Hostname: page-component-cd9895bd7-dk4vv Total loading time: 0 Render date: 2024-12-25T17:10:08.095Z Has data issue: false hasContentIssue false

WANTON GODDESSES TO UNSPOKEN WORTHIES: GENDERED HERMENEUTICS IN THE CHU CI ZHANGJU

Published online by Cambridge University Press:  12 June 2018

Monica E. M. Zikpi*
Affiliation:
Monica E. M. Zikpi 馬思清, University of Oregon; email: mzikpi@uoregon.edu.
Rights & Permissions [Opens in a new window]

Abstract

The influential Chu ci zhangju 楚辭章句, the earliest received edition of the foundational poetry anthology Chuci 楚辭, performs a distinct gender bias in its exegesis of deities, and this bias accords with the Eastern Han ideology of the editor Wang Yi 王逸 (2nd c. CE) more than with immanent features of the original Warring States texts. The gender bias is an essential feature of Wang Yi’s canonization of the Chuci, and it lays the groundwork of the allegorical tradition of interpreting the Chuci. This paper analyzes the zhangju presentation of archetypal Chuci texts to elucidate the hermeneutic transformation of gender and religion in early China, comparing the Eastern Han exegeses with earlier and later interpretations, immanent textual features, and fresh perspectives on Warring States and Han culture that have emerged from archeological evidence. The analysis demonstrates that the Chuci zhangju treats the male deities more literally than the female deities, reflecting the reduction in status of goddesses in late Han discourse. The history of gender ideology is an essential critical lens for understanding the Chuci and the tradition that emerged from it.

提要

著名的《楚辭章句》是現存最早的《楚辭》傳世注本,其對於神靈精怪的解釋有明顯的性別偏見。編者王逸的偏見符合東漢的意識形態,卻與戰國的《楚辭》正文內在意義有所落差。《楚辭章句》的性別偏見也奠定了《楚辭》寓言闡釋傳統的基礎。本文在比較東漢註解與早期詮釋、後世註疏、文本內在特徵、現代考古研究等的基礎上,闡明古代中國性別與宗教觀在詮釋上產生的變化。研究指出《楚辭章句》中男性神靈的解釋與女性神靈的詮釋相比要來得直接,反映出漢代晚期女神地位下降的事實。研究性別意識的演變是理解《楚辭》及其傳統的重要視角 。

Type
Articles
Copyright
Copyright © The Society for the Study of Early China and Cambridge University Press 2018 

Incorporating fantastic, mythical, celestial, and sacred imagery, the hermeneutically vexing collection of Chu-style verse known as Chu ci 楚辭 is a fountainhead of the Chinese literary tradition as well as an essential source for studies of early religion. Authorship of the core poems is attributed to the legendary ur-poet Qu Yuan 屈原 (fourth–third century b.c.e.?), exiled nobleman of the Warring States kingdom of Chu 楚; but it was not until centuries after Qu Yuan's death that Wang Yi 王逸 (second century c.e.) of the Eastern Han dynasty established the basis of the now traditional interpretation of the collection.Footnote 1 Because Wang Yi's Chu ci zhangju 楚辭章句 commentary edition is the earliest received text of the Chu ci, and because its composition was historically and geographically closest to the original poems, the Chu ci zhangju remains an essential, inevitable source. The interpretation presented in the Chu ci zhangju is certainly not the only possibility, yet due to its unique status, its subtlest peculiarities have staggering implications for literary history.

This article reassesses the Chu ci zhangju by focusing on one of its peculiarities: the unequal exegeses of male and female deities. The analysis finds that Wang Yi's interpretation is gendered according to late Han imperial ideology and does not necessarily follow from immanent features of the original Chu ci or the culture of Warring States Chu. For the Chu ci zhangju treats deities quite differently depending on their gender, and apparently not in the way that later scholarship assumes. The commentary gives female deities a far more explicitly allegorical interpretation than the male deities; most of the male deities are explained with completely straightforward literalness, while the female deities are often explained as allegorical figures for mortal men. But the Chu ci zhangju is actually far less allegorical than later editions, which may interpret any or all of the deities allegorically regardless of gender. From the shifts between the textual layers of the Chu ci emerges a curious history of anachronistic circular reading, in which the ideological freight of later interpretative discourse is projected back onto the interpretation's sources. Discerning the gender bias of the Chu ci zhangju will thus clear the way for a more open-minded reading of the Chu ci, and will also demonstrate the significant implications of the gender politics of interpretation for scholarship on Chinese literature and cultural history.

If you have read the Chu ci poems, especially the Han compositions usually included in the anthology, or if you have read almost any premodern or modern Chu ci commentary, it may seem rather obvious that the female deities do in fact relate in a figural way to the mortal men of this-world politics. After all, “Li sao” 離騷 (Encountering Sorrow), “Jiu zhang” 九章 (Nine Compositions), and other poems include explicit references to governance, and they contain many allusions to mythical or historical rulers and ministers. “Li sao” also plays with gender, clearly figuring the speaker, who is usually identified with Qu Yuan, as a woman with beautiful moth-eyebrows. It is therefore only a small step to read Qu Yuan's presumed ruler, Chu Huaiwang 楚懷王 (King Huai of Chu, r. 328–299 b.c.e.), into the poem, and religious significance out of the poem.Footnote 2 But if we look at the poem without the common sense of the past 2,000 years, a common sense that actually emerged with the Song dynasty interpretation of Wang Yi's Han dynasty edition of the Chu ci, there is really no reason to read any of the “Li sao” deities allegorically. Huang Linggeng 黃靈庚, Ishikawa Misao 石川三佐男, and Constance Cook have put forth coherent interpretations by reading the female deities as female deities, powerful beings that the speaker encounters on his/her way to Heaven and/or the world of the dead.Footnote 3 These readings do not require negating the explicitly political passages of “Li sao,” for as archeological evidence has increasingly demonstrated, religion and politics were hardly separable for the people of Warring States China.Footnote 4 And in fact it is just this kind of literal religious and political reading which is most widely present in the Chu ci zhangju commentary. If it were not for the cases where female deities are explicitly read as figures for mortal men, the commentary would provide a relatively straightforward reading of the divine encounters as demonstrations of Qu Yuan's god-like power-virtue. It is only later editions that rationalize the religious significance of all the divine imagery by identifying everything as indirect representation of Qu Yuan's activity in secular government affairs.

Among the Chu ci texts, the “Jiu ge” 九歌 (Nine Songs) and the early zhao 招 (summoning) poems are generally considered to represent Warring States Chu religious culture in relatively original form.Footnote 5 The “Songs” narrate ritual proceedings and erotically charged encounters with various deities, while the “Da zhao” 大招 (Great Summons) and “Zhao hun” 招䰟 (Summoning the Soul) represent the ritual calling-home of a recently deceased person's wandering soul. In “Yuan you” 遠遊 (Far Roaming), often considered a later composition but attributed to Qu Yuan in the Chu ci zhangju, the speaker departs from the mundane world and journeys to the transcendent realm of the taichu 泰初 (Supreme Origin).Footnote 6 None of these texts explicitly narrate mundane matters of state administration or political history. The “Nine Compositions,” on the other hand, are complaints of the wronged, misunderstood courtier (i.e., Qu Yuan) who proclaims his perfect trustworthiness and enduring loyalty to his tragically deluded lord (i.e., King Huai). “Li sao” includes the imagery of the “Songs” and the themes of the “Compositions” in a longer narrative of a far-roaming celestial journey in pursuit of divine beings and meizheng 美政 (fair governance).Footnote 7 “Li sao” has been the object of intense exegetical debate, giving rise to varying interpretations, some relatively literal, but most strongly figural, in which political history is substituted for references to divine or otherworldly entities. The figural interpretation of “Li sao,” prototypically represented in later imitation poems included in the Chu ci, has in turn been used to explain the “Nine Songs,” the summoning poems, and “Far Roaming,” so that they too could be understood as allegories of Qu Yuan's this-worldly quest for fair governance.Footnote 8 The basic transition from ritual archetype to literary interpretation is apparent in the differences between the summoning poems: from first-person narration of the ritual in “Great Summons,” to third-person description of the ritual in “Summoning the Soul,” to literary adaptation of the ritual theme in the Han dynasty addition “Zhao yinshi” 招隱士 (Summoning the Recluse).

A commentary does not simply explicate what the text denotes, it posits meaning in terms of contemporary discourse, and thus is as much an index of the belief systems of its time as it is a key to immanent meaning. The Chu ci zhangju canonized the allegorical aspect of Wang Yi's interpretation as the traditional approach to reading the female deities. Its application was expanded to explain the male deities as well in later versions, such as in the Five Ministers 五臣 commentary on the Wenxuan 文選 (c. 714–742) and Hong Xingzu's 洪興祖 (1090–1155) Chu ci buzhu 楚辭補注 (c. 1154). Wang Yi's work was the necessary precursor to Zhu Xi's 朱熹 (1130–1200) strictly rhetorical reading, and has deeply informed all subsequent reinterpretations, including readings of the “doubting antiquity” approach.Footnote 9 In light of new methods and archeological evidence that have allowed a clearer picture of Warring States culture to emerge, many scholars now reject the allegorical interpretations of the “Nine Songs” and even of “Li sao”; but still, the allegorical impulse of 2,000 years has remained continually relevant, and the Chu ci zhangju is still indisputably the most essential source for Chu ci scholarship.Footnote 10 A thorough reassessment of Wang Yi's hermeneutic in terms of Han dynasty culture is in order.

This article begins to meet this need by presenting a critical discourse analysis of the Chu ci zhangju exegeses of deities. I will focus on the archetypal Chu ci poems “Li sao,” “Nine Songs,” “Nine Compositions,” and “Far Roaming,” comparing the Chu ci zhangju exegeses of deities in these poems to the narrative context of the original texts as well as to the Song dynasty sub-commentary on the Chu ci zhangju by Hong Xingzu. This article is focused on these selections of the Chu ci zhangju for several reasons. First, these works represent archetypal themes of the anthology that are combined in the principal poem “Li sao.” In the oldest recorded sequence, which differs from most editions today, “Li sao” is first in the anthology, “Nine Songs” is third, “Nine Compositions” is fifth, and “Far Roaming” is sixth. The sequence is a hierarchy of prestige that is also an interpretive proposition: the first poem, called “Li sao jing” 離騷經 (the Classic of Encountering Sorrow) in Wang Yi's edition, is poetically expanded and hermeneutically expounded in the following poems, as indicated by the word zhuan 傳 (commentary) appended to the following titles in some early editions. The zhangju commentaries on the archetypal poems, moreover, represent the full diversity of exegetical possibilities realized in the zhangju with regard to gender and deities, from straightforward literalism to the most strained allegoresis. It is these exegetical possibilities of the “commentary” poems that are read back into the “classic” “Li sao” in the zhangju, regardless of their dates of composition. Secondly, the zhangju exegeses of “Li sao” and “Nine Songs” are clearly related to the explicit political content of other poems included in the anthology, most of them historically later imitations, such as the “Jiu bian” 九辯 (Nine Modes), “Qi jian” 七諫 (Seven Remonstrances), “Jiu huai” 九懷 (Nine Cares), “Jiu tan” 九歎 (Nine Sighs), and Wang Yi's own composition, “Jiu si” 九思 (Nine Longings). The “Nine Compositions” is one of these more explicitly political works and is typically representative of them. The zhangju commentary projects the interpretation of the archetypal poems, as elaborated in the later derivative compositions, back onto the archetypes; isolating the archetypes will allow me to clearly demonstrate that the derivative interpretation is not in fact an inevitable one. Another early poem, “Tian wen” 天問 (Heavenly Questions)—early both in terms of historical attribution and anthology placement, fourth in Wang Yi's sequence—is not only formally and structurally distinct from the other Chu ci, but also not amenable to the kind of allegorical interpretation applied to the other poems in the anthology; as a list of rhetorical questions, it lacks the narrative dimension required of the allegorical hermeneutic (discussed below). The zhangju interpretation of “Heavenly Questions” is accordingly limited to the glossing of difficult words and tracing of allusions, with some generalized speculation on Qu Yuan's emotional state. The summoning poems are based on an archetypal soul-summoning ritual but contain few references to deities; they will be discussed as relevant. The remaining Chu ci, “Buju” 卜居 (Divination) and “Yufu” 漁夫 (The Fisherman) are straightforward pseudo-historical anecdotes and not relevant to this inquiry. In sum, the article focuses on the poems that are most pertinent to the study of religion and the genesis of allegory.Footnote 11 First, I will briefly contextualize the Chu ci zhangju in terms of late Han discourses of religion, government service, and gendered cosmology, and discuss the preface, in which Wang Yi explains his principles of interpretation. Then I will discuss the case study texts, beginning with the master text “Li sao,” with consideration of marginal and ambiguous interpretations. The analysis demonstrates how the Chu ci zhangju allegorically interprets the goddesses so as to present Qu Yuan as the paragon of a fading ideal of Han dynasticism: selfless service to the patriarchal cosmic imperium. This selfless imperial Qu Yuan is not the original speaker of the archetypal Chu ci, but he is a key player in nearly two millennia of Chinese literary history.

Methodology

The allegorical character of the Chu ci zhangju interpretation has been criticized in modern times, and Wang Yi's commentary has been derided for centuries for its many lacunae and inaccuracies, but it is nevertheless the essential gateway to the Chu ci, to which all received editions and interpretations owe an incalculable debt.Footnote 12 In his preface to his revised Song dynasty edition of the Chu ci, Zhu Xi accuses previous scholars—Wang Yi is his implied target—of making arbitrary comparisons and forced interpretations. Yet his own interpretation of “Li sao” is even more exclusively concerned with the this-worldly ruler–subject relationship, and it relies heavily on essentially arbitrary comparisons to the point that it explicitly negates the reality of the religious—that is, literal—dimension of the narrative.Footnote 13 As a more recent example, in the introduction to his beautiful translation of the Chu ci, David Hawkes complains that Wang Yi “is a very unreliable guide” and is “emphatically not” one of the “great scholars and exegetes” of the Eastern Han.Footnote 14 And yet, his interpretation and commentary are profoundly indebted to Wang Yi.Footnote 15 For example, in the first stanza of his translation of “Li sao,” the words “I passed from the womb” are clearly not entirely derived from the original text, which has only jiang 降 (descend), but is more literally derived from Wang Yi's commentary: “I came out of my mother's body” (下母之體).Footnote 16 Gopal Sukhu has attempted a total re-interpretation of “Li sao” based on a rejection of Wang Yi's explanation of jiang, demonstrating both the enormous impact of minute details of the commentary and, in the nevertheless derivative portions of his interpretation, the difficulty of actually reading without it.Footnote 17 As it is the earliest extant edition, reliance on Wang Yi's work is simply inevitable.Footnote 18

Chu ci zhangju was transmitted in Hong Xingzu's Song dynasty sub-commentary edition, the Chu ci buzhu. Much of the allegorical character attributed to Wang Yi's work is in fact to be found in later commentaries and the sub-commentary included in the Chu ci buzhu; if we isolate the Chu ci zhangju commentary and examine it carefully, we find that it is only allegorical in certain instances—especially the exegeses of female deities. In other words, although the Chu ci buzhu preserves Wang Yi's work and presents itself as an editorial elaboration of the Chu ci zhangju rather than as a fundamental reinterpretation of the original texts, it is in fact quite different in its overall treatment of divine entities. The Chu ci zhangju, like the original Chu ci texts upon which it comments, has been an object of allegorical interpretation. The task of this essay is to trace the historical process of allegoresis as it has been discursively instantiated in the textual history of the Chu ci. While the “original texts” cannot be fully disentangled from the Chu ci zhangju and the Chu ci buzhu in which they have been transmitted, they must be read bi- or tri-focally, not only in terms of the Warring States context of composition but also in terms of the late Han and Song contexts of reception.

Like the Chu ci buzhu in which it was transmitted, the Chu ci zhangju is also a composite text; it was completed in the Eastern Han dynasty (25–220 c.e.) but incorporates the work of earlier scholars, including the King of Huainan 淮南 and patron of the Huainanzi 淮南子 Liu An 劉安 (179–122 b.c.e.), the bibliographer Liu Xiang 劉向 (77–6 b.c.e.), and the historian Ban Gu 班固 (32–92 c.e.). It reflects a Han dynasty debate about the proper interpretation of the already archaic texts attributed to Qu Yuan.Footnote 19 Wang Yi is unreserved in his praise of Qu Yuan, and so a task he set himself in writing a Chu ci commentary was to overcome the criticisms of the poetry and its author leveled by earlier court scholars such as Yang Xiong 揚雄 (53 b.c.e.–18 c.e.) and Ban Gu. Qu Yuan's legendary suicide was a major concern, but Ban Gu also particularly pointed out the goddess Fufei 宓妃 as an example of the non-canonical, and therefore immoral, content in the Chu ci.Footnote 20 To overcome their criticism, Wang Yi draws on Liu An's earlier praise of Qu Yuan's spiritual transcendence. He refutes Yang Xiong and Ban Gu by making Qu Yuan out to be a spiritual adept—though not one who consorts with goddesses—and a loyal servant of the state.Footnote 21 In doing so, his work registers important conceptual shifts that had developed in official scholarly discourses between the Warring States and the late Han: a reconfiguration of the state's relation to the spiritual world, a moralizing insistence on commitment to government service, an ideological subordination of women and discursive restriction of their agency, and a reduction in status and/or re-gendering of most female deities.Footnote 22 Contextualizing the Chu ci zhangju within these ideological differences and transformations offers considerable precision in clarifying Wang Yi's hermeneutics.

In the half-millennium between the late Warring States and the Eastern Han, significant transformations occurred in spiritual thought and official religious ideology, as can be seen in the shifting, overlapping discourses of ritual, deity, and cosmology.Footnote 23 Furthermore, the religious culture of Warring States Chu was in many ways distinct from the contemporaneous culture of the central plains, but it was the ancient rituals and texts of the central plains that were most esteemed by Wang Yi and other court scholars of the Eastern Han.Footnote 24 Between the time of Qu Yuan and the time of Liu An, the Qin and Han empires had profoundly changed the form of court religion and the political structure of society. And between Liu An and Ban Gu, the ritual reform of the late Western Han again marked significant changes in the relationship between the state and the gods.Footnote 25 According to Michael Puett, the concept of self-apotheosis arose in direct relation to the crumbling of the Zhou political order and its rituals, while the Han ritual reform was a rejection of early imperial aspirations to theomorphic kingship such as those asserted by Qin Shihuangdi 秦始皇帝 (the First Emperor of Qin, 259–210 b.c.e.) and Han Wudi 漢武帝 (Han Emperor Wu, 157–87 b.c.e.). The monistic correlative cosmology most associated with the Han dynasty likewise arose as a critique of court theism and rulers' pursuits of immortality and omnipotence through personal contact with deities.Footnote 26 By Wang Yi's time, the emperors' appropriation of the language and iconography of deity to themselves could be taken for granted,Footnote 27 but also, “claims of ascension became associated with those groups who opposed the empire.”Footnote 28 For Liu An, contemporary of Emperor Wu, Qu Yuan could very well have been a paragon of self-cultivation whose work expresses an experience of the spiritual transcendence of mundane political struggles and the attainment of god-like power.Footnote 29 For Ban Gu and for Wang Yi, to be such would be heretical and possibly treasonous.

Men's attitudes toward state service underwent similar changes over the course of the Han. Whereas “the dominant ethos of the Western Han was one of public service” that emphasized a man's impartial duty to his father and ruler, in the Eastern Han the balance of power shifted to local elite families, with the consequence that “members of the political elite became increasingly disaffected from the state and withdrew from official life.” At the same time, the importance of the mother–son bond was increasingly emphasized, representing the social valuation of a private alternative to public service.Footnote 30 The dominant ethos, however, is not the only or official attitude toward state service. If, as John Major has asserted, the Huainanzi was a reactionary text that “proposed solutions to the political problems of a world that no longer existed,”Footnote 31 then Liu An's praise of a transcendent Qu Yuan must be understood as similarly anachronistic. Similarly, Wang Yi's Eastern Han interpretation conveys the interests of the waning imperial court, and thus a rejection of the turn toward the maternal bond and away from impartial public service. In other words, Wang Yi responds to Liu An's heterodox praise and Ban Gu's classicist critique with his own residual imperialistic ideology: the cultivation of spiritual power is praiseworthy when it is undertaken as a service to one's king, rather than as a transcendence of court politics. Selfless service to the state is the ultimate value.Footnote 32

Consider the commentary to “Nine Modes,” which followed directly after “Li sao” and was called zhuan 傳 (commentary) in Wang Yi's edition. The “Nine Modes” does indeed resemble “Li sao,” and like “Li sao” the final section relates the speaker's cosmic flight among various deities and last thought of the earthly world. The Chu ci zhangju commentary does not explain any of these deities as allegorical figures, and none of them appear to be female. The poem concludes with a statement that brings together celestial and earthly rule: “Having obtained August Heaven's favor, I return to [ensure] my lord's good health” (賴皇天之厚德兮,還及君之無恙). Wang Yi's comment to this final section is as follows:

言己雖陞雲遠遊,雖從百神,志猶念君,而不能忘也。

This says, “although I ascend on the clouds and roam far, although I follow the hundred deities, still my intention is to think of my lord, and I cannot forget him.”Footnote 33

Because the “Nine Modes” is, in Wang Yi's presentation, a commentary on “Li sao,” it makes sense to assume the same literalism is allowable in reading the cosmic flights of “Li sao.” And in most cases, it is; but not, for Wang Yi, in the cases of female deities. Wang Yi reads the “Li sao” deities according to his understanding of Song Yu's 宋玉 “Nine Modes,” in which there are no goddesses and Qu Yuan's true intention is always to serve his lord. In Wang Yi's reading, the proper objective is the court of one's lord, not the natal home. This is made explicit in the “Li sao,” commentary, which has Qu Yuan “coming down out of his mother's body” in the beginning, and “following [loyal minister] Peng Xian to his dwelling place” (從彭咸而居處) in the end.Footnote 34 Neither the mother's body nor the loyal minister of the state are explicit in or essential to the poem; they are supplied by Wang Yi's commentary, and it is entirely possible to read the trajectory of the poem differently.Footnote 35

The Eastern Han particularity of Wang Yi's reading is also apparent in his frequent practice of integrating images into the discourse of official Han Confucianism, an ideology of correlative yin-yang 陰陽 cosmology and orientation toward harmony with an intentional Heaven that is typified in the Chunqiu fanlu 春秋繁露 (c. second century b.c.e.) attributed to Dong Zhongshu 董仲舒 (179–104 b.c.e.).Footnote 36 In contrast to earlier cosmologies, the yin-yang theory of Han Confucianism consistently elevates yang 陽 over yin 陰, heaven over earth, the ruler over the ministers, and male over female, and tends toward strong gender essentialism.Footnote 37 In parallel development, over the course of the Han the social roles of women were gradually restricted and diminished,Footnote 38 including the role of the female spirit-medium.Footnote 39 In the official religious practice of the state, a few anthropomorphic or abstract gods of heaven gradually supplanted other ancient deities, many of whom were transformed from female to male in the historical record, even as the emperor assumed command of the gods himself.Footnote 40 The Chu ci zhangju commentary's correlation of hemerological system, gender, and philosophical yin and yang is typical of the Han Confucian approach to texts;Footnote 41 correlative reasoning and the cultivation of yin and yang essences is often invoked in the commentary to explain the speaker's actions. But the Han cosmology is biased in favor of the masculine, and the Chu ci zhangju is biased in applying it, invoking cosmological correlations to explain the female deities but never using them to explain male deities.

On a thematic level most of the Chu ci texts share enough general characteristics of the Han dynasty Five Classics that they are in fact easy objects for the classicist hermeneutic lens: they idealize moral virtue as political power, describe a golden-age history and a fallen present, advocate for promotion by merit, and use historical precedents for persuasion. The classics also address gender, although they were never as narrowly proscriptive of real women's lives as is often assumed.Footnote 42 And in fact Wang Yi's commentary doesn't take issue with the various roles of female characters in the Chu ci, unless the female character is a goddess. Where the Chu ci zhangju finds female spirit-mediums and court dancers, it never interprets them as allegorical signs. But some aspects of the Chu ci fall outside the scope of the Han classical lens entirely; most obvious are the fabulous erotic quests for divine companionship. The preferred texts of the Han court scholars represent the idealized Zhou culture of the central plains and offer little guidance for dealing with the wanton goddesses and celestial flights of the Chu poetry. Ban Gu's criticism of these extra-canonical, un-classical elements was a significant concern for Wang Yi, as we can see from Wang Yi's response to it.Footnote 43 It is with regard to these problematic aspects, the basic essence of the “Nine Songs” and a significant feature of “Li sao” and “Far Roaming,” where the allegorical approach emerges most clearly in Wang Yi's interpretation. For Liu An, patron of the Huainanzi and first known commentator on “Li sao,” and for some people of his time such as Han Emperor Wu and the Marquis and Lady of Dai 軑 (buried in the tomb now known as Mawangdui 馬王堆 c. 163 b.c.e.), female deities were clearly worthy objects of veneration and emulation, as well as sought-after sources of spiritual power.Footnote 44 We can imagine that Liu An's lost commentary on “Li sao” would have been fairly literal; substantial relics of the literal interpretation are indeed to be found in the extant Chu ci zhangju.Footnote 45 But for Wang Yi of the Eastern Han, the female deities required a different interpretive strategy: cosmological allegory.

Without engaging the long-standing theoretical debate about Chinese allegory, I will offer a heuristic definition of allegory for the present study.Footnote 46 Allegory explicitly relates a sign in the text to some other narrative that is inappropriate to the obvious literal context. This apparent inappropriateness is central to most descriptions of the “allegorical” interpretation of early Chinese classics including the Chu ci.Footnote 47 So, to say that fragrant plants “are used to symbolize virtue” (所以象德) in a poem where the speaker cultivates fragrant plants immediately following a declaration of his/her corresponding neimei 內美 “inner beauty” and xiunen g 脩能 “cultivated bearing,” and to explain moreover that the noble men of old really did wear fragrant plants on their clothes, is not exactly allegorical;Footnote 48 likewise, to say that “the sun rules as a king” (日爲王者) in a poem about a solar deity entitled “Dong jun” 東君 (Lord of the East) is not exactly allegorical.Footnote 49 It does not require breaking out of the literal narrative of the text to put forth these interpretations. But to say that a particular deity is a goddess who is used as a figure for virtuous men in reclusion, when the explicit context contains no mention of or narrative requirement for recluses, is to do something quite different, hermeneutically speaking. It is this second kind of interpretation, which refers the sign of the text to some other thing outside of the literal narrative context, that I will call allegorical. In this I am not claiming that allegorical interpretations are always wrong; only that they assert the indication of a secondary referent that is not explicit and not absolutely necessary. Due to the complicated nature of “literal” meaning, this definition does not produce conclusions about the original, intentional, meanings of the poets and commentators, but it will allow us to re-read the Chu ci zhangju commentary in terms of what it actually says, and to discern where its exegeses are in tension with what the primary texts and later sub-commentaries actually say.

This method of analysis does not require us to distinguish between fantasy and reality or fiction and biography. Instead of reference, the issue is rhetoric: what discursive moves are made? Most readers of “Li sao” have understood it to be more or less historically referential; but can we really assume that the decidedly constructed story of Qu Yuan's political disappointment is more “real” than the story of the speaker's celestial courtships?Footnote 50 Is that distinction valid across all historical readings? If we read in terms of a post-Song dynasty metaphysics or modern common sense, we might allegorize instinctively. But if we read the archetypal Chu ci in the context of early Chinese culture, it is difficult to demonstrate any need to allegorize so strictly. Given the abundant evidence that the people of early China assumed the existence of all manner of extra-human powers, it would be absurd to insist that Wang Yi's understanding of Qu Yuan's biography must be primarily secular or this-worldly—much less the archetypal Chu ci poems themselves.Footnote 51

Where, then, does the Chu ci zhangju reach beyond the literal narrative of the text for explanation? In order to comprehensively present Qu Yuan as a dutiful servant of the state, as Wang Yi claims to do in his introduction, some aspects of the poetry need to be allegorical, and some don't.Footnote 52 Wearing herbs and flowers is acceptable within an appropriate cosmological context to Wang Yi; so is flying in a dragon-drawn chariot to meet tiandi 天帝 (the God-Emperor of Heaven); but erotic encounters with female deities are not. In Wang Yi's commentary on “Li sao” all unquestionably female characters except Nüxu 女嬃 are translated into mortal men, either ministers or courtiers in reclusion.Footnote 53 Nüxu, the sole exception, is glossed as Qu Yuan's elder sister in both “Li sao” and “Xiang jun” 湘君 (The Lord of the Xiang River).Footnote 54 Other mortal women elsewhere in the Chu ci are also explained literally, such as the spirit-medium and seductive dancers of the summoning poems.Footnote 55 The variants on this principle, which I will discuss below, are some of the female deities of the “Nine Songs” and “Far Roaming.” But while the Chu ci zhangju does not explain all the “Nine Songs” goddesses as figures for mortal men, it does explicitly compare them to a mortal man, Qu Yuan, and it makes them symbols of cultivation rather than lovers pursued. The seemingly anti-political “Far Roaming,” on the other hand, notably lacks the layer of commentary attributed to Wang Yi until the point in the poem at which goddesses appear: at this point Wang Yi explains that they represent mortal women with a pragmatic and subservient role in court politics.Footnote 56 There are no goddesses to interpret in the “Nine Compositions,” but the Chu ci zhangju commentary here offers some stronger hints toward a figural reading of male deities. Thus, the Chu ci zhangju commentary constructs a hermeneutic bridge between the potentially allegorical goddesses of the original “vulgar lyrics” of Chu rituals (其詞鄙陋),Footnote 57 the fully allegorical goddesses of the literary “Li sao,” and “Far Roaming,” and the androcentric worldly politics of the “Nine Compositions” and later imitations. The structure and content of the Chu ci zhangju suggests that the allegorical reading of female deities is an Eastern Han addition to previously more literal exegeses that circulated in the Western Han and earlier, and that the allegorical reading of male deities is an even later addition.

I do not devote much ink to cases of reading historical and quasi-historical people related to Qu Yuan's biography (e.g., King Huai, Zilan 子蘭 [Prince Orchid], Zijiao 子椒 [Prince Pepper]) into the text, because, as we shall see, the Chu ci zhangju does not use gods or goddesses to incorporate these historical referents. While the specific historical people are unnecessary according to the literal narrative, and so certainly allegorical according to my definition, these instances only demonstrate that Wang Yi interpreted the Chu ci biographically, a point that is hardly in doubt. The interpolation of historical actors is what Pauline Yu calls “contextualization,” a hermeneutic practice applied to the Shi jing 詩經 that Wang Yi emulated.Footnote 58 Reading Nüxu as Qu Yuan's elder sister rather than as a spirit-medium can also be understood in terms of contextualization. But Nüxu appears in the narrative before the speaker's departure from mundane politics, and her role is to offer advice in the form of historical perspective rather than to elude the speaker's celestial quest, so reading her as a human sister does not require a significant departure from the literal narrative. In contrast, reading the clearly divine Fufei as a worthy man in reclusion is contextualization that further requires stepping outside of the literal narrative of the goddess quest.

The gender imbalance of Wang Yi's interpretation first appears in the Chu ci zhangju preface. In this text Wang Yi expounds his view of Qu Yuan's compositional technique, which furnishes the principles of his hermeneutic method.

離騷之文,依詩取興,引類譬諭,故善鳥香草以配忠貞,惡禽臭物以比讒佞,靈脩美人以媲於君,宓妃佚女以譬賢臣,虬龍鸞鳳以託君子,飄風雲霓以爲小人。

The literary pattern of “Li sao” relies on the classical Songs in selecting evocative juxtapositions and draws from natural categories in making figural comparisons. Thus, auspicious birds and fragrant herbs go together with the loyal and steadfast, evil birds and stinking things are compared to slanderers and toadies, “spiritually distant” and “fair one” are matched to the ruler, Fufei and Jiandi are compared to worthy ministers, dragons and phoenixes convey noble men, and whirlwinds and rainbows are used for petty men.Footnote 59

Most significant for the analysis below is Wang Yi's assertion that specific deities, the goddesses Fufei and Jiandi 簡狄, are figures for ministers—that is, mortal men—while the ruler is figured not by named deities but rather by laudatory epithets. The epithets may be used in address to deities, as they are in some of the “Nine Songs,” but they are not the names of particular deities. In other words, language that can be used to address the gods is also used to praise a ruler, while language that names specific known deities is used to refer to ministers. The specific known deities are both definitely female, while the gender of the epithets is ambiguous. No male deities are mentioned in this framework, in spite of their ubiquity in the poems. Here and throughout the Chu ci zhangju, Wang Yi does not go so far as to say that gods are figures for rulers, but he does explicitly state that goddesses are figures for ministers.

“Li sao”

It may be helpful to first establish the most basic literal narrative of “Li Sao” so as to identify clearly where the Chu ci zhangju and later commentaries depart from or add to it. The poem begins with an introduction, including divine genealogy, auspicious names, and cultivation of virtues.Footnote 60 Then the speaker makes a statement of political intent: to change the king to be like the ancient kings. He/she tells of being mistreated and yet continuing to cultivate virtues, including spiritual qualities reflected in bodily adornments. The speaker undertakes a preliminary journey by horse-drawn chariot. Then Nüxu appears and offers pragmatic advice. Not satisfied, the speaker travels south and addresses the god-king Chonghua 重華, who offers further advice. Concluding that the times are bad, the speaker departs on the first celestial journey to the gate of heaven, this time flying in a chariot drawn by mythical beasts. Decrying the dirty, jealous world, the speaker courts various divine or mythical women. Failing in these advances, he/she laments the king's continued ignorance. The speaker then obtains divinations from Lingfen 靈氛 and Wu Xian 巫咸, and concludes that everyone has been sullied except him/herself. The speaker takes a final flight to the distant edges of the mythic cosmos, and ultimately renounces the earthy kingdom.

Words that are not included anywhere in this literal narrative include jun 君 (lord/ruler), chen 臣 (minister), shi 士 (courtier), yin 隱/yi 逸 (recluse), and yin 陰. Yang 陽 is used only once, as part of the name of the deity Gaoyang 高陽. There are many clear instances of immanent figure in such phrases as “Acorus does not perceive my inner nature” (荃不察余之中情兮), since acorus, a flower, is not sentient.Footnote 61 But because the celestial flight passages are prefaced by explicit rejections of the worldly status quo and don't contain the key words Wang Yi uses to explain them, and because other early texts describe the same journey locations and deities in terms of celestial geography and religion, it is difficult to establish conclusively that a secular narrative of court intrigue is immanent in these passages.Footnote 62 Although the literal narrative of “Li Sao” is not exactly a simple plot line, if we assume as the people of early China did that the world is full of ghosts, nature spirits, heavenly deities, mythical gods and goddesses, dragons, and other extra-human powers, then at least we do not have to look far beyond the explicit content to find “real” meaning.Footnote 63

In sum, “Li Sao” can be understood as an anonymous narrative or a specific biography without requiring additional layers of meaning: the fictional speaker or the real Qu Yuan renounces mundane political wrangling to seek spiritual power and transcendence. In Wang Yi's reading, Qu Yuan does seek spiritual power and contact with deities, but he does not renounce his lord or worldly governance; this turn is introduced by hanging an allegorical narrative of continual service to the state on the literal narrative of courting deities, and specifically by substituting unspoken worthy courtiers for the female deities.

Wang Yi's commentary reads the divine female characters as figures for mortal men on the grounds that both women and ministers are yin 陰.Footnote 64 This formula allows Wang Yi to conflate the exegeses of women and goddesses as it suits his overall interpretation. So it explains that zhongnü 衆女 (the assembled women) means zhongchen 衆臣 (the assembled ministers); gaoqiu zhi wu nü 高丘之無女 (the high hill is without women) means that Chu is without worthy ministers; qi wei shi qi you nü 豈唯是其有女 (how could only here have women) asks how only Chu could have ministers to seek out; and qiu nü 求女 (pursue women) means to seek for tongzhi 同志 (men of like aspiration) to serve the ruler, that is, potential ministers.Footnote 65 Yousong zhi yinü 有娀之佚女 (the fair woman of Yousong) is explained as a reference to Jiandi, the consort of the god Diku 帝嚳, who “was a mate to a sage god-emperor and gave birth to a worthy son, so she is used as a figure for loyal worthies” (配聖帝,生賢子,以喻貞賢); i.e., she is an allegory of mortal men.Footnote 66 In a straightforward demonstration of this gender bias, to the lines “I command Fenglong to ride the clouds—to seek Fufei's abode [for me]” (吾令豐隆椉雲兮,求宓妃之所在), Wang Yi comments that Fenglong is yunshi 雲師 (the Master of Clouds), and Fufei is “a goddess, used as a figure for recluses” (神女,以喻隱士). Thus,

言我令雲師豐隆,乘雲周行,求隱士清潔若宓妃者,欲與幷心力也。

This says, “I command the Master of Clouds, Fenglong, to ride the clouds around, seeking recluses as pure as Fufei; [I] want to combine [our] efforts.”Footnote 67

The master of clouds is literally acceptable as a companion for Qu Yuan, whereas a female water deity is allegorical. The female-yin-minister formula allows Wang Yi to relate the female deities to the apparently extraneous worthy ministers and potential-ministers-in-reclusion. This interpretation may have seemed both erudite and commonsensical to scholars of the late Han and after, but it is not necessarily required of the text.

On the other hand, the unmarked,Footnote 68 apparently male deities of the “Li sao” celestial journey, including Gaoyang 高陽, Di 帝, Fenglong 豐隆, Dixin 帝辛, Huang 皇, Chonghua 重華, Xihuang 西皇, and the inhabitant of the “Spring Palace” 春宮 (identified in the commentary as Dongfang qingdi 東方青帝 “the Green God-Emperor of the East”), are not given any allegorical explanation at all, but are simply identified as gods and described in terms of the literal narrative of the journey. None of these deities are explained as “a figure for the ruler” as one might expect from the yin-yang analogy applied to the goddesses. For example, changhe 閶闔 in the Chu ci zhangju is simply tianmen 天門(the gate of heaven), whither Qu Yuan ascends “to present his plaint to the God-Emperor of Heaven” (將上訴天帝).Footnote 69 However, the Chu ci buzhu, citing the Eastern Han dictionary Shuowen jiezi 說文解字 (121 c.e.) and an even later commentary on the Wenxuan, concludes that “Qu Yuan, too, uses changhe [the gate of heaven] as a figure for the gate of the ruler” (屈原亦以閶闔喻君門也).Footnote 70 This implies that the God-Emperor of Heaven himself is a figure for King Huai far more clearly than anything in the Chu ci zhangju, whatever Wang Yi's intention may have been. Zhu Xi makes this implication explicit: “this is a metaphor for seeking a great ruler but not meeting him” (蓋求大君而不遇之比也).Footnote 71

The expanded allegorical approach to “Li sao” is not, however, entirely a Song dynasty novelty. There are hints in the Chu ci zhangju that male deities may be correlated with mortal kings. Near the beginning of the first heavenly journey in “Li sao,” the speaker declares: “[I] want to linger a while by these spirit-wrought engravings—[But] the day so swiftly advances to evening” (欲少留此靈鎖兮,日忽忽其將暮). This is the point in the narrative at which the speaker departs from terrestrial geography in a dragon-drawn chariot to begin a voyage among the celestial deities. According to the Chu ci zhangju, the ling 靈 of lingsuo 靈鎖 “spirit-wrought engravings” is “used as a figure for his ruler” (以喻君), while suo 鎖 refers to a gate that is engraved with a design of interlinked rings, and specifically to the entrance of the living quarters of the King of Chu.Footnote 72 But then, Wang Yi adds, “it is also said that ling is where deities are present” (一云:靈,神之所在也). Ignoring for a moment the second comment, let us examine the first to see if this could be the key to reading all the male deities as figures for mortal rulers. The word ling first appears in the Chu ci in one of the names of the speaker of “Li sao”: “[He] called me Spiritual Balance” (字余曰靈均).Footnote 73 Wang Yi glosses this first ling as shen 神 (divine), and further elaborates that “as to what nourishes things in balanced harmony, nothing is more divine than Earth” (養物均調者,莫神於地).Footnote 74 Ling, in this case, is a divine quality associated with earth and yin. In its second appearance, ling is part of an epithet, lingxiu 靈脩, which can be awkwardly translated according to Wang Yi's gloss as “Spiritually Distant”; this epithet describes the spiritual farsightedness of the ruler's power-virtue (能神明遠見者,君德也), “so it is used as a figure for his ruler” (故以諭君).Footnote 75 But the same comment also clearly distinguishes the gods from the king:

言己將陳忠策,內慮之心,上指九天,告語神明,使平正之,唯用懷王之故,欲自盡也。

This says, “I [Qu Yuan] am laying out a loyal plan from the innermost thoughts of my heart. Above I point to the Nine Heavens, informing the gods that they may verify it: purely for the sake of King Huai, I will give my all.”

Grammatically, Spiritually Distant/King Huai is not among the gods addressed in either the poem or the comment. Thus, the figural epithet Spiritually Distant evokes a general divine quality, as it does in the name Spiritual Balance. The ruler's spiritual farsightedness must refer to King Huai's god-like ability to perceive what is distant in space and time (a regrettably underutilized ability, we are to infer). Later, “Spiritually Distant” is again attributed to King Huai, in the comment “Spiritually Distant refers to King Huai” (靈脩,謂懷王也), in the context of the speaker's declaration of his/her unchanging preservation of old virtues in a changed world. This corrupt new world is described with such worldly words such as min 民 (the common people) and shi su 時俗 (the vulgar customs of this time).Footnote 76 Thus, when we finally arrive at the “spirit-wrought engravings,” the word ling has been associated with both the speaker and his king via auspicious epithets, but it has not been used to allegorically identify spirits or gods as substitutions for mortal men. If we furthermore note that the “spirit-wrought engravings” is the point whence the poem's speaker departs from the corrupt earth on the journey to heaven, and that throughout the Chu ci zhangju commentary it is petty mortal men who impede this journey and powerful gods who assist it, then the secondary comment that “‘spiritual’ describes a place where deities are present” does not seem like a contradiction: certainly deities should be present at the place where the god-like human recipient of Heaven's Mandate resides. Although the exegesis of “spirit-wrought engravings” and other instances of ling do not indicate generalized allegoresis in Wang Yi's interpretation, they provide hints for later commentators.

A few male deities are given figural explanations in the Chu ci zhangju, but they are not clearly defined as substitutions for other beings; instead they are given demonstrative value in terms of Qu Yuan's power-virtue. So Feilian 飛廉 is identified as fengbo 風伯 “the Earl of the Wind” while “the wind is commands, used as a figure for the ruler's orders” (風爲號令,以喻君命).Footnote 77 This could imply that Feilian is a figure for King Huai's human deputy, but does not make that explicit. Indeed, the commentary goes on to simply reiterate that Feilian himself is in charge of the ruler's orders:

言己使清白之臣如望舒先驅求賢,使風伯奉君命於後,以告百姓。

This says, “I make pure minsters like Wangshu lead the way in searching for worthies, and make the Earl of the Wind receive the ruler's orders behind, announcing them to the people.”

(Wangshu is allegorical and probably female; see below). This does not identify “the ruler” as King Huai; it may be that in this case the ruler is a ruler of heaven.Footnote 78 It is followed by an even more literal alternate reading: the speaker needs the force of the wind to ride on dragons and clouds.Footnote 79

The strongest instance of possible allegory with respect to a male deity is leishi 雷師 “the Master of Thunder,” who in the poem “tells me that it is not ready” (雷師告余以未具).Footnote 80 According to the Chu ci zhangju, “thunder is the feudal lords, as an evocation of the ruler” (雷爲諸侯,以興於君). This is unusual in that the metaphor indicated by wei 爲 is further identified as the rhetorical figure xing 興, “evocative juxtaposition.”Footnote 81 It is one of few xing noted in the Chu ci zhangju commentary. According to the pattern of evocative juxtapositions in Shi jing exegesis, from which Wang Yi draws this term, the thunder should precede the appearance of the ruler in the poem; but no ruler is explicitly present in these or the following lines of the poem. Instead Wang Yi uses the thunder as a pretext to introduce the ruler, via the feudal lords, into the commentary as the first of a series of impediments on the journey to heaven:

言己使仁智之士如鸞皇先戒百官,將往適道,而君怠墮,告我嚴裝未具。

This says, “I make good and wise courtiers like magical birds to go ahead and alert all the officials, and am about to embark on the way; but the ruler is idle, and tells me the equipage is not complete.”Footnote 82

King Huai is not the object of this quest; in both the poem and the Chu ci zhangju commentary, the object is literally a heavenly di 帝 (God-Emperor). And while the commentary states that thunder, a force of nature, represents the feudal lords, it does not quite state that the deity Master of Thunder is himself a feudal lord, or that “the ruler” is King Huai. In Wang Yi's comment, the ruler is the logical subject of the verb gao 告 (to tell), although in the poem it is the Master of Thunder. To make this interpretation coherent we could read “Master of Thunder” as “master of feudal lords,” i.e. the ruler; but then the rhetorical figure is mechanical substitution rather than evocative juxtaposition. Thus, the Master of Thunder can certainly be read as a figure in Wang Yi's allegory, but it is not unproblematic to do so. It is possible, and simpler, to read the Master of Thunder and “the ruler” as deities in the literal quest narrative, as Wang Yi could well have done given the conceptual parallels between worldly and celestial governance during his time.Footnote 83

The pattern is clear: the Chu ci zhangju commentary on “Li sao” has a strong gender bias in its explanation of the deities. The exegeses of female deities are strongly and explicitly allegorical, while those of the male deities are quite literal or only obliquely allegorical. Even some deities that are not explicitly gendered seem to fit this pattern. For example, in “Li sao” there is a deity associated with the sun, Xihe 羲和. The Chu ci buzhu identifies Xihe as a female deity, so we may expect her to be allegorical. However, to the line where this name appears, the Chu ci zhangju simply adds the literal gloss, “Xihe is the charioteer of the sun” (羲和,日御也). There is no allegorical explication of Xihe.Footnote 84 But it is quite probable that Wang Yi was not reading Xihe as a female deity. While the Shanhai jing 山海經, the source of the Chu ci buzhu explanation, describes Xihe as female, it does not describe her as the charioteer of the sun, but rather as the mother of ten suns, and Hong Xingzu elsewhere asserts that Wang Yi “did not see the Shanhai jing” (不見山海經).Footnote 85 In most other Han texts, Xihe, or rather the Xis and the Hes, are apparently male, identified with historical officials of the calendar, and hence “control of the sun,” another possible translation of ri yu 日御.Footnote 86 Wang Yi makes a similar move in his interpretation of Peng Xian or Peng and Xian 彭咸 as a Shang dynasty minister, rather than as divinized ancestors, transcendents, and/or spirit-medium demigods.Footnote 87 Furthermore, in the Chu ci zhangju commentary on the “Nine Songs” poem “Lord of the East” it appears that the sun deity is male and his own charioteer.Footnote 88 In the Chu ci zhangju Xihe is not explicitly gendered and not allegorical; but it is likely that Wang Yi assumed the administrative Xis and Hes and/or the Lord of the East in his commentary.

On the other hand, Wangshu 望舒 is identified as the charioteer of the moon and may be feminine due to association with the yin moon. The Chu ci zhangju does not make the gender of Wangshu explicit, but circumstantial evidence suggests it. The Chu ci buzhu cites the Huainanzi to note that the charioteer of the moon is also known as Xian'e 纖阿;Footnote 89 the clearly feminine graphic variant Xian'e 孅阿 appears in “Zixu fu” 子虛賦 by the Western Han poet Sima Xiangru 司馬相如 (197–117 b.c.e.). The Tang scholar Sima Zhen 司馬貞 (697–732) annotates this passage as it appears in the Shi ji 史記 biography of Sima Xiangru 司馬相如列傳 as follows:

服虔云,纖阿爲月御,或曰美女姣好貌。又樂產曰,纖阿,山名,有女子處其岩,月歷岩度,躍入月中,因名月御也。

Fu Qian says, “Xian'e is the charioteer of the moon. It is also said that she was a beautiful woman of fair countenance.” And Yue Chan said, “Xian'e is the name of a mountain. A young woman dwelled on its crags, and when the moon was passing over she leaped into the moon, and so was named the charioteer of the moon.”Footnote 90

Xian'e also appears in the Chu ci in the Han imitation poem “Si gu” 思古, one of the “Nine Sighs” by the late Western Han scholar Liu Xiang. In this case Wang Yi remarks that Xian'e “was a skilled charioteer of old” (古善御者) and a figure for a worthy courtier.Footnote 91 Wang Yi was familiar with the Huainanzi and Xian'e, and both the moon and the name Xian'e would have feminine associations.Footnote 92 So it is no surprise that, according to the Chu ci zhangju, “the lunar body is bright, so it is used as a figure for the purity of the minister(s)” (月體光明,以喻臣清白). The following comment, “I make pure minsters like Wangshu lead the way in searching for worthies,” makes the allegory of ministers-as-Wangshu explicit.Footnote 93 This is appropriate, according to Wang Yi's formula, if the deity is female. It is also a bit awkward, as the allegorical explanation means that the speaker, Qu Yuan, is making pure ministers (men like himself) seek for worthies (other men like himself) in order to demonstrate his own pure worthiness, and in the meantime commands the ruler's divine helper Feilian (see above). Perhaps we are meant to extrapolate that the ruler is King Huai and that Feilian, who follows Wangshu, is a figure for a mortal player in King Huai's administration. But if that is the case, why does the Chu ci zhangju not provide any allegorical explication for Xihe, who is mentioned earlier in this section of the text? I believe it is because Wang Yi's presentation of “Li sao” does not even attempt to construct an internally coherent and strictly mundane allegorical narrative of Chu government as later commentators have assumed, because it is solely concerned with, and in fact quite consistently succeeds in, constructing a spiritual but androcentric cosmos for Qu Yuan to roam. It may be as Miyano Naoya 宮野直也 has demonstrated: although Wang Yi obviously reads the Chu ci in terms of Qu Yuan's biography, he treats the Chu ci as a classic, and so applies the techniques of Han scholasticism to expound the general and absolute significance of each line of text, and does not actually attempt to make the narratives correspond to the real history and intentions of the particular authors.Footnote 94 The Chu ci zhangju constructs an autobiographical account of Qu Yuan to explain “Li sao,” an autobiography that includes deities of nature and heaven, cosmic flights, and spiritual cultivation—but not spiritual cultivation in the form of divine courtship or contact with female deities. In denying the literal significance of the female deities, the Chu ci zhangju is able to insert mortal courtiers throughout, and thus to keep Qu Yuan oriented toward service to the state even as he seeks spiritual power.

The “Nine Songs”

Throughout the “Nine Songs,” the Chu ci zhangju does not add any allegorical explanation to the male deities, but the Chu ci buzhu explicitly describes many of them as figures for the ruler. The Chu ci buzhu, in Hong Xingzu's notes or in citations of the Five Ministers' Wenxuan commentary, identifies Donghuang Taiyi 東皇太一 (the Supreme One of Eastern Majesty), Yun zhong jun 雲中君 (the Lord in the Clouds), Da siming 大司命 (the Great Arbiter of Fate), Shao siming 少司命 (the Lesser Arbiter of Fate), Dong jun 東君 (the Lord of the East), and Shan gui 山鬼 (the Mountain Spirit) as figures for the ruler, even though none of these deities are allegorically explained in the Chu ci zhangju.Footnote 95 On the contrary, according to the Chu ci zhangju, Qu Yuan quite literally serves the Supreme One and the rest in order to demonstrate his spiritual cultivation and ritual mastery, while also taking the opportunity to lament that the king does not make use of this—Qu Yuan's—virtue.Footnote 96 Hebo 河伯 (the River Earl), far from being a figure for King Huai or any earthly ruler, is explained as Qu Yuan's divine friend in the Chu ci zhangju. It might not be a coincidence that the Chu ci zhangju explains Qu Yuan's intention to xiangyou 相友 (befriend) the god in the context of an explanation of why 女 (woman/women, or a variant of ru 汝) should be read as ru 汝 (you, informally). The reason, Wang Yi assures us, is because the Earl's rank is comparable to Qu Yuan's, so the informal address is acceptable in this case.Footnote 97 Hong Xingzu does not reject this reading and does not explain the Earl as a figure for a ruler. For not all of the “Nine Songs” deities are mortal kings in the Chu ci buzhu: Hong Xingzu asserts that Xiangjun 湘君 “the Lord of the Xiang River,” who in the Chu ci zhangju is male and not a figure for any other entity, is female, and moreover, applying Wang Yi's yin-yang formula, is “used as a figure for a worthy minister” (以喻賢臣).Footnote 98 The Chu ci zhangju rationalizes Qu Yuan's devotion to the “Nine Songs” deities to some extent, but only the Chu ci buzhu rationalizes the deities themselves.

The pattern for the exegesis of female deities is inconsistent in the “Nine Songs” commentary. However, the inconsistency is compensated for by the manner in which concrete historical referents are incorporated into the narrative. Many of the deities are referred to as jun 君 (lord), either as part of their name or in an address within the poem. The Chu ci zhangju, in contrast to later editions, never reads jun as a reference to a female deity; in cases where context would suggest that it ought to refer to a female deity, the Chu ci zhangju asserts that it refers distinctly to the ruler or specifically to King Huai. In the Chu ci zhangju the Lord of the Xiang River, who is often female in later interpretations including the Chu ci buzhu, is explicitly gendered male as the river-deity spouse of the two brides intended for Shun 舜.Footnote 99 He is not allegorical, although later in the Chu ci zhangju commentary the speaker of the poem turns his attention to a different “lord,” the King of Chu.Footnote 100 In others of these poems as well, historical referents may be incorporated, usually Qu Yuan and sometimes King Huai, but in the Chu ci zhangju exegeses they are never explicitly identified with the entreated gods and appear to be separate characters. In order to read the gods as figures for the ruler, we would have to take an extra interpretive step and presume that an allegorized epithet is the same as the titular god. For example, in “The Mountain Spirit,” King Huai enters the commentary only at the appearance of lingxiu (Spiritually Distant) in the primary text, and lingxiu is not identified as the female Mountain Spirit. The same is true of gongzi 公子 (my prince), who is identified as the quasi-historical Chu high official Prince Pepper, but not as the Mountain Spirit or any other character in the literal narrative context; and of jun (my lord/ruler), who is identified as King Huai, but never “used as a figure for King Huai.”Footnote 101 In some of the other “Nine Songs” poems that according to the Chu ci zhangju refer to male deities, jun and gongzi are explained as forms of address to the gods, but not as references to the ruler, King Huai, or Prince Pepper.Footnote 102 The interpolations of allegorized historical context are introduced directly in the pattern X wei 謂 (refers to) Y, and do not use the word yu 喻 (figure), which is so commonly used elsewhere in the Chu ci zhangju and in the allegorical explanations of these same deities in the Chu ci buzhu. The strictly literal reading of Wang Yi's commentary to the “Nine Songs” is that Qu Yuan actually performs or has mediums perform sumptuous rituals for various deities to purify and empower himself, while remaining mentally fixated on service to the king of Chu.

Xiang furen 湘夫人 (the Ladies of the Xiang River), are the most unambiguously female of the “Nine Songs” deities according to the name. In the Chu ci zhangju they are not explicitly allegorical in the manner of Fufei and the other “Li sao” goddesses, although they are in the Chu ci buzhu.Footnote 103 Instead, they are defined as the daughters of Yao 堯 and brides of Shun, and then compared to Qu Yuan throughout the commentary: like him, they are pure and loyal, they long for a sage-ruler, and they die by casting themselves into the water. The relative literalness of these divine women is justified in Wang Yi's hermeneutic, first, because the preface to the “Nine Songs” asserts that these poems are adaptations of vulgar folk rites (not completely original compositions like “Li sao”), and second, because the commentary makes it appear that Qu Yuan contemplates these goddesses as models of self-cultivation, and does not court them as he does the goddesses in “Li sao.”Footnote 104 In fact, at the end of the poem, the point at which the speaker leaves behind parts of his or her clothing and prepares an offering of fragrant plants, apparently as love tokens and/or to requite the goddess's blessing,Footnote 105 the Chu ci zhangju commentary suddenly changes theme to assert that Qu Yuan seeks for noble worthies in reclusion.Footnote 106

The other probably female deity of the “Nine Songs” is the Mountain Spirit.Footnote 107 “The Mountain Spirit” is the most strongly allegorized poem of all the “Nine Songs,” although the spirit herself is not explicitly allegorical in the Chu ci zhangju. The commentary reads King Huai for “Spiritually Distant,” Prince Pepper for “my prince,” and forces of nature as political agents, as in the “Li sao” exegesis. In the beginning, it presents the goddess as a mirror for Qu Yuan, who cultivates his purity in obscurity in the mountain wilds, like her. But after the appearance of Spiritually Distant, the commentary seems to forget about the Mountain Spirit and to the end of the poem refers everything to Qu Yuan and his political career. In the first line, the ren 人(person) of “As if someone is here—on the mountainside” (若有人兮山之阿)Footnote 108 is referred to the Mountain Spirit, but later the “person” of “The person in the mountains—fragrant as sweet pollia” (山中人兮芳杜若) is referred to Qu Yuan.Footnote 109 A rigorous allegorical reading of all of the poem's nature imagery follows the penultimate line of the poem, and although it does not go so far as to assert that the Mountain Spirit is a figure for a worthy recluse, it makes it clear that the whole poem can be understood as a coded critique of worldly governance on the pattern of Wang Yi's “Li sao” interpretation.Footnote 110

A hint toward, or pre-figuration of, the allegorical interpretation of male deities is found in the Chu ci zhangju treatment of the solar god Lord of the East. Near the end of the poem the Chu ci zhangju commentary notes that “the sun rules as a true king” (日爲王者). This may appear allegorical. But the name of the deity is jun (lord/ruler), the Chu ci zhangju explanation does not use the word yu (figure), and the exegesis is concerned with general principles of kingship rather than with identifying a covert reference to a specific king. Indeed, although the Chu ci zhangju does bring King Huai and other historical figures into the “Nine Songs” commentary at several points, in the “Lord of the East” commentary, no historical actors are mentioned at all. Preceding comments literally narrate the descent of rishen 日神 (the sun deity) with no hint of allegory, and in following comments ri 日 (the sun), not wang 王 (the king), is the grammatical subject, who serves as a model, not a figure, for the ruler.

日爲王者,王者受命,必誅貪殘,故曰舉長矢,射天狼,言君當誅惡也。

The sun rules as a king. A kingly ruler receives a mandate, and accordingly must eliminate rapacity and malice, so it says “raise the long arrow and shoot the Wolf of Heaven.” This means the ruler must eliminate evil.

日誅惡以後,復循道而退,下入太陰之中,不伐其功也。

After the sun has eliminated evil, he resumes his natural way and retires, descending into the great darkness [where the spirits dwell], and does not boast of his deed.Footnote 111

The shooting of the Wolf of Heaven, identified with the star Sirius, refers to symbolic actions performed in the ritual that this text or its ur-text accompanied.Footnote 112 That Hong Xingzu cites a Yi jing 易經 commentary in the exegesis of this passage further suggests the mytho-ritual nature of the interpretation.Footnote 113 This interpretation is, strictly speaking, prognostic, not allegorical in the sense of the discussion above. But it may have provided a clue for later interpreters.

The effect of the Chu ci zhangju treatment of the “Nine Songs” is to affirm Qu Yuan's spiritual cultivation and demonstrate his commitment to his king. It feminizes Qu Yuan by comparison with the purity and physical beauty of the goddesses, but also distances him from what appears to be erotic contact with divine women. To keep him politically engaged while religiously active, it makes the reading of divine women in “Li sao” as ministers or courtiers in reclusion—that is, as men like Qu Yuan, who cultivates his spiritual power in exile and compares himself to a woman in the king's harem—appear to derive naturally from Qu Yuan's own intention in repurposing the “Nine Songs.”

The “Nine Compositions”

The words jun (lord/ruler) and chen (servant/minister) do not appear at all in the primary text of “Li sao,” but they appear twenty-two and six times, respectively, in the “Nine Compositions.” Thus, the tendency to explain “Li sao” and “Nine Songs” as political texts is not ungrounded; the Chu ci zhangju simply reads “Li sao” and “Nine Songs” as if they are more like the “Nine Compositions” and derivative works, and less like the early summoning poems and the liturgical music in which the “Nine Songs” are rooted. For in the Compositions, the metaphorical political narrative is apparently immanent and appropriate. Moreover, although there are feminized speakers, beautiful ladies, and matchmakers in the Compositions, there are no goddesses. The closest resemblance to a goddess is cini 雌蜺 (female rainbow), in “Bei hui feng” 悲回風. “Female rainbow” is identified as a goddess elsewhere in the Chu ci zhangju, but this word can only be understood as a spatial marker rather than a deity in this case: “I ascend the steep cliffs of a mountain—Rest at the highest point of the rainbow” (上高巖之峭岸兮,處雌蜺之標顛).Footnote 114 Even if we translate it “rest at the highest point of Goddess Rainbow,” the speaker is clearly using the name to evoke a physical space, not to refer to a deity.Footnote 115 All the deities in the Compositions are male, and there are far more historical ministers than deities.

The first of the Compositions, “Xi song” 惜誦, opens with a brief address to the gods in the manner of the “Li sao” oath to the Nine Heavens, and then turns to a litany of complaint that twice includes the unambiguously political common expression shi jun 事君 (to serve my lord). After this introduction of the main theme, the speaker relates a meng 夢 (dream) of climbing to Heaven. The Chu ci zhangju explains, expanding on the words of the lishen 厲神 (afflicted spirit) who divines the dream,Footnote 116 that trying to climb to heaven is similar to wanting to serve his lord: both are futile efforts.Footnote 117 The explicit framing of dream interpretation immanently suggests secondary, figural significance and comes immediately after the “serving the ruler” theme has explicitly appeared in the primary text. There is no dream in “Li sao,” and neither is there a “lord/ruler”; but it becomes easy to interpolate the rational government service narrative based on thematic similarities with the “Compositions” if one assumes as Wang Yi did that they were written by the same person to express the same intention.

Similarly, in “Ai Ying” 哀郢, the Chu ci zhangju explains that huangtian 皇天 (August Heaven) “is used as an evocation of the ruler” (以興於君).Footnote 118 As in “Li sao,” this does not identify Heaven as the ruler, but it places them in narrative juxtaposition in the manner of Shi jing exegesis, implying the ruler will soon appear in the text. It also evokes the distinctly imperial discourse of huangdi 皇帝 and tianzi 天子, August God-Emperor-as-Son of Heaven. In this context, the juxtaposition is largely borne out by the original text, for the “evocative juxtaposition” of “August Heaven” is immediately followed by a description of the common people being forced to flee the disaster that has befallen the kingdom of Chu, the clear implication being that the king is losing Heaven's mandate.

The most explicit correlation of cosmos and empire is in the Chu ci zhangju presentation of “She Jiang” 涉江, in which the Kunlun 崑崙 mountain range is equated with the sacred mingtang 明堂 (Bright Hall).Footnote 119 This correlation allows the reader to map the mythical macrocosmos of the Chu ci onto the magical microcosmic space, itself constructed in the Han according to early myths of divinized politics, in which the emperor mediated between Heaven and Earth to regulate his realm.Footnote 120 In “She Jiang,” the only deity mentioned is Chonghua (i.e., the sage god-emperor Shun), but there are many lines dedicated to listing mortal exemplars of ministerial rectitude such as Bi Gan 比干 and Wu Zixu 伍子胥. It is not terribly jarring, then, that the Chu ci zhangju commentary on this poem cites a variant interpretation of a line that explains shan 山 (mountains) as a figure for ministers, and ri 日 (sun) as a figure for the ruler.Footnote 121 What is surprising is that the Chu ci zhangju commentary asserts this latter explicit allegory as huo 或 (someone else's) interpretation, and does not explicitly extend the narrative of Chu government to the interpretation of the male deities in “Li sao.” Of course, doing so would imply that the speaker of “Li sao,” who commands gods, magical beasts, and forces of nature including the very sun, is not exactly subservient to the emperor.

“Far Roaming”

The gendered orientation of Wang Yi's hermeneutic is most obvious in the Chu ci zhangju treatment of “Far Roaming,” which preserves a commentary composed earlier and only edited by Wang Yi.Footnote 122 This commentary is written in rhyming verse and expounds the spiritual language quite literally—even the various female deities that enter the narrative at the end of the text. And yet, Wang Yi's occasional additions to the commentary, which are written in the same prose style as the Chu ci zhangju commentary on “Li sao,” “Nine Songs,” and “Nine Compositions,” do allegorize these female deities. Wang Yi does not explain them as substitutions for mortal men as in “Li sao,” but he does humanize them and give them an unambiguously subservient role in worldly politics. The journey of “Far Roaming” shares a great deal of its narrative structure, mytho-geographic space, and deities with “Li sao,” including considerable shared language. The major difference is that the speaker of “Far Roaming” expresses hardly a thought for earthly politics, and the journey is more successful: the heavenly gates are opened, and the journey proceeds to the ultimate destination without worldly obstruction. No female deities appear until the very end of the journey, and up to this point the Chu ci zhangju, mostly in the older pattern of rhyming tetrasyllabic verse commentary, simply elaborates poetically on the transcendent spiritual glories of the quest. But when the speaker has triumphantly arrived in Heaven with his entourage of divine beings and is enjoying the moment, he suddenly glimpses the jiuxiang 舊鄉 (old home), just as in the narrative of “Li sao” and with extremely similar language. It is at this point that King Huai appears in the exegesis, in a comment on a line that also includes the word huai: “My servants yearn and my heart grieves” (僕夫余心悲兮). The commentary simply supplies King Huai as the object of grief and does not attach him via any figural deity: “[I] long for my fatherland, and mourn for King Huai” (思我祖宗,哀懷王也).Footnote 123 King Huai is apparently already deceased at this point, and his memory is only a worldly distraction from the speaker's true goal of transcendence. In “Far Roaming,” unlike in “Li sao,” this goal is explicitly attained, and it is attained with the accompaniment of female deities. In the primary text, the speaker and the goddesses have a relatively cooperative relationship: “I convey a message by phoenix to welcome Fufei” (騰告鸞鳥迎宓妃); “The Two Maidens direct the Nine Shao songs” (二女御九韶歌); and “Goddess Rainbow arcs gracefully, curving around” (雌蜺便娟㠯增撓). The goddesses are part of a scene of harmonious general revelry with sexual overtones.Footnote 124 It is only at this point, and only in the newer prose commentary, that the Chu ci zhangju supplies a truly allegorical interpretation:

屈原得祝融止己,即時還車,將即中土,乃使仁賢若鸞鳳之人,因迎貞女若洛水之神,使達己於聖君,德若黃帝,帝堯者,欲與建德成化,制禮樂,以安黎庶也。

Qu Yuan was stopped by [the god] Zhurong, and having turned back his chariot he is about to go to the earth of the middle kingdom; so he commands benevolent and worthy men like phoenixes, and then welcomes chaste women like the spirit of the Luo River [Fufei], making them advance him to a sagely ruler—one whose power-virtue is like that of the Yellow God-Emperor and God-Emperor Yao, who wants to join together to establish virtue and complete the moralization of the realm, to regulate the rites and the music, and thus to pacify the common people.Footnote 125

In the commentary, the phoenixes and goddesses are substitutions for mortal humans. As the goddesses are simply well-placed servants that Qu Yuan can use to advance himself to the real power-holder, the comment brings to mind the marriage politics and inner-palace manipulations of the Han imperial court. The goddesses also have a much more explicitly subservient role in the commentary than in the primary text; whereas in the main text only the phoenixes have the role of go-between, in the commentary Fufei is reduced to the status of intermediary as well.

Hong Xingzu notes that the ending of “Far Roaming” is unique in that speaker doesn't plan to commit suicide.Footnote 126 This assumes that “Li sao” is an exclusively political declaration of suicidal intent, an interpretation that originates in the Chu ci zhangju. But we could just as well interpret “Li sao” through “Far Roaming.” Considering that after being stopped by Zhurong 祝融 the “Roaming” speaker enters a realm of pervasive yin imagery (water, flood dragons, cold, north, and darkness are dominant in this section), the ending may be seen as quite similar to the “Li sao” conclusion. If Pengxian in fact refers to Peng and Xian, ancestral spirit-mediums or realized spiritual adepts as has been proposed, then the “Li sao” speaker's departure, even his presumed plunge into the water, can be a seen as a pursuit of ancestors, spirits, or immortals: as an apotheosis, an attainment of god-like power in the manner of “Far Roaming.”Footnote 127 Moreover, in “Far Roaming” it is only at this point that the speaker encounters female deities, and he encounters them as partners in transcendent aesthetic enjoyment: this amoral and erotic implication is foreclosed in the Chu ci zhangju allegorical commentary on “Li sao” and “Far Roaming.”

Conclusions

I will make a final point about the interpretation of divine beings that are explained allegorically, although they might not be considered full gods and goddesses: the dragons and phoenixes. The similarity of the Chu ci zhangju exegeses of the phoenixes and of the goddesses suggests that for Wang Yi, the divine women are mythology, not religion. The dragons are inconsistently allegorical; sometimes they represent the speaker's “dragon-like power-virtue” (德如龍) but sometimes they are treated quite literally.Footnote 128 They are not quite referents to something else beyond the literal narrative, since the speaker has asserted his continual cultivation of virtue, and one must assume only a god or highly cultivated person could ride dragons. The dragons are like the lesser male deities of “Li sao”—only weakly allegorical. The phoenixes and related magical birds—yi 鷖, luan 鸞, huang 皇 (凰), fengniao 鳳鳥, fenghuang 鳳皇—are either concretely related to the speaker's chariotFootnote 129 or figures for virtuous shi 士 (courtiers) that Qu Yuan commands.Footnote 130 In general, birds are strongly allegorical in the same way as the female deities. From a structural perspective, the first bird mentioned in the Chu ci is clearly an immanent metaphor, which may have suggested extending figural interpretation to the other similar images.Footnote 131 But why, hermeneutically or ideologically, would phoenixes would require more interpretative steps than dragons? Perhaps, due to their conventional associations with marriage and erotic longing, or with the soul's journey after death, the birds are not suitable company for a Han gentleman on an ultimately political outing.Footnote 132 Or perhaps, since phoenixes are a particularly auspicious political omen, they simply don't fit the end-times narrative of Qu Yuan's exile under benighted Kings Huai and Xiang. In any case, it is clear that goddesses and phoenixes are similar in that they require more figural explanation than the male gods, or, to state it more strongly, that the goddesses, in contrast to the gods, are classed among the mythological creatures.Footnote 133

Wang Yi's commentary does not attempt to completely fictionalize the spiritual and religious imagery of the poems, but it does subordinate possible references to erotic spiritual pursuits or yinsi 淫祀 (lewd rites)Footnote 134 to the ritual propriety of late Han dynastic ideology; that is, from early discourses in which spirit-mediums may erotically court female deities as part of spiritual appeasement and spiritual adepts or departed souls may encounter goddesses on the journey of transcendence, into a discourse in which a male courtier may appropriately cultivate his spiritual powers only chastely and with the ultimate intention of serving the state. The goddesses of the Chu ci, wanton and elusive, excluded from the orthodox Han canon, dismissed by the state ritualists, and criticized by Wang Yi's formidable predecessor Ban Gu, do not easily fit into this vision. This is not an extreme case of allegoresis, as slippage and blurring between the language of politics and the language of religion is immanent in the text and organic to early Chinese culture. But considering the relative rationalization of politics that had occurred between Warring States Chu and the Eastern Han, especially the official marginalization of wu spirit-mediums and Chu religion, Wang Yi's odd insistence that the female deities of “Li sao” are not deities like the male deities appears to be a case of anachronistic moralization.Footnote 135 This moralization is founded on just a few words: Ban Gu's example of a particular goddess as evidence for the non-canonical (immoral) import of “Li sao,” and a canonical precedent for equating (female) with chen (servant/minister). When combined with the interpretation of “Peng Xian” as a loyal minister rather than a spiritual exemplar, Wang Yi was able to use the goddesses to demonstrate that Qu Yuan always served his lord single-mindedly until the end, refuting Ban Gu's criticism and bringing the Chu ci into the canon of ethical political texts.

We don't have to read “Li sao” the way Wang Yi did, or the “Nine Songs” the way Hong Xingzu thought Wang Yi did; even Wang Yi didn't read the whole Chu ci that way. If we don't correlate gender with yin-yang, we can arrive at an interpretation like Zhu Xi's, in which the divine women are purely rhetorical, mere tokens for the true object pursued: a sought-after ideal ruler.Footnote 136 The possibility of reading this way is introduced in Wang Yi's commentary, in which the goddesses are already not themselves. On the other hand, we can also look for a perhaps earlier and more compelling reading, in which the female deities are real and ardently sought, and their presence in the world of the poem is fully coherent and meaningful in itself. In this reading, the goddesses do not need to be replaced with ministers or hermits, but can be exemplars of beneficent or vengeful power, divine guides for the journey to heaven, or ancestral models of martyrdom and apotheosis. If allegories are the mirrors of ideology,Footnote 137 then reexamining foundational texts such as the Chu ci zhangju, excavating them for what lies beneath the accumulated ideological performances of history, allows us to reimagine the past, and to discern how interpretation and writing participate in the discursive transformation of society and spiritual life. Wang Yi's edition of the Chu ci is a highly influential case of interpreting culture to serve the interests of the state; doing so required reading female spiritual power out of history.

Footnotes

The English abstract has been added. An erratum notice detailing this change has also been published (https://doi.org/10.1017/eac.2018.24)

I would like to thank the members of my panel at the 2016 AAS conference, Michael Schimmelpfennig, Wang Yugen, Feng Jun, and the journal's anonymous reviewers, for their informative suggestions and criticism on earlier versions of this work. All errors that remain are my own.

References

1. For what little is known about Wang Yi’s life, see Linggeng, Huang 黃靈庚, Chu ci zhangju shuzheng 楚辭章句疏證 (Beijing: Zhonghua shuju, 2007), 14Google Scholar; and Schimmelpfennig, Michael, “The Quest for a Classic: Wang Yi and the Exegetical Prehistory of his Commentary to the ‘Songs of Chu’,” Early China 29 (2004), 111–62CrossRefGoogle Scholar, at 141.

2. Whether or not the experience of a spirit-filled cosmos as discursively constructed in the Chu ci is properly called “religious” is a significant question. For this paper I accept Mu-chou Poo’s broad definition of religion as “belief in the existence of extra-human powers”; see In Search of Personal Welfare: A View of Ancient Chinese Religion (Albany: State University of New York Press, 1998), 56Google Scholar.

3. Linggeng, Huang, Chu ci zhangju shuzheng and Chu ci yu jianbo wenxian 楚辭與簡帛文獻 (Beijing: Renmin, 2011)Google Scholar; Misao, Ishikawa 石川三佐男, “Cong Chudi chutu bohua fenxi Chu ci ‘Jiu ge’ de shijie” 從楚地出土帛畫分析《楚辭.九歌》的世界 in Zhongguo Chu ci xue 1 中國楚辭學第一輯, ed. China Qu Yuan Studies Association (Beijing: Xueyuan, 2002), 205–18Google Scholar, and Gudai Chu wangguo guoce yu Chu ci ge pian ji Zhanguo Chu zhushu deng wenxan de guanxi” 古代楚王國國策與《楚辭》各篇及戰國楚竹書等文獻的關係, Zhongguo Chu ci xue 19 (2013), 137Google Scholar; Cook, Constance, Death in Ancient China: The Tale of One Man’s Journey (Leiden: Brill, 2006), 2627CrossRefGoogle Scholar.

4. Cook, Death in Ancient China, 41–42; Major, John S., “Characteristics of Late Chu Religion,” in Defining Chu: Image and Reality in Ancient China, ed. Cook, and Major, (Honolulu: University of Hawai‘i Press, 1999)Google Scholar; Poo, Mu-chou, “Imperial Order and Local Variation: The Culture of Ghost in Early Imperial China,” Acta Orientalia Academiae Scientiarum Hungaricae 56.2–4 (2003): 295308Google Scholar; Guolong, Lai, Excavating the Afterlife: The Archaeology of Early Chinese Religion (Seattle: University of Washington Press, 2015)Google Scholar; and Sou, Daniel, “Living with Ghosts and Deities in the Qin 秦 State: Methods of Exorcism from ‘Jie 詰’ in the Shuihudi 睡虎地 Manuscript,” in From Mulberry Leaves to Silk Scrolls: New Approaches to the Study of Asian Manuscript Traditions, ed. McDaniel, Justin Thomas and Ransom, Lynn (Philadelphia: University of Pennsylvania Libraries, 2015)Google Scholar.

5. Most scholars (I dare say all of them) follow Wang Yi’s preface in understanding the Nine Songs as more or less rooted in Chu rituals; the issues of debate are whether the rituals were popular or courtly, purely religious or politically coded, and where exactly the texts fall between the poles of transcribed liturgy and literary imitation. For an overview, see Xiaoqiong, Li 李曉瓊, “Chutu wenxian yu Chu ci ‘Jiu ge’ jin shi nian zhi yanjiu” 出土文獻與《楚辭・九歌》近十年之研究, Journal of Chuxiong Normal University 楚雄師範學院學報 23.11 (2008), 3942Google Scholar.

6. See Kroll, Paul W., “On ‘Far Roaming’,” The Journal of the American Oriental Society 116.4 (1996), 653–69CrossRefGoogle Scholar.

7. It is likely that in fact “Far Roaming” derives the journey topos and much of its form from “Li sao,” but the dating of the texts is much in debate. See Kroll, “On ‘Far Roaming’,” 653–54.

8. Some or all of the “Nine Compositions” may have been composed after “Li sao” and can be classed among the derivative “Li sao” imitation works, but Wang Yi attributed them all to Qu Yuan. See Hawkes, David, Songs of the South: An Anthology of Ancient Chinese Poetry by Qu Yuan and Other Poets (Harmondsworth: Penguin, 1985), 152–54Google Scholar. See also 過常寶, Guo Changbao, Chu ci yu yuanshi zongjiao 楚辭與原始宗教 (Beijing: Zhongguo renmin daxue, 2014), 117Google Scholar.

9. Pauline Yu argues that the influence of Wang Yi’s interpretation even extended beyond the Chu ci tradition to shape the hermeneutics of the Shi jing, in The Reading of Imagery in the Chinese Poetic Tradition (Princeton: Princeton University Press, 1987), 15Google Scholar.

10. On the significance of the allegorical reading of the Chu ci in terms of political discourse and literary history, see Yuyu, Zheng 鄭毓瑜, Xingbie yu jiaguo: Han Jin cifu de Chu Sao lunshu 性别與家國:漢晉辭賦的楚騷論述 (Shanghai: Sanlian shudian, 2006), 2Google Scholar, 4, 185. Archaeological evidence has been used to vigorously refute the allegorical interpretation by, for example, Huang Linggeng in Chu ci yu jianbo wenxian; and Ishikawa in “Chu ci ‘Jiu ge’ de shijie” and “Gudai Chu wangguo guoce.” See also Sukhu, Gopal, The Shaman and the Heresiarch: A New Interpretation of the Li sao (Ithaca: State University of New York Press, 2012)Google Scholar.

11. On the classification and relation of “classic” and “commentary” in the Chu ci zhangju, see Naoya, Miyano 宮野直也, “O Itsu Soji shoku no chushoku taito ni tsuite” 王逸『楚辞章句』の注釋態度について, Nippon Chūgoku Gakkai hō 日本中国学会報 39 (1987), 8497Google Scholar, 90. On the textual history of the Chu ci zhangju and the sequence of the anthology, see Hawkes, Songs of the South, 28–41, and Huang Linggeng, Chu ci zhangju shuzheng, 6–7.

12. The unsatisfactory yet essential quality of the Chu ci zhangju is canonized in the preface of the Qing Siku quanshu 四庫全書 edition of Chu ci zhangju. For a strong criticism of the allegorical interpretation, see also Anne Birrell’s review of Three Elegies of Ch’u by Waters, Geoffrey in Bulletin of the School of Oriental and African Studies, University of London 50.3 (1987), 584–86Google Scholar.

13. Xi, Zhu, Chu ci jizhu 楚辭集注 (Shanghai: Shanghai guji, 1979), 2Google Scholar, 15.

14. Hawkes, Songs of the South, 28.

15. Kroll makes a similar observation about Hawkes’ translation (“On ‘Far Roaming’,” 656), and of course the same could be said for most translations of classical texts.

16. Hawkes, Songs of the South, 68; Xingzu, Hong, Chu ci buzhu (Beijing: Zhonghua shuju, 2010), 3Google Scholar; Huang Linggeng, Chu ci zhangju shuzheng, 31. On jiang, see Guo Changbao, Chu ci yu yuanshi zongjiao, 79; and Sukhu, The Shaman and the Heresiarch, 35–37, 39–40. Other modern translations fall between the generality of jiang and the precision of Wang Yi’s gloss: see, e.g., “was born,” in Watson, Burton, Columbia Book of Chinese Poetry: From Early times to the Thirteenth Century (New York: Columbia University Press, 1984), 54Google Scholar; “descended,” in Rouzer, Paul, Articulated Ladies: Gender and the Male Community in Early Chinese Texts (Cambridge, MA: Harvard University Press, 2001), 5CrossRefGoogle Scholar; “came down,” in Owen, Stephen, An Anthology of Chinese Literature: Beginnings to 1911 (New York: W.W. Norton, 1996), 162Google Scholar.

17. See Sukhu, The Shaman and the Heresiarch, 81–85, and the trenchant review by Schimmelpfennig, Michael in The Journal of Asian Studies 73.4 (2014), 122–23CrossRefGoogle Scholar.

18. See Schimmelpfennig, “Quest for a Classic,” 140–41, 162; cf. Hawkes, Songs of the South, 29.

19. Schimmelpfennig, “Quest for a Classic,” 140–41, 162; Guo Changbao, Chu ci yu yuanshi zongjiao, 120–28.

20. Cited in Hong Xingzu, Chu ci buzhu, 49–51; see also Huang Linggeng, Chu ci zhangju shuzheng, 1964.

21. Schimmelpfennig, “Quest for a Classic,” 139, 142; and Naoko, Yada 矢田尚子, “Handai Qu Yuan pingjia zhi bianqian” 漢代屈原評價之變遷 in Zhongguo Chu ci xue 19 (2013), 283–94Google Scholar. On the religious world of the scholar-official in the late Han—the milieu of Wang Yi—and its mingling of Confucian and proto-Daoist forms, see Bujard, Marianne, “State and Local Cults in Han Religion,” in Early Chinese Religion, Part I, vol. 2, ed. Lagerwey, John and Kalinowski, Marc (Leiden: Brill, 2009), 810–11Google Scholar.

22. On the status of female deities, see Birrell, Anne, Chinese Mythology: An Introduction (Baltimore: Johns Hopkins University Press, 1993), 162–64Google Scholar. On the discursive restriction of women’s agency, see Nylan, Michael, “Golden Spindles and Axes: Elite Women in the Achaemenid and Han Empires,” in The Sage and the Second Sex: Confucianism, Ethics, and Gender, ed. Li, Chenyang (Chicago: Open Court, 2000), 212–13Google Scholar; Lisa Raphals, “Gendered Virtue Reconsidered: Notes from the Warring States and Han,” in The Sage and the Second Sex, 236–37; Esther Sunkyung Park (Esther S. Klein), Women of Action in Zuozhuan and Shiji (M.A. Thesis, University of Oregon, 2002); and Zheng, Xiucai, “From Zuozhuan to Shiji: Changes in Gender Representation in Sima Qian’s Rewriting of Stories,” Chinese Literature: Essays, Articles, Reviews 36 (2014), 149–74Google Scholar.

23. See Wang, Aihe, Cosmology and Political Culture in Early China (Cambridge: Cambridge University Press, 2000)CrossRefGoogle Scholar; Joachim Gentz, “Language of Heaven, Exegetical Skepticism and the Re-Insertion of Religious Concepts in the Gongyang Tradition,” in Early Chinese Religion, Part I, vol. 2, especially 836–38; and Puett, Michael, To Become a God: Cosmology, Sacrifice, and Self-Divinization in Early China (Cambridge, MA: Harvard University Press, 2002)Google Scholar. For specific reference to imagery and symbolism, see Tseng, Lillian Lan-ying, Picturing Heaven in Early China (Cambridge, MA: Harvard University Press, 2011), 36CrossRefGoogle Scholar.

24. On the differences between Chu and central plains religious cultures during the Zhou era, see Guo Changbao, Chu ci yu yuanshi zongjiao, chap. 1.

25. Puett, To Become a God, 117–19, 223–24, 245, 318–19.

26. Puett, To Become a God, 317, 321.

27. The polysemy of the words di 帝 and huang 皇 is significant, as their appropriation by human rulers at the end of the Warring States period corresponds to a conceptual shift in the understanding of gods and kings. I have translated di 帝 as god-emperor rather than “thearch” to make both the pre-imperial and imperial meanings of the word visible. Huang 皇, jun 君, and even chen 臣 and shi 士 raise similar philological issues; I have done my best to translate each instance appropriately according to the historical context of its use. Cf. Sterckx, Roel, “Religious Practice in Qin and Han,” in China’s Early Empires: A Re-appraisal, ed. Nylan, Michael and Loewe, Michael (Cambridge: Cambridge University Press, 2010), 427Google Scholar.

28. Puett, To Become a God, 319; see also Raz, Gil, “Chinese Religion from the Han to the Six Dynasties,” in The Wiley-Blackwell Companion to Chinese Religions, ed. Nadeau, Randall Laird (Sussex: John Wiley & Sons, 2012), 62Google Scholar.

29. Schimmelpfennig, “Quest for a Classic,” 121; see also Tseng, Picturing Heaven in Early China, 149–52, 167. This essay does not attempt to clearly differentiate immortality, transcendence, and apotheosis, first, because these are English terms projected onto Chinese discourse, and second, because the distinctions between various kinds of transcendence in Chinese discourse are by no means consistent and clear-cut; see Poo, In Search of Personal Welfare, 213. Puett’s sharp distinctions between these terms are not always solidly grounded, as his reviewers have pointed out. See reviews by Nylan, Michael in The American Historical Review 108.4 (2003), 1117–18CrossRefGoogle Scholar; and Loewe, Michael in Journal of the American Academy of Religion 73.4 (2005), 1249–51CrossRefGoogle Scholar.

30. Brown, Miranda, “Sons and Mothers in Warring States and Han China, 453 bce.–220 ce,” Nan Nü 5.2 (2003), 137–69CrossRefGoogle Scholar; see 167 and passim.

31. Major, John S., Heaven and Earth in Early Han Thought: Chapters Three, Four, and Five of the “Huainanzi” (Albany: State University of New York Press, 1993), 51Google Scholar.

32. See Guo Changbao, Chu ci yu yuanshi zongjiao, 124–26. Yin Guoming 殷國明 argues that the original gender ambiguity and eroticism of “Li sao” were reinterpreted and stabilized as representations of political loyalty and patriotism due to patriarchal ideology; see Yin Guoming, “Bian ‘Sao’” 辨 “騷” Huadong shifan daxue xuebao 華東師範大學學報 zhexue shehui kexue ban 哲學社會科學版 2010.3, 52–56.

33. Hong Xingzu, Chu ci buzhu, 196; Huang Linggeng, Chu ci zhangju shuzheng, 739.

34. Hong Xingzu, Chu ci buzhu, 3, 13, 47; Huang Linggeng, Chu ci zhangju shuzheng, 31, 175, 550.

35. E.g., Huang Linggeng, Chu ci zhangju shuzheng, 550–53; Guo Changbao, Chu ci yu yuanshi zongjiao, 77–83, 88–89.

36. On the gradual development of this cosmology, see Guangde, Sun 孫廣德, Xian Qin liang Han yinyang wuxing shuo de zhengzhi sixiang 先秦兩漢陰陽五行說的政治思想 (Taipei: Taiwan shangwu yinshuguan, 1993)Google Scholar, chaps. 1 and 2; Queen, Sarah A., From Chronicle to Canon: The Hermeneutics of the Spring and Autumn, According to Tung Chung-Shu (Cambridge: Cambridge University Press, 1996)CrossRefGoogle Scholar, chap. 9; Michael Nylan, “Yin-yang, Five Phases, and qi,” in China’s Early Empires, 409; and Aihe Wang, Cosmology and Political Culture, 124–25. With respect to gender in particular, see Lihua, Gu 顧麗華, Handai funü shenghuo qingtai 漢代婦女生活情態 (Beijing: Shehui kexue wenxian, 2012), 2836Google Scholar, 79–80; Raphals, Lisa A., Sharing the Light: Representations of Women and Virtue in Early China (Albany: State University of New York Press, 1998)Google Scholar, chap. 6; and Hinsch, Bret, Women in Early Imperial China, 2nd ed. (Lanham, MD: Rowman & Littlefield, 2011)Google Scholar, chap. 8.

37. Beata Grant, “Gender,” in The Wiley-Blackwell Companion to Chinese Religions, 399; Raphals, Sharing the Light, 153–65; Jialin, Bao 鮑家麟, “Yinyang xueshuo yu funü diwei” 陰陽學說與婦女地位, in Zhongguo funü shi lunji xuji 中國婦女史論集續集, ed. Jialin, Bao (Taipei: Daoxiang, 1991), 3754Google Scholar.

38. Raphals, Sharing the Light, chaps. 9–10; Zhou, Yiqun, “Virtue and Talent: Women and Fushi in Early China,” Nan Nü 5.1 (2003), 142Google Scholar.

39. Grant, “Gender,” 400, 404–407; Fu-shi Lin, “The Image and Status of Shamans in Ancient China,” in Early Chinese Religion, Part I, vol. 1, 437, 440. According to Gu Lihua, in the Han female spirit-mediums were ubiquitous and influential, but also of low social status; Gu Lihua, Handai funü shenghuo qingtai, 164–66.

40. Raz, “Chinese Religion from the Han to the Six Dynasties,” 55; Birrell, Chinese Mythology, 162, 182. Gong Weiying 龔維英argues that Donghuang Taiyi 東皇太一, the first titular deity of the Chu ci “Nine Songs,” was gradually transformed from female to male; see “‘Jiu ge’ zhushen Donghuang Taiyi xingbie kao” 《九歌》主神東皇太一性别考, Yunmeng xuekan 雲夢學刊 1990. 2, 8–11. On the transformation of the status of ruler and deities between the “Nine Songs” and Eastern Han court liturgy, see Taniguchi Hiroshi 谷口洋, “Xihan ‘Jiaosi ge’ shijiu zhang yu ‘Jiu ge’” “西漢《郊祀歌》十九章與《九歌》in Zhongguo Chu ci xue 19 (2013), 235–46.”

41. See Aihe Wang, Cosmology and Political Culture, 124–25; Harper, Donald, “Warring States Natural Philosophy and Occult Thought,” in The Cambridge History of Ancient China: From the Origins of Civilization to 221 BC, ed. Loewe, Michael and Shaughnessy, Edward L. (Cambridge: Cambridge University Press, 1999), 857Google Scholar, 866. For a thorough discussion of Wang Yi’s cosmological interpretation of the speaker’s supposed birth date, see Liangfu, Jiang 姜亮夫, Chu ci tonggu 楚辭通故 Part 1, in Jiang Liangfu quanji 姜亮夫全集, vol. 1 (Kunming: Yunnan renmin, 2002), 105–19Google Scholar.

42. See, e.g., Alice Yao on the highly variable construction of gender identity in ancestor cults of pre-dynastic China, in Engendering the Ancestors Through Death Ritual in Ancient China,” in The Oxford Handbook of the Archaeology of Death and Burial, ed. Tarlow, Sarah and Nilsson, Liv Stutz (Oxford: Oxford University Press, 2013), 592–93Google Scholar; and Nylan, “Golden Spindles and Axes.” See also Hinsch, Women in Early Imperial China, 168. On the idealizing, textual construction of gender identity, see Raphals, Sharing the Light, passim; and Rouzer, Articulated Ladies, 34.

43. Hong Xingzu, Chu ci buzhu, 48–50; Schimmelpfennig, “Quest for a Classic,” 135–38, 149–50.

44. See Cahill, Suzanne on the Queen Mother of the West 西王母 during the Han dynasty, in Transcendence & Divine Passion: The Queen Mother of the West in Medieval China (Stanford: Stanford University Press, 1993), 1724Google Scholar; on the dangerous power of her followers, see 21, 23. For an interpretation of the early Chu ci in terms of the Mawangdui paintings, see Ishikawa, “Chu ci ‘Jiu ge’ de shijie.” In Huainanzi, chap. 2: “Chu zhen xun” 俶真訓, 13, the zhenren 真人 “realized man” is shown to command many of the deities of “Li sao,” both male and female, in the course of spirit-travel, and in “Lan ming xun” 覽冥訓, 9 the goddess Nü Gua 女媧 ascends to heaven riding thunder and dragons, a display of brilliance and power that is strikingly similar to the ascents in “Li sao” and “Far Roaming.”

45. Schimmelpfennig, “Quest for a Classic,” 149; and Ichiro, Kominami 小南一郎, “Wang Yi Chu ci zhangju zai Handai Chu ci zhushi shi shang de diwei” 王逸《楚辭章句》在漢代《楚辭》注釋史上的地位, translated by Ping, Liu 劉萍, Guji zhengli yu yanjiu 古籍整理與研究 6 (1991), 277–85Google Scholar; cf. Kominami, , “O Itsu Soji shoku o megute: Kandai shoku no gaku no ichi sokumen” 王逸‘楚辭章句’をめぐって-漢代章句の學の一側面, Tōhō gakuhō 東方學報 63 (1991), 61114Google Scholar.

46. The question of allegory in Chinese literature has been taken up by Pauline Yu in The Reading of Imagery, Saussy, Haun in The Problem of a Chinese Aesthetic (Stanford: Stanford University Press, 1993)Google Scholar, and Longxi, Zhang in Allegoresis: Reading Canonical Literature East and West (Ithaca: Cornell University Press, 2005)Google Scholar.

47. For a paradigmatic example, see Waley, Arthur, trans., The Book of Songs (New York: Grove Press, 1960, originally published 1937)Google Scholar, Appendix I: “The Allegorical Interpretation”; with respect to the Chu ci, see Waley, , The Nine Songs (London: Allen and Unwin, 1955), 16Google Scholar. For the contrary view, cf. Waters, Geoffrey R., Three Elegies of Ch’u: An Introduction to the Traditional Interpretation of the Ch’u Tz’u (Madison: University of Wisconsin Press, 1985), 1619Google Scholar. Inappropriate or arbitrary (i.e., inorganic) collocation is central to the modern devaluation of allegory, as elucidated in Gadamer, Hans Georg, Truth and Method, 2nd edition (London: Bloomsbury, 2013), 6473Google Scholar: “allegory assumes not the kind of original metaphysical affinity that a symbol claims but rather a co-ordination created by convention and dogmatic agreement” (68). Strictly speaking, Gadamer and the modern literary critics he invokes are primarily concerned with allegory as a technique of original composition that reflects a tradition of dogmatic reading, while this article focuses on the historically earlier part of that formulation, the tradition of dogmatic reading (allegoresis) as applied to the Chu ci.

48. Hong Xingzu, Chu ci buzhu, 5, 9; Huang Linggeng, Chu ci zhangju shuzheng, 55, 112. There is an injunction regarding the wearing of fragrant sachets in the Liji 禮記 chapter “Neize” 內則 5; fragrant plants used in perfume and medicine were found in a Chu-area Western Han tomb at Mawangdui, and are now in the collection of the Hunan Provincial Museum.

49. Hong Xingzu, Chu ci buzhu, 75; Huang Linggeng, Chu ci zhangju shuzheng, 927.

50. Cf. Schimmelpfennig, “Quest for a Classic,” 138.

51. Poo, In Search of Personal Welfare, 5–7, 202, 207–8. Eugene Y. Wang demonstrates the difficulty of maintaining seemingly common-sense distinctions, such as that between real and unreal, with regard to Chu religious imagery in Ascend to Heaven or Stay in the Tomb?: Paintings in Mawangdui Tomb 1 and the Virtual Ritual of Revival in Second-Century B.C.E. China,” in Mortality in Traditional Chinese Thought, ed. Olberding, Amy and Ivanhoe, Philip J. (Albany: State University of New York Press, 2011), 7576Google Scholar. In Searching for Spirit: Shen and Sacrifice in Warring States and Han Ritual and Philosophy,” Extrême-Orient, Extrême-Occident 29 (2007), 2354CrossRefGoogle Scholar, Roel Serckx demonstrates the cultural construction and “non-fixity” of the spirit world (40). As a point of comparison, consider transubstantiation. A priest asserts that a cup of wine is really the blood of Christ; believers accept the literal truth of the assertion. An analysis of the wine will not help an observer understand the priest’s assertion, much less refute it, because the referent of “the blood of Christ” does not exist in a conventional mode of being. Analysis can, however, clarify the relation between the signs “wine” and “blood.”

52. Hong Xingzu, Chu ci buzhu, 1–2, Huang Linggeng, Chu ci zhangju shuzheng, 1–11.

53. Er yao 二姚 (The Two Yao [Women]) may appear to be an exception, as the commentary could be strictly interpreted to read them as mortal women rather than as goddesses, and it is less explicit in insisting that they are figures for ministers. But the transformation of ancestors into deities can be safely assumed, and certainly the allegorical narrative of this section has been made so abundantly clear by this point in the commentary that they must be included among the worthy ministers that Qu Yuan seeks in the comment to this line; see Hong Xingzu, Chu ci buzhu, 34; Huang Linggeng, Chu ci zhangju shuzheng, 412. See also Jiang Liangfu, Chu ci tonggu, 445. Cf. Huang Linggeng’s Chu ci yu jianbo wenxian, in which the women are divine ancestors pursued as models on the speaker’s way to his own deification (39, 92), and Ishikawa’s interpretation in “Gudai Chu wangguo guoce,” in which the speaker seeks to bring down the Mandate of Heaven, on behalf of his king, with assistance from goddess intermediaries (19, 36).

54. Recent scholarship makes the case that she should be understood as a wu 巫 spirit-medium; see Huang Linggeng, Chu ci zhangju shuzheng, 249–50; Guo Changbao, Chu ci yu yuanshi zongjiao, 97. In the “Nine Songs,” Nüxu is inserted into the poem via the zhangju commentary on the basis of language that is also used to describe her in “Li sao” (Hong Xingzu, Chu ci buzhu, 61; Huang Linggeng, Chu ci zhangju shuzheng, 813). According to Huang Linggeng’s analysis, Wang Yi himself may well have understood Nüxu as both an elder sister and a spirit-medium, but the zhangju only records the sister part, and that is the exclusive part discussed in later tradition. Cf. Sterckx: “Given that, in Han sources, spirit mediums are often female, one wonders whether the hostility or bias toward them might not have been inspired by an antipathy towards women possessing secretly transmitted knowledge rather than by outright disdain for the esoteric methods they endorsed” (“Religious Practice in Qin and Han,” 416–17). I translate wu as “spirit-medium” rather than “shaman,” which has become conventional in Chu ci scholarship, in order to avoid invoking inappropriate connotations from contemporary religious cultures whence the name and concept of shamanism is derived—running the risk, of course, of invoking equally inappropriate connotations from nineteenth-century Anglo-American Spiritualism, but at least that particular discourse is not as entrenched in Sinology. Cf. Boileau, Gilles, “Wu and Shaman,” Bulletin of the School of Oriental and African Studies 65.2 (2002), 350–78CrossRefGoogle Scholar.

55. See the commentary’s literal treatment of wu spirit-mediums throughout “Li sao” and the “Nine Songs,” many or all of whom must be assumed to be female given Wang Yi’s explicit gloss of wu in “Summoning the Soul” (Hong Xingzu, Chu ci buzhu, 197); and the literal treatment of female dancers in the summoning poems (e.g., Chu ci buzhu, 197–98, 205–6, 209–11, 221–23).

56. On attribution of the parts of the commentary, see Kominami, “Wang Yi Chu ci zhangju zai Handai,” 282–83; cf. “O Itsu Soji shoku o megute” for a slightly different version of the argument.

57. Hong Xingzu, Chu ci buzhu, 55; Huang Linggeng, Chu ci zhangju shuzheng, 745.

58. See Pauline Yu, The Reading of Imagery, chap. 2, 76 and passim. Cf. Miyano, “O Itsu Soji shoku,” 89–90.

59. Hong Xingzu, Chu ci buzhu, 2–3; Huang Linggeng, Chu ci zhangju shuzheng, 9–11. For a different translation of this passage, see Schimmelpfennig, “Quest for a Classic,” 150.

60. Guo Changbao reads the opening lines not as the speaker’s self-introduction, but as the speaker’s invocation of ancestral spirits; see Chu ci yu yuanshi zongjiao, 83, 86.

61. Hong Xingzu, Chu ci buzhu, 9; Huang Linggeng, Chu ci zhangju shuzheng, 112. Wang Yi reads this particular figure as metonymy: the lord of men wears fragrances, so he is yu 諭 “figured” by a fragrant herb.

62. For example, see Major’s discussion of the Kunlun mountains in Heaven and Earth, 150–58. For a comparison from the visual arts, see Tseng, Picturing Heaven in Early China, 200–205, 233.

63. For example, Guo Changbao has interpreted the whole poem as a lyric modeled on Chu rituals. While he assumes the biography of Qu Yuan to be the background of “Li sao,” he explains all the particular elements of the poem in terms of ritual acts and their attendant emotions, and he does not use them as substitutions for specific elements of a narrative of Chu government; see Chu ci yu yuanshi zongjiao, chap. 4, 90 and passim.

64. Hong Xingzu, Chu ci buzhu, 14, 64; Huang Linggeng, Chu ci zhangju shuzheng, 191, 832. The locus classicus for this formulation is in the Yi jing 易經 wenyan 文言 commentary on hexagram kun 坤: “The dark force [yin] possesses beauty but conceals it. So must a man be when entering the service of a king. He must avoid laying claim to the completed work. This is the way of the earth, the way of the wife, the way of one who serves [i.e., a minister]” (陰,雖有美含之,以從王事,弗敢成也。地道也,妻道也,臣道也). Translation of Cary F. Baynes and Richard Wilhelm (bracketed insertions mine, for clarity), in The I Ching or Book of Changes (Princeton: Princeton University Press, 1950, 1977), 394Google Scholar. Sun Guangde notes that this layer of interpretation is a very late addition to the Yi jing and that there is no corresponding list of “ways” for yang; see Yinyang wuxing shuo 18–19. The yang of heaven, husband, and ruler is easy to infer, but this textual absence in the canon may have discouraged Wang Yi from asserting the yang correspondences in his own work. See also Jiang Liangfu, Chu ci tonggu, 11.

65. Hong Xingzu, Chu ci buzhu, 30, 35, 42; Huang Linggeng, Chu ci zhangju shuzheng, 377, 430, 497. In some of these instances and in others throughout the texts, some of the 女 (woman/women/girls) could be read as ru 汝 (you); I have translated according to Wang Yi’s readings for clarity.

66. Hong Xingzu, Chu ci buzhu, 32; Huang Linggeng, Chu ci zhangju shuzheng, 400.

67. Hong Xingzu, Chu ci buzhu, 31; Huang Linggeng, Chu ci zhangju shuzheng, 385.

68. For a discussion of the discursive markedness of gender, see Judith Butler, Gender Trouble: Feminism and the Subversion of Identity (New York: Routledge, 1990), 12–17; and for specific reference to Chinese language (albeit as present in modern usage), see Farris, Catherine S., “Gender and Grammar in Chinese: With Implications for Language Universals,” Modern China 14.3 (1988), 277308CrossRefGoogle Scholar. The concept of markedness implicitly underlies Anne Birrell’s discussion of goddess nomenclature (Chinese Mythology, 163), as well as Hinsch’s discussion of cosmology in Women in Early Imperial China, chap. 8. The markedness of feminine deity names and the general androcentrism of Han dynasty discourse leaves little doubt that these unmarked deities would be de facto understood as male.

69. Hong Xingzu, Chu ci buzhu, 29; Huang Linggeng, Chu ci zhangju shuzheng, 367.

70. Hong Xingzu, Chu ci buzhu, 29; Huang Linggeng, Chu ci zhangju shuzheng, 367. On Han representations of the Gate of Heaven and its religious significance, see Tseng, Picturing Heaven in Early China, 205–33.

71. Zhu Xi, Chu ci jizhu, 16.

72. Hong Xingzu, Chu ci buzhu, 26–27; Huang Linggeng, Chu ci zhangju shuzheng, 332–33.

73. For a very different reading of this line, see Guo Changbao, Chu ci yu yuanshi zongjiao, 84–86.

74. Hong Xingzu, Chu ci buzhu, 4; Huang Linggeng, Chu ci zhangju shuzheng, 40. On the philological problem of ling 靈 and the verifiability of Wang Yi’s glosses, see Kósa, Gábor, “The Shaman and the Spirits: The Meaning of the Word ‘Ling’ in the ‘Jiu ge’ Poems,” Acta Orientalia Academiae Scientiarum Hungaricae 56.2–4 (2003), 275–94CrossRefGoogle Scholar.

75. Hong Xingzu, Chu ci buzhu, 9; Huang Linggeng, Chu ci zhangju shuzheng, 118; cf. Guo Changbao, Chu ci yu yuanshi zongjiao, 125.

76. Hong Xingzu, Chu ci buzhu, 14–15; Huang Linggeng, Chu ci zhangju shuzheng, 186–96.

77. Hong Xingzu, Chu ci buzhu, 28; Huang Linggeng, Chu ci zhangju shuzheng, 353.

78. This possibility is illustrated in Sukhu’s interpretation of this passage in the context of Han chenwei 讖緯 literature; see The Shaman and the Heresiarch, 16.

79. Hong Xingzu, Chu ci buzhu, 28; Huang Linggeng, Chu ci zhangju shuzheng, 353.

80. Hong Xingzu, Chu ci buzhu, 28; Huang Linggeng, Chu ci zhangju shuzheng, 357.

81. On xing, which I have translated with the appropriately vague term “evocative juxtaposition” following Pauline Yu’s thorough analysis, see Yu, The Reading of Imagery, 57–60, 105–8, 112–14; Schimmelpfennig, “Quest for a Classic,” 151–54; and Zhengguo, Kang 康正果, Fengsao yu yanqing: zhongguo gudian shici de nüxing yanjiu 風騷與豔情:中國古典詩詞的女性研究 (Zhengzhou: Henan renmin, 1988), 1624Google Scholar.

82. Hong Xingzu, Chu ci buzhu, 28, Huang Linggeng, Chu ci zhangju shuzheng, 357–58; on birds, see below.

83. Cf. Sukhu, The Shaman and the Heresiarch, 16; Sterckx, “Religious Practice in Qin and Han,” 427.

84. Hong Xingzu, Chu ci buzhu, 27; Huang Linggeng, Chu ci zhangju shuzheng, 335.

85. Shanhai jing jiaozhu 山海經校注, ed. Yuan Ke 袁珂 (Shanghai: Shanghai guji chubanshe, 1980), 10.381 (“Da huang nan jing” 大荒南經); Hong Xingzu, Chu ci buzhu, 21.

86. E.g. Sima Qian 司馬遷, Shi ji 史記 (Beijing: Zhonghua shuju chuban, 1973), 1.16–17 (“Wudi benji” 五帝本紀) and 26.1257 (“Li shu” 歷書); in Han shu 漢書 “xihe” commonly refers to an historical government office that takes its name from the Xis and Hes, e.g. Ban Gu 班固, Han shu (Beijing: Zhonghua shuju chuban, 1962), 12.351 (“Pingdi ji” 平帝紀), 19.721–22, 731 (“Baiguan gongqing biao, shang” 百官公卿表上) and 30.1734, 1775 (“Yiwen zhi” 藝文志). In the Han shu account of Bing Ji 丙吉 (d. 55 b.c.e.), the xihe officials are even described as “driving/managing” 乘 the four seasons, which is similar to “driving/controlling” 御 the sun (ibid., 74.3139 [“Wei Xiang Bing Ji zhuan” 魏相丙吉傳]). See also Zhu Xi, Chu ci jizhu, 15; Birrell 124–25; and Jiang Liangfu, Chu ci tonggu, 186–91.

87. Cf. Huang Linggeng, Chu ci zhangju shuzheng, 175–78; and Guo Changbao, Chu ci yu yuanshi zongjiao, 97–98.

88. Hong Xingzu, Chu ci buzhu, 75; Huang Linggeng, Chu ci zhangju shuzheng, 929.

89. This passage is not part of the received Huainanzi and probably comes from one of the lost chapters.

90. Sima Qian, Shiji, 117.3009n7 (“Sima Xiangru liezhuan” 司馬相如列傳).

91. Hong Xingzu, Chu ci buzhu, 309; Huang Linggeng, Chu ci zhangju shuzheng, 2620.

92. See Jiang Liangfu, Chu ci tonggu, 66, 199.

93. Hong Xingzu, Chu ci buzhu, 28; Huang Linggeng, Chu ci zhangju shuzheng, 351.

94. Miyano, “O Itsu Soji shoku,” 95–96.

95. E.g., Hong Xingzu, Chu ci buzhu, 57, 59, 69, 70, 72, 74–76, 79; Huang Linggeng, Chu ci zhangju shuzheng, 771, 791, 873, 889, 894, 919–22, 951. The zhangju glosses the “Lord in the Clouds” as Fenglong, who appeared un-allegorically in “Li sao”; see Hong Xingzu, Chu ci buzhu, 59.

96. E.g. Hong Xingzu, Chu ci buzhu, 57, 69, 75; Huang Linggeng, Chu ci zhangju shuzheng, 771, 873, 922.

97. Hong Xingzu, Chu ci buzhu, 76–77; Huang Linggeng, Chu ci zhangju shuzheng, 932. Huang Linggeng calls this “arbitrary explanation” without further comment on this line (932), but later asserts that meiren 美人 (fair one(s)) of this poem refers to women who were offered in human sacrifice as brides of the river god (943–944). Wang Yi, however, asserts that in this case “Fair One” is an epithet for Qu Yuan.

98. Hong Xingzu, Chu ci buzhu, 60; Huang Linggeng, Chu ci zhangju shuzheng, 797.

99. Hong Xingzu, Chu ci buzhu, 59–60; Huang Linggeng, Chu ci zhangju shuzheng, 796.

100. Hong Xingzu, Chu ci buzhu, 61; Huang Linggeng, Chu ci zhangju shuzheng, 811.

101. E.g., Hong Xingzu, Chu ci buzhu, 80, 81; Huang Linggeng, Chu ci zhangju shuzheng, 959, 964, 966, 968.

102. E.g., Hong Xingzu, Chu ci buzhu, 57, 59, 59, 60, 66, 72; Huang Linggeng, Chu ci zhangju shuzheng, 771, 791, 794, 801, 846, 901.

103. Hong Xingzu, Chu ci buzhu, 65, Huang Linggeng, Chu ci zhangju shuzheng, 836.

104. Hong Xingzu, Chu ci buzhu, 55; Huang Linggeng, Chu ci zhangju shuzheng, 742–46.

105. Jade pendants and articles of clothing were offered to gods and ghosts alike in the Warring States, to thank them, possibly by conferral of rank, after they had been appeased by sacrifice of blood and liquor; see Cook, Death in Ancient China, 88–100.

106. The suddenness of this awkward transition in the commentary makes me wonder if Wang Yi’s choice of 遠者 over the variant 遠渚 was not a deliberate effort to insert the recluses and distance Qu Yuan from the dangerously erotic implications of disrobing and lingering on the riverbank (Hong Xingzu, Chu ci buzhu, 68; Huang Linggeng, Chu ci zhangju shuzheng, 869.

107. On the identity and gender of the Mountain Spirit, see Tian Peipei 田培培 and Zhongzhen, Shi 施仲貞, “‘Jiu ge Shangui’ zhong ‘shangui’ zhi xinkao” 《九歌 ⋅山鬼》中《山鬼》之新考, Journal of Sichuan Minzu College 四川民族學院學報 25.5 (2016), 7986Google Scholar. Arthur Waley translates this deity as male (Nine Songs, 53), but Wang Yi’s description of the Mountain Spirit uses language that typically describes a woman’s physical beauty. Furthermore, attribution of the adjective yaotiao 窈窕 to the deity and explicit citation of the Shi jing poem “Guan ju” 關雎, in which yaotiao is undoubtedly feminine, represent the Mountain Spirit as female; see Hong Xingzu, Chu ci buzhu, 79; Huang Linggeng, Chu ci zhangju shuzheng, 951.

108. Following Wang Yi’s reading; cf. Huang Linggeng, Chu ci zhangju shuzheng, 948–49.

109. Hong Xingzu, Chu ci buzhu, 79, 81; Huang Linggeng, Chu ci zhangju shuzheng, 948, 965.

110. Hong Xingzu, Chu ci buzhu, 81; Huang Linggeng, Chu ci zhangju shuzheng, 967. This comment is marked as “someone else’s” reading and is in the rhyming tetrasyllabic verse form of the older annotation style. The older commentary is usually first in Wang Yi’s presentation, but in this case it is presented after the paraphrase. See Kominami, “Wang Yi Chu ci zhangju zai Handai,” 282–83.

111. Hong Xingzu, Chu ci buzhu, 75; Huang Linggeng, Chu ci zhangju shuzheng, 927.

112. On such rituals regarding another deity of the “Nine Songs” and as discerned in archeological evidence, see Harper, “Warring States,” 870.

113. Hong Xingzu, Chu ci buzhu, 76; see Yingjing hexagram ䷍ Dayou 大有, xiangzhuan 象傳, 1.

114. Hong Xingzu, Chu ci buzhu, 159; Huang Linggeng, Chu ci zhangju shuzheng, 1705.

115. Cini 雌蜺 (Female Rainbow) is included among the goddesses in the Chu ci zhangju commentary on “Far Roaming,” where this word also appears (Hong Xingzu, Chu ci buzhu, 173; Huang Linggeng, Chu ci zhangju shuzheng, 1842), but in the commentary on “Nine Compositions” there is no language that identifies this word as a deity or even as divine. The commentary on “Roaming,” however, probably pre-dates Wang Yi and the Eastern Han (see discussion and citations below).

116. Huang Linggeng argues that the lishen in this case is not the Plague Spirit but the ghost of someone killed in battle, which is closer to Wang Yi’s interpretation and better suits the context; hence I have translated it “afflicted spirit”; see Chu ci zhangju shuzheng, 1306–7.

117. Hong Xingzu, Chu ci buzhu, 124–25; Huang Linggeng, Chu ci zhangju shuzheng, 1307, 1313.

118. Hong Xingzu, Chu ci buzhu, 132; Huang Linggeng, Chu ci zhangju shuzheng, 1392.

119. Hong Xingzu, Chu ci buzhu, 129; Huang Linggeng, Chu ci zhangju shuzheng, 1346.

120. On the significance of the Bright Hall in the Han, see Tseng, Picturing Heaven in Early China, chap. 1.

121. Hong Xingzu, Chu ci buzhu, 130; Huang Linggeng, Chu ci zhangju shuzheng, 1368.

122. See Kominami, “Wang Yi Chu ci zhangju zai Handai,” 282–83, and the table in “O Itsu Soji shoku o megute,” 73.

123. Hong Xingzu, Chu ci buzhu, 172; Huang Linggeng, Chu ci zhangju shuzheng, 1830.

124. See Kroll, “On ‘Far Roaming’,” 659. I have translated cini 雌蜺 as “Goddess Rainbow” following the Chu ci zhangju verse commentary.

125. Hong Xingzu, Chu ci buzhu, 172–73; Huang Linggeng, Chu ci zhangju shuzheng, 1837.

126. Hong Xingzu, Chu ci buzhu, 175.

127. For the wide variety of interpretations that have been proposed for Pengxian or Peng and Xian, see Guo’en, You 游國恩, Li sao zuanyi 離騷纂義 (Beijing: Zhonghua shuju, 1982), 121–28Google Scholar and 499–506; and Huang Linggeng, Chu ci zhangju shuzheng, 175–78.

128. E.g., Hong Xingzu, Chu ci buzhu, 25, 42–43, 45 46; Huang Linggeng, Chu ci zhangju shuzheng, 321, 504, 520, 523, 533; and throughout the “Nine Songs.”

129. E.g., Hong Xingzu, Chu ci buzhu, 25, 44; Huang Linggeng, Chu ci zhangju shuzheng, 325, 516.

130. E.g., Hong Xingzu, Chu ci buzhu, 28, 29, 34, Huang Linggeng, Chu ci zhangju shuzheng, 357, 358, 409.

131. Hong Xingzu, Chu ci buzhu, 16; Huang Linggeng, Chu ci zhangju shuzheng, 207.

132. On phoenixes, see Cook, Death in Ancient China, 145–46.

133. So far as I know, phoenixes did not receive sacrifice or devotion in early imperial China, whereas dragons did. On the reduction of the status of goddesses from religion to mythology, see He, Lin 林河, Jiu ge yu Yuan Xiang minsu 九歌與沅湘民俗 (Shanghai: Sanlian shudian, 1990)Google Scholar. Lin argues that all female deities had disappeared or been reduced to the status of legends by Qu Yuan’s time (46); but in the Western Han the funerary banners of Mawangdui, records of worship of the Jade Maidens (yunü 玉女), and the popularity of the cult of the Queen Mother of the West, for example, suggest that the transition was much slower and mostly restricted to the official rites of the imperial court. The apparently low status of goddesses in the Chu ci is as likely to be an effect of the Chu ci zhangju as it is a feature of the original texts.

134. Hong Xingzu, Chu ci buzhu, 55; see also Hawkes, Songs of the South, 19, and Sukhu, “Monkey, Shamans, Emperors, and Poets: the Chu ci and Images of Chu during the Han Dynasty” in Defining Chu. Cf. Goldin, Paul Rakita, The Culture of Sex in Ancient China (Honolulu: University of Hawai‘i Press, 2002)Google Scholar, chap. 1.

135. On the rationalization of politics and the status of wu see Sterckx, “Religious Practice in Qin and Han,” 416–17; Raz, “Chinese Religion from the Han to the Six Dynasties,” 405; Aihe Wang, Cosmology and Political Culture, 168. The transformation of emperorship “into a vehicle for the ethical ideas [of the reformist scholars]” has clear consequences for the practitioners of erotically charged rites; see Fu-shi Lin, “The Image and Status of Shamans,” 446. Whalen Lai reads the Shi ji story of Ximen Bao destroying the wu cult of the River Earl as an example of Han official culture reimagining ancient religious rites in rationalistic terms; see Looking for Mr. Ho Po,” History of Religions 29.4 (1990), 335–50CrossRefGoogle Scholar.

136. Zhu Xi, Chu ci jizhu, 17.

137. Fletcher, Angus, Allegory: The Theory of a Symbolic Mode (Ithaca: Cornell University Press, 1964), 369Google Scholar.