Women,Rituals,and the Domestic-Political Distinction in the Confucian Classics

Abstract

In this article, I show that women are depicted in the early Confucian texts not primarily as undertaking household duties or nurturing children but rather as partaking in rituals of mourning and ancestor worship. To make the argument, I analyze, besides the more philosophical texts like the Analects and the Mencius, texts known as the “Five Classics,” which describe women in their social roles in much more detail than the former. What women’s participation in rituals reveals, I contend, is that the domestic-political distinction does little to illuminate the philosophical vision offered by the early Confucian texts. Relatedly, while women’s involvement in communal religious rituals has also been noted about early Greece, the political import of such participation is even more pronounced in the Confucian case. Specifically, I show that, by embodying intergenerational continuity, the mourning and ancestor rituals that women partake in are foundational to the Confucian state.

Scholarship on women in early China and early Chinese thought has, by emphasizing the continuity between the domestic and the political spheres in Confucianism, shown that women influence the political realm by helping instill virtue at home (Chan 2000; Goldin 2000; Raphals 1998; Rosenlee 2007). Even while rejecting a sharp distinction between the domestic and the political, the continuity argument presupposes a conventional conception of the domestic, one in which women take care of children and the household, and a conventional conception of the political, centered around the (male) business of government. But women are portrayed, in the early Confucian texts, primarily as participants in rituals of marriage, mourning, and sacrifice to ancestors, not as homemakers. Furthermore, mourning and ancestor rituals embody the value of intergenerational continuity, and this value is foundational to the Confucian conception of the political. Or so I argue in this paper. To make this argument, I analyze, besides the more philosophical texts like the Analects and the Mencius, texts known as the “Five Classics.”

Scholars writing on early Greece have also revealed women’s involvement in religious and ritual ceremonies central to the lives of their communities (Alexiou 2002; Goff 2004; Holst-Warhaft 1992; Honig 2009; Loraux 1990; Patterson 2007; Saxonhouse 1983), upending the conventional view of Greek women as relegated to the private sphere of the household (the oikos). I use this account of early Greece as a point of comparison to reveal the distinctiveness of the Confucian view concerning mourning and ancestor rituals, how these rituals complicate the distinction between the domestic and the political spheres, and the political import of women’s participation in them.

The paper is divided into three parts: In the first part, I discuss the role of women in texts like the Analects, the Mencius, and the Xunzi and make the case for a turn to a different set of texts, known as the “Five Classics,” especially the Rites Classics, which I will also say more about. In the second part, I illustrate women’s salient participation in all aspects of rituals of marriage, mourning, and ancestor worship as depicted in the Rites and underscore the rarity, conversely, of depictions of women in their household-related roles, including their roles as mothers. In the third part, I show how these rituals are important for the political realm, using arguments concerning women’s participation in rituals in early Greece as a point of comparison. What the concluding inquiry also reveals is that the domestic-political distinction does little to illuminate the philosophical vision offered by the early Confucian texts.

Women in the Philosophical Texts

References to women in the early Confucian philosophical texts are few and far between. They include, among others, the discounting of a woman official (Analects 8.20), the idea of the distinction (bie 別) between husband and wife and male and female (Mencius 3A.4; Xunzi 5:115), and the idea that the wife obeys the husband when he follows rituals (Xunzi 12:95). On the basis of a systematic analysis of the relevant references in the Analects and the Mencius, Chan (2000) concludes that the conception of gender is “primarily a functional distinction assigning women to inner/domestic duties, and men to outer/public duties” (115). The fact that all officials mentioned in these texts are men, and the explicit discounting of the female official mentioned in the preceding, makes it clear that women are not thought to belong to the government bureaucracy. On the other hand, the texts say little about the domestic activities of women—Chan glosses these as involving “nurturing the children, cooking . . . and other household work” (117).

While women in the Mencius are described as responsible for looking after in-laws and the aged (5B.5, 7A.22), and one passage in the Xunzi mentions mothers feeding their babies, ¹ women are not explicitly described in the early Confucian philosophical texts as doing child-rearing or household work. ² On the other hand, in a couple of passages, including in the reference to Xunzi mentioned in the preceding, women are associated with ritual activities. Take Mencius 3B.3; the passage goes as follows: “A feudal lord takes part in the ploughing to support the grain for sacrificial offerings. His wife takes part in sericulture to provide the material for sacrificial dresses.” ³ The first point to make here is that the idea of men farming and women weaving (nan geng nü zhi 男耕女織), to which I will return later, becomes a classic statement of the gender division of labor in early China. ⁴ What interests me here, however, is that in addition to this passage encapsulating a “paradigmatic instance of the gender distinction,” as Chan (2000) points out, it also encapsulates how both husband and wife participate, in a complementary fashion, in “the important rite of making sacrifice” as she also points out (118). Or take Mencius 6B.6, which says that “The wives of Hua Chou and Ch’i Liang, being supreme in the way they wept for their husbands, transformed the practice of a whole state.” Chan (2000) uses this passage to show that women are “capable of moral judgments” and that their “domestic role should not hinder them from attaining the [highest ethical] ideal of chun-tzu” (124). Note here that what is getting glossed as domestic is the practice of mourning.

While the philosophical texts help us ponder the importance of these rituals in general, a topic to which I turn in the last part of the paper, they do not provide details about how these rituals function and about women’s roles in them. But they do give us a clue as to where to look for such details: the passage cited in the preceding from Mencius 3B.3, with men farming and women weaving, is actually presented as a quote from the Rites (li 禮). Another passage from the Mencius (3B.2), cited by Chan and which mentions a mother’s advice to her marrying daughter not to disobey her husband, also attributes that idea to the Rites. While li 禮 is also a reference to rituals in general, the second passage, as will be shown later, finds a strong correspondence in the Ceremonials (Yili 儀禮), while the first passage finds echo in both the Zhou Rites (Zhouli 周禮) and the Rites Records (Liji 禮記), ⁵ the three forming the sanli (三禮) or the “three Rites canons” as grouped together by the Eastern Han commentator Zheng Xuan (127–200) (Nylan 2001, 174).

It is to these Rites canons that I propose to turn in what follows. I do not mean to suggest that texts like the Analects and the Mencius straightforwardly build upon texts like the Rites such that one could match the former to the latter one-to-one; the challenges of dating these texts militate against any attempt to establish simple lines of influence. While the philosophical texts are conventionally thought to be products of the Warring States period (479–221 BCE), scholars of early China have argued that they have been edited and put together in their received form during the Western Han Dynasty (206 BCE–9 CE). On the other hand, the Five Classics, which include, in addition to the Rites, the Odes (Shijing 詩經), the Changes (Yijing 易經), the Documents (Shangshu 尚書), and the Annals (Chunqiu 春秋)—a chronicle of events in the state of Lu in the Spring and Autumn period—are thought to have been adopted as the basis for official learning starting with the Western Han yet their content has been traced to different periods: for example, the Yili is traced to the Warring States period (Boltz 1993; Nylan 2001), the Zhouli to the “late Warring States . . . or very possibly imperial Qin dynasty (221–206 BCE)” (Elman and Kern 2009, 1), and the Liji to the Eastern Han (25–220) (Boltz 1993; Nylan 2001). ⁶ My exploration of the Rites canons is thus only meant to illuminate the philosophical texts in a loose sense, focusing on a set of ideas about rituals that are consistent with these texts and gesturing toward changes in time as the Warring States period is brought to an end through the imperial unification efforts first of the Qin and, more lastingly, of the Han.

Lest it is thought that by focusing on texts about rituals, my inquiry is predetermined to lead to an argument about rituals, I should finally clarify that only the Yili is wholly focused on rituals. The Liji includes, in addition to much commentary on rituals, discussions of virtue broadly understood, education, punishment, and even territorial organization (see Nylan 2001, 185–188). The Zhouli consists of an idealized description of government during the (Western) Zhou dynasty; it is divided into six sections, only one of which concerns rituals (the others concern general administration, education, military affairs, penal affairs, and crafts) (Boltz 1993, 24). Furthermore, I will bring in the other Classics as relevant to the discussion at hand.

Women in the Rites

What I aim to do in what follows is to pick up on references made in the previously mentioned passages of the Mencius to the Rites canons and use these as an entryway into the latter texts, exploring how women are described in them. I will show that women are represented as full participants in rituals of marriage, mourning, and sacrifice and that they are not relegated to secondary or female-only tasks. I will also show that the various Confucian texts under consideration rarely describe women in their motherly or otherwise household-related roles. The exception here is weaving, but weaving is also ritual-related and should not be described as domestic in any case—I will elaborate on the latter point in the third part of this paper.

I start with the Yili, which details the rituals of the shi 士 (variously translated as “literati,” “scholar-officials” or “men-in-service”), the lowest members of the aristocracy. These include rituals of marriage, mourning, sacrifice, interstate visits, capping ceremonies for coming-of-age young men, visits between shi, archery conquests, banquets, drinking ceremonies, and imperial audiences. Women make appearances in the first five, and only to a significant extent in the first three of these, namely sacrifice, mourning, and marriage rituals.

Rituals of sacrifice are directed at ancestors, and usually referred to as “ancestor worship,” although “worship” does not quite capture what is at stake in these rituals, namely the “maintenance of family ties through ritual means in the hope that one’s forbears might continue to aid their descendants” (Bokenkamp 2002, 388 cited in Cook 2009, 237n1; see also Brashier 2011). They take place in temples, and the corresponding ceremonies involve, as do mourning rituals, divination to determine the best day for burial or sacrifice, various impersonation rituals in which the living personate the dead, and various food and drink offerings among the ranks of the participants.

Participants in rituals of mourning and of sacrifice are both male and female, with the women playing no less of a central role than their male counterparts, or so I aim to show in what follows. For example, when divination is undertaken to determine the best day for burial, the result is announced both to the head of the household (zhuren 主人) and to his wife (zhufu 主婦) (§XII, XXVIII, 19: m-n). ⁷ During the preparations for sacrifice, the hostess is in charge of the cooking stoves while the host watches the sacrificial animal being killed (§XV, XXXIV, 6: a-b). During the ceremonies, they both place and present the food (albeit in different manners), both offering wine to the personator of the deceased (§XV, XXXIV, 8). In the sacrifice ceremony for low-rank shi, one of the female relatives offers chestnuts and dates (§XV, XXXV, 6: c) and the senior among the male relatives offers roasted meat (§XV, XXXV, 6: e). Then, the hostess offers wine to the host and he reciprocates, after which the guests offer it to both (§XV, XXXV, 6: 9–11).

In the various stages of these rituals, men come first: the host before the hostess, the male guests before the female guests, and the male ancestor before his wife. ⁸ But women are not relegated to less important or less visible aspects of these rituals (such that it could be argued that they play a secondary or subordinate role in them). The gender hierarchy is counterbalanced by hierarchies of rank and age: for example, the hostess receives a part of the same cut of the pig that the host, the liturgist, the oldest of the brothers, and the temple keeper get (§XV, XXXVI, 8: g-k), while everyone else gets a different cut (§XV, XXXVI, 8: l). At the other end of the hierarchy, it is not necessarily women who occupy the lowest positions: the only instance in which women are singled out for the seemingly lowest task is when they remove the food from the room at the very end of the sacrifice ceremony as undertaken by low-rank shi (§XVII, XXXIX, 43: c-d). Even then, however, this is preceded by assisting officials (siguan 司官)—ostensibly men—moving the offerings away (§XVII, XXXIX, 42: a). Women are not even singled out for crying; men stomp and cry too (§XIII, XXIX, 5:g). While the example from the Mencius (6B.6) that Chan cites mentions two wives’ virtuous and inspiring weeping for their husbands, this should not be taken to suggest an association between women and mourning lamentation here–-as is true, at least with regards to “excessive mourning,” of Greek texts “from Solon to Pericles and beyond” (Honig 2009, 11; see also Goff 2004, 31).

The Yili’s account of ancestor and mourning rituals does reveal, as the preceding shows, that women’s roles are distinguished from those of men. Furthermore, women are grouped together: for example, if it is the mother who has died, servants from the women’s apartment wash the corpse (§XIII, XXXI, 8: c). ⁹ In ancestor ceremonies, men personate men and women women (§XIV, XXXIII, 5: b). Finally, the texts specify where men stand and where women stand (§XV, XXXIV, 7: a; §XVI, XXXVII, 9: b-f). With all of this said, the two groups are not segregated. They clearly interact in all aspects of the rituals. There is no suggestion that men and women should not occupy the same space, let alone that temples are the domain of men alone. ¹⁰

In marriage rituals too, women, as mothers and mothers-in-law (not just as brides), take center stage. While it is the father who instructs the messenger who makes first contact between the groom’s and the bride’s families, if the father is dead, it is the mother who gives the instructions to the messenger (§II, IV, 8:15a). The bride is counseled by both her father and her mother (§II, IV, 8:14d). The wedding rituals focus on the introduction of the bride to both of her in-laws: thus the morning after the bridal ceremony, the bride is introduced to both and serves both food (§2, IV, 1–5). If the in-laws are dead, she goes to the ancestral temple and sacrifices to them (§II, IV, 6–7). When the groom visits the bride’s house later on, he pays his respects to both of her parents. He is invited to drink by the father and is presented with food by the mother (§II, IV, 8: 16g-k).

Note that the passage just referenced about soon-to-be brides being counseled by their parents mentions the need for women to be obedient but leaves the object of the obedience (in the counsel by the father) unspecified as ming 命 (or instructions) and (in the counsel by the mother) specified as gongshi 宮事, which could mean duties of either household or temple (§II, IV, 8:14b-c). This is in contrast with the Mencius passage (3B.2) cited by Chan, which mentions mothers advising their daughters as they go out to get married, on the authority of the li 禮, that they “should not disobey their husbands” and which concludes with the statement that “it is the way of a wife or concubine to consider obedience and docility the norm.” Nowhere is there a statement about obedience being the way of wives in the Yili.

Statements about obedience and docility are much more in line with ideas laid out in the Liji, which, as previously stated, is thought to have been composed much later (around 100 CE) and during a very different historical context. The Liji indeed lists faithfulness (xin 信) (IX, §III, 7) and obedience (shun 順) as the virtues of the wife (XLI, 10). ¹¹ Another reference in the Mencius to the rites proscribing men and women from touching each other (which Mencius offers an exception to) (Mencius 4A.17) is also echoed in the Liji (III, 6: 31–33; XVIII, §II. Part II, 19; see also XXXI:30). The Liji even mentions that women should cover their faces when they go out (X, §I:12). But while the Liji is clearly more invested in emphasizing the difference between men and women (nan nü zhi bie 男女之別) (XIV, §I.5; XLI, 3; XVII, §I, 13; XXIII, 7), and while it uses the language of “inner” and “outer” (nei wai 內外)—to which I return below, it primarily and substantially features women in discussions of mourning rituals, as well as sacrifice and marriage rites (V, §I, 20; X, §I, 20; XII, 11; XV, 26; XVIII, §I, Part II, 25; XXVII, 38). ¹² While, like the Yili, it assigns women distinct roles and positions in these rituals (II, §II, pt II.5; XVIII, §II, Part II, 38; XIX, §I, 5–7; XXI, §I, 6; XIX, §I,13, 20; XXII, 4; XXXI, 10), there are explicit mentions of the necessity of women’s presence in them (V, §I.7). Women are also not said to be physically segregated from men during ritual ceremonies, drinking and eating together (IX, §III, 7); indeed one passage suggests that it is precisely—indeed only—in sacrifice rituals (ji 祭) that women and men are allowed to exchange drinks (XXVII, 35; see Raphals 1998, 224). Again as in the Yili, wailing is performed by both men and women (Bk II, §I, Part III, 45, XIX, §I, 5) and regulated for both (II, §.I, Part I, 28). I should also mention here that a few passages in the Liji associate women with weaving and dyeing (IV, §I, pt III, 12; IV, §II, pt I, 19; IV, §II, pt III, 9), supervised by the ruler’s wife in her palace (XXI, §II, 7), while the ruler is responsible for farming (XXI, §II, 6)—more on the weaving versus farming metaphor below.

In the Zhouli, almost all of the (explicitly) female positions occur in its first section, where they make up the administration of the queen’s palace: there are women responsible for prayers (nüzhu 女祝), mostly for sacrifices undertaken by the queen (I:134), and women scribes (nüshi 女史) who write out orders by the queen and accompany her in ritual ceremonies (I:134). ¹³ Tasks associated with women also involve the reception of foreign visitors and funerals. Besides the positions in the queen’s palace, the general category of artisans also includes women (nügong 女工) (I: 51). Women do seem to be specifically associated with lamentations (I:131; III:77)—although it is not obvious that all lamentations mentioned elsewhere (for example in III: 94, 95, 104) are undertaken by women. ¹⁴ Be that as it may, women in the Zhouli, as I have just argued in the cases of the Yili and of the Liji, are associated first and foremost with rituals of sacrifice—in the Zhouli’s case, as a continuation of the queen’s role in these. They are not associated with child care or other domestic chores, even though the Zhouli lists descending ranks of women operating under the queen (one might have expected that at least some of these would be said to be in charge of the young princes).

Staying with the topic of child care and returning to the Liji, a few paragraphs of the latter, unlike anything in the Yili or the Zhouli, are actually devoted to the birth of a child (X, §II, 16–31) and include a discussion of the proper animal to sacrifice for the occasion and the selection of women appointed to care for the child—literally “feeding mothers” (shimu 食母). ¹⁵ The passage clarifies that the son of a great officer (dafu 大夫) had such a nurse but that the wife of shi fed her own child (X, §II, 30). There are a few other references in the text to cimu 慈母, “wet nurses” or (given that their role seemingly extends beyond nursing) “foster mothers” (V, §I, 26; XIII, §II, 13). ¹⁶ On the one hand, these references to wet nurses suggest that the dearth of other representations of women nourishing and raising their children can be attributed to the fact that these tasks were delegated to lower-class women, with whom the texts at hand—whether the philosophical texts or the Rites canons—were simply not concerned. On the other hand, the preceding passage indicates that women of the shi class—precisely the class that much of these texts are in fact concerned with—nurse their own children.

In any case, it is clear that the themes of birth, reproduction, and childcare make rare appearances in the Classics. ¹⁷ The stated purpose of marriage, in the Rites canons (Yili §II, IV, 8:13a–b; Liji I, § II, Part III, 11) and in the commentaries on the Annals (Zuozhuan: Wen 2.7; Guliang: Zuang 24.1), is primarily ancestor sacrifice, not reproduction per se. Also related is the rarity of passages that detail specific motherly dispositions: Chan cites a passage in the Liji that associates mothers with affection (and fathers with reverence [zun 尊]) (XXIX, 29). The Zuozhuan includes a passage describing fathers as dutiful (yi 義) and mothers as kind (ci 慈) (Wen 18.7b). And one Ode mentions mothers being depended on (yi 依) (and fathers looked up to [zhan 瞻]) (#197, repeated in #202). Both Erin Cline and Anne Kinney have argued that childbirth (and childhood) emerges as a significant theme in early China only during the Han dynasty (Cline 2015, 66; Kinney 2004, 9).

The lack of emphasis on maternal roles might suggest that these roles were so central that their importance went without saying. ¹⁸ The same would be true of household work more generally. Note that this alternative reading, in which the texts’ lack of focus on domestic activities is simply an indication of these activities being taken for granted, depends on a prior—and reasonable—assumption that women perform all these tasks (the absence of references to women serving in government would not be similarly taken to be a sign of this service’s ubiquity). It might thus first help here to clarify what I take the goal of this paper to be: my aim is not to reconstruct women’s roles in early China but rather to ask what image of the exemplary (often elite) woman these (elite) texts present. The question is not whether women were mothers or took care of their houses but rather whether this is the primary image that the prescriptive texts of the period chose to project about them.

Even then, the question remains whether the texts’ focus suggests a normative preference (i.e., that the texts focus on women’s participation in rituals because they view it as more important than domestic work) or a merely strategic choice to focus on those roles that are more likely to be overlooked—without being necessarily more important. Note here that the Liji, an Eastern Han text, does emphasize certain aspects of gender, especially the physical separation between men and women, more than earlier texts do. Does it do so because physical separation becomes important in the Han, or because people had just become lax about it, such that what was always important could no longer be taken for granted? Note also that, to the extent that the early Greek texts are more likely to associate women with domestic roles, including baking, childbearing, nursing, and the maintenance of the household (Saxonhouse 1985, 37–91), the question would also arise as to whether they do so because these activities could not be taken for granted, as they were—on the alternative reading—in the Chinese case. In short, this line of reasoning raises more questions than it can answer.

In any case, the way in which the texts gloss the purpose of marriage lends support to the idea that the prioritization of ritual is a normative, not strategic, choice. Specifically, the texts characterize marriage as being for the sake of sacrifice to ancestors, not reproduction per se, which stands in contrast to what Goff (2004) argues is the purpose of marriage in the Greek case—namely “to produce heirs for the husband’s line” (30). ¹⁹ Relatedly, mourning practices in the Rites Classics partake of rituals of ancestor worship, not those concerning childcare—which have barely any presence in these elite texts. Goff’s assertion that birth and death are viewed as analogous in Greece as in “many other cultures,” and that this analogy is what sets the stage for women’s role in funerals (31), does not readily apply to the Confucian texts surveyed here.

It is thus on the basis of such explicit, and distinctive, ideas about the purpose of marriage—and the purpose of family and filial piety, to which I will turn later—that I interpret the dearth of references to childcare as a result of a normative prioritization, making women standard bearers of and central participants in rituals that commemorate the deceased. Women are also bearers of children, but the bearing and rearing of children were potentially viewed as aimed at such commemoration. Indeed, the rearing of children can be delegated by (some) elite women to women of lower ranks, whereas commemoration is not to be similarly delegated. The passage in the Liji about the birth of a child, referenced in the preceding, continues with the—distinct—education of girls and boys. ²⁰ While boys learn reading and calculation, as well as music, dance, archery, and chariot driving, girls are taught the following by their governess (mu 姆) at the age of ten:

the arts of pleasing speech and manners, to be docile and obedient, to handle the hempen fibres, to deal with the cocoons, to weave silks and form fillets, to learn (all) woman’s work, how to furnish garments, to watch the sacrifices, to supply the liquors and sauces, to fill the various stands and dishes with pickles and brine, and to assist in setting forth the appurtenances for the ceremonies. (X, §II, 36–7)

Women’s education, this passage makes clear, centers on education in sacrifice rituals and the weaving work associated with it. In her commentary on this passage, Rosenlee (2007) glosses this as the “realm of domestic skills and household management” (82) but—as she also goes on to wonder (85)—does the idea of the domestic, understood as household management, really capture what is described in the passage? I turn to this question in the third and last part of the paper.

Rituals, Intergenerational Transmission, and the Domestic-Political Distinction

The account I presented above of women’s meaningful participation in sacrifice and related rituals pushes against understanding women’s idealized role in the early Confucian texts as restricted to a narrow domestic realm, wherein they raise children and take care of household chores. Instead, their duties extend to taking part in rituals that ultimately center (even in the case of marriage) on temples and their commemorative function. The question that remains, and which this section addresses, concerns the upshot of this argument: in what way do rituals of sacrifice matter socially and politically, and in what way, by implication, is women’s participation in them important? Furthermore, to return to the questions I asked at the end of the preceding section, how should we think of these rituals in relation to the distinction between what is domestic/private and what is political/public?

I will argue that rituals of mourning and sacrifice are foundational to the Confucian political community because they embody the concern with intergenerational transmission or, more simply, lineage. I will first start, however, with the contention that what women’s role in rituals helps us see is that the domestic-political distinction sheds little light on the philosophical vision offered by the early Confucian texts, for the rituals they participate in do not fit into either of these categories nor are they best understood in relation to the distinction between the two.

The comparison with the early Greek case is instructive here. The distinction between the political community, or polis, and the household, or oikos, is foundational for early Greek thought. In early Greek poems and tragedies, this opposition “was often expressed in sexual images of the male warrior and the female householder” (Saxonhouse 1983, 372). ²¹ Aristotle’s famous philosophical definition of the political is made precisely in contradistinction with what relates to the household: in the first book of his Politics, Aristotle uses this opposition to elicit the distinguishing traits of political relationships. The distinction is, relatedly, crucial for understanding the place of women: women appear in the Politics in Aristotle’s discussion of the oikos; a more elaborate treatment of the oikos can be found in Xenophon’s Oeconomicus, which is concerned with wealth and property management and figures women within this inquiry.

With this said, as I mentioned previously, scholars have argued against the idea that the ancient Greeks relegated women to the domestic sphere by revealing women’s involvement in religious ceremonies central to the lives of their communities. The crux of the argument lies in the identification of these ceremonies as public. Pomeroy (2011) thus argued that “Religion was the major sphere of public life in which women participated” (75). Patterson (2007) also views women’s participation in religion (as well as in matters relating to property and inheritance) in Periclean Athens as participation in “key areas of public life” (169). Goff (2004) argues that rituals allowed women “public voice” and “enabled them to organize aspects of their lives independently” (2).

Note that many of these rituals, while public, build upon women’s roles at home. Goff (2004) argues that “ritual events and structures often purveyed and confirmed traditional, restrictive notions of female identity” (4). The argument here relates to how women participated in rituals related to fertility and, by association, agriculture (3; see also Pomeroy 2011, 78) as well as childbirth. Women also participated in mourning rituals, with death considered “particularly suitable for women’s management” (Goff 2004, 31) because it was viewed as analogous to birth (Pomeroy 2011, 44). Furthermore, women’s role in various rituals “dramatizes” their domestic roles, including weaving and the carrying of water (Fantham et al. 1995, 96; Goff 2004, 4). Some of the rituals Goff (2004) discusses are also women only, and others involve the display of unmarried women (5).

In short, while the identification of the rituals that women participate in as public suggests a transcendence of the household, or oikos, the relevant rituals also depend on the oikos and its definition of women’s roles. Furthermore, even while they are public, the question arises as to whether these rituals can be categorized as part of the political community, or polis, proper, since the polis is defined, as is suggested in Aristotle’s analysis, through activities centered on the legislative roles of (male) citizens. Goff (2004) describes the rituals in which women participate as “parapolitical” (6). Blundell (1995) argues that “Oikos and polis were linked together by a common thread of religious observance” (160), thus avoiding assigning religion exclusively to either realm. Pomeroy (2011) argues that in Athens, “cult was subordinate to” the state—while also viewing the former as an “integral part” of the latter (75). Historical instances of rituals being the subject of state legislation provide further evidence of cult’s subordination to the state: Alexiou (2002) has argued that restrictions on laments in funerals, particularly on the part of women, were a sign of “a new society, often culminating in democracy, . . . establishing itself” (17); according to her, reforms to these rituals in Athens “involved a gradual transfer of ritual, and of all the motive feeling attached to it, from the ancestor of the clan cult to the hero of the state cult” (19). In short, in the Greek case, religious rituals complicate the distinction between oikos and polis, potentially by suggesting a public realm distinct from the sphere of the polis proper, without—and this is the crucial point here—undermining the relevance of the distinction. Indeed, this religious public realm, while separate from the operations of the state proper, is reminiscent of the oikos and arguably subordinate to the polis.

In the Chinese case, rituals that women participate in do not emphasize distinctive female roles, let alone “domestic” ones. Even though women are restricted to a subset of rituals (they do not take part in archery contests, for example), this subset does not explicitly include childbirth, nor is tending to, and mourning the dead, presented as a particular realm for women, let alone for its association with childbirth. While in the Yili women are in charge of cooking the sacrificial animals, they are not assigned roles such as crying or cleaning, reminiscent of a distinctive sphere of private activity (which is then brought out in the open during such rituals, as in the Greek case).

Zhou (2010) contrasts the Chinese “home where one engaged in daily practices of kinship-centered moral precepts and religious ceremonies” as “the site for the most fundamental education in Zhou society” with “the sacred grove, the gymnasium, the wrestling school, and the symposium” of the Greeks (147). But note that it is not the home per se but the ancestral temple that is the site of much of the central ritual activity discussed in this essay. For Zhou, the ancestral temple belongs to the “domestic ritual space” because it does not involve “extrafamilial homosocial bonding” (150). While true, it is important to note here that the family of concern is the extended family or clan. As the descriptions of rituals in the second half of this essay indicate, participants were at different degrees of kinship from each other. Furthermore, as just suggested, the rituals performed in the temple do not reflect activities that are related to the household. Indeed, temple-centered ritual activities are the kinds of activities also undertaken by the ruler. As Zhou points out, earlier on during the Zhou period, the ruler’s ancestral temple was “the site for the full gamut of events and ceremonies related to the state’s political and military affairs” such as appointments and calendars (113). In the texts from the Warring States period discussed in this paper, the ancestral temple, and the sacrifices associated with it, continue to be described as crucial to the state; this is clear from the Zhouli’s inclusion of rituals, especially those focused on sacrifice, as part of the first chapter on “general administration.” As the Zuozhuan puts it, “The great affairs of the domain lie with sacrifice and warfare” (Cheng 13.2). The philosophical texts also all associate the ruler—not in his personal capacity but explicitly in his capacity as a ruler—with ceremonial sacrifice, ceremonial attires, and ancestral temples. ²²

I take the preceding to suggest that the whole domestic-political distinction does not help make sense of rituals central to the Confucian political community and as such does not clarify how this community works more generally. It is true that the early Chinese texts use characters that are usually translated as private and public—respectively, si (私) and gong (公)—and it is true that the pair becomes important during the Warring States period (Brindley 2013, 3; see also Nylan 1996, 7 and Sato 2020). However the two do not relate to distinct spheres of activity, but rather to two kinds of ethical disposition, both of which are used to describe, among others, governmental affairs: the contrast is between what is “private, unofficial, partial, particular, individual, and divided” and what is “objective, impartial, universal, impersonal, and unified” (Brindley 2013, 2). ²³ This contrast does not, furthermore, shed light on the place of women, since there is no association between women and what is si. ²⁴

One could point to another pair of concepts—namely, gong and jia (家), with jia meaning family, household, or clan. This pairing is deployed in the “Liyun” chapter of the Liji, which states that when the proper path is followed, all-under-Heaven will be gong (tianxia wei gong 天下為公), whereas when it is not, all-under-Heaven will be jia (tianxia wei jia 天下為家) (VII, §1, 2–3). The contrast here is reminiscent of the contrast between impartiality and partiality in that all-under-Heaven is either treated as an object of universal care or as belonging to the ruler and his family. Jia is not typically used to signal women’s distinctive realm, nor is it typically contrasted with gong. ²⁵

A final relevant distinction is the one to which Chan refers—namely, the one between what is “inner” (nei 內) and what is “outer” (wai 外). This distinction becomes systematic, as well as systematically gendered, only during the Han dynasty, as Raphals (1998, 195–234) and Rosenlee (2007, 72) have argued. ²⁶ It is central to the Liji—recall that the Liji is a Han text. It is arguably no coincidence that with administrative centralization—following the period of imperial disintegration and fluid state boundaries that characterize the Warring States period—come clearer demarcations of various realms of social life, ²⁷ as well as of various social categories (including gender). ²⁸ This would explain why it is in the Liji and the Guliang ²⁹ —that is, in Han texts, and thus later than the other texts discussed in this essay—that we find a concerted emphasis on women’s physical separation from men and on an inner-outer distinction that distinguishes women and men. ³⁰ Note again, however, that, as I mentioned previously, physical separation does not apply to the realm of rituals, even in the Liji. ³¹ Furthermore, notwithstanding the emergence of an inner-outer distinction, it might still be worth asking whether it amounts to anything like the Greek distinction in which the political realm is defined precisely through its difference from the realm of the family.

Here, it might be helpful to return to the idea, mentioned at the beginning of this paper in relationship to the Mencius, that “men till and women weave.” ³² This becomes a classic statement of the gender division of labor during the Han Dynasty when “the figure of the female silk weaver” becomes “exemplary” (Chin 2014, 194). Not only does this trope potentially suggest a domestic-public dichotomy, since weaving is done at home and tilling outside, but it is also reminiscent of the Greek association between women and weaving: weaving was considered, in early Greece, “the quintessential female accomplishment” (Blundell 1995, 141) and was the central activity associated with women on vase paintings (Fantham et al. 1995, 103). My contention, however, is that the resemblances between the two cases hide more differences than might first appear.

In the Greek case, the weaving woman is not contrasted with the tilling man; farming in early Greece was mostly done by slaves. The weaving woman was ostensibly an elite, free woman whose husband was probably a small landowner. The latter participated in the polis as a citizen-legislator. The woman’s weaving was part of the oikos and aimed at clothing the members of the household and decorating the home; Blundell (1995) writes that “some households would have been completely self-sufficient in this respect, even producing their own wool” (141). Most farming in early China was, by contrast, done by free men (Yates 2001, 315). Furthermore, weaving in the Chinese case was part of the larger economy, including, as mentioned previously, for the production of ceremonial dress. Bray (1997) argues that “Although men worked outside the house and women inside, both were thought of as equally productive members of society,” contributing to “the welfare of the common people and the strength of the state” (183). ³³ She also cites Michel Cartier’s argument that weaving was not only for the household itself, to ensure basic needs and self-sufficiency, but that it also created “surplus value” (184, citing Cartier 1984). Chin (2014) mentions debates earlier during the Han period that emphasized women’s roles “as industrious workers in the expansionist economy” and that were only later overtaken by the emphasis on women’s “exemplary ethical roles as wives and mothers” (194) and concomitantly on the moral exemplarity of the female weaver.

While elite women in early China did not need to weave for economic benefit, if weaving is generally considered a productive activity (at least earlier on), then it is hard to view it as domestic, at least in the way that weaving readily signals the sphere of the oikos in the Greek case. This contrast perhaps pushes the question back to the relationship between the oikos and the larger economy in early Greece, but this is precisely the point: our understanding of the activities in which women partake depends on our interpretation of how the economy at large—in the case of weaving—works. The importance of women’s role in rituals in both early China and early Greece similarly depends on the broader relationship between rituals, society, and the state. It is this broader relationship that sheds light on the question of whether the inner-outer distinction translates into a private-public distinction. My concern is that arguments, mentioned at the outset, about Confucian government being an extension of the family, and about women making good subjects by making good children through their rearing and educative roles, ³⁴ operate by accepting this distinction without probing its bases. While similar arguments have been made about early Greek thought, wherein scholars have shown that women contribute to (Salkever 1986), help define (Saxonhouse 1983, Saxonhouse 1986), and are governed by the polis (Dietz 1985), it could be argued that the operating distinction between polis and oikos, which they complicate (rather than reject), is already well-established.

Let me suggest, finally, one reason why the private-public distinction is not readily applicable to the rituals discussed in this essay: lineage. It is the lineage dimension of family, and the concern with intergenerational transmission that undergirds it, that motivates the emphasis on mourning and ancestors’ rites, over any other rites, at the level of the state and the level of (extended) family units. And it is to this argument to which I turn in closing.

Consider the concept that is universally believed to be foundational to Confucian thought: xiao 孝, usually translated as “filial piety,” embodies the reverence children owe their parents. Three aspects of filial piety are important to note for my purposes here: first, in all of the Confucian texts of concern, filial piety is directed toward fathers and mothers (fumu 父母). ³⁵ This fact pushes against the argument that Confucian filial piety is a patriarchal notion, useful politically because it trains in obedience “to the patriarch” and then to the ruler as Knapp (1995) argues (221). ³⁶ While this argument would complement the understanding of women’s roles as being focused on child-rearing and household chores, the emphasis on both fathers and mothers makes clear that filial piety does not presume a patriarch. ³⁷ The second relevant point to make about filial piety is that the Confucian philosophical texts emphasize it significantly more than they emphasize the other side of the relationship, namely parental love. ³⁸ This fits with filial piety’s extension to both fathers and mothers, indicating that the emphasis is on the generational hierarchy more than the gender one. ³⁹ Finally, filial piety is expressed not just as respect for living parents but very much also in mourning and its accompanying rituals, as shown in the examples from the philosophical texts that Knapp discusses (Analects 1.11, 2.5, 14.40, Mencius 3A.2)—and as the discussion of mourning rituals in this paper reveals. As Mencius puts it, “Keeping one’s parents when they are alive is not worth being described as of major importance; it is treating them decently when they die that is worth such a description” (4B.13). Furthermore, while Knapp emphasizes “extended self-deprivation” in mourning for parents (214), what is also distinctive of this mourning is another aspect that also comes out in his discussion, at least of the Analects, namely the maintenance of the way of deceased parents unchanged during this extended period (Analects 4.20, 19.18).

My suggestion is thus that filial piety centers on preservation and continuity, on revering parents as a way to revere intergenerational transmission. Nuyen (2004) has argued that the father, in the Confucian view, “is not an independent figure of authority,” but “the father has his own father and the latter his and so on,” such that the father is “a father figure representing tradition” (209). Nuyen’s argument might be reformulated as both father and mother representing tradition, with respecting both of them being a way of respecting tradition. Ames (2011) argues similarly that filial piety can be viewed not just as respect for one’s own parents but also as respect for the whole lineage that one’s parents represent (233). This concern for intergenerational transmission would explain the lack of emphasis on motherly affection, or parental duties more generally (in comparison with that on filial piety), ⁴⁰ as well as the focus on mourning rituals and the maintenance of ways of life after death (compared to the significantly lesser emphasis on rituals of birth and reproduction). It is not that tending to the young is not important—who else would care for the old?—but rather that what it means to care for the young is precisely to teach them to venerate and continue what came before.

It is true that scholars of early China have pointed to the decline during the Eastern Zhou period of the role of lineage and the rise of social mobility outside of kinship structures (Hsu 1965; Li 2013, 171–75). But this decline does not completely subordinate kinship to the state during the Warring States period. For the decline concerns what Gassmann (2009) calls the “geneatactical” system of the Zhou, which involves a certain pattern of assigning ancestors, and not genealogy more generally (94–95). Cook (2009) also shows that ancestor worship does not disappear but gradually transforms, with ancestors’ sphere of influence shifting from the “clan” to “an extended-family unity in a single location” (241). In short, while some rules of allegiance and filial piety changed during this period, the importance of the family and of genealogy does not disappear, nor does its primacy vis-à-vis the state. As Zhao (2015) argues, “all of the territorial states that emerged during late Eastern Zhou . . . had developed from lineage-based city-state predecessors without neck-breaking political or societal disruption” (92); ⁴¹ he concludes that the state in China was always “shaped by Confucianism’s kinship-based understanding of . . . politics” (92).

In my interpretation here, the Confucian “kinship-based” conception of the political means not that the family gets deployed for strategic purposes by the state, as when emperors choose to present themselves as fathers to their people—ostensibly because family images resonate. I understand it rather to signal the importance of family as lineage, embodying the goal of intergenerational transmission, which continues to be an important end of the state; this is evidenced in the focus in texts of the period on filial piety, reverence to the past, ⁴² tradition, ⁴³ and the upkeep of rituals—sacrifice and mourning rituals in particular—as goals of government. As Fingarette (1972) puts it, the Confucian vision is a “political vision” of “an emerging unity among men” as well as “a philosophical vision, even a religious one” centering on “community as rooted in the inherited forms of life” (69).

Instead of homology between the state and the family (Zhou 2010, 17), my suggestion thus is to view the state as a vehicle for the family understood as lineage. ⁴⁴ If we see the family in this light, rather than in the light of a sphere of the domestic, then women stand out not primarily in their capacities of bearing and minding children but in their capacities of bearing and minding tradition; furthermore, allowing tradition to flourish in this way is a goal with which the government itself is entrusted. Sacrifice rituals are not subordinate to other state goals; they are central to the purpose of having a political community. By partaking in them, women partake in activities foundational to the Confucian conception of the political.