Tian et al. 2022). In their view, the Transeurasian languages are closer to a ‘sprachbund’—a group of languages that have influenced each other due to geographical proximity—than a genetically related language family (Tian et al. 2022).
2. Archaeological Critique: Critics pointed out that the direct links between the material culture of farming societies in the West Liao River basin (e.g., Hongshan culture) and other regions like the Korean Peninsula and the Japanese archipelago are very tenuous, and there is a lack of archaeological evidence to support a large-scale migration of farmers (Tian et al. 2022; Miyamoto 2022). Instead, they argued that the introduction of agriculture to the Korean Peninsula and Japan showed a more complex pattern of small-scale group movements and the selective adoption of farming techniques by indigenous hunter-gatherers (Jeulmun and Jomon peoples) (Miyamoto 2022).
   Miyamoto (2022) offers a more specific archaeological counterargument to the ‘Farming/Language Dispersal’ hypothesis. He argues that the spread of the Transeurasian languages cannot be explained by a single agricultural dispersal model and that the dispersals of Japonic and Koreanic were separate events driven by different factors at different times.
   • Dispersal of Japonic: Miyamoto links the spread of Japonic to the transmission of a specific pottery-making technology. This technique (using wide clay slabs, external joining, smoothing surfaces with wooden tools) forms a lineage from the Pianpu culture of Liaodong (c. 2700 BCE), through the Gonggwiri-type pottery of the northern Korean Peninsula and the Mumun pottery culture of the south, to the Yayoi culture of Japan. He argues that the continuity of this technological tradition demonstrates the dispersal route of the Japonic language, emphasizing that the initial spread from Liaodong to the Korean Peninsula was not related to the spread of agriculture (Miyamoto 2022).
   • Dispersal of Koreanic: In contrast, he associates the spread of Koreanic with a separate wave of migration that occurred much later, around the 5th century BCE. This migration was not driven by agricultural factors but was triggered by political pressure from the eastward expansion of the Chinese Yan state. These migrants introduced the rolled rim vessel culture and the slender bronze dagger culture to the Korean Peninsula, and he explains that the Koreanic language they brought gradually replaced the Japonic language that was previously spoken on the peninsula (Miyamoto 2022).
     Thus, Miyamoto’s model shows that language dispersal does not necessarily accompany agricultural spread and can be driven by a complex mix of social factors like technological transmission or political migration, thereby challenging the linear hypothesis of Robbeets et al.
3. Genetic Critique: While acknowledging the existence of shared genetic ancestry among speakers of Transeurasian languages, critics argued that the patterns and timing of admixture do not fit a simple model of farmer expansion from a single center in the West Liao River basin (Tian et al. 2022). They contended that the ‘Amur-like ancestry,’ presented as key evidence by Robbeets et al., is a very old genetic component found not only in the West Liao River basin but also among a wide range of hunter-gatherer populations across Northeast Asia, making its exclusive association with a specific farming group problematic (Tian et al. 2022). They also pointed out that in modeling the genetic composition of Korea and Japan, the Robbeets team selectively adopted certain hypotheses (e.g., sequential influence from Hongshan and Upper Xiajiadian cultures) despite the insufficient resolution of their data (Tian et al. 2022).

These critiques conclude that the ‘triangulation’ by Robbeets et al. was a failed attempt, as the three lines of evidence do not actually point to the same conclusion. Instead, Tian et al. argue that the linguistic, genetic, and cultural similarities in these regions are the result of a complex, multidirectional interaction network (a network model) that continued for thousands of years, rather than a tree-like divergence from a single origin. This debate reveals not just a difference in data interpretation but a fundamental methodological divide between attempts to explain human history with simple, clear branching models and those that seek to understand it through complex, entangled network models.

 

Part 3: The Genetic Verdict – Tracing Human Movement Across the Millet Frontier 

The key to settling the ‘Farming/Language Dispersal’ debate ultimately lies in the question: “Did people move with the agricultural technology?” The advancement of ancient DNA analysis provides an unprecedented tool to directly trace the movements and admixtures of past populations. This section synthesizes the vast ancient genomic data accumulated from the West Liao River and Yellow River basins to verify whether population movement—demic diffusion—actually occurred between the two regions. Genetic evidence reveals the direction and scale of macro-level population movements, while stable isotope analysis provides micro-level substance to this big picture by tracing the life trajectories of individuals.

 

3.1 The Genetic Tapestry of Neolithic North China 

Before significant interaction began, Neolithic North China was not a genetically homogeneous space. A distinct north-south genetic cline was already established across East Asia by the Early Neolithic (Wang et al. 2021), and within North China, at least three distinct ancient population groups existed.

1. Yellow River Farmers: This population was centered in the middle and lower Yellow River basin. They were millet farmers with a distinct ancestral lineage (often referred to as ‘Neolithic Yellow River’ or NYR) that maximized in this region (Ning et al. 2020; Wang et al. 2021). They formed the demographic base of the Central Plains.
2. Amur River Hunter-Gatherers: This population was widely distributed in the northern regions from the Amur River basin to the Russian Far East. They had an ancestral lineage (‘Ancient Northeast Asian’ or ANA, also termed ‘Amur-like’) that was clearly distinct from the Yellow River farmers (Ning et al. 2020; Mao et al. 2021). They were the descendants of deep-rooted hunter-gatherers in this region.
3. West Liao River Populations: The Neolithic populations of the West Liao River basin, located geographically between the other two groups, exhibited a mixed genetic profile. They possessed genetic components from both the Yellow River farmers and the Amur River hunter-gatherers (Ning et al. 2020).

These three groups were the main ‘actors’ that shaped the population dynamics of North China, and the interactions and gene flow among them created the complex genetic landscape of the Late Neolithic and Bronze Age.

 

3.2 Evidence of Demic Diffusion: Population Movement from the West Liao River to the Central Plains 

The most direct genetic evidence clearly shows that there was population movement from the West Liao River basin to the Yellow River basin. A 2020 study in Nature Communications by Ning et al., which analyzed 101 ancient genomes from North China, provided decisive data on this issue. According to this study, during the Late Neolithic and Bronze Age, “ancestry related to West Liao River farmers spread southward into the Yellow River Basin” (Ning et al. 2020).

The timing of this southward movement precisely coincides with the archaeological period of agricultural intensification and the emergence of complex societies like the Longshan culture (Ning et al. 2020). Analysis of Longshan period genomes from the Middle Yellow River Basin (MYRB) confirmed a significant influx of new genetic elements different from the preceding Yangshao population. This new component included not only ancestry related to populations from the Lower Yellow River Basin but also ancestry related to northern steppe pastoralists who would have been in contact and admixed with West Liao River groups (Huang et al. 2025, Advance online). This indicates that during the Longshan period, the Central Plains region transformed into a more genetically heterogeneous society through population influx from the outside, and one of the sources of this influx was the West Liao River basin to the north.

 

3.3 A Two-Way Street: Reciprocal Gene Flow from the Central Plains to the West Liao River 

The population movement from the West Liao River to the Central Plains was not a one-way process. Genetic evidence clearly shows that “reciprocal gene flow” also existed from the Yellow River basin to the West Liao River basin (Ning et al. 2020). This implies that the interaction between the two regions was a complex exchange, not a simple migration or conquest.

Y-chromosome haplogroup analysis, which traces paternal lineages, provides specific evidence for this northward movement. Analysis of prehistoric human remains from the West Liao River basin shows an increase in genetic diversity over time, particularly with the influx and increasing frequency of haplogroup O3, which is strongly associated with Yellow River farmers and speakers of Sino-Tibetan languages (Cui et al. 2013). Specifically, the frequency of haplogroup O3 in the later populations of the West Liao River basin reached 37% (Cui et al. 2013). The genetic composition of the Lower Xiajiadian culture population, a Bronze Age culture in the West Liao River basin, clearly shows there was an influx of migrants from the Central Plains (Cui et al. 2013). Recent studies also reaffirm that the Central Plains region acted as a key hub for ancient population movements, and that Late Neolithic Yellow River (YR_LN) related genetic ancestry spread extensively to surrounding areas, including the West Liao River basin (Wan et al. 2025, Advance online).

This bidirectional gene flow illustrates the process by which the two agricultural centers competed and interacted, gradually integrating into one large demographic network.

 

3.4 Corroboration through Isotope Analysis: Tracing Individual Life Trajectories 

While ancient DNA reveals population-level gene flow, stable isotope analysis is a powerful tool for proving the movement of the individuals who constituted that flow. In particular, strontium isotope ratios (87Sr/86Sr) are highly effective in identifying non-local immigrants, as they record the geological characteristics of the region where an individual spent their childhood in their tooth enamel (Bentley, Price and Stephan 2004).

Studies analyzing human remains from the Late Neolithic Longshan culture sites of Haojiatai and Wangchenggang in the Yellow River basin have consistently found that a significant portion of the population (about 25%) were non-locals, not native to the area (Shen et al. 2024; Zhao et al. 2015). This suggests that the major political centers of the Central Plains during this period grew by actively incorporating outside populations.

What is more intriguing is that a significant number of these non-locals were female (Shen et al. 2024). This raises the possibility that inter-regional marriage exchange, following patrilocal customs, was an important social mechanism for population movement and gene flow. In other words, the phenomenon of demic diffusion, which appears as a genetic shift in a large population, may not have been accomplished solely through dramatic events like the conquest by warrior groups or the mass migration of entire clans. Rather, the accumulation of steady, structured movements of individuals through marriage networks sustained over hundreds of years could have led to macro-level changes in the genetic landscape. Thus, isotope analysis gives social substance to the phenomenon of demic diffusion by restoring the life stories of individuals hidden behind the grand narrative.

 

Part 4: Synthesis and Implications 

Synthesizing the archaeological, linguistic, genetic, and isotopic evidence presented thus far, it becomes clear that the spread of millet agriculture in Neolithic North China was not a simple transmission of technology but a process of ‘demic diffusion’ accompanied by the actual movement of people. However, this process was not a unidirectional, conquest-driven expansion but the result of complex and multi-layered interactions. These interactions fundamentally reshaped the genetic landscape of North China and became a decisive moment in laying the demographic foundation for later Chinese civilization.

 

4.1 Synthesis of Evidence: People Moved with the Plow 

The answer to the user’s core question—whether only farming methods spread from the West Liao River to the Central Plains, or if people moved as well—is clear. Multiple independent ancient DNA studies consistently prove that there was significant, bidirectional population movement between the West Liao River basin and the Yellow River basin throughout the Late Neolithic and Bronze Age (Ning et al. 2020; Cui et al. 2013; Wan et al. 2025, Advance online). This population movement is the very definition of demic diffusion. Strontium isotope analysis further solidifies the genetic conclusions by showing that this population-level gene flow was carried out by individuals who moved through social mechanisms such as marriage exchange (Shen et al. 2024; Zhao et al. 2015).

Therefore, the conclusion that ‘people moved with the plow’ is firm. However, the confirmation of demic diffusion as a mechanism does not automatically validate the entire grand narrative of the ‘Farming/Language Dispersal’ hypothesis proposed by Robbeets et al. The reality shown by genetic data is far from a linear dispersal (the ‘Millet Arrow’) that neatly explains the distribution of five major language families originating from a single center in the West Liao River basin. Rather, the archaeological and genetic reality is closer to a ‘Millet Matrix’ of multidirectional exchange, where Yellow River farmers, West Liao River farmers, and surrounding hunter-gatherer groups became intricately entangled, contributing to each other’s gene pools. As Miyamoto (2022) shows, even the dispersal of languages was likely a complex process that occurred at different times, driven by various factors such as agricultural diffusion, technological transfer, and political migration.

 

4.2 Broader Implications: The Formation of a North China Meta-Population 

The long-term interaction and admixture between the two millet farming centers, the West Liao River and the Yellow River, had a profound impact on the population history of North China. The northern genetic component (related to West Liao River/Amur River) and the southern genetic component (related to the Yellow River basin), which were clearly distinct in the Early Neolithic, began to mix actively by the Late Neolithic.

Through this process, a new, much more homogenized genetic landscape was created in North China (Ning et al. 2020). This mixed ‘meta-population’, that is, the dense, admixed population formed from the fusion of West Liao River and Yellow River populations (formed after c. 4000 BP), became the demographic foundation for the Bronze Age civilizations of the Central Plains, namely the Erlitou, Shang, and Zhou dynasties (Wang et al. 2021). And their genetic legacy continues down to the modern northern Han Chinese (Du et al. 2024).

The Central Plains region is often called the ‘cradle of Chinese civilization’ and a ‘melting pot’ where various cultures merged. Ancient genome research shows that this melting pot phenomenon did not suddenly begin in the historical era. The first true melting pot event was the process in the Late Neolithic when the northern farmers of the West Liao River and the southern farmers of the Yellow River met and fused genetically. This event was more than just an exchange of agricultural technology; it was a fundamental demographic event that established the genetic substrate upon which the historical identity of the North China region would be built for the next several thousand years. Therefore, the interaction between the millet farmers of these two core regions can be assessed as the decisive process that constituted the very formation of the North China population.

 

References 

Barton, L., et al. (2009) ‘Agricultural origins and the isotopic identity of domestication in northern China’, Proceedings of the National Academy of Sciences, 106(14), pp. 5523–5528.

Bennett, A. E., Liu, Y. and Fu, Q. (2025) Reconstructing the Human Population History of East Asia through Ancient Genomics. Cambridge: Cambridge University Press.

Bentley, R. A., Price, T. D. and Stephan, E. (2004) ‘Determining the ‘Local’ 87Sr/86Sr Range for Archaeological Skeletons: A Case Study from Neolithic Europe’, Journal of Archaeological Science, 31(4), pp. 365–375. https://doi.org/10.1016/j.jas.2003.06.003.

Cui, Y., et al. (2013) ‘Y Chromosome analysis of prehistoric human populations in the West Liao River Valley, Northeast China’, BMC Evolutionary Biology, 13, p. 216. https://doi.org/10.1186/1471-2148-13-216.

Du, P., et al. (2024) ‘Genomic Dynamics of the Lower Yellow River Valley since the Early Neolithic’, Current Biology, 34(17), pp. 3996–4006.e11. https://doi.org/10.1016/j.cub.2024.07.063.

He, K., et al. (2017) ‘Prehistoric evolution of the dualistic structure mixed rice and millet farming in China’, The Holocene, 27(12), pp. 1885–1898.

Huang, Z., et al. (2025, Advance online) ‘Ancient Genomes Reveal Complex Population Interactions in the Middle Yellow River Basin during the Late Neolithic Period’, Genomics, 117(4), 111061. https://doi.org/10.1016/j.ygeno.2025.111061.

Jia, X., et al. (2016a) ‘Human-environment interactions within the West Liao River basin in Northeastern China during the Holocene Optimum’, Quaternary International, 426, pp. 10-17.

Jia, X., et al. (2016b) ‘The transition of human subsistence strategies in relation to climate change during the Bronze Age in the West Liao River Basin, Northeast China’, The Holocene, 26(5), pp. 781-789.

Liu, L., et al. (2015) ‘Plant domestication, cultivation, and foraging by the first farmers in early Neolithic Northeast China: Evidence from microbotanical remains’, The Holocene, 25(12), pp. 1965-1978.

Lu, H., et al. (2009) ‘Earliest domestication of common millet (Panicum miliaceum) in East Asia extended to 10,000 years ago’, Proceedings of the National Academy of Sciences, 106(18), pp. 7367–7372.

Ma, Z., et al. (2016) ‘Early millet use in West Liaohe area during early-middle Holocene’, Science China Earth Sciences, 59(8), pp. 1554-1561.

Mao, X., et al. (2021) ‘The Deep Population History of Northern East Asia from the Late Pleistocene to the Holocene’, Cell, 184(12), pp. 3256–3266.e13. https://doi.org/10.1016/j.cell.2021.04.016.

Matuzevičiūtė, G. M. and Liu, X. (2021) ‘Prehistoric Agriculture in China’, Oxford Research Encyclopedia of Environmental Science, Oxford UP.

Miyamoto, K. (2022) ‘The Emergence of ‘Transeurasian’ Language Families in Northeast Asia as Viewed from Archaeological Evidence’, Evolutionary Human Sciences, 4, e3. https://doi.org/10.1017/ehs.2021.49.

Ning, C., et al. (2020) ‘Ancient Genomes from Northern China Suggest Links between Subsistence Changes and Human Migration’, Nature Communications, 11, 2700. https://doi.org/10.1038/s41467-020-16557-2.

Robbeets, M., et al. (2021) ‘Triangulation Supports Agricultural Spread of the Transeurasian Languages’, Nature, 599(7886), pp. 616–621. https://doi.org/10.1038/s41586-021-04108-8.

Shen, Y., et al. (2024) ‘Interplay of Dietary Diversity and Mobility of the Late Neolithic People in the Central Plains of China: Multi-Isotope Analysis at the Haojiatai Site’, Anthropological Science, 132(2), pp. 85–104.

Tian, Z., et al. (2022) ‘Triangulation Fails When Neither Linguistic, Genetic, nor Archaeological Data Support the Transeurasian Narrative’, bioRxiv, 9 June.

Wan, W., et al. (2025, Advance online) ‘Ancient Genomes in Southwest China Reveal Genetic Interactions among Diverse Populations in the Historical Period’, BMC Biology, 23, p. 174. https://doi.org/10.1186/s12915-025-02299-4.

Wang, C.-C., et al. (2021) ‘Genomic Insights into the Formation of Human Populations in East Asia’, Nature, 591(7850), pp. 413–419. https://doi.org/10.1038/s41586-021-03336-2.

Wang, J., et al. (2019) ‘Plant exploitation of the first farmers in northwest China: Microbotanical evidence from Dadiwan’, Quaternary International, 529, pp. 3–9.

Wang, J., et al. (2021) ‘Relationship between C4 biomass and C4 agriculture during the Holocene and its implications for millet domestication in Northeast China’, Geophysical Research Letters, 48(1), e2020GL089566. https://doi.org/10.1029/2020GL089566.

Wang, X., et al. (2025a) ‘Multi-Isotope Analysis Reveals Human Millet-Based Diets and Limited Mobility in the Central Plains of China ca. 5000 Years Ago’, International Journal of Osteoarchaeology, 35(3), e3407. https://doi.org/10.1002/oa.3407.

Yang, M. A., et al. (2020) ‘Ancient DNA Indicates Human Population Shifts and Admixture in Northern and Southern China’, Science, 369(6501), pp. 282–288. https://doi.org/10.1126/science.aba0909.

Zhang, M. and Fu, Q. (2020) ‘Human Evolutionary History in Eastern Eurasia Using Insights from Ancient DNA’, Current Opinion in Genetics & Development, 62, pp. 78–84. https://doi.org/10.1016/j.gde.2020.05.009.

Zhao, C., et al. (2015) ‘Strontium Isotope Analysis and Human Mobility from Late Neolithic to Early Bronze Age in the Central Plain of China’, in CAA 2015: Keep the Revolution Going, Archaeopress, pp. 783–788.

Zhao, Y. and Fan, Z. 趙越雲, 樊志民 (2016)「粟·黍·豬論原始旱作農業類型的形成與發展」,《中國農史》, 35(6), pp. 29–43, 64.