Research from Archival Case Records : Law, Society and Culture in China [1 ed.] 9789004271890, 9789004271883

Citation preview

Research from Archival Case Records

The Social Sciences of Practice The History and Theory of Legal Practice Series Editor Philip C. C. Huang (University of California, Los Angeles and Renmin University of China)

VOLUME 1

The titles published in this series are listed at brill.com/ssop

Research from Archival Case Records Law, Society, and Culture in China Edited by

Philip C. C. Huang and Kathryn Bernhardt

LEIDEN | BOSTON

Library of Congress Cataloging-in-Publication Data Research from archival case records : law, society, and culture in China / Edited by Philip C. C. Huang and Kathryn Bernhardt.   p. cm. — (The social sciences of practice 1)  Includes bibliographical references.  ISBN 978-90-04-27188-3 (hardback : alk. paper) — ISBN 978-90-04-27189-0 (e-book) 1. Law—China— History. 2. Justice, Administration of—China—History. 3. Practice of law—China—History. I. Huang, Philip C., 1940– editor of compilation. II. Bernhardt, Kathryn, editor of compilation.  KNN122.R47 2014  349.51—dc23

2014002473

This publication has been typeset in the multilingual ‘Brill’ typeface. With over 5,100 characters covering Latin, ipa, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities. For more information, please see brill.com/brill-typeface. issn 2214-952X isbn 978 90 04 27188 3 (hardback) isbn 978 90 04 27189 0 (e-book) Copyright 2014 by Koninklijke Brill nv, Leiden, The Netherlands. Koninklijke Brill nv incorporates the imprints Brill, Brill Nijhoff, Global Oriental and Hotei Publishing. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, ma 01923, usa. Fees are subject to change. Brill has made all reasonable efforts to trace all rights holders to any copyrighted material used in this work. In cases where these efforts have not been successful the publisher welcomes communications from copyright holders, so that the appropriate acknowledgements can be made in future editions, and to settle other permission matters. This book is printed on acid-free paper.

Contents Author Biographies  xii Series Foreword  xvi Editor’s Introduction  1 Philip C. C. Huang

PART 1 Analytical Approaches: History of Practice · Women’s History · Local Administration · Discourse Analysis · Case Records as Ethnographic Evidence  23 1 The History-of-Practice Approach to Studying Chinese Law (Introduction to Chinese Civil Justice, Past and Present)  25 Philip C. C. Huang Practice as Opposed to Theory: Legal Formalism and the History of Practice of American Law  26 Practice as Opposed to Representation: Qing Law  27 Practice as Opposed to Institutions: Male and Female Inheritance Rights and Their Actual Operation  30 The History of Practice vs. Formalist Theory  31 Practical Moralism  32 Divorce Law Practices and the System of Court Mediation  36 The Third Realm and Centralized Minimalism  39 Community Mediation under Minimalist Governance  42 References  45 2 Women and Property in China, 960–1949, Introduction and Conclusion  47 Kathryn Bernhardt Introduction  47 The Issues  47 Source Materials  51 The Song Baseline  53 Conclusion  53 References  56

vi

contents

3 Illicit Bureaucrats  57 Bradly W. Reed Preface  57 The Issues  62 Past Scholarship  66 Administration and Bureaucracy  66 State and Society  73 Corruption  78 References  84 4 From Oral Testimony to Written Records in Qing Legal Cases  89 Yasuhiko Karasawa Introduction  89 The Status of Depositions in Qing Legal Procedure  90 Writing Legal Testimony in the Context of Literary Culture  93 Records of Oral Testimony Written in the Vernacular  97 Composing Testimony at the Local Level: A “Directly Examined” Case from Beijing  103 Conclusion  110 Character List  112 References  114 5 Abortion in Late Imperial China: Routine Birth Control or Crisis Intervention?  119 Matthew H. Sommer Introduction 119 Past Scholarship 124 The Demographic Historians: Routine Birth Control to Limit Family Size?  124 The Gender Historians: Autonomous Fertility Control by Elite Wives?  135 Abortion in Qing Legal Records  142 Abortion in Qing Law  142 The Practice of Abortion as Seen in Qing Legal Cases  147 The Cost and Accessibility of Abortion  155 Inconvenient Pregnancies Carried to Term  163 Unsafe Abortion in Modern China  165 Traditional Abortion Methods in Hebei, 1928  165 Traditional Abortion Techniques since 1949  168 The Persistence of Unsafe Traditional Abortifacients  170 Conclusion  175 Glossary  177 References  180

contents

PART 2 Buying and Selling of Land · Homicides  191 6 Customary and Judicial Practices as Seen in Criminal Sales of Land in Qing Manchuria  193 Christopher Isett Sources and Methods  195 The Criminalization of Customary Practice in Manchuria  196 The Sale of Qing Land to Commoners  197 Rural Agents, Peasant Defiance, and the Politics of Local Compromise  206 Adjudication in the Face of Criminal Customary Acts  210 References  213 7 Guoshi Killing: The Continuum of Criminal Intent in Qing and Republican China  216 Jennifer M. Neighbors Guoshi Killing in the Qing Dynasty  217 Guoshi Killing as Accidental Killing  218 Guoshi Killing as Negligent Killing  222 Non-Guoshi Negligence  225 Killing at Play  229 The Republican Codes  236 Guoshi in the Republican-era Criminal Codes  236 Republican-era Guoshi Cases  241 Standard Negligence  242 Accident, Negligence, or Neither?  245 Negligence or Intention?  248 Conclusion  250 References  251

PART 3 Tax · Education · Local Governance  255 8 Between the State and the Village: Land Taxation and “Substantive Governance” in Traditional China  257 Huaiyin Li Introduction  257 The County Government and the Xiangdi  263 The Xiangdi’s Delivery of Taxes  264 Tax Prompting by the County Government  266

vii

viii

contents

Village Regulations on Taxation in the Late Qing  267 Village Regulations on Taxation in the Republican Period  269 Village Regulations  269 Noncompliance  271 The Tax-Prompting Xiangdi  273 The Collection of Enclave Taxes  276 Conclusion  278 References  283 9 Village-State Cooperation: Modern Community Schools and Their Funding, Haicheng County, Fengtian, 1905–1931  286 Elizabeth VanderVen The Setting: Fengtian Province and Haicheng County  288 Funding the New Community Primary Schools  290 Costs of Establishing and Operating Community Primary Schools  290 Methods of Funding Primary Schools  292 Multi-Village Relationships: Creativity, Cooperation, and Conflict  301 References  311 10 Power Networks and State-Society Relations in Republican China  314 Danny Hsu Local Governance in Late Imperial China  317 Xinmin County in Early Republican China  319 The Appointment of Ward Officials  320 Power Networks in Xinmin County  322 Conflict between Village Headmen  323 Village Residents versus Village Headmen  324 Villagers versus Villagers  326 Sub-County Administration in the Early Republic  327 Sichuan and the National Government  328 Changes in Local Administration  329 Power Networks and the Ward Office  332 Power Networks and the Township-Village Office  336 Conclusion  340 References  340

contents

ix

PART 4 Concubinage · Spousal Abuse · Transnational Families  343 11 Ceremony and the Definition of Marriage under Republican Law  345 Lisa Tran Ceremony in Social Practice  346 The Ceremony for a Main Wife  346 The Ceremony for a Minor Wife  348 A “Main Wife” Ceremony for a Concubine  350 Ceremony in Early Republican Law  353 The Legal Space for Concubinage in the Early Republic  360 Ceremony in the 1929–30 Civil Code  364 From Consent to Complicity under GMD Law  368 References  371 12 Spousal Abuse: Divorce Litigation and the Emergence of Rights Consciousness in Republican China  374 Margaret Kuo The Prevalence of Intolerable Cruelty Divorce Litigation  377 From Grievance to Injustice: “Naming, Blaming, and Claiming”  380 “Please Protect Women’s Rights”: Cao Xiuzhen’s Pleas  382 Intolerable Cruelty Defenses: Marriage Finance, Economic Hardship, and Socioeconomic Interpretations of Rights  387 “Marriage and money are not easy matters”: Lü Linger v. Ding Maotou  390 “How can such trivial reasons be allowed to split us up?”: Zhu Xu shi v. Zhu Er  393 State Approaches to Intolerable Cruelty Cases: Judicial Outcomes Affirm a Modern Conjugal Patriarchy  395 State Validation of Women’s Grievances Narrowly Construed  396 Severity and Proof: Guo Zhou shi Granted Divorce after Severe Beating  398 Habitual Abuse: Li Yourong Receives a Divorce after Repeated Abuse  401 Individual Rights and the Ironic Affirmation of Modern Conjugal Patriarchy   403 References  404

x

contents

13 Law, Gongqin, and Transnational Polygamy: Family Matters in Fujian and British Malaya, 1855–1942  408 Huey Bin Teng Between Two Worlds: The Making of Chinese Customary Law in Malaya  415 Establishment of Common Law in Malaya  416 Exceptions to the Common Law: The Making of Chinese Customary Law  417 The Uncustomary Chinese Customary Law: Polygamy and Formalities of Marriage  424 Jiating shi 家庭事: Law and Family Life across Geographical Boundaries  437 Tongyangxi  437 Christians  446 Mediation and Enforcement: The Gongqin in Cross-Border Conflicts  451 The Ups and Downs of a Gongqin  452 Implicating the Gongqin (gongqin bian shizhu 公亲变事主)  454 Conclusion  456 References  457 People Interviewed  457 Works Cited  457

PART 5 Past and Present: Local Administration · Court Mediation  461 14 Centralized Minimalism: Semiformal Governance by Quasi-Officials and Dispute Resolution in China  463 Philip C. C. Huang The Evidence  464 The Late Qing Baodi County Example  464 The Republican Shunyi Example  467 Tax Administration in Late Qing and Republican Huailu County  470 Educational Administration and Village Schools in Haicheng County in the Northeast  471 Yamen Administration in Ba County, Sichuan, in the Qing  473 Centralized Minimalism  475 Confucianized Legalism  479

contents

xi

Bureaucratization and Minimalism in Contemporary China  481 References  486 15 Court Mediation in China, Past and Present  490 Philip C. C. Huang The Ideology of Mediation in the Qing  492 The Actual Practice of Qing Courts  494 Mediation in the Republic  497 The Ideology of Mediation in Post-1949 China  500 The Practice of Court Mediation in Post-1949 China  504 No-Fault Mutual Consent Divorce Cases  504 No-Fault “Tort” Cases  508 Both Parties Equally at Fault  509 Both Parties Bearing Equally Legitimate Claims or Obligations  511 Between Mediation and Adjudication  513 The Nature of Contemporary Chinese Judicial Mediation  518 The Qing, the Republic, and Post-1949 China  522 The Logic of Chinese Court Mediation  523 References  526 Interviews  526 Case Records  526 Books and Articles  527

Postscript  531 16 How a “New Legal History” Might Be Possible: Recent Trends in Chinese Legal History Studies in the United States and Their Implications  533 Chenjun You Introduction: An Intellectual Earthquake?  533 Westerners’ Misunderstandings of and Reflections on Traditional Chinese Law  534 Judicial Archives and Research on Chinese Legal History  538 Chinese Legal History Studies and the Social Sciences  542 Discovering a Historical Sense in the Meeting of Empiricism and Theory  545 Stones from Other Hills May Serve to Polish the Jade of This One 他山之石, 可以攻玉: The UCLA Research Group’s Achievements and Chinese Introspection  553 Conclusion  558 References  559

Author Biographies Philip C. C. Huang taught at UCLA from 1966 to 2004, advancing to “Professor, Above-Scale” in 1991, and has taught at the Renmin University of China, in the Law School and the School of Agricultural Economics and Rural Development, since 2005. He was the founding director of UCLA’s Center for Chinese Studies from 1986 to 1995, the (founding) editor of Modern China: An International Journal of History and Social Science from 1975 to the present, and the (founding) editor of 中国乡村研究 (Rural China: An International Journal of History and Social Science) from 2003 to the present. His major publications are his trilogy on rural China: The Peasant Economy and Social Change in North China, 1985 (awarded the Fairbank prize of the American Historical Association); The Peasant Family and Rural Development in the Yangzi Delta, 1350–1988, 1990 (awarded the Levenson prize of the Association for Asian Studies); and 超越 左右: 从实践历史探寻探寻中国农村发展出路 (Beyond the Left-Right Divide: Searching for a Path of Rural Development in China from the History of Practice), in Chinese only, 2014. And his trilogy on Chinese civil justice: Civil Justice in China: Representation and Practice in the Qing, 1996; Code, Custom, and Legal Practice in China: The Qing and the Republic Compared, 2001; Chinese Civil Justice, Past and Present, 2010. All the books in English have been published in multiple printings and editions in Chinese. Kathryn Bernhardt is Professor Emerita of History at the University of California, Los Angeles. She is the author of Rents, Taxes, and Peasant Resistance: The Lower Yangzi Region, 1840–1950 (Stanford University Press, 1992; awarded the 1992 John K. Fairbank prize of the American Historical Association) and Women and Property in China, 960–1949 (Stanford University Press, 1999) and co-editor (with Philip C. C. Huang) of Civil Law in Qing and Republican China (Stanford University Press, 1994). She has served as the co-editor of Modern China: An International Journal of History and Social Science from 1998 to the present. Danny Hsu teaches at Dalian University of Foreign Languages. He was the Resident Director of the Dalian study center for Brethren Colleges Abroad from 2007 to 2012 and also formerly taught world history at Saginaw Valley State University. His research interests include Chinese legal history, state-building and state-society relations, civil society, and more recently, Sino-Western cultural relations.

author biographies

xiii

Christopher Isett is Associate Professor of History at the University of Minnesota. He is author of “Social History and Rehistorizing the Great Divergence Debate in Qing and World History” in Haihui Zhang ed., A Scholarly Review of Chinese Studies in North America (Asia, Past and Present, Association for Asian Studies, 2013), State, Peasant, and Merchant in Qing Manchuria, 1644–1862 (Stanford University Press, 2007) and co-author with Robert Brenner of “England’s Divergence from China’s Yangzi Delta: Property Relations, Microeconomics, and Patterns of Economic Development” ( Journal of Asian Studies, 2002). He is currently working on a study of wellbeing in eighteenth-century China using anthropometric data gathered from homicide records and a study of economic growth in Taiwan from 1949 to 1993. Yasuhiko Karasawa is Professor of History at Ritsumeikan University, Japan. His major publications in Chinese history are “Shindai Chūgoku no chihō saiban bunsho” (Local-level case records in Qing China), Rekishi to Chiri 636 (2010); “Qingdai de suzhuang ji qi zhizuozhe” (Legal plaints and their writers in the Qing), Beida falü pinglun 10–1 (2009); and “Between Oral and Written Cultures: Buddhist Monks in Qing Legal Plaints,” in Robert E. Hegel and Katherine Carlitz eds., Writing and Law in Late Imperial China: Crime, Conflict, and Judgment (University of Washington Press, 2007). He is currently working on the social and cultural roles of official notaries working for local governments at the interface between oral and written cultures in late imperial China. Margaret Kuo is Associate Professor of History at California State University, Long Beach, where she teaches courses on Chinese history, women’s history, and history and theory and serves on the editorial board for The History Teacher. She is the author of Intolerable Cruelty: Marriage, Law, and Society in Early Twentieth-Century China (2012). Her current research examines the history of missionary photography in China. Huaiyin Li is Professor of Modern Chinese History at the University of Texas at Austin. Author of two books on rural China and a third book on Chinese historiography in the twentieth century, he is working on two new projects that investigate the renewal of the Chinese state since 1600 and the political economy of Maoist China, respectively.

xiv

author biographies

Jennifer M. Neighbors is Associate Professor of History at the University of Puget Sound. She is currently completing a book manuscript on homicide law in Qing and Republican China. Bradly W. Reed (B.A. University of Oregon, M.A. University of Washington, PhD University of California, Los Angeles) teaches late imperial and modern Chinese history at the University of Virginia. He is the author of Talons and Teeth: County Clerks and Runners in the Qing, Stanford University Press, 2000; and articles on late imperial administration and law. He is currently working on the Qing judicial process in relation to homicide cases. Matthew H. Sommer (BA Swarthmore College, MA University of Washington, Ph.D. UCLA) teaches Chinese history at Stanford University; he previously taught at the University of Pennsylvania. He is the author of Sex, Law, and Society in Late Imperial China (Stanford University Press, 2000); his second book, near completion, will be entitled Polyandry and Wife-Selling in Qing Dynasty China: Survival Strategies and Judicial Interventions. Huey Bin Teng is a doctoral candidate in Chinese history at the University of California, Los Angeles. Her research focuses on migration, gender, and transnational families, as well as the mutual interaction between China and Southeast Asia. In her dissertation on “Gender, Power and Transnational Families: Fujianese Migration in the Republican Period (1911–1949),” she draws on a wide variety of research materials (including court case records, family letters, diaries, genealogies, and oral histories) to investigate the complex internal dynamics of transnational polygamous families, in which the husband has one wife in his hometown in Fujian, China, and another in his new home in British Malaya. Lisa Tran (BA, Pomona College; MA, Ph.D., UCLA) is Associate Professor of Modern Chinese History at California State University, Fullerton. Her research focuses on women and the law in twentieth-century China. With fellowship support from the American Council of Learned Societies, the American Philosophical Society, and the Fulbright Program, she has conducted archival research in Beijing, Shanghai, and Nanjing. Her publications have appeared in various scholarly journals, including those published in China and France.

author biographies

xv

Elizabeth VanderVen is an independent scholar and historian of late imperial and modern China. Her research interests include education, schooling, rural society, and the culture and history of wines and spirits. She is the author of A School in Every Village: Educational Reform in a Northeast China County, 1904–1931 (University of British Columbia Press, 2012) and has taught at Reed College, Rutgers University, Camden, and Pepperdine University. Chenjun You 尤陈俊 received his PhD from Peking University Law School in 2010 and is Associate Professor in the Law School, Renmin University of China 中国人民大学. He served as the editor-in-chief of 北大法律评论 (Peking University Law Review) from 2008 to 2009. His research fields include Chinese legal history and the sociology of law. He is the author of the monograph 法律知识的文 字传播: 明清日用类书与社会日常生活 (Written Propagation of Legal Knowledge: The Encyclopedias for Daily Use and Social Life in the Ming and Qing Dynasties) and the editor or co-editor of several volumes of articles published in Chinese. His publications also include more than thirty academic articles in various journals in China, South Korea, and the US. He is currently completing a book manuscript on litigation and society during Ming and Qing China.

Series Foreword Philip C. C. Huang 黄宗智

The social sciences in China and the U.S. have come to be rather heavily dominated by abstract theorizing divorced from practical realities. What this series proposes to emphasize instead is actual economic and legal, and historical and social practices, and the theoretical logics evidenced therein. The theoretical works included in the series proceed not from theory to practice, but rather from practice to theory; the empirical studies included are ones of important theoretical implications. We propose to include selected major works in each of five sub-series, to be published simultaneously in both English and Chinese, or, where the work is already available in one language (English or Chinese), then its translation into the other. The five sub-series include one each in the history and theory of legal practice, the economic history and economics of practice, and the social history and sociology of practice. The fourth series consists of broader crossdisciplinary works in historical political economy, in the tradition of the likes of Adam Smith, Karl Marx, and Max Weber. The fifth series includes major innovations evident in Chinese economic, legal, social, and political-economic practices that have yet to receive full theoretical elaboration. The typical social science study today proceeds from a certain theoretical position, and asks a question derived from that particular perspective, with the intention of proving (or, sometimes, disproving) the posited “hypothesis.” This may be done explicitly or implicitly, but always with a host of assumptions, often unspoken, even unconscious. Since almost all existing social science theories have originated from the West, such an epistemological approach often comes with the belief that the Western experience (e.g., capitalism and democracy and formalist law) and the theoretical abstractions therefrom are universally applicable. When applied to the developing non-Western world, the research can become mainly a matter of finding inadequacies in the subject country or region, often with implicit or explicit advocacy for development in the direction of the Western “model.” This is most evident, for example, in the disciplines of economics and jurisprudence, the most formalistic and ideological of the “modern” disciplines. Our “social sciences of practice,” however, calls for inverting that epistemological process, to proceed not from (Western-originated) theory (and

series foreword

xvii

hypotheses derived therefrom), but rather from the practical realities of the subject country. Non-Western countries, faced with the economic-military as well as cultural-intellectual expansion of the West in modern times, have been placed unavoidably into a historical situation of conflicting tugs between tradition and (Western) “modernity,” the indigenous and the imported, East and West. What exist in practice are almost of necessity paradoxical when seen from the perspective of Western theory. We start with practice because, unlike theory, practice is anchored in the subject country’s own social-economic and political contexts, perspectives, and discourses. And problems seen through practice rather than theory are far more likely to be of indigenous concern to the subject country itself rather than just theoretical / epistemological concern to the West. Practice, in fact, tells first and foremost about paradoxes—or the co-existence of pairs of seemingly contradictory and mutually exclusive realities that are both true and real, like (globalized) commercialization without development, or capitalism without democracy, or Westernized formalist legal codes without their judicial practice. It challenges the very validity of commonly used (Western) theories that presuppose a cause and effect relationship between those, and therefore cries out for new theoretical formulations. In addition, unlike theory, which tends to be dominated by deductive / formal (and Western-originated) logic, and hence insists on logical consistency, practice readily accommodates the seemingly contradictory. It demands conceptualization and theorizing in terms of its own logic—such as, for example, China’s “feeling for the stones while crossing the river.” The perspective from practice is one that calls for taking historical process as the point of departure, and for theoretical formulations therefrom. Such practical and theoretical concerns do not mean, however, the simple rejection or disregard of Western social science theory, but rather deliberate dialoguing with existing theory, and also deliberate borrowing from and developing the West’s multiple alternative theoretical traditions. Such research can in fact highlight the need to combine “mainstream” formalist theoretical traditions with alternative Western theoretical traditions to conceptualize Western experience itself—e.g., combining the jurisprudential traditions of “classical orthodoxy” with legal pragmatism (and Marxism and postmodernism) to understand the past and present of American law, or the Smithian with the Marxist and substantivist to understand the economic history of the West. Better yet, to uncover the operative theoretical logics of those combinations that exist in practice and to search for ways to go beyond the either/or juxtaposition of binaries that are seen to be mutually exclusive.

xviii

huang

Our series proposes to include such works in economics, law, sociology, and political economy, as well as works on such Chinese innovations that have already been made in practice but have yet to receive full theoretical formulation and attention. Our goal is to develop a series of works guided by problematics and theoretical conceptions that are better suited to Chinese concerns and realities than typical “mainstream” formalist studies.

Editor’s Introduction Philip C. C. Huang This introduction will begin with some background information and then turn to short summaries-commentaries of each article in the volume, before concluding with some general observations about what the different pieces share in common. Background Since 1988, I have turned my main research energies from rural social-economic history to legal history. A major reason for the change was the availability of large quantities of rich local government archival materials, especially legal case records, which had yet to be seriously explored. Before that, my studies had been based mainly on firsthand field survey reports, interview transcripts and quantitative data, especially the household- and village-level information gathered by Japanese researchers in the 1930s using modern social-science methods. Those kinds of materials are not usually available to scholars of West European history, given the fact that peasant economies had largely dis­ appeared before the coming of modern social science, unlike in China where the peasant economy has persisted well into the 20th and the 21st century. When I came into contact with the local archival materials during my year-long stint in China in 1980, I realized that China researchers had a unique opportunity to draw on both social science research and local government records. The latter had propelled some of the path-breaking works of Europeanists, such as those of Georges Lefebvre ([1934] 1959; Huang 1985: 33) and, later, of the second-generation Annales scholar Emmanuel Le Roy Ladurie (1974). China researchers, I thought, could do more by drawing on both local government records and modern social science data. I therefore published an article in 1982 urging China scholars-colleagues to take advantage of the opportunity afforded by the even richer source materials available to us (Huang 1982). By the late 1980s, my own work on rural social-economic research was coming to a natural pause, having completed The Peasant Economy and Social Change in North China (1985) and The Peasant Family and Rural Development in the Yangzi Delta (1990), and the follow-up article “The Paradigmatic Crisis in Chinese Studies: Paradoxes in Social and Economic History” (1991). I was also beginning to rethink the implicit materialism of my own earlier work.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_�02

2

Huang

The special appeal of legal case records for me was that they contained rich information not only about behavior, but also representations, and the interactions between those two dimensions. As I wrote in 1998 in reflecting upon my own turn of research interests: Legal history has held special appeal for me precisely because it compels us to deal not only with actions but also with representations, and not only with practical realities but also with ideals. Legal documents [i.e. case records] arguably articulate more than most other kinds of sources the logics of both customary practice and official ideology and of the relationships between them. . . . Legal records have shown me the importance of representation, but they have also reminded me of the crucial difference between genuine evidence and fraudulent evidence and between truth and fabrication. . . . (Huang 1998: 200–201). At the same time, I realized that earlier scholarship on Chinese legal history, due to the lack of case records, had tended to rely too much on sources explicating official ideology rather than legal practice. Even the best available research relied at best on collections of exemplary verdicts (e.g., Shiga Shūzō (1981, 1984), or compendia of selected major cases handled by the central-level Board of Punishment (e.g., Bodde and Morris, 1967). It lacked solid grasp of how the courts actually operated, especially with respect to so-called “minor matters” pertaining to “households, marriage, and land” (户婚田土, what we would today term “civil cases”). Most scholars simply accepted the official constructions, assuming that there were relatively few “minor matters” cases, or at least that those were of little importance—which is very far removed indeed from the realities shown by the case records. I have therefore emphasized especially the approach of starting with case records and legal practice, though certainly not to the neglect of codified texts and other legal sources. In hindsight, my emphasis on attention to both representation and practice, and how they inter-relate, is related to my own background of studying household- and village-level rural social-economic history. Entering into legal history from this background, I naturally developed a historical perspective of attending to both—to material bases as well as thought and mentality, to social-economic context as well as agency, to historical tendency as well as individual choice, and to institutional design as well as actual operation. The Japanese field survey data, it turns out, also provide rich information on villagers’ lawsuits, as well as community and kin mediation, to help supplement and contextualize official case records.

Editor ’ s Introduction

3

After deciding to enter into this field of research, I had the good fortune of meeting up with a group of outstanding colleagues and students. First was my spouse Kathryn Bernhardt, who too was just turning from the study of rural history to legal history, with special attention to what law meant to the common people. We in fact shared a remarkable amount in common. Then, the new Center for Chinese Studies which was founded in 1986 (in response to my offer from Princeton University), and of which I became founding director, soon attracted a critical mass of exceptional students who were also interested in this area of inquiry: Bradly Reed, Matthew Sommer, Christopher Isett, and Yasuhiko Karasawa,1 and together we quickly formed a vibrant research community. Then, from 1991 on, Kathryn Bernhardt and I succeeded in obtaining a large Luce Foundation grant, which the Foundation was kind enough to allow us to spread over the course of ten years, lending us important support for the acquisition of research materials, conferences and student support over and above the institutional commitments from UCLA. After the four students listed above, a “second generation” of outstanding students came, quite a number of whom were interested in legal history, including Jennifer Neighbors (who had taken an MA under Bradly Reed at the University of Virginia and was thus both in name and fact a second generation student), Elizabeth VanderVen, Lisa Tran, Margaret Kuo (who came with a JD degree and experience as an attorney), and Huey Bin Teng. In addition, between the two generations came Huaiyin Li (from China) and Danny Hsu. The seven of them formed a second critical mass after the first. This volume contains selected representative works from the eleven former students mentioned above, along with Kathryn Bernhardt and myself. Every article is based on archival case records, gathered from months to years of research at Chinese local government archives, each amounting to no less than several hundred legal cases, and each with conclusions based on such evidence. To use Kathryn Bernhardt and myself as an example, we together collected the archival records of a total of 2200 cases. In 2005, after completing our own research and writing, we donated-sold the materials to the East Asian Library of Stanford University (where Matthew Sommer teaches), totaling 180 volumes of bound xerox copies (a total of 36,500 sheets) and 135 reels of microfilm (see the inventory list appended to this Introduction). We and James Cheng, at that time Curator of UCLA’s East Asian Library, had ambitious plans to build on the basis of the research materials used by our students and us the largest collection of case records and local government archives outside 1 In addition to others in social and economic history.

4

Huang

of China, but unfortunately the plan had to be abandoned when James was subsequently hired away by the Harvard-Yenching Library. Interested scholarsstudents can, however, use the “Huang-Bernhardt Collection” at Stanford as a resource, the inventory list of which is appended to this introduction.

The Articles

Part One. Analytical Approaches: History of Practice, Women’s History, Local Administration, Discourse Analysis, and Case Records as Ethnographic Evidence Archival materials, of course, cannot be separated from the researcher’s approach and sense of problem. This volume begins with five pieces that each contains broad discussions and illustrations of distinct approaches. The first summarizes my own major research findings and understandings acquired from more than two decades of research into these archival records. Legal studies in general tend to emphasize codified texts more than actual legal practice. Ever since I found in the archival records a striking difference between how the legal system was represented in the code (and in official pronouncements) and how it actually operated, I have placed that disjunction at the center of my inquiry. I have sought to understand the implicit logics of each of the two dimensions, as well as how they interacted over time. What is included here had begun as a lecture to explain my “history of practice” approach, later revised and expanded into an article, and subsequently further revised to become the introductory chapter of the third volume of my trilogy on Chinese civil justice from the Qing to the present. I emphasize in particular how representation and practice are different or even contradictory, and yet together form a paradoxical and yet mutually defining whole to make up the legal system. One theme revealed by that combination is an abiding “practical moralism”—of high moral values along with practical considerations of what works and does not work—in Chinese legal thinking from the Qing down to the present. The article illustrates these larger points with concrete illustrations drawn from cases in several different spheres of civil justice. Kathryn Bernhardt’s work focuses on women’s property rights from the Song through the Republic. On the basis of a large body (some 438) of Qing and Republican-period inheritance cases, plus relevant legal codes and court verdicts, and a collection of Song dynasty cases, she demonstrates a surprising finding. Most people have equated imperial China’s property inheritance system with equal division among sons (household division 分家), and have viewed the patrilineal succession 承祧 system as but another manifestation of

Editor ’ s Introduction

5

the same principles and system, with little or no change over time. Bernhardt demonstrates, however, that when viewed from the standpoint of women, we find that inheritance and succession were in fact two very different systems. As many as one-third of all women either had no brothers (as a daughter) or no sons (as a mother). For those women, the household-division system (which involved only sons) mattered little, but the succession system was crucial. And that system went through hitherto un-recognized changes between the Song and the Republic. In the Song, women without brothers and women without sons could inherit the father’s or husband’s property. In the Ming, however, the law required that widows without sons must establish a nephew as the patrilineal successor/heir 过继, such that widows and daughters themselves no longer had the right to inherit the husband’s/father’s property. Widows now could only wield custodial rights on behalf of the adopted nephew heir. In the Qing, however, widows, especially “chaste widows” 守节孀妇, were allowed first in legal practice to reject her deceased husband’s closest nephew as heir (termed 应继, or the “ought-to-be heir”), and could choose from among all the patrilineal nephews the one she most preferred (termed 爱继, or the “affectionate heir”). This expanded legal right of widows was written into law in the midQing. Under the early Republican Supreme Court 大理院, the scope of choice for widows was further expanded, giving them broader discretion still in the choice of an heir. But then, with the coming of the Guomindang government’s new Civil Code of 1929–1930, the patrilineal succession system was completely set aside, replaced by a new imported legal regime of equal inheritance rights for sons and daughters. But those new provisions of the law mattered little in actual practice in the countryside, where the old system of partible inheritance among sons continued to operate unchanged. Such changes, of course, have profound implications not only for the history of women in China, but also for the history of property rights in general. What is included in this volume is the introduction and conclusion of Bernhardt’s monograph, which summarizes the analytical framework and themes of the entire monograph, as well as her views on women’s history. Bradly Reed’s contribution examines the actual operations of locallevel administration on the basis of the Baxian (Ba county) archives (on “internal administration” 内政). Past scholarship had been deeply influenced by official constructions, characterizing local government clerks and runners as “yamen vermin” 衙蠹. Reed demonstrates that in actual operation, these local government personnel were actually a paradoxical mix—though (the majority were) not formally recognized by the state and therefore “illicit,” they were nevertheless indispensable functionaries of a bureaucratic apparatus almost in the Weberian sense of bureaucracy (hence “illicit bureaucrats”). In their

6

Huang

handling of official affairs like judicial administration and tax collection, they exhibited both the Weberian (modern rational) bureaucratic characteristics of division of labor, professionalization, institutionalization, and formalization, as well as the non-bureaucratic characteristics of “prebendal” offices (that come with property and income) and personalistic relations. These together made up a system that was neither completely formal nor completely informal, but somewhere in between. Our understanding of “corruption” in the Chinese administrative system had been limited to the binary of the statutory and the corrupt, to the neglect of what occurs outside the bounds of formal rules and regulations but were still widely accepted as customary administrative practices, to be distinguished from truly abusive corruption. The selection here comes from the preface and introduction of Reed’s monograph. The fourth article is by Yasuhiko Karasawa who, from the time he first entered our graduate program, was very much concerned with discourse analysis. What is included here is his work based on a study of cases from the local Beijing area that were under the jurisdiction of the Board of Punishment and “directly examined” 现审 by it. Those materials contain large numbers of transcripts of the initial oral testimonies of suspects along with the paraphrasing or recounting of those (especially in the final “confession” 招供) in the final written report on the case. They therefore permit a detailed analysis of the changes introduced in the transition from oral testimony to written record. Karasawa shows first how local dialects and colloquialisms, and coarse or crude language, are changed into concise official mandarin in the process of transcription. More important, for the final written report, contradictory testimonies are eliminated or altered to present a coherent narrative, even subjected to literary manipulations to form a consistent whole. As Wang Youhuai 王又槐, a well-known 18th-century private legal secretary advised in the chapter on “Narrating Testimonies” 叙供 of his Important Points for Handling Cases 办案要略, a magistrate must see to it that the information he sends upward supports the verdict he recommends, lest his handling of the case be questioned by upper levels of the judicial bureaucracy. In terms of analytical approach, Karasawa has long emphasized, in the manner of Kurosawa’s film Rashomon, how “truth” may be highly elusive and represented differently in different versions. This section concludes with the article from Matthew Sommer on abortion as practiced in the Qing, perhaps the best example to date of the use of legal case records as ethnographic evidence on the social lives of common people. Abortion has been quite widely studied in recent years in two distinct groups of scholarly literature (demographic studies and women’s studies), each arguing that it was quite readily available and quite commonly used in the Qing.

Editor ’ s Introduction

7

But both bodies of literature have been based mainly on inferences from notations in medical treatises about abortion methods and drugs; neither contains any concrete case examples of actual abortions, not even one. What Sommer does, first of all, is to comb legal case records for contemporary evidence on how and why abortion was practiced. On the basis of 31 such cases from all over China, which record a total of 24 completed attempts at abortion, 3 interrupted attempts, and 4 situations in which the woman sought but failed to find a way to abort, Sommer shows that 17 of the 24 women who completed attempts died, and at least 2 others are recorded as having fallen seriously ill for months afterward (no details are available on the others). Almost all of the 27 women who attempted abortion did so for reasons of “social crises,” 15 of them to maintain their “chaste widow” status (lest they be deprived of property claims, for example), and 11 to conceal forbidden incest. In addition, the costs recorded (in 8 of the cases) varied from 3 to 7 silver taels, this at a time when an agricultural laborer’s annual wage ran 6 to 7 taels. In other words, abortion by pre-modern methods, most especially by the use of “abortifacients,” was physically dangerous and monetarily expensive, and was usually a last-ditch resort in crisis situations. Evidence from the Republican period gathered in a 1928 study by J. Preston Maxwell, British medical missionary and Professor of Obstetrics and Gynecology at Peking Union Medical College confirms the above picture, as does anecdotal evidence on traditional abortion methods from post-1949 rural China. It is therefore not surprising that the Qing code contained a 1740 substatute that specifically punished those illicit sex offenders who caused the death of the woman involved by arranging for the use of abortifacients (substatute 299–11: 用药打胎以致堕胎身死): they would be punished by analogy to the statute on killing by administering poison 比照以 毒药杀人. In Republican law, similarly, induced abortion was forbidden, leading Maxwell to title his study of abortion by traditional methods in general “On Criminal Abortion in China.” Overall, the evidence amassed by Sommer is so compelling that it can only leave us wondering how and why some scholars have managed to construct a picture so contrary to social realities. Putting matters positively, the article is a fine demonstration of how legal case records can serve as powerful material evidence for the social lives of common people in historical periods when such evidence is not otherwise available. Part Two. Buying and Selling of Land, and Homicides Christopher Isett employs the archival case records kept at the Liaoning Provincial Archives (of the joint court sessions of the Shengjing Board of Revenue 盛京戶部 and the Imperial Household Department in Shengjing

8

Huang

盛京內務府) and at the Jilin Municipal Archives (of the Bodune 伯都納 military yamen) to detail the practice of land buying and selling in northeast China (Fengtian Prefecture 奉天府 in the Qing), originally a protected preserve for the ruling Manchus prior to immigration by the majority Han people. By official court order, the land of Manchu bannermen could not be sold, only leased out (to local Han people). In actual practice, however, “conditional selling” 典卖 of land occurred very frequently, with the dian-maker (seller) and the dian-holder (buyer) representing the transaction as renting/leasing land, and the added payment(s) that came with market appreciation of land values 找价 as rent increases. The “legitimacy” (and enforceability) of the transactions relied not on the laws of the state but rather on the customary practices of society—community recognition, the use of middlemen, and kin or community mediation in the event of disputes. However, if disputes should go on to the official courts, the two parties would then be faced with possible suppression or punishment. From the point of view of the judicial officials on the scene, they were faced with two simultaneous concerns: one was to follow the letter of the law, and the other was to make allowances for people’s livelihood and the maintenance of social stability. Under the unavoidable tensions between the two concerns, local officials followed different strategies of action depending on the circumstances: they may enforce the letter of the law strictly (confiscate the land and punish the offender), act flexibly (confiscate the land but not punish the offender), or allow appropriate compromises (allow the offender to continue to farm the land, or do so for a given period of transition). What we see, therefore, are not only the disjunctures between law and practice, but also the interactions and accommodations between the two. Jennifer Neighbors’ dissertation studies in detail homicide-related laws of the Qing and the Republic, focusing especially on the differences between the two systems in their conceptualizations of homicidal intent and the legal practices stemming therefrom. It proceeds along two axes of comparison: between the codified texts of Qing law and of Guomindang Republican law (practically the same as modern Western Continental law), and between codified text and legal practice. What is included in this volume is her analysis of “negligent killing.” It demonstrates first how very different the two conceptualizations were: Qing law included under negligent killings 过失杀 completely accidental killings, or what “the ears and eyes could not reach, and what contemplation could not attain” 耳目所不及, 思虑所不到, while Republican law, under the influence of (Western) Continental Law’s conception of negligence, included only acts with some degree of fault. What is demonstrated here is not only Qing law’s broader conception of responsibility in such “killings,” but more importantly Qing law’s finer distinctions of intent, under

Editor ’ s Introduction

9

its “six killings” categorization: premeditated murder 谋杀 (as with poison), intentional killing 故杀 (as in the heat of the moment), killing in an affray 斗殴杀 (further differentiated by the nature of the weapon used), killing at play 戏杀 (further differentiated by whether dangerous games were involved, as in fencing or boxing), mistaken killing 误杀, including killing from negligence (as in playing with fire or bow and arrow), and, finally, accidental killing 过失杀 (as by a runaway horse or cart). It was precisely because Qing law drew finer and more concrete distinctions in degrees of intent than the newly imported concepts and their two simple categories of intentional and negligent killing that Republican judges tended in actual legal practice to continue to rely more on Qing categories than the new Republican ones. Such a finding about Qing legal thinking, of course, runs directly counter to the conventional (and Weberian) notion that traditional Chinese law tended to be limited to concrete descriptions of fact situations and was unable to engage in abstract conceptualization. Part Three. Tax, Education, and Local Governance Parts three and four of this volume are studies of the late Qing and the Republic, mainly in the first four decades of the 20th century. The three articles in Part Three are concerned mainly with issues of local governance, raising questions and criticisms of the conventionally employed binary between state and (rural) society to demonstrate a very different historical reality. Huaiyin Li has engaged in a book-length study of Huailu 获鹿 county of Hebei province, using its rich late-Qing and Republican archival materials, especially those pertaining to tax collection, to explore in detail rural governance in that critical transition period between the old and the new. What is included here is a revised version of Chapter 5 of his monograph, focusing on three different methods of tax collection used in the county: in one, the local xiangdi 乡地 (comparable in functions to the xiangbao 乡保 elsewhere, but here staffed by villagers in rotation) first advances the tax payments due (using village public funds or loans) 乡地垫款 and then collects monies from the local tax-paying households. In the second, taxed households pay directly to the county government (at its collection station 自封投柜), and the xiangdi’s role is limited to that of collecting delinquent payments 催粮 乡地. In the third, used for village land owned by people outside the county (called 寄庄地, or non-resident land / enclaves), taxes are collected by the supra-village sheshu 社书 (responsible for maintaining tax registers) in a tax farming 包买 arrangement with the government. Of the three, Li demonstrates that the first was the most effective and engendered the fewest disputes, mainly because it was based on “village regulations” 村规 and the voluntary

10

Huang

participation of the villagers. For Li, this model tells about “substantive governance” (in contrast to formalistic governance by bureaucratized state organs and officials) and demonstrates the inadequacies of the conventional binary of state vs. society. Elizabeth VanderVen similarly uses a late-Qing and Republican-period county archive (Haicheng 海城 county in Liaoning province of the northeast), but her focus is on village education. The article here comes from her monograph’s chapter on the financing of village education. On the basis of surprisingly rich source materials, she demonstrates that in this sphere, at least, the relationship between state and village at this time was not a simple one of resource extraction or state control, but two-way cooperation, in which the villages demonstrated active agency propelled by their own community interests. Haicheng county established village-level “public (community) primary schools” at the beginning of the 20th century in response to calls for such from the government (already by 1908, there were 333 such schools in the county), but those village schools received no government funding and relied on their own community resources, including land, public fund raising, private donations, and student tuition. (The taxes paid by the village to the government, including the extra-statutory special levies 摊款, all went into the county government coffers, of which only small amounts were spent for higher-level public schools, all well above the level of the village). VanderVen provides rich information about both funding and expenditures of the village schools, culled mainly from the dispute “cases” involving education handled by the county government, and also from the reports submitted by the county government’s Education Promotion Bureau 劝学所 that drew on the participation of local elites. On the basis of such evidence, VanderVen demonstrates the cooperation between state and society and the active agency of village communities, so very different from what a state vs. society dichotomy might lead us to expect. Danny Hsu’s article draws on county archives to study administrative litigation in the Republican period. Unlike most other research on administrative litigation, Hsu’s focuses on the sub-county level. It is based on the archives (a total of 11,000 juan [bundles] of materials) of Xinmin 新民 county, kept at the Liaoning Provincial Archives in Shenyang, and 200 corruption cases of the early Republican period contained therein, and another 50 cases of the later Republican period from the provincial Civil Administration Department 民政厅 of Sichuan, kept at the Sichuan Provincial Archives. Xinmin county established the sub-county qu 区 or “ward”-level administration in 1908, whereas Sichuan did not do so until 1935. As Hsu demonstrates, what is shown by these archival records of administrative litigation involving official corruption is not a simple matter of tensions stemming from the penetration of gov-

Editor ’ s Introduction

11

ernmental authority into society with “modern state-making” (as is assumed by many), but rather the more complex interplay between governmental authority and different village-level interest networks, some of which were opposed to the new ward heads (and township heads) while others supported them. What the records show are crisscrossing tensions and conflicts among multiple power groupings and interests that go beyond any simple opposition between state and society. For this reason, Hsu advocates using a power network analysis rather than a state vs. society frame to analyze the process of modern state-making. Part Four. Concubinage, Spousal Abuse, and Transnational Families Lisa Tran studies concubinage under Guomindang law, in comparison with Qing and early Republican law. In the Qing, in both law and in social practice, concubines 妾 were “minor wives,” though lesser in status than the main wife, but still recognized by law and given certain rights—for example, when the main wife died, if there were no son, the concubine who remained “chaste” could enjoy the powers of choosing an heir and wielding custodial power over the husband’s property. Guomindang law, by contrast, adopted the legal principle of monogamy, this despite the widespread social practice of concubinage, especially among the upper classes. By the letter of the new law, concubines had no status, becoming almost non-persons; they were legally not a wife, not even a minor wife such as under the Qing. Yet, at the same time, the law must somehow deal also with the social realities of large numbers of mistresses and de facto concubines in addition to wives. Law-makers (a number of whom had concubines themselves) attempted to erect a workable principle for differentiating between legal and not legal marriages by focusing on the wedding ceremony: if a public ceremony were held with at least two witnesses, then the ceremony was legal and the woman was a wife; if not, then not. This standard helped to distinguish between those ceremonially married and those not, but it also had multiple consequences that were not intended: since, in social practice, some kind of ceremony usually accompanied the taking of a concubine by a man, the concubine could use the fact of a ceremony to claim legal status as a wife; at the same time, however, she faced the prospect of conviction for bigamy for that very same reason. Thus, codified law and unintended legal practice both serve to illustrate the confusing and contradictory realities of a transitional period between the old and the new. Margaret Kuo’s evidence, mainly from cases of the Hebei Superior Court, tells first of all about how very difficult it was for a woman to obtain divorce under Republican law, despite its allowance for “intolerable cruelty” as legally acceptable grounds for divorce. Of the five divorce lawsuits she narrates in

12

Huang

detail, three were denied, despite clear evidence of physical abuse (in two of the three; the other involved a husband who gambled and was an opium addict while the wife had to work in a factory to support the family), and also despite the persistence of the wives in appealing to the Superior Court. Of the two cases in which the plaintiff wives obtained court approval for divorce, one involved a woman who was nearly killed by her husband and had wounds to prove the abuse. In the other, the husband was a repeat offender, having previously agreed to a court-mediated settlement (after severely injuring his wife) to refrain from any further abuse. Republican judges, as Kuo observes, tended to sympathize with the husbands’ arguments that they had paid a heavy brideprice and should not be left without a wife for just “minor or occasional injuries.” And judges, as Kuo further notes, imposed strict standards of proof and a very narrow definition of what constituted sufficient cruelty, accepting only abuse “of the direst conditions” as grounds acceptable for divorce. Therefore, only a small minority of women seeking divorce on grounds of abuse succeeded in gaining court approval. Nevertheless, it does seem clear that these women could not have made similar claims in the Qing, much less persist with appeals to upper-level courts. To that extent, Kuo is justified in her argument that the coming of the new Republican law shows new rights consciousness on the part of women, rights that accompanied the arrival of the “modern, liberal form of conjugal patriarchy.” Huey Bin Teng’s article is about “two-headed families” (liangtoujia 两头家) formed by married men from Fujian who migrated to Malaya to work and then married a local wife there. Those two-headed families were subject to two legal systems: at one end in Fujian, they were governed first by the Qing code and then by the Guomindang code. By the Qing code, the original wife was the main wife, and the second wife, a concubine or minor wife, each with different legal rights (for example, if the husband died and there were no son or only a minor son, the main wife enjoyed custodial rights, even if the son were the issue of the concubine and not herself. If the main wife died, then the minor wife enjoyed those rights.) As for the Guomindang law that followed, the second wife had no legal status or claims at all. At the other end of these two-headed families in British-ruled Malaya, on the other hand, the operative British colonial law was based partly on British common law, but also deferred in principle to pre-existing local custom. The pertinent result here under those twin principles was that both wives were treated as legal wives, giving them equal rights, without distinction between main and minor wife. Under those conflicting laws, once the husband died, conflicting claims between the two ends of the family were inevitable. There arose as a consequence the institution of the (clan or community) mediator (gongqin, 公亲) who tried to work between the two sides and

Editor ’ s Introduction

13

the two conflicting legal systems to arrive at compromises acceptable to both. Teng’s sources include lawsuit records, mediation records, and private documents and letters. Her materials show graphically the complex differences and points of conflict between the two legal systems, thereby illuminating both. They also provide rich ethnographic information on the social lives of these distinctive “transnational” families. It gives us the most finely textured look to date at the distinctive social-legal phenomenon of the liangtoujia.

Part Five. Past and Present: Local Administration and Court Mediation The fifth part contains research that is concerned with both the past and the present. American historians studying Chinese history rarely engage with major contemporary issues the way many American U.S. historians do—perhaps in part because of a sense of marginality (of “foreign area studies”), the more so among Chinese-Americans, and also perhaps in part because of “ivory tower” values of pure scholarship. In the China field, moreover, post-1949 China is generally considered the exclusive domain of social scientists, most especially political scientists, and historians who stray across that divide face the problem of being seen as “interlopers.” Perhaps for those reasons, very few of our students in the UCLA program have ventured to study contemporary China. Even I myself, despite a strong emotional attachment to China, generally only thought about but did not write about contemporary China. However, in this past decade of teaching in China after retirement from UCLA in 2004, I have turned to writing mainly in Chinese and for a Chinese audience, and have found myself drawn irresistibly into contemporary issues, in part because of the sense that the research in which I am engaged is of central concern and importance to China and in part because of my own conviction that a historical perspective is the best one from which to think about the present. This does not mean that I have come to write public commentaries, but that my scholarly concerns have now taken on a deeper engagement with questions of immediate contemporary relevance. The two articles included here are among my early efforts to join history to the present. The first article begins by summarizing the evidence that my students and I have accumulated about basic-level Chinese governance in the past to demonstrate that “centralized minimalism” 集权的简约治理 has been a major and abiding method of Chinese rule, with highly centralized administrative power at the top but minimalist approaches at the bottom. The latter tendency is shown by the broad reliance on semi-official personnel who were nominated by the localities (and confirmed by the county government) and who operated without salary or bureaucratic paperwork. Those quasi-officials, at the

14

Huang

critical juncture of the state and the village, were in fact identified with the interests both of the government and of their communities. The county government as a rule left them to function on their own, intervening only in the event of complaints or disputes (or change of personnel). It was an approach to administration that was very judicial—with the government intervening only in the event of disputes—something evident in the spheres of taxation, public security, and judicial administration. It is evident even in the way the Qing county magistrate related to the different offices of his yamen (intervening only to resolve disputes). It is also evident in the “village heads” 村长 system introduced at the beginning of the 20th century and in the establishment of new public village schools at that time. In (what Chinese historians term) the “modern” 现代 (i.e., 1912–1949) and the contemporary 当代 periods (i.e., since 1949), the apparatus of the state did undergo a considerable degree of elaboration and “bureaucratization” (in the Weberian sense). But, at the same time, much has remained of the old approach, even in the contemporary period, as can be seen in village governance during the collective era (with village heads and party-branch secretaries coming generally from the village community itself and funded by the community). In the Reform era, the same basic method remains all the more evident with the withdrawal upward of party-state control and the increasingly widespread resort to the “letters or visitations / appeals upward (to higher levels of the administrative apparatus)” 信访、上访 system for handling tensions between local society and the government. This “centralized minimalism” method of local administration cannot be understood in terms of mainstream Western theories that do not as a rule consider the overlapping intermediate realm between state and society. This method of minimalist governance, most especially the use of state-initiatives with community participation for public services, is something that may yet find much contemporary relevance—a useful resource in the search for a distinctive path toward Chinese “modernity.” The second article is an overview synthesis of Chinese uses of court mediation, past and present. The article first distinguishes between genuine mediation that is voluntarily accepted by the disputants and ostensible mediation that is imposed against their will. This distinction places into perspective the much exaggerated propagandistic claims made for court mediation in the Maoist period. By examining genuinely mediated cases, we can discern how very different the operative judicial principles and legal reasoning for mediation are from formalist law. The latter proceeds from certain given premises (e.g., individual rights), employs deductive reasoning to make them applicable to all fact situations, and sets up an unavoidably adversarial system of right vs. wrong, winner vs. loser. The former, by contrast, emphasizes the mediatory

Editor ’ s Introduction

15

ideal of peace-making compromise, insists on illustrating abstract principles with concrete fact situations, and is always concerned with practical, workable solutions. This kind of approach can avoid the absolutizing tendencies exhibited by legal formalism and also the simply retrospective tendencies of some strains of legal pragmatism or empiricism. Such Chinese mediation has in fact been a source of inspiration for the pursuit in recent years in the West for modes of “alternative dispute resolution” (ADR). It has shown the strength of avoiding the polarizing tendencies of the Western adversarial legal system, but also the weakness of fuzzying up clear-cut cases of right and wrong (dubbed in Chinese “mixing up wet mud” 和稀泥). The modern court mediation system—a legacy of the Maoist period—has proven to be of much broader use than Western ADR, but it has also exhibited a tendency toward excessive resort to high-handed methods, such as the coercive “mediated reconciliations” in divorce cases of the Maoist period, which since the turn of the century have largely fallen into disuse. Nevertheless, the Chinese court mediation system (as well as extrajudicial mediation) still demonstrates much vitality, and might yet be able to combine well with newly imported formalist law. One useful approach, I suggest, might be to employ the following principle: in disputes that do not involve fault, use mediation; in those that do involve clear-cut right and wrong, use the Western mode of legal thinking to adjudicate and protect individual rights. Such an approach, of course, would be consistent with the practical-moralism mode of thinking in that it begins with the concrete fact-situation. The inclusion in this final section of only my own contributions is something forced by circumstances and not selected by choice. I hope that in the years to come, more and more Chinese as well as Western students will take on such historically based contemporary research. The volume to follow this one, based on students and younger scholars I have taught in China this past decade, will include more examples of such work than does this volume. The last article, by Chenjun You, a leader of the younger generation of Chinese scholars, provides a Chinese perspective on the work of our UCLA group. YOU is sophisticated beyond his years, is well acquainted with research outside the mainland, most especially in the U.S. and in Taiwan, and reflects that new tendency among the younger generation of Chinese scholars. He places our group’s work into the larger context of globalized studies of Chinese legal history and exhorts his Chinese colleagues to develop new materials and methods to advance mainland Chinese research. I would echo his exhortation here with an observation I made elsewhere earlier: the future vitality of Chinese legal history research depends very much on what happens with such research inside China: just imagine a situation in which American scholars

16

Huang

no longer pay attention to American legal history, leaving its study to just a (relatively small) group of Chinese specialists who write only for one another! The vibrancy of the field of Chinese legal history, in other words, depends very much on the engagement and quality of Chinese research.

Some Common Threads

In the context of the 1980s, there seemed little need to articulate just what it was that we were doing; it seemed enough to just do it. What we dug out from archival materials went counter to much that had been assumed about Chinese legal history. Most of us simply assumed that the evidence we gathered would speak for itself and that other scholars, like us, would form empirical judgments on the basis of the evidence. We were aware, of course, of the new theoretical fad of postmodernism, a central tenet of which is the denial of the possibility of proof for being a positivist claim predicated on faulty modernist assumptions about the truthseeking capacities of scientific research. Such postmodernist sensibilities, without doubt, have contributed much that is positive, and we ourselves have been deeply influenced by its insights about the neglect of subjectivity and its influence on one’s perception of evidence, the faulty belief in supposedly “scientific” quantitative data, the denigration of tradition and its continued relevance, and the implicit Western-centrism of modernism. But we never thought that those very sound critiques would be carried to an extreme—to the denial of the relevance of evidence and the reduction of all “facts” to subjective constructions. Clifford Geertz in fact likens the American court process to just contestations between two opposed sides and their hired guns, without consideration of the presence of a judge and a jury who more often than not act in good faith to arrive at conclusions about what is true on the basis of the evidence (Geertz 1983). And Edward Said, likewise, went to the extent of reducing all Western research on the non-Western world as just exercises in self-validation of the West and the modern, arguing that all assertions of “fact” are finally but representations of different varieties and degrees of subjectivity (Said 1978, especially pp. 272–273). Those extremes, we thought, were passing fads that would soon give way when good common sense returns. Little did we know that they would be powerfully reinforced by other major tides: the explosion of available information to an extent that cheapens all information as little more than what can be accessed by a click of the computer; the powerful alliance with neo-conservative orientations that share with postmodernism a similar disregard for evidence, even if for different reasons (once God’s

Editor ’ s Introduction

17

revealed fundamental truths or other kinds of faith-based truths have been grasped, there need be no attention to mere worldly evidence—such as on the supposed presence of weapons of mass destruction in Iraq). Even then, I personally was to be surprised even more by how those powerful tendencies in the U.S. would come to be magnified many-fold in China. First because of the political system’s continued insistence on ideological control, resulting ironically in a strong tendency to either ideologize all academic theories or else to reject them in toto for being ideological. Then also because of the novelty of many Western theoretical ideas and methods such that many who are influenced by them exhibit the kind of absolutist belief only the newly converted cling to. Indeed, many neo-liberals in China today make their neoconservative brethren in the U.S. look sophisticated and open-minded. And postmodernism would be propelled in China not so much by a genuine epistemological questioning of modernist assumptions, but rather by the deeply felt nationalistic impulse to “de-center the West.” The combination of those powerful currents have made for an academic environment in China even less inclined to careful examination of evidence than in the U.S. Add to that the material conditions of Chinese academic life: so underpaid as to require dependence on royalty payments by the thousand words (gaofei 稿费) to supplement one’s income, over-controlled by an education bureaucracy that understands only countable quantity and not genuine quality, imposed not just on professors but even on graduate students (in quotas for publication in bureaucratically graded journals), and the bureaucratic organization of learning into compartmentalized spheres that separate, for example, “legal theory” 法理 and “legal history” 法史 into segmented, mutually exclusive pursuits. Together these and other factors have made for an academic environment in which American fads turn into ideologies, American weaknesses into institutionalized bureaucratic practice, and American epistemological excesses into nationalistic commitments. The result is an even more pervasive neglect of careful attention to evidence. Even so, many good Chinese (as well as American) scholars continue to abide by their sense of truth and reality, and much good work is produced, despite an incredibly adverse environment. Such scholars, however, rarely engage in explaining just what it is that they do, and most steer clear of theoryideology. The result is the tendency for the unscholarly and the bureaucratic to dominate theoretical and methodological discussions. Given the current academic climate in China, and also in the U.S. if to a lesser degree, I have felt keenly that we need very much to sort out and articulate just what is it that we do and why. As part of that endeavor, I have paid even more attention to the available theoretical literature than I had in the

18

Huang

past (even though some empiricist historians, I know, had already found my work overly engaged with theory). I attempted at first to identify an established school of legal theory that would be consistent with our own tendencies, but my review of the existing literature past and present has not turned up any school that spells out what our own group have come to take for granted through our scholarly practice: a basic commitment to the ideals of scholarship and of truth-seeking, an emphasis on legal practice even more than theory, a basic sense of the importance of both historical background and social context for understanding law, as well as an engagement with theory and with contemporary relevance. The publication of this volume, the first in a series of such, is intended to illustrate what we do and what implications it might carry for legal theory and for present-day law-making. Let me just summarize briefly some of the basic commonalities in the work included in this volume. First and most obviously, the researches begin with archival case records. That tells about a basic outlook that the study of legal texts alone is not enough; law must be seen in conjunction with what is actually done. Only thus can we grasp what law really means for people’s lives. The latter concern, in turn, requires that we examine the social context and social bases of what we study, to grasp how law could have different meanings for peasants as opposed to the urban elite, for women as opposed to men, for the underclasses as opposed to the privileged. From that approach and perspective comes our view of legal change, historically and in the present. Abstract texts can disregard social realities, but legal practice cannot. No law can function well if it runs counter to social practices or is far removed from social reality. In those situations, there is either simply disuse or a process of accommodation of one to the other. Equal inheritance rights for male and female, for example, could not operate in a socialeconomic context of a peasant economy in which most women married out of the village and the parents in their old age could only depend on their sons for support. Yet it is also the disjunction between legal ideal and legal practice that can propel both social and legal change in a two-way interaction between law and society. For these same reasons, legal case records are major sources for ethnographic evidence about the daily lives of common people, evidence that would not otherwise be available for historical periods when the common people themselves, in contrast to the literate elites, left little record of what they thought and did. Thus, Kathryn Bernhardt is able to prove conclusively how women’s property rights changed greatly during the imperial era, when most past literature assumed no change. And Matthew Sommer is able to prove conclusively that abortion could only have been a dangerous and expensive option chosen

Editor ’ s Introduction

19

only in crisis situations, this despite the two bodies of influential literature that have insisted otherwise. Case records, we have seen, have also provided new insights into the operations of local government, land transactions, village taxation, and education reform. Thus Bradly Reed provides conclusive evidence that yamen functionaries, long denigrated as corrupt and abusive by official representations, in fact provided regular and necessary services under customary standards that fell somewhere between the statutory and the unacceptable. And Christopher Isett is able to demonstrate how bannerland was regularly bought and sold conditionally in northeast China, this despite their false representations as rental transactions. As for villages, Huaiyin Li, Elizabeth VanderVen, and Danny Hsu have together demonstrated the inadequacies of a binary construction of state vs. society when it comes to Chinese village taxation, education reform, and power relations. The evidence points to a large intermediate sphere where communities and the state could cooperate and to power networks that crisscross both. Tran and Kuo, again on the basis of case records, demonstrate how very differently the law operated vis-à-vis women seeking divorce and concubines pursuing their interests, despite the formal provisions of the law. And Huey Bin Teng, finally, shows the true social and legal implications of two different legal systems operating in conflict upon the transnational families of Fujian and Malaya. Finally, the first of my own two articles demonstrates a basic operative principle of Chinese administrative practice, that of using judicial methods in minimalist administration—one manifestation of the continued inseparability of law and administration 政法 in Chinese history, past and present. The second shows how Chinese community and court mediation operate by judicial principles very different from modern Western law, how that system changed over time, and how it has continued to remain vital down to the present. The heart of our method, and what we share in common, is perhaps the very simple point that these research findings should be judged by their evidence, regardless of what one’s theoretical and ideological persuasions might be. Unspoken here is also the conviction that arguments and concepts need to be built from evidence, rather than driven by theoretical “hypotheses” that then seek out evidence in support. Our method is to go from evidence to theory and back to evidence, not the reverse. As for fuller articulation and illustration of (what I would term) the “historical-social study of legal history” and of “historical-social jurisprudence” (i.e., the theoretical implications of such historical-social study of law) than what has been roughly outlined above, we leave to the volumes to follow this one.

20

Huang

References Bodde, Derk and Clarence Morris. 1967. Law in Imperial China, Exemplified by 190 Ch’ing Dynasty Cases. Cambridge, Mass.: Harvard University Press. Geertz, Clifford. 1983. “Local Knowledge: Fact and Law in Comparative Perspective.” In Clifford Geertz, Local Knowledge: Further Essays in Interpretive Anthropology. New York: Basic Books, pp. 167–234. Huang, Philip C. C. 1982. “County Archives and the Study of Local Social History: Report on a Year's Research in China." Modern China, 8.1 (January): 133–43. ———. 1985. The Peasant Economy and Social Change in North China. Stanford: Stanford University Press. ———. 1990. The Peasant Family and Rural Development in the Yangzi Delta. Stanford: Stanford University Press. ———. 1991. “The Paradigmatic Crisis in Chinese Studies: Paradoxes in Social and Economic History.” Modern China, 17.3 (July): 299–341. ———. 1998. “Theory and the Study of Modern Chinese History: Four Traps and a Question." Modern China, 24.2 (April): 183–208. Le Roy Ladurie, Emmanuel. 1974. The Peasants of Languedoc. Tr. John Day. Urbana: University of Illinois Press. Lefebvre, Georges. [1934] 1959. Les paysans du Nord pendant la revolution Francaise. Bari: Laterza. Said, Edward. 1978. Orientalism. New York: Pantheon. Shiga Shūzō 滋贺秀三. 1981.《清代诉讼制度における民事的法源の概述的检 讨》(A general analysis of the origins of civil law in the litigation system of the Qing)《东洋史研究》40.1: 74–102. ———. 1984 《清代中国の法と裁判》(Law and justice in Qing China). 东京: 创 文社。

Appendix “Philip Huang-Kathryn Bernhardt Collection Checklist,” Stanford University East Asia Library (斯坦福大学东亚图书馆藏黄宗智-白凯档案库清单) 县级档案 巴县档案。土地、债务、婚姻（奸情）、继承类，1760–1859年，共300 起案 件。复印件，20卷，共3,996 页。

Editor ’ s Introduction

21

顺天府宝坻县档案, 1810年代到1900年代。缩微胶卷。 共135 盘, 333卷 。 顺天府宝坻县档案。土地、债务、婚姻（奸情）、继承类。118起案件。1810 年代到1910年代。复印件，8卷，共1621页。 淡水－新竹档案。“民事”档案（据戴炎辉分类编目）。1830年代到1890年 代。复印件，30卷，共6915页。 民国时期（四川）宜宾县、（浙江）乐清县、（江苏）吴江县民事案 件。120起案件。复印件，2132张。 顺义县档案，1910年代到1930年代。67卷，9305页。包含128起民事案件、刑 事案件、有关区-村政府档案、司法统计材料、区政府按户登记材料以及各 种社会经济调查报告。多年来用作研究生档案使用锻炼材料。 双城县档案目录, 1912–1937年。2万卷档案的案件目录。4卷，909页。 小计：120 卷, 24878页, 135盘缩微胶卷 北京市档案馆: 北京市地方法院， 1920年代到1940年代。告状与判决书。 离婚案件，225起，11卷，2025页。 婚姻案件，96起，4卷，490页 继承案件，156起，19卷，3300页 赡养案件，55起，3卷，300页 小计：532起，37卷，6415页。 第二历史档案馆： 大理院案件（判决书） 离婚与婚姻案件，1914–1918， 83起，3卷，515页。 继承与赡养案件，58起，3卷，525页。 京师高等审判厅案件（判决书）

22

Huang 离婚与婚姻案件， 1913–1925年，233起，8卷，1660页

大理院与京师高等审判厅 有关物权案件（多是继承案件，也有债务和典权案件） ，1912–1924年，246起 （大理院107起，京师高等审判厅139起） ，13卷，2540页。 小计：620起，27卷，5240页。 总计：共180卷，36533页，135盘缩微胶卷。

Part One

Analytical Approaches: History of Practice · Women’s History · Local Administration · Discourse Analysis · Case Records as Ethnographic Evidence

∵

chapter 1

The History-of-Practice Approach to Studying Chinese Law (Introduction to Chinese Civil Justice, Past and Present)* Philip C. C. Huang The study of legal history generally emphasizes legal codes more than the actual workings of the law, and legal theories more than actual legal practice. My own research these past decades however has found that what is codified text in the law is one thing, and what is done is often quite another thing. Codified law may ignore social economic realities, but in actual implementation it must accommodate those realities. A grasp of the past and present of law therefore requires attention not to just one dimension or the other, but to both and to their interactions over time. Tensions between “representation” and “practice” are in fact often the motive force that drives changes in law. To grasp that process of change, it is best to begin by looking at real practices rather than to assume, as so many do, that practice will necessarily correspond with codified text. The method I wish to advocate is therefore the study of the “history of practice.” The word “practice” here carries mainly three overlapping yet somewhat different meanings. First is practice in its conventional sense, namely, practice as opposed to theory, referring mainly to action. This meaning is close to and yet not quite the same as a second meaning of “practice,” as used in the sense of my 1996 book Civil Justice in China: Representation and Practice in the Qing, where it is juxtaposed against “representation.” Both of these meanings should be distinguished from a third meaning of “practice,” as in Pierre Bourdieu’s (1977) usage, where practice is juxtaposed mainly against institutions and structures. In the tradition of the Chinese Revolution, “practice” was used in a sense close to yet narrower than conventional usage, and the term referred mainly to the problem of how to apply “universal” (Western) Marxist theory to the Chinese Revolution. But since that usage spotlights the issue of disjunction between theory and practice (and therefore required the intermediation of * This was first published as the introduction to my book, Civil Justice in China, Past and Present, Rowman & Littlefield, 2010. Minor alterations have been made for publication in this volume.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_�03

26

Huang

“Mao Zedong thought”), it is directly related to the central theme of this book. My own work has stressed that the disjunction between code and practice is not just a matter of a gap between imported Western law and Chinese practice, but also between China’s own representation of the law and its actual practice. Bourdieu’s concept of “logic of practice,” on the other hand, calls for searching out the “logic” contained in the process of practice, rather than in institutional structures, in order to uncover the fundamental characteristics of a society and to transcend the longstanding problem in Western thought of binary oppositions between the subjective and the objective, and structure and agency.1 In addition to these three meanings of “practice,” I wish to place particular emphasis on the historical dimension. Mao Zedong’s idea of practice attended to time and space, and yet was also intended to be a universal transcending time and space. My “history of practice,” by contrast, aims mainly to spotlight the historicity of the humanistic and social spheres of life. Bourdieu, of course, emphasized process, and distinguished it from institutions and structures, but his concept of process paid little attention to history, and was mainly the conceptual tool of an anthropologist focusing on contemporary society.2 The difference between “history of practice” and simple “practice” is that the former incorporates not only the kinds of practices outlined above but also their interactions over time with theory, representation, and institutions, as manifested also in practice. Let me turn to illustrate these points with examples.

Practice as Opposed to Theory: Legal Formalism and the History of Practice of American Law

What will be explained here is first the difference between practice (meaning action) and theory. Chinese scholars often emphasize how Western theories are abstractions from Western experience and do not accord with Chinese reality. That point, of course, contains its grain of truth, but what I wish to 1 Clearly, “practice” as used here overlaps with and yet is different from Aristotle’s usage in his three-way distinction of theoria, praxis, and poiesis (production). It is broader than Marx’s usage of “revolutionary praxis” to mean the merging of theory and practice. It is also broader than James Scott’s recent usage (Greek mêtis) to mean mainly the perfecting of techniques through repeated use (Scott 1998). 2 To be sure, Bourdieu’s “habitus” seeks to deal with cumulative experience as an intermediary between structure and agency, but the concept of history used in this book is much broader than Bourdieu’s.

The History-of-Practice Approach to Studying Chinese Law

27

emphasize here is that Western theory, especially theory that has been turned into ideology sponsored by the power of the state, also does not accord with the West’s own history of practice. America’s so-called “classical orthodoxy” tradition of legal theory is a good example. Its origins are usually traced to Christopher Columbus Langdell, who became dean of Harvard School of Law in 1870. Because this tradition emphasized especially the universality and scientific nature of law and the use of deductive theory to arrive at absolute truths, scholars have also dubbed it “legal formalism.” From the 1870s to the 1920s, it did in fact occupy a dominant position in the American world of legal studies. Yet at the same time, it was challenged from the start by Langdell’s colleague at Harvard Oliver Wendell Holmes (later a Supreme Court justice). Holmes stressed especially the historical nature of law, rather than its universality beyond time and space; he also argued that law must be tested in actual use and evaluated on the basis of its practical consequences. Scholars generally consider Holmes the founder of the American tradition of legal pragmatism. This tradition was followed in the 1920s by the new legal realism. (Grey 1983–1984; Tamanaha 1996; Hull 1997) At the level of legal practice, Supreme Court decisions have not come simply from the “classical orthodoxy” tradition of theory, but rather mainly from the long-term tug-of-war between orthodox formalists and their pragmatist-realist challengers. The former dominated the court until the 1920s until, under the presidency of Franklin D. Roosevelt, the latter came to occupy a majority among the nine Supreme Court justices. Many important American labor and welfare decisions of the Court come from that period. (Wiecek 1998) The non“orthodox” jurists would maintain their upper hand in the Court until, under the neoconservative administration of George W. Bush, jurists of the orthodox persuasion once more occupied the majority. Looking back at the history of practice of American law, it is clear that its real essence lay not in any single theoretical tradition, but rather in the long-term coexistence and competition among its different theoretical traditions.

Practice as Opposed to Representation: Qing Law

Because of the apparent disconnect between Western theories and Chinese realities, some Chinese scholars have advocated the use of China’s own concepts to generalize Chinese experience, but what I want to point out here is that China’s own representations can also be at odds with its history of practice. My first volume, about Qing justice, emphasized how Qing representations and practice diverged from one another and yet were combined to

28

Huang

form a single system. That is to say, between representation and practice there were disjunctures and tensions, yet also mutual adaptation and joining together. More concretely, Qing representations give us a three-point picture. First, there were few civil lawsuits. In the official ideology of the state, such lawsuits should not happen and, even if they did, they were only about “minor matters,” which the central government would not be unduly concerned with and would leave largely to the local governments to deal with “on their own authority” (zili 自理). Furthermore, most “good people” (liangmin 良民) would not become involved in a lawsuit; if they did, then they usually did so only on the instigation of evil “litigation mongers” (songgun 讼棍) and “litigation masters” (songshi 讼师). Finally, when magistrates dealt with such minor lawsuits, they usually acted in the manner of “father-mother officials” ( fumuguan 父母官) handling disputes among children, employing mediatory methods, and edifying the “children people” (zimin 子民) with moral principles, without resorting to adjudication by law. Needless to say, these representations were closely tied to the Confucian ideology of benevolent government (renzheng 仁政). They have profoundly influenced past scholarship. However, archival records of actual cases demonstrate a very different picture. First, “civil” lawsuits accounted for about one-third of all cases at the county yamen level. This is my conclusion based on records from the three counties of Baxian in Sichuan, Danshui-Xinzhu in Taiwan, and Baodi in Shuntian prefecture. The so-called “minor matters” cases in Qing official representations in fact accounted for a significant and important proportion of cases handled by the local yamen. Second, most litigants were common people who went to court out of necessity and for the sake of protecting their rightful claims. From a total of 628 cases, I identified the background of precisely 500 litigants, of whom 189 were common peasants, 20 rural hired workers, 51 common landlords, 82 common town residents, 25 degree holders, and 33 merchants. The remainder were variously big landlords, big money-lenders, corporate entities, and so on. These were not consistent with the images given in official representations of litigants. Third, when the county courts dealt with such disputes, they either allowed the disputes to be settled by community or kin-group mediation or they made decisions on the cases at court in accordance with the law (without excluding, of course, appropriate considerations of qingli 情理, or human compassion/ relations and moral principles). The magistrates very rarely conducted mediation at court. Of the total of 628 cases examined, 221 went on to a formal court session (the others being resolved by community or kin mediation, or simply went unresolved). Of those, just 11 underwent a kind of arbitration, in which

The History-of-Practice Approach to Studying Chinese Law

29

the magistrate ordered both sides to compromise. All others were decided in clear-cut judgments for or against. From the actual records of cases, at least, Qing magistrates very rarely indeed undertook mediation in the manner of official representations. A moment’s thought will make clear that a magistrate was unlikely to, indeed could not have, undertaken the “Ma Xiwu style” of handling cases— referring to the manner in which Ma Xiwu, a superior court judge of the central Shaan-Gan-Ning base area (who was later to become vice president of the Supreme People’s Court), conducted on-site investigations and discussions with neighbors and kin and others concerned in order to arrive at a resolution, often through mediation—that characterized Maoist courts of the pre-1949 “Liberated Areas” and the post-1949 period before the Reforms. They had neither the intent nor the leisure. They generally simply “adjudged at court” (zuo­ ting pan’an 坐庭判案). Furthermore, even though in their writings they liked to represent themselves as morally superior gentlemen, in accordance with the habit of official discourse of the time, they were in reality the bottom rung of a complex bureaucracy, in which the safest behavior for the sake of their own advancement was to act according to law and to regulations. Moreover, litigants who persisted all the way to a formal court session were usually among the most truculent, or else entirely committed to their rightful claims in the belief that the other side had violated them. Such people could not easily be persuaded by mediation. Under those circumstances, the most realistic thing to do for the magistrate was to adjudicate outright. On this point, the famous Qing jurist Wang Huizu was the clearest and most realistic. As he put it, “While adjudication is done by law, mediation is done by human compassion.” The magistrate is someone who adjudicates by law (duan’an 断案). It is the kin and the neighbors who mediate, not the magistrate. He explains further, “When it is a matter of law, then there has to be a clear-cut position for or against. But when it is a matter of human compassion, then right and wrong can be compromised some.” What he means is that once the case is at court, the magistrate can only act according to the law, distinguish clearly between right and wrong. That way, the losing side is likely to harbor lasting animus, and the two sides are likely to remain hostile toward each other for a long time, unlike the cordial resolution that can be arrived at through mediation by friends and kin, and no lasting animosity results. Precisely because he believes that the magistrate must adjudicate according to the law, he admonishes them to regularly study the code, “whenever one has time after a day’s work, one should study carefully one or two statutes and substatutes. Within a few months, one will be familiar with the essentials.” (Cited from Huang 1996: 205–6.)

30

Huang

By the formalist logic of Western Continental Law, the Confucian ideals and actual Qing legal practice would seem to be contradictory (and therefore cannot coexist); but for Chinese jurists, there was no problem here of logical inconsistency. Confucian moralizing has to do with the ideals of the legal system, while actual application can allow for practical principles and for adjudication, even if those seem opposed to the stated Confucian moral ideals. In Chinese legal reasoning, Confucian moralistic representations are about what ought to be, while the practical provisions and the adjudicatory actions of the judges are about adapting to real-life circumstances that cannot fully accord with moral ideals. Actual realities determine certain kinds of actions, just as Confucian teachings point to a picture of an ideal world. The main argument of my book Civil Justice in China: Representation and Practice in the Qing can be summarized this way: in the history of practice of Qing law, what was said was one thing; what was done was another; and when the two were joined together, that was yet another thing.

Practice as Opposed to Institutions: Male and Female Inheritance Rights and Their Actual Operation

“History of practice” includes also the meaning, à la Bourdieu, of distinguishing between institutional structure and actual operation. For example, the Guomindang’s 1930 Civil Law adopted from German law the principle of gender equality in inheritance, but in actual operation, the new law was not applied in the countryside. This is a conclusion I reached on the basis of 247 cases from the four counties of Shunyi (Republican Hebei province), Wujiang (Jiangsu), Yibin (Sichuan), and Yueqing (Zhejiang). The reason was simple. At the time, most rural women “married out” into other villages, and parents in their old age had to be maintained by sons who remained in the natal villages. Their land therefore could not go to their daughters. For this reason, the courts of the Guomindang period in practice for the countryside generally only recognized the inheritance rights of the sons. But lawmakers did not revise the text of the code accordingly; instead they permitted the legal provisions (institutional structure) to coexist with incongruent practice, amounting, in effect, to closing one eye over rural customary practice. This disjunction between codified legal provision and judicial practice was not dealt with formally in legislation until the 1985 Law of Succession of the People’s Republic of China, when the right to inheritance was linked by law to the obligation for old-age maintenance. That way, rural sons were to inherit their parents’ property because they fulfilled the obligation for their old-age

The History-of-Practice Approach to Studying Chinese Law

31

maintenance, not because they were male; daughters, if they fulfilled the obligation for the parents’ old-age maintenance, could likewise inherit their parents’ property. Thus did legislators both uphold the principle of gender equality in inheritance rights and attend to rural realities, thereby resolving the inconsistency between institutional structure and legal practice. This was a resolution that was arrived at only after many years of actual practice. It was a principle that cannot be found in Western law. At least on this matter, legislators of the People’s Republic were not limited to the simple ideology of wholesale Westernization. Yet in the Reform period, that ideology has once more become predominant. But the history just summarized above demonstrates for us legislative innovation and its possibility.

The History of Practice vs. Formalist Theory

The reason I have chosen to emphasize the history of practice as the approach to the study of law is because formalism and its mode of thought predominate in today’s legal studies (as well as the various social sciences). Formalism is more concerned with theory than with reality, and it emphasizes especially deductive logic, aspiring to attain thereby universal truths that stand above the particularities of time and space. This kind of absolutist propensity lends itself readily to ideologizing by the state, as has happened time and again in Western history, beginning with the Reign of Terror after the French Revolution. Nineteenth-century imperialism and recent American neoconservatism are other particularly notable examples. Max Weber can be seen as one of the best representatives of the Western modernist tradition, and also one of its best analysts. He made clear the deepseated premise and belief of modern Western civilization, the key being its view of formal rationality. As Weber points out, the key difference between modern Western law and other legal traditions consists in its “formal rationality.” In Weber’s view, legal formalism in the modern Western Continental tradition demands logical consistency in codified law and also between codified law and its application to concrete fact situations. (Weber, [1968] 1978: 657.) Qing law, by comparison, in Weber’s view, was substantivist, or instrumentalist, concerned more with the ruler’s wish to maintain social order than with the protection of individual rights, and therefore also more prone to arbitrariness (Weber, [1968] 1978: 844–48). For Weber, even Anglo-American common law is a kind of “empirical justice,” and not the rational-formalist law he approves of, for it is based on precedents, not the universal principle of rights,

32

Huang

and it relies on a jury comprised of common folk, not specialists employing legal logic. (Weber [1968] 1978: 976, 891.) Weber’s views have profoundly influenced the study of Chinese law. In accordance with his views, scholars have regarded Qing law as containing only concrete and particular regulations, not universal abstract principles/norms. At the same time, they think that the Qing had little or no civil law to speak of. Even the fine study done by Derk Bodde and Clarence Morris (1967) was not able to rise above this kind of presupposition. And scholars have maintained that Qing judicial practice, because it did not employ the kind of reasoning used in formalist law, did not adjudicate in any true (namely, Weberian) sense of the word (see, for example, Shiga 1981: 74–102). Scholars have also argued that revolutionary China had no civil law at all to speak of. True Chinese civil law has existed only since the Reform period, with the massive importation of Western laws (William C. Jones 1987: 309–31). (Before that, only the Guomindang’s laws might be considered genuine law.) I will demonstrate at length that the true difference between Qing law and Continental formalist civil law consists not in Qing law’s lack of provisions to guide legal judgments but rather in its insistence on embedding its conceptual system in concrete fact situations for the purpose of resolving actual problems. Qing law did not try to develop abstracted principles of universal applicability; rather, it seemingly postulates that only when joined to real legal practice can abstract principles be properly clarified and carry real meaning and applicability.

Practical Moralism

Let me illustrate with an analytical theme of mine developed from the historyof-practice approach. It should be said at the outset that what I mean by the “history-of-practice” method is not a purely “empiricist” method of research. In my opinion, mere accumulation of empirical information does not have a great deal of meaning. Experience must be linked up with theoretical concepts. What I wish to advocate in method is to develop from the history of practice theoretical concepts appropriate to Chinese realities. At the same time, by “history of practice” I also do not intend a purely retrospective approach, but one that is accompanied by prospective (i.e., forward-looking) moral visions. What the method emphasizes is to search out, through actual history, not mere theoretical speculations, but avenues for pursuing given moral ideals (e.g., harmonious human relations, social justice, people’s rights).

The History-of-Practice Approach to Studying Chinese Law

33

This book will demonstrate in detail that a particularly important characteristic of Chinese law is that, with respect to the duality of experience and theory, it privileges experience, but not to the neglect of theory. What it seeks is the close interlinking of concept and empirical situation. In contrast to the European Continental tradition of formalist law represented by Weber, Chinese law has sought to embed abstract concepts in concrete fact examples, rather than to seek to elevate abstractions into universal principles independent of concrete fact situations. For example, on property rights, traditional Chinese law did not abstract the principle of private property rights in the manner of modern Western law, but rather, within the circumstances of the time, provided for legal punishments against “fraudulently selling [another’s] land or house” (daomai tianzhai 盗卖田宅) (as for example by falsely representing others’ property as one’s own to sell or forcibly occupying another’s property), against “stealing-eating the fruits of another’s plot or garden” (shanshi tianyuan guaguo 擅食田园瓜果) and other such invasions of others’ properties, and against “the young using [familial] wealth without [parental] authorization” (beiyou sishan yongcai 卑幼私擅用财), and so on. On marriage, as a further example, Qing law did not advance the abstract principle of marriage as a contract, but rather provided for punishments against various false representations and violations in a marital agreement, such as “promising again to marry another” (zaixu taren 再许他人), or “to represent a crippled person as not” (you canji zhe, wangzuo wu ji 有残疾者、妄作无疾), or “to forcibly marry before the agreed-upon time” (qiyue weizhi er qiangqu 期约未至而强娶), or “to deliberately not marry when the agreed-upon time arrives” (qiyue yizhi er gu weiqi 期约已至而故违期), and so on. The same approach was used with respect to debt and inheritance, the other two main spheres of civil law. Some scholars (including Weber) have believed therefore that traditional Chinese law attended only to specific concrete situations and lacked abstract concepts and principles, but this belief is based on a misunderstanding. Qing law’s difference from modern Western formalist law consists not in its inability to formulate abstract principles about nonconcrete problems, but rather in its different mode of thinking as to how to join experience and theory. Formalism demands that abstract universal principles be created through legal (deductive) logic, while Qing law asks that abstract principles be embedded in concrete examples. A very good illustration is Qing legislation on homicides. All the laws are organized around a very abstract issue—intent. Homicides were separated into six grades, depending on the degree of intent. The severity of punishment varied with the degree of deliberate intent (Xue Yunsheng [1905] 1970: 849–57). The distinctions employed, as we will see, were in fact finer than

34

Huang

the mere two categories of intentional killing and negligent killing adopted by the Guomindang from German law. Precisely for that reason, Republican judges in actual practice often turned back to Qing law for its concepts and distinctions to adjudge cases (for detailed analyses and documentation, see Neighbors 2004). Those Qing distinctions were not purely abstract concepts (intentional or unintentional), but rather came from closely joining together abstract concepts with concrete fact situations. These are points I make at some length in this book in chapters 6, “Civil Adjudication, Past and Present,” and 8, “Whither Chinese Law?” At the same time, Qing law was definitely not just a retrospective law, based only on what has happened in the past. (Some have criticized American legal pragmatism on precisely these grounds for its lack of a clear agenda for lawmaking, arguing that it is, in the end, mainly just a reaction against “classical orthodoxy.”) Rather, Qing law was a legal system with a strong prospective vision. Its view of the future of society rested on its moral visions, as for example the ideal of a harmonious society (of course, Confucian discourse equated this kind of ideal with the ancient era of the sage kings), in which there would be no lawsuits among the people, in which disputes would be resolved entirely by moral principles, and in which even when lawsuits happen, they would be resolved by the “father-mother officials” through moral edification. While harboring such moral ideals, Qing law in practice also very pragmatically established a judicial system for dealing with disputes over what it termed “minor matters” (roughly equivalent to the modern “civil” category), relying first on community and kin mediation, and then the “third realm,” involving an interaction between community mediation and court intervention (more below), and, finally, if the dispute was still not resolved, then by the county courts “on their own authority.” And magistrates, while they espoused Confucian moral discourse, in actual practice consistently employed the very practical method of outright judgments (duan’an 断案), clearly distinguishing between right and wrong according to law. I call this combination “practical moralism.” This is a theme that I developed in my first volume, Civil Justice in China: Representation and Practice in the Qing; the present volume develops the theme further in chapters 6 and 7. The purpose of using “practical moralism” to conceptualize Qing law is first of all to highlight its merging of moral representations with practical action, the two being at once inconsistent and yet combined together. This is what I characterize as: what this system said was one thing, what it did was another, but these two combined was yet another thing. Another purpose is to spotlight its rather distinctive mode of thought, which contrasts sharply with Weberian formal rationality.

The History-of-Practice Approach to Studying Chinese Law

35

This mode of thought was evident even in the Maoist period. To be sure, Maoist rule built a totalistic state ideology, making even more absolutist claims than Western modernism. But at the same time, it carried a tradition akin to pragmatism, which we can term its practice-orientation. That was born mainly of the Chinese Communist Party’s own experience in revolution, in reaction against the doctrinaire Marxism of the party’s early period. After the failure of the Great Revolution (in 1927), the party turned to the countryside, with which most party members of the time were not well acquainted, thereby setting the stage for rethinking the relationship between theory and experience. Later, in the War of Resistance period, large numbers of intellectuals from the coastal areas who made the trek to the Yan’an area had little understanding of the local situation and were, in fact, not even able to communicate with the local peasant “masses,” thus creating a critical problem for the party: how to unite the two groups. That was part of the historical background for Mao’s essay “On Practice.” He called for the intellectuals to go deep into the countryside, first to obtain “perceptual knowledge” and identify with the “class feeling” of the peasants, and then to raise the level of their understanding with the help of (Marxist) theory to arrive at a higher level of knowledge, which was in turn to be tested in actual practice. Based on that kind of revolutionary epistemological outlook, the entire party developed the demand to “investigate and do research” (“no investigation, no right to speak”) (Mao [1941a] 1971: 196), resulting in a particular style of operation for the entire party. Even today, many Chinese social science instructors take their students down to the countryside to do on-site investigations of concrete conditions, something hard to find among social science instructors in the United States. This is a point that I have made in two published essays in Chinese (Huang Zongzhi [Philip C. C. Huang] 2005a, 2005b). Even under the wholesale transplanting of Western laws in the Reform period, we can still discern the continuation of the original mode of thought that privileged practice. Here we might briefly introduce the illustrative example of tort law discussed in this book. It was imported from German law; Qing law contained no provisions about torts. The crucial concept is “wrongful acts” (i.e., violations of others’ rights). In the manner outlined by Weber, this principle starts from the doctrine of natural rights, deducing therefrom a host of provisions: wrongful acts stem from the fault of violating others’ rights, and fault results in the obligation for compensation; if there is no fault of violating another’s rights, then there can be no compensatory obligation to speak of. Chinese law, however, has lent all this a different interpretation. As I will show in detail (chapter 7), the 1986 “General Principles of Civil Law” lent this a distinctive reading based on the Chinese practical-moralism mode of

36

Huang

thought. First, in light of the empirical reality that civil damages occur both in disputes that involve fault and those that do not, the law must deal with both kinds of situations. For this reason, the “General Principles” provides, on the one hand, for obligations for compensation in the event of fault in violating another’s rights and, on the other hand, also provides that even in the absence of fault, the party involved might have to bear obligation for appropriate civil compensation. From the point of view of the formalist mode of legal thought, these two are contradictory provisions that cannot coexist. But for Chinese lawmakers, the key point is the reality of the different fact situations, and law has merely provided appropriate provisions accordingly. Logical consistency is simply not an issue. In my view, this is another example of how China can create its own distinctive modern law.

Divorce Law Practices and the System of Court Mediation

The history of the practice of divorce law reveals the same mode of thinking, and furthermore shows the nature of the very distinctive court mediation system that has continued to this day. In the Mao Zedong period, in response to specific historical needs, the system of court mediation was developed to operate over and above community mediation. The origins lay in the need to deal with strong rural opposition to the party’s early radical promise of freedom of divorce (to be granted upon either party’s petition); in the face of that opposition, the decision was eventually made to deal with disputed divorces case by case through mediation, in order to minimize tensions between the party and the peasantry. After a sustained period of practice, the legal principle emerged of using the quality of the emotional relationship of the couple (ganqing 感 情)3 as the standard to decide whether to grant divorce or not. The key question became whether the relationship of the couple “has truly ruptured.” That way, it was thought, the party could both avoid divorces based on so-called “bourgeois” cavalier attitudes of “liking the new and tiring of the old” (xixin 3 The Chinese term ganqing has no exact English equivalent. In the semiofficial Foreign Languages Press version, this stipulation is translated, “In cases of complete alienation of mutual affection,” which is close. But ganqing allows for gradations—“very good” (ganqing henhao 感情很好), “poor” (buhao 不好), “ruptured” (polie 破裂), and so on—while “mutual affection” does not. Here and later in this book I will render ganqing as “emotional relationship,” or simply “relationship,” which seems to me to come closest to capturing the meaning of the Chinese term.

The History-of-Practice Approach to Studying Chinese Law

37

yanjiu 喜新厌旧, referring especially to revolutionary cadres who came to prefer women comrades to their village wives), while still carrying forward the revolution against “feudal marriage” practices (that ignored the feelings of the couple)—such as polygamy, slave girls, tongyangxi,4 parentally imposed mates, and purchased wives. For this, a distinctive court mediation system was invented in the Mao period, developing in time an entire set of methods, procedures, and style of work. Judges were to go deep into rural village communities; “interview” ( fangwen 访问) the “masses” (kin, neighbors, and the local Party organization); and investigate and do research, in order to arrive at an understanding of the marriage’s background and present condition and analyze the roots of the couple’s marital problems (hunyin maodun 婚姻矛盾). They were then to intervene actively, using all kinds of methods, including political education, organizational pressure, material incentives, and so on, to try to save the marriage of the couple. The purpose was to arrive at “mediated reconciliations” of the great majority of the divorce disputes. On the basis of 336 cases from a North China county (A county) and a Jiangnan county (B county), this book documents in detail the above system. In the Reform period, this Maoist “Ma Xiwu style” of court actions, especially its more authoritarian parts, have fallen into relative disuse, but the other parts of the court mediation system that developed alongside divorce law practice, especially with disputes involving no-fault fact situations, continue today to form an important part of the Chinese legal system. Chapters 4 and 5 examine this historical process. If we shift our perspective at this point from the “what was” of the “history of practice” to the “what ought to be” of China’s search today for its own “modernity,” the standard of “whether the emotional relationship of the couple has ruptured” formulated during the course of the practice of divorce law may be seen as the “logic of the history of practice,” or of “modernity,” contained within Chinese marriage law. It is different from the “freedom of marriage” principle; it is a legal principle born of years of legal practice, and it was not formally incorporated into the codified law until the 1980 Marriage Law. In my view, there is every reason for this “tradition” born of China’s modern revolution to be developed further today in conjunction with the from-reality-to-legal-principle-to-practice mode of thinking of Chinese law. At the same time, what is shown here is the practice-based spirit of lawmaking in the People’s Republic: 4 Neither of the two commonly used English terms for tongyangxi—“child bride” and “adopted daughter-in-law”—is quite accurate, as I have pointed out elsewhere (Huang 2001: 160n). “Child bride” suggests that the girl was married very young, which was generally not the case. “Adopted daughter-in-law” is also misleading, for no formal process of adoption took place. I therefore use the original Chinese term throughout.

38

Huang

legal principles are incorporated into codified law only after an extended period of trial through practice. Western legal formalism takes the abstracted principle of rights as its point of departure and asks that it be applied to all fact situations, resulting thereby in a strongly adversarial legal system. However, in the real world, there are both disputes with fault and disputes without fault (in the sphere of divorce the West itself has since the 1980s turned to an approach that sets aside considerations of fault). In point of fact, Western legal systems today, faced with the high incidence of lawsuits stemming from the adversarial nature of its legal systems, are trying to find alternative ways to resolve disputes. China has accumulated much more experience in this regard than the West, and can develop further in this direction. Chapters 4, 5, 7, and 8 explore this question in detail. At the same time, the tradition of mediation and of practical moralism no doubt carry with them a tendency to scramble up (“mixing up wet mud,” he xini 和稀泥, in Chinese parlance) right and wrong and not distinguish clearly between disputes that involve violations of the law from those that do not, and between those with fault and those without. They can therefore lapse easily into treating one as if it were the other. Where the litigants involved are of unequal power, the system can easily slop over into abuse of power and connections. The importing of Western laws that are based on rights can be a corrective to such a tendency, and should be used in fact situations involving fault to clarify and protect rights, just as the Chinese mediation tradition can be employed in no-fault fact situations as a corrective for the excessively adversarial tendency of Western legal systems that insists on separating out right from wrong and winner from loser. In my view, consistent with the start-from-concrete-reality mode of Chinese legal thought, what might be done today is to adopt the following guideline: in fact situations that do not involve fault, employ mediation, including court mediation, because it has proven to be most effective in such situations; where fault is involved, then adjudicate by law to protect rights and thereby incorporate the strengths of Western law. At the same time, blind transplanting of Western systems should be avoided. One notable example is the recent changes adopted with respect to rules of evidence, which have completely discarded the accumulated experience of past practice under the People’s Republic, doing away with the system of having judges bear the principal responsibility for investigation and evidence gathering (characterized in Chinese usage as “judge’s authority-ism,” zhiquan zhuyi 职权主义), and adopting instead the West’s system of leaving the burden of proof and evidence to the litigants (what is termed in Chinese “litigant-ism,” dangshiren zhuyi 当事人主义). The intention for the changes was to protect

The History-of-Practice Approach to Studying Chinese Law

39

the rights of litigants to present evidence themselves. That is well and good for criminal law where the state is an interested party, but it is not entirely apposite for civil law that deals with disputes among citizens. Furthermore, in the sphere of divorce law practice, because of the lack of necessary accompanying institutional practices, especially that of the use of witnesses, what has resulted is a system that is simply not working as intended, with strong tendencies toward utter disregard of substance in favor of formalistic procedures. Bureaucratic attention to form, without regard to substance, has been among the worst problems. This kind of practice may be considered the negative unanticipated consequences of the changes in evidence law. This fact is documented in chapter 5 on the basis of forty-five recent cases from R county in the south. The point is, practice and the history of practice can be positive, and a resource for searching out China’s own kind of modernity, but they can also be negative, running counter to intended moral purposes. What is being emphasized here is the difference between a practicebased historical outlook and a theory-based historical outlook. The latter can lapse easily into an utter disregard of reality, and if put to actual use, can result in a host of unintended negative consequences such as those just outlined. The former, however, tends to stay closer to reality. If combined with forward-looking moral values, the logic evinced in the history of practice can serve as a better guide for practical reforms. This too is part of what I mean by “practical moralism.”

The Third Realm and Centralized Minimalism

This book presents another example of an analytical concept derived from history-of-practice methodology, to bring out another thread of tradition that has present-day implications. In chapter 5 of my Civil Justice in China: Representation and Practice in the Qing (1996), I highlighted “the third realm” in the Qing system for dealing with disputes. What community and kin did when faced with disputes was mediation, while what magistrates did in formal court sessions was adjudication; between the two there existed a vast third realm. Once a disputant filed a lawsuit, it would trigger an interactive process between the societal system and the official system. On one side were the renewed efforts at community and kin mediation, and on the other, magisterial comments on the plaints, counterplaints, and petitions of the litigants. Those comments were generally either publicly displayed or conveyed to the litigants, or the litigants saw them through other means and channels. These comments would often directly influence the ongoing societal mediation.

40

Huang

If the litigants should reach agreement at that point, then they would petition the court to withdraw the case, and the court would almost without exception grant the request. In this way, the dispute would be resolved in this semiformal third sphere through the interaction of the societal and the official systems. In another article from the same perspective, “ ‘Public Sphere’/‘Civil Society’ in China? The Third Realm between State and Society” (Huang 1993a), I pointed out how state and society formed an either/or binary in Jurgen Habermas’s very widely discussed concept of “public sphere,” as well as in the later concept of “civil society.” The history of practice in China was very different, for state and society overlapped and interacted across broad historical time spans. For that reason, I advanced the concept of the “third realm.” The present volume further illustrates that concept with the history of practice of Chinese basic-level governance. Modern Western theorizing about state/society relations, under the influence of the assertion of bourgeois rights against the state in the French Revolution, has developed a deep-seated analytical assumption about the either/or opposition between state and society. That presupposition is strikingly evident in theories ranging from Weber to Habermas, and in fact may be said to be present in virtually all of the theorizing in the field of historical sociology. The “state” refers mainly to the formal bureaucratic system, and “society” to informal endogenous organizations. Rarely is the sphere in between the two considered, resulting in a view of state and society as an either/or binary. But in the history of practice of Chinese governance, what obtained much more often were interactive or joint state-society semiformal operations, outside the scope of direct control by the centralized state apparatus. This kind of semiformal governance was manifested first of all in the xiangbao 乡保 system, positioned at the crucial nexus of state and society—in nineteenthcentury Baodi county, each xiangbao generally oversaw about twenty villages. Different from formal bureaucrats, they were generally quasi-officials nominated by the community and then approved by the county yamen, who performed their duties without salary or paperwork. But different also from simple endogenous societal systems, they were formally approved and appointed by the state, and bore definite responsibilities to the state apparatus. That is why I refer to them as semiformal quasi-officials. Another similar example is the village-head (cunzhang 村长) system initiated near the end of the Qing, which was similar to the xiangbao system. A further example is village education, also initiated in the last decade of the Qing, in which the villages themselves generally furnished the school buildings (often the village temple) and hired the teachers, the government providing only the blueprint for education without appropriating any funds (VanderVen 2005). The origins of China’s modern

The History-of-Practice Approach to Studying Chinese Law

41

mass education system lay precisely in this kind of joint state-community effort. Yet another example is the Quanxuesuo (劝学所 Education Promotion Offices) of the time, whose members were nominated by the localities and then confirmed by the county magistrate, also examples of semiformal governance. Like the xiangbao and the village heads, members of the Quanxuesuo were quasi-officials who performed their functions with little or no paperwork, were positioned outside the yamen’s formal apparatus, and assisted the magistrate in overseeing basic-level education (VanderVen 2005). Still another example is the merchants associations that arose in the late Qing, which similarly were semi-official, semi-popular organizations, formed on the instigation of the state and participation of the local merchants. The operating mode of these quasi-officials was very minimalist in terms of the method of administration, most of them having neither salaries nor paperwork, and with the formal state apparatus intervening only in the event of disputes or change of personnel. For example, the xiangbao generally acted as they liked; unless there were complaints against them from villagers or there was to be a personnel change, the county yamen would not intervene in their operation. For this reason, what we know about them comes mainly from county archival records of disputes involving them or their replacements. The same applies to the village heads and the Quanxuesuo personnel. (Of course, from the 1920s onward, we also have materials from anthropological field investigations.) Even the management of the county yamen itself went by these same methods. What we know about the actual operations of various offices ( fang 房) of the yamen comes mainly from litigation records, such as disputes within a given office over the control of the office or among different offices over their respective powers and revenues (Reed 2000). The magistrate intervened only then, in a manner similar to his intervention in village governance—using dispute resolution as the principal method. I call these administrative methods “centralized minimalism,” distinct both from the formal official bureaucratic system and from informal societal organizations, and possessing their own logic and method of governance. Chapter 3 focuses on this tradition of the history of practice. (See also Huang 2008.) Although my “third realm” conceptualized the space occupied by this sphere, it did not explain the manner and logic in which it actually operated. The subtitle of chapter 3, “Semiformal Governance by Quasi-Officials and Dispute Resolution in China,” is intended to be a preliminary conceptualizing of its history of practice. This kind of administrative method came first from the minimalist governance ideal of Confucianism and was evident in the Qing state’s promise to not raise taxes with the continued expansion of population (shengshi ziding, yong bu jiafu 盛世滋丁、永不加赋). Another aspect of this

42

Huang

is the limit set by the Qing government on the numbers of clerks and runners a yamen could employ. Later, when faced with continuing increase of population and the real requirements of governance, it developed gradually this system of basic-level semiformal governance under the centralized system at the top. The system originated from long-term historical practice, not from any governing ideology, whether the Confucian ideology of minimalist governance or the more sophisticated formulation of “Confucianized Legalism” from Ch’ü T’ung-tsu (1961). It was born of the interaction between ideology and administrative requirement in the history of practice, and not simply of any ideology. Here, if we turn once more from the “what was” of the historical practice to consider the “what ought to be” of systemic social or political reform, what “centralized minimalism” stands for is an approach that might possibly be relevant for the present. The needs of the welfare state of the new era, of course, would render obsolete a good deal of the content of the old minimalist tradition of governance, but the semiformal administrative method and the stateinitiation-societal-participation model of minimalism might yet have a role to play—as for example in the provision of public services and in China’s continued search for a distinctive political modernity.

Community Mediation under Minimalist Governance

Finally, a word about the community mediation system that is closely related to minimalist governance. On the one hand, there was the concept and method basic to Chinese governance that communities should deal with disputes over “minor matters” by themselves, that the state would intervene only when society itself was not able to resolve the problems. On the other hand, there was the development, endogenous to the tightly-knit communities based on a peasant economy, of a system for resolving disputes: respected individuals within the communities would come forth to listen to and consider the points of view of both sides, and then seek to work out, first separately with each and then together with both, compromise resolutions acceptable to both sides. In the process, both the laws of the state (guofa 国法) and what people termed “reason/moral principles” (daoli 道理) would be considered, but the main purpose was to work out compromises for the sake of maintaining good human relations (renqing 人情). Then, on the basis of the voluntary agreement of both sides, a mediated resolution would be reached. It might employ “apologies” (peili daoqian 赔礼道歉), oral promises or written agreements, a shared meal, and so on, to lend the mediated resolution a ritualized kind of confirmation.

The History-of-Practice Approach to Studying Chinese Law

43

This set of concepts and methods was seen by the state as an integral part of its method of governance, and it was also a long-standing custom of crucial importance for the countryside. What is surprising is that despite the past century’s repeated rejections of the Chinese legal tradition, the same concepts and methods have managed to continue, first in the Guomindang period, and then also in the collective era—even though mediators changed from the community’s most respected people to its state-approved cadres, and even though the principle and method changed from the past “compromise/human relations first, law and reason/moral principle second,” to “law-policy first, human relations and reason/moral principle second”—the basic principle and method, that the community would resolve its own disputes, remained. In the Reform period, even though the massive migration of peasant-workers (nongmin gong 农民工) has impacted the entire system in unprecedented ways, still, community mediation has shown persistent and powerful vitality and has received the determined support and approval of the state. Today, despite the powerful ideological current of wholesale Westernization, it remains a critical component of the Chinese legal system, also its most distinctive aspect. This is the main topic of chapter 2, “Community Mediation, Past and Present.” This book will start with village community disputes and mediation to set the larger social context for civil justice. It will then move on to the politicaleconomic context to analyze the two traditions of informal and semiformal governance in the history of practice and explore their present-day relevance. Next the book turns to the history of divorce law practice to bring out the origins, myths, and realities of court mediation, and then to consideration of recent changes in rules of evidence as they have impacted divorce law practice to explain the negative consequences of a blind copying of Western institutional arrangements. It will be shown that while the history-of-practice may be reasonable and positive, it can also be unreasonable and negative. From there the book turns to the consideration of how civil adjudication and court mediation, despite radical changes in their theoretical underpinnings, reveal in practice substantial continuities, as well as a host of innovations both conceptual and methodological. They reveal also the “practical moralism” mode of thinking and its present-day relevance. The final part is a prospective look at the (Chinese conception of the) problem of “modernity” to address the issue of choices and directions for Chinese law in the future. The conclusion provides a quick summary of the major empirical findings of the book. The method of starting with the history of practice advocated here is obviously connected to the tradition in Chinese law of privileging experience and practical use. What that tradition reveals is a mode of thought different from

44

Huang

modern Western formalism. This book suggests that it is a mode of thinking that can serve as the basis for constructing China’s own modern epistemological method and theorizing. It can be used to transcend the either/or binary of experience/theory; it can be the key to joining closely theoretical concept with experience. Once joined to theory, one can rise above the simple descriptive, retrospective, and particularistic character of experience; at the same time, once joined to experience, one can recognize the historicity of theory, and avoid absolutizing and ideologizing tendencies that disregard time and space.5 My intention here is not to reject completely Weber’s type of formalist theory and logic, because the building of new theoretical concepts needs to be done through a dialogue with formalized theory, and the systematizing of new concepts also needs the assistance of formal logic. What I object to is only their absolutizing and universalizing. My attitude toward positivist theory and research methods (and to postmodernist theories as well) is basically the same. Practice obviously is not so clear or simple as formal theory; its strength is that it is closer to, more in accord with, reality. At the same time, the history of practice often embodies (though not necessarily so) coexistence and compromise, or mutual adaptation, between theory and action, and thus it will not be so unitary or one-sided as theory, but rather more tolerant of differences. In the examples given above, the history of practice, in its process of reconciling theory and reality, often embodies a certain kind of practical wisdom. It is the latter that can serve today as a possibly useful resource. In epistemological terms, my point of view may be summed up this way: experience is one thing and theory another, but, joined together, they make up something more than either one or the other. The same applies to practice and theory: from the point of view of what is, practice is one thing; theory, representation, or institution another; but coexisting, interacting, combining, and differentiating in the course of the history of practice, they make up something more than either one or the other. From the point of view of the pursuit of what ought to be, the history of practice and the logic contained therein might be seen as a resource for more realistic and intelligent choices of what paths to follow in the pursuit of given moral ideals.

5 I have explored some of these issues also in Huang Zongzhi [Philip C. C. Huang] 2007b.

The History-of-Practice Approach to Studying Chinese Law

45

References Bodde, Derk and Clarence Morris. 1967. Law in Imperial China, Exemplified by 190 Ch’ing Dynasty Cases. Cambridge, MA: Harvard University Press. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Translated Richard Nice. Cambridge: Cambridge University Press. Ch’ü, T’ung-Tsu. 1961. Law and Society in Traditional China. Paris: Mouton. Grey, Thomas G. 1983–1984. “Langdell’s Orthodoxy.” University of Pittsburgh Law Review 45: 1–53. Huang, Philip C. C. 1996. Civil Justice in China: Representation and Practice in the Qing. Stanford: Stanford University Press. ———. 1993. “ ‘Public Sphere’/ ‘Civil Society’ in China? The Third Realm between State and Society.” Modern China, 19, 2 (April): 216–40. ———. 2001. Code, Custom and Legal Practice in China: The Qing and the Republic Compared. Stanford: Stanford University Press. Huang Zongzhi [Philip C. C. Huang] 黄宗智. 2005a. “认识中国——走向从实践出 发的社会科学” (Studying China: toward social sciences that start from practice), 中国社会科学, no. 1: 85–95. ———. 2005b. “悖论社会与现代传统” (Paradoxical society and modern traditions), 读书, no. 2: 3–14页. ———. 2007. “中国法庭调解的过去和现在” (Court mediation in China, past and present), 清华法学, v. 10:37–66. Hull, N. E. H. 1997. Roscoe Pound and Karl Llewellyn, Searching for an American Jurisprudence. Chicago: University of Chicago Press. Jones, William C. 1987. “Some Questions Regarding the Significance of the General Provisions of Civil Law of the People’s Republic of China.” Harvard International Law J. 28, 2 (Spring): 309–31. Mao Zedong. 1941 [1971]. “Preface to Rural Surveys,” in Mao, 1971: 194–97. ———. 1971. Selected Reading from the Works of Mao Tse-tung. Beijing: Foreign Languages Press. Neighbors, Jennifer Michelle. 2004. “Criminal Intent and Homicide Law in Qing and Republican China.” Ph.D. diss., University of California, Los Angeles. Reed, Bradly W. 2000. Talons and Teeth: County Clerks and Runners in the Qing Dynasty. Stanford: Stanford University Press. Scott, James. 1998. Seeing Like a State. New Haven: Yale University Press. Shiga Shūzō 滋贺秀三. 1981. “清代诉讼制度における民事的法源の概括的检讨” (The origins of civil law in the Qing litigation system), 东洋史研究, v. 40, no. 1. Tamanaha, Brian Z. 1996. “Pragmatism in U.S. Legal Theory: Its Application to Normative Jurisprudence, Sociological Studies, and the Fact-Value Distinction.” American Journal of Jurisprudence 41: 315–56.

46

Huang

VanderVen, Elizabeth. 2005. “State-Village Cooperation: Modern Community Schools and Their Funding, Haicheng County, Fengtian, 1905–1931.” Modern China 31, 2 (April): 204–35. Weber, Max. 1978. Economy and Society: An Outline of Interpretive Sociology, vol. 2. Berkeley: University of California Press. Wiecek, William M. 1998. The Lost World of Classical Legal Thought: Law and Ideology in America, 1886–1937. New York: Oxford University Press. Xue Yunsheng 薛允升. 1970. 读例存疑重刊本, 黄静嘉 ed., 5 volumes. 台北: 中文 研究资料中心.

chapter 2

Women and Property in China, 960–1949, Introduction and Conclusion* Kathryn Bernhardt Introduction Past scholarship has presented a static picture of property inheritance in China, mainly because it has taken as its primary focus men, whose rights in fact changed little over the centuries. When our focus shifts to women, however, a very different and dynamic picture of property rights emerges. Women’s rights to property changed substantially from the Song through the Qing and even more dramatically in the twentieth century under the Republican Civil Code. It is through an examination of those transformations in women’s claims that we can best discern the larger changes taking place in property rights as a whole. This book is thus at once a study of women’s rights to property specifically and a study of property rights in general. It is also a study that would not be complete without treating both the imperial period and the Republican period. Imperial and Republican inheritance laws were based on radically different concepts of property, the full implications of which cannot be truly appreciated when each period is studied separately. When the two are examined in conjunction, however, each serves to illuminate the other: the distinctive characteristics of the property logic of each period become clear only when studied against the property logic of the other. The Issues As is well known, inheritance in imperial China was governed by the principles and practices of household division ( fenjia 分家): equal division among sons of the father’s property. Women, it is generally assumed, had no inheritance rights. At most, an unmarried daughter would be provided with a dowry, if the family could afford one, and a widowed mother would be provided with

* The selections here are drawn from Women and Property in China, 960–1949 (Stanford University Press, 1999). Used with permission of Stanford University Press, www.sup.org.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_�04

48

Bernhardt

old-age maintenance, but neither had the right to an independent share of the property. As is also well known, household division was accompanied by the principles and practices of patrilineal succession (chengtiao 承祧): a man had to be succeeded by a son for ritual as well as for property purposes. If he did not have a son of his own, he had to adopt one to carry on his line and to continue the ancestral sacrifices. Patrilineal succession, it is generally assumed, reinforced household division. They were but two aspects of the same phenomenon— inheritance by sons only. This understanding of household division and patrilineal succession has given rise to a static picture of the inheritance regime of late imperial China, and understandably so. So long as attention is fastened on fathers and sons, one would indeed be hard put to find evidence of significant change from the Song through the Qing, for their rights in fact changed little. But as this book will demonstrate, the conventional picture leaves out large parts of the story and distorts important parts of the remainder. And it does so because it fails to consider property from the point of view of women in their different capacities as daughters, wives, and concubines. Seen in that light, household division and patrilineal succession are revealed as separate processes with different implications for property inheritance. The principles and practices of household division came into play when a man had birth sons, and those of patrilineal succession when he did not. Moreover, the rules of succession changed in important ways in the Ming and Qing. Of the two, although household division was the much more common form of inheritance in imperial China, patrilineal succession was by no means insignificant. Something on the order of one family out of every five did not have sons who survived to adulthood.1 Thus, inheritance in as many as a fifth of families in imperial times took the form not of household division, but of patrilineal succession. For a woman, patrilineal succession mattered even more because of her membership in two separate families, her natal and her marital, during the 1 Only an adult son could become his father’s full patrilineal heir, meaning that if a son died before reaching adulthood (20 sui), he could no longer be considered his father’s patrilineal heir and another would be needed to take his place for the father’s line to continue. As Ted Tedford found in his study of 41 lineage groups in Tongcheng county, Anhui, from 1520 to 1661, 17 percent of married men had no sons who survived to adulthood (1995: 62, 79). Liu Ts’ui-jung reports a similar rate, 17–24 percent, among five lineages in central and South China from the fourteenth through the nineteenth centuries (1995: 105, 107). All together, of the 23,029 married men in their two studies, 19 percent (4,348) did not have birth heirs.

Women and Property in China, 960–1949

49

course of her life. As a daughter, a woman stood about a 6 to 12 percent chance of being born into a family with no surviving sons.2 And as a wife, she stood roughly a 20 percent chance of being married to a man who had no birth heirs. Thus as many as one woman in three was either a daughter without brothers or a wife without sons (or both), and likely to be involved in patrilineal succession sometime in her life. Equally important, litigation over inheritance in imperial times was overwhelmingly over patrilineal succession, not household division. The reasons for the discrepancy will be examined later. Suffice it to note here that in the collection of 430 Song to Qing inheritance cases on which this study draws, lawsuits over the adoption of an heir for a sonless man outnumbered those over household division by four to one. That disproportionately high incidence of succession suits is also reflected in the relative attention accorded to each process in the Qing code: household division is covered in just four brief statutes and substatutes totaling a little over 200 characters, compared with 11 laws totaling some 1, 100 characters on succession. Patrilineal succession was thus constituted as a legal problem in a way that household division was not. To anticipate our story, the entry point for the analysis of the late imperial period centers on situations in which the inheritance claims of women have to be considered in their own right, because of the absence of brothers in the case of daughters and the absence of a husband and sons in the case of sonless widows. It is these women, as daughters and wives in the absence of men, who bring out in sharpest relief the different implications of patrilineal succession. Seen from their points of view, property rights turn out to have been very far from static in imperial China, as conventional wisdom would have it. The first big change came in the early Ming, with the adoption of the legal requirement that all sonless families establish a lineage nephew to be the patrilineal heir to the father, or, in the terms of this book, the adoption of “mandatory nephew succession.” That development cost women dearly, seriously diminishing the property claims of both daughters and widows in the ensuing years.

2 Available fertility and mortality data suggest that married men in the Ming and Qing periods had on average three to four children who survived to adulthood (J. Lee et al. 1995: 173–80; Liu Ts’ui-jung 1995: 99–100; Telford 1995: 67). By genetic change alone (and assuming for simplicity’s sake a sex ratio of 100 and not 105), we would expect 12.5 percent of those with three children and 6.25 percent of those with four to have all daughters.

50

Bernhardt

For a daughter, the adoption of mandatory nephew succession in the early Ming meant a virtual loss of any right to inherit in the absence of brothers.3 Simply put, whereas in the Song a daughter was legally entitled to inherit the family property should her parents die without any sons, whether biological or adopted, under the rule of nephew succession, the claims of nephews took precedence over her claims. A daughter’s likelihood of inheriting property by default was very remote. For a widow, the new rule meant, if not a total loss, at least a severe contraction in her inheritance rights. Where once she stood to inherit all of her husband’s property in the absence of sons, she now had merely custodial powers over it, responsible for preserving it for her husband’s heir, one that she herself was now legally obligated to adopt. Moreover, initially under the rules of mandatory nephew succession, she had no choice but to adopt the lineage nephew most closely related to her husband. In time, however, in a change driven in great measure by the growing power of the chaste widow ideal, that requirement was dropped. As is well known, the Ming and Qing saw the rise and solidification of the cult of female chastity. For a widow, the insistence that she not remarry after her husband’s death turned out, somewhat surprisingly, to be empowering, at least as far as inheritance was concerned. In legal practice, Ming and Qing judges, acting out of the conviction that a chaste widow deserved the heir of her choice, consistently allowed the widow to reject her husband’s closest nephew. Then, in the mid-Qing, the state adopted formal legislation granting her the right to choose freely from among all of the lineage nephews. That expansion in the range of a widowed wife’s custodial powers within the nephew succession regime was the second big change to come in women’s property and inheritance rights in the late imperial period. The custodial rights of widowed concubines over property also expanded as a result of the growth of the chaste widow ideal. For a concubine, as we shall 3 My use of the term “rights” to describe women’s property claims in the imperial period is based on the analysis of magisterial adjudication in Philip Huang’s 1996 book, Civil Justice in China: Representation and Practice in the Qing. There he demonstrates that, though the Qing state did not have an abstract conception of rights in the Western sense of absolute rights protected by law and independent of the will of the ruler, the Qing code nevertheless contained numerous stipulations that local magistrates consistently used to uphold legitimate property and contractual claims from infringement by others. Conversely, as he also demonstrates, litigants sought recourse to the courts to safeguard their property. The practical consequence of the legal system was therefore the protection of legitimate claims, and to that extent, one can speak of the existence of “rights” (see especially Huang’s chap. 4). For the imperial period, I use “rights” in this sense of rights in practice.

Women and Property in China, 960–1949

51

see, widow chastity proved to be the great equalizer, erasing the status distinction that had previously prevented her from sharing any of the wife’s claims on her husband’s property. By the Qing, a woman’s status as a wife or a concubine came to matter less than whether she was a chaste widow. A concubine came to be entitled to the same rights accorded to any chaste widow, including custodial powers over her deceased husband’s property and the right to adopt the heir of her choice. When we turn our attention to the transition from imperial to Republican law, we find patrilineal succession as a crucial point of change between the old and the new. In the early Republican period, although the Qing laws on mandatory nephew succession remained in force on paper, the interim Supreme Court, or Daliyuan (大理院), chose to interpret them in such a way as to grant widowed wives completely autonomy in the selection of an heir, even if she chose to go outside of her husband’s lineage nephews. In so doing, it effectively overturned the basic principle of nephew succession. That was an important change, introduced within the conceptual frame of the old system. The old system and its concepts were finally overturned by the Republican Civil Code of 1929–1930. Adopting a single new inheritance regime based on the Western concept of individual property, the code removed patrilineal succession from any relevance to inheritance. It did not mandate the appointment of a male heir for a deceased sonless man, nor did it recognize the property claims of patrilineal kin. At the same time, in keeping with its emphasis on gender equality, the code granted women the same inheritance rights as men in principle. The full implications of those changes in the laws can best be understood through an examination of legal practice as revealed in court case records. The new code was, after all, superimposed on a society long accustomed to operating by household division and patrilineal succession. And it was in the courtroom where the new legal principles came into direct conflict with ages-old established social practices. The result was a complex picture, neither one of simple radical change suggested by the letter of the laws nor one of simple continuity suggested by the powerful persistence of old practices. Rather, contestations between the new and the old manifested themselves at specific points of tension, with varied implications for women in their different capacities. For all the lawmakers’ good intentions, women lost old powers even as they gained new ones. Source Materials To understand the changes in property rights over time, we must look beyond the codes themselves and study the law in action through court case records.

52

Bernhardt

For the imperial period, the successive dynastic codes, by themselves, offer little evidence of change. At most, they reveal subtle revisions of wording and the addition of new substatutes whose implications are not readily apparent. For the Republican period, attention on the civil code alone can all too easily lead to an exaggerated picture of change and a neglect of the practical effects of some Western-derived principles. For court cases from the imperial period, the book draws on three different types of materials. The first is the original archival records for 68 inheritance-related cases of the Qing period. Those records contain all manner of documentation generated during litigation, notably plaints, counterplaints, magistrate instructions, and court judgments, and involve five different jurisdictions: Qufu county, Shandong, from the 1710’s to the 1890’s; Baxian county, Sichuan, from the 1760’s to the 1850’s; Baodi county, Shuntian prefecture, from the 1830’s to the 1900’s; Danshui subprefecture and Xinzhu county, Taiwan, from the 1840’s to the 1890’s; and Taihu subprefecture, Jiangsu, in the 1870’s. In addition, I have drawn on published collections of original court decisions. Some of these works incorporate the judgments of numerous officials (the best-known of this sort being the Collection of Lucid Decisions by Celebrated Judges [Minggong shupan qingmingji 名公书判清明集] of the Song period). Others present a single official’s judicial rulings. These latter collections, usually put out by the officials themselves right after the expiration of a term in office, consist at most of verbatim reproductions of instructions (pici 批词) and judgments (tangduan 堂断), and provide only partial documentation of court cases. Without the litigants’ plaints for the essential background, it is often difficult to get a complete picture of any particular case. But for my purposes, that disadvantage is more than balanced by the fact that the authors, with eventual publication in mind and out of a desire to showcase their own legal acumen and moral wisdom, tended to write longer and more detailed rulings than was normally the case. They also tended to offer lengthy explanations for their decisions and extended commentary on laws, again something normally not found in the archival documents. Finally, I have relied on narrative accounts of lawsuits set out in the diaries and autobiographies of local officials. Composed in a storytelling fashion, a typical account begins with the nature and cause of the dispute, continues to the official’s interrogation of the litigants, and ends with his resolution of the suit. Interspersed throughout the narrative are the author’s personal reflections on the case and the applicable laws. Like the published judicial rulings, these accounts are invariably one-sided and self-glorifying, but they too permit us to see how officials themselves read and understood the law.

Women and Property in China, 960–1949

53

For the Republican period, the book relies principally on the original archival court records for 370 inheritance cases. Of that number, 96 are appeals cases heard by the Daliyuan, China’s highest court, in the 1910’s and 1920’s, and 134 are appeals cases heard by the Capital Superior Court ( Jingshi gaodeng shenpanting 京师高等审判厅) located in Beijing, also in the 1910’s and 1920’s. The remaining 140 sets of records cover cases originating at the Capital Distict Court ( Jingshi difang shenpanting 京师地方审判厅) and its successor, the Beijing District Court (Beijing difang fayuan 北京地方法院), from the 1910’s to the 1940’s. In roughly half of those cases, the litigants appealed the district court’s decision to higher courts, with the result that for some of the more hotly contested disputes the various court judgments alone ran to more than 200 pages. The Song Baseline A word, finally, about the Song, with which this book opens. Past scholarship, mainly Japanese, has for a variety of fortuitous reasons come to see the Song as an exception to imperial China, as a period when a daughter enjoyed independent inheritance rights to property under a half-share law that supposedly entitled her to half of what a son got at the time of household division. Even though scholars have accepted this “law” as fact, it seems to me that none has provided a satisfactory explanation for why the Song should have stood apart from the rest of imperial history. This book begins with a reexamination of the extant evidence and arguments about the Song in order to construct a solid baseline from which to assess the later changes. Readers should be forewarned that the discussion will, of necessity, be a dense one, given the weight of past scholarship and the need for a close scrutiny of all of the available evidence. That chapter concludes that there was no “half-share law” in the Song and indeed could not have been. Instead, the principles of patrilineal succession applied, and women enjoyed inheritance rights only by default, in the absence of brothers and sons. What set the Song apart from the Ming and Qing was that there was as yet no mandatory nephew succession, with all that that implied for women’s inheritance rights. Conclusion A focus on women in their various roles shows that household division and patrilineal succession were two separate processes and conceptual complexes

54

Bernhardt

with different implications for property inheritance. The one governed inheritance when a man had birth sons, and the other when he did not. For women, it made all the difference whether they were women in the presence of men or women in the absence of men. Women’s rights in household division did not change in the imperial period; from the Song on, they possessed only a claim to dowry and to maintenance. But their rights in patrilineal succession changed substantially. The imperial period, as we have seen, was characterized by three distinctly different regimes of patrilineal succession. First, in the Song, women could still inherit by default in the absence of men. Patrilineal succession had not yet become a universal legal requirement. In the early Ming, women’s rights underwent a sharp contraction with the adoption of mandatory nephew succession. A daughter could no longer inherit in default of brothers, but had to defer to all of her paternal cousins out to fourth cousins. Similarly, a widow no longer had the right to inherit in default of sons, but was merely to serve as the custodian of the property, holding it intact for the required heir (the lineage nephew most closely related to the deceased), whom she herself was now legally obligated to adopt. In the mid-Qing, finally, a widow’s custodial powers expanded greatly with new legislation that permitted her free choice among lineage nephews. In that, the law came to recognize what had been longstanding practice in the late Ming and early Qing: judges had been rewarding widows for their chastity by giving them greater latitude in the selection of an heir. The group most heavily impacted by these changes in women’s rights was the father-husband’s agnatic male kin. From no rights of inheritance in the face of a surviving widow or daughter, they came to take precedence in the rigidly fixed system of mandatory nephew succession in the early Ming. Although they continued to retain their rights in the Qing, they lost ground as the code was amended to give widows the exclusive say in which nephew was to inherit. Past scholarship has not grasped these changes because it did not separate out patrilineal succession from household division, and it did not separate the two because it considered inheritance principally from the perspective of men. In that light, the two processes merely reinforced each other as two sides of the same coin of inheritance by sons. That is the basic view of both Niida Noboru and Shiga Shūzō, arguably the two giants in the field. Both assume a complete congruency between household division and patrilineal succession, and both, as a consequence, assume an essential continuity in inheritance throughout the imperial period. It is only when patrilineal succession is separated out from household division and analyzed on its own terms that we can fully grasp the patterns of change in inheritance in imperial China.

Women and Property in China, 960–1949

55

For the Republican period as well, it is essential to see the two as separate systems with different conceptual underpinnings and different consequences. To be sure, changes in one affected the other, but inheritance as a whole could not be completely revamped without overturning the separate logics and processes of both. The early Republican period was a time of transition. The Qing code and its provisions on inheritance remained in force, adopted by the early Republican authorities as the law of the land. The Daliyuan therefore operated within the legal frameworks of household division and patrilineal succession. Yet, at the same time, it interpreted the old laws to give widows complete autonomy in the choice of a successor. If by the mid-Qing, a widow no longer had to follow the lineage order in her selection of a nephew as heir, she could now bypass a nephew altogether. The high court’s rulings effectively put an end to mandatory nephew succession. That was the most important change in inheritance in the Daliyuan period, and it came wholly within the laws on patrilineal succession. The Guomindang lawmakers were determined to overturn the very logic of the old inheritance regime, not just to reinterpret it. They focused their energies on patrilineal succession, in their view the source of the “feudal” ideas and practices that disenfranchised women. If they dismantled patrilineal succession and replaced it with individual property (as opposed to family property) and gender equality (as opposed to inheritance by sons only), they assumed, they would deal a death blow to the old inheritance regime and women would thereby gain the same rights as men. What actually happened ran counter to their expectations in several ways. First, by failing to target household division as a separate process, they unwittingly allowed it to continue. Their assumption was that granting women equal inheritance rights would spell the end of sons-only household division. But, in fact, their Western-derived inheritance theories took effect only upon the death of the property owner, with women inheriting equally only postmortem. That in effect gave legal sanction for old household division practices to continue under the rubric of gift-giving during the property owner’s lifetime. A father could disinherit his daughters simply by parceling out his property as gifts before his death. As a result, daughters did not gain the inheritance rights the lawmakers had intended for them. Second, the lawmakers took away the custodial powers that a widow had enjoyed under earlier law. Once her husband died, his estate passed in shares to his heirs as separate individuals, regardless of her wishes. She no longer had the right to adopt an heir as a way to secure her control over his property. To be sure, the lawmakers granted a widowed wife a set portion of her

56

Bernhardt

husband’s property, but that gain in inheritance came at the cost of her custodial powers over his entire estate. Their elimination of patrilineal succession had an even greater impact on widowed daughters-in-law and widowed concubines, for the loss of custodial rights was not balanced by any gain in inheritance rights to their husbands’ property. These issues did not occur to the lawmakers because they did not consider inheritance from the point of view of women in their different capacities. The “modern” law of the Guomindang therefore had mixed consequences for women. There was no simple transition from a regime of no property rights for women to a regime of full property rights for women, as the lawmakers intended. Instead, the practices of household division persisted, albeit in a different legal guise. And the custodial powers enjoyed by women under patrilineal succession vanished completely, only to be partially offset by the acquisition of inheritance rights. In the end, women lost even as they gained under the Republican Civil Code. It might be well to reflect briefly, by way of closing, on the implications of this book for women’s history. When I began this study, I was not at all certain whether it would merely tell an untold part of the story of inheritance or whether it would have broader implications for our understanding of late imperial and Republican inheritance in general. Now at the end of the project, I can say that the focus on women led me to an entirely different understanding not only of women’s inheritance, but of the very logics and consequences of the two conceptual complexes governing inheritance. The women’s story, then, is not just about women, but about rethinking the subject of inheritance as a whole. References Harrell, Stevan, ed. 1995. Chinese Historical Microdemography. Berkeley: University of California Press. Huang, Philip C. C. 1996. Civil Justice in China: Representation and Practice in the Qing. Stanford, Calif: Stanford University Press. Lee, James, Cameron Campbell, and Lawrence Anthony. 1995. “A Century of Mortality in Rural Liaoning, 1774–1873.” In Harrell, pp. 163–82. Liu Ts’ui-jung (Liu Cuirong). 1995. “A Comparison of Lineage Populations in South China, ca. 1300–1900.” In Harrell, pp. 94–120. Telford, Ted. A. 1995. “Fertility and Population Growth in the Lineages of Tongcheng County, 1520–1661.” In Harrell, pp. 48–93.

chapter 3

Illicit Bureaucrats* Bradly W. Reed Preface As a graduate student, I was initially drawn to the county clerks and runners in Qing dynasty China by my discovery of how little anyone actually seemed to know about them. True, they were mentioned in nearly all the scholarly works on local government during the Qing, but only rarely did these citations extend beyond a brief, caricatured description of them an undifferentiated mass of corrupt ne’er-do-wells, single-mindedly bent on the pursuit their own self-interest to the detriment of both state and society. Despite the fact that it was the clerks and runners who carried out most of the administrative work in the county yamen, there was little or no information available as to who these people actually were, where they came from, how they were organized, or how they went about performing their work in the yamen. This book began as an attempt to redress this situation. My intent was to get past the simplified representations of clerks and runners as stock villains in order to better understand their role in county government and the part their actions might have played in the relationship between the Qing state and local communities. What I wanted to do in the process was to traverse the distance between the statutory norms dictated by the central government in Beijing and daily administrative practice at the local level in order to discover how a local yamen actually operated. I subsequently determined that the best way to accomplish this task was not to provide a comprehensive sketch of county yamens throughout the empire, but to render a more detailed account of a single yamen. Before I could begin, however, I first had to confront the problem sources. Previous studies of local administration under the Qing have drawn almost exclusively from three bodies of documentation: county and provincial gazetteers written under the auspices of either the government or local gentry; the collected writings of various officials in the form of handbooks and memoirs; and compendiums of central-level documents including edicts, * This chapter is drawn from “Preface” and “Chapter One” of Bradly W. Reed, Talons and Teeth: County Clerks and Runners in the Qing Dynasty, Stanford University Press, 2000.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_�05

58

reed

memorials, and administrative regulations. But while these sources are adequate for descriptions of formal structure and official policy, they are of considerably less value in revealing the minutiae of quotidian practice in the yamen. To reach that level of detail, it is necessary to delve into county government archives. Until quite recently, however, historians have been faced with an egregious lack of precisely this type of source material. In this respect, it is ironic that the historical record of the Qing government is so abundantly rich at the central level and yet so very nearly barren at the level of county government. Although this disparity has been quite frustrating for historians, the fact remains that very few county government archives from the Qing period have survived in more than mere fragments. Prior to the mid 1980’s, the only county level archive of any size available to Western scholars was that of Danshui and Xinzhu (Danxin) counties in Taiwan. The Danxin archive has provided scholars such as Mark Allee, David Buxbaum, Dai Yanhui, and Philip Huang with a wealth of information on Taiwanese society and the workings of the Qing legal system. But in matters pertaining to the yamen staff, it is far too thin to serve as the basis for an in-depth study. Fortunately, in the last decade Western scholars have gained increasing access to the largest and most comprehensive Qing local government archive known to exist in China, the archives of Ba county (Baxian) in Sichuan Province. Although supplemented with material from other available archival and published sources, this study draws primarily from material gathered from this unique resource. A brief description of the collection and some of the problems associated with its use is therefore in order. Following the fall of the Qing dynasty in 1911, the Ba county archive was largely neglected by Republican era county governments until World War II, when it was moved to a temple outside the city of Ya’an to protect it from Japanese bombing (Esherick and Ye 1996). Forgotten there for nearly two decades, the mildewed and pest-eaten archive was re-discovered 1953 and moved to the University of Sichuan in the provincial capitol of Chengdu. Although scholars at the university had begun preliminary cataloguing and transcription of the archive’s 113,000 files ( juan), in 1965 the entire collection was moved to the Sichuan Provincial Bureau of Archives (Sichuan sheng dang’an guan), where it is presently housed. Preservation, restoration and microfilming of the damaged and fragile documents, however, did not begin in earnest until 1980. Since then, the archive has been made progressively more available to Western and Chinese scholars alike. Archivists at the Sichuan Archives initially cataloged the Ba county collection by reign periods, each of which they then subdivided into six sections:

Illicit Bureaucrats

59

Administration (neizheng); Economy ( jingji); Military (bing); Culture and Education ( jiaoyu); Law (sifa); and Foreign Relations (waijiao). Files relating to the yamen staff are located chiefly within the administrative section of each reign period. These files provide information on a wide variety of issues pertaining to the yamen’s clerks and runners: recruitment and appointment procedures, employment records, personnel reports to superior yamens, dossiers on individual clerks and runners, disciplinary proceedings, intra-yamen regulations and procedures, directives from superior yamens, grievances against yamen personnel lodged with the magistrate by local residents, and, above all, case records of disputes among yamen personnel. Although the archive contains administrative files from nearly all of the Qing dynasty’s ten reign periods, the survival of documents from periods prior to the nineteenth century is far too fragmentary to provide the information necessary for a detailed description of administrative practice. The archival survival rate picks up considerably in the nineteenth century, however, with by far the richest section being that of the Guangxu reign (1875–1908). This period alone accounts for nearly half of the archive’s 4,500 administrative files (Wang 1988, 10). My initial research was thus based on a sampling of over five hundred files culled from the Guangxu collection over an eighteen-month sojourn in the archive. The count of five hundred is somewhat misleading, however, in that not all files are discrete or self-contained. A single dispute, for example, might continue over several individually catalogued files. Conversely, many files contain multiple cases and records; often as many as fifty to eighty discrete items are contained in a single file. Thus, while the sample was not exhaustive, it nevertheless yielded sufficient material to provide a descriptive account of how a county yamen was organized and operated on a daily basis. In order to expand this account beyond the confines of the Guangxu period, documents from earlier reign periods were then gathered during a return visit to the archive. Although these files are far fewer and less complete than those of the Guangxu reign, they nevertheless added depth to my initial description without substantially altering it. Thus, while earlier documents reveal differences in areas such as the number of personnel employed or details in the organization of those personnel, the basic patterns of informal and practice remained constant. But while these documents provide an unprecedented wealth of detail, their use is not without the same sort of difficulties and pitfalls confronting any historian who utilizes local archives. One of these difficulties stems from the highly localized nature of the customary and informal practices which this study attempts to apprehend. Information as to yamen structure and

60

reed

procedures, for example, is very rarely provided in a comprehensive manner by one or even several documents. Aside from those few instances where an individual clerk or runner had reason to explain a particular practice to an incoming magistrate, this information usually comes in the form of piecemeal references scattered across a large number of files, often in highly localized idiomatic expressions. Moreover, due to the informal nature of many yamen procedures and the fact that such practices were often in violation of Qing law, statutory descriptions of administrative procedure are of little value as a point of reference. What we are dealing with here is very much a form of local knowledge that was never intended for yamen outsiders. As a result, the historian often finds her or himself in the position of an archeologist working with bits and pieces of scattered evidence in an attempt to reconstruct a reasonably comprehensible picture of how thing worked. In the process, however, one inevitably encounters gaps in the documentary record which render at least portions of this picture in frustrating degrees of shadow. A second difficulty in using archival records as an historical source lies in the discursive nature of many documents and the problem of interpreting subjective portrayals of events and individuals. This problem becomes most acute in records of intra-yamen disputes and conflicts between yamen staff members and local residents. As in all conflicts brought before a higher authority for resolution, it must be remembered that the testimony given in these disputes is above all a form of representation. What a disputant states is a subjectively constructed version of events fashioned for a particular audience, the judge, in order to win a favorable decision. In a Chinese magistrate’s court, the subjective element of testimony was all the more salient because it was frequently not a particular act itself that mattered so much as the magistrate’s perception of the actor’s moral character and intent, as well as the circumstantial context of the act. In some cases, certain objective claims made by an individual can be authenticated through reference to other sources: whether or not a particular person was employed in the yamen at a particular time, or whether an inter-divisional agreement on allowable procedures had been produced, for example. In a similar manner, the likelihood that an accused clerk or runner was guilty of charges brought against him may be increased, though by no means proven, by the discovery of multiple instances of similar accusations in the past. But even here, we must avoid assuming that any given representation of events was substantively true. Just because a magistrate decided in favor of one account over another, moreover, does not in itself establish its veracity. Like all individuals, magistrates were fallible, subject to their own prejudices and interests

Illicit Bureaucrats

61

and, thus, to manipulation as well as simple error. Nor should we privilege our own judgement by assuming that we, removed from events by more than a century, are in any better position to judge the truth than was the presiding magistrate. As tempting as it may be to play amateur sleuth in the archives, the fact remains that the “truth content” of many individual documents, particularly in regard to representations of motive and character, remains beyond our grasp. But whether an account is true or false is ultimately of less importance to us than are the accounts themselves as objects of inquiry. The substantive object of a dispute, the language used by the disputants, and the venue and manner in which a dispute was presented for resolution can all provide useful information. For example, that disputes between runners over the right to collect fees from a specific type of legal case were frequently presented to the magistrate for resolution indicates that fee collection had achieved a degree of customary legitimacy despite its formal illegality. Similarly, that a particular claim or accusation was made in the first place implies that, quite apart from its veracity, it fell within the realm of what the speaker anticipated would be accepted by the magistrate as at least plausible. Regardless of whether or not an individual or group of yamen employees were considered by others as honorable men, the fact that such claims to honor and rectitude were made at all is therefore significant in and of itself. Equally significant are the values, either declared or implied, upon which such claims were based and the discursive elements used to express those values. Subjective representations of events and persons were not simply fabricated out of a vacuum. Rather, to be effective they had to conform to culturally conditioned norms and expectations as well as with the dictates of personal interest. By focusing on this aspect of disputes and the representations thereof, we are in position to understand the interplay between culture, ideology, and personal interest upon which the social world of the county yamen was based. It might be argued that the portrait of yamen practice I create in these pages is overly self-contained and without sufficient reference to broader national contexts and historical forces. In mapping out the intricacies of the Ba county yamen, I indeed give scant attention to momentous events such as the intrusion of Western imperialism, the rebellions of the mid-nineteenth century, the Tongzhi Restoration, the New Policy Reforms, or the Rights Recovery Movement, all of which had a decisive impact on the Qing government at both the provincial and central levels. This is not a product of oversight, however, but is due instead to the simple fact that these events are rarely reflected in archival material at the county level. Despite the influence these events had in shaping the contours of Chinese history and the fate of imperial

62

reed

government in the nineteenth and early twentieth centuries, from the standpoint of clerks and runners in the county yamen, they were largely irrelevant to the way things worked. If this gives the impression that local government in many ways operated in isolation from national events, the machinations of the central court, and officially sponsored efforts at political reform, it is not an inaccurate representation.

The Issues

Before turning to the organization, practices, and lives of Ba Country clerks and runners, in this chapter I set the stage by placing the study in a theoretical and analytic context. I begin by considering the problematic position of yamen clerks and runners within the Qing administrative system in general. I then discuss several of the key issues raised in subsequent chapters, the ways in which past scholarship has dealt with these issues, and how fresh approaches based on recently available sources can provide greater insight and understanding of how local government operated in the Qing dynasty. The imperial government of the Qing dynasty (1644–1911) could not have functioned without the clerks and runner of the county yamen. Working under the direct supervision of the county magistrate on the lowest rung of the Qing state’s vast administrative apparatus, clerks and runners performed such rudimentary yet critical tasks as the drafting, copying, and filing of documents and records, the recording of cadastral surveys, the registration of deeds and land sales, the acceptance and transmittal of petitions to the magistrate, the posting of official notices and proclamations, the issuance of licenses and permits, the delivery of communications to and from superior yamens, as well as all the sundry duties associated with the magistrate’s court. In addition to these administrative duties, clerks and runners also served as one of the primary mechanisms for the enforcement of state authority in areas such as arrest and detention, public security, and, of course, the collection of taxes. In this capacity, yamen employees functioned, as the oft quoted phrase put it, as the “talons and teeth” of the local magistrate (weiguan zhi zhaoya). Despite the pivotal role they played in the administration of the empire, however, Qing officials regarded yamen clerks and runners with open contempt. In striking contrast to the moral character and integrity attributed to officials from the magistrate upward, local clerks and runners were nearly universally reviled as the most cunning and venal of scoundrels. “All the present suffering under Heaven is caused by clerks,” as one nineteenth-century commentator was moved to write. “Their arrogance is greater than that of any

Illicit Bureaucrats

63

official while gentry and scholars alike acquiesce to their every desire. The common people upon whom they feed thus have no redress and can do nothing but die in lonely ditches by the roadside” (HCJSWB 24: 8a). Antipathy toward yamen clerks and runners among imperial officials was certainly not a new phenomenon in the Qing. As early as the Song dynasty (960–1279), officials regularly expressed a hostility that was itself deeply rooted in the ideological and social transformations of that era. On the one hand, the steady consolidation and centralization of imperial authority in the post-Tang empire created a need for more administrative personnel at the county level than ever before. Yet at the same time, office holding itself was becoming increasingly linked to success in the civil service examinations and association with the elite literati culture which those examinations helped engender. Even in the Northern Song (960–1125), neither clerks nor runners were considered to be officials (guan) insofar as that term generally referred to individuals appointed to office by the central government. Clerks and runners were instead recruited locally to perform specific duties in the county yamen and were thus not required to hold a degree before assuming their positions. As a result, an occupational as well as social fissure developed between officials and members of the yamen staff, a fissure that continued to grow over the remainder of the late imperial period (Lo 1987, 24). The appearance of this fissure was closely tied to the emerging NeoConfucian orthodoxy in the Song upon which the late imperial social and political order rested. According to this view, political authority was the prerogative of the “superior man” ( junzi) whose personal character and moral qualifications for office holding had been recognized by state fiat in the form of an official degree. Clerks and runners, on the other hand, were specialists, administrative laborers whose value lay in their narrowly defined technical expertise and skill. Indeed, it was precisely this utilitarian function and narrowness of competence that disqualified yamen employees from consideration for higher office (Ocko 1983, 133–135, 176). Without the benefit of broad literary training in the Confucian canon and the inculcation of normative social values and political ethics implied thereby, members of the yamen staff were regarded as manifestly unfit for the responsibilities of authority. As the purview of the “small man” (xiaoren), yamen service thus came to be associated with menial, socially degrading forms of labor undertaken only by the destitute or those who desired to use the proximity to authority to further their wicked and corrupt designs. Attitudes such as these, rooted in ideology, training, and social distance, were further compounded by practical problems of administrative dependence and control. For although clerks and runners were regularly vilified as corrupt

64

reed

and self-serving, commentators throughout the late imperial era recognized that they were nevertheless essential for administrative operations due to their technical expertise and familiarity with local conditions, customs, and people. With the social and institutional lines between the yamen staff and ranking officials firmly established, the question then became how to maintain control over these “yamen vermin” ( yadu) and minimize the leverage they possessed by virtue of their functional position within the administrative structure. The first line of defense in this regard was simply to restrict the number of yamen employees working in any given office or yamen. To this end, the central government established strict quotas and limits on terms of service in the closing decades of the Northern Song. Since clerks and runners were not centrally appointed and thus not subject to the control system of review, sanctions, and rewards imposed on appointed officials, responsibility for the actual supervision and disciplining of yamen employees was placed directly upon the magistrate, who was held personally accountable for any instances of wrongdoing on the part of his staff. Reliance on both quotas and the personal authority of the magistrate would remain hallmarks of the statutory control system for the rest of the imperial era. The problem with this system was that it was unworkable from the very beginning. The Wang Anshi reform policies of the eleventh century, though failing in many respects, succeeded in vastly expanding the workload of county government by drawing a much larger number of tasks within the ambit of state administration. Thus, while the enforcement of quotas was becoming something of an obsession at the Song court, the actual requirements of county government were forcing magistrates to employ workforces far in excess of these legally stipulated quotas (Liu 1967, 331). The disparity between statutory workforce and required labor was only exacerbated in subsequent centuries by demographic growth, the greater socio-economic complexity of the empire, the declining ratio of local administrative units to population, and the continued absorption by county yamens of many administrative duties previously performed by local communities (Kracke 1953, 47; Liu 1967, 318; Skinner 1977, 24). As a result, not only was the magistrate’s already dubious ability to exercise effective control over his staff made still more tenuous, but substantive control over many administrative operations fell increasingly to individuals whose very presence in the yamen was a violation of the statutory system. If this state of affairs caused grave concern among officials at the central court, their misgivings were only increased by the manner in which yamen clerks and runners supported themselves. Provided with little or no compensation for their services by the state, yamen personnel from the Southern Song

Illicit Bureaucrats

65

onwards depended instead on the ubiquitous customary fees (lougui, guifei) collected from anyone having business with the yamen, whether voluntary or otherwise. It was this officially proscribed practice which was above all responsible for the reputation of clerks and runners for avaricious self-interest and corruption. As nineteenth-century prefect He Gengsheng said of yamen employees in general, “They only care for profit and know nothing of gratitude or honor. If an official shows them leniency, therefore, it will only give rein to their desires. But if one treats them with severity, they will be filled with dread and awe and cease their nefarious cunning” (MLS 4: 30). He Gengsheng’s admonition that state officials adopt the most draconian of measures to protect the emperor’s subjects from the rapaciousness of the state’s own functionaries was by no means an idiosyncratic point of view; similar opinions were regularly expressed by officials at all levels of government in the Qing dynasty. Although acknowledged as administratively necessary, clerks and runners nevertheless remained presumptively corrupt as individuals and beyond the pale of official status both institutionally and ideologically. Over the course of the eighteenth and nineteenth centuries, however, the empire was becoming an altogether different place than it had been when this administrative system was devised. Demographic growth and diversification, commercial expansion, and the progressive monetization of both the economy and tax system were all combining to increase the administrative workload of the county yamen to the point where statutory provisions for manpower and financing were made all but irrelevant. In the face of such pervasive change coupled with a static statutory system, county government could only remain operational by the continued growth of informal methods and personnel that legally should not have existed. In the process, there developed a system of local administration about which the central government had only the vaguest knowledge and over which it exercised even less control. Considering the centrality of clerks and runners to local administration under the Qing, we know remarkably little about them. Although they are mentioned briefly in nearly every study of county government, the tendency has been to accept the portrayals of them given by official and elite sources as self-interested and corrupt individuals. But in sharing the biases and pejorative assumptions of these sources, we have also tended to share the Qing central government’s ignorance of the informal practices that sustained county government during the Qing. This study attempts to remedy this situation through an examination of the organization and daily practices of yamen clerks and runners in a single locale, Ba county, Sichuan, in the nineteenth century. My use of the term “system” to describe these informal methods of administration is deliberate. For although they operated outside of the statutory

66

reed

system, and often in direct violation of the legal code, yamen clerks and runners exhibited a remarkably consistent degree of organization and rationalization in the form of internally formulated and enforced rules and procedures. As this implies, my approach here is to examine these practices not simply in terms of their deviation from the statutory norm, but as a fully functioning system of local administration which becomes visible only by going beyond formal structures and representations. In other words, my concern here is not with how government was supposed to work, but rather with how local administration actually operated. Along with its empirical value, moreover, a detailed examination of the organization and practices of yamen clerks and runners opens the door to a reconsideration of the conceptual frameworks traditionally used to analyze the relationship between the late imperial state and Chinese society.

Past Scholarship

Administration and Bureaucracy Until recently, the general trend of Qing administrative studies was to focus on statutory structure rather than practice, with particular emphasis on institutional methods of imperial control over officials. Rarely, if ever, did these studies venture below the level of the magistrate and his official responsibilities to consider the less formal aspects of local administration or the broader social context within which local government operated (Mayers 1878; Brunnert and Hagelstrom 1912; Hsieh 1925; Metzger 1973; Wu 1970). The outstanding exception to this pattern was the work of Hsiao Kung-ch’uan, whose Rural China: Imperial Control in the Nineteenth Century, first published in 1960, provided a detailed study of the articulation between government and local society during the Qing dynasty. As his title implies, however, Hsiao’s concern was with mechanisms of state control at the village level. For the most part, he therefore left the county yamen unexamined. It was not until the subsequent publication of Ch’ü T’ung-tsu’s Local Government Under the Ch’ing, in 1962, that we got our first good look inside the local yamen. Although Ch’ü’s pioneering work remains the indispensable starting point for research in the field, his study is limited by several of the same features that made it such a breakthrough at the time of its publication. Culling from the broadest possible array of central government documents, local gazetteers, administrative handbooks, and the personal memoirs of magistrates and provincial officials, Ch’ü attempted to provide a comprehensive account of local administration throughout the Qing empire. In this attempt

Illicit Bureaucrats

67

he succeeded in providing us with a veritable catalogue of structures, statutes, practices, and anecdotal detail. The shortcoming here, however, is that Ch’ü’s description is on such a broad canvas as to fit everywhere in general, and nowhere in particular. A second and more serious limitation of Ch’ü’s work is that despite its richness of detail, his analytical focus remains on what he saw as the dysfunctional aspects and structural irrationalities of the Qing administrative system. Although he devotes a full chapter each to clerks and runners, his concern with them is primarily on how these administrative inadequacies resulted in corruption within their ranks and the impact which that corruption had on the magistrate’s effectiveness as an administrator. One result of this is that Ch’ü fails to capture either how clerks and runners were organized in the face of inadequate statutory provisions or how the resulting informal methods of operation hung together to provide a functioning system of local administration. A number of outstanding monographs published since Hsiao and Ch’ü have added enormously to our understanding of local government under the Qing (see, for example, Watt 1972, 1977; Ocko 1983; Huang 1985; Rankin 1986; Schoppa 1989; Min 1989; Zelin 1984). Yet the degree to which these studies have given serious attention to the roles played by yamen clerks and runners has remained negligible. When the yamen staff does make an appearance, it is generally as a parochial interest group whose corrupt activities placed them in conflict with either the magistrate, local communities, or both. Aside from a paucity of primary source material, the lack of scholarly attention given to yamen clerks and runners can in part be traced to the difficulty of fitting them into the Weberian conceptual schema that have so deeply informed studies of Chinese bureaucracy and administration. In this regard, John Watt’s work on the local magistrate and the county yamen in the early Qing period is exemplary (Watt 1972, 1977). Watt added appreciably to the work of Ch’ü T’ung-tsu by highlighting the informal social functions of local government as opposed to its role as part of the formal structure of state authority. According to Watt, the yamen served as a locus of tension between what he viewed as a secular trend towards political centralization and structural rationalization on the one side, and the social forces leading to a diversification of local power on the other (Watt 1972, chap. 7). Critical to this tension, as Watt interpreted it, was the fiscal weakness of the Qing state and the subsequent reliance of local administration on informal sources of revenue (Watt 1977; see also Zelin 1984, chap. 2). Because of this reliance, fiscal operations in the yamen retained what he referred to as a non-rational, non-bureaucratic nature which left local administration economically dependent on local society, thus blocking the aforementioned

68

reed

trend toward centralization and rationalization of administrative authority. In making these observations, Watt’s gaze remained steadily on the magistrate, whom he described as being held hostage between these two competing processes of development (Watt 1972, 211). His consideration of yamen clerks and runners was therefore limited to the ways in which they, as a specifically local interest group, both impinged on the magistrate’s role as a representative of state interests and acted as a brake on further administrative rationalization (Watt 1977, 375–376, 384). Watt’s stress on the county yamen’s dependence on informal financing effectively underscored several of the contradictions within the Qing administrative system. His argument is flawed, however, by his identification of the centralizing features of the early Qing state as constituting a secular trend towards a form of Western-style bureaucratic rationalization. Together with the further assumption that functional rationalization takes place only as a result of formal structural changes as dictated by the central government, Watt concluded that insufficient financing inevitably led to a failed attempt at creating a structural differentiation between state administration and society (Watt 1977, 375). This may well be true. But in the process of making this claim, he disregards the possibility that late imperial China represented not a failed model of Western development so much as it did an alternative pattern of bureaucratic government. His delineation of the irrational elements of local administration under the Qing thus tells us very little as to how this particular form of administration operated, how it managed to survive for so long, or how it might have responded to social, political, and economic changes in the later half of the Qing dynasty. As mentioned above, at least part of the problem here is the uncritical use of Weberian categories. Weber characterized a modern bureaucratic organization as being both legal and rational (Weber 1978, 956–969; Bendix 1960, 423–430). It is rational in that utilitarian administrative duties are clearly tied to specific offices, the functions of which are placed in a well defined hierarchy of responsibility and authority. Such a system is also legal insofar as this rationalization is based on formally legislated and codified organizational and operational regulations to which all members of the organization can appeal and by which all their activities as office holders are delimited and evaluated. Of paramount importance in this regard is that both office and the means of administration with which it is invested are legally separated from the person of the office holder himself. Unlike patrimonial office holding, office-holding within a legal-rational bureaucracy cannot be appropriated as a form of private property. It cannot be bought, sold, inherited or otherwise disposed of by the individual. Nor does the bureaucratic office holder derive his income from

Illicit Bureaucrats

69

the perquisites of office as a form of benefice, but is instead compensated for his services by the state in the form of a regular salary. Although not explicitly, it is this ideal type to which Watt refers when he speaks of “formal structures” or “rationalized bureaucratization” within the state administrative system. Although it has been made many times before, I will risk belaboring the point still further by stressing that Weber presented his description of modern bureaucratic administration as an ideal type which, by definition, did not actually exist in its totality in any historical period. To the extent that this ideal type did correspond to a particular geographical and temporal location, it was based on what Weber saw as the trajectory of European, largely German, development in the twentieth century. Both of these points raise a number of difficulties in applying Weber’s concepts to non-Western historical contexts in general, and the Chinese case in particular. The first difficulty lies in determining exactly what is meant by bureaucracy in the first place.1 If the term is taken to mean either a style of rule based on official functionaries or a collective term referring to those officials, our discussion would clearly be limited to ranking office holders appointed by the central government on the basis of their success in the civil service examinations. It was this body of administrative personnel for whom qualifications and assignment to office, promotion, and disciplinary procedure were all formally determined by codified regulations. It was also clearly this group that was, and still is, referred to by the term “official” or “office holder” (guan) within the Qing system. But if we take bureaucracy more broadly to apply to a form or style of administration, then the lines of demarcation become much less clear. In terms of their functional roles in areas such as tax collection, the maintenance of public order, the operation of the judicial system, as well as the routine flow of documents, yamen clerks and runners undeniably served as an element of bureaucratic administration. Once we include clerks and runners in our definition of bureaucracy, however, a number of problems arise. In the Weberian scheme of things, a bureaucracy is rationalized to the extent that it depends on formalized regulations which both guide the official’s behavior and serve as a source of authority to which he can appeal. To the extent that these rules have a legal basis, it can be assumed that they are codified and enforced by a central government authority. In the context of late imperial China, however, formalized regulation of administrative behavior was restricted to ranking officials and did not extend to the yamen staff at the bottom of the administrative hierarchy. How, then, did this stratum of 1 For a useful discussion of this issue, see Albrow 1970, chapter one.

70

reed

bureaucratic personnel carry out their business in the yamen? How were these people recruited, organized, and assigned specific duties? How, if at all, were they disciplined? How were disputes among them settled? And upon what criteria were their careers in the yamen based? Such questions lead to the issue of the relationship between administrative practice and the legal statutes designed to guide such practice; i.e. administrative law. In the broadest sense, all law in China was administrative in that there was no formal separation between the administrative and judicial functions of government. As noted by Leon Vandermeersch, the system of law laid out in the penal codes of various dynasties served as an administrative tool designed to maintain public order in the interests of the state (Vandermeersch 1985, 14–15; see also van der Sprenkel 1962, 70). If, on the other hand, we take a more functional approach and define administrative law as a body of codified regulations and guidelines specifically dictating the duties, organization, and requirements of administrative personnel, things look a bit different. The statutes and disciplinary regulations of the Qing contain a large number of quite detailed stipulations pertaining to centrally appointed officials, but beneath the level of magistrate there was very little that could be described as formal administrative law. The reason for this was relatively simple: Control over all yamen employees was the personal responsibility of the magistrate. To the extent that codified regulation of this staff did exist, it was found not in the Disciplinary Regulations (Chufen zeli) but in the Qing penal code, where it took the form of severe punishments for criminal transgressions such as extortion, bribery, embezzlement, and the maltreatment of prisoners. The problem, of course, was that the official quotas were woefully inadequate to meet the needs of local administration and were therefore regularly exceeded by the informal employment of personnel over whom the magistrate had little effective control. Insofar as this extra-statutory body of functionally bureaucratic personnel exercised authority outside of the direct, personal supervision of the magistrate, we must inquire into the relationship between bureaucratic authority and power, as well as the sources of legitimacy which might have sustained this system of administration. Weber himself viewed authority as a specific instance of power, one in which power, defined as the ability to impose sanctions and deprivations, is legitimized or institutionalized in such a way as to make compliance voluntary (Weber 1978, 212–214; see also Wrong 1995, 35–41). Specifically bureaucratic authority, in Weber’s view, stems from its impersonal and rationalistic basis which separates the person of the office-holder from the authority and functional utility of the office itself (Gerth and Mills 294–295; Weber 1947, 130–131).

Illicit Bureaucrats

71

It is at this point that the imperial Chinese bureaucracy departed sharply from Weberian typologies. It is not necessary here to engage the debate as to whether the legitimizing principles of Confucian ideology constituted a form of charismatic, traditional, or legal authority. It is sufficient to recognize that the magistrate’s authority flowed neither from the utility of his office within the bureaucratic structure nor from his technical skill at discharging his duties. Rather, it flowed from his identification with the highly personal Confucian moral ideals underwriting both the authority of the imperial state, which the magistrate represented in microcosm, and that of the broader stratum of social elites at the local level. Of course, this does not mean that such authority, as an instance of legitimized power or domination, was devoid of coercive force simply because of the socio-ethical context in which it was set. As Pierre Bourdieu has argued, legitimizing ideologies mask, and thereby help produce, a mis-recognition of coercive power, both on the part of the power holder as well as its subject, by placing it within a cognitive structure or world view wherein submission to its authority is represented as natural, right, and just (Bourdieu 1990, 123–141). But in contrast to the authority of officials, whatever power yamen personnel held by virtue of their functional position and technical expertise lay outside the realm of legitimately sanctioned authority. As James Liu has argued, it was in part the vulnerability of appointed officials and local elites to the functional power of clerks and runners that led to the ideological and social debasement of yamen work by placing its utilitarian function outside the boundaries of Confucian legitimacy (Lo 1987, 24). Corollary to this representation of legitimate authority was the need to control yamen clerks and runners through penal sanctions and the personal supervision of the magistrate to prevent them from using their power in ways that might undermine the paternalistic ideal of the Confucian political order and the legitimacy of imperial institutions. Yet the span of formal bureaucratic supervision under the Qing was far too narrow to provide for effective control over the large number of clerks and runners actually working in county yamens across the empire. How, then, was the de facto power of yamen personnel exercised? And what, if any, alternative sources of legitimacy were used to sustain this system at the local level? As these questions indicate, the holistic application of Weberian categories to the context of local administration in late imperial China inevitably confronts a number of conceptual contradictions and ambiguities. This does not imply, however, that we need to dispense with Weberian theory altogether in our attempt to understand the workings of local government in late imperial China. On the contrary, Weber provides several very powerful analytic tools

72

reed

that can be utilized quite fruitfully in this context. The following chapters will demonstrate, for example, that the organization, division of labor, and procedures of yamen clerks and runners were to a large extent bureaucratically rationalized in the sense that they were based on a body of operational rules to which all yamen employees could appeal. These same rules were also evident in areas of yamen employment such as recruitment, training, jurisdiction, advancement, discipline, and a hierarchical system of rank and authority. Yamen employees, moreover, also established norms of technical competence and behavior in the workplace which, by comparison, were far more rationalized and specific than anything seen in the official bureaucracy. Yet none of these methods of organization or rules of operation were based on codified laws or formal administrative regulations legislated by the central government for the simple reason that no such body of official regulations existed. Yamen rules were instead based on customary procedures which were for the most part formulated and enforced by yamen clerks and runners themselves. Such practices, moreover, should not be dismissed simply as deviations from a legally defined, and therefore more legitimate norm of administrative behavior. To do so would be to miss the degree to which these practices were integral to daily administrative practice within the county yamen. Whatever central government officials may have felt about them, at the local level these informal regulations served as a form of customary administrative law and were frequently presented as such to the magistrate himself in the course of intra-yamen disputes. Within this realm of informal practice, yamen clerks and runners sought to validate their positions and careers by articulating claims to the social and political utility of their role as bureaucratic functionaries and administrative specialists. To the extent that this informal system and its justifications were accepted by successive magistrates, despite the violation of formal statutes, clerks and runners achieved what might be called an illicit legitimacy within the interstices of structural inadequacies and the demands of local administration. In addition to these rationalized and rule driven elements, however, yamen practice was also defined by decidedly non-rational forms of association such as kinship bonds, patron-client ties, and factional formations. In previous studies of Qing local administration, these associations have not only eclipsed entirely the more rationalized aspects of yamen practice but have also been taken as prima facie evidence of the fundamentally self-regarding and deviant behavior of clerks and runners. The tendency here has been to view rationalized and particularistic actions in strictly Weberian terms as fundamentally

Illicit Bureaucrats

73

oppositional and mutually exclusive. Insofar as yamen practice developed outside the boundary of formal structure, however, it is possible, even necessary, to set aside this dichotomy. As the following chapters demonstrate, the county yamen operated not only as an administrative institution, but also as a social sphere where, in the absence of legally defined administrative norms, action was guided by culturally sanctioned modes of particularistic association and practice. Furthermore, although bonds of kinship and factional formation were frequently used to subvert the system of informal yamen rules and procedures, they were more often used to formulate and enforce those same rules. In these instances, the rational and non-rational elements of yamen practice worked together to produce a system of local administration that was functional as well as durable. By recognizing this interaction, we can better understand aspects of Qing local government that might otherwise remain paradoxical. We will also be in a better position to understand the relationship between the yamen and the local community. State and Society Another reason clerks and runners of the county yamen have received so little attention can be traced to the analytic framework most often employed in studies of Chinese state and society in the late imperial era. Recent work in this area has generally concentrated on the patterns of tension, as well as symbiosis, between the legalistic authority of the state, represented at the county level by both the magistrate and local institutions such as the bao-jia (mutual responsibility) and village officer systems, and the less formally defined but nonetheless potent social authority of gentry elites (Rankin 1986; Kuhn 1975; Ch’ü 1962; Watt 1972; Rankin and Esherick 1990). Yet regardless of whether this relationship is apprehended as one of conflict, cooperation, or a mixture of the two, the tendency has been to conceptualize the field of local authority and dominance in essentially bi-polar terms. In the process, the ambiguous yet critical role of non-elites, such as the yamen staff, in mediating the relationship between state and society has been largely overlooked. This tendency is clearly exemplified by the ongoing debate as to where the line of demarcation should be drawn between late imperial Chinese state and society. Roughly speaking, past scholarship has seen this line move from the magistrate (Hsieh 1925) down to the gentry (Hsiao 1960; Ch’ü 1962) back to the magistrate (Watt 1972) and, finally, back down to a point somewhere between the two (Rankin 1986). How clerks and runners might have fit into these conceptual schema has consequently remained problematic, though the

74

reed

general solution has been to simply place them on one or the other side of the divide and have done with them. Watt, for example, identified the emergence of the yamen staff in the early years of the Qing as a “dangerously disconnected group, cut off from the officials and secretaries by social and educational barriers and from the rural people by their urban life and work” (Watt 1972, 154). Yet despite his recognition of this essential ambiguity, he nevertheless places clerks and runners firmly in the constellation of local power groups resisting state intervention in local affairs. Watt’s treatment of the yamen staff is based on his conclusion that while clerks and runners belonged to the formal (state) structure of administration, their primary interests remained centered on the informal (societal) function of the yamen insofar as those functions served their own personal and localized interests (Watt 1977, 372). Although the issue plays but a small part in her study of elite activism in Zhejiang Province, an opposing view is offered by Mary Rankin, who insists that clerks and runners, despite being residents of the local community, functioned primarily as representatives of the state. Having carved out a source of profit at the bottom margins of the state apparatus, she maintains, yamen staff members were fundamentally opposed to the interests of local power groups, especially when those groups sought to undermine the status quo through political reform in the late nineteenth and early twentieth centuries (Rankin 1986, 19, 132). In contrast to both Watt and Rankin, a third group of scholars including Prasenjit Duara (1988), Philip Huang (1985), and Philip Kuhn (1975, 1978), has begun to consider yamen personnel as a separate and critical fulcrum point between state and society. Kuhn, for example, describes the yamen staff during the late Qing as a distinct and well-defined interest group cut off from both the official bureaucracy and local elites by education, training, and personal interest. This trend, he believes, was not appreciably altered in the Republic by the formal incorporation of clerks and runners into the state bureaucracy since they continued to compete with local elites for control over resources to the detriment of state interests (Kuhn 1975, 262, 279). Duara has elaborated on much the same dynamic in his description of low-level administrative personnel in the Republican period as evolving into what he calls “entrepreneurial state brokers.” Operating outside the range of effective control by the state even as they wielded formal state authority, the proliferation and predatory behavior of these individuals eventually resulted in a pattern of growth which Duara identifies as “state involution,” a process which he argues contributed to the de‑legitimation of the Nationalist government (Duara 1988, chap. 2).

Illicit Bureaucrats

75

Although Duara and Kuhn are primarily concerned with developments in the Republican era, the potential autonomy of low-level administrative functionaries which they describe might well be applied to the Qing period as well. If most clerks and runners operated beyond the reach of formal regulatory control and in such numbers as to render personal supervision by the magistrate ineffective, it stands to reason that they, both individually and as a group, developed a degree of independent power viz-a-viz the state, which they nominally served, and the local community within which they operated. The following chapters will, in fact, supply ample evidence of just this sort of autonomy. The trouble with this approach, however, is that it continues a long-standing tendency in the field to view state and society as dichotomous entities locked in perpetual conflict, often to the point where these structures themselves assume a position of historical agency independent of the individuals and groups which composed them. At its best, this conceptual framework has yielded insightful and frequently provocative work, such as in the recent debates concerning the emergence of a civil society and public sphere in late imperial and republican era China.2 Yet even here, in the hands of scholars cognizant of the dangers of cross-cultural generalizations, the approach lends itself to the application of analytical constructs developed out of the European experience by assuming a clear separation of cultural, social, and state institutions and practices which did not exist in late imperial China (Rowe 1993). Rather than replicating this tendency by once more attempting to locate clerks and runners somewhere between the state and society dichotomy, it seems much more fruitful to reformulate the conceptual scheme itself. Philip Huang has made a valuable step in this direction in his recent work on civil justice in the Qing dynasty (Huang 1993; 1996). Within the Qing justice system, Huang identifies an intermediate sphere judicial practice, where the formal apparatus of the magistrate’s court interacted with elements of informal community mediation to form what he refers to as a third realm of local dispute resolution. In this, Huang persuasively argues that despite the Qing code’s lack of formally constituted civil law, in practice the court and code were nevertheless regularly used by local residents as an avenue of redress in civil disputes. One of the benefits of Huang’s approach is that it forces recognition that late imperial state and society frequently operated not as discrete entities but instead through areas of overlap and interpenetration. To apprehend this 2 For the most succinct summation of the issues in this debate, see articles by Huang, Rowe, Rankin, and Wakefield in the symposium issue of Modern China 19, 2 (April) 1992.

76

reed

sort of interaction, moreover, it is necessary to move beyond an examination of formal structures and codified law to consider how these structures and codes worked in the often paradoxical realm that existed between official representation and daily practice. The following chapters build upon this formulation, but also attempt to push it yet a bit further. In doing so, I draw upon S. N. Eisenstadt’s concept of free-floating resources. In his work The Political Systems of Empires, Eisenstadt describes these resources as “manpower, economic resources, political support, and cultural identifications (which are) not embedded within or committed beforehand to any primary ascriptive-particularistic groups” (Eisenstadt 1963, 27). The existence of such free-floating resources serves to create what Eisenstadt refers to as a “reservoir of generalized power” which could be used by different social or occupational groups in the pursuit of varying goals. Eisenstadt incorporates the availability of these resources into his taxonomy of historical empires. Insofar as he describes the availability of free-floating resources as a product of a given socio-political structure, however, his description remains static. This shortcoming has recently been overcome by sociologist Michael Mann in his work on the sources of social power. In Mann’s view, societies are best seen not as unitary or bounded social systems or structures but rather as “multiple overlapping and intersecting sociospatial networks of power” rooted in ideological, economic, political, and military relations (Mann 1986, 1–2). In addition to revealing themselves through social interaction, these networks of power operate as well through the organizational and institutional means available to individuals and groups in their efforts to achieve human desires. While encompassing Eisenstadt’s notion of free-floating resources, Mann’s work has the benefit of focusing less on structure and more on interaction between individuals, the nature of which is responsive to changes in circumstances. Rather than treating late imperial state and society as dichotomous and juxtaposed structures, whether interpenetrating or oppositional, I approach them instead as nests of socially differentiated institutions, practices and resources, all of which exhibited varying degrees of monopolistic control and social fixity and were thus subject to varying degrees of appropriation and competition. By resources, I mean those factors, both material and symbolic, from which individuals draw in their daily efforts to secure income, prestige, and sustainable livelihood. In the context of county government and local authority in late imperial China, these resources were as varied as the control over the material means of administration (records, files, seals, jails, etc.), organizational and disciplinary control over the yamen staff, access to and strategic use of the magistrate’s court, and the symbolic capital generated

Illicit Bureaucrats

77

by association with either imperial institutions or Confucian ideological orthodoxy. Access to such resources was conditioned by an individual’s physical, functional, as well as symbolic location within specific governmental and societal networks, both formal and informal; e.g. a clerk or runner’s functional indispensability, a magistrate’s position as an appointed official and representative of imperial authority, or the role of local gentry as community leaders and defenders of community interests. Resources, moreover, were rarely unitary in the benefits they might be made to yield by different individuals or groups. If the magistrate’s court served the government as a symbolic font of imperial authority, for example, it also served clerks and runners as a purely economic resource while at the same time providing local residents with an alternative avenue of dispute resolution. Although this approach is considerably hazier and analytically less elegant than the traditional dichotomy of state and society, it nevertheless more closely approximates the untidy realities of social and political relations at the county level in late imperial China. Rather than a neatly defined line of demarcation between state and society across which interaction occurs, we instead see county government as an area of negotiation, exchange, and competition, where informal and formal elements of administration combined with social institutions, customary practices, and personnel as well as group interests to produce highly localized patterns of cooperation as well as conflict. This approach reduces the importance of the traditional dichotomy between agency and structure by reminding us that terms such as state, society and community are themselves theoretical constructs. Furthermore, although such constructs imply static sets of relations and structures, in actuality no such permanence existed. Social and political relations, as well as the practices through which they are realized, are instead in a constant state of flux and change as individuals and groups maneuver to take advantage of newly available resources and strategies. The ability of yamen clerks to gain economic benefit from legal cases, or the ability of local elites to influence the operations of the yamen staff, for example, are both instances of the use of specific resources by specific groups for particular purposes. The availability of these resources, for individuals as well as groups, itself changes in accordance with systemic changes such as population increase, an expansion of commercial economic activity, the level of litigation occasioned by social conflict, and the ability of the government to maintain control over the means of administration. In addition, the approach requires us to focus less on abstracted structures and more directly upon on the historical agents themselves, those individuals and groups of individuals who acted within structured sets of circumstances, expectations, and opportunities even as their actions constantly

78

reed

worked to create, sustain, as well as modify those same structures. As this suggests, we thus need to bring a more anthropological perspective to our consideration of local administrative practice than has previously been the case. By doing so, we can highlight not only the structures and group interests that have traditionally concerned sociologists, and historical sociologists in particular, but also the ways in which the operation of a given political‑administrative system unfolded within the context of the culture in which it is located. This is particularly true of yamen practices that have traditionally fallen under the rubric of corruption. Corruption The image of corruption among yamen clerks and runners has been one of the most durable shibboleths in our perceptions of county government under the Qing dynasty. Given the pervasiveness of such images in Qing source material, this is hardly surprising. What is curious is the surprisingly few number of studies that have undertaken to explore this facet of Qing local administration in any depth or detail. At least part of the reason for this lacunae is the sheer slipperiness of the concept of corruption itself and its resistance to precise definition and analysis (see Heidenheimer 1989; Roy 1970; Deysine 1980; Theobald 1990). Among European and American social scientists, the general trend over the past century has been away from a view of political or administrative corruption as indicative of systemic moral decay to that of corruption as a form of deviant behavior on the part of individual officials (Heidenheimer 1969, introduction). Recent scholarship has thus attempted to define corruption by contrasting specific actions with either established laws (legalistic definitions), declared ideals of public office (public office definitions), or public perceptions of official behavior (public interest definitions). At the root of all these efforts and the behavioral models which they have produced, lies Weber’s idealized notion of rationalized bureaucracy and its separation of private-regarding interest from the institutional and social functions of public office. There are a number of problems with this approach, particularly in its applicability to historical and non-Western contexts. To begin with, as Robin Theobald notes, the approach is flawed from the start by an overly precise conception of bureaucratic office holding which fails to take into account the discretionary power of any public office holder and other such practical elements blocking complete rationalization (Theobald 1990, 11). Along similar lines, James Scott has argued that some form of corruption is, in fact, integral to almost all political systems, an element which is overlooked by the tendency to give normative value to declared standards of orthodox political conduct (Scott 1972, 4).

Illicit Bureaucrats

79

Scott, moreover, is one of a number of Western social scientists who have begun to look at corruption not as a degenerative or pathological phenomenon but rather as a facilitative pattern of behavior (Leff 1964; Leys 1965; Huntington 1968; Nye 1978). Focusing primarily on developing nations in the twentieth century, these scholars have called into question the application of moral as well as behavioral standards to national contexts where such norms either do not exist or have remained only partially internalized. Following Colin Leys’ argument that corrupt actions can often serve quite positive purposes, Scott has suggested that what is often called corruption may represent a channel for the satisfaction of demands without which more formal political and social formations could not survive (Scott 1972, preface viii). In this regard, he continues, regularized patterns of corruption may be seen as an alternative, informal political system operating alongside formally constituted structures and publicly declared norms (Scott 1972, 2). Many of these same trends are evident in the scholarship on corruption in Chinese government as well. Western observers in China during the nineteenth century, for example, saw the payment of customary fees (luogui) to local officials and yamen employees alike as indicative of systemic corruption and moral decline (Brenan 1897; Alabaster 1960). Nearly a century later, the noted scholar of imperial Chinese bureaucracy Etienne Balazs echoed this same assumption by regarding corruption as a defining feature of late imperial Chinese society, a view which, he admitted, left him puzzled as to how such a system could have continued to function for so long (Balazs 1965, 68, 73). More recently, scholars have moved away from such broad assumptions to focus attention more narrowly on political and administrative corruption in the People’s Republic of China (PRC), particularly on the post-Mao era of economic reform (Chan and Unger 1982; Liu 1983, 1985; Ma 1989; Oi 1989; Myer 1989; Lee 1990; Tang 1996; White 1996). Set within the context of increasingly lax administrative supervision and control on the part of both the communist party and the central government, this scholarship has combined an impressive array of methodologies ranging from structuralist and culturalist explanations to economistic analyses such as rational-choice theory and principal-agent/cost-benefit models. Like those arguing for the facilitative aspects of corruption in other national contexts, moreover, many of these studies have approached administrative corruption in modern China as a form of rational and even economically positive form of behavior. Gordon White, for example, has concluded that the de-politicization of both the bureaucratic elite as well as of society at large since the death of Mao Zedong has rendered corruption an accepted means of getting ahead in an environment characterized by relative scarcity and increasing commercialization (White 1996, 158).

80

reed

Although these studies have produced quite valuable insights into the linkage between corruption and economic development in the PRC, similar methodologies have enjoyed only limited success in those few cases where they have been applied to the imperial period (e.g., Marsh 1962; Kiser and Tong 1992). The most persistent problem remains that confronting all scholars of administrative corruption, defining the term corruption itself. Even among those who view corruption as a potentially positive factor in national development, the question of what actually constitutes corruption remains only vaguely or arbitrarily resolved. As Anne Deysine comments in her review of the social science literature on the subject, the Gordian knot of definition has gradually been passed over by most political scientists in favor of an implicit agreement that everyone seems to know, subjectively at least, what corruption is (Deysine 1980, 448). In the China field this view is represented by James Myers’ claim that despite differences in detail, all definitions of corruption share as a common theme “the notion of deviation by public officials from some standard of rectitude” (Myers 1989, 197). But if this solution has freed political scientists to move on to less intractible aspects of corruption in the modern period, it nevertheless begs questions that should be of central concern to historians: By whose standard of rectitude was corruption measured or assessed by contemporaries and in what instances and by whom were such standards applied? Those few historians who have addressed the issue of yamen corruption in the Qing period have, by and large, been reticent to confront these questions directly. Ch’ü T’ung-tsu, for example, saw the “deviant” behavior of yamen clerks as at least in part the result of a dissonance between cultural values and social position. “In other words, when one has no legitimate means to secure the desired values—pecuniary reward, career opportunity—owing to one’s unprivileged position, one tends to seek these values by illegitimate means” (Ch’ü 1962, 197; see also Liu 1967). To this explanation Ch’ü adds that corruption among yamen personnel might also be examined as a variant form of normative behavior. As in other occupations, if the majority of group members are engaged in specific sorts of activity, they will exert pressure upon others to conform. “Thus, what the government and the public consider deviant behavior or corruption may be regarded as conformity to the occupationally defined norms of behavior” (Ibid.). But Ch’ü does not pursue the implications of this provocative observation. Maintaining his focus on the dysfunctional elements of Qing bureaucracy at the local level, he instead goes on to replicate the views of his primary sources that endemic corruption among yamen clerks and runners constituted an insurmountable barrier between the people and imperial government.

Illicit Bureaucrats

81

The obvious but rarely acknowledged problem with this uncritical use of source material was emphasized by Adam Lui’s suggestion that the widespread reputation of yamen employees for corruption is due at least in part to the fact that those who left the historical record—officials, scholars, and elites—were exactly those with both the reasons and means of castigating them in a public manner. Yet despite this rare insight, Lui himself then goes on to conform to that same record by describing yamen underlings as “selfish cowards (and) petty nuisances of society which the public could not do without but could possibly tolerate” (Lui 1979, 24–25). As described above, the infamous reputation of clerks and runners was, and still is, based primarily on the collection of customary fees from anyone having any contact with the yamen. Since the collection of such fees was expressly forbidden by Qing law, and since all clerks and runners relied to a greater or lesser extent upon these fees, it makes a certain amount of sense to conclude that clerks and runners were indeed corrupt. Although this might ring true with our own twentieth-century Western sensibilities in regard to the use of public office for personal gain, it nevertheless relies on a standard, expressed in either strictly legalistic or ethical terms, which is provided to us by the empowered segments of Qing society. Aside from constituting an unwarranted generalization of elite representations and normative standards to all of Chinese society, these broad assumptions as to the venality of clerks and runners presuppose a clear distinction between public and private resources and structures. As this study will demonstrate, however, such a separation did not exist in local administrative practice. Nowhere does this become more clear than in the operations of the magistrate’s court. According to the views of scholars, officials, and emperors throughout the Qing period, the magistrate’s court was manifestly not intended to serve as a venue for the resolution of civil disputes between local residents. The intended purpose of the court, as well as the legal code in general, was instead to reinforce imperial authority by maintaining the cosmic balance of order and chaos through the judicious assignment of appropriate punishment for specific offences. In serving this purpose, the magistrate’s court functioned not only as the site of penal sanction, but also, and even more importantly, as a ritual center. In this respect, the court, along with all its trappings and language, symbolically represented imperial authority and its disinterested alignment with higher principle and justice, in this case read as the maintenance of the Confucian socio-political order. The collection of customary fees by yamen clerks and runners threatened to undermine this legitimizing function of the court by publicly associating the administration of justice with private gain. Even more alarming was the belief

82

reed

regularly expressed by officials and local gentry alike that clerks and runners attempted to further their pecuniary interests by duping local residents into going to court over trivial matters and then forcing them to pay ruinous levels of fees. Such practices, particularly the formal presentation of disputes which should ideally be settled amicably through informal mediation, could only result in a disruption of civic harmony. Clerks and runners were thus seen not only as avaricious, but as an active threat to the very basis of the social order upon which both gentry and imperial authority rested. Early Western studies of Qing law accepted these portrayals of yamen clerks and runners almost verbatim. When combined with the widely held view of the Qing code as an arbitrarily applied and entirely penal instrument of state authority, many scholars concluded that the system was incapable of providing what they considered to be an adequate and fair administration of justice (Ch’ü 1962, 1965; van der Sprenkel 1977; Bodde and Morris 1967; Unger 1976). Received wisdom in this area also held that because of the relentless dunning of fees from anyone even remotely connected to formal legal proceedings, local residents regarded involvement with the yamen court as a ruinous plague to be avoided at all costs. More recent scholarship in this field has begun to challenge many of the Western biases and broad assumptions of these early studies (Buxbaum 1971; Alford 1984, 1986; Ocko 1988; Sommer 1994; Bernhardt and Huang 1994; Huang 1993, 1996). Most importantly, this new generation has demonstrated that to understand the significance of the Qing legal system, as either a repository for normative social ideals or as an embodiment of state authority, it is necessary to go beyond the formal structure and provisions of the Qing code to examine how those provisions were utilized in the context of judicial practice. The present study adds to this growing body of knowledge by revealing the centrality of the magistrate’s court not only to the judicial system, but also to many of the yamen’s basic administrative functions. In this respect, the fees collected from legal cases served as a critical source of revenue providing both the personal income necessary to maintain a large staff on a full-time basis, as well as the operational financing for departments and divisions throughout the yamen. Far from constituting a dysfunctional form of corruption, legal case fees often served as the fiscal cornerstone of local administration and as such they were regularly defended as right and proper before the magistrate himself in the course of intra-yamen conflicts. To accept elite and official portrayals of yamen corruption at face value runs the risk of overlooking the crucially important functional role of these extrastatutory and informal practices. My reason for belaboring this point here is not to rehabilitate clerks and runners or to rescue them from their historical

Illicit Bureaucrats

83

reputation. As the following chapters will amply testify, members of the yamen staff frequently engaged in practices that were considered to be corrupt by even the narrowest of definitions on the part of contemporaries.3 My purpose, rather, is two-fold. First, an effort must be made to distinguish between standard practices which, despite their technical illegality, served to sustain local administrative functions and those more purely extractive or private-regarding acts which deviated from even these informal procedures and norms. Admittedly, the line between the two was often blurred, sometimes purposefully so. But to ignore the distinction altogether merely for the sake of analytic clarity robs the term corruption of any meaningful application. Even more significantly, it also clouds the actual relationship between local government and society and renders the continuation, much less the rationalization, of formally illicit practices incomprehensible. Along these same lines, I also suggest that the portrayal of yamen corruption provided by elite and official sources should itself be made subject to interrogation. Failure to do so inevitably privileges the normative value of elite standards and formal administrative structure and thus blinds us to the essential role played by the informal, extra-legal realm of practice. It also prevents us from seeing the ways in which the morally embroidered image of yamen corruption was itself very much a structured portrayal, based upon and articulated through Confucian discourse. Insofar as that discourse was used to help maintain a particular social and political hierarchy, images of corruption must also be seen as resources utilized in the defense of specific interests and power structures. This does not mean that I am adopting an entirely materialistic view of the role of ideology or ideologically driven discourse. Although such discourse was at times used consciously by certain individuals as a tool to achieve specific ends, ideology needs to be taken much more seriously than a purely materialist approach would suggest. The effectiveness of ideological discourse as a resource in any given interaction depended above all on the degree to which it was accepted as reflective of broader cultural perceptions of what constituted a just society. That such ideologically based definitions of the proper order of things tended to support a particular pattern of dominance and was most 3 Published collections of archival documents offer abundant examples of the often ingenious methods used by yamen personnel to squeeze more from the administrative process. That these examples usually constitute the entirety of archival samples relating to clerks and runners, however, underscores the degree to which Chinese archivists and scholars making the selection regard clerks and runners as nothing more than corrupt elements. See, for example, Qingdai Baxian dang’an huibian; Qingdai Qian Jia Dao Baxian dang’an xuanbian.

84

reed

often articulated by the privileged elements of the socio-political order does not lessen the influence of those definitions in shaping cultural values and expectations at either end of the socio-economic continuum. In other words, ideological values played a critical role in structuring an individual’s perceptions of the social world, his place within that world, as well as what he saw as the realm of the possible. In contrast to many materialist interpretations of ideology, there is no basis here for an a priori assumption on the part of the historian that the articulation and use of these values in any given instance was necessarily disingenuous or used as a cover for a somehow more real or genuine intent (Bloch 1975, 2). Neither can we assume that less privileged members of society did not at times interpret and use elements of these same perceptual categories in creative ways to claim a certain degree of legitimacy and prestige for themselves and their occupations or that these claims were in any way less earnest that those made by social and political elites. In the following chapters, I thus consider portrayals of yamen corruption not as strictly factual accounts, but rather as discursive strategies drawing upon orthodox Confucian concepts. In this light, we will see how stereotypic images were frequently employed as a method of character assassination by clerks and runners themselves in the course of intra-yamen disputes. More significantly, we will see that in the process of utilizing these images, yamen employees appropriated elements of Confucian orthodoxy in order to establish their own sets of internalized norms and standards. In this manner a formally illicitly stratum of administrative personnel could frame their frequently illegal practices as an honorable occupation and positive contribution to local government and society. References Alabaster. 1960. “Notes on Chinese Law and Practice Preceding Revision.” Journal of the North China Branch, Royal Asiatic Society, 37: 139–141. Albrow, Martin. 1970. Bureaucracy. New York: Praeger. Alford, William. 1984. “Of Arsenic and Old Laws: Looking Anew at Criminal Justice in Late Imperial China.” California Law Review, 72.6: 1180–1256. Allee, Mark A. 1994. Law and Society in Late Imperial China: Northern Taiwan in the Nineteenth Century. Stanford: Stanford University Press. Balazs, Etienne. 1965. Political Theory and Administrative Reality in Traditional China. London: University of London. Bendix, Reinhard. 1960. Max Weber: An Intellectual Portrait. Garden City: Doubleday Anchor.

Illicit Bureaucrats

85

Bernhardt, Kathryn, and Philip C. C. Huang, ed. 1994. Civil Law in Qing and Republican China. Stanford: Stanford University Press. Bloch, Maurice. 1975. “Introduction.” in Political Language and Oratory in Traditional Society, edited by Maurice Bloch. London: Academic Press. Bodde, Derk, and Clarence Morris. 1967. Law in Imperial China: Exemplified by 190 Ch’ing Dynasty Cases. Philadelphia: University of Pennsylvania Press. Bourdieu, Pierre. 1989. “Social Space and Symbolic Power.” Sociological Theory, 7.1: 14–25. ———. 1990. The Logic of Practice. Stanford: Stanford University Press. Brenan, B. 1897. “The Office of District Magistrate in China.” Journal of the North China Branch of the Royal Asiatic Society, 32: 36–65. Brunnert, H. S., and V. V. Hagelstrom. 1912. Present Day Political Organization in China. Shanghai. Buxbaum, David C. 1971. “Some Aspects of Civil Procedure and Practice at the Trial Level in Tanshui and Hsinchu from 1789 to 1895.” Journal of Asian Studies, 30.2: 255–279. Chan, Anita, and Jonathan Unger. 1982. “Grey and Black: The Hidden Economy of Rural China.” Pacific Affairs, 55.3: 452–471. Ch’ü T’ung-tsu. 1962. Local Government in China Under the Ch’ing. Cambridge: Harvard University Press. ———. 1965. Law and Society in Traditional China. Paris: Mouton. Dai Yanhui. 1979. Qingdai Taiwan zhi xiangzheng (Local Government in Taiwan During the Qing Period). Taibei: Lianjing chubanshe. Deysine, Anne. 1980. “Political Corruption: A Review of the Literature.” European Journal of Political Research, 21.2: 447–462. Duara, Prasenjit. 1988. Culture, Power, and the State: Rural North China, 1900–1942. Stanford: Stanford University Press. Eisenstadt, S. N. 1963. The Political Systems of Empires. New York: The Free Press. Esherick, Joseph W., and Ye Wa. 1996. Chinese Archives: An Introductory Guide. Berkeley: University of California Press. Esherick, Joseph W., and Mary Backus Rankin. 1990. Chinese Local Elites and Patterns of Dominance. Berkeley and Los Angeles: University of California Press. Gerth, H. H., and C. Wright Mills, ed. 1946. From Max Weber: Essays in Sociology. New York: Oxford University Press. HCJSWB: Huangchao jingshi wenbian (Collected Qing Memorials on Statecraft). 1963 reprint. Taipei: Guofeng chubanshe. Heidenheimer, Arnold J., Michael Johnston, and Victor T. LeVine. 1989. Political Corruption: A Handbook. New Brunswick: Transaction Publishers. Hsiao, Kung-chuan. 1960. Rural China: Imperial Control in the Nineteenth Century. Seattle: University of Washington Press.

86

reed

Hsieh Pao-chao. 1925. The Government of China, 1644–1911. Baltimore: The Johns Hopkins Press. Huang, Philip C. C. 1985. The Peasant Economy and Social Change in North China. Stanford: Stanford University Press. ———. 1993. “Between Informal Mediation and Formal Adjudication: The Third Realm of Qing Justice.” Modern China, 19.3: 251–298. ———. 1996. Civil Justice in China: Representation and Practice in the Qing. Stanford: Stanford University Press. Huntington, Samuel. 1968. Poltical Order in Changing Societies. New Haven: Yale University Press. Kiser, Edgar, and Xiaoxi Tong. 1992. “Determinants of the Amount and Type of Corruption in State Fiscal Bureaucracies: An Analysis of Late Imperial China.” Comparative Political Studies, 25.3: 300–331. Kracke, E. A. 1953. Civil Service in Early Sung China, 960–1067. Cambridge: Harvard University Press. Kuhn, Philip A. 1975. “Local Self Government Under the Republic.” In Conflict and Control in Late Imperial China, edited by Frederic Wakeman and Carolyn Grant, 257–298. Berkeley: University of California Press. Lee, Peter. 1990. “Bureaucratic Corruption in the Deng Xiaoping Era.” Corruption and Reform, 5.1: 29–47. Leff, N. H. 1964. “Economic Development Through Bureaucratic Corruption.” American Behavioural Scientist: 8–14. Leys, Colin. 1965. “What is the Problem About Corruption?” Journal of Modern African Studies, 3.2: 214–224. Liu, James T. C. 1967. “The Sung Views on the Control of Government Clerks.” The Journal of the Economic and Social History of the Orient, 10.2–3: 317–344. Liu, Alan. 1983. “The Politics of Corruption in the PRC.” American Political Science Review, 77: 602–623. ———. 1985. “The Dragon’s Teeth of Mao Tse-tung: Cadres in Mainland China.” Issues and Studies, 21.8: 12–23. Liu Min. 1983. “Qingdai xuli yu guanliao zhengzhi (Clerks and Bureaucratic Administration in the Qing).” Xiamen daxuebao, 3: 75–82. Lo, Winston W. 1987. An Introduction to the Civil Service of Sung China. Honolulu: University of Hawaii Press. Lui, Adam Y. C. 1979. Corruption in China During the Early Ch’ing Period, 1644–1660. Hong Kong: University of Hong Kong. Ma, Stephen K. 1989. “Reform Corruption: A Discussion of China’s Current Development.” Pacific Affairs, 62.1: 40–52. Mann, Michael. 1986. The Sources of Social Power. Cambridge: Cambridge University Press.

Illicit Bureaucrats

87

Marsh, Robert M. 1962. “The Venality of Provincial Office in China and in Comparative Perspective.” Comparative Studies in Society, 4.4: 454–466. Mayers, William Fredrick. 1878. The Chinese Government: A Manual of Chinese Titles Categorically Arranged and Explained. Shanghai: American Presbyterian Mission Press. Metzger, Thomas A. 1973. The Internal Organization of Ch’ing Bureaucracy: Legal, Normative, and Communicative Aspects. Cambridge: Harvard University Press. Min, Tu-ki. 1989. National Polity and Local Power: The Transformation of Late Imperial China. Cambridge: Harvard University Press. MLS: Muling shu (Selected Writings of Local Officials). 1868 edition. Myers, James. 1989. “China’s Modernization and Unhealthy Tendencies.” Comparative Politics, 21.2: 193–214. Nye, J. S. 1978. “Corruption and Political Development: A Cost-Benefit Analysis.” In Political Corruption, A Handbook, edited by Arnold J.Heidenheimer et al., 564–578. New Brunswick: Transaction Publishers. Ocko, Jonathan K. 1983. Bureaucratic Reform in Provincial China: Ting Jih-ch’ang in Restoration Kiangsu, 1867–1870. Cambridge: Harvard University Press. Oi, Jean. 1989. “Market Reforms and Corruption in Rural China.” Studies in Comparative Communism, 22.2–3: 221–233. Rankin, Mary Backus. 1986. Elite Activism and Political Transformation in China, Zhejiang Province, 1865–1911. Stanford: Stanford University Press. Rowe, William T. 1993. “The Problem of ‘Civil Society’ in Late Imperial China.” Modern China, 19.2: 139–157. Roy, Edward Van. 1970. “On the Theory of Corruption.” Economic Development and Cultural Change, 19.1 (October): 86–110. Schoppa, Keith R. 1982. Chinese Elites and Political Change: Zhejiang Province in the Early Twentieth Century. Cambridge: Harvard University Press. ———. 1989. Xiang Lake: Nine Centuries of Chinese Life. New Haven: Yale University Press. Scott, James C. 1972. Comparative Political Corruption. Englewood Cliffs: Prentice Hall. Shiga, Shuzo. 1974. “Criminal Procedure in the Ch’ing Dynasty, part 1.” Memoirs of the Research Department of the Toyo Bunko, 32: 1–45. ———. 1975. “Criminal Procedure in the Ch’ing Dynasty, part 2.” Memoirs of the Research Department of the Toyo Bunko, 33: 115–138. ———. 1976. “Criminal Procedure in the Ch’ing Dynasty, part 3.” Memoirs of the Research Department of the Toyo Bunko, 34: 16–26. Skinner, G. William. 1977. “Cities and the Hierarchies of Local Systems.” In The City in Late Imperial China, edited by William G. Skinner, 275–352. Stanford: Stanford University Press.

88

reed

Sommer, Matthew H. 1994. “Sex, Law and Society in Late Imperial China.” Ph.D Dissertation, UCLA, 1994. Tang Shengming. 1996. “Official Corruption in a Developing Economy: The Case of the People’s Republic of China.” Humanity and Society, 20.2: 58–71. Theobald, Robin. 1990. Corruption, Development and Underdevelopment. Durham: Duke University Press. Unger, Roberto M. 1976. Law in Modern Society: Toward a Criticism of Social Theory. New York: The Free Press. van der Sprenkel, Sybille. 1977. Legal Institutions in Manchu China: A Sociological Analysis. New York: Humanities Press. Vandermeersch, Leon. 1985. “An Enquiry into the Chinese Conception of Law.” In The Scope of State Power in China, edited by Stuart R. Schram. London: School of Oriental and African Studies. Wakeman, Frederic Jr. 1993. “The Civil Society and Public Sphere Debate: Western Reflections on Chinese Political Culture.” Modern China 19, no 2: 108–38. Wang Qing. 1988. Sichuansheng dang’anguan, guancang dang’an gaishu (A General Description of the Holdings of the Sichuan Provincial Archives). Chengdu: Sichuansheng shehui kexueyuan chubanshe. Wang Yanyuan. 1989. “Lun Qingdai duli (Clerical Corruption in the Qing).” Liaoning daxue xuebao, 5: 53–56. Watt, John R. 1972. The District Magistrate in Late Imperial China. New York: Columbia University Press. ———. 1977. “The Yamen and Urban Administration.” In The City in Late Imperial China, edited by William G. Skinner, 353–390. Stanford: Stanford University Press. Weber, Max. 1947. Social and Economic Organization. New York: The Free Press. ———. 1978. Economy and Society. Berkeley: University of California Press. White, Gordon. 1996. “Corruption and the Transition from Socialism in China.” Journal of Law and Society, 23.1: 149–169. Wrong, Dennis H. 1995. Power: Its Forms, Bases, and Uses. New Brunswick: Transaction. Wu Jiyuan. 1993. “Shilun qingdai liyi de zuoyong he diwei (The Position and Functions of Clerks and Runners in the Qing).” Qingshi yanjiu, 3.1: 47–55. Wu, Silas. 1970. Communication and Imperial Control in China: Evolution of the Palace Memorial System, 1693–1735. Cambridge: Harvard University Press. Zelin, Madeline. 1984. The Magistrate’s Tael: Rationalizing Fiscal Reform in EighteenthCentury Ch’ing China. Berkeley: University of California Press. Zhao, Shiyu. 1989. “Liangzhong butong de zhengzhi taidu yu Ming Qing xuli de shehui diwei (Two Different Conceptions of Government and Social Position of Clerks and Runners during the Ming and Qing).” Zhengzhixue yanjiu, 1: 50–56. Zheng, Qin. 1988. Qingdai sifa shenpan zhidu yanjiu (The Judicial System of the Qing). Changsha: Hunan jiaoyu chubanshe.

chapter 4

From Oral Testimony to Written Records in Qing Legal Cases* Yasuhiko Karasawa Introduction A criminal case is an event before it is a case record. The written record, with its implicit claims to truthfulness, is in fact the result of a complex manipulation of the original process of investigation, interrogation and analysis that constitutes court proceedings. In order to illuminate that process, this chapter examines one kind of document that was essential to the record of criminal cases in the Qing dynasty. These are the written records of oral testimony taken from defendants and key witnesses, which were written down separately based on the notes taken at their depositions, and then woven into the case reports composed to explain the case and justify the magistrate’s verdict. These documents, called “formal testimony,” constituted key pieces of the evidence made available to the superior judicial officials who reviewed all serious crimes. The most important were likely to be criminal confessions. I find two broad influences on criminal case recordings that derive from Qing-dynasty legal culture and literary convention respectively. First, the requirements of judicial review of all serious felonies made legal specialists composing drafts of formal testimony strive for consistency in details, wording, and plot among the stories of different parties. In those obligatory judicial reviews, the credibility of testimony woven into the narrative of the case report played a decisive role in persuading senior authorities to accept the magistrate’s suggested sentence. Second, the broader written culture of the Qing supplied models of narrative fashioning that played a crucial role in producing * Originally published in Charlotte Furth, Judith T. Zeitlin, and Ping-chen Hsiung, eds., Thinking with Cases: Specialist Knowledge in Chinese Cultural History (University of Hawai’i Press, 2007, pp. 101–22). Used with permission of University of Hawai’i Press. A different Japanese version of this essay appeared in 1995 as “Hanasukoto to kakukoto no hazamade: Shindai saiban bunsho ni okeru kyōjutsusho no tekusutosei (Between speech and writing: Textuality of the written records of oral testimony in Qing legal cases),” Chūgoku: Shakai to Bunka, 10: 212–250.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_�06

90

Karasawa

a highly formalized and concise style of reporting oral testimony. In particular, vernacular stories as a literary genre provided scribes who worked at county government offices ( yamen) and who were part of its audience with the lexicon and other linguistic tools to take oral testimony originally uttered in local dialects or regional languages and record it in the uniform vernacular style derived from Mandarin speech (guanhua). We usually find such formalized, concise style in the only surviving written records of oral testimony in local-level case records kept stored at the county offices. These official case files do not include original depositions or drafts written prior to the final, decisive version, probably because it was not necessary to expose these documents to the eyes of senior authorities. However, one archive of local-level case records—the so called “cases directly examined by the Board of Punishments” (Xingbu-xianshen, referred to as “directly examined”)—contained in Beijing’s Board of Punishments archive reveals the process of revising drafts of written testimony into final form. “Directly examined” cases involved felonies that occurred in two counties surrounding Beijing and in Beijing itself.1 The Board of Punishments, probably because it was the empire’s supreme judicial authority responsible for the security of the emperor and court, left detailed records of the interrogation process of local cases that took place in the capital. Because of their more detailed documentation, “directly examined” cases tell us how the interrogations were carried out and how the final draft of testimony was formalized. Through the process of recording depositions and producing successive drafts of testimony based on the interrogations, the raw dossier evolved into a polished, consistent final draft, without contradictions or ambiguity. The raw dossier itself never finalized the “truth” of the case; the valid legal case claimed truth when it arrived at a narrative that offered one and only one reconstruction of what actually happened. Using material from the Beijing Board of Punishments, this chapter shows how a legitimate legal case record of testimony was constructed, involving linguistic manipulation with the help of narrative devices derived from the written culture in the Qing.

The Status of Depositions in Qing Legal Procedure

Qing legal procedure is characterized by its well-organized system of obligatory reviews. Let me briefly summarize this system in order to show why witnesses’ testimonies occupied a key position in legal case records. All legal cases, great 1 Shiga 1974: 23; 1984: 26–27. On the detailed procedure of handling cases arising within and around Beijing, see Na 1992: 295–380.

From Oral Testimony to Written Records in Qing Legal Cases

91

and small, were first heard at the county office where the magistrate served as judge.2 For serious cases requiring penalties including or exceeding temporary exile, a magistrate would file a formal case report including verified facts and the “suggested sentence” (ni) according to law. He sent this, along with the convicted criminal in person, to his senior official, the prefect. If the prefect verified the facts and found the suggested sentence reasonable, the case record, along with the criminal, was sent to his senior official, the judicial commissioner at the provincial capital. After investigating the documents and questioning the criminal, the judicial commissioner, if he found the suggested sentence proper, forwarded the case to the governor of the province. A noncapital case carrying a suggested sentence of temporary exile could be approved and carried out on the governor’s authority. Homicide cases and those that required penalties of life exile or military servitude were sent to the Board of Punishments in Beijing. Only with its endorsement were the penalties finalized. Finally, cases requiring capital punishment had to be approved by the emperor. In this fashion, the establishment of penalties including and exceeding temporary exile required successive review at several levels of government.3 The obligatory review system served to check any errors in the judgment of the crime. At each stage of review, documents and the testimony of the criminal and witnesses were carefully examined. If any doubts arose, the case would be returned to a lower level with severe admonition. A mishandled case, if found, would damage a magistrate’s future career. Thus, the obligatory review system always placed pressure on the magistrate to be as careful as possible with all serious criminal procedures. (If a dissatisfied litigant appealed his case, a magistrate might even be ordered to send up to superiors the records of lesser “civil procedure” cases (xishi), which were not usually the subject of obligatory review).4 Therefore, in documenting a case, the primary requirement for a convincing case report was to provide consistent and firm testimony from the parties and witnesses. But the reason testimony was so central requires further explanation. The legal historian Shiga Shūzō argues that in contrast to modern Western criminal procedure, in which factual issues are decided by “adjudication,” Qing criminal procedure emphasized decisions based on the confession of the offender. 2 The following explanation of judicial procedure is based on Shiga 1974. For the original Japanese version, see Shiga 1984: chapter 1. 3 Buoye 1995 gives a detailed analysis of homicide reports in eighteenth century China. 4 “Lesser” cases include disputes over land, debts, marriage, and inheritance. Magistrates’ adjudication given to these “civil” disputes was also part of Qing justice system. For civil justice in the Qing, see Huang 1996.

92

Karasawa

This difference derives from differing expectations as to the role of the judge. “Adjudication” in the Western sense, according to Shiga, is simply a judgment or estimation, made by a chosen person or body of persons, which, once delivered, is accepted as a substitute for the truth itself. On the assumption that the truth itself can be grasped only by God, [Western] “adjudication” is intended to satisfy the social need to know the truth on a matter requiring prompt and adequate handling. . . . [T]he said judgment must be seen to come as close to the truth as human ability permits. In contrast, in Qing China, the task of the judge (usually the magistrate) was to expose the truth itself. A judge in the Qing had to induce a defendant to tell the truth using any means, including trickery or torture, on the grounds that the person who committed the crime knows best what s/he did. Therefore, in Qing criminal procedure proof of guilt always had to include the criminal’s written confession. “No one could be pronounced guilty of a crime he did not confess.”5 Since guilt had to be established by a confession, the record of testimony constituted the most crucial part of the case report. Guidelines for producing written records of oral testimony were laid out in A Complete Book on Happiness and Benevolence (Fuhui quanshu), a handbook for local magistrates compiled in the seventeenth century. Such testimony records consisted of two kinds: the “original deposition” (gongzhuang or caogong), and the “formal testimony” (zhaozhuang). The first is the record of what the parties and witnesses said during each interrogation. After several stages of interrogation, a magistrate would ascertain to his own satisfaction what had actually happened. The final draft, or “formal testimony,” was composed to sum up these conclusions in the principals’ own words. This formal testimony was read aloud in front of the criminal, who was expected to sign it (usually by a cross mark for men or a circle for women) to signify a guilty plea.6 Witnesses’ statements were transcribed by scribes at the government office at each interrogation. “Original depositions” were the products of this stage. 5 Shiga 1975: 121–23. Citations come from 122. Also see Shiga 1986. 6 Shiga 1975: 120–1. Shiga’s original work in Japanese does not make a distinction between gongzhuang and caogong. See Shiga 1984: 69. However, the English version translates gongzhuang as “original deposition,” while the caogong is translated as “draft deposition.” I verified with actual records at the Number One Historical Archive in Beijing that original depositions were also called caogong. In this paper, therefore, I do not make distinction between gongzhuang and caogong.

From Oral Testimony to Written Records in Qing Legal Cases

93

Then the draft testimony of each party was further composed either by scribes under the guidance of the magistrate’s private secretaries or by the secretaries themselves.7 These intermediate drafts are called “draft depositions” in this chapter. After revising these drafts, the secretaries crafted the formal testimony version to be used for the case report reviewed by superiors. The prospect of review encouraged private secretaries to pay attention to crucial requirements in composing testimony, such as the appropriate order of confessions, the consistency of events recounted in confessions, and skillful construction of the narrative plot or story. These conditions imposed by Qing judicial principles and procedure helped shape the social and cultural setting in which testimony was composed.

Writing Legal Testimony in the Context of Literary Culture

Written testimony was full of administrative terms and legal clichés, but reading it as a technical language would fail to reveal the cultural context that also shaped it. The magistrates’ private secretaries who composed testimony lived within the world of written culture that also produced examination and other literary essays and vernacular fiction. A number of guidebooks for legal private secretaries illuminate shared concepts of good literary style in Qing China. One of these, Important Points for Handling Cases (Ban’an yaolüe), a technical guide for legal specialists written in the eighteenth century, includes an essay titled “How to Compose Testimony” (Xugong) that may be taken as representative. It offered guidelines under six headings: 1) putting events in order (qianhou cengci); 2) logical development (qichengzhuanghe); 3) foreshadowing and correspondence between events (maifu zhaoying); 4) highlighting the key points and making a transition (dianti guomai); 5) summarizing points and adding supplementary accounts (xiaona buwo); and 6) writing with skill and laying out the story ( yunbi buju).8 7 See Reed 2000 for a detailed explanation of clerks’ activities during the Qing. See Ch’ü 1988 [1962]: 93–115 on the crucial role played by private secretaries in local administration. 8 Wang Youhuai, Ban’an yaolüe, pp. 501–6. Other similar handbooks for private secretaries include the following titles: Essentials for Handling Homicide Cases (Ming’an yaolüe); On Sending a Copy of a Confession to Higher Authorities (Zhaojie shuo); Mu Han, My Limited Opinions on How to Clarify the Law (Mingxing guanjianlu); Bai Yuanfeng, Required Reading for Convening Courtrooms (Qintang bidu); Wang Youhuai, Indispensable Readings for Legal and Taxation Secretaries (Xingqian bilan); Wan Weihan, Guidelines for Secretaries on Law and Taxation (Xingqian zhinan); Gang Yi, Proposing Formulas for Investigating Cases (Shenkan nishi); Judicial Important Points (Yuanshu lunyao); and Essentials for Legal Secretaries (Xingmu yaolüe).

94

Karasawa

“Putting events in order” requires careful attention to the sequence of each party’s testimony, to show clearly how an event developed and let the reviewing official grasp the point of the case. “Logical development” is about how to make all aspects of the description of an event consistent. This idea of consistency is encapsulated in three terms: qing, xing, and jie.9 Qing indicates both the facts of the case and people’s states of mind. Xing refers to the physical circumstances involved. Jie means the plot or the details of the story of the case. According to “How to Compose Testimony,” every case must have its own facts (qingjie),10 while each party must have his/her own circumstances (qingxing). The reason the two sides in the case came into conflict constitutes the facts (qing) of the case. The explanation of how the case developed forms the plot or the details ( jie) of the story. The description of what each party thought and wished on the day of the crime explains the state of mind (qing) of the parties. The narration of how the parties got involved manifests the physical circumstances (xing) of each party. It is important that the plot (qingjie) of the case mirror the respective states of mind and physical circumstances (qingxing) of the parties involved. In other words, the narration of the plot of the case must logically organize (canhe) the different states of mind and physical circumstances of each party. The choice of elements used in narrating the testimony from beginning to end is made through the process of logical organization. “Foreshadowing” (in guideline 3 above) warns against a sudden appearance of anybody in the testimony without prior identification to prepare the reader. The “correspondence of events” enjoins that such things as the words used to describe the event, dates and places, the location of injuries on a corpse, and the narration of circumstances should correspond in each witness’ testimony without contradiction. However, since everybody involved has a different position and status, composing testimony also requires making distinctions in the order of events in each party’s narration, in the language used, and in the manner of speech of different individuals. The fourth technique, “highlighting the key points,” is a skill of drawing attention to the core of the text. For example, in the description of a corpse, the phrase “it is determined that the cause of death was . . .” will lead to a description of the injuries. In a murder confession the criminal’s description of his act should be followed by a statement of whether there was intent to kill. In “making a transition” the writer must make clear sequences such as how 9 10

Most of the available guides explain consistency in testimony by using these three terms or some combination thereof. Depending on context, qingjie can mean either of two distinct English words: “facts” and “plot.”

From Oral Testimony to Written Records in Qing Legal Cases

95

that event turned into this event, how an escaped criminal was caught and interrogated, and how the case first handled by one magistrate was taken over by his successor. “Summarizing points” (guideline 5) means condensing insignificant matters into a single phrase of only a few words. “Adding supplementary accounts” suggests that any doubtful points should be explained further at appropriate places within the record of testimony. “Writing with skill” and “laying out the story” (guideline 6) are the keys to making a case report the kind of literary work that will satisfy reviewers. The language is clear and fluent. Not a single word is superfluous or rambling, and every sentence reveals a vigorous style. Testimony written in this way is well organized and coherent. An expert at “laying out the story” can keep in mind all aspects of the case; he can gather the threads of the various parties’ testimonies in a consistent manner so that the narrative is concise, with no loose ends, yet complete, omitting nothing important. Summing up, the author of “How to Compose Testimony” insists that these guidelines are identical with those for composing literary essays. “This whole process is similar to that of reading and writing a literary essay, in that you need to ponder what is the main theme of the essay, which passage contains the main message, and which specific words convey its essence.”11 This advice in fact contained many points of similarity with the guidelines for writing examination essays that had to be mastered by every candidate for a civil service position. Authors of guidelines for composing testimony were well aware of this. “How To Compose Testimony” begins: “Composing [literary] essays means writing meaningful opinions on behalf of the sages; composing a record of courtroom testimony means expressing views on behalf of mediocre plebeians” (zuowen dai shengxian liyan xugong dai yongsu dayi).12 “Literary essays” here means examination essays. This introductory sentence in elegant parallel prose style comparing the two types of writing lent prestige to the court scribe’s work. Tang Biao, author of two guidebooks for examination candidates, offered students advice closely echoing that in “How to Compose Testimony.” A good examination essay should crystallize its theme in one summation.13 Tang’s “method of forward projection” ( yufu fa), like “foreshadowing” in “How to Compose Testimony,” recommended the use of foreshadowing to provide textual connections (mailuo) linking each part.14 In addition, Tang’s handbook 11 12 13 14

Wang Youhuai, Ban’an yaolüe, 501. Wang Youhuai, Ban’an yaolüe, 501. Tang Biao, Fushi Shanyoufa, 26b. Tang Biao, Dushu zuowenpu, 7.91.

96

Karasawa

discusses other methods such as “logical development,” “highlighting the key points,” and “making transition,” all of which emphasize interconnections within a text.15 To highlight the importance of these integrative devices, Qing treatises on writing examination essays frequently used the metaphor of an “artery” (mai) to represent linkages within a text.16 For Tang Biao, the method of “correspondence of events” was particularly crucial.17 These points of similarity between the desired literary style of examination essays and that of records of court testimony would not surprise most legal secretaries, who in practice were likely to be people who had repeatedly failed to pass the examinations, becoming technical legal experts as a second-choice occupation.18 The aforementioned advice was meant for writing composed in classical Chinese. However, vernacular fiction also supplied models for legal composition, and it also was prized for criteria of internal consistency and integration of parts of a narrative into a thematic web. Qing champions of vernacular fiction even claimed their genre equaled the examination essay in sophistication of literary technique.19 By showing that vernacular fiction could be analyzed and appreciated according to criteria developed for the prestigious essay, fiction critics argued for vernacular fiction as serious literature. Traditional Chinese fiction criticism paid particular attention to the internal consistency within a text, employing such concepts as lianguan and guanchuan (both mean “internal consistency”). Fiction critics praised a work if each of its parts (or units of narrative) was systematically orchestrated to interconnect and shape the overall structure.20 Narrative units had to be interwoven to form a “dense fabric.”21 As with the examination essay, the textual inter­connections in a fictional narrative were metaphorically signified by the 15 16

17 18

19 20 21

Tang Biao, Dushu zuowenpu, 8.106, 9.130–9.131. Wang and Zhang eds. 1992: 203–206. Tang Biao, Dushu zuowenpu, 7.91, 9.119. See also Liu Xizai, Yigai (Outline of the arts): “jingyi gai” (outline of the examination essays), 173–176. Liu was a scholar-official in the nineteenth-century. Tang Biao, Dushu zuowenpu, 7.87–7.88. All literate people shared the training for classical essay writing in their elementary level education. Elman (2000) points out that preparation for examinations profoundly influenced the literate mentality in late imperial China. Rolston 1990: 17–29. Plaks 1990: 90–91. Plaks 1990: 92–93, 95. Plaks points out that metaphorical use of fabric and sewing was a major part of Chinese fiction criticism. The popular proverb “tianyi wufeng” (the clothes worn by people in Heaven are seamless) is used to describe a successful effort to “put events in order” (qianhou cengci). See Gang Yi, Shenkan nishi, 7a; Yuanshu lunyao, the third page of “How to Compose Testimony.”

From Oral Testimony to Written Records in Qing Legal Cases

97

term mai (artery). “Logical development” and “foreshadowing” served the goal of integrating the sequence of units. Critics of vernacular fiction also cared about the linkage among narrative units that were located in different parts of the fiction, considering it necessary in helping the reader to grasp the entire structure. The ideal of “interconnection from start to finish” (shouwei xiang­ ying), as Andrew Plaks explains, implied “the ability of the master narrator to arrange his narrative details so as to impart to his text the impression of a tight web of interrelated elements, each of which takes on its full significance only in the total context of all the other narrative units brought together in his coordinated structure.”22 Thus we find that lexicon of six guidelines found in “How to Compose Testimony” borrowed heavily from terminology used by fiction critics.23 “Laying out the story,” “highlighting the key points,” “making a transition,” “putting events in order”—all these techniques ensure that, like a fine work of fiction, the case text maintains a structural coherence as though it were a tightly woven fabric in which narrative threads (xiansuo), running inside like “arteries” (mai), interconnect to function as narrative units, ultimately shaping the entire text.24

Records of Oral Testimony Written in the Vernacular

Written testimony was supposed to be organized along lines closely resembling those of essays in classical Chinese, the written language commonly used for most purposes. Why, then, was the vernacular style of writing employed in composing testimony, instead of the classical style? “How to Compose Testimony” explains the reason: The style of testimony must not be classical. Every single statement must sound like a popular proverb and every single word must be 22 23

24

Plaks 1990: 95–97. The “how to read” genre of fiction commentary (some examples of which are available in English in Rolston 1990) includes a number of terms such as zhaoying, xiansuo, buju, fu, mai, qichengzhuanghe, bi, etc. In guidelines for composing testimony, the idea of narrative thread (xiansuo) provided a skill crucial to achieve a structural integrity. “Once these threads are clear enough to be seen as one thread, all parties’ mouths would be tuned to speak one voice” (ru yixian chuancheng yikou tongsheng). See Gang Yi, Shenkan nishi, 7a; Yuanshu lunyao, the second page of “How to Compose Testimony.”

98

Karasawa

understandable by everyone, so that the testimony can convey the tone of ordinary speech. I have seen testimony which included classical words such as zhi, ji, and er and passages from the [Confucian] Classics. Such testimony gives no sense of vulgar plebeian talk.25 In order to give a written record of testimony the appearance that it had actually been uttered by ordinary people, classical Chinese, which was linguistically distinct from spoken language, could not be employed. After all, “composing a record of courtroom testimony means expressing views on behalf of petty plebeians.”26 Frequent use of popular proverbs in speech, in fact, enabled testimony to have the appearance of vulgar diction. In the famous eighteenth-century vernacular novel The Dream of the Red Chamber (Honglou meng), the speech of uncultured protagonists is full of vulgar proverbs. The author, Cao Xueqin, expressed his dissatisfaction with the artificiality of other novels in which utterances of maidservants were filled with the flavor of classic language.27 Thus the use of proverbs in plebeian testimony came from a common linguistic practice conspicuous in the writings of both self-conscious authors of fiction and skilled legal secretaries. Behind the literary convention was an ideological principle nurtured by examination culture. Classical language expresses the spirit of the sages.28 Therefore the speech of uneducated common people should not be expressed in sagely language, because such people were not supposed to have internalized the sages’ virtue and spirit.29 In sum, employment of the written vernacular for recording testimony was necessary to give testimony the realistic appearance of vulgar plebeian talk. In fact, written vernacular was much closer to what people actually spoke in daily life than the classic Chinese. However, the ideological basis of adopting the written vernacular was rooted in a linguistic ideal that the language of sages and of

25

26 27 28

29

Wang Youhuai, Ban’an yaolüe, 506. This guideline was emphasized by almost all authors of handbooks. Bai Yuanfeng specifically warned not to use literary particles such as hu, zhe, and ye. See his Qintang bidu, 60a. Wang Youhuai, Ban’an yaolüe, 501. Ōta 1988 [1965]: 293. This notion is emphasized in a seventeenth-century manual for education that aimed to train children eventually to pass the civil service examinations. See Cui Xuegu, Shaoxue, 6b. Tang Biao repeatedly stresses this point in his Dushu zuowenpu, juan 7.

From Oral Testimony to Written Records in Qing Legal Cases

99

people who had a grasp of their spirit must be separated from the language of vulgar, uneducated people.30 Let us turn to the linguistic characteristics of testimony produced by the use of the written vernacular. When the oral mode of communication met the written one, the requirement of consistency meant that certain linguistic elements, such as marks of dialects and regional languages, disappeared from official documents. In addition, the record of testimony, though written in the vernacular style, had to avoid coarse use of language. “How to Compose Testimony” emphasizes, Testimony must not be too coarse. Slang expressions, such as profane and insulting remarks and obscene details in adultery and rape cases must not be transcribed verbatim. Those expressions and scenes can be summarized with simple words such as “to abuse and revile” (hunma), “to have illicit sex” (chengjian), and so on. Failure to do such summarizing will provoke the reproach of senior officials [who review the case].31 Testimony thus does not record what people actually said because the primary concern was the review process. The gap between speech and written records of testimony can be seen in surviving “formal testimony” documents from local archives in China. Every example of testimony, for instance, included in the Baxian archives (Baxian dang’an), the Nanbu County archives (Nanbuxian dang’an), both in Sichuan, and the Nanling County archives (Nanlingxian dang’an) in Anhui, is a final draft of composition. This is also the case with the Danshui Subprefecture and Xinzhu County archives (Danshuiting Xinzhuxian dang’an, usually abbreviated as “Dan-Xin dang’an”) in Taiwan. But no single example of written testimony in the Dan-Xin archives from nineteenth century Taiwan is a verbatim record of Hakka or Southern Min languages, which were the major spoken languages on the island. Aside from a few rare examples, we cannot find remnants of those spoken languages in the written records of testimony in Dan-Xin cases.

30

31

Li Tiaoyuan’s Danmo lu introduces an episode about the use of patois in the civil service examination. In 1735 (Yongzheng 13), an examination candidate who participated in the Shaanxi provincial examination wrote a Song dynasty-style edict praising sages (songsheng biaowen) with “Shaanxi patois” (Shaanxi tuyu), with the intention to make fun of Shaanxi native candidates. He received severe punishment for insulting the ceremonial civil service examination. See Danmo lu, 10.153–10.154. Wang Youhuai, Ban’an yaolüe, 506.

100

Karasawa

Clerks, who usually were employed in their native counties, had no problem understanding what local people stated in their common mother tongue. But even prior to the stage of composing drafts in the written vernacular, it was necessary to simultaneously translate what people were saying in the courtroom into Mandarin in order to write their utterances down on paper.32 First of all, many words in Hakka and Southern Min have no exact correspondents in Chinese logographs. Second, a scribe who recorded confessions at a county office might have chosen either of two approaches: he could record what his colleague orally translated into Mandarin for magistrates who understood neither Hakka nor Southern Min, or he could translate what he heard stated in these regional languages into Mandarin in his brain at the time of courtroom interrogation (not when he composed the written drafts).33 Testimony composed for official preservation and for senior officials’ review of the case, of course, was composed in the written vernacular. However, given that the vernacular style of written language based on Mandarin was the only available tool to record oral speech, initial recording of what people said during the

32

33



When Li Xiucheng, one of the leaders of Taiping Rebellion, was seized by Zeng Guofang, Zeng’s private secretaries could not understand Li’s local patois (tuyin). They had to have Li himself write his answers to the questions they put to him during interrogation. See Luo 1982: 14. This example indicates that it was difficult for an interrogator to write down what local people said if he did not possess the skills to translate simultaneously between dialects and Mandarin. Shiga 1984: 57. In Zhangzhou Prefecture in Fujian, ordinary people did not understand Mandarin (guanyin) and, thus, clerks at government offices served as translators during interrogation. Clerks presumably altered what people actually said. See Zhangzhou fuzhi, 38.9a. Considering this kind of abuse, Wang Huizu, an eighteenth-century private secretary and later an official, suggested that a newly appointed local magistrate should hire a native child to keep him company and allow the child to speak local patois (tuyin) freely. The magistrate thus would soon begin to understand the local dialect spoken by parties in the courtroom and not have to rely on clerks’ translations. This, however, might not work unless the magistrate was as diligent as Wang Huizu. See his Xuezhi yishuo, 1.269–1.270.  Acquiring a certain fluency in spoken Mandarin was required for officials, although such acquisition was not always at a level to satisfy the emperors. The Yongzheng emperor, in an edict, mentions his difficulty in understanding the speech of officials from Guangdong and Fujian. A number of textbooks that taught spoken Mandarin were available during the Qing period. Shayizun’s Zhengyin juhua, one of the popular titles, includes a section describing the scene of a trial, indicating that the author assumed officials-to-be would be his main audience. For Yongzheng’s edict, see Qinding xuezheng quanshu, juan 59 (Yongzheng 6), and Daqing shilu, juan 72 (Yongzheng 6).

From Oral Testimony to Written Records in Qing Legal Cases

101

interrogation was also inevitably done with the written Mandarin vernacular, thereby translating from one regional dialect to Mandarin in the scribe’s brain. In addition, local expressions from certain dialects or regional languages were a target of censorship in composing testimony. As drafts were composed, dialect elements might be eliminated. A suggestion from a Qing manual for private secretaries recommends, In testimony, although it is permissible to use vulgar language, it ought to be comprehensible to everybody. [Annotation:] Dialect ( fangyu), patois (tuyu), and overly colloquial phrases (shifen cunsu zhi yu) must not be used. Only if a certain patois [phrase] is available [to express the situation] and cannot be deleted, a pertinent annotation must be added in the testimony so that the reader can understand what that phrase means.34 The fact that actual records of testimony in the Dan-Xin archive rarely reveal vestiges of dialect indicates that this suggestion was indeed followed by scribes and secretaries.35 Although technical barriers and censorial requirements interfered with the use of “patois” utterances, occasional remnants of dialect words can still be detected in Dan-Xin case records.36 However, examples of “patois” are highly exceptional, and most of the vocabulary in written records of testimony is Mandarin. How, then, did scribes in the county offices learn Mandarin vocabulary? Aside from the tutelage of apprenticeship in the county offices itself, vernacular fiction was a logical model for representing dialogue in writing.37 34 “Ban’an” (Handling cases), in Xingmu yaolüe, 1a. Bai Yuanfeng explicitly states that records of testimony should only use Mandarin. See his Qintang bidu, 61a. 35 Homicide reports to the upper authorities also had to avoid using dialect when referring to the fatal injuries on a body. See Zhang Wuwei, Jiangqiu gongji lu, 46a. This is a handbook left by a prefect of the early nineteenth century. 36 Dan-Xin dang’an 35506 leaves some Hakka vocabulary in the record of testimony. Hakka vocabulary in Xunzhou and Wuzhou (Guangxi) also appears in the writings of Li Xiucheng, one of the leaders of the Taiping Rebellion, most of whom came from Guangxi. See Luo 1982: 21–23, 45–47. 37 The courtroom scene in Zhengyin juhua, a Mandarin textbook, reads like a drama script: the story goes on with a successive pattern in which the magistrate questions and the witnesses answer. This vivid depiction not only provided officials-to-be with a useful example of what kind of language magistrates should use in the courtroom, but also supplied scribes at county offices with a tool to document courtroom records in Mandarin vernacular. It is possible that clerks at county offices also could learn Mandarin through these textbooks. See Shayizun, Zhengyin juhua, 75a–86a.

102

Karasawa

The occupation of clerks served to absorb both semiliterate people with a limited ability to read and write as well as fully literate people who had failed the civil service examinations. Clerks in county offices in any region of China during the Qing period were exposed to Mandarin-based written vernacular through the reading of fiction. As Patrick Hanan mentions in his pioneering work on Chinese vernacular stories, as popular fiction spread in late imperial China, its vernacular language became gradually standardized to reach an empire-wide audience, including readers whose mother tongues were not Mandarin. Thus although the written vernacular had its linguistic basis (grammar and syntax) in Beijing dialect, the so-called “normative power of genre” worked to eliminate locally flavored vocabulary, including that used only in the region of Beijing.38 The normative power of genre also operated in the writing down of testimony in legal case records. The presumed readers of these records—senior imperial officials— came from different places throughout China. Written records of testimony had to use a standardized vocabulary in common use to be readily accessible to officials of various regional backgrounds. Scribes and probably also the private secretaries supervising them were provided chances to acquire common vocabulary and Mandarin syntax through reading vernacular fiction aimed at such semi- and/or fully literate people throughout China. Using such lexicon in common use, these scribes and private secretaries could then exercise censorship over how the first draft record ought to be written and what vocabulary should be employed in successive drafts. The normative power over the choice of language—what Jack Goody would call one of the “consequences of literacy”—became available to exert pressure on testimony.39 In sum, records of testimony, though sometimes revealing “patois” vestiges that can be transcribed with Chinese logographs, are filled with Mandarin lexicon. The bulk of any given case record was written in the classical style with which any official would be familiar. But the complicated multilingual 38 39

Hanan 1981: 8. Goody and Watt 1968; Goody 1977. Goody describes “consequences of literacy” as something positive that enables society to develop civilization. Rawski, in her seminal work on “functional literacy” in Qing China, seems to share Goody’s perspective. See Rawski 1979. However, as my use of the term “censorship” shows, I see writing as assisting in the ideological reinforcement of domination. While Goody praises the spirit of criticism and cumulative knowledge enabled by scrutinizing writing, I would like to emphasize the ideological use of the scrutiny of written language. Goody’s later important works do not alter his original idea on “consequences of literacy”. See Goody 1986; 1987. For a criticism of Goody from this perspective, see Street 1984.

From Oral Testimony to Written Records in Qing Legal Cases

103

condition of Qing China made it necessary that even testimony recorded in spoken style be written in another common written language to provide for efficient understanding by officials from different parts of China.

Composing Testimony at the Local Level: A “Directly Examined” Case from Beijing

The written records of testimony included in local-level case records heretofore available (such as Dan-Xin archives, Baxian archives, Nanbu County archives, Nanling County archives, Baodi County archives, and Taihuting archives) are all formal testimony. None of these archives appear to have preserved the original depositions, which were the primary records of confessions. The cases that were directly examined by the Board of Punishments in Beijing are an exception to this pattern. In order to maintain the security of the imperial capital, these local Beijing cases were investigated either by the Warden’s Offices of the Five Wards Dividing the Imperial Capital (Wucheng bingmasi) or by the Offices of General-in-chief Commander of Infantry and Five Police Brigades in the Imperial Capital (Tidu jiumen bujun xunbu wuying tongling yamen). If the suggested sentence included or exceeded temporary banishment, the case was sent directly to the Board of Punishments for further investigation and trial. A number of directly examined cases preserve records of original depositions, revised drafts of testimony in official format, and the final draft of testimony for judgment. Through these cases it is possible to trace exactly how private secretaries and scribes fashioned their written narratives out of what the parties in a case said at different stages of interrogation. The process is best illuminated by following one case from beginning to end over a nine-month period in 1905.40 On the eighteenth day of the second month of Guangxu 31 (1905), Wang Guocai (aged forty-five) filed a report in the Warden’s Offices of the Five Wards located in Beijing, saying that his eldest son, Wang Qunzi (aged twelve), had been found dead on the previous day. On the twenty-ninth day of the same month, the murderer, Wang Tiegun (aged nineteen), was arrested based on another report from Wang Guocai, in which

40

Xingbu dang’an, Sichuansi (Sichuan Bureau) section—(abbreviated as XBDA/SC hereafter) 19961. Citations give the case file’s serial number. A bureau with a prefix of one province name usually worked on examining felony cases that occurred in that particular province. Only a “directly examined” case forwarded to the Board, after each bureau drew a lot, was assigned to one of these bureaus. See Shiga 1984: 58.

104

Karasawa

he revealed that he had heard that Tiegun had been reproached by his mother, Mistress Wang Yang (aged forty-six), for his murder of Wang Qunzi.41 According to Tiegun’s final formal testimony confessing to the crime, he and Wang Guocai’s wife (that is, Qunzi’s mother), Mistress Wang Yu (aged fortyfive), had committed adultery. Their relationship was discovered by the boy Qunzi during the first month of the same year, and both of them were afraid that Qunzi would tell his father. They decided to kill Qunzi and carried out the premeditated murder on the fourteenth day of the second month. Although at first Mistress Wang Yu denied both adultery with Tiegun and conspiring in her son’s murder, under interrogation she finally admitted that she had planned the murder of her son and had cooperated with Wang Tiegun to carry it out. Tiegun was sentenced to death by strangulation; the sentence of Mistress Wang Yu, which originally had been strangulation, was reduced to life imprisonment by the court assizes (chaoshen) of the same year.42 The first round of interrogations of Wang Tiegun was carried out over several weeks between the tenth day of the third month and the ninth day of the fourth month. Original depositions were drawn up on the basis of these sessions. Two preliminary drafts of Tiegun’s testimony were written based on these. The second round of interrogations of Tiegun took place between the eighteenth day of the fourth month and the twelfth day of the fifth month. The third and final draft, Tiegun’s formal testimony, was dated the twenty-eighth of the fifth month. This draft was copied verbatim on the fourteenth and again on the twenty-second day of the sixth month, probably because they were produced for delivery to different officials. The first round of interrogations of Mistress Wang Yu was conducted between the ninth day of the third month and the twelfth day of the fourth month. An original deposition based on her cumulative testimony (which covered everything she had confessed up to that point) was written some time in the fifth month. This first draft was revised to produce a second draft. Between the twenty-eighth day of the fourth month and the twenty-ninth day of the fifth month a second round of interrogations of Mistress Wang Yu was carried out. The final (third) draft, her formal testimony, was dated the twentyninth of the fifth month, and included her affirmation that all she described on that date was true. Two verbatim copies of this formal testimony dated the fourteenth and the twenty-second days of the sixth month joined that of Wang Tiegun in the dossier. 41 42

Although all parties in this case were surnamed Wang, this does not mean they belonged to the same family, rather, it is because they came from a single-lineage community. XBDA/SC 19961.

From Oral Testimony to Written Records in Qing Legal Cases

105

But there is another set of records of testimony from Mistress Wang Yu in which she did not admit guilt. In one piece of this testimony, dated the twentyninth of the fourth month, she simply recounted how her son had disappeared and how the investigation had ensued, giving the impression that she had nothing to do with her son’s death. In another piece of independent testimony dated the first of the fifth month, she denied adultery and conspiracy to murder. Finally, there is testimony from her, based on a rough draft dated the same day, in which she admitted to adultery with Wang Tiegun. However, a month later (the twenty-ninth of the fifth month), Mistress Wang Yu signed a final “formal testimony” document admitting guilt and declaring her earlier statements had been false. This record shows that the stories told by the two principal offenders, Wang Tiegun and Mistress Wang Yu, differed subtly throughout their interrogations. But differences explicit in the original depositions were completely expurgated from the formal testimony records. This shows in a small matter, the date when Wang Qunzi discovered their adultery. Neither of the lovers was able to give a specific date. “I do not remember which day that was” (buji riqi) is usually used in their depositions. However, in both their formal testimonies, the specific date “the 25th day of the first month of this year” suddenly appears. Second, and more important, throughout their interrogations, the two lovers did not agree about who was responsible for initiating the plot. The conclusion of both formal testimonies was that Mistress Wang Yu was the instigator. But earlier drafts show each of the two denying responsibility. On her deposition recorded on the first day of the fifth month, Mistress Wang Yu claimed she tried to protect Qunzi from her lover’s threats. Tiegun and I continued having illicit sex. We were accidentally discovered by Qunzi. It was the [blank] day of the first month. Tiegun threatened Qunzi. I told Qunzi, “You don’t have to be frightened. I am here [to protect you].” From then on, Tiegun frequently asked me to help plan to kill Qunzi.43 A report from ten days later states that she again recounted the scene to emphasize that it was Wang Tiegun who had first proposed the murder. Discovering Tiegun lying in my room, Qunzi told Tiegun, “I will tell my dad.” I told Qunzi, “Tiegun is just sleeping in my room. Don’t get nervous. Nothing else is going on.” Tiegun told me that [if Qunzi told my husband 43

XBDA/SC 19961, deposition of Mistress Wang Yu (first day of the fifth month, Guangxu 31).

106

Karasawa

his life would be lost]. “Let’s kill Qunzi.” I told him that “it is impossible for Qunzi to tell [his father].” Thereafter, Tiegun still tried to continue illicit sex, but I refused to do so. Tiegun told me, “I am still afraid that Qunzi will tell [his father]. I will kill Qunzi.” I replied, “If you would like to continue having illicit sex (xujian) with me, please kill Qunzi.”44 On the next day (the twelfth), she continued her testimony. Tiegun came to my house declaring that still he would like to kill Qunzi. I didn’t agree at first. [But] Tiegun didn’t agree [with me on letting Qunzi live]. I said, “If you want to kill [Qunzi], just do it.” He asked me to give him a rope. I gave him two ropes. It was the evening of the 14th day. On the morning of the 15th day, after I told Qunzi to [go out to] buy noodles, Tiegun came to my house again and said, “I will kill Qunzi today. The idea [of murder] can be attributed to me alone.” I also agreed with him. He told me to go with him to watch [whether the scene of murder was being watched by others]. I didn’t see Qunzi. When Tiegun followed Qunzi, I came home and combed my hair.45 In contrast, Wang Tiegun, at his interrogation on the eighteenth day of the fourth month, blamed his mistress. I continued illicit sex with Mistress Wang Yu as usual. One night I went out for a pee and was discovered by Wang Qunzi. When I entered the house, I forgot to lock the door. I went back to bed with Wang Yu. Wang Qunzi opened the door and came in and discovered [us]. He said he would tell Wang Guocai. Since Wang Yu was very afraid that Wang Guocai would find out about [her adultery], she proposed the idea of killing Wang Qunzi to prevent his telling. Originally she tried to kill [Qunzi] at home. Wang Yu bought two baked cakes and put poison in them. . . . But because she was afraid that Qunzi would vomit, she didn’t dare to have Qunzi eat them. In the evening of the 14th day of the second month, she told me,

44

45

XBDA/SC 19961, deposition of Mistress Wang Yu (eleventh day of the fifth month, Guangxu 31). The statement in parentheses was probably added by clerks (or any unknown superiors) as a “correction” in the margin of the document. XBDA/SC 19961, deposition of Mistress Wang Yu (twelfth day of the fifth month, Guangxu 31).

From Oral Testimony to Written Records in Qing Legal Cases

107

“Wang Guocai will be back pretty soon. It would be better to kill [Qunzi] as soon as possible.”46 On the twenty-fourth day of the fourth month, Wang Tiegun repeated the same story, again emphasizing his own hesitation to commit murder. Wang Qunzi told me that he would tell Wang Guocai when Guocai came back. Therefore, Mistress Wang Yu decided to kill [Qunzi] and told me to carry out the murder. I didn’t agree at first. Wang Yu told me to take Qunzi to the mountain when I carried coal on my back and push him over the cliff to die. I didn’t agree.47 In his last deposition, dated the twelfth day of the fifth month, Wang Tiegun reconstructed the conversation between him and Mistress Wang Yu on the eve of the murders as follows: Because Qunzi discovered our illicit sex accidentally and Wang Guocai also was coming back soon, we were afraid that Qunzi would tell [Wang Guocai]. Wang Yu often consulted with me about how to kill Qunzi. [But] we hadn’t yet committed murder. At that time, if we had delayed the murder any more, Qunzi would have tipped off [Wang Guocai] soon. . . . Wang Yu brought up her earlier adultery with Wang Fuling again [when her husband had beat her seriously]. She told me, “Since discovering our adultery, Qunzi has lost appetite and has not made any response to what I said. He just wishes to tell [his father] about our relationship.” Wang Yu started to want me to kill Qunzi. I didn’t agree at first. [But] I finally agreed [to kill Qunzi] when she said that Qunzi was her son and, therefore, she would be responsible for everything.48 Using these documents we have no way to determine which of the lovers actually first proposed the murder or who took the lead in planning it. The Board’s final case report on this case was lost from the archives, making it impossible to learn how the Board weighed the conflicting evidence on this important issue. Both a deposition and a formal testimony from the victim’s 46 47 48

XBDA/SC 19961, deposition of Wang Tiegun (eighteenth day of the fourth month, Guangxu 31). XBDA/SC 19961, deposition of Wang Tiegun (twenty-fourth day of the fourth month, Guangxu 31). XBDA/SC 19961, deposition of Wang Tiegun (twelfth day of the fifth month, Guangxu 31).

108

Karasawa

father, Guocai, survive, in which he blamed his wife for being responsible for the murder of their son.49 For whatever reason, a version closer to what Wang Tiegun had confessed was adopted as the official one. The conclusive formal testimony of Wang Tiegun (dated twenty-eighth of the fifth month) says, On the 25th day of the first month of this year, I came out of her house to have a pee at sunrise. I forgot to lock the door when I went back to her room. I still lay relaxing under the quilt. Suddenly Wang Qunzi came in and discovered [us]. He declared that he would inform his father [of our illicit sex]. I immediately got up and ran away. The next day I went to Mistress Wang Yu’s house. She told me about her earlier illicit sex with Wang Fuling. She expressed her fear that if Wang Qunzi informed Wang Guocai [of our adultery], our lives would be lost. Therefore, she decided to slay Wang Qunzi to prevent him from informing and consulted with me. She asked me to abduct Wang Qunzi and put him in a box for carrying coal and throw him into the Hun River when I passed it. Before I consented to this idea, she also told me another idea—that she was going to put the powder of a poisonous insect in a cake and feed it to Wang Qunzi at home. I feared that after eating it Wang Qunzi might vomit from the poison. Neighbors would hear [his vomiting]. Thus, we didn’t dare try this idea. On the evening of the 14th day of the second month, I went to her house again to have illicit sex with her and stayed there. Wang Yu told me, “Wang Qunzi, since discovering our illicit sex, has always felt anger in his heart, he refuses to answer me, and he shows no appetite. The day after tomorrow Wang Guocai is coming back. If he discovers [our illicit sex], he will tell your father. Both you and I will be ruined. It would be better to come up with a plan to kill Wang Qunzi to avoid a disaster.” I asked her how we should kill him. Wang Yu consulted with me and decided that the next morning I would let my younger brother, Wang Tiechui, go home from my hut in Wopeng to have breakfast. She would give Qunzi one foreign yuan at sunrise and tell him to go to Wulitun to buy noodles. He would definitely pass by Wopeng. When Qunzi passed by, I would block him [on the path] and force him to eat the poisoned cake. If Qunzi ate it and lost conscious to die, I would tie up his corpse with a rope prepared on the way and carry him on my back to a remote place to dispose of it. If he wouldn’t eat it, I would strangle him to death with the prepared rope. 49

XBDA/SC 19961, deposition of Wang Guocai (twenty-fourth day of the fourth month, Guangxu 31); formal testimony of Wang Guocai (fourteenth day of the sixth month, Guangxu 31).

From Oral Testimony to Written Records in Qing Legal Cases

109

Hearing what she planned, I still hesitated. Mistress Wang Yu also said that Wang Qunzi was her son whom she had given birth to and she alone would take full responsibility if the murder were discovered. She told me not to be afraid. I was muddled temporarily (yishi hutu) and agreed to her plan.50 This is exactly the story that is told in Mistress Wang Yu’s formal testimony dated the twenty-ninth day of the fifth month. The fact that in some parts of the two “final testimonies” not only the content of the story but also the wording are identical means that the person(s) responsible for composing the written record of testimony did not follow some specific instructions given in the essay “How to Compose Testimony.” The essay states, [S]ince everybody has a different position and status, composing testimony requires exercising the practice of imaginative identification with her or him (sheshen chudi) in order to give the impression that s/he is actually speaking. If the order of events varies in each party’s narration, the language used and the manner of speech must be different from each other. Only a skilled composer of testimony can organize testimony in which each party speaks in a different manner without creating any contradictions with each other.51 Clearly the composer of testimony in this case failed to “give an impression that s/he is actually speaking.” However, at the same time, the writer loyally employed the other part of the method of “correspondence of events” in the sense that all details of the stories told by the witnesses correspond with each other. Following another method explicitly suggested by the essay “How to Compose Testimony,” sexually graphic details in this case were indicated by the single word jian (illicit sex). Sexual acts were always referred to this way, not only in formal testimony, but also in all depositions given by Mistress Wang Yu and Wang Tiegun.52 We have no clues about the exact language they used 50 51

52

XBDA/SC 19961, formal testimony of Wang Tiegun (twenty-eighth day of the fifth month, Guangxu 31). Wang Youhuai, Ban’an yaolüe, 504. Jin Shengtan, a literary critic of the seventeenth century, insisted that protagonists in fictional stories should employ manners of speech suitable to their character and social condition. See Jin 1985 [ca. 1641]: 10. See Sommer 2000 on the boundary surrounding jian drawn by the Qing state and its ideological implications.

110

Karasawa

when they described their relation. Here we see that even the original depositions were subjected to standardization, eliminating vulgarisms that ordinary people might have used. Taken together with the standardization integral to literacy that mandated the employment of the written vernacular, these conventions helped blur local dialects and everyday idioms in records of testimony, making all language used in testimony throughout the empire look alike. The belief that the crafts of writing vernacular fiction, successful examination essays, and records of testimony shared common ground shaped the cultural setting that conditioned the composition of testimony during the Qing period. All literate people shared the same theory and practice of writing, whether they used the literary or the vernacular style. Although in theory writing depositions was supposed to imitate actual speech, in practice speech was transformed by linguistic manipulation to achieve consistency—an integral part of literacy. The cultural setting was characterized by the social domination of oral culture by written culture. Conclusion In the Qing legal tradition, what common people directly said did not primarily make up their “formal testimony.” What commoners spoke was to be scrutinized to satisfy the literate culture’s requirement of consistency. The state, which was the largest institution benefited by the ideological use of literacy, also imposed the requirement of consistency in order to satisfy legal categories. What commoners said should be transformed into written drafts with an emphasis on consistency in details of dates, action, and motivation of different parties. The original depositions of parties that exhibited internal contradictions, major and minor, were unified to produce a single version of the story.53

53

Ginzburg 1983 [1966] describes the dynamic process by which an agricultural cult rooted in peasants’ oral traditions was transformed into witchcraft by the force of cultural hegemony, the inquisitors’ perception, in sixteenth- and seventeenth-century Italy. Sabean 2001 analyzes the narrative structure of bureaucratic documents in early modern Germany that include records of courtroom interrogation, concluding that the bureaucratic narrative produced a single explanation of what happened. In the European field, historians have been successful in reconstructing the mental universe of oral culture by paying attention to the profound gap between interrogators and peasants—the gap that subtly appears in the inquisition records. See Ginzburg 1980 [1976]; Ginzburg 1989; Sabean 1984. On the early use of confession records of those who participated in religious rebellions during the Qing, see Naquin 1976; Perry 1980.

From Oral Testimony to Written Records in Qing Legal Cases

111

The requirement of consistency and the rejection of contradictions, first of all, derived from the Chinese legal tradition, in which the credibility of testimony played the most crucial role in establishing the truth of a case. Consistency in the written record of oral testimony was also derived from standards that were integral to literacy. By considering the interaction between oral and written modes of communication found in the process of composing testimony, we can locate that testimony in its larger social and cultural context. As Jack Goody points out, one of the “consequences of literacy” was scrutiny.54 For Goody, the ability to scrutinize written materials facilitated the growth of a spirit of criticism. However, he does not address the dominating power of literacy when it encountered oral culture.55 The utterance produced by the oral mode of communication would fade immediately in such an encounter. Transcribing what was spoken gives us access to what was transmitted by oral communication. However, at the same time, the scrutinizing force, which is an integral part of literacy, tends to distort our view of oral culture. The process of composing testimony in the Qing created an interface where the oral tradition met the written one. The spoken language was recorded in the written vernacular language. However, given the technological condition of Chinese languages in the late imperial period, it was impossible to record verbatim what witnesses said in court. Not all spoken vocabulary had corresponding logographs; in addition, spoken languages in south China were quite distinctive from Mandarin. Technological barriers were imposed by the nature of the Chinese language in writing down what was expressed orally. Linguistically standardizing effects (especially on the lexicon) were an inevitable result of the use of the written vernacular. Packaging an event with a simple cliché was a common writing practice for those who composed written records of oral testimony in the Qing. This practice operated to draw a border around what had happened by dismissing details regarded as irrelevant by officials. As I stated earlier, in most parts of final drafts of written testimony not only the content of the story but the wording is also exactly the same. For the sake of establishing legal facts, both the writing practice and the employed style in written testimony served to make the case correspond with the codified category of crime. Molding what commoner defendants said into one uniform story was standard practice for judicial authorities.

54 55

Goody and Watt 1968; Goody 1977. Street 1984; Goody 1986.

112

Karasawa

However, the essay “How to Compose Testimony” states that “[t]he testimony must not be false. If the story has foundation, it will be firm. False words will be absolutely laid bare.”56 This advice means that Qing judicial authorities did not think that they were falsely fabricating stories by fashioning the oral statements of defendants into a single written narrative version. The Qing legal tradition was based on the fundamental presupposition that the “truth” of the case could be discovered in the person who committed the crime. Therefore, if we interpret the textuality found in the process of composing testimony as evidence of purposeful falsification of cases, we will fail to understand the way the Qing state tried to establish its ideological basis. Qing legal procedure was not constructed upon the “falsification” of what people said. Establishing consistent details in written testimony was an inevitable result when speech was converted to the written word. Textuality of recorded testimony, shaped through transforming speech into written narrative, involved the state’s ritualistic use of commoners’ voices to confirm the “truth” that it established. The final draft of written testimony, which in essence describes exactly what people did, tends to render invisible the dominating forces exercised by the written mode of communication over the oral one. Written records of testimony used the writing skills that literate people were familiar with. Those skills included not only the writing patterns of examination essay style, but also those of vernacular fiction, in that specific care was taken with both the form and content of protagonists’ speech. Those who produced testimony records were people who were familiar with the form of the examination essay and with the literary devices and language employed in vernacular fiction. Composing testimony in Qing China meant a sharing of writing practices to reconstruct “reality” that once had been represented orally within the larger written culture.

Character List

ban’an Baodixian Baxian buji riqi buju 56

辦案 寶坻縣 巴縣 不記日期 布局

Wang Youhuai, Ban’an yaolüe, 507.

From Oral Testimony to Written Records in Qing Legal Cases buwo canhe caogong chaoshen chengjian Danshuiting dianti fangyu fu gongzhuang guanchuan guanhua guanyin guomai Honglou meng hu zhe ye hunma jian jie jingyi gai lianguan mai maifu mailuo Nanbuxian Nanlingxian ni qianhou cengci qichengzhuanghe qing qingjie qingxing ru yixian chuancheng yikou tongsheng sheshen chudi shifen cunsu zhi yu shouwei xiangying songsheng biaowen Taihuting tianyi wufeng

補斡 參合 草供 朝審 成姦 淡水廳 點題 方語 伏 供狀 貫串 官話 官音 過脈 紅樓夢 乎者也 混罵 姦 節 經義概 聯貫 脈 埋伏 脈絡 南部縣 南陵縣 擬 前後層次 起承轉合 情 情節 情形 如一線串成異口同聲 設身處地 十分村俗之語 首尾相應 頌聖表文 太湖聽 天衣無縫

113

114

Karasawa

Tidu jiumen bujun xunbu wuying tongling yamen tuyin tuyu Wang Fuling Wang Guocai Wang Qunzi Wang Tiechui Wang Tiegun Mistress Wang-Yang Mistress Wang-Yu Wopeng Wuzhou Wucheng bingmasi Wulitun xiansuo xiaona Xinzhuxian xing Xingbu-xianshen xishi Xugong xujian Xunzhou yamen yishi hutu yufu fa yunbi Zeng Guofan zhaoying zhaozhuang zhi ji er zuowen dai shengxian liyan xugong dai yongsu dayi

提督九門步軍巡捕五營統領衙門 土音 土語 王福齡 王國才 王群子 王鐵錘 王鐵棍 王楊氏 王于氏 窩棚 梧州 五城兵馬司 五里屯 線索 消納 新竹縣 形 刑部現審 細事 敘供 續姦 潯州 衙門 一時糊涂 預伏法 運筆 曾國籓 照應 招狀 之及而 作文代聖賢立言敘供代庸俗達意

References Bai Yuanfeng 白元峰. 1841. Qintang bidu 琴堂必讀 (Required reading for convening courtrooms).

From Oral Testimony to Written Records in Qing Legal Cases

115

Buoye, Thomas. 1995. “Suddenly murderous intent arose: Bureaucratization and benevolence in eighteenth-century Qing homicide reports,” Late Imperial China, 16–2: 62–97. Ch’ü, T’ung-tsu. 1988 [1962]. Local Government in China under the Ch’ing. Cambridge, Mass.: Harvard University Press. Cui Xuegu 崔學古. Ca. Seventeenth century. Shaoxue 少學 (Learning for youths). In Tanji congshu 檀几叢書 (Encyclopedia of Tanji). Ed. Wang Zhuo 王啅, vol. 2: juan 9. Daqing shilu 大清實錄 (Veritable records of the great Qing dynasty). Dan-Xin dang’an 淡新檔案 (Danshui Subprefecture and Xinzhu County archive). Microfilm copy, University of California, Los Angeles. Catalogued by Dai Yanhui 戴炎輝. Elman, Benjamin. 2000. A Cultural History of Civil Examinations in Late Imperial China. Berkeley: University of California Press. Gang Yi 剛毅. 1889. Shenkan nishi 審看擬式 (Proposing formulas for investigating cases). N.p.: Jiangsu shuju. Ginzburg, Carlo. 1983 [1966]. Night Battles: Witchcraft and Agrarian Cults in the Sixteenth and Seventeenth Centuries. Baltimore: Johns Hopkins University Press. ———. 1980 [1976]. The Cheese and the Worms: The Cosmos of a Sixteenth-Century Miller. Baltimore: Johns Hopkins University Press. ———. 1989. “The inquisitor as anthropologist.” In his Clues, Myths, and the Historical Method, pp. 156–164. Baltimore: Johns Hopkins University Press. Goody, Jack. 1977. The Domestication of the Savage Mind. Cambridge: Cambridge University Press. ———. 1986. The Logic of Writing and the Organization of Society. Cambridge: Cambridge University Press. ———. 1987. The Interface between the Written and the Oral. Cambridge: Cambridge University Press. ——— and Ian Watt. 1968. “The consequences of literacy.” In J. Goody, ed., Literacy in Traditional Societies. Cambridge: Cambridge University Press. Hanan, Patrick. 1981. The Chinese Vernacular Story. Cambridge, Mass.: Harvard University Press. Huang Liuhong 黃六鴻. 1694 (date of Preface). Fuhui quanshu 福惠全書 (A complete book concerning happiness and benevolence). Huang, Philip C. C. 1996. Civil Justice in China: Representation and Practice in the Qing. Stanford: Stanford University Press. Jin Shengtan 金聖嘆. Ca. 1641. “Xu san 序三” (The third preface [to The Water Margin]). In Jin Shengtan quanji 金聖嘆全集 (Complete works of Jin Shengtan), 1985, pp. 9–12. Yangzhou: Jiangsu guji chubanshe.

116

Karasawa

Li Tiaoyuan 李調元. 1881. Danmo lu 淡墨錄 (Record of skilled civil service examination papers). In Congshu jicheng. Shanghai: Shangwu yinshuguan, 1939. Liu Xizai 劉熙載. 1873 (date of Preface). Yigai 藝概 (Outline of the arts). Shanghai: Shanghai guji shudian, 1978. Luo Ergang 羅爾綱. 1982. Li Xiucheng zishu yuangao zhu 李秀成自述原稿注 (Annotation to the drafts written by Li Xiucheng). Beijing: Zhonghua shuju. Ming’an yaolüe 命案要略 (Essentials for handling homicide cases). Qing edition. Mu Han 穆翰. 1845. Mingxing guanjianlu 明刑管見錄 (My limited opinions on how to clarify the law). Na Silu 那思陸. 1992. Qingdai zhongyang sifa shenpan zhidu 清代中央司法審判制度 (System of justice at the highest level in Qing China). Taipei: Wenshizhe chubanshe. Naquin, Susan. 1976. “True confessions: Criminal interrogations as sources for Ch’ing history,” National Palace Museum Bulletin, 11–1: 1–17. Ōta Tatsuo 太田辰夫. “ ‘Kōrōmu’ no gengo 紅楼夢の言語 (The language of The Dream of the Red Chamber).” In Ōta Tatsuo, Chūgokugo shi tsūkō 中国語史通考 (The study of the history of Chinese language), 1988: 289–96. Tokyo: Hakuteisha. Perry, Elizabeth. 1980. “When peasants speak: Sources for the study of Chinese rebellions,” Modern China, 6–1: 72–85. Plaks, Andrew. 1990. “Terminology and central concepts.” In Rolston 1990. Qinding xuezheng quanshu 欽定學政全書 (Imperially sponsored collection of writings by education commissioners). Ca. 1773 ed. Rawski, Evelyn. 1979. Education and Popular Literacy in Ch’ing China. Ann Arbor: University of Michigan Press. Reed, Bradly. 2000. Talons and Teeth: County Clerks and Runners in the Qing Dynasty. Stanford: Stanford University Press. Rolston, David. 1990. “Sources of traditional Chinese fiction criticism.” In Rolston 1990: 3–34. ———, ed. 1990. How to Read the Chinese Novel. Princeton: Princeton University Press. Sabean, David. 1984. Power in the Blood: Popular Culture and Village Discourse in Early Modern Germany. Cambridge: Cambridge University Press. ———. 2001. “Peasant voices and bureaucratic texts: Narrative structure in early modern German protocols.” In Peter Becker and William Clark, eds., Little Tools of Knowledge: Historical Essays on Academic and Bureaucratic Practices, pp. 67–93. Ann Arbor: University of Michigan Press. Shayizun 莎彝尊. 1853. Zhengyin juhua 正音咀華 (Appreciating the essence of proper pronunciation). Shiga Shūzō 滋賀秀三. 1974. “Criminal procedure in the Ch’ing dynasty—With emphasis on its administrative character and some allusion to its historical antecedents (I),” Memoirs of the Research Department of the Toyo Bunko, no. 32: 1–45.

From Oral Testimony to Written Records in Qing Legal Cases

117

———. 1975. “Criminal procedure in the Ch’ing dynasty—With emphasis on its administrative character and some allusion to its historical antecedents (II),” Memoirs of the Research Department of the Toyo Bunko, no. 33: 115–138. ———. 1984. Shindai Chūgoku no hō to saiban. 清代中国の法と裁判 (Law and justice in Qing China). Tokyo: Sōbunsha. ———. 1986. “Chūgoku hō bunka no kōsatsu: soshō no arikata o tsūjite 中国法文化 の考察—訴訟のあり方を通じて (The inquiry of Chinese legal culture: through the way of legal proceedings),” Hōtetsugaku nenpō: 37–54. Sommer, Matthew. 2000. Sex, Law, and Society in Late Imperial China. Stanford: Stanford University Press. Street, Brian. 1984. Literacy in Theory and Practice. Cambridge: Cambridge University Press. Tang Biao 唐彪. N.d. Dushu zuowenpu 讀書作文譜 (A handbook for reading classical essays and writing successful examination essays). Taipei: Weiwen chubanshe, 1976. ———. N.d. Fushi Shanyoufa 父師善誘法 (A handbook for fathers and teachers to instruct sons to pass the civil service examinations). In Wuzhong yigui 五種遺規 (Five volumes of left regulations). Ed. Chen Hongmou 陳宏謀. Taipei: Dezhi chubanshe, 1963. Wan Weihan 萬維翰. 1774. Xingqian zhinan 刑錢指南 (Guidelines for secretaries on law and taxation). Wang Huizu 汪輝祖. 1793 (date of Preface). Xuezhi yishuo 學治臆說 (Subjective views on learning governance). In Rumu xuzhi wuzhong 入幕須知五種 (Five volumes of essential knowledge for those entering private secretary work). Ed. Zhang Tingxiang 張廷驤. 1892. N.p.: Zhejiang shuju. Wang Kaifu 王凱符 and Zhang Huien 張會恩, eds. 1992. Zhongguo gudai xiezuoxue 中國古代寫作學 (Scholarship on writing in traditional China). Beijing: Zhongguo renmin daxue chubanshe. Wang Youhuai 王又槐. 1793. Ban’an yaolüe 辦案要略 (Important points for handling cases). In Rumu xuzhi wuzhong (Five volumes of essential knowledge for those entering private secretary work). Ed. Zhang Tingxiang. 1892. N.p.: Zhejiang shuju. ———. 1814. Xingqian bilan 刑錢必覽 (Indispensable readings for legal and taxation secretaries). Xingbu dang’an 刑部檔案 (Board of Punishments Archive), held at the Number One Historical Archives in Beijing. Xingmu yaolüe 刑幕要略 (Essentials for legal secretaries), Qing edition. In Rumu xuzhi wuzhong (Five volumes of essential knowledge for those entering private secretary work). Ed. Zhang Tingxiang. 1892. N.p.: Zhejiang shuju. Yuanshu lunyao 爰書論要 (Judicial important points). Qing manuscript.

118

Karasawa

Zhang Wuwei 張五緯. 1812 (date of Preface). Jiangqiu gongji lu 講求共濟錄 (Record of seeking the public benefit). Zhangzhou fuzhi 漳州府志 (Gazetteer of Zhangzhou Prefecture). 1877. Taipei: Dengwen yinshuaju reprint, 1965. Zhaojie shuo 招解說 (On sending a copy of a confession to higher authorities). Jiaqing manuscript. In Ming Qing gongdu miben wuzhong 明清公牘秘本五種 (Five volumes of secret manuals for making official documents in the Ming and Qing periods). Eds. Guo Chengwei and Tian Tao, 1999. Beijing: Zhongguo zhengfa daxue chubanshe.

chapter 5

Abortion in Late Imperial China: Routine Birth Control or Crisis Intervention?* Matthew H. Sommer Introduction1 Two sets of prominent historians have argued—from very different kinds of evidence, and with very different goals—that abortion was routinely practiced in late imperial China. The first set consists of demographic historians, notably James Z. Lee, Li Bozhong, and Wang Feng.2 They claim that the Chinese demographic regime has long been characterized by systematic birth control, in which abortion played a significant role. Since Chinese people have practiced “rational” family planning for centuries, there was no need for China to * Originally published in Late Imperial China 31, 2 (Dec. 2010): 97–165. Used with permission of The Johns Hopkins University Press. 1 Sincere thanks to all who commented on drafts of this paper, especially Che-chia Chang, Hill Gates, Christopher Isett, Lillian Li, Thomas Mullaney, Arthur Wolf, and Yi-Li Wu; my graduate students Wesley Chaney, Annelise Heinz, Ying Hu, and Yvon Wang; Barbara Welke and other participants in the Hurst Legal History Institute at the University of Wisconsin; my cousin Joan Kreider MD, who shared her professional expertise in gynecology; and two anonymous referees. The following people helped me obtain sources: Ying Hu, Christopher Isett, Karasawa Yasuhiko, Janet Theiss, Xue Zhaohui, Karen Young, and the staff at the First Historical Archive of China and the Sichuan Provincial Archive; the following people helped me interpret difficult texts: Che-chia Chang, Mei-yu Hsieh, Dongfang Shao, and Yi-Li Wu. The research here was funded by the CSCPRC, the American Philosophical Society, Stanford’s Dean of Humanities and Sciences and UPS Endowment, and the Japanese Ministry of Education. 2 The key texts are Lee and Wang, One Quarter of Humanity, and Li Bozhong, “Duotai, biyun, yu jueyu” (Abortion, contraception, and sterilization). Li Bozhong claims that by the Qing, natural fertility had disappeared in Jiangnan, due to widespread and effective abortion and other means of fertility control, and that Jiangnan was representative of wider trends (see discussion below). Lee’s earlier book (with Cameron Campbell), Fate and Fortune in Rural China, does not mention abortion or contraception per se, instead emphasizing marital restraint, infanticide, and neglect as “standard methods of family planning” (99). The addition of abortion and contraception in One Quarter of Humanity is largely based on Li Bozhong’s article, and these scholars’ citations of each other show that they share the same agenda.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_�07

120

sommer

undergo a modern demographic transition away from natural fertility like that experienced by the West; indeed, according to Lee and Wang, it was a simple and seamless process for Chinese people to adapt their own long-standing practice of birth control to the post-Mao state’s new regime of population control.3 The second set consists of gender historians, notably Francesca Bray and Charlotte Furth.4 Their focus is the politics of reproduction within the polygynous elite household; and their claims are far more nuanced and limited in scope than those of the demographers. Specifically, Francesca Bray argues that elite women were able to use abortifacients (euphemized as emmenagogues) to terminate early pregnancies (euphemized as “menstrual blockage”) because regular menstruation was deemed the foundation of women’s health and fertility. This technology of reproductive control enabled elite wives to avoid childbearing and to displace the biological duties of the wife/mother role onto the concubines and maidservants of their households. Despite their many differences, both sets of historians believe that at least some people in late imperial China used abortion for routine birth control. This belief would seem to presuppose that traditional methods of abortion (and in the case of the historical demographers, contraception as well) were safe, effective, and readily available to those who wished to use them; indeed, Li Bozhong makes these claims explicitly.5 If true, these claims would have terrific implications for a host of sub-fields of Chinese history, including demography, gender, and sexuality, but also law and medicine. Moreover, as Lee and Wang suggest, whether and how the Chinese practiced fertility control in the past should inform the way we understand population policy in the PRC today. For all of these reasons, the practice of abortion in late imperial China deserves closer scrutiny. The arguments of both groups of historians are quite appealing, and that appeal enhances their persuasive power. The historical demographers envision people in late imperial China engaging in a definitively modern form of behavior long before Europeans did; their vision is a variation on the theme of Chinese parity with (or even superiority to) the early modern West that has informed the works of William T. Rowe, Kenneth Pomeranz, and others. For their part, the gender historians envision at least some women in late imperial China enjoying a remarkable degree of agency and even autonomy in 3 Lee and Wang, One Quarter of Humanity, 9–10, 133–35. 4 Bray, “A Deathly Disorder” and Technology and Gender; Furth, “Blood, Body, and Gender” and A Flourishing Yin. 5 Li Bozhong, “Duotai, biyun, yu jueyu.”

abortion in late imperial china

121

their reproductive lives, albeit at the cost of exploiting other women. (“Not all women were victims of their wombs. Women could and did act to control their own fertility.”)6 Their vision is a variation on the theme of elite female empowerment and self-realization within the formal structures of Confucian patriarchy that has informed the works of Dorothy Ko, Susan Mann, and others. Many of us would like to believe in both of these visions. But I believe they are wrong, at least with regard to abortion. To avoid misunderstanding, let me stipulate that I agree with past scholarship on several points. I agree that a number of purported methods of abortion were known, the most important by far being abortifacient drugs. For example, as many have observed, the most famous encyclopedia of traditional medicine, Li Shizhen’s sixteenth-century Bencao gangmu (Compendium of Materia Medica), lists a large number of herbal, animal, and mineral substances that supposedly can be used to induce abortion.7 I also agree that people sometimes attempted to induce abortion, and that at least some of these methods worked as intended, at least some of the time. Where I disagree with past scholarship is about the safety, efficacy, and availability of these methods—and, therefore, the practical circumstances in which women actually resorted to them. I believe that traditional abortifacient drugs (the principal means employed) were dangerous and unreliable, and that access to them and their use required specialized knowledge and often a fair amount of money. Under the circumstances, abortion constituted neither an effective means of family planning nor a tool for female empowerment as understood by these scholars. On the contrary, abortion was an emergency intervention in a crisis. Textual sources from the late imperial era generally locate abortion in the context of either a medical crisis, in which pregnancy endangered a woman’s health, or a social crisis, in which pregnancy threatened to expose a woman’s extramarital sexual activity. Abortion in the context of both sorts of crisis can be found in the MingDynasty novel Jinpingmei cihua (Plum in the Golden Vase). In the first episode of abortion (in Chapter 33), Ximen Qing’s wife Wu Yueniang is more than five 6 Bray, Technology and Gender, 276. 7 Li Shizhen, Bencao gangmu xin jiaozhuben (New annotated edition of Compendium of materia medica, hereafter cited as BCGM), 1:264–66, and individual entries throughout the text. For an overview of this text, see Nappi, The Monkey and the Inkpot; for an overview of traditional means of abortion in China, see Maxwell, “On Criminal Abortion in China”; Rigdon, “Abortion Law and Practice in China”; Li Bozhong, “Duotai, biyun, yu jueyu”; and Tian Yanxia and Jiao Peimin, “Zhongguo gudai duotai kaolüe” (Abortion in premodern China); for the spectrum of moral opinion on abortion in China, see Nie Jing-Bao, Behind the Silence.

122

sommer

months pregnant; after wrenching herself in a fall on the stairs, she suffers severe abdominal pain and her health is believed in danger. She consults a female herbal practitioner, who gives her two “large black pills” of unidentified ingredients to take with wine, in order to induce abortion. The abortifacient works—Wu Yueniang expels an intact male fetus—and she recovers from her illness. Given the novel’s karmic logic, this episode serves to link Ximen Qing’s incorrigible promiscuity with his inability to produce an heir: his licentiousness breeds infertility, because of damage to his health but also as retribution for his crimes.8 In the second episode of abortion (in Chapter 85), Ximen Qing’s notorious concubine Pan Jinlian becomes pregnant by an affair with Ximen’s son-in-law Chen Jingji. (By this point in the novel, Ximen has already died from an overdose of aphrodisiac.) Jinlian is anxious to terminate her pregnancy in order to conceal her incestuous adultery, so her lover procures an abortifacient from Dr. Hu, a specialist in women’s medicine. Dr. Hu intuits what Chen is after, but at first he coyly boasts of his skill in promoting fertility and healthy childbirth. When Chen makes clear what he wants, the physician feigns horror: “Nine out of ten come to me seeking medicine to secure a fetus (antai). Why on earth would anyone want to abort a fetus (datai)? I have no such medicine!”9 Only after Chen offers a hefty bribe does Dr. Hu drop his pretense and provide him with an abortifacient compound of guaranteed efficacy.10 Jinlian takes the drug and expels the fetus (again male) into a chamber pot. This abortion too 8 McMahon, Causality and Containment, 102; for an overview of medicine in the novel, see Cullen, “Patients and Healers in Late Imperial China.” 9 Jinpingmei cihua, 85:2b. I follow Yi-Li Wu’s advice in translating antai as “secure” (rather than “pacify”) the fetus, since this intervention’s purpose was to prevent miscarriage (personal communication). 10 Pan Jinlian’s abortifacient includes the striped blister beetle mylabris (banmao), gansui, yuanhua, liquid mercury, ox knee (niuxi), and musk (shexiang) ( Jinpingmei cihua, 85:2b). Several of these substances are extremely toxic and are now restricted by the governments of the PRC, Taiwan, and Hong Kong; if ingested in any quantity this potion might be fatal, and by itemizing its ingredients, the author may intend to convey a sense of horror and black humor. Jinlian suffers no ill effect from the abortion: her karmic retribution arrives later in the form of Wu Song, who kills her. For the Taiwan and PRC lists of controlled materia medica, see Ko, “Causes, Epidemiology, and Clinical Evaluation of Suspected Herbal Poisoning,” tables 1 and 2; for the Hong Kong list, see Schedule 1 of the Chinese Medicine Ordinance (Cap. 549 of the Laws of Hong Kong), Government of Hong Kong, “Bilingual Laws Information System,” http://www.legislation.gov.hk/eng/index .htm. Official efforts to restrict toxic abortifacient herbs (such as gansui and yuanhua) date back to at least the Yuan Dynasty—see Yuan dian zhang (Decrees and regulations of the Yuan Dynasty), 57:39a–40b.

abortion in late imperial china

123

c­ arries an ironic twist: throughout the novel, Jinlian longs to become pregnant by Ximen Qing; but when she does finally become pregnant, it is by Chen Jingji, and she must abort the fetus. The episode of Wu Yueniang’s abortion exemplifies the scenario of medical crisis, when a pregnant woman’s health is believed in jeopardy. (A variation is when a woman has been so exhausted or her health so impaired by multiple pregnancies that she fears carrying another to term.) The medical texts cited by Francesca Bray and Charlotte Furth document therapeutic abortions of this kind. In contrast, Pan Jinlian’s abortion exemplifies the scenario of social crisis, when a woman attempts to terminate a pregnancy in order to prevent the exposure of adultery. This is the scenario found in Qing legal texts, as well as in modern medical reports of unsafe abortion in China. Jinpingmei is fiction; but the two scenarios of medical and social crisis account for most (if not all) documented instances of abortion in late imperial China. Another point stands out: in both episodes, it is necessary to buy the abortifacient drugs from medical practitioners. In other words, the novel portrays abortion as a field of specialized knowledge, rather than some sort of commonplace women’s lore.11 In addition, anthropologists have documented a third context for abortion, through interviews with elderly rural women about life before modern birth control: some women burdened by poverty and physical hardship would try to terminate pregnancies in order to avoid bearing yet more children (which Lee and Wang might call “early stopping”). Fearing the disapproval of husbands and in-laws, they would do this in secret. Because these women feared the health consequences of further childbirth, their behavior bears some resemblance to the accounts of medical crisis found in textual sources. The rest of this paper divides into three sections. First, I evaluate the claims of the demographic and gender historians, with particular attention to the weakness of their qualitative evidence; I suggest that their own evidence actually undermines their arguments. I also review relevant findings by anthropologists. Second, I present new evidence from Qing legal sources that suggests a very different perspective on the questions of safety, reliability, and accessibility of abortion, and on the context in which women actually attempted abortion. Among other things, Qing legal cases show that even women who actively sought the means for abortion were not necessarily able to get them. In the 11

Some historians of the West suggest (controversially) that contraceptive/abortifacient herbs were found in every housewife’s garden and that their use was part of a ubiquitous women’s lore passed from mother to daughter (e.g. Riddle, Eve’s Herbs). This was certainly not the case in China.

124

sommer

final section, I survey medical reports about the persistence of unsafe abortion by traditional means in the modern era, right down to the present day. These medical reports complement the legal cases: both kinds of evidence argue strongly against complacency about traditional methods of abortion in China.

Past Scholarship

The Demographic Historians: Routine Birth Control to Limit Family Size? James Lee and Wang Feng argue that the premodern Chinese fertility regime was characterized by self-conscious, purposeful birth control to restrict fertility through “late starting, early stopping, and long birth intervals.”12 The key to this regime, they say, was “marital restraint” (that is, low coital frequency) that resulted in low marital fertility. But they also point to the role of “a wide variety of traditional reproductive technologies,” including “various herbal medicines for contraception and a wide variety of abortive techniques,”13 as well as infanticide and deliberate neglect to dispose of superfluous children. Indeed, they must claim that birth control was widespread in order to justify their estimated fertility rates, which are far lower than most scholars have argued, given that marriage at an early age was essentially universal for Chinese women. Lee and Wang stake these claims mainly on quantitative data gathered from population records, but their analysis of those data has provoked controversy. As Arthur Wolf, Christopher Isett, and others have demonstrated, Lee and Wang’s analysis is weakened by their obfuscation of the difference between total fertility and total marital fertility;14 their inclusion of the Qing

12 13

14

Lee and Wang, One Quarter of Humanity, 88. Lee and Wang, One Quarter of Humanity, 91; cf. 106: “Chinese couples also had access to traditional contraceptive and abortive technologies”; 139: “Chinese households in the past . . . required that married couples limit their fertility either through sexual restraint or contraception, a policy continued at the state level today. Such collective strategies forced Chinese parents to kill or abort some children and nurture others according to collective economic and social constraints.” Their goal is to make China’s fertility rates appear lower than those in Europe—see Brenner and Isett, “England’s Divergence from China’s Yangzi Delta,” 619; and Isett, State, Peasant, and Merchant in Qing Manchuria, 189–90, 368–69. As Wolf and Engelen explain, “China’s marital fertility may well have been lower than Europe’s, but its total fertility was far higher” because nearly all Chinese women married whereas many European women did not, and because average age at marriage of Chinese women was lower than that of European women (Wolf and Engelen, “Fertility and Fertility Control in Pre-Revolutionary

abortion in late imperial china

125

imperial nobility, who were Manchus (not Han Chinese), and constituted a highly exceptional urban elite; their use of incomplete data on rusticated bannermen, who had strong motives to avoid being counted; and other factors.15 Wolf and Engelen show that according to the best evidence, fertility rates in premodern China actually conform to worldwide norms for natural fertility: This discovery does not prove that the Chinese refrained from deliberate birth control, but it adds greatly to the burden of proof that Lee and Wang must bear. They must either generalize their thesis to include most historical societies with early and nearly universal marriage, or they must give reasons why birth control in China was necessary to achieve a level of fertility produced by involuntary means in many other societies.16 Wolf and Engelen use household registers from Taiwan (the most accurate record of any premodern Chinese population) to test key aspects of Lee and Wang’s argument: The received view of Chinese fertility is that most couples made no effort to control their fertility because they wanted as many sons as possible. It predicts that the more children a woman has borne in the past, the more she will bear in the future because her fertility history indexes her fecundity. Lee and Wang’s view is that since most couples wanted only a limited number of sons, they spaced their births to achieve but not overshoot this mark. Their position predicts that any relationship between past and future performance is bound to be negative. Couples who reproduced too rapidly when young would have made an effort to slow down later, whereas those who reproduced too slowly when young would have made an effort to speed up later.17 The Taiwan data confirm the received view: the general pattern is that couples had as many children as possible, instead of stopping at some desired

15 16 17

China,” 348). Lee and Wang “report total fertility rates for China, but they never ­compare them with British or European total fertility rates. Although they repeatedly contrast Chinese and European marital rates, they never risk comparing their total rates” (Wolf and Engelen, “Fertility and Fertility Control,” 349). Arthur P. Wolf, “Is There Evidence of Birth Control”; Isett, State, Peasant, and Merchant, 190; Wolf and Engelen, “Fertility and Fertility Control.” Wolf and Engelen, “Fertility and Fertility Control,” 366. Wolf and Engelen, “Fertility and Fertility Control,” 362–63.

126

sommer

­ aximum; whatever limit existed was apparently imposed by nature. Wolf and m Engelen conclude that birth control, if practiced at all, did not have significant demographic impact. In this paper, I focus on the qualitative evidence, because any quantitative data from pre-1949 China not backed by robust qualitative evidence should be treated with skepticism. As Wolf comments, “birth control on the scale suggested would be like having an elephant in the living room. There would be evidence of its presence everywhere.”18 But Lee and Wang offer little qualitative evidence. Their method, rather, is to claim that their data prove the practice of birth control, and then to speculate about what forms birth control might have taken. With regard to “marital restraint,” for example, they cite an esoteric textual tradition that advocated sexual moderation for the sake of male health and longevity; but they provide no example of any such text advising restraint in order to limit family size, nor any evidence that this textual tradition either reflected or influenced actual behavior.19 What of their claim that the Chinese employed “a wide variety of abortive techniques”? For specific practices, Lee and Wang cite the list of abortifacient drugs found in Bencao gangmu and Li Bozhong’s article about fertility control in the Yangzi Delta (which I discuss below).20 For the supposed ubiquity of abortion in Chinese society, they invoke the sociologist Fei Xiaotong: According to the well-known Chinese ethnographer Fei Xiaotong, by the early twentieth century abortion was not only widely known and used in some locales, but a woman who did not know how to use abortion to prevent a birth was laughed at by fellow villagers as a “foolish wife.”21 18 19

20

21

Wolf, “Is There Evidence of Birth Control in Late Imperial China?” 151–52. Lee and Wang, One Quarter of Humanity, 90–91. They also cite a study from Thailand as evidence that “Asian couples continue to follow a pattern of coital frequency considerably lower than elsewhere,” 189–90. In his review of their book, Patrick Heuveline points out that “the copious Western literature about the adverse health consequences of onanism does not imply restraint from it, perhaps quite the contrary”; similarly, the actual practice of marital restraint “is hardly demonstrated by the ‘copious literature on the need to limit sexual activity’ to enhance health” cited by Lee and Wang. They also cite a “forthcoming” article by Hsiung Ping-chen that appears never to have been published (as of this writing), and an article about Song-Dynasty tales of karmic retribution for abortion and infanticide; the latter seems relevant only insofar as it refutes their claim that infanticide “has not been considered immoral” in China (Lee and Wang, One Quarter of Humanity, 61). Lee and Wang, One Quarter of Humanity, 92.

abortion in late imperial china

127

Given Fei’s high reputation, this statement appears to constitute powerful evidence. But if one checks what Fei actually wrote on the subject, a different picture emerges. Whatever its merit, the “foolish wife” anecdote comes not from “some locales,” but from Dayaoshan in central Guangxi—and the people in question were not Han Chinese, but rather a subgroup of the Yao ethnic minority. Fei himself makes the specific context of his observation perfectly clear, so to imply, as Lee and Wang do, that he is speaking about China in general is misleading at best.22 What then does Fei Xiaotong say about abortion among the Chinese? He does mention in two of his books that abortion was practiced by the Chinese, but I have been unable to find in his work any concrete evidence about techniques, their effectiveness, or their actual incidence. In his classic Peasant Life in China, Fei mentions abortion in passing without providing any specific information.23 In his theoretical study entitled Shengyu zhidu (The System of Childbirth), Fei lists abortion, infanticide, and neglect as alternative means to restrict family size, citing published studies about various parts of the world (for example, Malinowski’s work on the Trobriand Islands) and anecdotes from his own fieldwork (including among the Yao people). Fei’s discussion of the Chinese focuses on infanticide and neglect; again, he offers nothing specific about abortion practices.24 In Shengyu zhidu, Fei Xiaotong does make the following cryptic comment about folk methods of contraception: “I believe the snails (luosi) that Gui Youguang’s mother ate, the fish and birds’ eggs ( yu niao dan) that women in Yangzi Delta villages ate, and other things of that kind were extremely common.”25 Fei provides no citation or any further information, but the “fish and birds’ eggs” bring to mind the tadpoles that Shanghai prostitutes used to swallow to prevent or terminate pregnancy. In 1958, after rigorous testing, tadpoles were “officially declared to have no contraceptive value”; in one study, more than 40 percent of the participating women became pregnant within

22

23 24 25

Lee and Wang cite Shengyu zhidu, where Fei clearly attributes this saying to the Hualan Yao: Fei Xiaotong, Xiangtu Zhongguo—Shengyu zhidu (Peasant life in China; The system of childbirth), 108; cf. 248–49. The anecdote first appeared in Fei’s report on fieldwork in present-day Jinxiu Yao Autonomous county, Guangxi, in the 1930s; see Fei Xiaotong, Liu shang Yaoshan (Six visits to Yaoshan), 58. Hsiao-t’ung Fei [Fei Xiaotong], Peasant Life in China, 33–34. Fei Xiaotong, Shengyu zhidu, 10–11. Fei Xiaotong, Shengyu zhidu, 11.

128

sommer

four months.26 As to the sixteenth-century literatus Gui Youguang, Fei must be referring to Gui’s essay about his mother. Gui’s mother married his father at the age of 16 sui (that is, 15 years old),27 and subsequently gave birth to seven children, the first after just a year of marriage. After finally weaning her seventh child, she complained to her servant women of being exhausted by childbirth, so one of them gave her two “snails” (luo) to swallow with water, saying that this would reduce the frequency of pregnancy. According to Gui, after taking the snails his mother became mute, and she died not long thereafter at the age of 26 sui (25 years old).28 Plainly, this anecdote does not constitute evidence of effective contraception, let alone “marital restraint.” If Gui’s mother had really had access to reliable birth control, she might have lived longer.29 In sum, the brief references in Fei Xiaotong’s work offer no support for Lee and Wang’s claims about the pervasiveness of effective contraception and abortion. What about other anthropologists’ fieldwork? Some does exist that is relevant to this inquiry. For example, elderly women in Sichuan told anthropologist Hill Gates that the methods of contraception and abortion known in their youth simply did not work.30 More detailed information is provided by Elizabeth Johnson, who interviewed rural women in Hong Kong’s New Territories in the early 1970s (when abortion was illegal in Hong Kong). According to Johnson, “none of the women mentioned abstinence as a means of birth control. This is considered unnatural.”31 As to contraception, There were apparently no effective ways of controlling conception before modern methods of control were introduced. My informants said either that women previously had no means of preventing conception or that they had not heard of any. One elderly woman asserted that she had 26

27

28 29

30 31

See Hershatter, Dangerous Pleasures, 173, 462, and sources cited therein. Any apparent contraceptive effect among prostitutes was probably the result of infertility caused by sexually transmitted diseases. A person is aged one sui at birth and gains another sui at each lunar new year; an age calculated in sui is on average one more than the same age in years old. Gui gives the dates of his mother’s birth and death, so we can calculate her age exactly. Gui Youguang, “Xianbi shilüe” (A record of my late mother’s life), 218. Nevertheless, Li Bozhong does cite this anecdote as evidence of effective contraception, on the grounds that Gui’s mother did not give birth again after swallowing the snails; he ignores the fact that by that time she was an invalid who would soon die. Li Bozhong, “Duotai, biyun, yu jueyu,” 180. Gates, “Footbinding and Handspinning in Sichuan,” 181. Johnson, “Women and Childbearing in Kwan Mun Hau Village,” 238.

abortion in late imperial china

129

s­ uccessfully prevented conception after her twelfth pregnancy by drinking a cup of unboiled water mixed with salt every night at bedtime.32 Some women, however, did attempt to induce abortion. Johnson interviewed two who had attempted abortion by traditional means a total of three times between them, with an apparent success rate of one in three: One middle-aged woman had tried in vain to induce an abortion with drugs bought from a traditional medicine shop. Another (age forty-five) had tried to abort her fifth pregnancy, taking various drugs and even deliberately taking a violent fall, but had failed. She succeeded in aborting her sixth pregnancy, taking a Chinese medicine in the fourth month.33 Since at least 15 percent of recognized pregnancies miscarry spontaneously (and poor nutrition, illness, and other factors can increase the miscarriage rate), there is no guarantee that the abortifacients worked even in the “successful” third attempt.34 One of Johnson’s most interesting findings is that her informants opposed abortion on the grounds that “it hurts the woman”: No one raised moral objections; the taking of life at this stage appears not to be considered immoral. Abortion is inadvisable, almost all the women said, because it is dangerous to the mother’s health. . . . The threat to the mother’s health was the overwhelming concern. Women expressed this folk wisdom with proverbs: “It is better to have three babies than one abortion”; and an abortion “is like picking an unripe papaya, which hurts the tree. It is better to pick it when ripe.”35 In sum, Johnson’s informants contradict any notion that traditional techniques of contraception and abortion were safe, reliable, or routinely used on a wide scale. Abortion was 32 33 34

35

Johnson, “Women and Childbearing in Kwan Mun Hau Village,” 237. Johnson, “Women and Childbearing in Kwan Mun Hau Village,” 236. Having a fever (an index for infectious disease) in early pregnancy or being underweight (an index for poor nutrition) both increase risk of miscarriage; in general, significant stress and fatigue increase risk of poor pregnancy outcomes—see Helgstrand and Andersen, “Maternal Underweight and the Risk of Spontaneous Abortion”; Kline et al., “Fever During Pregnancy and Spontaneous Abortion”; Maconochie et al., “Risk Factors for First Trimester Miscarriage” (my thanks to Joan Kreider MD for this information). Cf. Wolf, “Is There Evidence of Birth Control,” 137–39. Johnson, “Women and Childbearing in Kwan Mun Hau Village,” 236–37.

130

sommer

sometimes attempted and occasionally may have worked; but one cannot imagine these women resorting to it in a casual way. A recent study by Hua Han contributes valuable new insights from interviews with elderly rural women in Xiaoshan, Zhejiang, about life before the introduction of modern birth control. Han emphasizes that “fertility was largely uncontrolled before the 1970s,” but her informants did report a number of folk methods that at least some women used “in an attempt to limit their number of offspring and thereby reduce the physical and financial burdens of child rearing.”36 They did so not as part of a collective family birth control strategy aimed at maximizing prosperity (as Lee and Wang envision), but rather as a desperate individual measure taken secretly, in defiance of husbands and ­in-laws.37 Moreover, there is little evidence that these methods actually worked; Han characterizes them as “often risky, dangerous, and ineffective.”38 Han’s informants mentioned three herbs that supposedly could prevent pregnancy and induce sterility, but “although many women apparently are aware of these plants and their effects, they rarely used them.” Only one of the ninety women interviewed had actually taken one of these herbs, hemazi. “Because her husband was very lazy and rarely worked, she felt it would be too hard on her to have another baby while working and raising her other children at the same time. Without telling her husband, she collected the plant, boiled it, and took it regularly.” The herb may have worked, since she bore no more children (having previously borne five). But the consensus among Han’s informants was that “the hemazi plant is usually not very effective and can be very harmful, especially if women accidentally take it during pregnancy.”39 Two other women had eaten large quantities of water chestnuts (biqi) in hope of avoiding pregnancy, but with no evidence of success: one of them bore a total of six children, the last at age 37; the other had a total of eight pregnancies (including one miscarriage), the last at age 39. These numbers suggest that water chestnuts had little if any contraceptive effect.40 The only means of abortion known to Han’s informants was self-injury, using the horizontal beating bar of a loom to strike the abdomen. Such methods are notoriously dangerous and unreliable: as Han puts it, “brutal, yet not 36 37 38 39 40

Hua Han, “Under the Shadow,” 322. Han, “Under the Shadow,” 333–34. Han, “Under the Shadow,” 332. Han, “Under the Shadow,” 328–329. The other two herbs are called zhushu and zimo. So far, I have been unable to find any other information about these three herbs. Han, “Under the Shadow,” 340. According to Wolf, the mean age of last birth for Chinese women before 1949 was 38—see “Is There Evidence of Birth Control,” 141.

abortion in late imperial china

131

very effective.”41 One “model mother” who had borne nine children (“because effective birth control methods weren’t available”) reported that the loom “rarely worked.”42 Again, although many women had heard about using the loom to attempt abortion, Han found only three who had done so themselves. Two reported that they had failed, whereas the third believed she had succeeded in aborting twice. But she had a total of fifteen pregnancies, including thirteen live births and the two miscarriages attributed to the loom; two out of fifteen is 13.3 percent, well within the rate of spontaneous miscarriage, so there is no reason to assume success in her case either. One woman who reported failure with the loom explained that “she did not want to have ten children but she had no choice, as no effective birth control methods were then available in the village or in the township.”43 To summarize, Hua Han documents the desire of poor rural women to avoid further childbirth, due to physical and economic hardship—and since at least some acted on that desire, they can be said to have exerted agency in an effort to control their fertility. But Han also shows that such attempts were both uncommon and largely ineffective.44 To support their claims, James Lee and Wang Feng rely on Li Bozhong, who accepts as “fact” their argument that “women controlled their fertility,” and sets out to explain how they did so.45 Li argues that techniques for preventing and terminating pregnancy “constantly improved” (buduan gaijin) over the late imperial era, “were superior to their Western equivalents,” and “became 41 42 43

44

45

Han, “Under the Shadow,” 331. Informants also mentioned “forceful physical activities, such as jumping,” 323. Han, “Under the Shadow,” 341–42. Han, “Under the Shadow,” 331. For another ethnographic account of dubious folk methods of abortion in Zhejiang, see Cao Jinqing et al., Dangdai Zhebei xiangcun de shehui wenhua bianqian (Modern social and cultural change in rural northern Zhejiang), 96–99. The authors conclude that poverty and illness were the real checks on population growth in this region, with poverty-driven infanticide being very common. As Cao Shuji and Chen Yixin point out, northern Zhejiang (the site of both studies) lies in the heart of the region where Li Bozhong claims effective abortion and other means of fertility control were widespread and easy to obtain—see “Ma’ersasi lilun he Qingdai yilai de Zhongguo renkou” (Malthusian theory and the Chinese population since the Qing Dynasty), 49. Han shows that in the PRC era, some educated urban couples began using abstinence to “stop early”; but this was part of the modern fertility transition, not traditional practice— “Under the Shadow,” 332, 334–38. Li Bozhong, “Duotai, biyun, yu jueyu,” 172. This article is listed as “forthcoming” in the bibliography of One Quarter of Humanity and was published in 2000 in a volume co-edited by James Z. Lee.

132

sommer

ever more widespread, being accepted by the great masses of the people (wei guangda minzhong suo jieshou).” He focuses on Jiangnan, but claims that this region was “representative” of wider trends.46 Li bases this progress narrative in part on a selection of medical texts dating from the Song Dynasty through the late Qing, noting that the number of abortifacients listed in these texts declines over time, and that some of the later texts warn of the toxicity of certain substances listed in earlier ones. He notes that some substances listed in older texts (like Bencao gangmu) have been shown by experiments to have abortifacient effect, but that others are now known to be dangerous (for example, the striped blister beetle mylabris, banmao) or ineffective (for example, donkey meat). Since later texts list fewer of the dubious substances, Li argues that techniques must have improved, enabling people to regulate their fertility with increasing efficacy. He also cites elaborate prescriptions that combined a number of substances described in classical pharmacopeia as having abortifacient effect. Since these individual ingredients were effective when used alone (something Li assumes without proving), he argues that their use in combination must have enhanced their efficacy, and that such use constitutes further evidence of improvement over time.47 Another possibility, however, is that the simultaneous use of many herbs was a shotgun approach that reflected doubt about the reliability of any single ingredient.48 There are a number of problems here. First of all, Li Bozhong’s selection of medical texts is arbitrary. It may be true that those he has selected list a decreasing number of dubious abortifacients over time. But the point is moot, since old classics like Bencao gangmu have remained in print and retained their authority right down to the present day. In traditional Chinese medicine (TCM) journals today, articles on specific drugs typically begin by citing Bencao gangmu, and physicians who specialize in the problem of adverse reactions lament the “blind” use of prescriptions from such old texts. For example, in a 2004 report on seventy-three cases of TCM poisoning that received emergency treatment at a hospital in Guizhou (including five women who used the blister 46 47 48

Li Bozhong, “Duotai, biyun, yu jueyu,” 172, 179, 196. Li Bozhong, “Duotai, biyun, yu jueyu,” 175–76. Stefania Seidlecky observes that Western folk remedies “list a myriad of herbs, to be taken separately or together, often over days or weeks, thereby suggesting that none was particularly reliable either as an emmenagogue or an abortifacient” (“Pharmacological Properties of Emmenagogues,” 96). Cf. Santow, “Emmenagogues and Abortifacients in the Twentieth Century,” 82, and Santow’s critique of Riddle, much of which applies equally well to Li Bozhong (Santow, Review of Eve’s Herbs).

abortion in late imperial china

133

beetle mylabris in attempts to induce abortion), the authors make the following observation: Medical practitioners always rely on Bencao gangmu and other texts of its genre when treating illnesses with drugs; but such texts are ancient, and unavoidably their information is incomplete, lacking in detail, and sometimes wrong. For example, there is no experimental or clinical evidence whatsoever that mylabris is effective for inducing abortion or treating rabies. Nevertheless, because Bencao gangmu says that it can be used for these purposes, people have continued to do so right down to the present, resulting in countless episodes of adverse reactions. The authors list several other examples of prescriptions from classic pharmacopoeia that have caused illness, blindness, or death in recent years.49 In the final section of this paper, I address the continued use in recent years of dangerous traditional abortifacients, including those whose disappearance from texts Li Bozhong hails as evidence of progress. But also, it is not hard to find late Qing medical texts that offer what now appears to be dubious and even dangerous advice. For example, nineteenthcentury annotations of the forensic manual used by all government coroners, Xi yuan lu (Washing Away of Wrongs), explain that the blister beetle mylabris (which contains the potent toxin cantharidin, banmao su) is an effective treatment for rabies precisely because of its efficacy as an abortifacient. The reason is that the bite of a rabid dog impregnates the bitten person with tiny dogs that will kill the host if allowed to grow. Chicken eggs steamed with mylabris, if eaten, will cause these dog fetuses to be passed in the urine in the form of bloody clots. The patient must continue eating mylabris-laced eggs until no more blood clots appear. According to the text, this therapy is an example of “using poison to attack poison” ( yi du gong du).50 (Cantharidin causes kidney failure, which probably explains the blood clots.) Incidentally, the same 49 50

Ouyang Ju and Jiang Youzhong, “Zhongyao zhongdu yuanyin fenxi ji yufang cuoshi” (TCM poisoning: Causes and prevention), 724. Xu Lian, Xiyuanlu xiangyi (Detailed meaning of Washing Away of Wrongs), 4: 19a–b; cf. Smith (“Chinese Blistering Flies,” 19): “Chinese theory [is that] the bite of a mad dog impregnates the person, who is not safe until the delivery of a fetal dog by way of the urinary passages”; according to Smith, mylabris and the “red lady” (or “bride”) cicada huechys (hongniangzi, hongniang chong) were used together “to induce abortion” of this kind. Versions of this idea appear in several medical classics, including Bencao gangmu, the common theme being that successful treatment of rabies requires the patient to pass tiny dogs or dog-shaped blood clots in the urine (e.g. BCGM, 1517).

134

sommer

forensic manual also provides the recipe for an antidote to mylabris poisoning. But this recipe includes lard (zhu gao); and toxicologists today warn that fatty foods will speed the absorption of cantharidin (the active agent in mylabris) and should be strictly avoided by anyone suffering from cantharidin intoxication. Therefore, it seems likely that the forensic manual’s antidote might actually exacerbate the life-threatening effects of mylabris poisoning.51 As these examples suggest, a wider reading of medical texts does not support Li Bozhong’s confidence in the march of progress. But there is another problem with Li’s use of evidence: the pharmacopoeia he cites show that their elite authors may have believed these medicines to have contraceptive or abortifacient efficacy; but they do not prove that the medicines actually worked, or that they were even used by a significant number of people. This point is vital: clearly, some people believed (and some continue to believe) that one or another remedy worked as a contraceptive or abortifacient—for example, tadpoles, snails, blister beetles, saltwater, water chestnuts. . . . But that belief in itself cannot be taken as proof of efficacy. Li argues that modern experiments have proven some of the drugs in question to have genuine therapeutic effect.52 But as Gigi Santow points out (in her skeptical review of John Riddle’s history of emmenagogues and abortifacients in the West), laboratory tests and clinical trials do not replicate the real life circumstances in which such drugs were used in times past; they often test chemicals that have been extracted from crude materials and administered in precise doses (something not possible with traditional preparations).53 Moreover, tests have also shown that many substances traditionally used for contraception or abortion (such as tadpoles) have no therapeutic value whatsoever. Li Bozhong admits that he has little empirical evidence about how widely such methods were actually used.54 But this inconvenient fact does not deter 51

For the antidote see Xu Lian, Xiyuanlu xiangyi, 4: 21b, and Ruan Qisi, Chongkan buzhu Xiyuanlu jizheng (Washing Away of Wrongs, reprinted with amendments, examples, and commentaries), 4: 19a; for cantharidin’s toxicity see Moed et al., “Cantharidin Revisited,” 1359. 52 Li Bozhong, “Duotai, biyun, yu jueyu,” 175. 53 Santow, Review of Eve’s Herbs; cf. Santow, “Emmenagogues and Abortifacients,” 82, Seidlecky, “Pharmacological Properties,” 106–09, and van de Walle, “Flowers and Fruits,” 185. 54 Li Bozhong, “Duotai, biyun, yu jueyu,” 179. Much of Li’s evidence comes from fiction, which he cites as if it were empirical fact; for example, he cites a single work of fiction to claim that massage abortion was “very common among the people” (minjian hen pubian) in Qing-Dynasty Jiangnan (184). He also conflates prescriptions to treat fetal death in utero (seen as a medical emergency) with elective abortion for birth control (188–90).

abortion in late imperial china

135

his claim, based on anecdotes from fiction and literati jottings, that by the Ming-Qing era the practice of abortion was “very common” (shifen pubian) throughout the Yangzi Delta, and that abortion services were “very easy to obtain” (shifen rongyi huode). By that time, the “great majority” of “birth control medicines” ( jieyu yaowu) in use were “common Chinese medicines, and not particularly expensive, which made them easy to obtain.”55 He makes similar claims about contraception and sterilization, while emphasizing the relative “maturity” and “widespread use” of abortion; overall, “traditional methods of fertility control became extremely popular.”56 Therefore, Li concludes, “the notion that premodern demographic behavior was simply ‘natural’ fertility is absolutely incompatible with historical facts.”57 But this conclusion rests on conjecture rather than hard evidence.58 It seems unlikely that abortion or contraception had a significant demographic impact. But that conclusion does not necessarily rule out the gender historians’ claims about the termination of pregnancies by elite women in polygynous households. What evidence do they offer to support those claims? The Gender Historians: Autonomous Fertility Control by Elite Wives? Two distinguished historians of gender, Francesca Bray and Charlotte Furth, have used the writings of male literati physicians to study fertility regulation among elite women in late imperial China.59 As Bray and Furth make clear, these physicians portray abortion as a risky intervention justified only by medical emergency: they consistently warn that abortifacients are dangerous, and reinforce that warning by recounting illnesses and fatalities caused by attempts to induce abortion with such well-known drugs as ox knee (niuxi),

55 56 57 58 59

Perhaps his strongest evidence of abortion as birth control is a brief essay by the fourteenth-century literatus Kong Qi, which Li cites to the effect that Kong’s wife often used abortifacients, as did the wife of Kong’s friend Li Hanjie (“Duotai, biyun, yu jueyu,” 180, 190). In fact, the point of Kong’s essay is to warn of the potential lethality of these “poisonous drugs” (duyao), which should be used, if at all, with great caution and only during the first two months of pregnancy. Kong cites Li Hanjie’s wife as a cautionary example, because she nearly bled to death after using an abortifacient—and she had already given birth more than ten times. Kong Qi, “Duotai dang jin” (Be cautious about abortion). Li Bozhong, “Duotai, biyun, yu jueyu,” 180, 189–90. Li Bozhong, “Duotai, biyun, yu jueyu,” 181, 196. Li Bozhong, “Duotai, biyun, yu jueyu,” 172; cf. 196: “Traditional fertility in late imperial China, in other words, was hardly ‘natural’ fertility.” Cf. Cao and Chen, “Ma’ersasi lilun he Qingdai yilai de Zhongguo renkou,” 49. Bray, “A Deathly Disorder” and Technology and Gender; Furth, “Blood, Body, and Gender” and A Flourishing Yin.

136

sommer

musk (shexiang), and cinnamon (rougui). If a woman’s life were at stake, however, physicians would prescribe abortifacients in an attempt to save her: without question, they prioritized the life of a woman over that of her fetus.60 Nevertheless, Furth and Bray also maintain that traditional medicines for menstrual regulation (known as tongjing yao, “drugs to unblock the menses”) gave elite wives a significant degree of control over their own fertility. When a woman’s period was late, these powerful medicines could be used to “unblock” her menses; and male physicians would cooperate in this effort, because they agreed that regularity was the key to women’s health and fertility. Moreover, some of the same ingredients in abortifacients were also used for tongjing yao. Bray and Furth believe, therefore, that elite wives could take advantage of these ambiguities to terminate unwanted pregnancies: in effect, early pregnancy was euphemized as “blocked” menses, to be treated with abortifacients euphemized as emmenagogues. Ambiguity is key, for no one, it seems, would explicitly acknowledge these interventions to be abortions.61 In my view, there are two problems here. First, their argument about elite wives’ use of abortifacients for fertility control seems at odds with the explicit warnings about the dangers of abortion found in their sources. But Furth and Bray never resolve this contradiction.62 Second, they provide no documented examples of women actually terminating unwanted pregnancies in the manner they propose. For inspiration, Furth and Bray cite Chor-Swang Ngin’s fieldwork on folk methods of menstrual regulation among women of Chinese descent in 60 Bray, Technology and Gender, 322–25, 335; Furth, A Flourishing Yin, 168–69, 252–56. 61 Some historians of Europe (e.g. Riddle, Eve’s Herbs) make similar claims about traditional means of birth control; but many others disagree (e.g. Flemming, Review of Eve’s Herbs; Santow, Review of Eve’s Herbs and “Emmenagogues and Abortifacients”; Seidlecky, “Pharmacological Properties”; van de Walle, “Flowers and Fruits”). As Heuveline comments (in his review of One Quarter of Humanity), “similar knowledge has been claimed for other populations, but the efficiency of the mentioned techniques has remained dubious except for some abortion practices that so endangered the mother that their use must have been limited” (517). Cf. Brodie (“Menstrual Interventions in the Nineteenth-Century United States,” 47–54) on the ambiguity between emmenagogues and abortifacients, and on the risks and dubious effectiveness of such herbs when used to induce abortion. 62 In “A Deathly Disorder,” Bray stresses the dangers of abortifacients; nevertheless, here too she affirms that “ ‘treating amenorrhea’ may in fact be a euphemism for administering an abortifacient” (241n11) and that “the ambiguities of the diagnosis of amenorrhea permitted a woman who suspected she might be in the early stages of pregnancy to exercise control over her fertility in the name of her long-term health” (249). Her book emphasizes the freedom of maneuver these drugs supposedly gave elite wives.

abortion in late imperial china

137

Malaysia; Bray also cites similar findings by Carole Browner, who interviewed women in Colombia.63 Browner notes that Colombian women “utilize the ambiguity about when a pregnancy begins” in order to “regulate fertility”: Women have created an interval after conception but before assuming the role of pregnant during which they practice fertility regulation. This allows them to undo pregnancies they do not want while escaping the sanctions, including their own guilt, associated with the concept of abortion.64 Similarly, Ngin explains that: The ambiguity in folk beliefs about when a developing fetus becomes human makes the period between a missed menses and the positive identification of pregnancy a time in which menstrual induction cannot be subjected to negative legal sanctions or labeled medically as abortion.65 Furth and Bray apply this idea to China: a seamless shift from the folk remedies of poor women in late twentieth-century Malaysia and Colombia, to elite women’s medicine in late imperial China, implies that the same logic operated in all three milieus. Citing Ngin’s doctoral dissertation, Furth observes that “menstrual therapies stressed fertility and procreation, but if a woman wished to have no more children, such medication could disguise an attempt at abortion”—in other words, “ ‘menstrual regulation’ thrived on ambiguity about the facts of a woman’s internal bodily state,” so that “the desire for an abortion could remain inarticulate and so relatively blameless.”66 After quoting this passage from Furth, Francesca Bray concludes: I would go further. . . . The medical sources show that orthodox gynecology provided elite women in late imperial China with an approved technology of reproductive control that offered certainly not total reproductive freedom but rather room to maneuver, the possibility of c­ ultivating the 63

64 65 66

Ngin, “Reproductive Decisions and Contraceptive Use in a Chinese New Village in Malaysia” and “Indigenous Fertility Regulating Methods among Two Chinese Communities in Malaysia”; Browner, “The Management of Early Pregnancy” and “Traditional Techniques for Diagnosis, Treatment, and Control of Pregnancy in Cali, Colombia.” Browner, “The Management of Early Pregnancy,” 28, 31. Ngin, “Indigenous Fertility Regulating Methods,” 40. Furth, “Blood, Body, and Gender,” 64–65.

138

sommer

role of mater, or social mother, maybe at the expense of that of an endlessly fruitful progenitrix, or biological mother.67 Bray’s larger argument is that an elite wife would practice birth control (using the rhetoric of menstrual regulation as a cover for terminating unwanted pregnancies) in order to facilitate a division of labor in which she monopolized the authority and prestige of social motherhood, while displacing the duties and risks of biological motherhood onto the subordinate women of her household (concubines and maidservants) who were also sexually available to her husband. In other words, “flexibility of reproductive maneuver was restricted to high-ranked women and depended on exploiting the inequalities within female hierarchies.”68 To succeed, however, this exploitative division of labor would presumably depend on an elite wife’s ability actually to avoid bearing children—that is, it presupposes that this “technology of reproductive control” actually worked, and also was safe enough for routine use. Otherwise, how much “room to maneuver” could it provide? But Bray avoids the question of efficacy, and what little evidence she does present gives little reason for confidence. Both she and Furth attribute great importance to an episode recorded by the seventeenth-century physician Cheng Maoxian, in which Cheng reports his treatment of his own wife during a difficult pregnancy.69 Her health had already been severely depleted by many previous difficult pregnancies when, at the age of 40 sui, she missed her period. Fearing that it would be dangerous for her to carry yet another pregnancy to term, Cheng administered “drugs to unblock the menses” (tongjing yao), but without effect. Bray terms this situation “a classic case of ‘possible pregnancy,’ ” and compares Mrs. Cheng’s dosing with tongjing yao to the ambiguous selfmedication of the Colombian women interviewed by Browner.70 I would call it instead a classic case of resorting to abortion in a medical emergency:

67 Bray, Technology and Gender, 334. 68 Bray, Technology and Gender, 336; cf. 347–68. 69 Bray, Technology and Gender, 323–24, 329; Furth, A Flourishing Yin, 252–55. The translation in Bray was also prepared by Furth (see Bray, Technology and Gender, 324n18), but it differs from the translation presented in Furth’s own book in a number of details. For the original text see Cheng Maoxian, Cheng Maoxian yi an (Cheng Maoxian’s medical cases), 3:4b–6a. 70 Bray, Technology and Gender, 329.

abortion in late imperial china

139

the original text leaves little doubt that this was an attempt to terminate the pregnancy.71 Some months later, Cheng’s wife began to hemorrhage and he thought she might miscarry. Believing that a miscarriage would be “a blessing in disguise,” Cheng tried to induce one by dosing her with “drugs to break the blood” (poxue yao). But this second attempt also failed, and she survived, eventually giving birth to a healthy baby boy. Cheng attributes this outcome to fate: the boy was fated to be born, and nothing could stop him. But there is a simpler explanation: Cheng’s wife was forced to carry her pregnancy to term because the drugs simply did not work. Moreover, her prior history of multiple depleting pregnancies indicates the very opposite of effective birth control; she reminds us of Gui Youguang’s exhausted mother, who was enslaved to pregnancy throughout her short married life. In commenting on the story of Cheng’s wife, Charlotte Furth concedes this point explicitly: “Menstrual regulating and Blood-breaking (poxue) formulas emerge as potentially dangerous yet unreliable abortifacients.”72 But neither Furth nor Bray tries to reconcile this conclusion with the argument that elite wives gained “room to maneuver” by using tongjing yao to terminate unwanted pregnancies. Significantly, neither Carole Browner nor Chor-Swang Ngin claims that the folk remedies used by their Colombian and Malaysian informants were actually effective. Instead, both scholars focus on the “perceived attributes” of those 71

After two or three doses of tongjing yao had no effect, Cheng and his wife “did not dare” (bu gan) use more—the text does not say why they did not dare, but Bray speculates that they had concluded she was pregnant and did not want to harm the fetus (Technology and Gender, 324, 329). Since the reason to use tongjing yao in the first place was that they “feared” (kong) she was pregnant, this interpretation does not seem plausible. It seems more likely they stopped because tongjing yao was dangerous in large doses and Cheng feared harming his wife further. Cheng’s account makes clear that his motive throughout was urgent concern for her safety, which he would not have jeopardized for the sake of yet another child. I am grateful to Yi-Li Wu for her expert help interpreting this text. 72 Furth, A Flourishing Yin, 255. Furth cites this episode as evidence that women had autonomous control over fertility drugs, but the original text provides little support for her claim. In several places in her two translations (the one used by Bray and the one in Furth’s own book), where the original text has no pronoun (because classical Chinese does not require one when context is clear), Furth inserts the pronoun “she” to suggest that Cheng’s wife was making decisions and taking action independently (hence, “even though this was a physician’s family, his wife managed her own menstrual regulating medicines”). Despite the absence of pronouns, I see no ambiguity in the original text: Cheng exercises authority as husband, physician, and author, and his supervision of his wife’s treatment is obvious (A Flourishing Yin, 255).

140

sommer

remedies, in order to understand the mentality of women who would continue to use them despite the availability of modern ones that presumably are more reliable. Their evidence suggests that folk remedies were far from effective at preventing or terminating pregnancies. In fact, one motive for both scholars’ research was to help develop more effective programs for delivering modern birth control to a wider population.73 It may be true that some elite women exploited the ambiguity of menstrual regulation to terminate unwanted pregnancies, as a form of birth control. Perhaps wealth gave them access to practitioners skilled enough to reduce the risks of these drugs to an acceptable level, so that they could be used for elective abortion. But Bray and Furth provide no clear evidence of anyone actually doing so—and without such evidence, their argument remains conjecture. Moreover, there is ample reason for skepticism. As Bray points out, some of the drugs used to “unblock the menses” (tongjing) were also used to “break blood” (poxue) or to “induce abortion” (datai), and there is certainly some ambiguity in the category “tongjing yao.” This term is usually translated as “emmenagogue”; but as we have seen, Cheng Maoxian’s wife took what Cheng called “tongjing yao” in her first attempt to abort, and in modern Taiwan, the terms “tongjing” (unblocking the menses) and “tiaojing” (regulating the menses) are sometimes euphemisms for abortion services.74 Nevertheless, it is not clear how much ambiguity between menstrual regulation and abortion existed in practice. Dosage would likely vary, depending on the purpose for which the drugs were intended; but Bray and Furth say little about dosage. Bray does note that the dose of ox knee root used to induce abor73

See especially Ngin, “Reproductive Decision and Contraceptive Use,” introduction and conclusion. Among Ngin’s informants, only 26.5 percent of rural women and 16.6 percent of urban women still used folk methods, because modern contraception was increasingly available (“Indigenous Fertility Regulating Methods”); the oldest cohort of rural women (aged 40–44, the most likely to have relied on folk methods) had an average of seven living children, suggesting that if they used birth control it was unreliable (Ngin, “Reproductive Decisions and Contraceptive Use,” table 9.1). In one study, Browner interviewed 207 pregnant women at a prenatal care center; 73 described their pregnancies as “unwanted.” 55 percent of the women with unwanted pregnancies had used folk methods in an attempt to “restore” menses, without success; so had 23 percent of women with pregnancies described as “wanted” at time of interview (Browner, “Traditional Techniques”). She notes that “acceptance of pregnancy may . . . follow repeated failure of folk abortifacients: as [one woman] explained, ‘I was going to have [the baby] because there just weren’t any more remedies left’ ” (Browner, “The Management of Early Pregnancy,” 31); another said, “I used them all but none of them worked for me, nor for my sister either” (ibid., 30). 74 Bray, Technology and Gender, 332n37.

abortion in late imperial china

141

tion in a case cited by the eighteenth-century physician Xu Dachun was some thirty to fifty times larger than the tiny amounts “usually prescribed”; and that patient ended up dying of hemorrhage.75 This observation seems to confirm that the same drug might be used very differently when intended as an abortifacient as opposed to a simple emmenagogue. Nor is it clear how casually such drugs would be used: some medical authorities argue that such powerful drugs should be avoided unless absolutely necessary. For example, as Bray herself points out, Xu Dachun cautions against using tongjing yao to treat amenorrhea, warning that “these are drugs that can kill.”76 It would be interesting to learn whether the fertility of elite wives differed from that of their husbands’ concubines to any significant degree. If Bray is correct, one would expect the fertility of wives to be consistently lower than that of concubines. Even if such a pattern could be established, however, it would not necessarily follow that the reason was the wives’ use of abortifacients. The standard excuse for taking a concubine (and in orthodox terms, the only acceptable one) was the wife’s apparent infertility. But even if the wife were fertile, such a pattern might simply reflect a preference on the part of husbands for having sex with concubines instead of wives (the marriage of husband to wife was arranged, whereas men could select concubines to suit personal taste). Bray’s larger thesis about a division of labor between social and biological motherhood in the polygynous household—which is perfectly plausible— does not require that wives used medicines for fertility control. After all, one of Bray’s main contributions is to show how polygyny could serve the interests of elite women, and as she herself points out, there are many documented examples of wives acquiring concubines for their husbands.77 Male authors portray this act as exemplary self-abnegation, in which a wife subordinated 75 Bray, Technology and Gender, 322n11. Cf. an 1870 case from America in which a woman accustomed to using small doses of oil of tansy as an emmenagogue suffered convulsions and nearly died after taking a triple or quadruple dose in an attempt to induce abortion. Many women who overdosed in this way did die (Brodie, “Menstrual Interventions,” 48). 76 Cited in Bray, “A Deathly Disorder,” 247–49, and Technology and Gender, 332, cf. 333. The drugs Xu cites were also used as abortifacients and some are potentially lethal, e.g. the “blue lady” blister beetle lytta ( yuanqing or qingniangzi, which, like mylabris, contains cantharidin) and the “red lady” cicada huechys. Xu Dachun, Xu Dachun yi shu quanji (Medical writings of Xu Dachun), 2:1846. 77 Bray, Technology and Gender, 356–58; cf. Patricia Ebrey, The Inner Quarters, 220–21; Susan Mann, The Talented Women of the Zhang Family, 189; Janice Stockard, Daughters of the Canton Delta, 48–69. By custom, acquiring a concubine depended on the wife’s approval; typically, a new concubine’s first act would be to serve tea to the wife, in a ritual

142

sommer

her own natural jealousy to her husband’s comfort—but that interpretation assumes that these wives wanted to have sex with their husbands.78 If a wife really wanted to avoid the biological duties of her role, what better way than to provide her husband with an attractive younger woman? This simple expedient would seem far preferable to dosing oneself with toxic drugs.

Abortion in Qing Legal Records

To sum up the discussion so far, the demographic historians and the gender historians provide little qualitative evidence to support their claims. Significantly, they fail to provide clear, documented examples of anyone actually using abortion (or any other means) for routine birth control, either to limit family size or to avoid the biological duties of the wife/mother role. Nevertheless, important questions remain. Who actually did attempt abortion, and under what circumstances? Probably the richest untapped source for investigating this topic is Qing legal case records, which sometimes document attempts at abortion. But to make sense of this evidence, we need a firm grasp of how abortion figured in Qing legal thinking. Abortion in Qing Law American scholars interested in the history of abortion in China have generally looked to Qing law to find out whether abortion was legal. This line of inquiry dates to a seminal article by Bernard Luk, in which he observes that the Qing Code contains no explicit, blanket prohibition of abortion.79 But Luk goes on to characterize that lack of prohibition as “the traditional Chinese freedom to

78

79

of ­obeisance (Jaschok, Concubines and Bondservants, 95; Watson, “Wives, Concubines, and Maids,” 240). See Patricia Ebrey’s speculations about the feelings of a hypothetical elite wife whose “fairly normal sex life” was interrupted by her husband’s acquisition of a teenage concubine (The Inner Quarters, 166–67). Luk, “Abortion in Chinese Law.” It would be wrong to assume indifference on the part of Qing lawmakers toward an unborn fetus. For example, the Qing code added extra penalties for physical assault that caused a woman to miscarry (ibid., 379). The code also prohibited the use of torture when interrogating a pregnant female suspect, and the execution of a pregnant female offender; a woman was to be allowed to give birth before enduring torture or execution. The point of both rules was to avoid harming the fetus. Xue Yunsheng, Duli cunyi chongkanben (A new edition of Lingering Doubts upon Reading the Substatutes, hereafter cited as DC), statute 420–00 and substatutes.

abortion in late imperial china

143

abort,” which supposedly gave a Chinese woman “far greater freedom than her Western sister to dispose of the contents of her womb.”80 Subsequent scholars have tended to invoke Luk’s misleading notion of “freedom.” For example, Francesca Bray—with Luk as her only source—goes so far as to attribute explicit legal rights to individual women in the Qing dynasty (something that Luk himself does not do): “A mother was above her child, born or unborn, in the hierarchy, so her life was more valuable and her right to dispose of her child was legally acknowledged.”81 In fact, the law of no dynasty acknowledged any such “right.” The frame of reference for this line of thinking is the modern American controversy over abortion. In the United States, abortion is consistently framed as a question of what role the government should play in securing individual rights: should it protect a woman’s “right to choose,” or a fetus’s “right to life”? But it makes no sense to apply this frame of reference to Qing law, which had no doctrine of rights or even a word for that concept, and which in principle acknowledged no limit to its reach into the lives of the emperor’s subjects. Only by setting aside the anachronistic framework of rights can we understand the place of abortion in Qing law. Whereas late imperial medical texts depict abortion as a risky intervention justified by a medical crisis, Qing legal texts portray abortion as a desperate and potentially fatal act occasioned by the social crisis of adultery. As far as I can tell, abortion appeared in Qing courts only in this context. Occasionally, for example, an adulteress who became pregnant was exposed despite a successful abortion, in which case she and her lover would be prosecuted for their illicit sexual relationship.82 In most cases that appear in the archives, however, a woman who became pregnant by illicit sex had attempted abortion to avoid exposure, and the abortifacient drugs had killed her. In this situation, prosecution would target whoever had provided the fatal abortifacient. The key text is a substatute promulgated in 1740, found in the Qing Code’s chapter on “forcibly causing another person’s death because of illicit sex” ( yin jian weibi ren zhi si): If a woman becomes pregnant because of illicit sex ( yin jian you yun) and, fearing discovery, discusses this with the man with whom she has had illicit sex, and then uses a drug to induce abortion ( yong yao da tai), with the result that she aborts and dies, then the man shall be sentenced 80 Luk, “Abortion in Chinese Law,” 382, 384. 81 Bray, Technology and Gender, 334. 82 E.g. Ba County Archive (hereafter cited as BX), case 8951, Daoguang 12.

144

sommer

to 100 blows of the heavy bamboo and life exile at a distance of 3,000 li, in analogy to the statute on “knowingly selling poison to another for the purpose of committing homicide” ( yi duyao sha ren zhi qing mai yao zhe). If the man and woman who have illicit sex are related within the mourning system, and the penalty for their sex offense is greater than life exile, then he should be sentenced according to the relevant statute against illicit sex [instead of the present substatute]. If a woman who commits illicit sex herself asks someone else to buy the abortifacient [that causes her death], and her partner in illicit sex does not know about the situation, then he shall be punished only for his offense of illicit sex [and not for causing her death].83 During the Qing Dynasty, a new law would typically be issued in response to requests from provincial officials who had encountered cases of a particular scenario not precisely covered by the code.84 (The ideal of Qing legislation was to anticipate every possible variation of crime and mandate precise penalties, in order to reduce the local magistrate to a fact-finder.) The promulgation of the 1740 substatute suggests that magistrates had had to judge a significant number of cases that fit this scenario. According to the eighteenth-century jurist Wu Tan, “previously, no law had specifically addressed this scenario, so provincial officials used to handle such cases in various ways, without any consistent policy. The present substatute was promulgated to meet this need.”85 One way that such cases were handled prior to 1740 can be seen in a case from 1737, in which a man who had provided his partner in adultery with an abortifacient that killed her was sentenced to beheading after the assizes, by analogy to the basic (and rather vague) statute against “forcibly causing someone’s death because of illicit sex.”86 Wu Tan comments that such a sentence “would be excessive, because these incidents typically arise from consensual illicit sexual relations in which no ‘forcible causing’ (weibi) is involved”; but he cautions that it would be too lenient to

83 84 85 86

DC, substatute 299–11. For an example of how this worked, see Sommer, Sex, Law, and Society in Late Imperial China, 124. Wu Tan, Da Qing lüli tongkao jiaozhu (Edited and annotated edition of Thorough Examination of the Qing Code), 810–811. Neige xingke tiben (Grand Secretariat routine memorials on criminal matters, hereafter cited as XT), #28–2, Qianlong 1.12.14 (cf. XT, #12–6 and XT, #30–19, both Qianlong 1.?); DC, statute 299–00.

abortion in late imperial china

145

sentence the man only for adultery, since that adultery “had caused a death.” Hence, the balance struck by the 1740 substatute.87 The substatute demonstrates an acute awareness that adultery might lead to abortion, and moreover that an abortion attempt might well be fatal. The specific analogy drawn by the substatute in order to determine the appropriate penalty underscores the perceived danger of abortion: an abortifacient drug is analogous to deadly poison (deadly for the woman, that is), and a man who provides an abortifacient that proves fatal to the woman corresponds to someone who knowingly abets murder by supplying its means. The implication is that the man should have known that an attempt to induce abortion entailed risk of death. As Wu Tan explains, The drugs used to induce abortion are all extraordinarily harsh (kulie) in nature, so that even one out of ten women who suffer their effects may not survive (shi wu yi sheng). In this scenario, although the adulteress’s misfortune is of her own making, the man with whom she had adultery first ruined her chastity and then took her life, so it would be inappropriate to let him off lightly [by sentencing him only for his crime of illicit sex].88 In proposing a fatality rate of over 90 percent, Wu Tan may have been exaggerating for rhetorical effect. Nevertheless, his basic point about the potential lethality of abortifacient drugs conforms to the rest of the available evidence. As an experienced senior jurist, Wu was presumably in a position to know what he was talking about. The 1740 substatute addresses only an adulterer who procured an abortifacient for his partner, causing her death; but a physician, herbalist, or midwife who supplied an abortifacient in this scenario would also receive severe penalties. Case records show that several different laws could be cited to sentence these individuals, and there appears to have been no absolutely consistent policy. For example, the Qing Code’s basic statute against “fighting” (dou ou) contains a clause that reads: “Whoever severs another person’s tendons, blinds another person in both eyes, aborts another person’s fetus (duo ren tai), or wounds another person with an edged weapon, shall be sentenced to two years of penal servitude.”89 In a 1767 case, this clause was cited to sentence

87 88 89

Wu Tan, Da Qing lüli tongkao jiaozhu, 810–11. Wu Tan, Da Qing lüli tongkao jiaozhu, 810–11. DC, statute 302–00.

146

sommer

a physician who had provided an adulteress with a fatal abortifacient.90 In other cases, we find the 1740 substatute being cited for this purpose, with the practitioner who provided the abortifacient receiving a sentence reduced by one degree, as “an accomplice” (in contrast with the “ringleader,” that is, the adulterer).91 In the early nineteenth century, the judiciary settled on a policy of sentencing providers of fatal abortifacients to adulteresses by analogy to a statute in the Qing Code’s chapter on “fraud” (zhawei). This statute assigns penalties for various kinds of malingering to avoid official responsibility, including a criminal defendant who purposely inflicts self-injury in order to evade interrogation and punishment. It adds that anyone who agrees to injure [such a defendant] for pay (shou gu zhai wei ren shang can zhe) shall receive the same penalty as the defendant; but if the defendant dies as a result of such injury, the person who accepted pay to injure him shall receive the penalty for “homicide in a fight” (dousha) reduced by one degree.92 The logic behind this analogy is that a woman who has become pregnant by adultery is a criminal, and for her to induce abortion to conceal her adultery constitutes an effort to evade punishment through self-injury; therefore, a physician or herbalist who sells her an abortifacient would, in effect, be accepting pay in exchange for injuring her to abet her evasion of responsibility. Hence, if the abortifacient kills the adulteress, then the final line of this clause should apply: the resulting penalty would be 100 blows of the heavy bamboo and life exile at a distance of 3,000 li, i.e. the same penalty imposed by the 1740 substatute on an adulterer who provided his partner with a fatal abortifacient. This neat equivalence in penalties was clearly intentional. Once again, the analogy underscores the perception of high risk: to induce abortion amounts 90

91

92

XT, #1000–3, Qianlong 32.6.20; XT, #1011–14, Qianlong 32.8.9. Because of its context (in the statute against “fighting”), Bernard Luk interprets this clause to refer to “abortionin-assault”—i.e. “the inadvertent act of causing a pregnant woman to abort” by physical assault (“Abortion in Chinese Law,” 373). But there is nothing in the statute’s language to limit its application to that particular scenario, and the case I cite here simply refers to it as “the statute against aborting another person’s fetus” (duo ren tai lü). E.g. a 1774 case in which a midwife who provided a fatal abortifacient for a pregnant prostitute and a man who helped convey the prostitute to the midwife were both sentenced in this way. Zhu Qingqi et al., eds., Xing’an huilan san bian (Conspectus of penal cases, with sequels), 2:1328. DC, statute 364–00.

abortion in late imperial china

147

to ­self-injury, and to sell an abortifacient is the equivalent of accepting pay in order to injure another person.93 This perception is reinforced by Qing editions of the official forensic manual, Xi yuan lu (Washing Away of Wrongs). These texts include descriptions of the corpses of women who had died after ingesting abortifacient drugs, along with instructions about how to determine the cause of death in such cases. For example, one edition describes the corpse of a woman from Le’an county, Jiangxi, who died of hemorrhage in 1794 after swallowing an abortifacient concoction of musk and Tibetan saffron (honghua).94 These forensic manuals also prescribe antidotes for poisoning by the blister beetles mylabris and lytta; both insects were believed to have abortifacient properties, and attempted abortion was (and remains) a major reason for such poisoning.95 To sum up, if we set aside the question of “rights,” we find that Qing legal texts portray abortion as a desperate and possibly fatal act attempted by a woman who had become pregnant in an illicit sexual relationship. The priority of Qing law was not to prohibit abortion per se, but to punish adulterers—as well as others who had abetted an attempt to induce abortion that caused the death of an adulteress. I should add, however, that I have seen no indication in any Qing legal text that abortion outside the context of adultery was commonplace, or was socially or legally acceptable. The Practice of Abortion as Seen in Qing Legal Cases I have come across quite a few legal cases in Qing archives that involved abortion. I did not have the foresight to copy them all; but I do have a sample of 26 that I consider representative.96 I supplement these 26 cases from the archives

93

94 95

96

See Zhu Qingqi et al., eds., Xing’an huilan san bian, 3:1930, for an 1817 case from Sichuan that set this precedent; cf. BX, case 1618, Xianfeng 10.9. In one 1738 case, when an attempt to induce abortion (with a decoction of musk) resulted in partial abortion with retained placenta, a midwife was brought in to try to save the woman, who died nevertheless. Although the midwife had played no role in inducing the abortion, she was sentenced to a beating according to the statute against “doing things that ought not to be done—severe cases” (bu ying zhong) (XT, #90–6, Qianlong 3.12.12). Ge Yuanxu, Xiyuanlu zhiyi (Choice omissions in Washing Away of Wrongs), 1:4a, 2:17a; cf. Ruan Qisi, Chongkan buzhu Xiyuanlu jizheng, 1:34a. E.g. Ruan Qisi, Chongkan buzhu Xiyuanlu jizheng, 4:19a; Xu Lian, Xiyuanlu xiangyi, 4:21b– 22a. For the continuing problem of unsafe abortion using mylabris, see the final section of this paper. 22 are “routine memorials on criminal matters” (xingke tiben, in which provincial governors reported major cases to the imperial center for review) from the eighteenth

148

sommer

with five others that are summarized in official casebooks.97 The total sample of 31 cases includes 24 completed attempts at abortion and three incomplete attempts (in which the woman was interrupted and prevented from continuing), as well as four cases in which women actively sought means of abortion but failed to obtain them. The obvious problem with using legal cases to assess social practice is that they tend to record only people who got in trouble. All of the abortion attempts I have seen in Qing legal records involved women who had become pregnant because of illicit sexual relations; and at least two-thirds of these attempts ended in death. If it were true that many married women routinely and safely induced abortion as part of an accepted regime of birth control, they would not likely appear in legal case records. But the problem with such a supposition is that they do not seem to appear in any other source, either. It is difficult to argue for the existence of a widespread social practice on the basis of no documented examples. The great advantage of these legal cases is that they do provide 27 documented examples—some very detailed—of actual attempts to induce abortion, revealing the social context, the means employed, and the consequences; moreover, they involve ordinary women, rather than the elite wives who are the focus of Ming-Qing medical texts. I assume this sample is skewed to some extent—I assume, for example, that most attempted abortions were never reported to the authorities, regardless of outcome. Even so, Qing legal records appear to provide the only documented examples we have of the actual practice of abortion in late imperial China, aside from the medical texts cited by Bray and Furth (which do not seem to support their argument). Moreover, many other cases that do not involve abortion attempts nevertheless shed light on the availability of abortion (and of birth control in general)—and this evidence proves (pace Li Bozhong) that even women who wanted an abortion could not necessarily get one. The evidence from legal cases may not allow for definitive conclusions. At very least, however, any theory about abortion in late imperial China should be able to account for this evidence. The following table shows the marital status of the 31 women in my case sample who attempted or actively sought abortion.

97

and nineteenth centuries, and four are cases from the local court archive of Ba county, Sichuan, from the nineteenth century. Zhu Qingqi et al., eds., Xing’an huilan san bian, 2:138 and 3:1930; Xu Lian and Xiong E, eds., Xingbu bizhao jia jian cheng’an (Cases in which the Board of Punishment judged by analogy to raise or lower penalties), 297–98, 710; Zhen Ze and Shen Zhanlin, eds., Jiangsu cheng’an (Jiangsu cases), 134–35.

abortion in late imperial china

149

Marital status of 31 women who attempted or sought abortion: Widow, not remarried:98 16 Wife whose husband is away: 5 Adopted daughter-in-law, has not yet consummated marriage: 3 Betrothed, not yet married: 2 Widow remarried to her brother-in-law:99 1 Single, not yet betrothed: 1 Buddhist nun: 1 Unclear: 2 One of the widows worked as a prostitute, and she is unique in this sample because her concern was to protect her livelihood rather than to keep her sexual relations secret—she feared that pregnancy would prevent her from earning a living. The other 30 women were all desperate to abort because they feared the consequences of having their extramarital sexual activity exposed. Why were they so afraid? We should not assume that these women’s families and communities necessarily shared the official and elite obsession with female chastity—it is questionable to what extent that value was embraced by ordinary people for its own sake.100 To be sure, at least some husbands and in-laws felt entitled to kill a woman who had committed adultery, although in such cases one can often discern ulterior motives.101 Significantly, however, in nearly all of the abortion cases, we can identify specific factors that greatly raised the stakes for the people involved, giving them extra cause for anxiety beyond simple concern about losing face. (Some cases include more than one such factor.) 98

Three of the women cited in the casebooks had been married (one can tell from their names) and almost certainly widowed (as suggested by the fact that no husbands are mentioned). I include them here as widows who had not remarried. 99 In this 1800 case, from Lingling county, Hunan, remarriage by levirate was mandated by the widow’s mother-in-law, in order to provide her surviving son with a wife. When the widow’s natal family found out, they were outraged and threatened dire consequences, prompting the widow to deny that the remarriage had taken place and (when she found herself pregnant) to attempt abortion using a decoction of rice paper plant (tongcao). This abortifacient herb, which killed her, had been provided by a midwife for a fee of 4.5 shi of unhusked rice. XT, #2126–3, Jiaqing 5.10.15 & #2111–7, Jiaqing 5.5.21. 100 See, for example, Wolf and Huang, Marriage and Adoption in China, 159–60. 101 In Ming-Qing law, a husband who caught his wife in the act of adultery could kill her and her partner with impunity, as long as he acted at once (DC, statute 285–00; cf. Sommer, Sex, Law, and Society, 42).

150

sommer

Factors that raised the stakes for women pregnant by adultery: Woman is a widow whose independence is at stake: 15 Woman is pregnant by incest: 11 Woman’s sexual partner is a hired laborer: 5 Woman is a Buddhist nun, her sexual partner a monk: 1 Independent widows stand out in this list not only because they may have had more opportunity than other women to pursue illicit affairs. Most widows remarried soon after the first husband’s death, out of necessity; but a widow whose husband had left her sufficient property would likely not remarry (unless she could obtain her in-laws’ permission to bring in an uxorilocal husband). Crucially, however, a widow’s independence (including control over her husband’s property and custody of their children) depended on her reputation for chastity. If she was caught committing adultery, then her in-laws could seize her property and children and either return her to her natal family or sell her to the highest bidder. In such instances, chastity discourse was a weapon to deploy in conflicts over very material interests.102 The second factor that stands out is incest, usually involving sexual relations between women and close relatives of their husbands, which if exposed could lead to disastrous social and legal consequences. The Qing Code mandated the death penalty for sexual relations or marriage between a woman and her husband’s brother, nephew, or uncle (even though levirate was far from rare in China)—and in several of these cases, that penalty was ultimately imposed. The death penalty also applied to at least some instances of sexual relations between a woman and a long-term hired laborer of her husband’s household, a scenario that also appears in abortion cases. In the case involving a monk and a nun, exposure resulted in both being stripped of clerical status and hence losing their livelihoods (in addition to penalties for illicit sex, which were more severe for clergy). In all of these scenarios, both partners shared an urgent interest in avoiding exposure, which justified the great risk that attempting abortion evidently entailed.103 What methods did these women use to attempt abortion, and how successful were they? Out of 27 attempts, one involved self-injury and we do not know the means used in one other; but 25 attempts involved abortifacient drugs. 10 cases list the specific drugs, and several explain exactly how they were used. 102 Sommer, Sex, Law, and Society, chapter 5 (esp. 172–77, 190–97). 103 N.B.: In Jinpingmei, Pan Jinlian is both a widow and pregnant by incest when she induces abortion.

abortion in late imperial china

151

Most of the drugs were taken orally, but some were used as vaginal suppositories or abdominal plasters. For example, an 1850 case from the local court archive of Ba county, Sichuan, involved the widow Tan Liu Shi, of Wan county in the same province. She became pregnant from adultery with a relative of her dead husband’s named Tan Huanda. During a visit to her natal home, her parents noticed her swelling abdomen and accused her of being pregnant. She denied it—but desperate to avoid discovery by her parents-in-law, she confided in a woman herbalist named Tan Fang Shi, begging her for an abortifacient and promising to pay her handsomely if it proved successful. Tan Fang Shi picked “a handful of ox knee (niuxi) root” behind her house (where it appears she had a garden), and gave it to Tan Liu Shi with instructions how to prepare it. Tan Liu Shi boiled the herb in water and drank its decoction; the next morning, she aborted the fetus and “threw it away.” But she could not stop hemorrhaging (xiebeng), and after five days she died. Before dying, she admitted her adultery, pregnancy, and abortion, but refused to identify her lover or the herbalist who had supplied the drug. Several years later, Tan Liu Shi’s son learned from herbalist Tan Fang Shi exactly what had happened, and he took revenge for his mother’s death by murdering her lover, Tan Huanda. This homicide was the focus of prosecution. Given the circumstances, Tan Liu Shi’s son (the killer) received a muchreduced sentence of penal servitude, which he served in Ba county (for this reason, a record of the case survives in the Ba county archive). For providing the fatal abortifacient, Tan Fang Shi was sentenced to 100 blows of the heavy bamboo and exile at a distance of 3,000 li by analogy to the statute punishing “anyone who agrees to injure [a defendant] for pay, [so that] the defendant dies as a result of such injury.” In the end, however, she avoided punishment because of an amnesty.104 We have already encountered the herb Tan Liu Shi used to induce abortion, ox knee, which is commonly used in traditional Chinese medicine today. It is well known to have abortifacient properties and is a common ingredient in such preparations (both decoctions and vaginal suppositories): like many such herbs, it causes uterine bleeding, and the classic pharmacopoeia recommend it for treating amenorrhea and expelling retained placenta or a stillborn fetus, in addition to inducing abortion. Ox knee can be fatal in large doses. Tan Liu Shi’s death is not unique: the medical texts cited by Bray and Furth

104 BX, case 1618, Xianfeng 10.9; DC, statute 364–00.

152

sommer

also recount cases in which women died after taking large doses of ox knee to induce abortion.105 A case from Haining county, Zhejiang, reported in 1739, involved the widow Xu Zhu Shi, who became pregnant while having an illicit sexual affair with her brother-in-law Lübing (45), who was a Buddhist monk. When the widow realized she was pregnant, she told Lübing “to get a drug so that she could induce an abortion” (nong yao datai). According to her son (who had known of her adultery but was afraid to confront her), “her abdomen was already big” at the time of her abortion attempt. Lübing later testified that at first he had refused to get an abortifacient; but she pressed him, going to his temple and refusing to leave until he complied. So finally he went to a pharmacy (yaodian) and bought three drugs: the red lady cicada huechys (hongniangzi), musk, and hawthorn fruit (shanzha); the record does not tell how much he bought or what they cost. He boiled these substances and gave the decoction to Xu Zhu Shi to drink. The result was severe abdominal pain, hemorrhage, and after thirty-six hours, death. Later, the magistrate asked Lübing, “Who taught you that abortifacient prescription?” The monk testified, “In the past I had heard people say that red lady cicada and musk can cause abortion (shi xia tai de), and that’s why I bought them. Nobody taught me a prescription, and the guy at the pharmacy didn’t know what I was going to do with the drugs.” (It is possible that the monk lied to protect someone who had advised him, but the magistrate accepted his testimony.) The monk was sentenced to immediate strangulation for illicit sex with the wife of his older brother, which superseded the crime of procuring the abortifacient.106 The drugs that killed Xu Zhu Shi are well-known traditional Chinese materia medica. Hawthorn fruit is used to treat a variety of conditions, including amenorrhea, and is generally considered safe.107 Huechys, however, is highly toxic; its properties are similar to those of the blister beetles mylabris and lytta. All three insects are listed by the Hong Kong, Taiwan, and PRC governments as controlled toxic medicinal substances, and all three can be fatal even in small 105 Furth, A Flourishing Yin, 168–69; Bray, Technology and Gender, 322. The fourteenthcentury literatus Kong Qi warns against using ox knee as an abortifacient suppository, because “few who do this survive” (bu zhi yu yunjue zhe xian yi) (“Duotai dang jin,” 93). For toxicity of ox knee, see Xia Liying et al., eds., Xiandai Zhongyao dulixue (Modern toxicology of Chinese materia medica), 470–71. 106 XT, #91–4, Qianlong 4.2.18. Also see Sommer, Sex, Law, and Society, 202–03; when I summarized this case in my book, I had not identified the ingredients of the abortifacient. 107 Hong Kong Baptist University Library, “Chinese Medicine Specimen Database,” http:// www.hkbu.edu.hk/lib/electronic/libdbs/scm_specimen.html.

abortion in late imperial china

153

amounts if used internally (the typical causes of death being kidney failure and internal chemical burns).108 Musk, a pungent glandular secretion of the musk deer, is well known in TCM as an abortifacient, but it can also be lethal if used internally.109 Musk appears again in a case from Jin’gui county, Jiangsu, reported in 1767, which concerns an adopted daughter-in-law named Zhu Shi (18 sui), who had not yet consummated marriage with her fiancé Chen Fulang (17 sui). She became pregnant during an affair with her in-laws’ hired agricultural laborer, a widower named Ni Weiliang (45 sui). Ni had worked for the Chen family regularly since the age of 18 sui, receiving an annual wage of five or six taels. Zhu Shi feared for her life if her in-laws discovered her adultery. Ni told her he had heard that “if a woman carries musk on her person, she will abort” ( furen shen bian pei dai shexiang jiu ke duotai), so she gave him 300 cash to buy musk. Ni asked his maternal cousin, a physician named Hua Qun, to buy a little musk, telling him he wanted it for scent. Zhu Shi wore the musk on her person for six weeks, but to no effect. Growing desperate, Ni finally told his cousin the truth and begged him for an abortifacient, saying that Zhu Shi’s life and his own were in danger. Agreeing to help, Hua purchased borax (pengsha) and camphor (zhangnao), mixed them with musk, and formed the mixture into three balls, telling Ni that if Zhu Shi would “insert them into her vagina, then she might be able to abort the fetus” (sai ru yinhu huoke duo de tai xia). (The record does not give the amount or cost of these ingredients.) When later interrogated, however, Hua Qun insisted that “I am not someone who normally does abortions for people” (pingri ye bu shi guan dai renjia datai de). Zhu Shi followed Hua’s instructions, but after several days she fell ill with high fever and intense abdominal and pelvic pain. Her suspicious ­mother-in-law 108 Both blister beetles are close relatives of lytta vesicatoria, the notorious “Spanish fly” (this insect’s reputation as an aphrodisiac stems from the fact that one symptom of cantharidin poisoning in men is priapism). Huechys apparently does not contain cantharidin, but its use in TCM is very similar to that of blister beetles. For huechys, mylabris, and lytta, see Li Xia et al., “Authentication of 31 Species of Toxic and Potent Chinese Materia Medica” and respective entries in Xia Liying et al., eds., Xiandai Zhongyao dulixue, and in Xu Guojun and Chen Jinquan, eds., Xianggang changyong youdu Zhongyao tujian (Illustrated guide to toxic traditional Chinese medicines commonly used in Hong Kong). 109 For musk toxicity, see Hu Lifa, “Shexiang zhongdu zhi jixing shen gongneng shuaijie er li” (Two cases of acute kidney failure caused by musk intoxication), which reports a case from Gansu in which a woman died of kidney failure after swallowing 3 grams of musk to induce abortion. A different source estimates the toxic dose to be 3 or more grams, with 6 grams being fatal—Ma Jianxin, “Changjian Zhongyao buliang fanying gai kuang” (Adverse reactions from some commonly used traditional Chinese medicines), 153.

154

sommer

noticed the smell of musk and interrogated Zhu Shi, who confessed everything; the Chens then fetched Zhu Shi’s brother, and she repeated her confession to him. Over the next few days Zhu Shi’s condition worsened, so her brother carried her to Ni’s home and left her there (the implication being that her death would be Ni’s responsibility); at the same time, her father-in-law notified the local baojia head, who reported the scandal to the county yamen. The magistrate sent runners to arrest Ni and a physician to attend Zhu Shi. This physician diagnosed Zhu Shi’s condition as terminal, but prescribed medicine “to secure the fetus” (antai) in an attempt to save her. Nevertheless, the next day (ten days after inserting the suppositories) Zhu Shi finally expelled the fetus and died. Ni was sentenced to strangulation after the assizes as “a [long-term] hired laborer who has illicit sex with close relative of the head of his household” (the penalty for this offense superseded that for procuring the fatal abortifacient). For supplying the abortifacient, Hua Qun’s sentence was two years’ penal servitude and 80 blows of the heavy bamboo (according to the abortion clause of the statute against “fighting”).110 A number of classic texts provide prescriptions for abortifacient suppositories that contain musk, like the one that killed Zhu Shi, often in combination with ox knee root.111 I have not found other abortifacient prescriptions containing borax or camphor, but camphor is highly toxic and, like musk, it can be lethal if used internally.112 The Qing legal cases I have seen confirm what other sources say, namely that drugs were by far the most common method used to attempt abortion.113 The results were grim: out of 24 women who completed abortion attempts, 17 died; of the seven who survived, six did succeed in inducing abortion, but at least two of them suffered severe side effects that made them ill for months (we lack detailed information on the others). The seventh survivor’s attempt failed. To summarize, the apparent success rate in this sample was one in four (although because at least 15 percent of recognized pregnancies miscarry spontaneously, 110 XT, #1000–3, Qianlong 32.6.20 and XT, #1011–14, Qianlong 32.8.9; I am grateful to Janet Theiss for sharing her notes on this case. 111 E.g. BCGM, 266, 711; Du Wenxie, Yao jian (Mirror of medicine), 107; Chen Jiamo. Bencao mengquan (Materia medica), 38. 112 A 1997 article reports a near-fatal, unsuccessful abortion attempt in which a 16-year-old girl who was six weeks pregnant swallowed camphor; this case apparently occurred in Ireland, where abortion access is strictly limited. The authors report the human lethal dose to be 50–500mg per kg body weight. Walter Rabl et al., “Camphor ingestion for abortion (case report),” 137–38. 113 Cf. Bray, Technology and Gender, 325.

abortion in late imperial china

155

the actual success rate was probably even lower); in contrast, the fatality rate was more than two out of three.114 This evidence suggests that traditional drugs were more likely to cause illness or death than to induce abortion safely. Of course, one must assume that negative outcomes would tend to be overrepresented in legal cases. Even so, this evidence helps us understand the warning by physician Xu Dachun that “these are drugs that can kill,” the observation by jurist Wu Tan that “even one out of ten women who suffer their effects may not survive,” and the assertion by Elizabeth Johnson’s informants that traditional methods were so dangerous that “it is better to have three babies than one abortion.”115 These cases certainly demonstrate a widespread awareness that abortion was possible (my sample represents Anhui, Fujian, Guangxi, Guizhou, Hunan, Jiangsu, Jiangxi, Shandong, Sichuan, Yunnan, Zhejiang, and Zhili), but also that there were many misconceptions and that actual methods of abortion constituted a fairly specialized field of knowledge. Many people, for example, seem to have been aware that musk has abortifacient properties (it was the smell of musk that made Zhu Shi’s mother-in-law suspicious), but they did not necessarily know how to use it to induce an abortion—let alone how to do so safely. For example, Ni Weiliang believed that a woman could abort by simply smelling musk (this is an old wives’ tale that persists in China today), and when that did not work he and his lover had no idea what to do. Moreover, all but one of the women who used abortifacient drugs had to obtain them from herbalists, midwives, or physicians, and the drugs cost money.116 The Cost and Accessibility of Abortion Since abortifacient drugs usually had to be purchased from people with some degree of specialized knowledge, an important factor in assessing the accessibility of abortion is price. How much did an abortion cost, and could most people afford one? The raw ingredients of abortifacients must have ranged widely in price: a common herb like ox knee may have been cheap (or even free, if one knew 114 Of these 24 women, at least 23 used abortifacient drugs, including all 17 who died. 115 Wu Tan, Da Qing lüli tongkao jiaozhu, 810–11; Xu Dachun, Xu Dachun yi shu quanji, 2:1846; Johnson, “Women and Childbearing in Kwan Mun Hau Village,” 237. 116 In the lone exception, an 1819 case from Shaoyang county, Hunan, a wife whose husband had long been away became pregnant during an affair with a hired laborer; she successfully induced abortion using a wild herb she picked herself, “local ox knee” (tu niuxi)— this term is used in different sources to refer to various plants, so it is not certain which one she used. XT, #2732–4, Jiaqing 24.11.17.

156

sommer

how to find it oneself), whereas exotic substances like musk, camphor, cinnamon, and liquid mercury presumably cost far more. Musk is the abortifacient most frequently mentioned in my case sample (it was used in five out of the ten cases in which we can identify the abortifacient, often in combination with other drugs); it has long been prized as a costly ingredient in perfumes, as well as in Chinese medicines. Unfortunately, we do not know the amounts of specific ingredients used in preparing the abortifacients described in legal cases, nor the exact prices paid for most of them. In the last case cited above, for example, Ni Weiliang used 300 cash to buy a small amount of musk. This price can be compared to Ni’s annual wage of five or six taels, which was worth about 4,000 cash at that time. But since we do not know exactly how much musk Ni bought (let alone its quality) or whether Hua Qun needed more to prepare the suppositories, it is impossible to evaluate this price. (Nor do we know the cost of the camphor and borax.) Because musk was so potent and so costly, presumably only very small amounts were used. Still, two abortifacient prescriptions recorded by a British medical missionary in the 1920s show a wide range: a decoction to be taken orally contained just 0.37 gram of musk, whereas a suppository contained ten times that amount—and this latter dose would likely be dangerous, since as little as 3 grams can be lethal if used internally.117 Fortunately, we do know the cost of abortion in eight legal cases.118 Cost of abortion in eight legal cases, with place and year: 1)

3 taels of 70-percent cash (= 1.5 shi of husked rice); Jin’gui county, Jiangsu, 1773 2) 1,500–4,500 cash; Yongbei subprefecture, Yunnan, 1774 3) 1,000 cash; Yanghu county, Jiangsu, 1781 4) 4,100 cash; Cangwu county, Guangxi, 1788 5) 4.5 shi of unhusked rice (= 2.25 shi of husked rice); Lingling county, Hunan, 1800 6) 2,000 cash; Gao’an county, Jiangxi, 1802 7) 5 taels of silver; Guzhou subprefecture, Guizhou, 1815 8) 5 taels of silver; Yuqing county, Guizhou, 1842

117 Maxwell, “On Criminal Abortion in China,” 18; Hu Lifa, “Shexiang zhongdu zhi jixing shen gongneng shuaijie er li.” 118 XT, #1280–4, Qianlong 38.8.24; XT, #1303, Qianlong 39.2.26; Zhen Ze and Shen Zhanlin, eds., Jiangsu cheng’an, 134–35; XT, #1763–12, Qianlong 53.2.24; XT, #2126–3, Jiaqing 5.10.15 & #2111–7, Jiaqing 5.5.21; XT, # 2171–2, Jiaqing 7.1.20; XT, #2586–13, Jiaqing 20.11.14; XT, #3323–6, Daoguang 22.5.28.

abortion in late imperial china

157

To put these prices in perspective: 1.5 shi of husked rice constituted approximately one year’s supply for one person; and annual wages for agricultural laborers during the eighteenth and early nineteenth centuries averaged about 3,000–5,000 cash or five to seven taels, depending on the medium of ­payment.119 Sometimes wives were sold for similar sums: for example, in a 1798 case from Yongshan County, Yunnan, a man pressed by poverty sold his wife (aged 25 sui, with a proven record of fertility) for just five taels of silver.120 Let us take a closer look at two examples. In the 1773 case from Jin’gui county, Jiangsu (the same county as the case of Zhu Shi and Ni Weiliang recounted above), the widow Zhang Yang Shi (27) became pregnant during an affair with her dead husband’s nephew, Zhang Bao. A friend of Zhang’s, Pan Yonglin, lived with his aunt Pan Zhang Shi (a widow aged 50), who was a midwife (wenpo). Zhang Bao asked Pan whether his aunt could induce an abortion, but Pan did not know; so he took Zhang to meet her. The midwife later testified that “I had never done an abortion for anyone before” (congqian bing mei you ti ren daguo tai de), but her grandmother had told her how to do so long ago; so she agreed to help. Pan Zhang Shi stuck a small piece of musk onto the end of a stalk of “ox knee grass” (niuxi cao) and secured it with silk thread; she had guessed by palpation that Zhang Yang Shi was about five months pregnant, so she used a stalk that was five cun in length (about 6.5 inches), because her grandmother had told her its length should be one cun per month of pregnancy.121 She then inserted this device into Zhang Yang Shi’s cervix ( fang ru chanmen) and left it there; and within twenty-four hours, the widow hemorrhaged, aborted, and died. Given the stalk’s length, it may have perforated the uterus.122 The midwife agreed to provide this abortion in exchange for a fee of three taels of “70-percent cash,” the equivalent of 2,100 standard cash. To pay this sum, Zhang gave Pan Zhang Shi one shi of husked rice (calculated as worth two taels of 70-percent cash, or 1,400 standard cash) and the balance in money. 119 Philip Huang (The Peasant Family and Rural Development, 184) estimates consumption using annual rice allowances in Songjiang county during the 1950s: 3–3.25 shi of unhusked rice per capita (not differentiating by age or sex). Unhusked rice converts to husked rice at a ratio of 2:1. For wages see Li Wenzhi et al., Ming Qing shidai de nongye zibenzhuyi mengya wenti (Sprouts of capitalism in Ming-Qing agriculture), 230–38. 120 XT, #2051–8, Jiaqing 3.2.27. 121 “Ox knee grass” (niuxi cao) is hyssop, a fragrant herb that has emmenagogic and abortifacient properties (Ernst, “Herbal Medicinal Products during Pregnancy,” 231). Despite their similar names in Chinese, this plant is apparently not the same ox knee (niuxi) of which the root is usually used. 122 Chanmen (“birth gate”) can mean either vagina or cervix, but in this instance it clearly refers to the latter.

158

sommer

From this information we can calculate the real value of the fee as 1.5 shi of husked rice. The people involved in this case were all poor peasants, for whom this would have been a huge expense. Of this fee, Zhang Bao first paid the midwife 0.4 tael (worth 280 standard cash) up front to use for buying the abortifacient’s ingredients; he gave her the rice and the balance of the cash after she had inserted the abortifacient device. Since Pan Zhang Shi picked the ox knee grass herself (on the verge of a nearby rice paddy), the down payment presumably covered the cost of the musk. In other words, more than five-sixths of the midwife’s fee compensated her for her knowledge of what to use and how to use it (knowledge that does not seem to have been commonplace), as well as for her discretion.123 If we accept Pan Zhang Shi’s testimony (as did the magistrate and his superiors), then abortion was not a routine part of her midwifery practice—indeed, she had never before even attempted one. Perhaps she was lying—perhaps, in fact, she routinely provided safe and effective abortions as birth control for married couples, and the fatality in this particular case was a fluke. But if that is true—if in fact abortion was widely practiced, socially and legally acceptable, and a normal part of a midwife’s job—then why would she lie? (And why didn’t her nephew know whether she could induce an abortion?) Her claim that this was the first abortion she had ever attempted in no way mitigated her penalty for complicity in Zhang Yang Shi’s death. One might apply the same reasoning to the doctor Hua Qun, who provided the suppository that killed Zhu Shi, but who also testified that he did not normally deal in abortifacients. If the midwife did lie about her lack of experience with abortion, then the most plausible explanation would be that she sought to protect other clients who had become pregnant out of wedlock. Here we are engaged in speculation. But if abortion was closely associated with adultery in social practice as well as in legal discourse, then on previous occasions this midwife may have secretly administered abortifacients to other women in trouble, and in her testimony she may have sought to shield them from prosecution and scandal by pretending to lack such experience. But that scenario is very different from the notion that abortion was practiced by married couples as routine birth control. Moreover, it is entirely possible that the midwife simply told the truth. Speculation aside, the fact that Zhang Bao agreed without protest to pay such a high fee indicates that he had no other option. He obviously did not have a menu of different abortion services available to compare prices. The second case, from Yongbei subprefecture, Yunnan, reported in 1774, does not record the exact price of an abortion, but it does indicate the range in 123 XT, #1280–4, Qianlong 38.8.24.

abortion in late imperial china

159

which this price fell. This case concerned a woman named Xiao Wan Shi (31), whose husband had been away for a long time. She helped to support herself by occasionally sleeping with a more prosperous neighbor, Li Xiu (35, a military licentiate who was married). He would pay her 100–200 cash each time they had sex, and they kept their relationship secret. After seven months, Xiao Wan Shi became pregnant and she asked Li to give her 4,500 cash to buy an abortifacient. When Li protested at the expense (saying that surely the drug would not be that expensive), she threatened to carry the child to term and expose Li as the father, regardless of the consequences; so he gave her 1,500 cash, promising to pay the balance later on. She bought and used an unidentified abortifacient drug, but although the abortion was successful and she survived, the drug made her ill with vaginal bleeding and she was incapacitated for several months before gradually recovering. During this illness, she asked her patron to give her the promised balance of 3,000 cash, so that she could pay for medical treatment; but Li refused, reasoning that because the abortion had succeeded, he no longer needed to fear being exposed. His selfishness enraged her, and after she finally recovered from her illness she refused to sleep with him anymore, instead taking up with another neighbor who agreed to help support her. Li became jealous, stalked her, and finally strangled her. Given her negotiations with Li Xiu, Xiao Wan Shi probably paid between 1,500 and 4,500 cash for the abortifacient (she had requested the larger sum, and we do not know whether the smaller sum Li gave her covered the full cost of the drug). This price range can be compared to the 100–200 cash Li paid each time they had sex: she would have had to sleep with him some eight to fifteen times just to earn the lowest price in that range.124 Given the limitations of the evidence on prices, any conclusions must be provisional. It is safe to say, however, that the eight abortions for which we do have prices would have been expensive for most people in Qing-Dynasty China (after all, even Li Xiu, the military licentiate, balked at paying 4,500 cash)— and for the poor, such prices might well have been prohibitive. These prices may reflect the high cost of some drugs; but in at least one case, the key factor seems to have been the sheer scarcity of abortion services, which enabled a midwife to demand an extortionate fee. Indeed, cost was by no means the only factor in accessibility. Several other cases in which people desperately sought means of abortion, but were unable to find them, show that even those who could pay did not necessarily have access to abortion. 124 XT, #1301–5, Qianlong 39.2.26.

160

sommer

In a 1769 case from Shulu county, Zhili, a woman named Su Shi who became pregnant by adultery in her husband’s absence committed suicide after her lover Zhao Heizi was unable to obtain an abortifacient drug for her to use. Su Shi had originally intended to commit suicide (to avoid the consequences of having her adultery exposed), but her lover stopped her by offering “to buy an abortifacient drug for you, so that you can get rid of it” (ti ni mai fu datai de yao lai da xia lai ba). But despite his efforts, Zhao “could not find any abortifacient drug” (xun bu chu datai yao lai), and Su Shi returned to her husband’s home and committed suicide by eating arsenic.125 Similarly, in a 1787 case from Fuyang county, Anhui, a man named Jia Debao (32) was unable to procure an abortifacient drug for his cousin Jia Dajie (19) after he had gotten her pregnant by adultery. When Dajie realized she was pregnant (Debao later testified), “she told me to buy a drug to use for an abortion” ( jiao wo mai datai yao), and he agreed; but it turned out that “there was nowhere to buy such a drug” (wu chu goumai), and after four months (as her abdomen began to show), he became desperate to avoid a scandal. Finally, he gave her rat poison mixed in wine (telling her it was an abortifacient), and she drank it and died. Since Jia Debao was a prosperous landowner (with two tenant farmers and a maidservant for his wife), the high price of an abortifacient was probably not the problem. Rather, he simply could not find one to buy.126 In a 1773 case from Xing’an county, Guangxi, the widow Tang Yang Shi (41) became pregnant by adultery with a casual laborer, Li Chengzhong (41), whom she had hired for three months at a wage of 400 cash per month. Tang Yang Shi had a “goddaughter” (gan nüer) Zhao Shi, whose older brother Zhao Tinglin was a doctor. When she realized that she was pregnant, Tang Yang Shi had her lover ask Zhao Shi’s husband, Tang Mingyi, to introduce her to Zhao Tinglin and persuade him to induce an abortion; she promised to pay Tang Mingyi one tael of silver for this service, and he agreed to help. But when Tang escorted the widow to Zhao’s home, the doctor refused to administer an abortifacient, saying “that is against the law” (shi fan fa de shi). Tang Yang Shi insisted that he give her an abortifacient; but he adamantly refused to do so, and when she stayed on at his house, refusing to leave, he had his mother spend the night with her. After two days of stalemate, Tang Mingyi came to fetch her home, informing her that her in-laws (who had learned of her pregnancy) had killed her lover, Li Chengzhong. Tang Yang Shi had a forceful personality and some financial means, yet she was unable to obtain an abortion. The doctor’s refusal is as tantalizing as it is 125 XT, #1104–18, Qianlong 34.4.6. 126 XT, #1760–11, Qianlong 52.12.17.

abortion in late imperial china

161

striking: unfortunately, his testimony does not record whether he knew how to induce an abortion; nor does it specify exactly what he believed to be “against the law” (that is, abortion in general, or just providing one for an adulteress). But his refusal was absolute. Tang Yang Shi evidently knew of no other way to get an abortion, because she ended up carrying the pregnancy to term and bearing a stillborn infant. She was then prosecuted and beaten for adultery.127 The particular circumstances in these three cases varied, and we have no way of knowing all the factors involved. But it is clear that these women would have terminated their pregnancies if they had had the means to do so. Moreover (to state the obvious), they lacked effective means of avoiding pregnancy in the first place. If a woman could not afford an abortifacient drug, or for any other reason could not obtain one, what else might she do? In a 1762 case from Renqiu county, Zhili, the widow Ma Shi (27) attempted abortion by self-injury after becoming pregnant while having an affair with her hired laborer Li An (25). She was desperate to prevent her brother- and sister-in-law from finding out about her adultery, because she feared they would expel her and seize her dead husband’s property and her children. Li An later testified, “she said, ‘We’re in trouble—I’m pregnant. Now don’t worry about me, just come quickly and step on my stomach.’ Then she lay down on her back on the kang, and I got on top and stepped on her two or three times.” But just then Ma Shi’s sister-in-law walked in on them and immediately realized that Ma Shi must be trying to abort a pregnancy. Ma Shi and Li An kowtowed to the sister-in-law, begging her to conceal their secret from her husband, but when she demurred the couple decided to commit suicide together. Li An then murdered Ma Shi, her sister-inlaw, and their two daughters, and tried (but failed) to commit suicide. Li was sentenced to dismemberment for the crime of killing more than three persons of a single household.128 The crudeness of this abortion attempt reflects the protagonists’ desperation and lack of options—one is reminded of Hua Han’s informants who attempted to induce abortion by striking their abdomens with the beating bar of a loom. Such methods would almost certainly fail, and obviously could be dangerous; but like Han’s informants, this couple evidently had access to no other way to terminate a pregnancy. It is intriguing, however, that her sister-in-law realized what was going on when she saw Li stepping on Ma Shi’s abdomen; apparently there was no other plausible explanation for such behavior. 127 XT, #1275–4, Qianlong 38.7.12. 128 XT, #795–6, Qianlong 27.3.26. For a detailed account, see Sommer, Sex, Law, and Society, 203–04.

162

sommer

This example makes me wonder about the many cases of female “suicide” that can be found in Qing legal archives.129 Many concern young peasant women who died after swallowing poisonous substances that were readily available in rural settings: for example, arsenic (used for pesticide and treating skin sores); pickling brine; certain wild plants, such as “bowel-breaking poison weed” (duanchang ducao);130 and in the late nineteenth century, opium. No doubt the great majority of these deaths were genuine suicides; but I suspect (without being able to prove) that at least some may have resulted from attempts to induce abortion on the part of women who did not have access to specialists, and who in desperation resorted to toxic materials ready at hand. One can certainly find this pattern of behavior elsewhere (such as in the United States before Roe vs. Wade).131 In such cases, the actual circumstances may have gone unrecognized, or families may have concealed them. For example, in an 1824 case from Lu county, Zhili, a woman pregnant by adultery died after swallowing white lead (guanfen)—her death was interpreted as a suicide, but the substance that killed her, though used for a variety of purposes, was also traditionally believed to have abortifacient properties. There is no way to know for sure what this woman intended, but she may actually have been trying to induce an abortion.132 In fact, one might extend this speculation to China today, where the suicide rate among young rural women remains shockingly high, with drinking pesticide being one of the most common causes of death. Might some of these deaths, too, be the result of crude attempts at abortion?133

129 In early twentieth-century Taiwan, suicide rates were far higher for women than for men, and were especially high among women in their late teens and early twenties (Margery Wolf, “Women and Suicide”). 130 Gelsemium elegans, also known as “big tea leaf” (da chaye), was used for both suicide and murder—for examples of its use by women, see XT, #3–3041, Qianlong 31.5.9; #6–2325, Qianlong 31.9.5; and #7–2754, Qianlong 31.11.16. This plant appears to have been widespread and well-known. 131 Reagan, When Abortion Was a Crime. Quinn Javers, who is completing a dissertation at Stanford on unnatural death in late Qing Ba county, has found that opium was the most common means of suicide. 132 XT, #2852–1, Daoguang 4.5.9. The classic pharmacopoeia list guanfen (also known as qianfen or fenxi) as an abortifacient—e.g. BCGM, 1/266, 336–38. 133 See discussion of unsafe abortion in modern China below. China is the only country on earth today where the suicide rate is higher among women than men; the rural rate is three times the urban rate, with suicides being concentrated in ages 15–34 (Xie Chuanjiao, “China’s Suicide Rate among World’s Highest”; Fan, “In Rural China, a Bitter Way Out”).

abortion in late imperial china

163

Inconvenient Pregnancies Carried to Term For every legal case I have seen that records an actual abortion attempt, there is a far larger number that record inconvenient pregnancies carried to term, with ruinous consequences for the women whose adultery was thereby exposed. The women in these cases include much the same cast of characters found in abortion cases: widows, brides, adopted daughters-in-law (who had not yet consummated their marriages), and wives whose husbands were away from home, as well as the occasional Buddhist nun. Consider the following three examples from memorials submitted to the throne in 1736: in Changzhou county, Jiangsu, the widow Qiu Shi’s pregnancy exposed her sexual relationship with her brother-in-law, leading to them both being prosecuted and sentenced to death for incest; in Linqing department, Shandong, a young bride named Luo Shi gave birth to a healthy daughter just four months after consummating her marriage, provoking her husband to murder both her and the infant; and in Nanjing county, Fujian, a young woman named Zhuang Shi became pregnant while her fiancé was away in Taiwan, exposing her illicit affair and causing her such despair that she ended up committing suicide rather than face the consequences.134 I could cite many more examples. Such cases can serve as a sort of control set, to compare with those that document actual abortion attempts: like Tang Yang Shi (who was refused help by a doctor and ended up carrying her pregnancy to term), these women surely would have terminated their pregnancies if they had had means to do so. Another thing is clear: if they had had effective means of contraception, they would not have found themselves in this predicament in the first place. Consider the 1845 case of the widow Hu Shi from Baodi county, Zhili. Hu Shi gave birth three years after her husband’s death, prompting her in-laws to attempt to expel her from her dead husband’s household; she resisted, denying that she had engaged in adultery and claiming to be mystified about how she had become pregnant. The dispute developed into a messy court case. No one believed the widow’s story, but she stuck to it, and in the end a mediated settlement gave her most of what she wanted. (It helped that no one could identify her partner in adultery.) Tellingly, the infant (a girl) died immediately after being born: at first, Hu Shi admitted in court that she had killed the newborn, which is what her in-laws also testified; later she changed her testimony and claimed the infant had simply died. The magistrate did not pursue the question of how the infant had died, perhaps because no one wanted her anyway.135 134 XT, #7–3, Qianlong 1.4.18; XT, #11–3, Qianlong 1.6.2; XT, #20–15, Qianlong 1.9.19. 135 Baodi County Archive, bundle 162, Daoguang 25.3.8; see Sommer, Sex, Law, and Society, 205–07 for a detailed account of this case.

164

sommer

An abortion would have saved Hu Shi an enormous amount of trouble. Indeed, like Elizabeth Johnson’s and Hua Han’s informants and Cheng Maoxian’s wife, she may have tried to abort, but failed. That is what happened in a 1788 case from Cangwu county, Guangxi, in which a woman took an unidentified oral abortifacient (for which her lover had paid 4,100 cash!) without effect. She ended up giving birth, and then strangling and burying the newborn in a vain attempt to cover up what had happened.136 References to infanticide are far more common in legal cases involving unwanted pregnancy than are references to abortion (in a ratio of at least four to one), suggesting that infanticide, when possible, may well have been the preferred solution to such problems. We get a hint of the calculations involved in a 1737 case from Badong county, Hubei, in which a woman named Tan Shi became pregnant by adultery during an extended stay at her natal home to care for her ill mother. Her lover paid her father to tolerate their relationship. The possibility that she might resort to abortion or infanticide crossed her lover’s mind, and he warned her father that he wanted the child. “If your daughter uses a drug to abort it ( yong yao da xia) or if you drown it after it is born, because you are afraid that her husband will find out, then I will go tell him the truth that you pimped your daughter and she got pregnant by me. Then everyone will get in trouble, including you.” This case came to official attention when Tan Shi’s lover murdered her husband, hoping to make her his own wife; at that point she was seven months pregnant. Her lover’s statement suggests that abortion and infanticide were the two possible ways to deal with an inconvenient pregnancy. In fact, Tan Shi did intend to return to her husband (for her, the illicit affair was a fling, not a longterm relationship), and she was not complicit in his murder. When interrogated in court, she explained that she had planned to delay returning to her husband until the following spring, because “after giving birth, it would have been easy to get rid of the evidence (bunan mieji),” and her husband would never be the wiser. This statement implies that she was contemplating infanticide, after having ruled out abortion.137 But infanticide required a woman first to give birth. Childbirth itself carried risk (given the high rate of maternal mortality); but also, childbirth was hard to conceal—and for a woman trying to keep a secret, this could be a major problem. For example, in a case from 1736, the widow Qin Shi from Anhui, who became pregnant with her hired laborer, waited until she gave birth and then immediately strangled the infant. But her alert neighbors heard the ­newborn’s 136 XT, #1763–12, Qianlong 53.2.24. 137 XT, #29–09, Qianlong 1.12.19.

abortion in late imperial china

165

cries, and her adultery was exposed. She committed suicide rather than face prosecution and expulsion by her in-laws.138 Something similar transpired in another 1736 case, from Hua county, Henan: the widow Liu Shi (who had affairs with two neighbor men) did away with her infant as soon as it was born, but her snooping in-laws found the corpse—and when the widow tried to run away, they hunted her down and killed her.139 Given the terrible risks they were running, why did Qin Shi and Liu Shi carry their pregnancies to term? The most plausible answer is that they simply had no choice.

Unsafe Abortion in Modern China

Traditional Abortion Methods in Hebei, 1928 Unsafe abortion did not disappear with the Qing Dynasty. On the contrary, it has continued to occur throughout the twentieth century, even in the postMao reform era. Unique insight into traditional abortion techniques practiced in North China comes from J. Preston Maxwell, a British medical missionary who served as Professor of Obstetrics and Gynecology at Peking Union Medical College. In 1928, Maxwell described in gruesome detail his clinical experience treating the adverse effects of such techniques, in his paper “On Criminal Abortion in China.”140 Maxwell describes four basic methods of abortion. The first is “violent manipulation of the uterus from the outside,” which he states was not very common or effective: “the procedure seldom succeeds in doing more than making the lower abdomen very tender for a time.”141 Massage abortion is still practiced in some parts of the world where abortion is banned; but the success rate is low and the risk of complications high.142 I have seen no evidence that massage abortion was widespread in China. 138 XT, #32–12, Qianlong 1 (rest of date unclear). 139 XT, #4–06, Qianlong 1.3.14. 140 Cf. citation of Maxwell by Rigdon (“Abortion Law and Practice in China,” 548, 559–60) and Hershatter (Dangerous Pleasures, 462–63). 141 Maxwell, “On Criminal Abortion in China,” 13. 142 “The vigorous pummeling of the woman’s lower abdomen is designed to disrupt the pregnancy but sometimes bursts the uterus and kills the woman instead” (Grimes et al., “Unsafe Abortion: The Preventable Pandemic,” 4). See Shepherd, Marriage and Mandatory Abortion among the 17th-Century Siraya, for massage abortion as apparently practiced by an aboriginal people of Taiwan.

166

sommer

The second and third methods cited by Maxwell are “insertion of foreign bodies into the cervix,” and “ecbolic acupuncture,” that is, insertion of needles through the skin and abdominal wall into the uterus. Both methods were used occasionally, according to Maxwell. But, like the proverbial coat hanger abortion, they entailed grave risk of life-threatening complications; in the absence of sterilization and antibiotics, such invasive procedures would be perilous even if performed “correctly.”143 Maxwell reports one case of a woman who had successfully induced abortion by inserting a chopstick into her cervix, after which she nearly died of septicemia; a second woman who attempted this kind of abortion failed, but she did die of tetanus. He also recounts three cases of needling gone wrong. In two cases the needle had broken off or was otherwise lost inside the abdomen, requiring emergency surgical extraction; neither woman succeeded in aborting her pregnancy. The second of these women (a 38-year-old peasant who had become pregnant by adultery while her husband was away) had first attempted herbal abortion in vain before engaging a “Chinese native doctor” to use the needle. (The resort to ever more invasive and risky methods after less dangerous ones fail is a well-documented pattern of unsafe abortion around the world.)144 In the third case, a 43-year-old widow (who had had eight previous pregnancies, and was now pregnant again, this time by adultery) did succeed in aborting, after being needled a total of four times; but then she died of peritonitis.145 The fourth method Maxwell cites is “ecbolic medicine,” that is, abortifacient drugs, which all the evidence suggests were the principal means employed. Maxwell’s informant, a practitioner of traditional medicine, provided him with three prescriptions for abortifacient compounds, and assured him that they were effective within the first three months after conception. The first prescription consists mainly of toxic insects: it includes fourteen carcasses of mylabris, two of huechys, 5.6 grams of leeches (shuizhi), and 1.9 of gadfly (mengchong), among other things, to be powdered, mixed in wine, and taken orally.146 This prescription would likely be dangerous, since Chinese medical journals report 143 Cf. Grimes et al., “Unsafe Abortion,” 4–5. 144 Santow, Review of Eve’s Herbs. 145 Maxwell, “On Criminal Abortion in China,” 14–17; Maxwell provides an x-ray of “an ordinary carpet needle” made of steel (7.25 cm. long) lodged inside a woman’s abdomen, and a photograph of the same needle after surgical removal. 146 Maxwell refers to huechys as “lady bugs” and to gadfly as “horse fly.” For toxicity of gadfly and leeches see Xu Guojun and Chen Jinquan, eds., Xianggang changyong youdu Zhongyao tujian and Xia Liying et al., eds., Xiandai Zhongyao dulixue, 481–83, 507–08.

abortion in late imperial china

167

fatalities in women who swallowed just three mylabris carcasses.147 The second prescription combines musk with a decoction of several herbs, to be taken orally in wine. The third prescription is for a vaginal suppository made of musk and other ingredients ground into a paste and put in a silk bag.148 Maxwell notes that “quite a number of cases have come into our hands suffering from the after-effects of these drugs, which are used all over North China.”149 (N.B.: This comment highlights geographical scope—but Maxwell does not say that the drugs were used safely or effectively, let alone that they were used for routine birth control to limit family size, or that the disastrous effects he describes were in any way exceptional.) He describes two cases of side effects caused by suppositories. In the first, the abortion succeeded, but caused permanent damage to the woman’s reproductive organs: “Since that time, she has had severe dysmenorrhea at each period, the pain ceasing with the passage of yellowish discharge and blood. The vagina is blocked up by strong fibrous tissue . . . and the uterus is buried in adhesions and bound up with stricture.” In the second case, the abortion failed—and Maxwell believes that this woman was near death, although he does not know how she ultimately fared. He also cites two other cases in which women died of kidney failure, although he does not specify the methods used.150 All of the women whose specific circumstances Maxwell reports had become pregnant by adultery. In this respect, his evidence bears a grim similarity to the Qing legal cases that document abortion attempts: the common

147 E.g. Sun Zhongguo and Guo Yingli, “Banmao zhongdu siwang wei jian chu banmao su yi li fenxi” (Analysis of a fatality by mylabris poisoning, in which examination failed to detect cantharidin); in this case, the beetles were powdered, mixed with other ingredients in sesame oil, and swallowed. Xi’an Jiaotong University’s forensic pathology website estimates a fatal dose to be 1.3–3.0 grams; another source gives 1.5 grams (Xu Guojun and Chen Jinquan, eds., Xianggang changyong youdu Zhongyao tujian, 188). One problem in using crude animal and herbal drugs is that the amount of active agent may vary depending on the season harvested, the age and sex of a specimen, the parts used, how it is prepared, etc. There can also be difficulty distinguishing superficially similar species, with dangerous results—there is a large literature on how to make positive identifications by microscopy, DNA analysis, and other means not available until recently: e.g. Xu Guojun and Chen Jinquan, eds., Xianggang changyong youdu Zhongyao tujian; Ko, “Causes, Epidemiology, and Clinical Evaluation of Suspected Herbal Poisoning”; Li Xia et al., “Authentication of 31 Species of Toxic and Potent Chinese Materia Medica.” 148 Maxwell, “On Criminal Abortion in China,” 17–18. 149 Maxwell, “On Criminal Abortion in China,” 18. 150 Maxwell, “On Criminal Abortion in China,” 18–19.

168

sommer

theme is women putting their lives in jeopardy because they feared the social consequences of extramarital pregnancy. It should be noted that Maxwell’s use of the term “criminal abortion” simply reflects the fact that induced abortion was prohibited under Chinese law at the time he was writing. Maxwell’s chief concern is “the terrible risk to the patient” posed by these methods, which he terms “a serious menace.”151 Of course, like legal cases, Maxwell’s evidence is anecdotal; and like a magistrate, a physician would likely learn about only those abortions that went dangerously wrong. Nevertheless, like the legal cases, his evidence argues against complacency about traditional abortion techniques; at very least, it suggests that the burden of proof lies with anyone who would claim such techniques were safe or reliable enough for routine use. Traditional Abortion Techniques since 1949 Since 1949, the government of the PRC has energetically promoted traditional Chinese medicine for ideological and economic reasons; and since the 1970s, it has also imposed compulsory birth control to limit population growth. Therefore, if safe and reliable techniques of abortion had been widely used before 1949, one would expect them to remain in widespread use today. What, then, has been the modern fate of traditional techniques? In recent decades, a number of herbal abortifacients have been tested under controlled clinical or laboratory conditions, sometimes in crude form but often as chemical extracts that can be injected. The greatest success, it seems, comes when certain traditional materials are used in conjunction with modern techniques. For example, in the 1970s an extract of snake-gourd root (tianhua fen) called tricosanthin was used (via injection) along with prostaglandins to induce early abortion, with a reported success rate of 81 percent. Later, however, tricosanthin was mostly replaced by mifepristone (RU-486), which gives far more satisfactory results. Mifepristone is the technique of choice for medical abortion in the PRC today, although it is occasionally supported by traditional therapies.152 For example, some physicians report successful use of ox knee root to dilate the cervix in conjunction with either mifepristone

151 Maxwell, “On Criminal Abortion in China,” 13. 152 Cheng Linan, “Medical Abortion in Early Pregnancy: Experience in China”; cf. Rigdon, “Abortion Law and Practice in China,” 548–549. For dangerous side effects of tianhua fen and tricosanthin, see Li Ling and Huang Guangzhao, “Jixing tianhua fen zhongdu shijian er li” (Autopsy report on two cases of acute tianhua fen poisoning).

abortion in late imperial china

169

or ­surgical abortion;153 another claims success using an abdominal plaster containing musk and herbs to improve the effectiveness and reduce the side effects of mifepristone.154 One can certainly find accounts of successful use of traditional prescriptions administered at modern TCM clinics.155 But medical journals also contain many reports of unsatisfying results or adverse effects. For example, one physician reports his attempt to induce abortion using two different traditional herbal decoctions made with musk, ox knee, and other ingredients: these medicines had no effect, and the patient ended up requiring a surgical abortion.156 Another physician used three different well-known traditional preparations to attempt abortion in 41 separate cases, with a failure rate of 100 percent. In a telling comment, however, he notes that “the preparations used in the experiment were screened for safety, and such dangerous ingredients as mercury and cantharides were removed.”157 (“Cantharides” are blister beetles, that is, mylabris or lytta.) It appears that by removing the obviously toxic substances, this physician had eliminated whatever abortifacient efficacy these preparations may have had. Similarly, two other physicians report their failure to induce abortion using just one carcass of mylabris, administered orally with warm water, as Bencao gangmu prescribes for expelling a dead fetus from the womb. They conclude that mylabris in safe dosage is utterly ineffective as an abortifacient, and that higher dosages would simply be too dangerous to attempt.158 This evidence is anecdotal, but it suggests that practitioners 153 E.g. Huang Guoping et al., “Liuchan yaowu de linchuang yingyong” (Clinical use of abortifacient drugs); Ji Xiaofen et al. “Niuxi zai rengong liuchanshu zhong de yingyong (Use of ox knee in surgical abortion). 154 Shang Jing, “Shexiang fengshigao tieji tigao yaowu liuchan chenggonglü de linchuang guancha” (Clinical observation of success rate in using a musk compound to induce abortion). In an 1875 case from Taicang Department, Jiangsu, a midwife used an abdominal plaster containing musk to induce abortion, resulting in the woman’s death—XT, #3751– 26, Tongzhi 13.12.18; XT, #3966–16, Guangxu 1.11.4. 155 E.g. Qin Yuhui and Li Peigan, “Hongcaotang kangzaoren 38 li” (38 cases using “red herbal decoction” to terminate early pregnancy); Wang Fengmin and Dai Zhannan, “Cui jing zhi yun tang kang zao zaoyun 34 li” (34 cases using “menstruation-encouraging and pregnancy-terminating decoction” to terminate early pregnancy). 156 Weng Yifeng, “Junji gongzhu pei shexiang duotai bu xia yi li” (A case of failed abortion using a musk compound). 157 Li Fu-kuang, “Problems Concerning Artificial Abortion through Oral Administration of Traditional Drugs,” 200. 158 Liao Anshui and Cheng Yuzhu, “Banmao zhongdu er li baogao” (Two cases of mylabris poisoning); cf. BCGM, 1:266.

170

sommer

who err on the side of safety have found traditional preparations of crude drugs less than satisfactory. In contrast, a case reported in 1986 concerns an 18-year-old high school student, five months pregnant, who begged her school doctor to help her. The doctor used a prescription containing one mylabris carcass, along with dried leeches, gadfly, and the herb qumai, the decoction of which was to be taken orally. When this prescription failed to induce abortion, the doctor had the student take it three more times, increasing the dose of mylabris in small increments, but still without effect. Finally, in frustration, the doctor administered a fifth dose using 30 whole mylabris carcasses—and that dose killed her. The pathologist who conducted the autopsy estimated that the amount of mylabris ingested had exceeded the fatal dose by more than five times.159 The Persistence of Unsafe Traditional Abortifacients These last three examples involving mylabris highlight the question of safety. Blister beetles are among the most dangerous substances traditionally used to attempt abortion. Nevertheless, as we have seen, many traditional abortifacient preparations include mylabris, its relative lytta, and/or the cicada huechys, sometimes in combination. We find mylabris in the abortifacient that Pan Jinlian used in Jinpingmei; it also appears (along with huechys) in one of the abortifacient recipes reported by Maxwell. Even today, there persists a widespread belief that mylabris can be used to induce abortion, as well as to cure rabies and cancer. As a result, poisoning and even death by mylabris ingestion remain a real problem. According to two faculty at Guangzhou TCM University, mylabris poisoning usually results from the use of “folk remedies” (minjian pianfang) in rural regions that are “relatively backward economically and educationally” ( jingji wenhua jiao wei luohou).160 But such notions are not limited to the rural poor—witness the following statement published in 2004 by a demographic historian at Renmin University: I once heard a woman say that because on one occasion she had happened to hold a piece of musk, she was no longer able to become pregnant (but since she already had four or five children, she was perfectly happy with this result). Also, a friend once told me that inserting a drug known as “mylabris” into the cervix (twisted in a piece of paper) is an 159 Editorial Committee, Linchuang yiliao huli shiwu fenxi (Analysis of errors in clinical care), 690–92. 160 Liang Jinquan and Wang Ningsheng, “Banmao de duxing fanying ji yuanyin fenxi” (Analysis of mylabris toxicity), 77.

abortion in late imperial china

171

effective means of preventing conception. Apparently in the old days prostitutes used this method, and after they quit prostitution they could still bear children without difficulty.161 If a professor at an elite university would credit such dangerous nonsense, we should not be surprised that some uneducated rural people might also believe it. Cantharidin (the active agent in blister beetles) is highly caustic, and can be used externally as a vesicant;162 in TCM mylabris is used in the same way, the powdered insect being applied to the skin (for example, to remove warts). But careless handling of the insect or inhalation of the powder can cause severe chemical burns—and for a woman to insert mylabris into her cervix would be a disaster. Of course, it is possible that desperate women have used it in this way in an attempt to prevent or terminate pregnancy—after all, in the United States before Roe vs. Wade, it was common for women to attempt abortion by douching with bleach or lye, even though that technique often proved fatal.163 I have not found statistics on how often mylabris poisoning caused by attempted abortion occurs today, so my evidence is anecdotal. But there is no shortage of reports from the last few decades about such incidents, including many fatalities. For example, a single hospital in Shandong provided emergency care for 88 separate cases from 1989 to 1996; a different hospital, also in Shandong, treated 42 cases from 1997 to 2001.164 Some poisonings result from attempts to treat dog bites or cancer, but many victims are women who hope to terminate pregnancy: according to a 2004 report from Zhangjiakou, Hebei, “mylabris poisoning occurs for various reasons, but by far the most common is that young unmarried women who become pregnant use it in a misguided attempt to induce abortion.”165 In a cursory survey of medical journals online, I found reports of such incidents from Beijing, Gansu, Guangdong, Guizhou, Hebei, Hong Kong, Hubei, Jiangxi, Qinghai, Shaanxi, Shandong, Sichuan, 161 Gao Wangling, “Ma’ersasi yinyingxia de renkou zhenglun” (The population debate in Malthus’s shadow), 16. 162 Moed et al., “Cantharidin Revisited.” 163 Reagan, When Abortion Was a Crime, 208–209. 164 Hu Ping and Hu Dechuan, “Banmao zhongdu 88 li baogao” (88 cases of mylabris poisoning); Jin Haiyan, “Banmao zhongdu 42 li” (42 cases of mylabris poisoning). A search in the Chinese Academic Journals Database (available through Stanford’s East Asian Library) on January 26, 2009, found 93 articles from 1980–2002 with both “mylabris” (banmao) and “poisoning” (zhongdu) in their titles. 165 Sun Zhongguo and Guo Yingli, “Banmao zhongdu siwang wei jian chu banmao su yi li fenxi,” 28.

172

sommer

Tianjin, and Yunnan.166 In the reports I have seen, every woman who succeeded in aborting subsequently died, whereas those who survived failed to abort.167 In a typical example, an article published in 1990 reports two separate cases of rural women in Hubei, both unmarried, who died of mylabris poisoning while attempting to induce abortion. The first woman, age 21, was seven months pregnant when she drank a mixture of powdered mylabris in wine, late one night while alone; by dawn she was dead. The other woman, age 22, was 166 For examples, see Chen Jianfei. “Zhongcaoyao zhongdu 50 li zonghe fenxi ji wenxian fuxi” (50 cases of poisoning with traditional herbal medicines: Analysis and literature review); Chen Jinhua, “Banmao zhongdu siwang de jianyan wu li baogao jian lun banmao duotai de weixian” (Five cases of death by mylabris intoxication: The danger of mylabris when used as an abortifacient); Cheng Kok-Choi et al., “A Fatality Due to the Use of Cantharides from Mylabris Phalerata as an Abortifacient”; Hou Rumiao and Ma Jiaji, “Jixing banmao zhongdu yi li baogao” (A case of acute mylabris poisoning); Hu Ping and Hu Dechuan, “Banmao zhongdu 88 li baogao”; Jin Haiyan. “Banmao zhongdu 42 li”; Jin Su, “Wu zhong changyong youdu chonglei Zhongyao de duxing ji jiejiu fangfa” (Five poisonous insects commonly used in traditional Chinese medicine); Liang Bingxue, “Banmao zhongdu er li baogao” (Two cases of mylabris poisoning); Liang Lei and Hu Yuchuan, “Banmao zhongdu yi li baogao” (A case of mylabris poisoning); Liao Anshui and Cheng Yuzhu, “Banmao zhongdu er li baogao” (Two cases of mylabris poisoning); Ouyang Ju and Jiang Youzhong, “Zhongyao zhongdu yuanyin fenxi ji yufang cuoshi” (TCM poisoning: Causes and prevention); Pan Qiying, “Jixing banmao zhongdu yi li baogao” (A case of acute mylabris poisoning); Tong Yaohua, “Banmao zhongdu de qiangjiu yu huli” (Emergency treatment of mylabris poisoning); Weng Xinzhi and Gao Chongji, “Banmao zhongdu—wenxian fuxi ji yi li baogao” (A case of mylabris poisoning); Xie Liansheng, “Qiantan jixing banmao zhongdu de zhengzhi” (Diagnosis and treatment of acute mylabris poisoning); Xu Chucai, “Jixing banmao zhongdu san li baogao” (Three cases of acute mylabris poisoning); Xun Xianwen, “Banmao zhongdu siwang er li” (Two cases of death by mylabris poisoning); Zhang Jizong et al., “Yong GC/MS jianding shiti zangqi zhong de banmaosu” (Autopsy use of GC/MS to identify cantharidin in organs, with analysis of one case); Zhang Junwen and Zhang Yuwu, “Zhongyao zhongdu 20 li baodao” (20 cases of poisoning with traditional Chinese medicines); Zhang Yigu and Huang Guangzhao, “Youdu dongwu zhongdu 4 li shijian baogao” (Autopsy report on four cases of poisoning with toxic animal products); Zou Weibo, “Koufu banmao duotai zhongdu zhi siwang de fayixue jianding” (Forensic diagnosis of fatal poisoning caused by oral ingestion of mylabris as an abortifacient). Also see Jiang Qingyu and Qi Yongmao, eds., Zhongyao buliang fanying (Adverse reactions to traditional Chinese medicine), 293–95. 167 There are also reports of mylabris poisoning resulting from attempts to treat amenorrhea or infertility, e.g. Jin Yacheng, “Banmao neifu zhi si yi li baogao” (A case of death caused by ingestion of mylabris), and Sun Zhongguo and Guo Yingli, “Banmao zhongdu siwang wei jian chu banmao su yi li fenxi.”

abortion in late imperial china

173

four months pregnant when she took six powdered mylabris beetles orally; she also applied plasters containing four more powdered beetles to her temples and lower abdomen. She died within twenty-four hours.168 A 1999 report from Guangdong describes an unmarried woman, age 21, who was three months pregnant when she drank about half an ounce of a decoction of mylabris carcasses. She too died within twenty-four hours, despite emergency treatment.169 Blister beetles are by no means the only culprits in such cases; one can also find reports of poisoning caused by musk and various herbal abortifacients. For example, a 1986 article reports a case from Guilin in which an unmarried woman aged 24, who was two months pregnant, used a bundle of white leadwort roots (baihuadan) tied in red string as a vaginal suppository. She had obtained this herb, known colloquially as “local ox knee” (tu niuxi), from an unlicensed TCM practitioner. Two days after inserting the suppository, she successfully aborted (in a public toilet), and then collapsed and was sent to hospital. Medical treatment failed, and several days later she died of kidney failure.170 Such accounts are shockingly similar to the scenarios found in both Qing legal cases and Maxwell’s 1928 report. Most portray women who resorted to desperate measures because they feared having their extramarital sexual relations exposed. The website of the Xi’an Municipal Government’s Committee for Population and Family Planning contains the following statement, entitled “Never take drugs to induce abortion on your own!” (qie ji zifu datai yao):171 Some women who become pregnant wish to abort for various reasons, but they are too embarrassed to go to a hospital for a clinical abortion, so 168 Xun Xianwen, “Banmao zhongdu siwang er li.” 169 Zou Weibo, “Koufu banmao duotai zhongdu zhi siwang de fayixue jianding.” 170 Zhang Cun, “Baihua dan duotai zhi baixiezheng jixing shen gongneng shuaijie siwang yi li baogao” (A case of death by septicemia and kidney failure from use of baihua dan as an abortifacient). For toxicity of baihua dan, see Xia Liying et al., eds., Xiandai Zhongyao dulixue, 253–55. For poisoning by musk (taken orally) see Hu Lifa, “Shexiang zhongdu zhi jixing shen gongneng shuaijie er li”; for herbs such as yuanhua and tianhuafen (used as vaginal suppositories), see Li Ling and Huang Guangzhao, “Jixing tianhuafen zhongdu shijian er li”; Ma Jianxin, “Changjian Zhongyao buliang fanying gai kuang”; Zhang Yigu and Huang Guangzhao, “Jiu zhong youdu zhiwu zhongdu de shijian baogao” (Autopsy report on cases of poisoning with nine species of toxic plants). 171 Xi’an Municipal Government Population and Family Planning Committee, “Qie ji zifu datai yao” (Never take drugs to induce abortion on your own), http://www.xianfp.gov .cn/Index/Catalog119/3105.aspx. The same statement appears on the websites of several other local governments.

174

sommer

they decide to take drugs to induce abortion on their own. Such action is extremely dangerous! It can easily lead to massive hemorrhage, partial abortion, or a variety of gynecological illnesses. This is a very common problem. Because the toxicity of abortifacient drugs is high, and the required dosages are also high, these drugs can easily cause poisoning. For example, in the countryside there are women who use the medicinal beetle mylabris to induce abortion. The average adult who takes just 0.6 gram of mylabris orally can have a severe toxic reaction; 1.3–3.0 grams can cause immediate death. . . . There are also women who insert ox knee, monkshood daughter ( fuzi), musk, yellow oleander leaf ( jiazhutao ye), or other herbs into the vagina hoping to induce abortion, but this method often causes severe vaginal hemorrhage and can endanger a woman’s life. There are other traditional Chinese medicines that can harm a pregnant woman if used to induce abortion. These include nux vomica (maqianzi), Jack-in-the-pulpit tuber (sheng nanxing), common monkshood mother root (sheng chuanwu), Kusnezoff monkshood root (sheng caowu), liquid mercury, croton seed (badou), centipede (wugong), dried leeches (shuizhi), common burreed tuber (sanling), and motherwort herb ( yimu cao).172 Here we have a straightforward account of unsafe traditional abortion in China today, including a list of many of the most dangerous substances long used for this purpose. It would be wrong, of course, to suggest that nothing has changed in China. There has been dramatic progress in maternal health over the past couple of decades in particular: according to government statistics, China’s maternal mortality rate dropped from 94.7 mother deaths per 100,000 live births in 1990 to 48.3 in 2004.173 Today, safe contraception and abortion are widely available in China—and as a result, the rate of unsafe abortion is considered negligible, especially compared to countries where abortion is banned or otherwise inaccessible. Nevertheless, death by unsafe abortion is notoriously underreported and its incidence is difficult to estimate, especially in developing countries.174 172 Many of these items are listed by the PRC and Hong Kong as toxic medicines subject to restriction. Monkshood daughter and Kusnezoff monkshood contain aconite; nux vomica contains strychnine. 173 WHO Representative Office in China, “Maternal and Child Health in China,” http://www .wpro.who.int/china/sites/mch/overview.htm. 174 World Health Organization, Unsafe Abortion; Grimes et al., “Unsafe Abortion.” I have been unable to find systematic reports on the incidence of death by unsafe abortion in

abortion in late imperial china

175

Moreover, given the extraordinarily high rate of suicide among young women in rural China today, it seems possible that these “suicides” include at least some deaths by unsafe abortion that have been unrecognized or deliberately misreported. As the Xi’an government website points out, most mylabris abortion poisonings seem to involve young rural women who are “too embarrassed” to seek a clinical abortion. Surveys suggest that rural migrants to cities are increasingly uninhibited about premarital sex, and are willing to seek safe contraception and abortion; but in home villages, attitudes may remain harsh. A group of rural women interviewed in Guiyang in 2001 explained that premarital pregnancy is taboo: “Nobody dares get pregnant before marriage! The old people would kick you out of the house if they found out! It is not an issue in the cities, but it matters in a village. The whole family would lose face!”175 One thing has not changed, evidently: at least some women still fear the exposure of their extramarital sexual activity to such an extent that they cannot risk the publicity that seeking a safe clinical abortion might entail. Instead they resort to unsafe traditional methods, with the results we have seen. The cases of poisoning by traditional abortifacients that occasionally surface in medical reports today are significant for our inquiry mainly because they shed light on how abortion was attempted in times past. These reports provide a glimpse in the present day of the practices that our older sources document. If unsafe abortion by traditional means is unusual in China today, that is fortunate indeed—but we must bear in mind that until very recently, such means were the only ones available. Conclusion WHO estimates that some 500,000 women die each year from pregnancy-related causes worldwide, and between one-fourth and one-third China; most reports on maternal mortality either exclude this cause from consideration or (apparently) include it in the category “other.” But in a striking exception, a report from Pingguo county, Guangxi, states that out of 31 women’s deaths related to pregnancy or childbirth in 1989–91, five deaths resulted from hemorrhage or infection in rural women who had “used herbal drugs to induce abortion on their own” ( yong caoyao zi xing duotai). Huang Runhong, “Pingguo Xian 31 li yunchanfu siwang qingkuang fenxi” (Analysis of 31 women’s deaths in pregnancy or childbirth in Pingguo county). 175 Zhenzhen Zheng et al., “Sexual Behaviour and Contraceptive Use among Unmarried, Young Women Migrant Workers in Five Cities in China,” 123.

176

sommer

of these stem from unsafe abortion. Of abortion-related deaths, 98 percent occur in Third World countries. Thousands of other women suffer serious ill health from the complications of unsafe abortion. . . . There should be no complacency, therefore, regarding the safety and reliability of folk remedies.176 Reports from the field add flesh to these figures. For example, in Uruguay (where abortion is illegal), during the years 1986–99, one poison center provided emergency care to 86 women made severely ill by herbal abortifacients.177 30 different plant species were identified among the abortifacients; 11 women had also attempted self-injury in addition to taking herbs. Only 23 of these women succeeded in inducing abortion; but 14 suffered multiple organ failure, nine required hysterectomy, and five died. Many other women suffered serious and lasting damage to their health. “Most of the abortions [that did succeed] occurred in patients with multiple organ system failure, and liver or blood diseases. These results strongly suggest that most of the abortions happen in the context of herbal intoxication.” In other words, abortion occurs as a result of poisoning. The report concludes that “ingestion of plants to induce abortion involves the risk of severe intoxication that could result in death or future reproductive complications.”178 Some historians have claimed that abortion was used for routine birth control in late imperial China; despite their differences, these scholars apparently assume that traditional methods of abortion were safe, effective, and readily available to women who wished to use them. If correct, these claims would have the most profound implications for our understanding of Chinese history. The Chinese might be the only people on earth to have enjoyed such effective fertility control before the advent of modern medicine. In fact, I believe, China was not so exceptional. The practice of abortion in China before 1949 actually looks quite familiar: it looks like unsafe abortion as practiced in the United States before Roe vs. Wade; it looks like unsafe abortion as practiced in Uruguay and many other countries today, often by means of herbal concoctions as well as invasive measures or self-injury.179 The folk methods that survive in China fit this picture very well, in that they are dangerous and unreliable.

176 177 178 179

Seidlecky, “Pharmacological Properties of Emmenagogues,” 96. Ciganda and Laborde, “Herbal Infusions Used for Induced Abortion.” Ciganda and Laborde, “Herbal Infusions Used for Induced Abortion,” 237–238. Grimes et al., “Unsafe Abortion.”

177

abortion in late imperial china

Safe and reliable contraception and abortion are among the great benefits of modernity, but like other such benefits they are not universally shared. Where they are not available, whether because of prohibition, stigma, poverty, or other factors, desperate women continue to resort to unsafe traditional methods—and this remains true even in China, despite that country’s overall progress in protecting maternal health.180 Glossary antai badou baihua dan banmao banmao su Bencao gangmu biqi buduan gaijin bu gan bunan mieji bu ying zhong bu zhi yu yunjue zhe xian yi chanmen Cheng Maoxian congqian bing mei you ti ren daguo tai de da chaye datai dousha duanchang ducao Duotai dang jin fang ru chanmen Fei Xiaotong fenxi

安胎 巴豆 白花丹 斑蝥 (斑蟊、斑毛、斑貓) 斑蝥素 《本草綱目》 荸薺 不斷改進 不敢 不難滅跡 不應重 不致于殞絕者鮮矣 產門 程茂先 從前並沒有替人打過胎的 大茶葉 打胎 鬪殺 斷腸毒草 《墮胎當謹》 放入產門 費孝通 粉锡

180 Estimates of worldwide mortality due to unsafe abortion vary widely: anywhere from approximately 70,000 deaths per year to more than twice that number (e.g. Grimes et al., “Unsafe Abortion”; World Health Organization, Unsafe Abortion; Population Council, “Safe Abortion,” http://www.popcouncil.org/topics/mnh_safeabor.asp#). For perspective, in the 1920s unsafe abortion killed an estimated 15,000 women every year in the United States alone (Reagan, When Abortion Was a Crime, 139).

178 furen shen bian pei dai shexiang jiu ke duotai fuzi gan nüer gansui guanfen Gui Youguang hemazi honghua hongniang chong hongniangzi Hualan Yao jiao wo mai datai yao jiazhutao ye jieyu yaowu jingji wenhua jiao wei luohou Jinpingmei cihua kong Kong Qi kulie Li Bozhong Li Shizhen luo luosi maqianzi mengchong minjian hen pubian minjian pianfang miulun wu chuan niuxi niuxi cao nong yao datai pengsha pingri ye bu shi guan dai renjia datai de poxue poxue yao qi qianfen qie ji zifu datai yao qingniangzi qumai

sommer 婦人身邊配帶麝香就可墮胎 附子 乾女兒 甘遂 官粉 歸有光 河馬子 紅花 紅娘蟲 紅娘子 花籃/藍瑤 叫我買打胎藥 夾竹桃葉 節育藥物 經濟文化較爲落後 《金瓶梅詞話》 恐 孔齊 酷烈 李伯重 李時珍 螺 螺螄 馬錢子 虻蟲 民間很普遍 民間偏方 謬論誤傳 牛膝 牛膝草 弄藥打胎 硼砂 平日也不是慣代人家打胎的 破血 破血藥 妻 鉛粉 切忌自服打胎藥 青娘子 瞿麥

179

abortion in late imperial china rougui sai ru yinhu huoke duo de tai xia sanling shanzha sheng caowu sheng chuanwu sheng nanxing Shengyu zhidu shexiang shi fan fa de shi shi xia tai de shi wu yi sheng shifen pubian shifen rongyi huode shuiyin shuizhi shou gu zhai wei ren shang can zhe taoren ti ni mai fu datai de yao lai da xia lai ba tianhua fen tiaojing tongcao tongjing tongjing yao tu niuxi wei guangda minzhong suo jieshou wenpo wu chu goumai Wu Tan wugong Xianbi shilüe xiebeng Xing’an huilan xingke tiben Xi yuan lu xun bu chu datai yao lai yang qian liang yuan yaodian yi du gong du yi duyao sha ren zhi qing mai yao zhe

肉桂 塞入陰戶或可墮得胎下 三棱 山楂 生草烏 生川烏 生南星 《生育制度》 麝香 是犯法的事 是下胎的 十無一生 十分普遍 十分容易獲得 水銀 水蛭 受雇債為人傷殘者 桃仁 替你買服打胎的藥來打下來吧 天花粉 調經 通草 通經 通經藥 土牛膝 為廣大民衆所接受 穩婆 無處購買 吳壇 蜈蚣 《先妣事略》 血崩 《刑案匯覽》 刑科題本 《洗冤錄》 尋不出打胎藥來 洋錢兩圓 藥店 以毒攻毒 以毒藥殺人知情賣藥者

180

sommer

yimu cao yin jian er weibi ren zhi si yin jian you yun yong caoyao zi xing duotai yong yao da tai yong yao da xia yu niao dan yuanhua yuanqing zhangnao zhawei zhongdu zhu gao zhushu zimo

益母草 因姦而威逼人致死 因姦有孕 用草藥自行墮胎 用藥打胎 用藥打下 魚鳥蛋 芫花 芫青 樟腦 詐偽 中毒 豬膏 朱術 子麽

References Abbreviations BCGM—Li Shizhen. Bencao gangmu xin jiaozhuben (New annotated edition of Compendium of materia medica), ed. and annotated by Liu Hengru and Liu Shanyong. Beijing: Huaxia chubanshe, 1998. BX—Ba County Archive. Chengdu: Sichuan Provincial Archive. Each case cited by serial number and Chinese date. DC—Xue Yunsheng. Duli cunyi chongkanben (A new edition of Lingering Doubts upon Reading the Substatutes), ed. and punctuated by Huang Jingjia. Taipei: Chinese Materials and Research Aids Service Center, 1970. Each law cited by serial number. XT—Neige xingke tiben (Grand Secretariat routine memorials on criminal matters) (“marriage and sex offenses”). Beijing: First Historical Archive. Each memorial cited by serial number and Chinese date. Other References Baodi County (Shuntian Prefecture) Archive (“marriage, sex offenses, and family disputes”). Beijing: First Historical Archive. Each case cited by number of bundle in which it was stored in 1992 and date of earliest document. Bray, Francesca. “A Deathly Disorder: Understanding Women’s Health in Late Imperial China.” In Knowledge and the Scholarly Medical Traditions, edited by Donald Bates, 235–250. Cambridge: Cambridge University Press, 1995.

abortion in late imperial china

181

———. Technology and Gender: Fabrics of Power in Late Imperial China. Berkeley: University of California Press, 1997. Brenner, Robert, and Christopher Isett. “England’s Divergence from China’s Yangzi Delta: Property Relations, Microeconomics, and Patterns of Development.” Journal of Asian Studies 61, 2 (2002): 609–62. Brodie, Janet. “Menstrual Interventions in the Nineteenth-Century United States.” In Regulating Menstruation, edited by Etienne van de Walle and Elisha P. Renne, 39–63. Chicago: University of Chicago Press, 2001. Browner, Carole [C. H.]. “The Management of Early Pregnancy: Columbian Folk Concepts of Fertility Control.” Social Science and Medicine: Medical Anthropology 14B, 1 (1980): 25–32. ———. “Traditional Techniques for Diagnosis, Treatment, and Control of Pregnancy in Cali, Colombia.” In Women’s Medicine: A Cross Cultural Study of Indigenous Fertility Regulation, ed. Lucile F. Newman and James M. Nyce, 99–123. New Brunswick, New Jersey: Rutgers University Press, 1985. Cao Jinqing et al. Dangdai Zhebei xiangcun de shehui wenhua bianqian (Modern social and cultural change in rural northern Zhejiang). Shanghai: Shanghai yuandong chubanshe, 2001. Cao Shuji and Chen Yixin. “Ma’ersasi lilun he Qingdai yilai de Zhongguo renkou” (Malthusian theory and the Chinese population since the Qing dynasty). Lishi yanjiu (Historical Research) 1 (2002): 41–54. Chen Jiamo. Bencao mengquan (Materia medica). Beijing: Renmin weisheng chubanshe, 1988. Chen Jianfei. “Zhongcaoyao zhongdu 50 li zonghe fenxi ji wenxian fuxi” (50 cases of poisoning with traditional herbal medicines: Analysis and literature review). Jiangxi Zhongyiyao (Jiangxi Journal of Traditional Chinese Medicine) 2 (1981): 45–48. Chen Jinhua. “Banmao zhongdu siwang de jianyan wu li baogao jian lun banmao duotai de weixian” (Five cases of death by mylabris intoxication: The danger of mylabris when used as an abortifacient). Yunnan yiyao (Yunnan Medical Journal) 1 (1981): 15–16. Cheng Kok-Choi et al. “A Fatality Due to the Use of Cantharides from Mylabris Phalerata as an Abortifacient.” Medicine, Science, and the Law 30, 4 (1990): 336–40. Cheng Linan. “Medical Abortion in Early Pregnancy: Experience in China.” Contraception 74 (2006): 61–65. Cheng Maoxian. Cheng Maoxian yi an (Cheng Maoxian’s medical cases). Shanghai: Shanghai gujishudian, 1982. Ciganda, Carmen, and Amalia Laborde. “Herbal Infusions Used for Induced Abortion.” Journal of Toxicology: Clinical Toxicology 41, 3 (2003): 235–239. Cullen, Christopher. “Patients and Healers in Late Imperial China: Evidence from the Jinpingmei.” History of Science 31 (1993): 99–150.

182

sommer

Du Wenxie. Yao jian (Mirror of medicine). Hong Kong: Shangwu yinshuguan, 1976. Ebrey, Patricia. The Inner Quarters: Marriage and the Lives of Chinese Women in the Sung Period. Berkeley: University of California Press, 1993. Editorial Committee. Linchuang yiliao huli shiwu fenxi (Analysis of errors in clinical care). Yanji: Yanbian renmin chubanshe, 1986. Ernst, E. “Herbal Medicinal Products during Pregnancy: Are They Safe?” British Journal of Obstetrics and Gynecology 109 (2002): 227–35. Fan, Maureen. “In Rural China, a Bitter Way Out.” Washington Post, May 15, 2007. Fei, Hsiao-t’ung [Fei Xiaotong]. Liu shang Yaoshan (Six visits to Yaoshan). Beijing: Zhongyang Minzu Daxue chubanshe, 2006. ———. Peasant Life in China: A Field Study of Country Life in the Yangtze Valley. New York: E. P. Dutton, 1939. ———. Xiangtu Zhongguo—Shengyu zhidu (Peasant life in China; The system of childbirth). Beijing: Beijing Daxue chubanshe, 1998. Flemming, Rebecca. Review of Eve’s Herbs, by John M. Riddle. Isis 90, 1 (1999): 102–03. Furth, Charlotte. “Blood, Body, and Gender: Medical Images of the Female Condition in China, 1600–1850.” Chinese Science 7 (1986): 43–66. ———. A Flourishing Yin: Gender in China’s Medical History. Berkeley: University of California Press, 1999. Gao Wangling. “Ma’ersasi yinyingxia de renkou zhenglun” (The population debate in Malthus’s shadow). Qing shi yanjiu (Qing Historical Studies) 1 (February 2004): 15–20, 57. Gates, Hill. “Footbinding and Handspinning in Sichuan.” In Constructing China: The Interaction of Culture and Economics, ed. Kenneth G. Lieberthal et al., 177–94. Ann Arbor: Center for Chinese Studies, University of Michigan, 1997. Ge Yuanxu. Xiyuanlu zhiyi (Choice omissions in Washing Away of Wrongs). Hubei guan shu chu, 1890. Government of Hong Kong. “Bilingual Laws Information System.” http://www.legislation .gov.hk/eng/index.htm (accessed February 2, 2009). Grimes, David A., et al. “Unsafe Abortion: The Preventable Pandemic.” The Lancet 368 (Nov. 2006): 1908–19. Gui Youguang. “Xianbi shilüe” (A record of my late mother’s life). In Gui Zhenchuan shi wen xuan (Selected writings of Gui Zhenchuan), 218–21. Nanjing: Jiangsu guji chubanshe, 2002. Han, Hua. “Under the Shadow of the Collective Good: An Ethnographic Analysis of Fertility Control in Xiaoshan, Zhejiang Province, China.” Modern China 33, 3 (2007): 320–48.

abortion in late imperial china

183

Helgstrand, Stine, and Anne-Marie Nybo Andersen. “Maternal Underweight and the Risk of Spontaneous Abortion.” Acta Obstetricia et Gynecologica Scandinavica 84, 12 (2005): 1197–1201. Hershatter, Gail. Dangerous Pleasures: Prostitution and Modernity in Twentieth-Century Shanghai. Berkeley: University of California Press, 1997. Heuveline, Patrick. Review of One Quarter of Humanity: Malthusian Mythology and Chinese Realities, 1700–2000, by James Lee and Wang Feng. American Journal of Sociology 106, 2 (2000): 516–18. Hong Kong Baptist University Library. “Chinese Medicine Specimen Database.” http:// www.hkbu.edu.hk/lib/electronic/libdbs/scm_specimen.html (accessed September 1, 2009). Hou Rumiao and Ma Jiaji. “Jixing banmao zhongdu yi li baogao” (A case of acute mylabris poisoning). Tianjin yiyao (Tianjin Medical Journal) 6, 9 (1964): 739. Hu Lifa. “Shexiang zhongdu zhi jixing shen gongneng shuaijie er li” (Two cases of acute kidney failure caused by musk intoxication). Zhongguo shenzangbing zazhi (Chinese Journal of Nephrology) 2 (1994): 32–33. Hu Ping and Hu Dechuan. “Banmao zhongdu 88 li baogao” (88 cases of mylabris poisoning). Shiyong xiangcun yisheng zazhi (Journal for the Practical Rural Physician) 4, 7 (1997): 41. Huang Guoping et al. “Liuchan yaowu de linchuang yingyong” (Clinical use of abortifacient drugs). Yiyao daobao (Medical report) 15, 5 (1996): 230–231. Huang, Philip C. C. The Peasant Family and Rural Development in the Yangzi Delta, 1350–1988. Stanford: Stanford University Press, 1990. Huang Runhong. “Pingguo Xian 31 li yunchanfu siwang qingkuang fenxi” (Analysis of 31 women’s deaths in pregnancy or childbirth in Pingguo County). Youjiang yixue (Youjiang medical journal) 20, supplement (1992): 33–35. Isett, Christopher M. State, Peasant, and Merchant in Qing Manchuria, 1644–1862. Stanford: Stanford University Press, 2007. Jaschok, Maria. Concubines and Bondservants: The Social History of a Chinese Custom. London: Zed Books, 1988. Ji Xiaofen et al. “Niuxi zai rengong liuchanshu zhong de yingyong (Use of ox knee in surgical abortion). Shizhen guoyi guoyao (Traditional Chinese medicine and pharmacy) 5 (2001): 446. Jiang Qingyu and Qi Yongmao, eds. Zhongyao buliang fanying (Adverse reactions to traditional Chinese medicine). Beijing: Zhongguo Zhongyiyao chubanshe, 1995. Jin Haiyan. “Banmao zhongdu 42 li” (42 cases of mylabris poisoning). Yaowu buliang fanying zazhi (Journal of Adverse Reactions) 5 (2002): 318–319. Jin Su. “Wu zhong changyong youdu chonglei Zhongyao de duxing ji jiejiu fangfa” (Five poisonous insects commonly used in traditional Chinese medicine). Gansu Zhongyi (Gansu Journal of Traditional Chinese Medicine). 16, 8 (2003): 39–42.

184

sommer

Jin Yacheng. “Banmao neifu zhi si yi li baogao” (A case of death caused by ingestion of mylabris). Jiceng Zhongyao zazhi (Primary Journal of Chinese Materia Medica) 16, 4 (2002): 63. Jinpingmei cihua (Plum in the golden vase). Hong Kong: Xianggang Taiping Shuju, 1982. Johnson, Elizabeth. “Women and Childbearing in Kwan Mun Hau Village: A Study of Social Change.” In Women in Chinese Society, ed. Margery Wolf and Roxane Witke, 214–41. Stanford: Stanford University Press, 1975. Kline, Jennie, et al. “Fever During Pregnancy and Spontaneous Abortion.” American Journal of Epidemiology 121, 6 (1985): 832–42. Ko, Richard J. “Causes, Epidemiology, and Clinical Evaluation of Suspected Herbal Poisoning.” Clinical Toxicology 37, 6 (1999): 697–708. Kong Qi. “Duotai dang jin” (Be cautious about abortion). In Zhizheng zhi ji (Candid record of the Zhizheng era), 93. Beijing: Zhonghua shuju, 1991. Lee, James Z., and Cameron D. Campbell. Fate and Fortune in Rural China: Social Organization and Population Behavior in Liaoning 1774–1873. New York: Cambridge University Press, 1997. Lee, James Z., and Wang Feng. One Quarter of Humanity: Malthusian Mythology and Chinese Realities, 1700–2000. Cambridge: Harvard University Press, 1999. Li Bozhong. “Duotai, biyun, yu jueyu: Song Yuan Ming Qing shiqi Jiang-Zhe diqu de jieyu fangfa ji qi yunyong yu chuanbo” (Abortion, contraception, and sterilization: Fertility control and its dissemination in Jiangsu and Zhejiang during the Song, Yuan, Ming, and Qing Dynasties). In Hunyin jiating yu renkou xingwei (Marriage, family, and population behavior), ed. Li Zhongqing [James Z. Lee], Guo Songyi, and Ding Yizhuang, 172–96. Beijing: Beijing Daxue chubanshe, 2000. Li Fu-kuang. “Problems Concerning Artificial Abortion through Oral Administration of Traditional Drugs.” Studies in Family Planning 4, 8 (1973): 200. Li Ling and Huang Guangzhao. “Jixing tianhua fen zhongdu shijian er li” (Autopsy report on two cases of acute tianhua fen poisoning). Tongji Yike Daxue xuebao (Journal of Tongji Medical College). 19, 6 (1990): 417–18, 440. Li Wenzhi et al. Ming Qing shidai de nongye zibenzhuyi mengya wenti (Sprouts of capitalism in Ming-Qing agriculture). Beijing: Zhongguo shehui kexue chubanshe, 1983. Li Xia et al. “Authentication of 31 Species of Toxic and Potent Chinese Materia Medica (T/PCMM) by Microscopic Technique, Part 1: Three Kinds of Toxic and Potent Animal CMM.” Microscopy Research and Technique 70 (2007): 960–68. Liang Bingxue. “Banmao zhongdu er li baogao” (Two cases of mylabris poisoning). Zhonghua neike zazhi (Chinese Journal of Internal Medicine) 6, 7 (1958): 721. Liang Lei and Hu Yuchuan. “Banmao zhongdu yi li baogao” (A case of mylabris poisoning). Sichuan Zhongyi (Sichuan Journal of Chinese Medicine) 4 (1989): 48.

abortion in late imperial china

185

Liang Jinquan and Wang Ningsheng. “Banmao de duxing fanying ji yuanyin fenxi” (Analysis of mylabris toxicity). Xin Zhongyi (New Traditional Chinese Medicine) 35, 7 (2003): 76–77. Liao Anshui and Cheng Yuzhu. “Banmao zhongdu er li baogao” (Two cases of mylabris poisoning). Jiangxi yiyao (Jiangxi Medical Journal) 3 (1986): 254–56. Luk, Bernard H. K. “Abortion in Chinese Law.” American Journal of Comparative Law 25 (1977): 372–92. Ma Jianxin. “Changjian Zhongyao buliang fanying gai kuang” (Adverse reactions from some commonly used traditional Chinese medicines). Hebei Zhongyi (Hebei Journal of Traditional Chinese Medicine) 23, 2 (2001): 152–54. Maconochie, N., et al. “Risk Factors for First Trimester Miscarriage.” British Journal of Obstetrics and Gynecology 114, 2 (2007): 170–86. Mann, Susan. The Talented Women of the Zhang Family. Berkeley: University of California Press, 2007. Maxwell, J. Preston. “On Criminal Abortion in China.” Chinese Medical Journal, (Jan. 1928): 12–19. McMahon, Keith. Causality and Containment in Seventeenth-Century Chinese Fiction. Leiden: E. J. Brill, 1988. Moed, Lisa, et al. “Cantharidin Revisited: A Blistering Defense of an Ancient Medicine.” Archives of Dermatology 137 (2001): 1357–60. Nappi, Carla. The Monkey and the Inkpot: Natural History and Its Transformations in Early Modern China. Cambridge: Harvard University Press, 2009. Ngin, Chor-Swang. “Indigenous Fertility Regulating Methods among Two Chinese Communities in Malaysia.” In Women’s Medicine: A Cross Cultural Study of Indigenous Fertility Regulation, ed. Lucile F. Newman and James M. Nyce, 25–41. New Brunswick, New Jersey: Rutgers University Press, 1985. ———. “Reproductive Decisions and Contraceptive Use in a Chinese New Village in Malaysia.” Ph.D. diss., University of California, Davis, 1985. Nie Jing-Bao. Behind the Silence: Chinese Voices on Abortion. Lanham, Maryland: Rowman & Littlefield, 2005. Ouyang Ju and Jiang Youzhong. “Zhongyao zhongdu yuanyin fenxi ji yufang cuoshi” (TCM poisoning: Causes and prevention). Shiyong Zhongyiyao zazhi (Journal of Practical Traditional Chinese Medicine) 20, 12 (2004): 724. Pan Qiying. “Jixing banmao zhongdu yi li baogao” (A case of acute mylabris poisoning). Zhongguo yikan (The Intermediate Medical Journal) 3 (1957): 40–41. Population Council. “Reducing Unsafe Abortion.” http://www.popcouncil.org/rh/rua .html (accessed June 4, 2009).

186

sommer

Qin Yuhui and Li Peigan. “Hongcaotang kangzaoren 38 li” (38 cases using “red herbal decoction” to terminate early pregnancy). Jilin Zhongyiyao (Jilin Journal of Traditional Chinese Medicine) 3 (2001): 24. Rabl, Walter et al. “Camphor Ingestion for Abortion (Case Report).” Forensic Science International 89, 1 (1997): 137–40. Reagan, Leslie J. When Abortion Was a Crime: Women, Medicine, and Law in the United States, 1867–1973. Berkeley: University of California Press, 1997. Riddle, John M. Eve’s Herbs: A History of Contraception and Abortion in the West. Cambridge, Mass.: Harvard University Press, 1997. Rigdon, Susan M. “Abortion Law and Practice in China: An Overview with Comparisons to the United States.” Social Science Medicine 42, 4 (1996): 543–60. Ruan Qisi. Chongkan buzhu Xiyuanlu jizheng (Washing Away of Wrongs, reprinted with amendments, examples, and commentaries). 1807 edition; reprint Taipei: Wenhai chubanshe, 1968. Santow, Gigi. “Emmenagogues and Abortifacients in the Twentieth Century: An Issue of Ambiguity.” In Regulating Menstruation, edited by Etienne van de Walle and Elisha P. Renne, 64–92. Chicago: University of Chicago Press, 2001. ———. Review of Eve’s Herbs, by John M. Riddle. Population and Development Review 24, 4 (1998): 869–75. Seidlecky, Stefania. “Pharmacological Properties of Emmenagogues.” In Regulating Menstruation, ed. Etienne van de Walle and Elisha P. Renne, 93–112. Chicago: University of Chicago Press, 2001. Shang Jing. “Shexiang fengshigao tieji tigao yaowu liuchan chenggonglü de linchuang guancha” (Clinical observation of success rate in using a musk compound to induce abortion). Henan Zhongyi (Henan Journal of Traditional Chinese Medicine) 23, 5 (2003): 30. Shepherd, John Robert. Marriage and Mandatory Abortion among the 17th-Century Siraya. American Ethnology Society Monograph Series Number 6, 1995. Smith, Frederick Porter. “Chinese Blistering Flies.” In Year-Book of Pharmacy, Comprising Abstracts of Papers Relating to Pharmacy, Materia Medica, and Chemistry, Contributed to British and Foreign Journals, from July 1, 1870 to June 30, 1871, 17–19. London: John Churchill and Sons, 1871. Accessed in Google Books online, original from Oxford University, digitized Nov. 14, 2006. Sommer, Matthew H. Sex, Law, and Society in Late Imperial China. Stanford: Stanford University Press, 2000. Stockard, Janice E. Daughters of the Canton Delta: Marriage Patterns and Economic Strategies in South China, 1860–1930. Stanford: Stanford University Press, 1989. Sun Zhongguo and Guo Yingli. “Banmao zhongdu siwang wei jian chu banmao su yi li fenxi” (Analysis of a fatality by mylabris poisoning, in which examination failed to

abortion in late imperial china

187

detect cantharidin). Hebei Beifang Xueyuan xuebao (Journal of Hebei Northern University) 21, 6 (2004): 28. Tian Yanxia & Jiao Peimin. “Zhongguo gudai duotai kaolüe” (Abortion in premodern China). Yixue yu zhexue (Medicine and Philosophy) 28, 3 (2007): 64–65. Tong Yaohua. “Banmao zhongdu de qiangjiu yu huli” (Emergency treatment of mylabris poisoning). Zhonghua huli zazhi (Chinese Journal of Nursing) 2 (1989): 82. van de Walle, Etienne. “Flowers and Fruits: Two Thousand Years of Menstrual Regulation.” Journal of Interdisciplinary History 28, 2 (1997): 183–203. van de Walle, Etienne and Elisha P. Renne, eds. Regulating Menstruation: Beliefs, Practices, Interpretations. Chicago: University of Chicago Press, 2001. Wang Fengmin and Dai Zhannan. “Cui jing zhi yun tang kang zao zaoyun 34 li” (34 cases using “menstruation-encouraging and pregnancy-terminating decoction” to terminate early pregnancy). Liaoning Zhongyi zazhi (Liaoning Journal of Traditional Chinese Medicine) 10 (1993): 28. Watson, Rubie S. “Wives, Concubines, and Maids: Servitude and Kinship in the Hong Kong Region, 1900–1940.” In Marriage and Inequality in Chinese Society, ed. Rubie S. Watson and Patricia Ebrey, 231–55. Berkeley: University of California Press, 1991. Weng Xinzhi and Gao Chongji. “Banmao zhongdu—wenxian fuxi ji yi li baogao” (A case of mylabris poisoning). Zhonghua yixue zazhi (National Medical Journal of China) 43, l (1957): 48–49. Weng Yifeng. “Junji gongzhu pei shexiang duotai bu xia yi li” (A case of failed abortion using a musk compound). Fujian zhongyiyao (Fujian Journal of Traditional Chinese Medicine) 21, 2 (1990): 13. Wolf, Arthur P. “Is There Evidence of Birth Control in Late Imperial China?” Population and Development Review 27, 1 (2001): 133–54. Wolf, Arthur P. and Theo Engelen. “Fertility and Fertility Control in Pre-Revolutionary China.” Journal of Interdisciplinary History 38, 3 (2008): 345–375. Wolf, Arthur P. and Chieh-shan Huang. Marriage and Adoption in China, 1845–1945, Stanford: Stanford University Press, 1980. Wolf, Margery. “Women and Suicide in China.” In Women in Chinese Society, ed. Margery Wolf and Roxane Witke, 111–42. Stanford: Stanford University Press, 1975. World Health Organization. Unsafe Abortion: Global and Regional Estimates of Incidence of Unsafe Abortion and Associated Mortality in 2003. Geneva: World Health Organization, 2007. WHO Representative Office in China. “Maternal and Child Health in China.” http:// www.wpro.who.int/china/sites/mch/overview.htm (accessed May 4, 2009). Wu Tan. Da Qing lüli tongkao jiaozhu (Edited and annotated edition of Thorough Examination of the Qing Code), ed. and annotated by Ma Jianshi and Yang Yutang, Beijing: Zhongguo Zhengfa Daxue chubanshe, 1992.

188

sommer

Xi’an Municipal Government Population and Family Planning Committee. “Qie ji zifu datai yao” (Never take drugs to induce abortion on your own). http://www.xianfp .gov.cn/Index/Catalog119/3105.aspx (accessed January 26, 2009). Xia Liying et al., eds. Xiandai Zhongyao dulixue (Modern toxicology of Chinese materia medica), Tianjin: Tianjin deji fanyi chuban gongsi, 2005. Xie Chuanjiao. “China’s Suicide Rate Among World’s Highest.” China Daily, September 11, 2007. Xie Liansheng. “Qiantan jixing banmao zhongdu de zhengzhi” (Diagnosis and treatment of acute mylabris poisoning). Gansu Zhongyi Xueyuan xuebao (Journal of the Gansu College of Traditional Chinese Medicine) 4 (1988): 19–20. Xu Chucai. “Jixing banmao zhongdu san li baogao” (Three cases of acute mylabris poisoning). Zhonghua neike zazhi (Chinese Journal of Internal Medicine) 6, 7 (1958): 720–21. Xu Dachun. Xu Dachun yi shu quanji (Medical writings of Xu Dachun). 2 vols. Beijing: Renmin weisheng chubanshe, 1988. Xu Guojun and Chen Jinquan, eds. Xianggang changyong youdu Zhongyao tujian (Illustrated guide to toxic traditional Chinese medicines commonly used in Hong Kong). Hong Kong: Shangwu yinshuguan, 1994. Xu Lian. Xiyuanlu xiangyi (Detailed meaning of Washing Away of Wrongs). Hubei guan shu chu, 1890. Xu Lian and Xiong E, eds. Xingbu bizhao jia jian cheng’an (Cases in which the Board of Punishment judged by analogy to raise or lower penalties). Beijing: Falü chubanshe, 2009. Xun Xianwen. “Banmao zhongdu siwang er li” (Two cases of death by mylabris poisoning). Zhongguo fayixue zazhi (Chinese Journal of Forensic Medicine) 3 (1990): 156. Yuan dian zhang (Decrees and regulations of the Yuan Dynasty). Beijing: Zhonghua shuju, 1990. Zhang Cun. “Baihua dan duotai zhi baixiezheng jixing shen gongneng shuaijie siwang yi li baogao” (A case of death by septicemia and kidney failure from use of baihua dan as an abortifacient). Guangxi Zhongyiyao (Guangxi Journal of Traditional Chinese Medicine) 1 (1986): 30–31. Zhang Jizong et al. “Yong GC/MS jianding shiti zangqi zhong de banmaosu” (Autopsy use of GC/MS to identify cantharidin in organs, with analysis of one case). Zhongguo fayixue zazhi (Chinese Journal of Forensic Medicine) 3, 4 (1988): 228, 251. Zhang Junwen and Zhang Yuwu. “Zhongyao zhongdu 20 li baodao” (20 cases of poisoning with traditional Chinese medicines). Xi’an Yike Daxue xuebao (Journal of Xi’an Medical University) 7, 2 (1986): 176–79. Zhang Yigu and Huang Guangzhao. “Jiu zhong youdu zhiwu zhongdu de shijian baogao” (Autopsy report on cases of poisoning with nine species of toxic plants).

abortion in late imperial china

189

Zhongguo fayixue zazhi (Chinese Journal of Forensic Medicine) 2, 1 (1987): 20–22, 64–65. ———. “Youdu dongwu zhongdu 4 li shijian baogao” (Autopsy report on four cases of poisoning with toxic animal products). Zhongguo fayixue zazhi (Chinese Journal of Forensic Medicine) 2, 2 (1987): 114–17. Zhen Ze and Shen Zhanlin, eds. Jiangsu cheng’an (Jiangsu cases). In Lidai pan li pan du (Collected judgments from various eras), vol. 8, ed. Yang Yifan and Xu Lizhi. Beijing: Zhongguo shehui kexue chubanshe, 2005. Zheng, Zhenzhen et al. “Sexual Behaviour and Contraceptive Use among Unmarried, Young Women Migrant Workers in Five Cities in China.” Reproductive Health Matters 9, 17 (2001): 118–27. Zhu Qingqi et al., eds. Xing’an huilan san bian (Conspectus of penal cases, with sequels). 4 vols. Beijing: Beijing guji chubanshe, 2004. Zou Weibo. “Koufu banmao duotai zhongdu zhi siwang de fayixue jianding” (Forensic diagnosis of fatal poisoning caused by oral ingestion of mylabris as an abortifacient). Xiandai linchuang yixue shengwu gongchengxue zazhi (Journal of Modern Clinical Medicine and Bioengineering) 5, 3 (1999): 203.

part two

Buying and Selling of Land · Homicides

∵

chapter 6

Customary and Judicial Practices as Seen in Criminal Sales of Land in Qing Manchuria Christopher Isett In his study of Qing law, Philip Huang (1996) argues that the threat of a lawsuit might prompt intensified efforts at mediation in villages. The possibility that a case brought to court would result in fines or corporal punishment, even for minor infractions of the Code, could push disputants to an out-of-court settlement. Magistrates also understood this and may have formulated their preliminary opinions on plaints in ways that pressured disputants to such a settlement (51–57). In this way, legal norms and customary practices served to reinforce each other, but only if there was some correspondence between the two. A magistrate was unlikely to urge or sanction a settlement when the actions in dispute were criminal. Huang’s observation, therefore, raises the question of what may have occurred when customary practices and the law were at odds. How did villagers resolve disputes when they could not take their case to court because their actions were criminal? What did magistrates do when they uncovered practices that were customarily legitimate yet codified as criminal? And, what do these instances add to our understanding today of Qing legal practice (both formal and informal—in the realm of the court and the community) as a whole? This short study examines the procedures and legal ramifications of criminal sales of what I shall call collectively Qing land (both manor land 莊地 and bannerland 旗地) to commoners in Manchuria in the eighteenth and nineteenth centuries.1 The frequency with which Qing land was sold to commoners suggests that villagers (bannermen 旗人, serfs 莊丁, and commoners 民人 alike) were able to secure these transactions without the assistance of the state. In other parts of China the customary exchange of land also depended upon the actions and practices of villagers to legitimize and secure land sales. But, villagers there could also turn to the state if irresolvable disputes arose. By

1 For a full account of the region’s history across the Qing, including economic, political, and social change, see Isett (2007).

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_�08

194

isett

contrast, the case of land sales in Manchuria points to the capacity of villages to regulate custom in opposition to the state. The case of Manchuria also points to ways in which customary practices might constitute a politics of opposition to the state that is yet to be fully explored in the study of Qing legal culture. Matthew Sommer’s (2005) study of polyandry and wife-selling is an example of a similar phenomenon. Though illegal, among the poor, selling wives was an accepted survival strategy that few outside the state condemned. To be sure, the practitioners of these customary practices may not have recognized or understood their actions as political or even oppositional. These practices were not on the order of rebellion, after all. They did, however, rise to the level of what James Scott calls the “weapons of the weak,” strategies of deception and opposition to different forms of authority and power (Scott 1985). In adjudicating these kinds of cases, judges found themselves navigating between the clearly stated principles of the law and the contrary principles of local custom. Whereas the law prohibited the sale of Qing land to commoners, customary practices sanctioned and facilitated it. Moreover, these practices originated in the north China communities from where the commoner migrants to Manchuria came and where the community and the state sanctioned them.2 The tensions between the law and local practice thus placed the magistrate in a difficult position. The ubiquity of these exchanges and the fact that communities sanctioned them, that many depended upon illegally purchased land for their livelihood, and that these sales were ordinarily legal throughout the rest of China could place the magistrate in a difficult position. How was he to sentence? What rulings were in the best interest of the state and community? I suggest that magistrates adopted a strategic approach to sentencing under these conditions. They were not free to sentence as they wished, yet their decisions might take account of conditions on the ground.

2 In an effort to minimize conflict and litigation, in 1753 the state added an amendment to the Code specifying that conditional sale contracts signed within the previous thirty years continued to be redeemable, unless the contracts specified otherwise; however any land sale older than thirty years was declared irredeemable, regardless of the contract terms (Jamieson 1921, 100; Zhang 1998, 289–290; Huang 2001, 73–74; Jing 1994, 69–70). The Board of Revenue stipulated in its regulations that all contracts thereafter specify a maximum ten-year time limit on the redemption (Zhang 1998, 290). A grandfather clause was written to allow those who had legitimately conditionally sold their land between 1723 and 1753 to redeem it at the original price. After 1753, the right to redeem land was limited to ten years from the date of the contract. In both cases, however, the state supported the dian-maker’s right to pay only the original sale price.

customary and judicial practices



195

Sources and Methods

This study uses legal records from the eighteenth and nineteenth centuries in conjunction with Japanese ethnographies from the twentieth century. In recent years, many studies of civil law have drawn on the so-called “trivial matters” or xishi (細事) cases. These are non-violent disputes over such matters as debt, property, and betrothal agreements and were handled almost exclusively at the local level by county magistrates. Few xishi cases survive for northeast China, however. I have only come across such cases from the Bodune (伯都納) military yamen, stored at the Jilin City Archive (吉林市 檔案館). Therefore, this study also makes significant use of the more abundant records that have survived from the joint court sessions of the Shengjing Board of Revenue (Shengjing hubu 盛京戶部) and the Imperial Household Department in Shengjing (Shengjing neiwufu 盛京內務府). These cases reside at the Liaoning Provincial Archive (遼寧檔案館). These sessions handled non-violent criminal activity on the imperial estates (huangzhuang 皇莊) in Fengtian (奉天). In content, they come close to resembling xishi cases, though the court that handled them was part of the imperial household bureaucracy. The benefit of these court records is that the investigations focus on how land was bought and sold, the ways settlers obtained access to manor land, and the way in which others colluded to conceal these illegal activities from higher officials. There are many methodological problems associated with using trial records as windows on plebian society and social change. First, court records are not equivalent to surveys. Without independent verification, we have no way of knowing if the incidences these records relate are representative of general conditions, practices, or trends. Second, court records are by definition accounts of social conflict and often irreconcilable differences. Alone, they often leave the impression that late imperial society had no extra-legal means for resolving disputes. There are ways, however, by which we can read past the implicit interests of the court records to conduct social history. James Scott, for instance, argues that we often have little choice but to re-capture plebian life and customary practices in the very documents that those seeking to stamp out such practices create (Scott 1990, 1–16). In making the arguments that I do about social change and practices, I try to take from Qing court records the snapshots they provide of rural life and the customary practices employed by peasants and their communities to manage their affairs, without letting the implicit biases of the source influence how I understand the representativeness and efficacy of such practices.

196

isett

In making those kinds of judgments, I rely heavily on Japanese ethnographic surveys of rural conditions in the twentieth century. While reading the Qing legal cases, I make much use of the latter precisely for the sorts of insights they provide into plebian life that Qing court records often only hint at. The practice of using twentieth-century ethnographic work in this fashion was pioneered by Japanese legal historians such as Shiga Shūzō (1967), who used rural surveys of Taiwan and north China in their studies of Qing law and legal custom, and continues to be used today by Japanese historians such as Kishimoto Mio (1997, 1998). In the US, other scholars have followed suit: Arthur Wolf and Huang Chieh-shan (1980) in their study of Taiwan, and Philip Huang (1985, 1996) in his studies of the rural economy and Qing legal system. Whereas there are few sources from the Qing that shed light on rural Manchuria, twentieth-century sources make for an embarrassment of riches. Japanese colonial interests in Manchuria mean that it is one of world’s most thoroughly documented peasant societies in the twentieth century. Japanese ethnographic investigations and surveys of economic conditions of China were the product of Japan’s attempt in the first half of the twentieth century to carve out an empire in Asia. In Manchuria, most surveys were conducted in the 1930s and 1940s by the research arms of either the Ministry of Commerce and Industry of the Japanese-run Manchukuo government (1931–1945) or the Southern Manchurian Railway, a quasi-official arm of the Japanese state. What these sources show—something the Qing court records were never intended to show—is how peasants went about regulating their affairs. Drawing from first-hand interviews, the reports describe such things as how peasants bought, sold, and rented land, how they divided property among sons and handled disputes between neighbors over adjoining fields, how they planted their fields and how much labor was required to tend these fields, and how they married off their sons and daughters. I read these surveys alongside the Qing sources to obtain a much fuller sense of rural life and customary practices and to fill in information that is absent in the Qing sources.

The Criminalization of Customary Practice in Manchuria

In 1680, the Kangxi emperor commanded regional officials in Manchuria to perform several tasks: to demarcate boundaries between bannerlands (旗地) and commoner lands (民地); to repatriate to north China all commoners who refused to be relocated to designated commoner villagers (民屯); and to enforce, thereafter, the complete segregation of banner populations and commoners in Manchuria (Sudō 1944, 180–181, 197–209, 225–235, 332–342,

customary and judicial practices

197

389–400). To stabilize communal boundaries and to prevent further invasion of Manchuria’s bannerlands, in 1689 the Kangxi emperor criminalized the exchange of land between bannermen and commoners. The sale of manor land to commoners was already prohibited under provisions found in the Ming Code and adopted in the Qing Code in 1646. It remained a crime to sell Qing land to commoners until the end of the dynasty. Migration to the northeast, however, was not entirely prohibited at that time. While the Kangxi emperor banned any commoner settlements beyond the Willow Palisades, an area that covered most of central and northern Jilin (吉林) and all of Heilongjiang (黑龍江), commoners were still permitted to reside in Shengjing or Fengtian and it appears that limited migration into these areas was permissible. Circumstances changed during the mid-eighteenth century. In 1740, the Qianlong emperor responded to the increasing presence of commoners in Shengjing and Jilin by reiterating the Kangxi policy of restricting but not banning migration (GZSL 1964, 5385 [356]; QH 1992, 1000 [158]). Qianlong demanded, however, that all commoners living in the northeast to leave the designated bannerlands and estates and to register in the civilian baojia (保甲) system. The continued failure of these policies to prevent the segregation of the commoner population in Manchuria, however, pushed the Qianlong emperor to criminalize commoner migration to Manchuria in 1750 (QH 1992, 1001 [158]).

The Sale of Qing Land to Commoners

It was against the backdrop of these policies that commoners came to settle in Manchuria and acquire land. In the eighteenth and nineteenth centuries, customary practice in China maintained that land could be sold outright in a practice known as juemai (絕賣; sometimes simai 死賣) or that it could be sold conditionally in a practice known as either dian (典) or sometimes huomai (活賣). In the former case, the seller held no customary or legal claim to the physical property and could not, therefore, redeem it at a later date. In the latter case, however, the seller or dian-maker (also known as the yezhu 業主) retained both customary and the legal right to redeem the property. The dianholder was required by both the Qing Code (statute 95) and customary practice to return the land to the dian-maker on the return of the conditional sale price (Jing 1994, 67). If the dian-maker was either unable or unwilling to redeem the land, then the dian-holder could purchase the land outright by paying what was called the zhaotie (找貼), a closing payment that transferred the rights of the land permanently to the buyer. So long as the land was not redeemed, the

198

isett

dian-holder could use the land as he wished. He could, of course, work the land himself. It was often the case, however, that the dian-holder rented the land out, either back to the dian-maker or to a third party. The dianholder could even sub-dian (轉典) the land to a third party so long as the dian-maker was notified. Qing records of homicide trials and criminal investigations into illegal sales of banner and manorial land suggest that in Manchuria most if not all sales started as conditional exchanges. Indeed, so ubiquitous was the exchange of banner and manorial land by the conditional sale that the Qing authorities spoke only of “commoners having conditionally purchased banner­ land” (mindian qidi 民典旗地) in their communications on the criminal sale of Qing land to commoners. When the Qing state finally formalized the prohibition against the sale of bannerland to commoners in the Code in 1808, jurists worded the new sub-statute to criminalize only the “conditional” sale of bannerland to commoners (95.1). Bannermen and serfs had good reason to sell their land to commoners conditionally. Knowing these sales were illegal, but in need of cash, they could raise money and still hope to redeem their property before they were found out by the authorities. The conditional sale was a means to retain control over land in an economy where there were few options outside farming. The seller, of course, hoped to buy back his land if his circumstances improved. Often they did not. But, because of the long delay between the initial conditional sale and final zhaotie payment, there were opportunities available to the seller to continue to make claims against the land. Under customary practice, sellers of land could return to buyers years later to demand supplementary cash payments, on the grounds that the value of the land in question had risen since the time of the sale (Kishimoto 1997). The most common was the zhaojia (找價), which, by custom, could be secured multiple times. In Manchuria, the state regarded all of these practices illegal, from the conditional sale to the zhaojia, when the commoners bought banner or manor land. Despite regulations and laws prohibiting the sale of banner and manor lands to commoners, the practice flourished. The Shengjing military-governor Fujun maintained in 1806 that only half of all banner households in Manchuria retain control of their property (HZ 1806.n.n). How was the criminal sale of land on such a scale possible? To answer this question we need to look both at how local communities concealed criminal land sales from the state and how they regulated these exchanges given that they could not easily turn to courts to resolve their disputes. In the second half of the eighteenth century, Manchuria’s communities sought to hide the illegal sale of Qing land by maintaining two records of every

customary and judicial practices

199

land sale.3 Within the community (or at least among those who witnessed and guaranteed the sale), and between buyer and seller, these exchanges were understood for what they were: most commonly, this meant a conditional sale (典). To state officials, however, these same land sales were represented as leaseholds (佃). Under the Qianlong emperor, the Qing state permitted the renting of bannerland and manor land to commoners in Manchuria, so long as the leasehold did not exceed three years. When the magistrates or rural agents demanded an explanation for the presence of commoners on Qing lands, they would be shown the leasehold contracts. To an unsuspecting or overly burdened official a false leasehold contract might on the surface appear to be a legal transfer of use rights. Testimony given in an 1815 criminal trial for the illegal sale of imperial manor land tells us how disguising an illegal sale as a leasehold was expected to work. Commoner Xing Dexiang (興得祥), the middleman to a series of illegal sales of imperial estate land to a number of commoners, testified before the Fengtian Prefect that the buyers and sellers of the land feared being caught and so purposely substituted zu (租) for dian in the contract. It was their intention, he openly confessed, to dupe local officials and their staff into thinking the land had been rented and not sold (LPA 1815, 24129, 25944). Court records show that defendants who were under investigation for the illegal sale of Qing land produced counterfeit leaseholds to support their claims, asserting before the court that the land was not conditionally sold, but rented. Savvy magistrates, on hearing such protestations, would push their investigation further to find whether the land was truly under leasehold. They may even notice discrepancies in the contracts, or between the contracts and what was testified in court. They would then summon the defendants, neighbors and village leaders, as well those charged with village oversight, for questioning. However, judging from the complaints of higher officials, local magistrates and rural agents charged with overseeing villages were on the whole unaware of what was going on. Fujun, who had extensive administrative experience 3 The practice of keeping two accounts of illegal land sales in Manchuria appears similar in some interesting ways to the practice common among peasants in China Proper of using both “white” (白) and “red” (紅) deeds (契) (Jing 1994, 67). The Manchurian practice of hiding the sale of Qing lands to commoners was different in both intent and appearance, however. First, the “white” deed did not conceal an illegal sale of land, but it assisted in evading the land tax. Second, the language of the “white” deed did not hide the fact that the land was sold by misrepresenting the transaction as something other than a sale. What is similar about the two practices is what stood behind them, however—the belief that communities could under most conditions be expected to handle and resolve disputes without recourse to the courts.

200

isett

in the northeast, had a good understanding of what was happening. In his 1806 memorial, he described in some detail how both bannermen and commoners falsified their contracts to conceal illegally sold land as rented land. He suggested that most local officials did not see past these deceptions. But, perhaps most problematic for him, Fujun complained that the various rural agents tasked with village oversight understood the deceptions and knew land was being illegally sold, but did not report any of this to the magistrates. The complicity of the rural agents, which lent some legitimacy to these illegal exchanges, placed magistrates in a difficult position when they investigated and tried these cases (HZ 1806.n.n). A late eighteenth-century investigation and trial for the illegal sale of imperial estate land to two commoners illustrates in detail how peasants sought to evade state detection and how they deployed convention, or customary practices, to regulate property claims that the state had criminalized. The saga began in 1772 when just over two hundred mu of land of an imperial estate in Gai prefecture (蓋州) was confiscated by the state. In this case, the serf who had conditionally sold the land in question heard that local officials were conducting a general survey. The serf feared being caught and so voluntarily reported the sale to local officials, an action that saved him from corporal punishment. Nevertheless, the land in question was confiscated as bannerland conditionally purchased by commoners (LPA 1794, 8820, 8857). In 1783, Su Yi (蘇義), the son of the now deceased serf who had illegally sold the land many years earlier, successfully petitioned the regional government for its return. Upon retrieving it, Su Yi was approached by the estate’s bailiff (莊頭). The bailiff had been asked by a commoner in the neighboring village named Chen Jun (陳均) to broker an arrangement. The court records relate how Chen Jun’s father had illegally bought the land in question from Su Yi’s father. But, when it was confiscated, Chen Jun’s father was never compensated. Chen Jun then asked the estate’s bailiff to mediate an arrangement whereby he would be compensated for the financial loss suffered by his father. The arrangement would give Chen Jun the right to cultivate the land rent-free for two years. With the bailiff acting as facilitator, Su Yi agreed. The land was turned over to the Chen household for two years and it was stipulated that no rent be demanded. In other words, all returns from the land would accrue to Chen for those two years. When this arrangement came to an end in 1785, Su Yi decided that he did not want the land back. He chose instead to sell conditionally just over 100 mu of the land to Chen Jun and a little more to Chen’s brother Chen Rong (陳榮). To evade detection by the state, counterfeit contracts were drawn up that falsely represented the conditional sales as leaseholds. As required by

customary and judicial practices

201

custom, middlemen and witnesses were summoned to broker and validate the exchange and publicly formalize the exchange of property by fixing their names to the contract. Over the next decade, Su Yi sank deeper into poverty. In 1787, Su approached both Chens and asked each for a zhaojia, which he received. Su Yi’s circumstances continued to deteriorate, however. The following year, in 1788, he asked for and received a second zhaojia. In 1794, Su Yi again approached Chen, this time through a middleman, and he asked for a third zhaojia and received it. On each receipt of the zhaojia, an amendment noting the receipt was added to the original contract. But, to continue the deception, the zhaojia payments were represented as “increases in rent.” Later in 1794, penniless and destitute, Su Yi fled the manor. He abandoned his two children and his wife. The wife died shortly afterward. The example of the Sus and Chens, with their complex relationship built upon land sales, debt, and obligation over two generations, reveals just how powerful a force customary practice and convention were in the daily lives of peasants. In this case, village custom and convention sufficed first to convince Su Yi to compensate the Chens for their father’s financial loss and then persuade the Chens to make multiple zhaojia payments to Su Yi when the latter faced economic hardship. Nothing in the law required Su Yi to compensate Chen Jun for the losses his father incurred. In the eyes of the Qing courts, the original sale was criminal and the Chen family had no recognizable claim to the land. Yet, through his actions, and perhaps persuaded by the bailiff, it can be seen that Su Yi recognized that until the original purchase price was returned the Chen family retained customary claims to the land their father had purchased. The normative issue at stake—whether or not the Su family owed the Chens for their father’s losses—arose from a complex sense of outstanding debt between families, spanning two generations. The Chens successfully played upon the normative value to get remuneration. Yet these obligations seemed to flow both ways. There was nothing in the law compelling Chen Jun to make the zhaojia. Not only was the land sale that formed the basis for Su Yi’s request for zhaojia payments illegal from the state’s perspective, the Qing Code actually forbade the payment of more than one zhaojia in 1730. It was, in short, the weight of convention and custom that persuaded the Chens to make multiple zhaojia payments to Su Yi, just as convention had pushed Su Yi to compensate the Chens for their father’s losses. The circumstances of the case also illustrates the role played by the community, in this case the bailiff. It demonstrates how land sale practices common to north China were seen as legitimate by village leaders in late eighteenth-century Manchuria, and accepted by the broader community.

202

isett

These complex customary arrangements for selling land had traveled from north China into southern Manchuria as part of migrants’ baggage. Once unpacked, they spread throughout the northeast. It could not have been easy for commoners to secure possession of land in this way. First, as we have seen, there was the problem of hiding the sales from the state. Second, there was the problem of how to secure property claims without the possibility of taking disputes to the state. The amount of land sold illegally in a Manchurian community of several clustered villages could run into the thousands of mu, and the time-span over which it was sold might cover several decades and cross generations. Thus, commoners had to be able to resolve disputes over exchanges that were purposely misrepresented and that arose several years if not decades after the fact. For commoners to risk purchasing Qing land, they must have felt confident that their claims to the land would be upheld by the community, or members of it. Commoners must therefore have had sufficient faith in the community’s ability and willingness to secure their claims. What was the source of this faith? The answer appears to lie in village institutions that anchored and maintained customary practices. Throughout much of rural China lineage heads and local and village notables had roles in mediating disputes (Huang 1996, 58–61). In Manchuria, lineages were weak and fairly uncommon. They certainly held little sway over village affairs (NSS 1937, 69, table 11). Consequently, the task of securing commoner property claims in Manchuria fell entirely to village notables. These were not necessarily powerful people, in the sense that they dominated or controlled village politics. But, they were more established villagers with standing and reputations for fairness and common sense, and people with “personal prestige,” as Hsiao Kung-chuan (1960) remarked (269). These people might even serve an administrative function in the countryside, and therefore, derive some of their authority from their association with the state (Huang 1996, 127–130). In Manchuria, such people were the zhuangtou who oversaw the estates (莊), the shoubao (守保) who policed within the boundaries of the bannerlands (qijie 旗界) and banner villages (旗屯), or the xiangyue (鄉約) who oversaw the commoner communities in the commoner zones (minjie 民界) (Yamamoto 1941). A facilitator had to be more than just a villager, however. To be an effective middleman, witness, or guarantor also required a reputation as someone whose word carried weight in the community. Describing land sales in the 1880s, the US consul to Niuzhuang (牛莊) wrote: “two respectable witnesses known to both parties must sign the paper; and there must be a surety who can replace the whole sum in question, if it be found that there was a flaw in

customary and judicial practices

203

the title deeds or some other irregularity in the transaction” (CR #338, dated May 19, 1887). Facilitators had to be in good standing or “respectable.” But, for a person’s words to carry weight, it was also important that he be considered by fellow villagers to be fair and impartial (MTK 1939, 31–32, 35–36, 106; Shimizu 1945, 5–6, 35–46; Hsiao 1960, 291). Saving face was one way to understand these requirements. Because villages were often small—in Manchuria they might have comprised twenty or so households—it was necessary to minimize social friction in all affairs. Unresolved anger and frustration of a few could disrupt village life and even threaten the general consensus necessary for the proper functioning of property systems and other socio-economic relations (Liang 1996, 120–125, 149–152, 165–166; Huang 1996, 62). It is telling, therefore, that villagers considered a good mediator to be someone who ensured that all parties saved face, even if the final decision significantly favored one party over another (see Hsiao 1960, 291–292). These phenomena continue in contemporary Dongbei (Yan 1996, 125–126). To preserve the appearance of disinterest and impartiality, villages developed norms of propriety that frowned upon paying middlemen or guarantors cash for their services. Cash payment suggested that the broker could gain financially, and such motives conflicted with his duty as an honest broker of transactions and fair mediator of disputes. In the 1930s, rules required that after major transactions, such as land sales, the buyer invite the middleman to feast at the next festival or celebratory occasion such as the New Year or the wedding of a son (Shimizu 1945, 6). Of course, we can see how middlemen accumulated a form of what Pierre Bourdieu (1977) called “symbolic capital” and prestige. This could be used to try to secure loans or obtain some position of village leadership (LPA 1794, 8857). In this regard, the invitation to a celebratory feast (rather than a private meal at a home) served to recognize and reinforce publicly the middleman’s important social role in the village, giving him the symbolic capital that he could “cash in” at a later date, while still maintaining his all-important aura of detachment and fairness. The facilitator or middleman had two functions. The first was to ensure that the exchange of land followed customary norms and expectations. Facilitators were village members who understood the “dos and don’ts” of local practice (NSS 1937, 5–8,153–156; TKK 1936, 77). The unwritten convention that facilitators and guarantors were village members made good sense in an illiterate society and in a legal system that not only required deeds as proof of transaction but also required verification of the deed’s authenticity. Brokers, middlemen, and witnesses (often the same person performed more than one of these tasks) policed customary practice, guaranteeing that agreements did not violate accepted village practices (NSS 1937, 190–194; MTDG 1940, 154–155;

204

isett

Shimizu 1945, 20–21). The participation of other villagers in negotiating, guaranteeing, and witnessing arrangements also made private transactions communal record. The public nature of the process was in turn a means for ritualizing these norms in a way that helped to establish a basis for shared normative behavior in the village. We can see how this might have worked by examining the land contract. The land contract was as much a statement of rights and obligations as it was an account of how the land sale was undertaken. The text of the land sale contract, which presents the agreement as a fait accompli, and therefore irrevocable, follows the exact sequence of tasks of the middleman who, according to Japanese ethnographers, began by verifying ownership, then clarified any preemptive purchase rights (xianmai quan 先買權), measured and evaluated the land, and finally witnessed the exchange of the cash price (MTDG 1935, 155– 156; TKK 1937, 186–187, 197; Shimizu 1945, 11–15). This step-by-step re-counting of the sale, which took up most of the text of the contract, may have been considered necessary in part because of the credence this telling of the “story” lent the contract. That was certainly the rationale behind the ubiquitous statement that the seller was selling the land because he was empty handed or in need of money. Those statements were clearly irrelevant to the rights and obligations transferred by the contract, yet they endowed the contract with a sense of authenticity. Whatever the purpose of such language, the contract clearly ratified that village expectations were met and did so in language that recounted the participation of villagers who understood the shared expectations of the community (MTDG 1935, 153–164; Shimizu 1945, 14, 20–22; TKK 1937, 187–92, 196–98, 253–59; NSS 1937, 117–120, 190–194). The second function of the guarantor and middleman was to mediate future disputes. That function of the facilitator in regulating village affairs is captured in the language of the land contract, which typically ended with the following admonition that “any dispute be taken to the middleman for handling” (LPA 1761, 165; 1791, 33312; 1855, 32993; JCA 1787, 21312). Twentieth-century investigations of customary practices conducted by the Japanese show that facilitators (middlemen and brokers) were bound both by convention and village expectation to remember the arrangement and to use that memory to mediate disputes (NSS 1937, 5–8, 153–156; TKK 1936, 77). The Manchurian village did not have its own “courts” for handling contractual arrangements and there was certainly nothing comparable to “common law” in the Anglo-Saxon tradition. But it clearly did have other means for regulating its own affairs. For this, the village relied heavily on the middlemen and witnesses (Huang 1996, 52–54). In this environment, practice was knowledge, and a working k­ nowledge

customary and judicial practices

205

of how things are done, as well as a memory of how they were done in the past carried force in human affairs (Liang 1996, 120–125, 149–152, 165–166). These social practices should not be idealized as an organic and therefore more just form of village rule in contrast to state law. Village practices were no guarantee against malfeasance, bullying, and extortion. Neither were village expectations that middlemen and witnesses were impartial and fair. Despite overlapping material interests of many villagers (or at least the propertied villagers) or village leadership in maintaining customary practices in land, the asymmetry between these practices and the law always opened space for abuse of both. Village elites, particularly those employed to oversee the countryside by the state, were often in a position to abuse the poorly healed members of the village (Huang 1996, 130–135). In Manchuria, serfs and rusticated bannermen who wished to retrieve land they had sold illegally to commoners without cost might be tempted to threaten exposure of illegal sales in the hope of forcing commoners to give up their land or simply to make a further zhaojia payment. The full story of Su Yi and the Chen brothers, examined earlier, nicely illustrates just how villagers might abuse their knowledge of the divergence between social practices and the law to do violence to customary norms (LPA 1794, 8820, 8857). The case record tells us that in 1794, Su Yi’s uncle, Su Chenggui (蘇成貴), receiving word of his nephew’s flight from the estate and the death of his nephew’s wife, went to care for their two abandoned children (LPA 1794, 8857). Upon arriving at the village, he was approached by the bailiff, Gu Lianggui (雇良桂). Gu told Su Chenggui that his nephew, Su Yi, had possessed some estate land but had been forced by poverty to sell it conditionally to the Chen brothers. Bailiff Gu would later admit in court that he had grievances with the Chens and therefore suggested to Su Chenggui that he bring a lawsuit accusing the Chen’s of being rent-delinquent tenants. The lawsuit of course threatened to expose the Chens’ illegal purchase of Qing land. This, Gu Lianggui assured Su Chenggui, would be enough to convince the Chens to return the land without the redemption payment because “commoners are not allowed to conditionally purchase bannerland.” Su Chenggui brought the suit and the Chens relinquished the land without further ado. Su Chenggui, in turn, reported to the yamen that a settlement had been brokered between the parties and that he wished to drop the lawsuit against his scofflaw tenants. The magistrate was under the impression that this was a simple case of outstanding rent and found the out-of-court settlement satisfactory and did not pursue the case further. The Shengjing Board of Revenue’s Farmland Department (農田司)—the office charged with

206

isett

o­ verseeing, among other things, estate lands attached to the imperial household and the Shengjing regional government—felt there was something fishy about the whole matter, however. It ordered further inquiries. At this point in time, the situation began to unravel fairly rapidly for Su Chenggui, Bailiff Gu Lianggui, and the Chens, who were sentenced for their roles in the land sales, the subsequent cover up, and the attempt at extortion. These sorts of abuses were common enough (Isett 2007, 115–120). Yet, despite the possibility for malfeasance, a remarkable number of commoners went ahead and purchased manor and banner land knowing that this was illegal. They necessarily had a certain degree of faith in the ability of villages to secure their rights and to hide these sales from the state. But, we should not conclude that the actions of villagers were occurring independently of the law. Indeed, the opposite was true. Villagers understood that these sales were illegal. The way they concealed the sales is evidence of this. More than anything, the law, in fact, required that peasants turn to self-regulating property exchange. The law, therefore, remained a powerful force in their daily lives, even when peasants did not and could not depend upon it to secure their claims.

Rural Agents, Peasant Defiance, and the Politics of Local Compromise

How did local officials respond to these developments? In the face of continuing immigration and village defiance of the law, officials were increasingly in the difficult position of having to maintain, and in some instances restore, the mandated agrarian regime, even as efforts to do so might disrupt local social order. At times the state appears to have been aware of the dilemma these circumstances posed. The Qianlong emperor reflected in 1776 that removing commoners from Fengtian would deprive them of their livelihood and commented on the negative implications of this policy (GZSL 1964 [1023]). In a startling report, regional administrator Duo Qing (多慶) noted that dispossessing commoners of their illegally acquired land and returning it to bannermen could provoke suicides, or worse (HZ 1805, [33. 2806–2808]). What could be “worse” was not said, but the implication of Duo Qing’s report is fairly clear. Duo Qing feared social unrest. Despite Qianlong’s misgivings, the prohibition against commoners occupying Qing lands persisted and the state continued to exert a significant amount of energy to this goal (Isett 2007, 130–137). Magistrates found themselves negotiating between the law and village practices. On the one hand, retaining the sovereign’s hegemony over the region remained the state’s primary objective, underwriting the Qing government’s

customary and judicial practices

207

drive to preserve Qing lands and prohibit commoner migration and settlement across the eighteenth and nineteenth centuries. On the other hand, regional and local officials also realized that securing state control locally might on occasion require striking a compromise with peasants in order to dissipate brewing troubles before they grew out of hand, or to preclude problems from arising altogether. In the final analysis, the loss of local control was potentially as much a threat to the Qing agenda as was the constant flow of illegal migrants into the northeast; as a result, at times political hegemony was best secured by granting concessions in order to prevent social unrest. A case of the criminal sale of manor land in Jilin illustrates the conflicted position in which local officials found themselves. According to the case record, the extent of illegal land deals on the Dengerlezhe (登額勒折) estate came to the attention of the Bodune military yamen in 1791 when the Bodune military governor, traveling through on inspection, asked bailiff Yu (余) whether his estate’s land was sufficient to meet its immediate needs and rent payments. Yu admitted that one-third of the 18,000 mu of land allotted to the manor had been illicitly sold to commoners by the estate’s serf population (JCA 1795, 65418). He was ordered to rectify the situation. When bailiff Yu moved to confiscate the land, and gather the title deeds peasants had used to record their sales, he was sued in court by seventeen commoners. They accused him of seeking to take their land forcibly. The lawsuit itself was quite remarkable and was an audacious move by the commoners. Commoner Hou Duo (候鐸), the principal plaintiff in the case, argued that he had possessed land on the estate for some time. He stated that he had arrived in Xincheng county in 1776 with his daughter and sons and purchased, by conditional sale, three “sections” of land from the serf Li Fang (李芳). In 1777, knowing that Hou had conditionally bought the land, bailiff Yu informed Hou that he was raising the rent on the estate land and that Hou would have to pay.4 Hou protested that since he had purchased the land conditionally he should not have to pay the rent. Bailiff Yu told Hou to take the matter up with serf Li Fang. Hou did this, telling Li that he was reluctant to pay the increased rent since he did not own the land outright. Li responded that he had no intention of redeeming the land and that Hou Dou should “raise the tax and acquire the land.” Hou Duo did so and, under the assumption that the land was now his and with Bailiff Yu’s knowledge. 4 Serfs were required to work and pay rent on their lands. The bailiff was charged with collecting this rent and was subject to punishment when he failed to meet his estate’s rent quota. When land was transferred to commoners, the bailiffs felt pressured to continue to collect the rent from commoners in order to meet their quotas.

208

isett

Having paid the rent on the land for several years, Hou Duo was angered when Bailiff Yu came to confiscate the land on the orders of the military governor. Hou and the other sixteen commoners living on Denglezhi manor brought their accusations against Bailiff Yu at the Xincheng prefectural yamen. The prefect, however, was less than sympathetic to their cause. He had the commoners beaten. To the prefect, the commoners’ case was not simply without merit, but criminal. As far as he was concerned, the land belonged to the state, the commoners were forbidden by law to possess it, and the bailiff was acting within his duties to confiscate it. Yu was ordered to continue with the confiscation of the land and the eviction of the commoners. Bailiff Yu, however, hesitated to implement this new command. He pointed out that if the land were returned to waste, the manor would be left with insufficient arable land to meet its rent quota. He also reported that since the suit was filed, the commoners had already prepared and planted the fields and the crops were just beginning to push through the topsoil. Fearing a disturbance if the land were confiscated, Bailiff Yu reported that he dare not make a move. This fear was well grounded, since Hou Duo and the other commoners had already demonstrated incredible nerve by filing their suit against him. Searching for a solution, Bailiff Yu suggested that the situation be resolved by granting the commoners permanent ownership of the land, and thereafter, simply collect the land tax from them, as was the state’s practice in China proper. Bailiff Yu was becoming an advocate for the commoners’ position rather than an enforcer of state policy. The case picks up in 1795, four years after the bailiff had been ordered to remove the commoners. At this time, the commoners were still farming their illegally purchased fields. A newly arrived prefect re-opened the case. The commoners were to return the red-registry land (紅冊地) without receiving cash compensation, but could hold onto the surfeit land (余地) for another five years without having to pay any rent. The former land was the original allotment given to the serf households when they were attached to the estate. It was seen as part of their patrimony and essential to their ability to farm and pay rent. The latter was land that serf households had opened from waste on their own initiative. It was considered less essential to the running of the estate. The new prefect argued that this arrangement would allow the commoners to recover the money they had spent when they purchased the land. The arrangement would also provide land to serfs who had returned to the estate, having heard the commoners were being evicted. The court record ends there. The case of Hou Duo illustrates how customary practice served on the one hand to articulate and establish claims to land within the village, and on the

customary and judicial practices

209

other hand, to challenge state power in the ongoing contestation over land. It demonstrates the ways in which rural agents such as zhuangtou not only recognized illegal purchases, but might have also become advocates for commoners when illegal land sales went to court. The case shows further how local and regional officials might have had to compromise and in so doing, make resolutions that did not adhere to the Code. The case of plaintiff Kong Xingdai (孔興岱) is another example of the same dilemma facing regional officials. In 1785, serf Kong Xingdai, who was attached to an imperial estate of the Plain Yellow Banner of the Imperial Household Department, went before the Shengjing Board of Revenue to reclaim land that had been confiscated from his family as “bannerland conditionally purchased by commoners.” In court, Kong Xingdai described how his father had divided the family holding of estate land into three equal shares between himself and his two brothers. He told also how the three Kong brothers gradually sold their inheritance to a number of commoners. Kong Xingdai claimed that in 1771 he and his brothers voluntarily reported the land to local officials as “bannerland conditionally purchased by commoners.” There was a survey of manor land underway at the time and by reporting the land voluntarily he claims they hoped to avoid punishment. Kong stated that the land was then confiscated and that he and his brothers were given ten years to repay the redemption price and thereby retrieve the land from the state (LPA 1786, 8499, 8508, 8512). After recounting this tale, an investigation was ordered to verify Kong’s story, which was soon found to be false. It was reported to the court that between the winter of 1751 and the winter of 1763, Kong Xingdai, his two brothers, and five other members of the Kong family had indeed conditionally sold 1,332 mu of red-registry and surfeit land to eight different commoners. In every land sale, the Kongs and the commoner buyers drew up false leasehold contracts to disguise the sales as leaseholds and in none of the cases did the Kongs redeem the land (LPA 1786, 8499). It was found, however, that the Kongs had in fact never reported the land to the state in 1771. Moreover, it was found that as early as 1766 the local yamen had uncovered these illegal sales. But, rather than confiscate the land, as would have been appropriate under the policy of the time, the magistrate had decided to allow the commoners to continue farming the Kongs’ land. He even had them pay the manor rent due on it. He then falsified yamen records to cover up this fact (LPA 1786, 8508). After uncovering these facts, the Shengjing Board of Revenue contemplated what to do. What complicated matters was the fact that the commoners in this case, as in that of Hou Duo, had been paying rent on the land for nearly twenty years. It was, at first, suggested that the illicitly exchanged red-registry land be forfeited entirely and the commoners receive half of the original sale

210

isett

price as compensation. It was also recommended, however, that the commoners be allowed to continue to cultivate the surfeit land, and thereafter, pay land taxes on it. The case was finally resolved by the forfeiture of all of the illicitly exchanged red-registry land without the payment of a redemption price but the granting to commoners of permanent rights to the surfeit land. The judge justified this last decision on the grounds that the commoners had been working the land for a long time and that to remove them would be excessive. To facilitate the payment of taxes, the judge ordered the acreage and location of each piece of surfeit land registered under the name of its new proprietor and that this registry be stored at the prefectural yamen (LPA 1786, 8499, 8521). In the face of continuing immigration, regional and local officials were increasingly in the difficult position of having to maintain, and in some instances restore, the mandated agrarian regime, even as efforts to do so often threatened to disrupt local social order. At times the state appears to have been aware of the dilemma posed, as when the Qianlong emperor reflected in 1776 that removing commoners from Fengtian would deprive them of their livelihood, with all the negative repercussions that it would entail (GZSL 1964 [1023]). The state sought to deal with this problem by occasionally ­compensating commoners whose land was confiscated (Isett 2007, 132–137). The settlement of the Kong Xingdai case, however, went further when it let the commoners become proprietors of a portion of the land they had purchased. In so far as customary norms and practices diverged from or challenged the mandates of the state, these norms supported an alternative vision of how to organize day-to-day social, political, and even economic life. Confronted with this alternative vision, the state reacted. On the one hand, retaining the sovereign’s hegemony over the region remained the state’s primary objective, underwriting the Qing government’s goal to preserve Qing lands and prohibit commoner migration and settlement across the eighteenth and nineteenth centuries. Regional and local officials, however, also realized that social stability might require striking a compromise to dissipate brewing troubles or to preclude problems from arising altogether. Social disorder was potentially as much a threat to the Qing agenda in Manchuria as was illegal settlement. At times political hegemony was best secured by granting concessions.

Adjudication in the Face of Criminal Customary Acts

Many of the details examined in this study of law and customary practice are specific to the region and cases examined. Yet, they resonate elsewhere. Magistrates throughout Qing China found themselves presiding over actions

customary and judicial practices

211

that were legally criminal yet customarily acceptable. Again, Sommer’s (2005) examination of wife sales is a clear example of this. When handling these sales, magistrates might follow the letter of the law. But, on occasion, they would also expressed sympathy with the plight of men and women forced into these exchanges by poverty and resolved cases in ways that contradicted the statutes. We cannot fully understand why in some of the cases examined here judges in Manchuria passed sentences that accorded to the law, while in other instances they chose to make accommodations to local practices. Court statements that explained decisions were not required and when given they were often thin on substance. They certainly cannot be taken at face value given that judges wrote their decisions to justify their actions to their superiors. It is possible, however, to think of sentencing typologies or orientations when examining court outcomes. Any single magistrate might manifest different orientations when confronting contradictions between legal principles and customary practices depending on the circumstances at hand. These ranged from a strict adherence to the Code to finding pragmatic solutions to the problem. They could include the following: The letter of the law: When conditions favored it, we might assume a judge would apply the letter of the law. In the case of the decisions discussed here, this was more common when sentencing was unlikely to cause significant discontent and dislocation of the population. Strict sentencing was also more likely when malfeasance, extortion, or corruption were involved and the judge felt it necessary to rectify the behavior of rural agents such as the zhuangtou, shoubao, or xiangyue. This was how the judge acted in the case of Su Chenggui and Bailiff Gu’s attempt to extort the Chens. Su and Gu were punished for bringing false lawsuits, the Chens were punished for illegally buying manor land, Su Yi was sentence in absentia for illegally selling manor land, and the witnesses to these land sales were punished for facilitating illegal land sales. The land was confiscated. In such cases, the judge stated his “rationale” by simply declaring the nature of the guilt and providing a punishment. The intent of the law: On most occasions of illegal land sales to commoners, we find that judges passed sentence only partly in accordance with the law. The most common practice was to confiscate the property but not inflict corporal punishment. Sometimes, as we have seen, commoners were “compensated” for the loss of their land by being permitted to continue farming for a period of time. From the judge’s perspective, the confiscation of the property was in itself significant punishment. Those punished were considered to be all the more pitiable for it. The sad condition of the disposed

212

isett

was sometimes given as rationale for not inflicting further punishment. This language, of course, echoed ideological representations of the empathetic Confucian state. Passing sentence in this way was especially easy when only small amounts of land and a few households were involved. There was no fear of social unrest and little worry for the livelihood of the few households now dislocated. They might be sent back to north China, shunted off to a commoner village, or even passed on to some other jurisdiction and another official. This sentencing accorded to the primary intent of the law, which was to keep manor and banner land out of the hands of commoners. Compromise and pragmatism: There were times, however, when the local official was stuck between a rock and a hard place. The law required both punishment and the confiscation of property. Yet, large numbers of people were found to have engaged in the criminal exchange of large amounts of Qing land. In these instances, dispossessing commoners would leave many destitute and discontented. When this happened, judges might find themselves burdened with a large number of lawsuits, or perhaps growing resentment and social unrest. Magistrates would then seek out a compromise that skirted the letter of the law, in some cases by a wide berth. Commoners might be permitted to remain on part of the land for a defined period of time. Or, the judge might decide to permit commoners to keep farming part or all of their illegally purchased land in perpetuity, and register the land for taxation. In such cases the magistrate could explain his decision again by drawing on ideological discourse that presented the official and the emperor as a “parent” to his subjects, capable of compassion and sternness. In each case, the decision was strategic rather than principled. The important point from the perspective of the courts was that officials always preserved the state’s prerogative to act in its own interests, thus preserving its ability to decide how it was going to manage these problems as they arose. The courts might yield to those standing before it, yet they always reserved the right to evict the next batch of illegal commoners that came along. This flexibility allowed the state to maintain its principles in a way that did not undermine its authority. Ideologically speaking, either action might be justifiable. On the one hand, professional norms, buttressed by neo-Confucian emphasis upon loyalty, obligated the official to apply the full letter of the law and will of the emperor. Conversely, neo-Confucian normative thought committed the bureaucrat to securing the “people’s welfare” (minsheng 民生), which meant at the very least ensuring access to food, clothing, and shelter (Rowe 2001, 187–189). These two norms could clearly work at cross-purposes. The fact that the officiating

customary and judicial practices

213

magistrate could draw upon either discourse, however, allowed him to dress what was often a strategic decision dressed up in normative cloth. That Qing magistrates approached adjudication flexibly in these and other cases should surprise no one. For all its sophistication, the pre-modern bureaucracy could not enforce judicial conformity despite its pretensions. That amount of supervision was beyond its capacity. Moreover, when it came to the day-to-day business of local courts, the Qing Code was quite often inadequate to the task. If we can say that the Code was very clear on what constituted a crime, we must also acknowledge that magistrates regularly applied its statutes and substatutes by analogy. The Code simply did not grasp the full complexity of the society it purported to describe. To be sure, legal practice recognized that misalignment in part by allowing for xishi. But, this did not help magistrates who regularly encountered deeply embedded customs, which were widely if not universally accepted by the local population, yet also rubbed against the Code. In these instances, the magistrate was left balancing his duty to enforce the Code, as it was written, and his desire to ease social conflict and so maintain local order. The two goals often worked at cross-purposes and the skillful magistrate weighed the possible consequences of his actions for the state, the local population, and his own career. References Legal cases from the Liaoning Provincial Archive, Shenyang, are cited as LPA. These cases were part of the Shengjing neiwufu (盛京內務府) collection. Materials from the Jilin City archive are cited as JCA. Memorials from the First Historical Archives’ Board of Revenue collection are cited as HZ. The year a document was created and the archive’s specific catalogue number are indicated following the appropriate abbreviation. All statutes in the Qing Code are referenced following Xue Yunsheng’s Duli cunyi. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. CR. United States Consular Reports, United States. Department of State Consular Dispatches: Newchwang, 1875–1906. Record Group 59.2.2 (Consular Correspon­ dences), National Archives Microfilm Publication No. 115, 7 reels. GZSL. 1964. 大清高宗皇帝實錄 (The veritable records of the Great Qing Gaozong Emperor). Reprint. Taibei: Taiwan huawen shuju. Hsiao, Kung-chuan. 1960. Rural China: Imperial Control in the Nineteenth Century. Seattle: University of Washington Press.

214

isett

Huang, Philip C. C. 1985. The Peasant Economy and Social Change in North China. Stanford: Stanford University Press. ———. 1996. Civil Justice in China: Representation and Practice in the Qing. Stanford: Stanford University Press. ———. 2001. Code, Custom, and Legal Practice in China: The Qing and the Republic Compared. Stanford: Stanford University Press. Isett, Christopher Mills. 2007. State, Peasant, and Merchant in Qing Manchuria, 1644– 1862. Stanford: Stanford University Press. Jamieson, George. 1921. Chinese Family and Commercial Law. Shanghai: Kelly and Walsh Ltd. Jing Junjian. 1994. “Legislation Related to the Civil Economy in the Qing Dynasty.” In Civil Law in Qing and Republican China. Ed. Kathryn Bernhardt and Philip C. C. Huang. Stanford: Stanford University Press, pp. 42–84. Kishimoto Mio 岸本美緒. 1997. “明清実態における‘找價回贖’問題” (The problem of “zhaojia” and “huishu” in the Ming and Qing). Chūgoku: Shakai to Bunka 中国: 社会と文化 12: 263–293. Liang Zhiping 梁治平. 1996. 清代習慣法: 社會與國家 (Qing customary law: society and state). Beijing: Zhongguo zhengfa daxue chubanshe. MTDG. 1935. 満州農村の実態: 中部満州の一農村において (Actual village conditions in Manchuria: regarding one village in Central Manchuria). Shinkyō (Changchun): Manshū teikoku daidō gakuin. MTK. 1939. 満州土地関係調査報告集錄 (Report on the investigation of Manchuria’s land relations). Shinkyō (Changchun): Chiseki seirikyoku. Niida Noboru 仁井田陞. 1943. 支那身分法史 (A history of China’s status law). Tokyo: Zayuhō kankōkai. NSS. 1937. 農村社会生活篇: 1934 年農村度実態調査報告書 (On village social life: report on the 1934 investigation of actual village conditions). Shinkyō (Changchun): Jitsugyōbu rinji sangyō chōsakyoku. QH. 1991. 清會典事例 (Collected statutes of the Qing). Beijing: Zhonghua shuju. Rowe, William T. 2001. Saving the World: Chen Hongmou and Elite Consciousness in Eighteenth-Century China. Stanford: Stanford University Press. Scott, James. 1990. Domination and the Arts of Resistance: Hidden Transcripts. New Haven: Yale University Press. ———. 1985. Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven: Yale University Press. Shiga Shūzō 滋賀秀三. 1967. 中国 家族法の原理 (The principles of family law in China). Tokyo: Sōbunsha. Shimizu Kinjirō 清水金二郎. 1945. 契の研究: 満支土地慣行規範において (A study of contracts: regarding the standard land norms of Manchuria and China). Kyoto: Daigadō.

customary and judicial practices

215

Sommer, Matthew H. 2005. “Making Sex Work: Polyandry as a Survival Strategy in Qing Dynasty China.” In Gender in Motion: Divisions of Labor and Cultural Change in Late Imperial and Modern China. Ed. Bryna Goodman and Wendy Larson. Lanham, MD: Rowman and Littlefield, pp. 29–54. Sudō Yoshiyuki 周藤吉之. 1944. 清代満州土地政策の研究 (Research into state policy toward Manchu land during the Qing). Tokyo: Kawade shobō. TKK. 1937. 土地関係並びに慣行篇: 1934 年南満州農村調査報告書 (On property relations and practices: report on the 1934 investigation of actual village conditions in southern Manchuria). Shinkyō (Changchun): Manshū tosho kabushiki kaisha. Wolf, Arthur P. and Huang Chieh-shan. 1980. Marriage and Adoption in China, 1845– 1945. Stanford: Stanford University Press. Xue Yunsheng 薛允升. 1970. 讀例存疑 (Lingering doubts while perusing the substatutes). Taipei: Chinese Materials and Research Aids Service Center. Yamamoto Yoshizō 山本義 三. 1941. “旧満州における鄉村土地の形態” (The form of village control in traditional Manchuria). Mantetsu chōsa geppo 満鉄調査月報 11: 1–56. Yan, Yunxiang. 1996. The Flow of Gifts: Reciprocity and Social Networks in a Chinese Village. Stanford: Stanford University Press. Zhang Jinfan 張晉藩. 1998. 清代民法論 (General account of civil law in the Qing). Beijing. Zhongguo zhengfa daxue chubanshe.

chapter 7

Guoshi Killing: The Continuum of Criminal Intent in Qing and Republican China* Jennifer M. Neighbors Criminal negligence—a concept of negligence that first appeared in continental European legal codes beginning in the sixteenth century (Fletcher, 1971: 415)—is a category of culpability that occupies the gray area between intent and accident in modern legal systems. Crimes committed with negligence involve a low level of mental culpability in that the perpetrator either did not exercise adequate caution in his or her actions or ignored the risk those actions posed to the well-being of others. Previous Western scholarship on Chinese law has largely ignored this portion of the graded intent continuum. Only Benjamin Wallacher, in a 1983 article, has delved into the complexity with which late imperial Chinese law conceptualized intent by positing the category of killing at play 戲殺 as roughly equivalent to the Western category of negligence (1983: 315–16). Both Wallacher and A. F. P. Hulsewé have shown that Chinese law as early as the Han dynasty contained finer gradations of punishable acts than its Western contemporaries: in addition to acts committed with intent, Han law and later dynastic law also punished unintentional but nonetheless culpable crimes that approximated later Western notions of negligence (Wallacher, 1983: 261– 67, 316; Hulsewé, 1955: 252–63). To be sure, the concept of negligence, part of a finely graded homicide continuum, was present in Chinese law since early times. But scholarship has not acknowledged the true complexity with which Chinese law grappled with abstract concepts. Criminal intent was conceptualized along a finely graded continuum of severity, from premeditated homicide 謀殺 and sudden intent

* First published in Modern China, August 13, 2013, doi:10.1177/0097700413497550. mcx.sage pub.com. Among the many people who have commented on this article in its various stages, I would like to thank in particular Matt Sommer, Bill Rowe, Philip Huang, Kathryn Bernhardt, Lisa Tran, and Margaret Kuo. I would also like to thank Lotus Perry and Elizabeth Chen for their advice on some especially tricky passages in my cases, as well as the scholars (especially Huang Yuansheng) and archivists who helped me to access case materials.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_009

guoshi killing

217

homicide 故殺 on the high end down to guoshi 過失 homicide, the least blameworthy of all homicide offenses under Chinese law, at the low end of the scale. Homicides committed through guoshi included those committed purely by accident as well as those committed through low-level negligence. In addition to the guoshi statute, late imperial Chinese law also included a variety of individual provisions, entailing increasingly serious punishments, that mapped out a finely graduated scale of increasing mental culpability. Criminal negligence was not a single category of offense under late imperial Chinese law. Chinese law addressed the concept of criminal negligence, but it did so by degrees. Beginning with the first Republican-era criminal code in 1912, the host of statutes that late imperial law had used to cover crimes of negligence was condensed into one general negligence statute, retaining the guoshi moniker that had previously been used to cover not only a portion of negligence crimes but also those committed by accident. This article will examine these ideological and statutory changes at work in actual homicide cases, revealing a Republican-era court system struggling to adopt and to adapt a series of legal codes less attentive to fine gradations of criminal culpability than their late imperial predecessors had been. Looking at Republican- and Qing-era guoshi crimes in comparison reveals that the treatment of criminal intent in Qing law was highly complex, more nuanced than that of Republican law, and more nuanced than that of many modern legal systems of the West. When considering the conceptual nature of Chinese homicide law, our common assumptions about the advancements and advantages of modern Western legal models are in need of refinement.

Guoshi Killing in the Qing Dynasty

Guoshi killing was the homicide offense with the lightest punishment under late imperial Chinese law.1 It encompassed a broad swath of criminal behavior in that it covered crimes committed with two states of mind: lack of intent (accidental crimes) and low-level negligence.2 Below I will discuss both aspects 1 See Xue Yunsheng’s commentary on the code: “Guoshi is the lightest of the six homicides” 過失乃六殺中之最輕者 (Xue, [1905] 1970: A. 292.11). 2 The term “guoshi” has been given several different English translations in the past by those examining late imperial criminal and homicide law. Most scholars have chosen to translate it as “accidental” (Alabaster, [1899] 1968; Bodde and Morris, 1967; Johnson, 1979, 1997; MacCormack, 1988, 1990, 1996). Wallacher, 1983, refers to guoshi as “non-culpable strict

218

neighbors

of guoshi killing in Qing law, as well as non-guoshi statutes that covered homicides committed through negligence. Guoshi Killing as Accidental Killing In this section, we will examine guoshi killings that fell under the definition of accidental acts. It is important to note that accidental deaths were usually considered the result of accidental crimes, and not accidental acts, under late imperial law. In other words, despite the absence of any criminal intent, the act was still considered a crime. The tenor of the guoshi statute in the late imperial codes and their commentaries emphasized the accidental nature of guoshi. Consider the following definition given to the term guoshi in codes of the Tang, Song, Ming, and Qing: Guoshi means that which the ears and eyes do not reach, that which contemplation does not attain 耳目所不及, 思盧所不到. For example, shooting wild animals, or for some reason throwing away bricks and tiles, and unexpectedly 不期 killing somebody; or, because one is climbing and in a dangerous step, slips and falls and involves others; or if riding a boat pushed by the wind, or riding a horse that starts; or driving a cart downhill and one’s strength is not adequate to stop it; or lifting a heavy object with others and one’s strength is not enough to support it such that one of the other lifters is injured; in all [of these instances] there is initially no harmful intention 凡初無害人之意, but by chance 偶 there results a person’s injury or death. (Xue, [1905] 1970: A. 292.00; modified from Great Qing Code, 1994: 19, 278, and Alabaster, [1899] 1968: 261.) The Tang and Song codes elaborated on these definitions, noting that “that which the ears and eyes do not reach” meant that “the ears do not hear sounds of people, the eyes do not see people coming out” 耳不聞人聲, 目不見人出 (Tanglü shuyi, 1996: 289; Song xingtong, 1984: 410; Johnson, 1997: 384). In addition, “that which contemplation does not attain” meant the incident occurred in “a secluded place, a place where there ought not to be people” 幽僻之所 liability” or “trespassory slip-up.” Marinus Meijer, 1978, translates it as “by mishap,” a translation that still fails to convey the full meaning of the Chinese term. As MacCormack, 1988: 68, notes, “no English term is an adequate rendering of guoshi. It covered cases both of death caused by accident without anyone’s fault and those of death caused through someone’s carelessness.” As a result, in the discussions that follow I will always include the Chinese term when discussing guoshi.

guoshi killing

219

其處不應有人 (Song xingtong, 1984: 410; modified from Johnson, 1997: 384). The Qing Board of Punishments echoed these definitions in a 1771 ruling: “All matters of guoshi arise out of the unexpected” 意外 (Xing’an huilan, 1886: 31.39b). Huang Liuhong, a county magistrate in Shandong and Zhili (Hebei) provinces in the early Qing, concurred, calling an incidence of guoshi “a case of pure accident; [it is] understood [to be] a case of which no sufficient previous warnings could have been given” (Huang, 1984: 335). Since a guoshi killing was unexpected, the exact means by which the victim died was irrelevant. According to the Board of Punishments, “It matters not if the deadly weapon is scalding water or fire, a metal blade, hand or foot, or some other object. One should always evaluate 核 whether or not it was a situation of ‘that which contemplation does not attain’ ” (Xing’an huilan, 1886: 31.42a). In the eyes of the Board, the lack of a mental element to the crime was of paramount importance. As a result, monetary redemption was available to all persons found guilty of accidental killing, and not just those who would normally be eligible for redemption, such as the only sons of elderly parents, the elderly themselves (over age seventy), and officials. Consider the following trio of guoshi homicide cases from the Beijing municipal area. All three cases involved runaway carts, and all three illustrate use of the guoshi homicide statute to prosecute accidental acts. The first case took place in Beijing in 1874. Yao Pengyuan, who worked as a water bearer, was crushed to death by the runaway cart of city resident Chong Gui (Xianshen anjian: Jiangsu 9421). According to the testimony of the latter: In the evening on the fifth day of the twelfth month, I was driving an empty [horse] cart. Upon reaching Wudaoying, I suddenly encountered a man carrying hides. . . . The horse’s eyes went back, [and] it was startled and ran. I hurried to rein [the horse] in but could not, and [the cart] turned over. Yao Pengyuan, who was pushing his water cart nearby, couldn’t get out of the way, and was knocked down by the horse. The cart wheel crushed him in the right ribs and such places . . . I caught up and helped Yao Pengyuan up. . . . Unexpectedly the next morning he died. . . . I ask only for compassion. All that I have stated is true. A street watchman in Wudaoying, Fu Lu, witnessed the accident and corroborated Chong Gui’s account. The Board of Punishments ruled the case one of guoshi killing, giving the following rationale: “Because the horse was startled and ran, the criminal suddenly could not rein him in and [the cart] turned over. . . . It seems that this was a situation of that which the ears and eyes do

220

neighbors

not reach, that which contemplation does not attain. There was no intention to harm, but by chance a person died.” The court thus ruled this case one of accident: Chong Gui did his best, but his horse was startled due to unforeseeable circumstances. In accordance with the statutes, the court issued a standard guoshi verdict: death was the initial sentence, but monetary redemption of punishment was to be allowed, with the money going to the victim’s family. Following codified Qing law, though there was no intent and minimal blame in this guoshi case (seen in the automatically redeemable death sentence), there was nonetheless a degree of responsibility assigned to the perpetrator (paying a fine to the victim’s family). In paying the victim’s family, the actor was still punished. The initial death sentence also allowed a symbolic nod to the principle of requital-in-kind 抵命, the notion that when a person killed, his life would be taken in return.3 In this case the court quoted directly from the guoshi law when providing a rationale for its sentence. The court did so routinely for guoshi cases in which the circumstances were deemed accidental and in which the actor was deemed as innocent as was possible for a convicted homicide offender under the Qing judicial system. In 1897, for instance, the Board reviewed the case of Tian Derun, whose runaway mule crushed to death Zheng Da, a middle-aged cook in a local factory (Xianshen anjian: Jiangsu 9434). Tian made a living transporting goods in a mule cart. On the day in question, he was passing through the Zhengyang Gate in Beijing when the cart’s rope broke. According to his testimony, “The mule’s eyes went back and startled, he ran. I was not able to hold him fast.” Tian fell from the cart after himself sustaining an injury and he ended up on the ground, unconscious. He quickly recovered, but in the meantime the mule had run off through the gate. Zheng Da was just outside the gate talking to a friend, Chen Er, who testified, “Suddenly from inside the gate came a mule cart without a driver. I ran clear, [but] Zheng Da had nowhere to go and was knocked to the ground by the cart shaft.” Zheng Da died, and the coroner ruled the cause of death to be injuries received when struck by the cart. When making its ruling, again the court quoted from the guoshi statute: “This was a situation of that which the ears and eyes do not reach, that which contemplation does not attain. There was no intention to harm, but by chance a person died.” Tian Derun’s mule was spooked and knocked him to the ground. He had no way to regain control of his cart. Zhang Da’s death was an accident. Tian Derun was sentenced to death but was allowed to redeem his punishment 3 The notion of requital-in-kind was well on the decline, at least juristically, in the late imperial era. See Meijer, 1980; Huang, 2002; and Neighbors, 2004.

guoshi killing

221

monetarily, as per the law. In this case, he had to sell off his mule and cart to the nephew of the victim to obtain the necessary funds. Again we have a person found guilty of a crime and forced to pay restitution for an accidental act. And in this case, paying that fine was a harsh penalty indeed—Tian Derun was a deliveryman and had to sell his means of livelihood to redeem his punishment. A 1904 Beijing case of accidental-guoshi homicide involved the death of Lu Yongtai, an elderly man engaged in small-scale trade, at the hooves of a runaway mule cart being driven by a commoner named Li Wang’er (Xianshen anjian: Huguang 13602). Li had come into the city to make a delivery, and was driving his reloaded two-mule cart home with a new load when out of the blue one of the mules bit the other. The mule on the receiving end bolted, and soon both mules were dragging the cart along, with Li Wang’er unable to control them 揪拉不住. Thereupon, they came across the hapless victim. Li Wang’er testified, “[I] called out for him to get out of the way 躲避. Unexpectedly, Lu Yongtai did not hear me and continued on, with the result that the mule load . . . ran him down.” The details of this account were confirmed by a local patrolman, and the autopsy report found no evidence of foul play. The court made inquiries of Lu Yongtai’s son, Lu Dianyou, and learned that the victim was not only nearsighted, but deaf as well, so he likely heard neither the warnings nor the clamor of his approaching demise. As in the cases above, the court ruled that “it really was a case of that which contemplation does not attain,” and that since there was no prior enmity between the parties, and since the death was brought about accidentally 偶, a sentence of guoshi killing was issued, with monetary redemption of punishment allowed.4 Even though the final penalty was the lightest possible under the Qing judicial system, Li Wang’er was still convicted as a criminal. Responsibility had to be assigned, even if in the end that punishment amounted to only a fine. As these cases show, the guoshi statute applied fault to cases involving no mentally culpable state of mind, no degree of intent. If it could be established that one person’s actions could be linked to another person’s death, then the guoshi statute came into effect. In other words, Qing law seems to have had looser requirements for establishing causality between actions and results than we will find in the laws of the Republic. At the same time, as will be seen below, Qing law’s conception of negligence and recklessness was much more detailed than that provided under Republican law.

4 Other cases involving similar use of the law include Xianshen anjian: Jiangsu 9433, Huguang 13592, and Jiangsu 9441.

222

neighbors

Guoshi Killing as Negligent Killing In addition to dealing with accidental crimes, the guoshi killing statute also covered situations that crossed the boundaries from accident into an arena of slightly greater culpability. In such instances, the case usually fulfilled part of the requirement for accidental guoshi killing (for example, it was something “which the ears and eyes did not reach”), but failed to fulfill the rest of the requirement for accidental guoshi killing (for example, not being a case of “that which contemplation does not attain”). A variation on this theme occurred in the 1826 case of Ni Fu, which was forwarded to the Board of Punishments from Jiangsu province (Xing’an huilan, 1886: 31.46a). According to the Board, Ni Fu and his friend Liu Ligang had been sitting “on a boat imbibing copious amounts of liquor and getting quite drunk. Eventually Liu Ligang decided he had reached his limit, so he went to go rest in the stern of the boat. [Ni Fu] pulled Liu Ligang by the belt and demanded that he get up and drink some more. Liu Ligang went to wrest himself away and replied that he [would] not drink [any more]. [Ni Fu] was unsteady in his drunken state and as Liu twisted away, Ni Fu slipped and fell on top of Liu,” killing him. After some deliberation, the Board of Punishments issued a sentence of guoshi killing in this case, commenting that “although [this] does not quite match with ‘that which the ears and eyes cannot reach’ it mostly matches with 大半吻合 ‘that which contemplation does not attain.’ ” As a precedent for this ruling the Board cited an 1823 case that also involved drunkenness. In that earlier case, a drunk man was being helped home by another man on a rainy evening. They were walking along a riverbank when the drunken man slipped on the slick path and fell, dragging his companion into the water. Only the drunken man survived, leading to a sentence of guoshi killing for the survivor. In this case the alcohol-impaired physical and mental capacities of the offender did not lead to a harsher penalty. Instead, partial fulfillment of the standard requirements for a guoshi sentence was adequate to receive a guoshi sentence. The court acknowledged that such cases were something more than accident, but that did not mean they were something other than guoshi. That point was made even more clearly in the 1812 case of Liu Lanzheng, from Guangxi province, when a ruling of guoshi killing was issued despite the circumstances being even more serious (Xing’an huilan, 1886: 31.45a). According to the Board, “Liu Lanzheng was hired by a lumber shop, moving and piling up some pine boards.” While Liu went to get a load of boards, a six-year old boy began to play on the partially completed pile of wood. Liu dropped his next load on top of the youngster, and the boy died soon thereafter from the injuries he sustained. The Board noted that Liu “was not paying enough attention” 放不及照顧, but still ruled this a case of guoshi. Clearly

guoshi killing

223

this was more than a simple accident, as Liu should have been more careful in carrying out his job. However, the guoshi statute was broad enough to include not just accidental crimes, but also crimes with a higher degree of fault. In addition to these cases sentenced under the guoshi statute, select additional statutes that covered crimes that went past the bounds of accident permitted the same minimal punishment as that assigned to accidental acts. Though not directly sentenced under the guoshi laws, by analogy these cases were considered guoshi crimes, too. Take, for example, the 1902 case of Zhang Er, an elderly nightsoil collector who was run into by a donkey cart owned by Ma Shuaibin in Daxing county outside Beijing (Xianshen anjian: Guizhou 22363). The cart owner testified about what happened on the day of Zhang Er’s death: [I] transported five shi of maize to be sold at the Kang family grain brokers on Dongju Street.5 I unloaded the maize and put the cart on Dong[ju] Street. I was headed to the grain broker entrance and had just reached the measures when unexpectedly this Zhang Er was at the side of my cart collecting manure. The shaft horse’s eyes went back and startled, it ran. . . . I went to look and saw Zhang Er lying on the ground. I helped raise him to a sitting position and [he] was still breathing. Witnesses confirmed this version of events. Zhang Er’s son came and took his father home, where he died. The Board ruled that although Zhang Er’s death was “not something the criminal could have anticipated” 非該犯意料所及, he should have known that “horses by nature are not docile” 馬性不馴. As a result, the Board considered Ma’s failure to tie up the horse “careless” 疏忽. Ma was sentenced under Article 234, “Domestic Animals That Bite or Kick People” 畜產咬踢人, which contained a provision for people who did not properly restrain their animals with the result that a person was injured or killed. If death did result, the code stated that the laws on guoshi killing should be used to sentence the criminal (Xue, [1905] 1970: A .234.00). Therefore Ma received the same penalty as that assigned to guoshi killing (in essence a fine), implying a low level of culpability. Here, then, we find a non-accidental crime—it was after all called a “careless” act by the Board—placed in the same category of culpability as guoshi killing. Thus, while the category of guoshi killing encompassed crimes of pure accident, the concept of guoshi was broad enough to involve acts that involved a low-level mental element, a degree of neglect. 5 A shi 石 is a unit of measure for grains equal to 120–160 lbs.

224

neighbors

Another case that involved a similar penalty came before the Board of Punishments in 1905 (Xianshen anjian: Yunnan 21690). In this case Geng Bozhang and Zhang Luozi, two young factory workers in Beijing, were suffering from persistent syphilitic lesions 髒瘡. According to their friend, Mao Wu, when the boils took on a foul odor, Mao Wu and another friend, He Yongshan, took them to Shuang Quan for treatment. Shuang Quan testified: [I am] seventy-two sui.6 I make a living running a baked cake shop in Daqiao outside the Chaoyang Gate, and I also cure venereal diseases. On the eighteenth day of the third month of this year, Mao Wu and He Yongshun brought Geng Bozhang and Zhang Luozi to my shop and asked me for a cure. I looked and saw that [they] both had sores. I used a recipe of calomel, eggplant, and grain to be burned [and inhaled] for a cure. Mao Wu paid me 6,000 文and everybody left. Unfortunately, despite this “cure,” the men’s condition continued to deteriorate after they had returned to the factory. They became so ill that Mao Wu took them to his grandfather, a Daoist priest named Mao Libin. According to his testimony, “Mao Wu and this He Yongshun pulled a cart carrying Geng Bozhang and Zhang Luozi to my home in the temple for me to take care of their illness. I consented, and saw that Geng Bozhang’s illness was grave. I questioned them and [learned of Shuang Quan’s treatment]. They brought Shuang Quan to see [the men].” According to Mao Wu, at this point Shuang Quan told the men “it was not serious and helped Geng Bozhang up and pressed at his abdomen.” Geng Bozhang died later that day, and Zhang Luozi died five days later at the temple. An autopsy was performed on both bodies, and the coroner found that they were poisoned by the medicine they had been given. The court ruled in this case according to the Qing Code Article 297, “Incompetent Doctors Who Kill or Injure People”: if the doctor “mistakenly fails to use the proper methods and because of this brings about death . . . if there was no evidence of the intention to harm, use the guoshi laws to sentence” (Xue, [1905] 1970: A. 297.00; modified from Great Qing Code, 1994: 281). Shuang Quan’s death sentence was thus redeemable, and he paid the fine to the families of the deceased. This case too, then, was one of guoshi that was not simple accident, but rather was assigned a low level of mental culpability. 6 The term sui 歲 designates age and is calculated based on the passing of the Chinese lunar new year. Furthermore, a person is considered one sui at the time of birth. Therefore, a person’s age under this system is usually one to two years higher than a person’s age as calculated according to the Western calendar.

guoshi killing

225

Similarly, in a 1791 case from Sichuan, Li Xiuyu used impure ingredients when concocting a medicine he sold to Wu Guixiang, and two people died as a result of taking it (Xing’an huilan, 1886: 33.28a). The court sought to decide this case according to Article 297, quoted above. Li Xiuyu was allowed to monetarily redeem a death sentence, but was sentenced additionally to 100 blows of the heavy bamboo and three months in the cangue for bringing about the deaths of more than one person. In this case of what could be called low-level negligence, the multiple deaths seem to have necessitated a more serious punishment than the standard guoshi sentence given such cases. Non-Guoshi Negligence Both the case of the improperly tied horse and the cases of the incompetent physicians suggest that Qing law possessed a complex conceptualization of the mental element involved in crime, far beyond a simple accident-versus-intent dichotomy. In addition to the guoshi-equivalent crimes that went beyond the realm of pure accident, the Qing Code contained a number of statutes that covered offenses considered unintentional but still more serious than accidental offenses. These crimes all warranted more serious sentences than guoshi killing, but less serious sentences than those for crimes involving harm that was committed intentionally. In essence, these crimes occupied the category of mid-level criminal negligence. The first such crime to be considered here is that of killing with a bow and arrow (Xue, [1905] 1970: A. 295.00). The statute in the Qing Code read: “For those who, without reason 無故, fire a bullet or shoot an arrow or throw tiles or stones toward a city or a place where people live . . . if death results, sentence to 100 blows of the heavy bamboo and exile to 3,000 li” (modified from Great Qing Code, 1994: 280). The statute on killing with a bow and arrow in the Ming Code (Da Ming lü jijie fuli, [1908] 1989: 19.30a) was the same, with the exception of one character. In the Ming Code, the first line of the statute just contains the character 故, reading: “For those who intentionally 故 fire a bullet. . . .” The commentary notes that the 故 here means “with no reason, but intentionally.” The character 無 was added to the Qing Code in 1725 for clarity (Xue, [1905], 1970: A. 295.00). Both versions of the statute, however, conveyed an awareness of one’s act and the possible consequences. A substatute to the article in the Qing Code, added in 1801, added killing with a fowling gun 鳥槍 to the statute (Xue, [1905] 1970: A. 295.01). Xue Yunsheng notes that this substatute incorporated into the law hunting deaths, wherein “firing [the gun] was not unintentional” 非無故 (Xue, [1905] 1970: A. 295.01). In other words, an offender whose crime fell under this law chose to fire a gun or shoot an arrow in a location where it was possible that a person could be injured or killed as

226

neighbors

a result. Such acts were not innocent accidents and not guoshi, and therefore required the harsher penalty of beating and exile. The codes also acknowledged that the crime of killing with a cart or horse was a more serious matter than guoshi killing. The relevant Qing law (Xue, [1905] 1970: A. 296.00) read: “For those who, without reason, gallop 跑走 on a horse or drive a cart very fast in streets or markets or market towns . . . if death results, [punish with] 100 blows of the heavy bamboo and exile to 3,000 li. If one gallops a horse without reason in the countryside or a wild area where there are [usually] no people . . . and causes death, punish with 100 blows of the heavy bamboo” (modified from Great Qing Code, 1994: 281). The commentary to the earlier Tang Code iteration of this statute had called killing with a cart or horse “rash” 唐突 (Tanglü shuyi, 1996: 324). Thus, if one recklessly galloped through a market or along a street and harmed others, punishment was due. Even if one was galloping in a remote area, galloping was inherently dangerous, and one should have been aware of the potential harm involved. In 1771, the Board of Punishments voiced its concern that crimes that should be adjudicated under this cart-or-horse law would be erroneously classified as guoshi crimes: “The article on killing with a cart or horse and the article on guoshi killing are the easiest to fall into confusion qianhun 牽混. . . . Examining the interlinear commentary of the guoshi law, there is the phrase ‘riding a horse that starts’ 乘馬驚走. But killing or injuring a person because a horse starts is different from galloping a horse without reason [and causing injury]” (Xing’an huilan, 1886: 31.38b). The Board continued, quoting from the guoshi statute to spell out exactly how killing with a horse or cart fell outside the bounds of accidental acts: “If one gallops a horse without reason and causes injury to another, that certainly is not something that cannot be anticipated 意料所 不及. Therefore only reduce the penalty for killing in an affray by one degree. . . .7 The phrase ‘riding a horse that starts’ in the guoshi interlinear commentary specifically means that the horse or mule is frightened by some external cause and the rider is unable to control [it]. If for no reason one hastens [the horse] with the result that the horse is startled and kills or injures a person, that is not ‘that which contemplation does not attain’” (Xing’an huilan, 1886: 31.39b).8

7 The penalty for killing in an affray in the Qing dynasty was strangulation after the assizes (Xue, [1905] 1970: A. 290.00). 8 Interlinear commentary was commentary to a statute in the Ming Code that was incorporated into the actual language of the statute in the Qing Code.

guoshi killing

227

The crime of killing with a cart or horse was considered slightly more serious than killing with a bow and arrow, and as a result burial costs were also required of one who committed this crime in addition to the mandated beating and exile. The rationale was provided in commentary to the Ming Code: whereas in the case of killing with a bow or arrow the injury was not within eyesight, in the case of running a person down with a horse or cart “the victim was right in front of [the perpetrator’s] eyes and [he] did not care to rein in [the horse]” 人在眼前全無控馭之心 (Da Ming lü fuli zhujie, 1993: 715). Finally, we have the category of losing control of a fire (Xue, [1905] 1970: A. 382.00). Under this statute any person who lost control of a fire, with the result that someone was killed, was punished with 100 blows of the heavy bamboo. The loss of control was always unintentional (intentionally setting fires was covered under a separate statute of the code, Article 383.00), but nevertheless the punishment mandated for those who committed this crime —a beating— indicates that the degree of blame assigned this offense was higher than that assigned to guoshi offenses. As with the two other statutes discussed above, this offense was more serious than that of guoshi, and thus necessitated a more serious punishment. To examine how these negligent homicide statutes were adjudicated in practice, let us first examine an 1826 case from Shandong province, the firearms case of Fang Xiaoliu (Xing’an huilan, 1886: 33.25a). Fang had been watching a funeral for a member of the Cui family. According to the Board’s report, Fang observed Gong Tianzhi, “who had been hired to fire a salute, take his gun and gunpowder and safely place them in the doorway,” where he thought they would be out of harm’s way. The Board continued: “The criminal [went over and] picked up the gun and fired it. Ma Chengtong, who was leaving his home nearby, was shot in the right temple.” Ma eventually died from his injury. The Board ruled that Fang ought to be sentenced according to the law on killing with a bow and arrow for “shooting toward a place where people live, and mistakenly 誤 injuring a person” (Xue, [1905] 1970: A. 295.01). Because the death occurred so long after the time of injury and when the victim had also fallen ill, the penalty of exile to 3,000 li was reduced to three years’ penal servitude, still a much more severe punishment than that for guoshi killing, and still placing this case firmly in the intermediate realm between accident and intent. Consider also the 1902 case of Liu Er, a teahouse owner who lived outside the Chaoyang Gate with his wife, Liu Chen Shi, and their daughter (Xianshen anjian: Fujian 10842). One night Liu Er fell asleep and left a fire burning in his tea stove. A shed caught fire and the house burned down. Although Liu Er was able to escape, his wife and daughter were trapped inside by the flames and

228

neighbors

died. The Board ruled that Liu Er had been remiss—he “neglected the fire” 遺火 and “lacked the proper care” 失檢 in putting out the fire before falling asleep. Liu Er was thus found guilty of losing control of a fire with the result that a person died (Xue, [1905] 1970: A. 382.00) and sentenced to 100 blows of the heavy bamboo. Such a crime, though not committed intentionally, arose out of carelessness rather than sheer uncontrollable accident or even low-level negligence, and as such the law imputed more serious blame to the offender. An additional case comes from Ba county, Sichuan in 1864, where a young man named Chen Lunfu was found guilty of neglecting a fire in the house of his master, Wang Xingshun (Baxian dang’an: 1467). The fire eventually raged out of control, destroying not only Wang Xingshun’s shop, but also a number of warehouses and homes on the same street. Most residents escaped, but two men residing with Wang’s neighbor were trapped in the fire and burned to death. For his neglect of the fire and the deaths that resulted, Chen was sentenced to whipping 鞭 and three months in the cangue. Finally, we have two cases from 1815. In the first case, from Jiangsu, Wang Liu was galloping a horse in the city and plowed down an elderly man who had just entered the street from a side alley (Xing’an huilan, 1886: 33.27a). Wang Liu had called out a warning, but much as in the case of Lu Yongtai above, the elderly man was deaf and did not hear him. However unlike the case of Lu Yongtai, this was no runaway horse or runaway horse cart. This was a man who was knowingly galloping down an urban street. This was clearly a case of “Killing with a Cart or Horse” (Xue, [1905] 1970: A. 296.00), in which a rider gallops his horse within the city walls, an act the code considered inherently dangerous and easily capable of causing harm. In a similar case from Heilongjiang, this time involving the death of a five-year-old child, the Board of Punishments described the actions of the perpetrator, Ji Changchun, as “careless, and moreover not ‘that which the ears and eyes do not reach, that which contemplation does not attain’” (Xing’an huilan, 1886: 33.27a). Ji’s actions clearly exceeded the bounds of guoshi killing, and thus he, too, was sentenced in accordance with the law for “Killing or Injuring with a Cart or Horse.” Both men received a punishment of 100 blows of the heavy bamboo and exile to 3,000 li. All of these cases reflect crimes that carried a variety of sentences that were more severe than the fine assessed to guoshi perpetrators. These crimes incurred a higher degree of blame, but nothing serious enough to mandate a capital sentence. The multiplicity of these offenses and their increasingly serious punishments show that degrees of intent under Qing law ran along a complexly graded continuum.

guoshi killing

229

Killing at Play The statute on killing at play also occupied the conceptual gray area between intent and accident in Qing homicide law. Though not a crime of intentional killing or intentional harm, it incurred a higher degree of responsibility (and thus a more serious punishment) than guoshi killing and the homicides-bynegligence discussed above. This crime occupied the category of criminal recklessness, involving an awareness on the part of the offender that his actions could cause harm to others and a “willful disregard” for that risk (LaFave, 2000: 249).9 The increased seriousness of killing at play over guoshi and mid-level negligence crimes was reflected in its punishment—a non-redeemable death sentence. The crime of killing at play, xisha 戲殺, was described in Article 292 of the Qing Code as “doing something capable of killing someone when playing, such as boxing or fencing” 以堪殺人之事為戲如比較拳棒之類 (Xue, [1905] 1970: A. 292.00; modified from Great Qing Code, 1994: 278). An entry in the Mulingshu provided a similar definition: “If a situation arose from mutual blows in a contest and death resulted thereby, that is called xisha” (Mulingshu, 1848: 14.1a). These pithy definitions appeared to limit the scope of this particular law to dangerous sparring. However, the Board of Punishments elaborated on the definition in 1811, noting that killing at play could occur when “hands, feet and other objects such as metal blades, all of which can kill people, are then used in play” (Xing’an huilan, 1886: 31.53b). Such a broad definition in effect classified one’s own appendages as deadly weapons, giving killing at play an extremely broad scope. Such a definition infused an unwritten notion of recklessness into the law: play was inherently dangerous; those who engaged in behavior that was capable of harm, or used their bodies in ways that were capable of harm, should be aware of the potential danger. If the actors carried out these dangerous activities and thereby caused a person’s death, they would be punished with delayed strangulation. In addition to the initial killing-at-play provision, the subsequent section of the statute described additional circumstances considered “equivalent to killing at play” 與戲殺相等. Those were: For example, knowing that the ford in a river is deep and muddy but falsely claiming that it is even and shallow, or that a bridge or ferry boat is rotten and leaky and not capable of getting people across and falsely 9 On the concept of recklessness in contemporary common law, as well as the relationship between recklessness and negligence, see LaFave, 2000: 246–56.

230

neighbors

claiming that it is firm, and by deceit causing people to cross, and as a result a person falls in and drowns or is injured, this is equivalent to killing at play. (Xue, [1905] 1970: A. 292.00/B; modified from Great Qing Code, 1994: 278.) Since these situations were “equivalent to killing at play,” they merited the same sentence of delayed strangulation. The circumstances described in this second section of the law, however, were more serious than simple killing in the course of roughhousing. They involved both awareness of imminent (not just potential) danger as well as willful deceit. This was malicious trickery, wherein the criminal foresaw a risk and consciously chose to disregard that risk.10 The term “killing at play” had originated in the Tang Code, where it was a graded offense (Morohashi, 1992: 11681.54). The Tang Code defined the first and lowest grade of killing at play as “agreeing to a game where strength is used in a friendly way, but death results” (Johnson, 1997: 382). The commentary later called this “play fighting” 戲鬥 (Tanglü shuyi, 1996: 288). For this crime, the standard homicide punishment of strangulation was reduced two degrees to a sentence of penal servitude. The second grade of killing at play involved riskier recreation wherein it was agreed “to use knives, or to ascend to a dangerous height, or to be on the brink of treading on thin ice or entering the water; since they are in a dangerous spot, they must warn one another” (Tanglü shuyi, 1996: 288; modified from Johnson, 1997: 382). If no warning was given and a person was killed, the standard homicide punishment was reduced by only one degree to a sentence of life exile. The third and highest grade, “causing death or injury through deception” 詐陷人之死 (Tanglü shuyi, 1996: 321; modified from Johnson, 1997: 451), was the same as what Qing law called “equivalent to killing at play” (Xue, [1905] 1970: A. 292.00, quoted above). This article was not classified under the homicide statutes in the Tang Code, but was instead found under the chapter on “Fraud and Counterfeit” (Johnson, 1997: chap. 25). Those convicted during the Tang dynasty under this law received the standard homicide punishment with no reductions and no redemption; thus, strangulation (Tanglü shuyi, 1996: 264).

10

The very last imperial law code, the never-promulgated Revised Qing Code of 1910, formally recognized the increased severity of the death-by-deception provisions, adding interlinear commentary to the code noting that death by deception was “more serious than killing at play” 較戲殺為重 (Qinding da Qing xianxing xinglü anyu, [1910] 1995: 532).

guoshi killing

231

However, this tripartite gradation in punishments was amended with the addition of a clause in the killing-at-play statute that permitted, for the first two grades of killing at play, redemption of punishment according to the accidental killing laws 依過失法收贖 (Tanglü shuyi, 1996: 288).11 Thus the first two grades of killing at play, both play fighting and its riskier version, were assigned the same punishment and were in effect collapsed into one grade. Furthermore, since under Tang law all guoshi homicides were allowed monetary redemption,12 Tang law provided the same punishment for guoshi homicides and killing at play. The rationale was provided in the commentary: criminal intent was absent in all cases. The third grade of killing at play discussed above was set apart from the other killing-at-play provisions by its punishment—since no redemption was allowed under this statute, a higher degree of culpability was assigned. By the Ming dynasty, however, a higher degree of blame and a more developed discussion of intent had entered the law for the general killing-at-play offenses. In the Ming, the death-by-deception statute of the Tang Code was classified for the first time under the killing-at-play article, and also for the first time both regular killing at play and death by fraud and deception merited the same punishment. No longer were killing-at-play crimes viewed in the same light as guoshi ones. Killing at play had entered the realm of more blameworthy homicides, as shown by the fact that it was now punished with the non-redeemable sentence of delayed strangulation. As a result, we now find the codes comparing killing at play with the more serious—from the point of view of intent—homicide category of killing in an affray (Wallacher, 1983: 279). For example, one Ming-era commentary to the Ming Code noted that “although play 戲 is not the same as fighting . . . nonetheless the killing or injury really was because one laid blows. Therefore, use the killing or injuring in an affray law to sentence” (Da Ming lü jijie fuli, [1908] 1989: 19.24a).13 In 11

12 13

Since only one fine is specified therein—120 jin—the commentary notes that no reduction of that sum will be allowed. A jin 斤 is a catty, a unit of measure equal to roughly 1.3 lb. That redemption was also in the amount of 120 jin. Wallacher, 1983: 281, calls the increase in the sentence assigned to killing at play “a marked switch from a subjective theory of culpability to an objective theory,” i.e., while Tang and Song law considered killing at play as a lesser crime warranting mitigation, in Ming-Qing law the lack of intent to kill mattered less and the harm done more, hence the increase in the penalty to death. If this were so, the guoshi statute should have also seen an increased penalty in the Song-Ming legal transition. I would argue instead that the change in penalty shows greater attention to the differentiation of degrees of intent along the homicide continuum. Although the perpetrator in killing-at-play cases had not

232

neighbors

addition, we find discussions of the potential degree of intent associated with such a crime. Another Ming-era commentary on the code, the Da Ming lü fuli jianshi, explained the term “xisha” as “allowing an exchange of blows vying in a contest to win or lose. During the Jin, this was called consenting to hurt one another: knowing that [one’s acts] were enough to cause harm and agreeing to [commit] the acts. Therefore, any death or injury does not not arise out of intent 非出于不意” (Xue, [1905] 1998: 490). Although the Qing Board of Punishments noted in 1817 that killing at play and guoshi killing “both lack the intent to harm” 皆無害人之心 (Xing’an huilan, 1886: 31.55a), the Qing Code continued to mandate a non-redeemable sentence of delayed strangulation for all varieties of killing at play. And it was in legitimizing the gap in the severity of punishment between xisha crimes and guoshi homicides—the difference between the death penalty and a fine—that the court was able to emphasize and articulate the conceptual difference between the two categories: “In the adjudication of cases, we cannot allow killing at play and guoshi killing to be dragged into confusion” (Xing’an huilan, 1886: 31.55b). The Board did so by explaining that “if there is no jesting situation with the person, and accidentally a person is killed, [such a situation] really is that which the ears and eyes do not reach, that which contemplation does not attain, and [such a homicide] is called guoshi” (Xing’an huilan, 1886: 31.53n). Conversely, the Board considered death due to play a crime which the perpetrator could have foreseen: “play, then, is something the ears and eyes and contemplation have already given attention to. . . . Killing a person could have been reckoned upon” (Xing’an huilan, 1886: 31.53b). Elsewhere, the Board commented that “most of the killing-at-play cases involve vying to see who is stronger. Such actions can bring about injury; there is not a lack of knowledge [on the part of the offenders about that possibility]” (Xing’an huilan, 1886: 31.47b). In the eyes of the law, roughhousing was inherently dangerous, and any person who chose to participate would be held accountable for the consequences. By punishing the unintentional crime of killing at play, and underscoring its conceptual difference from accidental killing, Qing law emphasized not only that there was a gray area between intent and accident, but that there were multiple shades of responsibility between those two poles. Granted, crimes of killing at play were committed in the absence of intent to harm, but so were intended harm, he purposefully engaged in obviously risky behavior, a reckless act that bordered on intentional harm. This was especially true of the malicious trickery cases considered “equivalent” to killing at play.

guoshi killing

233

crimes of guoshi killing. And there was a difference in culpability between guoshi killing and killing at play, be the guoshi killing one of accident or lowlevel negligence. This point was reinforced by the fact that the xisha statute was home to the intent-hued homicide category of death by deception. Evidence from local-level Qing dynasty homicide cases indicates that the majority of cases adjudicated under the killing-at-play statute were concerned with one of two issues: whether the situation was one of play or guoshi, and whether the situation was one of play or affray. The former concern is illustrated in an 1810 case forwarded to the Board of Punishments by the Shaanxi governor concerning Shi Guangde, who injured and killed Song Lisheng while the two men were cavorting about and wrestling (Xing’an huilan, 1886: 31.56a). The provincial court, quoting the guoshi homicide law, declared the death “truly what contemplation cannot attain,” and ruled it a guoshi killing. The Board of Punishments disagreed, overturning this ruling in favor of a killing-at-play verdict, saying that the roughhousing “certainly was not ‘that which the ears and eyes do not reach’ . . . and similarly not ‘that which contemplation does not attain.’” Instead, the court ruled, “it matches exactly the laws on killing at play.” Clearly Shi Guangde did not intend to harm Song Lisheng, but was instead engaged in good-natured roughhousing with the victim. However, the danger of such roughhousing, as spelled out in the law, was something that Shi Guangde could have, and should have, known. By his willing participation in an activity that posed such a clear hazard to others, Shi Guangde automatically satisfied the requirements for conviction under the killing-at-play laws, and bore a higher degree of blame and responsibility for the death than that accorded under the accidental homicide laws.14 Frequently, however, the point of confusion in death-by-play cases had nothing to do with the issue of accident or low-level negligence, but was instead concerned with the possibility of non-congenial fighting and assault, in other words the difference between killing at play and killing in an affray. An example is found in a case from 1804, when the body of a John Doe was discovered slumped on the ground just north of the Dongsi Arch in Beijing (Xianshen anjian: Jiangsu 9409). Soon after the body was found, a man named Fang Zhigang identified it as that of his brother, Fang Da, who had worked as an assistant in the barbershop of Sun Cheng in the same neighborhood as the arch. In addition to working for Sun Cheng, Fang Da lived at his home along with another shop assistant named You Da and a street patrolman 看街兵 14

Wallacher, 1983, translates this case as well as numerous others from the Xing’an huilan.

234

neighbors

named Jin Liu. Sun Cheng told the court that after breakfast on the day he went missing, Fang Da complained of pains in the abdomen and then took the day off. Sun Cheng assumed that Fang Da had been assaulted sometime later. The autopsy report indicated that Fang Da died sometime between nine o’clock and eleven o’clock in the morning, and listed the cause of death as “internal injuries to the upper abdomen.” The court, in discussing the evidence, doubted that a fatal assault could take place in broad daylight in a public place without being noticed. The court reasoned: if Fang Da died so soon after he departed the shop, and if his body was found so close to the shop, wouldn’t it make the most sense if he had been injured in the shop? Both Jin Liu and You Da were brought in for interrogation. Jin Liu testified: I and the other shop assistants, Fang Da and You Da, when [we] woke up in the morning, Fang Da would make wisecracks, like that I was a son-ofa-bitch. I also ridiculed [him]. It was always this kind of “Are you scolding me?” talk in the morning when we got up. I went to enter Fang Da’s bunk and grabbed up his thumbs and pushed forward. Fang Da tried to get up but was unsteady . . . and pulled me down so that I fell on top of him. In trying to struggle free, [I] kneed him in the abdomen. Then we both let go and clambered up out of the bed, still exchanging a few jokes. You Da saw it all. You Da confirmed this tale in his own testimony, and both he and Sun Cheng averred that there “was no animosity” 並無仇隙 between Jin Liu and the deceased. The court, noting that “at no point had the playing become a fight” 始總並無爭毆情形, used the killing-at-play law to sentence Jin Liu to delayed strangulation. Because of the circumstances under which Fang Da’s body was discovered, the primary concern of the court in this case was to establish whether the homicide was caused by assault or in the course of play. Once it was established that no ill-will existed between Jin Liu and Fang Da and that the two men had been engaged in roughhousing rather than genuine strife, the appropriate situation had been identified (i.e., play and not affray) and the court could then issue a ruling of killing at play. A similar case arose in Beijing in 1899 (Xianshen anjian: Zhili 1255). On the day in question, Wu Dexi, a small-scale merchant 做小買賣 from Shuntian prefecture, had gone into town on business. He testified that on his way home that afternoon, he ran into four acquaintances: Tang Yutian, Chen Wenshan, Ji Gui, and En Jian. According to Wu Dexi:

guoshi killing

235

[They] were there tossing around a stone lock. Because I like to practice, I went to play with them. [We] spread out, tossing and catching. A large crowd of onlookers formed a ring around us. I played for two rounds. Just then En Jian threw and I caught the stone lock. When I turned and tossed [it] to Tang Yutian, suddenly an unidentified man came walking from the south. We saw that the situation was not good. At once we called out [for him] to get out of the way. Tang Yutian rushed to catch [the lock] but did not make it and the man did not get out of the way. . . . [It] hit the man right on top of the head. It could not have turned out much worse for the man. Though the various players went to get help, the victim died the following day. Four days later a farmer named Yang Chun identified the victim as his son, Yang Niuzi, and told the authorities his son had been sent to Beijing on a shopping errand. According to Yang Chun, “[I] heard rumors that my son had been beaten to death by a group of ten men.” Wu and the others denied that such an assault had taken place, and the coroner confirmed that the victim died after a stone struck him on the head. Once it was established that no affray had occurred, Wu was sentenced to delayed strangulation according to the laws on killing in the course of play or roughhousing (Xue, [1905] 1970: A. 292.04). His fellow defendants were sentenced to eighty blows of the heavy bamboo according to the article on “doing that which ought not to be done” 不應為 (Xue, [1905] 1970: A. 386.00; Great Qing Code, 1994: 359). The debates in these last cases reveal that not only was killing at play considered more serious than guoshi killing, but it involved a higher degree of responsibility, a mental element only slightly different from that involved in killing in an affray. In fact, the mental culpability involved in the two crimes was so close that they were assigned the same initial sentence, strangulation after the assizes. The differentiation in severity of punishment, and thus the fine gradation between the two crimes, is revealed when examining what occurred after the initial sentence, and during the standard review process for capital crimes. The death sentence for killing at play would have been commuted to a sentence of exile or penal servitude after the case passed its first Autumn Assizes (Meijer, 1967: 29; Huang, 1991: 182; Xue, [1905] 1970: A. 18.04, A. 411.27). Late imperial law considered killing at play a kind of reckless homicide, only one step down the ladder of criminal responsibility from crimes of intentional harm.

236

neighbors

The Republican Codes

In modern Western law, the foundation of Chinese law of the Republican period, guilt in a crime requires not only a criminal act (actus reus), but also criminal intent or a criminal state of mind (mens rea).15 The underlying idea is expressed by the Latin phrase actus not facit reum nisi mens sit rea, or “an act does not make one guilty unless his mind is guilty” (LaFave and Scott, 1986: 212). Of the two parts, modern Western law focuses more on the latter: “the concern of the criminal law is with the level of intentionality with which the defendant acted, in other words with what the defendant intended, knew, or should have known when he acted. . . . The mental element required by the definition of any crime, therefore, is of central concern” (Kadish and Schulhofer, 1989: 217–18). It is the individual’s will to commit the act that is of primary importance, of equal importance with the act itself and more important than any external circumstances surrounding the act. Early drafts of the 1912 Provisional Criminal Code (the first Republicanera criminal code) as well as subsequent drafts and editions of the Criminal Code of the Republic of China confirmed this framework. The commentary to Article 13 of the General Principles section of the 1911 draft of the code prepared by Shen Jiaben (Co-Commissioner for the Revision of Laws) and the Legal Revision Committee noted, “without criminal intent 犯意 there is no crime” (Da Qing xin xinglü, 1911: A. 13). Other jurists concurred: “to constitute an offense there must be both the act and intent” (China Law Review: 3, 3 [1927]: 133). Therefore, if there was only the intent and no act was committed, no crime had been committed. Conversely, and most importantly for us here, if there was only an act and no intent was present, no crime had been committed. The emphasis of Republican law, then, differed from that of Qing law. Republican law thought in terms of the act only in conjunction with criminal intent, while Qing law, in part because of the lingering influence of the principle of requital-in-kind, provided for the punishment of criminal acts even in the absence of any criminal state of mind. This was accomplished through the category of guoshi killing. As a result, the scope of punishable acts was wider in the Qing dynasty than during the Republican era: accidental killing was a crime in the Qing dynasty, while it was not a crime in the Republican era. Guoshi in the Republican-era Criminal Codes The category of guoshi killing survived the transition from Qing to Republican law, but its definition did not survive intact. As we have seen, in Qing law the 15

See LaFave and Scott, 1986: chap. 3; and Williams, 1961: chap. 1.

guoshi killing

237

guoshi statute itself covered not just cases of low-level negligence, but also cases of accidental killing. In addition, select other categories of killing that involved a slightly higher degree of culpability—categories in which the code noted carelessness or lack of proper attention—could be analogized to the guoshi statute. Beginning with the code of 1912, the term “guoshi” underwent a shift in meaning. Instead of using it to describe situations that were for the most part unforeseeable, as in the Qing, Republican codes used it to describe the exact opposite: a situation that not only could have been, but should have been foreseen. This marked a potentially confusing departure from the Qing dynasty use of the term “guoshi killing.” This change also meant that accidental killings were no longer to be prosecuted. The loss of this category narrowed the scope of punishable acts under Republican law. The change in the meaning of “guoshi” also meant a change in the codified treatment of negligent killing. While the Qing Code used a multiplicity of statutes to cover crimes of negligence, under Republican law there would be only one statute for such offenses. The graded scale of negligent acts found in late imperial law—from acts with such a minor degree of carelessness that they could be analogized to accidental killing to acts that bordered on recklessness or intent—would be no more. Now all negligent acts were subsumed under one article of the criminal code. The shift from Qing to Republican conceptions of guoshi and negligence was a murky one, and one made all the more difficult since no definition was provided for the term “guoshi” in its two main appearances in the first Republican-era criminal code, the Provisional Criminal Code of 1912. That code simply read: Article 13: Unintentional acts will not be punished, with the exception of those committed in negligence (guoshi). (Zhanxing xin xinglü, [1912] 1988: A. 13) Article 324: Those who kill or injure another through negligence (guoshi) will be punished as follows: 1) if death or serious injury results, a fine of 500 yuan or less; 2) if incapacitation results, a fine of 300 yuan or less; 3) if light injury results, a fine of 100 yuan or less. (Zhanxing xin xinglü, [1912] 1988: A. 324)16 16

In the 1918 and 1919 draft criminal codes and the Criminal Codes of the Republic of China (1928 and 1935), the punishment for negligent homicide was modified to incorporate the option of sentencing the offender to two years or less of imprisonment (Xingfa dier ci

238

neighbors

Since a precise definition for guoshi was not spelled out in the code, the door was left open in the early years of the Republic for jurists to use the term in the same way it had been used in Qing courts, i.e., to apply the category of guoshi killing to both accidental and low-level negligent homicides.17 The legal specialists who composed revised drafts of the Provisional Criminal Code during the 1910s were aware of this problem, and complained that in the 1912 Code, “the scope of [the terms] ‘intention’ 故意 and ‘negligence’ 過失 was never determined” (Xingfa dier ci xiuzheng an, [1918] 1973: A. 19). Noting that countries such as Italy, Russia, Siam, and Switzerland all included individual statutes defining negligence, they proposed that Chinese law do likewise (Xingfa dier ci xiuzheng an, [1918] 1973: A. 19). In the meantime, jurists had to make do with definitions provided in rulings by the Supreme Court 大理院 (Daliyuan). For example, in 1914 the court issued a ruling explaining the difference between crimes committed with intention and those committed through negligence, noting that negligent acts “lack criminal intent 犯意; however, because attention was not paid a crime was constituted. Therefore [to determine if it is] negligence, the standard is whether or not [the offender] recognized the result of his act. If he knew there would be such a result, and still recklessly 悍然 committed [the act], that cannot be called guoshi” (Guo, 1946: 54). Therefore, if one realized the possible results of one’s act and still carried it out, then one committed the act and achieved the result on purpose, and thus intentionally. However, if one did not realize the result of one’s act, it was committed in negligence.

17

xiuzheng an, [1918] 1973: A. 291; Criminal Code of the Republic of China, 1928: A. 291; Criminal Code of the Republic of China, 1960: A. 276). The earlier (1912) version, which simply imposed a fine on the offender, seems to mirror late Qing attempts to eliminate symbolic punishments for other forms of homicide. For example, in the revised Qing Code of 1910 the death penalty was eliminated for killing-at-play offenses and replaced with a sentence of three years of penal servitude. For guoshi crimes the initial death sentence was also eliminated in favor of an automatically redeemable sentence of penal servitude (Qinding da Qing xianxing xinglü anyu, [1910] 1995: 532–33). Thus, the reintroduction of a penal servitude option for guoshi offenses in the 1928 code indicated a return to earlier Qing models in terms of sentencing. Commentary on a 1907 draft version of the eventual Provisional Criminal Code of 1912 provides not clarity but rather further confusion, noting that situations of guoshi arose when the perpetrator “did not know to pay attention” 不知注意 (Da Qing xinglü zongze cao’an, 1907: A. 13). This was no break from Qing precedent, as it implied not a conscious lack of attention, but rather a situation in which the perpetrator was not aware that he even needed to pay attention. Such situations comprised the bulk of Qing-era accidentalguoshi crimes.

guoshi killing

239

The court issued a slightly different definition of guoshi the next year: “To constitute negligence, the requirement is that one did not pay attention to facts of which one could have foreseen. If it was something that could not have been foreseen 不能預知, then there was no way to pay attention, and there cannot be a problem of negligence” (Xingfa zongze, 1944: 58). While the first ruling had described the difference between intention and negligence, the second ruling noted the difference between negligence and pure accident, in effect marking part of the difference between the Qing-era definition of the term guoshi and the new Republican-era definition. Republican jurists no longer believed that the results (such as death or injury) of an act could serve as the sole grounds for issuing a verdict of criminal responsibility. Thus we find the commentators to the 1918 Draft Criminal Code critiquing the laws of the past wherein, according to the authors, the result 結果 of an act was what mattered most, not the intent (Xingfa dier ci xiuzheng an, [1918] 1973: A. 22). Notions of cosmic imbalances and requital were now absent from the law as guoshi no longer included accidental acts, but instead consisted solely of acts wherein the criminal was found to have been remiss in paying attention.18 Modern Western legal codes often use the “reasonable man standard” to explain this notion of negligence: if a reasonable man would have been aware enough or careful enough to prevent the commission of the crime, then the perpetrator must be held negligent (and thus at fault). This standard appeared in some Republican-era negligence cases.19 Not until the official promulgation of the Criminal Code of the Republic of China (1928) did a formal definition of guoshi debut in the legal code. Finally the term had a codified article which provided a standard definition for the term: Article 27: An act is done in negligence when the actor, although not acting intentionally 非故意, but under circumstances when he should have or could have given attention, does not do so 雖非故意但按其情節應 18

19

See, for example, a 1933 Supreme Court ruling emphasizing the difference between negligence and accident. In this case the court ruled that if an event was outside the scope of potential awareness, it was not a case of negligence (Zhonghua minguo xingfa pan jieshi yi quanwen, 1972: 84). The court ruled that if an event “really could not be [foreseen], then there ought not to be criminal responsibility [for the event].” Acts that were unforeseeable were no longer to be considered crimes. The term is translated as 自然人or 通常人. See Zhonghua minguo xingfa pan jieshi yi quanwen, 1972: 729; Daliyuan xingshi panjue quanwen huibian, n.d.: 2663; and Xingfa zongze, 1944: 57.

240

neighbors

注意並能注意而不注意. . . . [Or] although he foresaw that [the act] could occur, firmly believed that it would not occur 雖豫見其能發生而 確信其不發生. (Modified from Criminal Code of the Republic of China, 1928: 6.) This article provided, in essence, two possible degrees of negligence. In the first, the actor did not foresee the results of his act, even though he should have. In the second, the actor foresaw what could happen, but believed it would not happen. This article followed immediately after the article that provided the legal definition of “intention” 故意, defined in part as follows: “An act is done intentionally when the actor, with regard to his act constituting a crime . . . foresaw 豫見 that it would occur, and moreover its occurrence was not contrary to his intentions” (modified from Criminal Code of the Republic of China, 1928: 6). Thus, perpetrators of both negligent and intentional acts could possess foresight that an offense would occur. The difference between the two hinged on whether the actor believed the criminal act would not occur, or did not oppose the occurrence of the act. These definitions brought crimes of negligence and crimes of intention into proximity for the first time. In the West, the concept of criminal intent (mens rea) usually did not include negligence. As noted by Glanville Williams (1961: 31), “negligence is not necessarily a state of mind; and thus these crimes are best regarded as not requiring mens rea.” Nonetheless, Williams went on to note that negligence “is a kind of legal fault, and in that respect [negligent acts] are akin to crimes requiring mens rea. . . . Responsibility . . . may be incurred by the mere neglect to exercise due caution, where the mind is not actively but negatively or passively at fault” (Williams, 1961: 31, 100). At the turn of the twentieth century, German jurists argued that the German term used to represent mens rea actually referred more generally to states of guilt, and thus encompassed negligence as well as intent (Fletcher, 1971: 413). So, though negligence by definition implied a lack of intent, it still described a generally culpable state of mind—a state of mind that by all reasonable standards should have taken care not to commit the act. By the early 1930s Republican legal thinkers were also broaching this issue, since it pertained to differentiating foresight (in the second part of the negligence statute) from criminal intent. An article in the journal Law Review (Falü pinglun) raised several possible theories for interpreting the concept of guoshi. Among these was the notion that guoshi crimes were closely tied with those committed with intent: “[Negligent] acts arise from a person’s intentions; it is only that the results of the act were not foreseen. On this point [negligence] is different from criminal intent 犯意; however as to the recognition of the act,

guoshi killing

241

in essence that is no different from [that involved in] criminal intent” (Ping, 1930: 1).20 In other words, a person who committed a negligence crime and a person who committed a crime of intention both purposefully carried out an act that resulted in harm. In the case of negligence the outcome may not have been desired or foreseen, but the actions themselves were still done willingly or on purpose. The proximity of negligence and intent in the Republican period illustrates another difference between the Republican definition of guoshi and its meaning under Qing law. In Qing law there was a very clear demarcation between crimes of guoshi and crimes of intent. They were in fact on opposite ends of the intent continuum, with all other categories of homicide in between. There was no need to expend the courts’ time drawing guidelines for how to separate the two categories. So large was the gulf between guoshi killing and intentional killing in the Qing that the issue of confusing the two categories two was never raised, be it in code, commentary, or legal case. Republican law, with its two muddy and ambiguous categories of mental culpability, gave jurists a much harder time. Republican-era Guoshi Cases To examine how changes in the scope and definition of the term guoshi played out in practice, we turn to court records of guoshi cases from the Republican period. The guoshi cases that came before the Republican courts can be divided into three categories. First are the cases where the Republican-era guoshi laws were used to sentence criminals for negligent homicides, in accordance with the definition of guoshi provided in the Republican code. Second are cases where the lower courts sentenced people under the guoshi laws for accidental homicides—homicides no longer punished under Republican law—or for other crimes not meeting the definition of negligence under Republican law. When this happened, the higher courts had to reassert the new Republican definition of the term to prevent accidental acts from being punished. Third are cases wherein the courts had difficulties differentiating between negligent homicide and intentional homicide, categories which until 1912 had been discrete conceptual categories at opposite ends of the intent spectrum, but which now, as the only two remaining homicide categories, had become increasingly difficult to differentiate.

20

See also Xingfa zongze, 1944: 58–59.

242

neighbors

Standard Negligence Cases of standard negligence under Republican law included the kind of offenses often adjudicated as guoshi under Qing law, the low-level negligence guoshi seen in many of the cases discussed in the earlier section on Qing law. A query sent by the Gansu High Court to the Supreme Court in 1920 provides a good example of the proper use of the new guoshi laws (Daliyuan jieshili quanwen, 1931: 1377). The case in question started when a man from Qingyang county, Gansu, left home just before a downpour. According to the case record, his neighbor “noticed through a crack in [the man’s locked] door that [the man’s] cave was flooding. [He] dug a small hole out below the door to let the water out and prevent any further damage.” However, when the man returned home, he blamed the neighbor and the hole the neighbor had dug for all the damage. A dispute arose and the man asked the local militia 保衛團 to become involved. It did, sending several men to investigate, one of them bringing a rifle 來福鎗 along. In the course of the investigations, the rifle-toting militiaman had an altercation with a bystander and ended up pistol-whipping him. As a result the rifle went off, killing the hapless man who had been the original victim of the flooding. The Supreme Court explained that the militiaman ought to be sentenced according to the laws on negligent homicide (Zhanxing xin xinglü, [1912] 1988: A. 324), agreeing with the assessment of the lower court that “[his] intention was only to grab the gun handle to beat [the bystander]” and that he had not intended to fire the gun. Such a ruling would be appropriate given the homicide statutes in effect at the time. The militiaman did not intend to shoot the gun (and was apparently unaware that it was loaded), and thus it was not a crime of intentional killing. He should, however, have been aware of the inherent danger of beating a person with a loaded weapon, and his failure to give proper attention to this danger made him guilty of negligent homicide. This case would have been considered one of low-level negligence guoshi and sentenced similarly in the Qing. In fact, the circumstances of this case were very similar to those in a case that came before the Board of Punishments in the early 1800s. In the latter case, a man purchased a used gun, unaware that it was still loaded. “When he placed the gun in a fire to burn off some rust, the gun went off,” killing someone (Xing’an huilan, 1886: 31.42b). The offender in this case was sentenced for guoshi killing. Another example of a case that accorded with the Republican-era guoshi laws is provided in the 1915 Supreme Court trial of An Jiren, a resident of Jianghua county in Hunan who was a member of a small local militia (Daliyuan xingshi panjue quanwen huibian, n.d.: 2663–66). On September 13, 1913, he and two colleagues had just arrested a criminal named Li Shangchu, who was

guoshi killing

243

involved in opium crimes. They were heading home when a group of bandits confronted them in an effort to free the prisoner. The bandits achieved their goal. Fighting ensued, during the course of which An Jiren fired a shot. The bullet killed not one of the bandits nor the escaped Li Shangchu, but rather an innocent cattle-herder. The Hunan High Court procurator, who was appealing an earlier ruling, argued that in this case “of course there was no intention 故意 [to kill the cattle-herder], however [An Jiren] did not give the attention that an ordinary person 通常人 ought to give, [and thus he] ought to bear responsibility for . . . guoshi killing.” The Supreme Court, in examining the case, agreed. The court first noted, “In this case the accused of course did not have the intention to kill Li Wang Shi; that goes without saying.” However, despite the turmoil of the situation, it did all take place in broad daylight 百晝 and An Jiren “ought to have paid attention, knowing that there was a person by the side of the road 明知道旁有人自應注意.” The Supreme Court issued a verdict that punished An Jiren with a fine of 500 yuan in accordance with the negligent homicide laws. This case, like the one above, accorded with Qing concepts of guoshi negligence. In their rulings on these cases the Republican courts emphasized carelessness or lack of proper attention on the part of the offender, quoting codified definitions of guoshi (“did not give the attention that an ordinary person ought to give”) much as their Qing counterparts had done. Consider also the May 1929 shooting death of Dong Tingsheng after bandits had fired on the local militia in Lucheng village, Wanping county outside Beijing (Beijing difang fayuan: J65–4: 450–452). Injuries were sustained on both sides, with one bandit fleeing the scene on a bicycle. Six days later the militia heard of another group of bandits nearby and went on patrol. On the southern border of the village a militiaman named Cheng Jintang encountered a man riding by on a bike. Suspecting he was the escaped bicycle-riding bandit 乘車逃脫之逸匪, Cheng ordered the man to dismount. According to Cheng, “I ordered Dong Tingsheng to get off the bike for inspection. He not only did not listen, instead [he] pulled out a handgun and fired at me repeatedly. I just fired three shots in the air. I did not fire at him.” Members of the local militia confirmed his story. However, others called in to give evidence in the case presented a different version of events, and Cheng Jintang’s claims were soon proven to be spurious. First, several people confirmed that Dong Tingsheng was not a bandit, but instead was an employee of a charcoal factory who had been out collecting payments 收取貨款. Then, the court called into question the feasibility of Cheng Jintang’s claim that Dong fired repeatedly as he passed by on the bike. The court noted that based on where the gun was found at the crime scene, it was questionable whether Dong had even been holding the

244

neighbors

gun when he was shot. In addition, the court identified the make of the gun as a Belgian Seven 比國七. This model was capable of firing only one cartridge at a time, requiring manual removal of the old cartridge and reloading to fire again. The court wondered how Dong Tingsheng could possibly be reloading and firing his gun and still be able to speed his bicycle past on the bumpy 凹凸不平 village roads. Furthermore, the court questioned even Cheng’s story of firing warning shots into the air: if being fired at, wouldn’t he have returned fire immediately? Having thus refuted every part of Cheng’s account, the court made its ruling: Dong Tingsheng was your typical commoner. He was not a bandit, and as has been noted above, he was not carrying a gun. Moreover, Dong Tingsheng was wearing a blue shirt, and it was not at all like the white one described on the escaped bicycle-riding bandit. . . . Attention ought to have been paid to such matters, and could have been paid . . . . but [Cheng] did not pay attention. Cheng Jintang was found guilty of negligent (guoshi) homicide—he should have known the harm that could result from firing at a person—and this ruling was upheld on appeal to the Hebei High Court. While a guoshi statute had survived the transition from Qing to Republican law, the host of non-guoshi negligence statutes to be found in Qing law, from killing with a bow and arrow to losing control of a fire, were subsumed with relative ease under the Republican-era guoshi statute. After all, both the Qing guoshi statute and those additional statutes had dealt with crimes of negligence. Thus the Qing-era statute “Incompetent Physicians Who Kill or Injure,” essentially a statute covering medical malpractice, found a new home under the category of guoshi homicide (Johnson, 1997: 281; Xue, [1905] 1970: A. 297.00). Take, for example, the 1929 case of Chen Qilan, who was brought before the Beijing Municipal Court for his part in the death of his patient, Xiong Xingchi (Beijing difang fayuan: J65–4/377–80). Xiong had come to the doctor for treatment of an infected leg. Chen testified that he did his best to cure the patient: “I reflected on the matter and I did nothing wrong. [Xiong Xingchi] had an infected lump on his upper left leg. The name of this disease is liuzhu 流注. I treated him by making incisions.” Chen went on to argue that Xiong had died not as a result of the operation, but because he caught a cold 感冒 while recovering. The court did not buy Chen’s story. It noted that after his surgery there was uncontrollable bleeding. The court brought in a medical

guoshi killing

245

expert from the Beijing German Hospital (Beiping Deguo yiyuan). This doctor, Ke Li Shi, testified that Xiong had been suffering from advanced-stage tuberculosis 進步之肺痨 as well as the “infected swellings,” the latter of which he likely obtained while taking care of his daughter, who was suffering from dysentery 痢疾. The patient bled, the doctor said, “because his blood vessels had been torn, damage caused during the operation.” Another doctor, Dong Ziwen, confirmed that the real illness was in the patient’s lungs and the swellings were a secondary matter. The Beijing court found Chen guilty of negligent homicide in the course of business (Article 291 of the 1928 Code) and ordered him to pay a fine of 100 yuan. The Hebei High Court changed this sentence to four months in prison, a change upheld by the Supreme Court the following year. In issuing its opinion, the Supreme Court came down harshly on Chen, noting in its official written opinion on the case, “The accused did not investigate the illness [properly], and heedlessly 貿然 cut open the infected lumps, so that the sick man’s blood vessels were irreparably damaged and he died.” Negligence was also the verdict in a 1941 case from the local court in Chongqing, Sichuan, which replicated the situation covered by the statute on killing with a cart or horse (Article 296.00) during the Qing (Sichuan gaodeng fayuan: 13482, 13484). This case involved a driver named Chai Guitang. On the day in question, according to witnesses, Chai was driving at a “high speed,” and “not paying attention” 不注意 to the road. He struck a five-year-old boy, leaving the boy dead from a gruesome injury to the head. The court found Chai guilty of guoshi homicide in the course of business. This ruling was upheld by the Sichuan High Court: “The appellant made his living as a driver. When driving a truck he ought to pay very close attention to relative distance. . . . And [he] certainly cannot use the huge size of the truck as an excuse for not paying attention. . . . It clearly is negligent killing in the course of business.” Accident, Negligence, or Neither? Given that the guoshi homicide statute had always included accidental homicides in the past, occasionally during the Republican period a Qing-esque accidental guoshi case would make its way through the local courts, only to be overturned at the provincial or national level. In addition, the consolidation of homicide offenses into the two main categories of negligent and intentional homicide often caused confusion on the part of the courts as to how to adjudicate an offense, leading to mistaken sentences of guoshi. Sometimes the courts’ quoting of guoshi laws revealed a confusion about which meaning of guoshi was in effect at the time—the Qing definition or

246

neighbors

the Republican one. Such was the case in a 1914 appeal filed by Chen Hongjiu at the Supreme Court (Daliyuan xingshi panjue quanwen huibian, n.d.: 1721). Chen had served as the sub-prefect 廳同知 in Songtao, Guizhou, during the last years of the Qing. During the turmoil following the fall of the Qing, Chen had fled to Hunan, taking his official seal and some government money with him. Another man, Zhou Lian, had been newly appointed to his post. Eventually Chen returned home, and there followed a major disagreement over who ought to hold the sub-prefectural office. Things escalated and eventually a melee ensued. Knives were wielded, shots rang out, and when the dust cleared two of Zhou Lian’s supporters had been shot and killed while another had sustained fatal stab wounds. The Guizhou High Court had found Chen Hongjiu guilty of, among other crimes, guoshi killing and injury for the deaths of Zhou Lian’s men. The rationale the High Court provided was that the deaths could not have been foreseen 意料所不及 and thus constituted guoshi homicides, paraphrasing the Qing Code article on guoshi killing. The Supreme Court struck down the ruling, calling it “inappropriate.” The Supreme Court was right—the definition that the High Court was using came from the Qing Code, and not the Provisional Criminal Code in effect at the time of the case.21 If the High Court wanted to call this a case of accidental homicide, then there was no need to sentence Chen Hongjiu for a crime in this matter, since, under Republican law, accidental homicides were no longer to be punished. If Chen was guilty of shooting at the troops on purpose, the statute on intentional killing would come into effect (see below). In either event, the guoshi law could not be used.22 Consider also the 1933 trial of two brothers from Shandong, Zhao Shigong and Zhao Shixun, who were proprietors of a shoe store and employers of an apprentice named Wang Kedong (Sifa gongbao: 1935.2.28). On the night of May 8, 1933, a fire started in a bicycle shop and spread to the shoe store, which was located upstairs. While Zhao Shigong made it out the door, Zhao Shixun and the apprentice “jumped out of the [shoe store’s] east window to escape.” During his fall, Wang Kedong sustained a head injury from which he died. Though blame for starting the fire was imputed to another man, Zhao Shigong and Zhao Shixun were found guilty of negligence for failing to take proper care of their apprentice. The Supreme Court overturned this ruling. It noted that the criminal code defined negligence as “failing to exercise the degree of care which he should have and could have exercised” (Criminal Code of the 21 Given that guoshi was nowhere defined in that new code, such confusion was understandable. 22 The Supreme Court provided a second, evidentiary ground for overturning the ruling—it was unclear who had actually killed Zhou Lian’s men.

guoshi killing

247

Republic of China, 1928: 6). However, the court ruled that in this case, given the circumstances of the fire, the fact that it was night, and that one of the brothers had chosen the same escape path as Wang Kedong, the brothers had done nothing wrong. Sentencing two men for negligent homicide on the basis of not properly saving a person from a burning building that they themselves were trying to escape would be more likely to appear under the Qing definition of guoshi killing. Under Republican law, this case should have fallen well outside the scope of negligence laws, unless the defendants had done something to start the fire or fan its flames; the Supreme Court corrected the misruling of the lower court and exonerated the defendants. Consider also the 1928 case of Kuai Desan for the guoshi killing of his neighbor, Gao Li Shi, and her daughter in Pi county, Sichuan (Sichuan gaodeng fayuan: 2232). Gao Li Shi, a mother with a history of suicide attempts, had been sent home to her natal family. On the day she died, she had gotten into an argument with her father’s neighbor, and it escalated. Gao Li Shi sustained a minor wound and went into a rage. Her daughter in tow, she jumped into a nearby river, drowning them both. The local court sentenced Kuai Desan to guoshi killing and ordered him to pay a fine of 500 yuan, 200 yuan to be paid to the court and 300 to Gao Li Shi’s husband. Such a fine constituted the maximum penalty allowed under the law (Zhanxing xin xinglü, [1912] 1988: A. 324). Neither Kuai Desan nor the widower were satisfied with the verdict, and both appealed to the High Court. Upon investigation, the High Court found that Kuai Desan had not forced Gao Li Shi to kill herself: “loud voices” may have been heard coming from his home, but Kuai had not pursued Gao Li Shi out the door and had not been there when she jumped into the river; witnesses had testified to this much. In addition, Gao Li Shi’s history of suicide attempts ought to be considered. The court thus ruled: “Gao Li Shi jumping in the river carrying her daughter was suicide 自盡. Kuai Desan was not in any degree responsible, and it cannot be called guoshi.” The court thereby reversed the lower court’s ruling, ordering that any funds paid revert to Kuai Desan. A crime may have been committed, but it did not conform to the definition of guoshi intended in Republican law. Kuai Desan had not been remiss in paying attention to something. Even if he did strike Gao, that constituted a crime of assault, not negligence.23 In this case and in others, lower courts encountered problems reconciling the multiplicity 23

The High Court held Kuai responsible for his injury of Gao Li Shi with an iron tool. In Qing times Kuai Desan could have been convicted under Article 299 of the Qing Code, “Forcing Another to Commit Suicide.” Under this statute, if evidence was found that the offender had frightened the victim (here, the injury with iron tool), a sentence of 100 strokes of the heavy bamboo was merited (Xue, [1905] 1970: A. 299.00).

248

neighbors

of old statutes with the pithy new Republican code. If a homicide was not intentional, the only avenue to justice seemed to be adjudication using the new guoshi laws. Negligence or Intention? The question of how to differentiate the new conceptual category of negligent killing from intentional killing presented an additional problem for jurists. As noted in the earlier discussion of the Republican-era legal codes, this problem likely originated in the tricky wording of the articles defining the terms “intentional” and “negligent” (as well as the initial absence of any such definitions in the codes). Part of the intentional homicide article in the 1928 criminal code (Article 26) read: “An act is done intentionally when the actor . . . foresaw 豫見 that it would occur, and moreover its occurrence was not contrary to his intentions” (modified from Criminal Code of the Republic of China, 1928: 6). The definition of the term “negligent” (Article 27) contained the following clause: “An act is done in negligence when the actor, . . . although he foresaw 豫見 that the act could occur, firmly believed that it would not occur” (modified from Criminal Code of the Republic of China, 1928: 6). Since both definitions included the word “foresight,” and because assessing a perpetrator’s state of mind was always a tricky matter, the courts bandied around the term, sometimes ruling an incident intentional and sometimes ruling it negligent. This would never have occurred under Qing law, where guoshi and intentional killing, instead of being neighbors, were placed on opposite sides of the broad spectrum of degrees of intent. A 1916 Supreme Court response to a query from Jilin province illustrates the Republican courts’ difficulties differentiating the two main homicide categories (Daliyuan jieshili quanwen, 1931: 431). In this instance, according to the court records, “There was enmity between A and B. B was sitting in a ring with C, D, and E. From a crack in the window, A shot at B.” Everyone in the room was injured, though none died. Sentencing A for his crime against B was simple: he had committed attempted murder. Sentencing A for his crime against the rest of the men, however, was more difficult. The lower court presented two relevant opinions on the matter. First, the court suggested that since the men were sitting together in a ring, the possibility of injuring all of the men was “of course something A could have foreseen 豫見.” Thus, he ought to be sentenced for attempted murder of the rest of the men, too. Second, the court suggested that “not only did A not have the intent to kill C, D, and E, he did not have the intent to injure [them]. However, because he was careless to the point that injury resulted 不注意致傷害, he ought to be sentenced in accordance with negligent injury laws.”

guoshi killing

249

The intention to harm one of the people in the room would seem adequate grounds to sentence A under intentional (attempted) homicide laws.24 And since the other injuries were inflicted in the course of an attempted homicide the negligent homicide statute would only seem applicable if it was twisted to apply to the negligent commission of a crime—failing to carry out a crime with enough accuracy. The laws on negligence were not meant to apply to such a situation. The Supreme Court therefore corrected the lower court’s ruling, explaining that it was the issue of “whether or not A had foresight [that would] determine the verdict.” If A had foresight, presumably then he ought to be sentenced according to the first option provided by the court, thus under the standard intentional (attempted) homicide laws.25 A case forwarded to the Supreme Court from Jiangxi province later that same year also reveals the problems the courts experienced in differentiating negligence from intent. In this case, four men strangled a fifth man and believed they had killed him (Daliyuan jieshili quanwen, 1931: 489). Thereafter, one of their number dumped the body into water to conceal the crime 拋屍滅跡. After the body was discovered, it was revealed that the man had not died from the strangulation but from the drowning. The Jiangxi court asked the Supreme Court for advice as to what exact crime had been committed by the person who had disposed of the body; even though the dumping killed the man, the perpetrator had intended to kill the man at the time of the strangulation, not when he was dumping the body. In the eyes of the Supreme Court in this case, one possible verdict was that in addition to a charge of attempted murder shared with the other three criminals involved in the case, the perpetrator was also guilty of the crime of Article 258: “damaging, abandoning, or stealing human remains” (modified from Provisional Criminal Code of the Republic of China, 1915: 65). Additionally, the Supreme Court ruled that although the perpetrator did not have the intent to kill 殺意 the victim when disposing of what he thought was a corpse, there may have been negligence involved (如有過 24

25

Such would have been the case under Qing law, where an offender was sentenced under the intentional homicide statute if he intended to kill one person but mistakenly shot another instead. See Xue, (1905) 1970: A. 292.00. On the surface this ruling accords with the intentional homicide statute discussed above. However, it also accords with the negligence statute, because the court does not show us how it determined that hitting the other three men was “not contrary to [the offender’s] intentions.” Nor does the court show us that it was not a case of the offender “believing that [the incident] would not occur.” In the end, the sheer presence of foresight itself was enough to merit use of the intentional homicide laws, even though according to the letter of the law more than just the presence of foresight was required to constitute an intentional act.

250

neighbors

失,仍應以過失殺論). This again would have been an instance of being negligent in the commission of a crime, rather than negligently acting to cause a crime. The confusion the courts displayed in these cases reveals that the courts clearly found permeable bounds between intentional homicide and negligent homicide. Qing law had provided situational guidelines to help identify the category of intent involved in a crime. And Qing law possessed a broad, detailed continuum of intent categories. In Republican criminal law, however, there were only two mental categories of criminal culpability. Differentiating clearly between the two was an important matter indeed. After all, the punishment for negligent homicide was merely a maximum fine of 500 yuan until 1928, and thereafter a maximum fine of 1,000 yuan or a maximum period of two years’ imprisonment (Zhanxing xin xinglü, [1912] 1988: A. 324; Criminal Code of the Republic of China, 1928: 79). The punishment for intentional homicide, on the other hand, was between ten years’ imprisonment and death (Zhanxing xin xinglü, [1912] 1988: A. 311; Criminal Code of the Republic of China, 1928: 77). That meant a difference of eight years’ imprisonment between the maximum penalty for guoshi homicide and the minimum penalty for intentional homicide, a rather large disparity in punishment. Given how ambiguous and easily confused the definitions of intention and negligence in the Republican codes were, mistakes were easy, as were miscarriages of justice. Conclusion Guoshi killing occupied the low end of the continuum of homicide offenses in Qing law. This category covered acts committed by accident, as well as those committed with a low level of negligence. The Qing Code included, in addition to the guoshi homicide statute, a large collection of crimes that were considered more serious than accidental and low-level negligent crimes, but less serious than crimes of intentional harm. Those statutes included killing with a bow and arrow (Article 295), killing with a cart or horse (Article 296), killing with pit bows (Article 298), and losing control of a fire (Article 382). These intermediate crimes were distinguishable by their punishments. Unlike the penalties for guoshi killing, these punishments were not automatically redeemable. However, they were still less severe than the sentences of capital punishment assigned to crimes of intentional harm. Finally there was the category of killing at play (Article 292), covering crimes only one small step down from those of intention, and meriting the same initial sentence as homicides committed during the course of an affray. Collectively, these numerous

guoshi killing

251

statutes represented the highly detailed and finely differentiated late imperial legal category of criminal negligence. With the introduction in 1912 of a criminal code based on modern European and Japanese models, acts committed without intent or negligence were no longer considered crimes. As a result, accidental killings, adjudicated in late imperial times under the guoshi statute, were no longer to be prosecuted. Furthermore, the numerous degrees of criminal negligence found in the late imperial codes, from the low-level negligence of guoshi offenses to the middlelevel negligence of killing with a bow and arrow and similar offenses, to the reckless negligence exhibited in killing-at-play crimes, would all be condensed into one general category of negligence. The late Qing and Republican lawmakers had set out to create a streamlined, simplified new criminal code characterized by clear concept-oriented statutes. Though the criminal code that resulted succeeded in containing almost solely concept-oriented statutes, those statutes were so broadly and ambiguously defined as to make differentiating even between negligence and intention a tricky matter. Unlike Qing law, wherein guoshi homicide and intentional homicide occupied opposite sides of the intent continuum, under Republican law they were the only two categories of mental culpability, and were defined in similar terms. So similar, in fact, were the definitions of the two categories that they were often confused by jurists, a problem never encountered in Qing times. Most jurists of the time cast Republican criminal law as a marked advance over that of the Qing. However, for homicide law such claims of advancement need to be qualified. A complex system was replaced by a simpler one, to be sure, but that simpler system had limitations when put into practice. The Qing system was different, but it was highly sophisticated. In this case, modernization in accordance with Western models did not necessarily mean progress. References Alabaster, Ernest. [1899] 1968. Notes and Commentaries on Chinese Criminal Law and Cognate Topics, with Special Relation to Ruling Cases. Taipei: Ch’eng Wen Publishing Co. Baxian dang’an [Ba County archives, housed at the Sichuan Provincial Archives]. Cited by case number. Beijing difang fayuan [Beijing Local Court, case records housed at the Beijing Municipal Archives]. Cited by case number.

252

neighbors

Bodde, Derk and Clarence Morris. 1967. Law in Imperial China: Exemplified by 190 Ch’ing Dynasty Cases. Philadelphia: University of Pennsylvania Press. China Law Review. Shanghai: Law Department of Suzhou University. Criminal Code of the Republic of China. 1928. Trans. S. L. Burdett (with the collaboration of Lone Liang). Shanghai: Shanghai linshi fayuan yinxing. Criminal Code of the Republic of China. Bilingual Edition. 1960. Trans. L. J. Fuller and H. A. Fisher, Jr. Taipei: Sino-American Legal Series. Da Ming lü fuli zhujie [Code of the great Ming with appended substatutes and collected commentaries]. 1993. Ed. Yao Siren. Beijing: Beijing daxue chubanshe. Da Ming lü jijie fuli [Code of the great Ming with collected commentaries and appended substatutes]. [1908] 1989. Revised Zheng Jifang. Yangzhou: Yangzhou guji shudian. Da Qing xin xinglü [New criminal code of the great Qing]. 1911. Shanghai: N.p. Cited by article number. Da Qing xinglü zongze cao’an. 1907. Comp. Shen Jiaben. Beijing: Falü guan. Cited by article number. Daliyuan jieshili quanwen [Collected explanations of the laws by the Supreme Court]. 1931. Ed. Guo Wei. Shanghai: Shanghai faxue bianyishe. Cited by appeal number. Daliyuan xingshi panjue quanwen huibian [Collection of criminal case verdicts from the Supreme Court] n.d. Unpublished manuscript. Ed. Huang Yuansheng. Taipei: National Chengchi Univ. Legal Research Center Collection. Fletcher, George P. 1971. “The theory of criminal negligence: a comparative analysis.” University of Pennsylvania Law Rev. 119, 3: 401–38. Great Qing Code. 1994. Trans. William C. Jones (with the assistance of Tianquan Cheng and Yongling Jiang). Oxford: Clarendon. Guo Wei. 1946. (Xinbian) Xingfa xue ge lun (Essays in the study of criminal law [new edition]). Shanghai: Faxue bianyishe. Huang, Liuhong. 1984. A Complete Book Concerning Happiness and Benevolence: A Manual for Local Magistrates in 17th Century China. Ed. and trans. Djang Chu. Tucson: University of Arizona Press. Huang Yuansheng. 1991. “Shen Jiaben falü sixiang yu wan Qing xinglü bianqian” (Shen Jiaben’s legal philosophy and the evolution of late Qing criminal law). Ph.D. diss., National Chengchi University Taiwan. ———. 2002. “Dizhi Zhongguo zuihou yibu chuantong xingfa dian: jianlun wan Qing xingshi fa jindai hua de guocheng” (The traditional criminal code in the last part of imperial China: discussion of the late Qing legal modernization process). In Gan Tiangui jiaoshou liu zhi zhushou lunwen ji (Collected essays in celebration of the 60th birthday of Professor Gan Tiangui). Taipei: Xuelin. Hulsewe, A. F. P. 1955. The Remnants of Han Law. Leiden: E. J. Brill. Johnson, Wallace. 1979. The T’ang Code, Volume 1: General Principles. Princeton, NJ: Princeton University Press.

guoshi killing

253

———. 1997. The T’ang Code, Volume 2: Specific Articles. Princeton, NJ: Princeton University Press. Kadish, Sanford H. and Stephen J. Schulhofer. 1989. Criminal Law and Its Processes: Cases and Materials. 5th ed. Boston: Little, Brown. Lafave, Wayne R. 2000. Criminal Law. 3rd ed. St. Paul, MN: West Publishing. Lafave, Wayne R. and Austin W. Scott, Jr. 1986. Criminal Law. 2nd ed. St. Paul, MN: West Publishing. MacCormack, Geoffrey. 1988. “The Tang and Ming law of homicide.” Revue Internationale des Droits de l’Antiquité 35: 27–78. ———. 1990. Traditional Chinese Penal Law. Edinburgh: Edinburgh University Press. ———. 1996. The Spirit of Traditional Chinese Law. Athens: University of Georgia Press. Meijer, Marinus. 1967. The Introduction of Modern Criminal Law in China. 2nd ed. Hong Kong: Lung Men Bookstore. ———. 1978. “The concept of ku-sha in the Ch’ing code.” Pp. 85–114 in L. Lanciotti (ed.), Il dirrito in Cina. Florence: Olschki. ———. 1980. “An aspect of retribution in Chinese law.” T’oung Pao 66: 4–5. Morohashi Tetsuji. 1992. Dai kanwa jiten (Great Chinese-Japanese dictionary). 13 vols. Taipei: Landeng wenhua shiwu gufen youxian gongsi. Mulingshu [Collection of essays on magistrates’ duties]. 1848. N.p. Neighbors, Jennifer. 2004. “Criminal intent and homicide law in Qing and Republican China.” Ph.D. diss. University of California, Los Angeles. Ping Ping. 1930. “Lun guoshi fan” (On negligence crimes). Falü pinglun 35, 9: 1–3. Provisional Criminal Code of the Republic of China. 1915. Trans. T. T. Yuen and Tachuen S. K. Loh. Beijing: Commission on Extraterritoriality. Qinding da Qing xianxing xinglü anyu [Annotated criminal code of the Great Qing, currently in use]. [1910] 1995. Comp. and revised Shen Jiaben. In Xuxiu siku quanshu (Complete library of the four treasuries continued). Shanghai: Shanghai guji chubanshe. Sichuan gaodeng fayuan [Sichuan High Court case records housed at the Chongqing Municipal Archives]. Cited by case number. Sifa gongbao. 1930–1935. Nanjing: N.p. Cited by date. Song xingtong [Song code]. 1984. Ed. Wu Yiru. Beijing: Zhonghua shuju. Tanglü shuyi [Tang code and commentary]. 1996. 2nd ed. Taipei: Taiwan shangwu yinshuguan. Cited by article number. Wallacher, Benjamin E. 1983. “The Chinese offense of homicide through horseplay.” Hanxue yanjiu 1, 1: 259–316. Williams, Glanville L. 1961. Criminal Law: The General Part. London: Stevens and Sons. Xianshen anjian [Immediate examination cases from the Board of Punishments collection at the First Historical Archives of China]. Cited by bureau and case number.

254

neighbors

Xing’an huilan [Conspectus of legal cases]. 1886. Shanghai: Tushu jicheng ju. Xingfa dier ci xiuzheng an [Second revised draft criminal code]. [1918] 1973. In Falü cao’an huibian. Taipei: Chengwen chubanshe. Cited by article number. Xingfa zongze [General principles of criminal law]. 1944. Ed. Zhao Chen. Shanghai: Shangwu yinshuguan. Xue Yunsheng. [1905] 1970. Duli cunyi (Lingering doubts upon reading the substatutes). Ed. and punctuated Huang Jingjia. 5 vols. Taipei: Chinese Materials and Research Aids Service Center. Cited by article number. Zhanxing xin xinglü [Provisional criminal code]. [1912] 1988. In Yang Honglie (ed.), Zhongguo falü fada shi (History of the development of Chinese law). Taipei: Shangwu yinshuguan. Cited by article number. Zhonghua minguo xingfa pan jieshi yi quanshu [Complete book of explanations and meanings of criminal case verdicts from the Republic of China]. 1972. Taipei: Hua’an.

Part Three

Tax · Education · Local Governance

∵

chapter 8

Between the State and the Village: Land Taxation and “Substantive Governance” in Traditional China1 Huaiyin Li Introduction Two contrasting approaches have shaped traditional interpretations of villagestate relations in imperial China. One presumes an autocratic state capable of penetrating all the way down to every village and household through the imposed groupings of rural households known as baojia for neighborhood surveillance and lijia for adult male registration and tax collection. These devices, we are told, allowed the state to use the rural agents as its tool to exert authority in local society. By putting all aspects of rural life under its direction and supervision, the state successfully prevented the growth of any forms of local autonomy or self-government.2 The other approach argues that autonomous communities operated against government control. The idea that Chinese society operated relatively autonomously was quite popular in the early twentieth century in both Western scholarship and Chinese writings. In his analysis of Chinese social organizations, for example, Max Weber consistently emphasized the existence of communal autonomy and its tension with the patrimonial monarchy. According to Weber, the autonomy and cohesion of Chinese villages stemmed from local self-governing bodies, which carried out duties such as road improvement, river dredging, local defense, criminal control, schooling, and funeral and burial services; no less important were the clan organizations in community 1 This article is adapted from Chapter 5 and parts of Chapters 1 and 12 of the author’s Village Governance in North China, 1875–1936 (Stanford University Press, 2005). 2 See, for example, Hsiao 1960; Ch’ü 1962; Fu 1993; and Andrew and Rapp 2000. The Chinese historians who embraced the notion of “Asiatic mode of production” tended to describe the traditional Chinese state as a form of despotism intrinsic to the stagnant Asiatic society (Wu Ze 1993; Gu Zhun 1982, 1999). The more “orthodox” Marxist historians on the other hand characterized the Chinese state as a centralized bureaucracy inimical to the emerging “democratic” ideas in the last phase of the “feudal” society in Chinese history (Hou Wailu 1979; Hu Rulei 1979).

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_�10

258

li

life, as evidenced in the supremacy of clan elders’ power and ancestral halls. It was clan solidarity based on the cult of ancestors that “withstood the ruthless encroachments of the patrimonial administration,” resulting in a constant clash between patrimonial rule from above and the clans’ strong counterbalance from below (Weber 1951: 86–87). Despite their contrasting views of local administration, the above two approaches have one presumption in common—that is, the dichotomous opposition between state and society. Local governance was perceived as a realm of either omnipresent state influence or predominantly local auto­ nomous practices. To a degree, this paradigm of dichotomous opposition between state control and local autonomy may be ascribed to the fact that earlier scholarship on local government was largely based on traditional source materials, mainly official documents, local gazetteers, and private writings. As Kung-chuan Hsiao complained, these sources are marred with “possible biases, inaccuracies, partiality, dishonesty, or carelessness,” for they were written from the standpoint of either the government or the literate, particularly the gentry (Hsiao 1960: vi–viii). It is no wonder that one often finds in those writings a juxtaposition of the idealized image of state control with various counter-ideals that were depicted as full of flaws. Moreover, in the absence of reliable empirical studies, scholars could only perceive and interpret the different forms of village governance in the context of theoretical constructs then available and appealing to them. The very absence of a formal government below the county level and the prevalence of self-governing bodies in local society prompted scholars to treat them as forms of self-government in opposition to the autocratic state. Likewise, in the 1950s and 1960s when the theory of “oriental despotism” was prevalent (see Wittfogel 1957), scholars tended to perceive the Chinese state as despotic and penetrative, and dismissed baojia and its variants merely as tools of state control that had nothing do with autonomy and self-government (Hsiao 1960; Ch’ü 1962; Balazs 1964; Watt 1972; and Fu 1993). Dissatisfaction with this paradigm has caused scholars to seek a sophisticated alternative construct for understanding the complexity of village-state relations in imperial China. The constant tension and conflicts between state and society did not preclude their mutual dependence in local administration.3 3 Philip Kuhn pointed out this possibility as early as in the 1970s, noting that administrators in imperial China tended to rely on local social organs as the basis for their control in order to produce the desired results, as indicated in the fact that baojia unit was usually based on a natural village rather than a group of 100 households as the state designed. The employment of local indigenous social organs, Kuhn suggested, produced the characteristic a­ mbiguity

between the state and the village

259

In the resolution of civil disputes in Qing China, for instance, many disputes were resolved by neither the informal method of community or kin mediation nor formal court adjudication, but instead by the working of a “third realm” of civil justice, in which both the formal and informal systems participated and interacted with each other (P. Huang 1993). In the day-to-day operation of the county government, for another example, the ostensibly “formal” county court employed a large number of “informal” or illicit clerks and runners who were not subject to any statutory administrative regulations. While acting as agents of the state, they were simultaneously an occupational group rooted in the local community, thus functionally mediating between state and society (Reed 2000: 248). Effective administration of local society, according to adherents of the Neo-Confucian ideal of self-governance, did not just depend on administrative systems and measures imposed from above. Equally important were cooperative practices and welfare programs in village communities. In the opinion of Chen Hongmou, a model bureaucrat in eighteenth-century China, the employment of local initiatives and tighter control by the state were complementary rather than incompatible in the actual practice of local governance (Rowe 2001: 335). Together, these recent findings suggest a new direction in which we can explore a more dynamic and complicated relationship between state and society. The dichotomy between state control and local autonomy that prevailed in earlier studies of local administration was inadequate to explain interwoven arrangements between state institutions and village in their actual operation. To interpret the complex realities of local governance, we need an alternative conceptual framework built on solid empirical researches. For this purpose, this study further moves attention from the activities of the bureaucrats and their underlings at and above the county level, which have concerned the aforementioned studies in recent years, to the governing process in village communities, focusing on Huailu county in the south-central Hebei plain in North China. Village administrations in late nineteenthand early twentieth-century Huailu employed neither the original statutory of sub-county administrative systems, in which the principles of control and autonomy were not entirely separable. However, without solid evidence to support this hypothesis, Kuhn quickly pointed out that the idea about a mutually supportive relationship between state control and local autonomy, found in the minds of Qing bureaucrats and literati, did not work at all in actuality because of the obvious weaknesses of the bureaucratic system, such as the rule of avoidance, which prevented the magistrate from developing interests in improving the welfare of local society (Kuhn 1975).

260

li

baojia and lijia systems that were officially enforced in early Qing nor illegal ­arrangements banned by the government. What prevailed in the local communities was a form of voluntary cooperation among villagers who shouldered administrative tasks that had been performed by the baojia and lijia personnel. The key position in the cooperation was the xiangdi, who acted as an intermediary between the county yamen and his village. Chosen from local dwellers by annual rotation, the xiangdi performed a variety of tasks delegated by the county magistrate. He was required, for example, to report local crimes and help yamen runners to arrest criminals or bring summoned disputants to court. He was also responsible for issuing official deeds, prompting the payment of deed taxes, and investigating untaxed deeds and unregistered land. And it was his duty to collect irregular levies or facilities on the magistrate’s instruction. The xiangdi performed the same functions as the rural agents under the previous baojia system. However, the xiangdi was not just an agent of the government. He also served the needs of fellow villagers and represented his community before the county yamen. This was evident in his roles in all activities pertaining to land taxation. Unlike the statutory tax system that required villagers to pay taxes individually, a common practice in Huailu and neighboring counties was for the xiangdi to pay in advance all of the taxes of the community members during the collection period, using public village funds or loans; he then collected his monies from individual households after the taxes had been paid. The villagers preferred this cooperative arrangement because the xiangdi’s collective payment of taxes saved them the time and expense of delivering the taxes individually. Moreover, it precluded the intrusion into local communities of tax-prompting yamen runners under the official tax system and also made it impossible for tax farmers from outside to extort additional taxes from individual taxpayers, a phenomenon not uncommon in many parts of North China. Subcounty administration in Huailu, then, was characterized by a variety of cooperative arrangements among community members. The key to understanding the prevalence of cooperation in local governance, I will argue, lay in the fact that most villages in the area under study were highly cohesive communities of predominantly owner-cultivators. Endowed with a secure ecological setting where the absence of frequent natural calamities minimized migration, the villagers developed over time tight kinship networks and a strong identity with the community. They cooperated in community projects that served the interests of all members and supported the enforcement of village conventions and shared principles. Such cooperative practices were of course not limited to the villages in Huailu and other counties of south-central Hebei. Similar cooperation existed in other areas as well, such as the lower

between the state and the village

261

Yangzi region and southeastern China, where comparable ecological conditions, property relations, and social networks prevailed (Li 2000). My examination of local taxation shows that the magistrate rarely interfered with the working process of those institutions when they functioned to generate the expected taxes and social order. He stepped in only when disputes arose that disrupted the normal operation of those institutions and when the community failed to mediate on its own. And the magistrate acted on those occasions only as an arbitrator. From the rulers’ point of view, the government’s assignment of tasks to local communities and the promotion of cooperative practices in the villages had two obvious advantages. First, it freed them from the mundane task of dealing with individual villagers in tax collection and police control and saved them the expense of hiring additional underlings to perform those tasks. Second and more important, it reduced improper and illicit practices in administrative activity. A deeply rooted conviction among the ruling elites throughout the imperial period was that the involvement of yamen underlings in local administration would inevitably result in illegal practices, for the self-interested and underpaid underlings were always in a position to engage in wrongdoing to enrich themselves at the cost of local people. Allowing the community to shoulder those official duties instead could solve the problem because local agents were always subject to the scrutiny of the community, and their abuses, if any, could be handled by the community itself. Therefore, the imperial rulers preferred to minimize government intervention in local governance and to encourage villagers’ voluntary cooperation in fulfilling their duties to the government. As long as the informal institutions of local communities proved able to meet the government’s need for tax income and local control, the state showed no inclination to extend its reach further down than the county level; instead it allowed local communities and their own agents to assume all administrative tasks of the government. The communal arrangements in Huailu villages, as we will find in this study, fell well within the scope of voluntary cooperation the state promoted. In this study, I use the term substantive governance to characterize the government’s noninterference, laissez-faire orientation and the predominance of informal practices in subcounty administration. This term emphasizes the fusion of government purposes into local, unofficial arrangements and distinguishes this reality from the long-used standard practices that prevailed in the formal, centralized bureaucratic system at and above the county level. Unlike formalistic administration, which ideally precludes informal elements and nonstandardized practices, substantive governance was a realm in which both the government and society participated and where governmental f­unctions

262

li

intermeshed with local arrangements.4 The primary goal of the state in this realm was to ensure that the public order be maintained and its financial needs be met to the extent that they would not jeopardize local stability. So long as these demands were satisfied, the state felt no need to involve itself in the process of local governance. Instead, in order to achieve its goals, the state opened the realm to local society and encouraged its participation when local initiatives did not infringe on state interests. Local society, too, found it to their advantage to develop cooperative, self-governing arrangements to deal with the government and minimize its often disruptive intrusion. We thus find in substantive governance a common ground where the interests of the state and the village society overlapped and where official functions and local institutions were inseparable. This was a realm, however, that must be distinguished from various illegal practices in local administration that encroached on the interests of the state and therefore suffered its ceaseless attacks and prohibitions. It should also be distinguished from forms of local “autonomy” that ruled out government influence from the community. This article examines the operational realities and dynamics of substantive governance by focusing on land taxation in Huailu county. Three different tax systems prevailed in different parts of Qing and Republican China, namely, the informal practice of voluntary cooperation in tax payment, which was most common in Huailu villages; the official system of self-delivery of taxes to the collection station (zifeng tougui), which was less popular in the county; and the illegal business of tax-farming, which in Huailu was limited to landholders who owed tax duties to a neighboring county. My examination of how these methods operated brings to light the critical role of village regulations on cooperative practices in taxation. Tax payment was most efficient where the village regulations obliged the xiangdi to advance taxes for his community and compelled community members to repay the xiangdi by a deadline. It was less smooth where people paid taxes individually under the xiangdi’s prompting. Least efficient was tax-farming, which caused most disputes between the tax 4 Max Weber’s interpretation of the Chinese legal system is suggestive to my conception of substantive governance. Weber describes the administrative system in imperial China as basically anti-formalist and of patriarchical character. Chinese rulers under this system tended to seek “substantive justice rather than formal law” (Weber 1951: 101–2). Weber’s assertion is not valid for the formal bureaucracy of the Chinese empire. The actual practice of civil justice at the county level, as we know now, was also largely based on legal codes and far from arbitrary (P. Huang 1996). However, the imperial rulers stipulated no concrete codes to regulate sub-country administration, which was based largely on established practices or initiatives from either the administrator or local society. As a result, local governance remained largely substantive rather than formalistic throughout the imperial and early republican periods.

between the state and the village

263

collectors and taxpayers in the absence of both village regulations and community mediation. This article begins with an analysis of the xiangdi’s performance in delivering taxes and the county yamen’s prompting. Then I look at how village regulations governing land taxation operated during the late Qing and Republican periods by examining disputes between the tax-advancing xiangdi and local taxpayers. The last two sections examine tax collection in villages under the tax-prompting xiangdi and on tax enclaves, respectively, which contrast with the cooperative practices in most villages. My concern is primarily with the changing relationships between the government and the village under different tax systems, as well as their implications for understanding the nature of governance in North China villages.

The County Government and the Xiangdi

Each year the land tax was collected in two periods. The first tax period (shangmang), in which half of the tax quota was to be collected, normally began on the second day of the second month of the lunar year and ended on the fifteenth day of the fourth month. The second tax period (xiamang), with another half of tax quota to be fulfilled, lasted from the second day of the eighth month to the fifteenth day of the tenth month. The deadline for the second period sometimes varied; it could be advanced to the first day of the tenth month or delayed to the first day of the twelfth month, depending on the actual process of tax collection as well as the government’s need for revenue (656.1.366, 1915–27).5 To start the collection season, the Revenue Office (hufang) (or the First Section during the Republican period) of the county government routinely prepared a notice on behalf of the magistrate and posted it on the gates of the wall surrounding the county seat as well as at public sites in market towns. At the same time, the office also sent to the xiangdi in each village a copy of the updated tax roll (zhengliang hongbo, literally, “tax-collecting red book”), and an official announcement of the tax-collecting method (zhengliang banfa) for the current year. The tax roll only listed the tax quotas of individual taxpaying households in the form of how many taels a household owed. It was the tax-collecting method that informed the taxpayer how many silver dollars 5 The Huailu county government archives of the late Qing and Republican periods, currently preserved at the Hebei Provincial Archives, are cited in this article by category number, subcategory number, file number, and the year when the file was created.

264

li

or ­copper coins were to be assessed for each tael of his tax quota. Both the notice and the announcement were to be posted at a public site in the village (656.1.426, 1915–34). During imperial times as well as in the first two Republican years, all taxes in Hebei (then Zhili) province were paid in copper cash. The tax quota listed in silver taels on the tax roll thus needed to be converted into copper cash. Since the conversion rate changed every year, so did the actual amount of taxes paid by landowners. It was widely believed that this method of tax payment, and the possible manipulation of the conversion rate, created opportunities for the collection of extra tax monies (e.g., H. Huang 1918; Li Hongyi 1977 [1934]). It is also worth noting that, aside from the statutory land tax quota, taxpayers had to pay a number of additional charges attached to the land tax. In Huailu county, it had been an established practice since 1831 that for each tael of the land tax a surcharge of 0.175 tael was assessed to cover the “melting fee” (huohao, the fee to cover the loss after melting irregular pieces of silver into standard bullion), the remittance fee, and the administration fee, and another surcharge of 0.0218 tael to cover the collectors’ expenses for stationery and meals. In addition, the taxpayer needed to pay three wen as a tax-receipt fee (656.1.103, 1913). To simplify tax payment, the Zhili provincial government decided to collect land taxes in silver dollars (yuan) instead of copper cash (wen), beginning on April 1, 1914. The taxpayers now needed to pay 2.30 yuan for each tael of tax quota; all the surcharges mentioned above were abolished, except the receipt fee (656.1.216, 1914; 656.1.396, 1915). But small taxpaying households with a tax of less than one yuan still needed to pay in copper cash. Every day the local chamber of commerce announced the conversion rate between the silver tael (liang) and silver dollars (yuan). It is hard to judge the extent to which the government clerks may have profited by manipulating the daily conversion rate at a cost to the taxpayers. The Huailu magistrate simply denied any extra collection of taxes resulting from the conversion rate in his report to the Financial Department of Zhili province in 1920. According to his assertion, the clerks’ activities during the collection season were under his frequent and close surveillance (656.1.1203, 1920). The Xiangdi’s Delivery of Taxes The majority of the xiangdi in Huailu delivered the tax monies promptly. Table 9, based on the monthly reports of tax collection available to me, shows the cumulative amount of taxes collected each month. In 1914, for example, the first tax period started on February 24. By the end of March, 26.61 percent of the county’s tax quota had been met. In other words, the xiangdi of the county

265

between the state and the village

had delivered 53.22 percent of the taxes in the first month for the first tax period. By the end of April, the county had met 32.9 percent of the tax quota for the whole year or 65.8 percent for the first period. The first period ended on May 6; it is hard to judge how much the xiangdi had paid by that deadline. By the end of that month, however, the xiangdi had turned in 49.11 percent of the whole year’s taxes or, in other words, almost all of what was owed for the first period. The same was true for the second tax period of that year. By the end of the first month (October), the xiangdi had delivered 76.78 percent of the taxes for the whole year or over 55 percent for the second period. Data for the next month are not available. However, by the end of December (the third month), the taxpayers in Huailu had paid almost all of the taxes owed. The data of 1917 and 1920 show no big difference from that of 1914, except that the two tax periods started later in 1917 and still later in 1920 (almost one month later than in 1914) (see table below).

Tax quota

January February March April May June July August September October November December

Monthly reports of cumulative land taxes collected in Huailu County, 1914, 1917, and 1920. 1914 1917 1920 Tax Tax Tax Collected Percent Collected Percent Collected Percent 54,878 100.00 53,272 100.00 53,167 100.00

0 0 14,605 18,055 26,951 0 0 0 0 42,131 – 54,403

0 0 26.61 32.90 49.11 0 0 0 0 76.78 – 99.68

0 0 1,026 9,959 19,063 26,624 0 0 0 0 32,120 38,361

0 0 1.93 18.69 35.78 49.98 0 0 0 0 60.29 72.01

0 0 0 1,481 16,431 26,631 0 0 0 0 35,407 41,778

0 0 0 2.79 30.90 50.09 0 0 0 0 66.59 78.58

Sources: Jingzheng diliang yuebaocegao (Draft monthly report on the collection of land tax) (656.1.216, 1914; 656.1.719, 1917; 656.1.1204, 1920). Note: Data on taxes collected in the third month of the second period in 1917 and 1920 (i.e., January of the next year) are not available.

266

li

In general, the taxpayers, primarily the xiangdi, delivered in the first two months of a collection period approximately 30 percent of their tax quotas for the whole year or 60 percent of their duties for the collection period. As a rule, by the end of the third month they had met almost 50 percent and even more of the whole year’s tax quotas or almost all of what was due in a given period. We may then surmise that by the mid of the third month—that is, the end of a collection period—the taxpayers had paid 40 percent of their taxes for the whole year or 80 percent for the period. Since the xiangdi usually delivered all of the money due in a single visit to the county seat during a collection season rather than in several visits, it is safe to say that the vast majority of xiangdi in Huailu advanced land taxes for his village or his pai before the deadline. Tax Prompting by the County Government In order to collect as much tax as possible by the due date, the county government took several steps. Two months after the start of collection, the magistrate normally instructed village heads in his county to “supervise and hasten” (ducui) the xiangdi’s tax payment by the deadline. If the xiangdi remained inactive, the village head had to report the situation to the magistrate. However, few village heads abided by the magistrate’s instructions at the expense of their relationship with the xiangdi; in fact I found no case in which the village head accused the xiangdi of delinquent tax payment (656.1.953, 1918–21). As a second step, the county office in charge of the land tax would submit to the magistrate a list of villages that had not yet paid their taxes. The magistrate would summon in batches all those dilatory xiangdi (usually one collective summons for five villages) to court, where he “instructed and urged” ( yucui) them to pay the tax promptly. This method, however, again proved to be ineffective. In an extreme case, the xiangdi of Fujiazhuang village, who was responsible for a tax quota of 41 taels for his village, was named in as many as six collective summonses during the first tax period of 1920 but never once went to court. Consequently, the tax office advised the magistrate to detain the xiangdi. The xiangdi immediately paid the taxes upon receipt of the notice, alleging that he had failed to obey the earlier collective summons because “he had been ill for over two months and just recovered only recently” (656.1.953, 1918–21). The magistrate also used the officers of the eight policing wards of the county, instructing them to send policemen to villages in their charge to “watch and prompt” (shoucui or zuocui) the xiangdi’s tax payment. If the xiangdi failed to pay the taxes in full by the deadline, the policemen then might either submit to the magistrate a list of the delinquent xiangdi or bring them to court outright (656.1.953, 1918–21).

between the state and the village

267

After the closing date of the collection period, the magistrate would use his own bailiff ( fajing or zhengwu jing in the 1930s) instead of the ward police to deal with delinquent xiangdi. Each bailiff was assigned several xiangdi to prompt (hence called jingcui). Those who failed to clear their debts were invariably accused. On receipt of the accusation, the magistrate always responded with a summons to the accused. In fact, to make their promptings effective, the bailiff often first asked for a summons on the delinquent xiangdi and then went to the village with the summons in hand to collect the tax (this was called piaocui or “to prompt with a summons”). And most xiangdi paid off their duties upon receiving the summons (656.2.1, 1921; 656.3.916, 1934). Those who failed to do so were without exception brought to court and detained there. Jia Jiren, the xiangdi of Zhuangke village, for example, was detained on September 16, 1917, for failing to meet his obligation for the first collection period. The magistrate required him to pay what was due within five days. Three days later, the xiangdi was released when his relatives paid all the taxes in arrears (656.1.757, 1917). The tax-prompting steps taken by the county government as delineated above do not mean that delinquent tax payment was widespread in Huailu. In fact, as Table 9 shows, the majority of the xiangdi in this county delivered the taxes in a timely manner. Those who failed to meet the deadline accounted for only one-fifth of all the xiangdi in the county. And these delinquent xiangdi also paid off their taxes shortly after the deadline under the tremendous pressure of government prompting. This pressure, however, was not the only reason behind the xiangdi’s full payment of taxes. Equally important was the effective functioning of cooperative arrangements for tax payment in the villages.

Village Regulations on Taxation in the Late Qing

A variety of local regulations (known to villagers as cungui) concerning the recruitment of the xiangdi existed in Huailu villages during the late Qing and early Republican periods (see Huaiyin Li 2000, 2001). In tax collection, likewise, various cungui regulated the xiangdi’s advance payment of taxes for his fellow villagers and the latter’s repayment to the xiangdi. Let us first consider village regulations on the supply of chaiyao during the late Qing period. The chaiyao was the only surtax in Huailu during the late Qing period. It was to cover the county government’s expenses in supplying services to the imperial court (locally known as dachai, major service, or huangchai, royal service) as well as its expenses for local projects (xiaochai, minor service). The amount of the chaiyao varied every year, depending upon the magistrate’s

268

li

actual needs and the year’s harvest. The collection of chaiyao was separate from that of the land tax. Unlike the land tax, which was always a fixed quota (listed in taels) for individual taxpaying households and was collected regularly, the chaiyao could be levied at any time and was assessed directly in copper cash (wen). In most Huailu villages, it was the xiangdi’s duty to advance the chaiyao for his community. Because of the irregularity of the chaiyao, disputes between the xiangdi and his fellow villagers were much more likely to occur over this surtax than over the regularized land tax. In fact, all of the late-Qing records of disputes over tax collection available in the Huailu archives pertain to the chaiyao. Villagers thus felt it necessary to create village regulations to govern their activities in advancing and repaying the chaiyao and to reduce related disputes. In general, those village regulations fall into two types. In one type, rich households were obliged to assist the xiangdi in his advance payment of the chaiyao. In Dongpingtong village, for example, all households with more than 80 mu of land were required to help the xiangdi advance the chaiyao. In 1875, four rich villagers refused to do so on the grounds that the xiangdi had not asked four other households with 80 mu of land to share the collective duty; they argued that he had violated the “old custom” ( jiuli). At the court session, the magistrate simply ruled that all the households in that village with a tax quota of 1.5 taels and above had to help advance the chaiyao. He chose to use the tax quota instead of actual landholding as the criterion for sharing the xiangdi service because the magistrate had the information on all households’ tax liabilities readily available from the records of his yamen, whereas the actual landholdings of the villagers involved in the dispute were difficult to verify (655.1.901, 1875). Another village, Lianhuaying, had an “old regulation” ( jiugui) that all the rich households had to assist the xiangdi to meet the “royal service” when imposed. According to a petition from the xiangdi of the village, a rich villager, named Li Er Pangniu, refused to honor his obligation in 1875. Li was immediately summoned to the court, where he promised to help (655.1.978, 1875). A third case comes from Gaoqian village. Its regulation demanded that, in addition to the xiangdi (locally known as hongming xiangdi, “the xiangdi in red name”), six more villagers were to be appointed each year as bangban xiangdi (assistant to the xiangdi) according to their landholdings. The six bangban, mostly the rich in the village, had to share the xiangdi’s duty of advancing land taxes as well as the chaiyao. If any failed to do so, the xiangdi had to make good on the tax shortage and then collect his money from the delinquent bangban. In 1884 a xiangdi named Du Yuanyuan had to do so for one of his bangban, one Liang Youren, and later brought suit against Liang for failing to repay him 44,000 wen (655.1.977, 1884).

between the state and the village

269

Another regulation required the xiangdi to advance the chaiyao on his own and then collect the money owed by individual households according to their landholdings, exactly as he did with the land tax. A case from Xiumen village illustrates the operation of this type of regulation. By the “old regulation” of this village, the xiangdi had to advance all the chaiyao for his fellow villagers and then apportion his duties to them according to their landholdings (andi paichai). In 1882, each mu of land in that village was to share 81 wen of the chaiyao the xiangdi had advanced. A 9 mu plot in this village, however, had recently been sold to Yao Yongqing of the neighboring Shijiazhuang village. Yao repaid the xiangdi of Xiumen village only 900 wen in land tax (100 wen per mu) but refused to reimburse the chaiyao 729 wen. The xiangdi thus sued Yao for “not acting in accordance with the regulation of [his] village” (bu an xiaode cungui xingshi). The magistrate issued a summons to Yao upon receipt of the plaint. Yao’s son, who appeared at the court session on behalf of his father, promised to repay the advanced chaiyao after returning home. Three days later, the xiangdi filed a petition to close the case, claiming that Yao had asked the elderly in the village to mediate the quarrel, had paid off all his chaiyao dues, and had pledged to “abide by the old regulation” (zunshou jiugui) in the coming years (655.1.913, 1882). By the late nineteenth century, these sorts of “village regulations” had been firmly established in many Huailu villages to govern the payment of the chaiyao, a surtax that was much more likely than the land tax to cause disputes between the xiangdi and his fellow villagers because of its irregularity. Any of the two parties in dispute who violated the regulation was thus subject to the accusation of the other in the language of cungui. The county yamen, likewise, always honored local regulations in its adjudication of disputes over the chaiyao.

Village Regulations on Taxation in the Republican Period

Village Regulations The county assembly’s decision in 1912 to freeze the rate of chaiyao and to collect it together with land tax eliminated disputes resulting from the irregular levying of that surtax and ended the working of related village regulations. Disputes over tax collection during the Republican period bring to light the role of the cungui in the collection of the land tax. In most Huailu villages the xiangdi advanced tax monies for all his fellow villagers during the collection season and then collected them from individual taxpaying households on a designated date at the end of the lunar year. The operation of this cooperative arrangement, however, necessitated the support

270

li

of village regulations. Without an effective cungui, both the xiangdi’s advance payment and the villagers’ reimbursement were unimaginable. The purpose of the cungui was to define the mutual obligations between the xiangdi and his fellow villagers; it demanded the former to advance taxes incumbent on the whole village or the pai during the collection season, using the village fund or money he had borrowed, while requiring the taxpayers to repay the xiangdi by the due date. The imposition and subsequent suspension of the shanhou liangjuan (levy for postwar rehabilitation) illustrate how the cungui worked. In compliance with the decision of the Zhili governor, the Huailu magistrate announced on September 2, 1921, the collection of that levy together with the land tax, which brought the tax burden to 2.99 yuan per tael. Many xiangdi thus advanced their taxes at that rate. On November 29, however, the governor, in response to resistance from the elite, decided to postpone the collection of the liangjuan to the next year; all the overcharged taxes, by the governor’s order, were to be applied to the coming year. The Huailu government ceased the collection of the liangjuan accordingly and restored the collection rate to the original 2.50 yuan per tael. This midstream change in the collection rate put all those xiangdi who had paid tax in advance in a difficult situation. First, although the xiangdi had advanced taxes at 2.99 yuan per tael, the taxpayers now insisted on repaying him at the reduced rate of 2.50 yuan per tael. Since in most villages the xiangdi used loans rather than public funds of the community to advance taxes and had to pay interest on the loan, repayment at the reduced rate would mean that he would be unable to pay off his loan. Second, in most Huailu villages the xiangdi service rotated annually among the taxpaying households. The incumbent xiangdi was only responsible for advancing the taxes for the current year and thus needed to collect all his monies at the end of the year. If the overcharged monies were applied to the next year as the provincial government required, much confusion would arise in tax advancement and repayment in the next year. Therefore many xiangdi delivered petitions to the magistrate for immediate reimbursement. Most of those petitions cited the “old regulation” on the xiangdi’s advance payment of taxes in their respective villages and then emphasized the two reasons described above for prompt reimbursement of the overpaid taxes. The magistrate’s initial reaction to those petitions was simply to instruct the xiangdi to collect the advanced monies at the new rate (2.99 yuan per tael). After so many petitions arrived, however, the magistrate changed his opinion and reported the matter to the Financial Depart­ ment of the p ­ rovince on December 14, requesting a reimbursement of the

between the state and the village

271

overcharged taxes as the xiangdi had requested. Three weeks later, the department approved the request and all the xiangdi received their overpaid taxes back from the government (656.1.1232, 1922). It is obvious from this case that village regulations on the advance payment of taxes by the xiangdi remained effective in most Huailu villages and were respected by the government in the early 1920s. Noncompliance However, there was no shortage of instances in Huailu villages in which the taxpayers violated the village regulations. Those violations generally fell into three categories: some villagers simply rejected the old cungui and insisted on the statutory method of tax payment; some xiangdi overcharged the taxpayers; and some taxpayers refused to repay the xiangdi. Let us begin with an example of the first category. Although the xiangdi system worked for the overall good of taxpayers, protecting them from potentially abusive outsiders, on occasion the villagers disregarded the preexisting cungui requiring advance payment by the xiangdi. This was especially true after the state reasserted the self-delivery system. In 1929 the Guomindang government announced the abolition of the sheshu system for the purpose of eradicating the ill-favored tax-farming business, which was widespread in the country and was often the monopoly of the sheshu. At the same time, it reaffirmed the self-delivery system in tax collection, as the Qing state had repeatedly done. This step, though aimed at wiping out the entrepreneurial tax farmers, nevertheless offered a pretext for some villagers to challenge village regulations and to shirk their duties in tax collection and repayment. A dispute thus erupted in Nanxinzhuang village in October 1929, when Xue Deyu, the village head, accused three fellow villagers of “violating the local regulation” (pohuai xianggui). According to his plaint to the magistrate, it had long been a practice in his village for the xiangdi to advance taxes for the whole village at collection time and then distribute the tax receipts to individual households with the help of a teacher to collect the money owed him. However, the petition went on, the three villagers, all “rascals and ruffians” (wulai guntu) and “determined to change the local regulation” ( fei gai xianggui bu yi), insisted on “delivering taxes individually” (ge na ge liang) and prevented the xiangdi from advancing taxes for all the villagers. However, most taxpayers in the village, the village head contended, were illiterate and unwilling to change the regulation. The three villagers’ violation had impeded the xiangdi’s discharge of his public duties. Upon receipt of the plaint, the magistrate instructed the ward police to look into the matter and to settle the dispute in accordance with the local regulation (656.3.57, 1929).

272

li

The threat to village regulations came not only from the taxpayers. The xiangdi could also violate established practices when collecting taxes. I found two cases in which the xiangdi overcharged taxpaying households. On December 27, 1926, three Liangs from Liangjiazhuang village submitted a plaint against Liang Jifu, the incumbent xiangdi of that village. They asserted that the government had announced a collection rate for the current year of 2.9 yuan or 11,252 wen (copper cash) per tael of tax quota. However, the xiangdi had assessed them extra amounts varying from 30 or 40 to 200 or 300 wen per tael. The magistrate asked the ward police to investigate. The police reported that, according to the village head, the xiangdi had made an accounting error. As a friend of both parties, claimed the village head, he had mediated the dispute and done the accounting again, and all the overcharged money had been returned to the accusers (656.2.852, 1926). A similar event occurred in Zhengjiazhuang village. In 1920, Wu Weiyi, the xiangdi of that village, received a court instruction at the beginning of the first tax period to set the collection rate for the year at 2.3 yuan per tael of quota plus 0.20 yuan per tael as the chaiyao (totaling 2.50 yuan per tael, with the police fee and water-work contribution on top of that). On December 18, four villagers, all surnamed Shi, filed a petition complaining that the xiangdi had failed to post the new rates at a public site. After advancing all the taxes, they said, the xiangdi collected his monies at 2.53 yuan per tael, which was 0.03 yuan higher than the announced rate, and declined their request for a reimbursement. On the magistrate’s instruction, the village head mediated the dispute and had the overcharged monies repaid to the four accusers (656.1.1243, 1920). It was possible, therefore, for the xiangdi to overcharge taxpayers when collecting his advanced monies. But this kind of abuse was not the rule. Under normal conditions, the xiangdi was unable to profit from tax collection, for his activities were under the close surveillance of his fellow villagers; not only did they know each other’s tax quotas, but the official notice of the collection rate was normally posted in a public place as required by the government. Everybody thus knew exactly how much they needed to repay the xiangdi. Those who paid the xiangdi’s overcharge would not hesitate to request a refund, and the xiangdi’s refusal to provide it would lead to a lawsuit. In these two cases the overcharges were quite minor, 0.26 to 2.6 percent of the taxes due in the first instance and 1.2 percent in the second. When the xiangdi demanded a higher surcharge, the taxpayers simply refused to repay the xiangdi. That was exactly what happened with the suspension of the shanhou liangjuan in 1921. In that instance, the taxpayers unanimously refused to repay the xiangdi at a rate almost 20 percent higher than the recently reduced

between the state and the village

273

one. Thus the xiangdi’s opportunities to profit from overcharging the taxpayers were limited. The incidence of taxpayers’ nonpayment was even lower. I found just one such case. In 1915, the xiangdi of Bailinzhuang village, who had advanced all the taxes for his fellow villagers, accused one Wang Xinshun of refusing to pay back the tax money the xiangdi had advanced. The magistrate summoned the accused immediately. Wang pledged at the court to pay his debt to the xiangdi and was released on bail (656.1.424, 1915). By and large, tax payment proceeded quite smoothly where the villagers cooperated under local regulations. Disputes resulting from the xiangdi’s extra collection or taxpayers’ nonpayment were rare and, when they did happen, were quickly mediated and adjudicated on the basis of the regulations. The Tax-Prompting Xiangdi Aside from the vast majority of Huailu villages where the xiangdi advanced taxes for households in his charge, in a small number of villages the xiangdi only prompted his villagers to pay taxes. In these villages every household paid its own taxes, according to the official requirement of “self-delivery” (zifeng tougui). Needless to say, the burden of the tax-prompting xiangdi was much lighter than that of the tax-advancing xiangdi, for he did not have to borrow money for the advance tax payment or collect his repayment from individual households. However, his obligation to the government was the same as that of the tax-advancing xiangdi. Like the latter, he was to see that all taxpaying households in his charge paid taxes in full and within a time limit. If any villager failed to pay taxes by the deadline, the xiangdi had to make good the shortage. Failure to do so would result in a court hearing and possible detention. The tax-prompting xiangdi’s relationship with the government thus was no different from that of the tax-advancing xiangdi. Both were held responsible for full payment of the taxes imposed on fellow villagers. Under either arrangement, the government dealt only with the single xiangdi rather than with individual taxpayers. Where the two kinds of xiangdi differed was in their relationship with taxpayers. Unlike the tax-advancing xiangdi, who seldom argued with his fellow villagers where village regulations worked smoothly, the tax-prompting xiangdi faced great tensions in his relationship with local taxpaying households. To avoid the burden of making good any tax shortage after the deadline, the xiangdi resorted to every means available to prompt his fellow villagers to pay their taxes before the deadline. After that deadline, the xiangdi had to advance any taxes still due, so that he could avoid a court hearing and, if

274

li

summoned to court, get released from detention. In the absence of communal cooperation in tax payment, no regulation was formulated to define the mutual obligations between the xiangdi and his fellow villagers, as we have seen under the tax-advancing xiangdi. Therefore, the tax-owing villagers also felt no pressure to repay the xiangdi on time. Instead, they would delay repayment as much as possible, and, in some instances, simply denied it completely. It is no wonder therefore that the incidence of disputes involving this type of xiangdi was exceptionally high, accounting for 72 percent of all the disputes over tax collection in Huailu county (thirty-four of forty-seven cases, excluding disputes over enclave taxes). These disputes resulted from either the xiangdi’s prompting or the taxpayers’ nonrepayment. Let us first consider the former. The tension between the tax-prompting xiangdi and his fellow villagers started at the very beginning of the tax collection season. To avoid advancing taxes for delinquent taxpayers after the closing date, the xiangdi would strongly encourage his villagers to pay their taxes. However, not every villager was able to pay on time, and, similarly, not every xiangdi was able to advance taxes for the delinquent households. Unlike the tax-advancing xiangdi, here the xiangdi lacked institutional support and a stable source of money to make the payment, such as public land or a village fund. For poor xiangdi who lacked sufficient credit, it was also difficult to ask for a loan to cover tax shortages. Thus there were always a number of xiangdi who failed to fulfill their duties by the deadline and were summoned and detained. Zhang Yongtai, a xiangdi of Lingdi village, was thus detained in December 1915, owing to his failure to pay off the tax balance for a villager named Zhang Taoqi. He was released only after his relatives and friends helped him clear the deficit (656.1.419, 1915). Given the trouble that tax shortages caused, the xiangdi urged taxpayers in his charge to pay taxes by any means. And the tension between the two parties escalated as the deadline approached. An effective way to press dilatory taxpayers to pay on time was to file a plaint against them before the closing date. Thus, on January 8, 1919, Liu Shunxing, the xiangdi of Yongbei village, filed a petition claiming that Wang Wenzai and another fifteen taxpayers in his village refused to pay taxes and ignored his repeated promptings in an insulting way. He therefore found it difficult to perform his public tasks. The magistrate responded with a summons to all sixteen villagers, who promptly paid their taxes upon receiving the summons (656.1.1084, 1919). In another instance, Shi Yutai, the xiangdi of Zaicheng, accused twelve fellow villagers of nonpayment. The taxpayers immediately paid what they owed after receiving the summons from the magistrate (656.1.965, 1918). Although filing a petition proved here to be an effective way to prompt tax payment, it inevitably undermined the xiangdi’s relationship with his fellow villagers.

between the state and the village

275

More often than not, the disputes between tax-prompting xiangdi and taxpayers resulted from the latter’s nonrepayment of money owed. In most cases, a delinquent villager quickly paid his debt to the xiangdi after the xiangdi submitted a plaint. Those who refused to reimburse the xiangdi, as the latter often complained, were the “bully-like” villagers. Such persons, according to the petition from a xiangdi of Dongtongping village, were “always reckless and did unlawful things. And no one dare to offend them.” Every year, claimed the xiangdi, they delayed tax payment, forcing the xiangdi to advance their taxes. And as a rule, they refused to reimburse the xiangdi (656.1.599, 1915). In Nantongye village, a villager named Wang Yulin not only refused to repay the xiangdi himself, but encouraged others to delay their repayment. When asked to repay his debt to the xiangdi, Wang claimed that he would repay nothing until he received a court ruling requiring him to do so (656.1.392, 1915). In such cases, the xiangdi had no option but to file a complaint to recover his money. Together, these two kinds of cases illustrate the tension between the taxprompting xiangdi and local community members in the absence of cooperative arrangements. Unlike the tax-advancing xiangdi who paid all of the households’ taxes in advance and therefore had no tax-prompting conflicts with them, the tax-prompting xiangdi found himself in intense conflicts with local residents; he had either to use all possible means to press the taxpayers to meet their obligation before a deadline or make good on the tax shortages out-of-pocket. But because of the lack of cooperative arrangements, the taxpayers felt no obligation to repay the xiangdi in a timely manner; some even denied their debts to the xiangdi. The incidence of disputes involving the taxprompting xiangdi thus was much higher than that involving the tax-advancing xiangdi. And naturally, villages with a tax-prompting xiangdi were much less solidary than those with a tax-advancing xiangdi. Nevertheless, we should not exaggerate the extent of such disputes. In fact most disputes over tax payment and repayment were settled through community mediation (xiangzhong paijie) before the xiangdi had to file a plaint; disputes that led to a lawsuit accounted for just a small proportion of all those disputes. And such lawsuits quickly came to an end after the magistrate’s initial reaction in the form of a notice to prompt tax payment or repayment or the issuing of a summons, in which the magistrate also instructed the village head to help handle the case. Once the village leader received the magistrate’s order, he had to look into the causes of the dispute and persuade the accused to pay his debt. Then he would report to the magistrate the result of his handling and ask for the case to be closed (e.g., 656.1.392, 1915). Community mediation thus played an important part in the resolutions of disputes involving the taxprompting xiangdi.

276

li

The Collection of Enclave Taxes

In Huailu, most land was owned by local residents and taxed by the county government. Yet there existed numerous nonresident enclaves ( jizhuangdi) of varying sizes owned by people of neighboring counties but still taxed by the Huailu government. A 1913 statistic indicates that the nonresident enclaves within the county totaled 13,435 mu, with a sum of 484 taels of tax quota, which accounted for only 2 percent of the total taxable land in the county and almost 2 percent of the county’s total tax quota (656.1.103, 1913). There were also numerous enclaves outside Huailu owned by Huailu residents and taxed by neighboring counties. In spite of the small amount of enclave land in Huailu county, conflicts between tax-collecting outsiders and local taxpayers were much more frequent than disputes between local xiangdi and their fellow villagers. The Huailu archives contain forty-one records of disputes between local enclave owners and outside collectors and twenty-two between Huailu collectors and enclave owners in neighboring counties. Together, the two kinds of disputes, totaling sixty-three, accounted for over 57 percent of the 110 formally filed disputes over tax payment between 1912 and 1936. And again most such disputes involved enclave owners in villages under the tax-prompting xiangdi system, where no xiangdi advanced the enclave taxes to the collector. The person responsible for collecting enclave taxes was usually the sheshu (a semi-official clerk at the supra-village level, who recorded changes in tax liabilities and helped the government compile tax rolls), for he was the only person who knew the real names and exact locations of taxpaying households on the tax roll. Occasionally the county government relied on its bailiff ( fajing) to collect taxes from the delinquent enclave owners as it often did with taxowing taxpayers under its direct jurisdiction. At the beginning of the collection period, the county government usually issued to the sheshu or the bailiff a joint-notice (huipiao), which was also stamped by the neighboring county government, authorizing him to go to the neighboring county and prompt the enclave owners there to pay their taxes. Aware that the taxpaying households often “refused to pay their taxes, counting on its outside location” ( yiyi geshu, kang bu wanna), the magistrate always allowed the prompters to report delinquent taxpayers to him for summons and even detention (655.2.12, 1921–27). However, the sheshu and the bailiff had tremendous difficulty prompting tax payment. When the deadline was due, they themselves often had to make good all the tax shortages for the delinquent enclave owners. Then they faced similar difficulty recovering the money they had advanced. One obvious r­ eason

between the state and the village

277

for the taxpayers’ nonrepayment, as mentioned in almost every petition from the sheshu, was that the taxpayers took advantage of him as an outsider; in their home places (zhangdi qiyi) they were under the jurisdiction of a different government (zhangshi geshu) (e.g., 656.3.1102, 1919; 656.2.439, 1924). As an outsider, the sheshu could not collect their monies by force, for the debtor was backed by his relatives and friends and even the whole community. The sheshu thus frequently complained that “being in the strange land, [they] could not demand repayment by force [shenju yidi, weibian yingtao]” (e.g., 656.3.455, 1931). Nor could the sheshu resort to community mediation to settle his dispute with a local debtor, for he was not a member of the community. To hasten the enclave owners’ tax payment and to avoid advancing taxes for them, many sheshu opted to submit a plaint against the taxpayers before the deadline, just as a tax-prompting xiangdi might do in dealing with his fellow villagers. This was especially true of sheshu who were in charge of villages with numerous enclave owners. When disputes with the delinquent taxpayers occurred, the only way for the sheshu or the bailiff to get his money back was to file a complaint. The incidence of filing a formal plaint against the enclave owners thus was exceptionally high. In the absence of community mediation, the magistrate usually issued a straight summons, instead of a repayment-prompting notice, to the delinquent taxpayer, at the request of his colleague in the neighboring county to whom the sheshu had filed a petition. In most cases, upon receiving the summons the delinquent taxpayer quickly traveled to the neighboring county to pay what he owed. The xiangdi’s role varied in such cases. In some villages, he was not held responsible for fellow villagers’ nonpayment. He merely petitioned to end the case. In other cases, the xiangdi had to share with the fellow villagers the duty of repayment. For instance, when the summoned taxpaying households failed to go to court, the xiangdi had to appear at the court for them. Zhang Ruiheng, the xiangdi of Tazhong village, was thus brought to court on April 4, 1916, when the policeman failed to find a delinquent villager, who owed 0.07 tael of tax money to a sheshu of Luancheng county. The xiangdi was released after pledging to find the missing villager and report back to the court within ten days (656.1.597, 1916). Sometimes the xiangdi was expected to assist the sheshu or a bailiff to collect his advanced monies. Not surprisingly, some xiangdi were reluctant to help the outsider and were charged with noncooperation (e.g., 656.1.968, 1918; 656.2.101, 1921). Han Luozhang, the xiangdi of Nanqiema village, even suffered detention for refusing to cooperate with Liu Heng, a sheshu from Luancheng county, who was to collect the taxes of all seventy-seven households in that village (totaling 20.973 taels).

278

li

Disputes over the payment of enclave taxes occurred either between an outside collector (the sheshu or the bailiff) and enclave taxpaying households who refused to repay the collector, or between the outside collector and the xiangdi who was in charge of the delinquent enclave taxpayer yet refused to assist the outsider. What drove the incidence of such formally filed disputes so high was not merely the fact that the two disputing parties were under different jurisdictions, as the sheshu often claimed in their petitions, but the lack of community mediation that resolved most disputes between the local xiangdi and his fellow villagers. For the outsider, unable to count on local mediation to settle the dispute, the only way to get his money back was to file a plaint against the delinquent taxpayer. The incidence of formally filed disputes between the sheshu and enclave taxpayers thus was even higher than that between the taxprompting xiangdi and his fellow villagers. In the latter case, disputes that were brought to court were greatly reduced by community mediation. Conclusion The foregoing examination of tax collection in late Qing and Republican Huailu reveals two basic facts. One is that three tax systems existed simultaneously in local villages, namely, the voluntary cooperation under tax-advancing xiangdi, the official system of self-delivery under the supervision of tax-prompting xiangdi, and the sheshu’s tax-farming business in enclaves that owed taxes to a neighboring county. The other is that the incidence of disputes over tax payment differed under different systems. Disputes were rare in cooperative communities where the xiangdi paid all taxes in advance on behalf of his fellow villagers and the latter repaid the xiangdi on time in accordance with local regulations. Disputes were more likely where the villagers paid taxes individually at the prompting of the xiangdi. The incidence was the highest where the taxpayers refused to cooperate with tax farmers from outside. The different incidence of disputes on land taxes entailed different degrees of government involvement in local administration. Under normal conditions, the magistrate left tax collection to informal agents and did not step in until the taxpayers failed to deliver taxes by a deadline or until the tax collectors disputed with the taxpayers. His role thus was minimal in villages under the cooperative xiangdi system where local regulations were effective in resolving disputes. The magistrate’s involvement increased accordingly as disputes mounted under the tax-prompting and tax-farming systems. The contrasting effects of different systems on the relationship between the government and the village explain why neo-Confucian authors since the

between the state and the village

279

Song dynasty (960–1279) had vehemently advocated voluntary cooperation as the best way of governance. They believed that such cooperative institutions, when working properly, would serve the interests of both the community and the government, for they reduced the government’s administrative costs while protecting local people from abusive outsiders. When disputes arose among local residents, the self-governing community could easily work out a compromise through mediation, which further reduced the scope of official involvement. The best form of government, they argued, was one that minimized its interference with local affairs and let the people govern themselves on the basis of voluntary cooperation (Wu Ding’an 2000; Cao Guoqing 1997; Wang Gaoxin 1995). Counter to this ideal of self-governance, of course, was the situation in which the government devolved all its administrative tasks to profiteering agents, such as tax farmers or yamen runners, whose only concern was to enrich themselves at the expense of the people and even the government, and whose predatory activities inevitably incurred conflicts with local people, which could never count on community mediation for a resolution. Compared to the desired method of cooperative self-governance and the counter ideal of entrepreneurial practices, the official system of self-delivery was depicted as a compromise aimed at ensuring the government’s tax revenues while precluding the profiteering activities of any third party between the government and taxpayers. However, the weakness of this system—that is, its reliance on the government’s effective control of rural society—often caused local magistrates to turn to local initiative, either self-governing or abusive practices, when the government failed to penetrate the countryside through the baojia network. The coexistence of the self-governing cooperation under the tax-advancing xiangdi with the formally imposed self-delivery system and the officially denounced tax-farming practice in Huailu villages, then, can be seen as a microcosm of different methods of governance in rural China. Although these methods had different images in official representations and Confucian discourse on local control, a pragmatic magistrate had no difficulty allowing their concurrent existence in village communities. From his point of view, these different systems were actually supplementary, rather than contradictory, for each occurred in its own context, and together they worked to secure the government’s tax revenues. The magistrate’s usual approach to local government thus was to tolerate and even encourage local initiatives so long as these informal practices met the expected goals of tax collection and social control. Although the informal practices emerged seemingly as a result of the government’s inability to impose its own designs on the villages, from the magistrate’s point of view doing so was both financially unnecessary (and even

280

li

c­ ounterproductive) and ideologically against the neo-Confucian doctrine of a minimal government and a self-governing society. It is in this context that we can understand the intriguing relationship between state law and village regulations on taxation. From the viewpoint of village communities, any action that deviated from their established rules was deemed unacceptable. “Ge na ge liang,” or individual delivery of taxes, though exactly what the state stipulated, was nevertheless seen as a violation of local regulations. From the state’s point of view, the village regulations were perfectly compatible with the state law when they facilitated tax collection, no matter how the local cungui deviated from the statutory method. This was why the magistrate endorsed the cungui, rather than the statutory practice, in his handling of disputes resulting from the villagers’ insistence on self-delivery. The substitution of village regulations for state law was, then, indicative of the substantive approach of local government, an approach that remained unchanged in land taxation as late as the early twentieth century. Thus village-state relations in China before the twentieth century cannot be viewed simply in terms of a dichotomous opposition between state and society. Except where gentry elites and/or lineage organizations were so strong as to resist government influences or so weak as to allow the intrusion of extortionate agents, in most parts of rural China, as in the villages in Huailu, informal agents and institutions of local communities routinely carried out administrative functions delegated by the government in the absence of salaried bureaucrats below the county level, and the government did not intervene until disputes arose within the community that threatened the normal functioning of the quasi-official institutions. Everyday governance in most parts of rural China can be characterized as a form of “substantive governance,” to accentuate the blurring of the distinction between the government and village communities in rural administration, and to distinguish that process from the formalistic administrative system based on systematic codes and a formal hierarchy of personnel, as we find in the Chinese bureaucracy at and above the county level. To further understand the nature of substantive governance, we need to ponder at this point why the Chinese state remained content with its limited reach into local society, and why it did not work harder to prevent the growth of localism that was believed fatal to imperial regimes in history (M. Mann 1988: 1–31). I propose that Confucian teachings on benevolent government and concerns about the misdeeds of local officials were not the only reason rulers minimized the government’s direct involvement in local administration. Equally important in shaping the pattern of local governance were the financial needs

between the state and the village

281

of the Chinese state, which had to do with its geopolitical setting. Unlike the extremely fragmented Europe in the late-medieval and early-modern periods, when the European monarchs had to expand and maintain a military force as large as possible in order to survive international warfare (Finer 1975; Ardant 1975), China had long since been under a unified empire that had established its cultural, diplomatic, and even military supremacy over the surrounding tributary states. Although the Ming and early Qing dynasties did face frequent threats from the Mongol tribes, in most times the Chinese empire enjoyed peace, prosperity, and superiority over the neighboring peoples. In the meantime, the enormous taxpaying population and the vast taxable territory constituted a tax base so huge that it was unimaginable in Europe. Thus, setting the tax rate at a very low level by European standards would generate enough revenue to support the state apparatus under normal conditions. This vast tax base in turn allowed the Chinese rulers to pursue the “light tax” (qingyao bofu) policy that epitomized “benevolent government” in Confucian political tradition. Obviously, it was China’s status as a unified empire and its huge tax base, more than anything else, that explained its low rates of taxes on land, which varied from 2 to 4 percent of land yield in the Qing empire as a whole. Moreover, unlike the European monarchs in the early-modern period, whose power was greatly offset by rivaling regional lords and autonomous self-governing organs that restricted their rights in taxation (Tilly 1975; Braun 1975), the Chinese political structure was highly centralized, and the ruler theoretically had absolute control over the bureaucracy down to the county level. Below the county yamen, the ruler had support from gentry elites who identified themselves with the monarch under the influence of Confucian doctrines, as seen in their stipulation and implementation of community covenants that placed tax payment as one of their priorities. Although the involvement of evil gentry in tax evasion and hidden land was a chronic headache for local administrators, the Chinese ruler had never felt a threat to or restriction of his power from any social groups comparable to that faced by his counterparts in Europe. Therefore, unlike a European monarch who had to combat various forms of localism in the course of taxation and centralization, the Chinese rulers, endowed with tax revenue large enough to support the government, felt no need to do so. In fact, so big was the tax base in Qing China at its height that it generated more income than it actually needed. Many times the tax surplus allowed the Qing rulers to announce a universal suspension of land taxes throughout the empire (Hu Chunfan 1984; He Ping 1998: 14–26). Tax riots, so rampant and perennial in early-modern European countries, did not become a big problem for the Chinese rulers until the early twentieth century, when

282

li

tax surcharges proliferated. The Qing state’s decision in 1712 to permanently freeze the head tax quota, a policy that was faithfully observed for the rest of its history, is understandable only in this context.6 We may then understand why the Chinese rulers were so reluctant to widen the tax base using such means as uncovering black land and taxing land deeds, for these measures were harmful to local communities and ran counter to the tenets of benevolent government; also, their benefits were insignificant when the readily available tax income was already sufficient. So what are the implications of substantive governance in local society for understanding the Chinese state? Preexisting theories of the Chinese state do not work perfectly here, be they oriental despotism, the Asiatic mode of production, patrimonial bureaucracy, or bureaucratic monarchy, for they deal only with the formal system of the state and overlook the actual practices at the village level. To appreciate the nature and functioning of local practices and their importance in the structure of the Chinese state, Michael Mann’s concept of “infrastructural power” is useful in this context. By “infrastructural power,” Mann means the ability of the state to penetrate civil society by collecting taxes and information, carrying out the government’s directives, and coordinating economic activities. Infrastructural power, according to Mann, contrasts with the “despotic power” exercised by ruling elites over various social groups, which was autonomous and even unlimited because the rulers did not need to consult the ruled. Mann further observes that while despotic power was strong in the histories of many societies, infrastructural power was relatively weak; the reverse is true of modern Western society, where strong infrastructural power has coexisted with despotic power (Mann 1984, 1986). As for imperial China, we may grossly classify its regular bureaucratic system as one of despotic power, and the informal agents at the village and supravillage level who carried out the duties in tax collection and local control as wielding infrastructural power. From Mann’s view, those who held the infra6 Although the increasing population and hence the expanding administrative expenditures prompted local officials to increase taxes using various excuses, Yongzheng emperor nevertheless attempted to regularize the collection of the irregular charges in order to limit the tax burden and set the newly regularized charges at a very low level for fear of violating the maxim of low taxation (Zelin 1984). In the last decades of the Qing, when military expenses and foreign indemnities skyrocketed, the ruler could only turn to other channels such as creating commercial taxes (e.g., lijin) and maritime custom duties, rather than increasing the land tax. As a result, the land tax remained reasonably low until the beginning of the twentieth century.

between the state and the village

283

structural power were of course part of the formal bureaucracy. He does not know that infrastructural power in imperial China was far from formalized and institutionalized, and was never part of the formal bureaucracy. A more complete picture of the Chinese state would acknowledge that both formal despotic power and informal infrastructural power constituted the Chinese administrative system. The Chinese state should be described as a combination of despotism, found in the highly centralized and standardized bureaucratic system at and above the county level, and a laissez-faire tradition in the decentralized, informal institutions in local governance. The state’s reliance on informal institutions in village governance, a pattern of local administration moralized in neo-Confucian discourse and evidenced in varying forms in Huailu county, was not as much a result of the incomplete centralization of the Chinese state or local resistance against state intrusion as the traditional wisdom on the Chinese state has assumed; rather, it was embedded in the geopolitical settings of the Chinese empire and the neoConfucian teachings on minimal and benevolent government. In the final analysis, what characterizes the village-state relationship in late imperial China, as seen in the core area of the North China macroregion, was more interdependence than paired opposition and confrontation. References Andrew, Anita M., and John A. Rapp, [eds.]. 2000. Autocracy and China’s Rebel Founding Emperors: Comparing Chairman Mao and Ming Taizu. Lanham, Md.: Rowman and Littlefield. Ardant, Gabriel. 1975. “Financial Policy and Economic Infrastructure of Modern States and Nations.” In Charles Tilly, (ed.) The Formation of National States in Western Europe, pp. 380–455. Princeton, N.J.: Princeton University Press. Balazs, Etienne. 1964. Chinese Civilization and Bureaucracy: Variations on a Theme. New Haven, Conn.: Yale University Press. Cao Guoqing 曹國慶. 1997. “Mingdai xiangyue tuixing de tedian” 明代鄉約推行的 特點 (Characteristics of the implementation of the xiangyue system in the Ming dynasty), Zhongguo wenhua yanjiu 中國文化研究 Spring. Cheng Pengfei 程鵬飛. 2000. “Wang Yangming zhixing heyi yu nangong xiangyue” 王陽明知行合一與南贛鄉約 (Wang Yangming’s theory of unity of knowledge and practice and village covenants in southern Jiangxi), Guizhou wenshi congkan 貴州文史叢刊, 3. Ch’ü, T’ung-tsu. 1962. Local Government in China under the Ch’ing. Cambridge, Mass.: Harvard University Press.

284

li

Duan Zichang 段自成. 1996. “Qingdai qianqi de xiangyue” 清代前期的鄉約 (The xiangyue in early Qing), Nandu xuetan 南都學壇 16 (5). Finer, Samuel E. 1975. “State- and Nation-Building in Europe: The Role of the Military.” In Charles Tilly, (ed.) The Formation of National States in Western Europe, pp. 380– 455. Princeton, N.J.: Princeton University Press. Fu, Zhengyuan. 1993. Autocratic Tradition and Chinese Politics. Cambridge, England: Cambridge University Press. Gu Zhun 顧準. 1982. Xila chengbang zhidu 希臘城邦制度 (The Greek city-states). Beijing: Zhongguo shehui kexue chubanshe. ———. 1999. Minzhu yu “zhongji mudi” 民主與“終極目的” (Democracy and the “ultimate end”). Beijing: Zhongguo qingnian chubanshe. Hauf, Kandice. 1996. “The Community Covenant in Sixteenth Century Ji’an Prefecture, Jiangxi,” Late Imperial China 17 (2): 1–50. He Ping 何平. 1998. Qingdai fushui zhengce yanjiu 清代賦稅政策研究 (A study of Qing tax policies). Beijing: Zhongguo shehui kexue chubanshe. Hou Wailu 侯外廬. 1979. Zhongguo fengjian shehui shilun 中國封建社會史論 (Papers on the history of feudal China). Beijing: Renmin chubanshe. Hsiao, Kung-chuan. 1960. Rural China: Imperial Control in the Nineteenth Century. Seattle: University of Washington Press. Hu Chunfan 胡春帆. 1984. “Shilun Qing qianqi de juanmian zhengce” 試論清前期的 蠲免政策 (The tax exemption policy in early Qing), Qingshi yanjiu ji 清史研究集, vol. 3. Hu Rulei 胡如雷. 1979. Zhongguo fengjian shehui xingtai yanjiu 中國封建社會形態 研究 (A study of the forms of feudal society in China). Beijing: Sanlian. Hu Qingjun 胡慶鈞. 2001. “Cong Lantian xiangyue dao Chenggong xiangyue” 從藍田 鄉約到呈貢鄉約 (From the Lantian village covenant to the Chenggong village covenant), Yunnan shehui kexue 云南社會科學, 3. Huang, Han Liang. 1918. The Land Tax in China. New York: Columbia University. Huang, Philip C. C. 1993. “Between Informal Mediation and Formal Adjudication: The Third Realm of Qing Civil Justice,” Modern China 19.3: 251–298. ———. 1996. Civil Justice in China: Representation and Practice in the Qing. Stanford, Calif.: Stanford University Press. Kuhn, Philip A. 1975. “Local Self-Government under the Republic: Problems of Control, Autonomy, and Mobilization.” In Frederic Wakeman Jr. and Carolyn Grant, eds., Conflict and Control in Late Imperial China, pp. 257–98. Berkeley: University of California Press. ———. 1990. Soulstealers: The Chinese Sorcery Scare of 1768. Cambridge, MA: Harvard University Press. Li, Huaiyin. 2000. “Village Regulations at Work: Local Taxation in Huailu County, 1900–1936,” Modern China 26 (1): 79–109.

between the state and the village

285

———. 2003. “Zhongguo xiangcun zhili zhi chuantong xingshi: Hebei sheng Huailu xian zhi shili” 中國鄉村治理之傳統型式：河北省獲鹿縣之實例 (Patterns of village governance in traditional China: the case of Huailu county, Hebei province). In Philip Huang, (ed.) Zhongguo xiangcun yanjiu 中國鄉村研究 (Rural China), no. 1. ———. 2005. Village Governance in North China, 1875–1936. Stanford: Stanford University Press. Mann, Michael. 1986. The Sources of Social Power: A History of Power from the Beginning to A.D. 1760. Cambridge, England: Cambridge University Press. ———. 1988. States, War, and Capitalism. New York: Basil Blackwell. Reed, Bradly. 2000. Talons and Teeth: County Clerks and Runners in the Qing Dynasty. Stanford, Calif.: Stanford University Press. Rowe, William. 2001. Saving the World: Chen Hongmou and Elite Consciousness in Eighteenth-Century China. Stanford, ca: Stanford University Press. Tilly, Charles. 1975. “Reflections on the History of European State-Making.” In Charles Tilly, (ed.) The Formation of National States in Western Europe, pp. 380–455. Princeton, N.J.: Princeton University Press. Übelhör, Monika. 1989. “The Community Compact (Hsiang-yüeh) of the Sung and Its Educational Significance.” In Wm. Theodore de Bary and J. Chaffee, (eds.). NeoConfucian Education. Berkeley: University of California Press. Wang Gaoxin 汪高鑫. 1995. “Zhu Xi zhengzhi sixiang chutan” 朱熹政治思想初探 (Preliminary research on Zhu Xi’s political thought), Anhui shixue 安徽史學, 2. Watt, John R. 1972. The District Magistrate in Late Imperial China. New York: Columbia University Press. Weber, Max. 1951 [1922]. The Religion of China: Confucianism and Taoism. Glencoe, IL.: Free Press. Wittfogel, Karl A. 1957. Oriental Despotism; a Comparative Study of Total Power. New Haven, ct: Yale University Press. Wu Ding’an 吳定安. 2000. “Zhuzi shecang zhi fa jiqi yingxiang” 朱子社倉之法及其 影響 (Zhu Xi’s promotion of community granary and its consequences), Jiangxi shehui kexue 江西社會科學, 12. Wu Ze 吳澤. 1993. Dongfang shehui jingji xingtai shilun 東方社會經濟形態史論 (Papers on the history of Eastern socioeconomic history). Shanghai: Shanghai renmin chubanshe. Zelin, Madeleine. 1984. The Magistrate’s Tael: Fiscal Reform in Eighteenth-Century Ch’ing China. Berkeley: University of California Press.

chapter 9

Village-State Cooperation: Modern Community Schools and Their Funding, Haicheng County, Fengtian, 1905–1931* Elizabeth VanderVen Many studies of state making in the developing world assume a binary dynamic, opposing an intrusive and modernizing state to defensive village communities.1 Certainly when it comes to the more extractive aspects of earlytwentieth-century Chinese state making, such as taxation and military conscription, this assumption has been well borne out.2 It thus might be expected that in 1904, after the Qing Ministry of Education (xuebu) ordered local communities, including villages, to set up new primary schools with their own funds, resistance and hostility resulted. Such a view is held by Sally Borthwick and John Cleverley, who both maintain that the new schools of the early twentieth century alienated the rural populace. Borthwick, especially, emphasizes that the new schools were extractive, asserting that their financing placed a heavy burden on community resources and was part of “an anarchic race to reach available sources of income before others carried them off” (Borthwick, 1983: 98; Cleverley, 1985: 34).

* Originally published in Modern China 31, 2 (April 2005); 204–35. I thank Kathryn Bernhardt, Richard Gunde, Philip Huang, Edward McCord, and Alexander Woodside for their thoughtful and constructive comments in helping me to prepare this article for publication. Thanks also to Joel Andreas, Margaret Kuo, Yasuhiko Karasawa, Jennifer Neighbors, Chaohua Wang, and Jiayan Zhang, who participated in discussions of the earliest versions of this article. Finally, I am most grateful to Alice Falk for editorial help. 1 Perhaps the best-known proponent of this binary model is James Scott, whose moral economy approach is derived from his studies of Southeast Asia (Scott, 1976). 2 Both Philip Kuhn and Philip Huang discovered that late Qing and early Republican statebuilding increased the tax burden on village society and set the stage for hostile relations between state and society. Kuhn characterizes the taxes that were intrinsic to the late Qing modernization project as “a major burden upon the populace” (Kuhn, 1978: 116–17). Huang, in his study of rural North China, also describes how the expansion of local government put a heavy financial burden on society (Huang, 1985: 275–91).

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_011

village-state cooperation

287

On the other side are those who suggest that state making (and the call for a modernized school system) ultimately brought about the breakdown of communities—or, at the least, the dissolution of practices that had long reinforced community ties, in particular religious life. For example, Prasenjit Duara describes how community life in the Shandong and Hebei countryside deteriorated as villages supported a flurry of state-mandated projects including schools, police, and administration (Duara, 1988: 149–51). On this view, schooling and educational reform activities should have disabled the strong ties and identifications of rural peoples with their own communities, replacing those loyalties with allegiances to a larger community based on the nation-state.3 This article draws on archival materials on educational reform in Haicheng county, Fengtian province, in the late Qing and early Republic to examine how rural communities funded and established the new community primary schools and how villagers and the state coped with problems encountered in launching them.4 I argue, against the two positions sketched above, that educational reform was not an either-or process. Communities could be simultaneously active and defensive in their response to state-guided educational reform. On the one hand, many rural communities in Haicheng county embraced state orders to establish new schools, showing an eagerness to participate in movements that had implications far beyond the borders of the village.5 On the other hand, the willingness of these communities to 3 Eugen Weber’s discussion of schooling’s conversion of rustic rural children into polished French citizens rests on the assumption that education “eased individuals out of the [local group’s] grip and shattered the hold of unchallenged cultural and political creeds” (Weber, 1976: 338). 4 In this article I use “community school” as a general term to identify the new primary schools that carried the Chinese appellation gongli (community funded). Their local funding mechanism distinguished these community schools from those that were officially funded (guanli) or privately funded (sili). Although community schools were established on several administrative levels, including the township (xiang) and the municipality (zhen), most were set up at the level of the village. I therefore generally call them “village schools” and refer to them by the name of the village. 5 In his 1993 article “Educational Reform and the Paradigm of State-Society Conflict in Republican China,” Thomas Curran suggests that in some areas, village communities welcomed the opportunity afforded by the state to modernize their schools and that village leaders as well as county and subcounty officials initiated educational reform and school building. Curran thus challenges the extractive state paradigm presented by scholars such as Duara and Kuhn, at least in this sphere (Curran, 1993: 45). Although his article is an overview that addresses the subject only in general terms, he persuasively makes clear that in-depth studies of educational reform in rural areas are needed.

288

VanderVen

participate in the state’s modernization enterprise not only did not lead to community ties breaking down but in many ways reinforced them. Up to now, scholars have assumed that rural communities did not play a significant role in China’s early-twentieth-century modernization process; when they try to better understand modernization during this period, they generally look first toward urban spaces and phenomena. For example, Joseph Esherick writes of the Chinese city, “If a nation was to modernize, the cities had to take the lead,” and “urban reforms . . . were arguably the most successful Chinese efforts at modernization in the early twentieth century” (Esherick, 1999: 1). Urban areas unquestionably played a vanguard role in modernizing China. But this scholarly emphasis has led to the village being largely ignored—an oversight that, even now, can be attributed to a long-standing presumption in Western scholarship that modernization comes from the breakdown of communities and is inevitably accompanied by a move toward urbanization. Such a view of modernization seems inadequate for China, where the roots of modernization can be found in the rural areas as well as in cities and where participation in state-guided modernization projects such as educational reform could actually strengthen community ties. Despite the potential financial burdens brought by new schools, in Haicheng county the mandate to build schools, accompanied by an increased state presence in local society, did not simply create hostility and resistance or lead to community breakdown. Instead, local communities participated willingly in the state’s modernization program and defined a role for themselves by taking the initiative to raise funds for primary schools, by forging partnerships with other villages to set up schools, and by seeking the county government’s guidance and mediation. Although rural communities in Haicheng had difficulty coming up with resources, the state demand that they build schools generally served to mobilize them, bringing residents together and strengthening their sense of collectivity. Furthermore, when these communities encountered difficulties, they reacted not by opposing state pressures but by demanding further state intervention. The result was that villagers, striving to improve life in their own communities but also choosing to embrace state-guided educational reform, were simultaneously marking their membership in two communities: their immediate rural community and a larger national community.

The Setting: Fengtian Province and Haicheng County

In the late imperial period, education in Fengtian (present-day Liaoning) was considered to be backward and underdeveloped compared with that in other

village-state cooperation

289

provinces, a reputation derived at least in part from the province’s lackluster record of producing holders of civil service examination degrees.6 In the late nineteenth and early twentieth centuries, however, increased migration from other parts of China, the development of railroads, and the rise of reformminded leadership gradually brought the Northeast—especially Fengtian— into line with the rest of the country (Enatsu, 1991: vi–ix; Gottschang and Lary, 2000: 2, 45–46; Isett, 1998: 8; Suleski, 2002: 5). An important result of this transformation, especially after the New Policies were implemented and the civil service examination system was abolished in 1905, was the rapid growth and development of new educational institutions. By 1908, 2,113 schools had been established across the province, including professional schools, normal schools, and middle schools (LJSZ, 1.300–303; Li Xiping, 1998: 208). While some of these schools were located in the provincial capital of Shengjing (Shenyang) or in prefectural and county seats, the vast majority (1,925) were lower primary schools in rural areas (LJSZ, 1.300– 303).7 Almost one-fifth (333) of these schools, officially known as “communityestablished primary schools” (gongli chudeng xiaoxuetang), were established in Haicheng county (HXZ, 1987: 483; LJSZ, 1.300–305). The relatively large number of schools in Haicheng county and its rich archival sources from the late Qing and early Republic on educational reform make it an ideal area in which to conduct a close-up study. The impressive spread of schools there in the early twentieth century can be attributed to the county’s location and its economy. Haicheng county was (and still is) located halfway down the Liaodong Peninsula, between the cities of Shengjing and Dalian. In the period under study, the county was bifurcated by the Japanesecontrolled Southern Manchurian Railway line, ensuring its easy integration 6 According to both Ho P’ing-ti’s and Benjamin Elman’s statistics on the numbers of jinshi candidates (those who obtained the “palace graduate” degree, the highest in the civil examination system) during the Ming and Qing dynasties, the Northeast makes a poor showing (Ho, 1962: 227–29; Elman, 2000: 699–700). Several Chinese scholars also describe education in Northeast China, prior to the abolition of the examination system, as backward and underdeveloped (Guo, 2001: 1; Qi, 1992: 2; Sun, Dong, and Zhang, 1993: 3). 7 To provide some context as to what these figures might mean, I have compared Fengtian with some other provinces. In 1907, Fengtian had more lower primary schools than more than two-thirds of China’s (then) twenty-three provinces. It had nearly twice as many lower primary schools as Jiangsu (626) and Zhejiang (632), the two provinces with the richest tradition of civil service success. Fengtian ranked fifth in number of lower primary students, behind Sichuan, Zhili, Hubei, and Shandong. The following year, Fengtian had the thirdhighest number of lower primary students; only Sichuan and Zhili had more (Jindai jiaoyu shi ziliao, 1981: 2.85; LJSZ, 1.300–303).

290

VanderVen

with the rest of the province and likely making it receptive to new ideas and reforms. Equally important, Haicheng county, a well-irrigated agricultural hub, has been dubbed by its own inhabitants the “rice bowl of southern Liaoning” (Liaonan liangcang); it is home to a vast and fertile plain that for centuries has produced corn, soybeans, sorghum, and paddy rice (HXZ, 1987: 1–2). In addition, early in the century a number of Japanese settlers resided in the county seat. Gavin McCormack suggests that this Japanese presence engendered an angry and patriotic response among locals that manifested itself in a variety of spheres, including education (McCormack, 1977: 94–95).

Funding the New Community Primary Schools

Costs of Establishing and Operating Community Primary Schools The Haicheng county government funded only a small number of official (guanli) schools, mostly located in Haicheng city, the county seat. These elaborately planned schools, equipped with dormitories, dining halls, and libraries and enjoying substantial annual operating budgets, were a league apart from the numerous community primary schools set up by local villages. In 1924, budgets for official schools ranged from about 3,000 yuan for the five upperlower primary schools to more than 8,000 yuan for the boys’ middle school (HXZ, 1924: 6.24–25), a considerable sum at the time. The bulk of this money came from rents on the county’s various landholdings, which were scattered among several of Haicheng’s villages, as well as from local taxes and assessments.8 However, as mentioned above, these assessments were used only for the official schools; community schools generally did not benefit. The task of funding the community primary schools fell to the villages or townships where they were to be located. Compared with the official schools, these schools had relatively modest operating budgets. Just the same, there 8 The county owned a total of 7,940 mu of land, acquired from a variety of sources. Almost half (3,823 mu) came from landholdings that had formerly belonged to the county’s imperial academy (shuyuan). Another 2,537 mu came from temple lands turned over to the public and then appropriated by the county. Village associations had donated 1,394 mu, and countyowned temple property provided another 95 mu. Finally, private individuals donated an additional 90 mu. In addition to maintaining the schools it established, the county funded the educational promotion bureau (quanxuesuo), the administrative office responsible for overseeing educational reform; it also reserved some money for subsidies, used mostly to aid private girls’ schools and some sishu (private informal schools) that needed financial help for a successful conversion into a regular school. After 1919, upon orders from the provincial office of education to the promotion bureau, the county began to use fees from marriage certificates (hunshu) to subsidize schools, particularly those for women.

village-state cooperation

291

were many costs involved and communities sometimes struggled to establish and maintain institutions that were required to adhere to a uniform set of standards. The new curriculum and teaching methods called for a host of potentially expensive purchases, such as clocks, maps, blackboards, and textbooks. In addition, all schools were required to maintain a playground (caochang) sufficient for at least thirty active children to run around on and perform calisthenics and drills. Items necessary for basic operation—which in Haicheng county, with its freezing winters, included a heated brick bed (kang), plenty of fuel, and even tea—also had to be paid for. Not surprisingly, the most basic requirement was a place to house the school. In the past, when lower-level education had gone largely unregulated by the state, schools could and did operate anywhere—in private homes, clan halls, even shops. The new system, in contrast, sought to separate schooling from other activities. Ideally, communities were to erect a brand-new school building where only schooling was offered. But since most communities had insufficient resources to build new structures, the Qing government permitted them to house schools in preexisting ones as long as no other activities took place in them at the same time. Although schools most commonly occupied temples, some communities did manage to construct new buildings from scratch. Few extant records detail the cost of constructing new buildings; an exception was provided to the county by the village of East Lianhuapao. In 1906, the village constructed an entirely new school on land donated by two of its residents. The building, including materials for the foundation, walls, and roof, as well as miscellaneous items such as nails and tools, cost 527 yuan. It was composed of three jian (bays), two to serve as the classroom and one as the teacher’s living quarters. Because it was similar in size to other school buildings being constructed or renovated at the same time, we can assume that the costs for other communities were likewise similar. Communities that did not receive land endowments or did not possess communal land generally rented property, with rents of course varying according to the land’s location, quality, and area. The Teng’aobao upper-lower primary school paid 100 yuan per year for a plot on which it built a school from scratch. Teng’aobao, a township, was significantly larger and more heavily populated than were most villages in the county, and rents there may have been relatively high. The village of Yantai, located 60 li to the southeast of the county seat, paid 40 yuan per year to rent a house of five jian for its school. The village of Houyingleshan, 50 li from the county seat, rented a tile-roofed house of seven jian for 13 so-called foreign yuan (yang yuan). Among the many other significant expenses borne by communities were those for the teacher’s salary, classroom furniture, and books. Although the regulations did not stipulate the amount teachers were to be paid, salaries

292

VanderVen

were fairly consistent across Haicheng county. From 1907 through the early Republic, most schools paid the teachers between 150 and 200 yuan per year, often with food and some supplies included.9 Furniture and equipment consumed a large part of a school’s budget. In 1906, the Teng’aobao School’s single largest expenditure was 200 yuan for a quantity of desks. Textbooks cost a total of 53 yuan. The school also employed laborers to construct its school building and hired a cart to haul the textbooks from the bookstore in the county seat back to the town (HXGDA, 6.17782). In addition, schools were required to set aside part of their budgets to pay for the numerous small items required for daily operations. These cost anywhere from a few jiao to a few yuan. In 1906, the Teng’aobao School management purchased a wide variety of supplies, including paper, glue (2.4 yuan), string (3 jiao), a desk clock (5.4 yuan), a hook for the door screen (4 jiao), and lamps, chalk, and bookshelves (10 yuan). Unlike most other schools, Teng’aobao provided its pupils with a noon meal, necessitating the purchase of kitchen supplies, including a knife, a basin for washing dishes, chopsticks, bowls, fuel, and grains such as sorghum (8 to 9 yuan per month), corn, and rice. These items, required by all primary schools to some degree or another, added up. In 1906, the total budget for the Teng’aobao School, not including the teacher’s salary, was 1,157.23 silver yuan (HXGDA, 6.17782). .

Methods of Funding Primary Schools To meet annual school budgets that sometimes exceeded 1,000 yuan, rural communities such as East Lianhuapao and Teng’aobao had to find the money wherever they could. Because any single source of funding rarely could sustain a school, the community primary schools almost always relied on a combination of sources, including community funds and resources, individual donations, and tuition fees. This practice appeared to be common across early-twentiethcentury China. Sally Borthwick has found that several primary schools listed in the Hangzhou prefectural gazetteer were “composite” schools, drawing revenue from a number of different sources. For example, funding for the Qiantang Yuhang Tiaoxi Combined Higher and Lower Primary School came from public property, voluntary contributions, fines, and boarding fees. The sources of funds for the Qixi Combined Higher and Lower Primary School were just as varied: temple property, rent, and a tax on the town’s cooks (Borthwick, 1983: 98–99). Similarly, Stig Thøgerson notes that at least one rural primary school in Zouping county, Shandong province, was supported by an endowment of 9 The role of inflation on teacher’s salaries is not clear. Before 1922, prices in Northeast China were fairly stable, but the currency in the Northeast later began to lose value rapidly owing to conditions in the New York and Japanese markets (McCormack, 1977: 192).

village-state cooperation

293

temple land, temple association funds, and a tax on water mills (Thøgerson, 2002: 47). State Funds Past scholarship has not made sufficiently clear the degree to which the new community primary schools may have benefited from government subsidies. Some scholars have suggested that the state provided new schools with subsidies and that the prospect of such subsidies encouraged communities not only to set up schools but also to compete over resources (Borthwick, 1983: 88–103). Marianne Bastid notes that community schools were eligible for state subsidies (Bastid, 1988: 69), and Sidney Gamble, in his study of Ding county, reports that in the experimental district, fourteen of sixty-three village schools received subsidies from the county government for “special excellence” (Gamble, 1954: 206).10 In contrast, Thomas Curran stresses that villages and rural communities generally did not receive state subsidies and in fact bore a double burden: they were required both to contribute extra taxes to support official schools and to establish schools in their own communities whose support was solely their responsibility (Curran, 1993: 38–39). Such a range of interpretations suggests that state subsidies for community schools were spotty but generally quite infrequent throughout China. In fact, the 1904 central regulations did not call for provincial, prefectural, or county governments to subsidize community schools (Shu, 1961: 2.417). In Haicheng county, the financial contribution of the state to village primary schools was insignificant; the schools did not as a rule receive subsidies from the county. In Haicheng, villagers were required to pay nonstatutory charges on the land tax (tankuan or mujuan) as well as household taxes collected and administered by the village leaders (huishou). In the early years of the twentieth century, nonstatutory charges were imposed by county governments on villages as a whole to help the Qing meet indemnity and military expenses and to fund new institutions at the local level. Both the collection and allocation of taxes for such local purposes were the responsibility of village leaders (Duara, 1988: 64–65; Huang, 1985: 278–80; LJSZ, 1.300; Li Xiping, 1998: 213).11 In Haicheng

10

11

The fourteen schools in Ding county, Hebei province, that received county subsidies were exceptional cases: they were located in the county’s experimental district, where there was special emphasis on implementing reform (Gamble, 1954: 5, 10, 206, 462). Huang points out that nonstatutory charges had different names in different localities (Huang, 1985: 278–80). In Hebei, where Prasenjit Duara focuses much of his research, the taxes were referred to as tankuan; in Fengtian province they were known as mujuan, while household taxes were called hujuan (LJSZ, 1.300–303).

294

VanderVen

county, the government supported only a few official schools and none of the community primary schools with these moneys. In villages, the village head (cunzhang) and leaders played a key role in funding and establishing the community primary schools. They often sat on school boards that gathered community funds, collected school fees from students and their families, supervised school budgets, and, subject to the county’s approval, appointed the teachers (Gamble, 1954: 189; HXGDA, 6.18281, 6.17776, 7.20919). The school board also served as a liaison between the village community and the county authorities, especially the educational promotion bureau (quanxuesuo), the county administrative office responsible for overseeing educational reform. Controlling school funding and other community resources may have provided village leaders with new opportunities for corruption and peculation, as past scholarship has suggested (Borthwick, 1983: 98; Duara, 1988: 112), but it should by no means be assumed that malfeasance always occurred. In Haicheng county, more than 300 schools were successfully operating by 1907. This success can be explained at least in part by the local community’s responsiveness to orders to establish new schools, and by the village leaders’ flexibility and creativity in utilizing community resources to fund the schools. Community Property Community property, most commonly land, was an important source of funding for many village primary schools in Haicheng county. In some cases, this land had been previously purchased by the village and its rental provided a continuous source of money for the newly instituted school system. In other cases, individuals donated land to the village as an endowment. Unfortunately, sources do not reveal exactly how often schools in the county were built on or funded by community-owned land. Usually records exist only if the use of the land was disputed in some way that required the intervention of the state—as happened at least twice (HXGDA 6.18608, 17999). The written documentation from these two conflicts illustrates how village communities responded to state orders to set up new schools. First, villages that possessed community resources often drew on them to establish and maintain the schools, showing a willingness that itself suggests their eagerness to participate in educational reform. Second, once villages had invested in the schools, they took serious measures to protect their investment. Significantly, these disputes show that village communities increasingly recognized that the county government could be an important mediator when their interests (here, their schools) were threatened.

village-state cooperation

295

The first case comes from Jinjiatai village, where village leaders purchased a piece of temple land from a private owner in 1894 (Guangxu 20). Many years later, sometime after the new school system was implemented, Jinjiatai village chose to build its new four-room primary school on a portion of this land. To raise additional funds for the school, the village rented out the remainder of its property. The school had been operating for only a year or two when a banner-land manager (zhuangtou) from another village made some claims to the land, annexed part of it, and then tried to rent it out.12 Not surprisingly, the Jinjiatai village head and other leaders passionately disputed this action and requested immediate intervention from the county government. After a quick investigation, the county authorities ordered the banner-land manager off the land so that the village school could continue to benefit from its rents (HXGDA, 6.17999). We today lack sufficient information to assess the competing claims to the land made by Jinjiatai village and the banner-land manager. But regardless of which were more legitimate, the county, in the interest of furthering education and easing tension, ruled in favor of the village. This case illuminates the importance of community-owned property to the development of educational reform in Haicheng county. More importantly, it demonstrates the substantial commitment that Jinjiatai had made to the school by devoting communityowned property to it; the village wanted the state to protect its commitment and the state obliged. The second case, from 1909, revolves around a conflict over the use of endowed land;13 it reveals how communities and their individual members encouraged and supported the new schools. Li Chunrong, a resident of Gaokanwan village and a student at the imperial college (jiansheng), donated land to Gaokanwan so it could set up a community primary school. When the actions of an unscrupulous monk threatened the school’s existence, Li, along with the village leaders, filed a report with the county. In it, Li recounted that many years earlier his late father had given 120 mu of land as alms to the Shunhe temple located in the neighboring village of Xiaolianggou. Li was responsible for administering the land, and in 1907 he allocated 20 mu to support the monk who occupied the temple and gave the remaining 100 mu to Gaokanwan village for its school. 12 13

Zhuangtou (bailiffs) were state agents appointed by the Qing government to manage land and collect rent in Northeast China (Isett, 1998: 84–85). Communities had used rents from land endowments to partially fund schools since as early as the Ming dynasty (see Rawski, 1979: 66–79; Schneewind, 1999: 301).

296

VanderVen

The village intended to farm its portion of land and use the harvest proceeds to fund the school. But without informing anyone, the monk, described by both Li and the village leaders as “evil” (e), pawned 30 mu of land, including 10 mu that belonged to the school. Both Li and the villagers sought the county’s assistance in removing the monk from the land (HXGDA, 6.18165). Although the dispute’s final outcome was not recorded, this case, like the one above, highlights the importance of donated property to a school’s survival. The Gaokanwan villagers clearly depended on the land to cover many of the school’s expenses. The villagers, along with the individual who endowed them with land, brought in the state to protect their own interests as a community. In many other instances, villages in Haicheng county used rents from collectively owned land to partially cover school costs. For example, Xiaomatou village owned 100 mu of community land, a relatively large amount, and applied the annual rent to many of the operating costs of its school. Henanliugongtun village owned 20 mu of community land, a plot that generated enough rent to offset students’ tuition fees. Similarly, Longtaipu village used the rental from approximately 17 mu of land to cover tuition (HXGDA, 6.18266). If a village was lucky enough to be able to support a school partially or fully with community-owned land, then clearly many parties would benefit. From the village’s point of view, when school expenses were covered by rent from community land—as in Gaokanwan village, where the imperial student, Li Chunrong, specified that his donated land should support the community school—then other collectively generated funds might be freed for other purposes. Individual families also benefited when, as in Henanliugongtun and Longtaipu, the village leaders chose to use rent to cover tuition fees. Families that did not have to dip very deeply into their own pockets would be more inclined to see the new schools as a boon to the community rather than a burden. Finally, the county government benefited from not having to expend energy and resources helping villages to find other alternatives or policing absenteeism engendered by high costs for individual families. Individual Donations Some schools received significant amounts of financial support from individual members of their community. For example, in 1907 two cousins, Liu Ximing and Liu Xisheng, who were “anxious to contribute to the public good” (jigong haoyi), footed much of the bill to build a new school in East Lianhuapao village, a community of about ninety households.14 At age 29, Liu Ximing was a graduate of the provincial police academy who worked for Haicheng county as 14

The number of households in 1924 was eighty-six; there are not comparable data for 1907, when the school was set up, but it is unlikely that this figure changed drastically.

village-state cooperation

297

a magistrate’s assistant. Liu Xisheng, at age 32, managed the East Lianhuapao School (HXGDA, 6.18107; HXZ, 1924: 2.2). Liu Ximing donated 2 mu of land for the school’s site and another 2.2 mu of quarry land to provide clay for its bricks. The rest of the money came from private donations that Liu Xisheng solicited from various villagers. In total, the cousins collected 727 yuan, a substantial contribution. As mentioned earlier, the cost of building the Lianhuapao Village School was 527 yuan; the remaining 200 yuan could go toward other expenses, such as the teacher’s salary. The residents of East Lianhuapao later expressed their appreciation for the Lius’ help by asking the county government to reward their efforts (HXGDA, 6.18107). The primary school in Daluyan village, which was jointly operated by Daluyan and the neighboring village of Hougangouzi, also benefited from individual donations. Wealthy households in both villages contributed 300 yuan to the school (HXGDA, 6.17776). Similarly, a number of individuals in the community donated between 1.5 yuan and about 25 yuan per person for the Teng’aobao primary school (HXGDA, 6.17782). Perhaps the most striking effort by individual community members was launched in the large village of Xintaizi; in 1913 one of the residents, Gao Yuanzhong, founded a girls’ primary school in a neighbor’s yard in the east of the village. It proved so popular that Gao set up another girls’ school in the village a few years later. He funded the school by campaigning vigorously for donations from individual members of the community. He himself raised about 500 yuan from fellow villagers. One relative, Gao Yuanzhi, donated 500 yuan of his own money. Others contributed more modest amounts, ranging from 110 yuan to as little as 10 yuan. In the end, eighteen different villagers gave 1,765 yuan for the two schools (HXGDA, 6.19135). It is clear that local communities often took the initiative and could be quite creative in finding ways to pay for the new schools mandated by the state, despite the financial hardships they incurred. Indeed, in the absence of any formal and organized system for funding and with little state-provided financial support, this individual and community response was crucial to the successful implementation of the new school system in the early twentieth century. Tuition Fees Regardless of their other sources of funding, community primary schools in Haicheng county invariably charged tuition fees.15 The practice of collecting 15

Even schools like the Xiaomatou School, which covered the cost of instruction with funds generated from rental of community lands, formally charged fees—which in this case did not have to come from individual pockets.

298

VanderVen

fees, as described below, was both sanctioned and, at least in theory, limited by the state. The way fees were handled reveals an increasingly complex interaction between the county government and local society. Although the collection of fees was left to the discretion of individual villages and schools, the state was to act as regulator and prevent local communities from imposing unreasonably high fees. But because the county government recognized the importance of fees to many schools’ operating budgets, it often overlooked the charging of very high amounts. Nevertheless, such (over)charging did not necessarily indicate that village leaders and school boards were exploiting the families of students. Indeed, village leaders sometimes subsidized student fees or tried to offset them with other funds. Only official primary and normal schools were to be free, according to the 1904 regulations (Borthwick, 1983: 107; Shu, 1961: 2.418). They set forth no formal policy for the community primary schools. Therefore, the question of whether fees should be assessed, and at what rate, rested with local authorities. Late in 1906, however, the Ministry of Education formally issued regulations allowing localities, “depending on local circumstances,” to charge tuition fees (HXGDA, 6.17858). At the same time, the ministry sought to prevent local officials and schools from exploiting students or corruptly overcharging them. First, the regulations emphasized that to the greatest degree possible, tuition charges should be avoided. Second, they set a maximum that students could be charged. Lower primary students were to pay no more than 3 jiao in silver per month or 3.6 yuan per year; upper primary students were to pay no more than 6 jiao per month or 7.2 yuan per year (Bailey, 1990: 103; Borthwick, 1983: 107; HXGDA, 6.17858). In the tenth month of 1906, the Fengtian Office of Education relayed these regulations to the province’s prefectures and counties. At that time, the lower primary students in Haicheng county were already, in accordance with a previous local order, paying about 5 yuan per year in tuition. Furthermore, the new regulations were not officially put into practice until the spring of 1907, when two school managers requested that the county fix a policy on school fees. The two managers, both from the Gengjiazhuangzi village upper-lower primary school, wanted the county to publish an official policy that would justify their collecting money from the upper-level students at the school. They admitted that when they opened the school in 1906, many parents hesitated to enroll their children because they feared heavy fees. Therefore, to attract students, the school initially used community funds to cover tuition and school supplies, but it could no longer afford to do so. The two managers pointed out that local regulations allowed schools to charge lower primary students up to 5 yuan in fees each year; they felt the policy for upper primary students should

village-state cooperation

299

be similar. They asked the county to set the fees as it saw fit, but insisted on the importance of issuing an order (HXGDA, 6.17858). If these managers are to be believed, the Gengjiazhuangzi school delayed collecting fees as long as possible and asked the county to set a policy only when community funds were no longer sufficient to support the students. It is also clear from the managers’ request that the local community saw the state as responsible not just for helping them to facilitate the educational reform but also for protecting their interests. In response, the promotion bureau issued a general policy that essentially reiterated the regulations issued by the Ministry of Education a few months earlier. Thus lower primary students were required to pay 3 jiao per month or a total of 3.6 yuan per year, and upper primary students 6 jiao per month or 7.2 yuan per year. The bureau went further and set a standard timetable for payments as well. Fees were to be collected twice a year, half on the Dragon Boat festival (the fifth day of the fifth lunar month) and half at the Mid-autumn festival (the fifteenth day of the eighth lunar month) (HXGDA, 6.17858).16 Despite the official order for schools to stay within the limits set by the Ministry of Education, most schools in the county did not lower their annual fees from 5 yuan to 3.6 yuan.17 In trying to enforce the regulations, the promotion bureau was caught between the interests of the local communities and the interests of the school boards. Lower fees or no fees would make the new schools more attractive to the public, but would undercut a school’s operating budget. For example, a school with thirty students paying 5 yuan each would receive a total of 150 yuan in tuition fees. If school management reduced fees to 3.6 yuan per student, the amount stipulated by the ministry, it would lose 42 yuan annually, an amount that would cover a significant portion of rent or purchase a considerable quantity of supplies. Because the county could not compensate schools for their lost revenue, it took the easier course and overlooked the enduring practice of charging 5 yuan. 16

17

The promotion bureau also took this opportunity to settle other questions regarding school expenses, outlining who should be responsible for what. The bureau officially stated that all expenses, except for miscellaneous expenditures for such items as school equipment, furniture, window paper, oil, and fuel, were to be paid for with community funds. The school itself was to be responsible for miscellaneous expenses. In a very small concession, the county agreed to provide, as a matter of course, some smaller items to each school: twice a year, it would distribute brooms and dusters; once a year, a watering can, dustpan, and spittoon. Borthwick also points out that many schools elsewhere in China charged more than the amounts fixed by the Ministry of Education (Borthwick, 1983: 107).

300

VanderVen

A representative sampling of sixty-six primary schools across the county (roughly one-fifth of all the schools) attests to the ubiquity of the 5-yuan tuition fee during the 1907 school year.18 In forty-five of the schools (68 percent), the students paid exactly 5 yuan each. The remaining twenty-one schools frequently showed minor variation from the standard 5 yuan. In six of the schools, students paid from 5.2 to 5.5 yuan; in another school, each student paid 4.9 yuan. In the other schools, fees were slightly lower, hovering at around 4.5 yuan per student. The only dramatic deviations were the fee of 6.25 yuan that students paid to attend the Dagaokan Village School and the 2.2-yuan fee that students paid at the Xiaorujumiao Village School. There is no explanation for these outliers (HXGDA, 6.18122, 18130, 18120, 18124, 18117, 18123, 18132).19 Although the 5-yuan tuition fee was almost universal, differences in how fees were assessed demonstrate that sometimes village leaders, the school board, or both sought to ease the financial burden on families with schoolage children. Of course, in some communities or villages families paid tuition fees out of their own pockets. In others, school boards sometimes provided money to “subsidize” tuition fees. For example, the school in Liansantun village officially “charged” each of its twenty-eight students 5 yuan in tuition, but the board provided an extra 150 yuan in subsidies specifically to offset those fees (HXGDA, 6.18117). Similarly, although each of the twenty-five students in Erdaobian village was charged 5 yuan, the village leaders covered 120 yuan of the total sum (HXGDA, 6.18117). Recall also that both Henanliugongtun and Longtaipu villages applied revenues from community-owned land toward tuition fees. Some schools, like the one in Haiqingwan village, tried to help students in other ways. The school itself was a joint effort by nine neighboring villages, each of which contributed community money toward its funding. Each of the thirty-five students attending the school paid 5.7 yuan, an amount somewhat higher than in other schools. But this school also received some subsidies from 18 19

I looked at least five schools from each area of the county: the north, northeast, east, southeast, southwest, and west. In only one case were school fees actually reduced. In the spring of 1907, in Kongjiatun village in the western part of the county, students were paying 5 yuan each. However, only a few months later, ten students were paying 2.5 yuan each and fifteen were paying 3.5 yuan. The fee reduction may have been in response to the regulations as well as to individual need. The school thus experienced a total loss in income of 47.5 yuan, and it may have compensated by lowering the teacher’s salary. In the spring of 1907, the teacher received 180 yuan annually. By the summer, there was a new teacher who was being paid only 100 yuan, a low salary by county standards.

village-state cooperation

301

all nine villages that were specifically used to provide meals for the students rather than to cover their tuition (HXGDA, 6.18120).

Multi-Village Relationships: Creativity, Cooperation, and Conflict

Villages that were too poor or too small to set up schools on their own joined with other villages, pooling their resources. How these alliances were forged, sometimes dissolved, and forged again also reveals the cooperative effort between state and society. Village communities often initiated relations with other villages but also welcomed the state’s help in facilitating and supervising the process. Past studies indicate that multi-village cooperation outside the realm of education occurred frequently in North China during the nineteenth and early twentieth centuries. Prasenjit Duara notes that multi-village irrigation associations were common there at least as early as the mid-1800s. Martin Yang observes that the village of Taitou in Shandong province had long cooperated with several other villages in a maintaining a public defense system. Philip Huang shows how in nineteenth-century Baodi county, a group of about twenty villages collectively nominated the xiangbao, a rural tax agent, who was then officially appointed by the county government (Duara, 1988: 27–28; Yang, 1945: 194–95; Huang, 1985: 24–225). But despite this history of multi-village cooperation to such ends, the phenomenon of villages linking up to build schools appears to have been new to the early twentieth century. Scholars who write on education in the Ming dynasty (1368–1644), for example, do not mention that villages cooperated in operating schools. Although the first Ming emperor ordered a school to be opened in every village, the standards were not nearly as strict as those required of primary schools in the early twentieth century. In the Ming, it was acceptable to simply install a teacher in the village; any available space could be used. A Ming-dynasty education intendant who ordered schools to be set up in every village and hamlet reportedly said that doing so “need not be overly troublesome. Some might choose convenient nearby cloisters, others might themselves build or use existing buildings and schools. Follow the willingness of the people and the local customs and establish them according to local circumstances” (qtd. in Schneewind, 1999: 293–94). In any case, Ming standards for village schools were such that almost all villages could afford to maintain their own. The same was true of old-style sishu, the private informal schools that dominated primary education in rural China during the late imperial period.

302

VanderVen

Almost all sishu adhered to a Confucian-based curriculum, but that was their only commonality. They might be attended by a single student or fifty students, and they were variously managed by a single teacher, a village, or a clan. Very little money was needed to set up and maintain sishu, which were not so highly regulated as the new primary schools. It was precisely the high degree of state regulation that accompanied the order to set up schools in the early twentieth century that made multi-village cooperation in education necessary for those communities too poor to establish schools on their own. Such cooperation sometimes succeeded and sometimes was fraught with tensions and frustrations, necessitating adjustment and compromise on the part of participating villages. When problems could not be resolved locally, the state stepped in to mediate and supervise. The partnership between the villages of Dayushubaozi and Qianyushubaozi (hereafter referred to as Dayushu and Qianyushu), one of the longest on record, exemplifies a fairly successful joint effort at school building. In 1906, the relatively well-to-do Dayushu village built a primary school, purchased equipment, and opened its doors to its school-age children. A few months later, the educational promotion bureau sent the students from Qianyushu village to the school, whose status officially changed from a regular village primary school to a “two-village cooperative lower primary school” (liangcun heli gongli chudeng xiaoxuetang). The partnership between the two villages endured harmoniously for twenty-three years. Dayushu shouldered most of the expenses of the school it had originally built, but the students from Qianyushu paid fees to offset some of the costs. In the seventh month of 1929, however, the harmony between the two villages was temporarily shattered. In the words of the Dayushu village leaders, by then the area had seen “cultural improvement” (wenhua jinbu) and the school system had become more sophisticated, causing substantial increases in its yearly expenditures. The extra school fees from Qianyushu were no longer enough. The village leaders reasoned that since the school was officially a “cooperative school” (heli xiaoxuetang), Qianyushu should take on a larger portion of its expenses. The Dayushu village leaders approached the Qianyushu village leaders with what they thought to be equitable terms for further cooperation. They proposed limiting Qianyushu’s contribution to a share of consumable items such as kerosene, paper, and pens, as well as part of the teacher’s salary. In addition, they wanted help in paying for repairs to the school’s building and equipment. Dayushu’s leaders suggested that the amount each village should contribute should be based on the average amount of its land tax. When Qianyushu’s leaders rejected these terms, Dayushu had no recourse but to seek the county’s intervention.

village-state cooperation

303

Once the state stepped in, the situation resolved itself quickly. The county ordered Wang Qinghui, a promotion bureau officer, to inspect the two villages and confirm Dayushu’s claims. Wang visited both villages, which were located 18 li to the west of the county seat, and met with their leaders. After some discussion, they agreed on terms satisfactory to each village. Wang recorded the terms in a contract to which each village head and leader—ten people in all—affixed either his seal or fingerprint. The contract stipulated that the students of Qianyushu should, as in years past, pay school fees. In addition, Qianyushu was to pay one-fourth of the regular school expenses; Dayushu was to pay the remaining three-fourths and also cover building repairs and equipment (HXGDA, 7.20919). It is not clear whether the final division of financial responsibility was in fact based on land tax, as Dayushu had originally requested. However, Dayushu was more than twice the size of Qianyushu, with eighty-four households and 2,600 mu of land; Qianyushu, with only 32 households, held 765 mu (HXZ, 1924: 2.30–31). In any case, both sides accepted the terms and the school continued to operate, at least until the Japanese occupied Northeast China in 1931. The successful partnership between the villages of Dayushu and Qianyushu was forged under state guidance. The state coordinated the initial partnership between the two villages, essentially ordering Qianyushu to join forces with Dayushu. This kind of state involvement aided the formation of cooperative arrangements that were beneficial to both villages. Qianyushu, too small to build a school on its own, was able to send its students to the Dayushu Village School. Dayushu in turn received its share of tuition fees and, after 1929, onefourth of the total operating expenses from Qianyushu. The state benefited as well, as supervising the negotiations between or among villages helped it closer to its goal of setting up schools for every village. The following example similarly shows how the state helped to smooth relations between villages; it also demonstrates how the villages themselves might initiate cooperative relations. In the fifth month of 1909, four villages in the remote western part of the county informed the county authorities, as regulations stipulated, of their cooperatively built school that had been open since the second month of that year. In their report, the head and leaders of Bajiazi village explained why they built the school. Since 1905, Bajiazi’s villagers had sent its students to the school in Hanjiabao village.20 But the Bajiazi students often failed to attend, complaining that Hanjiabao was too far away. To solve the problem, Wang Zeru, a county educational promotion bureau officer, encouraged the Bajiazi village leaders to 20

Bajiazi gave the date as 1905, while Hanjiabao reported it as early 1906.

304

VanderVen

set up their own school. Because Bajiazi was not large enough to do this on its own, it pooled its resources with several nearby small villages, including Dongtaoyuan village, which also had been sending its children to Hanjiabao. The new school had already employed a teacher and ordered desks, chairs, a podium, and a blackboard—all elements required for a school to have formal operating status. Although the equipment had not yet arrived at the time the village leaders filed their report, the school, attended by eighteen students, was operating provisionally (HXGDA, 6.18281). Soon after receiving Bajiazi’s report, the county government sent Wang to the villages to confirm its veracity. Wang told the county magistrate that the Hanjiabao School was a good school—it had thirty students and everything was in working order. But he agreed that it truly was too far from Bajiazi, and therefore he supported Bajiazi’s request to end their relationship. Wang was also impressed at the excitement, despite their poverty, displayed by the three villages that had recently joined forces with Bajiazi. Wang reported that the school could soon expect an increase in enrollment and that the temporary equipment, though crude, was at least suitable (HXGDA 6.18281). With all matters thus settled, the Bajiazi Village School was officially launched. While this arrangement worked perfectly for Bajiazi and Dongtaoyuan, it dealt Hanjiabao a serious financial blow. Of all the outlying villages that had been sending students to Hanjiabao’s school, Bajiazi was the best off.21 When Bajiazi withdrew, Hanjiabao in its own estimation lost two-thirds of its income as well as the tuition fees from the six Bajiazi and Dongtaoyuan students who left the school. Hanjiabao’s village head and leaders quickly conveyed these concerns to the county authorities. But the bureau officer had anticipated Hanjiabao’s reaction and had already moved to find a solution. Wang informed the county authorities of several other small villages nearby that could begin cooperating with Hanjiabao and send students to its school. The case ended with the county supporting Bajiazi’s new cooperative arrangement and urging Hanjiabao to pool its resources with those of several nearby villages.

21

It is not clear how many villages had been sending students to the school in Hanjiabao. Bajiazi had 105 households and owned 1,010 mu of land. As Wang observed, the villages with which it later cooperated with were significantly smaller and poorer. Dongtaoyuan, for instance, had only 30 households and 270 mu; Hanjiabao had 39 households and 760 mu. But as Wang also indicated, if it pooled its resources with those of the four neighboring villages of Fuliyuan (21 households, 210 mu), Nanxiaohulitai (27 households, 230 mu), Xixiaohulitai (23 households, 620 mu), and Beixiaohulitai (32 households, 450 mu), they together could maintain a school (HXZ, 1924: 2.36–38).

village-state cooperation

305

This case, like that of Qianyushu and Dayushu, reveals the high degree of state involvement in multi-village relations. Although the source of the original impetus for Bajiazi to cooperate with Hanjiabao is unclear, the state intervened quickly once Bajiazi’s students began cutting class. Its intervention was welcome, at least in Bajiazi, because it gave the village a chance to justify ending its partnership with Hanjiabao and joining forces with other villages. This case also highlights that the county government and the villages had a mutually reinforcing relationship. The county depended on these multi-village alliances. From its point of view, if students failed to attend class and if villages could not afford to set up schools, its educational program was being undercut. By mediating and trying to find solutions that benefited as many villages as possible, it ensured that the reform would be implemented more smoothly. Not all disputes were not so easily resolved, however. In at least one instance, a partnership that two villages entered into on their own ended in a bitter feud that ultimately required provincial intervention, a relatively rare occurrence. In the early spring of 1906, the village of Daluyan, 35 li northwest of the county seat, set up a lower primary school. Lacking its own school, the neighboring village of Hougangouzi sent ten children to Daluyan. From the beginning, both villages contributed the funds needed to run the school, in proportion with the number of students from each village attending it. Since twenty children from Daluyan attended the school, that village was responsible for two-thirds of the funding and Hougangouzi one-third (HXGDA, 6.17776). The first five years of cooperation passed uneventfully. Ten terms were successfully completed, and the first class graduated after completing the five years of schooling specified in the regulations (Shu, 1961: 2.418). But in early 1911 the relationship between the two villages soured when some students from Hougangouzi began attending the school in the nearby village of Gushuzi. Daluyan’s leaders informed the county of the situation, stating that although the village had originally intended to set up a school on its own, its scant resources led it to initiate a partnership with Hougangouzi.22 After five years with no “dissenting words,” it was an unwelcome surprise when Hougangouzi’s village leaders “intentionally” sent students instead to Gushuzi (HXGDA, 6.17776). Upon hearing Daluyan’s tale, the county magistrate, Zhao Peiguang, ordered Hougangouzi to continue paying one-third of the school’s operating expenses. When Hougangouzi refused, the Daluyan village head and committee again 22

According to the 1924 gazetteer, Daluyan had seventy households and landholdings of 1,100 mu; Hougangouzi had fifty-five households and 700 mu (HXZ, 1924: 2.54).

306

VanderVen

tried to persuade Hougangouzi to return to the original arrangement. Hougangouzi still refused, and Daluyan once more requested the county’s intervention. In the meantime, the village heads of Hougangouzi took a step rarely taken by villages in Haicheng county: bypassing the county authorities, they appealed to the provincial authorities and explained why it was not in Hougangouzi’s best interests to maintain its partnership with Daluyan. At some point after their cooperation began, Hougangouzi discovered that Daluyan was simply inconvenient to get to. Students had to travel 4 li each way, a long journey made worse in the summer months by heavy rains that caused the water from the nearby Guchengzisha River to rise and flood the road to a depth of several feet. The flooding so frightened the students that they returned home; and while the Hougangouzi village leaders had attempted to persuade the children to continue attending the school in Daluyan, they had been unsuccessful. The refusal of many of Hougangouzi’s students to go to school in Daluyan did not mean that they were averse to the idea of school itself, as the village leaders pointed out. Soon the students discovered that they had been “seeking afar what was actually near” (shejin qiuyuan) and of their own volition began attending school in the village of Gushuzi. Gushuzi was only half a li from Hougangouzi, and its school appealed to Hougangouzi’s students. As the Hougangouzi village head reported: We do not need to persuade the children to attend the school [in Gushuzi]. They all want to go there. We have already discussed this with Gushuzi and are paying them the proper allocation according to the rules. Both villages agree with this plan. It really is more suitable for the students to attend the closer school. (HXGDA, 6.17776) The Hougangouzi village leaders were not concerned only about the distance between Hougangouzi and Daluyan. They also believed that the quality of schooling at Daluyan was substandard—and they had some reason for concern. In 1906, not long after the two villages entered their partnership, the Daluyan School received an unfavorable inspection report from a county promotion bureau officer. He noted that the teacher, who was over the age of 50, followed the “old ways” of education. The report further criticized the teacher for his errors in mathematics and even alleged that he moonlighted as a quack doctor. Finally, the school was unacceptably dirty: the classroom was deep in dust and covered with cobwebs (HXGDA, 6.18130). Hougangouzi’s belief in the

village-state cooperation

307

marked superiority of the Gushuzi School tallied with comments made about it by the same bureau officer. After his inspection, he reported that the facility was in good working order and that the teacher was clear, followed new methods, and possessed good math skills (HXGDA, 6.18130). In addition, Hougangouzi’s leaders intimated that Daluyan was not as strapped for funds as it professed to be. Although Daluyan’s leaders claimed that the school needed the additional enrollment from Hougangouzi, Hougangouzi’s leaders said that at least ten Daluyan children who should have been paying fees to the village school were attending an illegal sishu. They thereby implied that if the sishu would close down and its students attend the village school instead, Daluyan would not need the fees from the Hougangouzi students. Their assertion was also in effect a rather serious allegation, for the county was making increasingly rigorous efforts to eliminate illegal sishu and there should not have been one in Daluyan in the first place. But the provincial authorities found Hougangouzi’s claims unpersuasive. After ordering the educational promotion bureau to investigate the statements of both villages, the province told the county to enforce its initial decision requiring Hougangouzi to continue paying one-third of the Daluyan School’s operating expenses. Fortunately for Hougangouzi, the county leadership soon changed. Later in 1911, Magistrate Zhao, who had been responsible for the county’s initial order, resigned from office and was replaced by Magistrate Guo (Jinxiu). Hougangouzi’s leaders, taking advantage of the shift, pressed the issue with Guo until he set aside Zhao’s decision and worked out a compromise. The compromise allowed Hougangouzi to reduce its previous contribution by half, to one-sixth of the Daluyan School’s operating expenses, while sending only five students there. This arrangement worked for a short time although leaders from both villages continued to grumble. Two years later, Magistrate Guo was promoted to a higher position and replaced by Magistrate Tian (Yushi), who supported his predecessor’s attempt at a compromise but decided to alter the arrangement slightly by raising Hougangouzi’s contribution from one-sixth to one-fifth. Hougangouzi would continue to send five of its students to the Daluyan School. This arrangement should have appealed to both sides. Although Daluyan would not receive as much from Hougangouzi as it had during the first five years of their partnership, it would not be solely responsible for the school’s costs. For its part, Hougangouzi, although unable to divorce itself completely from Daluyan, had its payments reduced significantly. It also was able to send at least some of its students to the school in Gushuzi as it preferred. Magistrate Tian, however, wary of the bitter relations that had developed between the two

308

VanderVen

villages, decided to guarantee an end to the matter by ordering each party to sign a written pledge that it would raise no further complaints. Daluyan’s leaders balked; and borrowing the previous tactics of Hougangouzi’s leaders, they went over Magistrate Tian’s head and appealed to the provincial authorities: If we had sufficient resources, we could have set up the school ourselves. Why would we need [Hougangouzi’s] contribution? It is only because our village’s income is insufficient that we entered a partnership to build the school. If we follow Magistrate Tian’s decision that (Hou)gangouzi pay only one-fifth, then our school will have to close. For this reason, we plead with the xiantai (provincial administration) and ask it to order the county to follow the previous order by which Gangouzi pays one-third. Do not change the original formula by which Gangouzi pays one-third and we pay two-thirds. Then this school can still be a product of cooperation. (HXGDA, 6.17776.) Upon receiving Daluyan’s request, the provincial education office ordered the county to conduct yet another investigation,23 in a final attempt to bring the two villages back together. The county even expressed a willingness to provide subsidies to Hougangouzi—though the state, as I noted above, almost never provided such financial support—if it would send students to the Daluyan School. But Hougangouzi’s unenthusiastic response led the county finally to conclude that the dispute was irreconcilable. The county judged that of the two villages, Daluyan was the more capable of independently running a school, and thus decided that Hougangouzi should continue to send its students to the Gushuzi School. Daluyan, surprisingly, accepted this decision and informed the county that since it could not return to its previous relationship with Hougangouzi, it would like to cooperate with the neighboring village of Xinhetun. The relieved county allowed Daluyan to operate its school independently or to merge with Xinhetun, as it preferred. Thus the seven-year partnership between Daluyan and Hougangouzi ended in 1913. Throughout the conflict, the school remained open and in 1924 was still

23

The provincial office speculated that other issues underlay the tension between the two villages but acknowledged the pointlessness of guessing what they might be.

village-state cooperation

309

operating.24 The Gushuzi School, too, was open in 1924 and presumably still attended by children from Hougangouzi.25 Although the partnership between Daluyan and Hougangouzi was largely unsuccessful, it raises several important points. First, as was also true of the previous two cases, when tensions arose between villages, the villages drew in the county government as a mediator. When the county failed to rule in Hougangouzi’s favor the first time, Hougangouzi called on the province to mediate. Daluyan responded in kind, bringing in county officials again and then, dissatisfied with the compromise, bringing in those at the provincial level. Second, the villages made it clear more emphatically than in the previous two cases that they were willing to carry out the state mandate to set up schools but lacked the financial means to do so. Recall that Daluyan told the province that it entered into a partnership with Hougangouzi only out of financial necessity. Such an assertion implies that Daluyan cooperated with Hougangouzi because it was doing its best to carry out state orders. It also implies that if Daluyan encountered difficulties in carrying out state orders, then the state had an obligation to step in and offer assistance. Third, despite the tensions between the two villages, it is clear that neither was averse to the idea of cooperation. From the beginning, Daluyan wanted to maintain its partnership with Hougangouzi. In the end, when their partnership collapsed, Daluyan chose to cooperate with the village of Xinhetun rather than close down its school. Similarly, Hougangouzi objected not to cooperation but to cooperating with Daluyan: it preferred Gushuzi as a partner. This attitude also suggests that many villages were enthusiastic about setting up new schools and participating in the state’s modernization project. Rather than simply carrying out state orders, the villages also recognized the benefits that came with new-style education. The Hougangouzi villagers behaved strategically; the village leaders and community acted to ensure that 24

25

By 1924, the school was officially called Gushuzi Citizens’ School (guomin xuexiao). All community primary schools were ordered to formally take the name Citizens’ School in 1915, but little changed except the name. Each community was still responsible for funding and maintaining its school and the content of education remained largely the same as before, although the years mandated for primary schooling were reduced from five to four. As of 1924, there was no school in Hougangouzi. There is no evidence of whether Daluyan maintained a successful relationship with Xinhetun, a village so tiny that it could be found neither on the county map nor in the county gazetteer.

310

VanderVen

Hougangouzi’s students were able to attend a better and more conveniently located school. Finally, this case illustrates the degree to which state institutions and local society cooperated to implement educational reform. When villages too poor to set up schools on their own pooled their resources with other villages, the state gained. Without having to make any investment itself, the state saw its educational program reach into areas to which they previously had not penetrated. These arrangements also benefited the villagers. More villages both gained access to the new-style education, thereby playing a key role in modernization, and also took advantage of new opportunities to engage with the state. Without question, the state program to set up primary schools drained community resources. Villages and townships bore a double burden, paying new taxes to fund the elaborate government schools located in the county seat and larger towns and bearing the costs of building and running schools in their own communities. Receiving little or nothing from county coffers, local communities had to locate funding on their own. In the face of these difficulties, communities in Haicheng county exhibited surprising initiative and effort in their attempts to implement the state reform program. They creatively pieced together resources from a multitude of sources; they drew on community property and, when possible, tried to offset expensive tuition fees with community subsidies. Individuals frequently donated land and money to the schools in their communities. The state, despite not providing money for community schools, played a prominent role in community mobilization. The communities called on the state to set fee policies and also to mediate when disputes arose over landholdings intended to support schools. Village communities in particular also brought in the state to initiate and supervise their efforts to cooperatively run schools with other villages. These arrangements produced both new tensions—between villages, rather than between village and state—and new opportunities. Most importantly, local initiative and the cooperative relations between state and society in education reveal that rural communities were crucial in China’s early-twentieth-century modernization. Although the process of educational reform was not smooth or uniform, it is clear that village communities, at least in Northeast China, supported it wholeheartedly. This article has focused on how educational reform was implemented in one county in Northeast China during the early twentieth century. Though certain factors may have been unique to Haicheng county—its optimal location, its economy, and the sizable Japanese presence in its county seat—statistics

village-state cooperation

311

from other counties and administrative areas in Fengtian province as well as ethnographic data from North China suggest that the pattern of educational reform that obtained in Haicheng County during this period was representative of what was taking place elsewhere. Throughout Fengtian, community schools were funded by a variety of means, including tuition fees, community property, and rents (LJSZ, 2.300–303). In North China, rural communities played an active role in implementing educational reform. They derived school funding from a ranges of sources, lightened the economic burden on families by offsetting tuition fees, and, when resources were scant, cooperated with other villages to set up schools (Nida, 1952–58: 1.120; Gamble, 1954: 206; Li Jinghan, 1933 [1986]: 207). While these observations are preliminary, they indicate the strong likelihood that in the early twentieth century, rural communities across China, like those in Haicheng county, enthusiastically participated in educational reform activities, thereby playing an important role in China’s modernization. Furthermore, they suggest that education was an area in which state and society shared interests and goals and enjoyed a cooperative relationship. References Bailey, Paul J. 1990. Reform the People: Changing Attitudes Towards Popular Education in Early Twentieth-Century China. Vancouver: University of British Columbia Press; Edinburgh: Edinburgh University Press. Bastid, Marianne. 1988. Educational Reform in Early 20th Century China. Trans. Paul Bailey. Ann Arbor: Center for Chinese Studies, University of Michigan. Borthwick, Sally. 1983. Education and Social Change in China: The Beginnings of the Modern Era. Stanford, CA: Hoover Institution Press. Cleverley, John. 1985. The Schooling of China: Tradition and Modernity in Chinese Education. Sydney: George Allen & Unwin. Curran, Thomas. 1993. “Educational reform and the paradigm of state-society conflict in Republican China.” Republican China 18, 2: 26–63. Duara, Prasenjit. 1988. Culture, Power, and the State: Rural North China, 1900–1942. Stanford, CA: Stanford University Press. Elman, Benjamin A. 2000. A Cultural History of Civil Examinations in Late Imperial China. Berkeley: University of California Press. Enatsu, Yoshiki. 1991. “The Rise of the Fengtian Local Elite at the End of the Qing Dynasty.” Ph.D. diss., University of Michigan. Esherick, Joseph W. 1999. “Modernity and nation in the Chinese city.” Pp. 1–16 in Joseph W. Esherick (ed.), Remaking the Chinese City: Modernity and National Identity, 1900–1950. Honolulu: University of Hawai’i Press.

312

VanderVen

Gamble, Sidney. 1954. Ting Hsien: A North China Rural Community. New York: Institute of Pacific Relations. Gottschang, Thomas R. and Diana Lary. 2000. Swallows and Settlers: The Great Migration from North China to Manchuria. Ann Arbor: Center for Chinese Studies, University of Michigan. Guo Jianping. 2001. Fengxi jiaoyu (Education in Fengtian). Shenyang: Liaohai chubanshe. Haicheng xian gongshu dang’an guan [Haicheng County Government Office Archives]. 1905–31. Liaoning sheng dang’an guan (Liaoning Provincial Archives). [Cited by catalogue and juan (file) number.] Cited as HXGDA. Haicheng xianzhi [Haicheng county gazetteer]. 1924. Haicheng City. Cited as HXZ, 1924. ———. 1987. Haicheng City. Cited as HXZ, 1987. Ho, Ping-Ti [He Bingdi]. 1962. The Ladder of Success in Imperial China: Aspects of Social Mobility, 1368–1911. New York: Columbia University Press. Huang, Philip C. C. 1985. The Peasant Economy and Social Change in North China. Stanford, CA: Stanford University Press. HXGDA. See Haicheng xian gongshu dang’an guan, 1905–31. HXZ. See Haicheng xianzhi. Isett, Christopher. 1998. “State, Peasant, and Agrarian Change on the Manchurian Frontier, 1644–1940.” Ph.D. diss., University of California, Los Angeles. Jindai jiaoyu shi ziliao [Materials on the history of modern education]. 1981. Beijing: Renmin daxue chubanshe. Kuhn, Philip A. 1978–79. “Local taxation and finance in Republican China.” Select Papers from the Center for Far Eastern Studies 3: 100–136. Li Jinghan. 1933 [1986]. Ding xian shehui gaikuang diaocha (A summary of the investigation of society in Ding county). Beijing: Zhongguo renmin daxue chubanshe. Li Xiping [ed.]. 1998. Liaoning jiaoyu shi (A history of Liaoning education). Shenyang: Liaohai chubanshe. Liaoning jiaoyu shi zhi ziliao [Materials on the history of Liaoning education]. 1990. 4 vols. Shenyang: Liaoning daxue chubanshe. Cited as LJSZ. LJSZ. See Liaoning jiaoyu shi zhi ziliao, 1990. McCormack, Gavin. 1977. Chang Tso-lin in Northeast China, 1911–1928: China, Japan, and the Manchurian Idea. Stanford, CA: Stanford University Press. Niida Noboru [ed.]. 1952–58. Chūgoku nōson kankō chōsa (Investigations of customary practices in rural China). 6 vols. Tokyo: Iwanami. Qi Hongshen [ed.]. 1992. Dongbei difang jiaoyu shi: Dongbei jiaoyu shi yanjiu congshu (A history of local education in Northeast China: compendium of the history of education in Northeast China). Shenyang: Liaoning daxue chubanshe.

village-state cooperation

313

Rawski, Evelyn Sakakida. 1979. Educational and Popular Literacy in Ch’ing China. Ann Arbor: Center for Chinese Studies, University of Michigan. Schneewind, Sarah. 1999. “Community Schools and Improper Shrines: Local Institutions and the Chinese State in the Ming Period (1368–1644).” Ph.D. diss., Columbia University. Scott, James. 1976. The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia. New Haven: Yale University Press. Shu Xincheng. 1961. Zhongguo jindai jiaoyu shi ziliao (Materials on the modern history of Chinese education). 3 vols. Beijing: Renmin jiaoyu chubanshe. Suleski, Ronald. 2002. Civil Government in Warlord China: Tradition, Modernization and Manchuria. New York: Peter Lang. Sun Jingyue, Dong Huiyun, and Zhang Xiuchun. 1993. Zhang Xueliang yu Liaoning jiaoyu (Zhang Xueliang and Liaoning education). Hong Kong: Xianggang tongze chubanshe. Thøgerson, Stig. 2002. A County of Culture: Twentieth-Century China Seen from the Village Schools of Zouping, Shandong. Ann Arbor: University of Michigan Press. Weber, Eugen. 1976. Peasants into Frenchmen: The Modernization of Rural France, 1870–1914. Stanford, CA: Stanford University Press. Yang, Martin. 1945. A Chinese Village: Taitou, Shantung Province. New York: Columbia University Press.

chapter 10

Power Networks and State-Society Relations in Republican China Danny Hsu Throughout the Qing Dynasty, a minimalist state with a small bureaucracy governed an immense country in both size and population. With regard to its ability to penetrate local society, the Qing state’s official bureaucracy reached only as far down as the county government, with the county magistrate as the lowest level salaried official. As a result of this arrangement, the county magistrate played an all-important role, responsible for carrying out the state’s orders in local society. Consequently, for the majority of the residents in the villages (and hence, over ninety percent of China’s population), the government meant the county yamen (Reed 2000, 198). The county magistrate could not perform his extensive duties single handedly. His efforts to secure tax revenues and maintain social control on behalf of the Qing state depended heavily on drawing local people into service in various capacities.1 And as the works of Ch’ü T’ung-tsu and Bradly Reed have shown, the Qing state’s reliance upon quasi-officials (such as clerks and runners) within the county yamen raises, among many others, the issue of the nature of the Chinese state and its relationship to local society. Qing officials often railed against yamen clerks and runners and the harm they did to effective local governance. They accused these quasi-officials of engaging in corruption, most typically through tax engrossment and extortionate fees for services. Moreover, many agreed that the size of the jurisdictions and the paucity of official staff at the county level rendered the county magistrate incapable of governing effectively. Thus, advocates of reforming the local government argued that in order to curb the rampant corruption and inefficiencies of local government, it was necessary to create a new sub-county infrastructure that would increase the number of officials and do a better job of systematizing the recruitment and bureaucratization of sub-county functionaries (Kuhn 1975, 263–276).

1 See below for a more detailed overview of the various ways in which the Qing state attempted to draw local people into service as informal agents.

© koninklijke brill nv, leiden, ���4 | doi ��.��63/9789004271890_012

power networks and state-society relations

315

Gaining greater control over local society and local government administration by establishing a sub-county administrative apparatus therefore became one of the most pressing issues in China’s early twentieth-century statemaking drive. One of the earliest and arguably most important innovations was the introduction of the sub-county ward (qu 区). Though the idea of the ward was first introduced during the last years of the Qing Dynasty, it was not until after 1915 that wards became the lowest formal unit of the Republican state, operating, in theory, under the supervision of the county government (Duara 1988, 61).2 Each county was divided into a number of wards in accordance with local conditions and population and each ward established a ward office that was responsible for a set number of villages and/or towns. Under the Guomindang, sub-county administration saw further development and elaboration, including revisions to the role of ward offices and the establishment of formal administration at the township-village level (xiangzhen 乡镇).3 Ideally, the Republican system of sub-county administration would be an improvement over late imperial local governance; sub-county officials could now at once be brought into greater bureaucratic accountability and enjoy a more regularized relationship with local society. As admirable as the new sub-county administrative apparatus was in theory, how did it work out in practice? Past scholarship on the impact of the ward in Republican local governance has highlighted some important themes. Philip Kuhn, Philip Huang, and Prasenjit Duara all agree that state-making and the introduction of the sub-county ward did not result in the Weberian ideal of the complete bureaucratization of the state.4 Throughout the Republican period, local government was continually plagued by so-called local bullies and evil gentry (tuhao lieshen 土豪劣绅) who, though acting as agents or at least semiagents of the state, were never subject to effective state control. In terms of state-society relations, all three scholars agree that state-making represented a threat to the village community and contributed to the deterioration of local society. While Kuhn and Duara suggest that villagers were exposed to unmitigated exploitation by bullies and “evil gentry” acting as state agents, Huang 2 In the first years of the Republican period, the ward was subsumed under the category of local self-government (difang zizhi 地方自治). Its regulations promulgated on December 29, 1914, under the title “Experimental regulations for local self-government” (difang zizhi shixing tiaoli 地方自治试行条例) (Yu 1918, 1.11.1). However, Philip Kuhn points out that in practice, the ward did not have any representative functions. Rather, it operated as an administrative arm of the magistrate (Kuhn 1975, 284). 3 See discussion below for more details. 4 See Philip Kuhn (1975), Philip Huang (1985), and Prasenjit Duara (1988).

316

hsu

argues that though there were villages that succumbed to exploitation, there were also villages that succeeded in resisting increased state intrusion. This chapter seeks to present a different perspective on state-society relations during the Republican period through an examination of sub-county administration and its operational realities.5 Archival materials from the Liaoning Provincial Archives in Shenyang and the Sichuan Provincial Archives in Chengdu allow us to examine more closely the nature of the Republican state at the sub-county level. For the early Republican period (up to 1931), this paper draws on case records from the Xinmin County archives currently housed at the Liaoning Provincial Archives. This collection contains over 11,000 files (juan 卷) that date from the last years of the Qing (starting in 1908) to 1931, when Northeast China was taken over by the Japanese army. The bulk of the material dates from the Republican period and deals with various aspects of county and sub-county administration, including personnel records, regulations, reports from the various bureaus, and most importantly, roughly 200 case records of accusations of ward corruption brought by residents of the villages within the jurisdiction of those wards.6 These cases, filed at the county government office, form the empirical backbone of the first half of this paper. The post-1935 section draws on a collection of 50 cases of ward- and township-village-level corruption and malfeasance from the Provincial Civil Affairs Department (Sichuan sheng minzhengting 四川省民政厅) archives, currently housed at the Sichuan Provincial Archives. While the archive catalogues (mulu 目录) indicate that there are over 200 files involving corruption and malfeasance allegations against county and sub-county officials, the actual number of cases in this collection is even higher; it is not uncommon for a single file to contain as many as six separate cases, with each case running well over 30 pages.7 In short, this substantial archival collection is a rich source of hitherto underutilized materials. 5 State-making and state-society relations are broad ranging topics that have been covered by many scholars. In this paper, my usage of the terms state-making and state-society relations will be limited to the context of county and sub-county governance and its interactions with local society at the village level. 6 While some files amount to no more than a few pages, a significant number contain over 50 pages, with detailed statements from both plaintiffs and defendants. Cases from the Xinmin archives will be cited by the letters XMDA, then catalogue number (mulu 目录) followed by juan number. 7 For example, a file labeled as containing a case against a magistrate could actually contain several unrelated cases against sub-county officials. Cases from the Provincial Civil Affairs Department will be cited first by the letters SPA, then the number 54 (following the Sichuan Provincial Archives’ cataloguing system), followed by juan number, and when necessary, a letter (i.e., a, b, c) at the end.

power networks and state-society relations

317

Up to the present, our understanding of the relationship between the local government and village has been grounded within the framework of mutual antagonism. To be sure, instances of cooperation and collusion (which is common to most—if not all—societies) also occurred at times between local state agents and villagers. For example, during the Qing, local leaders sometimes secured favorable treatment (e.g., overlooking illicit community activities such as salt manufacturing) from county runners in exchange for monetary compensation. Nevertheless, these interactions still fall within the framework of antagonism between state and society because the local leaders’ central motivation for colluding with individual runners was to check state intrusion (Reed 2000, 193–199). However, as materials from Xinmin County and Sichuan show, the relationship between sub-county government and local society was marked by more than just antagonism and the possibility of cooperation and collusion. Though the new sub-county administrative apparatus was intended to establish a crucial new dividing line between state and society, in reality the new government offices became arenas where individuals and groups of villagers contested with each other as well as with other state agents. The village did not unanimously resist the state, nor was it systematically subjected to unmitigated exploitation. In other words, the established paradigm of state versus society, with the possibility of occasional cooperation and collusion, simply does not work. Rather, from the view of the Republican sub-county administration, state-society relations were characterized by crisscrossing patterns of power networks that were formed in a number of ways and along various lines of interest that defy simple dichotomization.

Local Governance in Late Imperial China

During the Qing Dynasty, the county yamen was the lowest formal administrative unit of the civil bureaucracy. No formal government existed below the county magistrate (Ch’ü 1962, 1–2). Nevertheless, in theory, the state’s baojia system appointed quasi-officials all the way down to the paitou (牌头) (the man responsible for ten households) within each village. But as Hsiao Kungch’uan has shown, the baojia system was plagued with so many problems that some emperors complained that it existed only in name, and not in practice (Hsiao 1960, 72–83). Philip Huang has also pointed out that in reality, the lowest level quasi-official in north China that the state attempted to exercise influence over was the xiangbao (乡保), a tax collection agent in charge of about 20 villages. However, the state’s influence over the xiangbao was limited, for the agent was nominated by community leaders to serve as a buffer between the village and the state (Huang 1985, 225–227). Furthermore, even

318

hsu

with the services of the xiangbao, the bulk of local governance (e.g., legal casework, administrative duties, and police work) still fell within the purview of the county yamen and its clerks and runners. Consequently, county runners traveling from the yamen to the various villages would often have to travel long distances (as far as 55 kilometers in some areas). Without the aid of modern transportation, the yamen’s ability to accomplish its duties with speed and efficiency was often less than ideal. Under such circumstances, the state’s ability to penetrate local society was constrained by the limits of its physical location. Though detailed information is unavailable, there is enough evidence to present a general picture of the structure of local governance in the Xinmin region in late imperial China.8 At the outset of the Qing Dynasty, the Qing rulers sought to maintain Manchuria as the geographic and social heartland of the Manchus. Thus, the Manchurian countryside was structured into rent-paying estates and bannerlands (reserved for Manchu nobles and demobilized bannermen), accompanied by a somewhat different system of local governance than seen in China proper. Commoners who had migrated from north China were bound in groups of ten to the rent-paying estates as serfs (zhuangding 庄丁), each serf receiving an initial allotment of 36 mu of land.9 From among the ten, a bailiff (zhuangtou 庄头) was appointed to manage the estate and collect rent on behalf of the state. Above the bailiffs were the prompters (cui­ ling 催领), who were responsible for overseeing the collection and remittance of rent from the bailiffs of the rent-paying estates, the Manchu bannerlands, and the five types of imperial estates (Isett 1998, 84–93). Though on the surface the presence of one bailiff for every estate suggests a great level of state presence within society, in reality, local control in Manchuria (as in China proper) was fraught with issues of inefficiency and corruption. First of all, the state viewed bailiffs as members of society who, having been delegated managerial responsibilities, were in need of state supervision (Isett 1998, 104–105). Thus, the role of the prompter was of greater importance in local governance. However, there were simply not enough prompters to effectively oversee rent collection and law enforcement.10 Without a sufficient staff of prompters, it is unsurprising that bailiffs often had plenty of room to 8 9

10

Xinmin was originally designated as a sub-prefecture in 1813, changed to prefectural status in 1902, and finally gained its status as a county in 1913. It needs to be pointed out that even though each serf received an initial allotment of 36 mu of land, subsequent land reclamation increased the total landholding of the average estate to anywhere between 5,400 mu and 16,000 mu of land (Isett 1998, 126, 164). For example, the Office of Accounts in Shengjing and the Jinzhou Imperial Household Department were staffed with only twelve prompters and two prompter captains between

power networks and state-society relations

319

conduct their business relatively free of state supervision. Consequently, complaints were frequently brought against bailiffs who had abused their position to amass wealth and gain local influence (Isett 1998, 135).11 Because of the state’s inability to effectively enforce its authority in this region, throughout the Qing period increasing Han Chinese migration and settlement in Manchuria led to increasing illegal purchases of Manchu land by Chinese peasants (Isett 1998, 44–45).12 These phenomena transformed the social and demographic landscape of the Manchurian region. Administratively, regions were subdivided into small units as soon as there were sufficiently large numbers of people. As early as 1653, Liaoyang and Haicheng counties (both within the newly formed Liaoyang Prefecture) were established as the first county-level administrative units in Manchuria (Edmonds 1985, 113–114).13 By 1877, South Manchuria had a total of 14 counties, signaling the end of this region’s period as China’s frontier territory (Hosie 1904, 163).

Xinmin County in Early Republican China

By the early twentieth century, changes in the socio-political landscape of the Xinmin region led the state to close the gap between formal administration and local society. In 1902, Xinmin’s status as a sub-prefecture (established in 1813) was changed, making it a prefecture (fu 府) of significantly smaller jurisdiction. The area not included in the new prefectural jurisdiction was divided into the first counties ever established in the Xinmin region—the northwestern section of Xinmin now formed Zhangwu County; the southern section of Xinmin (the area west of Raoyang River) formed the newly established Zhen’an

11

12

13

them to supervise 126 imperial grain estates as well as the cotton and indigo estates, fruit orchards, and salt pans (Isett 1998, 119–120). Though Isett’s study takes the entire Manchurian plain as its focus, his descriptions of the socio-political evolution of southern Manchuria are very applicable to the Xinmin region. Xinmin local histories indicate that during the Qing, there were considerable amounts of bannerlands and government estates (hongce di 红册地), though the exact numbers are not listed (Xinmin fuzhi 新民府志 1909, 97). When fraud was discovered by the state, land from illegal exchanges between Chinese and Manchus was confiscated. To get around this, Chinese peasants often created fraudulent leasehold contracts to conceal the illegal transaction (Isett 1998, 144–145). At this point, the exact relationship between the county government (once it was established) and the prompters is unclear. No scholar working on Manchuria has dealt with this issue and county gazetteers have yielded little useful information.

320

hsu

County (in 1914, the name was changed to Heishan County); and the area east of the Raoyang River was designated as Liaozhong County in 1906. Formalized administration within the newly designated Xinmin Prefecture was also established in 1902. A head police bureau was established in Xinmin city (seat of the prefectural government) and six police patrol bureaus (xunjing ju 巡警局) were established throughout the countryside. By 1906, the six police patrol bureaus had been consolidated into five bureaus. Two years later, the bureaus were re-named police wards, with branch stations attached to each ward. In 1910, a new ward was created, bringing the number of wards back up to six (Xinmin xianzhi 新民县志 1992, 54–59). In 1913, the second year of the Republican period, Xinmin’s status was changed to that of a county. Xinmin County now had a total area of about 3,314 square kilometers and a population of roughly 400,000 divided into the six police wards (Xinmin xianzhi 1992, 132). By 1923, state presence within Xinmin County was further strengthened by the division of the entire county into eight administrative wards, with each ward administered by a new ward office (qugong shu 区公暑) (Xinmin xianzhi 1926, 83). And finally, sometime between 1926 and 1929, the six police wards were increased to eight police wards, with most of the police ward offices located in the same village as the administrative ward office (XMDA 1.141).14 The Appointment of Ward Officials In theory, the ward office was to be one of the primary units of local selfgovernment (difang zizhi 地方自治), with its officers nominated by the ward’s local population.15 According to the “Experimental regulations for local selfgovernment” (difang zizhi shixing tiaoli 地方自治试行条例) promulgated on December 29, 1914, local residents of each ward were to present to the county magistrate a list of nominees from which ward managers (qudong 区董) and self-government personnel (zizhi yuan 自治员) would be appointed. Each 14 15

Though it is not explicitly stated in the archival records, presumably the police wards and administrative wards shared the same jurisdiction. The idea of self-government was first formulated during the last years of the Qing. However, by 1911, the local assemblies of the self-government system often directly challenged the authority of the county bureaucracy. As a result, on February 3, 1914, selfgovernment was abolished throughout the country (Hao 1964, 72). However, on December 29 of the same year, the Yuan Shikai government promulgated a new set of regulations on local self-government intended to bring about a greater degree of coordination between the county yamen and local society (Kuhn 1975, 279; Yu 1918, 1.11.6–7).

power networks and state-society relations

321

ward office could have anywhere from six to ten personnel (depending on the size of the ward) and each office would be headed by a ward manager.16 Local residents nominated twice the number of self-government personnel positions open for that ward and three nominees for the ward manager position. From that list, the county magistrate would then appoint the ward’s personnel and ward manager to a two-year term of service. While self-government officers could be renominated for a second term, ward managers were limited to one term (Yu 1918, 1.11.1–4). In this way, the state ensured a limited measure of popular participation while maintaining its overall control of the appointment process. Though on the surface the 1914 self-government regulations might seem like merely a more codified version of the local quasi-officials of the Qing, a closer examination reveals a significant trend that would increase throughout the Republican period: the state’s push to impose a formal level of coordination in sub-county governance. In addition to traditional categories of locallyadministered activity (e.g., education, road maintenance, and promotion of agriculture), the ward manager also assisted the county office with many administrative duties (Kuhn 1975, 279). Ward managers prepared motions and proposals for the county magistrate (ti’an zhunbei 提案准备), implemented resolutions passed by the magistrate (xian zhishi hezhun zhi yijue shixiang 县知事核准之议决事项), and also carried out tasks entrusted by the county magistrate (xian zhishi weituo banli shixiang zhi zhixing 县知事委托办理 事项之执行). Ward managers were also salaried officials and even had the power to hire paid assistants (zuoli yuan 佐理员) (Yu 1918, 1.11.4–6).17 At the outset, Xinmin County’s ward administration did not fulfill the prescriptions of the 1914 self-government regulations. For the most part, the ward office was primarily a unit of the local police, and not self-government. Dossiers of ward policemen in Xinmin County clearly show that these officers 16

17

The 1914 self-government regulations set the theoretical standard for ward administration until the promulgation of the County Organization Laws (xian zuzhi fa 县组织法) on September 15, 1928. Though modifications were made, in essence, the 1928 laws held a vision of ward administration very similar to the one laid out in the 1914 self-government regulations. More substantive changes in the law did not come about until 1933 and 1934 (Zhu 1942, 80–91). Kuhn argues that the codified changes were merely reflections of changes that had already taken place (Kuhn 1975, 284–286). The regulations do not distinguish between ward manager assistants and self-government personnel. Furthermore, the regulations also do not specify the salary rates of ward managers.

322

hsu

were salaried state agents assigned to their posts by the county magistrate (XMDA 1.171).18 By 1923, however, Xinmin County had formally separated police and administrative duties by establishing administrative wards alongside police wards within the county. In accordance with Article 2 of the 1914 self-government regulations, Xinmin County was formally divided into eight administrative wards (Yu 1918, 1.11.1). However, the ward office essentially remained an extension of the county government into local society. Work reports from the ward offices in 1926 reveal that the ward headmen (quzhang 区长) cited for exemplary service were frequently involved in assisting county magistrates with the deliberation (pingyi 评议) of lawsuits, apprehending and arresting bandits, as well as participating in activities such as the promotion of agriculture and education.19 In addition, ward headmen were not nominated by ward residents, but rather appointed directly by the provincial governor.20 The work reports reveal that some ward headmen had held their posts for a period of five years, over twice as long as the two-year term of service stipulated in the 1914 self-government regulations (XMDA 3.3371).

Power Networks in Xinmin County

Low-level state functionaries in Chinese history have long had the image of being lowly figures, mainly preoccupied with receiving self-serving monetary gains through a variety of corrupt practices. Studies of Qing county government are filled with descriptions of the “yamen vermin” (yadu 衙蠹) and their venal ways. The Xinmin County archival records also document with great detail the types of corrupt activities that took place in Republican-era local government. But even more importantly, as the following cases demonstrate,

18

19

20

Policemen were trained at the local police instruction bureau (jingcha jiaolian suo 警察 教练所) for a period of six months, in accordance with the state’s regulations on local police schools (Xinmin xianzhi 1926, 155; Yu 1918, 2.13.14–16). Archival records show that the head of each ward office was always referred to as the ward headman (zhang 长) and not as the ward manager (dong 董). While the term zhang can be applied to both officials (e.g., county magistrates) and non-officials (e.g., village headman), the term manager (dong) never applies to an official, but rather shares connections with the gentry-manager (shendong 绅董) social category that Mary Backus Rankin discusses (1986). County magistrates also had the power to appoint ward headmen, as will be shown with the case involving Zhang Zhixin in the following pages.

power networks and state-society relations

323

the charges of ward official corruption reveal several power networks that counter our existing paradigms of state-society relations in Chinese history. Conflict between Village Headmen On October 20, 1928, seven village headmen, led by main village head Zhang Jingde, lodged a collective plaint accusing fifth ward headman Su Zelin of 10 counts of corruption. Zhang reported that since Su’s appointment in July of 1927, he had never bothered to live within the ward compounds, but rather conducted his operations from his own residence. The plaintiff also claimed that Su had failed to carry out an order to evict a certain villager from a piece of land that was being confiscated by the government and turned into public land (gongdi 公地). Instead, the ward headman allowed the villager to continue living on the land as long as the latter paid him a fee of 500 yuan. The plaint also alleged that Su misappropriated (qintun 侵吞) some 6,000 yuan that had been designated to fix the ward office building. Furthermore, Su was accused of employing personal friends (siren 私人) as ward-level functionaries who then used their powers to smuggle opium (XMDA 2.5130). The accusations brought against ward headman Su Zelin include many elements common to the literature on corruption by low level officials: misappropriation of government funds, extortion, bribe-taking, and hiring personnel based on personal ties. The evidence shows that the Republican state-making efforts to bring about a greater degree of official bureaucratization of local government did not necessarily eradicate the various abuses and informal networks that were common during the Qing Dynasty. As clear-cut as the case may seem from the initial plaint filed by Zhang Jingde, matters became complicated as the county yamen carried out its investigation. As expected, ward headman Su Zelin gave a defense of his conduct and denied all charges of wrongdoing. More surprisingly, all of the depositions taken by the case investigator also supported Su’s claim of innocence. Testimonies given by neighboring village headman Liu Yutian and village resident Peng Jingyun claimed that Su Zelin was an innocent man, not guilty of any of the accusations made against him. Furthermore, the policemen investigating the case also filed four reports on their inability to depose various people. The investigators reported that when they entered the village, village headmen and family members claimed that the people who were to be deposed were out of town visiting relatives. As a result, on February 19, 1929, the provincial governor ruled that ward headman Su Zelin was innocent because the charges of corruption could not be substantiated (XMDA 2.5130). Though the depositions do not add much to our knowledge of the details of the case, they do shed light on the inter-village dynamics of local society.

324

hsu

The plaint against ward headman Su Zelin was filed by a coalition of seven different village headmen who unanimously agreed that the ward headman was guilty of corruption. Far from helpless, village headmen not only confronted wily ward officials by filing plaints, but also enlisted the support of other village heads to present a stronger case to the adjudicating officials. Yet at the same time, Zhang Jingde and his supporters also faced opposition from villages outside of their coalition. At this point, the framework offered by the state-versus-society dichotomy simply breaks down. The point of conflict was not between the state and local society, rather, it was between constituents of local society over the guilt or innocence of a ward official. And while one simply cannot determine the ultimate truth from the evidence available in the case records, it is certainly possible that certain villages supported ward headman Su Zelin’s claim of innocence because they benefited from their connection with the ward headman, while other villages were excluded from these privileges. Village Residents versus Village Headmen Ordinary villagers also voiced their grievances against higher authorities. On January 30, 1925, Liu Yisan (a resident of Xiaohuangdi village) and Li Lingkui (a resident of Gaojiatun village) filed a plaint with the provincial government accusing seventh ward headman Zhang Zhixin and his assistant Xue Yaode of several counts of corruption.21 The main complaint of the petitioners against the ward headman was that he often flouted the orders of the county government and, most importantly, appointed men with bandit connections to ward defense militia (ziwei tuan 自卫团) positions. The plaintiffs alleged that the ward headman was fully aware that his some of his subordinates were in cahoots with bandits. The main charge leveled against Xu Yaode, the ward headman’s assistant, was that he often acted as the judge in civil and criminal cases at the ward level, instead of transmitting them to the county office. For example, in homicide cases, Xue would allow the alleged murderer to settle the case with a payment of a few hundred yuan. He would then force through a private settlement and not report the case to the county office (XMDA 1.469).

21

Allegations were also leveled against defense militia officer (ziwei tuanyuan 自卫团员) Yao Enxi. However, the details of the charges against Yao have been left out of this discussion because the case focused primarily on ward headman Zhang Zhixin and his assistant Xue Yaode.

power networks and state-society relations

325

Similar to the case against ward headman Su Zelin, the case against ward headman Zhang Zhixin and his assistant Xue Yaode was also complicated by divisions at the village level. The original petition was filed by Liu Yisan of Xiaohuangdi village and Li Lingkui of Gaojiatun village and backed up by depositions from five residents of Kuguashan village, two residents of Dahuangdi village, and two other people whose village affiliations were not made known. A few months after the plaintiffs submitted their allegations, from late March to early April of 1925, a second set of depositions were taken and submitted, this time by a total of 11 village headmen who testified that all the charges presented by the plaintiffs were false. In a joint statement, the village headmen singled out the plaintiff Liu Yisan and claimed that he filed charges against Zhang Zhixin and his assistant because of a personal grudge (xiexian 挟嫌). Allegedly, Liu Yisan was attempting to get revenge because the ward headman had detained him in the previous year for obstructing efforts to form a defense militia within the ward. In short, according to the second set of depositions, ward headman Zhang Zhixin was completely innocent. The case record ended with a document dated May 8, 1925, from the provincial police department dropping all charges against ward headman Zhang Zhixin (XMDA 1.469). The case against ward headman Zhang Zhixin, which pitted villagers against their own village headmen, again raises important questions about our existing frameworks for understanding state-society relations in Chinese history. Village headmen have generally been portrayed by historians as protectors of the village, and mediators between the local government and local society. Past scholarship has argued that during the Republican period, when traditional village structures remained intact, villages succeeded in resisting state intrusion and exploitation; when traditional village leadership disintegrated, however, then villages suffered from rapacious local bullies and evil gentry (Huang 1985, 259–291; Duara 1988, 217–243). But as the aforementioned case has shown, rather than protecting the village from the state, the village headmen involved in the dispute instead upheld the claims of the accused state agent. In other words, village headmen did not always align themselves with the interests of the villagers. And even if all 11 village headmen who supported ward headman Zhang Zhixin were merely examples of “local bullies” who had infiltrated the villages after the traditional village leadership had vacated their posts, the response of the villagers still defies our conventional understanding of what happens under such circumstances. In other words, the village did not simply succumb to intrusion and exploitation. In fact, villagers could be quite resilient in pressing their charges to higher levels of state authority, even without the support of their village headmen.

326

hsu

Villagers versus Villagers Another major form of conflict surrounding ward administration involved different networks of villagers. Sometime during 1927, a group of villagers with names withheld (niming 匿名) filed a petition against Zheng Yunge, the third ward’s Xiaotaiping (小太平) village branch police station (fenzhusuo 分驻所) chief (xunzhang 巡长).22 According to the petitioners, Zheng Yunge allowed numerous villagers to set up and operate gambling dens (paijiu baoju 牌九 宝局) from their residences in return for monthly fees. Furthermore, he also allegedly allowed villagers in Taiping (太平) village to conceal weapons and munitions (junhuo 军火) in exchange for fees. According to the investigation carried out by county health office (weisheng gu 卫生股) officer Liu Shengnian, all the charges against the police chief were denied by the villagers. Depositions given by Mafang village headman Su Jingqi and assistant headman (cunfu 村副) Han Guangsheng testified that the residences in question did not double as gambling dens. The investigator also reported that there was nobody in the village by the name of Wang Luoming, whom the original plaint had implicated along with the police chief. Village leaders and residents from Mafang village, Majiatao village, and Zhaojiatao village (16 people total) also submitted a joint deposition that defended the innocence of police chief Zheng Yunge. Finally, a deposition given by a resident of Taiping village testified that Zheng never extorted any money from the villagers in exchange for allowing them to conceal weapons. Interestingly, despite all the depositions taken in this case that claimed police chief Zheng was innocent of all corruption charges, the county police chief (the adjudicating officer in this case) believed otherwise. In the end, on May 21, 1927, the county police chief declared that even though there was no direct evidence, there must be some reason or cause that led to the accusations against Zheng Yunge Thus, the order was given to remove police chief Zheng Yunge from his post (XMDA 1.1910). On the one hand, it is intriguing how charges that lacked any supporting testimony could prevail against a coalition spanning several villages, including village leadership as well as common villagers. On the other hand, it is not difficult to imagine why the county police chief believed that the accusations must have had some basis in reality. If the plaintiffs’ charges were true, then it should not be surprising that the villages implicated would band together to deny all charges of wrongdoing by the police chief. Protecting the police chief also served to protect the village’s interests in continuing their weapon 22

The date of the original plaint cannot be discerned from the archival records. However, the final verdict was given on May 27, 1927.

power networks and state-society relations

327

concealment. Otherwise, the weapons would have to be reported and turned into the county government, thereby depriving the various villagers engaged in this activity of greater armed protection, revenue from black market weapon sales, or some other benefit.23

Sub-County Administration in the Early Republic

Early Republican reformers envisioned the ward administrative unit as a key component in the effort to formalize local administration below the county office. However, as we have seen, in practice the ward office did not function exactly as expected; corruption and malfeasance continued to plague local administration. More importantly, the introduction of the ward office created an arena where members of local society contested with each other in ways that simply cannot be explained by the binary that still dominates much of our understanding of Republican state-society relations. Were the crisscrossing patterns of power networks prevalent in Xinmin County in fact a unique phenomenon, perhaps a reflection of the unusually disorderly nature of the Dongbei region during the warlord period? Did crisscrossing power networks exist, for example, in regions that were more directly under the control of the Guomindang? To answer these questions, the second half of this paper examines the Guomindang’s efforts to establish sub-county administration in post-1935 Sichuan, and its impact on state-society relations. After its move to the interior of China, the Guomindang state continued to expand its reach and power into local society in many different fields, including industry, education, media, and local governance.24 In particular, new developments in the realm of Sichuan’s sub-county administration sought to substantially increase the reach of the state in a region where little formal administration below the county level had existed prior to 1935. Yet, even in this core region of Guomindang control, one 23

24

Phil Billingsley has shown that guns were not only valuable possessions for bandit gangs, they were closely looked after and represented important markers of social distinction within the world of banditry (Billingsley 1988, 110–112). Thus, it is conceivable that the weapons allegedly being concealed by the villagers could yield a handsome profit on the black market. It is also a possibility that the weapons could have been used for village defense. Whereas much of the earlier scholarship in the West has portrayed the Guomindang’s move to the interior as the beginning of the end for an embattled regime doomed for failure, recent scholarship has argued that the expansion of state authority continued in many different fields (see Esherick 2002, 9–14; Liu 2002; Bian 2005).

328

hsu

could still see the existence of power networks that defied the conventional state-society oppositional binary.

Sichuan and the National Government

While warlordism was endemic in practically every region of China during the early Republican years, few provinces experienced the type of intra-provincial civil wars waged in Sichuan between various warlords vying to expand their area of control. During the first few years after the fall of the Qing, Sichuan was at least nominally part of the Republic. Yuan Shikai’s declaration of his emperorship in 1915, however, invoked opposition from several Sichuan regional leaders, who then joined the anti-Yuan rebellion spearheaded by the forces of Cai E in Yunnan Province. According to Kristin Stapleton, the anti-Yuan rebellion and Yuan Shikai’s subsequent death in the summer of 1916 encouraged a process of political disintegration in Sichuan. From 1916 until the arrival of the Guomindang in 1935, no central regime exercised much control over the province. What took place during those years was a series of civil wars between various warlords, each with his own independent government (Stapleton 2000, 188–190, 201–202).25 The arrival of the Fourth Front Red Army in northern Sichuan in late 1932 brought an end to Sichuan’s period of de facto independence. Though the initial arrival of the Communist force led by Zhang Guotao and Xu Xianqian did not attract much notice from the regional warlords (who were embroiled in their own civil wars with one another), the expansion of Communist power in the region gradually led to increased military clashes with the warlords (Kapp 1973, 87–90). Mounting losses and the financial strain of having to sustain military campaigns against both other warlords and the Communists eventually led Liu Xiang, the provincial governor at the time, to ask for military support from Nanjing in late 1934 (Kapp 1973, 92–98; Liu 2002, 98). The arrival of the Staff Corps, the first central government military force to enter Sichuan, signaled the beginning of the Guomindang’s efforts to exert its authority over the province. Through the Staff Corps, the central government maneuvered to weaken the Sichuan military and bring it under central control. Native Sichuan armies were sent to the frontlines of the battle with the 25

Sichuan was officially incorporated into the national government in 1927. However, regardless of the public declaration of affiliation with the newly established Guomindang regime in Nanjing, the warlords continued to exercise de facto independence in many areas.

power networks and state-society relations

329

Communists, while Guomindang troops were stationed in the vital economic and political zones of the province. This tactic relegated the provincial warlords to the peripheries of the provinces and a shrinking sphere of influence. The central government then embarked on a registration program for native Sichuan military regiments in order to control enrollment figures and reduce the size of the local armies (Liu 2002, 99). Upon its arrival in 1935, the central government also initiated a series of reforms designed to revamp Sichuan’s administrative apparatus. The first step taken was the establishment of a new provincial capital in Chongqing on February 2, 1935, along the lines of regulations for provincial government organization promulgated by the central government in July of 1934. Next, the major warlords ceded the territories under their control—also known as garrison areas (fang qu 访区)—along with the right to collect taxes in those areas.26 In an effort to further improve administration, Sichuan was divided into 18 special administrative inspection districts (xingzheng ducha zhuanyuan qu 行政督察专员区), thus establishing a layer of administration between the county and the province. Each district would be staffed by an inspector whose duties included doubling as a magistrate in the county where his office was located, commanding military forces, and, perhaps most importantly, supervising and investigating other magistrates within his jurisdiction (Kapp 1973, 106–111). In essence, the administrative inspection districts were intended to enable the central and provincial governments to monitor county governments more effectively. Changes in Local Administration Another Guomindang change was the establishment of sub-county ward administration. As the previous section of this paper has already shown, though the ward office was first introduced as a unit of local self-government in 1914, in practice it functioned as an office of the bureaucracy. The de facto official status of the ward office was finally changed in 1934, when the Guomindang formally bureaucratized the ward as an administrative level of government (Chang 1936, 246). The Guomindang’s drive to officially—and fully—bureaucratize the ward office was motivated by its quest to annihilate the Communists in the so-called bandit suppression provinces. Regardless of motivation, formalizing the ward office did have the effect of establishing an

26

By 1935, the system of garrison areas had already been greatly weakened as campaigns by Liu Xiang and the Communists weakened the power of the other warlords and reduced their original territories (Kapp 1973, 106).

330

hsu

overall standard of local sub-county administration for the rest of the country (Kuhn 1975, 286–287; Chen 1942, 16; Zhu 1942, 89–90). Though evidence from local gazetteers indicates that the first ward offices were established in Sichuan sometime after the Guomindang’s promulgation of the County Organization Law of 1928 (Xian zuzhi fa 县组织法), the majority of counties did not establish ward offices until after the arrival of the Guomindang in 1935 (Sichuan shengzhi 1992, 71). Zizhong County (资中县), to highlight just one example, was organized in 1935 into four wards. The newly established ward office was staffed by several ward officers (quyuan 区员), an office clerk (shiwuyuan 事务员), an office scribe (lushiyuan 录事员), several ward personnel (quding 区丁), and several patrol officers (xunguan 巡官), all of whom were under the supervision of the ward headman (Zizhong xianzhi 1995, 576–577). Altogether, by the beginning of 1938, 472 ward offices had been established throughout the 134 counties in Sichuan (Wang Chunying 2004, 130). In 1939, one year after the formal establishment of Chongqing as the wartime capital, the Guomindang introduced the New County System (Xinxian zhi 新县治) to further enhance coordination efforts between the county office and lower administrative units. The two biggest changes to local administration under the new system were the incorporation of the ward office into the bureaucratic apparatus of the county government and the establishment of the township-village office (xiangzhen gongsuo 乡镇公所) as the new lowest formal level of administration. Thus, whereas the ward office functioned as a distinct administrative level of government before 1939, it would now function as a supplementary organ (buzhu jiguan 补助机关) to the county office. Under the New County System, ward officials became officials of the county government, even though they continued to perform the same functions and operate out of their ward offices (Zhu 1942, 101).27 Having eliminated the ward’s status as a formal level of administration, the New County System shifted the focus of sub-county administration to the

27

At the same time, magistrates could also dissolve the ward office altogether and replace it with new supervisory wards (zhidao qu 指导区) staffed by supervisory officials (zhidao yuan 指导员). Ward supervisory officials performed essentially the same functions as ward officials; the former operated directly from the county government office. It was believed that the since ward supervisory officials operated directly from the county government office, their physical proximity would therefore enable magistrates to exercise greater supervision of their activities (Zhu 1942, 101; Li 1943, 104; also see Hsu 2007, chapter 7, for actual cases involving ward supervisory officials).

power networks and state-society relations

331

township-village office.28 Of course, the concept of the township-village office was actually first introduced in 1928 under the County Organization Law as part of the Guomindang’s earlier efforts to establish a state presence at the village level (ZMXFD 1934, 281). However, recent research has shown that in the late 1920s and 1930s the township-village office idea was not implemented throughout all of China. When it was implemented, it neither necessarily adhered to the state’s prescriptions nor fulfilled the state’s desire to establish a greater bureaucratic presence at the village level.29 Thus, the emphasis of the township-village office under the New County System can be viewed as the continuation of the Guomindang’s long-standing effort to expand state authority down to the village level. In terms of substantive changes, the New County System expanded the number of personnel at the township-village office and designated each official a set, albeit paltry, salary. From 1940 to 1945, township-village offices throughout Sichuan averaged between 10 and 13 office personnel, including a township-village headman (xiangzhen zhang 乡镇长), an assistant headman (fu xiangzhen zhang 副乡镇长), and several support personnel (zhuli ganshi 助理干事; shiwuyuan 事务员).30 Township-village headmen received a monthly salary ranging from 30 yuan (1940) to 70 yuan (1945), while office support personnel received a monthly salary from as little as 3 yuan to as much as 45 yuan (Cao 2002, 35).31 As early as the end of 1935, the changes to Sichuan administration outlined above caused at least one British observer to comment The rule of the formerly independent Szechwan [Sichuan] warlords has been wholly destroyed, and Central Government troops and officials

28

29

30 31

Villages comprising more than 100 households would be formed into a xiang, while villages with less than 100 households were to be combined to form a xiang. Market towns with more than 100 households would be formed into a zhen (Li 2005, 215). For example, Li Huaiyin’s work has shown that township-village headmen (xiangzhang 乡长) in Huailu County in Hebei not only performed functions not originally prescribed by the law (mainly tax collection), they also served more as representatives of their own communities than agents of the state (Li 2005, 216, 264). Philip Huang has noted that in Shunyi County in Hebei, the township-village was a “nebulous” unit that was not provided for in the county budget and existed mainly on paper (Huang 1985, 287–288). See also Xu and Zhang (2003). 1940–1945 are the years for which data are available. See Xin (2005) for an overview of the implementation of the New County System throughout China.

332

hsu

hold every position of importance, strategic or political, throughout the province (E. R. Leach cited in Kapp 1973, 108–109). In short, in light of the absence of any meaningful central government presence during the warlord period, the changes brought about by the Guomindang after 1935 were an important achievement in the goal to reintegrate Sichuan into the national government. Sichuan during the wartime period, then, represents an important contrast to the Xinmin County case example examined in the first half of this paper. Whereas Xinmin County was located in a region rife with warlord politics during the early Republican period, Sichuan during the wartime period was the core region of Guomindang authority and control. Yet, as the following cases make clear, crisscrossing power networks also existed in the heartland of Guomindang power. Power Networks and the Ward Office On March 29 of 1939, Zhu Liewen (a.k.a. Zhu Hanwen), a villager from an undisclosed village in Neijiang County (内江县) was dragged out of a local teashop and placed in detention by ward headman Tang Jiwen (唐继文) from the second ward (di’er qu 第二区) on suspicion of involvement with bandits. A few days later, ward headman Tang had Zhu executed. Sometime afterwards, the wife of the deceased Zhu Liewen filed a suit against Tang with the Neijiang branch court (difang fayuan 地方法院) and an order was eventually passed down to have Tang brought to court for questioning.32 Apparently, however, the county government ignored the Neijiang branch court’s orders, thus effectively stalling the case. Interestingly, on July 12, around the same time that the widow of Zhu Liewen began appealing to various levels of the government, several Neijiang gentry representatives (shenshi daibiao 绅士代表) from the second ward filed a petition with the provincial government, praising ward headman Tang Jiwen’s great work performance.33 In their petition, these gentry members argued that ward headman Tang was doing an excellent job driving out bandits and criminals from his ward jurisdiction. As a result of his good work, they claimed, peace was finally returning to the area. If given just a little more time, 32

33

The date of the original suit against ward headman Tang is unknown because the file is missing from the archival record. However, details of the original suit are referred to in subsequent petitions against Tang Jiwen. Plaints submitted to the provincial government were usually sent to the provincial governor’s office for final adjudication.

power networks and state-society relations

333

the ward headman could very well finish the task of cleaning up the ward. The petitioners concluded by imploring the provincial government not to pay any attention to the false accusations against the ward headman (SPA 54.9131.c). Up to this point, it is arguably unclear whether there were any power networks at work in this case. After all, regardless of Zhu Liewen’s true status, it is conceivable that ward headman Tang’s gentry supporters were merely coming together to defend an official who they believed was innocent. Perhaps Tang’s supporters were not harboring any hidden agenda or purpose at all. However, on December 25, a third and final party entered the picture and further complicated the case. In contrast to the first group of gentry members, who wholeheartedly vouched for Tang’s innocence, a different group of five gentry members from the second ward now lodged a plaint accusing the ward headman of committing all sorts of evil acts. In particular, the plaintiffs highlighted two matters. First, they alleged that Tang had fined several locals for opium possession but then failed to report those funds or make a clear accounting for how those funds were subsequently used. Second, the plaintiffs accused the ward headman of embezzling money designated for the salary of ward policemen. While accusations of officials embezzling money were quite common during this period, it is the subsequent details of the charge that perhaps provide us with some indication of the plaintiffs’ motivation and interest in lodging the plaint against the ward headman in the first place. Specifically, the plaintiffs argued that upon his arrival in office, ward headman Tang fired many policemen but continued to collect and pocket their salaries for himself. Most importantly, Tang also filled some of the police positions with people from his home in Wusheng County. As a result, only two of the current ward policemen were local residents (SPA 54.9131.c). It is interesting to note that the plaintiffs emphasized the fact that ward headman Tang removed several local residents from their positions as ward policemen and replaced them with outsiders. Undoubtedly, village communities that were fortunate enough to have their own residents placed in key positions within the ward office enjoyed at least some benefits and protection. Conversely, the replacement of these local residents with outsiders at the very least exposed the local community to the potential threat of predatory behavior by officials, who, having no personal ties to the community, probably did not have much vested interest in the well-being of that community. Thus one wonders whether the removal of these local residents from their positions somehow adversely affected the interests of these petitioners and the networks they represented, thereby moving them to take action. In the end, the case concluded rather anticlimactically. In February of 1940, the Sichuan provincial government, the final adjudicating organ in this

334

hsu

case, received notice that ward headman Tang Jiwen had left his office without first securing permission and now could not be located. As a result, Tang was removed ex post facto from his office, a judgment the provincial government repeated in its final statement regarding this case on March 3, 1940 (SPA 54.9131.c). While the archival record cannot provide any definitive answers regarding the motivations and aims of the parties involved, what is clear is the existence of two coalitions of local gentry members (and one widow) who were set against each other in their evaluation of the performance of ward headman Tang Jiwen. Thus, we again confront a scenario that the dominant framework of Republican era state-society relations cannot explain. In other words, instead of presenting a case of clear-cut antagonism between state and society, this case reveals the ways in which the local state could at once generate both support and opposition from different interest groups. The following case, also from Neijiang County, further illustrates how the operation of a local government office could become a site of contention for different societal factions. On August 5, 1938, a group of 18 gentry members from the fourth ward jurisdiction filed a petition with the provincial government supporting the innocence of ward headman Diao Hong (刁宏). The defenders of Diao Hong were responding to a plaint filed sometime earlier accusing the ward headman of crimes related to extortion, operating opium dens, and improprieties in carrying out the conscription of soldiers.34 In their detailed response to the charges raised by the ward headman’s accusers, the gentry members defended Diao’s innocence and argued that the accusations were the work of lowly ruffians attempting to stir up rumors. In response to the charge that ward headman Diao allowed certain people to evade conscription for a fee, the gentry plaintiffs denied that this could be possible. Their argument was simply that draft evasion was illegal. In response to the charge that ward headman Diao forced residents to pay fees for conducting mediations, Diao’s defenders argued that since they had relatives who served regularly as mediators, they knew that none of them collected any illegal fees for their services. The only fee collected was a stamp (yinhua 印花) fee of two jiao (角), which they argued was in keeping with government regulations. Soon afterwards, 13 representatives from different villages responded to ward headman Diao’s defenders by filing a new petition that detailed even more crimes than the original plaint. Though these plaintiffs raised a total of 34

The original plaint against Diao Hong is either missing from the archival records or misplaced somewhere in the file, which runs several hundred pages and contains numerous separate cases.

power networks and state-society relations

335

nine charges against the ward headman, three are especially worth noting. First, these plaintiffs argued that regarding the conscription of soldiers, ward headman Diao colluded with several local baojia heads to allow certain people to evade military service for a payment of 12 yuan.35 However, according to the plaintiffs, while some villages paid Diao the money, their quotas were still unfulfilled. In other villages, soldiers were conscripted but their names were never registered, thereby allowing Diao and his partners to conscript more soldiers than the official quota. Second, in response to Diao’s gentry supporters’ explanation of fees related to mediation services, these plaintiffs reiterated that collecting fees for assisting in mediations was prohibited. Moreover, they continued, since mediation is between two people, there should be no need for a stamp fee. And even if they had to pay a stamp fee, why couldn’t the people just apply the stamps themselves instead of having to give money to officials to apply the stamps for them? Last, the plaintiffs concluded by stating that the ward headman knew that he was not popular with local residents and that it would be difficult for him to hold on to his position for long. Thus, he has sought to make alliances with all sorts of ruffians in hopes to gain their protection. After these initial petitions, the case then proceeded through several more rounds of petitions and investigations by the Sichuan Province second administrative inspection district official (Sichuan sheng di’er xingzheng ducha zhuanyuan 四川省第二行政督察专员), Chen Zhixue. The pattern of one group of plaintiffs refuting allegations raised by the other side repeats throughout the remainder of the case, including one round of arguments and counter-arguments published in two local Neijiang newspapers.36 After all the effort put in by the plaintiffs and investigators, sometime after February 15 of 193937 the Sichuan provincial government issued a final judgment declaring that ward headman Diao Hong would be issued a reprimand (shenjie chufen 申诫处分)—the lowest punishment an official could receive under the “Law 35

36 37

Those who paid would be issued a receipt. Then, a fake name was supposed to be registered in a book and sent to the county government to demonstrate the fulfillment of the conscription quota. Because of the lengthy nature of this case (over 100 pages), the other petitions will not be discussed in this paper. Unfortunately, one cannot discern a clear date from the document containing the final judgment. However, since the final investigation report available in the case record was dated February 15, 1939, one can only presume that the final judgment was handed down soon after.

336

hsu

on Disciplinary Action against Public Servants” (Gongwuyuan chengjiefa 公务 员惩戒法) (SPA 54.9130.e).38 The opposing petitions discussed in this case stand out because of the degree to which the different groups of plaintiffs, beyond just presenting their argument either for or against the ward headman, were actually engaged in an adversarial dialogue with one another. The first petition that supported Diao’s innocence, for example, was actually a response to each of the charges raised by an earlier petition. Then the next petition questioned, among other things, the explanation given by Diao’s supporters over the issue of stamp fees and mediations. This pattern of argument and counter-argument continued for the duration of the case. One could argue that the particularly adversarial posture of the plaintiffs involved strongly indicates that the core conflict in this case was not so much between state and society as it was between different constituents in local society and their relationship to the local state. The petitions also stand out because the plaintiffs themselves indicated that there were broader allegiances at work within the ward jurisdiction. The first group of gentry plaintiffs, for example, were not neutral but rather clearly had some personal interest in defending the ward headman Diao’s innocence. After all, in refuting the allegations of Diao’s involvement with imposing fees for mediations, the plaintiffs were also defending the innocence of many of their relatives, who served as mediators. The second group of plaintiffs alleged that ward headman Diao had his own network of baojia leaders that assisted him in allowing people to evade military service for a fee. Lastly, both groups accused the other of representing a broader alliance of ruffians that sought to either harm or protect the ward headman. To sum up, the highly adversarial posture of the different plaintiffs and the indications of broader allegiances that included both the ward official and many members of society paints a complicated picture of state-society relations that defies simple dichotomization. The relationship between state and society is better characterized by the existence of complex webs of power networks that neither wholly represented the state nor society. Power Networks and the Township-Village Office In 1939, the Guomindang began implementing the New County System throughout the regions under its control to further enhance the state’s reach 38

This punishment had the effect of a warning without any other penalty attached (ZMXFD 1934, 251–252; Kong 1993, 491–494). For a detailed discussion of the efforts of citizens to get redress against abusive officials in Republican Sichuan, see Hsu (2007), chapters six and seven.

power networks and state-society relations

337

into local society. Under this new system, the township-village office represented the lowest point of intersection between state and society. Supporters of the New County System claimed that it represented a dramatic improvement in local governance that would eliminate the shortcomings of previous local administrative schemes (Li 1943, 6–12). What kind of impact did this shift have on the power networks that we saw operating in the previous cases? While much more research remains to be done on the actual operation of townshipvillage administration, available evidence indicates that, like the ward office, township-village offices could also become sites of contention between various power factions. One case from Ziyang County (资阳县), filed in December of 1940 and finally concluded in June of 1944, illustrates just how much the establishment of the township-village office could divide the interests of local society. Certainly the following case does not follow the same pattern as the other cases discussed in this paper. First of all, whereas this paper has focused on complaints against sub-county officials, the following case was lodged against the county magistrate of Ziyang County. Second, this case did not involve contradictory petitions by different groups of plaintiffs. In other words, only petitions lodged against the county magistrate are found in the case record. However, the case is germane to our discussion because disputes over the zoning of the new township-village administrative unit were one of the central issues raised by the plaintiffs. More importantly, even though the plaints did not suggest divisions within local society, the case investigations clearly revealed that the establishment of township-villages in Ziyang County was in fact fraught with conflict and disagreement between different interest groups in society. In December of 1940, 14 baojia headmen from various villages submitted a petition to the provincial governor’s office regarding recent changes in the zoning and establishment of the new township-village office. The plaint, short on details, argued that Ziyang County Magistrate Liu Fang (刘芳) recently made zoning changes to new township-villages that were not reasonable and not in accordance with the desires of the people. The plaintiffs argued that these changes would make their jobs very difficult, but offered no specifics to support their claim. Soon after, another group of plaintiffs submitted a petition against the county magistrate that highlighted their own discontent with the recent zoning of the county’s township-villages. These plaintiffs claimed that the magistrate neither consulted with the local population nor followed legal regulations (faling guiding 法令规定) when it came to determining where township-village offices would be established. Further petitions that included similar complaints were submitted throughout the remainder of the case (SPA 54.9125).

338

hsu

The unanimity of the plaintiffs’ discontent over the zoning of the county’s township-villages might lead one to believe that this was a clear instance of wrongdoing by magistrate Liu Fang. However, the several rounds of investigations that followed the petitions revealed a far more complicated situation at the village level. Two investigation reports, in particular, are worth highlighting. The first investigation, filed by the number-two special administrative inspection district official (di’er xingzheng ducha qu zhuanyuan 第二行政督察区 专员), Chen Zhixue, on December 26, 1940, gave a detailed description of the township-village zoning process in Ziyang County. According to the investigation report, the decision to zone the county into township-village districts had in fact gone through several steps. First, the magistrate convened local leaders, including gentry members, members of local legal associations (fatuan 法团), ward headmen, and village leaders, for two days of discussions. At the conclusion of the discussions, it was decided that the county would be re-zoned to form 29 township-villages. Next, small groups were formed to discuss and determine the locations of the new township-village offices. Finally, a general meeting was convened to finalize the decisions. According to Chen’s report, as a result of the changes, some who had been in positions of power now lost their privileges. For example, the location of the Cheng Shi township-village office was a matter of great contention. Originally, the office was to be located in Cheng Bei Village. However, because some complained that the location was inconvenient, another round of meetings was convened, which resulted in the selection of the Dongguan region as the final location for the township-village office. In his conclusion, the investigator stated that because a minority was still unhappy with the final decision, they decided to lodge a suit against the magistrate. But according to Chen Zhixue’s report, the magistrate was not guilty of any wrongdoing. Just over a year later, on February 7, 1942, another investigation report from the second administrative inspection district revealed that since December of 1940, Ziyang County had undertaken three more rounds of re-zoning at the insistence of the local population. Apparently, after Ziyang County was zoned into 29 township-villages at the insistence of several local gentry, the county was re-zoned to form 39 township-villages. Then, at the insistence of the local population, the county was zoned again, this time into 45 township-villages. Finally, due to inconsistencies in each region’s number of baojia units, the county was zoned at last into 49 township-villages. After nearly four years and over 200 pages of plaints, investigations, and comments, Magistrate Liu Fang was absolved of any wrongdoing related to the zoning of the township-villages. Regarding the other charges, the provincial

power networks and state-society relations

339

government only found the magistrate guilty of committing an administrative error (xingzheng shang zhi guoshi 行政上之过失) related to his handling of the bandit situation in Ziyang County. No specific punishment was passed down (SPA 54.9125). Until now, our understanding of the impact of state-making on local society in Republican China has largely been dominated by explanations that have highlighted either village dissolution or successful resistance to state intrusion. In either case, the relationship between state and society has been characterized by antagonism. Although it is undeniable that the expansion of state authority brought disastrous consequences on some communities while others were better able to thwart or minimize the penetration of the state, neither of these scenarios can adequately explain the disputes that took place over the formation of township-villages in Ziyang County. The case investigation reports strongly suggest that Ziyang County residents did not view the establishment of a new local government office as an unwanted intrusion of the state. Rather, it seems that different communities were in fact competing for the right to have a township-village office established in their vicinity. The first investigation report, for example, claims that due to the re-zoning of the county, some had lost their positions of power, thus causing them to seek ways to regain those positions. Others apparently wanted to have easier access to their township-village office. Though there is no way to confirm the veracity of these statements, the February 1942 investigation report on the numerous zoning efforts seems to support the first investigator’s claims. It is important to note that with each subsequent effort to re-zone the county, the number of township-villages—and hence, townshipvillage offices—increased. One would hardly think that if residents of Ziyang County viewed the presence of a new government office as inimical to their interests, they would call for re-zonings that continually pushed the number of township-village offices upwards from an initial 29 to the eventual 49 offices. Instead, the increasing numbers seem to suggest the people viewed the presence of a township-village office as something that was beneficial. Unfortunately, without further evidence we can only speculate as to why the residents of Ziyang County acted the way they did. Regardless of their motivations, the expansion of state power down to the township-village level in this case did not seem to generate conflict where we have come to most expect it—between state and society. Rather, and perhaps surprisingly, the conflict that arose was between different local communities and interest groups that were contending for a closer association with the state’s administrative apparatus.

340

hsu

Conclusion The introduction of sub-county administration to the Republican state apparatus was designed to foster a greater degree of bureaucratic control, as well as facilitate the transmission and execution of directives from higher levels of the government. The expansion of the state’s formal administrative apparatus sought to bring the reach of the state down to the village level and eliminate the ills that supposedly plagued the informal administration of the Qing county yamen. And while Republican era state-making did produce important changes in local administration and in the relationship between state and society, a closer look at the operational realities of the sub-county administration reveals a far more complicated picture of state-society relations than has previously been recognized. Namely, rather than bringing about a heightened antagonism that resulted in local society’s resistance to or disintegration from increased state intrusion (both of which are well-documented scenarios), the introduction of sub-county administration created an arena where members of local society contested with each other in ways that have not yet been documented in the scholarly literature on Republican state-making and state-society relations. As we have seen, this was true in the warlord region of Dongbei (Xinmin County) during the early Republican period, as well as in Sichuan, the heart of the Guomindang administration after 1935. In short, alliances and power networks were formed in a number of ways, along different lines of interest that defy any simple dichotomization. Thus, the picture of Republican era state-society relations that begins to emerge from this close-up view of the operational realities of sub-county administration is complex, and deserves further research based on archival records. References SPA XMDA ZMXFD

Sichuansheng dang’an guan 四川省档案馆 (Sichuan provincial archives). [Cited by collection number (54) and juan number.] Xinmin dang’an 新民档案 (Xinmin county archives). In Liaoning Provin­ cial Archives. [Cited by catalogue number and juan number.] Zhonghua minguo xianxing fagui daquan 中华民国现行法规大全 (A comprehensive collection of the laws and regulations of the Republican government currently in effect). 1934. Nanjing: Shangwu yinshuguan.

power networks and state-society relations

341

Bian, Morris. 2005. “Building State Structure: Guomindang Institutional Rationalization during the Sino-Japanese War, 1937–1945,” in Modern China, 31.1 (January): 35–71. Billingsley, Phil. 1988. Bandits in Republican China. Stanford, CA: Stanford University Press. Cao Chengjian 曹成建. 2002. “20 shijie 40 niandai Sichuansheng xinxianzhi xia difang zizhi de shixing” 20世纪40 年代四川省新县制下地方自治的施行 (The implementation of local self-rule under the new county system in Sichuan in the 1940s), in Xinan jiaotong daxue xuebao 西南交通大学学报 (Journal of Southwest Jiaotong University), 3.2: 26–38. Chang, C. M. 1936. “A New Government for Rural China: The Political Aspect of Rural Reconstruction,” in Nankai Social and Economic Quarterly, 9.2 (July): 239–295. Chen Boxin 陈柏心. 1942. Zhongguo xianzhi gaizao 中国县制改造 (The reform of China’s county governance). Shanghai: Guomin tushu chubanshe. Ch’ü, T’ung-tsu. 1962. Local Government in China under the Ch’ing. Cambridge, mA: Harvard University Press. Duara, Prasenjit. 1988. Culture, Power, and the State: Rural North China, 1900–1942. Stanford, CA: Stanford University Press. Esherick, Joseph. 2002. “War and Revolution: Chinese Society during the 1940s,” in Twentieth-Century China, 27.1 (November): 1–37. Hao Zhaogong 郝兆巩. 1946. Zhongguo xianzhishi 中国县制史 (A history of county governance in China). Taipei: Hongye shuju. Hosie, Alexander. 1904. Manchuria: Its People, Resources, and Recent History. London: Methuen and Co. Hsiao, Kung-ch’üan. 1960. Rural China: Imperial Control in the Nineteenth Century. Seattle, wA: University of Washington Press. Hsu, Danny. 2007. “Impeachments and Administrative Litigation in Qing and Republican Law.” Ph.D. dissertation, University of California, Los Angeles. Huang, Philip. 1985. The Peasant Economy and Social Change in North China. Stanford, CA: Stanford University Press. Isett, Christopher. 1998. “State, Peasant and Agrarian Change on the Manchurian Frontier, 1644–1940.” Ph.D. dissertation, University of California, Los Angeles. Kapp, Robert A. 1973. Szechwan and the Chinese Republic: Provincial Militarism and Central Power, 1911–1938. New Haven, Ct: Yale University Press. Kong Lingji 孔令纪, [ed.]. 1993. Zhongguo lidai guanzhi 中国历代官制 (The bureaucratic system of China’s past dynasties). Jinan: Jilu shushe. Kuhn, Philip. 1975. “Local Self-Government Under the Republic: Problems of Control, Autonomy, and Mobilization.” In Frederic Wakeman, Jr., and Carolyn Grant, eds., Conflict and Control in Late Imperial China. Berkeley, CA: University of California Press.

342

hsu

Li, Huaiyin. 2005. Village Governance in North China, 1875–1936. Stanford, CA: Stanford University Press. Li Zonghuang 李宗黄. 1943. Xinxianzhi zhi lilun yu shiji 新县制之理论与实际 (The theory and practice of the new county system). Shanghai: Zhonghua shuju. Liu, Lu. 2002. “A Whole Nation Walking: The ‘Great Retreat’ in the War of Resistance, 1937–1945.” Ph.D. dissertation, University of California, San Diego. Reed, Bradly W. 2000. Talons and Teeth: County Clerks and Runners in the Qing Dynasty. Stanford, CA: Stanford University Press. Sichuan shengzhi 四川省志 (Annals of Sichuan province). 1992. Chengdu: Sichuan renmin chubanshe. Stapleton, Kristin. 2000. Civilizing Chengdu: Chinese Urban Reform, 1895–1937. Cambridge, mA: Harvard University Press. Wang Chunying 王春英. 2004. “Shilun minguo zhengfu jiceng zuzhi,” 试论民国政 府基层组织 (Exploratory essay on Republican local government organization), in Sichuan daxue xuebao 四川大学学报 (Journal of Sichuan University), no. 16: 128–135. Xin Zhi 忻知. 2005. “Gesheng shishi xinxianzhi tuixing difang zizhi chengji zong jiantao,” 各省实施新县制推行地方自治成绩总检讨 (A critical examination of the implementation of local self-rule of each province under the new county system), in Minguo dang’an 民国档案 (Republican archives), no. 3: 39–45. Xinmin fuzhi 新民府志 (Xinmin prefectural gazetteer). 1909. Taipei: Chengwen chubanshe. Xinmin xianzhi 新民县志 (Xinmin county gazetteer). 1926. Taipei: Chengwen chubanshe. ———. 1992. Shenyang: Shenyang chubanshe. Xu Xuyang 徐旭阳 and Zhang Taishan 张泰山. 2003. “Kangri zhanzheng shiqi Hubei xinxianzhi de tuixing yu jiceng zhengquan biange,” 抗日战争时期湖北新县制的 推行与基层政权变革 (The practice of the new county system and the reform of grassroots government in Hubei province during the anti-Japanese war period), in Huazhong keji daxue xuebao, shehui kexue ban 华中科技大学学报社会科学版 (The journal of Huazhong University of Science and Technology, social science edition), no. 5: 115–124. Yu Baoxuan 于宝轩 [ed.]. 1918. Neiwu faling ligui jilan 内务法令例规辑览 (Compilation of the laws and regulations of the ministry of internal affairs). [Cited by volume, category, page number]. Zhu Zishuang 朱子爽. 1942. Zhongguo xianzhi shigang 中国县制史纲 (A historical outline of governance in China). Chongqing: Zhongguo wenhua fuwushe. Zizhong xianzhi 资中县志 (Zizhong county gazetteer). 1995. Chengdu: Sichuan cishu chubanshe.

part Four

Concubinage · Spousal Abuse · Transnational Families

∵

chapter 11

Ceremony and the Definition of Marriage under Republican Law Lisa Tran Ceremony has always been a factor in distinguishing marriage from other forms of unions. In early Republican China, the ceremony was just one of several rituals jurists considered in determining whether a union constituted marriage. In contrast, the Guomindang (GMD) civil code of 1929–1930 made the ceremony the exclusive criterion for ascertaining whether a union constituted legal marriage. The law left open the particular protocols to be followed, only requiring that the ceremony be “open” and witnessed by a minimum of two persons.1 However, in the early twentieth century, the social rituals associated with the acquisition of wives and that of concubines occasionally overlapped. In light of the courts’ liberal interpretation of what constituted a legal ceremony, the unintended consequence of the ceremony-based definition of marriage was that some concubines would gain legal recognition as a wife, and concubinage would sometimes be prosecuted as bigamy. This would have been impossible in the early Republic, but became very possible after the implementation of the civil code in 1929–1930. The Qing distinction between main and minor wife had enabled late imperial jurists to distinguish between legal marriages and semi-marital arrangements like concubinage. The Daliyuan, which functioned as the highest court in the early Republic, preserved the Qing legal space for concubinage, enabling it to continue to shelter the custom of concubinage from the disciplinary arm of the law. Like their predecessors, early Republican jurists viewed a man’s relationship with his concubine as an informal—but never legal—marriage. As long as jurists reserved a space within the law for semi-marital arrangements without granting them legal validity as full-fledged marriages, concubinage could continue to escape the laws on bigamy. 1 Shortly after the enactment of the civil code, the GMD state introduced new guidelines for wedding ceremonies, intended to be supplements to Article 982. However, lawmakers agreed that it was premature at this point to require that all wedding ceremonies conform to their guidelines. For the time being, they opted to simply adhere to the letter of the law and leave the particulars to individual preference.

© koninklijke brill nv, leiden, ���4 | doi 10.1163/9789004271890_013

346

Tran

That space disappeared when GMD lawmakers omitted any mention of concubinage in the new civil code. Concubinage no longer legally existed, and a woman who was a concubine was by definition not a wife; GMD law made no room for an intermediate legal category. That some concubines were recognized by the court as legal wives owed itself to the confluence of three developments in the early twentieth century: the privileging of the ceremony requirement by jurists operating under the civil code, the simplicity of the legal standards for a ceremony under Article 982, and the widespread practice of holding a ceremony when a concubine entered the household.

Ceremony in Social Practice

According to the Confucian classics, most notably the Book of Rites (Li ji), Etiquette and Ritual (Yi li), and Discourses in the White Tiger Hall (Baihu tong), a number of rituals collectively known as the Six Rites had to be followed in the proper order for a couple to be considered married. The Confucian texts stipulated specific protocols to be followed in each stage, and only careful adherence to the guidelines made a union a marriage. Although the Six Rites were not observed in the acquisition of a concubine, some type of ceremonial ritual was performed to mark her entry into the household. By the Qing, a concubine was considered to be a minor wife, and the ceremony acknowledged her semimarital status. However, the performance of a ceremony in and of itself did not constitute a full legal marriage; for that, completion of the Six Rites was necessary. The Ceremony for a Main Wife Since the Han (206 B.C.–220 A.D.), the Six Rites had dictated the rituals to be followed in betrothal and marriage.2 Although there were regional variations, it was generally acknowledged that the Six Rites included the following events.3 In the first rite (nacai), the prospective groom’s family made an offer of marriage by dispatching a gift-bearing matchmaker to the woman’s parents; acceptance of the gift constituted agreement to marry. The second rite (wenming) involved acquiring the woman’s full name and the year, month, day, and hour of her birth. In the third rite (naji), the couple’s horoscopes were compared to 2 Only marriage customs among the Han Chinese are discussed in this chapter; marriage rites among the ethnic minorities of China varied widely and are not reviewed here. 3 Ruan Changrui, Zhongguo hunyin xisu zhi yanjiu (Research on Chinese marriage customs) (Taibei: Taiwan shengli bowuguan chubanbu, 1989): 20–32; Ma Zhisu, Zhongguo de hunsu (Marriage customs of China) (Taibei: Jingshi shuju, 1981): 7–16.

Ceremony and the Definition of Marriage under Republican Law

347

predict compatibility or conflict. If all was satisfactory, the man’s family paid the woman’s parents a previously agreed upon sum; the presentation of this “betrothal gift” (pincai) completed the fourth rite (nazheng) and sealed the betrothal. The setting of a wedding date concluded the fifth rite (qingqi), and in the sixth (qinying), the bride was escorted to and welcomed into her new home in a bridal sedan chair (huajiao) accompanied by a troupe of musicians.4 By the Song, the Six Rites had been condensed into three: the selection of the bride (caize),5 which merged the first two rites; the payment that concluded the betrothal (nabi),6 which combined the next three rites; and the welcoming of the bride into her new home (qinying), the only one of the original Six Rites to remain intact. Although now reduced to half their number, the marriage rites were still collectively referred to as the Six Rites.7 Patricia Ebrey’s and Susan Mann’s research on marriage customs in late imperial China indicate that the Six Rites were in fact observed in practice, at least among the elite. Drawing from two manuals that describe social customs in Kaifeng and Hangzhou during the Northern and Southern Song, respectively, Ebrey discusses in great detail the rituals involved in the selection of the bride, the exchange of gifts that marked the betrothal and the celebrations on the wedding day.8 Besides completion of the betrothal and wedding rites, a banquet for relatives and friends was usually held after the ceremony to celebrate the union, and a marriage document (hunshu) was drawn up to serve as a written record.9 Mann’s discussion of discourses on marriage from the mid4 For a detailed discussion of the Six Rites, see Xu Chaoyang, Zhongguo qinshufa suyuan (The origins of China’s family law) (Shanghai: Shangwu yinshuguan, 1934): 112–118; Vermier Y. Chiu, Marriage Laws and Customs of China (Hong Kong: The Chinese University of Hong Kong, 1966): 7–10. 5 This first abbreviated rite is also referred to as nacai (Feng Shaoli and Chen Guohui, eds., Jiushi hunsu (Traditional-style marriage customs) (Jiulong: Jinhui chubanshe, 1991): 6. 6 This second abbreviated rite is also referred to as nazheng (Feng and Chen 1991: 6). 7 Chiu 1966: 4–7; Feng and Chen 1991: 6. 8 Patricia Buckley Ebrey, The Inner Quarters: Marriage and the Lives of Chinese Women in the Sung Period (Berkeley: University of California Press, 1993): 82–96. 9 The marriage document was not officially issued by a government agency but instead privately drawn up by the matchmaker and the parties involved. Blank copies of the marriage document could be readily purchased from a bookstore (Chiu 1966: 12). The form contained spaces for the names of the bride and groom, the persons who introduced the couple ( jie­ shaoren), the persons who presided over the ceremony (zhuhunren), and the witnesses (zhengmingren); often, the names would be filled in beforehand. Completion of the marriage document, along with the giving of betrothal gifts, sealed the engagement (Xu Sida, Lihun falun (Legal treatise on divorce) (Tianjin: Yishi baoguan, 1932): 122). On the actual day of the wedding, each person listed on the form would affix his or her seal to the document.

348

Tran

Qing notes the renewed interest in the Confucian classics and the emphasis on the marriage rituals prescribed in the texts.10 Although Mann focuses more on the discourse than the practice of marriage rites, her study does suggest that the greater attention mid-Qing intellectuals paid to the Six Rites may have been reflected in social practice as well. Studies by Chinese scholars reveal the continuing influence of the Six Rites in shaping marriage practices in contemporary China. Feng Shaoli and Chen Guohui’s survey of popular marriage customs in twentieth-century Guangzhou and Jiangsu shows that betrothal and wedding rituals continued to closely follow the Six Rites, albeit with some modifications and additions. The most apparent change was the renaming of the abbreviated Six Rites; the three rites to which the original Six Rites had been reduced in the Song were now referred to as tiqin, dingqin, and jiehun.11 Despite the name change, however, the marriage rituals in Guangzhou and Jiangsu generally retained the form and order of the Six Rites. Broader in geographic scope, Ma Zhisu’s study of marriage customs in the early twentieth century covers most of the provinces in China as well as Taiwan; Beijing, Tianjin, and Nanjing are treated in separate ­chapters.12 Although Ma’s detailed study reveals a dizzying array of marriage rituals, they all reflect the influence of the Six Rites; society may have elaborated the betrothal and wedding rituals, but they retained the Six Rites as the core. The Ceremony for a Minor Wife The Six Rites were to be observed only in the acquisition of a main wife; they did not apply in the case of a concubine. The Book of Rites differentiated between main wife (qi) and concubine (qie) on the basis of adherence to the complex rituals associated with betrothal and wedding; any woman who was married without observing the Six Rites was considered a concubine by default.13 While the Six Rites were not performed in the acquisition of a concubine, a few ceremonial rituals did mark a concubine’s entry into her new household. In most cases, a matchmaker was engaged to find a suitable candidate, although a concubine could be acquired without the services of one. Usually a contract—whether in the form of a marriage document or a bill of sale—was drawn up to formalize the relationship. As with the entry of a main wife, an 10

11 12 13

Susan Mann, “Grooming a Daughter for Marriage,” in Rubie S. Watson and Patricia Buckley Ebrey, eds., Marriage and Inequality in Chinese Society (Berkeley: University of California Press, 1991): 204–229. Feng and Chen 1991: 6–15. Ma 1981. James Legge, tr., Li Chi: Book of Rites (New Hyde Park, NY: University Books, 1967) vol. 1: 479.

Ceremony and the Definition of Marriage under Republican Law

349

auspicious day was selected for the concubine’s arrival. On the agreed upon date, a small green sedan chair was dispatched to the concubine’s home to transport her to her new residence. When she arrived, she was to enter through the side door, never the main entrance. She did not perform the ceremonial obeisances or pay her respects to the ancestral tablet with her new master; the main ceremonial act a concubine performed involved a kowtow to the main wife and any other concubines already in the household. No banquet was held to celebrate the occasion.14 Whereas the wedding rites were intended to commemorate the main wife’s incorporation into her new husband’s kinship networks, the purpose of the rituals marking a concubine’s arrival was to underscore and reaffirm the new member’s subordinate position in her master’s household. Anecdotal evidence indicates that some concubines continued to enter in this simple and subdued manner in the Republic. In her memoir, Su-Hua Ling describes the entrance of “Sixth Mother”15 as a relatively quiet affair: Now the hall was filled with people—Father, Mother, Aunt, and the other two mothers. The new woman gently knelt down before the ancestor’s shrine. She rose and knelt three times. After worshipping the ancestors, the new woman began to pay respect to Father and Aunt. . . . Then she went to pay respect to Mother and Third and Fifth Mothers.16 Absent are many of the rituals characteristic of a main wife’s entrance into the household. From all appearances, Sixth Wife arrived at her new household of her own accord; her master apparently had not made any arrangements for a wedding sedan to escort her to his home. Although Sixth Wife bowed before the ancestral shrine, her master did not perform the ritual with her, as he would have done with a main wife. What little ceremony there was involved the newest concubine kowtowing to her superiors, including the main wife and all the concubines who had preceded her. Firecrackers were lit to welcome the Sixth Wife, but as Ling explains, their purpose was “to drive away the evil spirits which came with her.”17 14 15

16 17

Feng and Chen 1991: 30–31. Children often referred to their birth mother as “Mother” and the other minor wives as Second Mother, Third Mother, etc. depending on the order each woman entered the household. Su-hua Ling [Shuhua Ling], Ancient Melodies (London: Hogarth Press, 1953): 57. Ling 1953: 56.

350

Tran

In addition, in lieu of a marriage document, a written pledge often stipulated the conditions under which the concubine entered her new home. And where the money exchanged for a main wife was cloaked as “betrothal gifts,” no such social niceties existed in the purchase of a concubine; the economic nature of the relationship was sometimes blatantly displayed by a bill of sale. As Vermier Chiu colorfully puts it: After the terms of the transaction are agreed upon by the parties, payment is made in full, and a date is fixed for “delivery” of the woman to the purchaser. In this connection, it may be said that the whole transaction is almost analogous to that of buying an article of merchandise from a shop, except that in the purchase of concubines payment in full is invariably made before delivery—no C.O.D. (cash or credit, on delivery) and no payment by installments.18 As the Six Rites concealed the financial aspects of marriage, their absence in the acquisition of a concubine lay bare the economic nature of the transfer of a concubine from her natal home to her master’s household. A “Main Wife” Ceremony for a Concubine By the twentieth century, the ceremonial rituals associated with the acquisition of a main wife and those associated with concubinage occasionally overlapped in social practice. In Wild Swans, Jung Chang describes in great detail the two wedding ceremonies of her grandmother, Yufang. Although Yufang was a concubine in her first marriage (to General Xue) and a main wife in her second marriage (to Dr. Xia), the ceremonies were remarkably similar. Indeed, Chang’s retelling of her grandmother’s two wedding days indicates that in the early twentieth century, it was not unusual for a concubine to go through the same elaborate ceremony as a main wife. In the eyes of society, the ceremony, whether for a main wife or a concubine, was a man’s opportunity to display his wealth and power. In the case of a concubine, it was also an expression of his magnanimity, as a lavish ceremony enabled the concubine’s family to save face after essentially selling their daughter. Chang mentions General Xue’s “promise to solemnize the liaison with a full wedding ceremony” to give her grandmother and great-grandparents “a considerable amount of face.”19 Chang’s account of her grandmother’s marriage as a concubine to General Xue indi-

18 19

Chiu 1966: 24. Jung Chang, Wild Swans: Three Daughters of China (New York: Anchor Books, 1991): 31.

Ceremony and the Definition of Marriage under Republican Law

351

cates that by the twentieth century, the ceremony did not serve as a clear mark of distinction between main wife and concubine. From the outset, General Xue made clear his intent to take Yufang as a concubine. As Chang explained, her grandmother’s father, who had schemed to bring the couple together, expected no more. Much to the father’s delight, the general sent betrothal gifts even though his union with Yufang was not a formal marriage. On the selected day, he dispatched to the Yang household eight men carrying a sedan chair bedecked with red embroidered silk and satin. Accompanied by a procession of people carrying plaques, banners, and lanterns decorated with images of a golden phoenix, considered the highest symbol for a woman, Yufang “was taken right around the town, visiting all four gates, as a full ritual demanded, with her expensive wedding gifts displayed on carts and in large wicker baskets carried behind her. . . .The pomp and ceremony made her feel she had gained prestige and esteem.”20 Later that evening, in the glow of red lanterns and with the musical clamor of drums, cymbals, and wind instruments in the background, a wedding ceremony was held according to local custom. General Xue and Yufang “performed a ceremonial kowtow to the tablets of Heaven and Earth. After this, they kowtowed to each other, then my grandmother went into the wedding chamber alone, in accordance with the custom, while General Xue went off to a lavish banquet with the men.”21 From all appearances, General Xue married Yufang in a ceremony befitting a main wife; it certainly bore no resemblance to the simple, quiet affair customarily associated with the acquisition of a concubine. As Chang explains, “This was very different from what a mere concubine would get—a small sedan chair draped in plain cotton of the unglamorous color of indigo, borne by two or at the most four people, and no procession or music.”22 Despite the lavish ceremony, however, Yufang’s social status was indisputably that of concubine. The ceremony could disguise the fact that Yufang was being married as a concubine, but it could not, at least in the eyes of society, change that fact. The marriage of Yufang as a concubine in a wedding ceremony remarkably similar to that held for a main wife does not appear to be an isolated incident. Ethnographic field surveys conducted by the Japanese South Manchurian Railway Company (Minami Manshū Tetsudō Kabushiki Kaisha, hereafter Mantetsu) in the Chinese countryside in the early 1940s provide invaluable glimpses of customary practices and daily life at the village level in the late

20 21 22

Chang 1991: 31. Chang 1991: 32. Chang 1991: 31.

352

Tran

Republican era.23 According to the interviews, a concubine’s entry into the household was often indistinguishable from that of a main wife. Villagers from Shajing in northeastern Hebei, for instance, explained that a matchmaker was employed in the acquisition of a concubine, suggesting that the procedures involved in the taking of a concubine continued to mirror those for a main wife. According to the Shajing villagers, “a ceremony is held for when the concubine enters her master’s household; it is the same as a marriage ceremony.”24 Other villagers confirmed that the ceremony for a concubine was “the same as that for a main wife.”25 The Mantetsu interviews suggest that at least in terms of ceremonial rites, a concubine shared much in common with a main wife. That did not mean, however, that a concubine enjoyed the same social position as a main wife. Despite the semi-marital nature of concubinage, there was no confusion about who filled what role. As Xu Chaoyang explained, when people referred to a man having two wives, what they really meant was that he had a main wife and a minor wife.26 Consequently, as much as the ceremonial rites for taking a main wife and a concubine sometimes overlapped, the social distinction between the two remained rigidly in place. A concubine could be married in accordance with the Six Rites, but in the eyes of society, she remained a concubine. By the Republic, the Book of Rites’ distinction between main wife and concubine based on observance of the Six Rites no longer held; in the eyes of society, what mattered now was the intent of the man, not the manner in which the woman had been acquired. That the ceremonial rituals involved in the acquisition of main wife and concubine grew increasingly similar by the Republic, then, should not be mistaken for any blurring of the very rigid social distinction between the two; whether in the Qing or the Republic, that line was always strictly observed. Indeed, the differences between main wife and concubine manifested themselves in a multitude of ways—from mundane details like where one could sit, to more significant matters such as claims to one’s own children. As Kathryn Bernhardt points out, it was not until after the implementation of the civil 23

24 25 26

The Mantetsu interviews consulted here were compiled from November 1940 to December 1942. For a discussion of the Mantetsu sources, see Philip C. C. Huang, The Peasant Economy and Social Change in North China (Stanford: Stanford University Press, 1985): 34–36. Niida Noboru, ed., Chūgoku nōson kankō chōsa (Investigations of customary practices in rural China) (Tokyo: Iwanami, 1952–1958) vol. 1: 250. Niida, vol. 1: 275. Xu 1934: 93.

Ceremony and the Definition of Marriage under Republican Law

353

code that a concubine finally gained legal recognition as the mother of her own children.27 The anecdotal evidence presented above highlights the paradoxical nature of Republican social attitudes toward the concubine. These examples illustrate on one hand the inflexible social distinction between main wife and concubine, and on the other hand, the similarities in the ceremonial rituals involved in the marrying of a main wife and a concubine. Early Republican law, which generally followed Qing practice, respected the line between the two roles. The legal distinction only became muddled when GMD lawmakers made the performance of an open ceremony the sole criterion for legal recognition as marriage in the civil code.

Ceremony in Early Republican Law

Although the Six Rites continued to influence marriage customs in the twentieth century, Republican lawmakers were reluctant to mandate them as legal criteria for marriage, particularly in light of the May Fourth critique of Confucianism and the increasing popularity of Western-style wedding ceremonies among the urban elite in cosmopolitan centers like Shanghai. Zhang Shen remarked in his treatise on China’s marriage laws that Republican lawmakers discouraged the use of “the traditional style of the feudal era” (fengjian shidai jiushi) in marriage and funeral ceremonies.28 While jurists were willing enough to recognize the legal validity of marriages that had been completed in accordance with the traditional marriage rites, they refused to specify the Six Rites as legal criteria for marriage. In form and spirit, early Republican law followed Qing precedent more than it heralded the GMD civil code when it came to defining the legal criteria for marriage. While early Republican jurists did not mandate observance of the Six Rites, the procedures they deemed valid mirrored the Confucian set of rituals for betrothal and marriage, which appeared in one form or another in Daliyuan explications of the legal requirements for marriage. Customs identified in Daliyuan rulings included the following: securing the services of a matchmaker, distributing wedding invitations, sending betrothal wealth to the bride’s family, completing a marriage document to officially seal the engagement and holding an appropriate form of ceremony. 27 Kathryn Bernhardt, Women and Property in China, 960–1949 (Stanford: Stanford University Press, 1999): 189. 28 Zhang Shen, Zhongguo hunyinfa zonglun (Summary of views on Chinese marriage law) (Shanghai: Shangwu yinshuguan, 1936): 78.

354

Tran

The Daliyuan shared the Qing assumption that betrothal was a prerequisite for marriage, and that the document formalizing that betrothal was a legally binding contract. In a series of rulings, the Daliyuan clarified two preconditions for an official betrothal. First, there had to be a betrothal document drawn up by a matchmaker that would then either be published in the local newspaper or privately agreed upon by all interested parties. Second, betrothal gifts had to be presented and accepted. Satisfaction of these two requirements made a betrothal legally valid and binding.29 As under Qing law, a betrothal contract could not be broken unless both parties agreed to its dissolution or there were legal grounds to do so.30 As rulings by the Daliyuan suggest, only some of the requirements for betrothal and marriage needed to be met, and the manner in which they were satisfied was left up to local custom. The Daliyuan clarified that while both betrothal wealth (pincai) and a marriage contract (hunshu) were requirements for legal marriage, satisfaction of one of the criteria was sufficient to make the union a valid marriage.31 The Daliyuan even accepted wedding invitations (huntie) as evidence that a betrothal had been concluded.32 Betrothal wealth could be in the form of cash payment or its equivalent in the form of material goods.33 As for the marriage document, judges were to rely on their own sense of what constituted a contract by local standards.34 Although the Daliyuan provided examples of acceptable documentation, it concluded that, in the absence of legal regulations, local custom would prevail. A betrothal, however, held no legal force as a marriage until the completion of an actual wedding ceremony. In this regard, early Republican jurists departed from Qing practice. The Daliyuan did not consider betrothal to be tantamount to marriage, as indicated by its rulings that corrected such an assumption made by some local officials. A magistrate from Wuding in Yunnan province was about to convict a man of bigamy based on the fact that he had been previously betrothed to another woman. The Yunnan Superior Court, however, pointed out that since the man never married the first woman, he did not com-

29

Guo Wei, ed., Daliyuan panjueli quanshu (Complete collection of Daliyuan judgments on important cases). [1933] Reprint. (Taibei: Chengwen chubanshe, 1972): 212, 214, 217. 30 Guo 1933: 214–218. 31 Shanghai diyi tequ difang fayuan (Shanghai First Special District Court). Shanghai Municipal Archives, Shanghai. [Hereafter SMA] Y5-1-38 (vol. 4, ch. 3): 31. 32 SMA Y5-1-38 (vol. 4, ch. 3): 31. 33 Ibid. 34 Guo 1933: 213; see also Guo 1933: 212, 218.

Ceremony and the Definition of Marriage under Republican Law

355

mit bigamy when he married the second woman.35 In another case, the Daliyuan explained that for a marriage to be legally recognized, there had to be a day when an appropriate ceremony was held (juxing xiangdang lishi zhi ri). The ceremony was to be performed according to local custom; it did not matter if the ceremony followed traditional or new rites.36 Any union failing to meet this requirement was simply not a formal marriage (zhengshi hunyin) in the eyes of the law.37 In 1920, the Department of Justice (Sifabu) reaffirmed the contract and gift requirements for betrothal and the ceremony requirement for marriage. However, it refrained from stipulating the particular ways or forms in which these requirements should be met, deferring instead to local custom.38 While the Daliyuan was flexible about how the legal criteria for marriage would be interpreted, it held firm that certain parts of the criteria had to be met in order for a union to be legally recognized as marriage. That was the central question in a homicide case heard by the Daliyuan in 1921. The defendant appealing his conviction had fatally stabbed a woman who was living with his father. Before the Daliyuan would consider the appeal, it wanted to know whether this crime fell under the provisions in the criminal code that imposed a stiffer penalty for homicides committed against an ascendant. To answer that question, the legal nature of the relationship between the defendant’s father and the victim had to be determined. If the victim had merely been cohabiting with his father, then the case would be adjudicated according to the standard laws against homicide. However, if she had been married to his father, then the case would be decided on the basis of the law regarding the murder of an ascendant. The Daliyuan returned the case to the lower court with instructions to determine the legal nature of the victim’s relationship with the defendant’s father by ascertaining whether the legal requirements for marriage had been met in their union. As with the rulings discussed above, the Daliyuan emphasized the rites associated with betrothal: “Whether or not the victim holds the status of wife vis-à-vis the appellant’s father should be determined by whether or not the legal requirements were met during the 35 36 37

38

Marc van der Valk, Interpretations of the Supreme Court at Peking: Years 1915 and 1916 (Taipei: Ch’eng-wen Publishing Co., 1968): 351–354. Zheng Yuanzou, Zhu Hongda, and Shao Zumin, ed., Xingfa jijie (Collected interpretations of the criminal code) (Shanghai: Shijie shuju, 1932): 426. Xingfa shiyong fenze (Application of the specific provisions of the criminal code) (Publication of Sifa xingzhengbu faguan xunliansuo [Bureau of Judicial Administration training institute for judges], n.d.): 124–125. Guo Wei, ed. Daliyuan jieshili quanwen (Complete texts of Daliyuan interpretations) (Shanghai: Shanghai faxue bianyishe, 1931): 797–798.

356

Tran

engagement period.”39 To guide the lower court, the Daliyuan listed certain actions­—the writing up of the marriage document formalizing the betrothal and the receipt of betrothal gifts—that would serve as evidence that a legally valid marriage had occurred. While the protocols identified by the Daliyuan may not have been followed in cases of simple cohabitation, they generally were followed in rituals of concubinage. In social practice, many of the rituals associated with a legal marriage were also performed when a concubine was married, making it difficult to discern whether a woman was being married as wife or concubine. In a report filed with the Beijing police department, Mrs. Yu neé De accused her son-in-law of relegating her daughter to the status of concubine.40 To support her statement that her daughter had been married as a wife, Mrs. Yu pointed to the role of a matchmaker and the use of a bridal sedan chair. After the marriage, however, Mrs. Yu learned that the woman who she had thought was her son-in-law’s elder sister was in fact his wife and that her daughter had been reduced to the status of concubine. Although Mrs. Yu insisted that her daughter had been married as a wife, the rituals she described could just as easily have been performed in the acquisition of a concubine. Another case filed with the Beijing police station in 1918 tells a similar story.41 In this case, it was the woman’s elder brother and sister-in-law, Zhao Sande and Mrs. Zhao neé Mou, who came to her rescue, supporting her claim that she had been married as a wife. The young woman’s alleged husband, Fan Yibo, even acknowledged in his statement to the police that wedding invitations had been sent and a horse-drawn carriage had been deployed to convey the bride to his home. As with the previous case, there was clear evidence that some type of ritual had been followed. However, whether those rituals meant that the woman was being married as wife or concubine was not as clear. The deference to local custom certainly blurred the line further between marriage and concubinage, but the Daliyuan retained its authority to be the final arbiter on any question concerning the legal validity of regional adaptations of the requirements for marriage. Further complicating matters were the different types of relationships that society considered marriages even though they did not normally go through the same procedures as conventional marriages. Jurists now had to decide whether such unions met the legal requirements for marriage. Early Republican law’s treatment of concubinage and uxorilocal marriages, in which the 39 40 41

SMA Y5-1-38b (vol. 1, ch. 17): 27. Beijing difang fayuan (Beijing District Court). Beijing Municipal Archives, Beijing. [Hereafter BMA] J181-19-13717. BMA J181-19-19200.

Ceremony and the Definition of Marriage under Republican Law

357

husband assumed residence in his wife’s household and carried on her father’s patriline, illustrates the arbitrary nature of the law in accepting or rejecting certain unions as valid marriages. In a case in 1919, the Daliyuan held that since the agreement made for a man to marry into and live in the household of his wife established marital relations, there was no need to write up a separate marriage document.42 Although different in form, uxorilocal marriages were in spirit and purpose similar enough to conventional marriages for the Daliyuan to consider them interchangeable. In another ruling, the Daliyuan added that even in such marriages in which the husband moved into his wife’s household, betrothal gifts still needed to be given.43 Failure to deliver the betrothal wealth by the agreed upon time, the Daliyuan ruled in 1919, constituted a breach of contract and could be used as grounds for nullifying the marriage contract.44 Yet the presentation of betrothal gifts in and of itself, the Daliyuan qualified, did not make the subsequent union a legal marriage. In a 1916 case from Anhui province, the Daliyuan dismissed the lower court’s assumption that bigamy had been committed in a case involving a concubine, deeming evidence that betrothal gifts had been exchanged as irrelevant to the matter. In this case, a man had, in his son’s absence, called upon a matchmaker to arrange for his daughter-in-law to be another man’s concubine. Betrothal gifts were given, and a marriage agreement completed. Evidence that the rituals associated with betrothal and marriage had been fulfilled prompted the Anhui Superior Court to wonder if the case should be tried under the laws against bigamy even though the married daughter-in-law was technically a concubine in her union with the second man. Responding in the negative, the Daliyuan categorized the crime as sale with consent.45 In a 1919 ruling, the Daliyuan acknowledged that while the “capital and goods” (zicai) given to acquire a concubine were sometimes referred to as “betrothal gifts” (caili), the law did not consider them to be “betrothal gifts for an engagement” (dinghun zhi caili).46 The same logic prevailed regarding the marriage certificate and the use of a matchmaker to find a concubine. Although the Yuan code started using the term “marriage certificate” (hunshu) in lieu of “marriage contract” (hunqi) in reference to concubinage, the law never lost sight of the fact that concubinage was at heart a financial transaction. Likewise, the intermediary who arranged the ­introduction 42 SMA Y5-1-38 (vol. 4, ch. 3): 31. 43 Ibid. 44 SMA Y5-1-38 (vol. 4, ch. 3): 32. This only applied during the betrothal period; the failure to deliver the betrothal wealth in a timely manner could not be used as grounds for annulment. See SMA Y5-1-38 (vol. 4, ch. 3): 34. 45 van der Valk 1968: 199–200. 46 SMA Y5-1-38 (vol. 4, ch. 3): 32.

358

Tran

was not considered to be in the same category as matchmakers who arranged matches for a main wife.47 In comparison to the Daliyuan’s rulings on uxorilocal marriages, the decisions in concubinage cases were based on a stricter interpretation of the legal criteria for marriage. With the former, the Daliyuan had been willing enough to accept as a legal substitute for a marriage document the simple agreement made to secure a uxorilocal son-in-law. Yet it was not so inclined to legally recognize what it conceded to be betrothal gifts given in the acquisition of a concubine. By emphasizing the financial over the ritual aspects of the exchange, the Daliyuan could ignore the semi-marital features of concubinage. In sum, the Daliyuan interpreted the legal criteria for marriage so that it could validate uxorilocal marriages but not concubinage. Yet as suggested in both this and other rulings, the Daliyuan could not avoid acknowledging the similarities between the protocols to be followed in the marrying of wives and of concubines. In response to a case involving a widow who, of her own volition, wished to remarry as a concubine, the Daliyuan instructed the lower court to base its decision on the existing laws governing widow remarriage. When a widow remarried as a wife or a concubine, the paternal grandparents or the parents of her deceased husband (or in their absence, her own paternal grandparents or parents) should preside over the wedding ceremony.48 Similarly, in a 1923 case that raised the question of whether a woman’s father had the legal prerogative to marry her off as a concubine, the Daliyuan explicitly applied the law upholding the authority of paternal grandparents and parents to arrange marriages.49 Although the Daliyuan used the more general phrase “to preside over” (zhuchi) instead of the more specific expression “to preside over a wedding ceremony” (zhuhun) to describe the father’s role in marrying his daughter off as a concubine, the comparisons between concubinage and marriage—which the Daliyuan had previously dismissed in its ruling on betrothal gifts—are inescapable. This does not, however, suggest a contradiction in the Daliyuan’s handling of concubinage, but rather the fundamental distinction it made between form and nature. While the Daliyuan was willing enough to concede that in social form, concubinage shared much in common with marriage, it consistently excluded concubinage from the law’s definition of marriage. Paradoxically, while the law refused to recognize concubinage as marriage, under certain circumstances, it acknowledged as a legal wife a woman who 47 48 49

Tao Yi and Ming Xin, Zhongguo hunyin jiating zhidu shi (History of Chinese marriage and family) (Guangdong: Guangdong chubanshe, 1994): 243. SMA Y5-1-38 (vol. 4, ch. 3): 33. SMA Y5-1-38 (vol. 4, ch. 3): 32.

Ceremony and the Definition of Marriage under Republican Law

359

had been acquired as a concubine. Just as in the Qing, a man retained the prerogative to promote a concubine to wife status after the death of his main wife, referred to as fuzheng.50 Since his main wife had died, the man was no longer considered to be married. He now had the option to make his concubine his main wife. In a 1917 ruling, the Daliyuan reaffirmed the right of a man to elevate a concubine to legal wife status after the demise of his main wife.51 Following late imperial practice, the Daliyuan acknowledged as a matter of course the legal status of a promoted concubine as wife. As in the Qing, a man’s word held the force of law; he could simply declare a favorite concubine his main wife, and the law recognized her as such. As the Daliyuan explained in a 1919 decision, in cases of fuzheng, no ceremonial rites need be performed unless required by local custom;52 in the eyes of the law, all that mattered was the man’s expressed intent to promote a concubine to main wife status.53 Another Daliyuan ruling in 1919 affirmed such arrangements as marriages exempt from the usual legal requirements for marriage. Unless local custom dictated that special rites be followed in the promotion of a concubine to main wife status, the law would not impose any additional procedures.54 In another case that appeared before the Daliyuan that same year, timing rather than custom resulted in the court recognizing as a legal wife a woman who it would otherwise have deemed a concubine. In the case at hand, the woman found out after the wedding that her husband was still married to another woman. She refused to accept the status of concubine, which was the only option the law offered her if she opted to remain in the household. Citing a Qing statute, the Daliyuan dissolved the second marriage and mandated the return of the woman to her natal family. Significantly, the Daliyuan qualified that if the first wife died before the second marriage had been dissolved, then the newly married woman would acquire the legal status of wife.55 Although these cases highlight circumstances that led to the recognition as marriage of what legally would have been considered concubinage, by no means does it signal any legal confusion between concubinage and marriage. 50 51

52

53 54 55

Bernhardt 1999: 169. Only the man who owned the concubine, however, enjoyed this privilege. In a 1914 judgment, the Daliyuan denied marital status to a concubine who had been promoted to main wife status by the household head’s relatives after his death. Guo 1933: 207, 219. In Guizhou, for instance, custom required that a ceremony be held in order for a promoted concubine to acquire main wife status. “Guizhou sheng zhi renshi xiguan” (People’s customs in Guizhou province), Falü pinglun (Sept. 20, 1925): 18. Guo 1933: 222. SMA Y5-1-38 (vol. 4, ch. 3): 33. SMA Y5-1-38 (vol. 4, ch. 3): 34.

360

Tran

The Daliyuan only granted wife status to a concubine after the death of the main wife, and then only if it reflected her master’s wishes. Even when presented with incontrovertible proof that a marriage had been concluded, the Daliyuan interpreted the legal requirements for marriage in such a way as to preclude the possibility of concubinage being granted legal validity as marriage.

The Legal Space for Concubinage in the Early Republic

Early Republican jurists continued the Qing practice of overlooking bigamy by treating it as concubinage. Qing jurists had handled cases involving a man with more than one wife by regarding the subsequent wives as concubines. If a suit was filed, then the court dissolved the later marriage and the woman was to return to her natal home.56 The Daliyuan followed Qing precedent, but made it explicit that a woman’s prior knowledge of her husband’s marital status would play a role in how it would handle a case of alleged bigamy. In a 1916 case, a man had married another woman while still married to his first wife. The Daliyuan explained that if the woman knew beforehand that the man she was marrying already had a wife, then the law considered her to have the status of concubine. The Daliyuan construed the woman’s prior knowledge of the man’s status as married to be informed consent to be his concubine. However, if the true nature of the man’s marital status had been withheld from her, then the marriage was dissolved and the second wife was to return to her natal family.57 Notably, the Daliyuan refrained from using the term “bigamy,” preferring instead to use language that emphasized the false pretenses under which the marriage had been completed. Indeed, the Daliyuan’s repeated use of the phrase “cover up by deception” (qishi) to describe the circumstances under which the marriage had been contracted highlighted the decisive role the woman’s ignorance of her intended’s married status played. Although Qing law had also ordered the dissolution of betrothals and marriages completed under false pretenses, the circumstances of this case would not have met the Qing definition of marriage by deceit. In another case in 1923, the Daliyuan declared that ignorance of the law did not cancel out prior knowledge of marital status. If, during the engagement period, the woman knew that her fiancé was already married but had been led 56 57

Xing’an huilan (Conspectus of penal cases) [1886] Reprint. (Taibei: Wenhai chubanshe, 1968): 40: 22a–25b. Xu 1932: 82.

Ceremony and the Definition of Marriage under Republican Law

361

to believe that she would assume the role of legal wife, then the law did not consider the resulting union bigamous.58 The woman in this case filed a suit for the dissolution of her marriage after she learned that, since she was the second wife, by law, she could not hold the position of main wife. Refusing to be a concubine, she invoked the law against bigamy and demanded a divorce. The Daliyuan rejected her suit, giving more legal weight to her prior knowledge of her husband’s married state than to her misunderstanding of the law. In keeping with its prior rulings, the Daliyuan regarded as a concubine any woman who knowingly married an already married man. Although the summary of the judgment made no mention of the custom of combined succession, the situation described suggests that this was most likely the situation. In the practice of combined succession, a man married two or more wives, one to continue his father’s patriline and the others to carry on the family line of one or more of his paternal uncles who had no sons of their own. The marriages could be conducted simultaneously or, as it appears in this case, at different times. Society considered all women married in this fashion to be “wives of equal standing” (pingqi), but, as Bernhardt points out, both Qing and early Republican law considered the later wives to be concubines. The Daliyuan did qualify in another case that if a subsequent wife married in the practice of combined succession did not know at the time of the engagement that her fiancé already had a wife and refused to accept her legal status as concubine, then she could seek divorce.59 That, of course, did not apply in the 1923 case above, in which the court dismissed the second wife’s suit for divorce on the grounds that she knew beforehand that her intended husband was already married. The decisive role of prior knowledge in shaping how the Daliyuan interpreted the law applied to other situations as well. In a 1920 case, the question of prior knowledge had to do with the man—it was the woman who was already married when the man had acquired her as his concubine. The court was exclusively concerned with whether or not he knew that she had a husband when he took her to be his concubine. The case details indicate that the woman had engaged in illicit sex with another man. The question facing the court was whether the concubine’s master had the right to press charges against her and her paramour. Article 12 in the 1914 Amendment Act made a concubine criminally liable for adultery by equating “a woman whose husband is still living” in Article 289 with “a concubine whose master is still living.” The second section of Article 294 of the code limited the right to press charges to 58 59

SMA Y5-1-38 (vol. 4, ch. 3, sec. 4): 37. SMA Y5-1-38 (vol. 4, ch. 3, sec. 4): 36.

362

Tran

the husband; after the passage of the Amendment Act, that right was extended to a concubine’s master. The Daliyuan ruled that the man in the case in question only had the right to charge his concubine with adultery if he had not known when he took the woman as his concubine that she already had a husband. If he had known the true circumstances of her marital status, the Daliyuan reasoned, then he possessed no legal right to file a complaint.60 Although the roles were reversed, with the man’s prior knowledge in question and the woman being twice married, the same logic prevailed. In the cases above, the woman’s prior knowledge of her husband’s married state robbed her of the right to sue for divorce; in this case, the man’s prior knowledge of his concubine’s married state would deny him the right to seek prosecution for her illicit sex crime. As these rulings attest, the Daliyuan did not categorically dissolve marriages that would appear to be bigamous, instead giving jurists the latitude to determine the applicability of the law against bigamy in a variety of situations. More often than not, Daliyuan rulings exempted cases that involved concubines from the law against bigamy. For instance, the Daliyuan held that if a woman did not learn of her husband’s married status until after the wedding, but she was willing to remain in the household as a concubine, then the court would allow it. Just as the Daliyuan construed a woman’s prior knowledge of the married status of her intended as willingness to be a concubine, it interpreted her acceptance of her husband’s married status after the fact as her consent to be a minor, rather than a main, wife. For example, in a 1917 case from Liyang in Jiangsu province, the court readily overlooked the legal fact that bigamy had been committed when a woman consented after the fact to being a concubine. In this case, a man inadvertently found himself married to two women. Years ago, the man’s first wife had been kidnapped during a burglary. Believing her to be dead, the man married another woman. However, his first wife was still very much alive; she managed to escape her captors and return home, only to find that her husband had already replaced her. Although the man was still legally married to his first wife, in the magistrate’s opinion no crime had occurred, since there had been no intent to commit bigamy. Yet the fact remained that the man had two wives—the first wife’s claim to legal wife status was indisputable, but what of the second wife? Referring to an earlier interpretation by the Daliyuan in which a second wife taken in the custom of combined succession was deemed a concubine for legal purposes,61 the magistrate reasoned by 60 61

SMA Y5-1-38 (vol. 2, ch. 23): 46. See Guo 1931: 245 for the 1916 Daliyuan ruling on the treatment of wives married in the custom of combined succession.

Ceremony and the Definition of Marriage under Republican Law

363

analogy that the second wife in the case at hand should also be considered a concubine.62 Receiving the magistrate’s report from the Jiangsu Superior Court, which refrained from comment, the Daliyuan added that the subsequent wife in this case could seek an annulment on the grounds of bigamy if she wished. But she could also remain in the household as a concubine, as the magistrate had suggested. As the Daliyuan offered, “if she of her own free will changes [her status from main] wife to minor wife, under the law, it is not particularly out of the question.”63 Under early Republican law, a woman’s informed consent to remain married to an already married man rendered that relationship concubinage and not bigamy in the eyes of the court. Indeed, in a number of rulings, the woman’s consent led the Daliyuan to tolerate a union as concubinage rather than punish it as bigamy. In 1919, the Daliyuan stated that in cases where “a man with a wife marries again, the wife married later, if she already knew [that her husband was already married] and was still willing to pass her days with him, would not be ordered to return to her natal home in accordance with the law, but should be considered to be a concubine.”64 In reference to a 1920 case, the Daliyuan explained that as long as the woman was willing to be a concubine, then the law would regard her as such.65 And in a 1923 case, the Daliyuan declared that if a woman had not known before the wedding that her intended was already married but subsequently accepted the situation, then the law dissolving such unions was not applicable.66 The woman’s consent was critical in determining whether a union would be dissolved on grounds of bigamy or tolerated as concubinage. This attests to the Daliyuan’s continued preservation of the legal space first created by late imperial law and later erased by the GMD civil code, a space which classified the concubine as a minor wife. In most of the cases above, the Daliyuan acknowledged that bigamy had been committed. But the court was willing to overlook that legal fact when the woman in the bigamous marriage agreed to be a minor wife. Her consent to the arrangement, regardless of whether it was expressed prior to the marriage ceremony or after the fact, barred an outside party from invoking the law against bigamy to challenge the union. In this respect, early Republican law continued to shelter concubinage from the laws against bigamy. As long as the court privileged the woman’s consent to be a concubine 62 Guo 1931: 340–341. 63 Ibid. 64 SMA Y5-1-38 (vol. 4, ch. 1): 31. 65 Guo 1931: 689. 66 SMA Y5-1-38 (vol. 4, ch. 3, sec. 4): 37.

364

Tran

over evidence of a ceremony, concubinage would continue to escape the laws on bigamy. All that would change with the implementation of the civil code.

Ceremony in the 1929–30 Civil Code

The civil code represented the cul