The Eternal Present of the Past
HSIAO_f1_i-xix.indd i 3/20/2007 2:10:25 PM China Studies
Published for the Institute for Chinese Studies University of Oxford
Editors Glen Dudbridge Frank Pieke
VOLUME 12
HSIAO_f1_i-xix.indd ii 3/20/2007 2:10:26 PM The Eternal Present of the Past
Illustration, Theater, and Reading in the Wanli Period, 1573–1619
By Li-ling Hsiao
LEIDEN • BOSTON 2007
HSIAO_f1_i-xix.indd iii 3/20/2007 2:10:26 PM Cover Illustration: Min Qiji. The nineteenth illustration of Xixiang ji. Color print. Wuxing: Min Qiji, 1640. Museum für Ostasiatische, Köln. Inv.-No. R61,2 [no. 19]. Photo: Rheinisches Bildarchiv, Köln, Germany.
This book is printed on acid-free paper.
ISSN 1570-1344 ISBN 978 90 04 15643 2
© Copyright 2007 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and VSP.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. printed in the netherlands
HSIAO_f1_i-xix.indd iv 3/20/2007 2:10:26 PM To my Parents, Who Continue the Tradition of Chinese Art and Literature as Puppeteers
HSIAO_f1_i-xix.indd v 3/20/2007 2:10:26 PM HSIAO_f1_i-xix.indd vi 3/20/2007 2:10:26 PM CONTENTS
Acknowledgments ...... ix List of Illustrations ...... xi
Introduction: Theater, Illustration, and Time ...... 1
Chapter One: Toward the Contextualization of Woodblock Illustration: A Critique of Art Historical Method ...... 5
Chapter Two: The Stage or the Page: Competing Conceptions of the Play in the Wanli Period ...... 38
Chapter Three: Performance Illustration ...... 87
Chapter Four: Performance as an Interaction with the Past .... 175
Chapter Five: Image as an Interaction with the Past ...... 202
Chapter Six: Reading as an Interaction with the Past ...... 251
Conclusion: The Role of the Publisher ...... 293
Appendix ...... 299
Glossary ...... 305
Bibliography ...... 313
Index ...... 335
HSIAO_f1_i-xix.indd vii 3/20/2007 2:10:26 PM HSIAO_f1_i-xix.indd viii 3/20/2007 2:10:26 PM ACKNOWLEDGMENTS
This book evolved over a period of sixteen years. It began to take shape in 1990, when I was a master’s degree student in Chinese art at Chi- nese Culture University in Taipei. I continued to test and develop my ideas while a doctoral student in Chinese art and literature at Oxford University and a faculty member at the Universities of Minnesota and North Carolina. During these years, I was extremely fortunate in my teachers, colleagues, and friends, and I bene ted enormously from the experience and the insight of others. Professor Shi Shouqian, later the director of the National Palace Museum in Taipei, encouraged me to take up the study of Chinese book illustration while I was his student at Chinese Culture University. Under his guidance, I acquired basic skills of visual analysis that have served me well ever since. The late Edward Chien, professor of history at Taiwan University, further encouraged me, and set me an example of scholarly rigor and commitment that I have tried—not with complete success—to live up to. His death from cancer in 1996 was a terrible loss to the profession of Chinese studies, as all his students felt at the time and continue deeply to feel. My fellow graduate student Yuecen Zhong cared for Professor Chien during his nal illness. Her sel essness during those painful months remains an inspiring memory. Xu Hong, then the chair of the history department at Taiwan University, and his wife, Wang Zhizhi, professor of history at Fujen University, were extremely supportive of my work and extended to me their warm and unwavering friendship. I value their friendship to this day. My years as a doctoral student at Oxford University were the most enriching of my life. I owe this experience to the teachers, the staff, and the students of the Institute for Chinese Studies. I would par- ticularly like to thank Dr. Jessica Rawson, warden of Merton College, who co-supervised my work in its early stages; Dr. Alison Hardie and Dr. Henrietta Harrison, then my fellow graduate students, with whom I shared many pleasant hours of study and conversation; Dr. Robert Chard, an important mentor in the classroom; and Dr. Taotao Liu, Dr. Shelagh Vainker, Dr. Andrew Lo, and Professor Craig Clunas, who read my doctoral dissertation and provided helpful guidance and criticism. I would particularly like to thank Professor Glen Dudbridge,
HSIAO_f1_i-xix.indd ix 3/20/2007 2:10:26 PM x acknowledgments
who supervised the research of my dissertation. I have tried to follow the example of his commitment to scholarly excellence, but it has been no easy task. This book represents my effort. Its weaknesses are my own, its strengths are largely his. My debt to Professor Dudbridge is incalculable. I had the further good fortune while at Oxford to serve as personal assistant to Professor Michael Sullivan. Over a period of years, I helped Michael catalog his extraordinary collection of contemporary Chinese paintings. Michael’s love of Chinese art—guileless, tireless, sel ess—a lifelong act of service—is something I will not forget. Michael and his lovely wife Khoan took me in and made me feel a member of their fam- ily. Whatever I learned about Chinese painting, Michael and Khoan’s devotion to each other was the great lesson of the innumerable hours I spent in their home. I arrived at the University of North Carolina in 2002. From the very rst, I felt that I had found my home as a teacher and scholar. I thank my departmental colleagues Sahar Amer, Jan Bardsley, Inger Brodey, Mark Driscoll, Miles Fletcher, Eric Henry, Wendan Li, Robin Visser, and Nadia Yaqub, for their encouragement, assistance, and friendship. Gang Yue, chairman of the Department of Asian Studies, has been a staunch supporter of my research method and direction, and an invalu- able source of sound advice. I am lucky to have his assistance and the bene t of his conversation. Hsi-chu Bolick, East Asian bibliographer at Davis Library, has been a help in too many ways to count; she is the ace up my sleeve. I completed this work with the help of fellowships and grants pro- vided by the Institute of Arts and Humanities and the University of North Carolina. I am indebted to both institutions. My husband Dr. David Ross—himself part of the bounty of Oxford—advised me on matters of style throughout. I have him to thank wherever the language of this book contributes to the success of its argument. We hope that this book will one day inspire our daughter, Hsiao-fei, to embark on her own study of Chinese art and literature.
HSIAO_f1_i-xix.indd x 3/20/2007 2:10:26 PM LIST OF ILLUSTRATIONS
1–1. “The Banquet.” In Yipengxue, by Li Yu. N.p.: unknown publisher, 1628–1644. Reproduced from GBXQCK, 3rd series, box 4, vol. 6. Shanghai: Shangwu yinshuguan, 1955–1957 ...... 36 3–1. Liang Chenyu. Yujue ji. Jinling: Fuchun Tang, 1581. Reproduced from GBXQCK, 1st series, box 6, 5:2.1a. Shanghai: Shangwu yinshuguan, 1954 ...... 91 3–2. Shi Hui. Baiyueting ji. Jinling: Shide Tang, 1589. Reproduced from GBXQCK, 1st series, box 1, 10:2.40b ... 91 3–3. Frontispiece of Diamond Sutra, from Dunhuang, 868. British Museum. Reproduced from Zhongguo meishu quanji: huihua bian, 20:2. Taipei: Jinxiu chubanshe, 1989 ... 94 3–4. Panel of a painted wooden screen, from the tomb of Sima Jinlong at Datong, Shanxi Province, 484. Height 81.5 cm. Datong City Museum, Shanxi, China. Reproduced from The Three Perfections, by Michael Sullivan, 21. New York: George Braziller, 1999, c. 1974 ...... 94 3–5. Quanxiang Sanguo zhi pinghua. Jian’an: the Yu family, 1321–23. Naikaku Bunko, Japan. Reproduced from Zhongguo gudian wenxue banhua xuanji, edited by Fu Xihua, 1:18–19. Shanghai: Renmin chubanshe, 1981 ...... 96 3–6. Wang Shifu. Xixiang ji. Beijing: The Yue family, 1498. Reproduced from Zhongguo gudian wenxue banhua xuanji, edited by Fu Xihua, 1:48 ...... 96 3–7. Picture of Four Beauties. Pingyang: The Ji family, 1115–1232. Reproduced from Chgoku kodai hanga ten, edited by Takimoto Kiroyuki, 89. Tokyo: Machida shilitsu kokusai hanga bijutsukan, 1988 ...... 98 3–8. Mural painting in the Hall of King Mingying in the Temple of Guangsheng in Hongdong County in Shanxi Province, 1324. Reproduced from Zhongguo meishu quanji: huihua bian, 13:91. Taipei: Jinxiu chubanshe, 1989 ...... 100
HSIAO_f1_i-xix.indd xi 3/20/2007 2:10:26 PM xii list of illustrations
3–9. Zhu Yu (attributed). Dengxi tu. Ming dynasty (1368–1644). Reproduced from Song Jin Yuan xiqu wenwu tulun, edited by Shanxi Shifan Daxue Xiqu Wenwu Yanjiusuo, 46. Taiyuan: Shanxi renmin chubanshe, 1987 ...... 103 3–10. Wang Shifu. Xixiang ji. Jinling: Qiaoshan Tang, 1592. Reproduced from GBXQCK, 1st series, box 1, 2:1.1a ...... 105 3–11. Chalou tu. In Tangtu mingsheng tuhui, edited by H ky , illustrated by Gyokuzan Shy y, Kumagata Funki, and Higashino Minji, 4.3b–4a. Japan: 1805 ...... 105 3–12. Painting of a theatrical performance. Painted during Guangxu’s reign (1875–1908). Reproduced from Jingju shizhao, edited by Lu Qing et al., 22. Beijing: Yanshan chubanshe, 1990 ...... 107 3–13. Mao’er xi. In Shenjiang shengjing tu, by Zunwen Ge zhuren, painted by Wu Youru, 1884. Reproduced from the reprinted 1884 edition. Taipei: Guangwen shuju, 1981 ...... 108 3–14. Qin bing liuguo pinghua. Jian’an: the Yu, 1321–23. Reproduced from Zhongguo gudian wenxue banhua xuanji, edited by Fu Xihua, 1:12–3 ...... 110 3–15. Wang Zhideng. Quande ji. Jinling: Guangqing Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 3, 7:1.12b–13a. Shanghai: Shangwu yinshuguan, 1955 ...... 112 3–16. Lu Huafu. Shuangfeng qiming ji. Jinling: Shide Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 4, 4:1.2a ...... 112 3–17. Duanfa ji. Jinling: Shide Tang, 1586. Reproduced from Chgoku gikyoku zenpon sanshu, edited by Kanda Kiichiro, 410. Kyoto: Shibun kaku, 1982 ...... 114 3–18. Photo of a scene in the Peking Opera Shizi jingfeng, Shang Xiaoyun as Hu shi and Shang Changrong as the Bandit Jinyan bao. Reproduced from Jingju Shizhao, edited by Lu Qing et al., 141 ...... 116 3–19. Lu Huafu. Shuangfeng qiming ji. Jinling: Shide Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 4, 4:1.28b ...... 116 3–20. Herong ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 1, 6:1.4a ...... 117
HSIAO_f1_i-xix.indd xii 3/20/2007 2:10:27 PM list of illustrations xiii
3–21. Zheng Guoxuan. Baishe ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 5, 6:2.35b ...... 119 3–22. Tang Xianzu. Zixiao ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 10, 2:3.15a ...... 119 3–23. Photo of “The Addressing Sleeve.” Reproduced from Secrets of Chinese Drama, by Cecilia Zung, 85. New York: Arno Press, 1980, c. 1964 ...... 120 3–24. Wang Tingna. Toutao ji. Xiuning: Huancui Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 2, 8:2.13b–14a ...... 120 3–25. Herong ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, 6:2.22a ...... 122 3–26. Zhou Lüjng. Jinqian ji. Jinling: Jizhi Zhai, 1608. Reproduced from GBXQCK, 2nd series, box 3, 4:2.24a ...... 122 3–27. Gao Ming. Pipa ji. Anhui: Wanhu Xuan, 1597. A facsimile of the Jiyi Tang edition. Reproduced from Chgoku kodai hanga ten, edited by Takimoto Hiroyuki, 109 ...... 124 3–28. Zheng Zhiwen. Qiting ji. Jinling: Jizhi Zhai, 1603. Reproduced from GBXQCK, 2nd series, box 3, 9:1.18a ...... 124 3–29. Shen Jing. Shuangyu ji. Jinling: Jizhi Zhai, 1573–1619. Reproduced from GBXQCK, 1st series, box 11, 1:1.29b ...... 126 3–30. Shen Jing. Shuangyu ji. Jinling: Jizhi Zhai, 1573–1619. Reproduced from GBXQCK, 1st series, box 11. 1:2.24a ... 126 3–31. Photo of a scene in the Peking Opera Cimu lei given in 1980s. Reproduced from Jingju shizhao, edited by Lu Qing et al., 198 ...... 128 3–32. Zheng Zhizhen. Mulian jiumu. Anhui: Gaoshi Shanfang, 1582, 1.2a. National Central Library, Taipei ...... 128 3–33. Xu Qiao. Liangjiang ji. Jinling: Jizhi Zhai, 1608. Reproduced from GBXQCK, 2nd series, box 5, 2:1.5 ...... 129 3–34. Gao Ming. Pipa ji. Jinling: Shide Tang, 1573–1619. 1.22a. National Central Library, Taipei ...... 129 3–35. Gao Ming. Pipa ji. Jinling: Shide Tang, 1573–1619. 1.2a. National Central Library, Taipei ...... 131
HSIAO_f1_i-xix.indd xiii 3/20/2007 2:10:27 PM xiv list of illustrations
3–36. Gao Ming. Pipa ji. N.p.: Zunsheng Guan, 1573–1619. 1.3b–4a. National Palace Museum, Taipei, Taiwan, Republic of China ...... 131 3–37. Gao Ming. Pipa ji. Xin’an: Wanhu Xuan, 1597. A facsimile of the Jiyi Tang edition. 1.1b–2a. H sa Bunko, Nagoya ...... 132 3–38. Gao Ming. Pipa ji. Jinling: Jizhi Zhai, 1598. 1.1b–2a. Kokuritsu K bunsho Kan, Tokyo ...... 132 3–39. Sishu ji. In Baneng zoujin. Jianyang: Airi Tang, 1573–1619. Reproduced from Chgoku kodai hanga ten, edited by Takimoto Hiroyuki, 106. Tokyo: Machida shilitsu kokusai hanga bijutsukan, 1988 ...... 133 3–40. Gao Ming. Nan Pipa ji. Anhui: Qifeng Guan, 1573–1619. 1.1b–2a. National Palace Museum, Taipei, Taiwan, Republic of China ...... 133 3–41. Photo of “The Concealing Sleeve,” demonstrated by Mai Lanfang. Reproduced from Secrets of Chinese Drama, by Cecilia Zung, 78 ...... 135 3–42. Wang Shifu. Li Zhuowu xiansheng piping Xixiang ji. Xiling: Tianzhang Ge, 1640, picture 3b–4a. National Central Library, Taipei ...... 135 3–43. Gao Ming. Pipa ji. Jinling: Shide Tang, 1573–1619. 1.5b. National Central Library, Taipei ...... 137 3–44. Gao Ming. Pipa ji. N.p.: unknown publisher, 1573–1619. 1.5b–6a. National Palace Museum, Taipei, Taiwan, Republic of China ...... 137 3–45. Yang Zhijiong. Lanqiao yuchu ji. N.p.: Wanyue Xuan, 1606. Reproduced from GBXQCK, 1st series 1, box 11, 5:2.44b–45a ...... 139 3–46. Tu Long. Tanhua ji. Hangzhou: Weng Wenyuan, 1573–1619. 1.7b–8a. National Central Library, Taipei ... 139 3–47. Gao Ming. Pipa ji. Jinling: Shide Tang, 1573–1619. 2.2b. National Central Library, Taipei ...... 141 3–48. Gao Lian. Yuzan ji. Jinling: Jizhi Zhai, 1599. A facsimile of Guanhua Xuan 1598 edition. Reproduced from GBXQCK, 1st series, box 11, 6:2.10b–11a ...... 141 3–49. Zheng Zhizhen. Mulian jiumu. Jinling: Fuchun Tang, 1573–1619. 3: 1.7a. National Palace Museum, Taipei, Taiwan, Republic of China ...... 142
HSIAO_f1_i-xix.indd xiv 3/20/2007 2:10:27 PM list of illustrations xv
3–50. Zhang Fengyi. Hufu ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 8, 3:2.27a ...... 142 3–51. Photo from a performance of the Peking Opera Gucheng hui. Lin Shusen plays Guan Yu. Reproduced from Jingju shizhao, edited by Lu Qing et al., 76 ...... 143 3–52. Xu Qiao. Liangjiang ji. Jinling: Jizhi Zhai, 1608. Reproduced from GBXQCK, 2nd series, box 5, 2:2.2b ...... 143 3–53. Gao Ming. Pipa ji. Jinling: Shide Tang, 1573–1619. 1.11b. National Central Library, Taipei ...... 146 3–54. Yang Zhijiong. Lanqiao yuchu ji. N.p.: Wanyue Xuan, 1606. Reproduced from GBXQCK, 1st series, box 11, 5:2.37b–38a ...... 146 3–55. Gao Ming. Pipa ji. N.p.: Zunsheng Guan, 1573–1619. 1.17b–18a. National Palace Museum, Taipei, Taiwan, Republic of China ...... 147 3–56. Tang Xianzu. Zixiao ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 10, 2:1.16a ...... 147 3–57. Mural painting from a tomb in Wuling Village in Xinjiang County in Shanxi Province, Yuan dynasty. Reproduced from Song Jin Yuan xiqu wenwu tulun, Shanxi Shifan Daxue Xiqu Wenwu Yanjiusuo, 51. Shanxi: Renmin chubanshe, 1987 ...... 149 3–58. Li Gongling. Xiaojing tu. The Metropolitan Museum of Art. Reproduced from Li Kung-ling’s Classic of Filial Piety, by Richard Barnhart, 120–1. New York: The Metropolitan Museum of Art, 1993 ...... 150 3–59. Liu Dui. Jiaohong ji. Jinling: Jide Tang, 1435. Reproduced from GBXQCK, 1st series, box 2, 10:10a ...... 150 3–60. Wang Tingna. Sanzhu ji. Xiuning: Huancui Tang, 1608. Reproduced from GBXQCK, 2nd series, box 2, 7:2.30b–31a ...... 152 3–61. Tu Long. Xiuwen ji. N.p.: publisher unknown, Late Ming. Picture 1.1a. National Central Library, Taipei ...... 152 3–62. Photo from a performance of the Peking Opera Chen Sanliang patang given in 1980s. Reproduced from Jingju shizhao, edited by Lu Qing et al., 198 ...... 154
HSIAO_f1_i-xix.indd xv 3/20/2007 2:10:27 PM xvi list of illustrations
3–63. Photo of a stage in the Temple of Dongyue in Dongyang Village in Linfen City of Shanxi Province, 1345. Taken by Wang Jianmin, reproduced from Song Yuan xiqu wenwu yu minsu, by Liao Ben, color plate 54 (between pages 200–202). Beijing: Wenhua meishu chubanshe, 1989 ...... 158 3–64. Ji Zhenlun. Wuhou qisheng ji. Jinling: Guangqing Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 5, 4:1b–2a ...... 158 3–65. Li Zicheng chengwang, Ninghe in Hebe Province, mid-Qing. Reproduced from Zhongguo meishu quanji: huihua bian, 21:77. Taipei: Jinxiu chubanshe, 1989 ...... 159 3–66. Mural painting from tomb no. 1 in the town of Baisha in Beiying County in Henan Province, Song dynasty. Reproduced from Song Yuan xiqu wenwu yu minsu, by Liao Ben, color plate 9 (between pages 40 and 41) ...... 159 3–67. Zhuangyuan tukao. N.p.: unknown publisher, 1607. Reproduced from Zhongguo gudian wenxue banhua xuanji, edited by Fu Xihua, 1:355 ...... 161 3–68. Wang Tingna. Toutao ji. Xiuning: Huancui Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 2, 8:1.12b–13a ...... 161 3–69. Lanling Xiaoxiao Sheng. Jinping mei. N.p.: publisher unknown, 1644. Reproduced from Quanben Jinping mei cihua. 2:63.picture 2 (between pages 1764 and 1765). Jiulong: Xianggang Taiping shuju, c. 1982, 1986 ...... 163 3–70. Wang Zhideng. Quande ji. Jinling: Guangqing Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 3, 7:2.25b–26a ...... 163 3–71. Photo of a scene of the Peking Opera Tan qinjia, Liu Gansan as Madam Hu, early 20th century. Reproduced from Jingju shizhao, edited by Lu Qing et al., 14 ...... 166 3–72. Photo of a scene in the Peking Opera Jinshan si, Mei Lanfang as White-snake and Zhu Guifang as Green-snake, early 20th century. Reproduced from Jingju shizhao, edited by Lu Qing et al., 44 ...... 168 3–73. Shen Cai. Qianjin ji. Jinling: Shide Tang, 1573–1619, 4.4b. National Palace Museum, Taipei, Taiwan, Republic of China ...... 172
HSIAO_f1_i-xix.indd xvi 3/20/2007 2:10:27 PM list of illustrations xvii
3–74. Liang Chenyu, Yujue ji. Jinling: Fuchun Tang, 1581. Reproduced from GBXQCK, 1st series, box 6, 5:1.13b .... 172 3–75. Gao Lian, Yuzan ji. Jinling: Jizhi Zhai, 1599. A facsimile of Guanhua Xuan 1598 edition. Reproduced from GBXQCK, 1st series, box 11, 6:2.19b–20a ...... 172 3–76. Zhou Lüjing, Jinqian ji. Jinling: Jizhi Zhai, 1608. Reproduced from GBXQCK, 2nd series, box 3, 4:2.2b ...... 173 3–77. Xu Qiao, Liangjiang ji. Jinling: Jizhi Zhai, 1608. Reproduced from GBXQCK, 2nd series, box 5, 2:1.10a ... 173 5–1. Unknown Artist, attributed to Zhou Wenju. A Court Concert (Heyue tu) or Femail Musicians Playing Before the Emperor. Ming dynasty (1368–1644), early 15th century. Handscroll, ink and color on silk. 41.9 184.2 cm. Buckingham Endowment Fund, 1950.1370 Overall. Reproduction, The Art Institute of Chicago, C20198 ...... 209 5–2. Gu Hongzhong (after). The Night Banquet of Han Xizai (Han Xizai yeyan tu). Ink and color on silk. 28.7 335.5 cm. Palace Museum, Beijing. Reproduced from Zhongguo meishu quanji: huihua bian. 2:128. Taipei: Jinxiu, 1989 ...... 210 5–3. Du Jin. Enjoying the Antiques (Wangu tu). Ink and color on silk. 126.1 187 cm. National Palace Museum, Taipei, Taiwan, Republic of China ...... 210 5–4. Zhou Wenju (after). Emperor Houzhu of the Southern Tang Playing Chess (Houzhu guanqi tu). 11th century. Ink and color on paper. 31.3 50 cm. Freer Gallery of Art ...... 213 5–5. Min Qiji. The nineteenth illustration of Xixiang ji. Color print. Wuxing: Min Qiji, 1640. Museum für Ostasiatische, Köln. Inv.-No. R61,2 [no. 19]. Phto: Rheinisches Bildarchiv, Köln, Germany ...... 216 5–6. Picture of Puppet Show (Kuilei tu). In Sancai tuhui, edited by Wang Qi. A facsimile of 1609 edition, 140a. Shanghai: Guji chubanshe, 1988 ...... 216 5–7. Tang Yin (attributed). Portrait of Yingying. Hou Zhenshang Zhai collection. Reproduced from Comprehensive Illustrated Catalog of Chinese Paintings, Vol. 2: Southeast Asian and European Collections, edited by Suzuki Kei, II-2, S 1–008. Tokyo: University of Tokyo Press, 1982 ...... 242
HSIAO_f1_i-xix.indd xvii 3/20/2007 2:10:27 PM xviii list of illustrations
5–8. Chen Juzhong (attributed). Portrait of Yingying. In Xixiang zalu, edited by Gu Xuanwei, 1.1a. N.p.: Zhongfang shuzhai, 1569 ...... 244 5–9. Tang Yin (attributed). Portrait of Yingying. In Xixiang ji kao, Wanli. Reproduced from Zhongguo gudian wenxue banhua xuanji, edited by Fu Xihua, 1:338 ...... 244 6–1. “Opening Scene” of vol. 1. In Mulian jiumu, by Zheng Zhizhen, 1:1.1a. Anhui: Gaoshi Shanfang, 1582. National Palace Museum, Taipei, Taiwan, Republic of China ...... 259 6–2. “Opening Scene” of vol. 2. In Mulian jiumu, by Zheng Zhizhen, 2:1.1a. Anhui: Gaoshi Shanfang, 1582. National Palace Museum, Taipei, Taiwan, Republic of China ...... 259 6–3. “Opening Scene” of vol. 3. In Mulian jiumu, by Zheng Zhizhen, 3:1.1a. Anhui: Gaoshi Shanfang, 1582. National Palace Museum, Taipei, Taiwan, Republic of China ...... 260 6–4. “Title Page” of vol. 2. In Mulian jiumu, by Zheng Zhizhen, 2:title page. Anhui: Gaoshi Shanfang, 1582. National Palace Museum, Taipei, Taiwan, Republic of China ...... 260 6–5. “Opening Scene.” In Xunqin ji, by Wang Ling. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 3, 4:1.1a ...... 261 6–6. “Opening Scene.” In Yuchai ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 5, 4:1.1a ...... 261 6–7. “Opening Scene.” In Zixiao ji, by Tang Xianzu. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 8, 6:1.1a ...... 262 6–8. “Opening Scene.” In Fenjin ji, by Ye Liangbiao. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 7, 9:1.1a ...... 262 6–9. “Opening Scene.” In Shuangzhong ji, by Yao Maoliang. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 4, 3:1.1a ...... 263 6–10. “Opening Scene.” In Xiangshan ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 1, 3:1.1b ...... 263
HSIAO_f1_i-xix.indd xviii 3/20/2007 2:10:27 PM list of illustrations xix
6–11. “Opening Scene.” In Baishe ji, by Zheng Guoxuan. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 5, 3:1.1a ...... 264 6–12. “Opening Scene.” In Baitu ji. Beijing: Yongshun shutang, 1470s. Reproduced from Ming Chenhua shuochang cihua congkan, edited by Shanghai Wenwu Baoguan Weiyuanhui, 12:1.1a. Beijing: Wenwu, 1979 .... 264 6–13. “Opening Scene.” In Lijing ji. Jian’an: Yu Xin’an, 1566. Reproduced from Sanbun jiryaku, Sento yowa, Reikyoki, 1.1a. Tenri: Tenri Daigaku Shuppanbu, 1980 ...... 265 6–14. “Opening Scene.” In Lizhi ji, by Li Dongyue. N.p.: Yugeng tang, 1581. Reproduced from Mingben Chaozhou xiwen wuzhong, 1.1a. Guangzhou: Guangdong Renmin chubanshe, 1985 ...... 265 6–15. “Opening Scene.” In Jinyin ji, by Su Fuzhi. Jinling: Jizhi Zhai, 1573–1619. Reproduced from GBXQCK, 1st series, box 3, 7:1.1b–2a ...... 266 6–16. “Opening Scene.” In Chundeng mi ji, by Ruan Dacheng. N.p.: Yonghuai Tang, 1633. Reproduced from GBXQCK, 2nd series, box 9, 6:1.1a ...... 266 6–17. “Opening Scene.” In Guanyuan ji, by Zhang Fengyi. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 7, 2:1.1a ...... 274 6–18. “Opening Scene.” In Shiyi ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 1st series, box 5, 2:1.1a ...... 274 6–19. “Zither from the Jade Emperor.” In Herong ji. Jinling: Fuchun Tang, 1573–1619. Reproduced from GBXQCK, 2nd series, box 1, 7:2.25b ...... 277 6–20. Portrait of Yingying. In Xixiang ji, by Wang Shifu. N.p.: Yan Ge, 1630. Reproduced from Zhongguo meishu quanji: huihua bian, 20:125 ...... 285
HSIAO_f1_i-xix.indd xix 3/20/2007 2:10:27 PM
INTRODUCTION
THEATER, ILLUSTRATION, AND TIME
This study draws together the various elements of Wanli culture—illus- tration, painting, theater, literature, philosophy—and examines their interrelation and intersection in the context of the drama publication. In her essay “Printing as Performance,” the scholar Katherine Carlitz suggests the logic of conceiving drama publication as a distinct genre, making the point that “the structure and prosody of drama gave rise to specialized printing conventions,” from which it follows that drama publishing should be studied “as a discrete category, separate from the late Ming publishing boom.”1 This study particularly focuses on drama illustrations because they epitomize most of the important trends and developments in illustration during the Wanli era (1573–1619). What is more, both the quantity and quality of plays published during the period are unmatched in all of Chinese history. During these years, then, drama illustration was at the very center of Chinese culture. Illustration illuminates not only the drama and printing culture, but aspects of late Ming philosophy, religion, morality, and aesthetics. There is thus the opportunity not only to explore the formal language of illustration, but also to reconstruct something of the late Ming worldview. This study is not primarily concerned with market forces, or the advance of technol- ogy, or the celebration of aesthetic achievement, all of which are legiti- mate but by now familiar avenues of exploration. The attempt, rather, is to reconstruct the cultural complexity of drama illustrations—that is, illustrations originally included in editions of printed plays—by recognizing their sometimes obscure and always subtle relation to the traditions of theater, literature, and printing. Only after establishing each illustration in relation to text, edition, and genre—and beyond these in relation to the culture in its full scope—can we approach the full complexity of its expression, for illustrations are carefully crafted to resonate within each of these contexts. In the main, illustrations have been studied as isolated instances of artistic expression, as if analogous
1 Katherine Carlitz, “Printing as Performance,” 269.
HSIAO_f2_1-4.indd 1 3/19/2007 10:23:32 AM 2 introduction
to paintings. The necessity of restoring illustrations to their original context—of conceiving them as essentially contextual—is the most fundamental emphasis of the present study. As drama illustrations are bound up with two different and con ict- ing media—the book and the stage—this study attempts to illuminate how these media responded to each other and collided with each other due to the sudden increase in quantity and quality of printed plays during the Wanli period.2 The collision of these two media brought the relation between printed plays and theatrical performance under the close scrutiny of contemporary literati and drama critics. There was a widespread anxiety, expressed by almost every drama critic in the late Ming, that the popularity of published drama was severing the traditional relation between the play and the stage, and that the new market for published drama was creating a temptation to write for the reader rather than for the actor and the theatergoer. Many literati rightly feared the overthrow of the theater as the dominant cultural and moral institution, and they struggled on numerous fronts to uphold the status of the printed play as a medium rooted in the traditions of the theater, and illustration became the primary eld of ideological battle. The most interesting questions—those with which the present study will largely be involved—are why this rearguard action against the incursions of a purely literary culture was so vehemently undertaken and how the literati went about the attempt to preserve the centrality of the theater within a rapidly burgeoning print culture. The answer to these questions, by no means obvious, has to do with the Chinese conception of the past as a perpetual moral and intellectual example. Scholar Stephen Owen’s Remembrances, a study of the experi- ence of the past in classical Chinese literature, makes all the salient points. If the master gure in Western literary discourse is metaphor, Owen argues that the comparable gure in Chinese literary discourse is “synecdoche, the part that leads to the whole, some enduring frag- ment from which we try to reconstruct the lost totality.”3 In terms of temporal conception, this implies that each moment beckons to all past moments and functions as a means of access to the larger temporal
2 In many American lms produced in the early and mid-twentieth century, the opening credits are framed by a proscenium or by theatrical curtains. This practice also demonstrates a reluctance to abandon the familiar conventions of the theater. When lm found its own footing as an autonomous medium, this practice was dropped. 3 Stephen Owen, Remembrances, 2.
HSIAO_f2_1-4.indd 2 3/19/2007 10:23:32 AM theater, illustration, and time 3
continuum. From the era of Confucius to the late nineteenth century, the impulse to recover and reclaim the past manifested itself equally in all expressions of Chinese culture: literature, art, politics, philosophy. More than anything else, this impulse—by turns denominated ‘Imi- tating the Past’ (Fanggu), ‘Copying the Past’ (Mogu), and ‘Recovering the Past’ (Fugu)—uni ed the culture as a single vast enterprise. But as Owen indicates, this incessant mindfulness of the past created “a gap of time, effacement, and memory” and resulted in a mood of loss and melancholy.4 The perpetual question of Chinese culture is how to bridge this gap. In the literary and artistic elds, especially during the Ming period, the most ambitious and imaginative thinkers pursued a seemingly impos- sible aim: the full recovery of the past, the full con ation of past and present. The anxiety and desperation that colors this attempt has much to do with the political circumstances of the era. The advent of the Yuan dynasty in 1280 constituted a traumatic break with the centuries- old traditions of China. For the rst time the entirety of China (as it was then constituted) came under the rule of invading Mogolian tribes who were not culturally Han (the term Han here connotes a cultural rather than an ethnic identity). Although non-Han northern tribes such as the Northern Dynasties, the Liao, and the Jin had established dynas- ties in the past, the Han had always maintained rule over the south.5 With the advent of the Ming dynasty in 1368 political supremacy was restored to the Han-cultured ruling class. Ming intellectuals naturally felt an urge to bridge the cultural disjunction represented by the Yuan era and to recreate a cultural continuity with the past. The imagina- tive labor of the next two centuries was pervaded by this ambition to ‘restore the past’ ( fugu). The theater was the principal locus of this ambition during the late Ming period. Theater was conceived not as a mere venue of enter- tainment, but as a temporally transcendent space capable of bringing the past to life within the present moment. The assumed interconnec- tion between historical reality and dramatic reality is all-important: it distinguished drama from other literary genres and made it a crucial conduit between the past and the present. This was not merely a matter
4 Ibid., 1–2. 5 One might argue that the Tang dynasty was not ruled by a Han-dominated class. But the Li family was ‘Han-ized’ (Han hua) and accepted as Han during the era of its reign.
HSIAO_f2_1-4.indd 3 3/19/2007 10:23:33 AM 4 introduction
of analogy (the stage is like the past) or memory (the stage recalls the past), but of outright identity, of ontological equation (the stage is the past). This notion is vaguely suggested by some of the era’s rhetorical ourishes. The passage connecting the stage and the backstage, for example, was called ‘Path through the Door of Ghosts’ (Guimen dao). The phrase ‘door of ghosts’ refers to the threshold separating the natural and supernatural worlds, but also, tellingly, to the threshold separating present and past. With similar rationale, the oldest extant bibliography of Yuan drama is titled The Book of Ghosts (Lugui bu). The actors, it is to be inferred, were considered vessels of ghostly posses- sion, and drama itself was understood not merely as a representation or simulacrum of the past, but as a re-enactment of the past, a ghostly reincarnation of the past. There was no sense in which the reincarnated historical episode was considered ontologically inferior or subsidiary either to the rst incarnation of events or to the present reality (in the Chinese tradition, the ‘ghost’ is not a faded residuum, as it is in the West, but rather a fully manifest entity, different but analogous to the living). The historical events brought to life on the stage were considered—strictly speaking—‘real.’ This metaphysical dynamic is so counter-intuitive to the modern metaphysician—not to mention to the modern theatergoer—that it will require a good deal of elaboration and explanation. These larger cultural and metaphysical dynamics explain the sud- den importance and prominence of drama illustration during the Wanli period. As chapter three explains, drama illustration increasingly assumed the trappings of the stage—full frontal address, stylized the- atrical gesture, various framing devices reminiscent of the proscenium. This type of illustration, termed ‘performance illustration’ and discussed in detail in chapter three, initiated a new phase in the tradition of the medium. In its attempt to export the conventions of the stage into the realm of literature, it represents an important development in the struggle to preserve the dominance of the theater and the metaphysics of the theater—what I call the ‘eternal present of the past.’ Though scholars have widely recognized performance illustration as a distinct and de ned phenomenon, its moral, cultural, and metaphysical signi - cance—its implication in the late Ming era’s most profound debate—has gone unnoticed. To recover the full meaning of illustration is, by the logic of synecdoche, to recover the full meaning of this debate and of the late Ming world that hung upon its outcome.
HSIAO_f2_1-4.indd 4 3/19/2007 10:23:33 AM CHAPTER ONE
TOWARD THE CONTEXTUALIZATION OF WOODBLOCK ILLUSTRATION: A CRITIQUE OF ART HISTORICAL METHOD
Woodblock book illustration has long been seen as one of China’s minor arts. It evoked little discussion in the Ming and Qing periods. It was—and is to this day—seen as secondary to the text it illustrates. The common view is that illustration depicts only the most obvious surface actions of the story, and that illustrated images are crude, repetitive, and conventional. In recent years, however, theoretical developments in the eld of art history have inspired a renewed interest in the relation between word and image,1 and the study of illustration has gradually attracted the attention of scholars working in different elds. In Read- ing Illustrated Fiction in Late Imperial China (1998), literary scholar Robert Hegel argues that illustration is no more repetitive and conventional than painting, and that in this respect it is not inconsistent with the aesthetic of the elite literati culture.2 Hegel sees illustration as essentially decorative, its purpose to “capture the richness and fullness of human sensual experience, not just the positions or movements of characters.”3 In Pictures and Visuality in Early Modern China (1997), art historian Craig Clunas argues that illustration embodies important trends in the non- elite visual culture of the late Ming period, and is, at the very least, a major cultural phenomenon.4 These scholars disagree as to whether illustration re ects elite taste or popular culture, but both see illustration as a signi cant part of the attempt to understand the culture of late imperial China, or ‘early modern China’ as some scholars prefer to
1 Norman Bryson’s groundbreaking study Word and Image: French Painting of the ancien régime initiated a new interest in the exploration of the relation between word and image. Inspired by this new trend, Fong Wen and Alfreda Murck edited a volume that explores the same relation in the eld of Chinese art titled Words and Images: Chinese Poetry, Calligraphy, and Painting. 2 Robert Hegel, Reading Illustrated Fiction Fiction, 164–289. Hegel proposes that illustra- tions “were re ective of the arts of the elite strata of society” (290). 3 Ibid., 184. 4 Craig Clunas, Pictures and Visuality, 29–41.
HSIAO_f3_5-37.indd 5 3/19/2007 10:23:47 AM 6 chapter one
call the period from the late-sixteenth century to the mid-seventeenth century.5 This study is inspired by this trend of academic interest but with an important methodological difference: it attempts to examine illustration not merely as image, but as illustration, which is to say, as image embedded in and responding to speci c textual contexts.
For Decoration or for Elucidation?
The conventional approach to the study of illustration is deeply rooted in the precedent set by the study of painting. As many of the scholars who pioneered the study of Chinese woodblock illustration are art historians by training and institutional af liation, it is no wonder that illustration came to be studied on the model of painting, and that the assump- tions that apply to painting—that images are essentially free-standing and self-suf cient, without textual and sequential entanglements—were equally applied to illustration. By emphasizing aesthetic criteria rather than the context of the publication, scholars have overlooked much that is most interesting about illustrations.6 While it is undeniably true that illustration is a branch of graphic art, in many cases imitating the characteristic motifs and styles of painting, illustrations have a very important purpose of their own, which is to illustrate in the sense given by The Oxford English Dictionary: “To elucidate (a description, etc.) by means of drawings or pictures.”7 To remove illustration from its context is to overlook this role of ‘elucidation,’ and to undermine both the meaning and the signi cance of illustration, while holding it to purely aesthetic standards that, for the most part, it does not even attempt to meet. Illustrations provide ornament, to be sure, but their principal
5 Clunas, for example, considers late Ming culture the earliest stage of Chinese modern culture. A review of Clunas’ contribution to the study of late Ming culture by Timothy Brook titled “Picturing Clunas: A Review Essay” can be found in volume 40 of Ming Studies (117–124). Kai-wing Chow likewise associates the late Ming period with early modern China in his study of printing during the 16th and 17th centuries as his book is titled Publishing, Culture, and Power in Early Modern China. 6 There are several scholars who do correlate the illustration and the text. Jan Fontein, for example, uses the text to illuminate the iconography of the illustrations in Wenshu zhinan tuzan and Wuxiang zhishi song, printed in the twelfth century. See “Sudhana’s Pilgrimage in Chinese Art,” in The Pilgrimage of Sudhana, 23–77. Julia K. Murray uses the text to reconstruct the original sequence of illustrations in Nü Xiaojing. See “The Ladies’ Classic of Filial Piety and Sung Textual Illustration,” 95–129. 7 OED, second edition, CD-Rom.
HSIAO_f3_5-37.indd 6 3/19/2007 10:23:48 AM toward the contextualization of woodblock illustration 7
function is to enter into relation with the text. Indeed, an illustration divorced from its accompanying text is no longer an illustration, but a mere print. To read the text without reference to illustrations, or to examine illustrations without reference to the text, is to dissolve the phenomenology of the late Ming publication. For our purposes it is important to note that despite their differ- ent specialties both Hegel and Clunas share this common view that illustration is a branch of the visual arts, and that its primary purpose is to make books more appealing and marketable.8 This study shares Hegel’s premise that illustration is a crucial element in the experience of reading, but takes exception to his unstated assumption that illustration and text are discrete elements, physically juxtaposed but intellectually and aesthetically separable. This study also rejects Clunas’s method of entirely removing illustration from the context of the published book. Both approaches see that illustration can be isolated and studied on its own terms, and that the marriage of text and illustration is more an economic happenstance than an artistic necessity. The tendency to divorce illustration and text is not speci c to Hegel and Clunas, but has been widespread in China and Japan since the early 1960s, when scholars began compiling anthologies of late-Ming and post-Ming illustration. The very concept of the anthology assumes that the mean- ing or value of each illustration is independent of its original context. Among the most famous of these anthologies are Min Shin e-iri hon zuroku (1980), edited by Japanese Sinologist Nagasawa Kikuya; Zhongguo gudai banhua baitu (1984), edited by scholar Zhou Wu; and Zhongguo banhua shi tulu (1988), likewise edited by Zhou. These anthologies include a few illustrations from each of dozens of different publications, spanning the spectrum of literary genres. Other anthologies utilize a more restricted organizing principle. Zhongguo gudian wenxue banhua xuanji (1981), edited by scholar Fu Xihua, for example, exclusively excerpts illustrations that originally appeared in literary works. Zhou Wu’s Huipai banhua shi lunji (1984) exclusively features illustrations in the style of the so-called ‘Hui School,’ and his Jinling gu banhua (1993) limits its selection to material originally published in Nanjing. Many other anthologies of woodblock illustrations were subsequently compiled, but most follow one of these
8 Robert Hegel, Reading Illustrated Fiction, 133. Both scholars approach the book as a material artifact: Hegel emphasizes sizes, font, paper, binding, and even fragrance. Clunas sees illustration as a signi cant expression of the ‘materiality’ that generally characterizes late Ming society.
HSIAO_f3_5-37.indd 7 3/19/2007 10:23:48 AM 8 chapter one
models. In all cases, the anthologies remove illustration from the context of the written work and encourage an approach to illustration that disregards the very idea of context. Consistent with this editorial approach—perhaps even in uenced by it—the study of illustration has been devoted to the analysis of print- ing and engraving techniques and to the assessment of aesthetic merit, always with a focus on the individual illustration. As a function of the dominant aesthetic approach, enormous attention has been given to the Huang family of Anhui. Over the course of the late Ming period, the Huang family created the distinctive style known as ‘Hui School’ and established illustration as a respectable art form. Zhou Wu’s Huipai banhua shi lunji is exclusively devoted to the work of the Huang family and the other engravers who worked in the same style, as is art histo- rians Kobayashi Hiromitsu and Samantha Sabin’s paper “The Great Age of Anhui Printing.”9 The scholar Wang Bomin’s Zhongguo banhua shi (1962) gives substantial space to the Huang family in its discussion of the Ming woodblock print, as well as to famous painters like Chen Hongshou (1599–1652) and Xiao Yuncong (1596–1673) who moon- lighted as illustrators.10 This near-exclusive interest in aesthetics has done a great deal to illuminate the very signi cant contributions of the Huang family and various painters, but it has done nothing to illumi- nate illustration as a broad eld of cultural expression. Moreover, the emphasis on the Anhui engravers has led to an overriding concern with regional styles as part of an attempt to distinguish the Anhui style from its counterparts in Jinling, Hangzhou, and Fujian.11 This methodology assumes that these printing centers were gated communities, when in fact they were knit together by a constant interchange of styles and ideas and by an immersion in a shared artistic tradition. Not only did books circulate from city to city, but illustrators and engravers were frequently offered work away from home. This methodology, then, misrepresents the mobile and dynamic nature of late Ming society.12
9 Kobayashi Hiromitsu and Samantha Sabin, “The Great Age of Anhui Printing,” 25–33. 10 Wang Bomin, Zhongguo banhua shi, 63–79. Zhou Xinhui, Zhongguo gu banhua tongshi: Chen Hongshou: 209–214, and Xiao Yunchong: 252–261. 11 Zhou Wu’s Huipai banhua shi and Jinling gu banhua are examples. 12 Robert Hegel acknowledges regional styles in the early Wanli period, but does not discuss how the styles differ. He agrees, however, that the relocation of skilled craftsmen from one area to another “resulted ultimately in the decline of clear differentiation between regional styles” during the Wanli period, and he attributes this result to the dominance of the Huizhou illustrators and engravers (Reading Illustrated Fiction, 196).
HSIAO_f3_5-37.indd 8 3/19/2007 10:23:48 AM toward the contextualization of woodblock illustration 9
The temptation to emphasize aesthetic and technical criteria is even greater when it comes to the study of color illustration, which rst appeared during the Wanli period. In many cases color prints closely resemble paintings and openly imitate painting technique and subject matter. Color prints are thus particularly amenable to the traditional approach of the art historian.13 Here again, however, art history has projected its disciplinary assumptions in a problematic fashion. Unques- tionably there is a visual analogy between painting and illustration, but this analogy does not get at the essential meaning or function of illustration. Consider, for example, the widespread critical attention that has been given to illustrations that originally appeared in the 1640 edition of Xixiang ji published by Min Qiji (1580–after 1661) in Zhejiang. In a precedent for later decontextualizations, the illustrations were long ago excised from the text and sold as objets d’art, and they now constitute all that remains of the original edition.14 Among the scholars who have been attracted by the visual beauty of this edition are art historians Kobayashi Hiromitsu, who catalogs the motifs of the illustrations,15 and art historian Wu Hung, who mines the illustrations for examples of the ‘metapicture’ (a painting that acknowledges itself as a painting), on which basis he theorizes the self-conscious ctiveness of art in the Chinese tradition.16 In all of these studies, craftsmanship and aesthetic concerns are emphasized to such an extent that the context of illustration—in this case, one of the arch-canonical plays of the Chinese tradition—is ignored. The fact that these illustrations originally appeared amid the complexities and nuances of the play is
13 Artistic primers like Shizhu Zhai shuhua pu utilize color prints to demonstrate paint- ing techniques, suggesting the congruity of the painting and color print traditions. The volume was printed by Hu Zhengyan with a preface dated 1633. It was reproduced by Duoyun Xuan in 1985. For an English catalogue of selected prints from Shizhu Zhai huapu see Joseph Vedlich, ed., The Ten Bamboo Studio. 14 The whole set of illustrations is now in the collection of Museum für Ostasiatische Kunst der Stadt Köln, Köln. Bettina Clever, museum archivist, says that the set origi- nally belonged to the collection of Dr. A. Breuer, but the museum has no information about when or where Dr. Breuer acquired the set. The set is reproduced in Hsi-hsiang chi, chinesische Farbholzschnitte von Min Ch’i-chi 1640. 15 Kobayashi Hiromitsu, “Mindai hanga no sda,” 32–50. Hiromitsu also studies carefully the ve illustrations drawn by Chen Hongshou in a 1639 edition of Xixiang ji published by Zhang Shenzhi. See “Kin Koju no hanga katsud (I & II),” 25–39 and 35–51. He further cooperated with Samantha Sabin to explore how woodblock prints relate to the Anhui (or Xin’an) style of painting during the early Qing period. See “The Great Age of Anhui Printing,” 25–33. The subject matter of his three studies clearly indicates his art-historical perspective. 16 Wu Hung, The Double Screen, 237–59.
HSIAO_f3_5-37.indd 9 3/19/2007 10:23:48 AM 10 chapter one
considered incidental. The scholar Dawn Ho Delbanco, meanwhile, attempts to match each illustration with a corresponding scene in the play. This work begins to see the text and illustration in relation, but without attempting to explain illustration and text in terms of one another. Delbanco concludes that the illustrations are merely decorative, which is to say that their function is essentially aesthetic.17 This study attempts to reverse this common tendency and to set an example by restoring illustration to its necessary context or rather con- texts. An illustration’s immediate context is three-fold: rst, there is the text in which it is embedded (e.g., Pipa ji by Gao Ming, ca. 1306–1359); second, there is the speci c edition in which it is embedded (e.g. the 1610 edition of Pipa ji published by Rongyu Tang); third, there is the literary genre that informs both text and publication (e.g. traditional Chinese drama).
The Text as Context
The relationship between illustration and text—to consider the rst of these contexts—has multiple aspects. Most basically, illustration attempts to represent in visual terms the overt action of the plot, thus lending literary abstraction an apprehensible basis in daily visual experience. In this respect illustration serves what might be called a ‘mimetic function,’ a term used in a strict sense to describe illustration’s depic- tion of textual content. In its mimetic function, illustration powerfully conditions the reception of the text by directing and circumscribing the reader’s freedom of imagination. What is described in the text as a ‘pond,’ for example, becomes something more speci c through the work of the illustrator—say a pond with a small wooden bridge and a pavilion islanded at its center. Deeply relevant to this function is the ‘layout’ of text and image. Layout indicates both the graphic con guration of the individual page and the sequential placement of images throughout the publication. Not surprisingly, illustrations that attempt to ful ll the mimetic function are deliberately laid out to comprise a visual narrative that runs parallel to the text. In this layout the image is displayed in a strip at the top of the page directly above the text it illustrates, making the parallelism between word and image unmistakable. Typically, this parallelism is maintained throughout the
17 Dawn Ho Delbanco, “The Romance of the Western Chamber,” 12–23.
HSIAO_f3_5-37.indd 10 3/19/2007 10:23:48 AM toward the contextualization of woodblock illustration 11
entire publication, with a single illustration at the top of each page or straddling half-folio pages. In Chinese, the term ‘pictures-above-text’ (shang tu xia wen) describes both the placement of the illustration above the text and the placement of such illustrations on each page in a continuous sequence. This arrangement was very popular from the earliest eras of illustration until the early Wanli period. The decline of this convention in the Wanli period suggests that the mimetic function of illustration was challenged, and that a new layout was required to accommodate a new function. The relationship between illustration and text, however, is not limited to the mimetic function. Performance illustration—to take the most signi cant example—entirely dispenses with the mimetic function in an attempt to embed the story within the ontological frame of the stage. Illustration of this type, which takes the liberty of depicting the story with all the trappings of stage performance, in some cases showing the proscenium itself, does not mimic but rather ‘performs’ the text. While mimetic illustration reinforces the plot structure of the play, performance illustration often departs entirely from the plot, seizing on a speci c scene, detail, or moment as somehow deeply symbolic and informative of the whole. In this respect, performance illustration not only borrows the trappings of the theater, but enacts something of the essence of the theater, for traditional Chinese stagecraft is more than anything the art of framing subtly symbolic moments; in this respect Chinese theater is closely analogous to the art of painting. Understand- ing illustration, then, rst depends on grasping its multiple functions and placing each within its necessary historical and cultural context. In the case of performance illustration, this requires contextualizing each image not only within the speci c text and the speci c publication, but also within the larger theatrical tradition. Whatever its function—mimetic or performative—illustration neces- sarily involves interpretation, augmentation, and revision of the text; in short, it necessarily involves commentary. Oftentimes, the illustrator maintains the basic mimetic or performative function while adding symbolic or allusive elements that function as a commentary on the text, bringing to bear the illustrator’s personals views of the characters and the action.18 This phenomenon became especially marked during the late Wanli period, by which time illustration had achieved a degree of
18 For a fuller discussion see Li-ling Hsiao, “Political Loyalty and Filial Piety,” 9–64.
HSIAO_f3_5-37.indd 11 3/19/2007 10:23:48 AM 12 chapter one
respectability and had begun to attract well-known painters who were accustomed to traf cking in interpretive subtleties. In this light it does not go too far to consider illustration a form of literary criticism and the illustrator a literary commentator.19 As early as 1548, the concept of illustration as commentary was spelled out in Yuan Fengzi’s preface to the novel Sanguo zhizhuan, which bluntly af rms that “illustrations comment on the narrative [. . .].”20 The illustrator and publisher Zhong Renjie, who painted and published Sisheng yuan written by Xu Wei (1521–1593), indicates clearly in the preface that “the pictures are meant to invoke the meaning of the play.”21 Like any literary commentator, the illustrator hopes to offer a compelling interpretation and to sway the opinion of the reader. At times the illustrator works in support of the author’s seeming intention, but at other times he disputes the author at every turn, accentuating marginal or subversive details and urging the reader’s attention in new directions. Thus the intentions of the author are quietly, but sometimes powerfully mediated by the illustrator.
The Edition as Context
We may now consider illustration’s second crucial context: the speci c edition of the text, which is to say, the physical entity of the book as published. Just as illustrations must be seen in relation to the text, so too they must be seen in relation to each other, and in relation to the general graphic arrangement of a given published work.22 As a rule
19 Li-ling Hsiao’s paper titled “Reading the Illustrator’s Reading” demonstrates the bene t of approaching illustration from this angle (137–151). 20 Feng, “Preface,” in Sanguo zhizhuan, preface 1a. In Escorial Museum, micro lm at the Harvard Yenching Library. Anne mcLaren, “Constructing New Reading Publics,” 171. 21 Zhong Renjie, “Introduciton to Sisheng yuan,” 864. 22 There are scholars who have attempted to study the illustrations of a single text, but they have not attempted to explore the interrelationship between these illustrations. Fontein, for example, has discussed the illustrations of “The Pilgrimage of Sudhana” in Wenshu zhinan tuzan and Wuxiang zhishi song. Max Loehr has studied the landscape woodblock prints in the Mizang quan, included in the 1108 edition of Tripitaka. See Chinese Landscape Woodcuts. Miya Tsugio has studied the illustrations in Mulian jiumu jing, printed in 1251 in China and then re-engraved and reprinted in 1346 in Japan. See his “Illustrated Scripture of the Story of Mokuren’s Salvation of his Mother,” 155–178. Julia K. Murray has studied illustrations of the life of Confucius in several editions published between the late- fteenth and mid-seventeenth centuries. See her “Illustrations of the Life of Confucius,” 73–134. Yao Dajuin has studied the illustrations in the 1498 edition of Xixiang ji. See his “The Pleasure of Reading Drama,” 437–92.
HSIAO_f3_5-37.indd 12 3/19/2007 10:23:49 AM toward the contextualization of woodblock illustration 13
illustrated books feature multiple illustrations, sometimes as few as ve, sometimes as many as two hundred.23 These illustrations often build and comment on one another, and together embody a uni ed and coherent interpretation of the text. In this respect they are very much like the literary commentary often included in the page margins of late Ming publications. These literary commentaries seem to involve fragmentary or isolated statements that merely respond to the plot, but on closer inspection almost always show a good deal of design and coherence, suggesting the commentator’s reading of the text and the bent of his aesthetic and moral philosophy. In order to arrive at an understanding of this kind of literary commentary, one must correlate each comment with the entire body of comment and develop a sense of the com- mentary as a system of thought. The same approach must be applied to the study of illustration. The 1610 edition of Pipa ji published by Rongyu Tang may be taken as an example. The play is a tragicomedy that centers on the con ict between political loyalty and lial piety. The late Ming commentator seizes on this con ict and relentlessly assaults the text’s assumption that political loyalty is a form of lial piety, and that service to the state ful lls the obligation to the family. Likewise, the illustrator systematically—even exclusively—centers his depiction on symbolic details that privilege lial piety and domesticity.24 In this case, the illustrator no less than the commentator functions as a critic, and the illustration becomes the visual expression of the illustrator’s ‘reading’ of the text. Illustration can thus be understood as a form of criticism: an attempt to form a coherent response to the complexities of a literary work. The reception of the individual illustration is largely determined by the arrangement of the entire series within the speci c publication. As the pictures-above-text layout began to fade in popularity during the late Ming period, publishers turned to alternate layouts such as the ‘pictures-amid-text’ (cha tu) and ‘pictures-before-text’ (guan tu). In the former con guration, illustrations appear on half-folio (dan mian quan ye) or double half-folio pages (shuang mian quan ye) without accompanying
23 The 1498 edition of Xixiang ji includes more than two hundred illustrations, while the 1639 Zhang Shenzhi edition of Xixiang ji includes only ve illustrations; the latter illustrations are by Chen Hongshou. 24 Li-ling Hsiao’s study of Rongyu Tang’s 1610 edition of Pipa ji attempts to demonstrate that both the commentary and illustration in the edition challenge the conceptions of political loyalty and lial piety advanced by the playwright Gao Ming. See “Political Loyalty and Filial Piety,” 9–64.
HSIAO_f3_5-37.indd 13 3/19/2007 10:23:49 AM 14 chapter one
text. Such illustrations appear at regular or irregular intervals throughout the publication. In the latter con guration, illustrations again appear on half-folio or double half-folio pages without accompanying text, but in this case the illustrated pages are clustered at the front of the text and function as prefatory material. The aesthetic and philosophical impli- cations of these different layouts have not been thoroughly addressed by scholars: the crucial point is that each of these layouts suggests a completely different interaction between illustration and text, and each deliberately attempts to induce a different reading experience.25 This attempt to control reader reception may seem tenuous and vague today, but during the late Ming period readers paid very close attention to illustration and allowed meaning to emerge from the tension between image and word. The pictures-above-text layout allows the reader to absorb the word and the image more or less simultaneously. The image and text thus immediately and continuously inform each other during the course of reading. Moreover, illustration guides the reader’s reception of the text at every stage, and the tension between word and image is thus maximized. The pictures-amid-text layout, which entails illustrations interspersed at intervals, poses an entirely different phenomenology. The strict parallel between illustration and text is broken down. In this case, the reading experience no longer involves a simultaneous apprehension of word and image; rather the image occasionally intrudes upon the word. Illustrations of this type are not interspersed randomly, but stra- tegically, in the attempt to keep the reader on an interpretive course at moments when there is a particular temptation to go astray. The logic of each illustration’s placement can be teased out by careful attention to the text: it will be seen to reinforce, or undermine, or emphasize, or complicate, or remind. What such illustrations sacri ce in terms of omnipresence, they make up for in terms of dramatic presence, in each instance taking the reader ever so slightly by surprise. This theoretical understanding, however, is by no means obvious, and by no means universally accepted. Robert Hegel, for example, proposes the pictures- amid-text layout as a device that provides “a break from the text and the con rming view of an illustration after turning a number of pages,”
25 Anne Burkus-chasson has an interesting discussion about how the different forms of traditional Chinese books informed the interpretation of Liu Yuan’s illustration of Lingyan ge. See “Visual Hermeneutics and the Act of Turning the Leaf,” 371–416.
HSIAO_f3_5-37.indd 14 3/19/2007 10:23:49 AM toward the contextualization of woodblock illustration 15
thus encouraging “the poor reader to struggle onward.”26 Hegel’s view suggests that illustrations are rest stops, intermissions analogous to those between the acts of a play, without an integral function in relation to the text. It implicitly de nes illustration in relation to the absence or suspension of the text, and conceives the illustration as purely decorative. This overlooks the likelihood that illustrators, by virtue of a lifetime’s exposure to drama, had the impulse to comment and interpret, and it undermines the intellectual ambition of the medium, as this study hopes to show. The pictures-before-text layout, meanwhile, attempts to predetermine an interpretive course and to x the tone of the reading to come. This layout serves illustration’s attempt to function as commentary on the text by preempting the reader’s formation of opposed ideas or interpretations. This preemptive capacity largely explains why the pictures-before-text layout became dominant during and after the late Wanli period, when illustrators had learned to challenge the text more ambitiously. Taking note of this preemptive dynamic, Robert Hegel discusses a frontispiece that appears in an 868 edition of The Diamond Sutra ( Jingang jing, or Vajra-Prajñâpâramitâ-Sûtra). It shows Subhûti (Xuputi) listening to Buddha. Hegel says, “The picture serves a prefatory func- tion to induce in the reader an appropriately serious frame of mind in approaching the text.”27 Hegel is surely correct that such illustrations attempted to induce a serious frame of mind, but they had more speci c objectives as well. They attempted not only to induce a speci c mood, but also to impose the intellectual and moral premises of the reading experience and to orchestrate in advance a speci c interpretation of the text itself.
Literary Genre as Context
As discussed above, illustration must be understood in relation to the text and in relation to the speci c edition; it must nally be placed within the context of the overarching literary genre, whether poetry, drama,
26 Robert Hegel, Reading Illustrated Fiction, 183. 27 Ibid., 166. Hegel does not expand on this theory but falls back on the view that this layout is more “aesthetically pleasing” (Ibid., 314).
HSIAO_f3_5-37.indd 15 3/19/2007 10:23:49 AM 16 chapter one