Unlike the Middle East, where sheep and goats were the earliest domesticates and “pigs have been cast in the role of a late and rather unimportant addition to the repertoire of animal domesticates” (Redding and Rosenberg 1998:65), in many areas of China, pigs were most likely the earliest domesticate – with the possible exception of dogs – and suids have been and continue to be the most important domestic animal. The study of the process of pig domestication is therefore an important research topic in China. This paper discusses the development of pig domestication in ancient China through a study of pig remains found in some Neolithic sites based on established methods of analysis - measuring teeth and bones of pigs, identifying the age structure of pig populations, calculating the ratio of suidae in faunal assemblages, and examining the archaeological background.
The beginning of pig domestication
To date, there are four archaeological assemblages dated to approximately 10,000 BP in China: the Yucanyan site in Dao County, Hunan Province; The Xianrendong and Diaotonghuan sites in Wannian County, Jiangxi Province; and the Nanzhuangtou site in Xushui County, Hebei Province. Phytoliths of domesticated rice, pottery, and stone and bone implements have been found in both the Yucanyan site located at latitude 25.5° N, and the Diaotonghuan site, located at latitude 28.5° N (Yan 1997). Pottery, bone and stone implements have also been discovered at the Nanzhuangtou site, which is located at 39° N latitude (The Baoding Institute of antiquity Management et al. 1992). Based on these data, the beginning of agriculture and pottery manufacturing can be traced back to approximately 10,000 years ago in China. However, it is important to note that the faunal remains found in these assemblages seem to all represent wild species, and as yet there is no evidence for the existence of domesticated animals (Huang 1966; Yan 1997; Yuan forthcoming).
Some scholars believe that the earliest domesticated pig was found at the Zengpiyan site, in Guilin City, Guangxi Province (Nelson 1998). But the data from the Zenpingyan site are very problematic. First, the date of the site still remains an issue. The site had been dated to about 5000 BP according to the preliminary report, and the cultural assemblage was classified as the early phase of the Late Neolithic culture (The Antiquity Team of Guangxi National Autonomous Region et al, 1976). Later, the site was dated to 10,000-7000 BP based on radiocarbon dates. However, these radiocarbon dates are also suspect (14C Laboratory, Department of history, Beijing University et al, 1982). Second, the location from which the pig bones were unearthed is unclear. The investigators of these materials reported that all pig bones came from the excavation in 1973 (Li et al, 1978). While some scholars suggest that all faunal remains from the 1973 excavation belong to the lower deposits in the site, dated to about 9000 BP (Tan 1990), others believe that the excavation in 1973 only reached the upper deposits at the site, dated to about 7000 BP (Qi 1991). Third, the identification of the pigs as domesticated is questionable. The reason that the pigs were identified as domesticated was merely based on the age profile - 65% of individual pigs were only 1-2 years old. However, the study didn’t include the measurements of length and width of teeth, the relationship between the age and condition of tooth eruption, etc. On the other hand, since most sites with clearly domesticated pigs contain an assemblage that mostly consists of one year old pigs, Li et al. argue that the dominance of 1-2 year old pigs indicates a low level of early domestication, i.e., the age of pigs was older in the earlier period of domestication (Li et al, 1978). Altogether this evidence is not quite strong enough to indicate the domestication of pigs. The studies of other materials from the site indicate that subsistence at Zengpiyan was based on hunting and gathering. It is questionable whether there could be some domesticated pigs in such economic system. The data from the Zengpiyan site are not conclusive enough to be used for the study of the origin of pig domestication in China.
We believe the earliest domesticated pigs found in China to date were discovered in the Cishan assemblage dated to approximately 8,000 BP. This assemblage is located at latitude 36.5° N, in Wu’an County, Hebei Province. The identification of domesticated pigs is based upon three criteria: tooth measurement, the slaughter age of individuals and the archaeological contexts from which the animals were found. At Cishan site, the average length of the pig’s lower third molar is 41.4 mm; the average width of the lower third molar is 18.3 mm. These measurements are similar to those of domesticated pigs for which the average length is about 39 mm and the average width 18 mm. In addition, a considerable majority (over 60%) of the pigs were killed at the age of 0.5 to 1 years, which seems to indicate intentional population control linked to incipient domestication rather than hunting. What is more, complete pig skeletons, which were identified to about one-year old, have been found buried in pits, in some cases beneath large amounts of carbonized millet (Zhou 1981).
Based on the above, we argue that the occurrence of domesticated pigs in Neolithic China was at least 2000 years later than that of agriculture and pottery manufacturing.
2. Preconditions for the origin of pig domestication.
We believe that the origin of pig domestication in Neolithic China could only occur under four conditions. The first condition is that the meat gathered by traditional hunting could not meet the demand, thus new approaches for acquiring meat resources had to be exploited. The second prerequisite is that there must have been a certain quantity of wild pigs and new born baby pigs living near human settlements making it possible to catch this species for husbandry. Third is the successful cultivation of certain cereals from planting to harvesting; such successful cultivation not only gave people more confidence in plant domestication, but also encouraged them to domesticate certain animal species. The last condition is that the output from cereal farming produced surplus, thus allowing for feeding of animals with the by-product of cereals.
Two important archaeological discoveries in China can be used to assess these preconditions in the Chinese domestication of pigs. The first is from the Cishan site dated to circa 8,000 BP, where more than 300 earth-wall pits of rectangular shape were discovered. Remains of cultivated foxtail millet were recovered from 80 of these pits. Based upon the volume of the carbonized foxtail millet remains, it is estimated that the quantity of foxtail millet would have been over 50,000 kg when it was stored. In addition, skeletons of pigs and dogs were found beneath the millet remains in some of these pits (The Antiquity Management Department of Hebei Province et al. 1981).
The second important discovery is based on the dietary analysis conducted on human and pig bones found in the Taoshi assemblage in Xiangfen City, Shanxi Province. The Taoshi assemblage is dated to circa 2000 BC, and also includes remains of foxtail millet (Cai et al. 1984). The carbon isotopes in these bones reveal that both humans and pigs consumed a large quantity of C4 plants (Cai et al. 1984). Since foxtail millet is a C4 plant, it can be inferred that both humans and pigs in the Taoshi assemblage consumed foxtail millet; or, to be more precise, human beings consumed foxtail millet, and the pigs probably were fed with the chaff of foxtail millet.
In regard to the subsistence strategies of Neolithic human beings in given environments with limited natural resources, We suggest that the economic development of such resources was often passive. In the pig domestication example, the four preconditions listed above seem to apply. First, as we have discussed elsewhere, the meat gathered by hunting did not meet demand and encouraged pig husbandry (Yuan 1999). In terms of the second condition, remains of wild boar have been found in both Paleolithic and Neolithic sites in China, so it can be inferred that this species was widely found in many areas. It is possible that humans may initially have attracted wild boars by accidentally providing food for them. This is probably the most crucial step in the successful domestication of pig. Third, animal domestication postdates farming in China demonstrating the preexisting familiarity of Neolithic period humans with domesticated products. The existence of plant domesticates allows for the fourth pre-condition, that there be a cereal surplus with which to feed domesticated animals. Such surplus was not possible at the initial stage of farming, and only feasible when farming had reached quite advanced level. Lastly then, the large quantity of foxtail millet remains found in Cishan circa 8,000 BP, clearly demonstrates that the output of cereal farming had reached the level necessary for cereals to have been used to feed animals in addition humans. Additionally, the isotopic analysis of the pig remains from the later Taoshi site indicates that pigs were in fact fed on millet, as expected.
3. A model for the development of animal domestication.
The process of emergent domestication has been discussed extensively in archaeological literature with much of this discussion focusing on the problem of defining domestication and identifying it in the archaeological record (Bökönyi 1989; Clutton-Brock 1989; Crabtree 1993; Davis 1987; Ducos 1978, 1989; Hecker 1982; Meadow 1989). These definitions of domestication tend to emphasize the control of animals’ breeding conditions (Bökönyi 1989:22), or the ownership of animals (Clutton-Brock 1989:7; Ducos 1978:54). But as Hongo and Meadow demonstrate in their analysis of suid remains from Çayönü Tepesi in Southeastern Anatolia, the emergence of domestication is a gradual process involving increased interdependence over time between human and animal populations (1998:87-89).
The process of emergent animal domestication has been characterized as a three-stage development in a number of different ways. Zvelebil and Rowley-Conwey (1984) propose a model which involves an ‘availability phase’, followed by a ‘substitution phase’, and finally a ‘consolidation phase’ (also see Zvelebil 1986). This model, developed to explain the secondary emergence of domestic agriculture in Northwest Europe, is based on the preexistence of agriculture in other areas and not useful for elucidating the primary domestication of pigs in the Chinese context. Hongo and Meadow (1998) suggest a developmental sequence from ‘hunting’ to ‘culture control’ to ‘domestication’ based on Hecker’s (1982) proposal that there are enigmatic situations – situations of cultural control - in which humans control or manipulate animal populations without producing the ‘hard’ evidence of animal domestication. This model is essentially similar to one proposed by Yuan (1999) which outlines a development in human-animal relationships from a ‘dependence type’, to an ‘initial exploitation type’, culminating in an ‘exploitation type’. While Hongo and Meadow (1998:89) propose that initial ‘cultural control’ may involved penned females interbreeding with feral males, we see no reason why this should necessarily be the case and therefore use Yuan’s terminology in this discussion (see Rappaport 1967; Redding and Rosenberg 1998; and Hayashida 1971 for the ethnographic basis of Hongo and Meadow’s proposal).
Based on a quantitative analysis of animal remains found in archaeological sites from Yellow River region in China (the majority of these sites date to the Neolithic period), we discover that ancient people exploited different types of meat sources. Table 1 lists the sites used in this brief discussion and characterizes the assemblage as one which indicates a ‘dependence type’, an ‘initial exploitation type’, or an ‘exploitation type’ meat acquisition pattern. The locations of these sites, and of others mentioned in this article, are shown on Figure 1.
First of all, when people initially became sedentary, their hunting and gathering activities often took place in surrounding districts. Naturally, people would have hunted and gathered faunal species that were available in the vicinity of their settlement. Although the hunting and gathering activities were the intentional subsistence strategies used by human beings, these activities relied upon the natural resources in the vicinity of the settled spot, thus we name this type of acquiring meat resources as the ‘dependence type’. We define ‘dependence’ as the hunting and trapping of the required supply animal species completely within the natural environment of human' settlement. The site of Nanzhuangtou in Hebei, designated by a triangle on table 1, exhibits the characteristics of ‘dependence type’ subsistence. The faunal assemblage includes high species diversity and an absence of any clear evidence for animal domestication (Yuan forthcoming).
A second type of meat acquisition was the combination of hunting and trapping wild animals and the initial cultural control and exploitation of certain domesticated species. Animal husbandry involves human beings interfering in the life cycle of certain animal species according to their own wishes. We name this exploitation, which increases humans' ability to survive resource instability, the 'initial exploitation type'. It is defined by the prevailing of hunting and fishing activities associated with initial animal husbandry. Seven sites in table 1 have collections that represent ‘initial exploitation type’ meat acquisition. These collections – marked by circle symbols on the table - are characterized by the sorts of equivocal archaeological evidence of domestication that prompted Hecker (1982) to initially question the usefulness of the domestication/ non-domestication dichotomy. The site of Cishan, discussed in the first section of this article, should be considered an example of advanced ‘initial exploitation type’ animal use as well.
Finally, a significant change occurred when animal husbandry became dominant, and the importance of hunting and trapping declined. At this stage, the meat supply came mainly from certain domesticated animals, although hunting and trapping might still have been supplementary activities. Generally speaking, animal husbandry requires more long-term management skills and planning than do hunting and trapping activities. Additionally, the dominance of animal domestication in terms of meat acquisition indicates an increasingly stable subsistence base for human groups although, on the contrary, catastrophes that affect domesticate herds may have a greater affect on human subsistence that those that affect communities of wild animals. We name this stage of meat acquisition the 'exploitation type', which is defined by the majority of the meat supply coming from animal husbandry, with more or less supplemental hunting and trapping of wild species that are available in the vicinity of the site. Table 1 lists 12 sites that have collections – indicated by square symbols – associated with ‘exploitation type’ meat acquisition.
The Chinese data support a model of emergent domestication that involves an intermediate stage of intensifying human-animal interaction before humans fully control the breeding conditions of the animal species. The equivocal data from the Cishan site indicate the earliest evidence in China of humans controlling the breeding of pigs at about 8000 BP. By no means, however, were different communities across Northern China necessarily at the same point in the development of animal domestication at the same time. As late as 4000 BP (Jiangzhai phase V) some sites seem to still exhibit the characteristics of ‘initial exploitation type’ meat acquisition while ‘exploitation type’ meat acquisition seems to have emerged as early as 6000 BP (Bancun phase II). The pace of change therefore seems to have varied from centuries to millennia in different parts of Northern China.
The pig was one of the earliest domesticated animals in China, and was the most important species in the subsequent agricultural societies across East Asia. This is a striking feature in ancient China. We have seen that the first evidence of domesticated pigs in Neolithic China dates to about 8000 BP. This domestication occurred long after the first plant domestication, probably as a passive reaction to meat scarcity and in the context of a surplus of agricultural products. The domestication of pigs likely progressed through three phases; ‘dependence type’, ‘initial exploitation type’ and ‘exploitation type’ meat acquisition patterns. By analyzing the quantity and species of animal remains found in archaeological sites, we discover that there was a gradual increase of domesticated animals represented by pig, accompanied by a gradual decrease of wild species. This phenomenon illustrates an increase in subsistence stability during the Chinese Neolithic. This kind of developmental trend is generally similar to other ancient cultures in the world.
We are very grateful to Dr. Tracey Lie-Dan Lu, a visiting scholar fellow the Institute of Archaeology, CASS, for initially translating this article into English. She has done an excellent job by correctly expressing our arguments in another language. We would also like to thank Dr. Mark Hall, the Museum of Niigata, Japan, and Dr. Zhao Zhijun, the Institute of Archaeology, CASS, for their valuable opinions on our paper. We would also like to thank Ms. Zhang Lie, the Institute of Archaeology CASS, for providing the illustrations for our paper.
Bökönyi, S. 1989. Definitions of animal domestication, in J. Clutton-Brock (ed.): 22-26. London: Unwin Hyman.
14C Laboratory, Department of History, Beijing University, et al. 1982. Shihuiyan diqu 14C yangpin niandai de kekaoxing yu Zengpiyan deng yizi de niandai wenti (The reliability of radiocarbon dates of samples collected from limestone regions and the dating of the Zengpiyan and other prehistoric sites). Kaogu Xuebao 1982 (2): 243-250.
Cai, L., et al. 1984. Tanshisan ceding he gudai shipu yanjiu (C13 analysis and human diet in ancient society), Kaogu 1984 (10): 949-955.
Clutton-Brock, J. 1989. The walking larder: Patterns of domestication, pastoralism, and predation. London: Unwin Hyman.
Crabtree, P. J. 1993. Early domestication in the Middle East and Europe, in M. Schiffer (ed.), Archaeological Method and Theory, Vol. 5: 201-246. Tuscon: University of Arizona Press.
Davis, S. 1987. The Archaeology of Animals. New Haven: Yale University Press.
Ducos, P. 1978. ‘Domestication’ defined and methodological approaches to its recognition in faunal assemblages, in R. H. Meadow and M. A. Zeder (eds.), Approaches to faunal analysis in the Middle East: Peabody Museum Bulletin 2: 53. Cambridge, MA:Peabody Museum of Archaeology and Ethnology.
Ducos, P. 1989. Defining domestication: A clarification, in Clutton-Brock, J. (ed.): 28-30.
Hayashida, S. 1971. Inoshishi to buta shoshite Nihon minzoku (Wild pigs, domestic pigs and the Japanese), Kokogaku Journal 52: 11-15.
Hecker, H. M. 1982. Domestication revisited: Its implications for faunal analysis, Journal of Field Archaeology 9: 218-236.
Hongo, H. & R. H. Meadow. 1998. Pig exploitation at Neolithic Çayönü Tepesi (Southeastern Anatolia), in Nelson, S. M. (ed.): 77-88.
Huang, W. 1966. Jiangxi Wannian Xianrendong Quanxinshi Dongxue duiji (The Holocene sediments of the Xianrendong Cave, Wannian, Jiangxi), Gu jizhui dongwu yu gu renlei, 7(3): 226-271.
Li, Y. et al. 1978. Guangxi Guilin Zengpiyan yizhi dongwuqun (Note on the animal fauna of the Zengpiyan site in Guangxi ), Gu jizhui dongwu yu gu renlei, 16(4): 244-254.
Mayer, J. J. et al. 1998. Evaluation of molar size as a basis for distinguishing wild boar from domestic swine: Employing the present to decipher the past, in Nelson, S. M.: 39-54.
Meadow, R. H. 1989 Osteological evidence for the process of animal domestication, in Clutton-Brock, J. (ed.): 80-89.
Nelson, S. M. (ed.) 1998a. Ancestors for the pigs: Pigs in prehistory – MASCA research papers in science and archaeology, Vol. 15. Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology.
Nelson, S. M. 1998b. Pigs in the Hongshan culture, in Nelson, S. M.: 99-107.
Qi, Z. 1991. Tan Guilin Zengpiyan Dongxue yizhi yu Huanan zaoqi xinshiqi wenhua de gege wenti (Discussing the Guilin Zengpiyan Cave Site and various issues concerning the early Neolithic in southern China ), in Fengkai County Museum et. al (eds.), Jinian Huangyandong Yizhi Fajue 30 Zounian Lunwenji (A Collection of Papers from the Symposium on the 30th Anniversary of Scientific Excavations at Huangyandong): 80-85. Guangdong: Guangdong luyou chubanshe.
Rappaport, R. 1967. Pigs for the Ancestors: Ritual in the ecology of a New Guinea people. New Haven: Yale University Press.
Redding, R. W. & M. Rosenberg. 1998. Ancestral pigs: A New (Guinea) model for pig domestication in the Middle East, in Nelson S. M. (ed.): 65-76.
Tan, S. 1990. Cong Guilin Zengpiyan zhugu kan jiazhu de qiyuan (The origins of pig domestication based on pig bones from the Zengpiyan site in Guilin), in Zhang, Z.(ed.), Zengpiyan yizhi yanjiu (The study of Zengpiyan Site): 240-242. Lijiang: Lijiang chubanshe.
The Antiquity Management Department of Hebei Province et al. 1981. Hebei Wu'an , Cishan yizhi (The Cishan site in Wu'an County, Hebei Province), Kaogu Xuebao 1981(3): 303-338.
The Antiquity Team of the Guangxi National Autonomous Region et al. 1976. Guangxi Guilin Zengpiyan Dongxue yizhi (The Guangxi Guilin Zengpiyan Cave site), Kaogu 1976(3): 175-179.
The Baoding Institute of Antiquity Management et al. 1992. Hebei Xushui Nanzhuangtou yizhi shijue jianbao (Preliminary report on the trial excavation at the Nanzhuangtou site, Xushui County, Hebei Province), Kaogu 1992(11): 961-966.
Yan, W. 1997. Woguo daozuo qiyuan yanjiu de xinjinzhan (New progress on studying the origin of rice cultivaiton in China), Kaogu 1997(9): 71-76
Yuan, J. 1999. Lun Zhongguo xinshiqi shidai jumin huoqu roushi ziyuan de fangshi (Man's meat-acquiring patterns in Neolithic China), Kaogu Xuebao 1999(1): 1-22
Yuan, J. 2001. Zhongguo xinshiqi shidai jiachu qiyuan de wenti (The problem of the origin of domestic animals in the Chinese Neolithic), Wenwu 2001(5):51-58
Yuan, J. Forthcoming. Hebei Xushui Nanzhuangtou yizhi dongwu guge yanjiu baogao (Research report on the animal bones from the Nanzhuangtou site, Xushui, Hebei). Unpublished manuscript.
Zhou, B. 1981. Hebei Wu'an Cishan yizhi de dongwu guhai (Animal remains found in the Cishan site, Wu'an County, Hebei Province), Kaogu Xuebao 1981(3): 339-346.
Zvelebil, M. 1986. Mesolithic prelude and Neolithic revolution, in M. Zvelebil (ed.), Hunters in transition: Mesolithic societies of temperate Eurasia and their transition to farming, 5-15. Cambridge: Cambridge University Press.
Zvelebil, M. & P. Rowley-Conwey. 1984. Transition to farming in northern Europe: A hunter-gatherer perspective. Norwegian Archaeological Review 17: 104-128.
The original paper was published in Antiquity，Vol.76, No. 293, pp. 724-732.