Empowered by Ancestors: Controversy Over the Imperial Temple in Song China (960-1279) 9888528580, 9789888528585

Citation preview

15mm

5mm

146mm

12mm

18.9mm

12mm

146mm

5mm

15mm

15mm

5mm

“Cheung knows the history and culture of China’s Imperial Temple system best and pulls together a decade of research to share his mature reflections. Most modern scholars have avoided this arcane institution; Cheung clarifies its role in Song political culture, its influence in late imperial China, and its legacy in contemporary constructions of cultural memory and legitimacy.” —Hoyt Cleveland Tillman, Arizona State University; coauthor of Cultural Authority and Political Culture in China: Exploring Issues with the Zhongyong and the Daotong during the Song, Jin and Yuan Dynasties

“Professor Cheung helps us wrap our minds around the weight Song Confucian scholars put on reviving ancient rituals. He does this by digging deeply into their positions on the arrangement of the Imperial Ancestral Shrine and placing their contentions in both political and intellectual contexts.” —Patricia Ebrey, University of Washington; author of Confucianism and Family Rituals in Imperial China: A Social History of Writing about Rites

235mm

Empowered by Ancestors: Controversy over the Imperial Temple in Song China (960–1279) examines the enduring tension between cultural authority and political power in imperial China by inquiring into Song ritual debates over the Imperial Temple. During these debates, Song-educated elites utilized various discourses to rectify temple rituals in their own ways. In this process, political interests were less emphasized and even detached from ritual discussions. Meanwhile, Song scholars of particular schools developed various ritual theories that were used to reshape society in later periods. Hence, the Song ritual debates exemplified the great transmission of ancestral ritual norms from the top stratum of imperial court downward to society. In this book, the author attempts to provide a lens through which historians, anthropologists, experts in Chinese Classics, and scholars from other disciplines can explore Chinese ritual in its intellectual, social, and political forms.

EMPOWERED by ANCESTORS

CONTROVERSY OVER THE IMPERIAL TEMPLE IN SONG CHINA (960–1279)

CONTROVERSY OVER THE IMPERIAL TEMPLE IN SONG CHINA (960–1279)

EMPOWERED by ANCESTORS

Cheung Hiu Yu is an assistant professor in the History Department at the Chinese University of Hong Kong. He has published widely on the intellectual and social history of the Middle Periods of China from the seventh through fourteenth centuries. He is the author of Dowry: Pattern and Significance of Daughters’ Property Ownership in Song China (2008, in Chinese) and a myriad of articles published in both English and Chinese scholarly journals.

History / China

Printed and bound in Hong Kong, China

Cheung Hiu Yu

Cover images: Song Taizu, 927–976. Kong Daofu’s (985–1039) sacrificial oration to his primal ancestor, Confucius. Courtesy of the National Palace Museum, Taipei.

5mm

15mm

Empowered by Ancestors

Empowered by Ancestors

Controversy over the Imperial Temple in Song China (960–1279)

Cheung Hiu Yu

Hong Kong University Press The University of Hong Kong Pokfulam Road Hong Kong https://hkupress.hku.hk © 2021 Hong Kong University Press ISBN 978-988-8528-58-5 (Hardback) All rights reserved. No portion of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publisher. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.

10 9 8 7 6 5 4 3 2 1 Printed and bound by Hang Tai Printing Co., Ltd. in Hong Kong, China

To my mother, Wong Pok, and the endless support of my wife, Niki Leung

Contents

List of Illustrations ix List of Figures and Tables x Acknowledgments xi Note to the Reader xiii Song Ancestors and Emperors in Their Temple Names and Personal Names xiv Introduction 1 Why the Song Imperial Temple? 3 The Imperial Temple: A Literature Review 6 Theoretical Consideration, Sources, and Structure 9 Section One: Imperial Temple: Early History and Main Controversies Chapter 1: Pre-Song Interpretations of the Imperial Temple Centralizing the Imperial Temple in Early China Controversies in the Tang Conceptions of the Imperial Temple Concluding Remarks

15 15 26 36

Chapter 2: Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 38 Song Ritual Institutions and Ritual Officials 39 Basic Setting of the Song Imperial Temple and Temple Sacrifices 41 Fraternal Succession and the Discourse of Filial Piety 47 Concluding Remarks 58 Section Two: Song Imperial Temple: Ritual Debates, Factional Politics, and Intellectual Diversity Chapter 3: An Examination of the 1072 Controversy over the Song Primal Ancestor 63 Launch and Development of the 1072 Primal Ancestor Debate 63 Intellectual Interest and Political Stance of Ritual Officials in the 1072 Debate 69 Concluding Remarks 83

viii Contents

Chapter 4: Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate DPATR and the Yuanfeng Scheme of the Imperial Temple The 1079 Zhaomu Debate: Lu Dian and the DPATR The 1079 Zhaomu Debate: Lu Dian Versus He Xunzhi and Zhang Zao Concluding Remarks

85 86 93 97 102

Chapter 5: Imperial Temple in the New Learning 104 An Appraisal of New Learning Ritual Scholarship 104 From Wang Anshi to Wang Zhaoyu: New Learning Conception of the Imperial Temple 110 Chen Xiangdao’s Conception of the Imperial Temple in the Lishu 115 Reconciliation and Codification: Ma Ximeng, Fang Que, and the Zhenghe wuli xinyi 121 Concluding Remarks 129 Section Three: Imperial Temple, Daoxue Scholars, and the Socialization of Temple Rites Chapter 6: Daoxue Conception of the Song Imperial Temple: Zhu Xi and His Disciples Zhu Xi’s Rediscovery of the 1079 Ritual Controversy New Understanding of the Spatial Arrangement of the Imperial Temple: Zhu Xi and Yang Fu Zhu Xi’s Conception of the Zhaomu Sequence Concluding Remarks

133 135 139 149 157

Chapter 7: Socialization of the Discourse of Temple Rituals in the Late Southern Song 159 From the Comprehensive Commentary to Other Southern Song Ritual Commentaries 160 Adoption of Temple Rituals in the Southern Song: An Emphasis on Orthopraxy 167 Concluding Remarks 181 Conclusion 183 Bibliography 191 Index 208

Illustrations

Illustration 1.1: A Qing Illustration of the Tang Imperial Temple 29 Illustration 2.1: Nie Chongyi’s Mingtang Illustration 44 Illustration 5.1: Visualization of the Arrangement of Spirit Tablets in the Di Sacrifice 117 Illustration 5.2: Depiction of Mingtang 119 Illustration 6.1: Zhu Xi’s Temple Scheme for Feudal Lords 141 Illustration 7.1: A Diagram of Seasonal Sacrifices in a Thirteenth-Century Encyclopedia 174

Figures and Tables

Figures Figure 4.1: Diagram of the Configuration of Eight Temples with Different Chambers 92 Figure 4.2: Zhaomu Sequence Suggested by Lu Dian in the 1079 Debate 96 Figure 5.1: Wang Anshi’s Perception of the Imperial Temple 111 Figure 6.1: Zhu Xi’s Spatial Setting of the Imperial Temple 140 Figure 6.2: Zhu Xi’s Depiction of the Imperial Temple in “Lunbenchao miaozhi” 143 Figure 6.3: Zhu Xi’s Visualization of the Imperial Temple’s Main Chamber 150 Figure 6.4: Zhu Xi’s Solution to the Seniority Problem in the Zhaomu Sequence 156

Tables Table 3.1: Officials Involved in the 1072 Primal Ancestor Debate: Ritual Interests and Political Stance 81 Table 5.1: A Survey of Ritual Writings Composed by New Learning Scholars 108 Table 7.1: A Comparison between Seasonal Sacrifices in the Family Precepts of the Lü Family and the Official Ritual Code Zhenghe wuli xinyi 171

Acknowledgments

One afternoon in 2011, while I was working on a paper about the arrangement of imperial ancestors in Chinese court rituals, I suddenly noticed that I was, in fact, working on a book project beyond the scale of a paper. Over the past years, I have devoted myself to the general research fields of Chinese court rituals and traditional ritual learning in China. This book concludes my work over the last decade. It would have never been possible without the tremendous encouragement and support from many teachers, colleagues, and friends. My first and sincere appreciation goes to my doctoral supervisor, Professor Hoyt Cleveland Tillman, who led me to walk through the fascinating intellectual landscape of the Middle Periods of China. With limitless generosity and patience, Professor Tillman guided me through the writing of the early drafts of this book. His guidance has made my research on Chinese court rituals a rewarding and promising journey. I have learned and am still learning to be a good historian from his erudition and high standards of academic writing. During my doctoral study at the Arizona State University, I was fortunate enough to study with a variety of eminent historians and scholars who educated me from different disciplinary angles. Professor Stephen MacKinnon enriched my knowledge of modern Chinese history and Chinese politics with his exceptional critical comments, which offered me new insights in understanding traditional Chinese history. Likewise, Professor Chen Huaiyu’s Buddhist seminar opened a new door for me to conceptualize Chinese intellectual history. Professor James Rush introduced me to Southeast Asian studies and comparative colonialism, which now constitute parts of my keen teaching interests. Professor Stephen Bokenkamp’s reading groups every Monday at his home provided me with excellent opportunities to communicate with other colleagues on the interpretation and translation of ancient Chinese texts. Additionally, I will never forget how much I have learned from Professor Stephen West’s lectures and reading seminars. His profound understanding of literary theories and the very nature of Chinese texts always guides me in reading ritual Classics and their commentaries. My master’s thesis supervisor, Professor Billy So Kee-long, led me to the study of Song history. I am indebted to him both academically and personally. Without his encouragement, I would not have become a historian. I would also like to express

xii Acknowledgments

my gratitude to my undergraduate mentor, Professor Tsang Shui-lung. Professor Tsang’s path-breaking research in Song military history exemplifies the perfect marriage between historical research and theoretical analysis. His death in 2003 is a great loss to the field of Song history. Throughout the writing of this book, I felt myself empowered by his teachings in my undergraduate classes. I also feel thankful to have the opportunity to discuss with and learn from a number of eminent historians, including Professors Deng Xiaonan, Huang Kuanchung, Charles Hartman, Peter Bol, Paul Smith, Ari Levine, Ho Koon-wan, Yin Hui, Chen Xi, Zhao Dongmei, Song Jaeyoon, Ji Xiao-bin, Chu Ming-kin, Liu Guanglin, and the anonymous reviewers of my book manuscripts. Their valuable comments and suggestions pushed me to expand my research scope and thus refine my research questions. Moreover, I would like to thank my colleagues at the Chinese University of Hong Kong for their continuous support in my pursuit of better scholarship. My gratitude goes to Professors Yip Hon-ming, Lai Ming-chiu, Cheung Sui-wei, Ho Pui-yin, Hsiung Ping-chen, Poo Mu-zhou, Leung Yuen-sang, David Faure, Puk Wing-kin, He Xi, and Lam Weng-cheong. I would also like to extend my gratitude to my Arizona State University colleagues including Zhang Dongling, Wu Yang, Chang Wenbo, Jennifer Bussio, Jiang Lijing, Wu Siyuan, and Wen Yu. Their criticisms of the early drafts of this book resulted in a more thorough research than I would have been able to conduct without their assistance. The main body of Chapter 3 first appeared as an article with the title, “Ritual and Politics: An Examination of the 1072 Primal Ancestor Debate in the Northern Song,” in Frontiers of History in China 13.3 (2018): 275–310. With the permission of the journal editors, it is collected in this book after some stylistic revisions. I would like to express my acknowledgment to Frontiers of History in China here. Additionally, I would like to acknowledge with great gratitude the financial support from the Early Career Scheme (ECS, Project No.: 24603017) Funding of the Research Grant Council (RGC) of Hong Kong for supporting me to write this book. Regarding the editing of my book manuscript, I am indebted to Eric Mok, Joan Vicens Sard, and Clara Ho, editors of Hong Kong University Press, as well as my research assistant Cheung Sze-ting, who assisted me in finalizing the book. Finally, I would like to express my deepest gratitude to my wife, Niki Leung, who offered constant encouragement and valuable suggestions. I cannot imagine the completion of this book without her sacrifice and tolerance. Her support always instills in me the courage to pursue higher achievements. If I have produced any valuable research, she shares them with me. Certainly, I am solely responsible for any mistakes in this book.

Note to the Reader

I have adopted the hanyu pinyin system in the romanization of Chinese characters in this monograph. Dates are presented in the Chinese calendar years in the form of year/month/day. Thus, 1072/3/6 stands for the sixth day of the third month of the fifth year of the Xining era under Emperor Shenzong’s reign. For the general periods and eras of Chinese dynasties, I have converted them into Western (Gregorian) calendar years. I have presented the references of ancient Chinese texts in the footnotes in the form of source, x: y. z. “X” refers to the volume of the source in its modern collections. Sometimes volume numbers are omitted for the sake of clarity. “Y” refers to the juan of the source in its traditional editions. “Z” refers to the page number of the source in either its modern or traditional editions. Regarding the names of Chinese and Japanese scholars in footnotes, their surnames precede their given names for the sake of clarity.

Song Ancestors and Emperors in Their Temple Names and Personal Names

Shengzu 聖祖, Zhao Xuanlang 趙玄朗 (????–????, a Daoist deity as the legendary ancestor of the Song imperial house) Xizu 僖祖, Zhao Tiao 趙朓 (828–874) Shunzu 順祖, Zhao Ting 趙珽 (851–928) Yizu 翼祖, Zhao Jing 趙敬 (872–933) Xuanzu 宣祖, Zhao Hongyin 趙洪殷 (899–956) Northern Song dynasty (960–1127) Taizu 太祖, Zhao Kuangyin 趙匡胤 (927–976, r. 960–976) Taizong 太宗, Zhao Guangyi 趙光義 (939–997, r. 976–997) Zhenzong 真宗, Zhao Heng 趙恆 (968–1022, r. 997–1022) Renzhong 仁宗, Zhao Zhen 趙禎 (1010–1063, r. 1022–1063) Yingzong 英宗, Zhao Shu 趙曙 (1032–1067, r. 1063–1067) Shenzong 神宗, Zhao Xu 趙頊 (1048–1085, r. 1067–1085) Zhezong 哲宗, Zhao Xu 趙煦 (1077–1100, r. 1085–1100) Huizong 徽宗, Zhao Ji 趙佶 (1082–1135, r. 1100–1125) Qinzong 欽宗, Zhao Huan 趙桓 (1100–1161, r. 1125–1127) Southern Song dynasty (1127–1279) Gaozong 高宗, Zhao Gou 趙構 (1107–1187, r. 1127–1162) Xiaozong 孝宗, Zhao Shen 趙昚 (1127–1194, r. 1162–1189) Guangzong 光宗, Zhao Dun 趙惇 (1147–1200, r. 1189–1194) Ningzong 寧宗, Zhao Kuo 趙擴 (1168–1224. r. 1194–1224) Lizong 理宗, Zhao Yun 趙昀 (1205–1264, r. 1224–1264) Duzong 度宗, Zhao Qi 趙祺 (1240–1274, r. 1264–1274) Gongdi 恭帝, Zhao Xian 趙顯 (1271–1323, r. 1275–1276) Duanzong 端宗, Zhao Shi 趙昰 (1268–1278, r. 1276–1278)

Introduction

The most obvious trace of perfect governance is seen in the Imperial Temple. 王道之可觀者，莫盛乎宗廟。 —Wang Bi 王弼 (226–249)1

Ancestral worship and related ancestral rituals played a central role in Chinese culture. Historically, ancestral rites and ceremonies in imperial China underwent both social and intellectual developments. Traditional Chinese—including elites— emphasized taking care of the world of ancestral spirits through funeral rites and sacrificial ceremonies.2 On the local level, ancestral rituals promoted by Confucian scholars progressively penetrated village societies through the spread of clan rules, family rituals, and social institutions.3 On the state level, central governments actively participated in the campaign of ritualizing society and eagerly promoted particular ritual norms. Considering the significant role played by ancestral rituals in connecting state and society, the current book explores the making of ancestral ritual norms by focusing on ritual debates in the imperial courts of Song China (960–1279). Generally, it argues that court ritual debates among Song scholarofficials (shidafu 士大夫) empowered them with cultural authority to confront the state and reshape society. Song China witnessed the beginning of a great transformation of ritual norms on both state and local levels. Along with the revival of Confucianism, Song scholar-officials actively participated in debates concerning various ritual affairs. 1. Zhouyi zhushu 周易注疏, Tang Song zhushu shisanjing 唐宋注疏十三經 (hereafter TSZSSSJ), Sibu beiyao edition 四部備要本 (Beijing: Zhonghua shuju, 1998), 1:3.43. 2. Francis Hsu’s early work analyzed the anxiety between the living people and their ancestors experienced in Chinese villages. See L. K. Francis Hsu, Under the Ancestors’ Shadow: Kinship, Personality, and Social Mobility in Village China (New York: Natural History Library, 1967), 131–99. Also see Stephen Bokenkamp, Ancestors and Anxiety: Daoism and the Birth of Rebirth in China (Berkeley: University of California Press, 2007), 60–94. 3. Patricia Ebrey’s detailed annotation of Zhu Xi’s 朱熹 (also spelled Chu Hsi, 1130–1200) Family Rituals and its repercussion throughout late imperial China persuasively demonstrates how various ideas and practices of family rituals—capping, wedding, funeral, and sacrificial offerings—gradually diffused into society through the circulation of Confucianized ritual texts. Patricia Ebrey, Chu Hsi’s Family Rituals: A Twelfth-Century Chinese Manual for the Performance of Cappings, Weddings, Funerals, and Ancestral Rites (Princeton, NJ: Princeton University Press, 1991), 153–77; also see Ebrey, Confucianism and Family Rituals in Imperial China: A Social History of Writing about Rites (Princeton, NJ: Princeton University Press, 1991), 9–13, 220–29.

2

Empowered by Ancestors

Thanks to these debates, a consensus gradually formed within the circles of Song scholar-officials: the court should rectify some key ritual norms to regulate relationships among the government, cultural values, and society. Confucian scholarofficials allied with the court to promote ritual from the top stratum downward. They considered imperial rituals the highest standard of ritual performance and understood the rectification of these rituals (zhengli 正禮) to be their responsibility. From the perspective of scholar-officials, imperial ancestral rituals served not only as a “pretense for cultural agendas”4 but also as a way of self-identification. In this light, court discussions and debates over imperial ancestral rituals transcended the private sector of imperial families and constituted one of the most heated issues in Song state policies.5 Specifically, I focus on the ritual discussions about a significant ritual architecture in Song China, that is, the complex of the Imperial Temple (taimiao 太廟). Spatially, the Imperial Temple emblematizes the succession of the ruling house through a display of the royal ancestral line.6 On the one hand, the Imperial Temple nearby the palace displays the authority of the ruling house by its grandiose appearance. In the orthodox Confucian setting of a royal capital, the Imperial Temple, the royal palace, and the State Altar of the Grain and Soil (shejitan 社稷壇) are arranged according to a fixed order, which is first documented in the Kaogong ji 考工記 (literally, Records of Artificers).7 The order reads: “the Altar is located on the right and the Imperial Temple on the left, the administrative palace is located in the front and the market place in the rear” 左祖右社，面朝後市.8 This placement identifies the Imperial Temple as one of the four fundamental architectural structures of a royal capital. At the same time, the Imperial Temple embodies the Confucian virtue of filial piety at the highest level of political realm: the emperor himself shows due respect to his ancestors through solemn sacrifices and ceremonies that are regularly 4. I have borrowed this phrase from Kevin E. Brashier, Ancestral Memory in Early China (Cambridge, MA: Harvard University Press, 2011), 348. 5. In her study of Qing 清 (1636–1912) court rituals, Evelyn Rawski distinguishes between private and public imperial rituals based on Qing official archives. Rawski, The Last Emperors: A Social History of Qing Imperial Institutions (Berkeley: University of California Press, 1998), 264–68, 277–85. However, most “private” ancestral rites defined by Rawski still had an empathetic function that aimed to arouse emotions among members of particular groups. Non-Confucian funeral rites adopted by the Qing rulers involved physiological stimuli that contributed to a shared experience of the symbolic power of ritual. In this light, imperial ancestral rituals are at least “public” to their spectators, as they cast empathetic effects on the spectators’ minds. 6. As put forward by Michael Loewe, the Imperial Temple demonstrated that “the imperial house was of a more permanent duration.” Michael Loewe, Problems of Han Administration: Ancestral Rites, Weights and Measures, and the Means of Protest (Leiden: Brill, 2016), 9. 7. The Kaogong ji is the earliest surviving record of Chinese architectural and handicraft industries. Some scholars have identified it with an official record that was composed by the Qi 齊 state during the Warring States period. In general, it conveys an imagination of the ideal architectural settings of the Zhou dynasty. During the Western Han dynasty, some scholars attached the Kaogong ji to the Rituals of Zhou and made it the latter’s last section, the Dongguan 冬官 (Winter Bureau). For a textual history of the Kaogong ji, see Feng Jiren, Chinese Architecture and Metaphor: Song Culture in the Yingzhao Fashi Building Manual (Hong Kong: Hong Kong University Press, 2012), 26–27. Also, Wen Renjun 聞人軍, Kaogong ji yizhu 考工記譯註 (Shanghai: Shanghai guji chubanshe, 1993), 138–53. 8. Zhouli zhushu 周禮注疏, TSZSSSJ, 2:41.411.

Introduction 3

performed in the temple. Additionally, the Imperial Temple serves as a symbolic microcosm that connects imperial clansmen. In view of the above, the Imperial Temple serves as an “intimacy-oriented political model” of Confucianism that connects ritual with politics between private and public spheres.9 In summary, the Imperial Temple and its relevant rituals crystallized the tension between cultural authority and political ends. In the Song context, scholar-officials utilized various discourses to conceptualize the Imperial Temple. These discourses included the political contributions and merits of imperial ancestors, the Confucian idea of filial piety, and the revival of ancient rituals. How did these discourses flourish in court ritual debates? To what extent were they measured and valued by Chinese scholar-officials? How did they shape the “rectification of rituals” within the circles of Confucian scholars on the social level? Answers to these questions constitute the main body of this book.

Why the Song Imperial Temple? Since the Japanese Sinologist Naitō Konan 内藤湖南 (1866–1934) proposed his famous “Naitō hypothesis,” historians have generally considered political interest as the core value of Song “factions” (dang 黨).10 As is well understood, “political factions” (pengdang 朋黨) played a central role in Song history and have attracted scholarly attention from a number of historians. Fan Zhongyan’s 范仲淹’s (989– 1052) Qingli Reforms (Qingli xinzheng 慶曆新政) and Wang Anshi’s 王安石 (1021–1086) New Policies (xinfa 新法) have generally identified Song factionalism with a direct political confrontation between reformists and anti-reformists, especially in the New Policies of the late Northern Song.11 Nevertheless, Song ritual controversy over the Imperial Temple reflects another dimension of Song factionalism that was more defined by intellectual than political 9. I borrow this phrase from Jiang Tao 蔣韜, “Intimate Authority: The Rule of Ritual in Classical Confucian Political Discourse,” in Confucian Cultures of Authority, ed. Peter D. Hershock and Roger T. Ames (New York: State University of New York Press, 2006), 30. 10. In the original version of the “Naitō hypothesis,” Naitō Konan argued for a shift in the nature of scholar-officials’ “faction” from martial relations to political interests in the Tang-Song transition. See Naitō, “Gaikakuteki To-So jidai kan” 概括的唐宋時代觀. Rekishi to chiri 歴史と地理 9:5 (1922): 1–12, especially 7–8. For more details of the Naitō hypothesis and its evolution in scholarship, see Hisayuki Miyakawa 宮川尚志, “An Outline of the Naitō Hypothesis and Its Effects on Japanese Studies of China,” The Far Eastern Quarterly 14.4 (1955): 533–52, especially 535–38. Zhang Guangda 張廣達, “Neiteng hunan de Tang Song biangeshuo jiqi yingxiang” 內藤湖南的唐宋變革説及其影響, Tang Yanjiu 唐研究 11 (Beijing: Beijing daxue chubanshe, 2005): 5–71. 11. For a thorough study of Northern Song factionalism, see Ari Levine, Divided by a Common Language: Factional Conflict in Late Northern Song China (Honolulu: University of Hawaii Press, 2008), 44–71. According to Levine, Northern Song politicians tended to conceptualize factionalism with polarized vocabularies for the purpose of persuading the emperor to support their interest groups and to expel their adversaries. Their factional rhetoric reflected their political interests on a conceptual level. In practice, Song factionalism was rather volatile—always changing with the times and the external political environment. Some historians have challenged Levine’s methodological choices, especially his focus on the use of terminology for factions. For example, see Hilde de Weerdt’s “Review of Divided by a Common Language: Factional Conflict in Late Northern Song China, by Ari Levine.” Journal of Asian Studies 69.2 (2010): 556–58.

4

Empowered by Ancestors

factors. This study argues that the ritual interests of Song scholar-officials were more associated with their scholarly backgrounds rather than with their political stances or affiliations. Song ritual discussions and debates on the Imperial Temple involved scholar-officials from various departments in the central bureaucracy, including chief councilors, the emperors’ private secretaries, academicians, and ritual officials from various ritual bureaus (lichen 禮臣). It is worth noting that Song ritual officials did not hold lifelong appointments. As such, they should not be simply treated as a monolithic group of professionals. Scholar-officials who had never served in ritual bureaus may have been as knowledgeable about ritual issues as ritual officials. Generally, ritual debates between scholar-officials reflected how these officials formulated and promoted particular intellectual interests. Through a thorough analysis of these interests, this study reveals the intellectual confrontation between Song scholar-officials behind the veil of political factions. Hence, it offers historians a new perspective to understand Song factionalism. As an intellectual discourse, Song ritual debates on the Imperial Temple is also associated with the identity of scholar-officials. Peter Bol’s significant work on the intellectual transitions in Tang (618–907) and Song China explores the formation of scholar-officials’ identity and the shift of a view of literati culture from literary accomplishment to ethical values.12 In the Song ritual discourses of the Imperial Temple, scholar-officials rendered a separate intellectual identity that transcended the boundaries of not only factional politics but also the strictly defined “schools” (xuepai 學派) of Song scholarship. In terms of intellectual identity, Song scholarofficials are more eclectic than historians have previously thought, if ritual interest is taken into consideration. Song ritual discourse of the Imperial Temple reveals some discrepancies between conservative and reformist ideas among scholar-officials. Nonetheless, that discrepancy does not necessarily concur with the conventional understanding of Song scholar-officials’ political and intellectual identities as represented in the Yuan-compiled Song shi 宋史 (official dynastic history of Song) and the Qing-compiled Song Yuan xuean 宋元學案 (Case Studies of the Learning of Song and Yuan Scholars). In this light, this study supplements and enriches Bol’s and other scholars’ research on the construction of Song literati identity as a kind of self-identification process along with the Daoxue 道學 movement.13 Ritual debates over the Imperial Temple also inspired some Song Confucian scholars to promote imperial ritual norms on the social level. Anthropologists have significantly explored and enriched the research field of Chinese ancestral rituals. Historians have also made substantial contributions from various perspectives.14 12. Peter Bol, This Culture of Ours: Intellectual Transitions in T’ang and Sung China (Stanford, CA: Stanford University Press, 1992). 13. Peter Bol, “Neo-Confucianism and Local Society, 12th–16th Century: A Case Study,” in The Song-YuanMing Transition in Chinese History, ed. Paul Jakov Smith and Richard von Glahn (Cambridge, MA: Harvard University Asia Center, 2003), 241–83. Also see Chen Wenyi 陳雯怡, “Networks, Communities, and Identities: On the Discursive Practices of Yuan Literati” (PhD diss., Harvard University, 2007), chap. 4. 14. Stephan Feuchtwang and Arthur Wolf, ed., Religion and Ritual in Chinese Society (Stanford, CA: Stanford

Introduction 5

Nonetheless, the intellectual origin of the pivotal role played by ancestral rituals in shaping kinship organizations of late imperial China still remains obscure to this date. Recent studies scarcely discuss the “medieval” origin of these ritual norms, especially compared with the rich literature on the formation and development of different kinship rituals in late imperial China. The interest in tracing that origin has driven me to venture into the field of imperial ancestral rituals. Through a careful examination of Song ritual texts, I argue that intensive court debates over the Imperial Temple during the Song dynasty had codified some ritual norms for the ancestral rituals of later periods to follow. In this light, an analysis of the ritual order of royal ancestors in the Imperial Temple could shed new light on how some ritual norms were rectified on the state level to create a Confucian model of ritual propriety for adoption on the social level, such as the way of ordering ancestors in genealogical writings. Therefore, a study of the Imperial Temple and related ritual controversies contributes to a better understanding of the Song conception of ritual “orthodoxy” and “orthopraxy.”15 Lastly, a study on the Song Imperial Temple provides us an opportunity to rethink state-elite and state-society relationships in Chinese history. Robert Hartwell’s classic argument on the longue durée transformation in the Middle Period of China emphasizes the “local turn” in the Tang and Song periods, which indicates a shift in educated elites’ focus from the central state to local society.16 This book reexamines and enriches the “local-turn” approach from the perspective of ritual studies. Since the Song dynasty, scholar-officials and ritualists (lixue jia 禮學家) had served as collaborators of emperors in formulating court rituals. Reciprocally, as Confucian scholars, these educated elites (shi 士) possessed adequate cultural capital that enabled them to construct ritual norms based on their own conceptions

University Press, 1974), 106–7; Joseph McDermott, The Making of a New Rural Order in South China (Cambridge: Cambridge University Press, 2013); Patricia Ebrey, Confucianism and Family Rituals in Imperial China; David Faure, Emperor and Ancestor: State and Lineage in South China (Stanford, CA: Stanford University Press, 2007). 15. For the concept of orthopraxy, see Catherine Bell, Ritual: Perspectives and Dimensions (New York: Oxford University Press, 1997), 191–97. Bell differentiates orthodoxic and orthopraxic ritual traditions and links rituals in the orthopraxic tradition with specific religious activities that sustain a holistic cultural heritage. In contrast to Bell, anthropologists such as James Watson emphasize the performative aspect of ritual orthopraxy and argue for more pluralistic and discursive understandings of ritual practices in nonreligious contexts. See James Watson, “Anthropological Analyses of Chinese Religion,” China Quarterly 66 (June 1976): 355–64, and Donald Sutton, “Ritual, Cultural Standardization, and Orthopraxy in China: Reconsidering James L. Watson’s Ideas,” Modern China 33 (January 2007): 3–21. 16. Robert Hartwell, “Demographic, Political and Social Transformation of China, 750–1550,” Harvard Journal of Asiatic Studies 42.2 (1982): 365–442. Other major contributors to the “local-turn” approach include Peter Bol, Robert Hymes, Paul Smith, and Hugh Clark, whose research mostly covers crucial areas of the southeastern coastal regions in China. Peter Bol, “The Rise of Local History: History, Geography, and Culture in Southern Song and Yuan Wuzhou,” Harvard Journal of Asiatic Studies 61.1 (2001): 37–76; Robert Hymes, Statesmen and Gentlemen: The Elite of Fu-Chou, Chiang-Hsi, in Northern and Southern Song (Cambridge: Cambridge University Press, 1986); Paul Smith, Taxing Heaven’s Storehouse: Horses, Bureaucrats, and the Destruction of the Sichuan Tea Industry, 1074–1224 (Cambridge, MA: Harvard University Press, 1991); Hugh Clark, Community, Trade and Networks (Cambridge: Cambridge University Press, 1991).

6

Empowered by Ancestors

of cardinal Confucian values.17 This symbiotic relationship between educated elites and rulers ensured the continuity and legitimacy of the Chinese monarchy for hundreds of years. Concerning ritual discourse on the Song Imperial Temple, ritual debates in the Northern Song consolidated a set of standardized codes that imbued temple rituals with Confucian values, especially filial piety. These standardized codes include oversight of regular and irregular temple sacrifices, the arrangement of ancestral chambers in the temple, sacrificial offerings, and ritual utensils. In the eleventh century, scholar-officials employed court ritual debates to assert their autonomy in modifying and rectifying these standardized codes. However, given the rise of monarchial power in the late Northern Song, monarchical authority gradually dominated ritual discussions on temple rituals. Correspondingly, educated elites of Southern Song conceptualized temple rituals within a socio-intellectual framework, under which they textually modified and codified the imperial rituals in genealogical records to symbolize their social prestige.18 After all, knowledge about temple rituals offered educated elites not only the cultural authority to confront monarchical power but also a means to empower themselves regarding their own pedigree.

The Imperial Temple: A Literature Review Since the 1970s, political and social historians have devoted considerable attention to how ancestral rituals have been institutionalized and politicalized according to Confucian doctrines in the Middle Period of China (seventh to thirteenth centuries). In recent decades, scholars have published extensive research regarding imperial ancestral rituals, especially on how these rituals were connected to various political implications such as legitimacy and monarchical authority. Japanese and Chinese historians have approached imperial ancestral rituals based on a binary conception of Chinese monarchs as both the Son of Heaven (tianzi 天子) and the emperor. The works of Sadao Nishijima 西嶋定生 and his student Kaneko Shūichi 金子修一 demarcated the political and ritual identities of Chinese emperors in a public-versus-private conceptual framework.19 Throughout the 1980s and the 1990s, Japanese historians gradually shifted their focus from a general overview 17. For a classic presentation of cultural capital theory, see Pierre Bourdieu, “The Forms of Capital,” in Handbook of Theory and Research for the Sociology of Education, ed. John Richardson (New York, Greenwood, 1986), 241–58. 18. To borrow terms from Robert Hymes, temple rituals underwent a shift from the top stratum of “court-oriented” authority downward to local “shi-oriented” authority in the Song period. Robert Hymes, “Sung Society and Social Change,” in The Cambridge History of China, Vol. 5, Part Two: Sung China, 907–1279, ed. John W. Chaffee and Denis Twitchett (Cambridge: Cambridge University Press, 2015), 621–60, especially 631–32. 19. Sadao Nishijima, “Kōtei shihai no shutsugen” 皇帝支配の出現, in Chūgoku kodai kokka to higashi ajia sekai 中国古代国家と東アジア世界 (Tokyo: University of Tokyo Press, 1983), 370–93; Kaneko Shūichi, Chūgoku kodai kōtei saishi no kenkyū 中国古代皇帝祭祀の研究 (Tokyo: Iwanami Shoten, 2006), 1–28, 431–52; Ogata Isamu 尾形勇, Zhongguo gudai de jia yu guojia 中國古代的家與國家 (Beijing: Zhonghua shuju, 2010), 205–31.

Introduction 7

of imperial ancestral rituals to concern with specific rites and ceremonies.20 The endeavors of Japanese scholars in exploring ritual details have been echoed by ritual historians in China since the 1990s. Ritual historians have examined the institutional changes of imperial rites and ceremonies.21 Ritual historians like Wu Liyu 吳麗娛, Lei Wen 雷聞, Li Hengmei 李衡眉, Chen Shuguo 陳戌國, Guo Shanbing 郭善兵, Zhang Wenchang 張文昌, Zhu Yi 朱溢, and many others have greatly enhanced our understanding of dynastic ritual codes, ritual institutions, and the Confucianization of state rituals in the Middle Period of China’s history.22 In contrast to Japanese and Chinese scholarship, ritual studies in Western languages focus more on how imperial ancestral rituals were conceptualized in their times. In his pioneering work on Tang imperial rituals, Howard Wechsler revealed how ancestral ceremonies were utilized as effective tools to sustain legitimacy by retaining the dynasty’s mandate from Heaven through a twofold worship of Heaven and ancestors.23 With its inspiring theoretical exploration, Wechsler’s monograph remains one of the most important references in understanding the political implications of Chinese imperial rituals. Among other relevant publications in the 1980s, only David McMullen’s study on Tang imperial rituals and Antonino Forte’s research on Tang ancestral buildings can rival Wechsler’s work in depth and scope.24 20. To list a few, Tozaki Tetsuhiko 戶崎哲彥, “Tōdai niokeru teikyū ronsōto sono igi” 唐代における禘祫論爭 とその意義, Tōhōgaku 東方學 80 (1990): 82–96; Tozaki Tetsuhiko, “Tōdai niokeru taibyō seido no bensen” 唐代における太廟制度の変遷, Hikone ronsō 彥根論叢 262–263 (1989): 371–90; Egawa Shikibu 江川式部, “Teigen nenkan no taibyōsōgi to tōdaikōki no reisei kaikaku” 貞元年間の太廟奏議と唐代後期の禮制改 革, Chūgoku shigaku 20 中國史学 (2010): 153–75; Kaneko Shūichi, Chūgoku kodai kōtei saishi no kenkyū, 141–430; Kojima Tsuyoshi 小島毅, “Kō shi seido no hensen” 郊祀制度の變遷, Tōyō bunka kenkyūsho kiyō 108 (1989): 123–219; Yamauchi Kōichi 山內弘一, “Hokusō jidai no kō shi” 北宋時代の郊祀, Shigaku zasshi 史學雜誌 92:1 (1985): 40–66; Yamauchi Kōichi, “Hokusō jidai no taibyō” 北宋時代の太廟, Jochi shigaku 上智 史学 35 (1990): 91–119; Nishioka Ichisuke 西岡市祐, “Shaku Tougyo: Daitō kaigenrei senshin u taibyōrei no senshinbutsu sonoichi” 《大唐開元禮》荐新於太廟禮の荐新物その一, Bulletin of the Sinological Society of Kokugakuin 国学院中国学会報 38 (1992): 74–90. 21. Gao Mingshi 高明士, “Lifa yiyi xiade zongmiao: yi zhongguo zhonggu weizhu” 禮法意義下的宗廟—以中國 中古為主, in Dongya chuantong jiali, jiaoyu yu guofa: jiazu, jiali yu jiaoyu 東亞傳統家禮、教育與國法：家 族、家禮與教育 (Shanghai: Huadong shifan daxue chubanshe, 2008), 23–86; Gan Huaizhen 甘懷真, Tangdai jiamiao lizhi yanjiu 唐代家廟禮制硏究 (Taibei: Taiwan shangwu yinshuguan, 1991); Gan, Huangquan liyi yu jingdianquanshi: zhongguo gudai zhengzhishi yanjiu 皇權、禮儀與經典詮釋：中國古代政治史研究 (Taibei: Ximalaya yanjiufazhan jijinhui, 2003); Huang Jinxing 黃進興, Youru shengyu : quanli, xinyang yu zhengdangxing 優入聖域：權力、信仰與正當性 (Taibei: Yunchen wenhuashiye gufenyouxian gongsi, 1994); Kang Le 康樂, Congxijiao dao nanjiao: guojia jidian yu beiwei zhengzhi 從西郊到南郊: 國家祭典與北魏政治 (Taibei: Daohe chubanshe, 1995). 22. Wu Liyu, Zhongji zhi dian: zhonggu sangzang zhidu yanjiu 終極之典：中古喪葬制度研究 (Beijing: Zhonghua shuju, 2012); Lei Wen, Jiaomiao zhiwai: Suitang guojia jisi yu zongjiao 郊廟之外：隋唐國家祭祀與宗教 (Beijing: Sanlian shudian, 2009), 72–100; Li Hengmei, Zhaomu zhidu yanjiu 昭穆制度研究 (Jinan: Qilu shushe, 1996); Chen Shuguo, Zhongguo lizhishi: suitang wudai juan 中國禮制史：隋唐五代卷 (Changsha: Hunan jiaoyu chubanshe, 1998); Guo Shanbing, Zhongguo gudai diwang zongmiao lizhi yanjiu 中國古代帝 王宗廟禮制研究 (Beijing: Renmin chubanshe, 2007); Zhang Wenchang, Zhili yijiao tianxia: Tang Song lishu yu guojia shehui 制禮以教天下：唐宋禮書與國家社會 (Taibei: Guoli Taiwan daxue chuban zhongxin, 2012); Zhu Yi, Shibangguo zhi shenzhi: tangzhibeisong jili bianqian yanjiu 事邦國之神祇：唐至北宋吉禮變遷研究 (Shanghai: Shanghai guji chubanshe, 2014). 23. Howard Wechsler, Offerings of Jade and Silk: Ritual and Symbol in the Legitimization of the T’ang Dynasty (New Haven, CT: Yale University Press, 1985), 1–106. 24. David McMullen, State and Scholars in T’ang China (Cambridge: Cambridge University Press, 1988),

8

Empowered by Ancestors

Compared to studies on Tang imperial rituals, studies on Song imperial rituals only proliferated after the 1990s. Patricia Ebrey’s early work examines the Song tradition of adopting portrait scriptures in imperial ancestral rituals.25 Song Jaeyoon’s recent monograph provides hitherto the most comprehensive description of some significant Song commentaries on the Confucian ritual Classics within the framework of state policy, including some key passages about imperial ancestral rituals.26 Mihwa Choi’s book studies Song death and sacrificial rituals performed in society and at the imperial court, with a special focus on their political messages.27 Although much scholarly attention has been devoted to imperial ancestral rituals, debates and discussions over these rituals have received scant attention from historians.28 Christian Meyer’s work on the eleventh-century court ritual debates over imperial rituals is an exception, as it focuses on the court debates and discussions through which imperial rituals were codified and instituted.29 Specifically, Meyer attempts to link his study of Song ritual debates to factional politics and the emergence of neo-Confucian philosophy during the same period. Meyer has painstakingly reconstructed neglected aspects of Song ritual history, especially some debates on Song state sacrifices and the ceremonial music used in these sacrifices. Despite his efforts it is difficult to say that he has successfully confirmed his observation on the relationships between ritual, politics, and intellectual campaigns. The lack of a persuasive explanation of these relationships should be attributed to the fact that ritual debates were not solely dominated by political factors during the Song dynasty.30 In ritual debates, intellectual factors were equivalent to or even more important than political factors. However, a thorough analysis of Song ritual debates in relation to their intellectual context is still absent in related fields. This

especially 113–58; McMullen, “Bureaucrats and Cosmology: The Ritual Code of T’ang China,” in Rituals of Royalty: Power and Ceremonial in Traditional Societies, ed. David Cannadine and Simon Price (Cambridge: Cambridge University Press, 1987), 181–236. Antonino Forte, Mingtang and Buddhist Utopias in the History of the Astronomical Clock: The Tower, Statue and Armillary Sphere Constructed by Empress Wu (Rome: Instituto Italiano per il Medio ed Estremo Oriente, 1988). 25. Patricia Ebrey, “Portrait Sculptures in Imperial Ancestral Rites in Song China,” T’oung Pao 83 (1997): 42–92. 26. Song Jaeyoon, Traces of Grand Peace: Classics and State Activism in Imperial China (Cambridge, MA: Harvard University Asian Center, 2015). 27. Choi Mihwa, Death Rituals and Politics in Northern Song China (New York: Oxford University Press, 2017). 28. This is even more obvious in the study of Song ritual history. Wechsler and McMullen both focus on Tang state rituals, as well as Gao Mingshi, Gan Huaizhen, and Kaneko Shūichi. Joseph McDermott’s edited volume concerning Chinese state rituals skips the Song period, regardless of the rich ritual texts that had been produced from the tenth to the thirteenth centuries. Joseph McDermott, ed., State and Court Ritual in China (Cambridge: Cambridge University Press, 1999). Also see Zhu Yi’s comments on Chinese and Japanese scholarship for a lack of attention to Song imperial rituals. Zhu, Shibangguo zhi shenzhi, 37. 29. Christian Meyer, Ritendiskussionen am Hof der nördlichen Song-Dynastie (1034–1093): zwischen Ritengelehrsamkeit, Machtkampf und intellektuellen Bewegungen (Sankt Augustin: Institut Monumenta Serica, 2008). Meyer later summarized the main ideas of his monograph in an English essay. Christian Meyer, “Negotiating Rites in Imperial China: The Case of Northern Song Court Ritual Debates from 1032 to 1093,” in Negotiating Rites, ed. Ute Hüsken and Frank Neubert (Oxford: Oxford University Press, 2011), 99–115. 30. Hilde de Weerdt has also pointed out this. See her review of Meyer’s book on Bulletin of the School of Oriental and African Studies 72.1 (2009): 206.

Introduction 9

study of Song debates over the Imperial Temple fills the void and thus contributes to a better understanding of the connotative meaning of Chinese rituals.

Theoretical Consideration, Sources, and Structure It is commonly said that Chinese emperors were sanctified and empowered by the spiritual power of their ancestors through the appropriate arrangement of Imperial Temples and temple rituals. The Geertzian reading of ritual acts as a manifestation of power within a theatrical state has some explanatory value in explicating temple rituals from a performative perspective.31 Nonetheless, James Laidlaw has argued that the Geertzian account of ritual in the Chinese context tends to overlook the complicated intellectual actions that were involved in the making of court rituals.32 In the case of the Chinese Imperial Temple, related rituals were performed within a conceptual framework of particular cultural references and agendas. Under most circumstances, neither these rituals nor their symbolic meaning matters. What matters are the connections between rituals and various cultural agendas. Anthropological studies of Chinese rituals focus more on the cultural agendas of village ancestral rituals. In an early study of Chinese village rituals, Stephan Feuchtwang and Arthur Wolf have claimed that the targets of traditional Chinese ancestral rituals could be aptly categorized into three different kinds of spiritual beings: ghosts, gods, and ancestors.33 The separation of ancestors from ghosts and gods to a large degree reconciled the tension between this-worldliness and the anxiety surrounding the afterlife, as well as contributed to a sense of familial solidarity among lineage members. Generally, the Imperial Temple served the same purpose as village ancestral rituals in providing a bridge between the living and their ancestors. Thus, some social historians would hastily assume that there was an intrinsic interplay of Imperial Temple rituals and village ancestral practices. However, it is necessary to bear in mind the danger of overestimating the communication between court ritual norms and diverse ritual traditions on the village level. In fact, rural traditions of Chinese ancestral rituals mostly evolved from relevant practices of late imperial China, which differed significantly from their earlier counterparts before the sixteenth century.34

31. Clifford Geertz, Negara: The Theatrical State in Nineteenth-Century Bali (Princeton, NJ: Princeton University Press, 1980), 98–136. 32. James Laidlaw, “On Theatre and Theory: Reflections on Ritual in Imperial Chinese Politics,” in State and Court Ritual in China, ed. Joseph P. McDermott (Cambridge: Cambridge University Press, 1999), 399–405. 33. Feuchtwang and Wolf, ed., Religion and Ritual in Chinese Society, 106–7. 34. For two exemplars of anthropological studies of Chinese village rituals after the sixteen century, see David Johnson’s study on the marginalized “entertainers” (yuehu 樂戶) in Shanxi 山西, Spectacle and Sacrifice: The Ritual Foundations of Village Life in North China (Cambridge, MA: Harvard University Press, 2009) and Liu Yonghua’s study on the “ritual experts” (lisheng 禮生) in Fujian 福建, Confucian Rituals and Chinese Villagers: Ritual Change and Social Transformation in a Southeastern Chinese Community, 1368–1949 (Leiden: Brill, 2013).

10

Empowered by Ancestors

Considering the aforementioned difficulties in studying Chinese imperial rituals, I argue that only if we see the formation of court ritual traditions as a dynamic process of intellectual endeavor in its historical context, can we understand it comprehensively. Beneath the apparently self-contained structure of ritual, researchers would confront the deep consciousness of those who set, performed, and manipulated court ritual for their own intellectual and political ends. Therefore, by focusing on how Song scholar-officials posited the Imperial Temple and related rituals, I aim to explore their mentality as ritual “manipulators” and their intellectual endeavor to transmit ritual ideas to a wider audience outside the imperial court. Methodologically, I embrace a contextual reading of different ritual texts to reveal their intra- and inter-relations. Through a contextual analysis of these texts, I attempt to approach the decision-making moment of the authors who produced them. It does not mean that I intend to speak on behalf of Song scholar-officials in explicating their ritual texts. Instead, my study aims at interpreting Song ritual texts from their contemporary perspective and minimizing the impact of our modern interpretations. Borrowing hermeneutic terms, I try to let the voice of the past horizon reveal itself in a contextual space that is less influenced by modernity. Furthermore, my study emphasizes the profound presentation and revisions of details regarding the Imperial Temple in Song ritual writings. Historians usually find liturgical details in dynastic ritual codes and commentaries on ritual Classics boring and insignificant. However, these details and commentaries were significant to Song scholar-officials. New-historicism argues that the “slippages, cracks, fault lines, and surprising absences in the monumental structures” of history deserves more attention.35 The “surprising absences” of liturgical details in the English studies of Chinese ritual history deserves some reflection. By focusing on ritual details, this book challenges one of the basic assumptions of Chinese ritual history: historians can portray a panorama of traditional rituals through an overview of some eye-catching elements, such as spectacular state sacrifices and some general policies of imperial rituals. This assumption implies a prescribed order that values dynastic ritual codes more than the liturgical details upon which these codes were established. In practice, liturgical details were much more important than official ritual codes. An investigation of these details in various ritual texts helps to fill in a missing link, especially in English studies of Chinese rituals. Concerning sources, most Chinese, Japanese, and Western works on Song ritual history rely on traditional historical sources, including the Yuan-compiled dynastic history of Song, the Xu zizhi tongjian changbian 續資治通鑑長編 (A Sequel to the Comprehensive Mirror for Aids in Governance in Detailed Version), and the sections of ritual affairs in the collections of Song official archives in the Song huiyao jigao 宋會要輯稿 (Collected Manuscripts on the Various Aspects of 35. Catherine Gallagher and Stephen Greenblatt, Practicing New Historicism (Chicago: Chicago University Press, 2000), 17.

Introduction 11

the Song Dynasty). My study is no exception. It relies primarily on the latter two, especially the categories of imperial rituals in the Song huiyao jigao. Additionally, my study uses the extant Song official ritual codes, such as the Zhenghe wuli xinyi 政和五禮新儀 (New Forms for the Five Categories of Rites of the Zhenghe Era) and the Zhongxing lishu 中興禮書 (Ritual Manual of the Revived Song), because they have codified some official regulations of Song temple rituals. More importantly, my research introduces Song ritual commentaries and annotations to the study of Song ritual history. None of the recent Western studies of Song rituals, so far as I know, has systematically used the rich repository of Song private commentaries on ritual Classics.36 These commentaries provide abundant sources for my research on the formation of Song ritual discourse—to list some of them: Nie Chongyi’s 聶崇義 (d. 962) Sanlitu jizhu 三禮圖集註 (Collected Commentary on the Illustrations of the Three Ritual Classics), Chen Xiangdao’s 陳祥道 (1053–1093) Lishu 禮書 (Ritual Manual), Wang Zhaoyu’s 王昭禹 (fl. 1080) Zhouli xiangjie 周禮 詳解 (Detailed Explanation of the Rituals of Zhou), Wang Yuzhi’s 王與之 (fl. 1242) Zhouli dingyi 周禮訂義 (Revised Explanations of the Rituals of Zhou), and Wei Shi’s 衛湜 (fl. 1205–1224) Liji jishuo 禮記集說 (Collective Commentary on the Book of Rites). Having been conventionally conceived as repetitive, pompous, and vacuous records of ritual details, these ritual commentaries serve as one of the key sources of my book. Indeed, a close reading of these commentaries leads to a thorough understanding of what the Song ritualists were thinking about while they were penning these words. In this light, commentaries on ritual Classics are not arcane materials of little significance. Rather, they are a fascinating manifestation of intellectual curiosity in a peculiar form—a form that has been well adopted and accepted by traditional Chinese scholars for thousands of years. To better present a panorama of the Song Imperial Temple controversies, I structure my research both chronologically and thematically. I have divided the whole story about the Imperial Temple into three sections. Section One includes Chapters 1 and 2. Chapter 1 addresses pre-Song interpretations of the Imperial Temple settings and the arrangement of the ritual order of ancestors from the thirteenth century BC to the end of Tang dynasty. This chapter classifies two important interpretations of the ritual status of imperial ancestors since early imperial China: one primarily emphasized ancestors’ political merits; the other emphasized the factor of seniority and hence the Confucian value of filial piety. Chapter 2 briefly introduces Song ritual institutions and ritual officials, as well as early Song ritual controversies over fraternal succession and the discourse of filial piety. Through its two chapters, Section One lays the necessary foundation for the ensuing analysis of Song temple discourses. Section Two, consisting of Chapters 3, 4, and 5, explores the disjunction between Song scholar-officials’ political stances and intellectual interests in terms of their 36. The only exception is Song Jaeyoon’s Traces of Grand Peace. However, as aforementioned, as a pioneering research it focuses on the conception of state policies in Song ritual commentaries but not on ritual history.

12

Empowered by Ancestors

opinions on temple rituals. Chapters 3 and 4 examine two influential ritual debates over the placement and sequence of Song imperial ancestors in the Imperial Temple in the 1070s, a period that overlapped with the heyday of the political reforms led by Wang Anshi and Song Shenzong 宋神宗 (r. 1067–1085). Drawing on these two case studies, I plan to bridge Song intellectual history with Song factionalism in a contextual way that helps elucidate how ritual debates both divided and integrated different groups of scholar-officials in the reform eras of the late eleventh century. Chapter 5 deals with the general intellectual background where the ritual debates of the 1070s were rooted. It discusses several crucial interpretations about the Imperial Temple raised by Wang Anshi’s disciples from the late eleventh to the early twelfth centuries. This chapter also illustrates how Wang Anshi’s disciples as ritualists elaborated and revised his ritual theory and thus contributed to the revival of ancient rituals under Song Huizong’s 宋徽宗 (r. 1100–1125) reign. Section Three, composed of Chapters 6 and 7, examines the Daoxue conceptions of the Imperial Temple and traces the intellectual origin of some key ritual norms in later Chinese societies to these conceptions. Chapter 6 focuses on the link between the eleventh-century ritual discussions on imperial ancestral sacrifices and the Daoxue conception of the Imperial Temple, represented by the prominent Daoxue scholar Zhu Xi and some of his best students in ritual scholarship. Chapter 7 analyzes the adoption of some ritual norms of the Imperial Temple in the Southern Song and Yuan societies. In the conclusion, I will discuss the repercussions of Song debates over the Imperial Temple in later periods, followed by a reflection on the modernization of Confucian ancestral rituals.

Section One Imperial Temple: Early History and Main Controversies

1 Pre-Song Interpretations of the Imperial Temple

Centralized states in early imperial China paid special attention to the meaning and power of imperial ancestral rituals. In extant Chinese Classics, a myriad of paragraphs and sections are found related to imperial ancestral rituals, such as the series of sacrificial odes and chants in the Book of Songs.1 Few of them, however, deal with the concrete setting of an early Imperial Temple prior to the first century. Archaeological excavations provide some evidence, which has been analyzed by recent studies.2 This chapter revisits the formation of various conceptions of the Imperial Temple in early imperial China by primarily focusing on the political and cultural implications of these conceptions. After the first century, these conceptions eventually converged into two major interpretations of imperial ancestral rituals: one emphasized political merits and the other emphasized Confucian value of filial piety. Although most interpretations about the Imperial Temple were developed during the Western Han (206 BC–AD 9), it is possible to trace the precursor of the Imperial Temple in the form of ancestral buildings to the pre-Qin 秦 (221–206 BC) period. An overview of pre-Qin ancestral buildings and related worship, therefore, is necessary before probing into the Western Han interpretations of the Imperial Temple.

Centralizing the Imperial Temple in Early China Following David Keightley’s interpretation of Shang ancestral worship, some scholars conceptualize early ancestral sacrifices as a symbolization of ritual “generationalism” underpinning the religious dynamic of ancient China.3 Ancient Shang people oriented themselves according to a particular cosmo-ritual order. In his study of the Late Shang 商 (ca. 1200–1045 BC) oracle bones and bronze inscriptions, Keightley 1. Edward Shaughnessy, Before Confucius: Studies in the Creation of the Chinese Classics (Albany: State University of New York Press, 1997), 165–96. 2. Timothy Danforth, “The Imperial Ancestral in China’s Western Han Dynasty: Institutional Tradition and Personal Belief ” (PhD diss., Harvard University, 2006), 42–52. 3. See, for instance, Rodney Taylor, The Religious Dimensions of Confucianism (New York: State University of New York, 1990), 10; Zhang Guangzhi 張光直 (Chang Kwang-chih), Zhongguo qingtong shidai 中國青銅時代 (Beijing: Sanlian shudian, 1983), 202–7.

16

Empowered by Ancestors

draws anthropologists’ and historians’ attention to what he refers to as the “ancestral landscape” of the Shang dynasty. Through a careful study of Shang spatial and calendrical structure, such as the Five-Ritual cycle and the quadrate order of the spatial system of sifang 四方, Keightley reveals a well-structured cosmological orientation within the Shang ritual rubric.4 According to Keightley, despite the shifting nature of Shang cosmology, it always pointed to a center. It was in the Shang centrality that political power and ritual divinity intermingled and developed into a cult of ritual politics. Architectural constructions and designs, such as the ya 亞-shaped configuration of ancestral architecture and tombs, have been cited to exemplify the centralization of imperial ancestral worship in ancient Chinese culture.5 In this light, Shang ancestral buildings symbolized the sovereign power of the Shang kings and also their spiritual connections with their ancestors. At the heart of Shang’s political order was the kings’ supreme power to perform certain ancestral rituals at a central site. From a broad perspective, Keightley’s reading tends to see Shang ancestral worship as a ritual reflection of what he called the “bureaucratic mentality” that characterized later Chinese culture and Confucianism.6 In practice, Shang ancestral rites became a way through which living people could sanctify their deceased ancestors. Reciprocally, ritual communications between Shang ancestors and their descendants contributed to the political supremacy of the Shang imperial house. Moreover, Shang ancestral worship routinized imperial ancestors into a hierarchical structure of generations. Shang ancestors were not only worshipped by the living; they were also “managed” by the living in terms of ritual performance. While the Shang ruling class monopolized the power to divinize ancestors, Shang ancestors could draw spiritual power from ancestral sacrifices to maintain their positions in the divine world, which accorded with their secular positions in the corresponding ancestral buildings.7 In this sense, Shang ancestral practices established the fundamental pattern of ancestral practices that gradually developed into a variety of generation-differentiating rituals in the succeeding dynasties. Among them the most influential one was the di 禘 sacrifice that aimed at making offerings to Shang

4. David Keightley, The Ancestral Landscape: Time, Space, and Community in Late Shang China (Berkeley: University of California Press, 2000). 5. Aihe Wang, Cosmology and Political Culture in Early China (Cambridge: Cambridge University Press, 2000), 37–46. 6. David Keightley, “The Religious Commitment, Shang Theology and the Genesis of Chinese Political Culture,” History of Religions, no. 17 (1978), 211–16. As Michael Puett pointed out, Keightley’s analysis of the Shang conception of ancestors was shaped by the Weberian conception of bureaucratic society. Michael Puett, To Become a God: Cosmology, Sacrifice, and Self-Divinization in Early China (Cambridge, MA: Harvard University Asia Center, 2002), 36–40. 7. Richard Von Glahn, The Sinister Way: The Divine and the Demonic in Chinese Religious Culture (Berkeley: University of California Press, 2004), 21–25.

Pre-Song Interpretations of the Imperial Temple 17

royal ancestors.8 Based on oracle bone inscriptions, it is known that the Shang practice of di paralleled royal ancestors with the supreme Thearch 帝.9 Other scholars focus on how the post-Shang ancestral rituals evolved in different ways. Timothy Danforth reveals how the fraternal succession in the Shang imperial lineage was replaced by the paternal succession in the Zhou 周 (1046– 256 BC), Qin, and Han dynasties. He also points out the terminological shift from the term zong 宗 in Shang oracle bone texts to the term miao 廟 in referring to postShang ancestral buildings and ritual activities performed within.10 Symbolically, by emphasizing the dominance of linear royal lineage in related rituals, the post-Shang change of succession patterns further “centralized” ancestors who were offered sacrifices in imperial ancestral buildings. Di as state sacrifices was usually held in the suburban Altars near imperial capitals (jiaosi 郊祀). Zhou imperial ancestral rituals offered a different picture. The archaeologist Zhao Huacheng 趙化成 distinguishes the Zhou cemetery system from the Shang’s and argues that the former emphasized more on a centralized sacrificial setting.11 Later dynasties inherited the Zhou concept of centralized ancestral sites. Mark Lewis argues that Qin and Han imperial capitals detached themselves from their local ties by concentrating on the state rituals within themselves.12 Xianyang 咸陽, Chang’an 長安, and Luoyang 洛陽 were all “ritually correct capitals” embodying a sense of centrality.13 The Han imperial architecture miniaturized the cosmos and granted it a center, which was the Imperial Temple of the Son of Heaven (tianzi zongmiao 天子宗廟). Since most records concerning the practice of Zhou rituals had been lost in the chaos of the transition from Qin to Western Han, Han scholars and officials generally understood the Zhou Imperial Temple based on the ritual texts of their times.14 Their relevant ritual knowledge was predominantly textual and lacking in material 8. Keightley, The Ancestral Landscape, 72–73. Although the origin of the di sacrifice could be traced back to Shang, only during the early Zhou period the kings of the central Zhou State started to monopolize the di sacrifice and practiced it as a worship made to the Heavenly Lord. See Robert Eno, The Confucian Creation of Heaven, Philosophy and the Defense of Ritual Mastery (New York: State University of New York, 1990), 23–28. Zhou bronze inscriptions document some liturgical details of the di sacrifice. See Huang Yifei 黃益飛, Xi Zhou jinwen lizhi yanjiu 西周金文禮制研究 (Beijing: Zhongguo shehui kexue chubanshe, 2019), 96–101. 9. Shen Wenzhuo 沈文倬, Zong Zhou liyue wenming kaolun 宗周禮樂文明考論 (Hangzhou: Hangzhou daxue chubanshe, 1999), 10–11. 10. Danforth, “The Imperial Ancestral in China’s Western Han Dynasty,” 16–18. 11. Zhao Huacheng, “Cong Shang Zhou jizhong gongmuzhi dao Qin Han dulilingyuanzhi de yanhua guiji” 從商 周集中公墓制到秦漢獨立陵園制的演化軌跡, Wenwu 文物 7 (2006): 41–48. However, extant archaeological findings are inadequate to prove the existence of a centralized Zhou cemetery setting prior to the Warring States period. 12. Puett, To Become a God, 237–41; Mark Lewis, The Construction of Space in Early China (Albany: State University of New York Press, 2006), 169–88. 13. Lewis, The Construction of Space, 183–84. 14. After the Qin disposal of some pre-Qin ritual texts, Confucian scholars in the Western Han collected and compiled extant ritual texts into the forms that later scholars used to name as “ritual Classics.” Considering the complexity of the compilation process of these ritual Classics, I will generally treat the texts in these Classics as Han records and conceptions of Zhou institutions.

18

Empowered by Ancestors

evidence. These Han-compiled ritual texts succinctly describe Zhou and pre-Zhou Imperial Temple settings. Among them of particular importance are: the Wangzhi 王制 (Regulations of the Kings), the Jitong 祭統 (Summary of Sacrifice), the Jifa 祭 法 (Law of Sacrifice), the Dazhuan 大傳 (Great Treatise), and the Rituals of Zhou 周 禮.15 The first four had been compiled into one single volume by Han Confucians, later known as different chapters of the Book of Rites 禮記.16 The Wangzhi records a five-tiered system of ancestral buildings that cascades from the Imperial Temple all the way down to the main chambers in the residence of commoners. It says: As for the Son of Heaven’s seven temples, there were three zhao temples and three mu temples, and that of his Primal Ancestor; comprising seven temples altogether. As for the feudal lords, there were two zhao temples and two mu temples, and that of his Primal Ancestor; comprising five temples altogether. As for the great ministers, there were one zhao temple and one mu temple, and that of his Primal Ancestor, comprising three temples altogether. Officials had one temple. The commoners presented their offerings in their main chambers.17 天子七廟，三昭三穆，與大祖之廟而七。諸侯五廟，二昭二穆，與太祖之廟 而五。大夫三廟，一昭一穆，與太祖之廟而三。士一廟。庶人祭於寢。

This five-tiered system is also recorded in the Jifa text, which provides concrete designations and settings for all these temples and other ancestral spaces. According to the Jifa, other ancestral spaces include the raised altar of dan 壇 and sacrificial yard of shan 墠.18 In contrast to the temples, where the tablets of those ancestors who still interacted with the living people in the form of ancestral spirits were placed, the dan altar and the shan yard were built to house those ancestors who were excluded from the ancestral space of interaction and could no longer affect the living.

15. Against traditional understanding of the Rituals of Zhou as a fabricated text designed by Han scholars, Jin Chunfeng 金春峰 argues that the text itself was correlated with the legalist culture and institutions of the Qin dynasty. See Zhouguan zhi chengshu jiqi fanying de wenhua yu shidai xinkao 周官之成書及其反映的文化與 時代新考 (Taipei: Dongda tushu, 1993), 113–97. For a summary and elaboration of Jin’s points, see David Schaberg, “The Zhouli as Constitutional Text,” in Statecraft and Classical Learning: The Rituals of Zhou in East Asian History, ed. Benjamin Elman and Martin Kern (Boston: Brill, 2010), 35–39. 16. The Book of Rites was a Han compilation of mostly the Warring States ritual writings. Some of its contents might be composed by Western Han scholars with expertise in collecting and researching the ritual legacy of the pre-Han periods. See Michael Nylan, The Five “Confucian” Classics (New Haven, CT: Yale University Press, 2001), 173–75; Michael Loewe, ed., Early Chinese Texts: A Bibliographical Guide (Berkeley: University of California, 1993), 293–97; Wang E 王鍔, Liji chengshu kao 禮記成書考 (Beijing: Zhonghua shuju, 2007). Since the Book of Rites is a late composed collection of pre-Han ritual texts, I regard all the “chapters” in the Book of Rites as individual works and have italicized all the titles. 17. Liji zhushu 禮記注疏 (hereafter LJZS), TSZSSSJ, 2:12.148. According to Zheng Xuan, the character qin 寢 in the last sentence referred to the main chamber of a house (shiqin 適寢). See LJZS, 2:12.148. I consulted James Legge’s translation here. See James Legge, trans., The Sacred Books of China: The Texts of Confucianism (New York: Gordon Press, 1976), 2:220. 18. At the dan altar and the shan yard, on occasions of prayers, sacrifices are offered. But if there is no prayer, there is no sacrifice. If a tablet is removed from the shan yard, the ancestral spirit in it is designated as a ghost and receives no sacrifice. LJZS, 2:46.511.

Pre-Song Interpretations of the Imperial Temple 19

The zhao 昭 and mu 穆 temples in the Wangzhi passage refer to a ritual principle that regulates most temple rituals. The ritual principle of zhaomu 昭穆 regulates the arrangement of ancestors’ spirit tablets (shenzhu 神主). Michael Loewe’s recent publication greatly advanced our understanding of the nature of the zhaomu sequence.19 In general, the zhaomu refers to a parallel setting of ancestral sequence, in which ancestors are located on the left zhao and right mu positions successively, with the “Primal Ancestor” at the center. The Primal Ancestor means the first ancestor of an imperial lineage and is usually designated as dazu 大祖 or shizu 始祖 in ancient texts (literally means, the “great ancestor” or the “primogenitor”).20 Since the Han dynasty, Confucian scholars had considered the zhaomu as a representation of the political succession from the founding ancestor of a royal family to their extant ruler.21 Imperial ancestors’ positions in the zhaomu sequence also signified their political legacy, especially their overall contributions to the empire. In this light, the zhaomu sequence in the Imperial Temple represents the authority of the imperial ancestors. As the most significant principle in the rubric of temple rituals, zhaomu had been frequently mentioned in the ritual discussions since the first century and then became a particularly controversial topic in later periods. The setting of seven Imperial Temples recorded in the Wangzhi and the Jifa offered the basic model for all later interpretations and imaginations of the Imperial Temple. The great Han Classicist Zheng Xuan 鄭玄 (127–200) considered the seventemple setting as a real Zhou setting, in which two tiao 祧 temples were established to permanently honor the two Zhou sage-kings, King Wen 文王 and King Wu 武 王.22 Zheng was possibly the first scholar who defined the two tiao temples in Jifa as permanent ancestral buildings based on political contributions (gong 功) and virtues (de 德). According to his reasoning, King Wen and King Wu of Zhou should be honored in tiao temples due to their great contributions to the founding of the dynasty. For a regular setting of the Imperial Temple, Zheng argued that it should only consist of five temples. 19. Michael Loewe, Problems of Han Administration, 4–14. Also see Loewe, “The Imperial Way of Death in Han China,” in State and Court Ritual in China, ed. Joseph P. McDermott (Cambridge: Cambridge University Press, 1999), 93. 20. Some scholars argue that shizu is a Han-invented term that has never appeared in pre-Qin sources. Li Hengmei, “Lidai zhaomuzhidu zhong shizu chenghu zhiwu lizheng” 歷代昭穆制度中始祖稱呼之誤厘正, Qiushi xuekan 求是學刊 (1995:3): 95–100; also see Gao Mingshi, “Lifa yiyi xiade zongmiao,” 38–39. 21. Archaeologists have not yet proved that zhaomu was a Zhou ritual practice. The majority of Chinese archaeologists argue that the zhaomu order was only an ideological construct of the Confucians after the Warring States. See, for example, Li Boqian 李伯謙, “Cong Jinhou mudi kan xizhou gongmu mudi zhidu de jige wenti” 從晉侯墓地看西周公墓墓地制度的幾個問題, Kaogu 考古 11 (1997): 51–60, especially 53–55. There is one particular set of ancestral buildings of the Qin 秦 State of the Warring State period that has been suspected of adopting the zhaomu order: the Majiazhuang 馬家莊 temple complex located at the Fengxiang 鳳翔 county of the modern Shanxi 陜西. However, a strict zhaomu order is indiscernible among the spatial arrangement of excavated ancient tombs. See, for example, Shanxisheng yongcheng kaogudui 陜西省雍城考古隊, “Fengxiang majiazhuang yihaojianzhuqun yizhi fajuejianbao” 鳳翔馬家莊一號建築群遺址發掘簡報, Wenwu 文物 2 (1985): 1–29. I am indebted to the talks with my colleague Lam Weng-cheong for my understanding on the archaeological accounts of the early history of zhaomu. 22. LJZS, 2:12.148; 2:46.511.

20

Empowered by Ancestors

Despite Zheng Xuan’s authority as an eminent Classicist, his theory of a regular setting of five temples for the Son of Heaven is clearly at variance with the Wangzhi record of seven temples. Furthermore, his argument is based on two suspicious apocryphal texts (weishu 緯書) of the second century, namely, the Junmingjue 鈞 命決 (Determination of destiny) and the Qimingzheng 稽命徵 (Verification of destiny).23 Regarding the number of the Zhou Imperial Temples, the Junmingjue and the Qimingzheng differ in their textual records. The Junmingjue records that the Zhou imperial family “had six temples, plus one that was left to the descendants” 周六廟，至於子孫七; the Qimingzheng records that “the Son of Heaven had five temples, two zhao, two mu, and one of the Primal Ancestor” 天子五廟，二昭二 穆，以始祖為五.24 Obviously, Zheng Xuan’s conception of a five-temple setting was based on his interpretation of these two apocryphal texts. Concerning the number of temples, the disjunction in the original text of Wangzhi and Zheng Xuan’s interpretation resulted in later controversies. Apart from the number of the Imperial Temples, the zhaomu order also triggered interesting interpretations of Han ritual texts. The Jitong text defines the zhaomu as a ritual tool to regulate relations between imperial family members. Therefore, if there is a ritual service in the Imperial Temple, all ancestors will receive their proper offerings in terms of their familial relations according to zhaomu. This is called “a ritual discrepancy based on the degree of intimacy” 親疏之殺.25 The Dazhuan, which was possibly compiled into the Book of Rites during the Western Han, says: “In a sacrificial ritual when all clan members assemble to share the food, their seating plan should be arranged according to zhaomu, in order to adopt ritual differentiations between the members” 旁治昆弟，合族以食，序以昭繆（穆） ，别之以禮義.26 Unlike the Wangzhi text, which solely relies on Han sources, the Jitong and the Dazhuan contain some pre-Qin sources about ancient aristocratic rituals. Hence, they both emphasize the role played by zhaomu in the differentiation of intra-clan relations within the aristocratic clans. Whereas the Jitong and the Dazhuan stress the zhaomu’s aristocratic origin, the Rituals of Zhou conceptualizes zhaomu based on a bureaucratic vision of Zhou imperial rituals. In this “constitutional document,”27 two offices under the Bureau of Spring (chunguan, 春官, the bureau in charge of ritual affairs) take charge of the zhaomu sequence in the Imperial Temple: the Vice Minister (xiaozongbo 小宗伯) and the Minor Scribe (xiaoshi 小史). According to the Rituals of Zhou and Zheng Xuan’s commentary on it, zhao and mu signified the positions of spirit tablets in the 23. Both texts are found in a Han collection of apocryphal texts titled Liwei 禮緯. The Liwei has been lost. But some of its excerpts are preserved in a Qing collection of Han apocryphal texts. 24. Ma Guohan 馬國翰 (1794–1857), Yuhan shanfang jiyishu 玉函山房輯佚書 (Shanghai: Shanghai guji chubanshe, 1990), 2:58.31 (Junmingjue); 2:54.25 (Qimingzheng). 25. LJZS, 2:49.536; Legge, The Sacred Books of China, 4:246–47. For a detailed explanation on the familial relations mentioned in Jifa, see Zhu Bin 朱彬 (1753–1834), Liji xunzuan 禮記訓纂 (Beijing: Zhonghua shuju, 1996), 729. 26. LJZS, 2:34.393; Legge, The Sacred Books of China, 4:61. 27. David Schaberg, “The Zhouli as Constitutional Text,” 33–63.

Pre-Song Interpretations of the Imperial Temple 21

Zhou Imperial Temple. The xiao zongbo office was responsible for “differentiating spirit tablets based on a zhaomu order in the Imperial Temple” 辨廟祧之昭穆. Zheng Xuan commented on this line: “Except the Primal Ancestor, (tablets of) ancestors who belong to the father’s line are designated as zhao; (tablets of) ancestors who belong to the son’s line are designated as mu” 自始祖之後, 父曰昭, 子曰穆.28 Zheng Xuan’s commentary on this very line of the Rituals of Zhou imposed a patriarchal idea upon zhaomu. Historically, Zheng’s conception of zhaomu originated from a significant Han ritual discussion on the Imperial Temple’s spatial arrangement. The ritual discussion occurred in the reign of Emperor Yuandi of Han 漢元帝 (Liu Shi 劉奭, r. 48–33 BC), when the chancellor (chengxiang 丞相) Wei Xuancheng 韋玄成 (d. 36 BC) and his colleagues submitted memorials to Yuandi in response to his inquiry about an earlier memorial submitted by Gong Yu 貢禹, the advisory counselor (jian dafu 諫大夫) of Yuandi’s court.29 Michael Loewe and Timothy Daforth have already examined this discussion based on the two great Han private histories, the Shiji 史記 (Historical Record) and the Han shu 漢書 (History of the Western Han). In the following, I would focus on how the reasoning used by Han Confucian officials was crucial in establishing a court-centered discourse of the Imperial Temple and thus triggered a series of interpretations in approaching temple rituals. In his memorial, Gong Yu advocated for a revival of the ancient arrangement of the Imperial Temple in the setting of seven temples. Although Gong Yu’s advocacy was approved by Yuandi, it had not been implemented before Gong’s death in 44 BC. In 40 BC, a series of memorials submitted by Wei Xuancheng and his colleagues reactivated the whole ritual debate.30 In the memorial titled “Huimiao yi” 毀廟議 (On the Abolishment of Temples), Wei and his colleagues suggested a revision of Gong Yu’s plan of seven temples. Wei personally regarded the arrangement of seven temples as inappropriate. He suggested abolishing the Imperial Temples of Huidi 惠 帝 (Liu Ying 劉盈, r. 195–188 BC), Wendi 文帝 (Liu Heng 劉恆, r. 180–157 BC), and Wudi 武帝 (Liu Che 劉徹, r. 141–87 BC) to maintain a maximum of five Imperial Temples.31 Wei further argued that the Zhou setting of seven temples was inapplicable to the Han context, because the latter did not have meritorious ancestors like King Wen and King Wu of Zhou.32 Wei Xuancheng’s memorial revealed a court-centered conception of the Han Imperial Temple. In Wei’s opinion, an ideal temple setting should “illustrate that 28. Zhouli zhushu, TSZSSSJ, 2:19.187. 29. Loewe, Problems of Han Administration, 47–53. Danforth, “The Imperial Ancestral in China’s Western Han Dynasty,” 112–68. 30. Loewe, Problems of Han Administration, 47–53. 31. Ban Gu 班固 (32–92), Han shu (Beijing: Zhonghua shuju, 1962), 73.3118. 32. Loewe and Deng Zhirui 鄧智睿 have briefly discussed the influence of Wei Xuancheng’s memorials on Han Yuandi’s ritual reforms concerning temple arrangements; see Loewe, Problems of Han Administration, 53–56. Deng, “Tianxia yijia dao yijia tianxia: yi tangsong miaoyi yu junwei qianghua wei zhongxin de taolun” 天下 一家到一家天下：以唐宋廟議與君位強化為中心的討論 (MA Thesis, National Taiwan Normal University, 2011), 21–26.

22

Empowered by Ancestors

there is a limit (in the number of temples) according to the degree of intimacy from close ancestors to distant ones” 親疏之殺示有終也.33 In the symmetrical distribution of the Imperial Temple, the primogenitor of the dynasty, who usually had a legendary origin, was placed at the center of the whole configuration as the Primal Ancestor, facing east. On the left of the Primal Ancestor would be the odd-numbered zhao ancestors, arranged in order of seniority, first the Primal Ancestor’s son, then his great-grandson, and so forth; on the right of the Primal Ancestor would be the even-numbered mu ancestors, first the Primal Ancestor’s grandson, then his great-great-grandson, and so forth. This ideal setting of the Imperial Temple, as some scholars have argued, reinforced the unity of an imperial clan and the Primal Ancestor as the origin.34 According to Wei Xuancheng, all Han imperial ancestors with zhao and mu designations could receive food offerings in the Primal Ancestor temple in the court sacrifice of xia 祫.35 Extant Han sources record a number of ritual discussions on the Imperial Temple. However, they fail to offer a concrete description about the spatial arrangement of the temples in reality. An argumentative essay written by the Eastern Han (25–220) scholar Cai Yong 蔡邕 (132–192) briefly describes the location of early Imperial Temples. Cai Yong argued that in ancient times the Imperial Temples were established near the “main chambers” (qin 寢) of deceased ancestors. As the temple stored the spirit tablets of imperial ancestors, the “main chambers” functioned as keeping the ancestors’ living households.36 However, after the First Emperor of Qin 秦始皇 (r. 247–220 BC) had separated the “main chamber” from the Imperial Temple and placed it near his imperial tomb, the Western Han emperors followed by establishing “main chambers” at the burial grounds near the Han capital Chang’an. However, Cai Yong failed to reveal the exact location of the Han Imperial Temple; he only mentioned that an ideal temple arrangement should follow the setting recorded in the Wangzhi.37 Only from the records of some post-Han geographical texts could we know that the first Han Imperial Temple of Gaozu was located within

33. Han shu, 73.3118. 34. Patricia Ebrey, “The Early Stages in the Development of Descent Group Organization,” in Kinship Organization in Late Imperial China: 1000–1940, ed. Patricia Ebrey and James Watson (Los Angeles: University of California Press, 1986), 27. 35. Han shu, 73.3118. Xia as one of the seasonal sacrifices appears in the Book of Rites, especially in the Wangzhi and the Zengziwen 曾子問 (Questions of Zengzi) chapters. Gongyang 公羊 scholars in the Han dynasty perceived the xia sacrifice as a collective offering to all ancestors in the Imperial Temple (heshi 合食)—a practice that could be traced back to the Shang sacrifice of yin 衣 (pronounced as yin 殷). See Shen Wenzhuo, Zong Zhou liyue wenming kaolun, 11. According to the Chunqiu gongyang zhuan 春秋公羊傳 (Gongyang Commentary on the Annals), the Duke Wen of Lu 魯文公 practiced a magnificent xia sacrifice in the second year of his reign. The Gongyang commentators admitted that this xia was a grandiose sacrifice, yet Duke Wen failed to promulgate it as a state sacrifice. Chunqiu gongyang zhuan zhushu, TSZSSSJ, 3:13.108. 36. Cai Yong, “Du duan” 獨斷, Caizhonglang ji 蔡中郎集 (Shanghai: Zhonghua shujiu, 1936), Waiji 外集:4.20a–b. 37. Cai Yong, “Du duan,” Waiji: 4.8a. Cai somewhat elaborated the Wangzhi passage by emphasizing that the Imperial Temple should be located within the imperial palace, inside its second exterior door (kumen 庫門) but outside the third one (zhimen 雉門).

Pre-Song Interpretations of the Imperial Temple 23

the Chang’an city and other Western Han temples were located near the tombs of Han emperors in the suburban areas of Chang’an.38 Another Eastern Han scholar, Wei Hong 衛宏, composed possibly the most detailed description of the temple of Han Gaozu, where the state sacrifice of xia was performed. It records: In the xia sacrifice that is performed once per three years, the (tablets of the) descendants of the Han imperial line are placed in the temple of Gaozu according to the zhaomu sequence. All ancestral spirits share the sacrificial offerings in Gaozu’s temple, with seats on both left and right sides. Gaozu sits on the north, facing south . . . His sons are designated as zhao ancestors; his grandsons are designated as mu ancestors. The zhao ancestors all sit on the southwest of the Primal Ancestor, beneath a curved screen; and the mu ancestors on the southeast, beneath a curved table.39 宗廟三年大祫祭，子孫諸帝以昭穆坐於高廟，諸隳廟神皆合食，設左右坐。 高祖南面……子為昭，孫為穆。昭西南，曲屏風；穆東南，皆曲几。

Wei Hong had documented the setting of the Han Imperial Temple. Yet, his records are not free from doubt for two reasons. First, the above passage is quoted from Qing scholars’ collections of Han institutional sources, which renders its authenticity somewhat doubtful.40 Second, even if the records were written by Wei Hong, they do not necessarily represent the concrete Imperial Temple setting in Han state sacrifices. Lacking in crucial details, such as the scale of the temple, Wei Hong’s records reiterate the temple settings that are documented in other Han texts, especially the Dazhuan text in the Book of Rites. Instead of recording the Han temple’s real configuration, Wei might have just paraphrased the idealized description of the Imperial Temple in the Han texts. A clear rebuttal to Wei Hong’s description can be found in Cai Yong’s critical essay on the arrangement of the Han Imperial Temple, in which Cai criticized the Han practice of ignoring the limit of seven temples and the zhaomu principle.41 Considering Cai Yong’s proficiency in both ancient rituals and contemporary history, his criticism of the Han Imperial Temple should be more reliable than Wei Hong’s records. Cai’s statement also reveals his orthodox thinking of ritual practice. He saw the Imperial Temple as a legacy of the Zhou mingtang 明堂 (literally, the 38. Among these geographical texts, the most reliable one is Sanfu huangtu 三輔黃圖 and Yan Shigu 顏師古’s annotation on Han shu. See Yang Kuan 楊寬, Zhongguo gudai lingqin zhidu yanjiu 中國古代陵寢制度研究 (Shanghai: Shanghai guji chubanshe, 1985), 20–21 and 201–28. 39. Wei Hong, Han jiuyi 漢舊儀, compiled by Sun Xingyan 孫星衍 (1753–1818) in Congshujicheng chubian 叢書 集成初編 (Beijing: Zhonghua shuju, 1985), 30. 40. The Qing scholars extracted records concerning Han institutions from Wei, Jin, and Tang sources, especially those encyclopedic collections (leishu 類書). However, these collections often fail to mention the sources of their excerpts, which greatly increases the difficulty of text dating. Wei Hong’s text should be traced back to the Jin scholar Sima Biao’s 司馬彪 (~246–306) Xu Han shu 續漢書. Tang collections such as the Chuxueji 初 學記 and the Yiwen leiju 藝文類聚 also quote it, with slightly different wordings. 41. Cai, “Zongmiao diehui yi” 宗廟迭毀議, Caizhonglang ji, 9.8a.

24

Empowered by Ancestors

Hall of Brightness), an architectural complex that Cai claimed to be the illustration of the Heaven’s order of governance to the King.42 Cai Yong’s words crystallized the Han Gongyang scholarship, in which the ideal administration of the Three Dynasties was utilized to justify contemporary intellectual and social reforms. Based on some pre-Qin interpretations of the Zhou ritual legacy, the Han Gongyang master Dong Zhongshu 董仲舒 (179–104  BC) developed his own theory of historical cycles and the transition of the Mandate of Heaven.43 The extant collection of Han Gongyang scholarship, the Chunqiu fanlu 春秋繁露, considers the Imperial Temple as a ritual space to perform the four seasonal sacrifices, respectively, the spring ci 祠, the summer yue 礿, the autumn chang 嘗, and the winter zheng 蒸 (sishi zhi ji 四時之祭).44 Han ritual texts describe the four seasonal sacrifices of the Imperial Temple differently. For example, the Jitong describes the yue sacrifice as a seasonal sacrifice in spring, but not in summer.45 Gongyang Confucians generally explained this discrepancy by arguing that yue was once a spring sacrifice in the Shang dynasty but was then modified into a summer sacrifice in the Zhou.46 He Xiu 何休 (129–182), the Han Gongyang scholar who annotated the Annals, defined the term yue as the boiling of premature crops in summer. Hence, the Zhou practiced yue sacrifice in summer by taking premature crops as one of the offerings.47 Gongyang scholars who compiled the Chunqiu fanlu shared with He Xiu the same reasoning and consolidated the understanding of the yue sacrifice as a temple sacrifice in summer. In anticipation of ritual reforms, these Gongyang scholars attempted to modify temple sacrifices according to their understanding of Zhou ritual legacy. The Han Confucians’ intention to implement an ideal setting of the Imperial Temple was eventually outweighed by economic factors. The “literary officials” (wenxue 文學) in the Yantie lun 鹽鐵論 (Discourse on Salt and Iron Policies) rightly indicated that the full implementation of a ritually satisfactory Imperial Temple setting bore a considerable cost, which created a massive burden for the Western 42. Cai, “Mingtang yueling lun” 明堂月令論, Caizhonglang ji, 10.1a–b. Mingtang in ancient times referred to a ritual structure that was built for offering sacrifices to Heaven. John Henderson, The Development and Decline of Chinese Cosmology (New York: Columbia University Press, 1984), 75–85. For a thorough study of mingtang, especially the evolution of its architectural forms, see Hwang Ming-chorng, “Ming-tang: Cosmology, Political Order and Monuments in Early China” (PhD diss., Harvard University, 1996), 7–10, 27–118; for the functions and meaning of mingtang in Zhou period, see Xue Mengxiao 薛夢瀟, “Zhouren mingtang de benyi chongjian yu jingxue xiangxiang” 周人明堂的本義、重建與經學想像, Lishi yanjiu 歷史研究 (2015.6): 22–42. 43. For a comprehensive introduction of Dong Zhongshu’s scholarship and the Han Gongyang tradition, see Sarah A. Queen, From Chronicle to Canon: The Hermeneutics of the Spring and Autumn, According to Tung Chung-shu (New York: Cambridge University Press, 1996), 13–38, 115–26, 187–201. 44. Su Yu 蘇輿 (1874–1914), Chunqiu fanlu yizheng 春秋繁露義證 (Beijing: Zhonghua shuju, 1992), 402. Shen Wenzhuo has argued that the zheng 蒸 sacrifice originated from the Shang practice of offering fresh and seasonal food to royal ancestors (jianxin 薦新). Shen Wenzhuo, Zong Zhou liyue wenming kaolun, 11–13. In Dong Zhongshu’s era most Han Confucians perceived the zheng sacrifice as one of the four seasonal sacrifices. 45. LJZS, 2:49.148. 46. Su, Chunqiu fanlu yizheng, 406. 47. Chunqiu gongyang zhuan zhushu, TSZSSSJ, 3:5.39.

Pre-Song Interpretations of the Imperial Temple 25

Han court.48 Indeed, the Han court might have considered reducing the expenditure on temple rituals—the best way was to scale down the original configuration of the Imperial Temple in the long term. Although the interregnum reign of Wang Mang 王莽 (r. 9–23) was characterized by an extravagant display of the Imperial Temple in the name of the Wang family, the ensuing Eastern Han dynasty promoted a more economical way to practice temple rituals.49 The Eastern Han court paid less attention to the grandiose practice of temple rituals. Archaeologists argue that the Western Han imperial burial grounds followed a parallel-oriented ritual order, which could be rarely observed in the Eastern Han cases.50 The Eastern Han court was also less interested in managing the correct configuration of the Imperial Temple, let alone an overwhelming revision of temple rituals. The declining interest might be attributed to the deteriorating financial situation of the Eastern Han Empire, as the court attempted to find a more economical way to practice ancestral rituals. Emperor Guangwu of Han 漢光武帝 (r. 25–57), the founder of the Eastern Han dynasty, began to place all preceding Han emperors in a single Imperial Temple at Chang’an, named the Gaozu temple of Han 漢高 祖廟.51 Guangwu and his officials might render this practice as a temporary solution, considering the constraints at the beginning of Guangwu’s reign. However, the practice of arranging all the imperial tablets within one temple turned out to be much more influential than Guangwu and his officials might have thought. When Emperor Mingdi of Han 漢明帝 (r. 57–75) died, he left an edict that prohibited the construction of any new temple for him.52 Instead, Mingdi’s tablet was placed in the temple of Emperor Guangwu to receive regular temple sacrifices. When Emperor Zhangdi 章帝 (r. 75–88) died, he also left an edict forbidding his successors to erect any new temple.53 The ways of how Guangwu arranged the tablets of preceding Western Han emperors and how Mingdi’s and Zhangdi’s tablets were arranged by their own wills became a new ritual norm of the Imperial Temple: the original 48. Wang Liqi 王利器, Yantie lun jiaozhu 鹽鐵論校注 (Beijing: Zhonghua shuju, 1992), 162–63. Also Essen Gale, Discourses on Salt and Iron: A Debate on State Control of Commerce and Industry in Ancient China (Taipei: Ch’eng-wen Publishing Company, 1967), 79. 49. Concerning the multiple Imperial Temples set up by Wang Mang for his ancestors, see Loewe, Problems of Han Administration, 56–62. Archaeological discovery offers us a clear layout of the complex of Wang Mang’s Imperial Temples. In an excavated site near the modern Xian city, archaeologists located twelve square enclosures that served as the sites of Wang Mang’s nine Imperial Temples and three uncertain ancestral buildings. See Xi Han lizhi jianzhu yizhi 西漢禮制建築遺址, ed. Xianyang shi wenwu kaogu yanjiusuo 咸陽市文物考古 研究所 (Beijing: Wenwu chubanshe, 2003), 6, Figure 1. 50. These archaeologists include Yang Kuan, Du Baoren 杜葆仁, and Li Yufang 李毓芳. See Lei Baijing’s 雷百景 and Li Wen’s 李雯 article for a brief summary of their arguments about the zhaomu setting of the eleven burial grounds of Western Han emperors. Lei and Li, “Xihan diling zhaomu zhidu zaitantao” 西漢帝陵昭穆制度再 探討, Wenbo 文博 (2008:2): 48; also see Shen Ruiwen 沈睿文, “Xihan dilinglingdi zhixu” 西漢帝陵陵地秩序, Wenbo (2001:3): 22, note 1. For the opposition that the arrangement of the eleven Western Han burial grounds disobeyed the zhaomu order, see Cui Jianfang 崔建芳, “Lun huangquan chuanchengguifan dui Xihan dilingbuju de zhiyue” 論皇權傳承規範對西漢帝陵佈局的制約, Kaogu yu wenwu 考古與文物 (2012:2): 60–64. 51. Fan Ye 范曄 (398–455), Hou Han shu 後漢書 (Beijing: Zhonghua shuju, 2003), 1.27–28. 52. Hou Han shu, 3.131. 53. Hou Han shu, 3.159.

26

Empowered by Ancestors

configuration of multiple temples had been replaced by multiple “chambers” (shi 室) within one temple, with each chamber containing a single emperor’s tablet. Indeed, this new ritual norm of multiple ancestral chambers was adopted by all post-Han dynasties. Alongside the simplified Imperial Temple, the Eastern Han court hardly examined the zhaomu sequence of the Han royal line, except in Emperor Guangwu’s reign, when Zhang Cun 張純 and a few other officials discussed the succession sequence of the Han imperial lineage.54 As Loewe pointed out, during the Eastern Han the zhaomu issue was raised “more for rhetorical purposes than with immediate application in practice.”55 In fact, no Eastern Han account of the Imperial Temple can rival Wei Xuancheng’s memorials in terms of ritual proficiency. Neither does Wei Hong’s description of the Imperial Temple nor Cai Yong’s criticism on temple rituals achieve the same degree of proficiency that Wei Xuancheng and other Western Han scholars have demonstrated in their earlier memorials.

Controversies in the Tang Conceptions of the Imperial Temple Ritual controversies over the Imperial Temple continued throughout the succeeding several centuries after the Han dynasty. Kaneko Shūichi concludes that emperors from the third to the sixth centuries emphasized temple rituals in order to highlight the legitimacy of imperial governance underpinning these rituals. The Imperial Temples of this period symbolized the supreme authority of governing imperial houses. During this period, the elevation of Wang Su’s 王肅 (195–256) theory of seven Imperial Temples was at the heart of relevant ritual discussions. Against Zheng Xuan’s theory of five Imperial Temples, Wang argued that a regular temple setting should have seven temples.56 In an essay titled “Shengzheng lun” 聖 證論, Wang questioned that Zheng Xuan’s annotation on the Wangzhi text diminished the Son of Heaven’s ritual status to the level of feudal lords by championing a five-temple setting as the regular setting.57 Although the Wei 魏 (220–266) dynasty still adopted a five-temple setting in the form of ancestral chambers, the Jin 晉 (265–420) dynasty adopted Wang Su’s theory of seven temples to honor the royal clan of Sima 司馬.58 Personal factors greatly contributed to the elevation of Wang Su’s theory, considering that Wang was the maternal grandfather of Sima Yan 司馬 炎 (236–290), the first emperor of the Jin dynasty. 54. Loewe, Problems of Han Administration, 28–30, 62–70. 55. Loewe, Problems of Han Administration, 28–30, 62–71. 56. Gao Mingshi, “Lifa yiyi xiade zongmiao,” 26–27. 57. LJZS, 2:12.148; Wechsler, Offerings of Jade and Silk, 45–49. 58. The essay has been lost. I quoted it here from the official dynastic history of Jin. See Fang Xuanling 房玄 齡 (579–648) et al., Jinshu 晉書 (Beijing: Zhonghua shuju, 1975), 19.602–8. Notably, the Wei dynasty constructed three permanent temples for Cai Cao 曹操 (155–220), Cai Pi 曹丕 (r. 220–226), and Cai Rui 曹叡 (r. 220–239). Later scholars criticized the Wei establishment of three permanent temples as a deviation from ideal Imperial Temple settings. See Wan Sitong 萬斯同 (1638–1702), Miaozhitu kao 廟制圖考, Siku quanshu zhenben 四庫全書珍本, Series 6 (hereafter SKQSZB), 66a–67a.

Pre-Song Interpretations of the Imperial Temple 27

The divergence between the theories of Zheng Xuan and Wang Su reached its culmination when Confucians of the fifth and sixth centuries joined in the debate over the number of the Imperial Temples. Two Classicists, Zhang Rong 張融 (444– 497) and Ma Zhao 馬昭, criticized Wang Su and advocated for a certain degree of flexibility in determining the number of temples. As they put it: “If some ancestors of the Son of Heaven made great contributions to the society, then the imperial house could build seven temples; if there were no such type of ancestors, then a five-temple setting was fair enough” 有其人則七，無其人則五.59 Possibly inspired by Zheng Xuan, Ma Zhao also quoted a passage from the apocryphal text Liwei in arguing that ancient dynasties did not strictly follow an arrangement of seven temples. Like temple rituals, state sacrifices performed at the suburban Altars near the capital also underwent profound changes during Wei and Jin periods. According to Kaneko Shūichi, emperors of the southern dynasties were more inclined to make sacrifices personally (qinsi 親祀), as compared with the “foreign” dynasties in the northern China.60 On the one hand, by personally participating in state sacrifices, the emperors of the southern dynasties presented themselves as the legitimate rulers of not only the territories of the southern dynasties, but also the suffering territories under the foreign dynasties. On the other hand, the emperors of the northern dynasties were less interested in the performance of the rituals of the “Central Plain” (zhongyuan 中原), mostly because they were weary of the complicated liturgical procedures in these rituals. The Northern Zhou 北周 (557–581) and the Sui 隋 (581–619) dynasties—the latter unified China after nearly three centuries of division—established a temple system consisting of four ancestral chambers. However, the dispute between the theories of Zheng Xuan and Wang Su remained unsolved. Ritual controversies concerning the Imperial Temple re-emerged after the collapse of Sui, when China was soon reunified into a centralized dynasty under the Tang dynasty (618–907). Historians commonly considered Tang as a period that witnessed the consolidation of state authority. Howard Wechsler argues that the intention of most Tang imperial rituals shifted from private secrecy to public exhibition.61 David McMullen notes that the major institution in charge of the Tang Imperial Temples had changed from the Court of Imperial Sacrifices (taichang si 太常寺) to the Court of the Imperial Clan (zongzheng si 宗正寺) in the mid-Tang.62 Other studies explored the “privatization process” of the Tang Imperial Temple and related rituals.63 Compared 59. LJZS, 2:12.148. 60. Kaneko, Chūgoku kodai kōtei saishi no kenkyū, 238–308, especially 258–60, 300–2. 61. Wechsler, Offerings of Jade and Silk, 107–235. 62. David McMullen, “Bureaucrats and Cosmology,” 208. Zhu Yi further elaborated on McMullen’s observation and took it as evidence of the privatization of the Imperial Temples in the Tang-Song transition. Zhu, Shibangguo zhi shenzhi, 195–234. 63. Gao Mingshi, “Lifa yiyi xiade zongmiao,” 23–86; Wechsler, Offerings of Jade and Silk; Kaneko, Chūgoku kodai kōtei saishi, 309–430; Tozaki Tetsuhiko, “Tōdai niokeru teikyū ronsōto sono igi,” 82–96.

28

Empowered by Ancestors

with the rich literature on the Tang Imperial Temple from the “public-versusprivate” perspective, studies on relevant ritual debates are scarce. By examining some crucial questions in these ritual debates, I seek to trace the development of two influential ritual discourses that revealed a tension between state authority and Confucian values in the eleventh century. The Tang Imperial Temple established a complete model for later dynasties in several aspects. First, it integrated a new, post-Solstice sacrifice into the traditional system of seasonal sacrifices, namely the laxiang 臘享 (literally, the “sacrifice in the last month of a year”).64 The system of four regular temple sacrifices evolved into a new system of “five annual sacrifices” (wuxiang 五享), including the spring ci, the summer yue, the autumn chang, the winter zheng, and the post-Solstice sacrifice of laxiang.65 Second, in addition to the annual performance of five regular sacrifices, the Tang court officially categorized the two state sacrifices of xia and di as templeoriented rituals and performed them in the temple.66 Third, the Tang court initiated the practice of placing multiple tablets of empresses in a single chamber, which basically reflected the personal interest of Tang emperors.67 Last and foremost, owing to the Tang emperors’ interest in performing imperial rituals, the arrangement of their imperial ancestors in the temple was more complicated than the cases of previous dynasties, thus generating new questions, approaches, and ritual discourses. Modeled on the Northern Zhou and Sui precedents, the early Tang court established four chambers within one Imperial Temple complex to house the spirit tablets of Tang imperial ancestors. In 618, Li Yuan 李淵 (566–635), the first Tang emperor, worshipped his ancestors up to the fourth generation and bestowed his grandfather Li Hu 李虎 (d. 551) a posthumous title (shihao 諡號) of Emperor Jin 景皇帝.68 Li Hu was also given the temple title of taizu 太祖 (Founding Ancestor) in conjunction with offerings to Heaven (peitian 配天) in suburban Altar sacrifices. Li Yuan’s father, Li Bing 李昞, was bestowed a posthumous title of Emperor Yuan 元皇帝. The other two distant ancestors of the Li family were bestowed the titles of the King of Xuanjian 宣簡王 (Li Xi 李熙, Li Yuan’s great-great-grandfather) and the 64. The laxiang originated from an early ritual practice in the first century that was influenced by the Han advocacy of the “Theory of Five Virtues” 五德終始說. In the beginning, the ritual itself referred to a ceremony through which emperors offered the preys that they got from imperial hunts to their ancestors in winter. It was not performed as a temple ritual until the mid-Tang. For the origin of the laxiang sacrifice, also see Chan Hok-lam 陳學霖, “Song Jin erdi yiqi ding tianxia: Xuanhe yishi kaoshi yize” 宋金二帝弈棋定天下──《宣和 遺事》考史一則, in Song shi lunji 宋史論集 (Taibei: Dongda tushu gongsi, 1993), 217–18, especially note 6. 65. The Tang official ritual code, the Da Tang kaiyuan li 大唐開元禮 (Kaiyuan Ritual Code of the Great Tang) documents the liturgical procedures of the five Tang annual sacrifices. Xiao Song 蕭嵩 et al. Da Tang kaiyuan li (Beijing: Minzu chubanshe, 2000), 842:37.1a–22b; 38.1a–13b. 66. There are four juan in the extant Da Tang kaiyuan li dealing with xia and di sacrifices, respectively, “Emperors making xia sacrifice in the temple” 皇帝祫享於太廟, “Responsible officials and liturgical procedures involved in the xia sacrifice in the temple” 祫享於太廟有司攝事, “Emperors making di sacrifice in the temple” 皇帝禘 享於太廟, and “Responsible officials and liturgical procedures involved in the di sacrifice in the temple” 禘享 於太廟有司攝事. Da Tang kaiyuan li, 842:39–42. 67. Zhu, Shibangguo zhi shenzhi, 207–9. 68. Wang Pu 王溥 (922–982), Tang huiyao 唐會要 (Beijing: Zhonghua shuju, 1955), 1.1.

Pre-Song Interpretations of the Imperial Temple 29 Illustration 1.1: A Qing Illustration of the Tang Imperial Temple

Source: Wang Sitong 萬斯同, “Tang chu simiao tu” 唐初四廟圖, Miaozhitu kao 廟制圖考, Siku quanshu zenben 四庫全書珍本, Series 6, 82a.

30

Empowered by Ancestors

King of Yi 懿王 (Li Tianxi 李天錫, Li Yuan’s great-grandfather).69 The whole setting basically corroborated Zheng Xuan’s theory of the Imperial Temple, consisting of four regular chambers and a permanent one for the Primal Ancestor. As the Tang court did not formally designate a Primal Ancestor, the Primal Ancestor chamber was left vacant for future decision. The primitive setting of the Tang Imperial Temple was the result of a relatively prudent decision. Li Yuan only traced the origin of the Tang imperial lineage back to Li Xi. Li Xi was indeed the most distant ancestor with whom the Tang imperial clan could confirm a clear connection.70 After Li Yuan’s reign, the Tang court started to rewrite the imperial lineage records and enrich them with some celebrated individuals from the aristocratic Li clan of Longxi 隴西李氏.71 The Tang court had officially recognized some of these individuals as imperial ancestors in temple rituals since the mid-seventh century. Although Li Yuan had bestowed the temple title of taizu to Li Hu, he had not officially recognized Li Hu as the Primal Ancestor. In fact, the early Tang court gave no clear instructions for the recognition of the Primal Ancestor. Given the adequate number of chambers in the Imperial Temple, the Tang court left the Primal Ancestor chamber vacant and stored Li Hu’s tablet in one of the six regular chambers. The Primal Ancestor issue became more important as time progressed, when the spirit tablets of more emperors were moved into the temple chambers in the succeeding decades. When Wu Zetian 武則天 (r. 684–705) interrupted the Tang reign, the number of temple chambers of the Li family in Chang’an was reduced from six to three.72 Correspondingly, Emperor Wu established a new Imperial Temple for the Li family in the eastern capital Luoyang. When Emperor Zhongzong 唐中宗 (Li Xian 李顯, r. 684, 705–710) reclaimed the emperorship in 705, the Imperial Temples of the Li family in both Chang’an and Luoyang were ill-managed, short of proper maintenance and management. To restore the dignity of the Li family, Zhongzong elevated the downsized Imperial Temple in Chang’an to its original scale. Moreover, he renovated the Imperial Temple in Luoyang in conformity with the dynastic scheme of dual capitals.73 The restoration of the Tang Imperial Temple in Chang’an triggered a new debate on the number of ancestral chambers, as well as the Primal Ancestor 69. Tang huiyao, 1.1. 70. Chen Yinque 陳寅恪, Tangdai zhengzhi shishu lungao 唐代政治史述論稿 (Shijiazhuang: Hebei jiaoyu chubanshe, 2002), 163–75. 71. The native place of the imperial Li family has been a controversial topic in Tang studies since Chen Yinque put it forward. Howard Wechsler agreed to Chen Yinque’s argument that the noble origin claimed by the Tang house was a deliberate fabrication. Howard Wechsler, “The Founding of the T’ang Dynasty: Kao-Tsu (Reign 618–26),” in The Cambridge History of China, Vol. 3, Part I: Sui and T’ang China, 589–906, ed. Denis Twitchett and John K. Fairbank (Cambridge: Cambridge University Press, 1979), 150–51. 72. For other changes of temple rituals under Wu Zetian’s reign, see Kaneko, Chūgoku kodai kōtei saishi, 325–31. 73. Qing scholars like Xu Song 徐松 (1781–1848) had already studied the general layout of Tang dual capitals based on the Song collections of related sources. Based on Qing scholars’ studies and new archaeological findings, modern scholarship mapped Tang dual capitals and the division of market places (shi 市) and residential

Pre-Song Interpretations of the Imperial Temple 31

issue. Officials like Zhang Qixian 張齊賢 and Yin Zhizhang 尹知章 (fl. 669–718) insisted that Li Hu should be officially recognized as the Primal Ancestor, because of his latent contributions to the foundation of the Tang dynasty. Other officials traced the Tang imperial line back to Li Hao 李暠 (351–417), who was the founder of the Western Liang 西涼 Kingdom in the early fifth-century China and had been officially recognized as a distant ancestor of the Tang imperial clan.74 Zhang Qixian and Yin Zhizhang argued that the Primal Ancestor title should be bestowed upon meritorious ancestors with palpable connections to the “origin of kingship” (wangji 王迹) of the Tang empire. Based on this idea of “origin of kingship,” the two officials claimed that Li Hu, as one of the “Eight State Pillars” (ba zhuguo 八柱國) of the Western Wei dynasty 西魏 (535–557), was a perfect candidate for the Tang Primal Ancestor, owing not only to his aristocratic background and renowned reputation but also his contribution in consolidating the Li clan and thus laying the cornerstone for the Tang. Li Hao, despite his remarkable political merits as the founder of a previous kingdom, contributed nothing to the Tang foundation. Therefore, he should not be considered as a potent competitor for the Primal Ancestor.75 In his memorial, Yin Zhizhang elaborated this meritbased discourse regarding the Primal Ancestor issue.76 He argued that the Primal Ancestor of a dynasty should be the most meritorious ancestor who could remain in the Imperial Temple for hundreds of generations. Other ancestors, with fewer merits, will be removed from the temple successively when they fall out of the limit of seven generations.77 Yin Zhizhang’s argument demonstrates how the merit-based discourse was emphasized in the early Tang context. Social transition in the seventh century could provide an explanation for this emphasis. As known, the early Tang court intended to underplay the influence of great aristocratic families in political and social realms.78 Along with the gradual disintegration of old aristocracy and related institutional factors, the Tang court and society increasingly regarded individual areas (fang 坊) on the level of Tang urban planning in a more concrete way. A classic modern reconstruction of Tang dual capitals is Hiraoka Takeo 平岡武夫, Tangdai de Changan yu Luoyang: ditu 唐代的長安與洛陽： 地圖 (Shanghai: Shanghai guji chubanshe, 1991). 74. Li Hao was a member of the Li clan of Longxi. 75. Additionally, Zhang Qixian argued that if Li Hao deserved the Primal Ancestor position, Gaozu and Taizong would have already bestowed him the relevant title of taizu, instead of giving Li Hu that title. However, since they denied bestowing Li Hao the title, there must be some reasons. Thus, Zhang indicated that to designate Li Hao as the Primal Ancestor would offend Li Hu’s spirit and also violate the wills of Gaozu and Taizong. Wang, Tang huiyao, 12.295. 76. Gao Mingshi claimed that the merit-based approach to ancestral rituals was invented by Tang Confucians to limit the monarchical power of the emperors. See Gao Mingshi, “Lifa yiyi xiade zongmiao,” 50–53. 77. Tang huiyao, 12.296. 78. Tang Changru 唐長孺, Weijin nanbeichao suitangshi sanlun 魏晉南北朝隋唐史三論 (Wuhan: Wuhan daxue chubanshe, 1992), 370–404. Nicolas Tackett’s recent study attributes the destruction of Tang aristocratic families to the rebellions in the late ninth century, represented by Wang Xianzhi 王仙芝 and Huang Chao 黃巢 (835–884). See Tackett, The Destruction of the Medieval Chinese Aristocracy (Cambridge, MA: Harvard University Asia Center, 2014), 187–234.

32

Empowered by Ancestors

reputation and personal achievement as criteria of excellence.79 The same criteria also applied to imperial ancestral rituals. Li Hu was a qualified Primal Ancestor not only because he was the grandfather of the first Tang emperor but also because he demonstrated himself as a powerful warlord with reputable military achievements. After all, it was Li Hu, rather than the distant ancestor Li Hao, who contributed to the rise of the Tang in a concrete, traceable way. When Emperor Ruizong 睿宗 (Li Dan 李旦, r. 684–690, 710–712) died in 716, his ritual position in the Imperial Temple brought a new controversy. As Ruizong and Zhongzong were brothers who had been successively enthroned, whether they should be considered as two generations in the Imperial Temple sparked debates.80 Sun Pingzi 孫平子, a commoner from the Henan 河南 prefecture, argued that fraternal succession should be treated as patrilineal succession, because the enthroned brothers were bound by the monarchical-subject relationship when they were alive. Therefore, Ruiong and Zhongzong should be considered more as successive emperors rather than brothers in ritual context.81 According to Sun, Ruizong and Zhongzong should be designated as separate generations and stored in the Imperial Temple. Other officials who served in the Court of Imperial Sacrifices, such as Chen Zhenjie 陳貞節 and Su Xian 蘇獻, considered Ruizong and Zhongzong as belonging to the same generation in terms of seniority.82 However, Chen and Su argued that it was more appropriate to place Zhongzong’s spirit tablet in a “subsidiary temple” (biemiao 別廟, referring to a chamber next to the main chamber of the Imperial Temple), since in fraternal succession only one of the brothers nominally succeeded their father’s emperor title. These officials argued that if Ruizong was ritually designated as the direct successor of his father, Emperor Gaozong 唐高宗 (Li Zhi 李 治, r. 649–683), Ruizong’s brother Zhongzong would be excluded from the line of succession worshipped in the Imperial Temple. In Chen’s and Su’s opinions, even though Zhongzong and Ruizong were successively enthroned when they were alive, it was inappropriate to place their tablets in the Imperial Temple undistinguishably, considering the possible confusion from the co-existence of two ancestors of the same generation. When the Tang court adopted Chen’s and Su’s advice in 716 and removed Zhongzong’s tablet from the Imperial Temple, most scholar-officials assumed that there was only one ancestor for each generation. It was ritually inappropriate to house multiple ancestors of the same generation in the same chamber of the temple.83 However, Tang scholar-officials failed to explain why it was Zhongzong’s 79. Chen Yinque has pointed out that a meritocratic system of civil service examinations was established during Wu Zetian’s reign. Chen, Tangdai zhengzhi shishu lungao, 182–83. 80. Gao Mingshi, “Lifa yiyi xiade zongmiao,” 42–44. 81. Liu Xu 劉昫 (887–946) et al., Jiu Tangshu 舊唐書 (Beijing: Zhonghua shuju, 1975), 25.952–53. Also see Ouyang Xiu 歐陽修 (1007–1072) and Song Qi 宋祁 (998–1061) et al., Xin Tangshu 新唐書 (Beijing: Zhonghua shuju, 1975), 200.5695. 82. Xin Tangshu, 200.5695–96. 83. As Chen and Su put it, given that Gaozong was a zhao ancestor according to the Tang genealogical sequence, and whose tablet was placed in the Imperial Temple, it is ritually appropriate to have two mu ancestors

Pre-Song Interpretations of the Imperial Temple 33

tablet, but not Ruizong’s, that should be removed from the temple. In 722, Emperor Xuanzong 唐玄宗 (Li Longji 李隆基, r. 713–756) decided to refurbish the temple into a nine-chamber configuration, adding back the tablets of Zhongzong and King of Xuanjian (Li Xi, who had been bestowed the more honorable posthumous title of Xianzu 獻祖 in 674) to the temple. Based on his personal understanding of the Confucian text Xiaojing 孝經 (Book of Filial Piety), Xuanzong claimed that only a nine-chamber setting could fully symbolize the “ultimate virtue of filial piety” 至 德之謂孝.84 In spite of the enduring tradition of a seven-chamber setting, which was coined in the Wangzhi text and had been practiced by the Tang court since 635, Xuanzong determined to create his own temple setting to “suit contemporary needs” 因宜以創制.85 Xuanzong’s manipulation of temple chambers reflected how monarchical power fundamentally shaped ritual practices. The ritual theories of Zheng Xuan and Wang Su could hardly impede Emperor Xuanzong’s reform of the Imperial Temple. Although Xuanzong’s officials possessed the power to interpret temple settings and related rituals, it was the emperor’s monarchical power that eventually empowered ancestors and determined their positions in the temple. The Great Rebellion launched by An Lushan 安祿山 (703–757) and Shi Siming 史思明 (703–761) in 755 disrupted further reforms on the Tang Imperial Temple. Understandably, the court neglected ritual affairs in the wartime during the later years of Emperor Xuanzong and the succeeding reign of his son, Emperor Suzong 唐肅宗 (Li Heng 李亨, r.  756–762). When Tang Daizong 唐代宗 (Li Yu 李豫, r.  762–779) was enthroned in 763, the court eventually designated Li Hu as the Tang Primal Ancestor.86 After Tang Daizong’s death in 779, the Tang ritual controversies over imperial ancestral rituals reached a new height.87 Daizong’s death led to a series of reassessments of the generational sequence in the Imperial Temple. The celebrated Confucian Yan Zhenqing 顏真卿 (709–785), who had served as the ritual commissioner (liyishi 禮儀使) since Daizong’s early reign, called for a return to a seven-temple configuration in 779.88 While the inclusion of Daizong’s tablet in the temple would definitely result in the removal of one distant ancestor’s tablet, Yan suggested the court revise the whole generational sequence in the temple and adopt a correct zhaomu order. Specifically, Yan Zhenqing quoted Han precedents to elaborate his suggestions.89 His criticism toward the Eastern Han practice of ancestral rituals targeted the Tang temple setting, especially Xuanzong’s scheme of (Zhongzong and Ruizong) in the opposite chamber of Gaozong’s. Xin Tangshu, 200.5696. 84. Jiu Tangshu, 25.953. 85. Jiu Tangshu, 25.953. 86. Jiu Tangshu, 25.954. 87. David McMullen, “The Death Rites of Tang Daizong,” in State and Court Ritual in China, ed. Joseph P. McDermott (Cambridge: Cambridge University Press, 1999), 150–96. 88. Jiu Tangshu, 25.954–55. 89. Yan Zhenqing, “Lun Yuanhuangdi tiaoqian zhuang” 論元皇帝祧遷狀. This memorial was submitted in the tenth month of the last year of the Dali 大歷 era (766–779), right after Daizong’s funeral. Yan Lugong wenji 顏 魯公文集, in Sibu beiyao 四部備要 (Shanghai: Zhonghua shuju, 1936), 228:2.33–34.

34

Empowered by Ancestors

nine ancestral chambers. Yan stated clearly that the posthumous titles of zu (祖, literally, Founding Ancestor) and zong (宗, literally, Exemplar Ancestor) should be reserved for those praiseworthy ancestors.90 According to Yan, although Li Bing had been given the temple title of the “Great Ancestor of Tang” (Tang Shizu 唐世祖), he was not that “great” as the title suggested, owing to his loose connection to the establishment of the Tang empire. At the end of his memorial, Yan questioned the abuse of the filial-piety narrative in conceptualizing temple rituals. In his words: “If a dynasty continues for hundreds of generations, should it worship the hundredth generations of its imperial ancestors to illustrate the virtue of filial piety” 假令傳祚 百代，豈可上崇百代以為孝乎?91 Eventually, Yan proposed a temple arrangement consisting of six regular zhaomu chambers and three permanent tiao chambers, with the latter housing the tablets of three meritorious ancestors—Li Hu as Tang Taizu, Li Yuan as Tang Gaozu, and Li Shimin as Tang Taizong.92 Like Yin Zhizhang, Yan Zhenqing emphasized merits in conceptualizing the Imperial Temple. However, Yan Zhenqing was not totally opposed to the concept of filial piety. It was the overstatement of filial piety in the temple controversies that perplexed him. Although Yan stated that Li Hu was the Primal Ancestor in usual temple sacrifices, he suggested that Li Hu’s spirit tablet be removed from the Primal Ancestor position and arranged according to zhaomu in the xia sacrifices, where all ancestral tablets were placed in the Primal Ancestor chamber once every three years. Instead, Li Xi, as Li Hu’s grandfather, should be placed at the center of the Primal Ancestor chamber to demonstrate the virtue of filial piety in the xia sacrifice.93 According to Yan, although Li Hu as the Founding Ancestor received the mandate from Heaven and was in conjunction with offerings to Heaven in suburban Altar sacrifices, his tablet should be temporarily removed from the Primal Ancestor position in the xia sacrifices, since xia involved ancestors who were more ritually superior. By temporarily giving the Primal Ancestor position to his grandfather Li Xi in the xia sacrifices, Li Hu’s ancestral spirit would set an example of filial submission to ancestors, because he demoted his own ritual status to a less privileged position.94 Yan Zhenqing’s attempt to balance the two factors of merits and filial piety in the Imperial Temple was later echoed by the great Confucian scholar Han Yu 韓愈 (768–824). In an 802 memorial, Han Yu suggested the court officially acknowledge Li Xi’s supreme ritual status in the xia sacrifices by placing Li Xi’s tablet in the Primal 90. Yan, “Lun Yuanhuangdi tiaoqian zhuang,” 33. 91. Yan, “Lun Yuanhuangdi tiaoqian zhuang,” 33. 92. The concrete arrangement of the Tang Imperial Temple after Tang Daizong’s death was recorded in an official ritual manual called the Tai Tang Yuanling yizhu 大唐元陵儀注. For a modern reconstruction of the Tang temple arrangement in the 780s, see Daito Motomisasagi Gi Chushinshaku 大唐元陵儀注新釈, Kaneko Shūichi ed. (Tokyo: Kyuko Shoin, 2013), 344, Figure 12. 93. Yan, “Miaoxiang yi” 廟享議, in Yan Lugong wenji, 2.34–35. 94. In Yan’s own words, this practice was to “diminish oneself to fulfill the intent of filial piety and to make due offerings to the ancestors in a respectful way” 屈已伸孝，敬奉祖宗. Yan, “Miaoxiang yi,” 2.34.

Pre-Song Interpretations of the Imperial Temple 35

Ancestor chamber, facing east.95 Correspondingly, Li Hu and other Tang ancestors should be arranged in different zhao and mu chambers. Han explained how seniority and filial piety were both factored in the xia sacrifices. According to Han, Li Hu’s ritual status in xia should be diminished out of his respect for his grandfather Li Xi, despite the former’s superiority in terms of merits.96 Compared to Yan Zhenjing, Han Yu was more affirmative in upholding the principle of filial piety. As Zhu Xi argued, Han was not trying to deny the meritbased reasoning in his argument. Han actually admitted that it was necessary to preserve Li Hu’s ritual status as the Primal Ancestor. Nonetheless, Han provided two explanations that helped to reconcile the tension between filial piety and merits in practicing temple rituals. First, Han claimed that different ancestral chambers in the temple were mutually independent in terms of ritual status.97 In other words, Taizu’s ritual status would not be compromised within his own chamber wherein only his spirit tablet was placed. Technically speaking, Taizu’s spirit monopolized the ritual sanctuary of his chamber and acted as the sole authority of the whole chamber. If the tablets of other Tang ancestors were moved into Taizu’s chamber, they could not rival Taizu’s ritual authority within Taizu’s own ritual space. Other chambers were subject to the same rule. Yet, other chambers seldom stored other tablets as the Taizu chamber would do in the xia sacrifices. Second, Han Yu considered the xia sacrifices an exception. Since xia sacrifices were held only once every three years, there were only a few occasions where the Primal Ancestor’s ritual status would be apparently diminished. Additionally, considering the Tang case, although Li Hu’s ritual status would be diminished in the xia sacrifices, his grandfather’s and his father’s ritual statuses were magnified by Li Hu’s personal achievements. In this light, Li Hu elevated his ancestors’ ritual status to a much higher level, thereby further confirming his supremacy as the Primal Ancestor. Other late Tang scholars shared the same interest in affirming the Primal Ancestor of the Tang Imperial Temple, but with different approaches. In 781, Chen Jing 陳京 memorialized to the court to defend Li Hu’s Primal Ancestor position.98 Furthermore, Chen Jing suggested safeguarding Li Hu’s Primal Ancestor status in xia and di sacrifices, when other ancestors were involved.99 Tozaki Tetsuhiko correctly pointed out that Chen Jing’s suggestion represented the growing conception of the Tang Imperial Temple as a “state temple.” Therefore, Chen Jing, together with some other officials, privileged Li Hu’s ritual status in xia and di sacrifices because Li Hu as an abstract “feudal lord” received the Mandate from Heaven, upon which his

95. Han Yu, “Di xia yi” 禘祫議, Han changli ji 韓昌黎集 (Beijing: Shangwu yinshuguan, 1958), 14.32. 96. Han, “Di xia yi,” 14.31–32. 97. See Zhu Xi’s annotation to Han Yu’s “Di xia yi,” which was attached as a footnote in Han’s “Di xia yi,” 14.32. 98. Xin Tangshu, 200.5712. 99. Xin Tangshu, 200.5713.

36

Empowered by Ancestors

sons established a “state.”100 Clearly, Chen Jing’s suggestion followed the merit-based approach used by officials like Yin Zhizhang, Yan Zhenqing, and Han Yu. As more deceased Tang emperors were worshipped in the Imperial Temple when time lapsed, the conflict between merits and filial piety became more intense. The merit-based approach gained its concluding recognition in 803, when the Tang court decided to confirm Li Hu’s Primal Ancestor position in most temple rituals, including the xia sacrifices.101 The ritual controversies concerning the Tang Imperial Temple seemed resolved. However, the issues brought out by temple controversies continued to shape ritual debates in the following several centuries, when new political and intellectual factors became prominent.

Concluding Remarks The Imperial Temple as an essential part of imperial ancestral rituals underwent a series of developments, from Zhou to Tang. Given the crucial role played by the Imperial Temple in the ritual discourse of sacrificial and funeral rituals, two relevant interpretations can be distinguished. The first conceptualized the Imperial Temple as a place to illustrate the personal merits of imperial ancestors. In this light, only those ancestors who contributed remarkably to the foundation of the dynasties were qualified to be permanently stored in the temple in either the name of Primal Ancestors or other designations such as Founding Ancestors and Exemplar Ancestors. The memorials of the Han official Wen Xuancheng and the Tang officials Yin Zhizhang, Yan Zhenqing, and Han Yu exemplified how this merit-based approach developed from the first to the ninth century. In particular, the Han Confucian Zheng Xuan’s theory of the tiao temples offered textual evidence for this approach, as Zheng argued that the tiao temples were set up to store imperial ancestors with unparalleled merits, such as King Wen and King Wu of Zhou. The second interpretation considered the Imperial Temple as a visualization of the Confucian virtue of filial piety. In this sense, the genealogical sequence of imperial ancestors mattered most as it reflected the strict order of seniority in filial practices. Han Confucians, especially the Gongyang scholars, interpreted the arrangement of temple ancestors as the manifestation of seniority and hence the very concept of filial piety. As Cai Yong mentioned, the most contentious issue of the Han Imperial Temple was that it had never succeeded in rectifying a correct sequence of the spirit tablets stored within. Since the naissance of temple rituals, the two interpretations on the Imperial Temple had intertwined with each other in related discussions. In the Tang dynasty, 100. According to Tozaki Tetsuhiko, the endeavor to designate Li Hu rather than Li Yuan as the Primal Ancestor since the 780s might also aim to soften the influence of the Great Rebellion by avoiding possible interpretation of rebellion as an effective way to win the Mandate from Heaven—a way through which Li Yuan established the Tang empire. Tozaki, “Tōdai niokeru teikyū ronsōto sono igi,” 82–96. 101. Xin Tangshu, 200.5716.

Pre-Song Interpretations of the Imperial Temple 37

scholar-officials like Yan Zhenqing and Han Yu embraced a compromising plan in addressing the Primal Ancestor issue. On the one hand, the meritorious ancestor Li Hu had been designated as the Primal Ancestor in the late eighth century. On the other hand, distant ancestors were honored in state sacrifices to symbolize the virtue of filial piety, when all ancestors were arranged together in the main chamber of the Imperial Temple. As the late Tang period witnessed a revival of Confucian values and relevant discourses, court officials and local elites emphasized more on the potency of imperial rituals in reviving Confucian values.102 The Northern Song dynasty continued the late Tang endeavor and witnessed a blossom of ritual debates and learning about imperial rituals.103 It was under this intellectual transformation that the Imperial Temple aroused new controversies.

102. For a general intellectual history about the late Tang revival of Confucian values among educated elites, see Bol, This Culture of Ours, 108–47. 103. Meyer, Ritendiskussionen am Hof der nördlichen Song–Dynastie (1034–1093), 103–297; Wu Wanju 吳萬居, Songdai sanlixue yanjiu 宋代三禮學研究 (Taibei: Guoli bianyiguan, 1999), 460–507.

2 Northern Song Conceptions of the Imperial Temple and Temple Sacrifices

The early Northern Song was profoundly influenced by the political and intellectual traditions of the Five Dynasties, especially in terms of cultural practices. Owing to socio-political problems, the warlord monarchy of the Five Dynasties had not established durable cultural norms to cope with its military governorship. A stable society and a strong central government as the two prerequisites for the revival of cultural norms were difficult to fulfill when China was split into various powers before the tenth century. In contrast to the chaos of the Five Dynasties, the Northern Song experienced a rise of literary culture and a passionate pursuit of ancestral rituals.1 Since the end of the tenth century, the Song court and its rulers had gradually devoted their attention to imperial ancestral rituals. The Song scholar Li Zhi 李廌 (1059–1109) claimed that the essence of imperial sacrificial rituals is “to provide the Son of Heaven a chance to gain appreciation from his subjects of all-under-Heaven and by doing so he can well serve his imperial ancestors” 天子得四表之歡心以事其先王者是也.2 Along with an increasing interest in ancestral rituals and related Confucian norms, the issue of the Imperial Temple came into the sight of the Song court.3 In the first half of the eleventh century, the arrangement of the Imperial Temple became a focus of the Northern Song ritual officials and ritual institutions.

1. By arguing that the power structure of the Five Dynasties (883–947) had a great impact on the political configuration of the Northern Song dynasty, Wang Gungwu introduced a dynamic perspective to conceptualize the early Song history. Wang Gungwu, The Structure of Power in North China during the Five Dynasties (Kuala Lumpur: University of Malaya Press, 1963), 2–6. 2. Li Zhi, Shiyou tanji 師友談記 (Beijing: Zhonghua shuju, 2002), 40. 3. For the increasing interests in ancestral sacrifices in the Northern Song, see Patricia Ebrey, Confucianism and Family Rituals in Imperial China, 53–56; Joseph McDermott, The Making of a New Rural Order in South China, 100–102; Cheung Hiu Yu, “Inventing a New Tradition: The Revival of the Discourses of Family Shrines in the Northern Song,” Journal of Song-Yuan Studies 47 (hereafter JSYS, 2019): 85–136; Azuma Jūji 吾妻重二, “Sōdai no kabyō to sosen saishi” 宋代の家廟と祖先祭祀, in Chūgoku no reisei to reigaku 中國の禮制と禮學, ed. Kominami Ichirō 小南一郎 (Kyoto: Hōyū shoten, 2001), 505–75.

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 39

Song Ritual Institutions and Ritual Officials Recognizing the significance of imperial ancestral rituals, the Song court established various institutions and posts to handle ritual affairs from the late tenth century to the first half of the eleventh century. The Commission of Ritual Affairs (taichang liyuan 太常禮院) was the most significant institution that took responsibility for the rectification and standardization of court rituals.4 Originally established as an ad hoc institution in the late eighth century, the Commission of Ritual Affairs became a standing ritual institution after the ninth century.5 When the Northern Song dynasty was established in 960, the Song court tended to follow the late Tang precedent in maintaining the Commission of Ritual Affairs as a standing institution. The Commission of Ritual Affairs was nominally under the supervision of the Court of Imperial Sacrifices, but it took charge of most ritual affairs in practice.6 The commission remained as a key institution in rectifying imperial rituals prior to the bureaucratic reforms in the late eleventh century, when the Song court finally affirmed the Court of Imperial Sacrifices as the only ritual institution responsible for rectifying imperial rituals.7 Regarding the Imperial Temple, the Commission of Ritual Affairs often worked with the Court of the Imperial Clan to codify regulations on the setting and rituals of the temple before the bureaucratic reforms. In the early Northern Song, the Court of the Imperial Clan was usually headed by imperial clansmen.8 In 1036, due to a rapid growth in the clansmen population, Emperor Renzong 宋仁宗 (r. 1022–1063) established a new institution, the Great Office of Clan Affairs (da zongzheng si 大 宗正司), to assist the Court of the Imperial Clan.9 The great office dealt with the education of imperial clansmen and provided allowances for poor clan members.10

4. Tuotuo 脫脫 (1314–1355) et al., Song shi 宋史 (hereafter SS) (Beijing: Zhonghua shuju, 1985), 12:164.3882–84. 5. For a detailed analysis of the institutional history of the Commission of Ritual Affairs, see Cheung Hiu Yu, “Zhuanda de xianzhi: Tang zhi Beisong taichangliyuan yangekao” 專達的閒職—唐至北宋太常禮院沿革考,” in Zhongguo gudai zhengzhizhidu yu lishidili: Yangengwang xiansheng bailing jinian lunwenji 中國古代政治制 度與歷史地理─嚴耕望先生百齡紀念論文集 (Jinan: Qilu shushe, 2020), 177–204. 6. Cheung, “Zhuanda de xianzhi,” 187–202. 7. For a succinct analysis of the bureaucratic reforms in the late eleventh century, see Paul Jakov Smith, “Shentsung’s Reign and the New Policies of Wang An-shih, 1067–1085,” in The Cambridge History of China, Vol. 5, Part I: The Sung Dynasty and its Precursors, 907–1279, ed. Denis Twitchett and Paul Jakov Smith (Cambridge, UK: Cambridge University Press, 2009), 457–64. 8. Ge Shengzhong 葛勝仲 (1059–1131), “Zongzhengsi shaoqing biji” 宗正寺少卿壁記, in Danyang ji 丹陽集, Wenyuange siku quanshu 文淵閣四庫全書 (hereafter SKQS), comp. Ji Yun 紀昀 (1724–1805), et al. (Shanghai: Shanghai guji chubanshe, 1987), 1127:8.3. For a general record of the composition of the Court of the Imperial Clan, see SHY, Zhiguan 3:20.1; SS, 12:164.3887. Under most circumstances, the Song court would select an individual from the imperial clansmen to serve as the Head of the Court of the Imperial Clan, especially after Zhenzong’s reign. SHY, Zhiguan, 3:20.2–4. Also see John W. Chaffee, Branches of Heaven: A History of the Imperial Clan of Sung China (Cambridge, MA: Harvard University Asia Center, 1999), 21–22. 9. SHY, Zhiguan, 3:20.16; SS, 12:164.3888–90. Chaffee, Branches of Heaven, 40–44, especially 42, 84–86. 10. Song Xi 宋晞, “Songdai de zongxue” 宋代的宗學, Aoyama Hakushi koki kinen Sōdai shi ronsō 青山博士古稀 紀念宋代史論叢 (Tōkyō: Seishin Shobō, 1974), 161–81.

40

Empowered by Ancestors

It also superintended the practices of ancestral sacrifice and other rituals in the Imperial Temple.11 Considering the pivotal role played by the Commission of Ritual Affairs in making court rituals, it is not difficult to imagine how ritual officials from the commission—usually being referred to as “erudites” in Song sources (liyuan taichang boshi 禮院太常博士)—participated actively in related Song ritual debates. Notably, there are two types of taichang “erudites”: one refers to the “titular offices” (jiluguan 寄祿官) in Song bureaucracy; the other refers to the functional positions of the Song commission system (chaiqian 差遣).12 The former governs the salaries and ritual ranking of officials. A taichang erudite in this sense represents a titular title that most civil officials would have been granted in their early careers. The title taichang here means the abbreviation of the titular office of the Court of Imperial Sacrifices 太常寺博士, an office with a salary rank (jie 階) of twenty-three.13 During the Northern Song, the taichang title related to the Court of Imperial Sacrifices indicated the salary of junior officials who passed through civil service examinations and served in the central government (you chushen 有出身). On the other hand, the taichang erudites in the Commission of Ritual Affairs were functional positions. It is the second type of taichang erudites who actively participated in ritual debates concerning the Imperial Temple. The taichang erudites in the Commission of Ritual Affairs comprised the main group of ritual officials who handled concrete matters, mostly the compilation and revisions of ritual codes and manuals. The head of the commission, who was usually designated as the prefect or the supervisor of the commission (zhi liyuan 知 禮院, pan liyuan 判禮院) in Song bureaucracy, was responsible for collecting opinions from the erudites for submission to the court. The fact that many celebrated scholar-officials in the Northern Song had served as the prefects and supervisors of the commission demonstrated the importance of these positions in terms of functionality. Prefects and supervisors of the Court of Imperial Sacrifices were rarely associated with the examination of imperial rituals, given their attributes as “titular offices.” Likewise, the court gentleman of temple fasting (taimiao zhailang 太廟齋 郎) and the head of temple chambers (taimiao shizhang 太廟室長) were also titular titles bestowed to the junior relatives of high-ranking officials through the recruitment channels of privileges (yin 蔭) and controlled sponsorship (ren 任 or baoren 保任).14 On some occasions, the court gentleman of temple fasting and the head of 11. Chaffee, Branches of Heaven, 103. 12. For the practice of the dual ranking system in Song personnel management, see Charles Hartman, “Sung Government and Politics,” in The Cambridge History of China, Vol. 5, Part II: Sung China, 907–1279, ed. John W. Chaffee and Denis Twitchett (Cambridge, UK: Cambridge University Press, 2015), 49–80, especially 59–62, 66–68. For an explanation of the entire mechanism of Song bureaucracy, see Winston Lo, An Introduction to the Civil Service of Sung China: With Emphasis on Its Personnel Administration (Honolulu: University of Hawaii Press, 1987), 58–70, 115–21. 13. Gong Yanming 龔延明, “Yuanfeng qianhou liang Song wenguan jilu guanjie duizhao biao” 元豐前後兩宋文 官寄祿官階對照表, in Songdai guanzhi cidian 宋代官制辭典 (Beijing: Zhonghua shuju, 2017), 759. 14. A number of examples can be found in Song official archives to demonstrate the bestowal of these two titles

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 41

temple chambers served as ritual practitioners in some temple sacrifices.15 Titular officials from the Commission of Ritual Affairs and the Court of Imperial Sacrifices, such as the ritual duty officials (lizhi 禮直), the court gentlemen for ceremonials (fengli lang 奉禮郎), and the great supplicators (taizhu 太祝), also helped to manage the performance of temple sacrifices and related ceremonies.16 However, as executive officers they had no right to participate in ritual discussions about the Imperial Temple. Scholar-officials from the Two Drafting Groups (liangzhi 兩制, Hanling academicians 翰林學士 and edict drafters of the Secretariat 中書舍人) also played a key role in Northern Song ritual debates. The conventional Song practice of ad hoc collective advisory meeting (jiyi 集議) promoted communication between ritual officials of the commission and officials of the Two Drafting Groups. In most collective advisory meetings, opinions of involved officials were collected and conveyed to the emperor as a package.17 Generally, officials from the Commission of Ritual Affairs and the Two Drafting Groups constituted the center of ritual discussions throughout the Northern Song.

Basic Setting of the Song Imperial Temple and Temple Sacrifices Song scholars and officials generally considered the Imperial Temple as the preserve of the royal family.18 The official history of the Song Imperial Temple could be traced back to 960, when the dynastic founder, Emperor Taizu 太祖 (Zhao Kuangyin 趙 as yin privilege. See Li Tao 李燾 (1114–1183), Xu zizhi tongjian changbian 續資治通鑑長編 (hereafter XCB; Beijing: Zhonghua shuju, 1986), 23.526, 44.930, 83.1893, 104.2424, 109.2549, 131.3100, 187.4509, 273.6697, 391.9508. For discussions about the Song yin privilege, see Umehara Kaoru 梅原郁, “Sōdai no on in seido” 宋 代の恩蔭制度, in Sōdai kanryō seido kenkyū 宋代官僚制度硏究 (Kyoto: Dōbōsha, 1985), 423–500; also see Winston Lo, An Introduction to the Civil Service of Sung China, 103–9. For the sponsorship system in personnel administration, see E. A. Kracke, Civil Service in Early Sung China: 960–1067 (Cambridge, MA: Harvard University Press, 1953), 102–89. 15. SHY, Zhiguan, 3:22.19–20, Li, 1:15.4; XCB, 83.1893. For a detailed summary of their duties, see Gong Yanming, Songdai guanzhi cidian, 303–4. 16. SHY, Zhiguan, 3:22.17. In terms of bureaucratic establishment, both the Commission of Ritual Affairs and the Court of Imperial Sacrifices would recruit some ritual duty officials for clerical works. The court gentlemen for Ceremonials and the great supplicators were titular titles under the supervision of the Court of Imperial Sacrifices, institutionally. These officials usually held other tasks in the commission system. However, Song official archives also records these officials’ participation in court rituals. See, for example, SHY, Li, 1:2.23, 1:2.26, 1:17.22; 1:17.61–63 (temple and altar sacrifices); SHY, Li, 1:4.5–7 (celestial sacrifices). The audience attendants (xuanzan sheren 宣贊舍人) from the Commissioner for Audience Ceremonies (hemen si 閤門司) also helped the performance of temple rites, especially as guides of civil and military officials who attended these rites. SHY, Zhiguan, 4:34.1–4. 17. Generally, Song emperors conveyed collective advisory meetings to discuss significant policies, especially ritual issues. For a brief discussion about the Song jiyi system, see Hirata Shigeki 平田茂樹, Songdai zhengzhi jiegou yanjiu 宋代政治結構研究 (Shanghai: Shanghai guji chubanshe, 2010), 12–13, 161–67; Cheung Hiu Yu, “Zhuanda de xianzhi,” 194–96. 18. When a Song clansman attempted to construct a family temple for his relatives in the 1050s, Ouyang Xiu claimed that only the Son of Heaven could construct an Imperial Temple. It was inappropriate for princes to build ancestral temples in their own fiefs. Fan Zhen 范鎮 (1007–1087), Dongzhai jishi 東齋記事 (Beijing: Zhonghua shuju, 1980), 58.

42

Empowered by Ancestors

匡胤, 927–976, r.  960–976) ordered officials to construct an Imperial Temple for the Zhao 趙 clan. Zhang Zhao 張昭 and Ren Che 任徹, two ritual officials who had previously served in the ritual institutions of the Five Dynasties, suggested the court to construct five temples in the form of chambers.19 Zhang and Ren argued that an appropriate setting of an Imperial Temple complex should include two zhao temples, two mu temples, and a Primal Ancestor temple—the last was reserved for a Zhao ancestor who had made great contributions to the foundation of the Song dynasty.20 Given the fact that Taizu himself had established the Song, Zhang and Ren advocated that the Primal Ancestor chamber should be left vacant for a period, until Taizu was deceased and his spirit tablet could be moved into the temple.21 The Song court adopted this plan and posthumously bestowed imperial titles upon Taizu’s ancestors up to the fourth generation: his great-great-grandfather Zhao Tiao 趙朓 (828–874) was given the title Xizu 僖祖 (lit. “the ancestor of joyfulness”), his great-grandfather Zhao Ting 趙珽 (851–928) was given the title Shunzu 順祖 (“the ancestor of obedience”), his grandfather Zhao Jing 趙敬 (872–933) was given the title Yizu 翼祖 (“the ancestor of guardianship”), and his father Zhao Hongyin 趙 洪殷 (899–956) was given the title Xuanzu 宣祖 (“the ancestor of propagation”).22 Zhang Zhao’s and Ren Che’s plan set the blueprint for the arrangement of the Song Imperial Temple. Essentially, their plan continued the conventional arrangement of the Imperial Temples practiced by the previous Five Dynasties.23 Zhang and Ren legitimized their arrangement by connecting it to the idea of ancient sagekings, such as Yao 堯, Shun 舜, and Yu 禹.24 Based on their own imagination of ancient ritual practices, they claimed on behalf of the court that their design of the Song Imperial Temple “perfectly accorded with the ritual texts” (shenhe liwen 深合 禮文) in both theoretical and practical ways.25 In the absence of archeological evidence, it is impossible to visualize the concrete arrangement of the five-temple setting suggested by Zhang Zhao and Ren Che. However, it seems that early Song scholars like Zhang and Ren had scant knowledge about the Imperial Temple’s architecture. Nie Chongyi, a taichang erudite who annotated an official illustration of ritual utensils and architecture, failed to provide any illustration for the Imperial Temple.26 While Nie had drawn several illustrations of other imperial ritual architectures, including the mingtang, the yuanqiu

19. SHY, Li 1:15.22; XCB, 1.8. 20. XCB, 1.8. 21. XCB, 1.8. 22. XCB, 1.10. For a brief introduction of these Song ancestors, see Chaffee, Branches of Heaven, 21–22. 23. Zeng Gong 曾鞏 (1019–1083) summarized Zheng’s and Ren’s blueprint of the Song Imperial Temple in his policy essay. Zeng Gong, “Benchao zhengyaoce: zongmiao” 本朝政要策: 宗廟, in Yuanfeng leigao 元豐類藁, Sibu congkan 四部叢刊 (hereafter SBCK) (Shanghai: Shangwu yinshuguan, 1919), 1680–89:49.6. 24. SHY, Li 1:15.22. 25. SHY, Li 1:15.22. The ritual texts mentioned in Zhang Zhao’s and Ren Che’s memorial seem to be Zheng Xuan’s comments on the Wangzhi chapter. 26. Nie Chongyi, Sanlitu jizhu, SKQS, 129: tiyao 提要.1a–3a.

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 43

圓丘 (literally, Round Altar), and the fangqiu 方丘 (literally, Square Altar), he missed the illustration of the Imperial Temple in his annotated illustration Sanlitu jizhu.27 A possible explanation to Nie Chongyi’s neglect of temple illustration is that he regarded the temple configuration as similar to that of mingtang. In his commentary next to the mingtang illustration, Nie stated that “the setting of the Imperial Temple resembled the one of mingtang” 宗廟制如明堂.28 He cited a sentence from the Tang scholar Jia Gongyan’s 賈公彦 commentary on the Rituals of Zhou to argue that both the Imperial Temple and mingtang were composed of five chambers, twelve halls, and four entrances.29 Nie’s mingtang illustration dovetails with the official temple setting suggested by Zhang Zhao and Ren Che in a sense that it portrays a structure of five chambers. Certainly, the mingtang illustration in Nie’s Sanlitu jizhu (see Illustration 2.1 on p. 44) does not reflect the accurate dimension of the Song Imperial Temple. It only demonstrates how ritual officials visualized imperial ritual architectures in the early Song. Since the establishment of the Song Imperial Temple in 960, the court had begun to sort out related liturgical details from the ritual precedents of previous dynasties. In 971, it finished compiling the first official ritual code of Song, namely the Kaibao tongli 開寶通禮 (Ritual Code of the Kaibao Era). This ritual code encompassed many ritual precedents of the Five Dynasties.30 Because the Kaibao tongli has been lost, it is impossible to know the details about the Song Imperial Temple. A clear fact is that Song ritual officials were frustrated about temple rituals. Christian Meyer has already discussed how the frustration about imperial rituals weighed in on court debates over suburban Altar sacrifices and the composition of ceremonial music (yayue 雅樂).31 Likewise, ritual practices in relation to the Imperial Temple served as a source of frustration for the Song ritual officials. The Song Imperial Temple was established, yet temple rituals were often performed in informal ways. For example, despite the clear recognition of laxiang as one of the five seasonal sacrifices since the mid-Tang, the early Song court did not perform laxiang regularly, considering that the ritual itself might be an additional burden to the emperors. Instead of laxiang, a winter sacrifice to state deities was performed in the southern suburban Altar, namely the labaishen 蠟百神.32 The Song court only rectified the system of seasonal 27. For the illustrations and Nie Chongyi’s comments, see Sanlitu jizhu, 129:4.2a–3a (mingtang of the Zhou dynasty); 4.11a–14a (yuanqiu and fangqiu); 4.24a–b (mingtang of the Qin dynasty). Mingtang referred to a divine architecture that was built for offering a sacrifice to the Heavens. John Henderson, The Development and Decline of Chinese Cosmology, 75–85. For a thorough study of the mingtang, especially the evolution of its architectural structure, see Hwang Ming-chorng, “Ming-tang: Cosmology, Political Order and Monuments in Early China,” 7–10, 27–118. Yuanqiu and fangqiu were generally regarded as the ritual locations for offering suburban Altar sacrifices, respectively, to the Heaven and to the Earth. 28. Nie, Sanlitu jizhu, 129.4.2a. 29. Nie, Sanlitu jizhu, 129.4.3a. 30. For a useful introduction of the history of the Kaibao tongli, see Lou Jin 樓勁, “Guanyu Kaibao tongli ruogan wenti de kaocha” 關於開寶通禮若干問題的考察, Zhongguo Shehui Kexueyuan Lishi Yanjiusuo Xuekan 中國 社會科學院歷史研究所學刊, no. 4 (Beijing: Shangwu yinshuguan, 2007): 420–30. 31. Meyer, Ritendiskussionen am Hof der nördlichen Song–Dynastie (1034–1093), 163–253. 32. SHY, Li 1:19.23.

Empowered by Ancestors

44 Illustration 2.1: Nie Chongyi’s Mingtang Illustration

Source: Nie Chongyi, Sanlitu jizhu 三禮圖集註, SKQS, 129:4.3a.

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 45

sacrifices in the late eleventh century, when Emperor Shenzong promoted full-scale ritual reforms on temple rituals. The Song court’s emphasis on subsidiary imperial ancestral buildings (yuanmiao 原廟, literally, “Origin Shrines”) in the first half of the eleventh century further upset the regular performance of temple rituals. Subsidiary imperial ancestral buildings were originally designed for special ritual campaigns, such as the Temple of Spectacular Numina (jingling gong 景靈宮) that was built in 1016 to cope with the ritual of receiving Heavenly Text in the name of the Holy Ancestor (shengzu 聖 祖) of the Zhao clan, a Daoist deity named Zhao Xuanlang 趙玄朗.33 This Daoistinfluenced temple soon became a ritual site for most imperial ancestral worship; since the temple’s settlement, several emperors made offerings to the portraits and statues of imperial ancestors therein.34 Although some ritual officials criticized the Daoist and Buddhist practices of ancestral worship within subsidiary imperial ancestral buildings as a deviation from Confucian ritual norms, these buildings prospered during the Northern Song.35 The emperors found these buildings more attractive than the Imperial Temple because these buildings allowed the practicing of religious ceremonies that prayed for ancestors’ better afterlives. The court soon acknowledged the legitimacy of subsidiary imperial ancestral buildings and related ceremonies. In 1033, under Emperor Renzong’s reign, the court even agreed to build an independent subsidiary ancestral building for empresses in the ancient name of bigong 閟宮.36 Under usual circumstances, the ceremonies that were held in bigong followed temple rituals. Some of these ceremonies could still be found in extant Song ritual codes, especially those that were compiled during Shenzong’s ritual reforms.37 The conflict between Imperial Temple rituals and Daoist and Buddhist ceremonies in subsidiary imperial ancestral buildings continued. Advocates of temple rituals usually found themselves at a disadvantage, given complexity and uncertainty in temple rituals in comparison with Daoist and Buddhist ceremonies. According to official ritual codes, Song temple rites and ceremonies included five categories, respectively the five seasonal sacrifices, the xia sacrifices, regular offerings to the imperial ancestors in the form of fresh and seasonal food (jianxin 薦新), monthly offerings (shuoji 朔祭), and the morning offerings in the temple (chaoxiang 朝享)

33. XCB, 79.1802. For a discussion on the Song practice of subsidiary imperial ancestral buildings, especially the Temple of Spectacular Numina, see Yamauchi Kōichi, “Hokusō jidai no shingyoden to keireikyu” 北宋時代 の神御殿と景靈宮, Tōhō gaku 70 (1985): 46–60; Peng Meiling 彭美玲, “Liang Song huangjia yuanmiao jiqi lisu yiyi qiantan” 兩宋皇家原廟及其禮俗意義淺探, Chengda zhongwen xuebao 成大中文學報, 52 (2016.3): 67–114, especially 82–83. 34. Ebrey, “Portrait Sculptures in Imperial Ancestral Rites in Song China,” 52–61; 65–67. 35. Scholars and officials with a deep Confucian imprint also criticized the use of portraits and other images of ancestors in subsidiary imperial ancestral buildings. See Cheung, “Inventing a New Tradition,” 92–96. 36. XCB, 112.2620. 37. For ceremonies that were performed in Song subsidiary ancestral buildings, see Ebrey, “Portrait Sculptures in Imperial Ancestral Rites in Song China,” 67.

46

Empowered by Ancestors

prior to the suburban Altar sacrifices.38 The extant Song ritual codes have preserved the ritual procedures, sacrificial utensils, animal and crop offerings, and the spatial arrangement of ritual sites in temple rituals in the form of liturgical manuals (yizhu 儀注). A typical liturgical manual that instructs the seasonal sacrifices conducted by the emperors at the Imperial Temple contains eight sections, as listed below: 1) Deciding the dates of sacrifices, usually by the Commission of Ritual Affairs and the Astrological Service (taishi ju 太史局) (shiri 時日); 2) A preparatory stage of fasting, usually held ten days before the formal sacrifice at the temple (zhaijie 齋戒); 3) The arrangement of sacrificial utensils, musical instruments, and the positioning of ritual practitioners and officers in the seasonal sacrifices (chenshe 陳設); 4) The route and arrangement of the emperor’s carriage from the main palace (in the Northern Song, it would usually be the Palace of Great Celebration 大慶殿) to the Imperial Temple; 5) The examination of sacrificial utensils, musical instruments, and animal and crop offerings before the day of sacrifice (shengshengqi 省牲器); 6) A preliminary ceremony of pouring the wine on the ground of the temple’s main hall, performed in the early morning of the day of sacrifice (chenguan 晨祼); 7) The formal offering of cooked food and ritual wine (kuishi 饋食); and 8) The return of the emperor’s carriage to the palace (chejia huannei 車駕還 內).39 The above list only covers major procedures and issues involved in the seasonal sacrifices held in the Imperial Temple. Considering the complexity of temple rituals, Song emperors who were exhausted by these rituals decided to suspend or even abandon some of them occasionally. A dramatic scenario that occurred when Song Taizu entered the Imperial Temple reflects the emperors’ general attitude toward the temple rituals. After Taizu saw the sacrificial utensils in the newly established Imperial Temple, he asked about their uses. When his courtiers answered that they were ritual utensils used in temple sacrifices, the emperor laughed out: “How could it be possible that my ancestors know these things” 吾祖宗寧識此?40 Instantly, Taizu ordered the courtiers to whip the sacrificial utensils away and asked for normal food offerings to his ancestors. Only after a second consideration did the emperor order the recovery of sacrificial utensils in the temple. 38. Among all Song ritual codes and official archives, the Zhenghe wuli xinyi (New Forms of the Five Categories of Rites of the Zhenghe Period) that was compiled in Emperor Huizong’s reign contained the most comprehensive record of temple rites and ceremonies. Zheng Juzhong 鄭居中 (1059–1123) et al., Zhenghe wuli xinyi, SKQS, 647:97–105. I will discuss more about this work in Chapter 5. 39. Zheng Juzhong, Zhenghe wuli xinyi, 102.1a–104.9b. 40. XCB, 9.211.

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 47

Song Taizu’s response to the sacrificial utensils in the temple was not a random incident. The Qing scholar Qin Huitian’s 秦蕙田 (1702–1764) criticism of the Song neglect of seasonal sacrifices revealed how the complexity of temple rituals hindered emperors from practicing ancestral worship properly.41 When an investigating censor (jiancha yushi 監察御史) criticized Renzong for his neglect of seasonal sacrifices in the temple, Renzong defended himself by saying that he “made offerings to the images of his three imperial ancestors within the palace day and night” 朕朝 夕奉三聖御容於禁中.42 Renzong’s preference for the more convenient image sacrifices in the palace demonstrated how temple rituals were underplayed in practice. In contrast to early Song emperors’ reluntance to practice temple rituals, discussions and debates about temple rituals among ritual officials flourished in the meantime. The growing interest among ritual officials could be attributed to their endeavors to rejuvenate temple rituals in accordance with Confucian norms. Other political and intellectual factors also weighed in. Among them the fraternal succession between the two founding emperors of the Song dynasty and the return of the discourse of filial piety in the 1040s served as two critical factors in shaping early Song discussions on the Imperial Temple.

Fraternal Succession and the Discourse of Filial Piety When Taizu died unexpectedly in 976/11, he had not yet designated an official successor to the throne. His younger brother Zhao Kuangyi 趙匡義 (939–997), later known as Emperor Taizong 太宗 (r.  976–997), succeeded the throne in times of crisis. The unexpected demise of Taizu and the presence of his two mature sons, Zhao Dezhao 趙德昭 (951–979) and Zhao Defang 趙德芳 (959–981), aroused suscipion about the legitimacy of Taizong’s succession. Some rumors circulated that Taizong murdered his brother Taizu to seize the throne.43 In most aspects, Taizong proved himself to be an experienced leader of the new empire. During his two decades of goverance, Taizong estalished a centralized government and ensured internal security by forming a bureaucracy dominated by civil officials. Owing to the Song dynasty’s military disadvandage in confronting the northern Khitan-Liao empire, Taizong eventually transformed the focus of the state policies from military expenditure to literary pursuits. He launched grandiose cultural campaigns, 41. Qin Huitian, Wuli tongkao 五禮通考 (Taibei: Xinxing shuju, 1970), 92.5651–53. 42. XCB, 142.3423. 43. Wen Ying 文瑩, Xiangshan yelu 湘山野錄 (Beijing: Zhonghua shuju, 1984), 74. Much research has been done on the unexpected death of Taizu and the succession crisis between Taizu and Taizong. See Lau Nap-yin and Huang K’uan-chung, “Founding and Consolidation of the Sung Dynasty under T’ai-tsu (960–976), T’aitsung (976–997), and Chen-tsung (997–1022),” in The Cambridge History of China, Vol. 5 Part I: The Sung Dynasty and its Precursors, 907–1279, ed. Denis Twitchett and Paul Jakov Smith (Cambridge, UK: Cambridge University Press, 2009), 242–44, also footnote 130. For a detailed Chinese study on the succession crisis, see Li Yumin 李裕民, “Jiekai ‘fusheng zhuying’ zhimi” 揭開“斧聲燭影”之謎, in Songshi xintan 宋史新探 (Xian: Shanxi shifan daxue chubanshe, 1999), 16–29.

48

Empowered by Ancestors

especially the compilation of voluminous literary projects.44 Accordingly, Taizong hired a large number of literati and civil officials through civil service examinations and other channels. Taizong’s inclination toward cultural campaigns differed from his brother Taizu, who was more interested in military affairs. Taizu established the Song empire; Taizong consolidated it through cultural endeavors. Confucian scholars and civil officials, especially those who had served in the southern states, were offered chances to participate in Taizong’s cultural campaigns. When Song Taizong died in 998, literati and civil officials generally admitted that he was a legitimate successor of Taizu and a governor who made great contributions to the inaugural empire. After the death of Taizong, his ritual designation soon became a controversial issue for the court. In 998/3, Li Zongne 李宗訥, as the prefect of the Commission of Ritual Affairs, led his commission colleagues to memorialize to the court, suggesting a revision of the designation of Song imperial ancestors.45 Li Zongne’s memorial opened a heated discussion about the sensitive issue of fraternal succession between Taizu and Taizong. According to Han ritual texts, scholar-officials from other bureaus insisted that Taizu and Taizong belonged to different generations. Zhang Qixian 張齊賢 (942–1014), the head of the Ministry of Revenue (hubu shangshu 戶 部尚書), argued that Taizong should be regarded as Taizu’s son in temple sacrifices; accordingly, Taizong’s son and the new emperor, Zhenzong 真宗 (r.  997–1022), should be regarded as Taizu’s “grandson,” despite the fact that he was the latter’s nephew by blood.46 Zhang Qixian’s opinion represented the conventional understanding of imperial succession, in which fraternal succession was often conceptualized as regular patrilineal succession to accord with the Confucian emphasis on filial piety. In the Tang controversy concerning the succession between Tang Ruizong and Tang Zhongzong, most ritual officials viewed the brotherly emperors as belonging to the same generation. In practice, the Tang court stored Zhongzong’s tablet in a subsidiary chamber to avoid placing multiple ancestors in the same zhaomu rank. However, in the Song context, the case was more complicated. As both Taizu and Taizong reigned the empire for a considerable period and acquired adequate political legitimacy through their dominant monarchical authority, it was difficult to arrange their tablets in the same way as the Tang court had done with those of Ruizong and Zhongzong. If Taizu and Taizong were not brothers, they definitely would deserve individual chambers. In order to address the ritual sequence concerning the fraternal succession between Taizu and Taizong, Emperor Zhenzong ordered the Commission of Ritual 44. For Taizong’s compilation projects, see Johannes L. Kurz, Das Kompilationsprojekt Song Taizongs (r. 976–997) (Bern: Lang, 2003); Kurz, “The Politics of Collecting Knowledge: Song Taizong’s Compilations Project,” T’oung Pao 87.4–5 (2001): 289–316. 45. SHY, Li 1:15.24; SS: 2566–68. 46. Zhang Qixian quoted some phrases from the Wangzhi and the History of the Former Han Dynasty to bolster his argument. SHY, Li 1:15.24.

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 49

Affairs to thoroughly examine related ritual texts. In the beginning, ritual officials of the commission suggested the court designate Taizu and Taizong as the same generation in terms of zhaomu. In other words, both Taizu and Taizong were designated either as zhao ancestors or as mu ancestors.47 A problem with this approach is that its reasoning is not grounded with solid evidence. Although ritual officials cited a phrase from the Chunqiu zuozhuan zhengyi 春秋左傳正義 (Corrected Meaning of the Zuo Commentary on the Annals), which stated that “the zhao and mu ranks of fathers and sons are different; the zhao and mu ranks of brothers are the same” 父子異昭穆，兄弟昭穆故同,48 this phrase actually came from a sub-commentary (shu疏) on the Zuo Commentary, annotated by the Tang Confucian Kong Yingda 孔穎達 (574–648).49 Because clear statement on the genealogical sequence of fraternal succession was absent in the main texts (zhengwen 正文) of any ritual Classics, Zhenzong’s ritual officials looked for precedents from other historical sources and ritual codes of previous dynasties, including the Shiji, the official dynastic history of Tang (the Jiu Tangshu), and the Sui 隋 (581–619)-compiled Jiangdu jili 江都集禮 (Collection of Ritual Regulations Compiled in Jiangdu).50 Ritual officials from the Commission of Ritual Affairs failed to satisfy Zhenzong. The emperor soon issued a new edict that called for a general discussion about the ritual arrangement of fraternal succession among officials at rank four and above from the Two Drafting Groups and the Department of State Affairs (shangshu sheng 尚書省).51 Participation of high-ranking officials from other bureaus and institutions, especially the Two Drafting Groups, converted the succeeding discussions into a collective advisory meeting. By involving more officials, Emperor Zhenzong attempted to reach a definitive settlement of the disputed issue of fraternal succession between his father Taizong and his uncle Taizu. In the collective advisory meeting held in the summer of 998, the majority of officials argued for the differentiation of Taizu’s and Taizong’s generations in temple rituals.52 They provided several justifications. Based on a merit-based reasoning, these officials claimed: “In ancient settings, Founding Ancestor and Exemplar Ancestor as ritual designations were designed to honor ancestors with great contributions and merits. Therefore, once the ancestors’ merits were confirmed, their honorable designations followed” 古者祖有功，宗有德，皆先有其實，而後正其 47. SHY, Li 1:15.25; 25; SS, 106: 2567. 48. SHY, Li 1:15.25. 49. For the original text in the Chunqiu zuozhuan zhengyi, see Kong Yinda, Chunqiu Zuo zhuan zhengyi, TSZSSSJ, 3:18.194. 50. While the ritual officials of the commission cited the Jiu Tangshu, they claimed that Ruizong and Zhongzong were placed at the same zhao rank in the Tang Imperial Temple. Yet, as I mentioned in Chapter 1, this was not entirely true. In practice, Tang Xuanzong adopted Chen Zhenjie’s and Su Xian’s compromising plan and removed Zhongzong’s tablet from the Imperial Temple in 716. By excluding Zhongzong from the line of imperial succession, the Tang court avoided possible ritual controversy that might have happened in the fraternal succession between Ruizong and Zhongzong. 51. SHY, Li 1:15.25. 52. The SHY fails to date the meeting. However, the XCB records that the final plan on the designations of Taizu and Taizong was promulgated in 998/6. So, I date the meeting to the summer of 998. XCB, 43.913.

50

Empowered by Ancestors

名.53 Because Taizu and Taizong contributed tremendously to the founding and consolidation of the Song dynasty, Zhenzong’s officials argued that the two emperors should be considered as separate generations in the temple for the purpose of illustrating their remarkable merits. Otherwise, if Taizu and Taizong are placed in the same generation, Taizong could not “represent his sole generation by himself ” 不得 自為世數—a practice that symbolically degraded Taizong’s ritual status as the head of Song imperial lineage.54 In other words, the designation “Exemplar Ancestor” by itself indicated the ritual authority of its recipient who should not be associated with any subsidiary positions in the Imperial Temple. Additionally, positing the relationship between Taizu and Taizong in a patrilineal way also symbolized an acknowledgment of their individual contributions. According to the officials from the Two Drafting Groups, placing the two emperors’ tablets in the same generation would underplay Taizong’s political contributions since it belittled his ritual status as subsidiary to Taizu. Additionally, officials from the Two Drafting Groups questioned previous accounts of a parallel setting of imperial tablets in fraternal succession, especially the Tang Confucian Kong Yingda’s commentary on this issue. They argued that Kong’s champion of a parallel setting in fraternal succession was only applicable to cases of feudal lords, such as the State of Lu 魯 in the Spring and Autumn period 春秋 (770–476 BC).55 But a Zhou feudal lord was not an emperor. Given the differences between the Son of Heaven and feudal lords in terms of status and the number of their ancestral temples, it was inappropriate for the Song court to follow the ritual practices of Zhou feudal lords. Moreover, although Kong Yingda’s commentary mentions that brotherly emperors can have the same zhaomu designation, it does not state that imperial ancestors in fraternal succession should not be distinguished from each other.56 When Taizu died, Taizong ritually treated his deceased brother as his father in every aspect, from mourning practices and suburban Altar sacrifices to Taizong’s edicts that described his relationship with Taizu.57 By appealing to Taizong’s ritual practices, Zhenzong’s officials suggested acknowledging Taizong’s own will in considering Taizong and Taizu as two generations.

53. SHY, Li 1:15.25. Based on a brief record of the whole ritual dispute in the SS, some historians interpreted the suggestions made by the officials from the Two Drafting Groups and the Department of State Affairs as a part of the final imperial edict issued by Zhenzong. See, for example, Deng, “Tianxia yijia dao yijia tianxia,” 64. However, in the complete record that was incorporated in the SHY version, it was clearly recorded that all these suggestions were submitted to Zhenzong by “his ministers and officials” 為人臣者. Additionally, in the beginning of the SHY record, it mentioned that the whole piece of writing was copied from a memorial submitted by officials, but not from an edict. 54. SHY, Li 1:15.25. 55. SHY, Li 1:15.25. 56. SHY, Li 1:15.25. 57. SHY, Li 1:15.26. For example, since Taizong had mourned for Taizu for twenty-seven days in a row (a simplified version of the ideal Confucian mourning practice of twenty-seven months that a son devoted to his deceased father), Zhenzong’s officials claimed that Taizong’s mourning practice demonstrated that he actually regarded his brother as his father.

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 51

Nevertheless, Zhenzong still hesitated to designate Taizu and Taizong as two separate generations after collecting more opinions from the collective advisory meeting. Having realized that Zhenzong had not yet made the final decision, some officials reiterated the original plan of placing Taizu’s and Taizong’s tablets in the same temple chamber. The Hanlin academician Song Shi 宋湜 (948–999) argued that there were numerous cases of fraternal succession from the Three Dynasties to Tang; among them there was not a single case to support the differentiation of ritual designations of two succeeding brotherly emperors.58 Particularly, Song Shi was unsatisfied with the expression that Zhenzong as Taizu’s nephew was the latter’s “filial grandson” (xiaosun 孝孫) in the ritual context.59 Perhaps Song’s straightforwardness impressed Zhenzong. The young emperor sent Song Shi’s memorial to the Commission of Ritual Affairs for a further discussion. The final memorial submitted by the commission fulfilled Zhenzong’s interest by suggesting an agenda that compromised the ritual positions of Taizu and Taizong in the Imperial Temple. On the one hand, the commission emphasized the zhaomu principle in differentiating different generations.60 Quoting from the Jitong text, the commission reiterated that zhaomu as a ritual tool differentiated familial relations, especially the relations between fathers and sons. Additionally, they cited an explanatory note from the official ritual code Kaibao tongli, which corroborated the “differentiation of the ritual positions of fathers and sons to highlight the origin of ancestry” 父子異位，以崇本也.61 On the other hand, the commission argued that zhaomu was only applicable to different genealogical generations. In the case of fraternal succession, the ritual designation of succeeding brotherly emperors should not be regulated by the same zhaomu designation that disciplined the designations of emperors in patrilineal succession. The commission cited a myriad of Tang and Jin precedents to explicate why in the case of fraternal succession “brothers should not adopt the same zhaomu principle that was used in the case of patrilineal succession” 兄弟不合繼 位昭穆.62 Alongside some historical precedents, the commission’s officials quoted Tang ritual manuals, especially those liturgical manuals of Tang court rituals such as the Jiaosi lu 郊祀錄 (Records of Suburban Altar Sacrifices) and the Xu qutai li 續曲台禮 (Continuation of the Code of Imperial Rituals), to substantiate their 58. SHY, Li 1:15.26. 59. SHY, Li 1:15.26. 60. SHY, Li 1:15.26. For the original text, see LJZS, 2:49.536; Legge, The Sacred Books of China, 4:246–47. 61. SHY, Li 1:15.26. 62. Early to the third century, the Jin Confucian He Xun 賀循 (260–319) had already expressed that in fraternal successions the brothers should use the same zhaomu designation. He raised a hypothetical case to reveal the problem of identifying the fraternal succession with patrilineal succession in arranging the zhaomu designation of ancestors: “if there are four brothers who successively ascended to the throne, when they die and their tablets are going to be stored in the Imperial Temple, does it mean that four distant ancestors should be removed from the temple” 比有兄弟四人相襲為君者，便當上毀四廟乎? See He Xun, “Dixiong bu he jiwei zhaomu yi” 弟兄不合繼位昭穆議 Quan Jin wen 全晉文, in Quan shanggu Sandai Qin Han Sanguo Liuchao wen 全上古三代秦漢三國六朝文, compiled by Yan Kejun 嚴可均 (1762–1843) (Taibei: Shijie shuju, 1963), 88.5b.

52

Empowered by Ancestors

argument.63 The quotations of Tang ritual manuals demonstrated how previous examples offered resources for the early Song solutions to ritual controversies. By referring back to Confucian Classics, especially the Book of Rites and the Gongyang Commentary on the Annals, officials from the commission eventually worked out a consistent reasoning for a co-chamber setting in fraternal successions.64 Zhenzong accepted the plan suggested by the commission and stored the spirit tablets of Taizu and Taizong in the same chamber, but in different seats (tongwei yizuo 同位異坐).65 In the ritual controversy concerning the fraternal succession between Taizu and Taizong, both sides appealed to Taizong’s political contributions and his authority as a dominant and separate ruler from his brother Taizu. Officials from the Two Drafting Groups worried that placing Taizu and Taizong in the same chamber and generation would belittle Taizong’s ritual status and hence his merits as subsidiary to Taizu. In the later phases of discussions, officials from the Commission of Ritual Affairs used the same reason to persuade Zhenzong to adopt a co-chamber plan. In their opinions, the court should place Taizong’s tablet next to Taizu’s in the same chamber to illustrate Taizong’s merits in consolidating the Song Empire. One important remark about the ritual controversy over the fraternal succession between Taizu and Taizong is that it crystallizes the tension between ritual agendas and political concerns. Zhenzong’s hesitation in determining Taizu’s and Taizong’s ritual designations can be attributed to his anxiety about the potential strain between the two emperors’ lineages. People in the early Song did realize the succession crisis from Taizu to Taizong.66 In Zhenzong’s early reign, because the political climate was still very tense and sensitive, any ritual arrangements related to the succession issue might trigger unpredictable political crisis. In this light, Zhenzong’s extreme caution with respect to his father Taizong’s temple position reflects his concern in balancing not only Taizong’s and Taizu’s ritual statuses but also the interests of Song clansmen who witnessed the court’s ritual policy. While Taizong’s line expectedly would monopolize the throne with Zhenzong’s ascendancy, Zhenzong decided to honor Taizu’s ritual status in the temple as a form of psychological compensation to Taizu’s clansmen and those who admired Taizu’s contributions as the “Founding Ancestor.” On the other side, Zhenzong also avoided a ritual belittlement of his father’s ritual status in the temple, lest it undermined his own political legitimacy. It is also worth noting that the ritual controversy in 998 demonstrated a clear inclination toward the merit-based approach in temple rituals. Predictably, ideal Confucian ritualists who anticipated full manifestation of filial piety in ancestral rituals criticized this approach. Liu Chang 劉敞 (1019–1068) castigated those ritual

63. SHY, Li 1:15.27. 64. SHY, Li 1:15.27. 65. XCB, 43.913. 66. Chaffee, Branches of Heaven, 26–27.

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 53

officials who argued for the co-chamber setting in 998 as “ridiculous” (wang 妄).67 From Liu’s viewpoint, as Taizong received the mandate from his brother Taizu, he should be recognized as the latter’s nominal son.68 Likewise, Yang Shi 楊時 (1053–1135), one of the early advocates of Song Daoxue movements, lambasted the merit-based approach in the 998 setting. In Yang Shi’s words: “If the descendants only make sacrifices to their meritorious ancestors, then these descendants are able to select ancestors to venerate” 若以為有功德然後祭，是子孫得揀擇其祖宗而尊 之也.69 For Yang, this utilitarian approach is anything but a genuine demonstration of filial piety. Zhenzong’s ritual officials primarily embraced a merit-based approach. However, the succeeding reigns of Renzong saw the rise of a discourse on filial piety. Zhenzong died in 1022. His wife acted as his son Renzong’s regent from 1022 to 1033, known as Empress Dowager Liu 劉太后. When Renzong officially acquired the full power as a mature emperor in 1033, he aimed at maintaining the stability that his mother left him. In the ritual aspect, Renzong approved a drafted proposal of the Commission of Ritual Affairs that suggested offering seasonal sacrifices to Taizu, Taizong, and Zhenzong permanently in both the suburban Altars and the Imperial Temple.70 This ritual arrangement was designated as the “worship of Three Sages” (sansheng binyou 三聖並侑) in Song official records. Particularly, Renzong and his ritual officials considered temple sacrifices to Zhenzong as an illumination of Renzong’s filial devotion to his father. The Secretariat-Chancellery (zhongshu menxia 中書門下) proclaimed Renzong’s regular offerings to Zhenzong as “the basis of ultimate filial piety” (zhixiao zhizong 至孝之宗).71 In the 1040s, along with the emergence of new situations, especially the deteriorating international relations with the Tangut kingdom on the northwest, Renzong initiated a series of political and institutional reforms, commonly designated as the Qingli Reforms.72 Accompanied by a revival of Confucian norms and culture, the Qingli Reforms challenged conventional practices and ideas in most political, intellectual, and cultural aspects, including the practice of imperial ancestral rituals. Thereupon, the Qingli ritual controversy over the arrangement of the Imperial Temple emphasized more on filial piety. In 1040, one year before the political implementation of the Qingli Reforms, Zhao Xiyan 趙希言, the auxiliary of the Imperial Archive (zhi mige 直秘閣), suggested the court reform the basic architecture of the Imperial Temple. According to Zhao, the original setting of the Song Imperial Temple compartmentalized its main 67. Liu, “Weixiong houyi,” 為兄後議, Quan Song wen 全宋文 (hereafter QSW) (Shanghai, Hefei: Shanghai cishu chubanshe, Anhui jiaoyu chubanshe, 2006), 59:1290.287–88. 68. Liu, “Weixiong houyi,” 1290.287. 69. Yang Shi, Yangshi ji 楊時集 (Beijing: Zhonghua shuju, 2018), 274. 70. XCB, 116.2732–34. 71. XCB, 116.2734. 72. For a comprehensive analysis of the origin and the development of Qingli Reforms, see Paul Jakov Smith, “A Crisis in the Literati State: The Sino-Tangut War and the Qingli-Era Reforms of Fan Zhongyan,” JSYS 45 (2015): 59–137.

54

Empowered by Ancestors

hall into sixteen segments. From east to west, fourteen segments were arranged in the form of seven chambers to house seven imperial ancestors. The easternmost and westernmost segments were labeled as the two subsidiary chambers.73 Zhao Xiyan was dissatisfied with this setting for two reasons. First, he criticized that the setting assembled all ancestors in a single hall of the temple, and hence failed to provide separate space for each imperial ancestor to enjoy offerings. Second, Zhao questioned why temple designations were not inscribed on the plaques of the chambers.74 In his opinion, the court should revive the ancient Zhou setting of the Imperial Temple by establishing seven temples to house imperial ancestors, separately, each comprising a main hall and a chamber. This was called the “separate configuration of temples” (dugong 都宮). The cost of refurbishing the Imperial Temple, as Zhao expressed, “is not that high in comparison with the cost of building palaces” 費比宮殿未足為多.75 It was important for the court, as Zhao put it, to refurbish the temple to illustrate Renzong’s virtue of filial piety. In responding to Zhao Xiyan’s criticism and suggestions, Song Qi 宋祁 (998– 1061), the deputy prefect of the Court of Imperial Sacrifices, questioned the revival of the dugong setting regarding the Imperial Temple. Song Qi indicated that the practice of placing all ancestors in a single temple had a long tradition. He further argued that it was not necessary to follow the dugong temple setting, which was coined by Zheng Xuan in his commentaries to the Wangzhi. Instead, Song claimed that it was more appropriate to house more ancestors in the temple than construct separate buildings for them.76 In particular, Song criticized Zhao Xiyan’s idea to establish a new tiao temple to house Xizu’s tablet. In Song Qi’s opinion, “there are only six generations of ancestors from Xizu to Zhenzong, it is not appropriate to establish a tiao temple” 僖祖至真宗，方及六世，不應便立祧廟.77 Moreover, although the tiao temple was permanent, Song Qi still considered it as a belittlement of Xizu’s ritual status. Song also argued that the exclusion of Xizu’s tablet from the Imperial Temple might set a bad example for the Song people, if they witnessed that the officially recognized “first ancestor” of the Song clan received little respect from his descendants. The court agreed to Song Qi’s plan to maintain the Imperial Temple setting in 1040/12. Clearly, the rhetoric used by Song Qi and Zhao Xiyan was different from what we have seen in the 998 ritual controversy over fraternal succession. Both Song and Zhao advocated the idea of filial piety in conceptualizing the arrangement of the Imperial Temple, despite the different conclusions they reached. James T. C. Liu mentioned that the Song mingtang sacrifices to Heaven and their ritual performance by Renzong embedded a strong sense of filial piety.78 I would argue 73. SHY, Li 1:15.29. 74. SHY, Li 1:15.29. 75. SHY, Li 1:15.29. 76. XCB, 129.3059–60; SHY, Li 1:15.29–30. 77. XCB, 129.3060. 78. Liu Zijian 劉子健 (James T. C. Liu), “Fengshan wenhua yu Songdai mingdang jitian” 封禪文化與宋代明堂祭

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 55

that Renzong’s officials drew on the same idea of filial piety to examine the arrangement of the Imperial Temple. In a 1050s ritual debate, Zhao Lianggui 趙良規, the prefect of the Court of the Imperial Clan, suggested another reform regarding the Imperial Temple based on an account of filial piety. His suggestion was rejected by the court.79 During Renzong’s reign, the court launched no fundamental reforms on the arrangement of the Imperial Temple. Renzong tended to maintain the temple setting and related rituals. Nonetheless, during and after the Qingli era, the elevation of the discourse on filial piety was obvious in the ritual discussions about the Imperial Temple. Song Minqiu 宋敏求 (1019–1079), who was active in the 1040s and 1050s, perceptively noted that the court granted Song Taizu an additional character of xiao 孝 (filial piety) in his posthumous title in the Qingli era.80 After Renzong’s death in 1063, the Commission of Ritual Affairs suggested building an additional chamber in the Imperial Temple to house Renzong’s spirit tablet. Given the condition that Taizu and Taizong were placed in the same chamber, Sun Bian 孫抃 (996–1064) as the commission’s senior officer highlighted the necessity to build an extra chamber.81 Nevertheless, Sima Guang 司馬光 (1019–1086), who served as a Hanlin academician at that time, opposed Sun Bian’s suggestion. Against building an extra chamber to house Renzong’s tablet, Sima suggested the court remove Xizu’s tablet from the Imperial Temple. He argued that Xizu lacked discernible contributions to the establishment of the Song dynasty and hence should not be permanently placed in the main hall of the temple. In other words, Xizu was not a real king who received a mandate from Heaven.82 According to Sima, as soon as Xizu’s tablet had been removed and placed in the subsidiary chamber, the tablets of other imperial ancestors could be altered accordingly and one chamber would be vacated to house Renzong’s tablet.83 This rearrangement also helped to affirm Taizu’s ritual status as the supreme ancestor in the temple. Based on Han, Jin, and Tang arrangements of the Imperial Temples, Sima asserted the necessity of facilitating a temple configuration consisting of three zhao and three mu chambers. For Sima, the removal of Xizu’s tablet “would not fail to accord with both ancient ritual codes and contemporary regulations” 於先王典禮及近世之制無不符合.84 Given Sima’s

天, Liang Songshi yanjiu huibian 兩宋史研究彙編 (Taipei: Lianjing chuban shiye gongsi, 1987), 3–9. 79. Zhao argued that Taizu’s tablet should be placed at the center of the temple’s main chamber, facing east. By doing so it would confirm his ritual status as the Song Primal Ancestor. Wang Juzheng 王舉正, the head of the Bureau of Ritual, disagreed with Zhao and suggested maintaining the conventional practice of leaving the central place of main chamber vacant. See XCB, 189.4568–69. 80. Song Minqiu, Chunming tuichaolu 春明退朝錄 (Beijing: Zhonghua shuju, 1980), 41. Song Minqiu himself disagreed with the court’s revision of Taizu’s posthumous title. He argued that the contribution made by Taizu is tremendous and hence should not be confined to the virtue of filial piety. 81. SHY, Li 1:15.34b–35. XCB, 198.4809. 82. XCB, 198.4810. 83. Sima Guang, “Fumiao yi” 祔廟議, in Wenguo wenzheng simagong ji 溫國文正司馬公集, Sibu chongkan chubian suoben 四部叢刊初編縮本 (hereafter SBCKCBSB) (Taibei: Shangwu yinshuguan, 1969), 46:26.240; the same essay can also be found in other editions of Sima Guang’s collected works. 84. Sima, “Fumiao yi,” Wenguo wenzheng simagong ji, 26.240.

56

Empowered by Ancestors

identity as a great historian, his argumentation was basically historical, emphasizing the temporality of the Imperial Temple arrangement. Against Sima’s contention, Sun Bian and his colleagues in the commission refuted that Xizu’s tablet played a crucial role in imperial sacrifices. With no surprise, they appealed to the cultural authority of the Three Dynasties. Although they admitted that the arrangement of the Imperial Temple had to change to suit contemporary needs, they devoted scant attention to the historical precedents that Sima had employed in his memorial.85 Instead, they claimed that in the original Zhou context “the designation of the Founding Ancestor referred not to the first king who received a mandate from the Heaven, but to the first feudal lord who received a fief from the son of Heaven” 所謂太祖，亦非始受命之主，特始封之君而已.86 Sun’s differentiation between the shouming zhi zhu 受命之主 (the first king who received a mandate from the Heaven) and the shifeng zhi jun 始封之君 (the first feudal lord who received a fief from the son of Heaven) was inspired by the late Tang ritual official Liu Mian 柳冕 (ca. 730–ca. 804). Liu stated that granting the title of “Founding Ancestor” to any officially recognized “first ancestor” was appropriate, regardless of whether that “first ancestor” was a king or a feudal lord.87 Possibly inspired by Liu’s conception of the “first ancestor,” Sun Bian argued that Xizu was the first Song ancestor who had been recognized and acknowledged by the court, despite his humble background and dubious status as a shifeng zhi jun.88 Considering the status quo of the Song Imperial Temple, Sun worried if the court blatantly abrogated Xizu’s ritual status by removing his tablet from the temple, it might defy the “ritual intent of ancient sage-kings” 先王之禮意.89 Emperor Yingzong 英宗 (r. 1063–1067), Renzong’s adopted son and his successor, was persuaded by Sun Bian. He decided to preserve Xizu’s chamber in the Imperial Temple throughout his reign. Other than Sima Guang and Sun Bian, only a few officials participated in the Imperial Temple debate in 1063.90 This contradicts sharply with the controversy over the posthumous title of Yingzong’s deceased biological father from 1065 to 1066, in which a larger number of officials were engaged. The ritual controversy was named as puyi 濮議 in Song sources and has been thoroughly studied by historians.91 In 85. SHY, Li 1:15.35; XCB, 198.4811. 86. XCB, 198.4811. 87. Jiu Tangshu, 213.4122. Gao Mingshi argued that the shifeng zhi jun should be the real founder of an empire (chuangye zhe 創業者). Yet, it is impossible to reach that conclusion solely based on Liu Mian’s writing. Liu merely stated that the shifeng zhi jun referred to the progenitor of a feudal lord and should be dignified as the taizu ancestor of the lord’s lineage 故雖諸候，必有先也，亦以尊太祖焉. He never said that the progenitor of the feudal lord must be a meritorious founder in political sense. Gao Mingshi, “Lifa yiyi xiade zongmiao,” 39. 88. SHY, Li 1:15.35; XCB, 198.4810. 89. SHY, Li 1:15.35. 90. Lu Shizong 盧士宗, one of Sima Guang’s colleagues in the Hanlin Institute of Academicians (xueshiyuan 學 士院), joined Sima in 1063 in supporting the removal of Xizu’s tablet from the temple. XCB, 198.4810. Some officials from the Two Drafting Groups also participated in the temple debate. Yet, their names were not revealed in either XCB or SHY. Generally, the temple debate in 1063 was limited to a small group of officials. 91. Among these studies I would like to mention Carney Fisher, “The Ritual Dispute of Sung Ying-tsung,” Papers on Far Eastern History 36 (1987): 109–38; Meyer, Ritendiskussionen am Hof der nördlichen Song–Dynastie

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 57

the puyi debate, a group of censors (yushi 御史) and remonstrators (jianguan 諫 官), led by Sima Guang and Lü Hui 呂誨 (1014–1071), urged Emperor Yingzong to honor his adopted father Renzong above his biological father, the Prince of Pu.92 In contrast, the ministers, led by Han Qi 韓琦 (1008–1075) and Ouyang Xiu, insisted that Yingzong’s personal affection for his biological father should be respected and ritually reflected in the posthumous title of the Prince of Pu.93 As previous studies of puyi have pointed out, both groups of censors and ministers made use of Confucian values to support their ritual claims. What previous studies have not pointed out is how filial piety was emphasized in the puyi controversy in the same way as it was in the 1063 debate over the Imperial Temple. In the beginning of puyi, a group of idealistic officials reminded Yingzong of his ritual affiliation with the imperial ancestors in the temple. Among these idealists, Sima Guang stood out and spearheaded to highlight how Yingzong as an adopted emperor should fulfill his filial duty toward ancestors in the “line of imperial succession” (datong 大統).94 In fact, Sima’s emphasis on Yingzong’s filial duty to Renzong in the puyi debate echoes with his affirmation of Taizu’s ritual status as the founding ancestor in the 1063 temple debate. The difference is that in 1063 Sima Guang argued for Taizu’s ritual status based on a merit-based approach, whereas in 1065 he adopted the rhetoric of filial piety to persuade Yingzong to fulfill his filial duty toward Renzong and his adopted mother, Empress Dowager Cao 曹太后. Sima advised Yingzong in earnest: if the emperor behaved as a filial son, “then the reputation of Your Majesty’s benevolence and filial piety would spread to eternity, and the virtue of your outstanding astuteness would reach all areas, the Imperial Temple would be secure forever, and the imperial descendants would receive the blessings (from their ancestors)” 則陛下仁孝之名流於萬世，英叡之德達於四表，宗廟永 安，子孫蒙福.95 Compared to his argument in the 1063 temple debate, Sima spoke more on filial piety in puyi and rhetorically related it to the stability of the Imperial

(1034–1093), 254–88, 439–48; Michael McGrath, “The Reigns of Jen-tsung (1022–1063) and Ying-tsung (1063–1067),” in The Cambridge History of China, Vol. 5, Part I: The Sung Dynasty and its Precursors, 907– 1279, ed. Denis Twitchett and Paul Jakov Smith (Cambridge, UK: Cambridge University Press, 2009), 340–43. A recent study focuses on the puyi issue and its political implications in Song factionalism. See Martin Kroher, “‘With Malice Toward None’ to ‘A House Divided’: The Impact of Changing Perceptions of Ritual and Sincerity on Elite Social Cohesion and Political Culture in Northern Song China, 1027–1067” (PhD diss., Harvard University, 2014), 406–91. 92. XCB, 205.4971–76. 93. Recognizing Yingzong’s will to honor his natural father, the Prince of Pu, Han and Ouyang suggested bestowing upon him ritual designations like “imperial father” (huang kao 皇考), instead of the conventional designation of “imperial paternal uncle” (huang bo 皇伯) that accorded with the Confucian norm of imperial adoption. McGrath, “The Reigns of Jen-tsung (1022–1063) and Ying-tsung (1063–1067),” 342. 94. McGrath traced puyi to an earlier memorial submitted by Sima Guang in the spring of 1064. Actually, that memorial was submitted by Lü Hui, addressed to Dowager Empress Cao. XCB, 199.4837. However, in somewhere else the XCB clearly records that Sima “solely and enthusiastically drafted a memorial” 獨奮筆立議 to advise the court at the beginning of the puyi controversy, possibly in the spring of 1065. XCB, 205.4971. 95. XCB, 200.4855. I consulted Martin Kroher’s translation here. See Kroher, “‘With Malice Toward None’ to ‘A House Divided,’” 412.

58

Empowered by Ancestors

Temple and ancestral spirits therein. His opponents, in fact, made the same correlation in their memorials.96 Yingzong’s reign is short-lived. His death in 1067 triggered a new dispute over the Imperial Temple. This time, the Commission of Ritual Affairs suggested removing Xizu’s tablet from the temple in order to offer a vacant chamber to house Yingzong’s tablet. Zhang Fangping 張方平 (1007–1091), the head of the Bureau of Ritual, championed the commission’s suggestion. Zhang regarded the arrangement of the Imperial Temple as something with mythical power, which not only regulated generations of the imperial clan, but also illuminated “the essence of benevolence and integrity in the utmost extremity” 極仁義之本.97 He further argued that Xizu’s tablet should be removed in di and xia sacrifices. The new Emperor Shenzong endorsed Zhang Fangping’s suggestions by removing Xizu’s tablet from the main chamber of the Imperial Temple. This temporary removal of Xizu’s tablet from the temple demonstrated the fleeting success of the merit-based approach in the late 1060s. However, controversies concerning the temple remained unsettled, given the increasing influence of the rhetoric of filial piety on imperial rituals. Later years of Shenzong’s reign saw the return of temple controversies in a subtler way, when Wang Anshi instated his New Policies and ritual affairs were shadowed by the intellectual interests of his personal scholarship.

Concluding Remarks Since the beginning of the Northern Song, ritual officials from the Commission of Ritual Affairs and other bureaus had argued over the ritual order of imperial ancestors in the Imperial Temple. From Zhang Zhao to Sima Guang, a continuation of the Tang interpretations on the Imperial Temple was observable in the early Song ritual debates over relevant topics. As exemplified in the 998 controversy concerning the fraternal succession between Taizu and Taizong, ritual officials generally embraced a merit-based interpretation of the temple arrangement. Nevertheless, the discourse on filial piety gained prominence along with Emperor Renzong’s reign after the 1030s. Ritual officials, especially those who had served in the Commission of Ritual Affairs, were more inclined to the filial piety discourse in explicating temple rituals. Xizu, a distant Song ancestor who made no contribution to the foundation of the Song empire, had been officially recognized as the “first ancestor” because the honoring of his ritual status symbolized the filial virtue of his descendants. The opinions of Zhao Xiyan, Song Qi, and Sun Bian exemplify this type of discourse, as well as those of later ritualists like Liu Chang and Yang Shi. 96. For example, in an early memorial concerning the puyi, Han Qi suggested the court start a discussion on the posthumous rituals concerning the Prince of Pu and his three wives. In that memorial, Han claimed that the emperor’s filial behavior was at the heart of his responsibility in the Imperial Temple. XCB, 201.4872. 97. SHY, Li 1:15.36. For the original version of Zhang Fanping’s memorial, see Zhang, “Taimiao tiao xizu yi” 太廟 祧僖祖議, QSW, 37:800.301.

Northern Song Conceptions of the Imperial Temple and Temple Sacrifices 59

A compromise between the two discourses of merits and filial piety characterized court decisions in the Song temple arrangement. Compared to Tang ritual debates, Song debates concerning the temple involved more political concerns. As the Song court set strict limits on clansmen’s access to political power, the Imperial Temple offered a type of ritual indemnity to clansmen. In this light, early Song controversy over the Imperial Temple reminds me of the juxtaposition between social structure and communitas made by the anthropologist Victor Turner. Turner has argued that structural marginality in political and social realms sometimes serves as the presupposition of ritual superiority.98 In the case of the Song Imperial Temple, it might be interesting to conceive that Song clansmen’s lack of political power was balanced by the affirmation of their ancestors’ ritual status in the Imperial Temple. Although Taizu’s descendants were barred from substantive power in the Northern Song,99 the advocacy of an elevation of Taizu’s ritual status would ease their discontentment with the monopoly of the throne by Taizong’s line. This may explain why ritual officials incessantly emphasized Taizu’s supreme role in the Imperial Temple based on a merit-based discourse. Even Liu Chang, an opponent of the merit-based discourse, admitted that the transition of the absolutist power from Taizu to Taizong had to be ritually “framed” (ge 格) by the genealogical sequence in the temple. Otherwise, it would result in a crisis of legitimacy, as people would ask where Taizong received his mandate and the right of governance.100 A failure to answer this vital question could be perilous to the future regime of Taizong’s descendants. To balance political concerns and ritual practices, Song ritual officials established official norms of temple arrangment to accord with their understanding of Confucian ritual norms, especially the idea of filial piety. However, a sophisticated interpretation of how filial piety should practically regulate temple rituals only emerged after Wang Anshi had initiated his New Policies in 1069. The intrinsic relations between Wang’s New Policies and his interest in revising the arrangement of the Imperial Temple led to a series of chain reactions in the late eleventh century, including the two momentous ritual debates on temple rituals in 1072 and in 1079.

98. As Turner succinctly summarized, “the structurally inferior aspire to symbolize structural superiority in ritual; the structurally superior aspire to symbolize communitas and undergo penance to achieve it.” Victor Turner, The Ritual Process: Structure and Anti-Structure (Chicago: Aldine Publishing Company, 1969), 203. 99. Chaffee, Branches of Heaven, 25–30. 100. Liu, “Weixiong houyi,” 41.6.

Section Two Song Imperial Temple: Ritual Debates, Factional Politics, and Intellectual Diversity

3 An Examination of the 1072 Controversy over the Song Primal Ancestor

During Song Shenzong’s reforms of administration and officialdom in the Xining and the Yuanfeng eras (Xifeng xinzheng 熙豐新政), the political context of ritual discussions had undergone major transformations. Two ritual debates that occurred respectively in 1072 and 1079 intensified the controversy over the Imperial Temple. In these two ritual debates, with sophisticated ritual rhetoric, ritual officials increasingly disagreed with each other about the ritual status of the Primal Ancestor and other ancestors in the Imperial Temple. This chapter examines the ritual interests of Song scholar-officials who were engaged in the 1072 debate and demonstrates how these interests revealed a new dimension of the factional politics in the late Northern Song. Traditional understanding of Northern Song factional politics regarded it as a dichotomous political struggle between the conservative officials and a group of reformists under Wang Anshi’s leadership.1 This understanding of Song factional politics is prevalent, but it is far from comprehensive. Through a close reading of the ritual debate over the Song temple setting in 1072, I supplement the conventional understanding of Song factional politics by revealing the intellectual interests of ritual officials, as well as how these interests transcended the narrow boundaries of factional politics.

Launch and Development of the 1072 Primal Ancestor Debate Song Shenzong’s reign has usually been regarded as a watershed in Song history. He was a forward-looking emperor who was eager for a fundamental change. In the spring of the third year of the Xining 熙寧 era (1070), Shenzong promoted Wang Anshi to the grand councilor (pingzhang zhengshi 平章政事, practically, a prime minister of the Northern Song Empire).2 With deep trust and full support from 1. Politically, Northern Song politicians tended to conceptualize factionalism with polarizing vocabularies for the purpose of persuading the emperors to support their own interest groups and to expel their adversaries. In practice, Song factional politics were rather complicated and engaged different interest groups. For a detailed analysis of Song factional politics and factionalism in political domain, see Luo Jiaxiang 羅家祥, Bei Song dangzheng yanjiu 北宋黨爭研究 (Taibei: Wenjing chubanshe, 1993). 2. For a brief introduction of the duties and power of the Song Grand Councilor, see E. A. Kracke, Civil Service in Early Sung China, 30–32.

64

Empowered by Ancestors

Emperor Shenzong, Wang began managing state affairs with the introduction of a series of reforms, those of which included reforms on imperial ancestral rituals.3 In 1072/3, the Secretariat-Chancellery launched the Primal Ancestor discussion in court through the submission of a formal memorial to the emperor.4 The memorial reads: All things originate from Heaven; man originates from his ancestor. The purpose of ancient temple setting is to keep the less related (ancestors in the sacrificial sequence) without forgetting them, to take care of the distant (ancestors) without leaving them . . . Considering the sequence of seniority and the priority of ancestral worship, even if the descendant has merit as great as that of the Sages, he cannot take precedence over his ancestor. This is the general Way of all-under-Heaven for thousands of generations. Since the imperial lineage prior to Xizu is untraceable, Xizu should be designated as the Primal Ancestor of the Song dynasty, the same as the Zhou progenitor Ji and the Shang progenitor Qi. Yet, nowadays Xizu’s temple is removed and his spirit tablet is placed in the subsidiary chamber. This kind of practice defies the virtue of filial piety and the spirit of serving the dead as serving the living, as it attaches the superior ancestor to his inferior offspring (in the ancestral space of the subsidiary chamber). Occasionally, there may be some historical precedents for such a practice; yet we find no proof in the Classics. Under the holy regime of Your Majesty, it is the right time to remodel ancestral rites on the basis of decency.5 萬物本乎天，人本乎祖。先王廟祀之制，有疏而無絕，有遠而無遺……若夫 尊卑之位，先後之序，則子孫雖齊聖有功，不得以加其祖考，天下萬世之通 道也。竊以本朝自僖祖以上，世次不可得知，則僖祖有廟，與契稷無以異。 今毀其廟，而藏其主夾室，替祖考之尊而下祔於子孫，殆非所以順祖宗孝 心、事亡如存之義。求之前載，雖或有然，考合於經，乃無成憲。因情禮 制，實在聖時。

As early as 1069, Wang Anshi and Emperor Shenzong had already initiated an imperial clan reform that aimed to cut imperial support for the clansmen who were 3. Extensive research has been done on the New Policies carried out by Wang Anshi. For example, see James T. C. Liu, Reform in Sung China: Wang An-shih (1021–1086) and His New Policies (Cambridge, MA: Harvard University Press, 1968), 4–7; Paul Smith, “Shen-tsung’s Reign and the New Policies of Wang An-shih, 1067– 1085,” 347–483; Peter Bol, This Culture of Ours, 212–53; Deng Guangming 鄧廣銘, Beisong zhengzhi gaigejia Wang Anshi 北宋政治改革家王安石 (Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2007), 154–241; Qi Xia 漆俠, Wang Anshi bianfa 王安石變法 (Shanghai: Shanghai renmin chubanshe, 1979), 100–168, 269–87; Ichio Higashi 東一夫, Ō Anseki shinpō no kenkyū 王安石新法の研究 (Tokyo: Kazama Shobō, 1970), 394–920. For a comprehensive literature review of Wang’s New Policies, see Li Huarui 李華瑞, Wang Anshi bianfa yanjiushi 王安石變法研究史 (Beijing: Renmin chubanshe, 2004), 327–599. 4. SHY, Li 1:15.37. 5. SHY, Li 1:15.37. Gu Donggao 顧楝高 (1679–1759) dated this memorial to the sixth year of the Xining era, which was 1073, see his Gu Donggao, Wangjinggong nianpu 王荊公年譜, Songren nianpu congkan 宋人年譜 叢刊 (Chengdu: Sichuan daxue chubanshe, 2003), 3.1999. Yet Li Tao 李燾 (1115–1184) provided a detailed explanation on the date of this memorial. Based on the compilation format of court documents, Li persuasively proved that the year of the submission of this memorial should be 1072, instead of 1073. See XCB, 232.5629.

An Examination of the 1072 Controversy over the Song Primal Ancestor 65

out of the “five mourning grades” (wufu 五服).6 The implication of that reform was financial in nature. After three years, when Wang Anshi’s reforms were on track and financial matters were primarily addressed by the newly established Finance Planning Commission (zhizhi sansi tiaolisi 制置三司條例司), Wang turned his eyes to the imperial ancestral rituals and asked for a reconsideration of the Imperial Temple layout; the above Secretariat-Chancellery memorial was indeed drafted by Wang himself.7 Wang Anshi’s draft was, on behalf of the Secretariat-Chancellery, submitted by the court to the Two Drafting Groups for further discussion. In 1072/4, one month after the submission of the first draft, the emperor had agreed, thereupon issuing an edict requesting the Hanlin Institute of Academicians to assemble the officials of the Two Drafting Groups to rediscuss the Primal Ancestor issue.8 During this discussion, the Hanling academician Yuan Jiang 元絳 (1008–1083) as well as a group of the edict drafters had championed Wang Anshi’s early proposal of changing the layout of the Imperial Temple. Consequently, they submitted a detailed memorial suggesting that Xizu ought to be honored as the Primal Ancestor and his temple to be established at the center of the entire Imperial Temple configuration. The reasoning illustrated in that memorial deserves our attention here. In essence, Yuan Jiang had argued that Xizu should be honored as the Primal Ancestor because it was technically impossible to trace back any imperial ancestor prior to Xizu. Yuan further implied that if someone argued for Taizu’s ritual superiority over Xizu in the Imperial Temple, then “everyone under Heaven would not be able to honor their ancestors, since descendants could surpass their ancestors in the ritual sequence based on personal merit and individual achievements” 是以天下之不復知尊祖， 而子孫得以有功加其祖考也.9 Like Wang Anshi, Yuan Jiang and his fellows argued that merit and personal achievements should not be overrated in ancestral worship as they contradicted the virtue of filial piety. For those forebears who made great achievements and significant contributions to the founding of the empire, such as King Wen and King Wu of Zhou, their achievements should be attributed to their “original lineage” (bentong 本統), which was the Zhou primogenitor Hou Ji 后稷. In this sense, Xizu of Song and Hou Ji of Zhou were alike in initiating the bloodline of great imperial houses. Genealogically, then, they should both be designated as the Primal Ancestor in 6. John W. Chaffee, Branches of Heaven, 68–71. 7. The draft can still be found in Wang Anshi’s collected works. See Wang, “Miaoyi zhazi” 廟議劄子, in Linchuan xiansheng wenji 臨川先生文集 (hereafter cited as LCJ), SBCKCBSB, 51:42.269. 8. XCB, 240.5838; SHY, Li 1:15.37. 9. SHY, Li 1:15.37–38. This memorial was also preserved in Yuan Jiang’s biography in Dongdou shilüe 東都事 略, with slightly different wordings. Wang Cheng 王稱, Dongdou shilüe, 130 juan in 4 vols. (hereafter cited as DDSL) (Taibei: Wenhai chubanshe, 1967), 3:81.4b. The fact that the memorial has been collected in Yuan’s personal biography further confirms that Yuan drafted its final version as a talented literatus. For a superb introduction of the DDSL as a historical source and its textual relationship with Li Tao’s XCB, see Chen Shu 陳 述, “Dongdou shilüe zhuanren Wang Shang Cheng fuzi” 東都事略撰人王賞稱父子, Zhongyang yanjiuyuan lishi yuyan yanjiusuo jikan 中央研究院歷史語言研究所集刊 8.1 (1939): 129–38.

66

Empowered by Ancestors

the Imperial Temple. Yuan Jiang and his fellows argued that seniority, rather than personal achievements, should take precedence in ancestral sacrifices, since only through the former could one recognize the ritual intent of ancestral worship itself. Yuan’s argument was opposed by his colleagues from the Two Drafting Groups, especially Han Wei 韓維 (1017–1098). In contrast to Yuan Jiang’s suggestion to elevate Xizu’s ritual status, Han claimed that Taizu should be recognized officially as the Primal Ancestor through the emphasis of one’s achievements and merits as qualifications of the Primal Ancestor title—a strategy that aimed to distract the emperor’s attention away from Xizu’s unparalleled role in initiating the imperial lineage. Han’s proposal basically followed Sima Guang’s earlier suggestion to remove Xizu’s spirit tablet from the Imperial Temple when Renzong died in 1063. Han insisted that the central position in the Imperial Temple should be reserved for the ancestor with the temple title of taizu because the title itself designated the first ancestor who had consolidated the empire. Since Xizu was not the real founder of the Song dynasty, the court should not replace the spirit tablet of Taizu with that of Xizu. Furthermore, regarding Yuan Jiang’s emphasis on the “original lineage,” Han pointed out that it was unknown to the Song court exactly who Xizu’s father was, which in other words meant Han Wei had implied that Xizu was not of heroic origins. Therefore, one could not equate Xizu with the ancestors of ancient dynasties such as Qi 契 and Hou Ji, who had possessed charismatic prestige and had noble origins.10 Han Wei also discussed the spatial arrangement of the Primal Ancestor in di and xia sacrifices. Interestingly, in an early memorial, Han claimed that the central position in the xia sacrifice should be left vacant.11 However, in his response to Yuan Jiang’s memorial in 1072, he admitted that Xizu’s tablet could be placed at the central position in these sacrifices to manifest the correct ritual order.12 It seems that Han Wei changed his mind and was attempting to adopt a negotiating strategy by making concession to the pro-Xizu officials in 1072. But his general interest undoubtedly was to downplay Xizu’s ritual status. Han Wei’s suggestion to elevate Taizu’s position in the temple architecture was further substantiated and elaborated by his colleague from the Hanlin Institute of Academicians, Sun Gu 孫固 (1016–1090). In comparison with Han Wei, Sun was more straightforward in his view on the Primal Ancestor issue. In his memorial, Sun simply asserted that Xizu did not deserve the Primal Ancestor position in the Imperial Temple. According to Sun, only those heroes who made great contributions to all-under-Heaven could be the chief recipients of temple sacrifices. Since Xizu’s contributions were so obscure—if any at all—his spirit should not be honored in an extraordinary way. 10. XCB, 240.5840–41; SHY, Li 1:15.37. For the original version of Han Wei’s proposal, see Han, “Yi Xizu miao zhuang” 議僖祖廟狀, Nanyang ji 南陽集, SKQS, 1101:22.4a–7b. 11. Han Wei, “Yi xiaxiang xu dongxiangwei zhuang” 議祫享虛東向位狀, Nanyang ji, 22.1a–3a. 12. Han, “Yi Xizu miao Zhuang,” Nanyang ji, 22.7b.

An Examination of the 1072 Controversy over the Song Primal Ancestor 67

Sun Gu’s memorial put an end to the first phase of the ritual debate among the officials of the Two Drafting Groups. At the end of his memorial, Sun wished the emperor either to follow his suggestion to rectify Taizu’s ritual status as the Primal Ancestor or to call for a further discussion involving officials from major ritual institutions. The court agreed to Sun’s second suggestion and sent his memorial, together with the ones from Yuan Jiang and Han Wei, to the Commission of Ritual Affairs and the Court of Imperial Sacrifices for a comprehensive review. Song Minqiu, the head of the Court of Imperial Sacrifices, told the court that since he had composed an essay on the placement of Xizu’s tablet in 1067, he dared not to make a conclusion about the ongoing debate for fear that his personal interest might affect the court’s decision.13 In the absence of Song Minqiu, the Commission of Ritual Affairs took more responsibility in examining the controversy over the issue of the Primal Ancestor. Thus, ritual officials from the commission played a major role in the succeeding discussion. In the second half of 1072, more ritual officials were drawn into the flames of the Primal Ancestor debate, thus beginning the second phase of discussion. Zhao Yiruo 趙彥若, the assistant prefect of the Commission of Ritual Affairs, raised a practical issue concerning the placement of the Primal Ancestor in 1072/7. Zhao pointed out that the court had to construct an additional chamber to store Xizu’s tablet if it really designated Xizu as the Primal Ancestor, given that there were only eight chambers in the existing Imperial Temple.14 Wang Jie 王介 (1015–1087), the sub-editor of the Imperial Archive (mige jiaoli 秘閣校理), immediately submitted a memorial against Zhao’s suggestion to expand temple configuration.15 Wang Jie’s major premise was that imperial ancestral worship should set a limit on the number of ancestors who could receive offerings; otherwise, the emperor would have to offer sacrifices to an endless list of ancestors. Ancient sages had established the seventh generation as the furthest genealogical limit for whom the royal house could build temples, and the ninth generation as the furthest generation to whom the royal house could offer sacrifices. “All of these are demarcated by ritual” 七廟 自顯祖之外，而祧亦猶九族，至高祖而止也，皆以禮為之界也.16 Additionally, Wang Jie echoed and developed Han Wei’s and Sun Gu’s meritocratic approach in conceptualizing the ritual status of the Primal Ancestor. He argued that the Primal Ancestor, by definition, should either be a feudal lord with a fief, or a king who had received the Mandate from Heaven to form a lineage.17 Since Xizu was neither, he was thereby unqualified for the Primal Ancestor position. Wang Jie thus suggested removing Xizu’s tablet from the Imperial Temple, for the reason that Xizu’s status failed to earn him the special treatment of permanent storage in the temple.18 13. XCB, 240.5848. 14. Zhao, “Qizeng taimiao jiushi yifeng xizu zou” 乞增太廟九室以奉僖祖奏, QSW, 84:1833.235. 15. XCB, 236.5748. 16. SHY, Li 1:15.41. 17. SHY, Li 1:15.41. 18. XCB, 236.5748–50.

68

Empowered by Ancestors

Liang Tao 梁燾 (1034–1097), Zhang Shiyan 張師顏, and Zhang Gongyu 張 公裕, all of whom had served in the Commission of Ritual Affairs and the Court of Imperial Sacrifices, further submitted a memorial in 1072/10. In this collective memorial, these ritual officials explicitly denied Xizu’s ritual status as the Song Primal Ancestor. Specifically, they objected to Yuan Jiang’s comparison between Hou Ji of Zhou and Xizu of Song. Hou Ji was a celebrated hero who had taught the people the art of agriculture, the very name Hou Ji itself—literally “the Lord of Millet”—proved that he was an enormously important figure who deserved worship by the Zhou people. Furthermore, the ritual officials claimed that Hou Ji’s contributions were recorded in the series of “Zhou Odes” 周頌 in the Book of Songs. In their opinions, the Mandate of Heaven (chengming 成命) in the “Zhou Odes” implied Hou Ji’s merit in initiating the bloodline of Zhou royalty.19 Xizu, in contrast, did not have any traceable contributions. In contrast to Zhao Yiruo, Liang Tao, and the two Zhangs, some younger colleagues in the ritual institutions sought for a compromise in the Primal Ancestor debate. Arguing against Han Wei, Sun Gu, and Wang Jie, Su Zhuo 蘇梲 vied for the promotion of Xizu to the Primal Ancestor position. Su suggested, however, that temple rituals involving Xizu’s spirit should be performed in a less solemn way since the “traces” (ji 迹) of Xizu’s contributions to the dynasty were obscure, especially compared with those of Qi and Hou Ji, the Primal Ancestors of the Shang and Zhou dynasties, respectively. In order to demonstrate that liturgical difference in temple rituals, Su suggested the court place Xizu’s tablet in the Temple of Spectacular Numina.20 By positing Xizu’s spirit in a Daoist sacrificial site, Su attempted to reconcile the discrepancy between Xizu’s deficiency in actual contributions and his supreme position in the genealogical sequence of the Song imperial family. Moreover, since the relocation of Xizu’s tablet would have far-reaching effects on the whole system of imperial sacrifices, Su Zhuo suggested the court follow the early Song practice of calling an ad hoc collective advisory meeting that would include comments and reports from officials of different ranks who came from the Department of State Affairs (shangshusheng zhusi baiguan 尚書省諸司百官), and, if possible, also seeking opinions from court fortune-tellers (bushizhe 卜筮者).21 The court, however, appeared to have ignored Su Zhuo’s suggestion, possibly because Wang Anshi himself preferred to keep the ritual discussion to a limited audience.22 Apart from Liang Tao and Su Zhuo, other ritual officials of lower ranks had also written memorials to the court. Yang Jie 楊傑, the archivist of the commission, together with Song Chongguo 宋充國, the associate manager of the commission, and Zhou Mengyang 周孟陽, an assistant ritual administrator of the Court of Imperial Sacrifices, asked for a rectification of Xizu’s ritual status and claimed that 19. XCB, 240.5849. 20. XCB, 240.5854–55. 21. SHY, Li 1:15.46. 22. Zhu, Shibangguo zhi shenzhi, 185.

An Examination of the 1072 Controversy over the Song Primal Ancestor 69

Xizu was the perfect candidate for the Primal Ancestor position. By distinguishing the designation of Primal Ancestor from that of the “Founding Ancestor” (taizu 太 祖), Yang, Song, and Zhou developed a new approach to addressing the tension that had developed between seniority and merit. They grounded their reasoning mostly on the Han Classicist Zheng Xuan’s commentary on the Rites and Ceremonies 儀禮. Zheng therein defined the “Great Ancestor” as the “first feudal lord who received the fief,” and the Primal Ancestor as the “first legendary ancestor who received a Mandate from Heaven” (ganshenling er sheng 感神靈而生).23 Because Xizu gave birth to the Song royal line, he should be recognized as the Primal Ancestor and placed parallel to the spirit of the “Thearch in Response to Birth” (ganshengdi 感生 帝) in court sacrifices.24 Zhang Heng 章衡 (1025–1099), another assistant ritual administrator of the Court of Imperial Sacrifices, also argued for the rectification of Xizu’s ritual status as the Song Primal Ancestor. According to Zhang Heng, since the imperial lineage prior to Xizu was untraceable, Xizu should be regarded as the Primal Ancestor to illustrate the virtue of filial piety.25 Second, as ancient sage-kings without exception traced their imperial lines back to their “original lineage,” the virtue of filial piety should be considered more important than personal merit in prioritizing genealogical sequence.26 These two points clearly refer back to Yuan Jiang’s memorial. Third, if Xizu’s tablet was to be placed on the right side of Shunzu’s (Xizu’s son) rather than at the center, it would directly violate the ethical values of filial piety, as it would degrade the father’s status by positioning him sideways. Last but not least, Zhang asserted that it was inappropriate to build a new Imperial Temple for Xizu because such practice violated the a priori spirit of ancestral ritual and was not recorded in any ritual Classic.27

Intellectual Interest and Political Stance of Ritual Officials in the 1072 Debate The debate over the Primal Ancestor continued up to the second half of 1072, yet the problem remained unsolved. Officials who served in the ritual institutions played a role in the second phase of the discussion since they had been accustomed to proposing and revising ritual agendas. Consequently, in the eleventh month of 1072, after a further discussion with the Secretariat-Chancellery and particularly with Wang Anshi, Emperor Shenzong officially designated Xizu as the Primal Ancestor

23. XCB, 240.5856. For the original text in the Rites and Ceremonies, see Yili zhushu 儀禮注疏, TSZSSSJ, 2:30.242. 24. XCB, 240.5858. 25. XCB, 240.5859. 26. XCB, 240.5859. 27. XCB, 240.5860.

70

Empowered by Ancestors

in the Imperial Temple.28 In the first month of 1073, Xizu’s spirit tablet was finally placed into the central Primal Ancestor chamber of the temple.29 The Primal Ancestor controversy in 1072 was initiated and concluded by Wang Anshi. In a court audience with Shenzong, Wang admitted that he found no direct support in the Confucian Classics for honoring an ancestor without merit. However, with an emphasis on a sense of integrity (yili 義理), Wang argued that the factor of heredity should take precedence over merit in ancestral worship.30 By reiterating the general statement of the Book of Rites mentioned in Yuan Jiang’s memorial that “all things originate from Heaven; man originate from his ancestor” 萬物本乎天，人本 乎祖, Wang legitimized Xizu’s ritual status at a metaphysical level. In his response to Shenzong’s inquiry about the Primal Ancestor issue, he explained that despite a lack of merit, Gun 鯀, the legendary father of King Yu 禹, was still qualified for a central position in suburban Altar sacrifices. Wang used this example of Gun to criticize the meritocratic approach, arguing: “The king always associates his Great Ancestor with Heaven in sacrificial practices; therefore, he makes offerings to both Heaven and his Great Ancestor in suburban Altar sacrifices. If the ritual status of the Primal Ancestor was determined by merit, how could Gun—an ancestor without merit— be honored by his son Yu” 王者天太祖，故配天以祖。若以有功，則郊鯀豈得 為有功也?31 By bridging Xunzi’s 荀子 (313–238 BC) theory of induction between Heaven and the taizu (Founding Ancestor) with the meritocracy-versus-heredity context of the 1072 debate, Wang deliberately co-mingled the ritual designation of “Founding Ancestor” with that of the Primal Ancestor; thus, he formulated his own Primal-Ancestor approach through a distinction between the Heavenly Way (tiandao 天道) and the human Way (rendao 人道).32 According to Wang, the human Way regulated the ritual positions of departed ancestors based on their individual achievements; yet, the first ancestor as the most distant ancestor should be honored according to the Heavenly Way. Hence, the designation of the Primal Ancestor solely marked the farthest ancestor, and the farthest ancestor’s ritual status was solely defined by his genealogical priority.33 The fact that Wang Anshi was inclined to elevate Xizu’s status in the Song imperial lineage reflected his interest in rectifying imperial ancestral rituals based on a fundamental principle of regulation—the Heavenly Way. It is noteworthy that this 28. XCB, 240.5861. 29. XCB, 242.5891. Xizu’s wife had been posthumously bestowed the empress title of “wenyi” 文懿 in 960. 30. XCB, 240.5864. 31. XCB, 240.5864. 32. The phrase that “the King always associates his Great Ancestor with Heaven in sacrificial practices” 王者天太 祖 first appeared in the “Lilun” 禮論 chapter of Xunzi. I am indebted to John Knoblock’s translation here. John Knoblock, Xunzi: A Translation and Study of the Complete Works, 3 vols. (Stanford, CA: Stanford University Press, 1994), 3:58. The Qing scholar Wang Xianqian understood the character “天” as a verb, which meant “to parallel Heaven” (peitian 配天). Wang Xianqian 王先謙 (1842–1917), Xunzi jijie 荀子集解, 20 juan (Beijing: Zhonghua shuju, 1988), 13.349. 33. Wang’s relevant argument was best preserved in his essay, “Jiaozong yi” 郊宗議, in the extant Longshu edition 龍舒本 of Wang Anshi’s collected works. Wang Anshi, Wangwengong wenji 王文公文集, 100 juan in 16 vols. (Beijing: Zhonghua shuju, 1962), 31.1a–2a.

An Examination of the 1072 Controversy over the Song Primal Ancestor 71

principle of regulation, with regard to ancestral ritual, had originated from Wang’s broader conception of “regulatory standards” (fadu 法度).34 In 1059, Wang submitted a memorial to Emperor Renzhong 仁宗 (r. 1022–1063), in which he proposed fadu as the core idea of his envisioned reform.35 Wang’s conception of fadu stemmed from his criticism of the conventional practice of Statecraft and also from his perception of ritual as a key means of disciplining society. In his essay “Liyue lun” 禮樂論 (On ritual and music), Wang alleged that the essence of ritual and music was reflected in the ancient model of regulatory standards, legislative codes, and administrative policies (fadu xingzheng 法度刑政). For example, he wrote: “The way of ancient sage-kings that can be transmitted to the succeeding generations in words and that can be put into effect is the regulatory standards, legislative codes, and administrative policies, rather than the abstract motions of their spiritual enlightenment” 是故先王之道可以傳諸言、效諸行者，皆其法度刑政，而非神 明之用也.36 Although the Song government promulgated numerous laws and codes in Wang Anshi’s time, Wang still regarded the government as suffering from a lack of regulatory standards, because in many circumstances the laws and codes “failed to suit the regulatory standards of ancient sage-kings” 方今之法度，多不合於先王 之法度故也.37 In distinguishing himself from the conventional Confucian conception of moral politics, Wang advocated that the government should regulate society through proper regulatory standards on the basis of concrete ancient codes and policies. However, what constitutes the core element of proper regulatory standards, aside from legislative codes (xing 刑) and administrative policies (zheng 政)? In an essay entitled “Laozi lun” 老子論 (On Laozi), Wang discussed the principle and various practices of the “Way” (dao 道).38 His main thesis was that Laozi was wrong to proclaim a theory of non-interference. According to Wang, even though the Way itself was indeed obscure and abstract, ancient sages could still manage the world by following its traces, the “four techniques” (sishu 四術), including ritual, music, legislative codes, and administrative policies (liyuexingzheng 禮樂刑政). Because 34. I consulted James Liu’s translation of fadu here. The term fadu literally means laws and measures. As Liu explained, Wang Anshi offered the term a general meaning of institutional controls to govern and regulate. Hence, he translated fadu into “regulatory systems.” Liu, Reform in Sung China, 42. Since my interpretation emphasizes the conceptual correlation between Wang Anshi’s fadu and ancient ritual standards, I translate fadu as “regulatory standards” throughout my book. 35. Wang, “Shang renzong huangdi yanshishu” 上仁宗皇帝言事書, LCJ, 51:39.243. For the significance of this letter in the formation of Wang Anshi’s political thought, see Higashi, Ō Anseki shinpō no kenkyū, 955–57. 36. Wang, “Liyue lun,” LCJ, 51:66.423. Liu, Reform in Sung China, 42. In this text and some others following, I consulted Liu’s translations with some modifications. Some scholars have examined the usage of “Ancient Kings” in Wang Anshi’s writings and argued that for Wang and other Northern Song Confucians this word conveyed no special meaning. Zhang Yuan 張元, “Cong Wang Anshi de xianwang guannian kanta yu Songshenzong de guanxi” 從王安石的先王觀念看他與宋神宗的關係, in Song shi yanjiuji 宋史研究集 (Taibei: Zhonghua congshu bianshen weiyuanhui, 1993), vol. 23, 273–99. Yet, the “Liyue lun” reveals that Wang regarded the institutions and regulations set up by “Ancient Kings” to be something concrete and practical. 37. Wang, “Ni shangdian zhazi” 擬上殿劄子, LCJ, 51:41.261; Liu, Reform in Sung China, 43. 38. Wang, “Laozi lun,” LCJ, 51:68.435–36.

72

Empowered by Ancestors

the “four techniques” embodied the traces of the Way, a qualified ruler would take up his responsibility to “regulate the ten thousand things” based on these traces 所以成萬物者也.39 Comparing Wang’s wordings in his two essays, “On ritual and music” and “On Laozi,” the heart of what he termed the “regulatory standards” was essentially “propriety and music.” Expressed more theoretically, it referred to an effective governing technique related to the Three Dynasties (sandai 三代) that was speculatively constructed by Han and post-Han Confucians.40 In other words, Wang assumed that the Song government could reestablish the regulatory standards of ancient sage-kings through a revival of their ritual traditions. In this light, Wang Anshi’s endeavor to rectify Xizu’s ritual status in the Imperial Temple constituted an important element of his broader scheme to reconstruct ancient regulatory standards.41 Therefore, Wang’s conception of imperial ancestral rituals, as we have already seen in the 1072 Primal Ancestor controversy, was characterized by a spirit of revivalism that emphasized the ritual intent and related design of ancient sage-kings. Wang Anshi’s interest in reviving the ritual intent of ancient sage-kings was echoed by his political followers in the reformist camp, especially Yuan Jiang. Yuan’s memorial similarly emphasized “the ancient sage-kings’ genuine intention of revering ancestors” (xianwang zunzu zhiyi 先王尊祖之意).42 Although Yuan Jiang was generally seen by historians of later generations as a “petty man” (xiaoren 小人) who always flattered Wang Anshi and enthusiastically embraced Wang’s New Policies,43 Yuan successfully projected a positive image to some of his contemporaries. His superb talent in composing decrees and edicts even won applause from his political enemies.44 Su Song 蘇頌 (1020–1101), a brilliant writer and astrologist, composed a tombstone epitaph for Yuan Jiang. Su depicted Yuan as an exemplar of high-ranking officials. To be sure, Song tombstone epitaphs typically engage in the rhetoric of flattery.45 However, the tombstone epitaph of Yuan Jiang also contains pertinent facts; particularly it contains information about his official career and social expectations about him. More importantly, since Su Song was more neutral in his political stance and did not assume clear party affiliation,46 his epitaph can provide historians a 39. Wang, “Laozi lun,” LCJ, 51:68.436. 40. Wang, “Zhouli yixu” 周禮義序, LCJ, 51:84.529. 41. For a general introduction of Wang’s tendency to reconstruct ancient regulatory standards, see Higashi, Ō Anseki shinpō no kenkyū, 937–40. 42. SHY, Li 1:15, 38. 43. DDSL, 3:81.5b; SS, 31:343.10907. Tang Jiong 唐炯 in a memorial explicitly designated Yuan Jiang as the “slave” of Wang Anshi. XCB, 237.5778. 44. Indeed, Yuan Jiang as a gifted writer was praised by most celebrities of his time. DDSL, 3:81.5b; SS, 31:343.10907. 45. For general features of Song epitaphs, see Angela Schottenhammer, “Characteristics of Song Epitaphs,” in Burial in Song China, ed. Dieter Kuhn (Heidelberg: Edition Forum, 1994), 253–306. 46. Under most circumstances, Su Song was neutral in the political conflicts in the court. However, he and his family had a better relationship with the circle of reformists. Su Shuo, one of the aforementioned ritual reformists, was Su Song’s brother. Moreover, Su Song’s epitaph was written by Zeng Zhao 曾肇 (1047–1107), the brother of Zeng Bu 曾布 (1035–1107). As well known, Zeng Bu’s political stance was more inclined to the reformist side. See Zeng Zhao, “Zeng Susikong muzhiming” 贈蘇司空墓誌銘, Qufu ji 曲阜集, SKQS, 1101:3.31b–41b. Su Song had also served in the Court of Imperial Sacrifices.

An Examination of the 1072 Controversy over the Song Primal Ancestor 73

more balanced description of Yuan’s political life, including his compromising tendency in factional politics, as well as his administrative talents.47 This detailed epitaph, however, does not mention the 1072 debate, which reflects Su Song’s indifference toward Yuan Jiang’s intellectual interests. Likely, Yuan Jiang drafted the memorial that championed Wang Anshi’s opinion on the Primal Ancestor issue in the fourth month of 1072. Li Tao asserted that Yuan Jiang’s memorial was “instigated by Wang Anshi” 安石所主.48 His assertion is not without evidence. Yuan’s memorial not only followed along Wang’s reasoning about the impossibility of tracing any imperial ancestor prior to Xizu, but it also employed the same example that Wang had used in answering Shenzong’s inquiry about Xizu’s lack of merit—namely, Gun as an ancestor without merit having still been honored by his son Yu.49 Other edict drafters who endorsed Yuan Jiang’s 1072 memorial included Xu Jiang 許將 (1037−1111), Wang Yirou 王益柔 (1015−1086), Chen Yi 陳繹 (1021–1088), and Zeng Bu. They all shared similar reasoning with Wang Anshi and Yuan Jiang, too. These officials of the Two Drafting Groups agreed upon a consensus to support naming Xizu as the Primal Ancestor, in accordance with the ritual practice of ancient sage-kings. In fact, however, these officials actually came from different political backgrounds and had varied stances—a discrepancy that Xu Jiang’s case exemplifies. As one of Shenzong’s most reliable agents and a gifted diplomat, Xu Jiang started his court service in the Academy of Scholarly Worthies (jixianyuan 集賢院) and the Commission of Ritual Affairs. In both institutions, he accumulated a wealth of ritual knowledge and learning of the Classics.50 Politically, Xu leaned more toward the reformist camp led by Wang Anshi, who endorsed Xu’s promotion to the Secretariat in the Xining era.51 Yet in front of Emperor Shenzong, Wang Anshi once severely criticized Xu Jiang for his lack of skill and tact in drafting edicts.52 Moreover, Xu was dismissed from his administrative position as the Prefecture of Kaifeng (zhi kaifeng fu 知開封府) after he had been impeached by two partisans of the reformist camp, Cai Que 蔡確 (1037–1093) and Shu Dan 舒亶 (1041−1103).53 In his later years, Xu also dissuaded Emperor Zhezong from excavating Sima Guang’s tomb as suggested by the two notorious New Party ministers Zhang Dun 章惇 (1035–1105) and Cai Bian 蔡卞 (1048–1117).54 It is difficult to conclude that Xu had close affiliations with the reformist camp. The official dynastic Song History described Xu as a 47. Su Song, “Taizishaobao Yuanzhangjiangong shendaobei” 太子少保元章簡公神道碑, Suweigong wenji 蘇魏公 文集, Wang Congce 王從策 et al. (Beijing: Zhonghua shuju, 2004), 52.781–83, 786–87. 48. XCB, 240.5861. 49. XCB, 240.5839; 240.5864; Wang, “Miaoyi zhazi,” LCJ, 51:42.269. 50. SS, 31:343.10908. 51. Zeng Minxing 曾敏行 (1118–1175), Duxing zazhi 獨醒雜誌, in Quan Song biji 全宋筆記 (QSBJ), comp. Zhu Yian et al. (Zhengzhou: Daxiang chubanshe, 2006), Series 4:5.123–24. 52. XCB, 10:238.5789–90. 53. SS, 31:343.10908. 54. As Xu put it: “to disinter one’s tomb is an inappropriate behavior under a reign of prosperous virtue” 盜墓非 盛德事. DDSL, 3:96.5b; SS, 31:343.10910.

74

Empowered by Ancestors

person who was apt to change his mind (with relative ease) because of an absence of a “determined vision” (dinglun 定論).55 In many ways, Xu manifested himself as more of a compromising politician than a diehard follower of Wang Anshi’s New Policies. Among the ritual officials who had championed Wang Anshi’s and Yuan Jiang’s ritual interest in reinstating Xizu as the Primal Ancestor, it was Wang Yirou who was regarded as an anti-reformer. Wang Yirou, who also endorsed Yuan Jiang’s memorial, maintained good personal relations with celebrated conservatives, such as Sima Guang and Shao Yong 邵雍 (1011−1077).56 Indeed, Wang Yirou was perhaps the first page-proof reader of Sima’s voluminous work, the Zizhi tongjian 資治通鑑.57 The Qing narrative of Song scholarship even identified Wang Yirou as one of the informal disciples of Sima Guang.58 Geographical factors may also explain Wang’s close affiliation with the conservative camp; Wang was born and raised in Luoyang, which at the time was the fortified base of both the retired conservative Sima Guang and the pro-conservative thinker Shao Yong.59 Moreover, in a court audience, Wang Yirou criticized Wang Anshi for camouflaging his evil intent with skillful interpretations of the Classics.60 Although Wang Yirou’s political stance differed from that of Wang Anshi, he displayed a similar level of hostility toward conventional intellectual ideas as Wang Anshi had done when criticizing traditional commentaries and annotations. According to Yin Zhu 尹洙 (1001–1047), the young Wang Yirou was fond of writing essays railing against the memorization of traditional commentaries and previous interpretations; instead, he preferred a straightforward explanation of the main ideas of Confucius.61 The Song shi also quoted Yin Zhu’s description of Wang Yirou as a talented writer.62 However, it deleted several key phrases of Yin Zhu’s original text that dealt with Wang Yirou’s criticisms of conventional interpretations of Confucian Classics. Accordingly, later historical writings fostered a stereotype of Wang Yirou by his political stance alone. Historians have tended to ignore his anti-traditionalist spirit as well as his personal affiliations with Wang Anshi.63 If 55. SS, 31:343.10923. 56. DDSL, 2:53.4b; SS, 28:286.9634. 57. SS, 28:286.9635. 58. Wang Zicai 王梓材 (1792–1851), Feng Yunhao 馮雲濠, comp., Gaoben Song Yuan xuean buyi 稿本宋元學案 補遺 (Beijing: Beijing tushuguan chubanshe, 2002), 105. 59. Zhou Mi listed the twelve core members of the Luoyang group of conservatives. They were Fu Bi 富弼 (1004– 1083), Wen Yanbo 文彥博 (1006–1097), Xi Ruyan 習汝言, Wang Shanggong 王尚恭 (1007–1084), Zhao Bing 趙丙, Liu Ji 劉几 (1008–1088), Feng Xingsi 馮行巳 (1008–1091), Chu Jianzhong 楚建中 (1010–1090), Wang Shenyan 王慎言 (1011–1087), Wang Gongchen 王拱辰 (1012–1085), Zhang Wen 張問 (1013–1087), and Sima Guang. Zhou Mi 周密 (1232–1298), Qidong yeyu 齊東野語 (Beijing: Zhonghua shuju: 1983), 20.367. 60. SS, 28:286.9635. 61. Yin Zhu, “Song Wang Shengzhi zanxiang yishou” 送王勝之贊善一首, Henan ji 河南集, SBCK, 45:5.23. 62. SS, 28:286.9635. 63. Evidence of their personal relations, especially their exchange of poems, can be found in the extant collected works of Wang Anshi. See Linchuan xiansheng wenji 臨川先生文集 (Hong Kong; Zhonghua shuju, 1971), 7.134; 16.220–21; 19.242.

An Examination of the 1072 Controversy over the Song Primal Ancestor 75

Wang Yirou’s intellectual interests were taken into serious consideration, it is not difficult to understand why this political conservative supported Wang Anshi and Yuan Jiang, instead of following conventional ritual practice throughout the Primal Ancestor controversy. Chen Yi and Zeng Bu, the other two edict drafters who also endorsed Yuan Jiang’s memorial, have been commonly regarded as reformers who following Wang Anshi’s order.64 Like Yuan Jiang, Chen Yi was a celebrated literatus whose literary ability attracted Emperor Yingzong’s attention. In time, he would be assigned to the examining editor of Court Records (shilujiantao 實錄檢討).65 In a memorial to Yingzong, Chen succinctly argued that one of the five “principles of governance” (guoshi 國是) should be a thorough study of antiquity (qigu 稽古), stating that “learning from antiquity can help understanding the present” 觀古所以知今.66 Chen’s interest and position in the 1072 ritual debate, then, is understandable when viewed in light of the rectification of Xizu’s Primal Ancestor position and its symbolization of the genuine pursuit toward antiquity. In general, Chen Yi regarded the rectification of titles and positions not as a mere quibble over words but rather as the representation of correct social order.67 Despite Chen Yi’s adherence to the reformist camp, he still had connections with the conservatives, such as the celebrated philosopher Shao Yong, who had asked Chen Yi to compose an epitaph for his father, Shao Gu 邵古.68 Likewise, Zeng Bu, despite his strict adherence to Wang Anshi and the New Policies, maintained lifelong friendship with his examination cohort (tongnian 同年) Su Shi 蘇軾 (1037– 1101), who frequently criticized Wang’s New Policies. Ritual officials who lent support to Wang Anshi’s plan also came from different political backgrounds; these included: Yang Jie, Song Chongguo, Zhou Mengyang, and Zhang Heng. Zhang Heng was the nephew of the reformer Zhang Dun and did not share his uncle’s political stance nor membership in Wang Anshi’s alliance. On one occasion, Zhang criticized an institutional problem in the personnel system that was caused by the Elective Bureau of Junior Officers (sanbanyuan 三班院). Although Wang Anshi supported the Elective Bureau, Zhang Heng insisted on his own views and confronted Wang Anshi directly in the Secretariat-Chancellery.69 Additionally, Zhang Heng was portrayed as the disciple of Chen Xiang 陳襄 64. SS, 30:329.10614–5; 39:471.13714–17. 65. SS, 30:329.10614. 66. Chen Yi’s memorial was lost. Fortunately, in Su Song’s collected works there is an epitaph of Chen that preserves some fragments of this memorial. Su Song, “Taizhong dafu chengong muzhiming” 太中大夫陳公墓誌 詺, Suweigong wenji, 80.911–12. 67. Chen Yi has expressed his idea about the relationship between titles, positions, and social order in an essay titled “Xinxiu dongfu ji” 新修東府記, a memorial writing that was dedicated to the newly constructed office of the Secretariat-Chancellery. This essay was collected in Lü Zuqian’s 呂祖謙 (1137–1181) Huangchao wenjian 皇朝文鑑, Lü Zuqian quan ji 呂祖謙全集, Huang Linggeng 黃靈庚 ed. (Huangzhou: Zhejiang guji chubanshe, 2008), 13:81.482–84. 68. Chen, “Shao Gu muming” 邵古墓銘, Huangchao wenjian, 143:743–44. 69. SS, 31:347.11008.

76

Empowered by Ancestors

(1017–1080), a Fujian conservative who vigorously criticized Wang Anshi’s Green Sprouts Policy (qingmiao fa 青苗法).70 Currently, no detailed description is extant in regard to what Zhang might have learned from Chen, but we do know that Chen Xiang had high expectations for Zhang Heng. In a letter to his early disciple, Chen encouraged Zhang to search for the intrinsic correlation between the Classics, the Sages, and Heaven.71 What Chen Xiang taught Zhang Heng resembled Wang Yirou’s emphasis on the main idea of ancient sages. In Chen Xiang’s opinion, studies of ancient rituals serve as an effective means through which one can approach the main ideas of ancient sages. Indeed, Fujian scholarship represented by Chen Xiang had been obsessed with ancient rituals since the 1040s.72 It is possible that Chen Xiang’s conception of ancient rituals may have later influenced Zhang Heng in his agreement with Wang Anshi about the rediscovery of the ritual intent of ancient sage-kings.73 Compared with Zhang Heng, Song Chongguo and Zhou Mengyang appeared to be less associated with factional politics in the 1070s. Both Song and Zhou had served in the Commission of Ritual Affairs in the 1060s.74 Song Chongguo was the son of Song Xiang 宋庠 (996–1066), the chief councilor of Emperor Renzong and also a ritual expert. In fact, Song Xiang was one of the main advocates for building private ancestral temples in the 1050s. In a 1050 memorial, Song Xiang stated: “As for the construction of ancestral buildings, the arrangement of the zhaomu sequence of ancestors, and the distinguishing of the noble and lowly, all have filial piety as their aim. Although someone may err by going too far in these practices, they are just overdoing filial piety” 夫建宗祏、序昭穆、别貴賤之等，所以為孝，雖有過 差，是過於為孝.75 Clearly, Song Xiang conceived the idea of filial piety as a unique criterion of ancestral worship. Accordingly, it would follow that there is nothing wrong with elevating ancestors’ ritual statuses, either. As the son of Song Xiang, Song Chongguo was in this sense preoccupied with his father’s notion of practicing “excessive” filial piety. In the 1072 debate, he argued for posthumous worship of Xizu as the Primal Ancestor in order to illustrate the origin and beginning of the Song royal lineage.76 The case of Yang Jie is more complicated. As a literatus who came from a region with profound Buddhist influence, Yang’s thought bears a clear Buddhist imprint.77 70. Chen Xiang, Guling ji 古靈集, SKQS, 1093:3.7a–16b. Zengbu Songyuan xuean 增補宋元學案, Huang Zongxi 黃宗羲 (1610–1695) and Quan Zuwang 全祖望 (1705–1755) comp. (Taibei: Zhonghua shuju, 1970), 5.b. The editors of Siku quanshu greatly praised Chen for his courage to oppose Wang Anshi’s opinion at the heyday of the latter’s power in the Xining period. See Chen, Guling ji, 1093: tiyao 提要.1. 71. Chen, “Song Zhangheng xiucai xu” 送章衡秀才序, Guling ji, 1093:18.9a–b. 72. For an interesting example of the Fujian scholars’ practice of ancient rituals, see the case of Chen Lie 陳烈 in performing funeral rites. Zhang Shizheng, Juanyou zalu 倦遊雜錄, in Song Yuan biji congshu 宋元筆記叢書 (Shanghai: Shanghai guji chubanshe, 1993), 6. 73. XCB, 240.5860. 74. XCB, 203.4911. 75. XCB, 169.4071–2; SHY, Li 1:12.1. 76. XCB, 240.5858. 77. Huang Qijiang 黃啟江, “Bei Song jushi Yangjie yu fojiao: jianbu Song shi Yangjie benzhuan zhique” 北宋居士

An Examination of the 1072 Controversy over the Song Primal Ancestor 77

His expertise in ritual studies, however, has been commonly recognized by his contemporaries. According to Zhao Shican 趙士虨, the editor of Yang Jie’s collected works, Yang was a brilliant poet. His poetry was deeply influenced by Ouyang Xiu, Wang Anshi, and Su Shi.78 Yang was also an “academic fellow” (jiangyou 講友) of the early Song pedagogical reformist Hu Yuan 胡瑗 (993–1059).79 For Yang Jie, then, his intellectual interests allowed him to freely transcend the political boundaries of factionalism. His proficiency in ritual covered a myriad of aspects, especially those in relation to ritual music for court sacrifices.80 Thus in the 1072 debate, Yang, together with Song Chongguo and Zhou Mengyang, made a claim for Xizu’s ritual supremacy in the Imperial Temple. However, they had also worked out a compromise that suggested the permanent preservation of Song Taizu’s tablet in the Imperial Temple as a way of symbolizing Taizu’s great contributions toward establishing the dynasty. The opposing side of the 1072 debate, which advocated the rectification of Song Taizu’s ritual status as the Primal Ancestor, found its roots predominantly in the conservative camp. Han Wei, Sun Gu, Liang Tao, and Wang Jie had all opposed Wang Anshi’s New Policies. Both Han Wei and Sun Gu had disagreed with Wang Anshi’s promotion to grand councilor at the beginning of Shenzong’s reign. Han Wei, formerly a close friend of Wang Anshi, became an anti-reformist in 1072 and had been recommended, instead of Wang Anshi, as a possible candidate for the grand councilor position by the anti-reform league, composed of Sima Guang, Han Qi, and Lü Gongzhu 呂公著 (1018–1089).81 Hence it came as no surprise that Han Wei was later labeled a core member of the conservative Yuanyou 元祐 (1086–1094) Party, with his name inscribed on the notorious Stele of Yuanyou Partisans (yuanyou dangjibei 元祐黨籍碑)82 erected by Cai Jing 蔡京 (1047–1126), the brother of Wang Anshi’s son-in-law.83 Similarly, Sun Gu was also an unflinching conservative and staunch member of the “anti-Wang Anshi” clique post-1070s, despite his early personal friendship 楊傑與佛教—兼補宋史楊傑本傳之缺, Hanxue yanjiu 漢學研究 (2003.6): 253–77. 78. Zhao Shican, “Wuwei ji xu” 無為集序, in Wuweiji jiaojian 無為集校箋, annotated by Cao Xiaoyun 曹小雲 (Hefei: Huangshan shushe, 2014), 1. 79. Gaoben Song Yuan xuean buyi, 27. 80. Yang, Wuweiji, 1099:15.8a–16a. 81. SS, 31:341.10874. 82. Although the Stele itself did not exist anymore, the names on it have been recorded by later scholars. Wang Chang 王昶 (1725–1806) preserved a list of Yuanyou partisans with their names inscribed on the Stele in his encyclopedic collection of epigraphy, the Jinshi cuibian 金石萃編. This annotated list is entitled “Yuanyou dangjibei xingmingkao” 元祐黨籍碑姓名考. See Wang Chang, Jinshi cuibian, Xuxiu siku quanshu 續修四庫 全書 (hereafter XXSK), compiled by by Xuxiu Sikuquanshu Bianzuanweiyuanhui 續修四庫全書編纂委員 會 (Shanghai: Shanghai guji chubanshe, 1995), 890:144.13a–33a. For a detailed description of Han Wei’s and Sun Gu’s careers as Yuanyou Partisans, see Lu Xinyuan 陸心源, Yuanyou dangrenzhuan 元祐黨人傳, XXSK, 517:1.12a–13b. 83. For Cai Jing’s life and his role in the Northern Song political struggles, see SS, 39:472. 13721–28. It is noteworthy that the Song shi record of Cai Jing’s life is inaccurate in many aspects. For a detailed analysis of Cai’s life, see Charles Hartman, “A Textual History of Cai Jing’s Biography in the Songshi,” in Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, ed. Patricia Ebrey and Maggie Bickford (Cambridge, MA: Harvard University Asian Center, 2006), 517–64.

78

Empowered by Ancestors

with Wang Anshi.84 When Shenzong asked for Sun Gu’s advice about Wang Anshi’s promotion to the grand councilor, he explicitly expressed disagreement.85 In 1072, Sun served as the edict drafter of the Tian Zhang Pavilion (Tianzhangge daizhi 天章 閣待制), a court position that provided him numerous chances to discuss state policies with the emperor. Sun Gu’s memorial in the 1072 debate was highly applauded by an important veteran conservative, Han Qi, mainly because it disagreed with the dominating opinion of Wang Anshi.86 More importantly, it severely criticized those officials who assumed that the ritual intent of ancient sage-kings was a static and unchangeable tradition. In Sun’s opinion, the ancient rituals were rooted in human nature and thus should be adjusted to suit contemporary needs. Sun Gu charged “those officials who overwhelmingly champion the ancient rituals and institutions are actually ignoring contemporary contexts” 所謂慕古而違當世之宜者也.87 Thus, Sun argued that concrete historical precedents from Han (202  BC–AD  220) and Tang experience were more compelling than the imagined practice of the Three Dynasties. Wang Jie’s intellectual interests coincided with Sun Gu’s, both of which cast doubt on rigid interpretations of ancient rituals. In contrast to Sun Gu’s emphasis on Han and Tang cases, Wang Jie quoted examples of ancient sage-kings to demonstrate why Taizu should be honored as the Primal Ancestor. It is worth noting that Wang Jie acknowledged that the Jifa 祭法 text in the Book of Rites might not reflect the ritual practice of ancient sage-kings.88 In a similar fashion, Sun Gu also doubted the validity of Jifa as a reliable source.89 Yet despite his hostility toward the New Policies, Wang Jie maintained a good relationship with Wang Anshi on a personal level and genuinely admired his scholarship. The appointment letter Wang Anshi later wrote for Wang Jie’s promotion, before the New Policies, demonstrated how Wang Anshi valued him to a certain degree. In the appointment letter, Wang Anshi applauded Wang Jie as a genuine

84. DDSL, 3:81.6a–7a. The Ming historian Ke Weiqi 柯維騏 (1497–1574) stated that Sun Gu would rather sacrifice his private friendship with Wang Anshi to avoid flattering Wang and taking Wang’s bait. Ke Weiqi, Song shi xinbian 宋史新編, Ershisishi waibian 二十四史外編, 152 vols. (Tianjin: Tianjin guji chubanshe, 1998), 96:115.7a. For a succinct introduction of the historiographical value of Ke’s Song shi xinbian, see Chan Hok-lam 陳學霖, “Ke Weiqi Songshi xinbian shuping” 柯維騏宋史新編述評, in Songshi yanjiuji 宋史研究集 (Taibei: Guoli bianyiguan, 1990), vol. 20, 489–526. 85. DDSL, 3:81.6a–6b; SS, 31:341.10874–75. 86. DDSL, 3:81.6a; SS, 31:341.10875. 87. Sun repeated this phrase twice in his memorial. XCB, 240.5841; 240.5844. 88. SHY, Li 1:15.41. 89. Sun argued that after the Qin (221–207  BC) destruction of ancient Confucian norms and text, the ritual Classics that were left were incomplete; sections of the text of the Classics were intermingled with the private writings and commentaries of Han Confucians. XCB, 240.5843. Wang’s and Sun’s suspicions regarding the record of Jifa largely reflected the sense of uncertainty in the minds of most mid-Song ritual officials. Suspicions on the authenticity of the Jifa continued to grow throughout the transition period from Xining and Yuanfeng to the end of Northern Song. A late Northern Song scholar, Lü Benzhong 呂本中 (1084−1145), who came from a family of scholastic tradition, doubted the Jifa text as Han fabricated. See Lü Benzhong, Ziwei zashuo 紫微雜說, QSBJ, Series 3:6.65–66.

An Examination of the 1072 Controversy over the Song Primal Ancestor 79

Confucian whose scholarship often conveyed a sense of integrity.90 Despite their different political stances, Wang Anshi still composed an elegy for Wang Jie after the latter’s death in 1076. In the elegy, Wang Anshi recalled his lifelong friendship with Wang Jie, which began from a tender age.91 Hence, it is not surprising that later historical narratives categorized Wang Jie as an “academic fellow” of Wang Anshi.92 Liang Tao was often in disagreement with his colleagues at the Commission of Ritual Affairs throughout his long career of ritual service. Although Liang Tao tended to downplay Xizu’s ritual status in the Imperial Temple, he also expressed his dissatisfaction with the Song arrangement of temple rituals. Unlike Han Qi and Sun Gu, who felt that the conventional setting of the Song Imperial Temple was decent enough, Liang had a longing for a perfect temple setting and relevant rituals that could match ancient models.93 Liang also took a more complicated stance on the argument surrounding ancient rituals and contemporary practices. Politically, Liang was conservative; intellectually, he supported the revival of ancient rituals. In terms of ritual interests, the gap separating Liang Tao and Wang Anshi was potentially narrower than that of Liang Tao and Han Wei, or Sun Gu. If political stance alone cannot explain the discrepancy between officials with respect to ritual understanding, then what else ought to be the key factor? The answer lies in the aforementioned ritual revivalism found in Wang Anshi’s discussion about the fadu of ancient sage-kings. Indeed, an advocacy for ancient rituals modeled on the Three Dynasties had been prevalent since the mid-eleventh century. Renzong’s reign, especially after the Qingli era, witnessed an increasing trend of discussing and promoting ancient rituals. The compilation of the Taichangyingeli 太 常因革禮 in 1065 clearly marked the interests of Song ritual officials in tracing the cultural legacy of the Three Dynasties.94 Although Ouyang and other ritual officials who had compiled the Taichangyingeli failed to stipulate a comprehensive ritual code based on the ritual intent of ancient sage-kings, they successfully launched a campaign that led to a conceptual association between Statecraft and ritual reforms in later periods.95 By integrating the 1072 Primal Ancestor debate into the context of ritual revivalism in the late eleventh century, we see how discussions in the post-Renzong period were characterized by the “gu” 古 (ancient) discourse and an increasing interest toward the revival of ancient rituals. For example, in both Sun Gu’s and Chen Yi’s 90. Wang, “Wang Jie mishucheng zhi” 王介秘書丞制, LCJ, 51:51.322. Despite Wang Jie’s personal affiliations with Wang Anshi, he disagreed with Wang’s reform and insisted on his political stand in an unflinching way. Gu, Wangjinggong nianpu, 3.1982. 91. Wang, “Wang Zhongfu xueshi wanci” 王中甫學士挽詞, LCJ, 51:35.229. 92. Wang Zicai named Wang Jie as an academic friend of Wang Anshi. Gaoben Song Yuan xuean buyi, 865, 874. 93. SHY, Li 1:15, 44. 94. Ouyang Xiu, Taichangyingeli, in Congshujicheng chubian 叢書集成初編 (Shanghai: Shangwu yinshuguan, 1936), 1043: preface.1. 95. Cheung Hiu Yu 張曉宇, “Cong bian Tang zhi li dao zhangxian xinyi: Bei Song qianqi xinli wenti yu Taichang yingeli” 從變唐之禮到章獻新儀──北宋前期新禮問題與太常因革禮. Hanxue yanjiu 漢學研究 37.1 (2019.3): 50–52.

80

Empowered by Ancestors

memorials written in the 1072 debate, “gu” referred to the early reign of the Three Dynasties. However, the difference between Sun’s and Chen’s understanding of “gu” epitomized the scholar-officials’ different conceptions of the cultural legacy of ancient sage-kings (xianwang zhi zhi 先王之治). In the heyday of Wang Anshi’s political reforms, it was certain that officials had different political stances toward the New Policies. However, given the growing reflection on the “gu” discourse, the ancient fadu, the differences in the ritual understandings among the participants of the 1072 debate were associated more with their different conceptions of ancient rituals than with their political stances. Therefore, the two worlds of political conflicts and intellectual interests did not necessarily overlap with each other. Most Song ritual officials appeared to have crossed the boundaries of established partisan politics in the discussions of the 1072 ritual debate. Even Han Wei and Sun Gu, who obviously championed the elevation of Taizu’s ritual status, quoted examples from the ritual legacy of ancient sage-kings (xianwang zhili 先王之禮) to justify their merit-based approach. After all, the discussions over the Primal Ancestor issue always involved a negotiation process in which different conceptions of ancient rituals competed for equal shares. The cultural superiority of ancient rituals was indisputable in the eyes of Song officials; the only question was to what extent these rituals could be reimplemented in their own times. It is noteworthy that the discussions on rituals were apparently not as intense as the famous political struggles often highlighted in the late Northern Song, thus the intellectual discrepancies underlying these discussions also appeared less obvious. To better illustrate these intellectual differences among the scholar-officials during the 1072 debate, we may categorize them into three basic groups: (a) those who highlighted Taizu’s merit and contributions for his eligibility as the Primal Ancestor; (b) those who regarded the rectification of Xizu’s ritual status as a revival of the original intention of ancient rituals; and (c) those who profusely quoted from the Classics to argue for a negotiated resolution of the tension between ancient ritual intention and contemporary utility. Table 3.1 (see p. 81) lists all three categories of officials and ritual officials with their basic political stances. Certainly, the above categories do not necessarily represent a fixed set of characteristics. Categorization, after all, only serves as an analytical tool to bring attention to the selected attributes.96 In general, those like Han Wei, Sun Gu, and Wang Jie, who supported the notion of Taizu becoming the Primal Ancestor, generally embraced a merit-based argument when defending Taizu’s ritual superiority. As Sun Gu and Wang Jie repeatedly emphasized, the previous practice of leaving the Primal Ancestor position vacant (to reserve it for Taizu’s spirit tablet) should be followed

96. James Liu once admitted that his classification of Song officials according to political behavior just indicated “a range of behavioral patterns which overlap.” Liu, Reform in Sung China, 71. It is also true here when intellectual criteria are applied. Officials, such as Liang Tao and Zhang Heng, can never be entirely reduced to stereotypes when we consider their different and even contradictory speeches and writings.

An Examination of the 1072 Controversy over the Song Primal Ancestor 81 Table 3.1: Officials Involved in the 1072 Primal Ancestor Debate: Ritual Interests and Political Stance Pro-Taizu officials

Pro-Xizu officials

Ritual officials who embraced a negotiated plan

Han Wei (C)

Wang Anshi (R)

Su Zhuo (~R)

Sun Gu (C)

Yuan Jiang (R)

Song Chongguo (R/C)

Wang Jie (~C)

Xu Jiang (R/C)

Zhou Mengyang (R/C)

Chen Yi (~C)

Yang Jie (R)

Wang Yirou (~C)

Liang Tao (C)

Zeng Bu (~R)

Zhang Gongyu (?) Zhang Shiyan (~R) Zhang Heng (~C)

Note: R: politically reformist; C: politically anti-reformist, conservative; ?: political stances unknown; ~R: pro-reformist; ~C: pro-conservative; R/C: political stance shifting between reformist and conservative.

because it was the “original design of Song ancestors” (zuzong zhiyi 祖宗之意).97 They argued that changes in the Primal Ancestor designation from Taizu to Xizu would cause a shift in the state’s principles of governance and would also increase government expenditure—a conservative point of view that was articulated by Han Wei and Sun Gu.98 Although ancient rituals of the Three Dynasties were generally regarded as prominent, their applicability was doubtful in the eyes of officials of the pro-Taizu camp. The conversation between Su Zhe 蘇轍 (1039–1112) and Lü Dafang 呂大防 (1027–1097) about whether the South Altar and the North Altar sacrifices should be combined exemplified the tension between ancient rituals and the conventional practices of Song ancestral rituals. When Lü Dafang questioned Su Zhe about whether or not an integrated suburban Altar sacrifice would fail to comply with the ritual intent of the Three Dynasties, Su answered: Nowadays, not only the practice of a combined state sacrifice but also other ritual practices discard the Three Dynasties’ model by following Han or Tang precedents. For instance, in ancient times, the Son of Heaven had seven Imperial Temples; today, the architectural complex of the temple has been modified to one single temple with nine chambers. In ancient times, temple sacrifices made offerings only to emperors and empresses; today, all the wives of the emperor could receive

97. SHY, Li 1:15, 44. 98. For the Song conservative attitude toward government expenditure, see Peter Bol, “Government, Society, and State: On the Political Visions of Ssu-ma Kuang and Wang An-shih,” in Ordering the World: Approaches to State and Society in Sung Dynasty China, ed. Robert Hymes and Conrad Schirokauer (Berkeley: University of California Press, 1993), 128–92; also Ji Xiao-bin 冀小斌, Politics and Conservatism in Northern Song China: The Career and Thought of Sima Guang (A.D. 1019–1086) (Hong Kong: Chinese University Press, 2005), 10–19, 181–85.

82

Empowered by Ancestors offerings. What matters is to meet contemporary needs; there is no need to follow the rules of the Three Dynasties in a stubborn way.99 今捨三代而從漢、唐者，非止一事矣：天子七廟，今乃一廟九室；廟祀一帝 一后，今諸后並配。事各適時，豈必三代！

In contrast to the officials in the pro-Taizu camp, pro-Xizu officials considered the elevation of Xizu’s ritual status to that of Primal Ancestor as being a symbolic revival of the ritual intent of ancient sage-kings. We do not know if there was a clear correlation between the rectification of Xizu’s ritual status in 1072 and Wang Anshi’s promotion of the New Policies during that period. However, in effect, Wang enhanced Xizu’s prestige at the expense of Taizu’s ritual status. By opposing Taizu’s elevation to Primal Ancestor status, Wang and his followers had, regardless of their original intention in initiating the 1072 debate, symbolically undermined Taizu’s ritual status as the ultimate authority of conventional practices in Song ancestral rituals. Wang Anshi himself might not have regarded the launch of the 1072 debate as a direct challenge to conventional ritual practices at his time, though he evidently did attach less importance to Song precedents than those of the ancient sage-kings in the Primal Ancestor issue. In an early memorial concerning other ritual norms in the Imperial Temple, Wang questioned the necessity to follow the model of the Three Dynasties and proposed maintaining the conventional practice established by previous Song emperors.100 In this light, Wang’s inclination toward Xizu in the 1072 debate might be attributed more to his hostility toward advocates of Taizu’s ritual superiority rather than to his personal views on conventional ritual practices. The pro-Taizu camp was, after all, mostly composed of Wang Anshi’s political opponents. Compared to the highly factionalized officials like Wang Anshi, Han Wei, and Sun Gu, other ritual officials who were involved in the 1072 debate, regardless of their political stances, reflected a general interest which sought compromise based on ancient rituals. The young ritual official Su Zhuo, who suggested keeping Xizu as the Primal Ancestor, invented a new ritual model based on a genealogical record from the Shiji. In his ritual model, Xizu’s spirit tablet would be attached to the tablets of the two most recent ancestors of the Zhao 趙 lineage—Baiyi 柏翳 and Zaofu 造父, in the state’s suburban Altar sacrifices. Su admitted that this kind of ritual practice “had not been recorded in the Classics; however, at least it would avoid offending ancestral spirits” 雖不合經，而免於瀆祖.101 By the same token, the other pro-Xizu ritual officials, Song Chongguo, Zhou Mengyang, and Yang Jie, suggested that the court separately designate Xizu as the Primal Ancestor and Taizu as the Great Ancestor. They argued that the temples of Taizu, Taizong, and Zhenzong 99. Su Zhe, Longchuan lüezhi 龍川略志 (Beijing: Zhonghua shuju, 1982), 8.51–52. Emphasis mine. 100. Wang, “Yi rumiao zhazi” 議入廟劄子, LCJ, 51:41.263. 101. SHY, Li 1:15, 46.

An Examination of the 1072 Controversy over the Song Primal Ancestor 83

should be preserved forever to be in line with Shang and Zhou practices of worshipping their ancestors as permanent ancestors (Shang: Taijia 太甲, Taiwu 太戊, Wuding 武丁; Zhou: King Wen and King Wu).102 Furthermore, although Liang Tao, Zhang Shiyan, and Zhang Gongyu quoted Renzong’s 1067 edict to argue for Taizu’s legitimacy as the Primal Ancestor, they also attempted to adopt an ancient ritual practice to ease the vexation that might be caused by Xizu’s absence in the Imperial Temple.103 Specifically, Liang and the two Zhangs suggested building a new “lateral temple” (biemiao 別廟) to house Xizu’s tablet. They claimed that this practice was in line with the Zhou practice, recorded in the Rituals of Zhou, of setting a tiao-preservation bureau (shoutiao 守祧) to accommodate the spirits of ancient ancestors.104 In the final analysis, the reasoning found in all three categories of these ritual officials converged toward an emphasis of ancient rituals. Rather, it was their different interpretations of the applicability of “ancient rituals” that distinguished them from one another in the 1072 debate. Hence, both sides were Confucian scholarofficials who sought to access and apply the rituals and teachings of ancient sages— in contrast to the “conventional Confucians” during the Song, who emphasized continuity of Han and Tang Confucianism and ritual practices.

Concluding Remarks Generally, historians look for political implications from the changes of arrangement and settings in the Imperial Temple. Considering Wang Anshi’s hostility toward the conventional Song regulatory standards at his time,105 his insistence in designating Xizu as the Primal Ancestor could be interpreted as seriously political. As Taizu had been commonly regarded as the chief architect of the conventional Song regulatory standards, by replacing Taizu’s tablet with Xizu’s tablet, Wang Anshi might attempt to undermine the symbolic authority of Taizu that was associated with the existing regulatory standards. In other words, the removal of Taizu’s tablet from the Primal Ancestor position might indicate a fundamental change in real policies. Nevertheless, without solid evidence it is difficult to prove this political implication, let alone the political stances of ritual officials involved in the 1072 debate. Instead of emphasizing the political meaning of the 1072 debate, my analysis demonstrates that the debate itself was a relatively independent event in which factional concerns appeared less relevant. The divergent views of ritual officials in the 1072 102. SHY, Li 1:15, 48. 103. XCB, 240.5850–52. 104. XCB, 240, 5852; also see SHY, Li 1:15, 45. For the original text concerning the functions of the shoutiao bureau, see Zhouli zhushu, TSZSSSJ, 2:17.168; 2:21.211. 105. Song scholar-officials usually referred to the conventional regulatory standards coined by Song Taizu and Song Taizong as “ancestral codes” (zuzong zhifa 祖宗之法). Nevertheless, as Deng Xiaonan 鄧小南 has convincingly demonstrated, a comprehensive understanding of “ancestral codes” among Song scholar-officials was only coined during Emperor Renzong’s reign. Deng Xiaonan, Zuzong zhifa: Beisong qianqi zhengzhi shulüe 祖宗之法：北宋前期政治述略 (Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2006), 340–423.

84

Empowered by Ancestors

debate actually revealed an intellectual perspective of Song factional politics that was associated more with different conceptions of antiquity and ancient ritual practices than with political struggles. In one of his early memorials, Wang Anshi proclaimed: “concerning the world today, my humble opinion is that we still need to restructure the government and society according to the regulatory standards of ancient sage-kings, yet take warning from the rulers of the Middle Periods” 然 竊恐今日之天下，尚宜取法於先王，而以中世人君為戒也.106 The 1072 debate revealed a latent collective consciousness underlying the mindset of many Song ritual officials after the mid-Song: a consciousness that aimed to restructure the state orthopraxy to reconstruct the ritual practices of ancient sage-kings that were recorded in the ritual Classics.

106. XCB, 217.5287. Wang Anshi here also implicitly criticized the contemporary regulatory standards of his time, arguing that it followed the bad examples of the Han and Tang rulers of the Middle Periods.

4 Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

The 1072 debate over Xizu’s ritual status in the Imperial Temple and the formal recognition of Xizu’s Primal Ancestor position foreshadowed a series of ritual rectification movements from 1077 onward. Officials and ritualists who advocated Xizu’s Primal Ancestor position were the forerunners who called for sweeping reforms of court sacrificial rituals during the late Xining and the succeeding Yuanfeng 元豐 era (1078–1085). Since the reforms were primarily concerned with the formulation of sacrificial rituals held at the suburban Altar and the Imperial Temple, they had been referred to as the “Yuanfeng regulations on the suburban Altar and temple rituals” 元豐郊廟奉祀禮文 in Song official records. The year 1078 marked the initiation of the Yuanfeng ritual reforms and witnessed the establishment of a new ritual department within the Court of Imperial Sacrifices, the Department of Prescribed Altar and Temple Rites (taichang jiaomiaofengsi xiangding liwensuo 太常郊廟奉祀 詳定禮文所, hereafter cited as DPATR).1 In the same year, the court expanded the Administrative Office of South Altar Affairs (tidian nanjiao shiwusuo 提點南郊事 務所) by subsuming into it the Editorial Board of the Regulations on the Hall of Brightness (bianxiu mingtangshisuo 編修明堂式所).2 Institutionally, these changes set the stage for the succeeding ritual reforms involving the meticulous discussions about a number of altar and temple rituals, including the concrete performance of court sacrifices, the symbolic meaning of the South Altar as a ritualized space, and ritual utensils and offerings used in altar and temple sacrifices.3 In Chapter 3, I argued that the discrepancy among ritual officials in the 1072 Xining debate was primarily associated with their different understanding of ancient rituals. That discrepancy narrowed at the time of the Yuanfeng ritual reforms. Although the ritual officials in the DPATR came from different political backgrounds, they reached a consensus on the presupposition of reviving ancient rituals. None of them considered their contemporary practice of altar and temple rituals to be decent and satisfactory. Therefore, the crucial issue in the Yuanfeng

1. XCB, 287.7012. 2. XCB, 287.7029. 3. SHY, Li 2.28.55. Also see XCB, 291.7124; 292.7136–37; 292.7138–39.

86

Empowered by Ancestors

ritual reforms shifted from the question of how ancient rituals should be understood to how they should be performed. In 1079, the Yuanfeng ritual reforms reached the culmination, when a significant debate concerning the ritual order of ancestors in the Imperial Temple broke out.4 In this debate, three ritual officials from the DPATR—Lu Dian 陸佃 (1042– 1102), Zhang Zao 張璪 (d. 1093), and He Xunzhi 何洵直 (jinshi. 1078)—disputed the ritual sequence of zhaomu of Song imperial ancestors in the temple.5 Notably, these ritual officials were generally identified with the reformist faction in traditional narratives. Their conceptions of the zhaomu of Song imperial ancestors not only shaped later understanding of spatial arrangement of temple ancestors but also the ritual representation of familial relations reflected in that arrangement.

DPATR and the Yuanfeng Scheme of the Imperial Temple Emperor Shenzong initiated his reforms on officialdom and bureaucracy at the beginning of the Yuanfeng era.6 Meanwhile, he turned his attention to imperial sacrificial rituals—an aspect that had disappointed him since the day of his enthronement. In 1078/1, Shenzong launched the Yuanfeng ritual reforms at the suggestion of a remonstrator named Huang Lü 黃履 (1030–1101).7 Shenzong regarded the ritual reforms as part of a general revival of ancient Statecraft—a practice inspired by Wang Anshi’s active reading of the Rituals of Zhou.8 It is recorded in the Wenxian tongkao 文獻通考 that a ritual official named Yang Wan 楊完 compiled a thirtyjuan 卷 collection to document all the memorials, regulations, and ritual writings 4. The Song shi dated the debate to 1078. SS, 106.2574. According to XCB, in 1079/1, Emperor Shenzong appointed Lu Dian to the DPATR to revise altar and temple rituals. Before 1079, Lu Dian served as one of the editors of the official edition of Shuowen jiezi 說文解字 (Explaining Graphs and Analyzing Characters). XCB, 296.7195. Therefore, it was only possible for Lu to participate in the debate after he joined the DPATR. Zhang Zao was only hired into the DPATR in 1079/1. So, I date the debate to 1079. XCB, 302.7349. 5. Compared to the 1072 Primal Ancestor debate, not much scholarly attention has been devoted to the 1079 debate and its influence. As far as I know, I am the first scholar who has presented related research findings. See Cheung Hiu Yu, “The 1079 zhaomu Debate: The Song Ritual Controversy over Ancestral Rites,” Western Branch Meeting of the American Oriental Society, 2012. After conducting more research on the 1079 debate, I integrated my revised argument into my dissertation, which was published online in 2015. I am pleased to learn that some Chinese scholars have recently started to pay attention to this debate when I came across an essay published in 2017. Hua Zhe 華喆, “Fuzi yilun: Bei Song Yuanfeng zhaomu zhiyi zai pinjia” 父子彝倫：北宋元豐昭穆之議再評價, Zhongguo zhexueshi 中國哲學史 (2017.3): 18–29. In his article, Hua affirmed some of my research findings in 2012 and 2015. Different from Hua’s research, which is based more on Classical studies, I devoted more attention to the debate’s historical context and its meaning in the development of related Song ritual discourse. Moreover, I thoroughly investigated a Southern Song source that contains valuable records of the 1079 debate, which Hua ignored in his research—despite the fact that he did mention the source in his essay. 6. Emperor Shenzong selected the era name Yuanfeng from several other choices based on Wang Anshi’s etymological study of characters. Ye Mengde 葉夢得 (1047–1118), Shilin yanyu 石林燕語 (Beijing: Zhonghua shuju, 1984), 5. 7. XCB, 286.6999. 8. See Peter Bol, “Wang Anshi and the Zhouli,” Statecraft and Classical Learning: The Rituals of Zhou in East Asian History, ed. Benjamin Elman and Martin Kern (Boston: Brill), 229–51.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

87

emerging from Shenzong’s ritual reforms.9 Unfortunately, this valuable record has been lost since the late thirteenth century. However, we can still construct the history of Yuanfeng ritual reforms based on the excerpts in the collected works (wenji 文集) of ritual officials and their personal commentaries on ritual Classics.10 The Yuanfeng ritual reforms played a pivotal role in Song ritual history. Emperor Shenzong led the whole process of ritual reforms. From 1078 to 1082, Shenzong assembled a number of celebrated Hanlin academicians and ritual officials into the ritual bureau of DAPTR, including Huang Lü, Li Qingchen 李清臣 (1032–1102), Wang Cun 王存 (1023–1101), Sun E 孫諤 (1065–1096), Chen Xiang, He Xunzhi, and Zhang Zao.11 Additionally, Shenzong commissioned ritual experts like Yang Wan to further examine the regulations drafted by the DPATR officials.12 He also ordered officials from other bureaus to review the tentative conclusions made by the DPATR officials. Compared to the ritual debates in the Xining era, the Yuanfeng ritual reforms reflected more of Emperor Shenzong’s personal will in reviving ancient rituals.13 Under usual circumstances, Shenzong instructed his ritual officials in the DPATR to express opinions on rituals that he considered problematic, especially on altar and temple rituals.14 Despite the diverse political backgrounds of the DPATR officials, they shared similar ritual interests in arguing for a revival of ancient rituals. Chen Xiang, the main drafter of the 1079 temple scheme and a myriad of other regulations of the Yuanfeng ritual reforms, was a typical conservative, as we have mentioned in 9. Ma Duanlin 馬端臨 (1254–1323). Wen xian tongkao (Beijing: Zhonghua shuju, 1986), 187.1598. The Song bibliographer Chao Gongwu 晁公武 (fl. twelfth century) documented a thirty-one juan in his Junzhai dushuzhi 郡齋讀書志. Yet, the Song bibliographer Chen Zhensun 陳振孫 (fl. 1211–1249) recorded Yang Wan’s work as thirty juan. Chen Zhensun, Zhizhai shulujieti 直齋書錄解題 (Taibei: Guangwen shuju, 1968), 5.15b. The extra juan in Junzhai dushuzhi might be a table of contents. See Sun Meng 孫猛, Junzhai dushuzhi jiaozheng 郡齋讀書志校證 (Shanghai: Shanghai guji chubanshe, 1990), 83–84. 10. In general, Yang Wan’s work continued the preceding practices of compiling ritual manuals in the form of yizhu 儀註 (ritual exegesis), yingeli 因革禮 (modification of rituals), and xinyi 新儀 (new rites and regulations). Concerning the Northern Song textual tradition of ritual codes, there was a tendency to favor the existing paradigm of ritual practices. Taichang yingeli, Lige xinbian 禮閣新編 (New Collections of Ritual Pavilion), and Qingli siyi 慶曆祀儀 (Sacrificial Ceremonies of the Qingli Era) all functioned as supplementary notes to the Song official ritual code of Kanbao tongli. However, Yang Wan’s work posed challenge to the conventional ritual paradigm that sometimes hindered the development of new ritual theories. Based on Yang Wan’s work, Su Song on behalf of the court compiled an official ritual code that summarized the accomplishments of Yuanfeng ritual reforms, titled the Yuanfeng xinli 元豐新禮 (New Ritual of the Yuanfeng Era). The Yuanfeng xinli textually integrated Yang Wan’s work into the Kaibao tongli. Ye, Shilin yanyu, 8. 11. XCB, 287.7012; Junzhai dushuzhi, 83. 12. XCB, 287.7012. 13. Historians have already noted the involvement of Shenzong’s own will in politics from Xining to Yuanfeng. In particular, it was reflected in the emperor’s attitude toward Wang Anshi and his selection of ministers. See Paul Smith, “Shen-tsung’s Reign and the New Policies of Wang An-shih, 1067–1085,” 447–64; Luo, Bei Song dangzheng yanjiu, 97–108. In a broad sense, Shenzong also guided the progress of the Yuanfeng reforms on officialdom and bureaucracy, as noted by the Song scholar Wang Yingchen 汪應辰 (1118–1176). Wang, Shilin yanyu bian 石林燕語辨, Ye, Shilin yanyu, 202. 14. Pang Yuanying 龐元英 (fl. 1080), Ouyang Xiu’s son-in-law, considered the revision of suburban Altar sacrifices as the main thesis of the Yuanfeng ritual reforms. See Pang, Wenchang zalu 文昌雜錄, QSBJ, Series 2:4.160–61.

88

Empowered by Ancestors

Chapter 3. Throughout his life, Chen strove to imitate ancient exemplars in order to “achieve a well-ordered world as great as the ancient one” 致治如古.15 Considering Chen’s interest in reviving ancient rituals, it is understandable that he enthusiastically participated in the Yuanfeng ritual reforms near the end of his life. Huang Lü was the exact opposite of Chen Xiang. As an apparently hardcore defender of the New Policies, Huang was notorious for making false accusations against conservative officials and sowing discord among reform leaders. In spite of his opportunist character, Huang Lü was a formidable scholar of ancient rituals, especially suburban Altar sacrifices. Historical records show that he played a key role in solving one of the most controversial problems with respect to Shenzong’s ritual reforms: whether or not the South Altar and the North Altar sacrifices should be combined.16 By tracing back to the ritual performance of the Three Dynasties, Huang argued that suburban Altar sacrifices with different configurations should be separately held at the South Altar and the North Altar.17 He Xunzhi and Zhang Zao shared views with Huang Lü in terms of ritual expertise and hostility toward conservatives. Both He Xunzhi and Zhang Zao were associated with the reformist camp. The elder brother of Zhang Zao, Zhang Huan 張環, was a close friend of Wang Anshi. Conservatives hence attributed Zhang Zao’s promotion to the Secretariat position to his brother’s personal affiliations with Wang.18 Politically, Zhang Zao advocated Wang Anshi’s New Policies and recommended Cai Bian 蔡卞 (1048–1117), Wang Anshi’s son-in-law, to the position of Lecturer of the Directorate of Education in 1082.19 Zhang further gained a bad reputation among the conservatives, because he collaborated with the notorious remonstrator Li Ding 李定 to prosecute Su Shi in the Wutai Inquisition of Su’s poetic writings (wutai shian 烏臺詩案).20 He Xunzhi leaned on the reformist camp as Zhang Zao did. He achieved his jinshi degree in the 1067 palace examination.21 Considering Emperor Shenzong’s burgeoning inclination toward reforms in 1067, the fact that he put He Xunzhi in the second place of the first rank among all of the jinshi graduates somewhat indicated He’s reformist tendency. In fact, when the conservatives regained power in the late 1080s, some of them explicitly denounced He Xunzhi for his reformist stance in Shenzong’s reign. Liu Anshi 劉安世 (1048–1125) argued that most scholar-officials 15. Liu Yi, “Chenxiansheng citang qi,” Guling ji, 25.29b; DDSL, 85.7a. Zhu Xi was aware of Chen Xiang’s interest in the general revival of ancient regime and collected the related writings of Chen Xiang in his edited volume of Song celebrated officials. Zhu Xi, Sanchao mingchen yanxinglu 三朝名臣言行錄, SBCK, 1094–101:14.1a–3b. 16. For a general portrait of this controversy, see Zhu Yi, “Cong jiaoqiuzhizheng dao tiandifenhe zhizheng: Tang zhi Beisong shiqi jiaosizhushenwei debianhua” 從郊丘之爭到天地分合之爭—唐至北宋時期郊祀主神位的 變化, Hanxueyanjiu 漢學研究 27.2 (2009): 282–88. 17. DDSL, 96.6b–7b; SS, 328.10573–4. 18. SS, 328.10569. Owing to Zhang Zao’s close affiliation with Wang Anshi, conservatives like Su Zhe, Liu Zhi 劉 摯 (1030–1097), and Wang Yansou 王巖叟 (1043–1093) charged him in the Yuanyou era, when the political atmosphere turned to their favor. XCB, 379.9213–14, 380.9230–33, 385.9373–74. 19. SS, 472.13728. 20. DDSL, 83.4a. 21. SHY, Xuanju 選舉, 5:2.10.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

89

in his time considered He’s words and deeds as vicious; Lu further criticized that He was not qualified for any reputable position in the court according to “public opinions” (gongyi 公議).22 Certainly, “public opinions” here referred to the judgment of the conservatives in the late 1080s, when the group of “Yuanyou conservatives” was at its culmination. In the 1070s, less-biased officials like Zeng Gong applauded He Xunzhi as “being able to respond to contemporary needs based on his interpretation of the Classics” 夫能據經之說適今之宜.23 Emperor Shenzong also recognized He Xunzhi’s erudition in Classics, especially in ritual Classics.24 Indeed, Shenzong personally appointed He to the DPATR in 1079/1, right before the initiation of the Yuanfeng ritual reforms.25 Unlike Chen Xiang, Huang Lü, He Xunzhi, and Zhang Zao, other DPATR officials displayed certain ambiguities concerning their political stances. Despite his inclination toward supporting Wang Anshi’s New Policies, Sun E, a high-ranking official who previously served in the Court of Sacrifices, showed his sympathy toward the conservatives and attempted to protect them from being persecuted by the grand councilor Zhang Dun during Huizong’s reign.26 Sun also had the courage to deny Wang’s interpretation of the Book of Documents at the height of the latter’s power.27 Likewise, Wang Cun, once a close friend of Wang Anshi, disagreed with Wang’s political reforms but criticized the persecution of the reformer Cai Que through the literary inquisition of Cai’s poems.28 Due to their ambiguous political stances, Sun E and Wang Cun were discriminated against by both conservative and reformist camps, and infelicitously had their names inscribed on the Stele of Yuanyou Partisans.29 Intellectually, Wang Cun was of the same type of ritual official as Huang Lü and Chen Xiang. He agreed that the South Altar and the North Altar sacrifices should be distinguished from each other by reclaiming the ancient configuration preserved in the Rituals of Zhou.30 More importantly, Wang Cun claimed

22. XCB, 431.10421. 23. Zeng, Yuanfeng leigao, 20.158. 24. In 1079/10/4, Emperor Shenzong promoted He Xunzhi to sub-editor of the Imperial Archive. In his promotion edict, Shenzong praised He for his adroitness and erudition. XCB, 300.7306. 25. XCB, 287.7012. 26. Whereas Sun advised Emperor Huizong on the danger of clique politics and suggested the court reconcile the reformers and the conservatives, he was persecuted by Zhang Dun and other reformist leaders. SS, 346: 10984. Concerning New Policies, Sun in particular realized the benefits of implementing the Hired Service System (muyifa 募役法 or mianyifa 免役法) in local administration. See Gaoben Songyuanxuean buyi, 824. 27. However, Sun E’s criticism of Wang Anshi’s commentaries on Classics, methodologically, was still confined within the analytical framework of the Wang Learning. According to the Qing scholar Wang Zicai, Sun preferred to criticize Wang based on Han Confucian commentaries. See Gaoben Songyuanxuean buyi, 825. Yet Wang’s ritual and Classical learnings were mostly characterized by their adaptation to the interpretations of Han Confucians. We will address this in detail in Chapter 5. 28. DDSL, 90.1b; SS, 341.10873. 29. Wang Chang, “Yuanyou dangjipei xingmingkao,” Jinshi cuibian, 144.15b (Wang Cun), 144.28b (Sun E); Lu Xinyuan, Yuanyou dangrenzhuan, 1.10a–b (Wang Cun), 6.11a (Sun E). 30. DDSL, 90:1a. Zeng Zhao, “Wang xueshi Cun muzhiming” 王學士存墓誌銘, in Mingchen beizhuan wanyan zhi ji 名臣碑傳琬琰之集, ed. Du Dagui 杜大珪 (fl. 1194), SKQS, 1092:30.12a.

90

Empowered by Ancestors

to have built a private family shrine (jiamiao 家廟) for his ancestors “in the ancient manner” (ru gufa 如古法).31 As one of the chief directors of the Yuanfeng ritual reforms, Li Qingchen unreservedly supported Emperor Shenzong’s plan “to follow the footprint, the Statecraft, and the cultural legacy of the Three Dynasties, henceforth creating a spectacular new order” 欲繼三代絕蹟制度文理，燦然一新.32 Like Wang Cun, Li Qingchen emphasized the importance of reviving ancient rituals within and outside the court. The young Li had already underscored the decisive role of rituals in the promotion and demotion of clerks (li 吏) in his response to the palace examination questions.33 Li’s political career was characterized by his long-term services in most ritual bureaus, which also symbolized the key role played by rituals in the broad spectrum of Song officialdom.34 In 1078, Emperor Shenzong asked the DPATR officials to formulate a new model of the Imperial Temple configuration as a crucial part of his ritual reforms. Several months later, the DPATR submitted a scheme to Shenzong and asserted that it perfectly corresponded to the regulations of the previously compiled Xining yi 熙寧儀, a liturgical manual summarizing temple and altar rituals of the Xining era. Despite the negligible differences between the DPATR scheme and the Xining yi,35 the former is a likely replication of Wang Anshi’s conception of the temple configuration, which had been successfully implemented after the 1072 debate. However, the DPATR scheme distinguishes itself from the Xining setting in a crucial way: it emphasizes the necessity to lodge spirit tablets in separate temples, rather than in one single temple. Hence, it proposes an architectural complex of multiple temples and focuses on the ritual sequence of these temples. In 1040, Zhao Xiyan had already purposed to establish multiple temples. But his proposal was turned down by Emperor Renzong.36 Four decades later, the DPATR officials reignited Zhao Xiyan’s idea by memorializing a new scheme of multiple temples to Emperor Shenzong, who responded positively and called for a further discussion on the scheme. The DPATR memorial reads: According to the Zhou setting, ancestral rituals of those who ranked above junior officials with honorable titles should be performed in this way: the tablets of grandfathers, fathers, and sons were placed in separate temples, in order to pay respect to ancestors and not to blaspheme them. As the Law of Sacrifices chapter in

31. Zeng Zhao, “Wang xueshi Cun muzhiming,” 30.16b–17a. 32. Chao, “Zizhengdian daxueshi ligong xingzhuang,” 資政殿大學士李公行狀, 487. 33. Chao, “Zizhengdian daxueshi ligong xingzhuang,” 485; DDSL, 96.3a. 34. Li Qingchen served successively at the Court of Imperial Sacrifices, the Commission of Ritual Affairs, the DPATR, and as the ritual practitioner of the mausoleum of Empress Gao (xuanren huanghou shanling liyishi 宣仁皇后山陵禮儀使), and the director of the Bureau of Rites throughout his career. Chao, “Zizhengdian daxueshi ligong xingzhuang,” 485b–489a. 35. For instance, in the DPATR scheme Shunzu’s tablet had already been removed from the Imperial Temple. SS, Zhi 59.2574. 36. XCB, 129.3059–60; SHY, Li 1:15.29.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

91

the Liji says, “A shi of the highest level has two ancestral temples.”37 The Annals of Spring and Autumn documents the temples of Duke Heng and Duke Xi of Lu.38 The Betrothal Gift chapter in the Rites and Ceremonies records that “someone received ritual coins from some temples.” The Question of Master Zeng reads, “When spirit tablets are taken from their temples or returned there, it is required to keep passengers out of the tablets’ way.”39 From kings and feudal lords to the junior officials with honorable titles, people of different social statuses would not avoid following the ritual practice (of performing ancestral sacrifices in separate temples). Only the junior clerks of feudal lords would lodge the tablets of their fathers in a single chamber and the tablets of their grandfathers in a single temple. The Later Han Emperor Guangwu defied the ritual practice of setting separate temples because of his excessive frugality. He merged all Han imperial ancestors from Gaozu to Emperor Ping in a single temple: their tablets were stored in different chambers yet they all resided in the same hall. This kind of ritual practices diminished the status of the Son of Heaven to the level of the Zhou junior clerks. Succeeding dynasties followed the conventional practice started by Emperor Guangwu and failed to reform it. Therefore, based on the traces of ancient ritual practices recorded in the Rites and Ceremonies, and the designations of the Imperial Temple in the Dictionary of Erudition, and the measurement documented in the Record of Technique, your servants respectfully submit the “Diagram of the configuration of eight temples with different chambers,” in which the Primal Ancestor temple is placed at the center, where other temples are assigned alternatively to its right and left sides according to the zhaomu sequence.40 周制，由命士以上，父子異宮，祖禰異廟，所以致恭而不凟也。〈祭法〉 曰：「適士二廟」；《春秋》書「桓宮、僖宮」；〈聘禮〉有之「某君受幣 於某宮」；〈曾子問〉曰：「主出廟，必蹕」。是人君達於命士，莫不然 也。惟諸候之下士，則父子同宮而居，祖禰共廟而祭。後漢光武儉不中禮， 合高祖以下至平帝為一廟。異室同堂，屈萬乘之尊，而俯同周之下士，歷代 因循不革。臣等以《儀禮》求其迹，以《爾雅》辯其名，以〈考工記〉約其 廣深，謹圖上八廟異宮，以始祖居中，昭穆為左右以進。

As the head of the DPATR, Chen Xiang drafted this memorial himself.41 The attached “Diagram of the configuration of eight temples with different chambers”

37. Shishi 適士 means shangshi 上士 in this context. That is, the government officer with the highest grade. Legge, The Sacred Books of China, 4:205. In addition to two ancestral temples, shishi could also build an altar for presenting seasonal sacrifices. 38. Duke Heng was Duke Xi’s grandfather. The fact that they possessed their own temples demonstrated the principle of separation in dealing with the placement of spirit tablets or temples. 39. Here, the DPATR scholars failed to quote the whole sentence and, consequently, obscure the meaning in the original context. The full sentence is: 主，出廟入廟，必蹕. The character bi 蹕 conveys a meaning of traffic control in ancient ritual Classics. See Zhu Bin, Liji xunzuan, 298–99. See also Legge, The Sacred Books of China, 4:325. 40. XCB, 292.7138–39. 41. XCB records that the DPATR memorial was sent to Chen Xiang for a further examination after its first submission to the court. However, I found an original draft of the DPATR memorial in Chen Xiang’s collected works. The text is exactly the same as the one I quoted here. See Chen, “Bamiao yigong,” Guling ji, 9.2a–b.

92

Empowered by Ancestors

Figure 4.1: Diagram of the Configuration of Eight Temples with Different Chambers

(bamiao yigong 八廟異宮) was portrayed by He Xunzhi and Zhang Zao.42 Most ritual officials in the DPATR endorsed the memorial. Primarily, these officials called for a renovation of the Song Imperial Temple by adding new extensions to the current temple complex. As shown, they cited evidence from the ritual Classics to advocate a configuration of multiple temples. According to the DPATR scheme, Xizu’s temple should be placed at the center of the whole arrangement, as he was designated as the Primal Ancestor in the 1072 debate. The temples of Yizu, Taizu, Taizong, and Renzong were placed on the right side, along with the mu sequence; on the left were the temples of Xuanzu, Zhenzong, and Yingzong, along with the zhao sequence. The spatial orientation should be like Figure 4.1, with all the temples arranged from north to south vertically.43 According to the classic interpretation of the “eight-trigram” (bagua 八卦) system in the Book of Changes, the Primal Ancestor seat on the north emblematized the supreme qian 乾 position; other positions were subjected to it. Theoretically, other temples should be placed on the south of Xizu’s temple to highlight Xizu’s ritual status in relation to his descendants. The conventional arrangement of the Song Imperial Temple, however, placed ancestral chambers along the east-west axis, with the Primal Ancestor facing east.44 Seemingly, the scheme of “eight temples with different chambers” attempted to reframe the spatial arrangement of Song imperial ancestors in the temple by adopting a new orientation along the north-south axis. 42. Tang Qinfu 湯勤福 suspected that there were two editions of this diagram, one by He Xunzhi and the other by Chen Xiang. See Tang Qinfu and Wang Zhiyao 王志躍, Song shi lizhi bianzheng 宋史禮志辨證 (Shanghai: Shanghai sanlian shudian, 2011), 434–35. My understanding is that He Xunzhi and Zhang Zao as DPATR colleagues worked hand in hand on this diagram and presented it to Chen Xiang for final submission to the court. Two pieces of textual evidence support my claim: (1) In Song Shi, it is explicitly recorded that He Xunzhi presented the diagram to the court 何洵直圖上八廟異宮; SS, 106.2574; (2) In Lu Dian’s criticism of this diagram, he stated that it was codified by “Zhang Zao and the others” 右張璪等所定圖. Lu, “Zhaomu yi” 昭穆議, Taoshan ji 陶山集, SKQS, 1117:6.13a. 43. SS, 106.2573. 44. Wan, Miaozhitu kao, 97a, 98a.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

93

By differentiating individual temples (in practice, chambers) from one another, it continued Wang Anshi’s endeavor to highlight Xizu’s ritual status. The DPATR scheme echoed Shenzong’s resolution to rectify the configuration of the Imperial Temple according to ancient rituals. In practice, the scheme of multiple temples and the corresponding change of their orientation could not be enacted without an exhaustive revision of relevant temple rituals. From 1078 to 1079, the DPATR officials led by Chen Xiang turned to examine concrete temple rites and ceremonies, such as the arrangement of monthly fresh offerings and ritual utensils used in temple sacrifices.45 However, as more scholar-officials were involved in relevant ritual debates, new concerns about the DPATR scheme emerged and resulted in a theoretical dispute about the nature of temples in 1079.

The 1079 Zhaomu Debate: Lu Dian and the DPATR The period from 1078 to 1079 witnessed the height of the Yuanfeng ritual reforms. Most revisions concerning altar and temple rituals were instated in this period. In 1079/1, Shenzong assigned remonstrator He Zhengchen 何正臣 to check the preparation of the spring sacrifice to be held in the Imperial Temple two months after.46 In the same year, a ritual official named Lu Dian was appointed to the DPATR. The young Lu Dian thoroughly challenged the DPATR scheme of the Imperial Temple and redefined temple rituals in such a provocative way that even the cleverest ritual officials in the DPATR found his ritual ideas elusive. A brief introduction of Lu Dian’s intellectual background should help us understand his provocative ideas. Lu was born to a poor family in the Shanyin 山陰 County of the Prefecture of Yue 越州 (modern Zhejiang).47 Owing to financial constraints, Lu grasped every chance of learning in a self-disciplined manner. It was said that Lu’s family was too poor to afford candles. As a result, Lu read books by the moonlight every night.48 In the late 1060s, the young Lu Dian had a chance to study with Wang Anshi when the latter lectured local scholars from 1064 to 1067 at Jiangning 江寧, a place near Lu Dian’s hometown. Lu always claimed himself as one of the disciples of Wang and reminisced in his late years about how Wang’s teaching shaped him.49 During the Xining era, Lu Dian ascended in the bureaucracy rapidly due to his early training of studies of Confucian Classics, especially his skillful interpretation of Wang Anshi’s scholarly works. His personal connection with Wang Anshi raised the eyebrows of some of his contemporaries.50 In 1072, Wang Anshi appointed Lu Dian to assist in the compilation of a new commentary on the Book of Songs, later 45. XCB, 292.7134; 292.7138; XCB, 299.7273–74. 46. XCB, 296.7197; 297.7232. 47. SS, 343.10917. 48. SS, 343.10917. 49. Lu, “Shenjun mubiao” 沈君墓表, Taoshan ji, 16.11b. 50. XCB, 211.4129.

94

Empowered by Ancestors

known as the Shijing yi 詩經義 (Meaning of the Book of Songs).51 Since then, Lu Dian’s expertise in interpreting Classics had been widely recognized by the scholarofficials of the reformist camp.52 The imperial edict concerning Lu’s promotion to the expositor in attendance (sijiang 侍講) praised him for being “fond of antiquity and familiar with the Classics” (haogu zhijing 好古知經).53 Moreover, in one of the edicts on Lu Dian’s promotion, Emperor Shenzong personally remarked that Lu was clever and erudite.54 Particularly, Lu Dian’s expertise in ritual scholarship was noticed by his contemporaries.55 His enthusiasm toward ritual details, such as the utensils and garments used in altar and temple sacrifices, resembled Han Confucians’ learning of rituals. After joining the DPATR in 1079, Lu shared his knowledge of ritual details to assist in the revisions of sacrificial rites and related utensils, including the ritual dishes used in the regular offerings of daily food in the temple (yapan 牙盤, literally, the “plate of teeth”), the fabric used to cover ritual vessels in the temple (shubu 疏布, literally, the “sparse cloth”), and the ritual garment of the emperor in altar sacrifices (daqiu 大裘, a specific fur coat with a dozen symbolic images).56 In most ritual discussions within the DPATR, Lu Dian kept pace with his colleagues. His emphasis on ritual details brought about the improvement of the original schemes designed by his colleagues. However, when Lu memorialized to the court a new scheme of the Imperial Temple in early 1079, he fundamentally challenged his colleagues’ ritual expertise. Lu Dian’s new scheme included two parts that were separately addressed in two memorials, respectively titled “Miaozhi yi” 廟制議 (Discussion on the temple structure) and “Zhaomu yi” 昭穆議 (Discussion on the zhaomu sequence). The first part of Lu Dian’s temple scheme was stated in the “Miaozhi yi.” In this memorial, Lu discussed different structural elements of the temple, including the towered gateways (taimen 臺門), the black threshold (xuankun 玄閫), the central yard (titang 提唐), the carved screen (shuping 疏屏), the four entrance doors (simen 四門), the stairs and doorsteps (jiujie 九階, zuoqi 左 墄, and xuanbi 玄陛), the decorated square timber (fuge 復格), the black column (heiying 黑楹), the red rafter (danjue 丹桷), the side windows and the cross-framed windows (daxiang 達鄉 and jiaoyou 交牖), the main wall (biyong 賁墉), the small room connecting to the main chamber (sheyi 設移), the multiple corridor 51. XCB, 229.5570. 52. Zeng Bu, Zenggong yilu 曾公遺錄, QSBJ, Series 1:8.194. Zeng Bu also mentioned Lu’s erudition in Classics and regarded him as a potentially talented official. Zeng, Zenggong yilu, 226. 53. Zeng Gong, “Lu Dian jian shijiang zhi” 陸佃兼侍講制, Yuanfeng leigao, 1680–89:21.162. 54. XCB, 298.7256. 55. Shenzong once claimed that “there has never been anyone like Lu, who could explicate rituals in such a detailed and clear manner, except Zheng Xuan and Wang Su” 自王鄭以來言禮未有如佃者. Songshi xinpian, 116:7a. 56. DDSL, 97.3b–4a; Chao Yuezhi 晁說之 (1059–1129), Chaoshi keyu 晁氏客語, QSBJ, Series 1:10.102; Lu, Taoshan ji, 5.1a–18a. Particularly, Lu Dian’s argument on the correct format of the daqiu demonstrates how well Lu addressed ritual details. Li Po 李朴 (1063–1127), Fengqingmin gong yi shi 豐清敏公遺事, QSBJ, Series 2:8.139; Fang Shao 方勺, Bozhai bian 泊宅編, QSBJ, Series 2:8.218–19.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

95

(chonglang 重廊), and the outer wall with the decoration of mountain paintings (shanqiang 山牆).57 Lu Dian cited most of these structural elements from a suspicious writing named Yi Zhoushu 逸周書 (Excerpts of Zhou Texts).58 None of these technical designations appeared in the main text of the three ritual Classics. Based on these imagined Zhou designations and his elaborations on their arrangement, Lu Dian constructed a model of the ancient Imperial Temple that was unfamiliar to his colleagues in the DPATR. If Lu could successfully integrate these designations into the Classical narrative of the Imperial Temple, he would not have perplexed his DPATR colleagues. However, it was the second part of Lu Dian’s scheme that fundamentally undermined his colleagues’ work and challenged the predominant conception of ritual order in the Song Imperial Temple. In the memorial titled “Zhaomu yi,” Lu Dian opposed the “Diagram of eight temples with different chambers” that was portrayed by He Xunzhi and Zhang Zao and later endorsed by other DPATR ritual officials.59 Lu argued that the zhao and mu lines in He Xunzhi’s diagram should be swapped with each other to embody the “order of seniority” (zunbei 尊卑). According to this order, senior ancestors always reside in zhao positions and junior ancestors reside in their exact opposite mu positions. Lu argued that the original diagram designed by He Xunzhi defied the right order of seniority. In He’s diagram, along the mu line, those ancestors on the right side of Xizu are the fathers of the ancestors on their exact opposite zhao positions: Yizu is Xuanzu’s father; Taizu and Taizong are brothers and Taizong is Zhenzong’s father; Renzong is Yingzong’s father.60 Against He Xunzhi’s conception of the imperial zhaomu sequence, Lu argued that the temples of Yizu, Taizu, Taizong, and Renzong should be arranged along the zhao line, since ancestors in these temples are the fathers of those who are situated in their exact opposite temples. Temples of Xuanzu, Zhenzong, and Yingzong should be arranged along the mu line, since ancestors in these temples are the sons of those who are situated in their exact opposite.61 The diagram in Figure 4.2 (p. 96) depicts Lu Dian’s proposed revision of the Imperial Temple. In summary, Lu Dian switched the positions of zhao and mu ancestors in He Xunzhi’s scheme. Meanwhile, he concurred with the configuration of eight temples. In Lu’s scheme, temples on the left zhao rank are reserved for the ancestors who are 57. Lu, “Miaozhi yi,” Taoshan ji, 6.1a–6a. 58. For example, see Lu, “Miaozhi yi,” 6.1b, 3a, 3b, 5a. 59. The extant version of the “Zhaomu yi” was preserved in Lu Dian’s collected works, Taoshan ji. The original copy of Taoshan ji was lost and then restructured after the twelfth century, which resulted in the disappearance of some essays. The extant Taoshan ji in the Wenyuange siku quanshu was a result of Qing editors’ endeavors to recollect Lu Dian’s scattered works based on the Ming Yongle dadian 永樂大典 (Collected Texts of the Yongle Era). 60. Lu, “Zhaomu yi,” 6.13a. The official dynastic history records that Shunzu was also categorized as a zhao ancestor in He Xunzhi’s diagram. SS, 106.2573. However, in Lu Dian’s memorial, which quoted He Xunzhi’s diagram as a reference, Shunzu was absent. Considering that Shunzu as the farthest ancestor should have already been removed from the Imperial Temple in Shenzong’s time, the compilers of SS probably made a mistake. 61. Lu, “Zhaomu yi,” 6.13b.

Empowered by Ancestors

96

Figure 4.2: Zhaomu Sequence Suggested by Lu Dian in the 1079 Debate

senior to their mu counterparts on the exact opposite; correspondingly, temples on the right zhao rank are used to house ancestors who are the sons of the ancestors in the left zhao temples. Throughout the “Zhaomu yi,” Lu Dian also criticized He and Zhang for ignoring the seniority factor and hence defying the ritual intent (liyi 禮意) of ancestral rituals. Lu claimed that He Xunzhi’s arrangement of imperial ancestors postulated that zhao and mu ancestors should only be moved along their own axes. In other words, all zhao ancestors can only shift along the zhao line. Likewise, all mu ancestors can only shift along the mu line. Lu challenged this fixed zhaomu pattern by referring to the arrangement of spirit tablets in the collective offerings of ancestors in the xia sacrifices. He made a hypothetical case about this in the “Zhaomu yi”: Suppose that within the family of a great minister, according to the generational sequence, the father and the great-grandfather generations are designated as zhao ancestors and the grandfather and the great-great-grandfather generations are designated as mu ancestors. In the collective offerings of ancestors in the xia sacrifices, all these ancestors will be seated in the same space, facing one another. The common reason is that the less senior should not take precedence over the more senior. Under such circumstance, would Zhang Zao and other ritual officials in the DPATR insist that zhao ancestors are always kept as zhao and mu ancestors are always kept as mu? If their scheme is adopted, then the great-grandfather will occupy the superior position, but the great-great-grandfather will occupy the inferior; likewise, the father will occupy the superior position, but the grandfather will occupy the inferior. This scheme therefore fails to accord with the intention of the zhaomu sequence as a ritual embodiment of the relationships between fathers and sons, in which zhao conveys a meaning of illuminating the inferior son, and mu conveys a meaning of revering the superior father.62 62. Lu, “Zhaomu yi,” 6.11b.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

97

假令大夫昭穆，以世次計，曾祖適為昭，高祖適為穆；父適為昭，祖適為 穆。同時合食，則將偶坐而相臨。義不得以卑而踰尊。則璪等將令「昭常為 昭，穆常為穆」乎？如此則曾祖居尊高祖居卑、父居尊祖居卑矣。非所謂父 昭子穆、昭以明下、穆以恭上之義。

In Lu Dian’s opinion, placing the grandfather and the great-great-grandfather along the mu line symbolizes a disconcerting scenario wherein “senior ancestors beg for respect from the junior ancestors who precede them in the zhaomu sequence” 實 屬父行乞于上世之次.63 Lu interpreted this setting as a violation of the spirit of filial piety. Although Lu admitted that the free shift of position between zhao and mu lines might cause some confusion in terms of designations, he was confident that his zhaomu scheme represented the correct sequence of seniority. In the final analysis, Lu Dian argued that zhao and mu as ritual designations symbolized the virtues of fathers and sons in temple sacrifices. As he said, “the zhao designation conveys a meaning of illuminating the inferior; the mu designation conveys a meaning of revering the superior” 昭以明下為義，穆以恭上為義.64 According to his understanding, father-ancestors should always be placed at the zhao positions to illuminate the son-ancestors on their opposite. In the Song Imperial Temple, the father-ancestors would refer to Yizu, Taizu, Taizong, and Renzong, who “illuminated” their sons—Xuanzu, Zhenzong, and Yinzong—in their opposite mu positions.

The 1079 Zhaomu Debate: Lu Dian Versus He Xunzhi and Zhang Zao Lu Dian’s “Zhaomu yi” set the platform for the 1079 debate. He Xunzhi and Zhang Zao did respond to Lu Dian’s “Zhaomu yi.” However, these responses had not been documented in official records. Moreover, unlike Lu Dian, He and Zhang left no personal writings that could be compiled into collected works. Fortunately, in a Southern Song collection of over fifty commentaries on the Book of Rites, I found a detailed record of the arguments used by both sides in the 1079 debate. The collection, titled the Liji jishuo (Collection of Explanations on the Book of Rites), offers a rich repository of materials concerning the 1079 debate, including the excerpts of Lu Dian’s two lost commentaries on the Book of Rites, the Liji jie 禮記解 (Explanations of the Book of Rites) and the Liji xinshuo 禮記新說 (New Explanations of the Book of Rites).65 Thanks to Wei Shi, the compiler of the Liji jishuo, it is possible to scrutinize the 1079 debate and the involved ritual arguments developed by Lu Dian, He Xunzhi, and Zhang Zao.

63. Lu, “Zhaomu yi,” 6.12a. 64. Lu, “Zhaomu yi,” 6.10b. 65. Liji xinjie probably is a revised version of the Liji jie. Wei Shi, Liji jishuo (hereafter LJJS), SKQS, 117: Mingshi 名氏.4a.

98

Empowered by Ancestors

The first section of the Liji jishuo starts with a quotation from Lu Dian’s “Zhaomu yi.”66 After the quotation, it follows Lu Dian’s summary of several counterarguments that all begin with the phrase “someone may argue that (shuozhe yue 說者曰 / huozhe yue 或者曰).” “Someone” here refers to either He Xunzhi or Zhang Zao (mostly He). The succeeding parts, which are cited from Lu Dian’s Liji jie and Liji xinshuo, then turn to Lu’s further responses to these counterarguments. For the sake of analysis, I have reorganized the Liji jishuo narrative into three sections, each of which contains a counterargument and Lu Dian’s response. The first counterargument goes as follows: Someone may argue that in the Zuo Commentary of the Annals, Da Bo and Yu Zhong were designated as zhao ancestors in relation to their father Taiwang.67 Yet, Hao Zhong and Hao Shu were designated as mu ancestors in relation to their father Wang Ji.68 Likewise, it was argued that Guan, Cai, Cheng, and Huo were designated as zhao ancestors in relation to their father King Wen.69 Yet, Han, Jin, Ying, and Han were designated as mu ancestors in relation to their father King Wu.70 Someone may also argue that because the Book of Documents records that “King Wen is a mu ancestor,” King Wen of Zhou always resided at the mu positions in the Zhou Imperial Temple. Accordingly, King Wu always resided at the zhao positions. When the tablet of Wang Ji was removed from the Zhou temple, King Wu was accordingly moved into Wang Ji’s temple as a zhao ancestor. Yet, King Wen remained at his mu position. While King Kang was moved into King Wu’s temple as a zhao ancestor, King Cheng remained at his mu position; while King Mu was moved into King Kang’s temple as a zhao, King Zhao remained at his mu position.71 說者或以《左傳》大伯、虞仲，太王之昭；虢仲、虢叔，王季之穆；管、 蔡、郕、霍，文之昭也；邗、晉、應、韓，武之穆也。又以《書》稱「穆考 文王」，乃謂文王世次居穆；武王世次居昭。王季親盡而遷，則武王入王季 之廟為昭，文王仍為穆；康王入武王之廟為昭，成王仍為穆；穆王入康王之 廟為昭，昭王仍為穆。

The reasoning underneath the above quote is that both zhao and mu as ritual designations were originally designed by Zhou people to indicate fixed positions of ancestors with reference to their ancestral temples. Therefore, zhaomu has nothing to do with the representation of patrilineal relationship. Against this counterargument, Lu Dian reiterated his fundamental standpoint that it is inappropriate to designate ancestors of the son generations as zhao ancestors and designate ancestors of the father generations as mu ancestors. He argued that this type of practice violates the ritual intent of zhaomu, which was designed by

66. Wei, LJJS, 30.28a; Compare the text with Lu Dian’s wordings in the “Zhaomu yi,” 6.10a. 67. Correspondingly, Ta Wang is a mu ancestor in relation to his sons, Da Bo and Yu Zhong. 68. Wang Ji is a zhao ancestor in relation to his sons, Hao Zhong and Hao Shu. 69. King Wen is a mu ancestor in relation to his sons, Guan, Cai, Cheng, and Huo. 70. King Wu is a zhao ancestor in relation to his sons, Han, Jin, Ying, and Han. 71. Wei, LJJS, 30.29b.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

99

ancient sage-kings to symbolize the correct order of seniority.72 To further elaborate the intent of ancient sage-kings, Lu differentiated the “genealogical sequence” (shici 世次) from the “ritual sequence of temple ancestors” (miaoci 廟次). By defining shici as the natural sequence of ancestors, Lu agreed that the zhao and mu designations corresponding to shici primarily indicate the relative positions of ancestors in their ancestral lines. Therefore, regarding the Zhou case: “from Hou Ji to King Wu, there are altogether sixteen ancestors in the Zhou ancestral line. This is called shici” 蓋 周自后稷至文武十有六世，此世次也.73 Lu Dian admitted that the ritual practice of the “removal of ancestors from the temple” (qian 遷) does not apply to the Zhou shici.74 In the shici system, the son of a zhao ancestor should be designated as a mu ancestor, and vice versa. In this light, Lu Dian argued that the phrase quoted by He Xunzhi and Zhang Zao from the Zuo Commentary, that “King Wen always resided at the mu positions” 文王世次居穆, should be read as “King Wen was a mu ancestor according to his position in the genealogical sequence.” Lu further cited the Jin commentator Du Yu 杜預 (222–285) to support his interpretation of the term shici. As Du Yu stated in his sub-commentary to the Zuo Commentary: “Ta Bo, Yu Zhong were zhao ancestors corresponding to the Zhou genealogical line according to their positions in the genealogical sequence” 以世次計，故大伯、虞仲，於周為昭.75 Lu Dian agreed to Du Yu’s understanding of shici as capturing the genuine meaning of shici in the main text of the Zuo Commentary. However, Lu Dian argued that miaoci was different from shici, as miaoci was concerned mostly with the ritual sequence of temple ancestors.76 In Lu’s words, although the genealogical sequence can be extended as long as new ancestors are added into the shici, the number of ancestors to whom one can make seasonal sacrifices in the Imperial Temple is limited. Under usual circumstances, the number of ancestors in the Imperial Temple should not exceed six. In other words, as the ritual indicator of the six ancestors in the temple—what I name as “the tail of shici,” miaoci symbolizes a ritual order that disciplines the intermediate space between ancestors and their descendants. In Lu Dian’s reasoning, ancestors would obtain tentative zhao and mu designations according to their positions in the genealogical sequence of shici. However, these designations would change when ancestors are moved into the Imperial Temple, along with the rearrangement of their temple sequence in the form of miaoci.77 72. Wei, LJJS, 30.30a. 73. Wei, LJJS, 30.30a. 74. The Sangfu xiaoji (Record in the Dress of Mourning) chapter in Liji describes the ritual practice of “qian” in the ancient clan system. According to the Sangfu xiaoji, the Honored Head (biezi 別子, a son other than the eldest son of a family) continued the bloodline of his high ancestor through a removal of his spirit tablet from the original family line after the passage of five generations. The removal of the Honored Head’s tablet indicated the birth of a new but shorter family line. TSZSSSJ, 2:32.376–77; Legge, The Sacred Books of China, 4:43. 75. Wei, LJJS, 30.30a. For the original text in Du Yu’s sub-commentary, see Zuozhuan zhushu 左傳注疏, TSZSSSJ, 3:12.134. 76. Wei, LJJS, 30.30b. 77. In Lu Dian’s own words, it means the shift of the zhaomu sequence (zhaomu yiyi 昭穆移易). Wei, LJJS, 30.30b.

100

Empowered by Ancestors

Regarding ancestors in the miaoci sequence, Lu Dian argued that they should be “served in the same manner as the living would be served” 以事生之禮事之.78 Among all the manners in the living world, the familial relation between father and son is the most precious one. Therefore, Lu Dian proclaimed that it is necessary to reallocate the zhao and mu positions of ancestors based on the miaoci order every time when the spirit tablet of a newly deceased ancestor is moved into the temple. In this way, the emperor on the throne can demonstrate his affection to his imperial ancestors and thus present a perfect model of filial piety for his subjects to follow. The second counterargument raised by He Xunzhi and Zhang Zao concerns the fu ritual of attaching tablets in the Imperial Temple (fumiao 祔廟).79 He Xunzhi cited a sentence from the Tangong 檀弓 (Wingceltis Bow) chapter of the Book of Rites, which states: “the next day the ritual of attaching the spirit tablet of the deceased ancestor to his grandfather’s is performed” 明日祔於祖父.80 In the Tangong case, the fu ritual of attaching the tablet of the deceased ancestor to his grandfather’s stirs a suspicion of ritual inappropriateness: the deceased ancestor as a grandson is directly attached to his grandfather, yet the ritual status of his father in the fu ritual is ignored. Based on this Tangong text, He Xunzhi and Zhang Zao made a logical deduction. If scholars insist that ancestors of father generations should always be seated at the zhao position to embody their superior ritual status, then the fu ritual is inappropriate, because it ignores fathers’ status in the Imperial Temple. However, since the fu ritual is unquestionably valid, given that it has been recorded in the ritual Classics, the presupposition that zhao and mu designations symbolize the hierarchical relations between fathers and sons must be invalid.81 Lu Dian refuted the second counterargument in a dexterous way. First, from the perspective of ritual performance, he illustrated that comparison should not be made between the fu ritual and the qian ritual as the latter deals with the removal of an ancestor’s tablet from the temple. Lu admitted that the fu ritual was an ancient ritual of the Zhou ritual matrix, as He and Zhang indicated. Nevertheless, concerning other ritual practices of mourning and grieving, Lu pointed out that “the ritual of attaching tablets is carried out right after the ritual of wailing” 卒哭而祔. He also pointed out that “the ritual of the removal of tablets is performed twelve months 78. Wei, LJJS, 30.30b. 79. According to the Sangfu xiaoji chapter in the Book of Rites, the fu ritual referred to the attachment of an individual’s spirit tablet to an ancestor of higher seniority in ancestral temples. The original text of the Sangfu xiaoji reads: “The tablets of the deceased maternal relatives (wives and concubines) should be attached to their grandmothers’; if there have been no such grandmothers’, these tablets should be attached to their great-greatgrandmothers’, according to the zhaomu sequence” 其妻袝於諸祖姑， 妾袝於妾祖姑，亡則中一以上而袝. See LJZS, 2:33.384. For a detailed discussion of the fu ritual, see Zhu Bin, Liji xunzuan, 133, 497. The term zhongyi 中一 means skipping one generation in the ritual of tablet attachment. James Legge misinterpreted this term in his early translations of Liji. See Legge, trans., The Sacred Books of China, 4:51. 80. Wei, LJJS, 30.31a. See Legge, The Sacred Books of China, 4:171. Translation modified. 81. Wei, LJJS, 30.31a. In terms of propositional logic, He and Zhang clearly adopted a modus tollens argument here, in which “the free switching of the zhaomu of temple ancestors” stands for P and “the fu ritual is inappropriate” stands for Q. In He’s and Zhang’s reasoning, the negation of Q (the fu ritual is not inappropriate) implies the negation of P, that is, the free switching of the zhaomu of temple ancestors.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

101

after the ritual of attaching tablets, when the son of the deceased ancestor begins to wear the mourning garment of the lian grade” 練而後遷廟.82 Hence, Lu Dian questioned how He Xunzhi and Zhang Zao could conceptualize fu and qian rituals in the same way, given that these two rituals are basically performed in different contexts. Lu concluded that it is nonsensical to use the fu ritual to explain the qian ritual and the formulation of the zhaomu sequence in relation to the removal of tablets. Furthermore, Lu urged He Xunzhi and Zhang Zao to consider the situation where the two spirit tablets of the son and the father coexist in the grandfather’s temple in the fu ritual. He referred again to the Zhou case to explicate this: When the tablet of King Mu was attached to that of his grandfather (King Kang) in the fu ritual, as Wang Ji had not yet been removed from the temple setting, the zhaomu sequence remained unchanged. Since King Mu and King Kang were both zhao ancestors according to the genealogical sequence, the former’s tablet was placed in the latter’s temple in the fu ritual. This is what has been called “to attach the spirit tablet of the deceased ancestor to his grandfather’s.” Because the grandson’s tablet was attached to that of his grandfather, he should not be regarded as the subject of his grandfather’s temple. Therefore, the attachment of the tablet of King Mu to his grandfather’s should be free from the criticism that the son’s ritual authority overpowered his father’s.83 且穆王初祔未練，則王季未遷，昭穆未動；與祖昭穆同班，則祔於康王之 廟，所謂祔於祖父也。祔於祖父，則非專其廟。而襲其處自無壓父之嫌。

While the possible danger of ritual inappropriateness in the fu ritual was dispelled, Lu Dian claimed that the fu ritual should not be used as a pretext for overlooking the seniority factor in conceptualizing the zhaomu sequence of temple ancestors. The third counterargument is related to the second one. He Xunzhi quoted the Han Confucian scholar Liu Xin 劉歆 (ca.  50 BC–AD  23) to substantiate his claim that zhao ancestors and mu ancestors are only allowed to shift along their own lines but never be permitted to cross the lines. As Liu Xin expressed: “If the grandson’s tablet is placed in his grandfather’s position, then the zhaomu sequence will be rectified. Therefore, the replacement of the zhao and mu positions of the grandfathers by the tablets of their grandsons refers to the ritual of the removal of tablets in the temple” 孫居王父之處，正昭穆，則孫常興祖相代，此遷廟之殺 也.84 However, Lu Dian interpreted the first part of Liu Xin’s quote as “to attach the grandson’s tablets to his grandfather’s position” 孫從王父之位, rather than “to place the grandson’s tablet in his grandfather’s position” 孫居王父之處.85 Lu argued that if the grandson’s tablet does not supersede but is just attached to his grandfather’s 82. Wei, LJJS, 30.31a. Both phrases can be found in the same Tangong chapter that He Xunzhi and Zhang Zao cited. The Qing scholar Zhu Bin’s annotation states that in the fu ritual the spirit tablets of the departed should be attached only to that of his grandfather with the same zhao and mu titles. Zhu Bin, Liji xunzuan, 133. 83. Wei, LJJS, 30.31a. Emphasis mine. 84. Wei, LJJS, 30.32a. For Lin Yin’s original wordings, see Wang Xianqian, Hanshu buzhu 漢書補注 (Yangzhou: Guangling shushe, 2006), 43.20. 85. Wei, LJJS, 30.32a.

102

Empowered by Ancestors

tablet in the temple, it is not the replacement of a zhao ancestor by another zhao ancestor. Basically, Lu considered He Xunzhi’s quote from Liu Xin as a description of the fu ritual, rather than the qian ritual. To summarize Lu Dian’s responses to He Xunzhi’s and also Zhang Zao’s counterarguments, there are three aspects to which Lu devoted special attention. First, he emphasized the intimate relationship between grandfathers and grandsons in the Imperial Temple context; second, he highlighted the difference between the rituals of fu and qian in referring to the Imperial Temple; and third, he conceptualized the zhaomu sequence based on a new concept of miaoci. Compared to Lu Dian’s wellstructured responses, the counterarguments made by He Xunzhi and Zhang Zao are relatively incomplete and deficient. Particularly, He Xunzhi failed to explain the difference between the ritual performance of fu and qian, as well as why the zhaomu order of genealogical sequence should be strictly adopted in the Imperial Temple. He Xunzhi frequently quoted phrases from the Classics. Yet, he could not make a consistent argument based on these quotes.86 Despite the clarity and consistency of Lu Dian’s proposed revisions of the zhaomu of the imperial lineage, the court ignored his revisions and eventually adopted the original scheme designed by He Xunzhi and Zhang Zao in restructuring the Imperial Temple. He’s and Zhang’s higher ranks in the DPATR and their roles in helping Emperor Shenzong to reform officialdom in 1080 might explain the court’s silence on Lu Dian’s zhaomu plan and the 1079 debate.87 An even more important reason is that their scheme was established upon the consensus of most DPATR officials. As I have argued in the beginning of this chapter, the intellectual discrepancy between Yuanfeng ritual officials became less obvious in the 1080s under Emperor Shenzong’s overwhelming personal interest in reviving ancient rituals. However, the “outlier” Lu Dian created new tension within the circle of Shenzong’s ritual officials. Controversies regarding other rituals gradually emerged after the 1079 debate. More “outliers,” such as Zeng Zhao and Chen Jian 陳薦, expressed different opinions on other imperial rituals, especially the performance of suburban Altar sacrifices.88

Concluding Remarks The 1079 zhaomu debate among Lu Dian, He Xunzhi, and Zhang Zao left us a rich legacy of Song ritual scholarship, through which we can not only examine the orthopraxy of temple rituals but also reflect upon the consolidation of particular cultural norms. By focusing on the zhaomu sequence in the Song Imperial Temple, Lu, He, 86. Some excerpts of He Xunzhi’s original responses to Lu Dian can be found in Liji jishuo. See Wei, LJJS, 30.33a–36b. 87. Both He Xunzhi and Zhang Zao were appointed by Emperor Shenzong to examine the archive of official reforms in 1080/6; see XCB, 305.7424. 88. XCB, 312.7563.

Yuanfeng Ritual Reforms and the 1079 Zhaomu Debate

103

and Zhang reactivated discussions on some important rules of Confucian families, such as the ritual affiliations between grandfathers and grandsons in ancestral worship. Through debates, the Song court ascertained the affiliations in the name of ancient rituals, but it discarded Lu Dian’s scheme that aimed at a faithful manifestation of patrilineal relations in ritual settings. In the eyes of He Xunzhi, Zhang Zao, Lu Dian, and other ritual officials within and outside the DPATR, the zhaomu in the Imperial Temple signified the generational sequence of the Song imperial line and represented the Song dynastic order. In this light, the ritual controversy surrounding the imperial zhaomu in 1079 stemmed not from an apparently dogmatic fanaticism about ritual formality, but from scholar-officials’ concern about the representation of familial relations in the form of imperial ancestral rituals. The 1079 debate also revealed the tension among ritual officials under the reformist camp. Lu Dian, He Xunzhi, and Zhang Zao were all related to Wang Anshi in one way or another. However, the intellectual tension among He, Zhang, and Lu in the 1079 debate reveals the same cross-factional tendency that we have observed in the 1072 Primal Ancestor debate, yet from a different angle. The tendency is even more pronounced if the political stances of other ritual officials in the 1079 debate are considered. In terms of ritual interests, the difference between “reformers” is more obvious than the one between “reformers” and “conservatives”—after all, reformers and conservatives are political labels based primarily on scholar-officials’ political inclinations. The exploration of scholar-officials’ ritual interests, however, reveals a different picture. In terms of ritual interests, neither did Lu Dian nor He Xunzhi and Zhang Zao adopt in their zhaomu discourse the merit-based approach that frequently appeared in the 1072 Primal Ancestor debate. Compared to the participants of the 1072 debate, Yuanfeng ritual officials placed more emphasis on the arrangement of ancestors in the Imperial Temple in terms of their conceptions of the order of seniority. These officials detached temple rituals from the political context of merits and contributions. Instead, they associated these rituals more with Confucian values. In the following chapters, I will shift from factional politics to ritual scholarship and investigate the intellectual background that bred the 1072 and the 1079 debates over the Imperial Temple—Wang Anshi’s New Learning (xinxue 新學) scholarship.

5 Imperial Temple in the New Learning

Toward to end of Emperor Shenzong’s reign, a wave of ritual reforms and related ritual debates scaled ritual scholarship to its new height. Court officials and scholars outside the court shared a common ground in establishing a new form of ritual learning. In the 1070s and 1080s, most ritual officials and ritualists were influenced by Wang Anshi’s scholarship and his new commentaries of the Classics, namely the Sanjing xinyi 三經新義, comprising the Zhouli xinyi 周禮新義 (New Meanings of the Rituals of Zhou), the Shijing xinyi 詩經新義 (New Meanings of the Book of Songs), and the Shangshu xinyi 尚書新義 (New Meanings of the Book of Documents).1 Historians tend to categorize these scholars into an independent Song intellectual tradition under the name of “New Learning,” in which Wang Anshi’s commentaries on the Classics served as the supreme authority for resolving ritual controversies. Nevertheless, scholars who were influenced by Wang Anshi’s Classical learning came from different intellectual backgrounds. It is unsurprising that their ritual interests are divergent, as we have seen in the case of Lu Dian and He Xunzhi in Chapter 4. Moreover, in their own commentaries on the ritual Classics, New Learning scholars rectified Wang’s ritual interpretations and rendered them new meanings. By exploring the diverse commentaries written by these scholars related to the Imperial Temple, I argue against the conventional interpretation of New Learning ritual scholarship as a simple elaboration of Wang Anshi’s individual learning. Instead, I argue that other New Learning scholars actually elaborated Wang’s conception of the Imperial Temple based on their own commentaries on related passages in the ritual Classics.

An Appraisal of New Learning Ritual Scholarship New Learning as a community is hardly a modern construction. When Wang Anshi was still in charge of state politics in the 1070s, Wang’s political opponents coined some terms to describe officials who supported Wang’s political reforms, such as 1. Cheng Yuanmin 程元敏 has arguably claimed that the original titles of the Sanjing xinyi should be Sanjing yi, comprising Shijing yi, Zhouli yi, and Shangshu yi. See Cheng, Sanjing xinyi jikao huiping 三經新義輯考彙評 (hereafter SSXYJKHP) (Shanghai: Huadong shifan daxue chubanshe, 2011), 759–67.

Imperial Temple in the New Learning 105

“new officials” (xinren 新人) and “kinship partisans” (qindang 親黨). The term “kinship partisans,” probably first raised by Liang Tao, conveys an implicit meaning that all major reformists engaged in Wang Anshi’s political reforms are his relatives by either blood or marriage.2 However, if one carefully examines the list of “kinship partisans” drawn up by Liang Tao, only two of them—Wang Anli 王安 禮 (1034–1095), Wang’s younger brother, and Xie Jingwen 謝景溫 (1021–1097), the brother-in-law of Wang Anli—can be identified as Wang Anshi’s kinsmen. The others are either Wang’s political allies or admirers of his scholarship. The former includes Cai Que 蔡確 (1037–1093), Zhang Dun 章惇 (1035–1105), Zeng Bu, Shu Dan 舒亶 (1041–1103), Lü Huiqing 呂惠卿 (1032–1111), An Tao 安燾 (jinshi. 1059), Pu Zongmeng 蒲宗孟 (1022–1088), Lü Jiawen 呂嘉問, and Zhao Tingzhi 趙挺之 (1040–1107); the latter includes Zeng Zhao, Lu Dian, Huang Lü, Shen Gua 沈括 (1031–1095), Ye Zuqia 葉祖洽 (1046–1117), Zhang Shangying 張商英 (1043–1121), and Peng Ruli 彭汝礪 (1042–1095).3 The above spectrum reveals two distinct types of Wang Anshi’s “kinship partisans.” The first type includes those reformist politicians who energetically participated in the promotion of the New Policies, such as Cai Que, Zhang Dun, Lü Huiqing, Lu Jiawen, An Tao, and so on. The second type refers to those scholars who considered Wang Anshi’s scholarship as more compelling than his political goals, such as Lu Dian, Shen Gua, and Peng Ruli. Although some reformist politicians had studied with Wang Anshi and helped Wang to revise the Sanjing xinyi, it was primarily the second type of “New Learning” scholars who inherited Wang Anshi’s ritual scholarship and developed it into an established intellectual tradition. As Peter Bol has emphasized, Wang Anshi believed in the coherence of the Classics and attempted to achieve this coherence through a comprehensive reading of a myriad of texts and materials.4 Early to the 1050s, Wang Anshi started his reflection on the study of Classic learning. Consequently, he formed an individualized academic discipline during his lecturing period at Jiangning, from 1064 to 1067.5 The Jiangning period of lecturing witnessed a sharp turn of Wang Anshi’s scholarship toward a comprehensive study of Confucian Classics. Although Wang 2. Wang Ruilai 王瑞來, Song zaifu biannianlu jiaobu 宋宰輔編年錄校補 (Beijing: Zhonghua shuju, 1986), 537. Modern historians have already discussed the affinal relations between the reformers surrounding Wang Anshi. Shen Songqin 沈松勤, Beisong wenren yu dangzheng 北宋文人與黨爭 (Beijing: Renming chubanshe, 1998), 184. 3. Interestingly, Liang Tao’s list did not include Wang Anshi’s son, Wang Pang 王雱 (1044–1076), and the husband of Wang Anshi’s younger sister, Shen Jichang 沈季長 (1027–1087). 4. Bol, This Culture of Ours, 228–29. Also see Peter Bol, “Reconceptualizing the Order of Things in Northern and Southern Sung,” in The Cambridge History of China, Vol. 5, Part II: Sung China, 960–1279, ed. John W. Chaffee and Denis Twitchett (Cambridge, UK: Cambridge University Press, 2009), 682–89, especially 685–87. 5. Among the numerous modern studies of Wang Anshi, Liu Chengguo 劉成國 offers the most detailed research about Wang’s lecturing period at Jiangning. By focusing on the regional characteristics of Wang’s Jiangning community, Liu tends to perceive Wang’s Jiangning disciples as an extension of the general Southern scholarship (nanxue 南學) in the Northern Song. Liu Chengguo, Biange zhong de wenren yu wenxue: Wang Anshi de shengping yu chuangzuo kaolun 變革中的文人與文學：王安石的生平與創作考論 (Hangzhou: Zhejiang daxue chubanshe, 2011), 148–69.

106

Empowered by Ancestors

had already composed a new commentary on the Book of Changes in 1058, it was not until the Jiangning period that he emphasized the internal coherence among the Classics.6 According to Lu Dian’s remembrance of Wang Anshi’s Jiangning lectures, Wang repeatedly emphasized the significance of the “basic structure” (dati 大體) of the ancient Way.7 In Wang’s opinions, the “basic structure” split into separate parts after high antiquity, but supposedly every piece of the existing Classics preserved a part of it. In other words, every piece of Classics shared a part of the ancient Way. Therefore, a careful integration of all the Classics would reveal the “basic structure” and the “oneness of the Way” (daozhiyi 道之一).8 In order to achieve the “oneness of the way,” Wang asserted the importance of studying the nuanced nature of the mind, which essentially resonated with the ethics-based ontology of the two Cheng brothers.9 In contrast to the Cheng brothers, Wang was less concerned about the philosophical interpretation of particular Classics, especially the Book of Changes. Instead, he preferred to perceive the Classics as an integrated whole that could be explained only in an interconnected way. To grasp the secret of the Way, one has to “see the entity of the Classics.”10 To seek the entity of the Classics, Wang further argued that the Book of Songs and the Three Ritual Classics—the Rituals of Zhou, the Book of Rites, and the Rites and Ceremonies—could be mutually interpreted, because ancient sages composed these texts based on the same principle.11 Wang’s approach of mutual interpretation, in addition to a comprehensive understanding of various intellectual traditions, is at the heart of his ritual scholarship. In general, Wang Anshi perceived the learning of rituals as a rather elusive goal for scholars to achieve. Therefore, he argued that when scholars studied the Classics, they should start with the easiest ones, such as the Book of Songs and the Book of Documents. Only if scholars finished studying these “preliminary” Classics, they could proceed to study the Book of Rites.12 Interestingly, despite Wang Anshi’s emphasis on ritual learning and his composition of the Zhouli xinyi, his contemporaries usually regarded him as a scholar who opposed ritual scholarship. In reviewing Wang Anshi’s ritual interest in the 1072 controversy over the Imperial Temple, Shao Bo 邵博 (d. 1158) stated that Wang “despised ritual scholarship and took an abnormal stand on related topics” 王荊 公薄禮學，又喜為異.13 Shao’s statement represented Song scholars’ conventional understanding of Wang Anshi’s ritual scholarship, in which Wang was depicted as a crafty interpreter of the Rituals of Zhou for his political ends.14 Against Wang Anshi’s emphasis on the Rituals of Zhou, some Song scholars even questioned the 6. Liu Chengguo, Jinggong xinxue yanjiu 荊公新學研究 (Shanghai: Shanghai guji chubanshe, 2006), 21–28. 7. Lu Dian, “Da Li Bi shu” 答李賁書, Taoshan ji, 12.7b–8b. 8. Lu, “Da Li Bi shu,” 12.8a. 9. Liu, Biange zhong de wenren yu wenxue, 162–65. 10. Wang, “Da Zengzigu shu” 答曾子固書, LCJ, 51:73.469. 11. Wang, “Da Wuxiaozong shu” 答吳孝宗書, LCJ, 51:74.474. 12. See Lu Dian, “Da Cui Zifang xiucai shu” 答崔子方秀才書. Taoshan ji, 12.12b. 13. Shao Bo, Shaoshi wenjian houlu 邵氏聞見後錄 (Beijing: Zhonghua shuju, 1997), 1.6. 14. See, for example, Ye Shi, Lijing huiyuan 禮經會元, SKQS ed., 92:1.4b.

Imperial Temple in the New Learning 107

authenticity of this ritual Classic.15 There is a famous story about Wang Anshi’s ritual scholarship, which records that Wang persuaded Emperor Shenzong to shelve the series of imperial lectures on the Book of Rites, owing to his ignorance of the ritual details recorded in it.16 The above story is unconvincing for some obvious reasons. Wang Anshi’s eminence and his proficiency in Confucian Classics were well recognized by his contemporaries, including his political opponents. It is implausible that Wang was ignorant about the Book of Rites. Moreover, in an early letter to his friend Zeng Gong, Wang explicitly expressed his interest in “composing some writings after reading the Book of Rites” 所云讀《禮》，因欲有所論著.17 In fact, Wang Anshi composed two scholarly monographs on the Book of Rites, namely the Lijing yaoyi 禮經要義 and the Liji faming 禮記發明. Although both works have been lost, their very existence contradicts the impression that Wang Anshi overlooked the learning of the Book of Rites. There is more textual evidence to substantiate the New Learning emphasis on a comprehensive training of ritual learning if Wang Anshi’s students and academic followers are taken into consideration. Most members of Wang Anshi’s circle had devoted sufficient and balanced attention to the Three Ritual Classics. In fact, a thorough study of ritual details recorded in especially the Book of Rites constitutes a great portion of New Learning ritual scholarship. Table 5.1 (see pp. 108–9) surveys the ritual writings of scholars whose learning was either affiliated with Wang Anshi’s scholarship or shaped by Wang’s advocacy of ritual learning. Among the twenty-two works listed above, only four of them (1, 4, 21, and 22) deal with the study of the Rituals of Zhou.18 In contrast, ten of them (2, 3, 5, 7, 9, 11, 12, 13, 17, and 20) focus on the Book of Rites and the ritual details recorded in it; two works (10 and 16) focus on the Rites and Ceremonies; four works (6, 8, 14, and 15) are comprehensive reviews of ritual theories and practices in all the Three Ritual Classics; one (18) examines the evolution of ritual music and the scales of instruments; and one (19) addresses the Yuanfeng ritual reforms. Statistically, the

15. For example, see Shao Bo, Shaoshi wenjian houlu, 3.23, and Hu Hong 胡宏 (1105–1155), “Jilun Zhouli” 極論 周禮, Huhong ji 胡宏集 (Beijing: Zhonghua shuju, 1987), 259–60. 16. Both Zhu Bian 朱弁 (1085–1144) and Lu You 陸游 (1125–1210) recorded this story in their pen-notes. According to Zhu and Lu, Wang was embarrassed by Shenzong’s question concerning the procedure of changing the mat of a deceased ancestor (yize 易簀)—a rite that was recorded in the Book of Rites. Zhu Bian, Quwei jiuwen 曲洧舊聞 (Beijing: Zhonghua shuju, 2002), 9.208; Lu You, Laoxuean biji 老學庵筆記, Song Yuan biji xiaoshuo daguan 宋元筆記小說大觀 (Shanghai: Shanghai guji chubanshe, 2007), Vol. 4, 3539. 17. Zeng, “Yu Wang Jiepu disanshu” 與王介甫第三書, Yuanfeng leigao 元豐類藁, SBCK, 1680–1689:16.127. Cai Shangxiang 蔡上翔 (1717–1810) dated this letter to the 1065 winter. Cai, Wang Jinggong nianpu kaolue 王荊公 年譜考略 (Beijing: Zhonghua shuju, 2006), 400; also see Li Zhen 李震, Zeng Gong nianpu 曾鞏年譜 (Suzhou: Suzhou daxue chubanshe, 1997), 216–17. Liu Chengguo dated the letter to 1064. See Liu, Wang Anshi nianpu changbian 王安石年譜長編 (Beijing: Zhonghua shuju, 2018), 672. 18. Lin Zhiqi 林之奇, a critic of Wang Anshi’s New Learning, wrote a commentary on the Rituals of Zhou based on the Zhouli xinyi. Because Lin regarded himself as an anti–New Learning scholar, I choose not to include his work in this survey. Wang Yuzhi, Zhouli dingyi, 93: preface.2b.

Empowered by Ancestors

108

Table 5.1: A Survey of Ritual Writings Composed by New Learning Scholars No Author

Title of the Ritual Writing

Editorial information Source

1

Wang Anshi 王安石

Zhouli xinyi 周禮新義

22 juan (excerpts preserved in YLDD, 16 juan)

2

Wang Anshi

Lijing yaoyi 禮經要義 2 juan (lost)

3

Wang Anshi

Liji faming 禮記發明

1 juan (lost)

4

Wang Zhaoyu 王昭禹 (fl. 1080)

Zhouli xiangjie 周禮 詳解

40 juan (preserved in SKQS, vol. 91 SKOS)

5

Fang Que 方愨

Liji jie 禮記解

20 juan (lost, excerpts ZZSLJT, 2.24b–25a; preserved in LJJS) SS, 202.5050

6

He Xunzhi 何洵直 (jinshi. 1078)

Lilun 禮論

1 juan (lost, excerpts preserved in LJJS)

7

Ma Ximeng 馬希孟

Liji jie 禮記解

70 juan (lost, excerpts ZZSLJT, 2.25a; preserved in LJJS) SS, 202.5050

8

Lu Dian 陸佃

Lixiang 禮象

15 juan (lost)

9

Lu Dian

Liji jie 禮記解

40 juan (lost, excerpts SS, 202.5049 preserved in LJJS)

10

Lu Dian

Yili yi 儀禮義

17 juan (lost)

SS, 202.5050

11

Lu Dian

Shuli xinshuo 述禮新說

4 juan (lost)

SS, 202.5050; LJJS, Mingshi.4

12

Lu Dian

Daqiu yi 大裘議

1 juan (preserved in TSJ)

SS, 202.5050; TSJ, juan 5

13

Lu Dian

Shenyi zhidu 深衣制度

No record

SCTSM

14

Chen Xiangdao Lishu 禮書 陳祥道 (1042–1093)

150 juan (preserved in SKQS)

SKQS, vol. 130; SS, 202.5050; ZZSLJT, 2.27b; JZDSC, 90

15

Chen Xiangdao Lili xiangjie 禮例詳解 10 juan (lost)

SS, 202.5050

16

Chen Xiangdao Zhujie Yili 註解儀禮

32 juan (lost)

SS, 202.5050

17

Chen Xiangdao Liji jiangyi 禮記講義

24 juan (lost)

YH, 39.33a

18

Chen Yang 陳暘 (1064–1128)

200 juan (preserved in SKQS)

ZBSYXA, 98.20b; SKQS, vol. 211

Yueshu 樂書

SKQS, vol. 91; SSXYJKHP vol.2; JZDSC, 81–82 JZDSC, 1094 LJJS, Mingshi 名氏.5

SS, 202.5050

SS, 202.5049

Imperial Temple in the New Learning 109

No Author

Title of the Ritual Writing

Editorial information Source

19

Yang Wan 楊完

Yuanfeng jiaomiao fengsi liwen 元豐郊廟奉祀禮文

30 juan (lost, excerpts WXTK, 187.1598; JZDSC, 83–84; preserved in YH, GLJ, and other Song GLJ, juan 9 anthologies)

20

Yang Xun 楊訓

Liji jie 禮記解

20 juan (lost)

GBSYXABY, 876

21

Zheng Zongyan Kaogong ji zhu 鄭宗顏 考工記註

1 juan (preserved in SKQS under the name of Kaogong ji jie 考工記解)

JYK, 129; SKQS, vol. 91

22

Gong Yuan 龔原

10 juan (lost)

SS, 202.5050

Zhouli tu 周禮圖

Index of sources: SKQS: Siku quanshu 四庫全書（上海古籍出版社影印文淵閣本）; JZDSC: Junzhai dushuzhi 郡齋讀書志（上海古籍出版社孫猛校證本）; ZZSLJT: Zhizhai shulujieti 直齋書 錄解題（台北廣文書局本）; JYK: Jingyi kao 經義考（中華書局本）; ZBSYXA: Zengbu Song-Yuan xuean 增補宋元學案（台北中華書局本）; GBSYXABY: Gaoben Song-Yuan xuean buyi 稿本宋元 學案補遺（北京圖書館出版社本）; SS: Songshi (yiwenzhi) 宋史藝文志（中華書局本）; WXTK: Wenxian tongkao 文獻通考（北京中華書局本）; LJJS: Liji jishuo 禮記集說（文淵閣四庫本）; GLJ: Guling ji 古靈集（四庫珍本叢書三集）; TSJ: Taoshan ji 陶山集（文淵閣四庫本）; YH: Yuhai 玉海（文淵閣四庫本）; YLDD: Yongle dadian 永樂大典; SSXYJKHP: Sanjing xinyi jikao huiping 三 經新義輯考彙評（程元敏輯本）; SCTSM: Suichutang shumu 遂初堂書目（中華書局本）.

110

Empowered by Ancestors

ritual scholarship of New Learning demonstrates a significant percentage of over 40 percent (10 out of 22) on the study of the Book of Rites. Although the Rituals of Zhou had occupied a privileged position in the civil service examinations since the 1070s, studies concerning the Book of Rites and its records of ritual details had never diminished within the New Learning circle. While New Learning scholars turned Wang Anshi’s highly individualized learning into a more inclusive “disciplinary matrix,” New Learning ritual scholarship underwent significant intellectual transformations.19 Taking the Imperial Temple as an example, I will demonstrate that New Learning ritual scholarship as a “disciplinary matrix” was usually characterized by complexity and ambiguity. Only near the end of Northern Song, under Emperor Huizong’s reign, the disciplinary matrix converged toward a common interest of reviving ancient rituals.

From Wang Anshi to Wang Zhaoyu: New Learning Conception of the Imperial Temple In the two ritual debates of 1072 and 1079, Wang Anshi, Lu Dian, and He Xunzhi demonstrated great interest in rectifying the arrangement of the Imperial Temple and temple rituals. As previously mentioned, political stances hardly mattered in these ritual debates. Within the New Learning community, intellectual factors contributed more to the differentiation of ritual interests concerning the temple. During the 1070s, along with the compilation of the Sanjing xinyi and its installation in the civil service examinations, the New Learning community expanded rapidly.20 Particularly, the official publication of the Zhouli xinyi in 1075 codified Wang Anshi’s ritual interests and his emphasis on the Imperial Temple. The Zhouli xinyi serves basically as a blueprint for the political and financial agendas of Wang Anshi’s New Policies. Wang’s commentaries on the first two sections of the Rituals of Zhou, Bureau of Heaven (tianguan 天官) and the Bureau of Earth (diguan 地官), deal with state policies in relation to administration, education, and finance.21 Song Jaeyoon has thoroughly studied Wang’s commentaries on these two sections. A less-studied section of the Zhouli xinyi is the Bureau of Spring 19. I borrow the term “disciplinary matrix” from Thomas Kuhn. In the 1969 Postscript to his classic work, The Structure of Scientific Revolutions, Kuhn proposed using the term “disciplinary matrix,” instead of his influential usage of “paradigm,” to designate the professional communication between the members of a scientific community. Thomas Kuhn, The Structure of Scientific Revolutions (Chicago: The University of Chicago Press, 1970), 182–87. Although Kuhn’s disciplinary matrix theory has been primarily adapted in the sociological studies of professionals, I find in Song New Learning scholarship the same regulatory factors. In contrast to the conventional understanding of New Learning as a loose composition of miscellaneous ideas, or an odd combination of Legalism and radical Confucianism, I tend to view it as a disciplinary matrix that has been characterized by a shared belief of ancient rituals and a complete set of intellectual codes, especially the discourse of “ancient sage-kings.” 20. XCB, 243.5917. Also see Kondō Kazunari 近藤一成, “Wang Anshi de keju gaige” 王安石的科舉改革, in Riben zhongqingnian xuezhe lun zhongguoshi: songyuanmingqing juan 日本中青年學者論中國史：宋元明清卷 (Shanghai: Shanghai guji chubanshe, 1995), 154–57. 21. Song, Traces of Grand Peace, 116–94.

Imperial Temple in the New Learning 111 Figure 5.1: Wang Anshi’s Perception of the Imperial Temple

(chunguan 春官), which crystallizes Wang Anshi’s understanding of court rituals and their cultural implications. In the Zhouli xinyi, Wang Anshi proclaimed that the top priority in state building was to construct an Imperial Temple.22 The most provocative idea in the Zhouli xinyi’s perception of the Imperial Temple is an introduction of yang 陽 and yin 陰 concepts in positing temple locations. Wang perceived the administrative place (chao 朝) in the south as a yang position and the market place (shi 市) in the north as a yin one; likewise, he perceived the Imperial Temple on the left side of the imperial palace as a yang position and the State Altar of Grain and Soil on the right side of the palace as a yin one (see Figure 5.1).23 Wang Anshi argued that the left position of the imperial palace as a yang position symbolizes “the dwelling of the humanly Way” 人道之所鄉.24 According to Wang, the Imperial Temple, where the deceased ancestors dwell, should be spatially arranged similar to the way of serving living humans. Therefore, it should be established on the left-yang side to convey the idea that ancestors in the temple were symbolically immortal (busi 不死).25 In coping with his idea of symbolic immorality, Wang Anshi advocated making regular offerings to temple ancestors.26 He further clarified the duties of different 22. Wang, Zhouli xinyi, SSXYJKHP, 1.22; XCB, 265.6493. 23. Wang, Zhouli xinyi, SSXYJKHP, 8.292 (the locations of the Imperial Temple and the State Altar of Grain and Soil), 18.612 (the locations of the administrative place and the market place). 24. Wang, Zhouli xinyi, SSXYJKHP, 8.292. 25. Wang, Zhouli xinyi, SSXYJKHP, 8.292. Wang Anshi’s conception of the Imperial Temple on the left of the palace as “the dwelling of the humanly Way” was inherited by later scholars, despite their hesitant acknowledgment. Cheng E 鄭諤, a contemporary of Wang Anshi, also named the Imperial Temple as “the dwelling of the humanly Way” in his private commentary on the Rituals of Zhou, the Zhouli jiangyi 周禮講義. The Zhouli jiangyi was lost; I quoted it from Wang Yuzhi, Zhouli dingyi, SKQS, 93:4.28a. 26. Wang, Zhouli xinyi, SSXYJKHP, 9.317.

112

Empowered by Ancestors

offices in the Zhou bureaucracy to construct a more systematic rubric of temple rites and ceremonies. Some of Wang’s clarifications are rather technical, such as the format of wine vessels and the service of the shoutiao office in preserving the relics of ancestors.27 Notably, he expressed special interest in strictly defining the zhaomu sequence of temple ancestors as Lu Dian did in the 1079 zhaomu debate. In his commentary on the duty of the xiaozongbo office, Wang Anshi stated that “the zhao designation conveys a meaning of examining the inferior; the mu designation conveys a meaning of respecting the superior” 昭穆者，昭以察下為義，穆以敬上 為義.28 Comparing Wang Anshi’s statement in the Zhouli xinyi to Lu Dian’s wording in his memorial “Zhaomu yi,” Wang’s Zhouli xinyi seemingly had offered his disciple Lu Dian a ready definition in constructing the latter’s own zhaomu theory against He Xunzhi and Zhang Zao. From Wang Anshi’s Zhouli xinyi in 1075 to Lu Dian’s “Zhaomu yi” in 1079, there was a clear continuity in the New Learning conception of zhaomu, in which zhao designations were rendered as superior.29 In his commentary on the xiaoshi office, Wang Anshi elaborated this hierarchical understanding of the zhaomu sequence used in temple rituals: “Father-ancestors are designated as zhao ancestors and son-ancestors are designated as mu. The sequence of the succession of father-ancestors and son-ancestors is named as generation. What comes from the accumulation of generations is named as genealogy” 父謂之昭，子謂之 穆；父子相代謂之世，世之所出謂之繫.30 According to Wang Anshi, “to ascertain the origin of a lineage, it is necessary to rectify the lineage’s genealogy and involved generations; in order to comprehend a genealogical order, it is necessary to distinguish between zhao and mu designations” 奠繫世，以知其本所出；辨昭穆，以 知其世序.31 Like Wang Anshi, his disciples and academic followers demonstrated a keen interest in the Imperial Temple in their ritual commentaries. Most of these commentaries have been lost. Among the surviving texts, Wang Zhaoyu 王昭禹’s (fl. 1080) Zhouli xiangjie 周禮詳解 (Detailed Explanations of the Rituals of Zhou) provides the most comprehensive explication of the Rituals of Zhou from the perspective of New Learning scholarship. A thorough study of Wang Zhaoyu’s personal 27. Wang, Zhouli xinyi, SSXYJKHP, 9.302, 9.321. 28. Wang, Zhouli xinyi, SSXYJKHP, 8.292. 29. In the “Zhaomu yi,” Lu Dian put it this way: “the zhao designation conveys a meaning of illuminating the inferior; the mu designation conveys a meaning of revering the superior” 昭以明下為義，穆以恭上為義. The interchangeable usage of the four characters, ming 明, cha 察, gong 恭, and jing 敬 in the Zhouli xinyi and the “Zhaomu yi” may convey a special meaning, if we consider the New Learning emphasis on the study of written characters. Indeed, the etymological origin of the two compound words “perspicacious investigation” (mingcha 明察) and “due respect” (gongjing 恭敬) can be traced back to a second-century supplementary text to the Analects, the Kongzi jiayu 孔子家語. For a general introduction of the Kongzi jiayu and its author, see R. Kramers, K’ung Tzu Chia Yu: The School Sayings (Leiden: Brill, 1950), 15–36; 54–90. For the original text, see Wang Su, Kongzi jiayu, SKQS, 695:3.19a. The same reference also appears in a Han-compiled annotation on the Book of Songs, the Hanshi waizhuan 韓詩外傳. See Han Yin 韓嬰 (fl. 150 BC), Hanshi waizhuan jishi 韓詩外傳集釋, comp. Xu Weiyu 許維遹 (Beijing: Zhonghua shuju, 1980), 205–6. 30. Wang, Zhouli xinyi, SSXYJKHP, 11.377. 31. Wang, Zhouli xinyi, SSXYJKHP, 11.377.

Imperial Temple in the New Learning 113

scholarly interest is difficult, owing to limited sources about him.32 Except for the Zhouli xiangjie, there are no other materials about Wang Zhaoyu’s life and thought. Only his courtesy name—Guangyuan 光遠—is known to us.33 A sixteenth-century edition of a Yuan scholar’s commentary on the Rituals of Zhou offers a little more information about Wang Zhaoyu’s native place. The first several pages of the commentary include a catalogue of major Song commentators on the Rituals of Zhou. According to that catalogue, Wang Zhaoyu was born and raised in Linchuan 臨 川—Wang Anshi’s hometown.34 The Song bibliographer Chen Zhensun mentioned that Song scholars studied the Zhouli xiangjie in preparation for the civil service examinations during Huizong’s reign.35 The Zhouli xiangjie as an examination manual echoed Wang Anshi’s earlier reforms on the central educational system in the 1070s.36 Together with other examination textbooks that were composed by New Learning scholars like Wang Pang, Cai Bian, Gu Tang 顧棠, Gong Yuan 龔原 (jinshi. 1063), and Geng Nanzhong 耿南仲 (jinshi. 1082), Wang Zhaoyu’s Zhouli xiangjie was circulated within the examination field throughout the late years of Northern Song. Nonetheless, it is a mistake to read the Zhouli xiangjie as merely an annotated explanation of Wang Anshi’s Zhouli xinyi, which would demean its scholarly value. Although Wang Zhaoyu frequently cited and paraphrased the Zhouli xinyi, he did suggest new understandings that expanded the latter’s perspective toward ancient rituals. Regarding the duties of Zhou officials in charge of the Imperial Temple, Wang Zhaoyu enriched Wang Anshi’s original explanations in the Zhouli xinyi. He was possibly the first scholar who realized the overlap of duties of the two offices of the Vice Minister (xiao zongbo) and the Minor Scribe (xiaoshi) in Zhou bureaucracy. According to the Rituals of Zhou, both the Vice Minister and the Minor Scribe were responsible for arranging the zhaomu sequence of imperial ancestors in the Zhou temple. A meticulous reader would probably ask why there were two officials offering the same service in the ideal Zhou bureaucracy. A failure to answer this question might have resulted in a collapse of confidence on the authenticity of the Rituals of Zhou and thus the textual foundation of New Learning ritual scholarship. While Wang Anshi failed to explicate this conflict of duty in the Zhou bureaucracy, Wang Zhaoyu offered a clear explanation in his Zhouli xiangjie. 32. Song Jaeyoon has already analyzed some crucial parts of the Zhouli xiangjie, especially those concerning education and social welfare. Song, Traces of Grand Peace, 169–317. 33. Wei, LJJS, mingshi.5b; also see Zengbu Songyuan xuean 增補宋元學案 (hereafter ZBSYXA), compiled by Huang Zongxi 黃宗羲 (1610–1695) and Quan Zuwang 全祖望 (1705–1755), 98.19b. 34. Qiu Kui 丘葵 (1244–1333), Zhouli buwang 周禮補亡, compiled by Gu Kejiu 顧可久 (1485–1561). Harvard Yenching Rare Book Collections, the Ming edition (1465–1620), printed by Li Qi 李緝 in wooden block prints, Zhi Zhouli xingshi 治周禮姓氏, 2a. 35. Chen Zhensun, Zhizhai shulujieti, 2.21b. 36. John Chaffee, The Thorny Gates of Learning in Sung China: A Social History of Examinations (New York: State University of New York Press, 1995), 76–77. Yang Tianbao 楊天保 demonstrated that before the 1060s Wang Anshi was more inclined to a utilitarian learning of achieving degrees and practicing administrative skills. Yang Tianbao, Jinling wangxue yanjiu: Wang Anshi zaoqi xueshu sixiang de lishikaocha, 1021–1067 金陵王學 研究：王安石早期學術思想的歷史考察, 1021–1067 (Shanghai: Shanghai renmin chubanshe, 2008), 145–82.

114

Empowered by Ancestors

According to Wang Zhaoyu, whereas the Vice Minister managed the zhaomu sequence of ancestors in the Zhou Imperial Temple, the Minor Scribe archived the genealogical documents of the Zhou lineage and involved generations (xishi 繫 世).37 By filling the rupture between the texts of the Rituals of Zhou with his own interpretation, not only did Wang ameliorate the New Learning conception of the zhaomu sequence, but he also explicated the main text of the Classics in a more productive way by which he challenged the conventional annotative practice of “leaving the suspicious text untouched” (cun er bulun 存而不論). In the Zhouli xiangjie, Wang Zhaoyu also explicated the tiao temples based on an analysis of the written character tiao 祧. By interpreting the left radical (shi 示, literally, demonstrate) and the right part (shao 兆, literally, symbolize) of the character tiao as semantic components, Wang argued that the zhaomu system naturally regulated the two tiao temples.38 In terms of ritual performance, the left tiao temple demonstrates the first zhao ancestor and the right tiao temple symbolizes the first mu ancestor. Regarding the Imperial Temple complex, multiple zhao and mu temples “arrange ancestors downward according to the principle of integrity” 以義率祖順而下之也; with the Primal Ancestor temple, the multiple zhao and mu temples “arrange ancestors upward according to the principle of benevolence” 以 仁率親等而上之也.39 By conceptualizing the tiao temples and the zhaomu system based on the Confucian ideas of benevolence and integrity, Wang Zhaoyu attached a moral value to the arrangement of temples. His interpretation particularly resonates with the saying of Confucius recorded in the Book of Rites: “seasonal sacrifices in the temple, such as chang and di, are used to symbolize the notion of benevolence through the zhaomu sequence” 嘗禘之禮，所以仁昭穆也.40 Ethically and introspectively, Wang Zhaoyu advocated the intimacy between ancestral worship and moral values. He connected the exterior ritual performance of zhaomu with the interior virtue of benevolence. Wang Zhaoyu’s interpretation resembles the moralistic reading of zhaomu within the circle of New Learning scholars. Wang Anshi, Lu Dian, and He Xunzhi all correlated zhaomu with the Confucian concept of seniority, albeit in different ways. Wang Zhaoyu’s introspective vision of zhaomu conceptualizes the ritual sequence of ancestral temples as a symbolization of Confucian virtues. Concurrently, Wang Zhaoyu challenged the merit-based approach in understanding the Imperial Temple. Song scholar-officials would dispute this on the priority of the two ideas of merits and seniority in temple rituals. Yet few of them would prioritize merits over 37. Wang Zhaoyu, Zhouli xiangjie, SKQS, 91:18.2b; 23.11a. Xi 系 refers to the imperial genealogical record, also called dixi 帝系; shi 世 refers to the genealogical record of the feudal lords, also called shiben 世本. See Jia Gongyan’s sub-commentary of the Rituals of Zhou. Zhouli zhushu, TSZSSSJ, 2:26.259. For a Han understanding of the lineage of ancient sage-kings, see the “Dixi” 帝系 chapter in the Dadai Liji 大戴禮記, Wang Pinzhen 王聘珍 (eighteenth century), Da Dai liji jiegu 大戴禮記解詁 (Beijing: Zhonghua shuju, 1983), 126–30. 38. Wang, Zhouli xiangjie, 91:19.21a. 39. Wang, Zhouli xiangjie, 91:19.21a. 40. Zhu Bin, Liji xunzuan, 746; James Legge, trans., The Sacred Books of China, 4:271.

Imperial Temple in the New Learning 115

the moral value of benevolence. In this light, Wang Zhaoyu’s Zhouli xinyi demonstrates how New Learning ritual scholarship offered new insights that contributed to a more ethical and metaphysical conception of ritual—a conception that frequently surfaced in later decades.

Chen Xiangdao’s Conception of the Imperial Temple in the Lishu Compared with Wang Zhaoyu’s Zhouli xiangjie, Chen Xiangdao’s 陳祥道 (1042– 1093) Lishu 禮書 (Ritual Manual) is a more ambitious enterprise that addresses numerous ritual issues and controversies in a thematic way. Chen Xiangdao, whom the Northern Song scholar Li Zhi praised as an extraordinary ritual scholar, received favor from Wang Anshi in Li’s early career.41 Because of his erudite knowledge in ritual learning, the court appointed Chen to the Court of Imperial Sacrifices as an erudite in 1089.42 Unfortunately, the erudite in the Court of Imperial Sacrifices was the highest court position Chen Xiangdao ever achieved.43 Chen’s implication in his father’s crime destroyed his political career.44 However, on the intellectual level, his ritual learning was well acknowledged by his contemporaries. The fact that Chen Xiangdao was promoted during the Yuanyou era shows how his achievement in ritual learning had left a deep impression on the minds of conservative scholar-officials. As his lifetime work, the Lishu crystallizes Chen’s ritual learning. In 1090, after Chen Xiangdao expanded the content of Lishu,45 Fan Zuyu 范祖禹 (1041–1098), a steadfast conservative, claimed that the Lishu surpassed Nie Chongyi’s Sanlitu jizhu in terms of ritual scholarship.46 The Yuan-scholar Yu Ji 虞集 (1272–1348) even considered the Lishu as a representation of the Yuanyou scholarship. In making such a claim, Yu certainly avoided mentioning Chen Xiangdao’s New Learning background.47 The Lishu is indeed a milestone research in the development of Song ritual scholarship. Its theoretical framework rests on Wang Anshi’s conception of reviving ancient rituals. Thus, it challenges the pragmatic approach taken by most ritual officials in composing ritual manuals.48 Not only does the Lishu explicate ancient ritual practices in detailed texts, but it also illustrates ritual garments, utensils, and concrete spatial settings in parallel with the texts in the beginning of each section. In 41. Li, Shiyou tanji, 2002), 33. For Chen Xiangdao’s relationship with Wang Anshi, see Cheung Hiu Yu, “Cong Yuan-Ming dixiu baiwushijuanben Lishu luelun Chen Xiangdao Lishu de jinxianguocheng ji yiyi” 從元明 遞修百五十卷本《禮書》略論陳祥道《禮書》的進獻過程及意義, Lishi wenxian yanjiu 歷史文獻研究 39 (2017.9): 291–300, especially 296–300. 42. Li Zhi dated Chen Xiangdao’s promotion to 1093. Yet, Li Tao dated it to 1089. XCB, 422.10210. 43. Gong Yanming, “Yuanfeng qianhou liang Song wenguan jilu guanjie duizhao biao,” 759. 44. Li, Shiyou tanji, 33. 45. XCB, 450.10808. 46. Fan Zuyu, “Qi kanxiang Chen Xiangdao Lishu zhazi” 乞看詳陳祥道禮書劄子, Fan taishiji 范太史集, SKQS, 1100:19.16a. 47. Cheung, “Cong Yuan-Ming dixiu baiwushijuanben Lishu,” 299. 48. Cheung, “Cong Yuan-Ming dixiu baiwushijuanben Lishu,” 297–99.

116

Empowered by Ancestors

particular, Chen Xiangdao’s illustrations provide a valuable record of the Imperial Temple configuration, including its walls, entrances, and basic structure. In juan seventy-one, Chen attached a precise diagram about how spirit tablets of imperial ancestors should be placed in the di sacrifice.49 With these illustrations (see Illustration 5.1 on p. 117), readers of the Lishu can visualize at least the basic spatial orientation of the temple. Chen’s discussion on the Imperial Temple contains several key topics. His explication of the general configuration of the Imperial Temple, in particular, serves as one of the best reviews of previous ritual controversies over the number of temples. On the one hand, by citing the Tang Confucian Kong Yingda’s subcommentary on the Book of Rites, Chen revealed that Zheng Xuan’s argument was grounded on some suspicious passages from Eastern-Han apocryphal texts rather than solid evidence from the Classics.50 On the other hand, Chen criticized Wang Su for his insistence on a strict configuration of seven temples. To distinguish himself from Zheng and Wang, Chen Xiangdao argued for a more flexible arrangement of temples by quoting other Han sources. One of them reads, “If the Son of Heaven has seven ancestors, there should certainly be seven temples; if the Son of Heaven has less than seven ancestors, a five-temple setting is enough. Yet, for the feudal lords, even if they have more than five ancestors, they cannot exceed the limitation of five temples” 天子七廟，有其人則七，無其人則五。若諸候廟制，雖有 其人，不得過五.51 In the excerpts of Chen Xiangdao’s other ritual writings, Chen argued that the seven-temple setting of the Son of Heaven symbolizes the furthest realm under the influence of the imperial clan’s virtue. In numerology, the numbers seven and five emblematize the “utmost benevolence and integrity” of the clan’s spiritual power 仁之至義之盡.52 In other words, seven and five are fixed numbers corresponding to temple construction. The emperor as the Son of Heaven should not degrade his Imperial Temples from seven to five, even though he does not have seven ancestors. Like Wang Zhaoyu, Chen conceptualized the tiao temples as the symbolization of the transition of ancestral spirits.53 By definition, the Son of Heaven with seven ancestors designates his fifth and sixth ancestors as the two tiao ancestors. Since Zhou feudal lords took their first ancestors as tiao ancestors, the character tiao in Zhou feudalism referred to “origin.”54 However, in contrast to Wang Zhaoyu, Chen argued that the two tiao temples should not count toward the seven temples of the Son of Heaven if there are more than seven ancestors.55 According to Chen, there is a fundamental difference between the tiao temple and the Imperial Temple. 49. Chen Xiangdao, Lishu, SKQS, 130:71:1b–2a. 50. Chen, Lishu, 130:67.13b. 51. Chen, Lishu, 130:67.14b; 67.17a–b. Wei, LJJS, v. 117, 30:25. 52. Wei, LJJS, 117.30.25a. 53. Chen, Lishu, 130:68.5a–6b. 54. Chen, Lishu, 130:68.5a. 55. Chen, Lishu, 130:67.17a.

Imperial Temple in the New Learning 117 Illustration 5.1: Visualization of the Arrangement of Spirit Tablets in the Di Sacrifice

Source: Chen Xiangdao 陳祥道, Lishu 禮書, SKQS, 130:71:1b–2a.

118

Empowered by Ancestors

When temple ancestors are removed from the main hall of the Imperial Temple and placed successively in the tiao temple, the dan altar, and the shan yard, their spirits shift from one sacrificial space to another, as well as the formality of related rituals. Therefore, Chen Xiangdao not only elaborated the New Learning conception of tiao temples, but he also redefined the very nature of regular and tiao temples as different parts of the same chain of ritual spaces for worshiping deceased ancestors. Chen Xiangdao’s explanation of temple configuration in Lishu also contributes to a more sophisticated understanding of temple rituals. Chen fairly assimilated and integrated the early interpretations of temple rituals. Tension sometimes emerges in his work of assimilation and integration. For instance, by revising the earlier model drafted by Nie Chongyi, Chen Xiangdao depicted the mingtang as a threeby-three architectural complex of five sacrificial chambers—in referring to the “five phases” of gold, wood, water, fire, and earth—and four Imperial Temples, in which cosmic elements were linked to temple sacrifices.56 Concerning the arrangement and removal of spirit tablets in temple rituals, Chen was especially interested in maintaining a certain degree of flexibility. To Chen, whether a temple or a tablet was “removable” depended primarily on the merits of the ancestor within.57 Thus, Chen indicated that ancestors with outstanding merits should be excluded from the removal list—a viewpoint in sharp contrast with Wang Zhaoyu’s argument in the Zhouli xiangjie.58 Chen Xiangdao also distinguished himself fundamentally from both Wang Zhaoyu and Lu Dian in the understanding of the zhaomu sequence in the temple. Since Chen’s arguments profoundly reflected the conceptual tension within New Learning ritual scholarship, I quote it at length here: The Imperial Temple may be renovated or destroyed, yet its zhaomu should never be altered in any way. While the Zuo Commentary on the Annals mentions Dawang as a zhao ancestor and Wang Ji as a mu ancestor, it also mentions King Wen as a zhao ancestor and King Wu as a mu. This proves that the zhaomu of genealogy should never be altered. The Grave Maker office Zhongren in the Rituals of Zhou is in charge of imperial graves and burials. In the burial ground the tomb of the first ancestor is always situated in the middle, and other tombs are arranged on both sides according to the left-zhao and right-mu setting. This proves that the zhaomu of burial grounds should never be altered. The Rites and Ceremonies records that after the sacrifice of “stop-wailing,” the tablets of deceased ancestors are attached to their corresponding temples according to the order of descent.59 As the Sangfu 56. Chen, Lishu, 130:89.1b; Nie Chongyi, Sanlitu jizhu, 129:4.2a–3a, 4.24a–b. 57. Chen, Lishu, 130:68.4b–5a. 58. Wang, Zhouli xiangjie, 91:18.2b. 59. In the Rites and Ceremonies, it records that the ritual of tablet attachment follows an order of descent (ban 班). But the text fails to mention how this ban is arranged in practice. The Qing ritualist Lu Wenchao 盧文 弨 (1717–1796), who annotated the Rites and Ceremonies, provided no sub-commentary for this phrase. The only thing we know about the term ban here is that it was written as pang 胖 in the Han New-Text writings, according to Zheng Xuan’s commentary. Zheng himself annotated the character pang as ci 次 (order). Lu Wenchao, Yili zhushu xiangjiao 儀禮注疏詳校, annotated by Chen Donghui 陳東輝 and Peng Xishuang 彭

Imperial Temple in the New Learning 119 Illustration 5.2: Depiction of Mingtang

Source: Chen Xiangdao, Lishu, SKQS, 130:89.1b.

120

Empowered by Ancestors xiaoji in the Book of Rites says, the ritual of attachment must be performed according to zhaomu, in which the tablets of grandsons are always attached to those of their grandfathers. This proves that the zhaomu in the ritual of tablet attachment should never be altered. If honorary titles or designations are bestowed in a sacrifice, the manager of serviceman60 will ask the ones who will be granted titles to proceed forward in the zhaomu order. Also, according to the Ji Tong chapter in the Book of Rites, whenever an honorary title or designation is bestowed, the people who will be granted titles should proceed alternately, in which a zhao is always aligned with a zhao, a mu is always aligned with a mu. This proves that the zhaomu in the ritual of granting titles should never be altered. Lastly, as the Da zhuan puts it, when all the family members gather to share the food, the seating plan should be arranged according to zhaomu. This proves that the zhaomu in the rite of food sharing should never be altered. Regarding ritual affairs for the living, such as granting titles and food sharing, the zhaomu sequence should not be altered; likewise, regarding ritual affairs for the dead, such as burial rites and the fu ritual, the zhaomu sequence should not be altered. Considering all these situations, it is plausible to know the zhaomu in the Imperial Temple.61 宗廟有迭毀，昭穆則一成而不可易。《春秋》《傳》言：「大王之昭，王季 之穆」。又言：「文之昭，武之穆」。此世序之昭穆不可易也。《周官冢 人》掌公墓之地，先王之葬居中，以昭穆為左右。此葬位之昭穆不可易也。 《儀禮》曰：「卒哭，明日以其班祔」。《禮記》曰：「祔必以其昭穆，亡 則中一以上」。此祔位之昭穆不可易也。司士：「凡祭祀賜爵，呼昭穆而進 之」。〈祭統〉凡賜爵，昭為一，穆為一，昭與昭齒，穆與穆齒，此賜爵之 昭穆不可易也。〈大傳〉曰：「合族以食，序以昭穆」。此合食之昭穆不可 易也。生而賜爵、合食，死而葬、附，皆以世序而不可易。則廟之昭穆可知 矣。

Chen’s main thesis here, that the zhaomu sequence should never be altered under any circumstances, contradicts Lu Dian’s zhaomu theory. Lu Dian insisted that the zhao and mu positions should be shifted in such a way that the superior zhao positions are always reserved for the father-ancestors. Although Chen Xiangdao shared with Lu Dian—as well as Wang Anshi and Wang Zhaoyu—the convention that zhao positions are superior to mu positions, he prioritized the ritual relationship between grandfathers and grandsons in temple rituals. Citing a phrase from the Chunqiu guliang zhuan 春秋穀梁傳 (Guliang Commentary on the Annals), Chen Xiangdao denounced the ritual practice of “inverse sacrifice” (nisi 逆祀) that appeared in the state sacrifices of Lu 魯 during the Warring States period. According to the Chunqiu 喜雙 (Taibei: Zhongyang yanjiuyuan Zhongguo wenzhe yanjiusuo, 2012), 285; John Steele, The I-Li, or Book of Etiquette and Ceremonial (Taibei: Ch’eng-wen Publishing Company, 1966), 2:93. John Steele interprets the term zuku 卒哭 as “stopping wailing.” Yet, according to Zheng Xuan’s commentary, the term actually refers to a specific sacrifice that is held after the three sacrifices of repose. For a thorough analysis of the zuku ritual, see Huang Yizhou 黃以周 (1828–1899), Lishu tonggu 禮書通故 (Beijing: Zhonghua shuju, 2007), 586–87. 60. Serviceman, sishi 司士, an office of the Zhou bureaucracy. 61. Chen, Lishu, 130:69.10a–b; also see Wei Shi’s quote in Liji jishuo, with slightly different wording. Wei, LJJS, 117:30.27b–28a.

Imperial Temple in the New Learning 121

guliang zhuan, in the spring of 625 BC, Duke Wen of Lu 魯文公 offered a sacrifice to his father Duke Xi of Lu 魯僖公 before he made sacrifices to his grandfather Duke Zhuang of Lu 魯莊公 in the Lu temple.62 Chen Xiangdao indicated that the Lu practice was a typical example of “inverse sacrifice” because it failed to follow the right order of sacrifice based on zhaomu. In Chen’s words, “a lack of zhaomu indicates an absence of ancestors. An absence of ancestors indicates disrespect for Heaven. Hence, a gentleman will not harm the dignity of his ancestors just because of his personal affection for his intimate relatives” 無昭穆，則是無祖也；無祖， 則無天也。君子不以親親害尊尊.63 By arguing that the dignity of ancestors (zunzun 尊尊) outweighs personal affection (qinqin 親親) in temple rituals, Chen Xiangdao claimed that when a newly deceased ancestor is designated as a zhao, his tablet should always be placed in a zhao temple, despite his father’s mu temple on the opposite side. Chen further compared the unalterable zhaomu principle to the ritual performance of “spiritual medium” (shi 尸). In the ritual performance, the son acts as a medium and sits on the superior south side in major ancestral sacrifices. His father, in contrast, stands on the inferior north side.64 However, the relatively inferior position of the father does not defy the virtue of filial piety, because the son as a medium incarnates his grandfather’s being in this ritual. The key point of Chen Xiangdao’s conception of the Imperial Temple is that zhaomu reorders the structure of familial hierarchy in the ritualized space of the temple. In the Lishu, Chen consolidated a ritual order that stresses the ritual status of grandfathers rather than fathers in temple sacrifices. After all, his zhaomu conception is grounded on the natural order of seniority that He Xunzhi and Zhang Zao had already emphasized in the 1079 debate.

Reconciliation and Codification: Ma Ximeng, Fang Que, and the Zhenghe wuli xinyi Along with the evolution of New Learning ritual scholarship, new understanding of the Imperial Temple emerged in the late eleventh century. Two New Learning scholars who specialized in the learning of the Book of Rites, Fang Que 方慤 and Ma Ximeng 馬晞孟, contributed significantly to this new understanding.65 As acclaimed by Song bibliographers, Fang Que’s twenty juan Lijijie 禮記解 supplements Wang Anshi’s Sanjing xinyi.66 Ma Ximeng earned his jinshi 進士 degree during the Xining era and was considered a diehard follower of Wang’s ritual learning.67 In contrast, Fang studied in the official school at Kaifeng and only gained access to officialdom 62. Chunqiu guliang zhuan zhushu, TSZSSSJ, 3:10.68. 63. Chen, Lishu, 130:69.9a. 64. Chen, Lishu, 130:69.11a. 65. Zhu Xi, Zhuzi yulei 朱子語類 (Beijing: Zhonghua shuju, 1986), 87.2227. 66. Zhizhai shulujieti, 2.24b–25a. 67. ZBSYXA, 98.19b.

122

Empowered by Ancestors

through a submission of his writings.68 Both Fang and Ma belonged to the intellectual community of the southerners (nanxue 南學), a community that Wang Anshi himself had helped to consolidate since the 1060s.69 Despite their New Learning backgrounds, Fang Que and Ma Ximeng distinguished their own explanations about the Imperial Temple from Wang Anshi’s. Whereas Wang Anshi and Wang Zhaoyu argued that the ritual status of temple ancestors should be managed according to transcendental concepts like yin and yang and moral conceptions like benevolence and integrity, Fang and Ma emphasized the factor of merits in conceptualizing the Imperial Temple. For example, Fang Que argued that the title Primal Ancestor refers exclusively to the dynasty’s founder, owing to his visible contributions and merits.70 A natural extension of Fang’s reasoning would lead to the conclusion that only Emperor Taizu is qualified as the Primal Ancestor of the Song dynasty, which blatantly contradicts Wang Anshi’s view in the 1072 Primal Ancestor debate. In a less straightforward manner, Ma Ximeng suggested that the position of the Primal Ancestor may be reserved for dynasty founders.71 Ma recognized the discrepancy within the text of the Book of Rites in describing the temple of the Primal Ancestor. As Ma argued, although the Wangzhi chapter highlights merits in determining the ritual status of the Primal Ancestor, the Jifa chapter underplays the factor of merits in ancestral worship. For Ma, it was acceptable for the Song court to render different definitions of “meritorious ancestors.” Although some ancestors are less “meritorious” than others, their ritual status can still be acknowledged based on the Jifa text.72 Not only did Ma’s skillful interpretation of the Wangzhi and Jifa texts reconcile the “meritocratic-hereditary” dilemma that had been triggered by the 1072 Primal Ancestor debate, but it also reminded his contemporaries of the textual tension within the Book of Rites. Although the Book of Rites contains detailed information about the Imperial Temple and temple rituals, it is essentially a collection of essays rather than a monograph underpinned by a consistent thesis. In revealing the textual tension within the Book of Rites, Ma Ximeng questioned the consistent, holistic, and hence simplistic depiction of the Imperial Temple’s nature in previous ritual theories. 68. Zhizhai shulujieti, 2.25a. 69. ZBSYXA, 98.19b. Geographically, Wang Anshi’s personal scholarship represented the scholarship of the Western Jiangnan Circuit (江南西路, modern western Jiangxi 江西). Prior to Wang Anshi, Ouyang Xiu (native place: Yongfeng, Jizhou 吉州永豐), and Li Gou 李覯 (1009–1059, native place: Nancheng, Jiangchang Military Prefecture 建昌軍南城) reflected the local intellectual tradition of the Western Jiangnan Circuit. The conflict between the intellectual community of the Southerners and the opposing Northerners (beixue 北學) is a less developed topic. Geographical discrimination and biases frequently appeared in court discussions concerning the prefectural quotas (jie e 解額) in civil service examinations. For some succinct descriptions of the Song quota system, see Edward Kracke, “Region, Family and Individual in the Chinese Examination System,” in Chinese Thought and Institutions, ed. John K. Fairbank (Chicago: University of Chicago Press, 1967), 251–68; Chaffee, The Thorny Gates of Learning in Sung China, 120–23; Thomas Lee, Government Education and Examinations in Sung China (Hong Kong: Chinese University Press, 1985). 70. Wei, LJJS, 117:30.24b–25a. 71. Wei, LJJS, 119:109.10a–b. 72. Wei, LJJS, 119:109.10b.

Imperial Temple in the New Learning 123

Additionally, Fang Que provided an influential explanation on the Imperial Temple based on a philosophical reading of zhaomu. He distinguished two kinds of zhaomu, the zhaomu of the deceased and the zhaomu of the living: The ritual of Imperial Temple is not only used to prioritize the zhaomu sequence of the deceased, it is also used to order the zhaomu of the living people. The temple system consisting of three zhao and three mu mentioned in the Wangzhi text refers to the zhaomu of the deceased. In contrast to it, the zhao and mu mentioned in the Jitong text refer to the zhaomu of the living people.73 宗廟之禮，非特序死者之昭穆，亦所以序生者之昭穆焉。〈王制〉所謂三昭 三穆，即死者之昭穆也。〈祭統〉所謂羣昭羣穆，即生者之昭穆也。

Elsewhere, Fang Que further articulated the difference between the zhaomu of the deceased ancestors and the zhaomu of the living. He suggested that the zhaomu sequence involved in temple sacrifices should be referred to as the zhaomu of the living, or the “secular zhaomu” (renzhi zhaomu 人之昭穆). By nature, the secular zhaomu is different from the zhaomu of the deceased ancestors, the “spiritual zhaomu” (shenzhi zhaomu 神之昭穆). The “secular zhaomu” is dominated by human emotions, especially familial affections. Since the strong tie between father and son reflects the highest degree of intimacy (qinshu 親疏) within a family, it regulates the secular zhaomu sequence. Fang used this explanation to explicate why some ritual scholars conceptualized zhaomu as an apparatus to differentiate the ritual status of fathers and sons in some rituals. In Fang’s opinion, these scholars were actually talking about the “secular zhaomu” when they were talking about the relationship between father and son.74 Nevertheless, in dealing with sacrifices in the Imperial Temple, Fang argued that the order of ancestors should be solely determined by the “spiritual zhaomu.” By indicating that the “spiritual zhaomu” was the dominating one between the two in temple rituals, Fang Que accentuated the role played by familial affection in related ritual discourses.75 His conclusion, that the zhaomu in temple rituals should be “spiritually oriented toward ancestors,” echoes Chen Xiangdao’s argument in the Lishu, yet contradicts Lu Dian’s understanding in the 1079 debate. The above discussions illustrate how various New Learning scholars contributed to a pluralistic understanding of the Imperial Temple and temple rituals. However, these scholars lacked opportunities to transform their understanding into more sophisticated ritual theories in the succeeding decades. The dominance of New Learning in both the intelligentsia and the examination realm receded in the Yuanyou era, when conservatives regained power under the regency of the Empress Dowager Xuanren 宣仁皇太后 (r. 1086–1093). The wave of criticism toward Wang Anshi’s Sanjing xinyi in 1086 stirred up a series of anti–New Learning campaigns 73. Wei, LJJS, 120:129.30b. 74. Wei, LJJS, 119:115.30a–b. 75. Wei, LJJS, 119:115.30b.

124

Empowered by Ancestors

during the late eleventh century. Although some of these campaigns were halted by the senior politicians from the conservative camp, they still impeded the development of the New Learning scholarship.76 As the royal pendulum swung back to the reformist side when Zhezong 哲宗 (r. 1085–1100) took over the court after the death of Empress Dowager Xuanren in 1093, New Learning soon recovered from the previous suppression. The ebbs and flows of New Learning were definitely correlated with the changes of political atmosphere. Nonetheless, political factors alone can neither explain the proliferation of intellectual discourse in times of crisis, nor the decline of learning in times of prosperity. The evolution of New Learning ritual scholarship illustrates how internal tension contributed to plural interpretations within the textual world of commentaries. However, it is impossible to demarcate the intellectual realm from the political one in a clear-cut manner. In 1100, the minister Cai Jing, together with Lu Dian, Huang Shang 黃裳 (1044–1130), and other ritual officials, submitted a new proposal about temple configuration, in which Song Shenzong’s ancestral chamber was elevated to the zhao position.77 By elevating Shenzong’s ritual status in the Imperial Temple, Cai Jing attempted to dignify the reformist emperor’s achievements in initiating the New Policies. After all, Cai’s proposal marked a relaunch of the New Policies, along with Emperor Huizong’s ascendancy to the throne. Huizong recognized the normative power of ritual innovations under his regime and regarded these innovations as a core part of a broader reformist program.78 In 1103/9, Huizong drafted a personal edict (shouzhao 手詔) emphasizing that “in the beginning of a king’s rule, rituals and music should be intentionally prioritized” 王者政治之端，咸以禮樂為急.79 To fulfill the statement in his 1103 edict, Huizong established a new ritual institution in 1107, the Bureau of Ritual Deliberation (yili ju 議禮局). The new bureau was responsible for implementing Huizong’s ritual innovations. On the one hand, ritual officials who worked in the Bureau of Ritual Deliberation collected ancient bronze ritual utensils to create a repository for the casting of archaizing alternatives.80 On the other hand, the bureau 76. A lecturer of the Imperial Academy (guozi siye 國子司業), Huang Yin 黃隱, was particularly active in these anti–New Learning campaigns in the 1080s. As Huang went too far in opposing Wang’s New Learning, some conservatives found his behavior unbearable and criticized him as hysterical and insane. XCB, 390.9496–9501; also ZBSYXA, 98.11a–b. For a detailed study of the Huang Yin incident, see Cheung Hiu Yu, “Cong Huangyin shijian zailun Yuanyou chuqi zhengju yu dangzheng” 從黃隱事件再論元祐初期政局與黨爭, Journal of Chinese Studies 中國文化研究所學報 66 (2018.1): 1–23. 77. SHY, Li 1:15.52. 78. Patricia Buckley Ebrey, Emperor Huizong (Cambridge, MA: Harvard University Press, 2014), 159–68, 243–52; Ari Levine, “The Reigns of Hui-tsung and Ch’in-tsung (1126–1127) and the Fall of the Northern Sung,” in The Cambridge History of China, Vol. 5, Part I: The Sung Dynasty and Its Precursors, 907–1279, ed. Denis Twitchett and Paul Jakov Smith (Cambridge, UK: Cambridge University Press, 2009): 606–9. 79. Huizong’s edicts concerning ritual reforms, as well as the relevant memorials of his ritual officials, were compiled by the Song court into a single juan that was placed before the main text of Huizong’s new ritual code; see Zheng Juzhong 鄭居中 (1059–1123) et al., Zhenghe wuli xinyi, SKQS, 647:shou 首.3a. 80. Xue Ang 薛昂, a senior ritual official in the Bureau of Ritual Deliberation, suggested the court send painters to places where ancient bronze ritual utensils had been excavated in order to document their appearances. See Zheng Juzhong, Zhenghe wuli xinyi, shou.14b. For more details on the ritual materials collected by the bureau,

Imperial Temple in the New Learning 125

made tremendous efforts to revive ancient rituals to match contemporary ritual practices. These efforts were embodied in a new ritual code compiled in 1113, the Zhenghe wuli xinyi.81 Many ritual regulations and codes had already been compiled before the Zhenghe wuli xinyi, but they paled by comparison with the new ritual code in terms of magnitude. The final version of the Zhenghe wuli xinyi, which was issued in 1113, reflects Huizong’s will to reorder the world based on a fivefold structure of rituals derived from the Rituals of Zhou: auspicious rites (ji 吉), inauspicious rites (xiong 凶), guest rites (bin 賓), military rites (jun 軍), and celebratory rites (jia 嘉).82 Given its structural resemblance to the Zhou ritual rubric, the Zhenghe wuli xinyi reveals the Song court’s effort to concretize the Zhou rituals. In his preface to the Zhenghe wuli xinyi, Huizong expressed his ambition to revive the Zhou cultural legacy: The Yin dynasty followed the ritual of the Xia: wherein it added to or subtracted from the ritual of the Xia should be known. The Zhou dynasty followed the ritual of the Yin: wherein it added to or subtracted from the ritual of the Yin should be known. Some other dynasties may follow the ritual of the Zhou after hundreds of generations, the ritual legacy of Zhou should be known. From the Zhou dynasty to the world of our times, more than a thousand years have passed. The Way has never been as obscure as nowadays . . . Concerning customs today, I imitate ancient Statecraft by adding to and subtracting from the Zhou ritual system according to the Way. First adopting the Way, then practicing and promoting it to the extent that people of hundreds of generations afterward will follow, and the ritual intent of hundreds of generations before will be reactivated. Thereupon, the ritual intent of our times matches that of the kings of hundreds of generations in a way like combining tallies together. When everything comes to a final unity, what the Analects has called “to inherit the Zhou legacy after hundreds of generations” becomes true.83 商因於夏禮，所損益可知也；周因於商禮，所損益可知也。其或繼周，百世 可知也。今天下去周千有餘歲，道之不明，未有疏於此時也……朕因今之 俗，倣古之政，以道損益而用之。推而行之，由乎百世之後，奮乎百世之 上。上等百世之王，若合符契。其歸一揆，所謂百世而繼周者也。 see Chen Fangmei 陳芳妹, Qingtongqi yu Songdai wenhua shi 青銅器與宋代文化史 (Taiwan: Guoli Taiwan daxue chuban zhongxin, 2016), 20–21; for Huizong’s collection of bronze ritual utensils and his promotion of ritual reform after 1107, see Patricia Ebrey, Accumulating Culture: The Collections of Emperor Huizong (Seattle: University of Washington Press, 2008), 150–204; see also Jeffrey Moser, “Recasting Antiquity: Ancient Bronzes and Ritual Hermeneutics in the Song Dynasty” (PhD diss., Harvard University, 2010), 200–214. 81. Ebrey, Emperor Huizong, 243–52. 82. According to the “Da zongbo” entry in the Rituals of Zhou, auspicious rites are used to serve deities and ancestral ghosts of the state in sacrificial affairs; funeral rites are used to express the sadness of the state; guest rites are used to strengthen the ties between the central court and other feudal states; military rites are used to intimidate the states; celebratory rites are used to look after the people. Zhouli zhushu, TSZSSSJ, 2:18.175–83. Notably, the original text in the Rituals of Zhou includes other ritual categories. As the fivefold ritual structure was codified by the Tang official code of Kanyuanli, a more accurate expression of the relationship between the Zhenghe wuli xinyi and the Rituals of Zhou should also involve a study of the intermediate role played by the Tang Kanyuanli in transiting the fivefold ritual structure. 83. Zhenghe wuli xinyi, 647: yuanxu 原序.3a–b.

126

Empowered by Ancestors

The frequently used term “baishi” 百世 comes from the Analects, referring to Confucius’ response to his disciple’s question on historical memory.84 The original intent of Confucius’ saying is that Zhou ritual legacies can be passed down to hundreds of generations. By citing the saying in his preface, Huizong asserted a transmission of the ritual legacies from Zhou to his own reign. The use of the term “baishi” also indicates that during the “hundreds of generations” between Zhou and Song, there was nothing worth mentioning in terms of rituals, let alone the great political achievements of Han and Tang empires. In Huizong’s rhetoric, only the one who can revive ancient rituals and the related rule of ritual (gu zhi zheng 古之政) deserves to be named as the successor of Zhou. In this light, the “ritual intent” that was transmitted directly from Zhou to Song and was documented in the Zhenghe wuli xinyi served as a representation of the transmission of the ancient Way in Huizong’s time.85 Similar to the “succession of the Way” (daotong 道統) concept that developed around the same period,86 Emperor Huizong in his preface to the Zhenghe wuli xinyi indicated a concept of “succession of ancient rituals” (litong 禮 統). To Huizong and his ritual officials, the concept of litong represented the core value of the Way. As Huizong articulated in his preface, his ritual enterprise and policies were accomplished only “by adding to and subtracting from the Zhou ritual rubric according to the Way” 以道損益而用之.87 In this light, the concept of litong reflected Huizong’s understanding of a transmission of ancient rituals from Zhou to his time. Alongside the preface, Emperor Huizong wrote a myriad of imperial edicts to direct the compilation of the Zhenghe wuli xinyi in the name of yubi 御筆 (imperially composed edicts).88 The emperor composed ten juan of “capping rituals” as an exemplar to demonstrate to the ritual officials in the Bureau of Ritual Deliberation how previous dynasties had endeavored to adopt ancient capping rituals.89 Certainly, Huizong was dissatisfied with previous dynasties’ rituals, especially Tang ritual practices. Hence, he proposed some revisions to the practice of capping rituals. 84. “People of hundreds of generations afterward would follow” 由乎百世之後 and “to match the kings of hundreds of generations” 上等百世之王 come from the Mencius. 85. For a thorough analysis of the evolution of the Song daotong concept, see Christian Soffel and Hoyt Cleveland Tillman, Cultural Authority and Political Culture in China: Exploring Issues with the Zhongyong and the Daotong during the Song, Jin and Yuan Dynasties (Stuttgart: Franz Steiner, 2012), 87–109, especially 90–94. Concerning the causes and effects of the Song construction of daotong, see James Liu, “How Did a NeoConfucian School Become the State Orthodoxy?” Philosophy East and West 23: 4 (1973): 483–505; Cho-Ying Li and Charles Hartman, “A Newly Discovered Inscription by Qin Gui: Its Implications for the History of Song ‘Daoxue,’” Harvard Journal of Asiatic Studies, 70.2 (2010): 387–448, especially 426–46. For some pre-Daoxue usage of daotong and Zhu Xi’s understanding and consolidation of the daotong idea, see Yu Yingshi, Zhu Xi de lishi shijie: Songdai Shidafu Zhengzhi Wenhua de Yanjiu 朱熹的歷史世界：宋代士大夫政治文化的研究 (Beijing: Shenghuo, dushu, xinzhi: Sanlian shudian, 2004), 7–35. 86. Li and Hartman, “A Newly Discovered Inscription by Qin Gui,” 426–33. 87. Zhenghe wuli xinyi, yuanxu.3b. 88. For studies on Huizong’s imperial edicts, see Ebrey, Emperor Huizong, 123–27; Li Rujun 李如鈞, “Yuduo zaishang: Song Huizong chao de wei yubi zefa” 予奪在上：宋徽宗朝的違御筆責罰, Taida lishi xuebao 臺大 歷史學報 60 (2017.12): 119–57. 89. Zhenghe wuli xinyi, 647: yuzhi guanli 御製冠禮 1.1a–10.7b.

Imperial Temple in the New Learning 127

Regarding the Imperial Temple and temple rituals, Huizong called for fundamental revisions. Having recognized Huizong’s enthusiasm to rectify temple rituals by approaching ancient Zhou rituals, ritual officials in the Bureau of Ritual Deliberation proposed revisions along this direction. For example, in 1108/5/10, Yu Li 俞㮚, an examining editor (jiantao 檢討) of the bureau, suggested a revision of existing sacrificial garments worn by scholar-officials in temple sacrifices. Furthermore, having drawn reference from the Zhou spirit of practicing filial piety, Yu Li emphasized the necessity to reinstall the tradition of family shrines in society. In his words: According to the Zhou practices, those with junior official status and above offered sacrifices at their family shrines, those with official clerk status and below offered sacrifices at their main chambers. Before building any dwellings, ancestral shrines had to be built first. Before making any daily utensils, sacrificial utensils had to be made first.90 周制，適士以上祭於廟，庶士以下祭於寢。凡營居室，必先建宗廟；凡造養 器，必先修祭器。

Yu Li’s suggestion was well recognized by Emperor Huizong. Half a year later, Huizong promulgated an imperial edict regarding family shrines, in which he announced that temple and shrine rituals of the present should resemble those of the antiquity.91 Nevertheless, the situation in Huizong’s time did not allow a full adoption of ancient temple and shrine rituals. The existence of subsidiary imperial ancestral buildings, especially the Daoist-influenced Temple of Spectacular Numina, was clearly in conflict with the Classical description of ancient temple rituals. Because of his fondness for Daoist rituals, Huizong did not agree to abolish the Temple of Spectacular Numina and other subsidiary imperial ancestral buildings in different Daoist and Buddhist monasteries. Eventually, Huizong’s ritual officials defined seasonal sacrifices in Song subsidiary imperial ancestral buildings and the morning offering in the Temple of Spectacular Numina as two “regular rites” (changyi 常儀) and thus compiled them into the Zhenghe wuli xinyi.92 The integration of these “regular rites” into the official ritual code under Huizong’s supervision after all reveals itself as a compromise of idea and reality. As an official ritual code, the Zhenghe wuli xinyi is characterized by its consonance with the New Learning doctrine in emphasizing the role played by rituals in ancient governance. In his personally composed edicts for and the preface to the Zhenghe wuli xinyi, Huizong repetitively used the term “ancient sage-kings” to 90. Zhenghe wuli xinyi, shou.7b. 91. Zhenghe wuli xinyi, shou.8a–b. 92. Zhenghe wuli xinyi, 106.1a–12a (seasonal sacrifices in Song subsidiary imperial ancestral buildings); 113.1a– 16a and 114.1a–4b (the morning offering in the Temple of Spectacular Numina). The extant juan 115 of the Zhenghe wuli xinyi includes a Daoist ritual practice that made offerings to the “Holy Ancestor” of the Song imperial line, Zhao Xuanlang. Venues of this ritual practice were local Daoist monasteries, rather than the Imperial Temple at Kaifeng. Notably, ritual officials who compiled the Zhenghe wuli xinyi did not name the ritual as a “temple ritual.” Zhenghe wuli xinyi, 115.1a–5b.

128

Empowered by Ancestors

describe the aims of his ritual agenda.93 The repeated uses of “ancient sage-kings” should be attributed to Huizong’s belief in the New Learning doctrine of legitimizing ritual reforms by appealing to the authority of ancient sage-kings. Although Huizong mentioned meeting contemporary needs in a few words concerning the compilation of the Zhenghe wuli xinyi, his ultimate goal stuck to “renewing the people in order to resemble the grandeur of the Three Dynasties” 作新斯人，以追 三代之隆.94 In his preface, Huizong quoted the phrase “to renew the people” from the edict of Wang Anshi’s posthumous promotion to the Grand Mentor (taifu 太傅), drafted by Su Shi. The phrase in its original context is in fact satirical.95 However, Huizong redefined the phrase by indicating that New Learning ritual scholarship did serve as an effective way to “renew the people.” Huizong also lamented that rituals had been performed improperly since the tenth century, when the cultural legacies of ancient sage-kings were abandoned during the governance of the Five Dynasties.96 He further criticized that contemporary scholars had indulged themselves in “conventional learning” (suxue 俗學) for a long time.97 More explicitly, Huizong quoted the exact term “conventional learning” from Wang Anshi’s preface to the Zhouli xinyi and criticized the same “conventional (ritual) learning” that had been charged by Wang two decades ago.98 Particularly, the influence of New Learning ritual scholarship in the Zhenghe wuli xinyi is palpable regarding the spatial arrangement of the Imperial Temple. The official layout of the Song Imperial Temple in this ritual code basically follows the agenda set by the New Learning scholars who accorded with Wang Anshi’s idea of temple rituals in the two ritual debates of 1072 and 1079. The Zhenghe wuli xinyi reconfirmed Xizu’s ritual status as the Primal Ancestor, as Wang Anshi had proposed in the 1072 debate. Moreover, the zhaomu of imperial ancestors in the xia sacrifice was an extension to He Xunzhi’s zhaomu plan in the 1079 debate, in which the tablets of Xuanzu, Zhenzong, Yingzong, and Emperor Zhezong were arranged along the zhao line, and the tablets of Yizu, Taizu, Taizong, Renzong, and Shenzong were arranged along the mu line in the temple.99 The confirmation of this arrangement in the Zhenghe wuli xinyi underscores the New Learning dominance in temple rituals. Since the 1072 debate and the compilation of Wang Anshi’s Zhouli xinyi, the

93. Huizong used the term ancient sage-kings at least twenty times in his preface and personal drafted edicts. See Zhenghe wuli xinyi, yuanxu.1a–3b; shou.5a, 6a, 7a, 10a–12a, 14a, 17a, 24a for some of his usage. 94. Zhenghe wuli xinyi, yuanxu.2a. 95. As Su Shi put it, Wang Anshi “renewed the people, as he took all the conventional teachings and legacies of previous intellectual traditions as worthless chaff ” 糠秕百家之陳迹，作新斯人. Su Shi, “Wang Anshi zeng taifu” 王安石贈太傅, Sushi quanji jiaozhu 蘇軾全集校注, ed. Zhang Zhilie 張志烈, Ma Defu 馬德富, and Zhou Yukai 周裕鍇 (Shijiazhuang: Hebei renmin chubanshe, 2010), 14:38.3774. 96. Zhenghe wuli xinyi, yuanxu.1b. 97. Zhenghe wuli xinyi, yuanxu.1b. 98. Wang, “Zhouli yixu,” 84.529. 99. Zhenghe wuli xinyi, 3.6a.

Imperial Temple in the New Learning 129

scholarly tradition of New Learning ritual scholarship had accumulated for four decades and finally manifested itself in Huizong’s ritual code.100

Concluding Remarks Since Wang Anshi’s time, scholars had tended to marginalize New Learning scholarship by neglecting its diversity and comprehensiveness. Taking the Imperial Temple as an example, this chapter explores how New Learning ritual scholarship as a disciplinary matrix had been more discursive and pluralistic than previously thought. I unfold the in-depth influence of the two approaches in conceptualizing the Imperial Temple and especially the zhaomu sequence within New Learning ritual scholarship. One approach defended merit-based approach in accounting for the Primal Ancestor position and emphasized the ritual relations between grandfather and grandson in positing zhaomu, represented by Chen Xiangdao, Ma Ximeng, and Fang Que; the other approach argued that zhaomu reflected the genealogical order of seniority, familial relations, and thus moral and transcendental values. Wang Anshi and Wang Zhaoyu were the staunch supporters of the latter approach. In fact, both approaches drew inspiration from previous Confucians’ ritual interpretations that could be traced to earlier periods. After all, none of these New Learning scholars revealed themselves as merely supporters of Wang Anshi’s personal scholarship. The fact that most New Learning scholars deemed Wang Anshi as the “scholarly master” (zongzhu 宗主) did not imply a closure of communication between themselves and other annotative traditions, nor a stagnancy of their pursuits to approach the Classics from their own perspectives.101 From Wang Anshi’s Zhouli xinyi to his disciples’ commentaries on the ritual Classics, and finally to the codification and the crystallization of New Learning ritual scholarship in the Zhenghe wuli xinyi, norms on temple rituals had been gradually established. In the late Northern Song, the direct transmission of ancient rituals from Zhou to Song became a shared convention within New Learning ritual scholarship. Eventually, the advocacy of a comprehensive revival of ancient rituals became the core doctrine of the New Learning “Daoxue” under Huizong’s reign. From the 1070s to the 1110s, the confrontation between the New Learning and other scholarly traditions not only revealed the tension between the former’s creative reading of the Classics and the latter’s opposition to that reading, but it also marked the development of a preliminary “learning of the Way” that centered on ritual scholarship. Unfortunately, the ritual writings of New Learning scholars have 100. It is also noteworthy to point out that the chief editor of the New Forms, Cheng Jiuzhong, was tied to Wang Anshi’s family by affinal bonds. Cheng’s younger brother Cheng Jiuzhong 鄭久中 was the son-in-law of Wang Pang. XCB, 485.11521. 101. The Qing scholar Zhang Xuecheng 章學誠 (1738–1801) proclaimed a famous statement in the Wenshi tongyi 文史通義, saying that “a scholar cannot study by himself without a real master; yet, [in the studying process] he must not be bounded by the biases of his master’s intellectual tradition” 學者不可無宗主，而必不可有門 戶. Zhang, Wenshi tongyi (Beijing: Zhonghua shuju, 1985), 523.

130

Empowered by Ancestors

been largely ignored in previous studies. The stigmatization of Wang Anshi and his academic followers after the Northern Song explains this indifference toward New Learning scholarship. A few Southern Song scholars, however, heeded to the impacts of New Learning ritual scholarship. Zhu Xi stands out among these scholars given that he has been conventionally perceived as the greatest opponent to the New Learning in Chinese intellectual history.

Section Three Imperial Temple, Daoxue Scholars, and the Socialization of Temple Rites

6 Daoxue Conception of the Song Imperial Temple: Zhu Xi and His Disciples

The Jurchen siege of Kaifeng in 1126 and 1127 caused tremendous chaos in this great imperial capital. After the Jurchen had captured and looted Kaifeng in the spring of 1127, many Song officials chose to leave the capital and looked for shelters in the southern capital of Yingtian 應天府 (modern Henan) and later Yangzhou 揚州, where the new emperor Song Gaozong 宋高宗 (r.  1127–1162) resided. A ritual official named Ji Ling 季陵, who served as the deputy prefect of the Court of Imperial Sacrifices during the wartime chaos, was one of those who chose to leave the capital. As a brave and responsible official, Ji Lin preserved and transported to Yangzhou the nine major ancestral tablets of Song emperors that were originally stored in the Imperial Temple in Kaifeng. A chronological account of the early Southern Song history records that Gaozong made sacrifices to these tablets in a Buddhist monastery in Yangzhou.1 When the Jurchen organized a powerful attack on Yangzhou and forced Gaozong to flee across the Yangtze River in early 1129, Ji Lin asked his personal guard Li Bao 李寶 to carry the imperial tablets of Song emperors to the relatively peaceful regions in the south of China. Unfortunately, in 1129/3, Li Bao lost the tablet of Taizu when he passed the warzone of Guazhou 瓜 洲, where the northern bank of the Yangtze River was located.2 The court immediately promulgated an edict to search for the lost tablet. However, the edict turned out to be a mere scrap of paper. The retrieval of Taizu’s tablet was not the top priority of the Song court during the wartime. The original tablet of Taizu was gone forever. In the end, Li Bao succeeded to transport the other eight imperial tablets across the Yangtze River. These tablets were temporarily stored in some Buddhist monasteries of several counties in Jiangxi and Zhejiang. From 1129 to 1137, Song imperial tablets traveled across Hongzhou 洪州, Qizhou 虔州, Wuzhou 婺州, and finally resided in Wenzhou 溫州. However, the chief administrators of Wenzhou hesitated to preserve these tablets, lest they should bear the extra responsibility for taking care of these invaluable tablets.3 In 1135, a primitive Imperial Temple was established 1. Li Xinchuan 李心傳 (1166–1243), Jianyan yilai xinian yaolu 建炎以來繫年要錄 (Beijing: Zhonghua shuju, 2013), 18.428–29. 2. Li Xinchuan, Jianyan yilai xinian yaolu, 20.456. 3. Zhongxing lishu 中興禮書, compiled by Xu Song (Shanghai: Shanghai guji chubanshe, 1995), 822:94.2b–3b.

134

Empowered by Ancestors

on the eastern side of the Ziyang Mount 紫陽山 in Linan 臨安, the new capital of the Southern Song dynasty.4 The new Imperial Temple comprised nine chambers and eleven rooms.5 In 1137, the remaining eight imperial tablets were conveyed to Linan and placed in the newly established Imperial Temple. In the 1130s and 1140s, major temple rituals were reimplemented, including seasonal sacrifices, morning and monthly offerings, di and xia sacrifices, along with a restoration of relevant ritual utensils. Throughout the Southern Song, the court continued to expand and renovate the Imperial Temple to store more tablets of the deceased emperors. Nevertheless, court debates over the Imperial Temple and temple rituals received scant attention after Gaozong’s reign. Compared to their Northern Song predecessors, Southern Song scholar-officials focused on the sacrificial texts that were used in temple rituals, instead of the arrangement of the Imperial Temple.6 The most famous Southern Song debate over the temple occurred in 1194, when the chief councilor Zhao Ruyu 趙汝愚 (1140–1196) removed Xizu’s spirit tablet from the Primal Ancestor position in the temple. The Daoxue master Zhu Xi severely criticized Zhao for his inappropriate removal of Xizu’s tablet. Zhu Xi claimed that Wang Anshi and Cheng Yi had already provided solid reasons to support the designation of Xizu as the Song Primal Ancestor several decades ago.7 However, Zhao Ruyu as a clansman of the Song imperial family had a reasonable motivation to replace the spirit tablet of Xizu with that of Taizu. When Song Xiaozong 宋孝宗 (r. 1162–1189) was enthroned, the crown shifted from Taizong’s line back to Taizu’s line. As one of Taizong’s descendants, Zhao Ruyu might want to dispel any doubt of his loyalty to the newly crowned emperor Song Ningzong 宋寧宗 (r. 1194–1224), a descendant of Taizu. Therefore, he elevated Taizu’s ritual status to the Primal Ancestor as an effective way to show his fidelity to Taizu’s lineage.8 The 1194 debate hence demonstrates how temple 4. Nan Song taimiao yizhi 南宋太廟遺址, compiled by the Hangzhou Municipal Institute of Cultural Relics and Archaeology 杭州市文物考古所 (Beijing: Wenwu chubanshe, 2007), 1–3. 5. Zhongxing lishu, 822:95.1b. 6. I consulted the online database of QSW for Southern Song ritual discussions about the Imperial Temple. For the two hundred and eighty-six Song essays with the keyword taimiao in their titles, only one of them concerns a ritual debate over the Imperial Temple. When the keyword Xizu 僖祖 is applied, there are two more pieces of relevant essays. Moreover, all three essays refer to the same ritual debate over the temple in 1194. See Chen Fuliang 陳傅良, “Xizu Taizu miaoyi” 僖祖太祖廟議, QSW, 268:6053.220; Zheng Qiao 鄭 僑, “Lun taimiao Xizu zhi wei zou” 論太廟僖祖之位奏, QSW, 273:6184.378; Zheng Qiao, “Lun Xizu dangli biemiao zou” 論僖祖當立別廟奏, QSW, 273:6184.379. Beyond the scope of keyword search based on QSW, there must be other sources concerning ritual debates over the Southern Song Imperial Temple. Nonetheless, the digital mining in the QSW database still provides us an overview of the Southern Song indifference to the temple and temple rituals. 7. Zhu, “Tiaomiao yizhuang” 祧廟議狀, Huian xiansheng zhuwengong ji, 58:15.225. 8. Some scholars have already noted the political implication under the 1194 debate. For example, see Zhang Huanjun 張煥君, “Song dai taimiao zhong de shizu zhi zheng: yi shaoxi wunian wei zhongxin” 宋代太廟中 的始祖之爭—以紹熙五年為中心, Zhongguo wenhua yanjiu 中國文化研究 (2006.2): 48–56. Yin Hui 殷慧 thoroughly studies Zhu Xi’s criticism of Zhao Ruyu’s demolition of Xizu’s ritual status in the 1194 debate and illustrates how specific liturgical details were used to maintain imperial solidarity. Yin Hui, Li li shuangzhuang: Zhu Xi lixue sixiang tanwei 禮理雙彰: 朱熹禮學思想探微 (Beijing: Zhonghua Shuju, 2019), 168, 374–79.

Daoxue Conception of the Song Imperial Temple

135

rituals were associated with real politics in times of political crises.9 It also demonstrates how Zhu Xi studied and utilized the ritual learning of New Learning scholars to serve his own interests. In his early work, Hoyt Cleveland Tillman had already noticed some similarities between Wang Anshi’s New Learning and the Daoxue doctrines in the thirteenth century.10 Tillman’s observation is compelling in terms of ritual learning, especially when interpretations concerning imperial ancestral rituals are involved. In the following, I will reveal how Zhu Xi’s ritual learning was inspired by the ritual campaign initiated by New Learning scholars, especially the Imperial Temple and its multifaceted representations of ancient rituals.

Zhu Xi’s Rediscovery of the 1079 Ritual Controversy Since the late eleventh century, New Learning scholars had developed a pragmatic framework of textual analysis, in which traditional commentaries and new interpretations were intertwined to respond to contemporary affairs. Likewise, Daoxue Confucians utilized traditional ritual commentaries and other texts to support their explanations on contemporary ritual affairs. Specialists in Chinese thoughts conventionally understand the Daoxue scholarship as speculative in nature and consider Daoxue as the opposite of the scholastic learning of Classics in the name of hanxue 漢學. Nevertheless, Daoxue has its own scholastic tradition, which is particularly reflected in its interpretation and annotation of the ritual Classics. Intellectual historians recognized Zhu Xi as the greatest synthesizer of the Daoxue ritual scholarship of Southern Song. The Comprehensive Commentary and Explanation of the Rites and Ceremonies (Yili jingzhuan tongjie 儀禮經傳通解, hereafter the Comprehensive Commentary) crystallizes Zhu Xi’s lifetime endeavors to rejuvenate ancient rituals. After Zhu was deported from the court in the Qingyuan prohibition of False Scholarship (Qingyuan dangjin 慶元黨禁) and returned to his hometown in 1197, he spent his late years compiling the Comprehensive Commentary.11 Unfortunately, Zhu failed to complete the Comprehensive Commentary and witness its publication. Right before his death, Zhu was still discussing the manuscript of it with Huang Gan 黃榦 (1152–1221), who was his

9. For a general discussion of Song political crises in succession, see Chaffee, Branches of Heaven, 25–30, 132–35, 179–81. For Zhao Ruyu’s role in court politics, see Chaffee, Branches of Heaven, 189–95. 10. Hoyt Tillman argues that both Daoxue and Wang Anshi shared some basic assumptions, including the idea of restoring antiquity and pursuing the Way. Hoyt Tillman, Utilitarian Confucianism: Ch’en Liang’s Challenge to Chu Hsi (Cambridge, MA: Harvard East Asian Monographs, 1982), 42–44. 11. Although the idea to compile a comprehensive ritual writing had crossed Zhu Xi’s mind several times in his early years, it was not until 1196 that Zhu decided to assemble his disciples to realize the compiling project. Shu Jingnan 束景南, Zhu Xi nianpu changbian 朱熹年譜長編 (Shanghai: Huadong shifan daxue chubanshe, 2014), 1249. The original name of the Comprehensive Commentary is Yili jizhuan jizhu 儀禮集傳集注. See Shu Jingnan, Zhu Xi nianpu changbian, 1288–89.

136

Empowered by Ancestors

disciple and son-in-law.12 After Zhu Xi’s death, Huang Gan supplemented the two sections of funeral and sacrificial rites based on Zhu’s explanatory notes about the Comprehensive Commentary. In 1223, a local official named Zhang Fu 張虙 published Huang Gan’s supplementary edition to the Comprehensive Commentary at Nankang 南康 (modern Jiangxi). However, Huang Gan’s work is preliminary. He collected a variety of Song commentaries on sacrificial rites and put them all in the Comprehensive Commentary, without a clear conceptual framework to schematize these commentaries. Eventually, a specialist in ancient rituals, Yang Fu 楊復 (fl. 1228), who was also Zhu Xi’s disciple, revised the section on sacrificial rites in the Comprehensive Commentary. The final product is the Sacrificial Rites: An Extension on the Comprehensive Commentary and Explanation of the Rites and Ceremonies (Yili jingzhuan tongjie: xujuan jili 儀禮經傳通解續卷祭禮, hereafter Sacrificial Rites). The Academia Sinica of Taiwan recompiled and published this extended edition.13 Yang Fu’s edition preserves more of Zhu Xi’s accounts of sacrificial rites and Yang’s interpretations of these rites. In conjunction with Huang Gan’s edition of the Comprehensive Commentary, which has been collected in the Qing compiled Siku quanshu, Yang Fu’s edition provides us a comparable set of data to examine Zhu Xi’s and his disciples’ discourse of the Imperial Temple.14 Before probing into Zhu Xi’s discussion about the temple, it may be useful to explicate his fundamental assumptions about imperial rituals. Zhu is essentially a revivalist in terms of ritual studies. However, as Julia Ching argues, Zhu’s revivalist tendency does not necessarily mean that he wanted to replicate ancient rituals without any modifications.15 Indeed, Zhu had explained the difficulty in adopting ancient rituals under contemporary circumstances.16 His magnum opus of family rituals, the Zhuzi jiali 朱子家禮, exemplifies his effort to simplify and standardize the complicated funeral and sacrificial rites of antiquity for both literati families and commoners.17 Nevertheless, regarding the domain of imperial rituals, Zhu believed that the ritual intent of the Three Dynasties, or, in his own words, “the fundamental bases of ancient rituals” (daben dayuan 大本大源), had to be maintained.18 Particularly, Zhu Xi identified two “fundamental bases of ancient rituals” worthy of clarifications by his contemporary scholars. In a conversation with his students, Zhu criticized two conventional practices of imperial rituals in his time. First, the court maintained the practice of offering collective sacrifices to the Heaven and the 12. Zhu Xi, “Yu Huangzhiqing shu” 與黃直卿書, in Huian xiansheng zhuwengong ji 晦菴先生朱文公集, SBCKCBSB, 58:29.462. Zhu Xi was dead in 1200/3/9. Zhu Xi wrote this letter to Huang Gan one day before his death in 1200/3/8. 13. Yang Fu, Yili jingzhuan tongjie xujuan: jili 儀禮經傳通解續卷祭禮 (YLJZTJ: JL), compiled by Ye Chunfang 葉 純芳 and Hashimoto Hidemi 橋本秀美 (Taibei: Zhongyang yanjiuyuan zhongguo wenzhe yanjiusuo, 2011). 14. For a general briefing of these two editions and their evolution in terms of textual history, see Ye Chunfang’s introduction to Yang Fu’s revision of the Comprehensive Commentary. YLJZTJ: JL, Introduction, 6–20. 15. Julia Ching, The Religious Thought of Chu Hsi (New York: Oxford University Press, 2000), 79–83. 16. Zhu Xi, Zhuzi yulei, 84.2178. 17. For a reliable English translation of the Zhuzi jiali, see Ebrey, Chu Hsi’s Family Rituals. 18. ZZYL, 84.2179.

Daoxue Conception of the Song Imperial Temple

137

Earth at the southern suburban Altar. Second, the Song Primal Ancestor did not reside in a building separate from the Imperial Temple. His tablet was placed in the main chamber of the temple, alongside the tablets of other ancestors. Zhu Xi called these two ritual practices as “extremely important affairs under Heaven that were still fettered by conventional ideas” 今天下有二件極大底事, 恁地循襲.19 Definitely, the rectification of the Primal Ancestor’s position in the temple was crucial to Zhu Xi’s understanding of ancient rituals. In echoing Wang Anshi, Zhu considered Xizu as the only legitimate Primal Ancestor of the Song Imperial Temple.20 When Sun Congzi 孫從之, one of Zhu Xi’s disciples, questioned Xizu’s contributions to the founding of the Song dynasty in a private talk with Zhu, Zhu made a powerful argument based on his new conception of ancestors’ merits. As Zhu argued, if one admitted that the career success of an educated elite should be attributed to his ancestors, by the same reasoning, one should also recognize how the “secret merits” (yingong 陰功) of imperial ancestors contributed to their descendants as emperors. Zhu Xi further asked, if scholar-officials’ success had nothing to do with their ancestors, why would the court bestow posthumous titles to scholar-officials’ ancestors after these officials had obtained official ranks?21 After all, to deny Xizu’s merits in giving birth to the Song dynasty was just as absurd as to say that scholar-officials did not need to acknowledge the blessings of their ancestors. Although Zhu criticized Wang Anshi for his “disturbance of the old practices” (bianluan jiuzhi 變亂舊制) in other aspects of ritual learning,22 he praised Wang’s interpretation of the Primal Ancestor designation. Prior to Zhu Xi, Cheng Yi had already applauded Wang’s endeavors in the 1072 debate to rectify Xizu’s status as the Primal Ancestor.23 Both Zhu and Cheng considered the Northern Song conservatives’ opposition to Wang Anshi’s ritual reforms as a representation of their partisan interests: adamant anti-reformers would reject Wang’s arguments on every account, no matter how well-reasoned these arguments were.24 As Zhu put it, since the proTaizu officials in the 1072 debate were overwhelmed with prejudice, they “failed to notice that self-reflection would be enough to tranquilize their minds, and thus they could have reached the same conclusion as Wang Anshi” 不知反之於已，以即夫 心之所安.25 Considering the Primal Ancestor controversy, Zhu argued that Wang Anshi overpowered all his opponents both rhetorically and intellectually. None 19. ZZYL, 90.2289. 20. ZZYL, 90.2305–6. 21. ZZYL, 107.2662. 22. Zhu, “Qi xiu sanli zhazi” 乞修三禮劄子, Huian ji, 14.212. 23. Zhu, “Mianzou yaomiao zhazi” 面奏祧廟劄子, Huian ji, 15.228; ZZYL, 107.2664. 24. ZZYL, 107.2662. Zhu Xi also noted that in the official collection of eminent memorials (Zhuchen zouyi 諸 臣奏議, later known as the Collections of Eminent Memorials of Song Dynasty 國朝諸臣奏議), Wang Anshi’s memorials in the 1072 debate were deliberately underplayed by the editors: they cited Wang’s memorials in small font. As a result, Wang’s argument looks more like a commentary to the whole issue. ZZYL, 107.2661, 2664. 25. Zhu, “Mianzou yaomiao zhazi,” Huian ji, 15.228.

138

Empowered by Ancestors

of the writings of the pro-Taizu camp could match Wang’s memorial on temple rituals.26 By articulating Cheng Yi’s judgment, that “Jiepu’s (Wang Anshi’s courtesy name) point of view after all surpassed conventional Confucians’ mediocrity” 介甫 所見, 終是高於世俗之儒,27 Zhu identified the pro-Taizu camp in the 1072 debate with the “drifting and vulgar shi” 流俗之士 whom Wang Anshi had described as “incapable” in his letter to Wang Hui.28 Not only did Zhu Xi highlight Wang Anshi’s contributions in the rectification of Xizu’s Primal Ancestor designation, but he also evaluated New Learning scholars’ commentaries on the Imperial Temple as inspiring. Particularly, Zhu Xi acclaimed Fang Que and Ma Ximeng for their interpretations of the Imperial Temple’s sequence.29 More importantly, Zhu Xi was the first scholar who acknowledged the significance of the 1079 zhaomu debate between the two New Learning scholars, He Xunzhi and Lu Dian. In an essay titled “Dixia yi” 禘祫議, Zhu claimed that Emperor Shenzong asked his officials to discuss ritual texts in order to “trace the grandeur of the Three Dynasties and correct the absurdity that had spread over for thousands of years” 遠迹三代之隆，一正千古之謬.30 Zhu’s wordings here perfectly echo Lu Dian’s 1079 memorial, in which Lu claimed that by introducing an appropriate order of the Imperial Temple the Song regulatory standards “can be compared with the excellent kingship of the Three Dynasties” 以齊三代盛王.31 After all, the 1079 zhaomu debate and Lu Dian’s related discourses inspired Zhu Xi to establish his own conceptions on the Imperial Temple and temple rituals. However, why did Zhu Xi—an opponent of Wang Anshi’s New Policies—turn to New Learning scholars in conceptualizing ancestral rituals? The factor of influence might be the first reason. Northern Song scholars’ endeavors to escape the New Learning influence reflected what Harold Bloom has called the clinamen revision of intellectual precursors.32 Cheng Yi and his disciples tended to interpret New Learning with an aim of breaking away from their precursor Wang Anshi. In practice, Cheng and his disciples considered New Learning as their main opponent and attempted to depart from the New Learning influence by establishing their own doctrines. Most of Cheng Yi’s disciples tended to downplay the conventions that had been shared with New Learning scholars, such as the similar pursuit of a learning of the Way. Cheng Yi’s disciple Yang Shi, who was also a specialist in New Learning scholarship, severely criticized Wang Anshi’s analysis of characters and his personal records of Shenzong’s regime.33 Yang’s hostility 26. In a conversation with his disciple, Zhu severely criticized Han Wei’s memorial in the 1072 debate as fragmentary and paradoxical that “does not deserve to be called a writing” 都不成文字. ZZYL, 107.2664. 27. Zhu, “Xiaotiezi” 小貼子, Huian ji, 15.225; ZZYL, 107.2664. 28. Wang, “Da Wang Shenfu shu” 答王深甫書, LCJ, 51:72.464. 29. ZZYL, 87.2227. 30. Zhu, “Dixia yi” 禘祫議, Huian ji, 69.1264; Yang, YLJZTJ: JL, 448. 31. Lu, “Zhaomu yi,” Taoshan ji, 6.13b. 32. Harold Bloom, The Anxiety of Influence: A Theory of Poetry (New York: Oxford University Press, 1973), 14. 33. Yang Shi, Shenzong rilubian 神宗日錄辨, Wangshi zishuobian 王氏字說辨, Yangshi ji, 104–60. Concerning Yang Shi’s criticism of Wang learning, see Xia Changpu 夏長樸, “Cong Lixinchuan Daominglu lun

Daoxue Conception of the Song Imperial Temple

139

toward Wang Anshi and his learning is understandable, considering that in the formative stage of a new tradition its proponents are usually more inclined to claim their legitimacy based on differences from rather than similarities with existing traditions. Given the prevalence of New Learning scholarship in the late Northern Song, Cheng Yi and Yang Shi chose to criticize it as a heretical doctrine in order to facilitate the elevation of their own scholarship.34 When the Daoxue development entered its transitional phase in Zhu Xi’s time, Zhu was less anxious about Wang Anshi’s scholarship than his Northern Song predecessors were. Therefore, he could value Wang Anshi and the New Learning scholarship in a more objective way. The second reason for Zhu Xi’s reconsideration of New Learning ritual scholarship had something to do with the elevation of the general discourse of revivalism among some Southern Song ritualists. As the greatest ritualist of his time, Zhu Xi claimed that the correct practice of imperial ancestral rituals would enlighten relevant social practices by fostering a commitment to Confucian ethical values, especially filial piety. Although in different ways, Zhu Xi’s claim reiterated the New Learning emphasis on reviving ancient rituals in the ritual debates several decades ago.

New Understanding of the Spatial Arrangement of the Imperial Temple: Zhu Xi and Yang Fu Zhu Xi praised Lu Dian’s endeavors in preserving ancient elements of the Imperial Temple. He complimented Lu Dian’s scheme of the Song temple, especially its allocation of the temple’s chambers, doors, walls, and other spatial arrangements.35 Based on the Jin (266–420) Confucian Sun Yu’s 孫毓’s record, Zhu Xi made a hypothetical case of the temple setting of feudal lords to explicate his understanding of the Imperial Temple configuration. According to Zhu’s model, in an ideal arrangement of temple configuration, all the five temples are located on the southeast of the palace. Each temple is enclosed by a “palace wall” named dugong 都宮.36 The temple of the Primal Ancestor is located in the north; the two zhao and the two mu temples extend southward in succession. The spirit tablet of the first ancestor who received a fief from the Son-of-Heaven is stored in the Primal Ancestor temple; the secondgeneration ancestor’s tablet is stored in the northern zhao temple; the third-generation ancestor’s tablet is stored in the northern mu temple; the fourth-generation Songdaidaoxue de chengli yu fazhan” 從李心傳道命錄論宋代道學的成立與發展, in Songshi yanjiuji 宋史研 究集 (Taibei: Zhonghua congshu bianshen weiyuanhui, 2006), vol. 36, 20–24. 34. Cheng Yi considered the New Learning as more harmful than the Buddhist teaching to true Confucianism. Cheng emphasized the significance of having a thorough rectification of Wang Anshi’s private learning, lest it corrupt young scholars. Cheng Hao, and Cheng Yi, Ercheng yishu 二程遺書, in Ercheng ji 二程集 (Beijing: zhonghua shuju, 1981), 38. 35. Zhu, “Dixia yi,” 69.1264; Yang, YLJZTJ: JL.448. 36. The Chinese characters of the name of the palace wall are the same as those being used to describe the separate configuration of temples.

140

Empowered by Ancestors

ancestor’s tablet is stored in the southern zhao temple; the fifth-generation ancestor’s tablet is stored in the southern mu temple. All five temples face south, each with entrances (men 門), halls (tang 堂), rooms (shi 室), and main chambers (qin 寢).37 In his annotation on Zhu Xi’s ritual writings, Yang Fu, as Zhu’s disciple, visualized Zhu’s temple scheme and named it as “the illustration of the five temples of feudal lords” (Zhuhou wumiao tu 諸侯五廟圖; see Illustration 6.1 on p. 141).38 It is difficult to retrieve the historical sources from which Zhu Xi developed his understanding of the Imperial Temple’s arrangement and its main structure. Zhu merely traced his hypothetical model of the temple back to the Jin Confucian Sun Yu and a sixth-century official ritual code.39 Except for these two texts, whose original copies had already disappeared in Zhu Xi’s time, Zhu did not provide any earlier references. However, for the arrangement of chambers, rooms, and halls within each temple, Zhu portrayed two diagrams. The first appeared in Zhu Xi’s memorial on di and xia sacrifices (“Dixia yi”), which was later collected in Yang Fu’s Sacrificial Rites. For the sake of clarity, I reportray Zhu Xi’s diagram in Figure 6.1:40 Figure 6.1: Zhu Xi’s Spatial Setting of the Imperial Temple

Despite its highly abstract form, the diagram conveys three general ideas. First, the entrance of an Imperial Temple is a wall gate called yuanmen 垣門. Second, there used to be some open areas named ting 廷 within the temple walls, which remind us of the courtyard space of Buddhist monasteries. Possibly, the outdoor space of an Imperial Temple in practice was smaller than that of Buddhist or Taoist monasteries in Zhu Xi’s time. However, the presence of a courtyard in the temple 37. I paraphrase Zhu Xi’s hypothetical model based on the record of Wei Shi’s Liji jishuo. See Wei, LJJS, 30.45a–b. It is worth mentioning that the extant Liji jishuo contains some valuable excerpts of both Zhu Xi’s and Yang Yu’s ritual texts, especially the latter’s Sacrificial Rites. 38. Yang Fu, Yili pangtong tu 儀禮旁通圖, SKQS, 104:6a. 39. The ritual code is the Sui-compiled Jiangdou jili. Wei, LJJS, 30.52a. 40. Zhu, “Dixia yi,” 69.1265; Yang, YLJZTJ: JL, 450.

Daoxue Conception of the Song Imperial Temple Illustration 6.1: Zhu Xi’s Temple Scheme for Feudal Lords

Source: Yang Fu 楊復, Yili pangtong tu 儀禮旁通圖 (SKQS ed.), 104:17.6a.

141

142

Empowered by Ancestors

still served some purposes from the performative perspective. In the Sacrificial Rites, Yang Fu added a piece of text about the procedures for transporting the spirit tablets of feudal lords’ ancestors from one temple to a “novel temple” (xinmiao 新 廟). The text itself was clearly retrieved from a related chapter in the Dadai Liji 大 戴禮記 (Records of Ritual by the Dai Senior).41 According to the record, when the feudal lords and his ministers arrive at the novel temple, they should follow the order of ritual practitioners and find their own places in the ritualized space of the temple. Supposedly, ministers without special assignments should also be presented during the performance. Since the indoor space of the temple is more sacred in nature, unauthorized ministers and officials will be prohibited from entering the divine inner space of the temple—that is, in most cases, the main hall or the main chamber. Therefore, a vast courtyard provides unauthorized officials an intermediate space to wait in solemn reverence, and effectively prevents any disturbance caused by a crowd within the temple. The third idea conveyed by this diagram concerns the bipartite structure of the archaic setting of the Imperial Temple. By recognizing the front temple structure and the rear main chamber as two key components of the temple, Zhu embraced Zheng Xuan’s interpretation of temple configuration in the latter’s commentary of the Yueling 月令 (Regulations of Different Months) chapter of the Book of Rites.42 Nonetheless, in an attached diagram to his memorial, “Lunbenchao miaozhi” 論本朝廟制 (On the configuration of the Song Imperial Temple), Zhu Xi illustrated the structure of the Imperial Temple with more details. I reportray the diagram in Figure 6.2 (see p. 143). We cannot blame Zhu Xi for drafting such a brief layout of the Imperial Temple structure, considering that he might never be able to visit a real one and study its inner structure. Compared with Figure 6.1, Figure 6.2 identifies more temple compartments, including two rooms, two subsidiary chambers, the front hall, and a new main chamber. The new main chamber, which Zhu Xi termed as a shi 室, replaces the rear main chamber (qin) in terms of functionality: it is where the spirit tablet resided. The rear main chamber of qin, which originally houses the tablet, becomes a “resting place” in this diagram. The shift in the connotation of qin from “main

41. Yang, YLJZTJ: JL, 410–14; “Zhuhou qianmiao” 諸侯遷廟, Wang, Da Dai liji jiegu, 198–202. The Qing scholar Wang Pinzhen argued that the ritual practice of transferring the tablet to the “novel temple” must be performed after the lian 練 period, but not within it. As the Guliang Commentary indicated, the temple with eaves changed and repainted (yiyan gaitu 易檐改塗) would first be an “abandoned temple” (huaimiao 壞 廟). Chunqiu guliang zhuan zhushu, TSZSSSJ, 3:10.67. The word “abolishment” here refers to the symbolic uninstallation of a temple, instead of physical destruction. The temple is transformed ritually into a “novel temple” after the lian sacrifice is performed. Accordingly, the transfer of the tablet to the “novel temple” is only possible in a renewed temple after the lian sacrifice. Hence, Wang claimed that the whole process of tablet transfer should be performed after the period of three years of mourning. Wang, Da Dai liji jiegu, 198. 42. Zheng’s commentary said, “for every Imperial Temple, the front area is called the temple, the back area is called the chamber” 凡廟，前曰廟，後曰寢. Zhu Bin, Liji xunzuan, 231; Legge misunderstood the meaning of qin here and translated it into “sleeping apartments.” Legge, The Sacred Books of China, 3:260.

Daoxue Conception of the Song Imperial Temple

143

Figure 6.2: Zhu Xi’s Depiction of the Imperial Temple in “Lunbenchao miaozhi”

Zhu Xi’s own commentary on the spatial arrangement of an Imperial Temple: “Each generation has its own temple. A temple should include an entrance gate, a hall, a main chamber, two rooms, two subsidiary chambers, a resting place, and walls surrounding the four sides” 一世各為一廟。廟有門、有堂、有室、有房、有夾室、有寢，四面有墻。* * Zhu, “Lunbenchao miaozhi,” Huian ji, 15.223.

chamber” to “resting place” reflects how Zhu Xi has revised his understanding of temple structure over time. Zhu Xi conceptualized the Imperial Temple based on his understanding of an ideal temple structure that he believed to have existed in ancient times. Such a temple does not exist in reality. A 1976 archaeological excavation of a Western Zhou ancestral architecture revealed the interior structure of a Zhou “temple-like” structure.43 In the core area of the excavated architecture where the “main chamber” (qin or shi) is supposedly located, there is only a corridor, or, to borrow Rudolf Arnheim’s word, an “extrinsic space.”44 This corridor, according to Wu Hung, “created discontinuity in space,” as it separates the central hall (tang) from other compartments.45 Other architectural components of the idealized temple configuration, such as the 43. See Wu Hung, Monumentality in Early Chinese Art and Architecture (Stanford: Stanford University Press, 1995), 87, Figure 2.7a (floor plan of the excavated temple) and Figure 2.7b (reconstruction of the temple). For a more comprehensive portrait of the temple, see Chen Quanfang 陳全方, Zhouyuan yu Zhou wenhua 周原與 周文化 (Shanghai: Shanghai renmin chubanshe, 1988), 37–69. 44. Rudolf Arnheim, New Essays on the Psychology of Art (Berkeley: University of California Press, 1986), 83. 45. Wu, Monumentality, 84.

144

Empowered by Ancestors

courtyard, the rooms, and the subsidiary chambers, are not identifiable in the excavated Zhou architecture of 1976. However, in Zhu Xi’s time, these components were necessary components of an ideal Imperial Temple, as Zhu portrayed in his diagrams. Concerning ritual practices in the temple, it was Yang Fu rather than Zhu Xi who contributed more to the standardizing performance of temple rituals considering Yang’s Sacrificial Rites. In his version of the Comprehensive Commentary, Yang Fu quoted the Dadai liji at length to explain his interpretation of temple rituals. According to Yang Fu, a feudal lord needs to make a retreat three days before he transfers his ancestor’s tablet to a “novel temple.”46 For the retreat practice, some ritual masters (zhu 祝 and zongren 宗人) and officials of different ranks will accompany the lord.47 On the day of transferring the tablet to the “novel temple,” the lord and all the attendants wear black garments. When they arrive at the original temple, officials stand in tight rows in front of the temple, resembling the array of a court audience. The zongren and the zhu will direct the whole process and make the formal speeches. The lord stands beneath the stairs of the hall, facing west, as his ancestor’s tablet is located on the east side of the main chamber.48 While the zongren says, “Please be ascended,” the lord ascends to the hall, accompanied by the zhu on his left side, with several ritual coins at the zhu’s hands as offerings. Then the lord bows to the north. Meanwhile, the zhu says on behalf of the lord: “The feudal lord X, the filial son, with auspicious ritual coins, humbly submits this petition to my ancestor, the imperial father Y. It is announced that the great spirit of the father Y will be moved from the existing temple to a new one. Announcement done” 孝嗣候某， 敢以嘉幣告于皇考某候，成廟將徒，敢告.49 Then, the lord and the zhu bow again to the main chamber of the temple (where the ancestor’s tablet is originally placed), descend from the hall, and stand at the foot of the hall stairs. Meanwhile, the ritual master who holds the clothes left behind by the deceased ancestor follows the zhu and descends from the hall to the stele (bei 碑) in the courtyard area.50 According to Li Rugui 李如圭 (jinshi. 1193), a Jiangxi scholar who specialized in concrete rites and ceremonies and helped Zhu Xi to compile the Comprehensive Commentary, the stele should be set up at some distance from the entrance gate of the temple and is erected for calendrical purposes.51 After a short and solemn prayer at the stele’s 46. 君前徙三日齋. Yang, YLJZTJ: JL, 410. For the original text, see Wang, Da Dai liji jiegu, 199. The Zhonghua shuju edition of Da Dai liji jiegu and the Ye-Hashimoto edition of Yang Fu’s Sacrificial Rites punctuated this phrase differently. The former reads 君前徙三日 as one phrase, which awkwardly means someone (the subject is missing) would offer a sacrifice in front of the lord. The latter reads the six characters 君前徙三日齋 as a whole, which means that the lord who performs the ritual should hold a “three-day sacrifice” before the removal of his ancestor’s tablet from the old temple. 47. Yang, YLJZTJ: JL, 410. Wang Pinzhen annotated both zhu and zongren as “masters who communicate directly with ancestral spirits” 接神之官. In other words, they do the reception work when the spirits descend to the temple area. Wang Pinzhen, Da Dai liji jiegu, 199. 48. Yang, YLJZTJ: JL, 411; Wang, Da Dai liji jiegu, 199. 49. Yang, YLJZTJ: JL, 411; Wang, Da Dai liji jiegu, 200. 50. Yang, YLJZTJ: JL, 412; Wang, Da Dai liji jiegu, 200. 51. Following Zheng Xuan and Zhu Xi, Li Rugui argued that the stele is used to “recognize the shadow of the

Daoxue Conception of the Song Imperial Temple

145

place, the lord, the zhu, the zongren, the “clothes-holding” person, and all other attendants of the ritual board the carriages and proceed to the “novel temple.” According to Yang Fu’s understanding, the zhu and the zongren serve as key mediums in the process of transferring spirit tablets. The zhu, in particular, as Yang Fu annotated, acts as “a guide of the ancestral spirit to be transferred” 祝所以導神 也.52 Therefore, the zhu stands as the beacon to lead the spirit that is emblematized in the relics handled by the person who also holds the clothes of the deceased ancestor (fengyifu zhe 奉衣服者). The zhu leads the procession walking from the old temple to the “novel temple.” After the lord and his attendants arrive at the “novel temple,” they perform rituals in another way. As this part is central to Yang Fu’s understanding of temple rituals, I quote it at length here: When they arrive at the novel temple, they first set up the sacrificial mat somewhere between the east of the window and the west of the door,53 and place the ritual vessel beneath the west interior wall (of the hall).54 Then they put the minced and marinated sacrificial meat in the west room,55 and align the washing utensils with the east cornice (of the temple’s outer wall):56 the distance from the washing utensils to the hall is calibrated based on the scale of the hall.57 The officials with duties enter the temple first, and stand in tight rows in front of it, resembling the array of a court audience. Then the zhu leads the “clothes-holding” person to enter the door; the lord follows them.58 While the person who holds the clothes is entering the door, all the attendants are stepping aside in reverence to let him pass. Then the “clothes-holding” person ascends to the hall, when all the others return to their own positions. Following the person who holds the clothes, the lord ascends to the hall and places the clothes on the mat, and puts the ritual coins on the east side of the long table.59 The lord stands, facing north, with the zhu standing on his left. The Sun and to conceive yin and yang” 識日景，知陰陽也. Li Rugui, Yili shigong 儀禮釋宮, SKQS, 103:15a. As Li Rugui’s only extant work, the Yili shigong reflects Zhu Xi’s ritual ideas, especially the spatial arrangement of ritual buildings. Scholars like Shu Jingnan even argued that the Yili shigong was co-authored by Li and Zhu. Shu, Zhu Xi nianpu changbian, 1290. What we clearly knew is that Li joined the compiling project of the Comprehensive Commentary no later than 1196. Shu Jingnan, Zhu Xi nianpu changbian, 1249. 52. Yang, YLJZTJ: JL, 412. Most of the annotation for this part was composed by Yang Fu. Zhu Xi only outlined the framework of the sacrificial section for the Comprehensive Commentary. 53. According to the Dictionary of Erudition, the space between a hu 戶 (door) and a you 牖 (window) is called yi 扆. Wang Pinzhen and Li Rugui regarded yi as the space between the east of the window and the west of the door. Wang, Da Dai liji jiegu, 200; Li, Yili shigong, 6a. 54. According Yang Fu, in the four seasonal sacrifices, when the sacrifice is performed within the hall, the sacrificial mat should be placed under its wall, and the ritual vessel is placed on the east. Yang, YLJZTJ: JL, 412. 55. Yang Fu annotated the fang here as the western room, since during the sacrifice the lord stays in the rightwestern room 房，西方也，諸候在右房也. Yang, YLJZTJ: JL, 412. 56. The character rong 榮 refers to the east cornice of an architecture. 57. Say, the hall is twenty feet deep, then the distance between the washing utensils beneath the east cornice (at the southeast corner of the temple) and the hall should also be twenty feet. 58. The left side is where the honorable guest resides. In this case, the “guest” is the spirit of the ancestor. Note that here the left and right positions are oriented from the perspective of the person who enters the door 以入為 左右. Wang, Da Dai liji jiegu, 201. 59. The long table, ji 几, is used to rest the spirit of the ancestor, which is emblematized by his clothes. Wang, Da Dai liji jiegu, 201.

146

Empowered by Ancestors assistant,60 with his hand washed, ascends to the hall and enters the [west] room. He prepares a sacrificial set by using the minced and marinated meat. The lord, with his hand washed, pours the liquor and makes an offering to the west where the sacrificial meat is placed. All the attendants return to their positions. The lord and the zhu bow again and then stand up. The zhu speaks the following words thrice: “The feudal lord X, the filial son, with auspicious ritual coins, humbly submits this petition to my ancestor, the feudal lord Y, whose great spirit is now able to rest in the renewed temple on the auspicious Z day of this month. Humbly submitted.” After a third bow, the lord enters the eastern subsidiary chamber, facing west, when the zhu enters the western subsidiary chamber, facing east. All the other attendants quickly step away from their paths, as gentlemen will do after a meal. Then the master of ceremony raises his hand and says: “Please return to your positions.” Then the lord steps back to his position, with the zhu following him and standing on his left. Then the ministers and various officials who have designated positions all return to their own positions. The zhu speaks the following words thrice: “The feudal lord X, the filial son, performs my ablution to make a bright offering with the sacrificial meat.” Then the lord and the zhu bow again: The lord returns to his position (in the east subsidiary chamber); the zhu leaves and return to his position (in the western subsidiary chamber). Then the master of ceremony declares: “The transfer of spirit tablet has been completed. Please attend the post-sacrifice banquet.” The lord exits the temple with his ministers, ritual masters, and all the other attendants. The practice of informing the ancestor is completed.61 至於新廟，筵於戶牖閒，樽於西序下，脯醢陳於房中，設洗當東榮，南北以 堂深。有司皆先入，如朝位。祝導奉衣服者乃入，君從。奉衣服者入門， 左。在位者皆辟也。奉衣服者升堂，皆反位。君從升。奠衣服于席上，祝奠 幣于几東。君北向，祝在左，贊者盥，升，適房，薦脯醢。君盥，酌，奠于 薦西，反位。君及祝再拜，興。祝聲三曰：「孝嗣候某，敢以嘉幣告于皇考 某候，今月吉日，可以徒于新廟，敢告。」再拜，君就東廂，西面。祝就西 廂，東面。在位者皆反走辟，如食閒。擯者舉手曰：「請反位。」君反位， 祝從，在左。卿大夫及眾有司諸在位者皆反位。祝聲三曰：「孝嗣候某，潔 為而明薦之享。」君及祝再拜，君反位，祝徹，反位。擯者曰：「遷廟事 畢，請就燕。」君出廟門，卿大夫、有司、執事者皆出廟門，告事畢。

As we can see, Yang Fu’s Sacrificial Rites fully exploits the liturgical details that are recorded in the Dadai liji, especially the sixth-century Classicist Lu Bian’s 盧 辯 (~557) commentary on it. The basic architectural components of the temple, including the entrance door, the main hall, the two rooms, and the two subsidiary chambers, are all documented in the Dadai liji. Moreover, it is Lu Bian who first pointed out that the sacrificial mat should be placed somewhere between the door and the window of the temple hall, prior to the arrival of the ancestral spirit in the 60. Zan 贊 here refers to the ceremonial assistant, not the religious reciter. In the sacrifices held in the mingtang hall, ministers and senior officials assist the ruler to perform the sacrifice, their wives assist the ruler’s wife 卿 大夫贊君，命歸贊夫人. See the Mingtang wei 明堂位 (The positions in the Luminous Hall) chapter in the Book of Rites. Zhu Bin, Liji xunzuan, 484; Legge, The Sacred Books of China, 4:33. 61. Yang, YLJZTJ: JL, 412–13; Wang, Da Dai liji jiegu, 200–202.

Daoxue Conception of the Song Imperial Temple

147

ritual of transferring tablets.62 Lu’s commentary also indicates that the temple of the feudal lord should include a left room and a right room.63 The Dadai liji, as well as the early Han dictionary Erya 爾雅, compartmentalizes the ancestral hall into a front-hall space and two subsidiary chambers (xiangfang 廂房).64 In Guo Pu’s 郭璞 (276–324) commentary on the Erya, the subsidiary chambers have been renamed to jiashi 夾室—a term that gained popularity in the Northern Song ritual debates concerning the Primal Ancestor position. Yang Fu carefully integrated these terms and related explanations in his Sacrificial Rites. Yang Fu’s explanations do not appear in Huang Gan’s earlier draft of the Comprehensive Commentary. Huang Gan’s edition substitutes the Dadai liji text cited in Yang’s Sacrificial Rites with passages and phrases from the Kaogong ji chapter of the Rituals of Zhou. After quoting the Wangzhi text concerning ancestral sacrifices of different social statuses, the Huang edition continues with the jiangren yingguo 匠人營國 (artisans’ city design) section of the Kaogong ji, discussing the setting of the Imperial Temple, its spatial relations with the palace and the infield altar, as well as the basic measurement of the temple hall.65 The whole section ends with an explanatory note of several ancient terms about different parts of the Imperial Temple, including beng 閍 (entrance door), tang 唐 (paths), and chen 陳 (lanes) in the temple space.66 Huang then began the next section with a discussion of the function of the tiao-preservation office in the Zhou bureaucracy. In contrast to the Huang edition, Yang Fu’s Sacrificial Rites replaces the Kaogong ji text with other passages from the Jifa text of the Book of Rites and the Kongzi jiayu.67 Additionally, the Sacrificial Rites cites the two Dadai liji chapters, the Zhuhou qianmiao 諸侯遷廟 and the Zhuhou xinmiao 諸侯釁廟, along with Lu Bian’s commentaries.68 Why did Yang Fu take the Kaogong ji part out of Huang Gan’s original edition and replace the Kaogong ji part with texts from other sources? A possible explanation can be attributed to the authors’ approaches and their different target audiences. When Huang Gan compiled his edition of the Comprehensive Commentary based on Zhu Xi’s instructions and guidelines, the Daoxue scholarship of ritual learning was still in its formative stage. New Learning ritual writings, such as Chen Xiangdao’s Lishu and Wang Zhaoyu’s Zhouli xiangjie, were still influential among ritual scholars. Zhu himself shared with New Learning ritualists the same confidence in the authority of the Rituals of Zhou.69 For the first generation of students who studied with Zhu Xi, the learning of the Rituals of Zhou was equivalent to, if not more important than, the learning of the other two ritual Classics in postulating 62. Wang, Da Dai liji jiegu, 200. 63. Wang, Da Dai liji jiegu, 200. 64. Wang, Da Dai liji jiegu, 201. 65. Huang Gan, Yili jingzhuan tongjie xujuan 儀禮經傳通解續卷 (hereafter YLJZTJXJ), 132:25a.5b–10b. 66. Huang, YLJZTJXJ, 25.10b. Chen here refers to the main lane that connects the entrance gate to the main hall. Li, Yili shigong, 15b–6a. 67. Yang, YLJZTJ: JL, 400–409. 68. Yang, YLJZTJ: JL, 410–15, shan 山: 19a–24b. 69. ZZYL, 86.2203.

148

Empowered by Ancestors

their own ritual scholarship. In this light, the proliferation of the learning of the Rituals of Zhou among the Southern Song Daoxue scholars was a natural response to the New Learning scholars’ monopoly on the related learning in the preceding decades. Therefore, the inclusion of the Kaogong ji section in Huang’s draft of the Comprehensive Commentary reflects the influence of New Learning ritual scholarship in a paradoxical way. On the one hand, Daoxue scholars at Zhu Xi’s time well recognized the textual authority of the Rituals of Zhou. On the other hand, the same group of scholars aimed to deconstruct the New Learning interpretations on the Rituals of Zhou and substituted them with the Daoxue understanding of the same text. The very idea of substitution reflects how Daoxue scholars were influenced by their New Learning predecessors in terms of ritual learning. Huang Gan was, in his very essence, a conservative scholar who devotedly followed Zhu Xi’s instruction in contending for the leadership of ritual learning with not only the New Learning community but also with other potential rivals of Zhu Xi’s scholarship.70 Therefore, Huang kept the Kaogong ji part in his edition to highlight the Daoxue emphasis on the Rituals of Zhou—a discipline that he inherited from his master Zhu Xi. Compared with Huang Gan, Yang Fu—as the most erudite ritualist of a later generation—was more inclusive and less afraid of adopting new ideas to modify Zhu Xi’s ritual learning. Instead of establishing a Daoxue disciplinary matrix of ritual learning—a work that had been largely completed by Zhu Xi himself, Yang was more concerned about the permanent solutions to the theoretical dilemmas in sacrificial rituals. He assumed that his target audience was those genuine Confucians who shared the same ritual learning with him (xili junzi 習禮君子).71 These Confucians could work with him to rectify several main issues in sacrificial rituals, such as the inner structure of the Imperial Temple.72 Yang Fu grounded his ritual learning primarily on Classicist theories.73 Yang composed his commentary on the Comprehensive Commentary by organizing his sources in a way that resembled the narrative of the Annals of Spring and Autumn. At the heart of Yang Fu’s Sacrificial Rites is a pragmatic intertwinement of historical and Classical texts. In this light, Yang’s work deserves more recognition not only because it provides additional comments based on Zhu Xi’s work but also because it reveals a set of new understanding of sacrificial rituals by restructuring Classical texts. 70. One of these potential rivals was the Yongjia School 永嘉學派 that was active in modern Zhejiang. On the one hand, the Yongjia community competed with Daoxue scholarship in what Hilde de Weerdt has called the “examination field” by developing new exam standards and curriculum. Hilde de Weerdt, Competition over Content: Negotiating Standards for the Civil Service Examinations in Imperial China (1127–1279) (Cambridge: Harvard University Press, 2007), 89–169. On the other hand, the Yongjia approach to rituals, which was usually regarded as utilitarian and instrumental, tended to read rituals in a historical way. Hence, the Yongjia scholars focused more on concrete performance and liturgical details than what Zhu Xi called the “original intent” of ritual practices. 71. Yang used the term in his preface to the Comprehensive Commentary. YLJZTJ: JL, xu.5. 72. As Yang Fu’s preface said, these main issues include suburban Altar offerings, mingtang sacrifices, the North Altar sacrifices, ancient and contemporary temple rituals, as well as di, xia, and seasonal sacrifices. YLJZTJ: JL, xu.4. 73. Yang, YLJZTJ: JL, Ye Chunfang, Introduction, 38–44.

Daoxue Conception of the Song Imperial Temple

149

While Zhu Xi and Huang Gan emphasized the structural differences between a temple (miao) and a main chamber (qing), Yang Fu enriched the abstract structure of “temple and main chamber” with the vivid description of the performance of temple rituals in the Dadai liji.74 Not only did the inclusion of the Dadai liji text and all its architectural terms reveal Yang Fu’s practical approach to explicating the temple structure, but it also provided a performative manual for the theoretical model constructed by his master Zhu Xi. The architectural terms indicated in Zhu Xi’s writings and diagrams gained practical meanings in Yang Fu’s text, as they were corroborated by ritual utensils, practitioners, and related movements. The tang is not an empty hall, but a hall with sacrificial mats and ritual vessels arranged at appropriate places. The western room is where the sacrificial meat will be placed. The lord and the ritual master zhu keep walking between the subsidiary chambers and the rooms during the ritual performance, while the mediums of ancestral spirits are always kept in the middle of the main hall. With new sources from Zhu Xi’s works and colloquial teachings, as well as other materials, Yang Fu and his Sacrificial Rites contextualize Zhu’ Xi’s imagination of the Imperial Temple in a telling way. In the Sacrificial Rites, temple rituals become more vivid and practical. Yang Fu also adopted the same practical approach in his annotation on different components of the temple complex, in which he carefully labeled their functions and how they were associated with the movement of ritual practitioners.75 An overlooked detail in the Sacrificial Rites is the concrete setting of the Imperial Temple’s main chamber. Fortunately, Zhu Xi attached a layout of the temple’s main chamber at the end of one of his memorials on temple rituals, as shown in Figure 6.3 (see p. 150). This layout conforms to Zhu’s hypothetical blueprint for the temples and chambers of Zhou feudal lords. Under normal circumstances, a single spirit tablet (or other relics) of the temple’s ancestor will be placed at the zhu position of the main chamber. However, in the cases of feudal lords and emperors, when a xia sacrifice is held, the tablets of all imperial ancestors will be assembled in the Primal Ancestor temple and arranged in a zhaomu order.

Zhu Xi’s Conception of the Zhaomu Sequence Zhu Xi recognized Lu Dian’s contributions in the 1079 debate about the Imperial Temple. However, he was dissatisfied with Lu Dian’s interpretation of the zhaomu sequence of Song imperial ancestors.76 Against Lu Dian’s zhaomu interpretation, Zhu reiterated that zhao and mu designations of imperial ancestors should never be altered, regardless of any change caused by the newly deceased ancestors. Zhu’s 74. Huang, YLJZTJXJ, 25.10a–b. 75. Yang, Yili pangtongtu, 2a–3a. 76. Zhu, “Dixia yi,” 69.1264; Yang, YLJZTJ: JL.448.

150

Empowered by Ancestors

Figure 6.3: Zhu Xi’s Visualization of the Imperial Temple’s Main Chamber

[Zhu Xi’s comments] The arrangement of temple’s main chamber is always like this: the spirit tablets reside beneath the west wall, facing east. While a xia sacrifice is held [in the temple], the Primal Ancestor’s tablet faces east [and occupies the zhu position], with the zhao tablets facing south [on its left] and the mu tablets facing north [on its right]. 廟室之制皆如此。其主皆在西壁下東向。祫則太祖東向，昭南向，穆北向。* * Zhu, “Yaomiao yizhuang,” Huian ji, 15.223. Based on Wei Shi’s quote of Zhu Xi in the Liji jishuo, I modified Zhu’s layout a little bit by marking where the zhu position is and the spatial arrangement of zhao and mu ancestors in the xia sacrifice. Wei, LJJS, 30.45a–b. It is noteworthy that Zhu’s orientation of the door and the windows concurs with the mainchamber layout recorded in the Rites and Ceremonies. See Li Rugui, Yili shigong, 5a. Yang Fu also portrayed a layout of the main chamber of Temples in his personal ritual writing. Yet, Yang’s layout contains only one window and fails to mark the position of spirit tablet, possibly because it refers to the main chambers of scholar-officials’ family shrines rather than that of Imperial Temples. Yang, Yili pangtongtu, 5a–b.

conception of zhaomu originated from his understanding of the very concept of genealogical order, which can be traced to his early writings. In a letter to Lu Jiuling 陸九齡 (1132–1180), the elder brother of Zhu Xi’s intellectual rival Lu Jiuyuan 陸九淵 (1139–1192), Zhu stated that the zhao and mu titles in the ritual of “temple abolishment” should be preserved, according to the ritual principle that “zhao ancestors are always kept as zhao, mu ancestors are always kept as mu” 昭常為昭穆常為穆.77 Zhu told Lu Jiuling that this principle was first raised 77. Zhu, “Da Lu Zishou” 答陸子壽, Huian ji, 36.569.

Daoxue Conception of the Song Imperial Temple

151

by a “ritualist” in the Northern Song. Despite the omission of the name in Zhu’s letter, the “ritualist” mentioned by Zhu Xi is clearly He Xunzhi.78 When Zhu Xi wrote this letter to Lu Jiuling in 1147, he was still developing his ritual studies.79 At the end of the letter, Zhu admitted that he was not familiar with the ritual Classics and lacked expertise in conducting meticulous textual analyses, which he termed as “kaozheng” 考證.80 It seems that Zhu tended to discuss the zhaomu issue with his friend Lu Jiuling in a compromising tone. However, consequent upon his accumulation of ritual knowledge in the following years, Zhu leaned more toward He Xunzhi’s zhaomu approach. In the later essay “Dixia yi,” Zhu argued that the general zhaomu sequence should be arranged in a way that the shift of zhao ancestors and mu ancestors are strictly limited to vertical movements along their own axes.81 By differentiating zhao and mu as two separate lines of the same genealogical order, Zhu summarized He Xunzhi’s zhaomu approach in the 1079 debate.82 In his late writings, including private conversations, essays, and memorials, Zhu repeatedly emphasized the principle that “zhao ancestors are always kept as zhao, mu ancestors are always kept as mu.”83 When Zhu Xi began to compile the Comprehensive Commentary, he developed a more sophisticated understanding of the zhaomu sequence. In their supplementary editions to the Comprehensive Commentary, Huang Gan and Yang Fu preserved some excerpts of Zhu Xi’s zhaomu ideas that were composed in Zhu’s late years. If comparison is made between the two supplementary editions of the Comprehensive Commentary, juan twenty-five of the Huang edition is equivalent to juan seven of the Yang edition in terms of content. Additionally, both editions discuss the establishment of the Imperial Temple in relation to its zhaomu order. They start with the familiar quote from the Rituals of Zhou as a section title, that is, “the differentiation of spirit tablets and temples is based on the zhaomu sequence” 辨廟祧之昭穆.84 As Zhu Xi formulated the basic structure of the Comprehensive Commentary, the fact that he cited the Rituals of Zhou at the beginning of this section demonstrated how he valued this ritual Classic in referring to zhaomu.85 As shown, it was He Xunzhi 78. Zhu Xi mentioned the exact same principle of zhaomu as He Xunzhi did in the 1079 debate. See Lu Dian, “Zhaomu yi,” 6.11b. 79. For the dating of this letter, see Chen Lai 陳來, Zhu Zi shuxin biannian kaozheng 朱子書信編年考證 (Beijing: Sanlian shudian, 2011), 154. 80. Zhu, “Da Lu Zishou,” 36.569. 81. Zhu, “Dixia yi,” 69.1266; Yang, YLJZTJ: JL, 454. 82. As Zhu put it, “for ten thousands of generations the zhaomu order should never be altered. How can it be like what Lu has said” 昭穆是萬世不可易，豈得如陸氏之說. ZZYL, 89.2283. 83. See, for instance, his letter to his disciple Ye Weidao 葉味道 (1167–1237). Zhu, “Da Ye Weidao” 答葉味道, Huian ji, 58.1054. In conversations with his other disciples, Zhu also expressed that the zhaomu order in the Imperial Temple should never be altered. ZZYL, 89.2283. 84. Huang, YLJZTJXJ, 25.1b; Zhouli zhushu, TSZSSSJ, 2:19.187. 85. Huang, YLJZTJXJ, 25.1b. Under usual circumstances, Zhu adopted phrases from the Rites and Ceremonies to be the titles of different sections. However, in this case, Zhu particularly cited a text from Rituals of Zhou to serve as a section title, because he considered this text as a reliable record of ancient temple rituals. Several times, Zhu asserted that both the Rituals of Zhou and the Rites and Ceremonies were the reliable sources in discussing ancient institutions and rituals. ZZYL, 86.2203, 2205. Yet, in some rare cases, he acknowledged

152

Empowered by Ancestors

who fundamentally shaped Zhu Xi’s understanding of zhaomu as a ritual legacy of antiquity. Nevertheless, Huang Gan’s and Yang Fu’s supplementary editions of the Comprehensive Commentary record also that zhaomu embodies a patriarchal idea of familial relationship, in which father-ancestors are associated with zhao ancestors and son-ancestors are associated with mu ancestors.86 In the later part of the Comprehensive Commentary, it cited a sub-commentary with a Zhou example to elaborate the zhaomu idea in a familial structure: The Zhou people honored Hou Ji as their Primal Ancestor and thus in particular built a permanent temple for him. Counting successively from Bu Zhu (Hou Ji’s son), Bu Zhu as the father was the first zhao ancestor, Ju as the son was the first mu ancestor, and henceforward zhao designated fathers and mu designated sons. When it came to King Wen, since he belonged to the fourteenth generation (of the Zhou family since Bu Zhu), genealogically the Zhou people called him a mu ancestor.”87 周以后稷廟為始祖，特立廟不毀，即從不窋已後為數，不窋父為昭，鞠子為 穆。從此以後，皆父為昭，子為穆，至文王十四世，文王第稱穆也。

The above text comes from Kong Yingda’s sub-commentary on the Rituals of Zhou.88 The question is why Zhu Xi, or Huang Gan and Yang Fu, quoted a Tang sub-commentary in the Comprehensive Commentary that clearly contradicted their own zhaomu ideas. An immediate answer is that Zhu Xi has included in the Comprehensive Commentary a collection of ritual theories, including theories contradictory to his own ones. In order to explain the contradiction, it is necessary to examine Zhu’s zhaomu conception with reference to his other ritual writings. Generally, Zhu Xi conceptualized zhao and mu as the spatial indicators of the directions of temples and spirit tablets. As he argued: “Zhaomu was originally named according to whether the temples were on the east or on the west and whether the tablets were facing south or north. At the beginning, the term had nothing to do with the designations of fathers and sons” 昭穆本以廟之居東、居西、主之向南、 向北而得名，初不為父子之號也.89 His justification followed: “if zhao and mu designate fathers and sons definitely, how can it be possible that the son of a mu ancestor be named as a zhao ancestor” 必曰父子之號，則穆之子又安得復為昭哉?90 In this light, although the Zhou court designated Zhou father-ancestors as zhao and son-ancestors as mu, zhaomu was not originally designed to indicate any paternal that some ancient rites were impractical. For instance, in a conversation with his disciples, he said Duke Zhou was careless in regulating the use of reddish jade ornaments and white jade disks in the funeral rite of lian 斂. ZZYL, 86.2233. 86. Huang, YLJZTJXJ, 25.1b; Yang, YLJZTJ: JL, 393. 87. Huang, YLJZTJXJ, 25.2; Yang, YLJZTJ: JL, 394. 88. Zhouli zhushu, TSZSSSJ, 2:19.187. 89. Zhu, “Dixia yi,” 69.1266. 90. Zhu, “Dixia yi,” 69.1266. Hua Zhe reads this as Zhu Xi’s provoking challenge to the conventional zhaomu pattern. See Hua, “Fuzi yilun: Bei Song Yuanfeng zhiyi zai pinjia,” 24–25.

Daoxue Conception of the Song Imperial Temple

153

relationship. Because of the regular pattern of patrilineal succession in the Zhou imperial line, it was natural for the Zhou court to bestow Zhou ancestors after the Primal Ancestor Hou Ji zhao and mu designations alternately, given its conformity to a neat pattern of father-ancestors as zhao and son-ancestors as mu.91 In other words, it was only by coincidence that the zhaomu of Zhou ancestors was arranged in such a neat pattern. Nevertheless, Zhu argued, on occasions when the regular mode of patrilineal succession is disturbed, the court should correspondingly alter the zhaomu to accommodate the temple setting, regardless of the natural order of seniority. Taking the Song imperial line as an example, Zhu championed the idea that Emperor Taizu and Emperor Taizong should be separated in terms of zhaomu, despite their sibling relationship. He emphasized that the orthodox practice of temple rituals gave priority to the monarch-and-subject relationship over the sibling relationship. According to the Confucian doctrine of the “rectification of names” (zhengming 正名), Zhu Xi argued that the two emperors Taizu and Taizong should be perceived as two separate generations in the Imperial Temple.92 In opposition to Lu Dian’s perspective that zhao and mu ancestors can shift freely between the zhao line and the mu line in the same manner as they shift between the right dan altar and the left shan yard, Zhu Xi questioned the presence of a zhaomu sequence outside the temple context. Zhu pointed out that all the ancient ritual writings record nothing about the arrangement of the dan altar and the shan yard.93 Therefore, he argued that dan and shan do not adopt a distinguishable zhaomu order, but they only bear the zhao and mu designations of ancestors therein. Unlike the arrangement of spirit tablets in the temple, which shifts along the zhao and mu lines, dan and shan are individual sacrificial spaces beyond the ritual limit of seven generations.94 Borrowing Lu Dian’s terminology, Zhu Xi distinguished the temple sequence of miaoci from the genealogical sequence of shici in the same manner as Lu did in his critique of He Xunzhi’s zhaomu theory. However, Zhu reached a different conclusion based on the distinction between shici and miaoci. According to Zhu, miaoci and shici are incommensurable because they belong to different ritual domains. Miaoci is symbolized by the zhaomu of the Imperial Temple; shici, which plays a role in regulating the order of non-temple ancestral spaces, especially the dan altar and the shan yard, solely adopts zhao and mu designations based on genealogical order. In Zhu Xi’s words, from the day when an imperial house established its temple, zhao and mu ancestors within the temple would be naturally sorted out by the order of their occurrences: the second ancestors the first zhao, the third the 91. ZZYL, 90.2298. 92. Zhu, “Xiaotiezhi,” 15:227. 93. There are indeed some statements concerning the orientation of the dan altar and the shan yard in the commentaries and sub-commentaries of ritual Classics. However, as Zhu said, these statements all originated from some questionable records of Han and Tang Confucians. Zhu, “Dixia yi,” 69.1266. 94. In this light, Zhu named the dan hall and the shan altar as the same thing, despite the obvious differences of their architectural features. Zhu, “Dixia yi,” 69.1266.

154

Empowered by Ancestors

first mu, and so forth.95 As zhao and mu represent fixed positions in the Imperial Temple, the ancestral lines of zhao and mu never intersect with each other.96 If a son occupies the zhao position in the temple and his father resides at a mu position across from him, the order of seniority remains unchanged. Zhu quoted the case of the King Cheng of Zhou 周成王 to explain this: “Although in King Cheng’s time his grandfather King Wen was placed at the mu position, King Wen was still more superior to King Wu in terms of ritual status. By the same token, the fact that King Wu was placed at the zhao position did not necessarily indicate that King Wu’s ritual status was higher than that of King Wen” 故成王之世，文王為穆而不害其 尊於武，武王為昭而不害其卑於文.97 The troubling case of designating the fatherancestor as mu and a son-ancestor as zhao, which once hampered He Xunzhi’s efforts to resolve the zhaomu controversy in 1079, is no longer a problem in Zhu Xi’s zhaomu theory. Since Zhu Xi articulated the immutability of zhao and mu positions in the Imperial Temple context, some scholars have understood Zhu’s zhaomu theory as a new model of Chinese patriarchy.98 However, a scrutiny of Zhu Xi’s ritual writings reveals that Zhu had absorbed and inherited a myriad of ideas from previous ritual discussions into his understanding of patriarchy, especially the ideas of New Learning scholars like He Xunzhi and Lu Dian. Both Zhu Xi and He Xunzhi agreed that zhaomu presupposes an adherence to the direct transition of zhao and mu designations from grandfathers to grandsons, rather than to sons. Yet, in comparison with He Xunzhi’s response to Lu Dian, Zhu Xi was more confident in challenging zhaomu as an embodiment of mere paternal relationship. In answering his disciple’s question about the arrangement of burial grounds, Zhu asserted that “the zhaomu sequence merely differentiates generations, but not seniority” 昭穆但分世數，不 分尊卑.99 Hence, he defined zhao and mu designations as generational markers in most circumstances. Additionally, Zhu Xi argued that zhao and mu designations in the Imperial Temple are self-contained symbols. In other words, each ancestor is treated with equal reverence in his own temple.100 By distinguishing the sacrificial spaces of different temples from one another, Zhu demonstrated how zhao and mu ancestors as individual spirits reside together in harmony, with no violation of the cherished Confucian norms of seniority. If, as Zhu Xi said, that zhao and mu designations are generational markers, how are they concretely arranged in the Imperial Temple? Yang Fu’s Sacrificial Rites records that Zhu had cast doubt on the correlation between the left zhao positions and the idea of superiority: “The ancient setting of seats defined either the west 95. 96. 97. 98.

Zhu, “Dixia yi,” 69.1266. Zhu, “Dixia yi,” 69.1267; Wenxian tongkao, 91.829. Zhu, “Dixia yi,” 69.1266–67. Lin Zhenli 林振禮, “Zhu Xi puxu fawei” 朱熹譜序發微, Zhongguo zhexueshi 1 (2001:7): 62–72, especially 64–65. 99. Zhu, “Da Chen Anqing” 答陳安卿, Huian ji, 57.1040. 100. Wei, LJJS, 30.46–47.

Daoxue Conception of the Song Imperial Temple

155

or the south as the superior direction. It is not necessarily the case that ancient people always took the left as the more superior position” 古人坐次，或以西方 為上，或以南方為上，未必以左為尊也.101 Zhu also argued that during the xia sacrifices, when all imperial ancestors are arranged in the main chamber of the Primal Ancestor temple, only the Primal Ancestor can reside in the central position beneath the western wall. All the zhao ancestors, since they are moved into the main chamber of the Primal Ancestor temple, reside side by side beneath the northern window, facing south; all the mu ancestors reside side by side beneath the southern window, facing north.102 In this way, Zhu Xi associated the two characters zhao and mu with spatial directions. As he stated, the zhao ancestors are designated as zhao ancestors because they are facing the brighter southern side in the xia sacrifice.103 Since the mu ancestors are facing the dim and obscure northern side in the same sacrifice, they are bestowed the mu designation, given the connotative meaning of the character mu as “complexity and distance” in the ancient dictionary of Shuowen.104 Notably, Zhu Xi’s conception of zhaomu in the xia sacrifice differs from his general understanding of zhaomu. Zhu considered xia as a separate performative ritual. According to Zhu Xi, he heard from his master that there are two different types of xia rituals: one refers to the seasonal sacrifices performed in summer, autumn, and winter; the other refers to the state sacrifices made by feudal lords, kings, and emperors in their Primal Ancestor temples.105 For the second type of xia sacrifice, which involves collective offerings to ancestors, ritual masters will rearrange the zhaomu of ancestors to illustrate the correct order of seniority—as He Xunzhi and his colleague Zhang Zao had argued in the 1079 debate.106 Since the spirit tablets of the abandoned temples are not involved in the state xia sacrifice, the rearrangement of zhaomu is sometimes confusing, considering the irregularity in the imperial line of succession and the relevant shift of tablets along zhao and mu lines. By tracing back to the Zhou practice of the xia sacrifices, Zhu introduced what may be called a “theory of vacancy” to resolve the problem arising from irregularity. Zhu took the xia sacrifice performed during the King Zhao of Zhou’s reign 周昭 王 as an example to demonstrate his theory. In King Zhao’s time, when Wang Ji’s tablet (father of King Wen) was removed and King Kang’s 康王 tablet was added to the zhao line in the xia sacrifice, King Wu’s tablet would supposedly be moved upward to take over Wang Ji’s position to make place for King Kang’s. However, since King Wen as King Wu’s father resided across from Wang Ji in the mu line, the upward move of King Wu’s tablet to Wang Ji’s position would possibly violate the order of seniority, as King Wen would face north to his son. In Chinese orientation, 101. Yang, YLJZTJ: JL, 586–92. 102. Wei, LJJS, 30.45b. 103. Wei, LJJS, 30.45b. 104. Wei, LJJS, 30.45a–46b. 105. Yang, YLJZTJ: JL, 552. 106. Wei, LJJS, 30.44; Yang, YLJZTJ: JL, 552.

156

Empowered by Ancestors

a subject facing north spatially conveys a sense of obedience to the northern one.107 Consequently, Zhu Xi suggested keeping King Wu’s tablet in its original position and leaving the zhao position of Wang Ji—across from King Wen—vacant to prevent a possible violation of the order of seniority. The shift can be graphically presented in Figure 6.4: Figure 6.4: Zhu Xi’s Solution to the Seniority Problem in the Zhaomu Sequence

By introducing a vacant position into the performance of the xia sacrifice, Zhu Xi maintained the original zhaomu sequence and simultaneously aligned it with the conventional understanding of seniority as well as his understanding of the shifting pattern of zhao and mu ancestors. From the perspective of ritual performances, Zhu’s work, especially the Comprehensive Commentary, synthesizes significant performative details and provides more information about the spatial arrangement of zhaomu than other Southern Song ritual texts. Basically, Zhu Xi’s conception of temple architecture follows a north–south orientation. However, he visualized the space within the main chamber of the temple in an east–west alignment, in which zhao and mu tablets extend eastward, from the windows to the east wall.108 Zhu Xi’s ritual writings, as well as Yang Fu’s and Huang Gan’s supplementary editions to the Comprehensive Commentary, offer a detailed depiction of the zhaomu setting of the temple’s main chamber. Nevertheless, we should not overestimate the level of specificity rendered by Zhu’s research. In a private conversation, Zhu admitted that the ancient setting of 107. Yang, YLJZTJ: JL, 551–52. 108. Apart from the sources we have discussed in the above, there are some discussions about the spatial arrangement of the main chamber in the Zhizi yulei. Most of them emphasize the correct perception of the zhu position, which should be located on the west, facing east. ZZYL, 90.2293, 2298.

Daoxue Conception of the Song Imperial Temple

157

the Imperial Temple’s main chamber was incomprehensible.109 In a personal letter to his disciple Guo Zicong 郭子從, Zhu Xi was frustrated about the zhaomu of burial grounds and failed to provide a compelling reason for some zhaomu orientation in clan burials.110 Zhu also acknowledged that he had difficulty in imagining the spatial arrangement of some ancient practices of the xia sacrifice, especially how multiple tablets were concretely placed in the Zhou performance of collective offerings, as over thirty ancestral tablets were involved in such a performance. Zhu doubted how the offerings would be possible, given the large amount of tablets and the relatively small space of the Zhou Imperial Temple.111 After all, the arrangement of the imperial zhaomu sequence remained a daunting problem for Zhu Xi throughout his later years.

Concluding Remarks Zhu’s perception of the general setting of the Imperial Temple demonstrated how his understanding of certain ritual ideas had been deeply influenced by previous ritual discourses, especially those launched by New Learning scholars in Northern Song ritual debates. By linking the Yuanfeng ritual debate in 1079 to his contemporary ritual controversy, Zhu Xi highlighted the importance of reviving ancient rituals that had already been emphasized by Northern Song ritual officials. Ritual officials with New Learning backgrounds, especially Lu Dian and He Xunzhi, as well as their master Wang Anshi, initiated a campaign of ritual revivalism. Zhu Xi continued that campaign and thoroughly discussed the spatial arrangement of an ideal Imperial Temple and the zhaomu order within. By doing so Zhu laid a theoretical foundation on which his disciple Yang Fu later added more performative details in his supplementary edition to the Comprehensive Commentary. As a collective work of Zhu Xi and his disciples, the Comprehensive Commentary resembles New Learning scholars’ endeavors to negotiate the tension between ancient ritual models and contemporary ritual practices regarding discussions on the Imperial Temple. In his acclaimed work of Song political and intellectual history, Yu Yingshi emphasizes the intellectual continuity between Wang Anshi and Zhu Xi. He argued that the Daoxue rubric initiated by the Cheng Brothers was a counter-movement to Wang Anshi’s New Learning.112 This is an important observation, which has been substantiated by new evidence in recent studies.113 My study on Zhu Xi’s conception of the temple further corroborates Yu Yingshi’s observation by arguing that the Daoxue rubric could be comprehended as a reaction to New 109. ZZYL, 90.2296. 110. “Da guozicong,” Huian ji, 63.1162. 111. ZZYL, 90.2296. 112. Yu, Zhu Xi de lishi shijie, 36–64. 113. See, for example, Cheung Hiu Yu, “The Way Turning Inward: An Examination of the ‘New Learning’ Usage of daoxue in Northern Song China,” Philosophy East and West 69.1 (2019.1): 86–107.

158

Empowered by Ancestors

Learning ritual scholarship. In terms of ritual revivalism, shared ideas and agendas about the Imperial Temple and temple rituals cut across the divides between the two intellectual traditions of New Learning and Daoxue. After the Daoxue ritual norms penetrated into the society throughout the thirteenth century, their intellectual origins were ignored by later generations. In this light, the story of how Zhu Xi understood temple rituals and constructed his own zhaomu theory is not a trivial topic of ritual studies, but instead it is a lens through which we can understand the intellectual correlation between Daoxue and New Learning.

7 Socialization of the Discourse of Temple Rituals in the Late Southern Song

Zhu Xi’s ritual learning and his explication of temple rituals left a significant legacy to his contemporaries. In Emperor Lizong’s 理宗 (r. 1224–1264) reign, ritual officials referred back to Zhu Xi and his emphasis on Yuanfeng ritual reforms in dealing with ritual controversies over the Imperial Temple. In 1231/9, the Imperial Temple in Linan was burned down in an accident. Du Zheng 度正 (1166–1235), the deputy prefect of the Court of Imperial Sacrifices and an admirer of Zhu Xi, submitted a memorial to the emperor and claimed that it was opportune to reexamine the temple arrangement after the fire accident. Considering the fire as a bad portent, Du advised the court to reconsider Zhu Xi’s earlier opinion in placing Xizu’s tablet at the center of the Imperial Temple. Du Zheng suggested two plans in his memorial. The first plan was an adapted version of Zhu Xi’s temple scheme; the second plan was a compromise between Zhu Xi’s scheme and the conventional temple settings in Emperor Lizong’s time.1 In a tone of regret, Du stated that the Yuanfeng ritual controversy in 1079 had not received adequate attention.2 Among the ritual officials who were involved in the Yuanfeng controversy, Du highlighted the contributions made by Lu Dian, possibly because Zhu Xi had mentioned Lu in the “Dixia yi.” Hence, Du suggested compartmentalizing the Song Imperial Temple in a way that the front areas of each chamber would be combined into one single space to host the offerings to ancestors in state sacrifices. However, after some discussions at a collective advisory meeting, the court rejected Du Zheng’s suggestion.3 Du Zheng’s proposed revision on the Imperial Temple in 1232 marked the decline of ritual officials’ interests in temple-related issues. No court debates over the Imperial Temple had left their traces in the official records of the thirteenth century. Instead, the advocacy of establishing ancestral halls (citang 祠堂) gained popularity in southern rural regions. While a great deal of scholarly attention focuses on the social history of ancestral worship in southern rural areas after the thirteenth century, scant scholarship explores the relationships between imperial ancestral rituals and the social norms of ancestral worship. Owing to the vital role 1. SS, 107.2589. 2. SS, 107.2590. 3. SS, 107.2590.

160

Empowered by Ancestors

of the Imperial Temple in the rubric of imperial ancestral rituals, I am going to examine the “socialization” of the discourse of temple rituals in Southern Song and Yuan periods. As the term socialization here refers to a differentiation between central court and local society, the socialization of temple rituals indicates a detachment of these rituals from their imperial context.4 Although most temple rituals as social practices had not been widely instituted until the sixteenth century, some new understandings about these rituals had been formally constructed by Southern Song scholars and educated elites in various ritual texts. Still, Zhu Xi’s Comprehensive Commentary serves as a point of departure for our adventure in these texts.

From the Comprehensive Commentary to Other Southern Song Ritual Commentaries In all fairness, the Comprehensive Commentary was a tremendous success. Not only did it crystallize the ritual learning of Zhu Xi and his direct disciples, but it also laid the cornerstone for the ritual scholarship of later Confucians who recognized themselves as Daoxue scholars. In the two major editions of the Comprehensive Commentary, Zhu Xi, Huang Gan, and Yang Fu generally understood the ritual intent of the Imperial Temple and related rituals based on a quote from the Doctrine of the Mean. As it is said: In spring and autumn, they repair and beautify the temple halls of their ancestors, set forth their ritual utensils, display their various robes and garments, and present the offerings in seasonal sacrifices. By means of the rituals of the Imperial Temple, they distinguish the imperial kindred according to the zhaomu sequence.5 春、秋修其祖廟，陳其宗器，設其裳衣薦其時食。宗廟之禮，所以序昭穆也。

The subjects of this passage refer to King Wu and the Duke of Zhou.6 Rhetorically, the revision of ancestral rituals in the Comprehensive Commentary is built upon the authors’ interpretations of the Zhou ritual rubric. However, as we have analyzed in Chapter 6, Zhu Xi, Huang Gan, and Yang Fu compromised the basic idea of Zhou ancestral rituals with concrete ritual practices that had been recorded in performative texts, such as the Jiangdou jili and the Dadai liji. In his celebrated commentary on the Doctrine of the Mean, Zhu Xi cited texts from the Jitong chapter to annotate the sentence that “by means of the rituals of the Imperial Temple, they distinguish the imperial kindred according to the zhaomu

4. For a discussion on the adaptability of the term socialization in modern studies of Chinese rituals, see Robert Weller, “Religion, Ritual, and the Public Good in China,” in Confucianism and Spiritual Traditions in Modern China and Beyond, ed. Fenggang Yang and Joseph B. Tomney (Leiden: Brill, 2012), 332, 343–45. 5. Yang, YLJZTJ: JL, 618. For the original sources, see Liji zhushu, TSZSSSJ, 2:52.569; I consulted James Legge’s translation here. See Legge, The Sacred Books of China, 4:310. 6. Zhu Bin, Liji xunzuan, 775.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 161

sequence.”7 Considering the significance of the Doctrine of the Mean in Zhu Xi’s philosophy,8 Zhu’s citation of a less authorized text in annotating a text under canonization reveals what Walter Benjamin called an “implicit interruption to the fundamental structure and context” of the established textual norms.9 By conceptually bridging the theoretical zhaomu notion mentioned in the Doctrine of the Mean and the operational zhaomu sequence described in Jitong, Zhu Xi initiated a synthesis of the “higher-order” ritual theories based upon a discourse of reviving ancient Zhou rituals and the “lower-order” ritual practices emphasizing consistency on the practical level.10 Theoretically, Zhu’s compilation of the Comprehensive Commentary gave the literati community access to the complicated theories of ancient rituals. His endeavors earned support from a number of Southern Song scholars, especially those who were also interested in remodeling their contemporary world through a revival of Confucian rituals. Consequently, discourses concerning the Imperial Temple in the thirteenth century saw the recurrence of intellectual synthesis initiated by Zhu Xi’s Comprehensive Commentary. These discourses embraced Zhu’s general understanding of the Imperial Temple as a crucial part of ancestral worship in symbolizing filial piety. In fact, opinions of scholars and ritualists in the late Southern Song followed the framework established by the Northern Song ritual officials and elaborated by Zhu Xi. “High-order” ritual theories with an imperial context, such as Xizu’s ritual status in the Imperial Temple, still constituted a crucial part of the related writings of Southern Song scholars. Nevertheless, when it came to “lower-order” ritual practices, these scholars and ritualists invented new discourses connecting the theories of temple rituals with their practices. I will analyze how the explications of temple rituals in two Southern Song ritual commentaries emphasize the performativity of rituals. The first piece of ritual text is the Zhouguan zongyi 周官總義 (Summary of the Rituals of Zhou), authored by Yi Fu 易祓 (1156–1240). Yi Fu’s contemporaries recognized him as a gifted but vicious scholar. Yi’s political affiliation with the powerful minister Han Tuozhou 韓侂胄 (1152–1207) and Han’s military adviser Su Shidan 蘇師旦 rendered Yi an infamous reputation as an opportunist.11 Despite his 7. Zhu Xi, Zhongyong zhangju 中庸章句, in Zhu, Sishu zhangjujizhu 四書章句集注 (Shanghai: Shanghai guji chubanshe, Anhui jiaoyu chubanshe, 2001), 31–32. The Daoxue scholar Yang Shi first linked the phrase from the Doctrine of the Mean to the Jitong text in his study of sacrificial rites (Wei, LJJS, 129.22a). The difference is that Yang Shi included both texts in the main body of the same paragraph, while Zhu Xi put the Jitong text as an annotation to the Doctrine of the Mean. 8. Soffel and Tillman, Cultural Authority and Political Culture, 52–86. 9. Walter Benjamin, Illumination (NY: Schocken Books, 2007), 151. 10. I borrowed the two terms here from Patricia Ebrey’s work on Chinese family rituals. Ebrey uses both “lowerorder” and “higher-order” in referring to ideas. She understands “higher-order” ideas as those about general Confucian ethics, such as filial piety and the authority of ancient sages. “Lower-order” ideas, in contrast, refer to Confucian scholars’ conceptions of ritual details. Ebrey, Confucianism and Family Rituals in Imperial China, 220. 11. Zhou Mi described how Yi flattered Su by writing him a draft edict of promotion, in which he compared Su with Confucius. Zhou, Qidong yeyu, 11.200. Also, Wang Kexi 王可喜, Wang Zhaopeng 王兆鵬, “Nansong

162

Empowered by Ancestors

bad reputation, Yi’s achievement in ritual learning was remarkable. For example, the Qing editors of Siku quanshu acknowledged that Yi’s Zhouguan zongyi had succeeded in introducing an intertextual analysis of Classics into the study of ancient rituals.12 No extant historical sources have suggested a direct link between Yi Fu and Zhu Xi. But Yi had some connections with the Daoxue fellowship. As a senior student of the Imperial College, Yi developed his interest in the Rituals of Zhou in youth.13 In 1194, Yi began to serve in the secretariat of Zhou Bida 周必大 (1126–1204), the Military Commissioner (anfushi 安撫使) of the Southern Jinghu Circuit (荊湖南路, modern Hunan 湖南 province).14 Zhou had patronized the Daoxue fellowship since the 1180s, when he had been promoted to the Grand Councilor position. Zhou Bida’s personal friendship with Daoxue leaders, such as Zhu Xi and Zhang Shi 張栻 (1133–1180), led to his sympathetic view of Daoxue scholarship and Daoxue-related scholars in the 1197 Qingyuan prohibition of False Scholarship. Although Zhou and Zhu Xi construed “the learning of the Way” ways and sometimes disagreed with each other about the means to promote Daoxue members, they partook in the same campaign of pursuing an intrinsic learning of the Way.15 Zhou Bida’s wide circle of acquaintances contributed to the spread of Daoxue scholarship to other scholars. As Hoyt Tillman has pointed out, the full-scale persecution of Daoxue scholars in the Qingyuan era revealed how the Daoxue fellowship as a loosely defined entity gradually gained attention from its contemporary allies and opponents.16 Zhou’s contribution to the long-term development of the Daoxue scholarship might be small, but he spoke for the Daoxue interest in the central government. Yi Fu’s early contact with Zhou Bida possibly accounts for his new interest in the Daoxue learning in ancient rituals. His extant work, the Zhouguan zongyi, serves as a testament to the influence of Zhu Xi’s ritual scholarship.17 Regarding Zhu Xi’s discourse on ancient rituals, Yi Fu was particularly attracted by Zhu’s hypothetical model about the temple of Zhou feudal lords. He cited Zhu Xi’s description of the model in considerable length in the Zhouguan zongyi.18 In explaining the setting of the Imperial Temple, Yi valued Zhu Xi’s model above and beyond the traditional ciren Yifu xingniankao” 南宋詞人易祓行年考, Zhongguo yunwen xuekan 中國韻文學刊, 19:4 (2005): 71–72. 12. In their own words, “to explicate the Classic texts based on Classics” (yijing shijing 以經釋經). Yi, Zhouguan zongyi, tiyao.2a. 13. Nansong guange xulu 南宋館閣續錄 (Beijing: Zhonghua shuju, 1998), 4:281. 14. Wang, “Nansong ciren Yifu,” 70. 15. See Yu, Zhu Xi de lishi shijie, 497–523, especially, 499–508. 16. Hoyt Cleveland Tillman (Tian Hao), “Ping Yu Yingshi de Zhu Xi de lishi shijie” 評余英時的朱熹的歷史世界, Shijie zhexue 世界哲學 (2004:4): 103–7. 17. Yi Fu, Zhouguan zongyi, SKQS, 92: tiyao.1a. 18. Yi, Zhouguan zongyi, 92:12.3a–5a. In addition to the Zhouguan zongyi, two other Southern Song commentaries cited Zhu Xi’s model in great length. One is Wei Shi’s encyclopedic work Liji jishuo; the other is the Zhouli jishuo 周禮集說 (Collective Explanations of the Rituals of Zhou), an anonymous work that was compiled by a Yuan scholar named Chen Youren 陳友仁. In a number of sections, the Zhouli jishuo quotes Wang Anshi’s Zhouli xinyi. The author might be one of Wang Anshi’s admirers in the late Southern Song. For the quoted part of Zhu Xi’s temple model, see Zhouli jishuo, SKQS, 95:4.40b–44b.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 163

interpretations. Although Zhu Xi mentioned several alternative interpretations about the Imperial Temple of the Son of Heaven in his own works,19 Yi Fu disregarded those interpretations and concluded his commentary by arguing that the Imperial Temple of the Son of Heaven should strictly follow Zhu Xi’s hypothetical model of the feudal lords’ temple.20 Yi appeared to have considered that traditional interpretations could be replaced by better alternatives, such as Zhu Xi’s temple model. Regarding the performance of concrete temple rites, Yi Fu remarked on the conventional performance of temple sacrificial rites of his time. Yi argued that the first two offerings of the nine libations of ritual wine (jiuxian 九獻) in seasonal sacrifices should be abandoned, because the performance of the two offerings assumes the ritual procedure of guan 祼, which refers to the pouring of wine on the ground. Yi considered this procedure as inappropriate to solemn rituals like seasonal sacrifices.21 While in Yi Fu’s time the guan rite had already been removed from the nine libations in the ancestral offerings to high-ranking officials, Yi questioned why the same rite was still performed in the Imperial Temple, which was clearly a more sacred space than the family shrines of high-ranking officials.22 Following Zhu Xi and Yang Yu, Yi linked the “lower-order” ritual practices of his contemporary world to the “higher-order” ritual theories in his ritual commentaries. The second piece of ritual text that reveals the influence of “lower-order” ritual practices is Wang Yuzhi’s Zhouli dingyi (Revised Explanations of the Rituals of Zhou). Wang Yuzhi was born in Leqing 樂清, in the prefecture of Wenzhou 溫州 (modern Zhejiang), where the regional scholarly tradition of the Yongjia School had a deep imprint. Wang’s Zhouli dingyi is the most comprehensive synthesis of Song commentaries on the Rituals of Zhou.23 It includes fifty-one commentaries covering a wide span of ritual writings from the eleventh to the thirteenth centuries. Albeit an admirer of the Daoxue scholarship, Wang Yuzhi had reservations about Zhu Xi’s ritual learning.24 The most compelling evidence is that he scarcely cited Zhu Xi’s work in the Zhouli dingyi. Considering that Zhu Xi’s Comprehensive Commentary was published two decades prior to the Zhouli dingyi, Wang’s oversight of Zhu Xi’s ritual learning is unusual.25 19. Wei, LJJS, 30:47b–48a. 20. Yi, Zhouguan zongyi, 12.5a. 21. Yi, Zhouguan zongyi, 92:12.14b–15a. 22. Yi, Zhouguan zongyi, 92:12.15a. 23. For an introduction of the Zhouli dingyi, see Song Jaeyoon, “Tension and Balance: Changes of Constitutional Schemes in Southern Song Commentaries on the Rituals of Zhou,” in Statecraft and Classical Learning: The Rituals of Zhou in East Asian History, 253. 24. According to Zhen Dexiu’s 真德秀 (1178–1235) preface to Zhouli dingyi, as a Yongjia scholar, Wang Yuzhi’s learning resonated with Cheng Yi and Zhang Zai’s scholarship. Zhen Dexiu, “Zhouli dingyi xu” 周禮訂義序, in Wang Yuzhi, Zhouli dingyi, QSW, 313:7170.166. 25. According to Zhen Dexiu’s preface, although a full version of the Zhouli dingyi had already been compiled in 1232, an officially authorized copy of it was only published in the second year of the Chunyou 淳祐 era (1242). Zhao Ruteng’s 趙汝騰 (d. 1261), the local governor of Wang Yuzhi’s hometown of Le Qing, submitted a printed copy of the Zhouli dingyi to the court in 1242. Zhao’s memorial and the court’s edict illustrate how

164

Empowered by Ancestors

In his preface, Wang claimed that his own Zhouli dingyi was written with close reference to Zhu Xi’s annotation format in the latter’s Lunmeng jizhu 論孟集 注 (Commentaries on the Analects and the Mencius).26 Intriguingly, while Wang was explaining his basic ideas about Zhou rituals, such as the priority of the Six Bureaus and the number of Zhou offices in the Rituals of Zhou, he tended to adopt the opinions from scholars other than Zhu Xi. He also had a special liking for the Yongjia scholars.27 Among the fifty-one commentaries that he had taken as references, twelve were composed by Yongjia scholars, including Xue Jixuan 薛季宣 (1134–1173), Chen Fuliang 陳傅良 (1137–1203), Liu Ying 劉迎, a scholar with the surname of Wang, Yang Que 楊恪, Chen Ji 陳汲, Huang Du 黃度 (1138–1213), Zheng Boqian 鄭伯謙, Cao Shuyuan 曹叔遠 (1159–1234), Lin He 林荷, Chen Wang 陳汪, and Li Jiahui 李嘉會.28 Geographical factors may explain Wang Yuzhi’s wide adoption of the Yongjia scholars’ commentaries on the Rituals of Zhou. Wang might have close affiliations with the Yongjia scholars who lived nearby his hometown and thus rendered their ritual writings more accessible to him. Still, it is necessary to ask why Wang Yuzhi overlooked Zhu Xi and the Comprehensive Commentary in the Zhouli dingyi. The answer lies in the text of the Zhouli dingyi. If one reads the explanatory note of the Zhouli dingyi, he would notice how Wang Yuzhi emphasized the importance of selecting appropriate commentaries for each passage in the Rituals of Zhou.29 In his preface to the Zhouli dingyi, Zhen Dexiu summarized the intention of Wang Yuzhi’s work as to reveal the “public heart” (gongxin 公心) of the Rituals of Zhou. Zhen Dexiu understood the “public heart” as a shared value of Ancient Kings and the Duke of Zhou, which sharply contradicted the “private heart” (sixin 私心) of later scholars and officials who contaminated the true meaning of the Rituals of Zhou by their deviant learning and political maneuvers.30 By juxtaposing the “public heart” and the “private heart,” Zhen implicitly criticized the Northern Song reformer Wang Anshi as a deviant scholar who had annotated the Rituals of Zhou with his “private heart” and made erroneous policies based on his deviant understanding of the Rituals of Zhou. However, Zhen Dexiu ignored the fact that Wang Yuzhi’s interest to reveal the “public heart” of the Rituals of Zhou was meant to include as many interpretative traditions in the Zhouli dingyi as possible. Alongside Daoxue scholars like Zhu Xi, Zhang Zai, and the Cheng Brothers, Wang Yuzhi quoted numerous commentaries from the Yongjia scholars who were not well-known but were more familiar with the ritual theories of the Rituals of Zhou. In a number of cases, Wang the state power and its local capillaries attempted to absorb regional scholarly traditions into a holistic project of ritual orthodoxy. Wang, Zhouli dingyi, SKQS, zoule 奏勒.1a–2b; diewen 牒文.1a–2a; zhouzhuang 州狀.1a–b. 26. Wang, Zhouli dingyi, tiaoli 條例.1a. 27. Wang Yuzhi discussed these ideas in his foreword to the Zhouli dingyi. See Wang, Zhouli dingyi, bianyan 弁言 1a–18a, especially 6a–9a, 12a–18b. 28. Wang, Zhouli dingyi, 93: xingshi 姓氏.3a–4a. 29. Wang, Zhouli dingyi, tiaoli.1a. 30. Zhen Dexiu, “Zhouli dingyi xu,” Zhouli dingyi, xu.1a–3b. The “private heart” in Zhen Dexiu’s Preface also refers to the self-interest of deviant rulers and politicians.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 165

modified the ritual theories of Yongjia scholars to suit contemporary needs. For the same reason, he frequently quoted New Learning commentaries whose authors clearly demonstrated a better understanding of the text of the Rituals of Zhou and the details of ancient rituals. Wang Anshi’s Zhouli xinyi and Wang Zhaoyu’s Zhouli xiangjie were sprinkled throughout the Zhouli dingyi. Wang Yuzhi also quoted Chen Xiangdao’s Lishu and Lu Dian’s Liji jie in his fine work. The Zhouli xiangjie demonstrates the same inclusive tendency in its explication of the Imperial Temple and temple rituals. Different from Yi Fu’s Zhouguan zongyi, which focuses on Zhu Xi’s hypothetical model of the Imperial Temple, the Zhouli xiangjie agrees to Wang Anshi’s rectification of temple’s orientation and regards the establishment of a centralized temple in the capital as a means of imposing the government’s influence over its people in various aspects, especially for administrative and enlightening purposes.31 In Wang Yuzhi’s opinion, the positioning of the temple with reference to other significant architectures in the capital—for instance, the imperial palace and the State Altar—is crucial to the promotion of effective governance.32 In one particular section, he expressed his concern about the height of the exterior wall of the temple and the State Altar, which he referred to as wei 壝 rather than dougong—the latter was used by Zhu Xi in the Comprehensive Commentary.33 Wang Yuzhi’s concern about the height of the temple was clearly associated with the temple’s physical location. The Imperial Temples of Northern Song and Southern Song were located in the city centers of both Kaifeng and Linan, nearby a populous area. The considerable height of the temples’ exterior walls, therefore, effectively protected the solemnity of the temples from possible disturbance from the populace.34 Additionally, Wang Yuzhi criticized the practice of making fresh food offerings to the Imperial Temple, which had been performed by most Song emperors. Wang considered the offerings of fresh food (jianxin) to temple ancestors as an inappropriate ritual that should only be held in the main chambers of subsidiary imperial ancestral buildings.35 In his words, the offerings of fresh food in a regular Imperial Temple with both eastern and western chambers contaminate temple rituals. Although Wang cited no contemporary examples in his commentary, he was definitely dissatisfied with the Song court’s use of seasonal food in temple sacrifices. Instead, Wang advocated for the use of “cooked meats” (fan 膰) in temple sacrifices. He basically followed the Yongjia scholar Xue Jixuan’s interpretation of the “cooked meats” as the symbol of the intimate relations between the sacrificer

31. Wang, Zhouli dingyi, 1.4b–5a. 32. Wang, Zhouli dingyi, 1.4b–5a. Throughout the Zhouli dingyi, Wang Yuzhi usually used the term “my humble opinions” (yu an 愚案) to distinguish his own opinions from those of other scholars. 33. Wang, Zhouli dingyi, 1.4b–5a. 34. In the Northern Song, officials sometimes suggested evacuating the population who lived near the Imperial Temple, lest their noise disturb the practice of temple rituals. See, for example, XCB, 143.3459. 35. Wang, Zhouli dingyi, 52.22b–23b.

166

Empowered by Ancestors

and the sacrificed.36 Both Xue Jixuan and Wang Yuzhi made use of the Chinese character shu 熟 in referring to the dual meanings of “cooked” and “intimacy”: fan as a “cooked meat” is used to symbolize the sense of intimacy involved in temple sacrifices. The interpretation of the term fan, as well as its connotative meaning, seems to be an adoption of the New Learning emphasis on the study of written characters, despite the fact that no New Learning commentaries had ever interpreted the term in this way. Apart from the concrete performance of temple rites, Wang Yuzhi was also inclined to adopt controversial temple issues that had frustrated previous scholars and ritual officials. By citing the commentary of another Yongjia scholar Huang Du, Wang argued that the seven-temple configuration of the Imperial Temple was a regular setting that had been interpreted differently by previous Confucians like Zheng Xuan and Wang Su.37 He cited Xue Jixuan’s commentary to articulate the hierarchical nature of the temple’s zhaomu.38 Wang Yuzhi also quoted Chen Xiangdao’s zhaomu argument in full length to demonstrate that the temple’s zhaomu should never be altered, although he mistakenly put it under Lu Dian’s name.39 In general, Wang Yuzhi paid more attention to Yongjia and New Learning commentaries in annotating phrases about temple rituals. While Yi Fu’s Zhouguan zongyi is reminiscent of Zhu Xi’s ritual theories, Wang Yuzhi’s Zhouli dingyi reveals itself more like an inclusive integration of commentaries on ritual practices. Likewise, the anonymous Zhouli jishuo also quotes some New Learning narratives from the commentaries of Wang Anshi, Wang Zhaoyu, and Chen Xiangdao to provide a consistent explanation of temple settings and rites. The author of the Zhouli jishuo even prioritized New Learning interpretations over Zhu Xi’s in annotating some phrases of the Rituals of Zhou.40 Without the intellectual endeavors of New Learning scholars, it is difficult to imagine how the Southern Song learning on the Rituals of Zhou would have played out. A typical comment on the Zhouli jishuo reads, “despite the criticism made by Song scholars on Wang Anshi’s Three New Meanings, Wang Zhaoyu had annotated the New Meaning of the Rituals of Zhou before and Lin Zhiqi had elaborated its ritual scholarship later. The Zhouli jishuo inherits these annotative traditions by quoting the New Meaning of the Rituals of Zhou and continues the transition of the text of Wang’s New Meaning” 蓋安石《三經新義》雖為宋人 36. Wang, Zhouli dingyi, 30.7a–8b. 37. Wang, Zhouli dingyi, 32.6a–7b. Huang Du’s commentary on the Rituals of Zhou is called the Zhouli shuo 周禮 說 (Discussion on the Rituals of Zhou). Ye Shi 葉適 (1150–1223) prefaced the Zhouli shuo and stated its intention to challenge Wang Anshi’s Zhouli xinyi. According to Ye Shi, Huang Du was so diligent to read the Zhouli xinyi that he compared it with other commentaries overnight when he was seventy-five years old, without noticing the sound of his water clock. Ye Shi’s interesting description shows how some Yongjia scholars struggled with the influence of Wang Anshi’s New Learning in the mid-Southern Song. Huang Du, Song Huang xuanxiangong zhouli shuo 宋黃宣獻公周禮說, XXSK, 78: yuanxu 原序.1a–b; 78: kaozheng 考證.1a–3a. 38. Wang, Zhouli dingyi, 32.7a. 39. Wang, Zhouli dingyi, 32.8a–8b. 40. See, for instance, the Zhouli jishuo’s annotation of the phrase “differentiation of tablets and temples based on the zhaomu order” 辨廟祧之昭穆. Zhouli jishuo, 4.39b–44b, especially 4.40a–b.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 167

所攻，而《周官新義》則王昭禹述之于前，林之奇述于後，故此書亦相承援 引，不廢其文也.41 What is embedded in the Zhouli dingyi and the Zhouli jishuo is an inclusive attitude toward various intellectual traditions before, during, and after Zhu Xi’s era. In terms of ritual learning, scholars developed ideas about temple rituals in various ways. Their endeavors facilitated the development of not only ritual scholarship but also standardized ritual norms that were later implemented in the society in both rhetorical and discursive forms.

Adoption of Temple Rituals in the Southern Song: An Emphasis on Orthopraxy Along with the compilation and spread of Southern Song ritualists’ commentaries, especially Zhu Xi’s Comprehensive Commentary, discourses on the Imperial Temple moved toward the practical aspects in the thirteenth century. The Imperial Temple rituals fascinated those local elites who attempted to organize local societies based on standardized ritual norms and related ritual performances. Temple rituals, especially the zhaomu sequence, formed the core of the discussions on ritual norms in the Southern Song society. Scholar-officials and local elites advocated the adoption of temple rituals, including the arrangement of burial grounds, the composition of prefaces to descent group records (zupu xu 族譜序), and the integration of the zhaomu concept in local societies. In fact, some scholar-officials in early to the late Northern Song had already initiated the campaign of rectifying daily life ritual practices based on temple rituals. Sima Guang’s Shuyi 書儀 (Letters of Ritual) explains why the correct practice of four family rites—capping (guan 冠), wedding (hun 婚), funeral (sang 喪), and sacrifices (ji 祭)—was all related to temple rituals in different forms. However, the decline of the tradition of building family shrines among scholar-official families in the Northern Song led to the abolition of temple rituals.42 Even so, as Sima correctly pointed out in the Shuyi, some elements of Song family rituals originated from the Imperial Temple. For example, the ancestral offering board that was used by Song scholar-officials in ancestral sacrifices was an alternative to the spirit tablet in the Imperial Temple.43 Furthermore, Sima argued that making sacrifices in separate chambers had been a shared tradition of the Imperial Temples and private ancestral temples since the Eastern Han.44 The only difference was that seasonal sacrifices 41. Zhouli jishuo, xu.2a. The Siku editors argued that Lin Zhiqi originally followed the scholarship of the Yuanyou (conservative) scholars. But Lin’s annotation on the Rituals of Zhou (Zhouli quanjie 周禮全解) also embraced Wang Anshi’s ritual scholarship. 42. In the Shuyi, Sima Guang repeatedly argued that ancient family rites were originally performed in ancestral temples. See Shuyi, SKQS, 142:2.3a (capping rites); 4.3a (wedding rites); 6.3a (funeral rites); 8.7b (sacrificial rites). Choi Mihwa argues that Sima constructed his ritual system in Shuyi based on a hierarchical conception of official ranks. Choi, Death Rituals and Politics in Northern Song China, chap. 3. 43. Shuyi, 142:7.8b. 44. Shuyi, 142:10.2b.

168

Empowered by Ancestors

were performed in the first months of the four seasons in the Imperial Temples, yet they were performed in the second months in private temples.45 A decade later than Sima Guang’s Shuyi, Cheng Yi came up with his own discourse of funeral rites based on some temple rituals. Not only did Cheng Yi design a blueprint of wooden ancestral tablets for scholar-officials that resembled the spirit tablets used in the Imperial Temple, but he also simplified the procedures of imperial seasonal sacrifices to suit the needs of his contemporaries.46 Most importantly, Cheng Yi stressed the supreme position of the Primal Ancestor in the solstice sacrifice, which was clearly influenced by Wang Anshi’s argument in the 1072 ritual debate over the Imperial Temple. Like Wang Anshi and Yuan Jiang, who emphasized the concept of “original lineage” in determining the Primal Ancestor, Cheng Yi composed a sacrificial text stating that the primary goal of solstice sacrifice was to trace the origin of ancestors.47 Notably, both Sima Guang and Cheng Yi criticized the conventional practice of geomancy in funeral rites, especially in the selection of burial grounds.48 Specifically, Cheng Yi questioned the practice of arranging burial grounds in accordance with the “five pitches of surnames” (wuxing 五姓), in which tombs and graves were arranged in a rhyming scheme of surnames’ consonants according to the five musical pitches of gong 宮, shang 商, jiao 角, zhi 徴, and yu 羽. The five musical pitches echo the five phases of elements (wuxing 五行). Considering this system of five phases as a heterodox practice, Cheng Yi advocated a return to the zhaomu order in burial grounds.49 Despite Cheng Yi’s hostility toward the geomantic practice of phases of elements, he adopted a geomantic method in helping Shao Yong select a burial ground for the latter’s father.50 The negotiation between geomancy and Confucian rites characterized the burial practices of the Northern Song, especially in the northern regions. In the Dilixinshu 地理新書, a geomancy manual compiled in 1057, there are several well-depicted zhaomu diagrams of burial grounds named wuyin zhaomu zang 五音昭穆葬, literally, the burial setting of the five pitches in the zhaomu order.51 These diagrams have combined the Confucian idea of zhaomu and the 45. Shuyi, 142:10.1a. 46. Cheng Yi, “Zuo zhu shi” 作主式 and “jili” 祭禮, Ercheng ji, 627–28. 47. Cheng Yi, “jili,” Ercheng ji, 628. 48. For a brief introduction of Sima Guang and Cheng Yi’s negative attitude toward geomancy, see Patricia Ebrey, “Sung Neo-Confucian Views on Geomancy,” in Meeting of Minds: Intellectual and Religious in East Asian Traditions of Thought, ed. Irene Bloom and Joshua A. Fogel (New York: Columbia University Press, 1997), 75–97; Liu Xiangguang 劉祥光, “Songdai fengshui wenhua de kuozhan” 宋代風水文化的擴展, Taida lishi xuebao 臺大歷史學報 45 (2010.6): 35–38. 49. Cheng Yi, “Zangfa jueyi” 葬法決疑, Ercheng ji, 628. 50. Shao Bowen 邵伯溫, Shaoshi wenjian lu 邵氏聞見錄 (Beijing: Zhonghua shuju, 1983), 20.221. Cheng Yi also attempted to introduce the geomantic idea of Nine Palaces (jiugong 九宮) into his scheme of ideal graveyards. See Ina Asim, “Status Symbol and Insurance Policy: Song Land Deeds for the Afterlife,” in Burial in Song China, ed. Dieter Kuhn (Heidelberg: Edition Forum, 1994), 331–32. 51. Wang Zhu 王洙 (997–1057) et al., Tujie jiaozheng dili xinshu 圖解校正地理新書 (Taibei: Jiwen shuju, 1985), 392–93. The wuyin setting of burial sites is similar to the wuxing arrangement. The Yuan-compiled Dahan

Socialization of the Discourse of Temple Rituals in the Late Southern Song 169

geomantic practice of wuxing. According to the Dilixinshu, the zhaomu sequence regulates the basic setting of burial sites. Furthermore, the Dilixinshu records an auspicious burial arrangement named guanyu zang 貫魚葬,52 in which tombs and graves strictly follow the zhaomu order.53 The influence of geomancy continued in the Southern Song practices of funeral rites. However, the rise of the conception of “descent groups” with common ancestors (zongzu 宗族) breathed new life into the educated elites’ campaign of rectifying family rites.54 Compared with Northern Song scholar-officials, Southern Song elites were more confident in constructing their family rites based on imperial exemplars. Temple rituals constituted a key part of the Southern Song discourses of family rites. During the thirteenth century, several types of writings embraced these discourses: family precepts (jiafan 家範), encyclopedic collections of daily-use rites and other knowledge (riyong leishu 日用類書), and prefaces to descent group records. Scholars have already noted the emergence of family precepts in the Southern Song.55 Most of these precepts list a series of concrete regulations to regulate their family members. A few of these family precepts discuss the intent of family regulations and rites. Lü Zuqian’s 呂祖謙 (1137–1181) Family Precepts in six-juan explicates how the family rites of scholar-officials should symbolize the solidarity of not only a family but also the descent group of the family. Lü specially heeded to the Dazhuan text in the Book of Rites, which provides the most detailed description about the setting of ideal descent groups. Although the Dazhuan text emphasizes that only imperial rulers are qualified to perform the di sacrifice to their ancestors, Lü regarded di as a daily ritual of ancestor worship for all descent groups. Thus, he argued that “the honoring of ancestors precedes the reverence toward descent groups” 蓋尊祖然後能敬宗.56 In this way, Lü Zuqian considered the di sacrifice, which was solely performed in the Imperial Temple, as the ritual intent of family precepts in “consolidating the descent group and regulating its junior members”

yuanling mizangjing 大漢原陵秘葬經 (The Han Secret Manual of Arranging Tombs) records how the Han setting of burial grounds was determined by the particular pitch to which the Han emperors’ surname (Liu) belongs. For an introduction of the practice of wuyin in imperial burial grounds, see Shen Ruiwen 沈睿文, Tangling de buju: kongjian yu jianzhu 唐陵的佈局: 空間與建築 (Beijing: Beijing daxue chubanshe, 2009), 42–43; Ina Asim, “Status Symbol and Insurance Policy,” 333. 52. Literally, the term refers to a burial arrangement that resembles a shoal of fish. 53. Wang, Dili xinshu, 390. The Dahan yuanling mizangjing also mentions guanyu zang and states that the “zhaomu order arranged in the form of guanyu is very auspicious” 昭穆貫魚葬大吉. Shen, Tangling de buju, 84. 54. I adopted the term “descent group” from Patricia Ebrey’s and James Watson’s definitions of various zu-related concepts in their edited volume, Kinship Organization in Late Imperial China, ed. Patricia Ebrey and James Watson (Los Angeles: University of California Press, 1986), 4–9. 55. Patricia Ebrey, Family and Property in Sung China: Yuan Ts’ai’s Precepts for Social Life (Princeton, NJ: Princeton University Press, 1984); Liu, Hui-chen Wang, The Traditional Chinese Clan Rules (Locust Valley, NY: J. J. Augustin, 1959). 56. Lü Zuqian, “Jia fan,” Donglai Lütaishi bieji 東萊呂太史別集. Lü Zuqian quan ji, 1:1.281. In other places Lü Zuqian reiterated that the reverence toward the same descent group conveys the meaning as honoring the common ancestor. “Jia fan,” 1.283.

170

Empowered by Ancestors

睦族治子弟.57 Lü further argued that a descent group, once consolidated, would proliferate naturally. Reciprocally, the proliferation of a descent group would dignify the group’s private temple.58 Along the same vein, if the descent group dispersed, then its temple would diminish. In this light, Lü traced the origin of descent groups to the ancient setting of the Imperial Temple: Ancient rituals were all designated in terms of “descent groups.” Hence, Boyi was appointed by (the sage-king Shun) as the director of ancestral sacrifices. The Rituals of Zhou records ritual practitioners who ranked from the offices of Chief Minister and Vice Minister in the Spring Bureau to the sacrificial offices of the royal domain. The very word “descent group” refers to “temple.” Ritual roots in the regulations reflect one’s personal affection to his relatives; these regulations would not be integrated without ancestral temples. Therefore, the differentiation of surnames and the unification of descent groups are all based on the establishment of ancestral temples.59 古之典禮者，皆以「宗」名之。故伯夷作秩宗，《周官》有宗伯，下及乎都 家，皆有宗人。宗者，廟也。禮始於親親之法，非廟不統。所以别姓、收 族，無不出於祖廟。

By correlating the Zhou-derived ideas of “descent group” and “temple” to the practice of establishing ancestral buildings in his time—mostly in the form of ancestral halls, Lü Zuqian rendered the Imperial Temple a social function outside its imperial context. In his opinions, the Imperial Temple has nothing to do with “differentiating surnames” (bie xing 別姓), considering the fact that all imperial clansmen share the same surname. In the later section of the Family Precepts, Lü Zuqian explained why family sacrifices to ancestors resulted from the Imperial Temple rituals. Lü quoted the Wangzhi phrase to state that Song-educated elites, equivalent to Zhou junior officers, should construct their own family “temples” or “shrines.” He claimed that the idea of family temples or shrines was coined by Daoxue masters like Cheng Yi and Zhang Zai.60 He also pointed out the two fundamental elements in family temples, namely the alternative arrangement of ancestors in zhao and mu designations and the use of wooden tablets in symbolizing ancestors.61 In fact, Lü Zuqian’s Family Precepts and the Song court’s regulation of temple rituals shared some crucial steps of ancestral sacrifices, such as the presacrifice fasting, the preparation and arrangement of sacrificial utensils, the offerings of cooked food, and the presence of assistant practitioners (zhuji 助祭). Certainly, Lü Zuqian altered some ritual details to distinguish family sacrifices from imperial sacrifices in terms of ritual specifications. For the sake of clarity, I tabulated some major differences in Table 7.1 (see p. 171): 57. Lü, “Jia fan,” 1.281. 58. Lü, “Jia fan,” 1.284. 59. Lü, “Jia fan,” 1.294. According to Jia Gongyan, the term “dujia” 都家 refers to three different levels of fiefs in the royal domain, respectively, the dadu 大都, the xiaodu 小都, and the jiayi 家邑. Zhouli zhushu, TSZSSSJ, 2:31.303. 60. Lü, “Jia fan,” 4.347. 61. Lü, “Jia fan,” 4.348.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 171 Table 7.1: A Comparison between Seasonal Sacrifices in the Family Precepts of the Lü Family and the Official Ritual Code Zhenghe wuli xinyi Family Precepts of the Lü Family

Official ritual code in the Zhenghe wuli xinyi

Dates of sacrifices*

Second month of the season

First month of the season

Pre-sacrifice fasting†

Fasting lasted for two days. The head of the family would fast in the family temple one day prior to the day of sacrifice.

Fasting lasted for ten days. The emperor would fast in the Imperial Temple one day prior to the day of sacrifice.

Offerings‡

Fruits, pickled vegetables, steamed buns, cooked rice, fish meat, and rice in thickened soup.

Raw meat of beef, pork, and lamb; fish meat and crop offerings. All served in ritual utensils (mostly the tripod vessel of ding 鼎 and the small baskets and containers of bian 籩, dou 豆, gui 簠, and fu 簋).

Assistant practitioners§ Brothers of the head of the family and his wife.

Clansmen, selected civil and military officials, and ritual officials from the Court of Imperial Sacrifices and other ceremonial departments

Notes: * In most ritual manuals, the first and second months of a season were called mengyue 孟 月 and zhongyue 仲月, respectively. Lü, “Jia fan,” 4.348; Zhenghe wuli xinyi, 647:102.1b. † Lü, “Jia fan,” 4.352; Zhenghe wuli xinyi, 647:102.2a–3a. ‡ Lü, “Jia fan,” 4.351; Zhenghe wuli xinyi, 647:104.1a–2a. § The wife of the head of the family is in charge of examining and cleansing ritual utensils before the day of sacrifice. Lü, “Jia fan,” 4.349. His brothers make the second and third offerings in the rite of “three offerings” 三獻 on the day of sacrifice. Lü, “Jia fan,” 4.350–51. There are other male family members and ritual practitioners (zhishi zhe 執事者) who help the head of the family to prepare for the sacrifice. However, Lü Zuqian failed to reveal their relations with the head of the family. Lü, “Jia fan,” 4.3a; 4.352–55. There is a complete juan in the official ritual code dealing with the responsibilities of assistant practitioners in temple sacrifices. See Zhenghe wuli xinyi, 647:105.1a–13b.

172

Empowered by Ancestors

Apparently, the family sacrifices in Lü Zuqian’s Family Precepts are simpler than the seasonal sacrifices in the Imperial Temple. Although Lü was unable to revise the Family Precepts before his death,62 he successfully constructed a consistent framework of ancestral sacrifices from the supreme di sacrifice to family sacrifices of educated elites, which accorded with the hierarchical system of the Imperial Temple in the Wangzhi. Lü Zuqian’s Family Precepts became the source of inspiration to relevant discourse of family ancestral buildings that appeared in later family precepts and liturgical manuals in the Southern Song. Zhu Xi’s Family Rituals proves to be the most influential one among these liturgical manuals. Patricia Ebrey has argued that Zhu Xi invented some rites and liturgical details in the Family Rituals.63 In fact, some of these rites were composed by Zhu based on his conception of the Imperial Temple. The obvious example is the spatial arrangement of ancestral tablets. As Zhu considered the ancestral hall as a replacement of family temples, he arranged the tablets therein from the west to the east, in alignment with his visualization of the tablets of imperial ancestors in the main chamber of the Imperial Temple.64 As two influential scholars of their times, both Lü Zuqian and Zhu Xi adopted in their regulations of family rites some rituals that originated from the Imperial Temple. Their discourse of family ancestral temples continued the ritual campaign of Sima Guang and Cheng Yi, advocating for a return to Confucian norms of sacrificial rites among both educated elites and commoners. Established regulations about the Imperial Temple and temple rituals serve as liturgical models to follow in Lü’s and Zhu’s ritual writings. Nevertheless, the writings of Daoxue scholars were targeted at a limited audience of scholar-officials and local literati. Less educated commoners had no access to Sima Guang’s Shuyi or Lü Zuqian’s Family Precepts, not to mention illiterate peasants.65 How could the revised sacrificial rites in the Daoxue writings reach a broader audience? The rise of encyclopedic collections of daily rites and knowledge in the thirteenth century offers a solution. These encyclopedic collections were designed by local educated elites as sourcebooks for commoners to learn the basic regulations of daily rites. Therefore, the compilers of these encyclopedic collections usually simplified the liturgical procedures of complicated Confucian rituals by providing clear illustrations. As most compilers were affiliated with the Daoxue circle, they introduced many Daoxue ideas in their encyclopedic collections.66 In the case of sacrificial rites, a local literatus Chen Yuanjing 陳元靚 integrated Cheng Yi’s ideas 62. According to Zhu Xi, the part of family sacrifices in Lü Zuqian’s Family Precepts is less complete than other parts. See Lü, “Jia fan,” 4.356–57. 63. Ebrey, Confucianism and Family Rituals in Imperial China, 106–7. 64. Zhu, “Yaomiao yizhuang,” Huian ji, 15.223l; also see Figure 6.3. 65. Evelyn Rawski’s early work has marshalled crucial materials about how popular literacy had arisen critically in Song China, especially the twelfth century. See Rawski, Education and Popular Literacy in China (Ann Arbor: University of Michigan Press, 1979). 66. For example, Zhu Mu 祝穆, the compiler of the encyclopedic collection Gujin shiwen leiju 古今事文類聚, was Zhu Xi’s disciple.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 173

of the scale of coffin and the role of the Primal Ancestor in the solstice sacrifice into his edited encyclopedia, the Shilin guangji  事林廣記 (Comprehensive Record of Affairs).67 A later Japanese edition of the Shilin guangji also included Cheng Yi’s illustration of the zhaomu arrangement of burial grounds.68 In the inceptive edition of the Shilin guangji, there is a diagram called “the diagram of seasonal sacrifices in the main chamber” (zhengqin shiji zhitu 正寢 時祭之圖). The diagram finds no precedents in previous Daoxue writings.69 This diagram, shown in Illustration 7.1 (see p.  174), which was portrayed by Chen Yuanjing himself, illustrates the placement of ancestral tablets from west to east in the order of great-great-grandparents, great-grandparents, grandparents, and parents. Chen admitted that the diagram was inspired by Zhu Xi’s Family Rituals. In fact, it clearly embodied Zhu’s conception of the spatial arrangement of the Imperial Temple’s main chamber. Other encyclopedic collections in the thirteenth century also quote ritual norms from Daoxue masters in constructing their own narratives of family sacrifices and sacrificial rites. None of them, however, is comparable to the Shilin guangji in terms of readability. Encyclopedic collections such as Gujin hebi shilei beiyao 古今合璧 事類備要 and Gujin shiwen leiju are cluttered with canonical quotes and historical precedents under different categories and sub-categories. They expect readers to possess adequate knowledge in the studies of Classics, history, and literary. A literate villager who reads the section of “Family Temple” in the Gujin hebi shilei beiyao would be baffled by the Tang exemplars of an Imperial Temple with different references from literary writings.70 However, he may find it easy to follow the liturgical procedures and spatial arrangement of seasonal sacrifices by looking at the detailed illustrations in the Shilin guangji. In practice, works like Shilin guangji aided the spread of Daoxue sacrificial rites to a broader audience on the level of “lower-order” ritual practices. The other routine of the Daoxue transition of temple rituals from the upper stratum downward emphasized the extension of the zhaomu principle from the Imperial Temple to other social applications outside the palace. In the Northern Song, scholar-officials like Lü Dalin 呂大臨 (1044–1091) had already mentioned the social application of zhaomu in differentiating familial relations and household identities.71 In Lü’s words, the zhaomu sequence represented the Confucian norm

67. Chen Yuanjing, Zuantu zengxin qunshu leiyao shilin guangji 纂圖增新群書類要事林廣記 (Beijing: Zhonghua shuju, 1999), 2b.51. 68. Chen Yuanjing, Xinbian qunshu leiyao shilin guangji 新編群書類要事林廣記 (Beijing: Zhonghua shuju, 1999), 9c.498. The original illustration made by Cheng Yi can be found in his “Zang shuo” 葬說, Ercheng ji, 628. The Japanese edition was reprinted based on a Yuan-compiled edition of 1325. For details of this Yuan edition and other Japanese editions, see Kenji Morita 森田憲司, “Guanyu zai riben de shilin guangji zhuben” 關於在日本的事林廣記諸本, Xinbian qunshu leiyao shilin guangji, 566–72. 69. Chen, Zuantu zengxin qunshu leiyao shilin guangji, 2b.51. 70. Xie Weixin 謝維新, Gujin hebi shilei beiyao: qianji 前集, SKQS, 939:69.8a–10a. 71. Wei, LJJS, 129.18b–19b.

174

Empowered by Ancestors

Illustration 7.1: A Diagram of Seasonal Sacrifices in a Thirteenth-Century Encyclopedia

Source: Chen Yuanjing 陳元靚, Zuantu zengxin qunshu leiyao shilin guangji 纂圖增新群書 類要事林廣記, 2b.51.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 175

of “distinguishing kin relations” (bierenlun 別人倫).72 Fan Zhongyan claimed that a correct zhaomu would help members of a descent group to “commemorate their ancestors” 追思祖宗.73 Su Xun 蘇洵 (1009–1066) stated that the zhaomu sequence formulated both the “great descent group” (dazong 大宗) and the “lesser descent group” (xiaozong 小宗) in the selection of sons from the branches (fang 房) of a descent group to continue the lineage.74 However, neither of them discussed the concrete ways by which kin relations and descent groups were associated with zhaomu. Further descriptions on the social applications of zhaomu only proliferated in the late Southern Song, when the required social conditions were ready on a regional level. Along with the rising local interest in descent groups, a new demand for the production and rectification of collective group rituals arose among local elites. Under this context, Confucian scholars started to consider zhaomu as a ritual tool to stabilize their conceptions of the social system of descent groups. In transferring the zhaomu sequence conventionally used in the Imperial Temple to a ritual of descent groups, Southern Song scholars had to make some theoretical revisions. The first concerned the “philosophization” of the zhaomu concept. The word philosophization here refers to a metaphysical conception of zhaomu as a transcendental concept that went beyond the context of the Imperial Temple. Zheng Genglao 鄭耕老 (1108–1172), a contemporary of Zhu Xi,75 perceived zhaomu in this philosophical way. He argued that the core idea of zhaomu was the “order” (ban 班) of zhao and mu designations in the setting of descent groups.76 According to Zheng, zhaomu represents a fixed ritual order in every ancestral space. This zhaomu argument apparently reiterates Zhu Xi’s definition of zhao and mu as fixed location markers in the Imperial Temple. It is possible that Zheng had read Zhu’s ritual writings and drawn some inspiration from them. Nevertheless, compared to Zhu Xi, Zheng put more emphasis on the universal applicability of zhaomu and the metaphysical ground upon which this ritual order was established. In a philosophical way, Zheng reconciled different records of zhaomu in the Book of Rites. On the one hand, Zheng saw the Wangzhi description of the temple system of the Son of Heaven as the “motion” (dong 動) of zhaomu, since it involves the ceaseless abolishment and displacement of ancestral temples. On the other hand, he argued that the regular temple system, the dan altar, and the shan yard, which are recorded in Jifa, reflects 72. Wei, LJJS, 129.19a. 73. Fan Zhongyan, “Xu jiapu xu” 續家譜序, Fanwenzheng ji 范文正集, SKQS, 1089: bubian 補編 1.23b. 74. Su Xun quoted from the Great Treatise chapter of the Book of Rites to explain the difference between the zhaomu of the “great descent group which the group head’s tablet will not be removed for a hundred generations” 百世不遷之宗 and that of the “lesser descent group which the group head’s tablet will be removed after five generations” 五世則遷之宗. See Su Xun, “Zupu houlu” 族譜後錄 and “Dazong pufa” 大宗譜法, Jiayou ji 嘉祐集, SBCK, 51:14.49–52, especially 50. 75. ZBSYXA, 4.11a. Ye Shi composed an epitaph for Zheng Genglao. Ye praised Zheng for “elucidating the Way of the sages and walking on the right track of Mean and Correctness” 推明聖人之道，歸於中正不偏. Ye Shi, “Fengyilang zhenggong muzhiming” 奉議郎鄭公墓誌銘, Yeshi ji 葉適集 (Beijing: Zhonghua shujiu, 1961), 15.280. 76. Wei, LJJS, 117:30.39a–b.

176

Empowered by Ancestors

the “constancy” (chang 常) of zhaomu’s “static essence” (dingti 定體).77 Therefore, Zheng understood zhaomu not only as a genealogical order but also an embodiment of the self-perpetuating cycle of stillness and motion through the practice of ancestral worship. As the zhaomu sequence incessantly absorbs new ancestors and displaces recent ancestors with the new ones, it conveys a ritual meaning of eternity. Zheng’s transcendental reading of zhaomu continues by stating that zhaomu could be interpreted as a symbolization of the Heaven and the two qi 氣 (pneuma), the energetic yang and the constant yin.78 As zhao ancestors are located at the left-odd positions (qi 奇, 1, 3, 5, .  .  .  ), they belong to the Heavenly yang order. Correspondingly, mu ancestors, who are located on the right-even positions (ou 偶, 2, 4, 6, . . . ), belong to the Earthly yin order.79 According to Zheng, since the parallel configuration of the odd yang order of zhao and the even yin order of mu are unalterable, the homogeneity among ancestors of the same order of zhao and mu is self-evident. The grandson, instead of his father, can serve as his grandfather’s corpse medium in funeral rites and the xia sacrifices because the grandson and the grandfather belong to the same order (tongban 同班). The grandfather of the zhao order takes the zhao grandson as his medium; the grandfather of the mu order takes the mu grandson as his medium. Only when the ancestor’s spirit approaches the medium of the same order can it be effortlessly attached to that medium.80 Concerning the formulation of the genealogy of families and descent groups, the conception of zhao and mu as transcendental ideas offers a theoretical base for the extension of the zhaomu discourse to different social contexts. However, it is necessary to figure out how the idea of zhaomu coped with the social structure of descent groups on the practical level. This task was accomplished by the Daoxue scholars and officials who championed the construction of family temples. Apart from Zhu Xi’s promotion of ancestral halls in the Family Rituals, Wei Liaoweng 魏 了翁 (1178–1237) as a Daoxue advocate emphasized how the ancient concept of descent groups was derived from the ritual ordering of communities under the same surnames. In this light, zhaomu serves as a ritual apparatus to regulate the internal relations within the same descent group based on the Confucian differentiation of the “major surnames” (zhengxing 正姓) and the “minor surnames” (shuxing 庶姓) of a descent group.81 In an epitaph, Wei Liaoweng pinpointed the importance of zhaomu in ordering temple rituals:

77. Wei, LJJS, 117:30.39b. 78. Some modern Chinese scholars, such as Xu Zibin 許子濱, tended to posit the zhaomu sequence in Chinese metaphysical terms. Xu, “lun zhaomu zhi mingming yiyi” 論昭穆之命名意義, Hanxue yanjiu 漢學研究 25.2 (2001): 329–46. Zheng Genglao’s argument pioneered this metaphysical reading of zhaomu. Zheng’s reading of zhaomu may also illustrate the Daoist influence on the “neo-Confucian” concepts during the Southern Song Daoxue movement. 79. Wei, LJJS, 117:30.40b–41a. 80. Wei, LJJS, 117:30.41a. 81. The two concepts of “major surname” and “minor surname” are based in Zheng Xuan’s commentary on the Dazhuan chapter of the Book of Rites. See Liji zhushu, TSZSSSJ, 2:34.394.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 177 When there is a ritual affair in the temple, all descendants of the descent group under the same surname should be ordered according to the zhaomu sequence. For the rites of offering ritual wine to one another and the bestowal of honorable titles, zhaomu should be employed to differentiate the order of participants. Sisters and other female relatives who participate in these rites should be ordered according to seniority. Zhaomu as a ritual order not only lays the foundation for the descent group and emblematizes its lineage, but it also resolves what is doubtful and brings to light what is abstruse.82 凡有事（於）廟，廟之子姓，各以昭穆為序。旅酬賜爵，昭穆各齒；内賓宗 婦，廟序以倫。盖不惟奠系明統，亦以別嫌明微。

Wei Liangweng cited the last phrase of the above passage from the Liyun 禮運 chapter of the Book of Rites. The original sentence reads: “Ritual is a great instrument in the hands of rulers. It is by ritual that the rulers resolve what is doubtful and bring to light what is abstruse” 禮者，君之大柄也，所以別嫌明微.83 By citing this phrase, Wei Liaoweng endowed zhaomu with the instrumental function in constructing and consolidating the conception of descent groups. Recognizing the importance of zhaomu in the conceptual construction of descent groups, Southern Song and early Yuan scholars frequently used the term zhaomu in their prefaces to descent group records. Generally, they indicated that the primary function of descent group records is to “order descent groups according to the zhaomu sequence” (xu zhaomu 敍昭穆). Xue Jixuan claimed that descent group records should be compiled to “order the zhaomu sequence of group members and differentiate group branches” 是故序昭穆，别宗祧.84 Without zhaomu to differentiate surnames, Xue further deduced that different clans could not be distinguished from one another.85 Huang Zhongyuan 黃仲元 (1231–1312), a Fujianese who witnessed the fall of the Southern Song dynasty and named the ancestral hall of his descent group as “Sijing tang” 思敬堂, traced the zhaomu of the Huang group back to the thirteenth generation.86 Chen Lü 陳旅 (1287–1342), a Yuan scholar, advocated the composition of descent group records in his time, because these records 82. Wei Liaoweng, “Andejun jiedushi zeng shaobao junwang Zhaogong xiguan shendaobei” 安徳軍節度使贈少保 郡王趙公希錧神道碑, Heshan ji 鶴山集, QSW, 311:7111.76. According to a conversation between Confucius and his disciple Zhengzi, the rite of lü chou 旅酬 [offering ritual wine] is not adopted in the mourning period after the death of a family member. Instead, it is performed in the form of a circulation of wine between the host of the family and his relatives in the post-mourning sacrifices. For a detailed explanation of this rite, see Liji zhushu, TSZSSSJ, 2:18.227–28, especially Kong Yingda’s sub-commentary on the related text. The term naibin 内賓 refers to female participants in the family rituals. See Jia Gongyan, Yili zhushu, TSZSSSJ, 2:45.365. 83. Liji zhushu, TSZSSSJ, 2:21.264. 84. Xue Jixuan, “Jiashi jiapu xu” 賈氏家譜序, Xue Jixuan ji 薛季宣集 (Shanghai: Shanghai shehui kexue chubanshe, 2003), 30.448. 85. Chen Gao 陳高 (1314–1366), a Yuan Confucian, argued that the compilation of descent group records aims to “differentiate branches within descent groups, order the ancestors in terms of zhaomu, clarify familial seniority, and distinguish different levels of intimacy among relatives” 别宗支，叙昭穆，定長幼，辨親疎也. Chen Gao, “Zupu xu” 族譜序, Buxi zhouyuji 不繫舟漁集, SKQS, 1216:10.1a. 86. Huang Zhongyuan, “Zuci sijingtang ji” 族祠思敬堂記, Siru ji 四如集, SKQS, 1188:1.4b–5a; also see Chen Gao, “Wushi shipu xu” 吳氏世譜序, Buxi zhouyuji, SKQS, 1216:10.7b.

178

Empowered by Ancestors

helped dispel the confusion of generations concerning descent groups with multiple branches. According to these scholars, the clarification of a correct zhaomu sequence among group members necessitates the composition of descent group records.87 Furthermore, some scholars argued that zhaomu emblematizes the prosperity of descent groups, regardless of their social and economic conditions. As the Yuan Confucian Li Cun 李存 (1281–1354) remarked in his afterword to a local clan’s descent group record: “What makes a person poor, depraved, rich, and elegant is destiny; what maintains the endless continuation of his group’s zhaomu is integrity” 夫貧賤富貴者，勢也；而昭昭穆穆雖百世不可絶者，義也.88 According to Li Cun, the zhaomu order in descent group records reveals the fact that all group members come from the same ancestor. Therefore, wealthy members in a descent group should not discriminate against the poor members, considering their shared origin of lineage.89 Southern Song and Yuan scholars attached considerable importance to the documentation of the zhaomu of group members. Sometimes they even regarded zhaomu as the only reason for the compilation of descent group records. In writing a preface to a Jiangxi descent group who prospered at the county Lulin 廬陵, the local scholar Wang Li 王禮 (1314–1389) argued that the meaning of “a descent group record is to clarify the zhaomu order” 譜在於明昭穆之次.90 He further stated that his preface to the descent group should not be the key focus because it was only composed to capture the endeavor made by this local descent group in revering their ancestors. In Wang’s opinion, the endeavor itself deserved more attention and respect from the Yuan-educated elites of Lulin. In addition, those educated elites who failed to compile a descent group record were usually described by Southern Song and Yuan scholars as “worse than birds and beasts” 禽獸不若也—a saying that was commonly attributed to Ouyang Xiu, who was also believed to be the inventor of the format of descent group records.91 In practice, the integration of the zhaomu idea into the descent group was an enduring problem for many local clans in Southern Song and Yuan. Supposedly, the zhaomu system is premised on the hereditary purity of a group under the same surname. In reality, members in a descent group usually came from different origins. Therefore, it was the exchange of memories between the members of descent groups that contributed to a unified conception of the descent group and related group 87. Chen Lü, “Dingshi shipu xu” 丁氏世譜序, Anyatang ji 安雅堂集, SKQS, 1213:4.19a–b. 88. Li Cun, “Ti Zhangshi zupu hou” 題章氏族譜後, Sian ji 俟庵集, SKQS, 1213:27.2a. 89. Li Cun, “Ti Zhangshi zupu hou,” 27.2b. 90. Wang Li, “Xialiu liushi zupu xu” 夏𣲖劉氏族譜序, Linyuan wenji 麟原文集, SKQS, 1220: hou ji 後集 1.9b. 91. For example, see Xie Weixin, Gujin hebi shilei beiyao: xuji 續集, 940:2.3b–4a; Zhu Mu, Gujin shiwen leiju: houji, 926:1.25b; Dai Liang 戴良, “Zhangshi jiacheng xu” 章氏家乘續, Jiuling shanfang ji 九靈山房集, SKQS, 1219:6.11a–b. I could not find in Ouyang’s collected works this aggressive charge toward those educated elites who failed to compile a descent group record. As far as I know, Wang Decheng 王得臣 (1034–1116) first mentioned this critical comment and put it in Ouyang’s mouth. Wang, Zhushi 麈史, in Song Yuan biji congshu 宋元筆記叢書 (Shanghai: Shanghai guji chubanshe, 1986), 73–74.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 179

records. The composition of a monolithic group record with heterogeneous clans is called tongpu 通譜.92 From Southern Song to Yuan, the practice of tongpu reached new heights and sparked the proliferation of local descent groups whose members came from different clans but shared the constructed memories of the common first ancestors.93 It thus became almost impossible for post-Song descent groups under the same surname to accurately clarify all the zhaomu orders of their group members’ ancestors. Due to this difficulty, Song and Yuan scholars often employed the word “deduction” (tui 推) when they mentioned zhaomu in their prefaces to descent group records. It is only by deductive reasoning that the compilers of descent group records could identify the zhaomu of the members of different branches in a constructed descent group. Fang Dazong 方大琮 (1183–1247), a Southern Song scholar, had already noted this problem in his preface to his own descent group record. As he said, the collection and compilation of scattered zhaomu records of the Fang clan aimed to “comprehend the origin and the development of the descent group by scrutinizing the records back and forth, as well as to count the number of descendants by putting them in order” 推而上下之，則知源流之脉絡，從而 列之，則知子孫之多寡.94 In constructing the history of the three divisions of the Fang clan who lived in the Putian 莆田 region of Fujian, Fang Dazong admitted that it was impossible to examine the zhaomu of every member of the Fang group. However, Fang Dazong claimed that all the three Fang divisions originated from the same ancestor Fang Hong 方紘, a Han official who suffered under Wang Mang’s governance.95 Instead of making a clear clarification of the earlier zhaomu records about previous Fang groups, Fang Dazong was more concerned about the proliferation of the newly constructed Fang descent group and a foreseeable recompilation of the group record in the future.96 By compiling descent group records, Fang Dazong believed that the zhaomu of the Fang group would extend forever. Given the heterogeneity of most descent groups in blood origins, some Song and Yuan descent groups attempted to establish a well-structured zhaomu sequence merely based on seniority. The Southern Song scholar Wang Yan 王炎 said that as 92. For a succinct description of the tongpu practice since the third century, see Gu Yanwu, Rizhilu 日知錄 (Shanghai: Shanghai guji chubanshe, 2012), 23.878–81. The difference between the two concepts of “descent group” and “clan” is difficult to recognize, owing to the confusing use of the Chinese characters zongzu 宗 族 (descent group or “lineage” in the context of late imperial China) and zu 族 (clan) in relevant sources. For a classic differentiation of the two concepts, Maurice Freedman, Chinese Lineage and Society: Fukien and Kwangtung (London: Humanities Press, 1966), 169. Also see Patricia Ebrey and James Watson, Kinship Organization in Late Imperial China, 5–6. 93. A lot of research has been done on the studies of descent groups and especially the more strictly defined “lineage” system that was based on shared assets. To list a few representative works: Maurice Freedman, Lineage Organization in Southeastern China (London: Athlone, 1958); Patricia Ebrey and James Watson’s edited volume, Kinship Organization in Late Imperial China; David Faure, Emperor and Ancestor; Chang Jianhua 常建華, Mingdai zongzu yanjiu 明代宗族研究 (Shanghai: Shanghai renmin chubanshe, 2005); Timothy Brook, “Funerary Ritual and the Building of Lineages in Late Imperial China,” Harvard Journal of Asiatic Studies 49.2 (1989): 465–99. 94. Fang Dazong, “Fangshi zupu xu” 方氏族譜序, Tiean ji 鐵菴集, SKQS, 1178:31.4b. 95. Fang Dazong, “Shu putian Fang sanpai juzu” 述莆田方三派聚族, Tiean ji, 32.3a. 96. Fang Dazong, “Fangshi zupu xu,” 31.4b.

180

Empowered by Ancestors

members of a descent group shared a common ancestor, the group members should be arranged according to a fixed order of zhao and mu designations.97 According to Wang Yan, the zhaomu of a descent group should follow the imperial zhaomu by adopting an ordered pattern that designates ancestors of two succeeding generations as zhao and mu separately. In other words, if a father in the descent group is designated as a zhao, then his son must be a mu. However, this genealogical sequence of zhaomu can easily be disturbed by the existence of multiple branches within a single descent group. The popular practice of adoption between male members of different branches (guo fang 過房) created serious disruption to the zhaomu of the same descent group. Hu Ciyan 胡次焱, a thirteenth-century literatus who experienced the Song-Yuan transition, discussed how the customs of adoption disrupted the genealogical zhaomu sequence within descent groups. Hu illustrated with three different social practices of adoption. In the first case, the head of the branch takes his grandnephew (zhisun 姪孫) as the adopted son. In this case, the adopted son’s elder brother has to call his brother as an “uncle.” In the second case, the head of the branch takes his great-grandnephew (zhizengsun 姪曾孫) as the adopted son. This case, in Hu Ciyan’s opinion, is even more absurd as the adopted son’s biological father has to call his son as an “uncle.” In the third case, the head of the branch takes his younger brother as the adopted son. As a result, the adopted son’s younger brother has to call his brother as a “nephew.” Hu Ciyan understood that the practice of adoption could help stabilize the descent group by enhancing the emotional ties between different branches. Nonetheless, he severely criticized the practice of crossgeneration adoption, because it disturbed the genealogical zhaomu within a descent group by elevating the junior generations to the superior, and vice versa.98 Apart from descent group records, Southern Song and Yuan scholars also advocated for a rectification of zhaomu in general clan rituals, especially in funeral rites. Since the mid-twelfth century, the practice of “clan cemetery burial” (zuzang 族葬) had gained popularity among elite families. Recent archaeological excavations reveal that the burial grounds of some descent groups and elite families in the Southern Song were arranged based on a strict zhaomu order. For example, Lü Zuqian’s clan burial ground in Mount Mingzhao 明招山 (modern Wuyi County 武 義, Zhejiang) included five generations of family members. According to the burial diagram of the Lü family, for the burial ground of each generation it strictly obeyed the zhaomu order.99 The excavated family cemetery of the Lü Zuqian corroborates Fang Dazong’s description of Lü Zuqian’s family cemetery: “Zhao and mu designations followed each other in every generation and lasted for over forty generations. Lü Zuqian sincerely adopted the practice of clan cemetery that is recorded in the 97. Wang Yan, “zongzi lun” 宗子論, Shuangxi leigao 雙溪類稾, SKQS, 1155:24a–b. 98. Hu Ciyan, “Lun guofang” 論過房, Meiyan wenji 梅巖文集, SKQS, 1188:5.4b–5b. 99. Zheng Jiali 鄭嘉勵, “Mingzhaoshan chutu de Nansong Lü Zuqian jiazu muzhi” 明招山出土的南宋呂祖謙 家族墓志, Tang Song lishi pinglun 唐宋歷史評論, ed. Bao Weimin 包偉民 and Liu Houbin 劉後濱 (Beijing: Shehui wenxian kexue chubanshe, 2015), 1:186–215.

Socialization of the Discourse of Temple Rituals in the Late Southern Song 181

Rituals of Zhou.” 每代之中，昭穆相從，凡四十餘，《周官》族墳墓之說，吕 氏深有取焉.100 I doubt whether Fang Dazong had a chance to access the burial diagram of the Lü family, since the diagram was supposedly not open for public view.101 Nonetheless, his description of Lü Zuqian’s family cemetery reflects the idea of regarding zhaomu as the necessary element for a decent clan burial.102 A typical layout of Southern Song and Yuan clan burial would arrange zhao and mu ancestors successively and alternately, in the same way as the Song court arranged imperial ancestors in the Imperial Temple.103 Eventually, the political zhaomu in the temple evolved into the social zhaomu that emerged in burial grounds, descent group records, and the prospering family rituals after the mid-twelfth century.

Concluding Remarks From the Imperial Temple to descent group records, various groups of elites utilized temple rituals in demarcating the political and cultural spaces between different social classes. During the Northern Song, discourses of temple rituals were still restricted within the imperial domain and correlated mostly with the state narrative of ritual orthodoxy. However, the tension between meritocracy and heredity, which characterized the Northern Song ritual debates on the Imperial Temple, was attenuated by the diversified understanding of temple rituals in various Southern Song and Yuan sources. Although Southern Song and Yuan scholars shared with their Northern Song predecessors the same rhetoric in reviving ancient rituals, some important changes occurred after the twelfth century. Southern Song scholars, especially those who were identified with Daoxue circles, considered moral cultivation as the key to the benefits of the whole society and the correct practice of ancestral rituals as a shortcut to the perfection of morality. Compared to Northern Song scholars, Daoxue scholars in Southern Song were more interested in the moralizing effect of ancestral rituals. Therefore, they highlighted the enlightening power of rectified ancestral rituals in consolidating various social organizations on the local level, especially the burgeoning system of descent groups in South China. Consequently, Southern Song discourses on temple rituals demonstrated a shift of focus from state to society, and from theoretical disputes to liturgical practices. “Lower-order” ritual practices were emphasized to cope with contemporary social 100. Fang Dazong, “Ci Fangguang libu ji zhufen” 辭方廣禮部及諸墳, Tiean ji, 33.6b–7a. 101. Family burial diagrams were more open for public view in later periods. Yuan local elites sometimes invited celebrated scholars to write epilogues for the burial diagrams of their families and descent groups. See, for example, Xie Yingfang 謝應芳, “Ba cuzang tu” 跋族葬圖, Guichao gao 龜巢稿, SKQS, 1218:14.6a–b. 102. The Northern Song scholar-official Su Song named the “clan burial” based on zhaomu as “a so-called decent practice of funeral rites” in writing his epitaph for a local official 餘皆昭穆相從，所謂以禮葬也. However, Su did not explain in what sense the practice is “decent.” Su Song, “Yingzhou wanshouxianling Zhangjun muzhiming” 潁州萬壽縣令張君墓誌詺, Suweigong wenji, 58.12b. 103. See, for example, the Yuan scholar Li Qi’s 李祁 record of the burial arrangement of a local family, in which the great-grandparents, grandparents, and parents of the family were alternately designated as zhao and mu ancestors. Li Qi, “Xueshi shifen ji” 薛氏世墳記, Yunyang ji 雲陽集, SKQS, 1219:7.21b–22a.

182

Empowered by Ancestors

issues. In a variety of social texts, including geomancy manuals, family precepts, encyclopedic collections of daily-use rites, and prefaces to descent group records, Southern Song scholars measured temple rituals in the form of family rites on the social level. The shift in the conception of zhaomu exemplified this new measurement. Primarily, Northern Song ritual officials perceived zhaomu as a core element of imperial ancestral rituals that were closely associated with the Imperial Temple.104 However, Southern Song scholars interpreted zhaomu from a broader perspective and bestowed upon zhaomu a regulatory function of maintaining a sense of connectedness among families and members of descent groups. The shift of focus from imperial court to social organizations reflects how Southern Song scholars recognized and constructed their identities as educated elites through the advocacy of specific ritual practices. Along with the rectification of temple rituals, Southern Song and Yuan scholars legitimized some group rituals in descent groups, thereby paving the way for the further implementation of these rituals in late imperial China.

104. Northern Song scholars put this in an explicit way. See Yang Jie, “Dixia hezhengweixu yi” 禘祫合正位 序議, Wuweiji xiaojian, 15.505–6; “Zeng Zhao Xingzhuang” 曾肇行狀, Qufu ji, 1011:4.17a; Yu Jing 余靖 (1000–1064), “Hanwu buyi chengzong lun” 漢武不宜稱宗論, QSW, 27:567.29.

Conclusion

This book provides a missing link in the history of the Middle Period of China. It demonstrates how ritual in the Song dynasty intertwined more with scholarofficials’ intellectual endeavors than with their political stances. Based on their own interpretations of imperial ritual traditions and related ritual commentaries, Northern Song ritual officials sought monarchical support to initiate a campaign of reviving ancient temple rituals. In particular, officials and scholars under the influence of Wang Anshi’s ritual scholarship emphasized the necessity of revising the layout of the Imperial Temple, in order to conform to the ancient setting that was recorded in the ritual Classics. Scholar-officials outside the New Learning circle also championed the New Learning advocacy of an idealized ancient Imperial Temple. Some of them were adamant opponents of Wang Anshi’s New Policies. The disjunction between scholar-officials’ political stances and their ritual interests provides a counterexample to the conventional understanding of Song factional politics as polarizing political groups. As I have demonstrated in my discussion of the 1072 debate on the Primal Ancestor of the temple, it was quite understandable for some late eleventh-century ritual officials to share a common interest with Wang Anshi and Emperor Shenzong in promoting ritual reforms—despite the conservative stances of these same ritual officials on the political level. In this light, this book illustrates how Song debates and discussions over the Imperial Temple and temple rituals differentiated scholar-officials’ ritual interests and shaped their identities on the intellectual dimension. The Song ritual controversy over the Imperial Temple was also a continuation of previous dynasties’ discussions and practices on the arrangement of the zhaomu sequence of temples. I have thoroughly surveyed this continuation from different perspectives. Since the Han period, the conflict between the two approaches of meritocracy and filial piety had been borne on the theories and practices of zhaomu. Song ritual officials developed both approaches in conceptualizing the zhaomu of imperial ancestors and the placement of the Primal Ancestor. Three New Learning ritual officials who were engaged in the 1079 debate over the zhaomu sequence—Lu Dian, He Xunzhi, and Zhang Zao—interpreted the zhaomu sequence according to their individual conceptions of ritual texts. On the one hand, the Confucian

184

Empowered by Ancestors

discourse of filial piety provided a general framework for a hierarchical account of zhaomu and the familial relation between ancestors of different generations. On the other hand, the perception of zhao and mu as genealogical markers indicated a typical merit-based approach, in which the positions of imperial ancestors were determined by their political merits, especially in the case of the Primal Ancestor. New Learning ritualists who had no chance to participate in court ritual debates also contributed to a discursive understanding of the Imperial Temple. In fact, their discursive interpretations of the temple and temple rituals not only blurred the disciplinary boundary of Wang Anshi’s ritual learning but also reflect considerable diversity and comprehensiveness. Through a series of ritual reforms in the late eleventh century, especially the sweeping reforms of imperial sacrificial rituals in the 1080s, New Learning ritualists drastically revised the imperial rituals of the late Northern Song. The reforms eventually led to Emperor Huizong’s pursuit of monumental ritual projects in the early twelfth century, including the promulgation of a new ritual code, the compilation of a new Daoist Canon, the call for a reform on liturgical music, and the endeavor to build an independent Luminous Hall of sacrifice.1 Eventually, immense attention to the Imperial Temple in the Northern Song resulted in the proliferation of related discourses in the Southern Song. The correlation between New Learning and Daoxue ritual scholarship is worth noting. In the canonic synthesis of Daoxue ritual scholarship, the Comprehensive Commentary and Explanation of the Rites and Ceremonies, New Learning and Daoxue converged on the Imperial Temple issue. Furthermore, along with the Daoxue interpretation of the Imperial Temple, the Southern Song and early Yuan witnessed a conceptual shift in the understanding of temple rituals from political agendas to social applications. I have termed this shift as “socialization” in this book. Southern Song and Yuan educated elites incorporated temple rituals into family precepts and lineage regulations for the consolidation of descent groups and lineage organizations. Since the Yuan period, in local ancestral halls and cemetery grounds, the tension between merit-based and filial-based approaches in understanding temple rituals was less intense. This is reflected in the obscure merits of lineage members from a political perspective, and because filial piety itself was more regarded as a merit in most lineage regulations.2 The “socialization” of temple rituals continued in late imperial China. In his pioneering study of Ming and Qing ritual studies, Chow Kai-wing has argued that “the ascendency of ritualism also contributed significantly to the growth of the lineageoriented ancestral cult, which helped reshape the relationship between the imperial

1. Ebrey, Emperor Huizong, 243–54, 160–65, 265–73. 2. For a general explanation of filial piety in Ming and Qing lineage and clan regulations, see Hui-Chen Wang Liu, “An Analysis of Chinese Clan Rules: Confucian Theories in Action,” in Confucianism in Action, ed. David Nivison and Arthur Wright (Stanford, CA: Stanford University Press, 1959), 63–96, especially 84–86.

Conclusion 185

state and the gentry at the local level.”3 The socialization of imperial ancestral cults among the gentry and the commoners accompanied the ascendency of what Chow has described as “ritualism.” The zhaomu discourse in the writings of Ming and Qing ritualists exemplify this phenomenon. For example, Ji Ben 季本 (1485–1563), a Ming Confucian who has been generally considered a follower of Wang Shouren’s 王守仁 (1472–1529) scholarship, criticized how previous Confucians from Lin Yin to Zhu Xi had misinterpreted zhaomu by confining it to the imperial domain. Possibly influenced by the Great Ritual Controversy (dali yi 大禮議) of the 1520s and the 1530s,4 Ji Ben advocated for a general revival of Confucian rituals. He also asserted that an ideal zhaomu reflecting “the ultimate virtues of all-under-Heaven” (tianxia zhi dadao 天下之達道) should be applied to families from any social background.5 Ji Ben argued that zhaomu could be practiced in local lineages and families on a smaller scale, as it was commonly practiced in state sacrifices and ancestral rituals in the Imperial Temple. A detailed analysis of Ji Ben’s zhaomu theory would be beyond the scope of this book. Nonetheless, it is worth noting that Ji paid special attention to the 1079 debate and Zhu Xi’s response to it.6 He agreed with Lu Dian that the differentiation of zhao and mu designations indicates a familial relationship between father- and son-ancestors. Regarding his contemporary practices of ancestral rituals, Ji Ben proposed a relaxation of regulations on the orientation of ancestral buildings, in which zhao and mu ancestors are not strictly arranged to the left and right sides of the Primal Ancestor. According to Ji, the zhaomu arrangement depends on the geographical landscape where the ancestral buildings are located.7 As Ji argued, the ultimate goal of adopting zhaomu and other temple rituals in local ancestral buildings is to unveil the virtue of filial piety in all circumstances. Ji Ben advocated for a general adoption of zhaomu in the ancestral-cult practices of the sixteenth-century society. At the same time, the Ming court was revising 3. Chow Kai-wing, The Rise of Confucian Ritualism in Late Imperial China (Stanford: Stanford University Press, 1994), 224. 4. Plenty of research has been done on the Great Rituals controversy of the mid-Ming period. Carney Fisher’s early work still stands out as the most detailed description of the controversy among Western-language works. See Fisher, The Chosen One: Succession and Adoption in the Court of Ming Shizong (Sydney: Allen and Unwin 1990). An interesting article examines the controversy by taking it as the final chapter of a peculiar practice of mass demonstration of Ming scholar-officials, namely fuque 扶闕 (literally, to prostrate in a kneeling position at the imperial gate for the purpose of presenting a compelling plea to the throne). John Dardess, “Protesting to the Death: The fuque in Ming Political History,” Ming Studies (2003:1): 86–125, especially 109–19. David Faure links the Great Rituals controversy to the general revival of Confucian rites and related social transformation after the 1530s. Faure, Emperor and Ancestor, 100–108. His work inspires me to understand Ji Ben’s writings about temple rituals. 5. Ji Ben, Miaozhi kaoyi 廟制考議, Siku quanshu cunmu congshu 四庫全書存目叢書 (Tainan: Zhuangyan wenhua shiye youxian gongsi, 1997), 105:16b. The phrase “the ultimate virtues of all-under-Heaven” comes from the Doctrine of the Mean, which originally refers to “the virtue of harmony” (he 和). 6. Ji, Miaozhi kaoyi, 105:17a–18b. 7. As Ji Ben put it, “It is appropriate to build temples according to the sequence of ancestors, but not the reverse— that is, to place ancestors into fixed temple space” 以人定廟，則可；以廟定人，則不可. Ji, Miaozhi kaoyi, 105:18b–19a.

186

Empowered by Ancestors

temple rituals on the central level. In 1534, Ming Emperor Shizong 明世宗 (Zhu Houzong 朱厚熜, 1507–1567, r.  1522–1567) revived the ancient arrangement of multiple Imperial Temples by establishing eight separate temples in the Beijing palace. However, an accidental fire in 1541 burned most of his newly established temples to ashes. In 1545, the Ming court re-adopted the original setting of placing multiple chambers in one main temple. In this re-construction of the Imperial Temple, Shizong claimed that the new arrangement of Ming imperial ancestors should not be restricted by zhaomu. Instead, the arrangement should follow the principle of human ethics (lunli 倫理).8 Shizong’s statement certainly reflects his personal interest in honoring his father King Xingxian 興獻王 (Zhu Youyuan 朱祐 杬). In effect, the statement enhances the correlation between temple rituals and the conception of filial piety, as Ji Ben had done in his analysis of the Imperial Temple. Together with Ji Ben, a group of Ming and Qing scholars were fond of the study of the Imperial Temple. These include Wan Sitong, Qin Huitian, Mao Qiling 毛奇 齡 (1623–1726), Hui Dong 惠棟 (1697–1758), Huang Yizhou 黃以周 (1828–1899), and Pi Xirui 皮錫瑞 (1850–1908). Modern scholars have identified most of these scholars with the acclaimed school of “evidential studies” (kaojü xue 考據學) of the High Qing period. Wan Sitong, Hui Dong, Mao Qiling, and Pi Xirui composed individual monographs to analyze the general arrangement of the Imperial Temple and the pivotal role it played in illuminating filial piety.9 Having collected a rich repository of records about the Imperial Temples of different dynasties, Qin Huitian particularly acknowledged the contributions of Song ritualists to rectifying temple rituals. Qin also valued the foreign dynasties’ practices of temple rituals more than Song practices. This may reflect the Qing court’s censorial surveillance on scholars’ publications.10 Ming and Qing scholars’ emphasis on temple rituals not only intended to propagandize Confucian rituals on the social level but also offered an intellectual response to the dynastic court’s manipulation of ritual norms. This type of manipulation was especially discernible when China was conquered by foreign dynasties. In her study of Qing imperial rituals, Evelyn Rawski has distinguished two principles of rule: rule by virtue and rule by heredity.11 The Qing rulers equated the legitimacy of governance with ritual consistency, which was characterized by a compromise between Confucian ritual norms and Manchu shamanism.12 Regarding the general 8. Zhao Kesheng 趙克生, Mingdai de guojia lizhi yu shehui shenghuo 明代的國家禮制與社會生活 (Beijing: Zhonghua shuju, 2012), 11–15. 9. Wan Sitong, Miaozhitu kao; Hui Dong, Di shuo 禘說, in XXSK, 105:2.3a–4b; Mao Qiling, Miaozhi zhezhong 廟 制折衷, Siku quanshu cunmu congshu (Tainan: Zhuangyan wenhua shiye youxian gongsi, 1997), 108:1.3b–17b (the number of temples of the Son of Heaven); 108:2.17a–22b (spatial arrangement of the Imperial Temples); Pi Xirui, Luli dixiayi shuzheng 魯禮禘祫義疏證, XXSK, 112:3a–5a. 10. For example, Qin complained that Song scholars complicated the practice of seasonal sacrifices at the Imperial Temple—especially when a comparison is made with the Liao practice of seasonal sacrifices. Qin Huitian, Wuli tongkao, 94.5789. 11. Rawski, The Last Emperors, 201–3. 12. Abundant research has been done on the Qing adoption of Confucian ritual norms. For a general description,

Conclusion 187

history of ancestral rituals from the sixth to the eighteenth century, it is interesting to note how the Qing rubric of imperial rituals actually reversed what Howard Wechsler has called the shift from “lineal ancestors” to “political ancestors.”13 By giving precedence to the lineal sequence of ancestry in the name of filial piety, Qing imperial ancestral rituals emphasized the numinous connections between emperors and their ancestors. For example, Qing accession rituals evidently shifted the focus from Heaven to filial piety after the seventeenth century.14 In contrast to Tang imperial rituals, Qing imperial rituals symbolized the supreme monarchy of the ruling clan of Aisin Gioro and its linear ancestry. What were the external factors for such a change in the Qing ritual rubric? There is no simple answer to this question. Clearly, the rise of monarchial power after the eleventh century contributed to the involution of most imperial rituals in late imperial China. Yet, monarchial power alone could not explain why early Qing conquerors intended to “Confucianize” their clan sacrifices and perform some of these sacrifices in the Ming Imperial Temple in Beijing. After an examination of Song discourse on the Imperial Temple, I would suggest that imperial ancestral rituals symbolized the ideological correlation between the cultural capital of educated elites and the monarchial power of Chinese emperors. Imperial ancestral rituals were not merely the manifestation of monarchial power. They possessed authority in their own right. Political scientist Lawrence R. Sullivan has argued that “power without authority is reduced to pure coercion with no overarching moral obligation to sanction it.”15 In the case of the Imperial Temple rituals, they provided a set of standardized codes that imbued political power with the moral authority of Confucian norms. Therefore, temple rituals were rooted not only in their political utilization but also in their moral authority in reconciling the tension between educated elites and rulers in the system of monarchical absolutism. Since the eleventh century, the Chinese conception of temple rituals is characterized by an increasing emphasis on the moral authority of imperial ancestral rituals. A transition of focus from political merits to the Confucian-based virtue of filial piety occurred during the Northern to Southern Song transition, along with the rise of monarchical absolutism.16 From the perspective of intellectual history, there was a shift of cultural authority from state to educated elites in terms of ritual see Rawski, The Last Emperors, 197–230; Mark C. Elliott, The Manchu Way: The Eight Banners and Ethnic Identity in Late Imperial China (Stanford, CA: Stanford University Press, 2001), 235–41. 13. Wechsler, Offerings of Jade and Silk, 136. 14. Evelyn Rawski, “The Creation of an Emperor in Eighteenth-Century China,” in Harmony and Counterpoint: Ritual Music in Chinese Context, ed. Bell Yung, Evelyn Rawski, and Rubie Watson (Stanford: Stanford University Press, 1996), 150–74. 15. Lawrence R. Sullivan, “Intellectual and Political Controversies over Authority in China: 1898–1922,” in Confucian Cultures of Authority, ed. Peter D. Hershock and Roger T. Ames (New York: State University of New York Press, 2006), 171. It is noteworthy that Sullivan’s distinction between power and authority has absorbed some ideas from Yves Simon’s and Carl Friedrich’s works. See Sullivan, “Intellectual and Political Controversies,” 196–97, notes 2 and 3. 16. For the rise of monarchical absolutism in the early Southern Song, see Teraji Jun 寺地遵, NanSò shoki seiji rekishi kenkyû 南宋初期政治歷史研究 (Hiroshima: Keisuisha, 1988).

188

Empowered by Ancestors

theories and practices. One may remember how Zhu Xi recognized the endeavor made by the New Learning scholar Lu Dian in constructing an ideal Imperial Temple based on Confucian understanding of filial piety. Likewise, Lu Dian’s temple model was well recognized by Ming and Qing scholars, because it complemented monarchical governance with the ritual norm of Confucian patriarchy. In reality, ritual practices in imperial China propagated political myths that “help structure an understanding of the political world and the public’s attitude to the various political actors that populate it.”17 The collaboration between the cultural authority of scholar-officials and the political power of rulers collapsed during the reign of the Republic of China, given the apparent decline of monarchical absolutism and the retreat of the scholarofficial stratum from the nexus of cultural hegemony and state power. Throughout the twentieth century, Confucianism underwent drastic changes during the Nationalist’s New Culture Movements in the 1920s and the Communist’s Cultural Revolution in the late 1960s and the early 1970s. During these movements, not only Confucian norms but also traditional rituals with Confucian imprints were severely charged. However, traditional rituals experienced a gradual resurgence after the closure of the political campaign against Lin Biao 林彪 (1907–1971) and Confucius in 1974 and later after the arrest of the Gang of Four (siren bang 四人幫) in 1976.18 On the ideological level, the official worship of Confucius in the Confucius Temple of Qufu 曲阜 on September 28, 2004, and also the founding of the first Confucius Institute in Seoul in the same year signify state endorsement of Confucian norms. Such endorsement also renders nationalist connotations to both. On the practical level, the emergence of “popular Confucianism” in contemporary China offers new possibilities for the revitalization of lively Confucian rituals, especially for sacrificial, funeral, and wedding rites.19 As contemporary China witnesses a rekindling of Confucian rituals, will the alliance between ritual authority and state power be restored? Will Chinese intellectuals re-empower Confucian rituals by bridging the conceptual gap between official Communist doctrines and renewed traditions of Confucian rituals? Considering the almost negligible influence of modern Chinese intellectuals in the making of cultural policies, would such an act of re-empowerment be meaningful in a cultural sense, or would it be distorted by the partygovernment of China? After all, is it necessary to restore Confucian rituals to their traditional forms if the restoration of these rituals is intertwined with the radical nationalist trend in modern China? All of these questions call for a rethinking of 17. David I. Kertzer, Ritual, Politics, and Power (New Haven, CT: Yale University Press, 1988), 13. 18. Kam Louie, Critiques of Confucius in Contemporary China (New York: St. Martin’s Press, 1980), 97–136. 19. For two studies on “popular Confucianism” from different approaches, see Guillaume Dutournier and Zhe Ji, “Social Experimentation and ‘Popular Confucianism’: The Case of the Lujiang Cultural Education Centre,” Perspectives 4.80 (2009): 67–81; Philip Clart, “Confucius and the Mediums: Is There a ‘Popular Confucianism,” T’oung Pao 89 (2003): 1–38. From the perspective of historical anthropology, Clart’s article examines the usage of “Popular Confucianism” in Taiwanese religious practices and discusses the intriguing concept of “religionized Confucianism”—a term that some anthropologists have arguably used to describe popular religious practices.

Conclusion 189

the history of Confucian rituals. As long as Confucian rituals are still regarded as a symbol of Chinese culture, related ritual reforms, debates, and discussions will sustain. In the final analysis, the millennial path from Song to contemporary China to revitalize rituals in the past reveals itself as an ongoing process of constructing, imagining, and manipulating China’s cultural memory.

Bibliography

Abbreviations DDSL. Dongdou shilüe 東都事略. Wang Cheng 王稱. Taibei: Wenhai chubanshe, 1967. LJJS. Liji jishuo 禮記集說. Wei Shi 衛湜. Wenyuange Siku quanshu 文淵閣四庫全書 edition, vol. 117–20. LCJ. Linchuan xiansheng wenji 臨川先生文集. Wang Anshi 王安石. Sibu congkan chubian suoben 四部叢刊初編縮本 edition. QSBJ. Quan Song biji 全宋筆記. Zhu Yian 朱易安 et al. Zhengzhou: Daxiang chubanshe, 2006. QSW. Quan Song wen 全宋文. Shanghai, Hefei: Shanghai cishu chubanshe, Anhui jiaoyu chubanshe, 2006. SBCK. Sibu congkan 四部叢刊. Compiled by Zhang Yuanji 張元濟. Shanghai: shangwu yin­ shuguan, 1919. SBCKCBSB. Sibu congkan chubian suoben 四部叢刊初編縮本. Compiled by Zhang Yuanji. Taibei: Shangwu yinshuguan, 1967. SHY. Songhuiyao jigao 宋會要輯稿. Compiled by Xu Song 徐松. Taibei: Shijie shuju, 1964. SKQS. Wenyuange Siku quanshu 文淵閣四庫全書. Compiled by Ji Yun 紀昀. Shanghai: Shanghai guji chubanshe, 1987. SKQSZB. Siku quanshu zhenben 四庫全書珍本. Compiled by Wang Yunwu 王雲五. Taibei: Shangwu yinshuguan, 1976. SS. Song shi 宋史. Compiled by Tuo Tuo 脫脫. Beijing: Zhonghua shuju, 1985. SSXP. Songshi xinpian 宋史新編. Ke Weiqi 柯維騏. Tianjin: Tianjin guji chubanshe, 1998. SSXYJKHP. Sanjing xinyi jikao huiping 三經新義輯考彙評. Cheng Yuanmin 程元敏. Shanghai: Hua dong shifan daxue chubanshe, 2011. TSZSSSJ. Tang Song zhushu shisanjing 唐宋注疏十三經. Compiled by Zhonghua shuju bian­ jibu 中華書局編輯部. Beijing: Zhonghua shuju, 1998. XCB. Xu zizhi tongjian changbian 續資治通鑑長編. Li Tao 李燾. Beijing: Zhonghua shuju, 1986. XXSK. Xuxiu siku quanshu 續修四庫全書. Compiled by Xuxiu Sikuquanshu Bianzuan­ weiyuanhui 續修四庫全書編纂委員會. Shanghai: Shanghai guji chubanshe, 1995. YLJZTJ: JL, the Yang Edition. Yili jingzhuan tongjie xujuan: jili 儀禮經傳通解續卷祭禮. Yang Fu 楊復. Compiled by Ye Chunfang 葉純芳 and Hashimoto Hidemi 橋本秀美. Taibei: Zhongyang yanjiuyuan zhongguo wenzhe yanjiusuo, 2011. YLJZTJXJ, the Huang Edition. Yili jingzhuan tongjie xujuan 儀禮經傳通解續卷. Huang Gan 黃榦. SKQS edition, vol.132.

192 Bibliography ZBSYXA. Zengbu Songyuan xuean 增補宋元學案. Compiled by Huang Zongxi 黃宗羲 and Quan Zuwang 全祖望. Taibei: Zhonghua shuju, 1970. ZZYL. Zhuzi yulei 朱子語類. Zhu Xi 朱熹. Beijing: Zhonghua shuju, 1986.

Primary Sources Anonymous. Zhouli jishuo 周禮集說. SKQS edition, vol. 95. Ban Gu 班固. Han shu 漢書. Beijing: Zhonghua shuju, 1962. Cai Shangxiang 蔡上翔. Wang Jinggong nianpu kaolue 王荊公年譜考略. Beijing: Zhonghua shuju, 2006. Cai Yong 蔡邕. Caizhonglang ji 蔡中郎集. Shanghai: Zhonghua shujiu, 1936. Chao Yuezhi 晁說之. Chaoshi keyu 晁氏客語. QSBJ, Series 1:10. Chen Gao 陳高. Buxi zhouyuji 不繫舟漁集. SKQS edition, vol. 1216. Chen Lü 陳旅. Anyatang ji 安雅堂集. SKQS edition, vol. 1213. Chen Xiang 陳襄. Guling ji 古靈集. SKQS edition, vol. 1093. Chen Xiangdao 陳祥道. Lishu 禮書. SKQS edition, vol. 130. Chen Yuanjing 陳元靚. Xinbian qunshu leiyao shilin guangji 新編群書類要事林廣記. Beijing: Zhonghua shuju, 1999. ———. Zuantu zengxin qunshu leiyao shilin guangji 纂圖增新群書類要事林廣記. Beijing: Zhonghua shuju, 1999. Chen Zhensun 陳振孫. Zhizhai shulujieti 直齋書錄解題. Taibei: Guangwen shuju, 1968. Cheng Hao 程顥 and Cheng Yi 程頤. Ercheng yishu 二程遺書. In Ercheng ji 二程集. Beijing: zhonghua shuju, 1981. Chunqiu gongyang zhuan zhushu 春秋公羊傳注疏. TSZSSSJ, vol. 3. Sibu beiyao 四部備要 edition, commentated by He Xiu 何休, sub-commentated by Xu Yan 徐彥. Chunqiu guliang zhuan zhushu 春秋縠梁傳注疏. TSZSSSJ, vol. 3. Sibu beiyao edition, com­ mentated by Fan Ning 范寧, sub-commentated by Yang Shixun 楊士勛. Chunqiu Zuo zhuan zhushu 春秋左傳注疏. TSZSSSJ, vol. 3. Sibu beiyao edition, commen­ tated by Tu Yu 杜預, sub-commentated by Kong Yingda 孔穎達. Dai Liang 戴良. Jiuling shanfang ji 九靈山房集. SKQS edition, vol. 1219. Du Dagui 杜大珪 ed. Mingchen beizhuan wanyan zhi ji 名臣碑傳琬琰之集. SKQS edition, vol. 1092. Fan Ye 范曄. Hou Han shu 後漢書. Beijing: Zhonghua shuju, 2003. Fan Zhen 范鎮. Dongzhai jishi 東齋記事. Beijing: Zhonghua shuju, 1980. Fan Zhongyan 范仲淹. Fanwenzheng ji 范文正集. SKQS edition, vol. 1089. Fan Zuyu 范祖禹. Fan taishiji 范太史集, SKQS edition, vol. 1100. Fang Dazong 方大琮. Tiean ji 鐵菴集. SKQS edition, vol. 1178. Fang Shao 方勺. Bozhai bian 泊宅編. QSBJ, Series 2:8. Fang Xuanling 房玄齡 et al. Jin shu 晉書. Beijing: Zhonghua shuju, 1975. Gaoben Song Yuan xuean buyi 稿本宋元學案補遺, compiled by Wang Zicai 王梓材 and Feng Yunhao 馮雲濠. Beijing: Beijing tushuguan chubanshe, 2002. Ge Shengzhong 葛勝仲. Danyang ji 丹陽集. SKQS edition, vol. 1127. Gu Donggao 顧楝高. Wangjinggong nianpu 王荊公年譜. Songren nianpu congkan 宋人年譜 叢刊, edited by Wu Hongze 吳洪澤, Yin Bo 尹波. Chengdu: Sichuan daxue chubanshe, 2003. Gu Yanwu 顧炎武. Rizhilu 日知錄. Gu Yanwu quanji 顧炎武全集, vol. 28. Shanghai: Shanghai gujichubanshe, 2011.

Bibliography 193 Han Wei 韓維. Nanyang ji 南陽集. SKQS edition, vol. 1101. Han Yin 韓嬰. Hanshi waizhuan jishi 韓詩外傳集釋, compiled by Xu Weiyu 許維遹. Beijing: Zhonghua shuju, 1980. Han Yu 韓愈. Han cangli ji 韓昌黎集. Beijing: Shangwu yinshuguan, 1958. Hu Ciyan 胡次焱. Meiyan wenji 梅巖文集. SKQS edition, vol. 1188. Hu Hong 胡宏. Huhong ji 胡宏集. Beijing: Zhonghua shuju, 1987. Huang Du 黃度. Song Huang xuanxiangong zhouli shuo 宋黃宣獻公周禮說. XXSK edition, vol. 78. Huang Yizhou 黃以周. Lishu tonggu 禮書通故. Beijing: Zhonghua shuju, 2007. Huang Zhongyuan 黃仲元. Siru ji 四如集. SKQS edition, vol. 1188. Hui Dong 惠棟. Di shuo 禘說. XXSK edition, vol. 105. Ji Ben 季本. Miaozhi kaoyi 廟制考議. Siku quanshu cunmu congshu. Tainan: Zhuangyan wenhua shiye youxian gongsi, 1997, vol. 105. Li Cun 李存. Sian ji 俟庵集. SKQS edition, vol. 1213. Li Po 李朴. Fengqingmin gong yi shi 豐清敏公遺事. QSBJ, Series 2:8. Li Qi 李祁. Yunyang ji 雲陽集. SKQS edition, vol. 1219. Li Rugui 李如圭. Yili shigong 儀禮釋宮. SKQS edition, vol. 103. Li Xinchuan 李心傳. Jianyan yilai xinian yaolu 建炎以來繫年要錄. Beijing: Zhonghua shuju, 2013. Li Zhi 李廌. Shiyou tanji 師友談記. Beijing: Zhonghua shuju, 2002. Liji zhushu 禮記注疏. TSZSSSJ, vol.2. Sibu beiyao edition, commentated by Zheng Xuan 鄭 玄, sub-commentated by Kong Yingda. Liu Xu 劉昫 (887–946) et al. Jiu Tangshu 舊唐書. Beijing: Zhonghua shuju, 1975. Lu Dian 陸佃. Taoshan ji 陶山集. SKQS edition, vol. 1117. Lu Wenchao 盧文弨. Yili zhushu xiangjiao 儀禮注疏詳校, annotated by Chen Donghui 陳 東輝 and Peng Xishuang 彭喜雙. Taibei: Zhongyang yanjiuyuan Zhongguo wenzhe yanjiusuo, 2012. Lu Xinyuan 陸心源. Yuanyou dangrenzhuan 元祐黨人傳. XXSK edition, vol. 517. Lu You 陸游. Laoxuean biji 老學庵筆記. Song Yuan biji xiaoshuo daguan 宋元筆記小說大觀, vol.4. Shanghai: Shanghai guji chubanshe, 2007. Lü Benzhong 呂本中. Ziwei zashuo 紫微雜說. QSBJ, Series 3:6. Lü Zuqian 呂祖謙. Donglai Lütaishi bieji 東萊呂太史別集. Lü Zuqian quan ji 呂祖謙全集, edited by Huang Linggeng 黃靈庚. Huangzhou: Zhejiang guji chubanshe, 2008. ———. Huangchao wenjian 皇朝文鑑. Lü Zuqian quan ji. Ma Duanlin 馬端臨. Wen xian tongkao 文獻通考. Beijing: Zhonghua shuju, 1986. Ma Guohan 馬國翰. Yuhan shanfang jiyishu 玉函山房輯佚書. Shanghai: Shanghai guji chu­ banshe, 1990. Mao Qiling 毛奇齡. Miaozhi zhezhong 廟制折衷. Siku quanshu cunmu congshu 四庫全書存 目叢書. Tainan: Zhuangyan wenhua shiye youxian gongsi, 1997, vol. 108. Nansong guange xulu 南宋館閣續錄. Beijing: Zhonghua shuju, 1998. Nie Chongyi 聶崇義. Sanlitu jizhu 三禮圖集註. SKQS edition, vol. 129. Ouyang Xiu 歐陽修. Taichangyingeli 太常因革禮. Congshujicheng cubian 叢書集成初編. Shanghai: Shangwu yinshuguan, 1936. Ouyang Xiu and Song Qi 宋祁. Xin Tangshu 新唐書. Beijing: Zhonghua shuju, 1975. Pang Yuanying 龐元英. Wenchang zalu 文昌雜錄. QSBJ, Series 2:4. Pi Xirui 皮錫瑞. Luli dixiayi shuzheng 魯禮禘祫義疏證. XXSK edition, vol. 112.

194 Bibliography Qin Huitian 秦蕙田. Wuli tongkao 五禮通考. Taibei: Xin xing shuju, 1970. Qiu Kui 丘葵. Zhouli buwang 周禮補亡, compiled by Gu Kejiu 顧可久. Harvard Yenching Rare Book Collections. Quan shanggu Sandai Qin Han Sanguo Liuchao wen 全上古三代秦漢三國六朝文. Compiled by Yan Kejun 嚴可均. Taibei: Shijie shuju, 1963. Shao Bo 邵博. Shaoshi wenjian houlu 邵氏聞見後錄. Beijing: Zhonghua shuju, 1997. Shao Bowen 邵伯溫. Shaoshi wenjian lu 邵氏聞見錄. Beijing: Zhonghua shuju, 1983. Sima Guang 司馬光. Wenguo wenzheng simagong ji 溫國文正司馬公集. SBCKCBSB edition, vol. 46. Song Minqiu 宋敏求. Chunming tuichaolu 春明退朝錄. Beijing: Zhonghua shuju, 1980. Su Shi 蘇軾. Sushi quanji jiaozhu 蘇軾全集校注, edited by Zhang Zhilie 張志烈, Ma Defu 馬 德富, and Zhou Yukai 周裕鍇. Shijiazhuang: Hebei renmin chubanshe, 2010. Su Song 蘇頌. Suweigong wenji 蘇魏公文集. Wang Congce 王從策 et al. Beijing: Zhonghua shuju, 2004. Su Xun 蘇洵. Jiayou ji 嘉祐集. SBCKCBSB edition, vol. 51. Su Yu 蘇輿. Chunqiu fanlu yizheng 春秋繁露義證. Beijing: Zhonghua shujiu, 1992. Su Zhe 蘇轍. Longchuan lüezhi 龍川略志. Beijing: Zhonghua shuju, 1982. Tang Qifu 湯勤福, Wang Zhiyao 王志躍. Song shi lizhi bianzheng 宋史禮志辨證. Shanghai: Shanghai sanlian shudian, 2011. Wan Sitong 萬斯同. Miaozhitu kao 廟制圖考. SKQSZB edition, Series 6, vol. 369. Wang Anshi 王安石. Wangwengong wenji 王文公文集. Beijing: Zhonghua shuju, 1962. ———. Zhouli xinyi 周禮新義. SSXYJKHP. Wang Chang 王昶. Jinshi cuibian 金石萃編. XXSK edition, vol. 891. Wang Decheng 王得臣. Zhushi 麈史. In Song Yuan biji congshu 宋元筆記叢書. Shanghai: Shanghai guji chubanshe, 1986. Wang Li 王禮. Linyuan wenji 麟原文集. SKQS edition, vol. 1220. Wang Liqi 王利器. Yantie lun jiaozhu 鹽鐵論校注. Beijing: Zhonghua shuju, 1992. Wang Pinzhen 王聘珍. Da Dai liji jiegu 大戴禮記解詁. Beijing: Zhonghua shuju, 1983. Wang Pu 王溥. Tang huiyao 唐會要. Beijing: Zhonghua shujiu, 1955. Wang Ruilai 王瑞來. Song zaifu biannianlu jiaobu 宋宰輔編年錄校補. Beijing: Zhonghua shuju, 1986. Wang Su 王肅. Kongzi jiayu 孔子家語. SKQS edition, vol. 695. Wang Xianqian 王先謙. Hanshu buzhu 漢書補注. Yangzhou: Guangling shushe, 2006. ———. Xunzi jijie 荀子集解. Beijing: Zhonghua shuju, 1988. Wang Yan 王炎. Shuangxi leigao 雙溪類稾. SKQS edition, vol. 1155. Wang Yuzhi 王與之. Zhouli dingyi 周禮訂義. SKQS edition, vol. 93–94. Wang Zhaoyu 王昭禹. Zhouli xiangjie 周禮詳解. SKQS edition, vol. 91. Wang Zhu 王洙. Tujie jiaozheng dili xinshu 圖解校正地理新書. Taibei: Jiwen shuju, 1985. Wei Hong 衛宏. Han jiuyi 漢舊儀, compiled by Sun Xingyan 孫星衍. Congshujicheng cubian. Beijing: Zhonghua shujiu, 1985. Wen Ying 文瑩. Xiangshan yelu 湘山野錄. Beijing: Zhonghua shuju, 1984. Xiao Song 蕭嵩et al. Da Tang kaiyuan li 大唐開元禮. Beijing: Minzu chubanshe, 2000. Xie Weixin 謝維新. Gujin hebi shilei beiyao: qianji 古今合璧事類備要前集. SKQS edition, vol. 939. ———. Gujin hebi shilei beiyao: xuji 古今合璧事類備要續集. SKQS edition, vol. 940. Xie Yingfang 謝應芳. Guichao gao 龜巢稿. SKQS edition, vol. 1218.

Bibliography 195 Xue Jixuan 薛季宣. Xue Jixuan ji 薛季宣集. Shanghai: Shanghai shehui kexue chubanshe, 2003. Yan Zhenqing 顏真卿. Yan Lugong wenji 顏魯公文集. Sibu beiyao edition, vol. 228. Shanghai: Zhonghua shuju, 1936. Yang Fu 楊復. Yili pangtongtu 儀禮旁通圖. SKQS edition, vol. 104. Yang Jie 楊傑. Wuweiji jiaojian 無為集校箋. Annotated by Cao Xiaoyun 曹小雲. Hefei: Huangshan shushe, 2014. Yang Shi 楊時. Yangshi ji 楊時集. Beijing: Zhonghua shuju, 2018. Ye Mengde 葉夢得. Shilin yanyu 石林燕語. Beijing: Zhonghua shuju, 1984. Ye Shi 葉時. Lijing huiyuan 禮經會元. SKQS edition, vol. 92. Ye Shi 葉適. Yeshi ji 葉適集. Beijing: Zhonghua shujiu, 1961. Yi Fu 易祓. Zhouguan zongyi 周官總義. SKQS edition, vol. 92. Yin Chu 尹洙. Henan ji 河南集. SBCKCBSB edition, vol. 45. Zeng Bu 曾布. Zenggong yilu 曾公遺錄. QSBJ, Series 1:8. Zeng Gong 曾鞏. Yuanfeng leigao 元豐類藁. SBCK edition, vol. 1680–89. Zeng Minxing 曾敏行. Duxing zazhi 獨醒雜誌. QSBJ, Series 4:5. Zhang Shizheng 張師正. Juanyou zalu 倦遊雜錄. Shanghai: Shanghai guji chubanshe, 1993. Zhang Xuecheng 章學誠. Wenshi tongyi 文史通義. Beijing: Zhonghua shuju, 1985. Zeng Zhao 曾肇. Qufu ji 曲阜集. SKQS edition, vol. 1101. Zheng Juzong 鄭居中 et al. Zhenghe wuli xinyi 政和五禮新儀. SKQS edition, vol. 647. Zhongxing lishu 中興禮書, compiled by Xu Song 徐松. Shanghai: Shanghai guji chubanshe, 1995. Zhou Mi 周密. Qidong yeyu 齊東野語. Beijing: Zhonghua shuju, 1983. Zhouli zhushu 周禮注疏. TSZSSSJ, vol.2. Sibu beiyao edition, commentated by Zheng Xuan, sub-commentated by Jia Gongyan. Zhu Bian 朱弁. Quwei jiuwen 曲洧舊聞. Beijing: Zhonghua shuju, 2002. Zhu Bin 朱彬. Liji xunzuan 禮記訓纂. Beijing: Zhonghua shuju, 1996. Zhu Mu 祝穆. Gujin shiwen laijiu: houji 古今事文類聚：後集. SKQS edition, vol. 926. Zhu Yizun 朱彝尊. Jingyi kao 經義考. Beijing: Zhonghua shuju, 1998. Zhu Xi 朱熹. Huian xiansheng zhuwengong ji 晦菴先生朱文公集. SBCKCBSB edition, vol. 58–59. ———. Sanchao mingchen yanxinglu 三朝名臣言行錄. SBCK edition, vol. 1094–1101. ———. Sishu zhangjujizhu 四書章句集注. Shanghai: Shanghai guji chubanshe, Anhui jiaoyu chubanshe, 2001.

Secondary Sources Arnheim, Rudolf. New Essays on the Psychology of Art. Berkeley: University of California Press, 1986. Asim, Ina. “Status Symbol and Insurance Policy: Song Land Deeds for the Afterlife.” In Burial in Song China, edited by Dieter Kuhn, 307–64. Heidelberg: Edition Forum, 1994. Azuma Jūji 吾妻重二. “Sōdai no kabyō to sosen saishi” 宋代の家廟と祖先祭祀. In Chūgoku no reisei to reigaku 中國の禮制と禮學, edited by Kominami, Ichirō 小南一郎, 505–75. Kyoto: Hōyū shoten, 2001. Bell, Catherine. Ritual: Perspectives and Dimensions. New York: Oxford University Press, 1997.

196 Bibliography Benjamin, Walter. Illumination. New York: Schocken Books, 2007. Bloom, Harold. The Anxiety of Influence: A Theory of Poetry. New York: Oxford University Press, 1973. Bokenkamp, Stephen. Ancestors and Anxiety: Daoism and the Birth of Rebirth in China. Berkeley: University of California Press, 2007. Bol, Peter. “Government, Society, and State: On the Political Visions of Ssu-ma Kuang and Wang An-shih.” In Ordering the World: Approaches to State and Society in Sung Dynasty China, edited by Robert Hymes and Conrad Schirokauer, 128–92. Berkeley: University of California Press, 1993. ———. “Neo-Confucianism and Local Society, 12th–16th Century: A Case Study.” In The Song-Yuan-Ming Transition in Chinese History, edited by Paul Jakov Smith and Richard von Glahn, 241–83. Cambridge, MA: Harvard University Asia Center, 2003. ———. “Reconceptualizing the Order of Things in Northern and Southern Sung.” In The Cambridge History of China, Vol. 5, Part II: Sung China, 960–1279, edited by John. W. Chaffee and Denis Twitchett, 665–726. Cambridge, UK: Cambridge University Press, 2009. ———. “The Rise of Local History: History, Geography, and Culture in Southern Song and Yuan Wuzhou.” Harvard Journal of Asiatic Studies 61.1 (2001): 37–76. ———. This Culture of Ours: Intellectual Transitions in T’ang and Sung China. Stanford, CA: Stanford University Press, 1992. ———. “Wang Anshi and the Zhouli.” In Statecraft and Classical Learning: The Rituals of Zhou in East Asian History, edited by Benjamin Elman and Martin Kern, 229–51. Boston, MA: Brill, 2010. Bourdieu, Pierre. “The Forms of Capital.” In Handbook of Theory and Research for the Sociology of Education, edited by J. Richardson, 241–58. New York: Greenwood, 1986. Brashier, Kevin. Ancestral Memory in Early China. Cambridge, MA: Harvard University Press, 2011. Brook, Timothy. “Funerary Ritual and the Building of Lineages in Late Imperial China.” Harvard Journal of Asiatic Studies 49.2 (1989): 465–99. Chaffee, John. Branches of Heaven: A History of the Imperial Clan of Sung China. Cambridge, MA: Harvard University Press, 1999. ———. The Thorny Gates of Learning in Sung China: A Social History of Examinations. New York: State University of New York Press, 1995. Chan Hok-lam 陳學霖. “Ke Weiqi Songshi xinpian shuping” 柯維騏宋史新編述評. In Songshi yanjiuji 宋史研究集 v.20, 489–526. Taibei: Guoli bianyiguan, 1990. ———. “Song Jin erdi yiqi ding tianxia: Xuanhe yishi kaoshi yize” 宋金二帝弈棋定天下 ──《宣和遺事》考史一則. In Song shi lunji 宋史論集, 217–18. Taibei: Dongda tushu gongsi, 1993. Chang Jianhua 常建華. Mingdai zongzu yanjiu 明代宗族研究. Shanghai: Shanghai renmin chubanshe, 2005. Chen Fangmei 陳芳妹. Qingtongqi yu Songdai wenhua shi 青銅器與宋代文化史. Taiwan: Guoli Taiwan daxue chuban zhongxin, 2016. Chen Lai 陳來. Zhu Zi shuxin biannian kaozheng 朱子書信編年考證. Beijing: Sanlian shudian, 2011. Chen Quanfang 陳全方. Zhouyuan yu Zhou wenhua 周原與周文化. Shanghai: Shanghai renmin chubanshe, 1988.

Bibliography 197 Chen Shu 陳述. “Dongdou shilüe zhuanren Wang Shang Cheng fuzi” 東都事略撰人王賞稱 父子. Zhongyang yanjiuyuan lishi yuyan yanjiusuo jikan 中央研究院歷史語言研究所 集刊 8:1 (1939): 129–38. Chen Shuguo 陳戍國. Zhongguo lizhishi: suitang wudai juan 中國禮制史：隋唐五代卷. Changsha: Hunan jiaoyu chebanshe, 1998. Chen Yinque 陳寅恪. Tangdai zhengzhi shishu lungao 唐代政治史述論稿. Shijiazhuang: Hebei jiaoyu chubanshe, 2002. Chen Wenyi 陳雯怡. “Networks, Communities, and Identities: On the Discursive Practices of Yuan Literati.” PhD dissertation, Harvard University, 2007. Cheung Hiu Yu 張曉宇. “Cong bian Tang zhi li dao zhangxian xinyi: Bei Song qianqi xinli wenti yu Taichang yingeli” 從變唐之禮到章獻新儀──北宋前期新禮問題與太常因 革禮. Hanxue yanjiu 漢學研究 37.1 (2019.3): 39–82. ———. “Cong Huangyin shijian zailun Yuanyou chuqi zhengjiu yu dangzheng” 從黃隱事 件再論元祐初期政局與黨爭. Journal of Chinese Studies 中國文化研究所學報 66 (Jan 2018): 1–23. ———. “Cong Yuan-Ming dixiu baiwushijuanben Lishu luelun Chen Xiangdao Lishu de jinxianguocheng ji yiyi” 從元明遞修百五十卷本《禮書》略論陳祥道《禮書》的進獻 過程及意義. Lishi wenxian yanjiu 歷史文獻研究 39 (Sept 2017): 291–300. ———. “Inventing a New Tradition: The Revival of the Discourses of Family Shrines in the Northern Song.” Journal of Song-Yuan Studies 47 (2019): 85–136. ———. “Ritual and Politics: An Examination of the 1072 Primal Ancestor Debate in the Northern Song.” Frontiers of History in China 13.3 (2018.10): 275–310. ———. “The 1079 zhaomu Debate: The Song Ritual Controversy over Ancestral Rites.” Conference paper. Western Branch Meeting of the American Oriental Society, 2012. ———. “The Way Turning Inward: An Examination of the ‘New Learning’ Usage of daoxue in Northern Song China.” Philosophy East and West 69.1 (2019.1): 86–107. ———. “Zhuanda de xianzhi: Tang zhi Beisong taichangliyuan yangekao” 專達的閒職─ 唐至北宋太常禮院沿革考. In Zhongguo gudai zhengzhizhidu yu lishidili: Yangengwang xiansheng bailing jinian lunwenji 中國古代政治制度與歷史地理─嚴耕望先生百齡紀 念論文集, 177–204. Jinan: Qilu shushe, 2020. Ching, Julia. The Religious Thought of Chu Hsi. New York: Oxford University Press, 2000. Choi, Mihwa. Death Rituals and Politics in Northern Song China. New York: Oxford University Press, 2017. Chow Kai-wing 周啟榮. The Rise of Confucian Ritualism in Late Imperial China. Stanford, CA: Stanford University Press, 1994. Clark, Hugh. Community, Trade and Networks. Cambridge: Cambridge University Press, 1991. Clart, Philip. “Confucius and the Mediums: Is There a ‘Popular Confucianism.” T’oung Pao 89 (2003): 1–38. Cui Jianfang 崔建芳. “Lun huangquan chuanchengguifan dui Xihan dillingbuju de zhiyue” 論皇權傳承規範對西漢帝陵佈局的制約. Kaogu yu wenwu 考古與文物 2 (2012): 60–64. Danforth, Timothy. “The Imperial Ancestral in China’s Western Han Dynasty: Institutional Tradition and Personal Belief.” PhD dissertation, Harvard University, 2006. Dardess, John. “Protesting to the Death: The Fuque in Ming Political History.” Ming Studies 1 (2003): 86–125.

198 Bibliography de Weerdt, Hilde. Competition over Content: Negotiating Standards for the Civil Service Examinations in Imperial China (1127–1279). Cambridge, MA: Harvard University Press, 2007. ———. Review of Divided by a Common Language: Factional Conflict in Late Northern Song China, by Ari Levine. Journal of Asian Studies 69.2 (2010): 556–58. ———. Review of Ritendiskussionen am Hof der nördlichen Song-Dynastie (1034–1093): zwischen Ritengelehrsamkeit, Machtkampfund intellektuellen Bewegungen, by Christian Meyer. Bulletin of the School of Oriental and African Studies, 72.1 (2009): 205–7. Deng Guangming 鄧廣銘. Beisong zhengzhi gaigejia Wang Anshi 北宋政治改革家王安石. Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2007. Deng Xiaonan 鄧小南. Zuzong zhifa: Beisong qianqi zhengzhi shulue 祖宗之法：北宋前期政 治述略. Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 2006. Deng Zhirui 鄧智睿. “Tianxia yijia dao yijia tianxia: yi tangsong miaoyi yu junwei qianghua wei zhongxin de taolun” 天下一家到一家天下：以唐宋廟議與君位強化為中心的討 論. MA Thesis, National Taiwan Normal University, 2011. Dutournier, Guillaume, and Zhe Ji. “Social Experimentation and ‘Popular Confucianism’: The Case of the Lujiang Cultural Education Centre.” Perspectives 4:80 (2009): 67–81. Ebrey, Patricia Buckley. Accumulating Culture: The Collections of Emperor Huizong. Seattle: University of Washington Press, 2008. ———. Chu Hsi’s Family Rituals: A Twelfth-Century Chinese Manual for the Performance of Cappings, Weddings, Funerals, and Ancestral Rites. Princeton, NJ: Princeton University Press, 1991. ———. Confucianism and Family Rituals in Imperial China: A Social History of Writing about Rites. Princeton, NJ: Princeton University Press, 1991. ———. Emperor Huizong. Cambridge, MA: Harvard University Press, 2014. ———. Family and Property in Sung China: Yuan Ts’ai’s Precepts for Social Life. Princeton, NJ: Princeton University Press, 1984. ———. “Portrait Sculptures in Imperial Ancestral Rites in Song China.” T’oung Pao 83 (1997): 42–92. ———. “Sung Neo-Confucian Views on Geomancy.” In Meeting of Minds: Intellectual and Religious in East Asian Traditions of Thought, edited by Irene Bloom and Joshua A. Fogel, 75–97. New York: Columbia University Press, 1997. ———. “The Early Stages in the Development of Descent Group Organization.” In Kinship Organization in Late Imperial China: 1000–1940, edited by Patricia Ebrey and James Watson, 16–61. Los Angeles: University of California Press, 1986. Ebrey, Patricia, and James Watson, eds. Kinship Organization in Late Imperial China: 1000– 1940. Los Angeles: University of California Press, 1986. Egawa Shikibu 江川式部. “Teigen nenkan no taibyōsōgi to tōdaikōki no reisei kaikaku” 貞 元年間の太廟奏議と唐代後期の禮制改革. Chūgoku shigaku 中國史学 20 (2010): 153–75. Elliott, Mark C. The Manchu Way: The Eight Banners and Ethnic Identity in Late Imperial China. Stanford, CA: Stanford University Press, 2001. Eno, Robert. The Confucian Creation of Heaven: Philosophy and the Defense of Ritual Mastery. New York: State University of New York, 1990. Faure, David. Emperor and Ancestor: State and Lineage in South China. Stanford, CA: Stanford University Press, 2007.

Bibliography 199 Feng Jiren. Chinese Architecture and Metaphor: Song Culture in the Yingzhao Fashi Building Manual. Hong Kong: Hong Kong University Press, 2012. Feuchtwang, Stephan, and Arthur Wolf, eds. Religion and Ritual in Chinese Society. Stanford, CA: Stanford University Press, 1974. Fisher, Carney. The Chosen One: Succession and Adoption in the Court of Ming Shizong. Sydney: Allen and Unwin, 1990. ———. “The Ritual Dispute of Sung Ying-tsung,” Papers on Far Eastern History 36 (1987): 109–38. Forte, Antonino. Mingtang and Buddhist Utopias in the History of the Astronomical Clock: The Tower, Statue and Armillary Sphere Constructed by Empress Wu. Rome: Instituto Italiano per il Medio ed Estremo Oriente, 1988. Freedman, Maurice. Chinese Lineage and Society: Fukien and Kwangtung. London: Humanities Press, 1966. ———. Lineage Organization in Southeastern China. London: Athlone, 1958. Gale, Essen. Discourses on Salt and Iron: A Debate on State Control of Commerce and Industry in Ancient China. Taipei: Ch’eng-wen Publishing Company, 1967. Gallagher, Catherine, and Stephen Greenblatt. Practicing New Historicism. Chicago: Chicago University Press, 2000. Gan Huaizhen 甘懷真. Tangdai jiamiao lizhi yanjiu 唐代家廟禮制硏究. Taibei: Taiwan shangwu yinshuguan, 1991. ———. Huangquan liyi yu jingdianquanshi: zhongguo gudai zhengzhishi yanjiu 皇權、禮儀 與經典詮釋：中國古代政治史研究. Taibei: Ximalaya yanjiufazhan jijinhui, 2003. Gao Mingshi 高明士. “Lifa yiyi xiade zongmiao: yi zhongguo zhonggu weizhu” 禮法意義 下的宗廟─以中國中古為主. In Dongya chuantong jiali, jiaoyu yu guofa: jiazu, jiali yu jiaoyu 東亞傳統家禮、教育與國法：家族、家禮與教育, 23–86. Shanghai: Huadong shifan daxue chubanshe, 2008. Geertz, Clifford. Negara: The Theatrical State in Nineteenth-Century Bali. Princeton, NJ: Princeton University Press, 1980. Gong Yanming 龔延明. “Yuanfeng qianhou liang Song wenguan jilu guanjie duizhao biao” 元豐前後兩宋文官寄祿官階對照表. In Songdai guanzhi cidian 宋代官制辭典. Beijing: Zhonghua shuju, 2017. Guo Shanbing 郭善兵. Zhongguo gudai diwang zongmiao lizhi yanjiu 中國古代帝王宗廟禮 制研究. Beijing: Renmin chubanshe, 2007. Hangzhoushi Wenwu Kaogusuo 杭州市文物考古所. Nan Song taimiao yizhi 南宋太廟遺址. Beijing: Wenwu chubanshe, 2007. Hartman, Charles. “A Textual History of Cai Jing’s Biography in the Songshi.” In Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, edited by Patricia Ebrey and Maggie Bickford, 517–64. Cambridge, MA: Harvard University Asian Center, 2006. ———. “Sung Government and Politics.” In The Cambridge History of China, Vol. 5, Part II: Sung China, 907–1279, edited by John W. Chaffee and Denis Twitchett, 49–80. Cambridge, UK: Cambridge University Press, 2015. Hartwell, Robert. “Demographic, Political and Social Transformation of China, 750–1550.” Harvard Journal of Asiatic Studies, 42.2 (1982): 365–442. Henderson, John. The Development and Decline of Chinese Cosmology. New York: Columbia University Press, 1984.

200 Bibliography Higashi Ichio 東一夫. Ō Anseki shinpō no kenkyū 王安石新法の研究. Tokyo: Kazama Shobō, 1970. Hiraoka Takeo 平岡武夫. Tangdai de Changan yu Luoyang: ditu 唐代的長安與洛陽：地圖. Shanghai: Shanghai guji chubanshe, 1991. Hirata Shigeki 平田茂樹. Songdai zhengzhi jiegou yanjiu 宋代政治結構研究. Shanghai guji chubanshe, 2010. Hsu, L. K. Francis. Under the Ancestors’ Shadow: Kinship, Personality, and Social Mobility in Village China. New York: Natural History Library, 1967. Hua Zhe 華喆. “Fuzi yilun: Bei Song Yuanfeng zhiyi zai pinjia” 父子彝倫：北宋元豐昭穆之 議再評價. In Zhongguo zhexueshi 中國哲學史 3 (2017): 18–29. Huang Qijiang 黃啟江. “Bei Song jushi Yangjie yu fojiao: jianbu Song shi Yangjie benzhuan zhique” 北宋居士楊傑與佛教─兼補宋史楊傑本傳之缺. Hanxue yanjiu 21:1 漢學研究 (2003.6): 253–77. Huang Jinxing 黃進興 (Huang Chin-shing). Youru shengyu: quanli, xinyang yu zhengdangxing 優入聖域：權力、信仰與正當性. Taibei: Yunchen wenhuashiye gufenyouxian gongsi, 1994. Huang Yifei 黃益飛. Xi Zhou jinwen lizhi yanjiu 西周金文禮制研究. Beijing: Zhongguo shehui kexue chubanshe, 2019. Hwang, Ming-chorng. “Ming-tang: Cosmology, Political Order and Monuments in Early China.” PhD dissertation, Harvard University, 1996. Hymes, Roberts. Statesmen and Gentlemen: The Elite of Fu-Chou, Chiang-Hsi, in Northern and Southern Sung. New York: Cambridge University Press, 1986. ———. “Sung Society and Social Change.” In The Cambridge History of China, Vol. 5, Part Two: Sung China, 907–1279, edited by John W. Chaffee and Denis Twitchett, 526–664. Cambridge, UK: Cambridge University Press, 2015. Ji Xiao-bin 冀小斌. Politics and Conservatism in Northern Song China: The Career and Thought of Sima Guang (A.D. 1019–1086). Hong Kong: Chinese University Press, 2005. Jiang Tao 蔣韜. “Intimate Authority: The Rule of Ritual in Classical Confucian Political Discourse.” In Confucian Cultures of Authority, edited by Peter D. Hershock and Roger T. Ames, 21–47. New York: State University of New York Press, 2006. Jin Chunfeng 金春峰. Zhouguan zhi chengshu jiqi fanying de wenhua yu shidai xinkao 周官之 成書及其反映的文化與時代新考. Taipei: Dongda tushu, 1993. Johnson, David. Spectacle and Sacrifice: The Ritual Foundations of Village Life in North China. Cambridge, MA: Harvard University Press, 2009. Kaneko Shūichi 金子修一. Chūgoku kodai kōtei saishi no kenkyū 中国古代皇帝祭祀の研究. Tōkyō: Iwanami Shoten, 2006. ———, ed. Daito Motomisasagi Gi Chushinshaku 大唐元陵儀注新釈. Tokyo: Kyuko Shoin, 2013. Kang Le 康樂. Congxijiao dao nanjiao: guojia jidian yu beiwei zhengzhi 從西郊到南郊：國家 祭典與北魏政治. Taibei: Daohe chubanshe, 1995. Keightley, David. The Ancestral Landscape: Time, Space, and Community in Late Shang China. Berkeley: University of California Press, 2000. ———. “The Religious Commitment, Shang Theology and the Genesis of Chinese Political Culture.” History of Religions, no. 17 (1978): 211–24. Kenji Morita 森田憲司, “Guanyu zai Riben de shilin guangji zhuben” 關於在日本的事林廣 記諸本. In Xinbian qunshu leiyao shilin guangji, 566–72. Beijing: Zhonghua shuju, 1999.

Bibliography 201 Kertzer, David I. Ritual, Politics, and Power. New Haven, CT: Yale University Press, 1988. Knoblock, John. Xunzi: A Translation and Study of the Complete Works. Stanford, CA: Stanford University Press, 1994. Kojima Tsuyoshi 小島毅. “Kō shi seido no hensen” 郊祀制度の變遷. Tōyō bunka kenkyūsho kiyō 東洋文化研究所紀要 108 (1989): 123–219. Kondō Kazunari 近藤一成. “Wang Anshi de keju gaige” 王安石的科舉改革. In Riben zhongqingnian xuezhe lun zhongguoshi: songyuanmingqing juan 日本中青年學者論中國史： 宋元明清卷, 136–66. Shanghai: Shanghai guji chubanshe, 1995. Kracke, E. A. Civil Service in Early Sung China: 960–1067. Cambridge, MA: Harvard University Press, 1953. ———. “Region, Family and Individual in the Chinese Examination System.” In Chinese Thought and Institutions, edited by John K. Fairbank, 251–68. Chicago: University of Chicago Press, 1967. Kramers, R. K’ung Tzu Chia Yu: The School Sayings. Leiden: Brill, 1950. Kroher, Martin. “‘With Malice Toward None’ to ‘A House Divided’: The Impact of Changing Perceptions of Ritual and Sincerity on Elite Social Cohesion and Political Culture in Northern Song China, 1027–1067.” PhD dissertation, Harvard University, 2014. Kuhn, Thomas. The Structure of Scientific Revolutions. Chicago: The University of Chicago Press, 1970. Kurz, Johannes L. Das Kompilationsprojekt Song Taizongs (r. 976–997). Bern: Lang, 2003. ———. “The Politics of Collecting Knowledge: Song Taizong’s Compilations Project.” T’oung Pao 87:4–5 (2001): 289–316. Laidlaw, James. “On Theatre and Theory: Reflections on Ritual in Imperial Chinese Politics.” In State and Court Ritual in China, edited by Joseph P. McDermott, 399–416. Cambridge, UK: Cambridge University Press, 1999. Lau Nap-yin 柳立言, and Huang K’uan-chung 黃寬重. “Founding and Consolidation of the Sung Dynasty under T’ai-tsu (960–976), T’ai-tsung (976–997), and Chen-tsung (997–1022).” In The Cambridge History of China, Vol. 5 Part I: The Sung Dynasty and Its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 206–78. Cambridge, UK: Cambridge University Press, 2009. Lee, Thomas. Government Education and Examinations in Sung China. Hong Kong: Chinese University Press, 1985. Legge, James, trans. The Sacred Books of China: The Texts of Confucianism. New York: Gordon Press, 1976. Lei Baijing 雷百景, and Li Wen 李雯. “Xihan diling zhaomu zhidu zaitantao” 西漢帝陵昭穆 制度再探討. Wenbo 文博 2 (2002): 48–51. Lei Wen 雷聞. Jiaomiao zhiwai: Suitang guojia jisi yu zongjiao 郊廟之外：隋唐國家祭祀與 宗教. Beijing: Sanlian shudian, 2009. Levine, Ari Daniel. Divided by a Common Language: Factional Conflict in Late Northern Song China. Honolulu: University of Hawaii Press, 2008. ———. “The Reigns of Hui-tsung and Ch’in-tsung (1126–1127) and the Fall of the Northern Sung.” In The Cambridge History of China, Vol. 5, Part I: The Sung Dynasty and Its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 556–643. Cambridge, UK: Cambridge University Press, 2009. Lewis, Mark. The Construction of Space in Early China. Albany: State University of New York Press, 2006.

202 Bibliography Li Boqian 李伯謙. “Cong Jinhou mudi kan xizhou gongmu mudi zhidu de jige wenti” 從晉侯 墓地看西周公墓墓地制度的幾個問題. Kaogu 考古 11 (1997): 51–60. Li, Cho-Ying, and Charles Hartman. “A Newly Discovered Inscription by Qin Gui: Its Implications for the History of Song ‘Daoxue,’” Harvard Journal of Asiatic Studies, 70.2 (2010): 387–448. Li Hengmei 李衡眉. “Lidai zhaomuzhidu zhong shizu chenghu zhiwu lizheng” 歷代昭穆制 度中始祖稱呼之誤厘正. Qiushi xuekan 求是學刊 3 (1995): 95–100. ———. Zhaomu zhidu yanjiu 昭穆制度研究. Jinan: Qilu shushe, 1996. Li Huarui 李華瑞. Wang Anshi bianfa yanjiushi 王安石變法研究史. Beijing: Renming chu­ banshe, 2004. Li Rujun 李如鈞. “Yuduo zaishang: Song Huizong chao de wei yubi zefa” 予奪在上：宋徽宗 朝的違御筆責罰. Taida lishi xuebao 臺大歷史學報 60 (2017.12): 119–57. Li Yumin 李裕民. “Jiekai ‘fusheng zhuying’ zhimi” 揭開“斧聲燭影”之謎. In Songshi xintan 宋史新探, 16–29. Xian: Shanxi shifan daxue chubanshe, 1999. Li Zhen 李震. Zeng Gong nianpu 曾鞏年譜. Su Zhou: Suzhou daxue chubanshe, 1997. Ling Zhenli 林振禮. “Zhu Xi puxu fawei” 朱熹譜序發微. Zhongguo zhexueshi 中國哲學史 1 (2001:7): 62–72. Liu Chengguo 劉成國. Biange zhong de wenren yu wenxue: Wang Anshi de shengping yu chuangzuo kaolun 變革中的文人與文學：王安石的生平與創作考論. Hangzhou: Zhejiang daxue chubanshe, 2011. ———. Jinggong xinxue yanjiu 荊公新學研究. Shanghai: Shanghai guji chubanshe, 2006. ———. Wang Anshi nianpu changbian 王安石年譜長編. Beijing: Zhonghua shuju, 2018. Liu, Hui-Chen, Wang. “An Analysis of Chinese Clan Rules: Confucian Theories in Action.” In Confucianism in Action, edited by David Nivison and Arthur Wright, 63–96. Stanford, CA: Stanford University Press, 1959. ———. The Traditional Chinese Clan Rules. Locust Valley, NY: J. J. Augustin, 1959. Liu, T. C. James (Liu Zijian 劉子健). “How Did a Neo-Confucian School Become the State Orthodoxy?” Philosophy East and West 23:4 (1973): 483–505. 7. ———. “Fengshan wenhua yu Songdai mingdang jitian” 封禪文化與宋代明堂祭天. In Liang Songshi yanjiu huibian 兩宋史研究彙編. Taipei: Lianjing chuban shiye gongsi, 1987. ———. Reform in Sung China: Wang An-shih (1021–1086) and His New Policies. Cambridge, MA: Harvard University Press, 1968. Liu Xiangguang 劉祥光. “Songdai fengshui wenhua de kuozhan” 宋代風水文化的擴展. Taida lishi xuebao 臺大歷史學報 45 (2010.6): 1–78. Liu Yonghua. Confucian Rituals and Chinese Villagers: Ritual Change and Social Transformation in a Southeastern Chinese Community, 1368–1949. Leiden and Boston: Brill, 2013. Lo, Winston. An Introduction to the Civil Service of Sung China: With Emphasis on Its Personnel Administration. Honolulu: University of Hawaii Press, 1987. Loewe, Michael, ed. Early Chinese Texts: A Bibliographical Guide. Berkeley: University of California, 1993. ———. Problems of Han Administration: Ancestral Rites, Weights and Measures, and the Means of Protest. Leiden: Brill, 2016. ———. “The Imperial Way of Death in Han China.” In State and Court Ritual in China, edited by Joseph P. McDermott, 81–111. Cambridge: Cambridge University Press, 1999.

Bibliography 203 Lou Jin 樓勁. “Guanyu Kaibao tongli ruogan wenti de kaocha” 關於開寶通禮若干問題的考 察. In Zhongguo Shehui Kexueyuan Lishi Yanjiusuo Xuekan 中國社會科學院歷史研究 所學刊 4 (2007): 420–30. Beijing: Shangwu yinshuguan, 2007. Louie, Kam. Critiques of Confucius in Contemporary China. New York: St. Martin’s Press, 1980), 84–136. Luo Jiaxiang 羅家祥. Bei Song dangzheng yanjiu 北宋黨爭研究. Taibei: Wenjing chubanshe, 1993. McDermott, Joseph, ed. State and Court Ritual in China. New York: Cambridge University Press, 1999. ———. The Making of a New Rural Order in South China, Vol. 1: Village, Land, and Lineage in Huizhou, 900–1600. Cambridge, UK: Cambridge University Press, 2013. McGrath, Michael. “The Reigns of Jen-tsung (1022–1063) and Ying-tsung (1063–1067).” In The Cambridge History of China, Vol. 5, Part I: The Sung Dynasty and its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 279–346. Cambridge, UK: Cambridge University Press, 2009. McMullen, David L. “Bureaucrats and Cosmology: The Ritual Code of T’ang China.” In Rituals of Royalty: Power and Ceremonial in Traditional Societies, edited by David Cannadine and Simon Price, 181–236. Cambridge: Cambridge University Press, 1987. ———. State and Scholars in T’ang China. Cambridge: Cambridge University Press, 1988. ———. “The Death Rites of Tang Daizong.” In State and Court Ritual in China, edited by Joseph P. McDermott, 150–96. Cambridge: Cambridge University Press, 1999. Meyer, Christian. “Negotiating Rites in Imperial China: The Case of Northern Song Court Ritual Debates from 1032 to 1093.” In Negotiating Rites, edited by Ute Hüsken and Frank Neubert, 99–115. Oxford: Oxford University Press, 2011. ———. Ritendiskussionen am Hof der nördlichen Song-Dynastie (1034–1093): zwischen Ritengelehrsamkeit, Machtkampf und intellektuellen Bewegungen. Sankt Augustin: Institut Monumenta Serica, 2008. Miyakawa, Hisayuki 宮川尚志. “An Outline of the Naitō Hypothesis and Its Effects on Japanese Studies of China.” The Far Eastern Quarterly 14.4 (1955): 533–52. Moser, Jeffrey. “Recasting Antiquity: Ancient Bronzes and Ritual Hermeneutics in the Song Dynasty.” PhD dissertation, Harvard University, 2010. Naitō, Konan 内藤湖南. “Gaikakuteki To-So jidai kan” 概括的唐宋時代觀. Rekishi to chiri 歴史と地理, 9:5 (1922): 1–12. Sadao Nishijima 西嶋定生. “Kōtei shihai no shutsugen” 皇帝支配の出現.” In Chugoku kodai kokka to higashi ajia sekai 中国古代国家と東アジア世界, 370–93. Tokyo: University of Tokyo Press, 1983. Nishioka Ichisuke 西岡市祐. “Shaku Tougyo: Daitō kaigenrei senshin u taibyōrei no sen­ shinbutsu sonoichi” 《大唐開元禮》荐新於太廟禮の荐新物その一. Bulletin of the Sinological Society of Kokugakuin 国学院中国学会報 38 (1992): 74–90. Nylan, Michael. The Five “Confucian” Classics. New Haven, CT: Yale University Press, 2001. Ogata Isamu 尾形勇. Zhongguo gudai de jia yu guojia 中國古代的「家」與國家. Beijing: Zhonghua shuju, 2010. Peng Meiling 彭美玲. “Liang Song huangjia yuanmiao jiqi lisu yiyi qiantan” 兩宋皇家原廟 及其禮俗意義淺探. Chengda zhongwen xuebao 成大中文學報 52 (Mar 2016): 67–114. Puett, Michael. To Become a God: Cosmology, Sacrifice, and Self-Divinization in Early China. Cambridge, MA: Harvard University Asia Center, 2002.

204 Bibliography Qi Xia 漆俠. Wang Anshi bianfa 王安石變法. Shanghai: Shanghai renmin chubanshe, 1979. Queen, Sarah. From Chronicle to Canon: The Hermeneutics of the Spring and Autumn, According to Tung Chung-shu. New York: Cambridge University Press, 1996. Rawski, Evelyn S. Education and Popular Literacy in China. Ann Arbor: University of Michigan Press, 1979. ———. “The Creation of an Emperor in Eighteenth-Century China.” In Harmony and Counterpoint: Ritual Music in Chinese Context, edited by Bell Yung, Evelyn Rawski, and Rubie Watson, 150–74. Stanford, CA: Stanford University Press, 1996. ———. The Last Emperors: A Social History of Qing Imperial Institutions. Berkeley: University of California Press, 1998. Schaberg, David. “The Zhouli as Constitutional Text.” In Statecraft and Classical Learning: The Rituals of Zhou in East Asian History, edited by Benjamin Elman and Martin Kern, 33–63. Boston: Brill, 2010. Schottenhammer, Angela. “Characteristics of Song Epitaphs.” In Burial in Song China, edited by Dieter Kuhn, 253–306. Heidelberg: Edition Forum, 1994. Shanxisheng yongcheng kaogudui 陜西省雍城考古隊, “Fengxiang majiazhuang yihaoji­ anzhuqun yizhi fajiejianbao” 鳳翔馬家莊一號建築群遺址發掘簡報, Wenwu 文物 2 (1985): 1–29. Shaughnessy, Edward. Before Confucius: Studies in the Creation of the Chinese Classics. Albany: State University of New York Press, 1997. Shen Ruiwen 沈睿文. Tangling de buju: kongjian yu jianzhu 唐陵的佈局：空間與建築. Beijing: Beijing daxue chubanshe, 2009. ———. “Xihan dilinglingdi zhixu” 西漢帝陵陵地秩序, Wenbo 3 (2001): 17–23. Shen Songqin 沈松勤. Beisong wenren yu dangzheng 北宋文人與黨爭. Beijing: Renming chubanshe, 1998. Shen Wenzhuo 沈文倬. Zong Zhou liyue wenming kaolun 宗周禮樂文明考論. Hangzhou: Hangzhou daxue chubanshe, 1999. Shu Jingnan 束景南. Zhu Xi nianpu changbian 朱熹年譜長編. Shanghai: Huadong shifan daxue chubanshe, 2014. Smith, Paul Jakov. “A Crisis in the Literati State: The Sino-Tangut War and the Qingli-Era Reforms of Fan Zhongyan.” Journal of Song-Yuan Studies 45 (2015): 59–137. ———. “Shen-tsung’s Reign and the New Policies of Wang An-shih, 1067–1085.” In The Cambridge History of China, Vol. 5, Part I: The Sung Dynasty and its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 347–483. Cambridge, UK: Cambridge University Press, 2009. ———. Taxing Heaven’s Storehouse: Horses, Bureaucrats, and the Destruction of the Sichuan Tea Industry, 1074–1224. Cambridge, MA: Harvard University Press, 1991. Soffel, Christian, and Hoyt Cleveland Tillman. Cultural Authority and Political Culture in China: Exploring Issues with the Zhongyong and the Daotong during the Song, Jin and Yuan Dynasties. Stuttgart: Franz Steiner, 2012. Song Jaeyoon. “Tension and Balance: Changes of Constitutional Schemes in Southern Song Commentaries on the Rituals of Zhou.” In Statecraft and Classical Learning: The Rituals of Zhou in East Asian History, edited by Benjamin Elman and Martin Kern, 252–78. Boston: Brill, 2010. ———. Traces of Grand Peace: Classics and State Activism in Imperial China. Cambridge, MA: Harvard University Asian Center, 2015.

Bibliography 205 Song Xi 宋晞. “Songdai de zongxue” 宋代的宗學, Aoyama Hakushi koki kinen Sōdai shi ronsō 青山博士古稀紀念宋代史論叢, 161–81. Tōkyō: Seishin Shobō, 1974. Steele, John. The I-Li, or Book of Etiquette and Ceremonial. Taibei: Ch’eng-wen Publishing Company, 1966. Sullivan, Lawrence R. “Intellectual and Political Controversies over Authority in China: 1898–1922.” In Confucian Cultures of Authority, edited by Peter D. Hershock and Roger T. Ames, 171–214. New York: State University of New York Press, 2006. Sun Meng 孫猛. Junzhai dushuzhi jiaozheng 郡齋讀書志校證. Shanghai: Shanghai guji chu­ banshe, 1990. Sutton, Donald. “Ritual, Cultural Standardization, and Orthopraxy in China: Reconsidering James L. Watson’s Ideas.” Modern China 33 (Jan 2007): 3–21. Tackett, Nicolas. The Destruction of the Medieval Chinese Aristocracy. Cambridge, MA: Harvard University Asia Center, 2014. Tang Changru 唐長孺. Weijin nanbeichao suitangshi sanlun 魏晉南北朝隋唐史三論. Wuhan: Wuhan daxue chubanshe, 1992. Taylor, Rodney. The Religious Dimensions of Confucianism. New York: State University of New Work, 1990. Teraji Jun 寺地遵. NanSò shoki seiji rekishi kenkyû 南宋初期政治歷史研究. Hiroshima: Keisuisha, 1988. Tozaki Tetsuhiko 戶崎哲彥. “Tōdai niokeru taibyō seido no bensen” 唐代における太廟制度 の変遷. Hikone ronsō 彥根論叢, no. 262–263 (1989): 371–90. Tian, Hao 田浩 (Tillman, Hoyt Cleveland). “Ping Yu Yingshi de Zhu Xi de lishi shijie” 評余 英時的朱熹的歷史世界. Shijie zhexue 世界哲學 (2004:4): 103–7. Tillman, Hoyt Cleveland (Tian Hao). Utilitarian Confucianism: Ch’en Liang’s Challenge to Chu Hsi. Cambridge, MA: Harvard East Asian Monographs, 1982. Tozaki Tetsuhiko 戶崎哲彥. “Tōdai niokeru teikyū ronsōto sono igi” 唐代における禘祫論 爭とその意義. Tōhō gaku 東方學 80 (1990): 82–96. Turner, Victor. The Ritual Process: Structure and Anti-Structure. Chicago: Aldine Publishing Company, 1969. Umehara Kaoru 梅原郁. “Sōdai no on in seido” 宋代の恩蔭制度. In Sōdai kanryō seido kenkyū 宋代官僚制度硏究, 423–500. Kyoto: Dōbōsha, 1985. Von Glahn, Richard. The Sinister Way: The Divine and the Demonic in Chinese Religious Culture. Berkeley: University of California Press, 2004. Wang Aihe. Cosmology and Political Culture in Early China. Cambridge: Cambridge University Press, 2000. Wang E 王鍔. Liji chengshu kao 禮記成書考. Beijing: Zhonghua shuju, 2007. Wang Gungwu. The Structure of Power in North China during the Five Dynasties. Kuala Lumpur: University of Malaya Press, 1963. Wang Kexi 王可喜, and Wang Zhaopeng 王兆鵬. “Nansong ciren Yifu xingniankao” 南宋詞 人易祓行年考. Zhongguo yunwen xuekan 中國韻文學刊 19:4 (Dec 2005): 69–73. Watson, James. “Anthropological Analyses of Chinese Religion.” China Quarterly 66 (June 1976): 355–64. Wechsler, Howard. Offerings of Jade and Silk: Ritual and Symbol in the Legitimization of the T’ang Dynasty. New Haven, CT: Yale University Press, 1985.

206 Bibliography ———. “The Founding of the T’ang Dynasty: Kao-Tsu (Reign 618–26).” In The Cambridge History of China. Vol. 3, Part I: Sui and T’ang China, 589–906, edited by Denis Twitchett and John K. Fairbank, 150–87. Cambridge: Cambridge University Press, 1979. Weller, Robert. “Religion, Ritual, and the Public Good in China.” In Confucianism and Spiritual Traditions in Modern China and Beyond, edited by Fenggang Yang and Joseph B. Tomney, 329–49. Leiden: Brill, 2012. Wen Renjun 聞人軍. Kaogong ji yizhu 考工記譯註. Shanghai: Shanghai guji chubanshe, 1993. Wu Hung 巫鴻. Monumentality in Early Chinese Art and Architecture. Stanford, CA: Stanford University Press, 1995. Wu Liyu 吳麗娛. Zhongji zhi dian: zhonggu sangzang zhidu yanjiu 終極之典：中古喪葬制度 研究. Beijing: Zhonghua shuju, 2012. Wu Wanju 吳萬居. Songdai sanlixue yanjiu 宋代三禮學研究. Taibei: Guoli bianyiguan, 1999. Xia Changpu 夏長樸. “Cong Lixinchuan Daominglu lun Songdaidaoxue de chengli yu fazhan” 從李心傳道命錄論宋代道學的成立與發展. In Songshi yanjiuji 宋史研究集, v.36, 1–63. Taibei: Guoli bianyiguan, 2006. Xi Han lizhi jianzhu yizhi 西漢禮制建築遺址, edited by Xianyang shi wenwu kaogu yanjiu­ suo 咸陽市文物考古研究所. Beijing: Wenwu chubanshe, 2003. Xu Zhibin 許子濱. “Lun zhaomu zhi mingming yiyi” 論昭穆之命名意義. Hanxue yanjiu 漢 學研究 25:2 (2001): 329–46. Xue Mengxiao 薛夢瀟, “Zhouren mingtang de benyi chongjian yu jingxue xiangxiang” 周人 明堂的本義、重建與經學想像, Lishi yanjiu (2015.6): 22–42. Yamauchi, Kōichi 山內弘一. “Hokusō jidai no kō shi” 北宋時代の郊祀. Shigaku zasshi 史學 雜誌 92:1 (1985): 40–66. ———. “Hokusō jidai no shingyoden to keireikyu” 北宋時代の神御殿と景靈宮. Tōhō gaku 東方學 70 (1985): 46–60. ———. “Hokusō jidai no taibyō” 北宋時代の太廟. Jochi shigaku 上智史学 35 (1990): 91–119. Yang Kuan 楊寬. Zhongguo gudai lingqin zhidu yanjiu 中國古代陵寢制度研究. Shanghai: Shanghai guji chubanshe, 1985. Yang Tianbao 楊天保. Jinling wangxue yanjiu: Wang Anshi zaoqi xueshu sixiang de lishikaocha, 1021–1067 金陵王學研究：王安石早期學術思想的歷史考察, 1021–67. Shanghai: Shanghai renmin chubanshe, 2008. Yin Hui 殷慧. Li li shuangzhuang: Zhu Xi lixue sixiang tanwei 禮理雙彰：朱熹禮學思想探 微. Beijing: Zhonghua Shuju, 2019. Yu Yingshi (Yü Ying-shih) 余英時. Zhu Xi de lishi shijie: Songdai shidafu zhengzhi wenhua de yanjiu 朱熹的歷史世界：宋代士大夫政治文化的研究. Beijing: Shenghuo, dushu, xinzhi: Sanlian shudian, 2004. Zhang Guangda 張廣達. “Neiteng hunan de Tang Song biangeshuo jiqi yingxiang” 內藤湖 南的唐宋變革説及其影響. Tang Yanjiu 唐研究 11. Beijing: Beijing daxue chubanshe, 2005): 5–71. Zhang Guangzhi 張光直 (Chang Kwang-chih). Zhongguo qingtong shidai 中國青銅時代. Beijing: Shalian shudian, 1983. Zhang Huanjun 張煥君. “Song dai taimiao zhong de shizu zhi zheng: yi shaoxi wunian wei zhongxin” 宋代太廟中的始祖之爭─以紹熙五年為中心, Zhongguo wenhua yanjiu 中 國文化研究 2 (Summer 2006): 48–56.

Bibliography 207 Zhang Wenchang 張文昌. Zhili yijiao tianxia: Tang Song lishu yu guojia shehui 制禮以教天 下：唐宋禮書與國家社會. Guoli Taiwan daxue wenshi congkan 國立臺灣大學文史叢 刊 vol. 142. Taibei: Guoli Taiwan daxue chubanzhongxin, 2012. Zhang Yuan 張元. “Cong Wang Anshi de xianwang guannian kanta yu Songshenzong de guanxi” 從王安石的先王觀念看他與宋神宗的關係. In Songshi yanjiuji 宋史研究集 v.23, 273–99. Taibei: Guoli bianyiguan, 1993. Zhao Huacheng 趙化成. “Cong Shang Zhou jizhong gongmuzhi dao Qin Han dulilingyu­ anzhi de yanhua guiji” 從商周集中公墓制到秦漢獨立陵園制的演化軌迹. Wenwu 文 物 7 (2006): 41–48. Zhao Kesheng 趙克生. Mingdai de guojia lizhi yu shehui shenghuo 明代的國家禮制與社會生 活. Beijing: Zhonghua shuju, 2012. Zheng Jiali 鄭嘉勵. “Mingzhaoshan chutu de Nansong Lü Zuqian jiazu muzhi” 明招山出土 的南宋呂祖謙家族墓志. In Tang Song lishi pinglun Diyiji 唐宋歷史評論第一輯, edited by Bao Weimin 包偉民 and Liu Houbin 劉後濱, 186–215. Beijing: Shehui wenxian kexue chubanshe, 2015. Zhu Yi 朱溢. “Cong jiaoqiuzhizheng dao tiandifenhe zhizheng: Tang zhi Beisongshiqi jia­ osizhushenwei debianhua” 從郊丘之爭到天地分合之爭──唐至北宋時期郊祀主神 位的變化. Hanxueyanjiu 漢學研究 27:2 (2009): 267–302. ———. Shibangguo zhi shenzhi: tangzhibeisong jili bianqian yanjiu 事邦國之神祇：唐至北 宋吉禮變遷研究. Shanghai: Shanghai guji chubanshe, 2014.

Index

Academy of Scholarly Worthies (jixianyuan 集賢院), 73 An Tao 安燾 (jinshi. 1059), 105 ancestral halls (citang 祠堂), 159, 170, 172, 176–77, 184 ancestral worship, 1, 16, 45, 47, 64–67, 70, 76, 103, 114, 122, 159, 161, 176 ancient sage-kings, 42, 56, 69, 71–73, 76, 78–80, 82–84, 99, 127–28 Annals of Spring and Autumn 春秋: Gongyang Commentary on the Annals 公羊傳, 22n35, 52; Guliang Commentary on the Annals 穀梁傳, 120, 142n41; Zuo Commentary of the Annals 左傳, 49, 98–99, 118 apocryphal texts (weishu 緯書): Junmingjue 鈞命決, Qimingzheng 稽命徵, Liwei 禮 緯, 20, 27, 116 assistant practitioners (zhuji 助祭), 170, 171 table 7.1 bamiao yigong 八廟異宮, 91–92, 92 fig. 4.1 black column (heiying 黑楹), 94 black threshold (xuankun 玄閫), 94 Book of Rites 禮記, 18n16, 52, 97, 106–7, 110, 116, 121–22, 175; Dazhuan 大 傳, 18, 20, 23, 169, 176n81; Jifa 祭 法, 18–19, 78, 78n89, 122, 147, 175; Jitong 祭統 18, 20, 24, 51, 123, 160–61; Liyun 禮運, 177; Tangong 檀弓, 100, 101n82; Wangzhi 王制, 18–20, 22, 26, 33, 42n25, 48n46, 54, 122–23, 147, 170, 172, 175; Yueling 月令, 142, Zengziwen 曾子問, 22n35

Book of Songs 詩經, 15, 68, 94, 106 branches (fang 房), 175, 177–180; adoption between male members of different branches (guofang 過房), 180 Bureau of Ritual Deliberation (yili ju 議禮 局), 124, 126–27 Cai Bian 蔡卞 (1048–1117), 73, 88, 113 Cai Jing 蔡京 (1047–1126), 77, 77n83, 124 Cai Que 蔡確 (1037–1093), 73, 89, 105 Cai Yong 蔡邕 (132–192), 22–24, 26, 36 Cao Shuyuan 曹叔遠 (1159–1234), 164 carved screen (shuping 疏屏), 94 censor (yishi 御史), 47, 57 central yard (titang 提唐), 94 Chen Fuliang 陳傅良 (1137–1203), 164 Chen Gao 陳高 (1314–1366), 177n85 Chen Ji 陳汲, 164 Chen Jing 陳京, 35–36 Chen Lü 陳旅 (1287–1342), 177 Chen Wang 陳汪, 164 Chen Xiang 陳襄 (1017–1080), 75–76, 87–89, 91, 91n41, 92n42, 93 Chen Xiangdao 陳祥道 (1053–1093), 11, 108 table 5.1, 115–121, 123, 129, 147, 165–66 Chen Yang 陳暘 (1064–1128), 108 table 5.1 Chen Yi 陳繹 (1021–1088), 73, 75, 79, 81 table 3.1 Chen Yuanjing 陳元靚, 172–73, 174 illu. 7.1 Chen Zhenjie 陳貞節, 32, 49n50 Chunqiu fanlu 春秋繁露, 24, 24n44 clan cemetery burial (zuzang 族葬), 180–81, 181n101

Index 209 collective advisory meeting (jiyi 集議), 41, 41n17, 49, 51, 68, 159 Commission of Ritual Affairs (taichang liyuan 太常禮院), 39–41, 39n5, 46, 48–49, 51–53, 55, 58, 67–68, 73, 76, 79 Comprehensive Commentary of the Explanation of the Rites and Ceremonies 儀禮經傳通解, 135–36, 135n11, 136n14, 144, 145n51, 147–48, 151–52, 156–57, 160–61, 163–65, 167, 184; Sacrificial Rites: An Extension on the Comprehensive Commentary of the Explanation of the Rites and Ceremonies 儀禮經傳通解續卷祭禮, 136, 140, 142, 144, 146–49, 154 containers of temple offerings: bian 籩, dou 豆, gui 簠, and fu 簋, 171 table 7.1 conventional learning (suxue 俗學), 128 court gentlemen for ceremonies (fengli lang 奉禮郎), 41, 41n16 court gentleman of temple fasting (taimiao zhailang 太廟齋郎), 40–41, 41n16 Court of the Imperial Clan (zongzheng si 宗 正寺), 27, 39, 55 Court of Imperial Sacrifices (taichang si 太 常寺), 27, 32, 39–41, 54, 67–69, 85, 115, 133, 159, 171 table 7.1 crop offerings before the day of sacrifice (shengshengqi 省牲器), 46 Dadai liji 大戴禮記, 142, 144, 146–47, 149, 160 Daizong of Tang 唐代宗 (r. 762–779), 33, 34n92 dan altar壇, 18, 118, 153, 175 Daoxue 道學, 4, 12, 53, 129, 133–35, 139, 147–48, 157–58, 160, 162–64, 170, 172–73, 176, 181, 184 daqiu yi 大裘, 94, 94n56, 108 table 5.1 decorated square timber (fuge 復格), 94 Department of Prescribed Altar and Temple Rites (DPATR, taichang jiaomiaofengsi xiangding liwensuo 太常郊廟奉祀詳定 禮文所), 85–87, 89–96, 102–3 Department of State Affairs (shangshu sheng 尚書省), 49, 68

descent group records (zupu 族譜), 167, 169, 177–82; great descent group (dazong 大宗), 175, 175n74; lesser descent group (xiaozong 小宗), 175 di sacrifice 禘, 16—17, 17n8, 28n66, 35, 58, 66, 96, 116, 134, 140, 169, 172; spirit tablets in the di sacrifice, 58, 66, 117 illu. 5.1 differentiating surnames (bie xing 別姓), 170, 177; major surnames (zhengxing 正姓) and minor surnames (shuxing 庶 姓), 176 dignity of ancestors (zunzun 尊尊), 121 Dilixinshu 地理新書, 168 Dong Zhongshu 董仲舒 (179–104 BC), 24 dugong 都宮: separate configuration of temples, 54; temple’s wall, 139. See also wei 壝 Du Yu 杜預 (222–285), 99 Du Zheng 度正 (1166–1235), 159 Duke Wen of Lu 魯文公, 121 Duke Xi of Lu 魯僖公, 121 Duke Zhuang of Lu 魯莊公, 121 eight state pillars (ba zhuguo 八柱國), 31 eight trigrams (bagua 八卦), 92 Empress Dowager Liu 劉太后 (969–1033), 53 Empress Dowager Xuanren 宣仁太后 (1032–1093), 123–24 Empress Dowager Cao 曹太后 (1016– 1079), 57 encyclopedic collections of daily-use rites (rirong leishu 日用類書), 169, 172–73, 182 entrance door of the temple (beng 閍), 147 erudites of taichang institutions (taichang boshi 太常博士), 40, 42, 115. See also Commission of Ritual Affairs; Court of Imperial Sacrifices Erya 爾雅, 147 examining editor (jiantao 檢討), 75, 127 Exemplar Ancestor (zong 宗), 34, 36, 49–50 expositor in attendance (sijiang 侍講), 94 Fang Dazong 方大琮 (1183–1247), 179–81

210 Index Fang Hong 方紘, 179 Fang Que 方愨, 121–23, 129, 138 Fan Zuyu 范祖禹 (1041–1098), 115 family precepts (jiafan 家範), 169–70, 171 table 7.1, 172, 182, 184 family shrine (jiamiao 家廟), 38n3, 90, 127, 150, 163, 167 filial-piety discourse, 15, 33–37, 52–55, 57–59, 65, 69, 76, 97, 100, 121, 127, 139, 161, 183–88. See also Primal Ancestor Finance Planning Commission (zhizhi sansi tiaolisi 制置三司條例司), 65 five mourning grades (wufu 五服), 65 Founding Ancestor (zu 祖), 34, 36, 47, 49, 52, 57, 69–70 fresh food offering (jianxin 薦新), 24n44, 45, 165 fu ritual of attaching spirit tablets in a temple 祔廟, 100–102, 100n79, 118n59, 120 Gang of Four (siren bang 四人幫), 188 Gaozong of Tang 唐高宗 (r. 649–683), 32 Gaozong of Song 宋高宗 (r. 1127–1162), 133–34 genealogical sequence (shici 世次), 99, 153, 180 Geng Nanzhong 耿南仲 (jinshi. 1082), 113 Gong Yuan 龔原, 108 table 5.1, 113 gongyang scholarship 公羊學, 22n35, 24, 36 grand councilor (pingzhang zhengshi 平章 政事), 63, 77–78, 89, 162 grand mentor (taifu 太傅) Great Office of Clan Affairs (da zongzheng si 大宗正司), 39 Great Ritual Controversy (dali yi 大禮議), 185 great supplicators (taizhu 太祝), 41, 41n16 Gu Tang 顧棠, 113 Guangwu of Han 漢光武帝 (r. 25–57), 25–26, 91 Gujin hebi shilei beiyao 古今合璧事類備 要, 173 Guo Pu 郭璞 (276–324), 147 Guo Zicong 郭子從, 157

Han Imperial Temple of Gaozu 漢高祖廟, 22–23, 25, 91. See also Imperial Temple Han Qi 韓琦 (1008–1075), 57, 77–79 Han Wei 韓維 (1017–1098), 66–68, 77, 79–82, 81 table 3.1 Han Tuozhou 韓侂胄 (1152–1207), 161 Han Yu 韓愈 (768–824), 34–37 He Xiu 何休 (129–182), 24 He Xun 賀循 (260–319), 51n62 He Xunzhi 何洵直 (jinshi. 1078), 86–89, 92, 92n42, 95–104, 108 table 5.1, 110, 112, 114, 121, 128, 138, 151, 153–55, 157, 183 He Zhengchen 何正臣, 93 head of temple chambers (taimiao shizhang 太廟室長), 40 Heavenly Way (tiandao 天道), 70 Holy Ancestor of Song (shengzu 聖祖, Zhao Xuanlang 趙玄朗), 45, 127n92 Hou Ji 後稷, 65–66, 68, 99, 152–53, 156 fig. 6.4 Hu Ciyan 胡次焱, 180 Hu Yuan 胡瑗 (993–1059), 77 Huang Du 黃度 (1138–1213), 164, 166, 166n37 Huang Gan 黃榦 (1152–1221), 135–36, 147–49, 151–52, 156, 160 Huang Lü 黃履 (1030–1101), 86–89, 105 Huang Shang 黃裳 (1044–1130), 124 Huang Yizhou 黃以周 (1828–1899), 120n59, 186 Huang Zhongyuan 黃仲元 (1231–1312), 177 Hui Dong 惠棟 (1697–1758), 186 Huizong of Song 宋徽宗 (r. 1100–1125), 12, 89, 110, 113, 124–29, 184 human Way (rendao 人道), 70 Imperial Archives (mige 秘閣): auxiliary of the Imperial Archive, 53; sub-editor of the Imperial Archive, 67, 89n24 imperially composed edicts (yubi 御筆), 126 Imperial Temple (taimiao 太廟), 1–6, 9, 12, 15, 17–43, 45–47, 50, 53–59, 63, 65–67, 69–70, 72, 77, 79, 81–83, 85–86, 90,

Index 211 91–95, 97–98, 100, 102–4, 106, 110–16, 118, 120–24, 127–29, 133–40, 142–44, 147–51, 153–54, 157–61, 163, 165–70, 171 table 7.1, 172–73, 175, 181–88; main chamber (qin 寢), 18, 22, 32, 37, 55n79, 58, 94, 127, 137, 140, 142–44, 149, 150 fig. 6.3, 155–57, 165, 172–73; temple doors, 22n37, 94, 140, 145–47 inverse sacrifice (nisi 逆祀), 120–21 Ji Ben 季本 (1485–1563), 185–86 Ji Ling 季陵, 133 Jia Gongyan 賈公彥, 43, 114n37, 170n59 Jiangdu jili 江都集禮, 49 Jiaosi lu 郊祀錄, 51 Kaibao tongli 開寶通禮, 43, 43n30, 51, 87n10 Kaiyuan li 開元禮, 28n64n66 Kaogong ji 考工記, 2, 2n7, 109, 147–48 King Cheng of Zhou 周成王, 98, 154, 156 fig. 6.4 King Kang of Zhou 周康王, 98, 101, 155, 156 fig. 6.4 King Wen and King Wu of Zhou 周文 王, 周武王, 19, 21, 36, 65, 83, 98–99, 98nn69–70, 118, 152, 154–56, 156 fig. 6.4 King Xingxian of Ming 興獻王, 186 King Zhao of Zhou, 98, 155, 156 fig. 6.4 Kong Yingda 孔穎達 (574–648), 49–50, 116, 152, 177n82 lanes within the temple (chen 陳), 147 Li Bao 李寶, 133 Li Bing 李昞, 28, 34 Li Cun 李存 (1281–1354), 178 Li Ding 李定, 88 Li Hu 李虎 (d. 551), 28, 30–37 Li Jiahui 李嘉會, 164 Li Qingchen 李清臣 (1032–1102), 87, 90, 90n34 Li Rugui 李如圭 (jinshi. 1193), 144–45, 144n51, 150 fig. 6.3 Li Yuan 李淵 (566–635), 28, 30, 34, 36n100 Li Zhi 李廌 (1059–1109), 38

Li Zongne 李宗訥, 48 Lizong of Song 宋理宗 (r. 1224–1264), 159 lineage and generations (xishi 繫世), 114 Lixiang 禮象, 108 table 5.1 Liang Tao 梁燾 (1034–1097), 68, 77, 79, 81 table 3.1, 83, 105 Liji faming 禮記發明, 107, 108 table 5.1 Liji jiangyi 禮記講義, 108 table 5.1 Liji jie 禮記解 (Lu Dian), 97–98, 108 table 5.1, 165; Liji jie (Fang Que), 108 table 5.1; Liji jie (Ma Ximeng), 108 table 5.1; Liji jie (Yang Xun 楊訓), 108 table 5.1 Liji jishuo 禮記集說, 11, 97–98, 109 table 5.1, 120n61, 140n37, 150 fig. 6.3, 162n18 Lijing yaoyi 禮經要義, 107, 108 table 5.1 Lili xiangjie 禮例詳解, 108 table 5.1 Lilun 禮論, 108 table 5.1 Lishu 禮書, 11, 108 table 5.1, 115–16, 115n41, 117 illu. 5.1, 118, 119 illu. 5.2, 121, 147, 165 Lin Biao 林彪 (1907–1971), 188 Lin He 林荷, 164 Liu Anshi 劉安世 (1048–1125), 88 Liu Chang 劉敞 (1019–1068), 52, 58–59 Liu Mian 柳冕 (ca. 730–ca. 804), 56, 56n87 Liu Xin 劉歆 (ca. 50BC–AD 23), 101–2 Liu Ying 劉迎, 164 Lu Bian 盧辯 (d. 557), 146 Lu Dian 陸佃 (1042–1102), 86, 86n4, 92n42, 93–106, 96 fig. 4.2, 108 table 5.1, 110, 112, 114, 118, 120, 123–24, 138–39, 149, 153–54, 157, 159, 165–66, 183, 185, 188; Lu Dian’s Liji jie 禮記 解 and Liji xinshuo 禮記新說, 97–98, 97n65, 108 table 5.1 Lu Jiuling 陸九齡 (1132–1180), 150–51 Lu Jiuyuan 陸九淵 (1139–1192), 150–51 Lunmeng jizhu 論孟集注, 164 Lü Benzhong 呂本中 (1084–1145), 78n89 Lü Dafang 呂大防 (1027–1097), 81 Lü Dalin 呂大臨 (1044–1091), 173 Lü Gongzhu 呂公著 (1018–1089), 77 Lü Hui 呂誨 (1014–1071), 57 Lü Huiqing 呂惠卿 (1032–1111), 105 Lü Jiawen 呂嘉問, 105

212 Index Lü Zuqian 呂祖謙 (1137–1181), 169–70, 171 table 7.1, 172, 180–81

outer wall of mountain paintings (shanqiang 山牆), 95

Ma Ximeng 馬希孟, 121–22, 129, 138 Ma Zhao 馬昭, 27 Mao Qiling 毛奇齡 (1623–1726), 186 merit-based discourse regarding Primal Ancestor, 15, 35–36, 49, 53, 57–59, 65–69, 80, 103, 114, 129, 183–84, 187; meritorious ancestors, 21, 31, 34, 37, 50, 52, 56n87, 118, 122; merits of ancestors, 3, 11, 70, 73, 137. See also Primal Ancestor Military Commissioner (anfushi 安撫使), 162 mingtang 明堂, 23, 24n42, 42–43, 44 illu. 2.1, 54, 85, 118, 119 illu. 5.2, 116n60 Minor Scribe (xiaoshi 小史), 20, 113–14 monthly offerings (shuoji 朔祭), 45, 134 morning ceremony of pouring the ritual wine (chenluo 晨祼), 46 morning offerings in the temple (chaoxiang 朝享), 45, 127, 134 multiple corridor (chonglang 重廊), 94–95

paths within the temple (tang 唐), 147 Peng Ruli 彭汝礪 (1042–1095), 105 Pi Xirui 皮錫瑞 (1850–1908), 186 personal affection (qinqin 親親), 121, 170 prefect and supervisor of the Commission of Ritual Affairs (zhi liyuan 知禮院/ pan liyuan 判禮院), 40, 48, 67 prefect and supervisor of the Court of Imperial Sacrifices (zhi taichang si 知 太常寺/pan taichang si 判太常寺), 40, 54, 133 Primal Ancestor 始祖/大祖, 18–23, 19n20, 114, 122, 129, 139, 149, 155, 168, 173, 185; Song Primal Ancestor temple/ chamber, 42, 91–92m 92 fig. 4.1, 134, 137–38; Tang Primal Ancestor, 30–37; the 1072 Primal Ancestor debate, 63–70, 72–85, 103, 122, 183; Zhou Primal Ancestor, 152–53 Pu Zongmeng 浦宗孟 (1022–1088), 105 puyi 濮議, 56–57, 56n91

New Learning (xinxue 新學), 103–5, 107, 108–9 table 5.1, 110, 112–15, 118, 121–24, 127–30, 135, 138–39, 147–48, 154, 157–58, 165–66, 183–84, 188 New Policies (xinfa 新法): 3, 58–59, 64n3, 74–75, 77–78, 80, 82, 88–89, 105, 110, 124, 138, 183; Green Sprouts Policy (qingmiao fa 青苗法), 76 Nie Chongyi 聶崇義, 11, 42–43, 44 illu. 2.1, 115, 118 nine libations of wine (jiuxian 九獻), 163 Ningzong of Song 宋寧宗 (r. 1194–1224), 134 novel temple (xinmiao 新廟), 142, 142n41, 144–45

Qin Huitian 秦蕙田 (1702–1764), 47, 186 Qingyuan prohibition of False Scholarship (qingyuan dangjin 慶元黨禁), 135, 162

order of seniority (zunbei 尊卑), 22, 36, 95, 99, 103, 121, 129, 153–56 original lineage (bentong 本統), 65–66, 69, 168

rectification of rituals (zhengli 正禮), 2–3, 39, 175; rectification of ancestors’ ritual status and positions, 69, 75, 77, 80, 82, 137–38, 165, 182 red rafter (danjue 丹桷), 94 regular offerings of daily food (yapan 牙 盤), 94 regular rites (changyi 常儀), 127 remonstrators (jianguan 諫官), 57, 86, 88, 93 removal of temples or temple ancestors (qian 遷), 99–102, 99n74, 118, 144n46, 147 Ren Che 任徹, 42–43 Renzong of Song 宋仁宗 (r. 1022–1063), 39, 45, 47, 53–58, 66, 76, 79, 83, 90, 92, 95, 97, 128

Index 213 Rites and Ceremonies 儀禮, 69, 91, 93, 106–7, 112, 118 ritual duty officials (lizhi 禮直), 41 ritual intent (liyi 禮意), 56, 66, 72, 76, 78–79, 81–82, 96, 98, 125–26, 136, 160, 169 ritual officials (lichen 禮臣), 4, 11, 38–43, 45, 47–49, 53, 58–59, 63, 67–69, 74–75, 78n89, 79–87, 92–93, 95–96, 102–4, 115, 124, 126–27, 157, 159, 161, 166, 171 table 7.1, 182–83 ritual sequence of temple ancestors (miaoci 廟次), 99–100, 102, 153 ritualists (lixuejia 禮學家), 5, 11–12, 52, 58, 85, 104, 139, 147–48, 151, 161, 167, 184–86 Rituals of Zhou 周禮, 18, 18n15, 20–21, 43, 83, 86, 89, 106–7, 110, 113, 125, 125n82, 147–48, 151–52, 170, 181 round altar (yuanqiu 圓丘), 42, 43n27 Ruizong of Tang 唐睿宗 (r. 684–690, 710–712), 32–33, 48, 49n50 sacrifice of stopping wailing (zuku 卒哭), 100, 120n59 Sanjing xingyi 三經新義, 104–5, 104n1, 110, 121, 123; Shangshu xinyi 尚書新 義, 104; Shijing xinyi 詩經新義, 104; Zhouli xinyi 周禮新義, 104, 106, 108 table 5.1, 110–12, 162n18 Sanlitu jizhu 三禮圖集注, 11, 43, 44 illu. 2.1, 115 scholar-officials (shidafu 士大夫), 1–6, 10–12, 32, 37, 40–41, 48, 63, 80, 83n105, 88, 93, 103, 114–15, 127, 134, 137, 167–69, 172–73, 183, 188 seasonal sacrifices (shiji 時祭), 24, 28, 43, 45, 46–47, 53, 91n37, 99, 114, 127, 134, 145n54, 155, 160, 163, 168, 171 table 7.1, 172–73, 174 illu. 7.1, 186n10; laxiang 臘享, 28, 28n64, 43 secret merits (yinggong 陰功), 137 Secretariat-Chancellery (zhongshu menxia 中書門下), 53, 64–65, 69, 75 shan yard 墠, 18, 118, 153, 175 Shao Bo 邵博 (d. 1158), 106

Shao Gu 邵古, 75 Shao Yong 邵雍 (1011–1077), 74–75, 168 Shen Gua 沈括 (1031–1095), 105 Shenyi zhidu 深衣制度, 108 table 5.1 Shenzong of Song宋神宗 (r. 1067–1085), 12, 45, 58, 63–64, 69–70, 73, 77–78, 86–90, 86n4, 93–94, 102, 104, 107, 124, 128, 138, 183 Shilin guangji 事林廣記, 173, 174 illu. 7.1 Shizong of Ming 明世宗 (r. 1522–1567), 186 Shu Dan 舒亶, 73, 105 Shuli xinshuo 述禮新說, 108 table 5.1 Shuyi 書儀, 167–68, 172 Shunzu 順祖 (Zhao Ting 趙珽, 851–928), 42, 69, 90n35, 95n60 side and cross-framed windows (daxiang 達 鄉/jiaoyou 交牖), 94 Sima Guang 司馬光 (1019–1086), 55–58, 66, 73–74, 77, 81n98, 167–68, 172 square altar (fangqiu 方丘), 42, 43n27 Song Chongguo 宋充國, 68, 75–77, 81 table 3.1, 82 Song Minqiu 宋敏求 (1019–1079), 55, 55n80, 67 Song political factions and factionalism, 3–4, 3n11, 12, 63n1, 72–77, 79–84, 87–90, 103 Song Qi 宋祁 (998–1061), 54, 58 Song Shi 宋湜 (948–999), 51 Song Xiang 宋庠 (996–1066), 76 Song Yuan xuean 宋元學案, 4; Gaoben Song Yuan xuean buyi 稿本宋元學案補遺, 74n58 stairs and doorsteps (zuoqi 左墄/xuanbi 玄 陛), 94 Stele of Yuanyou Partisans (Yuanyou dangjibei 元祐黨籍碑), 77, 77n82, 89 Su Shi 蘇軾 (1037–1101), 75, 77, 88, 128 Su Shidan 蘇師旦, 161 Su Song 蘇頌 (1020–1101), 72–73, 87n10 Su Xian 蘇獻, 32, 49n50 Su Xun 蘇洵 (1009–1066), 175 Su Zhe 蘇轍 (1039–1112), 81 Su Zhuo 蘇稅, 68, 81 table 3.1, 82

214 Index subsidiarys imperial ancestral buildings (yuanmiao 原廟), 45, 45n33n37, 127, 165 subsidiary temple or chamber (biemiao 別 廟), 32. See also subsidiary chambers subsidiary chambers (bieshi 別室 or xiangfang 廂房), 48, 54–55, 64, 142, 143 fig. 6.2, 144, 146–47, 149 suburban Altars (jiaosi 郊), 23, 27, 43n27, 51, 85, 137; Altar sacrifices, 17, 28, 34, 43, 46, 50, 70, 81–82, 88, 102 succession of ancient rituals (litong 禮統), 126 succession of the Way (daotong 道統), 126, 126n85 Sun Bian 孫抃 (996–1064), 55–56, 58 Sun Congzi 孫從之, 137 Sun E 孫鄂 (1065–1096), 87, 89 Sun Gu 孫固 (1016–1090), 66–68, 77–82, 81 table 3.1 Sun Pingzi 孫平子, 32 Sun Yu 孫毓, 139–40 Taichangyingeli 太常因革禮, 79, 79n95 Taizong of Song 宋太宗 (r. 976–997), 47–53, 55, 58–59, 82, 92, 95, 97, 128, 134, 153 taizu 太祖: taizu of Tang, 28, 30, 34–35; taizu of Song, 41–42, 46–53, 55n80, 57–59, 65–67, 69–70, 77–78, 80–83, 92, 95, 97, 122, 128, 133–34, 137–38, 153. See also Founding Ancestor Temple of Spectacular Numina (jingling gong 景靈宮), 45, 45n33, 68, 127 textual analysis (kaozheng 考證), 151 Three Dynasties (sandai 三代), 24, 51, 56, 72, 78–82, 88, 90, 128, 136, 138 three offerings on the day of sacrifice (sanxian 三獻), 171 table 7.1 tiao temple 祧廟, 19, 34, 36, 54, 114, 116, 118; tiao-preservation bureau (shoutiao 守祧), 83, 112, 147 towered gateways (taimen 臺門), 94 Two Drafting Groups (liangzhi 兩制), 49–50, 52, 56n90, 65–67, 73; Hanling academicians 翰林學士, 41, 65; edict

drafters of the Secretariat 中書舍人, 41, 65, 73, 75 Vice Minster (xiaozongbo 小宗伯), 20, 113–14, 170 wall-gate (yuanmeng 垣門), 140 Wan Sitong 萬斯同 (1638–1702), 26n58, 186 Wang Anli 王安禮 (1034–1095), 105 Wang Anshi 王安石 (1021–1086), 3, 12, 58–59, 63–65, 64n3, 68–80, 81 table 3.1, 82–84, 86, 88–90, 93, 103–7, 105n5, 108 table 5.1, 110–15, 120–23, 128–30, 134–35, 137–39, 157, 164–66, 168, 183–84 Wang Cun 王存 (1023–1101), 87, 89–90 Wang Ji of Zhou 王季, 98, 98n68, 101, 118, 155–56, 156 fig. 6.4 Wang Jie 王介 (1015–1087), 67–68, 77–80, 81 table 3.1 Wang Li 王禮 (1314–1389), 178 Wang Mang 王莽 (r. 9–23), 25, 179 Wang Pang 王雱 (1044–1076), 105n3, 113 Wang Shouren 王守仁 (1472–1529), 185 Wang Su 王肅 (195–256), 26–27, 33, 112n29, 116, 166 Wang Yan 王炎, 179–80 Wang Yuzhi 王與之 (fl. 1242), 11, 107n18, 111n25, 163–66 Wang Zhaoyu 王昭禹 (fl. 1080), 11, 108 table 5.1, 110, 112–16, 118, 120, 122, 129, 147, 165–66 Wang Yirou 王益柔 (1015–1086), 73–76, 81 table 3.1 wei 壝, 165 Wei Hong 衛宏, 23, 26 Wei Liaoweng 魏了翁 (1178–1237), 176–77 Wei Shi 衛湜 (fl. 1205–1224), 11, 97, 120n61, 140n37, 150 fig. 6.3, 162n18 Wei Xuancheng 韋玄成 (d. 36 BC), 21–22, 26 Wenxian tongkao 文獻通考, 86, 109 table 5.1 Wu Zetian 武則天 (r. 684–705), 30, 30n72

Index 215 xia sacrifice 祫, 22–23, 22n35, 28n66, 34–36, 45, 58, 66, 96, 128, 134, 140, 149; the zhaomu of imperial ancestors in the xia sacrifice, 150 fig. 6.3, 155–57, 176 Xiaozong of Song 宋孝宗 (r. 1162–1189), 134 Xie Jingwen 謝景溫 (1021–1097), 105 Xining yi 熙寧儀, 90 Xizu 僖祖 (Zhao Tiao 趙朓, 828–874): Xizu as the Primal Ancestor, 64–69, 73–77, 79–83, 85, 92–93, 128, 134, 134n6, 137–38, 159, 161; Xizu’s spirit tablet and chamber, 54–56, 58, 70. See also Primal Ancestor Xu Jiang 許將 (1037–1111), 73, 81 table 3.1 Xu qutai li 續曲台禮, 51 Xuanzong of Tang 唐玄宗 (r. 713–756), 33, 49n50 Xuanzu 宣祖 (Zhao Hongyin 趙洪殷, 899–956), 42, 92, 95, 97, 128 Xue Jixuan 薛季宣 (1134–1173), 164–66, 177 Xun Zi 荀子 (313–238 BC), 70, 70n32 Yan Zhenqing 顏真卿 (709–785), 33–34, 36–37 Yang Fu 楊復 (fl. 1228), 136, 140, 141 illu. 6.1, 142, 144–49, 150 fig. 6.3, 151–52, 154, 156–57, 160 Yang Jie 楊傑, 68, 75–77, 76n77, 81 table 3.1, 82, 182n104 Yang Que 楊恪, 164 Yang Shi 楊時 (1053–1135), 53, 58, 138–39, 138n33, 161n7 Yang Wan 楊完, 86–87, 87n9, 109 table. 5.1 Yang Xun 楊訓, 108 table 5.1 Yantie lun 鹽鐵論, 24–25 Ye Zuqia 葉祖洽 (1046–1117), 105 Yi Fu 易袯 (1156–1240), 161–63, 161n11, 165–66 Yi Zhoushu 逸周書, 95 Yili yi 儀禮義, 108 table 5.1 Yin Zhizhang 尹知章 (fl. 669–718), 31, 34, 36 Yin Zhu 尹洙 (1001–1047), 74

Yingzong of Song 宋英宗 (r. 1063–1067), 56–58, 75, 92, 95, 128 Yizu 翼祖 (Zhao Jing 趙敬, 872–933), 42, 92, 95, 97, 128 Yu Ji 虞集 (1272–1348), 115 Yu Li 俞㮚, 127 Yuan Jiang 元絳 (1008–1083), 65–70, 72–75, 81 table 3.1, 168 Yuanfeng regulations on the suburban Altar and temple rituals 元豐郊廟奉祀禮文, 85, 109 table 5.1 Yueshu 樂書, 108 table 5.1 Zeng Bu 曾布 (1035–1107), 72n46, 73, 75, 81 table 3.1, 105 Zeng Zhao 曾肇 (1047–1107), 72n46, 102, 105 Zhang Cun 張純, 26 Zhang Dun 章惇 (1035–1105), 73, 75, 89, 105 Zhang Fangping 張方平 (1007–1091), 58 Zhang Fu 張虑, 136 Zhang Gongyu 張公裕, 68, 81 table 3.1, 83 Zhang Heng 章衡 (1025–1099), 69, 75–76, 81 table 3.1 Zhang Huan 張環, 88 Zhang Qixian 張齊賢 (the Tang ritual official), 31 Zhang Qixian 張齊賢 (942–1014), 48 Zhang Rong 張融 (444–497), 27 Zhang Shangying 張商英 (1043–1121), 105 Zhang Shiyan 張師顏, 68, 81 table 3.1, 83 Zhang Zao 張璪, 86–89, 86n4, 92, 95–103, 112, 121, 155, 183 Zhang Zhao 張昭, 42–43, 58 Zhao Kuangyi 趙匡胤 (r. 960–976). See taizu of Song Zhao Lianggui 趙良規, 55 Zhao Ruyu 趙汝愚 (1140–1196), 134, 135n9 Zhao Shican 趙士虨, 77 Zhao Tingzhi 趙挺之 (1040–1107), 105 Zhao Xiyan 趙希言, 53–54, 58, 90 Zhao Yiruo 趙彥若, 67–68 zhaomu 昭穆, 19–21, 19nn20–21, 25–25n50, 26, 33–34, 48, 50–51, 76, 91, 99n77, 100n79n81, 112–14,

216 Index 118, 120–21, 123, 128, 129, 149–58, 160–61, 166–69, 173, 175–86; the 1079 zhaomu debate, 85–86, 86n5, 93–98, 138, 154, 101–3, 121, 123, 128, 135, 138, 149, 151, 154, 155, 157, 159, 183, 185; secular and spiritual zhaomu, 123; Wuyin zhaomu zang 五音昭穆葬 (guanyu zang 貫魚葬), 168–69 Zheng Boqian 鄭伯謙, 164 Zheng Genglao 鄭耕老 (1108–1172), 175–76, 175n75 Zheng Xuan 鄭玄 (127–200), 19–21, 26–27, 30, 33, 36, 54, 69, 116, 118–120n59, 142, 166, 176n81 Zheng Zongyan 鄭宗顏, 108 table 5.1 Zhenghe wuli xinyi 政和五禮新儀, 11, 46n38, 121, 125–29, 171 Zhenzong of Song 宋真宗 (r. 997–1022), 48–54, 82, 92, 95, 97, 128 Zhezong of Song 宋哲宗 (r. 1085–1100), 124, 128

Zhongxing lishu 中興禮書, 11, 133n3 Zhongzong of Tang 唐中宗 (r. 684, 705–710), 30, 32–33, 48, 49n50 Zhou Bida 周必大 (1126–1204), 162 Zhou Mengyang 周孟陽, 68, 75–77, 81 table 3.1, 82 Zhouguan zongyi 周官總義, 161–63 Zhouli dingyi 周禮訂義, 11, 111n25, 163–66 Zhouli quanjie 周禮全解, 167n41 Zhouli tu 周禮圖, 108 table 5.1 Zhouli xiangjie 周禮詳解, 11, 108 table 5.1, 112, 114–15, 147 Zhujie yili 註解儀禮, 108 table 5.1 Zhu Xi 朱熹 (1130–1200), 1n3, 12, 35, 88n15, 130, 134–40, 141 illu. 6.1, 142–44, 147–49, 150 fig. 6.3, 151–67, 172–73, 175–76, 185, 188 Zhuzi jiali 朱子家禮 (Zhu Xi’s Family Rituals), 1n3, 136, 172–73