Bulls for sacrifice,cows for work? Morphometric models suggest that female cattle were used for traction in the Chinese Bronze Age Late Shang dynasty (ca. 1300

Abstract

Training animals to pull agricultural equipment and wheeled transport significantly shaped and advanced human economic systems. In this context the use of large domestic animals such as cattle was a milestone event in human history, part of what Sherratt memorably termed the Secondary Products Revolution: the use of the products of live animals such as milk, wool and traction power. It is commonly assumed that male cattle were generally preferred for traction because of their greater strength compared with females, and the importance of the latter for breeding and, in some societies, for milk and for dairy products, but surprisingly little is known of this aspect of the Secondary Products Revolution in prehistoric China. Here we apply established morphometric models to 10 assemblages of cattle bones from Chinese Bronze Age (ca. 2000–221 BCE) contexts. Our results indicate a process of intensified cattle labour exploitation at this time and, intriguingly, we also observe the earliest labour employment of female cattle during the Late Shang dynasty (ca. 1300–1046 BCE). It is proposed that female cattle may have been required for traction because of the large numbers of male cattle, especially bulls, that were sent for ritual sacrifice. Such a strategy reveals a sophisticated social management, upon which the Late Shang civilisation eventually developed.

Keywords

Bronze Age cow traction morphometrics North China ritual sacrifice Shang dynasty

Introduction

As a type of highly-valued versatile animal, domestic cattle (Bos taurus) could provide not only primary products such as beef and other carcass products but also the ‘secondary products’ (Sherratt, 1981) of the live animal: milk, traction labour and manure. The earliest domestication of cattle on current evidence dates to around 8000 BCE in the Near East (Helmer et al., 2005; Larson and Fuller, 2014; Larson et al., 2014; Zeder, 2008). The exploitation of cattle strength for pulling heavy equipment such as ploughs and wheeled vehicles was previously thought to have emerged in the fourth millennium BCE in the Near East and Europe (Greenfield, 2010; Sherratt, 1981). Recent studies, however, have suggested that the exploitation of cattle for traction may have been practised by Neolithic early farming communities in the western Balkans in the sixth millennium BCE (Gaastra et al., 2018), and even started as early as 8000 BCE in Syria and Turkey (Helmer and Gourichon, 2008; Helmer et al., 2018).

The training of cattle for draught purpose had profound economic and socio-political impacts on prehistoric societies in the Old World (Bogucki, 1993; Greenfield, 2010; Sherratt, 1983). The employment of the labour of large cattle had the potential to bring greater returns for less effort for these societies than by providing their own labour to carry loads or pull simple tools for tillage. Not only the intensification and expansion of agriculture by greater use of plough technologies, but also the facilitation of long-distance trade communications by cart transport, became practicable in prehistory (Greenfield, 2010). The development of cattle traction in the Old World was one of the main drivers of increasing densities and distributions of population, of increasing social hierarchies and, ultimately, of urbanisation (Bogaard, 2004; Bogucki, 1993; Sherratt, 1981, 1983).

Considering the potential significant contribution of cattle traction to the development of ancient societies, it is crucial to develop systematic and robust criteria for recognising the exploitation of cattle for labour in prehistory (i.e. in periods without written records). Unambiguous artefactual evidence such as models and pictographic evidence showing cattle pulling a plough or cart is very rare (e.g. Falkenstein, 1936; Piggott, 1992: 18; Sherratt, 1981), so identifying traction-related bone pathologies in archaeology (the Pathological Index or PI approach) has emerged as a promising approach for this purpose (Bartosiewicz et al., 1997; De Cupere et al., 2000; Johannsen, 2005; Lin et al., 2018; Thomas, 2008). Based on a control sample of cattle bones of modern individuals with known life and work histories, an effective morphometric model on lower limb bones has been established to distinguish not only between traction and non-traction cattle in archaeological bone samples, but also whether they were males, castrates or females (Lin et al., 2016), allowing us to further explore how ancient societies selected and managed their cattle for traction purposes. Such morphometric models have been successfully applied to reveal cattle traction histories in Neolithic south-eastern Europe (Gaastra et al., 2018), and Mediaeval eastern Africa (Gaastra and Insoll, 2020).

The systematic investigation of cattle traction using the PI approach has only been conducted recently in China; this has demonstrated that cattle were being exploited for their labour in the Bronze Age (Lin et al., 2018). The Chinese Bronze Age (ca. 2000–221 BCE) is well-known for its majestic bronze objects (e.g. ritual vessels, weapons) and delicate jades and other indicators of the development of strict social hierarchies that were developed especially in the Late Shang dynasty (ca. 1300–1046 BCE) (Li and Tang, 2018: 277; Liu and Chen, 2003: 24; Wang, 2014: 3). During the Shang dynasty, wheat from western Eurasia became a valued significant crop, augmenting the existing millet-based cultivation system that had been practised since Neolithic North China (Jones et al., 2011; Lee et al., 2007; Zhao, 2014). The use of animal traction for ploughing has been suspected as a significant driver of the Bronze Age agricultural economy in China, but whether specific sex and/or age groups of cattle were intentionally selected for traction use has not been systematically investigated. The selection and training of a specific cattle group for a specific economic role such as plough and/or cart traction would represent a significant specialisation in animal management, with potential implications for present understanding of the complexity of the Chinese Bronze Age economy more widely. This paper aims to fill this research gap by using the morphometric approach to investigate how cattle were used through the major phases of the Chinese Bronze Age: the Erlitou period ca. 1750–1530 BCE, the Early Shang dynasty ca. 1600–1300 BCE, the Late Shang dynasty ca. 1300–1046 BCE and the Western Zhou dynasty 1046–771 BCE.

Methods and materials

Methods

In the field of zooarchaeology, measurements of specific bone elements to explore domestication processes through size changes in ancient cattle populations were used effectively by for example Fock (1966) and Degerbøl and Fredskild (1970), but the most widely employed system for taking measurements on specific bone elements is that published by Von Den Driesch (1976). Her system has been used successfully to investigate animal domestication, breed difference and sex dimorphism (e.g. Albarella, 1997; Grigson, 1989; Holmes, 2014; Telldahl et al., 2012; Thomas, 1988; Thomas et al., 2013). Few of these studies, however, have used bone measurements to address questions about cattle traction, so Lin et al. (2016) used a control sample of bones from cattle of known histories to investigate whether measurements of their metapodials (metacarpals and metatarsals, respectively the front and rear ‘cannon bones’) could be used to distinguish between traction and non-traction individuals. We used the Von Den Driesch measurement ‘Bd’ (the maximum breadth of the distal epiphysis of cattle metacarpals and metatarsals) against ‘e/D1’ (e = the greatest breadth from medial condyle ridge to medial articulation edge; D1 = minimum depth of the medial condyle).

The control group of animals known to have been used for traction consisted of 65 specimens of modern Romanian Grey and Brown castrates, and Jersey castrates. This was contrasted with 103 specimens of animals from diverse modern meat/dairy breeds not used for traction including Romanian Grey and Brown males, Chillingham males, Scottish Highland males and females, Swedish Mountain males and females, Red Danish males and females, Chinese Yellow females and castrates, Icelandic females, Jersey females, Piebald Jutland females and so forth. Palaeolithic wild aurochs (i.e. pre-domestication animals) were added to the latter group. Considering the genetic continuity of taurine cattle in northern China from the Bronze Age to present day (Cai et al., 2014), it was concluded that modern Chinese Yellow cattle were the most suitable candidates to serve as a reference breed to compare with the Chinese archaeological samples, though only a limited number of them were available for study.

Using the known work histories and age/sex compositions of the control groups, appropriate statistical analyses of the Bd and e/D1 measurements of their metapodials were used to establish the discrete cattle groups shown with different shadings in Figures 2 to 9 (Lin et al., 2016). The archaeological cattle samples deriving from different phases within the Chinese Bronze Age were then plotted against these distributions. The line ‘a’ in these figures marks the lower limit of modern traction cattle and the line ‘b’ the upper limit of modern meat cattle and prehistoric wild cattle.

Materials

The archaeological specimens derive from 10 Chinese Bronze Age sites in the Yellow River region of northern China, commonly regarded as the heartland of successive Chinese civilisations (Figure 1). The middle and lower Yellow River valley in particular witnessed dramatic developments in social complexity, marking the emergence of early states and major later dynasties of northern China (Liu, 2004a: 30). The bones were generally excavated from cultural layers and ash-filled pits. The total sample consisted of 216 specimens: 109 metacarpals and 107 metatarsals (Table 1).

Figure 1.

Location of the 10 Chinese Bronze Age sites from which the material reported here derives: 1. Dinggong; 2. Daxinzhuang; 3. Guanzhuang; 4. Erlitou; 5. Yanshishangcheng; 6. Lijiayao; 7. Donglongshan; 8. Zaoshugounao; 9. Zhougongmiao; 10. Anban.

Table 1.

The number and chronological distribution of cattle metapodials from Chinese Bronze Age sites used in this study.

Period	Site	Metacarpal	Metatarsal
Erlitou period (c.1750–1530 BCE)	Erlitou	26	9
	Dinggong	1	1
	Donglongshan	1	1
Early Shang dynasty (c.1600–1300 BCE)	Daxinzhuang	7	5
Early Shang dynasty (c.1600–1300 BCE)	Yanshishangcheng	14	15
Late Shang dynasty (c.1300–1046 BCE)	Daxinzhuang	2	–
	Zaoshugounao	24	15
	Zhougongmiao	2	17
Shang dynasty* (c.1600–1046 BCE)	Daxinzhuang	9	4
Shang dynasty* (c.1600–1046 BCE)	Dinggong	3	3
Early Western Zhou (c.1046–900 BCE)	Zhougongmiao	9	14
Late Western Zhou (c.900–771 BCE)	Lijiayao	6	17
	Zhougongmiao	2	3
	Zaoshugounao	–	1
	Anban	2	1
Western Zhou* (c.1046–771 BCE)	Guanzhuang	1	–
Western Zhou* (c.1046–771 BCE)	Zaoshugounao	–	1

The general Shang and Western Zhou dynasties are recorded here as there is no further information to clarify whether these samples are associated with the early or late periods.

Results

Morphometric models of cattle metapodials deriving from the Erlitou, Early Shang, Late Shang and Western Zhou periods of the Chinese Bronze Age are displayed and discussed in chronological order below in comparison with the traction and non-traction control samples described earlier. For each period a scatter plot of measurements from the studied metacarpals in illustrated and discussed first and the metatarsals second.

The Erlitou period

For metacarpals (Figure 2), we can see that most of the Erlitou samples fall within the ranges of modern meat females and meat males. There is no sample above the ‘b’ line reaching the absolute traction zone with respect to ‘e/D1’ based on modern morphometric models, but there is one specimen outside the modern female meat range and another one outside the modern male or castrated meat group and within the traction area. The two different types of modern Chinese Yellow meat cattle – females and castrates – help roughly separate the archaeological assemblage on ‘Bd’ at the value around 65 mm. From the perspective of ‘e/D1’, the majority of archaeological elements of this period possess greater values than those of the Chinese Yellow specimens in the control sample.

Figure 2.

Scatterplot of ‘Bd’ against ‘e/D1’ for cattle metacarpals of the Erlitou period, compared with control groups (Lin et al., 2016). Modern Chinese Yellow meat cattle are shown as asterisks (black = female and orange = castrate). The control samples indicate that there are no traction animals below line ‘a’ and no meat cattle or wild cattle above line ‘b’.

Compared with the front lower limb bones, there are fewer metatarsals for the Erlitou Period available for analysis, but while the sample is not closely clustered, no specimen is clearly located above line ‘b’, in the definite traction zone (Figure 3). Most of the Erlitou cattle metatarsals are located between lines ‘a’ and ‘b’, an interval shared by the traction and non-traction groups of the control sample. The two modern female specimens with a ‘Bd’ value lower than 60 mm suggest that archaeological specimens below this value likely derive from female cattle as well.

Figure 3.

Scatterplot of ‘Bd’ against ‘e/D1’ for cattle metatarsals of the Erlitou period, compared with control groups (Lin et al., 2016). Modern Chinese Yellow meat female cattle are shown as black asterisks. The control samples indicate that there are no traction animals below line ‘a’ and no meat cattle or wild cattle above line ‘b’.

The Early Shang dynasty

All the cattle metacarpals of the Early Shang dynasty are within the zones of modern female and male/castrate meat animals (Figure 4). Group separation within the whole assemblage is rather ambiguous and none of these elements expresses a clear traction feature. In terms of ‘Bd’, distal breadth, all of the archaeological specimens are within the range of modern Chinese Yellow meat cows and steers.

Figure 4.

Scatterplot of ‘Bd’ against ‘e/D1’ for cattle metacarpals of the Early Shang dynasty, compared with control groups (Lin et al., 2016). Modern Chinese Yellow meat cattle are shown as asterisks (black = female and orange = castrate). The control samples indicate that there are no traction animals below line ‘a’ and no meat cattle or wild cattle above line ‘b’.

The Early Shang cattle metatarsals in Figure 5 display a different model from that of the metacarpals. Two groups can be easily seen from the scatterplot – three specimens in the top half of the traction zone, and the remainder sitting in the meat cattle range. Both the front and hind legs share a similar sample size for analysis, but it is surprising that none of the metatarsals of this period are located below line ‘a’. In addition, most of the archaeological assemblage show values of ‘Bd’ greater than 60 mm, where modern Chinese Yellow cows lie.

Figure 5.

Scatterplot of ‘Bd’ against ‘e/D1’ for cattle metatarsals of the Early Shang dynasty, compared with control groups (Lin et al., 2016). Modern Chinese Yellow meat female cattle are shown as black asterisks. The control samples indicate that there are no traction animals below line ‘a’ and no meat cattle or wild cattle above line ‘b’.

The Late Shang dynasty

Specimens from the general Shang dynasty are presented here as well as those from the Late Shang period. As illustrated in Figure 6, diverse sub-clusters, indicated by the morphometric analysis on metacarpals, can be observed for the Late Shang period as well as for the general Shang dynasty. Even though the bulk of archaeological samples lies within the boundaries of modern non-draught breeds, many are located outside these zones, extending upwards with regard to the ‘e/D1’ measurement. While many are located beneath line ‘a’, there are two specimens in the area above line ‘b’. Considerable numbers of specimens are sitting between the two lines, where specimens from traction animals may overlap with the non-traction groups. A large number of samples would be identified as females if we followed the separation rule between modern Chinese Yellow meat cows and steers, that is, the value about 65 mm on ‘Bd’.

Figure 6.

Scatterplot of ‘Bd’ against ‘e/D1’ for cattle metacarpals of the Late Shang and general Shang dynasty, compared with control groups (Lin et al., 2016). Modern Chinese Yellow meat cattle are shown as asterisks (black = female and orange = castrate). The control samples indicate that there are no traction animals below line ‘a’ and no meat cattle or wild cattle above line ‘b’.

The Late Shang metatarsals are illustrated in Figure 7. Many specimens are in the ranges of modern non-draught females and males/castrates, but even more are situated outside, between lines ‘a’ and ‘b’. While two metatarsals are below line ‘a’, there are three specimens above line ‘b’. A striking number of specimens have values below 60 mm on ‘Bd’, where the modern Chinese Yellow female cattle are located.

Figure 7.

Scatterplot of ‘Bd’ against ‘e/D1’ for cattle metatarsals of the Late Shang and general Shang dynasty, compared with control groups (Lin et al., 2016). Modern Chinese Yellow meat female cattle are shown as black asterisks. The control samples indicate that there are no traction animals below line ‘a’ and no meat cattle or wild cattle above line ‘b’.

The Western Zhou dynasty

Morphometric results on the sample of cattle metacarpals from the general Western Zhou, together with those can be assigned more specifically to either the Early or Late Western Zhou sub-periods, are presented in Figure 8. Most of the specimens do not have values exceeding those of modern meat cattle and the identified patterns between the Early and Late Western Zhou show no significant difference. While some specimens are below line ‘a’, one individual is located above line ‘b’. The others are all clustered between the two lines. The value of 65 mm on ‘Bd’ also helps divide these archaeological specimens into two groups.

Figure 8.

Scatterplot of ‘Bd’ against ‘e/D1’ for cattle metacarpals of the Western Zhou (together with available Early and Late Western Zhou periods), compared with control groups (Lin et al., 2016). Modern Chinese Yellow meat cattle are shown as asterisks (black = female and orange = castrate). The control samples indicate that there are no traction animals below line ‘a’ and no meat cattle or wild cattle above line ‘b’.

In the case of the metatarsals of this phase, a rather large number of specimens are located outside the modern non-traction zones (Figure 9). While most of the assemblage lies between lines ‘a’ and ‘b’, there are a few individuals above the ‘b’ line. The mixed locations of the specimens suggest that there is no clear difference between the Early and Late Western Zhou.

Figure 9.

Scatterplot of ‘Bd’ against ‘e/D1’ for metatarsals of the Western Zhou (together with available Early and Late Western Zhou periods), compared with control groups (Lin et al., 2016). Modern Chinese Yellow meat female cattle are shown as black asterisks. The control samples indicate that there are no traction animals below line ‘a’ and no meat cattle or wild cattle above line ‘b’.

Discussion and conclusion

Cattle traction in the Chinese Bronze Age

The morphometric models produced by control samples of metapodials (metacarpals and metatarsals) of modern domestic and ancient wild cattle (Lin et al., 2016) indicate that the ‘e/D1’ index on the Y-axis functions well in terms of recognising animals used for traction, while the ‘Bd’ values on the X-axis mainly differentiate diverse sex groups (males, castrates, females). It should be noted, however, that the ‘Bd’ values could also be influenced by draught activity, as traction could, to some extent, broaden the distal breadths of metapodials as a result of the pressure on the lower limbs from pulling a plough through soil. This proviso has to be borne in mind.

According to Matolcsi (1970), complete metapodials can serve as a good indicator to estimate an animal’s withers height. For the Chinese Bronze Age, there is no significant difference for the greatest length of available complete metacarpals among sub-periods (One-way ANOVA, p = 0.824; Erlitou: n = 3, mean = 205.6 mm; Shang: n = 17, mean = 206.1 mm; and Western Zhou: n = 5, mean = 203.0 mm). This may indicate that the Chinese Bronze Age adult cattle were broadly of similar size, giving confidence in the use of morphometric models to distinguish between traction/non-traction and male/castrate/female groupings within the overall population through the different periods of interest.

In the Erlitou period, none of the assemblage, neither metacarpals nor metatarsals, displays a clear traction characteristic. While some of the specimens are located between the ‘a’ and ‘b’ lines that delineate mixed traction and non-traction groups, the others are below the ‘a’ line, where the non-traction control groups are located. During the Early Shang period, three specimens of metatarsals are recorded in the absolute traction zone, implying that the use of cattle for traction may have started to a limited degree in this period; females and males/castrates are present, but the majority of them cannot be securely categorised as likely to belong to animals used for traction. The Late Shang dynasty, by contrast, witnessed a significant development in cattle management, as the morphometric analysis indicates that a fairly large number of cattle metapodials derive from animals used for traction. Such findings are in line with previous observations based on Pathological Index analysis (Lin et al., 2018). Furthermore, the present analysis suggests that this group included female cattle. In terms of the metacarpals, it seems many more specimens with female characteristics can be recognised following the ca. 65 mm separation on ‘Bd’ between females and males/castrates. While some cows beneath line ‘a’ tend to be the ones with no traction history, many between lines ‘a’ and ‘b’ indicate their potential use for traction by expressing greater ‘e/D1’ than the control female non-traction group. Likewise, the practice of cow traction can also be recognised in the metatarsals, as a number of specimens display ‘Bd’ values under ca. 60 mm and much greater ‘e/D1’ values. The implication is that Late Shang farmers were training female cattle for draught purposes.

The use of cattle for traction continued in the Western Zhou dynasty. Whilst it is difficult to identify traction in the small sample of metacarpals, the Western Zhou metatarsals express clear traction features. For the latter, apart from those individuals above the ‘b’ line with a clear traction character, many located between the ‘a’ and ‘b’ lines also reveal certain traction traits. It seems that both female and male cattle can be associated with traction from the perspective of ‘e/D1’. The implication is that cattle management practices regarding traction use stayed consistent through the whole Western Zhou dynasty.

It is also worth noticing in the periods when there is clear morphometric evidence for cattle traction, the Shang and Western Zhou dynasties, the metatarsals examined here express more marked traction features than the metacarpals from the same phases. The rear limbs of cattle provide the major forward propulsion for animals pulling heavy equipment, putting more pressure on the metatarsals than on the metacarpals (Thomas, 2008; Thomas et al., 2021). The observed traction features on Chinese Bronze Age metatarsals are consistent with traction-induced pathologies on cattle of known life histories.

Traction use revealed from cattle age profile

The evaluation of other factors such as the age profile of the different assemblages throws further light on the morphometric models. Even though a full evaluation of age profiles at each site was not possible because of the paucity of cattle skulls and dentition, preliminary data were available as follows: Erlitou – 56.23% older than 26 months (Yang, 2006: 31); Daxinzhuang – two out of the seven mandibles older than 8 years; Zhougongmiao – half samples older than 4 years (Lin et al., 2013); Zaoshugounao – over four fifths older than 2 years old (Li et al., 2020); Donglongshan – 1 senior and 3 adults out of the 11 specimens (Yang, 2011: 372). The indications are that older cattle are consistently more represented in the assemblages from the Late Shang phase onwards. Considering the time and energy consumed in traction training, maintaining experienced draught cattle to an older age seems likely, even though exploiting other secondary products such as their milk may also have been practised. Moreover, according to the study of a semi-feral herd of Chillingham cattle (Thomas et al., 2021), there is no strong correlation between the biological variable of age and metatarsal broadening, which suggests that the present morphometric models, especially for those of metatarsals, are more attributable to traction use than to ageing.

Socio-economic implications of the Late Shang cow traction

Morphometric models reveal many female cattle were trained for traction purposes from the Late Shang dynasty. The Late Shang, also known as Yin Shang, starts from about 1300 BCE when the king Pan’geng chose Yin as the capital (Li, 2002). Compared with the frequent capital-changing events during the Early Shang, the Late Shang consistently maintained Yin as its capital (Wang, 2014: 326), enabling the state to develop consistently. Two major characteristics of these societies were the maintenance of a strict social hierarchy and the performance of sacred ritual human/animal sacrifices that served to legitimate the divine authority of the elite (CEEC, 1986: 400; Wang, 2014: 20). A lot of cattle were maintained during this period, and a considerable number of them ended in ritual sacrifice (Wei, 1986; Wen and Yuan, 1983: 232–234; Xie, 1985: 38). Critical components of these rites included the burials of large beasts such as cattle, and the use of cattle scapulae as ‘oracle bones’ for divination (Wei, 1986; Figure 10). To satisfy the sacred gods, it appears that certain cattle were saved from agricultural work and reserved exclusively for sacrifice (Song, 1986). Interestingly, the Chinese character for ‘animal sacrifice’ – ‘Xi’ – referred to animals of a single colour, and the character ‘Sheng’ to their intactness, that is not being castrated (Xu, 1981; Zheng, 1992). To fulfil the large scale of animal sacrifice during the Late Shang (Lv, 2015), a significant number of intact bulls had to be taken out of agricultural system for sacrifice, leading to a shortage of working cattle. The implication of the present study is that Late Shang societies were having to train female cattle for traction, despite the negative effects of this strategy on for example milk production and on the effective strength of the traction cohort. Plough cattle were an enormous investment for pre-modern societies, Roman and mediaeval written records, for example, demonstrating that they were difficult to maintain for those communities due to the inadequate feed available through the difficult seasons (Barker, 1988; White, 1970: 283–284). Such investment in female cattle and the strategy of ‘bulls for sacrifice and cows for work’ represented a balance between the needs of the ritual and food-producing domains of the Late Shang state.

Figure 10.

A series of ritual-related records in the Late Shang-period Yin ruins. (a) Two cattle buried in a ritual pit at North Wuguancun (Cass, 1987; Image courtesy of Prof. H Yue, CASS). (b) An oracle bone of cattle scapula at Southeast Dasikong (Cass, 2019; Image courtesy of Prof. H Yue, CASS). (c) Oracle bones of cattle scapulae buried together in an ash pit at West Xiaotun (Liu, 2004b).

In sum, the Chinese Bronze Age witnessed the intensification of cattle labour exploitation. Whilst the morphometric models do not show absolutely clear traction use during the Erlitou period, there are hints of it in the material, and it has to be remembered that light draught work might not leave recognisable traces on bones. The ‘illogical’ employment of cows for traction in the Late Shang period is explicable in terms of farmers having to respond to the elites’ demands for large numbers of intact bulls for ritual sacrifice. Zooarchaeological methods of the kind illustrated here have the potential to provide new insights into the interactions between the agricultural systems and ideological practices, that together sustained the Late Shang civilisation.

Footnotes

Acknowledgements

We are grateful for the assistance of Xingshan Lei and Huiqi Cao during the analysis period and for the very helpful comments of the anonymous reviewers.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This paper was reshaped from part of the PhD research of ML in the University of Cambridge, and has been supported by a series of grants including the SJTU research grant, Humboldt Fellowship, CSC Cambridge Overseas Trust, Wenner-Gren Dissertation Grants, Clare Hall Research Awards and Dorothy Garrod Memorial Trust.

ORCID iD

Minghao Lin

References

Albarella

(1997) Shape variation of cattle metapodials: Age, sex or breed? Some examples from mediaeval and postmediaeval sites. Anthropozoologica 25–26: 37–47.

Barker

(1988) Archaeology and the Etruscan countryside. Antiquity 62: 772–785.

Bartosiewicz

Van Neer

Lentacker

(1997) Draught Cattle: Their Osteological Identification and History. Tervuren, Belgium: Musee Royal de l’Afrique Centrale, Sciences Zoologiques.

Bogaard

(2004) The nature of early farming in central and south-east Europe. Documenta Praehistorica 31: 49–58.

Bogucki

(1993) Animal traction and household economies in Neolithic Europe. Antiquity 67(256): 492–503.

Cai

Sun

Tang

, et al. (2014) The origins of Chinese domestic cattle as revealed by ancient DNA analysis. Journal of Archaeological Science 41: 423–434.

Cass

(1987) Anyang Wuguancun beidi Shangdai jisikeng de fajue [Excavation of the Shang ritual pits at North Wuguancun, Anyang]. Kaogu [Archaeology] 12: 1062–1070.

Cass

(2019) Anyang Yinxu Dasikongcun dongnandi 2015-2016 nian fajue baogao [Excavation of the Southeast Dasikong at Anyang Yin ruins, 2015-2016]. Kaogu Xuebao [Acta Archaeologica Sinica] 4: 503–564.

CEEC (1986) Zhongguo Dabaike Quanshu: Kaoguxue [Chinese Encyclopaedia: Archaeology]. Beijing, China: Chinese Encyclopaedia Press.

10.

De Cupere

Lentacker

van Neer

, et al. (2000) Osteological evidence for the draught exploitation of cattle: First applications of a new methodology. International Journal of Osteoarchaeology 10: 254–267.

11.

Degerbøl

Fredskild

(1970) The Urus (Bos primigenius Bojanus) and Neolithic Domesticated Cattle (Bos taurus domesticus Linne) in Denmark. Copenhagen: Kommissionar: Munksgaard.

12.

Falkenstein

(1936) Archaesche Texte aus Uruk. Berlin-Leipzig, Germany: Deutsche Forschungsgemeinschaft.

13.

Fock

(1966) Metrische Untersuchungen an Metapodien einiger europäischer Rinderrassen. PhD Dissertation, University of Munich, München.

14.

Gaastra

Green

Vander Linden

(2018) Gaining traction on cattle exploitation: Zooarchaeological evidence from the Neolithic western Balkans. Antiquity 92(366): 1462–1477.

15.

Gaastra

Insoll

(2020) Animal economies and Islamic conversion in eastern Ethiopia: Zooarchaeological analyses from Harlaa, Harar and Ganda Harla. Journal of African Archaeology 18: 1–28.

16.

Greenfield

(2010) The secondary products revolution: The past, the present and the future. World Archaeology 42(1): 29–54.

17.

Grigson

(1989) Size and sex: Evidence for the domestication of cattle in the Near East. In: Milles

Williams

Gardner

(eds) The Beginnings of Agriculture. British Archaeological Reports, International Series 496. Oxford: Archaeopress, pp.77–109.

18.

Helmer

Blaise

Gourichon

, et al. (2018) Using cattle for traction and transport during the Neolithic period: Contribution of the study of the first and second phalanxes. Bulletin de la Société Préhistorique Française 115(1): 71–98.

19.

Helmer

Gourichon

(2008) Premières données sur les modalités de subsistance à Tell Aswad (Syrie, PPNB moyen et récent, Néolithique céramique ancien) – Fouilles 2001-2005. In: Vila

Gourichon

Choyke

, et al. (eds) Archaeozoology of the Near East VIII. Lyon, France: Maison de l’Orient et de la Méditerranée, pp.119–151.

20.

Helmer

Gourichon

Monchot

, et al. (2005) Identifying early domestic cattle from pre-pottery Neolithic sites on the middle Euphrates using sexual dimorphism. In: Vigne

J-D

Peters

Helmer

(eds) The First Steps of Animal Domestication: New Archaeozoological Approaches. Oxford: Oxbow Books, pp.86–95.

21.

Holmes

(2014) Does size matter? Changes in the size of animals throughout the English Saxon period (AD 450–1066). Journal of Archaeological Science 43: 77–90.

22.

Johannsen

(2005) Palaeopathology and Neolithic cattle traction: Methodological issues and archaeological perspectives. In: Davies

Fabis

Mainland

, et al. (eds) Diet and Health in Past Animal Populations: Current Research and Future Directions. Oxford: Oxbow Books, pp.39–51.

23.

Jones

Hunt

Lightfoot

, et al. (2011) Food globalization in prehistory. World Archaeology 43(4): 665–675.

24.

Larson

Fuller

(2014) The evolution of animal domestication. Annual Review of Ecology Evolution and Systematics 45(1): 115–136.

25.

Larson

Piperno

Allaby

, et al. (2014) Current perspectives and the future of domestication studies. Proceedings of the National Academy of Sciences of the United States of America 111(17): 6139–6146.

26.

Lee

G-A

Crawford

Liu

, et al. (2007) Plants and people from the early Neolithic to Shang periods in north China. Proceedings of the National Academy of Sciences of the United States of America 104(3): 1087–1092.

27.

Tang

(2018) Qingtongqi yu Zhongguo Qingtong Shidai [Bronzes and the Bronze Age of China]. Hefei, China: University of Science and Technology of China Press.

28.

Lin

Luan

Fang

, et al. (2018) Pathological evidence reveals cattle traction in north China by the early 2nd millennium BC. The Holocene 28(8): 1205–1215.

29.

Lin

Miracle

Barker

(2016) Towards the identification of the exploitation of cattle labour from distal metapodials. Journal of Archaeological Science 66: 44–56.

30.

Lin

Chong

Lei

(2013) Zhougongmiao Shangzhou shiqi juluo dongwu ziyuan liyong chushi [Preliminary analysis of animal exploitation of the Zhougongmiao site during the Shang and Zhou periods]. Kaogu yu Wenwu [Archaeology and Cultural Relics] 3: 39–47.

31.

Liu

(2004a) The Chinese Neolithic: Trajectories to Early States. Cambridge: Cambridge University Press.

32.

Liu

Chen

(2003) State Formation in Early China. London: Duckworth.

33.

Liu

(2004b) Lun Yinxu jiagu de maicang zhuangkuang ji xiangguan wenti [Burying and related questions of the oracle bones in the Yin ruins]. Sandai Kaogu [Three-Dynasty Archaeology] 1: 354–370.

34.

(2002) The Xia-Shang-Zhou chronology project: Methodology and results. Journal of East Asian Archaeology 4(1–4): 321–333.

35.

Zhang

Wang

, et al. (2020) Animal use in the late second millennium BCE in northern China : Evidence from Zaoshugounao and Zaolinhetan in the Jing River valley. International Journal of Osteoarchaeology 30: 318–329.

36.

(2015) Shangren liyong huangniu ziyuan de dongwu kaoguxue guancha [Zooarchaeological analysis of the use of cattle during the Shang dynasty]. Kaogu [Archaeology] 11: 105–111.

37.

Matolcsi

(1970) Historische Erforschung der Körpergröße des Rindes auf Grund von ungarischem Knochenmaterial. Zeitschrift für Tierzüchtung und Züchtungsbiologie 87: 89–137.

38.

Piggott

(1992) Wagon, Chariot and Carriage: Symbol and Status in the History of Transport. London: Thames and Hudson.

39.

Sherratt

(1981) Plough and pastoralism: Aspects of the secondary products revolution. In: Hodder

Isaac

Hammond

(eds) Pattern of the Past: Studies in Honour of David Clarke. Cambridge: Cambridge University Press, pp.261–305.

40.

Sherratt

(1983) The secondary exploitation of animals in the Old World. World Archaeology 15(1): 90–104.

41.

Song

(1986) Jiaguwen qian zi shuo [The Qian character in oracle bone inscriptions]. In: Hu

(ed.) Jiaguwen yu Yinshangshi: 2 [Oracle Bone Inscriptions and the History of Late Shang: Volume 2]. Shanghai, China: Shanghai Ancient Books Press, pp.65–84.

42.

Telldahl

Svensson

Götherström

, et al. (2012) Osteometric and molecular sexing of cattle metapodia. Journal of Archaeological Science 39(1): 121–127.

43.

Thomas

(1988) A statistical evaluation of criteria used in sexing cattle metapodials. Archaeozoologia 2(1, 2): 83–92.

44.

Thomas

Bellis

Gordon

, et al. (2021) Refining the methods for identifying draught cattle in the archaeological record: Lessons from the semi-feral herd at Chillingham Park. International Journal of Paleopathology 33: 84–93.

45.

Thomas

Holmes

Morris

(2013) “So bigge as bigge may be”: Tracking size and shape change in domestic livestock in London (AD 1220–1900). Journal of Archaeological Science 40(8): 3309–3325.

46.

Thomas

(2008) Diachronic trends in lower limb pathologies in later medieval and post-medieval cattle from Britain. In: Grupe

, et al (eds) Limping Together Through the Ages: Joint Afflictions and Bone Infections. Rahden, Westf: Verlag Marie Leidorf, pp.187–201.

47.

Von Den Driesch

(1976) A Guide to the Measurement of Animal Bones From Archaeological Sites. Cambridge: Peabody Museum of Archaeology and Ethnology, Harvard University.

48.

Wang

(2014) Zhongguo Kaoguxue Dacidian [Dictionary of Chinese Archaeology]. Shanghai, China: Shanghai Cishu Press.

49.

Wei

(1986) Cong Jiaguwen cailiao zhong kan Shangdai de yangniuye [Cattle keeping in the Shang Dynasty revealed from oracle bone inscriptions]. In: Zhang

Zhu

(eds) Zhongguo Xumu Shiliao Ji [Collected Works of Chinese Liverstock]. Beijing, China: Science Press, pp.154–159.

50.

Wen

Yuan

(1983) Yinxu Buci Yanjiu: Kexue Jishu Pian [Studies of Oracle Bone Inscriptions in Yin Ruins: Science and Technology]. Chengdu, China: Sichuan Social Science Academy Press.

51.

White

(1970) Roman Farming. London: Thames & Hudson.

52.

Xie

(1985) Zhongguo Yang Niuyang Shi [The History of Cattle and Sheep Keeping in China]. Beijing, China: Agricultural Press.

53.

(1981) Shuowen Jiezi Zhu [Interpretation of the Book of Shuowen Jiezi]. Shanghai, China: Shanghai Ancient Books Press.

54.

Yang

(2006) Henan Yanshi Erlitou Yizhi de Dongwu Kaoguxue Yanjiu [Zooarchaeological investigation of the Erlitou site, Yanshi, Henan]. MA Thesis, CASS, China.

55.

Yang

(2011) Shangluo Donglongshan [The Donglongshan Site in Shangluo]. Beijing, China: Science Press.

56.

Zeder

(2008) Domestication and early agriculture in the Mediterranean Basin: Origins, diffusion, and impact. Proceedings of the National Academy of Sciences of the United States of America 105(33): 11597–11604.

57.

Zhao

(2014) Zhongguo gudai nongye de xingcheng guocheng: fuxuan chutu zhiwu yicun zhengju [The formation of Chinese ancient agriculture: Evidence from floated plant remains]. Disiji Yanjiu [Quaternary Sciences] 34(1): 73–84.

58.

Zheng

(1992) Zhouli: Kaogongji [The Rites of Zhou: Records of Craft and Technology]. Beijing, China: Chinese Publishing House.

Bulls for sacrifice,cows for work? Morphometric models suggest that female cattle were used for traction in the Chinese Bronze Age Late Shang dynasty (ca. 1300–1046 BCE)

Abstract

Keywords

Introduction

Methods and materials

Methods

Materials

Results

The Erlitou period

The Early Shang dynasty

The Late Shang dynasty

The Western Zhou dynasty

Discussion and conclusion

Cattle traction in the Chinese Bronze Age

Traction use revealed from cattle age profile

Socio-economic implications of the Late Shang cow traction

Footnotes

Acknowledgements

Funding

ORCID iD

References