Introduction

Stratification involves the unequal distribution of resources among social groups, which is maintained and strengthened through institutions and interpersonal relationships so that the dominant group/s gain/s control over social resources (McLeod 2013). The emergence of ranked and patriarchal Chinese culture from the relatively egalitarian Neolithic to the hierarchical Bronze Age dynasties (c. 1600–221 BCE) laid the foundations for the marked class and gender inequalities seen in historical times and even today in certain aspects in the region (Dong et al. 2017; Huang 2004; Zhou et al. 2019, 2021a, b).

The Zhou Dynasty is the first well-documented period of China, representing a distinctive position in political and cultural history. The Confucian tradition, which has influenced Chinese society for millennia, traces its roots to this period (Feng 2013). The Zhou Dynasty is divided into Western Zhou (1046–771 BCE) and Eastern Zhou (770–221 BCE). The Eastern Zhou can be divided into two phases: the Spring and Autumn (770–476 BCE) and the Warring States (475–221 BCE) (historians have differing perspectives on the dates in the two phases, here we reference the records in Shi Ji (史记, Records of the Grand Historian)) (Si 1959).The decline of central royal authority under the Zhou emperor from the Eastern Zhou opened a half-century of warfare among states until the unification of the Qin Dynasty in 221 BCE, which established China’s first empire (Li 2007; Feng 2013). The Eastern Zhou was a crucial period in the political, economic, and cultural transformation of ancient Chinese society. During the late Spring and Autumn to the Warring States period, the well-known intellectual movement of “百家争鸣 (the Hundred Schools of Thought)” emerged, and the pioneers of Confucianism such as Confucius and Mencius were prominent during this period (Li 2007; Yang 2016). This school of thought had an influence on the development of the class- and gender-based stratification during this period.

The people in the Eastern Zhou were divided into two major classes, the nobles, comprising the 周天子 (zhoutianzi [Zhou emperor]), 诸侯 (zhuhou [vassals]), 卿大夫 (qingdafu [high official]) and 士 (shi [scholar-gentry]), and the commoners and other people of lower class, who were in a subordinate position (Li 2007, p. 160; Yang 1981, p. 94). Different classes had different rights, and these class-based differences were also expressed in their mortuary rituals, such as the quality and quantity of coffins, grave goods, and sacrificial animals as defined by social regulations (Bi 2010; Li 1990; Lin 1990; Song 2011; Yuan 2014; Zhang 2012; Zhang and Mei 2021). One particularly striking class-based difference is the increasing prevalence of sacrificial human companions within noble burials, which reflects the intensification of social stratification during the Eastern Zhou period (Huang 1987, 2004; Wang 2013).

Patriarchal gender institutions were possibly established during the Bronze Age, from which time females suffered the dual oppression of patriarchy and class (Du 1998; Geng 2017; Zhou 2006). This change in the female’s social status happened gradually over time and then by the Eastern Zhou period they were severely disadvantaged (Du 1998; Geng 2017; Zhou 2006). It is important to note that gender intersects with other identities in addition to class, including age, marital status, ethnicity, and other social factors (Geller 2009; Lindsey 2020; Meskell and Preucel 2007; Sofaer 2006; Yao 2021). For example, Han (1999) argues that the implementation of the patriarchal system highlighted the legitimate status of the wife. As we can see in burials, wives were commonly buried together with their husbands in equal parallel positions, while concubines were often sacrificed (Han 1999). Most sacrificial people in Eastern Zhou archaeological sites are young females, so scholars have hypothesized that the majority of sacrificial companions would be concubines and 女乐 (nvyue [females who entertain the noble class with their singing and dancing skills]) (Gu 1982; Huang 1997; Mi 2014; Shandong Museum 1977).

However, more scholars realize now that burial treatments are not unbiased reflections of the social status of individuals during their lifetime, but the material expressions of the status of the deceased at a specific time plus manipulations by the living (Cannon et al. 1989; Dong et al. 2019; Parker Pearson 1993; Shelach 2008; Underhill 2000). Therefore, solely relying on mortuary analysis to make inferences on the development of class and gender stratification can sometimes lead to incorrect conclusions. Historical records mostly focus on nobles and are therefore biased as well. Fortunately, bioarchaeological studies in recent years, especially the application of methods such as stable isotope dietary analysis and paleopathology, are providing direct information on the diet and health of past populations, which can contribute to the understanding of the intersection of class and gender identity during their life course (e.g. Dong et al. 2017; Reitsema and Vercellotti 2012; Stone and Netzer Zimmer 2022; Umezaki et al. 2016; Zuckerman et al. 2023).

The recent excavation of the site of Dahan (大**), an Eastern Zhou cemetery with well-preserved skeletons of nobles and sacrificial companions (Shandong Provincial Institute of Cultural Relics and Archaeology and Tengzhou City Administration of Cultural Heritage 2021), provides an invaluable opportunity to investigate the intersection of class, sex (as a proxy for gender), and mortuary rituals during this period. In this study, we analyzed stable isotope values of human and faunal bone samples, combined with evidence from paleopathological analysis, historical records and archaeological context on class- and sex-based bias to investigate: (1) the diet for the overall Dahan cemetery; and (2) the influence of class and gender on diet.

Social stratification during the Eastern Zhou

Social roles of sacrificial human companions

In ancient China, there were two main types of human sacrifices. 人殉 (renxun, sacrificial human companion) is essentially a type of grave good that ensures the burial owner can continue to enjoy services after death, while 人牲 (rensheng, sacrificial human offering) represents an offering for deities or ancestors (Huang 2004; Yang and Zhu 2018). In the archaeological context, sacrificial human companions are usually found with complete skeletons, and many have their own coffins and grave goods, whereas sacrificial human offerings are usually incomplete, with almost no coffins or grave goods (Huang 2004; Yang and Zhu 2018). Scholars agree that the relationships between the sacrificial companions and the burial owners are based on pre-existing relationships during life (Feng 2005; Gu 1982; Huang 2004; Yao 1960). Moreover, there are also status-based differences among the sacrificial companions as shown by their burial treatment largely depending on their social closeness to the burial owners (e.g. Gu 1982; Jiqing Highway Cultural Relics Archaeology Team **uhui Branch 1993; Shandong Museum 1977; Yao 1960). The practice of sacrificing companions has been prevalent in Shandong region since the Shang dynasty and continues into the Eastern Zhou period (Huang 2004, p. 187; Wei and Song 1998; Yin 2012, 2015). Thus, a comparative study of the burial owners and sacrificial companions can provide valuable insights into their relationships, and the social roles of sacrificial companions in the Eastern Zhou society.

Gender-based stratification

Gender-based stratification was evident in the Eastern Zhou period, as reflected in the Guo Yu (国语, Discourses of the States) “男女之别, 国之大节也, 不可无也 [the distinction between males and females is a fundamental aspect of the state, which cannot be ignored]” (Xu 2002, p. 147). The Yili (仪礼, Book of Etiquette and Ceremonial) also states “妇人有三从之义, 无专用之道, 故未嫁从父, 既嫁从夫, 夫死从子 [women have three feminine obedience [sic], and they have no independent identity, so they obey their father before marriage, obey their husband after marriage, and follow their son after their husband dies]” (Ruan 1980, p. 1106). There have been a considerable number of studies that examine the historical records on gender-based bias of the Eastern Zhou period (e.g. Du 1998; Zhou 2006). Archaeological investigations focused on gender-based differences in ancient China arose in the early twenty-first century (Johnson 2004). Additionally, studies focusing on burial practices (e.g. Ariko 1989; Han 1999; Jiang 2004; Li and Li 1997; Liu 2009; Long 2000; Wang 1982; Yong 2004), especially grave goods (e.g. Chen 2005, 2006, 2009; Yang and Sun 2016; Yang et al. 2020), also remark on gender-based differences during the Eastern Zhou.

Diet and social stratification

As the Li Ji (礼记, Book of Rites) states “夫礼之初, 始诸饮食 [the beginning of rituals began with food and drink]” (Ruan 1980, p. 1415). Food was a vital part of the social rituals during the Eastern Zhou period, hence was heavily influenced by this social stratification of class and gender in Eastern Zhou society. Nobles and commoners were referred to in the historical records mainly as “肉食者 [meat-eaters]” vs. “藿食者 [huo-eaters, huo refers to soybean leaves (Guo 2004)]” or “粱 [liang, foxtail millet] eaters” vs. “糟糠 [zaokang, chaff (zao refers to distiller’s grains and grain husk, kang refers to chaff and bran)] eaters.” In the Zuo Zhuan (左传, The Commentary of Zuo), there are many records of meat-eaters, such as “肉食者鄙, 未能远谋 [the meat-eaters (i.e. the high-status officials) are short-sighted and unable to plan ahead]” (Yang 1981, p. 182). The Shuo Yuan (说苑, Garden of Stories) also records “肉食者已虑之矣, 藿食者尚何与焉 [meat-eaters are already considering the policies, there is no need to get huo-eaters involved]” (Liu 1987, p. 271). In Eastern Zhou society, foxtail millet was regarded as a valued food that noble people had the privilege to consume (Qi 2001), while the commoners mostly consumed soybeans, soybean leaves, and chaff (Wang 1988, p. 65; Gao et al. 2010, p. 712). The use of specific sacrificial animals is also dictated by people’s class (Book of Rites, Ruan 1980, p. 1337), these rules are confirmed by zooarchaeological evidence (e.g. Song 2011). Gender-based dietary differences can be found in historical records too, such as “七年, 男女不同席, 不共食 [when reach seven years old, males and females do not sit together nor share food]” (Book of Rites, Ruan 1980, p. 1471); and males might consume more dogs and females more pigs (Lv 2016, p. 8, 71). These studies suggest that gender also has an impact on diet in the Eastern Zhou, possibly resulting in separate dining or differential consumption in food types.

Materials and methods

Archaeological background

Dahan cemetery (34°56′44.6″ N, 117°14′56.8″ E) is located in Tengzhou City, Shandong Province, China (Fig. 1). In response to looting at this site in January 2017, the Shandong Provincial Institute of Cultural Relics and Archaeology and Tengzhou City Administration of Cultural Heritage carried out rescue excavations from 2017 to 2020. The cemetery covers an area of approximately 7,000 m2 and extends over a long period from the late Spring and Autumn to the end of Warring States (770–221 BCE). A total of 48 large-sized burials and 150 small-sized burials were excavated (Fig. 1), and more than 3,000 artifacts including pottery, bronze, jade, stone, bone, gold, and iron were found. The 48 large-sized burials span four phases: late Spring and Autumn (phase A); early Warring States (phase B); middle Warring States (phase C); and late Warring States (phase D). Some large-sized burials have one tomb ramp and are roughly oriented towards the east. Nearly half of the large-sized burials have 1–10 sacrificial companions around the ercengtai (二层台, earthen ledges, such as Fig. 2A–B). The inscription “郳大司马 [the high-status official of the Ni State]” on the bronze vessels from burial M43 (Fig. 2C), indicates that these large-sized burials are of nobles from the Ni State (郳国) during the Eastern Zhou period. The 150 small-sized burials all date to the end of Warring States, later than the large-sized burials (Shandong Provincial Institute of Cultural Relics and Archaeology and Tengzhou City Administration of Cultural Heritage 2020, 2021). This study focuses on large-sized noble burials.

Fig. 1
figure 1

Map showing the location of the Dahan cemetery (34°56′44.6″ N, 117°14′56.8″ E) and the distribution of burials of the site (blue represents burials analyzed in this study)

Full size image
Fig. 2
figure 2

A Burial M57 and B M208 at Dahan cemetery (North is up); C some bronze vessels from burial M43, the inscriptions indicate this burial belonged to a high-status official of the Ni State (郳国) during the Eastern Zhou

Full size image

A total of 112 human skeletons were excavated from the noble burials (i.e., large-sized burials) of the Dahan cemetery, of which 83 individuals (35 nobles and 48 sacrificial companions) could have their sex estimated (**ao 2021). There are 14 females and 21 males in the noble group, and 44 females and 4 males in the sacrificial group. Most noble females were older (older than 35 years, n = 12/14), while the majority of sacrificial females were young adults (younger than 35 years, n = 34/44). Comparatively, the proportion of young noble males (n = 10/21) was higher than young noble females, and the 4 sacrificial males were all older adults (**ao 2021).

Materials

Stable isotope analysis samples include human and animal bones from the large-sized burials. Human bone samples were taken from 72 individuals from 31 burials that had suitable bone preservation (Fig. 1), including 31 noble individuals, and 41 sacrificial individuals (from 13 burials) (see Table S1 in Online Resource 1 for details). Of the 72 individuals, 47 were female (12 nobles, 35 sacrificial individuals), 21 were male (18 nobles, 3 sacrificial individuals), and 4 could not have their sex estimated (1 noble, 3 sacrificial individuals). The age-at-death of these individuals include subadults (n = 2, < 18 years), young adults (n = 32, 18–34 years), middle adults (n = 32, 35–49 years), old adults (n = 3, > 50 years), unaged adults (n = 2, > 18 years), and one individual fell within the range of both subadult and young adult (16–20 years). The standards used for sex and age-at-death estimation are described in Online Resource 2. Faunal samples were taken from 37 animals from 16 burials, and include 9 pigs, 6 cattle, 8 sheep/goats, 8 deer, 5 dogs, and 1 fish (see Table S2 in Online Resource 1 for details).

For paleopathological analysis, all observable human skulls from large-sized burials were analyzed. Fifty-four individuals could be observed for cribra orbitalia (CO), including 14 noble individuals (7 females, 7 males) and 40 sacrificial individuals (32 females, 3 males, and 5 who could not have their sex estimated). Seventy-nine individuals could be observed for porotic hyperostosis (PH), including 33 noble individuals (13 females, 20 males) and 46 sacrificial individuals (37 females, 4 males, 5 could not have their sex estimated).

Methods

This study primarily applied stable isotope analysis, supplemented by paleopathological analysis (cribra orbitalia and porotic hyperostosis).

Reconstructing paleodiet using stable isotope analysis

Stable isotope dietary analysis of past populations is based on the premise that “you are what you eat” (Kohn 1999). Different foods have different isotopic compositions, which can enter the bodies of humans and animals through the food chain. Thus, stable isotope analysis can identify the food sources of past humans and animals (Britton 2017; Lee-Thorp 2008), and provide information on the pattern of subsistence (Bogaard et al. 2013; Dong et al. 2021; Ma et al. 2021; Wang et al. 2022), management of livestock (Chen et al. 2016; Qiu et al. 2022; Szpak 2014), and social stratification (Cheung et al. 2017a, b; Dong et al. 2017; Reitsema and Vercellotti 2012).

C3 and C4 plants have different photosynthetic pathways with C4 plants having higher δ13C values than C3 plants (van der Merwe 1982). During the Eastern Zhou period in the Shandong region, C3 plants mainly included wheat, soybeans, and most wild vegetations, and C4 plants mainly include foxtail millet, broomcorn millet, and their byproducts (e.g. straw) (Chen et al. 2018; Guo et al. 2019; ** et al. 2013; Ma et al. 2019; Wang et al. 2011; Wang 2018; Wei et al. 2018). Animals fed on C3 or C4 plants will also have different carbon isotope values. Therefore, it is possible to measure the δ13C values in human and animal bone collagen to infer the type of plant and animal food they consumed. Nitrogen isotopes show a trophic level effect, where δ15N values are enriched by 3–6‰ at each level (Hedges and Reynard 2007; O’Connell et al. 2012). Furthermore, the isotopic fractionation of nitrogen is susceptible to environmental and human physiological influences (Ambrose 1991; Bogaard et al. 2013; Dong et al. 2022; Katzenberg and Lovell 1999; Schurr 1998). The δ15N values in animal bone from Dahan are analyzed to provide a baseline reference for potential meat sources for humans; δ15N values in human bone collagen are used to estimate the level of animal protein in the diet, and physiological stress during their lifetime.

Bone collagen preparation and stable isotope analysis were carried out at the JoInRLESA (Joint International Research Laboratory of Environmental and Social Archaeology, Shandong University, Qingdao, China). The specific methods of collagen preparation, isotopic analysis, statistical analysis, and diagenesis assessment are described in Online Resource 2. The use of calibration standard reference materials, check standard reference materials, and duplicate samples in this study are shown in Tables S3, S4, and S5 (Online Resource 1). The standard uncertainties of the carbon and nitrogen stable isotope values are ± 0.1‰ and ± 0.2‰, respectively (Online Resource 3).

Paleopathological analysis: cribra orbitalia and porotic hyperostosis

CO refers to porous lesions on the orbital roof, and PH refers to porous lesions on the cranial vault. These skeletal lesions are often considered non-specific indicators of chronic disease and nutritional deficiency generally associated with anemia, including acquired (e.g. iron-deficiency anemia) and genetic hemolytic anemia (e.g. sickle cell disease and thalassemia) (Angel 1966; Cybulski 1977; Walker et al. 2009; Oxenham and Cavill 2010; Rivera and Mirazón Lahr 2017). Other conditions such as vitamin deficiencies (e.g., scurvy and rickets), infectious diseases, trauma, and inflammation can also cause CO and PH (Ortner and Ericksen 1997; Schultz 2001; Wapler et al. 2004; Cole and Waldron 2019; O’Donnell et al. 2020). Paleopathological analysis was conducted by the third author. The standards used for identification of CO and PH are described in Online Resource 2.

Results

Stable isotope analysis

The δ13C and δ15N values for human and faunal bone samples are shown in Fig. 3 (raw data is given in Tables S1 and S2 in Online Resource 1). Ninety-three samples yielded acceptable collagen for analysis, including 61 human and 32 faunal samples.

Fig. 3
figure 3

Scatter plot of δ13C and δ15N values of human and faunal bone collagen from the Dahan cemetery

Full size image

Stable isotope values of human bone collagen

Of the 61 human samples, 26 were from nobles and 35 from sacrificial individuals (Table 1). The δ13C values ranged from –16.8‰ to –9.2‰ (mean: –12.6 ± 2.1‰), and the δ15N values ranged from 5.2‰ to 11.9‰ (mean: 8.5 ± 1.8‰).

Table 1 Stable isotope ranges and medians for human bone collagen samples from the Dahan cemetery
Full size table

Class-based

Both the carbon and nitrogen isotope values of the noble group were significantly higher than that of the sacrificial group (Fig. 4 and Table S6 in Online Resource 1). Neither δ13C values (z = –0.798, p = 0.432) nor δ15N values (z = –0.772, p = 0.462) of nobles changed significantly over time, but the sacrificial companions showed a statistical decrease of δ13C values from the Spring and Autumn (phase A) to the Warring States period (phase B, C, and D) (z = –2.476, p = 0.012), while no significant change were found in δ15N values (z = –1.743, p = 0.085) (Fig. 5 and Table S7 in Online Resource 1). There was no significant difference in δ13C values between nobles and sacrificial companions during the Spring and Autumn period (z = –1.457, p = 0.155), but by the Warring States period, the difference has become more obvious (z = –3.786, p < 0.001); while the δ15N values showed significant differences both in the Spring and Autumn (z = –3.908, p < 0.001) and the Warring States period (z = –3.895, p < 0.001) (Fig. 5 and Table S8 in Online Resource 1).

Fig. 4
figure 4

Scatter plot of class-based δ13C and δ15N values of human bone collagen from the Dahan cemetery (ellipses represent 68% confidence intervals of each group). Specific labeled individuals are discussed below

Full size image
Fig. 5
figure 5

Box plot and scatter plot of phase-based δ13C and δ15N values of human bone collagen from the Dahan cemetery (Phase: A = Late Spring and Autumn, B = Early Warring States, C = Mid Warring States, D = Late Warring States)

Full size image

Sex-based

There were no significant differences in δ13C values (z = –0.846, p = 0.406) nor δ15N values (z = –0.350, p = 0.754) between noble males and noble females (Fig. 6 and Table S9 in Online Resource 1). However, statistically significant differences were found between noble females and sacrificial females in both δ13C values (t = 2.590, df = 36, p = 0.014) and δ15N values (t = 6.425, df = 36, p < 0.001) (Fig. 6 and Table S9 in Online Resource 1). As there were only three sacrificial males available for analysis, it was not possible to assess sex-based differences within the sacrificial group.

Fig. 6
figure 6

Box plot and scatter plot of sex-based δ13C and δ15N values of human bone collagen from the Dahan cemetery

Full size image

Stable isotope values of faunal bone collagen

The δ13C values of faunal bone samples ranged from –24.4‰ to –8.2‰ (mean: –16.1 ± 3.8‰) and δ15N values ranged from 3.2‰ to 11.4‰ (mean: 5.7 ± 2.1‰), with large variations in carbon and nitrogen isotope values among species and within domestic species (Fig. 3 and Table 2). The δ13C and δ15N values of pigs, cattle, and dogs were relatively close. Sheep/goats had lower δ13C values than those of pigs, cattle, and dogs, but δ15N values were similar. One sheep/goat with the very high δ13C and δ15N values (from burial M206) was a subadult, likely reflecting an unweaned diet. The carbon and nitrogen isotope values of deer were lower than other animals with the narrowest range of variation. As the only aquatic animal in the sample, the stable isotope value of fish (grass carp) was distinct from other species, showing the lowest δ13C value (–24.4‰).

Table 2 Stable isotope ranges and medians for faunal bone collagen samples from the Dahan cemetery
Full size table

Paleopathological analysis

Both CO and PH were identified among the Dahan individuals. The prevalence of individuals with CO was 14% (2/14) and 60% (24/40) in the noble group and sacrificial group, respectively. The prevalence of PH was 21% (7/33) and 37% (17/46) in the noble group and sacrificial group, respectively. In terms of sex, the prevalence of CO and PH in noble females was 29% (2/7) and 15% (2/13), respectively. For noble males, the prevalence of PH was 25% (5/20), and no individuals with CO (0/7). The prevalence of CO and PH in sacrificial females was 56% (18/32) and 38% (14/37), respectively. For sacrificial males, the prevalence of CO and PH was 33% (1/3) and 25% (1/4), respectively.

Discussion

Diet

The isotope values of the Dahan people suggest that they consumed both C3 and C4 foods, meanwhile there was considerable variation among individuals in both cereal intake and meat consumption. During the Eastern Zhou period, as mentioned above, the major crop in the Shandong region is foxtail millet, wheat, soybeans, and broomcorn millet. The human stable isotope results support this general conclusion, while some people consumed large amounts of wheat, soybeans, or other C3 foods, and some consumed more C4 foods such as foxtail millet.

Both historical records (Ruan 1980, pp. 224, 2666, 2768; Wang 1988, p. 67) and zooarchaeological studies (Fan 1999; Li et al. 2020; Song 2008a, b, 2014; Song and Hao 2016; Song and Yan 2015; Yuan and Yang 2005) indicate that the meat sources of the Shandong population in the Eastern Zhou period were diverse, including domestic animals (such as pigs, cattle, sheep/goats, dogs, and chickens) and wild animals (mainly deer, plus some fish and shellfish). Those species were all present at Dahan and probably constitute part of the Dahan population’s diet (Ding 2022). Faunal isotopic data shows that pigs, cattle, sheep/goats, and dogs all consumed various amounts of C4 foods (millets and its byproduct) and C3 foods (wild grass). Moreover, the δ15N values of these domestic animals are close to those of the Dahan individuals. The historical evidence combined with the isotopic data show that the main sources of meat of the Dahan population were domestic animals such as pigs, cattle, and sheep/goats. In addition, there are significant differences in meat consumption among individuals as reflected in the highly variable δ15N values.

Class-based differences

Noble group vs. sacrificial group

Taken as a whole, the noble group had significantly higher δ13C and δ15N values than the sacrificial group, suggesting that on average their diet includes more C4 food (millets or animals raised on C4 plants, such as pigs). The diet of sacrificial companions contained more C3 foods, which, combined with their lower δ15N values and the plant resources available in Shandong during the Eastern Zhou, indicates that their diet likely comprised more wheat and/or soybeans (Yoneyama et al. 1986).

The class-based differences in the diet of the Dahan population are also supported by paleopathological evidence. There was a higher prevalence of CO and PH in the sacrificial group compared with the noble group. CO and PH can be caused by multiple etiologies, but anemia is considered the common causative factor, especially iron-deficiency anemia (e.g. Oxenham and Cavill 2010). Animal foods are rich in iron and more readily absorbable, while plant foods contain less iron and it is in a form that is not easily absorbed by the intestines (Buzała et al. 2016). Although it has been suggested that iron bioavailability varies little among cereal types (e.g., rice, wheat, maize) (Cook et al. 1997; Yang et al. 2016), a recent meta-analysis indicates that consumption of millets may reduce the iron-deficiency anemia (Anitha et al. 2021). The prevalence of CO and PH in the sacrificial group is likely associated with less consumption of meat or millets. However, other causes of anemia such as parasite infections (e.g., hookworm), chronic diseases (e.g., tumors), cannot be fully excluded (Handelman and Levin 2008; Bechir et al. 2012; Ludwig et al. 2013).

The lower social status of sacrificial companions was further reflected in their dietary changes through time. Archaeobotanical studies have shown that wheat was introduced to the Shandong region no later than the Longshan culture period (c. 2500–2000 BCE) but remained a small component of the diet until the Zhou period, when wheat rapidly became a predominant staple (Guo and ** 2019). There was no significant change in the diet of the Dahan nobles over time, but there was a notable shift in the sacrificial group: specifically, the δ13C values significantly decreased from the Spring and Autumn to the Warring States, indicating that the sacrificial companions probably consumed more C3 foods such as wheat and/or soybeans over time. Similarly, stable isotope analysis from ancient Zheng Han City (郑**故城) showed that wheat consumption increased substantially in the Warring States and this was mostly restricted to the lower-status individuals (Zhou et al. 2017). Zhou et al. (2017) attributes this shift to the external risks of this period: wars or disasters led to a food supply crisis, which forced the lower-status individuals to consume more of the low-value staple of wheat. Hence, both Zhou et al.’s (2017) and our study suggest that there are status-based differences in the spread of wheat, with the lower-status (in this case sacrificial companions) being forced to accept and consume wheat during this time.

Inter-noble group differences

It has been argued that there are nobles of different statuses in the Dahan cemetery from the archaeological context (Shandong Provincial Institute of Cultural Relics and Archaeology and Tengzhou City Administration of Cultural Heritage 2020). However, we did not find differences in stable isotope values based on burial variables (e.g. tomb ramp, burial orientation, chamber volume, funeral wares, waist pits (腰坑, yaokeng), and sacrificial companions, see Tables S10-S16 in Online Resource 1). Nevertheless, there were several noble individuals (M57, M208, M209, and M201) worthy of further discussion who have lower nitrogen isotope values than most nobles and some sacrificial individuals (Figs. 4 and 7).

Fig. 7
figure 7

Scatter plot of δ13C and δ15N values of nobles and sacrificial human companions from the Dahan cemetery grouped by burials. The nobles of burial M39 and M42 did not yield collagen that met the quality criteria

Full size image

M57 and M208 are the two largest burials in the Dahan cemetery, with the possibility that M208 was the king named “Ke (克)” of the Ni State (郳国) (Hao et al. 2022; Shandong Provincial Institute of Cultural Relics and Archaeology and Tengzhou City Administration of Cutural Heritage 2020). The noble individual M57 was a middle-aged adult, and M208 was a young adult male. Potential explanations for their low isotope values include but are not limited to (1) dietary habits, where an individual’s preference for food differs from others in the noble class. They may have been influenced by the cultural concept of health at that time, which considered plant foods as a healthy option compared with meat (Deng and Liu 2020; Zhao et al. 2022); and (2) sociopolitical-economic factors. Deer have lower δ15N values (3.6 ± 0.2‰, n = 8) in comparison with domesticated animals (6.5 ± 2.0‰, n = 23), and it is possible that M57 and M208 consumed more wild resources (such as deer) than other nobles. Interestingly, burial M208 is also found with the highest number of deer bones among the noble burials (Ding 2022).

The noble individual M209, a young adult male, has carbon and nitrogen isotope values that were not only lower than that of his sacrificial companion but also lower than those of most sacrificial individuals in the other burials (Figs. 4 and 7). However, the mortuary treatment (one inner coffin and one outer coffin, and two surrounding sacrificial individuals), and the quality and quantity of existing grave goods (3鼎[ding, tripods]) indicate that his status was high before death. Isotopic values of bone collagen can only reflect the long-term average dietary conditions of the last decades before death (Hedges et al. 2007), and thus, the high-quality foods consumed after a shift in social status may not be reflected in the isotopic signal of bones. The contradiction between the mortuary treatment and the stable isotope values may imply that his status was achieved from his own efforts (i.e. achieved status, see Giddens and Sutton 2014, p. 161). There were many weapons buried with him, including bronze swords, spears, dagger-axes, and arrowheads suggesting that this individual may have increased his status through military achievements. During the Eastern Zhou, social order was turbulent and wars were frequent, and there are several cases of individuals acquiring elevated status through their capabilities, such as famous historical figures Zou Ji (邹忌) of the Qi State (齐国), Zhang Yi (张仪) of the Qin State (秦国), and Shen Buhai (申不害) of the Han State (**国), all of whom were typical examples of 布衣卿相 [a high official coming from a commoner’s family] in this period (Hsu 1965). However, other potential causes that could lead to lower carbon and nitrogen isotope values of M209, such as personal preference of a more vegetarian diet, cannot be fully excluded. In addition, the noble individual M201 (a middle adult female) also showed lower carbon and nitrogen isotope values than most nobles (Figs. 4 and 7), but her burial treatment was not significantly different from that of the other nobles. She has no PH and CO as well.

Inter-sacrificial group differences

The δ13C values of the sacrificial companions range from –16.8‰ to –9.3‰ and δ15N values range from 5.2‰ to 9.7‰, which reflect large variations within the group (Fig. 7). Sacrificial individuals of burials M39, M57, and M208 stand out as they have higher δ13C and δ15N values, which are contributing to the large variation in isotope values for the total sacrificial group (Fig. 7). Although M39 does not have isotopic data for the noble individual, both M57 and M208 have at least one sacrificial individual who has higher δ15N values than the nobles in the burial. As discussed above, M57 and M208 are the two largest burials in the cemetery, with the most sacrificial companions, 5 and 10 individuals, respectively (Fig. 2A–B). Burial M208 also has the widest range of both δ13C and δ15N values compared with the other burials, suggesting that there are large dietary differences among the sacrificial individuals in this burial. Of these, individuals M208-6 and M208-7 have the highest isotopic values, and interestingly they also have more existing grave goods than others in this burial (Fig. 2B). Burial M39 has four sacrificial companions, and abundant grave goods, including bronze ritual vessels, jade, crystal, and other artifacts of high value (Shandong Provincial Institute of Cultural Relics and Archaeology and Tengzhou City Administration of Cultural Heritage 2021). These archaeological findings suggest that the nobles of burials M39, M57, and M208 may have had a high social status, and the diet of the sacrificial companions may have benefited from the resources of the nobles compared with sacrificial individuals from other burials. In particular, the individuals with higher nitrogen isotope values (e.g. M39-3) may have had a closer relationship with the nobles (e.g. wives, concubines, or attendants of the nobles).

Individuals M42-1 (child, 8–10 years old) and M208-8 (female, young adult) are the only burials with flexed legs in the entire cemetery. M208-8 has the lowest δ15N value among the sacrificial group and M42-1 also has lower isotope values than most sacrificial individuals (Fig. 4). These two individuals consumed more C3 foods and less animal protein, suggesting that they probably had a lower social status, possibly slaves of the nobles. However, the diet of M42-1 may also represent a typical childhood diet, more work needs to be done to investigate this hypothesis.

Gender-based differences

As we discussed earlier, gender intersects with age, class, ethnicity, and many other factors. Here, we use “sex” as a proxy for “gender” based on dietary differences to investigate gender-based stratification during the Eastern Zhou period. Our discussion below will center around four social groups: noble males, noble females, commoner (including sacrificial) males, and commoner (including sacrificial) females.

We did not find sex-based dietary differences between noble males and noble females at Dahan. This finding is also supported by another study of an Eastern Zhou population. Stable isotope values from 23 noble individuals of the Guo State (虢国) cemetery in Shangshihe (上石河) showed that the high-status individuals and the older adults (both males and females) consumed more protein-rich foods, but no sex-based dietary differences were found (Zhou et al. 2021b). At the same time, noble females shared similar mortuary treatments as noble males. Some scholars believe that this represented superficial equality, where females’ social status actually dependent on their husbands and their families of origin (e.g. Yang and Sun 2016; Yang et al. 2020).

Due to the very skewed sex distribution of the sacrificial group with a higher proportion of females and only three males, we cannot fully assess sex differences in diet among the sacrificial group. However, the sex and age distribution of the sacrificial group with a predominance of young sacrificial females, reflects the possible marginalization of females in Dahan society. Other bioarchaeological research has shown sex-based dietary differences in commoners or the lower classes during the Eastern Zhou period. The study on the ancient Zheng Han City indicates that the social status of females declined significantly during the Eastern Zhou period in the Central Plains of China, as evidenced by increasing sex differences in diet, height, and mortuary rituals (Dong et al. 2017). Similar sex differences in commoners have been found at many other Eastern Zhou sites, males almost all having higher δ13C values than females to varying extents, however, smaller differences exist in δ15N values (e.g. Ling et al. 2010, 2019; Pei et al. 2008; Tang et al. 2018; Tao et al. 2020; Zhou et al. 2017). Li and **g’s (2020) mortuary study on the commoner burials of the Qi State shows that males’ social wealth and status were overall higher than that of females as well. For the sacrificial class, fewer Eastern Zhou sites have been studied using bioarchaeological methods. Most of the sacrificial individuals at Eastern Zhou sites of Songzhuang (宋庄) and Liujiawa (刘家洼) were young females, and their diet was comprised of less socially valued foods (Wei et al. 2021; Zhou et al. 2019, 2021a).

Both carbon and nitrogen isotope values of noble females were significantly higher than those of sacrificial females, and sacrificial females also showed s higher prevalence of CO and PH, indicating that noble females consumed more C4 foods and meat. Some scholars have suggested that the roles of these sacrificial females were that of 女乐 (nvyue, females who entertain the noble class with their singing and dancing skills) (Gu 1982; Huang 1997, 2004; ** 1994; Mi 2014). Furthermore, historical studies show that the official prostitutes in ancient China began with the establishment of 女闾 (nvlv) in the Eastern Zhou period (Wang 2004a, pp. 25–40; Liu 2012), which also reflects the miserable fate of lower-class females during this time. Overall, females of the Eastern Zhou were in a subordinate position, and their social status was closely tied to their husbands and families of origin, and the lower-class females had an especially harder time than their noble female counterparts. It is worth mentioning that historical records show that a small number of females were involved in political and military activities (Gu 2005; Wang 2004b). A few Dahan noble females were also found with weapons, such as individual M122, who was buried with one sword, three dagger-axes, and one arrowhead, possibly suggesting that she was involved in military activities during her lifetime.

Looking at all the evidence from the Eastern Zhou period, there are sex-based differences in diet, health, and mortuary rituals, but these are specific to class. There are no significant sex differences in the diet of the noble class, while males in the commoner or lower class may have consumed more high-value foods, including C4 foods and meat compared with females of the same class. These class-specific sex differences are also reflected in mortuary rituals (based on the burial variables, such as grave goods). In recent decades, scholars recognize the fluidity and diversity of identity, and advocated that gender roles are not static, but can vary and change over time and space, and is intersectional and related to other aspects of identity (e.g. Geller 2009; Meskell and Preucel 2007; Sofaer 2006). Isotope and mortuary analysis among the Eastern Zhou sites, confirmed the class differences in sex-based food consumption and mortuary rituals. In particular, gender stratification in the Eastern Zhou likely had a greater impact on individuals of low status compared with high-status individuals.

Conclusions

In this study, we investigated the stable isotope values of the nobles and sacrificial companions in the Dahan cemetery, to examine the relationship among diet, class, and sex during the Eastern Zhou period. The results show that Dahan people had a mixed C3 and C4 diet, with some consuming large amounts of C3 foods (mainly wheat and soybeans), whereas some consumed more C4 foods (mainly millets). The amount of meat intake was also variable.

Class-based stratification had a significant impact on diet. Nobles had better diets with advantages in obtaining socially valued foods such as foxtail millet and meat resources, while the sacrificial companions generally had poor diets, being restricted in meat consumption and eating more wheat and/or soybeans. The relationship between the nobles and sacrificial companions varies among burials. For example, sacrificial individuals from burials M39, M57, and M208 may have benefited from the resources of the nobles in contrast to other burials. However, most sacrificial individuals may not have close relationships with the nobles as reflected in their large dietary differences. M209 may represent an individual who attained social status later in his lifetime potentially due to military achievements. There were sex-based differences in diet, but these were specific to class. There were no significant differences between noble males and noble females, whereas the lower-class males may have better diets compared with lower-class females. The predominance of young females buried as sacrificial companions also reflects the marginalization of lower-class females in the Eastern Zhou period.

This study has investigated the dietary differences between nobles and sacrificial companions during the Eastern Zhou period at a large cemetery-wide level. It provides direct and valuable insights into the influence of social stratification on the diet of Dahan population. This study demonstrated that the intersectional nature of social identities needs to be taken into account to better understand any differences in diet of these populations. Our study also shows that estimating an individual’s status solely based on mortuary rituals or stable isotopic data can be problematic. Social mobility among different classes was possible during the Eastern Zhou, and the social status of individuals can change during their life course. Future stable isotope analysis of the teeth (for diet during childhood) and ribs (for diet during the years before death) of special individuals will help to further confirm these findings.