1. Introduction

The origins and spread of millets and rice, the major staples of ancient China, have implications regarding the development and spread of cultures, language groups and ethnic groups across eastern Asia (Bellwood, 2005; Blench, 2005; Sagart, 2008; Robbeets, 2015). While wet rice agriculture is arguably the most productive form of landuse and has supported dense and growing populations (e.g., Ellis and Wang, 1997), it was millet agriculture that underpinned the rise of the Chinese state in the Yellow River (Liu and Chen, 2012) and is therefore equally significant for studies of past demography. This paper reviews the evidence for the early cultivation, domestication and spread of Chinese millets and rice agriculture from ca. 7000 to 2000 BC. The review is based on the empirical archaeobotanical evidence for plants preserved on archaeological sites, usually by charring; but other plant remains, less diagnostic of domestication, e.g. phytoliths and starch grains, and artefacts indicative of probable cultivation and harvesting activities are also considered. The review draws upon the Asian Crops Archaeobotanical Database (AsCAD, Stevens et al., 2016), expanded from an existing Rice Archaeological Database (Fuller et al., 2011; Silva et al., 2015). A critical approach to dating has been taken and spurious, untrustworthy radiocarbon dates have been rejected. The half-life has been standardized to 5568 before calibration (Stuiver, 1982), as many dates from China reported in the 1980s and 1990s assumed a half-life of 5730, and this has sometimes resulted in miscalibrations in the literature (Lu, 1999: 9). Where radiocarbon dates are not available, dating has been proposed by cultural association to a cultural phase, with a critical approach taken to the radiocarbon dating of that phase on other sites.

There can be no doubt that agriculture, in most instances, supports higher population densities and rates of population growth than foraging. This has led to the observation that agriculture was more or less a necessary requisite for cultural traditions that have survived into the later Holocene (with the exception of extremely marginal environments) (Richardson et al., 2001), and that most linguistic diversity derives from language lineages that were spoken and spread by early farmers (Bellwood, 2005). Binford (1968) introduced the idea that the transition to farming was part of a related set of post-Pleistocene adaptations, in which humans adapted to richer biotic environments and increased demographic pressure through the exploitation of a wider range of lower-value resources, including the ancestors of seed crops, and ultimately agriculture. What has become clear through recent empirical evidence, however, is that the initiation of cultivation occurred much earlier than the domestication of crops and, in its early stage, was disconnected from true agricultural subsistence, in which domesticated taxa came to dominate caloric subsistence (Asouti and Fuller, 2013; Fuller et al., 2014; Maeda et al., 2016). A key implication is that, contrary to the notion of a single rapid Neolithic demographic transition, which has tended to be dated by conflating the start of cultivation and eventual domestication (e.g., Guerrero et al., 2008; Bocquet-Appel, 2011), there rather was a protracted series of developments in the Neolithic that culminated in agriculture, which in turn supported much larger populations. In the current contribution, we decouple the start of cultivation from the transition to major food surpluses that is implied by reliance on agriculture and domesticates. While incipient cultivation undoubtedly supported higher sedentism and some population increase over pure foraging, it was the transition to an agricultural economy and the establishment of true domesticated crops that is likely connected to unprecedented population increase.

The widely discussed language-farming dispersal hypothesis rests on the premise that increased population densities within early farming societies lead to emigration in search of new land for agriculture and settlement, which simultaneously dispersed various languages (Bellwood and Renfrew, 2002; Bellwood, 2005). Over time this led to the distribution of related languages, ultimately the language families of farmers, over wide regions. Robbeets (2017, this volume) argues that the expansion of Transeurasian languages, including the ancestors of Japonic and Koreanic languages, arose from Panicum miliaceum-based agricultural societies that developed in northeast China. Meanwhile, speakers of languages ancestral to Old Chinese are inferred to have been present in the Yellow River Valley cultivating millets, especially Setaria italica, and some rice. On the basis of similarities in culture and ancestral shared vocabulary for S. italica and rice, Sagart (2008; 2011) has suggested that the origins of Austronesian, which dispersed to Taiwan from the East Asian mainland, lie within or near the Shandong peninsula, rather than the Lower Yangtze Region as postulated by Bellwood (2005). Both Hmong-mien and Austroasiatic have been suggested to be connected to rice origins and dispersal from the Yangtze basin (e.g., Bellwood, 2005; Sagart, 2011), but there has been little detailed correlation of the linguistic and archaeobotanical evidence.

2. Method and philosophy: a critical archaeobotany of domestication and early agriculture

The primary approach represented by this paper is a critical assessment of current archaeobotanical evidence, alongside complementary artefactual evidence of cultivation-related tools. This is not a restatement of commonly held opinions on the Chinese Neolithic, in particular with regards to pre-Yangshao cultures, before 5000 BC. All of these Early Neolithic cultures, which we deal with in this paper, traditionally tend to be regarded as early farming societies (e.g., Liu and Chen, 2012; Underhill, 2013; Shelach and Teng, 2013 and see primary references in Tables S1–S4), but this relies on assigning them agricultural economies more on assertion and assumption rather than on empirical evidence. That such sites were bigger and have more investment in house structures than earlier sites is clear, but this shift towards increased sedentism does not prove year-round occupied villages, nor does it make them farming settlements comparable to those which followed—except by imposing outmoded ethnographic models on the diversity of the pre-agricultural world (see Asouti and Fuller, 2013 for a similar critical perspective on the Near East). Much of this stems from the tendency to equate the presence of domesticated crops, or more precisely, plants undergoing domestication, with fully agricultural societies. This is a recurring problem in many parts of the world and in discussions of agricultural origins (Zeder, 2015; Smith, 2015).

We take as our baseline the clear definitions of cultivation, domestication (a genetic status of a crop), and agriculture from Harris (1989; see more recently Asouti and Fuller, 2013; Harris and Fuller, 2013). Cultivation is seen as an activity of soil preparation, sowing, and harvesting, and in its broadest sense has been widely practiced by many hunter-gatherer societies throughout the Holocene and before, in the sense of niche construction (Smith, 2015; Fuller et al., 2014). Agriculture, in contrast, is about economic dependence, encompassing a predominant reliance on cultivation—usually centered on species that are domesticated, e.g. changed genetically from their wild ancestors. Smith (2001) suggests that agriculture means that more than 50 % of calories come from cultivated resources, and while precise estimates of past diets are always a challenge, this remains a useful rule of thumb. Taking southwest Asian evidence as an example, the earliest cultivation was established across several sub-regions between 9500 and 9000 BC (if not earlier), but at that period no crops show morphological domestication traits. These do not rise to dominance until after 8000 BC, only becoming fixed in wheat and barley around or after 7000 BC (Fuller et al., 2014). Sites that have produced quantified archaeobotanical assemblages comprising more than 50 % cereals do occasionally appear in the Levant by 9000 BC, but on the whole, most early cultivating sites had plant economies dominated by wild foods, with cereal-dominated assemblages only becoming the norm from 7000 BC (Maeda et al., 2016). For the Near East, it is clear that the transition from early cultivation through domestication and agriculture took in excess of 2000 years. Our aim in synthesizing the evidence for China has been to look critically at the available archaeobotanical data and dating. As was the case in the Near East, we have assumed that the transition from early cultivation to agriculture took a minimum of 2000 years, perhaps longer, and the largest demographic increase came at the end of this process. The interpretations presented below follow this reasoning, and to avoid excessive repletion, we will not continually note where this differs from orthodox treatments of Chinese prehistory.

The pre-domestication cultivation stage represents the transition from the gathering of food plants to preparing the soil and sowing them. Initially these plants are morphologically wild, but during this period, genetic and morphological changes occur which “adapt” the plant to the cultivation regime. These biological changes are defined as domestication. A recent synthesis of many major crops from around the world allowed the start and finish of domestication to be mapped comparatively across species (Fuller et al., 2014). For cereals, this study demonstrated two significant results. Firstly, it showed that domestication occurs over extremely protracted periods of time, ranging from 1500 to over 3000 years. Secondly, it confirmed that grain size increased concurrently with changes in shattering, the transition from wild grain dispersal to reliance on human planting. One implication is that grain size change is a useful proxy for the domestication processes when evidence for seed dispersal is not available, as is the case for the millets. While some size increase is noted for millets from early sites (e.g. Zhao, 2004; Liu and Chen, 2012; Barton, 2009: 174–180; Crawford et al., 2016), what is needed are extended sequences demonstrating directional evolution. What these data tend to indicate, as reviewed below, is that the domestication process for millets continued well into the Yangshao period, and this suggests a need to reconsider when agricultural economies and associated demographic shifts took place.

We also differentiate phases in the spread of crops beyond their initial areas of cultivation. Initially, cultivation technologies emerged where wild cereal stands are prolific, replacing wild harvesting. A second phase sees cultivation spread, along with proto-domesticated cereals still undergoing morphological evolution, into areas within the general ecological limits of the wild ancestor, encroaching on and replacing wild cereal stands. At this stage we might expect increasing human population density. The final stage is the spread of fully domesticated crops, along with cultivation technologies, into areas where wild stands were absent or not widespread enough to have formed a major seasonal staple. This third stage is that posited to have spread the language families of early farmers.

3. Centers of cultivation and domestication of millets

Two species of millet were domesticated in China. Foxtail millet, Setaria italica, descended from Setaria viridis, is widespread throughout northern/central China, Central Asia and Europe, but appears to have been domesticated in China (Eda et al., 2013). Panicum miliaceum was also domesticated within northern China, but the wild progenitor is disputed, although Panicum miliaceum subsp. ruderale represents one possibility (De Wet, 2000).

Given the general absence of charred archaeobotanical evidence before 6500 BC, the longevity of millet consumption is currently based on conjecture. Shizitan, Shanxi, dated from 12,000 to 9,600 BC, is one of the few sites where charred grains have been identified, along with millet starch associated with grinding stones (Bestel et al., 2014). The use of millets between 9000–7000 BC is also affirmed through millet starch from grindstones at Nanzhuangtou, Hebei and Donghulin (Yang et al., 2012a; 2015). They appear dominated by Setaria or S. italica-like starch, but given current limitations on reference collections (Yang et al., 2012b), these cannot be accepted as definitive evidence for cultivation or domestication.

Based upon current archaeological evidence, five “centers” of millet cultivation are distinguished (A–E below; see Fig. 1 and Supplementary Materials, Table S1).

A. Peiligang, northern Henan

The Peiligang culture (excluding Jiahu, which is regarded as peripheral) dates to the 7th millennium BC. Radiocarbon dating suggests that sites with good evidence for charred millets occur no earlier than 6500–6000 cal. BC (Liu and Chen, 2012), the earliest evidence relating predominately to reported foxtail millet (Setaria viridis/italica), recovered from three sites to the south of the Yellow River (Fig. 1; Table S1). While other sites in this area, between 6500 and 5000 BC, produced charred remains of wild foods—e.g. walnut, jujube, and acorns—flotation was rarely conducted, so small remains of millet are unlikely to have been recovered. However, many sites did yield evidence for cultivation in the form of sickles, spades and/or hoes (Table S1; Zhu, 2013: Table 9.1; Liu and Chen, 2012: Table 5.3).

View Full Size

Figure 1

Distribution of early sites with archaeobotanical finds of millet and/or rice with median ages between 8000 and 4500 BC. Morphologically wild and domesticated rice differentiated. Regional cultures indicated: A. Peiligang, B. Cishan, C. Houli, D. Xinglongwa, E. Dadiwan, F. Lower Yangtze Neolithic, G. Upper Huai/ Han Neolithic, H. Middle Yangtze Neolithic.

Citation: Language Dynamics and Change 7, 2 (2017) ; 10.1163/22105832-00702001

• Download Figure
• Download figure as PowerPoint slide

Sedentary villages, with increased evidence of cultivation tools, subsequently developed in this region during the Early Yangshao period after 5000 BC (Zhu, 2013). A western Yangshao variant (Early Banpo) and an eastern variant (Hougang period I) are recognized. Recent archaeobotanical research shows increased grain size for Setaria still occurring between the Early and Late Yangshao (Zhao, 2015), suggesting that, while well-established cultivation economies were present, the evolution of domestication traits in millets was still ongoing.

4. Centers of cultivation and domestication of rice

The origins of Asian rice have received considerable archaeological and genetic attention in recent years (e.g., Fuller et al., 2010, 2016; Gross and Zhao, 2014; Castillo et al., 2016; Choi et al., 2017). In this discussion, we concern ourselves only with the origins of subspecies japonica, which is unambiguously domesticated from wild populations in eastern Asia. The earliest undisputed sites with remains of rice associated with cultivation can be divided into three regions: the Lower Yangtze (F), the Middle Yangtze (H) and the Lower Hanshui Valley/Upper Huai Valley (G) (Fig. 1; see Supplementary Materials, Table S2). Differences in material culture and the nature of the earliest field systems between the Middle and Lower Yangtze argue independent developments of rice cultivation (Fuller and Qin, 2009; Makibayashi, 2014). The Middle-Lower Huai River in northern Anhui/western Jiangsu and the Lower Hanshui River in southern Henan could also represent distinct centers.

F. Early rice of the Lower Yangtze and Lower Huaihe River

The evolution of domesticated rice from wild rice is archaeobotanically only well documented for the Lower Yangtze, where large assemblages of spikelet bases have been used to track the domestication process (Fuller et al., 2014). Such empirical evidence is better for the middle and later part of the process (Fuller et al., 2009, 2014; Crawford, 2012). The analysis of the proportion of wild (shattering), domestic (non-shattering) and immature rice spikelet bases from these sites over time provides a clear picture of rice domestication as a nearly linear trend of increasing grain size and increasing proportion of non-shattering spikelet bases between 6000 and 3000 BC (Fig. 2).

View Full Size

Figure 2

Illustration of episodes of evolution of domestication in Lower Yangtze rice, including percentage of non-shattering (top) and grain width (bottom)after Fuller, 2014

Citation: Language Dynamics and Change 7, 2 (2017) ; 10.1163/22105832-00702001

• Download Figure
• Download figure as PowerPoint slide

The earliest dated sites with good evidence for rice use cluster around 7000 BC and are associated with the Shangshan culture (Table S2). Little evidence from macro-remains is available from these sites, so they might represent the beginnings of cultivation or groups who still subsisted solely on gathered wild rice. However, phytolith studies (Wu Y. et al., 2014) put Shangshan at the beginning of a trend in phytolith metrics that led to increasingly domesticated rice plants in later sites like Kuahuqiao and Tianluoshan.

The Shangshan sites were likely only seasonally occupied, with a high reliance on wild foods, but with increased sedentism seen in later phases (Liu and Chen, 2012: 63; Jiang, 2007). The ensuing Kuahuqiao culture is dated from around 6000–5400 BC, with at least some cultivation tools (ZPIACR, 2004), while rice grains and spikelet bases indicate cereals in an early stage of pre-domestication cultivation (Fuller et al., 2014). Extensive use of wild foods, notably acorns and Trapa water chestnuts, is similar to the subsequent Hemudu culture (Fuller and Qin, 2010). Unlike the Shangshan, Kuahuqiao appears fully sedentary (Liu and Chen, 2012: 70–72, 158–160).

By 5000 BC, the major sites of Hemudu and Tianluoshan had appeared in the Yuyao region east of Hangzhou, while the Early Majiabang culture emerged north of Hangzhou Bay. The rice spikelet bases from Tianluoshan fall near the middle of the domestication episode, with a shift to the domesticated type outnumbering the wild form around 4650 BC (Fuller et al., 2009). Majiabang and Hemudu grains were plumper than those of earlier Kuahuaqiao but yet to reach the size seen in later sites (Fuller et al., 2010; 2014). Wooden hoes and bone spades from Kuahuqiao and Hemudu were suitable for working softer wetland soils, but later were displaced by stone, suited to harder clays (Xie et al., 2015). By around 4000–3800 BC (Late Majiabang period), the percentage of domestic as opposed to wild spikelet bases at Caoxieshan reached 70–80 % in some samples, representative of full domestication. Simultaneously, wild foods like acorns and Trapa largely dropped out of the diet from the late Majiabang era onwards (Fuller et al., 2010). Rice agriculture is further indicated by the preserved field systems of Caoxieshan, Chuodun and later sites, indicating investment in landscape modification for agricultural systems (Fuller and Qin, 2009). While the Hemudu animal economy focused on deer hunting and fish from fresh water wetlands (Zhang Y., 2014; Nakajima et al., 2010), an age profile shift in pigs suggests ongoing pig domestication during the 5th millennium BC (Zhang Y., 2014).

Parallel processes of early cultivation and domestication of rice can be suggested for societies in the Lower Huai river valley, which joins the lower Yangtze. Slightly predating the Hemudu sites, rice grain impressions, along with stone spades or shovels, are reported from the Houjiazhai culture site of Shuangdun on the Huaihe River in northern Anhui from 5300 to 5050 BC (Zhang and Ren, 2005). Systematic archaeobotanical evidence for assessing domestication status is presently missing. Recently the Shunshanji Neolithic culture, with evidence for rice older than 6000 BC, has been recognized somewhat further north (Nanjing Museum, 2016). However, whether this may come to represent a separate pre-domestication cultivation event or was tied to the spread of cultivation from the Lower Yangtze is presently unclear.

5. The integration of rice and millet agriculture, 4000–3500 BC

There is no evidence to suggest that millets and rice were domesticated simultaneously within a single region, or brought into pre-domestication cultivation by a single culture. While it is likely that Setaria italica and Panicum miliaceum became cultivated and domesticated together, the data for the earliest cultivation stages is too incomplete to assess their eventual integration. Only in the case of Shandong’s Houli culture might all three have been cultivated together early on, but the dating of cultivation here is based on the least conclusive evidence of any region, with no evidence supporting a full trajectory to domestication continuing from Houli. The empirical archaeobotanical evidence therefore points to distinct separate processes of plant domestication for Chinese millets and rice, and probably in multiple regions. The fragmentary evidence suggests that millet domestication took millennia, as was the case with rice, wheat, barley and other crops (Fuller et al., 2014); millet cultivation was thus not different in terms of the pace of domestication (contra Bettinger et al., 2010a). Some of these starts to cultivation may be evolutionary dead-ends, such as Dadiwan millets, Jiahu rice or Houli. In all cases, the earliest part of the process is most obscure, while the tail end, as agricultural dependence and unambiguous sedentism emerged, is better dated.

In the period leading up to 4000 BC, domesticated rice and millets began to spread. In the case of rice, this seems to have been more restricted to intraregional filling in; thus there is an absence of sites south of the Yangtze. By contrast, millet cultivation spread extensively, especially with the expansion of the Early Yangshao cultural tradition.

View Full Size

Figure 3

Distribution of sites with archaeobotanical finds of millet and/or rice with median ages between 4500 and 3000 BC. Selected sites with both millet and rice labelled: (1) Nanjiaokou; (2) Baligang; (3) Huitupo; (4) Chengtoushan.

Citation: Language Dynamics and Change 7, 2 (2017) ; 10.1163/22105832-00702001

• Download Figure
• Download figure as PowerPoint slide

It is at the end of this first expansionary phase, around 4000 BC, that we find the first evidence for the integration of millet and rice agriculture into the economies of single communities (Fig. 3; see Supplementary Materials, Table S3). Largely this represents the uptake of rice by millet farmers, who were culturally more expansive, with numerous Yangshao sites having evidence for both millets and rice. The earliest direct AMS date on rice in the core Yangshao area is 4000–3800 cal. BC, at Nanjiaokou, Sanmenxia (western Henan), from an assemblage dominated by foxtail millet (Qin and Fuller, 2009). Other early Yangshao rice finds could date back to 4500 BC, but lack direct dates. The southern expansion of the Yangshao culture, seen at Baligang and nearby Huitupo in the Upper Han River, starts c. 4300 BC with evidence for cultivation of rice, Setaria and Panicum (Deng et al., 2015; Weisskopf, 2014). Further south in the Middle Yangtze, foxtail millet has been found at Chengtoushan, otherwise dominated by rice agriculture, 4300–4000 BC (Nasu et al., 2012). It is notable that Lower Yangtze agriculture, well documented up to ca. 2000 BC, focused exclusively on rice, without evidence for millets or soybeans.

6. The southern and western spread of domesticated rice and millet agriculture after 3500 BC

By 3500 BC, the spread of rice and millets appears much more rapid, in particular after 3000 BC, when a significant spread of cereals into regions which had previously seen no cultivation of grain crops occurred (Fig. 4). Prior to 3000 BC, Panicum and Setaria spread beyond Gansu into western Qinghai and south to western Sichuan (D’Alpoim Guedes and Butler, 2014; Chen F. et al., 2015). There is further evidence for rice within Gansu (e.g., An et al., 2010). The first cultivation in western Sichuan, from ca. 3300 BC, is based exclusively on millets and associated with the Majiayao expansion out of Gansu (derived from the Yangshao culture), with rice added after 2700 BC during the Baodun phase, presumably from a separate expansion up the Yangtze (D’Alpoim Guedes, 2011; Fuller et al., 2010; D’Alpoim Guedes et al., 2013). Further finds of millet from the Tibetan Plateau, Qamdo Karuo, start from c. 2800 BC, but are inferred as traded from a lower elevation (D’Alpoim Guedes and Butler, 2014; D’Alpoim Guedes et al., 2014), although it should be noted this site has stone sickles indicative of harvesting (CPAM, 1985).

View Full Size

Figure 4

Distribution of sites with archaeobotanical finds of millet and/or rice with median ages between 3500 and 2000 BC. Selected sites labelled: (1) Baodun, (2) Qamdo Karuo, (3) Baiyangcun, (4) Haimenkou, (5) Dadunzi, (6) Shifodong, (7) Gantuoyan, (8) Wenjiatun, (9) Jitap-ri and Masan-ri, (10) Daechonri [problematic], (11) Oun and Sangchon B, (12) Kazahari, (13) Ryugasaki, (14) Nabatake (ca. 850 BC), (15) Khok Phanom Di, (16) Non Pa Wai.

Citation: Language Dynamics and Change 7, 2 (2017) ; 10.1163/22105832-00702001

• Download Figure
• Download figure as PowerPoint slide

The Baodun culture is the ultimate source of the spread of mixed rice and millet agriculture into Yunnan from around 2500 BC onwards (D’Alpoim Guedes et al., 2013). Rice, foxtail millet and broomcorn millet are found on Neolithic sites at Baiyangcun dating between 2500 and 1750 BC (Stevens/Fuller, unpublished data) and later at Haimenkou from 1700 BC (Xue, 2010), Dadunzi (Jin H. et al., 2014) and Shifodong (Zhao, 2010b), both dating from 1500 to 1000 BC (see Table S4).

As for south of the Yangtze, much of the limited evidence relates to the arrival of rice, at a few sites in Guangdong and Fujian dating generally from around 2500 BC (Table S4; Fuller et al., 2010; Zhang and Hung, 2010). Early populations in these areas focused on nuts (acorns, Canarium) and vegecultural plants, like palm pith starch, wild bananas and Chinese arrowroot (Yang et al., 2013; Yang, pers. comm.). In Guangxi, foxtail millet and rice co-occur at Gantuoyan, probably dating to the 2nd millennium BC (D’Alpoim Guedes et al., 2013).

The other significant movements of rice and millet(s) were to Taiwan and perhaps coastal southern China via maritime routes. Given the absence of millet cultivation in the Lower Yangtze, it is likely that the first cereals in Taiwan derive from further north, such as the Shandong Peninsula, and were carried with the dispersal of maritime-focused groups, whose presence in Shandong is seen through marine shell midden sites from the Dawenkou through Longshan periods (Yuan et al., 2002). This hypothesis is further supported by shared customs like tooth evulsion seen in burials (Sagart, 2008). The earliest finds of foxtail millet and rice from lowland sites in western and northern Taiwan date from between c. 2700 and 2300 BC (Tsang, 2005; Hsieh et al., 2011; and see Supplementary Materials, Table S4). Panicum miliaceum may also have been present, given its later importance amongst highland Formosan tribes (Fogg, 1983). In addition to cereals, Taiwanese pigs are genetically derived from central China, unlike pigs further south or in the Pacific (Larson et al., 2010). Maritime cultures, perhaps from Shandong, carrying Chinese cereals to Taiwan may also have transmitted farming among the coastally focused fisher-hunter-gatherers of Fujian and eastern Guangdong, which have strong cultural links to each other and to Taiwan (Jiao, 2007), and later perhaps onwards through parts of island and mainland southeast Asia (Bellwood, 2011; Higham, 2014).

This dispersal of rice and millet together into the tropical far south of China provided the passage for cereal agriculture, predominately rice with some foxtail millet, into mainland southeast Asia, perhaps as early as 2500–2000 BC (Weber et al., 2010), but most notably after 2000 BC (Table S4). Likewise, the spread of millets into northern Gansu resulted in their eventual dispersal into Central Asia by at least 2000 BC (Frachetti et al., 2010; Spengler et al., 2014; Table S4). The movement of wheat and barley (along with sheep and cattle) into China is of a likely similar date (Dodson et al., 2013; Barton and An, 2014).

7. The eastern spread of domesticated rice and millet agriculture after 3500 BC

The migration of agriculture out of northeast China brought the cultivation of millets to south-eastern Siberia, just beyond Jilin and Heilongjiang, by 3500–3400 BC (Sergusheva and Vostresov, 2009), millets to the Korean peninsula in the Middle Chulmun period by 3500–3000 BC (Crawford and Lee, 2003; Lee, 2011) followed by rice around 1500 BC (Ahn, 2010), and after some delay both crops to Japan after 700–600 BC.

The movement of foxtail and broomcorn millet into Korea has only a few direct radiocarbon determinations on grains from South Korean sites, suggesting their introduction between 3500 and 3000 cal. BC (Crawford and Lee, 2003; Lee, 2011; Table S4). Early North Korean sites, between 3500 and 2000 BC, include Jitap-ri and Masan-ri where only Setaria italica was found, but identifications may be unreliable (Lee, 2011; Kim, 2014). The date of the arrival of rice in Korea is less easy to establish. Early finds from Daechon-ri associated with the arrival of millet are uncharred and hence possibly more recent (Lee, 2011). Other early sites like Oun-1 or grains found in peat also have dating problems or include the possibility that the rice may be wild (Crawford and Lee, 2003; Ahn, 2010). Currently, much of the earliest evidence seems to center around the late 2nd millennium BC, the earliest dates being around 1300–1000 BC (Table S4) when agricultural settlements, stone harvesting knives, wooden tillage and pounding tools are recovered from archaeological sites (Ahn, 2010).

The arrival of foxtail and broomcorn millet, shortly after 3500 BC, was probably either across the sea or around the coast via Liaoning into North Korea, as suggested by Ahn (2010). An earlier dispersal of just millets from northeast China would be congruent with the posited language expansion of Proto-Japano-Koreanic (Robbeets, 2015). Ahn (2010) also suggests that rice entered Korea along a similar northern route via the Liaodong Peninsula, and while to date there is only one site recorded with rice, Wenjiatun 3000–2600 BC in Liaoning (Miyamoto, 2009), this route has stronger archaeological support.

The final spread of millets and rice into Japan is particularly poorly dated. The earliest two direct dates, both on rice grains from Kazahari, place the introduction of rice agriculture between 980 and 380 cal. BC (Table S4; D’Andrea et al., 1995), but have wide error margins and are in poor agreement with each other. The younger suggests a date of between 780 and 380 cal. BC, and it might be noted that direct dates of 810–550 cal. BC from Ryugasaki on Panicum miliaceum (Miyata et al., 2007) suggest a similar introduction date for millets.

8. The demographic and environmental impact of rice and millet farming (3500–2000 BC)

The evidence reviewed above indicates that agricultural economies based on millets in northern China and rice in the Yangtze basin were well established by the start of the 4th millennium BC. In many parts of the Yellow River Basin, rice was also grown by millet farmers, but appears to be a minor component of agriculture in the Yangshao culture (e.g., Lee et al., 2007; Fuller and Zhang, 2007). While the protracted domestication process began probably prior to 6000 BC for both millets and rice, and plausibly independently across as many as 8 regions, the impact of early cultivation on human economies and demography appears minimal initially. Where evidence is available, wild foods largely outranked early cultivated rice (e.g., Fuller et al., 2009; Fuller and Qin 2010; Zhao, 2010a; Deng et al. 2015). The drivers of population growth and expansion through migration must be sought in the period when domesticated cereals and agriculture became established between 4500 BC and 3000 BC. There are two aspects of population expansion that should be considered: internal packing, as sites became larger and denser with more of the available land between them colonized for agriculture, and external expansion, the migration of farmers in search of new land in new regions.

The internal packing of growing populations is demonstrated in the distribution of archaeological sites and by evidence for deforestation, implying acquisition of land for agriculture. As a visual comparison, the distribution of millet-growing sites indicates that the total geographical spread of millet only expanded slightly after 4500 BC and only a little more after 3000 BC (Figs 1, 3, 4; Stevens et al., 2016), but that the total number of sites increased. This pattern is even more striking in comprehensive maps based on survey data (Liu, 2004; Li et al., 2009; Wagner et al., 2013). The other indication is massive and sustained reduction in forest cover, estimated from palynological evidence compiled across the region (Ren, 2007). Pollen-derived estimates of forest cover for the Yangtze basin (including the Huai River) and the Yellow River Basin (Middle and Lower) show that forest cover rose through the early Holocene, presumably due to warmer and wetter climate, with peaks reached around 6000 BC, when forest cover was around 90 % for the Yangtze and 50 % for the Yellow River. Subsequently, there is a marked and sustained reduction in forest (Fig 5A). After 4000 BC, deforestation stopped in the Yangtze basin, but became more pronounced across north China. While there might be some climatic influence, as northern China was more adversely affected by increased aridification (especially 3500–2000 BC), we would argue that differences in agricultural practices between the two regions were more important. Millet production can be most readily raised by colonizing new land, whereas rice output can be increased by intensification of practices within the same unit of land, which has been seen within archaeological and archaeobotanical data in both the Middle and Lower Yangtze regions (Fuller and Qin, 2009; Weisskopf, 2014; Weisskopf et al., 2015).

View Full Size

Figure 5

Proxies for demographic change: (A) deforestation trends estimated from regional pollen databases (after Ren, 2007); (B) total increase in recorded archaeological sites across north China (after Wagner et al., 2013); (C) regional site counts per century, geometric scale (Upper Yellow River data from Li et al., 2009; others from Wagner et al., 2013); (D) regional site counts per century, geometric scale (data from Li et al., 2009)

Citation: Language Dynamics and Change 7, 2 (2017) ; 10.1163/22105832-00702001

• Download Figure
• Download figure as PowerPoint slide

Population growth is evident from the increase in the density of archaeological sites in the landscape and their size range. Several studies of the substantial published archaeological surveys in central and northern China indicate rising site numbers and the growth of larger sites during the course of the Chinese Neolithic (Li et al., 2009; Wagner et al., 2013). The largest compilation across northern China (Yellow River, Inner Mongolia and northeast China) demonstrates a massive increase in total site counts between 8000 and 2000 BC (Fig. 5B). In order to explore this data on a more regional scale, site counts have been averaged per century to account for the different timespans of cultural phases (Fig. 5C–5D). Note that the site count (vertical axis) is a geometric scale, and thus the overall slope of the trends represents approximate geometric (exponential) growth in site number.

In addition to an increasing number of sites, the sites themselves got bigger. While data is not systematically available across all of China, central Henan may be taken as an exemplar (after Liu, 2004). In the Peiligang period (7th millennium BC), the largest site was around 6 hectares, while the median site size was 1 ha. During the Yangshao period (5000–3000 BC), the median site size rose to around 5 ha, the largest being around 40 ha. In the Longshan period (2500–1900 BC) the largest was c. 55 ha, but during the Erlitou period, Erlitou itself reached 300 ha (ca. 1700 BC), with several sites attaining sizes between 30 and 75 ha, although the median was around 3 ha. Thus, the growth of population led to an increased number of village sites, the growth of some urban centers, and agricultural expansion through deforestation.

It is in this context that we need to consider the pressures that promoted the expansion of farming via colonization of new regions, the prime mover that underpins the expansion of farmers and their languages (Bellwood, 2005). As both millet and rice farming populations expanded simultaneously, expansion between millet- and rice-focused regions was constrained. A northward expansion of millet farming was limited by climate, as rainfall and the number of growing days became inadequate for millet farming; these climatic constraints likely intensified with increased aridification leading up to 2000 BC. Thus the main outlets for the migratory expansion of millet farmers was to the west and south (along the edge of the Tibet-Qinghai highlands), and eastwards into the Korean Peninsula, far eastern Russia, and potentially, by maritime links, to Taiwan and non-agricultural southeast China. These archaeologically appear as the main areas of expansion, especially during the 3rd millennium BC. Presumably, the initial pulse in population growth early in the Yangshao-Dawenkou era (4th millennium BC) was focused on internal expansion and deforestation, with the second pulse increasingly leading to external migration.

Rice farming was less expansive initially. The ability to intensify wet rice agriculture in the Yangtze basin and its tributaries potentially absorbed considerable population growth through intra-regional population packing. Additionally, the high labor costs of wet rice agriculture probably made its translocation by small frontier populations formidable, and potentially also reduced the attraction of adopting such systems to other societies, creating friction to its dispersal (Fuller and Qin, 2009). Instead, rice only spread to southeast Asia once less labor-demanding rainfed (dry) rice systems evolved (Fuller et al., 2011; Castillo et al., 2016).

9. Concluding remarks: implications for language dispersal

Early cultivation practices are unlikely to have caused major shifts in the distribution of population and, by association, culture or language. Rather, the archaeological evidence for the development of agriculture in China argues for major demographic expansions, which can be expected to have spread major language families geographically, to have taken place from the end of the domestication process. These dispersal processes took place out of the central plains of China, including the middle and lower basin of the Yangtze and lower basin of the Yellow River. Two major demographic pulses are identified that are associated with the establishment of agriculture; the first began around 4000 BC, intensifying between 3000–2500 BC. The first pulse witnessed the spread of millets westward and eastward, while the second pulse saw major expansions of rice and millet southwards by three major routes (Yunnan, Guangdong, and to the southeast). Thus, we have five geographical trajectories of agricultural spread, and perhaps it is no coincidence that five language families have been postulated as having experienced agriculturally based expansions out of greater China (Sino-Tibetan, Austronesian, Austroasiatic, Hmong-Mein and Japano-Koreanic). Alternative theories purporting to explain geographical derivations from Sino-Tibetan, such as the highly diverse Tibeto-Burman language zone of Northeastern India through Yunnan (e.g., van Driem, 2012), should be rejected, as they are clearly contravened by the archaeological evidence (for an explicit quantitative test of fit with rice data, see Silva et al., 2015). Nevertheless, there is much work to be done to tie together the linguistic evidence for proto-language farming vocabularies, or other associated material culture, and the archaeological geographies of early farming. What the archaeobotanical evidence can provide is a map, in time and space, of subsistence systems, where cultivation began, domestication evolved and agriculture became established as part of demographic expansion.

Supplementary materials

Table S1 lists sites with archaeobotanical evidence for millet and/or evidence for cultivation and dating. Table S2 shows sites with rice. Table S3 lists earliest sites with co-occurrence of millets and rice. Table S4 lists sites with rice and/or millets from Western China, Tibet, Central Asia, Korea and Japan.

References

Ahn, Sung-Mo. 2010. The emergence of rice agriculture in Korea: Archaeobotanical perspectives. Archaeological and Anthropological Sciences 2: 89–98.

An, Chengbang, Duxue Ji, Fa-hu Chen, Guanghui Dong, Hui Wang, Weimiao Dong, and Xueye Zhao. 2010. Evolution of prehistoric agriculture in central Gansu Province, China: A case study in Qin’an and Li County. Chinese Science Bulletin 55(18): 1925–1930.

Asouti, Eleni and Dorian Q Fuller. 2013. A contextual approach to the emergence of agriculture in Southwest Asia: Reconstructing Early Neolithic plant-food production. Current Anthropology 54(3): 299–345.

Barton, Loukas. 2009. Early Food Production in China’s Western Loess Plateau. PhD dissertation, University of California, Davis.

Barton, Loukas and Chengbang An. 2014. An evaluation of competing hypotheses for the early adoption of wheat in East Asia. World Archaeology 46(5): 775–798.

Barton, Loukas, Seth D. Newsome, Fa-Hu Chen, Hui Wang, Thomas P. Guilderson, and Robert L. Bettinger. 2009. Agricultural origins and the isotopic identity of domestication in northern China. Proceedings of the National Academy of Sciences of the U.S.A. 106(14): 5523–5528.

Bellwood, Peter. 2005. First Farmers. Oxford: Blackwell.

Bellwood, Peter. 2011. The checkered prehistory of rice movement southwards as a domesticated cereal—from the Yangzi to the Equator. Rice 4(3–4): 93–103.

Bellwood, Peter and Colin Renfrew (eds.). 2002. Examining the Farming/Language Dispersal Hypothesis. Cambridge: McDonald Institute for Archaeological Research.

Bestel, Sheahan, Gary W. Crawford, Li Liu, Jinming Shi, Yanhua Song, and Xingcan Chen. 2014. The evolution of millet domestication, Middle Yellow River Region, North China: Evidence from charred seeds at the late Upper Paleolithic Shizitan Locality 9 site. The Holocene 24(3): 261–265.

Bettinger, Robert L., Loukas Barton, and Christopher Morgan. 2010a. The origins of food production in North China: A different kind of agricultural revolution. Evolutionary Anthropology 19: 9–21.

Bettinger, Robert L., Loukas Barton, Christopher Morgan, Fa-hu Chen, Hui Wang, Thomas P. Guilderson, Duxue Ji, and Dongju Zhang. 2010b. The transition to agriculture at Dadiwan, People’s Republic of China. Current Anthropology 51(5): 703–714.

Binford, Lewis R. 1968. Post-pleistocene adaptations. In Sally R. Binford and Lewis R. Binford (eds.), New Perspectives in Archaeology, 313–341. Chicago: Aldine.

Blench, Roger. 2005. From the mountains to the valleys: Understanding ethnolinguistic geography in Southeast Asia. In Laurent Sagart, Roger Blench, and Alicia Sanchez-Mazas (eds.), The Peopling of East Asia, 31–50. London: Routledge.

Bocquet-Appel, Jean-Pierre. 2011. The agricultural demographic transition during and after the agriculture inventions. Current Anthropology 52 (S4): S497–S510.

Castillo, Cristina C., Katsunori Tanaka, Yo-Ichiro Sato, Ryuji Ishikawa, Bérénice Bellina, Charles Higham, Nigel Chang, Rabi Mohanty, Mukund Kajale, and Dorian Q Fuller. 2016. Archaeogenetic study of prehistoric rice remains from Thailand and India: Evidence of early japonica in South and Southeast Asia. Archaeological and Anthropological Sciences 8(3): 523–543.

Chang, Kwang-chih. 1986. The Archaeology of Ancient China. 4th ed. New Haven: Yale Press.

Chen, Baozhang, Zhang Juzhong, and Houyuan Lu. 1995. Discovery of rice phytoliths in the Neolithic site at Jiahu, Henan province and its significance. Translation by G. Leir. Chinese Science Bulletin 14(40): 1186–1191. [in Chinese]

Chen, Fa-hu, Guanghui Dong, Dongju Zhang, Xinyi Liu, Xia Jia, Chengbang An, and Martin K. Jones. 2015. Agriculture facilitated permanent human occupation of the Tibetan Plateau after 3600 BP. Science 347(6219): 248–250.

Choi, Jae Young, Adrian E. Platts, Dorian Q Fuller, Yue-le Hsing, Rod A. Wing, and Michael D. Purugganan. 2017. The rice paradox: Multiple origins but single domestication in Asian rice. Molecular Biology and Evolution 34(4): 969–979. doi: 10.1093/molbev/msx049.

Cohen, David J. 2011. The beginnings of agriculture in China. A multiregional view. Current Anthropology 52(S4): S273–S293.

CPAM (Commission for the Preservation of Ancient Monuments, Tibet) and Sichuan University Department of History. 1985. Karou: A Neolithic Site in Tibet. Beijing: IACASS. [in Chinese]

Crawford, Gary W. 2012. Early rice exploitation in the lower Yangzi valley: What are we missing? The Holocene 22(6): 613–621.

Crawford, Gary W., Xuexiang Chen, Fengshi Luan, and Jianhua Wang. 2016. People and plant interaction at the Houli Culture Yuezhuang site in Shandong Province, China. The Holocene 26 (10): 1594–1604.

Crawford, Gary W., Xuexiang Chen, and Jianhua Wang. 2006. Discovery of fossilized crops from the Houli site of Yuezhuang in the Jinan Region of Shandong. Dongfang Kaogu 3: 247–251. [in Chinese]

Crawford, Gary W. and Gyoung-Ah Lee. 2003. Agricultural origins in the Korean peninsula. Antiquity 77(295): 87–95.

Crawford, Gary W. and Chen Shen. 1998. The origins of rice agriculture: Recent progress in East Asia. Antiquity 72(278): 858–866.

D’Alpoim Guedes, Jade. 2011. Millets, rice, social complexity, and the spread of agriculture to the Chengdu Plain and southwest China. Rice 4: 104–113.

D’Alpoim Guedes, Jade and Ethan E. Butler. 2014. Modelling constraints on the spread of agriculture to Southwest China with thermal niche models. Quaternary International 349: 29–41.

D’Alpoim Guedes, Jade, Ming Jiang, Kunyu He, Xiaohong We, and Zhanghua Jiang. 2013. Site of Baodun yields earliest evidence for the spread of rice and foxtail millet agriculture to south-west China. Antiquity 87: 758–771.

D’Alpoim Guedes, Jade, Guiyun Jin, and R. Kyle Bocinsky. 2015. The impact of climate on the spread of rice to north-eastern China: A new look at the data from Shandong Province. PLoS ONE 10(6): e0130430. doi: 10.1371/journal.pone.0130430.

D’Alpoim Guedes, Jade, Hongliang Lu, Yongxian Li, Robert N. Spengler, Xiaohong Wu, and Mark S. Aldenderfer. 2014. Moving agriculture onto the Tibetan plateau: The archaeobotanical evidence. Archaeological and Anthropological Sciences 6: 255–269.

D’Andrea, A. Catherine, Gary Crawford, Masakazu Yoshizaki, and Takehisa Kudo. 1995. Late Jomon cultigens in northeastern Japan. Antiquity 69: 146–152.

De Wet, J.M.J. 2000. Millets. In Kenneth F. Kiple and Kriemhild F. Ornelas (eds.), Cambridge World History of Food, 112–121. Cambridge: Cambridge University Press.

Deng, Zhenhua, Ling Qing, Yu Gao, Alison Ruth Weisskopf, Chi Zhang, and Dorian Q Fuller. 2015. From early domesticated rice of the Middle Yangtze basin to millet, rice and wheat agriculture: Archaeobotanical macro-remains from Baligang, Nanyang basin, Central China (6700–500 BC). PLoS ONE 10(10): e0139885. doi: 10.1371/journal.pone.0139885.

Dodson, John R., Xiaoqiang Li, Xinying Zhou, Keliang Zhao, Nan Sun, and Pia Atahan. 2013. Origin and spread of wheat in China. Quaternary Science Reviews 72: 108–111.

Eda, Meiko, Ayumi Izumitani, Katsuyuki Ichitani, Makoto Kawase, and Kenji Fukunaga. 2013. Geographical variation of foxtail millet, Setaria italica (L.) P. Beauv. based on rDNA PCR–RFLP. Genetic Resources and Crop Evolution 60(1): 265–274.

Ellis, Erle C. and Siming Wang. 1997. Sustainable traditional agriculture in the Tai Lake Region of China. Agriculture, Ecosystems, and Environment 61: 177–193.

Flad, Rowan K., Jing Yuan, and Shuicheng Li. 2007. Zooarchaeological evidence for animal domestication in Northwest China. In David B. Madsen, Fa-Hu Chen, and Xing Gao (eds.), Late Quaternary Climate Change and Human Adaptation in Arid China, 163–199. Amsterdam: Elsevier.

Fogg, Wayne H. 1983. Swidden cultivation of foxtail millet by Taiwan aborigines: A cultural analogue of the domestication of Setaria italica in China. In David N. Keightley (ed.), The Origins of Chinese Civilization, 95–115. Berkeley: University of California Press.

Frachetti, Michael D., Robert N. Spengler, Gayle J. Fritz, and Alexei N. Mar’yashev. 2010. Earliest direct evidence for broomcorn millet and wheat in the central Eurasian steppe region. Antiquity 84: 993–1010.

Fuller, Dorian Q. 2011. Pathways to Asian civilizations: Tracing the origins and spread of rice and rice cultures. Rice 4: 78–92.

Fuller, Dorian Q, Tim Denham, Manuel Arroyo-Kalin, Leilani Lucas, Chris J. Stevens, Ling Qin, Robin G. Allaby, and Michael D. Purugganan. 2014. Convergent evolution and parallelism in plant domestication revealed by an expanding archaeological record. Proceedings of the National Academy of Sciences of the U.S.A. 111(17): 6147–6152.

Fuller, Dorian Q, Emma Harvey, and Ling Qin. 2007. Presumed domestication? Evidence for wild rice cultivation and domestication in the fifth millennium BC of the lower Yangtze region. Antiquity 81: 316–331.

Fuller, Dorian Q and Ling Qin. 2009. Water management and labour in the origins and dispersal of Asian rice. World Archaeology 41(1): 88–111.

Fuller, Dorian Q and Ling Qin. 2010. Declining oaks, increasing artistry, and cultivating rice: The environmental and social context of the emergence of farming in the Lower Yangtze Region. Environmental Archaeology 15(2): 139–159.

Fuller, Dorian Q, Ling Qin, Yunfei Zheng, Zhijun Zhao, Xugao Chen, Leo Aoi Hosoya, and Guo-Ping Sun. 2009. The domestication process and domestication rate in rice: Spikelet bases from the Lower Yangtze. Science 323(5921): 1607–1610.

Fuller, Dorian Q, Yo-Ichiro Sato, Cristina Castillo, Ling Qin, Alison R. Weisskopf, Eleanor J. Kingwell-Banham, Jixiang Song, Sung-Mo Ahn, and Jacob van Etten. 2010. Consilience of genetics and archaeobotany in the entangled history of rice. Archaeological and Anthropological Sciences 2(2): 115–131.

Fuller, Dorian Q, Jacob van Etten, Katie Manning, Cristina Castillo, Eleanor Kingwell-Banham, Alison Weisskopf, Qin Ling, Yo-Ichiro Sato, and Robert J. Hijmans. 2011. The contribution of rice agriculture and livestock pastoralism to prehistoric methane levels: An archaeological assessment. The Holocene 21(5): 743–759.

Fuller, Dorian Q, Alison R. Weisskopf, and Cristina Cobo Castillo. 2016. Pathways of rice diversification across Asia. Archaeology International 19: 84–96.

Fuller, Dorian Q and Hai Zhang. 2007. A preliminary report of the survey archaeobotany of the Upper Ying Valley (Henan Province). In School of Archaeology and Museology, Peking University, and Henan Provincial Institute of Cultural Relics and Archaeology (eds.), Dengfeng wangchenggang yizhi de faxian yu yanjiu (2002–2005) [Archaeological Discovery and Research at the Wangchenggang Site in Dengfeng (2002–2005)], 916–958. Zhengzhou: Great Elephant Publisher. [in Chinese]

Gross, Briana L. and Zhijun Zhao. 2014. Archaeological and genetic insights into the origins of domesticated rice. Proceedings of the National Academy of Sciences of the U.S.A. 111(17): 6190–6197.

Guerrero, Emma, Stephan Naji, and Jean-Pierre Bocquet-Appel. 2008. The signal of the Neolithic demographic transition in the Levant. In Jean-Pierre Bocquet-Appel and Ofer Bar-Yosef (eds.), The Neolithic Demographic Transition and Its Consequences, 57–80. Dordrecht: Springer.

Harris, David and Dorian Q Fuller. 2013. Agriculture: Definition and overview. In Claire Smith (ed.), Encyclopedia of Global Archaeology, 104–113. New York: Springer.

Hedges, Richard E.M., Rupert A. Housley, Christopher Bronk Ramsey, and Gert Jaap van Klinken. 1992. Radiocarbon dates from the Oxford AMS system: Archaeometry datelist 14. Radiocarbon 34(1): 141–159.

Higham, Charles. 2014. Early Mainland Southeast Asia: From First Humans to Angkor. River Books: Bangkok.

Hsieh, Jaw-shu, Yue-ie Caroline Hsing, Tze-fu Hsu, Paul Jen-kuei Li, Kuang-ti Li, and Cheng-hwa Tsang. 2011. Studies on ancient rice—Where botanists, agronomists, archeologists, linguists, and ethnologists meet. Rice 4: 178–183.

Hosoya, Leo Aoi, Michèle Wollstonecroft, Dorian Q Fuller, and Qin Ling. 2010. Experimental pilot study of peach/apricot kernel detoxification: For reconstruction of Chinese early rice farmers broad spectrum subsistence strategy. In Keisuke Makibayashi and Megumi Uchikado (eds.), Studies of Landscape History of East Asian Inland Seas, 457–464. Kyoto: Research Institute for Humanity and Nature.

Hu, Yaowu, Shougong Wang, Fengshi Luan, Changsui Wang, and Michael P. Richards. 2008. Stable isotope analysis of humans from Xiaojingshan site: Implications for understanding the origin of millet agriculture in China. Journal of Archaeological Science 35: 2960–2965.

Jiang, Leping. 2007. Zhejiang Pujiangxian Shangshan yizhi fajue jianbao. Kaogu 9: 7–19. [in Chinese]

Jiao, Tianlong. 2007. The Neolithic of Southeast China Cultural Transformation and Regional Interaction on the Coast. Youngstown: Cambria Press.

Jin, Guiyun, Wenwan Wu, Kesi Zhang, Zebing Wang, and Xiaohong Wu. 2014. 8000-year old rice remains from the north edge of the Shandong Highlands, East China. Journal of Archaeological Science 51: 34–42.

Jin, Hetian, Xu Liu, Rui Min, Xiaorui Li, and Xiaohong Wu. 2014. Early subsistence practices at prehistoric Dadunzi in Yuanmou, Yunnan: New evidence for the origins of early agriculture in southwest China. In Anke Hein (ed.), The ‘Crescent-shaped Cultural-communication Belt’: Tong Enzheng’s Model in Retrospect: An Examination of Methodological, Theoretical and Material Concerns of Long Distance Interactions in East Asia, BAR International Series 2679, 133–140. Oxford: Archaeopress.

Kim, Minkoo. 2014. Rice, ethnicity and the collation of archaeobotanical data for prehistoric Korea. Archaeologies 10(2): 168–193.

Lanehart, Rheta E., Robert H. Tykot, Anne P. Underhill, Fenghui Luan, Haiguang Yu, Hui Fang, Cai Fengshu, Gary Feinman, and Linda Nicholas. 2011. Dietary adaptation during the Longshan period in China: Stable isotope analyses at Liangchengzhen (southeastern Shandong). Journal of Archaeological Science 38(9): 2171–2181.

Larson, Greger, Ranran Liu, Xingbo Zhao, Jing Yuan, Dorian Fuller, Loukas Barton, Keith Dobney, Qipeng Fan, Zhiliang Gu, Xiao-Hui Liu, Yunbing Luo, Peng Lv, Leif Andersson, and Ning Li. 2010. Patterns of East Asian pig domestication, migration, and turnover revealed by modern and ancient DNA. Proceedings of the National Academy of Sciences of the U.S.A. 107(17): 7686–7691.

Lee, Gyoung-Ah. 2011. The transition from foraging to farming in prehistoric Korea. Current Anthropology 52(S4): S307–S329.

Lee, Gyoung-Ah, Gary W. Crawford, Li Liu, and Xingcan Chen. 2007. Plants and people from the Early Neolithic to Shang periods in North China. Proceedings of the National Academy of Sciences of the U.S.A. 104(3): 1087–1092.

Lee, Gyoung-Ah, Gary W. Crawford, Li Liu, and Xingcan Chen. 2011. Archaeological soybean (Glycine max) in East Asia: Does size matter? PLoS ONE 6(11): e26720. doi: 10.1371/journal.pone.0026720.

Li, Xiaoqiang, John Dodson, Jie Zhou, and Xinying Zhou. 2009. Increases of population and expansion of rice agriculture in Asia, and anthropogenic methane emissions since 5000 BP. Quaternary International 202(s1–2): 41–50.

Liu, Li. 2004. The Chinese Neolithic. Trajectories to Early States. Cambridge: Cambridge University Press.

Liu, Li and Xingcan Chen. 2012. The Archaeology of China. Cambridge: Cambridge University Press.

Liu, Li, Gyoung-Ah Lee, Leping Jiang, and Juzhong Zhang. 2007. The earliest rice domestication in China. Antiquity 81(313). Available online at http://www.antiquity.ac.uk/projgall/liu1/ (accessed June 26, 2017).

Lu, Houyuan, Jianping Zhang, Kam-biu Liu, Naiqin Wu, Yumei Li, Kunshu Zhou, Maolin Ye, Tianyu Zhang, Haijiang Zhang, Xiaoyan Yang, Licheng Shen, Deke Xu, and Quan Li. 2009. Earliest domestication of common millet (Panicum miliaceum) in East Asia extended to 10,000 years ago. Proceedings of the Natural Academy of Sciences of the U.S.A. 106(18): 7367–7372.

Lu, Tracey Lie Dan. 1999. The Transition from Foraging to Farming and the Origin of Agriculture in China. BAR International Series 774. Oxford: British Archaeological Reports.

Maeda, Osamu, Leilani Lucas, Fabio Silva, Ken-Ichi Tanno, and Dorian Q Fuller. 2016. Narrowing the harvest: Increasing sickle investment and the rise of domesticated cereal agriculture in the Fertile Crescent. Quaternary Science Reviews 145: 226–237.

Makibayashi, Keisuke. 2014. The transformation of farming cultural landscapes in the Neolithic Yangtze area, China. Journal of World Prehistory 27(3–4): 295–307.

Millard, Andrew R. 2014. Conventions for reporting radiocarbon determinations. Radiocarbon 56(2): 555–559.

Miyamoto, Kazuo. 2009. Noukounokigenwosaguru [In search of the origin of agriculture]. Tokyo: Yoshikawa kōbunkan. [in Japanese]

Miyata, Yoshiki, Takanobu Kojima, Akiko Matsutani, Shin Onbe, and Toyohiro Nishimoto. 2007. Oldest common millet from western Japan: Carbonized material adhering to pottery from the Ryugasaki A site, Shiga prefecture. In Mineo Imamura and Kenichi Kobayashi (eds.), Bulletin of the National Museum of Japanese History 137, 255–265. Chiba: National Museum of Japanese History. [in Japanese with English abstract]

Nanjing Museum and Jiangsu Province Institute. 2016. Shunshanji. Beijing: Science Press. [in Chinese]

Nasu, Hiroo, Haibin Gu, Arata Momohara, and Yoshinori Yasuda. 2012. Land-use change for rice and foxtail millet cultivation in the Chengtoushan site, central China, reconstructed from weed seed assemblages. Archaeological and Anthropological Sciences 4(1): 1–14.

Pechenkina, Ekaterina A., Stanley H. Ambrose, Ma Xiaolin, and Robert A. Benfer Jr. 2005. Reconstructing Chinese Neolithic subsistence practices by isotopic analysis. Journal of Archaeological Science 32: 1176–1189.

Pei, Anping. 2013. The Pengtoushan culture in the Middle Yangzi River Valley. In Anne P. Underhill (ed.), A Companion to Chinese Archaeology, 497–509. Oxford: Blackwell.

Qin, Ling and Dorian Q Fuller. 2009. Discussion on plant remains of the early and middle Yangshao Period from Nanjiaokou Site, Henan Province and related topics. In Henan Provincial Institute of Archaeology and Cultural Relics (ed.), Sanmenxia Nanjiaokou, 427–435. Beijing: Science Press. [in Chinese]

Ren, Guoyu. 2007. Changes in forest cover during the Holocene. Vegetation History and Archaeobotany 16: 119–126.

Ren, Shinan. 1996. Several major achievements in Early Neolithic China, ca. 5000 BC. Translation by Wing Kam Cheung. Kaogu 1996: 37–49.

Richardson, Peter J., Robert Boyd, and Robert Bettinger. 2001. Was agriculture impossible during the Pleistocene but mandatory during the Holocene? A climate change hypothesis. American Antiquity 66(3): 387–411.

Robbeets, Martine I. 2015. Proto-Transeurasian: Where and when? Man in India 95(3–4): 921–946.

Robbeets, Martine I. 2017. Austronesian influence and Transeurasian ancestry in Japanese: A case of farming/language dispersal? Language Dynamics and Change 7(2): 210–251.

Sagart, Laurent. 2008. The expansion of Setaria farmers in East Asia: A linguistic and archaeological model. In Alicia Sanchez-Mazas, Roger Blench, Malcolm Ross, Ilia Peiros, and Marie Lin (eds.), Past Human Migrations in East Asia: Matching Archaeology, Linguistics and Genetics, 133–157. London: Routledge.

Sagart, Laurent. 2011. How many independent rice vocabularies in Asia? Rice 4(3–4): 121–133.

Sergusheva, Elena A. and Yury E. Vostretsov. 2009. The advance of agriculture in the coastal zone of East Asia. In Andrew S. Fairbairn and Ehud Weiss (eds.), From Foraging to Farming: Papers in Honour of Gordon C. Hillman, 205–219. Oxford: Oxbow Books.

Shelach, Gideon and Mingyu Teng. 2013. Earlier Neolithic economic and social systems of the Liao River region, northeast China. In Anne P. Underhill (ed.), A Companion to Chinese Archaeology, 37–54. Oxford: Blackwell.

Silva, Fabio, Chris Stevens, Alison Weisskopf, Cristina Castillo, Ling Qin, Andrew Bevan, and Dorian Q Fuller. 2015. Modelling the origin of rice farming in Asia using the Rice Archaeological Database. PloS ONE 10(9): e0137024. doi: 10.1371/journal.pone.0137024.

Smith, Bruce D. 2001. Low-level food production. Journal of Archaeological Research 9(1): 1–4.

Smith, Bruce D. 2015. A comparison of niche construction theory and diet breadth models as explanatory frameworks for the initial domestication of plants and animals. Journal of Archaeological Research 23(3): 215–262.

Song, Jixiang. 2011. The Agricultural Economy during Longshan Period: An Archaeobotanical Perspective from Shandong and Shanxi. PhD dissertation, University College London.

Spengler, Robert, Michael Frachetti, Paula Doumani, Lynne Rouse, Barbara Cerasetti, Elissa Bullion, and Alexei Mar’yashev. 2014. Early agriculture and crop transmission among Bronze Age mobile pastoralists of Central Eurasia. Proceedings of the Royal Society B 281: 20133382.

Stevens, Chris J., Charlene Murphy, Rebecca Roberts, Leilani Lucas, Fabio Silva, and Dorian Q Fuller. 2016. Between China and South Asia: A Middle Asian corridor of crop dispersal and agricultural innovation in the Bronze Age. The Holocene 26(10): 1541–1555. doi: 10.1177/0959683616650268.

Stuiver, Minze. 1982. A high-precision calibration of the AD radiocarbon time scale. Radiocarbon 24(1): 1–26.

Stuiver, Minze and Henry A. Polach. 1977. Reporting of ¹⁴C data. Radiocarbon 19(3): 355–363.

Tsang, Cheng-hwa. 2005. Recent discoveries at a Tapenkeng Culture site in Taiwan: Implications for the problems of Austronesian Origins. In Laurent Sagart, Roger Blench, and Alicia Sanchez-Mazas (eds.), The Peopling of East Asia, 63–73. London: Routledge.

Underhill, Anne P. 2013. Introduction: Investigating the development and nature of complex societies in Ancient China. In Anne P. Underhill (ed.), A Companion to Chinese Archaeology, 1–12. Oxford: Blackwell.

Van Driem, George. 2012. The ethnolinguistic identity of the domesticators of Asian rice. Comptes Rendus Palevol 11(2): 117–132.

Wagner, Mayke, Pavel Tarasov, Dominic Hosner, Andreas Fleck, Richard Ehrich, Xiaocheng Chen, and Christian Leipe. 2013. Mapping of the spatial and temporal distribution of archaeological sites of northern China during the Neolithic and Bronze Age. Quaternary International 290–291: 344–357.

Weber, Steve, Heather Lehman, Timothy Barela, Sean Hawks, and David Harriman. 2010. Rice or millets: Early farming strategies in central Thailand. Archaeological and Anthropological Sciences 2(2): 79–88.

Weisskopf, Alison Ruth. 2014. Millets, Rice and Farmers: Phytoliths as Indicators of Agricultural, Social and Ecological Change in Neolithic and Bronze Age Central China. BAR International Series 2589. Oxford: Archaeopress.

Weisskopf, Alison and Dorian Q Fuller. 2014. Peach: Origins and development. In Claire Smith (ed.), Encyclopedia of Global Archaeology, 5840–5842. New York: Springer.

Weisskopf, Alison, Ling Qin, Jinlong Ding, Ping Ding, Guoping Sun, and Dorian Q Fuller. 2015. Phytoliths and rice: From wet to dry and back again in the Neolithic Lower Yangtze. Antiquity 89(347): 1051–1063.

Wu, Wenwan, Xinghua Wang, Xiaohong Wu, Guiyun Jin, and Pavel E. Tarasov. 2014. The early Holocene archaeobotanical record from the Zhangmatun site situated at the northern edge of the Shandong Highlands, China. Quaternary International 348: 183–193.

Wu, Yan, Leping Jiang, Yunfei Zheng, Changsui Wang, and Zhijun Zhao. 2014. Morphological trend analysis of rice phytolith during the early Neolithic in the Lower Yangtze. Journal of Archaeological Science 49: 326–331.

Xie, Liye, Steven L. Kuhn, Guoping Sun, John Olsen, W. Yunfei Zheng, Pin Ding, and Ye Zhao. 2015. Labor costs for prehistoric earthwork construction: Experimental and archaeological insights from the Lower Yangzi basin, China. American Antiquity 80(1): 67–88.

Xue, Yining. 2010. A Preliminary Study of Plant Remains from Haimenkou, Yunnan, Jianchuan. Master’s thesis, Peking University.

Yang, Xiaoyan, Huw J. Barton, Zhiwei Wan, Quan Li, Zhikun Ma, Mingqi Li, Dan Zhang, and Jun Wei. 2013. Sago-type palms were an important plant food prior to rice in southern subtropical China. PLoS ONE 8(5): e63148. doi: 10.1371/journal.pone.0063148.

Yang, Xiaoyan, Zhikun Ma, Jun Li, Jincheng Yu, Chris Stevens, and Yijie Zhuang. 2015. Comparing subsistence strategies in different landscapes of North China 10,000 years ago. The Holocene 25(12): 1957–1964.

Yang, Xiaoyan, Zhiwei Wan, Linda Perry, Houyuan Lu, Qiang Wang, Chaohong Zhao, Jun Li, Fei Xie, Jincheng Yu, Tianxing Cui, Tao Wang, Mingqi Li, and Quansheng Ge. 2012a. Early millet use in northern China. Proceedings of the National Academy of Sciences of the U.S.A. 109(10): 3726–3730.

Yang, Xiaoyan, Jianping Zhang, Linda Perry, Zhikun Ma, Zhiwei Wan, Mingqi Li, Xianmin Diao, and Houyuan Lu. 2012b. From the modern to the archaeological: Starch grains from millets and their wild relatives in China. Journal of Archaeological Science 39(2): 247–254.

Yuan, Jing and Rowan K. Flad. 2002. Pig domestication in ancient China. Antiquity 76(293): 724–732.

Yuan, Jing, Liang Zhonghe, Wu Yun, and Jia Xiaobing. 2002. Shell mounds in the Jiaodong peninsula: A study in environmental archaeology. Journal of East Asian Archaeology 4: 1–4.

Zeder, Melinda A. 2015. Core questions in domestication research. Proceedings of the National Academy of Sciences of the U.S.A. 112(11): 3191–3198.

Zhang, Chi and Hsiao-chun Hung. 2010. The emergence of agriculture in southern China. Antiquity 84: 11–25.

Zhang, Chi and Hsiao-chun Hung. 2013. Jiahu 1: Earliest farmers beyond the Yangtze River. Antiquity 87: 46–63.

Zhang, Juzhong and Qikun Ren. 2005. Past and future prospects in Anhui prehistoric rice agricultural research. Translation by Elaine Wong. Agricultural Archaeology 2005: 156–160.

Zhang, Xun, J. Wang, Z. Xi, and Sui Qiu. 2003. Studies on ancient human diets. Kaogu 2: 62–75. [in Chinese]

Zhang, Ying. 2014. Animal Procurement Strategies in Neolithic of the Yangtze River Basin: Zooarchaeological and Ecological Approaches. PhD dissertation, University College London.

Zhao, Zhijun. 2004. Cong Xinglonggou yizhi fuxuan jieguo tan Zhongguo beifang hanzou nongye qiyuan wenti (Addressing the origins of agriculture in north China based on the results of flotation from the Xinglonggou site). Dongya Kaogu 12: 188–199. [in Chinese]

Zhao, Zhijun. 2010a. New data and new issues for the study of origin of rice agriculture in China. Archaeological and Anthropological Sciences 2(2): 99–105.

Zhao, Zhijun. 2010b. Report on the archaeobotanical remains unearthed at the site of Shifodong. In Yunnan Sheng Kaogusuo and Chengdu Shi Wenwu Kaogu Yanjiusuo (eds.), Shifodong in Gengma County, 368–373. Beijing: Wenwu Chubanshe. [in Chinese]

Zhao, Zhijun. 2011. New archaeobotanical data for the study of the origins of agriculture in China. Current Anthropology 52(S4): S295–S306.

Zhao, Zhijun. 2015. Archaeobotanical evidence for the transition to millet farming in North China. Paper presented at the conference “Dialogue of Civilizations: Comparing Multiple Centres of Early Civilizations of the World,” Beijing, April 2015.

Zhao, Zijiun and Juzhong Zhang. 2009. The report of flotation work at the Jiahu site. Kaogu 8: 86–95. [in Chinese]

Zheng, Yunfei, Gary W. Crawford, and Xugao Chen. 2014. Archaeological evidence for peach (Prunus persica) cultivation and domestication in China. PLoS ONE 9(9): e106595. doi: 10.1371/journal.pone.0106595.

Zhu, Guangyao, Cheng Zhu, Guang Yue Shi, and Zhibin Sun. 2008. Relations between man and nature reflected by the evolution of the Neolithic production tools in the Three Gorges region of Yangtze River, China. Chinese Science Bulletin 53(S1): 95–106.

Zhu, Yangping. 2013. The early Neolithic in the Central Yellow River valley, c. 7000–4000 BC. In Anne P. Underhill (ed.), A Companion to Chinese Archaeology, 171–193. Oxford: Blackwell.

ZPIACR (Zhejiang Provincial Institute of Cultural Relics and Archaeology). 2004. Kuahuqiao. Beijing: Cultural Relics Publishing House. [in Chinese]