Information, Territory, and Networks: The Crisis and Maintenance of Empire in Song China 0674088425, 9780674088429

Citation preview

Information, Territory, and Networks

Harvard East Asian Monographs 388

Information, Territory, and Networks The Crisis and Maintenance of Empire in Song China

Hilde De Weerdt

Published by the Harvard University Asia Center Distributed by Harvard University Press Cambridge (Massachusetts) and London 2015

© 2015 by the President and Fellows of Harvard College Printed in the United States of America The Harvard University Asia Center publishes a monograph series and, in coordination with the Fairbank Center for Chinese Studies, the Korea Institute, the Reischauer Institute of Japanese Studies, and other faculties and institutes, administers research projects designed to further scholarly understanding of China, Japan, Vietnam, Korea, and other Asian countries. The Center also sponsors projects addressing multidisciplinary and regional issues in Asia.

Library of Congress Cataloging-in-Publication Data De Weerdt, Hilde Godelieve Dominique, author. Information, territory, and networks : the crisis and maintenance of empire in Song China / Hilde De Weerdt.    pages cm — (Harvard East Asian monographs ; 388) Includes bibliographical references. ISBN 978-0-674-08842-9 (hardcover : alk. paper)   1.  China—History—Song dynasty, 960–1279.  2.  China—Foreign ­relations—960–1644.  3. China—Boundaries.  4. Imperialism.  5.  Elite (Social sciences)—China—History.  I.  Title.  II.  Series: Harvard East Asian monographs ; 388. DS751.D4 2015 951'.024—dc23 2015005309 Index by David Prout   Printed on acid-free paper Last figure below indicates year of this printing 20 19 18 17 16 15

In Memory of Oka Motoshi

Contents

List of Figures, Maps, and Tables

ix

Preface and Acknowledgments

xv

List of Abbreviations Introduction

xxiii 1

Part I Contemporary Dimensions of Empire: The Court 1 The Dissemination of the Archives and the Formation of the Late Imperial Archival Mentality

35

2 Court Gazettes and Short Reports

76

Part II Transhistorical Dimensions of Empire: The Chinese Territories 3 The Reconstitution of Empire in Empire Maps

107

viii contents

Part III Margins, Borders, and Frontiers 4 Strategic Discourse: Building Frontiers in the Public Domain

167

5 The Multiplexity of Premodern Borders

233

Part IV Imperial Information Networks 6 The Notebook Phenomenon

281

7 Informant Networks and Literati Identities

325

8 Representing the Foreign Other

395



427

Conclusions and Prospects

Appendix 1: Supplementary Tables 439 Appendix 2: A Note on Topical Markup 467 Bibliography 469 Index 501

Figures, Maps, and Tables

Figures 1.1

The process by which archives evolve into different types of archival and historical compilations 40 3.1 “The General Survey Map of Chinese and Non-Chinese from the Past through the Present” 138 3.2 “The Map of the Wei Dynasty and the Northern States” 142 4.1 The hierarchy of strategic places in the Upper Yangzi and Huai regions 187 4.2A–B The subheading “Reunification” under the subject heading “Dealing with non-Chinese” in Liu Dake’s Bishui qunying, with source materials consisting of selected quotations from Liangchao shengzheng 198 4.3A–B Passages in Liangchao shengzheng corresponding to those quoted in Liu Dake’s Bishui qunying 199 4.4 Relationship between the main text and the chronological and topical indexes in Liangchao shengzheng 200 5.1 “The Map of the Empire during the Yuanfeng Reign [1078–85]” 244 6.1 The number of Song biji printed editions, by the century in which they were printed 292 6.2 The distribution of Song biji editions, by type of printer 297

x

6.3

figures, maps, and tables

The number of government imprints of Song biji, by century of printing, showing the subcategories of government printers 299 6.4 Methods used to qualify for officialdom among Song biji authors, by percentage 308 6.5 The number of authors of biji printed in Song times for whom no entry data exist in the China Biographical Database Project (CBDB) 310 6.6 Relationships involved in the production of Wang Mingqing’s Waving the Duster 319 7.1 Wang Mingqing, Waving the Duster: informant network by citation frequency 336 7.2 Sima Guang, A Record of Hearsay from Su River: informant network by citation frequency 340 7.3 Zhang Shinan, Records of Official Travel: informant network by citation frequency 342 7.4A Wang Mingqing, Waving the Duster: author network to 959 344 7.4B Wang Mingqing, Waving the Duster: author network, 1060–1109 345 7.5 Wang Mingqing, Waving the Duster: informant network, 1100–1219 346 7.6 Zhang Shinan, Records of Official Travel: informant network, 1150–1266 347 7.7 Wang Mingqing, Waving the Duster: temporal distribution of entries, by reign period 348 7.8 Zhang Shinan, Records of Official Travel: temporal distribution of entries, by reign period 350 7.9 Topic map for Wang Mingqing’s Waving the Duster 381



figures, maps, and tables

xi

Maps 4.1 4.2 5.1 5.2 5.3 5.4 6.1 7.1 7.2A 7.2B 7.3A 7.3B

Overview of the Southern Song frontier region 182 County and prefectural gazetteers in three Southern Song private collections 222 The distribution of zhai in Northern Song prefectures, based on the Summa of the Military Classics 246 Prefectural and county seat walls in Huainan, based on Hua Yue’s survey 263 Prefectural and county seat walls in the Middle Yangzi region, based on Hua Yue’s survey 264 The region proposed for the construction of ninetyfour mountain stockades in Huainan East 271 Southern Song prefectures where biji were printed 295 Informant network for Sima Guang’s A Record of Hearsay from Su River, showing residential and office addresses 361 Informant network for Wang Mingqing’s Waving the Duster, by residential address 362 Informant network for Wang Mingqing’s Waving the Duster, by office address 363 Informant network for Zhang Shinan’s Records of Official Travel, by residential address 364 Informant network for Zhang Shinan’s Records of Official Travel, by office address 365

Tables 5.1 6.1

Numbers of fortified settlements, by type, in the Loess Plateau Southern Song imprints of biji, by first publication date of the title

247 294

xii

6.2 7.1 7.2 7.3 7.4 7.5 8.1 8.2 8.3 8.4 8.5

8.6

figures, maps, and tables Zhang Hui’s data on the official careers of Song biji authors 309 Informants in officialdom in Sima Guang’s, Wang Mingqing’s, and Zhang Shinan’s networks 354 Proportion of court versus local and regional officeholders among authors and interlocutors in Waving the Duster 356 Survey of informants for whom there is information on their method of first obtaining office 357 Proportion of informants identified by type for whom there is entry information (from Wang Mingqing’s Waving the Duster) 358 Proportion of informants identified by type for whom there is entry information (from Zhang Shinan’s Records of Official Travel ) 358 Frequency of terms used in descriptions of the Jurchen people and the Jin state in Wang Mingqing’s Waving the Duster 399 Frequency of terms used in descriptions of the Song people and the Song state in Wang Mingqing’s Waving the Duster 400 Frequency of occurrence and Z-score for the top-ranked collocates associated with lu (虜) in Wang Mingqing’s Waving the Duster 402 Frequency of occurrence and Z-score for the top-ranked collocates associated with Jin (金) in Wang Mingqing’s Waving the Duster 403 Frequency of terms associated with the Jin state and people in the Siku quanshu edition of Wang Mingqing’s Waving the Duster and in the collated Zhonghua shuju edition 406 Frequency of occurrence of the terms yi 夷, di 狄, man 蠻, and rong 戎 in Wang Mingqing’s Waving the Duster, by text type 407



figures, maps, and tables

xiii

Appendix 1: Supplementary Tables 1 2 3 4 5 6 7 8 9a 9B 9c

List of texts and select art objects donated to the early Southern Song court (1131–59) 440 List of requests for donations of texts by the early Southern Song court (1132–59) 444 Number and size of texts and select art objects donated to the early Southern Song court, by year (1131–59) 448 Number of general and individual requests for donations of texts issued by the early Southern Song court, per year (1132–59) 449 Reports by Song envoys on their journeys to the Xia, Liao, and Jin, in the holdings of three private collectors 450 Xia, Liao, and Jin gazetteers and records in the holdings of three private collectors 454 Histories and memoirs of the loss of the north in the holdings of three private collectors 457 Comprehensive gazetteers in the holdings of four private collectors 458 Thematic overview of xuzhi (“what you need to know”) titles listed in major Song catalogs and bibliographies 460 Total number of xuzhi (“what you need to know”) titles listed in table 9A, by subject area 466 Proportion of xuzhi (“what you need to know”) titles on border affairs in three private catalogs and the Song shi Yiwen zhi 466

Preface and Acknowledgments

I

n one of the first lectures I attended in the late 1980s, I and my eighty or ninety classmates in Chinese studies were told that for us to research the history of modern and contemporary China was like a blind person touching an elephant: our views would vary and change depending on which part of the enormous creature we touched and a systematic overview would remain elusive. The story was based on a well-known parable and the moral ­applies no less to the history of pre-nineteenth-century Chinese history. Since I embarked upon this project, and even now, after having achieved a first milestone, I have often felt I was faced with a grey wall. No doubt some will feel that I have touched too much or that I did not press hard enough before moving on to an adjacent area; I plead guilty on both counts. By electing to present in these chapters an overview of the social production and circulation of information about court proceedings, border policy, and the geography of the Chinese territories, I aim to bring into focus interrelated areas of literati interest whose significance in the larger context of the his­tory of imperial elite identities and Chinese political culture would remain inconspicuous when treated in isolation from each other and from the growing social production and circulation of notebooks, examination manuals, or letters. To trace the connections among texts reporting and commenting on recent dynastic history and current affairs, I have experimented with d ­ ifferent ways of touching and feeling and making inroads. This ­results in an

xvi

preface and acknowledgments

unevenness in scope and method among chapters and, in particular, between the first three parts—which focus on the production of information about places associated with the imperial state—and the last part, which delves into the connections between these texts and the growing numbers of notebooks written and published in the twelfth and thirteenth centuries. By interpreting these texts in the aggregate as well as through individual cases, I hope to have charted itineraries through adjacent grey zones from which historians have retreated even when those areas were in sight. In some ways the present book, Information, Territory, and Networks, builds upon my earlier work on the civil service examinations. When I came across references to and quotations from Song archival compilations and dynastic histories, contemporary notebooks, and a Song historical atlas in examination manuals and encyclopedias compiled for students preparing for the examinations, I was already confronting the question of how prefectural schoolteachers, private tutors, and editors of commercially published textbooks acquired access to such materials. In one or two chapters in this book I have returned to those examination questions and manuals. Whereas my work on the intellectual history of Song examinations can be seen as an attempt to read examination questions and essays in the context of the textbooks that served as curricular guides, this new book started as an effort to dig into the textual layers of the manuals themselves. This round of textual archeology adds new meaning to some of the earlier work; for example, the analysis of historical atlases and military geographies in conjunction with examination materials helps explain how geographical knowledge may have been tested in examinations and why the editors of geographical texts could claim that their works were relevant for students preparing for the examinations. The examinations profoundly shaped Chinese political culture and in ways we have little understood. However, as will be shown throughout the chapters in this book, the cultural and political significance of the production and circulation of information about current affairs extended beyond the examination field into the broader social world of Song literati. The history of this book was very different from that of my earlier book, Competition over Content: Negotiating Standards for the



preface and acknowledgments

xvii

Civil Service Examinations in Imperial China (1127–1279). Competition over Content was to a large extent planned, and it benefited from years of relatively carefree exploration of little-used primary-source materials. When the time came to write acknowledgments, I could only arrange the names of those who had helped in various ways in a list, as in film credits; I had little experience and no words in which to convey heartfelt gratitude. Information, Territory, and Networks, by contrast, proceeded on borrowed time and evolved in col­ laboration with many others. My greatest debt remains to Peter Bol, who suggested I publish my first venture into the production and circulation of archival compilations in 2005. More important, he never saw the need to rein in my inclinations to roam widely even when I was rather conflicted about them. I owe no lesser debt to Cynthia Brokaw, whose pointed comments on my writing and that of others are reflected in much of my work. Several chapters and parts of chapters were first conceived as contributions to conferences, workshops, and panels. I thank staff at the Fairbank Center for Chinese Studies and Wilt Idema for sponsoring not one but two workshops on the early history of printing. I coorganized the first with Joe Dennis in 2005 and the other with Lucille Chia in 2007; needless to say, both workshops and the work that has come out of them would have been impossible without their initiative and collaboration on each occasion. I first discussed the circulation of court archival compilations in the 2005 workshop and benefited from the comments of all participants. I thank Joanna Handlin Smith for having taught me much about editing and revising in the process of having this first work published in the Harvard Journal of Asiatic Studies; her contributions were incorporated in chapter 1. The staff, fellows, and visitors to the Stanford Humanities Center in 2006–7 were a source of inspiration in many ways; much of the research and writing for chapters 3, 4, and 5 was undertaken in their midst. Their enthusiasm and the daily conversations we shared over lunch continued to be a source of motivation beyond our time together. I owe special thanks to Matthew Tiews, John Bender, Nicole Coleman, Robert Barrick, Kären Wigen, Matt Somers, Matt Jockers, Christian Henriot, Christine Guth, Troy Jollimore, Linda Zerilli, Mark Edward Lewis, Zhou Yiqun,

xviii

preface and acknowledgments

William Tronzo, Richard Vinograd, Thomas Mullaney, and the other staff and colleagues who made the Humanities Center a most congenial place to read, think, and write. Whereas the annual meetings of professional organizations had once been the venue at which to condense and present research laid out in long dissertation chapters, in the last seven years they have turned into regular deadlines by which to write up ongoing research. I thank the panel organizers who invited me onto panels to present work in progress at the meetings of the Association for Asian Studies, the American Historical Association, the Social Science History Association, and the Society for the History of Authorship, Reading and Publishing. They are, in chronological order starting in 2005, Han Seunghyun, Michael Chang, Nick Tackett, Joe Dennis, Fei Si-yen, Chen Song, and Anne Chao. Comments from discussants and members of the audience at these panels are gratefully acknowledged here. In addition to those already mentioned, I wish to thank in particular Robert Hymes, Tobie MeyerFong, Arthur Waldron, Ann Waltner, Michael Szonyi, Paul McLean, and John Mohr. Invitations to workshops at the École française d’Extrême-Orient (on noncommercial printing, 2009), Yale University (on conflict and negotiation, 2009), and Harvard University (on Chinese manuscript culture, 2010) were instrumental in shaping my research on the publication of notebooks in manuscript and print (chapters 1, 6, and 7) and on the representation of the SongJin conflict in notebooks (chapter 8). For invitations and feedback I owe much to Michela Bussotti, Jean-Pierre Drège, Christian Lamouroux, Christian Jacob, Roger Chartier, Hans van de Ven, Haun Saussy, Tina Lu, Chloe Starr, Tian Xiaofei, and Li Wai-yee. As the footnotes will attest, I have learned no less from colleagues who let me read and comment on their work; for getting me involved in this way kudos to Ann Blair, Javier Cha, Wang Li, and Matthew Mosca. The questions and comments I received at other workshops and during invited lectures have helped me refine arguments, elaborate where I take too much for granted, and develop new perspectives on familiar materials. I thank my hosts and all those who posed questions during my presentations on social networks, empire maps



preface and acknowledgments

xix

and map reading, book collecting, and empire and information at (in reverse chronological order) National Tsing Hua University (Taiwan); the Karl Jaspers Centre for Advanced Transcultural Studies and the Faculty of Philosophy at the University of Heidelberg; the Foreign and Commonwealth Office (London); the University of Arkansas; the Warburg Institute at the University of London; the East/Southeast Asian History Seminar at the School of Oriental and African Studies (University of London); the Université de ParisDiderot; the Asian Studies Seminar at the University of Cambridge; the Medieval History Seminar, the Department of History, the Oxford e-Research Centre, and the China Research seminar, all at the University of Oxford; the History Honor Society at Illinois Wesleyan University; the History Department at the University of Manchester; the Exzellenzcluster “Gesellschaftliche Abhängigkeiten und soziale Netzwerke” at the University of Trier; the Center for the Humanities and the Public Sphere at the University of Florida; Osaka City University; the Koninklijke Musea voor Kunst en Geschiedenis/ Musées royaux d’Art et d’Histoire in Brussels; Pembroke College (Oxford); the International and PRC Conference for Song Studies at Yunnan University and at Wuhan University; the Fair­bank Center for Chinese Studies at Harvard University; the Academy of East Asian Studies at Sungkyunkwan University; the Bath Royal Literary and Scientific Institute, the University of Warwick; the Book History Research Network at the Institute of En­glish Studies (University of London); the Humanities Center at Stanford University; the American Association for Chinese Studies Conference at Vanderbilt University; the Library History Seminar XI at the University of Illinois at Urbana-Champaign; and the Humanities Initiative at the University of Tennessee at Knoxville. This long list of debts accrued would be woefully incomplete without my acknowledging the material as well as the intellectual support of foundations, research organizations, and home institutions. I am grateful to all who have contributed time, financial, and moral support as well as intellectual input. None should go unmentioned, and certainly not Dagmar Schäfer and the wonderful team of researchers she gathered around her at the Max Planck Institute for the History of Science in Berlin. The six months spent over two

xx

preface and acknowledgments

summer sessions in 2010 and 2011 finally allowed me to bring the bits and pieces that had been lingering around for years together in one manuscript. The spring of 2011 also marked the beginning of a new project that aims to continue the work started in part IV and expand it into a comparative project. The basis for this work was laid in 2008–9, when I was given time to read through notebooks and experiment with digital analysis, largely through the support of the National Endowment for the Humanities (NEH), the Oxford University Press John Fell Fund, the British Academy, and the Chinese Studies subject group within the Faculty of Oriental Studies, Oxford. Special thanks are due to Jim Turner of the NEH for his patience and helpfulness. James Cummings, Sebastian Rahtz, Lou Barnard, and Martin Wynne from Oxford University Computing Services coaxed me into taking on XML-TEI and I am glad they did. I hope that the students who joined in at a later stage feel the same way; Lincoln/Lik Hang Tsui, Chen Yunju, You Zixi, and Li Yun-Chung made the journey into team territory thoroughly enjoyable. In the final stages of editing the manuscript I received valuable help and advice from Ho Hou Ieong; most of the revised figures and maps are his work. From my first steps into the territory of political communication I have been very fortunate to receive regular feedback and support from senior and junior colleagues involved in an international project on the same topic. Deng Xiaonan (Beijing University), Huang Kuanchong (then at Academia Sinica), Hirata Shigeki (Osaka City University), and Christian Lamouroux (École des hautes études en sciences sociales, Paris) set up an informal international network that has met regularly and discussed work on “the new political history”; I cannot thank them enough for the difference they are making by inspiring new scholarship in their own institutions and beyond. I started reading about the histories of news, information, archives, and the socialization of elites at the History Department of the University of Tennessee and, as a student trained predominantly in Chinese studies, gained much from the questions posed by historians working on other parts of the world. Tom Burman, Palmira Brummett, J. P. Dessel, Michael Kulikowski, and Tina Shephard-



preface and acknowledgments

xxi

son had me read and think in new ways. Aleydis van de Moortel helped me rethink relations between imperial cores and local places, and Miriam Levering and Suzanne Wright joined me in a crossdisciplinary series of workshops on the history of printing. In recent years, colleagues at the Oxford Global Middle Ages Initiative have similarly inspired me to think beyond the boundaries of familiar territory; for invitations and discussions I am especially grateful to Catherine Holmes, John Watts, Mark Whittow, Lesley Abrams, Chris Wickham, and Naomi Standen. Time has been too short to follow up on many of the suggestions they have provided. Pembroke College at Oxford became a multidisciplinary setting in which to explore the meaning of concepts and the uses of methods. Mark Fricker (Biology) suggested I set up a workshop on networks at the College and this made possible a visit and formative conversations with John Padgett. Mark also directed me toward clusters of network research at the university. Chen Min (Computer Science) and Eamonn Molloy (Management) shared their interests in humanities visualization and in the history of space. For comparative historical perspectives I turned to Adrian Gregory and Christopher Melchert. Their contributions have meant a lot. For good cheer and raw criticism I must acknowledge in addition Chu Ping-tzu, Ruth Mostern, Karl Gerth, Tia Thornton, and Rana Mitter. My mother, siblings, nieces and nephews, aunts, uncles, and cousins have shown amazing support and enthusiasm for the stubbornness with which I have pursued my interests throughout the years. I realize that this is a gift that should not too easily be taken for granted. I thank Mary and Simon Lucal for their unrelenting love and care and Neula Kerr-Boyle for inserting levity at all times. C. Scott Walker, Digital Cartography Specialist at the Harvard College Library, did the final preparation of the maps in chapter 7 on the basis of material provided by the author, and Hou Ieong Ho assisted in the preparation of the figures in that chapter. I also thank the editors and publishers of Hanxue yanjiu 漢學 研究; Harvard Journal of Asiatic Studies; Imago Mundi: International Journal for the History of Cartography (www.tandfonline.com);

xxii

preface and acknowledgments

Library Trends; T’oung Pao: International Journal of Chinese Studies; Brill Publishing; and Librairie Droz for allowing me to reproduce, summarize, and rework parts of the articles and chapters I published with them. The questions of how to research the reception history of state documents and how to analyze the communication networks of literati led me to explore a range of digital methodologies. The core data on which the arguments of the final chapters are based can be accessed and interactively explored by following the relevant links at http:// chinese-empires.eu/reference/information-territory-and-networks/. I also add there a brief explanation of the format of the data and the history of the transformation of the initial methodologies into the MARKUS text analysis and reading platform.

Abbreviations

CBDB CSJC-CB j. QSW QYTFSL SBCK SHY SS SKQS WXTK XZZTJ ZZSL ZZYL

Peter Bol et al., China Biographical Database Project Congshu jicheng chubian jinshi Zeng Zaozhuang and Liu Lin, eds., Quan Song wen Qingyuan tiaofa shilei Sibu congkan Xu Song, Song huiyao jigao Tuotuo, ed., Song shi Siku quanshu Ma Duanlin, Wenxian tongkao Li Tao, Xu Zizhi tongjian changbian Chen Zhensun, Zhizhai shulu jieti Zhu Xi, Zhuzi yulei

I n t roduc t ion

I

n the Chinese popular imagination and in survey histories, the Song dynasty (960–1279) is commonly envisioned as an age of military and political weakness. The Song court is frequently referred to as “the small court” (xiao chaoting 小朝廷) in titles of and subheadings in academic histories, especially in discussions of the second half of its three-hundred-year rule, after it had signed a diplomatic settlement with the Jurchen Jin court (1115–1234), in which it agreed to relinquish all claims to the Chinese territories north of the Huai and Yangzi Rivers. The derogatory label was not the invention of later scholar-officials celebrating the reunification of the Chinese territories under Yuan (1279–1368), Ming (1368– 1644), Qing (1644–1911), or republican rule. Rather, it was directly adopted from one of the more vociferous Song critics of the policy of appeasement with the Jin court. In 1138 Hu Quan 胡銓 (1102–80) ended a fiery memorial he had written in opposition to Qin Gui’s 秦檜 (1090–1155) scheme to have Emperor Gaozong 高宗 (r. 1127–62) assent to Jin demands with the following words: “How would it be possible to seek life in the position of a minor court [xiao chaoting]!”1 Hu Quan swore it would be better to end one’s life rather than accept the terms proposed by the Jin court. In an ironic twist of history, Hu Quan’s phrase became 1. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.208. The first number refers to the number of the installment; Huizhu lu was published in four installments.

2 introduction

an epithet for the Song dynasty in modern historiography. Closer analysis of the original context reveals that his incendiary antiforeign rhetoric and his demands for the execution of the main peace advocates and the immediate mobilization of Song troops were not intended as an admission of weakness. Nor were these comments a descriptive statement about a change in the political status of the Song polity on the part of a Song official. Rather, they read and were read as a call to arms to defend and restore Song territory. This call reportedly caused a stir in the streets of the Song capital2 and continued to be celebrated in notebooks and histories throughout the twelfth and thirteenth centuries. Hu Quan’s memorial and its special place in Song as well as later Chinese history testify to the maintenance and continuity of imperial traditions in twelfth- and thirteenth-century Song China. The question of how empire is maintained is the central focus of this book. Rather than regarding the twelfth and thirteenth centuries as a period of political weakness and division and great cultural and intellectual but “inward-turned” achievement, I propose that these centuries should assume a prominent place in an account of the remarkable continuity of empire in Chinese history. The continuity of empires is crucially dependent on how they are maintained and revived in times of crisis. The geopolitical crisis of 1127 initiated the final period, in which core Chinese territories became an object of lasting contention among multiple states. By focusing on the sources and channels of information that kept the imperial vision alive during the last 150 years of Song rule, I aim to highlight commonly overlooked factors that were critical in the revival and maintenance of empire in the last millennium of Chinese history.

The Problem of Empire in Chinese History Some years ago a colleague in Chinese history observed, “The Chinese did not have an empire; only the British had one.” The statement was offered in response to my discussion of the construction 2. Bi Yuan, Xu zizhi tongjian (Zhongguo jiben guji ku digital ed.), 121.10a.



introduction

3

of a political imaginary in twelfth-century Chinese atlases and maps and readings thereof (see chapter 3). Further discussion led us to the conclusion that at best “the Chinese had a problem with empire.” This latter statement expresses a sentiment that many Chinese historians share about the application of a Western category like “empire” to pre-nineteenth-century Chinese dynasties and their polities. One widely observed problem, and the one that my colleague was referring to, is that many Chinese politicians as well as Chinese subjects had a problem with imperialist policy. Imperialist policy is commonly associated with Western imperialism and, in particular, the expansionist regimes of the nineteenth and twentieth centuries that competed among one another for control over the non-Western world. “Imperialism” and “imperialist” imply an expansionist will to empire, a drive toward territorial expansion, the deliberate exploitation of new territories, and the subjugation of peoples and states. And, indeed, throughout Chinese history, we find that at times when expansionism was advocated and in evidence (e.g., under Emperor Wu of the Han in the second century BCE or under the Song emperor Shenzong in the late eleventh century) many cautioned against it. They did so for political and for cultural reasons. Political advisors and critics since the Han dynasty had warned that expansion was not sustainable and would cause the reigning dynasty to become stretched out too far and collapse. Historically minded policy advisors such as Sima Guang added that dynastic collapse had time and again resulted in multistate rule. Critics of territorial and economic expansionism also relied on an argument of cultural difference: Chinese dynasties ruled by nonviolent means, by attracting others to enter voluntarily into the orbit of civilization and by engaging them in an economy of tributary relations, or else by keeping others at bay in order to maintain civilizational differences. By and large expansionist agendas did not persist over time. There is, however, a critical difference between “imperial” and “imperialist” policy and practices. The former refers to policies and practices that help form and maintain empires but is not necessarily associated with military, economic, or cultural expansionism. Histor­ ically, early empires emerged without the strategic and expansionist

4 introduction

vision that imperialism implies. They also tended to be more concerned with the core, whereas imperialist regimes became more closely involved in the periphery.3 Throughout this book I use “empire” as an analytical category and not as a translation of any Chinese term in particular. “Empire” here stands for a type of political formation, relatively large in size, in which a political core (the emperor, the court, and the central bureaucracy) exerts control over vast territories and diverse populations through the mediation of elite brokers who assist the core in the extraction of resources in the form of revenue, corvée labor, and military service.4 Elite brokers engage in unequal relations with the core and tend to be segmented. Distinctions are maintained between political, economic, and religious elites in the provinces and in frontier zones, and different compacts bind different elites to the core so that their authority is negotiated via the center. Empires thrive on relationships of inequality and difference and, in that sense, they are theoretically distinct from nation-states. In Chinese history, successive dynasties kept religious elites out of the examination system and certified them through mechanisms that bestowed limited authority on them but also kept them out of the regular administration. They also attempted to incorporate frontier elites, especially local leaders in non-Han communities, through compacts that differed from those concluded with political and local elites in the Chinese heartland. From the twelfth century onward, lower-level degree holders, alongside others whose status was not recognized through the granting of privileges, became more visibly involved in the management of local welfare; the scope of these intermediaries’ involvement in the running of the empire was also subject to negotiation and renegotiation with the core. Even though Chinese empires may at times have been less ethnically diverse and less segmented than Western Asian empires such as the Abbasid or Ottoman empires, the strategies and processes of negotiation with intermediary elites in evidence in Chinese history 3. Münkler, Empires, 8, 152; Motyl, “Is Everything Empire?” 244. 4. Barkey, Empire of Difference; Burbank and Cooper, Empires in World History; Doyle, Empires; Münkler, Empires; Eisenstadt, The Political Systems of Empires.



introduction

5

can (and ought to) be compared with the intermediary levels of negotiation that were critical in the formation and maintenance of empires elsewhere. The application of such a definition raises a second problem with empire for Chinese history, namely, the imposition of an analytical framework that could itself be considered imperialist. This problem arises both from the wider debate surrounding the history and contemporary relevance of empires and from issues of translation and the representation of Chinese history in comparative studies of empire. As I have written elsewhere, in their discussion of Chinese history, core monographs in comparative studies of empire reproduce analytical and rhetorical structures that have resulted in the historiographical neglect of key events and structural transformations in imperial Chinese history.5 Analytical features such as the foregrounding of select empires and periods as ideal types (typicality) and the search for divergence have helped perpetuate Eurocentric and Sinocentric analyses. In this book I focus on an atypical case in the comparative study of empires: the Song Empire after its ruling elite lost control over the northern half of its territory (1127–1276). I examine the individual and collective commemoration of the Jingkang crisis (1126–27) and explain how the reaction of literati to the geopolitical crisis that ensued transformed the relationship between court and provincial elites in the area of political communication. (“Literati” here refers to those who were literate and conversant in the cultural skills requisite of the scholar-official; they form a subset of what twentieth-century scholars regard as the imperial elite, which includes others influential at local, regional, and central levels. The elites referred to in this work are predominantly literati elites.) The first five chapters trace the story of the systematic appropriation of bureaucratic genres associated with courtly textual and graphic production, and the last three examine the articulation of an imperial mission among literati and the scope of literati networks of textual exchange. My study confirms the critical role that literati have been 5. De Weerdt, “Shmuel N. Eisenstadt and the Comparative Political History of PreEighteenth-Century Empires.”

6 introduction

assigned throughout the history of Chinese empires by Shmuel Eisenstadt and others who followed in his wake. However, whereas these scholars assert that literati elites had predominated since the early empires, while other elites were relatively weak, I propose that the literati culture and identity and the structure and geography of communication networks taking hold in the twelfth and thirteenth centuries played a key role in the continuous maintenance of imperial traditions from the twelfth century onward.

A Structural Transformation in Political Communication Sim a Gua ng a nd the Problem of Unified Rule I have thus far established that we can examine the history of the Song (and other) Chinese polities as empires, but that there are several problems with the concept of empire in Chinese history. In addition to those, there is yet another problem that lies at the heart of this monograph, namely, the difficulty of maintaining political unity over all the Chinese territories. It was articulated in exemplary fashion by Sima Guang 司馬光 (1019–86), one of the best known historians in Chinese history. In 1061 he presented a memorial to the Song emperor Renzong 仁宗 (r. 1023–63) in which he pointed out that multistate rule had historically been dominant in Chinese history: “In these seventeen hundred or so years [from the move of the Eastern Zhou capital in the eighth century BCE until the foundation of the Song], there have been only five hundred or so in which the realm was united.”6 The transmission history of Sima Guang’s memorial is a powerful indicator of the larger change in political communication that takes place between the eleventh and thirteenth centuries. Sima Guang wrote these words before he started work on Zizhi tongjian 資治通鋻 (The comprehensive mirror for aid in government, 1084), a comprehensive history from 403 BCE until 959, the 6. Sima Guang, Zhuanjia ji, 21.4b.



introduction

7

year before the founding of the Song dynasty. His observation about the longue durée of Chinese history, articulated in 1061, appears to have been a source of inspiration for the kind of history he set out to write in The Comprehensive Mirror. For, even though he had originally been commissioned to compile a compendium of precedents from previous reigns, he decided to write a longue durée history.7 Such histories had seldom been written since the days of Sima Qian 司馬遷 (145–90 BCE), but it was in Sima Guang’s view a more solid basis upon which to form long-term policy than the administrative encyclopedias that dissected history into bits and lacked a long-term perspective.8 The main audience for Sima Guang’s original statement was Emperor Renzong. He impressed upon the emperor the frequent incidences of imperial collapse while urging him to devote himself energetically to the affairs of the imperial state. The history lesson was cautionary and took place in a context of policy debates on the pros and cons of the centralization of state power and relations with Khitans and non-Han peoples along the northwestern and southern periphery of Song territory. Despite his cautionary tone, Sima Guang noted that he was writing during a high point of imperial rule. The memorial continued that Emperor Renzong was in a fortunate position: “Ever since our dynasty pacified the region east of the Yellow River, for more than eighty years, there have been no internal or external disturbances. And so there has been no time of peace more glorious than ours ever since the Three Dynasties [of antiquity].”9 Sima Guang’s observation about the longue durée of Chinese history was repeated many times in the twelfth and thirteenth centuries, but after 1127 its import changed. After the fall of the Song capital of Kaifeng in 1126–27 and the division of the Chinese territories into two halves, the north ruled by the Jurchen Jin dynasty and the south by the Song, Sima Guang’s observation that imperial unity is easily lost materialized once again. The memorial from 7. Sung, “Between Tortoise and Mirror,” chap. 1, esp. 17–36. 8. For examples during the period of division, see Dien, “Historiography of the Six Dynasties Period.” 9. Sima Guang, Zhuanjia ji, 21.4b.

8 introduction

which the translated passages were taken was included in several anthologies, not only of memorials but also of prose texts more generally, in the twelfth and thirteenth centuries.10 A particularly striking example of the greater publicity given to the idea of the prevalence of disunity is a stone stele entitled “Diwang shao yun tu” 帝王紹運圖 (Diagram of rulers in succession).11 Completed around 1247, the stele traces the succession of Chinese dynasties and their rulers from the beginning of civilization through the Song present, starting with the Yellow Emperor and ending with the reign of Emperor Lizong 理宗 (r. 1225–64).12 This updated version was based on a similar table drafted by Huang Shang 黃裳 (1146–94) around 1190. Huang Shang drafted the table while serving as tutor to the future emperor Ningzong 寧宗 (r. 1195– 1224), but copies of this and other charts he compiled were preserved in his family’s collection in Sichuan. Through the intervention of Wang Zhiyuan 王致遠, who hailed from Eastern Zhejiang and who served in a number of local and regional offices, a stele edition based on Huang’s earlier draft was established in Suzhou Prefecture in the mid-thirteenth century and it remains there to this day. At first glance, the stele embodies the core characteristics associated with genealogies of any kind, family, dynastic, or intellectual. Along a vertical axis 197 rulers are arrayed, suggesting both legitimate succession and continuity over time.13 10. The full memorial was, for example, included in Lü Zuqian, Song wen jian (SKQS), 48.15b–27b (1179), Zhao Ruyu, Song mingchen zouyi, 1.5a–17b (ca. 1186), and Lou Fang (SKQS), Chonggu wen jue, 17.9a–12a. 11. Literally, “shao yun” means “to continue operation(s).” I have chosen to translate this as “Diagram of Rulers in Succession,” because alternate names for chronological charts like the one on the stele suggest that chronological sequence and transmission (as opposed to overseeing the natural operation of the realm, which is implied by the choice of “yun”) was the core feature of these charts. See, for example, the chronological chart included in the late Southern Song textbook Ruxue shuyao entitled “Diwang chuantong xi tu” 帝王统系之圖 (Table of rulers transmitting the line). Ruxue shuyao, juan 1. 12. My discussion of the 1247 stele is based on a transcription of the text by Zhang Xiaoxu (“ ‘Si da Song bei’ gaishu”). 13. For similar medieval genealogies of rulers, see Grafton and Rosenberg, Cartographies of Time, 34–35.



introduction

9

When read in conjunction with the inscription at the bottom, a tension emerges between the ideal of continuity (which is what it was intended to capture) and history as perceived by the stele’s twelfth- and thirteenth-century authors. The inscription quotes Sima Guang’s memorial at length but ends with the line that there had been only about five hundred years of unified rule; the celebratory words at the end of the original text were left out. Around the time when the stele went on show in Suzhou, Wang Yinglin 王應麟 (1223–96) singled out this one line as a key moment in a historical overview of the administrative organization of Chinese states: “In these seventeen hundred or so years, there have been only five hundred or so in which the realm was united.”14 Sima Guang’s memorial had now been condensed to one sentence. The historical lesson it conveyed was, by the time Wang Yinglin compiled his encyclopedia, no longer solely or even primarily intended for the emperor. It was recollected by and for literati at large. Sima Guang’s memorial had become part of a repertoire of texts that had once been the exclusive domain of the court and high officialdom but that now moved decisively in the hands of the literati. How this shift in the production of texts regarding the ruling dynasty’s affairs took place, what motivated literati interest in them, and what consequences this shift had for the history of Chinese political culture are the core questions addressed in the chapters that follow.

Structur a l Ch a nge a nd the Jingk a ng Cr isis The twelfth century has long been regarded as a period of structural change. More precisely, it has been regarded as the culmination of a longer-term process of social change spanning the period from the late eighth to the thirteenth centuries. Even though there are disagreements about the timing, nature, and scope of social change, social and intellectual historians tend to agree that the criteria for elite status changed during the course of the ninth through the eleventh centuries and that local interests were more clearly and more voluminously articulated among elites from the twelfth century 14. Wang Yinglin, Yuhai, 18.38a–b.

10 introduction

onward.15 It is generally accepted that by the thirteenth century, the political elite had changed from the hierarchically defined and capital-based aristocracy of the Tang Empire, to an elite of official servants recruited through examinations and focused on the capital in the eleventh century, and to an elite dominating local society in substantial part on the basis of its educational and cultural credentials thereafter. By the end of the twelfth century, elites living under Song rule had come to accept that learning, however defined, took precedence over other determinants of elite status such as family background. They, moreover, claimed learning not only as the prerequisite for office holding, for, in tandem with the expansion of the role of the civil service examinations in recruitment for office, learning and the demonstration of it through the production and exchange of texts also became a marker of literati status broadly conceived.16 Was the investment in local ties among elites in the twelfth century accompanied by changes in political communication between court and provincial elites? What effect did any such changes have on political institutions and political imaginaries? My answer to such questions is that the broadening and provincialization of scholar-official elites coincided with changes in the relationships and structures of information exchange between court and provincial elites. In these chapters I examine how changes in political communication between court and literati and among literati accommodated long-term absence from metropolitan areas and also question, on the basis of literati cultural production and communication patterns, how localized they had become. 15. For an early articulation of the nature of changes between the Northern and Southern Song, see Hartwell, “Demographic, Political, and Social Transformations of China”; Hymes, Statesmen and Gentlemen. For an overview of East Asian and European language scholarship, see Luo Yinan, “A Study of the Changes in the ‘Tang-Song Transition Model.’” Representative examples of different kinds of critiques of the localist turn include Bossler, Powerful Relations; Lee Sukhee, “Negotiated Power.” While most of the debate has focused on the nature of changes during the Song period, the ninth and tenth centuries have recently also come under closer scrutiny; see esp. Tackett, “Great Clansmen, Bureaucrats, and Local Magnates.” 16. The standard articulation of this transition is Bol, This Culture of Ours.



introduction

11

I pursue this shift in the relationship between court and literati in the area of political communication in three ways. At the level of institutional history, I examine how the general trend toward centralization from the tenth through the early twelfth centuries built on and reshaped pre-Song institutions. Throughout the first five chapters it also becomes evident that central institutions such as the court libraries, the archives, the Memorials Office, or the departmental examinations in the capital became the operating ground for elite networks that connected elites in the capital to literati across the Song territories. We also see whether and how the infiltration of elite networks into these institutions affected central control over the circulation of information and knowledge about the administration of the Song Empire. This question is pursued in part through an investigation of the impact of access to court archives and archival modes of operation on elite writing. Second, at the level of legal history, I document the increasing regulation of publishing in terms of the frequency of regulations and of the more detailed specification of regulations by genre and by means of publication. The stricter regulation of the manuscript and print publication of archival compilations, single-sheet state documents, court gazettes, policy essays, and texts on border affairs is, throughout the first five chapters, interpreted in the context of their growing publication and circulation. One of my central concerns in this monograph is to discuss these two paradoxical trends together: increasing legal control over manuscript and print publication, on the one hand, and the escalation of the private, commercial, as well as local government publication of ephemera and archival and recent historical compilations, on the other. The first aspect, let us call it censorship, has been covered extensively in existing scholarship in East Asian and European languages. Censorship is usually discussed either in comprehensive histories focusing on a continuous history of government intervention in publishing, or in case studies focusing on the often more complex personal and political relationships that led to infamous interventions.17 Here the 17. See, for example, An Pingqiu and Zhang Peiheng, Zhongguo jinshu daguan; Chan, Control of Publishing in China; Hartman, “Poetry and Politics in 1079.” On the other

12 introduction

key question is how we can best interpret the ambivalent position of central and local governments toward the dissemination of state documents and political news, an ambivalence that is not only evident in the lack of systematic implementation of existing regulations but also in the issuing of countervailing regulations on appropriate rewards for collectors donating or allowing the copying of forbidden materials. These chapters also discuss the longer-term implications of central and local governments’ ambivalent position for the history of Chinese political culture. Third, at the level of the cultural history of publishing and reading, I show that a marked shift took place in the production of texts relating to the administration of the Song Empire. In the book’s individual chapters on the publishing and dissemination of court gazettes, archival and historical compilations, maps and atlases, and military and administrative geographies, we see a general shift from the almost exclusive court supervision and publication of these genres toward their widening private and commercial production from the late eleventh and twelfth centuries onward. Such a trend has been observed in the production of some genres (for example, gazetteers, encyclopedias, and examination manuals) in past scholarship.18 Here, this trend is also correlated to a further burst in the publication of notebooks and the collected works of individuals—these are the most promising sources we have to explore the individual and collective reception of current affairs. Its broader applicability raises further questions. Why did a government whose initial policies of centralization exceeded those of its Han and Tang predecessors retreat to a less interventionist position in the twelfth century? Did the dissemhand, Kai-wing Chow has commented that governments were on the whole lax about the circulation in print of current affairs in the sixteenth and seventeenth centuries (Chow, Publishing, Culture, and Power in Early Modern China, 252). Chow further argues that the private and commercial printing of current affairs resulted in a “complete alienation from the imperial government” (ibid., 145). Here I will suggest that a similar phenomenon in the twelfth and thirteenth centuries did not necessarily lead to such alienation but rather to the reinforcement of literati commitment to the imperial mission. 18. See, e.g., Hargett, “Song Dynasty Local Gazetteers”; De Weerdt, “Aspects of Song Intellectual Life” and Competition over Content, esp. chap. 5; Mostern, Dividing the Realm, 90–99.



introduction

13

ination of bureaucratic genres affect literati identities and political loyalties? How did literati read and respond to bureaucratic literature? And, what was the role of woodblock printing, in the eyes of both Song authors and a modern historian, in literati communication? This technology too was first used to disseminate all manner of texts from the late eleventh and twelfth centuries onward. Structural change tends to come about as a result of longer-term developments and critical events that offer a sudden opportunity for a change in power relationships to take hold. In this study, I show that the events of 1126–27, referred to as the Jingkang crisis, played a pivotal role not only, as I suggested earlier, in bringing about longer-term social change but also in consolidating developments in political communication that had been gaining momentum since the latter half of the eleventh century. A brief overview of these events and their impact is presented below. Between 1125 and 1127 the Song capital of Kaifeng, home to an estimated population of 1,400,000, repeatedly came under the attack of Jurchen armies. Memoirs of the occupation of the city circulating in the years and decades that followed recollect how the ransom in gold and silver demanded by the Jurchens led in the first instance to the impoverishment and oppression of the city’s residents. When it became gradually clear that the demand for millions of ounces of gold and silver and tens of thousands of bolts of silk and heads of livestock could not be met, the city was emptied out of all valuables and large numbers of its population perished, were captured, or fled. Thousands of women were handed over in compensation for unmet payments, tombs were robbed, and wooden buildings demolished when firewood became in short supply during the snowy winter months. When the occupation came to an end in early 1127, the Jurchen army escorted hundreds of carts carrying not only gold, silver, and silk but also books, paintings, vessels, and other precious objects and antiques. An estimated fifteen thousand captives followed suit.19 The events of these years were referred to by contemporaries and later historians as the takeover/disorder/shame/disaster of the 19. Ebrey, “Introduction.”

14 introduction

Jingkang reign ( Jingkang zhi bian / luan  / chi / nan / huo 靖康之變、靖 康之亂、靖康之恥、靖康之難、靖康之禍)—the era name Jingkang or “peaceful prosperity” had been chosen in 1126 in the hope that peace would be restored and the control of the Song over its territories reestablished. In the years and decades that followed, the Song court relocated to the southern city of Hangzhou and agreed to a peace deal that effectively divided the former Song territories into two halves. The northern half, including the former capital, was now ruled by the Jurchen Jin dynasty. Following its downturn, from the 1120s onward, Kaifeng, one of the largest cities in the medieval world, never regained the prominence it had once held. The war with the Jurchens did not lead to an economic downturn, however. The population in both the north and the south recovered (with the exception of some areas, particularly in the new border zone), and judging from the Southern Song state’s fiscal income and sustained urbanization it appears that the twelfth and thirteenth centuries were broadly speaking not a time of socioeconomic crisis. Nevertheless, the loss of the north left an indelible impression on the minds of those who witnessed the events and remained for future generations who lived in the south “a shame to be washed away.” In this sense the twelfth and thirteenth centuries were a time of crisis. I suggest that this event and similar events taking place in other parts of the Song Empire in the 1120s and the longer-term territorial division that followed led to a stabilization of structural transformations in the relationship between court and imperial elites in the south. This transformation in the social sphere and in political communication in turn had long-term effects for the maintenance of imperial rule in later Chinese history.

The Literati Adoption of an Imperial Mission A significant and often overlooked feature of the twelfth-century surge in literati textual production outlined above is the articulation of an imperial mission. According to Herfried Münkler, “All empires that have lasted any length of time have chosen as their



introduction

15

self-justifying objective a world-historical task or mission that confers cosmological or redemptive meaning on their activity.”20 Such a mission is, in Münkler’s analysis of the Roman Empire, mostly directed at the political elites at the imperial center and motivates them to transcend personal interests and to take on duties that can only be realized in the long term. In early Chinese history the pivotal cosmological role of the ruler was theorized and justified in texts and ritual practice. Cosmological schemes were used to differentiate between the domain of civilization and the boundaries of non-Chinese social life. More generally, Mark Edward Lewis, in Writing and Authority in Early China (1999), has shown that the viability of a unified empire in the last centuries BCE depended on its inscription as an imaginary realm in a classical canon and on historical, administrative, philosophical, literary, and encyclopedic texts.21 The scholars and advisors who designed this imaginary 20. Münkler, Empires, 84. The emphasis on graphic, historical, and archival texts in the articulation of the elite mission in this monograph differs from that of Chinese and comparative studies of empire, which consider ideological factors equally key in the consolidation and maintenance of empires but tend to define these as broad cultural orientations—abstract traditions rather than historically experienced identities. Doyle (Empires) emphasizes public legitimacy. Mann (The Sources of Social Power) sees ideology as a key dimension of social power. And for Eisenstadt (The Political Systems of Empires) “cultural programmes” are an important variable in explaining the different histories of Eastern and Western polities. Münkler’s approach to imperial mission allows for greater analytical and historical specificity. 21. Lewis’s work shifted attention from the detection of the formal characteristics of empire (the military conquest of wide territories and many peoples; the exploitation of the conquered through taxation, tribute, or conscription; imperial projects; imperial symbols and institutions; imperial elites and collaborating classes) to the historically and culturally specific conceptualization of empire as a crucial factor in its formation. More recent scholarship (Wechsler, Offerings of Jade and Silk; McMullen, State and Scholars in T’ang China) has highlighted the role of Buddhist, Daoist, and classical/ Confucian texts and rituals in the legitimation of Sui and Tang imperial rulership and in the construction of a political and cultural common ground among the elite. The standardization of commentarial traditions in Wujing zhengyi 五經正義 (The correct meaning of the Five Classics) and other large-scale court-sponsored textual projects were, as shown in the work of David McMullen, a powerful example of the importance attributed to shared textual knowledge in the creation of an imperial elite. The first Song rulers similarly endorsed large-scale textual projects in an effort to generate

16 introduction

realm justified the ruler’s authority and the project of unification, but, in defining the scope of the harmonious sociopolitical order that should prevail, this interpretive elite also staked out a role for themselves and those who followed in their footsteps. Many of the synthetic, historical, and literary texts compiled in the formative years of the Qin, Han, and Tang empires became canonical and thus continued to exert influence over literati imaginations during the Song and later dynasties. The questions of which of these texts were particularly influential and how they were read are taken up in several chapters and especially in the chapters on the representation of the Chinese territories, borders, and foreign others (chapters 3–5, 8), in which the interest in “The Tribute of Yu” chapter of The Book of Documents is highlighted. It is one of my key contentions in this monograph that by the beginning of the second millennium different kinds of information contributed to the conceptualization of empire. The imperial vision was no longer solely informed by the coherent and durable knowledge of the classical canon and the large administrative, historical, literary, or medical compendia sponsored by successive dynasties’ founding emperors; it was continually reinforced by historical and current information about the Song polity. On the basis of an investigation into changes in the production and circulation of Song archival and current information, I propose that the relative retreat in the court’s position referred to in the previous section brought with it a strengthening of elite commitment to the imperial state and not, as has been argued, a turn away from the center and imperial government. The Jurchens’ lasting occupation of the northern half of the Song territories questioned the ability of the court and the central bureaucracy to uphold the mission. During this prolonged geopolitical crisis, the mission was stressed more than ever, and political networks dedicated to the recovery of the north spread from traditionally legitimate political actors (those in active service) to those in the provinces anxious about the court’s dedication to the larger project of territorial reunification. Lower-level bureaucrats as elite support and recruit generations of men schooled in officially sanctioned editions of texts.



introduction

17

well as local elites monitored and studied personnel and policy shifts continually. The recovery of the north did not seem on the immediate horizon for most of the duration of the Southern Song dynasty, and when hopes were raised they were quickly disappointed, as in the costly campaigns of the early 1160s and the early 1200s.22 The imperial mission can, therefore, not be explained exclusively in terms of direct elite interests; it required the formation of attachments to state and territory that transcended the interests of self, family, and locality. Did the meaning of the imperial mission change as a result of the broadening and provincialization of the political elite in the course of the twelfth and thirteenth centuries? Through an examination of the commemoration of territorial loss and its articulation in a wide range of cartographic and geographic work in chapters 3–5, I argue that the enduring occupation of the north led to a prioritization of the recovery of the full extent of the Chinese territories and an attachment to “a normative empire.” The dimensions of this idealized polity were determined on the basis of universal (cosmological, topographical, and classical) grounds as well as historical precedent. The foregrounding of such territorial concerns in elite commitments in the capital as well as in the provinces finally also raises questions about broader changes in political culture. In particular it raises questions about the role of loyalty (zhong 忠) in the self-perception and behavior of Song literati. Political and intellectual historians have observed that loyalty—a central virtue in Chinese political theory since the Warring States philosophers’ reflections on the relationship between ruler and minister—was defined in more exclusive terms in the eleventh century. James Liu and Naomi Standen have argued that the historiography of eleventh-century politicians such as Ou­ yang Xiu 歐陽修 (1007–72) and Sima Guang provides clear evidence that attitudes toward loyalty were hardening in the wake of the period of imperial disintegration and contending states covering the mid-eighth through the late tenth centuries.23 Standen further 22. Gong Wei Ai, “The Reign of Hsiao-tsung,” 713–20; Davis, “The Reigns of Kuangtsung and Ning-tsung,” 791–812. 23. James T. C. Liu, Ou-yang Hsiu; Standen, Unbounded Loyalty, chap. 2.

18 introduction

demonstrates that the interpretation of loyalty in historical texts changed from that of a quality of a personal relationship to that of a relationship between a subject and the ruling dynasty. Others have argued that Neo-Confucianism contributed toward the idealization of loyalty toward emperor and dynasty to the extent that it became a litmus test of the literatus and standard feature of late imperial political culture best observed in the dramatic displays of loyalists refusing to serve new masters in the early Yuan, the early Ming, the early Qing, and even the beginning of the republican era. On the other hand, Jennifer Jay has shown that the significance of loyalty has often been overstated, mostly as the result of twentiethcentury nationalist historiography.24 In chapter 8 I examine some of the contexts and meanings in which loyalty is discussed in the sources examined there. I will also propose there that a sense of empire in a very territorial sense determined the meaning of dynastic loyalty for the majority of the elite who, as Jay conclusively demonstrates, did not take the course of martyrdom or engage in other radical displays of dynastic loyalism. Dynastic legitimacy and thus loyalty to emperor and dynasty alike was for many premised on the dynasty’s dedication to the goal of restoring the normative dimensions of the homeland.

Networks and the Geography of Communication A second significant feature of the surge in textual production about the affairs of the reigning dynasty is the structure and geographic scope of the networks within which such texts were exchanged and commented upon. Networks, fluid and dynamic configurations of social relationships, cut across the central and local government institutions charged with communication between the capital and the localities. The interaction between institutional frameworks and social networks is a critical concern in this work since access to court news and state documents depended on per24. Jay, A Change in Dynasties.



introduction

19

sonnel with institutional affiliations to the central bureaucracy and on broader networks in the capital and in the provinces. Networks were equally critical in both the manuscript and print publication of primary state documents and secondary discourses about them. As private and commercial printing took off in the eleventh and especially the twelfth century, the embeddedness of authors and editors in broader social networks, or their ability to create ties with peers, often determined to what extent their work circulated. The critical examination of the social networks underlying the publication of specific titles can thus help us better understand access and publication, but in order to explore the structure and the dynamics of networks more systematically those genres in which social relationships were recorded and recalled need to be taken into consideration. Letter collections and notebooks are the sources most suited for this purpose and for a historical examination of the role of elite networks in political communication. They are the best sources we have to read about individual and collective interpretations of Song dynastic history and current affairs and to investigate the reception history of state documents. Judging from the extant record, the sharing and the publication of letters in Song times proceeded on an unparalleled scale in Chinese and world history. Basing his calculations on a tagging of all published personal letters included in a recent edition of all Song prose texts, Chu Ping-tzu has counted 17,957 letters by 611 authors in total.25 From my review of a wider selection of epistolary materials including more formal genres in the same collection, I estimate the total number of extant letters for the Song period to lie between 20,000 and 25,000. By contrast, less than 1,000 exist for the entirety of the Tang period. Similarly, only a few hundred remain from the Liao and Jin periods.26 Bernard Gowers estimates the number of letters surviving from 25. Chu Ping-tzu, “Beyond the Personal.” This number does not include genres of official communication such as qi 啟, biao 表, zou 奏, etc. 26. These estimates are based on the letters included in the large prose collections for these periods (and their supplements), Quan Tang wen and Quan Liao Jin wen—in the case of Quan Liao Jin wen, the estimate also includes some of the collections on which it was based. I acknowledge the assistance of Chu Mingkin for the Quan Liao Jin estimate.

20 introduction

twelfth-century Western Europe in the low thousands.27 A similar unprecedented growth took place in the publication of notebooks, as will be discussed in chapter 6. In this monograph I have opted to start with a case study of a series of notebooks in which the author regularly noted down the source of the written quotations or oral information he chose to discuss. From my analysis of the social and geographical distribution of informants as well as their temporal coordinates (see chapter 7) we can begin to construct hypotheses about the structure and the geographic scope of literati networks in the twelfth and thirteenth centuries. Such an analysis calls into question prevalent models for understanding political communication in imperial China. Within William Skinner’s macroregional model of Chinese geography, further discussed in chapter 7, political decisions were transmitted from the center and moved down a strictly hierarchical chain of administrative units. Within this chain each lower-level unit was made to fit within one and only one higher-level unit and, so, was effectively controlled by the center. In Skinner’s model, politics also took place in standard- and intermediate-level marketing towns at a remove from the bureaucratic chain of command. Here, local government and local elites intermingled and jointly led a variety of local welfare operations.28 In such a model there is no room for communication and exchange among elites resident in standard, intermediate, or central market towns that were not adjacent within the central place model of marketing towns. Political communication outside of the bureaucratic hierarchy of administrative places is then limited to local marketing patterns, patterns that were restricted by environmental conditions that divided the Chinese territories up into discrete physiographic regions. The geography of elite communication remains a largely unexplored field, and only a preliminary step can be undertaken here. At a minimum I hope to suggest that cross-regional communication 27. Gowers, “Comparing Letters and Letters-Writers.” 28. Skinner, The City in Late Imperial China, 307–46. For a comprehensive critique of the macroregional paradigm, see Cartier, “Origins and Evolution of a Geographical Idea.”



introduction

21

was a common phenomenon among literati in the twelfth and thirteenth centuries. As others have suggested with regard to the relative neglect of inter- and supraregional economic exchange in Skinner’s work, political communication and politics similarly cut across the hierarchies of formal administrative organization. This finding raises questions about the correlation between information exchange and other types of relationships. If we grant that elite marriage patterns became geographically more restricted in the post1127 period and that elites invested more heavily in local ties and local interests, did communication networks similarly contract, or did they, conversely, expand? The history of communication networks in this monograph is informed by social network analysis. How I have engaged network methodologies and the expanding body of literature in historical network analysis is presented in more detail elsewhere.29 In addition to social network analysis, the research on which this book is based combines traditional philological and historical methods with the use of prosopographical databases, historical geographic information systems, digital text analysis, and corpus linguistics. In view of recurring debates about digital and quantitative methods in the humanities, I have written a separate and extended explanation of my adaptation of digital methods in this work and the value of such methods for cultural history and humanities research in general.30 Finally, I have read the notebooks discussed in this monograph not so much as sources from which relational data can be extracted but as purposeful recollections of meaningful social relationships, conversations, and engagements with the writings of named authors and texts. This implies that political communication was not their sole or even their main aim. Rather, because of the diversity of the con­cerns and interests addressed, these sources allow us to ask in broader terms how literati identities were shaped and reshaped in the course of the Song dynasty. Through memories of reading and conversation, authors of the period evoked shared concerns and a multiplicity of political and cultural discourses and also articulated 29. De Weerdt, “Two Frameworks.” 30. De Weerdt, “Digital Interpretations” and “Isn’t the Siku quanshu Enough?”

22 introduction

their position within this context. In this way, notebook authors gave voice to the concrete social relationships and the cultural and political discourses within which individual elite identities were fashioned. My goal here is not to write a biographical narrative on this basis—only a full translation of individual texts would do justice to these sources in this respect. However, by analyzing in the aggregate the recollections recorded in the notebooks of lesserknown twelfth- and thirteenth-century authors I can begin to rewrite the cultural history of literati identities in the process of formation. On this basis we can also question the presumption that elite collaboration with the imperial court can be best understood with reference to overarching ideologies such as Confucianism or, from the twelfth century onward, Neo-Confucianism.

Communication and Information The question of how the circulation of political information and political communication among officials and other sectors of the population impacts on state formation, imperial integration, and national identity has in recent years received much attention from historians working on various parts of the early modern and modern world. Below I clarify my use of a second set of terms that would have been unfamiliar to Song literati. As in other recent work, I will also seek here to dislodge political information and communication and the technology of printing from a narrative in which they are linked to the formation of the modern nation-state.31 31. In contrast to the limited or nonexistent coverage of communication in general and of the sources and channels of political information in particular in comparative work on preindustrial empires, historians working on various parts of the early modern world have recently explored the dynamics of state formation and imperial integration from the perspective of social and political communication. Areas of research of particular relevance include the relationship between social communication and politics (de Vivo); the role of “information” and printed communication in the creation of a sense of nationhood (Berry, Frankel, and Anderson); the adaptation of statist models in public discourse (Berry and Frankel); and the relationship between local and imperial information systems (Bayly). Berry, Japan in Print; de Vivo, Information and Communication in Venice; Frankel, States of Inquiry; Bayly, Empire and Information.



introduction

23

“Political communication” is a rather malleable term. I use it here in the same broad sense as Filippo de Vivo does in his work on fifteenth- and sixteenth-century Venetian politics: “the circulation of information and ideas concerning political institutions and events.”32 Political communication in this sense is here also considered a political act. In de Vivo’s work, communication was politics in the very concrete sense that it was an arena in which the government’s attempts to preserve secrecy clashed with various social actors’ need for information and alliance building in the wake of the early sixteenth-century military crisis. De Vivo also shows that the tacit recognition of leaking as part of the political process was always challenged. Even though the government condoned and occasionally made use of leaking and pamphleteering, it reasserted its policy of secrecy once the political or diplomatic crisis had passed. A second implication of this definition of “political communication” is that communication implied secrecy.33 In the maritime empire of Venice in the sixteenth and early seventeenth centuries, a state like the Chinese Empire, with a relatively large ruling class but with strict boundaries defining who could exercise political power, secrecy was to some extent a myth, but it was a myth that was essential for the maintenance of another myth—that of harmony among the top of the administrative hierarchy, and unanimity in political decision making. The city of Venice was imagined to speak in one voice by Venetians and other Europeans alike. Secrecy had been part of the constitution of Chinese empires ever since the politicians laying the foundations for the establishment of the Qin Empire made it an essential part of political practice. As in early modern Venice, successive imperial bureaucracies were imagined by imperial elites and foreign interpreters alike to carry out the one will of the emperor. Historically, crises of confidence occasioned by military and/or diplomatic defeat revealed a growing tension between the policy of secrecy and elite social and political needs. As in early sixteenth-century Venice, in twelfthcentury Song China military and diplomatic defeat led to an outpour 32. De Vivo, Information and Communication in Venice, 2. 33. Ibid.

24 introduction

of publications about current affairs relating especially to personnel issues and military and diplomatic policy. The implementation of the civil service examinations and the broader cultural impact of them also stimulated demand for current political information. The exponential growth of the number of candidates preparing and sitting for the civil service examinations and the attested role of current information in the examinations led to an increased demand for information about dynastic history and court policy. As will be shown in each chapter of this book, John Chaffee’s estimates of the increase in candidates sitting the lowest level examinations—from 20,000 to 30,000 in the early eleventh century to 79,000 one century later and 400,000 or more for the Southern Song alone by the mid-thirteenth century34—correspond with growing evidence of the private and commercial publication of archival materials, state documents, and other types of current information. More widespread publication and use of such materials at each stage counterpointed government protestations of the divisive potential of policy documents and the threat they posed to security and unity should they be disclosed to the Song Empire’s powerful northern neighbors. Communication could no longer be imagined as a top-down affair in which the court instructs the people: the very genres and channels of official communication became sites of negotiation. Ambivalence about publicity always remained; as we shall see, the same individuals advocating policies that supported the imposition of secrecy were also found to be leaking information or to be benefiting from leaks. By exploring the tensions between secrecy and publicity, this work proposes that secrecy and unanimity proved to be myths worth preserving because they were, alongside the dissemination of the reigning dynasty’s current affairs, contributing factors to the consolidation of empire at a critical moment in Song history. “Information” here appears in two senses. First, an understanding of information as fragmentary and ephemeral bits is one that has become taken for granted in the world we live in. Under the influence of such disciplines as cybernetics, systems theory, the computer 34. Chaffee, The Thorny Gates of Learning, 36.



introduction

25

sciences, artificial intelligence, and, more recently, information economics, we have come to think of information as something akin to energy: it is characterized by the uninterrupted flow of fragmentary and ephemeral bits.35 Even though the speed and flow of information thus conceived are unparalleled, information in the sense of fragmentary and ephemeral bits is not a twentieth-century phenomenon. For example, Mary Berry intentionally refers to seventeenthcentury Japanese reference genres as virtual information libraries. She argues that the early modern Japanese library consisted of ephemeral information, detailed and filled with numbers and continually updated.36 This meaning of “information” is also applicable in the context of Song China. Our understanding of the term applies in particular to its bureaucratic operations. The constant movement of reports is the lifeblood or the energy of bureaucracies: reporting information often of an ephemeral and fragmentary nature, adding on to reports and circulating them, and eventually turning reports into archival materials and indexing them for future retrieval are operations that are shared across bureaucracies. Max Weber famously defined bureaucracy as “the exercise of control on the basis of knowledge.”37 When leaked systematically from the inner quarters of the bureaucratic world, bureaucratic genres and communication practices also took on a higher profile in literati textual production and political communication, as will become evident in parts I and II. I will also use “information” in its second and older sense. Before the emergence of the disciplines mentioned above, “information” meant “the imparting of learning.” It was close in meaning to what we refer to as knowledge and learning, with encyclopedic scope, coherence, and durability as attributes. “Information” in this meaning was also part of the Song political and intellectual life, because historical and administrative encyclopedias, notebooks, 35. Scholle, “What Is Information?”; Nunberg, The Future of the Book. Ann Blair similarly focuses on the management of the bite-size type of information in Too Much to Know. 36. Berry, Japan in Print. 37. Weber, Economy and Society, vol. 1, 339; quoted in Burke, A Social History of Knowledge, 118–19.

26 introduction

and prose anthologies brought together state documents and archival materials for both short- and long-term use. The prolific use of woodblock printing in the dissemination of political information during the Song period, which I discuss in the following chapters, raises the question of how these findings relate to the more voluminous work on the role of information and printing in the creation of a sense of nationhood in early modern and modern times. Communication and publishing were, in Mary Berry’s reading of early modern private Japanese archives, important catalysts in the formation of the Japanese nation.38 Building on a broad review of the plethora of sources of social and cultural information (booksellers’ catalogs and travel guides; maps and gazetteers detailing streets, trades, and products; rosters listing those in power; encyclopedias of the natural world; etc.) available to Japanese of all walks of life, Berry argues that such sources provided the basis for “a sense of nationhood: an integral conception of territory, an assumption of political union under a paramount state, and a prevailing agreement about the cultural knowledge and social intercourse that bound ‘our people.’”39 Even though commercial publishing played a dominant role in the creation of this virtual information library, Berry notes that the state provided the lexicon and the framework for all kinds of bestselling investigative writing as cadastral and cartographic surveys provided the model.40 Oz Frankel similarly conceptualizes government-sponsored printed social investigative reports in nineteenth-century England and America as a print archive that functioned differently from earlier government repositories in that social reports were made available for public use and comment. Frankel also sees the circulation of such reports and other parliamentary papers as a force shaping public discourse in both the content and modes of conversation and steering British state- and American nation-building efforts.41 38. Berry, Japan in Print. 39. Ibid., 248. 40. Ibid., 44. 41. Frankel, States of Inquiry, 29.



introduction

27

My own study places the political and identity-shaping powers of print publishing firmly in the context of the history of empire. It thus relaxes the connection between print information and nationhood in Berry’s or Frankel’s work or the paradigmatic articulation of the link between print capitalism and nationalism in Benedict Anderson’s Imagined Communities. It is primarily concerned with the circulation of political information among imperial elites and the role of political communication solidifying the imperial mission at a time of crisis.42 The first three parts of this book are intentionally structured around a typology of places associated with empire: court, borders, and territory. This is not an exhaustive list. The capitals and local jurisdictions were also associated with empire.43 This foregrounding of court, territory, and borders is the result of a research strategy designed to capture the opposite of a contemporary and related development in textual production in the twelfth and thirteenth centuries: the expansion and commercialization of information about local places (administrative subdivisions at different levels). Part IV shifts attention to the ways in which oral and written information 42. Some intellectual and literary historians have found in the xenophobic language of twelfth-century authors the burgeoning of a proto-nationalism. I will argue below that xenophobic language in select, albeit widely circulated, political and literary genres was part of a rhetoric of empire calling for the restoration and maintenance of imperial control over all of the Chinese territories and is not evidence of the acceptance of a de facto multinational world (chapter 8). Tillman, “Proto-­Nationalism in Twelfth-Century China?” Modern criticism of the work of Lu You and Liu Kezhuang similarly reads the patriotic tone of some of it as an early manifestation of nationalism. On Lu You and his poetic work, see, e.g., Qi Zhiping, Lu You zhuan lun; Qiu Minggao et al., Lu You pingzhuan; Guo Guang, Lu You zhuan; Zhu Dongrun, Lu You zhuan; Li Zhizhu, Lu You shi yanjiu. For a critical interpretation of the discourse on the non-Chinese in the work of Ye Shi, see Kondō Kazunari, “Sōdai Eika gakuha Yō Teki no ka-i kan.” In his recent dissertation Yang Shao-yun reinterprets the discourse on the non-Chinese in the work of select Tang and Song intellectuals as a way to focus Chinese identity on morality. Yang Shao-yun, “Reinventing the Barbarian.” 43. Recent articles and monographs in progress by Stephen West, Christian de Pee, Benjamin Ridgway, and Ari Levine will transform our understanding of the cultural history of Song cities. On gazetteers, see especially, Hargett, “Song Dynasty Local Gazetteers,” and Dennis, “Early Printing in China.”

28 introduction

covering all places associated with empire were brought together in notebooks, the logs of reading and conversation kept by increasing numbers of literati that were for the first time circulated in print in significant numbers between 1150 and 1250. Whereas the first three parts of this book thus examine the production and circulation of information about the court, the Chinese territories, and the borders respectively, part IV analyzes notebooks as textual embodiments of communication networks and explores their role in the commemoration of places associated with the Song Empire. Part I examines the role of the court in the dissemination of current affairs. It explores in two chapters the channels and networks through which the Song dynasty’s archives and court gazettes began to reach wider audiences. Both chapters in part I grapple with the paradox of the court’s imposition of policies of secrecy and its tacit acknowledgment of the diffusion of its archives. In chapter 1, I examine how and to what extent the vast Song archives were rendered legible in private and commercial editions. This and subsequent chapters further examine the impact of statist models on elite discourse. I propose that the release of state documents reinforced bureaucratic models of argument that spilled over from political and historical writing into other areas of social activity. Chapter 2 questions why measures proposed to enhance secrecy were not implemented even in the case of the more time-sensitive material in gazettes and takes a closer look at networking needs outside the capital. Part II examines how literati were taught to situate the Song present in a teleological history and on a normative map and argues that the construction of the transhistorical dimensions of the Song Empire was a crucial component in the fashioning of a literati imperial mission. Chapter 3 analyzes how the Song state was represented cartographically after the 1120s. Through a close reading of a historical atlas that circulated in a handful of commercial editions in the twelfth and thirteenth centuries, I explain the systematic and methodical overlay of present and past place names in Song cartography as part of a mission in which present jurisdictions are shown as continually part of a transhistorical empire rather than as elements in a truncated state or as carriers of local memory.



introduction

29

The map became for the first time in Chinese history part of a repertoire of tools of the collective memory. Literate audiences were taught how to produce a normative spatial image of the Chinese territories and a logic for reading it. The discrepancy between the ideal map and the status quo provided impetus to the activation of the imperial mission, as a growing readership of scholars and aspiring officials could now appropriate the symbolic power of the map in its printed form, a power that had once been the prerogative of generals and emperors. Following the analysis of the representation of the Chinese territories as a whole, part III examines the role of the expanding interest in borders and cross-border relations among literati. Chapter 4 argues that borders had become a central preoccupation among literati on the basis of a bibliographical comparison of works on border affairs before and after the Song dynasty and a survey of the presence of such works in twelfth- and thirteenth-century private collections. This chapter further suggests that literati interest in border affairs was a key factor in the systematization of publishing regulations on state documents. Governmental concern over the public disclosure of interstate and military information was commonly cited as a justification for policies and legislation aimed at the preservation of secrecy. Chapter 5 takes up the question whether the increased production and circulation of information on border affairs was accompanied by a reconceptualization of boundaries. From a comparison of eighth- and ninth-century descriptions of Tang borders with one mid-eleventh- and one early thirteenth-century rendering of Song borders, this chapter shows how new topographical models were added to a mix of models. Through a comparison between the 1044 description of Song borders in Summa of the Military Classics and the post-1127 private survey, this chapter further examines whether and how earlier topographical approaches could be applied to the new riverine border of the Huai River and in what ways the relocation of the border to the central heartland affected the literati’s perceptions of their role in border defense. The first three parts of this book focus on the formal and informal communication networks through which state archival documents,

30 introduction

court gazettes, diplomatic reports, military texts, and cartographic materials circulated, and through which was upheld and continually reinforced a vision of a unified empire whose existence depended on the participation of elites in the imperial project. These parts are principally concerned with the production and printing of current political information and to a lesser extent, in the case of maps and court gazettes, with its reception history. By following this work further into a systematic analysis of literati communication in notebooks (biji 筆記) we gain, in the final part IV, a better understanding not only of the reception of court-produced materials or their derivatives but also of horizontal communication patterns among lower-level officialdom and literati operating on the fringes of officialdom. Biji is a form of writing that was characterized primarily by the way in which data were collected and recorded; its authors emphasize that their works were the cumulative result of the personal collecting, recording, and evaluating of information acquired by hearsay and/or reading. Belonging to a genre whose very constitution is based on intertextuality and conversation, biji lend themselves particularly well to an analysis of the social and political networks through which information was exchanged and through which ties and identities were fashioned. The first chapter in part IV tabulates the temporal, geographical, and topical distribution of the increasing number of imprints of Song notebooks between the eleventh and thirteenth centuries. It further explores the sociopolitical dimensions of the Song biji phenomenon by examining the changing social and political backgrounds of those compiling and printing notebooks throughout this period. Even though particular attention is devoted to the circulation of notebooks in print, this chapter illustrates the close connection between manuscript and print publication through a case study of the networks involved in both the manuscript and print publication of Wang Mingqing’s Huizhu lu 揮塵錄 (Waving the duster). This case study suggests that the manuscript and print publication of notebooks were expressions of the networking impulse, which always started with a smaller group of personal connections but remained open to the incorporation of empire-wide ties.



introduction

31

Chapter 7 presents a case study of Wang Mingqing’s Waving the Duster. I here address the broader question to what extent and in what ways individual notebook compilers participated in an empirewide network of writers and readers. On the basis of a network analysis of the oral informants cited in Wang’s notebook, I analyze how large, how geographically diverse, and how well connected his network of informants was. This chapter also maps the temporal and topical distribution of the written sources cited in Wang’s notebook; it further examines centrality measures for authors or titles as potential markers of the authority and canonicity of particular types of materials. In this and the following chapter I further inquire into the broader political and cultural discourses articulated in Wang’s notebook and aim thus to throw new light on the cultural history of elite identities in imperial China. The conceptualization of empire was shaped by texts and conversations about the center, the peripheries, and the totality of the Chinese territories, the experiences of officials in the field, and the reports they and others produced thereof, but a sense of empire was also defined by the representation of the other beyond the imperial homeland. Basing my analysis on Wang Mingqing’s notebooks, in chapter 8 I extend earlier work on early Chinese as well as Song representations of ethnic others. Through a linguistic survey of the diverse representations of the other in the same series of notebooks, we can get a sense of the availability of the diplomatic texts translated and discussed in the work of Herbert Franke, Wang Gungwu, David Wright, and others44 to literati audiences, and an understanding of the new meanings that traditional ethnonyms discussed by Poo Muchou and Nicola Di Cosmo45 acquired. We also gain insight into the ways in which the other, specifically the Jurchens, were discussed in relation to the self in contexts including but not limited to the diplomatic and by literati not directly involved in 44. Franke, “Treaties between Sung and Chin”; Tao Jing-shen, “Barbarians or Northerners,” 69; Wang Gungwu, “The Rhetoric of a Lesser Empire”; Wright, The Ambassadors Records, and From War to Diplomatic Parity in Eleventh-Century China. 45. Poo Mu-chou, Enemies of Civilization; Di Cosmo, Ancient China and Its Enemies.

32 introduction

policy making. This chapter concludes with a discussion of how different representations of the other figured in literati attachments to the imperial mission. In sum, this book proposes a new answer to the question of the continuity of empire in the second millennium of Chinese history by surveying and analyzing the production, circulation, and, to some extent, the reception of current affairs by literati networks consisting of officials and scholars on the fringes of officialdom. Drawing upon the historical sociology of empire, the history of empire and information, and the cultural history of the book and reading, it addresses such questions as the political meaning of communication and of the use of print, and the role of literati communication networks in the formation and maintenance of an imperial mission. To what extent the formation and expansion of such networks in Song times structured the relations between literate elites and the state in later centuries (or to what extent such networks dissolved) remains to be explored in comparative and longue durée work; in many ways this investigation is just a starting point.

Pa r t I Contemporary Dimensions of Empire The Court

On e The Dissemination of the Archives and the Formation of the Late Imperial Archival Mentality

P

roducing written documents, collecting them, and organizing them into a structured and durable archive are activities we tend to associate with bureaucracies.1 The history of bureaucratic organizations suggests that these three operations do not coincide. In medieval England, written documents increasingly substituted oral testimony and memory attached to material objects in the twelfth century. It was not until the thirteenth century that kings (their chief fiscal advisors in particular) and, following them, bishops, barons, knights, and peasants began to hold on to the documents that related to their business. Even by the end of the century, royal servants and monks were at a loss to find written evidence in the royal treasure chests and monasteries dispersed throughout the kingdom of such a crucial matter as English claims to Scottish land. Ease of access to a permanent archive at the central level was not considered a priority, not by those at the center of political power and not by local power holders.2 Even though the medieval English scenario may appear surreal from an idealized perspective on the imperial Chinese bureaucracy, in Chinese his1. This chapter has been adapted from De Weerdt, “Byways in the Imperial ­Chinese Information Order.” 2. Clanchy, From Memory to Written Record, 152–53.

36

contempor ary dimensions

tory as well the production and collection of documents and their organization into accessible court archives did not coincide. The eleventh and twelfth centuries, however, marked a significant change, particularly in the latter area. Whereas parchment, ink, and wax technology spread the uses of writing in England and other parts of Western Europe in the twelfth century, in China it was paper, ink, and woodblock printing that provided the medium through which all manner of texts, including court archival documents and compilations, reached the political elite living in the Song capital and across its territories. The size of the literate elite in the aggregate can be estimated at some tens of thousands to hundreds of thousands of men over the course of the same century. Women, excepting female secretaries serving the emperor and other female members of the imperial clan,3 were excluded from the discussion of political affairs in writing. Written records and collections thereof had become essential to bureaucratic operations and to the legitimation of political authority in the centuries leading up to the unification of the Chinese territories in the third century BCE and under the early empires. The emperors and officials of the Song dynasty could rely on centuries of experience in the establishment of historical and dynastic archives; the archival institutions that emerged, however, differed from those of the past in the expansion and regularization of archival operations. More crucially, they provided access, by imperial fiat or through leaking, to the record of the ruling dynasty. Access to and the use of the dynastic record by elites outside the court and the local bureaucracy increased in the course of the eleventh and twelfth centuries and was unprecedented in Chinese history and arguably in world history. This chapter is principally concerned with the questions of how and why archival documents and compilations were leaked and disseminated beyond court and bureaucracy, and in what ways the accessibility of the archival record, limited though it may have been to many, affected literati thinking and writing about history, government, and other topics. 3. Deng Xiaonan, “Yanying zhi jian.”



late imperial archival mentality

37

Court Archives and the Institutionalization of Archival Operations The Song court relied on an information network that spanned the empire. Vertically, it extended down to the subadministrative levels of local society through village elders and mutual responsibility groups; and horizontally, into the border areas of Song territory and even the territory of neighboring states. Reports and memorials from the regional and capital offices, and records of court activities and high-level policy debates became part of the Song court’s central archives. The Song dynasty established an archival state, relying on the models of pre-Song Chinese empires. To a greater extent than ever before, documents were stored, sorted, and retrieved according to standardized procedures.4 The institutional history of the early Song period (the late tenth and eleventh centuries) was characterized by a strong drive toward centralization. This is evident in the institutional arrangement of the exchange of information between the court and local administrations. Soon after the establishment of the dynasty, the Memorials Office (Jinzouyuan 進奏院) was brought under the control of the central bureaucracy. The office, located in the capital, was a central node in the transmission of documents between the court and local governments. It oversaw the transmission of memorials and reports from the local level to the various boards and offices at the capital and was also responsible for the dissemination of court documents to the local levels of the bureaucracy. The eighth-century counterparts of the same office, the Capital Liaison Offices (also Jinzou­ yuan), had been managed by the appointees of regional commanders, who had autonomous control over sizable parts of the Tang Empire.5 The Tang Capital Liaison Offices had served as intelligence offices for the regional commanders. The second Song emperor, Taizong 太宗 (r. 976–97), established one central Memorials 4. For a full discussion of central and local archives and archival methods and stan­ dards in the Song, see Wang Jinyu, Songdai dang’an guanli yanjiu. 5. Li Bin, Tangdai wenming yu xinwen chuanbo, chap. 1.

38

contempor ary dimensions

Office,6 officially housed under the Chancellery (Menxia sheng 門 下省) and staffed by court officials. Accordingly, the mission of the office changed to servicing the court’s interest in the gathering, sorting, and selective dissemination of information. The Memorials Office was a central transfer point in the empirewide communication and archival network. It served as a gatekeeper in the upward transfer of information. Reports from local officials, as well as tax registers, accounting books, and legal d ­ ossiers, were sent to the office and then forwarded to the appropriate board or other central agency. The receiving agency reviewed the materials, made recommendations for further action, and selected materials for archival storage. Items were stored after three to ten years and were periodically culled. The central archives were housed in one agency during the Northern Song period. After the relocation of the court to Hangzhou, two separate archives, one for the Six Boards and another for the Three Departments (the Secretariat-Chancellery and the Department of State Affairs) and the Bureau of Military Affairs, were established.7 The preservation and sorting of local and central documents supported the work of court officials and clerks in central bureaucracy offices. Ministers and their subordinates consulted the archives in preparing for policy discussions, as did court academicians charged with compiling archival collections and writing the dynasty’s history. Moreover, the Memorials Office played a crucial role in the downward transmission of official news and records. Imperial decrees were sent down through the office and archived locally, as were law codes and irregularly published administrative and penal regulations. The Memorials Office also circulated the court gazette (chaobao 朝報, dibao 邸報) to officials in the provinces on a regular basis, according to one report, every ten days.8 The gazette included excerpts of decrees and memorials, listings of assignments and 6. Zhu Chuanyu, Songdai xinwen shi, 14. 7. For an overview of the institutions and operations of the central archives in Song times, see Wang Jinyu, Songdai dang’an guanli yanjiu. 8. The gazette was published more frequently, with the interval being changed from every five days to every day. Papers were sent out in bundles to local governments every ten days. Zhu Chuanyu, Songdai xinwen shi, 19, 42–45.



late imperial archival mentality

39

dismissals, a schedule of court audiences, and other news items related to the imperial family and court activities. Local and central archives complemented each other to a certain degree. Local archives provided material to fill in lacunae in court holdings. They were also important in the compilation of local descriptions (tujing 圖經, fangzhi 方志), which were submitted to court. The materials collected in the archives provided the raw material for a series of historiographical interventions. The progressive abstracting and reworking of the archival record into chronologically or thematically arranged compilations rendered the mass of information stored on shelves legible to the court itself; these interventions also opened the archives up, albeit in a mediated fashion, to those who would have been excluded from them in earlier centuries. As the process of abstracting progressed, the compilations became the archives. The materials and even the earlier larger compilations on which the final compilations were based were discarded. The incentive to keep the mass of originals further declined because the leaking and printing of archival compilations provided backup copies when fires or other disasters destroyed court holdings. Court archival compilations can be subdivided into two types according to frequency of compilation: ad hoc commissions and regularly published compilations. The court occasionally commissioned individual court officials and famous scholars to participate in ad hoc projects. One example is Cefu yuangui 冊府元龜 (Models from the archives, 1013), a vast collection of official documents commissioned under Emperor Zhenzong 真宗 (r. 998–1022). Unlike those discussed below, this compilation was transhistorical and borrowed heavily from secondary sources. It provided the emperor and his advisors access to a huge amount of textual information based on an elaborate classification scheme inherited in part from courtsponsored encyclopedic works compiled under the Sui and Tang dynasties. Like these works it fulfilled a didactic function. Each subject heading contained “exempla or paradigma, whose heuristic binding power comes from a meticulous negotiation between multiple concrete cases and the categorical synopsis in the prefaces.”9 9. Sung, “Between Tortoise and Mirror,” 140–41.

Court Diary Office (Qijuyuan/Jizhu’an)

Daily Calendar Office (Rilisuo)

History Institutes (Shiyuan)

Collected Statutes Office (Huiyao suo)

Fig. 1.1 The process by which archives evolve into different types of archival and historical compilations.

sagely government and precious instructions compilations (shengzheng/baoxun) [commissioned]



History Institutes (Shiyuan)

veritable records dynastic history (shilu) ( guoshi)

collected statutes (huiyao)



court diaries daily calendar (qijuzhu) (rili)



late imperial archival mentality

41

More important in the preservation of Song archival materials and their timely dissemination was the ongoing work of compiling archival and historical collections by permanent offices or regularly assembled committees. Figure 1.1 represents the process of digesting and assimilating state documents into these compilations. Court diaries (qijuzhu 起居注) were comprehensive, continuing records of activities at court as well as important events gathered from local reports. The primary sources for these diaries consisted of the notes taken by imperial diarists (xiuzhuguan 修注官), whose job it was to follow the emperor around and record his activities and exchanges; records of conversations between the emperor and high officials; and reports submitted by all offices in the capital. Capital offices sent in their reports at regular intervals ranging from every five days to every year, depending on the office and the type of report requested. The Court Diary Office (first Qijuyuan 起居院, later Jizhu’an 記注案) was a permanent office housed under the Secretariat-Chancellery (Zhongshu-Menxia 中書門下). When abiding by office rules, its compilers sent in monthly diaries to the Historiography Institute (Shiguan 史館) for storage or further processing. They arranged these monthly diaries in chronological order by year, month, and day. Using a template first set in 1095 and thereafter changed periodically, they entered the activities for each day, starting with the emperor’s comings and goings and the review of memorials submitted by the Six Boards and the Bureau of Military Affairs and ending with the meteorological and astronomical observations of the day. Reports concerning auspicious and inauspicious phenomena and local mores from across the empire, as well as population statistics, were entered at the end of each month or each year.10 10. This paragraph and the overview below follow Cai Chongbang, Songdai xiushi zhidu yanjiu. For the compilation of court diaries, records of administrative affairs, and the daily calendar, see also Wang Deyi, “Songdai de qijuzhu yu shizhengji zhi yanjiu” and “Songdai de rili he yudie zhi yanjiu.” The compilation of the Song dynastic history is briefly discussed in Lamouroux, Fiscalité, comptes publics et politiques financières. For a selective overview of twentieth-century Chinese historiography on Song official history projects, see Liu Liankai, “Songdai xiu shi zhidushi yanjiu shuping.”

42

contempor ary dimensions

The daily calendar (rili 日曆) was the work of editors and assistant editors (zhuzuolang, zuolang 著作郎, 左郎) attached to a permanent Daily Calendar Office (Rilisuo 日曆所). The office at first came under the Compilation Bureau (Bianxiuyuan 編修院) of the Chancellery, but later reported to the Draft History Branch (Guoshian 國史案) in the Palace Library (Bishusheng 秘書省). The process of compilation, though occasionally interrupted, as under the regime of Qin Gui 秦檜 (1090–1155), was continual.11 The editors at the Daily Calendar Office combined the court diaries submitted by the Court Diary Office and other materials, such as records of administrative affairs (shizhengji 時政紀) regularly submitted by the Bureau of Military Affairs and the Secretariat and biographical materials for the denizens of the court and high officials, into more coherent chronological accounts. These in turn became the basis for the historical accounts of one or several reign periods. From time to time, usually upon the death of an emperor, the editors combined the draft calendars into a set. The veritable records (shilu 實錄) abstracted the materials collected in the combined calendars of an emperor’s reign into a more concise documentary history. They supplemented court archival materials with private biographical and historical records and included, along with the biographies of deceased prominent figures appearing in the calendar, monographs such as descriptions of neighboring states and peoples. The committees that compiled the veritable records and the draft histories based on them were, with the exception of a short period in the eleventh century, not housed in permanent offices. They assembled by decree at frequent intervals and were housed in offices regularly charged with the storage and compilation of archival records. Throughout the Southern Song period, for example, the court called into existence Veritable Records Institutes (Shiluyuan 實錄院) and Draft History Institutes (Guoshi­y uan 國史院) and placed them under the authority of the Palace Library and its Historiography Institute. Veritable records were compiled for a total of fourteen emperors in a continuous 11. The Court Diary Office went unstaffed between 1143 and 1155. See Hartman, “The Making of a Villain,” 70.



late imperial archival mentality

43

run from the first emperor, Taizu 太祖 (r. 960–76), down to Li­ zong 理宗 (r. 1225–64). Only for the short reigns of the last four emperors, covering the years 1265–79, were veritable records not produced.12 The court invested less often in the drafting of dynastic histories. These combined the archival records of several reigns, with the number of reigns varying. Examples are Liangchao guoshi 兩朝國史 (Draft history of the two reigns), which concerns Emperors Renzong 仁宗 (r. 1023–63) and Yingzong 英宗 (r. 1064–67); Zhongxing sichao guoshi 中興四朝國史 (Draft history of the four reigns of the intermediate revival); and the unfinished Jiuchao guoshi 九朝國史 (Draft history of the nine reigns [of the Northern Song period]). The draft histories formed the basic materials for the dynastic history of the Song house and followed the same layout as all previous dynastic histories. Relying heavily on the annalistic record, the biographical records, and the monographs in the veritable records of each reign period, they condensed this material and filled in the gaps with extra biographies and monographs. Like the veritable records, the draft histories of the Song were complete through the reign of Emperor Lizong. There were precedents for the genres I have reviewed thus far in the Tang dynasty and earlier, but the Song court departed from Tang practice by institutionalizing archival and historical compilation in permanent or regularly established bureaus. This was also the case for the compilation of collected statutes (huiyao 會要). The collected statutes classified state documents according to a scheme of ten to twenty broad subject categories, including the imperial court, sumptuary law, ritual, the civil service with its supporting examination system, the economy, punishments, the military, and foreign peoples. Within each category excerpted documents were then arranged chronologically. Offices for creating these classified archival 12. For a discussion of the compilation of The Veritable Records of Qinzong, see Hartman, “The Reluctant Historian.” For an extensive recent study on the compilation, sources, and genre conventions of the Song veritable records, see Xie Guian, Song shilu yanjiu; because this work appeared after I completed the manuscript, I have not incorporated its findings.

44

contempor ary dimensions

collections (huiyao suo 會要所) were established between the early eleventh and thirteenth centuries and oversaw the compilation of documentary sources covering Song history through the reign of Lizong. The Collected Statutes Office was for most of the relevant time period subordinate to the Palace Library, as were the Veritable Records and Draft History Institutes.13 The five types of compilations discussed above (the court diary, the daily calendar, the veritable records, the draft history, and the collected statutes) were predominantly court compilations. Private initiatives in these genres were rare, and even these were undertaken by officials affiliated with other court compilation projects. An example is Guochao huiyao zonglei 囯朝會要縂類 (The collected statutes of the reigning dynasty comprehensively excerpted and classified, 1204–8),14 which Zhang Congzu 張從組 compiled while serving as a compiler in the Palace Library, the Draft History Institute, and the Veritable Records Institute even though assembling The Collected Statutes was not part of his official job assignment. Matters were somewhat different in the compilation of collections on “sagely government” (shengzheng 聖政) and “precious instructions” (baoxun 寳訓). These compilations were typically the work of court officials, but in several cases an official undertook them on his own initiative, submitting them to court upon completion. They were based on a careful selection of excerpts from the records of administrative affairs (shizhengji) kept by high court officials in the Bureau of Military Affairs and the Boards, as well as court diaries and daily calendars. Variously covering a short reign period in one reign, an entire reign, or several reigns, they were presented as an officially sanctioned selection of imperial pronouncements and policies. As no special offices were set up to work on these compilations, they appeared sporadically.15 13. In addition to Cai Chongbang, Songdai xiushi zhidu yanjiu, see Wang Jinyu, Songdai dang’an guanli yanjiu. 14. Cai Chongbang, Songdai xiushi zhidu yanjiu, 168–70. 15. Wang Deyi, “Songdai de shengzheng he baoxun zhi yanjiu.” For a comparison of the historiographical features of precious instructions and other archival compilations with the work of Ouyang Xiu and Sima Guang, see Sung, “Between Tortoise and Mirror.”



late imperial archival mentality

45

The Myth of Secrecy Beginning in the eleventh century and with growing frequency and explicitness in the twelfth, the Song court issued regulations on the circulation of the archival and historical compilations discussed above.16 With the exception of the collections about sagely government and precious instructions, all official archival compilations were banned from circulation outside of their respective offices and the court repositories where completed copies were stored. The prohibitions make little specific reference to court diaries or daily calendars, but as all compilations drawing on them could not circulate, nor could documents be taken outside the archival and historiography offices, we may infer that the ban covered these materials as well. Until the late eleventh century, the regulations governed the storage and internal transmission of materials, the movements of copyists, and the internal correspondence of the officials of the historiography institutes. They made no reference to the deliberate external circulation of court archival and historical compilations. Concerns over the copying and printing of veritable records, draft histories, and collected statutes first emerged in a decree issued in 1090. There is no evidence that printed copies of either type of compilation circulated at this time, but the anxiety that print technology would lead to the widespread dissemination of secret court documents across the Song Empire and abroad, as expressed in the decree, was characteristic of these years.17 This anxiety abated somewhat over time, as we can see in the publication of Qingyuan tiaofa shilei 慶元條法事類 (The classified laws of the Qingyuan period) around 1202. This code mentions the printing of collected statutes only because these circulated in print. Veritable records and draft histories did not circulate as printed compilations, and The Classified 16. For a chronological list of Song publishing regulations on court archival and historical compilations, see De Weerdt, “Byways in the Imperial Chinese Information Order,” 155–56. 17. SHY, Xingfa, 2.38. Egan, “To Count Grains of Sand”; De Weerdt, “What Did Su Che See?”

46

contempor ary dimensions

Laws apparently recognizes that fact. However, copying and excerpting from the latter sources was punishable to the same degree as the printing of collected statutes. This may have had to do with the fact that, as discussed further below, excerpts from the veritable records and draft histories were by this time routinely included in printed encyclopedias and history textbooks. Clues about what had led to the commercialization of state documents and archival compilations appear in memorials submitted in 1122 and 1201.18 The commercial reedition of Lin Fu’s 林虙 ( j. 1097) Shenzong huangdi zhengji gushi 神宗皇帝政績故實 (Truths regarding the government of Emperor Shenzong) under the title Cichang xin fan 辭場新範 (A new model for the examinations), described in the memorial of 1122, illustrates the historical link between the growth of commercial publishing in the late eleventh and twelfth centuries and the simultaneous expansion of the number of students preparing for the civil service examinations. During the Chongning 崇寧 reign (1102–6), Lin Fu compiled a series of notes on the reign of Emperor Shenzong, based on sources handed down to him by his uncle Lin Xi 林希 ( j. 1057). Lin Xi had been involved in the compilation of The Veritable Records of Shenzong and had also edited Precious Instructions of the Two Reigns (covering the two preceding reigns of Emperors Renzong and Yingzong). Lin Fu divided all the materials he collected into one hundred categories, reportedly in imitation of the precious instructions compiled for the first five Song emperors. Besides the original and commercial titles mentioned above, his compilation circulated as Shenzong shengzheng 神宗聖政 (Shenzong’s sagely government), Yuanfeng shengzheng 元豐 聖政 (The sagely government of the Yuanfeng period), and Shenzong baoxun 神宗寳訓 (Shenzong’s precious instructions).19 It is likely that the original title, Truths regarding the Government of Emperor Shenzong, was changed after Lin Fu had submitted his work to court. The only alternative title specifically mentioned in the 1122 18. SHY, Xingfa, 2.87; ibid., Xuanju, 5.25. 19. Wang Yinglin, Yuhai, 49.10b–11a; Zhang Ruyu, Qunshu kaosuo, 17.14a; Chao Gongwu, Junzhai dushuzhi jiaozheng, 269, 1229; SS, 203.5107; Liu Zhaoyou et al., Song shi “Yiwen zhi” shibu yiji kao, 355.



late imperial archival mentality

47

memorial, A New Model for the Examinations, suggests that abbreviated archival collections such as Lin’s found an eager audience among students competing in the civil service examinations, or, at least, that commercial bookshops targeted this group. The 1201 memorial suggests that, in the metropolitan examinations, examiners presumed familiarity with major events and figures covered in the draft histories, collected statutes, and veritable records. This requirement caused a ripple effect in schools and examination halls across the empire. Twelfth- and thirteenth-century examination manuals incorporated state documents, thus reflecting the emphasis on students’ ability to discuss policy questions on current affairs.20 At the same time, the 1201 report also suggests that access to state archival materials was far from universal: The draft histories of our dynasty and all materials such as the veritable records and collected statutes should be very carefully protected. There are detailed legal restrictions on the collection of these materials by private persons. However, the sons and brothers of high court officials sometimes gain illicit access through their fathers and brothers, and powerful families violate the ban on copying them. As for poor scholars in distant places, how can they know these sources! Recently, however, events occurring in the present reign of our dynasty have been cast into policy questions without revealing their origins and outcome. This is tantamount to making poor scholars in distant places responsible for becoming thoroughly familiar with books they have never seen. This does not accord with human feeling! I ask that from now on, when policy questions relate to events occurring in the present reign, they should be clearly formulated and the origins and outcome of those events be properly divulged. I hope that in this way poor scholars from rustic places will not be experiencing difficulties in isolation.21

High court officials, such as court academicians and ministers serving as examiners in the metropolitan examinations, had ready access to archival materials, but, as the author suggests, this was not 20. De Weerdt, Competition over Content, chaps. 4 and 6. 21. SHY, Xuanju, 5.25.

48

contempor ary dimensions

necessarily the case for provincial candidates. The request that examiners be asked to clarify wherever reference is made to archival materials to which not all candidates have access acknowledges that those with access would continue to allude to current affairs. This is the closest one can come to an admission that secrecy had become a myth. The myth mattered. Secrecy was embodied in the Chinese name of the Palace Library, Bishusheng (bi /m   i means secret), just as segretari in sixteenth-century Venice were in name supposed to protect the secrecy of the documents under their control.22 Secrecy protected the sanctity of the imperial person and was also seen as crucial in preventing factional politics both at court and beyond. The tacit and gradual accommodation to the leaking and reproducing of archival compilations may be explained by a gradual appreciation of the informational and networking needs of literati throughout the empire. This pragmatic adjustment in court-literati relations in the area of political information may have been crucial in ensuring the continued collaboration of the vastly expanded elite of examination candidates.

Text Technologies The court’s accommodation to the leaking and dissemination of archival compilations needs to be read in the context of abundant evidence of the copying, reproduction, and further dissemination through manuscript copying and commercial printing in the twelfth and thirteenth centuries. Bans on the copying, possession, and printing of archival materials proved ineffective against the range of text technologies that were marshaled to render the record of the ruling dynasty legible and useful to literati. Not only examiners, but also the tens of thousands who prepared for the examinations, those who taught them, those who prepared textbooks for them, and those who, like the amateur geographers discussed in chapter 4, worked as independent scholars in search of patrons, felt the need for primary source materials on questions of 22. De Vivo, Information and Communication in Venice, 50–51.



late imperial archival mentality

49

current relevance. Compilations of archival materials also proved useful to the thousands of officials working in central government offices whose positions did not come with the privilege of access to the archives or those posted in the provinces. They too looked for cases, anecdotes, or authoritative quotes attributable to Song emperors and statesmen while composing reports of various kinds. This section on text technologies examines the processes and methods through which the manuscript archival compilations in court repositories were delivered to readers in a range of material objects. Text technologies here not only include the manuscript and print production of derivative texts but also the collecting, classifying, and excerpting of both originals and derivatives.

Collecting The centralization and standardization of archival operations and the issuing of prohibitions on the dissemination of the court’s archival record reinforced the authority attributed to those who controlled it. The connection between political authority and control over the archive became all the more visible at times when court control over the dynastic archive was challenged. Such challenges mainly came from two directions. First, the number of private collectors and their holdings grew in the course of the eleventh and twelfth centuries; so I begin by assessing to what extent archival compilations appeared in private collections and to what extent private holdings in this area were visible and/or accessible to literati in the capital and beyond. Second, court archives and libraries suffered recurring losses due to fires and theft; major losses further resulted from the dispersal of court possessions during the occupation of Kaifeng in the 1120s. In the decades that followed the Song court’s relocation to the south, the recovery of its archival record became a high priority. The restoration of the archival record was a precondition for the Song court’s continuation of its own history, and the comprehensive search for its complete restoration was a sign of its refusal to give in to the logic of the dynastic cycle according to which the victor wrote the history of the preceding dynasty by appropriating its records. In the second part of this section on collecting, the history of collection campaigns

50

contempor ary dimensions

is briefly reviewed to determine whether these should be read as indicators of the centralization of archival holdings or as an acknowledgment of a distributed archive. The number of private collectors, many with libraries of at least ten thousand juan, went up significantly during the Song dynasty. Recent estimates of the number of documented collectors range from over five hundred to seven hundred. In contrast, the numbers for Tang dynasty collectors vary between thirty and one hundred.23 We know of at least sixty catalogs of Song private collections,24 but only three of the extant ones are complete. They are You Mao’s 尤袤 (1127–94) Suichu tang shumu 遂初堂書目 (Catalog of the Pavilion of Following One’s Original Intent, ca. 1190), Chao Gongwu’s 晁公武 (ca. 1104–83) Junzhai dushu zhi 郡齋讀書志 ([Chao] Junzhai’s record of books read, 1151–1240s), and Chen Zhensun’s 陳振 孫 (ca. 1186–ca. 1262) Zhizhai shulu jieti 直齋書錄解題 ([Chen] Zhizhai’s annotated catalog, ca. 1249). Veritable records and collected statutes were abundant in private collections of state archival materials. All three of the collectors just mentioned possessed uninterrupted sets of the veritable records from the reign of Emperor Taizu through the late Northern Song or early Southern Song. Chao Gongwu’s set was complete through the reign of Emperor Zhezong 哲宗 (1086–1100; completed ca. 1135); You Mao’s went up to the reign of Emperor Qinzong 欽宗 (1126–27; 1168); and Chen Zhensun’s, reflecting the later date of his catalog, included Gaozong’s (1197, 1202) and Xiaozong’s Veritable Records (1202). Not only did the collectors hold the veritable records for each reign; they also owned copies of various editions. You Mao held three editions of Shenzong’s Veritable Records, including the Shao­x ing edition completed in 1136; Chen Zhensun held two. Chen further notes that the revised edition put out under the reformist regime of Zhezong, who wished to revive Wang Anshi’s 王安石 (1021–86) 23. Endymion Porter Wilkinson puts the number of Tang collectors between twenty and thirty. See his Chinese History, 265. Li Yu’an and Chen Chuanyi (Zhongguo cang­ shujia cidian) list 42 names for the Tang and 149 for the Song. On the higher estimates of Tang and Song collectors, see Ren Jiyu, Zhongguo cangshulou, vol. 1, 698, 750; Fu Xuan­ cong and Xie Zhuohua, Zhongguo cangshu tongshi, vol. 1, 215, 227, 350. 24. Fan Fengshu, Zhongguo sijia cangshu shi, 121–24.



late imperial archival mentality

51

legacy, was widely disseminated. With both the original text in black ink as well as new additions in red ink and deletions marked in yellow, this version exerted a broad appeal.25 All the collectors held, in addition to complete runs of the veritable records, comprehensive sets of collected statutes. All three collections cover the period from Taizu through the reigns of Emperors Gaozong or Xiaozong. Chao Gongwu and Chen Zhensun both owned a combined edition that included all previous editions. Chao lists among his holdings The Collected Statutes of the Reigning Dynasty Comprehensively Excerpted and Classified, a combined edition of documents from the first eleven reigns, classified by subject area. A sequel to this work, compiled by the noted historian Li Xinchuan 李心傳 (1166–1243), appears in Chen’s catalog under the same title as Zhang Congzu’s work. Printed in Sichuan, the sequel added state documents from two additional reigns, and was accordingly given the alternative title Shisan chao huiyao 十三朝會要 (The collected statutes of thirteen reigns).26 The draft history also figures prominently in the three col­ lections. Draft histories of individual reigns were often combined into cumulative editions. Most collectors seem to have had access only to cumulative editions. Thus Chao Gongwu and Chen Zhensun held only cumulative titles, while You Mao, as an editor in the Historiography Institute, was able to get hold of draft histories of individual emperors or draft subsections of these histories. Both You and Chen owned complete sets of the draft histories for the Northern Song period. Since the last of these titles, Sichao guoshi 四 朝國史 (The draft history of the four reigns), was not completed until 1186 and the first edition of a combined Southern Song draft history did not come out until the 1250s, their holdings were remarkably current.27 25. ZZSL, 4.129–30. 26. Ibid., 5.163. For a fuller description of this title, see Cai Chongbang, Songdai xiushi zhidu yanjiu, 167–68. For an investigation into the titles and scope of both Zhang Congzu’s and Li Xinchuan’s compilations, as well as the presence of these texts in the current edition of Song huiyao, see Miao Shumei and Wang Yunhai, Song huiyao jigao: chong ru, 440–59. 27. Cai Chongbang, Songdai xiushi zhidu yanjiu, 126–48.

52

contempor ary dimensions

Court diaries and daily calendars were less in evidence but not absent from private collections. The legal record suggests that there was less interest in the lengthy and undigested court diaries. In these genres collectors focused on obtaining the latest editions for which the veritable records had not yet been compiled. All three collectors owned copies of the daily calendar of Gaozong’s 高宗 reign (r. 1131– 62), which overlapped with the careers of Chao Gongwu and You Mao. You Mao possessed five titles in the daily calendar genre and, most likely due to the time he spent as an editor at the Palace Library, had also obtained a copy of the latest edition of the court diary, Xiaozong qijuzhu 孝宗起居注 (Xiao­zong’s court diary). Additional evidence that twelfth-century Song scholar-officials sought and found access to collected statutes, veritable records, and draft histories through to the latest compilations comes from Wang Mingqing’s 王明清 (1127–after 1214) notebooks. Wang had benefited from his family collection but had to reconstitute most of it. In Huizhu lu 揮麈錄 (Waving the duster), a set of serialized notebooks compiled between the 1160s and the 1190s, he mentions that he “consulted” ( yue 閱) the collected statutes28 and he refers to draft histories and veritable records up to the reign of Huizong. When compared to the titles listed in the “Bibliographical Survey” of Song shi 宋史 (The history of the Song dynasty; completed in 1345), which was based on the holdings of court libraries, it becomes evident that in the twelfth and thirteenth centuries private holdings in state archival materials kept pace with their production at court.29 Wu Cheng 吳澄 (1249–1333) was only slightly exaggerating when he wrote in his summation of the major achievements of the fallen Song dynasty: “During the three hundred years of the Song, printing became a business and printed editions spread across the empire. All that was stored in the court collections became available in private collections.”30 28. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.7. 29. Tables of private collectors’ holdings and court holdings as reflected in the “Bibliographical Survey” are included in De Weerdt, “Byways in the Imperial Chinese Information Order,” 160–69. 30. Wu Cheng, Wu Wenzheng ji, 34.19a. Quoted without reference in Zhang Xiumin, Zhongguo yinshuashi, 58.



late imperial archival mentality

53

Who were these collectors and how did they assemble their collections? Private literati collections were not limited to the higher echelons of the bureaucracy. Of the four collectors mentioned thus far, You Mao was the only one who spent a considerable amount of time at court. Chen Zhensun, a native of Huzhou 湖州 (in presentday Zhejiang), did most of his collecting as a local administrator in the 1220s and 1230s. In 1244 he was nominated to the post of vicedirector of education ( guozi siye 國子司業) in the central government and subsequently retired as academician-in-waiting at the Baozhang Pavilion (Ningzong’s library, established in 1226). This last honor came toward the end of his life, by which time his collection was already complete. Chao Gongwu’s collecting efforts followed the same pattern. Having moved to Jiazhou 嘉州 (now Chengdu) after the Jin invasions of the 1120s, he obtained the jinshi degree in 1132 and served in a series of local positions. By the time he was nominated to a position in the capital to become censor (shi yushi 侍禦史) in the early 1160s, he had already finished a first draft of his catalog in 1151. The way in which these two men formed their collections demonstrates the importance of exchanges among literati in the building of collections. Many top-ranking collectors granted access to their collections and some built facilities to house visitors and provided stipends to interested students.31 Chao Gongwu inherited a large collection from one of his superiors in office. Jing Du 井度, who hired Chao as his personal secretary while serving as fiscal intendant for Sichuan in the early 1140s, left him his collection right before his death.32 Jing Du never played a role at court; his collection is another example of the sizable collections amassed by local officials. Chen Zhensun supplemented his book buying in local book markets with visits to fellow collectors in the places where he held office. A few let him copy materials to supplement his own library.33 31. Zhang Xiumin, Zhongguo yinshuashi, 204–5. 32. Pan Meiyue, Songdai cangshujia kao, 162–64. 33. Qiao Yanguan lists the small number of titles for which Chen Zhensun mentions specifically that he either bought them or copied them. See his Chen Zhensun xueji, 33.

54

contempor ary dimensions

Lending and copying facilitated the transmission of large collections such as court archival and historical compilations. An edict from 1158 prohibiting the private transmission of veritable records by high court officials to their families addresses the widespread practice of the transfer of court compilations into private collections.34 The private collections of several high court officials and others were renowned for their strength in archival materials. Such was the case of Zheng Yin 鄭寅 (d. 1237), a relative of Zheng Qiao 鄭樵 ( j. 1169), an administrator in the Bureau of Military Affairs under Ningzong. Zheng Yin himself rose to the position of director of the Left Office of the Department of State Affairs in the mid1230s. Chen Zhensun consulted Zheng’s collection while developing his own.35 Wang Mingqing should have inherited the large collection assembled by his father, the eminent court historian Wang Zhi 王銍 (?–ca. 1144). Due to a series of losses incurred through the intervention of a local official taking advantage of the dislocation caused by war and due to the alleged greed of family members, he instead rebuilt the family collection through purchase, copying, and exchanges with other collectors. He mentioned more than thirty collectors and shared contact details in his notebooks with others with an interest in text (as well as art) connoisseurship. His contacts ranged from a lowly clerk to a degree holder who had held court positions and the descendants of former grand councilors. Most had lived in places where Wang had lived or served in office. In his notebooks Wang Mingqing put his personal and family collection and the collections of others on display. He listed items of particular interest including court documents. By transcribing the text he also published some rare archival materials. He listed, for example, a single-sheet memorial with the imperial approval of Emperor Yingzong 英宗 (r. 1063–67) (obtained from a contemporary by the name of Liang Caifu 梁才甫), a letter drafted by Emperor Gaozong, an abbreviated copy of a court diary compiled by 34. SHY, Zhiguan, 18.66; Yaolu 180.540. 35. On Zheng Yin’s collection and Chen’s descriptions of it, see Pan Meiyue, Songdai cangshujia kao, 216–18. Qiao Yanguan, Chen Zhensun xueji, 28, 33.



late imperial archival mentality

55

Zeng Bu 曾布 (1036–1107), and a memorial submitted under Zhenzong’s 真宗 reign (r. 997–1022). Among the transcribed materials were Xu Fu’s 徐俯 (1075–ca. 1141) record of Huizong’s words, a hymn performed by court officials for Huizong, and a memorial from Zhenzong’s reign. Many of these materials did not necessarily originate in court archives, but may have been copies held by the original authors and their family members. The attraction of such items, especially items like the court diary of Zeng Bu, lay in the fact that they either had been source materials for the court’s archival and historical compilations or could serve to amend the record as it stood. The private collections of archival materials and the willingness of literati to reveal their holdings in catalogs and notebooks demonstrate at a minimum that the court and the central bureaucracy proved unable to stem the flow of its archival record. The myth of secrecy is further exposed through the court’s relationship with collectors small and large. Court officials usually only took action against the private copying and possession of archival materials when the perpetrators were already under investigation on more serious political charges. Although court officials sometimes confiscated materials subject to restricted access and prosecuted those who held them illicitly, the practices of the institutes charged with protecting these materials provided justification for the illegal activities of private collectors. Officials in the Palace Library and Historiography Institute relied on private collectors to supplement deficiencies in the court repositories. The court solicited private donations of books or urged private collectors to submit titles for copying by government clerks. Archival compilations frequently figured among the titles recovered during collection campaigns.36 36. This and the following paragraphs are based on De Weerdt, “The Discourse of Loss,” where the collection campaigns, the motivations behind them, and the response they elicited are discussed in greater detail. In his study of the Southern Song Palace Library, John Winkelman discusses the initiatives undertaken during Gaozong’s administration to restore the imperial collections. He mentions numerous examples of donations of draft histories, veritable records, collected statutes, and precious instructions. See his “The Imperial Library in Southern Sung China,” 28–29. See also Wang Jinyu, Songdai dang’an guanli yanjiu, 116.

56

contempor ary dimensions

Searches for items missing from the palace libraries had been conducted repeatedly since the establishment of the Song dynasty in the late tenth century. The best documented campaigns were those conducted in the 1130s and 1140s, which were motivated by the restoration government’s desire to recapture the dynasty’s record following the dispersal of the court depositories in the late 1120s. Early efforts to reconstitute the holdings of the Song Imperial Library focused on archival and historical compilations such as the veritable records, the draft dynastic histories, the collected statutes, and abridged compilations of “precious instructions.” In 1131, for example, the new Imperial Library in Shaoxing took in partial and complete copies of the archival compilations that had been produced up to 1126, including the veritable records, the draft histories, the collected statutes, and the precious instructions of the first six reigns of the Song dynasty. And in 1132 and 1133 vice-directors of the Imperial Library and other court officials in Lin’an (Hangzhou) prompted the emperor to issue decrees requesting contributions from specific collections and collectors. In all but one case, the requests specifically demanded copies of the draft histories, the veritable records, and the collected statutes (appendix 1, table 2). The methods of acquisition illustrate the extent to which the court accommodated the collection practices of the scholar-official elite. In decrees targeted exclusively at the collections of families or individuals the court suggested that the owners submit the works for copying. The originals were to be returned to the owners upon the completion of their duplication. To those who submitted titles the emperor granted rewards. At first, rewards were decided upon on a case-by-case basis, but in 1146 a graded scale of rewards based on earlier models was devised, printed, and broadly disseminated. Emperor Gaozong then invoked a scale developed under the second Song Emperor, Taizong, as a model, but the scale was modified in accordance with the changes that had taken place in Chinese book culture during the intervening 150 years. It differentiated between officials and scholars (shiren 士人). Officials who made large donations were given promotions; scholars who made large donations were either permanently or temporarily absolved from taking the prefectural civil service examinations. They thereby obtained direct



late imperial archival mentality

57

access to the triennial examinations at the metropolitan level. The addition of the category of the scholar-collector underscored the court’s recognition of the importance of the rapidly expanding class of examination candidates in the circulation of cultural goods. The response to calls for contributions had at first been lukewarm. After more than ten years of successive calls for donations and copies the Imperial Library still lacked a substantial number of titles that were, according to its librarians, available in private collections. Petitioners then attributed the lack of success to the unwillingness of private collectors to contribute their treasures in the absence of clear and substantial rewards. Xu Du 徐度 (fl. 1150), who had served in the very offices that produced the veritable records and other court archival materials, similarly noted the unwillingness of elite families to lend their materials to the court.37 Wang Mingqing confirmed what is also obvious from the official list of donors: the total number of donors in the early stages was relatively small (appendix 1, tables 1 and 3).38 More drastic measures were put into place under the librarianship of Qin Xi 秦熺 (?–1161) in the mid-1140s. Apart from the aforementioned scale of rewards, Qin Xi put through measures shifting the burden of the campaign from the library itself to local officials who were now required to submit regular reports on their collection efforts.39 According to Mo Shuguang 莫叔光 ( j. 1163), who directed the Imperial Library in the mid-1180s when the fear of retribution for criticizing the son of Councilor Qin Gui had long subsided, Qin Xi’s directorship led to the restoration of the imperial collections.40 An upsurge in donations is also evident from tables 1 and 3 in appendix 1. These results came at a substantial cost, however. Qin Xi’s demand for greater authority for the Imperial Library in its supervision of the collection efforts of local administrators was not simply an answer to a collection campaign gone dormant; it fit into a concerted effort to scrutinize private collections and weed out 37. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.68–69. 38. Ibid., 1.9–10. 39. SHY, Chong ru, 4.27–28. 40. Ibid., 4.31; also Chen Kui, Nan Song guan’ge lu, xu lu (1178), 174.

58

contempor ary dimensions

those materials that were deemed out of sync with the new regime’s propeace stance. According to contemporary accounts the families of prowar advocates burnt entire collections, or, more frequently, letters, memorials, and historical accounts that could be deemed subversive in the eyes of local administrators or librarians at the Imperial Library. The materials were in several cases burnt as a matter of precaution, typically when the owners learned about the persecution of a political ally.41 Despite targeted attacks on selected political enemies, there is also evidence that suggests that the court stuck to its acquisition policy. One of the donors listed in the court’s register was Xu Zhong 許中 ( j. 1100), whose contribution of a title in the collected status category was also highlighted by Chen Zhensun. Chen Zhen­ sun, who as we have seen above also collected in this area, pointed out, not without self-interest, that the revised edition of one of the collected statutes, Zhenghe chongxiu guochao huiyao 政和重修國 朝會要 (The collected statutes of the reigning dynasty revised during the Zhenghe reign [1111–18]), was made possible by the family of Xu Zhong, a former vice-minister of military affairs, who allowed the court to copy an earlier edition in its collection that Xu Zhong had helped to compile.42 Particularly noteworthy is that, in this and several other cases,43 the court “borrowed” the private copies for duplication and then returned them to the families who “owned” them. The history of the relationship between court and elites in the management of the dynasty’s archives sketched above suggests that emperors, archivists, and court officials recognized that authority had come to depend on the acceptance of a distributed archive. This distributed archive was partial and limited in scope, but by rewarding scholars and officials with titles and promotions even the most 41. For further discussion of the effect of Qin Gui’s and Qin Xi’s regime on the court’s and private collections, see De Weerdt, “The Discourse of Loss”; Hartman, “The Making of a Villain,” 93–94, 99–102. 42. ZZSL, 5.162. Cf. SHY, Chong ru, 4.23. 43. Winkelman, “The Imperial Library in Southern Sung China,” 29.



late imperial archival mentality

59

notoriously authoritarian among Song politicians ensured, at a basic level, the continuity of the Song archival record and thus Song imperial rule, and, more subtly, the cooperation of the literati in that enterprise.

Cl assif y ing a nd E xcer pting In her work on early modern European text management Ann Blair identifies storing, sorting, selecting, and summarizing as the four core operations in information management past and present. In addition to storing (collecting), sorting (classifying) and selecting (excerpting) were also core technologies that rendered the archives legible to readers in twelfth- and thirteenth-century Song China. Private collectors turned archival genres into bibliographic subdivisions and listed other types of works under the new subdivisions. In this section I will examine first the implications of this kind of appropriation of the central archives. Archival compilations were further rendered useful to literati readers through excerpting in encyclopedic texts that culled from these and a range of other texts. This intermingling of court archival compilations with other sources raises questions about both the status of official compilations among literati readers and the impact of archival operations on literati thinking and writing about history and government. The methods of classification in the catalogs of private collections reflect both the importance of court archival genres to the collections and the way in which these sources were integrated into literati scholarship. The private catalogs of You Mao, Chao Gongwu, and Chen Zhensun used draft dynastic histories, veritable records, and court diaries as bibliographic subdivisions. The categories used by private collectors typically accommodated a wider variety of works, of both official and nonofficial origin, than their definition as official court archival and historical genres implies. You Mao’s “dynastic history” section includes the official dynastic draft histories, biographical compilations, memoirs, collections of imperial decrees and other imperial writings, and court diaries. The juxtaposition of court and nonofficial sources in the dynastic history sections

60

contempor ary dimensions

( guoshi or zhengshi 正史) underscores the fact that dynastic history, though accepted as authoritative, was ranked alongside other works that supported, supplemented, or contested official accounts. The listing of collected statutes under the rubric “encyclopedias” (leishu 類書, leishi 類事) illustrates how court archival compilations had become appropriated by literati readers. The court promoted the compilation of classified collections of official documents before the first collected statutes of the dynasty appeared. Models from the Archives became the model for the collected statutes, the first of which was commissioned by Renzong in 1030.44 Presented to the emperor in 1044, this latter compilation included materials covering all of Song history between 960 and 1043. The collected statutes were at first intended for internal use only. The list of around twenty main categories gave the emperor and high court officials ready access to relevant historical documents within the welter of rapidly accumulating archives and library collections. By the twelfth century, Models from the Archives and collected statutes were being used as scholarly reference tools and for didactic purposes outside the court. They appear in the rapidly growing encyclopedia sections of private catalogs alongside other learning and reference aids. In both You Mao’s catalog and Zhao Xibian’s mid-­ thirteenth-century supplement to Chao Gongwu’s catalog, collected statutes are intermingled with literary and historical encyclopedias and primers, such as the well-known commercial encyclopedias Shilei fu 事類賦 (Classified prose poems), Shiwu jiyuan 事物紀元 (The record of the origins of things), and Gujin shiwen leiju 古今事文類聚 (Classified collection of affairs and texts from the past through the present). The first two of these titles became available in print in the twelfth century, the last in the thirteenth.45 The inclusion of court collections alongside titles commonly perceived as composition tools 44. Yamauchi Masahiro, “Sappu genki to Sō kaiyō.” 45. Ming-sun Poon lists a printed edition of The Record of the Origin of Things in “Books and Printing in Sung China,” 338. For a description of the genealogy of the current editions of this text, see Gao Cheng, Shiwu jiyuan (ca. 1078–85), 1–2. The Beijing Library has a printed edition (1146) of Classified Prose Poems. A Song printed edition of Classified Collection of Affairs and Texts from the Past through the Present is listed in Qian Zeng, Shugu tang Song ban shumu, 6b.



late imperial archival mentality

61

and learning aids suggests that the former were copied and purchased for their practical applications in literati discourse. Seen in the larger context of book production and collecting between the eleventh and the first half of the thirteenth centuries, the interest in court archival and historical compilations fit into a larger change in reading interests. Historical materials, especially those written by contemporary authors about the present dynasty, occupied a much larger part in the cultural production of the Song than in that of the early imperial period. According to Liu Zhao­ you, 90 percent of all the titles in the history section of the “Bibliographical Survey” were written by Song authors.46 The percentages of history titles in Song catalogs far outnumber those in the earlier bibliographical survey of Sui shu 隋書 (The Sui dynastic history).47 This trend holds true whatever the size of collections and the background of the collectors. The percentage of history titles in the four Song catalogs mentioned thus far range between 19.3 and 31 percent; only 12.5 percent of the titles included in the Sui bibliographical survey are history titles. A further step in the dissemination of archival compilations was their circulation in piecemeal format. Encyclopedias containing excerpts from collected statutes, veritable records, draft dynastic histories, and collections of “precious instructions” and of exempla of “sagely government” were printed in the twelfth and thirteenth centuries and marketed as examination manuals or writing aids. In all likelihood such encyclopedias reached a much wider audience than the voluminous manuscript or print editions of the original texts upon which they drew. Archival compilations were just one among many types of resource that were parsed for bits of information relevant to particular subjects. Note takers culled through literary collections by famous authors, the dynastic histories of previous dynasties, private historiographical works, administrative compendia, and classical and religious texts and then fit the pieces together under varying types and numbers of rubrics. 46. Liu Zhaoyou et al., Song shi “Yiwen zhi” shibu yiji kao, 24. 47. A table with holding numbers is included in De Weerdt, “Byways in the Imperial Chinese Information Order,” 169.

62

contempor ary dimensions

The range of archival compilations that the editors of commercial encyclopedias had access to, firsthand or secondhand, can be best illustrated on the basis of an entry from Gujin hebi shilei beiyao 古今 合璧事類備要 (The full essentials of the past and the present brought together and classified). Compiled by Xie Weixin 謝維新 (fl. mid-13th c.), this work was printed commercially around 1257.48 In the entry on “The Assistant Recipient of Edicts” (“Dufu chengzhi” 都副承㫖), the compiler draws upon the veritable records, the combined edition of the draft histories of the last four emperors of the Northern Song period (Sichao zhi 四朝志), and three consecutive compilations of the collected statutes: The Collected Statutes of the Four Reigns Continued, The Collected Statutes of the Intermediate Revival, The Collected Statutes of the Xiaozong Reign, and The Sagely Government of Xiaozong (Xiaozong shengzheng 孝宗聖政). These sources were used to allow the reader to trace the history of this position (and changes in the background of candidates) over time on the basis of short primary source quotations.49 Such excerpts could be used in a variety of circumstances. Bits of information now and then were picked up and shared to entertain and impress. The origins of particular customs and bureaucratic routines were a common topic of conversation among educated elites. Wang Mingqing’s notes on conversations with friends and texts read contain a good number of trivia of this kind. One of the earliest encyclopedias to quote from collected statutes may have served just this purpose. Titled The Record of the Origins of Things, it was compiled and circulated in print in the late eleventh century (ca. 1078–85). A reprint from 1197 is on record. One can find within it the earliest instance of all sorts of institutions and practices, and it is indeed still used as a reference work among historians nowadays. For example, it traces examination practices such as “posting seat assignments” and “printing examination questions” back to their earliest appearance in the collected statutes. The pre-Ming editions of Song encyclopedias usually refer to “the collected statutes” 48. Song prints are preserved at Beijing Library and Shanghai Library. 49. Xie Weixin, Gujin hebi shilei beiyao, houji, 17.1a–b.



late imperial archival mentality

63

(huiyao), “the collected statutes of the reigning dynasty” ( guochao huiyao 國朝會要), or to specific editions—such as Qiandao Xu sichao huiyao 乾道續四朝會要 (The Qiandao reign edition of the collected statutes of the four reigns, continued), Zhongxing huiyao 中興會要 (The collected statutes of the intermediate revival), or Xiaozong huiyao 孝宗會要 (The collected statutes of Xiaozong’s reign). Extant editions of The Record of the Origins of Things dating to the Ming and Qing dynasties have changed these references to “the collected statutes of the Song dynasty” (Songchao huiyao 宋朝會要), an anachronistic reference to the collected statutes that was not in use during the Song dynasty. The change was not made in the entry “end of the evening curfew” ( fang ye 放夜), showing what the original text may have looked like.50 Apart from the stuff for social exchanges, collections of archival materials also provided textbook material to students training in examination learning. Lin Jiong 林駉 (fl. 1210s), a teacher in Zhang­ zhou Prefecture 漳州 (Fujian), noted that his students found the history of earlier dynasties tedious and demanded classes on Song history.51 In response to these needs, Lin culled texts from the veritable records, collected statutes, draft histories, and precious instructions. The encyclopedia that resulted from his lecture notes further includes materials from digests of other genres of official documents, such as anthologies of memorials52 and Li Tao’s 李燾 50. Gao Cheng, Shiwu jiyuan, 8.430. These changes were apparently already made in the Ming editions on which the Zhonghua shuju text is based. Zhang Zhihe recently argued that, judging from the anachronistic way in which it refers to the Song collected statutes, The Record of the Origins of Things was written in Ming times. See his brief research note, “Shiwu jiyuan chengshu yu Mingdai kao.” However, Zhang fails to demonstrate that “Songchao huiyao” was a common name for Song collected statutes in Ming times and also overlooks several references to Record of the Origins of Things and its author in Song sources, such as Xie Weixin’s preface to The Full Essentials of the Past and the Present Brought Together and Classified. 51. Lin Jiong, Huang jian jian yao, preface. The original is held at Beijing University Library. The preface is also transcribed in Zhang Jinwu, Airi jinglu cangshu zhi, 26.9a. For the textual history of Ultimate Essays on Origins and Developments ( from the Past to the Present), see De Weerdt, Competition over Content, app. 52. “Memorials” (zouyi) probably refers to the widely published anthology Guochao zhuchen zouyi 國朝諸臣奏議, compiled by Zhao Ruyu 趙如愚 in 1186. This anthology

64

contempor ary dimensions

(1115–84) documentary history of Northern Song government, Xu Zizhi tongjian changbian 續資治通鋻長編 (Collected documents for a continuation of “The comprehensive mirror for aid in government”). This work, titled Xinjian jueke Gujin yuanliu zhilun 新箋决 科古今源流至論 (Ultimate essays on origins and developments from the past to the present, newly annotated for examination success), was completed by the late 1220s or early 1230s and published in three series. The earliest printing may date to 1237. The mixing of selections from different types of sources, such as we have seen in the works published by Lin Jiong and Xie Weixin, is typical of twelfth- and thirteenth-century examination manuals. Examples include Lidai zhidu xiangshuo 歷代制度詳說 (Detailed explanations of institutions throughout the ages, early 1180s), Bishui qunying daiwen huiyuan 璧水群英待問會元 (The epitome of eminent men responding at the Imperial College, ca. 1245), and Qunshu huiyuan jie jiang wang 群書會元截江網 (A net to unite and order the massive amounts of information in all books, ca. 1250). Their compilers quoted extensively from court archival and historical compilations, private histories of the Song period, anthologies of memorials, and policy questions and response essays.

M a nuscr ipt a nd Pr int Publishing Wu Cheng attributed the dissemination of materials that were previously stored in court repositories to the wider impact of print technology in Song times. Wang Mingqing observed that, as a result of local government investment in printing in the twelfth century, collectors had spread and collections grown: When peace prevailed the families of the political elite [shidafu] such as the Qi from Nandu, the Shen from Liyang, the Li from Lushan, the Chen from Jiujiang, and the Wu from Fanyang all were famous for their collections of books; these have now all been scattered. Many of the prefectures where I have been in recent years print texts; included 1,627 memorials submitted by 245 officials during the Northern Song. Cf. Balazs and Hervouet, A Sung Bibliography, 122–23.



late imperial archival mentality

65

it is also easy to obtain copies for further copying. The families of those who have but the slightest success in officialdom invariably possess several thousands of volumes [juan].53

Contemporary witnesses thus testified to the impact of printing in the broader distribution of both printed texts and manuscript copies thereof. Was this another expression of the anxieties and high expectations raised by the mechanical reproduction of texts? To what extent and in what ways was print technology used in the dissemination of Song archival materials? Manuscript remained the dominant medium for the production, leaking, and further dissemination of archival compilations in their entirety. Court archival and historical compilations were typically not printed in full. To my knowledge, only one comprehensive set of collected statutes was made available in print: Li Xinchuan’s combined edition of all Song collected statutes through the reign of Ningzong, The Collected Statutes of the Reigning Dynasty Comprehensively Excerpted and Classified, printed in Sichuan. Whether the printing was a government or commercial project is unclear. According to Chen Zhensun, the woodblocks were later transferred to the Directorate of Education in Hangzhou, which may suggest that they had been confiscated, but there is no record that either the printer or the author were punished for breaking the ban on the printing of collected statutes in The Classified Laws of the Qingyuan Period.54 Other sources claim that the court ordered Li Xinchuan to undertake the project and even had primary materials sent to him. If this is true, the transfer of the woodblocks would then have been normal procedure.55 Smaller collections of official documents, such as the “sagely government” and “precious instructions” collections, circulated 53. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.10. 54. ZZSL, 5.163. Zhang Xiumin (Zhongguo yinshuashi, 196) reads Chen’s statement as an indication that court orders prohibiting the private and commercial distribution of court archival compilations were not strictly observed. 55. Cai Chongbang (Songdai xiushi zhidu yanjiu, 168) interprets Li Xinchuan’s biography in The Dynastic History of the Song in this light. See also Chaffee, “Sung Biographies,” 209, 213. Miao Shumei and Wang Yunhai (Song huiyao jigao: chong ru, 455–56) give a more extensive account of the court’s arrangements with Li Xinchuan.

66

contempor ary dimensions

more widely in print than the collected statutes. Like their Tang model, Zhenguan zhengyao 貞觀政要 (Essentials of government from the Zhenguan reign, ca. 729), they were originally conceived as manuals for use in court lectures and the emperor’s instruction, but were also used by the literati as manuals on administration. The Song selections were imperially commissioned or submitted by private individuals throughout the dynasty. When, as was sometimes the case, they appeared before the final official edition of the imperial diaries of a given reign, they relied on the running drafts held at the respective offices responsible for compiling the diaries.56 They thus drew on the same sources that went into the veritable records and collected statutes, but, being both smaller and more current, they held greater appeal for commercial printers and students. In 1020 Emperor Zhenzong gave his imprimatur for a selection of documents and recorded statements on his government, Tianxi shengzheng ji 天禧聖政紀 (Record of the sagely government of the Tianxi reign [1017–21]). He ordered that it be printed and disseminated throughout the empire.57 Lin Fu submitted his compilation, Truths regarding the Government of Emperor Shenzong, to the court in the first decade of the 1100s and, according to the 1122 edict discussed earlier, it circulated commercially under various titles. Chen Zhensun provides several other examples of printed and commercial editions of these types of sources. He saw, for sale in Putian 莆田 (Fujian), an illustrated edition of sagely instructions from the first three Song emperors’ reigns, Sanchao xunjian tu 三朝 訓鑑圖 (Illustrations of the instructions and models from the three reigns), in a Northern Song edition printed by the Left Storehouse (Yufu 御府), the office charged with the handling of state revenue.58 56. For example, Tianxi shengzheng ji 天禧聖政紀 appeared in 1020. See Wang Deyi, “Songdai de shengzheng he baoxun,” 7. The daily calendar for this time was, however, not begun until Renzong’s ascension in 1022. See Cai Chongbang, Songdai xiushi zhidu yanjiu, 41. 57. Wang Deyi, “Songdai de shengzheng he baoxun,” 7. 58. ZZSL, 5.163. Wang Mingqing also mentions this printed edition under the title Sanchao baoxun 三朝寳訓. Huizhu lu (Zhonghua shuju ed.), 2.53. For a critical evaluation of Wang’s comments, see Liu Zhaoyou et al., Song shi yiwen zhi shibu yiji kao, 223–25.



late imperial archival mentality

67

This title is mentioned in several other Song sources. According to Guo Ruoxu 郭若虛 (ca. 1020–after 1075), it consisted of about one hundred illustrated events in ten chapters. Distribution was at first limited to high officials and members of the imperial family.59 Chen Zhensun mentions a similar compilation, Sanchao zhengyao 三朝政要 (Essentials of government from the three reigns), in an eleventh-century printed edition. This title was reissued under official auspices in 1138.60 Chen also lists two commercial copies of precious instructions of the reigns of Emperors Gaozong and Xiao­ zong.61 The first title, Gaozong Xiaozong shengzheng bianyao 高宗孝 宗聖政編要 (Essentials of the sagely government of Emperors Gaozong and Xiaozong), attests to the rapid turnover of these materials. According to Chen, these collections on sagely government were compiled during the Qiandao (1165–73) and Chunxi (1174–89) reign periods of Emperor Xiaozong. There must have been a slightly longer time lag, as the original court compilations on which the com­ mercial editions were based were completed in 1166 and 1192 (that is, just after the Chunxi reign had come to an end), respectively.62 Chen’s copy was a digest of these court compilations, complete with original imperial prefaces presumably written by Xiaozong and Guangzong. In Chen’s estimation, this commercial edition targeted examination candidates. Another commercial edition of The Sagely Government of Xiaozong that Chen lists, similarly aimed at examination candidates, was—in his view—the more comprehensive of the two editions. However, it is not clear whether these two editions were printed. 59. Guo Ruoxu, Tuhua jianwen zhi (ca. 1075), 6.7a–b. This reference is mentioned in Murray, “Didactic Illustrations in Printed Books,” 428. The inclusion of this title in Chen Zhensun’s catalog, along with the comment that he saw a copy for sale in Putian, refutes Murray’s conclusion that this title did not circulate outside of the court. 60. ZZSL, 5.163–64. 61. Ibid., 5.168–69. 62. Wang Deyi, “Songdai de shengzheng he baoxun,” 11–13. Chen Zhensun may have been copying from an entry in an imperial catalog. “The Song Bibliography Survey” uses the same wording in dating Essentials from the Sagely Government of Gaozong. Cf. Liu Zhaoyou et al., Song shi yiwen zhi shibu yiji kao, 298, 304–5; note that in the Zhong­ hua shuju edition the title has been changed to Essentials of the Sagely Government of Emperors Gaozong and Xiaozong. SS, 203.5103.

68

contempor ary dimensions

An extant combined edition of the collections on sagely government for the reigns of Gaozong and Xiaozong suggests that Chen Zhensun’s copies may have been commercially printed editions or copies of such editions. Zengru mingru jiangyi Zhongxing liangchao shengzheng 增入名儒講義中興兩朝聖政 (The sagely government of the two reigns of the intermediate restoration, with explanations by famous scholars) was printed in small format in Jianning Prefecture, which comprised Jianyang County, the site of at least fifty commercial printshops.63 The commentaries attributed to well-known scholars, as advertised in the title, reflect the pedagogical nature of the book. By including commentary from authoritative scholars the editors were reproducing the pedagogical setting for which compilations of this genre were intended. In 1163 Emperor Xiaozong had charged Lu You 陸游 (1125–1209) with the compilation of the precedents of Emperor Gaozong. Upon his departure from the capital, barely a few months after he had started the work, Lu You dictated from memory the draft he had started work on; this work circulated separately. Lu You’s draft and the comments he appended to each entry were selectively included in the explanatory sections of the full commercial edition. Just as the new emperor was to be instructed in the use and interpretation of exemplary decisions and model behavior, so readers in the provinces were exposed to the exemplary selection and interpretation of short passages from the archival record.64 The combination of comments from several scholars was part of a set of marketing strategies aimed at students. In addition to providing explanations (literally, “lectures on the meaning” jiangyi) of the selected documents, this edition inserted, for ease of reference, topical headings in the top margins and navigation markers listing the reign period and year in the corners. It also included stress and punctuation marks to facilitate reading. The value of this edition as a reference tool is evident in the cross-indexing supplied. At the front are a chronological and a classified table of contents. 63. See Chia, Printing for Profit, 76, 279–83. Hartman discusses this text in “The Making of a Villain,” 83. For a brief description of the printed edition preserved at the National Library in Taibei, see Wang Deyi, “Songdai de shengzheng he baoxun,” 26. 64. Kong Xue, “Lu You ji Gaozong shengzheng cao.”



late imperial archival mentality

69

The former shows the years covered in each chapter (starting with Jianyan 1 [1127] and ending with Chunxi 15 [1188]). The latter is divided into fifteen major subjects, each further subdivided into subtopics. Under each subtopic is a chronologically arranged list of entries, each summarizing an event in a phrase. The main text is organized chronologically, but the topic headings in the upper margins correspond with the subjects listed in the classified table of contents. Using these keys, readers interested in topics as varied as imperial book collecting, military expenses, the operations of the Censorate, the general Zhang Jun, or the chancellor Qin Gui could find several passages on each in the main text by locating the relevant subject in the classified table of contents, checking the year mentioned under the entry, and then matching the entry name to the topic heading in the chapter devoted to that year. The navigation markers enabled the reader to flip to the correct chapter without having to check the chronological table of contents.65 An example of how this might have worked is provided in my discussion of the question of reunification in chapter 4. As in early modern Europe, the application of print technology led to few true innovations in the presentation of the text. Reference books, notebooks, subject headings, annotations, and the use of images, markup, and punctuation were all attested well before the more widespread application of print technology from the eleventh century onward. However, as in sixteenth- and seventeenth-century Europe, “printing spread familiarity with the trappings of consultation reading to larger and more diverse audiences.”66 Beyond the 65. Some of these features of the Song print were deleted in more current editions, such as that in the Wanwei biecang collection (reprinted in Xu xiu siku quanshu). The Wanwei biecang edition has the same page layout, but omits the navigation markers (highlighting topical divisions, noting the date on each page, and separating the commentary from the main text) and the punctuation marks. The text of the Wanwei biecang edition also omits derogatory references to non-Chinese peoples. Although fewer chapters remain of the text of the Song edition in the National Library in Taibei, this contains chapter 45, which is not included in the more current editions. In the classified table of contents, reference is made several times to Chunxi 16 (1189) and the start of the reign of Emperor Guangzong. However, none of the editions I have seen covers the last year of the Chunxi reign period in the main text. None contains prefaces or postfaces. 66. Blair, Too Much to Know, 46.

70

contempor ary dimensions

archival compilations associated with the Historiography Institute at the court, other types of compilations of state documents were also disseminated in print. A good example is the collection of memorials by an individual author or by a group of senior statesmen, many of which appeared in print.67 In sum, more so than providing access to archival compilations in their entirety, printing served to familiarize readers with the use of such materials in digested format. All of the encyclopedias used to illustrate the practice of excerpting discussed above appeared in print in the twelfth and thirteenth centuries, as did most of the examination manuals. Notebooks, of which estimates range between 150 and 500 depending on the criteria used (see chapter 6), played a comparable role. Wang Mingqing’s series of notebooks, Waving the Duster (four installments), which appeared in at least six printed editions, paraphrased material from the same archival compilations as the encyclopedias, supplemented the published archive through full transcriptions of single-sheet originals, and demonstrated how the discussion of such materials fit into oral and written exchanges among the educated elite (see chapter 7).

The Archival Mentality Within the expanding repertoire of genres disseminating information about the Song government, archives and archival operations (the selecting, sorting, and compilation of documents) played an increasingly prominent role. Song literati came to assume that thinking, writing about, and practicing government should be based on referenced primary documents (even when accessed through second-hand sources) and that these documents should be comprehensively collected and digested to facilitate their use. Literati thinking about history and government was gradually characterized by an “archival mentality.” 67. For a fuller discussion of the printing and circulation of collections of memorials during the Song period, see De Weerdt, “Byways in the Imperial Chinese Information Order,” 180–84.



late imperial archival mentality

71

The development of this archival mentality was the unintended outcome of the bureaucratization of those who were not considered part of the bureaucracy. An indication of the formal exclusion of students and scholars from officialdom is the fact that in the extant commercial edition of The Sagely Government of Gaozong and Xiaozong “the examinations” (keju 科舉) is a subheading under “the learning of scholars” (ruxue 儒學) and is thus separate from “the ranking of officials” ( guanzhi 官職). As the gradual expansion of the use of civil service examinations in the eleventh century created a large and durable pool of aspiring officeholders, emperors and court officials set in motion a process whereby the ideological tools of government were shared with the elite in the provinces. The precedents of the founding emperors were, as Deng Xiaonan has shown, a necessary component in the formulation of policy proposals at court.68 The gifting of such collections to court officials and then the imperial endorsement for their circulation empire-wide was no doubt a powerful means to represent the exemplary government of Song rulers and statesmen. Song precedent was in literati writing typically referred to in order to prove the legitimacy of an argument. Selecting the most appropriate precedent and invoking it inevitably tied those involved in the exercise to the center of political power. Two aspects of the archival collections and their use in literati discourse rendered the relationship between the central government and scholar-officials ambivalent. The pastness of the collections created a distance between the court as the present hub of administration and past courts as a living memory of best practice. The memory of the founding emperors proved most authoritative. The legacy of Taizu and Taizong who had established a reunified empire was a par­ ticularly powerful body of precedent for those writing about history and politics throughout the Song period. The popularity of the sagely government collections of the two emperors who restored Song rule in the south, Gaozong and Xiaozong, in the thirteenth century similarly allowed for the critical application of authoritative models of the Song past to present decision making. 68. Deng Xiaonan, Zuzong zhi fa.

72

contempor ary dimensions

The fragmentary nature of precedent and the selection of brief primary source quotations similarly allowed for some flexibility in their use in policy proposals whether at court or in examination and study halls. Even though the sagely government compilations in their entirety provide some sense of change over time in an emperor’s or councilor’s decisions, precedents were in literati writing adduced in ad hoc fashion. In the absence of standard long-term narratives, users of these classified collections could adapt their selections to support the argument they wished to make. It is, for example, striking that among the twenty entries Lu You recollected from the first draft of The Sagely Government of Gaozong, the one relating to the civil service examinations concerned Gaozong’s decision to decline the prerogative to make the final ranking of the candidates. The emperor was quoted as saying: “I delegate the selection of candidates to the chief examiner, so there will be no mistakes.” Lu You praised the emperor’s decision in 1128 as abiding by the standard of fairness because it would have the effect of avoiding the interference of his personal entourage in recruitment for the field bureaucracy; he also used this opportunity to criticize the interference of the chief councilor in the examinations in subsequent years.69 This effect of the fragmentary nature of the evidence was enhanced by a process of historical layering described in greater detail in chapter 4. In brief, those discussing policy at court or in the provinces used primary source evidence from the classical past (antiquity), the pre-Song dynastic past (Qin through Five Dynasties history), and the Song past and present in a structured fashion. Cases from each of these layers were mustered to support the recommended proposal or interpretation through either positive or negative example. In addition to the example provided in discussions about war and peace in chapter 4, the following case from a thirteenth-century examination manual elucidates how the archival mentality worked. 69. Kong Xue, “Lu You ji Gaozong shengzheng cao,” 37. Kong’s article includes a full transcription of the original text, which survives only as an excerpt in Xie Jin, Yongle dadian 永樂大典, 12929.1a–6a.



late imperial archival mentality

73

Under the subject heading “exchanging envoys” the editors of The Epitome of Eminent Men Responding at the Imperial College included an essay arguing emphatically against the exchange of envoys at the time of writing: Great generals advance and do not retreat. In this way the authority of the Central Kingdom can be spread. Envoys are refused and not welcomed. In this way the schemes of the foreigners can be stopped. Why is this so? We rely on generals to hold the enemy off. Envoys are what they rely upon to spy on us. Should our generals tarry and not advance, their envoys will intimidate us and come back. Then the position of the realm [tianxia] may not even compare to the sorrow of today.70

The case against the danger posed by the policy of entering peace negotiations was based on two historical examples drawn from the period of division. Fu Jian 苻堅 (338–85), ruler of the Former Qin 前秦 (351–94), which was situated in the northwestern part of the Chinese territories, sent the envoy Guo Bian 郭辨 in preparation for the incorporation of the small state of Yan 燕 (Former Yan, 337–70), which occupied the northeastern part. And similarly, the northern state of Wei 魏 (386–534) sent Li Shun 李順 as an envoy as it started to move against Liang 凉 (Northern Liang, 397–439) in the west. The point of these cases was obvious: Yan and Liang were incorporated following their acceptance of these overtures. Moving on to more recent episodes in Song diplomatic history, the author of this model essay argued that the peace negotiations under the emperors Renzong, Zhenzong, and Gaozong had been conducted under radically different circumstances from the ones that applied at the time of writing in the early thirteenth century. Peace negotiations had in the past only been attempted when the Song court found itself in a position of relative strength vis-à-vis the other. Examples were adduced that demonstrated the effect of the projection of military strength on the enemy. Emperor Taizu had 70. Liu Dake, Bishui (Nanjing Library ed.), 75.15a.

74

contempor ary dimensions

been admired for his military valor by a Khitan envoy. Emperor Zhenzong responded to a violation of the peace settlement reached with the Liao by a show of military strength. And the peace agreements reached under Renzong were based on military fortifications along the northwest. The more recent case of Emperor Gaozong may have presented a conundrum as he was credited for having saved the Song dynasty but had done so through a settlement in which the occupation of the north was formalized. The author argued that this settlement had been acceptable because the Song court could then still rely on the prowess of the generals who had saved the day and whose fame only increased over time: Yue Fei 岳飛 (1103–41), Zhang Jun 張俊 (1086–1154), Han Shizhong 韓世忠 (1089–1151), and Liu Yi 劉綺 (1098–1162). The historical truth that may be inferred from this was that peace is acceptable under conditions of equality between both partners. Any imbalance in power, however, makes peace a selfdefeating strategy. As the author perceived there to be a major imbalance, with Song being the weaker party, peace agreements could not be entertained at the time of writing. Emperor Gaozong could thus be upheld as a model for current decision making not because he ultimately accepted a negotiated settlement but because he supposedly had understood the conditions under which peace was acceptable. The accompanying passage from the sagely government compilation covering his reign could be interpreted in this light: “In the eighth year of the Shaoxing period [1138] a foreign envoy came to Changzhou. The emperor said: ‘Even though the peace agreement has already been concluded, still we cannot relax our military preparations.’”71 As illustrated in this example, the dissemination of classified compilations of precedents had the dual effect of strengthening the identification of officials and elites with the dynasty and with the project of reunification associated with its founding emperors on one hand, and providing them with the tools and incentive to evaluate the performance of the current central government on the other hand. The text technologies used to disseminate dynas71. Ibid., 75.16b.



late imperial archival mentality

75

tic archival materials thus had a transformative effect on the political and scholarly discourse of elites beyond the court and the bureaucracy. The making, storing, and broader social uses of archival records were, also in Chinese history, processes separated in time. Even though the making and storing of records went back as far as preimperial times and their use in court histories similarly predated the formation of the Qin Empire, it was not until the eleventh and twelfth centuries that literati outside court circles gained access to and began to make use of court archival compilations. The social and political uses of archival records, which followed upon the centralization and standardization of archival operations in the tenth and eleventh centuries, were dependent upon the social and cultural transformation of the political elite from a metropolitan aristocracy to a provincialized literate elite whose status was tied to the civil service examinations. Literati networks infiltrated into the bureaus set up to organize and control the archival record, so that in the course of the eleventh and twelfth centuries scholar-officials in the provinces collected archival compilations and excerpted from and printed digests of them. Access to the dynastic archival record shaped thinking and writing about dynastic history and government and other topics. As we shall see in chapter 7 the ability to use the archival record in conversations and social writing also became part of what it meant to be a cultured person in Song literati circles. The court continued to be the main producer of archival compilations and retained the focus of elite political attention through their reproduction in the provinces. The dissemination of the archival record turned out to be a source of dynastic legitimation even though it could be used by provincial literati to voice opinion on contemporary government policy.

T wo Court Gazettes and Short Reports

T

he tension between secrecy and publication was, in the case of the court’s archival materials, typically resolved through the court’s tacit acknowledgment of the practice of leaking.1 In retrospect, the benefits of allowing the expanded reading public mediated access to these materials outweighed the cost of implementing bans on their circulation. The circulation of archival compilations enhanced literati identification with the dynasty and tied the large and ever-growing numbers of scholar-officials to the court, evoking it as the hub of imperial administration (see chapter 1). Because archival materials covered the dynastic past and because the compilations had usually gone through multiple stages of editing, the stakes were lower than in the case of state documents relating to current affairs. In this chapter I investigate the tension between secrecy and publicity in the circulation of court gazettes (usually chao­bao 朝報, dibao 邸報). The gazette was a court publication listing new appointments, dismissals, foreign missions, and excerpts 1. This chapter is revised from De Weerdt, “ ‘Court Gazettes’ and ‘Short Reports.’ ” In reviewing the material, I found some inaccuracies in translation (I earlier referred to county magistrates when the material clearly refers only to prefectural magistrates and circuit level administrators); these have been updated. More important, I tried to follow up on the suggestion by one of the anonymous reviewers of the article to refocus the argument. I regret I did not have sufficient time to make the desired changes then, and hope the new version is an improvement and engages the significance of leaking more directly. I hereby thank this reviewer for his/her comment.



court gazettes and short reports

77

of edicts and memorials, as well as a schedule of audiences. I ask who and what institutional processes determined editorial policy and content, how gazettes circulated, who the intended and actual readership was, and whether the dynamics of centralized control and the intervention of literati networks were also visible in the production and dissemination of the gazettes.

Bureaucratic Information Flows The content, compilation, editing, and circulation of court gazettes were regulated by imperial decree. In the absence of the gazettes themselves, such regulations reveal the court’s perception and management of the communication of its affairs with lower-level bureaucrats and the populations they governed. This section reviews what regulations were implemented in the course of the Song dynasty and what they reveal about the official view of political communication. First, what did the typical layout and content of the gazettes look like? The standard definition of the layout of the court gazette is based on Song dynasty models. According to a memorial submitted in 1173, the court gazette covered court political affairs, excerpts from edicts and memorials, appointments and dismissals in the bureaucracy, court audiences and departures from the court, and punishments and rewards.2 This representation of the court gazette as a court publication excerpting major policy decisions, listing personnel decisions, and publishing the schedule of imperial activities corresponds to later characterizations of the genre.3 Even though originals from the Song period or even full reproductions of Song gazettes have not come down to us, we can catch a glimpse of the coverage and presentation of the materials by reading through excerpts from them. In poetry and letters scholars and officials commented on news covered in the gazettes. From their interests we can glean that personnel issues such as appointments, demotions, retirements, and deaths were well covered. Hong Mai 2. Zhu Chuanyu, Songdai xinwen shi, 24; SHY, Zhiguan, 2.51. 3. Mittler, A Newspaper for China? 187–207.

78

contempor ary dimensions

洪邁 (1123–1202) described the format in which personnel moves were typically announced. The court’s list of new appointments of prefectural magistrates simply recorded who was sent where replacing whom. Lists of resignation requests similarly summarily noted the magistrate’s current jurisdiction, name, and reason for leave (e.g., illness).4 Readers also commented on policy issues focusing on border affairs and issues of interest to regional and local administrators such as disaster relief, suggesting that reports, memorials, and decrees on these matters were also included. A seldom used but revealing source that provides more direct access to material covered in gazettes is Li Tao’s 李燾 (1115–84) Xu Zizhi tongjian changbian 續資治通鋻長編 (Collected documents for a continuation of “The comprehensive mirror for aid in government”). In his elaborate but often overlooked footnotes, this historian of Northern Song government indicated where he drew upon gazettes for the main narrative. He also provided long excerpts from the gazettes as supplementary material when an alternative source had been used in the main body of the text. The gazettes were typically referred to by specific date (reign name/year/month/day). From a survey of the gazettes quoted for events under Zhezong’s 哲宗 (1086–1100) reign it is evident that presentations by high court officials, decrees, memorials, lists of appointments, and other types of government documents such as notices of leave (xiebiao 謝表) were transcribed in the gazettes. Issues relating to personnel management (including, in addition to the above, banishments, recommendations, and tensions between individual officials as well as broader factional disputes) appear to have been highest on the agenda judging from this source. Crises such as large-scale flooding in the Yellow River plain also merited sustained coverage. Li Tao’s notes further show that in many cases the name of the speaker/ author was not revealed. He made some educated guesses or otherwise noted that it was unclear to him who might have presented the proposal under consideration.5 4. Hong Mai, Rongzhai suibi, wubi, 4.878–79. 5. XZZTJ (Hanji quanwen ziliaoku digital edition), 404.9838–39, 406.9878–84, 451.10831–32, 493.11719, 495.11783–84, 503.11988–89, 504.12001, 505.12038–39, 12042,



court gazettes and short reports

79

This strategy may very well have been intentional and was picked up on by readers in the provinces. Zhu Xi 朱熹 (1130–1200), for example, noting that in one case he had received a report that prompted him to be careful about releasing recent work, added that he could not tell from whom the attack had come.6 This report was mentioned around 1195–96 in a letter to Cai Yuanding 蔡元定 (1135– 98), a disciple of Zhu Xi’s, who, like his teacher at this time, did not hold office.7 Both men then resided at home and were under severe government scrutiny due to attacks on the Learning of the Way movement led by Zhu Xi. The nondisclosure of speakers and authors may have been intended to constrain the mobilizing effect that news items carried in a highly networked political environment. The report affected the timing of Zhu Xi’s publication plans for Explanations and Diagrams of the Great Ultimate (Taiji tushuo 太極圖說) and similar works, and it raised obstacles for a more direct counteroffensive on the part of Zhu Xi and his supporters against its point of origin in the present or the future. Not only the layout but also the compilation and distribution process were clearly delineated in Song regulations. Under Emperor Taizong 太宗 (r. 976–97) the Capital Liaison Offices (Jinzouyuan 進奏院), which regional commanders set up in the Tang capital in the eighth and ninth centuries, were brought under the control of the central bureaucracy. Taizong’s institutional reorganization of the exchange of information between the court and local administrations reflected the center’s reassertion over information control. Whereas Tang Capital Liaison Offices had served as intelligence offices for regional commanders, the Song Memorials Office’s mission was to serve the court’s interest in the gathering, sorting, and selective dissemination of information.8 507.12087–89, 510.12148–49, 514.12209, 12232, 515.12246, 12249–50, 516.12276–77, 517.12299–300, 12302, 12314, 519.12346. 6. QSW, 349.330. 7. The letter is dated and annotated in Chen Lai, Zhuzi shuxin biannian kaozheng, 414. 8. For the history of the Capital Liaison Offices during the late Tang and Five Dynasties, see Fukui Nobuaki, “Godai Shikkokuki no shinsōin” and “Tōdai no shinsōin.” He argues that Capital Liaison Offices continued throughout the tenth century. Their functionality was less pronounced during the Later Liang dynasty, but they went

80

contempor ary dimensions

The reports that the personnel of the Memorials Office sent down to regional and local government offices were typically subject to prepublication review. Already in 999, Emperor Zhenzong 真宗 (r. 997–1022) set a precedent for the review of the gazettes by the Bureau of Military Affairs, the paramount central government agency in Song times.9 During times of relaxation, the editorial process still required that the supervising secretaries ( jishizhong 給事中) attached to the Chancellery review the gazettes prior to their transmission. This set of procedures was also understood to be the norm among scholar-officials. Zhao Sheng 趙升 outlined the review process in his lexicon of administrative terms (Chaoye lei yao 朝野 類要, or Classified essentials in and out of court, ca. 1236).10 Zhu Xi similarly suggested in a conversation that supervising secretaries should be able to check personnel decisions before word got out.11 The standardization of content and the centralization of the editorial process went hand in hand with the regularization of the gazettes’ dissemination. Memorials Office personnel were assigned to all prefectures, ensuring the transmission of court gazettes to all parts of the empire.12 The circulation of the gazette was legally limited to officials serving at court and at all levels of the local government. The Memorials Office published gazettes regularly, with the interval changing over time from every five days to every day. The distribution may have taken longer. According to one report (1048), the Ministry of Military Affairs, which was in charge of transportation and postal services, distributed the final copies once every ten through a revival and restructuring during the Latter Tang. He further shows that in the last part of the Five Dynasties period branch offices were especially important as bases for the three dynasties in the capitals of the southern kingdoms. See also Wang Jing, “Chaoting he fangzhen de lianluo shuniu.” 9. Zhu Chuanyu, Songdai xinwen shi, 18–24; Umehara Kaoru, “Shinsōin o megutte,” 80–81. 10. Zhu Chuanyu, Songdai xinwen shi, 18–24; Umehara Kaoru, “Shinsōin o megutte,” 80–81. 11. ZZYL, 128.3071. 12. The circulation of the court gazettes in border prefectures received higher scrutiny than in other prefectures. In 1167 Emperor Xiaozong issued an edict prohibiting the circulation of the court gazettes in Xuyi Prefecture and other border prefectures. SHY, Zhiguan, 2.51.



court gazettes and short reports

81

days. According to another report (1098), Memorials Office personnel sent out the court gazettes every three days.13 Whether or not the court gazettes were printed during the Song dynasty cannot be conclusively demonstrated. Historians focusing on the history of the court gazettes during the Ming and Qing dynasties generally hold that Song gazettes circulated in manuscript only and were not distributed in print until late imperial times. Historians touching specifically on the history of Song gazettes have mustered several pieces of evidence suggesting that gazettes may have appeared in print as early as the eleventh century. A decree of 1071 stipulated that the budget of the Memorials Office, whence the gazettes were dispatched, should include annually one thousand strings of cash for “the carving of woodblocks, paper, and ink.”14 Even though these moneys were not specifically earmarked for the printing of the gazettes, it is likely that the gazettes would have been included in the communications to be printed and sent down to the routes and lower administrative levels. The printing of the gazettes may have started earlier in the eleventh century. A later opinion, submitted in 1176, charged that the printing of communications had fallen under the Board of Punishments (Xingbu 刑部) before 1071 and asked that the printing operations in the Memorials Office be stopped and returned to the relevant office under the board.15 Despite this effort, printing operations and subsidies for printing activities in the Memorials Office continued in the twelfth and thirteenth centuries.16 The fact that other government docu13. Zhu Chuanyu, Songdai xinwen shi, 19, 42–45; Umehara Kaoru, “Shinsōin o megutte,” 98. 14. SHY, Zhiguan, 2.46. 15. Ibid., Xingfa, 1.10. The board was put in charge of the printing of decrees in 1024. XZZTJ (Hanji quanwen ziliaoku ed.), 102.2368; see also Zhu Chuanyu, Songdai xinwen shi, 39. 16. For further discussion of the records cited above and for additional evidence, see Umehara Kaoru, “Shinsōin o megutte,” 83; Zhu Chuanyu, Songdai xinwen shi, 38–42. Kubota writes emphatically that Song gazettes must have been circulating in print—a view that is in part based on the circulation of other official communications in print but that also reflects the intent to regard them as a structural equivalent to the modern printed newspaper in the construction of national identities. Kubota Kazuo, “Sōdai no chūō jōhō no chihō dentatsu ni tsuite”; I thank Deng Xiaonan for sharing with me

82

contempor ary dimensions

ments intended for dissemination to local authorities, for example, in the legal domain, were printed further supports the case for potential dissemination in print.17 The processes of standardization, centralization, and regulari­ zation describe the official network and the official view of political communication. The official network was based on a set of regulations structuring the flow of information. Official regulations defined the institutions and procedures for the gathering, compilation, review, and dissemination of court news. Such regulations were part of an official discourse of communication that prescribed a direct and transparent flow of information from the center to the public. This public was imagined as an “audience,” literally “the listening of the masses” (minting 民聼, zhongting 眾聼, qunting 群聼). In official discourse, the court gazette was seen as a channel for the transmission of news first to officialdom and through it to the population at large. The news covered in the gazette was distributed for political guidance; it was to be listened to (ting 聼) and accepted as it represented a digest of decisions and events authorized by the central government for the instruction of officials and, through their administration, the population. The orders excerpted in the gazette kept local administrators informed about new measures and demanded their collaboration; they also stipulated which ones were to be posted for broader dissemination.

Leaking The thirty prohibitions I have identified on the distribution of the court gazette and the separate circulation of news items included in the gazette (such as individual decrees and memorials) suggest that Song emperors and central government officials from the reign of Renzong 仁宗 (r. 1022–63) onward were aware that existing procethe papers presented at the workshop at which he presented this paper. Also, Kubota Kazuo, “Sōdai ni okeru seichoku no dentatsu ni tsuite.” 17. See also Kubota Kazuo, “Sōchō ni yoru shomin e no jōhō dentatsu to insatsu bunka.”



court gazettes and short reports

83

dures and codes regulating the flow of information were continuously being violated.18 Who was involved in these violations and what did they consist of? Commercial publishers and examination candidates were targeted in official reports on the spread of news genres such as recent histories, archival compilations, and administrative encyclopedias.19 They are, however, remarkably absent in official reports and regulations on the production and distribution of court gazettes. This absence does not correspond to a lack of interest on the part of booksellers in the gazettes; there were booksellers and peddlers in the capital who sold gazettes.20 Rather, it reflects the dominance of government personnel in the business in gazettes. Official memoranda emphasized the commercial interests of government clerks in particular. A prohibition from 1175, for example, exposed the profit motives of ministerial and Memorials Office personnel involved in the smuggling of court news and its distribution through the official postal network.21 In 1224 similar allegations were made. According to a report clerks in various offices were contracted to gather information in the interest of particular prefectures and charged fees in the amount of 2,000–3,000 min 緡. Such an amount would have been sufficient to purchase a house (or a substantial part of it) in most prefectures.22 The price for a printed book at the time was about 150–300 wen 文 (1,000 wen equaled one min).23 18. De Weerdt, “ ‘Court Gazettes’ and ‘Short Reports,’ ” online edition, 199, table 1. 19. De Weerdt, “What Did Su Che See?” and “Byways in the Imperial Chinese Information Order.” On examination essays and manuals, see Chia, Printing for Profit, 121–23; Liu Hsiang-kwang, “Yinshua yu kaoshi” and “Printing and Examinations.” For a discussion of twelfth- and thirteenth-century regulations in the context of the competition over the definition of examination standards, see De Weerdt, Competition over Content, esp. chap. 5. 20. Extant records suggest that court gazettes were on sale in the capital in 1127. Huang Zhuoming, Zhongguo gudai baozhi tanyuan, 63–65. Zhou Mi noted that papers were also on sale in the southern capital of Hangzhou. Zhu Chuanyu, Songdai xinwen shi, 83. Zhou Mi, Wulin jiushi, 6.15a. 21. SHY, Xingfa, 2.118. 22. Ibid., 2.145; Umehara Kaoru, “Shinsōin o megutte,” 77. Umehara gives Jiading 17.2.9 as the date; this should read Jiading 17.4.9. 23. Cheng Minsheng, Songdai wujia yanjiu, chaps. 2, 8, and 13, esp. 374.

84

contempor ary dimensions

The very existence of the official postal network spanning the empire helps explain the apparent absence of commercial publishers in the distribution of gazettes (especially their derivatives, the short reports) outside the capital. The details of the distribution process and payment methods cannot be verified in the existing sources, however, the centrality of Memorials Office personnel in official accusations suggests that they played a key role in the business. Memorials Office personnel had a final hand in packaging the materials to be sent from the court to government offices in the provinces, such as the official court gazettes and imperial orders. Likewise, they opened the materials sent back from the provinces. Whereas personnel of the Military Board and soldiers managed the actual transportation of the mail, Memorials Office personnel were most likely to include nonofficial papers in the mail and would also have had the opportunity to manage payment through the postal network. The operation of the postal network, and the fact that it linked the producers of the short reports to an existing readership (local administrators who were already readers of the official papers), thus obviated the development of a parallel commercial distribution network. The centrality attributed to Memorials Office personnel in the circulation and modification of reports in official discourse expresses a double anxiety on the part of some central government representatives. There were specific concerns about the early release and reliability of the content of gazettes, but there were also broader concerns about the trustworthiness of the official distribution channels themselves. In response to this latter concern measures targeting the personnel of the Memorials Office were proposed and implemented. A reward system for the exposure of those responsible for lapses, a mutual responsibility system whereby groups of clerks were made liable for any potential misbehavior by all the others, and hefty punishments for violators of proper procedures were among the available options.24 Leaking was not only the business of Memorials Office personnel. When Gaozong’s 高宗 (r. 1131–62) court was in the pro24. See the decrees listed in De Weerdt, “ ‘Court Gazettes’ and ‘Short Reports,’ ” online edition, 199, table 1. Examples include the decrees of 1098, 1110, 1135, 1189, and 1193.



court gazettes and short reports

85

cess of rebuilding, it was alleged that meeting quarters lacked security to the extent that all sorts of people could gain access and smuggle information out of government offices.25 Private spies operated in the central government offices, gathering and copying information from various offices in the capital. According to one official investigation conducted around 1193, one servant made a living by regularly infiltrating into the inner grounds of the palace, gathering leaked reports from government offices, picking up hearsay, and supplementing all this evidence with his “personal opinion.”26 The news value and hence the marketability of gazette reports were much higher compared to other genres relating to the affairs of the court. Gazette reports were sold in public spaces, “on the street and in markets” ( jieshi 街市),27 and news could be transmitted there faster than through the official distribution channels. The capital cities of Kaifeng and Lin’an witnessed a proliferation of sites of exchange; the prospect of ongoing court business entering such spaces, particularly prior to its official release, occasionally led to the escalation of fears about the spread of rumor. The power of rumor is at the heart of the first documented reference to “short reports,” a term that soon gained wider currency to refer to unauthorized newssheets carrying the same kinds of information as the court gazette. On April 16, 1156, Gaozong issued an edict in which he aimed to silence the rumor that the policy of appeasement with the Jin Empire was about to be revised. Emperor Gaozong declared: I believe that putting the army at rest and the people at peace expresses the great virtue of a sovereign. Being true to one’s words and maintaining close relationships have proven to result in great benefits in the past and the present. Therefore, I, based on my own convictions, decided on a policy of appeasement [with Jin]. The former 25. SHY, Xingfa, 2.148. 26. Ibid., 2.125–26. 27. Wang Zao, attrib., Jingkang yaolu, 9.180, and Jingkang yaolu jian zhu, 9.970. See also SHY, Xingfa, 2.125.

86

contempor ary dimensions

councilor Qin Gui, with his many great talents, merely approved. How could it be that the debate will depend on his life or death! Recently ignorant folks, because they think that everything came from Qin Gui and do not know that it all came from me, have been spreading superficial stories, confusing the public [zhongting]. They even go so far as to fabricate imperial decrees calling for the return of former officials and for memorials to be sent to the throne. They randomly discuss border affairs. I am really shocked by this.28

The accusation of the falsification of decrees was in this case related to factional disputes. The edict was issued about five months after the death of Qin Gui 秦檜 (1090–1155), the much hated architect of the negotiated peace settlement between Song and Jin. News about the death of Qin Gui opened up prospects for those opposed to his policies who had been sent away from court and kept in the provinces during his two-decade-long regime. The opposition used newssheets to overhaul the upper echelons of the bureaucracy and to mobilize their members in the provinces, asking them to send in appeals for a revision of court policy on border affairs. Emperor Gaozong’s speech elicited further investigation into the channels through which political rumor was transmitted. One of the court diarists and editorial directors, Zhou Linzhi 周麟之 (1118–64), reported: Now when the emperor issues decrees and promulgates orders, the thunder is loud and the winds fly, there are some crooked individuals who make up deceitful stories to mislead the public [qunting]. Like what was mentioned the other day about “employing former officials,”29 these superficial words caused a stir, but no one knows whence they came. I have investigated the cause for this. These are all based on privately obtained short reports. These so-called short reports originate in the Memorials Office. Its personnel composes them. In recent years whenever doubtful events come up that people at court or in the provinces do not know about, Memorials Office personnel always vie to write them up on small sheets. They immediately report 28. Bi Yuan, Xu zizhi tongjian (Zhongguo guoji ed.), 131.3470. 29. This is a direct reference to Emperor Gaozong’s proclamation translated above.



court gazettes and short reports

87

them nearby and faraway. These things are called short reports. For example, they cover things like today someone was called in for an audience, someone was dismissed, or someone was moved into a new position. They oftentimes make something out of nothing.30

This report, as well as other contemporary testimony, described the appearance of “short reports” as a recent phenomenon. The chronology of prohibitions suggests that even though there had been similar publications before (prohibitions against the circulation of separate newssheets and the forging and unauthorized printing of orders go back to 1031 and 1070 respectively), the sale of newssheets did not become an established business until the second half of the twelfth century. Short reports sold well in the capital, because they reported court news faster than the court gazettes and also included news items that had not yet been subjected to the editorial review of the secretaries in the Chancellery or the Bureau of Military Affairs. Zhou Linzhi added that at the time of writing it was commonplace for court officials as well as prefectural magistrates reading official reports to comment, “Oh! The short report on this has already arrived!” In Zhou Linzhi’s report the producers and consumers of the short reports are located within officialdom. The violation of the proper procedures within the bureaucracy has in his view consequences that have a larger impact. Unauthorized news about personnel changes and policies challenged the official view of direct and unilateral communication between the center and the provinces. Only if the court set up a system of punishments and rewards to eradicate the publication of short reports would the official order of communication prevail: “When the commands issued by the court are broadcast to the realm [tianxia], the realm will be able to learn about them, but will not be able to spy them; the realm will be able to trust them and not to fake them. Then the state will be respected and the public [minting] will be unified.”31 30. Zhou Linzhi, Hailing ji, 3.2b–3b. Italics in the translation are mine. 31. Ibid., 3.3b.

88

contempor ary dimensions

Gazettes and Literati Networks It would have been obvious to those receiving and requesting gazettes of both the official and semiofficial kind that the practice of leaking could hardly be blamed on clerks and servants working in central government offices. Scholars and officials bought, read, and shared court gazettes as well as short reports. Gazettes issued officially and their derivatives were part of a variety of materials on current affairs for which there was a growing demand in the course of the Song period. How and in what sorts of relationships were reports shared? What issues did literati respond to? Did the sharing of court gazettes challenge political unity in the way Zhou Linzhi had suggested? The presence of archival compilations in private collections was the tail end of a process of leaking and transmission that we can trace only in part. Due to their more ephemeral nature gazettes did not appear in catalogs, the odd exception notwithstanding. Some scholar-official households kept long runs and used them in political debates, as is evident from the example of Song Minqiu 宋敏求 (1019–79). In 1069 Song Minqiu was able to settle a court debate and alleviate the concerns of Emperor Shenzong 神宗 (r. 1068–85) about the proposed date for a major ritual by citing a precedent that he had been able to locate in a court gazette dated to 1036, which he had found in his family’s collection. He was able to show, based on this evidence, that the fifteenth day of the month had been used in the past and that the end of the month should be considered a less appropriate time for the sacrifice to Heaven. Wu Zeng 吳曾, collecting noteworthy anecdotes documenting the origins of various court practices and institutions in the 1150s, deduced from this story that the Song family collection contained court gazettes from the early days of the dynasty through to the day when Song Minqiu submitted the memorial that settled the debate.32 This case demonstrates at once the persuasiveness of archival evidence in eleventh-century court debates and the celebration of the effective application of archival skills among scholars and officials one century later. 32. Wu Zeng, Nenggai zhai man lu, 2.18. Quoted in Sogabe Shizuo, Shina seiji shūzoku ronkō, 353–54 (Sogabe provides a slightly different reference).



court gazettes and short reports

89

Song Minqiu’s ability to settle a debate in this way may have been unusual in the last decades of the eleventh century, but by the twelfth century there was a sense that access to the court gazette was necessary to remain current and connected. Local officials at the prefectural and the county level can be seen to share gazettes with acquaintances and were credited for doing so in letters and poems. Fang Yue 方岳 (1199–1262) wrote a poem titled “The County Sends the Court Gazette,” suggesting that local magistrates shared the gazettes with prominent retired officials.33 When former officials ousted from the court or candidates waiting for assignments at home were unable to receive gazettes, they relied also upon contacts elsewhere to forward copies. Hu Anguo 胡安國 (1074–1138), for example, thanked “Minister Lü” for sending him the gazettes of the previous five months and asked him to keep sending them.34 Zhu Xi complained on occasion about the slow speed with which the gazette reached him. His expectation that it serve as a source for new information that would prompt a further response on the reader’s part is further palpable in an occasional complaint that recent papers did not contain anything extraordinary.35 For others like Sun Yingshi 孫應時 ( j. 1175) gazette news served as an invitation to write to a better-informed acquaintance for more information about the goings-on in the capital and the movements of mutual acquaintances.36 References to court gazettes and short reports became more frequent in private correspondence (i.e., letters not related to bureaucratic business even though such letters could be shared with scholars and officials) from the twelfth century onward. During the same centuries a subgenre of poetry devoted to reflections on “reading the gazette” developed. Such “gazette poems” and private letters are not only the main sources for the reconstruction of the contents of the gazettes and their reception; they also formed secondary 33. Fang Yue, Qiuya ji, 11.8b. 34. Zhu Xi, Zhu Xi ji, vol. 7, 81.4165. Van Ess, Von Ch’eng I Zu Chu Hsi, 139–40. As van Ess points out (ibid.), “Minister Lü” probably refers here to Lü Haowen 呂好問 (ca. 1064–1131). 35. QSW, 250.108. 36. Sun Yingshi, Zhuohu ji, 7.14b.

90

contempor ary dimensions

news genres in which literati exchanged feelings and opinions on the news reported in the gazettes. In the correspondence of Zhu Xi, who spent some time in office but more on (mostly voluntary) leave, we can observe the mobilizing power attached to reports of both an official and semiofficial nature. The gazettes emerge in these letters as an essential source for observing the political climate at the top. The reported personnel decisions and the decrees and memorials chosen for broader dissemination led to further speculation, networking, and strategizing. In his correspondence with Liu Guangzu 劉光祖 (1142–1222), Zhu Xi revealed that the gazette had allowed him to detect in Liu Guangzu an ally, a politician who had revealed in memorials published in the gazettes his willingness to clearly differentiate right from wrong.37 The alliance with Zhu Xi’s movement cost Liu Guangzu; his support for advocates of the Learning of the Way and his impeachment of their opponents led to his dismissal from court in the late 1190s.38 When Liu Guangzu was sent away from court, Zhu Xi kept up correspondence with him, expressing his sympathy in response to reports about successive dismissals.39 Through the dissemination of the gazettes factional disputes spread across elite networks that spanned the capital and the provinces. Dry notices of new hires and recent retirements were read as a barometer indicating times of opportunity and danger for allies in the provinces. Zhu Xi reported the good news of the appointment of a friend, Zhan Tiren 詹體仁 (1143–1206), to his student and intellectual collaborator Cai Yuanding in 1186, but sounded a note of caution when writing to another student, Huang Gan 黃榦 (1152– 1221), then a prefectural schoolteacher in Fuzhou 福州, in 1193.40 In the latter exchange he noted cryptically that much depended on more senior connections. He confided to another acquaintance in 1199 that reading gazettes since he had fallen out of favor had turned 37. QSW, 250.92–93. 38. Yu Yingshi discusses Zhu Xi’s efforts to remain up to date while out of office and his correspondence with Liu Guangzu; see Yu Yingshi, Zhu Xi de lishi shijie, vol. 2, 248–60, 288–300. 39. QSW, 250.92–93, 246.28. 40. Ibid., 249.286, 337.



court gazettes and short reports

91

him anxious and dispirited. The rhetoric of the decrees and reports selected for broader dissemination in those years had left him wondering at what his critics at court would stop. The signal he picked up in the gazettes was then so strong that Zhu Xi cautioned his correspondent against requesting any writing from him so as to avoid future trouble.41 The networking that was set off by court gazette news resulted in political action. The mobilization of networks through writing about court gazette news can in several cases be directly associated with identifiable political goals. Peng Guinian 彭龜年 (1142–1206) sent a letter to Councilor Liu Zheng 留正 (1129–1206) after he read a short report announcing the return of Jiang Teli 姜特立 (1125–?) to the court of Emperor Guangzong 光宗 (r. 1190–94) around 1193.42 Councilor Liu had engineered Jiang Teli’s dismissal from the post of administrator of the Palace Postern in 1189. Jiang Teli was a trusted advisor to the emperor; their relationship was based on Jiang’s service during the years the emperor had spent in the palace of the crown prince. In his letter Peng Guinian wrote that the announcement had been proven false the day after he read it, but he argued that its appearance was a sure sign that Jiang’s supporters were preparing the way for his return to court. He added that he had surveyed reactions to the announcement: few expressed concern and many appeared excited when it came out. He proposed that the councilor open an investigation into the matter, prosecute those responsible for the circulation of the short report, and use the opportunity to bring in like-minded men and oust Jiang’s supporters. He concluded that if such actions were not taken, the short report’s announcement was bound to materialize in the near future. The maintenance of political networks was also pursued through the exchange of poetry commenting on recent reports. Poems were sent to celebrate the bureaucratic achievements of friends and acquaintances, and thus renew connections with men on the rise. “Gazette” poems also expressed sympathy and solidarity 41. Ibid., 249.17. 42. Peng Guinian, Zhitang ji, 12.5a–8a. Cf. Yu Yingshi, Zhu Xi de lishi shijie, vol. 2, 518–19; Schirokauer, “Liu Cheng,” in Franke, Sung Biographies, 626.

92

contempor ary dimensions

with friends who were ousted, demoted, or simply passed over. Dai Fugu 戴復古 (1167–?) opened a poem he sent to Gong Feng 鞏豐 (1148–1217) with the lines, “I have been reading the court gazette frequently; no assignments have come to you.”43 Dai proceeded to console his friend that his literary work would preserve his name for posterity. He compared him with the celebrated authors Yang Xiong (53 BCE–18 CE) 楊雄 and Meng Jiao 孟郊 (751–814), who had established a lasting reputation for themselves despite checkered bureaucratic careers. Dai also extended his sympathy to Zhao Fan 趙蕃 (1143–1229): “In the court gazettes during this time, I frequently noticed that you were given temple guardianships. Temple guardians don’t earn a lot, and, once you’re poor, what can you do about it! Pick some touchwood [Fomes fomentarius; longevity fungus], you can eat it too; you should write a song on picking touch­ wood.”44 Dai’s recommendation of the retired life (that is what picking touchwood and writing a song about it stand for) was in this case based on the knowledge that Zhao Fan had written a poem of support for Li Zhi 李埴 (1161–1238), vice-minister of ritual, who had recently been ousted from court. Dai’s poem was then not only an expression of sympathy for his friend’s low bureaucratic rank, but also a warning for him to lie low while the campaign against Li Zhi was under way. Dai’s message is a clear indicator of the factionalist politics that shaped both the production and the reception of the court gazettes. The language within which literati interpreted and commented on court news reproduced the factionalist rhetoric that characterized court politics since the eleventh century, especially during times of intense factional dispute.45 By extending the analysis of factional politics into the Southern Song period, Yu Yingshi and Shen Songqin have lent support to the observation of the seventeenthcentury historian and philosopher Wang Fuzhi 王夫之 (1619–92), who wrote that factional politics became the literati mode of being 43. Dai Fugu, Shiping shi ji (SKQS), 2.5a; (SBCK) 2.5b. 44. Ibid. (SKQS), 1.22a; (SBCK) 1.21a. 45. On the language of factionalism in the eleventh century, see Levine, Divided by a Common Language; Hartman, review of Divided by a Common Language.



court gazettes and short reports

93

in the early Southern Song period.46 Even though factionalism was a characteristic of court politics under previous dynasties, and was perceived as such by Song literati and later historians, based on mostly a few notorious instances of factional struggle involving bureaucrats and court attendants,47 Wang Fuzhi’s observation captures the transition to an imperial politics in which factionalist struggle became a regular feature of the bureaucracy and structured the political lives of literati across the empire. “Gazette” poems and private letters further suggest that the circulation of the court gazettes played an important role in the expansion and consolidation of factionalist political networking in imperial China. Reading the court gazettes and sharing them in full or in excerpts in poems and letters were ways of keeping up with court politics, expressing opposition to political enemies, and seeking political allies, especially for those who were marginalized in the provinces. In “Remembering My Past” Fang Hui 方囘 (1227–1307) captures the thinking of poor scholars who look up to the powerful: “Together they read the court gazettes, frustration and anger overtakes them.”48 According to Fang Hui, their anger was targeted at those who were in positions to formulate and review state policy: “Sometimes the censors find wrong what is right, discussions about state affairs are flawed.” Liu Kezhuang 劉克莊 (1187–1269) expressed his views of court policy in more confrontational language. In one of several “gazette” poems, he wrote: I wish we would opt for the pure criticism of Han times, And fully get rid of the corrupt tendencies of Tang times. The goatsucker produces terrible and evil sounds, Wild tigers will only be ashamed at death.

欲取漢清議, 盡投唐濁流。 鬼車鳴甚惡, 猛虎死方羞。

46. Wang Fuzhi, Song lun, 10.201; Shen Songqin, Nan Song wenren yu dangzheng, 7. 47. Levine, Divided by a Common Language, chap. 2. 48. Fang Hui, Tongjiang xu ji, 5.26b.

94

contempor ary dimensions

Fragrance and foul smell have to be clearly distinguished. How I dread the impact of bad policy planning!

芳臭須㬰判, 哀哉不善謀。49

The use of wild birds and beasts as metaphors for political enemies was a feature of the factionalist discourse of Song times, as was the use of historical analogy to identify the inheritors of political virtue and the heirs of vice. The division of the scholar-official class into those who exude good smell and those who stink underscores the parallels he drew between internal and external enemies. Non-Chinese political elites ruling northern Chinese territories had traditionally been portrayed in similar ways, but in his series of poems on the court gazette, Liu Kezhuang directs attention to the enemy within, and the connection between the rule of the enemy within and the power of the Mongols to the north. Liu Kezhuang wrote this poem in 1238, when the Mongols had begun to successfully occupy Song territory and court officials were caught in debates on whether to pursue a policy of war or appeasement. Along with Wang Mai, to whom this poem was addressed, Liu Kezhuang was sent away from court in 1237 for his posthumous criticism of Councilor Shi Miyuan 史彌遠 (1164–1233). He was assigned to a temple guardianship, which he still occupied in 1238.50 In his view, those who had been affiliated with Shi Miyuan were behind the demotions. Liu Kezhuang shared his readings of the court gazette with Wang Mai 王邁 (1185–1248). Wang Mai responded in like fashion: On Reading the Recent Report on Jiang Xian’s 蔣峴 [ j. 1196] Ousting with Liu Qianfu [Kezhuang] (和劉編修潜夫讀近報蔣峴被逐) Reading the report we are delighted and compose poems together.

讀報欣然共賦詩,

49. Liu Kezhuang, Houcun ji, 11.15a; Liu Kezhuang, Houcun xiansheng da quanji, 11.8b. 50. Cheng Zhangcan, Liu Kezhuang nianpu, 156–69.



court gazettes and short reports Since antiquity the paths of loyalty and treachery have run a different course . . . Even though one piece of weed has been cut out, It has been a while since the fragrant orchids have withered away. I am reminded of Master Kangjie’s [Shao Yong 邵雍 1011–77] words: “If you want to accomplish things, don’t let people knit their brows.”

95

古来忠佞各殊岐 . . . 惡草剪除雖一快, 芳蘭銷歇已多時。 懐哉康節先生語, 作事莫教人縐眉。51

Wang Mai read the news about the ousting of Jiang Xian, who wrote the censorial report that had led to his and Liu Kezhuang’s dismissals from court, as a sign of hope. Despite the fact that Wang’s allies were still on the periphery of political power, the court gazette proved Shao Yong’s dictum about the connection between achievement and perseverance true and was itself the medium that continuously kept the hopes of scholar-officials in the provinces alive. These examples of the shared reading and commenting on the gazette demonstrate that there was a mutually reinforcing relationship between factionalist politics and the maintenance of channels of communication through which court gazettes and secondary news genres circulated. Local literati relied on allies serving at court or in local government offices for the gazettes. By spreading the news in conversations and through correspondence they also helped build support for political causes. In “Zheng Ning [鄭寧] Shows a Border Report, My Quickly Written and Playfully Submitted Response,” Liu Kezhuang acknowledged his source and used the infor­mation contained in the report to ridicule Song policy toward the Mongols.52 In 1234, the last Jin emperor abdicated. Song forces moved north, but withdrew back to the south upon the arrival of Mongol contingents. Around this time Zheng Ning, who did not 51. Wang Mai, Quxuan ji, 16.8b. 52. Liu Kezhuang, Houcun ji, 10.3a–b; Cheng Zhangcan, Liu Kezhuang nianpu, 129–36.

96

contempor ary dimensions

hold an official title, shared a report with Liu Kezhuang which stated that Mongol armies were mobilizing troops and descending toward the Huai River, the border with Song. In his response to Zheng Ning, Liu Kezhuang exposed the ruling faction’s aversion to read the signs and engage the Mongol forces. Liu Kezhuang had established a reputation as a hawk and was the author of a large body of political poetry advocating the restoration of a unified empire under Song control (see chapter 3). Zheng Ning’s sharing a border report about Mongol activity with him was therefore most likely motivated by a similar political conviction. Gazettes consolidated ties among those who were already closely connected and whose fortunes were dependent on each other. Other evidence suggests that they also fueled discussion about issues that transcended the immediate interests of preestablished networks and in this way provided the potential for literati to form new ties. Judging by reader responses in poems and letters, decisions on war and peace were second only to personnel changes. News about war and peace became the basis for celebration, for the mustering of past memories depicting an alternative to the current political situation, or even for the outright mobilization of opposition. Unsurprisingly, gazette poems tend to be critical of court policy and advocate a more aggressive stance toward the Song Empire’s neighbors: Reading the Court Gazette in the Second Month (二月閲邸報) I have heard of the tens of thousands of soldiers on the frontier, Surrendering and turning their weapons on the immigrants from the north who are returning to us. Surrender is the wrong policy; the state is no longer the state; Burning our own fields leaves no grain; people eat people . . . Bookish students express their frustration and their vain incorruptibility;

聞道邉頭數萬兵, 倒戈歸我我遺民。 處降失䇿國非國, 清野無糧人食人 . . . 書生憂憤空頭白,



court gazettes and short reports Of course there must be officials planning in the interest of the dynasty.

97

自有經綸社稷臣。53

While poems capture frustration, anger, and criticism, letters show how scholars and officials translated their reactions to the news in court gazettes and short reports into political action. They sent letters to sympathetic politicians in the capital or submitted memorials to the throne. Wang Zhiwang 王之望 (1103–70), for example, responded to a memorial reporting a flood and famine in Zhedong Circuit excerpted in the court gazette. The proposal he submitted to the throne included a summary of the investigative procedures and support measures that were required by law and that should be followed in this case.54 Among the corpus of letters on court gazette news in Zhu Xi’s correspondence a small number address issues of either local governance or broader political significance. Letters commenting on a recent report regarding the tendency for prefectural magistrates to prioritize balancing local budgets over public welfare,55 questioning regulations regarding the appropriate mourning dress for Emperor Xiaozong,56 or protesting a request to let a certain general retire suggest that from Zhu Xi’s perspective such issues should be of concern to and thus open to further debate among local officials and their literati interlocutors.57 That news about personnel changes need not always concern factional disputes is evident in the latter case. The main reason why Zhu Xi shared his opposition to this request with a colleague with whom he discussed gazette news several times in the 1190s related to its potential impact on those active in the field rather than the individual concerned. Requests for withdrawals and retirement

53. Wang Mai, Quxuan ji, 14.22a–b. 54. Wang Zhiwang, Hanbin ji, 7.25a–26b. Quoted in You Biao, “Songchao de dibao yu shizheng,” 109. 55. QSW, 250.76, 244.346–47. 56. Ibid., 245.13. 57. Ibid., 250.112.

98

contempor ary dimensions

reported in the gazette became issues of broader public concern. This is also evident from reports on Zhu Xi’s repeated attempts to decline nominations for office he had received. Such refusals of office were used as evidence against him by his political enemies and regarded as signs of his ambition to monopolize officialdom. Zhu Xi complained about the repeated denial of his requests to friends and colleagues. (Interestingly, he observed that the response by short report reached him before the messenger carrying the official response returned.) In a letter addressed to the grand councilor he further commented on recent restrictions imposed on regulations for leave and retirement that appear to have followed one of his requests and that were announced in the court gazette.58 Even though gazettes did not directly function as a platform for a dialogue among central government and local political elites, sharing and commenting on gazette news in other genres provided scholars and officials with a channel for the discussion of broader governmental and cultural concerns. As Zhu Xi’s personal experience with denied requests of leave suggests, those secondary communication channels also allowed for a linkage and an explicit negotiation between the personal and the sociopolitical aspects of literati life. The personal and emotional dimensions of the secondary discourse on gazette news most clearly illustrate the divergence between the official and the broader literati discourse on court gazettes.59 The themes addressed in poems and letters (personnel changes, the question of war and peace, and decrees relating to 58. Ibid., 244.224, 250.98. 59. I distinguish between “official discourse” and “literati discourse,” because these broad terms allow us to interpret the varied responses of the same group of people, and even the same individual, to the dissemination and discussion of court news in a systematic way. Officials serving at court, or provincial officials responding to central directives, resort to a different set of concepts and evaluative tools (official discourse) than officials serving locally and corresponding with peers, officials sent away from court, or local scholars with ties to officialdom and an interest in court affairs (literati discourse). The same individual articulated different perspectives on the circulation of court news depending on his current position, targeted audience, and the speech or written genre in which he expressed his view. As chancellor, Zhou Bida 周必大 (1126–1204) authored a prohibition on short reports in 1188, but when governing in the provinces in 1174, he read a decree included in a short report and contributed to its further dissemination by



court gazettes and short reports

99

particular administrative decisions) overlapped with those outlined in official descriptions of the genre, but the readings captured in poems and letters diverge from the normative reception envisioned as consensus in the official discourse of bureaucratic genres. News about personnel changes did not merely serve as direct orders to individual officials to move in the desired direction, or as a notification to officialdom about changes at the top; in literati discourse it became a valuable source of information on the current whereabouts of family members, friends, and colleagues and the celebration of or expression of dismay over their bureaucratic fortunes. In their readings the listings of personnel changes in the gazettes lost their original shape as a bureaucratic register; individual news items were transposed into the contexts of personal and political networks. Within the contexts of personal and political relationships, literati assigned meaning to individual news items and expressed a variety of emotions and opinions in response. In “gazette” poems literati expressed sympathy for and worry about family members who were directly or indirectly impacted by personnel changes. Cheng Gongxu 程公許 ( j. 1211), for example, was prompted to write a long poem when he read in the court gazette that a new general had been sent to Sichuan.60 Two of Cheng’s nephews had been serving in the Sichuanese frontier zone for several years. A recent letter from them described their hardships as well as those of the local population who suffered from military attacks and the levying of surtaxes. News of the appointment of a new general brought a glimpse of hope, but also reminded Cheng of his nephews’ prolonged difficulties and uncertain future. The gazette was also read as a reminder of past friendships and a source for the maintenance of friendship ties. While living in retirement, Gao Zhu 高翥 (fl. mid-12th c.) wrote, “I read the court gazette frequently to remember my old friends.”61 While attending quoting it in a letter to a friend. Zhou Lun, Zhou Yiguo Wenzhong gong nianpu, 5879; Zhou Bida, Wenzhong ji, 192.14b; 151.1b; SHY, Xingfa, 2.123. 60. Cheng Gongxu, Cangzhou chen fou bian, 5.6b–8a. 61. Gao Zhu, Jujian ji, 24b–25a.

100

contempor ary dimensions

a lantern festival party in 1111, Chao Yuezhi 晁說之 (1059–1129) read in the gazette about a friend’s tenure in Jiazhou 嘉州 (Cheng­ du).62 It had been almost two decades since Chao and his friend passed the civil service examinations in 1082. Chao sent him a poem remembering the joys they shared in the past and jokingly urged him not to let the lanterns shine on his grey hair. The practice of writing and publishing about one’s response to news items suggests that the broader communicative function of the gazette had become acknowledged. For literati elites gazettes had become a site for the construction of a class-based empire-wide imagined community by the twelfth century at the latest. The interest in gazettes expressed a more general interest in current affairs among the twelfth- and thirteenth-century literati. They read them for personal reasons and to network with colleagues and friends across the vertical hierarchies of the official communication network. Letters such as the one submitted by Peng Guinian demonstrate how the circulation of court news in the gazettes and short reports enabled scholar-officials to reshape their roles in communications with the center, and thus in the body politic. In contrast to the roles of receivers of central directives and transmitters of local information assigned to them in official discourse, lower-ranking officials submitted their reflections on a variety of policies announced in the gazettes. The proliferation of argumentative writing, especially concerning administrative questions, has long been considered a characteristic of Song intellectual life.63 Some, such as the Yongjia scholars, proceeded to theorize the participation of lower-ranking officials and scholars and articulated a new interpretation of Confucian political theory. Consultation at the top of the body politic (that is, between emperor and court officials) had become standard political creed by the early eleventh century.64 The idea of an expanded body of participants in administrative deliberation and 62. Chao Yuezhi, Jingyu sheng ji, 6.7a–b. 63. Bol, “Reading Su Shi,” 2. 64. The Tang dynasty classic on government, Essentials on Government from the Zhenguan Era (Zhenguan zhengyao 貞觀政要, ca. 729), embodied this creed.



court gazettes and short reports

101

policy review that included local officials, retirees, exiles, and scholars represented a new vision of the imperial body politic.65 To what extent can this vision of the expanded body politic be substantiated in literati discourse on news transmitted in the gazettes? The actual readership of the court gazettes and short reports extended beyond the readership legitimated in official discourse. Even though the vast majority of surviving Song texts on the court gazette were authored by men who held office (often high office) at some point in their careers, poems and letters by non-office-holding scholars suggest that they also read and commented on court gazettes. Chen Chun 陳淳 (1159–1223),66 for example, who was nominated for a post late in life but died before he arrived on the job, intimated in a letter to a friend that he kept up on personnel changes through the court gazette.67 Moreover, non-office-holding scholars and students read court gazettes secondhand through excerpts and commentary sent to them by officeholders and retired officials. Cai Yuanding, a student of Zhu Xi’s who never passed the civil service examinations and never held office, learned about personnel changes and court policy through his correspondence with Zhu Xi.68 The origins of the court gazette have been variously traced to the Han, Tang, and Song dynasties. This historical debate is currently conducted within a teleological framework and is shaped by the desire to connect the emergence of the daily national newspaper to the imperial past.69 The context within which court gazettes originated and circulated suggests, however, that there is no direct link 65. De Weerdt, Competition over Content, 134–36. 66. I am adopting the dates given by Satō Takanori, “Shin Jon no gakumon to shisō—Shu Ki jūgaku izen.” For a discussion of the controversies surrounding Chen Chun’s dates, see 49, n.1. 67. Chen Chun, Beixi daquan ji, 24.5b. 68. Zhu Xi, Zhu Xi ji, xuji, vol. 9, 2.5177. Chen Lai dates this letter to 1192. Zhuzi shuxin biannian kaozheng, 357. 69. De Weerdt, “ ‘Court Gazettes’ and ‘Short Reports,’ ” 168–69.

102

contempor ary dimensions

between court gazettes and popular mass media, at least not in the form of national dailies. Gazettes did not serve as a forum for public participation and were not concerned with the representation of anything other than the court and officialdom. Yet their reception history powerfully underscores how the leaking, reproduction, fabrication, and sharing of court news fit into a larger transformation of the imperial information order. Through the sharing of court news and matters of concern to particular networks or the broader class of scholar-officials, men in and out of officialdom created and sustained horizontal network ties that linked them to one another as well as to the court. The court remained central to the political identities of Song literati, and was rendered all the more significant because it became visible and accessible as the hub of imperial administration to the politically ambitious throughout the empire. The standardization, centralization, and regularization of the gazette in the tenth and eleventh centuries were at first part of a concerted effort to redirect the flow of communication and ensure more effective central control over regional power holders. These efforts marked a major transition in imperial political culture not only because of their success at instituting an empire-wide system of news distribution, but also because, over time, the regular supply of court gazettes, alongside the dissemination of other bureaucratic genres such as local gazetteers and administrative maps, made the empire legible to the most important stakeholders in the empire. There are good reasons why the reading of and commenting on court gazettes became a common literati activity in the eleventh century and why short reports became a business in the mid-twelfth century. Both of these trends emerged during a period that has been defined by a “localist turn” or the reorientation of literati strategies from the capital to their local communities. This may suggest that the increasing availability of news about court events and decisions in the localities facilitated the localist turn. Literati maintained an interest in court policy, especially in personnel decisions and the question of war and peace as demonstrated in their readings of the gazettes. The dissemination of gazettes and other news sources about current affairs obviated the need for maintaining a presence or personal networks (such as marriage ties with high officialdom)



court gazettes and short reports

103

in the capital and facilitated the double pursuit of imperial politics and welfare management/cultural entrepreneurship in their hometowns. As suggested in their readings, literati read gazettes to follow up on broader political and administrative concerns. They also read them to keep track of family, friendship, and career ties across long distances and across the official-scholar divide. More clearly than in the case of archival compilations, the circulation of court gazettes demonstrates that the pressures of social and political networking kept the demand for information about the court high. Even though the risk associated with the broader circulation of information about new appointments and current decision making was perceived to be more immediate, the nature of elite networks mitigated against the full implementation of measures proposed to enhance security. The circulation of the gazettes shows how they served as a catalyst for communication within networks formed by men with varying ties to officialdom ranging from current post holders at various levels of the bureaucratic ladder, those demoted or expelled, those awaiting appointment, retirees, aspiring candidates seeking patronage, and local elites without degrees or positions. Even though the official stance denied the legitimacy of the circulation of court gazettes beyond their intended audience of current post holders, individual political actors were bound to depend upon the gazettes and their derivatives at various stages of their careers. At the same time, the use of the gazettes as a networking tool confirmed the centrality of the court and positioned it at the center of competing literati networks. In his work on the dissemination of political news in sixteenthand seventeenth-century gazettes, novels, and plays, Wang Hung-tai has shown that personal and factional disputes brought to light in court gazettes also spilled over into other media.70 He argues that just as the court was rendered visible and accessible in Ming and Qing gazettes, novels and plays created the stage on which viewers and readers of all sorts of social backgrounds could engage with court politics and imperial society at large. The Song evidence, and perhaps the later and far more plentiful evidence for the late Ming 70. Wang Hung-tai, “Information Media, Social Imagination, and Public Society.”

104

contempor ary dimensions

and Qing periods as well, suggests that differences between media carried important implications for the kinds of identities they could sustain. The sharing of court gazettes depended on literati networks that were continually reinforced through the appearance of a wide range of news about court and imperial administration. Plays and novels tended to focus on particular cases often relating to personal vendettas and powerfully underscored the role of gazettes as sources of rumor. Even though we may, along with Ming and Qing critics, accept the power of play and novel in conjuring a social imaginary that allowed viewers to identify with the social or mimic the political, gazettes were politically more significant as the medium that tied court and imperial elites through regular and uninterrupted exposure and participation, and that by the same token sustained networks throughout the empire.

Pa r t I I Transhistorical Dimensions of Empire The Chinese Territories

Three The Reconstitution of Empire in Empire Maps

W

hen Song literati fashioned an imperial mission for themselves, the Song court and its dynastic past remained a central focus. As shown in part I, in the archival compilations covering the history of the dynasty and in court gazettes the Song dynastic house and its court continued to be represented as the hub of political activity even as these genres fell into the hands of literati readers and editors in the twelfth and thirteenth centuries. The imperial mission was not only founded on the dynastic past and present; it was legitimated on the basis of universal claims that placed Song rule in a long history going back to the origins of civilization and on a transhistorical map whose spatial dimensions were defined by presumed inherent features of the Chinese common­ wealth throughout history. This chapter examines the transhistorical features of the imperial polity to which Southern Song literati committed themselves. It does so by discussing the production and reading of “empire maps”—graphic representations of the entirety of the Chinese territories. Maps are part of a repertoire of tools that states throughout history have used to grasp and enhance control over complex social realities.1 Early Chinese states were no exception. Along with cadastral surveys, population censuses, and all manner of reports, the first Chinese builders of a unified empire used maps both for practi1. Scott, Seeing Like a State, 77–83, and passim.

108

tr anshistorical dimensions

cal purposes and as symbolic expressions of power and authority. The large concrete map of the unified territories set up in the tomb of the First Emperor of Qin, and the maps used by Han emperors in similar attempts to incorporate regional centers of power, set off a continuous tradition of mapping projects aimed at showing and bringing about the unity of lands occupied by and surrounding the states and settlements that had formed in the second and first millennium BCE. Such tools of simplification and control did not remain the exclusive domain of political and military elites at court. In this chapter, by means of a historical investigation of their social production and uses, I analyze maps as reflections of the changing relationship between the imperial core (the court and the central bureaucracy) and local stakeholders. I first trace the shift from the court to the literati production, circulation, and transformation of empire maps and then explore the different media used in map production, asking to what extent the use of a medium such as woodblock printing resulted from the social shift in production and reading. I also question whether and how the use of different media such as stone steles, rubbings, or printed codices was related to new interpretations and uses of maps.2 From there, we turn to the ways in which map readers viewed the spatial dimensions of the Song Empire and defined its place in Chinese history. I will propose that the definition of and insistence on the spatial features of a transhistorical “Chinese Empire” mark a significant shift in the history of elite political imaginaries and identities. During the twelfth century ambitious central projects to map the land, population, and resources of every corner of the empire were gradually scaled down, not only because of technological limitations and the limited capacity of the premodern state but also because of concerted literati efforts to make mapping part of the repertoire of tools used to exert power and define the identity of the literatus. Empire maps were not the only genre to which political elites had been exposed and at which they tried their hand. References to various genres of imperial maps, including maps for river 2. This chapter is based on De Weerdt, “Maps and Memory” and “The Cultural Logics of Map Reading.”



empire maps

109

and water conservation, military defense, commercial taxes, disaster relief, and wall construction, hint at the broader range of objectives to which mapping techniques were applied—and at a high resolution when compared to mapping elsewhere in the medieval world.3 However, empire maps served broader ideological goals than the more focused administrative goals of local government mapping projects. As we shall see, in their use of maps as a technology to dif­ ferentiate between Chinese and non-Chinese, Song and late imperial Chinese elites resembled their European counterparts. Mapping the empire was for them, as it was for British empire builders, about self-definition.4

Court Mapmaking through the Eleventh Century Maps carry political meaning and were understood to do so ever since they were first written about in Chinese historical records. The trope of the founding emperor reading a map of the empire, first articulated in the early centuries of the first millennium, symbolized the political meaning of the map. It also powerfully underscored the fact that maps conferred exclusive authority on the individual who interpreted them in the early history of their use. For the same reason access to administrative maps was subject to conventions restricting their circulation.5 Maps of the empire were pri3. Pan Sheng, “Songdai dilixue de guannian,” 123–36. 4. In the British case, surveying colonial territory did not only (and not necessarily) serve the practical goal of rendering territory more legible and more easily exploitable through locational accuracy but also defined surveyors as men of science constructing an empire of reason in a world of unreason. Edney, Mapping an Empire, esp. chaps. 9–10. 5. Even though there appears to be a tacit understanding among historians of Chinese cartography that the circulation of maps was restricted by law, there appears to be very little evidence for legal prohibitions on maps through the Song period. There were prohibitions on esoteric charts in the Tang and Song codes, but maps were not specifically targeted. The circulation of maps appears to have become of concern when foreign envoys were involved. Shen Gua 沈括 (1031–95) tells the story of a court official collecting and subsequently burning the local maps that Korean envoys picked up from local government offices while traveling through. Shen Gua dates the story to

110

tr anshistorical dimensions

marily produced for a readership of emperors, court advisors, and generals and were seldom commented upon outside the context of the court before the twelfth century. Their restricted readership and circulation help explain why neither originals nor reproductions have survived. The founding emperors of the Qin (221–206 BCE), Han (206 BCE–220 CE), and Jin (265–316) empires were associated with largescale mapping projects; subsequent unifiers were similarly portrayed with maps as tools and symbols of imperial power. Early references dating to the second century BCE tied the mapping of the empire to universal kingship. Maps of the empire were presented exclusively for the emperor’s gaze. Cartographic control made it possible to survey all territories at once and was a metaphor for imperial sovereignty. According to Sima Qian 司馬遷 (ca. 145–ca. 86 BCE), the First Emperor’s (r. 221–210 BCE) tomb contained within it a monumental map. Made of natural materials and metals, this relief map reportedly depicted major mountains and rivers.6 Through the use of natural resources associated with mountains and rivers the map represented metonymically the natural features of the realm. In turn, the mapped rivers and mountains, as the supposedly immutable natural base layer of the realm, represented the empire as a whole. The First Emperor’s relief map was unique and enclosed within a mausoleum that, so learned Sima Qian, had within it mechanic replicas of archers that would shoot at intruders. It was never reproduced. In it we can read a first and dramatic expression of the imperial imperative to make territory legible to and inseparable from imperial control. It remained unnamed and is still sealed within the First Emperor’s tomb complex, and so never became a prototype of the genre of the empire map. the Xining 熙寧 era (1068–77). Possibly as a result of this event, Emperor Shenzong 神 宗 (r. 1068–85) endorsed a prohibition on the practice of providing local maps to travelers. Shen Gua, Mengxi bitan jiaozheng (Shanghai guji chubanshe 1987 ed.), 13.467–68; SHY, Zhiguan, 22.9. Shen Gua’s anecdote is frequently cited as an example of a policy of secrecy restricting map use. See, for example, Li Qi, “Woguo Songdai ditu dang’an gongzuo de fangfa,” 64. 6. On monumental maps, see Jacob, The Sovereign Map, 39–46. For a description of the First Emperor’s tomb, see Yee, “Chinese Maps in Political Culture,” 78.



empire maps

111

In Song scholars’ accounts the history of imperial mapping began with the first attested uses of maps in the administration of empire. In 117 BCE Emperor Wu 武帝 (r. 141–87 BCE) determined the dimensions of three princely fiefs on the basis of a map or maps of the earth ( yudi tu 輿地圖).7 When, more than one hundred years later, a clan member attempted to restore the rule of the Han dynasty, he too referred to “a map of the earth.” During a conversation with his advisor Deng Yu 鄧禹 (2–58) in 26 CE the newly installed Emperor Guangwu 光武 (r. 25–57) unrolled such a map and pointed out: “If the prefectures and fiefs of the realm are like this, and we have only recovered one piece of it, how could you say earlier that there was insufficient cause for me to be concerned about the realm?”8 Both the terminology ( yudi tu) and the association between the map of the empire and concern for the realm resonated throughout later imperial history. Emperor Guangwu singled out “the prefectures and fiefs” ( junguo 郡國) as the map’s most distinctive features. This term, capturing the primacy of administrative organization in constructing empire, also turned into a generic name for maps of the empire.9 The Guangwu anecdote related the emperor’s reading of a map of the empire to the political ambition of restoring a unified world and was thus particularly relevant for literate elites in the twelfth and thirteenth centuries. An oft-quoted literary piece,10 Lü Zuqian’s 呂祖謙 (1137–81) “Han yudi tu” 漢輿地圖 (“The Han Imperial Maps”), com­memorated the imperialist ambition of an emperor who ultimately succeeded in restoring central authority over all the territories that had belonged to his dynastic forefathers: “We can describe Guangwu’s ambition to achieve restoration as follows: As soon as he unrolled the map of the earth, an area of several thousands of li entirely entered into his mind’s eye. How could it have been possible to only start aiming for one prefecture after one has already 7. Sima Qian, Shi ji, 60.2110. 8. Fan Ye, Hou Han shu, 16.600. 9. In one twelfth-century atlas, Handy Geographical Maps throughout the Ages (Lidai dili zhizhang tu; 1989 ed. Song ben Lidai dili zhizhang tu), administrative maps of early Chinese empires are called “maps of the prefectures and fiefs” ( junguo tu). See below. 10. Zhu Mu, Gujin shiwen leiju, qianji, 13.7b–10a; Xie Weixin, Gujin hebi shilei beiyao, qianji, 5.2b; Shi xiansheng aolun zhu, xuji, 8.1a–4b.

112

tr anshistorical dimensions

obtained one settlement, and to only start thinking about one region after one has already obtained one prefecture!”11 Imperial mapping was not limited to the creation of empire maps. I differentiate between the “imperial map” as a collective noun for all kinds of maps generated under imperial rule (the subject of The Imperial Map)12 and the “empire map” as a genre defined by the extent of the physical area mapped. Empire maps covering the full extent of the areas claimed by an imperial power thus emerge as a genre in Chinese history from the Han dynasty onward. Song emperors bringing about or advocating unified rule in the tenth and twelfth centuries were, like the Han emperors Wu and Guangwu, portrayed as readers of empire maps. Emperor Taizu 太祖 (r. 960–76) allegedly used “a map of the earth” ( yudi tu) to plan his advance into the Sichuan region.13 Emperor Xiao­zong’s 孝宗 (1163–89) early ambitions to prepare for a campaign to recapture the northern territories lost to Jin were epitomized in a lacquer screen that on one side charted his field administration and on the other displayed “a map of the Chinese and non-Chinese” (huayi tu 華夷圖), one of the two most common types of empire maps (see below). As in the anecdote about Emperor Guangwu, the map in Xiaozong’s case served as a reminder of an unaccomplished mission; he reportedly also recommended that copies of it be shared with high court officials (see chapter 4).14 By Xiaozong’s time the type of map (huayi tu) the emperor had on his screen had become a more common sight not only for the high court officials to whom he recommended it but also for many literati in the provinces. Xiaozong’s early reign and anecdotes such as the one related above were fondly remembered among late Song elites because the second Southern Song emperor’s response to the empire map matched the feelings they attached to it. 11. This mock preface was part of a set of texts submitted for a polymathes examination. Lü Zuqian, Donglai ji, waiji, 4.23b; Lü Zuqian, Lü Zuqian quanji (Zhejiang guji chubanshe ed.), 1.676. 12. Implicitly this definition is used throughout the volume. It is most clearly articulated in Edney, “The Irony of Imperial Mapping.” 13. Du Zheng, Xingshan tang gao, 6.1b. 14. For a translation of a relevant passage and references see p. 210 and accompanying note 72.



empire maps

113

Up until the late eleventh century, however, viewing empire maps had more or less been the prerogative of emperor and court. The first Song emperors continued and expanded the mapping initiatives of the Sui and Tang courts.15 Local and regional jurisdictions were required to submit maps and registers on a regular basis; the registers were to be sent along with the maps in leap years (i.e., every few years) and contained summary data on population numbers, household property, taxes, official staff, and changes on jurisdictional boundaries.16 Gazetteers gathering data of interest to the court such as population figures, local products, relative distance to neighboring jurisdictions, local government offices and infrastructure, cultural and religious institutions, local customs, natural and strategic features, and notable inhabitants were also sent up. Initially they were sent at irregular intervals and may not have conformed to a standard layout, but by the first decade of the eleventh century the court demanded that they be sent at similar intervals as the maps and registers and conform to a standard layout.17 The local maps were the basis upon which draftsmen drew empire maps at court. The first such map symbolizing the unification of the realm under Song rule was completed in 993; a monumental piece of work on one hundred bolts of silk, it was stored in the Palace Library.18 Similar projects followed until the 1070s, when the polymath and politician Shen Gua 沈括 (1031–95) directed a mapping project that resulted in two survey maps of the empire (a large-scale map and a smaller, presumably shrunken, version) and eighteen detail maps, one for each of the circuits. The circuits were the highest level of the field administration and their commissioners had periodically been charged with the compilation of regional maps.19 Shen Gua’s project was the last investment of the Song government in large-scale mapping. His mapping activities inaugurated a 15. The Sui court requested maps from local jurisdictions; under Tang rule such maps were to be sent in according to a regular schedule ranging from three to five years. Pan Sheng, “Songdai dilixue de guannian,” 56, 66–69; Xin Deyong, “Tangdai de dilixue.” 16. Pan Sheng, “Songdai dilixue de guannian,” 63–64. 17. Ibid., 69–73; Mostern, Dividing the Realm, esp. 90–99. 18. Pan Sheng, “Songdai dilixue de guannian,” 109. 19. Ibid., 112–15.

114

tr anshistorical dimensions

historical shift in the production of maps. Not only was the survey map purportedly copied and circulated, but also Shen Gua discussed techniques for making and copying maps in his notebook, Mengxi bitan 夢溪筆談 (Notes by Dream Creek). In the twelfth century, mapmaking and discussions about maps were no longer solely for the more technically inclined, like Shen Gua and the surveyors who assisted him in his work; increasingly, literati read and discussed maps of various types and those less familiar with mathematics and cartographic techniques also tried their hand at the graphic representation of the geography of the empire (see chapter 4). Other historians have recently highlighted a similar transition in the production of other types of geographical knowledge. As local officials turned to scholars for the compilation of gazetteers or as underemployed scholars took the task upon themselves, gazetteers gradually morphed from summary bureaucratic lists into richer works supplemented with the achievements and literary production of both scholars and officials. Writings from local authors or celebrating local sites filled the pages under traditional subjects, and chapters anthologizing poems, essays, inscriptions, and documents relating in one way or another to the area were also added.20 The reinvented gazetteers, now increasingly called descriptions (zhi 志) rather than guides ( jing 經), were copied, printed, traded, and collected among elites with an interest in publications with a local theme.21 Gazetteers were read and cited for scholarly, administrative, and recreational reasons and were carriers of local pride and social status. Because they were situated in networks of gazetteers linking those of lower and higher administrative levels, and comprehensive surveys with “specialized gazetteers” focusing on a particular institution or topographical feature, and because literati collected and interpreted them as titles in a virtual library that covered the extent of the empire, gazetteers also belonged to a range of information genres that firmly tied the local to the 20. For examples of these trends, see Fan Chengda, Wujun zhi 吳郡志 (1192) and Zhao Buhui, Xin’an zhi 新安志 (1175). James Hargett describes these changes in more detail: Hargett, “Song Dynasty Local Gazetteers.” Peter Bol compares changes in the com­pilation of local descriptions to other types of publications with a local focus; see “The Rise of Local History.” Adapted from De Weerdt, “Regional Descriptions,” 124–29. 21. On the printing of gazetteers, see Dennis, “Early Printing in China.”



empire maps

115

empire in literati consciousness (see chapter 4). Literati interests were similarly better represented in the empire-wide gazetteers privately compiled from the twelfth century onward. The barebones hierarchical listing of jurisdictions, distances, features, and basic statistics then made way for extensive collections gathering inscriptions and literary texts about the places and features included.22 Nevertheless, the administrative logic of the early gazetteers structured around the official configuration and measurement of diverse places left its imprint on the literati appropriation of the bureaucratic genres of both the gazetteer and the empire map.

Southern Song Empire Maps Like the local and empire-wide gazetteer, the empire map became more visible and began to play a more prominent role in twelfthcentury Chinese society. The earliest extant examples of maps covering the extent of a Chinese Empire also date from the twelfth and thirteenth centuries. These examples allow us to treat in far greater detail what empire maps looked like and how and why they were published. Following a general exploration of these questions, this section also enters the discussion of how the dissemination of empire maps affected literati intellectual and political culture. The first maps of the “Chinese Empire” to circulate publicly in the twelfth century were based predominantly on two models: maps tracing the tracks of Yu the Great (Yugong tu 禹貢圖, Yuji tu 禹跡圖) and maps of the Chinese and non-Chinese (huayi tu). The first group alluded to the mythical tour of the empire related in the chapter titled “The Tribute of Yu” (“Yugong” 禹貢) in Shujing 書經 (The book of documents), one of the Five Classics.23 In the course of the Han dynasty, this collection of texts became a standard component in the education of literate elites, thereby acquiring an 22. Guo Shengbo, “Tang Song dili zongzhi.” 23. The chapter is translated in Legge, Sacred Books of the East, vol. 3, pt. 1, The Shu King, 63–76. For a recent discussion of this text, see Lewis, The Flood Myths of Early China, 30–33 and passim.

116

tr anshistorical dimensions

authority that imputed to it not only permanent truth but also political significance. As a classical text, “The Tribute of Yu” (which most likely belongs to a group of texts added to The Documents corpus around the time of unification under the Qin and Han dynasties) was considered the embodiment of enduring principles ( jing 經) and seen as indispensable for the understanding of the order of the world and, in a practical sense, a guide to the organization of imperial space.24 The nine domains of the Chinese territories described in “The Tribute of Yu” became the paradigm for representing the empire by means of its major administrative subdivisions ( jiuzhou 九州).25 “Map of ‘The Tribute of Yu’” and “map of the nine prefectures/ regions” became generic names for empire maps. Through their connection to the classic text such maps anchored the goal of territorial unification in the presumed original condition of Chinese civilization. Use of the title “Map Tracing the Tracks of Yu” is, however, not documented before the Song dynasty, from which period several examples have come down to us. These maps were incised on large steles and then circulated as rubbings; they also circulated in print. They were then associated with the same irredentism that accounted for a revival of interest in Guangwu’s maps of the empire. “The tracks of Yu” referred to a passage calling for the recovery of lost territory: “Recover Yu’s tracks; do not let go of the things of old!” from the classic Zuo zhuan 左傳 (Zuo’s commentary; ca. 4th c. BCE).26 On maps of Yu’s tracks, all the territories covered by Yu during his mythical tour of the empire were overlaid with Song and foreign place-names. They were drawn in such a way as to validate Song claims to territory lost to its northern neighbors, first the Liao and later the Jin, with Song sites and foreign sites shown where they ought to have been, rather than where they were in Southern Song times. 24. Lewis, Writing and Authority, 297–300; Nylan, The Five “Confucian” Classics, chap. 3. 25. For a more detailed description of pre-Song works in this subgenre, see De Weerdt, “Maps and Memory.” 26. Kong Yingda, Chunqiu Zuo zhuan zheng yi, in Shisan jing zhushu (2000), Duke Ai 1, 57.991; Yang Bojun, Chunqiu Zuo zhuan zhu, vol. 4, Aigong 1, 1606; Legge, The Chinese Classics, vol. 5, The Ch‘un Ts‘ew, with the Tso Chuen, 792, 794.



empire maps

117

“Maps of the Chinese and the non-Chinese” (huayi tu) became another frequently used designation for empire maps between the Tang and the Song dynasties. This second group of maps extended the model of Yu’s administrative division of the empire to set the empire into a wider geographical context by indicating the large numbers (ranging on different maps from dozens to hundreds) of foreign places and peoples that lay on its fringes. The first milestone in this subgenre was Jia Dan’s 賈耽 (729–805) “Hainei huayi tu” 海内 華夷圖 (“Map of Chinese and non-Chinese within the seas”).27 Although the two types of map differed in the attention given to the surrounding non-Chinese states and peoples, they both projected the same image of an enduring civilization whose geographical coordinates were fixed universally even though they had in historical reality been subject to change. They thus projected an image of a “Chinese Empire”: an ideal image of a commonwealth of the lands defined since antiquity as the nine regions, which were deemed inviolable and to which foreign states paid tribute.28 The extant maps appear either on large stone steles placed in public spaces or in printed books of various genres. In these media maps were reproduced officially, privately, and commercially. The publicity that maps of the empire were given from the twelfth century onward marks a transition in the history of Chinese cartography and is part of a larger development in political culture. 27. For a more detailed analysis of Jia Dan’s map, see De Weerdt, “Maps and Memory.” After “Maps and Memory” was written, I came across Xin Deyong’s reinterpre­ tation of the nature of Jia Dan’s map. It has traditionally been accepted that his map included contemporary as well as former place-names, in the old way, but that it improved on the traditional model of the empire map by using color to differentiate between them. Red was used for contemporary place-names and black for old placenames. Xin Deyong’s recent analysis of Jia Dan’s biographies and traditional readings of these sources proposes that the distinction between contemporary and old placenames was only made in the accompanying notes and not on the map itself. Xin Deyong, “Shuo Fuchang shike ‘Yuji tu’ yu ‘Huayi tu,’ ” 7–12. I do not find Xin Deyong’s evidence conclusive, but it suggests that it is possible that the color distinction was not applied to the map itself. 28. The centrality of tribute in the organization of the world is already apparent in the differentiation of five zones in the world model articulated in “The Tribute of Yu.” Lewis, The Flood Myths, 31–32.

118

tr anshistorical dimensions

Before the twelfth century maps of the empire were drawn on silk or wooden strips or were constructed out of concrete materials such as soil, metal, and wood. Concrete maps such as the map of the empire in the First Emperor’s tomb or the wooden model of the empire reportedly made in the fifth century tend to be fixed in one particular location. Large silk maps appear to have been portable, but the materials and techniques used constricted their circulation. These maps were unique items, drawn on the expensive medium of silk and executed by hand. Manual production made exact reproductions virtually impossible and any kind of copy hard to come by. No hand-drawn silk maps of the empire have come down to us, but several stelae dating between the 1120s and the 1240s capture what single large maps looked like. Detailed empire maps on stelae were in existence by the late Northern Song period. In 1121 the magistrate of Rong County 榮縣 sponsored the erection of a stele titled “Jiuyu shouling tu” 九域守 令圖 (“Map of the jurisdictions of the empire”) (130 by 100 cm), which marks more than 1,400 jurisdictions down to the county level and is thus to our knowledge the earliest existing comprehensive map that is based on the county level. He acknowledged that similar maps drawing on official empire maps were in circulation but that this map was updated to reflect frequent changes in administrative organization; it was acknowledged that not all recent changes could be represented.29 References to empire maps located outside of the court increased from the late Northern Song period onward. It is furthermore likely that the production of empire maps on stelae drew on traditions of stelae production in the Chang’an area going back to the late Tang.30 There is, nevertheless, little doubt that the local sponsoring of map stelae first became a frequent and widespread phenomenon following the collapse of Song power in the north. Famous examples of Song stele empire maps postdating the Jurchen invasions and occupation include the 1136 “Yuji tu” 禹跡圖 (“Map 29. Zheng Xihuang, “‘Jiuyu shouling tu’ yanjiu,” 35–40. 30. This is one of the main arguments in Xin Deyong, “Shuo Fuchang shike ‘Yuji tu’ yu ‘Huayi tu.’ ”



empire maps

119

tracing the tracks of Yu”), the 1142 “Yuji tu” (which refers to an original Chang’an edition), and the 1136 “Huayi tu.”31 The 1136 maps each measured roughly 80 by 80 cm and were cut on both sides of a stone slab, possibly because they were principally used for the reproduction of rubbings.32 The slab was cut following the breakdown of Song power in the north, under the regime of Liu Yu 劉豫 (1073–1143), who had declared independence from Song and collaborated with the Jurchens.33 Insufficient evidence remains regarding the motivation of those involved in the project but it seems reasonable to read the project as evidence that, both in the north and the south, empire maps on stone preserved the memory of a unified empire. The power of the map as a symbol of the imperial mission is most directly conveyed in the 1247 stele “Dili tu” 墬理圖 (“Map of the administrative organization of the empire”), sponsored by Wang Zhiyuan 王致遠. Wang Zhiyuan obtained the map from the family members of Huang Shang 黃裳 who had drawn a set of charts and maps and submitted them to the future Emperor Ningzong in 1190.34 The text placed at the bottom of this large stele (ca. 197 by 101 cm) pointed out that in Chinese history periods of division had been more prevalent than times of unity, and reminded its readers that the Song founders did not consider their mission complete until the (never-accomplished) recapture of the Sixteen Prefectures along the Great Walls separating the Chinese and their northern neighbors. This mission could still be accomplished provided the emperor and those serving him devoted themselves to domestic reform and, on this basis, the recovery of the north. 31. These maps and those mentioned below are reproduced and briefly described in Cao Wanru, Zhongguo gudai ditu ji, nos. 54–66, 70–72, 82–83 (“nos.” in references to this work refer to map numbers). 32. This argument is made in Cao Wanru, Zhongguo gudai ditu ji, 41, and challenged in Xin Deyong, “Shuo Fuchang shike ‘Yuji tu’ yu ‘Huayi tu,’ ” 36. 33. Twitchett and Smith, The Cambridge History of China, vol. 5, pt. 1, 657–59. 34. This is recorded in the postface incised underneath the stele map. Qian Zheng and Yao Zhiying, “Dili tu,” 46–47; Pan Sheng appears to suggest that the maps were presented to Guangzong. Pan Sheng, “Songdai dilixue de guannian,” 118. This does not seem to be supported in the cited source, however. SS, 393.12000–12001.

120

tr anshistorical dimensions

Another rare surviving example testifying to the continued popularity of maps showing the pre-1127 Song Empire in the thirteenth century is “Yudi tu” 輿地圖 (“Map of the earth,” ca. 1265– 74). This map survives as a (damaged) rubbing (207 by 196 cm) drawn most likely from a large stone administrative map. It displays not only all administrative units from the county level up, but also jinshi quotas, mountain passes, ethnic groups, and surrounding polities. The inclusion of overland and especially maritime transportation routes from Mingzhou 明州 Prefecture can be read as an indication that this map, discovered among the holdings of Tōfuku-ji 東福寺 (Rikkyoku-an 栗棘庵) in Kyoto, originated in this port city famous for its maritime exchanges with Japan.35 Local sponsorship of politically relevant maps is evident from the 1154 “Luguo zhi tu” 魯國之圖 (“Map of the state of Lu”; 171.5 by 88.4 cm). Depicting the city walls, palaces, temples, rivers and mountains, trees, and birds of Confucius’s home state, this type of map was most likely a significant object in ritual performances in local government schools. Even though it too went back to Northern Song models, its transfer to stone in the 1150s suggests that it gained added significance at a time when the present location of Confucius’s home fell under Jin sovereignty. This map was long held at the Wuchang 武昌 Xingguo 興國 prefectural school.36 In the thirteenth century Xie Ao 謝翱 (1249–95) wrote about a similarly titled map in the Dinghai 定海 county school (Mingzhou Prefecture).37 Local government schools and local government personnel played a significant role in the proliferation of such maps on stelae as well as in print. The 1142 “Yuji tu,” for example, was revised and cut by a prefectural teacher named Yu Chi 俞篪. The proliferation of empire maps following the retreat of the court and many of its subjects to the south is even more evident in printed materials of the twelfth and thirteenth centuries. During the twelfth century printed maps of the empire appeared in books in various subject categories. Maps featuring old and contemporary 35. Huang Shengzhang, “Songke yudi tu zongkao,” esp. 57. 36. Cao Wanru, Zhongguo gudai ditu ji, 4, no. 51. 37. Pan Sheng, “Songdai dilixue de guannian,” 142–43.



empire maps

121

place names were included in works of classical exegesis and pedagogical tools providing instruction in the interpretation and memorization of the classics. Examples include Liujing tu 六經圖 (Charts of the six classics) and Tang Zhongyou’s 唐仲友 (1136–88) Diwang jingshi tupu 帝王經世圖譜 (Maps and tables illustrating the governance of rulers). The former contains maps, charts, tables, and illustrations translating and synthesizing the classics in graphic formats. The first woodblock edition dates from 1165, but the earliest extant copy is a pocket edition printed later during the Southern Song period.38 The latter, sometimes considered an encyclopedia, similarly consists of maps, tables, and charts elucidating classical texts and synthesizing data relating to the administrative system of the pre-Qin era. Tang Zhongyou compiled it during the second half of the twelfth century, but the first printed edition, still extant, appeared in 1201. Maps of the empire also appeared in Buddhist texts. Fozu tongji 佛祖統紀 (A comprehensive account of Buddhist patriarchs), a general history of Buddhism modeled on the genre of the dynastic history and covering the period between around 981 BCE and 1265 CE, features three maps, one of them an administrative map of the Song Empire from about 1121 delineating circuits and showing prefectures. The first printed edition appeared toward the end of the Song dynasty. Historical works such as Lü Zuqian’s 呂祖謙 (1137–81) Wudai shi xiangjie 五代史詳節 (Detailed excerpts from “The history of the Five Dynasties”)—part of his series on the dynastic histories, titled Shiqi shi xiangjie 十七史詳節 (Detailed excerpts from the seventeen dynastic histories)—also included maps.39 A Song edition of this work features a map of the states occupying the Chinese territories in the tenth century; a layer showing the circuits (the largest administrative subdivisions) of the Northern Song Empire is displayed on top of the map, correlating the former independent states to subdivisions in the Song Empire. The publishing of maps and their availability to literati readers promoted greater interest in the map as an object of scholarly dis38. For a study of stele and woodblock editions of this text, see Ren Jincheng, “Muke Liujing tu chukao.” 39. Cao Wanru, Zhongguo gudai ditu ji, 12, no. 163.

122

tr anshistorical dimensions

course. This trend is visible in both the publication of atlases and in scholarly engagement with published maps. A map of the empire was included in the 1181 edition of Cheng Dachang’s 程大昌 (1123– 95) atlas “Yugong” shanchuan dili tu 禹貢山川地理圖 (Maps of the topography and geography in “The tribute of Yu”). A handful of editions of Lidai dili zhizhang tu 歷代地理指掌圖 (Handy geographical maps throughout the ages),40 an atlas with more than forty maps, about which more will be said in the following section, were printed between the 1130s and 1270s. Other atlases included Hunyi nei wai jiangyu tu 混一内外疆域圖 (Maps uniting the territories in the core and on the periphery), printed privately around the midthirteenth century but now no longer extant.41 These publications testify to the connection Christian Jacob has made between printing technology and the emergence of the genre of the atlas.42 Just as Elizabeth Eisenstein established a correlation between the printing press and increased intellectual activity more gener­ally,43 Jacob also surmises that the woodblock and copperplate reproduction of maps intensified scholarly discourse on geography. The availability of large numbers of identical printed copies across great distances (when compared to manuscript copies) and the prospect that corrections and additions could be incorporated in follow-up editions of similar volume and geographical spread stimulated participation in scholarly activity and publishing. Even though print often served as a medium to preserve and revive old maps and mapping techniques, it provoked a growing readership to contribute to the discussion and correction of outdated models and the addition of new ones.44 40. My translation of the title strays somewhat from the literal meaning of zhizhang. This term means “demonstrable in the palm of the hand” and, thus, easy to understand and clear. “Handy” captures the original metaphor, meaning “convenient” and “within reach”; its semantic field is stretched somewhat in my use of it here to include the meaning of “within easy intellectual reach.” 41. Yao Mian, Xuepo sheren ji, 38.1a–2b; Pan Sheng, “Songdai dilixue de guannian,” 139. Pan Sheng gives an alternate title, Hunyi huayi jiangyu tu 混一华夷疆域图. 42. Jacob, Sovereign Map, 66–76. 43. Eisenstein also applies this more general argument to the effect of print on cartography (Printing Revolution, 200). For a more extended treatment of her argument, see her Printing Press as an Agent of Change. 44. Jacob, Sovereign Map, 67.



empire maps

123

In twelfth-century Song China the publication of maps in various media (stelae, wooden models, manuscript, and print) went hand in hand with a broader amateur discourse on the making and reading of maps. An apt example is Zhu Xi’s 朱熹 (1130–1200) correspondence on Cheng Dachang’s above-mentioned atlas. Zhu Xi first wrote several of his friends asking them to send him a copy of both text and maps of the atlas (he reinforced the latter request when the noted collector You Mao 尤袤 [1127–94] first sent a copy without them) since he had not been able to purchase a printed copy when first learning of it. After he had eventually managed to acquire a copy printed at the Quanzhou 泉州 prefectural school he addressed a letter to the author himself in which he raised several questions about the maps. In these comments Zhu Xi noted that the description of Nankang’s features in the atlas did not match his own personal observation of the mountains and rivers of the place where he was serving in office. Like many of his contemporaries Zhu Xi created his own maps and discussed mapmaking with his students and peers (see chapter 4). For him, as for Cheng Dachang,45 the discussion of the geography of the features mentioned in the classic was not a narrowly construed philological exercise. A careful and critical reconstruction of the map in the classic with reference to the topographical features of the present (rivers and mountains) was a crucial component in the imperial project to delineate the proper boundaries between the Chinese and the non-Chinese.46 Finally, the growing impact of the civil service examinations on literati learning needs to be acknowledged as a factor stimulating interest in maps and atlases. Twelfth-century Chinese atlases were instructional tools with reference value. Their emergence coincided 45. See, for example, the memorial with which Cheng Dachang presented his work to court, reprinted in the still extant copy of Yugong lun printed at the Quanzhou prefectural school. Consulted in microfilm at the Beijing Library. 46. Pan Sheng, “Songdai dilixue de guannian,” 219–22. For Zhu Xi’s letter to Cheng Dachang, see QSW 245.382–83. Zhu Xi proved similarly critical of other maps and also commented on the poor quality of the local maps he was presented with when serving as a magistrate in Nankang. See De Weerdt, “Regional Descriptions,” 131–32. Ge Zhaoguang notes the shift toward a general elite emphasis on the clear differentiation between Chinese and non-Chinese ethnic markers in “Songdai ‘Zhongguo’ yishi de tuxian.”

124

tr anshistorical dimensions

with the compilation and printing of various kinds of instructional and reference works ranging from administrative and historical encyclopedias to prose anthologies and style manuals.47 Such works targeted students preparing for the civil service examinations, whose numbers were in the range of the hundreds of thousands as suggested by the estimated 400,000 who presented themselves for the lowest-level prefectural examinations by the mid-thirteenth century.48 A contemporary witness, Fei Gun 費衮 (ca. 1192), concluded that Handy Geographical Maps throughout the Ages was just such a work, targeted at candidates and boys and men studying for the examinations. He wrote that the maps in the atlas would be of use to those preparing for the policy response session of the examinations, in which candidates were asked to respond to questions relating to present administrative and cultural concerns on the basis of evidence gathered from the past and the present.49 That this advertisement for the atlas was not without foundation is evident from the range of questions touching upon border affairs (see chapter 4) and the models and techniques used in empire maps.50 The famous Yuan pedagogue Cheng Duanli 程端禮 (1271–1345) confirmed the pedagogical value attributed to both Handy Geographical Maps throughout the Ages and Maps of the Topography and Geography in “The Tribute of Yu” when he listed both titles as must-reads in his much-acclaimed program of study Dushu fennian richeng 讀書分年 日程 (Graduated study schedule, 1310s).51 The examinations led to the expansion of the reading public; this expansion fostered and was fostered by the development of commer47. De Weerdt, Competition over Content and “Encyclopedia as Textbook.” 48. Chaffee, Thorny Gates of Learning, 35. 49. Fei Gun, Liangxi man zhi, 6.9a. In Competition over Content I discuss manuals used in the twelfth and thirteenth centuries in preparation for the policy essay session. 50. Pan Sheng discusses a small set of questions on the appropriateness of the paradigm of the northern and southern boundary lines attributed to the Tang monk Yixing. The model is discussed below. Pan Sheng, “Songdai dilixue de guannian,” 177. Some other examples are given in ibid., 216–17. 51. Cheng Duanli, Dushu fennian richeng, 2.5b.



empire maps

125

cial printing. There is no doubt that the printing of atlases and the addition of maps in other types of works were commercial investments rendered viable by the prospect of a large readership. That this was the case can be illustrated by the early publication history of Handy Geographical Maps throughout the Ages, the first printed historical atlas and the focus of the next two sections of this chapter. At least five or six editions of this atlas appeared in the course of the Song dynasty; different printers published the work, updating information in the maps, altering the attribution of authorship, and adding new prefaces. The original work may date back to the turn of the twelfth century, just before the death of Shui Anli 稅安禮, who was credited as the author in some editions and in contemporary accounts. In the only extant Song edition, which may date back to the 1130s,52 the noted scholar, writer, and politician Su Shi 蘇軾 (1037–1101) is listed as the compiler. Already in the twelfth century readers, including the Neo-Confucian leader Zhu Xi, disputed this attribution,53 and nowadays it is seen as little else than a gimmick.

Defining Imperial Space in the Historical Atlas Maps can imbue us with a sense of order.54

How were maps read in the twelfth and thirteenth centuries? What defined the spatial dimensions of the empire depicted on stele maps and on printed sheets? A close reading of a set of instructions included in Handy Geographical Maps throughout the Ages can produce some answers to these questions. In this section I propose that this atlas and printing on smaller sheets more generally contributed to the formation and dissemination of a set of rules for reading empire maps. 52. For a discussion of the dating of the only extant Song edition, see Cao Wanru’s preface in Song ben Lidai dili zhizhang tu, 3. 53. Zhu Xi, ZZYL, 138.3278. 54. Harmon, You Are Here, 18.

126

tr anshistorical dimensions

Instructions for R e a ders Handy Geographical Maps throughout the Ages is typically described as an atlas consisting of forty-four maps of the empire covering the entirety of Chinese history as it was imagined at the time of writing.55 It helped the novice reader navigate Chinese spatial history in two ways: first, by organizing the maps in a general-to-specific and chronological scheme, and, second, by articulating a step-bystep trajectory through the map. Its organization exhibits many of the formal features that we have come to associate with the atlas. A twofold editorial scheme underlies the sequence of the maps. First, general maps are placed in front of time-specific maps. The first two maps are general, intended to provide an overview of the most important administrative divisions throughout time and an orientation toward the main topographical features (mountains, rivers, and lakes). The subsequent maps focus on specific periods in Chinese history. This generalto-specific organizational principle is repeated toward the end of the atlas, where four general maps are placed in front of a series of maps covering the Song period. These four maps cover the correlations of the empire with the constellations and the capitals from antiquity to the Song dynasty, and the last one shows the northern and southern boundary lines formed by mountains and rivers, which, according to the monk Yixing 一行 (683–727), were the natural foundations of the borders demarcating the “Chinese Empire” from peripheral peoples and states.56 In each case the placement of the general maps informs the reading of the maps that follow. They provide the framework within which the time-specific maps are to be read. This strategy is pedagogical, not only in the formal sense that it moves from the abstract to the concrete or from the introductory survey to a more in-depth treatment, but also in a more ideological way 55. The following discussion will be based on the one remaining Song edition. There are minor differences between this edition and the two Ming editions. 56. The boundary lines defined by Yixing were added to the base layer of maps in some Song atlases, but there was also discussion about the validity of this model for defining the natural boundaries of the “Chinese Empire.” Pan Sheng, “Songdai dilixue de guannian,” 175–78.



empire maps

127

that refers the time-specific back to universal conceptions of imperial territory. Second, the time-specific maps are organized into a chronological sequence systematically covering all major transitions in the political history of the Chinese territories. The atlas takes the reader on a historical tour of Chinese geography starting with the mythical division of the land into the nine regions ( jiuzhou 九州) under the culture hero Di Ku 帝嚳, going through the preimperial period and all dynasties up to the Song dynasty. The atlas thus becomes a comprehensive and well-organized geographical and historical archive. The pedagogical imperative was foregrounded in yet another way in early printed atlases. Attached to the first map in Handy Geographical Maps throughout the Ages, “Gujin huayi quyu zongyao tu” 古今華夷區域揔要圖 (“The general survey map of Chinese and nonChinese from the past through the present”), are a set of instructions called “supplementary explanations.”57 The explanations were reproduced on two pages following the map. As I have shown elsewhere,58 these explanations derived from annotations that may go back to earlier copies of the “Map of the Chinese and non-Chinese” type. However, the transfer of text blocks from large-scale individual maps to the smaller-size printed page led to the integration of the text blocks that had been attached to specific locations on the map and dispersed across the map surface. The editor compiled the textual information into a coherent guide to the map. In the atlas the blocks of transcribed text are ordered into a hierarchical sequence and each block is marked off by a large-font heading. The headings list the main aspects and features on the map the reader should note: 1. Mark administrative subdivisions past and present 2. The size of past and present geography 3. Mark the Great Wall 4. Mark the northern barbarians 5. Mark the Khitan 6. Mark the western territories 57. The two pages of text are referred to as “supplementary explanations” in the table of contents; Song ben Lidai dili zhizhang tu, 5. 58. De Weerdt, “The Cultural Logics of Map Reading.”

128

tr anshistorical dimensions

7. The rise and fall of the western Tibetans 8. Mark Liaodong 9. Mark the eastern barbarians 10. The southwestern barbarians 11. Mark the Xia Kingdom 12. The five prefectures of Gan and Liang 13. Mark Rinan 14. Mark the five ranges 15. Mark India 16. The Xidong Man 17. The names of the states in Hainan 18. The names of the states in the western territories 19. Mark the four rivers

The insertion of headings in large font and the imperative voice turns the map on the preceding page into an object to be analyzed. (I have marked most of the features on a digital map, the result of an overlay of geographic information system [GIS] data and manual markup onto a reproduction of the first map in the atlas.)59 Readers are to locate the features to be “marked” on the map. The Chinese term bian 辨, which I translate here as “mark,” literally means “to distinguish.” Among Song scholars, bian carried an overtone of connoisseurship.60 Those evaluating objects from the past were engaged in sorting them out from a mass of objects and in defining their essential features (bian gu 辨古). Similarly, by marking features on the map in order, viewers applied a hierarchical reading to the map and learned to appreciate what features the cartographer and the geographical tradition from which he constructed the map had deemed essential in the definition of the empire. The instructions, as summarized in the list above, start out with a general description of the dimensions of the empire. Administrative subdivision defines the empire. The history of Chinese space starts with the demarcation of nine zones in antiquity. The nine zones rep59. De Weerdt, “Reading Instructions.” 60. The term appears in titles on connoisseurship such as Yan shi bian gu tu 晏氏辨 古圖 (Mr. Yan’s illustrations for validating antiquities) mentioned in You Mao, Suichu tang shumu, 68a.



empire maps

129

resent ordered space, unity through division, and thus become the reference point for all future units of administrative subdivision. Even though the notes reveal that the types and numbers of administrative units multiply through history, all such units go back to the original divisions of Chinese space. The instructions specify that the units to be marked on the map include twenty-three Song circuits (lu 路) and all Song prefectures. They also underscore the continued relevance of the nine zones, and the norm they impose on the administration of space, by listing Song prefectures under them. The second heading, appended to the first, draws attention to the size of the empire. Its size is to be gauged in two ways: the width and length of Chinese space and the location of non-Chinese. The notes acknowledge that of the many non-Chinese peoples and states listed on Jia Dan’s ninth-century map most have been dropped. Readers are informed that only those who present a challenge are to be noted. From the general picture of administrative subdivision and size, the instructions move on to the Great Wall. Its importance for the reader is underscored by its visually prominent symbolic presence on the general map. It is also the only feature that appears on every geographical map in the atlas. Contrary to Arthur Waldron’s suggestion that a continuous wall is a twentieth-century myth founded on the European orientalist imagination,61 this inclusion on all maps suggests that the Great Wall was in Song times imagined as a marker of empire, an imperialist claim to the territories that fell south of it. The notes acknowledged the wall as a human construct and cited evidence that, along with Waldron’s, suggests that no continuous physical wall was in evidence in the twelfth century. However, from its presence and prominence throughout the atlas we can infer that the Great Wall, along with natural features to be discussed below, had become part of the empire’s timeless base layer. From a series of physical walls in the first centuries BCE, it was conceptually transformed into a continuous artificial barrier and, in a further step, into a naturalized feature demarcating the empire. 61. Waldron, Great Wall of China. Nicolas Tackett shows that walls did exist in Song times and played a role in defining borders: “The Great Wall and Conceptualizations of the Border.”

130

tr anshistorical dimensions

After viewing the extent of the Great Wall the reader is invited to survey the general surroundings of the empire. As the prominence of the Great Wall implies, the peoples living to the north of the Chinese territories merit attention first. As the most formidable challenge during the Northern Song period, the Khitan are singled out as the foremost threat in the north. They can be spotted just north of the Great Wall on the map, even though the Liao Empire they began to build under Taizu 太祖 (r. 907–26) occupied territory south of the imaginary wall. The instructions then take the reader to the west. The notes list western states, several of which are specially designated on the map. “Xi Qiang shengshuai” 西羌盛衰 (“The rise and fall of the Tibetans”) traces the history of the most persistent challenge to the west of the Chinese territories and refers the reader to their location all along the empire’s western edge.62 According to the instructions, the reader should then move across the Chinese heartland to the east. In Liaodong 遼東, the territories east of the Liao River drawn as a wedge across the Great Wall as it makes its way toward the Korean Peninsula, the reader is to focus on two historically important political forces: the Korean kingdom of Koryo˘ (Goryeo) 高麗 and the Jurchen state. The map was originally drawn prior to the Jurchen invasions of the 1120s and still lists the Jurchens as one among various peoples living in the northeast; the notes still describe them as a people that had for the moment accepted Song sovereignty. South of Liaodong, the reader is made aware of the existence of “eastern barbarians”; among them Japan is singled out as the only state to have sent visitors in present times. The peoples living in the southwestern parts of the empire are also to be noted, but like their eastern counterparts are represented as a lesser threat than their counterparts in the north and west. Even though the account of successive attempts to incorporate them in Chinese empires from the Qin through the Song appears to acknowledge failure, the notes underscore that as of the 960s they had all proclaimed adherence to the Song state. After having the reader scout the general surroundings of the empire, the instructions list a series of more specific geographic 62. On Northern Song policies toward Tibetan domains, see Smith, “Irredentism as Political Capital.”



empire maps

131

areas along the western, southwestern, and southern periphery of the empire, including the Xia 夏 Kingdom (1032–1227), the Gansu 甘肅 corridor, northern Vietnam, India, and the settlements of unregistered populations under autonomous rule in a broad area stretching across the present provinces of Hunan 湖南, Jiangxi 江西, Guangdong 廣東, and Guangxi 廣西. This part concludes with two lists of place-names: those of the states in the west and the islands in the South Sea. The lists appear to have been copied from an older map for further reference only; many of these names do not appear on the map. Finally, the instructions turn to sets of natural features of the mainland that carried particular significance in the definition of the empire. The instructions appended to the main administrative map mention only the four rivers: the Yellow River, the Yangzi River, the Huai 淮 River, and the Ji 濟 River. Like the nine zones, these rivers were part of numerical sets of spatial features that had come to define the entirety of the Chinese territories by early imperial times at the latest. The headings accompanying the second general map in Handy Geographical Maps throughout the Ages complete the general instructions by adding other sets of rivers, the origins of the Yellow River and the Five Peaks (Mount Song 嵩 in the center, Mount Tai 太 in the east, Mount Heng 衡 in the south, Mount Hua 華 in the west, and Mount Chang 常 [Heng 恆] in the north).63 These features appear on both the general maps in front and most other maps throughout the atlas. As shown on the digital reconstruction, the flow of the three rivers extends along an east-west axis, and the five mountains are located along a north-south course. These natural features thus reinforce the fact that the map showing the full extent of the Chinese territories is more than human artifact and political aspiration. The features highlighted in the instructions combine to shape a permanent base layer of the empire. This base layer orients the viewer, shows the contours and defining features of the Chinese territories, 63. Song ben Lidai dili zhizhang tu, 11. The stele also includes this passage. Cao Wanru, Zhongguo gudai ditu ji, 44, no. 17; cf. Chavannes, “Les deux plus anciens spécimens,” 229–30. Mount Chang refers to Mount Heng; the text was amended accordingly in Zhang Ruyu, Qunshu kaosuo, qianji, 59.15a.

132

tr anshistorical dimensions

and preserves a timeless and normative form of the empire. The map of the empire consists at its basic level of an administrative grid, the Great Wall, networks of rivers and mountains, and peoples and states on the periphery whose vague identities define the solidity of the empire. For Song cartographers the map of the empire defined Song territory. That viewers shared this notion of the equivalence of map and territory is evident from linguistic evidence: “map” (ditu 地圖) was also used in the sense of “territory,” as in the phrase “to return the old map/territory” (huan jiu ditu 還舊地圖).64

The Dissemination of a Cultur a l Logic for R e a ding the Empir e As shown above, the editor(s) of Handy Geographical Maps throughout the Ages defined a set of rules for reading the empire maps included in the atlas. To what extent were these rules replicated in twelfth- and thirteenth-century map reading and the interpretation of the spatial dimensions of the Song Empire? As a first step toward answering this question and before moving on to a discussion of the reception history of maps at the end of the chapter, I examine how this set of rules began to circulate beyond the confines of the printed atlas. Handy Geographical Maps throughout the Ages was quoted in various contemporary works, and some of these contemporary uses of the atlas allow us a glimpse into its reception history. Its reproduction in one of the better-known encyclopedias of the thirteenth century, Zhang Ruyu’s 章如愚 ( j. 1196) Qunshu kaosuo 群書考索 (Investigations into multitudes of books), suggests, along with the atlas’s publication history, that Handy Geographical Maps throughout the Ages superseded earlier work on the geography of the Chinese territories. Zhang Ruyu spent most of his life teaching at home.65 His encyclopedia was the result of a lifetime of research and teaching. 64. Cheng Gongxu, Cangzhou chen fou bian, 9.18a. 65. Later biographical records note that Zhang Ruyu spent years as an educator at the highest institution of learning in the empire and as a lecturer at court, but, as Peter Bol has shown, these accounts cannot be corroborated and were probably added to new editions of the encyclopedia to heighten the profile of its compiler. Bol, “Zhang Ruyu,” 645–50.



empire maps

133

The earliest edition was printed in the first half of the thirteenth century by Mr. Cao’s Zhongyin Academy (Cao shi Zhongyin shuyuan 曹氏中隠書院), most likely the name of a print shop located in Jinhua 金華 Prefecture. By 1248 this edition was picked up by another printer, who claimed that the first edition was produced inexpensively but that the layout was too crammed; this edition aimed to render the material more accessible.66 Zhang Ruyu’s encyclopedia may have been very popular with examination candidates; by the early fourteenth century an abbreviated edition reportedly circulated “for the convenience of those preparing for the examina­ tions.”67 Zhang’s work, like other work of its kind, covered a wide variety of topics ranging from the classics and the histories to finance, law, bureaucratic organization, ritual, the military, border affairs, astronomy, and geography.68 The geography section took up seven juan (chapters). Zhang divided it up into administrative subdivisions (zhoujun 州郡), customs ( fengsu 風俗), strategic places ( yaohai 要害), population (hu­kou 戶口), registers (banji 版籍), maps of the empire ( yudi tu 輿地圖), land distribution (tianzhi 田制), and water conservation (shuili 水利). With two entire chapters devoted to it, administrative organization was the largest of the sections. It offered an extensive account of the spatial order of the empire and a history of its spatial organization, based on a virtual replication of Handy Geographical Maps throughout the Ages, albeit without most of the maps.69 66. Ibid., 646–47. 67. Wu Shidao, Jingxiang lu, 13.5a; quoted in De Weerdt, “Aspects of Song Intellectual Life,” 5–6. 68. Peter Bol traces the compilation and publication history of the encyclopedia and demonstrates that the encyclopedia grew in size in later editions. The section on geography (dili men 地理門) discussed here appears in the first series and was most likely part of the earliest layers of the encyclopedia; it appears in the 1248 Song edition. For a comparative table of the 1248 and later editions, see Bol, “Zhang Ruyu,” 659–60. 69. Zhang Ruyu refers to Lidai dili zhizhang tu only once, when, true to his goal of collating various sources on the same topics in Qunshu kaosuo, he compares its text to that of variant sources; Qunshu kaosuo, qianji, 59.18a. In a later supplement to the encyclopedia, information not derived from the atlas is added under the topic of geography; ibid., xuji, juan 49–52.

134

tr anshistorical dimensions

Like Shui Anli, Zhang Ruyu first introduced the reader to the universal base layer of the empire. He reorganized the sequence of the general instructions somewhat by placing all the content from the general survey maps in front. He started with the correlation of jurisdictions to the constellations and reproduced a copy of the diagram of the twenty-eight constellations included in Handy Geographical Maps throughout the Ages. Then followed the text of the other general maps. Zhang reproduced the instructions appended to “The General Survey Map of Chinese and Non-Chinese from the Past through the Present” and subsequent maps, listing the rivers, mountains, and administrative place names readers were to “mark” on the map of the empire. The sequence of the instructions followed that in the atlas, with the exception that all instructions relating to the periphery of the empire were dropped. This omission may be due to the knowledge that the peoples and states on the periphery were discussed in more detail elsewhere in the encyclopedia under the subject “Non-Chinese” (Yidi men 夷狄門).70 Zhang’s adoption of the reading instructions in Handy Geographical Maps throughout the Ages suggests that he considered them a better introduction to the geography of the empire than earlier alternatives such as the geographical treatises in the dynastic histories or the comprehensive gazetteers of the empire. He also copied the atlas’s chronological tour through the history of imperial territory. The headings for Zhang’s written overviews of the spatial extent and history of all major dynasties corresponded exactly to the map titles in Handy Geographical Maps throughout the Ages. With minor changes and additions, the text matched the explanatory text, and sometimes the tables, attached to the atlas maps. Excepting the diagram of the constellations, maps of the empire do not appear to have been reproduced in Investigations into Multitudes of Books. Maps were more difficult to reproduce than most other kinds of figures; their reproduction required a higher investment of time and effort and resulted in a higher cost. Beyond considerations of cost, the absence of the graphic representations of the empire may also suggest that readers could come across maps of the 70. Ibid., bieji, juan 22.



empire maps

135

empire in other settings and that the more crucial task of the encyclopedist and printer was to teach them how to decode this genre of map. Once memorized, the instructions could also serve as a mnemonic aiding the student to reconstruct a map visually even in the absence of the physical artifact. The reproduction of the map reading instructions and explanatory texts preserved and promoted a cultural logic for imagining and interpreting the map of the empire. The notes under “Mark administrative subdivisions past and present” mapped Song administrative subdivisions onto the nine zones of antiquity and thus reproduced the overlay of present and past in Song maps of the empire. The preservation of the atlas’s sequence from general to time-specific maps similarly reinforced a logic in which the changing incarnations of administrative division were read against the backdrop of a constant form of imperial territory. The administrative geography section in Investigations into Multitudes of Books covered only Handy Geographical Maps throughout the Ages. Judging from the atlas’s wholesale adoption in this private encyclopedia, the notion gained currency between the twelfth and thirteenth centuries, that in the historical geography of the empire readers could detect both a timeless spatial order and historical trajectories in which that order was lost and restored. The cultural logic embedded in the reading instructions is characterized not only by the features to be noted and the layers to be imagined, but also by what is left unsaid in the text and unmarked on the maps.71 Even though during the twelfth and thirteenth centuries large numbers of maps of the empire were available in the public domain in print, stele inscriptions, and rubbings, none, to my knowledge, represented the Song Empire’s actual extent, and none represented the coexistence of two empires, the Jin in the north and the Song in the south.72 Maps of the Liao and Jin empires circulated 71. For a reading of silences in Japanese maps, see Yonemoto, “Silence without Secrecy?” 72. One possible exception is the atlas Liuhe zhangyun tu 六合掌運圖, listed in Chen Zhensun, Zhizhai shulu jieti, 8.266. According to Chen Zhensun this atlas included forty maps, starting with a map of the footsteps of Yu and ending with a map depicting the area of “the northern territories” (beidi 北地). This suggests that it may have included

136

tr anshistorical dimensions

separately,73 and the geographical knowledge of those empires extended far beyond what was represented on the maps of the empire in circulation. The continued circulation of maps showing the Northern Song administrative subdivision was not due simply to the conservatism or cost-consciousness of commercial printers; rather, it was due to the assumption that the map of the Song Empire ought to represent what the empire should look like and inspire readers to desire the spatial order that the Song founders had been able to restore, albeit with the exception of the Sixteen Prefectures.

Mapping Time Understanding “The Tribute of Yu” is not the same as under­standing the administrative organization of the empire in the present. —Zhu Xi74

So far I have focused on how Song mapmakers mapped space. With its forty-four maps, most of which bear a reference to a specific time period in Chinese history in their titles, Handy Geographical Maps throughout the Ages is also concerned with the mapping of maps of the empire as well as a map depicting the northern territories under separate rule. This atlas is no longer extant and is not mentioned elsewhere. 73. Two maps of Liao territory are included in Ye Longli 葉隆禮 ( j. 1247), Qidan guozhi 契丹國志. They are reproduced in Cao Wanru, Zhongguo gudai ditu ji, plates 112 and 113, and are discussed in the same work, 26–27. See also Huang Shengzhang and Wang Qianjin, “Zuizao de yifu Xi Xia ditu”; Hu Yubing, “Hanwen Xi Xia ditu wen­ xian shuyao.” That maps of Jin territory were also in circulation is clear from a private collector’s description of a book titled Jinguo Cheng’an xuzhi 金國承安須知. He notes that besides information on the bureaucracy of Jin it also included one or more geographical maps. Chao Gongwu, Junzhai dushuzhi jiaozheng, fuzhi, 1132–33. 74. Zhu Xi, ZZYL, 79.2025. Quoted in Xin Deyong, “Shuo Fuchang shike ‘Yuji tu’ yu ‘Huayi tu,’ ” 32. Cf. Zhu Xi’s discussion of Cheng Dachang’s work on “The Tribute of Yu” in the context of contemporary topographical features, discussed in Pan Sheng, “Songdai dilixue de guannian,” 219–20.



empire maps

137

time. Time was mapped both sequentially (in time slices) and through the overlay of different time layers on the same map. These two seemingly contradictory ways of treating time were fundamental features of Song administrative thinking and, in contrast to earlier arguments, not evidence of premodern ahistorical thinking but rather evidence of a “thinking in time,” a historical thinking invested in continually linking the past (both idealized and histor­ icized) and present. Was Handy Geographical Maps throughout the Ages the first historical atlas? To the historicist mind the timeless base layer of natural and cultural features explored in the previous section appears to contradict the very idea of the historical atlas. And indeed, Jeremy Black, a prominent historian of historical atlases, has concluded that “China was not central to the development of historical atlases.”75 Black based his conclusion on the observation that in their modern incarnation historical maps “were conscious historical statements dependent on a sense of the past as a separate sphere, one that was of relevance and could be interrogated, but which remained separate.”76 The contradiction between a timeless base layer and maps for different periods of time did not, however, lead to the confusion of past and present in Song maps. The separation between past and present was relative (as it remains in the postmodern present), but mapmakers clearly differentiated between the past and present layers in two ways. First, as already indicated above, the atlas included a series of period-specific maps that follow the small number of general maps in chronological order. In the period-specific maps the editors depicted how the organization of space at various points in time either approximated or, more frequently, deviated from the ideal order depicted in the general map. The two Song maps serve as an illustration. The first of these, “Taizu huangdi zhaozao zhi tu” 太祖皇帝肇 造之圖 (“Map of Emperor Taizu founding the dynasty”), shows the 75. Black, Maps and History, 4. 76. Ibid., 8; also 6.

Fig. 3.1 “The General Survey Map of Chinese and Non-Chinese from the Past through the Present” (“Gujin huayi quyu zongyao tu”), from Song ben Lidai dili zhizhang tu. Photograph courtesy of the Tōyō bunko, Tokyo.



empire maps

139

territory that Taizu (r. 960–76) acquired from the preceding Later Zhou 後周 dynasty (951–60). His state is surrounded by the competing states of the Liao, Wu 吳 (908–78), the Eastern Han 東漢 (951–79), the Southern Tang 南唐 (937–75), the Southern Han 南 漢 (917–71), Shu 蜀 (934–65), and the territories held by Zhou Baoquan 周保權 (952–95). The second of the Song maps, “Shengchao Yuanfeng jiuyu tu” 聖朝元豐九域圖 (“Map of the nine zones during the Yuanfeng reign in our dynasty”), shows the result of the Song founders’ campaigns. In contrast to the small number of oppos­ing large states on the other map, this map shows a unified empire consisting of twenty-three administrative jurisdictions. These circuits constitute an atomized world; each is drawn like an island in the sea of the empire. They appear on the same plane; none of the jurisdictions are bolded. The competition, in boldface, has moved to the northern and western periphery where the Khitan Liao and the Tangut Xia states define the limits of the unified world. This map shows that by the Yuanfeng reign (1078–85) the Song government had come close to restoring the normative grid of imperial administration (“the nine zones”), even though the extension of the Liao Empire below the line of the Great Wall continued to be a historical problem. In the normative order of administrative subdivision captured in the first general map in the atlas (fig. 3.1, “The General Survey Map of Chinese and Non-Chinese from the Past through the Present”), the Sixteen Prefectures then under Khitan rule were shown as Chinese territory, and the Khitan, who had been breaching the imaginary line of the Great Wall since the tenth century, were moved to a location north of the Wall. Time was also mapped through a process of overlaying. This is most evident in “The General Survey Map of Chinese and NonChinese from the Past through the Present.” The base layer includes the names of all Song circuits and prefectures. Superimposed on it are the names of select prefectures and circuits from other time periods. Pre-Song place-names are marked off by little circles on top and carry time references such as “antiquity” (pre-Qin), “Han,” “Jin,” and “Tang.” The circles are most likely a translation to the print medium of an older manuscript technique. We know that the early imperial maps of Pei Xiu 裴秀 (224–71) and Jia Dan displayed

140

tr anshistorical dimensions

present and past names and that Jia Dan may have applied color to distinguish between them. Color printing is not attested during the Song period; the color code was therefore replaced by the use of dots superimposed on past geographical names. The selection of the time layers suggests that in general maps the cartographers of the empire were mostly interested in mapping places representing dynasties that had controlled unified empires for extended periods of time. Whereas the mapping of different time layers may appear selfevident in a map that claims to cover the broad outlines of administrative geographies throughout the course of Chinese history, the systematic application of the overlay in the atlas is less so. For example, all seven pre-Qin maps display the base layer of Song prefectures. In these and subsequent maps, place-names related to the time period in the map title are marked by black dots. As shown in “Chunqiu lieguo zhi tu” 春秋列國之圖 (“Map of the Spring and Autumn states”), the overlay of present and past place-names stretches both past space and present time. On the one hand, present prefectures appear to occupy space that, especially in the south, cannot be correlated to past jurisdictions. By placing southern administrative terms on a map of antiquity, the space of the south is stretched far beyond what the historical records could attest. Most of the south was not incorporated into larger administrative structures for centuries, but this technique suggested that the scope of the early Chinese commonwealth was comparable to that of the pre-1127 Song Empire. On the other hand, present prefectures are shown within the boundaries of the Spring and Autumn states, also in the south. The present of Song prefectures is thus stretched far back in the past, suggesting that the Song Empire had been inherited from the preimperial past. The overlay of present and past place-names and jurisdictions was systematic and methodical in Song historical cartography. It was systematic because many of the printed historical maps, including those in Liujing tu, Diwang jingshi tupu, “Yugong” shanchuan dili tu, “Yugong” shuo duan 禹貢說斷 (Evaluating explanations of “The tribute of Yu”), Chunqiu fenji 春秋分紀 (Analyses of the Spring and Autumn period), and Wudai shi xiangjie, superimposed a layer



empire maps

141

of past places on a layer of Song place-names. It was methodical because the layers were visibly differentiated and past features were selected on the basis of their relevance to the map topic and their significance in present historical memory. Past and present are thus mapped as both separate and coexisting. There is no conflation here of antiquity and present reality, but a recognition that the present is part of a layered past and that the past is experienced through the present. The coexistence of and the negotiation between past and present place-names on Song maps were part of a broader cultural strategy to make sense of spatial disorder. It was a strategy that informed administrative thinking and practice in various articulations. Policy proposals, either in the form of memorials submitted by officials or of the policy response essays written by examination candidates, also consisted of an assemblage of different temporal layers (see chapter 4). The authors typically moved between discussing present problems and recalling moments in the far and more recent past. Imagining and reconstituting spatial and political order required a constant reconfiguring of the past and the present. This ordering imperative in administrative thinking stimulated the production of sites in which the past could be retrieved. During the twelfth century the print archive became a privileged site for the discovery of the past. Editors and commercial printers were engaged in an ongoing competition to produce and reproduce archival compilations that at once covered the imperial past and the Song present comprehensively and rendered it easily accessible to the visiting reader (see chapter 1). The historical atlas was similarly conceived of as a comprehensive archive. It aimed to cover the whole past by producing at least one map for each dynastic period in Chinese history, but it also claimed to make specific moments and places of the geographical past readily accessible to the contemporary reader. It allowed for a chronological reading as well as for the consultation and combination of select past and present layers. Editors enhanced the reference value of the atlas by adding brief overviews of the political history and spatial organization of each dynasty mapped in textual and tabular formats. As shown in “Yuan

Fig. 3.2 “The Map of the Wei Dynasty and the Northern States” (“Yuan Wei Beiguo tu”), from Song ben Lidai dili zhizhang tu. Photograph courtesy of the Tōyō bunko, Tokyo.



empire maps

143

Wei Beiguo tu” 元魏北國圖 (The Map of the Wei dynasty and the Northern states”) in figure 3.2, the atlas took on the form of a wellorganized and multimedia geotemporal archive. The right column summarizes the history of the Wei (386–534) dynastic house, measures its duration in generations, explains its administrative organization, and concludes with the military clashes that brought its rule to an end. The left column is a comparative chronological table that juxtaposes the reigns of Wei emperors with those of rulers of other dynasties governing other parts of the Chinese territories at the time. Time is thus mapped simultaneously through the overlay of past and present layers on the map and through the display of the table of attributes of past layers. Time and space are coextensive: the political history in narrative and table help interpret the organization of space; conversely, the map shows the spatial form of political history. However, this map also shows that in Handy Geographical Maps throughout the Ages the spatial form of political history always maps onto the base layer of the “Chinese Empire” and so consistently refers back to the question of order and disorder.

Memory and Destiny: Responding to Imperial Cartography How did readers respond to maps? Did they follow a trajectory or identify features similar to those identified in the reading instructions of the atlas and encyclopedia discussed above? Did map readings change over the course of Song history? Did the symbolic value of empire maps change once literati turned their hands to them in greater numbers? Capturing and translating the effect of maps on readers in the past is a difficult task. The reception history of maps remains uncharted, mostly because of the inherent difficulty of the task: “The way maps were used is probably the most difficult single aspect, since when people look at maps they leave no visible marks on the maps themselves. Their vision is invisible to us.”77 This history is crucially 77. Jacob, “Toward a Cultural History of Cartography,” 192.

144

tr anshistorical dimensions

important, however, for—whatever uses maps are put to—in each viewing and reading a relationship is established between the viewer and the places mapped. As Matthew Edney writes, “The exertion of intellectual and instrumental power over territory through maps is an act of self-empowerment that elicits an emotional response: pride, gratification, belonging, affection, and pleasure but also perhaps fear and anxiety. And it is in this emotional register that we can identify, at last, an understanding of the nature of ‘the imperial map.’”78 Despite the inherent difficulties, I turn to questions relating to the reception of maps and the relevance of emotions therein on the basis of a series of close readings of poems written on the viewing of an empire map, the best sources we have to interpret the settings, readings, and emotions literati attached to empire maps.

R e a ding Empir e M a ps (c a. 800 –1000) A small number of literati poems occasioned by a viewing of an empire map date to the late Tang period. Few accounts remain of the impression Jia Dan’s map made on its viewers; the rare exceptions include two poems from the ninth and the tenth centuries titled “Examining ‘The Map of Chinese and Non-Chinese.’” These were occasional poems in which the authors reflected upon the map in its entirety, presented the conditions under which the map came into their view, and articulated the reactions it provoked in themselves or others. The first was written by Cao Song 曹松, who is mostly known as a ninth-century poet but who also gained fame as one of a group of septuagenarians to pass the civil service examinations in the waning years of the Tang dynasty. The second, by Wu Qiao 伍喬, who served under one of the short-lived dynasties that emerged after the fall of the Tang dynasty in 907, dates from the tenth century. These poems, written on the occasion or in memory of a map viewing, express the readers’ affirmation of the prominence and coherence of the Chinese territories. 78. Edney, “The Irony of Imperial Mapping,” 32.



empire maps

145

Examining “The Map of Chinese and Non-Chinese” (觀華夷圖) With a touch of the brush the earth can be shrunk; Unrolling the map I encounter peace. The Chinese occupy a prominent position; under what constellation do we find the border areas? In minute detail we can discriminate all peaks, on a small scale we can observe all within the seas. I have for a long time wondered about the places I haven’t been; One by one it seems that I have gone through them.

落筆勝縮地; 展圖當晏寜。 中華屬貴分; 逺裔占何星? 分寸辨諸岳; 斗升觀四溟。 長疑未到處; 一一似曽經。79

Examining “The Map of Chinese and Non-Chinese” (觀華夷圖) The difficulty of having to say farewell touches this very essence; You should know that being close together issues from the heart. First I distinguish all the states drawn minutely; gradually I see the outline of all within the four seas on the map. The roads passing through the passes extend through the terrain of Chu, The mountains of Shu tower high, spilling over into the green colors of Qin. From the tip of the brush are entirely revealed the conditions of the human world, High I hang [the map] in my family home.

别手應難及此精; 須知攢簇自心靈。 始於毫末分諸國; 漸見圖中列四溟。 關路欲伸通楚勢; 蜀山俄聳入秦青。 筆端盡現寰區事; 堪把長懸在户庭。80

79. Cao Yin et al., Quan Tang shi, 716, no. 19; http://www.guoxue.com/qts/qts_0716/. 80. Ibid., 744, no. 14; http://www.guoxue.com/qts/qts_0744.htm.

146

tr anshistorical dimensions

Both poems are written in a reassuring tone. In the first, Cao Song expresses the confidence that comes with the ability to visualize space and thus gain control over it. He wrote at a time when the Tang Empire was breaking apart as military governors took control over sizable parts of the territory, yet he managed to capture the moment of peace that the map, and more specifically the act of unrolling the map, conjured. His second couplet attributes the sense of peace evoked by the map to the centrality of the Chinese territories. In Cao Song’s verse the poetic persona is cast as a reader who distinguishes between the prominence of the Chinese territories on the map and the marginality of the non-Chinese by rhetorically denying the latter a fixed place in the cosmos. Whereas each Chinese jurisdiction was traditionally correlated with a constellation, the poet is here implying that border areas occupied by nonChinese populations could not be fixed permanently on the map in this way.81 Turning away, then, from the radical alterity of nonChinese territories, Cao Song used his poem to trace a personal map of familiar places. The reader relates how the features he first noticed were the peaks, a reference to the Five Peaks (Wu Yue 五岳) that, in the mythical geography of the classical texts, defined the topography of the central plains or the Chinese heartland.82 From the mountains the reader made his way to other locations whose names were familiar, but whose place in the empire could only be understood after they had been visited on the map: “I have for a long time wondered about the places I haven’t been; / One by one it seems that I have gone through them.” 81. The practice of correlating constellations to jurisdictions is attested by several Han dynasty texts. For the early history of this practice, see Henderson, “Chinese Cosmographical Thought,” 208–10; Needham and Wang, Science and Civilisation in China, 3: 544–45. Nicola Di Cosmo, in his analysis of Sima Qian’s writing on foreign peoples, argues that this Han historian correlated the location of the Xiongnu (sometimes translated as the Huns) to specific constellations and thus integrated them into the history and cosmology of the Chinese territories. Di Cosmo, Ancient China and Its Enemies, 304–6. 82. The Rites of Zhou and other Han texts list five peaks associated with the four cardinal directions and the center. The names of the peaks varied, but their role in defining the Chinese territories was a common assumption in these and later lists.



empire maps

147

In the second poem Wu Qiao reconstructs a similar process of discovery. His eyes rested first on the various states on the periphery, before being drawn to the core. For Wu Qiao, too, the global view directed him to the empire at the center, where he could trace a trajectory among familiar places and reconstruct links between the regions of the empire that had recently been broken up by the rise of regional warlords. Wu Qiao’s poem attributes to the map of the empire a therapeutic effect similar to the calm and peace invoked by Cao Song. Wu’s final couplet resolves the tension, exposed in the opening lines, as the dread of separation is dissolved in the more permanent display of a map that projects territorial continuity. These early examples of map-inspired poetry go beyond mere geographical description to give an aesthetic and emotive appreciation of the reassurance conveyed by the map to the map reader. The poems of Cao Song and Wu Qiao reproduced the normative image of the empire in the map. The normativity of the map also gave it predictive powers. For Sikong Tu 司空圖 (837–908), a contemporary of the poets, who wrote a brief prose piece on the map of Jia Dan, the map also brought repose. From the rubble of destruction he witnessed around him, Jia Dan’s map was the only thing whose survival merited attention. In Sikong Tu’s reading, the map contained within it the information necessary to restore the unity of the empire. For him, the map’s depiction of places in relationship with each other provoked strategic thinking: by “rolling it open and examining it in full” he noted, “one can predict victory and defeat.”83 Sikong Tu contemplated and found beauty in the strategic value of the map; during the ninth century the map of the empire was not yet the symbol for political action that it became for literate elites in the twelfth and thirteenth centuries.84

83. Sikong Tu, Sikong Biaosheng wenji, 4.5a. For a broader exploration of maps, poetry, and politics in the mid-Tang, see Wang Ao, “Cartographies Actual and Imagined.” 84. For discussions of maps in strategic writing of the Warring States era (ca. 500– 221 BCE), see Yee, “Chinese Maps in Political Culture,” 72–73.

148

tr anshistorical dimensions

R e a ding Empir e M a ps (c a. 1100 –1300) One can reach at the same time the extreme end of Ao in the south and Ji in the north; then one can see that the remote past will be the future present. —Xue Jixuan 薛季宣 (1134–73, writing about an empire map)85

References to maps among the literate elite became more frequent during the Song dynasty. The number of poems written about maps also increased. Below I first review which maps literati referred to in poems, in which settings they encountered maps, and what emotions they attached to the encounter with the empire map in general. I then turn to the question of which geographical features captured their attention. By focusing on the metaphors used to describe these features, I also aim to capture the effect that these poems might have had on their readers. The Testimony of Tears Empire maps appear with greater frequency in Song poems. Map viewings were by Song times not only remembered in poems written to commemorate that specific occasion; poems written for other occasions—such as farewell parties, birthday celebrations, and visits to friends—also recorded observers’ impressions on seeing a map. Accordingly, the settings within which maps were viewed also diversified. Maps were read by individuals who placed themselves in the solitary settings of dark rooms in which lanterns illuminated the map’s image; they were also viewed jointly among friends and colleagues. Similarity of title suggests that at least some of the maps presented in Southern Song poems were like those in Shui Anli’s Handy Geographical Maps throughout the Ages. “The Map of the Nine Zones from the Xiangfu Period” (1008–16) referred to by Lu You below is a case in point. 85. Xue Jixuan, Langyu ji, 8.9a–b. Also discussed in Pan Sheng, “Songdai dilixue de guannian,” 140. For further contextualization of Xue Jixuan’s cartographic work, see chapter 4.



empire maps

149

Sighing over the Book (書嘆) This scholar has tears that cannot be wiped And then sees “The Map of the Nine Zones from the Xiangfu Period” [“Xiangfu jiuyu tu”].

書生有泪無揮處 , 尋見祥符九域圖。86

For Shui Anli’s edition of Handy Geographical Maps throughout the Ages, a slightly later version of “The Map of the Nine Zones” was used. Apart from the correction of some administrative placenames, however, we may assume that the map in Lu You’s possession would have embodied much the same material as the original administrative written text, The Gazetteer of the Nine Zones referred to in the title of another of Lu’s poems: Reflections: On the 28th Day of the 9th Month I Get Up by the Fifth Drum, Pick Out a Book from the Shelf, Find It Is Gazetteer of the Nine Zones [ Jiuyu zhi], and Cry (九月二十八日五鼓起坐抽架上書得九域志泣然有感) I have been around for seventy years, but my heart has remained as it was in the beginning, Unintentionally I spread the map, and tears come gushing forth.

行年七十初心在, 偶展輿圖泪自傾。87

Both map and gazetteer had been commissioned by the Song court at the same time in the early 1010s as complementary documents. The map in Shui Anli’s atlas was based on a revision of the gazetteer completed in 1080. It may be supposed that it was the revised version that Lu You pulled off his shelf, for this version appears to have been readily available during Lu You’s time, to judge from the frequent excerpts from it in contemporary encyclopedias and from the number

86. Fu Xuancong et al, Quan Song shi, vol. 41, 2236.25696. 87. Ibid., vol. 40, 2188.24948.

150

tr anshistorical dimensions

of copies in private collections.88 The Gazetteer itself did not contain any maps, but Lu clearly was familiar with maps as well as with the text of the gazetteer and could easily imagine the book as a map spread out before his eyes. In the final decades of the Southern Song dynasty, Fang Yue 方岳 (1199–1262) referred to another map in Shui Anli’s atlas. This map of the northern and southern boundary lines formed by mountains and rivers was designed by the monk Yixing, even though original copies of this map do not appear in the record. Yixing’s map is another type of empire map: the northern and southern networks of mountains and rivers drawn on it represented for Yi­xing and his later interpreters the natural foundations of the borders that demarcated the “Chinese Empire” from peripheral peoples and states. In a letter written for the official he served, Fang Yue invoked this type of empire map to express his frustration that the natural borders displayed on it were now beyond Song control: “I rest my hand on the map of the two boundary lines formed by mountains and rivers— this really leads to bitter tears.”89 A poem by the Buddhist monk Wenxiang 文珦 (1210–after 1280) demonstrates that another type of empire map included in Handy Geographical Maps throughout the Ages evoked similar sentiments of loss: Viewing “The Imperial Capitals through the Ages on the Map of the Nine Zones in ‘The Tribute of Yu’” (觀禹貢九州歴代帝王國都地理圖) The rivers and mountains covering ten thousand li, how often does [this territory] decline and flourish!

萬里江山幾廢興!

88. The gazetteer is listed in all three Song private catalogs; it was also in the Jiankang prefectural school library collection. Chen Zhensun, Zhizhai shulu jieti, 8.239–40; Chao Gongwu, Junzhai dushu zhi kaozheng, 8.344; You Mao, Suichutang shumu, 40a; Zhou Yinghe, Jingding Jiangkang zhi, 33.10a. Because the Southern Song government did not publish updated editions for self-evident domestic and diplomatic reasons, the 1080 gazetteer of the empire remained in general circulation. 89. Fang Yue, Qiuya ji, 23.27a.



empire maps Reading the map it is only appropriate that I beat my breast in distress It is difficult to find out about the three kings and two thearchs; What can one say about the Han and the Six Dynasties! Nowadays the Central Plains are entirely held by the north; The deep valleys of past times have probably turned into tumuli. From this we can see that present and past are like a dream; Illusions in a land of dreams, how can they be relied upon!

151 覽圖直合拊吾膺。90 三王二帝皆難問; 兩漢六朝何所稱! 此日中原全拱北， 異時深谷或為陵。 看来今古皆如夢， 夢境虛無豈足慿。91

Even though the monk concludes that the discrepancy between map and reality underscores one of the fundamental truths of Buddhism (the illusion of all existence), the poem suggests that for him too the map served as a painful reminder of the mismatch between it and the division between north and south at the time of writing. Poems written in the post-1127 period more generally bore testimony to the tears that empire maps evoked in their readers. Such poems reflected the disconnection between the normative empire map and what their readers saw as the disorder of space brought about by the signing of the peace treaty in 1141. Zhang Yuangan 張元幹 (1067–1143), for example, remarked wistfully in his poem “Jianyan ganshi” 建炎感事 (“Remembering the events of the Jianyan period [1127–30]”): I carefully examined a map of the empire, I’d rather see it once more implemented on the ground.

檢校輿地圖, 寧復見施設。92

90. Some editions read “zhen 真” (“really”) instead of “zhi 直”; Wenxiang, Qianshan ji, 10.4b. 91. Fu Xuancong et al., Quan Song shi, vol. 63, 3324.39634. 92. Fu Xuancong et al., Quan Song shi, vol. 31, 1784.19895.

152

tr anshistorical dimensions

The map itself had become a powerful reminder of loss. For Lu You 陸游 (1125–1210), too, one of the most celebrated poets of the twelfth century, maps evoked tears: Examining a Map of Qin and Shu at Night (夜觀秦蜀地圖) This map by the light suddenly meets my eye Grey-haired and wandering, I feel miserable in the extreme.

燈前此圖忽到眼, 白首流落悲涂窮。93

At Night I Look at the Map of North of the Huai That Zixu Obtained (夜觀子虡所得淮上地圖) Tears fully shed, by the light I look at the map.

泪盡燈前看地圖。94

Two generations after Lu You, Liu Kezhuang 劉克莊 (1187–1269) testified to the continuing power of tears as symbols of the trauma resulting from the enduring loss of the northern territories. Reflections on the Past (感昔) This lowly functionary is overcome by solitary anger By the nightly window frame, tears flowing, I look at the map.

螻蟻小臣孤憤意, 夜窗和泪看輿圖。95

The Empire as Home Feelings of anxiety were closely intertwined with the opposite feeling of belonging or at least a desire to belong. This feature of twelfth- and thirteenth-century poems about empire maps is encapsulated in the recurring metaphor of the home. For Edney, British “imperial maps” differ from other state maps in that they exert power through exclusion: the colonies are mapped 93. Ibid., vol. 39, 2167.24564. 94. Ibid., vol. 40, 2189.24964. 95. Ibid., vol. 58, 3035.36175.



empire maps

153

not for their inhabitants but for agents of empire whose relationship with the indigenous population is by its very nature antagonistic.96 Mapping projects associated with nation-states, by contrast, generate meanings and feelings that tie inhabitants to the nation.97 The responses of twelfth-century Song Chinese readers fell somewhere in between these two worlds. Southern Song poets wrote about territories in the north that now were home to Jurchen others, as well as Han Chinese who had accepted their rule, in ways that construed all these territories as quintessentially Chinese regardless of their distance from the core. They wrote about the trauma of loss in ways that bespoke an identification with something larger than the Sixteen Prefectures or the northern territories themselves. Elite identification with a universal “Chinese Empire” was celebrated in poems expressing feelings of anxiety and belonging generated in response to the empire map. Most of the twelfth- and thirteenth-century poets cited below had not themselves experienced the warfare and displacement of the 1120s. However, they participated in a discourse in which the empire, in its normative full extent, was presented as a “home,” albeit one in crisis. Frustration over the loss of territory was articulated through the imagery of homelessness. Memories of loss and homelessness were tied to the presence of the empire map. At the same time, the main features of the empire map opened up avenues for the recollection of those parts of the empire imbued with particular significance. Just as the home is remembered through those parts that have left the deepest impression on its former dwellers, so the empire, when lost, is remembered through those features celebrated in the textual tradition and highlighted on the empire map. For Southern Song readers the empire map on which these features were pulled together into a coherent and united empire also contained the hope for the gradual recovery of its constituent parts and its eventual reconstruction. Chen Yinglong 陳應龍 ( j. 1223), who was born long after the Song court’s move to the south, wrote a poem that explicitly equates frustration over the loss of the north with nostalgia for a lost home. 96. Edney, “The Irony of Imperial Mapping,” esp. 40. 97. Winichakul, Siam Mapped, esp. 138.

154

tr anshistorical dimensions To Xu Xian (Yingte) (呈徐憲應特)

In youth I handled the brush, For a while I was a scholar of textual exegesis. In the middle of the night my sword’s bright rays shine forth, Its bright rays shoot through the imperial avenues of the sky. I want to dispel the emperor’s sorrow; I light a lantern and read the map. Even though the landscape in the south is exquisite, I still miss my old home.

弱冠弄筆硯， 耻为章句儒。 夜半劍气發， 精光射天衢。 欲分天子憂， 張燈閱地圖。 七閩山水秀， 我亦思故廬。98

Home is here, as elsewhere in Chinese poetry and world literature, by definition the old and familiar ( gu 故). It provides a place in which to rest and, in the case of imperial scholar-officials, to escape to from the physical movement associated with study, examinations, or government service. It is there at the beginning of human life and remains its center. The “old abode” is the place to which the poet desires to return. As Yi-Fu Tuan’s reflections on the psychology of place illustrate so beautifully, the home as an ideal is associated with a past of safety and reliability.99 Feelings associated with the home, or the loss thereof, were projected onto a larger plane and allowed for the formation and dissemination of feelings of attachment to the normative empire. The map that reminds the reader of the territories lost in the north in the 1120s provokes feelings of homelessness. The subject of Chen Yinglong’s poem, like Lu You’s and Liu Kezhuang’s, is a solitary scholar in the dark of night. In “Sighing over the Book” Lu You locates his subject on a solitary boat drifting in a desolate landscape 98. Fu Xuancong et al., Quan Song shi, vol. 51, 2737.32233. I am following Quan Song shi; other editions read “Wu 吳” (“from Wu”) instead of “cheng 呈.” Jiajing Ningde xianzhi 嘉靖寧德縣志 (The gazetteer of Ningde County compiled in the Jiajing reign), the original source for this poem, clarifies that this poem was indeed written for Xu Xian. Ibid. (1990), 4.13a. 99. Tuan, Space and Place, 128–48.



empire maps

155

of rain, darkness, and the sorrowful sounds of autumn. In these poems a place called home can emerge only in the light of the lanterns illuminating a map of the empire of old. The map is thus at once a marker of loss and a physical reminder of home that keeps the desire for a return alive. In another poem, “Remembering the Events of the Jianyan Period,” Zhang Yuangan asks: Where do I place my home? The problem lies in building by the midnight moon. To the familiar mountains I have to return before long, How could I be concerned about trekking through the frost and snow!

何處置我家， 患在建午月。 故山盍早歸， 豈憂踐霜雪。100

In this passage Zhang has collapsed personal and imperial geography. In the preceding couplet of this poem, already quoted, he had plotted the destiny of the dynasty on a map of the empire (“I carefully examined a map of the empire, / I’d rather see it once more implemented on the ground”); here he commits himself to the reconstruction of his former home in the mountains of old. As we saw in the reading instructions in Handy Geographical Maps throughout the Ages, the Five Peaks defined the foundation layer of the empire; four of those peaks were located in the north. Home was the place that allowed the traveler to forget the pain of getting there. The analogy, by implication, was that any consideration of the hardships that would be involved in recovering the empire in full should be overcome by the strength of desire for the homeland. A sense of the collapse of personal and imperial geography is also created through the mirroring of individual solitude and the drifting of a displaced population. Zhang Yuangan continued: The south by nature is frivolous, Hit by a bow, my heart is fractured.

三呉素輕浮， 傷弓更心折。

100. Fu Xuancong et al., Quan Song shi, vol. 31, 1784.19895.

156

tr anshistorical dimensions

In all directions everywhere massive waves, 四顧皆驚波， Pressing on and crying bitterly in unison. 蒼黃共嗚咽。101

For Chao Gongsu 晁公溯 (fl. 1130s–60s) as well, the geography experienced on the ground was one of disorientation. His poem “De Dongnan shubao luanhou Dongdu guju you cun er zhoubei songjia wu huizhe” 得東南書報亂後東都故居猶存而州北松檟亦 無毁者 (“Upon receiving the report from the southeast that my old home in the Eastern Capital survived the upheaval and that none of the tomb sites north of the prefecture has been destroyed”) speaks of waves in all directions, a footloose population that can no longer be reached and calmed by the sounds of court music that used to issue forth from the capital in the north. Chao wrote: The population that is left behind cries to Heaven, Their ears having lost their sharpness, they cannot hear the palace bells of Eternal Joy. . . . . . . . . . . . . . . . . . . . . How can our generation have the firmness of metal and stone, When they fear they will not see the years of full splendor! Lost when leaving the home, I can no longer tell north from south; Where is the old homeland, where the mountains meet the sky?

遺民相對向天泣， 耳冷不聞長樂鐘。 . . . . . . . . 世人寧有金石堅， 定恐不見全盛年。 出門恍惚忘南北， 故國何在山連天? 102

Chao’s poetic persona loses all sense of direction when the home that was the center, and around which the world itself was oriented, is lost. All sense of direction vanishes once the most basic directions of north and south—the directions that structure a political world in which the emperor faces south—can no longer be distinguished. An ability to structure the world around oneself depends on the recovery, 101. Ibid., vol. 31, 1784.19895. 102. Ibid., vol. 35, 1997.22395.



empire maps

157

in memory first, of the “old homeland” ( guguo 故國). As in Zhang Yuangan’s poem above, as well as in “Map of the Two Boundary Lines Formed by Mountains and Rivers” of Fang Yue’s letter, the homeland here is defined by the massive and eternal mountain ranges that separate the Chinese world from the non-Chinese periphery. Mountains were not the only feature that served as a metonym for the empire. The empire, like the home, was remembered through a set of features. Yi-Fu Tuan has listed some of the icons that combine into a modern image of home: “the attic and the cellar, the fireplace and the bay window, the hidden corners, a stool, a gilded mirror, a chipped shell.”103 Although the literate home of Chinese imperial times consisted of rather different parts from those enumerated by Tuan, it too was imagined through its parts: its gates, its courtyard, the women’s quarters, the surrounding vegetation. A sense of empire and the image of its past were similarly created through the description of significant parts. In Southern Song poems on empire maps, the palace grounds and the capital cities had particular significance in the imagining of the empire. As suggested in a passage from the poem by Chao Gongsu translated above, the Jin army’s capture of the Song court’s musical instruments and performers symbolized the loss of empire, and the loss of court music contributed to the homelessness of the displaced population.104 Almost a century later, Hu Zhonggong 胡仲弓 (fl. 1250s), a lesserknown author living in the final decades of the Song dynasty, was also led back to the palaces and tombs of the former capital city in the north when he saw a map of the Central Plains. Viewing “A Handy Map of the Central Plains” (觀中原指掌圖) The moon is bright and illuminates the imperial tomb steles, The imperial palaces and terraces have lost their old foundation.

月明空照八陵碑， 玉館瓊樓失舊基。

103. Tuan, Space and Place, 144. 104. On the political significance of court music in the early twelfth century, see Lam, “Huizong’s Dashengyue,” 395–452.

158

tr anshistorical dimensions

Toward the old homeland I cannot bear to turn my head, The western wind blows all over, the old palaces overgrown.

故國不堪回首處， 西風滿地黍離離。105

Hu Zhonggong described the loss of empire through the poetic tropes of ruined palaces and abandoned palace grounds.106 The former home and homeland were now mere “archives of fond memories and splendid achievements,” places whose actual appearance (disrepair and ugliness) did not matter because of the feelings of attachment they had engendered over time.107 The loss of the home, on the other hand, exposed it to corrupting influences; the western wind stood for the invasions from the periphery and the lack of care for the features of the homeland. Homelessness was expressed in images that conveyed the opposite of what the home represented: disorientation, movement, displacement, and ruin. Conversely, the recovery of the map of the empire would have to start with the restoration of those parts most closely identified with it. In a poem presented to a friend who was called to serve the court, Zhang Yuangan identified those as the capital cities: I once manually collated a map of the empire, Going through time, a thousand years, using lead, yellow, and red. This man is exceptional, how would he need this! I would like to see him take back the eastern and western capitals.

書嘗手校輿地圖， 上下千載鉛黄朱。 斯人魁磊豈假此! 愿見克復東西都。108

105. Fu Xuancong et al., Quan Song shi, vol. 63, 3336.39835. 106. In the last line quoted above the poet alludes to “Shu li” 黍離 (“The millet abundant”), a poem from Shijing 詩經 (The book of songs). In this poem an officer of Zhou laments the decline of an old Zhou capital, traditionally interpreted as Hao 鎬, and describes the dilapidation of the former palace buildings. For a translation, see Legge, Shi King, in Sacred Books of the East, vol. 3, 439. 107. Tuan, Space and Place, 154. 108. Fu Xuancong et al., Quan Song shi, vol. 31, 1784.19900.



empire maps

159

The eastern capital is the main Song capital Kaifeng, but all other capitals, one for each of the cardinal directions, were also located north of the Yangzi and Huai Rivers, that is, in the lost territories. In Handy Geographical Maps throughout the Ages the four capitals were all permanent features in the empire’s administrative foundation and thus essential to the maintenance of spatial order. The images of violated palace grounds and capitals in disarray returned the reader to the map, which provided the design for the reconstruction of the ransacked home. The map showed the capitals and the mountains to which the poets’ memories had returned. Apart from the capitals and the court, other parts of the empire also became a metonym for it. Hu Zhonggong’s “A Handy Map of the Central Plains” probably also included the Sixteen Prefectures, which the Song emperors had been trying, again and again since the founding of the dynasty in 960, to recover by force or negotiation, without success. Before the establishment of Jin sovereignty in the north, the provinces had been crucial to the Liao economy. The Liao court had steadfastly refused to return them to Song control.109 Seeing the names of the lost Sixteen Prefectures on the map served as a constant reminder of the incompleteness of the Song administrative structure. The general maps in Handy Geographical Maps throughout the Ages show the Sixteen Prefectures set squarely within the “Chinese Empire”; the time-specific maps show how the Khitan Liao had become the acknowledged power over these prefectures in 938 and that they were still in control in the 1080s when the Song maps included in the atlas were drawn.110 The contested status of the prefectures is illustrated most vividly in “Benchao huawai zhoujun tu” 本朝化外州郡圖 (“The map of 109. In Chinese historiography Shi Jingtang 石敬瑭 (r. 936–42), the founder of the short-lived Jin 晉 dynasty (936–46), is usually blamed for the loss of the Sixteen Prefectures. Shi Jingtang recognized Liao control over the prefectures in 938. The significance of this event is also marked in Handy Geographical Maps throughout the Ages. The Khitan are first shown breaching the imaginary divide of the Great Wall and taking control of the Sixteen Prefectures on the map showing the Chinese territories during the Jin dynasty. “Shi Jin ji qiguo tu” 石晉及七國圖, in Song ben Lidai dili zhizhang tu, 74–75. For a modern cartographic interpretation of this event, see Franke and Twitchett, The Cambridge History of China, vol. 6, 71. 110. Song ben Lidai dili zhizhang tu, 94.

160

tr anshistorical dimensions

autonomous jurisdictions during our dynasty”). Along with other prefectures on the western and southwestern periphery, the Sixteen Prefectures were specially marked as jurisdictions, in which “even though they pay taxes, population registers are not sent up to the Board of Revenue.”111 The atlas’s editor recognized that these jurisdictions had fallen under Liao control, but was contesting the fact by suggesting that they remained under the influence of the Song Empire. The Sixteen Prefectures had—even before the loss of the entire north—stood for the empire as a whole. For viewers of the map, they were a focal point in the planning of recovery. Zhang Yuangan reminded readers of the poem last quoted: In his mind long-term plans are clear; His learning is all-encompassing. He knows very well that disaster resulted from [the scheme] to take You and Ji [jointly with the Jurchens]. Yet feels that a ruler of valor will retake Qing and Xu.

胷中逺略指諸掌， 表裏拄腹撑腸書。 深知禍起取幽薊， 頗覺氣王吞青徐。112

The places referred to in the penultimate line, Youzhou and Jizhou, were two of the Sixteen Prefectures, and “take You and Ji” had been the catchword for recovering the Sixteen Prefectures ever since the founding of the Song dynasty in the tenth century.113 Even though Zhang Yuangan, along with the other critics of the reformist government that dominated the court in the early decades of the twelfth century, blamed the loss of the north on the ill-conceived plan to join forces with the Jurchens in the pursuit of the recovery of the Sixteen Prefectures, the ideal remained. Zhang, however, after 111. Ibid., 94. 112. Fu Xuancong, Quan Song shi, vol. 31, 1784.19900. 113. Shao Bowen 邵伯溫 (1057–1134) briefly reviews discussions about retaking the Sixteen Prefectures in Wenjian lu (in Shao shi wenjian lu, SQKS ed.), 6.8b. Note that in the Zhonghua edition, the phrase “You Ji” appears as “You Yan 幽燕,” which was also commonly used as a reference to the Sixteen Prefectures. Shao Bowen, Shao shi wenjian lu (Zhonghua shuju ed.), 6.53.



empire maps

161

studying the map of the empire, noted that this ideal now had to be pursued by Song forces by way of Qingzhou and Xuzhou, both prefectures located south of the Sixteen Prefectures. Zhang Yuangan’s poem, and those of the other poets quoted here, reveals how in twelfth- and thirteenth-century China, maps and writing about maps provided a space for critics of appeasement. Time and again Lu You depicted his frustration over the political status quo by evoking an expressionist tableau: a solitary man in a dark and empty space illuminated by white light mulling over a map or a book. The map was a symbol of lasting frustration. At the same time it was an expression of the Song literati desire for action. As Lu You put it, “I have been around for seventy years, but my heart has remained as it was in the beginning.” Those in Song China who wrote emphatically about their experiences on viewing empire maps associated the map with men of valor who were ready for action. Their expressions of tears and anger, their imaginary drawing of swords in the deep of night, their denunciation of bookish scholars and glib politicians, and their romantic visions of the trip home all sprang from emotions unleashed by the sight of empire maps. These men were responding directly to the imperative embedded in the ideal empire maps, and they were to become patriotic heroes in the eyes of later readers of the poems.114 The fact that dozens of maps of the empire from the twelfth and thirteenth centuries remain whereas none drawn before this time have survived is no coincidence. The shift from court to literati 114. Lu You and Liu Kezhuang were among a group of poets who were remembered for their hawkish politics and the articulation of irredentism in poetry. Modern biographies of Lu You are numerous and cast him as a nationalistic hero. Modern criticism of the work of Lu You and Liu Kezhuang similarly reads the patriotic tone of some of it as an early manifestation of nationalism. On Lu You and his poetic work see, for example, Qi Zhiping, Lu You zhuan lun; Qiu Minggao et al., Lu You pingzhuan; Guo Guang, Lu You zhuan; Zhu Dongrun, Lu You zhuan; Li Zhizhu, Lu You shi yanjiu. In the introduction to the annotated edition of Liu Kezhuang’s songs (ci), Ouyang Daifa and Wang Zhaopeng (Liu Kezhuang ci xin shi ji ping, 1–2) emphasize this poet’s continuation of “the nationalist tradition” in poetic writing, which they traced back to another Southern Song poet, Xin Qiji 辛棄疾 (1140–1207).

162

tr anshistorical dimensions

production and the use of the print medium for the dissemination of empire maps help explain what may otherwise appear to have been the sheer bias of survival. Literati mapmakers and map readers placed their work within traditions of official mapmaking and did not claim to have significantly altered the empire map, even though many among them may have simplified mapmaking techniques considerably. This simplification was deliberate, as maps were drawn and annotated to convey the normative spatial dimensions of the “Chinese Empire” and to situate individual polities in transhistorical time. The cultural logic of reading empire maps was not only articulated in the maps themselves, but was also reinforced in secondary genres such as encyclopedic texts and poetry reproducing map interpretations. The use of print technology in the dissemination of empire maps changed the ways in which they were read. The insertion of empire maps in commercially published books best illustrates their commodification: from unique carriers of imperial authority, they turned into standardized images of the empire reproduced in atlases, works of classical exegesis, political history, and Buddhist historiography and were converted back into printed text in encyclopedias such as Investigations into Multitudes of Books. Print not only made maps of the empire more widely available, it presented them in formats that made them easy to manipulate, allowed them to be taken to a desk or held in hand, and thus rendered them available for consultation. As is evident from its title, Handy Geographical Maps throughout the Ages included maps of the empire that were “demonstrable in the palm of the hand”—the same term was at the time also applied to other printed maps. Publishers thus claimed to produce maps that provided convenient, clear, and reliable information about the empire. Maps were added to printed books as aids in the study of history and in classical exegesis. The atlas stored information from heterogeneous sources into formats such as the map and the table that allowed for the visualization and integration of fundamental markers of space and time. Maps that were “demonstrable in the palm of the hand” also instilled in the minds of their readers an image of the empire. Song writers placed the production of maps of the empire in a continuous



empire maps

163

history. They traced the origins of current maps of the empire back to models of geographical analysis and mapmaking developed in preimperial and early imperial times. The titles of Song maps directly referred back to the models of “the map tracing the tracks of Yu” and “the map of Chinese and non-Chinese” around which they constructed the history of empire maps. Such map titles corroborate Song historians’ view that the maps included in commercial publications reproduced an image of the empire that had been the normative standard since the beginning of (Chinese) civilization. The pedagogical drive in the world of commercial printing in the twelfth and thirteenth centuries ensured that first-time map readers would learn how to recreate and preserve that image. Reading instructions accompanied the graphic part of the map in Handy Geographical Maps and were further disseminated in commercially printed encyclopedias such as Investigations into Multitudes of Books. Beyond the simplifications of earlier models first developed in monumental manuscript maps, it is doubtful that printers changed the image of empire significantly; the map of the empire remained a symbol of political unification, administrative organization, and the centrality of the “middle kingdom.” Not only the atlas that produced the instructions and the encyclopedias that reproduced them but also the poems written about maps contributed to the formation of a cultural logic of reading and responding to maps of the empire. The poems of the more popular poets such as Lu You and Liu Ke­ zhuang were more than mere emotional responses to an occasional map viewing. They were also secondary sets of instructions, reinforcing the instructions given in the printed atlas and suggesting to their readers how to respond to particular aspects and features of maps they might encounter. Although the maps featured in the commercial editions of the Handy Geographical Maps throughout the Ages had been drawn before the loss of the northern territories, their circulation is first recorded only when they became powerful reminders of the empire as it had been, and as it ought to be restored. Maps were now associated with memory and destiny, not with the reality on the ground. Among hawkish scholar-poets such as Lu You and Liu Kezhuang the depiction of close confrontations with empire maps in the privacy

164

tr anshistorical dimensions

of the home was a call to political action, either in support of the imperial desire for the immediate recovery of the northern territories, as in the early years of the reign of Emperor Xiaozong, or as a challenge to a government settling for negotiated peace. As articulated in the poetry about map reading, as well as in other genres of writing, the call added a sense of urgency to politics: the restoration of the empire presented as the home that needed to be kept whole became the standard by which policy was to be evaluated. In this sense it amounted to an ultimatum that could not be easily overridden because it was based on deeply held beliefs of what the empire was, which were captured in the map’s normative base layer of natural and administrative features.

Pa r t I I I Margins, Borders, and Frontiers

Fou r Strategic Discourse Building Frontiers in the Public Domain

B

orders and the fields of knowledge related to the study of their role in government, in security and military strategy and in history, became a preoccupation for many Song literati. This chapter examines what sustained literati interest in border affairs. I first demonstrate how politicians and teachers used the civil service examinations to promote interest in fields such as historical geography, military geography, and border affairs. Second, by uncovering the work of amateur mapmakers in the twelfth and thirteenth centuries I show that the incorporation of these fields in literati learning extended beyond the scope of the examinations. The expertise claimed by some of these men suggests that the shift from the court to literati production of maps and gazetteers was a phenomenon limited to a small group of scholars. To gauge the impact of their work and the broader interest of literati in border affairs, we will then turn to the question of what place materials relevant to border affairs (maps, envoy reports, local and comprehensive gazetteers, ethnographic works) occupied in literati collections and discussions recorded in notebooks and correspondence. The thesis of a twelfth-century structural transformation in political communication has so far been constructed on the basis of the dissemination of state documents and the expanded support for the publication of state documents by literati. In this chapter I extend the analysis of this shift in the social and cultural production of state documents in two directions. First, I ask whether and to what

168

margins, borders, and frontiers

extent literati became involved in the production of original, selfauthored texts on border affairs as opposed to the editing and repackaging of leaked documents. This question will be pursued through an investigation of the social production of maps in this chapter and of military treatises in chapter 5. Second, I discuss the cultural impact of bureaucratic genres and statist frameworks on literati discourse on border affairs. By examining the literati adap­ tation of statist frameworks, we find a further explanation for the paradox central to this book: the central government’s insistence on the secrecy of court affairs (especially border policy) and the simultaneous proliferation of publications and collections on the same.

Border Affairs in Military Studies The study of war received new impetus during the Song dynasty. The number of works classified under the category of “military works” (bingshu 兵書) produced during the three hundred years of Song rule more or less equaled the number of works known to have been compiled during the preceding fifteen centuries.1 In an early study of the art of war in Chinese antiquity, Peter Boodberg wrote of the Song period: Memorials to the throne, admonitions to reform, military manuals, new commentaries on ancient classics and essays on methods through which a “they shall not pass” courage could be instilled into the weakening Chinese legions began to pour into the capital from every quarter of the vast empire. Libraries were ransacked in the search of incunabula on military art and committee after committee was appointed by the court to edit military texts.2

Even though Boodberg may have slightly exaggerated the activity of compilation committees throughout the course of the Song reign, there is no doubt that scholars and officials invested heavily in the 1. Liu Shenning, Zhongguo bingshu zongmu, lists 428 Song titles and 483 titles produced prior to 960. Many titles in the latter category are variant titles of the military classics. 2. Boodberg, “The Art of War in Ancient China,” 1.



str ategic discourse

169

compilation of new titles as well as in the editing and circulation of earlier military texts. It was under the auspices of Emperor Shenzong 神宗 (r. 1068– 85) that seven military texts were canonized and became standard not only in military examinations but also in the strategic thinking of literate elites. Questions about the military classics were raised in the regular jinshi examinations, suggesting that they were also considered necessary reading material for civil administrators. Individual and complete versions of the seven military classics did well on the market, then as now, and were sold in various editions. Private collectors listed multiple editions of these works. The fact that all three extant private catalogs list all titles, and many in competing editions featuring the annotations of a growing group of commentators, suggests that the military classics had become a basic staple among private collectors.3 The growing interest in military affairs was also reflected in the specialization of military knowledge and growth in fields that were not well covered in the military classics. Such was the case for border defense and the geography of border regions. According to the most complete bibliography of prerepublican military books, Song authors published eighteen works on military defense, while only two relevant titles, published around the turn of the eighth and in the ninth century, respectively, date from the pre-Song era. Based on this source, the growth in the output of border defense titles between the Tang and Song dynasties (900%) far outstripped the overall increase in the production of military titles (about 480%). Similarly, the compilers of this bibliography list four Song titles on the geography of border areas and no specialized works in this field prior to the twelfth century. Even though these statistics are problematic for several reasons, they are indicative of an unmistakable trend.4 3. Chao Gongwu, Junzhai dushuzhi jiaozheng, 631–46; ZZSL, 12.359–63; You Mao, Suichu tang shumu, 56a–57a. 4. Works that discuss border strategy but are not limited to this topic are not included. This number therefore underestimates the production of texts covering this topic in depth, such as military encyclopedias or comprehensive military treatises. In many cases we have only titles, and that makes classification a hazardous undertaking.

170

margins, borders, and frontiers

Borders rarely figured in the military classics and were not singled out as a topic deserving concentrated analysis. In contrast, the resurgence of interest in military theory and analysis between the tenth and the thirteenth centuries was informed by the overarching question of how to defend and extend Song frontiers. Apart from the publication of specialized titles, print shops sold encyclopedias and general military treatises that included sections on border defense and the geography of border regions and prose compilations that included policy questions, essays, and memorials on the same topics.

The Tang-Song Legal Transition Before moving on to a discussion of what the increase in the production of titles on military studies in general and border affairs in particular meant and where it came from, it is important to briefly recall the history of government regulations on their publication. The history of these regulations roughly follows the trajectory of increasing regulation over the course of the Song dynasty on the dissemina­ tion of archival compilations, court gazettes, and newssheets, outlined in earlier chapters. Border affairs were, however, a particularly sensitive area; hence, the tension between court and literati and between secrecy and publicity played itself out more visibly in this field. Changes in publishing regulations between the Tang lü 唐律 of 737 (The Tang code, originally Lü shu 律疏, or The code and subcommentary), which were largely reproduced in the early Song code, (Song) Xing tong (宋) 刑統 (The penal code, 963), on one hand, and Qing­yuan tiaofa shilei 慶元條法事類 (The classified laws of the Qing­yuan period) of about 1202 and later Song legal pronouncements, on the other hand, illustrate how much had changed in the legal domain as well as in the production of materials deemed sensitive by central governments. The Tang Code prohibited the private possession (and thus the private publication) of astrological works, military treatises, esoteric materials, unofficial calendars, and some divination manuals.5 It also included a provision that the punishment 5. In an earlier discussion of publishing regulations in The Tang Code I included maps in this list. Upon rereading this passage, I came to the conclusion that maps were not



str ategic discourse

171

for the dissemination of confidential (defined as relating to military and policing campaigns) and sensitive information (defined as secret but not of major importance to state security, such as observations of anomalies) was increased one degree when the information was shared with foreign envoys.6 The Tang Code thus forbade the possession of types of materials that had been considered detrimental to the preservation of the sociopolitical order since the early imperial period.7 Official documents and other literature relating to contemporary court policy and foreign affairs were not specifically targeted. The dissemination of such news abroad was a concern, but, as it was not disseminated in print, the concern focused on foreign envoys visiting the court and the capital of Chang’an where such information was concentrated. The Code did not differentiate between the dissemination of sensitive information orally, in manuscript, or print. In contrast to the limited discussion of publishing regulations under the section on administrative rules in The Tang Code, The Classified Laws of the Qingyuan Period added a separate section on official documents which, among other things, stipulated the punishments for the printing of policy essays touching upon foreign affairs and the printing of all official documents related to border affairs.8 These regulations were the result of almost two centuries of incessant legislation prohibiting the unauthorized publication of materials on borders affairs. As I have shown in greater detail elsewhere,9 the main concern driving this legislation was the expanding domestic discussion of included as “tu” most likely referred to charts and not maps in this context. See De Weerdt, “What Did Su Che See in the North?” 494. 6. Johnson, The T’ang Code, 2.77–79. Cf. Niida Noboru, “Keigen jōhō jirui,” 446–48. 7. Astrological and esoteric materials were outlawed in 267 under the Western Jin dynasty. More bans on these materials followed between the fourth and the sixth centuries. See An Pingqiu and Zhang Peiheng, Zhongguo jinshu daguan, 13–26. Hok-lam Chan wrote that The Tang Code also banned the transcription and distribution of “gov­ ern­ment statutes” and “national histories” but did not cite relevant evidence. Hok-lam Chan, Control of Publishing in China, 2. 8. QYTFSL, 8.145–48, 16–17.333–79, esp. 146, 364–65. 9. For a more extensive discussion of these prohibitions and tables listing them in chronological order, see De Weerdt, “What Did Su Che See?”

172

margins, borders, and frontiers

Song court policy regarding questions of war and peace with northern neighbors. High court officials proposing regulation on the publication of materials relating to border affairs from the eleventh century onward rested their case on the prospect of their dissemination abroad, but neither they nor other sources provide evidence for the presumed widespread export of sensitive materials to Liao, Xi Xia, or, later, Jin. The heart of the Song’s security problem lay in the commercial publication of materials pertaining to current state affairs for the escalating number of students preparing for the examinations. Besides the examinations there were also broader changes in political culture that heightened the tension between the court’s need for secrecy and the literati appetite for news about the borderlands and especially those parts of the north considered to form part of the Song Empire. References to the north were fraught with ambiguity. In the eleventh century the Song court acknowledged the Liao occupation of the Sixteen Prefectures, territories that were administratively and culturally considered part of the Chinese territories. In the 1140s it acknowledged the Jurchen occupation of all territories north of the Huai River. The north carried ambivalent meanings. For literati it was the carrier of historical memory and an object of desire. To the court it had also proven a potential threat to its existence. In sum, literati discourse on the northern homeland provided both the impetus for the systematization of Song publishing regulations and the reason behind their limited effect.

Crises, Tests, and Grand Strategy Song officials claimed that students in the capital and beyond were exposed to recent materials on border affairs. In prefaces commercial publishers also let it be known that their atlases and anthologies were especially helpful for students preparing for the examinations. I assumed in earlier work that claims about the utility of a historical atlas to examination candidates were most likely a gimmick. Below I revisit the question of whether and how geographic knowledge mattered in examination preparation. Did official examiners or those who wrote policy questions in an unofficial capacity raise questions



str ategic discourse

173

about historical and military geography? Were students equipped to respond to such questions? After considering both the types of questions and the kinds of geographical knowledge tested, I turn to the question to what effect historical and military geographical texts and diplomatic history were discussed in examination literature.

Policy Questions The proliferation of titles on border affairs and the government bans on them were to some extent the direct and somewhat ironic result of the Song emperors’ decision to turn the civil service examinations into the dominant and most prestigious route of access to officialdom and, from the mid-eleventh century onward, to make policy essay questions a decisive component in civil service tests. As policy questions were meant to cover concerns of the greatest contemporary relevance, questions relating to border affairs and military preparedness featured prominently on the agenda, especially during times of high alert. In a question set for students at a private academy Wei Liao­ weng 魏了翁 (1178–1237) wrote: Nowadays is there another matter more important to examiners than border affairs? In the past there was this saying that foreigners attacking each other signals good luck for the Chinese territories. Today, is that good luck or misfortune? Some say that this is a unique opportunity we have not had since the Jianyan reign [1127–30] and that we cannot lose it. Others say that we can securely rely on the proven efficacy of repeated peace agreements since the Shaoxing reign [1131– 62]. Still others say that our inadvertent mobilization of troups in the Kaixi reign [1205–7] should be a warning and an example that cannot be repeated. Yet another opinion holds that we should not mobilize our troops nor send payments and so entrust matters to circumstance and call this self-preservation. Some say that as the foreigners under pressure are about to be annihilated new ones have arisen and this is certainly not the time for peace. Some say that, given that the foreigners have not yet succumbed and would like to make peace arrangements with us, we should also consider the lesson of events since the Xuanhe [1119–25] and Jingkang [1126–27] reigns. Some hold that we

174

margins, borders, and frontiers

can lodge war in self-preservation; others that we can hold the enemy to peace agreements, and still others that we can appease them for the moment while awaiting the moment that they will devour each other. What are the strengths of these various arguments?10

Wei Liaoweng most likely drafted this question some time between 1215 and 1235, when Jurchens and Mongols fought each other in the northern territories formerly held by the Song court. The emergence of the Mongols as a new contender on the scene raised questions not unlike those senior statesmen had faced when the Jurchens advanced into Khitan Liao territory in the early twelfth century. The decision to side with the Jurchens against the Khitans was later seen to have backfired as the Jurchens turned against their erstwhile allies, took the Song capital of Kaifeng in 1126–27, and occupied the northern half of the Song Empire. Wei Liaoweng’s question illustrates two features of the examination questions on border affairs preserved in the oeuvre of Song authors.11 First, such questions frequently address current challenges or opportunities. In addition to the impact of the Jurchen-Mongol conflict on Song policy, questions cover several issues: relations with Xi Xia and strategic policy for the border areas in the mid-eleventh century,12 the appropriate response to the Xia-Jin conflict around 1212–25,13 border skirmishes following the peace agreement concluded between Song and Jin in 1164 (and thus most likely referring to renewed tension in the early 1200s),14 strategic planning follow10. Wei Liaoweng, Haoshan ji, 93.9a–b; QSW, 310.231. Italics in the translation are mine. 11. The following observations are based on a reading of relevant questions preserved in the work of about two dozen authors. This is not an exhaustive list. It is based on questions on border affairs indexed under the section on examination questions in the classified index of the Siku quanshu series (Siku quanshu wenji pianmu fenlei suoyin xueshu wen zhi bu, 2338–60). 12. Frank, “National Security in Northern Song,” 32; Twitchett and Smith, The Cambridge History of China, vol. 5, pt. 1, 300–316. Zheng Xie, Yunxi ji, 13.12a–13a, QSW, 68.34; Zhang Fangping, Lequan ji, 34.12a–b, QSW, 38.16. 13. Franke and Twitchett, The Cambridge History of China, vol. 6, 207–10, 259–61; Wu Yong, Haolin ji, 33.14b–16b; QSW, 316.343–44. 14. Yuan Xie, Jiezhai ji, 6.10a–11b. For a brief overview, see Twitchett and Smith, The Cambridge History of China, vol. 5, pt. 1, 793–96.



str ategic discourse

175

ing successes against Jin armies and the temporary recapture of Kaifeng by Song forces in 1234, and the best course of action against the Mongols following the annihilation of the Jin Empire.15 Other questions do not make explicit reference to specific events but address issues of immediate relevance to military planning, such as military geography or funding and the incorporation of volunteer defense forces. This is particularly the case after the 1130s, when questions about military geography make a regular appearance. All these questions addressed the geography of the new frontiers along the Yangzi and Huai Rivers and the seacoast. The shift in interest from the military geography of the north and west to the core regions of what had been the Song Empire is even more marked in Southern Song examination manuals. In Bishui qunying daiwen huiyuan 璧水群英待問會元 (The epitome of eminent men responding at the Imperial College, ca. 1245) the only strategic locations covered are Sichuan (Shu 蜀), the Middle Yangzi (Xiang 襄), the Lower Yangzi (Jiang 江), the Huai River (Huai 淮), and the seacoast (hai 海); the northern and western borders no longer appear as topics of immediate concern.16 Even when examiners set broader questions about military organization and strategy or identified historical cases, students were reminded of the need to make the past relevant to the present, as in Wu Yong’s question about the distribution and rotation of armies: Gentlemen, starting from the past try to trace for me the origins and dev­elopments in full; go back to the very beginnings and observe the variations. Go over the geographical conditions of the Han, Jin, and Three Kingdoms period. Investigate the military achievements of all great generals of our dynasty. Examine our army’s force in the Xiang, Shu, Jiang, and Huai regions. Compare all of this to the past to make it trustworthy and make it useful to the present so that it can be put in practice. Write this out in detail. Do not make the excuse that students of the classics and scholars are not familiar with military strategy.17 15. Twitchett and Smith, The Cambridge History of China, vol. 5, pt. 1, 855–72; Wu Yong, Haolin ji, 33.11b–14b; QSW, 316.341–43. 16. Liu Dake, Bishui (Nanjing Library ed.), juan 73. 17. Wu Yong, Haolin ji, 33.11b; QSW, 316.341.

176

margins, borders, and frontiers

Some examiners added further urgency to their final instructions and impressed upon their examinees the primacy of practical solutions in reversing the perceived crisis in military security: “If the examiner asks unsubstantial questions, you will all respond with inconsequential talk. What we lack today is not prose. I implore you to deduce why our dynasty finds itself in a weak position without military victories and what should be prioritized among those things that today can be used as effective remedies and immediate solutions.”18 Besides the emphasis on utility to the present circumstances (shiyong 世用; bian yu shi 便於時), questions on border affairs were in the aggregate also characterized by the relatively broad range of occasions for which they were set. Even though we lack information about the level of examination for which many surviving questions were set (editors preparing manuscript texts for print publication may have been partially responsible for this), the small sample includes questions drafted for private school examinations, local prefectural examinations, avoidance examinations, Imperial College internal and jinshi examinations, regular departmental jinshi examinations, palace examinations, and decree examinations.19 This suggests that the opening statement in Wei Liaoweng’s question to private school students may well have held true for examiners at all levels. That military affairs was high on the agenda for examiners and candidates participating in examinations at the capital is also evident from the extensive coverage of the subject in examination manuals, which were based on trendsetting policy essays written at the Imperial College or at the departmental examinations. Policy questions on themes related to military administration and border affairs can be grouped into sets of questions on different but related areas of security planning. 18. Fang Dacong, Tie’an ji, 27.16a; QSW, 322.281. 19. Tian Xi, Xianping ji, 22.13a; Hu Su, Wengong ji, 29.3a–5a; Zhang Fangping, Lequan ji, 34.11a–12b; Yang Wanli, Chengzhai ji, 97.15a–16a; Cheng Bi, Mingshui ji, 5.9b– 11b (QSW, 298.29–30); Wei Liaoweng, Haoshan ji, 93.4a–14b; Hong Zikui, Pingzhai ji, 9.3b–5a; Wu Yong, Haolin ji, 33.1a–3b, 11b–14b; Xu Yinglong, Dongjian ji, 10.2b–6b, 14a–18a; Yang Fang, Zixi ji, 8.28a–31a; Cexue shengchi, essay 1A, 1B.



str ategic discourse

177

1. Wei Liaoweng’s question is an example of the grand strategy type of question, in which candidates were asked to evaluate the pros and cons of the three policy options of war, self-defense (which called for self-strengthening and the defense of areas under occupation), or peace agreements (zhan 戰, shou 守, he 和). 2. Military organization was also an area of broad interest. It ­included questions on the type of army necessary for success and the distribution of power between the central government and field generals. Song armies mostly consisted of professional soldiers stationed in both the capital and, after Shenzong’s reign, in increasing numbers across the Song Empire. Candidates were tested on their ability to evaluate the pros and cons of professional armies based on their knowledge of the record of the imperial armies and the use of conscription armies in earlier dynasties; they were also asked to consider whether and to what extent local volunteers and militias, loyalist forces under Jin occupation, or refugees from the north could be integrated in the Song military. Emperor Taizu’s ­decision, later repeated by his Southern Song counterpart Gaozong, to constrain the power of generals who had contributed to the reunification of the Chinese territories and centralize military power at the court was brought up for reconsideration in questions asking to what extent field generals should be entrusted with military decisions and what kind of distribution of power would bring about the best results in moments of crisis. This was a question that was also hotly debated at court in the thirteenth century, and in the final decades the court moved toward a policy of greater autonomy for military commanders; in keeping with early Song precedent, however, the Tang system of regional commanders was never implemented.20 3. Funding was a key concern and one that reflected the very real pressure that military provisions put upon the imperial budget. 4. The role of geographic considerations in strategic planning had long been a theoretical concern for Chinese military thinkers and politicians and continued to be a topic of both theoretical and applied interest. In the latter category, questions about 20. Huang Kuangchong, “Wan Song chaochen dui guoshi de zhengyi,” 110–12.

178

margins, borders, and frontiers

the relative importance of and the relationship between border regions in defensive and offensive planning were of particular interest. Wu Yong’s concluding remarks quoted earlier, for example, encouraged examinees to compare the relative mil­itary strength of the four regions of greatest concern to the Southern Song state (Sichuan, the Middle Yangzi, the Lower Yangzi, and the Huai River region). The preoccu­pa­ tion with the new frontier regions was not shared by all, and some examiners noted that the role of the capitals in military planning should be reconsidered in view of the gradual demotion of capital defense since the heyday of the Northern Song, when Kaifeng was home to hundreds of thousands of soldiers.21 5. Finally, interstate relations within the broader East Asian region were also a topic deemed fair game for examiners. Questions in this category focused on whether conflicts between other states (such as Xia and Jin, Liao and Jin, or Mongols and Jurchens) could be exploited through Song collaboration with one or the other and thus result in better treaty conditions, territorial gains, or improved security. Occasionally such questions also delved into the difficult matter of the ­reliability of intelligence in making such decisions.

In the eyes of Zeng Jian 曾堅, a thirteenth-century teacher whose examination tips were shared with the broader reading public upon the printing of Dace mijue 答策秘訣 (Secret tricks for responding to policy questions) in the commercial printing center of Jian’an, examination questions not only fell into clearly distinguishable subject categories but also shared questioning strategies. In the late twelfth- and early thirteenth-century examination material he had reviewed, questions on military geography and strategic planning typically referred to events from the Spring and Autumn period, the Three Kingdoms period, and the Six Dynasties.22 This emphasis on the military history of periods of division between 21. Liu Dake, Bishui (Nanjing Library ed.), 73.7a–b. 22. Zeng Jian, Da ce mijue, “xingshi”; for a further analysis of this work, see De Weerdt, Competition over Content, 79–80, 391–92, 422–24.



str ategic discourse

179

southern and northern states coincides with a broader research interest in these topics among Song scholars (see chapter 5) and was prompted by the parallel between conditions then and those following the division of north and south after the 1141 peace treaty between Song and Jin. Zeng Jian’s comments further suggest that questions about military planning could best be answered with some reference to historical geography. Whereas he recommended that students refer back to the long-held truth that successful reunification campaigns started from either Shandong or Guanzhong, both areas in the north, it appears that by the twelfth and thirteenth centuries an examination of the new southern frontier regions and their proven and current potential for defensive and offensive planning had become a more attractive prospect for Southern Song scholars and officials.

Milita ry Geogr a ph y The increased production and circulation of geographic knowledge in the public domain were reflected in greater attention to the subject in policy questions and essays. The presence of historical atlases such as Handy Maps in a prefectural school library and the claim in one of its prefaces that this was essential reading for examination candidates may at first appear puzzling; there is no evidence that suggests that map skills were tested on the examinations. Even though candidates were not asked to draw maps or identify places on a blind map, questions and essays composed after the Jurchen occupation of the north demonstrate that a lack of basic geographic knowledge of the empire could cost candidates dearly. Below I examine how the geography of border regions as well as the military geography of the interior were addressed in questions and essays. Geographical questions tended to focus on the new frontier zone along the Yangzi and Huai Rivers. Such interest was strategic: it was born out of the recognition that the territory occupied by the Song did not reach to the boundaries of the normative “Chinese Empire.” Territory had become a core element in the definition of the “Chinese Empire”:

180

margins, borders, and frontiers

The northern boundaries of our dynastic founders extended north to Yan. Nowadays, they extend north to the Huai. The western boundaries of our dynastic founders extended west to Xia. Nowadays, they extend west to the Han. The boundaries of our dynastic founders can truly be considered boundaries. How can our boundaries be considered boundaries!23

A more detailed knowledge of the geography of the new frontier region was required to plan for defense and recovery. Sizing up the discrepancy between past and present, discussing the hierarchy of strategic places, and planning possible routes of advance were all aspects of the broader discourse on the geography of empire. For example, in a question challenging conventional wisdom that in military planning primacy was always given to the inherent strategic advantages of natural features, Wei Liaoweng challenged his students to consider whether the Southern Song court could continue to rely on the natural advantages offered by riverine boundaries or whether it should map out a plan for reconquest: Ever since the imperial carriage went south, the extent of the territory registered officially reaches from Mingzhou and Yue in the east, Qiongzhou and Yaizhou in the south, Min and Bo Mountain in the west, to the Huai and Han Rivers in the north. This equals only about three parts of the twelve regions demarcated by Shun, four parts of the thirteen occupied by the Han, and seven parts of the nineteen prefectures of Jin. A state like this is truly at peril. So some opinion makers say that the shame has to be washed clean and the rightful status of each has to be made clear. However, based on a con­sideration of strength and weakness and advantage and disadvantage, there may be regret when we move. Is this truly so? On land we have important passes, and as far as waterways are concerned we hold the Yangzi River. North of the Yangzi River there are, furthermore, the Huai and the Han Rivers. These are all features by which Heaven has demarcated north and south. If we now do not maintain these and pursue the enemy between the two rivers, then it is uncertain 23. Liu Dake, Bishui (Nanjing Library ed.), 72.3a–b.



str ategic discourse

181

whether we will win or lose. Is it the case that men of the past also focused on geographic advantage when discussing state affairs? In recent years the enemy has crossed the trenches, intruded into our territory north of the Yangzi, spied into our territory south of the Han River, and they have also stolen strategic places in He, Cheng, Feng, Liang, Yang, and Da’an in the west. Can we then still rely on geographic advantage?24

Beyond the prefectures, rivers, and mountains that defined the contours of Southern Song territory, Wei Liaoweng implicitly expected some familiarity with border counties and prefectures that had been affected by Mongol intrusions in the west. His question furthermore implies that basic routes of incursion were also part of the kind of knowledge of military geography that could be expected of those of minor military rank and that, by extension, the literati should possess: If we cannot but go for war, then from Huai and Cai we can get a view of Chen; from the sea we can attack Qing and Qi; from Xiang and Shaan we can move into Xu; from Ru we can move on to Luo; from Song and Guo we can shake Hedong; when leaving Shang from Qinfeng we can encircle Shaanxi. Even a minor officer can sketch a plan like this. However, since the restoration we have held Henan and then lost it; we occupied thirteen prefectures in Shaanxi and then relinquished them.25

Wei Liaoweng hinted that there were issues other than military geographic planning that had to be settled for any campaign to result in more lasting success. It is worth noting, however, that his plotting of prefectures in relation to their strategic importance to entire regions was a widely shared characteristic of Song geographic 24. Wei Liaoweng, Haoshan ji, 93.13a–b; QSW, 310.233–34. For Song-Mongol relations in the 1230s, see Twitchett and Smith, The Cambridge History of China, vol. 5, pt. 1, 863–67. 25. Wei Liaoweng, Haoshan ji, 93.14a; QSW, 310.234.

Map 4.1 Overview of the Southern Song frontier region. The five zones were at times collapsed into three (A, B, C), oftentimes as suggested in the text, by eliminating the coastal zone and either by combining the Huai and Lower Yangzi regions (A) or by dropping the Lower Yangzi.



str ategic discourse

183

thinking. Ni Pu, the lone scholar who drew his own empire map, similarly plotted potential routes of invasion in his letter addressed to Emperor Gaozong.26 Readers of the stele and printed empire maps discussed in chapter 3 would have been able to trace such routes on their maps as those too were reliant on a prefectural model of the empire. The frontier region not only was an object of analysis at the school of former statesman Wei Liaoweng, but also received extensive coverage in commercial examination manuals like The Epitome of Eminent Men. Its compilers approached the military geography of the empire through a survey of each of the five subregions into which the frontier region was customarily subdivided in the twelfth and thirteenth centuries. The essays and additional primary sources provided differing answers to questions about the relative importance of and relationship among each of the regions. Some sketched more detailed textual maps and addressed the strategic hierarchy of places within each subregion. The five subregions were Sichuan (C), the Middle Yangzi (B), the Lower Yangzi (A bottom), the Huai region (A top), and the maritime coast (black) (map 4.1).27 Each region was considered to play a distinctive role in Southern Song border affairs. Some saw the Middle Yangzi as key to the continuity of the Song state; others considered it important but not crucial for securing the core of the Song state in the southeast, especially the capital located in the Lower Yangzi region.28 The Huai region was commonly perceived as the screen that granted or prevented access to the main riverine axis of the Yangzi. The authors of the sample essays included in the examination manual debated whether resources should be devoted to this first line of defense or to the main second line provided by the Yangzi. Other candidates argued that even though the other regions possessed intrinsic strategic 26. Ni Pu, Ni Shiling shu, 6a–8b and passim; QSW, 242.79–80. 27. This map is based on the description of the border regions in Li Zhen, Zhongguo zhengzhi guofang shi, 382–83. 28. Liu Dake, Bishui (Nanjing Library ed.), 73.11a–b. For a summary description of the three zones and an analysis of early thirteenth-century debates about it, see Huang Kuanchong, “Wan Song chaochen dui guoshi de zhengyi,” chap. 3.

184

margins, borders, and frontiers

importance, the maritime route had become increasingly critical. “As soon as Tongzhou turns, we not only have to worry about Weiyang, Zhen and Tai will also be in turmoil. Once Zhen and Tai are in turmoil, on what will Su, Chang, and dependent prefectures depend?”29 Tongzhou 通州 was located at the mouth of the Yangzi; the candidate who wrote the essay from which this quote was taken observed that should the northern armies make their way through Tongzhou Prefecture, the route to key economic centers further down the Yangzi River and just south of it would lie wide open because those were not areas that ranked high on the list of places with a strong military presence. The threat was made real on the grounds that Jin missions had begun to take the maritime route and had been accused of invasions along the coast in the twelfth century.30 Further west, Sichuan also had its advocates for primary importance among the frontier zones. Some noted that the unification and consolidation of the Song Empire in the eleventh century had depended on the pacification of the Sichuan area: Concerning what preparations need to be made for our geographical situation today, what is of the greatest urgency? For the above reasons, I daresay that Shu should be treated with the greatest urgency. Our dynastic forefathers pacified Shu thoroughly. They tended to send the best of their generation to take care of matters there. Zhang Yong 張詠 [946–1015] was sent there twice. Zhenzong said that with Yong governing Shu there would be no trouble to the west. Zhao Bian 趙抃 [1008–84] was sent to Shu three times. Shenzong commented: “I heard you are going to Shu. All will follow and it will be easy to govern there.” It was all like this. It was because they chose Yong for Shu, not because they chose Shu for Yong. It was because they chose Bian for Shu, not because they chose Shu for Bian. Nowadays we again have to take note of the fact that our dynastic founders emphasized Shu, then our geographical position will be solid.31 29. Liu Dake, Bishui (Nanjing Library ed.), 72.8b. 30. Ibid., 74.10a–b. 31. Ibid., 72.9a.



str ategic discourse

185

The writer of this essay not only testified to the importance of the more remote border region of Sichuan but also emphasized that, in military planning, territorial considerations should outweigh other factors such as the personal preferences of generals and high officials. He attributed to earlier emperors the insight that a relatively isolated area like Sichuan needed men with the skills and personalities to address its particular territorial challenges. Opinions were thus divided and shifted over time as to which zones deserved primary attention and therefore the highest investment in fortification and troops. Despite differences in emphasis, however, many thought of the entire frontier region as a connected structure. The importance of particular regions was typically based on their relationship to other regions. The coastal zone’s direct connection to the lower Yangzi region and to the economically most advanced areas in Liangzhe elevated its importance in the eyes of some.32 In a decree examination question Hong Zikui 洪咨夔 (1176–1236) emphasized both the centrality of the JingXiang (Middle Yangzi) region as well as its reliance upon Sichuan and the Huai region: “As for the major strategic areas in the empire, we take Shu [Sichuan] as the head and the Huai [region] as the tail, but the waist and back is Jing-Xiang.”33 The waist and back metaphorically stood for the area most critical to the coherence of 32. Ibid., 74.10a–b. 33. Hong Zikui, Pingzhai ji, 9.3b; QSW, 307.195. For further discussion of this view and work on the Southern Song border in the Sichuan area, see He Yuhong, “Jingshi yishi yu Nan Song Chuan-Shaan bianfang shidi lei wenxian,” esp. 30–31. In the materials quoted in the section under Jing-Xiang the compilers of Bishui qun­ ying further explained to their readers that arguments advocating the importance (but not the primacy) of this region in frontier defense foregrounded its position within a broader context (mailuo 脈絡); arguments emphasizing that securing Jing-Xiang was sufficient for securing the empire, on the other hand, were based on its inherent strategic position (xingshi 形勢). As an example of the former, they quoted the military officer Liu Ziyu 劉子羽 (1096–1146), who had made the case for the importance of the Middle Yangzi region in the early Southern Song period on the grounds that it provided access to Sichuan and Shaanxi in the west (which constituted Song border regions in the eleventh century, see chapter 5), the Han and Mian Rivers (which flowed into the northern territories and were thus central arteries in communications between north and south), and the capital region in the east. Liu Dake, Bishui (Nanjing Library ed.), 73.11a–13b.

186

margins, borders, and frontiers

a multipart whole. The entire zone was referred to as “the three borders” (sanbian 三邊) and, as noted by an anonymous candidate, none could be considered to the exclusion of the others; the interrelatedness of these three border regions in the security of the Song state was from this point of view also the reason why the military history and geography of the Six Dynasties was of questionable use to Song statesmen.34 As none of the Six Dynasties had ruled over the extent of territory under Southern Song control, a critical analysis of the strategic choices their ruling elites had made was of limited use. Absent from the geography of the new frontier was the capital itself. The relative insignificance of the new capital, Lin’an [Hangzhou], in the mapping of strategic places constituted a major shift. Kaifeng had been selected as the new capital in part because of its large military population in the tenth century; its military population further increased as the first Song emperors concentrated their efforts at military centralization in the capital itself. Even after Emperor Shenzong reduced the number of military encampments in the city in the second half of the eleventh century, the essential role of Kaifeng and the subsidiary capitals in the defense of the political core of the empire was still a recognized principle. By the thirteenth century it had become clear that the capital’s role in military defense had declined precipitously. Some candidates tried to reverse the trend by pointing out that even though the border regions were strategically important, the capital constituted the very foundation of the empire. Its defensibility had been adversely affected by the policy to pad the border with imperial troops and to empty out the interior.35 Just as the border regions were conceived of as interlinked parts in a larger body, natural features, prefectures, and garrisons within each region were also plotted in an interlinked strategic hierarchy. One candidate, for example, argued that in military preparation the 34. Liu Dake, Bishui (Nanjing Library ed.), 73.6a–b. 35. Ibid., 73.7a–b. For a discussion of the prominence of the military in Kaifeng, see Kubota Kazuo, Songdai Kaifeng yanjiu, esp. chaps. 1–3, 8.



str ategic discourse

UPPER YA N G Z I

CORE DEFENSE ZONES

187

MAIN STRONGHOLDS

LOWER HAN RIVER QINLING M O U N TAI N S

F R O N T: FA N C H E N G X I A N G YA N G

BACK: JINGMEN

SANGUAN UPPER HAN RIVER ZAOJIAO

DOWNSTREAM

ALL PREFECTURES

MAIN STRONGHOLDS Sizhou

Right: Huainan East

Xuyi

Tianchang

Shouzhou

Huai Region

Caizhou Left: Huainan West

Xizhou

Fuguang

Huangpi

Fig. 4.1 The hierarchy of strategic places in the Upper Yangzi and Huai regions.

Upper Yangzi and Lower Yangzi/Huai regions should come first. These regions could be best defended if resources were distributed with the following schema in mind (fig. 4.1).36 Each riverine subregion was conceived of as a frontier zone with one or two main strongholds (passes, garrison towns, counties, or military prefectures) that could be supported by and provided protection for one or two subsidiary strongholds in the same subregion. This made for 36. Liu Dake, Bishui (Nanjing Library ed.), 73.4a–5a.

188

margins, borders, and frontiers

a relatively sparse network of strongholds, which may have also rendered the frontier porous due to insufficient connectivity across the mountainous and riverine frontier. The focus on garrison towns, counties, and prefectures and the hierarchical relationship among them in a defense network is broadly representative of military strategic thinking among Song literati and contrasts with the higher resolution found in the work of military experts like Hua Yue to be discussed in the next chapter. Regardless of the difference in resolution in discussions of the frontier zone, essays on the military geography of the frontier were unanimous in their insistence on the ultimate objective of territorial reunification. Natural features were selected not only for their imputed role in military defense against Jurchens and Mongols but also for their ability to project Song power into the northern territories, hence the focus on the Qinling Mountains, the Han River, and the Huai border in the example given above and in Wei Liao­weng’s questions quoted earlier. In sum, identification with the full extent of the normative “Chinese Empire” had become a widely shared feature of literati identity in the twelfth and thirteenth centuries. Shared control over the Chinese territories was experienced as a shame to be borne and washed away by all. The incorporation of military geography and border affairs in examination curricula conflicted with the court’s insistence on secrecy on these matters but corresponded with a need felt by scholars and officials at all levels to be better informed about the territorial problem at the core of the Southern Song state: Are they [those who focus on only fortifying and defending the Huai and Middle Yangzi regions] not ignoring that the land of Qi [northeast] is also our land! And that the territory of Zhou [central plains] is also our territory! . . . Let us not consider today’s boundaries the boundaries; let us take our founding fathers’ boundaries as the boundaries. Then, when the scope is vast and the institutional framework broad, our policy for military defense will be complete.37 37. Ibid., 72.3b.



str ategic discourse

189

Histor ic a l R e asoning Ours is not a history course—and so to readers we repeat that this book is not a history book. It is, in fact, not even about the stories it tells. It is about how to use experience, whether remote or recent, in the pro­ cess of deciding what to do today about the prospect for tomorrow.38

The new geography of the Southern Song border was for Song literati closely connected to the military and political history of the places it encompassed. They were trained to address policy questions through historical reasoning. Diplomatic and strategic questions were no exception. How was past experience used to discuss contemporary geopolitical problems? Was the Southern Song condition considered analogous to past geopolitical crises? The uses of the past illustrated in twelfth- and thirteenthcentury examination manuals remind one of the uses of the past modeled in Thinking in Time. This book, from whose preface the epigraph to this section was drawn, grew out of a course on the uses of history intended for future political practitioners in training at the John F. Kennedy School of Government. Written in response to the still relevant feeling that American politicians had, since the 1970s, turned into “a host of people who did not know any history to speak of and were unaware of suffering any lack, who thought the world was new and all its problems fresh,”39 its mission to make future administrators analyze and reimagine history for practical purposes corresponds closely to that of imperial Chinese historical and administrative textbooks between the eleventh and thirteenth centuries. Thinking in time was a prerequisite for official political discourse in imperial China and for those who were training to master it and demonstrate their mastery of it in examination essays. The cultural impact of the examinations extended well beyond the examination halls, and thinking in time also became a cultural skill, 38. Neustadt and May, Thinking in Time, xxii. 39. Ibid., xi.

190

margins, borders, and frontiers

practiced in social interaction with similarly motivated elites (see chapter 7). Song students acquired skills in historical reasoning while learning to compose policy essays required in the examinations. The typical policy essay was structured around the discussion of relevant cases from both preimperial and prior dynastic history, from the history of the Song dynasty itself, and from contemporary authors, in order to construct a proposal relevant to the question posed. Moreover, examination manuals included under each heading, besides sample essays, a section on primary sources. The primary source selections were organized according to the structure of evidence required in the policy essay moving from classical words to abide by ( jingzhuan geyan 經傳格言) and facts from successive dynasties (lidai shishi 歷代事實), to more recent types of materials under events from “our dynasty” (huangchao dianzhang 皇朝典章), key proposals from former righteous men (xianzheng lunjian 先正 論建), and selections from collected writings (wenji jinghua 文集菁 華). The addition of the collected writings shows that beyond the official record political literacy also assumed a familiarity with recent discourse on administrative issues.40 What were the uses of the past in political discourse that abided by these rules? On the basis of a broad exposure to examination essays and classified historical collections, the late Robert Hartwell distinguished between different attitudes toward the past in Song political discourse. Whereas classicism (the imitation of past models) and moral didacticism (the evaluation of events and personalities on the basis of moral criteria) were traditional frameworks within which to cast policy proposals, historical analogism (“the 40. This description is based on the presentation of model essays and primary sources in Liu Dake, Bishui (Nanjing Library ed.). The typical policy essay should, according to a thirteenth-century examination manual, exhibit the following structure: (1) the establishment of the basic argument to start off (liyi faduan 立意發端); (2) an examination of the remote past to make the argument grander ( jigu weiyi 稽古偉議); (3) the use of Song models to construct a sound plan ( fazu jiayou 法祖嘉猷); (4) striking passages from contemporary prose (shiwen jingduan 時文警段); (5) parallel phrases (qiyu pianzhu 綺語駢珠); (6) the proposal of policy advice for the immediate present (dangjin xiance 當今獻策); (7) a conclusion that adds meaning (shengyi shoujie 生意收結).



str ategic discourse

191

comparative study of similar historical phenomena” as a guide for contemporary policy) had become a prevalent mode for organizing and interpreting the past.41 In Hartwell’s analysis, the collection and analysis of historical events analogous to present conditions served at least three purposes. Past events provided direct models of success and failure. At a more analytical level, past facts and events could be aggregated quantitatively to establish long-term trends. Last, the repeated observation of patterns in distinct events could be used to establish functional relationships among categories of phenomena.42 A core principle of this mode of historical-cumadministrative analysis is the dual regard for general principles of history and administration and for their modification in specific circumstances. The features Hartwell attributed to this kind of historical analysis can be readily spotted in the essays and archival materials on border affairs in twelfth- and thirteenth-century examination manuals. As an example of the use of quantitative data, crossdynastic and dynastic trends in the size of the imperial troops were used to question the criteria by which effective military power should be measured.43 The review of historical cases, encouraged in the format of the policy questions and essay, also fostered the detection of broader principles and their application to current policy problems. In an essay that reviewed the peace settlements of the 1120s, 1140s, and 1200s the author concluded based on the different outcomes in each case that peace can only obtain when the negotiating parties are equally strong or equally weak.44 The treaty with Jin signed under Emperor Gaozong resulted in short-term benefits as both parties then pursued peace from a position of strength. Similarly, in the candidate’s reading, the treaty signed after the Song invasions into Jin territory in the early thirteenth century ended in short-term 41. Hartwell, “Historical Analogism, Public Policy, and Social Science,” 694. 42. Ibid., 719–22. 43. E.g., Wu Yong, Haolin ji, 33.2b; QSW, 316.335; De Weerdt, Competition over Content, 101, 145–48. 44. Liu Dake, Bishui (Nanjing Library ed.), 75.15a–16b.

192

margins, borders, and frontiers

stability as both parties came to the negotiations in comparably weak positions. The debacle in Song military and diplomatic history in 1126–27, conversely, resulted from the Song court’s willingness to settle for peace at a time when it faced a far stronger enemy. If this principle that peace requires a balance of power between the negotiating powers held true, the author cautioned, the current government, which was in his estimation neither as strong as in the 1140s nor as weak as in the 1200s, should be careful to heed the lesson of 1126–27 and thus not settle for peace (presumably with the Mongols, as this essay was most likely written in the 1230s or 1240s). On this basis, he argued, envoys should not be sent and the military should be mobilized instead. This essay illustrates very effectively how a series of cases could be used to induce a general principle for interstate relations with immediate application to a current policy problem: should envoys be sent to the Mongols or not? The author refrains from the moral didactic paradigm that could have been easily applied in this case. After the death of Qin Gui 秦檜 (1090–1155), the architect of the 1141 peace treaty with Jin, which consolidated the Jin hold on the northern territories, his lack of moral character and poor treatment of meritorious generals were widely blamed for the loss of the north. Qin Gui is here not depicted as a villain but appears, implicitly, to have been justified in having opted for negotiations in the late 1130s and early 1140s.45 At a basic level, policy questions raised cases as examples of different policy options and invited examinees to explain the rationale behind failure or success. In a question of the “grand strategy” type, inquiring about the pros and cons of peace or war with Jin, for example, Yuan Xie asked examinees to consider cases that may have appeared difficult to reconcile: Why did the fourth-century state of Jin 晉 never accept any overtures from Wei 魏 and go to the extreme of burning the latter’s proposals? On the other hand, why did the Tang emperor Taizong, whose military skill received high acclaim, humble himself and make peace with the Turks? How did the 1005 45. On the representation of Qin Gui in Song sources, see Hartman, “The Making of a Villain,” and Hartman and Li, “A Newly Discovered Inscription by Qin Gui.”



str ategic discourse

193

treaty of Shanyuan come about? Was it the result of Song unwillingness to lower itself to fight the Liao and was it signed with a view to protect the northern population? Conversely, was the peace treaty concluded in 1141 a mistake by which Grand Councilor Qin Gui lost the north? Why did Emperor Xiaozong in 1164 go back to the exchange of envoys between the two powers after an earlier promise to avenge the shame Song had incurred under his father’s rule? Was this choice in the end not clear evidence of the efficacy of the policy of appeasement? Readers were asked to examine these examples in light of what the examiner believed to be the guiding principle in relationships between Chinese and non-Chinese in antiquity: noninteraction. They were in the end confronted with the urgent question whether the Song court, faced with invasions from the north, should sue for peace or whether it should go to war. Could it accept escalating treaty demands, or could it muster the resources for war? Could it use peace as an expedient and so gain time for self-strengthening? How could self-strengthening measures be undertaken without raising the enemy’s suspicions?46 Policy questions on border affairs thus mirrored court debates. Even when the slant of the examiner’s questions suggested which line of argument he preferred (most likely self-strengthening in preparation for war, in Yuan Xie’s case),47 preparation for such questions necessitated familiarity with different points of view and with the cases used by all sides. Manuals and annotated compilations of historical and archival materials were intended to satisfy this need. As is evident from these examples, the past in imperial Chinese political discourse was layered. Cases were to be drawn from clearly dif­ferentiated periods in Chinese history. As in the transhistorical empire maps discussed in chapter 3, and as in Yuan Xie’s question, each piece of information is located in antiquity or in dynastic time. The first serves as a knowledge base from which to construct normative frameworks, and later dynastic and Song cases provide realworld experience from which to formulate contemporary policy. Layering in historical and administrative reasoning deserves our 46. Yuan Xie, Jiezhai ji, 6.10a–11b. 47. Huang Kuanchong, “Wan Song chaochen dui guoshi de zhengyi,” 16, 79.

194

margins, borders, and frontiers

attention, not only because it runs counter to the supposition that imperial Chinese uses of the past are “premodern” and thus postulate convergence between past and present but also because it allows us to see that the past could be used to multiple effect. Examples drawn from different temporal layers could serve different purposes within the overall argument. Primary source material in pre-Song imperial history could provide parallels or contrasts with contemporary experience, whereas material drawn from the textual legacy of antiquity or foundational periods in the more recent Song past could add legitimacy to an argument. Some examples illustrate this. As Zeng Jian noted, Six Dynasties events were often cited for further analysis in questions on Southern Song border defense. The interstate system of the fifth and sixth centuries and the geographic location of the southern dynasties turned this period into a favorite object of study for Song historians and policy makers who found themselves similarly facing empires ruled by non-Chinese elites to the north.48 The military and diplomatic history of the period and Six Dynasties debates about strategy and military geography were examined closely as the analogy with Song might give this history the strongest predictive power. Nevertheless, many questions and essays questioned the utility of the parallel. Six Dynasties examples were often invoked to establish the uniqueness of the present situation of the Song state. In a model examination essay selected in The Epitome of Eminent Men one author argued against the conclusion that the suitability of the Huai area as the focal point of the Song defense was evidenced in Sun Quan’s 孫權 (181–252) ability to rely on it to establish the first of the Six Dynasties.49 Sun Quan ruled the independent state of Wu 吳 between 222 and 252. The author argued that present local conditions should determine the focus of the Song military and not 48. For a summary list of such titles, see He Yuhong, “Jingshi yishi yu Nan Song Chuan-Shaan bianfang shidi lei wenxian,” 38–39, 42–43. See also Chen Aiping, “Nan Song dui Liuchao nan-bei junshi duiyi jingyan de lilun yanjiu.” 49. Liu Dake, Bishui (Nanjing Library ed.), 73.5b–7a. In another question Cheng Bi noted that in identifying strategic places along the Huai and Yangzi Rivers great care should be exercised in using Six Dynasties discussions as they lacked necessary detail. Cheng Bi, Mingshui ji, 5.27b–29a; QSW, 298.41–42.



str ategic discourse

195

past or theoretical characteristics associated with particular places. The parallel between the two was also unsuitable because the appropriate base of comparison should be empires at their high point rather than small states in dire circumstances. Others contended similarly that the differences between the two periods were so substantial that this legacy was of little help in formulating policy. Another anonymous candidate argued that Song resources in the south were vastly superior to those available centuries earlier; its border strategy could thus not be held to the same constraints.50 In contrast to the mixed utility of the experience of prior dynasties, some parts of the Song past had become standard reference points. This was particularly true of the early Song emperors (Taizu and Taizong) and of the first two emperors ruling from the southern capital of Lin’an (Hangzhou). As the founders and revivers of the Song house they had set out the basic parameters for subsequent policy; the effective use of the founders’ precedent legitimated policy proposals. In a model essay calling for a resolute and consistent policy against the Mongols, the following early Southern Song precedents were adduced as relevant source material: Gaozong said: “The foreigners have already retreated. We now have to start gradually planning for the recovery [of the north].” The Emperor [Xiaozong] said: “The scholar-officials no longer dare to discuss the restoration. They do not realize that their home originally covered one hundred mu, fifty mu of which was forcefully occupied by others. Would they not register their case and demand it back?”51

Such pronouncements by the two emperors who laid the foundations for the continuity of the Song dynasty in the south were marshaled in support of a more aggressive policy toward the north in the thirteenth century. Emperor Xiaozong’s pronouncement in particular was a strong indictment against those among the literati 50. Liu Dake, Bishui (Nanjing Library ed.), 73.2b–3a. For a different example, ibid., 72.6a–7a. 51. Ibid., 75.14a–15a.

196

margins, borders, and frontiers

who failed to see the empire as their home (see chapter 3) and resist occupation. The compiler’s reference to the source from which these two quotations were taken allows us to trace how such materials may have been selected and how their import was changed in the process of historical layering. The interlinear notes indicate that both passages came from Zhongxing liangchao shengzheng 中興兩朝聖政 (The sagely government of the two reigns of the intermediate restoration) (fig. 4.2). The sagely government collections were abbreviated editions of court archival collections and as such were a desirable shortcut to the Song political record (see chapter 1). The collection of important political events in the reign of the first two Southern Song emperors appeared to have been of particular interest in the thirteenth century, as the survival of an annotated commercial edition, Zengru mingru jiangyi huang Song Zhongxing liangchao shengzheng 增入名儒講義皇宋中興兩朝聖政 (The sagely government of the two reigns of the intermediate restoration, with explanations by famous scholars), printed in Jianning, suggests. Two of the passages marked in figure 4.2 can indeed be traced back to this edition (fig. 4.3); one could not be located, most likely because the volume for the year in which it occurred did not survive.52 Locating passages relevant to the topic of “recovery” or “reunification” (shoufu 收復, also huifu 恢復), the subheading under which the model essay was classified, would have been relatively straightforward. As shown in figure 4.4, the commercial edition of Liang­ chao shengzheng featured a chronological index that referred the reader to the years and months covered in each chapter (starting with Jianyan 1 [1127] and ending with Chunxi 15 [1188]) and a topical index that noted underneath the brief titles for the entries listed under each main heading the year under which it could be found in the main text. To further assist the reader in locating the relevant passage in the corresponding chapter, the publisher provided block titles in the upper margins which corresponded with the entry 52. The chapters for the year quoted for Xiaozong’s statement, Longxing 1 (1163), ( juan 42–43) do not survive, but, under the relevant rubric “huifu,” the index includes some references for this year.



str ategic discourse

197

titles listed in the classified table of contents. This process is illustrated in figure 4.4. There are minor variations in wording and the original passages are slightly longer but by and large it appears that the compiler of the examination manual may have had this or a comparable edition of The Sagely Government of the Two Reigns at hand. The recurring references to this collection in The Epitome of Eminent Men and comparable references to the Gaozong and Xiaozong reigns in thirteenthcentury policy questions help explain the context for the commercial production of the annotated edition of The Sagely Government of the Two Reigns in the thirteenth century. By the turn of the thirteenth century the reigns of the first two Southern Song emperors had become standard reference points for students and politicians just as the early reigns of the Northern Song period had become paradigmatic in the course of the eleventh century.53 There is, however, a significant difference between the archival collection and the archival sections in examination manuals and administrative encyclopedias in the representation of classified archival material. In their selection of material from primary archives, manuals reduce the diversity and contradictory nature of the primary archive. When checking the “reunification” section in the classified index in The Sagely Government of the Two Reigns, the reader could find a broad range of relevant events and imperial pronouncements on the issue. Given the chronological arrangement of the entries, the reader could also detect shifts in the position of both the emperors and their advisors over time. The corresponding section in the examination encyclopedia focused on those passages that supported the case for a more aggressive stance toward northern polities. Changes in attitude or comprehensive assessments of an entire reign could typically not be induced from the available primary source selections. This suggests that in the process of historical layering cases served goals other than the analysis of patterns and trends. The emphasis on the reigns of founding emperors such as Gaozong and Xiaozong illustrates not only the interest in recent Song history among students and politicians but also the use of the pronouncements and decisions of the 53. Deng Xiaonan, Zuzong zhi fa.

C

D B

A

D

Figs. 4.2A–B The subheading “Reunification” under the subject heading “Dealing with non-Chinese” in Liu Dake’s Bishui qunying. The markup highlights the end of a passage (A) on Gaozong’s and Xiaozong’s interest in the recovery of the north in a sample examination essay, source materials consisting of selected quotations from Liangchao shengzheng (B and C) (see Zengru mingru jiangyi huang Song Zhong­ xing liangchao shengzheng), and the interlinear source citations (D) (Liu Dake, Bishui qunying [Nanjing Library edition], 75.14a–15a).

Figs. 4.3A–B Passages in Liangchao shengzheng corresponding to those quoted in Bishui qunying. From Zengru mingru jiangyi huang Song Zhongxing liangchao shengzheng (National Central Library edition), 17.3a; 57.13a.

C

A

Main Text

C

B

Topical Index

D

Fig. 4.4 Relationship between the main text and the chronological and topical indexes in Liangchao shengzheng. The entry “Do not forget reunification” (B) under the subheading “Reunification” (D) in the topical index is shown in the top margin of the main passage. “Six” underneath the entry title in the topical index (C) refers to the sixth year in the Chunxi reign period (also noted in the chronological index); the corresponding juan number (A) can be looked up in the chronological index and then located quickly using the short-title reference in the margin of the page on which the main text appears. From Zengru mingru jiangyi huang Song Zhongxing liangchao shengzheng.

A

Chronological Index

B



str ategic discourse

201

founding emperors to legitimate the proposals advocated in their policy essays and memorials. In sum, in policy essay writing literati were trained to apply historical reasoning to contemporary policy questions on border affairs. Students were taught to think and construct arguments through the quotation and analysis of primary sources drawn from distinct periods in Chinese history including the recent past and contemporary policy debates conducted in official circles or privately. Cases were selected according to a variety of criteria. Comparability to present conditions was one of them, as illustrated in the preference given to the military history of periods of division in thirteenthcentury questions and essays. The above review of the cases cited in questions and essays on military and border affairs suggests that, besides the testing of universal principles and the analysis of historical trends, cases were also used to differentiate the past from the present and to legitimate present policy options through the selective interpretation of authoritative layers of the past. The application of historical layering in Song political discourse ensured that individual cases or sets of cases could be adduced to multiple effect.

Mapmaking in Twelfth- and ThirteenthCentury Song China Twelfth-century literati were not only responding to questions about historical and military geography in mock and real examination questions; private scholars also dedicated themselves to the compilation of geographical data and to drawing their own maps. Who were these private scholars? Why did they compile their own historical and military geographies, and how did such work fit in with their careers and other interests? How did their goals and methods compare to those of the court compilers who had monopolized this work in the past? Several among these amateur cartographers were “plain-cloth” commoners who obtained neither degree nor office. Those who found patrons interested in their work and in presenting it to the court were on occasion given honorary titles. Others obtained

202

margins, borders, and frontiers

examination success but saw their careers halted for personal reasons or because they lacked the requisite recommendations. One of the better, yet still poorly, documented examples in this group was Ni Pu 倪樸. According to his biographer Wu Shidao 吴師道 (1283–1344), a scholar who took great interest in men who had in his estimation shown practical ability in the service of society, Ni Pu was a man of singular ambition who had his mind set on useful knowledge.54 And the most useful to the fledgling Southern Song was, in Ni Pu’s mind, the geography of the Chinese territories. For Ni Pu and for many others who took up geography and cartography in the twelfth and thirteenth centuries, these fields of knowledge were of the greatest use in planning for the preservation of the Song state as well as for the greater goal of the reunification with the north. Ni Pu’s dedication to the latter cause may have predated his immersion in geography. In the late 1150s or early 1160s and as a commoner he addressed a letter to Emperor Gaozong outlining the reasons and necessary preparations for an immediate restoration campaign.55 When this policy proposal did not produce the desired effect, he set out to work on Yudi huiyuan zhi 輿地㑹元志 (Gazetteer bringing the empire together), a compilation that eventually grew to be forty juan long. Now no longer extant, it was reported to be a compilation of “the things that those deciding on military affairs should know, such as the mountains, rivers, and strategic places in the empire, and population numbers.”56 In Ni Pu’s own account, this work was the result of years of scouring through historical records and an attempt to trace, methodically and comprehensively, changes in administrative nomenclature and the location of administrative places throughout time, noting features such as passes, fords, places where battles were fought and treaties concluded, and other historical 54. Wu Shidao, Jing xiang lu, 6.1a–b. 55. Ni Pu, Ni Shiling shu, 1a–20a. Ni Pu writes in this letter that it has been thirty years since the demise of the old capital/central plains. Since Kaifeng fell in 1127, one might conclude that this letter was written in the late 1150s or early 1160s (ibid., 18a). Elsewhere he writes that it has been twenty to thirty years since a peace settlement was reached between Song and Jin (ibid., 5b). If one takes this to refer to the 1141 treaty, then the 1160s is the most likely time frame for this letter. QSW, 242.76–87. 56. Song Lian, Puyang renwu ji, xia, 10a–13a, esp. 11a.



str ategic discourse

203

sites. As in other work of this nature, all historical places were correlated to present place-names; references to primary sources were provided to document all conclusions.57 This research was also the basis for an empire map through which Ni Pu sought the patronage of prominent scholars in his hometown, Wuzhou 婺州, a prefecture known for practical scholarship that came mostly in the form of histories of institutions and administrative theory. The map also became the vehicle through which Ni Pu sought support beyond Wuzhou. In a letter seeking patronage from Zheng Boxiong 鄭伯熊 ( j. 1145), who served as prefect in Wuzhou about 1175–76, Ni Pu wrote that he had compiled a transhistorical map of the Chinese and non-Chinese that measured over one zhang (approx. 3–4 meters). Its length allegedly rendered all features more clearly visible than on comparable survey maps of the empire. He added that this work had already received local support from the Pan family, most likely Pan Jingxian 潘景憲 (1134–90) and his father Pan Haogu 潘好古 (1101–70), who had Ni’s larger map transferred to a screen for easier consultation. Ni Pu borrowed this map and offered it to Zheng Boxiong for his perusal in the hope that it would “show to you my dedication,” and that “I am solely engaged in useful scholarship.”58 Zheng Boxiong was a reputable scholar from Wenzhou, a prefecture that had, like Wuzhou, obtained a reputation as a center of intellectual activity. His political and intellectual connections could have proven very useful to the ambitious Ni Pu. Ni Pu presented his work as qualitatively different from the types of geographical knowledge possessed by those who had received a general education. His letter was an open defense for expert knowledge. He wrote that the many years he had dedicated to this project, more than ten up to the time of writing, had turned him into an expert on the subject. Attempts to share his work with peers had been unsuccessful; most, he claimed, were in pursuit of the kind of general knowledge required for the examinations and preparation for general 57. Ni Pu, Ni Shiling shu, 20a–22b, esp. 21a–b. QSW, 242.88–90. 58. Ni Pu, Ni Shiling shu, 22b. QSW, 242.90. For more on Wenzhou and Zheng Boxiong, see De Weerdt, Competition over Content, 46–50, 113, 168; Wang Yu, Yongjia xue­pai yu Wenzhou quyu wenhua, esp. 144–49.

204

margins, borders, and frontiers

administration, and those who understood the value of his work did not have time to engage with it. Ni Pu tried to persuade those from whom he sought support by claiming that expert knowledge had been the basis for civilized society since antiquity. In the past, knowledge essential to the government and defense of civilized society such as archery, charioteering, and calendrical sciences resulted from the specialization, “the dedication to one art” ( ye yiyi 業一藝), of culture heroes such as Yi 羿, Wang Liang 王良, or Xihe 羲和. In Ni Pu’s view, specialization in historical and military geography occupied a similar place in the preservation of civilized society in the mid-twelfth century. His own work on the geography of Chinese empires was the result of specialization over a long period of time and would become foundational for strategic planning. Such knowledge, acquired over the course of a lifetime, was conventionally transmitted within the family, from father to son. Ni Pu, who according to Song Lian was a rather difficult and stubborn personality who remained single well into adulthood, instead proposed more contemporary channels of transmission. By presenting his work to local prefects, such as Zheng Boxiong and Zhou Kui 周葵 (1098– 1174; prefect of Wuzhou 1160–62), and scholars who had gained a reputation for their interest in applied historical and geographical scholarship, such as Chen Liang 陳亮 (1143–94) and Pan Jingxian, he hoped to find support for its completion and publication. The claims made for the originality of and general lack of interest in his work were most likely not unusual in letters seeking the patronage and/or recommendation of men in office. Ni Pu’s Yuan and Ming biographers were keen on turning him into a rare exemplar who sought useful knowledge to preserve and unify the Chinese territories and who was misunderstood by a complacent court and sycophantic contemporaries, and thus read Ni Pu’s claims literally. Song Lian dramatized Ni Pu’s work by depicting him in his biography at work in his studio: “He hung the map on the wall, his finger pointed and his mind calculated which places could be used for battle and which cities could be used for self-defense.”59 59. Song Lian, Puyang renwu ji, xia, 11b.



str ategic discourse

205

However, Ni Pu’s search for a complete historical geography of Chinese empires and his interest in mapping were shared among a number of scholars. The compiler of Handy Maps, Shui Anli, who is virtually unknown to us, may have been similarly motivated. In the mid-1140s Wu Xie 呉澥, a commoner from Fuzhou 撫州, submitted his Yunei bian lidai jiangyu zhi 宇内辨歴代疆域志 (Gazetteer marking our territories through the ages). Around the same time, in 1146, Yu Zhe 余嚞 submitted a work listing the natural features, products, customs, and strategic places of the empire in the order of Song county and prefectural names.60 This may be the same Yu Zhe who obtained a jinshi degree in 1184 and gained notoriety as a prefectural school teacher for submitting a memorial requesting the execution of Zhu Xi 朱熹 (1130–1200). If so, this work was submitted well before the author had obtained official recognition and was part of a large body of scholarship that Yu Zhe produced in such fields as historical geography, astronomy, classical exegesis, and Song institutional history. Other examples include Wang Xixian 王希先 and his father Wang Xia 王珨, whose Huangchao fangyu zhi 皇朝方域志 (Gazetteer of our dynasty’s territory) was regarded as “the most detailed of administrative geographies” by Chen Zhensun.61 Wang Xia had started the work, and his son, who moved to Wuzhou where Ni Pu had been active, completed it. According to Chen, this work was divided into two parts: the lists ( pu 譜) gave an overview of past facts (shishi 事實) and correlated them with present place names; the gazetteer proper (zhi 志) provided a description of contemporary administrative organizations but linked them to the spatial divisions of the past. No further information about this source is available, but Chen Zhensun’s brief description suggests that by means of these lists major political and military events or classified facts62 could be placed on a contemporary map because the Wangs had provided the present 60. Wang Yinglin, Yuhai, 15.45a. 61. ZZSL, 8.241. 62. This interpretation is in part based on the use of comparable divisions in the presentation of historical geographical data. In Diwang jingshi tupu Tang Zhongyou provided both maps and lists ( pu) for the analysis of historical geographic information. The lists consist of tables in which all regions are described on the basis of a set of topics ranging

206

margins, borders, and frontiers

coordinates for them. In Chen Zhensun’s estimation this provided a most useful arrangement. Facts listed in chronological order could be seen in the spatial configuration of the comprehensive unified gazetteer, which provided a layered map of both past and present administrative organization. As seen in chapter 3, temporal layering was a standard feature of empire maps and historical atlases. It was a powerful tool in the hands of those who wished to highlight the anomaly of the Southern Song condition, but it was also a research tool for politicians and scholars who believed that historical lessons were essential in planning for the future. Under twelfth- and thirteenth-century conditions, past political and military history, especially the histories where southern states were pitted against northern states, could only become useful when the spatial coordinates of past places could be correlated with present locations. Geographic knowledge and mapmaking was in Song times not only the pursuit of self-proclaimed loners such as Ni Pu or of families such as the Wangs. Maps were shared and mapmaking techniques were discussed in private conversations. Zhu Xi appears to have been confronted from time to time with questions about maps and mapmaking techniques. The notes recording such conversations are difficult to interpret, given the absence of the material objects themselves or of more extensive technical treatises. The following passage suggests, nevertheless, that Zhu Xi and his conversation partners possessed both familiarity with empire maps and confidence in the ability of any scholar to draw them: If you want to make a map of the administrative organization, there are three types: one is to write the prefectural names, another one is to write the county names, and the last one is to write the names of mountains and rivers. And still, when you make a map, you need to use the position, length, and width of each prefecture in accordance with its geographical shape, cut pasted paper sheets.63 from geographic location relative to the other regions, topography (rivers and mountains), and communication with the capital, to type of soil and tax contributions. 63. ZZYL, 2.30. The awkward phrasing in the translation adheres to the original Chinese text.



str ategic discourse

207

The meaning of the latter sentence becomes clearer in the context of other contemporary comments on amateur mapmaking. One potential source of inspiration for Zhu Xi and his students was Xue Jixuan’s 薛季宣 (1134–73) cartographic work. Even though Xue’s mapping efforts were never published in finished form, Zhu Xi appears to have seen a draft map. In response to another student’s question about Xue Jixuan’s Jiuyu tu (Map of the nine regions), Zhu Xi commented that Xue’s empire map was executed in too small a font and appeared too detailed.64 Zhu Xi’s reading method for maps conformed to his rules for reading other textual materials that focused on the underlying principles and the cohesion of the text and warned against the distraction of detail. The map’s layout should guide the reader’s eye toward the main features such as the three principal rivers (Yellow River, Yangzi, and Huai) and the general cohesion of the mapped places and features. The high resolution of Xue’s map obfuscated the significance of the foundational layer of primary natural features like the main rivers and made the cohesion of the empire map, which depended upon the prominence of these main features, inaccessible. Xue Jixuan shared his map with others as well. In a letter to Chen Fuliang 陳傅良 (1137–1203) describing his progress on a project to map the empire, he revealed that the map was originally made up of individual sheets for the broader regions that constituted the empire: As far as the maps of the regions [zhoutu] are concerned, I have received the two sheets for Jingzhou 荆州 and Nanjiao 南交. I expect to send them to you for your perusal soon after they have been copied. The drafts for Yangzhou 楊州 and Jizhou 冀州 have not been updated and [for] Liangzhou 梁州 and Heyi 和夷 the places have not yet been annotated. We have not yet started Youzhou 幽州 and Yongzhou 雍州.65

64. Ibid., 2.31, 79.8. 65. Xue Jixuan, Langyu ji, 24.15b. QSW, 257.273–75, esp. 275. Italics are mine.

208

margins, borders, and frontiers

Posthumous biographies of Xue Jixuan make reference to an unfinished draft titled Jiuzhou tuzhi 九州圖志 (Map gazetteer of the empire [the nine regions]).66 The work described in the letter to Chen Fuliang was most likely part of this unfinished project. Xue Jixuan’s work and his personal reflections on “zhoutu” illustrate that this term does not necessarily refer to a prefectural map (i.e., a map of one prefecture) but was in Song times also used to refer to empire maps. Such empire maps were administrative maps rendered as constellations of the original nine zones in which the Chinese territories were believed to have been subdivided in antiquity, and/or maps showing the jurisdictional layout of the Chinese territories in Song times. In a series of three poems celebrating a “zhoutu” Xue wrote: “One can reach at the same time the extreme end of Ao in the south and Ji in the north; / then one can see that the remote past will be the future present.”67 Even though some have read Xue’s work as evidence of a Song orientation toward the mapping of local places,68 the broader context of his work and thinking suggests that even if his maps were based on a more detailed rendering of local places (as Zhu Xi suggests), “zhoutu” as prefectural maps were understood to be just pieces in a larger puzzle making up the “zhoutu” or map of prefectures of the empire. Just as the terminology commonly used to name the local gazetteer ([zhou]zhi) as well as the standardization of its subdivisions reflected its membership in a net of gazetteers covering in theory the entirety of the empire, so the prefectural map (zhoutu) was by definition part of an empire-wide map of prefectures (zhoutu). Xue Jixuan’s use of antiquarian names for administrative subdivisions in his drafts is echoed in his poetry. In the drafts, the Southern Song frontier regions of the Lower and Middle Yangzi and Sichuan (Yangzhou, Jingzhou, Liangzhou) are referred to by their old choronyms and put in the context of similar subdivisions north (Youzhou and Yongzhou) and south (Nanjiao); in his reflections on the map it becomes clear that this rendering of the empire 66. Xue Jixuan, Langyu ji, 35.37b, 49b. 67. Ibid., 8.9a–b. 68. See, for example, Pan Sheng, “Cong Songdai shiwen kan yousi,” 75.



str ategic discourse

209

map allows the reader to see the layer of past jurisdictions as the norm and to accept the need for the restoration of prefectures currently under non-Chinese rule in the future. The geographies by commoners like Ni Pu, Wu Xie, and Wang Xixian, their attempts to have their work recognized by officialdom and the court (successful in the case of Wu Xie and Wang Xixian), and the broader discussion of empire maps and mapmaking indicate the rising importance of geographic and cartographic literacy among the educated elite during the Song dynasty. The construction of a growing number of empire map steles further suggests that geographic knowledge was considered essential to the formation of the Song scholar. Prefectural schools played an important role in the creation and preservation of map stelae as documented in Xin Deyong’s review of the growing number of such stelae produced in the twelfth and thirteenth centuries.69 As Xin Deyong also underscores, schoolteachers took the initiative in having these maps carved in stone because of their value in the analysis of classical texts such as “The Tribute of Yu.” Whether the incontestable connection with classical scholarship implies that reading maps such as “Yuji tu” as historical maps is anachronistic is, however, questionable.70 Judging from printed atlases such as Handy Maps and the geographical works described above, their creators demonstrated a keen interest in historical and military geography. Scholars’ engagement with contemporary military strategy may in part explain the interest in the exegesis and cartographic interpretation of “The Tribute of Yu.” All of the above works included present jurisdictions and did so in part with the aim to make the military history of past dynasties accessible to the Song scholar. As we saw in the previous section examiners and students could use such knowledge and the demarcation of strategic places and related markup of population figures and physical features in responding to questions about military strategy. The connection between geographic scholarship, mapmaking, and military studies is further evident from the research profiles of several of the men engaged in geographic scholarship. For example, Ni Pu was 69. Xin Deyong, “Shuo Fuchang shike ‘Yuji tu’ yu ‘Huayi tu,’ ” esp. 23–27. 70. Ibid., 16.

210

margins, borders, and frontiers

most famous for his memorial to Gaozong, in which he outlined the benefits of immediate military action against Jin and the procedures necessary to ensure success for the proposed military campaign. Chen Fuliang, with whom Xue Jixuan discussed his map, was the author of the earliest comprehensive history of military organization through the Song present (Lidai bingzhi 歷代兵制, Military institutions through the ages). Wang Qia may have been the author of a history of military administration under the Western Han (Xi Han bingzhi 西漢兵制, The military institutions of the Western Han).71 The nexus of interests in military planning evidenced in these works suggests that their creators shared the imperial mission articulated by readers of gazettes, maps, and gazetteers. The more activist quest for reunification was, in the case of its most adamant advocates such as Ni Pu, controversial, but for these advocates model emperors of the Song dynasty remained a focal point in its pursuit. By the thirteenth century Emperor Xiaozong 孝宗 (r. 1163–89) was remembered as the most activist among Song emperors. Like the founding emperors of the Han and Northern Song dynasties he was depicted with an empire map at hand planning for the unification of all Chinese territories: On the evening of the kuichou day of the seventh month of the first year of the Qiandao reign [February 15, 1165], the emperor went to Xiande Palace. Behind the emperor’s seat was a big lacquer screen. All administrative circuits were separately painted on it. For each, it mentioned the supervisory commissioners and the prefects in two columns; the rank and full name of all incumbents were marked with yellow tags. The emperor showed it to Hong Shi and the others present and said: “I had this screen newly made. On the back is a map of the Chinese and non-Chinese [huayi tu]. It makes for very convenient reading. In your offices you ministers can also rely on this.”72

Emperor Xiaozong thus appeared in charge of his field administration but at the same time with his gaze firmly fixed on those terri71. Based on ZZSL, 12.362. 72. Song shi quan wen, 24 xia, 10a.



str ategic discourse

211

tories currently not under his administration. He also held up for his ministers a vision of reunification, inviting them to see the map as a frame of reference. Others reminiscing about Xiaozong’s administration in the decades following the emperor’s death saw “his ambition for restoration” manifested in the establishment of a “Bureau for Restoration” (Huifu ju 恢復局) and his reading of the map of the Chinese and non-Chinese.73 Such depictions of an activist emperor fit in with private geographical projects that similarly constructed the Chinese territories as a configuration of administrative places focused on the prefecture. The meaning of the image of the activist emperor was contested. Some, like the senior politician Wu Yong 吳泳 ( j. 1208), warned against the idealization of Xiaozong’s reign among his thirteenthcentury contemporaries and pointed to the gradual approach advocated by those who came to be heralded as the principal embodiments of a prowar stance, including Zhu Xi and Zhang Shi 張栻 (1133–80). They also pointed to the lessons to be drawn from the subsequent history of defeat as a result of Han Tuozhou’s campaign against the Jurchens in the early 1200s. It was perhaps because the image of the emperor holding the empire map had become an issue of concern as well as celebration that some among the literati began to portray themselves not only as readers of empire maps but also as their creators and transmitters. Their investment in the imperial mission was attested by the continued coverage of and research on northern jurisdictions as well as southern ones. In sum, the prominence of private compilations in the twelfth and thirteenth centuries was a marked shift away from the dominance of court-commissioned gazetteers in the first half of the Song dynasty. This shift in authorship was accompanied by some modifications in the genre of the comprehensive gazetteer, such as greater attention to the literary representation of geographical features.74 Such changes were the by-product of the literati appropriation of a genre that was originally intended as an overview of all jurisdictions 73. Liu Shiju, Xu Song biannian zizhi tongjian, 9.5a; Wu Yong, Haolin ji, 19.15a; see also Xin Deyong, “Shuo Fuchang shike ‘Yuji tu’ yu ‘Huayi tu,’ ” 34. 74. Guo Shengbo, “Tang Song dili zongzhi.”

212

margins, borders, and frontiers

according to a standard list of common indicators. The administrative logic of the gazetteer left an imprint on private geographical compilations and private maps. As suggested in Zhu Xi’s remarks, the configuration of jurisdictions, especially at the prefectural level, was a foundational principle of geographic thinking. This principle of administrative subdivision was also the first of the reading instructions printed on the introductory pages of Handy Maps (see chapter 3). Private mappers of the empire did not seek to overturn the paradigm according to which court mapping had operated so far. The continuity between court-sponsored and private gazetteers was reflected in the recognition and titles some of their authors received from the court. Private gazetteers and maps were powerful demonstrations of the literati defense of the imperial paradigm at a time when it appeared that it had come under threat. If there was evidence of tension between the private scholar reconstructing the geography of the Chinese Empire and the Song court, it was mostly due to the insistence of private mapmakers and their readers that the Song court take immediate action either through troop mobilization (as in the case of Ni Pu) or through a financial and military recovery plan that would lead to a coordinated Song offensive (as in the case of Hua Yue to be discussed in chapter 5). Maps and mapmaking were symbolic of literati allegiance to the restoration of Song control over all of the Chinese territories.

Collecting the Frontier at Home The study of border affairs encompassed a range of fields and genres including military and historical geography, military studies, diplomatic history, and the ethnography of the peoples living on the periphery of the Song territories. I have shown the former to have been directly relevant to students preparing for the examinations. We have also seen that the former fields became an area of specialization for a number of scholars outside of officialdom. Did these fields and the ethnography of foreign others garner broader literati interest? The catalogs of private collectors provide here, as in the case of



str ategic discourse

213

the circulation of court archival materials discussed in chapter 1, a window into the range and volume of materials that informed Song literati about the extent and organization of Song territory, the geographical, sociopolitical, and cultural features of the polities that surrounded them, and the relationship between the Song court and its northern rivals. Tables 5–7 in appendix 1 represent the holdings, as reported in those catalogs, of embassy reports, gazetteers, and histories of what had been the Song court’s main rivals in the north up to the lifetime of each collector. Below I first evaluate the access to information about the surrounding polities, questioning how knowledge about others compared to knowledge about Song state and society. Next I examine what literati holdings of local gazetteers suggest about their territorial interests. Were the local gazetteers, as has been suggested, markers of local pride and local elite identities? Or did they for literati, as well as for the state, also fit into an empire-wide database of local circumstances?

R eports on Li ao a nd Jin in Pr i vate Collections All the collectors whose catalogs I surveyed and report in appendix 1, tables 5–7, held copies of Song reports of missions to Liao and Jin. They also invested in collections anthologizing and excerpting reports on missions and descriptions of foreign polities. In this area, too, currency and detail mattered. The tables contain some duplicate entries, particularly in You Mao’s 尤袤 (1127–94) catalog (those are designated by their grouped placement in the tables). Multiple copies of the same item in part reflect the use of different titles for the same item in manuscript copies, but collectors also acquired variant copies on purpose. Chen Zhensun noted that one of his copies of Fu Bi’s report came with an appendix containing transcripts of all letters exchanged between the two states.75 That such copies may have been targeted at men interested in the broader history of diplomacy is attested to in Chao Gongwu’s comment that his copy of the proceedings for the exchange of birthday wishes in 1030 contained a supplement listing the names of emissaries and 75. ZZSL, 7.203.

214

margins, borders, and frontiers

assistants of the last twenty-five years as well as a description of various items of diplomatic protocol.76 Within the context of tense Song-Jin relationships firsthand reports and intelligence from the north were in high demand. Their appeal was augmented by supplementary materials that catered to the appetite for news about the border among the reading public. Chen Zhensun’s copy of Er Yang guichao lu 二楊歸朝錄 (The Yang brothers’ return to the Song court) included an appendix containing “several dozens of items of intelligence on the enemy’s affairs.”77 The potential use of such information is evident from an examination question asking candidates to evaluate the reliability of intelligence from the north.78 Candidates’ access to such materials and discussion of it in broad terms is also demonstrated in an essay included in a commercial encyclopedia claiming an intelligence report as a source of information on the Mongols.79 On the whole You Mao’s and Chao Gongwu’s holdings of envoy reports to the Jin court were relatively small. You’s and Chao’s holdings of compilations that contained excerpts from Song reports and state letters as well as other types of records on the polity, geography, and customs of the Jurchens (appendix 1, table 6) would have made up for this apparent gap. For example, Chao Gongwu’s copy of Xu Mengxin’s Sanchao beimeng huibian 三朝北盟㑹編 (Collection of documents regarding relations with the north during three reigns) would have given him access to what is still regarded as the best resource on the military and diplomatic history of SongJin relations through Gaozong’s reign (r. 1131–62). The supplement listed in both Chao’s and Zhen’s catalogs, Beimeng jibu 北盟集補 (Supplement to the collection of documents regarding relations with the north), is no longer extant. Like the other two collectors, You Mao collected extensively on the tumultuous history of SongJin relations in the period between the 1120s and 1130s and obtained some titles on the genealogy and practices of the Jin court. More76. Chao Gongwu, Junzhai dushuzhi jiaozheng, 282–83. 77. ZZSL, 7.217. 78. Wei Liaoweng, Haoshan ji, 93.9a–11a; QSW, 310.231–32. 79. Liu Dake, Bishui (Nanjing Library ed.), 75.18a.



str ategic discourse

215

over, he appears to have been collecting reports until shortly before his death, as attested by the presence of a report on Zheng Ruxie’s 鄭汝諧 mission from the early 1190s.80 Most of these titles are no longer extant. Some of the remaining texts have been heavily edited throughout the centuries. Nevertheless, from catalog descriptions and occasional contemporary references and excerpts we can still catch a glimpse of the features of some of the texts listed in the tables. As Christian Lamouroux has argued, Song representations of the northern territories differed from earlier accounts on the peoples and political entities to the north of the Chinese territories in that they assumed cultural and administrative similarity as well as difference. Song authors described Liao and Jin territories like Song territories in terms of their administrative organization. They still applied some of the cultural stereotypes about northern peoples (for example, the representation of non-Chinese peoples as animal and not entirely human), but, overall, the description of northern societies was based more on the cultural categories deployed in the regional descriptions of Song territory (tujing 圖經, fangzhi 方志) than on those formerly used in ethnographic descriptions of non-Chinese peoples.81 The use of similar categories in the description of Song and non-Chinese territories conveyed a sense of cultural continuity, a sense of continuity that lay reflected in the concept of “the north” as both encompassing Liao or Jin territory on the one hand and expressing Song attachment to the same territory on the other hand.82 Song comprehensive and local gazetteers and descriptions of Liao and Song society share a bureaucratic logic. Chen Zhensun’s and Chao Gongwu’s brief descriptions and characterizations of their books on Liao and Jin society suggest that they dealt primarily with the imperial family, court ritual, bureaucratic organization, regional foods and customs, and the administrative subdivision and natural features of Liao and Jin territory. As in Song gazetteers, 80. You Mao, Suichu tang shumu, 46a. 81. Poo Mu-chou discusses the representation of foreigners as animals in Enemies of Civilization, 83–84. 82. Lamouroux, “De l’étrangeté à la difference,” 109.

216

margins, borders, and frontiers

maps were included in several titles. Su Song’s 蘇頌 compilation included a map outlining the trajectory up north along the postal relay network ( yicheng ditu 驛程地圖, map of the relay course).83 Chen Zhensun’s copy of Yanbei zalu 燕北雜録 (Miscellaneous notes on the north) included a map of the border forts (“Xi zheng zhai ditu” 西征寨地圖, “Map of forts on a western expedition”).84 Chao Gongwu noted that Jinguo Cheng’an xuzhi 金國承安須知 (Things to know about the State of Jin during the Cheng’an reign [1196–1200]) also came with administrative maps (地理圖 dili tu).85 Geographic information was most likely also provided in nongraphic textual form in such titles as Xiongnu xuzhi 匈奴須知 (Things to know about the [present-day] Xiongnu), whose main subject headings were, according to Chao Gongwu, administrative geography (dili) and administrative organization ( guanzhi 官制).86 Even though the full text no longer exists, quotations from this title and a comparable one for the administrative geography of the Jin Empire ( Jinren jiangyu tu 金人疆域圖, Maps of the territory of the people of Jin) suggest that the location of and distances among places were a primary focus in such sources.87 Maps of northern states were probably also available and collected in single-sheet format. You Mao and Chen Zhensun listed copies of “Qidan jiangyu tu” 契丹疆宇圖 (“Map of the territory of the Khitan”), and the latter described it as “a record of the places inhabited by the Khitan barbarians as well as all the places that were lost to the Central State (Zhongguo).”88 To what extent these collectors’ holdings were representative of contemporary scholars’ interests is impossible to gauge. They were, however, not exceptional. References to and excerpts from envoy 83. Wang Yinglin, Yuhai, 58.42a–b. The full preface of the text is included in the com­ piler’s collected works and here too reference is made to the map: Su Song, Su Weigong wenji, 66.2a. 84. ZZSL, 5.139. 85. Chao Gongwu, Junzhai dushuzhi jiaozheng, 1133. 86. Ibid., 284–85. 87. Pan Sheng has collected references to these sources in commentaries on Sima Guang’s Zizhi tongjian. Pan Sheng, “Songdai dilixue de guannian,” 194–96. 88. ZZSL, 8.267.



str ategic discourse

217

reports and other records on northern polities appear in notebooks and encyclopedias such as Wang Mingqing’s 王明清 (1127–after 1214) Huizhu lu 揮麈錄 (Waving the duster), Wu Zeng’s 吳曾 (fl. 1127– 60) Nenggai manlu 能改漫錄 (Unconstrained notes from Nenggai Studio), Zhou Mi’s 周密 (1232–1308) Qidong yeyu 齊東野語 (Rustic words from East Qi), Yue Ke’s 岳珂 (1183–1234) Ying shi 桯史 (Rec­ ords recorded by the pillars), Cheng Dachang’s 程大昌 (1123–95) Yanfan lu 演繁露 (Extending “The rich dew [of The Spring and Autumn Annals]” ), Zeng Zao’s 曾慥 (fl. 1136–47) Leishuo 類說 (Clas­ sified explanations), or Xie Weixin’s 謝維新 (fl. 1257) Gujin hebi shilei beiyao 古今合璧事類備要 (Complete essentials of classified affairs matching past and present). The 1170 report Lanpei lu 攬轡錄 (Holding the reins) was excerpted in Huang Zhen’s 黃震 (1213–80) Huangshi richao 黃氏日抄 (Mr. Huang’s daily notes), Yue Ke’s Kuitan lu 愧郯錄 (Jottings made in awe of [the State of ] Tan),89 and Hu Zi’s 胡仔 (fl. 1147–67) Tiaoxi Yuyin conghua 苕溪漁隱叢話 (Recorded conversations from Tiao Stream’s fisher-hermit). The excerpted content ranges from information about the trajectory to historical facts (e.g., a discussion of Jin reign names) and other trivia that were regular topics of interest in notebooks. These excerpts were meant to entertain connoisseurs with coverage of topics such as means of exchange, fauna and flora, local customs, and inkstones.90 The coverage of trivia about the north in notebooks, encyclopedias, and anthologies suggests at a minimum that their compilers regarded them as part of a repertoire of sources on which the cultivated scholar should be able to draw. Sharing such information was not so uncommon as occasional complaints about the stinginess of particular collectors might suggest. Chen Yuan 陳淵 (d. 1145) commented that he obtained his copy of another work on border affairs attributed to Fu Bi, Hebei anbian ce 河北安邊策 (Policy essays on pacifying the border in Hebei), from local official Yu Zicai 89. In his preface Yue Ke explained that he saw the ability of Tan’s leaders to fulfill Confucius’s questions on their administrative history as a standard that states in the past and literati in the present could hardly match. Kuitan lu, xu. 90. E.g., Wu Zeng, Nenggai zhai man lu (1979), 2.31; Zhou Mi, Qidong yeyu, 16.298– 99; Zeng Zao, Leishuo (Beitu), 13.37a–39a.

218

margins, borders, and frontiers

喻子才. He interpreted Yu’s efforts to disseminate titles such as these as a sign of his broader concern for the well-being of his generation and expressed his intent to borrow and copy more from Yu’s collection.91 Like maps, envoy reports belonged to a repertoire of texts and material objects that served as reminders of the imperial mission for many literati. In addition to his poems on reading empire maps, discussed in chapter 3, Lu You also wrote down his reflections on reading Fan Chengda’s Holding the Reins: “In the evening I read Fan Zhineng’s [Chengda’s] Holding the Reins. He said that when the older men in the central plain saw the emissary, many broke out in tears. I wrote a short poem in sympathy.”92 In this poem the author further expressed regret over the fact that brave warriors like Yue Fei 岳飛 (1103–41) are no longer chosen to lead Song troops and that the support for reunification in the north is not acted upon down south.

Ga zetteers a nd M a ps of the Song Empir e What access did private collectors have to publications on the spatial organization of the Song and the Chinese territories more broadly? All three private collectors possessed historical maps and multiple atlases. Besides Shui Anli’s Handy Maps, Chen Zhensun held another altas titled Liuhe zhangyun tu 六合掌運圖 (Maps holding together the passage of time in the world). Like Handy Maps it consisted of some forty maps starting with the administrative subdivisions described in “The Tribute of Yu.” Unlike any other extant historical atlas, however, Chen noted that this anonymous work focused on the geography of north and south. Following the historical map were maps of “the three territories in the north and south after the [Southern Song] restoration.”93 This most likely referred to maps of the Jin, Xia, and Southern Song states. Strategic readings of the empire map placed at the beginning were further elaborated in the final part of the atlas, which demarcated “the 91. Chen Yuan, Motang ji, 22.7b. 92. Lu You, Jiannan shi gao, 25.24a. 93. ZZSL, 8.266.



str ategic discourse

219

passes and strategic places along all the borders as well as the boundaries of the northern territories.” You Mao held copies of three better-known historical atlases: Tang Zhongyou’s 唐仲友 (1136–88) Diwang jingshi tupu 帝王經世圖譜 (Maps and lists showing the governance of rulers), Liujing tu 六經圖 (Charts of the Six Classics), and Handy Maps. Administrative maps of the Song Empire and comprehensive gazetteers were also of interest to Song literati. Chen Zhensun listed two administrative maps of the Tang and Northern Song empires. The Jiankang prefectural school counted a Tang Empire map among its map holdings. Transportation maps may have become a basic staple for the traveling scholar. According to Li Dongyou’s 李東有 memoir of Lin’an, a relay station along the official road that passed south of West Lake sold printed maps of the road system (chaoting licheng tu 朝京里程圖, map showing distance to the capital) that were in high demand among those visiting the capital. Upon examining the extent of its coverage, one of the travelers had left the following lines on the wall of the station: “By White Tower Bridge gazetteers are sold, with the distance between stops and stations clearly marked. Why do they only speak of the road to Lin’an and why do they not count the distance to the central plains?”94 Such maps became only occasionally registered property, as in the case of You Mao’s map of postal relay stations to the northern capital (most likely the former Liao capital) and the Jiankang prefectural school library’s copy of Jiangxing tu 江行圖 (Riverine transportation map).95 Comprehensive gazetteers summarizing basic data about all jurisdictions in the empire according to the state’s hierarchy of administrative places were, judging from extant catalogs, a musthave or at least a must-consult. Song state-sponsored gazetteers typically listed changes in the nomenclature, size, or number of 94. Li Dongyou, Gu Hang zaji (no longer extant in full), quoted in Li E, Song shi jishi, 96.2312. See also Tian Rucheng, Xihu youlan zhi yu, 2.5b. 95. ZZSL, 8.239, 241 (Tang shidao tu 唐十道圖 and Yudi tu 輿地圖); Zhou Yinghe, Jingding Jiankang zhi, 33.10a (Yuanhe junxian tu 元和郡縣圖); You Mao, Suichu tang shumu, 52a (Beidu yicheng tu 北都驛程圖).

220

margins, borders, and frontiers

jurisdictions, the distance between them, their main natural features such as mountains and bodies of water, population numbers, and goods sent up as tribute to the court. As shown in table 8 (appendix 1), all collectors held copies of Yuanfeng Jiuyu zhi 元豐九域志 (Comprehensive gazetteer of the Yuanfeng reign). All individual collectors held copies of several other Tang and Song comprehensive gazetteers. Moreover, besides the compilations sponsored by the Northern Song court, all collectors invested in the equally hefty volumes compiled by Song commoners in the twelfth and thirteenth centuries. Wang Xixian 王希先 and Wu Xie 吳澥, discussed above, belonged to a group of literate elites without examination degrees or without office who dedicated themselves to geographic and cartographic scholarship and sought to promote their work to potential patrons by highlighting its place in discussions about border affairs. The interest in their work suggests that readers accepted their claim that new geographic research was needed. Even when empire maps did not necessarily portray how the contours of the Song state had changed since the heyday of the Northern Song, strategic planning for defense and offense required a reexamination particularly of those areas along the Yangzi and Huai Rivers that had been transformed from core areas into frontier zones. A spatial turn in the Song imagination of the Middle Kingdom is also evident in the interest in local gazetteers. The transformation of pre-twelfth-century map guides (tujing), whose layout was primarily determined by the interest of the central government in the kinds of basic data needed for the compilation of the comprehensive gazetteers mentioned above, into local descriptions (zhi ) with greater coverage of the literary and cultural life of local places, and the greater participation and investment of local elites in the compilation of gazetteers have been read primarily as signs of a localist turn. The localist turn captures a significant moment in Song and imperial history when the capital was no longer the sole focal point of elite networking. Native or residential places became another and, for many elite families, the main focal point of the social, political, and cultural life. Nevertheless, an investigation of the genre as well as its presence in private collections suggests that a



str ategic discourse

221

gazetteer was typically but one part in a hierarchically organized collection of administrative places. Gazetteers rendered the empire visible by rendering local places ever more visible. The map that follows embodies an overview of gazetteers of counties and prefectures that were preserved in Southern Song private collections (map 4.2). It represents an approximate geographic distribution of the places covered in the county and prefectural gazetteers held by You Mao (51), Chen Zhensun (89), and Chao Gongwu (33). The map shows that the north is, with the exception of present and former capitals (Chang’an and Kaifeng) and some border prefectures, very poorly represented. The emphasis on the south does, however, not simply reflect the better socioeconomic and cultural conditions along the southeast coast or in some interior areas such as the Chengdu plain. Substantial coverage of counties and prefectures in Huainan and Jinghu suggests that peripheral areas were of equal interest as the socioeconomic and cultural cores. Huainan and parts of Jinghu became the new northern frontier of the Southern Song state upon the fall of Kaifeng and territories north of the Huai and Han Rivers, and, as we will see in this and the next chapter, thus became objects for strategic analysis and political debate. The absence of substantial numbers of counties and prefectures on the map does not necessarily imply that no gazetteers were compiled for them in Song times (dozens of known titles are not included in these collections); absences also reflect the lack of interest on the part of collectors or a lack of access. Collectors’ efforts to acquire gazetteers in large numbers from across the empire suggest that for them they formed part of a virtual comprehensive gazetteer that depicted the empire at a higher resolution than the summary comprehensive gazetteers of the empire could. In sum, literati collections and notebooks suggest that familiarity with the contours of the Chinese territories and foreign states was expected of the Song literatus. Song literati gathered local descriptions and information about foreign states. They did so by adapting bureaucratic frameworks, fitting local descriptions into a larger empire-wide whole, and matching the description of others

Map 4.2 County and prefectural gazetteers in three Southern Song private collections.



str ategic discourse

223

to the description of Song jurisdictions. The investment in border affairs and geography attested to in the work of Song examiners and mapmakers thus appears to have resonated among Song literati more broadly.

Administrative Reporting and the Bureaucratic Imagination Gazetteers, envoy reports, administrative and military maps, policy questions, and essays are all genres shaped by the language and the logic of the state and its bureaucracy. The dissemination of these materials through official, private, and commercial publishing opened the discussion of border policy up to readers beyond intended readerships, inside and outside officialdom. In the process of transmission the conventions of these genres intensified the cultural bureaucratization of Chinese society as bureaucratic formats and taxonomies set the tone for the interpretation of the geography of the Chinese territories and border zones as well as Song relations with border polities. A particularly apt example of this process is the history of the xuzhi 須知 or “what one needs to know about” genre. The recurrence of this little known genre in the catalogs of private collectors prompted the investigation I detail below, into its origins and into the prevalence of titles related to border affairs in it. Knowledge about border territories and states had become normalized among literati. I will return to the implications for court-literati relations at the end of the chapter. The earliest use of “xuzhi” in a book title may, judging from some catalogs, date to the third or fourth century, when Xu Xun 許遜 allegedly compiled a work entitled Taishang jingming yuan buzou zhiju Taixuan dusheng xuzhi 太上淨明院補奏職局太玄都省 須知 (An amended list of ranks and bureaus from the Pure Brightness Directorate of the Highest—What one needs to know about the Inspectorate of Great Mysteriousness). More likely, however, is the finding that this work was compiled by a twelfth- or thirteenthcentury follower of the Daoist Pure Brightness (Jingming) sect who attributed it to the earlier Daoist sage, presumably to place it within

224

margins, borders, and frontiers

a lineage with greater longevity and authority. Xu Xun had become a cult figure for Song and Yuan adepts of Pure Brightness.96 Regardless of the question of authenticity, this Daoist work demonstrates just how influential the bureaucratic model of reporting and digesting information had become. Scholars of Chinese religions (mostly Daoist religions, but to some extent also Indian and Chinese Buddhism) have explored the bureaucratic organization of deities and the bureaucratic modes of communication among deities and between humans and deities in Daoist narratives and ritual practices in some detail. An Amended List of Ranks and Bureaus from the Pure Brightness Directorate of the Highest further testifies to the adaptation of bureaucratic reporting genres in Daoist compendia. It lists in summary fashion the civil and military offices and ranks in the bureaucracies of the other world, explains the differences in the nomenclature among officers of different rank, and provides templates for the writing of such bureaucratic document types as memorials, report sheets, public instructions, tallies, short notices, and seals. Lists of bureaucratic titles were commonly used in Daoist rituals; knowledge about the administrative organization and communicative practices of the other world were also believed to aid believers in obtaining meritorious titles.97 Such an overview of the civil and military organization of a Daoist theocracy may have been inspired by a regular routine for gov­ernments at all levels. The first reliable published example of a xuzhi is Cao Fan’s 曹璠 Xuzhi guojing 須知國鏡 (What one needs to know—The mirror of the state). The work is no longer extant and only scant reference is made to it in the surviving literature. According to Wang Yinglin, Cao, who resided in the capital as holder of the title of left militant guard (zuo wuwei 左武衞),98 had brought together in this work basic data submitted by the prefectures about taxation, population (in households), products, official salaries, natural environment, and the ethnic makeup of the population. According to this review, the only one to have come down to us, 96. Kohn, Daoism Handbook, 417–18, 421, 448. 97. See, for example, ibid., 278–79, 313. 98. Hucker, A Dictionary of Official Titles, 574.



str ategic discourse

225

this work gave the reader the impression that all essential information about these areas of administration had been included.99 From the eleventh century onward xuzhi appear with greater frequency in the record. In Song times the title format (topic title, followed by xuzhi) becomes standardized, suggesting that this type of report had acquired the characteristics of a genre. What these characteristics were is difficult to determine and can only be gauged from a review of bibliographic entries, thin descriptions, and a few remaining examples. Bureaucratic xuzhi cover all levels of the administrative hierarchy, and examples cover a wide range of areas of administration, from tax administration and policing, to ritual and the handling of legal cases. County and prefectural offices kept xuzhi, which most likely consisted of instructions for the various types of organizations they controlled (such as charitable granaries), a description of the yamen’s routines, and perhaps also basic data on population and taxation. The gazetteer of Chicheng (Chicheng zhi 赤城志), a regional description of Taizhou 台州 compiled in the early 1220s by Chen Qiqing 陈耆卿 (1180–1236), refers to a county xuzhi when describing the walls of Linhai 臨海, one of its subsidiary counties.100 When the governor of Jiankang Prefecture established a granary around the same time to better regulate rice prices in an area that had been struck by the rising cost of basic staples, its mission and rules of operation were enshrined in a xuzhi. The granary’s rules of operation were to be sent up to the court but also became part of the prefecture’s archive (in this case in the form of a stone stele) whence it was copied into the prefectural gazetteer.101 Chizhou Yongfeng qianjian xuzhi 池州永豐錢監須知 (What one needs to know about the cash-producing county Yongfeng in Chizhou) most likely covered the operations of an industrial county charged with the manufacture of copper coins in Chizhou Prefecture.102 Occasionally circuit intendants were also asked to submit reports that contained summary information about the particular tasks 99. Wang Yinglin, Yuhai, 15.35a. 100. Chen Qiqing, Chicheng zhi, 2.5a. 101. Zhou Yinghe, Jingding Jiankang zhi, 23.3b–5b, 33.23b. 102. Cf. Wang Qing, “Bei Song Chizhou.”

226

margins, borders, and frontiers

they had been assigned. In 1042 all border routes in the north and west were asked to take stock of their weaponry and report back to the State Finance Commission. The reports were to become the basis for regular factsheets about the total military capacity of the border prefectures; this requirement may have been extended to all routes within a few years.103 Song catalogs and bibliographies further testify to the existence of similar route-level reports taking stock of particular activities and the work of dedicated commissions. For example, Zhihe fayun cha yan xuzhi 至和發運茶鹽須知 (What one needs to know about the tea and salt tribute, from the Zhihe period) appears to have been a brief on the tea and salt monopoly compiled in the period between 1054 and 1056, and Liangzhe zhuanyun xuzhi 兩浙轉運須知 (What one needs to know about fiscal contributions from Liangzhe) a summary of the (rice) taxes levied in Liangzhe.104 Routes were not formal jurisdictions with a permanent staff overseeing the administration of entire regions; the thematic focus of these reports reflects the areas of administration assigned to those commissioned to oversee route-level offices. Apart from information on government monopolies such reports may have also included advice on how to coordinate policing across an entire region as suggested by the inclusion of Fujian taozei xuzhi 福 建盗賊須知 (What one needs to know about banditry in Fujian) in You Mao’s private collection.105 At court xuzhi reports were used in debates about policy priorities; such summaries were based on reports submitted by lowerlevel jurisdictions as well as on those compiled by various central government bureaus. The latter kept a record of court regulations that applied to them, internal guidelines, and basic data, as attested in occasional references to them in contemporary historical 103. I am following Wang Yinglin (Yuhai, 151.54a) in reading “zhulu” rather than “tianxia” (XZZTJ [SKQS], 157.14b; [Hanji quanwen ziliaoku ed.] 157.3807). The two versions in combination suggest that the requirement, originally only targeted at prefectures bordering on polities with which the Song state had a tense relationship, was extended to all routes in ca. 1045. 104. You Mao, Suichu tang shumu, 26b; Ji Zengyun, Zhejiang tongzhi, 254.6b; Zheng Qiao, Tongzhi, 65.37b. 105. You Mao, Suichu tang shumu, 29a.



str ategic discourse

227

and bibliographical works. Under the reign of Emperor Huizong, Chen Guan 陳瓘 (1057–1122) submitted for discussion a report titled Guoyong xuzhi 國用須知 (What you need to know about the state budget). Chen set out to demonstrate that the vast majority of the state’s resources were then devoted to the military buildup along the western frontiers, a development he considered detrimental to the state’s fiscal health.106 Such reports may already have formed part of central accounting practices in the first decades of the Song reign as suggested by Neizangku xuzhi 内藏庫須知 (What you need to know about the Palace Storehouse), which was submitted by Liu Chenggui 劉承珪, who had been commissioned to supervise it, in 1012.107 Contemporary accounts noted that Liu’s compilation included all regulations relating to the treasury since its establishment and a rec­ord of all matters under its control.108 Court offices similarly kept track of legal pronouncements relating to their work and their internal affairs. In his bibliographic treatise, which was substantially based on the catalogs of others, Zheng Qiao listed such titles as Neiyiku xuzhi 内衣庫須知 (What you need to know about the Special Gifts Storehouse),109 Ranyuan xuzhi 染院須知 (What you need to know about the Dyeing Service),110 Sitian jian xuzhi 司天監須知 (What you need to know about the Directorate of Astronomy), Jing­ling gong xuzhi 景靈宫湏知 (What you need to know about the Jing­ling Temple), and Renzong shanling xuzhi 仁宗山陵須知 (What you need to know about Emperor Renzong’s tomb). Other xuzhi appear to have offered general information about the court or particular areas of administration such as Chaotang xuzhi 朝堂須知 (What you need to know about court 106. Yue Ke, Ying shi, 14.158–59. 107. Hucker, A Dictionary of Official Titles, 353. General accounting reports (kuaiji lu) were compiled throughout the Song period and can also be found in private collections. For a discussion of these records, see Fu, “A Study of Governmental Accounting in China,” esp. 299–332. 108. SHY, Shihuo, 51.2; XZZTJ, 79.1805. 109. Hucker, A Dictionary of Official Titles, 346. This office under the Palace Administration and the Palace Domestic Service stored fine silks and other materials to be used as gifts to imperial family members, foreign visitors, and the like. 110. Ibid., 271. This was one of the workshops under the Directorate for Imperial Manufactories (Shaofu jian).

228

margins, borders, and frontiers

palaces) and Zhiyu xuzhi 治獄須知 (What you need to know about judging legal cases).111 Bureaucratic xuzhi appear to have mostly consisted of internal instructions, relevant legal and policy regulations, and basic information crucial for the operation of particular offices or the conduct of particular business. Some originated as individual documents, as in the case of the granary instructions, but most became part of larger collections as they were sent up the hierarchy of administrative places and central government bureaus. They were living collections that were often absorbed into more formal arguments and texts and were only seldom compiled into a publishable format. Those outlining office routines and administrative practices may have served practical purposes for officials. They could serve as a memory aid for those working in or rotating into a particular office and as a checklist used for reviewing purposes by superior bureaus. Based on the limited data compiled in tables 9A–9B (in appendix 1) we may infer that bureaucratic xuzhi (those xuzhi concerned with the operation of the Song administration) were the largest subcategory within a diversifying genre. Distilling essential information in the format of a brief itemized report was an activity that was, however, not only applied to the specific concerns of particular offices; the range of topics to which literati applied reporting techniques expanded to include scholarly activity (history and classical learning), household management (elementary education and food), medicine, and religious practice. The thematic range of the reporting genre coincides with the expansion of the market in textbooks, encyclopedias, and manuals in the same subject areas between the eleventh and thirteenth centuries. Compared to perhaps the closest educational genre of the guide (zhinan 指南), which navigated the reader through more advanced and specialized knowledge and practical skills (such as writing), the xuzhi were typically restricted to a set of basic instructions for novices. Just as the bureaucratic report brought together regulations and basic data to facilitate supervision and decision making, the xuzhi listed in each area nothing 111. Zheng Qiao, Tongzhi, 64.17a, 18b–19b, 68.16a; SS, 205.5212.



str ategic discourse

229

more than the basic tenets that the average reader “needed to know” or basic data that would facilitate decision making. One of the earliest surviving examples of a xuzhi whose subject matter is not directly linked to administration is Lü Xiaqing’s 吕夏卿 ( j. 1042) Tangshu xinli xuzhi 唐書新例須知 (What you need to know about “The new form of ‘The Tang history’”). In this short piece, a supplement to a larger work on Tang history, Lü Xiaqing provided an anatomy of The New Tang History, listing numbers and names of emperors, ministers, civil and military servants, jurisdictions, taxation data, money supply, amnesties, and book collections, and counting how many biographies of different types it contained.112 Duyi xuzhi 讀易須知 ( What you need to know when reading “The Changes”) similarly provided an overview of the main features and topics of The Changes in just over thirty entries. The itemization of content in a relatively small number of entries (at least when compared with administrative encyclopedias or classified anthologies) appears to have been a distinguishing feature of the xuzhi. Chen Zhensun, who showed greater interest in the features of the books he had owned or seen than modern critical bibliographers typically give him credit for, noted that Luting xuzhi 虜廷 須知 ( What you need to know about the barbarian court) featured twenty-one entries and that Xi Xia xuzhi 西夏須知 ( What you need to know about Xi Xia) provided an overview of the Tangut state in only fifteen.113 The fact that the size of the vast majority of the titles listed in table 9A was one fascicle only further underscores that a brief survey of key indicators was the distinguishing feature and the appeal of this genre.114 112. For the place of this work in reviews of Xin Tang shu, see Balazs and Hervouet, A Sung Bibliography, 67. 113. ZZSL, 5.140. 114. From table 9A (appendix 1) it is evident that there is one outlier, Jiannan xuzhi 劍南須知, which consisted of ten juan. This work, no longer extant, was a compilation of both older historical material on the south ( juan 1–6) and new material gathered by the author himself. One entry from it on the horse trade during the 1060s and 1070s received the praise of Song and later historians alike as material not covered in standard historical accounts. Little is known about the author, Song Ruyu 宋如愚, other than that he was a native of Sichuan who attempted the examinations without success. WXTK, 205.11b.

230

margins, borders, and frontiers

The circulation of official reports and the adaptation of reporting templates to information of broader social and cultural interest are in China, as in England, America, or Japan, indices of modernity. Their circulation suggests that citizens have the right to information but the expansion of governmental interest in citizens’ lives at the same time demonstrates the nation-state’s regulative powers. Regardless of the incompleteness of both past and contemporary titles, duplication, and cataloging errors in its database, WorldCat’s holdings of Chinese xuzhi provide an indication of the exponential growth of the genre in the twentieth century. WorldCat, the largest union catalog in the world, contained only 26 records for xuzhi titles with a publication date between 1000 and 1800, 47 records for the period 1800 to 1899, and 1,696 for titles produced between 1900 and 1999. A similar but even more pronounced trend holds for guides (zhinan), which according to raw WorldCat results, grew from 62, to 73, and then to 15,788 during the same chronological intervals.115 The subjects for which the summary report had provided a template for information sharing in imperial times are still covered by twentiethcentury publishers. Administration, medicine, religion, and education remain core concerns, as illustrated in the publication of such titles as Wei guan xuzhi 为官须知 (The basics of becoming a government official, 2003), Linchuang yongyao xuzhi 临床用药须知 (The basics of administering medicine, 2001, 2005), Zisu bibei fojiao yishi xuzhi 緇素必備佛敎儀式須知 (The basics of Buddhist rituals essen­ tial for lay persons and clergy, 1976), and Xuexiao xuanyong keben ji xuexi cailiao xuzhi 學校選用課本及學習材料須知 (The basics for selecting textbooks and learning materials for use in schools, 2002). The main difference is that Chinese citizens of all walks of life have in the last century gained access to essential information about not only those topics but also social security abroad (Fei Meiguo gongmin shenqing shenghuo buzhujin xuzhi 非美國公民申請生活補助金須知 [Social security by the United States Social Security Administration, 1996]), legal protection (Quanli yu bianhu: fanzui xian­yiren ji jiashu xuzhi 权利与辩护: 犯罪嫌疑人暨家属须知 [Rights and defense: 115. Advanced search for 须知 and 指南 in title, limited to the time periods noted above and by language (Chinese), March 12, 2011, http://www.worldcat.org/.



str ategic discourse

231

The basics for criminal suspects and their family members, 2005]), business opportunities (Zuixin Taiwan waixiao huopin mulu ji chu­ kou maoyi xuzhi 最新臺灣外銷貨品目錄及出口貿易須知 [The latest Taiwan export articles and export handbook, 1972]), or genderspecific concerns (Nüxing xuzhi 女性須知 [What women need to know, 2006]). In the Chinese case reporting genres and reporting procedures had already become part of imperial elite culture from the eleventh century onward. Xuzhi on bordering polities and peoples took up the highest percentage of xuzhi in private collections (appendix 1, table 9C). This underscores the centrality of this concern among literate elites. Regrettably, none of these titles is extant, but the above description of Things to Know about the [Present-Day] Xiongnu suggests that in these titles non-Chinese polities were written up according to categories (such as administrative geography and administrative organization) that had become standard in the description of Chinese states. These titles thus incorporated a new understanding of northern polities as dynastic states with the attributes of Chinese polities. Such a representation of the Song state’s northern competitors underscored the need for a careful examination of available foreign policy options. These need no longer be exclusively based on the history of relations between central Chinese states and nomadic confederacies in the north but could now be discussed, as the reexamination of Six Dynasties military and political history suggests, on the basis of the history of interstate relations in the Chinese territories themselves. The dissemination of these and similar genres discussing Song border affairs in the twelfth and thirteenth centuries ensured at a minimum that literati shared the central government’s understanding of the nature of the enemy and the need for appropriate planning despite remaining opposition to the policy of appeasement. Between the eleventh and the thirteenth centuries, the Song government generated a great deal of information about the borders that separated Song territory from the political entities to the north and to the west. Officials addressed a wide variety of questions

232

margins, borders, and frontiers

about border areas, neighboring political entities, and Song border policy. The archives of the central administration contained maps, embassy reports, diplomatic letters, and policy proposals by civil and military officials. Occasionally, such information was systematically organized and compiled into large archival compendia. The court adopted a policy of secrecy regarding such materials, inheriting the legalist notion that the sharing of information of strategic value outside official circles jeopardized imperial authority. It defended and reinforced this policy of secrecy throughout the course of the Song dynasty by issuing bans on the circulation and commercial publication of Song documents on border affairs and current affairs more generally. The list of restrictions on circulation contrasts with an equally well-documented history of the official, tacit acknowledgment of the presence of such sensitive materials in private book collections and of their use in examinations at all levels of the school and examination hierarchy as well as in the manuals produced for students preparing for the policy essay session of the examinations. This paradox was never resolved. The insistence on secrecy allowed for the regular surveillance of sensitive materials. The proliferation of leaked materials and the compilation of new materials based on statist models had, however, created space for literati participation in border affairs. In the process, a sense of belonging to a “Chinese Empire” whose boundaries could not be defined by diplomatic negotiation had become part of the identity of the literatus.

Five The Multiplexity of Premodern Borders

I

n chapter 4, I showed how and why official documents, geographic sources, and examination materials on border affairs were produced and circulated at an increasing rate among literate elites in the twelfth and thirteenth centuries despite legislation against their publication. In this chapter I examine whether the growing interest in border affairs corresponded with changing perceptions and analyses of the periphery. This inquiry is not meant to be exhaustive, since border policy and literati perceptions of and interactions with border regions differed over time, according to policy orientation and depending on the topography of border regions.1 Rather, this chapter illustrates, on the basis of three military compendia and encyclopedias completed between the eighth and thirteenth centuries, that Song borders and understandings of them were multiplex. In the increased production and circulation of treatises on border affairs in Song times we find traces of the construction of new borders and new ways of constructing them alongside the reproduction of the earlier models. Even though there is no evidence for a linear transformation in the understanding of borders, the question remains, how 1. There is a large and growing body of scholarship on Song foreign relations, especially on Song-Xia, Song-Liao, Song-Jin, and Song-Mongol relations. A useful introduction to some of this literature can be found in Rossabi, China among Equals, and Twitchett and Smith, The Cambridge History of China, vol. 5, pt. 1.

234

margins, borders, and frontiers

portrayals and conceptualizations of the border were affected by the shrinking of the area under the effective control of the Song government. Following up on recent scholarship documenting the empowerment of local elites and local governments, the chapter closes with an examination of the value placed on local knowledge in the construction of imperial borders and the relation between centralizing and localizing tendencies reflected therein.

Margins: Ethnocyclopedic Writing Borders rarely figured in the military classics and were not singled out as a topic deserving concentrated analysis. The emergence of “bianfang” 邊防 (border protection) as a category in late Tang archival organization and its growing visibility in Song dynasty administrative encyclopedias and military compendia provide a window into interpretations not only of border protection but also of borders. In his Tong dian 通典 (Comprehensive institutions), a large encyclopedic history of imperial administration from antiquity to the Tang present, Du You 杜佑 (725–812) devoted sixteen chapters ( juan) to what had by the mid-eighth century become a principal concern for the Tang court: “the protection of the borders” (bianfang). This may have been the first time that this Chinese phrase, standard in modern Chinese, became a higher-order concept in Chi­ nese administrative compendia. The foregrounding of the concept around the turn of the ninth century certainly does not imply that questions relating to border defense had not been singled out for special reflection prior to 801, when Comprehensive Institutions was completed. Du You traced the history of border defense back to Zhou times and showed particular interest in Han dynasty debates on the subject. In early Chinese records on the relations between Chinese and non-Chinese states borders were conceptualized on the basis of varying criteria. Nicola Di Cosmo traces a trajectory in the imagination and use of borders from criteria based on strategy during the Warring States period, diplomacy under the early Han Empire, to ecological and cultural difference in the first comprehensive history of Chinese civilization,



multiplexity of premodern borders

235

Sima Qian’s Records of the Historian.2 These varying ways of interpreting the borders of states and later the Qin and Han empires co­ existed in Song political culture with models based on topography and administrative rationality. Du You’s chapters on border defense consisted for the most part of a survey of the peoples and states that had appeared in Chinese records throughout history. Divided into four generic types of nonChinese peoples (the eastern peoples, the southern peoples, the western peoples, and the northern peoples), the entries briefly described each ethnicity’s genealogy, tribal subdivisions, language, culinary preferences, social customs, death rituals, political organization, temperament, rough geographical location and movement, and its relations with surrounding non-Chinese peoples as well as with Chinese dynasties. In the latter category Du You took particular note of clashes with Chinese forces and intrusions into Chinese territories. As the example of the Uighurs, translated below, suggests, the types and amount of information presented differed in each case:3 The Uighurs were located north of the Xueyantuo 薛延陀 [SyrTardush] and lived by the Poling 婆陵 River,4 about 16,900 li from Chang’an. The number of soldiers reached as high as 50,000. Early on they belonged to the Tuque [Turks]. First there was one Shijian Aijin 時健俟斤. When he died, his son Pusa 菩薩 took over. In the beginning of the Zhenguan reign [626–49] of our Great Tang dynasty he rose up together with the Xueyantuo against the Turk Xieli 頡利 Khan [Illig Khan, 620–30].5 They invaded the northern border area of the Turks. Xieli sent his cavalry to punish them and waged battle with them by the Tianshan Mountains. He was greatly defeated and the members of his tribe were enslaved. The Uighur subsequently led his people to the Xueyantuo. He assumed the title of Huojielifa 活頡 利發 and still sent envoys to submit tribute. 2. Di Cosmo, Ancient China and Its Enemies. 3. Du You, Tong dian, 200.5491–92. 4. Following a suggested emendation by Wu Baogui, “Tong dian bianfang dian zhengwu.” 5. Wechsler, “T’ai-tsung,” 221–22, 230–31; Drompp, Tang China and the Collapse of the Uighur Empire, 20–21; Pan, Son of Heaven and Heavenly Qaghan, 171–79.

236

margins, borders, and frontiers

The Uighurs’ territory is sandy and salty. They have big goats whose feet are as long as five cun. When the Xueyantuo were defeated, their great leader of Silifa rank, Tumidu 吐迷度, led his tribe in submitting a petition and asked that they be placed on the population registers. Since the Turks have gone into decline, their state has gradually flourished. Their head of state is also called khan. In 727 they sent a minister, Meiluchuo 梅祿啜, to our court and presented famous horses on that occasion.6

Du You’s approach to the border may be called ethnocyclopedic. Ethnocyclopedic thinking, in the Chinese case, inventories nonChinese peoples as comprehensively as possible and characterizes each according to a set of ethnographic factors. Du You listed over 190 peoples and states. The list gave a full inventory of the peoples that appeared and disappeared on the fringes of Chinese empires throughout the course of recorded history. Brief characterizations based on various sources, supplemented, especially when information was scanty, with archetypical descriptions of the barbarian that were broadly applied across ethnic divides, such as “meat-eating and dressed in leather” (shirou yipi 食肉衣皮), combined to form a picture of a world of difference.7 Difference operates here on two levels. First, it opposes Chinese and non-Chinese. This difference is captured in the phrase “Huayi” 華夷. Du You prefaced the entire section on “the protection of the borders” with an elaborate contrast between the peoples living at the core of the known world, the Hua, and those living on the fringes (the yi). The opening sentence establishes the centrality of the 6. In an article on problematic passages in the border defense section of Comprehensive Institutions, Wu Baogui demonstrated on the basis of parallel passages in other Tang historical texts or later texts based on Tang documents that Du You most likely was mistaken in his assumption that the official presenting horses was a Uighur and not a Turkic representative (“Tong dian bianfang dian zhengwu”). In the original text Du You added a footnote alerting the reader to the discrepancy between his reading and that of his undisclosed sources, which all took Meilu to be a reference to a Turkic administrative title. See also Pan, Son of Heaven and Heavenly Qaghan, 287–92. 7. For a similar characterization of dynastic histories and encyclopedic works like Tong dian as accounts systematically articulating ethnic stereotypes, see Abramson, Ethnic Identity in Tang China, 41.



multiplexity of premodern borders

237

Chinese territories: “Within what is covered above and sustained below, and on that whereupon the sun and the moon shine, the Chinese [Hua Xia] live in the middle of the earth; there, living things receive the right energy [qi].”8 The geographical centrality of the Chinese and the natural benefits that were believed to come with that provided a rationale for the presumed superiority of this people’s intelligence and natural morality, the products of the soil it inhabited, and its political organization. The unity and coherence of the Chinese people and the continuity of its settlement and history were opposed to the teeming diversity of the non-Chinese peoples, their movement, and the discontinuity of the histories of individual ethnicities. Difference therefore not only signified the opposition of Chinese and non-Chinese, it also captured the multiple paths of deviation from the norm among non-Chinese that, by contrast, seemed to render Chinese civilization homogeneous. In the face of the other, the discontinuous history of dynastic rule, the partitioning of the territories at the core, and the cohabitation of Chinese and non-Chinese within it were substituted for a unitary history of the Hua Xia people. Du You’s portrait of the world lent a high profile to the Chinese at the core through the opacity of the peoples at the fringes. His ethnocyclopedic thinking was part of a tradition that lived on in maps of the Chinese and non-Chinese (Huayi tu 華夷圖), one of two types of empire maps around which both the history and the practice of cartography were constructed in Song times (see chapter 3). In Du You’s treatment the history of the Chinese became unitary through the shared experience of dealing with the other on the fringes of the Chinese world. Ethnocyclopedic thinking was therefore intimately connected with the discussion of border policy. An eighth-century military treatise was the main source for Du You’s interpretation of the history of Chinese border policy. In the conclusion framing the encyclopedic list of peoples, he quoted at length from Liu Kuang’s 劉貺 (?–759) Wu zhi 武指 (Military guide), a concise text reviewing and evaluating “those who have discussed borders” ( yi bian zhe 議邊者) past and present. 8. Du You, Tong dian, 185.4978.

238

margins, borders, and frontiers

The conclusion Liu Kuang drew from reviewing the history of border defense debates was that strict separation between Chinese and non-Chinese should be the basic principle in border affairs; failure to abide by this principle characterized the history of Chinese policy. Separatism as a principle of foreign policy may appear contradictory to the cosmopolitanism and multiculturalism that have become associated with the Tang Empire, but in his exploration of ethnic anxiety and conflict in Tang society Marc Abramson suggests that even strong endorsements of ethnic assimilation in the first century of Tang rule should be read in a larger context of ethnic stereotyping and separation: “As a rule, the actions of the state often belied its assimilationist rhetoric, leading to the conclusion that such rhetoric often served particular exigent goals within the broader context of a separatist-oriented ethnic policy.”9 The adoption of strict separation as the criterion by which the formulation and implementation of border policy ought to be measured led Liu Kuang to revise its history. He contrasted his reading to the views of Han politicians whom he regarded as the most reliable guides to the early history of border defense. Yan You 嚴尤, one of Wang Mang’s 王莽 (r. 8–23) advisors, faulted Zhou policy for not maintaining and enforcing a hierarchical relationship between Chinese and non-Chinese (on the model of that between master and servant); he faulted Qin policy not for repelling the non-Chinese but for bringing the state to disaster in its attempt to do so; and he faulted the policy of the first two hundred years of Han rule for first exhausting the state under the rule of Emperor Wu 武帝 (r. 141–87 BCE) and for subsequently resorting to policies of appeasement that were based on mistaken assumptions about commonalities between Han and Xiongnu society.10 Liu Kuang revised Yan’s judgment: When [the non-Chinese] caused disorder, [the Zhou] did not see that as a reason to tire their troops; when [the non-Chinese] submitted, 9. Abramson, Ethnic Identity in Tang China, 142. 10. For a brief description of Yan You’s views, see Yang, “Historical Notes on the Chinese World Order,” 23, 29.



multiplexity of premodern borders

239

[the Zhou] did not see that as a reason to slacken their preparations. They made their protective measures more stringent and their passes more impenetrable so that when [the non-Chinese] passed the border posts, there were victory announcements and intelligence was obtained; when [the non-Chinese] penetrated deeply [into Zhou territory] military honors were achieved and non-Chinese were exterminated. This way [the non-Chinese] were unable to invade when they wanted and they were unable to serve [the Zhou court] when they wished.11

Liu Kuang turned the alleged failure of the Zhou to establish a hierarchical relationship between Chinese and non-Chinese into an example of their ability to abide by the first rule of border policy, that is, splendid isolation through military preparedness. In Liu’s reckoning, Zhou policy was topnotch. Wall building merited the Qin a higher ranking than Yan You allowed because the walls were barriers designed to keep the barbarians out. Liu reversed the verdict on the much maligned Qin policy and suggested that by building walls the First Emperor of Qin was implementing instructions embedded in the oracular classic The Changes: “Kings and dukes establish strategic barriers to protect their state.”12 The walls were built in accordance with this instruction from the classic, and, in Liu’s reading, they effectively kept the Xiongnu from bringing their herds south into Qin territory. Liu Kuang reserved his scorn for the arsenal of policies proposed and implemented under Han rule. In addition to the misguided aggressive policies of Emperor Wu, he chided the policy of marrying off Han princesses to Xiongnu leaders, the granting of titles to Xiongnu tribal leaders, and the acceptance of Xiongnu settlers in Han territory. These latter policies illustrated that successive Han regimes failed to appreciate the fundamental difference between Chinese and non-Chinese. Marrying off a Han princess in the hope that the family ties thus created would prevent aggression was based on the moral reasoning fostered by Han society; 11. Du You, Tong dian, 199.5499. 12. Yijing, kan 坎 hexagram, tuan 彖 commentary; cf. Wilhelm, The I Ching, 532.

240

margins, borders, and frontiers

Liu argued that Xiongnu history, in particular the infamous anecdote about Maodun’s 冒頓 (r. 209–174 BCE) assassination of his own father in 209 BCE, demonstrated that such reasoning was alien to Xiongnu psychology. Politicians not only failed to appreciate the world of difference between Han and Xiongnu society, but also “abandoned sameness and pursued difference.”13 Laxity along the border was based on the acceptance of a multicultural world; the acceptance of difference was the reversal of the first principle of border policy attributed to the sage rulers of Zhou: absolute difference should exclude nonChinese from the norms and practices that structured relationships in the Chinese world. The application of social rules constituting familial and political hierarchy to Han subjects defined their identities as civilized Chinese; extending these rules to non-Chinese led to an explosive mix, because their aggressive natures combined with the desire to become Chinese. This mix, not the inherent characteristics of non-Chinese society, was the cause for foreign occupation. Mixing had the opposite effect on the Chinese. Corrupted by nonChinese fashions and practices, they lost the impetus to protect civilization. This view of the negative and destabilizing impact of the periphery on the metropolitan center was shared by other Tang representatives of the state.14 Liu Kuang developed such a thesis in Military Guide on the basis of an analysis of the history of early imperial border policy. Du You elaborated on its contemporary rationale and implications: During the Kaiyuan [713–41] and Tianbao [742–56] reigns of our dynasty the realm was peaceful, but border generals fished for favors and competed for meritorious campaigns. In the western corner we had the Qinghai campaign, and in the northeast we stationed troops in the Tianmen Mountains; in the western desert we had the Battle of Daluo [Talas], and in Yunnan the call to cross the River Lu. Several hundreds of thousands of men died in foreign territories.15 13. Du You, Tong dian, 199.5501. 14. Abramson, Ethnic Identity in Tang China, 171–78. 15. Du You, Tong dian, 185.4980–81.



multiplexity of premodern borders

241

The appended footnote clarified that those responsible for these disastrous losses were generals of foreign origin and those who asso­ ciated with them: Geshu Han 哥舒翰 (?–756), a Turkish general, allegedly sent 20,000 troops to death in Turfan; An Lushan 安祿山 (?–757), of Sogdian descent, lost 100,000 in campaigns against the Khitan and Xi peoples in the northeast; Gao Xianzhi/Go Seon-ji 高仙芝 (?–756), of Korean origin, was held responsible for the demise of 70,000 troops in the famous Sino-Arab Battle of Talas in 751; and Yang Guozhong 楊國忠 (?–756), whose relationship to the imperial con­sort Yang Guifei 楊貴妃 (719–56) allowed him to rise through the ranks and share her connections to the likes of An Lushan, incurred the heaviest loss, of over 100,000 men in the south. The experience of such heavy losses in the mid-eighth century led Chinese historians and officials such as Liu Kuang and Du You to reconsider both Tang border strategy and its Han models. Liu Kuang also reversed the verdict on Ban Gu 班固 (32–92), whom he selected as a second and alternative representative of Han discussions on border strategy. Unlike other Han politicians and generals known for their campaigns into Central Asia (such as Huo Qubing 霍去病 [140–117 BCE]) or for their advocacy of negotiated peace (such as Liu Jing 劉敬 [ca. 200 BCE]), Ban Gu wrote a scathing review of Han border policy through the first century CE. He divided discussions on the topic up into two types of strategy: (1) negotiated peace through the exchange of Han women and goods, generally advocated by high court officials; (2) military campaigning into foreign territory, generally favored by the military establishment. He deemed both strategies ineffective because the former had led to the Xiongnu exploitation of Han resources and the weakening of border control, and the latter to Han losses and increased invasions from the north. The two eighth-century historians agreed with Ban Gu’s assessment, but noted that like Yan You he missed the larger point. Ban Gu had proposed that the proper way to treat foreigners was to abide by ritual and apply the same standards of integrity and social hierarchy that governed Chinese society. For Liu Kuang and Du You ritual does not reach the other side. The belief in the universality of ritual practice, which its defenders based on the authority of

242

margins, borders, and frontiers

Confucius, had to be replaced with the belief in the radical incompatibility of Chinese and non-Chinese social practices—a belief for which Liu Kuang sought Zhou precedent, even though the surviving record yields ambivalent results in this regard.16 In contrast to the display of wealth and bounty that has come to be associated with the tribute missions in Han times and later, Liu Kuang imagined the inverse of the rituals of civilization for any occasion when non-Chinese presented themselves at court: When non-Chinese come to court, they should be seated outside of the gates. The interpreters should be deputed to feed them. Because they are like birds and beasts we should not allow them to know great smells and wonderful tastes. When we obtain their instruments they should not be placed in palaces and temples. When we receive their tribute goods, because they are simply red thorn arrows and animal skins, we should not consider them as gifts or goods. Since the benefit is small, the compensation should be in keeping.17

In the end, Liu Kuang and Du You imagined a world in which no court visits would be necessary. Military Guide proposed that the best border strategy was a strict policy of apartheid: Chinese men are foot soldiers. They are at an advantage in territories with natural barriers. Barbarians are infantry. They are at an advantage in the plains. They are good at sudden strikes; we are good at strong defense. We should not pursue them and we should not compete with them. When they come we should seal strategic passes so as not to let them in. When they leave, we should block strategic passages so as not to let them return. We should strike them with long lances, and wait for them with powerful bows. This not to seek conquest over them, but only to hurt them. . . . This way, what need is there for receiving them according to ritual, and what use is there for competing with them in righteousness! 18 16. Di Cosmo, Ancient China and Its Enemies. 17. Du You, Tong dian, 199.5502. 18. Ibid. For another translation, based on a later version of this text in Ouyang Xiu’s New Tang History, see Wright, “The Northern Frontier,” 57.



multiplexity of premodern borders

243

Borders: Administrative Zoning The maps included in the early twelfth-century commercial atlas Handy Geographical Maps illustrate the coexistence of the ethnocyclopedic model of the border and a model based on the administrative zoning of border areas. Whereas the former model appears in the transhistorical map of the Chinese territories, the latter model is implicit in “Shengchao Yuanfeng jiuyu tu” 聖朝元豐九域圖 (“The map of the empire during the Yuanfeng reign [1078–85],” fig. 5.1) in which the borders of Chinese jurisdictions are clearly drawn. The Song Empire is drawn as the configuration of twentythree bounded circuits/routes (lu 路), the highest level of administrative division. The administrative units constituted an atomized world; each was drawn like an island in the sea of the empire. Com­peting states and peoples, in boldface, appeared on the north­ ern and western periphery, where the Khitan Liao and the Tangut Xia states, without geobodies, defined the limits of the unified world. Borders similarly appeared in the mid-eleventh-century military treatise Wujing zongyao 武經縂要 (Summa of the military classics, 1044) as the perimeters of individual administrative jurisdictions. The “Border Protection” section consists of the administrative geography of the circuits that border on non-Song peoples and states. Under each of the circuits are listed the civil prefectures, with subordinate counties, or the units of military administration, in declining order of rank and size: the walled city (cheng 城), the fort (zhai 寨), and garrison (bao 堡). The border zone is defined by the location and distance of these units relative to each other and to the negotiated borders with the Tanguts and Khitan, the bodies of water that demarcate linear borders, or the Great Wall. The connections between the administrative units emerge in the description of the width of the roads linking them, in the distribution of beacon towers, and in the systematic surveying of postal and guard stations per jurisdiction, detailing the numbers of foot soldiers stationed there, the distance and location to the closest nodes in the four directions, and occasionally, their main functions, especially when these were limited to the inspection of traveling merchants.

Fig. 5.1 “The Map of the Empire during the Yuanfeng Reign [1078–85]” (“Shengchao Yuanfeng jiuyu tu”), from Song ben Lidai dili zhizhang tu. Photograph courtesy of the Tōyō bunko, Tokyo.



multiplexity of premodern borders

245

Based on a map showing the distribution of zhai (the most v­ isible of these subdivisions) mentioned in Summa of the Military Classics per prefecture, fortified military settlements were clustered in the northern parts of the northern routes and in the southwestern parts of the southwestern routes (map 5.1). This pattern emerges even more clearly when the location and numbers of zhai and zhaipu (encampment) per region are included.19 In the north, zhai, as well as cheng and bao, emerged in significant numbers around the middle of the eleventh century, as the Song court began to develop a line of defense against the Xia state. The first wave of fortification (ca. 1040–60) was in part a response to similar efforts on the Xia side of the border territories. According to one Song estimate, the Xia rulers completed a network of fortifications, ultimately consisting of about three hundred nodes, between 1038 and 1048.20 Apart from their defensive functions, these fortifications also fit into offensive war planning. Some of the Xia forts were built in Song territory and successfully held for decades, serving as bases in the continuous skirmishes that characterized SongXia relations. By the end of the second wave of Song fortification building in the mid-1080s, Song had more or less caught up with Xia in the region of the Loess Plateau (roughly covering parts of the provinces of Shanxi, Ningxia, Shaanxi, Gansu, Inner Mongolia, and Qinghai) (table 5.1). This table also shows that the majority of the fortified settlements established in the eleventh century were built in this region (note the predominance of the northwestern routes, esp. Qinfeng 秦鳳). The emergence of a dense network of fortifications in the Song-Xia border area represented a shift in the focus of Song fortification efforts from the north to the northwest, as negotiations tended to yield more stable boundaries in the areas contested by Song and Liao. Song fortifications continued to emerge in the northwest in the late 1090s and early 1100s, with another 19. De Weerdt, “Information, Territory, and Networks—Extra Materials,” maps 5.2a–e. 20. Sun Wei, “Bei Song shiqi Huangtu Gaoyuan diqu chengzhaibao tixi,” 10. The following description is based primarily on Sun Wei’s master’s thesis.

Map 5.1 The distribution of zhai in Northern Song prefectures, based on the Summa of the Military Classics.



multiplexity of premodern borders

247

Table 5.1 Number of fortified settlements, by type, in the Loess Plateau Circuits

Walled city Garrison Fort (cheng) (bao) (zhai ) Total

Northern Song Hedong 0 Yongxing 2 Qinfeng 5 Xihe 1 Subtotal 8 Xia (9)

8 37 45 7 32 41 115 43 163 24 9 34 154 121 284 [sic] (35) (37) 300+ [sic]

Source: The information is abstracted from Sun Wei, “Bei Song shiqi Huangtu Gaoyuan diqu chengzhaibao tixi,” 27, table 5.

eighty-four datable military settlements built after a brief reprieve under the more conservative regime of the Yuanyou period. Archeological surveys so far confirm the density of the fortification and communication network in the northwestern circuits. Recent surveys also provide a sense of the scale and features of Song and Xia fortified settlements. According to Sun Wei’s calculations of the average perimeter of 117 settlements identified in northern Shaanxi, the outer walls of fortified cities (52) measured 1,762 meters in circumference, those of forts (55) 1,019 meters, and those of garrisons (10) slightly smaller at 1,002 meters. The numbers for 28 military settlements situated in the eastern part of Qinghai were 1,287 (18), 1,180 (1), and 466 (9) meters, respectively.21 Compared to the average length of county walls in this area, those of fortified cities were about one-third larger, whereas those of forts were of similar size. The archeological surveys suggest that all fortifications were enclosed by walls. The walls were predominantly made of stamped earth, but stone slabs were also used, either as the sole material or in combination with stamped earth. Remnant lengths of wall suggest that the height could vary from 2 to 8 meters and the width from 1.5 to 6 meters. Only a few fortified settlements 21. Sun Wei, “Bei Song shiqi Huangtu Gaoyuan diqu chengzhaibao tixi,” 7, table 1; also 35, 41. Sun’s data are based on Guojia wenwuju, Zhongguo wenwu ditu ji—Shaanxi fence and Zhongguo wenwu ditu ji—Qinghai fence.

248

margins, borders, and frontiers

possessed double or triple walls, but many fort walls were equipped with wall extensions for added protection.22 The size and layout of the military settlements suggest that they were more than fortified military camps. Archeologists have proposed that the border settlements replicated standard features of the midsize Chinese city: a small number of city gates opened up onto the main thoroughfares, which divided the city into distinct parts. One of the better-preserved examples of the forts, Linqiang 臨羌 (located in Haiyuan 海原 County in present-day Ningxia), was originally a Xia stronghold but fell to the Song military in 1098.23 The remains suggest that the fort was subdivided into at least four functional zones: administrative, religious, commercial, and artisanal. From the tax quota for these military subdivisions we can further infer that the settlements, especially the larger ones, were self-sufficient and had become part of a colonized border network with military and civilian functions. The mixed nature of the fortified towns was also apparent in the constitution of its military forces. In addition to regular Song forces (the Palace Guard) (ranging from a few hundred to a few thousand), the settlements were the basis for local militia consisting of immigrants and local non-Chinese groups who were recruited to open up land and assist in warfare when needed.24 Borders emerge in Summa of the Military Classics as a result of the Song and Xia states’ systematic effort from the mid-eleventh century onward to build dense fortification and communication networks in the Loess Plateau. The specification of distances to border demarcations ( jie 界) and between network nodes demonstrates that a certain amount of surveying went into the development and maintenance of this network.25 Even though the Song-Xia border may not have been as stable as the Song-Liao border, the persistent efforts at border demarcation and defense through settlement in the Song-Xia border 22. Wall extensions were part of walled cities more generally. See, for example, Steinhardt, Chinese Imperial City Planning, 6–8; Kiang, Cities of Aristocrats and Bureaucrats, 144–49. 23. Sun Wei, “Bei Song shiqi Huangtu Gaoyuan diqu chengzhaibao tixi,” 44–45. 24. McGrath, “Military and Regional Administration in Northern Sung China,” 183–84. 25. For a cursory overview of Song surveying and mapping methods, see Wu Chengyuan, “Songdai cehui shi kao.”



multiplexity of premodern borders

249

zone were in several respects compatible with the policies Song pursued in border negotiations with Liao. The Song state’s attempt to fix a linear border in the Hedong 河東 area in the mid-1070s and the negotiators’ systematic use of geographic tools such as maps and distance measurement26 between border features in multiple directions correspond with the ongoing development of a dense border network of military jurisdictions whose local territorial boundaries were understood to coincide with state boundaries. The network was, however, not imagined as a permanent natural feature set; it was part of an imperial strategy that expansionist politicians at the Song court used to move the border up north or west when circumstances allowed. As such the network shares some of the characteristics of the frontier, in its early modern European sense of fortified political boundaries and in its American sense of expanding civilized settlement.27 Success was, however, limited, and imperialist ambitions were time and again restrained by more conservative forces at the court. However, the concept of the militarized border as a launching pad for conquest left a legacy in Song strategic thinking and reemerged in a different guise after the Song court lost the northern territories to Jurchen forces in 1127.

Frontiers without Walls? Recently, bookstores have been selling such items as Sin­ cere Discussions of the Northern Campaign and Remedies for Achieving Security. These works consist of letters and reports sent in by Gong Rizhang and Hua Yue and their discussions engage the situation at the border.28

The report from which this epigraph is taken led to the prohibition in 1213 of Beizheng dangyi 北征讜議 (Sincere discussions of the northern campaign) and Zhian yaoshi 治安葯石 (Remedies for 26. Lamouroux, “Geography and Politics”; Mostern, “Apprehending the Realm,” 154–56. 27. For a discussion of these two models of analyzing the frontier, see Power and Standen, Frontiers in Question, introduction. 28. SHY, Xingfa 2.138.

250

margins, borders, and frontiers

achieving security).29 All copies and woodblocks should have been destroyed, but like other Song titles that were targeted in this way they survived through incorporation in larger compendia or circulated under other titles. Due in part to its commercial success, Hua Yue’s Remedies for Achieving Security survived, unlike many of the other works on border affairs and military geography discussed in chapter 4. The two banned texts exemplify the interest in strategic discourse among Song readers and reflect the changes in the production of works on military strategy in the twelfth and thirteenth centuries. Whereas Comprehensive Institutions and Summa of the Military Classics were large encyclopedias with limited circulation, primarily based on the state’s archival record, and compiled by high court officials, these texts were produced by two men without illustrious court careers. They consisted of essays and short entries on border policy and military affairs dedicated, as convention dictated for policy advice, to the emperor, who was the sole authority that could adopt them for empire-wide implementation. In his preface, addressed to Emperor Ningzong 寧宗 (r. 1195– 1224), Hua Yue asked that the emperor distribute his program for border defense to the highest court officials including vice-councilors, remonstrators, censors, and personal advisors. He suggested further that, if approved at the central level, it also be forwarded to regional officers.30 As the history of its publication suggests, the strategies Hua Yue proposed in Remedies for Achieving Security traveled well beyond the circles of the highest political and military leadership. 29. Hua Yue’s Remedies for Achieving Security was long deemed lost, but a Yuan manuscript including this title was rediscovered in the late eighteenth or early nineteenth century. This manuscript (now at Beijing Library), titled Record of a Northern Mission (Beizheng lu 北征錄), and another Yuan manuscript (now at Nanjing Library) were the source texts for the punctuated edition included in a combined edition of his extant works published in 1993 (Cuiwei nanzheng lu, beizheng lu he ji ). This edition includes a short biography of Hua Yue, a textual history of the two titles it incorporates, and transcriptions of prefatory and other bibliographical materials. For further discussions of Hua Yue and this work, see Li Chuanyin, “Lun Hua Yue,” 63–69; Wang Lianbin, “Songdai bingshu ji qi junshi lunli sixiang,” 173–75. 30. Hua Yue, Cuiwei Beizheng lu qianshuo, 148.



multiplexity of premodern borders

251

Distribution in print had by the thirteenth century become a strategy resorted to by those whose policy proposals were blocked by powerful councilors or court factions.31 Hua Yue’s proposals differed from their predecessors in their advocacy of a new set of policies toward the management of the new border along Huai and Yangzi.32 We will turn to the author’s critique of contemporary border policy and the strategy and methods proposed in his work after a brief discussion of the textual history of Sincere Discussions of the Northern Campaign and Remedies for Achieving Security.

The Te xtua l History of T wo Milita ry Tr e atises Hua Yue enjoyed some fame in the early thirteenth century. In 1205, while still a student in the Military Academy in the capital, he submitted a letter in which he criticized the sudden mobilization for war under Councilor Han Tuozhou 韓佗冑 (?–1207) and called for the dismissal of the councilor and his closest allies. The councilor responded to this example of student activism by imprisoning Hua Yue, first in the capital, and later in Jianning 建寧 Prefecture. After the defeat of the Song forces and the subsequent decapitation of Han Tuozhou, Hua Yue returned to the Military Academy and passed the military examinations. He briefly served in an official capacity, as an officer in the Palace Command (Dianqiansi 殿前司), but, as the 1213 prohibition suggests, soon ran afoul of the administration. Even though he was hailed as a loyal subject by the time The Song Dynastic History was written, very little biographical detail remains about the last years of his life, except for the elliptic comment in his official biography that he was publicly executed under the administration of Shi Miyuan 史彌遠 (1164–1233), another councilor who came under attack for his authoritarian style of rule. 31. For another example see Zhen Dexiu’s printing of failed examination papers discussed in De Weerdt, Competition over Content, 211. 32. For a discussion of the Southern Song court’s management of the new frontier, see Mostern, “From Battlefields to Counties.”

252

margins, borders, and frontiers

Even less is known about the second person mentioned in the 1213 report quoted in the epigraph, Gong Rizhang. According to a later gazetteer of Fujian Province, one Gong Rizhang, hailing from Putian 莆田 County, was enrolled as a government school student around 1196 and later became a prefectural school teacher.33 The quoted excerpt implies that Sincere Discussions of the Northern Campaign was Gong Rizhang’s work and Remedies for Achieving Security Hua Yue’s; this is how later interpreters have read the passage.34 However, one contemporary source attributes Sincere Discussions of the Northern Campaign to Hua Yue and excerpts under this title a text that is identical to an essay included in “Ten Policy Essays on Pacifying the Barbarians,” a collection of essays included in a combined edition of Hua Yue’s work on border affairs.35 This combined edition, titled Beizheng lu 北征錄 (Notes for a northern campaign), is not listed in Song catalogs, but exists in a manuscript edition dated to the Yuan dynasty. I treat both Remedies for Achieving Security and Sincere Discussions of the Northern Campaign as the work of Hua Yue because of this attribution and because of the overlap between the texts in Remedies for Achieving Security and the one essay quoted directly from Sincere Discussions of the Northern Campaign. The correspondence between the latter essay and those in “Ten Policy Essays on Pacifying the Barbarians” provides further ground for the argument that Sincere Discussions of the Northern Campaign and “Ten Policy Essays on Pacifying the Barbarians” were alternate titles for the same text. Gong Rizhang must have been involved in some capacity in the publication of the two collections, since he is also listed, under the suggestively distorted name of Gong Rihua (hua 華 as in Hua Yue), as the author of Sincere Discussions of the Northern Campaign in the bibliographic treatise of The Song Dynastic History.36 I surmise that, 33. Xie Daocheng, Fujian tongzhi (1737), 35.3a. 34. See, for example, Yang Guoyi, “Nan Song jinshu,” 120. 35. Xie Weixin, Gujin hebi shilei beiyao, bieji, 4. The corresponding text can be found in Hua Yue, Cuiwei beizheng lu, in Zhongguo jingdian bingshu, vol. 2, 1921–24. 36. SS, 208.5380.



multiplexity of premodern borders

253

like other students and prefectural school teachers, Gong, who hailed from a county in Fujian relatively close to Hua Yue’s location of exile and the place where he wrote his essays on border affairs, was involved in the preparation of this work for commercial publication. Jianning, where Hua Yue was sent to be held under official surveillance, was also a known center of commercial printing in the twelfth and thirteenth centuries.37 It is very likely that these titles were singled out because of the authors’ criticism of Han Tuozhou’s military campaign against the Jin Empire in 1206, and the court’s military policy more generally. Given the paucity of information on Hua Yue’s later career, the particular circumstances are impossible to trace. However, because of its high profile and commercial success Hua Yue’s work offers a unique window into the strategic discourse of Song scholar-officials and their conception of the border. Presenting sets of policy essays advocating war, peace, or selfdefense and circulating them may not have been very unusual in the twelfth and thirteenth centuries. In “Ten Policy Essays on Pacifying the Barbarians,” the first part of Notes for a Northern Campaign, the author adopted a format that appears to have become common in twelfth- and thirteenth-century writing on border affairs. Other collections of ten essays on this topic include Li Shun­ chen’s 李舜臣 (fl. 1160s–70s) Jiangdong shi jian 江東十鋻 (Ten historical cases on the region east of the Yangzi), Li Daochuan’s 李道 傳 (1170–1217) Jiangdong shi kao 江東十考 (Ten historical investigations into the region east of the Yangzi), and Meiqin shi lun 美芹十論 (Ten expositions humbly presented) attributed to either Xin Qiji 辛 棄疾 (1140–1207) or Huang Dui 黃兌 ( j. 1172). The number ten carried a sense of a complete or exhaustive treatment. Li Shun­chen and Li Daochuan, father and son, focused their essays on historical examples of successful battles waged from the south of the Chinese territories and on successful defense strategies employed by southern regimes throughout Chinese history, respectively. The author of Ten Expositions Humbly Presented outlined a prowar program and discussed those areas in border policy that policy makers at the 37. Chia, Printing for Profit.

254

margins, borders, and frontiers

Southern Song court should in his view prioritize to achieve the recovery of the north; Hua Yue followed in his track. “Ten Policy Essays on Pacifying the Barbarians” starts out with two essays on recruitment and personnel management. This placement was deliberate. Calling for a change of personnel was part and parcel of court politics, but, unlike contemporary politicians, who tended to frame such calls in the moralistic rhetoric of factionalism, Hua Yue urged the emperor to replace bookish scholars with “heroes and brave men” ( yingxiong haojie 英雄豪傑) who had personal experience with the topography of the south and in-depth knowledge of the strategic places along the Huai 淮 River. Such men trained in traditional military thinking and conversant with the technology and strategic thinking of contemporaries would lead the way to the eventual recovery of the northern territories and the reunification of the empire. In a letter addressed to Emperor Ningzong, which precedes the essays proper, he described the reunification in metaphors of family and lineage: “joining the subscelestial world into one family, and uniting non-Chinese and Chinese into one whole.”38 Such language underscores the common assumption of the natural belonging of north and south into a larger patriarchal community. The other essays address four larger topics in pairs. The second pair of essays deals with the problem of the Jin cavalry, a key military challenge posed by the Jin armies, and explains how to face and trap cavalry. The third set of essays focuses on the goals of the Song military effort, that is, how to recapture lost territory and how not to give it up. The last two pairs of essays address administrative procedures to enhance military performance (rewards and punishments for successes and failures and the protection of military secrets) and broader areas of policy of the greatest importance to the military effort (fiscal policy and the horse administration charged with the supply of horses to the military). If we rely on the date of the prefatory matter, Hua Yue completed Remedies for Achieving Security shortly after “Ten Policy 38. Hua Yue, Cuiwei Beizheng lu qianshuo, 23.



multiplexity of premodern borders

255

Essays on Pacifying the Barbarians.” Whereas the latter is a coherent treatise on major policy questions, the former reads like an encyclopedic list of forty-one points for border defense and the administration of the military. Like other contemporary administrative encyclopedias, the entries are organized under larger subject headings, nine in total, covering state planning, major tasks in border defense, defeating the enemy, schemes for the military leadership, weaponry, information gathering, admonishing the military leadership, strategies for border protection, and managing relations between soldiers and commoners. Like other commercial publications of an administrative nature, these military treatises attest to the interest in current information in Song political and strategic discourse and the greater availability of such information to literate elites. Hua Yue wrote that, even though he still found himself in exile in Jianning, he compiled this collection in reaction to the dismal news from the borders that reached him on a daily basis.39 Hua Yue’s ability to receive reports from the border areas in the hilly parts of Fujian on a regular basis illustrates the ability of officials and nonofficials alike to receive current information. It was also the result of the conviction of many military and civil officials and literate elites that discussions about military and political strategy toward bordering states should be based on up-to-date information.

Str ategic Pe ace a nd Ter r itor i a l E x pa nsion Commitment to the preservation and restoration of the normative dimensions of the Song Empire did not, however, imply a prowar stance. Hua Yue regarded peace negotiations and the recognition of the sovereignty of the Jin dynasty over the northern territories that such negotiations implied as an unavoidable policy choice for the Song court in the early thirteenth century. In Remedies for Achieving Security he unequivocally identified “Peace Negotiations” as “The One Major Strategy for the Army and the State”—the quoted phrases 39. Ibid., 22, also 147–48.

256

margins, borders, and frontiers

are the title and subtitle of the first chapter in Remedies for Achieving Security. The antiwar position of Hua Yue may appear paradoxical when read in juxtaposition to the vision of reunification evoked elsewhere. Hua articulated his commitment to the recuperation of the northern territories in a language and imagery shared by the more hawkish among his contemporaries: I have learned that emperors cannot avoid songs about the native place [guxiang 故鄉] and an average person cannot forget thoughts about the old soil [jiu tu 舊土]. This is because the maintenance of tombs is not due to the effort of an individual ancestor and the enjoyment of fields and gardens cannot come about in a single day. Once we packed and moved, carrying our children, bandits turned our homes into camps and dined on our cattle and horses. The grains we left in our storehouses became a food source for the bandits, and our treasures of gold and silk filled their bags. Our elders fell along the road and our youngsters were left behind in ditches.40

As in the poetry of Lu You 陸游 (1125–1210), Chen Yinglong 陳應龍 ( j. 1223), or Zhang Yuangan 張元幹 (1067–1143),41 Hua Yue here impressed upon his reader the memory of the homeland, historical claims to territory that come with transgenerational land ownership or clan settlement, and the feelings of homelessness captured in the image of abandoned relatives along the road and the appropriation of family property. He lent authority to the imagination of empire as homeland and the concomitant desire to recapture the northern territories and did so by invoking authoritative sayings about native place attributed to emperors and sages. In the first line of the passage quoted above he makes a casual reference to a song attributed to Emperor Gaozu 高祖 (256–195 BCE), the first emperor of the Han dynasty. Gaozu allegedly composed the song during a banquet in 195 BCE. This was shortly after the establishment of the Han dynasty, when challenges to the newly 40. Ibid., 76. 41. See De Weerdt, “The Cultural Logics of Map Reading.”



multiplexity of premodern borders

257

founded empire pitted the image of unified empire against the more familiar one of independently ruled kingdoms. Gaozu passed through his hometown of Pei 沛 (now Jiangsu, Pei County) upon his return from a campaign launched to quell an uprising, an experience that allegedly inspired the following: “A great wind came forth, the clouds rose on high. / Now that my might rules all within the seas, I return to my native place [ guxiang]. / W here can I find brave men to protect the four directions!”42 These lines were well known through their inclusion in key sources such as Han shu 漢書 (The history of the Han dynasty) and the literary anthology Wen xuan 文選 (Selections of texts). Just as Gaozu returned home to seek out men whose attachment to their native place could be replicated in a comparable willingness to defend the empire “in all directions,” Hua Yue evokes images of the home, both the ideal home whose imprint always remains and the ransacked home whose portrayal stirs to action. Readers were thus persuaded to “hold onto the land” (shou di 守地), the theme of the essay that opens with the above lines. Hua Yue also creatively interpreted words attributed to Confucius, who, according to The Analects, had said: “Average persons are attached to soil.”43 In the original context of the passage, persons who care about the soil are compared negatively to the morally upright who care about virtue. Hua Yue drops the pejorative meaning and transposes the statement from the level of ethics to the domain of human psychology: just like great emperors of the past, average persons cannot but have feelings of attachment for their native soil. The naturalness of attachment to home and native place turns, as we shall see below, in Hua Yue’s reading from a sign of the provincialism of the average human being into a legitimate basis for territorial strategies. Hua Yue thus appeared to add momentum to the renewed mobilization of emotional ties to the northern homeland in the early 1200s, yet his treatises were written in defense of appeasement. He noted that appeasement provoked opposition both in officialdom 42. Translation adapted from Watson, The Columbia Book of Chinese Poetry, 68. 43. Lunyu IV.11.

258

margins, borders, and frontiers

and among the literate elite more broadly. He attributed the strong opposition to appeasement in contemporary opinion to a shared sense of shame. The unreturned bodies of the two last Northern Song emperors who had died in captivity in Jin territory and the loss of territory became an obsession for scholars and officials who believed that the shame this had brought upon the dynasty could only be washed away through the determination to fight till death. Hua Yue’s portrayal of a sizable unified opposition was exaggerated and served as a foil for his defense of the policy. He set out with great rhetorical flourish that shame over territorial loss and unreturned bodies ought not to be central to the political agenda. Of higher urgency were readily observable phenomena such as the death of generals and soldiers, the starvation of the population in the border territories, and the pressure on male relatives among the commoner population to seek the battlefield at all cost. Hua Yue saw the death toll and desperation as signs of the failure of offensive tactics. He countered the principal reasons for a more aggressive approach (the inherent unreliability of non-Chinese regimes, and the nefarious impact of peace on the morale of the military leadership and on military preparedness) and called on the reader to ignore the voices of a coalition of officials, unemployed scholars, and warrior-like generals. Shame and anger were in his view not a safe guide in setting policy and waging war. He compared the Song state to a gambler who had fallen on hard times and whose only chance for survival and the recovery of lost wealth was to hold on to the little he had left and to plan for a future gamble. However, even for a vocal advocate of peace like Hua Yue, peace was a strategy that would ultimately be substituted for military invasion in order to restore the empire’s normative territorial dimensions. Unlike other advocates of temporary peace, who proposed to buy peace in order to rebuild the state’s tax base and military force, Hua Yue proposed to use a time of peace to build an intricate defense system that would turn the border between Song and Jin into a dense frontier network that would serve as both a line of defense against invasion and an expanding frontier, a launching pad for the eventual recuperation of the northern territories.



multiplexity of premodern borders

259

The Milita ry Chorogr a ph y of the Hua i-Ya ngzi Frontier It was no coincidence that the only essay from Sincere Discussions of the Northern Campaign to be excerpted in a contemporary encyclopedia focused on territory. Titled, “Obtaining Territory” (“De di” 得地), it reflected Hua Yue’s more general attempt to reconceptualize defense strategy in the Huai region. This reconceptualization was based on the recognition that the region had by the midtwelfth century effectively become a frontier. For Hua Yue, border defense as applied to this region consisted of the systematic repossession of territory conceptualized as a network of fortified settlements, roads, waterways, and the land that connected these strategically important features. He substituted “territory” (di 地) for “walled fortifications” (cheng 城) as the main object of analysis in defense strategy. He turned the latter into a subcategory of the former, and, given that earlier military works had already paid systematic attention to cities, he instead turned his attention to a detailed chorography of the Huai region. The geographical focus of Hua Yue’s chorography was on the region south of the Huai River encompassing the territories that had in successive treaties been assigned to Song control. This was a deliberate choice. Hua Yue argued that current conditions required a reorientation in military geography and intelligence. He criticized the contemporary obsession with the prefectures directly north of the Huai (Haizhou 海州, Sizhou 泗州, Suzhou 宿州, and Bozhou 亳州 in the eastern part of the Jin Circuit of Nanjing 南京 and Tangzhou 唐州, Dengzhou 鄧州, Chenzhou 陳州, and Caizhou 蔡州 in this circuit’s western part). The preoccupation with these prefectures reflected the desire among a sizable group in the military and in officialdom for the recovery of the central plains but had in his view left the Song borders insecure. Surveying the areas north of the Huai River had facilitated the invasion of Song troops there, but the inability to secure the border region south of the Huai time and again led to invasions by Jin troops who circumvented the routes taken by the northbound Song troops.

260

margins, borders, and frontiers

Hua Yue therefore drafted an outline of a military chorogaphy of Huainan 淮南 (comprising both the Song circuits of Huainan East and Huainan West) and Jing-Xiang 荊襄 (comprising Jinghu 荊湖 North and Jingxi 京西 South). These two regions fell south of the Huai River, but the Huai River ran along only a small part of the northern boundary of Jingxi South Circuit. The region that extended beyond the source of the Huai was also referred to as Hanzhong 漢中, after the Han River, which was its major river. This draft surveyed the fortified walls of prefectures and counties, mountain and river stockades (shanshuizhai 山水寨), strategic spots, and overland and river routes large and small. The draft survey and his request to the court to further extend it by requiring military and civil officials stationed in these regions to participate in a similar systematic survey of the land were the necessary building blocks in the construction of a tight defense system. The river played a paramount role in the conceptualization of the frontier. Rivers and other natural features such as mountain ranges had traditionally been used to define borders. During the eleventh century, as Christian Lamouroux has shown, court officials considered the Yellow River a natural line of defense against Liao forces. In the words of one official, “Our dynasty relies on the Yellow River to provide . . . an obstacle to the barbarians’ cavalry.”44 Lamouroux further suggests that the association of the Yellow River with the stabilization of the empire during the eleventh century foreshadowed the adoption of a river frontier after the Jurchen invasions of the 1120s. Song literati indeed considered the Huai River, and to a lesser extent the Han, to be of great significance to the protection of the Song state. In Hua Yue’s writings on border defense, however, the river frontier acquires a meaning larger than the physical course of the river. Whereas the eleventh-century author drew attention to the Yellow River’s natural ability to stop the advance of enemy cavalry, he instead notes the Huai River’s interconnectedness with overland and water routes up north. The river thus becomes part of a frontier network. 44. Lamouroux, “From the Yellow River to the Huai,” 561.



multiplexity of premodern borders

261

The Huai River has five river branches that link up to roads in enemy territory. . . . If our military commanders can determine their depth and width and find ways to guard them, if they can truly strengthen military preparations in accordance with their flow and advantages, then we will be able not only to maintain safety in Huainan but also to recover the territories north of the Han River with ease.45

The Huai was a frontier that kept the prospect of the expansion of the Song realm alive. For this reason Hua Yue expressed strong disagreement with that line of strategic thinking in the early thirteenth century that called for a withdrawal to the Yangzi River. The Han River provided direct access to rivers in the north. Once its course in Song territory had been sealed from Jin attacks, it could play a strategic role in the gradual process of the recovery of territories directly north of the Huai. In ways reminiscent of the representation of the border in Summa of the Military Classics, Hua Yue’s conceptualization of the border that separated Song and Jin thus combined elements of the frontier as a secured zone and as an expanding frontier for the Song polity. Hua Yue’s model deviated from Northern Song models of the border in some crucial respects. He pointed to the reliance on defense strategies based on walled fortifications as one of the main reasons for Song defeats against invading Jin forces. He recognized the historical significance of walled settlements in the defense of the Huai region. Following up on earlier twelfth-century work on the defense of walled cities such as the titles included in Shou cheng lu 守城錄 (Record of city defense),46 Hua Yue listed for his readers tried techniques for city defense. The main point lay, however, elsewhere. To underscore the bankruptcy of models of border defense based on fortified towns enclosed by walls, he described the condition of prefectures and counties in the Huainan and Huai-Han (or Jing-Xiang) regions.

45. Hua Yue, Cuiwei Beizheng lu qianshuo, 293. 46. For a study and translation, see Subrenat, “Recherches sur l’histoire militaire de la Chine du XIIe siècle.”

262

margins, borders, and frontiers

He counted twenty prefectures in the former, only ten of which had seats with walls in adequate condition. He noted that the seats of four prefectures including those located on the borderline of the Huai River, such as Xuyi 盱眙 (Huainan East, now northeastern Anhui) and Anfeng 安豐 (Shouzhou 壽州, Huainan West), were in bad disrepair and could no longer serve a role in defense. Of the county seats in the region, four had fully functional walls; the walls of at least eight stood with major gaps. A similar situation prevailed further west. Among the twenty-two prefectures in the Huai-Han region, which in Hua Yue’s count had a total population of over 14 million, only six or seven had walls. Of the counties in the region only eight or nine were walled. Hua Yue concluded that other ways to protect the population had to be found since walled prefectural and county seats no longer offered security for the majority of the population47 (maps 5.2–5.3). The road conditions were such that even in places where walled enclosures could have served the surrounding rural population they proved to be unreachable. In Hua Yue’s view a more adequate model for the defense of the border population and of the land they lived on ought to be based on a network of mountain and river stockades. Stockades were safe havens built on flat mountain tops and on river or lake islands. Their isolated location would make them theoretically defensible without the cost of building and maintaining long stretches of artificial walls. They could host populations ranging from a few thousand to up to several tens of thousands. During the wars with the Jin in the twelfth century, local populations in the Huai region had set up stockades in the hope that their strategic location on mountain tops or in the middle of the water would either leave them unharmed or provide an advantage in defense. In Anfeng, for example, the local strongman Sun Li 孫立 gathered a local militia, set up a water stockade, and helped prevent the invasion of the forces under the command of Jin Emperor Wanyan Liang 完顏亮 (1122–61) in 1161. The Song court rewarded him for his efforts with a banner.48 47. Hua Yue, Cuiwei Beizheng lu qianshuo, 172–73. 48. Huang Kuanchong, Nan Song difang wuli, 209, 222, 227–28.

Map 5.2 Prefectural and county seat walls in Huainan, based on Hua Yue’s survey. Black areas with full boundary lines designate prefectural seats with walls, grey areas with full boundary lines walled county seats. Black areas demarcated by dotted boundary lines represent prefectural seats with walls in disrepair, and grey areas demarcated with dotted lines are county seats without fully functional walls. A larger map with place-names can be seen in De Weerdt, “Infor­ mation, Territory, and Elite Networks—Extra Materials.”

Map 5.3 Prefectural and county seat walls in the Middle Yangzi region, based on Hua Yue’s survey. Black areas demarcated by dotted boundary lines represent prefectural seats with walls in disrepair, and grey areas demarcated with dotted lines are county seats without fully functional walls. A larger map with place-names is found in De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.”



multiplexity of premodern borders

265

Court and local officials endorsed the construction of stockades or collaborated with the local population. On the other hand, as Huang Kuanchong has demonstrated, many were suspicious of the power that local strongmen wielded over the local militia in the stockades. The court attempted to harness that power by appointing stockade heads, a measure that did not necessarily alleviate the tensions between local strongmen, magistrates, and court officials. The enthusiasm for stockades was thus not a novelty in Song political and military thinking. From the available literature, however, it appears that Hua Yue’s effort to turn the stockades into a systematic defense network complete with their own arsenal of customized weaponry was intended to move the support for such local militia to another level.49 Apart from the deficiencies of city walls, Hua Yue based his claims for the superiority of stockades on the topography of the region. He generalized that streams and marshes dominated the land­ scape of the eastern part of Huainan (Huaidong 淮東), whereas mountains and forests were more common in the western part (Huaixi 淮西). Consequently, water stockades (shuizhai) were most suitable in Huaidong. Hua Yue calculated that forty-nine stockades ought to be built in this region and outlined a network of places running from the eastern to the western part of the border. He noted that special precautions were necessary in the case of eleven of the chosen locations because they were prone to flooding in the summer and fall. Reviewing the events of the past year, when Jin armies invaded Huainan in response to Han Tuozhou’s campaign into Jin territory, Hua Yue concluded that the large numbers of deaths in Huaidong could have been prevented if the state had supported the construction of a network of stockades. Water and mountain stockades could have 49. Huang Kuanchong is mainly concerned with the relations between state and local society and between local and central government as represented in the history of stockades during the Southern Song period. Tao Jing-shen’s article (“Nan Song liyong shanshuizhai”) documents several cases of stockade construction in the Huai region and Sichuan, demonstrating that such local militia existed well before the famous case of Yu Jie’s use of them in his war against the Mongol invasions in the late Southern Song period. Unfortunately, there is no work describing life in the stockades or the history of individual settlements.

266

margins, borders, and frontiers

provided a safe haven for the population; they could also have played a crucial role in keeping water transportation routes open for grain shipments into the devastated region.50 As in Huaidong, the population of Huaixi had suffered major losses in the recent past because mountain sites had not been readied for stockade settlement and because the loss of Song control over mountain passes had isolated local populations condemning them either to Jin violence or to starvation. At the time of writing, Hua Yue estimated, ninety-four mountain stockades ought to be built there, linking mountain ranges east and west. Efforts appear to have been made to build and expand mountain stockades in the region, but for Hua such efforts lacked the careful planning necessary to turn the stockades into an effective line of defense. He noted that six Huaixi locations where stockades were to be built had no access to water, rendering them unsuitable for human settlement. To prevent gaps in his proposed linked network of stockades, water would have to be diverted to these locations.51 For the stockades to fully exercise their potential, they had to be built according to a set of basic principles and tried methods, equipped with customized weapons, and linked up into a broader regional communication network. Hua Yue summarized the construction process of a mountain stockade as follows: On high mountain slopes that cannot be easily ascended and that are flat on top and thus inhabitable, ramparts should be built taking advantage of the mountain’s features, all roads and paths should be cut off, and walls with arrow holes ought to be added. The population surrounding the mountain should be mobilized to transport provisions. They should lead the young and the elderly up the mountain, build houses, and clear the trees and bushes.52

He also outlined rules for the building of water stockades: 50. Hua Yue, Cuiwei Beizheng lu qianshuo, 172–74. 51. Ibid. 52. Ibid., 173.



multiplexity of premodern borders

267

Wherever islands are found surrounded by water in the middle of rivers, lakes, and marshes and where they are unreachable from the shore and suitable for human habitation settlements ought to be built taking advantage of the water’s features. The banks [of the island] should be raised with sandstone to prevent ships from coming and going. The population on both sides of the river ought to be mobilized to raise cattle and horses and help provision the militia.53

In Sincere Discussions of the Northern Campaign he referred to thirty-six secret methods for building mountain stockades and twenty-seven for setting up water stockades. For water stockades, he listed the advantages of different types of locations and techniques to sabotage the progress of enemy warships. Those located in shallow water, for example, should use wooden contraptions to help stall storied warships; those living near deep water could use river vegetation to tie up the oars of enemy ships. Vegetation on the water surface could also be used to mess up the wheels propelling ships; placing iron hooks on the riverbed was a technique that served to trap ships that reached deep in the water or to prevent others from reaching deep enough to gain sufficient forward momentum.54 For mountain stockades, water was essential. The author recommended that in places without water supply rainwater be stored in bamboo vats or wells dug and pumps installed to transport water from other locations. The recommended guerilla tactics for use in mountain stockades included hurling projectiles made of stone or clay to knock off the protective headgear of forces climbing up the mountainside and sending fiery missiles to expose advancing forces. Bows drawn from behind ramparts could be used to cut off the path of ascending enemy forces, while arrow shields could be used to prevent enemy arrows from striking the stockade forces. These and other techniques and weapons were listed in full in a preliminary section in the chapter on “the three major tasks in frontier defense” in Remedies for Achieving Security. It listed the thirty-six “mountain stockade 53. Ibid. 54. Ibid., 77–79.

268

margins, borders, and frontiers

implements” referred to in Sincere Discussions and augmented the list of techniques for water stockades to thirty-nine.55

Loc a l K now ledge a nd Loc a l Orga niz ation The emphasis on territorial control and consolidation required a different kind of relationship between central control, on the one hand, and local knowledge, on the other. While moving away from the landscape of prefectural and county city walls and into the thick of water and mountain stockades it became clear to Hua Yue that the familiar categories of bureaucratic information included in comprehensive and even local gazetteers lacked the level of detail necessary for the construction of a continuous frontier zone. The model he proposed was heavily reliant on local knowledge of a kind seldom captured in the gazetteers compiled by both resident and visiting elites. The map he constructed on the basis of field surveys and the knowledge of the mobile population at the lower level of the social spectrum is at such a high resolution that it has proven impossible to reconstruct it in full, based on the surviving geographic record. Stockades played a dual role in Hua Yue’s strategic thinking. They offered protection for the local population in ways similar and superior to walled cities, but their location near strategic passes and along water transportation routes would also enable their militias to block the advance of enemy forces, to facilitate communication across Song territories, and thus to play a role in the defense of the entire region. Responsive to the concerns of those who were critical of independent armed forces, Hua Yue proposed that the court retain control over the militia by appointing stockade leaders. Each stockade should be led by a “stockade official” (zhaiguan 寨官), and each group of ten stockades should be supervised by a stockade general (zhaijiang 寨將). These men were to be recruited from among the local population, with priority given to local strongmen who had proven themselves in the organization of local defense. Government support for the stockades was, Hua promised, the best 55. Ibid., 171.



multiplexity of premodern borders

269

way to cut the Song military budget and to provide adequate protection to the border population. Like court officials who had written about the benefits of stockades in local defense earlier on, Hua Yue emphasized a collaborative relationship between local militia and the mercenary army. The troops could help organize local militia and provide training and logistical support. The stockades, on the other hand, supported the operations of the troops stationed along the Huai River and across the region. In Hua Yue’s border defense network local militia bore the primary responsibility in the defense of settlements, and Song troops were to be stationed along major roads, by courier stations, and in the plains and the wilds to prevent enemy passage.56 He acknowledged that troops had already been stationed along some of the major roads and in strategic transportation nodes, but pointed out that in many border counties lots of minor roads as well as some major roads leading up north across the Huai had not yet been surveyed and put under military surveillance. As in the case of the stockades, the problem was not that military commanders were unaware of the need to protect roads and communication nodes, but rather that no systematic attempt had been made to survey the entire region and to station troops based on its current topography: “Overall, generals and border commanders in charge of border defense all have their own opinions on how to guard river crossings and strategic sites. They send reports to the Bureau of Military Affairs and memorials to the court in large numbers, but I have never heard of any that offer a full and systematic description.”57 He asked the court to send down orders to military commanders in Huainan and Hanzhong to systematically survey strategic passes and roads and to develop plans to guard them in accordance with their features and the surrounding environment.58 56. Ibid., 175. 57. Ibid., 178. 58. In the defense of narrow roads, for example, military planners should have soldiers take up positions in the surrounding mountains; in the case of wide roads that could transport large numbers of enemy soldiers at greater speed, Song forces should be stationed in the valleys.

270

margins, borders, and frontiers

To convey the urgency of the task and as an example of his own surveying and intelligence work in the area, Hua Yue listed several roads and strategic locations that required immediate attention. For example: From the lesser-known road in Baizhang 白漳, to Henglin 橫林, Shoulu Mountain 手爐山, Chang Lake 長湖, Baizhu Stream 白竹澗, Gulou Mountain 鼓樓山, Gushi Lake 古石潭, Yangpo 陽陂, entering the border of Nanzhang 南漳 County [Xiangyang 襄陽, Jingxi 京西 South], the roads are even and linked up. There are virtually no steep and hazardous passages. If enemy troops were to set out during the night and surprise us, how could Qianjiang 潛江 [Jiangling 江陵, Jinghu North] and Jianyang 建陽 [Jianning 建寧, Fujian 福建] ever resist? The two previous examples show that these two roads in the Jing and Xiang areas should not be left unguarded.59

Hua Yue impressed upon his readers that topographical surveys conducted by military personnel required the collaboration of ordinary locals such as woodcutters and shepherds, who knew the local roads and paths best. Military commanders and local officials all too frequently overlooked such local sources of information. They took on special importance in Hua Yue’s military strategy since coopting them would help fill out many unknowns in the Song map of the border. A detailed topography of the border region based on local knowledge of the unbeaten path was the precondition for the construction and maintenance of a tight frontier. The construction of a social network of information locally could also help prevent local collaboration with enemy spies. Map 5.4 is a modern translation of the textual map of the network of stockades in Huainan East proposed in Remedies. A more precise location of most of the place-names listed could not be provided because most of the lakes, mountains, and other place-names could not be found in local gazetteers or other historical and contemporary data sets. In contrast with his contemporaries, Hua Yue only listed a few administrative place-names at the county or prefectural 59. Hua Yue, Cuiwei Beizheng lu qianshuo, 179.

Map 5.4 The region proposed for the construction of ninety-four mountain stockades in Huainan East. Based on Hua Yue, Cuiwei Beizheng lu qianshuo, 184.

272

margins, borders, and frontiers

level, and these allow us to trace only a rough trajectory of the planned stockades. This map captures not only our inability to reconstruct historical maps for this period at a higher resolution but most likely also conveys the unavailability of geographical information at this resolution to the political elite. Empire maps did not go beyond the county level and major mountains, rivers, and lakes. Local gazetteers contain a higher level of detail but also proved insufficient as an index to local place-names. Hua Yue’s call for field surveys and a new mapping of the topography of the frontier envisioned a map that would show not only the relationship among nodes in the administrative hierarchy but also the relationship among nodes of strategic importance and, to an extent, reflecting local hierarchies. Control over local routes was a crucial element in Hua Yue’s new geography of the border. Since the region consisted mainly of flat land where horses might travel easily (this in spite of the perception that the watery south made the advance of cavalry impossible there), strategies had to be developed to stop the superior Jin cavalry from reaching towns. Hua listed a variety of guerilla-style techniques that could be used to stop the Jin cavalry along the roads. For example, for major roads on which enemy cavalry could travel back and forth, he proposed the “hidden bamboo trap” ( fu quan 伏筌). It consisted of bamboo stalks cut to form spears, which were then buried in the ground and covered with wisteria. If a horse stepped into the trap, the bamboo spears would shoot up from the clingy wisteria and hit the horse. A variant of this method was the “horse pull” (ma tuo 馬拖). This involved a rope strung across the width of the road and with a spear attached to it. If a horse had its feet caught in the rope and tried to break loose, the spear attached to the rope would jump up and spike the horse’s legs and belly.60 The defense of the network of stockades, roads, and strategic passes depended on a fast and dense intelligence network. Preparations would be most effective when advance notice of enemy movements were relayed quickly and reliably. Hua Yue devoted an entire chapter to intelligence gathering and communication in which he outlined several complementary methods. He emphasized in par60. Ibid., 66–67.



multiplexity of premodern borders

273

ticular the importance of “amassing intelligence” ( ju tan 聚探). He reported that, at the time of writing, rules were in effect that required soldiers stationed in one border jurisdiction to transmit emergencies to a small number of neighboring places. He objected that this often led to breaks in communication flows and proposed instead that a network of border relay stations (bianpu 邊鋪) be established. Following standard practice in postal communication, the stations would be assigned a fixed number of men; he suggested that a quota of thirty per station be established. Instead of the prevailing practice of sending men out in different directions, he proposed that each station would report only to the next station in the chain. He listed thirty locations for border relay stations in Huaidong and Huaixi. He grouped them in teams of three and assigned each team the task to cover specific areas across the border. Hua deemed most qualified for this risky job men hailing from families who had made a reputation in cross-border smuggling and trade. Relying on local knowledge, he recommended fourteen families who were famous for “crossing the Huai River to steal horses, crossing the Han River for transporting salt, and smuggling copper and other goods.”61 Needless to say, all these activities were prohibited by Song law; punishments were set for those who were found carrying on illegal trade and rewards were given to those who helped in their arrest.62 Under the present conditions, however, the expertise that these men who “considered the area south of the Yellow River their family property”63 had acquired would prove of critical value in Song attempts to acquire up-to-date information about Jin troop movements. To subsidize the operations of spies in border regions and deep into Jin territory, Hua supported large upfront payments and bonuses. Those spies who were found guilty of malfeasance were to be punished in accordance with military law. In sum, in Hua Yue’s work on border defense, the Huai region was conceptualized as a frontier network consisting of stockades, 61. Ibid., 258. 62. See Quan Hansheng, “Song Jin de zousi maoyi”; De Weerdt, “What Did Su Che See?” 63. Hua Yue, Cuiwei Beizheng lu qianshuo, 258.

274

margins, borders, and frontiers

mountain passes, roads, and rivers whose density would define and secure Song territory and its northern border, and whose organizational features would serve as a model and a launchpad for the gradual recovery of the northern territories. This model continued Northern Song efforts to construct continuous border networks, but supplemented the nodes of walled fortifications with stockades. In this model there was, in ways congruent with broader developments in the relationship between state and local society in the twelfth and thirteenth centuries, a greater emphasis on local ownership of border defense. In Hua Yue’s vision the state coordinated information gathered locally and endorsed the organization of local self-defense among local strongmen. Theoretically this model would have allowed the weaker Southern Song state to develop the dense defense network and expanding frontier more commonly associated with the reformist Northern Song regimes. In the three administrative and military compendia from the early ninth, mid-eleventh, and early thirteenth centuries discussed in this chapter, we encounter three different approaches to border protection and three different conceptualizations of state boundaries. Du You opted for an ethnocyclopedic description of the transhistorical borders of Chinese empires. The detailed cataloging of heterogeneity beyond Chinese civilization in his ethnocyclopedic model fit into a strategic position that was characterized by the advocacy of nonengagement militarily as well as diplomatically. The ethnocyclopedic model remained current between the twelfth and fourteenth centuries, for it was embodied in Song maps of the empire and replicated, albeit under the revised title “Non-Chinese Peoples in the Four Directions,” in Ma Duanlin’s 馬端臨 (1254–1325) encyclo­ pedia of 1308, Wenxian tongkao 文獻通考 (Comprehensive investigation of textual records).64 The Summa of the Military Classics of 1044 embodies a different conceptualization, with its vision of a dense border network, evident in its survey of the infrastructure of the Song-Xia border. This 64. WXTK, “Siyi kao,” juan 324–48.



multiplexity of premodern borders

275

vision emerged from eleventh-century Song court debates and was advocated by those who supported reformist and self-strengthening policies. The topography of Song territory displaces the ethnography of the other in this model. Even though this model served to delineate and defend administrative borders, its uses moved beyond the hermetic closure of strategic traffic nodes proposed by Liu Kuang and Du You. During the waves of fortification in the eleventh and early twelfth centuries the clustering of military settlements turned the border region into a frontier that could be mobilized when an expansionist mood prevailed at the center. As Hua Yue suggested, the eleventh-century model presented the border as a network of walled fortifications; given the poor condition of the walls in most border jurisdictions this model, particularly when applied to the Huai border region, lacked density. But after 1127 the Song court no longer faced the need to defend long land borders in the sparsely populated northern and northwestern periphery of its former empire. The border was defined primarily by natural features, such as the course of the Huai River, in what had been the Song heartland. Expanding on the frontier network of the eleventh-century north, Hua Yue envisioned not a network of walls, as in the Summa’s account, but an internal frontier network that would exploit the riverine and mountainous topography of the region and that, in keeping with the more general retreat of the Song state, would capitalize on local intelligence and military organization. Hua Yue’s emphasis on comprehensive and high-resolution surveying of the Huai border zone and the development of an internal frontier differentiated his work not only from Tang and Northern Song antecedents but also from competing contemporary positions. Besides topography, contemporary military strategists also turned to history when analyzing border affairs, especially the history of the Six Dynasties, when north and south were similarly held by competing dynasties. For men such as Li Shunchen and Li Dao­ chuan the history of campaign strategy and of actual battles between north and south was the best guide to border planning. For these historians too, however, geography was a crucial factor in learning lessons from history. These works rendered the history and

276

margins, borders, and frontiers

historical geography of the Six Dynasties period relevant to Song policy debates about border policy in the region formerly occupied by the southern dynasties. Others, like the author of Ten Expositions Humbly Presented and the examiners and amateur geographers discussed in chapter 4, shared an interest in military geography but, as Hua Yue’s critique of contemporary military geography pointed out, they focused on trajectories to the north based on a network of prefectural nodes and major rivers and mountains.65 This approach had failed to deliver, and failed again when Song invaded Jin in the 1230s and for a short time recaptured Kaifeng in 1234. “Consolidating territory” (shou di) was thus meant to substitute a more systematic approach to territorial control for the looser network of fortifications characteristic of earlier models of territorial expansion and control. The high resolution and the novel emphasis on the inclusion of local knowledge in Hua Yue’s military geography is evident from a comparison of his work with Cheng Dachang’s Beibian bei dui 北邊 備對 (Notes for a response on questions about the northern border, 1191). The latter work resulted from the embarrassment that Cheng Dachang suffered as a court lecturer in the mid-1170s. In response to Emperor Xiaozong’s request to shift the focus of their conversation from “the mountains and the rivers of the heartland” to “the geography of the lands of the northern peoples,” he later remembered he had had only this to say: “The barbarians do not have writing. What has been transmitted in the histories lacks in detail and reliability. How dare I force what I do not know into a form of knowledge?”66 In retirement he set out to gather whatever materials he could find to make up for his earlier ignorance. Cheng Dachang admitted that his was by no means a full account. In this small notebook we find listed seven mountains, two passes, the desert, rivers and other bodies of water, the great walls, and foreign peoples; he also noted in general the presence of centers of power like capitals and courts and the lack of city walls. The sources for this compilation were limited to historical and classical 65. Yan Bai, The Political and Military Thought of Xin Qiji, 214, 271–85. 66. Cheng Dachang, Beibian bei dui, preface, 1a.



multiplexity of premodern borders

277

materials; archival materials and Song compendia do not appear to have been consulted, perhaps in part because of the engagement with historical and philological debates in Cheng Dachang’s published work. Even though Cheng Dachang shared with Hua Yue and his Song contemporaries a focus on the geographical features of border regions, the list of features in his work illustrates the novel interest in high-resolution surveying, in the extension of such surveying to an entire zone, and in the examination of the relationships among a wide range of features. The emphasis on territorial control in the Song models of the border discussed above is representative of a broader trend. Marc Abramson has argued that the emergence of a sense of loyalty based on ethnic difference had a lasting impact on the course of imperial Chinese history and facilitated the maintenance of an ethnoculturally unified empire after the late Tang. And Naomi Standen, even though much more critical of attempts to make ethnicity a defining category in the historical analysis of Five Dynasties transborder movement, similarly notes that attitudes hardened toward transborder crossings from the eleventh century onward.67 Such work ties into earlier scholarship that identified the critique of politicians serving multiple dynasties in Song historiography with the emergence of a paradigm of dynastic loyalty. The reinforcement of dynastic loyalty over the course of the eleventh century fostered political unity as well as acts of demonstrative loyalism, but the crisis of the twelfth century demonstrated that loyalty to the dynasty in and of itself proved insufficient. Even though dynastic loyalty has been claimed as the most significant and lasting contribution of Song politicians and historians to late imperial political culture, attachment to territory emerged as an equally important if not primary characteristic of Song political culture. In the eyes of literati who held different views on particular border strategies, dynastic legitimacy came to depend upon the court’s prospects for recovering the normative dimensions of the empire. Finally, the combination of the territoriality and the multiplexity of Song borders should caution us against equating them with 67. Standen, Unbounded Loyalty, esp. chap. 6.

278

margins, borders, and frontiers

the national boundaries constructed by early modern and modern nations. Even though linear borders were established and physically marked in Song times, they were not implemented systematically, nor were they deemed to separate the geobodies of equal states. Borders remained in Song times frontiers of civilization and embodiments of an imaginary universal “Chinese Empire” pitted against the transience of peoples and states beyond. The twelfthcentury crisis ensured that the defense of the imperial mission was taken up by elites across the Southern Song territories, and particularly in the core economic and cultural centers in the south.

Pa r t I V Imperial Information Networks

Si x The Notebook Phenomenon

T

he previous five chapters have shown that a structural transformation in political communication accompanied the social transformation and expansion of the political elite between the eleventh and thirteenth centuries. The institutions developed to centralize communication in the eleventh century, such as the Memorials Office or the bureaus in the Historiography Institute, were infiltrated by literati networks spanning court, capital, and local jurisdictions. Regulations and prohibitions on the production and dissemination of state documents and archival compilations and various materials on border affairs proliferated but could not stem the flow and reflected the growing dissemination of current affairs through woodblock printing as well as manuscript production. Genres that had been firmly associated with the court in earlier centuries, such as empire maps, archival compilations, military encyclopedic texts, or gazetteers, were in the twelfth and thirteenth centuries adopted and adapted by literati settling and writing in the provinces. Their interest in these genres was facilitated by the expansion of the civil service examinations. Also crucial for the consolidation of the shift from court control to literati participation in the production of texts relating to the state of the dynasty were the events of the mid-1120s and the establishment of the Jurchen Jin dynasty in the north. The genres that literati turned to were predominantly concerned with the history and geography of the former Song Empire or of a normative “Chinese

282

imperial information networks

Empire” and were dedicated to the restoration thereof. The secondary discussion of maps, gazetteers, envoy reports, and court gazettes and border reports in letters and poems similarly gave voice to an imperial mission. In the chapters that follow I pursue the question of how such texts were received more in depth. The following three chapters address the question of how widely the exchange of genres relating to the business of the empire reached, socially and geographically. In line with earlier work on the redefinition of literati learning during the Song dynasty, these chapters also address the question to what extent such texts and the knowledge that could be gained from them figured in twelfth- and thirteenth-century literati learning. These larger questions are pursued here on the basis of a genre of literati writing and social exchange that broke through during these same centuries: the notebook. Notebooks have seldom been interpreted in their entirety either as single texts or in the aggregate.1 Instead, they have become a treasure trove from which historians and literary scholars select anecdotes and factual information one rarely comes across in other genres of classical Chinese writing. In this and the following chapters notebooks are approached in two ways. First, notebooks included records of texts seen and read and of conversations about them and so provide an ideal context within which to pursue questions about the dissemination and reception of court genres. Second, notebooks commemorated conversations and encounters as well as engagements with authors past and present and were exchanged with contemporaries; they are thus an ideal entry point for historians aiming to understand how literati communication networks operated across the three centuries spanning Song rule. In this chapter, I survey the social production of notebooks, examining how notebooks turned into a genre, who wrote them, 1. Exceptions include Bol, “A Literati Miscellany”; Zhang Hui, Songdai biji yanjiu. After I completed this manuscript some significant new contributions appeared: Dudbridge, A Portrait of Five Dynasties China; Jack Wei Chen and Schaberg, Idle Talk; Zou Fuqing, Tang Wudai biji yanjiu. See also De Weerdt and Chu, The Making of the Literati Notebook.



the notebook phenomenon

283

what institutions and types of publishers printed them, and how topical interests related to the changing profiles of authors and printers between the tenth and the thirteenth centuries. The focus on production in print (at the cost of an extended analysis of a thriving manuscript culture) is justified to the extent that it provides the best entry point into an examination of the maximal extent of literati communication networks. Authors and those supporting publication of notebooks claimed repeatedly that the transfer to this medium was motivated by the desire to make the work public, to share it with literati outside the smaller circle of those with whom one exchanged manuscript texts. In the words of Qiu Su 丘橚, who praised the efforts of Hong Mai’s son in the dissemination of his late father’s notebooks in print, “to share it with all [zhong 眾], and from here on people [ren 人] can read it and families [hu 戶] can transmit it.”2 In chapter 7, I pursue the question of the scope of literati networks further through an analysis of the infor­ mants who contributed to one particular notebook, Wang Ming­ qing’s 王明清 (1127–after 1214) Huizhu lu (Waving the duster). In chapter 8, we return to Song-Jin relations and explore the role of the events of the 1120s in later twelfth-century literati political identities through Wang Mingqing’s recollection of them.

Biji: The Formation of a Genre? Much is ill-defined and indeterminable about “written notes” (referred to below as biji 筆記). As befits a genre that tends to evoke associations of random and wandering observations, scattered anecdotes, and disparate conversations, definitions and classifications have remained elusive. In the few attempts made to clear up its long history, modern Chinese scholars have applied evolutionary models to sketch a literary history that stretches back from antiquity to the early twentieth century. Even though the link with republican-era biji writing is thus held in abeyance, the history of a continuous tradition of biji writing is very much the product of a postimperial 2. Hong Mai, Rongzhai suibi, 981.

284

imperial information networks

historiography that leaves questions regarding the historical development of genre conventions and the cultural history of notebook production, distribution, and reception untouched.3 The unusually intractable nature of the genre is reflected in the widely differing counts of the number of biji known to have been produced during the Song dynasty (960–1279): modern estimates range from over 100 to 708.4 Regardless of whether one employs a strict or more liberal definition of the genre, the number of notebooks produced in Song times testifies to the acceleration of a trend. (A strict definition would include only works in related bibliographic categories such as xiaoshuo 小説 [short narratives] or zajia 雜家 [miscellaneous expertise], or only works compiled as random notes excluding those that have been grouped together as fictional narratives [biji xiaoshuo]; a more liberal definition would include all sorts of historical and literary reporting genres, such as historical memoirs on specific events or places and literary criticism.) It was also only from the twelfth and thirteenth centuries onward that dozens of biji titles appeared in print. In spite of the insoluble problems of numbers, this much can be established: biji turned into a genre in the course of the eleventh and twelfth centuries. Well over a dozen publications explicitly carried biji and associated terms (suibi 隨筆 or random jottings, bilu 筆錄 or written rec­ ords, bitan 筆談 or written conversations) in their titles,5 and many more authors attested to a shared understanding of a form of writing 3. See, for example, Zheng Xianchun, Zhongguo biji wen shi, and Liu Yeqiu, Lidai biji gaishu. 4. Zhang Hui includes 155 titles in Songdai biji yanjiu. The editors of the ongoing series Quan Song biji 全宋筆記 intend to publish about 500 titles in their collection of all Song biji; preface by Fu Xuanzong 傅璇琮, ser. 1, vol. 1, 1. An earlier series of biji, titled Lidai biji xiaoshuo jicheng 歷代筆記小説集成, included 188 Song titles: Zhou Guangpei, Songdai biji xiaoshuo. A recent master’s thesis lists 708 titles; Sun Li, “Song­ dai biji fenlei kaobian,” chap. 2. For an overview of recent editions and scholarship, see Zheng Jimeng, “Jinnian lai Songdai biji yanjiu shuping.” 5. Zhang Hui, Songdai biji yanjiu, 1–2. The term “biji” had been in use to differentiate regulated types of composition from “free-style” writing. For a discussion of the relevant chapter in Wenxin diaolong 文心雕龍 (The literary mind and the carving of dragons), see Owen, Readings in Chinese Literary Thought, 272–77. By Song times it also



the notebook phenomenon

285

that was characterized primarily by the way in which data are collected and recorded rather than by topical content. These works, which have come to be classified as biji only in recent times, cover a wide range of texts distributed across the topical categories used in imperial book catalogs and bibliographies; their authors emphasize, however, that they were the cumulative result of the personal collecting, recording, and evaluating of information acquired by hearsay and/or reading. Despite later attempts to subdivide notebooks into distinct thematic types, such as Liu Yeqiu’s oft-quoted subdivision of the corpus of biji texts into stories, histories, and philological works, Song authors legitimize biji writing as a form of writing that, unlike the more heavily scripted historical, literary, or philosophical genres, is defined by a broad and free choice of subject matter (boxue 博學, or broad learning), a nonstructured mixture of topics (za 雜, sui 隨, man 漫), the personal verification of the content by the author/compiler, and the itemized presentation of all verified content (tiao 條 or item, entry, shi 事 or affair, event). Song literati regarded the critical reading and verification of the evidential basis of statements in Song notebooks as a feature that distinguished Song learning from that of previous ages: For scholar-officials [shi] it is not broad learning that is difficult; rather, it is being able to reflect carefully and to draw clear distinctions that is difficult.6 Among those who lived in the past there were of course men who devoted themselves to books and hunted down written records and who zealously tried to accumulate them until they turned grey. However, they went for the ordinary and the conventional. In their reading and writing they sought after the obvious [what was before their eyes], they did not try to understand it deeply. They transmitted errors and uncertainties and they referred more specifically to the kind of free-style writing that could be found in individually circulating notebooks. 6. The phrase “shen si ming bian” (審思明辨) anchors what appears to be a statement about the distinction of Song scholars in the classics, and specifically in a chapter of The Records on Ritual that had become central in Song intellectual discourse. Legge, The Li Ki, 318. Harbsmeier, Thesaurus Linguae Sericae, text ID: LJ 0.0.31.2.0.20.

286

imperial information networks

did not make connections through classification because they did not focus on research. In the case of Mr. Gong, he was famous for his scholarship and writing among contemporaries. All men of importance vied to have him as their protégé.7 Starting from the classics, philosophical texts, and histories, there were none he did not read. Of the River Charts and the Luo Diagrams [explanations of the origins of the cosmos], inscriptions on mountains and tombs, regional terms and local gazetteers, Buddhist and Daoist texts, and those written by poets and litterateurs, there were none of which he did not take note. As for discussing precedent, amending historical facts, distinguishing pronunciations, evaluating and analyzing writing, he would without fail restore the truth even in the case of short expressions, so that scholars would know the evidential basis. This [notebook] is the result of what he got out of his reading while living in retirement; [his findings] were then recorded and therefore called “written notes” [biji].8

Liu Dong’s 劉董 1201 preface to Gong Yizheng’s 龔頤正 notebook Jieyin biji 芥隱筆記 (Written notes from the Jieyin Studio, ca. 1162) addresses the then much-debated question of what kind of learning should be the hallmark of the literatus. Broad learning based on a critical examination of all genres of textual materials and the ability to convey such critical learning in oral and written discourse were in Liu Dong’s eyes what distinguished the Song literati from those who had come before them. The notebook that contained the fruits of Gong Yizheng’s scholarly labor was the result of a commitment to research and the rejection of mere reproduction. An awareness of the genre conventions in terms of technique, content, and form can be found in prefaces and postfaces such as Liu Dong’s quoted above or the following by Liu Changshi 劉昌詩, author of Lupu biji 蘆浦筆記 (Notes written in Lupu, 1213): 7. The latter expression of praise is a literal quote from Han Yu’s tomb inscription for Liu Zongyuan and thus associates both the preface and the two authors whom it brings together with the twelfth-century enthusiasm for ancient prose ( guwen). Liu Dong here also compares the notebook’s author to arguably the most critical among late Tang guwen authors. For a Song edition of this text, see Han Yu, Zhu wengong jiao Changli xiansheng wenji, 32.4b. 8. Gong Yizheng, Jieyin biji, preface by Liu Dong, CSJC-CB, 33.



the notebook phenomenon

287

I serve by the edge of the sea and have little else to do besides selling salt. In my office I am the only civil servant; there are no colleagues coming by. As I am living in a remote village there are also no attractive young scholars to engage in discussion with. I long for home, but as I am poor I could not bring my personal affairs. Sitting up by the lantern light, I just go through books to amuse myself. I trace the origins and follow the developments in all such subjects as the commentaries of former scholars, the historical events of the successive dynasties, mistakes in texts, and changes in administrative geography. If I cannot be satisfied in my mind, then I leave it and do not dare to provide an explanation. I investigate in other sources and pursue my work with vigor; when I occasionally find a trace, this pleasure is then a source of joy. I have been dreading that this will be lost. Therefore, I have brought together what I have seen and heard in the past and written it down in a fascicle. In total there are over one hundred items gathered in ten chapters. Those that have not yet been verified I leave for a supplemental compilation. How dare I, a lonely scholar with little knowledge, do such a thing? I seek approval from more enlightened scholars. It is my wish that they could amend and correct it.9

As in the case of Liu Changshi’s written notes, biji typically resulted from a process of accumulation and sorting. At times of leisure or in retirement, scholars and officials followed up on things read, seen, or heard about history, geography, philology past and present, and more through further investigation in written sources. They transferred those pieces of information that had been researched and vetted into a carefully packaged volume. Notes were presented as affairs (shi), items, or entries (tiao, ze 則), often counted, and in several cases presented in packages of round numbers such as roughly one hundred items per installment. Liu Changshi was most likely very familiar with the genre and with publishing conventions such as the presentation of packages of notes in installments. Note taking was an ongoing process of engagement with all topics that mattered to the literati elite and that defined who they were; the mention of a supplement (xubian 續編) may well have been an 9. Liu Changshi, Lupu biji, preface, 1.

288

imperial information networks

e­ xpression of the desire of this low-level functionary to follow into the footsteps of peers like Wang Mingqing who similarly started with a volume of one hundred entries in 1166 and followed up with three more installments of Waving the Duster (in volumes of roughly fifty, one hundred, and two hundred entries) in the 1190s. Even earlier, in 1149 Chen Shan 陳善 had edited some of his notes in a volume of one hundred entries titled Menshi xinhua 捫蝨 新話 (New explanations bluntly presented); he followed up in 1157 with another installment of one hundred entries. Chen Shan wrote that he was motivated to publish the second volume because “those who love stuff” (haoshizhe 好事者) had taken an interest in his earlier notes. That notebooks had a communicative and networking value of a special kind, one that was principally limited to the literati class itself, was also recognized by Liu Changshi. In a postface he justified his personal investment in the printing of his notebook by pointing to the same anonymous public of aficionados: In [Ye Mengde’s 葉夢得 (1077–1148)] Shilin yanyu 石林燕語 [Shilin’s casual conversations] we come across many historical facts and past anecdotes or great sayings and inspiring achievements past and present; it is a work of broad scholarship. However, Mr. Wang Huaiyu 汪懷玉 [Wang Yingchen 汪應辰 (1118–76)]10 pointed out mistakes for each item; it is truly difficult to write. I have not read much. I entrust this to writing to test myself. Those who love stuff have recently tried to obtain it but I cannot give them my handwritten notes. After two years, in the autumn of 1215, I took out my own salary to have woodblocks cut in the county studio of Liufeng. This is not something to transmit but I wish to learn about the mistakes contained herein.11

Notes Written in Lupu is thus presented as both a critical examination of information received and a source for further critical examination. The author and printer was referring to an existing audience 10. Wang Yingchen wrote a work entitled Shilin yanyu bian 石林燕語辨 (Points about “Shilin’s casual conversations”); two printed editions of this work are included in De Weerdt, “Biji in Print Database.” 11. Liu Changshi, Lupu biji, postface, 83.



the notebook phenomenon

289

of men who asked for his work; the not uncommon practice of supplementing or correcting the mistakes of other notebook writers suggests that the author could indeed insert himself within a network of peers with a similar interest in conversing across distances about our affairs (shi, the affairs of interest to the cultured elite). Notebooks were about affairs examined and presented in itemized form (shi) but they were also about “loving affairs” (haoshi), about maintaining an interest in the affairs of the cultured elite. The term haoshizhe, used here and elsewhere to refer to the readership of notebooks, carried ambivalent meanings. It often occurs in a pejorative sense, denoting those among the elite with an obsessive interest in elite preoccupations, whether it be art objects, gossip, or indeed the itemized knowledge contained in notebooks. Even though the hao­ shizhe were a target for some notebook writers who criticized their lack of discriminating taste, their dilettantism, or their forging of texts,12 the existence of those suspected of being busybodies and of getting a thrill out of the exposure to and the acquisition and use of preselected snippets of information proved to be a boon for notebook producers. Liu Changshi was not the only Song author to have his own notes printed. His is a very telling example of how printing had become recognized by the Southern Song as the preferred medium to reach a broader public of peers. The desire to communicate broadly and to establish networks beyond the local community led to an investment that would allow the author to circulate identical copies to peers and to keep his original notes.

An Anatomy of Song Printed Notebooks As related in the previous section, the growing numbers of notebooks written during the Song period, the references in the titles of such works to a form of writing that consisted of the gathering of 12. See, for example, Hong Mai, Rongzhai suibi, 4.53, 8.101–3, 9.114, 13.176; xubi 6.284, 8.318; sanbi 6.497–98, 9.532, 10.547–48, 15.608; sibi 3.659, 4.677–78, 8.720, 9.734; wubi 10.949–50.

290

imperial information networks

miscellaneous notes, and the formation of a set of characteristics described in their paratexts suggest that notebooks had turned into a genre in the course of the Song dynasty.13 In order to get a better understanding of how this genre developed, who the notebooks’ authors were, where they hailed from, and when the genre began to attract a wide readership, I have collected a data set of Song printed editions of notebooks. My exclusion of manuscript editions, noted in the opening section of this chapter, should be taken into account when evaluating the conclusions drawn from this data set. The focus on printed editions has allowed the identification of which social actors and institutions invested the most in the circulation of notebooks, and where, when, and why they did so. On the basis of the history of notebooks traced below I will also propose that a new information regime emerged in the course of the eleventh and twelfth centuries. This regime was characterized not only by the literati’s adaptation of court genres such as archival compilations, newssheets, gazetteers, and maps, but also by the expanded publication and circulation of literati collections (besides notebooks, also collected writings [wenji 文集]) that operated on the margins of officialdom but also usually reflected back on central and local government.

The Sa mple At the core of the database on which the calculations and correlations in this chapter are based is a set of 121 records. Each record in this set contains a variety of data on one Song printed edition of a biji title. It contains records only for editions that I have personally consulted or that I have seen described in secondary materials as a Song print;14 copies are not entered separately (in cases where mul13. For the original and longer version of this section see De Weerdt, “The Production and Circulation of ‘Written Notes’ (biji).” Several arguments summarized here are based on tables that are included only in the longer version. 14. De Weerdt, “Biji in Print Database.” The data set benefited greatly from the work of Ming-sun Poon (“Books and Printing in Sung China [960–1279]”) and Liu Lin and Shen Zhihong (Xiancun Song ren zhushu zonglu). Not all the editions listed in these two bibliographies were included because some entries did not constitute separate editions



the notebook phenomenon

291

tiple copies of Song prints do not provide evidence of separate editions, only one record exists in the database). Admittedly, some of the editions entered here are reproductions of other editions, as printers oftentimes based a new set of blocks on older blocks or older printed editions. The difference between an edition in the stricter sense of a unique rendering and a (re-)production is in many cases impossible to tell. The difference is immaterial for the purposes of this chapter, as I am principally concerned with the frequency and geographical origin of productions rather than a textual history of editions. The data set includes both extant Song prints and those titles for which contemporary or later evidence attests that one or more editions were printed between the tenth and thirteenth centuries. I am employing a narrow definition of biji, including notebooks that report stories, contain reading notes, and record conversations on a range of topics but excluding fictional narratives and reporting genres such as historical and geographical memoirs, encyclopedic works, envoy reports, and doctrinal records of conversations. The 121 records in the data set refer to 83 unique titles; for 53 of those titles Song printed editions are extant.

Tempor a l a nd Spati a l Distr ibutions This sample demonstrates that the vast majority of printed Song biji first appeared in print during the Southern Song period (1127–1279). At least 115 editions in the database date to this period whereas only two can be traced back to the first half of the Song; the date of printing for the other four is ambiguous. Further analysis of the time of printing by century shows that only one title may have appeared in print during the eleventh century; all other titles were printed in the twelfth and thirteenth centuries. Even though the precise century of printing cannot be determined for twenty-two titles (most of which can be categorized as Southern Song prints due to material characteristics, author and completion dates, or and others were based on inconclusive evidence. The database is under revision and as this book went into press included 121 editions.

292

imperial information networks

NUMBER OF PRINTED EDITIONS

80

70

60

50

40

30

20

10

0

Unknown

11

12

13

CENTURY OF PRINTING

Fig. 6.1 The number of Song biji printed editions, by the century in which they were printed.

prefatory materials), production in print more than doubled in the last half of the Southern Song period (fig. 6.1).15 The increase in the printing of biji titles in the thirteenth century is in part due to the appearance of biji titles in series. One such series, Zhuru mingdao 諸儒鳴道 (All scholars advocating the Way), first appeared in print in the second half of the twelfth century and featured three titles that are included in the database; the reprint of this series in 1235, as well as the publication of two other series, Ruxue jingwu 儒學警悟 (Insights into the learning of scholars) in about 1201 (with six biji titles) and Baichuan xuehai 百川學海 (All streams of the sea of learning) in about 1273 (with twenty-six), added significantly to the circulation of biji titles in print. However, more than half of the titles produced in print during the thirteenth century (thirty-eight at least) were produced as individual titles, a number that by itself also exceeds the total number printed during 15. Here and elsewhere throughout this chapter thirteenth-century editions include only Song editions printed before the end of Song rule in 1279.



the notebook phenomenon

293

the twelfth century. This number is a minimum, since a substantial number of the editions for which a century cannot be determined most likely date to the thirteenth century. A closer examination of those editions for which a definite or approximate year (or year range) of printing exists further suggests that the century between 1150 and 1250 witnessed the greatest expansion.16 The growth in the final three decades of the Song period was almost exclusively due to the publication of All Streams of the Sea of Learning.17 The growth in absolute numbers of printed editions in the thirteenth century was not a direct reflection of growth in the number of biji written during this period. The correlation of the century of completion with the century of printing for all printed editions demonstrates a difference in the interests of twelfth- and thirteenth-century printers.18 Twelfth-century printed editions were mostly of twelfth-century works, while thirteenthcentury printers diversified by publishing eleventh- and twelfthcentury works in addition to more recent titles. The search for and printing of early Song memoirs written by high court officials, which, as we will see further below, lies behind these numbers, coincided with a more general interest in “the models set by the founding fathers” (zuzong zhi fa 祖宗之法).19 This interest was, however, conditioned by the abiding need to keep abreast of current affairs and new trends in literati culture. The share of titles produced after the fall of Kaifeng in 1127 within the total number of Southern Song printed editions listed in table 6.1 is indicative of Song literati appetite for up-to-date information. 16. Timelines of datable biji printed editions for the periods 1150–1250, 1050–1150, and 1200–1280 (figs. 1–3) are included in De Weerdt, “The Production and Circulation of ‘Written Notes’ (biji)—Tables and Figures.” 17. Ibid., fig. 3. 18. I am here considering only printers, not editors or publishers in the broader sense. We often do not know who the editors were; publishers would encompass a larger category of people including manuscript publishers who are not considered here. I discuss manuscript publishers in De Weerdt, “Continuities between Scribal and Print Publishing in Twelfth-Century Song China.” 19. For a discussion of this concept and its use in both Northern Song and Southern Song politics, see Deng Xiaonan, Zuzong zhi fa.

294

imperial information networks

Table 6.1 Southern Song imprints of biji, by first publication date of the title Tang 2 Five Dynasties 4 Northern Song 27 Northern Song–Southern Song transition 6 Southern Song 75

The place of printing can be established for about half of the titles in the 121 records in the data set, that is, 59 editions out of 121. Excluding those editions for which only the province/route is given (six records), map 6.1 plots all the prefectures in which biji were printed in Song times. The map illustrates that some of the conclusions that have been drawn for printing in the twelfth and thirteenth centuries more generally also apply to the printing of biji: printing occurred in a relatively large number of prefectures (over twenty). Even though there is a clear concentration in the coastal routes of East and West Zhe 浙, Huainan 淮南 East and West, Jiang­nan 江南 East and West, and Fujian 福建, there are also some examples of the printing of biji in more outlying and less welldocumented jurisdictions in Jingxi 京西 and Guangnan 廣南. The edition of Shen Gua’s 沈括 (1031–95) work compiled and printed by the Jinghu South Route Finance Commission is not included;20 the region between Jiangnan and Guangnan therefore appears as a blank on the map even though some activity is attested here as well. This map further highlights a key difference in the role of commercial and other types of printers in the dissemination of biji: commercial printers of biji were concentrated in only two prefectures (the well-known printing centers of Lin’an 臨安 [Hangzhou 杭州] and Jianning 建寧 [in Fujian]) where they published relatively large numbers in the genre.21 In sum, the printing of notebooks took off 20. See note 25 below. 21. The activities of the commercial printers of Hangzhou, Jianning, or Chengdu have been well documented in recent work on the history of book production in Song times. Very useful studies include those by Poon, “Books and Printing in Sung China”; Zhang Xiumin, Zhongguo yinshuashi; Chia, Printing for Profit; and, Edgren, “Southern Song Printing at Hangzhou.”

Map 6.1 Southern Song prefectures where biji were printed.

296

imperial information networks

between 1150 and 1250, predominantly but by no means exclusively in the southeast, and thus testifies to a more general rise in the printing of court and literati genres in the south following the Song loss of the northern territories.

The Pr inters Who participated in the printing of notebooks? What motivated different kinds of printers and what kinds of notebooks did they publish? What kinds of relationships and networks were mobilized to see notebooks through to print publication? From the fifty editions for which we have some information regarding the type of printer, it appears that governmental institutions and officials played a substantial role in the dissemination of biji in print (fig. 6.2). This conclusion is also borne out in map 6.1, as governmental printers were mostly responsible for the printing of biji outside of the well-known printing centers of Hangzhou and Jianning—and even in these jurisdictions governmental agents participated in the production of biji. Before moving on to a discussion of the role of the government (local governments in particular) in the dissemination of notebooks, it is important to note that the boundaries between the different types of printer (commercial, government, and private or “household”) were blurry.22 Even in cases where the name of the printer is beyond doubt (e.g., in the case of Lu Ziyu 陸子遹 [fl. ca. 1220] or Yue Ke 嶽珂 [1183–1240], both of whom printed multiple biji titles and other works in addition), scholars have differed in their classification of them as either government or private printers. While both Lu Ziyu and Yue Ke have been profiled as private “household” printers, there is no doubt that in Lu Ziyu’s case all titles were printed during his tenure in office and on government premises. In Yue Ke’s case all titles were printed in Jiaxing 嘉興, where he held office and around the time of his tenure there. He, moreover, refers to the district school as the location of printing in a preface to one 22. For further discussion on this see McDermott, “From Collection to Archives.”



the notebook phenomenon

297

PRINTER TYPE

NUMBER OF PRINTED EDITIONS

80

70

57%

60

50

40

30 19% 20

15%

10

0

3%

Unknown

Private

Possibly Private

3%

2%

1% Commercial

Possibly Commerical

Government

Possibly Government

Fig. 6.2 The distribution of Song biji editions, by type of printer.

of the titles (not a notebook) he published there.23 Regardless of whether private funds were involved, holding office enabled several authors to see their work into print. Besides Liu Changshi, who used his salary to have his work published in county premises, Zhao Yanwei 趙彥衛, the author of Yunlu manchao 雲麓漫鈔 (Unrestrained jottings from the Cloudy Mountain Foot), also admitted in 1206, in a preface to a reprinted and augmented edition of his written notes, that he had taken advantage of his assignment to the magistracy of Huizhou to have his work printed. Like Liu Chang­shi, he justified his decision by pointing to the demand for it; quite a few interested men had come to ask for a copy of an earlier edition he had printed, but since the demand outstripped the supply he had a new edition cut in the Huizhou prefectural school. This earlier edition too had been printed in a prefectural school while he was serving as prefect of Suizhou 隨州 in 1202–4.24 23. See his preface to Eguo Jintuo cuibian 鄂國金陀粹編 (The best of Yue Fei), quoted in Wang Guiping, Jiakeben, 74. 24. Zhao Yanwei, Yunlu manchao, xu, 2.

298

imperial information networks

I have classified all the titles in the Biji in Print Database that were printed in government premises (yamen and schools) as government prints; in cases of doubt (such as in the case of Yue Ke) the printer type has been left blank. This uncertainty in part reflects the fact that scholar-officials could at different stages in life be private and government printers and that government prints in many cases reflect the interests of individual magistrates or of their associates and acquaintances. The differentiation between commercial, government, and private printers is also primarily based on the status of the printer involved rather than the printer’s motivation. Commercial interest may have been a motivation for government and private printers—at least in one case a preface writer wrote, on the occasion of the reprint of Shen Gua’s Mengxi bitan 夢溪筆談 (Notes by Dream Creek), that the prefecture had benefited financially from the first edition.25 The breakdown of the total number of government imprints by subtype of government printer reveals some interesting particularities of biji prints when compared to other categories of printed books. The omissions in figure 6.2 are especially noteworthy. Printed editions of biji were virtually never issued by the top govern­ mental printing agencies in the empire. The Directorate of Education (Guozi jian 國子監) cannot be associated with any of the printed editions on record; the route-level agencies charged with the supervision of regional and local printing were only exceptionally involved in the compilation and printing of a notebook.26 In contrast, the vast majority was published at the prefectural level (mostly in prefectural schools) and only an occasional title was published in a county school not identified as the main prefectural educational institution. Figure 6.3 shows that the number of titles printed at the 25. Shen Gua, Mengxi bitan, preface by Tang Xiunian 湯脩年 (1166), 1087. 26. In a preface to a printed edition of Hong Mai’s Rongzhai xu bi 容齋續筆, He Yi 何異 mentions that he had a selection of passages from Hong Mai’s Yijian zhi 夷堅志 compiled and printed in the “Financial Office of South of the Lake” (Huyin jitai 湖陰之 計臺); this is most likely a reference to the Fiscal Commission of Jinghu South (Jinghu nan lu 荊湖南路). Hong Mai, Rongzhai suibi, preface by He Yi, 979–80. He Yi once held the position of fiscal supervisory official of Jinghu South (Hunan zhuanyun panguan 湖南轉運判官). SS, 401.12166.



the notebook phenomenon

299

NUMBER OF PRINTED EDITIONS

80

70

60

50

40

30

20

10

0

11

Unknown

12

13

CENTURY OF PRINTING

County School

County

Route

Prefectural School?

Prefectural Prefectural Prefecture? School Educational Officer

Prefecture

Unknown

Fig. 6.3 The number of government imprints of Song biji, by century of printing, showing the subcategories of government printers.

local level continued to increase in the course of the thirteenth century while the share of government imprints as a percentage of total output remained stable during this time. Why did local governments, often strapped for cash and unable to fund public welfare projects, invest in the printing of works that were often presented as random collections? Unlike encyclopedias, literary anthologies, or collected classical commentaries these works could not be marketed for their comprehensive and systematic coverage of topics relevant to students preparing for the examinations—a substantial component of the book-reading public in Song times. Notebooks also differed from collected writings (wenji 文集), the complete oeuvre (excluding monographs) of political and cultural icons and increasingly of lesser-known literati, histories, medical treatises, anatomies of natural and cultural objects ( pu 譜), or geo-

300

imperial information networks

graphical works in their casual rather than focused discussion of topics. All of these latter topics belonged to the repertoire of interests of the cultured male in twelfth- and thirteenth-century China. Notebooks resemble in some ways modern-day blogs; they freely interact with a wide range of previously published and unpublished materials, intend to keep readers and interlocutors current, make no claims to present material exhaustively, but offer running commentary on and corrections of others’ writings and comments on a selective range of topics. The continuation of local government activity in the printing of notebooks fits in with the recent revaluation of the active role of local governments in local welfare throughout the Southern Song period, a time that had heretofore been more commonly characterized as the era of local elite activism.27 From the prefaces written for government editions it emerges that the printing of selected works and the maintenance of woodblocks had by the twelfth and thirteenth centuries come to be regarded as part of the responsibilities of the prefect. The reprint of All Scholars Advocating the Way in Shaoxing in 1235 was based on old blocks. Huang Zhuangyou 黃壯 猷 wrote that he had found the old blocks and brought in carvers to restore them so that new copies could be printed. Prefectural schools served as both the place for the production of printed books and as a depository for blocks; this dual role is reflected in the verbal phrase “to carve and to deposit” (kanzhi 刊置; kezhi 刻置), which was typically used in combination with references to official premises such as the prefectural school.28 Prefects and others attached to the local bureaucracy turned to publishing for a variety of reasons. Tang Xiunian 湯脩年, professor in the Yangzhou Prefectural School, wrote that an earlier print of Shen Gua’s Notes by Dream Creek had benefited the prefectural coffers, but he noted that little benefit had then accrued to the school. 27. Lee Sukhee’s dissertation (“Negotiated Power”) and book of the same title, Negotiated Power, critically evaluate the work of Robert Hymes and others on the turn from state to local elite strategies and underscore the participation of local magistrates in community projects in Mingzhou. 28. See, for example, Pang Yuanying, Wenchang zalu, 1.



the notebook phenomenon

301

The reprint appears to have been intended to rectify this situation; the project was now designed to benefit scholars both financially and intellectually.29 The storage and provision of desirable titles were also seen as activities that would attract literati and thus place prefectural institutions of learning at the center of both local and broader cultural exchange. Local teachers were actively engaged in the procurement and preparation of titles, just as they were in the preparation and dissemination of maps (see chapter 3). Chen Yingxing 陳應行, who had been teaching in Quanzhou, was actively engaged in seeing the work of Cheng Dachang 程大昌 (1123–95) into print. He tried to have Cheng Dachang’s Yugong lun 禹貢論 (Discourses on “The tribute of Yu”) printed on prefectural grounds, and later also promoted the publication in print of Yanfan lu 演蕃露 (Elaborating on “Rich dew”) after having approached an intermediary close to Cheng Dachang for copies of this and other work by him.30 Titles were selected through local connections and on the basis of their relevance to literati interests. Local connections were in the case of these projects usually defined by peer and family relationships. Chen Yingxing had come across Cheng Dachang’s work when the latter was serving as prefect in Quanzhou (1180–82). He admitted that he had not been able to meet with the prefect as Cheng Dachang had been too busy with administrative matters. Chen Yingxing therefore went about the business of seeing the prefect’s work into print through the intervention of an intermediary, a fellow jinshi graduate [tongnian 同年]. Peer relationships among successive officials in the same locale were also important in the completion of printing projects. A Song printed edition of Mizhai biji 宻齋筆記 (Mizhai’s written notes) came with prefaces written by one prefect and two vice-prefects of Fuzhou 撫州. Similarly, He Yi 何異 ( j. 1154), who had printed an edition of Hong Mai’s Yijian zhi 夷堅志 (Record of the listener) as financial commissioner and 29. Shen Gua, Mengxi bitan, preface by Tang Xiunian. 30. Cheng Dachang, Yanfan lu, preface by Chen Yingxing; Cheng Dachang, Yugong lun, preface by Peng Chunnian. The first item is included in the series Ruxue jingwu; the second item exists in a stand-alone Song prefectural edition printed at the Quanzhou prefectural school. Both editions are held at the Beijing Library. The first preface is also included in Quan Song biji, ser. 4, vol. 9, 158.

302

imperial information networks

who endorsed a reprint of the same author’s Rongzhai suibi 容齋 隨筆 (Rongzhai’s jottings) with a preface, wrote that he had served in the same prefecture where Hong Ji now proposed to reprint the work of his grandfather’s brother. He also referred to friendships with two of Hong Mai’s descendants, whom he credited for having given him access to Hong Mai’s work in the past.31 As Hong Ji’s printing of a deceased relative’s work suggests, family relationships were instrumental in the printing of notebooks. Xie Caibo’s 謝采伯 Mizhai’s Written Notes had been printed privately, but as the font size used in this edition was deemed to be too small his son endeavored to have a new edition printed in the prefectural school. He showed copies to two colleagues and thus obtained support for the government edition.32 Local magistrates and their personnel thus printed their own work and that of family members and acquaintances, but they also positioned themselves as crucial partners to local elites by promoting the work of local luminaries. In 1134, while serving in Tong­ xiang 桐鄉, Zhou Tingyun 周庭筠 contributed from his own salary to print a notebook by Li Ruqian 李如篪. Li Ruqian completed his notebook while living a life of leisure there in 1132. Titled Dongyuan congshuo 東園叢説 (Ongoing conversations from Master Dongyuan), it contained in the author’s own account a digest of notes he had accumulated about subjects ranging from the classics, the histories, philosophical works, and prose collections to astronomy and geography. He underscored that it would be of use to learned gentlemen, and Zhou Tingyun shared this hope but rather emphasized the fact that Li Ruqian had formerly stayed at the Imperial College and was on familiar terms with all those who had been top of the class in the capital at the time.33 Local connections clearly mattered, but the connections involved in printing projects also transcended local boundaries. When Hong Ji, for example, brought up benefit to a local readership as the rationale for printing Rongzhai’s Jottings, another contributor to the 31. Hong Mai, Rongzhai suibi, preface by He Yi, 979–80. 32. Xie Caibo, Mizhai biji, preface by Wang Zongdan. 33. Li Ruqian, Dongyuan congshuo, prefaces by Li Ruqian and Zhou Tingyun.



the notebook phenomenon

303

project countered with the claim of its potential greater impact on the scholar-official class across the empire. In his 1212 preface, He Yi quoted from the letter in which Hong Ji asked him for an endorsement of his project: It has been over forty years since my grandfather’s brother, Mr. Wenmin, came to this place following his tenure as imperial diarist. I am very fortunate to have followed in his footsteps after a long while. While administrative business was slow, I took Wenmin’s random notes, starting from the four installments of sixteen juan each, and ending with the fifth installment, his last written work, which is only ten juan, and had them all printed in the prefectural school so that I could show them to the people here [bangren 邦人]. I imagine capturing a manifestation of him, just as if he is here. Could you inscribe it for me?

He Yi added, “Through this work one can investigate precedents, broaden one’s experience, verify errors, and lubricate the tip of one’s brush [enhance one’s writing]. This is truly an opportunity for scholars to improve their learning. How could it only be for the consolation of the people of Ganzhou upon his departure?”34 Qiu Su similarly welcomed the printing of this work as a benefit to all.35 The work of Hong Mai and others was printed with local government support for its presumed relevance to all scholars. Notebooks addressed a key question in twelfth- and thirteenth-century society; they provided different templates of what defined the male scholar. By capturing the sayings, readings, and experiences of individual men, they portrayed how these men’s lives ought to be perceived and provided a framework for others aspiring to literati status. Hence, the printing of Hong Mai’s notebook was not only an act of commemoration of the man in real life but also a source of emulation and a reference for further conversation for all aspiring to scholar status. Even though the kind of broad scholarship and critical examination reflected in the notebooks of Hong Mai, Chen Shan, Liu 34. Hong Mai, Rongzhai suibi, preface by He Yi, 979–80. 35. Ibid., preface by Qiu Su, 980–81.

304

imperial information networks

Changshi, and Zhao Yanwei was a prominent trend in the twelfth and thirteenth centuries, different definitions and additional aspects of literati status were highlighted in other notebooks. Earlier notebooks tended to be focused on court life and thus captured political behavior at the highest levels, and other works, such as Le shan lu 樂善錄 (Record of finding joy in goodness, ca. 1164) and Zijing bian 自警編 (Collected notes on self-vigilance, ca. 1224) were promoted for their relevance to the moral development and broader social responsibilities of the literati and their families. Moreover, as Wang Zongdan 王宗旦 underscored in his preface to Mizhai’s Written Notes, writing and leaving a legacy of learning that can be transmitted to family and the broader elite was the ultimate litmus test for the scholar-official: “When it comes to the hobbies of elites [shidafu] in later life, there are few who do not lose sight of their initial intent. When we read Mizhai’s own preface [to this work], it says that he leaves this book to his sons and grandsons. From this we can tell that he did not abandon learning in old age.”36 The promotion of the utility of a work to all scholar-officials involved an intervention on the part of men other than the original authors. In some cases selecting a work for publication involved censorship. During his tenure as military commissioner in Sichuan (1189–92), Jing Tang 京鏜 (1138–1200) supported a prefectural edition of Wu Zeng’s 吳曾 Nenggai manlu 能改漫錄 (Nenggai’s unrestricted records). Jing Tang had no apparent direct connection to Wu Zeng or his son Wu Fu 吳復, who had helped compile the work in the 1150s.37 He recommended the work to readers interested in categorized affairs (haoshizhe) for its breadth but added that some entries had been cut. The cuts were necessary because “in the past enemies [of the author] had picked one or two items that should not have been included and argued that these were things that were inappropriate; subsequently the work did not circulate. . . . Now we 36. Xie Caibo, Mizhai biji, preface by Wang Zongdan. 37. There is an indirect tie between Jing Tang and Wu Zeng, who were both connected to Lou Yue 樓鑰 (1137–1213). This connection was established through a query designed by Michael Fuller included in CBDB.



the notebook phenomenon

305

have eliminated what should not be included and we preserved what should not be discarded.”38 Wu Zeng’s well-documented support for Qin Gui 秦檜 (1090–1155) may have been the reason why his work was no longer deemed acceptable after the death of the grand councilor. Local governments maintained their share in the total number of printed editions (about 25%) throughout the twelfth and thirteenth centuries. In contrast, the number of commercial imprints rose significantly in the course of the thirteenth century. Thirteenth-century commercial imprints take up at least 60 percent of the total number of commercial imprints known to have been published in Song times (this number would be higher were we to exclude those imprints for which the century of printing is unknown; the twelfthcentury number also includes one edition of a notebook that was included in a printed edition of a collected oeuvre).39 Most of these imprints have been attributed to two commercial establishments: the printing shops of Chen Daoren 陳道人 and the Yin 尹 family in Lin’an. According to Søren Edgren,40 both were active around the same time, in the last half of the Southern Song period. A small number of biji titles have also been attributed to unidentified commercial publishers in the printing center of Jianning. The growth in the commercial printing of biji titles (a number that does not include the larger series that cannot be reliably attributed to commercial printers) and their marketing in major bookshops of the capital further attests to the growing appeal notebooks exerted among a reading public eager to become and remain up-to-date and interested in the sharing of stories and anecdotes about ordinary and not-so-ordinary lives. The small number of private imprints across time may suggest that commercial imprints lagged behind 38. Wu Zeng, Nenggai man lu, preface by Jing Tang. 39. Su Shi’s Dongpo zhilin was also included in what is presumed to be a twelfth-century commercial edition of his collected works. See An Ruixuan, “Songren biji yanjiu,” 83–89. See also Huiren, “Songkan Su Shi quanji kao.” 40. Edgren (“Southern Song Printing at Hangzhou,” 36–38, 73) lists only one biji title for the Chen print shop; his list excludes several editions (of Huizhu lu 揮塵錄 or Rongzhai suibi 容齋隨筆, for example) that are no longer extant but that are mentioned in catalogs.

306

imperial information networks

both government and private imprints in the twelfth century and that commercial publishers only became a significant supplier in the thirteenth century. However, given the small number of imprints that can be traced to private printers (five only, with one uncertain) and the uncertainty surrounding the vast majority of Southern Song titles, it is too risky to draw firm conclusions about the role of private printers in the dissemination of biji in print. As I have argued in more detail elsewhere,41 commercial and government printers differed only marginally in the topics most covered in the notebooks they printed. Minor differences also applied over time. The history and the current affairs of the reigning dynasty are the top primary subject category in both twelfth- and thirteenth-century printed editions; philology, learning, and storytelling remain top areas of interest as well, even though their relative ranking changes over time with storytelling declining and learning rising in interest in the course of the thirteenth century. Both commercial and government printers published notebooks focusing on court history and politics, philology, storytelling, and learning. However, when commercial printers began to market biji they were more likely to publish on court history and politics and showed relatively little interest in those titles mainly devoted to philological commentary and learning. This state of affairs could have resulted from a greater willingness among commercial printers to invest in more sensitive political commentary that would sell well among the book-buying public in the capital, which included not only the resident court and central government bureaucracy but also visiting regional and local officials, and, once every three years, candidates sitting for the metropolitan and palace examinations. Private printers appear to have been interested in notebooks on court history politics and philological questions, but the very small numbers we have available for private printers raises an insurmountable barrier for a further anatomy of private printing. In sum, during the twelfth and thirteenth centuries prefectural governments and increasingly commercial printers printed notebooks for literati readers across the Song Empire. The notebooks 41. De Weerdt, “The Production and Circulation of ‘Written Notes’ (biji).”



the notebook phenomenon

307

they chose to print covered roughly the same topics. I have given some reasons, based on a close reading of prefaces and postfaces, for the conspicuous role of prefectural schools and prefects in the dissemination of notebooks (their financial interest, the cultural standing of the prefecture and its contribution to literati culture, prefects’ new understanding of their administrative duties, and family obligations). Further explanations for local government investment in literati notebooks will emerge from the analysis of their authors’ social backgrounds below and the closer reading of Wang Mingqing’s notebooks in chapter 8.

The Sociopolitic a l Backgrounds of Notebook Authors The connection between notebook writing and officialdom can be explored through the extensive data on office holding made accessible through the China Biographical Database Project (updated edition dated July 5, 2011; CBDB). What percentage of notebook writers had served in officialdom? What percentage had served in high officialdom and may have acquired information about the court and high officialdom firsthand? How did those who served in office obtain their posts? Were family connections paramount or were biji authors among the substantially increased number of Song officials who entered officialdom through the civil service examinations? The questions that can currently be answered about career backgrounds are thus limited by the structure and contents of CBDB. Moreover, the CBDB data are by no means exhaustive in their coverage.42 By comparing the results obtained through the interlinking of Song Biji in Print and CBDB to a similar investi­ gation and by interpreting the results in light of known deficiencies, 42. For an earlier assessment, see De Weerdt, “Mapping Communication from Mingzhou: Networks of Correspondence.” This is a report presented in a workshop on the utility of the database for social science research. The database has been substantially updated since this report was first written in July 2007 and now includes, for example, data drawn from the most important biographical dictionary and index of Song persons, Wang Deyi’s Songren zhuanji ziliao suoyin.

308

imperial information networks

2 4%

Degree Examination

9

6

16.5%

11%

Decree of Special Grace

Regular Examination

Yin Privilege

37 68.5%

Fig. 6.4 Methods used to qualify for officialdom among Song biji authors, by percentage. Absolute numbers are listed above the percentages on the chart. Methods used only in one case (including religious qualifications and recommendation) are not included (five instances); pre-Song authors (five in total, with qualifications mentioned in only one case, regular and decree examinations) are also excluded. Authors could use multiple avenues to officialdom; in some cases the figure therefore includes multiple methods utilized by the same author.

I hope to offer some sound observations regarding career patterns of notebook authors. The results presented below suggest, on the one hand, that the backgrounds of notebook authors may not have been as one-dimensional as the high profile of the authors of the better-known Song notebook authors suggest; on the other hand, the prosopography of the authors also demonstrates that notebook authors frequently maintained close ties to officialdom and increasingly belonged to the lower-level elite, who either held low-level bureaucratic positions or aspired to literati status. Figure 6.4 shows the methods of entry for forty-seven out of a total of seventy-three unique authors (no entry information was



the notebook phenomenon

309

Table 6.2 Zhang Hui’s data on the official careers of Song biji authors Number of Official Jinshi Songshi biji included rank degree biography Northern Song Northern Song–   Southern Song Southern Song Total

40

30

23 (22+1)

25

20 67 127

10 40 80

6 (5+1) 17 (14+3) 46 (41+5)

2 10 37

Source: Zhang Hui, Songdai biji yanjiu, 47.

available for the remaining authors in CBDB; in total sixty-eight out of seventy-three authors are included in CBDB). It shows that, at the very least, nearly half (thirty-six) of all authors of printed biji passed some form of civil service examination; the vast majority (thirty-two) did so through the regular jinshi examinations. The chart further shows that the two other principal routes to acquire the prestige and power attached to official rank, inheritance and recommendation, played a relatively minor role both in absolute numbers and as a percentage. Moreover, when these data are plotted along a timeline based on the year in which the qualification was obtained,43 it appears that a jinshi degree, even though the principal qualification for notebook authors throughout the Song period, became a less prevalent attribute of notebook authors during the last century of Southern Song rule. When compared to the data Zhang Hui compiled for the authors of 127 Song notebook titles (table 6.2),44 the data presented above support the conclusion that a larger number of authors obtained jinshi degrees during the Northern Song period. Given the data in this table I would also propose that, in contrast to Zhang Hui’s data, which show the percentage of men entering officialdom through the examination route declining by over 30 percent, authors 43. De Weerdt, “The Production and Circulation of ‘Written Notes’ (biji).” 44. This is out of a total sample of 155 biji titles—the sample data are not differentiated by method of production and include some genres such as geographical and historical memoirs not included in my sample.

NUMBER OF AUTHORS

310

imperial information networks 30 24

25

20

15

10

7 5

5 1 0

Tang

2 Five Dynasties

Northern Song

Northern SongSouthern Song

Southern Song

CENTURY OF PRINTING

Fig. 6.5 The number of authors of biji printed in Song times for whom no entry data exist in CBDB.

whose notebooks appeared in print continued predominantly to acquire positions in officialdom through the examinations. However, this finding may be less significant in view of the fact that the majority of authors for whom we do not have further information were also writing in Southern Song times (fig. 6.5). This conclusion coincides with Zhang’s that the percentage of authors for whom we find no indication of official service steadily rises from the 1130s onward. This may be read as a sign of a transition in the social background of notebook writers: from high-ranking officials to lower-ranking officials and those without official rank or even an examination degree. Did the authors who acquired official rank hold particular types of positions that may have made their observations more relevant to their contemporaries? Does their office holding follow the trend from high offices to regional and local ones suggested above, and does such a trend bring about changes in subject matter? CBDB contains 567 posting records for fifty-three authors; for all but eight of them it includes data on multiple postings. Further analysis of the individuals holding court and lower-level posts lends support to the view that notebook authors with top-ranking positions and years of experience at court had been predominantly serving the



the notebook phenomenon

311

Northern Song court. Notebook authors who held court positions during the Northern Song period obtained at least five court offices in 75 percent of the cases (twelve out of sixteen), whereas only 33 percent of Southern Song notebook writers (seven out of twenty-one) held on to court positions at the same rate. Even though it is likely that more notebook authors in the twelfth and thirteenth centuries did not serve in officialdom (see above), a comparison between those serving at the Northern Song court and those serving the Southern court further suggests that the absolute number of printed authors with experience at court was higher in the Southern Song than in the Northern Song (twenty-one vs. sixteen)—the hypothesis of a shift from court to local officeholders and non-officeholders should thus be modified. Similarly, in the eleventh century, notebooks were authored by men with regional and local office-holding experience, albeit to a lesser extent than in the following centuries. Rather than a shift, we appear to be witnessing a broadening of the sociopolitical background of notebook authors. Most individuals with court positions had held lowerlevel positions at some point.45 All but one of the Northern Song authors who had held lower-level positions went on to higher-level positions (suggesting the dominance of high court officials), but nearly one-third of Southern Song authors who had held local office (eight out of a total of twenty-six) do not appear to have secured more prestigious positions at court, suggesting that lower-level officials as well as non-officeholders were increasingly represented among those authors who had their notebooks printed. However, from the topical coverage tabulated in the previous section we can also infer that a higher number or a higher percentage of authors whose official careers may have been halted at the local level does not directly translate into a shift from court politics to local concerns. 45. Tables showing numbers of court, regional, and local offices are included in De Weerdt, “The Production and Circulation of ‘Written Notes’ (biji).” Tables showing the number of regional and prefectural postings of printed biji authors by type of posting and tables displaying the number of court postings by type of posting for Northern Song printed biji authors, Southern Song printed biji authors, and those active during the transition from Northern to Southern Song are not reproduced here but can be obtained from the author. The postings data can also be accessed on the site accompanying this book.

312

imperial information networks

The Overlapping Networks of Scribal and Print Publishing The above analysis of the authors of notebooks printed in Song times has led to the conclusion that their sociopolitical background diversified in the course of the twelfth and thirteenth centuries and that authorship became an attribute of those at the lower echelons of the scholar-official class. How did those of lesser fame see their work through to publication? What kinds of relationships and networks facilitated publication in manuscript and print? Did authors whose reading and conversation logs were printed participate in different kinds of social networks than those whose works belong to the majority of titles for which no printed editions are on record? Did publishing in manuscript and print carry similar or different meanings? I pursue these questions on the basis of an analysis of the publication history of Wang Mingqing’s serially published notebook Waving the Duster, tracing how author, users, and eventually entrepreneurs published successive editions in the course of one century and underscoring the important role of user publication in this process. Following Harold Love,46 I distinguish between different modes of scribal publication (author, user, and entrepreneurial publi­ cation). In twelfth-century Song China, readers, who had already been involved in the publishing of manuscript texts, also turned into publishers of print texts. In Love’s definition, user publication starts with “the edition of one,” the copying of a text for private use and its inclusion in a commonplace book or personal miscellany. Personal copies were the starting point for further transmission. Friends or acquaintances shared copies more widely either through chain copying or by sending copies to their network.47 Scribal or manuscript publishing is thus used here in the sense of admitting a text to circulate within a social circle so that it ceases to be the private possession of a closed group of individuals, rather than in the 46. Love, The Culture and Commerce of Texts. 47. Ibid., esp. 79–83.



the notebook phenomenon

313

sense of producing large numbers of copies. The early history of Chinese printing suggests that author, user, and entrepreneurial publication operate in both the scribal and print medium, and thus prompt the question whether the relationships among these different modes of publishing change in an ecology where print becomes a widely available option. Again following Love, we may test the hypothesis of whether the continued use of an older medium at a time when the more public medium of print is in ascendancy should be read as a challenge to the prevailing hierarchy of modes of communication and, ultimately, to the legitimizing fictions of the state (and, we could add, of cultural elites).

Author Public ation In 1166 Wang Mingqing completed what would become the first in a series of four notebooks. In the short preface appended to this first compilation of notes, titled Waving the Duster, Wang noted that it was the product of many conversations with friends and relatives who came to visit during the many days of leisure he enjoyed while living in Kuaiji 會稽 (now Shaoxing 紹興). Like authors of notebooks before and after, Wang added that he kept track of the more interesting discussions and verified the details in his notes by consulting various accounts. As the notes accumulated he could not bear to discard them and decided, benefiting from a friend’s insistence, to edit and circulate them in an edition of one hundred entries. Almost thirty years later, in 1194 (or 1193), 1195 (or 1194),48 and 1196– 97, this first compilation was supplemented with three other installments of around two hundred, fifty, and one hundred entries respectively. Up until the late thirteenth century Waving the Duster was printed in at least six editions and in various locations including Hangzhou 杭州, Jianning 建寧, and Jiaxing 嘉興.49 48. Xia Shaohui, “Wang Mingqing Huizhu lu kaoshu.” 49. For the most extensive account of extant and lost Song prints, see Wang Ming­ qing, Huizhu lu, chuban shuoming and Qing bibliographic notes (Zhonghua shuju ed., 364–91). The count includes a minimum of three separate editions of the first three

314

imperial information networks

Wang Mingqing played a key role in the dissemination of his note­book. In the prefaces and postfaces to the early editions of Waving the Duster, he is portrayed as the agent who initiated the transformation of handwritten notes into a compilation with a title. He shared this compilation with friends and acquaintances and entrusted them with the further publication of his work, even soliciting prefaces and other front material for the purpose. He also acknowledged the print publication of the notebook in prefatory material attributed to the author himself. Wang’s brief inscription (shi 識) to the first installment, dated to 1166, explains both how his collection of notes came into being and how it acquired a title and so started its transformation into a published work. The author noted down (bilu 筆錄) those items that were of interest and that had been double-checked, he gathered the handwritten notes in a case, and when the notes had accumulated he transformed them into a book with a title: Waving the Duster. Despite Wang’s remark that “this is not the stuff of a book,” the act of titling his notes and the addition of a dated inscription were the first in a series of paratextual interventions that removed the handwritten notebook from the author’s desk and circulated it to gradually expanding circles of readers.50 In a later full-length postface (ziba 自跋) to the same first installment, written in 1185, Wang Mingqing recounts the origins of the notebook in the same way. He claims credit for the title, but traces the origins of the notebook’s publication to an old friend, Cheng Kejiu 程可久. He describes Cheng as a scholar of some renown. Cheng, then in the position of assistant magistrate (xiancheng 縣丞) of Dexing County 德興 in Raozhou 饒州, liked the notebook very much upon a first reading, copied it himself (shoulu 手錄), and added installments, one extant Song edition of the full series (Longshan shutang), one no longer extant Song print undertaken by the Chen family in Lin’an referred to by Huang Pilie, and a heavily modified edition included in the commercial series Baichuan xuehai printed in the late Song period. Whether the Xiuzhou edition of the first series referred to in a colophon in the Longshan edition refers to the same or a different edition from the one(s) mentioned by Wang Mingqing (preface to the second installment) and Zhao Bujian (comments on the combined edition) is unclear. 50. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.42.



the notebook phenomenon

315

another inscription to the text. From that point onward, Wang Ming­ qing wrote, the notebook circulated (liuchuan 流傳).51 The loss of authorial control hinted at in the brief mention of Cheng Kejiu’s role seems to have been the result and most likely the desired outcome of the author’s attempts to show and present manuscript copies to close friends. Cheng Kejiu noted in his 1169 postface, the earliest dated endorsement of the text after Wang Ming­ qing’s first inscription, that Wang surprised him by getting the work out for him.52 Cheng Kejiu’s endorsement was subdued. He expressed some doubt as to whether this was the best use of Wang Mingqing’s time. Rather than a case of a friend hand-copying and then disseminating the work of an unsuspecting author, the publication of Waving the Duster appears to have been eagerly pursued by the author in two ways. Friends were both coauthors and copublishers of the notebook. As I will argue in chapter 7, and as acknowledged in his prefaces, the text was shaped by a network of interlocutors and contemporary authors who provided the material for Wang’s notes. Some among Wang’s closest friends were also entrusted with the publication of his work. Zhao Bujian 趙不譾, a lifelong friend of Wang Ming­qing, accepted a charge he acknowledged came from the author himself “to let the entire realm share in the knowledge of unfamiliar events.”53 The gift of a manuscript of one’s own work to a friend or an acquaintance, either as an opportunity to copy or in the form of an actual manuscript copy, was in the context of twelfth-century publishing an invitation to participate in its transmission. Close friends showed they knew how to interpret such a gift by circulating and thus publishing it further among other close friends or by making the necessary arrangements for its broader dissemination in print (by having woodblocks cut in the case of Zhao Bujian). Wang Mingqing also explicitly expressed his support for the notebook’s broader dissemination in print beyond the small group of 51. Ibid., 1.43–45. 52. Ibid., 1.42. 53. Ibid., 4.328–30, esp. 330.

316

imperial information networks

initial readers, albeit with conventional expressions of humility. In a preface to the second installment, written in 1194, Wang Ming­qing writes that his notes went into print relatively soon after its initial compilation.54 Authorial initiative and the author’s ability to mobilize personal connections were a crucial factor determining which works would circulate in print. The self-financing of the print production of Notes Written in Lupu and Unrestrained Jottings from the Cloudy Mountain Foot, and the involvement of peers, disciples, and junior family members in the local government production of other notebooks discussed above suggest that authors and their immediate relationships were directly involved in the mechanical reproduction of written notes. Self-advertising in prefatory and other paratextual material, typically attributed to the emerging commercial printing workshops, formed part of the strategy by which aspiring authors sought support for their work. In addition to seeking out friends and acquaintances to collaborate in the circulation of the work, he also invited those friends and acquaintances to assert the significance and use of the notebook in prefaces and postfaces. Besides the prefaces by Cheng Kejiu and Zhao Bujian, some early editions of Waving the Duster featured rather unusual supporting materials such as a personal letter of endorsement for his work from two of the most prominent twelfth-century historians. In his 1185 postface he highlighted his connections to Li Tao 李 燾 (1115–84) and his son Li Hou 李垕 (decree exam 1172; d. ca 1179).55 He presented himself as a disciple of the former and a close friend of the latter and recalled a visit with the elder Li. He recounted that, after he, Wang Mingqing, had presented the eminent historian of Song history with his work on recent Song court affairs, the latter put him to the test by asking him about a series of recent events. Alleg54. Huizhu lu (Zhonghua shuju ed.), 223. In a later preface Zhao Bujian attests to the print circulation of the first installment. In his notes on a Song edition, the Qing bibliographer Zhang Yuanji 張元濟 transcribes an inscription stating that the first series had been printed in Xiuzhou 秀州, the prefecture in which Wang Mingqing’s family resided. These texts are included in Huizhu lu (Zhonghua shuju ed.), 330, 368. 55. Ibid., 1.44–45.



the notebook phenomenon

317

edly, Wang showed he was up to the challenge by providing a full account of each event, over twenty in total. Thereupon Li invited Wang Mingqing to join him at court and become his aide, an offer that Wang declined. In a move that came naturally to the historian aspiring to back up his stories with original primary sources but that stands out as unusual in the world of Song printing, Wang included a personal letter from the younger Li attesting to the invitation and expressing the interest of both men in Wang’s notebook. In this letter, undated but most likely written in the mid1170s,56 Li Hou relates that he had just finished reading Waving the Duster with “the old man” (referring to Li Tao) and that they were both very impressed with it. Li Hou also noted that some of the more extraordinary among his court historian colleagues similarly delighted in it. The letter appears to have been unedited: it contains the more casual opening and closing phrases of informal letters (unlike those contained in printed editions of collected writings) and Li Hou signed in a familiar style as “your covenant brother” (qixiong 契兄). In sum, even though Wang Mingqing did not publish his notebooks himself, he played a pivotal role in their dissemination by mobilizing friends, peers, and former colleagues, who took further steps to publish his work in manuscript and in print.

User Public ation a nd the Net wor k ing Impulse Besides the key role of the author in the publication of the notebook, the prefatory material of Wang Mingqing’s notebooks also provides a glimpse into the networks sustaining both manuscript and print publishing. Close friends were readers and turned into manuscript and print publishers of the text the author shared with 56. Huizhu lu (SBCK ed.), qianlu, 4.12a–b. The letter refers to Li Hou’s service in a national historiography office and must have been written before Li’s death around 1179, a death that Wang Mingqing also refers to. He held the position of compiler and editor for the Dynastic History and Veritable Records Institute (Guoshi shilu­y uan bianxiu jiantao guan 國史實錄院編修檢討官) in 1175–76; the letter most likely dates from this period.

318

imperial information networks

them. Below I discuss the social backgrounds of those involved in the user publication of Waving the Duster and the relationships among them to underscore the role of literati networks and lowlevel officials in the growing circulation of notebooks from the twelfth century onward. Among those included in the first group of readers were Cheng Kejiu and Wang Qiu 王璆. Like Cheng Kejiu, Wang Qiu obtained a copy of the first installment. He lent his copy to Guo Jiude 郭九悳, who copied it himself (shouchao 手抄). Guo Jiude was familiar with the notebook before he met Wang Mingqing and later looked the author up when serving in office in a neighboring prefecture. Guo Jiude held the position of local government school instructor in Gaoyou 高郵 Military Prefecture (Huainan East) when he contributed a postface; Wang Mingqing was then in Taizhou 台州, just south of Gaoyou.57 It is noteworthy that, with the exception of Li Tao, Li Hou, and the colleagues with whom they shared their copy, all in the inner circle of Wang Mingqing’s readers either served in local government (Guo Jiude, Cheng Kejiu, and Zhao Bujian) or are not on record as having held office (Wang Qiu58 and Wang Yuxi 王禹錫); those in local office mostly held minor positions (professor, vice-magistrate, and county magistrate). From their prefaces it emerges that this was a tight-knit group: Cheng Kejiu, Li Hou, and Zhao Bujian were close friends of Wang Mingqing since youth. Li, Zhao, and Wang used to visit one another and had maintained their relationship over time. Wang’s 1185 postface was in some ways a tribute to Li Hou upon the latter’s passing. Wang Qiu is cited as an informant in the notebook and must therefore also have been a direct acquaintance.59 Through him Guo Jiude joined the network, expressing the hope in his preface that the connection with Wang Mingqing never be cut off. The connections among these men are shown in figure 6.6. 57. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.42–43. 58. Guo Jiude and Wang Mingqing do not mention any titles for Wang Qiu, suggesting that he did not hold rank or office. Ibid., 1.43; 1.28. In CBDB one position is listed for Wang Qiu; ID 10045. 59. Ibid., 1.28.



the notebook phenomenon

319

D iscip le sh ip WAN G Q I U

LI TAO

L I H OU

WANG MINGQING

GUO JIUDE

WAN G Y U XI ZH AO BUJ IAN

Tex t Ex c hange

Frie n d sh ip CHENG KEJ IU

Fig. 6.6 Relationships involved in the production of Wang Mingqing’s Waving the Duster. The dotted and straight lines indicate how the connections move from weaker to stronger ties (from the exchange of manuscripts and writing to lifelong friendships and family ties).

Wang Mingqing’s text initially circulated in a small community of friends and personal relations—albeit that in imperial Chinese society these relationships were often mediated through political networks (the connection between the Wang and Li families was mediated through office holding, and Guo Jiude met Wang Ming­ qing during his tenure in Gaoyou). The same personal networks that shaped and were shaped by the exchange of manuscript texts also mobilized resources for the printing of these same texts. This appears to have been particularly the case for the lowerlevel officeholders in the network: Cheng Kejiu and Zhao Bujian (as well as Wang Mingqing himself ) appear to have been most instrumental in setting off the wave of printed editions of Waving the Duster. Their resourcefulness in this regard was part of a larger trend in twelfth- and thirteenth-century Song China: as noted

320

imperial information networks

above prefectural yamen and schools in particular became the sites where at least 25 percent of Song printed editions of notebooks were produced. In the case of Wang Mingqing’s Waving the Duster, local government officials’ engagement in the business of user publication clearly happened with the endorsement of the author, so that early printed editions of Waving the Duster were the product of a combination of manuscript author publishing, user manuscript publishing, and user print publishing. The use of both manuscript and print and their convergence in the same social networks in twelfth-century China contrasts with the social uses of the print and manuscript medium in seventeenthcentury England, a period when, for the first time, English literate elites, like their twelfth-century Chinese counterparts, lived in a world where print publishing was a commonly available option. Harold Love observes a clear divide between those writing for manuscript publication and those writing for print publication in ­seventeenth-century England. Few authors wrote for both media; their choice of medium correlated with different conceptualizations of writing. Manuscript authors tended to conceptualize writing as the medium for an intimate relationship, as an organic and free process of expression; politically, the scribal text occupied a critical and oppositional place within elite discourse. Printing, originally a more closely controlled medium and associated with “the public utterance of the gentleman,”60 was able to assume a role like that of the scribal medium only after it began to give voice to the latter medium’s oppositional forms of discourse. As a result of the further expansion of print publishing, government control then also became less effective. In twelfth-century Song China, manuscript did not function as the opposite of print publication. The authors and users publishing in manuscript also provided copies of the manuscript text for its transposition onto woodblocks. They considered xylography the best way through which a text could be shared beyond the circle of 60. Love, The Culture and Commerce of Texts, 189.



the notebook phenomenon

321

known acquaintances to others similar in outlook. Even though print was, as in Europe, a more formal means of publication that required further editing of the text, it did not, as in the European case, necessitate the construction of a new relationship between professional publishers and authors. Differences in technology played a role in the different relationship between manuscript and print media in twelfth-century Song China and seventeenth-century England. Xylography was a technology widely distributed across the Chinese landscape, required less of an investment than typography, and could thus more easily remain under the purview of text producers. Perhaps because the initial investment was lower, there was less of an interest in legal protections for the printers.

Scr iba l User Public ation: The Court The choice of medium did not correlate with the formation of particular literati networks in imperial China. Neither could the use of a more secretive medium such as manuscript be directly related to an oppositional stance. Woodblock printing was considered to be a technology that allowed the author to communicate with a more general public of scholar-official elites. Growing numbers of lowerlevel literati invested in the printing of all manner of materials that enabled them to create ties both real and imaginary with an empirewide elite. Printing was intended and perceived to possess the ability to create community among elites because it allowed for a quantitative change in one’s ability to communicate with those belonging to the same class. In practice it differed from manuscript publishing in degree more so than in kind. Unlike printing, manuscript circulation was then and later associated with the transmission of secretive knowledge through exclusive lineages (for example in skilled professions such as medicine and performance), but scribal publishing was not limited to intimate or sensitive materials. The court was most likely the largest manuscript publisher, acquiring books throughout the empire in manuscript through government-funded copyists and collators. Even though we know very little about the economics of text production

322

imperial information networks

(and learn nothing in this regard from Wang Mingqing), we can surmise that given the small number of copies involved for each title, manuscript production was a more cost-effective means of reproduction for the court; it also allowed for greater control over the text, as copyists and collators could intervene in the process of transmission through deletion, collation, and standardization. The textual history of Wang Mingqing’s Waving the Duster provides additional support for the view that, despite some large-scale court printing projects, by the late twelfth century the Song court continued to rely on manuscript reproduction to stock its own collections even when printed editions could have been requested. Song printed copies of Waving the Duster included two official communications forwarded from the Veritable Records Institute (Shilu yuan 實錄院), the first issued on August 15, 1195, and the second, just about fifty days later, on October 5. Both documents noted that pursuant to a decree of 1188 ordering the compilation of Gaozong huangdi shilu 高宗皇帝實錄 (The veritable records of Emperor Gaozong) the office set up to undertake the project had gone in search of all sorts of documents held by the families of court officials who had served under Emperor Gaozong 高宗 (r. 1127–62). The first communication noted that the office had learned of Wang Mingqing’s First and Second Installments of Waving the Duster (Huizhu qian- houlu 揮麈前後錄) and ordered Wang—in the case of the later communication, with greater urgency—as controller-general 通判 of Taizhou 泰州, to have the work copied, checked, and then forwarded to the office.61 These communications provide rare insight into the ways in which court librarians and historiographers requisitioned works lacking in court libraries and archives. General requests to fill the libraries had been issued throughout the course of Song history, typically after fires or other disasters had damaged collections. Such requests often went accompanied with promises of honorary titles and monetary compensation for those making valuable contribu61. Wang Mingqing, Huizhu lu (SBCK ed.), qianlu, xu, 1a–3b.



the notebook phenomenon

323

tions.62 The documents included in extant editions of Waving the Duster show how more general imperial orders for searches were followed up with the family members of former court officials (Wang Mingqing’s father, Wang Zhi 王銍, had served at court during Gaozong’s reign).63 These documents further show that the court continued to rely on the manuscript reproduction of texts even when, as in the case of Waving the Duster, they may already have become available in print. Court collections are, therefore, not a reliable indicator of the distribution of manuscript versus print editions in private or institutional collections.64 The court relied on its own mechanisms of user publishing: it had its own copies made for the court libraries, where the works thus reproduced could be entered into the library catalogs and used by those allowed access. Court efforts to keep up with books produced in the provinces were indicative of its inability to control the book market; however, the dominance of literati networks in publishing ensured that scholars and officials kept each other in check, as suggested by Jing Tang’s edition of a controversial notebook. To conclude, in the Chinese as well as in the English case, manu­ script and print publishing were situated at different ends of the scale ranging from intimacy to formality. Sensitive materials, such as political commentary produced by identifiable living speakers, were in the first instance circulated within an in-group of intimates. In twelfth-century Song China many such materials made the transition to the print medium within a relatively short span of time, in the case of Wang Mingqing within his own lifetime. Even though it is safe to assume that conventions of the print medium affected the reproduction of the manuscript text, such traces of intimacy were preserved within the printed text as the inclusion of an unedited personal letter, the attribution of quotations from conversations 62. For a recent discussion, see De Weerdt, “The Discourse of Loss.” 63. For a detailed account of the careers and social relations of Wang Mingqing’s immediate ancestors, see Zhang Jian, “Wang Zhi ji qi jiazu shiji kaobian.” 64. See De Weerdt, “Byways in the Imperial Chinese Information Order.”

324

imperial information networks

to personal friends, or the listing of borrowers of unreturned books suggest.65 The transposition of calligraphic text onto the woodblock may have facilitated this process. Notebooks emerged during the Song dynasty as a quintessentially literati genre. Lower-level officials and scholars became increasingly involved in writing them. Court offices showed no interest in publishing them (and in some cases banned individual titles), but local governments and commercial printers printed increasing numbers in various locations in the twelfth and thirteenth centuries. The authorial pursuit of publication as well as personal, political, and local ties can be shown to have played a crucial role in their dissemination in print. For those serving in the lower echelons of the bureaucracy and for local literati associating with them, the manuscript and print publication of notebooks were expressions of the networking impulse, which always started with a smaller group of personal connections but remained open to the incorporation of regional and supraregional relations through which core elements of literati identity could be circulated. The subtle changes in the social status of published authors did not bring about a transformation in the overall topical coverage. Contrary to earlier assumptions of a local turn in accordance with their changing power base, it appears that authors who did not rise high or who did not rise at all in the bureaucracy did not turn away from the history of the Song court and its predecessors.

65. Examples of this last category of additions are discussed in De Weerdt, “Continuities between Scribal and Print Publishing in Twelfth-Century Song China.” The above section is based on this article, but the article covers twelfth-century perceptions of the relationship between manuscript and print and accounts for the more widespread adaptation of the print medium in user publication during Song times in greater detail.

Se v e n Informant Networks and Literati Identities

I

n the previous chapter I argued that the authorship of notebooks expanded socially over the course of the twelfth and thirteenth centuries and that local institutions and commercial enterprises were significantly involved in their printing. This chapter examines the communication networks embedded within notebooks in order to address the same questions about the social and geographical backgrounds of those whose work or whose words notebook authors chose to recall. The very constitution of notebooks was based on intertextuality and conversation. Due to this feature the genre lends itself particularly well to an analysis of the networks through which information was exchanged. Notebook authors referred with some regularity to the interlocutor who had provided a noteworthy piece of information, to the source of published and unpublished texts, or to the collector who had provided access to a particular item of interest. From the conversations recalled in notebook literature we can reconstitute the social networks within which their authors placed themselves. And from the relationships constituting these networks we can rewrite the history of literati identities.1 Using Wang Mingqing’s Waving the Duster as a case, I aim to shift the cultural history of literati identities away from broader generalizations about the dominant 1. On the meaning of and relation between social relations and identity, see White, Identity and Control.

326

imperial information networks

role of a classical or Confucian repertoire in norm integration and about localism as the dominant paradigm for scholars from the twelfth century onward. By linking the social history of commu­ nication networks to the history of the discourses expressed in them, this chapter and the next show how a twelfth-century person created the identity of a literatus through the artic­ulation of interests and positions within the context of concrete historical relationships. Within the conversations recalled by Wang Mingqing 王明 清 (1127–after 1214), self, family status, literati culture, court organization and operations, and Song dynastic history were the clusters on which attention converged.

Conversations The notebooks by Sima Guang 司馬光 (1019–86), Wang Mingqing, and Zhang Shinan 張世南 (13th c.) on which this chapter is based contain reports on conversations between the authors and other individuals or small groups of individuals (some of which consist of reports on conversations by others at a remove) alongside engagements with texts and their authors. How can we interpret such rec­ ords of conversations as a history of literati identity? Historians and literary critics have interpreted records of and manuals on conversation in two distinct ways. Some have read them as guides to the cultural norms of the upper classes, which then filtered down over time. In his work on the cultural history of language and communication in early modern Europe Peter Burke has called attention to the various ways in which Italian, French, and English manuals in the art of conversation regulated private and public speech. He further proposed that manuals and the representation of conversation in fiction increasingly shaped and standardized cultural norms after they were reproduced in greater numbers in print.2 Others have read conversations in historical records as ways of creating exclusive relationships and establishing communities. 2. Burke, The Art of Conversation, esp. chap. 4.



liter ati networks and identities

327

Patricia Meyer Spacks best captured the tenor of the 1998 special issue on conversation in Social Research when writing “conversation is a form of relationship, not just a form of speech.”3 Writing about the representation of conversation in eighteenth-century English biography and fiction, she alerted us to the predominantly public function of conversation in its literary representation. Conversation was, for people such as Samuel Johnson or James Boswell, in the interest of the common good; for others more interested in intimate and private matters, conversation was still a social activity guided by rules. The rules and the participants determined what could be said and also what had to be disguised and repressed. Conversation represented a “self ” that had been created for the purposes of smooth social interaction.4 Within this interpretation, “the participants matter more than the utterance.”5 The social gatherings recalled in notebooks included examples of what norms and rules partners to the conversation should conform to. Wang Mingqing and his contemporaries occasionally put such rules and norms into perspective by recording their transgression, as in the case of the scholar Dong You 董逌 (fl. 1120s). Dong You was one of a number of guests invited to a social event organized by Cai You 蔡攸 (1077–1126), who oversaw the Imperial Library in the 1120s. Cai You treated his guests to melon slices and asked them at the party to come up with past stories about melons. Most held back and allowed Cai You to come up with the best references, except for Dong You, a collator, who came up with one event after another, all of which were unfamiliar to the attendees. All the precedents he was able to muster proved to be based on historical fact and he garnered much praise for his performance. As expected, however, the public defeat of his superior also led to Dong You’s transfer days later.6 The representation of a self was, as in the case of eighteenthcentury literary depictions of conversations, central to the note3. Spacks, “Private Conversation, Public Meaning,” 611. 4. Ibid., 613–14. 5. Ibid., 614. 6. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.29.

328

imperial information networks

books of twelfth- and thirteenth-century authors such as Wang Mingqing. There are quantitative indicators suggesting that self-­ representation was central to note taking (through the number of explicit references to the author’s own voice).7 There are also several scenes in which the author illustrates the appropriate communication of breeding and wit by personal example. Wang Mingqing recalled in the first installment of Waving the Duster how, as a child in 1136, he had impressed Zhu Dunru 朱敦儒 (1081–1159) and Xu Du 徐度 (fl. 1130s–50s), who were both employed as proofreaders at the Imperial Library at the time and colleagues of Wang Mingqing’s father, with his ability to answer their questions on the draft dynastic histories.8 He also recalled a subsequent encounter, which took place over two decades later: When Dunli [Xu Du] had become vice-minister [in the Ministry of Personnel],9 I once happened to visit him. Suddenly he posed a question to the guests, “How come that the place where I now find myself is called the Vice-Minister Bridge?” I immediately responded that the bridge was called after Lang Jian [郎簡, 975–1063], a man from Hang­ zhou, who had served at the court of Renzong [仁宗, r. 1023–63]. He served as vice-minister of the Ministry of Works and lived in this neighborhood. Because people appreciated his virtue they called the bridge after him. He further asked what is meant by “Lang Jian demonstrating his virtue”? I replied, “In his biography in The Draft History of the Two Reigns it says that his style name is Jianzhi. In Wang Anshi’s collected works there is a poem to Lang Jian in which he highly praises his wisdom.” 10

7. When measuring explicit self-references against the numbers of references to all other persons mentioned in the entire notebook (not only informants but all personal names referred to throughout the notebook), I found that references to Wang Ming­ qing figured among the top three of over two thousand names mentioned independent of method of calculating centrality. De Weerdt and Ho, “Towards a New Platform for Tagging, Extracting, and Analyzing Named Entities in Classical Chinese Text Corpora,” pt. B, “Tagging Huizhu lu with MARKUS: Preliminary Results.” 8. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.39. 9. One such position is listed in the career record of Xu Du in CBDB (no. 3382). 10. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.39–40.



liter ati networks and identities

329

The questioning continues, and Wang Mingqing continues to shine. His recollection of his own performance exemplifies how the archival mentality discussed in chapter 1 was not only expressed in writing but also in gatherings. The representation of the author’s performance during these social gatherings is more than mere instantiation of the familiar tropes of the precocious child or the memorious scholar; the author here reenacts a scene in which he demonstrates in the presence of older men of high sociopolitical status his mastery over a body of politically relevant court history and his ability to participate in the social world of those whose rank he would never attain. Besides familiarity with the history of the Song court and the men who served it, wit was also on display and shared in Wang Mingqing’s recollections of his social world. The same Xu Du who quizzed friends and colleagues on the historical records of the Song dynasty was elsewhere cited as the source for a humorous anecdote involving Su Shi 蘇軾 (1036–1101). When Su Shi received visits from men seeking his support who were sent to him by his younger brother Su Che 蘇轍 (1039–1112), he cautioned them, I am reminded of a story. In the past there was a man who opened up tombs. He spent an enormous amount of energy to make it through to the cavity. In the cavity there sat a man, stark naked, who said to the thief, “Have you not heard that this mountain is called Shouyang and that I am Boyi? How could there be anything here?” The thief left dissatisfied and went to another mountain. When he had dug about halfway through, he suddenly felt the naked man from the previous day patting him on the back, “Don’t open it. Don’t open it. This is my younger brother’s tomb.”11

The story of the brothers Boyi and Shuqi, who fled the tyranny of the last Shang emperor but starved themselves to death on Shou­ yang Mountain, unwilling to take salaries from the new Zhou dynasty, was common lore. Here Su Shi invoked the story to decline requests from those who sought recommendations from him 11. Ibid., 4.300.

330

imperial information networks

and his brother while simultaneously hinting at the political reasons for their refusal: in the charged political climate of the late eleventh century the Su family had little to offer to those hoping to chance upon a major prize. Recalled decades later, Su Shi’s words also bore testimony to the wit of the men who shared the story then. The conversations collected in notebooks were, besides being shapers of norms and vehicles for self-fashioning, also forms of political networking. We see here at work a political culture that is fundamentally interactional and conversational. Political culture is in this sense not limited to the people’s or social groups’ ideas about and attitudes toward political structures; it involves relationships among people “pursuing politically salient ends.”12 In notebooks we come across a range of ways of building (and breaking) politically salient relationships: besides the negotiation of patronage relationships and anecdotes about political networking, notebook authors shared stories about court figures and military and political upstarts, jokes about inappropriate language use, or cautionary tales about the dangers of not inviting the right people and overstepping one’s official rank. Being included in literati conversations and being perceived to be included in them was of such significance that exclusion could lead to retaliation of a kind that touched all who participated in it and that, as an event that percolated through the ranks of the bureaucracy, reverberated throughout the empire. Wang Mingqing transmitted the following anecdote, which attests to the political weight attached to social gatherings among elites in the provinces. In the early 1100s Liu Qi 劉跂 ( j. 1079), whose father Liu Zhi 劉摯 ( j. 1057) had been included on the blacklist of Yuanyou politicians compiled under the reformist regime of Cai Jing 蔡京 (1046–1126), retired to Dongping 東平 (present-day Shandong Province). There, Wang Xuanzan 王宣贊, a local military official with a reputation for hospitality who harbored the desire to establish connections with the Liu family, invited him to a party. Another member of the local elite, a man by the name of Li Yannian 李延年, wished to be included among the prominent guests, but the host refused his 12. McLean, The Art of the Network, 31–32.



liter ati networks and identities

331

request. Li Yannian had earlier lost office due to a criminal conviction. In an attempt to clear his name he traveled to the capital Kaifeng, where he met up with Wang Fu 王黼 (1079–1126), an old acquaintance. Wang Fu then held the post of vice-censor-in-chief (yushi zhongcheng 御史中丞) and was thus in a position to make or break careers. When asked about recent events in Dongping, Li Yannian told Wang Fu about the banquet hosted by Wang Xuanzan and mentioned that the attendees had discussed recent events relating to the Song dynastic family (the contents of which do not appear in Wang Mingqing’s account). Wang Fu later asked Li Yannian to write down what had been said at the banquet. Li Yannian reportedly spiced up the language to such an extent that the vicecensor decided to arrest all involved forthwith. Liu Qi was banished and all attendees, down to the entertainers, were punished. Wang Mingqing concluded, “This was all the result of one guest not being able to eat.”13 The story was relayed to the author by Zhao Jun 趙濬 (fl. 1154) and Liu Dong 劉董, both men for whom no court offices are on record and who may have belonged to the growing number of literati in the twelfth century who, like the notebook author himself, created and maintained literati identities by sharing stories about political fortunes.

Citation Patterns and Social Relations In the conversations recorded in notebooks, literati communicated cultural norms appropriate for their readership, staged a self that conformed to those norms, and recalled meaningful social and political relationships. What kinds of relations were recalled? What can an analysis of the informants whose words were entered into a notebook suggest about the social and political world of their author? A focus on literati communication patterns over the long run clarifies continuities and transformations in the transition from the more court-centered world of the tenth and eleventh centuries to the more localized literati life of the twelfth and thirteenth centuries. 13. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.180–81.

332

imperial information networks

By comparing the relationships recalled in Wang Mingqing’s notebook to that of two eleventh-century and thirteenth-century authors I will argue that the changes in the authorship of notebooks discussed in chapter 6 brought about gradual but significant changes in the voices represented in notebook literature, albeit that voices of authoritative political figures remained prominent.

Wa ng Mingqing, Sim a Gua ng, a nd Zh a ng Shina n For the analysis of informants I selected five notebooks and three notebook authors based on a review of a large sample of Song notebooks consisting of all notebooks contained in the Song Biji Print Database and additional notebooks that may have circulated in manuscript only.14 These include Wang Mingqing’s Waving the 14. With the help of three assistants I evaluated the extent to which all notebooks in the sample referenced oral and written sources and chose from among those shortlisted a set that allowed for a chronological spread and a diversity of backgrounds. The electronic text of the five notebooks was systematically and comprehensively encoded in XML-TEI. The goal was to collect all sources of information for each of the entries and to keep track of the kind of information contained in it. All notebook entries were therefore coded for the following information: (1) authors; (2) titles of books and other texts; (3) other informants, including interlocutors and secondary interlocutors (those whose conversations were relayed through an interlocutor), collectors, and patrons (who provided access to or ordered written materials); (4) oral and written quotations (where this could not be determined, the quote attribute ambivalent is used); (5) the main topical categories of each entry; (6) the time period discussed in each entry (temporal markup). Citation information concerns the sources Wang Mingqing referred to directly and typically does not include secondary citations within the quoted texts or passages of reported speech. Most of these elements were given attributes that allow for further analysis. For titles we provide an indication of the genre to which the title belongs and we note whether it was cited, quoted, or just mentioned; when relevant, we indicate whether the title is analytical (a piece in a larger body of work) or above monograph level (series). Titles known under different names (usually abbreviated titles) are also given a standardized title as an attribute. Temporal markup covers all references to reign periods and emperor’s names (including both ancestral and tomb names). The topics assigned to each entry were not based on a prior classification system; I developed the categories while reading the entries in Waving the Duster, and then regrouped them in a hierarchy moving from general topics to subtopics. Additions are being made to this taxonomy as other notebooks are similarly encoded. The topics do not necessarily reflect Song categories or conceptual schemes; I and the graduate students who helped tag additional notebooks decided to work with categories that best



liter ati networks and identities

333

Duster, Tou xia lu 投轄錄 (Keeping guests, before 1196–97),15 and Yuzhao xin zhi 玉照新志 (New record from the Jade Shine Studio, 1198); Sima Guang’s late eleventh-century notebook Sushui jiwen 涑水記聞 (A record of hearsay from Su River); and Zhang Shinan’s early thirteenth-century Youhuan jiwen 游宦紀聞 (Records of official travel). It is impossible to tell how typical or atypical these may be without a similar analysis and comparison of a much larger sample. On the basis of the review, however, it appears that this selection may capture a trend, with early Song notebooks containing a higher rate of court-based notebooks and authors with extensive court experience and Southern Song notebook authors adapting the court-based notebook to the interests of literati serving in official and other capacities in the provinces. Wang Mingqing was the son of an eminent court historian, Wang Zhi 王銍, and several of his paternal and maternal ancestors had held local, regional, and/or central government positions. Wang Mingqing himself, however, did not rise high on the bureaucratic ladder. He did not pass the examinations and served only in a handful of minor mostly advisory positions. He held a temple captured the content in our reading for both analytical and practical reasons. As we intend to identify topic clusters and analyze them in a present-day cultural historical analysis, it appears justified not to work with more traditional library classification schemes—those also run the risk of imposing a scheme that would not have coincided with the interests of the note taker. Practically speaking, we hope that, once included in the range of criteria by which readers can analyze, browse, or search the notebooks in an online environment, the primary and secondary tags (up to five in total per entry) will offer present-day readers new ways to access and read sources that, like notebooks, contain snippets of information on a wide range of topics. All personal names were to the extent possible linked to CBDB. By linking into this vast and growing prosopographical database, we gain access to extensive biographical, family, and career information on all informants. All place-names in informants’ biographical and career information were further mapped using the historical GIS data in China Historical GIS (CHGIS). These data are or will be freely available. By using XML-TEI for the notebooks and by thus creating data sets within the original texts we intend to ensure that, once the notebooks are published online, the results presented below can be easily tested, and all results can be interpreted within the context of the full text. 15. Wang Mingqing refers to Keeping Guests in the last installment of Waving the Duster (1196–97) and in New Record from the Jade Shine Studio (1198).

334

imperial information networks

guardianship and in the early 1190s worked successively as a controller for the Office of Miscellaneous Purchases at the Court of the Imperial Treasury, an administrative assistant to a military commis­ sioner, and was also assigned to the office of comptroller general in Taizhou 泰州. A decade later, he served as a consultant on the staff of a commissioner in Western Zhe. None of these posts kept him busy for long, and he appears to have devoted a substantial amount of time in the latter decades of his life to researching and compiling notebooks. Waving the Duster appeared in four installments totaling around 450 entries; the other two notebooks were smaller and counted around 49 and 86 entries each. A Record of Hearsay from Su River was in all likelihood based on notes about court events taken by the famed politician and historian Sima Guang in the late eleventh century. It consists of around 450 entries. Sima Guang may have taken the notes in preparation for a history of the Song dynasty that was intended as a sequel to his more famous survey of Chinese history up to 960. This notebook was not printed until around 1145 in the commercial printing center of Jianyang. It was then part of the general interest in the recovery of materials that had allegedly become rare as a result of the war with Jin. Prior to its first documented commercial circulation, in 1136, Councilor Zhao Ding 趙鼎 (1084–1147) asked for and received imperial permission to assign the noted historian Fan Chong 范沖 (fl. 1090s–1130s) the task of recovering and arranging these notes.16 Like Wang Mingqing, Zhang Shinan is mostly known to us through his notebook. He was a younger contemporary of the former, active between the late twelfth and early thirteenth centuries. Even though the exact titles and levels of the government posts he and older male relatives may have held are unclear, his notes suggest that both he and his father served in a number of local posts. In Zhang Cong’s reading of the work, the Zhangs were most likely a family of local repute but were not tied into the networks of families with distinguished careers at court.17 Records of Official Travel 16. For a further discussion of this work and a ban that was issued under Qin Gui, see De Weerdt, “The Discourse of Loss,” 415–16. 17. Zhang Cong, “To Be ‘Erudite in Miscellaneous Knowledge.’ ”



liter ati networks and identities

335

is now divided into ten juan and consists of fewer than 150 entries (108 in Zhang Cong’s count). As suggested by the title, the notes captured scenes from the life of a man traveling from one minor assignment to the next, in his early years accompanying his father and in later years pursuing his own career. Zhang spent a considerable amount of time in the southwestern region of Sichuan and also lived in what is now Jiangxi, Hunan, and Fujian.

Citation Net wor ks In the first step of this comparison of the structure of informant networks, I examine how extensive or constrained the scope of each network was and to what extent particular informants dominated in the memory of the note takers. This line of questioning allows us to gauge whether literati followed broader cultural trends detected by modern scholars, for example, the ascendancy of literary icons such as Su Shi or moral philosophers such as the Cheng brothers or Zhu Xi, or whether their preferred sources of information were indicative of other trends. Through an analysis of the temporal distribution of informants we can further determine whether literati interests focused on older texts (for example, classical texts, as is often assumed) or on more recent publications. Wang Mingqing’s network of informants was even by modern standards large. The combined network of interlocutors and authors in the four installments of Waving the Duster contains 309 unique nodes (informants). The number of informants from whom he collected and recollected information grew substantially over time. The first 100 entries were based on information obtained from 37 authors, 19 interlocutors, and 7 collectors or 61 informants in total; there was only very limited overlap between the two categories (two instances only). The subsequent installment of 200 entries, published around 1194, was based on the contributions of 191 unique informants including 104 authors, 74 interlocutors, and 21 collectors. There is again little overlap between the different types of contributors (16 instances); the overlap between the earlier and the later contributors is also limited: 21 informants, or about one-third, cited in the first installment reappeared in the second installment.

(11) (1) (5)

(1)

24: 0.32% [1/308], 4.34% [24/553]

15: 0.97% [3/308], 8.14% [45/553]

11: 0.32% [1/308], 1.99% [11/553]

(110) (80) (21) (1)

1: 68.83% [212/308] 38.34% [212/553]

+ + + + + +

= = = = = =

Publisher

Patron

Collector

+

+

Interlocutor/Secondary Interlocutor

Fig. 7.1 Wang Mingqing, Waving the Duster: informant network by citation frequency. Frequencies rise from 1 at the periphery to 2, 3, 4, 5, 7, 8, 11, 15, and 24 closer to the core. The total number of informants excluding Wang Mingqing is 308, and the total number of citations excluding self-citations is 553. Other views of the citation data for this and the other notebooks can be seen in De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.” The data can be linked interactively to the text and biographical information on the informants from De Weerdt et al., “Notebooks.”

(19) (17) (2)

2: 18.18% [56/308] 20.25% [112/553]

3: 6.17% [19/308], 10.31% [57/553]

4: 1.62% [5/308], 3.62% [20/553]

5: 1.3% [4/308], 3.62% [20/553]

7: 1.3% [4/308], 5.06% [28/553]

8: 0.97% [3/308], 4.34% [24/553]

Author



liter ati networks and identities

337

Across the four installments (450 entries) there are 168 unique authors, 117 unique interlocutors, 38 unique collectors, 24 unique secondary interlocutors (those whose conversations were relayed through an intermediary interlocutor), and a smaller number of other types of informants (patron, publisher, and the speaker, i.e., the author of the notebook himself ). Literati probably regarded the extent of one’s network and its expansion over time as an asset. This is confirmed in Wang Ming­ qing’s other extant notebooks. Keeping Guests relays in its small number of entries the words of 36 unique informants, only 11 of whom appear in Waving the Duster; another 8 are also cited in New Record from the Jade Shine Studio. Similarly, New Record from the Jade Shine Studio mentions 79 unique informants, over one-third of whom had not appeared in any of Wang Mingqing’s previous notebooks. Over half had made an appearance in Waving the Duster (45), and a few more had been mentioned in Keeping Guests (7). These two notebooks, compiled around the same time as the later installments of Waving the Duster, thus bring the total of Wang Mingqing’s informant network up to 353 individuals. Figure 7.1 shows the full extent of the informant networks embedded in Wang Mingqing’s Waving the Duster and arranges the nodes (informants) encircling Wang Mingqing in the center in the order of the frequency with which they are cited. A higher frequency corresponds with closer proximity to the center of the network. The structure of this diagram is similar to the pattern underlying the smaller networks invoked in his other two notebooks.18 He cited most informants only once. The number of those he cited more frequently declines gradually and only a small number occupy the core. The number of citations for those at the core is relatively small, suggesting that diversity rather than canonicity mattered to the notebook author. In Waving the Duster, Wang Mingqing cited only five informants more than ten times. Three among the five were family members. He cited his father most frequently, followed by Zeng Shu 曾紆 (1073–1135), who was his grandfather on his mother’s 18. For the other diagrams, see De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.”

338

imperial information networks

side, and Fang Zi 方滋 (1102–72), his father-in-law. Xu Du, whom along with Zeng Shu he cited fifteen times, was, as shown in the exchanges translated above, an older acquaintance. Even though the author network contains a number of well-known eleventhcentury authors with higher citation frequencies (Su Shi and Ou­ yang Xiu 歐陽修, 1007–72), their number is small and the total number of citations typically does not exceed 10 with Su Shi (15), the only exception. Out of the total number of citations (553 if we exclude all self-citations), the percentage of citations for the informant with the highest number of citations is only 4 percent and the percentage for all those with 10 or more citations is 14.5 percent. A frequency analysis of the informant networks in Wang Ming­qing’s other notebooks demonstrates minimal differences in frequency rates among informants. The names of those at the core there similarly suggest that he accorded senior family members and a select group of eleventh-century authors and politicians some level of primacy. Only Wang Zhi ranks high in all three notebooks, while authors such as Su Shi and Ouyang Xiu or family members like Zeng Dun 曾惇 (fl. 1136) receive slightly more attention than other informants. The large number of individuals (almost exclusively men) from whom Wang Mingqing recalled snippets of information may appear unsurprising in the context of the scope and organization of the imperial bureaucratic system. During the Song dynasty men sat examinations with large numbers of fellow students and joined in social celebrations when successful. Those who passed or who obtained regular positions in officialdom traveled back and forth to the capital where they may also have had the opportunity to participate in collective ceremonies at court or to join in smaller festivities with peers. In the case of Wang Mingqing, however, it is worth bearing in mind that he did not pass the civil service examinations and never obtained significant office; the minor assignments he obtained also appear to have only covered a short span of time. In the twelfth and thirteenth centuries Wang Mingqing and many others like him took careful note of who told them what and where they had obtained information. Recalling and preserving relationships became a preoccupation, and sharing one’s ongoing



liter ati networks and identities

339

engagement in literati networks through the circulation and publication of notebooks became a pastime for upper-class men. Wang Ming­qing benefited from the relationships he was drawn into in youth through his father’s and father-in-law’s assignments in the capital and elsewhere. It is significant, however, that it was not Wang Ming­qing’s father Wang Zhi who sought to publish his notes but rather his less successful son. The absence of canonization appears to be a more widely shared feature of informant networks in notebooks. The distribution of citation frequencies in Sima Guang’s and Zhang Shinan’s notebooks suggest a similar structure with a larger number of informants contributing small amounts of information, a gradual reduction in the number who contribute more, and no evidence of the dominance of authoritative figures at the core of the network. Figure 7.2 shows the informant network in A Record of Hearsay from Su River. Like Waving the Duster this notebook was constituted of about 450 entries. The author cites 154 unique informants, nearly all of whom are direct contacts (interlocutors). At the core of the frequency network are three interlocutors whom Sima Guang cited more than ten times: one is his father, the others appear to have been court contacts—Pang Ji 龐籍 (988–1063) had an illustrious career in part thanks to his relationship with Sima Guang, and Su Yan 蘇兗, about whom less is known, provided lots of insider information about court politics under Wang Anshi 王安石 (1021–86). As in Wang Mingqing’s Waving the Duster, the voices of those who appear more frequently do not dominate; Pang Ji’s contributions account for 5.5 percent out of the total number of citations (291) and all contributions of those with more than ten references account for 13.4 percent. Zhang Shinan’s network was smaller, at 83 unique informants, reflecting also the much smaller scope of the notebook itself. Compared with Wang Mingqing’s shorter notebooks, the total number of informants for the one hundred or so entries is of the same order. However, in contrast to Wang Mingqing, who cited authors and interlocutors in more or less equal measure, and Sima Guang, who mostly noted down orally transmitted information, Zhang Shinan’s informant network in Records of Official Travel consists of twice as

16: 0.65% [1/153], 5.57% [16/287]

12: 0.65% [1/153], 4.18% [12/287]

11: 0.65% [1/153], 3.83% [11/287]

9: 0.65% [1/153], 3.14% [9/287]

(15) (86) (1)

1: 66.67% [102/153] 35.54% [102/287]

=

+

Collector

Interlocutor/Secondary Interlocutor

Author

Fig. 7.2 Sima Guang, A Record of Hearsay from Su River: informant network by citation frequency. Frequencies rise from 1 at the periphery to 2, 3, 4, 6, 7, 8, 9, 11, 12, and 16 closer to the core. The total number of informants excluding Sima Guang is 153, and the total number of citations excluding self-citations is 287.

(1) (26)

2: 17.65% [27/153] 18.82% [54/287]

3: 6.54% [10/153], 10.45% [30/287]

4: 3.92% [6/153], 8.36% [24/287]

6: 0.65% [1/153], 2.09% [6/287]

7: 0.65% [1/153], 2.44% [7/287]

8: 1.31% [2/153], 5.57% [16/287]



liter ati networks and identities

341

many authors (58) as direct contacts (19 interlocutors and 6 collectors); about half of the total number of contributions came from authors. The citation frequency network shown in figure 7.3 exhibits the same pattern as in the earlier notebooks by Sima Guang and Wang Mingqing. The most frequently cited informant was an acquaintance by the name of Cheng Jiong 程迥 ( j. 1163) whose contributions accounted for 7.9 percent out of the total (126). Cheng Jiong lived in Poyang 鄱陽 ( Jiangxi), which was the home town of Zhang Shinan.19 Cheng Jiong does not appear to have had an illus­ trious career, but he was well respected as a scholar of the classics; the more pronounced interest in classical philology in Records of Official Travel, when compared to the other notebooks, may well have had something to do with Zhang’s exposure to the more senior scholar. The contributions of all top informants (those with more than four citations) account for 13.5 percent. This suggests that, even though there is some overlap between the lists of top informants in Wang Mingqing’s and Zhang Shinan’s notebooks (both Su Shi and Hong Mai 洪邁 [1123–1202] appear among the authors with higher citations frequencies), informant networks in notebooks were shaped by diversity and direct relationships more so than by the authority of canonical authors. Were notebook authors drawing on recent or more dated sources of information? Interlocutors are by definition contemporaries, since they contributed information directly and orally; the temporal analysis is therefore limited to the authors in the network. The analysis of author networks is particularly significant in the case of Waving the Duster and Records of Official Travel because authors constituted the largest group of informants for Wang Ming­ qing and Zhang Shinan.20 Both authors drew written information 19. Zhang Shinan, Youhuan jiwen, 6.53. 20. Introducing a temporal dimension to the network analysis of information exchange is a difficult challenge. In most cases we do not know when the information exchange took place, but we can assign index years to all informants to map the temporal distribution of all sources cited. In light of the interest in contemporary and Song dynasty materials noted in the chapters in part I, this latter type of analysis allows us to gauge the currency of the sources quoted or to identify which time periods attracted Song note takers’ attention. For the temporal diagrams we used the CBDB index year

(42) (21) (4)

1: 81.71% [67/82] 56.3% [67/119]

=

= +

+

Patron

+

Interlocutor/Secondary Interlocutor

Author

Fig. 7.3 Zhang Shinan, Records of Official Travel: informant network by citation frequency. Frequencies rise from 1 at the ­periphery to 2, 3, 4, 7, and 10 closer to the core. The total number of informants excluding Zhang Shinan is 82, and the total number of citations excluding self-citations is 119.

2: 8.54% [7/82], 11.76% [14/119]

3: 3.66% [3/82], 7.56% [9/119]

4: 3.66% [3/82], 10.08% [12/119]

7: 1.22% [1/82], 5.88% [7/119]

10: 1.22% [1/82], 8.4% [10/119]



liter ati networks and identities

343

principally from recent publications. Wang Mingqing included only a dozen authors whose lifetimes predated the Song dynasty; none of these contributed more than two references (fig. 7.4A). Over time the number of authors who had not lived during Song times declined; the majority of references to pre-960 sources appear in the first installment. The number of authors cited rises steeply as we move from those active during the earlier decades of the Song dynasty to those active immediately before and during the author’s lifetime. The number of authors who were active in the first century of Song rule (960–1059)21 was double that of the preceding centuries, and the number nearly doubled again, to 43, in the next fifty years (1060–1109) (fig. 7.4B). This number roughly includes all those active during the eleventh century.22 That number rose by another 50 percent during the next century, a number that includes all those active during Wang Mingqing’s lifetime (1110–1219, adding ten to fifteen years either side to include those who could have been directly known to Wang Mingqing). When all informants active ­between 1100 and 1219 are included, as in figure 7.5, it is evident that the vast majority of informants were active during Wang Ming­ qing’s lifetime (237 between 1100 and 1219; 204 between 1127 and 1219). Similarly, Zhang Shinan’s notebook contained a relatively small number of pre-Song and early Song citations; more than half of the citations came from contemporary, twelfth- and thirteenthcentury sources (fig. 7.6). The interest in contemporary witnesses is even more evident in Sima Guang’s notebook: none of his informants were at the end of their careers prior to 1000 and all were thus potential direct contacts. The emphasis on contemporary informants in all three notebooks corresponds with an emphasis on more recent content. We can get some sense of the temporal coordinates of the topics discussed in each (sixty years of age or year of death) for the informants, or, where this was missing, an index year assigned according to similar standards (principal years of activity) by ourselves. 21. For additional diagrams, see De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.” 22. This makes the assumption that the CBDB index year is based on the sixtieth year of the biographee.

(1) (1)

(129)

1: 6.55% [11/168] 4.3% [11/256]

Author

Fig. 7.4A Wang Mingqing, Waving the Duster: author network to 959. The shaded areas show the number of authors active prior to 960. The percentages represent their proportion of the total number of authors and citations, and the numbers in brackets indicate the frequency of occurrence of citations by the authors in question.

(23)

2: 1.19% [2/168] 1.56% [4/256]

(9)

(1)

(3)

(1)

15: 0.6% [1/168], 5.86% [15/256]

(129)

1: 17.26% [29/168] 11.33% [29/256]

Author

Fig. 7.4B Wang Mingqing, Waving the Duster: author network, 1060–1109. The shaded areas show the number of authors active from 1060 to 1109. The percentages represent their proportion of the total number of authors and citations, and the numbers in brackets indicate the frequency of occurrence of citations by the authors in question.

(23)

2: 5.36% [9/168] 7.03% [18/256]

(9)

3: 1.79% [3/168], 3.52% [9/256]

5: 0.6% [1/168], 1.95% [5/256]

8: 0.6% [1/168], 3.13% [8/256]

24: 0.32% [1/308], 4.34% [24/553]

15: 0.65% [2/308], 5.42% [30/553]

11: 0.32% [1/308], 1.99% [11/553]

8: 0.65% [2/308], 2.89% [16/553]

(110) (80) (21) (1)

1: 48.38% [149/308] 26.94% [149/553]

+ + + + +

= = = = =

Publisher

Patron

Collector

Interlocutor/Secondary Interlocutor

Author

Fig. 7.5 Wang Mingqing, Waving the Duster: informant network, 1100–1219. All informants are shown. The shaded areas show the number of informants active from 1100 to 1219. The percentages represent their proportion of the total number of informants and citations, and the numbers in brackets indicate the frequency of occurrence of citations by the informants in question.

(19) (17) (2) (11) (1) (5) (1)

2: 12.99% [40/308] 14.47% [80/553]

3: 4.22% [13/308], 7.05% [39/553]

4: 1.3% [4/308], 2.89% [16/553]

5: 0.97% [3/308], 2.71% [15/553]

7: 1.3% [4/308], 5.06% [28/553]

(42) (21) (4)

1: 47.56% [39/82] 32.77% [39/119]

=

= +

+

Collector

+

Interlocutor/Secondary Interlocutor

Author

Fig. 7.6 Zhang Shinan, Records of Official Travel: informant network, 1150–1266. All informants are shown. The shaded areas show the number of informants active from 1150 to 1266. The percentages represent their proportion of the total number of informants and citations, and the numbers in brackets indicate the frequency of occurrence of citations by the informants in question.

2: 2.44% [2/82], 3.36% [4/119]

3: 2.44% [2/82], 5.04% [6/119]

4: 2.44% [2/82], 6.72% [8/119]

10: 1.22% [1/82], 8.4% [10/119]

Fig. 7.7 Wang Mingqing, Waving the Duster: temporal distribution of entries, by reign period.

SHENZ ONG

YINGZ ONG

RE NZ O NG

Z H EZ ON G

SHAOXI

QI ANDAO

XIAOZONG

GU ANGZONG

LONGXING

SHAOXING

JIANYAN

JI NGKANG

CHUNXI

GAOZ ONG

QINZO NG

HU IZ ON G

XUANHE

ZHENGHE

DAGUAN

CHONGNING

JI ANZHONG JINGGUO

YUANFU

SHAOSHENG

YUANYOU

YUANFENG

XINING

ZHIPING

JI AYOU

ZHIHE

HUANGYOU

QINGLI

1041 1049 1054 1056 1064 1068 1078 1086 1094 1098 1101 1102 1107 1111 1119 1126 1127 1131 1163 1165 1174 1190

紹熙

淳熙

乾道

隆興

紹興

靖炎 

靖康

宣和 

政和

大觀

崇寧

建 中靖國 

元符

紹聖

元祐

元豐

熙寧

治平

嘉祐

至和 

皇祐

慶曆

Fig. 7.8 Zhang Shinan, Records of Official Travel: temporal distribution of entries, by reign period.

SHENZONG

ZH EZO N G

紹興 隆興 乾道 淳熙 紹熙 慶元 嘉泰 嘉定 寶慶 紹定

1127

1131

1163

1165

1174

1190

1195

1201

1208

1225

1228

SHAOXING

LONGXING

QIANDAO

CHUNXI

GUANGZONG SHAOXI

QINGYUAN

JIATAI

JIADING

BAOQING

SHAODING

XUANHE

靖炎

H U IZ ON G

1119

J INGYAN

1111

宣和

政和

1107

ZHENGHE

CHONGNING

大觀

SHAOSHENG

1102

DAGUAN

紹聖

1094

崇寧

元祐

1086

YUANYOU

1078

GAOZ ON G

XIAOZ ONG

NINGZONG

LIZONG

352

imperial information networks

notebook by tabulating the temporal distribution of the entries. The temporal distribution is based on a tabulation of all time references in the form of emperor names and imperial reign names. In the case of Waving the Duster, the reigns from Emperor Shenzong 神宗 (1067– 85) on are best covered, and the interest in more recent reigns (Huizong 徽宗 [r. 1100–26], Gaozong 高宗 [r. 1131–62], and the very brief reign of Qinzong 欽宗 [1126–27]) increases in later installments (fig. 7.7).23 The same applies to Records of Official Travel, in which the later reigns from the early Southern Song onward are covered most extensively (fig. 7.8). Notebook compilers thus showed greater interest in information with greater news value. Not only current affairs but also new interpretations of the most relevant recent Song history were the kinds of topics which Wang Mingqing and his publishers deemed most appealing to their readers. The temporal structure of the informant network and the temporal distribution of entries correlate to a lesser extent in Sima Guang’s notebook. Even though the author and his sources refer more frequently to the reign periods covering the last four decades of his life, and in particular to the reformist Qingli 慶曆 and Xi­ning 熙寧 periods, he invoked the founding emperors with even greater frequency.24 The implications of this require further close reading and interpretation, but it may well be the case that the authority assigned to the words and deeds of the founders of the reigning dynasty, which has recently been analyzed as a core feature of Northern Song politics,25 lost some of its rhetorical force in the course of the twelfth century. Indeed, the work of Wang Mingqing and Zhang Shinan suggests that literati gradually adopted the emperors who revived the Song in the south as new sources for authoritative statements (figs. 7.7–7.8). In sum, a systematic analysis of the sources cited in the selected notebooks suggests structural similarities in the authors’ citation 23. Further tables, showing distributions by installment and distributions by emperor names, are included in De Weerdt, “Information, Territory, and Elite Networks— Extra Materials.” 24. See tables for Sushui jiwen in De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.” 25. Deng Xiaonan, Zuzong zhi fa.



liter ati networks and identities

353

networks across the centuries. Their information networks tended to be large and diverse, showed no sign of the dominance of individual sources of information, and focused on contemporary informants and recent history with the focus of attention shifting from the early and then late Northern Song to the early reigns of the Southern Song in the course of the twelfth and thirteenth centuries.

The Sociopolitic a l Structur e of Infor m a nt Net wor ks What were the social and political attributes of those who were cited in notebooks ranging from the eleventh through the thirteenth centuries? The question of the sociopolitical backgrounds of informants becomes particularly relevant in view of the prevalence of con­ temporary informants in the communication networks discussed in the previous section. At present the attributes we can explore are to a large degree restricted to the presence and levels of office holding, access routes to office, and family and other connections. Despite the fact that the questions are thus restricted by the structure, contents, and coverage of the prosopographical database used (CBDB), the presence and absence of career information as well as the kinds of positions held by informants allow us to inquire further into the effect of the provincialization of notebook authorship on the networks represented in them. In other words, how did the networks recalled by high court officials compare to those recalled by literati with minor or without office? Officeholders constituted a very substantial group among the informants in the information networks of all notebook writers in Song times (table 7.1).26 Given the backgrounds of the notebook authors, with Sima Guang having held office at all levels of the bureaucratic hierarchy and the other two having held minor local offices, it is to be expected that peers were well represented. The 26. The data are based on CBDB and are most likely to be an underestimate. CBDB now includes data from the best biographical reference sources available, but information about office holding will most likely be augmented and revised in the future as primary sources are mined for further biographical information.

354

imperial information networks Table 7.1 Informants in officialdom in Sima Guang’s, Wang Mingqing’s, and Zhang Shinan’s networks

Notebook author and title

Proportion of officeholders in each network

Sima Guang, A Record of Hearsay from Su River 101/154 65.65% Wang Mingqing, Waving the Duster 201/309 65.05% Zhang Shinan, Records of Official Travel 34/83 40.96%

percentage of officeholders among informants in general does not appear to have dropped significantly over time; the lower percentage in Zhang Shinan’s work reflects in part a higher number of anonymous and pre-Song informants. Informants of lesser standing such as fishermen or woodcutters were referred to by occupation rather than by their full name. Despite the apparent continuous role of officeholders, significant changes took place in the career patterns of informants between the eleventh and thirteenth centuries. The vast majority of Sima Guang’s informants with office held court offices (93 out of 101).27 The average number of posts held by all Sima Guang’s informants with office is 13.56. Nearly half of them had held local, regional, and court offices (47.5 percent). Sima Guang thus appears to have interacted seldom with those who only held local office. Even though we do not have office data on about one-third of his informants, from checks on those informants it is evident that several of them also had court affiliations and served in security or other service roles at court. 27. For a table showing the temporal distribution of informants with court office in Sushui jiwen, see De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.”



liter ati networks and identities

355

Court office remains a common attribute among the infor­ mants with office in the later notebooks, but in both networks the percentage of court officeholders declines, the percentage of those holding local offices rises, and the average number of posts goes down. Moreover, these trends can be shown to accelerate in the course of the twelfth century. Nearly one-quarter of Wang Mingqing’s informants with office held local or regional offices only (24.38 percent). The percentage of those with three or more court offices is 69.65 percent, down from 92 percent in Sima Guang’s case. Moreover, the percentage of court officeholders among informants active prior to 1129 is 75 percent but 65.8 percent for those active thereafter. If we move the cut-off up to 1139–40, the discrepancy widens further with 75 percent of those active up to 1139 holding court offices and 62.5 percent holding court affiliations thereafter. The reverse holds for local office holding. Only one-fifth among those active prior to 1129 held local office, but that ratio goes up to 26.5 percent for those active after 1130 and to 31.8 percent among those active after 1140. Similar ratios emerge from Zhang Shinan’s informant network: 23.5 percent of his informants held local office only. The average number of postings for all informants with office is 10.56 for Wang Mingqing’s informants and 10.2 for Zhang Shinan’s. This number goes down further for those active in 1128 or later, to 9 for Zhang’s informants and 9.22 for Wang Mingqing’s. Similarly, the percentage of those rising through the ranks from the bottom to the top is halved and drops to 21.89 percent for Wang Mingqing’s informants and 17.65 percent for Zhang Shinan’s.28 In the later notebooks, informant career patterns differ depending on the type of informant. Those in direct contact with the notebook author, that is, interlocutors (and to a lesser extent collectors), were less likely to hold office and less likely to hold high office than those cited as authors and than those included in Sima Guang’s notebook as conversation partners. These findings are very clearly visible 28. Tables showing temporal distributions of informants with court office and with local office for these two notebooks are included in De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.”

356

imperial information networks

Table 7.2 Proportion of court versus local and regional officeholders among authors and interlocutors in Waving the Duster Court

Local and regional

Total (out of 201 with office)

Total (per type)

Authors 85 18 111 111/168 76.5% 16.2% 55.22% 66% Interlocutors 30 18 52 52/117 57.7% 34.6% 25.8% 44.44% Note: The total number by type in column 4 is slightly higher than the sum of columns 2 and 3 because in column 2 only those officials holding three or more offices are counted. I implemented this measure to exclude those who obtained only honorary titles.

in table 7.2. The calculation of percentages of office holding by type of informant there suggests that there was a strong correlation between authorship and court office (85 out of 168 unique authors, or 50.6 percent) in Wang Mingqing’s network. In Zhang Shinan’s the correlation is somewhat lower at 41 percent. This could be a consequence of the domination of high officeholders in the world of Song publishing, but in the absence of reliable figures on the social history of authorship for this and earlier periods the evidence here is inconclusive. For now I would argue that the high profile of authors with court office in these networks reflects the interests of the note takers themselves. Authors with court office are more likely to provide firsthand and useful information about the court and those who live and work at the center of the Song world. As we will see further below, current affairs and the history of the Song court were frequently discussed topics in the notebooks examined here; those who had had access to the court were for men such as Wang Mingqing and Zhang Shinan better sources for such information. Due to their success such men, among whose number were included such celebrated authors of literary works as Su Shi and Ouyang Xiu, would also have been models worth studying, discussing, and emulating. A similar difference between authors and interlocutors holds when we examine entry information by type of informant. The way in which office was first obtained (method of entry) serves as a further indicator of social standing and career potential. The survey



liter ati networks and identities

357

Table 7.3 Survey of informants for whom there is information on their method of first obtaining office (from all three notebook [biji] networks) Notebook (biji) author and title

Regular Proportion with examinations Yin privilege entry information

Sima Guang, A Record 55 18 77/154 of Hearsay from Su River 71.42% 23.38% 50% Wang Mingqing, 121 12 171/309 Waving the Duster 70.7% 7% 55.3% Zhang Shinan, 24 2 32/83 Records of Official Travel 75% 6.25% 38.55% Note: The table shows the overall percentage of office-holding informants in the listed note­ books for whom there is entry information (column 3) and the percentages for the officeholders’ two most common avenues to office (columns 1 and 2).

table of entry information on informants across the three networks suggests that ratios for those for whom we have this information (and who were thus most likely officeholders at one point) did not vary greatly over time (table 7.3). From tables 7.4 and 7.5 it is, however, evident that by the twelfth and thirteenth centuries authors were much more likely to have entry information registered than interlocutors. Interlocutors were thus over time less likely to hold office. The majority of interlocutors for whom we do not have information about examination degrees and official positions were most likely friends and acquaintances who did not hold any bureaucratic positions or who occupied auxiliary posts only. Zhang Shinan and select other notebook authors such as Shen Gua also recognized commoners and figures who had been marginal in literati discourse as informants. Shen Gua seems to have been aware that his interest in the expertise of commoners was less common among notebook authors: “I have not presumed privately to make a record of anything concerning our sagely emperor’s policies and imperial gover­ nance, or matters associated with the imperial palace and court departments.”29 29. Egan, “Shen Kuo Chats with Inkstone and Brush,” 135.

358

imperial information networks

Table 7.4 Proportion of informants identified by type for whom there is entry information (from Wang Mingqing’s Waving the Duster) Secondary Authors Interlocutors Collectors interlocutors Informants (main role) Informants with entry  information Percentage of informants   with entry information

146 111

112 33

31 15

17 10

76% 29.5% 48.3% 58.8%

Table 7.5 Proportion of informants identified by type for whom there is entry information (from Zhang Shinan’s Records of Official Travel ) Secondary Authors Interlocutors Collectors interlocutors Informants (main role) Informants with entry information Percentage of informants with entry information

55 24

17 4

4 3

6 1

43.64% 23.53% 75% 16.67%

We may therefore conclude that Wang Mingqing’s and Zhang Shinan’s network of interlocutors mostly consisted of men linked horizontally, men who met as colleagues in various levels of the local administrative hierarchy or as prominent men of letters. Such men shared information about a wide variety of subjects including historical and classical scholarship, contemporary literary and textual criticism, family life, connoisseurship, stories of retribution, and administration as befitted men of literati standing. If it is the case that authorship had generally been associated with higher official rank, the drive of lower-elite men such as Wang Mingqing and Zhang Shinan for seeing their notes to publication can also be read as a strategy to establish relationships and seek the symbolic and sometimes real social capital attached to authorship. The presen-



liter ati networks and identities

359

tation of manuscript notebooks to acquaintances in office (see chapter 6) can then be seen as equivalent to the presentation of maps and geographical scholarship to potential patrons in a position to make recommendations for office (see chapter 4). The above comparison of select data on the sociopolitical backgrounds of informants across time underscores that both local and court office remained common attributes over time. As in the case of the producers of printed notebooks discussed in the previous chapter, we detect in the data examined above not so much a shift away from the court as a broadening of the sociopolitical background of those whose words and deeds were recorded and shared in notebooks. Even though I have thus sounded a note of caution against the tendency to interpret twelfth- and thirteenth-century data as evidence of both the provincialization and debureaucratization of elites, it is obvious that there are significant differences between Sima Guang’s network of informants and the notebooks of Wang Mingqing and Zhang Shinan. Court office was no longer common among interlocutors in the later network (but was clearly still a factor in Wang Mingqing’s interlocutor network). There was also a marked decline in the significance of the yin privilege whereby officials of a higher rank could obtain office for family members or protégés. Given that Sima Guang operated mostly among high offi­ cialdom in the eleventh century he may have had greater exposure to families who obtained office through protection of this kind. Rather than seeing these changes as another instance of familiar transitional trends, I propose that the comparison between the three networks may help explain why and how notebook writing, reading, and publishing became more common among lower-level and non-office-holding elites in the twelfth and thirteenth centuries. As Sima Guang’s notebook suggests, early notebooks were records of current court affairs that were obtained in conversations and by reading the work of contemporaries. Like other bureaucratic and semibureaucratic genres, they were gradually adapted by provincial literati in the course of the late eleventh and twelfth centuries. They maintained many of their original features but they increasingly included the voices of those of lesser rank, men who were eager to network with literati across the empire. In light of this it is

360

imperial information networks

also worth noting that Sima Guang’s notebook had reportedly become very scarce by the early twelfth century; its popularity and survival were due to the interest it sparked both at court and in the provinces from the 1130s onward.30

The Geogr a phic Distr ibution of Infor m a nts Song dynasty notebook authors developed an extensive network of informants, focused on contemporary sources, and drew from authors with substantial experience at court and in the provinces as well as from peers with lesser careers or without formal connections to officialdom. What was the geographic scope of the literati networks embedded in the selected notebooks? Were informants dispersed across the extent of Song territories or did they cluster in specific jurisdictions or regions? By plotting the geographic distribution of informants we can gauge the extent to which notebooks brought literati across the realm together, as some claimed or hoped in their prefaces, and investigate whether the greater visibility and participation of literati in their native places or adopted homes correlated with regionally defined communication patterns. In order to address such questions relating to the geographical range of informant networks I mapped both the residential (including native place) and office addresses of all those informants in the notebooks discussed above for which this information is available in CBDB. Even though the data are somewhat misleading (we do not know in most cases to what extent residential or tenure dates would have occurred before the author’s exposure to the informant), the maps serve as indicators of how geographically concentrated or dispersed each notebook writer’s network may have been. Sima Guang’s contacts mostly hailed from the Central Plain, and to a lesser extent from what is currently Jiangsu, Zhejiang, Sichuan, and Jiangxi (map 7.1). Those who served in office similarly mostly served in the Central Plain and in locations along the Yangzi River, but some also served further south in what is now Guangdong 30. De Weerdt, “The Discourse of Loss in Private and Court Book Collecting,” 415–16.

Map 7.1 Informant network for Sima Guang’s A Record of Hearsay from Su River, showing residential and office addresses. Each hexagon represents a county or prefectural seat where at least one informant was resident.

Map 7.2A Informant network for Wang Mingqing’s Waving the Duster, by residential address. Each hexagon represents a county or prefectural seat where at least one informant was resident.

Map 7.2B Informant network for Wang Mingqing’s Waving the Duster, by office a­ ddress. Each hexagon represents a county or prefectural seat where at least one ­informant was active.

Map 7.3A Informant network for Zhang Shinan’s Records of Official Travel, by residential address. Each hexagon represents a county or prefectural seat where at least one informant was resident.

Map 7.3B Informant network in Zhang Shinan’s Records of Official Travel, by office address. Each hexagon represents a county or prefectural seat where at least one ­informant was active.

366

imperial information networks

and Guangxi. As could be expected from a senior politician who had served at court for extended periods of time, Sima Guang’s network ranged across the extent of the Northern Song Empire. Even though Wang Mingqing never served in high court office and held only advisory local posts, the geographic extent of his network similarly reaches far and wide. This pattern holds even if we limit the analysis to direct contacts (interlocutors). Wang Ming­ qing’s interlocutors lived and worked predominantly along the southeast coast; many also hailed from the region around the former capital Kaifeng (map 7.2A). The base of interlocutors expanded between the mid-1160s, when the first installment was completed, and the mid-1190s, when the rest of the work was published.31 The range of the network expands when we include all jurisdictions where informants served. Map 7.2B marks all prefectures and counties where authors and interlocutors were sent. Several among Wang Mingqing’s direct contacts had experience in more peripheral areas such as the south, southwest, and the central circuits. Wang’s oral informants had served in jurisdictions in all but two Southern Song circuits; when all contemporary authors are included, all Southern Song circuits were covered by Wang’s informant network. Zhang Shinan’s network of informants in general and of interlocutors in particular was, as reflected in Records of Official Travel, numerically much smaller; it was also geographically more limited. Most of the cited interlocutors lived or had been registered in various places along the southeast coast. Among those, the small number who served in office did so in locations dispersed across the northern and central parts of the southern Song territories; the far south and the western borderlands were not included in this network. Despite its more limited range when compared to the larger networks of Sima Guang and Wang Mingqing, Zhang’s network too shows an absence of the geographic clustering of informants. In other words, the communication networks examined here were at a minimum cross-regional in scope and included informants hailing from and working in different circuits or, beyond the limits 31. See additional map in De Weerdt, “Information, Territory, and Elite Networks— Extra Materials.”



liter ati networks and identities

367

of administrative geography, different physiographic regions (maps 7.3A and 7.3B).

The Geogr a ph y of Politic a l Communic ation The proposition that by the twelfth century literati communication networks extended, even for those at the lower rungs of the hierarchy, across vast distances and multiple regions has implications for the frameworks within which we conceptualize spatial control and political integration in Chinese history. Perhaps the most influential model for thinking about space and empire in Chinese history is still William Skinner’s adaptation of central place theory and regional systems theory.32 In his early work Skinner argued that the basic building blocks of Chinese communal life, market towns, were nested in ever more encompassing marketing systems culminating in an empire-wide network that parallels and reinforces the hierarchy of administrative jurisdictions. By the 1970s Skinner rejected the model of the empire-wide integration of marketing structures and argued that physiographic features such as rivers and topography constrained the hierarchy of nested economic systems. He proposed that Chinese socioeconomic history (and by extension all Chinese history) can be best approached through the ups and downs of a small number of physiographic macroregions, which did not coincide with the more commonly used jurisdictional subdivisions. A major weakness of the later Skinnerian model, already identified in Carolyn Cartier’s incisive review of Skinner’s work and its reception history, is its inability to explain the coherence and continuity of the Chinese territories. The insistence on the discreteness of the macroregions raises the question how empire-wide territorial coherence and the maintenance of imperial traditions can be explained. Others have noted that Skinner’s model neglects or underestimates inter- and supraregional economic exchange; the same could be said about the modeling of politics. 32. For a more extended discussion of Skinner’s central place theory and a potential alternative (social network analysis), see De Weerdt, “Two Frameworks.”

368

imperial information networks

Leaving aside the valuable insight that socioeconomic and cultural behaviors operate according to logics that both differ (at lower levels of human settlement) and overlap/interact (at the higher levels) with bureaucratic/administrative structures and dynamics (Skinner’s analysis of the bureaucratic labels affixed to administrative jurisdictions is particularly revealing in this regard), political behavior receives only limited attention in Skinner’s analysis of the spatial organization of Chinese empires. Skinner’s understanding of political behavior focused on either administrative/bureaucratic behavior or nonbureaucratic local leadership in welfare operations. The former fits into a strict hierarchical model of discrete administrative organization where each lower-level unit is made to fit within one and only one higher-level unit and so is effectively controlled by the center. The latter is in his analysis part of the marketing structure and so brings elites of different types together in the teashops of the standard- and intermediate-level marketing towns. The central market towns that were absorbed within the jurisdictional structure also become the scene for the intermingling and joint leadership of local government and local elites. Missing within this picture is any consideration of communication and exchange among elites resident in standard, intermediate, or central market towns that were not adjacent within the hexagonal structures of the central place model of marketing towns.33 The geography of elite communication remains an unexplored field. The preliminary results of the informant networks examined above suggest that cross-regional communication may have been a common phenomenon among imperial elites. As others have suggested with regard to the relative neglect of inter- and supraregional economic exchange in Skinner’s work, I would suggest that political communication and politics more broadly similarly cut across the hierarchies of formal administrative organization. By further examining the dynamics of cross-regional and empire-wide literati communication in the Song dynasty and beyond we should be able to examine further whether literati political communication played a crucial role in the long-term maintenance of empire in the second millennium of Chinese history. 33. Skinner, The City in Late Imperial China, 307–46.



liter ati networks and identities

369

Titles, Topic Domains, and Identities I have shown earlier that notebook authors increasingly represented the voices of scholars with little or no direct experience in officialdom, albeit that the selected notebooks also suggest that throughout the course of the Song period court officeholders retained a central place among informants. Below I survey the titles, genres, and topics discussed in the selected Song notebooks in order to sketch the cultural settings within which their authors staked out a position for themselves. I will propose on the basis of the results obtained so far that literati authors in the twelfth and thirteenth centuries merged court and literati genres, thereby destabilizing the authority of the official record, but at the same time consolidating court, family, and literati culture (and not local or regional community) as the core foci in literati identity.

Titles a nd Title Cl asses The titles quoted, cited, or mentioned in notebooks are a first point of entry into the kinds of conversations recorded and recollected. I will first describe the overall size and structure of the reference network for each notebook and then reflect on the genres34 that appear central in them. The citation network of titles in the twelfth- and thirteenthcentury notebooks examined here exhibits a similar pattern to that of the informant networks discussed above.35 The total number of unique titles cited is relatively large (336 recorded for Waving the Duster, 102 for Keeping Guests and New Record from the Jade Shine Studio, and 130 in Records of Official Travel); most sources are only 34. I assigned genre designations on the basis of indigenous named text types (e.g., ci lyric, rhapsody, memorial, or edict) or on the basis of our own categorization of texts (e.g., divination text, Buddhist text, Daoist text). Genre is in this context thus roughly equivalent to title classes. Figures showing the distribution of genres in each of the notebooks discussed in this chapter are included in De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.” 35. Diagrams showing the distribution of titles in the notebooks are included in De Weerdt, “Information, Territory, and Elite Networks—Extra Materials.”

370

imperial information networks

cited once and a small number is cited more frequently. None of the titles at the core of the network dominates (none of the titles accounts for more than 10 percent of the total number of citations). The titles cited most frequently are generic titles such as “poem” or “edict” (these genre names are only marked as a title when they are referred to as such in the original text and a more specific title is not given in the notebook). These generic titles do not usually refer to the same text. One’s ability to muster and comment on a broad range of texts appears therefore to have been just as important as one’s ability to draw from an extensive network of informants. In all cases the number of titles cited is far greater than (and at least double) the number of authors mentioned. Overall, notebook authors referred to archival and historical compilations most frequently. This was particularly the case for Wang Mingqing and Sima Guang. In Waving the Duster Wang Ming­qing referred to Song draft dynastic histories and veritable records as well as the smaller category of (privately compiled) Song histories frequently, underscoring his investment in the historical rec­ ord of the reigning dynasty (sixty-nine citations in total). In addition to Song historical compilations Wang Mingqing also often used Song primary documents such as edicts and memorials (seventy-four citations in total). History more broadly defined, including the history of earlier periods, was also of great interest (thirty-five citations). The titles cited most frequently in Waving the Duster provide further evidence that archival and historical compilations such as the draft dynastic histories of selected reigns and the veritable records discussed in chapter 1 were available outside court libraries and were used in literati discussions of political figures and events. Wang Mingqing cited the dynastic histories of the Northern Song reigns (Sanchao guoshi 三朝國史, The history of the three reigns, and Liangchao guoshi 兩朝國史, The history of the two reigns, covering the reigns from Taizu 太祖 [r. 960–76] through to Yingzong 英宗 [r. 1064–67]) more frequently than any other source. Sima Guang’s notebook drew from an extensive network of inter­ locutors; the number of written sources was, however, comparatively small (twenty-four) and drawn from a narrower range of genres. Two compilations of models of sagely government (shengzheng 聖政)



liter ati networks and identities

371

and standard court documents such as edicts and memorials were the main sources cited in Sima Guang’s work. The citation frequencies in A Record of Hearsay from Su River were, unlike the other citation networks, heavily skewed toward the top-ranked sources with the top two titles accounting for nearly half the total number of citations and the top four for two-thirds. As all of these fell within the category of court official documents, it becomes apparent that A Record of Hearsay from Su River was in the first place a record of interactions among high court officials and of Song policy documents. For its twelfth-century readers it became a history of late tenth- and eleventh-century political practice through the personal lens of one of its most prominent practitioners and a repertoire of cases for use in twelfth-century literati discussions about Song court politics. Wang Mingqing, for example, wrote that Sima Guang’s notebook had been a frequently used source for the original edition of Shenzong shilu 神宗實錄 (The veritable records of Shenzong’s reign) and that its contributions were subsequently edited out in the revised edition undertaken in the mid-1090s.36 He also corrected the occasional error, noting, for example, that the version of the story about the fight that Emperor Taizu had set up between the top graduates of the palace examination to decide their rank, as reported in Song Minqiu’s notebook, was more reliable than the version reported in A Record of Hearsay from Su River.37 Second on the list of most frequently referenced genres are notebooks themselves. Wang Mingqing cited notebooks, personal notes, and memoirs about sixty times in Waving the Duster; these are also among the most frequently referenced genres in his other notebooks. Zhang Shinan similarly refers to several notebooks. The prominence of these texts in the later notebooks underscores the extent to which the authoring of notebooks and allied genres such as the memoir (of one’s career, an event, or a place) had become a network phenomenon. 36. Wang Mingqing, Yuzhao xin zhi, 1.1; Levine, Divided by a Common Language, 14–15. 37. Wang Mingqing, Yuzhao xin zhi, 4.64.

372

imperial information networks

Wang Mingqing commented on the notebooks of others and previous installments of his own notebooks. His interest in the notebooks of others underscores the extent to which the notebook was understood to be a conversational genre: citations from and corrections to other notebooks suggested that behind the note taking was an individual engaged in ongoing conversations with peers. Based on conversations and further research, he also revisited, corrected, and augmented entries from his earlier notebooks not only within the Waving the Duster series but also in his other notebooks. For example, Shen Gua, whose notebook Mengxi bitan 夢溪筆談 (Notes by Dream Creek, 1087) contained among many subjects obser­vations on different types of official documents and processes of official communication, was called upon for advice when Wang Mingqing obtained a memorial approved by Emperor Yingzong but without the full name of the councilor or vice-councilor who would have submitted it.38 From Shen Gua’s notes Wang Mingqing was able to deduce the meaning of a missing signature: it indicated that it had been submitted as a regular memorandum for which at the time only the seal of the officeholder was necessary. He added that he was uncertain how such reports were signed at the time of writing and thus invited others to find out and report back. Wang Mingqing’s notes explained in this case the material object whose unknown author raised questions for him and presumably also the collector from whom he obtained it. His comment also threw light on the meaning of Shen Gua’s short note on the different ways in which regular and more urgent business were reported and processed. To us it demonstrates how very applicable the brief observations of Song scholars on various aspects of bureaucratic communication could be. Self-referentiality had become a core feature of the notebook genre by the Southern Song. This self-referentiality is perhaps nowhere better illustrated than in Zhang Zi’s 張鎡 (d. 1207) Huang­ chao shixue guifan 皇朝仕學規範 (Models for official service and learning from the reigning dynasty). A late twelfth-century printed edition (preface 1176) featured a list of recommended titles at the front, titled “A List of Books Compiled for Models for Official Service 38. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.69–70.



liter ati networks and identities

373

and Learning from the Reigning Dynasty” (Huangchao shixue guifan suo bian shumu 皇朝仕學規範所編書目). This list consisted of one hundred titles produced since the beginning of Song rule and deemed to be essential reading for those aiming to serve in gov­ ernment or to cultivate the qualities that were considered essential for the political elite. A dozen titles of biographies of worthy officials arranged by reign period are listed first, followed by selected collected works (wenji 文集), recorded conversations ( yulu 語錄), notebooks (tanlu 談錄, biji 筆記, zazhi 雜志, zalu 雜錄, bilu 筆錄, keyu 客語, etc.), memoirs ( yishi 遺事), comments on poetry (shihua 詩話), and classified notebooks (leishi 類事, leiyuan 類苑), covering the period of the early Song up to the time of writing as suggested by the inclusion of Hong Mai’s first two collections of stories Yijian zhi 夷堅志 (1160, 1166)39 and Lü Zuqian’s 呂祖謙 (1137–81) notes (Lize wenshuo 麗澤文說, Discussions on composition from Lize). No fewer than one-third of these titles were notebooks in the strict sense; if shihua (which focus more exclusively than notebooks in the broad sense on comments on poetry and literature) and yulu (which tend to focus on the conversations of a teacher with disciples and are often recorded by disciples and thus differ from notes taken by the author on conversations and personal observations on reading and site visits) were to be included the total number of notebooks in this list is well over half. Apart from Models for Official Service and Learning from the Reigning Dynasty, an earlier classified notebook titled Huangchao shishi leiyuan 皇朝事實類苑 (Classified collection of affairs from the reigning dynasty; ca. 1145) similarly included around twenty notebooks in its list of fifty to sixty titles40 and an additional small number of shihua and yulu. This title was included in the list of titles consulted in Models for Official Service and Learning from the Reigning Dynasty. Around the same time that Zhang Zi compiled 39. Inglis, “A Textual History of Hong Mai’s Yijian zhi,” 311, 323. 40. I counted forty-nine titles in the list provided at the end of the Shanghai guji ed. (Jiang Shaoyu, Songchao shishi leiyuan, pp. 1033–34). The Siku editor wrote, however, that Jiang Shaoyu cited over sixty titles in this work (SKQS ed., abstract; also Shanghai guji ed., p. 1032).

374

imperial information networks

his notes, according to his preface in the first place for his own personal development but also for the benefit of others, Li Yuangang 李元綱 (fl. 1165) compiled a similar notebook, titled Houde lu 厚德錄 (A record of enriching virtue), which also drew on over twenty notebooks and other historical, archival, and encyclopedic sources covering the history of the Song period. None among the three compilers of these notebooks appears to have held high office: Li Yuangang was a teacher, Jiang Shaoyu held a small number of local positions, and Zhang Zi held mostly titular positions. Their career profiles were thus comparable to Wang Mingqing’s. The elevation of notebooks as a principal source in texts providing “models” for governance highlights several features of the genre. First, when collecting and editing their personal notes Song literati were engaged in a shared endeavor. They read and commented upon the notes and conversations of others in order to make visible the standards of cultured behavior and their own embodiment of the requisite characteristics of the literatus. Second, the centrality of notebook literature also has implications for how we understand the literati reception of the court archival and historical official sources. Stephen Owen and others have proposed that notebooks destabilized the authority of official historiography. Historians in the employ of the court aimed to establish or at least give the impression of the historically true; in this regime of truth there was no room for factual error and speculation. Notebook authors, by contrast, proved willing to engage information that had not yet been fully verified. They by and large intended to lend credibility to the stories and factoids shared, but allowed space, in varying degrees, for the probable.41 The emphasis on the destabilizing effect of notebook literature gives a new twist to the older standard understanding of the relationships between official and private historiography: the latter was in the words of several notebook authors intended to “supplement” (bu 補) the former. This rendering is somewhat euphemistic but it also captures an important aspect of literati attitudes toward official historiography and state documents. The reference frameworks of Song 41. Owen, “Postface.”



liter ati networks and identities

375

notebooks show that literati used such sources with some regularity and in combination with notebooks and other genres of private historiography. By pointing out errors, hidden agendas, and oversights they destabilized the authority of such sources to some extent. Yet, at the same time literati thereby acknowledged state documents and archi­val compilations as common reference points, as did Wang Mingqing in his story about his ability to recall biographical details from the draft histories. There was always some tension in the relationship between court and literati genres, but there was a mutuality to it: court historians requested the notebooks of Wang Mingqing and others and incorporated those in future editions of the draft histories. Here too then we see the court adapting to a new informa­ tion regime, one in which literati not only adapt court genres but also produce on a large scale in genres interacting with the court’s record. The third and fourth major genres represented in the work of Wang Mingqing and contemporaries suggest that notebooks were but one among several literati genres in which lower-level literati were increasingly heard and seen. Wang Mingqing frequently referred to biographical materials, including biographies in histories (zhuan 傳), tomb inscriptions (muzhi ming 墓誌銘), and life histories (xingzhuang 行狀) in Waving the Duster (twenty-eight citations). The selected notebook authors similarly drew extensively from collective and individual biographies ( yanxing lu 言行錄, mingchen zhuan 名臣傳). As the latter titles suggest, biographies provided models of exemplary and not so exemplary family lives and official careers. In the notebooks, biographical accounts also became sources for discussion and debate about the ups and downs of individuals and the achievements of their family members. The literary output of peers formed a final major genre cluster in the source base of Wang Mingqing’s notebooks. He referred frequently to poems (shi 詩, ci 詞, and fu 賦) in Waving the Duster (forty-eight citations); more so than other genres, poems were quoted in full or in part. He drew upon prose genres such as letters (shu 書, qi 啓), inscriptions ( ji 記), and more generally the collected writings of one or more authors ( ji 集, wenji, jiaji 家集) in equal measure (about fifty citations). The weight attached to collected writings (wenji) and poetry talk (shihua) in the lists of sources cited

376

imperial information networks

in Models for Official Service and Learning from the Reigning Dynasty underscores that the lives as well as the literary output of better- and lesser-known Song authors had become common reference points for Song literate elites. The upward trend in the referencing and preservation of twelfth- and thirteenth-century collected writings and notebooks on poetry can therefore be seen in a similar light as the increasing numbers of notebooks and references to notebooks during the same period. Even though we will have to wait for further research on the cultural and social history of text production for this period to learn whether the listing and footnoting of titles in these genres and the upward trend in their production coincide with a social broadening of their authorship and readership comparable to that attested in the case of printed notebooks above, we may conclude that these genres played an increasingly visible role in literati communication and networking. The interest in the letters sent by others and in correspondence etiquette more generally is a particularly revealing example of the weight attached to the social value of prose and poetic writing. Wang Mingqing, for example, noted that standards had fallen in recent times: Older conventions stipulated that at the end of a letter one wrote “this is the letter” or “this is the end.” To show slightly more respect, one wrote “I once more show my respect.” Even for those of high standing and great reputation one would not write more than “I kowtow” or “I once more show my respect.” Unless it was for a father or a brother one would not add characters upside down. For those of the rank of councilor or vice-councilor or higher, one would add the two characters taihou 台候 [showing concern for the addressee’s well-being], but one would not dare do this for others. One can check this in our predecessors’ The Correspondence of Famous Ministers. Now it is all different. This is truly regrettable.42

The decline of such standards appears to have been a shared concern, albeit that the awareness of the standards themselves went 42. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.36.



liter ati networks and identities

377

hand in hand with the publication of manuals for letter writing and annotated collections of model letters.43 Such models had been around since Tang times at the latest, but by Wang Mingqing’s lifetime they appeared in print and were targeted at growing numbers of literate elites interested in learning about the most appropriate ways to write to others of various rank and gender for different needs.44 The major clusters of interest identified in the citation network of Waving the Duster recur in Wang Mingqing’s two other notebooks and Zhang Shinan’s notes, albeit with different emphases. Moreover, the wide variety of minor genres appearing in the later notebooks suggest that despite these emphases authors like Wang Mingqing drew on a wide range of other materials including gazetteers, legal texts, or family instructions. When we compare Wang Mingqing’s source base to that of Zhang Shinan in Records of Official Travel, it is apparent that even though draft dynastic histories, veritable records, official reports, and notebooks were also of interest to him, he turned more frequently to poetry, pharmacopeia, gazetteers, and a court-sponsored rhyme book. Pharmacopeia belonged to a genre of medical literature that figured prominently among the titles printed by commercial book providers and that found a place in the average private collection. Basic fluency in medical discourse had become a desirable quality of the civilized person, and medicine was therefore a recurrent topic in the later notebooks examined here. The other sources high on Zhang Shinan’s list helped explain the unfamiliar terms and practices Zhang encountered on his outings. In his notes Zhang Shinan related to a mobile audience of literati readers the personal experiences of a traveler who had seen various parts of the Song Empire and who had read about the experience of others who had done the same. Despite the continuities already described, some trends emerge from the comparison between Sima Guang’s A Record of Hearsay 43. Tsui, “Writing Letters in Song China.” 44. See, for example, Sheng Song qianjia mingxian biao qi, Song print preserved at Beijing Library.

378

imperial information networks

from Su River and the later notebooks. Two of the major clusters of interest in the later works are absent from Sima Guang’s. Notebooks and the prose and poetry of peers are of little or no significance in A Record of Hearsay from Su River. Even though other eleventh-century notebooks might differ substantially from A Record of Hearsay from Su River in this respect (an analysis of Shen Gua’s Notes by Dream Creek would show a very different constellation of sources), it appears that the gradual expansion of the number and social background of notebook authors (including greater numbers of lower-level literati) led to the emphatic inclusion of a wider variety of genres. For these elites, engagement with genres more closely associated with what Bin Wong has called the middle realm— the space that intersects with both the public and the private—also appears to have become a greater preoccupation.45 Letters, inscriptions, poems, and notebooks were not necessarily private and frequently crossed over into the public realm, as suggested by Wang Mingqing’s revelations on patronage seeking. Reading and commenting on others’ works in these genres became the medium by which lower-level literati such as Wang Mingqing and Zhang Shinan connected with peers. The court-oriented genres of official documents and archival compilations and the biographical texts that dominated in the genre map of A Record of Hearsay from Su River retained a presence in the later notebooks, albeit that the proportion of citations from bureaucratic documents fell as the note-taking habits of the eleventh-century courtier were adapted for the self-representation of lower-level literati.

Topic M a ps From an entry-by-entry analysis46 of Wang Mingqing’s Waving the Duster it emerges that court, family, and literati life were for him the core areas of concern. Below I briefly consider what implications 45. Wong, “Social Order and State Activism in Sung China.” 46. A note on the methodology is included in appendix 2. It is worth noting that the approach followed here is intentionally not based on a literal interpretation of the headings assigned in some editions of the notebook. Collections of fragmentary texts were



liter ati networks and identities

379

such a configuration of interests has for our understanding of the relationship between court and literati and of literati identities. Wang Mingqing devoted much attention to the court and the central bureaucracy (fig. 7.9), focusing on such topics as the emperor and the imperial family; the Song court and its environs (the palaces, the capital city, and its challengers on the periphery); the organization and operation of central offices; the words and actions of prominent eleventh- and twelfth-century politicians (esp. Qin Gui 秦檜 [1090–1155], whose reign had come to an end ten years prior to the publication of the first installment of Waving the Duster); in Song times regularly organized topically and/or presented with headings and subheadings in order to make their contents more easily accessible. Encyclopedias and anthologies were organized by main and subcategories, and headings were also used to annotate passages in these texts. Printers applied this paratextual grammar to notebooks as well. Some editions of Wang Mingqing’s Waving the Duster come with a table of contents that listed each entry in numerical order and provided brief titles for all. These titles summarized in succinct lines the central topic and storyline; in many cases they did so according to standard rhetorical patterns that captured both the particularity of the entry and the more general type of the entry. In Waving the Duster, for example, entries about the origins of particular practices were noted as “X FROM Y STARTED” (X zi Y shi, X 自 Y 始); the gist of entries about trivia or rare individual or family achievements was presented in the following manner “DURING THIS DYNASTY OR OTHER TIME PERIOD// PARTICULAR FEAT// NUMBER OF PEOPLE OR INSTANCES” (benchao 本朝 / guochao 國朝 // X // N UMBER); fuller accounts of an event or life were encapsulated in the form “X FROM BEGINNING TO END OR DEVELOPMENTS” (X benmo 本末 / shoumo 首末 / yange 沿革). Similar headings and patterns appear in an early edition of Songchao shishi leiyuan. The hermeneutics of the original headings is an important part of textual and cultural history. The original headings and their sequential arrangement help us navigate through the text and help us interpret the text according to guidelines provided by its author or editor. Reading or scanning through the notebook with the aid of these early reference tools results in a reading very different from that of most historians, who have mined the miscellaneous content of notebooks through the help of concordances or, in recent years, natural-language (keyword) searches. The latter tools have allowed historians to identify and gather together passages on the basis of particular characters or phrases, but the use of smaller lexical units in scanning the text has led to a fragmentation of the text and the loss of an overview of the subjects and arguments covered in the notebook in the aggregate. In other words, the original content map has been forgotten and the present-day reader in most cases is not replacing that map with a topic map that may help capture the content of the entire notebook for present-day historical analyses.

Fig. 7.9 Topic map for Wang Mingqing’s Waving the Duster. The diagram shows all topics applied six or more times in my interpretation of individual entries. Larger ovals represent higher frequencies, with cuts set at 100 + (2 topics), 80 + (1), 50 + (1), 40 + (1), 20 + (8), 10 + (23), and 6 + (20).

382

imperial information networks

and the ties that linked the court and central government to the provinces (recruitment, communication, and government ethics). Despite many episodes relating to local government in general and areas of government, such as security and water management, that linked local and central hierarchies of administration, Wang Ming­ qing’s notebook was on the whole court-centered. Questions relating to local government and, occasionally, local characteristics came up (even more so in Zhang Shinan’s notebook), but these did not become the ground for a celebration of local identities. In these and other genres that were increasingly practiced and published in the twelfth and thirteenth centuries, literati related local places and local practices to regional and empire-wide structures. The overall focus on the court was counterbalanced by a discussion of the interests and practices of literati men and the fate of their families, not by a focus on the local. In the following discussion I present some examples from the three clusters and conclude by explaining how court and literati life were closely entwined in the minds of twelfth-century literati. Imperial Sovereignty The virtue of the emperor is a frequently recurring topic throughout the four installments of Wang Mingqing’s Waving the Duster. He represented emperors throughout Song history as exemplars of moral excellence, whose suitability for succession was confirmed in auspicious signs and whose words and actions conformed with literati models of proper interaction. A series of anecdotes about Emperors Gaozong and Xiaozong 孝宗 (r. 1163–89), whose assumption of the throne had been less than straightforward, suggested that their ascendancy was justified by both suprahuman interventions and the recognition by others of their worth: In the winter of the second year of the Jianyan reign, Gaozong fled from Jiankang to avoid the northern invaders and went to Zhedong. When he first crossed the Qiantang River on his way to Xiao Mountain, some stood by the road to pay their respects. They held a sign saying, “Zhao Bushuai [趙不衰, 1107–79] [Zhao (the name of the



liter ati networks and identities

383

imperial family) Will-Not-Perish], scion of the dynastic family, and others greet you.” The Emperor was very pleased, and addressed his entourage, “With an omen like this, I should not worry.” He ordered that Bushuai be elevated three ranks. Even though [the emperor] had fled to sea, from this point onward the realm was by and large settled. Bushuai was the father of Jun 俊 [趙善俊, 1132–95]. This is equivalent to the sign of Song victory when Emperor Taizong conquered Hedong. At this time there was also the omen revealed through Zhao Li [趙立 (Zhao Will-Stand)] and Bi Sheng [畢勝 (Eventual Victory)] during the selection of craftsmen for the imperial boat.47

This and other episodes echoed the stories about legitimate succession reported in the court compilations and commercial reproductions of The Sagely Government of Emperors Gaozong and Xiaozong, even though Wang’s selections were not included in these works.48 Wang Mingqing shared the story of Zhang Hao 張浩, a doctor in Xiuzhou 秀州, who reported to him personally that he had seen a brightness lighten up the evening sky just before the birth of the child that would later become Xiaozong.49 He included two other personal witnesses attesting to the special protection enjoyed by the emperor who passed away in 1194, shortly before the last installment of Wang Mingqing’s notebook was completed. Zhao Zidao / Yanwo 趙子導 / 彥沔 told him that the emperor had been chosen as crown prince from among ten boys. He would have come in second if not for the fact that the fatter boy who had been nominated first messed up his chances by kicking a cat that 47. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 4.269. 48. The extant edition of Zengru mingru jiangyi Zhongxing liangchao shengzheng 增入 名儒講義中興兩朝聖政 (The sagely government of the two reigns of the intermediate restoration, with explanations by famous scholars) contains a section in the categorical index titled “according with the mandate of Heaven” ( fu ming 符命) that includes omens. This category contains two stories about omens at a future emperor’s birth, both in the form of a brightness lightening up a dark space. Ibid., fenlei shimu, 1a; relevant stories can be found in 2.17b, 27.12a. Wang Mingqing records a similar witness account about an omen occurring on the day the future Emperor Xiaozong was born, but this account differs in details and involves different witnesses. 49. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 4.270.

384

imperial information networks

had wandered into the palace. Emperor Gaozong thereupon changed his mind, observing, “This cat just happened to pass by, why s­uddenly kick it? How could someone who acts so rashly assume a heavy responsibility?”50 And later on in the prince’s career, as relations with the Jurchens became tense, a fortune teller allegedly predicted the abdication of Emperor Gaozong and Xiaozong’s early ascendance to the throne. Li Zhengzhi 李正之, who held office in Chuzhou 處州 in 1161, related to Wang Mingqing that a local fortune teller wrote the following on a piece of paper when asked to predict the outcome of the hostilities between Song and Jin: “He [Emperor Xiaozong] is in place.” The following year Xiaozong’s enthronement was announced.51 When sharing such stories, Wang Mingqing, his interlocutors, and his readers confirmed the legitimacy of the ruling house and their loyalty to it. At the same time, they cast the emperor as a persona who spoke to them and whose interests closely coincided with their own. The emperor acted with wisdom and foresight, as demonstrated in Emperor Gaozong’s decision to demote the prince who kicked a cat. He was also portrayed as a witty interlocutor on a par with those whose words featured elsewhere in the notebook. Perhaps the example that best encapsulates how the emperor emerges in the pages of the notebook as an embodiment of literati wit is the following: During the Dazhong Xiangfu reign [1008–16] Zhangsheng [章聖; Zhenzong 真宗 (r. 998–1022)] sacrificed in Fenyin. When they [the imperial procession] arrived at the foot of Mount Tai several tens of thousands of people had gathered to watch. They crowded the streets so that even when the emperor’s arrival was announced, [the procession] could not move forward. The emperor inquired about this with his entourage. Someone responded, “Village people fear the local commander. He can bring them under control.” An order was then issued to summon him forthwith. Shortly afterward a youngster wearing a green robe came forward on a galloping horse and an50. Ibid., 4.270–71. 51. Ibid., 4.271.



liter ati networks and identities

385

nounced at the top of his voice, “An official is coming!” They dispersed in a hurry. The emperor joked, “Am I not an official?” [我不是 官人邪?].52

The figure of the emperor that emerges from the passages recorded in Wang Mingqing’s notebook is of course not that of a mere official. Imperial sovereignty is central to Wang Mingqing’s notebook and it is represented very effectively through the use of direct speech in this and many other examples. The emperor spoke in his own voice, but he spoke and joked in a register familiar to a literati readership. He was fallible, as shown in the episodes about Emperor Huizong’s love and capacity for drink53 or Xiaozong’s falling through a hole in a Buddhist bell tower.54 In these scenes fallibility is unproblematic because the emperor compensates for it through a commitment to uphold the normative standards of good government. Even Emperor Huizong, who was in later historiography taken to task for the extravagant spending under his regime, appears in several episodes as a paragon of literati expectations. The notebook author quotes in full several inscriptions commissioned by the emperor for the grand parties held at his palaces.55 In these Huizong guided his guests to the rooms where antiquities were stored and introduced them to the vessels, books, paintings, and religious inscriptions that had been laid out for them. He prepared and served tea for those invited and further treated them to music, dance, snacks, wine, and the company of exquisite palace women. When the guests appeared ready to take their leave, the emperor retained them urging them to stay until the wine had taken effect (bu zui wu gui 不醉無歸). At his request the guests composed lyrics in turn until suddenly he asked Cai Jing to recite some lines from a lyric the latter had composed on the occasion of a similar party 52. Ibid., 2.141. Italics are mine. 53. Ibid., 4.276–81, 301–2. 54. Ibid., 3.224. For another discussion of the reconciliation of the emperor’s flaws in the representation of sovereignty, see Jack Wei Chen, The Poetics of Sovereignty. 55. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 4.276–81.

386

imperial information networks

organized by Emperor Zhezong 哲宗 (r. 1086–1100) more than twenty years earlier. Cai Jing noted he could no longer remember the lines, thus providing an opportunity to Huizong to impress his guests with a fine memory. Huizong promptly recited the lines he had heard only once twenty-four years earlier. The lines then became the occasion for emperor and minister to remember a relationship that had lasted through two decades. Cai Jing stopped his description of the party here, noting “I request to commemorate this event for later generations so that it will be known that today’s party was not simply about food and drink.”56 The figure of the emperor thus comes close to the ideal of the literatus: he is the perfect host entertaining his visitors with delicacies and curiosities, impressing them with the aesthetic qualities of his palace complex, challenging and outdoing them in cultural competition, and reminding them of the relationships that matter and that make men into what they are. Councilors such as Wang Anshi, Cai Jing, or Qin Gui are time and again shown to fall short of expectations, but the actions of emperors selected for inclusion in Wang’s notebook underscore that imperial sovereignty stands for moderation, the protection of the common people, and discernment to select the best for office and to restrain them when necessary. Wang Anshi was overruled when he tried to protect a calligrapher who inscribed inappropriate words on the lapel of a female singer57 or when he proposed a ritual that would in the emperor’s mind have shown insufficient deference on the emperor’s part toward a hermit the latter had recommended.58 Gaozong saw through slanderous remarks cast upon Zhang Jiucheng 張九成 (1092–1159) when the latter was accused of having gone over to the Jurchen side.59 The Fate of Prominent Families The continuity and the fragility of family traditions formed a second cluster of interest in Wang Ming­ qing’s notebook. The stories Wang selected drew attention to the 56. Ibid., 4.279. 57. Ibid., 4.295–96. 58. Ibid., 4.266–67. 59. Ibid., 3.253.



liter ati networks and identities

387

decline of elite families from prior dynasties and even from the more recent Song past in ways that substantiate modern scholarship on the demise of Tang aristocratic families in the course of the ninth and tenth centuries: “The Cui, Lu, Li, Zheng, and the Wei and Du from south of the city walls continued to hold on to posts and titles across generations and were prominent in their time. Coming down to our dynasty they no longer produced men of note.”60 In the face of such fragility of family fortunes the author set out to document cases of continuity and provided examples of the ways in which families had maintained status. During the Five Dynasties there were three men with the surname of Lü who became vice-minister. Their clans all had progeny that served our dynasty in the capacity of councilor. Lü Qi [呂琦] was viceminister of war during the Tianfu reign of the Jin dynasty [936–44]; his great-grandson Wenhui [文惠 / Duan 端] was councilor to Tai­ zong. Lü Mengqi [呂夢奇] was vice-minister of war during the Chang­xing reign of the Latter Tang dynasty [930–33]; his grandson Wenmu [文穆 / Mengzheng 蒙正] was councilor to Taizong, and his great-grandson Wenjing [文靖 / Yijian 夷簡] was councilor to Renzong. Their success was the greatest as already set out in the first installment. Lü Xianxiu [呂咸休] was vice-minister of revenue during the Xiande reign of the Zhou dynasty [954–59]; his seventh-generation grandson Zhengmin [正愍 / Dafang 大防] was councilor to Zhezong. This is extraordinary!61

Continuity was difficult to achieve, but the examples of the accomplishments of some families in placing male relatives in high positions simultaneously or across generations suggested that lasting success in office and cultural prominence across generations were attainable. Wang Mingqing continued his note on the demise of the great Tang families with a description of the naming strategies through which families had held their members together and had 60. Ibid., 1.20–21. For modern accounts of the demise of the great Tang clans, see Johnson, “The Last Years of a Great Clan”; also Johnson, The Medieval Chinese Oligarchy; Tackett, The Destruction of the Medieval Chinese Aristocracy. 61. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.105.

388

imperial information networks

thus provided the means by which they could network and support one another and be identified by others. Like other prominent families in the early Song, the Lü adopted the practice of a shared character in the first names of all male progeny so that they could all be identified as members of a particular generation. Other families became distinguished through the addition of epithets to their names; Wang pointed out that such epithets could derive from the location or particular features of their residences (a tree, gardens, the proximity to a bridge or a gate, or the name of a neighborhood), the official title of one of their family members, and, for those who had moved, the name of their native place. The list of families whose reputations should be known to the civilized elite of Song times was extensive: “Families like the Cai of Putian, the Xiao of Baisha, the Hu of Piling, the Shi of Kuaiji, the Chen of Fanyang, the Wang of Xin’an, the Shen of Wuxing, the Bao of Longquan, these are all reputable families of the present day. Moreover, there are still several families in the capital whose names derive from their neighborhood; they can’t be listed in full.”62 Wang Mingqing singled out scholarship as the route to prosperity chosen by his relatives. The Zeng family, to whom the Wangs were related by marriage, invested heavily in education. Wang Ming­ qing recalled the words of his maternal grandfather, who had shared with him the hardships of his forefathers. Zeng Gong 曾鞏 ( j. 1057) and one of his brothers sat and failed the examinations several times, but Zeng Gong continued to teach his younger brothers, five in total. Their repeated attempts became the object of ridicule as a neighbor circulated the following poem about them, “Once every three years, the examinations are held, / failing two talents of the Zeng family. They resemble pairs of swallows in between the eaves; / as one pair takes off, another comes flying.” Their efforts paid off, as six men from the family eventually passed the local and the departmental examinations in the late 1050s.63 Wang Mingqing further showed that examination success could be maintained over time through the example of the Zhang family from Pucheng. They had produced 62. Ibid., 1.20–22. 63. Ibid., 2.154.



liter ati networks and identities

389

top-ranking candidates in the departmental, palace, avoidance, Imperial College, and Kaifeng prefectural examinations and continued to cultivate top-ranking graduates after they moved away from Pucheng in Fujian. They had their own stele of successful candidates carved in a similar way to the steles produced for all successful candidates at the Imperial College.64 As in the case of the anecdotes about the Song princes selected to become emperors, destiny and worth came together in the stories about family fortunes. The success of the Zeng brothers had been foretold by a commoner woman. The wife of Zeng Bu 曾布 (1036– 1107) in turn predicted the success of the family of a protégé of her husband. Zeng Bu met Li Zhuan 李撰 (1043–1109), a teacher in the local government school, while serving as a magistrate in Zhending. The Li family was of moderate means. One day Mrs. Zeng invited the Li family and the family of a military commissioner, surnamed Song, to a party. The families provided a lesson in contrast. Mrs. Song arrived dressed in pearls and jade and looked splendid; Mrs. Li wore a worn-out dress and looked somewhat disheveled. The son of Mrs. Song was a beauty to behold while the sons of the Li family appeared rather dull even though they spoke with elegance. The other invitees were amused by the sight of the Li family, but Mrs. Zeng gently observed, “Even though the teacher may now be poor, his sons are endowed with outstanding talent which at a different time will outrank that of others. The son of the military commissioner looks tidy and neat but his talent is pedestrian.” It turned out that four among the five sons of Li Zhuan passed the examinations and two rose to the top ranks of the central government. By contrast, the son of the military commissioner only rose to the post of audience usher.65 The historical record, court gazettes, and notebooks attested to rapidly shifting fortunes of individuals and their families. Fortunetelling continued to be part of historical practice. Prognostication was shown to have facilitated the making of difficult decisions and, retrospectively, after the prediction had materialized in fact, helped 64. Ibid., 1.22. 65. Ibid., 2.166.

390

imperial information networks

explain the course of events.66 In Waving the Duster natural signs and predictions, regularly made by beings outside the circles of literate male scholars and officials such as women, priests, monks, and commoners engaged in their daily trades, provided ample evidence that the seeds of prosperity and the warnings of disaster can be slow to manifest themselves but can be detected in the course of events. Literati Life The core aspects of literati life as Wang Mingqing portrayed it in Waving the Duster are well illustrated in the story of Wang Shen 王莘 (fl. 1100s–1110s) and Song Huizhi 宋惠直. In the 1100s Wang Mingqing’s grandfather served as the prefectural magistrate of Jiujiang, where he encountered Song Huizhi, who then served as an assistant magistrate in the subordinate County of Dehua 德化. Song Huizhi was the son of Song Jingzhan 宋景瞻, who had sat the jinshi examinations in the same year as Wang Shen.67 The tongnian connection and the son’s appetite for learning were sufficient cause for Wang Shen to take the son on. He excused him from his clerical duties and trained him instead for the notoriously difficult polymathes (hongci 宏詞) examination, allegedly setting questions for him on a daily basis. He introduced Song Huizhi and his work to several influential men of the time, and, in due course, Huizhi did indeed pass the examination, a feat that few managed in Song times. Wang Mingqing further commented that subsequently Song Huizhi and later his son enjoyed quite a reputation, but that he had lost touch by the time of writing and that it was possible that the Song family had forgotten about their indebtedness to the Wang family.68 Learning, composition, literary exchanges, the critical appraisal of others’ work (Wang recalled from memory some lines from a song and recorded his grandfather’s comments on it), patronage, and peer and family networks (their formation as well as their dissolu66. Adler, “Chu Hsi and Divination”; Liao Hsien Huei, “Exploring Weal and Woe.” 67. According to Zhang Jian (“Wang Zhi ji qi jiazu”), Wang Shen sat the jinshi examination at the departmental level in 1067. See also Zhang Minghua, “Wang Shen kao.” 68. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 3.239.



liter ati networks and identities

391

tion) come together in Wang Mingqing’s recollection of this story. Collecting, borrowing, and printing books, alongside the critical interaction with texts and artifacts, were also frequent topics of discussion; the notebook further served as a vehicle for the circulation of rare manuscript texts (see chapter 1).69 Learned conversation, already discussed above, was seen as not only a requisite skill of the gentleman, but also an important attribute of the elite woman and her family. When the wife of Qian Chen (d. 1161) 錢忱 visited Empress Qinsheng 欽聖 (1046–1101), she and a group of other women who had come to the palace from the same region noticed an inscription at the entrance of a palace which read “受釐殿.” The other women misinterpreted the second character, reading it by the most common of its readings “li.” Mrs. Qian, who was a granddaughter of the prominent mid-eleventh-century statesman Tang Jie 唐介 (1010–69), corrected them noting that in this case 釐 should be read “xi,” meaning bliss. She added that “receiving bliss” was in this case a reference to a hall by the same name in the Han dynasty Palace of Weiyang, “receiving bliss in the Main Hall [of the Weiyang Palace].”70 Qinsheng was pleased and commented, “The sons and daughters of good families are in the end distinguishable hereby.”71 We can now correlate the networks of contemporary interlocutors and of Song texts discussed in the first part of this chapter with the clusters of interests. Wang Mingqing drew from the court archival and historical compilations produced at the court and highlighted how his notes on conversations and transcriptions of primary texts could supplement the archival record of the Song dynastic house and its bureaucracy. In subsequent centuries Wang’s notebook has therefore frequently been seen as a work of history.72 69. De Weerdt, “Manuscript and Print in Twelfth-Century Song China.” 70. Sima Qian, Shi ji, 84.2502. 71. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.171. 72. This conception is widely shared and often an unspoken assumption, in large part due to the way in which historians use notebooks as “historical materials” (shiliao). For some examples that relate directly to Wang Mingqing’s Huizhu lu, see Yan Yongcheng, “Shilun Wang Mingqing de bu shi chengjiu”; Wu Xiaoping, “Huizhu lu yu Wang Mingqing de xueshu chengjiu.” For a critical review of the Siku editors’ review of Song

392

imperial information networks

By placing his notes in the broader context of the circulation of notebooks in the twelfth and thirteenth centuries and through an integral overview of the topics addressed in the notebook we can see that the kind of history in which Wang Mingqing was mostly interested related to very specific, oftentimes period-specific, concerns such as imperial sovereignty and legitimacy, territorial sovereignty (the periphery), the development of bureaucratic organization, nomenclature, communication during the Song and up to the time of writing, the performance of and relationships among prominent families across the realm, or the critical evaluation of the most influential councilors in recent political history. Such concerns were not simply of a historical kind, especially not if history is then understood as an act of individual scholarship or family specialization divorced from broader political and social interests. By placing and reading conversations about court, bureaucracy, elite families, and cultural life and discursive practices next to one another it becomes obvious that these cannot be compartmentalized as distinct areas of interest, as is commonly done in descriptions of notebooks (in part because they are used as “source materials” in distinct disciplines with distinct research interests). By reading and conceptualizing them together we can begin to appreciate how learned and sometimes witty interactions with named peers and named texts about all of these concerns in combination shaped and represented the identities of their authors. These identities were constructed interactively through the social networks recollected in notebooks and through the circulation of texts relevant to shared concerns and practices. Song literati were, or at least aspired to be, well connected. This applies not only to those who frequented the court but also to those who only held occasional minor office or no office what­ soever. The example of Wang Mingqing and the other notebook notebooks considered to be “historical materials,” see Ding Haiyan, “Song ren shiliao biji yanjiu.”



liter ati networks and identities

393

authors examined here suggests that they tapped into a large and diverse network of sources and informants. The analysis of the social backgrounds of the informants whose words were captured in Wang Mingqing’s notes suggests that officeholders continued to have a dominant voice and also that the role of high officeholders declined in favor of lower-ranking peers. Geographically, the literati networks captured in notebooks ranged empire-wide in some cases and cross-regionally at a minimum. Because of their geographical range, literati networks could play a critical role not only in the dissemination of the expanding numbers and genres of literati texts but also in the shaping of literati identities across the extent of the Song Empire. The notebooks of better- and lesser-known twelfth- and ­thirteenth-century authors allow us to read identities in the process of formation. In their preoccupation with the organization and history of the court, the prominent families of the realm, the influence of major political and cultural figures, trivia, the verification of omens, rare texts and objects, and wit, they caution against the presumption that literati collaboration with the imperial court can be best under­stood with reference to overarching ideologies such as Confucianism or, from the twelfth century onward, Neo-­ Confucianism. The scarce appearance of classical texts and pre-Song history in the information networks diagrammed and discussed here questions the extent to which broader generalizations about norm integration through a classical repertoire alone resulted in an elite dedicated to the maintenance of empire. Rather, these texts suggest that even as many of them were settled locally rather than in the capital, supra­local concerns relating to court, the imperial elite, and Song political history remained primary subjects of debate. In Song times notebooks were not only of historical but also, and more important, of social and political relevance as the communicative relationships embedded in them helped form and consolidate identities at both the individual and the group level. The desire of a notebook author such as Liu Changshi writing from an isolated county on the southern periphery to communicate with peers across the realm and his decision to self-fund the publication

394

imperial information networks

of his notes for this purpose (see chapter 6) are a telling illustration of their role in the representation of literati identities empirewide. Notebook authors also pursued political ends, both through the networking that was part and parcel of the constitution of the notebook and the manner in which the texts circulated, and through the development of a position on shared issues of concern. To further pursue the latter line of inquiry I will next turn to a close reading of Wang Mingqing’s comments and conversations relating to the relations of the Song with its foremost northern rival, the Jin Empire.

E igh t Representing the Foreign Other

T

he previous chapters have shown that compilations from the archives of the Song dynasty and documents on its ongoing business were circulating at an increasing rate from the late eleventh century onward. As evidenced in their notebooks, the literati readership of such compilations constructed communication networks that brought them into contact with peers who had lived and worked in locations across the Song Empire. This chapter is an examination, on the basis of Wang Mingqing’s Waving the Duster, of what political imaginaries and loyalties were articulated in notebook literature. Its focus is on the representation of foreign others and on the political values that Wang Mingqing and his interlocutors ascribed to Song Chinese in the face of foreign threat. Song literati showed great interest in compilations on border affairs by the late eleventh century. This interest only increased in the twelfth century, when the Song dynasty became involved in a century-long interstate conflict. The successful Jin occupation of the central plains in 1126, and the series of battles and negotiations between Song and Jin military and civil personnel that ensued over the next century,1 led to a continuous stream of publications touch1. The proceedings of the military and diplomatic maneuvers on the part of the Song and Jin military and diplomatic corps have been the subject of many studies. See, for example, Thiele, “Der Abschluss eines Vertrages”; Franke, “Treaties between Sung and Chin” and “The Chin Dynasty.”

396

imperial information networks

ing upon border affairs. Song relations with its northern neighbors became a shared concern and a central topic of debate. As a record of conversations and interactions with a wide range of texts compiled over the course of three decades, Wang Mingqing’s Waving the Duster allows us to trace in some detail how a twelfth-century literatus responded to contemporary debates on the question of war and negotiated peace. Disparate records of Song-Jin conflict, negotiations, and prowar literature may confirm for some readers of Wang Mingqing’s notebooks the impression that notebooks lack overall narrative or argumentative structure.2 This chapter supports the view that notebook authors also articulated clear positions on the key questions raised in conversations and contemporary literature and did so by bringing together a variety of voices.3

Violence and Civilization in Waving the Duster Reading through the hundreds of entries in Wang Mingqing’s work, it becomes clear very quickly that he made no attempt to standardize references to the Jin state or Jurchen people. Terms ranging from neutral generic references to foreign peoples, the polite terms of address customary in diplomatic language, and blatantly xenophobic rhetoric occur in its pages, probably reflecting to a large extent the words used by Wang Mingqing’s sources. Through a systematic analysis of the language used to refer to the Jin and Jurchen people,4 I will first establish that derogatory terms and acts 2. On the lack of narrative structure in notebooks, see Jack Wei Chen, “Introduction” to Chen and Schaberg, Idle Talk, 4. 3. Peter Bol has written about the intellectual agenda of Zhang Lei’s 張耒 Mingdao zazhi 明道雜志 (Clarifying the Way: A miscellany), and Ron Egan very recently also wrote about the articulation of a coherent political stance in Zhou Hui’s 周煇 Qingbo zazhi 清波雜志 (Miscellaneous notes by one who lives near the Gate of Clear Wave). Bol, “A Literati Miscellany and Sung Intellectual History”; Egan, “Authorial Intent in Song Period biji.” 4. I examined the frequency of individual terms within the entire corpus and also investigated the frequency and the statistical significance of expressions that occur along with these terms (collocates or patterns that occur immediately before and after



representing the foreign other

397

of violence dominated in the descriptions Wang Mingqing recalled and then examine the effect that particular expressions might have had on contemporary readers. Previous scholarship has already highlighted the fact that both Song politicians and their Liao and Jin counterparts recognized the significance and effect of changing the terminology for foreign places and foreigners. The intersection of language and conflict in imperial Chinese history has been central to the small body of work on eleventh- and twelfth-century diplomatic exchanges in East Asia. In the early 1980s, contributors to the edited volume titled China among Equals drew attention to the role of language not so much in interstate conflict but in negotiated settlements of conflict between the Song court and its northern neighbors. Wang Gungwu and Tao Jingshen noted that in their dealings with the Liao and Jin states Song envoys resorted to “the rhetoric of a lesser empire” or a “new diplomatic language of equality”5—this in contrast with the more conventional Chinese assumption of cultural superiority and the expectation of deference on the part of tribute-bearing foreigners. They saw this shift in diplomatic rhetoric reflected in the use of terms of address based on relationships of fictitious kinship between the terms under investigation). The statistical significance of a collocate is measured through a score that indicates how strongly related the two are in the corpus. Such a measure is especially relevant when the term and a collocate occur together at a rate that is unusually high relative to the corpus, suggesting that these terms tend to be clustered together and seldom appear in other combinations. The measure used here is the Z-score, which measures “how unlikely it is that the focus and collocate are unrelated.” I used Xaira, developed by Lou Barnard. The program and documentation can be accessed at http://xaira.sourceforge.net/. Corpus linguistics can further help us address questions such as the relationship between (written and speech) genre conventions and language. In order to measure how significant a role context may play in particular usages I marked the genre of the source text (e.g., edict, memorial, memoir, conversation report, survey history, war announcement) and the type of situation described in the passage and then examined frequencies of usage relative to the type of source text and the situation associated with it. I differentiated between the type of source text quoted by Wang Mingqing directly (see chapter 7) and the type of source text for the specific passage in which the term occurs. Since the former does not capture quotations within quotations, the latter can be more significant here. 5. Tao, “Barbarians or Northerners,” 69; Wang, “The Rhetoric of a Lesser Empire.”

398

imperial information networks

rulers (younger brother/older brother; nephew and uncle—with the Song ruler occasionally occupying the junior position), terms of address that appear to acknowledge the political equality of “the northern court” and “the southern court.” Another indication of the acceptance of a new diplomatic language was the renaming of border jurisdictions with seemingly derogatory names such as Weilu 威虜 (showing military power to the caitiffs), Polu 破虜 (breaking the caitiffs), and Jingrong 靜戎 and Pingrong 平戎 (pacifying the barbarians).6 This initial foray into the historical meaning of shifting terminology for ethnic and political others raises many questions. How did diplomatic texts compare to other types of texts (historical, administrative, literary) on conflict and negotiation? Should we assume that deferential language and patterns (projecting an image of equality) were reserved for negotiations and treaties? Or, conversely, were the classical stereotypes of the enemies of civilization (Poo Mu-chou’s term for the representation of the cultural other in ancient civilizations)7 dominant in elite (and perhaps popular) discourse? Should the revisionist reading of Southern Song political history that suggests that proponents of peaceful negotiation represented, over time, a dominant voice and, at least, the only practical solution imply that diplomatic discourse spilled over into or corresponded with a real­ politik approach to foreign relations among the educated elite?8 To some extent Song perceptions of the foreign other acknowledged a multistate world. Such a conclusion could be drawn on the basis of the terms of respect and neutrality used in diplomatic parlance and of the appearance of gazetteers and other texts that analyzed the states to the north (Liao, Jin, and Gaoli) using frameworks typically applied to the Song territories (see chapter 4). Similarly, the dynastic name was frequently used in references to the most powerful state to the north of the Southern Song territo6. Tao, “Barbarians or Northerners,” 69–70; XZZTJ, 58.1301. 7. Poo, Enemies of Civilization. 8. Rossabi, China among Equals, introduction. A paper by Wang Ruilai that similarly explained collaboration as a practical strategy caused heated debate at the 2008 International Conference on Southern Song History in Hangzhou. Wang Ruilai, “Yu sui haishi wa quan?”



representing the foreign other

399

Table 8.1 Frequency of terms used in descriptions of the Jurchen people and the Jin state in Wang Mingqing’s Waving the Duster

a Term Frequency 

Jin (金) (221) Jinren 金人 30 Da Jin 大金 22 Jinguo 金國 5 lu 虜 153 waiguo 外國 1 beiguo 北國 0 beichao 北朝 2 beiren 北人 3 Nüzhen 女真 1 Qidan 契丹 13 yi 夷 34 di 狄 12 man 蠻 4 rong 戎 21 chou 醜 4 dian 酋 10 diren 敵人 2 Note: The term 金 (Jin) has a high rate of occurrence, to a significant extent due to its multiple meanings in this text. a Numerical tallies from Huizhu lu (Zhonghua shuju edition).

ries (table 8.1). On the whole, Wang Mingqing and his sources tended to use terminology that appears at first sight to be politically neutral: Jinren 金人 (people of Jin), Da Jin 大金 (the Great Jin), and Jinguo 金國 (the state of Jin).9 Such terms are similar to those used to refer to the Song polity and its habitants, even though terms of self-reference were far less common (table 8.2). Ethnonyms in use at the time to refer to the Jurchen people who had established this state (such as Nüzhen 女真, or Ruzhen 如真/汝真) seldom appear in Waving the Duster. There is only one occurrence of 女真, and the reference to it points to an edict dating to the late Northern Song 9. The total for “Jin 金” (224) includes uses of this term other than as a reference to the Jin dynasty, such as “gold,” or its use in other combinations such as place-names.

400

imperial information networks

Table 8.2 Frequency of terms used in descriptions of the Song people and the Song state in Wang Mingqing’s Waving the Duster a Term Frequency 

tianxia 天下 142 Song (宋) (87) Da Song 大宋 7 Songren 宋人 1 Songmin 宋民 1 wo Song 我宋 5 wo chao 我朝 1 wo renmin 我人民 1 chaoting 朝廷 88 guochao 國朝 20 nanchao 南朝 3 nanguo 南國 0 nanren 南人 b 0 zhongguo 中國 13 zhongyuan 中原 12 Note: The term 宋 (Song) has a high rate of occurrence, to a significant extent due to its multiple meanings in this text. a Numerical tallies from Huizhu lu (Zhonghua shuju edition). b There are two occurrences of Jiangnan ren 江南人, referring to the native origins of particular people: “he was someone from south of the Yangzi River.”

period, when the contours of the Jin state may not yet have seemed relevant to the Song court.10 The Jin state could, moreover, not be defined primarily along ethnic lines; a variety of ethnicities, including former Song subjects of Han Chinese origin, were part of the Jin military and the Jin state. The widespread use of the term “people of Jin” (even in unsympathetic portrayals of their actions) suggests that people along the northern edges of the Song Empire had become associated with the courts that claimed to rule them rather than with tribal groups and confederations or their leaders. However, the prevalence of politically neutral designations for the Jin state and its people should not be read as an indication that this state and its people had become an equal partner in the eyes of 10. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.182.



representing the foreign other

401

Song politicians and literati. Indeed, the most frequently occurring term to refer to Jurchen and Jin individuals (including heads of state), groups of individuals, or the Jin military or court is lu 虜 or “caitiff, slave” (table 8.1), a pejorative term that classifies them as inferior beings devoid of civilization and autonomy. The people of Jin were mostly portrayed in the four installments of Wang’s notebook as invaders and the violent occupiers of Song territory. They are cast first and foremost as military actors; we do not see or hear them as political actors.11 The emphasis on the military aspect of the Jin state and its subjects is most clearly illustrated in the verbs associated with Jin individuals and collectivities. Jinren 金人 occurs most frequently with verbs indicating the illegitimate violation and appropriation of territory and goods: terms such as fan 犯 (invading), qinfan 侵犯 (invading), ju 據 (occupying), kou 寇 (robbing),12 lue 掠 (robbing), shao 燒 (burning), po 破 (breaking), or jin qu 盡取 (taking everything). Similarly, lu 虜 is often associated with the military: qi 騎 (cavalry), bing 兵 (soldiers), and jun 軍 (army) rank among the terms most frequently paired with it. Similar to Jinren 金人, with which it is used interchangeably even within the same entry, the more generic and frequently used term luren 虜人 (caitiff people) tends to occur with verbs indicating occupation and violation ( fan 犯 [invading], qinfan 侵犯 [invading], and kou 寇 [robbing]) and with verbs indicating their presence in Song territory ( yi ru 已入 [had already entered], yi zai 已在 [were already at], and du 渡 [crossed over]) (tables 8.3 and 8.4).13 The people of Jin were cast as the instigators and perpetrators of conflict not only through the description of their military advances into territory claimed by Song but also by their portrayal as inveterate violators of agreements. Passages discussing an agreement or treaty (meng 盟 or yue 約) depict Song and Jin actors in adver11. For a discussion of a similar emphasis on the military features of noncivilized peoples in Roman texts, see Wells, The Barbarians Speak, chap. 5, esp. 100. 12. Note the high Z-score for kou 寇 with Jin (金), suggesting the close association of this term with the subject Jin. 13. In these instances luren 虜人 refers to the Jin people, since the passages in which luren appears concern events occurring in the last years of the Northern Song and the early years of the Southern Song period.

402

imperial information networks

Table 8.3 Frequency of occurrence and Z-score for the top-ranked collocates associated with lu (虜) in Wang Mingqing’s Waving the Duster Frequency Score 37 24.9 12 42 10 11.8 8 5.1 6 11.5 5 6.9 4 5.9

Collocate ren 人 qi 騎 shi 使 zhong 中 zhu 主 bing 兵 jun 軍

sarial roles: Jin people are most frequently cast in the act of breaking agreements (bei meng 背盟),14 whereas Song people are called upon to form a united front (tongmeng 同盟).15 There are, however, significant differences in assigning responsibility for the breakdown of agreements. In memoirs and narratives of military events written by Song authors, the blame is readily assigned to Jin forces. In other texts destined for a domestic audience Song officials and commoners are criticized for seeking a peace agreement (in Hu Quan’s 胡銓 [1102–80] fiery 1138 memorial excoriating the Jin and those willing to settle for peace with them, discussed below) or encouraged to join together in an alliance to fight the Jin (in Wang Zhi’s 王銍 [?–c. 1154] draft war announcement, discussed later). In petitions submitted to or drafted for the Jin military command in the last years of the Northern Song, Song literati and statesmen proved more willing to assign some of the blame to the wrong-headed advice of previous Song court officials and to ambiguous referents such as “previous rulers.”16 14. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.213, 2.220. 15. Ibid., 3.246, 4.281. 16. This was in keeping with the public acknowledgment of mistakes performed by the emperor himself. Franke, “Treaties between Sung and Chin,” 74. The Jin emperor used violations of previous treaties as a justification of Jin invasions. See, for example, Levine, “The Reigns of Hui-tsung and Ch’in-tsung,” 634. Elsewhere, Wang Mingqing’s sources showed an understanding for Jin officers’ decision to punish Song breaches of agreements: Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.127.



representing the foreign other

403

Table 8.4 Frequency of occurrence and Z-score for the top-ranked collocates associated with Jin (金) in Wang Mingqing’s Waving the Duster Frequency Score 30 16.2 22 16.9 5 4.4 4 14.5

Collocate ren 人 da 大 guo 國 kou 寇

Wang Mingqing not only commemorated conflict but also diffused language that mobilized against the northern invaders. The wanton violence associated with foreigners was the driving force behind the mobilization of the civilized realm. It also provided space for the articulation of an active role for the literati in the conflict between Song and Jin. Wang Mingqing quoted at length a text on military institutions drafted by his father. It read, “Our founding fathers, with sagelike civil and military government, ruled the universe and flogged and lashed [castigated] the non-Chinese [siyi 四夷].”17 Wang Zhi drafted this announcement for the supreme area commander of the Western Circuit (Xidao du zongguan 西道都總管), Wang Xiang 王襄, who had been charged with mobilizing troops in the last hours of the Northern Song court as Jin cavalry reached the capital of Kaifeng.18 Wang Mingqing counterbalanced memories of Jin invasions and associations between the people of Jin and the threat of war with the written testimony of Song calls for war, threats to Jin, and preparations for war under the first emperor of the restored Song court, Emperor Gaozong. The opening line of a war announcement written by Wang Zhi set the tone for a last-minute attempt to incite the population to rise up against the invading Jin troops: “Violating agreements and succumbing to our sovereignty, [alternating between these] is the

17. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 4.286. 18. On this assignment see SS, 23.431.

404

imperial information networks

inherent nature of uncivilized peoples [ yidi 夷狄].”19 According to Wang Zhi, a prominent historian, this finding was proven by the relations between Chinese and non-Chinese peoples documented since the beginning of recorded history. This general truth was to serve as a guarantee that the sacrifices that Song subjects were asked to make would not be for naught. In the face of all manner of brutalities visited upon official and commoner, young and old alike, this announcement projected that commoners and office­ holders would naturally be provoked to rise up against the nonChinese aggressors: People can tell the difference between obedience and disobedience, and from the four corners the sound of their coming forward can be heard; the literati know the difference between gratitude and revenge, and from all over the sound of their mobilization reverberates. In order to realize our long-term goal we have to break through this enduring encirclement. We have to speedily mobilize our tiger [fierce] troops so as to completely sweep away the dog and sheep [subservient] masses. The bands of howling tribesmen will then be exterminated and the types who went along with them must then be annihilated.20

The announcement called for a united front in which all men “embody the state with utmost sincerity, and show their devotion to their ruler with deep integrity so that their loyalty will be carried by the spirits and their might will be familiar to the uncivilized tribes [yilu 夷虜].” The use of animal metaphors has historically been a common linguistic technique in discriminatory ideologies. Within ideologies that differentiate between higher and lower types of beings, assigning particular groups to lower-ranked life forms has the effect of transferring value judgments associated with the referent.21 By blending metaphors in discourse, as in the lines quoted above where “dogs and sheep” can stand in for “howling tribesmen” or Jin people, 19. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 3.245. 20. Ibid., 3.246. Italics are mine. 21. Musolff, “What Can Critical Metaphor Analysis Add?” 2.



representing the foreign other

405

the properties and actions associated with animals like dogs and sheep are conceptually linked with Jin.22 Jin people can then be imagined as a subservient species to be driven off by their masters. Similarly, “slave” peoples are to be “flogged and lashed.” Centuries after the Song-Jin conflict had come to an end, the power of these metaphors and associated derogatory language was still recognized. The compilers of the Siku quanshu series (SKQS) edited all terminology that could give offense to the Manchus (the alleged descendants of the Jurchens) and incite tension between Han and non-Han subjects. They included Waving the Duster in the collection, but a reader relying on the Siku edition would come away with a rather different interpretation of the representation of SongJin conflict in Wang Mingqing’s work. As shown in table 8.5, the editors removed pejorative designations such as “slaves,” which were dominant in extant pre-Qing editions, and changed derogatory uses of archaic terms for foreigners. The politically more neutral terms were either fully preserved in the Siku edition, or increased in usage because the editors considered them more suitable substitutes for more compromising originals. Terms such as “enemy” (diren 敵人, di 敵, chou 仇) replaced more offensive terminology. Table 8.5 also testifies to the continued use of archaic terminology that, as Wang Zhi underscored, went back to the beginning of recorded history. This claim masks the different context from which this terminology derived its meaning in Song discourse on nonChinese states and peoples. Wang Zhi’s war announcement and another text he wrote on military institutions were unusual in their heavy and exclusive use of the more generic and archaic categories for non-Chinese peoples, most notably the set of four terms ( yi 夷, di 狄, man 蠻, rong 戎) that were, in accordance with their definition in The Record of Ritual (Liji 禮記), identified with the four cardinal directions (the eastern Yi, the southern Man, the western Rong, and the northern Di), but which, in varying combinations with each other and with numbers referring to an entire set (such 22. Ibid., 5. There is also a key difference in the use of metaphors between twelfthcentury extreme xenophobic rhetoric and Nazi ideology: Song authors do not imagine the possibility of a world without non-Chinese.

406

imperial information networks

Table 8.5 Frequency of terms associated with the Jin state and people in the Siku quanshu (SKQS) edition of Wang Mingqing’s Waving the Duster and in the collated Zhonghua shuju edition Term Jinren 金人 Da Jin 大金 Jinguo 金國 lu 虜 Nüzhen 女真 Qidan 契丹 yi 夷 di 狄 man 蠻 rong 戎 chou 醜 meng 盟 dian 酋 diren 敵人 a The

Frequency (Siku quanshu edition)

Frequency (Zhonghua shuju edition) a

47 22 5 2 0 14 18 (3) 2 4 16 0 12 1 49

Change

30 increase in SKQS 22 no 5 no 157 deletions in SKQS 1 minor 13 minor 34 deletions in SKQS 12 deletions in SKQS 4 no 21 minor 4 deletions in SKQS 16 minor 10 deletions in SKQS 2 increase in SKQS

collated Zhonghua shuju edition of Waving the Duster is based on a Song edition.

as four or nine), also referred to non-Chinese peoples in general.23 Zhenzong’s 真宗 (r. 998–1022) recognition, referred to above, that the use of some of these terms in the naming of Song border garrisons proved offensive to Liao was an early indication of the fact that these terms could carry pejorative overtones. The distribution of these terms in the various genres excerpted in Wang’s notebook illustrates that they were the preferred terminology in those types of texts aiming to arouse animosity against Jin (including the passages from Wang Zhi’s war announcement, history of military institutions commissioned by Gaozong in 1130, and Hu Quan’s 1138 critique of the pursuit of peace with Jin) (table 8.6). Song authors also used archaic terms and general categories for non-Chinese peoples in general reference without overt verbal 23. Poo, Enemies of Civilization, 46.



representing the foreign other

407

Table 8.6 Frequency of occurrence of the terms yi 夷, di 狄, man 蠻, and rong 戎 in Wang Mingqing’s Waving the Duster, by text type Source text type

Total occurrence of terms

Number of entries

War announcement 6 1 Military institutions survey 7 1 Pro-war memorial 3 1 Subtotal 16 3 Other genres a 20 15 Note: The last column shows the number of notebook entries in which these terms occur. a “Other genres” includes memorials, envoy reports, poems, memoirs, anecdotes, conversations, and notebook entry copy.

aggression, as in the prefatory comments by a Song envoy that he left a record of his mission to Qara-hoja (Gaochang 高昌) in order to serve as a guide to establish connections with “all non-Chinese” (literally, the nine Yi and the eight Man), or in stable standard expressions such as “establishing harmonious relations with the nonChinese” (he rong 和戎) or “border defense” ( yurong beibian 禦戎 備邊).24 However, the sensitivity to the use of these terms, shared by Liao, Jin, and by nineteenth-century British diplomats,25 may derive from their association with text types such as war announcements and dissenting memorials. The sheets that carried such texts may have been printed for broad distribution; victory announcements were disseminated in large numbers, and news about Hu Quan’s letter to Emperor Gaozong (discussed later) allegedly caused a stir in the capital upon its submission.26 These texts gave offense not only by what they said but also by their illocutionary force and their perlocutionary effects: the imperative to rise up against the invader and the mobilization of troops led by civil and military elites loyal to the dynastic and imperial ideal. The texts and conversations collected in Waving the Duster suggest that the concept of Song hegemony over the known world was 24. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.39, 2.165, 2.183, 2.203, 4.281. 25. Lydia He Liu, Clash of Empires. 26. Bi Yuan, Xu zizhi tongjian (Zhongguo jiben guji ku digital ed.), 121.10a.

408

imperial information networks

unassailable. Although the reality of a multistate world resulted in a worldview in which the formation of bureaucratic states in the north with subjects of their own was acknowledged, the imperial ideal of the ruler of the central state as the sovereign of Chinese and non-Chinese peoples (tianxia zhi ren 天下之人 [the people of the realm], Yi-Xia 夷夏) alike remained unchallenged. In his war announcement, Wang Zhi reaffirmed that the emperor possessed parental feelings for all: “The emperor watches over all and treats all life as his own children. Always concerned, his heart goes out to both Chinese and non-Chinese. Constantly worried, he continues the legacy of the founding fathers of our dynasty.”27

Negotiation and the Rhetoric of Empire Despite misgivings about the reliability of negotiations and treaties, Wang Mingqing commemorated successive instances in which Song emperors, envoys, court officials, and self-identified degree holders approached and engaged with Liao and Jin rulers, civil officials, and military commanders. The texts selected in Waving the Duster provide a window into what was negotiable for the Song court from Wang Mingqing’s perspective; their hidden transcripts further reveal what was not. In select anecdotes Wang Mingqing underscored the willingness of Song rulers to satisfy the material demands of the military and political elites of the north. In a passage about the rule of Emperor Renzong 仁宗 (r. 1023–63) he quoted from the personal memoirs of Li Zunxu 李遵勗 (988–1038) (Li Hewen yishi 李和文遺事, Memoirs of Li Hewen): Renzong once wore a beautiful jade belt. Those assisting him all stared at it. When he returned to the palace, the emperor said to his courtiers: “Why are you all staring at my belt without end?” They responded: “We have never seen something this extraordinary.” The emperor said: “I should leave it to the head of the northern barbarians 27. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 3.245.



representing the foreign other

409

[luzhu 虜主].” Those around him said: “This is one of the most precious items in the empire; it would be a shame to give it to the foreigners [waiyi 外夷].” The emperor responded: “In the central kingdom [Zhongguo 中國] we regard the welfare of the people as a treasure. How could this be worth cherishing?” All the officials cheered, “Long live the emperor!”28

This exchange was reminiscent of and may have been modeled upon the behavior attributed to the wise rulers of antiquity, who similarly held personal material wealth and comfort in low regard. It may also have invoked the image of clever heads of state who used the gift of material wealth as a tool to pacify competing states and even to bring about their moral and political decline. The lengths to which Qinzong 欽宗 (r. 1126–27) was prepared to go to stave off the collapse of the Northern Song court were commemorated in much starker language: At this time an imperial edict has been clearly promulgated demanding that a full search for gold and silver be conducted in order to repay the kindness by which the Great Jin let the people live. It states that we should fully exert ourselves and not be constrained by human attachments. Whenever something can be used to repay the Great Jin, even if [as personal as] hair and skin, we should not hold onto it. As long as there is something, it should be fully obtained. Our officials have repeatedly issued warnings. As for those things that have been designated forbidden, even if the collections are small, they should be obtained on the basis of martial law. Thus thorough measures for a full search have been taken and it is not that we are not exerting ourselves. Ranging from the ritual vessels in the ancestral temple to the headdress of insignificant folks, we are fully exhausting all possessions and intend to offer them for reuse. The emperor has moreover said, “If there is anything with which I can repay the Great Jin, I will not hold on to it even if it were hair and skin [of a most personal nature.]”29 28. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 1.6. 29. Ibid., 2.126–27; the emperor’s words are phrased slightly differently in Duan Guangyuan’s letter, ibid., 2.129. For a fuller list of the concessions made, see Levine, “The Reigns of Hui-tsung and Ch’in-tsung,” 642.

410

imperial information networks

Renzong had been portrayed in an act of personal disregard for luxury, an act that simultaneously showed his low regard of the foreigners who were deemed to be primarily motivated by material desire. In a letter submitted by Duan Guangyuan 段光逺 during the occupation of Kaifeng in 1126–27, from which the above quote was taken, Emperor Qinzong urged the entire population to give up all valuables including the most personal and sacred of items such as the ritual vessels in ancestral shrines “in order to repay the Great Jin.” This letter was addressed to a Jin marshal (Yuan­shuai 元帥) after Jin forces had attacked and then withdrawn from the capital. Their withdrawal was dependent on the handover of all things of value in the capital; Duan Guangyuan and two others who wrote similar letters at the same time also knew that the Song emperor was being held hostage. Huang Shicheng 黃時偁, Xu Kui 徐揆, and Duan Guangyuan wrote as private persons. Huang and Duan identified themselves as holders of the jinshi degree, whereas Xu Kui was a student at the Imperial College.30 They clarified that they did not hold office at the time of writing. They petitioned to reassure the commander that Jin demands were being implemented and asked for considerate treatment of those residents who had been accused of withholding possessions. They also requested the return of their emperor, appealing to the moral conscience of the addressee as well as the history of neighborly relations between Song and Jin. In the words of Xu Kui: You, Commander, embody the Great Jin emperor’s virtue of loving life. You are often concerned about the hardships of the common people. When your great army trekked across great distances and headed straight for the central plains, they never made killing their 30. Huang’s title is Imperial College student in Xu Mengxin, Sanchao beimeng hui­ bian, 78.588; in Li Xinchuan’s Jianyan yilai xinian yaolu (2.40) he is also referred to as a jinshi degree holder. Xu Kui was accorded a biographical entry in the Song dynastic history under the category of “the loyal and righteous” (zhongyi 忠義). His biographer there mentions that Xu Kui led a group of students in protest of Qinzong’s arrest. His protest led to his violent death at the hands of Jin soldiers. SS, 447.13179–80.



representing the foreign other

411

business. This is how they excelled at caring for the people. All things considered, you, Commander, possess the virtue to preserve the imperial altars and the humaneness to preserve human life. However, to take our lord [jun 君] hostage for the sake of gold and silver, this is like caring for the children but embarrassing their elders. How does it differ from not caring for them?31

And in the words of Duan Guangyuan: I have come across the following passage in Zuo’s Commentary on the Spring and Autumn Annals: “To be close to the benevolent and to be on good terms with one’s neighbors, this is a boon for a state.”32 I have also come across another passage in the “The Meaning of Missions” in The Record of Ritual: “If one regards wealth lightly and treasures ritual, then the people will be submissive.”33 When I read these words I always have to ponder over them. I put the book down and sigh in admiration, “Who could accomplish such a thing except for a great sage?” After having considered it, I realize that in the past Emperor Taizu 太祖 [r. 960–76] received Heaven’s clear mandate, and acceded to the throne by being offered it. Holding the seals of emperorship, he became the lord of all. His accomplishments were held up, and brilliant government continued for over two hundred years, infiltrating the east, covering the west, spreading across the south, and flourishing in the north. All across were spread our prefectures and counties. Unfamiliar corners and far-removed lands became our neighbors. We visited and communicated with them; spreading a network of roads. As for those for whom integrity and attention for ritual mattered gravely and whose kindness was deep and virtue abounding, in comparison with other states, only the emperor of the Great Jin could match up to this.34

31. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.69. The italics are mine. 32. Zuo zhuan, Yin’gong 6. Cf. Legge, The Ch’un Ts’ew, 20–21. 33. This passage appears to be a slight rewording from current editions of Liji. E.g., Liji zhushu 禮記注疏, in Chongkan Songben shisanjing zhushu fu jiaokanji 重刋宋本十 三經注疏附校勘記, 聘義, 48.1029. Cf. Legge, The Li Ki, book XLV, Phing Î, no. 8. 34. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.128–29.

412

imperial information networks

Praise for the kindness and discipline of Jin commanders and soldiers, for the integrity of the Jin emperor, and for the Jin state’s respect for the ritual of friendly missions in these letters sharply contrasts with the recurring themes of Jin aggression and their violation of treaty agreements elsewhere in Wang Mingqing’s recollections. The authors further portray the invasion as a legitimate response to the corruption of Song court officials ruling in recent times. To Wang Mingqing, this language did not appear obsequious. He accorded these letters high acclaim and saw them as exemplary embodiments of loyalty to the Song, since the praise for Jin here symbolized an act of protest and a demand that the Jin leadership live up to the behaviors and standards already ascribed to them in the letters. Wang noted that he had selected these letters, which he reportedly found decades later in the private collection of an unidentified relative or acquaintance, from among a plethora of memoirs and diaries compiled by literati in the last years of the Kaifeng court for a particular reason: “There were men who set aside difficulty and conflict and put their lives on the line like this. I admire their loyalty and nobility and fear that with time passing this may no longer be transmitted.”35 Remarkably, in the third installment of Waving the Duster Wang Mingqing accorded similar praise to two opinion pieces that Qin Gui 秦檜 (1090–1155) submitted around the same time, in 1126. Wang remarked in words that bear some resemblance to other expressions of appreciation of the early career of tyrannical statesmen: “The intent of the words [in these memorials] is deeply loyal; the wording is also very unusual. If this is truly by Huizhi [Qin Gui], if he had not returned to the councilorship in the Shaoxing [紹興, 1131–62] period to rule the state single-handedly, to specialize in killing and intimidation, then would it be fair not to say that he had been a wise man?”36 Qin Gui earned this rare if subdued form of praise because of two proposals in which he set out his solution to the crisis the court faced when the capital could no longer be defended against invading 35. Ibid., 2.125, 2.130. 36. Ibid., 3.245.



representing the foreign other

413

Jin troops. Qin Gui’s proposals read as draft letters for the Jin leadership in which the Song court humbly requests a return to friendly relations. Qin proposed that the court persuade the Jin military leadership that the best way forward was to ensure the common interests of both Song and Jin. He reminded his intended readership of the bond between Song and Jin forged in their common defeat of Liao. He further noted that the loyalties of both Jin officers and the Song population would prove to be insurmountable obstacles for the Jin should they try to further extend their control over Song: As for Song rule over the entire realm, for nine reigns now virtue has prevailed. They can measure up to the Han and Tang dynasties and are substantially different from the two Jin 晉 dynasties. When we examine carefully those making plans [for the Jin] today, they are for the most part men who have previously served the no longer extant state of the Great Liao. The reason why they make plans and establish policies to annihilate Song is not because they are loyal to the Great Jin; they use the annihilation of the Great Song only as a means to take revenge. How could they not recognize that the Great Liao were annihilated through the joint effort of the Great Jin and the Great Song! Should the Great Song be annihilated, will the Great Jin be able to control them?37 Now if the Great Jin really could annihilate Song, the [Song] imperial magnanimity cherished [in the land] between Hebei and Hedong will not be forgotten. If the Song can in the end not be annihilated, this [the attempt to annihilate it] will make the imperial clansmen and elites of great virtue mobilize the realm and exert the state’s energies to move northward. Then even those among the people between Hebei and Hedong who may in the past have been brought under submission will also abandon the Great Jin and return to Song. Now there naturally are the states of the north and of the south, and their territory is very different. The [Later] Jin were annihilated by the Khitans, but Emperor Shizong [世宗, r. 955–59] of the Zhou [周] dynasty [951–60] reestablished the Three Passes;38 this was to repay the wrath of the Jin throne. And so if today the Zhao 37. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 3.242. 38. For the military successes of the later Zhou in the north, see Standen, “The Five Dynasties,” 129.

414

imperial information networks

[趙] imperial clan were to be annihilated, it would not have to fall upon members of the Zhao clan to take revenge. Heroic men of the central plains are also bound to take revenge for the central kingdom. I, Gui, exert myself and risk my life to explain to you, Commander, the significance of abolishing the state in order to clarify the harm and benefit for the two states.39

Qin Gui’s proposals, like the letters submitted to the Jin military leadership, were letters of request. In this text type the benefits of treaties with a non-Chinese state and errors on the side of the Song leadership (emperor included) and its military could be admitted. Such letters could be read as acts of loyalty on a par with those of men who refused to negotiate and resisted militarily in the field. Even though terms of respect were here substituted for their more derogatory equivalents,40 these letters were perceived to serve the same goal through different means: the continuity of Song rule and the territorial integrity of the Song Empire. The latter goal was less emphatically stated in the letters of request but formed the core of their hidden transcript. Duan Guang­ yuan’s letter reminded the Jin commander of the vast Song administrative network and the Jin state’s proper position as a neighbor on the periphery of what had supposedly been the Zhou Empire of antiquity. Qin Gui, who later pushed through an agreement to cede the northern half of the empire, and whose reputation became unrescuable afterward, began his 1126 address with a reminder of the grand dimensions of the Song Empire: “Ever since Song established unity in the central kingdom, its territory continues on for over several tens of thousands li, . . . [and] its children have proliferated and fill all the land between the oceans.”41 Reminders of the continuity of empire in Chinese history underscored what was not negotiable in dealings of any sort with the Jin state. Throughout Wang 39. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 3.243. 40. Twenty-one out of twenty-two occurrences of the polite form of address Da Jin 大金 occur in these two entries. 41. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 3.242.



representing the foreign other

415

Mingqing’s collected notes on the written and oral legacy of the events of 1126–27, Song voices determined the terms of engagement and negotiation. Wealth of any kind could be given up and polite letters of requests could be submitted, but Song emperorship and the territorial integrity of the central kingdom could not be surrendered. Matters had taken a different turn by the time successive installments of Waving the Duster were compiled and printed, but the words of those who practiced the rhetoric of empire at the most difficult of times remained worth commemorating after the Song court had moved south.

The Rejection of Negotiated Peace The changed circumstances of the Song court after the occupation of the north resulted in the polarization of prowar and propeace opinion. Upon his return from captivity in the north and his rise to high office at Gaozong’s court in 1138, Qin Gui became a front player in negotiations with Jin, resulting in the conclusion of a treaty in 1142. Qin Gui’s brokerage of negotiated peace at the cost not only of the more conventional payments of silver and silk but also of all territory formerly held by Song north of the Huai River in the east and all territory north from the southern borders of Tangzhou 唐州 and Dengzhou 鄧州 in the west provoked sharp criticism. This criticism could not be silenced even by Gaozong’s public recognition of Qin Gui’s contributions to Song-Jin relations at the time of the latter’s death in 1155. The acceptance of a treaty that ignored the hidden transcript of earlier attempts at negotiation led to vitriolic rejections of negotiated peace and high-pitched antiforeign rhetoric among literati. In the second installment of Waving the Duster, Wang Mingqing copied a memorial submitted in 1138 by Hu Quan, then junior compiler in the Bureau of Military Affairs (Shumiyuan bianxiuguan 樞密院編修官), in which the latter indicted Wang Lun 王倫 (1084–1144), who had been sent as an envoy to the Jin court, and Qin Gui, whom he saw as the engineer of the scheme to have Gaozong

416

imperial information networks

assent to Jin demands.42 Hu Quan denounced those in favor of accepting Jin demands in the conventional rhetoric of court politics as well as in extreme xenophobic language. Many at the imperial court used terms such as “caitiffs” (lu) and “foreigners” ( yi, di, man, rong) derogatively. Hu Quan’s language can, however, be situated near the far end of the scale of antiforeign rhetoric. The repetition and intensity of metaphors that classified Jin people as less than human matched the more radical solution to the Jin occupation that Hu Quan was aiming for. Hu Quan repeatedly referred to the Jurchen as “ugly slaves” (choulu 醜虜), a term that, within the scope of Waving the Duster, occurs only in his memorial. Chou 醜 was highly correlated to the term lu 虜 in the corpus (Z-score of 38); it may therefore not have been commonly used in written discourse, and its exclusive use in this combination therefore would have stood out and rendered this reference to Jin people memorable. The only other occurrence of chou 醜 in the notebook provides an indication of the kind of conceptual link that may have been established in the minds of the memorial’s audience. In an excerpt of a story about the unwavering loyalism of Zhao Li 趙立 (1094–1130), his biographer claimed that “should Li have lived, he would have exterminated the ugly sort.”43 Terms like “dog-barbarians” (quan-rong 犬戎) were similarly an indicator of extreme antiforeign rhetoric as it too occurs rarely in prose but emphatically in Hu Quan’s memorial. The context of the only other instance in which it occurs establishes, as in the case of 醜虜, a link with extermination as the only legitimate course of action.44 Wang Yande 王延德 (936–99), while on a mission in Gaochang, learned that a Liao emissary had tried to raise suspicions with the king of Gaochang that the Chinese (Han) were trying to obtain intelligence on this mission. This provoked the following outburst on Wang’s part: “The dog-barbarians inherently cannot 42. For a brief discussion of the immediate context of this letter, see Tao, “The Move to the South,” 679–80. 43. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 2.200. 44. There is one other instance in which this term occurs in a grandiloquent banquet poem under Emperor Huizong. Ibid., 2.121.



representing the foreign other

417

obey the central kingdom. This [emissary of Liao] here is an infiltrator; I want to kill him.”45 The object of Hu Quan’s rhetoric was not limited to the Jin but included those whom he perceived to be their internal collaborators. The extreme rhetoric was intended to persuade the emperor of the illegitimacy of any agreement granting Jin demands that challenged Song sovereignty over its former territories. Hu Quan’s memorial further underscored the nonnegotiable nature of Song sovereignty by seeking the death sentence of those at court who were pursuing such an agreement. Fears over the loss of sovereignty and territory were raised by the arrival at court of a Jin emissary whose presence resulted in the request that the Song refer to itself as “Jiangnan” 江南 (South of the Yangzi River). According to Hu Quan this condition was a clear indication that the Jin intended to “turn us into its servants and concubines” and “turn us into another Liu Yu.”46 (Liu Yu 劉豫 [1073–1146] had been established as a puppet for Jin in the north in 1130.)47 Arguing against the advice of Wang Lun, who advocated acceptance of these terms in order to bring about the return of the coffin of the former emperor, Huizong 徽宗 (r. 1101–25), who had died in captivity, and the eventual recapture of the central plains, Hu Quan wrote: A three-foot-tall boy who is most ignorant would turn enraged should he be asked to pay respect to the dogs and pigs. Now those at our grand court one after the other pay their respect to the dogs and pigs. How is it possible that you, Emperor, could bear to do what a small boy would be ashamed to do! . . . Even if it were the case that the Jin could truly be appeased, and all is as [Wang] Lun claims, for what sort of a ruler will later generations in the realm hold you, Emperor? Moreover, the ugly barbarians come up with new schemes all the time, and Lun further assists them through treachery. It is thus clear that the [imperial] coffin will not be returned, the empress will 45. Ibid., 1.39. 46. Ibid., 2.206. 47. For a brief overview of Liu Yu’s Qi state (1130–37), see Tao, “The Move to the South,” 657–59, 674–77.

418

imperial information networks

not come back, the former emperor Qinzong will not come home, and the central plains will not be obtained. As soon as this knee is bent, it will no longer be extended.48

Hu Quan stressed that his was not the position of an embittered politician and that he represented the voice of a broadly shared anger: “The people of the realm grind their teeth and spit [at Lun]”; “This [not taking revenge for the loss of the state] is of concern to officials throughout the realm”; “The one thing that is on everybody’s tongue, from the bureaucrats to the army and the common people, is the wish to eat [Wang] Lun’s flesh”; “Those in the know all say that [due to Qin Gui’s control over attendant officers in the Censorate and Department of State Affairs (Tai Sheng shichen 臺省 侍臣)] there are no [real] people at court.”49 The solution Hu proposed was, as already suggested in his representation of popular feeling toward Wang Lun, the execution of those colluding with the Jin, the detention of the Jin emissary, and the mobilization of Song troops: “I humbly wish that the heads of these three men [also including Sun Jin 孫近 ( j. 1103), accused of being Qin Gui’s second-hand man] be severed and that they be hung from a pole on Gao Street [publicly displayed as a warning to the foreigners]. I further wish that the emissary be retained, charged with noncompliance with ritual, and that a punitive expedition be organized in due course.”50 Hu Quan concluded with a statement of his personal and absolute commitment to the cause of empire: “If it cannot be so, I would rather go to the eastern seas and drown myself. How would it be possible to seek life in the position of a minor court [xiao chaoting 小朝廷]!”51 Perhaps a rejection of the terms of address that Song diplomats and politicians were growing accustomed to (the use of Great Jin 大金 became mandatory for all Song offices as of late 1141;52 “country town” biyi 敝邑, or 48. Ibid., 2.207. 49. Ibid., 2.207–8. 50. Ibid., 2.208. Gao street was a street in Han Chang’an frequented by foreigners. 51. Ibid., 2.208. 52. Franke, “Treaties between Sung and Chin,” 81.



representing the foreign other

419

“insignificant state,” was adopted in official exchanges with Jin in 1141),53 Hu’s “minor court” became in an ironic twist of history an epithet for the Song court in modern historiography.

From Loyalty to Loyalism Following the submission of his memorial, Hu Quan faced the wrath of Qin Gui, but he eventually made a comeback and was later given the honorary name “loyal and honest” (Zhongjian 忠簡). Loyalty was a shared value and its meaning was therefore contested. Qin Gui himself had also been given the posthumous title of “loyal and wise” (Zhongxian 忠獻). Within a political culture that put a premium on loyalty, Wang Mingqing set out the criteria for its appropriate use. He did so by contrasting historical cases. He also collected exemplars, repeating in their life stories a script of loyalist behavior and advocating for the public recognition thereof. It will also become clear that discussions about the meaning and exemplars of loyalty were not merely philosophical but were motivated by one’s stance on the issue of Song-Jin relations. The meaning of loyalty had already undergone a major shift in the first century of Song rule. Switching allegiance between courts and serving multiple rulers was usual practice until the mid-tenth century. The emergence of two major powers (Liao and Song) in the late tenth century resulted in a hardening of attitudes against those who valued personal loyalties over loyalty to one court or one emperor. This new model of loyalty was justified not only in philosophical discourse but also in critical histories of tenth-century courts and biographies of political and military elites.54 We could interpret Hu Quan’s assertion that suicide was preferable to the acceptance of Jurchen demands as an example of the impact of the eleventh-century redefinition of loyalty. His letter suggests, however, that loyalty was determined by other overarching values. Loyalty was due to an emperor and a 53. Ibid., 79. 54. Wang, “Feng Tao”; Standen, Unbounded Loyalty, 59–61.

420

imperial information networks

court that defended Song sovereignty over the realm and all its territories. The Song emperor therefore had to insist at all cost on governing “the realm” (tianxia) rather than “the area south of the Yangzi River.” An emperor’s failure to do so required that officials and commoners alike demonstrate loyal protest even to the point of self-sacrifice. The legacy of Qin Gui was an almost unavoidable question for those like Wang Mingqing who came to maturity in the last decade of his rule. Despite his critical outlook on the vindictive politics of the time, Wang Mingqing had, in the second installment of his notebook, given some credit to the opinion pieces Qin Gui had signed in 1126 in which he appeared uncompromising in his commitment to the continuation of a unitary Song state (see above). In later installments, however, the author withdrew this endorsement and cast doubt upon Qin Gui’s loyalty. He related a story by You Jiuyan 游九言 (1142–1206), who had learned from one of Qin Gui’s colleagues at the time, Ma Shen 馬伸 (1097–1129), that he was the author of these opinion pieces. Qin Gui had been hesitant about Ma Shen’s proposal, and Ma Shen therefore had all his colleagues at the Censorate exert pressure on Qin Gui, who was head of the Censorate at the time. Qin Gui finally gave in and signed. Ma Shen later became a victim of Qin Gui’s paranoid style of government, but by sharing this story You Jiuyan and Wang Mingqing confirmed that “the loyal achievements of Ma Shen subsequently became clear at the time [of Qin Gui’s death].”55 Qin Gui’s perfidy had thus been shown to date back to the last days of Qinzong’s regime. Emperor Qinzong himself had shown a commitment to the recovery of all Chinese territories. Wang Mingqing recollected an exchange between Qinzong and Ren Shenxian 任申先 (fl. 1140) in which the emperor remembered that, under his predecessor Huizong, Ren had proposed a plan to wage a war against a weakened Liao. The hope had been that these were the ideal circumstances for Song to retake the northern Sixteen Prefectures, which had been occupied by the Liao throughout the reign 55. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 4.310.



representing the foreign other

421

of the Song dynasty (see chapter 3). The emperor conferred an honorary jinshi degree on Ren Shixian and commented: “Had we acted according to your words, how could we not have obtained [the provinces of] Yan and Yun!”56 By contrast, Qin Gui was shown time and again to lend support and forgiveness to those who had surrendered to Jin or who recommended the acceptance of Jin conditions to the Song court.57 Those who had first recommended Qin Gui to the court were similarly reported to have moved back and forth between Jin and Song and were thus similarly suspected of Jin sympathies.58 Qin Gui may have enjoyed the trust of Emperor Gaozong and was long considered a loyal servant of the restored Song court, but for Wang Mingqing and his interlocutors loyalty was determined by a commitment to territorial sovereignty. Qin Gui became an antihero, and those whom he had convicted for treason were upheld as new models of loyalty. Most controversial were the generals who had advocated war against the Jin forces. Wang Mingqing included a full transcript of the witness account that had been used to convict Yue Fei 岳飛 (1103–41). The transcript detailed conversations between the assistant controller-general Wang Jun 王俊, who made the deposition, and his superior, which suggested that the latter was being recruited by Yue Fei into a plot against the court. Wang Mingqing concluded that this was a fabrication obtained under duress and thought it important to share with peers the kind of evidence that had led to Yue Fei’s summary execution. Even though Yue Fei’s name had been cleared, Wang Mingqing expressed the hope that this miscarriage of justice be remembered and learned from.59 Apart from the better-known politicians and generals whose loyalty or disloyalty had been the object of much discussion, Wang Mingqing also commemorated the words and actions of several lesser-known individuals who had shown loyalty to Song in the face 56. Ibid., 4.267. 57. Ibid., 3.256–67, 4.311. 58. Ibid., 4.318. 59. Ibid., 4.313–18.

422

imperial information networks

of Jin threats and enticements. These were individuals whose words and actions varied, with some preferring death and others a retreat from political life in the face of Jin invasions, but whose stories broadly shared two characteristics: a refusal to accept any Jin overtures that violated Song sovereignty and an ability to persuade others to do the same. The recollection of these stories in the last decades of the twelfth century suggests that loyalty carried new meaning for lower-level literati such as Wang Mingqing. Eleventh-century politicians and historians had shown an interest in loyalty as a moral concept and evaluated the performance of past political actors based on the notion of an enduring commitment to service to one dynasty or one ruler only.60 The stories that were circulated in the post-1127 era turned to a normative script which can be better described as loyalism. The stories outlined a course of action that started with brave military resistance in the face of extreme stress, the public rejection of surrender, and defiance in the face of death justified by a personal endorsement of loyalty to emperor and court until the bitter end. The script would not be complete until such words and actions were publicly and officially commemorated. Loyalism had thus already become a shared script for provincial elites as well as high-ranking political and military officials well before the SongMongol conflict of the mid-1200s. From a collector friend, for example, Wang Mingqing had obtained a sheet in which an anonymous author recounted the story of three valiant men; their single-minded loyalty, righteous anger, and outspokenness justified in his view further disseminating their stories. In these stories, as in imperial Chinese political and historical writing more generally, direct quotations attributed to the protagonists are used at length and at critical moments, a technique that established a more direct relationship between the reader and the main protagonists and rendered their words and actions more memorable. For example, Wang Bing 王稟 (d. 1126), commander in chief of Tai­yuan, was celebrated for having broken the siege of the city of Taiyuan, charging into the Jin camp, and slaying dozens of enemies. When it 60. Jennifer Jay (A Change in Dynasties) similarly explores the construction of loyalism in accounts of resistance to the Mongols in both contemporary and later sources.



representing the foreign other

423

appeared that the pacification commissioner and the circuit inten­ dants were about to surrender, Wang Bing confronted them with five hundred of his men threatening to kill any who would surrender: “Do you want to be officials?” They said, “Yes.” Bing said, “When you achieve things for the court, then you can serve in officialdom.” He also asked them, “Do you want rewards?” They responded, “Yes.” Wang Bing said, “When you resist the enemy for the court, then you can earn rewards.” He continued, “Since you want office and rewards, you should exert yourselves to the utmost to serve the dynasty with loyalty. If there are any among you who speak of surrender, what should be done?” The soldiers raised their weapons and said, “We would wish to kill them.” Again he asked, “If I were to speak of surrender, what should be done?” The soldiers responded, “We would also request to kill for this reason.” Once more he asked, “If the pacification commissioner and the circuit intendants were to speak of surrender, what should be done?” The soldiers said, “We would also request to kill them for this reason.” 61

After this exchange none dared go back to the subject of surrender. Wang’s army and the townspeople fought back, but after conditions had worsened to the point that the inhabitants started cooking wooden bows and arrows and eating human flesh they finally accepted defeat. Local men reportedly set themselves on fire carrying the image of the second emperor of the Song dynasty. Wang Bing himself set his family on fire and waited for Jin soldiers to meet him in his front hall. When they arrived, they asked him to surrender and offered him a drink. Wang Bing threw the wine in the soldiers’ face and spoke his last words, “How could I drink the wine of uncivilized bandits!”62 Wang Mingqing shared the sense of an anonymous source that the actions of men who had proven their loyalty in action ought to be commemorated. They recommended that the recognition for loyalism consist of an entire package: a conferral of honorary 61. Wang Mingqing, Huizhu lu (Zhonghua shuju ed.), 3.247. 62. Ibid., 3.248.

424

imperial information networks

names, a burial equivalent to that accorded to those of the highest ranking at the court, the erection of shrines with real-life images, the engraving of the events in which they had proven their worth on stone steles, the transcription of these narratives in Song dynastic history, and the bestowing of appropriate tokens of honor to their descendants.63 In sum, Wang Mingqing’s notes rejected the kind of loyalty that Qin Gui had shown Emperor Gaozong and commemorated instances of radical commitment to the defense and restoration of Song sovereignty. He pointed to the documented support for territorial restoration by Emperor Qinzong, who had become a symbol for the need for mobilization following his capture and removal to the Jin capital. Wang Mingqing further bore personal testimony to the success of Song forces in the more recent battles of the 1160s. He recounted the Song victory of Caishi and wrote that, years after the battle, soldiers who had participated in it still remembered how Song naval commanders had through skillful planning driven out the Jin emperor’s vastly larger armies. Regrettably, they also remembered that they had obtained little reward for their accomplishments.64 Not only had Song forces shown that they were up to the task during this battle, but also it had become evident from conversations between Song and Jin forces during these years that the latter were staffed by many whose loyalty was questionable at best. Soldiers captured by Song armies who had been drafted into the Jin armies after the occupation of the northern territories were alleged to share pro-Song sentiments: “On the day we left home our elders told us, ‘When you spot the armies of the southern dynasty, do not by any means move forward to meet the enemy, just surrender.’ ”65 Stories such as these and others relating to missed opportunities in the recruitment of bandit forces suggested that Song should capitalize on the forces at its disposition.66 63. Ibid., 3.248. 64. Ibid., 3.262–63. 65. Ibid., 3.262. 66. Ibid., 3.258–59.



representing the foreign other

425

By presenting fragments and anecdotes from various sources and different interlocutors Wang Mingqing’s notes suggest how prominent an issue Song-Jin relations had become in the conversations of provincial elites. The author’s position in this debate emerges from his selection of stories and the summary comments he attached to them. Read next to each other, the prowar sympathies of Wang Mingqing and some of his interlocutors emerge, and the memories of Jin violence are counterbalanced with memories of loyalty and commitment to restoration. Multivocality was to an extent the raison d’être of notebooks; entries were often cast as transcripts of conversations and texts in which the voices of the original speakers were preserved true to their original intent. In Waving the Duster, the foreign other is thus represented in a variety of texts and in a range of terminologies. References to the other range from the high-polite form in petitions, and the more neutral “people of Jin” and more generic terms with mild pejorative connotations in narratives, to the vitriolic dehumanizing terminology deployed in calls for war and for resis­ tance to peace negotiations. The multistate world in which Song politicians and educated elites had found themselves from the beginning of the dynasty may have resulted in the common association of the Jurchens in the north with the bureaucratic state they had developed by the 1120s. At the same time, Song politicians and literati used more generic and archaic terms to label the foreigner and added dehumanizing qualifiers to these terms in genres intended to rouse the population, officials, and the ruler to military action and to demand the execution of advocates for peace. Notebook authors such as Wang Mingqing intervened in debates about shared concerns, and such interventions were a significant part of literati networking. Wang Mingqing’s selections and interventions in Waving the Duster reflect both a broader ongoing concern about Song policy on border affairs and toward Jin in particular, and a mild endorsement of prowar opinion. When read against one another, the different voices discussing Song-Jin relations of

426

imperial information networks

c­ on­flict and negotiation harmonize. Their shared logic is the rhetoric of empire. This rhetoric allows for moderate antiforeign language, the acceptance of negotiation subject to Song control, and the vitriolic rejection of negotiation and the persecution of negotiators if and when the sovereignty of the ruler and his control over the Chinese territories are at stake. Long after the terms for peace with Jin were decisively accepted at the court under Gaozong and Qin Gui and reaffirmed in 1161, Wang’s notes testified to the continued circulation of texts demanding in different pitches the restoration of empire.

C onc lusions a n d Prospe c t s

F

rom a long-term historical perspective, the twelfth and thirteenth centuries are generally considered to have been times of relative prosperity. Despite recurrent natural and human-induced disasters, economic and social historians tend to agree that both in the north and the south the population recovered in due course from the war and displacement of the 1120s and 1130s, and population statistics on both sides showed an upward trend. Urbanization was uneven but in many parts of Song China (and to some extent in Jin China) the number of cities and urban residents grew and urban networks spread into rural hinterlands.1 While the Song state may have been less interventionist after 1127 than in the eleventh century, it appears to have been able to generate and sustain sufficient tax income to cover substantial annual payments to its northern neighbors, maintain a large army, and support a court and bureaucracy as sizeable as it had been in the early days. War with the Jurchens did not lead to an economic downturn, but the loss of the north left an indelible impression on the minds of those who witnessed the events and remained for future generations who lived in the south “a shame to be washed away.” In this sense the twelfth and thirteenth centuries were a time of crisis. The Song court and political elites operated under the cloud of a prolonged

1. For a general overview of urban history between the Tang and Song periods, see De Weerdt, “Chinese Cities.”

428

conclusions and prospects

geopolitical crisis both in real terms when faced with invasions from the north and as a result of the living memory of the unaccomplished recovery of the northern territories. Social and intellectual historians have in recent decades argued that Song military and diplomatic defeat led to a “turn inward” in the twelfth century, a turn that was characterized by a reduction in scope of the concerns that preoccupied both the state (central and local) and the literati.2 From a capital-oriented empire that aimed to centralize power in the hands of its top bureaucratic elites, the Song turned into a state with reduced ambitions accommodating the interests of local elites in local management and the accompanying Neo-Confucian ideology of moral self-cultivation. From the surviving textual legacies of the eleventh and twelfth centuries it appears indeed that local interests and the moral discourse of advocates of the Learning of the Way gained an unprecedented visibility after the 1130s. This kind of localization was however not accompanied by regional fragmentation, as had been the case during earlier times of military crisis. No doubt the hierarchical spatial organization of the state, its bureaucratic administration, and the institutionalization of regular civil service examinations were all factors that militated against fragmentation. These are the factors that historians and sociologists commonly point to in order to explain the presumed continuity of unified rule throughout Chinese history, but these, alongside other factors such as Confucian ideology, have proven to work equally well at geographically more reduced scales. During periods of multistate rule, such as during the tenth century, rulers tended toward the very same bureaucratic technologies. This book proposes that during the twelfth century localization went hand in hand with the consolidation of a commitment to unified rule over the Chinese territories among literati in the provinces. The precise dimensions of the territories that were considered to have historically been Chinese were a matter of debate, but twelfthand thirteenth-century stelae and printed maps along with their reading instructions suggest that at a minimum the Chinese terri2. I review recent social and intellectual historical scholarship in more detail in “History” and “Recent Trends in American Research in Song Dynasty History.”



conclusions and prospects

429

tories covered all territories up to where the walls were imagined to have been and including the Sixteen Prefectures in the north, down to where the most southward of the Five Mountains was located in the south, east to the coastline, and west to where the course of the major east-west rivers ended. The normative dimensions of the Chinese Empire that were thus constructed need not include areas on the western, southern, and northern periphery that had been incorporated under the unified empires or smaller states of the past, but the maps, and the literature surrounding them, emphatically underscored that the northern and southern halves in which the Song and Jin courts had divided the land were an anomaly that required literati attention. Song military and diplomatic defeat, not only in the 1120s but also in the decades that followed, led to an outpour of publications about current affairs, some leaked from government offices and others produced in growing numbers by literati. Single-sheet official documents, court gazettes, and archival compilations were leaked, sold, and repackaged. Literati wrote, copied, and bought maps, atlases, military treatises, military geographies, diplomatic treatises and reports, border and intelligence communications, and gazetteers. In their poetry, letters, and notebooks they fashioned identities that tied them to the places that constituted the empire (court, capitals, borders, and the Chinese territories). This book has not covered the variety of such publications exhaustively. I have aimed instead to interpret parallel developments in the authorship and publication of a range of state-related documents and collections as indicative of a structural transformation in political communication. Below I first briefly review the institutional, legal, and cultural manifestations of this transformation and its key features. The key questions are twofold. First, was the paradigmatic shift toward localism among elites in the twelfth century (social change) accompanied by changes in political communication between court and provincial elites (political change)? What effect did any such changes have on political institutions and political imaginaries? Second, how can we best explain the simultaneous development of centralization and censorship on the one hand, and the breakthrough of private and commercial printing on the other?

430

conclusions and prospects

I will conclude with some observations on the implications of the Jingkang crisis and its aftermath for Chinese history and the history of empires.

The Crisis and Maintenance of the Song Empire It is a key contention of this book that the Jingkang crisis led to the stabilization of a structural transformation in the relationship between court and scholar-officials not only in the social sphere but also in political communication. The localist focus of the growing numbers of literati was counterbalanced by a change in the relationships and structures of political communication that created more space for the involvement of literati in the discussion of current affairs and placed the court and its business at the center of literati attention. There are three main levels in which this change revealed itself: at the institutional level, at the level of legal history, and at the level of text production. First, at an institutional level, this change manifested itself in the porousness of institutions that had been set up to centralize communication in the early decades of the Song dynasty but that became ever more visibly subject to the infiltration of literati networks from the late eleventh century onward. In the tenth and eleventh centuries the Song rulers implemented a series of policies of centralization that were far more radical than those implemented under their Han, Sui, and Tang predecessors. For example, the institution of the civil service examinations as a three-tiered system culminating in the palace examinations overseen in theory by the emperor himself has been seen as a crucial step in a long-term process of growing autocratic rule. There is no doubt that Song policies and political theory provided solid ground for the absolute primacy of the imperial person in all decision making. However, in the course of the late eleventh and twelfth centuries the early policies of centralization aimed at the institutionalization of imperial control over regional and local power holders were gradually refitted to allow for two-way traffic between the court and local elites. Institutions such as the civil service exami-



conclusions and prospects

431

nations, the archives, and the Memorials Office, which were in the early decades of Song rule designed to reassert imperial control over the bureaucracy and local literati, took on an additional role in communications between the court and the provinces by the twelfth century. They developed into sites through which lower-level officials and literati gathered information about the court and its policies, discussed it, and fed it back to the center. For example, the Memorials Office (Jinzouyuan) changed from intelligence offices for regional commanders set up in the Tang capital in the eighth and ninth centuries to a central bureau whose mission was to serve the court’s interest in the gathering, sorting, and selective dissemination of information in the late tenth century. This mission proved difficult to accomplish. Memorials Office personnel and intermediaries who made their way into the palace grounds to gather information were time and again accused of selling gazette reports and other court news even before it had been authorized and sent through the appropriate bureaucratic channels. The intended audience of the official court gazettes (prefectural magistrates) shared them with local literati, and private derivatives called “short reports” were on sale on the streets of the capital cities. By the twelfth century, gazettes were quoted in private correspondence and a new subgenre of poetry capturing the reader’s impressions “upon reading the gazette” emerged. The practice of writing and publishing about one’s response to news items suggests that the broader communicative function of the gazette had become acknowledged. For literati, gazettes had become a site for the construction of a class-based empire-wide imagined community by the twelfth century at the latest. The reading of gazettes among the twelfthand thirteenth-century literati expressed a more general interest in current affairs. They read them for personal reasons and to network with colleagues and friends across the vertical hierarchies of the official communication network. As argued in chapter 2, the wider dissemination of the court gazettes also coincided with an explicit call for the inclusion of scholars and students in the body politic in some circles.

432

conclusions and prospects

Second, at the level of legal history, structural change manifested itself in the overall tendency on the court’s part to balance a growing body of publishing regulations with the acknowledgment of literati information needs. Between the eleventh and mid-thirteenth centuries, the Song government repeatedly issued bans on the leaking, transmission, and publishing of the different types of official documents and compilations discussed in the first five chapters. The regulations targeted single-sheet documents, court gazettes, archival compilations, Song dynastic history drafts, and memorials and examination essays on border affairs. Changes in Song legislation on what were considered to be sensitive materials reflected both the widening scope of materials that circulated beyond their intended audiences and the increased use of the print medium in their dissemination. The repetition of such regulations over time suggests that their implementation was intermittent and ineffectual. As shown through­ out the chapters of this book, the targeted materials were available to collectors and readers for consultation and reuse. Private collectors included in their collections and their catalogs archival compilations at various stages of their transformation into dynastic histories alongside other prohibited materials. Compilers and commercial printers of examination encyclopedias included excerpts from the latest editions of the archival compilations and draft histories in their works, as did notebook authors such as Wang Ming­qing and Zhang Shinan. Wang Mingqing also transcribed single-sheet documents into his notebooks, which he had printed in suc­cessive install­ments. The military treatises of Hua Yue were incorporated into encyclopedias and circulated under alternative titles following the ban of 1213. Commercial printers also gathered policy essays and memorials on border affairs into essay collections and classified encyclopedias, where they remain to bear testimony to the types of questions and the range of materials students could be expected to tackle. Such infractions were prosecuted—and they also went unnoticed. Moreover, the practice of copying prohibited materials and their preservation in family collections were tacitly acknowledged by the court, as it rewarded collectors for allowing court copyists access to these materials when original copies went missing in the palace libraries.



conclusions and prospects

433

In retrospect, we see that the tacit acknowledgment of the circulation of prohibited state documents had a significant effect on the maintenance of imperial structures and traditions. Secrecy and publicity can, also in retrospect, be seen as parallel processes that ensured the continued collaboration of growing numbers of literati. Access to current affairs had become important to literati for at least three reasons. First, familiarity with archival collections and Song dynastic history was necessary in political discourse and political practice. Second, beyond the state bureaucracy, familiarity with Song dynastic history and current affairs was needed to prepare for the policy essay session in the examinations and had also become standard in literati conversation. Third, literati depended upon the gazettes and their derivatives at various stages of their careers and used information about new appointments and recent policy decisions in networking behavior. The focus of book and media historians on prohibitions and censorship obscures the importance of publicity as an increasingly significant aspect of state control in Song times. The benefits to the Song state of a repertoire of control consisting of both the insistence on secrecy and the acknowledgment of publicity were several. The circulation of unedited reports and therefore more sensitive materials such as single-sheet documents and court gazettes placed the court firmly at the center of literati networks and interests. By positioning the court at the center of competing networks these genres confirmed its political authority even when they fell in the hands of unauthorized readers. The continued centrality of the court in literati discourse during the twelfth century is also illustrated in Wang Mingqing’s series of notebooks, in which the author and his interlocutors comment profusely on the Song court and continue to perceive family prestige as a function of achievement at court. Literati engagement with the official record also strengthened identification with the dynasty. Wang Mingqing and his informants drew substantially from Song archival and historical compilations but also shared and transcribed edicts, petitions, and letters not included therein. The circulation of notebooks provided an avenue to discuss the Song official record and to supplement and revise it with other relevant materials.

434

conclusions and prospects

The disclosure of the archive remained ambivalent. It created space for the evaluation of particular Song politicians’ performance and approach to current problems—in Wang Mingqing’s case, the reign of Qin Gui and the policy of appeasement. On the other hand, the insistence on secrecy on the court’s part allowed for the occasional prosecution of those whose words might revise the rec­ ord in ways inconsistent with the line of the reigning regime, as in the case of Li Guang and others whose notebooks and histories were confiscated in the 1140s. Secrecy and the always present threat of the implementation of the rules imposed to protect it were the stick complementing the court’s acknowledgment of literati access to its record. Third, the court’s retreat from a more interventionist position in scholarly discourse was further manifested in a remarkable shift in the production of texts relating to its affairs. Throughout the first half of Song history compilations of Song archival materials, digests of imperial pronouncements, maps and gazetteers of the Chinese territories, and administrative and historical encyclopedias were mostly compiled by officeholders and were commissioned by the court. Following the crisis of the 1120s, private scholars, many of them without examination degrees and official posts, set out to work on large-scale projects in the provinces. They compiled historical and military geographies, maps of the Chinese territories, archival compilations, administrative encyclopedias, and local gazetteers, thus adopting and adapting the genres that had heretofore been monopolized by the court and the bureaucracy. In addition, secondary discourse on Song affairs proliferated in the twelfth century as the compilation and publication of notebooks (records of conversations and reading notes) and correspondence by lowerlevel literati took off. The breakthrough of the print medium can similarly be dated to the twelfth century.3 It was only then that it began to be used for all manner of written texts and increasingly for the notebooks and collected writings of Song scholars and those whose works they most admired. 3. On a shift in the production of maps and gazetteers, see also Mostern, Dividing the Realm in Order to Govern.



conclusions and prospects

435

The developments outlined above were already gaining momentum in the late eleventh century, but it was the crisis initiated in the 1120s that allowed for a structural transformation to take hold. In the decades that followed the readership for materials relating to the history and policies of the Song court kept expanding and literati resident in the provinces became more visible among the producers of the texts that defined the standards for literati membership. The enduring decline of the court’s position in the field of publishing, and more broadly in the management of local communities or in the setting of examination and curricular standards, ironically appears to have brought with it a strengthening of literati commitment to the imperial state and not, as has been argued, a turn away from the center and imperial government. The articulation of a commitment to the imperial state was a striking feature of the maps, atlases, comprehensive histories, anthologies of policy essays, and historical and administrative encyclopedias produced by literati in the twelfth and thirteenth centuries. The empire maps, for which the earliest surviving examples date to the Southern Song period, engraved on large stone stelae or carved on woodblocks by private and commercial printers to facilitate wider transmission among the literate elite, outlined the contours of the Chinese commonwealth as they should be, encompassing all former Song territories. Twelfth- and thirteenth-century inscriptions accompanying the graphic representations of the Chinese territories as a unified political entity acknowledge the historical fact that times of divided rule have been dominant in the history spanning the disintegration of the early Zhou state in the eighth century BCE up until the time of writing. At the same time, these maps and the poems written about them express a resolve among the elite to keep the recovery of the north and the full restoration of Song imperial rule on the political agenda.4 In a thirteenth-century textbook, Ruxue shuyao 儒學樞要 (The essentials of the learning of scholars), the basics of literati knowledge are defined as the administrative geography of the Song Empire pre-1127, the chronology of Chinese dynasties and their rulers (including all Song era names for all Song 4. De Weerdt, “Maps and Memory” and “The Cultural Logics of Map Reading.”

436

conclusions and prospects

rulers up to Ningzong), and an overview of cosmological concepts that in combination legitimized the place of the Song court in the succession of Chinese dynasties. At the same time, the editor of this work tied the Song court’s legitimacy to the repossession of the northern homeland. Interestingly, the newly influential NeoConfucian tradition does not figure prominently in its pages. As I argued in chapter 8, the prioritization of territorial concerns also had an impact on other aspects of political culture. Loyalty to emperor and dynasty became defined by the court’s commitment to the restoration of the Song homeland. Finally, the structure and the geographic scope of the networks within which texts were exchanged were a second significant feature of the surge in literati textual production about the affairs of the reigning dynasty. Prevalent models such as William Skinner’s macro­ regional model have led to the analysis of political communication in imperial China in two more or less disconnected realms, the topdown administrative hierarchy and local nodes of exchange. The geography of elite communication remains a largely unexplored field. The geographic distribution of networks of information sharing as recorded in correspondence and in notebooks of the twelfth and thirteenth centuries suggests that cross-regional communication was a common phenomenon among imperial elites. This may further imply that, even if we grant that literati marriage patterns became geographically more restricted in the post-1127 period, and that they therefore invested more heavily in local ties and local interests, this did not preclude an ongoing political engagement with the center and even a strengthening of an attachment to the empire. In sum, the geopolitical crisis of 1126–27 brought about a restructuring of the information order. Literati became the main producers and consumers of all manner of texts relating to the history and current affairs of the Song dynasty and increasingly turned to woodblock printing for the dissemination of these texts. In these texts the recovery of the north continued to be a core theme. The communication networks through which these texts were circulated and discussed ranged widely across the vast extent of the Southern Song Empire. Court and central government institutions lost ground in the field of cultural production. However, they con-



conclusions and prospects

437

tinued to exercise control due to the demand for primary texts emanating from the center. Central government and court personnel played an important role in the dissemination of single-sheet state documents, archival materials, and gazettes. The dissemination of political information and the use of printing to this end need not be associated with a public sphere. The large-scale leaking of current affairs to officeholders and non-officeholders alike can in the case of late imperial Chinese history be seen as crucial elements in the consolidation of empire. What are the implications of the Jingkang crisis and its aftermath for the history of empires and especially the representation of Chinese history within it? First, by inserting an atypical case into the sequence of Chinese empires that have been at the core in comparative studies of empire, we have restored some measure of historicity to the narrative of the maintenance of imperial traditions in Chinese history. Sima Guang’s celebrated observation of the prevalence of discontinuity serves as a reminder that the prolonged post-1127 geopolitical crisis became a watershed in Chinese history. Following the reunification under Mongol rule in the late thirteenth century, periods of divided rule lasted a few decades at most, and the c­ enturieslong break-up of the Chinese territories as portrayed in Song maps became a thing of the past. Even though violence, coercion, and the institutional innovations of the Yuan and early Ming courts also contributed to this change, elite collaboration with Chinese and non-Chinese regimes on the winning side was also typically justified by their ability to restore territorial unity. Eisenstadt and those who followed in his wake rightfully underscored the role of the literati elite in the history of Chinese polities. My investigation of the literati elite’s response to the Jingkang crisis and the restructuring of the information order that ensued suggests their role became crucial in the formation and maintenance of vast unified empires from the twelfth century onward. Second, this case also suggests that the adoption of an imperial mission, seen as a key component by Münkler in the maintenance of empire, was for the literati the result of a collective response to

438

conclusions and prospects

the 1127 crisis. Crisis moments and the response to them by different social groups should take up a greater role in the comparative history of empires and other types of polities. Even though longerterm developments underlay the growing prominence of the literati (especially the validation of the civil service examinations and the exponential rise in the numbers of men sitting and preparing for them), it was only in the post-1127 era that they began to articulate an imperial mission in textual and graphic representations of an idealized Chinese Empire. Third, the imperial mission was further sustained by literati communication networks that stretched not only vertically from the lower units of administrative organizations to the center of the imperial administration but also horizontally across regional boundaries. Travel and hosting travelers were a common feature of literati life as students went in search of teachers and took trips to sit the examinations, and younger males traveled along with family members taking up positions away from home, or as those appointed to official posts moved between assignments, home, and the capital. Growing corpora of correspondence and notebooks further remain as testimony to relationships forged near and far. The structure and geography of elite communication networks, the identities they sustained, and their dynamics over time should be explored in greater detail in order to examine their role not only in the long-term maintenance of empire in the second millennium of Chinese history but also in the maintenance or contraction of polities elsewhere.

A ppe n di x 1 Supplementary Tables

Zhang Mao: former general of the Right Imperial Insignia Guard Tang Kai police chief in Jinyuan County, Chuzhou Prefecture Huang Meng court gentleman for ceremonial service (official rank)

the families of court officials

Shaoxing 1

Shaoxing 1

court officials

Mrs. Wang, wife of Zhang Mao

Shaoxing 1

Shaoxing 1

He Kezhong jinshi

in accordance with the award granted to Huang Meng (rank promotion)

5 certificates (rejected); rank promotion

SHY-CR 4:21 “Taizu huangdi shilu,” 50 juan; “Taizong huangdi shilu,” 80 juan; “Zhenzong huangdi shilu,” 150 juan; “Renzong huangdi shilu,” 200 juan; “Yingzong huangdi shilu,” 30 juan; “Tiansheng nanjiao lubu ceji,” 11 ce “Li shu,” “Kaiyuan li,” “Yi jing,” “Li yi,” SHY-CR 4:21 “Cui Tongdian,” “Kaibao tong li,” “San li tu,” “Jiaomiao fengsi liwen,” “Guochao huiyao,” “Liu dian li ge xin bian,” “Taichang yinge li,” “Daguan li shu,” “Liujia haofa,” “Zhenghe xu bian huiyao,” “Kaiyuan li baiwen,” “Taichang xin li,” “Jiangdu ji li,” “Qutai li,” “Zong fan qing xilu,” “Kaiyuan li yi zuan,” “Wu li jingyi”

10 certificates

Reward

rank promotion

SHY-CR 4:21

SHY-CR 4:20–21 a

Source

SHY-CR 4:21

“Chongxiu guochao huiyao,” 300 juan

“Taizu Huangdi shilu,” 4 ce; “Guochao ­baoxun,” 12 ce; “Mingchen liezhuan,” 2 ce; “Guochao huiyao,” 3 ce Shilu, huiyao, guoshi zhi, 222 ce

Donor Donor occupation/title location Donation

Shaoxing 1 (1131)

Date Donor

Table 1 List of texts and select art objects donated to the early Southern Song court (1131–59)

440

Wei Xu

Lin Yan

Xu Zhong

Mao Gangzhong

Baoyue

Shaoxing 2

Shaoxing 3 (1133)

Shaoxing 3

Shaoxing 5 (1135)

Shaoxing 5

Shaoxing 2

He Lin (the family of He Zhu) Zeng Min

Shaoxing 2 (1132) records from several reign periods, over 2,000 juan In Chen Kui, cited as 2,678 juan

SHY-CR 4:21–22; also Lu Youren, Yanbei zazhi, shang, 5b SHY-CR 4:22; also Chen Kui, Nan Song guan’ge lu, xu lu 3.22 SHY-CR 4:22; also Li Xinchuan, ­Jianyan yilai xinian yaolu 56.747a SHY-CR 4:23

jinshi

imperial writings by Taizong and other Wuhu County in materials Taiping Prefecture gentleman for atimperial writings by the imperial father, tendance (official ­imperial paintings and calligraphy rank) (7 scrolls), shilu, huiyao, guoshi, and classical texts (2,122 juan) prefect of Jingjiang “Zhenghe chongxiu guochao huiyao,” SHY-CR 4:23 “Zhenghe xiuding haofa,” “Xuanhe chong­ xiu lubu ji,” 3 titles gentleman for fos“Jian gu tuji” (compiled in 1040 at the SHY-CR 4:24 tering temperance Guanwen Palace under Emperor Renzong) (military rank) monk military treatises: “Li Wei gong bi sheng ji,” SHY-CR 4:24; “Bing qin,” “Shui jing,” “Wulue yaoyi,” also Li Xinchuan, “Guanzi,” “Qingtian ji,” “Mozi,” “Guiguzi,” ­Jianyan yilai xinian “Feng yun lun,” “Xin shu,” “Yuju tongguan yaolu 91.294b mijue,” “Anbian ce,” “Liu bin ji,” “Ping hu ce,” and formation blueprints, 39 titles

former vicechamberlain for ceremonials

Pingjiang 5,000 juan Prefecture

(continued)

subbureaucratic assignment

rank promotion

honorific title and official assignment

honorific title

honorific title and official assignment honorific title

441

Shaoxing 15 Qin Zhenqing

Shaoxing 15 Chen Yi

Shaoxing 15 Li Dazhao (1145)

Shaoxing 13 Shen Jiayou (1143) Shaoxing 13 Lu Zai

Li Deguang

Shaoxing 9 (1139)

gentleman for court calligraphy, 1 scroll service (official rank); prefect of Jianzhou jinshi Mingzhou 756 juan (all items of use to the Imperial Library) jinshi Anyue calligraphy, 4 pieces County in Puzhou

SHY-CR 4:27

SHY-CR 4:27

exemption from examinations payment

Shi Su, Jiatai Guiji zhi 16.29b; also SHY-CR 4:26 (see List of Requests for donations, appendix table 2) SHY-CR 4:27 payment

13,000+ juan

payment

honorific title

Reward

SHY-CR 4:25

SHY-CR 4:25

SHY-CR 4:24; also Shi Su, Jiatai Guiji zhi 16.29b

Source

“Chunqiu sanzhuan,” Directorate edition

“Zhenzong huangdi yulu” and paintings of meritorious officials, 2 ce

“Cefu yuangui,” and other titles, 11,515 juan In Shi Su, cited as 8,546 juan

judicial investigator for the High Court of Justice (official rank) Also a commoner jinshi Wujiang County in Pingjiang Prefecture

Zhuge Xingren

Shaoxing 5

Table 1 (continued)

Donor Donor occupation/title location Donation

Date Donor

442

a

273 stone inscriptions

The abbreviation SHY-CH is used here to cite the Song hui yao, Chong ru section.

gentleman-litterateur (official rank)

Shaoxing 29 He Lin (1159)

51 titles (all items missing from the Imperial Library) Fenghua the imperial collections of Shenzong Shaoxing 16 Chen Taichu (1146) County in and Zhezong, 118 ce Mingzhou Prefecture righteous gentleman calligraphy, 1 piece Shaoxing 17 (1147) (military rank) Shaoxing 17 Qian Yunkui gentleman for meri2,990+ juan (all items missing from the torious achievement Imperial Library) (official rank); viceprefect of Jiande County (Yanzhou) jinshi calligraphy, 1 piece Shaoxing 18 Wu Jie (1148) Shaoxing 18 Guo Shixin gentleman for mericalligraphy, 1 piece torious achievement (official rank), administrator for public order (Chengdu) his collection Shaoxing 25 Chen Youdi gentleman for meri(1155) torious achievement (official rank) Shaoxing 25 Su Cao jinshi Meizhou “Su Yuanlao’s Literary Collection,” 25 ce; paintings and calligraphy, 3 scrolls Shaoxing 25 Wang Yan jinshi Pengzhou paintings, 15 scrolls

loyal gentleman (military rank)

Shaoxing 15 Zhang Lun rank promotion

rank promotion rank promotion

exemption from examinations rank promotion

SHY-CR 4:28

SHY-CR 4:29 SHY-CR 4:29

SHY-CR 4:29

exemption from examinations exemption from examinations official assignment

SHY-CR 4:29

SHY-CR 4:29

SHY-CR 4:29

official assignment

SHY-CR 4:29

SHY-CR 4:29

rank promotion

SHY-CR 4:27

443

three monasteries

Shaoxing 3 (1133) (S)

Shaoxing 3 (S) Shaoxing 3 vice-director of the Right Office (G)

(in the Department of Ministries), Liu Cen

Buddhist monastery in Wuhu County, Tai­ ping Prefecture

Shaoxing 2 vice-director (1132) of the Imperial (S) a Library, Hong Taiyan

Han Lü (Han Qi’s grandson)

Yu Shen (ex-minister from Fuzhou); Zhao Tingzhi (ex-minister from Quanzhou); Xue Ang (ex-assistant councilor from Yanzhou) Lin Shujia (ex-assistant councilor from Huzhou)

Targeted Targeted locations collectors

Date Issuer

SHY-CR 4:22–23

SHY-CR 4:22

SHY-CR 4:22

imperial writings; draft histories, veritable records, and collected statutes; materials falling in the four bibliographic divisions (classics, history, masters, and collected writings) Han Qi’s letters, titled “Erfu zhongyi,” 100 juan missing books

SHY-CR 4:22 b

draft histories and veritable records; the materials of Cai Jing

Types of materials requested Contents Source

Table 2 List of requests for donations of texts by the early Southern Song court (1132–59)

444

Secretariat, Wang He

Shaoxing 9 imperial recorder,

(G)

Historiography Institute, Fan Chong Shaoxing 9 Historiography Institute (1139)

Shaoxing 6 decree in response to senior (1136) compiler in the (G)

Shaoxing 5 (1135) (G) Shaoxing 5 (S)

Library, Zeng Tong Shaoxing 4 imperial recorder, Chancellery, (1134) Chang Tong

Shaoxing 3 vice-director of the Imperial (S)

the families of high officials

collected statutes of the Song dynasty

monographs from the draft history of the Shenzong period; and annals, monographs, and biographies from the draft history of the Zhezong reign

missing veritable records for the years 1092–93

(continued)

SHY-CR 4:25

SHY-CR 4:25

SHY-CR 4:24

SHY-CR 4:24

Zhezong’s veritable records

Wuzhou

Zhao Mingcheng’s family collection (former auxiliary in the Hall of the Dragon Diagram) officials, scholars, and commoners

SHY-CR 4:24

SHY-CR 4:23

SHY-CR 4:23

all sorts of writing

draft histories, veritable rec­ ords, precious instructions, collected statutes

draft histories and veritable records of the Shenzong and Zhezong reigns, dynastic statutes, documents

all government the people schools

Hong Ji (former library official)

445

Shaoxing 13 (G)

Shaoxing 13 (G)

Shaoxing 12 (1142) (S) Shaoxing 13 (1143) (G) Shaoxing 13 (S)

emperor

the fiscal inten- scholars and dants in all commoners routes

Lu Zhi’s family collection (Shaoxing Prefecture)

local scholars

missing titles

Yu Shen (ex-minister Directorate of Education from Fuzhou) [cf. editions Shaoxing 2]

SHY-CR 4:26; also Shi Su, Jiatai Guiji zhi 16.29b (see appendix table 1) SHY-CR 4:26

SHY-CR 4:25–26

approval of Xiang’s request to have catalogs of missing titles printed and distributed across the empire order to fiscal intendants SHY-CR 4:26–27 to investigate collections in their subordinate prefectures and counties and give out appropriate rewards

request for scale of rewards

SHY-CR 4:25

Types of materials requested Contents Source

Table 2 (continued)

Targeted Targeted locations collectors

provisional all prefectures commissioner of Xuyi Prefecture, Xiang Zigu

emperor

Date Issuer

446

b

a

supervisor, Imperial Library, Qin Xi

route intendants and prefectural magistrates

Sichuan

all administra- private collectors tive subdivisions order based on Qin Xi’s request to have local administrators report on their collecting activities, to implement the policies designed by the Imperial Library, and to give rewards to donors missing items; original pieces the promulgation and of calligraphy from the Jin distribution in printing and Tang dynasties of a reward scale designed by Qin Xi communication that the Imperial Library was to check into local collecting efforts on a monthly basis reminder of an earlier instruction to local officials to make book collecting a priority admonition that the ­reward scale was to be implemented more enthusiastically

The abbreviation (S) in the first column (the date column) indicates a specific request; the abbreviation (G) a general request. The abbreviation SHY-CH is used here to cite the Song hui yao, Chong ru section.

Shaoxing 29 (1159) (G) Shaoxing 29 (G)

Shaoxing 16 (1146) (G) Shaoxing 16 (S)

Shaoxing 15 (1145) (G)

SHY-CR 4:29

SHY-CR 4:29

SHY-CR 4:28; Chen Kui, Nan Song guan’ge lu, xu lu 22 SHY-CR 4:28–29

SHY-CR 4:27–28; for Taizong’s precedent see SHY-CR 4:15–17

447

ce

juan

1

Shaoxing 29 (1159)

273 inscriptions

Note: The numbers of titles, ce, and juan should be added together for an indication of the total number of the donations in a given year. Titles are listed only when the number of ce or juan is not recorded. Similarly, the number of ce is given only when the number of juan is not recorded.

5+ 3 2

Shaoxing 1 (1131) Shaoxing 2 (1132) Shaoxing 3 (1133)

(unknown) 253 ce 810 juan (unknown) 7,678 juan 3 titles 2,222 juan 7 scrolls 39 titles 11,535 juan Shaoxing 5 (1135) 3 (8,556 juan) Shaoxing 9 (1139) 1 2 ce 1 title 13,000+ juan Shaoxing 13 (1143) 2 Shaoxing 15 (1145) 4 2 scrolls 756 juan 3 pieces 51 titles 118 ce Shaoxing 16 (1146) 1 1 title 2,990+ juan Shaoxing 17 (1147) 2 Shaoxing 18 (1148) 2 1 piece 1 scroll 18 scrolls; 25 ce Shaoxing 25 (1155) 3 (unknown) (collection)

titles

Number of Year donations Size of donations

Table 3 Number and size of texts and select art objects donated to the early Southern Song court, by year (1131–59)

448

General requests

1

1 1

1

1 3

Individual requests

Note: “General requests” refer to those that were targeted for constituencies across the empire. “Individual requests” were addressed to individual families, groups of families and religious organizations, or specific counties, prefectures, or circuits.

Shaoxing 2 (1132) Shaoxing 3 (1133) 1 Shaoxing 4 (1134) 1 Shaoxing 5 (1135) 1 Shaoxing 6 (1136) 1 Shaoxing 9 (1139) 2 Shaoxing 12 (1142) Shaoxing 13 (1143) 3 Shaoxing 15 (1145) 1 Shaoxing 16 (1146) 1 Shaoxing 29 (1159) 2

Year

Table 4 Number of general and individual requests for donations of texts issued by the early Southern Song court, per year (1132–59)

449

Daidou fengshi lu (戴斗奉使録) Cheng yao lu (乘軺録) Shengchen guoxin yulu (生辰國信語録) Liushi xi xing lu (劉氏西行錄) Fu Wenzhong baguo yulu (富文忠八國語錄) Fu gong yulu (富公語録) Qingli fengshi lu (慶歴奉使録) Fu gui fengshi lu (富貴奉使録) Fengshi bielu (奉使别錄) Fu gui fengshi bielu (冨貴奉使别録) Qingli zhengdan guoxin yulu (慶歴正旦國信語錄) 1041 1042

Liu Huan 劉渙

Fu Bi 富弼

X X

X

Chao Gongwu, Junzhai dushu zhi

ca. 1042

ca. 1042 1 ca. 1042 X

Fu Bi 富弼

Fu Bi 富弼 Fu Bi 富弼

Yu Jing 余靖 1043 1

X

ca. 1042 1 X ca. 1042 X

X

Fu Bi 富弼 Fu Bi 富弼

1

1

1 X 1

ca. 1009 ca. 1030

Lu Zhen 路振 Kou Jian 冦瑊

X

2

Wang Shu 王曙 1006; 1009

You Mao, Shuichu Title Author Date Juan tang shumu

Table 5 Reports by Song envoys on their journeys to the Xia, Liao, and Jin, in the holdings of three private collectors

X

X

X

Chen Zhensun, Zhizhai shulu jieti

450

6 (5)

1 1

8 (6)

Xuanhe shi Jin lu (宣和使金錄) Lian Nanfu 連南夫 1124 Jingkang fengshi lu Zheng Wangzhi 1125–27 (靖康奉使録)   鄭望之

Subtotal for the Xia and Liao dynasties

Qidan jianghe ji anonymous post-1048 1 (契丹講和記) Liu Yuanfu fengshi lu Liu Chang 劉敞 ca. 1055 X (劉原父奉使録) Wang Jiefu songpan lu Wang Anshi? ca. 1063 X (王介甫送伴録)   王安石? Xining zhengdan guoxin lu Dou Bian 竇卞 1075 1 (熙寧正旦國信錄) Jie pansong yulu (接伴送語錄) Shen Jichang 1076 1   沈季長 Fuxiu jushi shi Liao lu Zhang Shunmin 1094 2 X (浮休居士使遼録)   張舜民;   Zheng Jie 鄭介 Shi Liao lu (使遼録) Zhang Shunmin 1094 X   張舜民 Zhang Yunzhai shi Liao lu Zhang Shunmin 1094 X (張芸齋使遼録)   張舜民 Shi Liao jianwen lu Li Han (李罕) [ca. 1103] 2 (使遼見聞錄)

(continued)

X X

7

X

X

X

451

Table 5 (continued) Chao Gongwu, Junzhai dushu zhi

Qian Qin Lu Changzhu Princess Qin Lu [1060s–1130s] X fengshi lu (錢秦魯長   秦魯長主 主奉使録) Zhang Zhongke fengshi Zhang Yi 章誼 1131 1 X Jinguo yulu (章忠恪奉使 金國語錄) Fengshi zalu (奉使雜錄) He Zhu 何鑄 1142 1 Guanpan rilu (館伴日錄) anonymous 1154 1 Shaoxing jianghe lu anonymous ca. 1131–62 2 (紹興講和録) Longxing fengshi shenyi lu Yong Xiji 1164 1 (隆興奉使審議錄)   雍希稷 Bei xing rilu (北行日錄) Lou Yue 樓鑰 1169 1 Lan pei lu (攬轡錄) Fan Chengda 1170 2 X   范成大 Qiandao fengshi lu Yao Xian 姚憲 1172 1 (乾道奉使錄) Fengshi zhi li lu (奉使執禮錄) Zheng Yan 鄭儼 1189 1 Pin Yan lu (聘燕録) Zheng Ruxie ca. 1193 X   鄭汝諧

You Mao, Shuichu Title Author Date Juan tang shumu

X

X

X X

X

X X X

Chen Zhensun, Zhizhai shulu jieti

452

Subtotal of other collected reports

X

X

X 11

1 2 2

Shi Yan lu (使燕錄) Yu Rong 余嶸 1211 1 Subtotal for the Jin dynasty 2 2 Hua rong luwei xin lu Su Song 蘇頌 1081 229+5 X (華戎魯衞信録) Sanchao bei meng huibian Xu Mengxin 1194 250 X (三朝北盟㑹編)   徐夢莘 Bei meng ji bu (北盟集補) Xu Mengxin? 1190s–1207 50? X   徐夢莘

453

X

2 X X

X

X

Chen Zhensun, Zhizhai shulu jieti

Xi Xia xuzhi (西夏須知) Liu Wenrun 劉温潤 1 X X Xi Xia zazhi (西夏雜志) X Xiaguo shuyao (夏國樞要) Sun Xun 孫巽 2 X Yuanyou fen jiang lu You Shixiong 游師雄 3 (元祐分疆錄) Subtotal for the Xia dynasty 2 2 Xiongnu xuzhi (匈奴須知) Tian Wei 田緯 1 X X Bian ting xuzhi (邊廷須知) Chen Fang 陳昉 1 Qidan xuzhi (契丹須知) 1 X Beiting zaji (北廷雜記) Zhao Zhizhong 趙志忠 10 X Yinshan zalu (隂山雜録) anonymous; 16   Zhao Zhizhong 趙志忠? Qidan lu (契丹録) anonymous; 1   Zhao Zhizhong 趙志忠? Qidan jiyi tongyao 1 X (契丹機宜通要) Qidan shiji (契丹事迹) 4 X Qidan shilu (契丹實録) 1 X Qidan chao xian liwu li X (契丹朝獻禮物例) Qidan zhi (契丹志) X

You Mao, Chao Gongwu, Shuichu Junzhai Title Author Juan tang shumu dushu zhi

Table 6 Xia, Liao, and Jin gazetteers and records in the holdings of three private collectors

454

Qidan huiyao (契丹㑹要) X Yanjing huiyao (燕京㑹要) X Yan bei lu (燕北録) X Yan bei quanjiang dili ji X (燕北全疆地里記) Yan bei zalu (燕北雜録) Wu Gui 武珪 5 Bian jue lu (辨鴂録) anonymous 1 Beibian yishi (北遼遺事) anonymous (Liao) 2 X Jinren wang Liao lu Shi Yuan 史愿 2 (金人亡遼録) Subtotal for the Liao dynasty 11 3 Zheng meng ji (征䝉記) Li Daliang 李大諒 1 X Fan Zhongxiong bei ji Fan Zhongxiong X (范仲熊北記) Bei ji (北記)   范仲熊 X Pingjiang lu (平江録) X Pingyi beizhong lu X (平議北中録) Jinren bei meng lu Wang Zao 汪藻 1; 7 X (金人背盟録) Jinguo xingcheng (金國行程) Wang Zao 汪藻 10 X Jinguo cheng’an xuzhi 1 X (金國承安須知) Jinren jieyao (金人節要) [Jin captive] 1 X Jinguo zhi (金國志) Zhang Di 張棣 2 Jinguo zhi (金國志) anonymous 1 Jinguo jieyao (金國節要) Zhang Hui 張匯 3

(continued)

X X X

6 X

X

X X

455

Table 6 (continued)

6

X

X

Chen Zhensun, Zhizhai shulu jieti

Jinren nanqian lu Zhang Shiyan 1 (金人南遷録)   張師顔; fake? Zaji Jinguo shi X (雜記金國事) Jinguo shixi (金國世系) X Jinguo wenju lu X (金國文具録) Songmo jiwen (松漠記聞) Hong Hao 洪皓 2 X Nüzhen shilu (女真實録) X Subtotal for the Jin dynasty 10 4 Bei zhi (北志) X Beizhong fangyan X (北中方言) Subtotal for unknown sources 2

You Mao, Chao Gongwu, Shuichu Junzhai Title Author Juan tang shumu dushu zhi

456

Total number of titles held

Bi rong ye hua (避戎夜話) Wei cheng zaji (圍城雜記) Nan gui lu (南歸録) Chaoye qianyan (朝 野僉言) Bei shou xing lu (北 狩行錄) Jingkang yaolu (靖康 要錄) Bei zheng jishi (北征 紀實) Bei shou yeshi (北狩野史) Jingkang shiyi lu (靖 康拾遺錄) Qixue lu (泣血録) Fan que ji (犯闕記) Guichao lu (歸朝録) 5 2

1

anonymous

Cai Tao 蔡絛

He Lie 何烈

2

3+1

1

Cai Tao 蔡絛

Ding Teqi 丁特起 Fang Guan 方冠 Yang Yaobi 楊堯弼; Yang Zai 楊載

X X

5

X

X

X

4

X X

X

1 1 1

X

Chao Gongwu, Junzhai dushu zhi

1

Shen Guan 沈琯 Xia Shaoceng 夏少曾?

Shi Maoliang 石茂良

You Mao, Shuichu Title Author Juan tang shumu

11

X X X

X

X

X

X

X X

X

X

Chen Zhensun, Zhizhai shulu jieti

Table 7 Histories and memoirs of the loss of the north in the holdings of three private collectors

457

Tang shidao sifan zhi (唐十道四蕃志) Huanghua sida ji (皇華四逹記) Subtotal for the Tang dynasty (Yuanfeng) jiuyu zhi ([元豐]九域志) Yudi guangji (輿地廣記) Taiping huanyu zhi (太平寰宇志) Tujing (圖經)

Yuanhe junxian zhi (元和郡縣志) Shidao zhi (十道志)

Li Fang 李昉; Li Zonge   李宗諤

Wang Cun   王存 Ouyang Min?   歐陽忞? Yue Shi 樂史

13

Liang Zaiyan   梁載言 Liang Zaiyan   梁載言 Jia Dan 賈耽

X X

38 200 98; 77 (incomplete)

X

3

X

X

X

10

10

Tang

Northern Song Northern Song Northern Song Northern Song

10

Tang

Tang

40

Li Jifu 李吉甫 Tang

You Mao, Shuichu Title Author Dynasty Juan tang shumu

X

X

X

X

1

X

Chao Gongwu, Junzhai dushu zhi

X

X

X

2

X

X

Chen Zhensun, Zhizhai shulu jieti

Table 8 Comprehensive gazetteers in the holdings of four private collectors

X

0

Zhou Yinghe, Jingding Jiankang zhi

458

Zhifang jiyao anonymous Northern (職方機要) Song Subtotal for the Northern Song period Fangyu shenglan Zhu Mu 祝穆 Southern (方輿勝覽) Song Yu gong jiangli guangji Yi Fu 易祓 Southern (禹貢疆理廣記) Song Wu Xie 吳澥 Southern Lidai jiangyu zhi (歴代疆域志) Song Yudi jisheng (輿地紀 Wang Xiang- Southern 勝)  zhi 王象之 Song Huangchao fangyu zhi Wang Xixian Southern (皇朝方域志)   王希先 Song Subtotal for the Southern Song period Kunyuan lu (坤元録) Wang Tai 王㤗 Wei Diyu ji (地輿記) Gu Yewang Liang   顧野王 Total 9

8

8

3

1 X X

10

2

X

X

200

X

3

200

X

5 X

X

3

X

10

6

43+7+20

40

1

0

1

459

Liang Zhe zhuanyun xuzhi (兩浙轉運須知) Jie yan xuzhi (解鹽須知)

Bureaucracy Chaojiyuan xuzhi (朝集院須知) Zhihe fayun cha yan xuzhi (至和發運 茶鹽須知) Zhi yu xuzhi (治獄須知) Fujian daozei xuzhi (福建盗賊須知) Xuzhi guojing (須知國鏡) (Tang) Pingzhi cang xuzhi bei (平止倉須知碑) Xuzhi (須知) (local) X X

1

1

X

?

X

X

X

X

X

X

Other

XX (multiple editions)

Chao Gongwu, Chen Zhensun, Zhou Yinghe, Zheng Qiao, Junzhai Zhizhai Jingding Tong zhi dushu zhi shulu jieti Jiankang zhi 通志

(1)

2

?

1

?

1

Song shi, You Mao, Yiwen zhi Suichu tang Titles Juan 宋史藝文志 shumu

Table 9A Thematic overview of xuzhi (“what you need to know”) titles listed in major Song catalogs and bibliographies

460

Chizhou Yongfeng qian jian xuzhi (池州 永豐錢監須知) Sitian jian xuzhi (司天監須知) Qiji xuzhi (騏驥須知) Wukuai xuzhi (吳會須知) Jingling gong xuzhi (景靈宫湏知) Renzong shanling ­xuzhi (仁宗山陵湏知) Chaotang xuzhi (朝堂湏知) Neiyi ku xuzhi (内衣庫湏知) Ranyuan xuzhi (染院湏知) Neizang ku xuzhi (內藏庫湏知) Guoyong xuzhi (國用須知) (central) Subtotal 11

X

1

2

X

1

1

X

1

2

X

1

2

X

1

(continued)

XX (multiple editions) 3+

X

1

(1)

X

1

X

X

1

5

X

1

461

Border affairs Xiongnu xuzhi (匈奴須知) (Qidan) Qidan xuzhi (契丹須知) Lu ting xuzhi (虜廷 須知) (same as Bei­ ting xuzhi 北廷須知) Beiting xuzhi (北庭 須知) (same as Lu ting xuzhi 虜廷須知) Liao ting xuzhi (遼庭須知) Chongxiu Liao ting xuzhi (重修遼庭須知) Bei bi xuzhi (北鄙須知) Xi Xia xuzhi (西夏須知) X

1

X

X

1 X

X

X

X

1

X

X

X

X

X

X

Chao Gongwu, Chen Zhensun, Zhou Yinghe, Zheng Qiao, Junzhai Zhizhai Jingding Tong zhi dushu zhi shulu jieti Jiankang zhi 通志

Table 9A (continued)

1

2

X

X

1

2

X

1

X

Song shi, You Mao, Yiwen zhi Suichu tang Titles Juan 宋史藝文志 shumu Other

462

History Tang shu xuzhi (唐書須知) (same as below) Tang shu xin li xuzhi (唐書新例須知) Huntian diwang wuyun tu gujin xuzhi (混天帝王五運圖古 今須知) Subtotal

Jinguo xuzhi (金國須知) Jinguo cheng’an xuzhi (金國承安須知) Xuzhi guojing (須知國鏡) (Tang) Jiannan xuzhi (劍南須知) Fujian daozei xuzhi (福建盗賊須知) Xingjun xuzhi (行軍須知) a Subtotal

1

1

1

?

1

X

5

X

10

?

X

2

1

?

1

X

5

X

X

1

X

3

X

1

X

3

6

3

X

X

X

(continued)

463

Elementary education Youxue xuzhi (幼學須知) (same as Chuxue xuzhi 初學須知) Chuxue xuzhi (初學須知) (same as Youxue xuzhi 幼學須知) Tongguan xuzhi (童丱須知) Tongmeng xuzhi (童蒙須知)

Medicine Zhi bing xuzhi (治病須知) Yong yao xuzhi (用藥須知) Shanghan xuzhi (傷寒須知) b Subtotal

X

3

(1)

X

1

X

5

5

(1–2)

1

1

Song shi, You Mao, Yiwen zhi Suichu tang Titles Juan 宋史藝文志 shumu

X

1

X

1

X

Chao Gongwu, Chen Zhensun, Zhou Yinghe, Zheng Qiao, Junzhai Zhizhai Jingding Tong zhi dushu zhi shulu jieti Jiankang zhi 通志

Table 9A (continued)

X

Other

464

?

(4)

1

3

2

11

1

X

3

X

7

4

6

1

1

X

1

X

1

X

1

b

For a brief review of editions and an argument defending a mid-eleventh-century completion date for this work, see Wang Ruozhao, “‘Xingjun xuzhi’ kao.” This item was stored as twenty-six woodblocks; twenty-six sheets would normally be the size of one or two juan. c Taishang jingming yuan buzou zhiju Taixuan dusheng xuzhi is not included because the century of its completion is in doubt. It too originally consisted of one juan. d These items were part of larger collections and may not have been circulating individually.

a

Total (excluding known duplicate titles)

Food Shipin xuzhi (食品須知) d Subtotal

Classical texts Du Yi xuzhi (讀易須知) Subtotal

Religion c Shishi xuzhi (釋氏須知) Danfang xuzhi (丹房須知) Subtotal

Wuxue xuzhi (務學須知) Subtotal

465

Number of titles

Bureaucracy 20+ Border affairs 14 Elementary education 4 History 3 Medicine 3 Religion 2 Food 1 Classical texts 1

Subject area

Table 9B Total number of xuzhi (“what you need to know”) titles listed in table 9A, by subject area

5/11 45.5%

Chen Zhensun, Zhizhai shulu jieti

4–5/7 3/4 3/6 57–71% 75% 50%

You Mao, Chao Gongwu, Song shi, Yiwen zhi Suichu tang shumu Junzhai dushu zhi

Table 9C Proportion of xuzhi (“what you need to know”) titles on border affairs in three private catalogs and the Song shi Yiwen zhi

466

A ppe n di x 2 A Note on Topical Markup

M

y reading of Wang Mingqing’s Waving the Duster and the other notebooks discussed in chapter 7 is based on a systematic entry-by-entry analysis of the entire notebook. Every entry was assigned one main and up to four secondary topics. The main topic was assigned at the level of the entry itself, and secondary topics could be assigned at either the entry level or at the level of specific passages within an entry. As I explained in chapter 7, the topics were user-generated. They were developed organically, based on a reading of the text, and were edited and arranged into a hierarchical taxonomy to facilitate navigation to the appropriate topic and to allow for analyses at different levels of aggregation. For example, we arranged topics under overarching subject fields such as scholarship, politics, religions, private life, and so forth, and then broke these down ­further into units such as history—by dynasty—Five Dynasties and the like. Keywords were designed to be as specific as possible but the level of specificity varies and in part depends on the expertise and interests of the reader who undertook the topical markup. After all entries had been assigned topics in this manner, all topical markup was extracted into a file and arranged according to frequency. It is important to note that higher-level topics in the taxonomy appear at greater frequencies as they are also counted for each lower-level topic that falls within their range. Some further editing and clean-up was necessary in order to remove topical categories that mostly served

468

appendix 2

navigational purposes in the taxonomy. The resulting topic map is shown in figure 7.9. The topical markup is here mainly used in the analysis of topical clusters in the aggregate or, in other words, as an overview of the general interests expressed in the full text of the notebook. It is important to observe that this method of textual analysis also serves other research functions. On the reader’s desktop, or when applied in an online environment for textual analysis, the topical categories can be used as a finding aid by themselves, or in combination with other types of markup (personal names, time period, titles and genres, or structural features of the text such as notebook title, chapter, or entry), to locate passages relevant to topics of interest to the reader. Second, markup can also be used to create contextualization tools showing the frequency of the topics associated with passages under examination. When reading a particular passage located through a search or when browsing through the notebook text, the reader can then get a sense of how prevalent (or how rare) the topic covered in a particular passage is in the notebook as a whole and move on to other relevant passages. Third, topical markup can also help elucidate the structure of the text when topics are visualized on the basis of their sequential appearance in the text. Readers can through such means detect in what parts of the notebook particular topics are concentrated and get a sense of how the author moved from one subject matter to another in an individual installment and across the entire text of the notebook.

Bibliography Primary Sources This list includes all titles published before the twentieth century. The first date reference is to the date of publication of the edition used. Dates in parentheses refer to the first publication or indicate the time of completion. For works included in the Siku quanshu series (SKQS), the 1983 edition by Taiwan Shangwu yinshuguan, Taibei, has been used. For most sources the year of completion and the time of the first publication await further research; in many cases, exact dates cannot be determined. For the collected works of many Song writers, centuries and decades have been specified; when Song printings of their collected writings are documented, this information has been given; the dates of the authors are given in the main body of the text. Bi Yuan 畢沅. Xu zizhi tongjian 續資治通鑒 (1801). Zhongguo guoji edition. Beijing: Zhongguo guoji guangbo chubanshe, 1993. Zhongguo jiben guji ku digital edition. Beijing Erudition Digital Technology Research Center, ed. Zhongguo jiben guji ku 中国基本古籍库. Cao Yin 曹寅 et al., eds. Quan Tang shi 全唐詩. Reprint of Yangzhou shiju edition. Yangzhou: Yangzhou shiju, 1701. Beijing guoxue shidai wenhua chuanbo 北京国学时代文化传播, ed. Guoxue 国学, 2000 (based on the 1701 reprint). Accessed Aug. 1, 2015. http://www .guoxue.com/qts/qts_xml.htm. Chao Gongwu 晁公武. Junzhai dushuzhi jiaozheng 郡齋讀書志校證. Supplement by Zhao Xibian 趙希弁; edited by Sun Meng 孫猛 and Wang Lixiang 王立翔. Shanghai: Shanghai guji chubanshe, 1990; reprint, 2005 (ca. 1151–1240s). Chao Yuezhi 晁說之. Jingyu sheng ji 景迂生集. SKQS (12th c.). Chen Chun 陳淳. Beixi daquan ji 北溪大全集. SKQS (ca. 1220s). Chen Kui 陳騤. Nan Song guan’ge lu, xu lu 南宋館閣錄, 續錄. Beijing: Zhonghua shuju, 1998 (1178). Chen Qiqing 陈耆卿. Chicheng zhi 赤城志 (early 1220s). Song Yuan fangzhi congkan 宋元方志叢刊. Beijing: Zhonghua shuju, 1990 (1261). SKQS. Chen Yuan 陳淵. Motang ji 默堂集. SKQS (ca. 1147).

470 bibliogr aphy Chen Zhensun 陳振孫. Zhizhai shulu jieti 直齋書錄解題. Shanghai: Shanghai guji chubanshe, 1987 (ca. 1249). Cheng Bi 程珌. Mingshui ji 洺水集. SKQS (13th c.). Cheng Dachang 程大昌. Beibian bei dui 北邊備對. CSJC-CB (1191). ———. Yanfan lu 演蕃露. In Ruxue jingwu 儒學警悟. Song printed edition, 1201 (1180). Beijing Library. ———. Yugong lun 禹貢論. Song printed edition, ca. 1177. Beijing Library. Cheng Duanli 程端禮. Dushu fennian richeng 讀書分年日程. Shandong: Shangzhi tang, 1871 (1310s). Cheng Gongxu 程公許. Cangzhou chen fou bian 滄洲塵缶編. SKQS (ca. 1241). Congshu jicheng chubian 叢書集成初編. Shanghai: Shanghai yinshuguan, 1935–37. Series reprint. Beijing: Zhonghua shuju, 1991. Dai Fugu 戴復古. Shiping shi ji 石屏詩集 (1230s–1240s printing). SKQS. SBCK. Du You 杜佑. Tong dian 通典. Beijing: Zhonghua shuju, 1982 (801). Du Zheng 度正. Xingshan tang gao 性善堂稿. SKQS (1210s–1230s). Fan Chengda 范成大, comp. Wujun zhi 吳郡志. Supplement by Wang Taiheng 汪泰亨 et al. Song Yuan fangzhi congkan 宋元方志叢刊. Beijing: Zhonghua shuju, 1990 (1192; 1228 printing). Fan Ye 范曄. Hou Han shu 後漢書. Beijing: Zhonghua shuju, 1965 (1st half 5th c.). Fang Dacong 方大琮. Tie’an ji 鐵庵集. SKQS (13th c. printing). Fang Hui 方回. Tongjiang xu ji 桐江續集. SKQS (13th–early 14th c.). Fang Xuanling 房玄齡. Jin shu 晉書. Beijing: Zhonghua shuju, 1974 (648). Fang Yue 方岳. Qiuya ji 秋崖集. SKQS (13th c. printing). Fei Gun 費袞. Liangxi man zhi 梁谿漫志. SKQS (1192). Feng Qi 馮琦 and Chen Bangzhan 陳邦瞻. Song shi jishi benmo 宋史紀事本末. 3 vols. Beijing: Zhonghua shuju, 1977 (1605). Fu Xuancong 傅璇琮 and Beijing daxue Guwenxian yanjiusuo 北京大學古文獻硏 究所, eds. Quan Song shi 全宋詩. Beijing: Beijing daxue chubanshe, 1991. Fu Xuancong 傅璇琮 et al., eds. Quan Song biji 全宋筆記. Zhengzhou: Da xiang chubanshe, 2003–. Gao Cheng 高承. Shiwu jiyuan 事物紀原. Compiled by Li Guo 李果; edited by Jin Yuan 金圓, Xie Weixin 謝維新, and Xu Peizao 許沛藻. Beijing: Zhonghua shuju, 1989 (ca. 1078–85). Gao Zhu 高翥. Jujian ji 菊澗集. Compiled by Gao Shiqi 高士奇. SKQS (early 13th c.). Gong Yizheng 龔頤正. Jieyin biji 芥隱筆記. CSJC-CB (1201 printing). Guo Ruoxu 郭若虛. Tuhua jianwen zhi 圖畫見聞誌. SKQS (ca. 1075). Han Yu 韓愈. Zhu wengong jiao Changli xiansheng wenji 朱文公校昌黎先生文集. Edited by Zhu Xi 朱熹. SBCK (12th c.)



bibliogr aphy

471

Hong Mai 洪邁. Rongzhai suibi 容齋隨筆. Edited by Kong Fanli 孔凡禮. Beijing: Zhonghua shuju, 2006 (1162 and 1202; 12th and 13th c. printings). Hong Zikui 洪咨夔. Pingzhai ji 平齋集. SKQS (13th c. printing). Hu Su 胡宿. Wengong ji 文恭集. SKQS (late 11th c.). Hua Yue 华岳. Cuiwei beizheng lu qianshuo 翠微北征录浅说. Annotated by Lan Shucheng 兰书臣 and Wu Ziyong 吴子勇. Beijing: Jiefangjun chubanshe, 1992 (ca. 1213). ———. Cuiwei nanzheng lu, beizheng lu he ji 翠微南征录北征录合集. Annotated by Ma Junhua 马君骅. Hefei: Huangshan shushe, 1993. ———. Cuiwei beizheng lu 翠微北征录. In Zhongguo jingdian bingshu 中国经典 兵书, vol. 2. Jinan: Shandong youyi chubanshe, 2002. Huang Gan 黃榦. Mianzhai ji 勉齋集. SKQS (13th c. printing). Ji Zengyun 嵇曾筠 et al. Zhejiang tongzhi 浙江通志. SKQS (1736). Jiang Shaoyu 江少虞. Songchao shishi leiyuan 宋朝事實類苑 (ca. 1145). Shanghai guji edition. Shanghai, 1981. SKQS. ( Jingxuan) Huang Song cexue shengchi (精選) 皇宋策學繩尺. Qing copy of a Song edition (mid-13th c.). Beijing Library. Kong Yingda 孔穎達, ed. Chunqiu Zuo zhuan zheng yi 春秋左傳正義. In Shisan jing zhushu 十三經注疏. Beijing: Beijing daxue chubanshe, 2000 (ca. 574– 648). Li E 厲鶚. Song shi jishi 宋詩紀事. Shanghai: Guji chubanshe, 1983 (1736). Li Ruqian 李如篪. Dongyuan congshuo 東園叢説. CSJC-CB (1132). Li Tao 李燾. Xu Zizhi tongjian changbian 續資治通鑑長編 (1183). SKQS. Hanji quanwen ziliaoku digital edition. Zhongyang yanjiuyuan zixun kexue yanjiusuo 中央硏究院資訊科學研究所, ed. Hanji quanwen ziliaoku 漢籍全 文資料庫. Taibei: Zhongyang yanjiuyuan jisuan zhongxin, 2000. Li Xinchuan 李心傳. Jianyan yilai xinian yao lu 建炎以來系年要錄 (ca. 1208). SKQS. Hanji quanwen ziliaoku digital edition. Zhongyang yanjiuyuan zixun kexue yanjiusuo 中央硏究院資訊科學研究所, ed. Hanji quanwen ziliaoku 漢籍全 文資料庫. Taibei: Zhongyang yanjiuyuan jisuan zhongxin, 2000. Liji zhushu 禮記注疏. In Chongkan Songben shisanjing zhushu fu jiaokanji 重刋宋本十三經注疏附校 勘記 (Nanchang: Nanchang fuxue, 1815). Hanji quanwen ziliaoku digital edition. Zhongyang yanjiuyuan zixun kexue yanjiusuo 中央硏究院資訊科學研究所, ed. Hanji quanwen ziliaoku 漢籍全 文資料庫. Taibei: Zhongyang yanjiuyuan jisuan zhongxin, 2000. Lin Jiong 林駉. Huang jian jian yao 皇鑑箋要. 1210s. Beijing University Library. Liu Changshi 劉昌詩. Lupu biji 蘆浦筆記. Beijing: Zhonghua shuju, 1986 (ca. 1213).

472 bibliogr aphy Liu Dake 劉達可. Bishui qunying daiwen huiyuan (xuanyao) 璧水群英待問會元 (選要) (ca. 1245). Naikaku bunko edition. Tokyo, 1509. Nanjing Library edition. Ming printing. Siku quanshu cunmu congshu 四庫 全書存目叢書, Zibu, vol. 168. Jinan: Qi Lu, 1997. National Central Library, Taibei. Early Ming edition. Liu Kezhuang 劉克莊. Houcun ji 後村集. SKQS (1270s). ———. Houcun xiansheng da quanji 後村先生大全集. SBCK (ca. 1270s printing). Liu Shiju 劉時舉. Xu Song biannian zizhi tongjian 續宋編年資治通鑑. SKQS (ca. 1280?). Liu Xu 劉昫. Jiu Tang shu 舊唐書. Beijing: Zhonghua shuju, 1975 (945). Lou Fang 樓昉. (Yuzhai xiansheng biaozhu) Chonggu wen jue (迂齋先生標註)崇古 文訣. Beijing Library edition (1220s printing). SKQS. Lu You 陸游. Jiannan shi gao 劎南詩稾. SKQS (1180s–1220s printings). ———. Weinan wenji 渭南文集. Beijing: Zhonghua shuju, 1976 (1220 printing). Lu Youren 陸友仁. Yanbei zazhi 硏北雜志. SKQS (early 14th c.). Lü Zuqian 呂祖謙. Donglai ji—bieji, waiji, fulu 東萊集—別集、外集、附錄. SKQS (1204). ———. Lü Zuqian quanji 呂祖謙全集. Edited by Huang Linggeng 黃靈庚. Hangzhou: Zhejiang guji chubanshe, 2005. ———. Song wen jian 宋文鑑 (1179). Zhonghua shuju edition. 3 vols. Beijing, 1992. SKQS. Ma Duanlin 馬端臨. Wenxian tongkao 文獻通考. 2 vols. Beijing: Zhonghua shuju, 1986 (late 13th–early 14th c.). Ni Pu 倪朴. Ni Shiling shu 倪石陵遺書. SKQS (late 12th c.; 1526 printing). Ouyang Xiu 歐陽修. Ouyang Yongshu ji 歐陽永叔集. 3 vols. Shanghai: Shangwu yinshuguan, 1933 (late 11th c.). Pan Zimu 潘自牧. Jizuan yuanhai 記纂淵海. SKQS (ca. 1209). Pang Yuanying 龐元英. Wenchang zalu 文昌雜錄. Beijing: Zhonghua shuju, 1958 (1082–85). Peng Guinian 彭龜年. Zhitang ji 止堂集. CSJC-CB (early 13th c.). Qian Zeng 錢曾. Shugu tang Song ban shumu 述古堂宋板書目. In Song Yuan ban shu­mu tiba ji kan 宋元版書目題跋輯刊. Beijing: Beijing tushuguan chubanshe, 2003. Qingyuan tiaofa shilei 慶元條法事類. Edited by Yang Yifan 楊一凡 and Tian Tao 田濤. Haerbin: Heilongjiang renmin chubanshe, 2002 (ca. 1202).



bibliogr aphy

473

Shao Bowen 邵伯溫. Shao shi wenjian lu 邵氏聞見錄 (1132). SKQS. Zhonghua shuju edition. Beijing, 1983. Shen Gua 沈括. Mengxi bitan 夢溪筆談 (1086–93). In Yuhaitang ying Song Yuan ben congshu 玉海堂景宋元本叢書. [China]: Guichi Liu shi, 1915. ———. Mengxi bitan jiaozheng 夢溪筆談校證. Annotated by Hu Daojing 胡道靜. 2 vols. Shanghai: Shanghai guji chubanshe, 1987. Sheng Song qianjia mingxian biao qi 聖宋千家名賢表啓. Song printing (early 13th c.). Beijing Library. Shi Su 施宿, comp. Jiatai Guiji zhi 嘉泰會稽志. Song Yuan fangzhi congkan 宋元 方志叢刊. Beijing: Zhonghua shuju, 1990 (1202). Shi xiansheng aolun zhu (Qianji, houji, xuji) 十先生奧論註 (前、後、續集). SKQS (13th c.). Sibu congkan 四部叢刊. Shanghai: Shangwu yinshuguan, 1922–36. Sibu congkan xubian 四部叢刊續編. Shanghai: Shangwu yinshuguan, 1934. Sikong Tu 司空圖. Sikong Biaosheng wenji 司空表聖文集. SKQS (10th c.; Song printing). Siku quanshu 四庫全書. Reprint. Taibei: Shangwu yinshuguan, 1983. Sima Guang 司馬光. Sushui jiwen 涑水記聞. Beijing: Zhonghua shuju, 1989 (late 11th c.). ———. Zhuanjia ji 傳家集. SKQS (12th c. printing). Sima Qian 司馬遷. Shi ji 史記. Beijing: Zhonghua shuju, 1959 (90 BCE). Song ben Lidai dili zhizhang tu 宋本歷代地理指掌圖. Shanghai: Shanghai guji chubanshe, 1989 (ca. 1130). Song Lian 宋濂. Puyang renwu ji 浦陽人物記. SKQS (14th c.). Song shi quan wen 宋史全文. SKQS (14th c.). Songji sanchao zhengyao 宋季三朝政要. SKQS (after 1280, late 13th c.). Su Che 蘇轍. Luancheng ji 欒城集. Edited by Zeng Zaozhuang 曾棗莊 and Ma Defu 馬德富. Shanghai: Shanghai guji chubanshe, 1987 (1090s–1100s). Su Shi 蘇軾. Su Shi wenji 蘇軾文集. Zhonghua shuju edition. 6 vols. Beijing, 1986 (12th c. printing). Yuelu shushe edition. Edited by Gu Zhichuan 顧之川. Changsha, 2000. Su Song 蘇頌. Su Weigong wenji 蘇魏公文集. SKQS (early 12th c.; late 12th c. printing). Sun Yingshi 孫應時. Zhuohu ji 燭湖集. SKQS (early 13th c. printing). Tian Rucheng 田汝成. Xihu youlan zhi yu 西湖遊覽志餘. SKQS (1547 printing). Tian Xi 田錫. Xianping ji 咸平集. SKQS (early 11th c.). Tuotuo 脫脫, ed. Songshi 宋史. Beijing: Zhonghua shuju, 1977 (1345). Wang Fuzhi 王夫之. Song lun 宋論 Taibei: Liren shuju, 1985 (late 17th c.). Wang Mai 王邁. Quxuan ji 臞軒集. SKQS (13th c.). Wang Mingqing 王明清. Huizhu lu 揮麈錄 (1166–97). SBCK. Zhonghua shuju edition. Beijing, 1961.

474 bibliogr aphy ———. Tou xia lu 投轄錄. Shanghai: Shanghai guji chubanshe, 1991 (1196–97). ———. Yuzhao xin zhi 玉照新志. Shanghai: Shanghai guji chubanshe, 1991 (1198). Wang Yinglin 王應麟. Yuhai 玉海. SKQS (1250s). Wang Zao 汪藻, attrib. Jingkang yaolu 靖康要錄. SKQS (ca. 1165). ———. Jingkang yaolu jianzhu 靖康要录笺注. Annotated by Wang Zhiyong 王智勇. Chengdu: Sichuan da xue chubanshe, 2008. Wang Zhiwang 王之望. Han bin ji 漢濱集. SKQS (late 12th c.). Wei Liaoweng 魏了翁. Haoshan ji 鶴山集. SKQS (1259). Wenxiang 文珦. Qianshan ji 潛山集. SKQS (1211). Wu Cheng 吳澄. Wu Wenzheng ji 吳文正公集. SKQS (14th c. printing). Wu Shidao 吳師道. Jing xiang lu 敬鄉錄. In Xu Jinhua congshu 續金華叢書. [China]: Yongkang Hu shi Meng xuan lou, 1924 (14th c.). Wu Yong 吳泳. Haolin ji 鶴林集. SKQS (13th c.). Wu Zeng 吳曾. Nenggai zhai man lu 能改齋漫錄. Shanghai: Shanghai guji chubanshe, 1979 (ca. 1127–60). Xie Caibo 謝采伯. Mizhai biji 宻齋筆記. CSJC-CB (early 13th c.). Xie Daocheng 謝道承, comp. Fujian tongzhi 福建通志. Edited by Hao Yulin 郝玉麟. SKQS (1737). Xie Jin 解縉 et al., comp. Yongle dadian 永樂大典. Beijing: Zhonghua shuju, 1986 (1404). Xie Weixin 謝維新. Gujin hebi shilei beiyao 古今合璧事類備要. Supplement by Yu Zai 虞載; edited by Xia Xiang 夏相. 80 vols. [China]: Xia Xiang, 1552 (1257). (Xin bian) Ruxue shuyao (新編) 儒學樞要. 13th c. Shanghai Library. Xiong Ke 熊克. Zhongxing xiaoji 中興小紀. SKQS (1185). Xu Mengxin 徐夢莘. Sanchao beimeng huibian 三朝北盟會編. 3 vols. SKQS (1196). Xu Song 徐松, ed. Song huiyao jigao 宋會要輯稿. 8 vols. Beijing: Zhong­hua shuju, 1957 (ca. 1809). ———. Song huiyao 宋會要 [digital edition]. Edited by Zhongyang yanjiuyuan 中央硏究院 and Harvard University. Taibei: Zhongyang yanjiuyuan jisuan zhongxin, 2003. Xu Wenjing 徐文靖. Yugong huijian 禹貢會箋. SKQS (1753). Xu Yinglong 許應龍. Dongjian ji 東澗集. SKQS (13th c.). Xue Jixuan 薛季宣. Langyu ji 浪語集. SKQS (late 12th c.). Yang Fang 陽枋. Zixi ji 字溪集. SKQS (13th c.). Yang Wanli 楊萬里. Chengzhai ji 誠齋集. SKQS (1170s–80s). Yao Mian 姚勉. Xuepo sheren ji 雪坡舍人集. Song ji zhenben congkan 宋集珍本 叢刋. Beijing: Xianzhuang shuju, 2004 (ca. 1216). Ye Tinggui 葉廷珪. Hailu suishi 海錄碎事. SKQS (mid-12th c.). You Mao 尤袤. Suichu tang shumu 遂初堂書目. SKQS (ca. 1190). Yuan Xie 袁燮. Jiezhai ji 絜齋集. SKQS (1228 printing).



bibliogr aphy

475

Yue Ke 岳珂. Kuitan lu 愧郯錄. SKQS (ca. 1214). ———. Ying shi 桯史. Beijing: Zhonghua shuju, 1981 (1214). Zeng Jian 曾堅. Da ce mijue 答策秘訣. Edited by Liu Jinwen 劉錦文. Yuan edition (mid-13th c.). Palace Museum Library, Taibei. Zeng Zao 曾慥. Leishuo 類說. Song-Ming edition (1136). Beijing Library. Zeng Zaozhuang 曾棗莊 and Liu Lin 劉琳, eds. Quan Song wen 全宋文. Shanghai: Shanghai cishu chubanshe, 2006. Zengru mingru jiangyi huang Song Zhongxing liangchao shengzheng 增入名儒講義 皇宋中興兩朝聖政. Southern Song printed edition. National Central Library, Taipei. Zhang Duanyi 張端義. Gui’er ji 貴耳集. SKQS (1246). Zhang Fangping 張方平. Lequan ji 樂全集. SKQS (1080s; 12th c. printing). Zhang Jinwu 張金吾. Airi jinglu cangshu zhi 愛日精廬藏書志. Qingren shumu tiba congkan. Beijing: Zhonghua shuju, 1990 (1826). Zhang Ruyu 章如愚. Qunshu kaosuo 群書考索 (13th c. printing). SKQS. Reprint of a Yuan edition. Beijing: Zhonghua shuju, 1992. Zhang Shinan 張世南. Youhuan jiwen 游宦紀聞. Beijing: Zhonghua shuju, 1981 (early 13th c.). Zhao Buhui 趙不悔, comp. Xin’an zhi 新安志. Song Yuan fangzhi congkan 宋元方 志叢刊. Beijing: Zhonghua shuju, 1990 (1175). Zhao Ruyu 趙如愚, comp. Song mingchen zouyi 宋名臣奏議 [alt. title: Guochao zhu­ chen zouyi 國朝諸臣奏議]. SKQS (1186). Zhao Sheng 趙升. Chaoye lei yao 朝野類要. SKQS (1236). Zhao Yanwei 趙彥衛. Yunlu manchao 雲麓漫鈔. Beijing: Zhonghua shuju, 1996 (1206). Zheng Qiao 鄭樵. Tongzhi 通志. SKQS (1149). Zheng Xie 鄭獬. Yunxi ji 鄖溪集. SKQS (11th c., 12th c. printing). Zhou Bida 周必大. Wenzhong ji 文忠集. SKQS (1206 printing). Zhou Linzhi 周麟之. Hailing ji 海陵集. SKQS (late 12th c.). Zhou Mi 周密. Qidong yeyu 齊東野語. Beijing: Zhonghua shuju, 1983 (1291). ———. Wulin jiushi 武林舊事. SKQS (ca. 1280). Zhou Yinghe 周應合. Jingding Jiankang zhi 景定建康志 (1261). Song Yuan fangzhi congkan edition. Beijing: Zhonghua shuju, 1990. SKQS. Zhu Mu 祝穆. (Xin bian) Gujin shiwen leiju (新編) 古今事文類聚. Edited by Fu Dayong 富大用. 3 vols. Beijing: Shumu wenxian chubanshe, 1991 (1246). Zhu Xi 朱熹. Zhu Xi ji 朱熹集. 10 vols. Chengdu: Sichuan jiaoyu chubanshe, 1996 (early 13th c.). ———. Zhuzi yulei 朱子語類. Edited by Li Jingde 黎靖德. Beijing: Zhonghua shuju, 1986 (1270).

476 bibliogr aphy

Secondary Sources Abramson, Marc Samuel. Ethnic Identity in Tang China. Philadelphia: University of Pennsylvania Press, 2008. Adler, Joseph A. “Chu Hsi and Divination.” In Sung Dynasty Uses of the I Ching, edited by Kidder Smith Jr. et al., 169–205. Princeton: Princeton University Press, 1990. Akerman, James R. The Imperial Map: Cartography and the Mastery of Empire. Chicago: University of Chicago Press, 2009. An Pingqiu 安平秋 and Zhang Peiheng 章培恒. Zhongguo jinshu da guan 中国禁 书大观. Shanghai: Shanghai wenhua chubanshe, 1990. An Ruixuan 安芮璇. “Songren biji yanjiu—yi suibi zaji wei zhongxin” 宋人笔记研 究—以随笔杂记为中心. Ph.D. diss., Fudan University, 2005. Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso, 1983. Balazs, Étienne, and Yves Hervouet, eds. A Sung Bibliography (Bibliographie des Sung). Hong Kong: Chinese University Press, 1978. Barkey, Karen. Empire of Difference: The Ottomans in Comparative Perspective. Cambridge: Cambridge University Press, 2008. Bayly, C. A. Empire and Information: Intelligence Gathering and Social Communication in India, 1780–1870. Cambridge: Cambridge University Press, 1996. Berry, David. “Digital Humanities: First, Second and Third Wave.” In Stunlaw: A Critical Review of Politics, Arts and Technology, January 14, 2011. Accessed May 22, 2012. http://stunlaw.blogspot.com/2011/01/digital-humanities-first-second -and.html. Berry, Mary Elizabeth. Japan in Print: Information and Nation in the Early Modern Period. Berkeley: University of California Press, 2006. Biot, Édouard, Stanislas Julien, and Jean-Baptiste Biot. Le Tcheou-Li ou Rites des Tcheou. 3 vols. Paris: Imprimerie nationale, 1851. Birkerts, Sven. The Gutenberg Elegies: The Fate of Reading in an Electronic Age. Boston: Faber and Faber, 1994. Black, Jeremy. Maps and History: Constructing Images of the Past. New Haven: Yale University Press, 1997. Blair, Ann. Too Much to Know: Managing Scholarly Information before the Modern Age. New Haven: Yale University Press, 2010. Bol, Peter K. “A Literati Miscellany and Sung Intellectual History: The Case of Chang Lei’s Ming-tao tsa-chih.” Journal of Song Yuan Studies 25 (1995): 121–52. ———. “Reading Su Shi in Southern Song Wuzhou.” East Asian Library Journal 8, no. 2 (1998): 69–102. ———. “The Rise of Local History: History, Geography, and Culture in Southern Song and Yuan Wuzhou.” Harvard Journal of Asiatic Studies 61, no. 3 (2001): 37–76. ———. This Culture of Ours: Intellectual Transitions in T’ang and Sung China. Stanford: Stanford University Press, 1992.



bibliogr aphy

477

———. “Zhang Ruyu, the Qunshu kaosuo, and Diversity in Intellectual Culture— Evidence from Dongyang County in Wuzhou.” In Qingzhu Deng Guangming jiaoshou jiushi huadan lunwen ji 庆祝邓广铭教授九十华诞论文集, edited by Tian Yuqing 田馀庆, 644–73. Shijiazhuang: Hebei jiaoyu chubanshe, 1997. Bol, Peter K., et al. China Biographical Database Project (CBDB) (2004–). Accessed May 22, 2012. http://isites.harvard.edu/icb/icb.do?keyword=k16229&pageid= icb.page12970. ———, et al. “China Historical GIS (2001–).” Accessed May 22, 2012. http://www .fas.harvard.edu/~chgis/. ———, et al. “Hartwell China History Project GIS (1992–2001).” Accessed May 22, 2012. http://www.fas.harvard.edu/~chgis/. Boodberg, Peter Alexis. The Art of War in Ancient China: A Study Based upon the Dialogues of Li, Duke of Wei. Berkeley: University of California, 1930. Bossler, Beverly. Powerful Relations: Kinship, Status, and the State in Sung China (960– 1279). Cambridge, MA: Harvard University Asia Center, 1998. Bridenbaugh, Carl. “The Great Mutation.” American Historical Review 68, no. 2 (1962): 315–31. Burbank, Jane, and Frederick Cooper. Empires in World History: Power and the Politics of Difference. Princeton: Princeton University Press, 2010. Burke, Peter. The Art of Conversation. Ithaca: Cornell University Press, 1993. ———. A Social History of Knowledge: From Gutenberg to Diderot. Cambridge: Polity, 2000. Burkus-Chasson, Anne. “Visual Hermeneutics and the Act of Turning the Leaf— A Genealogy of Liu Yuan’s Lingyan Ge.” In Printing and Book Culture in Late Imperial China, edited by Cynthia Joanne Brokaw and Chow Kai-wing, 371–416. Berkeley: University of California Press, 2005. Burt, Ronald S. “Bandwidth and Echo: Trust, Information, and Gossip in Social Networks.” In Networks and Markets, edited by J. E. Rauch and A. Casella, 30–74. New York: Russell Sage Foundation, 2001. Cai Chongbang 蔡崇榜. Songdai xiushi zhidu yanjiu 宋代修史制度研究. Taibei: Wenjin chubanshe, 1991. Cao Wanru 曹婉如 et al. Zhongguo gudai ditu ji: Zhanguo–Yuan 中国古代地图集: 战国–元. Beijing: Wenwu chubanshe, 1990. Cartier, Carolyn. “Origins and Evolution of a Geographical Idea: The Macroregion in China.” Modern China 28, no. 1 (2002): 79–142. Cha, Javier. “Data-Intensive Methods and the Logic of Scientific History.” Paper presented at the Annual Conference of the Association for Asian Studies, H ­ awaii, March 31–April 3, 2011. Chaffee, John W. “Sung Biographies, Supplementary Biography No. 2: Li HsinCh’uan.” Journal of Sung-Yuan Studies 24 (1994): 205–15. ———. The Thorny Gates of Learning in Sung China: A Social History of the Examinations. Albany: State University of New York Press, 1995 (1985). Chan, Hok-lam. Control of Publishing in China, Past and Present. Canberra: Australian National University Press, 1983.

478 bibliogr aphy Chavannes, Edouard. Les deux plus anciens spécimens de la cartographie chinoise. Hanoi: Schneider, 1903. Chen Aiping 陈爱平. “Nan Song dui Liuchao nan-bei junshi duiyi jingyan de lilun yanjiu” 南宋对六朝南北军事对峙经验的理论研究. Shayang shifan gaodeng zhuanke xuexiao xuebao 沙洋师范高等专科学校学报, no. 3 (2006): 29–31. Chen, Jack Wei. The Poetics of Sovereignty: On Emperor Taizong of the Tang Dynasty. Cambridge, MA: Harvard University Asia Center, 2010. Chen, Jack Wei, and David Schaberg, eds. Idle Talk: Gossip and Anecdote in Traditional China. Berkeley: University of California Press, 2013. Chen Lai 陈来. Zhuzi shuxin biannian kaozheng 朱子书信编年考证. Shanghai: Shanghai renmin chubanshe, 1989. Chen Ming 陈铭. Ouyang Xiu zhuan 欧阳修传. Guangzhou: Guangdong gaodeng jiaoyu chubanshe, 1998. Cheng Minsheng 程民生. Songdai wujia yanjiu 宋代物价研究. Beijing: Renmin chubanshe, 2008. Cheng Zhangcan 程章灿. Liu Kezhuang nianpu 刘克灿年谱. Guiyang: Guizhou renmin chubanshe, 1993. Chia, Lucille. Printing for Profit: The Commercial Publishers of Jianyang, Fujian (11th–17th Centuries). Harvard-Yenching Institute Monograph Series. Cambridge, MA: Harvard University Asia Center, 2002. Chinnery, Colin. “Bookbinding.” Accessed Feburary 15, 2007. http://idp.bl.uk /education/bookbinding/bookbinding.a4d. Chow, Kai-wing. Publishing, Culture, and Power in Early Modern China. Stanford: Stanford University Press, 2004. Chu, Ping-tzu 祝平次. “Beyond the Personal.” Invited lecture at Institute for Chinese Studies, University of Oxford, Oxford, UK, May 10, 2012. Clanchy, M. T. From Memory to Written Record: England, 1066–1307. Oxford: Basil Blackwell, 1979; reprint, 1993. Cohen, Daniel J., and Roy Rosenzweig. “Introduction: Promises and Perils of Digital History.” In Digital History: A Guide to Gathering, Preserving, and Presenting the Past on the Web, edited by Daniel J. Cohen and Roy Rosenzweig, 1–17. Philadelphia: University of Pennsylvania Press, 2006. Davis, Richard L. “The Reigns of Kuang-tsung (1189–1194) and Ning-tsung (1194– 1224).” In Cambridge History of China, vol. 5, pt. 1, The Sung Dynasty and Its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 756–838. Cambridge: Cambridge University Press, 2009. de Pee, Christian. “Cycles of Cathay: Sinology, Philology, and Histories of the Song Dynasty (960–1279) in the United States.” Fragments: Interdisciplinary Approaches to the Study of Ancient and Medieval Pasts 2 (2012): 35–67. de Vivo, Filippo. Information and Communication in Venice: Rethinking Early Modern Politics. Oxford: Oxford University Press, 2007. De Weerdt, Hilde. “Aspects of Song Intellectual Life: A Preliminary Inquiry into Some Southern Song Encyclopedias.” Papers on Chinese History 3 (1994): 1–27. ———. “Biji in Print Database.” In Communication and Empire: Chinese Empires



bibliogr aphy

479

in Comparative Perspective, 2011. Accessed May 22, 2012. http://www.chinese -empires.eu/analysis/database/. ———. “Byways in the Imperial Chinese Information Order: The Dissemination and Commercial Publication of State Documents.” Harvard Journal of Asiatic Studies 66, no. 1 (2006): 145–88. ———. “Chinese Cities, 600–1300.” In The Oxford Handbook of Cities in World History, edited by Peter Clark, 292–308. Oxford: Oxford University Press, 2013. ———. Competition over Content: Negotiating Standards for the Civil Service Examinations in Imperial China (1127–1279). Harvard East Asian Monographs. Cambridge, MA: Harvard University Press, 2007. ———. “Continuities between Scribal and Print Publishing in Twelfth-Century Song China—The Case of Wang Mingqing’s Serialized Notebooks.” East Asian Publishing and Society 6, no. 1 (2016): 54–83. ———. “ ‘Court Gazettes’ and ‘Short Reports’: Official Views and Unofficial Readings of Court News.” Hanxue yanjiu 漢學研究 27, no. 2 (2009): 167–200. ———. “The Cultural Logics of Map Reading: Text, Time and Space in Printed Maps of the Song Empire.” In The Cultural Logics of Map Reading: Text, Time, and Space in Printed Maps of the Song Empire, edited by Lucille Chia and Hilde De Weerdt, 239–70. Leiden: Brill, 2011. ———. “Digital Interpretations.” Communication and Empire: Chinese Empires in Comparative Perspective, Feb. 5, 2014. Accessed April 1, 2014. http://chinese-empires .eu/blog/digital-interpretations/. ———. “The Discourse of Loss in Private and Court Book Collecting in Imperial China.” Library Trends 55, no. 4 (2007): 404–20. ———. “History: Song Studies.” In A Scholarly Review of Chinese Studies in North America, edited by Zhang Haihui, 23–53. Ann Arbor, MI: Association for Asian Studies, 2013. ———. “Information, Territory, and Networks—Extra Materials,” April 18, 2014. Accessed April 1, 2015. Available from https://www.academia.edu/attachments /33510144/download_file. ———. “Isn’t the Siku quanshu Enough? Reflections on the Impact of New Digital Tools for Classical Chinese.” Communication and Empire: Chinese Empires in Comparative Perspective, Feb. 20, 2014. Accessed April 1, 2014. http://chinese -empires​.eu/blog/isnt-the-siku-quanshu-enough​-ref lections​- on-the​-impact​ -of-new-digital-tools-for-classical-chinese/. ———. “Mapping Communication from Mingzhou: Networks of Correspondence.” Paper presented at the conference “The Prosopography of Middle Period China: Using the Chinese Biographical Database,” University of Warwick, UK, Dec. 14, 2007. http://www.humanities.uci.edu/eastasian/SungYuan/Downloads /Warwick/DeWeerdt.pdf. ———. “Mapping Communication from Mingzhou: Networks of Correspondence.” Journal of Sung-Yuan Studies 38 (2009): 184–86. ———. “Maps and Memory: Readings of Cartography in Twelfth- and ThirteenthCentury Song China.” Imago Mundi: International Journal for the History of Cartography 61, no. 2 (2009): 145–67.

480 bibliogr aphy ———. “The Production and Circulation of ‘Written Notes’ (biji).” In Imprimer sans profit? Le livre non commercial dans la Chine impériale, edited by Michela Bussotti and Jean-Pierre Drège, 19–48. Geneva: Droz, 2015. ———. “The Production and Circulation of ‘Written Notes’ (biji)—Tables and Figures.” Oxford University Research Archive (2010). Accessed May 22, 2012. http://ora.ox.ac.uk/objects/uuid:ea1a17a3%E2%80%932e27%E2%80%934a13 -b8d8-ed90692ff893. ———. “Reading Instructions for an Early Printed Map of the Chinese Empire: A Digital Reconstruction.” Oxford University Research Archive (2007–9). Accessed May 22, 2012. http://ora.ouls.ox.ac.uk/objects/uuid:c0c281cd-4c2d-46ea -94b0-f7758cf0c3e6. ———. “Recent Trends in American Research in Song Dynasty History: Local Religion and Political Culture.” In Taiwan Song shi yanjiu wang 臺灣宋史研究網, Feb. 1, 2006. Accessed May 23, 2012. http://www.ihp.sinica.edu.tw/~twsung /breview/subpage/02/files/Recent_Trends_in_American_Research_in_Song _Dynasty_History.pdf. ———. “Regional Descriptions: Administrative and Scholarly Traditions.” In Treasures of the Yenching: Seventy-Fifth Anniversary of the Harvard-Yenching Library: Exhibition Catalogue, edited by Harvard-Yenching Library, Patrick Hanan, and Mikael S. Adolphson, 122–53. Cambridge, MA: Harvard-Yenching Library Harvard University, 2003. ———. “Shmuel N. Eisenstadt and the Comparative Political History of PreEighteenth-Century Empires.” Asian Review of World Histories 4, no. 1 (2016), forthcoming. ———. “Transformative Journeys: Travel and Culture in Song China by Cong Ellen Zhang (Review).” Bulletin of the School of Oriental and African Studies 74, no. 3 (2011): 44–46. ———. “Two Frameworks for Understanding Spatial Control and Political Integration in Chinese History.” Communication and Empire: Chinese Empires in Comparative Perspective, March 18, 2014. Accessed April 1, 2014. http://chinese -empires.eu/blog/historical-network-analysis-and-the​- dynamics​- of-literati​ -communication/. ———. “What Did Su Che See in the North? Publishing Laws, State Security, and Political Culture in Song China.” T’oung Pao: International Journal of Chinese Studies 92, nos. 4–5 (2006): 466–94. De Weerdt, Hilde, and Chu Mingkin. The Making of the Literati Notebook (9th– 13th c.). Special issue of East Asian Publishing and Society 6, no. 1 (2016). De Weerdt, Hilde, and Hou Ieong Ho. “Towards a New Platform for Tagging, Extracting, and Analyzing Named Entities in Classical Chinese Text Corpora.” Paper presented at the conference “Letters and Notebooks as Sources for Literati Communication in Chinese History, 900–1300,” Pembroke College, Oxford, UK, Jan. 10, 2014. De Weerdt, Hilde, Hou Ieong Ho, Chen Yunju, Li Yun-Chung, Lik Hang Tsui, and You Zixi. “Notebooks.” In Communication and Empire: Chinese Empires in



bibliogr aphy

481

Comparative Perspective, 2011. http://chinese-empires.eu/reference/information -territory-and-networks/. Deng Xiaonan 邓小南. “Yanying zhi jian—Songdai shangshu neisheng guankui” 掩映之間—宋代尚書內省管窺. Hanxue yanjiu 漢學研究 27, no. 2 (2009): 5–42. ———. Zuzong zhi fa: Bei Song qianqi zhengzhi shulüe 祖宗之法:北宋前期政治 述略. Beijing: Sanlian shudian, 2006. Dennis, Joseph. “Early Printing in China Viewed from the Perspective of Local Gazetteers.” In Knowledge and Text Production in an Age of Print: China, 900– 1400, edited by Lucille Chia and Hilde De Weerdt, 105–34. Leiden: Brill, 2011. ———. Writing, Publishing, and Reading Local Gazetteers in Imperial China, 1100– 1700. Harvard East Asian Monographs. Cambridge, MA: Harvard University Press, 2015. Di Cosmo, Nicola. Ancient China and Its Enemies: The Rise of Nomadic Power in East Asian History. Cambridge: Cambridge University Press, 2002. Dien, Albert. “Historiography of the Six Dynasties Period (220–581).” In The Oxford History of Historical Writing, vol. 1, Beginnings to ad 600, edited by Andrew Feldherr and Grant Hardy, 509–34. Oxford: Oxford University Press, 2011. Ding Haiyan 丁海燕. “Songren shiliao biji yanjiu—cong Siku quanshu zongmu dui Songdai shiliao biji de pingjia tanqi” 宋人史料笔记研究－从《四库全书总目》对 宋代史料笔记的评价谈起. Zhongzhou xuekan 中州学刊, no. 1 (2004): 112–16. Doyle, Michael W. Empires. Cornell Studies in Comparative History. Ithaca, NY: Cornell University Press, 1986. Drège, Jean-Pierre. “Des effets de l’imprimerie en Chine sous la dynastie des Song.” Journal Asiatique 282, no. 2 (1994): 409–42. Drompp, Michael Robert. Tang China and the Collapse of the Uighur Empire: A Documentary History. Brill’s Inner Asian Library, vol. 13. Leiden: Brill, 2005. Drucker, Johanna. “Humanities Approaches to Graphical Display.” Digital Humanities Quarterly 5, no. 1 (2011). Accessed May 22, 2012. http://digitalhumanities .org/dhq/vol/5/1/000091/000091.html. Drucker, Johanna, and Bethany Nowviskie. “Speculative Computing: Aesthetic Provocations in Humanities Computing.” In A Companion to Digital ­Humanities, edited by Susan Schreibman, Raymond George Siemens, and John Unsworth, 431–47. Malden, MA: Blackwell, 2004. Dudbridge, Glen. A Portrait of Five Dynasties China: From the Memoirs of Wang Renyu (880–956). Oxford: Oxford University Press, 2013. Ebrey, Patricia Buckley. “Introduction.” In Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, edited by Patricia Buckley Ebrey and Maggie Bickford, 1–27. Cambridge, MA: Harvard University Asia Center, 2006. Edgren, Søren. “Southern Song Printing at Hangzhou.” Bulletin of the Museum of Far Eastern Antiquities, no. 62 (1989): 1–212. Edney, Matthew H. “The Irony of Imperial Mapping.” In The Imperial Map:

482 bibliogr aphy Cartography and the Mastery of Empire, edited by James R. Akerman, 11–45. Chicago: University of Chicago Press, 2009. ———. Mapping an Empire: The Geographical Construction of British India, 1765– 1843. Chicago: University of Chicago Press, 1997. Egan, Ronald C. “Authorial Intent in Song Period biji: The Case of Zhou Hui’s 周煇 Qing Bo Za Zhi 清波雜志 (Miscellaneous Notes by One Who Lives Near the Gate of Clear Wave).” Paper presented at the conference “Letters and Notebooks as Sources for Literati Communication in Chinese History, 900–1300,” Pembroke College, Oxford, UK, Jan. 9, 2014. ———. “Book Printing and Resistance to It in Song Dynasty China.” Paper presented at the conference “The Early Development of Chinese Print Culture,” Cambridge, MA, April 29, 2005. ———. “Shen Kuo Chats with Inkstone and Brush.” In Idle Talk: Gossip and Anecdote in Traditional China, edited by Jack Chen and David Schaberg, 132–53. Berkeley: University of California Press, 2013. ———. “To Count Grains of Sand on the Ocean Floor: Changing Perceptions of Books and Learning in the Song Dynasty.” In First Impressions: The Cultural History of Print in Imperial China (8th–14th Centuries), edited by Lucille Chia and Hilde De Weerdt, 33–62. Leiden: Brill, 2011. Eisenstadt, S. N. The Political Systems of Empires. New Brunswick, NJ: Transaction Publishers, 1993 (1963). Eisenstein, Elizabeth L. The Printing Press as an Agent of Change: Communications and Cultural Transformations in Early Modern Europe. Cambridge: Cambridge University Press, 1979. Elliott, Mark C. The Manchu Way; The Eight Banners and Ethnic Identity in Late Imperial China. Stanford: Stanford University Press, 2001. Emirbayer, Mustafa. “Manifesto for a Relational Sociology.” American Journal of Sociology 103, no. 2 (1997): 281–317. Emirbayer, Mustafa, and Jeff Goodwin. “Network Analysis, Culture, and the Problem of Agency.” American Journal of Sociology 99, no. 6 (1994): 1411–54. Ess, Hans van. Von Ch’eng I zu Chu Hsi: Die Lehre vom Rechten Weg in der Überlieferung der Familie Hu. Asien- und Afrika-Studien der Humboldt-Universität zu Berlin, Bd. 13. Wiesbaden: Harrassowitz, 2003. Fan Fengshu 范凤书. Zhongguo sijia cangshu shi 中国私家藏书史. Zhengzhou: Daxiang chubanshe, 2001. Fang Hanqi 方汉奇. Zhongguo xinwen shiye tongshi 中国新闻事业通史. Beijing: Zhongguo renmin daxue chubanshe, 1992. Faure, David, and Helen F Siu, eds. Down to Earth: The Territorial Bond in South China. Stanford: Stanford University Press, 1995. Ferguson, Niall. Empire: How Britain Made the Modern World. London: Allen Lane, 2003. Fish, Stanley. “The Digital Humanities and the Transcending of Mortality.” In Opinionator Blog, New York Times, 2012. Accessed Jan. 9, 2012. http://­opinionator​



bibliogr aphy

483

.blogs.nytimes.com/2012/01/09/the-digital-humanities-and-the-­t ranscending​ -of-mortality/. ———. “Mind Your P’s and B’s: The Digital Humanities and Interpretation.” In Opinionator Blog, New York Times, 2012. Accessed Jan. 23, 2012. http://opinionator .blogs.nytimes.com/2012/01/23/mind-your-ps-and-bs-the-digital​-humanities​ -and-interpretation/. ———. “The Old Order Changeth.” In Opinionator Blog, New York Times, 2011. Accessed Dec. 26, 2011. http://opinionator.blogs.nytimes.com/2011/12/26/the-old​ -order-changeth/. Fort, Bruce. “American Slave Narratives: An Online Anthology, 1996–98.” Accessed May 22, 2012. http://xroads.virginia.edu/~hyper/wpa/wpahome.html. Frank, Paulo. “National Security in Northern Sung: A Subject for the Civil Service Examination.” Papers on Chinese History 1 (1992): 22–38. Franke, Herbert. “The Chin Dynasty.” In Alien Regimes and Border States, 907– 1368, edited by Herbert Franke and Denis Twitchett, 215–320. Cambridge: Cambridge University Press, 1994. ———. “Treaties between Sung and Chin.” In Études Song: In memoriam Étienne Balazs, edited by Françoise Aubin, 55–84. Paris: Mouton, 1970. Franke, Herbert, and Denis Twitchett, eds. Cambridge History of China, vol. 6, Alien Regimes and Border States, 907–1368. Cambridge: Cambridge University Press, 1994. Frankel, Oz. States of Inquiry: Social Investigations and Print Culture in NineteenthCentury Britain and the United States. Baltimore: Johns Hopkins University Press, 2006. Franzosi, Roberto, and John W. Mohr. “New Directions in Formalization and Historical Analysis.” Theory and Society 26, nos. 2–3 (1997): 133–60. Fu Lehuan 傅乐焕. Liao shi congkao 辽史丛考. Beijing: Zhonghua shuju, 1984. Fu, Philip Yuen-Ko. “A Study of Governmental Accounting in China: With Special Reference to the Sung Dynasty (960–1279).” Ph.D. diss., University of Illinois at Urbana-Champaign, 1968. Fu Xuancong 傅璇琮 and Xie Zhuohua 谢灼华. Zhongguo cangshu tongshi 中国藏 书通史. 2 vols. Ningbo: Ningbo chubanshe, 2001. Fukui Nobuaki 福井信昭. “Godai Shikkoku ki no shinsōin” 五代十国期の进奏 院. Ōsaka shiritsu daigaku tōyōshi ronsō 大阪市立大学东洋史论丛 14, no. 63–76 (2005). ———. “Tōdai no Shinsōin—Tō kōhanki ‘hanchin teisei’ no ichi sokumen” 唐代 の进奏院―唐后半期「藩镇体制」の一侧面. Tōhōgaku 东方学, no. 105 (2003): 47–62. Ge Gongzhen 戈公振. Chatu zhengli ben Zhongguo baoxue shi 插图整理本中国报 学史. Shanghai: Shanghai guji chubanshe, 2002. Ge Jianxiong 葛剑雄. Zhongguo gudai de ditu cehui 中国古代的地图测绘. Beijing: Shangwu yinshuguan, 1998. Ge Zhaoguang 葛兆光. “Songdai ‘Zhongguo’ yishi de tuxian 宋代’中国’意识的 凸显.” Wen shi zhe 文史哲 280, no. 1 (2004): 5–12.

484 bibliogr aphy Gerritsen, Anne. Ji’an Literati and the Local in Song-Yuan-Ming China. Leiden: Brill, 2007. Gong, Wei Ai. “The Reign of Hsiao-tsung (1162–1189).” In Cambridge History of China, vol. 5, pt. 1, The Sung Dynasty and Its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 710–55. Cambridge: Cambridge University Press, 2009. Gould, Roger V. “Uses of Network Tools in Comparative Historical Research.” In Comparative Historical Analysis in the Social Sciences, edited by James Mahoney and Dietrich Rueschemeyer, 241–69. Cambridge: Cambridge University Press, 2003. Gowers, Bernard. “Comparing Letters and Letters-Writers: China and Latin Europe in the Twelfth Century.” Paper presented at the conference “Letters and Notebooks as Sources for Literati Communication in Chinese History, 900– 1300,” Pembroke College, Oxford, UK, Jan. 9, 2014. Grafton, Anthony, and Daniel Rosenberg. Cartographies of Time. New York: Princeton Architectural Press, 2010. Guillory, John. “How Scholars Read.” ADE [The Association of Departments of En­ glish] Bulletin 146 (2008): 8–17. Guo Guang 郭光. Lu You zhuan 陆游传. Zhengzhou: Zhongzhou shuhua she, 1982. Guo Shengbo 郭声波. “Tang Song dili zongzhi cong diji dao shenglan de yanbian” 唐宋地理总志从地记到胜览的演变. Sichuan daxue xuebao 四川大学学报, no. 6 (2000): 85–92. Guojia wenwuju 国家文物局. Zhongguo wenwu dituji—Qinghai fence 中国文物地 图集—青海分册. Beijing: Zhongguo ditu chubanshe, 1996. ———. Zhongguo wenwu dituji—Shaanxi fence 中国文物地图集—陕西分册. Xi’an: Shaanxi ditu chubanshe, 1998. Harbsmeier, Christoph, ed. Thesaurus Linguae Sericae: An Historical and Comparative Encyclopaedia of Chinese Conceptual Schemes. Accessed May 22, 2012. http:// tls.uni-hd.de/. Hardt, Michael, and Antonio Negri. Empire. Cambridge, MA: Harvard University Press, 2000. Hargett, James M. “Song Dynasty Local Gazetteers and Their Place in the History of Difangzhi Writing.” Harvard Journal of Asiatic Studies 56, no. 2 (1996): 405–42. Harmon, Katharine A. You Are Here: Personal Geographies and Other Maps of the Imagination. New York: Princeton Architectural Press, 2004. Hartman, Charles. “Divided by a Common Language: Factional Conflict in Late Northern Song China (Review).” Journal of Song-Yuan Studies 40, no. 1 (2010): 141–49. ———. “The Making of a Villain: Ch’in Kuei and Tao-Hsueh.” Harvard Journal of Asiatic Studies 58, no. 1 (1998): 59–146. ———. “Poetry and Politics in 1079: The Crow Terrace Poetry Case of Su Shih.” Chinese Literature: Essays, Articles, Reviews (CLEAR) 12 (1990): 15–44.



bibliogr aphy

485

———. “The Reluctant Historian: Sun Ti, Chu Hsi, and the Fall of Northern Sung.” T’oung Pao: International Journal of Chinese Studies, no. 89 (2003): 25–57. Hartman, Charles, and Choying Li. “A Newly Discovered Inscription by Qin Gui: Its Implications for the History of Song Daoxue.” Harvard Journal of Asiatic Studies 70, no. 2 (2010): 387–448. Hartwell, Robert M. “Demographic, Political, and Social Transformations of China, 750–1550.” Harvard Journal of Asiatic Studies 42, no. 2 (1982): 365–442. ———. “Historical Analogism, Public Policy, and Social Science in Eleventh- and Twelfth-Century China.” American Historical Review 76, no. 3 (1971): 690–727. He Yuhong 何玉红. “Jingshi yishi yu Nan Song Chuan-Shaan bianfang shidi lei wenxian” 经世意识与南宋川陜边防史地类文献. Shixueshi yanjiu 史学史研究, no. 1 (2007): 30–36. Henderson, John. “Chinese Cosmographical Thought.” In Cartography in the Traditional East and Southeast Asian Societies, edited by David Woodward and John Brian Harley, 203–27. Chicago: University of Chicago Press, 1994. Hermann, Albert. Die Westländer in der Chinesischen Kartographie. Stockholm: Lithographic Institute of the General Staff of the Swedish Army, 1922. Himmelfarb, Gertrude. “A Neo-Luddite Reflects on the Internet.” Education Digest 62, no. 6 (1997): 50. Hitchcock, Tim. “Academic History Writing and the Headache of Big Data.” In Historyonics, 2012. Accessed January 30, 2012. http://historyonics.blogspot .co.uk/2012/01/academic-history-writing-and-headache.html. Hockey, Susan. “History of Humanities Computing.” In A Companion to Digital Humanities, edited by Susan Schreibman, Raymond George Siemens, and John Unsworth, 3–19. Malden, MA: Blackwell, 2004. Hostetler, Laura. Qing Colonial Enterprise: Ethnography and Cartography in Early Modern China. Chicago: University of Chicago Press, 2001. Hsu, Mei-Ling. “The Qin Maps: A Clue to Later Chinese Cartographic Development.” Imago Mundi 45 (1993): 90–100. Hu Dajun 胡大浚 and Ma Lanzhou 马兰州. “Qishi nian biansai shi yanjiu zongshu” 七十年边塞诗研究综述. Zhongguo wenxue yanjiu 中国文学研究 3 (2000): 88–92. Huang Kuanchong 黃寬重. Nan Song difang wuli: difangjun yu minjian ziwei wuli de tantao 南宋地方武力:地方軍與民間自衛武力的探討. Taibei: Dong da tu shu gong si, 2002. ———. “Wan Song chaochen dui guoshi de zhengyi—Lizong shidai de hezhan, bianfang yu liumin” 晚宋朝臣對國是的爭議—理宗時代的和戰、邊防與流民. Master’s thesis, National Taiwan University, 1975. Huang Shengzhang 黄盛璋. “Songke Yuditu zongkao 宋刻舆地图综考.” In Zhongguo gudai ditu ji: Zhanguo–Yuan 中国古代地图集:战国—元, edited by Cao Wanru 曹婉如 et al., 61–64. Beijing: Wenwu chubanshe, 1990. Huang Shengzhang, and Wang Qianjin 汪前进. “Zuizao de yifu Xi Xia ditu—‘Xi Xia dixing tu’ xin tan” 最早的一幅西夏地图—〈西夏地形图〉新探. Ziran kexue yanjiu 自然科学研究 11, no. 2 (1992): 177–87.

486 bibliogr aphy Huang Zhuoming 黄卓明. Zhongguo gudai baozhi tanyuan 中国古代报纸探源. Beijing: Renmin ribao chubanshe, 1983. Hucker, Charles O. A Dictionary of Official Titles in Imperial China. Stanford: Stanford University Press, 1985. Huiren 慧仁. “Song kan Su Shi quanji kao 宋刊苏轼全集考.” In Douban, Dec. 8, 2009. Accessed May 22, 2012. http://www.douban.com/group/topic/8952833/. Hymes, Robert. Statesmen and Gentlemen: The Elite of Fu-chou, Chiang-hsi, in North­ ern and Southern Sung. Cambridge: Cambridge University Press, 1986. Inglis, Alister D. “A Textual History of Hong Mai’s Yijian Zhi.” T’oung Pao: International Journal of Chinese Studies 93, no. 4 (2007): 283–368. Jacob, Christian. “Toward a Cultural History of Cartography.” Imago Mundi 48 (1996): 191–98. ———. The Sovereign Map: Theoretical Approaches in Cartography throughout History. Edited by Edward H. Dahl. Chicago: University of Chicago Press, 2006. Jay, Jennifer W. A Change in Dynasties: Loyalism in Thirteenth-Century China. Bel­ lingham: Western Washington University, 1991. Johnson, David G. “The Last Years of a Great Clan: The Li Family of Chao Chün in Late T’ang and Early Sung.” Harvard Journal of Asiatic Studies 37, no. 1 (1977): 5–102. ———. The Medieval Chinese Oligarchy. Boulder: Westview Press, 1977. Johnson, Linda Cooke. Women of the Conquest Dynasties: Gender and Identity in Liao and Jin China. Honolulu: University of Hawaii Press, 2011. Johnson, Wallace. The T’ang Code. Princeton: Princeton University Press, 1979. Kiang, Heng Chye. Cities of Aristocrats and Bureaucrats: The Development of Medieval Chinese Cityscapes. Honolulu: University of Hawaii Press, 1999. Kohn, Livia. Daoism Handbook. Leiden: Brill, 2000. Kondō Kazunari 近藤一成. “Sōdai Eika gakuha Yō Teki no ka-i kan” 宋代永嘉学 派葉適の華夷観. Shigaku zasshi 史学雑誌 88, no. 6 (1979): 51–79. Kong Xue 孔学. “Lu You ji Gaozong shengzheng cao” 陆游及高宗圣政草. Shixue yuekan 史学月刊, no. 4 (1996): 32–38. Korzec, Michel. Het voelen van de draak: de opening van het district Yonghe en andere verhalen uit het midden van China. Amsterdam: B. Bakker, 1986. Kramer, Michael. “The Fetishization of Data: The Data-Reality Conflation and the Role of the Digital Humanities.” In Issues in Digital History, 2012. Accessed February 2, 2012. http://www.michaeljkramer.net/issuesindigitalhistory/blog /?p=559. Kubota Kazuo 久保田和男. “Sōchō ni yoru shomin e no jōhō dentatsu to insatsu bunka—Hō to rei o chūshin to shite” 宋朝による庶民への情报伝达と印刷文 化—法と礼を中心として. Paper presented at the conference “Guoji Songshi yantao­ hui ji Zhongguo Songshi yanjiuhui di shisan jie nianhui” 国际宋史研讨会暨中 国宋史研究会第十三届年会, Kunming, China, July 29–Aug. 1, 2008. ———. “Sōdai ni okeru seichoku no dentatsu ni tsuite—Genpō kaisei izen o



bibliogr aphy

487

chūshin to shite” 宋代における制勅の伝达について—元豊改制以前を中心と して. In Sōdai shakai no nettowāku 宋代社会のネットワ一ク, edited by Sōdaishi Kenkyūkai 宋代史研究会, 197–232. Tokyo: Kyūko Shoin, 1998. ———. “Sōdai no chūō jōhō no chihō dentatsu ni tsuite—Teihō o chūshin to shite” 宋代の中央情报の地方伝达について－邸报を中心として. In Tang Song shiqi de wenshu chuandi yu xinxi goutong guoji xueshu gongzuofang 唐宋时期的文 书传递与信息沟通国际学术工作坊. Beijing: Beijing University, 2007. ———. Songdai Kaifeng yanjiu 宋代开封研究. Translated by Guo Wanping 郭万平 and Dong Ke 董科. Shanghai: Shanghai guji chubanshe, 2010 (original Japanese edition 2007). Kuhn, Philip A. Soulstealers: The Chinese Sorcery Scare of 1768. Cambridge, MA: Har­vard University Press, 1990. Lam, Joseph S. C. “Huizong’s Dashengyue, a Musical Performance of Emperorship and Officialdom.” In Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, edited by Patricia Ebrey and Maggie Bickford, 395–452. Cambridge, MA: Harvard University Press, 2006. Lamouroux, Christian. “De l’étrangeté à la difference: Les récits des missaires Song en pays Liao (Xiè S.).” In Récits de voyages asiatiques: Genres, mentalités, conception de l’espace: Actes du Colloque Efeo-Ehess de Décembre 1994, edited by Claudine Salmon, 101–26. Paris: École Française d’Extrême-Orient, 1996. ———. Fiscalité, comptes publics et politiques financières dans la Chine des Song: Le chapitre 179 du Songshi. Bibliothèque de l’Institut des Hautes Études Chinoises, vol. 33. Paris: Collège de France, 2003. ———. “From the Yellow River to the Huai: New Representations of a River Network and the Hydraulic Crisis of 1128.” In Sediments of Time, edited by Mark Elvin and Ts’ui-jung Liu, 545–84. Cambridge: Cambridge University Press, 1998. ———. “Geography and Politics: The Song-Liao Border Dispute of 1074/75.” In China and Her Neighbours: Borders, Visions of the Other, Foreign Policy 10th to 19th Century, edited by Sabine Dabringhaus, Roderich Ptak, and Richard Teschke, 1–28. Wiesbaden: Harrassowitz, 1997. Lee, Sukhee. “Negotiated Power: The State and Elites in 12th–14th Century China.” Ph.D. diss., Harvard University, 2009. ———. Negotiated Power: The State, Elites, and Local Governance in Twelfth- to Fourteenth-Century China. Cambridge, MA: Harvard University Asia Center, 2014. Lee, Thomas H. C. Government Education and Examinations in Sung China. Hong Kong: Chinese University of Hong Kong Press, 1985. ———. Handbook of Oriental Studies. Vol. 13, Education in Traditional China, a History. Leiden: Brill, 2000. Legge, James. The Chinese Classics. 5 vols. Taibei: SMC Publishing, 1991 (1893–1895). ———. The Ch’un Ts’ew with the Tso Chuen. Hong Kong: Hong Kong University Press, 1960. ———. The Li Ki. Edited by F. Max Müller. Vol. 27 of The Sacred Books of the East. New York: Dover Publications, 1966.

488 bibliogr aphy ———, ed. The Sacred Books of China. Edited by F. Max Müller. Vol. 3 of The Sacred Books of the East. New York: Dover Publications, 1962. Lemercier, Claire. “Analyse de réseaux et histoire.” Revue d’ histoire moderne et contemporaine 52, no. 2 (2005): 88–112. Leung, Irene S. “Felt Yurts Neatly Arrayed, Large Tents Huddle Close: Visualizing the Frontier in the Northern Song Dynasty (960–1127).” In Political Frontiers, Ethnic Boundaries, and Human Geographies in Chinese History, edited by Nicola Di Cosmo and Don J Wyatt, 192–219. London: Routledge, 2003. ———. “The Frontier Imaginary in the Song Dynasty (960–1279): Revisiting Cai Yan’s ‘Barbarian Captivity’ and Return.” Ph.D. diss., University of Michigan, 2001. Levine, Ari Daniel. Divided by a Common Language: Factional Conflict in Late Northern Song China. Honolulu: University of Hawaii Press, 2008. ———. “A House in Darkness: The Politics of History and the Language of Politics in the Late Northern Song, 1068–1104.” Ph.D. diss., Columbia University, 2002. ———. “The Reigns of Hui-tsung (1100–1126) and Ch’in-tsung (1126–1127) and the Fall of the Northern Sung.” In Cambridge History of China, vol. 5, pt. 1, The Sung Dynasty and Its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 556–643. Cambridge: Cambridge University Press, 2009. Lewis, Mark Edward. The Early Chinese Empires: Qin and Han. History of Imperial China. Cambridge, MA: Belknap Press, 2007. ———. The Flood Myths of Early China. SUNY Series in Chinese Philosophy and Culture. Albany: State University of New York Press, 2006. ———. Writing and Authority in Early China. SUNY Series in Chinese Philosophy and Culture. Albany: State University of New York Press, 1999. Li Bin 李彬. “Tangdai wenming yu xinwen chuanbo” 唐代文明与新闻传播. Ph.D. diss., Zhongguo renmin daxue, 1998. Li Chuanyin 李传印. “Lun Hua Yue” 论华岳. Chizhou shizhuan xuebao 池州师专 学报, no. 2 (1996): 63–69. Li Qi 李琪. “Woguo Songdai ditu dang’an gongzuo de fangfa yu guanli zhidu” 我国 宋代地图档案工作的方法与管理制度. Gansusheng jingji guanli ganbu xueyuan xuebao 甘肃省经济管理干部学院学报 18, no. 2 (2005): 63–65. Li Yu’an 李玉安, and Chen Yuanyi 陈传艺. Zhongguo cangshujia cidian 中国藏书家 辞典. Wuhan: Hubei jiaoyu chubanshe, 1989. Li Zhen 李震. Zhongguo zhengzhi guofang shi 中国政治国防史. Taibei: Taiwan shangwu yinshuguan, 1986. Li Zhizhu 李致洙. Lu You shi yanjiu 陆游诗研究. Taibei: Wenshizhe chubanshe, 1991. Liang Gengyao 梁庚堯. Songdai shehui jingji shi lunji 宋代社會經濟史論集. Taibei: Yunchen wenhua shiye gufen youxian gongsi, 1997. Liao, Hsien-Huei. “Exploring Weal and Woe: The Song Elite’s Mantic Beliefs and Practices.” T’oung Pao 91, nos. 4–5 (2005): 347–95. Liberman, Mark. “The ‘Dance of the P’s and B’s’: Truth or Noise?” In Language Log, Jan. 26, 2012. Accessed Jan. 26, 2012. http://languagelog.ldc.upenn.edu/nll /?p=3730.



bibliogr aphy

489

Lieberman, Victor. “Introduction to Modern Asian Studies.” Modern Asian Studies 31, no. 3 (1997): 449–61. ———. “Transcending East–West Dichotomies: State and Culture Formation in Six Ostensibly Disparate Areas.” Modern Asian Studies 31, no. 3 (1997): 463–546. Liu, Hsiang-kwang 刘祥光. “Printing and Examinations: The Circulation of Study Aids in Song China.” Paper presented at the workshop “The Early Development of Chinese Print Culture.” Fairbank Canter for Chinese Studies, Harvard University, Cambridge, MA, April 29, 2005. ———. “Yinshua yu kaoshi: Songdai kaoshi shiyong cankaoshu chu tan” 印刷与 考试:宋代考试用参考书初探. Song shi yanjiu ji宋史研究集 31 (2001): 151–200. Liu, James T. C. Ou-Yang Hsiu: An Eleventh-Century Neo-Confucianist. Stanford: Stanford University Press, 1967. Liu Liankai 刘连开. “Songdai xiu shi zhidushi yanjiu shuping” 宋代修史制度史 研究述评. In Songdai zhidu shi yanjiu bainian, 1900–2000 宋代制度史研究百年, 1900–2000, edited by Bao Weimin 包伟民, 267–80. Beijing: Shangwu yinshuguan, 2004. Liu Lin 劉琳, and Shen Zhihong 沈治宏. Xiancun Song ren zhushu zonglu 現存宋 人著述總錄. Chengdu: Ba Shu shushe, 1995. Liu, Lydia He. The Clash of Empires: The Invention of China in Modern World Making. Cambridge, MA: Harvard University Press, 2004. Liu Shenning 刘申宁, ed. Zhongguo bingshu zongmu 中国兵书总目. Beijing: Guofang daxue chubanshe, 1990. Liu Yeqiu 刘叶秋. Lidai biji gaishu 历代笔记概述. Beijing: Beijing chubanshe, 2003 (1980). Liu Zhaoyou 刘兆佑, and Guoli bianyiguan Zhonghua congshu bianshen weiyuan­ hui 国立编译馆中华丛书编审委员会. Songshi “Yiwen zhi” shibu yiji kao 宋史艺 文志史部佚籍考. Taibei: Guoli bianyiguan Zhonghua congshu bianshen weiyuanhui, 1984. Liu Zijian 刘子健. See Liu, James T. C. Love, Harold. The Culture and Commerce of Texts: Scribal Publication in SeventeenthCentury England. Amherst: University of Massachusetts Press, 1998. Luo, Yinan. “A Study of the Changes in the ‘Tang-Song Transition Model.’” ­Journal of Song Yuan Studies 35 (2005): 99–127. Mann, Michael. The Sources of Social Power, vol. 1. Cambridge: Cambridge University Press, 1986. May, Jon, and N. J. Thrift. Timespace: Geographies of Temporality. Critical Geographies. London: Routledge, 2001. McCarty, Willard. “Looking Backward, Figuring Forward: Modelling, Its Discontents and the Future.” June 2007. Accessed May 22, 2012. http://www.mccarty .org.uk/essays/McCarty,%20Looking%20backward.pdf. ———. “What Is Humanities Computing? Toward a Definition of the Field.” 1998. Accessed May 22, 2012. http://www.cch.kcl.ac.uk/legacy/teaching/dtrt/class1 /mccarty_humanities_computing.pdf. McDermott, Joseph. The Ascendance of the Imprint in China.” In Printing and

490 bibliogr aphy Book Culture in Late Imperial China, edited by Cynthia Joanne Brokaw and Chow Kai-wing, 55–104. Berkeley: University of California Press, 2005. ———. “From Collection to Archives: The Life of Family Imprints in Huizhou, 1500–1700.” Paper presented at the conference “Imprimer autrement: Le livre non commercial dans la Chine impériale.” École Française d’Extrême-Orient, Paris, June 12, 2009. McGann, Jerome. “The Rationale of HyperText.” 1995. Accessed May 22, 2012. http://www2.iath.virginia.edu/public/jjm2f/rationale.html. McGrath, Michael Charles. “Military and Regional Administration in Northern Sung China (960–1126).” Ph.D. diss., Princeton University, 1982. McLean, Paul Douglas. The Art of the Network: Strategic Interaction and Patronage in Renaissance Florence. Politics, History, and Culture. Durham, NC: Duke University Press, 2007. McMullen, David. State and Scholars in T’ang China. Cambridge: Cambridge University Press, 1988. Miao Shumei 苗书梅, and Wang Yunhai 王云海, eds. 2001. Song huiyao jigao: chong ru 宋会要辑稿:崇儒. Kaifeng: Henan daxue chubanshe, 2001. Mittler, Barbara. A Newspaper for China? Power, Identity, and Change in Shanghai’s News Media, 1872–1912. Harvard East Asian Studies Monographs. Cambridge, MA: Harvard University Asia Center, 2004. Miyakawa, Hisayuki. “An Outline of the Naitō Hypothesis and Its Effects on Japanese Studies of China.” Far Eastern Quarterly 14, no. 4 (1955): 533–52. Mohr, John W. “Network Methods, Historical Texts, and the Formal Analysis of Institutional Systems.” Paper presented at the 162nd Annual Meeting of the American Historical Association, Chicago, Jan. 5–8, 2012. ———. “Soldiers, Mothers, Tramps and Others: Discourse Roles in the 1907 New York City Charity Directory.” Poetics 22 (1994): 327–57. Mohr, John W., and Vincent Duquenne. “The Duality of Culture and Practice: Poverty Relief in New York City, 1888–1917.” Theory and Society 26 (1997): 305–56. Moretti, Franco. Atlas of the European Novel, 1800–1900. London: Verso, 1998. ———. Graphs, Maps, Trees: Abstract Models for Literary History. London: Verso, 2007 (2005). Mostern, Ruth. “Apprehending the Realm: Territoriality and Political Power in Song China, 960–1276 CE.” Ph.D. diss., University of California, Berkeley, 2003. ———. Dividing the Realm in Order to Govern: The Spatial Organization of the Song State. Cambridge, MA: Harvard University Asia Center, 2011. ———. “From Battlefields to Counties: War, Border, and State Power in Southern Song Huainan.” In Battlefronts Real and Imagined: War, Border and Identity in the Chinese Middle Period, edited by Don Wyatt, 227–52. New York: Palgrave Macmillan, 2008. Motyl, Alexander J. “Is Everything Empire? Is Empire Everything?” Comparative Politics 38, no. 2 (2006): 229–49. Münkler, Herfried. Empires: The Logic of World Domination from Ancient Rome to the United States. Translated by Patrick Camiller. Cambridge: Polity, 2007. Musolff, Andreas. “What Can Critical Metaphor Analysis Add to the Understand-



bibliogr aphy

491

ing of Racist Ideology? Recent Studies of Hitler’s Anti-Semitic Metaphors.” Critical Approaches to Discourse Analysis across Disciplines 2, no. 2 (2008): 1–10. Needham, Joseph, and Wang Ling. Science and Civilisation in China, vol. 3, Mathematics and the Sciences of the Heavens and the Earth. Cambridge: Cambridge University Press, 1959. Neustadt, Richard E., and Ernest R. May. Thinking in Time: The Uses of History for Decision-Makers. New York: Free Press; London: Collier Macmillan, 1986. Niida Noboru 仁井田陞. “Keigen jōhō jirui to Sōdai no shuppanhō” 庆元条法事 と宋代の出版法. In Chūgoku hōseishi kenkyū 中国法制史研究, 445–65. Tokyo: Tōkyō daigaku shuppankai, 1981 (1935). Nugent, Christopher M. B. “The Circulation of Poetry in Tang Dynasty China.” Ph.D. diss., Harvard University, 2004. Nunberg, Geoffrey, ed. The Future of the Book. Berkeley: University of California Press, 1996. Nylan, Michael. The Five “Confucian” Classics. New Haven: Yale University Press, 2001. Ouyang Daifa 欧阳代发, and Wang Zhaopeng 王兆鹏, eds. Liu Kezhuang ci xin shi ji ping 刘克庄词新释辑评. Beijing: Zhongguo shudian, 2001. Owen, Stephen. “Postface: ‘Believe It or Not.’ ” In Idle Talk: Gossip and Anecdote in Traditional China, edited by Jack Chen and David Schaberg, 217–23. Berkeley: University of California Press, 2013. ———. Readings in Chinese Literary Thought. Harvard-Yenching Institute Monograph Series. Cambridge, MA: Harvard University Press, 1992. Pan Meiyue 潘美月. Songdai cangshujia kao 宋代藏书家考. Taibei: Xuehai chubanshe, 1980. Pan Sheng 潘晟. “Cong Songdai shiwen kan yousi yu shenglan dui Songdai ­d ituxue fazhang de yingxiang” 从宋代诗文看幽思与胜览思想对宋代地图学 发展的影响. Zhongguo lishi dili luncong 中国历史地理论丛, no. 2 (2010): 69–83. ———. Ditu de zuozhe ji qi yuedu: Yi Song-Ming wei hexin de zhishishi kaocha 地圖 的作者及其閱讀:以宋明為核心的知識史考察. Nanjing: Jiangsu renmin chubanshe, 2013. ———. “Songdai dilixue de guannian, tixi yu zhishi xingqu” 宋代地理学的观 念、体系与知识兴趣. Ph.D. diss., Beijing University, 2008. Pan, Yihong. Son of Heaven and Heavenly Qaghan: Sui-Tang China and Its Neighbors, Studies on East Asia, vol. 20. Bellingham: Center for East Asian Studies, Western Washington University, 1997. Pearson, Mike, and Michael Shanks. Theatre/Archaeology. London: Routledge, 2001. Perdue, Peter C. China Marches West: The Qing Conquest of Central Eurasia. Cambridge, MA: Belknap Press, 2005. Poo, Mu-Chou. Enemies of Civilization: Attitudes toward Foreigners in Ancient

492 bibliogr aphy Mesopotamia, Egypt, and China. SUNY Series in Chinese Philosophy and Culture. Albany: State University of New York Press, 2005. Poon, Ming-Sun. “Books and Printing in Sung China (960–1279).” Ph.D. diss., University of Chicago, 1979. Power, Daniel, and Naomi Standen. Frontiers in Question: Eurasian Borderlands, 700–1700. Themes in Focus. New York: St. Martin’s Press, 1999. Qi Zhiping 齐治平. Lu You zhuan lun 陆游传论. Shanghai: Gudian wenxue chubanshe, 1958. Qian Zheng 钱正, and Yao Shiying 姚世英. “Dili tu bei” 墬理图碑. In Zhongguo gudai ditu ji: Zhanguo–Yuan 中国古代地图集:战国–元, edited by Cao Wanru 曹婉如, 46–49. Beijing: Wenwu chubanshe, 1990. Qiao Yanguan 乔衍琯. Chen Zhensun xueji 陈振孙学记. Taibei: Wenshizhe chubanshe, 1980. Qiu Minggao 邱鸣皋, Nanjing daxue 南京大学, and Zhongguo sixiangjia yanjiu zhong xin 中国思想家硏究中心. Lu You pingzhuan 陆游评传. Nanjing: Nanjing daxue chubanshe, 2002. Quan Hansheng 全漢升. “Song Jin de zousi maoyi” 宋金的走私貿易. Lishi yuyan yanjiusuo jikan 歷史語言研究所集刊 11 (1944): 425–47. Ramsay, Stephen. Reading Machines: Toward an Algorithmic. Urbana: University of Illinois Press, 2011. Ren, Jincheng 任金城. “Muke Liujing tu chukao” 木刻六经图初考. In Zhongguo gudai ditu ji: Zhanguo–Yuan 中国古代地图集:战国–元, edited by Cao Wanru 曹婉如, 61–64. Beijing: Wenwu chubanshe, 1990. Ren Jiyu 任继愈. Zhongguo cangshulou 中国藏书楼. 3 vols. Shenyang: Liaoning renmin chubanshe, 2001. Robins, G. L., and P. Pattison. “Interdependencies and Social Processes: Generalized Dependence Structures.” In Models and Methods in Social Network Analysis, edited by Peter J. Carrington, John Scott, and Stanley Wasserman, 192–214. Cambridge: Cambridge University Press, 2005. Rossabi, Morris, ed. China among Equals: The Middle Kingdom and Its Neighbors, 10th–14th Centuries. Berkeley: University of California Press, 1983. Satō Takanori, 佐藤隆则. “Shin Jun no gakumon to shisō—Shu Ki jūgaku ki” 陈 淳の学问と思想—朱熹从学期. Daitō Bunka Daigaku Kangakkai shi 大东文化大 学汉学会志 29 (1990): 138–52. ———. “Shin Jun no gakumon to shisō—Shu Ki jūgaku izen” 陈淳の学问と思 想—朱熹从学以前. Daitō Bunka Daigaku Kangakkai shi 大东文化大学汉学会志 28 (1989): 44–64. Schirokauer, Conrad. “Liu Cheng.” In Sung Biographies, edited by Herbert Franke, 626. Wiesbaden: Steiner, 1976. Schnapp, Jeffrey, Peter Lunenfeld, and Todd Presner. “Digital Humanities M ­ anifesto 2.0.” 2009. Accessed May 22, 2012. http://www.humanitiesblast.com/manifesto /Manifesto_V2.pdf.



bibliogr aphy

493

Scholle, David. “What Is Information? The Flow of Bits and the Control of Chaos.” In Democracy and New Media, edited by Henry Jenkins, David Thorburn, and Brad Seawell, 343–64. Cambridge, MA: MIT Press, 2003. Scott, James C. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press, 1998. Scott, John. ­Social Network Analysis: A Handbook. Newberry Park, CA: Sage, 2000. Sewell, William H. Logics of History: Social Theory and Social Transformation. Chicago: University of Chicago Press, 2005. Shanghai Xinsijun lishi yanjiuhui Yinshua yinchao fenhui 上海新四军历史硏究会 印刷印钞分会. Zhuangding yuanliu he buyi 装订源流和补遗. Beijing: Zhongguo shuji chubanshe, 1993. Shen Songqin 沈松勤. Nan Song wenren yu dangzheng 南宋文人与党争. Beijing: Renmin chubanshe, 2005. Shields, Anna M. “Gossip, Anecdote, and Literary History: Representations of the Yuanhe Era in Tang Anecdote Collections.” In Idle Talk: Gossip and Anecdote in Traditional China, edited by Jack W. Chen and David Schaberg, 107–31. Berkeley: University of California Press, 2014. Skinner, G. William. The City in Late Imperial China. Stanford: Stanford University Press, 1977. ———. Marketing and Social Structure in Rural China. Ann Arbor, MI: Association for Asian Studies, 2001. ———. “Mobility Strategies in Late Imperial China: A Regional Systems Analysis.” In Regional Analysis, edited by Carol A. Smith, 327–64. New York: Academic Press, 1976. Skocpol, Theda, and Margaret Somers. “The Uses of Comparative History in Macro­ social Inquiry.” Comparative Studies in Society and History 22, no. 2 (1980): 174–97. Smith, Paul Jakov. “Irredentism as Political Capital: The New Policies and the Annexation of Tibetan Domains in Hehuang (the Qinghai-Gansu Highlands) under Shenzong and His Sons, 1068–1126.” In Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, edited by Patricia Buckley Ebrey and Maggie Bickford, 78–130. Cambridge, MA: Harvard University Asia Center, 2006. Sogabe Shizuo 曽我部静雄. Shina seiji shūzoku ronkō 支那政治习俗论考. Tokyo: Chikuma Shobō, 1943. Spacks, Patricia Meyer. “Private Conversation, Public Meaning.” Social Research 65, no. 3 (1998): 611–30. Standen, Naomi. “The Five Dynasties.” In Cambridge History of China, vol. 5, pt. 1, The Sung Dynasty and Its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 38–132. Cambridge: Cambridge University Press, 2009. ———. Unbounded Loyalty: Frontier Crossing in Liao China. Honolulu: University of Hawaii Press, 2007. Steinhardt, Nancy Shatzman. Chinese Imperial City Planning. Honolulu: University of Hawaii Press, 1990. Subrenat, Jean-Jacques. “Recueil sur la défense des villes (守城录 par Chen Gui 陈规

494 bibliogr aphy et Tang Shou 汤寿), recherches sur l’histoire militaire de la Chine du XIIe siècle.” Ph.D. diss., Université de Paris, 1968. Sun Li 孙励. “Songdai biji fenlei kaobian” 宋代笔记分类考辨. Master’s thesis, Shanghai shifan daxue, 2004. Sun Wei 孙伟. “Bei Song shiqi Huangtu Gaoyuan diqu chengzhaibao tixi yanbian yanjiu” 北宋时期黄土高原地区城寨堡体系演变研究. Master’s thesis, Shaanxi shifan daxue, 2005. Sung, Chia-fu. “Between Tortoise and Mirror: Historians and Historiography in Eleventh-Century China.” Ph.D. diss., Harvard University, 2010. Svensson, Patrik. “Humanities Computing as Digital Humanities.” Digital Humanities Quarterly 3, no. 3 (2009). Accessed May 22, 2012. http://digitalhumanities .org/dhq/vol/3/3/000065/000065.html. Tackett, Nicolas. The Destruction of the Medieval Chinese Aristocracy. Cambridge, MA: Harvard University Asia Center, 2014. ———. “Great Clansmen, Bureaucrats, and Local Magnates: The Structure and Circulation of the Elite in Late-Tang China.” Asia Major 21, no. 2 (2008): 101–52. ———. “The Great Wall and Conceptualizations of the Border under the Northern Song.” Journal of Song-Yuan Studies 38 (2008): 99–138. Tao, Jing-shen 陶晉生. “Barbarians or Northerners: Northern Sung Images of the Khitans.” In China among Equals: The Middle Kingdom and Its Neighbors, 10th– 14th Centuries, edited by Morris Rossabi, 66–86. Berkeley: University of California Press, 1983. ———. “The Move to the South and the Reign of Kao-tsung (1127–1162).” In Cambridge History of China, vol. 5, pt. 1, The Sung Dynasty and Its Precursors, 907–1279, edited by Denis Twitchett and Paul Jakov Smith, 645–709. Cambridge: Cambridge University Press, 2009. ———. “Nan Song liyong shanshuizhai de fangshou zhanlue” 南宋利用山水寨的 防守戰略. Shihuo yuekan 食貨月刊 7, nos. 1–2 (1977): 1–10. ter Haar, B. J. “Newly Recovered Anecdotes from Hong Mai’s (1123–1202) Yijian zhi.” Journal of Sung-Yuan Studies 23 (1993): 19–41. ———. Telling Stories: Witchcraft and Scapegoating in Chinese History. Leiden: Brill, 2006. ———. The White Lotus Teachings in Chinese Religious History. Leiden: Brill, 1992. Thiele, Dagmar. “Der Abschluss Eines Vertrages: Diplomatic Zwischen Sungund Chin-Dynastie, 1117–1123.” Bulletin of the School of Oriental and African Studies 35, no. 2 (1972): 440. Thomas, William G. “Computing and the Historical Imagination.” In A Companion to Digital Humanities, edited by Susan Schreibman, Raymond George Siemens, and John Unsworth, 56–68. Malden, MA: Blackwell, 2004. Also available online at http://digitalhumanities.org/companion/view?docId=blackwell /9781405103213/9781405103213.xml&chunk.id=ss1-2-5. Tillman, Hoyt Cleveland. “Proto-Nationalism in Twelfth Century China? The Case of Chen Liang.” Harvard Journal of Asiatic Studies 39, no. 2 (1979): 403–28.



bibliogr aphy

495

Tsien, Tsuen-hsuin. Paper and Printing. Edited by Joseph Needham. Vol. 5.1 of Science and Civilization in China. Cambridge: Cambridge University Press, 1985. Tsui, Lik Hang [Lincoln]. “Writing Letters in Song China (960–1279): A Study of Its Political, Social, and Cultural Uses.” Ph.D. diss., University of Oxford, 2015. Tuan, Yi-Fu. Space and Place: The Perspective of Experience. London: Edward Arnold, 1977. Twitchett, Denis, ed. Cambridge History of China, vol. 3, Sui and T’ang China, 589–906, pt. 1. Cambridge: Cambridge University Press, 1979. Twitchett, Denis, and Michael Loewe. Cambridge History of China, vol. 1, The Ch’ in and Han Empires, 221 B.C.–A.D. 220. Edited by D. Twitchett and J. K. Fairbank. Cambridge: Cambridge University Press, 1986. Umehara Kaoru 梅原郁. “Shinsōin o megutte—Sōdai no bunsho dentatsu seido” 进奏院をめぐって—宋代の文书伝达制度. Shūjitsu joshi daigaku shigaku ronsō 就実女子大学史学论集 15 (2000): 69–130. Vivier, Brian. “Chinese Foreign Trade, 960–1276.” Ph.D. diss., Yale University, 2008. Waldron, Arthur. The Great Wall of China: From History to Myth. Cambridge: Cambridge University Press, 1992 (1990). Walton, Linda A. Academies and Society in Southern Sung China. Honolulu: Univer­ sity of Hawaii Press, 1999. Wang, Ao. “Cartographies Actual and Imagined: Imperial Grand Maps and MidTang.” Paper presented at the conference on Middle Period China, 800–1400. Harvard University, Cambridge, MA, June 5–7, 2014. Wang Deyi 王德毅. “Songdai de qijuzhu yu shizhengji zhi yanjiu” 宋代的起居注 與時政記之研究. In Di er jie Guoji Hanxue Huiyi lunwenji (Lishi kaogu zu) 第 二屆國際漢學會議論文集(歷史與考古組), edited by Di er jie Guoji Hanxue Huiyi lunwen ji bianweihui 第二屆國際漢學會議論文集編委會, 981–1005. Taibei: Zhongyang yanjiuyuan lishi yuyan yanjiusuo, 1989. ———. “Songdai de rili he yudie zhi yanjiu” 宋代的日曆和玉牒之研究. Guoli Taiwan daxue lishixi xuebao 國立臺灣大學歷史學系學報 10–11 (1984): 119–37. ———. “Songdai de shengzheng he baoxun zhi yanjiu” 宋代的圣政和寳训之研究. Song shi yanjiu ji 宋史研究集 30 (2000): 1–26. Wang Guiping 王桂平. Jiakeben 家刻本. Nanjing: Fenghuang chubanshe, 2002. Wang, Gungwu. “Feng Tao: An Essay on Confucian Loyalty.” In Confucian Personalities, edited by Arthur F. Wright and Denis Twitchett, 123–45. Stanford: Stanford University Press, 1962. ———. “The Rhetoric of a Lesser Empire: Early Sung Relations with Its Neighbors.” In China among Equals: The Middle Kingdom and Its Neighbors, 10th–14th Centuries, edited by Morris Rossabi, 47–65. Berkeley: University of California Press, 1983. Wang, Hung-tai. “Information Media, Social Imagination, and Public Society during the Ming and Qing Dynasties.” Frontiers of History in China 5, no. 2 (2010): 169–216.

496 bibliogr aphy Wang Jing 王静. “Chaoting he fangzhen de lianluo shuniu: shitan Zhong-, Wan-Tang de jinzouyuan” 朝廷和方镇的联络枢纽:试谈中晚唐的进奏院. In Zhengji kaocha yu xinxi qudao: yi Songdai wei zhongxin 政绩考察与信息渠道:以宋代为重心, edited by Deng Xiaonan 邓小南, 235–73. Beijing: Beijing daxue chubanshe, 2008. Wang Jinyu 王金玉. Songdai dang’an guanli yanjiu 宋代档案管理研究. Beijing: Zhongguo dang’an chubanshe, 1997. Wang Lianbin 王联斌. “Songdai bingshu ji qi junshi lunli sixiang” 宋代兵书及其 军事伦理思想. Junshi lishi yanjiu 军事历史研究, no. 2 (1996): 169–75. Wang Qing 汪青. “Bei Song Chizhou ‘Yongfeng jian’ jianzhi kao” 北宋池州“永丰 监”监址考. Chizhou ribao 池州日报, June 24, 2008. Wang Ruilai 王瑞來. “Yu sui haishi wa quan? Song Yuan zhi ji de liang nan jueze” 玉碎还是瓦全？宋元之际的两难抉择. In Zhongguo Nan Song shi guoji xueshu yantaohui—ji Nan Song dingdu Lin’an (Hangzhou) 870 zhounian jinianhui lunwenji 中国南宋史国际学术研讨会—暨南宋定都临安 (杭州) 870周年纪念会论 文集, vol. 2, 259–77. Conference proceedings, Hangzhou, 2008. Wang Yu 王宇. Yongjia xuepai yu Wenzhou quyu wenhua 永嘉学派与温州区域文化. Beijing: Shehui kexue wenxian chubanshe, 2007. Watson, Burton, ed. The Columbia Book of Chinese Poetry: From Early Times to the Thirteenth Century. New York: Columbia University Press, 1984. Watts, John. The Making of Polities: Europe, 1300–1500. Cambridge: Cambridge University Press, 2009. Weber, Max. Economy and Society: An Outline of Interpretive Sociology. New York: Bedminster Press, 1968. Wechsler, Howard. Offerings of Jade and Silk: Ritual and Symbol in the Legitimation of the T’ang Dynasty. New Haven: Yale University Press, 1985. ———. “T’ai-tsung (reign 626–49) the Consolidator.” In Cambridge History of China, vol. 3, Sui and T’ang China, 589–906, pt. 1, edited by Denis Twitchett, 188– 241. Cambridge: Cambridge University Press, 1979. Wells, Peter S. The Barbarians Speak: How the Conquered Peoples Shaped Roman Europe. Princeton: Princeton University Press, 1999. White, Harrison C. Identity and Control: How Social Formations Emerge. Princeton: Princeton University Press, 2008. Wilhelm, Richard. The I Ching. Translated by Cary F. Baynes, Bollingen Series XIX. Princeton: Princeton University Press, 1950. Wilkinson, Endymion Porter. Chinese History: A Manual. Cambridge, MA: Harvard University Asia Center, 2000. Winichakul, Thongchai. Siam Mapped: A History of the Geo-Body of a Nation. ­Honolulu: University of Hawaii Press, 1994. Winkelman, John. “The Imperial Library in Southern Sung China, 1127–1279: A Study of the Organization and Operation of the Scholarly Agencies of the Central Government.” Transactions of the American Philosophical Society, n.s. 64, no. 8 (1974): 61. Wong, R. Bin. “Social Order and State Activism in Sung China: Implications for Later Centuries.” Journal of Sung-Yuan Studies 26 (1996): 229–50. Woodward, C. Vann. “The Jolly Institution.” New York Review of Books, May 2,



bibliogr aphy

497

1974. Also available online at http://www.nybooks.com/articles/archives/1974 /may/02/the-jolly-institution/?pagination=false. Wright, David C. The Ambassadors Records: Eleventh-Century Reports of Sung Embassies to the Liao. Papers on Inner Asia. Bloomington: Indiana University, Research Institute for Inner Asian Studies, 1998. ———. From War to Diplomatic Parity in Eleventh-Century China: Sung’s Foreign Relations with Kitan Liao. Leiden: Brill, 2005. ———. “The Northern Frontier.” In A Military History of China, edited by David A. Graff and Robin Higham, 57–79. Boulder: Westview, 2002. Wu Chengyuan 吴承园. “Songdai cehui shi kao” 宋代测绘史考. Cehui gongcheng 测绘工程 7, no. 1 (1998): 71–79. Wu Xiaoping 吴晓萍. “Huizhu lu yu Wang Mingqing de xueshu chengjiu”《挥麈录》 与王明清的学术成就. Anhui jiaoyu xueyuan xuebao 安徽教育学院学报, no. 4 (1999): 24–25, 28. Wu Yugui 吴玉贵. “Tongdian ‘Bianfang dian’ zhengwu”《通典》“边防典”证误. Wen shi 文史 70, no. 1 (2005): 149–84. Xie Guian 谢贵安. Song shilu yanjiu 宋实录研究. Shanghai: Shanghai guji chubanshe, 2013. Xin Deyong 辛德勇. “Shuo Fuchang shike ‘Yuji tu’ yu ‘Huayi tu’ ” 说阜昌石刻《禹迹 图》与《华夷图》. Yanjing xuebao 燕京学报, no. 28 (2010): 1–44. ———. “Tangdai de dilixue” 唐代的地理学. In Tangdai diyu jiegou yu yunzuo kongjian 唐代地域结构与运作空间, edited by Li Xiaocong 李孝聪, 439–60. Shanghai: Shanghai cishu chubanshe, 2003. Yamauchi Masahiro 山内正博. “Sappu genki to Sō kaiyō” 册府元龟と宋会要. Shigaku kenkyū 史学研究 103 (1968): 20–39. Yan Bai, and Xin Qiji. The Political and Military Thought of Xin Qiji, 1140–1207, with a Translation of His Ten Discussions (Meiqin Shi Lun). Vol. 39, Chinese Studies. New York: Edwin Mellen Press, 2005. Yan Yongcheng 燕永成. “Shilun Wang Mingqing de bu shi chengjiu” 试论王明清 的补史成就. Shixueshi yanjiu 史学史研究, no. 3 (2009): 49–57. Yang Guoyi 杨国宜. “Nan Song jinshu: Beizheng dangyi, Zhian yaoshi” 南宋禁书: 北征谠议、治安药石. Guji yanjiu 古籍研究, no. 2 (1999): 120. Yang Bojun 杨伯峻. Chunqiu Zuo zhuan zhu 春秋左传注. Beijing: Zhounghua shuju, 1981. Yang, Lien-sheng. “Historical Notes on the Chinese World Order.” In The Chinese World Order: Traditional China’s Foreign Relations, edited by John K. Fairbank, 20–34. Cambridge, MA: Harvard University Press, 1968. Yang, Shao-yun. “Reinventing the Barbarian: Rhetorical and Philosophical Uses of the Yi-Di in Mid-Imperial China, 600–1300.” Ph.D. diss., University of California at Berkeley, 2014. Yee, Cordell. “Chinese Maps in Political Culture.” In Cartography in the Traditional East and Southeast Asian Societies, edited by David Woodward and John Brian Harley, 71–95. Chicago: University of Chicago Press, 1994.

498 bibliogr aphy ———. “Taking the World’s Measure: Chinese Maps between Observation and Text.” In Cartography in the Traditional East and Southeast Asian Societies, edited by David Woodward and John Brian Harley, 96–127. Chicago: University of Chi­ cago Press, 1994. Yonemoto, Marcia. “Silence without Secrecy? What Is Left Unsaid in Early Modern Japanese Maps?” Early Modern Japan (2006): 27–39. You Biao 游彪. “Songchao de dibao yu shizheng” 宋朝的邸报与时政. Zhongzhou xuekan 中州学刊 144, no. 6 (2004): 108–11. Yu Rubo 于汝波, and Li Xingbin 李兴斌, eds. Zhongguo jingdian bingshu 中国经 典兵书. Jinan: Shandong youyi chubanshe, 2002. Yu Yingshi 余英时. Zhu Xi de lishi shijie: Songdai shidafu zhengzhi wenhua de yanjiu 朱熹的历史世界:宋代士大夫政治文化的硏究. Taibei: Yunchen, 2003. Zhang, Cong. “To Be ‘Erudite in Miscellaneous Knowledge’: A Study of Song (960–1279) Biji Writing.” Asia Major, 3rd ser., 25, no. 2 (2012): 43–77. ———. Transformative Journeys: Travel and Culture in Song China. Honolulu: University of Hawaii Press, 2011. Zhang Hui 张晖. Songdai biji yanjiu 宋代笔记研究. Wuchang: Huazhong shifan daxue chubanshe, 1993. Zhang Jian 张剑. “Wang Zhi ji qi jiazu shiji kaobian” 王铚及其家族事迹考辨. Wenxue yichan 文学遗产, no. 2 (2012). Accessed May 22, 2012. http://wxyc.literature .org.cn/journals_article.aspx?id=1084. Zhang Minghua 张明华. “Wang Shen kao” 王莘考. Fuyang shifan xueyuan wenxue yuan 阜阳师范学院文学院, no. 3 (2009): 4–9. Zhang Xiaoxu 张晓旭. “ ‘Si da Song bei’ gaishu” “四大宋碑”概述. Wenbo zazhi 文博 杂志, no. 2 (1991): 29–35. Zhang Xiumin 张秀民. Zhongguo yinshuashi 中国印刷史. Shanghai: Renmin chubanshe, 1989. Zhang Zhihe 张志和. “Shiwu jiyuan chengshu yu Mingdai kao” “事物纪原”成书 于明代考. Dongfang luntan 东方论坛, no. 4 (2001): 59–60. Zhang Zhongzhi 张忠智, and Zhuang guiying 庄桂英. “Ren zai shanyin, xin xi biansai—tan Lu You de biansai shi” 人在山阴,心系边塞—谈陆游的边塞诗. Zhongguo yunwenxue kan 中国韵文学刊, no. 1 (2004): 52–56. Zheng Jimeng 郑继猛. “Jinnian lai songdai biji yanjiu shuping” 近年来宋代笔记 研究述评. Gansu shehui kexue 甘肃社会科学, no. 4 (2008): 37–41. Zheng Xianchun 郑宪春. Zhongguo biji wen shi 中国笔记文史. Changsha: Hunan daxue chubanshe, 2004. Zheng Xihuang 郑锡煌. “ ‘Jiuyu shouling tu’ yanjiu” “九域守令图”研究. In Zhongguo gudai ditu ji: Zhanguo–Yuan 中国古代地图集:战国–元, edited by Cao Wanru 曹婉如, 35–40. Beijing: Wenwu chubanshe, 1990. Zhonghua wenhua fuxing yundong tuixing weiyuanhui Siku quanshu suoyin bianzuan xiaozu 中華文化復興運動推行委員會四庫全書索引編纂小組, ed. Siku quanshu wenji pianmu fenlei suoyin—xueshuwen zhi bu 四庫全書文集篇目 分類索引—學術文之部. Taibei: Taiwan shangwu yinshuguan, 1989. Zhou Guangpei 周光培. Songdai biji xiaoshuo 宋代笔记小说. 24 vols. In Lidai



bibliogr aphy

499

biji xiaoshuo jicheng 歷代笔记小说集成. Shijiazhuang: Hebei jiaoyu chubanshe, 1995. Zhou Lun 周纶, and Diao Zhongmin 刁忠民. Zhou Yiguo Wenzhong gong nianpu 周益国文忠公年谱. Chengdu: Sichuan daxue chubanshe, 2003. Zhu Chuanyu 朱傳譽. Songdai xinwen shi 宋代新聞史. Taibei: Zhongguo xueshu zhuzuo jiangzhuweiyuanhui, 1967. ———. Xian Qin Tang Song Ming Qing chuanbo shiye lunji 先秦唐宋明清傳播事 業論集. Taibei: Taiwan shangwu yinshuguan, 1988. Zhu Dongrun 朱东润. Lu You zhuan 陆游传. Beijing: Zhonghua shuju, 1960. Zhu Shangshu 祝尚书. Song ren bieji xu lu 宋人别集叙录. 2 vols. Beijing: Zhonghua shuju, 1999. Zou Fuqing 邹福清. Tang Wudai biji yanjiu 唐五代笔记研究. Beijing: Zhongguo shehui kexue chubanshe, 2013.

Index Note: Page numbers in italics indicate illustrations or tables. Abbasid Empire, 4 Abramson, Marc, 238, 277 administrative zoning, 243–49, 244, 246, 247 An Lushan, 241 Anderson, Benedict, 27 archival mentality, 35–44, 49–59, 70–72, 329 archives, 430–31; classification methods of, 59, 88; excerpts from, 59–62, 64–70; indexing methods of, 64, 68–69, 196– 97, 200; information networks of, 37– 44, 40; local, 39, 47–48, 53; in medieval England, 35–36; operation of, 37–44, 40, 49–59; secrecy of, 45–48, 65, 76, 80– 87; selecting materials for, 39. See also libraries atlases, 123, 149–50; historical, 125–36, 218–19; printing technology and, 122. See also maps Baichuan xuehai (All streams of the sea of learning), 292–93 Ban Gu, 241 Bayly, C. A., 22n31 Beizheng dangyi (Sincere discussions of the northern campaign), 249–56, 259– 68, 263, 264. See also Hua Yue Berry, Mary Elizabeth, 22n31, 25–27 bian (mark, distinguish), 128 bianfang (border protection), 234, 243. See also border affairs biji (written notes), 30, 281–326; authors of, 307–11, 308, 309, 310; database of, 290–91, 298, 332, 353; distribution of, 291–96, 292, 294, 295, 297; genres of, 283–

85, 325, 331–32, 369n34, 378; printers of, 295, 296–307, 297, 299; publishing networks of, 312–24, 319; terms associated with, 284–85, 373 Bin Wong, 378 biographies, 375 Bishui qunying daiwen huiyuan (Epitome of eminent men responding at the Imperial College), 64, 73, 183, 185n33; policy debates using, 194, 197, 198, 199 Black, Jeremy, 137 Blair, Ann, 59 Board of Punishments (Xingbu), 81 Bol, Peter, 114n20, 133n68, 396n3 Boodberg, Peter, 168–69 border affairs, 167–70, 212–23, 231–32; administrative zoning of, 243–49, 244, 246, 247; ethnocyclopedic writing on, 234–43; examination questions on, 173–201; local knowledge and, 268–74, 271; military chorography of, 259–68, 263, 264; representing ethnic others and, 395–426 Boswell, James, 327 British Empire, maps of, 109n4, 152–53 Buddhism, 15, 121, 224, 286 bureaucracy, 367–68, 430–31; administrative reports of, 223–31; archival mentality of, 35–44, 49–59, 70–72, 329; deities organized as, 224; factionalism in, 92– 95, 103; functions of, 35; information flow in, 77–104; lexicons of, 80; Weber on, 25 Bureau of Military Affairs (Shumiyuan), 42, 44, 80, 415–16 Burke, Peter, 326

502 index cadastral surveys, 26, 107 Cai Jing, 330, 385–86 Cai You, 327 Cai Yuanding, 79, 90, 101 Cao Fan: Xuzhi guojing (What one needs to know—The mirror of the state), 224 Cao Song, 144–47 Capital Liaison Offices (Jinzouyuan), 37, 79 Cartier, Carolyn, 367 cartography. See maps Catalog of the Pavilion of Following One’s Original Intent (Suichu tang shumu), 50, 59–60, 213, 226, 460–466. See also You Mao CBDB. See China Biographical Database Project Cefu yuangui (Models from the archives), 39, 60 censorship, 11–12, 250, 304, 429, 433–34; errors in, 93–94. See also “secrecy” of documents Chaffee, John, 24 Chancellery (Menxia sheng), 38, 80. See also Secretariat-Chancellery Chao Gongwu, 50–53, 59; envoy reports of, 214; maps mentioned by, 221; Zhao Xibian’s supplement to, 60; Junzhai dushu zhi (Junzhai’s record of books read), 50, 213–16, 450–66 Chao Gongzu, 156, 157 Chao Yuezhi, 100 chaobao. See court gazettes Chen Chun, 101 Chen Fuliang: Lidai bingzhi (Military ­institutions through the ages), 210 Chen Guan: Guoyong xuzhi (What one needs to know about the state budget), 227 Chen Liang, 204 Chen Qiqing, 225 Chen Shan: Menshi xinhua (New explanations bluntly presented), 288, 303–4 Chen Yinglong, 153–54, 256 Chen Yingxing, 301

Chen Zhensun, 54, 65–68, 229; career of, 53; maps mentioned by, 135n72, 205–6, 215–16, 218–19, 221; Zhizhai shulu jieti (Zhizhai’s annotated catalog), 50–51, 59, 213, 214, 450–66 Cheng Dachang: Beibian bei dui (Notes for a response on questions about the northern border), 276–77; Yanfan lu (Elaborating on “Rich dew”), 301; Yugong lun (Discourses on “The tribute of Yu”), 301; “Yugong” shanchuan dili tu (Maps of the topography and geography in “The Tribute of Yu”), 122, 123, 124 Cheng Duanli, 124 Cheng Gongxu, 99 Cheng Jiong, 341 Cheng Kejiu, 314–15, 318–20, 319 China Biographical Database Project (CBDB), 307–11, 310, 333n14, 353, 360 Chizhou Yongfeng qianjian xuzhi (What one needs to know about the cashproducing county Yongfeng in Chi­ zhou), 225 Chu Ping-tzu, 19 citation networks, 352–68; frequency of informants in, 335–52, 336, 344–51; geographic distribution of, 367–68; ­location of informants in, 360–67, 361–65; officeholding informants in, 353–60, 354, 356–58; titles quoted in, 369–78 civil service examinations, 46–47, 430–31; historical reasoning in, 189–201, 198– 200; manuals for, 62–64, 71; maps and, 124–25; military affairs and, 169, 179– 88; policy questions on, 172–79; procedures for, 62–63, 72; scholar-collectors and, 56–57 collected statutes (huiyao), 40, 43–44, 52, 60–63 Collected Statutes Office (Huiyao suo), 44 collected writings (wenji), 190, 290, 299, 373, 375–76

index 503 Compilation Bureau (Bianxiuyuan), 42 Comprehensive institutions (Tong dian), 234–36, 240–42, 250, 274 Confucianism, 257, 326; NeoConfucianism and, 18, 393, 428 corpus linguistics, 21, 397n4 court diaries (qijuzhu), 40, 41; as bibliographic subdivision, 59; in private collections, 52; of Zeng Bu, 55 Court Diary Office (Qijuyuan), 40, 41, 42 court gazettes (chaobao), 38–39, 76–104, 149–50; distribution of, 81–82, 102–4; literati networks and, 88–104; material in, 77, 78; poems of, 91–99; sale of, 83, 85; similarity to newspapers, 81n16, 101–2 Dai Fugu, 92 Daily Calendar Office (Rilisuo), 42 daily calendars (rili), 40, 41–42, 52 Daoism, 15, 223–24, 286 Deng Xiaonan, 71 Deng Yu, 111 de Vivo, Filippo, 22n31, 23 Di Cosmo, Nicola, 31, 234–35 Di Ku, 127 “Dili tu” (Map of the administrative ­organization of the empire), 119 Dong You, 327 Draft History Institutes (Guoshiyuan), 42 Du You, 275; Tong dian (Comprehensive institutions), 234–36, 240–42, 250, 274 Duan Guangyuan, 410–12, 414 Duyi xuzhi (What you need to know when reading “The Changes”), 229 dynastic histories (guoshi), 40, 43, 47, 370; as bibliographic genre, 59–60 Edgren, Søren, 305 Edney, Matthew, 144, 152–53 Eisenstadt, Shmuel N., 6 Eisenstein, Elizabeth, 122 empire, 2–6; diplomatic rhetoric of, 397– 98; rhetoric of, 408–19, 425–26 empire maps, 107–64, 138, 142, 201–12,

218–23; British, 109n4, 152–53; poems about, 144–61; prefectural maps and, 139, 208, 221, 222, 263; reading of, 125– 36, 143–61; of Xiaozong, 210–11; Xue Jixuan on, 148; of Yuanfeng reign, 139, 243, 244 encyclopedias (leishu), 60–63; Peter Bol on, 133n68; ethnographic, 234–42, 274; of Zhang Ruyu, 132–34 England: empire maps of, 109n4, 152– 53; medieval archives of, 35–36; ­seventeenth-century elites of, 320– 21, 327 envoys, 171; exchange of, 73, 192–93; ­reports of, 214–18, 223 Epitome of eminent men responding at the Imperial College. See Bishui ­qunying daiwen huiyuan Er Yang guichao lu (Yang brothers’ return to the Song court), 214 ethnic groups, 144–47; descriptive terms for, 396–408, 399, 402, 403, 406, 407, 416–17; intermarriage among, 239–40; maps of, 112, 117, 120, 129–30, 138, 139– 40, 163, 237; noninteraction of, 193, 238; reports on, 213–18, 236–37; representations of, 395–426 ethnocyclopedic writing, 234–42, 274 excerpting methods, 59, 68 Fan Chengda, 218 Fan Chong, 334 Fang Hui, 93 Fang Yue, 89, 150, 157 Fang Zi, 338 Fei Gun, 124 First Emperor’s tomb map, 110, 118 Five Mountains, 131, 429 Fozu tongji (Comprehensive account of Buddhist patriarchs), 121 Franke, Herbert, 31 Frankel, Oz, 26, 27 Fu Jian, 73 Fujian taozei xuzhi (What one needs to know about banditry in Fujian), 226

504 index Gao Xianzhi (Go Seon-ji), 241 Gao Zhu, 99–100 Gaoli state, 398 Gaozong, Emperor, 1, 56, 68, 382–83; daily calendar of, 52; Jin policies of, 74, 85–86, 191–92, 195, 214; Qin Gui and, 85–86, 415–16, 424; sagely government of, 71–72; veritable records of, 50, 322; on Zhang Jiucheng, 386 Gaozu, Emperor, 256–57 gazetteers, 114–15, 149–50, 215–23, 431; county, 221, 222, 263, 264; of Ni Pu, 202–3; prefectural, 139, 208, 221, 222, 246, 263, 264; of Wu Xie, 205; of Yuanfeng reign, 220 Geshu Han, 241 Gong Feng, 92 Gong Rizhang, 249–50, 252–53 Gong Yizheng, 286 Gowers, Bernard, 19–20 Great Wall, 119, 127, 129–30, 139, 243 Guangwu, Emperor, 111 Gujin hebi she lei bei yao (Full essentials of the past and the present brought t­ogether and classified), 62 “Gujin huayi quyu huyao tu” (General survey map of Chinese and non-­ Chinese from the Past through the Present), 138, 139–40, 163 Gujin shiwen leiju (Classified collection of affairs and texts from the past through the present), 60–61 Guo Jiude, 318, 319 Guo Ruoxu, 67 Han Shizhong, 74 Han Tuozhou, 251, 253, 265 Handy geographical maps throughout the ages. See Lidai dili zhizhang tu Han dynasty, 3, 256–57; history of, 115– 16, 257; maps of, 108, 111–12 Hartwell, Robert, 190–91 He Yi, 301–3 historical reasoning, 189–201, 198–200

Historiography Institute (Shiguan), 41, 42, 51, 55, 70, 281 History Institutes (Shiyuan), 40 Hong Ji, 302–3 Hong Mai, 77–78, 373; citations of, 341; Rongzhai suibi (Rongzhai’s jottings), 302–4; Yijian zhi (Record of the listener), 301–2 Hong Zikui, 185 Hu Anguo, 89 Hu Quan, 1–2, 402, 406, 407, 415–20 Hu Zhonggong, 157–59 Hua Xia people, 237 Hua Yue, 249–58, 268–77; Beizheng dang­yi (Sincere discussions of the northern campaign), 259–68, 263, 264; Beizheng lu (Notes for a northern campaign), 242, 252–53; “Ten Policy Essays on Pacifying the Barbarians,” 252–54; Zhian yaoshi (Remedies for achieving security), 249–56, 267–72 Huang Dui, 253 Huang Gan, 90 Huang Kuanchong, 265 Huang Shang, 8 Huang Shicheng, 410, 412 Huang Zhuangyou, 300 Huizhu lu (Waving the duster), 30–31, 52, 70, 332–33, 467–68; citation network of, 335–39, 336, 341–43, 344–46, 348–49; descriptions of Jurchens in, 396–408, 399, 402, 403, 406, 407, 416–17; location of informants in, 362, 363, 366; officeholding informants of, 353–60, 354, 356–58; publication history of, 288, 312–25, 334; representing ethnic others in, 395–426; titles quoted in, 369–72, 374–78; topic map of, 378–92, 380–81. See also Wang Mingqing Huizong, Emperor, 385–86, 417 Hunyi nei wai jiangyu tu (Maps uniting the territories in the core and on the periphery), 122 Huo Qubing, 241

index 505 imperial sovereignty, 382–86 indexing methods, 64, 68–69, 196–97, 200 India, 131 information: within bureaucracies, 77– 104; court gazettes and, 77–104; definitions of, 24–26; editorial policies and, 77, 80; leaking of, 82–87 information networks, 37–44, 40 Jacob, Christian, 122 Japanese libraries, 25–27 Jay, Jennifer, 18, 422n60 Jia Dan, 139–40; “Hainei huayi tu” (“Map of Chinese and non-Chinese with the seas”), 117, 129, 144–47 Jiang Shaoyu: Songchao shishi leiyuan (Classified collection of affairs from the reigning dynasty), 373–74 Jiang Teli, 91 Jiang Xian, 94–95 Jin Empire, 14, 159n109, 281, 408–26; border conflicts with, 258–62, 265–66, 276; ethnic groups in, 400; Gaozong’s policies on, 74, 85–86, 191–92, 195, 214; maps of, 118–19, 130, 135–36, 216; Qin Gui’s negotiations with, 85–86, 192, 193, 412–15; reports on, 213–18; urbanization of, 427. See also Jurchens Jing Du, 53 Jing Tang, 304, 323 Jingkang crisis, 5, 13–14, 430, 436–38 Jinguo Cheng’an xuzhi (Things to know about the State of Jin during the Cheng’an reign), 216 Jinren jiangyu tu (Maps of the territory of the people of Jin), 216 Jinzouyuan. See Memorials Office Jiuchao guoshi (Draft history of the nine reigns), 43 Jiuyu zhi (Gazetteer of the nine zones), 149–50 Johnson, Samuel, 327 Junzhai dushu zhi (Junzhai’s record of

books read), 50, 213–16, 450–66. See also Chao Gongwu Jurchens, 174; descriptive terms for, 396– 407, 399, 402, 403, 406, 407, 416–17; ethnonyms of, 399–400; invasions of, 260; Manchus and, 405; reports on, 213–18. See also Jin Empire Kaifeng, 159; Jurchen occupation of, 13, 49, 410 Keeping guests (Tou xia lu), 333, 337, 369, 377 Khitans, 7, 73–74, 130, 241, 413. See also Liao Empire Korea, 130, 241 Kubota Kazuo, 81n16 Lamouroux, Christian, 215, 260 Lang Jian, 328 Learning of the Way movement, 79, 90 Left Storehouse (Yufu), 66 Lewis, Mark Edward, 15 Li Daochuan: Jiangdong shi kao (Ten historical investigations into the region east of the Yangzi), 253, 275–76 Li Dongyou, 219 Li Hewen, 408–9 Li Hou, 316–18, 319 Li Ruqian: Dongyuan congshuo (Ongoing conversations from Master Dongyuan), 302 Li Shunchen: Jiangdong shi jian (Ten historical cases on the region east of the Yangzi), 253, 275–76 Li Tao, 63–64, 316–17, 319; Xu Zizhi tongjian changbian (Collected documents for the continuation of “The comprehensive mirror for aid in government”), 78 Li Xinchuan, 51; Guochao huiyao zonglei (Collected statutes of the reigning dynasty comprehensively excerpted and classified), 44, 65 Li Yannian, 330–31

506 index Li Yuangang: Houde lu (Record of enriching virtue), 374 Li Zhengzhi, 384 Li Zhi, 92 Li Zhuan, 389 Li Zunxu, 408–9 Liang Caifu, 54 Liangchao guoshi (Draft history of the two reigns), 43, 328, 370 Liangzhe zhuanyun xuzhi (What one needs to know about fiscal contributions from Liangzhe), 226 Liao Empire, 7, 73–74, 159–60, 243, 413, 420–21; border fortifications of, 248; descriptive terms of, 397–98, 416–17; maps of, 130, 135–36, 139, 216; reports on, 213–18; Yellow River and, 260 libraries, 49–59, 88–101, 213–18; Japanese, 25–27. See also archives Lidai dili zhizhang tu (Handy geographical maps throughout the ages), 122–27, 131–37, 143, 155, 162–63, 212; ethnocyclopedic writing and, 243; poems on, 148–51, 157–59 Lidai zhidu xiangshuo (Detailed explanations of institutions throughout the ages), 64 Liji (Record of Ritual), 405–6, 411 Lin Fu: Cichang xin fan (New model for the examinations), 46, 47; Shenzong huangdi zhengji gushi (Truths regarding the government of Emperor Zhenzong), 46–47, 66 Lin Jiong, 63, 64 Lin Xi, 46 literati, 367–68, 390–92; court gazettes and, 88–104; discourse of, 98n59, 326– 31; empire maps and, 144–61, 218–23, 222; imperial elites versus, 5, 10; imperial mission of, 14–18, 107; notebooks of, 281–89, 293, 294; private libraries of, 49–59, 88–101; in seventeenth-century England, 320–21, 327 Liu Changshi, 297; Lupu biji (Notes written in Lupu), 286–89, 303–4, 316 Liu Chenggui, 227

Liu Guangzu, 90 Liu, James T. C., 17 Liu Jing, 241 Liu Kezhuang, 93–96, 152, 161n114, 163–64 Liu Kuang, 275; Wu zhi (Military guide), 237–42 Liu Qi, 330–31 Liu Yeqiu, 285 Liu Yi, 74 Liu Yu, 119, 417 Liu Zhaoyou, 61 Liu Zheng, 91 Liu Zhi, 330 Liu Ziyu, 185n33 Liuhe zhangyun tu (Maps holding ­together the passage of time in the world), 218–19 Liujing tu (Figures of the six classics), 121, 219 Lizong, Emperor, 43–44 local knowledge, 234, 268–74, 271 Lou Yue, 304n37 Love, Harold, 312, 313, 320 loyalty (zhong), 17–18, 412, 414, 419–25 Lu You, 27n42, 68–69, 256; on civil ­service examinations, 72; on Fan Chengda, 218; on maps, 148–51, 154– 55, 161, 163–64 Lu Ziyu, 296 “Luguo zhi tu” (Map of the state of Lu), 120 Lü family, 387 Lü Xiaqing: Tangshu xinli xuzhi (What you need to know about “The new form of ‘The New Tang History’”), 229 Lü Zuqian, 373; Han yudi tu (Han imperial maps), 111; Wudai shi xiangjie (Detailed excerpts from “The history of the Five Dynasties”), 121 Ma Duanlin: Wenxian tongkao (Com­ prehensive investigation of textual ­records), 274 Ma Shen, 420 manuscripts: copying of, 64–65, 312–15,

index 507 317–25; European, 19–20, 36, 59, 321, 323 maps, 107–64, 218–23; of border defenses, 246, 263, 264, 271; chronological information on, 136–43, 138, 142; of ethnic groups, 112, 117, 120, 129–30, 138, 139– 40, 163, 237; Great Wall on, 119, 127, 129–30, 139; making of, 109–15, 140, 161–62, 201–12; monumental, 110, 113; political meaning of, 109–10; reading of, 125–36, 143–61; on stelae, 117–20, 209; as tool of collective memory, 29; of Wei dynasty, 141–43, 142. See also empire maps May, Ernest R., 189–90 Meiqin shi lun (Ten expositions humbly presented), 253–54, 276 Memorials Office (Jinzouyuan), 79–80, 84, 281, 431; centralization of, 37–39; court gazettes published by, 80–81; leaks from, 83–84; “short reports” of, 86–87 Meng Jiao, 92 Mengxi bitan (Notes by Dream Creek), 114, 298, 300–301, 372, 378. See also Shen Gua methodological concerns, 378n46, 396n4, 467–68 Military Affairs Bureau (Shumiyuan bian­ xiuguan), 42, 44, 80, 415–16 military chorography, 259–68, 263, 264 military geography, 179–88, 182, 187, 212 military studies, 168–70, 209–10, 212, 237–42, 274–78. See also Wujing zongyao Ming dynasty, 1, 18, 437; court gazettes of, 81, 103–4 Ministry of Military Affairs (Bingbu), postal network of, 80–81, 83, 84, 219, 273 Mo Shuguang, 57 Mongol Empire, 94–96, 437; Jurchen con­ flicts with, 174; Song policies toward, 192 Münkler, Herfried, 14–15, 437–38

Nazi xenophobia, 405n22 Neizangku xuzhi (What one needs to know about the Palace Storehouse), 227 Neo-Confucianism, 18, 393, 428 Neustadt, Richard E., 189–90 New record from the Jade Shine Studio (Yuzhao xin zhi), 333, 337, 369, 377 Ni Pu, 183, 202–6, 209–10; Yudi huiyuan zhi (Gazetteer bringing the empire together), 202–3 Ningzong, Emperor, 8, 53, 119, 250 notebooks. See biji Notes by Dream Creek. See Mengxi bitan Ottoman Empire, 4 Ouyang Xiu, 17, 338 Owen, Stephen, 374 Palace Library (Bishusheng), 42–44; private collectors and, 55–56; You Mao and, 52 Pan Jingxian, 203, 204 peace negotiations, 73–74, 191–92; Hua Yue on, 255–56, 258; rejection of, 402, 415–19, 426; Wang Mingqing on, 408–26 Pei Xiu, 139–40 Peng Guinian, 91, 101 pharmacopeia, 377 poetry, 60–61, 91–99, 375–76 Poo Mu-chou, 398 Poon, Ming-sun, 60n45 population censuses, 107 postal network, 80–81, 83, 84, 219, 273 precious instructions compilations (baoxun), 40, 44, 46, 65–66 print technology, 13, 36, 64–70, 83, 140, 320–21 Pure Brightness (Jingming) sect of Daoism, 223–24 Qian Chen, 391 Qiandao Xu sichao huiyao (Qiandao reign edition of the collected statutes of the four reigns), 63

508 index Qin Gui, 1, 42, 57, 305, 379, 386; death of, 86, 415; Hu Quan on, 418; Jin negotiations by, 85–86, 192, 193, 412–16; legacy of, 420–21, 424 Qin Xi, 57 Qin Empire, 23 Qing dynasty, 1, 18, 405; court gazettes of, 81, 103–4 Qingyuan tiaofa shilei (Classified laws of the Qingyuan period), 45–46, 65, 170, 171 Qinzong, Emperor, 50, 420–21; as Jin hostage, 409–12, 418, 424 Qiu Su, 283, 303 Qunshu huiyuan jie jiang wang (Net to unite and order the massive amounts of information in all books), 64 Record of hearsay from Su River. See S­ushui jiwen Records of official travel. See Youhuan jiwen Record of the Origins of Things (Shiwu jiyuan), 60, 62–63 Ren Shenxian, 420–21 Renzong, Emperor, 6, 408–10; Cefu yuangui (Models from the archives) and, 39, 60; dynastic history of, 43; peace negotiations of, 73, 74; precious instructions of, 46; veritable records of, 46 reports, 223–31; of envoys, 214–18, 223; short, 84–89, 91, 97, 98n59, 431 Roman Empire, 15 Rossabi, Morris, 397 Ruxue jingwu (Insights into the learning of scholars), 292, 301n30, 435–36 sagely government compilations (shengzheng), 40, 44, 65–67, 196; of Gaozong, 67, 71–72; of Tianxi reign, 66; of Xiao­ zong, 62, 67, 71–72 Sanchao guoshi (History of the three reigns), 370 Sanchao xunjian tu (Illustrations of the

instructions and models from the three reigns), 66–67 Sanchao zhengyao (Essentials of government from the three reigns), 67 “secrecy” of documents, 45–48, 65, 76, 80–87; censorship and, 433–34; imperial authority and, 232; in Tang lü, 170– 71. See also censorship Secretariat-Chancellery (ZhongshuMenxia), 41. See also Chancellery Shao Yong, 95 Shen Gua, 109n5, 113–14; Mengxi bitan (Notes by Dream Creek), 114, 294, 298, 300–301, 372, 378 Shen Songqin, 92–93 shengzheng. See sagely government compilations Shenzong, Emperor, 3, 169, 371 Shi Jingtang, 159n109 Shi Miyuan, 94, 251 Shilei fu (Classified prose poems), 60–61 shilu. See veritable records Shisan chao huiyao (Collected statutes of thirteen reigns), 51 Shiwu jiyuan (Record of the origins of things), 60, 62–63 short reports, 84–89, 91, 97, 98n59, 431 Shui Anli, 125, 149, 150, 205 Shujing (Book of documents), 115–16 Sichao guoshi (Draft history of the four reigns), 51 Sikong Tu, 147 Siku quanshu series (SKQS), 405, 406 Sima Guang, 3, 6–9, 17, 437; Zizhi tongjian (The comprehensive mirror), 6–7. See also Sushui jiwen Sima Qian, 7, 146n81; on First Emperor’s tomb map, 110, 118; Records of the Historian, 235 Sincere discussions of the northern campaign (Beizheng dangyi), 249–56, 259– 68, 263, 264. See also Hua Yue Sixteen Prefectures, 119, 136, 139, 153, 159– 61, 172; recapturing of, 420–21, 429

index 509 Skinner, William, 20–21, 367–68 Song Huizhi, 390 Song Lian, 204 Song Minqiu, 88–89 Spacks, Patricia Meyer, 327 Standen, Naomi, 17–18, 277 stelae, 8–9; maps on, 117–20, 209 Su Shi, 125, 329–30; citations of, 338, 341 Su Yan, 339 Suichu tang shumu (Catalog of the Pavilion of Following One’s Original Intent), 50, 59–60, 213, 226, 460–66. See also You Mao Sui shu (Sui dynastic history), 61 Summa of the military classics. See Wujing zongyao Sun Jin, 418 Sun Quan, 194–95 Sun Wei, 247 Sun Yingshi, 89 Sushui jiwen (Record of hearsay from Su River), 333, 334; citation network of, 339, 340; location of informants in, 360– 66, 361; officeholding informants of, 353–55, 354, 357, 359–60; titles quoted in, 370–71, 377–78. See also Sima Guang Taishang jingming yuan buzou zhiju Tai­ xuan dusheng xuzhi (Amended list of ranks and bureaus from the Pure Brightness Directorate of the Highest), 223–24 Taizong, Emperor, 71, 383; military skill of, 193, 195 Taizu, Emperor, 71, 195, 370, 411; maps of, 112, 137–39; veritable records of, 50 Talas, battle of, 240, 241 Tang Jie, 391 Tang Xiunian, 300–301 Tang Zhongyou: Diwang jingshi tupu (Maps and tables illustrating the governance of rulers), 121, 219 Tang Empire, 10; draft histories of, 43; maps of, 117

Tang lü (Tang code), 170–71 Tianxi shengzheng ji (Record of the sagely government of the Tianxi reign), 66 Tibetans, 130 Tillman, Hoyt Cleveland, 27n42 Tong dian (Comprehensive institutions), 234–36, 240–42, 250, 274 Tou xia lu (Keeping guests), 333, 337, 369, 377 “Tribute of Yu.” See “Yugong” Tuan, Yi-Fu, 154, 157 Turks, 235–36 Uighurs, 235–36 Unrestrained jottings from the Cloudy Mountain Foot (Yunlu manchao), 297, 303–4, 316 veritable records (shilu), 42–43, 370; as bibliographic subdivision, 59; of Renzong, 46 Veritable Records Institute (Shiluyuan), 42, 44, 322 Vietnam, 131 Waldron, Arthur, 129 Wang Anshi, 50–51, 328, 386 Wang Bing, 422–23 Wang Fu, 331 Wang Fuzhi, 92–93 Wang Gungwu, 31 Wang Hung-tai, 103 Wang Jun, 421 Wang Lun, 415–18 Wang Mai, 94–95 Wang Mingqing, 54, 57, 283, 392–94, 420; on book collections, 64–65; on Jin peace negotiations, 408–26; social background of, 333–34; Tou xia lu (Keeping guests), 333, 337, 369, 377; Yuzhao xin zhi (New record from the Jade Shine Studio), 333, 337, 369, 377. See also Huizhu lu (Waving the duster)

510 index Wang Qia: Xi Han bingzhi (Military institutions of the Western Han), 210 Wang Qiu, 318, 319 Wang Shen, 390 Wang Xiang, 403 Wang Xixian, 209, 220; Huangchao fangyu zhi (Gazetteer of our dynasty’s territory), 205 Wang Xuanzan, 330–31 Wang Yande, 416–17 Wang Yingchen, 288 Wang Yinglin, 9 Wang Yuxi, 319 Wang Zhi, 54, 323, 333, 339; draft war ­announcement of, 402–5 Wang Zhiwang, 97 Wang Zhiyuan, 8, 119 Wang Zongdan, 304 Waving the duster. See Huizhu lu Weber, Max, 25 Wei Liaoweng, 173–74, 176, 177, 180–83 Wei dynasty map, 141–43, 142 wenji (collected writings), 190, 290, 299, 373, 375–76 Wenxiang (monk), 150–51 Wright, David, 31 Wu, Emperor, 3, 111, 238–39 Wu Baogui, 236n6 Wu Cheng, 52, 64 Wu Qiao, 144, 145, 147 Wu Shidao, 202 Wu Xie, 209, 220; Yunei bian lidai jiangyu zhi (Gazetteer marking our territories through the ages), 204 Wu Yong, 175, 178, 211 Wu Zeng, 88, 305; Nenggai manlu (Nenggai’s unrestricted records), 304 Wujing zongyao (Summa of the military classics), 29, 243–45, 244, 246, 248, 250, 261. See also military studies “Xiangfu jiuyu tu” (Map of the nine zones from the Xiangfu period), 148–49 xiaoshuo (short narratives), 284

Xiaozong, Emperor, 68, 193, 384; collected statutes of, 62, 63; court diary of, 52; defense policies of, 276; maps of, 112; reunification plans of, 210–11; sagely government of, 62, 71–72; veritable records of, 50 Xie Ao, 120 Xie Caibo: Mizhai’s Written Notes, 302, 304 Xie Weixin, 62, 63n50, 64 Xin Deyong, 117n27, 209 Xin Qiji, 161n114, 253 Xiongnu people, 238–40 Xiongnu xuzhi (Things to know about the Xiongnu), 216, 231 Xi Xia, 243; fortifications of, 245–49, 247; maps of, 131, 139 Xi Xia xuzhi (What one needs to know about Xi Xia), 229 Xu Du, 57, 328, 329, 338 Xu Fu, 55 Xu Kui, 410–11 Xu Mengxin: Sanchao beimeng huibian (Collection of documents regarding ­relations with the north during three reigns), 214 Xu Xun, 223–24 Xu Zhong, 58 Xue Jixuan, 148, 208–9; Jiuyu tu (Map of the nine regions), 207; Jiuzhou tuzhi (Map gazetteer of the empire), 208 xuzhi (“what one needs to know”) genre, 216, 223–31, 460–66 xylography, 320–21. See also print technology Yan You, 238 Yanbei zalu (Miscellaneous notes on the north), 216 Yang Guifei, 241 Yang Guozhong, 241 Yang Xiong, 92 Ye Mengde: Shilin yanyu (Shilin’s casual conversations), 288 Yingzong, Emperor, 43, 46, 54

index 511 yishi (memoirs), 373, 408–9 Yixing (monk), 126, 150 You Jiuyan, 420 You Mao, 50–53, 123; envoy reports of, 214; maps mentioned by, 216, 219, 221; Suichu tang shumu (Catalog of the ­Pavilion of Following One’s Original Intent), 50, 59–60, 213, 226, 450–66 Youhuan jiwen (Records of official travel), 333–35, 432; citation network of, 339– 43, 342, 347, 350–51, 352; location of informants in, 364, 365, 366–67; officeholding informants of, 353–60, 354, 357, 358; titles quoted in, 369–71, 377–78 Yu Yingshi, 92–93 Yu Zhe, 205 Yuan Xie, 192–93 Yuan dynasty, 1 Yuanfeng reign, 46; empire map of, 139, 243, 244; gazetteer of, 220 “Yuan Wei Beiguo tu” (Map of the Wei dynasty and the Northern states), 141– 43, 142 yudi tu (maps of the earth), 111, 112, 120 Yue Fei, 74, 218, 421 Yue Ke, 296–97 “Yugong” (Tribute of Yu), 16, 115–16, 122, 301; Wenxiang on, 150–51; Xin Deyong on, 209; Zhu Xi on, 136 “Yugong” shanchuan dili tu (Maps of the topography and geography in “The Tribute of Yu”), 122, 123, 124 “Yuji tu” (Map tracing the tracks of Yu), 115–16, 118–19, 163 yulu (recorded conversations), 373 Yunlu manchao (Unrestrained jottings from the Cloudy Mountain Foot), 297, 304, 316 Yuzhao xin zhi (New record from the Jade Shine Studio), 333, 337, 369, 377 zajia (miscellaneous expertise) genre, 284 Zeng Bu, 389; court diary of, 55 Zeng Gong, 388 Zeng Jian, 194; Dace mijue (Secret tricks

for responding to policy questions), 178–79 Zeng Shu, 337–38 Zengru mingru jiangyi huang Song Zhongxing liangchao shengzheng (Sagely government of the two reigns of the intermediate restoration, with explanations by famous scholars), 196, 197 Zengru mingru jiangyi Zhongxing liang­ chao shengzheng (Sagely government of the two reigns of the intermediate restoration, with explanations by famous scholars), 68 Zhan Tiren, 90 Zhang Cong, 334–35 Zhang Congzu, 44, 51 Zhang Hao, 383 Zhang Hui, 309–10 Zhang Jiucheng, 386 Zhang Jun, 74 Zhang Lei: Mingdao zazhi (Clarifying the Way: A miscellany), 396n3 Zhang Ruyu: Qunshu kaosuo (Investigations into multitudes of books), 132–35, 162 Zhang Shi, 211 Zhang Shinan, 432. See also Youhuan ­jiwen (Records of official travel) Zhang Yong, 184 Zhang Yuangan, 151–52, 155–61, 256 Zhang Zi: Huangchao shixue guifan (Models for official service and learning from the reigning dynasty), 372– 74, 376 Zhao Bian, 184 Zhao Bujian, 315, 318–20, 319 Zhao Ding, 334 Zhao Fan, 92 Zhao Li, 416 Zhao Sheng: Chaoye lei yao (Classified ­essentials in and out of court), 80 Zhao Yanwei: Yunlu manchao (Unrestrained jottings from the Cloudy Mountain Foot), 297, 304, 316 Zheng Boxiong, 203, 204

512 index Zheng Bu, 55 Zheng Ning, 95–96 Zheng Qiao, 54; Tong zhi, 227–28, 460–65 Zheng Ruxie, 215 Zheng Yin, 54 Zhenguan zhengyao (Essentials of government of the Zhenguan reign), 66 Zhenzong, Emperor, 39, 55, 80, 184, 384, 406; Lin Fu’s compilation for, 46–47, 66; peace negotiations of, 73–74 Zhezong, Emperor, 50–51, 78, 386 Zhian yaoshi (Remedies for achieving ­security), 249–50 Zhihe fayun cha yan xuzhi (What one needs to know about the tea and salt tribute, from the Zhihe period), 226 Zhizhai shulu jieti (Zhizhai’s annotated catalog), 50–51, 59, 213, 214, 450–66. See also Chen Zhensun Zhonghe chongxiu guochao huiyao (Collected statutes of the reigning dynasty revised during the Zhenghe reign), 58 Zhongxing huiyao (Collected statutes of the intermediate revival), 63

Zhongxing liangchao shengzheng (Sagely government of the two reigns of the intermediate restoration), 196, 199 Zhongxing sichao guoshi (Draft history of the four reigns of the intermediate revival), 43 Zhou Baoquan, 139 Zhou Bida, 98n59 Zhou Hui: Qingbo zazhi (Miscellaneous notes by one who lives near the Gate of Clear Wave), 396n3 Zhou Kui, 204 Zhou Linzhi, 86–87, 88 Zhou Tingyun, 302 Zhou Yinghe: Jingding Jiankang zhi, 150n88, 219n95, 225n101, 458–65 Zhou dynasty, 238–39 Zhu Cunru, 328 Zhu Xi, 89–91, 97–98, 101, 211; Cheng Dachang and, 123; on mapmaking, 206–8, 212; on “Tribute of Yu,” 136; Yu Zhe and, 205; Taiji tushuo (Explanations and diagrams of the great ultimate), 79 Zhuru mingdao (All scholars advocating the Way), 292, 300

Harvard East Asian Monographs (most recent titles)

. Wilt L. Idema, Wai-yee Li, and Ellen Widmer, eds., Trauma and Transcendence in Early Qing Literature . Barbara Molony and Kathleen Uno, eds., Gendering Modern Japanese History . Hiroshi Aoyagi, Islands of Eight Million Smiles: Idol Performance and Symbolic Production in Contemporary Japan . Wai-yee Li, The Readability of the Past in Early Chinese Historiography . William C. Kirby, Robert S. Ross, and Gong Li, eds., Normalization of U.S.-China Relations: An International History . Ellen Gardner Nakamura, Practical Pursuits: Takano Chōei, Takahashi Keisaku, and Western Medicine in Nineteenth-Century Japan . Jonathan W. Best, A History of the Early Korean Kingdom of Paekche, together with an annotated translation of The Paekche Annals of the Samguk sagi . Liang Pan, The United Nations in Japan’s Foreign and Security Policymaking, – : National Security, Party Politics, and International Status . Richard Belsky, Localities at the Center: Native Place, Space, and Power in Late Imperial Beijing . Zwia Lipkin, “Useless to the State”: “Social Problems” and Social Engineering in Nationalist Nanjing, – . William O. Gardner, Advertising Tower: Japanese Modernism and Modernity in the s . Stephen Owen, The Making of Early Chinese Classical Poetry . Martin J. Powers, Pattern and Person: Ornament, Society, and Self in Classical China . Anna M. Shields, Crafting a Collection: The Cultural Contexts and Poetic Practice of the Huajian ji 花間集 (Collection from among the Flowers) . Stephen Owen, The Late Tang: Chinese Poetry of the Mid-Ninth Century ( – ) . Sara L. Friedman, Intimate Politics: Marriage, the Market, and State Power in Southeastern China . Patricia Buckley Ebrey and Maggie Bickford, Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics . Sophie Volpp, Worldly Stage: Theatricality in Seventeenth-Century China

Harvard East Asian Monographs . Ellen Widmer, The Beauty and the Book: Women and Fiction in NineteenthCentury China . Steven B. Miles, The Sea of Learning: Mobility and Identity in NineteenthCentury Guangzhou . Man-houng Lin, China Upside Down: Currency, Society, and Ideologies, – . Ronald Egan, The Problem of Beauty: Aesthetic Thought and Pursuits in Northern Song Dynasty China . Mark Halperin, Out of the Cloister: Literati Perspectives on Buddhism in Sung China, . Helen Dunstan, State or Merchant? Political Economy and Political Process in s China . Sabina Knight, The Heart of Time: Moral Agency in Twentieth-Century Chinese Fiction . Timothy J. Van Compernolle, The Uses of Memory: The Critique of Modernity in the Fiction of Higuchi Ichiyō . Paul Rouzer, A New Practical Primer of Literary Chinese . Jonathan Zwicker, Practices of the Sentimental Imagination: Melodrama, the Novel, and the Social Imaginary in Nineteenth-Century Japan . Franziska Seraphim, War Memory and Social Politics in Japan, – . Cynthia J. Brokaw, Commerce in Culture: The Sibao Book Trade in the Qing and Republican Periods . Eugene Y. Park, Between Dreams and Reality: The Military Examination in Late Chosŏn Korea, – Nam-lin Hur, Death and Social Order in Tokugawa Japan: Buddhism, Anti-Christianity, and the Danka System . Patricia M. Thornton, Disciplining the State: Virtue, Violence, and State-Making in Modern China . Vincent Goossaert, The Taoists of Peking, – : A Social History of Urban Clerics . Charo B. D’Etcheverry, Love after The Tale of Genji: Rewriting the World of the Shining Prince . Michael G. Chang, A Court on Horseback: Imperial Touring & the Construction of Qing Rule, – . Carol Richmond Tsang, War and Faith: Ikkō Ikki in Late Muromachi Japan . Hilde De Weerdt, Competition over Content: Negotiating Standards for the Civil Service – ) Examinations in Imperial China ( . Eve Zimmerman, Out of the Alleyway: Nakagami Kenji and the Poetics of Outcaste Fiction . Robert Culp, Articulating Citizenship: Civic Education and Student Politics in Southeastern China, – . Richard J. Smethurst, From Foot Soldier to Finance Minister: Takahashi Korekiyo, Japan’s Keynes . John E. Herman, Amid the Clouds and Mist: China’s Colonization of Guizhou, –

Harvard East Asian Monographs . Tomoko Shiroyama, China during the Great Depression: Market, State, and the World Economy, – . Kirk W. Larsen, Tradition, Treaties, and Trade: Qing Imperialism and Chosŏn Korea, – . Gregory Golley, When Our Eyes No Longer See: Realism, Science, and Ecology in Japanese Literary Modernism . Barbara Ambros, Emplacing a Pilgrimage: The Ōyama Cult and Regional Religion in Early Modern Japan . Rebecca Suter, The Japanization of Modernity: Murakami Haruki between Japan and the United States . Yuma Totani, The Tokyo War Crimes Trial: The Pursuit of Justice in the Wake of World War II . David M. Robinson, ed., Culture, Courtiers, and Competition: The Ming Court (1368– 1644) . Calvin Chen, Some Assembly Required: Work, Community, and Politics in China’s Rural Enterprises . Sem Vermeersch, The Power of the Buddhas: The Politics of Buddhism During the Koryŏ Dynasty ( – ) . Tina Lu, Accidental Incest, Filial Cannibalism, and Other Peculiar Encounters in Late Imperial Chinese Literature . Chang Woei Ong, Men of Letters Within the Passes: Guanzhong Literati in Chinese History, – Wendy Swartz, Reading Tao Yuanming: Shifting Paradigms of Historical Reception ( – ) . Peter K. Bol, Neo-Confucianism in History . Carlos Rojas, The Naked Gaze: Reflections on Chinese Modernity . Kelly H. Chong, Deliverance and Submission: Evangelical Women and the Negotiation of Patriarchy in South Korea . Rachel DiNitto, Uchida Hyakken: A Critique of Modernity and Militarism in Prewar Japan . Jeffrey Snyder-Reinke, Dry Spells: State Rainmaking and Local Governance in Late Imperial China . Jay Dautcher, Down a Narrow Road: Identity and Masculinity in a Uyghur Community in Xinjiang China . Xun Liu, Daoist Modern: Innovation, Lay Practice, and the Community of Inner Alchemy in Republican Shanghai . Jacob Eyferth, Eating Rice from Bamboo Roots: The Social History of a Community of Handicraft Papermakers in Rural Sichuan, – . David Johnson, Spectacle and Sacrifice: The Ritual Foundations of Village Life in North China . James Robson, Power of Place: The Religious Landscape of the Southern Sacred Peak (Nanyue 南嶽) in Medieval China . Lori Watt, When Empire Comes Home: Repatriation and Reintegration in Postwar Japan . James Dorsey, Critical Aesthetics: Kobayashi Hideo, Modernity, and Wartime Japan

Harvard East Asian Monographs . Christopher Bolton, Sublime Voices: The Fictional Science and Scientific Fiction of Abe Kōbō . Si-yen Fei, Negotiating Urban Space: Urbanization and Late Ming Nanjing . Christopher Gerteis, Gender Struggles: Wage-Earning Women and Male-Dominated Unions in Postwar Japan . Rebecca Nedostup, Superstitious Regimes: Religion and the Politics of Chinese Modernity . Lucien Bianco, Wretched Rebels: Rural Disturbances on the Eve of the Chinese Revolution . Cathryn H. Clayton, Sovereignty at the Edge: Macau and the Question of Chineseness . Micah S. Muscolino, Fishing Wars and Environmental Change in Late Imperial and Modern China . Robert I. Hellyer, Defining Engagement: Japan and Global Contexts, – . Robert Ashmore, The Transport of Reading: Text and Understanding in the World of Tao Qian ( – ) . Mark A. Jones, Children as Treasures: Childhood and the Middle Class in Early Twentieth Century Japan . Miryam Sas, Experimental Arts in Postwar Japan: Moments of Encounter, Engagement, and Imagined Return . H. Mack Horton, Traversing the Frontier: The Man’yōshū Account of a Japanese Mission to Silla in – . Dennis J. Frost, Seeing Stars: Sports Celebrity, Identity, and Body Culture in Modern Japan . Marnie S. Anderson, A Place in Public: Women’s Rights in Meiji Japan . Peter Mauch, Sailor Diplomat: Nomura Kichisaburō and the Japanese-American War . Ethan Isaac Segal, Coins, Trade, and the State: Economic Growth in Early Medieval Japan . David B. Lurie, Realms of Literacy: Early Japan and the History of Writing . Lillian Lan-ying Tseng, Picturing Heaven in Early China . Jun Uchida, Brokers of Empire: Japanese Settler Colonialism in Korea, – . Patricia L. Maclachlan, The People’s Post Office: The History and Politics of the Japanese Postal System, – . Michael Schiltz, The Money Doctors from Japan: Finance, Imperialism, and the Building of the Yen Bloc, – . Daqing Yang, Jie Liu, Hiroshi Mitani, and Andrew Gordon, eds., Toward a History beyond Borders: Contentious Issues in Sino-Japanese Relations . Sonia Ryang, Reading North Korea: An Ethnological Inquiry . Shih-shan Susan Huang, Picturing the True Form: Daoist Visual Culture in Traditional China . Barbara Mittler, A Continuous Revolution: Making Sense of Cultural Revolution Culture . Hwansoo Ilmee Kim, Empire of the Dharma: Korean and Japanese Buddhism, – . Satoru Saito, Detective Fiction and the Rise of the Japanese Novel, – . Jung-Sun N. Han, An Imperial Path to Modernity: Yoshino Sakuzō and a New Liberal Order in East Asia, –

Harvard East Asian Monographs . Atsuko Hirai, Government by Mourning: Death and Political Integration in Japan,

–

. Darryl E. Flaherty, Public Law, Private Practice: Politics, Profit, and the Legal Profession in Nineteenth-Century Japan . Jeffrey Paul Bayliss, On the Margins of Empire: Buraku and Korean Identity in Prewar and Wartime Japan . Barry Eichengreen, Dwight H. Perkins, and Kwanho Shin, From Miracle to Maturity: The Growth of the Korean Economy . Michel Mohr, Buddhism, Unitarianism, and the Meiji Competition for Universality . J. Keith Vincent, Two-Timing Modernity: Homosocial Narrative in Modern Japanese Fiction . Chong-Bum An and Barry Bosworth, Income Inequality in Korea: An Analysis of Trends, Causes, and Answers . Jamie L. Newhard, Knowing the Amorous Man: A History of Scholarship on Tales of Ise . Sho Konishi, Anarchist Modernity: Cooperatism and Japanese-Russian Intellectual Relations in Modern Japan . Christopher P. Hanscom, The Real Modern: Literary Modernism and the Crisis of Representation in Colonial Korea . Michael Wert, Meiji Restoration Losers: Memory and Tokugawa Supporters in Modern Japan . Garret P. S. Olberding, ed., Facing the Monarch: Modes of Advice in the Early Chinese Court . Xiaojue Wang, Modernity with a Cold War Face: Reimagining the Nation in Chinese Literature Across the 1949 Divide . David Spafford, A Sense of Place: The Political Landscape in Late Medieval Japan . Jongryn Mo and Barry Weingast, Korean Political and Economic Development: Crisis, Security, and Economic Rebalancing . Melek Ortabasi, The Undiscovered Country: Text, Translation, and Modernity in the Work of Yanagita Kunio . Hiraku Shimoda, Lost and Found: Recovering Regional Identity in Imperial Japan . Trent E. Maxey, The “Greatest Problem”: Religion and State Formation in Meiji Japan . Gina Cogan, The Princess Nun: Bunchi, Buddhist Reform, and Gender in Early Edo Japan . Eric C. Han, Rise of a Japanese Chinatown: Yokohama, 1894–1972 . Natasha Heller, Illusory Abiding: The Cultural Construction of the Chan Monk Zhongfeng Mingben . Paize Keulemans, Sound Rising from the Paper: Nineteenth-Century Martial Arts Fiction and the Chinese Acoustic Imagination . Simon James Bytheway, Investing Japan: Foreign Capital, Monetary Standards, and Economic Development, 1859–2011 . Sukhee Lee, Negotiated Power: The State, Elites, and Local Governance in TwelfthFourteenth China

Harvard East Asian Monographs . Foong Ping, The Efficacious Landscape: On the Authorities of Painting at the Northern Song Court . Catherine L. Phipps, Empires on the Waterfront: Japan’s Ports and Power, 1858–1899 . Sunyoung Park, The Proletarian Wave: Literature and Leftist Culture in Colonial Korea, 1910–1945 . Barry Eichengreen, Wonhyuk Lim, Yung Chul Park, and Dwight H. Perkins, The Korean Economy: From a Miraculous Past to a Sustainable Future . Heather Blair, Real and Imagined: The Peak of Gold in Heian Japan . Emer O’Dwyer, Significant Soil: Settler Colonialism and Japan’s Urban Empire in Manchuria . Martina Deuchler, Under the Ancestors’ Eyes: Kinship, Status, and Locality in Premodern Korea . Joseph R. Dennis, Writing, Publishing, and Reading Local Gazetteers in Imperial China, 1100–1700 . Catherine Vance Yeh, The Chinese Political Novel: Migration of a World Genre . Noell Wilson, Defensive Positions: The Politics of Maritime Security in Tokugawa Japan . Miri Nakamura, Monstrous Bodies: The Rise of the Uncanny in Modern Japan . Nara Dillon, Radical Inequalities: China’s Revolutionary Welfare State in Comparative Perspective . Ma Zhao, Runaway Wives, Urban Crimes, and Survival Tactics in Wartime Beijing, 1937-1949 . Mingwei Song, Young China: National Rejuvenation and the Bildungsroman, 1900-1959 . Christopher Bondy, Voice, Silence, and Self: Negotiations of Buraku Identity in Contemporary Japan . Seth Jacobowitz, Writing Technology in Meiji Japan: A Media History of Modern Japanese Literature and Visual Culture . Hilde De Weerdt, Information, Territory, and Networks: The Crisis and Maintenance of Empire in Song China . Elizabeth Kindall, Geo-Narratives of a Filial Son: The Paintings and Travel Diaries of Huang Xiangjian (1609–1673) . Matthew Fraleigh, Plucking Chrysanthemums: Narushima Ryūhoku and Sinitic Literary Traditions in Modern Japan . Hu Ying, Burying Autumn: Poetry, Friendship, and Loss . Mark E. Byington, The Ancient State of Puyŏ in Northeast Asia: Archaeology and Historical Memory . Timothy J. Van Compernolle, Struggling Upward: Worldly Success and the Japanese Novel 394. Heekyoung Cho, Translation’s Forgotten History: Russian Literature, Japanese Mediation, and the Formation of Modern Korean Literature 395. Terry Kawashima, Itineraries of Power: Texts and Traversals in Heian and Medieval Japan 396. Anna Andreeva, Assembling Shinto: Buddhist Approaches to Kami Worship in Medieval Japan