Polyandry and Wife-Selling in Qing Dynasty China: Survival Strategies and Judicial Interventions 9780520962194

Citation preview

Polyandry and Wife-Selling in Qing Dynasty China

This page intentionally left blank

Polyandry and Wife-Selling in Qing Dynasty China Survival Strategies and Judicial Interventions

Matthew H. Sommer

UNIVERSIT Y OF CALIFORNIA PRESS

University of California Press, one of the most distinguished university presses in the United States, enriches lives around the world by advancing scholarship in the humanities, social sciences, and natural sciences. Its activities are supported by the UC Press Foundation and by philanthropic contributions from individuals and institutions. For more information, visit www.ucpress.edu. University of California Press Oakland, California © 2015 by The Regents of the University of California

Library of Congress Cataloging-in-Publication Data Sommer, Matthew Harvey, 1961- author. Polyandry and wife-selling in Qing dynasty China : survival strategies and judicial interventions / Matthew H. Sommer. pages cm Polyandry. “Getting a husband to support a husband”. Attitudes of families, communities, and women toward polyandry. The intermediate range of practice — Wife-selling. Anatomy of a wife sale. Analysis of prices in wife sales. Negotiations between men in wife sales. Wives, natal families, and children. Four variations on a theme — Polyandry and wife-selling in Qing law. Formal law and central court interpretation from Ming through high Qing. Absolutism versus pragmatism in central court treatment of wife sales. Flexible adjudication of routine cases in the local courts. Includes bibliographical references and index. ISBN 978-0-520-28703-7 (cloth : alk. paper) ISBN 978-0-520-96219-4 (e-edition) 1. Married women—China—Social conditions—Case studies. 2. Polyandry—China—Case studies. 3. Rural poor—China— Case studies. 4. China—Social conditions—1644-1912. I. Title. HQ684.S66 2015 306.0951'09032—dc23 2015017190

Manufactured in the United States of America 24 23 22 21 20 19 18 17 10 9 8 7 6 5 4 3 2 1

16

15

In keeping with a commitment to support environmentally responsible and sustainable printing practices, UC Press has printed this book on Natures Natural, a fiber that contains 30% post-consumer waste and meets the minimum requirements of ansi/niso z39.48–1992 (r 1997) (Permanence of Paper).

For my wife, Ih-hae Chang, with affection and gratitude

This page intentionally left blank

contents

Acknowledgments Conventions in the Text Map: Provinces of China Proper within the Qing Empire, circa 1800 Introduction

ix xiii xiv 1

part one: polyandry 1. “Getting a Husband to Support a Husband”

23

2. Attitudes of Families, Communities, and Women toward Polyandry

55

3. The Intermediate Range of Practice

86

part two: wife-selling 4. Anatomy of a Wife Sale

117

5. Analysis of Prices in Wife Sales

149

6. Negotiations between Men in Wife Sales

179

7. Wives, Natal Families, and Children

211

8. Four Variations on a Theme

243

part three: polyandry and wife-selling in qing law 9. Formal Law and Central Court Interpretation from Ming through High Qing

277

10. Absolutism versus Pragmatism in Central Court Treatment of Wife Sales

308

11. Flexible Adjudication of Routine Cases in the Local Courts

341

Conclusion

376

Appendices A–E Character List Notes References Index

385 395 415 451 469

acknowled gments

Any project of this scale is to some extent collective in nature, and many people and institutions have helped me bring it to fruition. It is gratifying and cathartic finally to be able to thank them all formally in print. Of course, I alone am responsible for the claims I make in this book. First, I must thank my teachers at Swarthmore, the University of Washington, and UCLA, especially Kathryn Bernhardt, Daniel Chirot, Kent Guy, Lillian Li, Elizabeth Perry, and, above all, my doctoral advisor Philip Huang. I also thank the many patient language teachers who taught me Chinese and Japanese. This book took shape through countless conversations with three comrades from graduate school: Christopher Isett, Karasawa Yasuhiko, and Bradly Reed. They helped me at every stage of the project and provided excellent company on several research trips to China. Chris and Brad read early drafts of some chapters, and they arranged for me to give talks at their universities. Yasuhiko included me on a research grant from the Japanese Ministry of Education that financed two research trips to Sichuan. More important, he has always been a reliable source of superior whisky, even in the most unlikely places, and he managed to calm me down whenever capricious and infuriating nonsense at the archives threatened to make me lose my grip. Several other colleagues provided valuable feedback on the manuscript at different stages of development: Kathleen Brown, Hill Gates, Jonathan Greenberg, Avner Greif, Margaret Kuo, Ben Nathans, and Arthur Wolf. I also wish to thank the three readers for the press—Gail Hershatter and two who remain anonymous—for their exceptionally conscientious and useful reports. It was impossible to follow everyone’s advice, for the simple reason that not everyone agreed. But I have tried to act ix

x

acknowledgments

on areas of strong consensus while remaining faithful to my own vision of what this book should be. The following people helped by inviting me to give talks, commenting on papers I presented at conferences, writing recommendation letters, or providing other mentorship: Francesca Bray, Cornelia Dayton, Neil Diamant, Paul Dresch, Joseph Esherick, Bryna Goodman, Kishimoto Mio, Dorothy Ko, Wendy Larson, Li Shuzhuo, Kam Louie, Susan Mann, Steve Miles, Susan Naquin, Fernanda Pirie, Michael Puett, Qiu Pengsheng, William Rowe, Judith Scheele, Terada Hiroaki, Ding Xiang Warner, Barbara Welke, Yang Binbin, and Wen-hsin Yeh. I also wish to express my appreciation to several other colleagues who have made an important difference over the years: at Penn, Fred Dickinson, Jeff Fear, Lynn Hunt, Bruce Kuklick, Lynn Lees, and Tina Lu; at Stanford, Melissa Brown, Gordon Chang, Marcus Feldman, Zephyr Frank, Estelle Freedman, Richard Roberts, Paul Robinson, and Kären Wigen. Special thanks to my senior colleagues Harold Kahn and Lyman Van Slyke for their generous mentorship and for the high standard they set during their many years of teaching Chinese history at Stanford. I have presented parts of this project in public lectures at many places over the past sixteen years or so, and I wish to thank my hosts at the Academia Sinica; the Chinese University of Hong Kong; the University of Connecticut; Cornell University; Dickinson College; Georgetown University; Harvard University; the Hurst Summer Institute in Legal History; the Johns Hopkins University; Tel Aviv University; UCLA; the University of Hong Kong; the University of Michigan; the University of Minnesota; the University of Oregon; the University of Oxford’s Centre for Socio-Legal Studies; Princeton University; the University of Toronto; the University of Washington’s Jackson School of International Relations; Washington University at St. Louis; and Xi’an Jiaotong University’s Institute for Population and Development. I also gave more informal workshops or presentations at the First Historical Archive in Beijing, the Nanchong Municipal Archive, Penn’s Center for East Asian Studies, the Sichuan Provincial Archive, and Stanford’s Clayman Institute for Gender Research. The staff at three archives have my profound gratitude for making available the original legal cases that are the foundation of this book: the First Historical Archive in Beijing (especially Yin Shumei and Zhu Shuyuan), the Sichuan Provincial Archive (especially Ma Xiaobin), and the Nanchong Municipal Archive (especially Hou Wenfeng). I would also like to thank Shao Dongfang, Xue Zhaohui, and the rest of the staff at Stanford’s East Asia Library for their constant and indispensible support. In addition, several colleagues in China provided warm hospitality during research visits: Deng Jianpeng, Li Shuzhuo (and his colleagues at Xi’an Jiaotong University), Li Zan, Zhao Weini, and especially the late Tian Tao, whose untimely death was a great loss for his friends and for the entire field of Chinese legal history.

Acknowledgments

xi

This project was possible because of generous financial support from the American Council of Learned Societies and the Social Science Research Council; the American Philosophical Society; the Committee for Scholarly Communication with the People’s Republic of China; the Japanese Ministry of Education; the University of Pennsylvania (the Center for East Asian Studies, the Dean of Arts and Sciences, and the History Department); and Stanford University (the Center for East Asian Studies, the Clayman Institute for Gender Research, the Dean of Humanities and Sciences, the Freeman-Spogli Institute for International Studies, the Stanford Humanities Center, and the UPS Endowment). While writing this book, I had the privilege of working with some extraordinary graduate students at Stanford, and I thank them for all that they have taught me: Wesley Chaney, David Cheng Chang, Meiyu Hsieh, Ying Hu, Quinn Javers, Qiao Zhijian, Philip Thai, Brigid Vance, and Yiwen Yvon Wang. Several have completed their doctorates, and their own books will be coming out before long. I would also like to thank the students who took my courses on “Gender and Sexuality in Chinese History,” “Law and Society in Late Imperial China,” and “Qing Legal Documents”; our discussions helped clarify my thinking about many issues. A number of old friends have patiently endured innumerable conversations about this book, and no doubt they share my relief that it is finally finished: Frank Borchert, Chris and Siu Li GoGwilt, Daniel Licht, and Ding Xiang and Chris Warner. I owe an enormous debt of gratitude to Jonathan Greenberg, who has been my running partner these past twelve years, and who heard about every chapter of this book as I was writing it, while we ran around the Stanford campus together. He has been an unfailing source of optimism and encouragement. This book concerns people who struggled to survive in circumstances that were always difficult and sometimes tragic, and reading about their lives has made me acutely aware of my own good fortune in enjoying economic security and a happy marriage. My wife, Ih-hae Chang, deserves most of the credit for our good fortune, and I dedicate this book to her, with affection and gratitude. I also thank my parents, who made everything possible; my late brother Andy, for his sense of humor and faith in me—I think about him every day; and my kids, Anne and Joseph, for making me proud to be their father, and for helping me keep things in perspective by never being terribly impressed by anything I do. Finally, I would like to thank Reed Malcolm, Stacy Eisenstark, Kate Warne, Julia Zafferano, and the rest of the staff at the University of California Press, for their calm and reassuring professionalism in producing my book. At their urging, I shortened the original manuscript by about 20 percent (mainly by reducing the number of legal cases cited), and I am reluctantly convinced that the book is better—or, at least, more readable—as a result. (My daughter asked me, “does that mean you had to throw out two or three years’ worth of work?”) Readers should be aware that I had even more evidence to support my claims than I was able to include.

xii

acknowledgments

Shortly after submitting the final draft of my book to the press, I read a news report about a recent case of brideprice fraud in rural Hebei that involved over a hundred Vietnamese women.1 A matchmaker representing an organized ring had arranged marriages for gullible poor bachelors in villages with high sex ratios for a brideprice of RMB 115,000 per wife (approximately US$18,600). The women subsequently absconded, along with the matchmaker. Despite some distinctly modern factors (the involvement of foreign women and the sheer scale of the scheme), this episode illustrates a kind of brideprice fraud known as “flying a falcon” (fang ying) that my sources document in the Qing dynasty (see Chapter 8). The reason such fraud can succeed is that men expect to pay high prices for wives—which shows that, in at least some regions, brideprice-heavy marriage is once again widespread and perhaps even the default form of marriage. It would be absurd, of course, to suggest that nothing has changed over the past century. But it is a validation of sorts (however disconcerting) to encounter such vivid evidence of the relevance of my research to conditions in China today. I hope that this book will be useful in helping people understand those conditions and perhaps to find more effective ways of ameliorating them.

conventions in the text

Where possible, I have provided the ages of the protagonists in each legal case at the time that case was prosecuted (except when otherwise noted). Ages are expressed in sui, which are, on average, one more than the same age when reckoned in “years old.” For example, a person aged twenty sui is probably nineteen years old. Chinese women’s names are rendered as found in original sources. Usually, a peasant woman had no given name of her own (aside from an indicator of birth order, such as “second daughter” or “older sister”). Instead, legal documents would identify her by the surname of her father and sometimes (if she were married or widowed) that of her husband, followed by the term “shi” (literally, “lineage”). For example, “Wang Li shi” refers to a woman whose father’s surname is “Li” and whose husband’s surname is “Wang.” (In very formal contexts, this might be rendered “Wang men Li shi,” i.e. “Mrs. Wang née Li.”) In the text, I have not translated or italicized “shi.” The Chinese lunar and Gregorian calendars do not align (the lunar New Year falling in early spring), but, for convenience, when referring to dates I have converted years to the closest Gregorian equivalent. When giving exact Chinese dates (e.g., in citation of cases), I provide the reign period followed by the year, month, and day: thus, “Qianlong 12.10.2” means the second day of the tenth month of the twelfth year of the Qianlong emperor’s reign. (For Qing reign periods, see Appendix A.) In citation of Chinese dates, “r” refers to an intercalary month, and “?” means that part of date is unknown. Chinese names and terms are romanized in Hanyu pinyin according to standard Mandarin pronunciation. All translations are my own, except where noted. xiii

Counties Imperial Capital

SHENGJING ZHILI

Baodi County

KOREA

Riv e

r

Beijing

Ye llo

w

SHANXI SHANDONG

N GANSU SHAANXI

JIANGSU

HENAN ANHUI

HUBEI Yang zi R ive r

Nanbu County SICHUAN

ZHEJIANG

Ba County HUNAN

JIANGXI FUJIAN

GUIZHOU

Xinzhu County YUNNAN

GUANGXI

GUANGDONG

0

VIETNAM

TAIWAN (FUJIAN)

0

100

200

300 mi

100 200 300 400 500 km

Provinces of China Proper within the Qing Empire, circa 1800

Introduction

T H E I S SU E S

This book uses more than 1,200 legal cases from the central and local archives of the Qing dynasty (1644–1912) to analyze polyandry, wife sale, and a variety of intermediate practices that mobilized a woman’s sexual and reproductive labor to help support her family. Its main setting is the countryside, and its protagonists are the rural poor. By exploring this field of social practice, I seek to document and understand the roles played by marriage, sex, and reproduction in the creative strategies by which people survived under conditions of overpopulation, worsening sex ratios, shrinking farm sizes, and agricultural involution. How did people live under these conditions? Polyandry and wife sale represent opposite ends of a spectrum of strategies to supplement household income and maintain subsistence. At the polyandry end of the spectrum, an outside male would be fully incorporated into a couple’s household as the wife’s second husband or the first husband’s sworn brother, and he would share the wife’s bed in exchange for helping support her family. Hence, polyandry was a strategy to keep the family together by expanding it, thereby raising the ratio of laborers to consumers. At the opposite end of the spectrum, a wife would be transferred permanently to her buyer’s household in exchange for cash payment. She would escape poverty and get a fresh start with a new husband, while her first husband (i.e., the seller) would secure an emergency infusion of cash. Children usually accompanied the wife into her new household, either temporarily or permanently. In short, wife sale was a strategy to survive by breaking up the family, in the process creating a new marriage. In the mid-range of the 1

2

Introduction

spectrum was a variety of more informal arrangements whereby a wife would have sexual relations with one or more other men, with her husband’s approval, in exchange for material support. If we focus on social practice among the poor (instead of normative discourse among the elite), no clear distinction can be drawn between marriage and the traffic in women in Qing dynasty China; on the contrary, the two categories overlapped and were mutually implicated to a great degree. It also becomes impossible to sustain the clear-cut binary distinction between marriage and sex work that was basic to Qing law and elite ideology. By emphasizing the impact of material exigency, I seek to bring the analytical perspective of class back into the picture. But I do not ignore gender: on the contrary, the ideologies of masculine solidarity that informed polyandry, wife sale, and related practices are as high a priority for this study as are the experiences and perspectives of women. With regard to legal history, this book provides an in-depth case study of the interplay of ideology and practice in the Qing judicial system. Qing law prohibited all of these practices under the rubric “illicit sexual relations” (jian), and the main source for my book is legal cases from the eighteenth and nineteenth centuries that I have collected in Chinese archives over the past twenty years. By incorporating large samples of records from both central and local courts—and this book is the first study of Qing law to do so—I show how magistrates charged with propagating and enforcing a fundamentalist Confucian vision of female chastity tried to cope with the social reality of widespread wife sales driven by poverty. This contradiction illuminates the expedient pragmatism of routine judicial practice but also the increasingly dysfunctional nature of the dynastic state in the face of mounting social crisis. Since these transactions were prohibited, they had to be regulated and enforced on the community level, without reference to the courts. In this respect, they were but a subset of a much broader field of illicit customary practice that flourished in defiance of prohibition. Our understanding of Qing “law” should expand to include this field of illicit custom and community regulation. Past Scholarship and the Approach of this Book In 1994, Dorothy Ko launched a revisionist wave of Chinese women’s history by vowing to write “against the May Fourth legacy.” For the reformers and revolutionaries of the May Fourth era, “the victimized woman in old China” symbolized everything wrong with the old society that would have to be overcome in order to remake China as a modern nation.1 This paradigm found its most powerful expression in the polemical fiction of Lu Xun, Rou Shi, and other May Fourth writers (some of whom wrote about polyandry and wife sale), but it continues to influence the portrayal of women in both historical writing and popular culture in China today.2 As Ko explains, the problem with the victimization paradigm is not that it is absolutely wrong—it is “not

Introduction

3

without its grain of truth”—but rather that it is an artifact of modern ideologies that obscures more than it reveals about the actual experiences and perspectives of women in prerevolutionary China.3 Over the past two decades, Dorothy Ko, Susan Mann, and others have produced a powerful body of scholarship that seeks to reclaim the agency of women in late imperial China (1368–1912) and to celebrate what was positive in their lives. A fundamental goal is to discover what footbinding, polygyny, the cult of female chastity, and other practices often said to epitomize victimization actually meant to the women who engaged in them. The best of this work transcends the dichotomy between victimization and agency to explore in a nuanced way how women constructed meaningful choices within the constraints of the Confucian gender system and tested the flexibility of those constraints without rebelling outright.4 The power of much of this scholarship derives from its use of women’s own writings to “correct the distortions inherent in the male gaze and to see how women themselves articulate value and meaning in a society dominated by Confucian norms.”5 But the inevitable result has been a near exclusive focus on literate elite women, mainly from the Yangzi Delta, who constituted a tiny minority of the population. Indeed, a basic part of Ko and Mann’s agenda is to disaggregate the overly broad, seemingly timeless category of “Chinese women” by zeroing in on a specific historical period, social class, and geographic region.6 This is a necessary and laudable goal, and it remains incumbent on others to expand the scope of inquiry with studies of women in other periods, classes, and regions.7 But we also need a fuller picture of men and masculinity in late imperial China, to complement and balance our increasingly rich understanding of women’s lives. My own research focuses on the rural poor, and on men as well as women, but the victimization paradigm is no more helpful for understanding their lives than for understanding elite female poets. By casting women simply as victims—and, by implication, men as victimizers—the old paradigm privileges gender over all other factors in a simplistic way that obscures the fuller complexities of human relations. Take the example of wife sale: on the face of it, for a husband to sell his wife would seem like the epitome of patriarchal exploitation. There is some truth to that characterization, because, after all, such sales were part of a pervasive traffic that made commodities of women’s bodies. But if one looks at what actually happened in a wife sale, the usual scenario is that one man (the buyer) would gain at the expense of another (the seller), and, since the motive to sell was almost always poverty, the wife’s move from one household to the next often resulted in substantial improvement in her security and standard of living. Furthermore, in most cases a sold wife took her children with her, leaving the seller alone. Under the circumstances, the “loser” was usually not the wife herself but rather her first husband, who would join the multitude of single men who made up the Qing

4

Introduction

underclass. Moreover, if one assumes a wife sale to have been simply a transaction between men in which the woman was a passive object, it is hard to explain why so many sales resulted from women’s demands to be sold, or how others were sabotaged by women who refused to be sold. In other words, if we presuppose the big story to be women’s victimization, we will fail to comprehend what actually happened. As this example should make clear, my analysis builds on the revisionist insight that there was real scope for female agency within the constraints of the old gender order. What I hope to add is an attention to survival strategies among the poor that involved non-normative and even non-patriarchal alliances, as well as a sympathetic effort to understand the experiences and perspectives of the men, as well as the women, who found themselves in such circumstances. My principal inspiration for this research project has come from two bodies of scholarship. The first is the classic social and economic history of China that focuses on the lived experience of the peasantry in order to understand the roots of social crisis and revolution. I have in mind especially the studies of rural north China by Elizabeth Perry, Philip Huang, Joseph Esherick, and Susan Naquin, who were inspired in part by the British Marxist and French Annaliste schools of social and economic history.8 I first encountered the ubiquitous “bare sticks” (guanggun—poor, single men) of rural China in Perry’s analysis of how endemic patterns of violence in Huaibei helped foster the Nian Rebellion. Thus, my interest in gender history began not with elite women but rather with the most despised and exploited men in China. Perry’s work also taught me that what the state or elite condemned as deviance might constitute a rational survival strategy for the people engaged in such behavior. Philip Huang’s analysis of how peasants enduring agricultural involution would mobilize family labor to produce handicrafts and engage in sidelines for diminishing returns has provided a basic framework for understanding my own evidence about polyandry, polyamory, and marital prostitution. One form of family labor was the sexual and reproductive labor of women, and one possible sideline was sex work.9 The second body of scholarship is the classic social anthropology of China that focuses on gender, kinship, and community at the village level. This work prioritizes the logic of social practice in local context over normative prescriptions and ideals, and in this respect it complements the historical scholarship cited above. Here, I have in mind especially the work of Arthur Wolf and Hill Gates but also of anthropologists such as Myron Cohen, Margery Wolf, and Janice Stockard.10 Arthur Wolf has used the Taiwan household registers in conjunction with fieldwork to analyze a variety of non-normative marriage forms, and his finding that some practices were stigmatized but nevertheless widespread, because they solved problems and met needs that normative ones could not, is one point of departure for my own study. Wolf and Chieh-shan Huang’s refreshingly frank assessment of

Introduction

5

the implications of female promiscuity for the marriage system has helped me see my own evidence more clearly, too.11 Hill Gates has documented the significance of female labor (especially handicrafts) for rural household incomes, the way female labor is subsumed by gender ideology into “obedience,” and the role of footbinding in disciplining and deploying that labor. Her findings complement Huang’s analysis of how involution pushed peasants into the market, and she brings a distinctly gendered perspective to that dynamic. Gates’s analysis of the incidence of brideprice-heavy marriage undergirds my own understanding of the economic logic of wife sale.12 With my perspective shaped by these two bodies of scholarship, I have come to this topic by way of the stories told in legal cases from the Qing archives. Given the illiteracy of the poor majority during the Qing, these cases are by far the most revealing sources about their lives that we are ever likely to find. The testimony they record (mediated though it is by the judicial process) is the closest we will ever come to hearing their own “voices.”13 Whereas my first book was roughly twothirds legal history and one-third social history, in the present book I reverse the balance. The type of social history I attempt here reflects the influence of the anthropologists, in that a basic priority is to document marginalized kinship practices and to analyze their logic in context, from the standpoint of the people involved. Fundamentally, this represents an effort to get past judicial categories and the orthodox values that informed them, as well as the “enormous condescension” of both May Fourth polemicists like Lu Xun and modern historians like Guo Songyi, in order to understand what people did, why they did it, and how they felt about it.14 This book also breaks new ground in Chinese legal history. The sharp contradiction between the widespread practice of polyandry and wife sale and the ideological mandates of the judiciary helps to expand our perspective on Qing law to include the pragmatic adjustments and compromises that magistrates had to make in dealing with routine cases. It also provides us with a deeper understanding of customary norms, rules, and practices that existed outside the formal judicial system—sometimes in harmony with it, but often in contradiction with it. This informal realm included community mediation of disputes over “minor matters of household, marriage, and land” (hu hun tiantu xi shi), most of which were settled out of court.15 But it also included a wide variety of prohibited practices that were common because they solved problems that approved practices did not. These practices had to be regulated at the community level, because to take them to court would guarantee trouble for the participants. The negotiation of wife sales, including demands for supplementary payments after a sale had been concluded, is a paradigmatic example. Others include the use of “white contracts” for land sales (“white,” because they lacked the red seals indicating registration and payment of transfer tax); the sale of Qing manorial land in Manchuria and Zhili

6

Introduction

(and native land in Taiwan, Yunnan, and other frontier zones) to Chinese migrants; the production and sale of salt outside the state monopoly; and the formation of collective brotherhoods. This illicit field of community regulation implies an alternative set of values and more-or-less conscious resistance to the state.16 How does our analysis of “Qing law” change, if we include this realm of illicit practice? Moreover, how does the perspective from this informal realm help explain social and political change in the dynasty’s last decades, when the imperial center’s power to impose its will weakened dramatically? Skewed Sex Ratios and the Traffic in Women The practices documented in this book were part of a pervasive traffic in women that affected every social class and most families in China during the Qing dynasty.17 This traffic was closely linked to the imbalance in the ratio between the sexes that has long prevailed in China: a shortage of women that has most severely affected poor rural communities, where the surplus of adult males might well exceed 20 percent.18 The stubborn persistence of skewed sex ratios is a profoundly important continuity in modern Chinese history. Although ratios improved somewhat during the Maoist era (1949–76), in recent years they have returned to levels not seen since the early twentieth century. According to the 2000 census, the sex ratio “at birth” (which reflects the effects both of sex-selective abortion and of infanticide) for the People’s Republic of China (PRC) overall was 117 males per 100 females, but eleven provinces exceeded 120, and three of these exceeded 135. These ratios resemble those from the 1930s as well as the data we have for scattered locales in the eighteenth and nineteenth centuries.19 Excess female mortality due to systematic discrimination has long been the main cause of the sex ratio imbalance, although in recent years sex-selective abortion has become a crucial factor. Scholars debate how common infanticide was before 1949, but there is no question that some did occur and that its incidence would rise in times of famine.20 Whatever the actual rate of infanticide, it is clear that childhood mortality was (and continues to be) far higher for females than for males, especially among the rural poor, and sex ratios actually worsen between ages one and four.21 Moreover, during periods of extraordinary hardship (such as the Great Leap Forward famine), sex ratios have suddenly spiked, showing that discrimination against daughters has been, in part, a crisis strategy to ensure the survival of sons. Such discrimination has taken a number of forms aside from outright infanticide and abandonment, the most important being relative quantity and quality of nutrition and health care. For example, infant daughters were often weaned earlier than sons, because earlier weaning would enable the mother to get started on a new pregnancy, in hope of a boy. But also, since it was understood that longer breastfeeding improved an infant’s likelihood of thriving, this was a higher priority for a son than for a daughter.22 Discrimination was not limited to infants, of course, and, even

Introduction

7

today, excess female mortality affects all age cohorts. For example, China is one of the few places in the world today where women commit suicide more often than men: Chinese women commit more than half of all female suicides worldwide, even though China accounts for only one-fifth of world population. Suicide is concentrated among young rural women, just as it was in the late Qing.23 If we seek to understand how sex ratios influence individual behavior, national and provincial data are less useful than specific, micro-level case studies. For example, in the Qing legal cases I use for this study, we often find extreme ratios of four or five males to one female among the protagonists: a woman, her husband, a son or two, plus one or more single men who are sleeping with her. Typically, there are no daughters in the picture. In such cases, the lone woman is the focal point in a web of relations among men, sometimes becoming the effective head of her extended household. Ironically, the high sex ratios in such milieus seem to have empowered at least some women in their relations with men (see Chapters 1 and 2). At the opposite end of the socioeconomic spectrum, we know that the households of gentry and wealthy merchants included many female servants and that most elite men had concubines (qie) in addition to one main wife (qi). An extreme (albeit fictional) example is the Jia household in the eighteenth-century novel Honglou meng (Dream of the Red Chamber). A surplus of young women is one of many luxuries enjoyed by the fabulously wealthy Jia family, and any Qing reader would have recognized this reverse sex ratio as a form of conspicuous consumption. Its most extreme manifestation is found in the hero Jia Baoyu’s famous garden sanctum, where he is the sole male, surrounded by a bevy of attractive girls in a sort of parody of the imperial harem.24 To some extent, the traffic in women exacerbated the raw imbalance in poor communities by exporting women to become servants and concubines in prosperous households while others became prostitutes in urban settings. At the same time, the practices documented in this book responded to the shortage of wives in poor communities by making a relatively small number of women available to a larger number of men. One reason to share a wife was that there simply were not enough wives to go around—and polyandry, polyamory, marital prostitution, and conditional wife sale all involved a husband sharing his wife with one or more other men, in exchange for material support. Wife sales and widow remarriage served a similar function, by recycling one woman through more than one marriage. Moreover, the poor, unmarried man—known in Qing legal discourse as a guanggun (which translates literally as “bare stick,” or more colloquially as “rootless rascal”)—played a central role in these scenarios.25 In polyandry, he was the outside male “brought in” by a poor couple either as a second husband or the first husband’s sworn brother; similarly, a couple’s outside partners in polyamory would also be single men. In most wife sales, the buyer was a single man who had never before married but, by hard work and good luck, had managed to save enough

8

Introduction

money to buy another man’s wife (this being a relatively inexpensive way for a man to marry). However, the buyer’s upward mobility created a new bare stick—the seller—because few men who sold wives could recoup the resources necessary to acquire another. We have surprisingly little scholarship on the traffic in women in late imperial China, but most of what we have focuses on servants and concubines who were purchased by elite households from their parents through brokers.26 The sale of children by their parents was perfectly legal in the Qing, no doubt because the elite wanted to buy these children.27 Moreover, polygyny served the interests not only of elite men but also of their wives: a man could have only one main wife, who would come from the same social background as he, and she would have nearly absolute authority over the inner quarters of their household. Polygyny enabled an elite wife to monopolize the prestigious role of social motherhood over all of her husband’s children while shifting much of the burden of bearing them onto the concubines and maidservants who were also sexually available to her husband. This division of labor constituted a remarkable example of class exploitation within a single family.28 It is important to bear in mind that elite men and women had a vital stake in the traffic in women and girls, even though normative discourse seldom explicitly acknowledged this fact. By the eighteenth century, bonded servitude no longer played a major role in the productive economy, but the traffic in women continued to play a key role in the biological and social reproduction of the elite.29 Elites aside, it is clear that the routine form of marriage practiced by many peasants (especially the poorest of them) was simply to sell a daughter to the groom’s family—even if, for reasons of face, the transaction was not always explicitly labeled a “sale.” In other words, the brideprice (caili or caili qian) paid by the groom’s family far exceeded any dowry, which was often trivial in value (if any was given at all). Daughters usually married out, whereas sons remained with their parents and brought in daughters-in-law, and therefore peasants understood the brideprice to be compensation to a woman’s parents for the cost of raising her. One can also assume that brideprice included compensation for the loss of the daughter’s labor.30 These facts were accepted by Qing officialdom. As the late Qing jurist Xue Yunsheng observed, “to sell one’s own daughter or sister in marriage (jiamai) to a man to become his wife or concubine is a legitimate form of marriage (ben shu hunyin zhi zheng).” Here, Xue purposely uses the colloquial term “to sell in marriage” (jiamai), which was also used for illegal wife sales.31 There was some regional variation in the incidence of brideprice-heavy versus dowry-heavy marriage, and a number of factors influenced their distribution, but it is clear that wealth and class played a major role in structuring these practices. Dowry was a status symbol because so many people could not afford it, and a lavish dowry was one means by which the elite converted material capital into symbolic capital, to show that they were rich enough and moral enough not to sell their daughters.32 Furthermore,

Introduction

9

most widow remarriage constituted a direct or indirect sale of the woman to her new husband (see Chapter 8). In the post-Mao era, with the end of collectivization and return to family farming, brideprice-heavy marriage has once again become widespread in at least some parts of rural China, while sex ratios have steadily worsened.33 One sign of this development is the return of a kind of marriage fraud that was common before 1949, in which a gullible man is duped into paying a high brideprice for a woman who then runs away. The targets, in villages with high sex ratios, are older bachelors who are cheated out of their savings. The reason such scams succeed is that these men expect to pay high prices to acquire wives, without any dowry in return, and since no local women are available they are willing to risk marrying an outsider.34 Given this larger context, it would be a mistake to assume that there was any clear practical boundary between “marriage” and “traffic”; on the contrary, the two categories overlapped and were mutually implicated to a very large degree. Th erefore, I contend, wife sale and the other strategies documented in this book should be understood as variations of the dominant pattern rather than as deviant exceptions, notwithstanding the fact that they were stigmatized and prohibited. One goal of this study is to establish the absolute centrality of the traffic in women to the Chinese marriage system. Stigmatized and Prohibited Forms of Marriage The practices documented in this book were all stigmatized to some degree, as well as being prohibited by Qing law. But they were not unique in either respect, and it is not clear that their stigma exceeded that of other unorthodox marriage practices. Many forms of marriage carried stigma, to the extent that they diverged from the normative ideal of “major marriage” (in which a grown-up bride would be transferred to her husband’s household in exchange for brideprice, sometimes bearing dowry), and some forms were also prohibited. But stigma and prohibition did not necessarily deter people from contracting such marriages. Moreover, perception of stigma might vary by region, social class, and even gender. Widow remarriage is a case in point. Neo-Confucian orthodoxy condemned remarriage as a violation of chastity, and the practice was unknown among the elite during the Qing. Among the poor, however, remarriage was normal, and high sex ratios made it easy for young widows to find husbands. But even among the poor, remarriage carried a certain stigma, expressed in a variety of customs and taboos. In many regions, a widow had to be delivered to her new husband at night, and she would have to mount the sedan chair some distance away from her first husband’s home and agricultural land (to avoid damaging its fertility). Members of her first husband’s village or lineage might waylay the sedan chair to demand “money to cover shame” (zhexiu qian)—a type of hazing seen in wife sales as well.35 Moreover, Ming and Qing law prohibited widow remarriage during the official

10

Introduction

mourning period of three years, although this law was seldom enforced.36 In fact, it was common for a poor widow to remarry immediately, so that the brideprice from her new marriage could be used to clear her husband’s debts or even to buy his coffin (most remarriage constituted a direct or indirect form of sale in which the in-laws received payment). Levirate—in which an unmarried man inherited his brother’s widow through remarriage—was found in many parts of China, even though Ming and Qing law prohibited it as incest to be punished by strangulation, and even though case records show that the death penalty was indeed imposed when it came to official attention.37 A number of other examples can be cited. Delayed transfer marriage (in which a bride delayed moving in with her husband for several years) was common in the Pearl River Delta, where sericulture made the labor of young women especially valuable to natal families. During the period of delay, brides would visit their husbands only on holidays and would avoid sleeping with them or eating food from the in-laws’ hearth. The Qing dynasty did not proscribe the practice, but local officials and elites found it horrifying and did their best to suppress it.38 The Qing did prohibit most marriage between Han Chinese and other peoples as a threat to “ethnic sovereignty.”39 But such prohibitions were difficult to enforce (except for urban banner garrisons), and in frontier zones intermarriage between Han men and indigenous women was common. In fact, “hanjian”—the modern word for “traitor”—originally referred to Han males who lived among indigenous peoples and “went native” by adopting their customs.40 Uxorilocal marriage—in which a husband would move into his wife’s household—was a frequent minority practice throughout China, and in some regions it accounted for as much as 15 percent of marriages. The typical uxorilocal husband was an orphan or younger son with no prospects, who would be “brought in” to marry a woman without brothers. Qing law permitted this form of marriage, but for men it was considered shameful because it required them to abandon their parents, and it was proverbial that a decent man would refuse such a marriage. Even “minor marriage” (adopting an infant daughter-in-law, or tongyangxi) was stigmatized, although it could be found all over China and was majority practice in some communities. This form of marriage saved money and also served the interests of the mother-in-law (by letting her raise her daughter-in-law herself, instead of bringing in an adult bride who might become a rival). But the adopted daughterin-law was popularly viewed as a pathetic, abused creature, and these marriages were notoriously unhappy (in part because childhood association tended to foster sexual aversion). In Taiwan (where the practice had been common), it disappeared as soon as socioeconomic change liberated youth from parental authority, and on the mainland it was prohibited by Communist marriage reforms.41 The practices documented in this book should be seen against this broader background of marriage forms that diverged from the normative ideal. They were

Introduction

11

stigmatized and sometimes criminalized, but each made sense in its own context; each, in its own way, was a solution for problems including the shortage of wives, poverty, and the high cost of brideprice in major marriage. Therefore, stigma and prohibition did not necessarily deter people from these practices, and such prohibitions were difficult to enforce. In the absence of violent crime, illegal marriage practices came to official attention only when someone directly involved was sufficiently upset to go to court. For this reason, I assume that the vast majority never left any record: the legal archives reveal only the tip of a huge iceberg. Marriage, Sex Work, and Queer Domesticity Another basic goal of this book is to challenge the normative distinction between marriage and sex work that was vital to elite lifestyle and orthodox ideology in the Ming-Qing era. Most scholarship on sex work in China has focused either on elite courtesans in the late empire or on brothel prostitution in modern urban settings. Shanghai is particularly well documented, with major studies by Gail Hershatter, Christian Henriot, and Catherine Yeh on the century of “semi-colonialism” from the Opium War to the Communist victory.42 Much of this scholarship has focused more on elite discourses about prostitution than on actual social practices; despite its excellence, it has relatively little relevance for the present book, given my very different focus on survival strategies of the rural poor and on the multifarious connections between marriage and sex work in that context. My own previous work on prostitution in late imperial China shows how an ageold regime of regulation, based on legal status distinctions, was replaced in the eighteenth century by a blanket prohibition that extended the free commoner (liang) standard of female chastity to all. Previously, prostitution had been tolerated as a hallmark of hereditary debased (jian) legal status, whereas any extramarital sexual activity had been prohibited to women of free commoner status (who constituted the great majority by the eighteenth century). In other words, the purpose of regulation was not to prohibit a given conduct but rather to require people to conform to the standards appropriate to their respective statuses. Beginning in 1723, however, the Yongzheng reforms expunged the debased status of the groups associated with prostitution, thereby eliminating their exemption to the prohibition of extramarital sexual relations. The practical result was to criminalize all prostitution.43 From the standpoint of the present study, a significant feature of the pre-1723 regulatory regime was its premise that debased status prostitution took place within marriage and was organized on a household basis, with a husband/father pimping wife and daughters; as in farming, the household was the unit of production. Confucian ideology and imperial law held marriage and prostitution to be irreconcilable opposites, but that standard applied only to free commoners and elites; in contrast, “prostitute households” (chang hu) of hereditary debased status represented a sort of mirror image, in which women were neither expected nor

12

Introduction

entitled to adhere to the free commoner standard of chastity. These women were supposed to marry within their caste: marriage meant sex work under the management of a husband instead of a father.44 The archival legal cases used for the present study all date from the era of prohibition, after 1723. But, even after prohibition, it appears that most retail sex work continued to be linked to marriage. As I explain in Chapter 3, most of the prostitutes found in Qing legal cases were married women who were helping to support their families. In this context, sex work actually supported marriage in that it enabled impoverished couples to survive without permanently separating. In fact, the same was true of the entire spectrum of polyandrous practices documented in this book: their common feature is that a wife, with her husband’s cooperation, would sleep with one or more other men in order to help support her household. Like many scholars who have studied prostitution in recent years, I see it primarily (if not solely) as a form of work,45 but unlike most, I emphasize the marital and familial context of this work. The focus of my study is not the urban brothel but rather the peasant household, where the familization of sex work through a variety of polyandrous arrangements was part of a menu of survival strategies for coping with poverty and agricultural involution. When farms shrank below the size necessary for autonomous subsistence, families would mobilize their own underemployed labor to engage in a variety of market-related activities, in addition to farming, in order to maintain household incomes.46 The purpose of polyandry and polyamory was to keep the family together, and usually also to stay on their land. From this standpoint, these arrangements can be seen as variations of the way that female labor produced marketable goods (such as cotton yarn and cloth) and can also be compared to the way semi-proletarianized peasants hired out excess labor to other farms. In effect, a wife’s sexual and reproductive labor was a commodity that she and her husband could sell or hire out to other men. However, “selling” and “hiring out” are not entirely accurate characterizations for this kind of sex work, because often these women would open their homes to the men who helped support their families in exchange for sex. Indeed, with full polyandry—which should be considered a form of marriage—the family would expand by “bringing in” an outside male either as the woman’s second husband or as her husband’s sworn brother. With this arrangement, sex work was domestic labor that took place within the family. Moreover, the “wholesale” services provided by women within polyandry and polyamory were not limited to sex but also included various forms of domestic caring work (cooking, making and mending clothes, etc.) in an extension of wifely duties within the household. In contrast, the landless peasants who engaged in retail sex work in urban settings approximated the condition of fully proletarianized peasants who subsisted entirely on wage labor. In thinking about the familization of sex work in China, I find inspiration in Luise White’s classic study of prostitution in colonial Kenya.47 By focusing on the

Introduction

13

economics of sex work—that is, seeing it as work and analyzing what women did with their earnings—White challenges an earlier generation of scholarship that highlighted the victimization of women. She emphasizes that the processes by which rural women entered urban prostitution reveal “the abilities of families and individual women to solve their problems through the mobilization of their own labor.”48 Kenyan prostitution in its many forms was “family labor”: it supported and reproduced families, it held poor families together, and it created families with women at their heads. Moreover, these women serviced migrant laborers who could not afford wives of their own and therefore sought them out as surrogates. To a certain extent, these laborers resembled the surplus males who participated in polyandry and polyamory in China. White’s title, The Comforts of Home, reflects the fact that Kenyan sex workers sold a range of domestic services in addition to sex, including “all that is legitimately available in marriage.”49 Although her focus is urban retail prostitution, and the circumstances of colonial Nairobi differed from those of rural China, many of her insights apply there as well. In my first book, I used the term “unorthodox households” to cover various scenarios of “people excluded from accepted patterns of marriage and household because of poverty and other factors bonding with each other in unorthodox ways to satisfy a range of human needs.”50 Another way to characterize these scenarios is “queer domesticity,” a term that historian Nayan Shah uses to describe nineteenthcentury Chinatown, San Francisco, where sex ratios among Chinese immigrants were very high. In this context, “queer” does not necessarily imply same-sex sexual acts (Shah also uses the term “perverse heterosexuality”) but rather a variety of alliances and living arrangements that were viewed by white municipal authorities with prejudice and alarm. To them, the largely male Chinese population appeared to observe no clear boundaries between families, nor clear parentage of children: Chinatown was a promiscuous milieu of bachelors and prostitutes, who inhabited dormitories, brothels, and opium dens, and posed a grave threat of contagion (both physical and moral) to the white population.51 In fact, the living arrangements found in Chinatown mirrored those found in high sex ratio contexts throughout China and its frontiers (the diaspora being an extension of the frontier), where it was common for men to form sexual alliances with each other, to share wives, or to partner with indigenous women. For our purposes, “queer domesticity” evokes the nonnormative (and often non-patriarchal) arrangements found on the polyandrous spectrum as well as the alarm that they provoked on the part of Qing ideologues. S OU R C E S

The main source for this study is Qing legal cases from central and local courts that I have collected in Chinese archives over the past two decades. I have discussed these categories of cases elsewhere, so here I confine myself to describing

14

Introduction

the samples used in this study.52 The central cases are xingke tiben: routine memorials in which provincial governors reported major criminal cases to the imperial center for review. (For the Qing penal system, see Appendix D; for the process of judicial review recorded in xingke tiben, see Appendix E.) Most of these cases concern homicide, but modern archivists have sorted them according to the background situation that framed the main crime, and that background situation was my focus in selecting cases. For this study, I have used more than 800 xingke tiben, all from the category “marriage and sex offenses” (hunyin jianqing): 303 involve wife sales of one kind or another, and the rest concern polyandry, polyamory, and marital prostitution—that is, the full range of polyandrous practices covered by the statute against “abetting or tolerating a wife or concubine having illicit sexual relations with another man” (zongrong qi qie yu ren tong jian—often abbreviated as zong jian). These cases come from all provinces of China Proper, although the traditional core provinces contribute the bulk.53 About 90 percent date from the Qianlong (1736–95) and Jiaqing (1796–1820) eras. This study also uses more than 400 local court cases, the majority coming from Ba and Nanbu counties (both in Sichuan), which have the richest local archives known (as of this writing) to survive from the Qing. Ba County has the largest collection by far, with more than 100,000 legal and administrative case files. During the Qing, Ba County included Chongqing, which was then, as it is now, the most important port on the upper Yangzi River. Given its key position in both administrative and commercial networks, Ba County was socially and economically far more complex than the other three counties in my sample. Moreover, Chongqing was a boomtown, its population more than quadrupling over the last century of the Qing to reach nearly one million. These distinctive features of Ba County gave a particular shape to the traffic in women there (see Chapter 4). Nanbu County was a far more ordinary place. Located some two hundred kilometers to the north (as the crow flies), it was linked to Ba County by the Jialing River, which joins the Yangzi at Chongqing. I also have smaller samples of local court cases from Baodi County, Zhili (a typical rural county on the north China plain, located some eighty kilometers southeast of Beijing), and Xinzhu County in northern Taiwan.54 Three hundred and forty-five of my local court cases concern wife sales; the rest concern polyandry and marital prostitution. Table 1 breaks down the local wife sale cases by date and county. The numbers in this table reflect not change over time but simply the fact that most of what survives dates from the late Qing as well as limits on my time and access at the archives in Sichuan.55 Since about three-quarters of my local cases come from a single county, it is fair to ask whether they are representative of either social practice or the administration of justice elsewhere. As far as I can tell, they are. The cases of wife sale, polyandry, and marital prostitution from Ba County are similar to those from elsewhere. To be sure, there was some regional variation in these practices. For

Introduction

15

table 1 Wife-Selling Cases from Local Archives by Date and County Years

Ba

Nanbu

Baodi

Xinzhu

Total

1750–99 1800–49 1850–99 1900–12 TOTAL

26 121 94 7 248

0 17 53 4 74

0 5 13 1 19

0 0 4 0 4

26 143 164 12 345

example, in Nanbu County it was normal for a wife sale contract to bear both the handprint and the footprint of the seller, whereas contracts from Ba and Baodi counties usually bear only a handprint. Contracts from Sichuan are usually longer than those from Baodi. In other respects, however, the three counties’ contracts are essentially the same. There was also local variation in terminology, especially for polyandry. However, the big story is not regional variation (which seems fairly minor) but rather the high degree of uniformity throughout China. With regard to adjudication, we should bear in mind that Ba County was a major administrative center, with Chongqing serving as county seat as well as headquarters for Chongqing Prefecture and the East Sichuan Circuit Intendant. Given the proximity of their superiors, it is unlikely that magistrates of Ba County could have deviated much from accepted practice. If anything, they probably took greater care to follow the rules than the magistrate of a remote county might have. But comparison with my smaller samples from other counties shows great consistency in how routine cases were adjudicated. Moreover, my Ba County sample includes judgments by several dozen different magistrates, all of whom hailed from other provinces. Most had already served elsewhere because, given this county’s importance, only men of proven competence would be posted there.56 (When I speak of “a magistrate” judging cases, I have in mind both the individual appointed to that office who bore responsibility for all decisions and the privately employed legal experts who advised him and did much of the actual work.)57 One purpose of personnel rotation was to standardize administration, including the administration of justice, and in this respect it seems to have been effective. This study is the first to use large samples of both local and central court records from the archives to study Qing judicial procedure. The juxtaposition of local and central cases highlights their very different qualities as historical sources. Xingke tiben focus on major crimes that required exhaustive investigation and systematic reporting according to a standardized format. For this reason, they consistently provide more detailed testimony and factual information than the local court records, the vast majority of which concern the routine adjudication of “minor matters” (xi shi) that did not have to be reported up the chain of command. But local archives contain the raw material of cases, including litigants’ plaints

16

Introduction

(bearing magistrates’ rescripts) and original contracts, and many of the latter were never intended for official gaze. Usually, such documents are not transcribed in xingke tiben. Local archives are especially valuable for the present study because they contain many contracts for wife sales and related documents (such as “request certificates” issued to matchmakers, and “submission contracts” exchanged for supplementary payments), which were submitted in evidence and confiscated when magistrates voided those transactions. My local court sample includes copies of 107 contracts for wife sales of various kinds, plus a comparable number of ancillary documents.58 Few such contracts survive outside local court archives, because there was no reason to preserve them beyond the lifetime of the individuals involved. In contrast, contracts for land sales, debt, and household division all retained value indefinitely as proof of title and might be preserved for generations. For this reason, vast numbers of those documents survive from the Ming-Qing era. A comparison of local and central court records also highlights two very different modes of adjudication practiced by county magistrates, depending on the kind of case. The provisional judgment of “major criminal cases” (zhongda anjian) required a strict, precise application of the Qing code in rigid deference to protocol. In death-penalty cases, magistrates prepared their reports for the eyes of the emperor, who in principle had the final say. In contrast, they judged routine cases involving only “minor matters” on their own authority. Routine adjudication was guided by a flexible pragmatism that prioritized concrete solutions over strict enforcement of the code. (Chapter 11 explores these issues.) After Qing legal cases, the most important sources for this study are two surveys of local customs from the early twentieth century.59 The most important is Minshi xiguan diaocha baogao lu (Report on an Investigation of Customs—henceforth, Investigation of Customs), which was compiled during the last years of the Qing and the first years of the Republic in preparation for drafting a modern civil code. This survey includes much information about unorthodox practices, and its descriptions of “getting a husband to support a husband” and conditional wife sale (to cite just two examples) are a priceless supplement to the evidence from legal cases. But we need to bear in mind the limitations of this survey: its reports vary widely in quality, and it leaves out most counties and many provinces altogether, including the far south and entire southwest of China. Therefore, it would be a mistake to assume that the practices in question did not exist in places not specifically mentioned.60 The second survey is Taiwan shihō (The Private Law of Taiwan), compiled by Japanese colonial authorities in the years immediately after annexation in 1895. This survey includes transcriptions of many contracts for wife sale, and it appears to be the largest existing collection of such documents outside local court archives. A limitation of both surveys is that they usually record only the most formally contracted practices that can be considered variations of marriage. Thus, for

Introduction

17

example, Investigation of Customs leaves out the version of polyandry that was framed by sworn brotherhood, as well as transactional polyamory. Its compilers either did not know about such arrangements or (more likely) considered them too deviant to mention in a report on “customs.” But since the Qing code criminalized the full range of practices covered by this book, we can turn to the legal cases for a more complete picture of what was going on. When using legal cases as sources for social history, one must always consider their inherent sampling bias. Since these practices were against the law, the participants generally had a strong interest in avoiding official attention. Moreover, by their very nature, legal cases usually record only people who got in trouble. This bias is especially strong in cases of homicide, which is a highly exceptional event in any society. Therefore, these sources inevitably give an exaggerated impression of the incidence of conflict and violence. People prone to conflict and violence will be heavily overrepresented, whereas harmonious relationships would rarely have left any trace in the public record.61 Even so, the kinds of trouble that ended up in court can also be very revealing, since they expose the tensions intrinsic to these practices and give explicit voice to what might normally be left unsaid. For example, the routine local cases show that the single most common reason for a wife sale to end up in court was that the seller demanded more money from the buyer after the sale had ostensibly been concluded. In Chapter 6, I examine these disputes to illuminate the perspectives of the men involved in such transactions (including the close parallel between wives and land in the male peasant imagination) as well as the role played by local communities in mediating what were, after all, illegal transactions. A basic challenge in interpreting the legal cases is to see beyond the judiciary’s criminal categories to understand why people behaved as they did and how they understood their own behavior. Therefore, it is helpful to read the legal cases together with the surveys, which record “customs” rather than “crimes.” This juxtaposition shows that the practices documented by both sources were widespread survival strategies that operated according to well-known customary rules rather than exceptional acts of deviance. Other primary sources for this book include published casebooks from the Qing and Republican-era gazetteers. In Chapter 10, I use casebooks to explore formal reasoning at the highest levels of the judiciary, the most important being the early nineteenth-century Xing’an huilan (The Conspectus of Penal Cases) and its sequels, which legal advisors to the Board of Punishment compiled as reference works for sitting magistrates. They contain brief summaries of judgments that involve complex fact situations, balance competing legal principles, or apply the code by analogy—for this reason, the cases they include tend to be unusual and are useful mainly for illuminating judicial reasoning.62 I also cite reference works for the Autumn Assizes that show how senior officials weighed aggravating and

18

Introduction

mitigating factors when advising the emperor in the final disposition of capital cases (see Appendix D.3). Gazetteers from the Qing rarely record unorthodox marriage practices, so they are of little use for this study. But many Republican-era gazetteers reflect a modernizing spirit of investigation, and they sometimes report “vulgar customs” (lou xi) that the compilers hoped to reform. Such reports can be useful (in conjunction with Investigation of Customs) for indicating the minimum geographic scope of certain practices—for example, levirate (see Chapter 1) and uxorilocal widow remarriage (see Chapter 8). A N OV E RV I EW

This book is organized thematically in three parts. Part One (Chapters 1–3) and Part Two (Chapters 4–8), on polyandry and wife sale, respectively, analyze these practices in a quasi-anthropological manner. Part Three (Chapters 9–11) examines ideology and practice within the judicial system, with primary focus on the treatment of wife sales. Chapters 1 and 2 document the two main frameworks for bringing an outside male into a family, namely a contract for “getting a husband to support a husband” and sworn brotherhood. They explore community attitudes toward these arrangements and the subject positions of the parties themselves: the ideology of masculine solidarity that informed the men’s relationship with each other, and the attitudes and experiences of the women who played an indispensible role in polyandry. Chapter 3 covers a wider range of practices on the polyandrous spectrum, including transactional polyamory (ongoing sexual-economic exchange with two to four regular male partners), marital prostitution, and conditional wife sale. Chapters 4–7 address the direct, open form of wife sale: an honest transaction negotiated directly between seller and buyer with the help of a matchmaker. Who engaged in such transactions and why? How were they negotiated, using what kinds of contracts? How were prices determined and how much were they worth? What kinds of disputes arose, and how were they mediated? Again, a key theme is the distinct interests and subject positions of the different parties: the patriarchal ideology that informed men’s attitudes toward wives and land as well as the range of women’s attitudes and the resources they might draw on either to provoke a sale or to sabotage one. Chapter 8 expands the inquiry to include several important variations: compensated divorce (i.e., an indirect sale brokered by the wife’s natal family), widow remarriage, sale in the husband’s absence, and fraudulent sale. It concludes by showing how fraud ties them all together. Chapters 9 and 10 dissect codified law and central court decisions (in major cases reported up the chain of command for review) in order to explain what constituted a criminal wife sale in the Qing dynasty. The key change since the former Ming was that the Qing extended prohibition to include wife sales motivated by

Introduction

19

poverty—that is, the vast majority of such sales. In addition, Chapter 10 explores the tension within the central judiciary between the orthodox, empowered view that condemned all wife sales as licentiousness, and a more pragmatic minority view that recognized the motive for most sales to be poverty. Chapter 11 shifts focus from the imperial center to the local courts, and from ideology to the practical adjudication of routine cases not subject to review. Again, the main focus is wife sale. How did magistrates reconcile the codified prohibition with the social reality of widespread sales driven by poverty? What does this case study of local court practice tell us about the Qing judicial system as a whole? The heart of this book is my detailed telling of the stories found in Qing legal archives. Many of the issues I address are difficult to quantify in any precise way, and, given the nature of the main evidence (hundreds of anecdotes from court cases), any conclusions must be somewhat provisional. But a clear picture does emerge, one that no other surviving sources can provide, of life on the margins during the Qing—and as I hope to show, the view from the margins casts the whole in a new and different light. The beauty of this material is the vivid, textured sense of people’s lives that it reveals: their passions, hopes, and fears; the stark choices that circumstances forced upon them; and their creative ways of getting by.

This page intentionally left blank

pa rt one

Polyandry

This page intentionally left blank

1

“Getting a Husband to Support a Husband” A country with a system of one man and many wives has to have a system of one woman and many husbands. Moreover, there have to be a certain number of men who have no wives, and those without wives are sure to be the poor. —pioneering feminist he-yin zhen, 1907 1

A C A SE O F P O LYA N D RY O N T H E N O RT H C H I NA P L A I N

In 1743, peasant Wang Yuliang realized that he could no longer feed his family. Therefore, he decided to use his wife to recruit into the family a man who could.2 Wang (aged 49) lived in Fangshan County, Zhili, about fifty kilometers southwest of Beijing; his household consisted of himself, his wife, Li Shi (41), his widowed mother, Fu Shi (79), two young sons, and a daughter. The six of them shared a oneroom house. Wang owned only four mu (about two-thirds of an acre) of poorquality land, so much of the family’s income depended on what he could earn by hiring out his labor (an example of “semi-proletarianization” typical of the north China plain).3 To make matters worse, for several years Wang had suffered from a chronic illness that made it difficult to keep down food, and he was bedridden much of the time. These circumstances prompted Wang to approach Hao Shixin (37), an immigrant from Neiqiu County, Zhili (about 350 kilometers to the southwest), who was working in the village as a casual laborer. Hao Shixin had neither land nor family, but he was strong and healthy. Wang proposed that Hao move in with Wang’s family and sleep with Li Shi in exchange for “farming and supporting the family,” and Hao readily agreed. At first, Wang’s wife refused to cooperate, but eventually he persuaded her. As Li Shi later testified, he told her “I have this sickness and I can’t take care of you anymore, so all we can do is get him to support us and get along as best we can.” 23

24

part one

Li Shi protested that they had only one room and sleeping platform in their house; so Wang explained, “Everyone will sleep together, but you don’t need to be ashamed.” She finally relented out of resignation and disgust, because if the family were to survive, they would need the help of some other man. Wang’s mother was unhappy, too: “[I] saw that my son had brought (zhao) Hao Shixin into our family. My son told me that he was going to let Hao Shixin live with us and sleep on the same kang4 with my daughter-in-law, so that he would farm our land and support our family. I said, ‘We may be poor, but how can we do something like that?’ ” But she, too, bowed to the inevitable. So it happened that Hao Shixin moved in with Wang Yuliang’s family, shared their sleeping platform, had sexual intercourse with Li Shi, and supported them as best he could by working their land and hiring out his labor. Neighbors later testified that everyone had had a pretty good idea what was going on, but no one interfered—after all, what better solution did they have to the family’s problems? But sometimes Hao could not get work, and the family continued to go short on food, provoking Wang Yuliang to complain and to abuse his wife; moreover, Wang was ill most of the time, and, by throwing up everything he ate, he was seen to be wasting quite a bit of all-too-scarce food. Wang had become a taxing burden to his family, something especially difficult to tolerate when all were going hungry. Finally, in the summer of 1744, Li Shi persuaded Hao Shixin to help kill her husband, so that they could be a couple and have a better life together. After the murder they were quickly found out, prosecuted, and sentenced to death—which is the only reason we know their story.5 How Widespread a Practice? This story illustrates the practice known as “getting a husband to support a husband” (zhao fu yang fu), a form of non-fraternal polyandry that, with some variation, appears to have been remarkably widespread among the poor in China during the Qing dynasty. How widespread? It is impossible to quantify the practice in any exact way, and I would not suggest that most people participated in such relationships. But it certainly was no isolated phenomenon. The story of Wang Yuliang’s family appears in a xingke tiben from the archival category “marriage and sex offenses,” where a common scenario is an impoverished couple being supported by one or more outside males in exchange for sexual privileges. From a judicial point of view, such behavior constituted the crime of “abetting or tolerating one’s wife or concubine to engage in illicit sexual intercourse with another man” (zongrong qi qie yu ren tong jian—often abbreviated as zong jian), for which the Qing code mandated ninety blows of the heavy bamboo (for the woman and both men) and compulsory divorce.6 The “marriage and sex offenses” category contains countless memorials related to this crime. It is also well represented in the archives of local courts.

Chapter 1. “Getting a Husband to Support a Husband”

25

In addition to criminal records, two early twentieth-century surveys of customs document the practice of “getting a husband to support a husband.” Investigation of Customs reports this “evil custom” in Fujian, Gansu, Hubei, Shaanxi, Shanxi, and Zhejiang, and The Private Law of Taiwan reproduces a contract for one such marriage.7 Many other sources also mention the practice. Therefore it seems safe to assume that, for every instance of polyandry mentioned in a legal case, there must have been a great many others that left no specific written record. Together, the legal cases and the surveys show that these practices occurred in every province of China Proper. The Big Picture Bearing this evidence in mind, let us consider the larger context for the story of Wang Yuliang. One precondition for the arrangement proposed by Wang was his ability to recruit Hao Shixin as a second husband for his wife. Hao was an ablebodied man with no wife or property of his own, who had migrated far from home in search of a livelihood, and Wang had no trouble at all persuading him to accept the proposal. There seems to have been no shortage of men like Hao, the ubiquitous “rootless rascals” or “bare sticks” (guanggun) at the bottom of Qing society. The larger context is the skewed sex ratio and concomitant shortage of wives among the rural poor that were already widespread and troubling phenomena by the mideighteenth century. One reason for men to share a wife was that there simply were not enough wives to go around, and, in some rural communities, as many as a fifth of adult males would never marry, even though marriage was universal for women.8 A second precondition was a pervasive market for women, specifically their sexual and reproductive labor. It required no wild stretch of the imagination for Wang Yuliang to come up with this solution to his family’s problems: when all else failed, his family had one more asset, namely his wife’s body. Nor was it really very difficult for other people to understand and accept the arrangement, Li Shi’s professions of distaste notwithstanding. This was a society in which it was both possible and easy—one can even say that it made sense—for a man to pimp or sell his wife in order to survive. A third precondition, of course, was the desperate poverty of Wang Yuliang’s family, exacerbated by his peculiar illness. The larger context is that there were many downwardly mobile families, living on farms too small to support themselves, who were turning to a range of desperate strategies in order to survive. In this particular case, we are reminded of Philip Huang’s analysis of the involutionary pressure on peasant families to mobilize underutilized labor and engage in sidelines and risky cash cropping in order to survive.9 Among other things, polyandry represented a familization of sex work. This context helps us make sense of the myriad legal cases where we find a wife taking one or more patrons who chip in to supplement her family’s income, with

26

part one

her husband either openly embracing her initiative or simply pretending not to notice. In these situations, sex work is usually not the only kind of work going on; rather, it is part of a portfolio of strategies that enable a family to get by. In that sense, it is typical of most sex work that goes on in the world, which is part-time, temporary, or seasonal activity designed to supplement other sources of income in order to support families. The women doing this work do not necessarily see themselves as “prostitutes”—that is, they do not necessarily see sex work as the most important or defining aspect of their lives.10 These three larger phenomena—the shortage of wives and consequent surplus of single men, the market for women’s sexual and reproductive labor, and the problem of widespread downward mobility and involutionary pressure on poor families—were connected, and at their intersection we find people like Wang Yuliang, Li Shi, and Hao Shixin engaged in survival strategies that combined elements of marriage and prostitution in a range of polyandrous forms. Some arrangements were formalized with matchmakers and contracts, or with some type of chosen kinship; others were more casual and depended on verbal agreement or more indirect ways of reaching an understanding (such as the husband turning a blind eye to what he knows his wife is doing). Each case tells a unique story. But among the countless anecdotes, we can discern common patterns and logic that make sense only when considered at the intersection of larger forces. M A R R IAG E , P R O ST I T U T IO N , A N D P O LYA N D RY

Qing legal cases reveal a variety of arrangements by which a wife, with her husband’s approval, would have sex with one or more other men in order to help support her family. Some scenarios look more like marriage, others more like sex work, and many like something in between.11 A number of variables can be used to assess a given scenario. First, how many outside sexual partners did the wife take, and how long did their relationship(s) last? At the marriage end of the spectrum, we find a stable long-term relationship between the wife and one partner in addition to her husband. Many of the relationships recorded in homicide cases ended in trouble within a year or two—but this evidence is misleading, because such sources inevitably exaggerate the incidence of conflict and violence. Despite this bias, I have found forty-five examples of polyandrous relationships that lasted at least four years; eleven of these relationships lasted more than ten years, and three lasted more than twenty years. Such long-term relationships were relatively harmonious (hence their stability), and they came to official attention only indirectly; for this reason, it is safe to assume that long-term relationships are heavily underrepresented in legal cases. The single longest polyandrous relationship I have found comes from a 1747 case from Jianyang County, Fujian, involving a peasant named Zheng Guoshun who

Chapter 1. “Getting a Husband to Support a Husband”

27

became blind, prompting his wife, Jiang Shi, to negotiate a sexual relationship with a younger man named Jiang Yilang (no relation) who would help work their farm. Eventually, Jiang Yilang moved in with the couple, and over time Jiang Shi bore two daughters of ambiguous paternity. After twenty-eight years, Zheng Guoshun died of natural causes, after which Jiang Shi and Jiang Yilang continued their relationship. They came to official attention only later, when an in-law’s interference provoked a violent quarrel; as long as Zheng Guoshun was alive, no one had bothered them.12 At the other end of the spectrum, we find multiple partners, each of whose relationship with the wife lasted only the duration of each “trick.” This scenario was a form of retail prostitution in which the husband acted as pimp and tout. Second, to what degree did the couple incorporate an outside male into their family, and how did they represent that relationship to themselves and others? At the marriage end of the spectrum, we find the outside male fully incorporated as a second husband by means of contract, kinship vocabulary, co-residence, resource pooling, the sharing of meals, and sometimes change of surname. For example, in Hubei “the second husband who was brought in” would adopt the first husband’s surname to formalize his integration into the family.13 In legal cases, we also find examples of the couple and their children adopting the outside male’s surname. In their dealings with this man, the couple does not maintain boundaries: a phrase that repeatedly appears in testimony is “bu fen nei wai”—literally, “they do not distinguish between inner and outer,” a reference to the inner female space of the household from which outside males were normally to be excluded. In other words, they treat him as a member of their family. At the prostitution end of the spectrum, however, the woman’s multiple sexual partners may be completely anonymous strangers. They are simply customers, and the couple is fully self-conscious about being engaged in prostitution. Third, what sorts of benefits were exchanged between the couple and the outside male(s), and were they exchanged in a “wholesale” or “retail” manner? At the marriage end of the spectrum, we find an ongoing exchange of a variety of different benefits over time; this is a “wholesale” exchange, in that we find no itemized calculation of compensation for each discrete sexual favor, and more is involved than just sex and money. The heart of this quid pro quo may well be an exchange of economic support for sexual relations, but, once incorporated into the family, the outside male will also partake of the entire package of domestic caring work performed by the wife for her family, including food preparation, mending and making clothes, cleaning, care for the sick, and so on. He also gains the less tangible benefits of membership in a family, including both chosen kinship (such as sworn brotherhood, or adopting the couple’s children as “gan qin,” something like a godfather) and the opportunity to have children of his own with the wife. For their part, the couple gains security through the pooling of labor, income, and whatever other resources the outside male can contribute on an ongoing basis.

28

part one

The benefits a wife provided her second husband were simply an extension of her ordinary tasks within the family. Paola Tabet’s description of the context of sex work in rural Niger applies equally to our Chinese cases of polyandry: “In the villages, giving sexual service is integrated with the other services women give in marriage: domestic labor, reproduction, and all the tasks allotted to women by the sexual division of labor.”14 Sex was just part of the package, and the second husband did not pay “by the trick” any more than did the first. Of course, at the opposite, prostitution end of the spectrum, we find straightforward, “retail” transactions: discrete acts of sex for discrete payments of money, which constitute the family’s cash income. The present chapter and Chapter 2 both focus on those stable, long-term arrangements that most closely resembled marriage: those contracted between a couple and a single outside male who joined their family as, in effect, a second husband who lived with them, pooled resources, shared the wife, and ate from the same hearth. There were two basic frameworks for such relationships. The first involved a formal contract (either written or verbal) for “bringing in” a second “husband”; the case of Wang Yuliang narrated above is an example of such a relationship. The second involved framing the polyandrous relationship in terms of chosen kinship, with sworn brotherhood between the two men being the most common pattern. Chapter 3 explores a broader range of strategies on the polyandrous spectrum, including transactional polyamory, retail prostitution with husband as pimp, contracting a wife to a brothel, and conditional wife sale. The common theme of all these practices is that a wife would sleep with one or more other men, with her husband’s permission, in order to help support her family. F O R M A L LY C O N T R AC T E D P O LYA N D RY

The term “getting a husband to support a husband” (zhao fu yang fu) generally referred to a formally contracted relationship. It should be understood as a form of marriage, despite its official illegality and its unacceptability to elite standards; it was certainly understood as such by its participants, and even by the community at large. What stands out is the formality and openness of the arrangements, in conscious imitation of more widely accepted forms of marriage. Contracts for “Getting a Husband to Support a Husband” Investigation of Customs notes that polyandry might be formalized through the use of matchmakers and written contracts. In Shaanxi, for example, The couple will talk it over and agree to ask a matchmaker to bring a second husband into their household to support the first husband (zhao fu ru jia, yi yang qian fu). They will draw up a “bringing in a husband contract” (zhao fu ju zi), which clearly states that “the second husband may not mistreat the first husband.”15

Chapter 1. “Getting a Husband to Support a Husband”

29

In other places, the polyandry contract might be verbal, instead of written, but in other respects the arrangements appear to be identical. In Gansu, for example, If a man takes a wife, but later on he grows old and weak, suffers from severe illness, or becomes impoverished, so that he cannot make a living, then with his permission his wife can go through a matchmaker to bring a second man into their household as a husband (zhao zhi qi jia wei fu); this second husband will take responsibility for all the needs of the household. The two sides will strike a verbal contract (kou tou qiyue) to settle questions such as which husband will get [as successors] any children who are born.16

In Liangdang County, Gansu (located on the border with Shaanxi), most contracts consisted of a verbal oath sworn by the outside male, who promised to respect the first husband as “older brother” and to support him for life; the oath would be ritualized by drinking wine and bowing before the first husband’s ancestral altar.17 In Qing legal cases, it is not always clear if the terms of an agreement were written down. Verbal contracts do seem to have been far more common than written ones, although the balance may have shifted by the twentieth century, when the surveys were conducted. I have the texts of two written contracts for “getting a husband to support a husband.” The Private Law of Taiwan reproduces the following specimen from 1869: Wang Yunfa hereby establishes this contract for getting a husband to support a husband (zhao fu yang fu zi). Years ago I married Li San’s daughter, Li Xiuliang, who is now aged twenty sui. We have lived together for four years. Xiuliang is filial in serving my parents and she takes care of the household without creating trouble or stirring up quarrels. It makes one very content to have such a good wife. Unfortunately, some time ago I contracted a disease and have become paralyzed. We are poor and have no source of income to meet our expenses. Although at the present time we are not starving, we have considered the fact that “there are three kinds of unfilial conduct [and the worst is not to have heirs]” (buxiao you san).18 When my wife’s youth expires, it will be impossible to have a son. After long discussions, we have decided that there is no other alternative: if we insist that Xiuliang preserve her chastity, the whole family will be threatened with starvation. The only solution is to get a husband to support a husband. We have, therefore, consulted a matchmaker, and it has been arranged for Wu Jiusheng’s first-born son, Wu Jinwen, to enter our family as an uxorilocal husband (dengmen jin zhui), and he [and Xiuliang] will become husband and wife (chengwei fuqi). We have, on this day, agreed that there will be no brideprice, but that Wu Jinwen should provide the family with 20 yuan a month to cover expenses. Regardless of how many sons and grandsons Wu and Xiuliang may have, they will be heirs to the Wang family as well as to the Wu family.

30

part one Both parties have reached this agreement voluntarily and without regret. Since spoken words alone are unreliable, I hereby establish this contract for getting a husband to support a husband, which can be presented as proof of our agreement. Dated Tongzhi 8.3.?, and signed with marks of husband Wang Yunfa, the matchmaker, the scribe, and two witnesses.19

This agreement closely resembles those described by Investigation of Customs as customary in several other provinces. We find the same basic vocabulary, including the ubiquitous “zhao fu yang fu,” which was used everywhere from Taiwan to Gansu. The Taiwan contract is written in the voice of the first husband, and begins with a narration of hardship that justifies the unorthodox transaction as a last resort undertaken with regret; in both respects it closely resembles contracts in which husbands sold or divorced wives, or parents sold (“adopted out”) their children. Contracts for widow remarriage also justify themselves with a preamble of hardship and regret. Such narratives of woe have a formulaic quality, but all the evidence suggests that no one was proud of the kinds of transactions documented by these contracts.20 The editors of the Taiwan survey and several contributors in Investigation of Customs point out that “getting a husband to support a husband” was a form of uxorilocal marriage, in that the new husband would be incorporated into his wife’s family. It closely resembled uxorilocal remarriage of a widow, who (with permission of in-laws from her first marriage) might “bring in” a second husband without changing her surname or surrendering her place in her first husband’s household.21 The latter practice was called “zhao fu yang zi” (getting a husband to support one’s sons) and other similar terms. As I explain in Chapter 8 below, uxorilocal remarriage should be considered a variation of polyandry in which the first husband simply happened to be dead. An important issue addressed by the Taiwan contract is the two men’s (and, by extension, their respective lineages’) claims to any sons born to the wife. Practice apparently varied quite a bit—and the matter was especially important where contracts set a time limit on the arrangement, meaning that eventually the second husband would leave the couple’s family. Investigation of Customs cites limits of anywhere from three to ten years in various places. In Gutian County, Fujian, these relationships were generally limited to a decade: “Any sons born within the ten years go to the second husband for him to raise. There are also some first husbands who lack sons, who stipulate that the first son born during the ten-year period will go to the second husband, but that any subsequent sons must become successors to the first husband himself.”22 This focus on progeny indicates there was more at stake in these formally contracted relationships than just sex and money. One of the important benefits being

Chapter 1. “Getting a Husband to Support a Husband”

31

exchanged was reproductive ability in the form of surrogate motherhood or fatherhood. The contracts described in Investigation of Customs all make clear that the second man was entitled to at least one of any sons produced by the relationship. We are reminded of a related form of surrogate motherhood, namely a husband’s “conditional sale” (dian) of his wife to a creditor for a limited period of time in order to bear him sons (see Chapter 3 below).23 The Taiwan contract adds surrogate fatherhood to the picture: any sons born out of polyandry should be successors to both husbands. The same provision is reported in Zhejiang: “If the first husband has no issue, then sons born to the second husband may serve as successors to the first husband as well.”24 Our second example of a written contract for “getting a husband to support a husband” appears in a 1917 legal case from Longquan County, Zhejiang. This case documents a polyandrous relationship that had been contracted under the following circumstances. Peasant Pan Zhaozhi was married to He Shi (then aged 36) and they had three sons, all of whom were still too young to work. When harvest failure led to hunger, the couple decided “to bring in” an outside male to help support their family. Pan’s younger brother acted as matchmaker, and he recruited a single man named Dong Zhiyou. The contract, dated 1901, reads as follows: Pan Zhaozhi hereby establishes this marriage contract for bringing in an uxorilocal husband (zhaozhui hun shu). I took in marriage my wife, He Shi, whose given name is [ . . . ], she is just 36 sui, and she has given birth to three sons all of whom are still little. Our household is impoverished, and the harvest was poor, so that we do not have enough food to eat and cannot provide our three daily meals, so that the whole family is prostrate from exhaustion. My wife and I have discussed the situation, and we have decided that it would be inappropriate to lack heirs and also that we cannot separate from one another (buhe fa si, weike fen zhen). Therefore, we decided that we should choose a hardworking and thrifty man to bring in as uxorilocal husband, in order to help support our family (zhao fu ruzhui bang jia). We begged the matchmaker, who has selected Dong Zhiyou to bring in as uxorilocal husband to help support our family. On this day, we have agreed through the matchmaker on a marriage price of exactly 13 foreign yuan. I have received that amount in full, and not a cent more is owed. After bringing in Dong Zhiyou to our household, my oldest son and second son will remain in the Pan lineage to carry on my family line, but my third son will take the name “Dong Liangren” and enter the Dong lineage to carry on the Dong family line. This way, both surnames can continue into the future. Once we have agreed, Pan and Dong will get along harmoniously as brothers. This agreement was reached after discussion, and neither side will change their mind or challenge this agreement. Fearing that spoken words are unreliable, I specially establish this marriage contract for bringing in an uxorilocal husband. May this marriage produce one hundred sons and one thousand grandsons. I hereby establish this marriage contract for bringing in an uxorilocal husband, to keep as proof.

32

part one On Guangxu 27.8.29, I, Pan Zhaozhi, hereby establish this marriage contract for bringing in an uxorilocal husband. Matchmaker: younger brother Pan Zhaochu Scribe: Xie Yibang

Several features of this contract deserve attention. First of all, it is clear that the couple had considered and rejected two other possible strategies for survival: selling their sons, and wife sale (hence the euphemistic reference to “lacking heirs” and “separating”). Instead, they decided to keep their family together by “bringing in” what the contract explicitly labels “an uxorilocal husband.” This is an excellent example of how polyandry was a survival strategy that kept a family together (by adding an outside male), in contrast with wife sale, which ensured survival by breaking up a family. Second, whereas the Taiwan contract specifies that the outside male should provide a monthly stipend, the Zhejiang contract instead stipulates a single, lump sum payment—the rough equivalent of a brideprice—and makes no reference to pooling resources. Third, the arrangements for carrying on both husbands’ family lines are creative and precise. This couple had a surplus of sons, which no doubt was one of the chief inducements for Dong Zhiyou to ally with them. He was authorized to adopt one of their sons, but he also gained conjugal privileges with He Shi and the chance to father more—hence the contract’s formulaic invocation of “one hundred sons and one thousand grandsons.” Fourth, by stipulating that “Pan and Dong will get along harmoniously as brothers,” this contract invokes the principle that fraternal bonds should encourage sharing and mute conflict—calling to mind the second major context for polyandry, sworn brotherhood. Despite the contract’s allusions to “bringing in” a new husband, testimony reveals that Dong Zhiyou did not actually move in with the couple but instead maintained a separate household nearby; He Shi divided her time between the two households, going back and forth on a regular basis. The couple’s third son moved in with Dong, as agreed, to be his son and heir, and He Shi and Dong adopted a daughter-in-law to be the boy’s bride. This arrangement continued for sixteen years, until Dong’s adopted son died in 1917. At this point, He Shi took the adopted daughter-in-law back to Pan’s household to be the bride of one of her older sons (an example of levirate), and she severed their ties with Dong Zhiyou. This act provoked Dong to file charges against the Pan family, for “seizing my daughter-inlaw by force” and “marrying a younger brother’s wife to his older brother.”25 He Shi’s actions show that, after her son’s death, she assumed the relationship between the Pan and Dong households to be at an end. By that time her age was fifty-two sui, and no doubt she believed that she had fulfilled her sexual and reproductive duties to Dong Zhiyou.26 He Shi’s actions also imply that she considered the adopted daughter-in-law to be part of her own uterine family rather than Dong’s patriarchal family.27

Chapter 1. “Getting a Husband to Support a Husband”

33

Another point stands out: although the contract records an agreement between two men, its text emphasizes that He Shi was a full partner in the decision, and in practice the arrangement depended entirely on her willingness to cooperate— indeed, to play an active role in both husbands’ households for sixteen years. At each stage, her attitude must have been decisive: in the decision to opt for polyandry (instead of selling sons or separating via wife sale), in her conjugal sojourning, and in the way she finally severed ties with Dong. Her primary loyalty lay with her first husband and sons, and her top priority was their survival as a family. It is obvious that the son whom Dong had adopted mattered more to her than her connection with Dong himself. This case suggests a larger conclusion that the bulk of my evidence supports: the success of polyandry and the other strategies discussed in this book depended in large part on the cooperation of the wife. Therefore, it was often the wife’s attitude that determined which particular strategy a couple chose to pursue. Two Examples of Formal Negotiation of Polyandry We find an example of the open, formal negotiation of polyandry in a case from Shaanxi in which a poor family contracted consecutive relationships with two different outside males; the record provides a detailed account of the second relationship. “Old Wang” (51) and his wife, Wen Shi (31), were originally from Pucheng County, but in 1748 they fled famine with their young son and made their way to Yijun County (about seventy-five kilometers to the northwest). Wang found work as a casual laborer, but he soon came down with tuberculosis and could no longer make a living. The couple coped by allying with an immigrant laborer named Li Wen’ge; they moved in with Li and “ate from the same hearth,” in exchange for which Wen Shi slept with him. This relationship continued for about a year, and it ended only when Li decided to return home to Shanxi, leaving the couple with no means of support. At this point, another immigrant laborer named Hei Jing noticed their difficulty. Hei was a widower from Yichuan County (about 150 kilometers to the northeast) who had no children and could not afford to remarry, and it occurred to him that “it would be less complicated and expensive just to contract an exclusive relationship (bao) with Old Wang’s wife.” He sent a mutual friend to propose that he replace Li Wen’ge as the couple’s patron, and they agreed. Hei then asked the village head to act as matchmaker to negotiate the terms. With this man as witness, Hei promised husband Wang to “support him for the rest of his life” (yang lao) and to raise his son to maturity and secure him a wife; Hei then presented Old Wang with a cloth jacket to seal the deal. In exchange, Wang promised to “yield” (rang) his wife to Hei—as she later testified, Wang “agreed to use me to bring in Hei Jing as husband” (jiang ming ba xiao furen zhao Hei Jing wei fu). The family moved in with Hei, Wen Shi began sharing his bed, and they “ate together as one family”

34

part one

(tong jia chi fan). In testimony, Wen Shi referred to this relationship as “getting a husband to support a husband.”28 The parties to this transaction clearly understood it as a form of marriage, in which (to use their terms) an outside male was “brought into” the family as a “husband.” An interesting detail is the second husband’s promise to secure his predecessor’s line of descent, by raising that man’s son and eventually providing him with a wife. This provision resembles the surrogate fatherhood included in the Taiwan contract; it also resembles a common feature of widow remarriage, in which the deceased husband’s son would accompany the widow into her new marriage, with guarantees of being raised to maturity and married, without being forced to take the second husband’s surname. The formality of this transaction reinforces its identity as a marriage that was seen as legitimate in the eyes of its participants and the local community. The formal elements include the use of a go-between to make the initial proposal; the engagement of a respected person of authority as matchmaker to negotiate and witness the terms; and the ritualized presentation of a gift to the first husband, to signify the second husband’s assumption of responsibility for his welfare. This case record does not specify whether the contract was written down; but either way, there is no questioning its formality or openness—indeed, the record suggests no sense of stigma on the part of any of the participants in this case.29 A case from Zhejiang reported in 1753 illustrates the use of a written contract for polyandry. Ma Shiyin (45), Lin Shi (31), and their son, Ma Ake (8), were landless peasants from Yongjia County in southeastern Zhejiang; poverty had driven them from home, and they ended up begging and sleeping in empty temples. Ma had acquired some sort of chronic illness, and they were having serious difficulty getting by. Just after the new year of 1753, they fell in with an itinerant fortune-teller, Mao Yuanfu (34), who had saved up several taels and was hoping to marry. Mao lent the family some money to buy rice, and they began migrating together and sharing meals. After two months, Ma proposed to formalize their relationship, yielding his wife to Mao in exchange for a promise of continuing support. Mao agreed, and though Lin Shi at first refused to cooperate, she soon conceded that they had no good alternative. Ma engaged a man they met on the road to write a “marriage contract for getting a husband to support a husband,” and he gave this document to Mao. That night, Lin Shi began sleeping with Mao, and the couple instructed their son to address Mao as “uncle” (shushu). From this time on, Mao covered all their expenses. Witnesses referred to this arrangement as “zhao fu yang fu” (getting a husband to support a husband), “zhao fu yang bing” (getting a husband to support an invalid [husband]), “zhao zhui yang bing” (getting an uxorilocal husband to support an invalid [husband]), and similar terms. In another six months, however, Mao had used up his savings, and Ma’s health had greatly improved. Ma began to talk about taking his wife back home to Yongjia

Chapter 1. “Getting a Husband to Support a Husband”

35

County and settling back into their old life, but Mao objected vehemently: “You already used Lin Shi to bring me into your family (ni yi ba Lin Shi zhao le wo)— how can you say you’re going to take her back?” Lin Shi told her husband privately that it would be impossible for them to leave Mao unless they could retract the marriage contract; so one day, when Mao was out telling fortunes, Lin Shi stole it from his pack and Ma burned it. The next day, Ma informed Mao that he and Lin Shi would depart the following morning; Mao angrily retorted: “You already ‘zhao’ed’ your wife to me (ni qizi zhao yu wo le), and I have a marriage contract to prove it, so if you want to take her back you should come up with the money to redeem (shu) her!” (Note here the use of “zhao” as a transitive verb with the wife as its object, as shorthand for “contracted your wife to me in a getting a husband to support a husband relationship.”) Mao then tried to produce the contract in order to cite its terms, discovered it was missing, and realized that the couple must have stolen it. The quarrel quickly turned into a fight, and Mao ended up killing Ma.30 Among other things, the latter case makes clear the powerful symbolic importance of the written contract in making an agreement valid and binding—even for illiterate people, and even when that transaction was prohibited by law. As long as Mao had their polyandry contract in his possession, Ma and Lin Shi felt they could not safely sever their connection with him—even though he could hardly take such a document to court, because there its only value would be as evidence of criminal “adultery.” We shall explore this paradox in detail with regard to contracts for wife sales in Chapter 4 below. But the potency of such documents clearly relates to their close resemblance to land sale contracts, which served as title deeds to sold land, and to their place in the common vernacular contract culture that unified prerevolutionary China. C HO SE N K I N SH I P A S A F R A M EWO R K F O R P O LYA N D RY

Investigation of Customs gives the impression that the dominant mode of polyandry involved a formally negotiated contract for the entire package of benefits to be exchanged. The legal cases certainly record many examples of such contractual agreements (both written and oral), as we have seen. But at least as common, it seems, were polyandrous relationships framed by chosen kinship, the most frequent scenario being sworn brotherhood between the two men. There are also many examples of the outside male pledging gan qin relations31 with one or more members of the couple’s household. If the couple had more than one son, then it was not unusual to have one of them pledge to the outside male (as a sort of godson) or even to be adopted by him. With sworn brotherhood, the outside male’s connection to the wife derived from his fraternal relationship with her husband. We find a paradigmatic example

36

part one

in a 1750 case from Tangyi County, Shandong, in which peasant Ding Bi (32) “pledged brotherhood” (bai xiongdi) with Yuan Congren (48) out of gratitude, because Yuan had helped him financially during an illness. Yuan was a single migrant who worked as a casual laborer. Ding Bi had no money to repay Yuan, and he felt acutely his inability to requite the other man’s charity. But Ding did have a wife—and, as Yuan later testified, I told him, like it was a joke, “I don’t want any reward from you. But we brothers are very close friends: if we share your wife, it will show that we’re really close to one another! (ni de xifu dajia huozhe, zhe cai shi xiang hao le).” At first, Ding Bi wouldn’t agree, but I often gave him a few hundred cash to spend, and when I got the chance, I would ask again [to sleep with her], so finally he agreed.

Material interests played a role here, but far more important, it seems, was Ding’s genuine gratitude. Yuan had offered unconditional aid, and this seemingly selfless act made Ding anxious to reciprocate. In the end, Ding seems to have agreed that true brotherhood meant sharing everything, even a wife. Eventually, the couple ended up living with Yuan, pooling resources, and farming together in another village, after Ding Bi’s uncle (shamed by their relationship with Yuan) forced them to leave home. In this way, polyandry framed by sworn brotherhood ended up supplanting “natural” kinship.32 Anthropologists traditionally characterize sworn brotherhood and other such arrangements as “fictive” kinship (also “symbolic,” “pseudo-,” etc.). But, as Kath Weston observes, such usage implies that the only real kinship consists of biological ties of blood and normative marriage organized around procreation. Her study of family formation among queer people in San Francisco argues that chosen kinship can create alliances that are far more meaningful and reliable than those based on the accident of birth, especially for those who have been rejected by (or separated from) birth families.33 Here, I follow Weston’s example in using the term “chosen kinship.” The emphasis on choice seems especially apt for late imperial China, where marriage was arranged between families without reference to the personal desires of bride and groom. In that world, chosen kinship relations such as sworn brotherhood functioned as surrogates for normative family; in addition, they might imply an alternative—a potential rival, even—to normative kinship, especially for individuals like Ding Bi, whose “natural” kin had rejected or failed to help them. Sworn Brotherhood in China: Historical and Anthropological Perspectives Previous historical studies of chosen kinship in China have focused on the large collective brotherhoods (often labeled “secret societies” or “Triads”) that were implicated in criminal entrepreneurship and rebellion. Scholarly interest originally focused on

Chapter 1. “Getting a Husband to Support a Husband”

37

the supposed revolutionary potential of such organizations, but later work emphasizes their similarity to mainstream social and religious practice.34 Historians agree that brotherhoods were “informal, popular institutions, created by marginalized men seeking mutual protection and mutual aid in a dangerous and competitive society.”35 Men who joined the Triads became part of “an extended family” with the “usual brotherly obligations,” including mutual support, financial aid for marriage and mourning, and food and shelter for fugitives from the law.36 Similarly, bandit gangs were bound by vows of brotherhood and organized according to fraternal hierarchy, which served the young “bare sticks” who typically joined them as a “surrogate family.”37 According to David Ownby, the Triads first appeared in the historical record in the early eighteenth century, when the devastation of dynastic transition, followed by mounting demographic pressure in core regions, fostered the marginalization and migration of unprecedented numbers of young men. They “founded brotherhood associations to seek mutual assistance and protection in a precarious world.”38 The same sorts of men often played the role of the outside male brought into polyandrous arrangements through sworn brotherhood. The same factors that fueled the proliferation of collective brotherhoods also helped foster these more intimate patterns of alliance on a smaller scale. In both versions, the appeal was much the same: mutual support and access to resources, to compensate for a lack of “natural” kinship and community. Several anthropologists have made useful observations about collective sworn brotherhoods based on fieldwork in Taiwan. Bernard and Rita Gallin define a sworn brotherhood as “an alliance of theoretically equal partners who utilize friendship for the purpose of security and potential advantage.” They posit two ideal types: an “affective” brotherhood is based on unselfish friendship, whereas an “instrumental” brotherhood is “a manipulative grouping” based on “economic and sociopolitical gain.” In practice, most groups combined affective and instrumental dimensions in varying proportions.39 David Jordan asks, “Why is it better to be a Chinese brother than a Chinese friend?”40 According to his informants, the rubric of brotherhood provides three specific advantages that are pertinent here. First, economic aid “is uniformly and eagerly described as inherent in the relationship, and this is often given as one of the axes of difference from relationships of friendship,” because if you are helping your “brother,” you are not helping an outsider at the expense of family. In fact, help in time of need becomes “compulsory.”41 Second, the ideology of brotherhood mutes conflict, because competition and exploitation are prohibited between brothers, and they will avoid open disagreement. Third, “sworn brothers assume an obligation towards the family members of their fictive brethren”—in effect, sworn brotherhood “allows associated families to treat selected outsiders as insiders.”42 These features of sworn brotherhood help explain why it would have been an effective framework for polyandry.

38

part one

Avron Boretz has analyzed sworn brotherhood among marginalized males on the fringes of the underworld, based on fieldwork in Taiwan and Yunnan. A key value in that milieu is yiqi (“honor” or a “righteous spirit,” characterized by “selfless generosity”): “Honor entails self-limitation; sacrifice; and a postponement of personal interest, need, or desire. . . . Yet honor is nevertheless a form of reciprocity that entails social and material, as well as moral, obligation. . . . A righteous act on one’s behalf raises the stakes of the relationship and incurs an obligation to respond.”43 What Boretz characterizes as “the mutual loyalty and obligatory generosity of brotherhood” specifically requires an open hand with money: “A man of prowess is big-hearted and generous; stinginess is by itself sufficient cause for ostracism.”44 And yet, Boretz emphasizes, “cash flow is a focus of anxiety” for his marginalized informants, “given their unstable sources of income and the expectation that whenever flush, a man should be generous with his sworn brothers.”45 Paradoxically, this form of honor requires an insouciant and even reckless generosity that defies insecurity. Another feature of sworn brotherhood is a pretense of indifference or contempt toward women. Boretz identifies the underlying cause of such misogyny as “women’s inherently destabilizing role in the exogamous, virilocal, patrilineal Chinese family”: [T]he patrilineally organized household and the male-dominated jianghu share the tendency to regard women as both highly valuable, hard-to-obtain commodities and (for that very reason) threats to male solidarity—the organizing principle and basis for the identity of the institution itself.46

To illustrate, Boretz describes the conduct of his informants at a hostess bar: One pays direct attention here primarily to one’s friends, sworn brothers, comrades—a man among men, it would be unmanly to allow oneself to be overly distracted by the feminine charms of the “drinking partners.” Such unmanly behavior is subject to derision—in putting one’s own sexual desire above male solidarity (zhong se qing you), one is displaying a pathetic weakness.47

This pretense of indifference toward women would reinforce bonds between brothers. These values are strongly associated with the marginalized hero of folklore and fiction known as the haohan (“good fellow”), who is typically portrayed as immune to the wiles and temptations of women, and also as casual and open-handed with money. As Boretz explains, “The stereotypical righteous man of martial prowess . . . is thoroughly physical in every way but the sexual. Where haohan had prodigious appetites for meat, wine, and combat, they are abstemious and circumspect when it comes to relations with females.” Nevertheless, “resisting temptation entails the repression, not the absence, of desire.”48 Literary scholar Kam Louie

Chapter 1. “Getting a Husband to Support a Husband”

39

emphasizes the repressed sexuality of the haohan’s “martial” (wu) masculinity: his sexual puritanism serves to sublimate and mask both the homoerotic desire between sworn brothers and the incestuous desire one might feel toward another’s wife. On occasion, sexual repression might erupt in extraordinary displays of misogynist violence. In a famous example from the novel Shuihu zhuan (Outlaws of the Marsh), the hero Wu Song rejects the sexual advances of his sister-in-law Pan Jinlian, and later he savagely murders her to avenge his brother’s death.49 We should bear in mind that such fiction was written by literati for a mainly elite readership that had little in common with the marginalized men who actually swore brotherhood as a survival strategy. To the extent that authors comprehended wu masculinity, they did so as outsiders, and the sexual repression of the haohan probably reveals more about the tensions of literati masculinity than about the social practice of sworn brotherhood. Real life was not necessarily so repressed: many sworn brothers actually did have sex with each other and/or with each other’s wives. As far as the putative incest taboo is concerned, we should remember that levirate remarriage—in which a single man would inherit his brother’s widow by remarriage—was far from unknown. The practice was known as zhuan fang: literally, “rotating a wife,” “bedrooms,” or “family branches.”50 This form of remarriage was a variation on fraternal polyandry, since it typically occurred in poor families that could afford a wife for only one son and therefore sought to recycle her if he died (hence the idea of “rotating”). Therefore, even though the sort of fraternal polyandry common in the Himalayas was not practiced by Han Chinese, it was by no means unthinkable for a Chinese man to marry his brother’s wife. Ming and Qing law criminalized levirate as a form of incest punishable by strangulation, but what was taboo for the judiciary and elite was not necessarily so for ordinary people who had other concerns.51 As the case of Yuan Congren and Ding Bi illustrates, we find the same sense of masculine honor described by Boretz within polyandry framed by sworn brotherhood. The apparent premise of these alliances was that the bond between brothers outweighed that between husband and wife, becoming the primary alliance within the triangular relationship. Brothers were supposed to share everything, hence the husband would act as if he thought nothing of sharing his wife with his sworn brother. Insofar as poverty-driven survival strategies like polyandry compromised the masculinity and authority of husbands, sworn brotherhood provided an alternative masculine paradigm that sought to compensate them through an idealized fraternal solidarity grounded in misogyny. Nevertheless, many cases show that such relations were not always free of selfishness and jealousy—although, to be sure, such vices are overrepresented in legal cases. None of the historians or anthropologists cited above mention access to sexual resources as a feature of sworn brotherhood. But David Schak’s account of a beggar

40

part one

community in Taipei records an example of polyandry within sworn brotherhood that closely resembles the relationships found in Qing legal cases. Chosen kinship was common among the beggars, because “being people without extensive real kinship ties or realistic expectations of as much assistance as they need from their relatives, they have used fictive kinship to increase the number of allies they have and the number of people who have obligations to them.”52 The beggar leader, Tiek-kou, had several sworn brothers, one of whom, A-iu:, was a demobilized soldier from the mainland. Tiek-kou’s wife built up a heavy gambling debt to A-iu:, and arrangements were made to pay off the debt by granting him conjugal privileges. Tiek-kou had no objections to this arrangement. In fact he was quite willing to go along with it. Not only was it an inexpensive way out of the debt, but it also demonstrated his magnanimity.53

In this way, the two men became “co-husbands” in addition to sworn brothers, and this arrangement lasted for several years, until A-iu:’s death. The initial motive for polyandry was the need to pay a debt, but a strong bond clearly existed between the two men, who continued to share Tiek-kou’s wife and to pool resources long after the debt had been cleared. Tiek-kou also “pledged” the first son of his third daughter to A-iu:. After this child was born, he was given A-iu:’s surname and registered at the District Office as his son. In this way, polyandry also secured the outside male’s line of descent.54 Tiek-kou’s “magnanimity” in both sharing his wife and pledging his grandson to A-iu: is another example of the “selfless generosity” that characterized honor between sworn brothers. One can easily imagine how the gambling debt might have provoked conflict between the two men, but instead they hewed to a higher ideal of brotherhood ultimately based on relative contempt for women; as one of Boretz’s informants explained, “you can’t let a woman cause a split among brothers.”55 It appears that Tiek-kou welcomed the opportunity to display his “magnanimity” in this manner, which brought him honor rather than shame within his own milieu.56 Affective and Instrumental Dimensions of Sworn Brotherhood Let us now turn to legal cases in which chosen kinship provided the framework for polyandry. In thinking about these cases, the Gallins’ distinction between affective and instrumental dimensions of sworn brotherhood is useful. In some relationships, the affective dimension clearly took precedence, at least at first. But all parties understood their instrumental potential, and relationships that began out of friendship could acquire more materialistic functions as needs and desires changed. The reciprocity implied by resource sharing might become the framework for a quid pro quo whereby sex was traded for economic support. The genius

Chapter 1. “Getting a Husband to Support a Husband”

41

of chosen kinship was that it had the flexibility to encompass a wide spectrum of purposes. In a case from Chenggu County, Shaanxi, reported in 1736, tenant farmer Wang Xiao (23) befriended a landless laborer Heng Quanlian (27). Heng came from a village some thirty kilometers distant, but he had been peddling wine throughout the area, and that is how he and Wang became acquainted. In 1729, the two men “acknowledged each other as brothers” (ren le xiongdi). According to Heng’s wife, Ren Shi (20), the two men treated each other “as if they were brothers of the same family” (dang ziji qin xiongdi yiyang). There is no evidence that their relationship had any specifically instrumental intent when it began. But, after three years, Wang agreed to alleviate his sworn brother’s worsening poverty by subletting part of his own farm to him. Heng and Ren Shi then moved in with Wang, who was single and lived alone. The couple’s only possessions were a few old pots and a single worn blanket—they had no money or grain—so Wang fed them and lent them money to help them get started; also, significantly, Wang paid a debt that Heng owed from having borrowed the brideprice to take Ren Shi in marriage. Payment of this debt likely gave Wang a sense of entitlement with regard to Heng’s wife. Heng deeply appreciated this charity, and he made it clear to Wang that he was welcome to share Ren Shi—he had them drink wine together, telling them that since they were “brother and sister-in-law,” there was no need to “avoid” one another. After a couple weeks of this encouragement, Ren Shi and Wang began sleeping together. Heng did not mind; as he told them, “you don’t need to try to hide it from me, as long as you make sure to take good care of me.” Wang’s motive in helping the couple may not have been purely altruistic, but only with the beginning of sexual relations, after several years of sworn brotherhood, do we find the clear articulation of a quid pro quo. This relationship was quite open. Heng’s younger brother and father-in-law both scolded him for letting “a single man” (dan shen Han) live on such intimate terms with his young wife, but Heng told them to mind their own business. Heng and Wang introduced each other to outsiders as “my good sworn brother” (bai ren de hao dixiong), explaining that “we are pledged in kinship as one family (ganqin jia).” Everyone understood the relationship for what it was: here chosen kinship served not to disguise polyandry but rather to provide a framework for understanding and talking about it.57 In a 1758 example from Ning Department, Jiangxi, we again see economic aid as a logical extension of sworn brotherhood, followed only later by sexual relations. Peasant Wang Changming (32) and his wife, Lin Shi (21), lived with their infant daughter and Wang’s elderly, invalid parents (note the high ratio of dependents to laborers in this household). Wang became good friends with a single immigrant from Guangdong, Chen Manqi (22), and the two men swore brotherhood.

42

part one

After this, Chen often came to visit Wang’s family, bringing wine and food to treat them; Wang was very poor, and he began borrowing money and food from Chen whenever he ran short. Wang could never repay these debts, but Chen never demanded repayment and never failed to help, and Wang’s whole family felt deeply grateful. According to Wang’s father, “my son treated him as if he were his own brother by blood.” According to Lin Shi, “my father and mother-in-law and husband all treated him as if he was our closest relative,” and they all instructed her to be friendly and “not to avoid him” (buyao bi ji). Soon, Chen found a moment alone with Lin Shi and made a pass at her; she rebuffed him and later complained to her husband and parents-in-law, but they assured her there was nothing to worry about: the important thing was that, no matter what Chen did, she should take care not to offend him, because the family needed his assistance to survive. When she asked her father-in-law how they would ever repay what they had borrowed from Chen, he simply assured her that “as long as you don’t offend him, there’s nothing to fear on that count.” Obviously, the repeated injunctions not to “offend” Chen implied that Lin Shi should sleep with him. She understood this message, and, the next time he visited, she accepted his advances, and they began having sex regularly. The family’s desire was to bind Chen closely to them by reinforcing the chosen kinship tie with a sexual relationship, but they also feared that refusal of sex might drive Chen away. Moreover, even if Wang and Chen never explicitly negotiated a contract, both sides had come to a clear understanding that Lin Shi’s sexual services would offset Chen’s “loans” of food and cash, and that as long as she slept with him, he would continue to help. The instrumental potential of sworn brotherhood was there from the beginning, even if latent, so that what began as Chen’s brotherly extension of aid transformed seamlessly into an exchange of sex for material support. As Lin Shi later testified, “because we were so poor, my father and motherin-law and husband all wanted to use me to sell illicit sex (mai jian).” It seems significant that Lin Shi here eschewed euphemism—from her angle, the transaction appeared quite straightforward.58 Sometimes, a husband swore brotherhood with another man as a self-conscious, intentional strategy to gain access to his resources; even then, however, the affective aspect of the relationship cannot be dismissed out of hand. For example, Zhu Gan (32) and his wife, Liu Shi (29), were peasants in He Subprefecture, Sichuan; they had no land, and Zhu worked as a casual laborer. They had two small children (a third had already died), and the family was too big for Zhu to feed on his own. In 1740, Zhu got to know Wang Hu (34), a single migrant from Guizhou who did odd jobs as an agricultural laborer, as a musician in funerals, and occasionally as a peddler. When Zhu saw that Wang had ready cash, he proposed that they swear brotherhood. Wang agreed because, as he later recalled, “I was on my own, far from home, and it was a good idea to have a friend. So we pledged and became brothers.”

Chapter 1. “Getting a Husband to Support a Husband”

43

That winter, Zhu borrowed 3 taels from Wang Hu to buy food. Zhu knew that he could not repay the loan, so he invited Wang to spend the night and sent Liu Shi to his bed. At first Wang was alarmed, but after she explained that this was her husband’s way to repay the debt, he relaxed, and they had sex. Thereafter, Wang would work at various jobs, and when finished would come stay with the couple, turning his pay over to them and sharing Liu Shi’s bed. In this way, he became part of the family; as several witnesses testified, “they did not distinguish between inner and outer” (bu fen neiwai). Zhu Gan’s older brother, Zhu Ming, observed Wang Hu living with the couple, and he was not happy. Zhu Ming later testified, I asked my brother, “Who is this guy? Why do you let him sleep over at your house, and let him come and go as he pleases, without distinguishing inner from outer?” My brother said, “His name is Wang Hu, he’s from Zunyi Prefecture, and he’s my good brother who’s pledged brotherhood with me. So there’s no need to treat him as an outsider (mei you fen shenme neiwai de).”

After six months, Wang persuaded the couple to move about a hundred kilometers west to Anle County, where they lived together and worked land belonging to a relative of his.59 The exchange between Zhu Gan and his brother is revealing: Zhu Ming clearly knew what was going on, and Zhu Gan did not deny it. Instead, he justified it in terms of sworn brotherhood. As Yuan Congren asserted in the first case narrated above, sworn brothers needed no boundaries, because they shared everything. A 1763 case from Xuyi County, Anhui, involving a three-year polyandrous relationship, provides explicit articulation of the distinct affective and instrumental dimensions of the fraternal bond. The outside male in this case, Tan Si (41), was a brutal man who had initiated sexual relations by forcing himself on the wife, Wang Shi (27), while her husband, Gao Shizhong (27), was away. Subsequently, however, Tan feigned a close and friendly fraternal relationship with Gao Shizhong, while also providing the couple with substantial material aid, in order to facilitate his continued sexual access to Wan Shi. Gao was happy to cooperate (Wang Shi much less so, though after she had been raped she no longer resisted), and he appears to be a clear example of a man who placed higher priority on his fraternal bond than on his bond with his own wife. In the end, however, Tan Si murdered Gao in an attempt to take Wang Shi for himself. When asked by the magistrate why Gao Shizhong had let him continue sleeping with Wang Shi, Tan Si dismissed Gao as a pathetic alcoholic who could not make ends meet: “Gao Shizhong was bitterly poor and could not earn a living, and also he really liked to drink wine. He depended on my support, and that’s why I could have illicit sex with his wife—he let me do whatever I wanted.” But Wang Shi insisted that her husband had been motivated by friendship:

44

part one My husband was not greedy for Tan Si’s money, and he could have beaten Tan Si in a fight if he had to. He did not let Tan Si sleep over and have illicit sex with me out of greed or fear. It’s true that my husband loved to drink wine and was foolish, but he was also a loyal and sincere man (wei ren zhonghou). Tan Si was his friend and they often drank wine together. That is why my husband allowed him to sleep over and have illicit sex with me.

Wang Shi’s account recalls the “magnanimity” displayed by the beggar leader Tiekkou in Schak’s ethnography. Wang Shi had not known that Tan Si had murdered her husband, but, when the facts were exposed, she confessed the details of their relationship and begged the magistrate to avenge her husband’s death. Her emphasis of her husband’s “loyalty and sincerity” underscored the cruel cynicism of Tan Si’s betrayal.60 Sharing Work Leads to Polyandry A pattern seen in a number of cases is that two men who found it economical and convenient to work together would end up pooling resources, forming chosen kinship bonds, and eventually sharing a wife. Each step seems to have led logically to the next. In a 1743 case from Ju Department, Shandong, two itinerant musicians found that they could make more money performing together than separately. One of the men, Yang Lun (36), was a peasant whose farm in Lanshan County had been wiped out by a flood, making him and his wife, Zhang Shi (31), refugees; Yang was able to make ends meet, barely, by singing. The other man, Zhou Qilong (31), seems to have been the more accomplished musician—he could play wind instruments as well as sing—but he was single. The two men began performing together and soon swore brotherhood; after some months, they decided to live together, and before long Zhou began having sexual relations with Zhang Shi. The whole relationship seems to have unfolded as a logical result of the two men’s decision to work together.61 In a second example, from 1753, a pair of itinerant barbers, Chen Ying (26) and Yu Hualong (26), met at the department seat of Guang Independent Department, Henan, where both had traveled to take advantage of the extra business likely because of the civil examinations then being held. The two hit it off and became friends, setting up their barber stalls together; one day they drank wine together and “pledged brotherhood,” and then Chen Ying invited Yu to join Chen and his wife and “eat from the same pot.” Yu agreed, and he began contributing his earnings to the common fund. He also lusted after Chen’s wife, Wang Shi (23), who is described in the case record as “young and attractive,” and since the feeling was mutual, the two soon began having sex. As Wang Shi later testified, “my husband obviously knew that I was having illicit sex with Yu Hualong, but because Yu Hualong shared all his earnings with my family to use together, my husband never interfered, and just let me do as I liked.”62

Chapter 1. “Getting a Husband to Support a Husband”

45

In a 1754 case from Lushan County, Hunan, a polyandrous household began to take shape when three carpenters began working together. Cui Huaibang (54) was married to Zhao Shi (52), whose son from a previous marriage, Mao’er (22), was also a carpenter; the couple also had a daughter (12) of their own. Cui became close friends with another carpenter, Zhang Fu (40), who was single and lived alone, and soon he invited Zhang to move in with his family, pool resources, and work together, reasoning that they would all be better off that way. Zhang agreed, and soon he began having sexual relations with Zhao Shi. She became very fond of him, and, with Cui’s permission, she adopted out her son Mao’er to Zhang. Cui did not object to the sexual relationship, since Zhang (in his own words) “was earning money to support his family”—indeed, as in many similar cases, the sexual relationship was a logical consequence of the other aspects of the relationship that had been negotiated and seems to have been implicit in the invitation to live together.63 A common theme of these various scenarios is the sense of masculine honor peculiar to sworn brotherhood, which eliminated boundaries between “inner” and “outer” and mandated an open-handed generosity that subordinated individual interests to fraternal solidarity. Brothers were supposed to pretend indifference toward money and women, even though both were scarce and highly valued resources. The basis for polyandry was a sharing of material resources and wives in an exchange that strengthened the primary alliance, that between men. Adoption as a Strategy to Bind the Outside Male to the Household As the last example above illustrates, sworn brotherhood was not the only form of chosen kinship used to frame polyandrous relationships. Another form seen in many cases is adoption: to bind an outside male to their family, a couple would offer to “adopt” (bao) or “pledge” (bai) one of their sons to him, thereby providing him with a son and successor of his own. This offer held powerful appeal, because for single men of this kind, a son was a supremely desirable but unobtainable good, far rarer and more precious than mere sex. Hence, as we saw in Schak’s anecdote about the Taiwan beggar community, the ultimate act of magnanimity was Tiek-kou’s gift of his own grandson to be his sworn brother’s son. In a case from Fuyang County, Anhui, reported in 1745, Liang Chou and his wife, Zhao Shi, were landless peasants who had one young son. A local man named Zhao Fuyue, who was single, lusted after Zhao Shi, and, hoping to gain access to her, he offered to help the family financially. He gave Liang 600 cash and later another 1500 cash; in return, Liang had their son “pledge” Zhao Fuyue as his “godfather” (ren zuo gan laozi), and in exchange for Fuyue’s agreement to continue taking care of the family, Liang also “promised” (xu) his wife to Fuyue. Both sides agreed to these terms, and so Fuyue moved in with the family, pooling resources and sleeping with Zhao Shi. These negotiations resemble the formally contracted polyandry discussed earlier in this chapter, but in this instance a verbal contract

46

part one

for “getting a husband to support a husband” was reinforced by chosen kinship. The pledge of kinship was meant to repay Fuyue’s financial assistance and to bind him to the family, while the promise of sex was intended to ensure that such assistance would continue. Liang Chou may have felt genuine gratitude toward Zhao Fuyue, but the instrumental nature of their verbal contract could hardly have been more explicit.64 We find a similar situation in a 1758 case from Shehong County, Sichuan. Xiang Gongping (in his thirties) had been a tenant farmer, but he lost his lease because of illness, compelling him, his wife, Wang Shi (33), and their three youngest sons to leave home and try to survive by begging. After two months, they were squatting at an abandoned temple and starving when they encountered a mendicant fortune-teller named Wang Xueyi (24). Wang noticed the children crying from hunger, and he offered the family two bowls of rice and beans. Profoundly grateful—but also desperate for further aid—Xiang praised Wang as a good man and invited him to join the family and travel and beg together; as inducement, Xiang made his youngest son pledge himself as Wang’s adoptive son and heir (jizi). Wang was so moved by this gesture that he agreed to Xiang’s proposal and immediately turned over his meager savings (some three pints of rice and over 100 cash) to Xiang for common use. Thus far, there had been no mention of sex, but after a few days together Xiang quietly told his wife to go sleep with Wang, in order “to bind his body to us” (ban zhu ta shenzi):65 “We have too many mouths to feed and no way to earn enough money to get by. But by telling fortunes, Wang Xueyi can earn a lot of beans and rice in a single day. We’d better keep him with us, so that we can all survive.” At first she refused, but her husband kept urging her, so finally she obeyed and went to Wang’s bedroll. Wang asked whether her husband would object, but she told him, “it’s my husband who told me to come here.” From that night on, she slept with Wang, while her husband and sons slept separately.66 In both of these cases, the husband took the initiative to “bind” a potential patron to his family by offering precisely what that man had little hope of otherwise obtaining: a son and a wife. Many examples from the archives demonstrate that these were powerful inducements to a single man outside the normative family system. The chosen kinship bond played a critical role in cementing such relationships, in making them real to the people on both sides of the bargain. Obviously, there was an instrumental quid pro quo from the beginning, but that did not prevent people from feeling a powerful emotional investment in the kinship bond thus created. A case reported in 1740 by the governor of Guangxi involved migrants in a remote frontier zone. In this case, Su Shi and her husband, Long Shengcai, were landless peasants who had abandoned their home village to farm hillside wasteland. After farming one spot for eight years, the land was exhausted, so they

Chapter 1. “Getting a Husband to Support a Husband”

47

migrated to a more remote site in Lingchuan County, where they opened new land to plant indigo, selling the harvest for grain. The couple had two small sons (one of whom they hired out) and were very poor. Their only nearby neighbor was a migrant male named Su Honglou (32 when they met him) who lived and farmed alone. He and the couple quickly became good friends and began trading work; since he was lonely, he spent a great deal of time visiting them. Since Su Honglou and Su Shi (who was then 35) had the same surname, Honglou “recognized” (ren) her as his “gan guniang” (in this context, guniang probably means “paternal aunt”). After six months, Honglou’s straw shack collapsed in a storm, so the couple invited him to move in with them; he accepted, and from then on (in Honglou’s words) “we lived together and recognized one another as family” and “pooled all earnings for common use.” Another seven months went by before Honglou and Su Shi began having sex. Long Shengcai did not object, telling them that as long as Honglou would help support Su Shi, he was welcome to sleep with her. From that time on (in Honglou’s words), “Su Shi and I slept together openly and didn’t worry about it, and he didn’t interfere, just as he had promised.” In this relationship, the affective and instrumental dimensions blend seamlessly: isolation and poverty drew the principals together in common struggle for survival on the frontier, and the accident of common surname only made it easier to cement their relationship with a pledge of kinship. The sexual dimension was the last piece of the relationship to fall into place, but to all involved it seems to have been a perfectly logical development, indeed one that may have been implied when the couple invited Honglou to move in with them. In this case, the sexual quid pro quo aimed to bind a single man all the more tightly to a family that needed his labor. It worked: in Honglou’s words, when Su Shi confirmed that her husband approved their relationship, “I simply worked as hard as I could at farming, and when I sold the indigo I wouldn’t keep even the slightest bit of money, instead I gave it all to Su Shi for the family’s support.”67 To sum up, a variety of scenarios might lead to polyandry. Despite this diversity, it is striking that hardly anyone evinced surprise when a sexual relationship eventuated. Clearly, there was a widespread, commonsense understanding of how these relationships were created and the principles on which they were based. A Qing Magistrate’s Insights about Sworn Brotherhood Unusual insight into the potential of chosen kinship for framing non-normative sexual relationships comes from Ji Qiguang, who served as founding magistrate of Zhuluo County, Taiwan, in the late seventeenth century. In an essay recorded in the 1720 gazetteer of then Taiwan County, Magistrate Ji denounced the “evil custom” (esu) of sworn brotherhood. As a frontier full of unruly male migrants, Taiwan was notorious for mafia-like collective brotherhoods and the political

48

part one

threat they posed.68 But, instead, Ji’s denunciation targeted the more prosaic bonds that united pairs of men. In Ji Qiguang’s view, “sworn brotherhood brings people together for illicit sexual relations” (meng zhe jian zhi mei ye): If A and B join together in brotherhood, then A’s mother becomes B’s mother, B’s younger sister becomes A’s younger sister, A’s wife becomes B’s elder sister-in-law (dasao), and B’s wife becomes A’s younger sister-in-law (dixi). After all, if the families keep a proper distance from one another, there is no point in this man becoming “brother” to that. Moreover, the dwellings of petty urbanites and peasants [i.e., the kind of people likely to swear brotherhood] lack the number of stories and rooms necessary to separate inner from outer in a proper manner. In such circumstances, sworn brothers will roam freely in one another’s homes as if they were each other’s true flesh and blood. The inevitable danger is that men and women will flirt with each other, resulting in intimate contact, and in the end all inhibition will be lost (gua li zhi xian bufu guji). Under the circumstances, how many could resist temptation (shei ru bingzhu zhi tianshen)? Would any man refrain from taking advantage of such easy opportunity for illicit conduct (shei shi zuo huai zhi nanzi)? There will be men who climb into “mother’s” bed, or who raise the curtain of “younger sister’s” boudoir—there will be men who pay lip service to a woman’s status as “elder sisterin-law” or “younger sister-in-law,” while in fact pursuing casual liaisons with those who share their desires.

Ji concludes that men who swear brotherhood cannot be trusted. “At first they may restrain themselves, but gradually they lose their inhibitions. When jealous competition over women leads to murder, the root cause is always this pretense of ‘brotherhood.’ ”69 As Ji Qiguang clearly understood, far from imposing any sort of incest taboo, the act of swearing brotherhood actually eliminated the boundary between “inner” and “outer” that was supposed to secure female chastity and household resources from outside threats. He portrays all this in a highly pejorative light, but his basic insight was absolutely correct, and it is possible that his comments on jealousy leading to murder reflect his experience as a magistrate having to deal with the sort of violence that we find in xingke tiben. In the next passage of his statement, Ji argues that sworn brotherhood tends to impoverish people, because the erasure of kinship boundaries gives unrelated parties the opportunity to plunder each other’s assets. Here too, if we look past his pejorative gloss, we can see that he is confirming what our other sources show: that sworn brotherhood provided an acceptable framework for mutual assistance and pooling of resources.70 Many legal cases show that fraternal solidarity was not always capable of containing the tensions created by polyandry, and the result was the trouble and violence that ended up documented in xingke tiben. One common scenario is that the relationship between the wife and the outside male would become the primary

Chapter 1. “Getting a Husband to Support a Husband”

49

alliance, with the husband increasingly sidelined. Another is that the married couple would try to sever their relations with the outside male after solving their immediate difficulties. A common theme is that women had minds of their own, and their feelings and views could not easily be ignored, irrespective of the misogynist contempt that was supposed to unite sworn brothers. In practice, if the wife preferred one man or developed antipathy toward the other, it might not matter how the men felt about each other. The prominence of chosen kinship in polyandry cases points to the fundamental definition of the family-household unit (jia) in Chinese society as a group of people who live together, pool resources, and eat from the same hearth. (Thus, household division between brothers traditionally culminated in a final ritual meal together, followed by the establishment of a separate cooking hearth by each sister-in-law for her own newly separate household.)71 Usually these factors coincided with the traditional kinship connections of blood and marriage, but they also seem to have helped define kinship.72 The pledging of kinship in polyandrous relationships ratified the facts of co-residence, resource pooling, and wife sharing. It was a strategy to reinforce trust and dependence. Q U E ST IO N S

Poverty as a Motive for Polyandry The principal motive for couples to recruit an outside male was poverty. Even when the wife’s adultery preceded economic support and was motivated by passion (there are many examples of this scenario), these were invariably poor people, and it was the promise of food, cash, or other resources that would win the husband’s acceptance of the relationship. For this reason, polyandry’s incidence and regional distribution probably correlated with those of poverty as well as with high sex ratios.73 Poverty was sometimes compounded by an illness or disability that prevented the husband from working, so that the family felt compelled to seek the help of another man. Investigation of Customs confirms this pattern. In Hubei, for example, “a woman whose husband is handicapped, so that he cannot make a living, will be allowed to bring in a second husband to support the first (ling zhao yi hou fu, yi fuyang qian fu).”74 When the husband’s illness or disability was the main motive for polyandry, the arrangement was often called “getting a husband to support an ill husband.” Poverty distinguishes the non-fraternal polyandry found in China from the socially acceptable, fraternal polyandry traditionally practiced in Himalayan societies. In the Himalayas, reasonably well-off families practiced polyandry in order to prevent household division and to concentrate and preserve their property across generations. It also tended to limit the birth rate in such families, since one woman

50

part one

can get pregnant only so often (whereas polygyny would tend to have the opposite effect). This important social benefit was reinforced by the existence of large establishments of celibate clergy. The relatively low population density that resulted tended to protect the delicate ecology of that high-altitude region.75 The situation in China was completely different: in China, the illegal polyandry practiced among the very poor was an inverted reflection of the legal polygyny practiced by elite men. Here, it is useful to recall the argument of Russian economist Aleksandr Chayanov that the fortunes of a peasant family would change in a cyclical pattern, depending on the ages of family members and the ratio of laborers to mouths that had to be fed. (Women should be counted as laborers, not mere consumers.) According to Chayanov, a young couple with small children would be the most vulnerable to hardship, and, in the majority of cases I have seen, it was precisely such couples who resorted to polyandry in order to survive. Sometimes there were elderly dependents as well as children.76 Of course, the rogue males who typically allied with these families were poor, too. But a single man had to feed only himself, so it was often possible for him to earn a little extra, as long as he could find work. Most of these men were simple peasants or laborers, but some had special skills by which they could earn a humble living: Qing case records include storytellers, fortune-tellers, thieves, carpenters, musicians, barbers, many peddlers and clergy, and even teachers. In some cases, the single man was simply the one with the wits and the initiative to keep the family fed. For example, in 1745 the governor general of Zhili reported a case of “one wife with two husbands” (yi fu liang fu) in which a “propertyless single man” (guang shen Han) named Zhang Liang had employed a variety of means to support a peasant couple and their two children for over a decade. When husband Dong Si (from Xingtai County) first invited Zhang Liang to share his wife, Zhang Shi, in exchange for supporting the family, Zhang Liang was working as a casual laborer; after he moved in with the family, he managed to get together enough cash to start a little business peddling cooked food, which the three adults worked at together. Several years later, when they ran short on funds, Zhang fed them by means of theft—eventually being beaten and tattooed for that crime—and later on, he supported the family for another year or two by peddling tobacco. Finally, he arranged for the entire family to move to a busy market town in Yongnian County, where they again peddled food and Zhang did odd jobs.77 Whatever his resources, a single able-bodied man could help alleviate the distress of a poor family with whom he allied. In Chayanovian terms, his addition to the family would improve the ratio of laborers to consumers. In exchange, he would benefit from the sorts of family life and female caring labor he otherwise could not hope to enjoy. How poor was “poor”? In some cases, families faced starvation or, at best, the prospect of breaking up in order to survive (with the husband selling off his wife

Chapter 1. “Getting a Husband to Support a Husband”

51

and children, one by one). Here, bringing in an outside male was a strategy not just to feed the family but also to preserve it as a family. In other cases, the situation was not so dire, but couples were poor enough to welcome the support a second man could provide. Poverty in and of itself explains only so much—after all, most people in Qing dynasty China were poor to one degree or another, but not all engaged in polyandry. Poverty constrained people’s options but did not necessarily determine which option they chose to pursue. Moreover, although wife selling and polyandry were solutions to similar problems, they had dramatically different effects on the people involved: polyandry expanded a family in order to keep it together, whereas wife sale broke up one family while laying the foundation for another by creating a new marriage. To understand why specific people pursued one strategy or another, we must look to the personalities and preferences of the individuals involved and the nature of the relationships between them as well as the circumstances in which they found themselves. Polyandry and the Chinese Marriage System Twenty-five years ago, when I first encountered in Qing legal cases the kinds of practices described in this chapter, I had no idea what to make of them. They certainly did not fit my image of “Chinese marriage.” Based on what I had read, I knew that, among the wealthy, polygyny was the rule: a main wife (qi), of the same social background as her husband, would be supplemented by concubines (qie) and maidservants (bi) purchased from less-well-off families. Among the broad majority of peasants, however, monogamy prevailed (although there was considerable variation).78 Dowry was a status symbol, and the daughters of more prosperous families incorporated the bridal gifts into the dowry they took with them to their husbands’ households.79 Among less prosperous families, many daughters were more or less sold for brideprice, although this cash transaction might be masked by face-saving gestures. I also knew that there were many surplus males at the bottom of society with no wives at all, but there was very little information about their lives. The evidence shows that, in addition to polygyny and monogamy, the Chinese system for exchange of women included a third pattern, polyandry, which was practiced by a significant section of the poor. The unifying theme of these diverse strategies is that a wife would have sex with one or more men other than her husband, with his approval, in order to help support her family. Polyandry and related practices depended on what officials saw as the sexual promiscuity of a small number of women with a larger number of men. It was an ironic mirror image of the female chastity that underpinned the polygyny enjoyed as a status symbol by elite males. To the official mind, this was sexual anarchy, a dangerous trend with politically subversive implications.80 It was this perspective

52

part one

that misrecognized custom as crime, polyandry as adultery, and survival strategies as sensual license. But—some may protest—if this behavior was illegal and frowned upon, how can it be considered part of traditional Chinese custom? This objection only begs another, more fundamental, question: what counts as “custom,” and who gets to decide?81 “Custom” (fengsu xiguan) has a positive connotation of normative behavior, especially in modern Chinese: it implies a value judgment, behind which lies a power structure. Therefore, although “delayed transfer” marriage was the norm in much of the Pearl River Delta, gentry stigmatized the practice, and officials tried to suppress it.82 Levirate was widespread, but Ming and Qing law criminalized it as a capital offense. More fundamentally, the celebration of female chastity combined with legal protection of polygyny was an ideological program that protected the interests and privileges of elite men. When a rich man who already had a wife bought a second woman to be his concubine, it counted as marriage, but when a poor woman with her husband’s approval contracted an uxorilocal marriage with a second man, that counted as adultery. Because “custom” has a positive connotation, people may resist applying this label to practices, however widespread, of which they are ashamed. In some Taiwan villages, “minor” marriage (the adoption of a daughter-in-law in early childhood) was the universal practice, yet villagers, when asked about “Chinese marriage” by a Western anthropologist, would invariably describe the “major” form, even though none of them practiced it themselves. Moreover, men involved in uxorilocal marriages often tried to conceal that fact from outsiders in order to avoid ridicule.83 We should bear in mind that the stigma associated with polyandry and the other practices analyzed in this book, as well as their criminalization, gave people strong incentives to conceal them and mean that any quantitative survey will inevitably underestimate their incidence. Polyandry was a minority practice, but, in weighing its significance, we should bear in mind, too, how few marriages were polygynous during the Qing. Rates of polygyny varied regionally and by class, and it was most common among the gentry and wealthy merchants, who constituted a tiny percentage of the population.84 The best available data suggest that, overall, well under 4 percent of married women were concubines, whereas the rest were main wives.85 But surely no one would argue that concubinage is trivial to our understanding of Chinese marriage. On the contrary, polygyny has usually been seen as paradigmatic, and it has consequently received a degree of scholarly attention far out of proportion to its actual incidence. For example, the foundational studies of marriage and gender roles in late imperial China by Patricia Ebrey, Dorothy Ko, Francesca Bray, and Susan Mann all focus exclusively on polygynous elite households.86 If polygyny was paradigmatic of Chinese marriage, then so was polyandry—and it seems likely that polyandry broadly defined (including transactional polyamory and marital pros-

Chapter 1. “Getting a Husband to Support a Husband”

53

titution) was by far the more common of the two practices. In fact, there is a logical connection between polygyny among the rich and polyandry among the poor, in that the transfer of women to rich households (to become concubines and maidservants) exacerbated the shortage of wives in poor communities. Late Qing feminist He-Yin Zhen saw this connection clearly. In her 1907 essay “On the Question of Women’s Labor,” she characterizes elite polygyny as “an indirect method of seizing the property of the poor” that, by worsening the imbalanced sex ratio, made polyandry a logical necessity: In the homes of high officials or large extended families, the number of concubines can reach more than ten. [In contrast,] in such places as Yancheng, Xinghua, and Gaoyou counties [in Jiangsu], among the lower classes one woman belongs to many men; or younger and older brothers share a wife. In Yangzhou, peasants who have many sons always provide a wife for the eldest, whereas the sons next in line can never marry. . . . [A] country with a system of one man and many wives has to have a system of one woman and many husbands. Moreover, there have to be a certain number of men who have no wives, and those without wives are sure to be the poor.

In sum, she argues, “the evil of the wealthy lies not only in raping the daughters of the poor, but also in tacitly stealing the wives of poor men.”87 He-Yin Zhen’s radical analysis exposed the sheer hypocrisy of an orthodoxy that would condemn the poor for failing to conform to elite ideals of gender performance and sexual propriety. Polyandry in its various forms was a survival strategy, one means by which “the little people” coped with the very big social and economic problems that have afflicted China over the past few centuries. To recapitulate, three larger forces that converged in these strategies were the skewed sex ratio and concomitant surplus of single men, the pervasive market for women’s sexual and reproductive labor, and the subsistence crisis of a growing number of rural families. At the intersection of these larger forces, we find the case examples narrated above. It seems likely that the three interrelated problems that framed polyandry were all getting worse over time—that is, they were affecting a growing number and probably a growing proportion of people, from the mid-Qing on. (For example, even if we assume that the percentage of “bare sticks” in the population held steady from 1700 to 1900, the number of such men would have more than tripled.) If that is correct, then the incidence of polyandry and related practices would likely have increased as well. The report from Gansu in Investigation of Customs contains a remarkable flash of insight: “[T]his evil custom [of ‘getting a husband to support a husband’] exists almost everywhere in the province (ji yu quan sheng jie ran). It seems likely, therefore, that to enforce the law and prohibit it would pose an immediate threat to the

54

part one

survival of a very large population.”88 Like He-Yin Zhen’s analysis, this frank statement cuts through all the moralizing rhetoric, right to the heart of the matter. Polyandry was a response to exigencies that had rendered the normative standards and values of the late empire irrelevant. One might even read it as a harbinger of their collapse.

2

Attitudes of Families, Communities, and Women toward Polyandry If my husband wants to bring Liu Yu into our household to live with us, what does that have to do with you? If anyone comes out here and talks nonsense again, then I’ll fight him to the death, and die at his house! — xia shi, a remarried widow aged 27, defying the senior men of her husband’s lineage 1

As to the authority of the husband, this has always been weaker among the poor peasants because, out of economic necessity, their womenfolk have to do more manual labor than the women of the richer classes and therefore have more say and greater power of decision over family matters. In sexual matters (xing de fangmian), poor peasant women also have relatively more freedom. Among the poor peasant class, triangular and multilateral relationships (sanjiao ji duojiao guanxi) are almost universal (jihu shi pubian de). With the increasing bankruptcy of the rural economy in recent years, the basis for men’s domination over women has already been weakened. — mao zedong, 1926: “report on the peasant movement in hunan” 2

In this chapter, we continue our analysis of the subject positions of the parties involved in polyandry, and of their families and communities. We begin with stigma, which was a significant factor in many cases but nevertheless is particularly difficult to evaluate. To state the obvious, stigma and even criminalization do not necessarily deter people from doing something if they have good reasons to do it. To what extent did stigma or shame deter people from engaging in polyandry? If they did engage in polyandry, how did stigma affect them—if, indeed, it affected them at all? Did families or communities care enough about stigma to cause these people trouble? Under what conditions did local lineage or village authorities interfere in such arrangements? 55

56

part one

Next, we address women’s agency and perspectives within polyandry. Why did women participate in polyandry? How did they understand and experience their situation? Under what circumstances did coercion occur, and to what extent did such relationships victimize women? How much scope was there for female agency in this context? Was polyandry ever good for women? This chapter concerns matters that are hard to quantify, and, given the nature of the surviving evidence, any generalizations must be impressionistic and somewhat tentative. Nevertheless, two basic points do seem clear. First, families, communities, and local village authorities rarely interfered with people who engaged in polyandry, even if they did not approve. On the contrary, the evidence suggests a high degree of indifference or even tolerance, in part because those who might complain could rarely offer a viable alternative to those concerned. Second, for obvious reasons, polyandry required at least a minimum of cooperation, if not enthusiasm, from the woman. Even if a woman was reluctant to resort to this measure, she had to come to the conclusion that polyandry was the least bad option realistically available to her and her family. Otherwise, it would be very difficult for the men to make this strategy work (the same was true of wife sales—see Chapter 7). Although some women were coerced (by violence and intimidation), this does not seem to have been the most common scenario, and such coercion was possible only when a woman’s natal family was unable or unwilling to intervene on her behalf. But also, in more than a third of my cases, it was the woman herself who took the initiative to recruit a man who would help support her family in exchange for sex. If a woman’s husband could not support their family, and if she did not want to leave him and their children, then she might see polyandry as the optimal solution to their difficulties. After all, polyandry was fundamentally a strategy to keep a family together by raising the ratio of laborers to consumers and loosening the Chayanovian straitjacket. The likely alternative would be to break up the family (by selling off children or the wife herself) in order to survive. In these circumstances, we find pragmatic women using sex in a strategic, unsentimental way to secure their families’ subsistence. But also, some of these relationships began with passionate extramarital affairs in which the women secured their husbands’ acquiescence by making their partners provide material aid. S T IG M A , L O S S O F FAC E , A N D O T H E R P R IO R I T I E S

The openness and formality of the contractual arrangements for “getting a husband to support a husband” described in the last chapter imply a high degree of social acceptability and even accountability—why else have witnesses, let alone a contract? Since Qing law prohibited these transactions, they could be enforced only on the community level, and in this respect they resemble other prohibited

Chapter 2. Attitudes toward Polyandry

57

transactions that were possible only because of community acceptance. Some of the commentary on “getting a husband to support a husband” found in Investigation of Customs confirms a high degree of community acceptance, at least in some regions, despite its authors’ condescension and disapproval. We learn that in Gansu, for example, “the first husband’s relatives never interfere in these arrangements, and there is no social stigma attached to them whatsoever.” In some parts of Gansu, these alliances were highly ritualized and publicly celebrated much in the manner of normative marriages.3 But the evidence from legal cases is mixed, and we find many examples of people who engaged in polyandry being subjected to gossip and ridicule. In the midrange of practices between formal, openly acknowledged polyandry and outright retail prostitution, especially, we find many couples trying to maintain secrecy or at least ambiguity about the wife’s sexual relations. There are also examples in which the husband or other family members try to maintain a degree of deniability by never openly acknowledging what is going on, although they know about it. In assessing the significance of stigma, we must bear in mind that many forms of marriage were stigmatized to one degree or another, and some were also criminalized. In this respect, polyandry and wife sale were by no means unique, and, as explained in the introduction, it is not clear whether the stigma attached to these particular practices exceeded that attached to others. Moreover, perceptions might vary by region, social class, and even gender. (For example, among men, taking a concubine was a prestigious status symbol, but even some wives who had agreed to be sold absolutely refused to become concubines.)4 Also, not everyone was equally sensitive to stigma, and some simply faced more pressing concerns; as many witnesses testified, worrying about “face” was a luxury that not all could afford. The following case illustrates the paradoxical quality of polyandry as a widespread, well known, and formally contracted alliance that many people were ashamed of all the same. In 1735, the harvest failed in Guyuan Subprefecture, Gansu, and by the New Year peasant Ha Qijun (57)5 and his family had begun to starve. Ha and his wife, Zhang Shi (50), decided to head south to Fengxiang Prefecture in Shaanxi, about a hundred kilometers away, where they hoped to find work and food; the couple’s son (13) and adopted daughter-in-law (7) went with them. After a day’s walk, the family stopped at a Guan Di Temple to shelter for the night and beg for food. There they encountered a teacher named Ye Ce (48), who worked at the temple; Ha knew him slightly from a previous visit during another famine many years before. Ye Ce was an immigrant from Wugong County in Shaanxi. He was not a typical “bare stick” in that he was literate enough to market himself as a teacher and had saved up a little money; even so, he had no wife or family, and certainly could not be called prosperous by any ordinary standard. But prosperity is a relative concept, and Ha Qijun decided to approach him. According to Ye Ce’s testimony, Ha told him, “Teacher, you have no family of your own; how

58

part one

would it be if I used my wife to bring you into my family, so we could all get by together (ba wo laopo zhao le ni, women yi tong guo rizi)?” Ha agreed, but it was not easy to persuade Ha’s wife. Zhang Shi testified: My husband told me, “There’s a teacher named Ye, I know him from before, and he has money. . . . I have it in mind to use you to bring him in and have him support our family so that we can survive (ba ni zhao le ta yanghuo women yi jia).” When I heard this talk I rebuked him, saying “I’m already old, and our son is already grown up, how could I do such a thing?”

Zhang Shi and her husband argued for four days until she finally gave in. As she later recalled, he told her, “If we lose this man, and don’t bring him into our family, then who else are we going to find who can help us? I’ve already acknowledged him as family, and there’s no one who will know what’s going on. If you don’t obey me, then I’ll die here, I refuse to go on!” Once the problem of immediate survival was solved, however, Ha Qijun became concerned that word of their new arrangement might spread back home, which was only a day’s walk away. So he asked Ye to join them in walking to Shaanxi. They ended up some 200 kilometers from home in Jun County, where Ye found work as a teacher and supported the family; they told neighbors that Ye was Zhang Shi’s brother, everyone in the Ha family called him “Uncle Ye” (Ye shu), and the sexual relationship between Ye and Zhang Shi was kept secret. This arrangement continued for about two years. One day Ha Qijun and Zhang Shi heard that the famine in Gansu had ended, and they insisted on returning home. After some argument, Ye agreed, but when they neared home, Ha told him they no longer wanted his company. Incensed, Ye Ce murdered him and then forced Zhang Shi to accompany him back to Shaanxi as his own wife. Two points are crystal clear. Everyone who testified in the case was clearly familiar with polyandry and used a consistent vocabulary to characterize it (“bringing in”/zhao a second man to “support”/yang the family). Moreover, there is no evidence in the case record that anyone actually ridiculed or otherwise bothered the couple because of their relationship with Ye Ce. Nevertheless, the couple found the relationship shameful—hence Zhang Shi’s reluctance, Ha’s anxiety about public exposure, their effort to use kinship of one kind (“Uncle Ye”) to conceal another (polyandry), and the fatal attempt to sever relations. The dispute that led to murder focuses our attention on the emotional as well as material investment that individuals could make in such alliances. In rejecting Ye, Ha had reneged on what had been stipulated as a permanent arrangement. Ye Ce stated: “That day he told me that since he was already old, after he died his wife would be mine. Only then did I agree to hand over the money I had saved.” Ye Ce obviously did not see himself as some sort of customer, to be rejected when his

Chapter 2. Attitudes toward Polyandry

59

patronage was no longer wanted. On the contrary, Ye saw himself as part of a family: “I had joined his family by marrying his wife (zhao zhei furen), and had supported the entire family for two years—but now he was going back home and wanted to get rid of me! So I decided to kill him instead.” He clearly did not share the couple’s shame about their relationship.6 To reiterate, legal cases record only relationships that ended in serious trouble, so they probably convey an exaggerated impression of the impact of stigma. We should also bear in mind that, when testifying in court, women would likely emphasize their sense of shame and devotion to chastity in the hope of receiving a more sympathetic hearing.7 In contrast, the social surveys convey little or no sense of stigma on the part of people engaged in these practices, despite the authors’ obvious disapproval. But also, there are plenty of cases in which couples were quite open and obviously did not care about the opinions of other people. Attitudes seem to have varied within families, too. Often the husband and his male relatives seem to have worried more about losing face than did the woman herself: her lack of chastity seems to have bothered them more than it bothered her. When the couple was part of a well-organized corporate lineage, then elders sometimes tried to change their behavior in order to protect family honor, but (as we shall see below) such intervention did not always succeed. Reproach by Family We do find examples of family members feeling embarrassed and reproaching those involved in polyandry. Typically, however, the same family members were in no position to provide a viable alternative, and therefore their objections were met with indifference or even defiant resentment. In a 1750 case from Zhaoyuan County, Shandong, Yang Chao’er (34) and his wife, Li Shi (27), were poor peasants with two small children. In the spring of 1748, they found that they could no longer make ends meet, and so the entire family walked to Li Shi’s natal village and asked her older brother, Li Wenbin, to take them in. A woman’s natal family had a recognized, customary role in protecting her interests after marriage, and the fact that most peasant women married men from villages within walking distance of their natal homes facilitated the maintenance of close ties.8 Therefore, it was natural for Li Shi and her family to seek her brother’s help in an emergency—and if a woman did not want to participate in polyandry, but was being pressured by her husband, then an appeal to her natal family would be her most obvious and effective resistance strategy. But Li Wenbin refused to help them. He was little better off than his sister’s family, and he also feared entanglement with his shiftless brother-in-law, who had a reputation for “liking to eat but being too lazy to work” and drinking heavily. Instead, the family found shelter with a neighbor, a single tenant farmer named Yu Sihan, who took pity on them. Yu lent them 200 cash to buy food, and he

60

part one

agreed to let them stay for the time being in one of the three rooms of his straw house. But Yang Chao’er failed to find steady employment, so his family again ran out of food, and meanwhile, Yu Sihan was pressing them for repayment. Finally, Yang Chao’er told Li Shi to go sleep with Yu in lieu of paying their debt. At first she refused, but Yang beat her; in desperation, she went back to her brother for help, explaining frankly what her husband wanted her to do: “I was hoping my brother would stand up for me.” But once again he rejected her: “Your husband is useless, but it’s no good complaining to me, because I can’t take care of you.” After that, Li Shi gave in. One night, she went to Yu’s room and told him (as she later recalled), “my husband owes you money and can’t pay you back, so I’ve come to keep you company (he ni zuo ban), on condition that you agree not to ask for the money.” This became an ongoing arrangement, in which Yu supported the family in exchange for sexual access to Li Shi. Either Yu would sleep with Li Shi in her room (while her husband and children went elsewhere), or in winter she would come to his room, where he had a kang to sleep on and warm blankets. This variation of polyandry was a little unusual, in that the refugee family moved in with the outside male, instead of vice versa, but in other respects the quid pro quo that united them in a single household closely resembled “getting a husband to support a husband.” The arrangement soon became widely known in the village, and Li Shi’s brother, out of embarrassment, finally tried to intervene: I saw my younger sister always going in and out of Yu Sihan’s room, and finally I couldn’t bear the sight of it any longer, so I went over and tried to talk to my brotherin-law. But he just said, “We poor people can’t afford to abide by those moral rules!” (anmen xiaohu renjia yi bude zhexie guiju), and he got angry at me for not minding my own business. After that, I didn’t bother them anymore.

Li Shi also resented her brother’s meddling: if he really cared so much, he should have helped them when she had twice begged for his help. She had not wanted to sleep with Yu Sihan in the first place, but it was too late for her brother to complain now.9 Yang Chao’er and Li Shi were certainly aware of the normative “moral rules” violated by polyandry, but, with their family’s survival at stake, such niceties were simply beyond their means. We find the same sentiment expressed in a 1747 polyandry case from Yongqing County, Zhili: when the husband was accused of “not wanting face” (bu yao lian)—that is, of being shameless—he retorted with a proverb: “If you have money, then you can have ‘face,’ but if you don’t have money, then you can’t have ‘face’!” (you qian jiu you lian, mei qian jiu mei lian). He added that if his wife did not sleep with the other man, then they would have no choice but to sell their son in order to pay their debts.10 The whole question of shame and stigma must be seen in this light.

Chapter 2. Attitudes toward Polyandry

61

We find a similar confrontation with family in a 1746 case from Nanchuan County, Sichuan. In this case, a landless agricultural laborer and occasional peddler named Huang Yinglin (43) was married to Xiong Shi (33). The couple had great difficulty making ends meet, and in early 1743, a few days after the New Year, Huang invited an acquaintance, Guang Ronglun (32), to move in with them and pool resources. Guang was a landless peasant who made a decent living peddling paper, and over the years he had saved up more than 20 taels; it was Guang’s savings that prompted Huang to invite him. Guang agreed, because he had no wife or family of his own and because he intuited that Huang’s invitation implied an offer of sexual relations with Xiong Shi. At Huang’s urging, Guang pledged himself to the couple as their “godson” (gan erzi)—an interesting variation of the adoption strategy—and then shared his savings and earnings with them. Initially, there was no explicit agreement about sex, but, after Guang moved in with the couple, Huang told his wife to sleep with him; she was reluctant, but he pointed out that they had no other way to repay Guang and secure his continued support, so she finally agreed. Then, as Guang later recalled: “Huang Yinglin told me: ‘You haven’t married yet, and since we’re all eating from the same pot, there’s no need for us to distinguish between yours and mine’ (women jiran yi guo chi fan, dajia ye fen bu de ni wo), and he said that I should sleep with his wife. . . . After that, she and I usually slept in the same bed.” The principals made no effort to conceal their arrangement, and word spread in the community. Huang Yinglin had an older brother, Yingxuan, but since they had divided their father’s household long before and ate from separate hearths, Yingxuan did not feel responsible for Yinglin and had declined to help him. (Once again, we find that “fictive” kin were more reliable in a crisis than “real” kin.) But when Yingxuan heard what was going on, he felt embarrassed and reproached his brother: I saw that Guang Ronglun did not distinguish inner from outer in his relations with my sister-in-law, and this looked very bad. . . . I advised my brother: “You have a young wife in your household, and you shouldn’t let this single man live with you.” But he said: “Guang Ronglun and your sister-in-law address each other as ‘mother’ and ‘son,’ and they get along well together. I’m very poor, and we rely on his money to survive, so I can’t afford to pay too close attention (ye bu gu de xuduo le).” . . . He told me there was no need for me to worry about it. I saw that my brother was content to do this kind of shameful thing, so it was hard for me to stick my neck out and make a scene about it. So I just left him alone.

Despite this euphemistic language, it is obvious that everyone concerned understood exactly what kind of relationship was framed by chosen kinship in this household. Once again, from the standpoint of the couple engaged in polyandry, stigma was far less important than the resources secured by this “shameful” means.

62

part one

This relationship lasted for two years, until Guang’s savings ran out and his inability to provide the same level of support provoked a quarrel. The couple told him to move out, so he demanded that they repay his money, but they had no money, and anyway, from their point of view, he had been fully compensated in the form of sex and the other benefits of sharing their household. When he refused to move out, Huang Yinglin went to the local rural agents (pailin) and complained that Guang had had illicit sex with his wife. Guang became frightened and moved out, so Yinglin withdrew his accusation and the rural agents did not pursue the matter. (Some months later, however, Guang and Huang got in a fight, and Huang died of his injuries.)11 Community Indifference and Tolerance How did neighbors and village authorities react to polyandry? The case records offer a variety of reactions, including a certain amount of opprobrium, but most significant, it seems to me, is the high degree of indifference and even tolerance found in so many. In the cases just cited, for example, it is obvious that polyandry was stigmatized to some degree and caused the relatives of its practitioners some embarrassment. But it is equally clear that no one in those communities was sufficiently offended to cause any trouble, and these relationships were allowed to continue for years without interference. One gets the impression that such arrangements were far from unusual, and although other people may have gossiped and snickered, they rarely felt either entitled or obliged to interfere. Either they did not seriously object, or they preferred not to inquire too closely into other people’s domestic affairs. Given the exigencies that made people resort to polyandry and the inability of others to offer a better solution, the default attitude seems to have varied between “live and let live” and “don’t ask, don’t tell.” The case records contain many examples of such attitudes, often vividly expressed in testimony. For example, in the last case cited above, the two local rural agents later testified that they knew Guang Ronglun had moved in with the Huang family and that “he and Huang Yinglin’s wife did not distinguish between inner and outer. . . . We thought this was inappropriate, but Huang himself told us that Guang is his ‘godson’ (gan erzi) and that his wife and Guang call each other ‘mother’ and ‘son,’ so we saw no good reason to intervene.” The rural agents clearly understood what kind of relationship Guang had with the couple—“not distinguishing between inner and outer” evidently being a standard euphemism for polyandry—but they would not meddle as long as the husband did not complain.12 Similarly, a 1775 case from Wucheng County, Zhejiang, involved a polyandrous relationship between Zhang Zongdai, his wife, Chen Shi, and Zhang’s distant relative Zhang Guobin. Fellow villagers gossiped about this unconventional household, but no one interfered. Later on, when interrogated by the magistrate, the neighbors explained as follows: “Zhang Guobin and Zhang Zongdai seemed to get

Chapter 2. Attitudes toward Polyandry

63

along well with each other, so it was inconvenient for us to go and meddle in their affairs, which were none of our business. . . . Anyway, everyone has their own affairs, so we never bothered them” (yin geren youshi, buceng qu chawen).13 In a 1736 case from Lujiang County, Anhui, that involved a seven-year polyandrous relationship, the local baojia14 head testified that the people characterized by the magistrate as “criminal adulterers” were just “simple hard-working people who never caused trouble.” As far as adultery was concerned, “that was their own secret affair (siqing), and since the woman’s husband never complained, there was no reason for me to interfere.”15 In a case from Ningdu Independent Prefecture, Jiangxi, a village tolerated a couple’s open polyandrous relationship with a single woodcutter for three years, including co-residence and resource-pooling. But then the woodcutter was caught robbing a fellow villager’s home, and he was prosecuted, beaten, and tattooed with the word “thief ” (qiedao) on his shoulder. After he returned, the couple’s neighbors insisted that they sever ties with this “bad man” and expel him from the village. Clearly, the community did not find polyandry particularly offensive, but theft was another matter.16 To summarize the discussion so far, we find that the usual scenario was for neighbors and community members to refrain from interfering in any serious way—and later, when questioned by authority, they would excuse themselves by saying (for example) that “it was none of our business,” that it was “inconvenient” (bubian) to get involved, that despite suspicion they had no real proof. In the case of rural agents and baojia heads, the most common testimony is that, although they had heard rumors, as long as the husband himself did not complain, they felt they had no basis for acting. In fact, the Qing code required rural agents and baojia leaders to report “criminal elements” (fei lei) in their communities, with penalties for failure to do so.17 Occasionally, this law was cited to punish local authorities who had tolerated polyandry. For example, in a 1756 case from Jinning County, Yunnan, the magistrate cited it to punish the local baojia head and rural agent: both men were stripped of their posts and beaten.18 But the legal definition of “criminal elements” is vague, and in polyandry cases, at least, it appears that magistrates usually did not enforce this law. Some cases suggest, however, that polyandry was less likely to be tolerated if the principals were outsiders. A 1747 case illustrates this pattern. Mao Bangcai (33) and his wife, Peng Shi (28), had three small sons (the oldest aged just 4 sui) as well as a daughter they had adopted out. In 1740, poverty compelled them to leave their home in Xiangxiang County, Hunan, in search of a livelihood. They eventually settled down in Zitong County, Sichuan (roughly 800 kilometers to the west, as the crow flies), where they met Hu Guanghua (43), a single peasant who, with his brother, had migrated from the same county some years before and now farmed land they held by conditional purchase. Mao and Hu Guanghua became friends on

64

part one

the basis of their shared native place ties. Before long, the family moved in and pooled resources with the brothers, in exchange for Guanghua’s sexual relations with Peng Shi; Guanghua’s brother was not happy about the arrangement, but he put up with it. The relationship was open, and everyone in the community became aware of what was going on. But after eight months, the local baojia head became concerned that such behavior on the part of “outsiders” might eventually provoke trouble, so he pressured the villager who had conditionally sold land to the Hu brothers to redeem their land and take it back. Then the baojia head forced the brothers and Mao’s family to leave the village.19 Similarly, in a 1746 case from Yunxi County, Hubei, Li Yonghai and his wife, Liu Shi (both in their early twenties), left Li’s home village (where, as an orphan, he had been raised by his uncle) and migrated to another village in the same county in search of work. They rented living quarters and tried to earn a living by hiring out their labor, but they had trouble making ends meet. Finally, they secured help from a neighbor, Zhong Keming (39), who was single and lived and farmed alone. Zhong made Li a direct proposition: “If you tell your wife to sleep with me, then I’m willing to provide your food and clothing.” This arrangement lasted for seven months, but, as gossip spread, the couple’s landlord became nervous that his tenants’ domestic arrangements might somehow get him in trouble. Finally, he evicted them from their lodgings and, with the help of neighbors, pressured them to leave the village. Zhong elected to go with them, and the three settled in a different village where Zhong managed to rent some land that they could farm together.20 Even in these two cases, we can detect a fairly high level of tolerance for polyandry: after all, no one interfered until seven or eight months had passed. In both, it seems, a crucial factor in triggering intervention was that some or all of the principals were outsiders for whom the community felt no responsibility. Rather than accept trouble from outside, it seemed preferable simply to expel the outsiders. Yet, again, the villagers who took action were not casual acquaintances but were individuals in positions of responsibility who feared they might be blamed should trouble occur. An Extraordinary Example of Lineage “Justice” When the individuals involved in polyandry were members of a large, wellorganized lineage, then the stigma of their behavior might provoke a stronger reaction. A 1748 case from Huangmei County, Hubei, provides an exceptionally dramatic example of family and community condemnation of a polyandrous relationship. The male protagonists in this case were all members of the Xiang lineage and lived in close proximity to one another. Peasant Xiang Zhenglu (48) and his wife, Hu Shi (40), had a long-standing, open polyandrous relationship with Zhenglu’s distant cousin, a widower named Xiang Wanhuai (36), who lived in the

Chapter 2. Attitudes toward Polyandry

65

same village as they; the two men addressed each other as “brother.” This relationship had continued for over four years and was known to everyone in the lineage. At least some of the other men in the Xiang lineage felt anger and resentment at this relationship, because “it damaged the lineage’s reputation” (baihuai menfeng). Zhenglu and Wanhuai had been repeatedly admonished by lineage elders, to no effect. Moreover, according to later testimony, both Zhenglu and Wanhuai had a reputation as worthless, dishonest reprobates who had “stolen pigs” and “slaughtered oxen.” Wanhuai was further accused of reneging three separate times after promising his daughter in marriage. It is hard to know how much of this later testimony to believe, since its obvious purpose was to blacken the men’s reputation in order to justify what had been done to them. One day in the spring of 1746, the senior men in the lineage (not including Zhenglu or Wanhuai) gathered at the ancestral hall to organize for the Tomb Sweeping Festival, which is dedicated to honoring ancestors and visiting their graves. Conversation soon turned to the way Zhenglu had shamed the lineage by permitting his wife to carry on openly with Wanhuai. Several men indignantly argued that Zhenglu should be put to death by the lineage; others disagreed, proposing instead that, if Zhenglu had broken the law, then he should be taken to court for prosecution. Finally, the men agreed that Zhenglu should be forced to sell his wife; moreover, Zhenglu’s older brother Xiangzhou should be publicly beaten, in order to teach him to be stricter with “family discipline” (jiajiao). On the day of the festival, the men gathered at their lineage’s cemetery, bringing pork and other offerings for the ancestors. A group seized Zhenglu’s brother Xiangzhou, tied him up, and beat him with a shoulder pole while admonishing him for not preventing his younger brother’s disgraceful behavior. Confronting Zhenglu with paper, brush, and ink, they demanded that he “draw up a handprint and footprint contract” in order to “sell off ” his wife. (Contracts for wife sale usually bear the handprint and sometimes the footprint of the husband/seller, affixed in the presence of witnesses to demonstrate his consent.) Threatened with violence, Zhenglu had no choice but to agree. But then the couple’s partner in polyandry, Wanhuai, intervened. Protesting loudly, he declared that any man who forced Zhenglu to sell his wife should be required to provide Zhenglu with his own wife or daughter to replace her; Wanhuai further threatened to file charges against the lineage elders and “make them all die at his hand.” The lineage men then turned on Xiang Huangzuo (56), because he was Wanhuai’s closest senior relative, and scolded him for allowing Wanhuai to behave in this outrageous manner. Huangzuo then scolded Wanhuai, but the latter defied and cursed him. Incensed and humiliated, and pressured by the lineage to act, Huangzuo finally enlisted his nephews’ help and together they beat Wanhuai senseless, tied him up, and buried him alive. Inspired by this decisive act, two other men seized Zhenglu, bound him, and buried him alive as well.

66

part one

Just then, Hu Shi, who had heard what was afoot, rushed to the scene wielding a knife in a vain, heroic attempt to save her husband’s life. Men seized her and beat her, and then, declaring that “this whore” should die too because she had already caused two men’s deaths, they buried her alive next to her husband and lover. More than twenty men of the Xiang lineage witnessed these three murders, but no one tried to prevent them. Later, in defiance of specific orders from the elders, Zhenglu’s father and brother reported the murders to the county yamen, triggering mass arrests and a huge investigation that resulted in the main perpetrators receiving the death penalty. In court, the killers and witnesses testified that they believed their actions justified, because “if there’s someone in the lineage who defies law and public order, then according to family law he can be put to death” (zu zhong bu gong bu fa de ren, jiafa keyi chusi de de). The Qing government did allow lineages considerable leeway to manage their own affairs, but their privileges did not include license to kill. The key factor in this case was the presence of a well-organized corporate descent group whose leaders were acutely sensitive about their extended family’s reputation and honor. The festival’s rituals gathered the lineage men together under the eyes of their ancestors, thereby setting up the confrontation but also reinforcing the hierarchy and solidarity of the lineage organization. In this context, the trio’s defiance was particularly offensive and provocative, generating an unbearable level of peer pressure on their closest senior relatives to act. This case provides a salutary reminder that patriarchy’s victims do include men, as well as women, who refuse to conform. But that context makes the defiant courage displayed by Xiang Wanhuai and Hu Shi all the more striking and poignant. They evidently felt no shame about their relationship—on the contrary, they tried to defend Xiang Zhenglu and to face down the lineage, despite the extraordinary tension and danger of the situation. In fact, the Xiang lineage itself was divided. The Tomb Sweeping Festival occurs annually, but more than four years had passed before anyone actually took action against the polyandrous trio. When they finally did take action, the consensus plan was to expel the woman, Hu Shi, as the discordant alien element who had disrupted the male solidarity at the heart of lineage power; after all, both of the men involved were surnamed Xiang. (Here, we may recall Avron Boretz’s observations about the misogynist basis for fraternal solidarity discussed in the previous chapter.) If Xiang Wanhuai had kept his peace, the matter would have ended with a wife sale. But by defying his lineage elders in the presence of the ancestral graves, Wanhuai radically escalated the dispute, provoking the extremists among the Xiang men and silencing the more moderate voices that had prevailed up to that moment. Moreover, although Xiang Zhenglu’s father and brother may not have approved of polyandry, they did not agree that he deserved death; the fact that they defied their elders by seeking justice at the yamen shows that they rejected

Chapter 2. Attitudes toward Polyandry

67

this application of “family law.” As counterintuitive as it may seem, therefore, I would suggest that even this hideous instance of patriarchal repression reveals a surprising measure of tolerance.21 An Example of Failed Lineage Intervention The extreme cruelty and inhumanity documented in the last case above appear to have been rare; certainly not all lineages were so determined to interfere, nor so harsh in passing judgment. In a 1797 case from Zhuji County, Zhejiang, a landless peasant couple formed a polyandrous relationship with an unmarried agricultural laborer in their village, Zhou Chenggui (34), after husband Zhou Chenghuan (26) fell ill with jaundice and was unable to work. Their arrangement (which included co-residence and resource pooling) was open, and everyone in the community knew about it. The two men were distant cousins and members of the Zhou lineage. Finally, concern about the family’s reputation prompted the lineage head (zuzhang) to intervene: They carried on without inhibition and there was ugly gossip. So I went to Zhou Chenghuan and questioned him and scolded him. But he told me that because of his illness, he and his wife had to rely entirely on [Zhou Chenggui’s] support in order to survive, and that they simply had no other choice. After that, I didn’t bother them any more.

Out of sympathy for the couple’s plight, the lineage head declined to press the matter, and no one else interfered. The arrangement continued for four more years.22 In a 1797 case from Hefei County, Anhui, we find an example of lineage intervention that had a decidedly different outcome, thanks to the forceful personality of the woman in question. In this case, peasant Liu Peirong (50) and his wife, Xia Shi (27, a remarried widow), were very poor; their household included two small children and Xia Shi’s invalid mother. Their closest neighbor was a distant cousin of Liu’s named Liu Yu (36), a single man who lived alone and farmed cotton on land he owned by conditional purchase. A polyandrous relationship developed in stages, over the course of a year: first, Liu Yu began socializing with the family and helping them in various ways, and the couple expressed their gratitude by having their son of 4 sui “pledge” Liu Yu as “godfather” (ganye); then Liu Yu put up 4000 cash to help the couple buy an ox, the use of which he would share; and finally, Xia Shi began a sexual relationship with Liu Yu behind her husband’s back, and the two became passionately attached to each other. When Peirong discovered his wife’s adultery, he reacted angrily, but Liu Yu immediately proposed that they merge their households and pool resources, pointing out that this arrangement would solve the family’s economic difficulties. Peirong agreed, and Liu Yu’s sexual relationship with Xia Shi continued and became very open.

68

part one

After six months, Peirong’s cousin Liu Peile learned what was going on, and he reported the shameful matter to other men of the Liu lineage. A delegation of five men representing the lineage confronted the couple, forcing them to admit the nature of their relationship with Liu Yu, and then seized Liu Yu and Xia Shi, bound them with ropes, and took them to the ancestral hall to confer about what should be done to punish them. At first, the lineage elders resolved to take the adulterers to the county yamen and press charges, but Liu Yu and Xia Shi begged for mercy; so instead, to avoid shameful publicity, it was decided “to deal with this according to family law” (jiafa chuzhi). The couple were ordered to terminate their relationship with Liu Yu, including their pooling of resources, and to have no further contact with him; given the couple’s poverty, cousin Liu Peile was ordered to advance 4000 cash to buy out Liu Yu’s share of the ox. The latter stipulation seems to have been Peile’s penalty for failing to mind his own business. But without Liu Yu’s help, the family soon found themselves in difficulty once again, and, with the New Year approaching, Peirong decided to sell their ox. At first he failed to find a buyer who would offer a good price, but then Liu Yu helped him find someone who agreed to pay the ox’s original sale price, making Peirong happy and relieved. After this, the couple resumed their former polyandrous relationship with Liu Yu, in defiance of the lineage. When word spread, the same five lineage delegates visited to investigate. But this time, Xia Shi confronted the men and boldly defied them, shouting: “If my husband wants to bring Liu Yu into our household to live with us, what does that have to do with you? If anyone comes out here and talks nonsense again, then I’ll fight him to the death, and die at his house!” (wo zhangfu yuan zhao Liu Yu tong zhu, yu nimen hegan? ruo zai youren chulai hun shuo, wo jiu pinming si zai ta jia). Intimidated by her defiance, the men quickly departed. They later testified: “Her husband Liu Peirong just stood there and didn’t make a sound. But Xia Shi told us to our faces that she would fight to the death with anyone who tried to stop them (na ge zuzhi ta jiu pinming)! We were afraid, because she was so unreasonable and made such a scene (pa ta sa po), so we all just left.” After that, no one from the Liu lineage bothered them again.23 Qing legal cases show that men took women’s threats of suicide very seriously. In this case, Xia Shi’s specific threat to die at the house of any man who presumed to interfere with her domestic arrangements amounted to a threat to curse that man and his household, to ruin their fortune and haunt them in perpetuity. Suicide was a weapon of the weak, to be sure, but desperate women did sometimes “stage” suicides in a manner calculated to bring calamity to those who had offended them. For example, I have seen many cases with the following scenario: a woman who had been sexually harassed or raped by a fellow villager was unable to persuade the men in her family to do anything about it, because of the shame that publicity would bring; in the end, she crept out at night and committed suicide at

Chapter 2. Attitudes toward Polyandry

69

the house of her tormenter—usually by hanging herself from a tree or even a roof beam of the house itself. Such an act had terrifying implications for those who were its target—and, as a practical matter, it guaranteed publicity and an investigation into the offensive behavior that had provoked the suicide.24 WOM E N ’ S AT T I T U D E S A N D AG E N C Y I N P O LYA N D RY

Xia Shi’s feisty defiance highlights the importance of women’s own attitudes and interests with regard to polyandry. How did women themselves view their role in these arrangements? How did the experience of these survival strategies affect their own attitudes and consciousness? To what extent did these practices constitute the oppression or exploitation of women? There are no simple answers to these questions, and it is difficult to generalize from a series of anecdotes about diverse individuals. Practices like polyandry were certainly part of the May Fourth stereotype of what Dorothy Ko has called “the victimized woman in old China,” reinforced by the polemical fiction of such writers as Xu Dishan.25 Of course, some real women were indeed victims, and we encounter some of those women in this book. As we have seen in several of the cases narrated above, women would be especially vulnerable to coercion if their natal families were unable or unwilling to help them. But the stereotype of unremitting victimization obscures more than it reveals, at least when it comes to people’s actual experience of the gender system in late imperial China. The evidence from legal cases reveals something both more complex and more interesting. First of all, it would be inaccurate to imply that the husband was always the subject of action, and the wife always its object. In more than a third of my polyandry cases, it was the wife herself who took the initiative in negotiating her extramarital sexual relations, often informing her husband only after the fact. But even when it was men who took the initiative, strategies like polyandry and wife sale usually seem to have required a woman’s cooperation in order to succeed. Coercion and Other Forms of Pressure A 1749 case from Yangyi County, Shandong, provides a dramatic example of one wife’s bitter resentment and resistance toward a husband who tried to force her to support him by prostitution. Liu Xian (36) was a landless peasant who earned a meager living as a boatman; he lived with his wife of twelve years, Zhang Shi (33), their two children, and his elderly mother. The marriage was fraught with conflict. Zhang Shi described her husband as “a shameless wastrel” who had lost their land and house through improvidence. In 1745, a customer who hired Liu’s boat offered money to have sex with Zhang Shi, and Liu agreed. This was the only instance of paid sex that she performed, and she bitterly resented it and refused to do it again. Later, Liu had to sell his boat to pay debts, leaving him without a livelihood. As a

70

part one

result, his mother had to go live with his married-out sister, and the couple and their children moved in with Zhang Shi’s sister, brother-in-law (a tailor), and elderly mother, who reluctantly agreed to lend them lodging. But the couple often quarreled because Liu kept pressuring Zhang Shi to sleep with men for money. Her refusal enraged him, and he became increasingly abusive. That winter, Liu sold off Zhang Shi’s loom for 1500 cash and spent it all, further exacerbating their difficulties, because her weaving had provided at least some income; from then on, she occupied herself with spinning. A few weeks later, however, Liu was unable to pay a debt of 200 cash (for a couple days’ rental of a donkey), so he pressed Zhang Shi to sleep with his creditor, one Hao Tong. According to Zhang Shi’s testimony, when Hao visited to collect his debt, Liu asked him for another 200 cash: Hao Tong said, “You haven’t paid me back for the donkey yet, how can you ask me to lend you more?” My husband answered, “I have no money to repay you, so why don’t you let ‘your sister-in-law’ here pay you back instead (ni jiao ni saozi huan ni ba)?” Hao Tong pulled out 200 cash and asked me, “Is that okay with you?” My husband told him, “Of course it’s okay!” I saw that he wanted me to do that shameful thing again, and I scolded him. When Hao Tong heard my reaction, he took his money and left.

Liu’s choice of words is revealing: by referring to Zhang Shi coyly as “your sisterin-law,” he was implying that Hao and he were “brothers” and invoking the fraternal sharing of resources—albeit in a particularly crass way. As this language suggests, Liu hoped to recruit Hao for a longer-term arrangement, perhaps framed by sworn brotherhood. But Zhang Shi had foiled his plan—so instead, Liu sold their daughter for 1000 cash to be an adopted daughter-in-law. (Note that this little girl’s sale price was only two-thirds that of the loom.) Zhang Shi hid the money from her daughter’s sale to prevent her husband from wasting it, provoking a confrontation in which he beat her viciously. This episode was the last straw for Zhang Shi: late that night she stabbed her husband to death, with the help of her mother and sister (an extreme but telling example of a woman’s natal family standing up for her). This case illustrates the reality of coercion but also the difficulty of making coercion work, as well as the possible results (and the all-too-real limits) of a wife’s resistance.26 In a number of cases, a wife reluctantly agreed to engage in polyandry or similar strategies only after her husband confronted her with the alternative of selling their children. In women’s testimony, this scenario is typically framed as a husband’s ultimatum that compels his wife’s capitulation. Nevertheless, it seems less like coercion per se than pressure of circumstance, which weighs not only on the woman herself but also on her whole family, circumscribing her husband’s options as much as hers. In these cases, selling sex may well have been the only way to avoid selling family members.

Chapter 2. Attitudes toward Polyandry

71

A 1750 case illustrates this point. Liu Mian (42) and his wife, Mao Shi (38), were poor peasants from Zhenyuan County, in eastern Gansu; they had a son aged 9 sui. By all accounts, the couple had a harmonious marriage. In the summer of 1748, the family fled famine and walked to Bin Independent Department, Shaanxi (about sixty kilometers to the southeast), where they borrowed lodgings in an earthen pit dwelling owned by a single man named Zhang Ninghuan (35), who earned a modest living making and peddling pottery. Liu Mian could not find work, his family had no food or money, and they were very hungry. Within days of arriving in Shaanxi, Mao Shi later recalled, “my husband told me that we had to sell our son, but I refused. Then my husband said, ‘If you won’t let me sell the boy, then what are we going to live on?’ ” He told her frankly that their only other option would be for Mao Shi to persuade Zhang Ninghuan to support them by sleeping with him: “He said ‘You will have to serve as his wife temporarily (ni zhanqie he ta zuo ge fuqi—literally, “be husband and wife with him”), so that he will support us for a while.’ I saw I had no other choice, so I agreed to do what he said.” Significantly, Liu Mian did not propose selling Mao Shi herself—the couple did not wish to separate. Mao Shi did not particularly like Zhang Ninghuan, let alone desire him, but as she later testified: “I saw that my whole family was going to die of hunger, so I had no choice but to obey my husband and do this thing.” Eventually, this ad hoc measure developed into a formal agreement for “getting a husband to support a husband,” which the two men and Mao Shi negotiated while they all lay in bed together, along with the couple’s son, on Zhang’s single kang. In her later testimony, when explaining why she “had no choice,” Mao Shi did not claim that her husband or Zhang had coerced her. Instead, she cited the danger of her family starving to death, her refusal to sell her son, and the lack of other options.27 Men’s Shame, Women’s Power The women who appear in these cases almost never seem to be passive victims. Far more typical than passive victimhood is the behavior of the woman named Yan Shi (38) in the following 1741 case from Weining Subprefecture, Guizhou. She and her husband ran a little wine shop in a market town with the help of a single migrant from Jiangxi named Sun Dongbi (31), who also sold wine at his own street stall. When the couple first met Sun, they were short on cash, so Yan Shi made a pass at him and began a sexual relationship; at the same time, she began regularly “borrowing” money from Sun (none of which she ever paid back) to supplement her own business income. Her husband deferred to her and did not object. Sun continued to live separately, but visited regularly for sex and meals. This relationship went on for ten years. It became something of a burden to Sun, who liked the sex but came to resent the constant drain on his meager funds,

72

part one

because he also felt obliged to send money home to his widowed mother. But Sun was afraid to refuse Yan Shi, because even the slightest delay in his payments would provoke her to make an embarrassing public scene at his stall. According to witnesses, Yan Shi did not care what others thought of her, and although people certainly did gossip, she seems to have been popular. One day, Sun found her sitting next to another man at her wine shop, chatting and laughing. Such intimacy angered Sun, who seems to have felt entitled to monopolize her charms; he called her a “whore” (changfu), told her not to expect any more money from him, and asserted that he would soon return home to Jiangxi. One gets a sense of Yan Shi’s formidable personality from her response, which she screamed before a crowd of witnesses: “And what are you going to do when you go home? You’re going to go home and fuck your mother, that’s what!” (ni huiqu zuo shenme, ni huiqu cao ni jia laoniang). We can be sure that the case record quoted Yan Shi accurately because her words provoked a street brawl in which Sun ended up killing her, and an exact assessment of motive was crucial to the adjudication of homicide.28 This case illustrates two subaltern forms of empowerment available to poor women that surface repeatedly in this sample of legal cases. The first is the power to use insults, public display, and gossip in order to shame men, who often seem much more concerned than the women about reputation and “face.” This distinctly female form of power in peasant society has been documented by a number of scholars in China and elsewhere. Margery Wolf, for example, describes how the women’s community in Taiwan villages could influence men’s behavior by exploiting their fear of losing “face” before neighbors and kin: “Even men who think themselves free to ignore the opinions of their women are never free of their own concept, face. It is much easier to lose face than to have face. . . . This is precisely where women wield their power.”29 The second subaltern form of empowerment seen here is an unsentimental view of sex as a valuable asset for a woman to trade with, in a continuum of various forms of sexual-economic exchange. My sense is that most of the women in these legal cases ended up developing such a view (if they did not have it to begin with), through the experience of marriage and subsequently helping to support their families with their bodies. This attitude implies a frank recognition of the instrumental nature of arranged marriage and an understanding that a woman’s domestic duties in the marital context were not necessarily so different from a wide range of ways that sex might be exchanged for material benefits in extramarital contexts. This attitude took for granted that the main purpose of both marriage and sexual relations (even when meaningful and pleasurable) was not individual fulfillment of a personal, egocentric nature but rather the preservation and reproduction of the family.30

Chapter 2. Attitudes toward Polyandry

73

Women Take the Initiative Who took the initiative in arranging polyandrous relationships? Many were negotiated between the men, and, as we have seen, some husbands resorted to violence or intimidation to induce their wives to submit. But in more than a third of my cases, it was the woman herself who took the initiative, sometimes obtaining her husband’s acquiescence only after the fact. This appears to make sense in broader anthropological terms. For example, a comparative analysis of “non-classical polyandry” in more than fifty societies worldwide concludes that polyandry is an adaptive strategy for coping with male-skewed sex ratios in which “female preference” may well be a factor. In general, the authors argue, “members of the sex that is in shortest supply can better realize their mate preferences since their scarcity puts them in a stronger bargaining position”—therefore, “polyandry in the context of a high operational sex ratio may represent strong female choice for high male investment by having two husbands or multiple fathers.”31 For some of the women in Qing legal cases, the primary motive for entering these relationships was emotional or sexual, with economic assistance enabling them to secure their husbands’ tolerance. But the majority of women in these cases, like Yan Shi, seem to have approached these matters with an utterly pragmatic and unsentimental attitude. They saw that it was up to them to ensure their families’ survival, so they did what they had to do. The way a wife might take matters into her own hands can be seen in a case reported by the governor of Sichuan in 1825. Huang Shi and her husband, Wen Zehan (26), were poor peasants in Hongya County who were having a difficult time getting by. With this in mind, Huang Shi initiated a sexual relationship with a laborer of their acquaintance named Jiang Laoyao (45) without telling her husband. She liked Jiang, but her principal motive was to secure him as insurance against future need. Five months later, Huang Shi’s family found themselves in truly dire straits, with nothing to eat, so she went to Jiang’s home and “borrowed” a large quantity of rice. When her husband asked where she had gotten it, she did not mince words; as he later testified, “she said that because our household was so bitterly poor, she had been having illicit sex with Jiang Laoyao, so that he would help us. I was desperate for his help, so I didn’t make a fuss.” After that, Jiang would openly visit their home to sleep with Huang Shi, and he provided regular assistance in the form of cash and grain. The following year, he moved in with them, and they pooled resources as a single household. Here, a woman saw an opportunity to relieve her family’s distress and took the initiative to do so. There does not seem to have been any particular passion between her and the outside male, although they got along well enough; for her, it was a strictly utilitarian arrangement.32

74

part one

A case from Yongqing County, Zhili, memorialized in 1817, reveals a similarly pragmatic attitude on the part of the wife. Tian Shi (38) was the wife of Liang Yuanji (48); they were peasants, not desperately impoverished ones, but certainly not prosperous either. The couple was friendly with one Xing Wancheng (26), a single man from Dong’an County (about twenty kilometers to the northeast) who supported himself as an itinerant storyteller. Husband Liang later testified: I’m not sure just when they started having illicit sex, but in 1812 I often saw Xing Wancheng joking and laughing with my wife, so I questioned her, and she admitted that she was having adultery with him. She also said that Xing Wancheng had money and could help us out, and since I was greedy for his financial help, I let them continue. He would pay my wife cash, I don’t know how much it added up to in all.

The couple came to depend on this income, and the arrangement continued amicably for the next five years. The neighbors later testified that they had heard rumors “that Liang Yuanji let his wife sleep with Xing Wancheng,” but “because it had nothing to do with us, we didn’t dare pay too much attention.” But then, as Xing Wancheng recounted, “I couldn’t make as much money as before, so I couldn’t afford to help them as much either. Because of that, Tian Shi started to act cool toward me.” The two men continued to be friendly, but Tian Shi insisted on keeping their relationship on a strictly transactional basis, and she refused to let Xing Wancheng sleep with her anymore. Xing greatly resented what he saw as her lack of “qing” (love, feeling), but she was unmoved; Tian Shi’s priority was to support her family, and whatever affection she may have had for Xing did not extend to providing sex for free. Xing pled with Tian Shi to resume their sexual relations, but when he had no cash to offer, she rejected him absolutely—so he stabbed her to death.33 The end of this story is typical of the legal cases in which the wife takes a ruthlessly pragmatic attitude toward her relations with the outside male: when he can no longer help her family, she cuts off the sex, so he murders her—and that murder is why the matter ends up recorded in the legal archives. To some extent, such cases parallel wife-killing cases, where typically some perceived failure of gender duty on the wife’s part (refusing sex, for example) provokes her husband to kill her. These outside males clearly developed some of the same possessive sense of entitlement that many husbands maintained toward their wives. A classic example of this scenario can be found in an 1817 case from Pingluo County, Gansu. Yang Shi was the wife of a hired laborer named Luo Guogui, who was often away from home for long stints either working or seeking work. Their household hovered on the very edge of subsistence, until Yang Shi finally initiated a sexual relationship with a neighbor, Ding Sanyi, who supplied cash to support her and her children while her husband was away. This relationship lasted four years. Luo Guogui eventually found out about it, but he did not object because he

Chapter 2. Attitudes toward Polyandry

75

had no better solution to their problems. After a couple of years, Yang Shi began a second sexual relationship with another neighbor, Wang Yaozong, who also chipped in to support her family. She did not bother to inform her husband or her other patron of this new arrangement. In the meantime, Ding Sanyi ran short of funds. When it became apparent that he could no longer help out as before, Yang Shi cut off his sexual access in an entirely unsentimental way, and concentrated on cultivating her other partner. Ding refused to accept her decision, and, after harassing her for some time, he eventually murdered her.34 The Unsentimental Pragmatism of Peasant Women How are we to understand the practical, utilitarian attitude that these women took toward their sexual relationships? In his path-breaking study of Communist marriage reforms in China, Neil Diamant argues that, in contrast with the prudishness of urban intellectuals and elites, peasant sexual culture was strikingly “open.” Living in circumstances that allowed little private space, peasants were relatively unaffected by either Confucian or bourgeois sensibilities that would impose sharp distinctions between public and private. Diamant cites Communist cadres’ shocked complaints that peasants “do not separate public from private” (bu fen gong si),35 language that echoes the ubiquitous complaint in Qing legal cases that couples involved in polyandrous and polyamorous relationships “do not distinguish inner from outer.” Women, in particular, tended to view marriage, separation, and divorce in practical no-nonsense terms; as some said, “I marry to eat; If I don’t eat, I leave,” and “Firewood and rice make a husband and wife, no firewood or rice, the two separate” (you chai, you mi, shi fuqi; wu chai, wu mi, liang fen li). Usually, they married to guarantee for themselves, and perhaps their families, a better life. When things did not work out as hoped, they often did not hesitate to break up.36

Thus, Diamant argues, it was peasant women who were best able to take advantage of the Communist Party’s marriage reforms. Specifically, the prohibition of traditional arranged marriage and the promulgation of divorce on demand empowered these women to exploit sex and marriage as strategies for individual upward mobility.37 The frankly materialistic sayings quoted by Diamant recall the logic of sexual contract that framed the remarriage of poor widows, as seen in Qing legal cases that I have discussed elsewhere. This economic calculus is most explicit in cases where a widow refused to consummate her new marriage sexually (in hope of annulling it), because the groom turned out to be poorer than she had expected. As one widow bluntly informed her new husband, her only reason to remarry was “to secure food and warmth” (tu ge bao nuan). “Now I’ve come to your home and you turn out to be this poor—what should I marry you for?”38

76

part one

This experience of marriage as sexual-economic exchange recalls Friedrich Engels’s controversial, classic argument that marriage under conditions of gender domination is a wholesale form of prostitution: as long as women lack access to income-producing work outside the home, they must trade their sexual and other domestic labor for economic security. Therefore, a wife “differs from the ordinary courtesan only in that she does not hire out her body, like a wage-worker, on piecework, but sells it into slavery once for all.”39 Engels saw the situation as inevitably and absolutely oppressive of women; in his opinion, women could be liberated only when their choice of partners was based on true love alone, without any material factors playing any role at all.40 But in rural China, at least, a candid appraisal of marriage as simply one form of sexual-economic exchange sometimes empowered women to use sex as an asset they could trade to their own benefit.41 In Diamant’s analysis, it was precisely peasant women’s lack of bourgeois or Confucian pretensions—pretensions that would tend to mask the instrumental nature of marriage—that made possible their strategic pursuit of advantageous alliances. Many rural women were feisty, assertive, and uninhibited in taking advantage of new opportunities to pursue their own individual interests; many were frankly materialistic in dumping husbands who had been foisted on them in arranged marriages, in favor of more advantageous matches. The skewed sex ratio and shortage of wives that prevailed in the countryside only enhanced their opportunities to negotiate better terms for themselves. This kind of clear-eyed strategizing did not foster a feminist revolution, but it did enable many individual women to pursue their own interests quite effectively, by exploiting sex and marriage as strategies for upward mobility. In the end, Diamant argues, the biggest losers in the Communist reforms were the “poor, bald, and unattractive men” who had nothing to offer these hardheaded women and therefore found themselves excluded from the new marriage market created by the reforms.42 Equally, in many Qing polyandry cases, it was the women’s unsentimental pragmatism that made it possible to recruit patrons and secure their families’ livelihood.43 Another factor stands out in the Qing legal cases: a wife’s economic contribution enhanced her position in the household, especially if it became her family’s chief means of subsistence. Most obvious is the woman’s empowerment in relation to the men in her life. The clearest example is the scenario of the ill or disabled husband, who ends up depending entirely on his wife and her sexual partner(s) for survival; often he becomes extremely deferential, and she can initiate as many extramarital relationships as she chooses. But also, picture the situation found in many cases, where one adult woman is at the center of a nexus of relationships between men. She has what they all want, so it should not be surprising to find them deferring to her. Mao Zedong himself made a pertinent observation, in his famous 1926 “Report on the Peasant Movement in Hunan.” Many readers will be familiar with the pas-

Chapter 2. Attitudes toward Polyandry

77

sage quoted as an epigraph for this chapter—except, perhaps, for the middle two sentences: “In sexual matters, poor peasant women also have relatively more freedom. Among the poor peasant class, triangular and multilateral relationships are almost universal.” These lines have been cut from nearly all published editions of Mao’s report, presumably because they cast an unflattering light on the revolutionary vanguard. Mao’s overall argument in this section of his report is that, by smashing landlord power, the revolution will overthrow “the whole feudal-patriarchal ideological system” depending on that power, including “the masculine authority of husbands.” But in this passage, he suggests that poverty itself has weakened the authority of husbands, freeing their wives to sleep with other men and resulting in a proliferation of polyandrous and polyamorous relationships. Moreover, Mao’s tone is approving: he appears to endorse these relationships, because he assumes they reflect women’s liberation from the “chains” of patriarchy.44 One does not have to share Mao’s naïve optimism about the liberating effects of promiscuity to sense that he grasped an important part of the truth. His statement can be compared to the 1743 testimony of a landless peasant from Shahe County, Zhili, whose wife had initiated a sexual relationship with another man: “Since I couldn’t support her myself, I had no choice but to let her do as she liked.”45 Poverty did not necessarily liberate women (who might find themselves having to serve more than one man), but it did undermine “the masculine authority of husbands” in that strategies like polyandry forced them to yield their wives to other men. A Woman’s Sense of Self-Worth How did women themselves see the value of their contribution to household maintenance via polyandry? The case records contain many vivid examples of women’s awareness that their husbands were (as one wife put it) “relying on my body in order to survive” (kao wo shenzi guo rizi).46 Such comments often emerged during quarrels, when a wife would defy her husband’s authority on the grounds that it was she who supported their family, not he. In a 1757 case from Xinyu County, Jiangxi, for example, manual laborer Jiang Xingsan (44) and his wife, Liu Shi (36), pooled resources with a sock-maker named Du Guisheng in a polyandrous relationship. At one point, Jiang discovered that a few cash he had put by were missing, and he accused his wife of stealing them. Angered, Liu Shi retorted: “Don’t you forget, I’m using my body to earn Du Guisheng’s money in order to support you! How is it possible for me to steal your money? (wo jiang shenzi yu Du Guisheng zhuan qian yang ni, zenme hai tou ni de qian).” What comes through in such statements—especially in women’s specific mention of their bodies—is a clear-eyed perception that their sexual relations constituted a form of labor and source of earning power directly comparable to the labor of men.

78

part one

A fascinating case memorialized by the governor of Hubei in 1762 illustrates such self-perception from a different angle. The three Hu brothers, Zhaixiang, Tingxiang (35), and Lianxiang, were peasants who had immigrated to Baokang County from their home in Tianmen County (located some 225 kilometers to the southeast). They lived and farmed together as a single household and were very poor. Only the middle brother, Tingxiang, had ever married; he and his wife, Yu Shi (32), had two young sons. (Note the skewed sex ratio: five males to one adult female.) A close neighbor was Liu Yuheng (43), a single man who had emigrated from Xiangyang County (about a hundred kilometers to the east) several years earlier. Liu farmed maize on some rented hillside land. He and the Hu brothers became friendly and got into the habit of trading work and helping each other out in various other ways. In the spring of 1755, Liu asked Hu Tingxiang to have his older son pledge himself to Liu as “godson” (gan erzi), and Tingxiang agreed. From that time on, Liu and the Hu family began “using kinship terms to address each other,” and the Hus treated him as family, “without distinguishing between inner and outer.” Liu became very friendly with Yu Shi. That winter the Hu family ran short on food; the brothers also hoped to rent land from Liu’s landlord to begin farming in the spring, but they had no money for the deposit. Liu, however, had a bit of extra, so Hu Tingxiang asked him for a loan. Liu demurred, however, because he was saving up to do some sort of petty trading. Early the next morning, Hu Tingxiang sent his wife alone to Liu’s hut: Liu was still in bed, and she went right in, stood before his bed, and told him that her husband had sent her to ask for the loan. Liu understood the couple’s intent, and at once he pulled her into bed and they had sexual intercourse. After finishing, Liu agreed to make the loan on condition that the Hu family let him move in with them so that he could have easy access to Yu Shi. Her husband agreed, so Liu lent the brothers 7000 cash plus nine shi of corn (perhaps six months’ food supply for this family), and then he moved into a side room of their house. Liu usually ate separately, but Yu Shi (on her husband’s instructions) prepared Liu’s meals, and he had sex with her in his room when he liked. Unlike most of the relationships documented in legal cases, the sexual nature of this one was never openly acknowledged; according to Yu Shi, “my husband just pretended not to notice what was going on.” As husband Hu Tingxiang later confessed, I saw my wife coming and going from his room at all hours, without any boundaries whatsoever (bufen bici). I knew they were having illicit sex, but I never saw it with my own eyes; and I was very grateful to him for lending us the corn and money, so I didn’t really feel that I could challenge him about it (bubian yu ta renzhen).

Tingxiang’s brothers later claimed in court not to have known about the sexual relationship (had they known about the sex, they asserted, they never would have

Chapter 2. Attitudes toward Polyandry

79

agreed to repay the entire loan). But their claim seems incredible, given the intimate terms on which the five adults were living in a four-room house and the obvious quid pro quo that had secured the loan, not to mention the brothers’ adamant refusal to pay interest on the loan. Did the neighbors know what was going on between Yu Shi and the man they referred to as her sons’ “gandie” (“godfather”)? Their testimony was evasive and none openly admitted it, but they could hardly have failed to notice that (in the words of one witness) “Liu Yuheng and Yu Shi routinely ignored the boundary between family and strangers, they did not separate inner from outer (bulun qinshu nei wai).” As we have seen, this and similar phrases seem to have been standard euphemisms for polyandry and polyamory. This particular household’s modus vivendi is an example of the “don’t ask, don’t tell” scenario, in which everyone finds it convenient to turn a blind eye. In the winter of 1758 (after three years of co-residence), the Hu family’s house was damaged by fire, so Liu Yuheng moved out and set up a hut for himself nearby, but he continued to have sex with Yu Shi. Meanwhile, the Hu brothers gradually repaid their debt to him. But in the winter of 1760, after the principal had been paid, Liu suddenly demanded interest in the sum of over one shi of corn. This demand outraged the Hu family, who refused to pay. They stopped treating him as kin and finally cut off relations with him altogether. Yu Shi was particularly angry, and she refused to sleep with Liu anymore: “I saw how unfeeling (boqing) he was and got very angry at him, so after that I wouldn’t let him have relations with me (laiwang) anymore.” Liu kept trying to persuade Yu Shi to sleep with him, but she rejected him and became ever more insulting in her manner of doing so. On the final occasion when he approached her, she later recalled, I cursed him for being so lacking in human feeling (wuqing) and told him, “We already repaid all the money and corn we borrowed from you, and now you quarrel with us, and try to cheat us by pretending we owe you interest! If you want to sleep with someone, you can take your own sister and sleep with her (hebu la ziji zimei tong shui)!”

She added that, if he ever touched her again, she would scream and accuse him of trying to rape her. At that, he tried to strike her, and in the process accidentally injured her younger son, who was a toddler; the child later died. (Ironically, Liu Yuheng may well have been this child’s father.) The most revealing aspect of this case is the quarrel that eventually led to the violence. The Hu brothers’ desire to save face had prevented them from negotiating explicit terms of exchange, and this ambiguity may have provoked a genuine misunderstanding. Why were they so angry at Liu Yuheng’s demand for interest? Well, presumably that is what they thought the sex was for, along with the other benefits of co-residence that Liu had enjoyed—and the quarrel with Liu casts

80

part one

doubt on their pretense of ignorance about the means by which their sister-in-law had serviced the debt. In fact, their arrangement with Liu was a sort of hybrid between polyandry and a conditional wife sale, in which a wife would service a loan by sleeping with her husband’s creditor (see Chapter 3). Perhaps the most surprising thing about the quarrel is that Liu himself seems to have forgotten the quid pro quo that had given him conjugal privileges in the first place. Apparently, Yu Shi’s affectionate treatment had convinced him that she was sleeping with him simply out of love and desire. As he later recalled, he provoked her last outburst by complaining about her lack of “feeling” toward him—“Before, you were so intimate (qingmi) with me! Why do you turn away from me now?”— and trying to take her by the hand. His confusion on this point is precisely what made Yu Shi so angry, and she was the angriest member of the household. Until their falling out, she does seem to have had genuine affection for Liu Yuheng. After all, he had become part of her family: he was her son’s godfather, with whom they had lived and pooled resources, whose meals she had cooked, and whose bed she had shared for five years. Indeed, for all practical purposes he amounted to her second husband, tacitly if not explicitly acknowledged by the whole family. But she did not consider their relationship a romance. Her anger focused on what she called his attempt to “cheat” her family by demanding interest, even though, as she put it, they had already repaid everything they had borrowed (i.e., the principal of the loan). In effect, by demanding interest from the Hu brothers, Liu was taking her for granted: he was dismissing the exchange value of everything she had done for him—and one can only imagine what five years of sexual intercourse, food preparation, and other services might have cost if priced in cash. Today, it is commonplace to believe that sexual relations should be motivated only by romantic love (or at least mutual desire), and in particular to condemn any openly instrumental or materialistic use of sex. This is certainly the implicit or explicit position of many who have condemned prostitution, from Engels right down to the present, and, in the United States today, consensual sexual relations between adults are prohibited only when money changes hands.47 Whatever its merit, such a perspective does not help us understand the anger of Yu Shi. For her, there seems to have been no contradiction between a relationship of “feeling” (qing) or “intimacy” (qingmi) and an instrumental quid pro quo. On the contrary: in her view, it was precisely Liu’s refusal to honor the instrumental nature of their relationship and to acknowledge the exchange value of all she had done for him that demonstrated his “lack of feeling” toward her.48 The conflict between Liu Yuheng and Yu Shi was, among other things, a conflict between rival understandings of “qing,” which literally means “feeling” or “sentiment” but is often translated as “love.” The famous literary ideal of qing is a sublime and passionate form of romantic love, epitomized by the Ming dynasty drama

Chapter 2. Attitudes toward Polyandry

81

Mudan ting (Peony Pavilion), in which the power of qing both kills the heroine and later brings her back to life.49 This literary paradigm is the complete antithesis of the sort of instrumental pragmatism that informed Yu Shi’s approach to Liu Yuheng: she had no intention of dying for love, and the qing that she expected of him required that he acknowledge and appreciate the material value of her services. But Liu had somehow deluded himself into believing that their attachment was primarily one of sentiment. Affection and Passion, Jealousy and Violence Pragmatism notwithstanding, we do find many examples in which women developed strong bonds of affection and passion with their partners, especially when the women had negotiated these relationships themselves. In such relationships, there seems to have been no necessary contradiction between sentiment and more instrumental impulses, and sometimes the former even trumped the latter. A case from Chongyi County, Jiangxi, illustrates this pattern. This case involved tenant farmers who had migrated to Jiangxi many years before from Huizhou Prefecture, Guangdong (about 150 kilometers to the southeast). In 1749, when this case was reported, Deng Zihua was aged 68 sui and his wife, Huang Shi, was 41 sui (they had no children); her brother Huang Xuewen, who was single, was aged 34 sui. The three had migrated to Jiangxi together. Not long after arriving in Jiangxi, Deng Zihua had come down with an unspecified chronic illness that made it difficult for him to work, and for this reason Huang Shi had initiated a relationship of sexualeconomic exchange with a neighbor named He Hongyuan (aged 57 sui in 1749), who was also a migrant from Guangdong. He Hongyuan was single, and he lived and farmed alone. In exchange for regular conjugal relations, He helped support Huang Shi and her husband, forming a stable and intimate relationship that lasted more than a decade. He Hongyuan did not actually move in with the couple, but he lived nearby and spent much of his time at their home. He got along well with Huang Shi and her husband, Deng Zihua, and, under the circumstances, Deng did not object to their relationship. Deng seems to have played a passive and secondary role, keeping out of the way and content to be supported through his wife’s efforts. In 1747, a single migrant from Guangdong named Peng Tingcai (then aged 32) moved to Chongyi County and became friendly with Huang Xuewen, who by then was living apart from his sister and working at a shop in a market town. Peng made and peddled incense powder for a living. Since both men were from the same part of Guangdong and had been born in the same year, they pledged kinship as “sameyear brothers” (tongnian xiongdi). (This variation of sworn brotherhood is common in rural Guangdong.)50 As Huang later testified, he was drawn to Peng because the latter was “a man of honor” (you yiqi de ren)—“honor” or “a righteous spirit” being precisely the quality that, according to Boretz, defines the solidarity performed by sworn brothers.51

82

part one

After several months, Huang suggested that Peng Tingcai move to his sister’s village, because there was woodland nearby with trees that could be used to make incense. He introduced Peng to his sister and her husband, and arranged for him to move in with the couple and share their meals in exchange for paying rent. Because Peng and Huang were “same-year brothers,” Peng told Huang Shi that she should become his “same-year older sister,” and she agreed. Before long, Peng and Huang Shi began having sex, in exchange for which Peng turned over all his money for Huang Shi to “manage” for him. Huang Shi’s amiable husband acquiesced to this relationship as well. From his later testimony, it seems that Huang Xuewen did not expect Peng Tingcai to end up having sex with his sister, and he remained ignorant of the sexual dimension of their relationship until much later. Nevertheless, his introduction of Peng to Huang Shi was intended to give them both access to each other’s resources, and it illustrates a point that anthropologist David Jordan’s Taiwan informants emphasized: “Sworn brothers assume an obligation towards the family members of their fictive brethren,” as well as to brothers themselves, in a manner that includes the sharing of resources.52 By the time Huang Shi began sleeping with Peng Tingcai, it seems that her longstanding partner, He Hongyuan, was less able to support her and her husband than before, and Peng soon replaced He as the couple’s main source of support. Nevertheless, Huang Shi had a loving relationship with He Hongyuan, and she did not reject him; on the contrary, she continued to sleep with him. Moreover, she remained openly affectionate and intimate with He, making shoes for him, inviting him over for meals and to drink wine at the New Year, sitting with him (as Peng later testified) “just as if they were husband and wife,” and even giving him part of Peng’s money. Peng became intensely jealous, but when he told Huang Shi to cut off her relationship with He Hongyuan, she refused. After a year, Peng murdered He, hoping thereby to gain exclusive conjugal rights to Huang Shi.53 In this instance, instead of simply replacing her first patron when he could no longer hold up his end of the sex-exchange bargain, Huang Shi expanded her polyandrous network by adding another partner, and she even went so far as to share the second patron’s resources with the first. In this way, a stable triangular relationship morphed into the sort of transactional polyamory discussed in Chapter 3 below. But this expanded network proved untenable because Huang Shi’s two partners failed to establish a cooperative relationship with each other. Since polyandry was a chosen relationship, it should be no great surprise if sometimes it compared favorably to marriages that had been arranged by household heads and matchmakers with little regard to the prospective spouses’ personal feelings. If all marriage was instrumental to a significant degree, there is no reason that sentiment within normative marriage should necessarily have outweighed sentiment within polyandry—quite the contrary, especially if the latter

Chapter 2. Attitudes toward Polyandry

83

relationship had been initiated by the woman herself, with a man of her own choosing. A case from Tieling County, Shengjing, reported in 1817, tells of an alliance that Sun Defu (49) and his wife, Tan Shi (35), negotiated with an ex-convict named Li Er (in his thirties); these people were all migrants to Manchuria from the north China plain who worked as casual laborers. The couple became friendly with Li, and when he proposed to move in and pool resources with them, they agreed. They rented a one-room hut with a single kang where the three adults and the couple’s two young sons all slept together. Tan Shi liked Li very much, and the two began having sex literally behind her husband’s back: one night, Sun woke up to find them having intercourse right next to him there on the kang. He uttered a few curses but otherwise did not interfere, because he feared it would be hard for his family to get by without Li’s help. From then on, Tan Shi and Li Er enjoyed sex openly, uninhibited by his proximity. She clearly preferred Li to her own husband. Within this triangular relationship, there is no question that the woman’s alliance with her lover had become the primary alliance. Eventually Sun did try to expel Li, under pressure from the landlord renting them their house (he felt nervous about harboring a conspicuously tattooed exconvict). But Li refused to leave, and Tan Shi defiantly took his side: if Li had to go, then she would go with him. Sun very much resented being sidelined in this way. In the end, he murdered them both, while they lay sleeping in each other’s arms. 54 In a case from Shaoyang County, Hunan, also reported in 1817, a single man named Liu Kuanze (35) became sexually involved with Wang Shi (33), the wife of his distant cousin Liu Shengmo (36). Kuanze was already friendly with the couple and often dropped by their home; one day he and Wang Shi took advantage of her husband’s absence to have sexual intercourse. Just as they were finishing, her husband arrived home and caught them in the act. He threatened to turn them in to the authorities for prosecution, but Kuanze kowtowed to him and promised to pay him if he let them off. Shengmo agreed. This became an ongoing arrangement, with Kuanze and Wang Shi frequently meeting for sex and developing a passionate relationship, which her husband Shengmo tolerated in exchange for regular contributions of cash and grain. The relationship continued for two years, until Kuanze found himself unable to continue his payments. At that point, Shengmo attempted to cut off the relationship; he became violent, beating his wife and threatening to kill her and her partner if they continued to see each other. The lovers’ passion was the driving force in this relationship. Wang Shi refused to give up her lover, and she ultimately chose him over her husband. In the end, they conspired to murder her husband, hoping to live as a married couple themselves.55

84

part one

This last case exemplifies yet another pattern seen in these legal cases, where the wife’s consciousness is transformed by the experience of her relationship with the outside male, especially when she has an abusive, violent husband. The second relationship is obviously preferable to her marriage, and in effect it allows her to imagine a different life, a way out of her trap. When pressed to terminate that relationship, she ends up turning against her husband. In the most extreme examples, she murders him, in hope of creating a better life for herself with the second man. An alternative also seen is for the pair to run away together. Such violence underscores the limits to women’s agency in these situations. We have identified certain areas of subaltern power and agency that played an important role in our spectrum of polyandrous practice. The evidence in legal cases cannot be explained without taking into account peasant women’s assertiveness, their pragmatic deployment of sex as an asset, their ability to influence men through shame and gossip, and the ways in which supporting their families with their bodies might empower them vis-à-vis the men in their lives. But we should not forget that the larger context that fostered such strategies was a pervasive market for women’s bodies and that the shortage of women fueling this market was the result of lethal and pervasive discrimination. As a result, it was possible—one can even say that it made sense—for a husband to share, prostitute, or sell his wife in order to survive. Nevertheless, to understand how these practices actually worked, we must reckon with the paradox of women’s poverty-driven power within strategies of sexual commodification. T H E C O L L A P SE O F B OU N DA R I E S

From the standpoint of Qing orthodoxy, marriage and prostitution constituted irreconcilable opposites. Marriage depended on the absolute chastity of a secluded wife—a clear separation of “inner” (nei) from “outer” (wai)—whereas prostitution implied the untrammeled promiscuity of a public woman. This basic binary distinction was vital to elite status and lifestyle as well as imperial ideology throughout the Ming-Qing period. But if we survey the Chinese marriage system from the bottom of the socioeconomic scale, that clear-cut binary distinction cannot be sustained. From this perspective, we learn that sex work in one form or another might even play a decisive role in the preservation of marriage and family. In poverty-driven polyandry, the distinction between marriage and sex work collapses, as a wife exchanges her sexual and other domestic labor with an outside male, with her husband’s approval, in order to help maintain her family. This strategy depended on a frank assessment of the instrumental nature of arranged marriage and of the fact that a woman’s domestic duties within marriage, including sex, were not necessarily so different from a wide range of ways that her labor might be exchanged for material benefits in extramarital contexts.

Chapter 2. Attitudes toward Polyandry

85

This collapse of boundaries requires us to rethink marriage, kinship, and gendered power relations within the household. If we look at the Chinese marriage system from the bottom up, then it suddenly makes sense, because polyandry falls into place as the necessary third piece of the puzzle, alongside the polygyny of the elite and the monogamy of the middling peasantry. If we define kinship as strategic practice, then we can account for the many chosen relationships of people who found it necessary to seek alliances outside the normative family system in order to survive. If we reckon the importance of poor women’s work—including sex work—to their families’ livelihood, then we begin to understand how those women apparently most exploited by traditional patriarchy might actually be able to negotiate their own sexual lives and even end up as the effective heads of their own households. This reconsideration does not mean we should glamorize the often sordid and desperate lives of the Qing dynasty poor. But it does require us to get past elite norms and judicial categories, to understand why these people behaved as they did and how they understood their own behavior.

3

The Intermediate Range of Practice It’s common for rural people to sell illicit sex because of poverty (xiangjian ren yin pin mai jian ye shi changshi). — testimony of yu hualong (26), an itinerant barber from yishui county, shandong, who was the outside male in a long-term polyandrous relationship based on sworn brotherhood, recorded in 1750 1

The evidence all says that extramarital affairs were common and tolerated if not approved. It does no good to argue that people in Haishan were poor and lacked the benefits of a Confucian education; that was the fate of peasants everywhere in China. The image of the Chinese woman as a virtuous prisoner concealed behind high walls must go the way of the view that all Chinese families were four-generation households ruled by benevolent patriarchs. — arthur wolf and chieh-shan huang, “marriage and adoption in china, 1845–1945” 2

This chapter examines the intermediate range of practices between formal polyandry and outright wife sale: a variety of arrangements whereby a wife would have sexual relations with one or more other men, with her husband’s approval, in exchange for material support for her family. The full spectrum is illustrated in Table 2. Proceeding from left (polyandry) to right (wife sale), one can visualize this spectrum as the progressive alienation of the wife from her husband, his household, and his home village. By “alienation,” I refer to separation and transfer of custody, but the psychological connotations of the term may also be relevant here. On one end of the spectrum, polyandry and polyamory often took place at the couple’s own home; in many cases, a basic purpose was to enable them to remain at home on their land. These were mainly rural practices, as were both forms of wife sale (on the other end of the spectrum), whereas the two forms of retail prostitution found at the center of the spectrum usually took place in urban settings, even though most of the people involved were peasant migrants who had become 86

Chapter 3. The Intermediate Range of Practice

87

table 2 A Spectrum of Practice from Formal Polyandry to Outright Wife Sale Formal polyandry: “getting a husband to support a husband” or sworn brotherhood

Transactional polyamory: 2–4 regular patrons help support family, without co-residence

Retail prostitution with husband as pimp

“Pledging”/ “contracting” wife into retail prostitution

Conditional wife sale

Outright wife sale, either direct or indirect

separated from the land. The midpoint of the spectrum represents a profound break: to the left of that point in the table, the woman would remain with her first husband and be part of his household (even as, in moving toward that midpoint, she would sleep with an increasing number of sexual partners and move from country to town); whereas to the right of that point, her husband would transfer her to the custody and control of others, either temporarily or permanently. In terms of female agency, a woman’s control over her own fate seems to have peaked at the two extreme ends of the spectrum, polyandry and wife sale. In contrast, her vulnerability to coercion and other abuse would be greatest in retail prostitution, in part because its urban setting took most women far from the kinship and community networks that might otherwise have supported them. My schematic presentation imposes a somewhat artificial clarity. Most of the cases discussed in this book can be placed on this spectrum. But details vary, and many blur the ostensibly neat categories outlined here, whereas others involve the simultaneous pursuit of more than one strategy, or a strategic shifting between them. Nevertheless, these are the patterns that emerge, and in this chapter we proceed from left to right along this spectrum to examine the four intermediate categories of practice seen there. T R A N S AC T IO NA L P O LYA M O RY I N RU R A L C H I NA

In addition to formal polyandry, many legal cases document a more casual arrangement that involved a similar quid pro quo but no permanent co-residence or resource pooling, nor a contract or chosen kinship. In this scenario, which I call transactional polyamory, a couple would recruit two or more patrons from the local community—the cases suggest a maximum of four or five—who would help support the couple in exchange for occasional sexual relations with the wife.3 The patrons would drop by to deliver supplies, perform labor, and have sex, sometimes spending the night and sharing a meal. These relationships tended to be more casual and fluid than formal polyandry, and the details vary, but some were stable

88

part one

and lasted for years. In other words, they were not the one-off exchanges of sex for cash that characterized retail prostitution; rather, this form of transactional sex took place in the context of ongoing relationships that often included emotional bonds. In terms of our spectrum, these relationships represent a step away from formal polyandry toward retail prostitution. Consider the following 1757 case from Pengze County, Jiangxi, in which a peasant couple engaged three single peasant men to support them in exchange for sexual access to the wife. Yang Shimou (40) and his wife, Hu Shi (37), had four surviving children: three sons (aged 14, 4, and 2), plus a daughter (6) who had been adopted out in infancy as a daughter-in-law. The family rented and farmed hillside land. Yang fell ill with jaundice, which compelled the couple to hire an unmarried peasant named Li Chugui (34) to help with farm work. But they lacked means to pay Li’s wages, so Hu Shi began having sex with him, on the understanding that Li’s labor would be compensated in this way; it was she who took the initiative to negotiate this understanding. After the term of Li’s hire was over, his relationship with the couple continued, but instead of working their land, he contributed food and wine. For farm work, the couple hired Chen Huanxi and Cheng Qida, who were single men in their twenties. Both slept with Hu Shi in lieu of wages, and, again, it was she who negotiated these terms. In short, a stable and ongoing arrangement developed in which three young, single men chipped in to help support a family, in exchange for regular sexual relations with the wife. None of the men moved in with the couple; instead, they would visit and occasionally spent the night. The couple’s dwelling had three rooms: Hu Shi slept in one, her husband and sons in another, and the visitors in the third. Hu Shi’s three sexual partners were all on good terms with each other and often visited together, sharing meals and taking turns having sex with her in her room. Indeed, the couple’s home became a social center for these men: they gained not only sex but also opportunities to socialize and something like a family life that otherwise would not have been available to them.4 The arrangement was quite open, and no one interfered. According to Yang’s younger brother, who lived nearby, “everyone knew” that Yang relied on his wife’s sexual relations with these men to support their family; the brother felt shame, but his only response was to stop visiting the couple. The couple’s neighbors testified that they knew what was going on, but they sympathized because of Yang’s poor health. Another factor, perhaps, was that this couple had three tough young men as patrons who could be counted on to defend them. The rural agent (dibao), who lived just over a kilometer away, testified that he too had heard about Hu Shi’s promiscuity, but as long as her husband did not complain, he “did not dare interfere.” This arrangement continued for several years, but tensions emerged as Yang found himself increasingly sidelined, tolerated but not respected. One day, Cheng brought over some pork and wine to share with Hu Shi and her other two patrons.

Chapter 3. The Intermediate Range of Practice

89

Yang had been out, and by the time he arrived home the others had eaten all the meat. He scolded his wife, but she scolded him back, accusing him of laziness (“Why should we save any for you?”), and Li Chugui defended her, telling Yang “a man like you should be satisfied ‘eating prepared rice’! How can you expect to eat meat too?” “To eat prepared rice” (chi xian cheng fan)—prepared by others, rather than through his own labor—means to be a parasite. These insults angered Yang: he complained that “my home has been completely taken over by you men,” and he vowed to file charges in order to sever ties with them. But Hu Shi had come to prefer her lovers to her husband, whom she resented for his increasingly abusive behavior; taking his threat seriously, she persuaded the three men to help her murder him. By the time of Yang’s death, as he clearly realized, Hu Shi had become the effective head of their household. It was she who had taken the initiative to secure her family’s subsistence by trading her sexual labor, and by doing so she had made herself the center of a web of relationships between four men. In the end, it was Yang’s inability to adapt to this realignment that provoked his murder.5 But not all husbands were so obstreperous; others accepted the necessity of yielding authority and status, and hence they remained on good terms with their wives and patrons. In a 1747 case from Jiading County, Jiangsu, a couple allied with four landless, single peasants. The couple resorted to this strategy after the husband came down with a “swelling disease” that limited his ability to work. The couple’s four patrons were all very poor, and they obtained resources for the couple mainly through theft. This arrangement lasted six years, during which two of the men were arrested, beaten, and tattooed for theft, but upon being released they resumed their relations with the couple as before. Throughout, the husband meekly accepted his secondary role, avoiding any conflict with his wife’s lovers, and she remained loyal to him, manipulating the other men in order to secure her family’s survival. Trouble arose when one of them failed to uphold his end of the bargain and she tried to sever relations with him, provoking a violent quarrel in which her other patrons intervened to defend her.6 A case reported by the governor of Shaanxi in 1745 illustrates another variation of this pattern. Wang Shi (36) and her husband, Yao Xian (62), were impoverished peasants who lived in Lintong County. Wang Shi had been sold to Yao as an adopted daughter-in-law at the age of 11 sui, when he was 37 sui, and their only child was a daughter who had herself already been adopted out. The couple lived in an earthen pit dwelling in a ravine with only two other households close by. One household consisted of the three Yao brothers (cousins of Yao Xian)—Chongjin (a casual laborer who was away much of the time), Han’er (47, a widower), and Cang’er (31)—along with Han’er’s son Mo’er (19). None of these men had wives. The second household consisted of Yao Simin (54), who was the paternal uncle of the three Yao brothers. Simin’s wife was long dead, and his son worked elsewhere as a

90

part one

laborer and seldom returned home. Simin lived by himself, but he seems to have been slightly better off than the other two households, and he was the dominant figure in the little community. Note the very high ratio of males to females, which is typical for such cases. This stark symptom of poverty exposes the expendability of female children, while simultaneously underscoring how one woman could become the focal point of a network of men. Wang Shi was the only female in this community of six men, and she had sexual relations with four of them. Simin, Mo’er, and Cang’er were all helping to support her and her husband by sharing food, money, and labor. She had begun sleeping with Simin after she and her husband failed to repay a loan of grain, and he had been supporting them ever since, for six years. Her connection with Simin was very open, and he would sleep over at their dwelling whenever he liked. The other two relationships began when Wang Shi borrowed money from each man and offered to sleep with him in lieu of repayment. These, too, had become ongoing relationships of sexual-economic exchange; at first she concealed them from her first patron, who was the other men’s uncle, but eventually all the men became mutually aware of their relations with her. All of these men lived within easy walking distance of the couple’s home. When the other men visited his wife, husband Yao Xian kept out of the way—he would go sleep in another earthen pit with the animals. He seems to have been rather cowed by his hardheaded young wife, and he depended on the income she procured from his younger, more vigorous cousins. This situation continued until Wang Shi had a falling out with her oldest patron, Simin, and then recruited the two younger men to help her murder him. A notable feature of this case is the assertiveness of Wang Shi. She may have begun married life as a purchased adolescent, but, by the time she began sleeping with Yao Simin, it was she who was running her household. Her relations with the other men were initiated and negotiated by her. The case follows the pattern of many, in which dependence on the wife’s body seems to enhance that woman’s power in her relations with other men, including her own husband. The fact that she was the only woman in this group of half-a-dozen men reinforced her assertiveness.7 Why did such cases end up in court? In xingke tiben like the ones narrated above, we find three basic scenarios of trouble that led to homicide. In the first, the husband refuses to accept a subordinate role, provoking a confrontation with his wife and/or her patrons. In the second, the woman’s patrons have a falling out among themselves, usually because one tries to monopolize her at the expense of the other(s).8 In the third scenario, the woman tries to cut ties with one of her patrons, usually because he has failed to hold up his end of the bargain. We must bear in mind that conflict, violence, and people prone to them will inevitably be overrepresented in legal cases. Nevertheless, these scenarios suggest the tensions

Chapter 3. The Intermediate Range of Practice

91

inherent in such arrangements and the potential difficulty of establishing an equilibrium along such unconventional, non-patriarchal lines. In all of these cases, presumably, other options had been considered and rejected in favor of this one. These couples chose this particular option over the others: in the beginning, at least, they wanted to stay together (thereby ruling out wife sale), and they were sufficiently uninhibited to exchange sex for economic resources. But they did not wish to bring any one outside male permanently into their household, nor did they wish (or need?) to resort to the opposite extreme of retail prostitution. Moreover, a factor in at least some cases was the woman’s enjoyment of her relationships with the other men. Occasionally, we find an arrangement that began with outright polyandry (in which just one outside male was fully incorporated into the family), but later the family again fell on hard times, so that another male had to be recruited—but, because the couple got along well with the first man, they were loath to sever ties, and therefore ended up adding a second man without getting rid of the first. In this scenario, we find a couple first opting for polyandry to avoid wife sale, and then expanding their polyandrous network in order to avoid breaking it up.9 These households all engaged in other work aside from sex work—farming, for most, but many cases also mention women’s handicrafts—and the wives’ sexual relations supplemented these other sources of sustenance. From this perspective, one can see polyamory as part of the larger portfolio of survival strategies in the context of agricultural involution whereby peasants mobilized “underutilized” forms of household labor in order to maintain subsistence levels. The particular strategy discussed here resembles the way peasants with farms too small to support their families would hire themselves out as short-term or part-time laborers while continuing to farm their own land. This process of partial or semi-proletarianization (to use Philip Huang’s terms) constituted a defining feature of the poor peasant economy during the Qing and Republican periods.10 Such peasants’ condition contrasted with the full proletarianization of those who had lost their land and survived solely through wage labor. In the scenarios we have discussed, a wife’s polyamorous relations in exchange for supplementary income represented a sexual version of partial proletarianization, which enabled a family to stay together on the land—in contrast with the full proletarianization seen when peasant couples became separated from the land and survived through retail prostitution in urban settings. But polyamory also represented an extension of the way women’s labor in marketable handicrafts provided a crucial supplement to farm income in the involuted peasant economy. In a 1798 case from Guizhu County, Guizhou, we find the same arrangement transplanted to an urban setting. Zhang Tianyun (33), who hailed from Pingyue County, worked as a porter in the provincial capital (present-day Guiyang). He was the uxorilocal husband of Tian Shi (20), who was the only child of widow Tian

92

part one

Wang Shi (68). Zhang had a hard time supporting his family, but he knew three young, single men from Pingyue County who had also migrated to town in search of work. They were good friends with Zhang and with each other, and they often visited the couple’s home. In the early spring of 1797, just before the New Year, Tian Shi began having sexual relations with the three men in exchange for their promise to support her family by providing grain and modest sums of cash when they visited her. Tian Shi’s husband and mother both approved of her initiative. The three friends knew of each other’s relations with Tian Shi, and the arrangement seems to have been friendly on all sides, while it lasted. Trouble occurred one day when the three visited and got drunk. One of the men tried to remove Tian Shi’s clothing in front of the others, offending her and provoking a fight. The other two men beat him up, and he later died of his injuries.11 The polyamory described in this section was mainly a rural practice that, like other strategies of partially proletarianized peasants, aimed to provide supplementary income to enable them to remain on the land. In this last case, however, a band of dislocated peasants pursued polyamory in the provincial capital. They bonded on the basis of shared native place, and it seems to have been a natural step to cement their relationship through sexual-economic exchange. This case provides a convenient transition to our next topic: retail prostitution in the context of marriage, which was almost always an urban phenomenon, even though the couples involved were usually migrants from the countryside. R E TA I L P R O ST I T U T IO N I N T H E C O N T E X T O F M A R R IAG E

Elite ideology in the Ming-Qing era held marriage and prostitution to be irreconcilable opposites. Nevertheless, all the evidence indicates a strong interdependence between marriage and prostitution in social practice. After all, the classic paradigm for prostitution before the Yongzheng reforms was the “prostitute household,” whose hereditary debased status provided the context for legally tolerated sex work. The most famous example was the “music households” (yue hu) targeted by the first reform edict in 1723. In that venerable paradigm, the husband/ father acted as manager and pimp while his wife and daughters supported the household through sex work and other stigmatized entertainment services.12 Despite criminalization, marital prostitution remained a pervasive pattern. I do not mean to suggest that most wives worked as prostitutes, but the majority of prostitutes—that is, women who engaged in retail sex work with multiple customers—do appear to have been married. Certainly, most of the prostitutes found in Qing legal records were married women doing sex work in order to help support their families. If these records accurately reflect social reality—and on this issue, at least, I have no reason to think otherwise—then most retail sex work actu-

Chapter 3. The Intermediate Range of Practice

93

ally supported marriage in that it enabled poor couples to survive without permanently separating.13 The case records document two common forms of retail prostitution that involved married couples.14 In the first, a couple worked together on their own, with the husband acting as tout to recruit customers to have sex with his wife. Such couples often worked out of rented rooms at an inn or tavern, with the proprietor’s collusion. The second form was brothel prostitution, in which a husband contracted his wife to the brothel-keeper for a fixed term during which the couple would be separated. Both forms of retail prostitution were mainly urban phenomena, being concentrated in market towns and cities where large numbers of clients might be found. Nevertheless, most of the couples found in such cases were peasants who had been separated from the land either permanently, by losing it altogether, or temporarily, through some acute crisis such as harvest failure. They might go to an urban area in order to seek customers but also to avoid having people back home know what they were doing (some women adopted suggestive pseudonyms); or they might find themselves stuck in an urban area as refugees and end up resorting to prostitution because they had no other means of livelihood. Even in big-city brothels, many (perhaps most) prostitutes were married. For example, an 1813 case from Ba County records a raid on a brothel located on Daijia Alley, in the Linjiangmen neighborhood of Chongqing, in which two pimps and twelve prostitutes were arrested, along with the yamen runner responsible for the neighborhood (he had been collecting protection money from the pimps).15 Ten of the twelve prostitutes were married; most were peasants from neighboring counties who had come to Chongqing with their husbands. Five women had been contracted into prostitution by their husbands, at least two of whom were invalids entirely supported by their wives. The other women had contracted themselves out of sheer desperation; two were widows, and five were married women whose husbands were missing or had gone off to work on some job, leaving them without means of support. Their situation suggests the lack of options for women who found themselves on their own in Chongqing.16 During the Republican period, Gail Hershatter observes, many of the prostitutes in Shanghai brothels were married women or widows, the percentage being highest among “lower-class prostitutes.”17 Surveys conducted in Shanghai in the late 1940s and 1950s found that anywhere from 32 percent to 54 percent of brothel prostitutes were married women or widows. Christian Henriot attributes these “high” percentages of married women to the influx of war refugees, but it is not clear why he assumes that percentages were lower at an earlier point in time.18 Significantly, the landmark studies by Hershatter and Henriot both focus on organized brothel prostitution; neither addresses the scenario discussed below, in which a husband acted as his wife’s pimp outside a brothel setting. Exact quantification is

94

part one

impossible, but this less institutionalized form of marital prostitution appears to have counted for a large share of the sex trade in places like Chongqing during the Qing dynasty. Retail Prostitution with Husband as Pimp A 1762 case from Chengwu County, Shandong, illustrates the basic scenario of a couple working on their own in retail prostitution. This case concerns the murder of a peasant named Yuan Liu who had been pimping his wife, Ming Shi. Her confession provides a succinct account of how the couple took up this line of work: My husband liked to eat but was too lazy to work, so he would not continue living with my older brother-in-law. Instead, he sold the land he had received in household division and spent all the money he got for it. . . . Then he wanted me to leave home with him and do these shameful things. There was nothing I could do but follow him. We stayed at Song Xianye’s inn. My husband gave me the name “Charming Jade” (Qiaoyu) and made me receive customers and sell illicit sex.

The couple had left their home village and moved to a market town in the next county over so as to avoid the humiliation of having kin and neighbors find out what they were doing. By the time of her husband’s murder, Ming Shi had slept with dozens of customers, and, when interrogated, she could remember the names of only two regulars who lived nearby. These two turned out to be the murderers: they had hoped by eliminating her husband to get free use of her and to live off the proceeds of her prostitution with others.19 A 1756 homicide case provides a glimpse of a similar arrangement. Li Gui (42) and his wife, Jia Shi, were peasants from Linqing Department, Shandong, who left home because of poverty, making their way to a market town in Yongnian County, Zhili, where they rented a room at a rate of 300 small cash per month, and they “made a living by selling illicit sex.” Shi Rong (40) worked at the inn. One night he invited his boss Li Youguang and Jia Shi to drink wine and eat donkey meat together. The three sat together on the kang and got drunk, Jia Shi sitting between the men and flirting with them, and Shi announced that he wanted to have sex with Jia Shi. But she rebuffed him, saying she did not want to sleep with a mere employee, preferring the boss instead—and with these words, she embraced Li Youguang. Li agreed to her proposition, provoking a drunken brawl with Shi Rong, and Li ended up stabbing Shi to death. During the fight, Jia Shi had passed out drunk on the kang and later claimed to remember nothing.20 We can find the same basic pattern in many cases from Ba County, of which the following 1840 example is representative. Peasant Liu Zhengwei took Zhu Shi (30) as wife in 1825, when she was aged 15 sui. According to Zhu Shi’s testimony, her husband was to blame for their poverty because of his laziness and “failure to work at his proper occupation” (bu wu zhengye). In 1836, the couple left home and

Chapter 3. The Intermediate Range of Practice

95

migrated to Chongqing (about fifty kilometers from their village), where they ended up making ends meet through prostitution. After four years, Zhu Shi’s brother tracked the couple down, discovered what had happened, and took her to the county yamen to file charges against Liu. The magistrate ordered the couple deported home under Zhu Shi’s brother’s supervision.21 It is not unusual in such cases for the wife to express antipathy and contempt for her husband, especially for his alleged “laziness” that had caused their poverty and made prostitution necessary in the first place. What comes through is a clear sense that these women disliked retail prostitution—and here we can discern a contrast with both polyandry and less formal polyamorous arrangements, which often took shape as a result of women’s own initiatives. Any generalizations based on this kind of evidence must be impressionistic. Nevertheless, it makes sense that many would have found retail prostitution relatively distasteful. In retail prostitution, women seem to have had little say over with whom they had sex; moreover, having sex with multiple customers created an inherently unpredictable situation that probably involved greater danger (of violence, disease, being cheated, legal trouble, etc.) than steady relations with one man or a small number of men. In addition, polyandry and polyamory often took place in circumstances where a woman could exercise a degree of control (including choosing her sexual partners), and, as we have seen, the formation of such relationships might enhance her power relative to the men in her life, even making her the effective head of her extended household. In contrast, retail prostitution usually took place in circumstances where women found themselves socially isolated and therefore vulnerable to coercion and exploitation by their husbands and especially by other parties—after all, these were mainly rural women who had migrated to urban centers far from the kinship and community networks that might otherwise have protected them. It is not surprising, then, that in many cases we find women trying to improve their security by cultivating a regular client deemed reliable and trustworthy, and husbands might well encourage this effort. A woman might form what amounted to a polyandrous relationship with one client that would proceed in tandem with retail prostitution, or she might seek to transition out of prostitution altogether through such an alliance. The same phenomenon occurs in the modern sex trade, when a woman engaged in retail prostitution secures a transactional relationship with a favorite customer, a strategic repositioning that may result in long-term connections and even marriage. Movement along the sex work spectrum in the direction of marriage appears to be a goal for many modern prostitutes, who share much the same motives and desires as the women in our Qing legal cases.22 A 1756 case from Henan illustrates how a prostitute’s regular customer might end up joining her and her husband in a polyandrous relationship. Zhang Shi (aged 47 sui in 1756) and husband Cheng Binyan (about the same age as she) were originally from Ru Subprefecture in central Henan. According to Zhang Shi’s

96

part one

testimony, Cheng had been orphaned as a teen, not long after marrying her, and then quickly used up his modest inheritance by drinking and gambling. In 1726, Cheng took her to the county seat of Xiangcheng County (about seventy-five kilometers to the southeast) and “forced” (yi le) her to work as a prostitute. She was 17 sui at the time. Cheng himself “sang opera songs” to supplement her income. For more than two decades, they made a fairly decent living in this manner and eventually purchased a boy and a girl for adoption. (Zhang Shi gave birth to no children herself—like many prostitutes, she may have been made infertile by sexually transmitted disease.) But by 1752, Zhang Shi had reached the age of 43 sui, and, as she later testified, she “had gotten old” and had “lost her looks,” so fewer men were willing to sleep with her. The couple fell into debt, and Cheng was forced to pawn his collection of opera costumes and props. The couple then recruited one of Zhang Shi’s few regular customers, a single, itinerant actor named Li Youcai, to “live together and pool resources.” As Zhang Shi later testified, “from then on, Li Youcai often slept with me and had illicit sex with me, and my husband never bothered us. Whenever Li Youcai earned any money from singing and acting, he turned it over to us to help pay our household expenses.” But Zhang Shi also continued to turn tricks when she could. The family (including Li) moved to the riverine port of Sheqidian (about 120 kilometers to the southwest in Nanyang County), in hope of finding more customers. They continued to have difficulty, however. Cheng resented his wife’s declining earning power, and he became abusive and violent. He proposed to sell off their adopted daughter, provoking a major quarrel with Zhang Shi, who refused to sell the girl. Moreover, Zhang Shi had come to prefer the mild-mannered Li Youcai to her increasingly abusive husband, whom she finally murdered. Li Youcai had simply been one of Zhang Shi’s customers, but, after being recruited into her household, he began to consider himself—and to be considered by the couple—as more than that. Zhang Shi’s testimony makes this clear. When the magistrate suggested that Li had “abducted and run off ” with her (because he continued to live with her and the children after her husband’s murder), she contradicted him: “That’s just not how it was. He had been living with my family and pooling resources with us all along, so he just stayed with us and continued to do so. There was certainly no ‘abducting and running off ’ involved.” Li Youcai echoed her testimony: “I was already living with them together as one family (yijia guohuo), so I just stayed with them.”23 We find a similar scenario in a 1738 case, also from Henan. Wu Si (50) and his wife, Chen Shi (50), were peasants from Changheng County, Zhili. In 1731, they migrated to a market town in Shangqiu County (about 120 kilometers to the southeast, in Henan), where they rented lodgings at an inn and Chen Shi began working as a prostitute to support their family. A young, single man named Han San (26) was Chen Shi’s regular and favorite client. In 1736, Han also began paying

Chapter 3. The Intermediate Range of Practice

97

to have sex with the couple’s daughter Xiaobo, then aged just 15 sui (i.e., 13 or 14 years old). This was Xiaobo’s initiation into prostitution, and she soon began sleeping with at least two other clients. But Chen Shi worried about the future. According to her subsequent testimony, because of her age (she had reached her late forties by then) she could not attract enough clients to earn a good living, but also she hoped to enable her daughter to quit prostitution and enjoy a better life than she herself had had. In 1737, she proposed that their client Han San take Xiaobo as wife, on the assumption that Han would undertake to support her and her husband as well. Han San enthusiastically agreed to this plan, but Wu Si demanded a brideprice of 10 taels for his daughter. Han managed to borrow a total of 7 taels, but Wu Si rejected this sum and treated Han rudely, declaring that he would sell the girl to someone else as concubine for a better price. Wu Si’s attitude made Chen Shi very unhappy, for several reasons. In her view, he cared only about short-term profit, never considering their long-term prospects; moreover, he expressed no concern for their daughter’s welfare, being content either to exploit her indefinitely or to sell her to the highest bidder. For six years he had relied on Chen Shi’s body for his living, nevertheless he refused to defer to her wishes. Finally, Chen Shi decided that she had no future with Wu Si and that it would be best to kill him, quit prostitution, and establish a new household by marrying their daughter to Han. She secretly proposed this plan to Han, offering him Xiaobo without brideprice, and Han agreed. With the help of a friend, he murdered Wu Si.24 In both of these cases, the crisis erupted over the wife’s desire to transition out of prostitution into a more viable long-term situation, and an important flashpoint for conflict with her husband was her wish to provide for her daughter. Both cases also illustrate how a woman’s age and ability to recruit customers might shape her calculations. Women who had long been able to support their families with retail prostitution began worrying about the future when they realized that they could no longer attract as many customers as before. The evidence suggests that women began to worry after they entered their forties, when their reproductive years had passed. In wife sales (as we shall see in the next chapter), the late thirties represented a rough upper age limit beyond which a wife could be difficult to sell, because the main priority of most prospective buyers was to have sons. In the prostitution cases, the concern was sex appeal rather than fertility, but the two factors seem to have aligned closely. Shifting between Strategies The following 1750 case from Sichuan illustrates how one couple might shift between various strategies, depending on circumstances, in the process demonstrating remarkable geographical mobility. It also highlights one wife’s desire

98

part one

ultimately to quit both prostitution and polyandry and become the wife of a more reliable man. In this case, landless peasant Yao Rufu (36) was born in Yibin County; in 1737, at the age of 23 sui, he migrated about forty kilometers east to Nanxi County, where he married Li Shi (28) as an uxorilocal husband (at the time she was just 15 sui). The following year, however, Li Shi’s father died and her mother remarried, leaving the couple to fend for themselves. By 1742, the couple had lost whatever assets they had inherited from Li Shi’s father, so they moved to a town on the Yangzi River in Pingshan County (about a hundred kilometers to the southwest), where Li Shi supported them through retail prostitution. As she later recalled, “that town was a busy port where many travelers and merchants came and went, and my husband would go entice (gouyin) men to come spend the night with me and have illicit sex.” They lived this way for over a year, until Yao’s older brother, Yao Ruhan, heard what they were doing, tracked them down, and fetched them home to Yibin County to help work his farm. But (according to Ruhan’s testimony), Yao Rufu was “too lazy to be a farmer,” and, just after the New Year, he took his family and ran off without telling his brother. The couple traveled to Changning County (about sixty kilometers southeast of Yibin), where Yao eventually found temporary work making roof tiles. The couple became friendly with a tenant farmer named Chen Guotai (28), who was originally from Jiang’an County (about forty kilometers to the northeast); Chen was single and lived alone, and he lusted after Li Shi, so he lent the couple money and often visited with gifts of wine and meat. After a couple of months, Li Shi began sleeping with Chen, and Yao did not interfere. In her words, “Chen Guotai became accustomed to coming and going (laiwang guan le). He often gave us some cash, or bought wine and meat to share with us, and then he would have illicit sex with me, all without deceiving my husband. . . . We would laugh and talk right in front of my husband and he didn’t mind at all.” By this time the couple had three children, and Yao Rufu found himself unable to support his family. After six months, he proposed that Chen move in with them, promising to “yield” (rang) Li Shi in exchange for Chen’s support. Chen agreed and moved in with the family, and he and Li Shi shared a bed, while Yao Rufu and the children slept separately. Chen paid for all their “food and fuel”; as Li Shi summarized, “my husband always preferred ‘to eat prepared rice’ ”—that is, to live off the labor of others. After several months, rumors spread and Chen began to feel “embarrassed” (bu hao yisi), so he proposed that they move to Yongning County (about sixty kilometers to the southeast) and rent land to farm together. Yao and Li Shi agreed, and they made the move, passing themselves off as a single family: Yao and his children adopted the surname “Chen” and pretended that he and Chen Guotai were brothers. When they first moved to Yongning County, Chen Guotai still had some sav-

Chapter 3. The Intermediate Range of Practice

99

ings, but these soon ran out, so that they depended entirely on farming to survive. Yao resented this decline in their standard of living as well as the hard labor now required of him, so he began pressing Li Shi to separate from Chen in order to “find someone else who has money, so we can have an easier life.” But Li Shi resisted this pressure, and Yao grew angry and abusive. She informed Chen Guotai what was going on, and she asked him to find some way for the two of them to become “long-term husband and wife” (changjiu fuqi). Matters came to a head when Li Shi berated Yao for his shamelessness: “Now that you’ve spent all his money, you want to kick him out—don’t you have any conscience at all?” Yao cut her with a knife, and Chen intervened and killed Yao with axe.25 This case illustrates the broad pattern we have seen, in which a peasant couple who were separated from the land would migrate to town and end up resorting to retail prostitution in order to make ends meet. But this form of prostitution was only one option in a spectrum of survival strategies, including more stable polyandrous or polyamorous arrangements. Individual couples might engage in more than one such strategy over a period of years, along with more prosaic ways to earn a living, such as hiring out labor and engaging in farm work. Many cases document a remarkable degree of geographic mobility as people moved from village to town, thence to other urban centers in different counties and even provinces, in search of a viable livelihood. A common theme is women’s anxiety about their long-term prospects and their desire to form stable households. Women’s anxieties and desires played an important role in shaping the shifts between strategies found in these narratives. A corollary is their resentment of husbands whom they came to perceive as parasites and as obstacles to their own security and happiness. “Pledging” a Wife into Brothel Prostitution The second form of retail marital prostitution found in legal cases required a husband to “pledge” his wife to a pimp or brothel-keeper, for whom she would work for the duration of her contract; during this time she would live apart from her husband, under the pimp’s supervision. (A variation involved pledging a daughter or daughter-in-law instead of a wife.) In structure, this kind of transaction had much in common with conditional wife sale (to be discussed below), but it appears far more frequently in Qing court records. The term of service for a wife contracted to a brothel was relatively short: usually no more than one or two years, and sometimes as little as a few months. Often, the pimp would provide the husband with a loan in exchange for his wife’s services, the principal of which would usually have to be repaid when her contracted term was over. But, in some cases, the pimp would pay her husband a share of the income she generated as a stipend, instead of (or in addition to) the loan. Generally speaking, the shorter the contract, the less the husband would receive.

100

part one

Such transactions blurred the boundaries between wife-selling, polyandry, and prostitution, combining elements of all three, and they highlight the ways in which the sex trade and marriage were mutually dependent. By serving as temporary contract prostitutes, wives could provide vital income that sustained household survival without permanent alienation from their husbands and children. The generic term for this kind of transaction seems to have been “diya” (or simply “ya”), which is usually translated as “mortgage” or “pledge.” (The term “bao”—which means to contract the exclusive use of something—was also used in some places.) When applied to land transactions, diya referred to the mortgage of land to secure a loan. Since the land did not actually change hands, diya must be distinguished from a conditional sale (dian) of land, in which the buyer took possession for the duration of the contract.26 With regard to women, however, “diya” was much closer to a conditional sale, because for the duration of the contract the woman would remain in the custody of the pimp, for whom her sexual labor provided interest payments on the money he had lent her husband. Once the agreed term was over, she would return to her husband. Investigation of Customs records this practice of pledging in Jiangsu, Anhui, and Zhili. In “various counties of northern Jiangsu,” the practice was known as “binding out” (chu kun) and was often done with unmarried daughters or adopted daughters-in-law, as well as wives. A “binding contract” (kun qi) would be drawn up, and the woman would be turned over to the brothel keeper for a term of either three or five years. The brothel-keeper would pay the “binding price” (kun jia) according to a schedule set down in the contract. For example, if the agreed price was 100 yuan, then he would pay 40 yuan in the first year, 30 in the second year, and 30 in the third year, with a new promissory note being written for the outstanding amount and modified in each successive year. It is not clear from the report whether the “binding price” had to be repaid in the end, but practice probably varied depending on the amount of money and the duration of the contract.27 According to investigator Zhu Chengyou, who filed his report in 1918, this “evil custom” was “not at all improper” (bing fei bu zhengdang) in the eyes of local people. But it often provoked litigation, because if disagreement arose, the woman’s in-laws or natal family would file charges accusing the brothel-keeper of forcing her into prostitution. Their real motive to file charges was that they wanted more money.28 We find much the same custom described in Yuhu County, Anhui, where it was called “binding a prostitute” (kun ji): “A wife or daughter is pledged to work in a brothel,” known as the “binding host” (kun zhu), for a period of twelve to sixteen months, in exchange for a loan. The terms of the transaction would be recorded in a “body binding contract” (kun shen ziju), and, at the end of the term, the loan would have to be repaid for the husband to regain custody of his wife. Contracts would stipulate that, if the woman fell ill or became pregnant, she would have to make up for lost time by working beyond the loan period.29

Chapter 3. The Intermediate Range of Practice

101

Similarly, in Tianjin County, Zhili, “poor families will pledge their wives or daughters to work in brothels in exchange for cash loans, in order to obtain the money necessary to feed themselves.” The borrower (i.e., the woman’s husband or parent) will provide the brothel-keeper with a promissory note that specifies the woman’s length of service and guarantees that the loan will be repaid in full at the end of that time: “Only then will the wife or daughter be allowed to leave the brothel.” The investigator notes that his information came from “many legal cases” involving this “immoral custom.”30 My richest evidence about the temporary contracting of wives into prostitution comes from Chongqing, where such transactions provided much of the workforce in brothels. I find examples in the entire chronological range of my Ba County case sample, from the mid-eighteenth century through the early twentieth century. Details vary, but the underlying logic of these transactions is the same as that documented elsewhere in China by the sources cited above. A case from 1759 involved a pimp named Wang Gui (24), who, together with his cousin, had brought two prostitutes to Chongqing by boat from Mao Independent Department (about 300 kilometers to the northwest). These women (aged 25 and 30) had both been “contracted” (bao) to him the previous year by their husbands in exchange for loans, for terms of ten and eleven months, respectively. Wang and his cousin were spotted by the Yangzi River patrol and arrested on suspicion of trafficking; after interrogation, they were transferred to the Ba County yamen for prosecution. To justify himself, Wang Gui submitted his contracts and explained how he had acquired custody of the two women—whom he consistently referred to as “girls” (yatou), despite their age:31 One of the girls is named Yang Baojie. On the 23rd day of the sixth month of last year, her husband Hei Yi agreed to contract her to me for 28 taels of silver. He has already paid me back 24 taels, and still owes 4 taels. According to the contract, I must return home by the twentieth day of the fourth month of this year, when the balance of the money and the girl will be exchanged. The other girl is called “Little Wu,” and she was also contracted to me on the 23rd day of the sixth month of last year by her husband Zi Er, who got 28 taels of silver. He has already paid me 20 taels, and still owes 8 taels. According to our contract, I must return home by the fifth day of the fifth month of this year, when he will pay me the balance of his debt in exchange for his wife. I contracted these two girls from their husbands and brought them here, as proven by the contracts—I am telling the truth. I beg your honor to examine the contracts and release me.

Wang’s account was corroborated by his cousin, the women, and the contracts. From his testimony, it appears that he may not have realized that the transactions were illegal; at any rate, his main priority was to convince the authorities that he had not abducted the women (potentially a capital offense)—hence his eagerness to have the magistrate peruse his contracts. Wang did not state explicitly why he

102

part one

had wanted to contract these women from their husbands, but as the head of the patrol noted (in his report to the magistrate), “it is obvious that they have been engaged in prostitution (mai chang).” The logic of these transactions is that the pimp’s income from the women’s sex work would constitute interest on the money borrowed by their husbands. For that reason, only the principal had to be repaid. Each husband would pay the final installment in return for his wife.32 Some Ba County cases sum up this formula with the following phrase: “When the money arrives, you take back your wife” (yin dao qu ren).33 We find a slightly different arrangement in a Ba County case from 1840. A Chongqing vegetable peddler named Liao Duan (originally a peasant in neighboring Changshou County) “contracted” his wife, Liao Zou Shi, to a widow, Li Yang Shi, who ran a small brothel in the city. Liao Duan wrote the contract himself; this barely literate document, which survives in the case file, is so full of wrong characters and malapropisms that only a rough translation is possible: I, Liao Duan, hereby establish this affidavit (chengren zi). The situation is as follows: I am from Changshou County. Because trade is poor, and I have no means of survival, I have fled to Ba County, where I now contract to Li Yang Shi’s brothel (chang men). The two sides have agreed face to face on a monthly wage of 1300 cash. This money has been paid in full, and nothing more is owed. Through mediators, we have agreed that this contract will terminate in the first month of the New Year. Fearing that spoken words alone are not reliable, I hereby establish this affidavit as proof. Mediators: Old man Qiao, Old man Zhang, Old man Liu Witness: Chen Gang On Daoguang 20.9.25 the person establishing this affidavit Liao Duan has written this himself.

As we can see, this arrangement was intended to last only a few months, to get the couple through the New Year; instead of a loan, the husband was to receive a monthly stipend that he apparently was not required to refund. Curiously, the contract nowhere directly mentions Liao’s wife, although the substance of the agreement is not in doubt. This case came to court because four days after the contract was signed, Liao went to the brothel and demanded more money. Li Yang Shi refused, an altercation ensued, and Liao filed an oral complaint at the county yamen accusing her of abducting his wife. Li Yang Shi submitted the contract to disprove Liao’s claims, and when the case came to a hearing, he admitted he had entered into the agreement voluntarily: The fact is, I went broke and couldn’t buy vegetables to sell anymore, so I told my wife to go to Li Yang Shi’s household to sell sex as a prostitute (mai chang). I was promised 1300 cash per month, and I drew up an affidavit acknowledging this. On the 29th,

Chapter 3. The Intermediate Range of Practice

103

I went back to Li Yang Shi’s home to demand (suoyao) more money, but she refused, so we had a big quarrel.

The magistrate ordered Liao beaten for “abetting his wife’s prostitution” (zong qi mai chang) and Li Yang Shi slapped for “harboring prostitutes” (zhao liu mai chang). But he took pity on Liao’s poverty and returned his wife to him, making the couple promise to stay out of trouble, and not confiscating the money Liao had been paid.34 We find much the same kind of arrangement in several cases from the very last years of the dynasty. In a case from 1909, for example, one Liu Hongshun (32) “pledged” his wife, Liu Ju Shi, to one Wang Xingsheng (42), who operated a brothel in Chongqing, in exchange for a “body pledge price” (ya shen jia) of 20 yuan. The contract survives: I, Liu Hongshun, hereby establish this pledging contract (zhi ya wenyue). The situation is as follows: In order to pledge my wife Liu Gou Shi to Wang Yusheng, I have asked Xiong Xingfa to act as mediator and negotiate to pledge her for exactly 20 yuan. I, Liu Hongshun, have already received this full amount, and not a cent more is owed. The time limit for my use of this money is one year, after which I must repay the full amount. If Liu Gou Shi should happen to suffer any illness or other misfortune during the term of contract, then both sides agree to submit to Heaven’s will, and may not say different. Any earnings or property [garnered by Liu Gou Shi during the term of this contract] shall remain the property of Wang, and I shall make no claim on them. In the future, when I take back Liu Gou Shi, there will be no mistake about these matters. Fearing that spoken words alone are unreliable, I hereby specially establish this pledging contract as proof. Witnesses: Li Xingfa Wang Hongshun Xiong Dingchen On Guangxu 34.12.12 this contract is established by Liu Hongshun [mark]

Like many wife sales, this transaction took place just before the New Year when, by tradition, most debts fell due. Six months later, however, Liu demanded more money from Wang and, when rebuffed, filed charges accusing the brothel keepers of forcing his wife into prostitution while he was away (his plaint claims that he had hired his wife out to work as a seamstress). In response, Wang submitted the contract to prove that Liu was lying. The magistrate confirmed that the brothelkeeper was telling the truth, but he ordered both him and a partner slapped, and out of pity for Liu’s poverty, he returned his wife to him without punishing him or making him pay back the loan.35 The use of contracts as evidence in these cases reminds us of one basic purpose of wife sale contracts (to be discussed in Chapter 4): to prove that the husband had sold his wife voluntarily—that the buyer had not abducted her. If such a case went

104

part one

to court, the contract would prove guilt of the lesser crime of wife sale (or in the present instance, pimping) but also exonerate the bearer from far more serious crimes, such as abduction. C O N D I T IO NA L W I F E S A L E

Conditional wife sale had much in common with pledging a wife into brothel prostitution, but the main purpose of conditional sale was surrogate motherhood (rather than retail sex work), and therefore the contractual term was usually much longer. Moreover, a conditionally sold wife would be expected to have only one sexual partner, her buyer. Therefore, this transaction represents a step farther along the schematic spectrum toward outright wife sale.36 As Chapter 6 below explains in detail, there were many close parallels between wife sales and land sales, including similarity of contractual forms and the common practice of demanding supplementary payments from the buyer in order to generate further income for the seller after the sale was ostensibly complete. The vast majority of wife sales were absolute and final, in that the transfer of the woman was intended to be permanent (the next five chapters of this book focus on transactions of that nature). But there was also an explicitly conditional form of wife sale that closely followed the model of conditional land sale: their common characteristic was that the “conditionally sold” (dianmai) asset could eventually be “redeemed” (shuhui) by the seller. In other words, conditional sale allowed a peasant household to leverage a vital asset to raise emergency credit while retaining the option of recovering that asset in the future. In a “conditional wife sale” (dian qi), a woman would be transferred to her buyer for a limited period of time, after which she would return to her first husband. Like outright wife sales, a conditional sale would be negotiated through a matchmaker and formalized in a contract written in the seller’s name that would stay in the buyer’s possession along with the woman. From the standpoint of polyandrous practice, conditional wife sale can be seen as yet another arrangement by which a wife would share another man’s bed in order to help support her family. The Classic Form as Practiced in Zhejiang Conditional wife sales (and especially the term “dian qi”) appear to have been especially common in the province of Zhejiang, for reasons that are not clear, although the practice certainly could be found in other parts of China as well (we return to this issue below). Investigation of Customs lists nineteen counties in Zhejiang where conditional wife sales were known to take place, and other sources document the practice elsewhere in the province.37

Chapter 3. The Intermediate Range of Practice

105

How did a conditional wife sale work? The following account is representative: In the middle and lower strata of Longyou County society, the custom of conditional wife sale is very common. The contract will specify the time limit, and the price will be very low. When the time is up, the wife will return to her husband, but any children she bears during the period of conditional sale will belong to the buyer, and the conditionally sold wife may not take them with her.38

As this passage indicates, the wife’s sexual and reproductive labor belonged to the buyer for the duration of the contract, and he was entitled to keep any offspring she bore him. Such wives no doubt contributed much else of value to their buyers’ households, including domestic caring work and economically productive labor. But in the classic form of conditional wife sale, the buyer’s main priority was reproduction. The report from Jingning County is explicit: In a conditional wife sale, the seller’s motive is money, while on the buyer’s part there are two possible motives. If he has not married and cannot afford a wife of his own, then he may conditionally buy another man’s wife as a temporary measure [in order to secure a son]. If he has married, but his own wife has borne no sons, then he may conditionally buy another man’s wife in order to have sons.39

In other words, a conditionally sold wife was expected to serve as a surrogate mother to provide offspring for a man who could not afford the greater expense of acquiring a permanent wife or concubine of his own. Some contracts for conditional wife sales even stipulate that the women must breastfeed any infants they bear for a minimum period of time in order to ensure that they thrive. For example, a Republican-era contract from Zhejiang contains the following language: Chen Awang voluntarily agrees to sell his wife, Zhang Shi, in conditional sale to Yang Lin’gao in order to bear him children (sheng nan yu nü). . . . During the term of conditional sale, any son or daughter Zhang Shi bears shall belong to Lin’gao for him to raise to adulthood, in order to continue the Yang family line. But Zhang Shi should breastfeed any infant she bears (zan you Zhang Shi rubu) for at least one full year, after which Lin’gao will care for the infant.40

In other words, Zhang Shi would be required to nurse her baby for at least one year, even if that time period exceeded the agreed term of conditional sale. Given the priority of reproduction, a conditionally sold wife would have to remain with her buyer for a substantial period of time—various sources report from at least three years (seen as the minimum length of time necessary for a woman to get pregnant, bear a child, and breastfeed it until weaned), up to a maximum of ten years.41 These were much longer time commitments than required when husbands pledged their wives into brothel prostitution. The time limit would be recorded in the contract, and the longer the period, the higher the price of the woman’s services.42

106

part one

Within this overall pattern, there were variations. A key question is whether the first husband would have to repay the conditional sale price in order to “redeem” (shu, shuhui, huishu) his wife when the contractual term ended. In most places, repayment seems to have been standard: in effect, the conditional sale price was a loan, and the woman’s sexual and reproductive services constituted interest, so that only the principal would have to be repaid. This form of the practice followed the model of conditional land sales exactly.43 In some counties, the first husband could get his wife back without repaying the conditional sale price, in which case that price seems to have been considerably lower than otherwise; in this scenario, the woman’s sexual and reproductive services constituted both principal and interest, so that her term of service ended when the entire debt had been paid in this medium. This practice was standard, for example, in Yongkang County, where conditional wife sales appear to have been unusually common, even by Zhejiang standards. But if the buyer did not wish to return the wife, then at the end of the designated period he could negotiate “to pay the conditional seller a supplementary payment,” whereupon the wife’s relationship with her first husband would be terminated.44 In other words, this final “supplementary payment” (here termed “zhao jia”) would convert the conditional sale into an outright sale, in the manner of the final zhao tie paid to render a conditional land sale permanent. As this detail shows, a wife bought in a conditional sale cost considerably less than one bought outright. When considering such arrangements, it is worth remembering how expensive credit was for the rural poor. Unless one had land to put up for mortgage or conditional sale, or generous relatives willing to provide a loan, the most likely source of credit would be a pawnshop, but pawnshops required collateral, and their terms were expensive. A common practice was for peasants to pawn winter clothing and bedding in spring, hoping to redeem them in the fall before the onset of cold weather. The Qing code limited interest on loans (including pawn) to 3 percent per month (i.e., an annual rate of 36 percent), also stipulating that the total interest could not exceed the amount of principal originally borrowed, but moneylenders generally ignored these regulations, which frustrated officials felt obliged to reaffirm repeatedly throughout the dynasty.45 Interest rates more than double the official maximum were not uncommon, because, in reality, “what the peasant pays is dictated by his necessities, and by the ability of the lender to take advantage of them.”46 Some people might avoid such expensive rates by participating in revolving credit associations, but participation required one to contribute a share of funds up front.47 People desperate enough to resort to conditional wife sale or contracting a wife into prostitution would be unlikely to have access to such resources. It is not clear how much stigma attached to conditional wife sales in Zhejiang. In some places, there seems to have been none at all, at least from the buyer’s

Chapter 3. The Intermediate Range of Practice

107

standpoint. In Lishui County, for example, the family of the conditional buyer would celebrate this relationship as “a formal marriage” (zhengshi hunyin), offering congratulations and continuing to treat the woman as the buyer’s wife (or concubine) even after she had returned to her first husband. If she bore the buyer a son, then she would even be recorded in the lineage genealogy, and the relationship would in every way “be seen as a real marriage” (shi zuo youxiao zhi hunyin).48 My sample of xingke tiben contains several examples of the classic practice of conditional wife sale in Zhejiang. The following two cases from the mid-eighteenth century are especially rich in detail; they add flesh to the descriptions found in the early twentieth-century surveys of custom. Conditional Wife Sale in Zhejiang—Example #1 The first case, from Renhe County, illustrates several strategies that, together with conditional sale, constituted a menu of related options. In this case, peasant Chen Yuanfu and his wife, Qi Shi, had married at some point in the late 1730s, when she was 14 sui; they had no children. At the age of 25, in 1746, Chen fell seriously ill and was no longer able to support himself or his wife, so just after the New Year he approached her father, Qi Yongchang, and asked for a compensated divorce. But Qi refused to get involved. As he later testified, “I told him, ‘My daughter has already married out, so how can I be expected to take care of her? Since you can’t support her anymore, I’ll leave it up to you two to figure out a solution.’ ” After being rebuffed by his father-in-law, Chen engaged a friend named Shen Erren to act as matchmaker and, through him, arranged a conditional sale to a local man named Yang Hongmao who had no wife. Yang, who was 35 sui at the time of this transaction, worked with his brother and a cousin making cakes and other sweets for a living. Yang agreed to pay 24 taels of silver for a ten-year term, after which Chen would be entitled to redeem Qi Shi—as she later testified, “I was hoping eventually to return to my first husband’s home.” The contract reads as follows: Chen Yuanfu hereby establishes this conditional sale contract (dian qi). The situation is as follows: I have a wife named Qi Shi. Now the harvest has failed, leaving me with no means to pay the land tax or to get enough food to eat (guan liang wu ban, kou shi wu du), and in addition I am handicapped by illness. For these reasons, I can no longer support my wife, and so I am willing to sell her conditionally to Mr. Yang as wife, so that he can secure his own line of descent. For conditionally selling my wife, I, Chen Yuanfu, shall receive 24 taels of silver, and when ten years have passed, I may redeem her. This sum of money has today been paid in full. In future, if any other party should raise this matter or threaten litigation, then I, the first husband, shall take full responsibility on my own, without in any way involving the conditional buyer. There is no deceit involved, and we have acted voluntarily; in future there will be no regret or change of heart. Fearing that spoken words alone are not reliable, I hereby establish this conditional sale contract to serve as proof forever.

108

part one

The three sides have agreed face to face that the term of conditional sale shall be ten years. On Qianlong 11.2.? Chen Yuanfu establishes this conditional sale contract [mark] Witness to conditional sale: father Chen Zihui [mark]

This contract was written by matchmaker Shen Erren, but (as Chen Yuanfu testified), “he feared that later there might be trouble, so he refused to write his name on the contract.” Shen later died and so could not be summoned to testify. His anxiety about future trouble suggests that, in his community, some controversy must have attached to conditional wife sale. Even so, it is obvious that such transactions were common, and there was a regular and well-understood way to arrange them. Chen’s father, Chen Zihui, was the only other witness to the transaction. Qi Shi’s father, Qi Yongchang, was not directly involved in the transaction, although by washing his hands of his daughter’s fate he presumably understood that some sort of wife sale would be inevitable. After the transaction, Chen Yuanfu visited Qi Yongchang to explain what had transpired, and presented him with 2 taels from the conditional sale price. According to Qi’s testimony, at first he refused this money, but eventually he allowed himself to be persuaded to accept it. An important feature of the contract is its mention of Yang Hongmao’s desire to secure his line of descent. In that respect, the transaction was a success: four years later, Qi Shi gave birth to Yang’s son. In the winter of 1749, however, Yang found himself unable to repay a loan from his neighbor Li Dahua (a single man of 46 sui), so instead he told Qi Shi to sleep with Li; there quickly developed a polyandrous relationship in which Li supplied most of the couple’s needs in exchange for regular sexual access to Qi Shi. This arrangement was very open: their landlord and neighbors all knew about it, and no one interfered. After a few months, however, Li’s resources also ran short, so Yang Hongmao announced that he would have to sell Qi Shi off: as she later recounted, “he told me that since he got me for 24 taels of silver, he was going to sell me for 24 taels of silver.” Here we have a close parallel to the secondary conditional sale of land (“zhuan dian,” rendered as “sub-dian” in the official translation of the Republicanera civil code), in which the conditional buyer would resell the land conditionally to a third man, in a maneuver akin to subletting an apartment.49 Yang’s proposal to resell Qi Shi followed a similar logic. But since he planned to demand the full amount of money that he himself had originally paid for her, such resale would have amounted to transferring her contract to a third party (not unlike selling debt). In the future, Qi Shi’s first husband, Chen Yuanfu, would presumably have to redeem her from this third man. But all this is moot, because Yang never got the chance to resell Qi Shi. After learning of his intentions, she asked her polyandrous partner Li Dahua to buy her, but Li could not come up with such a large sum of money, so he proposed to her

Chapter 3. The Intermediate Range of Practice

109

instead that they simply murder Yang in order to get him out of the way. Since Qi Shi feared the prospect of being sold to a stranger, she eventually agreed to this plan, which Li executed by giving Yang wine laced with arsenic. Suspicion immediately fell on Li and Qi Shi (since everyone knew about their sexual relationship), and they were arrested and, after confessing, convicted of murder. A crucial factor in sentencing Qi Shi was whether she and Yang had been legally married (since more severe penalties applied to a wife who murdered her own husband than to someone who murdered an unrelated party). For this factor we are fortunate, because it explains the case record’s exhaustive documentation of the conditional sale. Since conditional wife sale was illegal, Qi Shi was sentenced as an unrelated party, rather than as Yang’s wife. In addition, her first husband was sentenced for conditional wife sale, and the money Yang had paid for her was ordered confiscated (including the 2 taels Chen had given her father).50 Qi Shi’s son by Yang was designated “a son born to adultery” (jian sheng zi) and was turned over to the custody of Yang’s brother (according to the principle that a son belongs to his father’s lineage).51 This case demonstrates three ways a woman could be leveraged for credit and cash income: compensated divorce, conditional wife sale, and polyandry. Since Yang Hongmao had acquired Qi Shi by conditional purchase, it seems to have been logical and easy for him to solve his subsequent financial difficulties through polyandry and resale. Conditional Wife Sale in Zhejiang—Example #2 In the second case (from Xuanping County, reported in 1769), one wife was conditionally sold twice in consecutive transactions. Jin Shi (39, from Yongjia County) had originally entered the household of peasant Wang Yunqi (42) at the age of 12 sui as an adopted daughter-in-law, and she had no surviving natal family. The couple had two sons and two daughters and because, as Wang later testified, “too many people were eating,” they were very poor. Moreover, for two years Wang had suffered a chronic illness that prevented him from working regularly. They sold two rooms of their house, but, by the early spring of 1768, they had run out of resources and faced the prospect of starvation. A couple of days after the New Year, Wang told his wife that “instead of starving to death together, it would be better to survive by going our separate ways” (e si zai yi chu buru gezi taosheng). With her agreement, he engaged two local men, Cai Jixian and Liu Guofa, to help find a conditional buyer for her. Cai and Liu negotiated Jin Shi’s sale to a single immigrant from Jiangxi named Hu Zixiu (40), who made a living peddling hemp cloth and had recently bought a piece of local land for the purpose of growing hemp. It was agreed that Jin Shi would join Hu Zixiu for six years in exchange for a “body price” of 6 taels, with the stipulation that she would take all four of her children with her to be supported by Hu for the duration

110

part one

of the arrangement. After six years, she and her children would return to Wang. A contract was drawn up, but Hu was able to pay only 1 tael in cash, so Wang agreed to accept a promissory note for the balance of 5 taels, because his chief priority was to secure food for his wife and children. Jin Shi entered Hu’s household immediately, but matchmaker Cai Jixian kept both the contract and the promissory note, with the understanding that they would be handed over to Hu only after he paid the outstanding 5 taels. The matchmakers acted for free, out of charity for the couple’s poverty, but Wang used half a tael to buy wine for them to drink at the conclusion of the exchange. The union was not happy because, as Jin Shi would later recall, Hu Zixiu “resented the fact that I was so old.” Apparently the matchmakers had been less than candid with him. Hu’s dismay at her age, 39 sui, reflects the fact that he had hoped for a son—the usual motive for buyers in these transactions—and, upon learning how old she was, he despaired of realizing this hope. After just a month, Hu left her and her children alone while he returned to Jiangxi to visit relatives. He was gone for several months, and in the meantime she ran out of food and had no idea when he would return. Finally, she took her children to the home of matchmaker Cai Jixian and made a scene, so Cai took her and the children back to her first husband Wang and handed over the conditional wife sale contract and the promissory note. Wang was very angry at what he saw as Hu’s violation of their agreement, so he burned the documents and began looking for another buyer. After a few days, Wang’s fellow villager Chen Wenze (43, also from Jiangxi) dropped by—having heard about recent events—and complained to Wang that he had no wife to cook for him or to watch his home while he worked in the fields. Wang immediately offered “to conditionally sell my wife to him for a few years,” on the same terms Hu had accepted. Chen objected that “if you want me to support your children for you, then 6 taels is too much to ask for your wife’s body price,” and Wang, anxious to feed his children, finally agreed to accept just half that amount. Cai and Liu again acted as matchmakers and drew up a new contract: Wang Yunqi hereby establishes this conditional marriage sale contract. Several years ago I took Jin Shi in marriage as my wife and she is now 39 sui. Because we are extremely poor, we have trouble getting our daily food, and now our family can endure hunger no longer. Therefore, I have begged matchmakers to draw up a contract in order to sell my wife conditionally to Chen Wenze for him to support for a period of six years. This day I have received from Chen exactly 3 taels of silver in payment for this conditional sale contract. I have received full payment and now establish this contract the very same day. During the six years that Jin Shi is supported by Chen any sons or daughters she bears will belong to the Chen family. When the six years are up, Jin Shi will return home as before. I hereby establish this conditional sale contract for Chen Wenze to keep as proof.

Chapter 3. The Intermediate Range of Practice

111

This contract is established on Qianlong 33.2.? by Yunqi [mark] Matchmakers: Cai Jixian and He Caiwen; scribe: Liu Guofa [mark]

The recorded testimony says nothing about Chen’s desire for children, but, as in the case from Renhe County, this contract makes explicit the reproductive purpose of conditional wife sale. The bargaining that preceded the transaction and the lack of any mention of repayment in the contract show that Wang was not expected to pay back the body price of 3 taels. Despite the wording of the contract, Chen could pay only half a tael up front, so he gave Wang a promissory note for the balance of 2.5 taels. Once again, Wang felt he had no choice but to agree because his overriding concern was the welfare of his children, whom he could not feed on his own. Later on, when Wang asked for the balance owed, Chen threatened him with a beating, so he desisted. Two months later, however, the first buyer, Hu Zixiu, finally returned to discover his house empty, and, upon learning what had happened, he demanded that Wang Yunqi either refund his down payment of 1 tael or return Jin Shi to him. Wang referred him to matchmaker Cai Jixian, who mediated a settlement whereby the second buyer Chen would be allowed keep Jin Shi in exchange for promising to pay Hu 1000 cash (the nominal equivalent of 1 tael) within a month. But when Hu came as scheduled to collect this sum, Chen asked for an extension of three months. Hu accused him of cheating and of “scheming to take another man’s wife,” and Chen retorted: “But I’ve already slept with the wife, and right now I don’t have any money for you—so what can you do about it? (qizi yi shui le, rujin meiyou qian ba ni, ni nai wo he).” The two men quickly came to blows, and Hu ended up stabbing Chen to death.52 Conditional Wife Sale in Other Regions As many sources testify, the classic form of conditional wife sale outlined above and especially the term “dian qi” were both strongly associated with Zhejiang. But one can find similar transactions elsewhere in China, sometimes called by the same vocabulary.53 A 1774 case from Tong County, Zhili, uses the term “dian” to refer to a proposed transaction that appears identical to Zhejiang practice.54 Similarly, an ethnography published in 1943 makes the following comment about marriage in Taiwan: “The middling classes and above strictly adhere to monogamy. But people of the lower classes do not. . . . For example, some men conditionally sell (dian) their own wives to other men, hoping by this shameful means to secure their own survival.”55 Elsewhere, different vocabulary was used. The practice of “marriage by pledge” (diya hun) in Fujian sounds exactly like conditional wife sale in Zhejiang. In Guangze County, for example: There is also the custom of “marriage by pledge,” in which a time limit is specified, and any sons born during that time will acknowledge the second husband as father.

112

part one When the time is up, the wife will be redeemed by her first husband and return to him. There are also some cases in which the first husband is not required to redeem his wife—it depends on the terms specified in the marriage contract.56

The early Republican-era anthology Qing bai lei chao (Qing Miscellany), which contains snippets of gossipy information from a variety of Qing texts, describes much the same practice in Gansu under the rubric “renting a wife” (zu qi). According to this account, wife-rental was a strategy for “a man who cannot afford to marry but who wants a son,” and the transaction, recorded in a written contract, would last for two or three years, or simply until the woman had borne a son. If the renter wished to extend the contract, he would negotiate an additional payment.57 We find something similar in Ba County. For example, a case from 1790 records the following transaction. Xiong Fu (34) and his wife, Zhang Shi (26), had migrated from home in Da Independent Subprefecture in 1787, and the couple had fallen on hard times. Unable to support his wife, Xiong arranged to “contract” her to one Ran Guangsi (in his fifties) as concubine, in exchange for 200 strings of cash for use as “business capital.” (Ran also hailed from Da Independent Subprefecture and had a small shop in Chongqing.) Zhang Shi and her little son then moved in with Ran, and she eventually bore him a daughter. This matter came to court three years later because Xiong began demanding more money from Ran (he may also have tried to take Zhang Shi back) and finally filed charges against him for “forcibly occupying” her. How should we classify this transaction? In testimony, neighbors referred to Zhang Shi as a “prostitute,” and she herself referred to her domestic arrangements as “getting a husband to support a husband,” a term that usually referred to full polyandry. The case record does not report a specific time limit on Zhang Shi’s cohabitation with Xiong (the contract does not survive), but if this had been an outright wife sale then witnesses would have called it that (“jia mai” or simply “mai”), and they did not. Instead, they used the term “bao,” implying a temporary but not necessarily short-term arrangement. Ran paid a large lump sum up front, Zhang Shi moved in with him (while Xiong did not), and her duties included bearing children. In substance, this transaction appears to be the same as what people in Zhejiang called “conditional wife sale.”58 Thus, although the phrase “dian qi”—“conditional wife sale”—may have been largely restricted to Zhejiang, the practice was found elsewhere too. Moreover, it was sufficiently well known to have been prohibited by both Ming and Qing codes.59 The advantage of a conditional sale (of either a wife or land) was precisely its temporariness, which provided flexibility for both sides and enabled the solution of several problems at once. Like the pawning of winter clothing over the summer, conditional sale was a way to raise emergency credit on a vital asset while retaining the option of getting it back. At the same time, it provided the services of a surro-

Chapter 3. The Intermediate Range of Practice

113

gate mother to a household that could not afford the price of a permanent wife or concubine. A SP E C T RUM O F S T R AT E G I E S

Returning to the schematic spectrum outlined at the beginning of this chapter, we are struck by the remarkable flexibility of this field of practice, which served to meet a wide range of needs for the people involved. This entire spectrum of strategies involved the mobilization of a woman’s sexual and reproductive labor, with her husband’s cooperation, in order to help support her household. The intermediate range covered in this chapter blurred the boundaries between marriage and sex work in ways that highlight the connections between these ostensibly separate and mutually exclusive categories. At the extreme ends of the spectrum, formal polyandry and outright wife sale, the purpose was to form stable, secure households, and here the priorities of reproduction and the entire package of female domestic labor seem to outweigh sex per se in importance. In this sense, the ends of the spectrum have much in common. Moreover, to the extent that a wife sale created a relationship between the two husbands in which the first might expect continuing aid from the second (rather than simply severing the first husband’s relationship with his wife), we can perceive a logical link between polyandry and wife sale. The paradigm is the peasant household attempting to secure its long-term survival and reproduction. As a result, there was real scope for meaningful affective ties to develop between the parties, and, in this context, women found themselves relatively able to take the initiative and exercise some control over their situations. In the middle of the spectrum, we find the far more mercenary scenario of retail prostitution, where the priority was cash income generated by individual sexual acts with many customers—and a woman’s relationship with each man might last no longer than a single encounter. Even though poverty was the basic motive for all of these strategies, poverty alone does not explain why a particular couple would opt for one strategy over another. Even the category “wife sale,” narrowly conceived, included several different options, including direct sale, compensated divorce, and conditional sale, as well as fraudulent versions of the first two transactions. How were such decisions made? Any claims about female agency in the context of sexual commodification must be carefully qualified; nevertheless, it was often the woman herself who had the decisive say. A striking exception to this generalization is retail prostitution in an urban setting. The couples who engaged in that kind of sex work typically were peasants who had migrated to town out of desperation, and desperation sharply narrowed their options. Moreover, dislocation and separation from kin and community

114

part one

meant that neither spouse was likely to have sympathetic allies close by from whom to seek help. Under the circumstances, such couples—and especially wives—became especially vulnerable to exploitation and other abuse. For this reason, it is not surprising that a woman might try to shift her position along the spectrum out of retail prostitution in the direction of polyandry, by cultivating a safer and more stable relationship with a regular customer.

pa rt t wo

Wife-Selling

This page intentionally left blank

4

Anatomy of a Wife Sale SELLING WIVES (mai qi): The main reason for this sort of custom is that men who are useless wastrels end up having difficulty making a living, and so eventually they sell off their wives. The wives are often perfectly willing to remarry, since they cannot rely on their husbands and can no longer tolerate the misery of living with them. If the man selling his wife can persuade some reprobate to act as matchmaker, and if his wife, her natal family, the prospective buyer, and the seller’s own senior relatives all agree, then the seller must place his handprint on a marriage contract and deliver it with his wife to the man buying her in marriage, and the buyer shall pay him an appropriate brideprice. —report from hunan in “investigation of customs” 1

A F I L IA L S O N SE L L S H I S W I F E

Just before the lunar New Year in early 1755, Yi Shangsong’s father died. For Yi, it was the last straw: to bury his father properly, he would have to sell his wife. Yi Shangsong was a peasant in his late twenties. An only son, he lived in poverty with his wife, Xia Shi (24), their little daughter, and his parents, who had long been ill. The family lived in Xiangtan County, Hunan. Xia Shi had married Yi Shangsong as an adopted daughter-in-law, and, according to her, “the two of us always got along well together.” They had always been poor, but their circumstances had worsened in recent years. As Xia Shi later recalled: My mother-in-law died, and my father-in-law was sick all the time, and we were really poor. Then, in the last lunar month of Qianlong 19, my father-in-law died too. Because my husband had no money to buy a coffin, he decided to sell me in marriage and use the money to give my father-in-law a proper burial. So he asked his cousin Yi Shangyan and my older brother Dahuai to act as matchmakers.

Since Yi was an only son, it made sense for him to seek the assistance of his older paternal cousin, Yi Shangyan (30), who was his closest lineage relative. Shangyan’s agreement to act as matchmaker signaled acceptance of Shangsong’s plan to sell 117

118

part two

his wife. It also made sense to approach Xia Shi’s natal family: it was the honorable thing to do, but also, if they opposed the sale, there would be trouble. For this reason, Shangsong asked his cousin to persuade Xia Shi’s older brother Dahuai to agree. (Since Xia Shi’s father was dead, her brother was head of her natal family.) According to Dahuai: Yi Shangsong asked his cousin Yi Shangyan to tell me that he wanted to sell my sister in marriage. I didn’t want to agree, but he pleaded with me again and again. I knew that Shangsong had always been poor, and that the reason he wanted to sell my sister was to serve his parents. In the end, I felt I had no choice but to go along with his plan, however reluctantly.

Dahuai also agreed to serve as matchmaker, so that both families would be represented. The two matchmakers approached an acquaintance, Zhou Wenxiao (29), who lived nearby and made a living selling sundry goods. Zhou’s wife had recently died, leaving him with an infant son, and he wanted to take another wife. Here is Zhou’s recollection of the matchmakers’ pitch: Xia Dahuai and Yi Shangsong’s cousin Yi Shangyan told me that Shangsong’s father had died of illness, and that he didn’t have any money for a proper burial, so he was willing to sell his wife Xia Shi in marriage for a brideprice (li yin) of only 12 taels of silver. At first I refused, because the woman they were offering me was a wife with a living husband (sheng qi). But Dahuai told me not to worry, because Xia Shi is his own younger sister, and he guaranteed that the couple was willing to separate because Shangsong was so poor that he couldn’t afford to take care of his father’s burial. I believed him, and so I agreed.

Zhou paid a brideprice of 12 taels, and received Xia Shi as wife, along with a sale contract written in her husband’s name. Since the matchmakers were the seller’s kin, and had acted out of sympathy, they refused payment. The whole matter was settled amicably; as a neighbor observed, “there was no deceit, and both sides acted voluntarily.” Two years passed, during which the principals remained on friendly terms. Then, during the winter of 1756–57, Xia Shi’s mother, Chen Shi, fell ill, and shortly after the New Year, Yi Shangsong decided to visit her and see how she was doing. For company, he took along a neighbor boy, Liu Kun (15). Xia Shi’s new husband, Zhou Wenxiao, visited Chen Shi that same day, bringing gifts of meat, fish, and wine. Chen Shi told her son Xia Dahuai to host their guests, and the men sat down together. But (as Dahuai would recall) after they finished the wine, Yi Shangsong said it was too bad there was so little wine, because he hadn’t drunk enough really to feel happy (chi bude changkuai). Zhou Wenxiao saw that Yi Shangsong wanted more wine to drink, so he acted as host and invited Yi Shangsong, Liu

Chapter 4. Anatomy of a Wife Sale

119

Kun, and me to Feng Renhong’s wine shop, where we drank four more flasks of wine and ate two more plates of food.

Zhou Wenxiao paid the bill, which came to six fen of silver (.06 tael) for four flasks of wine, a plate of tofu, and a plate of pork intestines. This scenario is highly suggestive: the prosperous buyer, acting as host, appears to have felt some sense of obligation toward the pathetic, impoverished seller, a sense that was matched (if not exceeded) by the latter’s sense of entitlement. It is no coincidence that Yi Shangsong happened to show up at his former mother-inlaw’s home on the very day that his wife’s new husband planned to visit with food and wine. After they left the wine shop, Dahuai intended to escort the visitors as far as the crossroads, but then a quarrel broke out. Here is his account: Yi Shangsong was pretty drunk, and he suddenly said to Zhou Wenxiao: “In the past I sold my wife to you, but I got only 12 taels in brideprice—that was not nearly enough! Now I have nothing left to live on, and I want you to make me a supplementary payment of a few more taels (hai yao zai zhao ji liang yinzi). Zhou Wenxiao answered: “It’s been two years since you married your wife to me, and now you want me to make you a supplementary payment? Who ever heard of such nonsense?” Yi Shangsong began to curse him, saying that he would fight him to the death (pinming); he also threatened to file charges against Zhou Wenxiao for “scheming to marry a wife with a living husband” (mou qu sheng qi). In his anger, Zhou Wenxiao punched Shangsong, and Shangsong fell down.

Dahuai and Liu had walked on ahead, and, before they could intervene, Zhou had strangled Shangsong. Zhou later confessed: Since I had taken in marriage a wife with a living husband, of course that put me in the wrong (qu le sheng qi yuan you bushi), so I was afraid that if Yi Shangsong really pressed charges against me I’d lose big. On the other hand, I knew that Yi Shangsong was an extremely impoverished and shameless guy (qiong ji wu lai de ren), and I feared that if I did pay him what he asked, then he would just keep trying to extort more money from me, and there’d be no end to the trouble and harm I’d suffer from him. So I decided to kill him, in order to get rid of this problem for the future.

Zhou threatened to implicate the other two in the murder and thereby won their promise not to report his crime. Then he stripped off Yi’s clothing and divided it between them. But word spread about what had happened, and the authorities arrested those involved. The principals quickly confessed. But the coincidence that buyer and seller had happened to visit their mutual mother-in-law on the same afternoon provoked the magistrate’s suspicions. Was this not evidence of a conspiracy to murder Yi Shangsong? Wasn’t it likely that Xia Dahuai was party to that conspiracy, since he “helped

120

part two

get Yi Shangsong drunk,” witnessed the murder, and failed to report it? The magistrate ordered Dahuai tortured with the ankle crusher (jia gun) and questioned him closely on this point.2 But Dahuai insisted that no one had thought it strange for the two men to drop by, even on the same day: “Both of them were my sister’s husbands, and we often socialized with them (dou shi xiaode meifu, shichang laiwang).” As to why he had failed to report the homicide, Dahuai explained that he feared being implicated, but also, “if I reported Zhou Wenxiao’s crime, he would pay with his life, and that would leave my sister with no means of support.” Eventually, the magistrate accepted these explanations. Zhou Wenxiao’s sentence for the crime of “purposeful homicide” (gu sha) was beheading after the assizes.3 Xia Dahuai and Liu Kun were each sentenced to one hundred blows of the heavy bamboo for failing to report the homicide. Xia Shi was also sentenced to one hundred blows of the heavy bamboo, according to the statute against “buying or selling a divorce, and thereby taking in marriage, with consent, the wife of another man” (mai xiu mai xiu he qu ren qi), which mandated that penalty for buyer, seller, and sold wife. She was permitted to redeem her beating with a cash fine, and ordered divorced from both husbands and “returned to her natal lineage” (gui zong). For acting as matchmaker, Yi Shangyan was sentenced to ninety blows. The same statute mandates confiscation of the brideprice and any matchmakers’ fees, but since Yi was dead and there was no way to recover the money, the magistrate waived confiscation. Since this case involved a capital crime, the magistrate had to report his provisional judgment up the chain of command, and eventually the governor of Hunan reported it to the emperor in one of the xingke tiben that survives in the archives in Beijing—which is the only reason we know that, in the winter of 1755, a peasant named Yi Shangsong sold his wife in order to buy a coffin for his father.4 The Wife as Asset of Last Resort The story of Yi Shangsong is typical of the more than 600 wife-selling cases I have collected in Qing legal archives. Each case record contains a unique anecdote, but this one illustrates many common features. The logic of the transaction, the way it was negotiated and documented, the motives and attitudes of the different parties, and the reasons for later conflict all can be seen in many other examples. For now, let us focus on Yi Shangsong’s reason for selling his wife. The immediate factor was the death of his father, which invoked the filial imperative to provide a good burial. But the underlying cause was poverty—and poverty of one degree or another was the motive for the vast majority of sales. According to the American missionary Arthur Smith, writing at the end of the nineteenth century, “the frequency of such sales may be said to bear a direct ratio to the price of grain.”5 In testimony, couples often explained their action simply as “separating

Chapter 4. Anatomy of a Wife Sale

121

in order to survive” (ge tao shenglu), and many wife sale contracts use similar language. Here are the reasons given for wife sales in court cases. I include only sales for which a specific reason is given:6 Reasons for wife sales, central cases (324 transactions in 304 cases):

Poverty: Marital discord (not including wife’s adultery): Husband’s illness: Wife’s adultery: Other:

243 (75%) 63 (19%) 41 (13%) 33 (10%) 9 (3%)

Reasons for wife sales, local cases (313 transactions in 300 cases):

Poverty: Marital discord (not including wife’s adultery): Husband’s illness: Wife’s adultery: Other:

259 (83%) 61 (19%) 46 (15%) 11 (4%) 3 (1%)

Here, the category “marital discord” does not include sales motivated by the wife’s adultery. In sales motivated by adultery, that is the only reason cited, so this category does not overlap with the others (a husband who sold his wife because of adultery may have been poor, but he did not cite poverty as a reason to sell her). Many of the other sales involve more than one motive: for example, a husband’s illness exacerbated a couple’s poverty; poverty worsened their discord; and so on. But the big story is poverty.7 Yi Shangsong faced conflicting imperatives: should he give his father a good burial, or try to keep his family together? Perhaps not every husband would have made the same choice as he, but his choice embodies the core Confucian concept that the primary alliance is that between fathers and sons, and that the ultimate purpose of marriage is filial piety.8 (It also epitomizes everything that the May Fourth radicals would reject in Confucianism, which they saw as sacrificing living people and future hopes for the sake of dead ancestors and obsolete values.) But Yi’s choice was complicated by the fact that he had no son. In some other cases, the choice was not so stark. For example, in a 1745 case from Yiyang County (also in Hunan), peasant Zhou Chengbin sold his wife in order to buy a coffin for his mother, who had just died, but his wife had already borne two sons, thereby fulfilling her principal function on behalf of his lineage.9 But Yi Shangsong had only a daughter. (We do not know her fate, and Yi may have sold her, too.) Therefore, by sacrificing his wife, Yi also sacrificed his chances of a son—and failing to secure patrilineal descent constituted a classic violation of filial piety. For this reason, his decision makes sense only as a last resort.

122

part two

How poor was Yi Shangsong? We lack much specific information, but at the time of his death he cut a pathetic figure. He stood under five feet tall (his corpse measured just 4’10”), and he was dressed in rags. Liu Kun testified that Yi’s pants were so full of holes that his genitals were partly exposed, and when offered those pants (after the murder), Liu declined because he considered them worthless. Also, the profile of Yi’s household before the wife sale reveals a low ratio of laborers (Yi and his wife) to dependents (his invalid parents and little daughter). The timing of the sale is also significant: the last lunar month. Most debts came due at the New Year, and it would be strange if a poor peasant like Yi did not have debts (perhaps including the cost of burying his mother, who had died not long before).10 Most of the wife sales in my case sample took place right around the New Year, confirming that debt was a major trigger of such transactions. It was in this context that Yi Shangsong’s father died: Yi may have been considering selling his wife for some time, but the filial imperative forced his decision. This imperative also served to rationalize the wife sale in the eyes of others, including Xia Dahuai and buyer Zhou Wenxiao. This emphasis on filial piety reminds us of the rhetoric of widow remarriage, which was similarly motivated by poverty but seen as a betrayal of the dead husband. Many a widow justified her remarriage by portraying it as a self-sacrifice that would secure her dead husband’s line of descent; she herself might prefer to starve to death rather than surrender her chastity, but, if she did so, then her husband’s children would also starve, thereby cutting off his posterity. Ostensibly, then, her decision to remarry was a lesser betrayal, motivated by a higher sense of loyalty.11 In Yi’s case, the death of a parent was the crisis that pushed an already poor household over the edge. But any sort of crisis might have that effect. Natural disasters provoked emergency sales of wives and children on a wide scale. In his classic study of famine in the eighteenth century, Pierre-Etienne Will notes that, on the north China plain, the failure of a single harvest could precipitate a subsistence crisis severe enough to cast thousands of peasants on the road as refugees, and sales of persons proliferated in the wake of harvest failures.12 This was still true in the early twentieth century, when British diplomat George Jamieson commented that “in times of famine and distress the sale of wives, who have given no cause for divorce, is common enough, and no question asked.”13 These observations are supported by many examples from the archives. In an 1823 case, peasant Wei Jing, his wife, and their daughter fled their home in Baodi County, Zhili, to escape flood and famine, ending up as refugees in neighboring Sanhe County. In the months that followed, Wei sold his daughter and then his wife, in the hope that all three might survive.14 In an 1813 case from Gansu, a woman named Xu Shi (25) provided the following testimony about how her husband came to sell her:

Chapter 4. Anatomy of a Wife Sale

123

There was famine at home in Zhongwei County, so my husband and I fled to Guyuan Department [150 kilometers to the south]. . . . We were hungry, and we could see that we were going starve to death. Finally, we talked it over and decided that my husband should sell me in marriage—that way, we could survive by going our separate ways.15

More than 10 percent of the couples in my sample of wife sale cases were famine refugees. In most of my cases, however, the precipitating factor in a wife sale was not some macro-level disaster but simply a stroke of bad luck that exhausted whatever slim surplus a household had possessed. After all, as historian Lillian Li observes, for many peasants “the line between ‘famine’ . . . and everyday, year-to-year subsistence and hunger was thin indeed.”16 For example, a house fire or robbery could push a poor family over that line, or household division between brothers might leave a newly independent family branch with too small a portion to support themselves.17 A key factor in up to 15 percent of cases was illness that prevented the husband from working (the percentage would be higher if alcoholism were included). We find men suffering from injuries, blindness, ulcerating sores, tuberculosis, paralysis, syphilis, “yellow-swelling disease” (hepatitis?), “chest pain disease,” “blood-vomiting disease,” “inability to keep down food,” and many other vaguely diagnosed ailments. We also find many references to “feng bing,” which in colloquial usage denoted a wide range of neurological and mental illnesses. In a case from Duchang County, Jiangxi, Bao Nianshi’s decision to sell his wife was precipitated by a strange illness that would come and go, rendering him incoherent and disoriented, and making it impossible for him to keep a job. Bao was a landless peasant who had worked as an agricultural laborer. His wife, Xiang Shi, had borne him five sons; the oldest two had been hired out, but the other three were too little to work. After her husband fell ill in 1776, Xiang Shi had to go out and beg for alms; they also received help from Bao’s nephew. In this manner they managed to get by for three years. But, by the winter of 1779–80, they had reached the end of their tether, and in a period of lucidity, Bao decided that he must sell his wife: “We had nothing to eat and we were going to starve to death. I had a lot of debts and no way to pay them. There was no other choice but to sell my wife.” Xiang Shi did not want to leave her husband, but he finally persuaded her that “instead of staying together and starving to death together, it would be better to separate in order to escape this situation and survive.”18 What all these examples have in common is a poor peasantry with little margin over subsistence. As Lillian Li explains: The economic consequence of a small negative event, such as a poor harvest or loss of one crop, could spell difficult social consequences, such as incurring debts, the sale of property, begging, the sale of wife or children, or infanticide. Such events were

124

part two sometimes so “normal” that it is hard to classify them as outcomes of a particular crisis rather than those of persistent scarcity.19

In 1932, economic historian R. H. Tawney summed up this “normal” condition with a famous analogy: There are districts in which the position of the rural population is that of a man standing permanently up to his neck in water, so that even a ripple is sufficient to drown him. The loss of life caused by major disasters is less significant than the light which they throw on the conditions prevailing even in normal times over considerable regions.20

It is not hard to imagine Yi Shangsong as Tawney’s drowning peasant, with the death of his father just when debts fell due being that fatal ripple in the water. In “normal times” like these, as well as in “major disasters,” a wife could serve as an asset of last resort. P R O F I L E S O F T H E P R O TAG O N I ST S I N W I F E S A L E S

We can use legal cases to develop rough profiles of the typical protagonists in wife sales, namely the seller, the buyer, and the sold wife. Xingke tiben tend to be more useful for this purpose than local cases, because they are written in a standardized format and include much detailed information presented in a systematic way. In contrast, local court records are less systematic and detailed, transcribed testimony tends to be brief and formulaic, and many files are incomplete. First, let us summarize the data from xingke tiben about sellers and buyers (see Appendix B.1). In these cases, the vast majority of sales consisted of peasant men in their early thirties selling their wives to other peasants of roughly the same age. The key difference between the men is that more than two-thirds of buyers had never previously married. Buying another man’s wife was a relatively inexpensive way to marry; nevertheless, the typical buyer had had to work hard and save for years in order to afford the purchase. In contrast, the typical seller had been married for over a decade by the time he sold his wife, and his age at first marriage was about the average for rural men (i.e., late teens for minor marriage, early twenties for major marriage).21 As this contrast suggests, most buyers had started out worseoff than most sellers, but the buyers were on an upwardly mobile trajectory, however modest their means, whereas the sellers had fallen on hard times, and many were truly desperate. A minority of buyers were widowers seeking “replacement wives” or married men seeking concubines (usually, they claimed, because their wives had not borne sons); such men were much better off, and they averaged five to ten years older than the typical buyer. For the typical buyer, another man’s decision to sell a wife provided the first opportunity finally to marry, establish a household, and have sons.

Chapter 4. Anatomy of a Wife Sale

125

Turning to the women in xingke tiben, the typical wife being sold was in her late twenties, so about five years younger than her seller and buyer. The seller was her first husband; they had been married since her mid-teens (i.e., for about eleven years); and she had had borne at least one child (an important selling point, since most buyers were hoping to have sons). Most of these women were in major marriages, but 13 percent were adopted daughters-in-law (there is only a single uxorilocal marriage in the sample). Eighty-three percent were sold to become the wives of their buyers, but 16 percent were sold to become concubines, and the latter women were slightly younger on average than those sold to be wives. The central case records considered so far come from all over China Proper. What about regional variation? The local case sample is less useful for detailed information on ages and children, but often we can determine the occupation or social background and the marital history of protagonists, which allows comparison with the central case sample. In this respect, Nanbu and Baodi counties (which, like most counties, were largely rural) match the central case sample closely (see Appendix B.2); nearly all of the sellers and buyers in these two counties were peasants or rural laborers. Ba County, however, presents a strikingly different social profile, reflecting the influence of the major port city and governmental center of Chongqing (see Appendix B.3). As elsewhere, the great majority of men who sold wives in Ba County were peasants or rural laborers (for a combined total of 70 percent). The buyers, however, present a very different profile. Peasants constitute the single largest group of Ba County buyers (45 percent), but, even if one adds rural laborers (4 percent), they make up less than half the total. Quite a few buyers were urbanites, in part because men from the rural hinterland who wanted to sell their wives sometimes came to Chongqing to look for buyers (in order to avoid shaming themselves in their local communities and also in hope of a better price); some rural couples came to the city in search of work or as refugees from famine, and they ended up resorting to wife sale when they ran out of options. The roster of buyers is notable for including many people who were relatively well off: shopkeepers, innkeepers, merchants, yamen clerks, minor degree holders (mostly purchasers of the jiansheng degree), and students for civil examinations together account for more than a quarter of the buyers in these cases. Some of the pimps also appear to have been quite well off. Few, if any, of these men could be considered members of “the elite”; rather, they constituted something like the petite bourgeoisie of Chongqing. Ba County buyers were also distinctive for their marital status and the purpose for which they were buying other men’s wives. Only 52 percent of the buyers in Ba County had never previously married, compared to more than two-thirds in the xingke tiben.22 The rest were either married men looking for a concubine (39 percent) or widowers looking for a second wife (9 percent). The men who bought concubines included many of the relatively privileged people from the above list—

126

part two

less than a third were peasants. Turning to the women, as elsewhere nearly all had been married only once (i.e., the seller was the first husband). What stands out is the high percentage of women who were bought to be concubines (36 percent) or prostitutes (8 percent), far higher than in either the central case sample or the smaller local case samples from Baodi and Nanbu counties. Only 16 percent of the women in central cases were sold to be concubines, and just two were sold into prostitution. Only 20 percent of the women in the Baodi cases and 16 percent in the Nanbu cases were sold to be concubines, and, as far as I can tell, none in either county were sold into prostitution. The high demand for concubines and prostitutes in Ba County reflects the highly commercialized market for women and sex in Chongqing. A striking feature of these data is that concubines (and men who bought them) are heavily overrepresented in all my samples of wife-selling cases. As noted in Chapter 1, the best evidence suggests that, during the Qing dynasty, an average of less than 4 percent of married women were concubines for China overall. The fact that 20 percent of wife sale cases from a poor county like Baodi involved the purchase of women to become concubines does not mean that 20 percent of married women in that county were concubines. Rather, these data show that wife sale was one of the main ways that men acquired concubines: the high end of the market for sold wives overlapped with the low end of the market for concubines. Such sales represented an especially cheap way to get a concubine, and they evidently appealed to upwardly mobile men who aspired to the trappings of higher status but were looking for a bargain. The overrepresentation of concubine purchase in these data probably also reflects the overrepresentation of urban people in the routine caseloads of local courts (i.e., cases originating in the county seat, rather than its rural hinterland).

N E G O T IAT I N G A W I F E S A L E

Bearing in mind the kinds of people who typically participated in wife sales, let us take a closer look at the transaction itself. How was a wife sale negotiated? Nearly all sales required at least one matchmaker. (On occasion, a wife sale was negotiated without a matchmaker by a seller and buyer who were previously acquainted, but this was rare.) The matchmaker was responsible for finding a buyer and negotiating the terms of sale; once the parties had agreed, the matchmaker would supervise a meeting at which a contract would be drawn up and exchanged for the brideprice (caili or caili qian, also referred to as “body price,” shenjia) before witnesses. After this meeting, the woman would be delivered to her buyer’s household, usually escorted by the matchmaker, in order to consummate the new marriage.

Chapter 4. Anatomy of a Wife Sale

127

Matchmakers Who acted as the matchmakers to negotiate wife sales, and how were they compensated? Sometimes there was only one matchmaker for a given transaction, but it was not unusual to have two or three, depending on what was involved in identifying a buyer. Most of the individuals acting in this capacity were not professional matchmakers but simply people known to the parties who agreed to perform this service. Most of the professionals who turn up in these cases were mature women, whereas nearly all of the non-professionals were men. Most often, a man seeking to sell his wife, or a man looking to acquire a wife, would approach some relative, neighbor, or acquaintance—usually a man who had some stature in the community (a senior relative, one’s employer or landlord, or perhaps an older man in the village)—and ask for help. Some individuals were known for performing such services, but also there seems to have been widespread knowledge about how to go about doing this. In about half of the cases for which I have specific information (and here I count only open sales that were completed, not fraudulent sales), matchmakers acted for free, out of compassion. Most who did so were kin or close friends of the seller—or occasionally of the woman or her buyer. We saw an example of this scenario in the case of Yi Shangsong, whose cousin and brother-in-law both agreed to act as matchmakers out of charity. Similarly, in a 1755 case from Jiangning County, Jiangsu, the matchmaker was the wife’s older brother; when approached by her husband, he agreed to act as matchmaker in order to rescue his sister from poverty and from an abusive marriage.23 In a 1766 case from Wenshui County, Shanxi, the three matchmakers were close friends of the seller, and they agreed to act for free (the buyer was also a mutual friend). Later, they testified as follows: Xue Yangwei told us that he was too poor to get by and could not support his wife, and asked us to act as matchmakers to sell Ji Shi in marriage. We knew that Zhang Dakang’s wife had died and he wanted to remarry, so we negotiated an agreement between them: Zhang Dakang paid Xue Yangwei 12 taels of silver, Xue Yangwei drew up a marriage contract, and Zhang Dakang took Ji Shi in marriage. Since we are good friends with both of them, we did not accept any of their money.24

Occasionally we find a matchmaker more tenuously connected to the principals who would be willing to act out of charity. In a 1754 case from Jinyun County, Zhejiang, the couple were famine refugees from Dongyang County (about a hundred kilometers to the north). They managed to find a man also from Dongyang County who agreed to broker the wife sale out of fellow feeling as “tong xiang”— that is, people from the same home area.25 At most, such matchmakers might receive a drink of wine (the drawing up of the contract and the exchange of brideprice for the wife were often ritually finalized with wine). For example, in a 1745 case from Yiyang County, Hunan, three

128

part two

maternal relatives of the seller who acted as matchmakers testified that “we did not receive any matchmaker fee—we drank wine on two occasions, but got nothing more.”26 It appears to have been the norm in much of rural China for middlemen in land contracts, loan contracts, and even marriage contracts to receive compensation only in the form of a feast, wine, or perhaps a modest gift—something other than money (which would have undermined the facilitator’s reputation for impartiality). By playing these roles, local men (and occasionally women) of respect would gain symbolic rather than material capital, accumulating a fund of prestige and good will that facilitated community relations and could be called upon in other circumstances. There was usually a small number of people in each village who regularly performed such services: individuals known for probity and trustworthiness, who often were leading members of their descent groups. Occasionally a local rural agent (xiangbao) would play this role.27 Unlike most such transactions, however, wife sales carried stigma and risk, and for this reason usually only close friends or relatives of the principals would agree to act for free—and they did so out of pity and charity, rather than to accumulate any sort of prestige. If others acted in this capacity, they typically had to be paid. Thus, in the second most common scenario found in legal cases, members of the community would act for a “matchmaker’s fee” (mei qian): either a share of the brideprice received by the seller, or a separate sum paid by the buyer. Matchmakers who worked for a fee usually had no close personal connection to the parties directly concerned but were fellow villagers who had acted in this capacity before and were considered trustworthy. Some were men with itinerant occupations that gave them wide contacts beyond the local community. Their fees ranged from a nominal gratuity (a couple hundred cash in “thank you money,” xie yin) to substantial percentages of the buyer’s total expenses. Such fees occasionally reached as high as 30 percent of total expenses, but most seem to have fallen in the range of 5–15 percent. Nevertheless, for a community member to demand a fee for brokering a wife sale could be controversial, especially if he were seen as exploiting his neighbor’s misfortune. A striking example is found in an 1807 case from Nanyang County, Henan. In this case, peasant Wang Zaofa was incapacitated by a paralyzing illness; he and his wife, Yan Shi, lived with his elderly father, and they were desperately poor. Finally, Wang’s father decided that the family “had no choice but to sell Yan Shi in marriage to get some money to live on,” and that this plan would be best for all concerned. His son and Yan Shi both agreed. Their neighbor Lu Huaiyou (52) acted as matchmaker and negotiated to sell Yan Shi to a stranger who was passing through for a “body price” of 8000 cash. Out of this sum, the Wangs paid Lu a fee of 1000 cash for his services. The next day, another neighbor named Li Mingjun (42) found out what had happened. “When I heard that Lu Huaiyou had helped break up their marriage

Chapter 4. Anatomy of a Wife Sale

129

because he was greedy for the matchmaker fee, I became very upset.” Li confronted Lu and cursed him for his selfishness; Lu cursed him back, and Li knocked him over. In falling, Lu hit his head, and he later died of the injury. What upset Li Mingjun was less the wife sale itself than the fact that Lu Huaiyou had profited from it. He saw this as unconscionable: everyone knew that Wang Zaofa was ill and his family destitute, and everyone knew how painful it would be to have to resort to wife sale. To Li’s mind, Lu Huaiyou had violated the basic norms of community by exploiting their neighbors’ desperation.28 Securing the Approval of Both Families A wise seller and a conscientious matchmaker would make sure to obtain the consent of everyone involved—including the wife herself, but also both the seller’s and the wife’s families—before proceeding. (Of course, not every seller was wise or every matchmaker conscientious, otherwise we would have far fewer court records of wife sales available for study.) Better than passive consent was securing the active participation of both families, to act as matchmakers, to “zhuhun” (i.e., formally authorize and preside over the wife sale), or to witness the contract. Active complicity on the part of senior members of both families would make it far less likely that anyone else from those families would dare to raise objections later on. Moreover, it was common for wife sale contracts to state explicitly that both families had been consulted and had agreed in advance of the transaction. We have already seen one example of a conscientious seller at the beginning of this chapter in the case of Yi Shangsong, whose first step was to persuade his cousin and brother-in-law to act as matchmakers on his behalf. Here we have the testimony of peasant Wei Zhongde (34), of Ningxiang County, Hunan, about how he went about selling his wife, Yin Shi, in 1776: My father is blind in both eyes, my mother is old and ill, and I have no brothers; my household is very poor and I can’t support them properly. So in the 2nd month of Qianlong 40 I talked it over with Yin Shi and told her I wanted to remarry her off to someone else, and she agreed. I also explained the situation to her older brother Yin Duren, and I asked my lineage members Wei Xixue and Wei Jisheng to act as matchmakers.

It turned out that a single man from a nearby village, Su Benliang, wanted to take a wife, and he had already asked Wei Xixue and Wei Jisheng to make inquiries on his behalf. (From this detail, it seems likely that these two men already had a reputation for performing such services.) The match was quickly made: They negotiated an agreement to marry Yin Shi to Su Benliang as wife for a brideprice of 30 taels of silver. On the 8th day of the 3rd month I wrote a marriage contract and asked my uncle Wei Jingshi to list his name on it in the role of “zhuhun,” and my wife’s brother Yin Duren also wrote his mark on it. Su Benliang gave me the brideprice, and then he took Yin Shi home in marriage.

130

part two

Since Wei’s own father was incapacitated, his uncle Wei Jingshi (60) was the closest senior member of his lineage in a position to act, so he was an ideal person to legitimize the transaction by agreeing to zhuhun. He and the matchmakers all acted for free, since the seller was their poor kinsman and the buyer was a friend. In this case, the seller took the initiative to make sure that everyone concerned would consent to the sale, but he also secured their active complicity by having senior members of both his own family and his wife’s witness the contract. He certainly could not be accused of any deceit. Witnesses’ testimony suggests that Wei Zhongde was a decent man caught in an unfortunate situation that was by no means his own fault.29 An 1819 case from Jin County, Shengjing, involved a similarly conscientious seller. Li Xiu (42), a landless peasant who made a living through short-term agricultural labor, was married to Guo Shi (35); they had two sons (aged 11 and 2). This family had always been very poor, but in the autumn of 1818 conditions worsened when Li came down with “thigh pain illness” and could no longer work. With the onset of winter, the family was reduced to begging, and Li confided in his uncle Li Hongmei (68), who had raised him, that if his family was to survive he would have to sell his wife. (Note his willingness to sell his wife but not his sons; there were no daughters in the picture.) Li Hongmei gave his consent, and since he had acted to zhuhun Li Xiu’s marriage in the first place, he agreed to perform this role again. Then Li Hongmei asked his friend Hou Jun to act as matchmaker, and Hou agreed to perform this service for free. He happened to know a man in a nearby village named Yan Xin (36) who was a childless widower and wanted to remarry. Hou and Li Hongmei visited Yan’s home and negotiated an agreement whereby Yan would buy Guo Shi as wife, and she would care for her infant until he was old enough to return to his father. Having reached this point in the transaction, Li Xiu visited his father-in-law, Guo Tianwei (58), to explain the exact terms of the proposed sale of his daughter, and asked permission to proceed. Guo agreed—as he later testified, “since I couldn’t help him myself, I didn’t try to dissuade him.” This discussion took place first thing in the morning on the chosen day; later that same day, the matchmaker brought the prospective buyer Yan to Li Xiu’s home to meet Guo Shi. For Yan, the purpose of this visit was twofold: to make sure that Guo Shi did not object to marrying him, and to assess her age and health. The visit was a success, and since Yan had brought his money, they immediately drew up a contract and made the exchange. Here we see both seller and matchmaker proceeding in a transparent and honorable way. They secured the consent of senior members of both the seller’s and the wife’s families, made sure that everyone clearly understood the terms of the transaction, and invited the prospective buyer to verify in person that he was not being cheated.30

Chapter 4. Anatomy of a Wife Sale

131

In a 1762 case from Luling County, Jiangxi, Wu Rongliang (25) decided to sell his wife of six years, Chang Shi (22), because “we were poor and my wife often quarreled with me.” She readily agreed (“we often went hungry, and I couldn’t stand it any longer”), so he notified her family, who consented, and, in exchange for his promise of 10 percent of the brideprice, Chang Shi’s grandmother agreed to “zhuhun.” Evidently, it was very important to Wu Rongliang to secure the participation of his in-laws. Then Wu engaged two professional matchmakers: a man, Zhou Heyou (40), and a woman, Chen Fengzhu (50), who worked as a team. The matchmakers’ first act was to confirm in person that the woman and her natal family had really given their consent. As Zhou later testified, I went to ask Chang Shi and her grandmother Wu Shi about it, and they both confirmed that they were willing. Only then did we open negotiations with Xiao Qinbei, who offered a brideprice of 52 taels. On the 20th, Xiao Qinbei paid 29 taels in advance, and Wu Rongliang drew up a marriage contract. At that time his wife’s younger brother, Chang Kongjian, was present and he wrote his mark on the contract, and Wu Rongliang gave him 5.2 taels of silver. It was agreed that the rest of the brideprice would be paid on the 22nd at the Wu lineage’s ancestral temple, and at that time a day would be chosen for [Chang Shi] to go to [Xiao Qinbei’s] home and consummate the marriage (guomen chenghun).

Chang Shi’s uncle also witnessed the contract on behalf of the natal family. Negotiations later fell through, however, because her brother Chang Kongjian suddenly demanded a larger share of the brideprice. The matchmakers had been promised an unknown fee to be paid after the transaction, but they were never paid.31 We find a final example of a conscientious matchmaker in a 1749 case from Huai’an County, Zhili. Zhang Xingzu and his mother, Zhang Cui Shi, decided to sell his wife, Liu Shi, because they were poor and the marriage was unhappy. (Zhang was serving as a soldier in Zhangjiakou and was seldom at home.) So Zhang Cui Shi asked one Hou Xian, who “has long made matches for people,” to seek a buyer on their behalf. As he later testified, Hou at first hesitated to arrange the marriage of a “wife of a living husband” (huoren qi), but since the seller’s mother and maternal uncle “were both willing to zhuhun, I decided it must be all right.” Hou Xian already knew of a potential buyer: he had recently been approached by a peasant named Di Huaixi, who was trying to find a wife for a friend of his, a sinicized Mongol known as Zhao Qiling. Zhang Cui Shi gave the matchmaker a “divorce contract” (xiu shu) that her son had written, and Hou took this to show Di and Zhao. Zhao agreed to pay a brideprice of 8 taels of silver. Before finalizing the sale, however, the matchmaker made a point of visiting Liu Shi’s cousin Gao Linding to seek his approval as representative of her natal family (her own parents were dead and she had no brothers). But since Gao was Liu Shi’s cousin on her mother’s side, he did not dare to authorize her sale; instead, Gao

132

part two

took Hou Xian to see Liu Shi’s paternal uncle Liu Zhaoji, who was her closest senior relative on her father’s side. Liu Zhaoji gave his approval, saying that “under the circumstances, it would be best for her to remarry in order to avoid suffering.” With the blessing of Liu Shi’s uncle, Hou proceeded to finalize the sale. The full brideprice went to the seller and his mother; the matchmaker acted for free, and the natal relatives asked for nothing.32 Exploitation by Unscrupulous Matchmakers Not all matchmakers were sympathetic friends or honorable members of the community. Some were professionals, and they stand out in the case records for profitseeking and downright exploitative behavior. In an 1820 case from Ba County, an impoverished man named Wu Gui decided to sell his wife, and he asked his landlord, Zeng Gui, for help; Zeng in turn introduced Wu to a woman in Chongqing named Chen Li Shi who made a living as a matchmaker. Chen Li Shi agreed to help in exchange for Wu Gui’s promise in writing to demand no more than 6000 cash, to grant her complete discretion over the negotiations, and to shield her from any trouble that might subsequently arise. The practical import of these terms was that the matchmaker could keep anything over 6000 cash as her commission. Indeed, as Chen Li Shi later testified, “four days later, I arranged for Chen Qifa to take Wu Gui’s wife as his wife, for a brideprice of 10,000 cash. Wu Gui received 6000 cash, and the remaining 4000 cash served as the matchmaker fee.”33 This case is an example of how a man trying to sell his wife might become vulnerable to exploitation, especially if abstracted from his own community (cases involving exploitation by unscrupulous matchmakers often involve peasants who have come to town, as seen here). A matchmaking commission of 40 percent was outrageously high, compared with the great majority of cases in my sample. But this case is by no means unique. In an 1827 case from Ba County, one Yang Huaxian came down with a “crippling illness” and fell into poverty; he decided to sell his wife, Da Shi, so that they could “separate in order to survive.” He engaged a pair of professional matchmakers in Chongqing, Feng Liang Shi and her son, Feng Er, who negotiated to sell Da Shi to a man from Hejiang County as wife, for a brideprice of 18,000 cash. Out of the brideprice, Yang Huaxian received only 8000 cash, with the remaining 10,000 being kept by the matchmakers as their commission. As in the previous example, the seller had accepted the matchmakers’ advance guarantee of 8000 cash, not realizing that they planned to negotiate a far higher price and pocket the difference.34 An 1887 case from Nanbu County illustrates a different sort of exploitation by a matchmaker, who in this case was a stereotypically thuggish yamen runner acting in collusion with an innkeeper. A young peasant named Xiang Dengping (18) was detained by county runners in connection with the arrest of his father for theft. He

Chapter 4. Anatomy of a Wife Sale

133

and his wife, Zhang Shi, were taken to the county seat and lodged at an inn near the yamen, where they were required to stay (at their own expense) until summoned by the magistrate. The innkeeper was an intimate of the runners, and he regularly provided lodgings for people summoned to court or released under guarantee, no doubt in exchange for paying kickbacks to the yamen staff. Over the next few months, the couple ran up an inn bill of seventeen strings of cash, and when they were finally told that the court no longer needed them (without ever having been summoned to testify), the innkeeper refused to let them leave until they had paid their bill. By this time, Xiang’s father had been transferred to the prefectural seat for retrial, and the couple had no money left. At this point Zhao Quan, the runner in charge of young Xiang’s case, intervened by offering to act as matchmaker and sell Zhang Shi in order to clear the debt. Under great pressure, Xiang reluctantly agreed to accept twenty strings of cash for his wife—that sum would pay his bill and leave a little extra for travel expenses so he could return home. Zhao then arranged to sell Xiang’s wife to one Song Ronggao for a brideprice of thirty-six strings of cash, which Song paid to Zhao. Zhao then paid off the couple’s bill at the inn, gave Xiang three strings of cash, and kept the remaining sixteen strings for himself as his matchmaker’s commission. Xiang later filed charges accusing the yamen runners and the innkeeper of forcing him to sell his wife and then stealing her brideprice. In testimony, Zhao Quan claimed he had simply been trying to help Xiang, but the magistrate was having none of it. He saw the seller and his wife as patsies who had been fleeced by predatory runners in collusion with the innkeeper. He ordered Zhao Quan beaten and dismissed, and also made him pay Xiang the sixteen strings of cash he had appropriated. The magistrate did order Xiang slapped for “selling a divorce” but also returned Zhang Shi to him. Buyer Song Ronggao was not punished, aside from losing the entire brideprice along with the woman.35 “Taking a Look” Many of the steps in arranging a sale were designed to minimize risk. In the same vein, it was common for a buyer to insist on seeing the woman himself, and this act of looking played a crucial role in closing a deal.36 (The practice was common in widow remarriage as well.)37 It is not always clear exactly what happened in such encounters, but testimony emphasizes the visual aspect: “to look” (kan), “to take a clear look” (kan ming), “to take a look for purpose of marriage” (kan qin), and so on. Thus, in the 1819 case from Jin County, Shengjing, cited above, the prospective buyer visited the seller’s home with the matchmaker and “saw Guo Shi’s face.” Pleased with what he saw, he agreed to her husband’s terms.38 But in an 1804 case from Ding Independent Department, Zhili, a deal fell through because the prospective buyer did not like what he saw.39

134

part two

What were buyers trying to see, when they “looked” at these women? All the evidence suggests that a woman’s age, as an index of her likely fertility, was by far the most important factor in determining her desirability for prospective buyers. In contrast, whether a woman was beautiful seems to have been irrelevant in all but a very few transactions. The few references to beauty appear in just two contexts: the market for concubines and the market for prostitutes, both of which are found at the high end of the price scale in wife sales. There was a clear sense that an unusually attractive woman might fetch a high price because of this specialized demand. In a 1743 case from Wuwei Department, Anhui, a man persuaded several people to help arrange the sale of his wife by telling them that, because she was very attractive, they “should be able to sell her as a concubine for 60 to 70 taels” and promising them a share. In the end, however, they found no one willing to pay that much, and the seller ended up accepting a mere 25 taels from a single peasant who wanted a wife.40 Occasionally, men bought women ostensibly to serve as wives but in fact with the intent of either pimping them or reselling them into prostitution, and these women are invariably described as young and attractive. An example is a sale that took place in Gu’an County, Zhili, in 1860. This woman was sold ostensibly to be the buyer’s wife, but he had other plans for her, in which her looks did play a role: “I saw that Jia Ernai was young and beautiful (shao ai), and realized I could make a lot of money by making her sell illicit sex.” After paying her husband a “body price” of 300 strings of “capital cash” (jing qian)—the equivalent of 150 strings of standard cash41—he took her to the more lucrative market of Beijing to pimp her, and he did succeed in making a lot of money. This case is an example of the highend scenario in which a pimp paid a high price for a young, attractive woman to serve as a prostitute.42 Similar examples can be found elsewhere in China. In an 1873 case from Ba County, a yamen runner decided to sell his wife because of poverty and asked two female matchmakers to help. They saw that his wife was “young and beautiful” and agreed, planning to take her downriver to Hubei and sell her into prostitution. The wife was willing to be sold, but when she learned what the matchmakers had in mind, she ran away and filed charges.43 I have no evidence of beauty (or, for that matter, ugliness) playing any role in the broad mainstream of wife sales, in which buyers were men of modest means looking to acquire a relatively inexpensive wife, as opposed to a concubine or prostitute. Even in the examples cited above, testimony paired a woman’s looks with her age in a formulaic way (“shao ai” means “young and attractive”), suggesting that beauty (or its lack) was defined in terms of age and significant primarily as a reflection of age. Instead, the most common reasons for buying a wife were the ones cited by a Sichuan peasant in 1736: “to manage the household and try for a son” (zhijia qiu zi).44 Therefore, a woman’s fertility and health were paramount considerations—and her age, as a likely correlate of fertility, played a major role in whether a man would buy her and how much he would pay. The late thirties appear

Chapter 4. Anatomy of a Wife Sale

135

to have served as a rough marker of the end of a woman’s reproductive years: 40 sui (i.e., 38 or 39 years old) is often mentioned as an upper limit. Arthur Wolf calculates that the mean age of last birth among Chinese women in pre-modern times was 38 years old (i.e., 39 or 40 sui), and many peasant women stopped menstruating even earlier, in their early or mid-thirties.45 Thus, in a 1754 case from Youxi County, Fujian, the buyer testified approvingly about a woman’s relative youth. The woman in question was just 35 sui (33 or 34 years old), but his specific word choice is significant: “I saw that Wang Shi wasn’t yet 40 sui, so I agreed to buy her.”46 In contrast, in a 1757 case from Pingyao County, Shanxi, the buyer was willing to pay only the “low price” of 20 taels for a woman of 37 sui, because “I saw that she was already pretty old.”47 As these comments suggest, a woman’s actual number of years mattered less than a buyer’s subjective appraisal of how young, healthy, and fertile she appeared. Older women might be hard to sell. In an 1808 case from Heyang County, Shaanxi, the would-be seller had difficulty finding a buyer for his 41-sui wife: “Everyone thought Zhou Shi was too old, and nobody would accept her in marriage.”48 In an 1805 case from Yuexi Subprefecture, Sichuan, a destitute husband hoping to sell his 48-sui wife in order to secure her survival, managed to do so only by accepting a purely token brideprice of 1200 cash.49 When older women did sell, special considerations played a role. In a 1780 case from Duchang County, Shanxi, a wife of 42 sui fetched the respectable price of 28 taels because of her excellent record of fertility: she had borne her first husband five surviving sons, and her buyer hoped she could still bear him a son as well (evidently she looked young and vigorous enough to persuade him that this was possible).50 In contrast, in an 1819 case from Huoshan County, Anhui, a woman of 47 sui managed to sell (albeit for what was termed a very low price) only because the buyer needed help caring for his elderly parents; for this man, procreation was secondary.51 Failure to “take a look” could lead to unpleasant surprises. In a 1747 case from Jian’an County, Fujian, peasant Fan Aidu offered to sell his wife, Miao Shi, for just 6 taels, and the matchmaker had no trouble finding a buyer named Wei Guangchun (a peasant who, at 37 sui, had never married). But Wei neglected to visit Miao Shi, and when she was delivered to his home he discovered that she was desperately ill with tuberculosis. He tried to force her husband to take her back, provoking a homicidal brawl.52 T H E W I F E S A L E C O N T R AC T

Once the terms had been agreed, the matchmaker(s) would host a meeting to draw up the contract and exchange it for the brideprice. The woman herself was not usually present at this meeting, but once the exchange had been made, the matchmaker would schedule her delivery to the buyer’s home.53

136

part two

Anthropologist Myron Cohen has pointed out that contracts from Qingdynasty Taiwan documenting “marriages deviating from the ideal” resemble contracts for the sale of land or other property far more than they do regular marriage contracts. The reason, he argues, is that “ritualization” is absent in such alliances and so “the particular interests of the parties to a marriage cannot be satisfied on the basis of the assumptions and expectations expressed and confirmed through the prestiged (sic) marriage ritual.” Therefore, the terms had to be spelled out concretely and in crass detail. Cohen cites the example of uxorilocal marriage, which was legal but stigmatized.54 Cohen’s observation is certainly correct. But the parallel between contracts for land sales and wife sales is so exact, as if one served as template for the other, that it suggests a specific connection between land and wives in the male peasant imagination. The Land Sale Contract as Template for Wife Sale Contracts The basic structure and language of contracts for land sales were uniform throughout prerevolutionary China.55 The phrases quoted here come from eighteenth- and nineteenth-century contracts from Shanxi that are in the author’s private collection, but they are typical of such documents found all over the country. A land sale contract is always written in the voice of the seller and bears his or her mark (a cross or “x” for men, sometimes a circle for women). It begins with the phrases “Zhang San (i.e., the name of the seller) hereby establishes this sale contract” (li mai qi ren Zhang San) and “the situation is as follows” (qing yin, or simply yin). Next comes an explanation that economic hardship has forced the seller to resort to the regrettable measure of selling ancestral land. Sometimes this explanation is perfunctory, such as “I have difficulty” or “I am broke,” but often it is more specific, such as “the harvest failed, so we have nothing to live on.” The apparent premise was that for a peasant to sell land was an undesirable and perhaps shameful measure to be taken only reluctantly. (Similar explanations appear in contracts for widow remarriage, uxorilocal marriage, the sale of children, and even household division, because all of these transactions violated normative ideals to some degree.) Such language reinforced the feeling that a peasant compelled to sell his land was the loser in an inherently unequal transaction. Next comes a statement that the seller has agreed through mediators and before witnesses to sell a given piece of land (its dimensions and location are recorded) to the named buyer, for what all have agreed is a fair market price (the amount is stated precisely). A contract for absolute sale (jue mai or si mai) then states that the sale price has been paid in full: for example, “on this day, silver and land have been openly exchanged, the full amount has been paid and nothing more is lacking,” or “the money has been paid in full and I have received it in person, nothing more is owed.” The aim of this language is to preempt future demands for more money,

Chapter 4. Anatomy of a Wife Sale

137

and most wife sale contracts contain nearly identical language. There often follows an explicit statement that the transaction is voluntary, without deceit or coercion, such as “in this transaction both parties are acting of their own free will, and there has been no pressure or coercion or anything else of that sort.” Next comes the seller’s guarantee against future trouble: for example, “if anyone should challenge or obstruct this sale then the seller will take full responsibility by himself.” This last phrase—“the seller will take full responsibility by himself ” (mai zhu yi mian chengdang)—is especially common. Land sale contracts often guarantee against interference by members of the seller’s lineage (who would have had a right to first refusal when the land was put on sale), and in wife sale contracts, similar language guarantees that the seller (and sometimes the matchmaker) will protect the buyer from interference by either the seller’s family or the woman’s natal family. Finally, every land sale contract closes with some version of the following boilerplate: “fearing that spoken words alone are not reliable, I hereby establish this sale contract to serve as proof ” (kong kou wu ping, li ci mai qi wei zheng). The same language is found in almost all wife sale contracts. There follows the date, the name and mark of the seller, and the names of mediators and witnesses. Examples of Wife Sale Contracts Details vary, but land sale contracts all follow this format, as do wife sale contracts. Some wife sale contracts are short and straightforward; others are more elaborate, occasionally exhibiting some literary pretension. The terms used to refer to such documents vary: some are euphemisms, such as “marriage contract” (hun shu), “divorce contract” (xiu shu), and “remarriage contract” (gaijia qi), whereas others are explicit, such as “wife sale contract” (mai qi qi). But all can be identified as contracts for wife sale because they name the buyer and specify the sum he paid for the woman. The following two examples, both transcribed in xingke tiben, are typical of basic, minimalist contracts for wife sales. The first, dated 1750, is from Jiyang County, Shandong, on the north China plain: I, Wang Zhun, hereby establish this sale contract (mai qi). The situation is as follows: Because I can no longer get by, I have asked matchmakers Luo Yong, Wang Shibao, and Jiang De’an to negotiate an agreement to sell my wife Jiang Shi to [widower] Wang Gongji as his replacement wife (ji qi), for an agreed price of 10,000 cash. This amount has been paid face-to-face, and I have willingly accepted it, without any coercion whatsoever, and therefore I establish this sale contract for [Wang Gongji] to keep as proof. Done, on Qianlong 15.3.16, by Wang Zhun, who establishes this sale contract [mark] Matchmakers: Luo Yong, Wang Shibao, and Jiang De’an Witnesses: Village Head (difang) Xin Kexun and Rural Agent (xiangyue) Yang Xiren

138

part two Verifier of contract (shi yue): Rural Agent Wang Qizheng Scribe: Wang De’an

Both buyer and seller in this case were peasants. An interesting feature of this contract is that three village notables (including two rural agents) are listed as witnesses and “verifier” of the contract; moreover, the matchmakers included members of both the seller’s and the wife’s families (the Wangs and the Jiangs).56 The second example, dated 1767, is from Nanhui County, Jiangsu, near Shanghai: I, Gu Shi, hereby establish this contract to sell my wife (mai qi qi). The situation is as follows: My household is poor and we cannot survive, so now I am willing to sell my wife, Su Shi, to Chi Chao to be his wife. I have received 4000 70% cash,57 and now Su Shi will follow him into his household and join him in marriage. Both of us are acting of our free will, and we will never go back on our word (yong wu yi yan). Fearing there will be no proof, I hereby establish this contract as proof. This contract to sell my wife is established on Qianlong 31.12.26 by Gu Shi [mark] Witnessed by Su Chuanguan [mark], Zhang Daguan [mark]

This contract was actually written by the seller, who had somehow managed to acquire some literacy, even though at the time of sale he was a homeless laborer; the buyer was a peasant who at the age of 52 sui had never married. The witnesses were the woman’s father and the seller’s employer, who also acted as matchmaker. Like so many wife sales, this one took place just before the New Year, when debts fell due.58 Both of these contracts are short and relatively free of boilerplate, but both follow the basic format of a land sale contract and contain the common features found in wife sale contracts throughout China during the Qing and Republican eras: the contract is written in the voice of the seller; he begins with a tale of hardship, in order to justify the unfortunate necessity of wife sale; he gives the name of the buyer and the amount of brideprice agreed by the parties; he confirms that this sum has been paid in full, and that he has received it in person; and, finally, he affirms that he is acting voluntarily and provides this contract to the buyer as proof. In the Nanhui County example, the seller further promises never to go back on his word, and such promises are another common feature of wife sale contracts. Each concludes with the seller’s name and mark, along with names and marks of matchmaker(s), witnesses, and (if applicable) the scribe. Nearly all the wife sale contracts I have seen bear a cross or other personal mark after the seller’s name, as well as his handprint and sometimes his footprint. The following is a longer, more detailed specimen from Ba County, dated 1808: I, Deng Zongjun, hereby establish this contract for voluntarily marrying off my wife (ziyuan jia qi). In years past I took Zhou Shi as my wife, and we were matched in

Chapter 4. Anatomy of a Wife Sale

139

childhood.59 But now the two of us no longer wish to stay together, because the harvest has failed and it is difficult to survive. Having no other choice, her family and mine have discussed the situation and decided that the two of us should separate in order to survive. We have invited Luo Gui to act as matchmaker, and he has negotiated for me to marry off Zhou Shi, who herself is perfectly willing (ziji xin’gan qingyuan), to Wu Rongzhong as wife. Through the matchmaker, Mr. Wu has agreed to pay a brideprice of exactly 8000 cash. On this day, Mr. Wu has paid that amount in full, and I, Deng Zongjun, have received it with my own hands; nothing more is owed me. Since I have authorized this marriage, Mr. Wu is now entitled to take Zhou Shi as his wife. No one from the Deng family or Zhou Shi’s natal family, regardless of whether they are now present or not, or whether they are old or young, nor any other relative of either lineage, may say anything different or otherwise interfere. The three families [i.e., Deng, Zhou, and Wu] are perfectly willing to agree, and there has been no coercion or deceit of any kind. Fearing that spoken words alone are not reliable, I hereby establish this contract voluntarily authorizing my wife to marry, for Wu Rongzhong to keep forever as proof.

This contract portrays the decision as undertaken by husband and wife together, in consultation with both their families. Here we see the tension inherent in these transactions between woman as agent and woman as commodity (to be explored in Chapter 7). In addition, the seller guarantees that no one from either family will interfere. The contract is dated Jiaqing 12.12.15 (just before the New Year) and bears Deng Zongjun’s mark (a cross), followed by names of the matchmaker, scribe, and five other witnesses, along with Deng’s left handprint and left footprint. His hand was cupped to leave a blank space on the paper inside the palm, in which are written the characters “perfectly willing” (xin’gan). The contract was folded in half along its vertical axis after the text’s ink was dry but the hand and footprints were still wet, so that they transferred to the right half of the document. In other words, the document was given to the buyer immediately after the hand- and footprints were applied, this being the final act in the transaction after the money had been handed over. The magistrate later canceled this contract by drawing a circle through its text in red ink.60 A fourth example, from Zhanghua County, Taiwan, dated 1891, is more elaborate than the other three, but it exhibits the same basic features: I, Huang Ji, of Gounei Village, Banxian Bao, in Zhanghua County, hereby establish this contract to sell my wife (mai fu zi). In the past, I went through a matchmaker to take Xiao Queniang as wife, and now she has reached the age of 16 sui, without bearing any children. My wife and I quarrel night and day over the most trivial things, without any resolution. The reason is that we are very poor and have trouble getting our daily food. Now, since I am short on money to cover my expenses, I am voluntarily willing to marry Xiao Queniang off to another man. I have engaged matchmakers to sell her in marriage to Zhou Hao, of Quanzhouxi Village, located in the same Bao, to be his wife. With the matchmaker present, and in accordance with current market

140

part two

prices (yi shi zhi), we have negotiated a wife-sale brideprice of exactly 30 large silver yuan, and this amount has been paid in full in the matchmaker’s presence. Since Xiao Queniang is sold to Zhou Hao as wife on this auspicious day, the two of them will be husband and wife forever, to bear sons and grandsons, all of whom will belong to Zhou Hao (ri hou sheng zi chuan sun xi Zhou Hao zhi wu), and will have no connection whatsoever to Huang Ji, who may not dispute this agreement or cause any other trouble or complication. After this sale, Huang Ji will have no connection with Xiao Queniang. There is no question about Xiao Queniang’s prior status or about the exchange taking place today or any other obstacle. In case of trouble, Huang Ji will step forward to take full responsibility, without involving Zhou Hao in any way. This agreement is completely voluntary for both parties, and neither will have any change of heart. Now, in order to have evidence of this agreement, I hereby establish this wife sale contract and give it to Zhou Hao to keep as proof. Note: on this day brideprice of exactly 30 large yuan of silver mentioned in this wife sale contract has been paid in full in the matchmakers’ presence.

The contract closes with the names of three matchmakers (one of them the seller’s brother) and the scribe, along with their marks, and the name and mark of seller Huang Ji (the scribe has an elaborate huaya signature, the others each have a circle with a dot inside). It is dated Guangxu 16.12—again, just before the New Year.61 This Taiwan contract is refreshingly free of euphemism, with its repeated use of the verb “sell” (mai) and its reference to “current market prices” (which closely mirrors the language used in many land sale contracts). The statement that sons and grandsons born in the future will belong to the buyer, if translated literally, actually says that they will be his “things”—wu—a term that usually refers to material objects. The contract is also striking for its frank account of marital discord exacerbated by poverty, which suggests that this wife wanted to leave her husband at least as badly as he wanted to sell her; this account strongly resonates with the testimony found in many legal cases. These four contracts represent the north China plain, the Yangzi Delta, Sichuan, and Taiwan, and they span nearly a century and a half. Their details vary, but they follow a common pattern, and in their basic features and some of their specific wording they closely resemble each other. They (and dozens of other examples) show that wife sale contracts were part of the uniform vernacular contract culture found throughout China Proper before the revolution. This uniformity underscores the fact that wife sale was not an isolated or particularly exceptional phenomenon but rather a widespread, routine transaction governed by well-known customary norms and rules. Drafting the Contract As these examples show, a wife sale contract had to be written in the voice of the husband and narrated in the first person in order to demonstrate his consent to the

Chapter 4. Anatomy of a Wife Sale

141

transaction. Ideally, the seller would actually write the contract himself. Several cases record that the seller copied a draft written beforehand by a scribe, underscoring the symbolic importance of the seller’s own hand.62 But, given the ubiquitous illiteracy among the poor, it was usual for a literate acquaintance or a hired scribe to write the contract, which would be formalized by the seller’s mark and handprint.63 Most contracts are written in the clear hand of a scribe, who is identified by name after the list of witnesses’ names. But examples written by the sellers themselves do survive: written in a childish, untutored hand, these documents are full of errors (characters are written incorrectly or replaced by homonyms of different meaning), and can be difficult to puzzle out. In an 1867 contract from Baodi County, for example, the seller has written “mei” (meaning “plum”) for “hui” (“regret”) and has substituted “mu” (“mother,” but which resembles the character “mei” meaning “every”) for “mei” (“matchmaker”); the text contains other errors, and the whole document is written in an uncertain scrawl, suggesting the effort of a marginally literate person to write down words being dictated, or perhaps to copy a model text. Regardless of who actually set brush to paper, it was absolutely critical that the seller himself apply his personal mark to the document in order to approve and actualize the agreement. This mark would be a cross or “x” under the seller’s name, almost always supplemented by his inked handprint and sometimes his footprint as well. These personal marks of the seller constituted the crucial element of a wife sale contract, and if necessary they carried contractual force even in the absence of written text. For example, in a 1740 case from Qishui County, Hubei, both seller (a migrant laborer) and buyer (a peasant) were illiterate, and no matchmaker or scribe was available, so they negotiated directly, and in lieu of a proper contract, the seller gave the buyer “a piece of paper with a handprint and a footprint.” This case is by no means unique.64 The application of the husband’s handprint to contracts for marital separation (and, more generally, of the seller’s handprint to contracts for the sale of persons) has a long history in China, dating back to the Song dynasty and perhaps earlier.65 A number of Ming sources document this practice. For example, during the Jiajing reign (1522–66), a “statute applied by analogy” clarified the procedure for a valid divorce: If husband and wife do not get along and are willing to divorce, they shall be allowed to do so. The divorce contract must be written by the husband himself, and must bear his handprint with his signature or mark written in the gaps between the fingers. A family member or the original matchmaker may write the contract for him only if he cannot write.66

Some Qing wife sale contracts bear similar iconography: between the fingers of the handprint or within the blank space created by cupping the palm of the hand, one

142

part two

finds such phrases as “voluntary and acting of my free will” (xin’gan qingyuan).67 Another Ming example appears in a collection of cases published in 1632 by the official Su Maoxiang. In this case, a wife sale was followed by attempted extortion and the filing of false charges against the buyer by the seller. Su notes that a contract bearing the plaintiff ’s handprint proved that he had sold his wife voluntarily.68 The handprint on a wife sale contract had nothing to do with the modern forensic purpose of using fingerprints to identify an individual.69 The handprints are usually heavily inked and solid black, so they could not have been used for this purpose anyway, and in some variations one finds an outline of the hand with writing inside, instead of an actual print. Rather, the preparation of the contract constituted a public ritual before witnesses named in the contract. This ritual climaxed with the physical act of placing the handprint, an act that embodied the husband’s voluntary agreement to the sale. According to Tai Yen-hui, “the palm would be completely opened to show that the husband would not reconsider and try to reclaim his wife.”70 The power of this ritual is attested by the vivid appearance of such documents: one feels intensely the presence of the man who, by affixing his dripping inked hand to paper, made his wife’s sale final. An 1825 case from Huan County, Gansu, illustrates the ritual significance of the handprint in sealing a wife sale. In this case, peasant Chang Erhu (29) found himself entrapped into selling his wife, Li Shi. Chang had already divided his own deceased father’s household with three brothers, and he was very poor. Li Shi’s father, Li Chengfang, who was very fond of her, worried about Chang’s ability to support her (in Chang’s words, “Li Chengfang resented that my poverty made his daughter suffer, and he wanted to marry her off to someone else”). Li often asked Chang Erhu to divorce her and “return” (tui) her to her natal family, but Chang refused. Finally, when Li Shi came back to her natal home for a visit, her father kept her there, and when, after a month, Chang came to fetch her, Li Chengfang would not let her leave. In the meantime, Li had engaged a matchmaker who arranged for a mutual acquaintance from a neighboring village, Guan Dapeng (59) to buy Li Shi as a new wife for his son (whose first wife had recently died). Guan (who was unaware that Chang Erhu had not consented to sell his wife) agreed to pay 6 taels of silver, which would constitute an exact refund of the brideprice Chang had originally paid for Li Shi when he married her.71 After a couple more weeks, Chang came again, and Li Chengfang made him agree to meet at the home of his brother-inlaw Wang Keyi (the husband of Chengfang’s younger sister) to talk things over— separately, Guan had been told to be there at the same time to pay the brideprice and watch the marriage contract be drawn up. Thus, when Chang Erhu arrived at Wang Keyi’s home, his father-in-law was waiting for him, along with the matchmaker and prospective buyer Guan Dapeng. For the first time, Chang was informed that his wife had been promised to another

Chapter 4. Anatomy of a Wife Sale

143

man; the brideprice Chang had originally paid for Li Shi would be refunded in full, but he had no choice in the matter. Chang reacted to his father-in-law’s announcement with shock and anger. Li Chengfang (who had prepared brush, ink, and paper) ordered Chang to write out a “wife sale contract” (mai qi wenqi) confirming the stated terms, but Chang claimed that he could not do so because he was illiterate, so Li wrote it for him. Once the contract was written, Li ordered his son-in-law to add his handprint and footprint to the document. Everyone present understood this to be the crucial moment, on which the success of the transaction depended— and Chang refused. Li began cursing him and tried to strike him. A fight ensued, in which Chang ended up stabbing his father-in-law to death.72 The failure of Li Chengfang’s fait accompli reveals the assumptions customarily framing a wife sale. Despite his arrangements, success required the husband’s handprint; without it, no one would accept a contract as valid, and the transaction could not proceed. If, after selling his wife, a man filed false charges against her buyer in an attempt to extort more money, the buyer might call his bluff by presenting the handprint contract in court. The contract constituted evidence of the crime of wife-selling, to be sure, but it also proved that the seller had acted voluntarily and therefore was equally culpable. A seller might try to preempt this defense by explaining away his handprint on the contract. Thus, in an 1825 case from Ba County, Li Dazhi filed charges against the Qi brothers, Youdong (who had acted as matchmaker) and Youliang (the buyer), accusing them of tricking him and forcing him to sell his wife. According to Li’s indignant plaint, the Qi brothers had gotten him drunk and taken advantage of his illiteracy and naïveté by inking his hand and pressing it onto a piece of paper without explaining what they were doing. Only later, after sobering up, did Li realize that he had unwittingly sold his wife. That was Li’s story—but the magistrate was not impressed.73 Once the seller had affixed his handprint, he would give the contract to the buyer along with the woman, in exchange for her brideprice. Some contracts describe this exchange explicitly: a specimen from Nanbu County dated 1836 contains the following statement: “Here the husband has clearly placed his footprint and handprint. With his left hand he delivers the woman, while with his right hand he receives the money.”74 The contract would stay in the possession of the buyer as his title deed to the woman (sometimes referred to as her “body contract,” shen qi). In this respect, too, wife sale contracts resembled land sale contracts, which functioned as title deeds to the sold land. Many cases illustrate how seriously people viewed this function of wife sale contracts. For example, in a Ba County case from 1821, the woman’s new husband became so weary of the seller’s demands for more money that he convened a meeting where, before witnesses, he destroyed the contract and then returned the woman to her first husband without demanding a refund. This destruction

144

part two

symbolized the second husband’s renunciation of the woman, which eight witnesses confirmed in an affidavit that he kept in place of the contract.75 If a sold wife was unhappy with her new husband, she might try to destroy the contract that documented his claim to her. In a 1754 case from Jinyun County, Zhejiang, Ge Shi burned her sale contract and then confronted her buyer with this fait accompli, demanding that she be allowed to leave him and return home. She threatened that, if he tried to stop her, she would file charges against him for “breaking up her marriage in order to steal a wife who already has a husband.”76 In other cases, women who ran away from their buyers first stole and destroyed their contracts, apparently on the assumption that without such documents these men could not force them to return.77 In an 1862 case, an unhappy wife filed charges at the Ba County yamen to stop her husband from selling her; as soon as the transaction was complete, she had managed to steal the contract and run away to the yamen, where she presented it in evidence.78 Finally, a number of cases record dramatic scenes in which someone intervened at the last moment to stop a sale by seizing and tearing up a contract that had just been drafted. Predictably, this maneuver would provoke uproar.79 A Summary Example The following case from Sha County, Fujian, memorialized in 1774, brings together many of the themes of this chapter. In this case, a man’s two attempts to sell his wife both fell through, due to his failure to secure the approval of both families and the consequent refusal of the prospective buyers to finalize the transactions. Problems related to contracts also played a role, especially in the failure of the second attempt. Peasant Chen Dequan (37) was orphaned at an early age and raised by his uncle Chen Qiwu, who provided him with a much younger adopted daughter-in-law, Lin Shi (23), for his wife. The couple had no children. Lin Shi later described her husband as a man who “liked to eat but was too lazy to work, and didn’t take care of our household.” The couple often went hungry and lacked proper clothing and other necessities. Finally, Dequan’s uncle Qiwu intervened; he could not support the couple himself, so he told Lin Shi’s older brother Lin Maoqing (the senior member of her natal family) to take her home for the time being. From then on Lin Shi lived with her brother, although Dequan often visited, staying for a few days each time. Five months later, just before the New Year (in the early spring of 1772), Dequan decided that he should sell his wife. A barber acted as matchmaker, negotiating a deal whereby peasant Zheng Huashi (64) would buy Lin Shi as a wife for his son, Hongxu (29) (the case record does not give the brideprice). The Zhengs made a down payment of 2000 cash in exchange for a contract, with the understanding that Lin Shi would join the Zheng household at the New Year in exchange for the

Chapter 4. Anatomy of a Wife Sale

145

balance owed. Before taking that final step, however, Zheng Huashi insisted that the matchmaker secure the blessing of Dequan’s uncle, Chen Qiwu. But Qiwu did not approve, and instead he threatened to file charges against the Zhengs for “plotting to take a married woman in marriage” (mou qu sheng qi). When they heard this threat, the Zhengs canceled the deal. Unfortunately, Dequan had already spent their down payment, so the Zhengs retained the contract for security, saying they would surrender it only after getting a refund. According to Zheng Huashi, “That Chen Dequan was lazy and irresponsible and really poor, and even though I asked him many times to repay my down payment, he never gave me a single coin.” Despite his uncle’s opposition, Dequan remained determined to sell his wife. Two months later, he managed to secure a second buyer. Peasant Lü Baolang had asked matchmaker Su Yulong to find a wife for his son Lü Tianfu (32), who had not yet married, and Su had heard that Chen Dequan wanted to sell his wife. The Lüs agreed to buy Lin Shi for a brideprice of 40 taels, and a meeting was scheduled to make the final exchange. Dequan’s literacy was marginal, so he asked an acquaintance named Wang Weifan, who was a military licentiate (wusheng), to write out a draft contract stating the agreed terms (which Dequan related orally), and then Dequan painstakingly “copied this draft in his own hand to draw up the formal contract.” He then took this formal copy to the meeting where he would add his mark and handprint. At the meeting, however, Lü Baolang refused to proceed. He had learned that Dequan had previously promised his wife to the Zheng family, and that the Zhengs still had a contract proving their claim to her. (Dequan and matchmaker Su Yulong had neglected to inform the Lüs of these facts.) Lü Baolang insisted that Dequan and the matchmaker retrieve the old contract from the Zhengs in order to ensure that the previous transaction had really been canceled. Also, when the new contract that Dequan had prepared was read out, Lü Baolang complained that no representative of Lin Shi’s natal family was listed as witness or guarantor, and instead of letting Dequan place his mark and handprint on it, he tore it up. Dequan assured him that Lin Shi’s brother Lin Maoqing approved the sale, but Lü would not take his word for it, and he demanded that a new contract be drawn up with Lin Maoqing as witness. Only after these conditions were met would the Lü family discuss the matter further. Shortly thereafter, Su and Dequan visited Lin Maoqing at his home and explained the situation. Su proposed that the other two accompany him to ask the Zhengs for the original marriage contract, but Dequan explained that the Zhengs would relinquish the contract only in exchange for a refund of their down payment, and he told Su to get 2000 cash from the Lü family so that they could “redeem” the marriage contract from the Zhengs. Su replied that since Lü Baolang no longer trusted him or Dequan, he was unlikely to agree to pay 2000 cash in advance unless Lin Maoqing would go along and personally guarantee the money

146

part two

(demonstrating his approval of his sister’s sale). By this time, however, Lin Maoqing was thoroughly disgusted with his brother-in-law—“you already cheated the Zheng family out of 2000 cash, and now you are trying to cheat the Lü family too!”—and he refused to help. Dequan became very angry, and the two men got in a fight, which ended when Maoqing beat Dequan to death.80 This case illustrates several key points. First, one sees the importance of securing the approval of both the seller’s family and the woman’s natal family in advance of a wife sale. Chen Dequan twice attempted to sell his wife (and he did so locally, so word got around), but he acted behind the back of his uncle, who had provided him with a wife in the first place. His uncle’s threat to file charges compelled the Zheng family to cancel the first sale, and, had Lü Baolang known of it, he certainly would have canceled the second sale, even in the absence of other concerns. Although Lin Maoqing did not try to prevent his sister’s sale, he never agreed to play an active role, and when pressed to do so he refused. The lack of a guarantee from the natal family helped scuttle the second sale. In addition, all of these negotiations had taken place without Lin Shi’s knowledge—Dequan was waiting until the last minute to tell his wife of his plan to sell her, and if she had refused to cooperate, it is unlikely that the sale would have succeeded, since she was living under her brother’s protection. Second, one sees the multifaceted significance of the contract and of the meeting at which the final exchange would take place. The Zheng family’s retention of the first contract posed an insurmountable obstacle to the second sale, because everyone understood the contract to be a title deed to the sold wife. The special effort by Dequan to copy out the second contract from a rough draft demonstrates the importance of having it written by the seller himself if at all possible. But this document had no standing until it had been finalized at the face-to-face meeting with the buyer, matchmaker, and witnesses, when Dequan would make his mark and handprint in exchange for the brideprice. Instead, the buyer tore it up, because (even though he was illiterate) he wanted the natal family’s approval in writing. This obsession with getting the contract absolutely correct did not relate to that document’s standing in court. Rather, it was geared toward ensuring that the transaction was honest and above-board, in order to prevent future trouble. T H E PA R A D OX O F T H E W I F E S A L E C O N T R AC T

To conclude this chapter, let us note the apparent paradox of the wife-sale contract. In late imperial China, most contracts (for land sales, business deals, debt, etc.) were designed to be presented in court, if need be, in order to enforce the transactions they documented.81 In contrast, a wife sale contract documented a crime, and the only reason we have so many originals to study is that magistrates often confiscated them in the course of prosecution. In my sample of local cases

Chapter 4. Anatomy of a Wife Sale

147

involving illegal wife sales that went to formal judgment, magistrates canceled two-thirds of these transactions, and, in most of those cases, one or more of the parties ended up with a beating (see Chapter 11). Under the circumstances, it was risky to present a wife sale contract in court. This fact raises the deceptively simple questions of what such a contract was really for, and how it was to be enforced. In this connection, I will make three points, all of which anticipate material covered in detail in later chapters. First, a contract for an open wife sale was specifically designed to minimize the risk of any dispute ending up in court. As we have seen, a basic function of such a contract was to document the consent of the seller: it was written in his voice, ideally in his own hand, and confirmed before witnesses with his handprint. An equally important function was to ensure what we might call the “informed consent” of the buyer—that is, to make sure he knew that he was buying “the wife of a living husband” and that he and the seller had a clear understanding, without any deceit or coercion. Such documents often record that the seller, the buyer, the wife herself, and their respective families had all agreed to the transaction, and the seller would guarantee against interference from any quarter. The list of matchmaker(s), witnesses, and guarantors (sometimes including local notables) reinforced this point: if any dispute were to arise, the parties should consult the people on this list. The purpose was to preempt any misunderstanding that might lead to trouble and cause it to become an object of official scrutiny. Such contracts were designed not to “stand up in court” but rather to avoid ending up in court at all.82 That was the first line of defense—to ensure that any dispute would be mediated without resort to the courts. But most people knew that wife sales were illegal, and they also knew that such cases did on occasion end up in court, often because sellers would file false charges against buyers in an effort to extort supplementary payments. Hence, the second line of defense was to protect the buyer against any accusation of forcing the husband to sell his wife, or of kidnapping or raping her, or of any other heinous capital crimes. If worse came to worst and the parties did end up in court, at least the buyer could prove the seller’s complicity and culpability. In other words, a basic reason for having the contract written in the voice of the seller and confirmed with his handprint was to make it difficult for him to frame and blackmail the buyer later on. The third point—and perhaps the most paradoxical of all—is that about a third of illegal wife sales that were subject to formal court judgment ended up being ratified by magistrates, who allowed the second marriage to stand. Generally speaking, it is axiomatic that courts cannot enforce a contract made for an illegal purpose,83 but, in these Qing cases, magistrates did precisely that. In these case files, we do not find the original contracts, because they were retained by the buyers: once a magistrate had ruled in this manner, the wife sale contract became, in

148

part two

effect, a legitimate marriage contract. Such retroactive legitimization resolved the paradox of the wife sale contract. How, then, was “legality” defined in Qing judicial practice? In these cases, “legality” as defined by a magistrate’s judgment contradicted and overruled “legality” as defined by the Qing code. In routine cases involving “minor matters,” Qing magistrates clearly had the authority and discretion to prioritize considerations other than codified law. (We shall examine this theme in Chapter 11.) Of course, two-thirds of the marriages resulting from wife sales did indeed end up being canceled, so the odds weighed heavily against a buyer who dared to take his contract to court. But we cannot rule out the possibility that some buyers were aware of the one-in-three chance that their marriages would be ratified.

5

Analysis of the Prices in Wife Sales Beat your wife, curse your wife, and if you run out of money then sell your wife (da lao po ma lao po, shou nei wu qian mai lao po). —proverb from hangzhou, late eighteenth century 1

How were prices in wife sales determined, and how much were they worth? I have collected dozens of wife sale prices from Qing legal cases, but their meaning is seldom transparent, and it can be difficult to determine their real value in local context. This chapter attempts to make sense of these prices and, in the process, to illuminate the economic logic of wife sales in comparison to the broader spectrum of marriage practices in late imperial China. To estimate the real value of wife sale prices, I begin by comparing them to the prices of other basic commodities (such as agricultural labor, grain, and draft animals), both in the aggregate and in several focused case studies. Such comparison establishes a basic and important fact: with few exceptions, these transactions really were sales, which involved very considerable sums of money given the standard of living of most of the people involved. At the same time, it is clear that wife sale constituted a relatively inexpensive way to acquire a wife. Probably the main factor depressing the prices in wife sales is that, in most cases, the seller had to accept the first offer he received. One finds a wide range of prices, especially in places of socioeconomic complexity like Ba County, with the highest prices being denominated in silver taels. In contrast, prices in a simpler and more typical locale like Nanbu County clustered in the mid-range and were usually denominated in cash (the typical medium of exchange between peasants). An analysis of the extreme ends of the price scale in Ba County reveals two very different kinds of logic that framed decisions and determined prices. Both extremes differ from the broad mid-range of prices found in most wife sales. In addition, since children often accompanied their mother into her new household, the terms of custody were a major factor in determining prices. 149

150

part two

In this chapter’s final section, I use a few exceptional references to dowry to illuminate, by contrast, the economic logic of wife sales as a subset of the wider spectrum of “brideprice-heavy” marriage practices that prevailed among the poor.2 Most of the practices in this wider spectrum can also be seen as sale of one sort or another (and sometimes were explicitly labeled as such), because the price paid for a wife far outweighed the value of any goods she brought with her. As a rule, dowry was not a factor in wife sales, because the wives being sold had not brought dowry from their natal families into their first marriages; in effect, their parents had sold them. Major marriage with dowry was the normative ideal and a status symbol that advertised the morality and prosperity of its practitioners, who were unlikely to end up in the dire straits that usually triggered wife sales. As a corollary, it seems to have been taken for granted that, if a man had not bought his wife from her parents in the first place, then he had no right to sell her either. In other words, if marital separation occurred, its terms reflected the terms of the couple’s original marriage, and in the unlikely event that a major marriage with dowry did end in divorce, that separation would not likely take the form of a wife sale. But all the evidence suggests that the vast majority of marital separations took the form of wife sales (direct or indirect) and were motivated by poverty. In these transactions, as in the original marriages that preceded them, dowry played no role whatsoever. A N OV E RV I EW O F P R IC E S I N O P E N W I F E S A L E S

As a first approximation, let us survey the known prices and compare them broadly to what we know of other prices. Readers can find detailed price information in Appendix C; here I summarize the highlights. I have many wife sale prices from both xingke tiben and local court cases. These two samples should be considered separately, because the xingke tiben come from all provinces of China Proper, whereas the local cases are concentrated samples from particular places (the vast majority from Ba and Nanbu counties in Sichuan). A limitation of both samples is that they are scattered over long periods of time, during which commodity prices, exchange rates (between silver and coin), and local socioeconomic conditions framing the traffic in women all presumably changed. For all of these reasons, it would seem unwise to use the aggregate data to draw any very precise quantitative conclusions. Here we focus on the immediate cost of open, direct wife sales (i.e., not including fraudulent sales or compensated divorces, both of which are covered in Chapter 8): these prices represent the entire sum spent by the buyer at the time of the transaction, including the brideprice itself plus any fees paid to matchmakers, scribe, or other parties. I do not include any supplementary payments that may have been paid after the sale was finalized. Some prices are expressed in taels (sil-

Chapter 5. Analysis of Prices in Wife Sales

151

ver being the usual medium of payment for large sums) and others in cash (the usual medium for smaller sums). In xingke tiben from the Qianlong through early Daoguang eras, the average wife sale price in taels is 20.3, and the average price in cash is 13,700.3 We can get a sense of the real value of these prices by comparing them to annual wages paid to agricultural laborers found in xingke tiben from the same period: most were paid in coin and fell in the range of 3000 to 7000 cash; wages paid in silver averaged about 5 to 7 taels.4 In sum, the average price of a wife sale found in xingke tiben would have paid roughly two to four years’ wages for an agricultural laborer (see Appendix C.1). We can compare wife sale prices in Ba County to the prices of oxen and unhusked grain (see Appendix C.2). In terms of real value, wife sale prices in Ba County are broadly comparable to those found in xingke tiben—details vary, but we find the same order of magnitude. The average price in wife sales paid in cash (8500 cash) approximated the average price paid in cash for cattle (including water buffaloes and common oxen of varying quality).5 In fact, the single highest price listed for cattle in Ba County was 27,400 cash, paid in 1792 for a large common oxen (huang niu) bull, which was probably so valuable because it could be used for stud.6 This amount is just under my highest price (in cash) in a Qianlong-era wife sale in Ba County, 28,000 cash, paid in 1780 by a minor degree holder for a young woman to serve as his concubine.7 But if one considers prices paid in taels (which include high-end sales of young women to serve as concubines), the picture changes: the average price in a wife sale, 24.5 taels, is just over four times the average price for cattle, 6 taels. In sum, the prices paid in wife sales might have bought anywhere from one to four head of cattle. To keep these figures in perspective, one must remember that, as draft animals, cattle were major capital investments and would rarely be slaughtered for food.8 In Ba County, the average price of unhusked rice paid in taels was 1.8 taels per shi; the average cash price was 1750 cash per shi.9 In eighteenth-century China, according to one estimate, “average annual per capita grain consumption (including children and adults) seems to have hovered around 2.5 shi of husked grain, or 5 shi unhusked.”10 This estimate does not distinguish by region or social class, and no doubt the poor consumed less. An average of about three shi of unhusked grain was necessary for basic subsistence, so 3–5 shi unhusked (i.e., 1.5–2.5 husked) probably represents the actual range of annual per capita consumption among the peasantry in normal times.11 On this basis, we can estimate that, in a normal year, the average wife sale price in Ba County would have bought roughly one to four years’ rice supply for one person, or up to one year’s supply for a single household. In a famine year, when the price of rice might rise by 500 percent or more, it would have bought far less. To reiterate, these data permit only the broadest of estimates. The main conclusion to draw from the discussion so far is that wife sale prices usually constituted

152

part two

substantial sums of money for the people involved: the equivalent of two or three years’ wages or food supply, or a couple head of cattle. Nevertheless, one can imagine such prices being within reach of a man of modest means, if he could save up over a number of years. T H E R E A L VA LU E O F W I F E S A L E P R IC E S : C A SE ST U D I E S

Given the difficulty of standardizing prices across any very wide geographic or temporal scale, it would be ideal to have concrete examples of what a given wife was worth in terms of some other commodity at a given time and place. Fortunately, in a small number of intriguing cases we find such evidence, which enables us to determine with some precision the purchasing power of a given price. Livestock and Coffins A few cases provide equivalent values in terms of livestock or coffins. Aside from land or a wife, a draft animal was the largest capital investment a male peasant would be likely to make (and not every peasant had one). In order to buy a wife, therefore, some men sold draft animals or simply used them to pay in kind. The reason coffins appear in some cases is that one motive for wife sale—a particularly laudable one, in terms of Confucian values—was to raise enough money to buy a good coffin for one’s parent or grandparent. As a result, these cases tell us how much a given wife was worth when exchanged for draft animals or coffins. Several examples follow below. In a 1736 case from Xuyong Independent Subprefecture, Sichuan, a brideprice of 20 taels was paid in the form of 13 taels of silver and a horse valued at 7 taels. The matchmaker’s fee was 0.5 tael, which the buyer paid separately in kind with a bolt of cloth. The woman in this case was 32 sui and had borne two surviving sons, but her obvious value was offset by the contractual stipulation that the buyer must raise her two sons for several years and then return them to their father’s lineage (I discuss such arrangements below). This relatively young wife of proven fertility cost her buyer roughly the equivalent of three horses; without her sons, she would have cost even more.12 In a 1754 case from Anlu County, Hubei, a peasant sold his ox for 6500 cash to raise part of the money to buy a woman and her little daughter. The total price was 30,500 cash, approximately double the average wife sale price found in my Qianlong-era xingke tiben. This relatively high price reflected the fact that the deal included the woman’s daughter, but also, it should be noted, this was a fraudulent sale, in that the buyer believed that the woman was a widow. This transaction cost the buyer roughly the equivalent of five head of cattle. We can assume that he

Chapter 5. Analysis of Prices in Wife Sales

153

would have paid considerably less had he known that he was buying “the wife of a living husband.”13 Similarly, in a 1757 case from Sui Department, Hubei, a brideprice of 18,000 cash was paid in kind, in the form of a horse worth 11,000 cash and an ox stated to be worth 7000 cash. (The ox later sold for only 4500 cash, so the man who sold his wife may have been cheated.) One of the protagonists in this case separately spent 9500 cash to buy a nice coffin for her dead husband. In other words, this wife cost a horse and an ox, or slightly less than the equivalent of two coffins.14 In a 1755 case from Shanghai County, Jiangsu, a peasant who sold his wife for 24 taels used 8 taels out of the brideprice to buy coffins for his paternal grandparents. At a rate of 4 taels per coffin, this wife was worth six coffins.15 In a similar case the following year, from Xiangtan County, Hunan, a peasant sold his wife for 12 taels mainly in order to buy a coffin for his father, who had recently died. We do not know the price of this coffin, but presumably it cost less than 12 taels—her brideprice was probably worth two or three coffins, depending on quality.16 Obviously, the price of livestock can vary widely (a racehorse costs more than a nag), and the same is true of coffins. But the relative costs of draft animals and wives in these concrete examples are broadly consistent with the other price data. More generally, these examples underscore the fact that wife sales involved sums comparable to other major investments, like draft animals and coffins. Indeed, given the mutual convertibility of wives, draft animals, and coffins, one can imagine them as parallel columns in a ledger, between which peasants made choices and allocated resources accordingly. The Value of a Woman’s Labor In a complicated case from rural Shangrao County, Jiangxi, prosecuted in 1748, we learn the prices of a number of different things. In this case, a peasant sold his 24-sui wife because of poverty, for a brideprice of 24 taels. For some time prior to the sale, the woman’s prospective buyer had been paying her father-in-law a stipend of 300 cash per month to help buy food, in exchange for being allowed to sleep with her. Not long after the sale, the buyer’s aunt sewed a pair of cloth shoes and sold them for 150 cash; this woman’s husband then used 60 cash to buy wine and get drunk. Thus, in a single case we have prices for a 24-sui wife, a monthly food allowance, a monthly supply of sex, a pair of shoes, and a drinking binge.17 These data give us a sense of the relative cash value of this wife—as well as an unusual glimpse of the cash value of women’s labor (both handicraft production and sex) and its potential contribution to household income.18 Let us assume hypothetically, for the sake of discussion, that one-third of the sale price of the shoes was profit. At that rate, according to the evidence internal to this case, the profit on one pair of shoes (50 cash) would have bought almost enough alcohol to

154

part two

get a man drunk. If we estimate this woman’s brideprice of 24 taels to be worth 18,000 cash (at an exchange rate of 750 cash per tael, which is about right for the 1740s), then it would have taken 360 pairs of shoes to pay for her brideprice. In other words, if she sold two pairs of shoes per week, she could have paid off her entire brideprice in less than four years. If the woman’s own household grew the cotton, spun the yarn, or wove the cloth used to make the shoes, then the rate of profit might conceivably have been higher. Alternatively, if her profit margin were smaller, or if she sold fewer shoes, then it would have taken that much longer to pay off her brideprice—but eventually she could have done so, as long as she remained healthy and productive. These estimates are entirely speculative, but they illustrate several important points. The brideprice paid for this woman was, in real terms, a significant sum for the people involved. But women’s commercially marketed handicrafts could add a significant margin to household income. This case also confirms what we have already seen in the context of polyandry: that a woman’s sexual labor could contribute vital income and might even support her household’s entire subsistence. The 300 cash per month paid by the prospective buyer for sexual access would add up to an annual stipend of 3600 cash, comparable to the average annual wage for agricultural labor. In other words, the brideprice spent to buy a wife was by no means an economic loss, at least not in the long run. Rather, it was an investment that could be recovered and might even turn a profit, as long as the wife remained healthy and could work. Most buyers in wife-selling cases sought “to establish a household” (chengjia)—they wanted wives for sex and reproduction, to be sure, but no doubt they also saw women as producers and understood the economic benefits of this investment. Brideprice as an Object of Struggle We gain further insight into the real value of the wife sale prices from cases involving struggle over the money after it had been paid. The typical poor peasant, by definition, would rarely encounter a large lump sum like the brideprice paid in a wife sale. On the contrary, his normal condition was indebtedness that waxed and waned over the course of the annual agricultural cycle. There were not many ways for peasants to raise such a sum; selling land, livestock, children, or a wife were the principal means of doing so, and these were all more or less drastic measures, not to be undertaken lightly. When a man sold his wife, he would almost always be paid in money—and these payments could be physically bulky and conspicuous. Most payments took the form of cash (qian, the familiar round coin with a square hole in the center), which came in “strings” of up to one thousand coins each, and a woman’s brideprice might cost several thousand or even several tens of thousand cash. The latter

Chapter 5. Analysis of Prices in Wife Sales

155

sums would be too heavy and awkward for one man to carry in a single load, and a number of legal cases note that buyers had to enlist friends to help deliver the brideprice, or that its delivery required more than one trip. Once the seller had received this money, its secure storage might prove difficult, because of its bulk but also because it would attract envious attention. Under the circumstances, it is not surprising that this money might become an object of struggle. If we look at xingke tiben in which wife sales were completed, we find that one third came to court because of homicidal violence provoked by disputes over brideprice that had been paid (not including disputes over supplementary payments). These disputes played out in a few basic scenarios. In one, the seller would entrust his matchmaker (usually a relative or close friend) to receive the money on his behalf, but then the matchmaker would keep or spend a substantial part of it for his own use, without permission.19 In a second scenario, the seller would store what was left of the brideprice (after settling debts and immediate expenses) in some supposedly secure place, but a family member or friend in the know would use the money without permission.20 In a third, a family member or friend would borrow all or part of the brideprice but fail to repay the loan. Sellers’ older brothers figure prominently in such cases, no doubt because their seniority made it hard to refuse them.21 A single example will illustrate this pattern. In an 1820 case from Dazhu County, Sichuan, peasant Luo Nianyao (29) sold his wife in marriage for a brideprice of 16,000 cash; the main reason for this sale was the woman’s unhappiness, and Nianyao was hoping to use her brideprice to take a new wife in her place. But, after he sold her, his older brother Nianzhao (with whom he had not yet divided their father’s household) insisted on borrowing the entire brideprice, promising to repay it later so that Nianyao could remarry. (The case record does not say how Nianzhao spent the money.) But whenever Nianyao broached the subject, his older brother became abusive, scolding and beating him. Finally, after three years, Nianyao arranged to take a new wife, but still his brother failed to repay him. Nianyao managed to borrow enough money to pay his new wife’s brideprice (amount unknown, but presumably about 16,000 cash), and he took her in marriage. But these loans would fall due at the New Year, and so in the eleventh lunar month, with growing anxiety, Nianyao proposed to Nianzhao that they sell some family land in order to repay his creditors. Evidently, land was the only family asset available that was of comparable value to the brideprice, a fact that underscores its significance. But Nianzhao refused permission to sell the land, provoking a violent quarrel in which Nianyao ended up killing his brother.22 When a man sold his wife, his relatives, friends, and neighbors might perceive her brideprice as a windfall that offered all sorts of tempting opportunities. The common denominator among these scenarios is the temptation posed by this large lump sum.

156

part two

A R E L AT I V E LY I N E X P E N SI V E WAY T O M A R RY

As we have seen, the prices in most wife sales represented significant sums for the people concerned. Nevertheless, buying another man’s wife in an open sale was considered a relatively inexpensive way to acquire a wife, and that was the source of its appeal to the typical buyer, a peasant who had never married or whose first wife had died. As one observer wrote in 1921, “where there is a scarcity of women a sale can be usually effected even in these modern days, and it is considered a cheap way of getting a wife by the poorer classes.”23 Occasionally, the testimony in legal case records makes this point explicit. In a 1736 case from Wenshui County, Shanxi, for example, a matchmaker approached peasant Wu Youzhang with an opportunity to buy the wife and infant son of another peasant named Wang Xi. Wu Youzhang was a single tenant farmer who had never married, and he was eager to get a wife if he could. The matchmaker urged Wu to make an offer, reminding him that this kind of marriage “would cost less money than getting a wife some other way,” and Wu agreed. As he later testified, “In the past, I could never afford to take a wife (qu bu qi furen). I realized that buying Wang Xi’s wife would be less expensive than other options, and on top of that she would bring a child with her. So I agreed, and I told the matchmaker to offer Wang Xi 24 taels.”24 By “getting a wife some other way,” the matchmaker presumably meant marrying a widow or acquiring someone’s virgin daughter through major marriage with brideprice (i.e., buying her from her parents). Apparently either way would have cost more than the amount Wu spent to buy Wang Xi’s wife, which both Wu and the matchmaker agreed was a bargain. We find similar testimony in a 1750 case from Liancheng County, Fujian, in which peasant Li Zhanlian sold his wife, Huang Shi, because of adultery. Her buyer was a peasant named Zhang Xuemei, whose first wife had died without issue. Zhang was an orphan who had been raised by his widowed aunt. Although he was ritual heir to his father and uncle both, he had neither a wife nor a son, and this caused him and his aunt great anxiety. But even though he wished to remarry, he had been unable to do so. His chance came when he heard that Li was asking “only” 30 taels for his wife: according to Zhang’s testimony, “when Li Zhanlian offered to sell his wife for a cheap body price, my aunt raised 30 taels of silver by selling some land,” and they used that amount to buy Huang Shi to be Zhang’s wife.25 Thirty taels is a higher price than the average open sale price paid in taels found in my sample of Qianlong-era xingke tiben. Moreover, the sale of land in order to buy a wife suggests the need for a major capital investment. But in this local context, evidently, 30 taels was a “cheap body price”: cheaper than other ways of getting a wife, and cheap enough to enable a poor widower to remarry. Zhang and his aunt

Chapter 5. Analysis of Prices in Wife Sales

157

had been weighing the cost of marriage for quite a while, but only with the sale of Li’s wife did they feel able to afford it. Perhaps the shortage of wives in this particular community was especially acute (which would explain both the woman’s relatively high price and Zhang’s utter lack of concern about stigma in buying an adulteress in open sale). This case reminds us that, in the eighteenth century, most markets were highly local, sex ratios varied widely, and the real value of a given sum in one time and place could be very different than in another time and place. The “Fire Sale” Logic of Most Wife Sales What made open wife sales a relatively inexpensive way to marry? Every price represented a compromise between the seller’s degree of desperation, his hope for a good price, and the local demand for wives. Demand, in turn, depended both on the relative availability of marriageable women and on a given woman’s desirability according to the criteria that mattered to potential buyers. In most sales, however, the husband trying to sell his wife had no choice but to accept the first offer he received. For this reason, sellers often had to settle for considerably less than they thought their wives were really worth. In a representative example from Ningdu Independent Department, Jiangxi, a landless peasant named Xie Bisheng sold his wife, Rao Shi. Their household had two little children and an elderly father to care for, plus Xie had contracted some sort of neurological or mental illness (feng bing) that made it difficult for him to hold down his job as an agricultural laborer. The decision to sell Rao Shi was made just after the New Year, in the spring of 1759. The first offer Xie received was “only” 28,000 cash. We do not know exactly how much this sum was worth in real terms at this time and place, but Xie rejected it out of hand. After a month had passed, however, no better offer had come along, and the family was running out of food. Xie found he had no choice but to accept the “low” offer of 28,000 cash.26 Like Xie Bisheng, most sellers had little bargaining power and had to accept whatever was offered. The feeling of having been taken advantage of provoked bitter resentment; it also reinforced the widespread assumption that sellers were the losers in wife sales and that buyers benefited at their expense. Such feeling certainly helped fuel subsequent demands for supplementary payments: when sellers demanded more money, they usually expressed indignation at having been underpaid. We see these forces at work in a 1796 case from Shilou County, Shanxi, in which peasant Nan Huaizhao arranged to sell his wife, Wei Shi, to a widower named Wang Peigan. As usual, the motive was poverty—Nan and Wei Shi were very poor, and they often went short of food. Wei Shi complained bitterly of their poverty and pestered her husband until he finally agreed that they should “separate in order to survive” (gezi tao sheng). The transaction included Wang’s promise to raise the couple’s infant son for five years before returning him to his father. Nan received

158

part two

14 taels and 3000 cash; he later testified that, even though this was not very much money, “at that time I thought only of survival, so I didn’t quibble about the body price.” Within three days, however, Nan had “spent the entire brideprice to redeem pawned belongings and buy food,” and, with his most pressing problems solved, he began to regret that he had accepted such a low price for his wife. “I suddenly regretted what I had done, and I decided to ask for a few more taels of body price as a supplementary payment” (yi shi aohui, yao zhao ji liang shenjia). Such feelings could provoke litigation, but also violence. Nan’s persistent demands for more money ended up provoking a fight with Wang, whom he killed.27 The Same Woman Sold Twice To assess the relative cost of different ways to take a wife, it would be ideal to have an example of a buyer considering “major marriage with brideprice,” widow remarriage, or open wife sale, and comparing the prices for each. But I have found no example of this kind—and I doubt that such a perfect market scenario often, if ever, occurred. Few (if any) of these transactions were part of a large, integrated market in which many women would be locally available for sale at the same time. In a fraudulent wife sale, the buyer would be duped into thinking he was marrying a widow. Therefore, the prices paid in such sales should approximate those paid in widow remarriage. Not surprisingly, average prices in my samples of fraudulent sales were substantially higher than those paid in open sales (see Appendices C.3 and C.4). Far more revealing than a macro-level price comparison, however, is a handful of cases in which one woman was sold more than once, in both open and fraudulent transactions. Following are four examples. First, in a 1736 case from Linjin County, Shanxi, a 29-sui wife was sold because of poverty, in an open sale, for a brideprice of 6 taels. Six years later, she had an affair and ran away with her lover, and later this man had her pose as his widowed sister-in-law and arranged her remarriage for a brideprice of 65 taels to a man who wanted a concubine in order to get a son.28 Second, in a 1794 example from Dingyuan County, Anhui, a wife was sold because of adultery, in an open sale, for a brideprice of 20,000 cash. But this new marriage was unhappy, so after one year her second husband resold her in a fraudulent transaction for a brideprice of 40,000 cash.29 Third, in an 1835 example from Ba County, a wife was sold because of poverty in an open sale for a brideprice of 4000 cash. She was unhappy in her new marriage, however, and she ran away. With the help of two matchmakers, she posed as a widow and arranged her own remarriage. Her third husband paid a brideprice of 10,000 cash, which she used to pay the matchmakers and to clear debts incurred in her flight.30 Fourth, in an 1876 case from Ba County, a wife was sold by her first husband because of poverty in an open sale for a brideprice of 14,000 cash. Later, her

Chapter 5. Analysis of Prices in Wife Sales

159

second husband died, and as a widow she remarried for a brideprice of 21,000 cash (which she used to clear her second husband’s debts), becoming her third husband’s concubine.31 In each case, the same woman fetched a far higher brideprice as a widow than when she was offered in open sale as “the wife of a living husband,” confirming the popular view that buying another man’s wife in an open transaction was a relatively inexpensive way to marry. T H E E X T R E M E E N D S O F T H E P R IC E S C A L E

Averages are misleading in this context, because wife sale prices ranged from very small to very large amounts of money, and the standard of deviation was high, reflecting the fact that these were usually isolated transactions rather than part of a generalized, unified market. (For example, prices denominated in taels in Ba County average 24.5 taels, but they range from as low as 2 taels to as high as 120— see Appendix C.2.) At the extreme ends of the price scale, we find different logics and criteria than in the broad mid-range that accounts for the great majority of transactions. The Low End of the Market At the bottom of the price scale, we find that desperation might force men to accept very low prices indeed. A dramatic example is a 1762 case from Linyou County, Shaanxi, in which Li Liangde sold his wife, Yang Shi. Li and his brother were landless agricultural laborers who had migrated from Fengxiang County some years before, in search of employment. In the late winter of 1761–62, in the last month of the lunar year, Li Liangde and his wife ran out of food, and they abandoned their home to beg in nearby villages. Li finally arranged to sell Yang Shi to another peasant, who was also quite poor, for just 1700 cash. (The couple had no children that we know of.) Then Liangde wandered off, and it is not clear what happened to him afterward. When Liangde’s younger brother Liangru heard what had happened, he was outraged—not so much that Liangde had sold his wife but that he had gotten so little for her: “I thought to myself, how is it possible that a woman could be worth only a thousand and some cash?” Liangru’s outrage may have been exacerbated by the fact that Liangde, as the oldest, had been the only brother to have married. Liangru could not find his brother, so he went to the buyer’s home, taking along three cousins (all of them landless laborers) for muscle, and demanded more money: “You illegally bought a married woman of my family (ni si mai wo jia huoren qi), and you paid only a thousand and some cash. You need to pay us a few taels more!” Rebuffed, Liangru and his cousins tried to lead away a horse and two donkeys belonging to the buyer’s uncle (their idea of a fair price for Yang Shi). But

160

part two

this maneuver provoked a brawl in which Liangru ended up killing one of the buyer’s cousins.32 A similar scenario unfolded in a Ba County case but ended with a negotiated settlement instead of violence. In the spring of 1813, peasant Wang Chaoyuan sold his wife, Feng Shi, because of poverty, for a brideprice of just 1500 cash. A few weeks later, however, Wang and his father Wang Longyong approached the buyer, Li Wenyu, and demanded more money. Wang Longyong explained that his son had acted foolishly and without permission, and that 1500 cash was a ridiculously low price. Unless Li Wenyu agreed to pay a much larger sum, Longyong would take him to court. In the end, Li agreed to make a supplementary payment of 11,200 cash—more than seven times the original price—in exchange for Longyong and Chaoyuan’s written pledge to cause him no further trouble. The Wang family’s maneuver may have been disingenuous, with the initial low price being designed to secure a buyer who would later be an easy target for extortion. To be sure, 1500 cash was a low enough price to provoke genuine indignation, but, in the end, the Wangs got a good deal, because the total of 12,700 cash that Li Wenyu ended up paying was well over the average wife sale price in Ba County. At any rate, the outcome suggests that even Li Wenyu recognized that 1500 cash was far from fair.33 Famine provided the context for some very low prices, because emergency sales of wives and children glutted the market, even as grain prices soared.34 For example, in the fall of 1751, the harvest failed in Dongyang County, Zhejiang; by winter, peasant Wang Yongsheng and his family had run out of food. Joining other famine refugees, Wang and his wife, Ge Shi, hit the road and walked to Jinyun County (about seventy-five kilometers to the south), carrying their baby daughter and begging in order to survive.35 But they were unable to obtain enough food, the family began to starve, and Wang fell seriously ill. As a witness later recalled, “there really was no way for them to survive.” Finally, just after the New Year, Wang arranged to sell his wife and daughter to one Shi Yiren, a peasant in Jinyun County whose own wife had died. A fellow famine refugee acted as matchmaker (for free), and Shi paid Wang a brideprice of just 2 taels. After the sale, Wang managed to walk home to Dongyang County, but just a few weeks later he died. Two taels (the equivalent of about 1500 cash at that time) was a low price, especially considering that it included the little girl as well as the wife. To put this sum in rough perspective: in Ba County during the Qianlong era, one shi of unhusked rice (about three or four months of subsistence for one person) in normal times cost about 1100 cash; but, in a famine, the price could spike to seven or eight times that much. Under the circumstances, Wang’s 2 taels may have bought no more than a few weeks’ worth of food. He did, however, secure the immediate survival of his wife and child, and that was probably his main goal in selling them.36 We find an even lower sale price in a highly unusual Ba County case from 1827, in which peasant Zhang Er sold his young wife, Zhu Shi (who was about 20 sui),

Chapter 5. Analysis of Prices in Wife Sales

161

because of poverty. She had been an adopted daughter-in-law, and their marriage had been consummated for less than two years when the sale took place. Zhang had come down with an unspecified illness, and the couple ran out of resources; Zhang’s father was dead, his mother had remarried, and he had no brothers or other family who could help. As he later explained, “I was sick for a long time and had no money for medical treatment, and we had no way to supply our daily food. My wife was suffering from hunger and couldn’t take it. Since I couldn’t support her, I just couldn’t bear to let her suffer anymore (xinli buren).” So, with Zhu Shi’s uncle acting as matchmaker, Zhang arranged to sell her in marriage to one Liu Gui for a purely token price of 200 cash, the lowest wife sale price in my entire sample of cases from all sources. In Daoguang-era Ba County, 200 cash amounted to just three or four days’ wages for casual short-term agricultural labor—nowhere near enough money to solve Zhang’s problems.37 In Zhang’s view (shared by the matchmaker and the buyer), this paltry sum did not really count as a brideprice at all, but simply as “tea and sweets money” (cha guo qian)— that is, a small gratuity. He had acted out of pity for his wife, not self-interest: his goal in letting her remarry was “to release her, so she could find a way to survive” (fang ta shenglu). By asking for so little money, he hoped to speed up the transaction, and perhaps to improve Zhu Shi’s chances of good treatment in her new household (since she would not have been “bought in”).38 The High End of the Market The high end of the price scale represented a very different kind of market, operating according to a very different logic. The typical seller at the high end was not in such a desperate hurry, and high prices were the result of protracted negotiations and competition from rival bids. These often took place in urban settings (such as Chongqing) and involved longer distances than usual. Therefore, although the sellers in high-end cases were usually poor, we find a fair number in which the impetus to sell came not from poverty per se but from marital incompatibility. In most wife sales, a woman’s looks seem to have been irrelevant, the main criteria considered by typical buyers being a woman’s age, record of childbearing, and health, as indices of her likely fecundity and ability to work. In contrast, the women in high-end transactions tended to be younger and (apparently) more physically attractive than most. Most of the highest prices in my sample were paid by men who bought other men’s wives for use as concubines. Not every price paid for a concubine was high, but most of the really high prices in the entire sample were for women who would become concubines. Only a relatively prosperous man could contemplate acquiring a second woman, so it is not surprising that buyers of concubines in my sample include examination candidates, students at state schools, and holders of minor civil and military degrees by purchase, as well as military officers, merchants,

162

part two

shopkeepers, and one yamen clerk. These men already had wives who had failed to bear children, and the stated reason they wanted concubines was to get sons. (Often they justified themselves by claiming that their parents had ordered them to acquire concubines in order to secure their posterity.) In other words, these were not the typical buyers found in the mid-range majority of sales. We see an extreme example of this pattern in an 1863 case from Ba County, in which a Chongqing merchant agreed to pay a whopping 130 taels for a teenage girl whom he sought to make his concubine. (Note that I do not include this price in Appendix C.2 because the transaction fell through.) The basic brideprice was to be 100 taels, but in addition the merchant agreed to pay two matchmakers a total fee of 30 taels. The girl in question was just 15 or 16 years old (17 sui), and it is clear from the case record that she was exceptionally attractive. Her first husband (who was twice her age) decided to sell her because she was unhappy and had repeatedly run away from home; according to her, she was unhappy because he had tried to force her into prostitution. The case came to court because the buyer wanted to make sure of his standing, so he and the woman’s husband petitioned the magistrate for permission to proceed with the transaction. In an emphatic rescript, the magistrate pointed out that “the crimes of buying and selling a divorce are explicitly prohibited by law,” and he denied their petition. After learning the magistrate’s response, the buyer canceled the deal.39 To put this sum in perspective, 130 taels is five times the average wife sale price in taels in my sample of Ba County cases; it is about twenty times the average annual wage in taels for agricultural labor in the early nineteenth century. One hundred and thirty taels approximates the gross annual rental income of a modest landlord,40 and that sum would have purchased a jiansheng (“Student of the Imperial Academy”) degree, which in the nineteenth century cost just over 100 taels.41 In other words, the men who could afford the high-end prices in our sample were indeed far better off than most. Their wealth could not compare, however, to that of the upper gentry, whose finances operated on an entirely different scale. For example, Chung-li Chang estimates the average annual income (from both official and unofficial sources) of a county magistrate in the Qing dynasty as 30,000 taels, and that of a governor or governor-general as six times that sum.42 But men from that rarefied social stratum presumably would not have stooped to buy another man’s wife, however attractive she might be—and if they had, they would have managed to stay out of court. High-end wife sales like this one suggest comparison with the high end of the slave trade in the antebellum United States: the “fancy trade” in beautiful, lightskinned girls and women bought for sexual exploitation. As Walter Johnson observes, in the New Orleans slave market the prices paid for such females “occasionally reached three hundred percent of the median prices paid in a given year” for slaves.43 Most slaves were purchased for economic exploitation. But “fancy”

Chapter 5. Analysis of Prices in Wife Sales

163

slaves were luxury goods, and the extraordinary prices paid for them represented ostentatious displays of wealth and power all the more potent for their flirtation with the boundaries of scandal. “For slave buyers, the bodies of light-skinned women and little girls embodied sexual desire and the luxury of being able to pay for its fulfillment.”44 A similar logic informed at least some wife sales in China. A variation of the high-end sale occurred when an attractive young woman was sold to a pimp who, while ostensibly taking her as his wife or concubine, in fact planned to put her to work as a prostitute.45 Another variation was when an infatuated customer bought a woman who was being pimped in this manner, to be his own concubine. These transactions, too, could involve very large sums of money; they represent just one of a number of areas of intersection between the market for wives and the market for sex. In 1871, physician Edward Henderson estimated the average price paid by brothel keepers in Shanghai for “a good-looking girl of from sixteen to twenty” to be two hundred silver dollars (even more if she could sing), but he also had personal knowledge of two wealthy men who had paid up to ten times that sum to buy unusually desirable women out of brothels to serve as their concubines. “Many rich Chinamen regard the payments of these large sums as public proofs of their superior wealth, and will shew (sic) wives so obtained with pride, boasting of the money they have cost.”46 The legal archives provide a number of examples of such high-end transactions. In a Ba County case from 1858, the Chongqing pimp Su Haishan sold his twentysomething concubine, Wang Shi, whom he was using as a prostitute, to one of her customers for a body price of more than 50 taels. This customer wanted to make Wang Shi his own concubine. Su Haishan had originally bought her for 35 taels from her first husband, Liu Xingfa. Liu was a migrant laborer from He Department (located about eighty kilometers north of Chongqing on the Jialing River), who had sold her because of poverty in the late winter of 1857, just before the New Year.47 We find an even more dramatic example in an “immediate examination” case from Beijing prosecuted by the Board of Punishment in 1906, near the end of the dynasty.48 This case involved a teenage girl from Datong County, Shanxi, who had circulated through a number of transactions and relationships over a period of six years before coming to official attention. She is described in the case record as very beautiful. Back in Shanxi, her widowed mother, Gao Zhang Shi, had sold her at the age of 13 sui to a man named Yang for 35 strings of cash, to serve as his concubine. (Note that 13 sui—i.e., 11 or 12 years old—was well under the average age of marriage for females, which came shortly after menarche at around 17 or 18 years old.)49 When she turned 14 sui, Yang began sleeping with her. After a year, however, Yang decided to divorce her back to her mother, because his attention to the concubine made his main wife extremely unhappy (in protest, she refused to wash or groom herself, and she often cursed and beat the girl).

164

part two

This divorce—an example of “compensated divorce” (see Chapter 8)—proved highly profitable for all concerned, except the girl herself. After she turned 16 sui, Gao Zhang Shi arranged to remarry her as wife to one Wang Chaoxian (a trader from Zhangjiakou, Zhili) for a brideprice of 90 taels. Out of this sum, Gao Zhang Shi paid 60 taels to Yang as compensation for the divorce, and she kept 30 for herself. Wang then took his new wife to Beijing, where he immediately resold her to one Bai Yichen as concubine for the astonishing body price of 240 taels. (This is the highest price in my entire sample of wife sales from all sources.) It had been Wang’s plan all along to resell the girl at a profit, and it had taken him less than a month to do so. His actions indicate an acute awareness of comparative prices: 90 taels was no doubt a great deal of money in Shanxi, but Wang knew that such a girl would fetch a far higher price in Beijing. Datong, Zhangjiakou, and Beijing were closely linked on one of the main trading routes used by Shanxi merchants, so it makes sense that Wang would have been familiar with all three places and would have had the opportunity to acquire this kind of knowledge about comparative prices.50 Bai gave his new concubine the work name “Little Cicada” (Xiaochan), and, with the help of a partner, he pimped her over the next three years at brothels in Beijing and Tianjin. When she realized what was happening, she protested and made a scene, but the two men whipped her into submission. When she was 19 sui, she managed to get word of her situation to her brother, who tracked her down in Beijing and filed charges with the Board of Punishment. The pimps were arrested and punished, and she was returned to her mother. We do not know what happened to “Little Cicada” after that, but it is safe to assume that her mother simply sold her in marriage yet another time.51 This story presents one of the most blatant examples of exploitation to be found in the context of wife selling. This girl circulated through multiple transactions between various people who sought to exploit her for sex and profit, showing that, in the right market conditions, an attractive young woman constituted a highly liquid asset. Her case illustrates several ways in which a woman could be sold: ordinary “marriage as sale” by her mother (to be Yang’s concubine, and later to be Wang’s wife); compensated divorce; open wife sale (by Wang to Bai); and finally prostitution (i.e., retail sale of sexual labor). Another interesting feature of the case is the dramatic increase in her market value as she matured and was transferred from small town Shanxi to the imperial capital: 35 strings of cash, 60 taels, 90 taels, and finally 240 taels. This last price was a big-city price paid by a pimp who expected to turn a handsome profit. The Broad Mid-Range of Prices in Nanbu County In my sample of wife sales from Nanbu County, Sichuan, we find neither the very high prices nor the very low prices occasionally found in xingke tiben and in cases from Ba County. For Nanbu County, I have forty-one prices for open wife sales

Chapter 5. Analysis of Prices in Wife Sales

165

denominated in cash, the average being 14,000 cash (see Appendix C.4). Only one sale in my entire Nanbu County sample, dating to 1884, was paid in taels: that price was 30 taels. These prices are in the same range as those found in Ba County and in xingke tiben (if one takes into account both tael and cash prices), so the commodity prices from those sources may provide a sense of real value. The lack of extreme prices, either high or low, and the fact that all but one price was denominated in cash, would seem to reflect Nanbu County’s relative lack of social and economic complexity. The contrast with Ba County is particularly clear. Ba County included both rural hinterland and the booming port city of Chongqing, which experienced explosive population growth over the course of the dynasty. Chongqing’s status as prefectural seat made it a center for official education and civil examinations, and its strategic location at the confluence of the Jialing and Yangzi rivers also made it a major focus of military logistics as well as civilian commerce. For all of these reasons, Ba County attracted large numbers of civil and military officials, their staff, gentry (and aspiring gentry), and merchants big and small—in other words, the very kinds of men who appear in the role of buyer in high-end wife sales. As a crucial node in a riverine network of transportation, Chongqing attracted people of all social classes—including the destitute rural migrants who might accept extremely low prices in wife sales. Whereas Ba County stood out for its social and economic complexity, Nanbu County was a relative backwater, and therefore a fairly typical place. In the Nanbu price sample we find, stripped of extremes, the broad mid-range of prices that seem to have characterized most wife sales in other places as well.52 T H E I M PAC T O F C H I L D C U S T O DY O N B R I D E P R IC E

When wives were sold, their children often accompanied them into new marriages. The basic reason is that sellers could not afford to support them or were otherwise incapable of providing proper care—as in the case of nursing infants, for example, who always accompanied their mothers. Outright Sale of Children Sometimes children were sold outright, along with their mother. If the seller were in a position to bargain, he would demand a higher total price than if he had sold the woman by herself. An 1868 case from Ba County illustrates the outright sale of a son along with his mother for a high brideprice. Zhou Xingtai had been married to Peng Shi for six years, and they had an infant son; Zhou’s elderly mother also lived with them. The family was very poor, and when Zhou fell ill and went blind, they ended up surviving on charity from relatives. Zhou, his mother, and Peng Shi agreed that their least bad option was for Zhou to sell Peng Shi and the baby, “to let them have a way to survive.”

166

part two

Through a matchmaker, Zhou arranged to sell them to a Chongqing man named Li Jingxiu (50), who was a yamen clerk. Li Jingxiu wanted to take Peng Shi as concubine and to adopt the baby boy as heir (sizi) because his own wife had no issue: as he later explained, “I had no son, so I bought one.” The contract makes explicit the baby’s permanent transfer to Li’s household: “My nursing son will follow his mother into the marriage and will become heir to Jingxiu’s lineage branch.” The total brideprice was 50 taels, about double the average price in taels found in Ba County. About a month after the sale, however, the baby died of smallpox. When Zhou heard this news, he and a cousin approached Li Jingxiu and demanded an additional 50 taels in compensation for the baby’s death. This demand suggests that the Zhou family may not have considered the baby’s transfer to be so absolute after all, regardless of what the contract said. In response, Li filed charges against Zhou, accusing him of extortion.53 Temporary Transfer of Children It was more common, however, to arrange the temporary transfer of children, to be raised by their mother and her new husband for a stipulated period of time before returning to their own father. This sort of arrangement was common in widow remarriage as well, the idea being that children so transferred would keep their own surname and eventually return to their father’s lineage. By thus protecting her dead husband’s line of descent, a widow might mitigate the betrayal that remarriage implied; indeed, some widows justified remarriage as the only realistic way to support their children and thereby secure the interests of the dead husband.54 In wife sales, the temporary transfer of children represented a compromise between the seller’s desire to secure his own line of descent and his inability to support them himself. The particular terms varied: in one case, a wife was pregnant when sold, and the buyer agreed to let her keep her baby (if it survived) for two years and then hand it over to the seller (but this requirement applied only if the infant were male);55 in a couple of cases, a baby was to be kept with its mother until weaned; in many cases, buyers agreed to support sellers’ children for a period of five or six years; and in others, buyers simply promised to raise the children until they had “grown up,” without specifying a time period. In every instance, if the seller wanted a guarantee that his children would keep his surname and eventually return to his custody, he had to defray the expense of child support by lowering the brideprice for the wife sale, or by allowing the buyer to retain some portion of the nominal price. Such compensation constituted a significant proportion of the brideprice—often as much as one-third or half of the woman’s full price if she were sold alone. A 1754 case from Fujian illustrates how the cost of child support might be factored into the calculation of brideprice. Shi Shimu (41) and his wife, Wang Shi (35),

Chapter 5. Analysis of Prices in Wife Sales

167

were landless peasants from Yongchun Independent Department who migrated to Youxi County with their 8-sui son, Azhi. Shi was unable to find enough work to support his family, and they were surviving by begging and sheltering at temples. By summer, the couple decided that this situation could not continue, so Shi negotiated to sell Wang Shi to Lin Shengzhi, a poor peasant of 39 sui who had never married and lived alone. Shi Shimu agreed to accept a low brideprice of 6 taels on condition that the couple’s son would accompany Wang Shi to be raised in Lin’s household for several years, after which he would return to his father. Shi Shimu wrote the contract himself. There was a delay, however, because Lin did not have enough money on hand to close the deal, and, after a couple of weeks, Shi decided to change the terms. He canceled the provision giving temporary custody of his son to Lin, and he raised the brideprice to 10 taels. Lin agreed to these changes, so Shi destroyed the original contract and wrote a new one. A temple curate witnessed the contract, but in fact Shi and Lin simply negotiated the terms face to face. The case record does not say exactly how long the couple’s son was to stay in the Lin household under the terms of the first contract. But, clearly, the initial brideprice was set artificially low in order to compensate the buyer for the expense of raising the boy. In this case, we can assign an exact figure to that compensation— 4 taels—because that is the amount added to the brideprice when Shi decided that his son would not follow his mother into Lin’s household after all.56 We find similar calculations in many other cases from across China. In an 1848 case from Ba County, Liu Hongwan (27) sold his wife, Mou Shi, because of poverty, to one Mu Shuzhai, who was proprietor of a little paper shop in Chongqing. The couple had a daughter and a son, and, in exchange for a low brideprice of 5000 cash, Mu agreed that “Mu Shi would bring with her a son to raise in my household.” No provision was made for the daughter, suggesting that Liu planned to sell her separately. The crude, barely literate wife sale contract, which survives in the case file, includes this stipulation: “Mou Shi will bring with her a son Liu Nian, who will follow his mother until he grows up. If Liu Nian falls ill with smallpox fever, then that is not the fault of the Mu family, and the Liu family may not complain.” After finalizing the transaction, however, Mu suddenly balked, and he refused to take the boy: “When I thought about how poor I was, I decided to deny that I had promised to let her bring her son into my household.” But Liu Hongwan could not afford to support his son either, so he demanded that Mu either take the boy as agreed or pay an additional 5000 cash for the woman’s brideprice. When Mu refused to pay, Liu’s widowed mother filed charges at the county yamen, accusing Mu and Mou Shi’s natal family of forcing her son to sell his wife and cheating him out of his money.57 As in the previous case, it is obvious that the seller had agreed to an artificially low brideprice—apparently just half of what he believed his wife was worth—in

168

part two

order to compensate the buyer for raising their son. In the seller’s eyes, the buyer’s refusal to take in the boy required a refund of that compensation, which constituted the balance of the woman’s full price. A case from Duchang County, Jiangxi, highlights the cost of child support from a different angle. In 1780, when poverty and illness forced Bao Nianshi to sell his wife, Xiang Shi, he negotiated for his three small sons to accompany her into her new marriage temporarily. The case record does not specify exactly how long they were to stay with their mother, but the arrangement was intended to last for several years at least. The buyer, Wu Lingyou, agreed to a nominal brideprice of 28 taels, but he was allowed to retain 10 taels to cover the expense of raising the boys, so in fact he paid only 18 taels. In this case, unusually, buyer and seller lived in the same village, not far from each other. The three boys would often visit their father and spend the night, and soon they were spending more time at their father’s home than at their mother’s. Finally, Bao told Wu that, under the circumstances, he was willing to take his sons back, but he asked Wu to remit the 10 taels retained for child support. Wu refused to pay, stating that it was not his fault that the boys wanted to return to their father, and no matter what they did, the original terms of the contract should hold. Bao had no choice but to return his sons to their mother, because without the 10 taels, he simply could not afford to keep them.58 Something similar occurred in a 1736 case from Xuyong Independent Department, Sichuan. When illness and poverty compelled peasant Zhao Ying (36) to sell his wife to a more prosperous fellow villager named Chen Wenzhang as concubine, the contract provided that Chen would raise Zhao’s two sons to adulthood while allowing them to keep their father’s surname and remain in his lineage. The brideprice of 20 taels was calculated to take this provision into account. But the two boys missed their father very much, and after a short time they returned to his household. Then Zhao asked Chen to compensate him for “child support expenses” and often visited with the boys to press this demand. Chen finally asked the matchmaker from the original sale to negotiate a new contract, in which Chen agreed to pay Zhao Ying an additional 6 taels. In return, Zhao promised to sever contact with Chen and his ex-wife.59 Sometimes a seller’s effort to arrange child support could be a deal breaker. In a 1736 case from Taiyuan County, Shanxi, peasant Chen San was leaning toward paying a brideprice of 6 taels for the wife of peasant Li Tong, but then Li added the condition that Chen must raise Li’s three children. Six taels seems like a low price, but the added burden of three children (whom Chen would eventually have to return to Li) made the proposal untenable for Chen. Li Tong was unwilling to sell his wife without provision for his children, and, as a result, the deal fell through.60 Efforts to reclaim children after the stipulated period sometimes provoked further disputes over money. We see this scenario in a case from 1798. Li Huailiang

Chapter 5. Analysis of Prices in Wife Sales

169

(32) and his wife, Wang Shi (25), were landless peasants from Yibin County, Sichuan, who had migrated to Yongshan County, Yunnan, in search of a livelihood. Li tried to make a living by hunting, with but little success. Finally, they negotiated to sell Wang Shi to a local man named Yang Gui (31) for a brideprice of 5 taels. The parties agreed that the couple’s infant son of 1 sui should accompany his mother until weaned. After a year, Li decided that his son should be old enough to be weaned, and he visited Yang Gui’s home to take custody of the boy. To Li’s surprise, Yang demanded money: “I’ve supported your son for a year. How can you expect to take him away just like that, for free?” Li retorted that the original agreement did not provide for any extra compensation and refused to pay; indignant, he grabbed Yang and dragged him into the street, intending to take him to the authorities. The two men fought, and Li ended up killing Yang.61 The xingke tiben reporting this case does not include a transcription of the contract, so we cannot determine exactly what terms were agreed in the original negotiations. But 5 taels would seem to be a very low price for a 25-sui woman of proven fertility, and Li clearly believed that he owed her buyer nothing. Yang’s demand for compensation may have been a clumsy attempt to keep the boy. Occasionally, care for an elderly dependent would be subsidized in the same manner seen here for children. In a fraudulent sale transacted in 1745 in Huoqiu County, Anhui, peasant Li Si (47) sold his wife, Shao Shi (33), because of poverty. No children were involved, but a condition for Shao Shi’s cooperation was that her elderly mother, Xu Shi, accompany her into the new marriage. The buyer, who believed he was taking a widow in marriage, accepted these terms, but to compensate for Xu Shi’s support, and with Li Si’s agreement, he retained 5 taels out of the nominal brideprice of 22 taels.62 In widow remarriage, too, if the first husband’s children were to retain his surname and some day return to his lineage, the cost of their care would have to be covered one way or another. In Anhui, for example, when a widow took children into her new marriage, the in-laws from her first marriage would have to provide her new husband with a separate written contract guaranteeing to cover the cost of child support. Provision of this “contract for providing meals” was the condition for allowing her children to retain their father’s surname.63 WA S D OW RY A FAC T O R I N W I F E S A L E S ?

Given that some Chinese brides took dowry into their marriages, it is worth asking whether dowry played any role in wife sales. Did these women have dowry from their natal families that they had brought into their original marriages, and then took with them into their new households when sold? If so, did the brideprice paid by the buyer represent compensation for assets that the wife was removing from

170

part two

her first husband’s household (instead of money paid for her person)? Also, did sold wives themselves receive all or part of the brideprice paid for them? (In normative major marriage with dowry, the nominal brideprice often was used by the natal family to buy goods for dowry and in this way would return with the bride to her husband’s family.) In other words, were the transactions I am calling “wife sales” actually something other than sales of persons? To summarize, dowry was not a factor in wife sales, with only rare exceptions. These exceptions are worth exploring, however, for the indirect way they illuminate the economic logic of wife sales—which were, in fact, sales of persons. Evidence in Contracts In my entire sample of central and local legal cases from Qing archives, I have 125 contracts for wife sales of one kind or another, and only one mentions dowry having been brought into the original marriage by the wife. But this 1849 contract from Ba County documents something other than a standard wife sale. This “contract for marrying off my wife, in obedience to my mother’s orders” is unique in several respects. (The case file contains no other documents, so we know little about its context.) Like a wife sale contract, it is written in the name of the husband, Ma Mingzhou. According to the text, Ma has been married to Zhou Shi for seven years, but because she “was not raised properly,” she has turned out to be a bad wife; her own father has reproved her several times, to no avail. Since Zhou Shi will not reform, Ma’s mother has ordered Ma to “marry her off to a different household,” and he has arranged through matchmaker Xie Zongwen to marry her to one Shi Guilu, who lives nearby, as wife. In exchange for his wife, Ma has received an unspecified quantity of cloth from Shi Guilu. The contract then states: “The Zhou family’s original dowry will be transferred in its entirety to the Shi family.” There follow the usual guarantees that Ma is acting of his own volition and will not change his mind, and that if any trouble develops it will be Ma’s responsibility. The contract closes by listing Ma’s maternal uncle as a guarantor, along with five witnesses. Superficially, this transaction resembles an ordinary wife sale, and confiscation of the contract (which explains its survival in the archive) shows that it was ruled illegal (no doubt because the first husband negotiated to marry his wife directly to the second husband without first divorcing her back to her natal family).64 Nevertheless, this contract is unique in my sample both for mentioning a previous dowry and for specifying that compensation has been paid in kind (rather than money), without giving the cash value of either; these features alert us that this transaction was no ordinary sale. In fact, it should probably not be considered a sale at all. We do not know the contents or value of the original dowry, but since it was still available after seven years to be transferred to Zhou Shi’s new husband, we can assume that it consisted of clothing or other durable goods, rather than consuma-

Chapter 5. Analysis of Prices in Wife Sales

171

bles such as food. Nor do we know the value of the cloth that Shi gave Ma in lieu of brideprice. But poverty was not the motive for this exchange, and it is highly unlikely that Ma gained a profit. Indeed, the contract’s use of the term “shuili” (literally “gift of water”), which refers to gifts in kind—as opposed to the usual “caili” (“gift of money”) or “shenjia” (“body price”), both of which refer to money— reinforces the implication that this transaction should not be seen as a sale. The cloth Ma received may have represented compensation for some amount his family originally paid Zhang Shi’s family in brideprice or contributed to their conjugal fund (i.e., indirect dowry), but, if that were the case, then the contract would probably say so. It seems more likely that the cloth had merely token value. Be that as it may, as far as Qing law was concerned, the direct transfer of the woman from one husband to the next, in exchange for any form of compensation, would have marked this transaction as the crime of “selling a divorce.”65 The fact that this woman had brought a dowry into her first marriage and would take it with her into her second marks her as a woman who could not really be sold; it also marks the second transaction as something other than a real sale. By contrast, the absence of any mention of dowry in any other wife sale contract in my sample suggests that those women had been married without it, and therefore could be sold. (One caveat: truly desperate people, such as famine refugees, could not afford to quibble. Under those circumstances, a wife would have to be sold simply for the sake of survival, regardless of whether she had originally come with dowry.) We do not know why this case ended up in court, but one clue is that the contract does not bear the name of anyone from Zhou Shi’s natal family as witness (despite the reference to the woman’s father reprimanding her). Therefore, it seems likely that the natal family was not consulted in the transaction and later filed charges in protest.66 Evidence in Legal Cases Not counting the Ba County contract discussed above, I have found only one wifeselling case in the archives (out of a total sample 651 cases) that records dowry having been brought by the wife into her original marriage. This reference to dowry appears in an 1802 case from Linshui County, Sichuan, in which peasant Jiang Panwan (39) sold his wife, Huang Shi (39), for a low brideprice of 3000 cash to be the second wife of widower Gan Yilin. The couple had been very poor and the decision was mutual: Jiang told Huang Shi that she should “remarry” (gaijia) so that they could “both secure a livelihood by going their separate ways,” and she agreed. The transaction was negotiated in a straightforward manner, apparently without acrimony. Six days later, however, Huang Shi returned to her first husband’s home with a demand. As Jiang later recalled, “Huang Shi said that when she first married me,

172

part two

her natal family had contributed clothing and accessories as dowry, but now that she had remarried, there were only two worn-out pieces of clothing left, and she demanded that I give her the brideprice money so she could make clothing.” Jiang was angered by her demand and beat her up; she later died of her injuries. Significantly, Huang Shi made this demand on her own initiative, not her new husband’s: she visited Jiang’s home without telling Gan Yilin, who found out only after her death that she had demanded her brideprice from Jiang. While we do not have a transcript of the sale contract, there is no evidence that the terms in any way diverged from the normal form of wife sales, in which the buyer paid cash in exchange for woman and contract. We do not know the exact terms of Huang Shi’s first marriage either, but it seems clear that she saw the dowry she brought into it as her own property, which she had the right to take into her new marriage and. if it had been used up, then she had the right to compensation from her first husband. Evidently she believed her dowry to have been worth at least the 3000 cash for which Jiang sold her. Put another way, Huang Shi’s demand amounted to an assertion that since Jiang had not bought her from her parents in the first place (because she had brought dowry), he had no right to sell her either, and so her brideprice belonged to her as compensation for her original dowry. But that principle had not been considered when the wife sale was negotiated, and clearly Jiang did not accept her point of view.67 We find an attitude similar to Huang Shi’s in another Sichuan case, this one from Jiajiang County and dated 1777, in which Xu Luyou and his father, Xu Deying, sold Luyou’s wife, Nie Shi, in a fraudulent transaction for 15 taels. Nie Shi had been 20 sui at the time of her first marriage, and there is no evidence that she came with a dowry. The witnesses agreed, however, that Nie Shi’s natal family had received no brideprice from the groom’s family, either; in other words, her parents did not sell her. This marriage seems to have been an example of what Hill Gates calls “no exchange” (i.e., neither brideprice nor dowry).68 Nie Shi’s subsequent sale by her husband took place after just one year of marriage, right after the New Year. The exact motive for her sale is not clear: her husband and parents-in-law claimed that she was “disobedient,” but she and her natal family insisted that she “never did anything wrong.” The sale’s timing implies a need to pay debts. Her in-laws did not bother to consult her natal family, and, when the Nies found out that she had been married off, they were outraged. A gang of nine Nie family men and their friends tracked her down and brought her back to her natal home (her buyer, who had been told she was a widow, did not resist). But when Nie Shi informed them that the Xu family had sold her for a brideprice of 15 taels, the Nie family men “became even angrier.” They proceeded to the Xu family’s home, planning to “beat them and make a scene in order to vent our anger,” after which they would take Xu Luyou and his father to the yamen for prosecution. They forced their way into the Xu family home, beat up both men (one of whom died), smashed things, and stole clothing and bedding.69

Chapter 5. Analysis of Prices in Wife Sales

173

The key feature of this case is the natal family’s indignation at Nie Shi’s sale, which stemmed in large part from the fact that they had not sold her in the first place. In other words, her natal family felt strongly that her in-laws had no right to sell her for brideprice because they had paid no brideprice for her when she married into their family. This linking of brideprice in wife sale to the original brideprice paid for a woman reflects the logic of compensated divorce, in which the compensation paid to the first husband was often explicitly calculated as a refund of all or part of the brideprice he had paid the natal family to marry her in the first place (see Chapter 8). As a corollary, lack of brideprice seems to have implied lack of entitlement to such compensation.70 Cases in which a Share of Brideprice Was Set Aside for the Wife In a handful of cases, part of the brideprice was set aside for the wife herself. What is the significance of this practice? Such provision was rare: I find just nine examples in my entire sample of more than 650 wife-selling cases from all sources (six in xingke tiben, two in cases from Baodi County, and one in an “immediate examination” case from Beijing). Five cases involved open sales, and four involved fraudulent sales (see Table 3). Since set-asides of this sort were so rare, it would seem unwise to attribute much significance to them. It is notable, however, that in every case for which we have this information, the stated purpose of the wife’s share was to pay for clothing (either to buy material to make clothes or to redeem clothing she had pawned). None of the case records call this money “dowry,” nor do any mention that dowry was taken into the women’s first marriages. But given the frequent use of cloth and clothing (often made by the bride herself) by rural people as a modest dowry, it is possible that these payments represent compensation to the women for clothing they had brought into their first marriages.71 It is also possible that the payments were made simply to secure the women’s cooperation. As Table 3 shows, the wife’s share ranged anywhere from 2 percent to 32 percent of the total brideprice, and the circumstances of these payments also varied dramatically. Sometimes the woman was paid directly, after the transaction had been finalized. In both the 1823 case from Baodi County and the 1884 one from Beijing, when each wife was about to depart for her new household, the seller himself gave a portion of the brideprice directly to her. These payments in these two cases were substantial, amounting to 21 percent and 20 percent, respectively, of the total brideprice.72 In neither case had such payment been stipulated by contract, but it seems likely that the wives had been promised these amounts to secure their cooperation. In the 1823 case, we know from testimony that the wife had bitterly resented her husband’s plan to sell her and agreed only reluctantly. The same logic may explain the wives’ shares in the four fraudulent sales, which could not have succeeded without their complicity. In the fraudulent sale that took place in

174

part two table 3 Sales in Which a Share of Brideprice Was Set Aside for the Wife Open Sales

Date

Place

1748

Shangrao Cty, 24 taels Jiangxi Ningdu Ind. Dept., 28,000 cash Jiangxi Baodi Cty 70 strings “east cash”*** Baodi Cty 90 strings “east cash” Beijing 80 strings “east cash”

1759 1823 1866 1884

Total brideprice

Wife’s share

Percentage

Stated purpose

4 taels*

17%

to make clothes

9000 cash*

32%

to make clothes

15 strings**

21%

to make clothes

3 strings 16 strings**

3.3% 20%

no comment to redeem pawned clothes

Fraudulent Sales Date

Place

Total brideprice

Wife’s share

1738 1754

Guangji Cty, Hubei Xiangyang Cty, Hubei Shanghai Cty, Jiangsu (same woman sold twice) Shidai Cty, Anhui

28 taels 30,000 cash

1755

1757

Percentage

Stated purpose

1 tael* 3500 cash

3.5% 12%

to make clothes to make clothes

24 taels

2 taels**

8%

to make clothes

16 taels

0.3 tael**

2%

no comment

18 strings cash

2 strings**

11%

to make clothes

* Buyer retains this amount from sale price, ostensibly for the woman. ** This amount is paid directly to the woman by seller or matchmaker after sale is finalized. *** “East cash” (dong qian) or “small cash” (xiao qian) was one of several non-standard forms circulating in the nineteenth century; it converted to standard cash at a rate of about 1000:160, so 70 strings would have been worth 11,200 standard cash (assuming that one string = 1000 coins). See King, 1965: 60–62.

Shidai County, Anhui, in 1757, the woman made a point of demanding two strings of cash from the matchmaker before she joined her new husband. If she had made a scene, the transaction would have fallen through, so it is not surprising that the matchmaker agreed to pay her before delivering the balance of the money to the seller.73 In at least three of these cases, however, the woman did not actually receive any payment. Instead, the contract stipulated that the buyer would retain a portion of the brideprice to pay for her clothing: in practice, he would pay the seller that much less. In the 1748 example from Shangrao County, the nominal brideprice was 24 taels, of which the buyer retained 10 taels. Of the latter amount, 4 taels were designated for the woman’s clothes and 6 were for child support (because she had brought her son by the seller into the new marriage to raise temporarily). But the

Chapter 5. Analysis of Prices in Wife Sales

175

woman later testified that she had received no money herself and that the buyer never spent any on her behalf.74 In the 1759 case from Ningdu Independent Department, the nominal brideprice was 28,000 cash, but the buyer actually paid the seller only 19,000, retaining the balance ostensibly for the wife’s clothes. But there is no evidence that she received any money.75 Occasionally the natal family of a sold wife received a share of her brideprice too. But such payments were also rare, and there is no evidence that they represented compensation for dowry that had been brought into the first marriage. Instead, when the natal family was paid, the purpose was to secure their cooperation. Dispute Settlements that Mandated Dowry Two unusual cases from Ba County shed light on the specific meaning of dowry as marking the opposite of sale. In each case, the settlement—one mediated privately and approved by the magistrate, the other imposed by court judgment—included payment of dowry for the woman to take into a new marriage, as a means of counteracting the stigma of what had happened to her and ensuring her a fresh start. In the first case, from 1864, Zhou Zongli suddenly married off his wife, Lai Xin’gu, after three years of unhappy marriage. This transaction does not appear to have been a wife sale. According to Zhou and his father, Xin’gu was disobedient and had “failed to keep the way of a wife” (bu shou fudao), although the Lai family denied these accusations and held her to be blameless. Zhou’s father had felt sufficiently upset by the couple’s discord that he forced his son and daughter-in-law to move out and live separately. Zhou Zongli blamed Xin’gu for this humiliating breach with his father, and in response he married her off to a carpenter named Hu Er as wife. Zhou Zongli had not bothered to consult either his own family or hers before taking this action, and when the Lai family found out what he had done, they fetched Xin’gu back to her natal home and threatened legal action. Mediators negotiated a settlement whereby, in exchange for avoiding prosecution, the Zhou family agreed to cancel Xin’gu’s marriage to Hu, yield all claim to her, and pay the Lai family 18,000 cash to finance a dowry for her to remarry respectably to someone else. The money was paid over before witnesses, and the settlement documented in a written contract, with one copy for each family; the Zhou family’s copy survives in the case file. This settlement did not satisfy Xin’gu’s uncles, however. They vented their anger by attempting to trash the home of Zhou Zongli’s father, provoking a brawl, and for this reason, it was the Zhou family in the end that filed charges. In judgment, the magistrate ordered both families to abide by the terms of the mediated settlement. In addition, he ordered Xin’gu’s grandfather (the head of her natal household) slapped for failing to restrain his sons from violence.

176

part two

A key question in this case is whether Hu Er paid any brideprice to Zhou Zhongli for Xin’gu. In his plaint, Xin’gu’s grandfather accused Zongli of receiving 20,000 cash for her, thereby portraying the transaction unambiguously as an illegal wife sale, in order to cast Zongli in as negative a light as possible. But Zongli denied receiving any payment, and in the formal court hearing no witness mentioned a brideprice or even used the word “sell” (mai) to refer to this transaction. Moreover, poverty was not Zongli’s motive to get rid of his wife, and the settlement makes it obvious that the Zhou family was not poor—after all, they had been willing to yield the woman and pay 18,000 cash in order to avoid going to court. It appears that Zongli told the truth: he simply gave his wife to Hu Er for free. Of course, to dispose of her in this manner (instead of returning her to her natal family) was a profound, calculated insult, and that is what so angered her family. (Note that, since Xin’gu had actually entered Hu Er’s household, people would assume that she had had sex with him.) The purpose of the 18,000 cash dowry was to compensate for this insult and to provide Xin’gu with an honorable fresh start in a third marriage—that is, to help overcome whatever stigma might attach to her for having been discarded so unceremoniously, which might obstruct her natal family’s effort to find her a new marriage. For a husband to pay compensation when divorcing his wife was the opposite of what usually happened. Typically, divorce required a woman’s natal family to pay compensation to her husband, the amount often calculated as a refund of the brideprice he had originally paid. For Zhou Zongli to finance a dowry for Xin’gu’s remarriage implies that, at the very least, he had not bought her from her parents, and she may even have brought some dowry with her when she married him. But the contract for the mediated settlement does not designate the 18,000 cash a refund of previous dowry. Instead, it is portrayed as an indemnity to the Lai family for the insult to their daughter, to persuade them not to press charges.76 In the second case of mandated dowry, which dates from 1832, peasant Lai Rongfa sold his wife, Cao Shi (an adopted daughter-in-law who was just 16 sui), because of poverty for a brideprice of 6000 cash. The buyer was Wu Fangji, a man of more than 50 sui whose 40-something wife had no issue, and who (“on my father’s orders”) sought to take a concubine to continue his family line. The transaction itself concluded amicably, but, although Cao Shi seems not to have minded leaving her first husband, she was very unhappy with her second, and after a couple of weeks she ran away to the home of her maternal uncle Hu Zaiying (who was her closest living natal relative). Hu then filed charges against Lai Rongfa and Wu Fangji at the Ba County court. The magistrate ruled the transaction to be an illegal wife sale, and he imposed an unusual and creative settlement. Since Cao Shi was unhappy with the new marriage, he returned her to her natal family, in the person of her uncle, and ordered the latter to arrange a proper new marriage for her. Her first husband Lai Rongfa

Chapter 5. Analysis of Prices in Wife Sales

177

could not refund the brideprice he had received (having spent it already), so the magistrate ordered him beaten, along with the two matchmakers (friends of Lai’s who had acted for free). The magistrate then ordered Cao Shi’s uncle Hu Zaiying to compensate Wu for the brideprice in Lai’s place—the logic seeming to be that, since Hu got the woman, he should pay the price. But in return for not being beaten (as mandated by statute), Wu “volunteered” to pay the entire sum back to Hu Zaiying, so that Hu could buy eight bolts of cloth as dowry for his niece’s new marriage. This odd transaction—Hu refunding the buyer’s money, and then getting it right back—took place in court under the magistrate’s supervision. In the affidavits (jiezhuang) that conclude the case file, Hu Zaiying pledged to marry Cao Shi off properly with a dowry of eight bolts of cloth, and he confirmed that he was “not permitted to accept any brideprice money” for her. Obviously, this magistrate assumed that the default mode of marriage for these people was sale, and his judgment underscores the fundamental similarity between (illegal) wife sale by a woman’s husband, and (legal, if stigmatized) marriage sale by a woman’s natal family. In this context, the mandated provision of dowry stands out as an extraordinary act of grace that reversed the usual economic logic of marriage. The magistrate’s ingenious alchemy converted the illegal brideprice from the wife sale into a dowry, guaranteeing that, the next time Cao Shi married, at least she would not be sold. The way the magistrate guaranteed that outcome implies a cynical appraisal of the motives of Cao Shi’s uncle, who came to court posing as her protector—but who obviously stood to profit by taking custody of her, if he had a chance to sell her off in marriage. In effect, the magistrate called his bluff: if Hu was really so concerned about his niece’s welfare, then he should prove it by putting up the money for her dowry.77 T H E E C O N OM IC L O G IC O F W I F E S A L E S

This chapter has documented several significant aspects of wife sales in Qing dynasty China. Most important, almost all of these transactions really were sales, in which buyers paid considerable sums of money: on average the rough equivalent of at least two or three years’ survival grain or agricultural wages, or two or three draft animals. Nevertheless, wife sales provided a relatively inexpensive means of acquiring a wife, because several factors tended to reduce prices: stigma and illegality depressed prices, to be sure, but more important were the inherent risks in marrying a woman whose first husband was still alive, along with the desperation that forced most sellers to accept the first offer received. Very different sorts of logic framed the extreme ends of the price scale: the low end took the “fire sale” logic of most wife sales to its farthest degree, whereas the high end was a luxury market for concubines and prostitutes not unlike the notorious “fancy

178

part two

trade” in American slavery. Finally, since children often accompanied sold wives into their new households, the terms of custody and the cost of child support had a major effect on prices. Major marriage with dowry was the opposite of sale: it was the normative ideal that conferred honor on the bride and her natal family, a status symbol associated with prosperity.78 Like other status symbols, I submit, its utility derived from the fact that it was out of reach for the poor majority. Dowry served to proclaim that “we are moral enough—and rich enough—not to sell our daughters.” Among the poor majority—certainly among the kind of people usually involved in wife sales—a dowry of real value was a rare exception. In this milieu, it appears that most marriages constituted sales of some kind: brideprice-heavy marriage, the sale of daughters to become concubines, widow remarriage, compensated divorce, and direct wife sale were all variations on a theme. The exceptional references to dowry described in the final section of the chapter illuminate by contrast the economic logic of wife sales. They underscore the fundamental similarity between wife sales and other forms of marriage that also constituted sales of some kind.

6

Negotiations between Men over Wife Sales SELLING WIVES AND THEN DEMANDING SUPPLEMENTARY PAYMENTS IS STRICTLY PROHIBITED (yan jin mai qi zhao jia): . . . Here in Jiangxi this is customary practice. It happens when shameless husbands and wives are pressed by poverty. The wife will mercilessly nag her husband, casually abandoning all regard for him; and the husband will ignore propriety, thinking only of the money he can get by selling her. . . . Then, as soon as they have cast aside their marriage, the first husband will seize on pretexts to extort supplementary payments (jie duan suo zhao) from the new husband. If the first husband gets even slightly less than he demands, he will file false charges of “illicit sexual relations” or “abduction” against the new husband, thereby disturbing and encumbering the courts. There is simply no end to false charges of this kind! Such evil customs (exi) make one’s hair stand on end in outrage! — from a public proclamation issued in the early qianlong era by the provincial judge of jiangxi 1

The focus of this chapter is the negotiations between men that framed wife sales. These negotiations occurred in two phases. The first, which preceded the transaction, involved measures to reassure prospective buyers by minimizing their risk of being subjected to extortion or ending up in court. First-phase negotiations had the character of self-fulfilling prophecy, however, because many sales were followed by a second phase of negotiations prompted by the seller’s demands for more money. The measures taken in the first phase aimed to prevent the second from taking place, but often they failed. There was widespread awareness that most wife sales were illegal and that the parties to these transactions risked punishment if they ended up in court. All negotiations proceeded in the context of this knowledge, which influenced calculations on both sides: to borrow Robert Mnookin and Lewis Kornhauser’s famous metaphor, the parties bargained “in the shadow of the law.”2 Indeed, when such 179

180

part two

cases ended up in court and received formal judgment, the most common outcome was that the second marriage would be canceled and one or more of the parties would receive a beating. But those risks were not evenly shared, because after a sale had been concluded, the buyer had far more to lose than the seller: the buyer stood to lose his new wife and the money he had paid, whereas the seller (after spending the brideprice) risked only a beating. This unequal distribution of risk made buyers vulnerable to extortion. There was also widespread awareness that men who had sold wives might demand more money from their buyers. If the buyers refused to pay, sellers might well threaten to file charges against them—false charges of kidnapping, rape, and other heinous crimes that carried penalties far more severe than those the Qing Code mandated for “buying or selling a divorce.” In fact, this was the single most common reason that wife sales ended up in court. Here was the “extortion” so dreaded by prospective buyers. But when considered from the seller’s point of view, such demands reveal something more than just crass opportunism: they reveal a close parallel between peasant men’s attitudes towards wives and towards land—and, specifically, between wife sales and land sales, which also often resulted in demands for “supplementary payments” (zhao jia). Like men who sold land, men who sold wives did not necessarily believe that they had given up all claim to those women. On the contrary, many acted as if selling a wife did not sever one’s relationship with her so much as it created a new relationship with her buyer: a relationship between two husbands in which the second had an obligation to help the first. (Here we can discern a clear conceptual link between wife sale and polyandry.) Given the illegality of wife sales and the likelihood of extortion, buying “a wife with a living husband” was widely understood to be a risky way to marry. The American missionary Arthur Smith, writing at the end of the nineteenth century, observed that “it is proverbial that a judicious man will never marry a woman who has a living husband, for the sufficient reason that he never can foresee the consequences, which are often serious.”3 Elsewhere, Smith noted that wife sales resulted in “perpetual brawls and lawsuits.”4 Understandably, many who sought wives were nevertheless wary of such transactions—hence the measures that had to be taken to reassure them. During the negotiations leading up to a sale, the man trying to sell his wife was at a disadvantage. As we have seen, these men were poor and many were desperate: they usually had to accept the first offer received. In addition, the well-known risks of wife sale increased the downward pressure on prices, with the lower cost of this form of marriage helping offset the higher risk. As a result, men who had sold wives often expressed bitter resentment at how little they had been paid, and this resentment only strengthened the sense of entitlement behind their subsequent demands.

Chapter 6. Negotiations between Men in Wife Sales

181

Once a wife sale had been concluded, the tables turned. Now it was the buyer who was vulnerable, because he had made a major investment to acquire a wife— usually his first—in hope of setting up a household and having sons. Many of these men had saved for years in order to marry for the first time, and if they lost both wife and investment, would they ever get another chance to marry? Under the circumstances, it is not surprising that buyers would try to appease sellers with at least one supplementary payment. ST R AT E G I E S T O M I N I M I Z E R I SK A N D P E R SUA D E T H E BU Y E R

Awareness that Wife Sales Were Criminal Anecdotal evidence throughout the case records shows that many people knew that wife selling was against the law. Sometimes, family members or friends would warn men who were negotiating a wife sale about the illegality of that transaction. In a 1797 case from Pucheng County, Fujian, a friend warned Xu Yilong that “it’s against the law to ‘sell a divorce’ for money” (de cai mai xiu shi fan jin de shi), so, despite his poverty, Xu decided not to sell his wife and broke off negotiations.5 In an 1803 case from Liangdang County, Gansu, Chen Liuyi told his father that he had bought another man’s wife; as his father later testified, “I knew that ‘buying or selling a divorce’ is against the law (mai xiu mai xiu shi fanfa de shi), so I scolded my son. But since the deal was done, there was nothing I could do about it.”6 In an 1819 case from Tongjiang County, Sichuan, Yang Zongyou was planning “to buy the divorce” of another man’s wife, but his older cousin Yang Zongzheng tried to dissuade him by warning that “to take a wife who already has a husband in marriage is against the law” (qu you fu zhi fu you wei li jin).7 As these examples show, many people even knew the legal term “to buy or sell a divorce” (mai xiu mai xiu). Magistrates consistently responded to plaints that reported wife sales by admonishing the plaintiffs about the illegality of this practice. Such rescripts would be posted publicly at the yamen, and they may have had a propaganda effect that reached beyond the individuals involved in a given case.8 Some local officials also issued proclamations to publicize the prohibition of wife selling and related practices.9 An example is the epigraph that opens this chapter. Given the close resemblance between illegal wife sale and the many other variations of marriage that also might be considered sales, some people must have been honestly confused about what was permitted by law. On occasion, an indignant buyer would respond to false accusations by arguing, “I took her in marriage properly and openly through a matchmaker” (ming mei zheng qu)—by which he meant that no coercion or deception had been involved and the brideprice had been paid in full. Nevertheless, the burden of evidence suggests that most people

182

part two

negotiating wife sales understood that what they were doing was against the law, even if they did not necessarily believe it to be wrong. How did this knowledge affect the negotiations that preceded a sale? For buyers, the risk of prosecution and the risk of extortion were closely linked, because it was the criminalization of wife sales that gave sellers much of their leverage. What measures were taken to allay buyers’ fears? We have already seen that wife sale contracts were designed to prevent misunderstanding, to shield the buyer from trouble, and to ensure that any dispute would be resolved through mediation and without resort to the courts. Two other strategies to reassure a skittish buyer (or matchmaker) were the use of “request certificates” (qing zi) and petitioning the magistrate for permission to undertake a sale. “Request Certificates” Buyers were not the only participants in wife sales who feared trouble. Matchmakers, too, would be vulnerable if a wife sale ended up in court; they had absolutely nothing to gain and also might lose whatever fees they had been paid, in addition to risking a beating. This vulnerability meant that matchmakers, as well, might fear extortion and the threat of false charges. (It also gave them a strong incentive to mediate out-of-court settlements to any disputes that might arise.) Therefore, if a matchmaker was not personally close to the seller, it was normal to demand that the seller provide a “request certificate,” or a written statement requesting and authorizing the matchmaker to act on his behalf. Sometimes a more elaborate term was used, such as “contract to prevent future trouble forever” (yong du houhuan wenyue) or “contract requesting and recognizing a matchmaker, and pledging never to change my mind or stir up trouble” (qing ren meizheng duhuan yong bu fanhui zi shi wenyue). Usually the matchmaker would keep this document after the sale concluded, because its purpose was to protect him or her against an accusation of acting against the wishes of the woman’s husband. It seems that professional matchmakers always demanded such certificates before agreeing to act (in wife sales and in arranging widow remarriage), but many people who provided this service more casually did so, as well.10 In many court cases, matchmakers testified that they had agreed to represent the seller only after being given such documents. In a 1765 case from Ba County, Peng Desheng testified that “Jiang Xuejing asked me to act as matchmaker in order to sell off his wife in marriage. I was afraid to get involved, but he wrote out a request certificate for me. Only then did I agree to act as matchmaker for him.”11 The reason so many original request certificates can be found in the archives is that matchmakers who found themselves in court would submit them in self-defense. A request certificate put the onus for the transaction on the seller himself. Here is a typical request certificate, which was used in a wife sale in Ba County in the early spring of 1858, just after the New Year:

Chapter 6. Negotiations between Men in Wife Sales

183

I, Chen Wannian, of this county, hereby establish this request certificate. In the past, our two families relied on a matchmaker for me to take Cao Shi as wife. Against my expectations, we have suffered hardship, with too many mouths to feed, and have fallen into poverty. Having no other choice, my wife and I discussed the situation, and we have repeatedly entreated the matchmaker Chen Hai to find a household to which I can marry off my wife. We must separate in order to survive, and it does not matter whether she marries nearby or far away. In the future, I, Chen Wannian, may not go back on my word, and if I change my mind and cause trouble, the matchmaker may take this certificate to the authorities, and I shall willingly admit my crime without complaint. I hereby establish this request certificate and give it to the matchmaker to keep as proof. Witnesses: Chi Xingnian, Wang Daye Scribe: Chen Hexing Done, on Xianfeng 7.1.14, by Chen Wannian [mark]

This text explicitly authorizes the matchmaker to present the request certificate in court to prove that the seller himself was responsible for the sale, which is exactly what the matchmaker in this case ended up doing.12 In several respects, request certificates resemble wife sale contracts: both are written in the voice of the seller, and sometimes in his own hand; both explain the reason for the wife sale and guarantee no future trouble. In addition, request certificates always emphasize that it was the seller who approached the matchmaker and did so of his own free will. But a request certificate provided to a matchmaker usually would not name the buyer or give the brideprice, because they were not yet known. An 1828 case from Ba County illustrates a variation on this use of the request certificate. A blacksmith named Chen Youfu decided to sell his wife (he had fallen ill, exacerbating their poverty, and she was very unhappy), so he engaged a married couple who offered matchmaking services; the husband was a waiter in a tavern, an ideal position for hearing news and making contacts. On their instructions, Chen prepared a request certificate and delivered it to them along with his wife, who would stay with the matchmakers until they found a buyer. But the matchmakers failed to find a buyer immediately, and Chen became nervous that his in-laws would find out what he was doing and cause trouble, so he changed his mind. The matchmakers released his wife to him only after he promised to pay them 600 cash to cover her room and board, but he could come up with only 300 cash, and they refused to return his request certificate until he paid the full sum. This dispute provoked a fight, and the parties ended up in court.13 If one function of a request certificate was to protect the matchmaker, another was to allay the anxiety of a prospective buyer by proving that the husband had agreed to sell his wife and that this matchmaker had been authorized to represent him. For this reason, it was not unusual for a matchmaker to obtain a request

184

part two

certificate after initiating negotiations with a prospective buyer. An 1832 case from Ba County illustrates this process. Here, Wu Fangji describes how he came to buy the wife of Lai Rongfa: Because my wife Liu Shi is already 42 sui but has not given birth, I obeyed my father’s order to take a concubine so as to continue our line of descent. Lai Rongfa, who lives over thirty li away, was very poor and could not get by, so he voluntarily asked Zhang Daxing and Tian Shitai to act as matchmakers to marry his wife Cao Shi to me as my concubine. At first I refused, but Daxing and Shitai explained to me that Rongfa truly was so poor that he could no longer get by. He and his wife had discussed this for a long time and were perfectly willing to go their separate ways in order to survive, but no one had yet been found to take her in marriage. They also showed me a matchmaker request certificate as proof of their claims. Only then did I finally agree. On the 28th day of the sixth month, with the matchmakers’ help I ascertained that Rongfa and Cao Shi were both willing to separate, and in face-to-face discussion we agreed on a brideprice of 6000 cash. Rongfa drew up a marriage contract, which I present here to the court, and on the 29th I received Cao Shi and we consummated the marriage.

Here is the text of the request certificate that the matchmakers showed Wu Fangji: I, Lai Rongfa, hereby establish this request certificate: In the past I went through a matchmaker and took in marriage the oldest daughter of Cao Yingfu. Unexpectedly, my household fell on hard times and it became difficult to supply our daily food (ri shi wudu). Therefore, my wife and I have voluntarily and repeatedly asked Zhang Daxing and Tian Shitai to act as matchmakers, and they have negotiated an agreement for Cao Shi to be separately remarried to Wu Fangji. Now that I, Lai Rongfa, have issued this request, no brother, uncle, or nephew of his lineage may ever make wild accusations, say differently, or cause trouble. In the future, if I ever make trouble or try to cheat Wu Fangji, then I hereby recognize that the two matchmakers Zhang and Tian can present this request certificate [to disprove my claims], and I shall willingly accept punishment. This is entirely voluntary and I have been subject to no coercion whatsoever. Fearing that spoken words alone are unreliable, I have specifically established this request certificate for the matchmakers to save as proof. Witnesses: Lai Xinfa [mark], Lai Renfa [mark] [the seller’s brothers] Scribe: He Shengpian Done, on Daoguang 12.6.2, by Lai Rongfa [mark]

As Wu Fangji’s plaint indicates, the main purpose of this document was to enable the matchmakers to prove that they were acting with the husband’s consent in order to reassure the prospective buyer. It states the buyer’s name, indicating that the matchmakers had already initiated negotiations with him before obtaining the request certificate in order to close the deal. At the same time, this document protected the matchmakers against accusation of acting without the husband’s authorization, and they retained it and later presented it in court for that very purpose.14

Chapter 6. Negotiations between Men in Wife Sales

185

Petitions for Permission to Sell a Wife Another strategy to minimize risk was to ask the magistrate for permission. Such petitions took the form of requests to “cun an”—literally, “to file for reference”—or to create an official record of something at the yamen. People might petition to cun an for various reasons, but the basic purpose was to create a record consisting of the petition and the magistrate’s rescript, which could be consulted in case of need. A magistrate’s agreement to create an official record indicated his approval. Such a petition was usually prompted by the buyer’s fear that either the seller might attempt extortion or the woman’s natal family might object. The buyer himself might submit such a petition, but in most of my examples the seller did so in order to persuade the buyer to agree to the sale. A magistrate’s agreement to create such a record meant that the buyer need not fear going to court if necessary and thus protected him against extortion. In an 1836 case from Nanbu County, we find a magistrate approving a request for permission to sell a wife. This case concerned a young man named Liu Guodong, whom various parties described as mentally deficient and incapable of supporting his wife, Yang Shi. The couple had initially been supported by Liu’s father, Liu Dashun, but after Dashun died their difficulties worsened dramatically. Liu’s widowed mother, Liu Sun Shi, and other elders of the Liu family finally concluded that the situation was untenable and decided to sell Yang Shi in marriage. She and her father agreed. Two members of the Liu lineage who acted as matchmakers identified a young widower named Huang Wan’guo (22) as a likely buyer. But Huang was anxious about marrying a “wife with a living husband,” so Liu Sun Shi agreed to submit a petition explaining that her son’s “illness” and poverty made it necessary for her to marry off her daughter-in-law to someone else, and asking permission to “create an official record” to that effect. The magistrate responded: “If Liu Sun Shi [feels that she must] marry off her daughter-in-law Yang Shi because of her son’s illness, then she may do as she pleases.” After seeing this rescript, Huang agreed to buy Yang Shi for a brideprice of 16,000 cash. The wife sale contract (of which an official transcript exists on file) is drawn up in the names of husband Liu Guodong, his mother Liu Sun Shi, and his two paternal uncles, and witnessed by three other men of the Liu lineage. The contract records the act of petitioning the magistrate: “Representatives of the Liu and Yang families entered the city and petitioned for permission for [Yang Shi] to remarry to someone else, so there is an official record that can be checked if necessary; only then did Wan’guo agree to the marriage.” Moreover, at the very end of the contract there is a transcription of the magistrate’s rescript granting permission. Later, this case came to court because the two uncles whose names appear on the contract demanded supplementary payments from the buyer; and when he

186

part two

refused to pay (feeling protected by the magistrate’s rescript), they filed charges accusing him of abducting Yang Shi. In self-defense, Huang Wan’guo explained all that had happened, referred the magistrate to the “official record” on file, and submitted the wife sale contract as supporting evidence. The case file is incomplete, but the fact that the contract was transcribed for the record (instead of being confiscated) indicates that the second marriage was ratified in conformity with the magistrate’s rescript granting permission.15 This example of a magistrate approving a petition for permission to sell a wife is unique in my case sample. The unusual circumstances would seem to explain this magistrate’s attitude, in particular the fact that the petition came from the husband’s mother, backed by her brothers-in-law, who portrayed him not only as poor but also as disabled. Magistrates usually rejected petitions “to create a formal record” for wife sales, and such petitions might backfire badly on those who submitted them. Magistrates would react with harsh rhetoric, citing the criminality of “buying or selling a divorce” and threatening dire consequences. We find a typical example in the following case. Yan Zongpin (33) was a runner at the yamen of Jiangbei Subprefecture, located directly north of Chongqing on the opposite bank of the Jialing River (which served as the northern boundary of Ba County). Yan and his wife, Qiu Shi, were very poor, and they had no children; they lived with Yan’s mother. In early 1873, in the last month of the lunar year, Qiu Shi managed to cross the river and make her way to the Ba County yamen, where she filed an oral complaint that her husband and mother-in-law wanted to sell her to two female matchmakers, who planned to take her “downriver” and resell her into prostitution (the matchmakers testified that, since Qiu Shi was “young and attractive,” she would be a good candidate for that work). At trial, however, Yan testified that he had been away on official duties and his mother had made these arrangements without his knowledge. The magistrate opted to return Qiu Shi to Yan, forbidding any further abuse of her, while ordering the two matchmakers slapped; no one else was punished. But within a month, just after the New Year, Yan was back at the Ba County yamen, where he submitted a petition asking permission to “create an official record” so that he could sell his wife in marriage. His previous testimony notwithstanding, this petition candidly admits that he had tried to sell her once before, because “I am desperately poor and simply cannot support her.” This time, he explains, he is trying to do things properly: first he discussed the situation with his wife and secured her consent; then he engaged a new set of matchmakers to negotiate her sale to a man named Yang Zehong, and Qu Shi “happily approves” of the match. Yang has already paid Yan in exchange for a “contract bearing my hand and foot prints.” But now the buyer “fears that later on I will go back on our agreement and make excuses to extort and cause trouble, and he is unwilling to proceed without official permission”—hence Yan’s petition.

Chapter 6. Negotiations between Men in Wife Sales

187

The rescript on Yan’s petition illustrates how magistrates usually reacted to such requests: “Both selling and buying divorce are strictly prohibited! (mai xiu mai xiu da gan li jin). Do not brazenly request permission to create an official record for such a purpose!”16 The case file ends there. A 1753 case from He Department, Sichuan, shows that a petition to “create an official record” might be prompted by knowledge both that wife sale was illegal and that magistrates sometimes approved wife sales anyway. In this case, Yuan Ren (a migrant from Guangxi who earned a living making wine) had arranged to buy Zhou Shi from her husband, Wen Shaoliang (the couple were destitute migrant laborers from Guizhou). But Yuan’s cousin Pan Pinshan stopped him with the following warning: “This woman is a wife with a living husband, so you can’t take her in marriage unless you petition the local magistrate and get official permission to do so.” Yuan Ren insisted that Wen and the matchmaker accompany him to the departmental yamen, where they filed a petition “to create an official record” for the proposed wife sale. But the magistrate happened to be out of town, and Yuan finally canceled the transaction and returned home.17 Some petitioners anticipated a negative reaction, so, when petitioning to create an official record in order to allay buyers’ fears, they tried to disguise the facts in order to forestall an outright rejection. I have several examples of such disingenuous petitions. A typical one comes from Nanbu County, dated 1869. The file begins with a petition from one Hong Zhengwen explaining that he is both poor and chronically ill; that his wife, Zhu Shi, is unhappy, disobedient, and refuses to remain married to him; and that both families have concluded it would be best for the couple to separate and “to let her go her own way.” If Hong recovers his health, he may choose to take another wife, and now he wants “to create an official record” in order to prevent her or her natal family from causing trouble in the future. But Hong makes no mention of any compensation for divorcing his wife, let alone the possibility that he might just sell Zhu Shi himself. The magistrate’s rescript is brief and to the point: “Since you both are willing to separate (liang yuan liyi), then you should rely on your two families and the matchmaker and witnesses from your marriage to make appropriate arrangements. It is not necessary to create an official record asking permission.” Here, the magistrate is citing the accepted standard for divorce under Qing law (see Chapter 9). The next document in the file is a plaint from the same Hong Zhengwen, filed two months later, in which his given name is written with different characters and he is listed as having a different age than in his previous petition—these discrepancies apparently represent a clumsy attempt to prevent the yamen staff from linking him with the man who had asked permission to create an official record about his divorce plans. In this new plaint, Hong claims that his wife’s natal family, a matchmaker, and a thug named Wang Dejin had conspired to kidnap his wife and sell her to Wang in marriage, all behind Hong’s back, and that when he discovered

188

part two

their nefarious deed and protested, they had beaten him up to deter him from doing anything about it. Unfortunately for Hong, he was immediately identified as the earlier petitioner, and the magistrate’s rescript is severe: “If in fact you have sold a divorce and are now practicing extortion, then you will be severely punished without mercy!” There follows a series of plaints from the buyer’s father, the woman’s father, and various natal relatives of hers, all agreeing that Hong had voluntarily sold his wife to Wang Dejin for fourteen strings of cash, and that he had already extorted a further 2000 cash from Wang in exchange for a written pledge never to cause trouble again. Hong had then filed false charges in an attempt to pressure Wang to pay him even more. The magistrate’s response: “To sell a divorce because of illness, and then attempt extortion, is an extremely disgusting sort of evil practice!” The file is incomplete, but it does contain the original wife sale contract bearing the prints of Hong’s left hand and right foot, so at least we can be sure that it was confiscated and the transaction canceled.18 These petitions show that petitioners understood wife sale to be illegal but also that magistrates occasionally permitted such transactions—why else would these people think it worthwhile to seek permission? In addition, such petitions reveal a strong and widespread belief that a wife sale motivated by poverty, transacted through a matchmaker without deceit or coercion, was reasonable and should be permitted. These are the elements consistently emphasized by such petitions, which are often accompanied by contracts submitted as supporting evidence that stress the same points. From this standpoint, a wife sale was no more shameful than widow remarriage—which, after all, usually involved a very similar transaction. This alternative perspective emerges in a remarkable 1853 case from Ba County, in which litigants had the audacity to contest the magistrate’s reading of the law. In this case, peasant Chen Youchuan filed a candid petition to create an official record of his purchase of Liang Shi, the wife of beggar Yang Zongpan, along with their little daughter. Chen’s reason for reporting this transaction was the usual buyer’s fear of future trouble: he wanted to create an official record so that “if Yang Zongpan or Liang Shi’s natal family collude to harm me, there will be a record on file to serve as evidence.” There is no hint in Chen’s petition that he understood wife sale to be illegal, but the magistrate was quick to enlighten him: “Selling and buying a divorce are both strictly prohibited by law!” Liang Shi’s father, Liang Zongzhi, and her first husband, Yang Zongpan, were summoned to testify. But these men begged the magistrate to reconsider. Liang Zongzhi explained that Yang had failed to support his family and that Liang Shi, out of desperation and hunger, had often returned to her natal home; in the end, Liang had persuaded Yang to sell his wife and daughter so that they could escape their hopeless situation. Yang, for his part, explained that he had sold

Chapter 6. Negotiations between Men in Wife Sales

189

his wife and daughter so that all three could “survive by fleeing our separate ways.” He even credited her new husband, Chen, for helping pay for his medical care in addition to paying Liang Shi’s brideprice. (Chen’s charity is a good example of a buyer’s sympathy for the seller as the loser in a wife sale.) Yang went so far as to challenge the magistrate’s rescript on the criminality of “selling a divorce”: “The fact is, I sold my wife in marriage because I had no means to support her, so my selling her has nothing to do with the law against ‘selling a divorce.’” In conclusion, both Liang and Yang urged the magistrate to “consider the facts” and refrain from interfering. The case file ends there, leaving us ignorant of how these people ultimately fared. But, clearly, they believed that they had done the right thing, regardless of what the magistrate might say about the law. Their argument articulated an understanding of right and wrong independent of legally codified norms.19 N E G O T IAT IO N S OV E R SU P P L E M E N TA RY PAYM E N T S

The single most common reason a wife sale ended up in court was that the seller (or occasionally a member of his family) would demand more money from the buyer after the sale was complete—often months and occasionally even years later. These scenarios account for 43 percent of the routine cases in my sample from local archives. Demands for supplementary payments also figure prominently in the disputes that provoked homicidal violence in my sample of xingke tiben.20 In making such demands, it was common for the seller to threaten to file charges, and sometimes he would actually do so.21 Sellers seldom intended to pursue their lawsuits all the way to formal judgment—rather, they hoped to intimidate buyers into paying them off. But a buyer wearied by incessant demands might call the seller’s bluff by filing a counter-plaint with the wife sale contract attached. In fact, it is clear that buyers would often make at least one supplementary payment. Given the risk involved in going to court, plus the cost of litigation, it was far safer and possibly cheaper simply to give the seller some more money.22 These supplementary payments could be substantial, as shown in Tables 4 and 5. Table 4 shows forty-four examples from the local court records. In each of these cases, we know that the buyer made at least one supplementary payment prior to litigation, and in seven of them, at least two payments were made (the largest number of documented payments in a single case being four). In at least thirteen cases, such payments equaled or exceeded the original brideprice. (Note that, thus far, we have focused on open, direct wife sales, but as these tables show, sellers would demand supplementary payments after compensated divorces and fraudulent sales as well.) Buyers made these payments mainly to avoid litigation but also—in at least some cases—out of a grudging sense of obligation to help the sellers, who were usually far worse off than they; one senses on the part of these

table 4 Supplementary Payments: Examples from Local Cases (One Payment Unless Noted) Year

County

Brideprice

Type of transaction

Total supplementary payment(s)

1808 1813 1815 1818 1819 1824 1826 1833 1836 1841 1843 1849 1851 1851 1852 1853 1853 1854 1854 1857 1857 1858 1858 1859 1860 1862 1862 1863 1865 1865 1867 1868 1868 1869 1869 1869 1870 1872 1872 1878 1881

Ba Ba Ba Ba Ba Nanbu Ba Nanbu Nanbu Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Ba Nanbu Ba Nanbu Nanbu Ba Ba Nanbu Ba Nanbu Ba Ba Baodi Ba Nanbu Nanbu Ba Nanbu Nanbu Nanbu Nanbu Nanbu

5 taels 1500 cash unclear 20 taels 5400 cash 6000 cash 7000 cash 4000 cash 12 strings cash 30 taels 100 taels 10 taels 8000 cash unclear 5000 cash 3 strings cash unclear 4000 cash unclear 5000 cash 5000 cash 50+ taels 20,000 cash 10 strings cash 2000 cash unclear 11 strings cash 4 strings cash 11 strings cash 5000 cash 20 taels 70 strings east cash 7 strings cash 6000 cash 14 strings cash unclear 11 strings cash 10,400 cash 12 strings cash 14 strings cash 6 strings cash

open sale open sale open sale open sale comp. divorce open sale comp. divorce fraudulent sale open sale fraudulent sale comp. divorce comp. divorce fraudulent sale comp. divorce open sale open sale open sale open sale open sale open sale open sale open sale open sale open sale fraudulent sale fraudulent sale open sale unclear open sale open sale comp. divorce open sale open sale open sale open sale open sale open sale fraudulent sale fraudulent sale open sale fraudulent sale

1899 1900 1910

Nanbu Ba Nanbu

10,000 cash 30 taels 15 strings cash

fraudulent sale fraudulent sale fraudulent sale

1000 cash 11,200 cash 1000 cash 24 taels 400 cash 7000 cash 3000 cash 4000 cash 2000 cash 3000 cash 120 taels (2 payments*) 15 taels** 2000 cash 26 taels (4 payments) 2000 cash 4 strings cash coffin for mother-in-law 1300 cash unclear 1600 cash (2 payments) 2000 cash unclear (at least 2 payments) 4000 cash 2200 cash (2 payments) 1200 cash 5 strings cash 11 strings cash 2500 cash 4400 cash 6000 cash 12 taels 5 strings east cash + jacket 1600 cash 1200 cash 2000 cash unclear 4000 cash 2600 cash 30 strings cash 6 strings cash 12 strings cash; 3200 cash (2 payments) 16,000 cash 20 taels + 3 strings cash 24 strings cash; 10 strings cash (2 payments)

* 100 taels paid by natal family, 20 by new husband ** Paid by natal family

Chapter 6. Negotiations between Men in Wife Sales

191

table 5 Supplementary Payments: Examples from xingke tiben Year

Province

Brideprice

Type of transaction Total supplementary Payment(s)

1755 1768

Jiangsu Jiangsu

28.4 taels 4000 70% cash

open sale open sale

1786 1798 1804 1805 1813

Shandong Zhejiang Sichuan Zhili Gansu

14,000 small cash 13,000 cash 13.5 taels 30,000 cash 52 taels

open sale open sale open sale fraudulent sale fraudulent sale

amount unclear 1 tael; 2 items of clothing to be pawned* 12,000 small cash 2000 cash 14 taels amount unclear 1200 cash; 800 cash*

* Two separate payments

buyers a combination of annoyance, exasperation, and pity. In each of these cases, however, the seller subsequently asked for even more money and was rebuffed, at which point he filed charges against the buyer, or the buyer filed charges against him instead. Table 5 shows seven examples from xingke tiben, all involving homicide, in which we know that the buyer made at least one supplementary payment before the seller’s further demands provoked a violent confrontation. This smaller sample from six provinces tells the same story as the three-county sample of local cases: the relative values of supplementary payments and brideprices are similar, with at least one payment exceeding the original brideprice; and in two cases we know that the buyer made two separate payments. It seems safe to conclude that for the seller to demand such payments (and for the buyer to pay at least once) constituted a fairly routine feature of wife sales. Land Sales in the Peasant Economy Cases that came to court because of demands for supplementary payments reveal a remarkably close parallel between male peasants’ attitudes toward wives and their attitudes toward land. To understand the perspective of a man who sold his wife, it is necessary to have a firm grasp of how land sales worked in the peasant economy. In theory, land sales were either conditional—the default form of sale—or absolute.23 In a conditional sale (dian mai, also known as “live sale,” huo mai), the sale price was less than the land’s full market value, and the seller retained the right to “redeem” (shu, shuhui, huishu) the land by refunding that sum within a fixed period of time, at which point the buyer was required by law and custom to return the land to the seller. But if the seller failed to redeem the land (a frequent outcome, given that peasants often sold land out of desperation and found it difficult to redeem it), then, in principle, the buyer could convert a conditional sale into an

192

part two

absolute sale by making one final payment (zhao tie) that would cover the difference between the original sale price and the land’s current market value. Given the long-term secular rise in land prices, however, the more time passed the more expensive this would be, so converting a conditional sale to an absolute one by paying the zhao tie was often not to the buyer’s advantage or within his means. In short, many land sales resulted in an ambiguous situation—with the land not redeemed, but the transaction not converted to absolute sale—that might continue indefinitely.24 As long as the buyer did not pay the final zhao tie, the seller was entitled to redeem his land. But if the seller could not afford to do so, it was common to demand a series of smaller “supplementary payments” (zhao jia) instead, sometimes long after the sale had occurred; the seller might threaten to redeem the land (at original sale price) unless the buyer made the supplementary payment. Buyers usually agreed, because the more time went by, the bigger the gap between the original price and the current market value. The timing of demands for supplementary payments correlated with the rise of land values, which sellers felt entitled to share.25 The Qing state repeatedly attempted to limit supplementary payments to a single, final zhao tie, but such attempts failed, and conditional sellers continued to demand and receive multiple supplementary payments through the end of the Republican era.26 (The persistence of this practice in defiance of the state shows that community norms and regulatory mechanisms had a much stronger influence than formal law.) To complicate things further, it was not uncommon for the conditional buyer to rent the land back to its seller, who would continue to cultivate it. In other words, conditional sale enmeshed land in a complex web of claims that limited its alienability, and it created a relationship between seller and buyer in which the seller felt entitled to a continuing stream of income from the land while retaining the option of redeeming it. In theory, land transferred by absolute sale (jue mai or du mai, also known as a “dead sale,” si mai) was not redeemable, and after paying the agreed price the buyer owed nothing more. In practice, however, it was not uncommon even after an ostensibly absolute sale for the seller to demand supplementary payments and to act as if it had in fact been conditional.27 Qing lawmakers tried to eliminate ambiguity by ordering that any transaction for which the contract did not explicitly state “absolute sale” be considered conditional by default. This policy reflected the fact that, in customary practice, the default form of land transfer was indeed conditional sale. It also protected the seller (seen as the weaker party) by placing the burden of proof on the buyer.28 Sellers and their descendants might demand repeated supplementary payments from buyers and their descendants, sometimes over a period of decades, and also typically claimed a permanent right to redeem the land if they chose to do so. Whereas the seller (and his heirs) felt a continuing sense of connection and entitle-

Chapter 6. Negotiations between Men in Wife Sales

193

ment, the buyer (and his heirs) would resent being the object of what felt, as time passed, more and more like extortion. It is not surprising that these negotiations provoked much litigation. The Parallel between Wives and Land Peasant land sales are relevant for our inquiry because we find exactly the same vocabulary and a number of analogous practices in wife sales, including the distinction between conditional and absolute sale and the practice of demanding supplementary payments even in ostensibly absolute sales. Relatively few wife sales were explicitly termed conditional. A typical contract for wife sale follows the format of a contract for an absolute land sale, stating that the brideprice has been paid in full and the seller’s relationship with the wife irrevocably severed. Nevertheless, it is obvious that many sellers felt a strong and continuing sense of entitlement vis-à-vis their wives. These men still referred to the women as their wives, even after having sold them—and, very often, the women also continued referring to these men as their husbands. In testimony, it was standard for a man who had sold his wife to identify her by saying (for example), “this Chen Shi is my wife” (zhei Chen Shi shi xiaode qizi)—not “ex-wife,” or “the wife I sold,” or “the woman who used to be my wife”—regardless of how much time had passed.29 Terms such as “the wife of a living husband” (huoren qi, shengren qi, or simply sheng qi) or “a wife who already has a husband” (you fu zhi fu), used all over China to refer to a sold wife, reinforce the sense that, in some fundamental way, the seller remained the woman’s real husband. When sellers demanded supplementary payments, their words reflect this sense of entitlement as well as burning resentment at the low prices they had had to accept. In a 1768 case from Nanhui County, Jiangsu, a seller named Gu Shi had already extracted two supplementary payments from his wife’s buyer, Chi Chao, when he demanded yet another. Chi scolded Gu for trying to cheat him out of his “hard-earned money”: “What makes you think I have so much money to give you?” But Gu cursed him, saying, “I sold my wife to you—do you really think she’s worth only that handful of taels you paid, and that you owe me nothing more? (nandao zhi zhide zhei ji liang yinzi, bu gai zhao xie qian me).”30 In a 1794 case from Gushi County, Henan, Quan Yugao demanded a supplementary payment in the following terms: “When I sold my wife to you, all I got for her was 5000 cash. Now I’ve got nothing left and can’t get by, so you should give me a few thousand cash more to live on.”31 In an 1813 case from Renshou County, Sichuan, the seller had had to accept a mere 1000 cash to close the deal, but he had no intention of settling for such a small sum; just four days later, he visited the buyer, who later testified: “He said that the brideprice he had previously agreed to was too small, and demanded that I give him an additional 2000 cash, only then would he leave us alone. He said that if I refused he would take his wife back.”32

194

part two

As the last example illustrates, many husbands seem to have assumed that they retained the option of redeeming or simply taking back their sold wives at some later date, even though those sales had supposedly been absolute and final. Sometimes, for example, a husband would persuade his wife to cooperate with a sale by promising to redeem her once he had saved up enough money to refund her brideprice. Or, after the sale had been transacted, he might inform her that he intended to redeem her as soon as he was able (although not every wife expressed enthusiasm at the prospect). As one seller later testified (in an 1802 case from Dayi County, Hunan): “It seemed really difficult to let go of our mutual feeling and relationship (qingfen nan she), so I told my wife that I had sold her temporarily only because of poverty, and in the future, after I have saved enough money, I would redeem her and get her back.”33 In other cases, if a wife’s natal family was outraged at her sale, the husband would promise to redeem her, as long as they helped defray the cost of doing so.34 There are also cases in which sellers demanded supplementary payments and, when rebuffed, threatened to take their wives back— such men believed that, unless satisfied, they could simply cancel a wife sale without refunding the original brideprice.35 These examples blur the boundary between absolute and conditional sales, showing that many a husband assumed that his claim to his wife had not been severed by the act of selling her, regardless of what her buyer might think. Buyers would complain of “extortion” and “blackmail” (suo, ezha), but in land disputes they complained in exactly the same way. What was the shared logic between sales of wives and of land? The logic framing peasant land sales was what Philip Huang has called “a precommercial ideal of permanence in landholding,” an ideal that reflected, among other things, all the labor the seller and his family had put into the land prior to the sale. It was this logic that held sold land to be permanently redeemable and empowered sellers to demand repeated supplementary payments. The conditional nature of most land sales extended survival credit to poor peasants who had sold land while securing further income and an indefinite opportunity to recover their land.36 In the poor peasant economy, land carried a meaning and value very different from those associated with commodities in a capitalist economy. As Huang explains, Land was much more than just a piece of property to be bought and sold. It was more even than the source of a man’s livelihood. It was the very basis for his sense of dignity, and for his status in the village community. It was what separated him from the sorry lot of the homeless who were forced to drift from place to place, and it embodied the very continuity of his patrilineal ancestry and descent, concretized in the family gravesite. Sales of land could thus be acts of profound symbolic meaning for the seller.37

Since the usual reason for peasants to sell land was economic desperation, they seldom viewed its sale as a truly equal exchange; rather, the buyer had taken

Chapter 6. Negotiations between Men in Wife Sales

195

advantage of and benefited from the seller’s misfortune. Thus, the buyer would traditionally hold a feast to thank the mediators and witnesses to the sale, whereas the seller was an object of pity and, possibly, scorn. Practically everything said here about land would apply equally to a wife. Among the peasantry, social status and economic viability depended above all on family farming, for which a man required both land and a wife; moreover, a man had to have sons to fulfill his duty to ancestors. Land was vital for material maintenance of a patriarchal family; equally, a wife was vital for that family’s biological and social reproduction. Moreover, in China full social adulthood came with marriage and reproduction (some might say that this continues to be true). Therefore, a man driven to sell his wife would inspire both pity and contempt—the latter sometimes expressed in organized hazing. Equally, a man who bought “the wife of a living husband” was seen as having benefited from and even taken advantage of his misfortune. This view was expressed in a folk proverb, “haohanzi bu qu huoren qi”: a real man—that is, a man of honor, of conscience—does not take in marriage the wife of another man who is still living.38 The parallel between land and a wife is expressed in other proverbs, such as, “if you have money, don’t take in marriage a living man’s wife; if you want land, don’t choose roadside land” (you qian bu qu huo ren qi, yao di bu yao shun dao de). The reasoning here seems to be that buying a wife with a living husband was risky, and so was buying roadside land (because crops would be vulnerable to damage and theft). Variations include, “if you buy land, don’t buy land by a river; if you take a wife, don’t take in marriage a living man’s wife” (mai di bu mai he wan di, qu qi bu qu huo ren qi) and “if you have money, don’t farm land with no tax grain quota; if you have money, don’t take in marriage a living man’s wife” (you qian bu zhong wu liang di, you qian bu qu huo ren qi). The reference to “land by a river” reflects the fact that, in southern China, waterways rather than roads were often the main thoroughfares. In the latter variation, the reasoning seems to be that wives with living husbands and land with no tax grain quota are both seemingly cheap deals with suspicious strings attached.39 In many parts of China, folk beliefs about the polluting effects of marital separation also reflect the parallel between wives and land. In Hunan, according to a Republican-era survey of customs, “when a woman with a living husband marries out . . . it is necessary both to write the contract and to have the woman mount the sedan chair at the roadside or the toilets [i.e., away from the home and agricultural fields], because it is said that no plants will grow where those things are done.”40 A proverb from Jilin held that “divorcing a wife will ruin your land, and there’s nothing you can do about it for the rest of your life” (xiu qi hui di, dao lao bu ji).41 Similar sayings can be found in Taiwan: “No grass will grow for three years on the land where a wife is divorced”; “No grass will grow for three years on the land where a divorce contract is written”; and “If the ink used to write a divorce

196

part two

contract is spilled on grass, the grass will wither and die.”42 The common theme is that destruction of a household’s biological fertility (through loss of the wife) would endanger the fertility of its land, through a kind of sympathetic magic. Of course, there are significant differences between a wife and a parcel of land (aside from the obvious fact that land is an inanimate object whereas even a sold wife could exercise some agency). Land is unmovable, so buyers for land were sought locally, and kin and immediate neighbors enjoyed the customary “right of first purchase.”43 In wife sales, however, as in conventional marriage, village exogamy seems to have been the rule, and the rare exceptions found in the case records suggest that selling a wife within one’s own village was a recipe for trouble. Another clear difference is that land properly maintained will retain its value indefinitely— hence the opportunity for a seller’s heirs to redeem sold land or demand supplementary payments from the buyer’s heirs even decades after a sale.44 In contrast, women are mortal, of course, and their fertility has an even shorter life. Thus, after wife sales, most demands for supplementary payments were made within one year of the sale, often within just a few weeks. A corollary is that a land sale contract retained its use as title deed as long as the land was not resold; for this reason, vast numbers of these documents survive from the Qing dynasty. But a wife sale contract retained value only for the lifetime of the woman and her buyer, so almost none of these documents survive outside local court archives.45 These differences notwithstanding, the shared vocabulary and contractual forms of sales for land and wives, and the demands for supplementary payments found in wife-selling cases, evidently reflect an attitude about wives fundamentally similar to peasants’ attitude about land. It is as if an asset as vital as land or a wife could not be alienated absolutely, no matter what superficial form a transfer might take; there was a fundamental connection, tied up with status, identity, survival, and morality, that could not be severed by the payment of money. In other words, to “sell” a wife (or a piece of land) did not sever one’s connection to that wife (or land) so much as it initiated a new and ongoing relationship between seller and buyer—a relationship in which the buyer was seen as morally obligated to assist the seller, whose misfortune had benefited him. This imbalance also meant that the buyer had far more to lose by going to court than the seller did, and that is why buyers often agreed to make supplementary payments. This logic of negotiation between men framed wife sales and fueled much of the related litigation. “Submission Contracts” and the Negotiation of Supplementary Payments When a buyer agreed to make a supplementary payment, it was standard practice for the seller to provide a “submission contract” (fu yue) or “acknowledgment contract” (ren yue) in exchange. There are many different terms for such documents, but they all share certain features because their purpose was to ensure that the

Chapter 6. Negotiations between Men in Wife Sales

197

seller would make no further demands for money or otherwise harass the buyer. This purpose is reflected in some of the more elaborate terms for these documents, such as “contract promising never again to demand supplementary payments”; “contract promising that I will never change my mind or attempt further extortion”; and “contract promising that I will never again use the pretext of having sold my wife to attempt extortion.” In this section, we start by examining three cases from Ba County that illustrate how supplementary payments were negotiated and documented. There are dozens of references to submission contracts in both local and central case records, and many original examples survive in the Ba and Nanbu County archives. These documents were part of the relatively uniform vernacular contract culture found throughout China before the revolution. As with wife sale contracts, however, they survive only as a result of being confiscated during prosecution, when a magistrate canceled the second marriage. (If instead he ratified the second marriage, then he would let the buyer retain the contract and any related documents.) For this reason, the following three cases all conclude with harsh judgments that demonstrate the risk of taking a dispute over a wife sale to court. These examples document the process of negotiating supplementary payments, but they also show why the threat to file charges was taken seriously. An 1854 case provides a classic example of an open sale followed by the extortion of a single supplementary payment. After seven years of marriage, Li Mingzhong (40) and his wife, Zhu Shi, decided to “separate in order to survive,” and with his mother’s permission Li arranged to sell Zhu Shi to Wan Er, a maker of musical instruments who lived in Chongqing. Two friends served as matchmakers, for free. The wife sale contract is written in the voices of Li and his mother, who acted to zhu hun: I, Li Yang Shi, and my son Mingzhong hereby establish this handprint contract in which we voluntarily authorize the marrying out [of Mingzhong’s wife, Zhu Shi]. The situation is as follows: in years past, Mingzhong took Zhu Shi in marriage as his wife, but now the family is poor, we have trouble getting our daily food, we are as impoverished as if washed clean by water (jia ru shui xi). Therefore, mother and son have discussed the situation and voluntarily agreed to ask our neighbors Ma Xinwen and Zhang Xingshun to negotiate to marry Zhu Shi to Wan Er as wife. Relying on the matchmakers, we have agreed in face-to-face discussion on a brideprice of exactly 4000 cash, and on this day we have received the full amount with our own hands. In the future no one from either natal or affinal families, whether now present or not, may say different or make excuses to cause trouble. Li Mingzhong and his mother will exert all their effort to oppose anyone who tries to do so. This agreement is made voluntarily by both [the Li and Wan] families, without any deception or coercion. Fearing that words alone are not reliable, we hereby establish this contract authorizing Zhu Shi to marry out in order to serve as proof.

198

part two Footprint, handprint, voluntary agreement [the contract bears a left footprint and a left handprint, with these words written inside] Neighbors [four men, including scribe] witnessing together

The contract bears the names and marks of Li Yang Shi (a circle) and her son Li Mingzhong (a cross), along with his hand- and footprints. Like other wife sale contracts, it states that the brideprice has been paid in full, the sale is final, and Li and his mother will shield the buyer from trouble. Six weeks after the sale, however, Li Mingzhong approached Wan Er to demand more money. After mediation by neighbors, Wan agreed to pay him 1300 cash in exchange for the following submission contract: I, Li Mingzhong, hereby establish this contract of voluntary acknowledgment and submission. The situation is as follows. Some years ago I took Zhu Shi in marriage as my wife. This year I could not earn a living and had no way to feed my family, and it was difficult to continue supporting my wife. Since I had no choice, on the second day of the ninth month, I begged Zhang Yuting to act as matchmaker to marry my wife off to Wan Er. Mr. Wan paid a brideprice of 4000 cash, which I received in my own hands in the presence of the matchmakers, witnesses, and neighbors. I was paid in full and not a single cash was still owed me. But in my foolishness I spent the entire sum, and then because I have no shame, I repeatedly made false accusations and tried to cheat Mr. Wan (bugu xiuchi, die ci wu pian). On the 24th day of the 11th month, Mr. Wan and the collected neighbors threatened to take the wife sale contract and report me to the authorities. Fortunately, Lu Dengxiao, Yang Zonghua, and Liu Yushan intervened as guarantors and persuaded Mr. Wan to give me 1300 cash in order to relieve my difficulty. Now, Lu, Yang, and Liu have set the condition that never again may I harass Mr. Wan or stir up trouble. If in the future, I once again act shamelessly to make false accusations and try to cheat Mr. Wan and stir up trouble, then Lu Dengxiao, Yang Zonghua, and Liu Yushan will take this contract to the authorities and report me for prosecution, and I am willing to accept punishment. Now, fearing that people’s hearts are not reliable as of old, I hereby establish this contract of my voluntary acknowledgment and submission and give it to Mr. Wan and the guarantors for them to keep as proof. Relying on neighbors [nine men, including scribe] to witness together

The document bears the name and mark of Li Mingzhong and is dated Xianfeng 2.10.21. Two of the men who had witnessed the wife sale contract played a role in mediating and witnessing this agreement as well.46 Several features of this document stand out and can be found in other submission contracts. It is written (like a wife sale contract) in the voice of the seller and is signed with his mark. The first few lines recapitulate the terms of the sale and confirm that the brideprice has already been paid in full. The next passage uses humiliating language (“In my foolishness . . . I have no shame . . . I tried to cheat . . .”) to extend this confession to include lies and extortion; the seller admits that

Chapter 6. Negotiations between Men in Wife Sales

199

the buyer would be perfectly justified in turning him in to the authorities. The buyer’s decision to withhold prosecution and instead to make a supplementary payment is portrayed as an act of mercy and generosity, reached through the intervention of disinterested and altruistic mediators. The condition is that the seller must promise never to repeat his “shameless” behavior. The text closes with confirmation that if the seller breaks his promise, the buyer may take this document to the authorities and turn him in for punishment. In sum, the submission contract is designed to exonerate the buyer of any wrong, to document the seller’s shameful misdeeds, and to humiliate and intimidate him into refraining from future extortion. It was negotiated before a large group of witnesses who were the buyer’s neighbors: nine men in all, of whom the three named in the text appear to have been important members of the community. This document records a ritualized public encounter in which the seller had to abase himself before the buyer and his neighbors in order to secure a supplementary payment. Nevertheless, he did secure that payment—and that, I think, is the key point. Having been paid off, Li Mingzhong left Wan Er alone for a little over a year. But in 1854, just after the New Year, Li approached Wan again to ask for more money. By this time, Zhu Shi was heavily pregnant with her new husband’s child, and news of her pregnancy is probably what prompted Li’s new demand.47 This time, however, Wan refused to pay him, so Li filed a plaint at the Ba County yamen accusing him of having had adultery with Zhu Shi and eloping with her, in the process stealing 5400 cash and many items of clothing and jewelry (Li attached an itemized list to his plaint). But Wan refused to yield, instead calling Li’s bluff by presenting the wife sale contract and submission contract in court. Wan would have been better off just paying Li a little more money. The magistrate’s harsh judgment—“Both buying and selling a divorce are crimes that deserve punishment!”—illustrates the risks of taking a dispute over a wife sale to court, especially for the buyer. The magistrate canceled both marriages, returned Zhu Shi to her natal family to be remarried (after giving birth), and ordered both Li and Wan beaten and put in the cangue for a month (an unusually harsh punishment for a routine wife sale, exceeding the penalty mandated by statute). The only hint of lenience was that the magistrate waived confiscation of the money Li had received. The second of our three Ba County cases involves an open wife sale followed by two supplementary payments; after being denied a third payment, the seller filed false charges. In the late winter of 1856–57, just before the New Year, Peng Jintang sold his wife, Chen Shi, to Shen Tingyuan. The transaction included the couple’s young son Shouxi, who was adopted by Shen. Here is the text of the contract for both Chen Shi’s sale and Shouxi’s adoption: I, Peng Jintang, hereby establish this contract to divorce and marry off my wife and to adopt out my son, promising that I will never change my mind or attempt

200

part two

extortion. The situation is as follows. I took Chen Shi in marriage as my wife, and she bore me a son, Shouxi. But because of poverty I cannot support them, and I have no relatives in my lineage or by marriage on whom we can rely, so we have both voluntarily agreed to separate in order to survive. I have asked Yan Hengtai to act as matchmaker to negotiate an agreement to sell my wife to Shen Tingyuan to be his wife. We have agreed on a brideprice of exactly 5000 cash, which I have received in full. . . . I hereby adopt out my son to follow his mother, and am willing to let Shen Tingyuan change my son’s name and raise him as his own, he may not return to my lineage. The future cannot be foreseen, but I may not attempt extortion. No one of either surname, including those now present and those not present, may cause trouble about this agreement. If someone should try to interfere, then I shall take full responsibility and am willing to answer for my crimes. I promise not to let any trouble affect the Shen family. There has been no coercion to force me to divorce and marry off my wife. Fearing that words alone are unreliable, I hereby establish this contract for divorcing and marrying off my wife, and for adopting out my son, combining both transactions in a single document, to serve as proof. [left handprint] This handprint serves as proof. Witnessed by [six men, including scribe] together present. Done on Xianfeng 6.11.10, by the person who establishes this contract to divorce and marry off my wife, and also to adopt out my son, Peng Jintang [mark], and by the one adopted out, Shouxi [mark]

It is striking that Shouxi’s name and mark appear on this document along with his father’s, because it appears that he was only a small child. Note that the opening line of this contract promises that the seller “will never change my mind or attempt extortion,” a promise that is repeated later in the text. Another line stands out as especially significant, given what happened later: “The future cannot be foreseen, but I may not attempt extortion.” This euphemistic statement pertains to the seller’s son, Shouxi: no matter what his fate may be, the seller will not try to exploit it for gain. Similar language in adoption contracts bars the birth parent(s) from taking action against the adoptive parent(s) should the child fall ill or die. The couple also had a little daughter, Chunxiu, but Shen had refused to adopt her. Peng Jintang used the girl as a pretext for demanding his first supplementary payment just a week after the sale had been finalized. Peng sent his father, Peng Yifa, with the little girl to visit Shen and press him to take the girl. When Shen again refused, Yifa demanded more money and made a big scene, threatening to abandon the girl at Shen’s home. Finally, Shen summoned the matchmaker and, before witnesses, agreed to pay an additional 1000 cash in exchange for the following document: I, Peng Yifa, hereby establish this contract to take custody and never again cause trouble, in order to prevent future disaster. The situation is as follows. My second

Chapter 6. Negotiations between Men in Wife Sales

201

son, Peng Jintang, at first was married to Chen Shi for several years, and they had one son and one daughter. But recently it has been our fate to fall on hard times, so that we have difficulty getting by. At the beginning of this month, we entreated our relative by marriage Yan Hengtai to act as matchmaker, and he arranged to remarry Chen Shi to Shen Tingyuan as wife. At that time, through the matchmaker it was agreed that Shen Tingyuan would pay exactly 5000 cash, which he paid over in full and we received in person with our own hands. It was originally agreed that Jintang’s son Shouxi would follow his mother and be adopted by Tingyuan as his son, to carry on his family line. But as far as Jintang’s daughter is concerned, Jintang would raise her himself. But a few days ago I, Yifa, took this daughter Chunxiu and delivered her to Tingyuan’s house in order to cause trouble. So once again, relying on the matchmaker and witnesses, an agreement was reached whereby they urgently advised Tingyuan to give me another 1000 cash, so that I would agree to take Chunxiu home to raise her ourselves. Having taken custody of her, my son and I promise never again to cause trouble or change our minds. We both agree willingly and of our own accord, without any coercion or deception. Fearing that spoken words alone are unreliable, I, Peng Yifa, specially establish this contract for taking custody. [mark] Witnessed by [seven men, including matchmaker Yan Hengtai, the scribe, and four of the same men who witnessed the wife sale contract] together present.

The contract bears Peng Yifa’s name and mark and is dated Xianfeng 6.11.20. Its language is less humiliating than that found in the previous example, although it does confirm that the Pengs used the child to practice extortion. Once again, we have a crowd of the buyer’s friends and neighbors acting as witnesses (seven men in all), including the original matchmaker and four of the men who had witnessed the wife sale contract. Here we see an example of how wife sales were regulated at the community level, and how disputes were resolved through mediation, without resort to the courts. Shortly thereafter, the son who had been adopted out, Shouxi, suddenly died of an illness. A couple days before the New Year, Peng Jintang visited Shen, made a scene (blaming Shen for his son’s death) and demanded more money. After some wrangling, Shen again summoned the matchmaker and witnesses and finally paid Peng 600 cash, in exchange for the following contract of submission: I, Peng Jintang, hereby establish this contract of submission promising that I will never again cause trouble. The situation is as follows. Last month, I asked Yan Hengtai to act as matchmaker to marry my wife, Chen Shi, to Shen Tingyuan as wife. I received 5000 cash and drew up a marriage contract as proof. Several days later, my father extorted another 1000 cash, and there is another submission contract as proof. Originally my wife and I had one son and one daughter, and when I married off my wife it was agreed that I would raise our daughter myself, but our son would follow his mother and be adopted by Shen Tingyuan, and there is a contract to prove it. But my son was ill, and unexpectedly he died early this month of his illness. When I heard this news, I once again went to Tingyuan’s home, taking advantage of the

202

part two

situation as an excuse to harass him and extort money. Tingyuan asked the collective membership of the tuan48 to talk reason to me, and they all scolded me for my wrongdoing and were about to turn me in to the authorities. But I myself realized that I was out of luck and my demands had no basis, and I feared the law and hoped to avoid being turned in. I repeatedly apologized and bowed, and so the tuan membership urged Tingyuan to give me 600 cash, which I personally received. I am perfectly willing to establish this contract of acknowledgment for them to keep, and I promise that in the future I will never again make excuses to cause trouble. If I repeat my previous behavior, then Wu Hengzhong will take responsibility. Tingyuan may take this contract and report to the authorities to have me prosecuted and punished, and I will accept punishment without complaint. This is done of my own free will, without any coercion or deception. Fearing there will be no proof, I specially establish this contract of submission and acknowledgment to be given to Tingyuan to save as proof. Witnessed by [eight men, including matchmaker Yan Hengtai, Wu Hengzhong, and the same scribe as in the previous document] together present.

This submission contract bears Peng Jintang’s name and mark and is dated Xianfeng 6.12.28. It is not clear who Wu Hengzhong was, but he played a key role in persuading Shen to make this payment by guaranteeing Peng’s future good behavior. This document is striking for its rhetorical groveling. Nothing could be clearer than Peng Jintang’s repeated promise not to harass Shen—nevertheless, less than a month later Peng returned to demand more money a third time. This time, however, Shen refused to pay, so Peng filed false charges at the subdistrict yamen.49 Shen was summoned by yamen runners, but no hearing was held because Peng failed to appear—he had had no intention of actually pursuing the lawsuit—but then Shen called his bluff by filing counter-charges against Peng at the main Ba County yamen, submitting the wife sale contract and both submission contracts in evidence. The magistrate’s runners hunted down Peng and arrested him. The magistrate canceled both marriages and returned the woman to her natal family for remarriage. He ordered Peng beaten for selling his wife, practicing extortion, and filing false charges, but he did allow Peng to keep all of the money he had received. The magistrate reprimanded Shen for buying “a woman who has a living husband” but declined to punish him further.50 In the third Ba County case, from 1808, we find that the seller’s demand for a second supplementary payment prompted mediators to force him to back down— at least temporarily. In this case, an impoverished peasant named Lu Yonglin decided to sell his wife, Liu Shi, so that he could clear his debts. The couple had left their village because of poverty and come to Chongqing hoping to rendezvous with a relative and find work. This plan fell through, so they headed for home, stopping at a market town where Lu found temporary work hauling coal in

Chapter 6. Negotiations between Men in Wife Sales

203

exchange for food. They ran up a bill at the inn where they were staying and had no money, so finally, in desperation, Lu decided to sell his wife. Lu gave a “request certificate” to an acquaintance named Ge Xianlong in order to persuade him to act as matchmaker, and Ge used this document to negotiate Liu Shi’s sale as concubine to Yi Quanwan for a brideprice of 5 taels. The transaction took place just before the New Year. After paying his debt, however, Lu had no money left to pay his way home, so after a week he approached Yi to demand more money. Yi reluctantly agreed to give Lu an additional 1000 cash, in exchange for the following “acknowledgment contract”: I, Lu Yonglin, hereby establish this contract of acknowledgment. The situation is as follows: Because of poverty we had no means to secure our daily food, so I voluntarily agreed to marry my wife off to Yi Quanwan as concubine, before matchmaker and witnesses. At that time, I drew up a handprint contract to serve as proof, and I received the entire brideprice in full. After that, however, I spent all of the money . . . and I had no money left to pay my travel expenses to return home. So I begged Chen Wenhui and other neighbors [of Yi Quanwan’s] to persuade him to give me 1000 cash more. From this day forward, I can never again make pretexts to cause further trouble. If I ever say differently, then I acknowledge that the witnesses will take this contract and report me to the authorities for punishment, which I agree to suffer without complaint. Now, because people’s hearts are false and changeable, I hereby establish this contract of acknowledgment and give it to Wenhui to serve as proof. Witnessed by [six men, including matchmaker Ge Xianlong and scribe]

This document bears Lu Yonglin’s name and mark and is dated Jiaqing 12.12.12. But Lu was not satisfied and he did not return home—just six days later, he got drunk and in the middle of the night visited Yi’s home, where he made a scene, demanded more money and, when rebuffed, insisted that Yi return “his” wife. Yi and his neighbors seized Lu and threatened to take him to the yamen and press charges. This threat seems to have sobered up Lu, and the next day he issued the following “contract of regret and submission” in exchange for being released: I, Lu Yonglin, hereby establish this contract of regret and submission. The situation is as follows: I was married to Liu Shi for several years, but because of drought and poor harvest [we fell on hard times and left home]. It became difficult to secure our daily food. We were on the road and I discussed the situation with my wife. We agreed to separate in order to survive, so we asked Ge Xianlong to act as matchmaker to divorce my wife Liu Shi and marry her to Yi Quanwan as concubine. The wife sale price was exactly five taels of silver, which Yi Quanwan paid in full. There is a marriage contract and a submission contract as proof. Later on, I spent all the money and late at night on the 18th I got drunk and went to Yi Quanwan’s home and caused a disturbance at his doorway. Mr. Yi called all the neighbors and they seized me and were ready to take me to the authorities. When I sobered up I realized with fear that

204

part two

I could not escape the law, so I begged Hu Dachao and the others to submit this contract of regret and submission on my behalf to the market town headman (chang shou), so that I could avoid punishment and trouble. If I ever again come to this market town and cause the slightest trouble, then the neighbors here will take this contract and report me to the authorities for prosecution, and I will not deny my crimes. Fearing that later on there will be no evidence, I hereby give this contract of regret and submission to the market town headman as proof. Done, on Jiaqing 12.12.19, by Lu Yonglin, the person who establishes this contract of regret and submission [mark] Witnessed by neighbors [at least nine men, including matchmaker Ge Xianlong, Hu Dachao, and two others who also witnessed the previous “acknowledgment certificate,” plus one who witnessed the “request certificate” Lu originally gave to Ge] together present

Once again, note the large group of the buyer’s neighbors, as well as the original matchmaker, who acted to impose and witness this agreement. This submission contract did not serve as a receipt, because Lu did not get any money; its purpose, rather, was simply to document Lu’s shameless behavior and to deter him from harassing Yi Quanwan anymore. But instead of going home, Lu raised the stakes by returning to Chongqing and filing charges against Yi and the matchmaker, accusing them of taking advantage of his illiteracy to trick him into selling his wife and “forcing me to put my handprint” on the wife sale contract. If Lu hoped to intimidate Yi into paying him, his gamble backfired, because Yi presented the handprint contract and the two submission contracts in court and reported that he had bought Lu’s wife in a voluntary open sale. When Lu Yonglin was hauled into court and confronted with these documents and with the testimony of the other principals to the transaction, he admitted that he had voluntarily sold his wife. The magistrate canceled both marriages and returned Liu Shi to her natal family; he allowed Lu to keep the money but ordered Lu, Yi, matchmaker Ge, and another witness to the sale all beaten.51 Sellers in land sales who received supplementary payments would often prepare submission contracts similar to those used in wife sales (an alternative was to add an amendment to the original sale contract). The parallels between these two documentary forms further underscore the homology between land and wives that we have already discussed. The following example from Taibei County, Taiwan, dated 1878, is written in the name of a widow and her two sons: I, Li Deng Shi, together with my sons Yushu and Decheng, hereby establish this contract to beg for a supplementary payment (zai ken zhao xi zi). We had land left to us by our husband/father [here the land is enumerated in detail]. All of this land was recorded clearly in the sale contract and in years past we already sold all of it by absolute sale to Chen Yiji and Chen Yirong. The contract was written, the price paid in full, and our connection to the land severed, so not a single inch of the land

Chapter 6. Negotiations between Men in Wife Sales

205

remains in our possession anymore. We would not dare to bring up the topic of supplementary payment in order to cause trouble, but we have recently fallen on hard times and we have no means of getting by and nowhere we can borrow money. . . . Therefore, we have no choice but to ask the middleman to beg Chen Yiji and Chen Yirong humbly on our behalf to give us a supplementary payment of 15 large foreign silver dollars. On this same day, we have personally received that amount in full. Henceforth, I and my sons and even my grandsons, no matter how poor or deprived we may be, will absolutely never again dare to beg for further supplementary payments or otherwise cause trouble. Both sides are acting voluntarily of our free will, and neither will have any regrets. Fearing that spoken words alone are not proof, we establish this contract to beg for a supplementary payment and give it to [the Chen brothers] to serve as proof.

The contract is “signed” by the middleman, Li Deng Shi, and her two sons with their marks (in Li Deng Shi’s case, a fingerprint).52 In this document, Li Deng Shi and her sons explicitly acknowledge that they had sold their land in an absolute sale and had been paid in full, and that under normal circumstances, they “would not dare” ask the Chen brothers for more money. They certainly had no legal standing to do so: Qing lawmakers never accepted the practice of supplementary payments even in conditional land sales, let alone absolute sales, and they repeatedly tried to suppress it (although without success). But there is no evidence in this document of any extortion or threats, let alone litigation. The crucial point, conveyed by the entire rhetorical posture of the text, is that this widow and her sons are to be pitied because they were forced to sell ancestral land and yet now are penniless, and the Chen brothers, having benefited from their misfortune, bear a moral obligation to help. Their agreement to do so is portrayed as noblesse oblige. Nevertheless, as in wife sales, their charity is conditioned on the petitioners’ promise that they will never “dare” to ask for another supplementary payment. Fear of extortion and legal trouble was certainly an important factor in buyers’ willingness to make supplementary payments, but we should not imagine it to be the only one, in wife sales any more than in land sales. The evidence indicates a widespread, if grudging, sympathy for sellers as the pathetic losers in these transactions—a sympathy shared even by many buyers. An 1813 case from Chongxin County, Gansu, records intriguing testimony about the circumstances in which peasant Zhao Chengdu managed to extract two supplementary payments from Guo Kuan, who had bought Zhao’s wife, Xu Shi, the year before. Here is the testimony of Li Xun, a neighbor of Guo’s who helped mediate between the two men: In Jiaqing 17.10, a man I didn’t recognize came to our village and argued with Guo Kuan, demanding money. A couple of other neighbors and I went over and separated them. On questioning them, we learned that this man was named Zhao Chengdu

206

part two

and was the former husband of Guo Kuan’s wife, Xu Shi. Zhao said that Guo Kuan had bought his wife Xu Shi and taken her away, but now Zhao had no money left, so he wanted Guo Kuan to help him by giving him some more money. Guo Kuan said “We used a matchmaker and drew up a marriage contract, and I paid 52 taels for her body price to buy her—that body price was a lot of money, but still you come and harass me!” Then Guo Kuan got out the contract and asked us to look at it, to prove that what he said was true.

Upon being confronted with the marriage contract before witnesses, Zhao Chengdu “had nothing more to say.” The contract clearly stated that Zhao had been paid in full and was owed nothing more. But the neighbors who had intervened pitied Zhao and urged Guo not to send him away empty handed: “We urged Guo Kuan to help Zhao Chengdu by giving him 1200 large cash, so he did. After receiving the money, Zhao Chengdu said nothing more and left.” Moreover, just one month later, when Zhao came back and demanded another supplementary payment, the neighbors took pity on him once again: “We urged Guo Kuan to give Zhao Chengdu 600 big cash more, and so he did.” Why did Guo’s neighbors pressure him to help Zhao, not once but twice? Perhaps they feared the consequences of not paying him: if Zhao were truly desperate, who knew what kind of trouble he might cause? Better to pay him, so he would leave quietly. But one senses more than a little sympathy for Zhao, a famine refugee who had migrated more than 200 kilometers from home, looking for work and begging, before finally being compelled to sell his wife. Certainly, his sold wife Xu Shi felt compassion for him: on Zhao’s second visit, when she saw that Guo Kuan had given him only 600 cash, she slipped Zhao another 200 cash behind Guo’s back. Moreover, it seems significant that Guo felt obliged to submit to his neighbors’ judgment and pay the sums they suggested. This episode illustrates the effect of peer pressure through mediation to enforce community norms.53 The Threat to File Charges When sellers demanded supplementary payments, they often threatened to file charges—and as we have seen, this threat alone might induce buyers to pay. Occasionally, when rebuffed, sellers actually did file charges, although their intent seems to have been to pressure the buyer, rather than to pursue a lawsuit all the way to a formal hearing and judgment. As the three Ba County cases just examined illustrate, however, this tactic could backfire badly. No one involved in a wife sale had anything to gain from submitting a dispute to formal judgment, unless the goal was to cancel the sale. Therefore, I assume that most disputes over supplementary payments never ended up in court but instead were mediated locally in the manner we have seen. The cases that ended up being pursued all the way to formal judgment were especially unusual.

Chapter 6. Negotiations between Men in Wife Sales

207

One reason it was risky to file false charges is that in these cases, at least, magistrates usually recognized them to be false. These plaints tend to be sensational and patently incredible. Moreover, they follow a stereotyped pattern, so that after reading a few, one learns what to expect. No doubt any magistrate or legal secretary with a modicum of experience would have recognized such charges for what they were. Here is a representative sample of skeptical rescripts written by six different magistrates on plaints in which men who had sold wives falsely accused the buyers of adultery, kidnapping, rape, and other crimes. They come from Ba and Nanbu counties and range in date from 1802 to 1878: It is obvious that first you sold your wife yourself, then you tried to extort more money but failed, and that is the reason you are now filing these false charges! (ming xi ziji jia mai, jin yin suozha bu sui, gu you shi nie).54 It is obvious that you colluded to sell your wife in marriage yourself, and now you are filing false charges because your subsequent attempts at extortion have failed—this is nothing but blackmail! (xian xi chuantong jia mai fu xiang xusuo busui nie ci wang kong, shu shu guizha).55 The claims in your plaint make no sense and are obviously fabrications. Await your summons and interrogation: if it turns out that you sold your wife’s divorce yourself, and later changed your mind and attempted extortion (mai xiu zhi hou fanhui tu suo), then I shall impose severe punishment without fail!56 If it turns out that you have taken cash to sell your wife’s divorce, and now are filing false charges in an attempt to extort more money, then you will be severely punished—there will be no lenience!57 The claims in your plaint are slippery and evasive. Most likely in fact you yourself colluded to sell your wife’s divorce, and later concocted these outrageous, sensationalist charges because your attempts at extortion failed!58 It seems likely that in fact you received money for selling your wife in marriage. . . . Your plaint makes no sense at all, and obviously is full of falsehoods. Await your summons so that the true story can be discovered through interrogation.59

These rescripts show that magistrates had a lively awareness of the practice of demanding supplementary payments after wife sales, and also of the tactic of filing false charges in order to raise the ante (see also the epigraph at the beginning of this chapter). They also make crystal clear that a plaintiff would be unwise to press such a suit to the bitter end. If no one stood to gain from pursuing a case all the way to formal judgment, then why did some people do so anyway? Why did mediation fail? Individual personalities no doubt played a role, as did anger and spite, as well as miscalculation and sheer stupidity. In an analogous context, Mnookin and Kornhauser make the following observation:

208

part two

One or both parties may be motivated in substantial measure by a desire to punish the other. . . . If the parties get heavily engaged in strategic behavior and get carried away with making threats, a courtroom battle may result, despite both parties’ preference for a settlement. Negotiations may resemble a game of “chicken” in which two teenagers set their cars on a collision course to see who turns first. Some crack-ups may result. [Also,] litigants typically overestimate their chances of winning.

Another reason mediation fails is that sometimes there exists “no middle ground on which to strike a feasible compromise.”60 By definition, legal cases usually record only people who got in trouble. At the risk of tautology, I would suggest that the sellers and buyers encountered in wifeselling cases tended to be unusually fractious, obstinate people. Otherwise, they would have found some reasonable way to compromise and would have left no trace in the archives. The case records are disproportionately populated by stingy buyers who refused to make even a single supplementary payment, and by greedy sellers who persisted in demanding more money, no matter how much they had already received. Calling the Seller’s Bluff We have discussed the seller’s sense of entitlement, his belief that the transaction had created a relationship between himself and the buyer that obliged the latter to help him. To some extent, the wider community shared this view. But there was a limit to how far buyers would indulge this sense of entitlement. I have fifty-one cases in which we know supplementary payments were made before a dispute came to court (see Tables 4 and 5 above). In forty-two of those cases, at least one payment was made; in eight cases, at least two payments were made; and in one other case, four payments were made—the largest number in any single case. Each of these cases ended up in court because the buyer continued to demand more money after the seller had reached his limit. These data suggest that many buyers were willing to make a single payment, more or less grudgingly, but only a few would go farther. A basic reason, no doubt, is that most buyers were men of very humble means, and there was a limit to how much they could pay. Moreover, the vast majority of these payments occurred within one year of the sale. Demands made later were more likely to be rejected, suggesting that the more settled and secure a buyer became in his new marriage, the less sense of obligation he felt toward his wife’s first husband. It seems likely, therefore, that if a seller would settle for a single supplementary payment, then the matter would never appear in court and hence would leave no trace in the historical record. Conflict flared when sellers refused to accept that there were limits to how much they could expect buyers to pay. The buyer would call the seller’s bluff, by meeting him in court—or, as in the exceptional cases

Chapter 6. Negotiations between Men in Wife Sales

209

recorded in xingke tiben, he might even murder the seller in order to eliminate what had become an intolerable burden. That is how the case of Guo Kan and Zhao Chengdu ended. In a previous section of this chapter, we saw how Guo’s neighbors mediated two supplementary payments to appease Zhao. But Zhao was insatiable, and he continued to pursue Guo even after the latter had moved his family to a different county in an attempt to escape. Zhao demanded 2000 cash that Guo did not have. As Guo later testified, He said that unless I gave him the money, he would never leave me alone. When I heard this I thought to myself, here we are at the end of the year, and I have no money left to give him. But still he comes, again and again, to harass me like this! I was really angry, and suddenly I decided to do him in, in order to avoid suffering his harassment again in the future.61

Similarly, in the case of Yi Shangsong that opened Chapter 4, we find this testimony from Zhou Wenxiao, who had bought Yi’s wife: I knew that Yi Shangsong was an extremely impoverished and shameless guy, and I feared that if I did pay him what he asked, then he would just keep trying to extort more money from me, and there’d be no end to the trouble and harm I’d suffer from him. So I decided to kill him, in order to get rid of this problem for the future.62

These comments show us the reverse of the seller’s sense of entitlement, namely the buyer’s realization that no matter what he did for the seller, it might never be enough. The following case from Zhejiang, reported in 1798, draws together a number of themes of this chapter. Zhou Shuangquan and his wife, Ji Shi, lived with Zhou’s father, Zhou Laoyi (70), in Sheng County. Landless peasants, their household was very poor; the men worked as agricultural laborers, and Shuangquan also left home for regular stints of begging. Finally, in the summer of 1795, when they were having a particularly hard time, Shuangquan sold his wife to an acquaintance named Yu Laochuan (another landless peasant who, at the age of 35 sui, had never married) for a “body price” of 13,000 cash. It was actually Zhou Laoyi’s idea to sell Ji Shi, but Shuangquan felt he could not oppose his father’s wishes. Six months later, just before the New Year, Zhou Laoyi visited Yu and demanded 10,000 cash. When Yu refused, Laoyi threatened to file charges against him, so Yu reluctantly gave him 2000 cash in exchange for “a submission contract promising that there will be no further supplementary payments.” Five months later, Zhou Laoyi broke this pledge by demanding that Yu pay 8000 cash more (that is, the balance of the 10,000 that Yu supposedly owed him). But Yu had no money left, so he refused to pay. But “I was harassed by him until I could not stand it anymore” (bei ta chanrao bu guo), and finally, in order to escape, Yu took Ji Shi to Dongyang County, where he found work on a managerial farm. After a month, however, the

210

part two

Zhous tracked the couple down and Zhou Laoyi continued badgering Yu, making public scenes and being an embarrassing nuisance. According to Yu, “finally, feeling that I had no choice, I promised to pay him, but I asked him to give me some more time.” But within a month, Laoyi renewed his demands and threatened to commit suicide and implicate Yu unless he paid immediately. “I hated Zhou Laoyi for constantly trying to extort money from me”—so, with the help of friends, Yu murdered him.63 Yu Laochuan had acted in good faith, and one supplementary payment had exhausted his resources. He tried to evade Zhou Laoyi but failed, and it finally dawned on him that no matter how much he paid the Zhous, he might never be rid of them. This is the flip side of the seller’s conviction that wife sale had bound the buyer in obligation to him: from the buyer’s perspective, their relationship had become an open-ended liability. I N T H E SHA D OW O F T H E L AW

Despite being prohibited, wife selling was far from rare, and it occurred all over China. Uniformity of practice shows that such transactions proceeded according to well-known customary rules and norms. But criminalization was not a trivial factor. On the contrary, the law always loomed in the background, influencing negotiations between sellers and buyers and their respective calculations of self-interest. Criminalization meant that contract enforcement and dispute resolution had to take place at the community level, without resort to the courts. However, buyers’ fear of prosecution made them vulnerable to extortion. Qing prohibition made the twin risks of prosecution and extortion inherent to wife sales, thereby helping to depress prices for “wives with living husbands.” Ironically, however, the low prices that sellers had to accept only strengthened their sense of entitlement, making demands for supplementary payments all the more likely. Faced with such demands, buyers had to weigh the cost of paying off the seller (and the risk that, if encouraged, he would return to demand even more) against the real danger of losing everything in a formal court judgment. Supplementary payments signified far more than just fear and opportunism, however. They reveal a homology between wives and land in the patriarchal peasant mentality also seen in everything from contractual forms to folk proverbs and taboos. They also reveal a widespread sentiment that men who sold wives deserved pity as the losers in these transactions. Nevertheless, criminalization shaped wife sales in predictable ways, even those that did not end up in court: at every stage, negotiations proceeded “in the shadow of the law.”

7

Wives, Their Natal Families, and Children [My husband] said to me: “You have a hard time with me; the children are thin, and you are miserable; it would be much better for you to be married to some kind man who would give you enough to eat. I will find such a one, and marry you to him, [along with our children].” . . . I assented to this, for I saw that the children would otherwise starve. . . . I did not cry at all when he left me, for I thought I could be no more wretched than I had been with him. — recollections of a woman from shantou, guangdong, as recounted by american missionary adele fielde in 1887 1

For a husband to sell his wife would seem to be the epitome of patriarchal exploitation and abuse. For this reason, the May Fourth radicals invoked wife sales as part of their powerful stereotype of “the victimized woman in old China,” who stood for everything wrong with the culture and society of the ancien régime.2 No doubt, some women who were sold did suffer exploitation and abuse. But, as with polyandry, most of the evidence about wife sales suggests something both more complex and more interesting than simple victimization. Consider the following testimony from a peasant woman named You Shi (24), which appears in a 1756 case from Changning County, Jiangxi: “I saw that my husband, Luo Fenghua, really could not support a wife, and that if I stayed with him I would end up starving to death. I realized that it would be better for us to go our separate ways and try to survive separately. So I agreed with his proposal to sell me in marriage.” Several months after the sale, You Shi’s first husband Luo Fenghua tried to take her back from her buyer, Liang Yunfa. But Liang refused to countenance Luo’s demands, and You Shi (who by this time was pregnant with her new husband’s child) felt no sympathy or loyalty toward Luo. As she later testified: While I was with Luo Fenghua, I was cold and hungry most of the time (duo shi dong e guo ri). When he decided to sell me in marriage, I was perfectly willing to leave him (xin’gan qingyuan). Ever since I married into Liang Yunfa’s household, I have lived 211

212

part two well. I never missed Luo Fenghua at all—when he sold me, my marriage bond with him was completely broken (en duan yi jue le).3

It is obvious from You Shi’s testimony that she played an active part in the decision and, moreover, that “being sold” had dramatically improved her quality of life. Far from feeling exploited or victimized, this woman felt lucky to have left her first husband for a second who could provide security and comfort. Her utter lack of sentimentality reflects the same pragmatic attitude toward sex and marriage that we have seen expressed by the women in many polyandry cases. Wife-selling cases reveal a spectrum of attitudes on the part of the women who were sold, but the basic story they tell is straightforward: as a practical matter, it was difficult for a wife sale to succeed without the wife’s cooperation. Therefore, in most of the sales that actually took place, the women themselves were willing and even active participants. Often, the wife’s attitude was decisive. Most surviving wife sale contracts explicitly record that the woman had agreed to the transaction, as had her natal family.4 Not all of these women felt enthusiastic about the prospect, but, like Adele Fielde’s informant (quoted in the epigraph), they believed that at least they had nothing to lose, and many, like You Shi, could expect an improved standard of living. In addition, there are many examples of women successfully preventing or sabotaging sales if they did not want to be sold. In most of these cases, the question of agency needs to be understood from at least two angles. Wife-selling was generally a family strategy, undertaken to solve family problems, but there was also opportunity for the wife to assert herself individually at a number of points in the process. In the most basic terms, it was an opportunity for individual women to pursue their own interests by escaping difficult situations in favor of potentially more secure ones. A husband might actually be more reluctant than his wife to take the drastic step of selling her, and indeed, it was sometimes the wife who demanded that her husband sell her so that she could have a chance for a better life in a different household. A man who had sold his wife was seen as the loser in this transaction, because he would thereby sink into the underclass of surplus males outside the family system. It was unlikely that he would ever recoup the resources necessary to marry again. Moreover, women often took their children with them into their new marriages, in which case it is the sellers who would be left alone, and disputes over child custody exposed in sharp relief the latent conflicts of interest between patriarchal and uterine families.5 Even after agreeing in principle to be sold, a wife might exercise a veto over subsequent negotiations if she felt they did not suit her interests. Most important, if the wife objected vehemently, the prospective buyer would probably withdraw, because he did not want legal trouble—nor did he want to marry a shrew, much less end up with an angry and vengeful ghost on his hands. Some women were

Chapter 7. Wives, Natal Families, and Children

213

forcibly sold against their clearly expressed wishes, but such cases are unusual, and they must be set against the many other examples in which women exercised an effective veto over whether or to whom they would be sold. If a wife were unhappy about the situation, for example, she could ask her natal family to go to court and file charges to prevent or cancel the sale. Sometimes a woman would run away to the yamen and petition the magistrate herself, but it was more effective to have her natal family act on her behalf, in order to avoid the appearance of impropriety that would attend filing charges against her own husband. Different versions of this scenario account for about a third of my county-level cases. I do not want to exaggerate the degree of agency enjoyed by women in these situations. When wives were sold, children accompanied them about 60 percent of the time—meaning that, 40 percent of the time, such women were separated from their children, and when wives resisted being sold, a frequent reason was refusal to leave their children. Going to court was no panacea: if a woman was divorced from both husbands and returned to her natal family by court order (part of the mandated penalty for “buying or selling a divorce”), the inevitable result was separation from her children, because a basic principle of patrilineal descent, codified in law, was that children belonged with their fathers. Thus, in many instances when a magistrate appears to be protecting a woman by returning her to her natal family, he is simultaneously severing her connections to her children.6 Moreover, a woman’s ability to assert her interests was circumscribed by her social context, and if she found herself isolated, without resort to natal kin or other allies, she might well be vulnerable to coercion and violence. W I V E S W HO D E M A N D E D T O B E S O L D

Let us begin at one end of the spectrum, with wives who actually pressured their husbands to sell them. If a woman wanted out of her marriage, what resources or strategies might she deploy to achieve that end? Consider the following 1759 case from Ba County, in which an unhappy wife succeeded in making her husband sell her to another man in marriage. Liu Xiangzhu (25) had been married to Luo Shi (25) for six years. According to Liu, She was disrespectful to me and unfilial toward my parents. She was always running away to her natal home and would disappear for days on end. One time, her older brother came over and beat her and scolded her for me, but she paid no attention to him. She would make a fuss and tell me I should marry her to someone else. She threatened to cut her throat or hang herself, and then she cut off all her hair. Finally, I had no choice but to marry her off.7

The case records invariably portray such women in a highly pejorative manner that draws on stock stereotypes of the bad wife.8 But one can read these accounts

214

part two

of “bad wives” like Luo Shi against the grain, for insight into their motives and subject positions. There was a straightforward, strategic logic to their behavior: a woman’s best strategy to make her husband release her was to refuse to play the expected role of a wife. After all, in the end Luo Shi got what she wanted: a divorce and a new marriage. Bad Wives on Strike Women who wanted out of their marriages would go on strike. They would refuse to perform household chores or to have sex with their husbands. They would refuse to get out of bed in the morning, or at night if their husbands came home late and wanted supper. Alternatively, a wife might get out of bed in order to burn the food on purpose so that she could present her husband with an inedible meal. Such behavior would be disparaged in male testimony as “laziness,” “disobedience,” and “defiance,” but it was calculated. When wives went on strike, they would harass their husbands verbally. In male testimony, this harassment was characterized as “making a fuss and scolding” (rang ma), “shrewishly making a scene” (po nao), “quarreling and fussing” (chaorang, chao nao), “complaining” (maiyuan), and “cursing” (zhouma). A wife might threaten to commit suicide, or even “to fight you to the death” (pinming), and she might punctuate such threats with dramatic gestures like cutting off her hair. Wives would perform this harassment as noisily and publicly as they could, so that the extended family and their neighbors and even village authorities would soon learn of it. In fact, they might go out in the street or to the village well and complain loudly to the community—especially to the community of women whose collective gossip and pressure might serve as a rough and ready check on objectionable male behavior.9 The purpose of such publicity was to humiliate and exhaust their husbands to the point of surrender. Unhappy wives would also leave home without permission, usually taking refuge with natal kin, who had a recognized role in protecting married-out daughters from abuse. They would complain bitterly to their natal families about their husbands’ poverty and incompetence; often they would stay away long enough to force their husbands to undergo the humiliation of having to fetch them home. Regardless of why a wife was unhappy, her natal family constituted her greatest resource and strongest ally against her husband and in-laws. Finally, unhappy wives would demand that their husbands divorce them or sell them in marriage. Faced with such a performance, many men would conclude, “A woman like this— what do I want her for, anyway?” (zheyang nüren, yao ta zuo shen). That is precisely the conclusion their wives wanted them to reach. To be sure, such strategies entailed a degree of risk, and some wives ended up being beaten or even murdered by their husbands. Homicide is a highly exceptional event, however, and it is far more likely that a husband would agree to sell

Chapter 7. Wives, Natal Families, and Children

215

an unhappy wife than that he would kill her. But the special requirements for reporting a wife killing up the judicial hierarchy make those case reports particularly useful for gaining insight into the mentality and strategies of unhappy wives. The main crime reported by most xingke tiben is homicide. The xingke tiben held at the First Historical Archive in Beijing are organized in broad categories according to the background situation that framed the main crime: family, debt, land disputes, peasant uprisings, and so on. The cases collected for the present study all come from the category “marriage and sex offenses.” In this category, the most common scenario of homicide is the killing of wives by their husbands, nearly always for some perceived failure of wifely duty: disobedience, defiance, refusing sex, unfilial treatment of parents-in-law, leaving home without permission, adultery, and so on. These cases include many examples of wives who were trying to provoke their husbands into letting them divorce and remarry—some flatly demanded to be “sold in marriage” (jia mai). The standard sentence for a husband who killed his wife was “strangulation after the assizes,” which meant that each sentence would be reconsidered in light of mitigating or aggravating circumstances at the annual Autumn Assizes (see Appendix D.3).10 In cases of wife killing, the Autumn Assizes rules mandated that, if a wife had provoked her husband with “defiant” or “unfilial” behavior, then his sentence would be commuted and his life spared. For this reason, xingke tiben reporting such cases would detail any misbehavior on the wife’s part that might have provoked him, and particular attention was paid to aggravating words or deeds that immediately preceded the homicide.11 The critical judicial importance of this evidence—because the husband’s life hung in the balance—required that it be recorded with an unusual degree of explicitness and precision (in contrast with the paraphrasing and standardization that characterizes much of the testimony “transcribed” in Qing case records). This was generally true of the “fighting words” that provoked homicides, but especially so in cases of wife killing, given the criteria for commuting sentences at the Autumn Assizes. The result is something close to a verbatim transcript of what a woman actually said, as recounted by her husband and other witnesses. Consider the following three examples, all taken from wife-killing cases. In a 1796 case from Pianguan County, Shanxi, peddler Zhang Fushouzi (25) killed his wife, Zhang Luo Shi (19), because she constantly complained about their poverty and demanded that he sell her in marriage. One night Zhang came home late from a long day of peddling and found his wife already asleep. He told her to wake up and make him supper, but she ignored him. He scolded her for her laziness, and she started cursing him: My wife wouldn’t back down, she yelled at me that she had already had enough suffering, she cursed me for coming back so late and asked why didn’t I just die out there

216

part two instead of coming home. I cursed her back, and she sat up and said that my household was so poor, she wanted me to sell her off in marriage, because she wasn’t willing to live with me anymore.12

In a 1797 case from Luochuan County, Shaanxi, a hired laborer named Liu Xinxian (30) beat and accidentally killed his wife, Chao Shi (18), because she complained about poverty and hunger and demanded that he sell her in marriage: That night I had already gone to bed, but my wife just sat there on the kang. Once again she complained that I had no work, and that I could not support a household and keep her alive, and she sat there crying. Then she told me that I should sell her in marriage. I sat up and argued with her, but she just complained and scolded me even more bitterly.13

Finally, in a 1798 case from Heyang County, Shaanxi, a landless peasant named Ma Zhenfu (39) killed his wife, Xing Shi (23), because she complained about their poverty, refused to have sex, and demanded that he sell her: Because we were really poor, and because I am stupid and clumsy, my wife would never get along with me, and she wasn’t willing to sleep with me (buken tong su). . . . She often quarreled with me and pressured me to sell her in marriage to some other man. . . . That night, I finished work and came home and saw that my wife had already gone to bed. I lit the lamp and told her to get up and mend my shirt, but she ignored me. Now, I’m nearly 40 sui and I still have no son, so I spoke to her nicely and tried to persuade her to have sex with me (yong haohua xiang nüren quandao), and I moved my pillow over next to her, thinking that we would sleep together. But my wife threw my pillow on the floor and pulled the blanket over her head to sleep. I thought to myself, every day she complains and quarrels and rejects me, and she’s not willing to make a family with me (buken he xiaode chengjia). I could see that there was no hope she would ever change. The more I thought about it, the more anger and hatred I felt.14

These case records are especially revealing because each climaxed with the husband killing his wife. But we see the same sort of behavior in many cases that resulted in a husband selling rather than killing his wife. In a 1747 case from Leiyang County, Hunan, peasant Liu Gongwei (24) sold his wife, Zhu Shi (24). She was unhappy about their poverty and wanted out of the marriage, and she pestered him until he finally gave in. As Liu later testified: “Because my household was very poor, Zhu Shi constantly quarreled and complained. Finally, I had no choice but to marry her to another man. Zhu Shi herself was willing to remarry.”15 Similarly, in a 1766 case, Zhang Wending (48) and his wife, Fu Shi (37), were landless peasants from Jiangling Cty, Hubei, who fled poverty to seek work in Leshan County, Sichuan. Not long after they arrived, their little son died of diarrhea, and, after his death, Fu Shi became fed up with her husband’s inability to

Chapter 7. Wives, Natal Families, and Children

217

support a family. As Zhang later testified, “Fu Shi told me that since I couldn’t support a wife, I should find some other man and sell her to him in marriage.” Zhang was reluctant to take this step, but she continued to pressure him. While they were in bed, she would curse him, saying: “On the one hand, you can’t support me, but on the other hand, you refuse to marry me off to someone else—are you telling me we’re just going to starve to death?” Finally, he agreed to sell her to another man.16 The men in these cases took women’s threats of suicide very seriously. In an 1810 case from Linyu County, Zhili, peasant Ma Siliang sold his wife of ten years, Liu Shi, after she jumped in a well and had to be pulled out by the neighbors. She was not injured (and she made sure to make her jump before witnesses who would rescue her), but her dramatic gesture had shaken her husband. As Ma later recalled: Liu Shi resented my poverty, and constantly stirred up a fuss and caused trouble, saying that she wanted to marry someone else in order to avoid suffering more in the future. At first I refused to listen. But she repeatedly threatened to kill herself, and I was afraid there would be trouble, so finally I agreed to sell her in marriage. . . . She was dead set on marrying someone else—she absolutely refused to continue suffering in poverty with me.

The couple’s two small children followed Liu Shi, leaving Ma on his own.17 “Laziness” and “Disobedience” Husbands who sold their wives because of poverty frequently complained that they had been “lazy,” “disobedient,” or “unskilled.” The phrase “likes to eat, but too lazy to work” (hao chi lan zuo) often appears in these men’s testimony, and this was evidently one of the most damning things one peasant could say about another.18 (Women who wanted to be sold sometimes used the same phrase to describe a wastrel husband; disappointed parents might characterize a ne’er-do-well son in the same terms.)19 The clear expectation of these husbands was that a wife had to pull her weight; she had to work, as well as eat. Such comments highlight the importance of women’s work in the peasant household. When men explained why they wanted to buy a wife, they typically cited their desire for a son, but, in addition to reproduction, it seems everyone took for granted that a wife had work to do—and peasant women’s work generally included both domestic tasks and the making of marketable handicrafts to supplement household income. These disparaging comments about women often focus on their failure to make clothing or lack of skill in doing so, with “she didn’t even know how to make her own clothes” being a particularly contemptuous observation.20 Spinning cotton, weaving cloth, and making and mending clothing were quintessential women’s work, important both to clothe the family and to supplement its income. Under

218

part two

conditions of involution, the income from such handicrafts became crucial to a household’s subsistence.21 Nevertheless, as Hill Gates observes, women’s work tended to be invisible (it is seldom directly acknowledged in legal cases), and this invisibility helps us make sense of husbands’ complaints that wives were “disobedient”: [One] reason for the invisibility of the work that women perform is that most of it occurs in the special, gendered form of “obedience” as subordinated kinswomen. . . . Girls do not work independently; they “help” their mothers. Daughters-in-law are seen to “help” their mothers-in-law, wives to “help” their husbands. The role of active worker who might take credit for specific output is sunk in the culturally much more significant kin relationship; the material output is veiled behind the submission to parental authority which motivates it. . . . One rarely hears a Chinese daughter or daughter-in-law who works under an older woman’s direction described as qinglao— hard-working; if she is praised, she is usually described as hen tinghua—very obedient.22

The corollary, of course, is that a wife who did not work hard was considered “disobedient” (bu tinghua), a term often coupled with “lazy” (landuo, lan zuo), almost as if they were synonyms. For example, in an 1810 case from Rong County, Guangxi, a husband testified that he had decided to sell his wife because of poverty but also because “she was always lazy and wouldn’t do what she was told.”23 In an 1801 case from Dazhu County, Sichuan, Huang Shi was said to be “lazy,” a “bad wife,” “defiant when scolded,” and “unwilling to endure poverty”; eventually she ran away from her husband’s home.24 The implication of such comments is that a woman’s “laziness” and “disobedience” exacerbated a household’s poverty. We should not take these husbands’ epithets entirely at face value. As we have seen, a woman might very well act “lazy” and “disobey” for a purpose—after all, these qualities were invariably attributed to unhappy wives who wanted out of their marriages. We also see this kind of behavior when a wife who had agreed to be sold found her new situation unsatisfactory and wanted to escape her second marriage. Consider the following two cases from Zhili. In a 1762 case from Bagou Subprefecture, Zhang Shi’s first husband resented that “she was lazy, she blamed me for my poverty, and she complained and made a fuss all the time”; but, after being sold, Zhang Shi found herself equally unhappy with her second husband, Cui Bo, who was almost as poor as her first, and she expressed her feelings in much the same way. When she refused to do housework, her new mother-in-law scolded her for being “a disobedient thing” (bu tinghua de dongxi—note that, in Chinese, it is a profound insult to call someone a “thing”/dongxi), so Zhang Shi shoved her; this unfilial defiance provoked Cui to beat Zhang Shi, saying “no wonder your first

Chapter 7. Wives, Natal Families, and Children

219

husband got rid of you!” She cursed him back, and he ended up beating her to death.25 In a 1777 case from Zhangjiakou Subprefecture, the buyer complained that his new wife resented his poverty; she was “lazy” and a bad cook; and she refused to obey his orders to make cloth shoes (a basic women’s handicraft). He decided he had been cheated.26 Similarly, in an 1820 case from Han Department, Sichuan, Liu Shi was dismayed to learn how poor her new husband, Ding Renyi, was, since she had agreed to be sold in the first place only in order to escape poverty. They often quarreled over her unhappiness, which manifested itself as “disobedience.” Matters came to a head when Ding scolded her for being “lazy” and in reply she cursed him, saying that she was not willing to live with him, and she started to pack up her clothes in preparation for leaving. He beat her to death.27 How did such women understand their wifely duty to obey and to work? A 1797 case from Lushi County, Henan, records a remarkably explicit statement of one woman’s understanding. In this case, a landless peasant named Yang Ming (38) ended up killing his wife, Liu Shi (28), because she complained of poverty, refused to do household chores, and demanded a divorce: My wife often quarreled with me and complained because I couldn’t earn enough money by hiring out my labor, and we had nothing to eat at home. She demanded that I divorce her so she could marry someone else. I had no choice but to put up with this. On the evening of JQ 1.12.10, I told my wife to mend my shoes because they were torn. She refused, saying that if I could not support her properly, then she wouldn’t do chores for me either (xiaode bu neng yang ta, ta ye bu gei xiaode zuohuo). I got mad and yelled at her, but she just carried on and demanded that I divorce her, and finally she started butting me with her head. I saw that my wife resented my poverty so badly that she wanted to marry some other man, and didn’t have even the slightest wifely feeling for me as her husband (haowu fuqi qingfen), and this made me extremely angry. I suddenly got the impulse to kill her.28

Note the stark contrast between Liu Shi’s perception and her husband’s. In explaining her refusal to mend his shoes, she articulated an explicit quid pro quo: his failure to fulfill his husbandly duty to support her freed her from her wifely duty to obey and do his household chores. She understood their marriage in terms of a pragmatic reciprocity, in which he had to hold up his side of the bargain. In contrast, he apparently believed she owed him affection and obedience based on a hierarchy of fixed, non-negotiable roles. In such cases, a woman’s refusal to obey or to work can be compared to a refusal to consummate a new marriage, which was typical strategy for a remarrying widow unhappy with her new husband. Since obedience, work (both domestic tasks and handicrafts), and sexual intercourse were all minimal expectations of a wife, to withhold them was a woman’s way to refuse a marriage and to express her desire to separate.29

220

part two

W I V E S W HO S A B O TAG E D S A L E S

If we shift our focus to the opposite end of the spectrum of women’s attitudes toward wife sale, legal cases document a number of ways that wives who did not want to be sold could prevent or sabotage a sale. A basic factor to bear in mind is that prospective buyers hoped to avoid trouble, and most already experienced some degree of anxiety at the risks inherent in buying “a wife with a living husband.” But also, as we already know, many of the women found in these cases had strong personalities and were not easily bullied. Very often, an ardent protest that came to the attention of the prospective buyer was sufficient to stop negotiations. In an 1863 case from Ba County, for example, peasant Xu Huanzhang (19) tried to sell his wife, Chen Shi (20), because of poverty—Xu’s father, who lived with the couple, had pressed him to take this measure. The matchmaker found a buyer willing to pay 10 taels. But Chen Shi vehemently rejected this plan and made an embarrassing public scene in the village. She caused such a commotion that the buyer backed out, and Xu and his father gave up their idea of selling her; instead, they sent her back to her natal home for a few weeks, in order to cool her anger.30 In Chapter 4, we discussed the common practice of allowing a likely buyer to visit and “take a look” at the woman who was offered for sale. This visit enabled him to assess her age and health (and hence her fertility) but also her temperament and attitude toward being sold. Wives understood the purpose of these visits, and they would size up the men in return. An assertive woman might take the opportunity to make her own views known, and this was one of the ways a wife might exercise a veto over such transactions. For example, in an 1807 case from Yiyang County, Jiangxi, a landless peasant named Bao Shifa decided he had no choice but sell his wife, Zheng Shi (30), in order to support their two little sons. She did not want to be sold (she refused to leave her children), and she absolutely refused to cooperate, but he proceeded to engage a friend from another village, Wang Songmao, to act as matchmaker and find a buyer. Wang, who did not realize that Zheng Shi had refused to be sold, visited one day with a prospective buyer, a cloth peddler surnamed Qiu, to “take a look for purposes of marriage.” But as Zheng Shi later testified, “when I realized why they were there, I scolded them and drove them away (jiang tamen cun chi zou qu).” Not surprisingly, the buyer refused to proceed. Bao Shifa angrily beat his wife for having spoiled the sale, so she took their children and ran away to her natal home.31 This episode suggests a limited parallel with the agency sometimes exerted by slaves who were put up for sale in the antebellum American South. Slave owners would try to persuade slaves to cooperate, because a show of resistance or a bad attitude might scare off possible buyers. At the auction house, buyers had a chance to inspect and question the slaves they were considering buying. These encounters enabled a mutual sizing-up, and a slave who did not like what he or she saw could

Chapter 7. Wives, Natal Families, and Children

221

behave in a manner calculated to deter a particular buyer from purchase. In this limited sense, a slave sale constituted a three-way negotiation between seller, buyer, and slave.32 If such subaltern agency was possible even within the extremely constrained circumstances of American slavery, it played a far greater role in Chinese wife sales. We have already seen that some women threatened suicide in order to compel their husbands to sell them. But other women threatened suicide in order to prevent a sale. Here we see another suggestive parallel with American slavery. In order to avoid being sold, slaves often ran away, hiding until they believed the danger had passed (we find the same strategy being pursued by Chinese wives who sought to avoid being sold). But the most intrepid or desperate slaves would threaten to mutilate or even kill themselves in order to ruin their cash value as commodities. By this ultimate act of self-determination, slaves would assert their agency as human beings against their status as commodities, forcing owners to desist from sale by threatening to destroy their very salability.33 Chinese wives who threatened suicide to avoid being sold seem to have had something different in mind: less a threat to destroy their husbands’ property than an assertion of their own status as good, loyal wives who did not deserve to be sold. Moreover, suicide implied the threat of public scandal and legal trouble (it was illegal to sell a wife in China, whereas slave sales were perfectly legal in the old American South), not to mention haunting by an angry ghost.34 A wife might also seek protection from senior members of her husband’s family who could order him to desist. Here is the testimony of Dai Shi, a landless laborer’s wife, recorded in an 1813 case from Liu’an Independent Department, Anhui: We were so poor that we had difficulty getting by, and my husband wanted to sell me in marriage. But I refused to agree, I cried and threatened suicide, and my husband’s older brother Zhao Feng and my father-in-law, Zhao Yunpei, who live separately from us, scolded my husband and made him stop. Then, four months later, my husband once again said that he wanted to take me away from home and sell me in marriage. I was upset and frantic because I did not want to lose my chastity, so I looked for my brother-in-law Zhao Feng to ask him to intervene, but he was away from home working on a job. My elderly father-in-law was sick in bed, so he told me to hide for a few days at my husband’s great aunt’s house and wait for my brother-in-law to come home and take care of things. So I went to stay with my husband’s great aunt. . . . On the afternoon of the 13th, my brother-in-law Zhao Feng and my husband together came to see me at great aunt’s home. Zhao Feng made my husband promise me that in the future he would never sell me in marriage. Only then did I agree to let them take me home.

Dai Shi’s husband was not happy—he told her, “just wait until we have nothing left to eat, I have no doubt you’ll agree to remarry then!” But he could not defy the senior members of his family, who united in forbidding him to sell his wife.35

222

part two

The Critical Role of the Natal Family If a wife wanted to avoid being sold, then by far the most common and effective strategy documented in the case records was for her to run away and take refuge with her natal family. Old stereotypes notwithstanding, it was normal for a married woman to maintain close ties to her natal family, which had a recognized role in protecting her interests.36 In the last case narrated above, Dai Shi’s natal family lived too far away for this to be practical, and if her senior in-laws had not intervened then she would have been vulnerable to coercion. In this respect, however, Dai Shi was unusual. Village exogamy was the rule, but it was common for women to marry out to nearby villages, in part because matchmakers relied on personal networks and face-to-face negotiations to arrange marriages. As recently as the 1980s, according to a national survey, most rural marriages took women no farther than twenty-five kilometers from their natal homes—a distance that might be walked in a single day, with effort—and most of anthropologist Ellen Judd’s Shandong informants (including elderly as well as younger women) had married within six kilometers of home.37 Proximity facilitated the maintenance of strong ties between married women and their natal families. As Judd explains, “keeping daughters nearby is a factor considered in making marriages, and a proper daughter should not wish to marry far away.”38 In the early years of her marriage, a woman was expected to receive continued nurture and support from her natal family; with the passage of time, roles might reverse, with the married daughter often providing support to her aging parents. Under the circumstances, it was natural and often quite easy for an unhappy wife to seek her natal family’s protection. No doubt, this helps explain why so many husbands and matchmakers made sure to consult a wife’s natal family before undertaking to sell her. When a natal family did agree to a wife sale, it was common to stipulate that the match be local, in order to protect their close ties with their daughter. In a 1759 case from Teng County, Shandong, Niu Qin decided that, because of poverty, he had to sell his wife of three years, Hu Shi (27). At first, she refused, telling him that “even if we die of poverty, we are husband and wife and must stay together.” Finally, Niu appealed to Hu Shi’s mother, Cao Shi, and older brother, emphasizing his desperate circumstances and promising them a share of the brideprice if they could induce Hu Shi to cooperate. Cao Shi agreed to help, on the condition that Niu find a local buyer, and she succeeded in persuading her daughter by assuring her that she would stay close to home and not lose touch with her natal family. The fact that Hu Shi agreed suggests that she may have feared separation from her natal family more than separation from her husband (her touching vow of conjugal loyalty notwithstanding). The matchmakers located a widower named Huo Qi (51), who was introduced to Hu Shi, “liked what he saw,” and agreed to pay a brideprice of 22,000 cash

Chapter 7. Wives, Natal Families, and Children

223

(of which Niu promised the natal family 6000 cash). Huo actually hailed from Qinghe County, some 200 kilometers to the north, but he concealed this fact from the matchmakers on the assumption that, once he had custody of Hu Shi, he could take her wherever he liked. After finalizing the sale, however, Huo hired a man with a cart to help take his new wife home, and her natal family learned from this man that their destination was Qinghe County. Cao Shi and her son were incensed at this betrayal—Qinghe County was too far away, but also they became alarmed at the possibility that Huo Qi might in fact be a trafficker who intended to resell Hu Shi into prostitution. The natal family insisted on canceling the deal.39 Running Away to the Natal Family In a 1755 case from Yongshun County, Hunan, peasant Xiang Shifu (32) had arranged through matchmakers to sell his wife, Tian Shi, to a stone mason named Liu Furong; the matchmaker assured Liu that “since her husband himself has authorized her sale in marriage, we guarantee that there will be no trouble at all.” Nevertheless, as that matchmaker later recalled, When the day came, Liu Furong brought the money to my house to wait for Xiang Shifu to bring his wife, so that the exchange of money for woman could be made face to face. We waited until evening, and finally we saw Xiang Shifu walking toward us alone. He told us that his wife had gone to her natal family’s home and refused to leave. He couldn’t fetch her back, so there was no way to complete the transaction. The deal was off. When Liu Furong saw that the marriage had failed, he took his money and went home.40

This case illustrates the most basic strategy for a wife who did not want to be sold: placing herself under the protection of her natal family. In a 1774 case from Shaoyang County, Hunan, agricultural laborer Deng Tianxiang (36) decided to sell his wife, Xu Shi (35). Their household had a severe Chayanovian profile, including two little sons and Deng’s elderly mother, and Xu Shi was relatively expendable because she had already fulfilled her function for his lineage by bearing him sons. Deng secured his mother’s permission to sell his wife, but he anticipated that Xu Shi would resist being separated from their children, so he pretended that he had hired her out temporarily as a domestic servant. However, as soon as Deng led his trusting wife away from their home, the little boys began crying for their mother, provoking a crisis of conscience for Deng’s mother, who sent after the couple and forbade her son to sell his wife. When Xu Shi realized that Deng had actually planned to sell her, she stormed out, returning to her natal home and complaining to her father. As Deng later testified, “After a few days, my father-in-law brought my wife back and confronted my uncle, making him guarantee that I would not be allowed to marry her off. My uncle scolded me

224

part two

severely, and I regretted what I had done. After that, I didn’t dare try to sell my wife in marriage again.”41 Something similar transpired in a 1788 case from Wugang Department, Hunan, in which peasant Yang Renwu (27) decided to sell his wife, Xue Yaogu, because of poverty. Anticipating her resistance, he engaged a matchmaker and secured a buyer without telling her. But before they could finalize the deal, Yaogu found out what was going on: “I heard that my husband had asked Tang Yaolong to act as matchmaker to sell me in marriage to Yang Laosi, so I went back to my natal home and told my father. My father refused to accept this, and scolded my husband. Then my husband dropped his plan to sell me.” After being scolded by his fatherin-law, Yang Renwu (by his own testimony) “did not dare” to try to sell his wife. Instead, he told the matchmaker to cancel the deal.42 These last two cases illustrate a pattern found in many others. The wife returns to her natal family and complains. Then the senior male member of her family confronts her husband or the senior male member of his family, scolding him and causing great loss of face. The wife’s natal family will let her return to her husband only after extracting a promise before witnesses that he will not sell her. In other words, the wife’s maneuver transforms her problem into a confrontation between the two families, one that invokes basic patriarchal values as well as family pride, with the attempt to sell the wife denounced as a shameful (and criminal) violation of the marriage contract that bound the two families together. In a 1799 case from Ningxiang County, Hunan, we see a wife seeking refuge with her natal family to avoid being sold under very different circumstances. Peasant Peng Wanxun had caught his wife, Sun Shi, in the act of adultery with a neighbor, turned her in to the yamen for prosecution and punishment, and received the magistrate’s permission to sell her off. Peng arranged through a matchmaker to sell Sun Shi to another peasant for a brideprice of 12 taels. But, although Sun Shi was happy to leave her husband, she refused to marry the prospective buyer on the grounds that he was too old; on the very day that the parties drew up the contract and finalized the transaction, she ran away to her natal home, where her mother and brothers gave her refuge. When the buyer found out what she had done, he canceled the transaction. Peng Wanxun was outraged by his wife’s behavior, but she had four tough brothers who were willing to defend her, so he washed his hands of her, and (in Sun Shi’s words) “after that, I had no more contact with my husband.” The protection of her natal family secured for Sun Shi a de facto divorce.43 In all of these cases, it is clear that the woman lived within easy walking distance of her natal family and could count on their support. Their proximity and support were the decisive factors that enabled her to prevent her husband from selling her.

Chapter 7. Wives, Natal Families, and Children

225

Filing Charges to Forestall a Sale In the above cases, we find natal families sheltering their daughters and directly intervening with her in-laws to forestall a sale. If direct intervention did not work or was not possible, a woman or her natal family might go to the local yamen and file charges.44 This scenario accounts for about a third of my sample of local court cases, and it constitutes the second most common reason that wife sales ended up in court. The Ba County archive contains many cases in which wives came to the yamen on their own to seek protection against husbands who wanted to sell them, albeit with mixed results. Such cases tend to involve couples who lived within close distance of the yamen (i.e., Chongqing and its immediate environs); typically, these were rural couples who had migrated to the city because of poverty at home, and for this reason the women’s natal families were not within reach—hence the need to seek official protection instead. Most of these women filed oral complaints on days when the magistrate was holding open court, rather than the written plaints typical of most lawsuits. Filing a written plaint cost money; moreover, any woman who filed one required an adult male guarantor willing to go on record to back her up.45 The women in these cases were not likely to have access to either money or a guarantor. In 1862, a woman named Liu Wu Shi (35) filed charges to prevent her husband, Liu Yongshun (with whom she had three children), from selling her in marriage to one Yu Tongshi, who had agreed to pay forty strings of cash for her to be his concubine. Liu and Yu had already exchanged contract for cash, but Liu Wu Shi refused to be sold. She had appealed to her neighbors, to no avail, so finally she managed to steal the contract and make her way to the yamen in order to throw herself on the magistrate’s mercy.46 The outcome of this case is not known, but some women who went to court on their own did manage to secure protection. In a 1788 case, Hu Shi went to the yamen to complain that her husband, Li Desheng, had arranged to sell her because of his illness and poverty; however, despite hardship, she did not want to remarry. In response to her complaint, the couple’s neighbors mediated a settlement in which Li and the prospective buyer agreed to cancel the transaction, and both men filed affidavits with the court to that effect.47 In other cases, magistrates ordered that those involved in negotiating the wife sale be punished and that the woman be granted a divorce and returned to her natal family.48 But the picture is decidedly mixed: some cases offer a far less sanguine perspective, suggesting that wives who filed charges against their own husbands could not assume they would get a sympathetic hearing in court. For example, in an 1840 case, Zeng Wu Shi filed an oral complaint that her husband, Zeng Chaichang, had beaten her in order to force her to submit to being sold. The forensic examiner

226

part two

reported bruises and broken skin on her face, but no “breaking injury” (i.e., broken bones or teeth or other internal damage). Therefore, the magistrate found that Zeng’s beating of his wife fell within the legal limit permitted a husband, and, after making Zeng promise not to sell his wife, he remanded her to Zeng’s custody.49 In an 1838 case, Ran Wang Shi filed a similar complaint against her husband, who admitted attempting to sell his wife because “she does not behave as a wife should.” In this case, the magistrate did not take the threat of wife sale seriously; instead, he ordered the husband slapped for “failing to control his wife” and returned her to his custody along with orders that he do better in the future. The magistrate ordered no punishment for Ran Wang Shi herself, on the apparent assumption that her husband would take care of that on his own.50 We have no information about how these two women later fared, but it seems unlikely that going to court had improved their security. As the last two cases suggest, a wife who complained about her husband on her own found herself in a weak position. The rules for filing lawsuits actually forbade magistrates from accepting such plaints (as well as plaints filed by sons against fathers, by younger brothers against older brothers, etc.) and both Ming and Qing codes mandated punishment for those who filed them, on the grounds that they violated normative Confucian hierarchy.51 Magistrates sometimes made exceptions, as we have seen. Nevertheless, without question, it was far more effective for a woman to have a senior natal relative file charges on her behalf. There was no perception of impropriety if a woman’s father, uncle, or brother filed charges against an abusive husband, and in marital disputes of all kinds it was typical for a woman’s interests to be represented in court by her natal family.52 In an 1829 case from Ba County, a young woman named Jiang Ergu did not want to leave her husband, Jiang Daiting, but he had been pressured by his mother to sell her. The buyer was a laborer named Ma Shihe. As Ergu’s mother-in-law later testified, At dawn, Ma Shihe arrived with a sedan chair to fetch Ergu, but she refused to obey. My son, the matchmaker, and I forced her to get in the sedan chair and the porters carried her to Ma Shihe’s home. But as soon as they arrived, Ergu ran away back to her natal home and told them what had happened. Then her father Jiang Rufa filed charges against us.

The magistrate ordered the transaction canceled and the buyer and matchmaker both slapped. Jiang Ergu was reunited with her husband, in exchange for his and his mother’s pledge never again to try to sell her.53 In an 1852 case from Ba County, Li Long Shi (22) visited her mother, the widow Long Peng Shi, and complained that her husband, Li Farong (26), was trying to sell her. The couple was very poor and had fallen heavily into debt, so Li had decided to sell his wife, but, anticipating her resistance, he delayed informing her until

Chapter 7. Wives, Natal Families, and Children

227

matchmakers had already secured a buyer and Li had exchanged a contract for a down payment. The prospective buyer, Wang Shuchuan (28), needed a concubine because his wife had so far failed to bear a son; Li Long Shi had borne at least two children, and Wang, impressed by her youth and proven fertility, had agreed to pay her husband a high brideprice of 55 taels. Wang had been told that the woman’s natal family approved the sale, but in fact they had not been consulted. Li Long Shi was certainly fed up with her husband and not necessarily averse to separation, but she did not want to be anyone’s concubine—hence her complaint to her mother. Upon hearing what was afoot, Long Peng Shi promptly filed charges against her son-in-law, the matchmakers, and the prospective buyer, thereby freezing the transaction. The magistrate responded by canceling both marriages and returning Li Long Shi to her mother’s custody. This judgment did separate her from her children, who remained with their father—but the record contains no information about how she felt about that result.54 Natal Families as Allies—A Summary Case In a final example, which may serve as a summary case, we find a woman’s natal family helping her first to escape an unhappy marriage and then to avoid being sold into a new marriage she did not want. In this 1890 case from Nanbu County, Jia Shi was the eldest daughter of peasant Jia Qiqing (45). In her teens she had married Yuan Zongdai, the son of Yuan Fengcai (44), but the marriage was not harmonious, and she often ran away to the safety of her natal home, refusing to return. According to Jia Shi, her husband was abusive: “After I joined his household he was always cruel to me, and that’s why I ran away to hide at my natal home.” The Yuans claimed that Jia Shi was disobedient and unfilial, and that her father (who resented the Yuan family’s poverty) encouraged and indulged her defiance. The two claims are by no means mutually exclusive, since, as we know, “disobedience” was a basic strategy for women seeking to exit unhappy marriages. Finally, Jia Qiqing and Yuan Fengcai asked two rural agents (baozheng) to mediate, and a meeting was convened at a temple. All sides agreed that, given the couple’s irreconcilable differences, Jia Shi should remarry and the Yuan family should make the arrangements; Jia Qiqing promised to cooperate, and in the meantime Jia Shi would stay with him. After six months, the Yuan family arranged to sell her in marriage to Zhao Dashun (68), whose grandson Guoxiao lacked a wife. Representatives of the three families met at the temple, Zhao Dashun “had a look” at Jia Shi, all sides agreed, and the exchange was made: the Zhaos paid a brideprice of 20,000 cash to the Yuans, plus 2000 cash to Jia Shi’s father, in exchange for a contract bearing Yuan Zongdai’s hand- and footprints, and Jia Shi’s uncle delivered her to the Zhao household. Only then did Jia Shi discover that her new husband, Zhao Guoxiao, was a “deaf-mute” (yaba)—it is not clear whether the Yuans had known this, but nobody

228

part two

had told the Jias, and Guoxiao’s grandfather had represented his family in all the negotiations. Jia Shi was horrified at her new husband’s disability, and within a week she snuck out of the house at night and fled back to her father’s home. Members of the Zhao family tried to fetch her back, but she “defiantly refused to return.” A second mediation attempt by the rural agents failed when the Yuans refused to cancel the marriage sale and refund the Zhaos’ brideprice. Jia Qiqing was incensed by the way his daughter had been treated, and he filed charges against the other two families, who then filed counter charges against him and each other. The magistrate ruled that, under the circumstances, the original mediated agreement had been appropriate and Jia Shi’s marriage to Zhao Guoxiao therefore legal. (He did not broach the question of whether concealing Guoxiao’s disability from the Jia family might have constituted marriage fraud, perhaps because the transaction had taken place between the Yuans and the Zhaos.)55 He reprimanded Jia Shi for her incorrigible defiance—“truly, she failed to behave as a wife should”— ordered her slapped, and returned her to the natal family; he instructed her father to marry her off to someone else and to use the new brideprice to refund the Zhao family’s 20,000 cash.56 He permitted the Yuan family, who really were quite poor, to keep what they had received from the Zhaos. Once again, the proximity of a woman’s natal family and their willingness to defend her proved decisive. With her father’s help, Jia Shi was able to force her inlaws to let her remarry, and, when she found her new husband not to her liking, her father’s willingness to go to court on her behalf freed her from entanglement with either family. She ended up in the custody of her father, who had carte blanche to arrange a new marriage that she would find acceptable. No doubt, it was unpleasant to be reprimanded and slapped; nevertheless, in a real way Jia Shi was the victor in this protracted struggle. However, if her natal family had not been available to help, there is no telling what might have happened to her.57 W I V E S S O L D AG A I N ST T H E I R W I L L

As we have seen, many women did not lack for resources or allies, and a wife who did not want to be sold might well be able to prevent her sale. But one can certainly find exceptional cases of wives sold against their will. The following three examples illustrate specific circumstances that might render a wife vulnerable to such coercion. Three Cases of Coercion In a 1738 case, Wu Shi (34) and her husband, Hu Cunpo (40 or older), were poor peasants from Ruichang County, Jiangxi, who had fled famine to Guangji County, Hubei (about seventy kilometers to the north). They had two sons with them (one aged 8 sui, the other an infant not yet weaned) and had been reduced to despera-

Chapter 7. Wives, Natal Families, and Children

229

tion. Hu decided that he had no choice but to sell his wife. But Wu Shi refused to be sold, so her husband arranged the transaction behind her back, engaging three acquaintances to act as matchmakers and providing a “request certificate” authorizing them to act on his behalf. However, after finding a buyer and mediating the exchange of brideprice for contract, the matchmakers murdered Hu Cunpo so that they could keep all the money for themselves. Then they informed Wu Shi that her husband had sold her and her sons, taken his money, and gone home to Jiangxi; they turned her over to her buyer. Wu Shi was shocked and distressed at this fait accompli. But she believed the matchmakers’ lies, because she knew that her husband had indeed been considering selling her. Why did coercion succeed in this case? The crucial factor is that Wu Shi was alone, far from home, without family or friends to help her. Under the circumstances, she had no choice but to cooperate, if for no other reason than to save her hungry children. Indeed, had her husband not been murdered, he might well have abandoned her exactly as the matchmakers claimed he had done.58 A 1747 case from Wuwei Independent Department, Anhui, presents a very different set of circumstances that made coercion possible. Zhu Shi, the wife of tenant farmer Qiu Yingyou (50), was notorious for her open adultery with Qiu Yingzhou (45), who was a distant relative of Yingyou’s. Yingzhou was a tough and intimidating man, and although Yingyou resented his wife’s adultery, he feared Yingzhou too much to confront him; besides, according to Yingyou’s testimony, his wife had always been “fierce” (qianghan) and “impossible to control” (guan ta bu xia). This case is notable for the brazen openness of the sexual relationship between Zhu Shi and Yingzhou, which continued for three years without anyone interfering in any direct way. Finally, Yingyou’s landlord became so disgusted that he withdrew Yingyou’s lease and kicked the couple off his land, forcing them to return to Yingyou’s home village—at which point, Yingzhou followed them and built a straw hut for himself near their house, where Zhu Shi would visit, eat meals, and spend the night as she pleased. At this point, Yingyou resolved, now that he was back on his home turf, to solve the problem once and for all by selling Zhu Shi. He enlisted the aid of his two brothers, and they in turn engaged three sympathetic neighbors to act as matchmakers for free. They secured a buyer, exchanged contract for brideprice, and set a date for a sedan chair to fetch Zhu Shi to her new husband’s home. When Zhu Shi discovered what was afoot, she made a noisy, public scene in the village and took refuge at Yingzhou’s hut—apparently she counted on her lover to face down her husband. But Yingyou was one step ahead of her; he had persuaded Yingzhou’s older brother, Qiu Yunzhu, to help. Yunzhu felt deeply ashamed of his younger brother’s conduct, and he agreed to be on hand when the sedan chair arrived, in order to prevent Yingzhou from interfering.

230

part two

Thus, on that fateful day, Yingyou was able to confront his wife with a bold show of strength, backed by his two brothers, the three matchmakers, and her lover’s own older brother. When she realized what was happening, she cut off her hair and began smashing the dishes and other household goods in Yingzhou’s hut, but the seven men dragged her out, forced her into the sedan chair, and escorted it out of the village. As she struggled with them, Zhu Shi caught sight of her lover, watching from a distance and helpless to intervene because he could not defy his own older brother, and she seems to have given up hope. When Zhu Shi arrived at the home of her buyer—a landless agricultural laborer who had gotten her for a bargain price—she informed him frankly that she had been sold against her will because of adultery. But he did not care, and she did not resist him; they consummated the new marriage that night. This case later ended up in court because Yingzhou, in a rage of frustration, committed a murder and then tried to use the corpse to implicate Yingyou falsely in order to get revenge. The truth quickly came out. Why did coercion succeed in this case? Because of her notorious conduct, Zhu Shi seems to have had no friends or allies apart from her lover, whose hands were tied by the intervention of his own older brother. She had nothing to gain from going to court, even if she had managed to do so. What about her natal family, her natural source of protection? According to testimony, she did have two brothers, but they lived “far away,” and she had had no contact with them “for many years.” All this was abundantly clear to Zhu Shi, and being sold off by force seems to have broken her spirit, and perhaps her sanity as well. She was able to testify coherently at the inquest, but later the yamen runners found it impossible to summon her for a second hearing. They reported to the magistrate that “Zhu Shi has gone mad: she pays no attention to her body or appearance, and she talks nothing but nonsense (man kou hu yan)—it is impossible to bring her in for interrogation.” An affidavit from her neighbors confirmed what the runners had reported: “She is frenzied and hallucinates—she has completely lost her senses” (kuang mi shi xing). It would be interesting to know exactly what sort of “nonsense” Zhu Shi uttered to persuade the runners she was mad. Whatever it was, they deemed it too indecorous to repeat in their report.59 The third example presents yet another set of circumstances that rendered a woman vulnerable to coercion. In this 1774 case from Hengyang County, Hunan, peasant Nie Yide’s older brothers forced him to sell his wife, Chang Shi, in order to compensate them after she accidentally started a fire that burned down their house. Nie did not want to sell his wife but felt he had no choice but to obey his brothers. Chang Shi, however, refused to cooperate, and (as we would expect) she ran away to her natal home and complained to her father, Chang Zhansan (66). But Nie Yide explained the situation to his father-in-law and begged him to permit the sale. Chang reflected (as he later testified) that, after all, his daughter was suffering

Chapter 7. Wives, Natal Families, and Children

231

from poverty and might be better off with a new husband. Finally he told Yide, “you decide what to do, and I won’t interfere.” Several days later, after brideprice and contract had been exchanged, the matchmaker and a cousin of Nie’s visited Chang Zhansan’s home and informed Chang Shi that they had persuaded her husband not to sell her after all, so she should not worry and they would take her home. She did not trust the two men and refused to go with them, but her own father lied to her, telling her that she should not worry because he guaranteed that they would not try to sell her off. But, on the road, she finally wormed the truth out of the matchmaker, whereupon she jumped in a lake in an apparent attempt at suicide. The two men pulled her out of the water and took her home, where she confronted her husband and insisted defiantly that she would rather starve to death than remarry. Her husband and his brothers watched her all night to prevent flight or suicide. The next day, Chang Shi continued her loud protests, refusing to eat and repeatedly trying to flee. By this time, her husband was torn by anguish and regret, and he begged his brothers to cancel the deal, but they insisted it was too late, because the contract had already been exchanged for the money. Then they tied Chang Shi to her bed and gagged her, so that they and the neighbors could all get some sleep—only to discover next morning that she had choked to death on the gag. What made Chang Shi vulnerable to coercion is that no one in either the Nie family or her own defended her. Even her father betrayed her by agreeing to the sale and then lying to her, although we have no reason to doubt that he genuinely believed she would be better off if sold. But Chang Shi’s sale was never completed, for obvious reasons, so the coercion she suffered can hardly be considered successful; if anything, her case underscores the difficulty of selling a wife without the wife’s cooperation. There is no way to know whether Chang Shi really would have committed suicide—her priority seems to have been deterring the sale, rather than actually dying. But all the evidence suggests that, if she had lived, she would have continued to resist. Moreover, her buyer had been falsely told that she had consented to the match, and once he discovered her true feelings, he would probably have canceled the deal.60 Isolation, Misogynist Violence, and the Limits of Female Agency In a 1737 case from Zhejiang, peasant Zhu Weiguan’s poverty reached desperate extremes after the harvest failed and he fell ill. He and his wife, Tang Shi (26), agreed to separate in order to survive, and both families reluctantly approved the decision. Zhu’s uncles and Tang Shi’s older brother then arranged a fraudulent sale (passing her off as a widow) to a single peasant named Li Yonglu. The natal family received a small share of the brideprice to acknowledge their cooperation. Tang Shi had cooperated fully, and she carefully concealed the facts of her marital history from Li Yonglu. But she found herself unhappy in her new household,

232

part two

which was also poor—after all, it was in order to escape poverty that she had agreed to remarry in the first place. Her unhappiness increased when her brother came to fetch her home for a visit, but her new husband refused to let her go, on the grounds that there was too much work to spare her at that time. Resolving to run away back to her natal home, she waited one night until Li was asleep, then crept out of the house and fled. She took several bolts of cloth, 5.4 taels of silver, and 400 cash belonging to her husband, in addition to a bundle of her own clothing. The problem with Tang Shi’s plan was that her new husband’s home was located in Changxing County, whereas her natal family lived in Fuquan County, over a hundred kilometers to the east across Lake Tai. Travel between the two counties took several days and required a boat. In this instance, the natal family had agreed to sacrifice their daughter’s proximity for the anonymity necessary to secure a lucrative fraudulent sale. Shortly before dawn, Tang Shi encountered two local men who had just gotten up: Zhao Yineng (33), a pig slaughterer by trade, described in testimony as fierce and intimidating, who was on his way with a friend (a casual laborer) to slaughter a pig. Zhao accosted Tang Shi, and she offered the men money if they would help hire a boat to take her to her natal home in Fuquan County. Realizing that this woman was traveling alone under suspicious circumstances, the men held her at knifepoint, stole her belongings, took her to an abandoned temple, and raped her. After that, they locked her in a shed and kept her captive for nine days, visiting every night to rape her. Rumors started to spread, however, and Zhao Yineng began to worry that Tang Shi might be discovered. Finally, the men hired a boat and delivered her to her natal home (repeatedly raping her along the way). The amazing thing is that the men did not murder her; instead, they assumed that Tang Shi would keep what had happened a secret, counting on her shame to protect them from being held to account. But when she arrived home, she immediately told her brother what had happened. He filed charges at the county yamen, and those who had abused her were arrested and severely punished (Zhao being beheaded). The fraudulent wife sale was prosecuted as a secondary offense, with Tang Shi herself sentenced to a beating for the part she had played.61 A basic argument of this chapter is that it was difficult to sell a wife against her will; therefore, most wife sales that actually took place did so with the cooperation of the wife. Tang Shi, too, had agreed to be sold, but her gruesome ordeal after she ran away reminds us of the limits of female agency within the traffic in women in late imperial China, and it warns against any sort of sanguine complacency. Here we have a young woman of spirit and decision, who participated fully in her own fraudulent sale, hoping to escape poverty and better her circumstances. When she found herself unhappy in her new marriage, she boldly seized the initiative and

Chapter 7. Wives, Natal Families, and Children

233

fled—and if her natal home had been located within walking distance, she might have succeeded. But her natal home lay too far away for quick passage on foot— and as many legal cases show, it was extremely dangerous for a young woman to travel any significant distance alone. In the event, she was lucky to escape with her life. To summarize: a woman might well prevent (or escape) an unwanted wife sale, but only if she had the courage to defy her husband, who might beat her. Courage alone might not suffice. Unless she had allies close at hand—her natal family, sympathetic in-laws, or the county yamen—she might be vulnerable to coercion and misogynist violence. The case of Tang Shi may be an extreme example, but it conveys a sense of the very real constraints that must have deterred many other women from taking such risks.62 CHILDREN IN WIFE SALES

Adele Fielde’s anecdote (quoted at the beginning of this chapter) reminds us that, when wives were sold, it was common for children to accompany their mothers into their new marriages, either temporarily or permanently. In both law and custom, a cardinal principle of Chinese patriarchy was that children belonged with their father; hence, a court-ordered divorce invariably separated the woman from her children. The primacy of patrilineal descent also meant that any “children born of adultery” (jian sheng nan nü) should become the responsibility of their father in order to protect the lineages of their mother’s husband and father from pollution.63 Moreover, wives who resisted being sold often did so because sale would mean separation from their children, whom their husbands intended to keep. (Unwillingness to sell children might also motivate a wife to agree to “bring in” an outside male to support her family through polyandry.) Nevertheless, the desperate circumstances of many couples that resorted to wife sale meant that children’s best hope of survival might be to follow their mother into her buyer’s household. In wife-selling cases, information about children is often incomplete or missing. In my entire sample, a total of 158 children are specifically mentioned as present in women’s families at the time of their sale. At least 93 of these children (about 60 percent) followed their mothers, either temporarily or permanently. If we disaggregate the sample of known children by sex, we find striking evidence of the power of son preference in Chinese society: 109 (69 percent) of the children were male, and 49 (31 percent) were female. These numbers yield an extraordinary ratio of 222 males per 100 females, which means that half or more of the girls actually borne by these women must be missing from the records. Daughters were less likely than sons to be mentioned in testimony, simply because they were deemed less important. But the extreme asymmetry of these figures must

234

part two

also reflect, at least in part, the many ways that poverty could exacerbate the demographic effects of son preference, a preference that derived from a complex of deeply rooted economic and cultural factors. Discrimination (especially with regard to nutrition and health care) as well as outright infanticide have long resulted in disproportionately high mortality for girls, a situation that persists in China today.64 Some of the missing girls, no doubt, were not dead but had been sold off to be servants or prostitutes, or had been given away as adopted daughtersin-law. We see the same phenomenon of skewed sex ratios on the micro-cosmic level in polyandry cases as well. The Uterine Family versus the Patriarchal Family To recapitulate, in my sample of legal cases, about 60 percent of the children known to be present at the time of wife sale followed their mothers into their new households, either permanently or temporarily. This transfer of children highlights the potential for conflict of interest between husband and wife, who had very different stakes in her sale. One way to understand such conflict of interest is through Margery Wolf ’s useful distinction between the men’s patriarchal family and the women’s “uterine family.” Wolf argues that, within the overall structure of the patriarchal family in China, daughters-in-law pursued their own interests by creating uterine families that consisted of themselves and their children, and pointedly excluded their husbands and in-laws. In Wolf ’s view, the uterine family served the patriarchal family up to a point, because both shared an interest in the bearing of children (especially sons). But she sees a strong if usually latent tension between the two, a tension that would break into the open when rival daughters-in-law encouraged their husbands to divide their father’s household. For Wolf, the successful Chinese woman was an insurgent who systematically hollowed out a space of her own, subverting patriarchy (by turning her husband against his brothers, and her sons against their father) even while paying lip service to it.65 Bearing Wolf ’s analysis in mind, we can see that a husband and wife might perceive the transfer of their children in the context of a wife sale in very different ways. As we saw in Chapter 5, there were two basic arrangements governing the custody of such children. In some cases, children were sold outright along with their mother and adopted by the buyer, who might pay a higher price. In others, the contract required the buyer to take temporary custody of children, promising to support them for a specified period and then return them to their father, whose surname they would retain; in this scenario, the buyer would pay a lower price in order to offset the cost of child support. I have examples of 54 children being transferred temporarily and 39 children being sold outright, a ratio of roughly three to two. Sons were far more likely to be

Chapter 7. Wives, Natal Families, and Children

235

transferred temporarily than daughters.66 Here is how these numbers break down by sex: Terms of custody for sons who followed their mothers:

Temporary transfer: Outright sale:

43 (77%) 13 (23%)

Terms of custody for daughters who followed their mothers:

Temporary transfer: Outright sale:

11 (30%) 26 (70%)

Sometimes sons had to be sold, if for no other reason than to secure their survival; but it was far more likely for a boy to be transferred temporarily than to be sold outright. Fathers sought to secure the eventual return of sons to their own lineages if they possibly could, in order to continue the family line; daughters, on the other hand, were more expendable. In fact, there are several cases in which one transaction included both types of transfer, with the wife sale contracts stipulating that sons will eventually return to their father whereas daughters are sold outright.67 The colloquial terms for permanent versus temporary transfer of child custody closely resembled those for absolute versus conditional sales of land. Permanent sale of children might be termed “si mai”—literally, “dead sale”—whereas a temporary transfer was “huo mai”—“live sale” (the same terms were used for absolute/ conditional land sales). Similarly, if a woman took her children into her new marriage permanently, she was said to have “taken them dead” (si dai), whereas in a temporary arrangement, she had “taken them alive” (huo dai).68 Here, the connotation of death versus life assumes the perspective of the children’s father: if sold permanently, they were “dead” to him, whereas if they would some day return, they remained “alive.” The same was true of land: if permanently sold, it was “dead” to the seller, but if conditionally sold, so that he might some day recover it, then it remained “alive” to him. From the seller’s standpoint, transferring his son to the temporary custody of his wife’s buyer represented a compromise between dispersing his family for the sake of their immediate physical survival and preserving it as a coherent line of descent for himself and his ancestors. Without survival, there would be no chance of continuing the family line, so the hope was to achieve the first priority without sacrificing the second. From the wife’s standpoint, however, when her husband sold her and transferred custody of their children, her uterine family effectively seceded from his patriarchal family in order to pursue its own interests by allying with her buyer’s patriarchal family. After all, it was the first husband who would end up alone, and the survival of his patriarchal family would depend on whether he ever actually

236

part two

managed to recover custody of his sons. (Of course, if he sold his sons outright, as happened in 23 percent of my cases, he would have no hope of doing so.) There is no way to estimate how often men succeeded, but anecdotal evidence from legal cases suggests that it was not always easy. Sometimes, when the time came, women simply refused to surrender their children.69 More often, however, tension focused on the efforts of fathers to maintain contact with their sons. Conflicts over Child Custody Many cases show that fathers who had transferred custody of their children temporarily made a serious effort to keep in touch with them. Such efforts reflect, in part, the seller’s sense of entitlement with regard to his wife’s buyer—that is, the sense that the sale of his wife had created a new and ongoing relationship between himself and her new husband. From the seller’s standpoint, this relationship entitled him to at least some material assistance, as reflected in demands for supplementary payments. But when children were in the picture, it also entitled the seller to visiting rights so as to keep in touch and monitor their welfare. Some of these relationships were amicable, being accepted with equanimity by the woman’s new husband. We find such a relationship in a case from Fenshui County, Zhejiang, in which landless peasant Song Wensheng was incapacitated by a seizure disorder (fengdian bing) and reduced to begging. In 1760, he sold his wife, Zheng Shi (aged 33 at the time), to peasant Chen Liangui (37) for a brideprice of 16 taels; the agreement included provision that Chen would raise the couple’s little son for six years and then return him to his father. Over the next four years, Song Wensheng frequently dropped by to see his son. Chen Liangui tolerated the presence of his wife’s ex-husband (perhaps because Song did not demand more money) and maintained a friendly relationship with him, sometimes feeding him when he visited. Unfortunately, Chen’s paternal uncle, Chen Yushu (55), found such contact extremely shameful, and when he saw Song, he would curse him and chase him away. But Liangui sympathized with Song and defended him, eventually provoking a bitter quarrel with his uncle. One evening, Yushu violently berated his nephew for having allowed Song to visit that day. Liangui reproached his uncle: “He came to give some rice to his son, I didn’t invite him, and I’ve never let him spend the night here. Why do you have to curse me like this?” The quarrel escalated into a fight, and Yushu beat his uncle so badly that the older man later died. Chen Yushu’s shame and anger aside, this case provides a striking example of how a wife sale accompanied by the temporary transfer of child custody could lay the basis for a stable, friendly relationship between the man who had sold his wife and her buyer. The way Liangui stood up to his uncle shows that he considered Song’s visits to his son to be innocuous and even worthy of praise; his uncle’s hostility and intolerance bothered him far more.70

Chapter 7. Wives, Natal Families, and Children

237

Other cases reveal significant potential for conflict, in which more than shame was at stake. To borrow Margery Wolf ’s conceptual scheme once again, a wife sale that temporarily transferred children set up a potential conflict of interest between three families: the two patriarchal families of the husbands, with the wife’s uterine family in between. Conflicts often seem to have focused on the treatment of children, with the first husband trying to assert a custodial prerogative over their welfare, even though he had contractually transferred custody to the second husband. In an 1819 case from Jin County, Shengjing, a landless laborer named Li Xiu (42) had sold his wife, Guo Shi (35), because “thigh pain illness” had prevented him from working and earning a living. The contract stipulated that the couple’s younger son, who was not yet weaned, would accompany Guo Shi to be raised by her new husband, Yan Xin (36, a widower whose first wife had died without issue), but would eventually return to his father’s lineage. After five months, Li Xiu had recovered his health and decided to look for work near his wife’s new village, in order to be near his infant son; he worried that Guo Shi did not look after the child properly. He happened to drop by for a visit, and Yan Xin was not there, but Guo Shi was home with the baby. She was not happy to see him and ordered him to leave before her new husband returned. Li Xiu scolded her for neglecting the baby, who was crying, and told her to breastfeed him. In reply she cursed him—“you’ve already sold me, so you can’t tell me what to do anymore!”—and ordered him off the premises. Outraged, Li Xiu beat her up, and she later died of her injuries.71 As this case suggests, a first husband’s efforts to monitor his son’s care might be perceived by the woman as unwelcome interference. She no longer felt subject to his authority—but also, she feared that his presence might poison her relations with her new husband and in-laws, just when she was trying to make a fresh start in a new household. If children were older, then conflict over custody might focus on the value of their labor. In an 1803 case from Liangdang County, Gansu, peasant Chen Guangwazi (30), his wife, Jin Shi (28), and their two children found themselves on the road, begging, after the harvest failed at home in Qin’an County. The couple agreed to “survive by separating,” and Chen arranged to sell Jin Shi and their infant daughter to a single peasant named Chen Liuyi for 4000 cash, with the stipulation that Liuyi would also support their son, Cai’er (6), for six years before returning the boy to his father. In a previous chapter, we have already discussed how the transfer of child custody affected the prices in wife sales: if children were transferred temporarily, it was common to reduce the price in order to compensate for the cost of raising them, whereas outright sale of children along with their mother generally indicated a higher price. In this case, the brideprice of 4000 cash for a wife and

238

part two

daughter seems very low, and it probably reflects the agreement to raise Guangwazi’s son for six years (in addition to the family’s sheer desperation). Nevertheless, after acquiring Jin Shi and her children, Chen Liuyi (a land-poor peasant) adjusted his household economy so that the entire family (including the little boy and Liuyi’s elderly father) worked at paper-making, the sort of labor-intensive commercial handicraft that was typical among the highly involuted peasantry. In this context, the boy’s labor was actually worth something—as Hill Gates has shown, children working in commercial handicrafts (notably, girls spinning cotton) could add significantly to a household’s marginal income.72 After several months, Chen Guangwazi visited the family to see his son, and the boy complained that Liuyi worked him too hard and often cursed and beat him. Guangwazi became angry, and he told Liuyi that he would take his son back immediately. But Liuyi refused, saying that the agreed period of six years was not yet up, and he accused Guangwazi of “breaking his promise.” Whereas Guangwazi had considered their agreement simply as a way to secure his son’s survival, Liuyi understood it to be a contract that bound the boy for a six-year term of servitude.73 In a 1792 case from Jingning County, Zhejiang, we see a very different sort of conflict over the custody of a son. In this case, peasant Liao Xinyuan (43) had sold his wife, Luo Shi (36), because of poverty to peasant Li Xiutong (63) for a body price of 20 taels, with the condition that Li agree to raise the couple’s only son, Liao Kaili, who was 8 sui at the time of the sale. The transaction concluded amicably, and over the next eight years Liao kept in close touch with his son and his wife’s new husband, who by all accounts scrupulously abided by the terms that had been agreed. But after eight years, the boy was 16 sui, and Li decided that it was time to return him to his father. Li complained of the expense of supporting Kaili, but a far more important factor was that Li—who by this time was 71—no longer wanted a young man of different lineage, on the verge of adulthood, to continue living in his household. He felt anxious about the integrity of his property and lineage; in effect, Liao Xinyuan’s effort to secure his own family line by placing his son in Li’s household was increasingly perceived by Li as a threat to his own patrilineal imperative. Li visited Liao Xinyuan to inform him of his decision, but Liao reacted with consternation: he was very poor, suffered chronic ill health, and could barely get by on his own—he feared for the boy’s safety but also felt that Liao was breaking his promise to raise Kaili to adulthood. The two men quarreled and got in a fight; the elderly Li got the worst of it, and later died of his injuries.74 Children might become the ties binding a first husband to his wife’s new household, reinforcing his sense of connection and entitlement. But they might also become a flashpoint for conflict, where the friction between contradictory interests of the three parties to a wife sale came to bear.

Chapter 7. Wives, Natal Families, and Children

239

T H E WOM A N ’ S AT T I T U D E A S T H E D E C I SI V E FAC T O R

Women’s subject positions within wife-selling were characterized by an unresolvable tension between wife as agent and wife as commodity. As Hill Gates puts it, with regard to the sale of daughters into marriage: “Even a bought person remains a person, with some rights and capacities that no nonhuman creature or object has.”75 Even a slave might exert some leverage, especially at the crucial juncture when an owner tried to market that slave to potential buyers.76 In Chinese wife sales, where did the balance lie between agency and commodification? No single formula can encompass all cases. Rather, a sliding scale of interest and emotion weighed multiple factors: a woman’s force of personality, her husband’s character, and the quality of their relationship; the proximity and attitude of her natal family or other allies; the degree of desperation a couple found themselves in; the presence or absence of children; and the availability of other men who might act as lovers, patrons, or buyers. A Sold Wife Reunites Her Family through Polyandry The following case draws together many of the themes of this chapter, in particular the decisive role that a wife’s attitude might play. Peasant Luo Yanchen (in his forties) was married to Wang Shi (28), and they had one son, Liangbaoer (5); the family lived in Hefei County, Anhui, but in the past Luo had found seasonal work at a tea farm in Huoshan County, where he got to know a single migrant laborer from Hubei named Zhang Chongxian (33) who also worked there. In 1738 there was drought and famine in their home county, so Luo and family walked to Huoshan County, begging for food and sleeping outside, and returned to the tea farm in search of work. Zhang was still working there, and they reconnected. Upon arrival, Luo fell ill, and he decided that selling his wife was the only way to secure his family’s survival. He offered to sell his wife to Zhang for a brideprice of 16 taels; the terms were agreed verbally, without recourse to a matchmaker or scribe; both men were illiterate, so Luo gave Zhang an otherwise blank piece of paper bearing his handprint and footprint to serve as a contract. Zhang promised to support the couple’s son until the age of 16 sui, at which point the boy would return to his father. According to Wang Shi’s testimony, she was very fond of her husband and did not want to leave him. She tried to talk him out of selling her: “Let’s just keep on begging together as a family and do our best to keep going (women dajia taofan ai shang qian qu)—anyway, whether we have food or not, I still want to follow you (you chi mei chi, wo shi yao genzhe ni de).” But for her sake and their son’s, he insisted: My husband said, “Zhang Chongxian and I are friends. Now I am sick, and I have only this one son. I’ve already come to an agreement with Zhang Chongxian that our

240

part two

son will follow you to be raised by him. He promises after two years to arrange for our son to go to school, and then when our son reaches 16 sui I can take him back. But if you refuse, then instead the three of us will all starve to death together. It would be better for you two to follow him and survive—and that way I can get a few taels to survive on, too.” I was crying, and I told him “Even if we starve to death, a husband and wife should stay together” (jiushi esi, fufu ye yao zai yi chu de).

But, eventually, “I realized I had no choice but to go with Zhang Chongxian, so I agreed.” After the transaction, Zhang continued to work at the tea farm, and Luo stayed in the area, often dropping by to visit his son; the two men remained good friends, and Wang Shi continued to sew and mend for Luo, who spent more and more time at their home. After about six months, however, Luo had spent the entire brideprice, so he approached Zhang once again; instead of demanding a supplementary payment, as we might expect, he asked if he could move in with them permanently. This proposal did not exactly thrill Zhang, but he agreed. In this way, a wife sale transmuted into a polyandrous relationship, so that the wife’s buyer ended up supporting and pooling resources with her entire family. Wang Shi played the critical role in this decision: she still felt close to her first husband, and she sought to reunite her family under Zhang’s patronage. Her account is revealing: Zhang “had no choice but to let my husband live with us” (zhi de liu zhangfu zhuzhe), because her willingness to cooperate with Zhang depended on his tolerance of Luo, whom she persisted in characterizing as “my husband.” Later on, in her courtroom testimony, Wang Shi summed up her relationship with her first husband Luo as follows: “As his wife, I bore him a son, and even after he sold me, right down to the present day, I have continued to make clothing and shoes for him to wear.” Bearing sons and making clothing were definitive duties of a wife, and by citing her loyal record on this score, Wang Shi was asserting that she had remained a good wife to Luo, wife sale and polyandry notwithstanding. After some months, Zhang decided to return home to Hubei, and he agreed to let Luo accompany Wang Shi and him and continue their arrangement, on condition that Luo pose as Wang Shi’s brother and keep his former relationship with her a secret. They stayed at Zhang’s sister’s home, and the two men earned money by casual labor. Some months later, however, the men quarreled, and Luo announced that he would take “his” wife back to Anhui. Zhang retorted that Luo would have to “redeem” her by refunding her brideprice: “Do you really think you can cheat me out of my money and get your wife back for free?” Their quarrel led to a fight, and Zhang killed Luo.77 This case is a variation on the theme of wife sale and polyandry being two options for solving the same sorts of problems. It also suggests that a seller’s con-

Chapter 7. Wives, Natal Families, and Children

241

tinuing sense of entitlement with regard to his wife’s buyer, if followed to its logical extreme, could actually lead to polyandry. The seller’s attempt to terminate the relationship by taking back “his” wife underscores his assumption that selling her had not severed their ties, that there was something inalienable about an asset so fundamental as a wife. The buyer’s response to this proposal—that the seller must “redeem” her if he wanted to take her away—suggests that he fell back on the terms of conditional sale as a formula for managing their separation. In this way, the relationship between these people moved along a continuum between the opposite extremes of outright sale and outright polyandry, finally returning to the intermediate scenario of conditional sale. Far from feeling a conflict of interest between her own uterine family and the patriarchal families of her two husbands, Wang Shi attempted to dissolve the boundaries between them by uniting them all in a single household. Her priorities were to ensure the survival of her first husband and their son and to keep the whole family together, and it is her attitude and feelings that played the decisive role in defining the relationship between her two husbands, and in fostering her first husband’s sense of entitlement. Poverty—A Partial Explanation at Best The point of departure for the story just narrated was poverty, exacerbated by drought, famine, migration, and illness. Wife sale and polyandry were not strategies undertaken by prosperous, comfortable people. But poverty alone cannot explain how these three individuals chose to solve their problems or how their story unfolded. In a crucial insight, Arthur Wolf and Chieh-shan Huang chide past scholars for resorting too casually to poverty as a blanket explanation for the behavior of Chinese peasants, in particular their practice of minor marriage (i.e., taking an adopted daughter-in-law instead of a grown-up bride). Poverty was certainly one factor in such decisions (since minor marriage cost far less than major marriage) but not necessarily the most important one, because the incidence of minor marriage did not correlate precisely with poverty. In Wolf and Huang’s view, adopting a daughter-in-law represented a “female strategy” that served to neutralize a potential threat to the mother-in-law’s uterine family: an adopted daughter-in-law would be raised by her mother-in-law, becoming, in effect, a member of that woman’s uterine family instead of an outside rival to it. In other words, the key factor in choosing minor marriage was the attitude and interests of the women who, as a practical matter, usually determined how their sons would marry.78 The vast majority of people in Qing dynasty China were poor, to one degree or another, but most did not engage in polyandry or wife-selling—or minor marriage either, for that matter. Also, although polyandry and wife-selling were both survival strategies driven by poverty, they had dramatically different effects on the

242

part two

people involved: polyandry kept a family together by adding an outside male, whereas a wife sale broke up the family and created a new marriage. Moreover, between the opposite poles of outright wife sale and outright polyandry, a variety of intermediate options existed. Granting that poverty was the basic reason for all of these strategies, why would individuals opt for any specific one over the others? A critical factor was the quality of the relationship between husband and wife, and very often, it seems, the wife’s attitude toward that relationship is what mattered the most. Did she want to stay with her husband, or did she prefer to leave him? If she did remarry, would she be able to take her children with her? If she had a lover whom she preferred to her husband, could he afford to buy her outright, or was helping to support her family the best he could do? Did she mind the stigma of having two husbands? The evidence clearly shows that it was difficult to make any of these strategies work without the wife’s cooperation. In particular, a woman could often prevent or sabotage a sale if she were unhappy about the prospect of being sold or displeased with the particular man to whom she had been promised. There are also cases in which wives refused to be sold, after which they recruited outside males to support their families through polyandry, with their husbands’ acquiescence. And there are cases like the last one narrated above, in which a sold wife managed to reunite her family through polyandry by making her cooperation with her second husband contingent on his acceptance of her first husband. As a practical matter, it might be the woman who had the final say.

8

Four Variations on a Theme

In considering wife sale thus far, our focus has been the open, direct form of the transaction: in other words, an honest exchange negotiated directly between the seller and buyer with the help of a matchmaker. This chapter broadens our inquiry to include four important variations of wife sale: compensated divorce (indirect sale facilitated by the natal family); widow remarriage; sale of a wife in her husband’s absence; and fraudulent wife sale. How did each work in practice, and what was the specific context for each? Why might people opt for compensated divorce instead of direct sale? What were the motives for fraud, and what forms did fraudulent sales take? Finally, what justifies inclusion of these kinds of marriage in the category “wife sale”? C OM P E N S AT E D D I VO R C E ( A . K . A . INDIRECT WIFE SALE)

With few exceptions, marital separation required both financial compensation for the husband and immediate remarriage for the wife. These two steps were linked, because the usual source of the first husband’s compensation was the new brideprice paid by the woman’s second husband. In other words, almost all marital separations constituted some form of wife sale, defined as the transfer of a wife from one husband to another in exchange for money. Broadly speaking, we can categorize wife sales as either direct or indirect. Direct sales often involved representatives of the woman’s natal family, whose approval might be formalized in their role as witnesses or even matchmakers, and who occasionally might even receive a share of the brideprice. But the defining 243

244

part two

characteristic of a direct sale was that the exchange of woman for money would take place between the two husbands, in a single transaction documented by a single contract. An indirect wife sale—which I also call “compensated divorce”—divided the exchange of woman for money into a sequence of two ostensibly separate transactions, divorce followed by remarriage, in which the woman’s natal family played an intermediary role. The first husband would return his wife to her natal family, along with a divorce contract, and he would either return or destroy their original marriage contract. The natal family would arrange a new marriage for their daughter and provide her second husband with a new marriage contract; they or their daughter would keep the divorce contract as protection against future trouble. The exact sequence of events varied, but, at some point in the process, the first husband would receive financial compensation, paid out of the second husband’s brideprice. Often, the second marriage would be arranged before the divorce was finalized, but sometimes a period of time elapsed between the two stages, during which the divorced woman would live in her natal household. Compensated divorce was well known throughout China, and it may have been as common as direct wife sale.1 The justification for calling compensated divorce a form of sale is that the first husband would not relinquish his wife without being paid, and the ultimate source of this payment was the second husband’s brideprice. In the most straightforward scenario, the natal family would obtain the new brideprice in advance of the divorce and simply transfer it to the first husband. An alternative was for the natal family to advance their own funds to compensate the first husband, later recouping this sum from the second brideprice as soon as a new marriage could be arranged. In both scenarios, the proceeds of the second marriage financed the dissolution of the first. Moreover, like contracts for direct wife sale, contracts for compensated divorce usually bear the first husband’s handprint.2 Terminology varied, but the most common term for a natal family’s payment of compensation in exchange for their daughter’s divorce was “to redeem the body” (shushen). The amount paid was known as the “body redemption price” (shushen jia) or simply “body price” (shenjia)—and this last term was also used in direct wife sales, interchangeably with the term “brideprice” (caili, caili qian). Such language is refreshingly frank about the nature of the transactions in question: “to redeem” (shu) is the same verb used for the payment of money to redeem conditionally sold land or pawned property; in other contexts, “to redeem the body” could mean to buy a slave’s freedom or to buy a woman out of prostitution by refunding the amount for which she had been “pledged” to the brothel keeper.3 An equally explicit term used in some regions was “mai nü gui zong”—literally, “to buy a daughter back to her natal lineage.”4 An 1891 case from Xinzhu County, Taiwan, provides a good basic account of how compensated divorce worked. Wu Laicheng (26) was married to Guo Cui-

Chapter 8. Four Variations on a Theme

245

niang, but the marriage was bitterly unhappy, and Cuiniang was on especially bad terms with her mother-in-law, Wu Lin Shi (61). Wu pressured Cuiniang’s father, Guo Lü, to accept a divorce. Guo’s plaint explains how this transpired: Several times they tried to pressure me to redeem her and remarry her to someone else, but I always refused. Finally, in the fourth month of this year, they asked their landlady Xu Chen Shi and a matchmaker to urge me to redeem her, in order to avoid matters developing in an unpredictable direction [a euphemism for suicide]. So I had no choice but to prepare a body redemption price (shushen jia) of 40 yuan, which I paid over via the matchmaker and Chen Shi to Lin Shi and her son, who received it in person. Then they drew up a body redemption contract (shushen ziju), and Laicheng wrote his mark on it, and his mother put her fingerprint on it. (I hereby submit it to the court as proof.) Then I took my daughter back and married her to Yang Rui.

Seven months later, Wu Laicheng and his mother filed suit accusing Guo Lü of abduction and Yang Rui of adultery (a transparent attempt to extort a supplementary payment). In response, Guo filed the above plaint, submitting his daughter’s “body redemption contract” in evidence, and yamen runners sent to investigate confirmed his account. In the end, the dispute was settled out of court before a formal hearing could be held.5 The case file does not say where Guo obtained the 40 yuan he paid to redeem his daughter. But, clearly, all parties understood that remarriage would immediately follow divorce, implying that the new brideprice was the source of the compensation. In other cases, the connection between divorce and remarriage is even more obvious. For example, in an 1858 case from Ba County, the woman entered her new husband’s household just two days after her divorce.6 In an 1836 case from Nanbu County, the woman’s divorce was followed on the very same day by a fraudulent remarriage (her new husband thought he was marrying a widow)—the two contracts bear the same date and were written by the same scribe, who was in on the fraud.7 In both cases, it is clear that the natal family had negotiated their daughter’s divorce and new marriage simultaneously, and had secured the second husband’s brideprice in time to use it to compensate the first husband when he relinquished her. Occasionally, a contract for a woman’s “redemption” names her new husband, making explicit his role in enabling her divorce. An 1896 specimen from Taiwan reads: I, Zhang Agui, of Xinbeijiao Village, hereby establish this contract for voluntarily letting my wife be redeemed out of our marriage. The situation is as follows: In previous years I took in marriage Wu Kuiniang. But now our household is not harmonious, and we are poor and have trouble obtaining our daily food, so I have no choice but to ask the matchmaker to arrange for her older brother Wu Hui to redeem her

246

part two

and take her back for 22 foreign silver dollars. This very day that sum has been paid over to me, Zhang Agui, and I have received the full amount in person. We have now agreed that she will be remarried (zhuan jia) to Lin Miao, of Xizhou Village, and they will be joined as husband and wife. This agreement is voluntary on both sides, and there is no coercion nor will either side change their mind or regret this. It will be like split bamboo, which can never be rejoined. There is no abduction or elopement involved. If any trouble should develop in the future, then Zhang Agui will exert all his strength to deal with it, and it will have nothing to do with Wu Hui at all. Fearing that spoken words are not reliable, and wanting to have evidence, I hereby establish this contract for redeeming my wife, and hand it over to serve as proof. Addendum: On this day I, Zhang Agui, in the presence of the matchmaker, have personally received exactly 22 foreign silver dollars, and I voluntarily agree for my wife to be redeemed and remarried to Lin Miao.

This contract bears Zhang Agui’s mark and is witnessed by the woman’s brother, Wu Hui. Presumably Wu drew up a second contract for her new husband, Lin Miao.8 Calculating the Amount of Compensation How was the amount of the first husband’s compensation calculated? It is not always clear, and practice certainly varied. Often, however, the compensation was simply a refund of the brideprice the first husband had originally paid for his wife. For example, in Mei County, Shaanxi, “the natal family will take their daughter back and return the brideprice her in-laws had paid for her.”9 In a 1742 case from Kaiping County, Guangdong, peasant Li Xuanman proposed to “divorce” (xiu) his unhappy wife: if her brother would “take her back and marry her off to someone else, then I can get a refund of her brideprice.”10 The practice of refunding brideprice confirms that these marriages had been sales of daughters by their parents— that is, “brideprice-heavy” marriage, to use Hill Gates’s term.11 In other cases, the first husband would be promised all or part of the brideprice from his wife’s new marriage, regardless of what he had originally paid for her. In Jingning County, Zhejiang, for example: The divorce contract will specify that in the future, when the wife remarries, a portion of her new brideprice will be paid to the first husband. . . . The amount he should be paid is decided at the time when he divorces his wife and is explicitly recorded in the contract. The woman’s natal family will give him a separate promissory note for that amount.12

An 1809 contract from Ba County stipulates that, when the woman’s new marriage is arranged, her new husband “will pay a brideprice of 5000 cash, and the matchmakers will pay the entire sum to her first husband.” It is not clear from the case record whether the new husband had already been identified at the time of the

Chapter 8. Four Variations on a Theme

247

divorce, or the first husband was simply promised that he would receive this amount at some future date.13 In an 1896 “body redemption contract” from Taiwan, the woman’s father agreed “to pay a body price according to the going rate,” determined through negotiations to be 200 large yuan of silver. In this instance, the woman’s father was fronting the compensation. Presumably, “the going rate” (shi zhi) that he agreed to pay reflected what he expected to receive for her new brideprice.14 Sometimes the natal family might try to exploit their intermediary role in order to profit. In an 1895 case from Baodi County, poverty compelled peasant Li Changchun (20) to “return” (tui) his wife, Wang Shi, to her natal family for remarriage. Wang Shi’s father, Wang San, negotiated a fraudulent new marriage, passing her off as a widow, to peasant Zhao Delu (28), who paid a “body price” of 110 strings of east cash. But Wang San paid his erstwhile son-in-law only fifty strings, assuring him that this was the total “body price” Zhao had paid, and he kept a profit of sixty strings for himself. Two months later, Li learned the truth and was outraged at how he had been cheated—he felt entitled to his wife’s entire new brideprice—and filed charges at the county yamen. The magistrate ruled this transaction to be an illegal wife sale (because the new marriage had been arranged before the divorce was finalized),15 but he opted to let the new marriage stand and punished no one. Nevertheless, he did redress what he saw as Li Changchun’s legitimate grievance. Wang San (who was also poor) had already spent the 60 strings of cash he had kept for himself, so the magistrate ordered Wang Shi’s new husband, Zhao Delu, to pay Li an additional twenty strings; he also ordered the matchmaker, who had negotiated both the divorce and the new marriage and was complicit in the fraud, to pay Li forty strings out of his own pocket. In other words, Li ended up receiving a total of 110 strings of cash, which was the full “body price” the natal family had received for his wife. This judgment shows that the magistrate understood the logic of compensated divorce and also expresses his compassion for the seller as loser in a wife sale.16 The redemption of adopted daughters-in-law represents a special case. The Private Law of Taiwan reproduces five contracts in which natal families redeemed adopted daughters-in-law who had reached the age of marriage (three specify 17 sui, a fourth 20 sui) but had not yet consummated their marriages; the stated reason for divorce in all five contracts is that the betrothed couple did not get along with each other.17 The amounts of compensation in these contracts are substantial and resemble those in other Taiwan contracts for compensated divorce, but since many adopted daughters-in-law were given as infants for free or for purely nominal sums, these prices presumably do not represent refunds of brideprice. How, then, were they calculated? Here, we can draw on Arthur Wolf and Chieh-shan Huang’s analysis of marriage practices in northern Taiwan, where adopted daughter-in-law marriage was

248

part two

particularly common. Peasants in Taiwan characterized the brideprice paid in major marriage as “compensation for the money invested in raising a daughter,” and, according to Wolf and Huang’s calculations, “the average price did in fact constitute an approximate compensation for the cost of raising the girl.”18 These observations concern “major marriage,” involving transfer of a grown-up bride from her natal home to that of her husband in exchange for brideprice. When asked about the advantages of taking an adopted daughter-in-law (“sim-pua” in Taiwanese) instead of a grown-up bride, informants spoke in terms of cash flow: Raising a sim-pua is like raising a pig. If you spend a little money and buy a little pig and then feed it each day, you don’t have to spend a lot of money at one time. When you sell the pig, you get a large sum of money. Raising a sim-pua is the same. If you raise a sim-pua, you have to feed her. But you don’t have to spend a lot of money when your son gets married.

Wolf and Huang conclude that an adopted daughter-in-law “was a kind of savings account” that enabled a family to avoid the large lump-sum payment necessary to acquire a grown-up bride.19 By extension, it is clear that a natal family hoping to redeem a grown-up adopted daughter-in-law would have to pay her in-laws a body price at the going rate, both to compensate them for the expense of raising her and (presumably) to ensure that they could afford another grown-up bride as her substitute. This would be true even if the natal family had received no brideprice for her in the first place. To borrow the analogy used by Wolf and Huang’s informants, redeeming an adopted daughter-in-law who had grown up was like buying a full-grown pig, which cost “a large sum of money.” Delayed Compensation in the Form of Supplementary Payments In a few cases, the first husband received no compensation from his wife’s natal family at the time of divorce. Instead, he managed to get it later from her second husband in the form of supplementary payments, and he had to approach the second husband directly in order to secure compensation. These cases beg the question of why a husband would yield his wife in the first place without guarantee of payment. The decisive factor, it seems, was the balance of power between the first husband and his wife’s natal family. An 1837 case from Ba County illustrates this scenario. A landless peasant named Li Tianyuan who worked as a tin smith was forced to divorce his wife, Shi Shi, and return her to her natal family without compensation. Li’s compulsive gambling exacerbated his poverty, and he had failed to support his wife properly. In 1830, the Shi family had staged a public intervention at which Li was forced to submit a humiliating “pledge to stop gambling and to

Chapter 8. Four Variations on a Theme

249

take care of my family” that was dictated by his in-laws. This document includes the following passage: Because I am addicted to gambling, I waste money and neglect my family’s food and clothing. My in-laws have scolded me many times but I have failed to reform. Finally, my wife’s older brother Shi Zhengdou seized me and tied me up in order to turn me in to the authorities for punishment, but I have begged the community to intercede and now I voluntarily submit this written pledge never to gamble again. In the future, if I repeat my bad behavior and fail to support my wife, if I ever play mahjong or dice again, then my wife’s older brother will take this pledge to the authorities and I shall submit to punishment without protest.

The pledge bears the names of seven witnesses, five being men from the natal family. But Li Tianyuan failed to reform, and two years later his wife’s family forced him to issue a contract renouncing their marriage. Li received no compensation; on the contrary, her natal family threatened to have him prosecuted for gambling (mere possession of gambling paraphernalia being subject to serious penalties).20 Their agreement to refrain from pressing charges was his only compensation. Three months later, Shi Shi married a widower named Yang Fengde, who was informed of her marital history. The Shi family did not ask brideprice, so they cannot be accused of profiting from their daughter’s divorce. Li Tianyuan may have felt intimidated by the Shi family, but he also resented the fact that he had been left “with neither wife nor money” (ren cai liang kong). Six months later, he approached her new husband and threatened to file charges of “forcibly seizing another man’s wife” unless he received a substantial brideprice. Yang refused to pay him, and instead threatened to call Li’s bluff by filing charges of his own. At this point, however, Yang’s neighbors intervened to mediate a settlement in which, at their urging, he agreed to pay Li 15 taels of silver “out of pity for his poverty.” In exchange, Li gave Yang a “submission contract to prevent disaster” bearing names of sixteen witnesses, including several members of Shi Shi’s natal family. In it, Li confesses that “because Shi Shi’s divorce came first and Fengde took her in marriage only later, I am well aware that their marriage is correct and my accusations false.” He promises no further trouble and offers to submit to “double punishment” if he breaks his word. A couple of years later, however, Li demanded a supplementary payment, and when Yang rebuffed him, he filed false charges at the county yamen. In response, Yang and Shi Shi’s brother filed counter-plaints and submitted the relevant documents to prove their claims. The magistrate ruled Shi Shi’s divorce and new marriage legal, and ordered Li beaten for attempting extortion and filing false charges. In this case, the Shi family was able to force Li to divorce his wife without compensation because of their realistic threat to have him prosecuted for gambling. Moreover, the Shi family could back their threats with muscle: Shi Shi had several

250

part two

brothers and male cousins who were willing to defend her interests and could also mobilize their community on her behalf. In contrast, Li appears to have been entirely on his own, which helps explain why the Shi family could treat him with such contempt. Nevertheless, it is striking that Yang Fengde’s neighbors persuaded him to pay Li so much money. Evidently, they judged Li’s grievance legitimate—no man should have to divorce his wife without compensation—and 15 taels represents what they considered a fair price. (It falls into the mid-range of prices in direct wife sales in Ba County during this period.) Mediated settlements reflected the wider community’s sense of fairness as well as pity for the men who were the ultimate losers in these exchanges.21 To conclude, let us consider the motivation for choosing this indirect form of wife sale over the direct form. First of all, compensated divorce carried less criminal liability and perhaps less stigma as well. In fact, as long as the wife was divorced and returned to her natal family before they collected her new husband’s brideprice, magistrates usually upheld such transactions (see Chapter 9). Anecdotal evidence suggests that many people understood the distinction. For example, in a 1903 case from Nanbu County, Li Guozheng’s relatives and neighbors warned that it was against the law to sell “a wife who has a living husband” and urged him to negotiate a compensated divorce. Heeding their advice, he persuaded her natal family to redeem her for nine strings of cash, and they arranged her new marriage themselves.22 Nevertheless, in most cases, it appears that the critical factor was the woman’s natal family: their concern for their daughter’s well-being and their ability to influence her husband’s behavior. We have already seen in the previous chapter how a natal family might help their daughter escape an unhappy marriage. Playing an intermediary role in a compensated divorce was another way they could help their daughter and minimize the risk that she might be mistreated. Many of these cases show natal families intervening, sometimes aggressively, to protect their daughters’ interests. But the natal family’s motivation mattered only if they were in a position both to influence their son-in-law’s actions and to secure the cash flow necessary for compensated divorce. If they had a good relationship with him, or could impose their will through force or threats, then they would likely opt to facilitate a compensated divorce (as long as they could manage its financing). If they were in a weaker position, they still might play a constructive role as witnesses or matchmakers to a direct sale. W I D OW R E M A R R IAG E

The forms of widow remarriage closely mimicked the unorthodox marriage practices covered in this study. Broadly speaking, we can divide widow remarriage into

Chapter 8. Four Variations on a Theme

251

two categories: those in which the widow left her first husband’s household to move into that of her second husband, which closely followed the model of wife sale; and those in which the widow “brought in” (zhao) a second husband without leaving her first husband’s household, which closely resembled polyandry. Underlying all these practices was the fact that a husband’s death did not release his wife from her bond to his household. As Tai Yen-hui observes, “when a widow wished to return to her own family or remarry, she first had to obtain what amounted to a divorce from her deceased husband’s family.”23 On the rarefied level of Confucian discourse, this bond was understood as an absolute, lifelong commitment that paralleled both filial piety and political loyalty: “to follow one husband to the bitter end.”24 On the mundane level of the poor peasant household, this bond was understood in practical terms: it had taken a big investment to get a wife in the first place, and that investment had to be recouped if she were to be released in order to remarry. Let us first consider remarriage in which a widow left her first husband’s household to move into that of her second husband. Remarriages of this kind generally followed the model of either direct or indirect wife sale. Needless to say, a dead man was in no position to sell his own wife. But his surviving family (or even the widow herself) could act in his stead, and the basic structure of these transactions was the same as that found in wife sales: separation plus remarriage in exchange for compensation, with the source of that compensation being the brideprice paid by the woman’s new husband. Widow Remarriage Modeled on Direct Wife Sale Investigation of Customs provides several candid descriptions of remarriage that followed the model of direct wife sale. In Licheng County, Shanxi, for example: When a widow remarries, an appropriate body price (shenjia) will be negotiated by her parents-in-law or brother-in-law who authorize the marriage, and only after they are paid can the marriage take place. This payment is called a “body price” (shenjia), and in fact this practice is no different from the buying and selling of people (yu maimai renkou wu yi).25

In a discussion of “vulgar marriage customs,” the 1936 gazetteer of Fengshan County, Guangxi, explicitly associates this form of widow remarriage with a husband’s sale of his wife: “There are two transactions that both amount to the same kind of traffic in persons (er zhong wei yi maimai shi): the first is the remarriage of a wife who has a living husband, and the second is the remarriage of a widow.” In both transactions, “the payment for the woman is referred to as ‘body price.’ ”26 The Private Law of Taiwan contains eleven contracts for widow remarriages that follow the model of direct wife sale.27 The following example from 1901 is typical:

252

part two We, Zhong Baoting of Simanshui Village, and Zhong Dailiang of Wumanshui village, hereby establish this contact for marrying out our sister-in-law. The situation is as follows: In the past, our brother Zhong Dinglang took the niece of Xu Cheng’er of Xindongshi Village as his wife. But because her husband died young, she cannot make ends meet on her own. Her given name is Chunmei and she is now 42 sui. Now, we, Zhong Baoting and Zhong Dailiang, voluntarily issue this contract to remarry our sister-in-law to Wen Delang of Binlanglin Village as his wife. On this day, relying on the matchmaker, we have met to discuss the terms and have agreed that the body price will be exactly 12 big yuan. That sum of money is now exchanged for this contract. If anything from the past should be unclear, then it will have nothing to do with Wen Delang; instead, we will exert all our strength to deal with any such challenge. After marrying out, our sister-in-law will forever be Wen Delang’s wife, and in the future we will never dare to say anything different, cause trouble, or change our minds. Fearing that spoken words alone are not reliable, we establish this contract for voluntarily marrying out our sister-in-law to serve as proof.28

The contract is witnessed by the matchmaker and scribe, and endorsed by the woman’s brothers-in-law. The reference to “anything from the past” being “unclear” is a euphemism for someone making claims about the woman’s prior marital status in order to challenge the validity of her new marriage (similar statements appear in many wife sale contracts). The language of this contract, the negotiations it records, and the structure of the transaction are all nearly identical to those found in wife sales, except that the woman is sold by her brothers-in-law in place of her deceased husband. Some contracts specify that the brideprice will be used on behalf of the dead husband—for example, an 1897 “contract for receiving the body price silver for widow remarriage,” also from Taiwan: The year before last my younger brother passed away, leaving his wife Zheng Shi, whose given name is Yuanniang, along with a son of three sui named Chen Zao. She is burdened with her husband’s debts, and mother and child have difficulty making ends meet. For this reason, she is willing to remarry to a respectable man. A matchmaker has negotiated for Zheng Yuanniang to remarry to Chen Qingfu as wife. On this day, Qingfu has prepared a body price of 34 big yuan of silver and has paid the entire sum to me, Chen Wen, into my own hands. I, in turn, have handed the money over to Zheng Shi so that she can pay off my younger brother’s debts (yi huan zhaiwu qing qi).29

An 1888 contract from Taiwan explains that the widow’s brother-in-law will use her brideprice “to take care of all of my older brother’s expenses (baoxiong zhi zhu fei).”30 Among the poor, a widow’s remarriage might well be negotiated before her first husband was even in the ground, and anecdotal evidence suggests that the brideprice was often used to pay for his coffin and funeral.31 In this way, the first husband might actually “receive” payment for his wife, despite being dead.

Chapter 8. Four Variations on a Theme

253

The stigma and taboos surrounding widow remarriage mirrored those for ordinary wife sale. In many regions, it was necessary to deliver the widow to her new household at night, as discreetly as possible. As the 1942 gazetteer from Chong’an County, Fujian, explains: “A widow who remarries must be delivered at night. She will be carried by lamplight until water has been crossed, and only then will she enter the sedan chair. Her feet must not touch the ground, because if they touch the ground the plants will wither and the soil will lose its fertility.”32 As in the case of a sold wife, the loss of the woman’s fertility to her husband’s household reverberated in a symbolic threat to crops and soil. Widow Remarriage Modeled on Compensated Divorce Another common form of widow remarriage closely followed the model of compensated divorce, in which the woman would be redeemed by her natal family and married off in a two-step process documented by separate contracts. Investigation of Customs records this version of widow remarriage in a number of regions. In Shanxi, for example, “When the husband of a married daughter dies, then if she is willing to remarry, her natal family must first pay an appropriate body redemption price (shushen jia) in order to redeem her and return her to her natal home. Then they will marry her to a different man.”33 Similarly, in Hubei, “When a woman’s husband dies, if her natal family has enough money, they can negotiate with her husband’s family and pay an appropriate sum of money to redeem their daughter. After that, she may remarry or remain chaste, and her husband’s family will not interfere.”34 The Private Law of Taiwan includes six contracts for the “redemption” of widows by their natal families. The following specimen from 1895 is typical: I, Li Ade of Xishi Village, hereby establish this contract for voluntary redemption. The situation is as follows: In years past, my older brother Li Agui took Lin Damei, the daughter of Lin Dingchang of Neibu Village, as his wife. But now Li Agui has died, and my sister-in-law Damei is young and should return to her natal home in order to remarry. Now, Lin Dingchang’s wife Chen Changmei has engaged Li Axin of Neibu Village [to serve as matchmaker], and he has explained that Chen Changmei is willing to pay 20 yuan in silver in order to redeem Lin Damei so that she can return to her natal family and remarry. Both sides are acting voluntarily, and we will never dare to change our minds. No relatives or others will ever dare to contradict this agreement or to cause trouble. Fearing that spoken words alone are unreliable, I hereby establish this contract for voluntary redemption and give it to [Chen Changmei] to keep as proof. Note: The original marriage contract is lost so it cannot be returned.35

As the addendum suggests, it was standard for the in-laws to return the original contract for the widow’s first marriage, which would then be destroyed. The redemption contract would be kept by the natal family or the widow, to protect against any challenge from her first husband’s family. As with compensated

254

part two

divorce, a second contract would be prepared for her new husband. The Private Law of Taiwan does not provide examples of these second contracts. But a “redemption contract” from 1905 names the second husband, so it is obvious that this widow’s new marriage had been negotiated in advance, along with her redemption, as part of a single package. The Japanese investigator who collected this contract notes that the new husband had paid the widow’s body price in advance, enabling her natal family to redeem her and then transfer her to him, apparently on the same day.36 In some regions, there was a hybrid practice in which the widow would be remarried directly from the household of her in-laws, who would receive her entire brideprice, but her natal family would choose her new husband. This was a one-step process involving only one contract. In Hunan, for example: “When a widow remarries, first her natal family must choose her husband. . . . Then the inlaws will negotiate a body price and draw up a contract for widow remarriage. Then the man taking her in marriage will pay the body price and cover all other expenses. The in-laws receive all the money.”37 In Hubei, the colloquial term for this practice was “the natal family chooses the man, and the in-laws get the money” (niangjia ze ren, pojia de qian).38 If the widow was being married directly from her first husband’s household, why did her natal family participate? The report from Hubei explains: This practice makes sense in terms of human feelings (renqing). The natal family can be expected to feel more concern for the fate of their daughter than her in-laws will, and the natal family will pay closer attention in choosing her new husband. Because they understand the important influence of her new husband on her happiness, the natal family is more likely to make a good choice.39

This hybrid practice seems to have been a compromise sometimes reached when a natal family cared deeply about their daughter’s security, and were in a position to assert themselves, but for whatever reason could not front the money to redeem her. In fact, this explanation applies more broadly to the choice of compensated divorce over direct wife sale. By playing an intermediary role, the natal family could minimize the risk that their daughter would end up with a man who might abuse her or fail to support her. Uxorilocal Remarriage as a Form of Polyandry Up to now, we have discussed forms of remarriage that required a widow to leave her dead husband’s household. An alternative was uxorilocal remarriage, in which the widow would “bring in” (zhao) a new husband without leaving her first husband’s household. To do this, a widow needed the approval of her in-laws from the first marriage, but that was often forthcoming because in certain circumstances uxorilocal remarriage offered an optimal solution to the problems of both the

Chapter 8. Four Variations on a Theme

255

widow and her in-laws. For this reason, uxorilocal remarriage was common throughout China.40 To understand the logic of uxorilocal remarriage, we must bear in mind two rules. As we have seen, if a widow were to leave her dead husband’s household, her in-laws would have to be compensated. But also, if a widow left her dead husband’s household in order to remarry, then by law and custom she would surrender all claim to his estate and their children, both of which remained in the custody of his lineage. For this reason, widows with property were particularly reluctant to remarry in the conventional way.41 In practice, it was not uncommon for a widow to take children into a new marriage if they had no other means of support, but, as in wife sales, they would often be expected eventually to return to their father’s lineage. Uxorilocal remarriage enabled a widow to get around both of these inconvenient rules, allowing her to maintain custody of her husband’s estate and their children while also enjoying the benefits that a new husband could bring.42 But there were also advantages from the standpoint of the dead husband and his lineage: like polyandry, uxorilocal remarriage was a strategy to preserve his household intact by enlisting the support of an outside male, whereas conventional remarriage (like a wife sale) would break up his household, putting both his children and their inheritance at risk. Conventional remarriage would also deprive the dead man’s aging parents of his wife’s support—a crucial consideration if they had no surviving sons. Moreover, if the deceased had left no son to carry on his family line, the uxorilocal husband could provide one: the first son he sired would be designated the dead husband’s heir, in a practice that parallels surrogate fatherhood within polyandry.43 The widow would continue to look after her dead husband’s ancestral tablets and graves. Moreover, it was common for the second husband to adopt the first husband’s surname (a practice sometimes seen in polyandry).44 These considerations induced many in-laws to permit widowed daughters-in-law to bring in uxorilocal husbands. Indeed, contracts for uxorilocal remarriage were often drafted in the name of the widow’s in-laws.45 Seen in this light, uxorilocal remarriage constituted a form of polyandry in which the first husband happened to be dead, but with whom the widow had by no means severed her ties, and the arrangement was designed to secure his interests and those of his lineage as well as her own. Terminology underscores the close parallel with “getting a husband to support a husband.” Colloquial terms for uxorilocal remarriage included “getting a husband to raise [the first husband’s] sons” (zhao fu yang zi or zhao fu fu zi), “getting a husband in order to give birth to sons” (zhao fu sheng zi), “getting a husband to carry on [the first husband’s] family line” (zhao fu chuan hou), and “getting a husband to support [the first husband’s] parents” (zhao fu yang qin or zhao fu yang lao).46 All of these terms emphasize the interests of the dead husband as well as of the widow herself.

256

part two

Conditional Widow Remarriage in Zhejiang In Zhejiang, there was yet another form of widow remarriage, based on the model of conditional wife sale (dian qi), which represents a sort of compromise between the alternatives of leaving or remaining in the dead husband’s household. This distinctive regional practice underscores the broader point that the various forms of widow remarriage closely resembled and were indeed based upon the spectrum of non-normative marriage practices analyzed in this study. In the classic version of conditional wife sale, a woman would move in with her conditional buyer and share his bed for a fixed period, during which any children she bore would belong to him; after the time was up, she would return to her husband. In exchange, her husband received a loan, which he would usually have to repay when the term was finished in order to “redeem” his wife. Her sexual and reproductive labor would constitute interest on this loan. According to Investigation of Customs, which documents the practice of conditional wife sale throughout Zhejiang, there was a variation in Fenghua County in which a widow who could not support her children would conditionally sell herself to a widower, or to a man who lacked a son but was too poor to take a permanent wife, for a period of up to ten years. According to the report, conditional remarriage was usually a strategy for widows who had children from the first marriage and did not want to be permanently separated from them, whereas a widow without children would simply remarry in the conventional way, leaving her first husband’s household altogether. Conditional remarriage did not necessarily require a woman to move in with her buyer, as long as she made herself available for conjugal relations.47 We find an example of precisely this practice in an 1885 homicide case from Yin County, which happens to border on Fenghua County (both are located in Ningbo Prefecture). In this case, a woman named Zhuang Ying Shi (aged 38 sui at the time of the transaction) was left in poverty upon her husband’s death, with no means of supporting their young son, so she decided “to sell herself conditionally to another man to be his wife” (dian gei yu ren wei qi). An older female cousin agreed to act as matchmaker and quickly found a peasant named Shi Jinyou, who at the age of 39 sui had no wife or family and was “hoping to take a wife by conditional sale.” Shi agreed to pay a “conditional sale price”—also referred to in testimony as the widow’s “body price”—of 60 foreign silver dollars in exchange for “having relations” (laiwang—a euphemism for sexual intercourse) with the widow for a term of ten years. When that time was up, she would “once again return to her first husband’s household.” The transaction concluded amicably, and Zhuang Ying Shi moved in with Shi Jinyou. But, the following spring, Shi fell ill and, because he could not work, he fell into debt and eventually lost his land, sinking into poverty. The household often

Chapter 8. Four Variations on a Theme

257

went short of food. Zhuang Ying Shi very much resented this situation, because she had agreed to the conditional sale precisely in order to avoid poverty. After considerable quarrelling, she finally left Shi Jinyou and returned to her dead husband’s household; she had lived with Shi for just over one year. Twice Shi Jinyou attempted to fetch her back, but she refused him. Then, one day, she remembered that she had left her fire tongs at his house, and she went to get them; when Shi could not produce the tongs (he had sold them), she first demanded money and then, when Shi refused to pay her, she seized some of his clothing in compensation. Shi became infuriated and beat her with a wooden stake; she later died of her injuries. Unfortunately, the terse record of this case leaves out much we would like to know—for example, it does not mention the widow’s in-laws, nor does it say who received the money that Shi paid for her, or what was done with it. But her son of 13 sui did not follow her into Shi Jinyou’s household, instead continuing to live at his own father’s home. Evidently, someone from the first husband’s family was on the scene to look after him, and, since the widow’s principal motive for seeking a conditional buyer was to provide for her son, it appears that the boy’s living expenses were being covered by the 60 silver dollars that Shi Jinyou had paid.48 The case of Zhuang Ying Shi again confirms what we have seen in many other cases, namely that peasant women were capable of using sex in an unsentimental, strategic way to bargain for resources from men. Clearly, this pragmatic widow assumed that her conditional buyer’s ability to support her during their years together was a basic condition of the deal, on top of the “body price” he had initially paid. She drove a hard bargain: once it became clear that Shi Jinyou could no longer put food on the table, she cut off his sexual access and moved out, almost nine years ahead of schedule, without refunding the brideprice he had paid. To her way of thinking, nothing required her to stay with a man who could not support her. Conditional wife sale, like the conditional sale of land, was a compromise strategy that enabled peasants to use a vital asset to raise short-term credit while retaining the option of eventually recovering that asset. The widow remarriage variant followed the same logic, enabling a widow to generate income vital to her family’s survival without permanently leaving her dead husband’s household. In the end she could return, and her temporary absence would have suspended but not voided her status there. However, this arrangement appears to have been unusual outside Zhejiang, at least as a formally contracted form of widow remarriage that received community recognition. A less formal and apparently far more common survival strategy was for widows simply to engage in transactional polyamory or retail prostitution. Many cases record rural widows trading sex for support from a handful of regular patrons, and the Ba County cases show that many of the prostitutes in Chongqing brothels were young widows.

258

part two

To summarize, the survival strategies of widows closely mirrored the spectrum of practices covered in this study, including direct wife sale, compensated divorce, and polyandry as well as conditional wife sale. Among the rural poor, almost all widows remarried, and they did so according to one of these strategies; for those who did not remarry, prostitution was a common resort. These facts underscore the mainstream relevance of our subject matter: it will not do to dismiss polyandry, wife sale, and the rest as deviant outliers. Rather, they constituted an integral part of a broader system of marriage and household formation.

T H E W I F E S O L D I N H E R H U SBA N D’ S A B SE N C E

Sometimes a man’s family would sell his wife in his absence. These sales were hybrid transactions. They resembled widow remarriage, in that the husband’s family would receive compensation in his place and sometimes truly believed him to be dead. They can also be understood as a form of indirect wife sale, because, like compensated divorce, such transactions usually involved two steps and often two separate contracts. In this scenario, a husband would travel far from home in search of work, leaving behind his wife to depend on her in-laws’ charity. During the husband’s absence, his family would struggle with poverty, and, without word of his fate, they might suspect that he had perished; eventually, they would decide to sell his wife. Investigation of Customs describes this practice in Shandong, where many poor peasants migrated seasonally or for longer periods in search of employment, often to Manchuria: In poor families there are some men who, after taking a wife, will leave home and travel far away in search of a living. Some may not return or even send home any news for several years. Their wives will be left with no means of feeding themselves and no way to survive. In this situation, a wife can ask her parents-in-law for permission to remarry, and if they agree, they will supervise her remarriage and will receive an appropriate body price for her.49

In the end, however, the woman’s husband might return, posing the problem of how to compensate him for the loss of his wife. These cases all seem to involve land-poor peasants who had become “proletarianized” in the context of agricultural involution: a worsening ratio of population to resources compelled men to hire out labor in order to maintain household incomes at subsistence level. The men in these cases were usually young (in their early twenties), married but without children, and whatever land they had inherited was inadequate to support their households. It was the need for work that induced men to leave home, and it was the poverty of the families left behind that forced the sale of their wives.50

Chapter 8. Four Variations on a Theme

259

What happened when these men did return home and discovered that their wives had been sold? According to a report from Gaoping County, Shanxi, “if, later on, the first husband does return home, then he must be paid the full brideprice specified in the contract for the second marriage, so that he can take another wife.”51 But given the poverty of most people in such circumstances, how could they avoid spending the brideprice? What if none of the money was left when the husband returned? Some families did try to preserve the brideprice, so that the absent husband would have means for a fresh start if he returned. In an 1815 case from Baodi County, peasant Liu Jinkui left home for several years in search of employment; in his absence, he expected his older brother Liu Lan to support his wife, Sun Shi. But after several years, during which Jinkui failed to send home word of his fate, let alone any money, his brother and Sun Shi finally concluded that he must have died; moreover, Liu Lan (who was also quite poor) wanted relief from the burden of supporting Sun Shi. Informing the community that Jinkui had died, Liu Lan arranged for his “widowed” sister-in-law to remarry for a “body price” of 230 strings of east cash. She agreed with this plan. However, two elders of the Liu lineage voiced concern, because there was no proof of Jinkui’s death. They did not oppose Sun Shi’s remarriage, but they insisted that Liu Lan invest her body price in land, so that, if Jinkui did return, he would not end up “with neither wife nor money.” After another two years, Jinkui did indeed finally return home. Thanks to the elders’ intervention, he received the full body price that his wife’s new husband had paid. Satisfied, Jinkui did not subject that man to any new demands.52 In an 1824 case from Nanbu County, we see a similar effort to preserve at least part of the brideprice for the absent husband. In this case, peasant Pu Hongfu (25) was married to He Shi, and the couple had no children. Because of poverty, Pu left home in search of work and did not return for three years; during that time, He Shi had no means of support other than family charity, and she became increasingly unhappy and defiant of her in-laws. Finally, Pu’s father and He Shi’s father agreed that the situation was untenable, and with He Shi’s approval they arranged for her to remarry as wife to a single peasant named Pu Changyin (no relation to her first husband). The contract for this marriage is written in the voice of Pu Hongfu’s father and is witnessed by representatives of his lineage, along with He Shi’s father and brother: Pu Tingmo hereby establishes this marriage contract to authorize marriage. The situation is as follows: My third son, Pu Hongfu, took He Congyuan’s daughter He Shi in marriage. But since childhood, Pu Hongfu would not behave properly according to his station. He has neglected his father and his wife and has wasted many years away from home without returning or sending word, so that we do not know whether he is alive or dead. He left his wife He Shi at home without supplying her food or

260

part two

clothing. In addition, there has been drought this year, and we are left without recourse. He Shi is willing to remarry, so I have engaged Pu Maochun to act as matchmaker and find a new household for her. The matchmaker has negotiated an agreement whereby she will marry Pu Nengyuan’s son Pu Changyin as wife. Matchmaker Pu Maochun has already visited He Shi’s natal family to present them with gifts of wine, and her father, He Congyuan, and brother He Sanchao have happily agreed to the match without any reservation. Through the matchmaker, we have agreed that Pu Nengyuan will pay exactly 6000 copper cash. Now he has paid this amount and I have received it. I, Pu Tingmo, hereby declare before the assembled witnesses that I will save 4000 cash for my son Pu Hongfu, so that if he does return home he can take another wife. In that case, he will not be allowed to cause any trouble for Pu Nengyuan or his son. The remaining 2000 cash will be used to pay for funerary garments (lao yi zhi zi) for my wife and myself. On this day the terms have been agreed and the marriage contract drawn up. Water, once spilled, cannot be gathered up (fu shui nan shou): what is done cannot be undone. If anyone surnamed He or Pu, from either natal or affinal families, or any relative near or far, should cause trouble, then I, Pu Tingmo, will take full responsibility, and it will have nothing to do with Pu Nengyuan or his son. Now, fearing that people’s hearts cannot be controlled, I specially establish this marriage contract, relying on the matchmaker, to serve as proof forever. Natal family: He Congyuan [woman’s father] [mark] He Sanchao [woman’s brother] [mark] Matchmaker: Pu Maochun [mark] Witnesses: [nine men from Pu lineage] Scribe: Pu Zhongyuan On Daoguang 4.6.14, Pu Tingmo [mark] establishes this marriage contract, which is true.

In this contract, Pu Tingmo acknowledges that the signatories do not know whether Hongfu is alive, and he berates his son for having abandoned his family without support. Nevertheless, he anticipates the possibility of his son’s return by setting aside two-thirds of the brideprice for him. In the event, Pu Hongfu did return home, just a month after his wife had remarried, but he was not satisfied with the terms negotiated in his absence. Through mediators, he persuaded He Shi’s new husband to pay him an additional 7000 cash, in exchange for a new handprint contract in Pu Hongfu’s own name to ratify the new marriage. This new contract reads as follows: Pu Hongfu hereby establishes this handprint contract for expelling my wife. The situation is as follows: I took He Shi as my wife, but she has repeatedly failed to behave as a wife should. She is unfilial to my parents, she does not respect her husband and master, and in years past she and I did not get along with each other. I had to leave home in order to earn a living and was away for three years. To my surprise, in my absence this wife of mine was even more defiant of discipline, running around as she

Chapter 8. Four Variations on a Theme

261

liked, contrary to morality (dong zou xi qu, you guai fenghua). Finally my parents agreed with her natal family that she should be released to become Pu Changyin’s wife. They had not expected that I would return home. After my return, I decided that since this wife of mine has never peacefully accepted her role, I am happy to issue this handprint contract, and through mediators Changyin has agreed to pay me another 7000 cash. I have received this full amount and nothing more is owed. After establishing this contract, it will be like water, once spilled, which cannot be gathered up: what is done cannot be undone. I promise never to cause any more trouble or make further demands on Pu Changyin. Now, fearing that people’s hearts are not as of old, I specially establish this handprint contract for him to keep forever as proof. Mediators: [names of ten men, of whom eight are surnamed Pu, one surnamed He, and one surnamed Li; four of the Pus also served as witnesses for the first contract] [left handprint] On Daoguang 4.7.24, Pu Hongfu hereby establishes this handprint contract [mark]

This contract effectively converted the transaction into a direct wife sale. Its rhetoric is strikingly different from the first contract that had been drawn up in Pu Hongfu’s absence by his father. Whereas Hongfu’s father had put the onus for the sale on his absent, irresponsible son, in the new contract Hongfu blamed everything on his wife’s alleged misbehavior: if only He Shi had been a loyal wife and a filial daughter-in-law, her sale would not have been necessary. In fact, this account of He Shi’s disobedience and defiance probably contains some truth (however selfserving it may seem), since it fits the model of an unhappy wife who sought to provoke her own sale by going on strike.53 In the following 1808 case from Changyuan County, Zhili, the absent husband’s family seems genuinely to have believed him to be dead; nevertheless, when he unexpectedly returned home, they did their best to compensate him. A landless peasant named Wang Xi had traveled to Henan in search of work, and in his absence, his wife, Zhang Shi, depended on his older brother, Wang Dong, for support. But Wang Dong was also very poor, and after three years with no news, he decided that his brother must be dead, so he sent Zhang Shi back to her natal family with instructions that they should arrange for her to remarry. Zhang Shi’s father negotiated her sale as concubine to peasant Zhang Guoliang (whose wife had no issue) for a body price of 160,000 “large cash”; half of this sum was handed over to Wang Dong, apparently as a refund of the original brideprice the Wang family had paid for Zhang Shi when she married Wang Xi. To everyone’s surprise, however, Wang Xi returned home just a month after his wife’s sale. His brother gave him 45,000 large cash, which was all that remained of the Wang family’s share of the brideprice. But Wang Xi was not satisfied, so he visited his wife’s new husband and demanded further compensation. Zhang

262

part two

Guoliang agreed to pay an additional 25,000 large cash, and in exchange Wang Xi confirmed his wife’s new marriage. He later used the compensation he had received to take a new wife.54 In a 1786 case from Ju Department, Shandong, the family was simply too poor to preserve the brideprice for the absent husband. In this case, peasant Wang Wencheng (27) had married Bo Shi (26) in 1779. The couple was very poor and had no children; they lived with Wang’s mother, Wang Jing Shi, and grandfather, Wang Youcong (82). In 1782, Wang Wencheng had left home alone and traveled to southern Manchuria in search of work, not returning for several years. He was able to survive as a casual laborer but did not save any money, much less send any home; in his absence, his family became desperate (witnesses testified that “they had nothing to eat”). Finally, Wang Youcong and Wang Jing Shi persuaded Bo Shi that it would be best for everyone if she remarried “in order to find a way to survive” (xun sheng lu). She agreed, as did her father Bo Zeng, and in the spring of 1783 she was married to peasant Zhang Cunliang (41) for a brideprice of 14,000 small cash. The Wang family received the entire brideprice. We have a transcript of the contract: Wang Youcong and his daughter-in-law Wang Jing Shi hereby establish this contract for remarriage. The situation is as follows: Grandson Wang Wencheng left home four years ago and we have had no news of him since then. He left his wife Bo Shi behind with no means of support and she is having a hard time getting by. Now we have invited matchmaker Chen Zhanggong to discuss the situation with us and he has negotiated a satisfactory match. Bo Shi is willing to remarry to Zhang Cunliang as his main wife. We have agreed on a brideprice of 14,000 cash, and woman and money have now been exchanged. Fearing that in future there would be no proof, we hereby establish this remarriage contract for Zhang Cunliang to save as proof. On Qianlong 47.4.16, Wang Youcong and Wang Jing Shi establish this remarriage contract.

The contract frankly explains that although Wang Wencheng may still be alive, his wife must remarry because of poverty. Its format closely resembles that for an ordinary wife sale, but the tale of woe explaining the reason for the woman’s remarriage focuses on the husband’s long absence, which exacerbates the household’s poverty. Instead of a husband selling his wife, his family is acting in his place. Six months later, however, Wang Wencheng returned home, and by that time the entire brideprice had been spent on food. Although Wang did not dare object to what his grandfather and mother had done in his absence, he could not help resenting that he had nothing to show for the wife he had lost. After a few months, he contacted Zhang Cunliang and threatened to file charges against him for “taking in marriage a wife with a living husband.” Wang’s brother-in-law and a close friend of Zhang’s stepped in to mediate an agreement whereby Zhang paid Wang

Chapter 8. Four Variations on a Theme

263

Wencheng a “supplementary payment” (zhao jia) of 12,000 small cash in exchange for a new contract in Wang’s name that confirmed Bo Shi’s marriage to Zhang. We have a transcript of this document, as well: Wang Wencheng hereby establishes this contract. The situation is as follows: I was unable to make a living, so on Qianlong 44.3.20 I left home and went east of the pass in search of work. That was four years ago. In my absence, my family had difficulty getting by, so my grandfather Wang Youcong remarried my wife Bo Shi to Zhang Cunliang as his wife, for an agreed brideprice of 14,000 cash. At that time, the brideprice was paid in full. Then, on Qianlong 47.10.20, I, Wang Wencheng, returned home and I asked for more brideprice. My younger sister’s husband Du Fanghe together with Ji Shengde mediated a settlement whereby Zhang Cunliang paid an additional 12,000 cash. Now, on Qianlong 48.3.7, he has paid that amount in full, and I will never regret or betray our agreement. Fearing that in future there would be no proof, I hereby establish this contract for Zhang Cunliang to save as proof. Qianlong 48.3.7, Wang Wencheng establishes this contract.

This settlement ended the dispute, and before long Bo Shi bore her new husband a son.55 A wife’s remarriage in her husband’s absence can be seen as a form of indirect wife sale, in that the absent husband’s family would sell her in his place, and he would be compensated only later (with the original brideprice and/or some supplementary amount added by the new husband). Such transactions most closely resemble a widow remarriage brokered by the widow’s in-laws. This resemblance underscores the similarity between widow remarriage and wife sale that we have already seen. The returned husband’s demand for more money is a variation of the demand for supplementary payments found in other forms of wife sale—some people referred to this demand as “zhao jia.” Moreover, this scenario resembles fraudulent wife sale in that the new brideprice would be solicited (some might say “extorted”) in exchange for a new contract to rationalize a transaction that had been based on a false premise, namely that the first husband was dead or otherwise out of the picture. If the new husband demurred, the first husband would threaten a lawsuit—and since the sale had occurred in his absence, he was likely to receive a sympathetic hearing in court.56 F R AU D U L E N T W I F E S A L E

In a fraudulent wife sale, the woman was usually represented as a widow, and outwardly the transaction resembled widow remarriage in every respect. The fact that hapless men could be duped into thinking they were marrying widows, when in fact they were buying “wives with living husbands,” confirms that widow remarriage was indeed a form of wife sale. Consider the following example.

264

part two

In 1838, charges were filed at the Ba County yamen accusing one Li Wenhan (30) of “kidnapping” and “seizing as his concubine” a married woman named Jiang Shi (22). The accusation came as a shock to Li, who had believed his new concubine to be a widow. As Li explained in his own petition, My wife Wang Shi is already 34 sui but has not given birth, so I obeyed my mother’s order to take a concubine in order to secure our line of descent. We had hired the carpenter Huang Shuangtai to add a room to our house, and he mentioned that in his neighborhood, there lived a widow named Wei Zhang Shi, whose son Wei Shilun had died of illness. Wei Shilun’s widow, Jiang Shi, was young and had no one to support her, and they had no way to provide their daily food. Huang Shuangtai said he was willing to act as matchmaker and arrange for me to marry Jiang Shi as my concubine. Wei Zhang Shi authorized the marriage, and Shuangtai wrote out a marriage contract for me to keep as proof (and which I now submit to the court), stating that Wei Zhang Shi would receive a brideprice of 9000 cash to pay for her son’s coffin. I honestly believed that Wei Shilun was already dead, and since his mother had authorized the marriage, I saw no reason for suspicion. So I agreed, and took Jiang Shi in marriage.

In fact, however, Jiang Shi’s husband, Wei Shilun (37), was still alive, and it was he who had engaged the carpenter Huang Shuangtai to find a buyer for his wife. The litigants eventually settled out of court, and the mediators’ petition asking for the magistrate’s approval explains what had happened: The truth is that Wei Shilun was crippled by a disease of the foot; his household was poor and his mother aged, and it became difficult to provide their daily food. He tried several times to divorce Jiang Shi and remarry her to someone else, in order to survive. But even if there were a matchmaker, the local men were fearful and would not dare to take in marriage a wife who had a living husband (ren wei sheng qi, bu gan cheng qu). So Shilun’s mother Wei Zhang Shi pretended that her son had already died, drew up a request certificate asking Huang Shuangtai to act as matchmaker, and thereby arranged for Jiang Shi to marry Li Wenhan, who lived far away, as his concubine. Wei Zhang Shi drew up a contract in her own name to authorize the marriage and conceal the truth from Li Wenhan.

All would have been well, except that Wei Shilun had neglected to secure the approval of his wife’s uncle Jiang Yuxuan, who was the senior member of her natal family. (Her father was dead.) When Yuxuan found out about the sale, he assumed that his niece had been mistreated, and he immediately filed charges against all involved. Li Wenhan finally learned the truth when the yamen runners came to summon him. The mediated settlement let the second marriage stand, which suited Jiang Shi. After all, she had kept her first husband’s existence a secret because, as she made clear in a petition of her own supporting Li Wenhan, she was content with her new

Chapter 8. Four Variations on a Theme

265

marriage and did not wish to return to Wei Shilun (a lazy wastrel who “worked at no proper occupation” and had “abandoned his mother and me, leaving us with nothing to eat”). Once she made her views clear, her uncle withdrew his objections.57 This case illustrates a number of basic features of fraudulent wife sales. In this instance, the stated motive for fraud was to make it easier to find a buyer, because the local men were too wary to buy a woman who had a living husband. (The other motive for fraud was to get a better price.) The contract prepared for the buyer would be indistinguishable from an ordinary contract for widow remarriage. Occasionally, a young woman might be posed as a never-married bride (who could fetch an even higher price than a widow), but this was far less common, because most of these women had already reached an age when it would be extremely unlikely for them never to have married.58 For obvious reasons, success depended on the wife’s cooperation, but it was also necessary to go outside the local community in order to find a buyer who did not know her and was unlikely to discover her background. It is impossible to estimate with any certainty the relative incidence of open versus fraudulent wife sales, but certainly the risk of fraud was well known, and it caused a certain amount of suspicion and anxiety among men who were offered genuine widows in marriage. Therefore, even though widow remarriage was extremely common and generally not prohibited, such transactions often involved measures to minimize risk and reassure anxious buyers similar to those found in open wife sales, especially if the widow was not a local woman.59 In my sample of wife-selling cases from the archives, fraudulent sales constitute 11 percent of local cases and 20 percent of xingke tiben. But I suspect that fraud is overrepresented in the court records, for two reasons. If a buyer really had been duped, then he faced little risk in going to court, as long as he was willing to give up the woman. More important, the experience of being cheated could easily provoke anger, conflict, and violence—which no doubt accounts for the fact that fraud is almost twice as likely to appear in xingke tiben (most of which are homicide cases) as in local court cases (nearly all of which concern routine “minor matters”). Fraudulent Wife Sale Followed by Extortion Aside from fraud, extortion is what prospective buyers of other men’s wives most feared—and many cases combine both nightmares. A common feature of fraudulent sale is that the first husband would visit the second husband after the new marriage had been consummated in order to confront him with the truth and demand more money—a variation on the usual extortion of supplementary payments after wife sales. In addition to threatening legal action, however, the first husband would offer the incentive of a new handprint contract that would rationalize the transaction as a direct wife sale. These confrontations were acrimonious and

266

part two

occasionally provoked violence. Sometimes the outraged second husband would immediately file charges against those who had deceived him. But most often, it seems, he would swallow his anger and trade a supplementary payment for a new contract. Usually such cases ended up in court because the first husband would persist in trying to extort even more money. A typical example is an 1860 case from Ba County, in which peasant Ren Tai, together with his widowed mother, his wife, Gao Shi, and their little daughter, had migrated to Chongqing from their home in Jijiang County (about sixty kilometers to the south) in search of a livelihood. In the spring of 1860, Ren fell ill, and with no means of support, he and Gao Shi agreed that he should sell her and their daughter in order to secure the whole family’s survival. However, as Ren later testified, “no one was willing to take in marriage a wife with a living husband,” so they decided that she should pose as his widowed younger sister. Ren took the alias “Cao Changchun,” and Gao Shi called herself “Yang Cao Shi”; through a matchmaker, they arranged for her to marry a casual laborer named Huang Wanshun, who had never previously married, for a very modest brideprice of 2000 cash. The couple’s daughter was included in the bargain, to be adopted by Huang. The fraudulent contract reads as follows: Older brother Cao Changchun hereby establishes this contract authorizing the remarriage of my widowed sister. The situation is as follows: My younger sister married Yang Tai as wife, but on the nineteenth of last month he died. For this reason, my sister Yang Cao Shi came to Chongqing and appealed to members of the lineage because she is unable to secure her food or make ends meet. She also has a daughter from Yang, who is named Weigu. Mother and daughter have no choice, therefore Yang Cao Shi has engaged Chen Wanshun to act as matchmaker to negotiate her marriage to Huang Wanshun as wife. Cao Shi’s daughter will follow her to be adopted by Wanshun as his daughter, and after she grows up he will arrange her marriage. Regardless of what her fate may bring, the Yang and Cao families may not interfere. . . . Wanshun has agreed to pay a brideprice of exactly 2000 cash, and now that amount has been paid over and received by me with my own hands. Now that this marriage had taken place, no-one whether present now or not may challenge it, and if anyone should make excuses to cause trouble, then I, Cao Changchun, will take full responsibility. Both families are acting voluntarily without any pressure or coercion. Now, fearing that people’s hearts are not as of old, I hereby establish this contract authorizing remarriage to serve as proof. Xianfeng 10.r3.6, established by Cao Changchun [mark] [names of six witnesses, including scribe]

The small amount that Ren Tai received for his wife and daughter is an index of the family’s desperation, and Ren was far from satisfied. Less than a month later, he confronted Huang Wanshun, revealed his true name and relationship to Gao

Chapter 8. Four Variations on a Theme

267

Shi, and demanded more money—since the names on the contract were false, Ren denied that Huang had any claim to his wife. When their dispute threatened to escalate into violence, Huang’s neighbors persuaded him to make peace by giving Ren Tai another 1200 cash, in exchange for a new contract acknowledging that Ren had voluntarily sold his wife: I, Ren Tai, hereby establish this contract of acknowledgment. In the past, I married the daughter of Mr. Gao, but recently we had difficulty getting by, and so my wife and I decided that we should separate in order to survive. We engaged Chen Wanshun to act as matchmaker and, pretending that my wife was my younger sister, he arranged for her to marry Huang Wanshun as wife. She entered Mr. Huang’s household one month ago. Now my mother and I have come to Mr. Huang’s home to cause trouble. Mr. Huang appealed to his neighbors to mediate but without success, so Wanshun seized me in order to turn me in to the authorities. I, Ren Tai, realized that I had no defense for my actions, so I appealed to the neighbors to prevent me from being turned in to the authorities. Previously, Mr. Huang paid 2000 cash, and now his neighbors have earnestly urged him to pay me an additional 1200 cash, to pay the travel expenses so that my mother and I can return home. I promise that I will never again cause trouble for Mr. Huang. If I do cause trouble, then Mr. Huang shall take this contract to the authorities and have me prosecuted, and I will submit to punishment without complaint. I promise never to break my word or challenge this agreement. Fearing that spoken words alone are not reliable, I hereby submit this acknowledgment agreement for Mr. Huang to keep forever as proof. [names of five witnesses, including scribe] Xianfeng 10.4.4, established by Ren Tai [handprint]

In contrast with the first contract (for alleged widow remarriage), this document is a straightforward example of an acknowledgment contract exchanged for a supplementary payment: it follows the standard format and contains an honest account of the facts, including the real names of the parties. It is unusual in two respects: it confesses the fraud that Ren Tai had perpetrated against Huang, and it bears the seller’s handprint, which gave it the force of a contract for direct wife sale. The new contract stipulates that Ren Tai and his mother should return home to Jijiang County and cause no more trouble, but, instead, they lingered in Chongqing, and after another month had passed, Ren again visited Huang to demand more money. This time, Huang categorically refused—he had no money left—so Ren filed charges at the Ba County yamen. In self-defense, Huang submitted both contracts, and the magistrate ordered Ren Tai slapped for perpetrating a fraudulent wife sale. But the magistrate also took pity on Ren, and, in accordance with Gao Shi’s wishes, he canceled the second marriage and returned both wife and daughter to Ren while allowing him to keep all the money he had received from Huang. The loser in this case was Huang Wanshun, for whom two months of Gao Shi’s company had cost 3200 cash.60

268

part two

In such cases, the willingness of the new husband to exchange a substantial supplementary payment for a new, truthful contract is strong testimony to the customary power of such documents. This power had nothing to do with their legitimacy in any official setting; after all, an honest contract for a direct wife sale had no more standing in court than a fraudulent contract disguising a wife sale as widow remarriage. Rather, it reflected the force of custom and community norms: to be valid in the eyes of the community, such a marriage had to be documented with a genuine contract bearing the handprint of the first husband to prove his voluntary agreement. We find a similar scenario in a case from Nanbu County, in which a yamen runner named He Xi was duped into paying a brideprice of 10,000 cash to take a “widow” named Zeng Shi as his second wife (his first having died). This transaction took place just after the New Year in the early spring of 1897. Five months later, a man named Jing Guangfu suddenly appeared at He’s home, accompanied by his father and several other men, introduced himself as Zeng Shi’s husband, and demanded money. Realizing that he had been a victim of fraud, He Xi enlisted two rural agents to negotiate on his behalf. He sought to return Zeng Shi for a refund, but the Jings did not want her back; instead they urged He to pay a new brideprice in exchange for a valid contract. He Xi finally agreed to pay 16,000 cash for the following document: I, Jing Guifang, with my son Guangfu (whose childhood name was Maogou), hereby establish this contract for divorcing, selling, and authorizing marriage (xiu mai zhuhun wenyue). The situation is as follows: In his youth, Guangfu took Jing Zeng Shi as wife, and they were married for several years. Contrary to expectation, this wife had not been disciplined properly as a child and does not behave the way a wife properly should. Husband and wife often quarreled, and we did not get along well for even a single day. She falsely accused her father-in-law of impropriety and secretly ran away from home. She was lured away and sold by Wang Fashen, and then resold by Ye Heshao under false pretenses to Mr. He, who did not know that he was being deceived. We did not learn the truth until the sixth month of this year, when, after searching high and low, we finally discovered the circumstances of Zeng Shi’s abduction and sale by Wang Fashen in the market town of Zhengjiachang. We appealed to the rural agent, who asked Mr. He whether he knew that Zeng Shi had been abducted and sold. Mr. He stated truthfully that he had not known, and that since her original husband had tracked her down, he was freely willing to take a loss on the brideprice he had paid for her and to hand He Zeng Shi over to the rural agent to be returned to her husband Mr. Jing. But we have thought it over and decided that since this woman has already lost her chastity, we no longer want her to be Guangfu’s wife. We are willing to divorce Jing Zeng Shi and sell her in marriage (xiu chu jia mai), never to return. We will not feel resentment or regret. Therefore, we personally wrote a request certificate, in order to engage Yang Fangjin and the rural agent to act as matchmakers, volunteering to divorce and sell (xiu mai) Jing Zeng Shi back to He

Chapter 8. Four Variations on a Theme

269

Tianxi to be his wife. We agreed on a brideprice of 16,000 cash, which we have received in full with our own hands. . . . If anyone from either her natal family or inlaws ever challenges this agreement, we, father and son, will take full responsibility. Zeng Shi’s husband Jing Guangfu is freely willing to divorce her, and to prove this he personally places his hand- and footprints on this contract, in order to divorce his wife and sell her to Mr. He. In the future, we promise not to cause any trouble. If either of us makes even the slightest challenge to this agreement, then Mr. He and the two rural agents may take this contract to court, and we shall admit our crime of “buying and selling divorce”—the brideprice will be confiscated, and we will be punished according to law. This we record as proof. Rural agents: [two men surnamed Jing] Present as witnesses: [six men, including three brothers of buyer] GX 23.7.4, marriage contract established by Jing Guifang and his son Guangfu [mark] [handprint and footprint]

This unusually elaborate contract confirms that He Xi had been a victim of fraud, but at the same time it disingenuously places the entire blame on Zeng Shi’s lack of chastity and on the traffickers who allegedly lured her into running away. In effect, the Jings denied any responsibility for the fraud (they did not admit receiving the initial brideprice of 10,000 cash) even as they confirmed that they no longer wanted Zeng Shi and were willing to “sell her divorce” to He Xi. In making the Jings promise to cause no further trouble, this contract explicitly invokes the statute against “buying or selling a divorce.” Evidently, He Xi knew the law (not surprising, since he was a yamen runner), and he was willing to use it to intimidate others, even as he violated it himself. But the Jings were not intimidated—later they demanded even more money from He Xi, who finally filed charges against them.61 Manipulation of Text to Take Advantage of Illiteracy Self-serving distortion of the facts seems to have been a common feature of contracts in fraudulent cases, even the supposedly truthful second contracts exchanged for supplementary payments. We find a striking example of the manipulation of text to take advantage of illiteracy in a 1754 case involving a landless migrant laborer named Shi Youde (43). Shi was originally from Cao County, Shandong, and his wife, Wang Shi (27), was from Hua County, Henan; she had been sold by her parents as a young girl to be a maidservant in a prosperous household, and in 1744 (when she was 17 sui) Shi had bought her out of servitude for a body price of 25 taels to be his wife. The couple had one little daughter. In the early spring of 1754, just after the New Year, they found themselves in Xinyang Department, Henan (some 300 kilometers south of Cao County), out of food and unable to find work, so they agreed “to separate in order to survive.”

270

part two

An acquaintance, peasant Liu Rui (37), agreed to help: he knew of a potential buyer who lived in Anlu County, Hubei (about a hundred kilometers southwest of Xinyang Department), in the same village as Liu’s younger sister and her husband. Shi Youde was too ashamed to participate directly in such negotiations, so he asked Liu to take Wang Shi and their daughter to Anlu County, broker a fraudulent sale, and then notify him when he had returned with the money. Liu hired a mule to carry Wang Shi and her daughter, and after a week they arrived at the home of Liu’s sister. Her husband, Jin Dalun (32), knew that a tenant farmer in the same village named Song Zhenhai was looking for a wife, but he pointed out that, since he and Song were not close, it would be necessary to recruit other villagers whom Song would trust. Jin explained the situation to Song’s landlord Cheng Liwan (36), a local rural agent Dong Shilu (50), and another villager Cheng Ruizhang (56), and invited them to act as matchmakers in exchange for a share of the brideprice. After securing Wang Shi’s assurance that she and her husband were both willing for her to be sold, these men agreed to help persuade Song Zhenhai and perpetrate the fraud. These men visited Song and told him that Wang Shi was a widow whose remarriage had been authorized by her “brother-in-law” Shi Youde. Song complained that Wang Shi’s “background is not clear” (laili buming): the men were promising him a marriage contract in Shi Youde’s name, but Shi himself was not present. To allay Song’s concerns, his landlord offered to guarantee both his money and the woman’s honesty, so that no matter what, he would suffer no loss. Since three village notables approved the marriage, and as far as Song knew they had never cheated anyone before, he agreed to pay a brideprice of 30,500 cash (calculated as equivalent to 37 taels). Song Zhenhai needed a few days to raise the necessary funds (among other things, he sold an ox for 6500 cash). Meanwhile, Cheng Liwan drafted the following contract: I, Shi Youde, hereby establish this contract for marrying off and granting custody of my widowed sister-in-law. Now, because my younger brother Shi Er has died of illness, I am willing for my younger brother’s widow Wang Shi to be taken in marriage as wife by Song Zhenhai, as negotiated by matchmakers Dong Shilu, Jin Dalun, and Cheng Ruizhang. On this day, Mr. Song has paid over 30,500 cash, which I, her older brother-in-law, have received in full with my own hands. This sum includes any ancillary expenses such as wine and snacks, presents, scribe’s fee, etc. Now that Wang Shi has remarried, no one may challenge this contract. Fearing that there will be no proof, I hereby establish this marriage contract for Song Zhenhai to keep as proof. Established on QL 19.2.30 Written, according to Shi Youde’s spoken words, by Cheng Liwan

But when rural agent Dong Shilu read the contract, he became angry. As he later testified, because Wang Shi was “a wife with a living husband” and Song Zhenhai

Chapter 8. Four Variations on a Theme

271

was being deceived, they all understood that the transaction was illegal. “On the contract they had written my name first, making me the lead matchmaker—they were trying to shift all the responsibility for the wife sale onto me, so that I would be blamed!” Dong seized the contract and insisted that Cheng Liwan write a new one without using his real name. In this second draft, Cheng Liwan removed his own name, along with those of Dong and Cheng Ruizhang; in their place, he wrote two fake names, “Jia Qian” and “Shi Yu.” Cheng knew that Song Zhenhai was illiterate and assumed that he would not notice the difference. The matchmakers gave Song this contract for a down payment of 2000 cash, and Wang Shi joined him in marriage; Song promised to pay the balance in a week. Before then, however, Song’s cousin Liao Wenqi dropped by to visit. Liao could read, so Song asked him to read the contract aloud. Song immediately noticed that three of the matchmakers’ names were missing, and instead it listed two nonexistent people. His suspicions aroused, Song interrogated Wang Shi, but she did not reveal the fraud: “I was afraid that if I told the truth, I’d have to return home and would have nothing to eat.” When the day came for Song to pay the balance of brideprice, all the men gathered at his home. First, he served them wine, but when they asked for the money, he informed them that he had discovered that the contract was “fraudulent” (pianren de), and he refused to pay unless they gave him a new, valid contract, plus a separate “written guarantee” (dancheng zi). An argument turned into a brawl (when the matchmakers tried to take Wang Shi back by force), and one of Song’s neighbors ended up being killed.62 In cases like these, literacy (or its lack) was a key factor in perpetrating fraud— and everyone seems to have realized this, given the central role of written contracts in any major transaction. There is no consensus on literacy rates in Qing China, but few peasants could read or write, at least not to any serious degree; nevertheless, literacy was available even in this humble village.63 Landlord Cheng Liwan was able to write contracts, and rural agent Dong Shilu could read them.64 The success of their scheme depended on Song’s inability to read, but even an illiterate peasant like Song was shrewd enough to understand his vulnerability to fraud, and he had recourse to a cousin who could read a contract aloud for him. “Releasing a Falcon” In a notorious variant of fraudulent wife sale, instead of the first husband trying to extort more money, the wife would simply run away and rejoin him. A 1749 case from Bo Department, Anhui, records a conversation in which one Wu Gan tried to persuade his friend Yang Er to attempt this kind of fraud. As Yang later testified, Wu urged him to pose his wife as a widow and “to pretend to sell her, in order to cheat someone out of his money.” When Yang hesitated, Wu assured him, “Don’t worry! All you have to do is persuade your wife to play along and stay in character

272

part two

(zuo ge ming se). Once we’ve cheated this guy out of the brideprice and have his money in hand, I guarantee you I’ll get her back for you.”65 The American missionary Arthur Smith described the same scheme: The man who wishes to sell his wife represents her as his sister, and declares that they are forced by hunger to part company. He reluctantly makes up his mind to sell her to some one who is in need of a wife, and who can get one more cheaply by this process than by any other. To this arrangement the woman tearfully assents, the money is paid to her “brother,” and he departs, to be seen no more. After a few days or a few weeks in her new home, the newly married “sister” contrives to steal out in the evening with all of her own clothes and as many more as she can collect, and rejoins her “brother,” setting out with him for “fresh woods and pastures new.” . . . In some districts this practice of “falconing” is exceedingly common, for the supply of gullible persons who hope to buy a wife at a cheaper rate than usual never fails.66

The term that Smith translated as “falconing” or “flying the falcon”—fang ying— was widely used to denote this particular form of marriage fraud.67 Another term is “releasing a pigeon” (fang gezi, fang fei ge), because both a hunting falcon and a homing pigeon will return, just as the wife would run away back to her first husband. Both terms are used today for this kind of swindle, which has staged a comeback, along with the sale of daughters as the default form of marriage in at least some parts of rural China.68 We find a vivid example in an 1819 case, in which a peasant couple from Bo Department, Anhui, attempted serial falconing. As Zhang Shi (30) later recalled, Because we were very poor and could not make ends meet, my husband proposed to sell me, and afterward he would abduct me back and look for someone else to sell me to, so that we could live off the profits from cheating people. I agreed. So together we went to Zhecheng County [across the provincial boundary in Henan], where a man named Li Fa bought me to be his concubine for a body price of 24,000 cash. [Note that this was a fraudulent sale, with Zhang Shi posing as a widow.] After that, though, Li Fa kept tight control over me, so there was no way for me to run away. Finally, my husband revealed to Li Fa that he had bought a wife with a living husband and threatened to file charges against him. Li Fa was afraid of trouble, so he returned me to my husband without demanding a refund.

Zhang Shi’s husband, Li Shun (32), hoped to repeat the trick. In the meantime, the couple moved in with a local peasant, Ma Congmeng (30), who helped support them in exchange for sex with Zhang Shi. Li Shun complained to Ma that it was impossible to find another gullible buyer because word had gotten around, and “now everyone here knows that we are scheming to cheat people”; he informed Ma that he planned to take Zhang Shi to a different county. But Ma had become attached to Zhang Shi, and he offered to buy her from Li Shun to be his own wife. Li refused, however, preferring the steady income that he expected from falconing

Chapter 8. Four Variations on a Theme

273

to a one-time payment in an honest deal. In the end, Ma murdered Li in order to take Zhang Shi for himself.69 In general, cases involving falconing came to court because something went wrong with the scheme—usually when the woman was caught trying to run away. In a 1755 case, peasant Ding Nanreng and his family became refugees after the harvest failed at home in Anhui, and before long they found themselves 300 kilometers away in Hubei. In desperation, Ding and his wife, Wang Shi (23), decided to pose her as a widow and sell her along with their little girl. They found a buyer willing to pay a high price of 58.5 taels for such a young widow and her daughter, but Wang Shi was unhappy at the prospect of marrying a stranger so far from home. To persuade her to cooperate, her husband promised that, after the sale, he would help her run away so that they might return home together: “If you can just hold out for a few days, I’ll come get you.” Wang Shi went through with the marriage. Some time later, Ding Nanreng visited her new home (posing as her brother-in-law) and secretly arranged that she and their daughter would sneak out that night and rendezvous with him and two friends, so that they could run away together. But her new husband awoke and gave chase, with several neighbors coming along to help. In the resulting brawl, Ding killed one of the villagers.70 As with the basic form of fraudulent wife sale, falconing could succeed only if the woman herself was willing to cooperate—and, sometimes, women changed their minds. In an 1813 case from Jimo County, Shandong, the sold wife became so anxious that her first husband would abduct her, as he had promised to do (“I was afraid he would just sell me again to someone else”), that she told her new husband the truth and swore she would not run away. When her first husband showed up at their home, she screamed and resisted him.71 In a similar situation, in a 1794 case from Dingyuan County, Anhui, the woman grabbed her first husband’s queue and began screaming “Kidnappers! Kidnappers!”72 In a 1745 case from Gao’an County, Jiangxi, when the woman caught sight of her first husband and his father, she immediately confessed the fraud to her new husband, told him who the two men were, and warned that they would try to abduct her.73 When women like these changed their minds, the decisive factor was the improvement in standard of living that had resulted from their new marriages. For example, in a 1751 case from Jie Independent Department, Shanxi, Yan Hongting (40) and his wife, Zhang Shi (28), had agreed on a falconing scheme in advance of her fraudulent sale: “Don’t worry,” he told her, “after I get the money, I’ll figure out a way to help you run away and rejoin me.” A widower named Cui Erheng (33) agreed to buy this young “widow” as his second wife for a brideprice of 41 taels. After entering Cui’s household, however, Zhang Shi began to have second thoughts: her first husband Yan was a landless peasant who barely made a living as a hired laborer, and during the twelve years of their marriage they had often gone hungry;

274

part two

in contrast, Cui—a decent man who treated her well—earned a good living as a small-scale merchant with several employees, and she found herself far more comfortable in his household. Thus, when Yan Hongting visited Cui Erheng’s home on a pretext, intending to take Zhang Shi away, she refused to cooperate. Twice more Yan visited to persuade her to leave, but she rejected him, and he became increasingly bitter at her change of heart. On the final occasion, he forced his way into the courtyard of Cui’s home, carrying a knife; when Zhang Shi caught sight of him, she cursed him: “You have no shame at all! What do you think you’re doing, coming here again? Get out!” Upon hearing this, Yan stabbed her to death.74 T H E I M P L IC AT IO N S O F F R AU D

Fraud ties this chapter together, because, from the buyer’s point of view, as long as the woman concealed her marital history, a fraudulent wife sale was indistinguishable from genuine widow remarriage. Fraud sometimes played a role in compensated divorce, when the natal family pretended their daughter was a widow, and the fact that both direct and indirect wife sales could be disguised as widow remarriage underscores the fundamental similarity between all three transactions. The purpose of fraud was to boost profit, because it was easier to sell a widow than “a wife with a living husband,” and widows typically fetched higher prices. The fact that all of these transactions had fraudulent versions confirms that they were indeed sales. A wife’s remarriage in her husband’s absence closely resembled widow remarriage: the structure of the two transactions was the same (with the husband out of the picture, his family acted in his stead), and some families honestly believed the absent husband to be dead when they arranged his wife’s new marriage. When the absent husband returned to demand more money from his wife’s new husband, his action mirrored the demand for supplementary payments that routinely followed wife sales. But the resemblance to fraudulent wife sale is particularly striking, because in both of these scenarios, when the husband revealed his presence and demanded more money, he would offer a new contract to rationalize his wife’s remarriage as an open, direct sale. This chapter has expanded our inquiry beyond open, direct wife sale to encompass the broader diversity of transactions whereby a wife was transferred from one husband to another in exchange for financial compensation. If we follow the money, it is clear that compensated divorce, widow remarriage, and wives’ remarriage in their husbands’ absence all constituted variations of wife sale. One implication of this finding is that wife sale formed an integral and logical part of the broader marriage system; and some variations (notably widow remarriage) were extremely common.

part three

Polyandry and Wife-Selling in Qing Law

This page intentionally left blank

9

Formal Law and Central Court Interpretation from Ming through High Qing A man who sells his wife in marriage because of poverty or illness should not be punished; nor should any second husband be punished if he has used money to buy another man’s wife in marriage, but whose case involves no illicit sex. —the ming dynasty censorate, 1569 1

Chen Yide sold his wife only because he was pressed by both illness and poverty, and the couple agreed to seek separate routes to survival. Be that as it may, to sell off one’s wife for cash constitutes the crime of “selling a divorce” (de qian jia mai ji shu mai xiu). —the magistrate of xinfeng county, jiangxi, 1774 2

The first two parts of this book focused on social practice: how polyandry, wife sale, and the intermediate spectrum of strategies related to survival logic as well as to the broader system of marriage and household formation in pre-revolutionary China. I have sought to document what people did, how they did it, and why, and to illuminate the subject positions of the different parties. A basic goal has been to show that these strategies were central to the Chinese system of marriage and household formation. In Part Three, our focus shifts to the treatment of these practices in late imperial law, with the main topic being wife sales. The reason to focus on wife sales is that they expose especially sharp contradictions between judicial imperatives and widespread social customs. These contradictions fostered disagreement among senior jurists about how to enforce the law and cast in high relief the tensions between ideology and practice in the Qing legal system. The present chapter explains formal law and how it changed from Ming through High Qing by looking at both the code and its application in “major cases” 277

278

part three

(zhongda anjian) reported up the chain of command for review (see Appendix E). This chapter makes two basic points. First, under the Ming dynasty, the great majority of wife sales and much marital prostitution were legally tolerated. Under the Qing, however, the scope of prohibition expanded dramatically, in concert with a new fundamentalism toward sexual behavior and gender roles as well as a heightened concern with imposing conformism on ordinary people. As a result, all prostitution and polyandry were redefined and criminalized as “illicit sex” (jian), as were wife sales motivated by poverty that involved no prior adultery. Among other effects, this criminalization of custom generated the vast archive of legal cases that made the present study possible. Second, despite this expanded prohibition, and despite the total ban on prostitution, Qing law did continue to permit some marital separations, including even some wife sales, under highly specific circumstances. The legal hairsplitting that distinguished criminal wife sale from permitted divorce depended on minute distinctions of ritual propriety. Such distinctions probably made little sense from the standpoint of the people who actually negotiated these transactions, since nearly all marital separations partook of the same basic economic logic.3

M I N G DY NA S T Y L AW

“Abetting or Tolerating” Illicit Sex and “Buying or Selling a Divorce” The Ming code’s chapter on “illicit sexual relations” (fan jian) contains the following statute, under the title “abetting or tolerating a wife or concubine committing illicit sex offenses” (zongrong qiqie fan jian). Given the central importance of this statute for Part Three of this book, we should pause to consider it in detail: If someone abets or tolerates his wife or concubine to engage in illicit sexual intercourse with another man, then the husband, the man who engaged in illicit sex, and the woman shall each receive 90 blows of the heavy bamboo. If someone forces (yi le) his wife, concubine, or adopted daughter to engage in illicit sexual intercourse with another man, then the husband or adoptive father shall receive 100 blows of the heavy bamboo and the man who engaged in illicit sex shall receive 80 blows of the heavy bamboo; the woman shall not be punished, and shall be divorced and returned to her natal lineage (liyi gui zong). If someone either abets or tolerates, or forces, his own birth daughter, or the wife or concubine of his son or grandson, to engage in illicit sexual intercourse with another man, then their penalties shall be the same as above. If someone uses money to buy or sell a divorce (yong cai mai xiu mai xiu), so that one man takes in marriage the wife of another with consent (he qu ren qi), then the first husband, the wife, and the man who bought the divorce shall each receive 100 blows of the heavy bamboo; the wife shall be divorced and returned to her natal lineage, and the brideprice shall be confiscated. If the man who buys the divorce and

Chapter 9. Formal Law and Central Court Interpretation

279

the wife use a stratagem to force her husband to divorce her, when he otherwise has no intent to sell her divorce, then the husband shall not be punished; but the man who buys the divorce and the wife shall each receive 60 blows of the heavy bamboo and one year of penal servitude. The wife’s sentence of penal servitude shall be redeemed and she shall be handed over to her first husband, who shall be permitted to sell her in marriage (cong qi jia mai). If the woman in question is a concubine rather than a wife, then the penalties shall be lowered by one degree. In all cases, matchmakers shall receive penalties one degree lower than those of the principal offenders.4

To summarize: the first clause prohibits men from permitting (or forcing) their wives, concubines, or daughters to engage in illicit sexual relations with other men (the abbreviated term for this offense was “zong jian,” or “abetting or tolerating illicit sex”), whereas the second clause prohibits men from selling wives or concubines in marriage to other men (the legal term for criminal wife sale being mai xiu, “to sell a divorce”). Under the Ming, the application of this statute was narrow, but the Qing dynasty reinterpreted it and dramatically expanded its scope. In 1646, the statute was incorporated into the first edition of the Qing code with only a single amendment (of which more below), and, throughout the Qing dynasty, central courts would apply it to major cases that involved wife-selling as well as polyandry, marital prostitution, and any other situation in which a husband either encouraged or tolerated his wife’s extramarital sexual relations. Nearly all of the xingke tiben consulted for this study cite this statute in their judgments: the wife-selling cases cite the first sentence of the statute’s second clause, and the polyandry, prostitution, and other cases cite the first sentence of its first clause. What was the logic of combining these two crimes in a single statute? When the statute was first drafted in the early Ming, the scenario of wife sale prohibited by the second clause was intended to be highly specific: a transaction in which a husband who had already “abetted or tolerated” his wife’s adultery then proceeded to sell her in marriage to her partner in crime. In other words, the statute’s second clause criminalized only wife sales that were a direct outgrowth of the crime prohibited by its first clause. The early Qing jurist Shen Zhiqi explains that “the reason the prohibition of ‘buying or selling a divorce’ immediately follows the prohibition of ‘abetting or tolerating illicit sex’ is that ‘buying a divorce’ was originally intended to cover only instances of first having illicit sex with another man’s wife and then taking her in marriage (ben wei xian jian hou qu zhe er she).”5 The main innovation of the Qing judiciary with regard to wife sales would be to reinterpret this statute in a manner that decoupled the two crimes and extended the second clause to cover wife sales involving no prior adultery and motivated solely by poverty. The result was a dramatic expansion of criminalization and increase in the number of prosecutions.

280

part three

Both crimes were understood to “break” (jue) the “bond of moral duty” (yi) between husband and wife. A classic principle of imperial law was that familial relationships came in two forms: those “joined by Heaven” (tian he) and those “joined by man” (ren he). The former comprised direct blood ties, as between parents and birth children, and in legal terms such ties could not be “broken” regardless of conduct. Therefore, although a father who pimped his own daughter was liable for punishment, she remained his daughter and he would not lose custody. But the relationships “joined by man,” such as marriage and adoption, would be nullified by conduct that violated the moral obligations inherent to them—hence the statute’s mandate that a wife or adopted daughter be returned to her natal lineage. From a broader patriarchal perspective, a husband who abetted his wife’s licentiousness betrayed the interests of husbands in general and therefore disqualified himself from custody over her.6 Limited Scope of the Ming Statute Such standards did not apply to everyone, however. Since this statute dates to the early Ming, its intended application was originally far narrower than it would become several centuries later, during the High Qing. The key point to remember is that this statute targeted offenses of “illicit sex” (jian), a category of crime defined in terms of the normative sexual morality and gender roles binding on free commoners (liangmin). For that reason, the conduct described in this statute was not prohibited to people of debased legal status, to whom the moral bond of duty broken by such conduct did not apply. The legal status category of “debased” (or “mean”) persons (jianmin) included various forms of unfree labor but also nominally hereditary groups associated with prostitution and other stigmatized entertainment, such as the infamous “music households” concentrated in Shanxi. Individuals from debased-status households were neither expected nor entitled to uphold the high standard of female chastity that was a definitive attribute of free commoner status. In fact, their unfree status traditionally implied an obligation to provide sexual service: female slaves and bondservants were sexually available to their masters regardless of marital status, whereas women of “music households” were expected to support their families by working as prostitutes, and it was normal for their fathers and husbands to act as their pimps. Therefore, it was taken for granted that prohibitions of “illicit sex” did not apply to such people. On the contrary, prostitution within the context of debased status was accepted, regulated, and even protected by the authorities.7 This was true, too, with the crime of “buying or selling a divorce,” which Ming law defined as a husband selling his wife to her partner in adultery after having tolerated their relationship. Where the Ming code addresses human trafficking in connection with debased status, it focuses on prohibiting the transfer (by sale, marriage, or adoption) of individuals of free commoner status to debased-status

Chapter 9. Formal Law and Central Court Interpretation

281

households. It also prohibits the debasing of free commoner individuals by selling them into slavery.8 The purpose of such measures was to police status boundaries rather than to prohibit trafficking per se. It was taken for granted, for example, that women of “music households” would marry or could be sold into other debasedstatus households. But even with regard to free commoners, the Ming prohibition of “buying or selling a divorce” was never intended to ban wife sales in general but only those that were a direct result of the prior offense of “abetting or tolerating” a wife’s illicit sexual relations. The second crime was an escalation of the first: hence, “abetting or tolerating” alone would bring ninety blows of the heavy bamboo, whereas “selling a divorce” would bring one hundred blows, and in either instance the woman would be returned to her natal household. In the late Ming, however, some jurists began to advocate a new and much broader interpretation that reflected a fundamentalist, uncompromising attitude toward female chastity. They argued that the prohibition of “buying or selling a divorce” should be reinterpreted to define wife-selling per se as a form of “illicit sexual relations,” regardless of motive, because the sexual union resulting from such a sale was illegitimate and therefore itself constituted adultery. In their view, it did not matter whether adultery preceded a sale. But the original, narrow interpretation prevailed during the Ming. In 1568, the Court of Judicial Review dismissed the proposed broad interpretation by pointing out how it contradicted other well-known provisions of the Ming code: If a husband and wife do not get along, then according to law they may divorce; by the same token, if a wife engages in illicit sex, then according to law, her husband is allowed to sell her off in marriage. In either case, if a second husband goes through a matchmaker and pays money to take the woman as his wife, their union does not constitute “illicit sex,” and is not prohibited.

In short, some forms of marital separation and remarriage, including some wife sales, were permitted and should not be confused with the offense of “buying or selling a divorce.” The following year, the Censorate reaffirmed the narrow interpretation, declaring that other scenarios of wife sale should be punished far less severely, if at all: Whoever sells his wife in marriage out of greed for money (tu cai jia mai) should be punished according to the statute against “doing inappropriate things,” and, depending on the circumstances, the amount paid may be confiscated. But a man who sells his wife in marriage because of poverty or illness should not be punished; nor should any second husband be punished if he has used money to buy another man’s wife in marriage, but whose case involves no illicit sex.

The Longqing emperor agreed: “The prohibition of ‘buying or selling a divorce’ is part of a statute against ‘illicit sexual relations’ (ben shu jian tiao). Therefore, it may

282

part three

not be applied except in cases that actually do involve [prior] illicit sexual relations.”9 The Statute against Conditional and Fraudulent Wife Sale Three other scenarios of wife sale were covered in the Ming code by the following statute: Any husband who accepts money to conditionally sell (dian) or hire out (gu) his own wife or concubine to serve as the wife or concubine of another man shall receive 80 blows of the heavy bamboo. Any father who conditionally sells or hires out his own daughter in this manner shall receive 60 blows of the heavy bamboo; [in either case] the woman shall not be punished. Anyone who falsely represents his wife or concubine as his sister in order to marry her off to another man (jiang qiqie wang zuo zimei jia ren) shall receive 100 blows of the heavy bamboo; the wife or concubine shall receive 80 blows of the heavy bamboo. Anyone who knowingly takes in marriage by conditional sale a woman who is already married shall in each instance receive the same penalty, and the woman shall be divorced. The brideprice shall be confiscated. If the buyer does not know the woman is already married, then he shall not be punished, and the brideprice shall be returned to him.10

Evidently, none of these scenarios were covered by the narrow interpretation of “buying or selling a divorce,” because none were envisioned as involving prior adultery; significantly, this statute is located in the code’s chapter on “household and marriage,” and it makes no mention of “illicit sexual relations” (jian). The statute does not specify that the woman be “returned to her natal lineage” (in contrast with the statute against “buying or selling a divorce”)—only that “the woman shall be divorced.” According to a commentary of the Wanli period (1573–1619), however, this phrase meant that “regardless of whether the woman was conditionally sold, hired out, or married off [fraudulently], she should not return to her first husband, nor should she stay with her second husband; instead, a daughter should be given to her parents and a wife or concubine should be returned to her natal lineage.”11 The first two offenses covered by this statute (conditional sale/dian and “hiring out”/gu) both involved the temporary transfer of a wife (or concubine) to another man to serve him in that same capacity. The Ming jurist Zhang Kai explains the difference between these two offenses as follows: “Conditional sale” in this context is similar to the conditional sale of land or a building: you exchange it for a price, and later you redeem it at the original price. “To hire” is just like hiring a boat or a cart: you simply pay by the day, and there is no need for it to be redeemed by its owner. . . . When a wife is taken by “conditional sale,” the relationship between buyer and woman lasts for years, and they become

Chapter 9. Formal Law and Central Court Interpretation

283

husband and wife, whereas when she is “hired,” she is employed for wages for a short time only.12

Although these offenses are not labeled “illicit sexual relations,” it seems likely that they were considered objectionable precisely because of their similarity to prostitution. In particular, it is not at all clear how “hiring out” one’s wife to serve as another man’s wife for a temporary period would be different from “abetting or tolerating” one’s wife to engage in “illicit sexual intercourse” with another man— and, in fact, I have not been able to find any example in Ming or Qing casebooks or Qing legal archives of the provision against “hiring out” actually being applied. The penalties are similar for the two offenses, except that a woman conditionally sold or hired out was not to be punished. In other words, in the offense of “abetting or tolerating” illicit sexual intercourse, the woman is envisioned as being equally culpable and motivated by licentiousness, whereas in the offense of conditionally selling or hiring out, the onus is entirely on the husband. The second clause of the statute, on fraudulent marriage, does not use the word “sell,” but the only likely motive for such a fraud would be to obtain the woman’s brideprice. Here, the woman is envisioned as complicit (and therefore to be punished), presumably because her cooperation was vital to the fraud’s success. As discussed in the previous chapter, Qing case records show that by far the most common form of marriage fraud was for a husband to pass off his wife as his widowed sister-in-law. The single example of this law being applied in the Ming dynasty that I have found (in a casebook from the early seventeenth century) involves a father and son passing off the son’s wife as a widow in order to sell her in marriage. In that case, the magistrate ordered the woman returned to her first husband and waived the mandated beating, out of pity for their wretched poverty, which had been their sole motive for selling her.13 In the Qing, fraudulent wife sales were occasionally punished by citing this statute, although central cases usually cite the statute against “buying or selling a divorce.” To summarize, Ming policy on wife-selling was highly specific and limited in scope. It involved a sliding scale of consequences, depending on the particulars of an offense. The most severe penalties applied to “buying or selling a divorce,” which was defined as the logical extreme of “abetting or tolerating” a wife to engage in “illicit sexual relations” with another man. It was the husband’s abdication of responsibility in facilitating his wife’s licentiousness that made this offense so bad; all three parties to the transaction were to receive the same penalty of one hundred blows of the heavy bamboo, and the woman was to be divorced from both men and returned to her natal lineage; matchmakers were to be beaten and all money confiscated. Slightly lower on the scale came fraudulent wife sale, followed by conditional sale and the “hiring out” of a wife or concubine; the details varied, but these three lesser offenses were not envisioned as involving prior

284

part three

adultery. Finally, an open wife sale motivated only by greed was to be punished rather mildly, according to the catch-all statute against “doing inappropriate things” (bu ying wei er wei), and whether or not to confiscate the brideprice was left to the magistrate’s discretion. But a wife sale motivated by poverty or illness was not to be punished at all; in such cases, a man who had bought another man’s wife in marriage was not to be punished either. In addition, a husband could actually get permission to sell his wife in marriage if she had been found guilty of offenses against chastity (of which more below). As we have seen, Qing archives show that the vast majority of wife sales were motivated by poverty. Assuming that was already true in the Ming, then Ming law did not prohibit the vast majority of sales—which no doubt explains why almost no examples of prosecution can be found from that era. No legal archives survive from the Ming, to my knowledge, but I have searched many Ming casebooks, and I find next to nothing relevant to the practices documented by this study. It was not until the High Qing that wife sale in particular became a significant problem for the judiciary. It is likely that such practices did become more common over the course of the Qing, as a result of worsening involution and poverty in the countryside. But their voluminous documentation in both casebooks and court records also reflects the dramatic expansion in the criminalization of nonnormative gender roles and sexual behavior that characterized High Qing fundamentalism. Q I N G I N N OVAT IO N S

To understand the Qing innovations addressed in this chapter, it is necessary first to understand the processes by which formal law changed during that dynasty. First and most obvious was the amendment of an old statute (lü) through promulgation of new substatutes (li), which are listed after the original statute in published editions of the Qing code. A new substatute would be proposed by provincial officials or by the Board of Punishment (usually to address scenarios of crime not covered by existing law) and then, if approved, would be promulgated by imperial edict. Old statutes remained in the code, superficially unchanged, to give an impression of timeless continuity, but the most recent substatutes constituted the living law of the dynasty, to be cited and applied in central cases. A second, subtler process for changing the law was reinterpretation of existing statutes (and of old legal terms and concepts) to apply them in new ways, without changing their text. If amendment of the code through promulgation of new substatutes is roughly analogous to the system of constitutional amendment in the United States, then reinterpretation of existing statutes can be compared to the Supreme Court’s occasional reinterpretation of the Constitution’s text to take new circumstances into account.

Chapter 9. Formal Law and Central Court Interpretation

285

My first book traced both processes of change in the shift from “status performance” to “gender performance” in High Qing law. A basic argument was that, in order to understand how Qing law changed over time and what laws were actually current, one must examine how central courts applied the code in actual legal cases. The best sources for understanding this process are central court cases and officially published casebooks such as The Conspectus of Penal Cases.14 If we look at how formal law treated polyandry and wife-selling in the Ming and Qing dynasties, we find that both processes were employed to change the law. But the most important changes occurred in the realm of reinterpretation: in particular, the criminalization of all prostitution (including everything I have termed “polyandry”) and of poverty-driven wife sales, which were the two key innovations of the Qing judiciary, took effect mainly through the reinterpretation of old Ming statutes rather than explicit amendment with substatutes. Expanding the Scope of Criminalization Under the Qing dynasty, jurists dramatically expanded the scope of the Ming statute against “abetting or tolerating” illicit sex and “buying or selling a divorce.” With regard to the statute’s first clause, the Yongzheng reforms eliminated the debased legal status categories heretofore associated with prostitution, in effect promoting those people to free commoner status. As I have shown elsewhere, this leveling maneuver had the effect of criminalizing all prostitution, by eliminating status-based exemptions and extending the free commoner standard of sexual morality and criminal liability to cover the previously exempt groups. As a result, the statute’s clause against “abetting or tolerating a wife or concubine to engage in illicit sexual intercourse with another man” now covered any prostitution that took place in a marital context—which, as we have seen, seems to have included most actual prostitution.15 What this meant in practical terms can be seen in the fate of the “music households,” who had been singled out by the first reform edict of 1723: the elimination of their debased status amounted to the criminalization of their livelihood. For example, in a 1743 case from Fenxi County, Shanxi, Yang Gui and his wife, Wei Sulian (“Pure Lotus”), were “a music household that had already reformed and changed from debased to free commoner status” (gai jian wei liang zhi yuehu) in response to the edict. But the loss of income from prostitution reduced them to penury, so Yang Gui “once again had his wife act as a prostitute.” The couple engaged in retail prostitution until one of their regular patrons arranged to contract the exclusive use of Wei Sulian’s sexual services and the couple moved in with him, initiating a polyandrous relationship. Their situation came to official attention when a former customer got in a fight with Yang Gui and their patron, and they killed him. Wei Sulian was sentenced to one hundred blows of the heavy bamboo for “violating an edict” by “illicitly acting as a prostitute” (si wei tu chang),

286

part three

and the county magistrate had his own salary docked for “failing to detect local prostitutes.”16 A 1736 case from Fengtai County, Shanxi, concerns a music household comprising a woman, her two sons, and an adopted daughter, Zhao Shi, who (after growing up) helped support the family by prostitution. At some point, Zhao Shi gave birth to a son whose father was unidentified. In 1723, the imperial edict had come down ordering all music households to “reform and change to free commoner status” (gai liang); in enforcing this edict, the local authorities prohibited former music households from “harboring unmarried women.” In response, one of the brothers, Li Lanyu, took Zhao Shi as his wife and claimed her son as his own, “in order to stop people’s mouths,” but she continued to work as a prostitute in defiance of the ban. The household became dependent on one particular customer, Zhao Jinlan, who paid for regular sexual access to Zhao Shi (another example of the shift from retail prostitution towards polyandry). But when Jinlan ran out of money, they rejected him, and he ended up murdering Li Lanyu in an attempt to take Zhao Shi for himself. The magistrate who judged this case found that Li Lanyu had “abetted and tolerated” his wife’s “illicit sexual relations” with his killer, so she bore no liability for the homicide. Instead, she was found liable only for the illicit sex and for joining the killer in flight (her penalties were waived because of an amnesty). The statute mandated that she be “returned to her natal lineage,” but she had none, so the magistrate ordered her sold in marriage by the official matchmaker and her body price confiscated.17 Both of these cases illustrate how former music households were affected by the criminalization of their livelihood. One household tried to “reform” but was compelled by poverty to revert to its former occupation; the other never stopped engaging in prostitution but attempted to disguise what had formerly been open arrangements through the subterfuge of marriage. In both cases, women who had engaged in permitted prostitution up to 1723 ended up being prosecuted for the free commoner crime of “illicit sexual relations” that were “abetted and tolerated” by their husbands.18 After 1723, there was no longer any legal space for tolerated prostitution, and the prohibition against “abetting or tolerating” illicit sexual relations was cited in all of the central cases of polyandry and marital prostitution used for this study. The Yongzheng reforms had the same effect on the prohibition of “buying or selling a divorce” by automatically extending it to cover the formerly debased groups. But also, in 1646, Qing lawmakers had added a key amendment at the end of the statute’s second clause, in the form of a “small character annotation” (xiao zhu): If a husband does not report his wife’s adultery to the authorities, and instead sells her in marriage (jia mai) to the adulterer, then the original husband shall receive 100

Chapter 9. Formal Law and Central Court Interpretation

287

blows of the heavy bamboo, and the man and wife who engaged in illicit sex shall each be punished to the full extent of this statute.19

Under the original, narrow interpretation of the prohibition that had prevailed in the Ming, this amendment would have been superfluous—indeed, redundant— because prior adultery was the only scenario that was supposed to count. The 1646 amendment is critically important because it signals the triumph of the new broad interpretation: it leaves no doubt that prior adultery is just one of the scenarios to which the prohibition of “buying or selling a divorce” would apply. Henceforth, all unauthorized wife sales would be covered by this prohibition, including those motivated solely by poverty. Qing central courts followed this principle consistently. Central case records show that the statute against “buying or selling a divorce” was rigorously applied to a wide range of scenarios of wife-selling that included no prior adultery.20 Moreover, even local magistrates judging routine minor cases would frequently cite the prohibition of “buying or selling a divorce” to denounce poverty-driven wife sales.21 AU T HO R I Z E D M A R I TA L SE PA R AT IO N I N Q I N G L AW

Qing law did continue to authorize husbands to divorce and even sell their wives under certain circumstances, despite the expanded prohibition of wife sales. (The situation was different for prostitution, which was entirely banned by the Yongzheng reforms.) Moreover, as we have seen, nearly all divorces required the payment of compensation to the first husband, the ultimate source of which was the new brideprice paid by the woman’s second husband when she remarried; therefore, it makes sense to consider most of these transactions indirect wife sales. How then did Qing magistrates distinguish criminal wife sales from authorized marital separation? Ming-Qing Divorce Law The Ming code’s chapter on “household and marriage” contains a prohibition against the unauthorized “expulsion of a wife” (chu qi), which indirectly explains the conditions under which such expulsion was permitted: If a husband expels his wife even though she does not deserve expulsion and there is no circumstance that constitutes breaking the moral bond between them (wu ying chu ji yi jue zhi zhuang), then he shall receive 80 blows of the heavy bamboo. Even if a wife has committed one of the seven offenses that justify expulsion (qi chu), as long as one of the three conditions forbidding expulsion (san bu qu) applies, then he may not expel her; if he does so, then his penalty shall be reduced by two degrees. The wife shall be recovered and reunited with her husband.

288

part three

This statute was incorporated into the Qing code in 1646. The “seven offenses that justify a wife’s expulsion” were failure to give birth to a son, licentiousness, failure to serve her parents-in-law, talkativeness, theft, jealousy (presumably of a concubine), and severe illness. But this apparent severity was mitigated to some degree by “the three conditions forbidding expulsion”: if a wife had properly mourned a parent-in-law, had stood by her husband as his condition improved from poverty to wealth, or no longer had any natal family, then he could not expel her, even if one of the seven reasons to do so was present. (The seven offenses and three conditions were enumerated in Ming commentaries and added to the Qing version of the statute as a small character annotation.) However, a Ming edict later codified as a Qing substatute specifies that these restrictions did not protect a wife from expulsion if she had committed adultery.22 Qing legal cases suggest that a husband’s legal prerogative to expel his wife was far less arbitrary than the code’s broad language would seem to suggest. For example, in a case from Bagou Subprefecture, Zhili, also memorialized in 1762, a man named Liu Facai decided that he no longer wanted his wife, Zhang Shi (29), because “she was lazy, she resented my poverty, and she complained and made a fuss all the time” (conduct that might fall into the category “talkativeness”). He went out into the street and shouted “I don’t want this woman anymore. Any man who wants her can have her!” A tofu peddler named Cui Po (30), who was too poor to afford a wife, happened to be walking by. He heard Liu Facai’s shouted offer, and after ascertaining that Liu was serious, and that Zhang Shi herself wanted to leave him, he kowtowed to Liu and took her home to be his wife. But the new couple later quarreled, and Cui ended up killing Zhang Shi. When this homicide came before the magistrate, he also punished Liu Facai for giving away his wife. No money had changed hands, so Liu was not guilty of “selling a divorce”; nevertheless, he was sentenced to eighty blows of the heavy bamboo under the statute cited above, for “expelling a wife without grounds for expulsion.” Evidently, “laziness” and complaining were not sufficient justification for such an extreme step—especially if undertaken in such a cavalier manner. Had Liu returned his wife to her natal family instead of just giving her away to another man, the magistrate would probably have shown more sympathy.23 The Ming statute on “expulsion of a wife” further provides that “if a husband and wife do not get along with each other, and they are both willing to separate (liang yuan li zhe), then they shall not be punished for doing so.” The Qing small character annotation adds a pragmatic consideration: “If in fact a couple has already separated, then it is difficult to force them to reunite.”24 This provision seems generally to have guided Qing magistrates in handling routine marital disputes, when couples sought formal permission to divorce. Again, however, “mutual willingness to separate” seems to describe the case just cited, but that did not save Liu Facai from a beating.

Chapter 9. Formal Law and Central Court Interpretation

289

Moreover, as we have seen, nearly every wife sale resulted from a couple’s “mutual willingness to separate.” But that fact in no way mitigated the offense of “buying or selling a divorce” as interpreted by central courts during the Qing. Th e principle is stated very clearly in the judgment of a 1777 homicide case from Xing’an County, Jiangxi, that involved an illegal wife sale: “This husband and wife were mutually estranged and willing to separate. Nevertheless, for a husband to sell his wife in marriage for money, and for a second man to buy her in marriage, despite knowing the facts of her situation, constitute the crime of ‘buying and selling a divorce.’ ” The parties were sentenced accordingly, and their sentence was approved on review.25 In routine cases at the local level, magistrates often made the same point in their rescripts to litigants. For example, Magistrate Xu Jinyong of Nanbu County made the following comments in two separate cases (dated 1877 and 1878, respectively): “If husband and wife do not get along they are permitted only to separate so that she returns to her natal lineage, but it is forbidden for him to sell her in marriage on his own, without authorization (si xing jia mai)”; “If husband and wife have an unharmonious relationship then the law permits them only to separate by mutual consent. To sell her divorce for money (shou cai mai xiu) is prohibited.”26 Occasionally, however, a local magistrate would make a mistake that required explicit correction by his superiors. We find an especially clear example in a 1736 case from Xuyong Independent Subprefecture, Sichuan, in which one Zhao Ying had sold his wife because of poverty and illness, after which his incessant demands for supplementary payments had finally provoked the buyer to kill him. The local magistrate initially ruled that, since Zhao and his wife had been “willing to separate,” the wife sale could be permitted under the code’s provision that “if a husband and wife do not get along with each other, and they are both willing to separate, then they shall not be punished for doing so.” On that basis, the magistrate absolved the wife and matchmaker of any crime (the seller was dead, and the buyer sentenced for homicide). But the Sichuan provincial judge rejected this ruling: With regard to “buying and selling of divorce,” the code’s chapter on “illicit sex offenses” contains a different law that is the correct one. According to that law the wife, matchmaker, and other parties should all be punished. But the subprefectural magistrate has instead cited the statute on “expelling a wife,” which does not fit the facts of this case.

In response, the magistrate issued a new sentence that strictly applied the statute against “buying or selling a divorce”: the wife and matchmaker were ordered beaten, she was returned to her natal family, and the brideprice was confiscated. This new sentence was approved on review.27 Under the law, wives had far less autonomy than husbands in initiating divorce. As the early Qing jurist Shen Zhiqi observed, “it is a wife’s duty to follow (cong) her

290

part three

husband; therefore a husband can expel his wife, but a wife cannot break her own tie with her husband (fu ke chu qi, qi bude zi jue yu fu).”28 A Ming edict (later encoded as a Qing substatute) provided that if a betrothal lasted five years but the groom had failed to accept his bride in marriage, or if a husband had left home for three years without returning, then a woman could petition a magistrate for permission to marry someone else, and her family could keep the brideprice originally paid for her.29 But even when a husband had been away for three years, his wife could not “remarry on her own without authorization” (shanzi gaijia). If she did so, she risked one hundred blows of the heavy bamboo. If a woman “abandoned her husband and ran away,” then she would receive one hundred blows of the heavy bamboo and her husband would be authorized to sell her off in marriage (cong fu jia mai). If she ran away from him and married someone else, then her sentence was strangulation after the assizes.30 The Qing retained all of these Ming provisions. Qing archives contain many examples of women who did remarry on their own accord. Some women remarried because husbands had been away for a long time, leaving them without adequate means of support. Presumably these women could have petitioned for permission to remarry, if they could wait the requisite three years, but I have come across only one who actually did so.31 It is far more common to find women being punished for acting on their own accord—usually after their first husbands finally came home and discovered what had happened. But if a woman’s in-laws had arranged her remarriage during her husband’s absence (a scenario examined in Chapter 8), then magistrates put the onus on the senior inlaw instead of the woman herself. In other cases, women who were unhappy with their marriages would simply run away and pretend to be widows, in effect arranging fraudulent wife sales for themselves (using the proceeds to pay off whatever debts they had accumulated since running away). If caught, these women were punished more harshly, although I have yet to see a case in which one was actually sentenced to strangulation. The Legal Treatment of Compensated Divorce The practice of compensated divorce was a difficult gray area for Qing magistrates.32 This form of marital separation certainly constituted “buying a divorce” in the literal meaning of those words, in that the husband had to be paid to secure his agreement to divorce his wife. Moreover, divorce usually required the wife’s immediate remarriage, so that the new brideprice could be used to pay off her first husband. Given the close resemblance between such divorces and direct wife sales, how did magistrates judge them when they came into court? A case judged by the famous Ba County magistrate Liu Heng33 in 1826 illustrates the circumstances in which a compensated divorce might be legal. In this

Chapter 9. Formal Law and Central Court Interpretation

291

case, a peasant named Liu Dean (42) had filed a petition saying that his young nephew, Liu Ruipin (16), had been cheated out of his wife, Zhang Shi: Taking advantage of my nephew’s youth and foolishness . . . [the matchmakers] tricked him into drawing up a contract of divorce, with his handprint as proof, and then they married his wife Zhang Shi to Huo Weiqian. They promised to pay my nephew 15,000 cash, but have not given him even a single coin. That night, they carried Zhang Shi off.

Magistrate Liu ordered a hearing, noting that “both buying and selling a divorce are serious violations of the law.” But the hearing produced a different account. Magistrate Liu summarized the facts as follows: Liu Ruipin was very ill and impoverished, with no means of getting by or supporting his wife. He and his wife were both willing to separate (liang yuan liyi), so he divorced his wife, returning her to her natal family. Then his wife’s uncle Zhang Mingli remarried Zhang Shi to Huo Weiqian, receiving a brideprice of 7000 cash. Because he pitied the former son-in-law’s illness and poverty, he gave the entire amount to Liu Ruipin. Ruipin’s uncle Liu Dean noticed that Huo Weiqian was fairly well supplied with clothing and food; so, hoping to extort money, he filed false charges that Ruipin’s marriage had been broken and his wife sold (chai jia), that Huo Weiqian had stolen Ruipin’s property, and so on.

Huo Weiqian had already made one supplementary payment of 3000 cash in an effort to appease Liu Dean. In the magistrate’s view, this was a case of extortion pure and simple, and he ordered Liu Dean beaten and held in custody until Huo’s 3000 cash had been refunded. The divorce and remarriage themselves were perfectly proper: Zhang Shi is the divorced and returned wife (xiu hui zhi qi) of Liu Ruipin; her uncle officiated over her remarriage, and Huo Weiqian went through a matchmaker to pay a brideprice to take her in marriage. There was no [buying or selling of divorce] involved; there is no need to punish any of them. Zhang Shi shall be returned to Huo Weiqian to continue as his wife.

In other words, this transaction did not constitute a criminal wife sale because the money and woman were not exchanged directly between first and second husband. Rather, Zhang Shi had first been “divorced and returned” to her natal family, who then married her to Huo; equally, the brideprice had been transferred to the first husband by her natal family, after he had already divorced her. The testimony makes clear that all the arrangements, including compensation of the first husband, had been negotiated together in advance. Nevertheless, because the second marriage had proceeded only after the divorce had already taken place, both transactions could be considered legal.34

292

part three

Permitted and Prohibited Divorces Compared The judicial distinction between permitted and prohibited divorce is cast in stark relief by a central case from Qiongshan County (on Hainan Island), Guangdong, memorialized in 1745, that provides three examples of compensated divorce. Only one was judged to be a criminal “divorce sale,” illuminating the specific criteria for doing so. This case did not involve violent crime, and it was reported up the chain of command because it involved a naval officer (shoubei) named He Chaochen. After He Chaochen (who was from Shunde County, Guangdong, on the mainland) took up his post on Hainan, he decided to take a concubine. He received a fraudulent offer to marry a young “widow” named Han Shi; in fact, she was the wife of one Xie Nasheng (37), who wanted to sell her because he was ill and they were very poor. As Xie later testified, “my mother and I talked things over and decided to sell Han Shi in marriage (jia mai) for a few taels of silver so that I could buy medicine to treat my illness; if I died, we could use the money to buy a coffin and pay my burial expenses.” Han Shi informed her own mother, Xie Shi, of this decision; Xie Shi then borrowed 5 taels (the amount of Han Shi’s original brideprice when she had married Xie Nasheng) and visited Xie and his mother in the company of a matchmaker to pay them this compensation in exchange for Han Shi’s divorce. Xie Nasheng drew up a “divorce contract” (tui shu) authorizing Han Shi to marry someone else, and Han Shi went home with her mother. Six months later, the matchmaker approached He Chaochen, who agreed to pay a brideprice of 6 taels of silver to take “the widow” Han Shi as his concubine. With this money, her mother was able to repay the 5 taels (plus interest) she had borrowed. The magistrate ruled that this first transaction “was different from selling a divorce”; indeed, considered by itself, it constituted a legal divorce followed by a legal remarriage. But it happened that Han Shi had a brother who was a soldier under He’s command, and the Qing code prohibited officials from marrying relatives of their subordinates, so she was ordered divorced from He Chaochen and returned to her natal lineage. The magistrate noted that “the circumstances of the transaction were forgivable,” and Xie Nasheng (the invalid first husband) was allowed to keep the compensation he had received for the divorce. In the second transaction, He Chaochen had a nephew back home in Shunde County who had not yet married, and He decided to buy a wife for him (He had been pleasantly surprised, he later testified, by how “cheap” women were in Hainan). He asked one of his subalterns, Yang Xingzhang, to act as matchmaker, and Yang happened to know a woman whose daughter, Liang Shi, wanted to end her current marriage. Three years earlier, Liang Shi had married a sailor from Chaozhou Prefecture (on the mainland) named Li Bohui, and their marriage contract stipulated that Li could not “take her back to his home across the sea.” But Li decided to go home anyway, and he agreed (in Liang Shi’s words) to “take back my

Chapter 9. Formal Law and Central Court Interpretation

293

brideprice and let me marry someone else” (shou hui caili ting ping xiao furen ling jia). Matchmaker Yang Xingzhang negotiated the following deal. First, He Chaochen paid a new brideprice of 9 taels to Yang, who transferred the money to Liang Shi’s mother; she, in turn, used 7 taels to refund the original brideprice that husband Li had paid for Liang Shi, whereupon he gave her a “divorce contract” and sailed home alone to Chaozhou. The following month, the matchmaker helped Liang Shi’s mother draw up a new contract to give He Chaochen, and escorted Liang Shi to He’s yamen. He Chaochen, too, had promised that Liang Shi would not leave Hainan, but after the transaction had been completed he revealed that in fact he planned to send her to Shunde County to marry his nephew. She refused to go, and, after a month, He summoned the matchmaker and sent Liang Shi back to her mother. Thereupon, her mother arranged yet a third marriage for her, to one Huang Shaoren, for a brideprice of 6 taels. She used this money, plus 3 taels she borrowed, to refund He’s 9 taels. The magistrate ruled that He Chaochen’s second transaction did indeed constitute the crime of “buying or selling a divorce,” and even though the deal had been canceled, he sentenced everyone involved according to that statute. He Chaochen was cashiered from the military and sentenced to one hundred blows of the heavy bamboo; Liang Shi and her mother received the same sentence, and Yang received the lesser beating mandated for matchmakers. Both brideprice and matchmaker’s fee were confiscated. Liang Shi’s first husband, Li Bohui, was long gone, so he escaped punishment. (But her most recent marriage was ruled legal, so she was permitted to stay with her third husband.) Since this case involved the military forces, it was reported up through the review system and the magistrate’s judgment confirmed. This judgment focused on a legally significant distinction between the first and second transactions. In each instance, the key was the order in which divorce, refund of the first brideprice, negotiation of the second marriage, and payment of the second brideprice had taken place. The magistrate noted that, in the first transaction, by the time Han Shi’s second marriage was negotiated, “a significant amount of time had already passed after her first husband had divorced her”; in addition, He Chaochen had believed her to be a widow, “so this was definitely not a matter of ‘knowingly buying a divorce’ (zhiqing mai xiu).” Han Shi’s mother borrowed the money she used to pay off first husband Xie, and even though everyone knew she would repay the loan using the proceeds of Han Shi’s second marriage, that transaction still lay in the future. In the second transaction, as the magistrate pointed out, He Chaochen “knew full well that Liang Shi was a married woman, yet he still took her to be his nephew’s wife.” Most importantly, husband Li Bohui did not issue her divorce contract until after the second marriage had been negotiated and he had been paid, and the

294

part three

money he received had come directly from Liang Shi’s buyer (via the matchmaker and Liang Shi’s mother). In other words, from a legal point of view, the first husband had been paid by the second to divorce his wife. The distinction between the first and second transactions is understandable in terms of judicial reasoning. But from the standpoint of the participants themselves, this distinction must have seemed highly technical, if they understood it at all. After all, in both instances the intent was the same: to use the proceeds from the second marriage to finance the dissolution of the first, by using the second brideprice to refund the first one. In the first transaction, deemed legal, husband Xie certainly believed that he was “selling” his wife; he explicitly testified that he had “decided to sell Han Shi in marriage for a few taels of silver.” The first transaction also shared with many criminal wife-sales the element of fraud, in that the woman was represented to her buyer as a widow. This case suggests, in other words, that the categories and distinctions that judges imposed on divorce and remarriage did not necessarily make sense in terms of social practice. Curiously, the third transaction (i.e., Liang Shi’s third marriage, after leaving He Chaochen) might also appear to be criminal according to these criteria. The magistrate seems to have allowed this marriage to stand mainly for practical reasons; at any rate, he did not explain his decision, and it was not challenged on review. In fact, the only party to the third transaction who went unpunished was husband #3, Huang Shaoren, who had played no role in the second transaction. Given the assumption that Liang Shi had to marry someone, it must have seemed expedient to let her stay with Huang. He Chaochen was clearly the main target of prosecution because he was an official, and he had already seen his career ruined in punishment for the second transaction. One wonders if the real reason He was prosecuted is that he had political enemies for whom his indiscreet trafficking in women simply provided a convenient avenue of attack.35 T H E S A L E O F U N C HA S T E W I V E S

In addition to divorce, the Ming code provided that a wife “may be sold in marriage by her husband” (cong fu jia mai) in at least three situations: if she had been convicted of adultery, if she had “abandoned her husband and run away,” or if she had schemed with a lover to force her husband to “sell her divorce” to him. The Qing code preserved all of these provisions.36 Each scenario constituted a criminal betrayal of the husband, so the law authorized him to sell her (or to keep her if he chose) after reporting her to the magistrate for prosecution. It was the husband’s righteousness and his betrayal by his wife that entitled him to sell her—in contrast with the crime of “buying or selling a divorce,” which Ming jurists defined as a husband’s abdication of responsibility by facilitating his wife’s licentiousness. In practical terms, authorizing a betrayed husband to sell his wife would help him

Chapter 9. Formal Law and Central Court Interpretation

295

recover at least part of the brideprice his family had originally invested in her, perhaps enabling him to remarry and start anew. Authorized Sale of an Adulteress by Her Husband In most wife sales, the woman herself had some say and her cooperation was necessary. In an authorized wife sale, however, the disgraced woman was an object of punishment, and it was far less likely that her wishes would have any influence. The following 1750 case from Liancheng County, Fujian, provides a vivid glimpse of the unhappiness that might result. Huang Shi (42) was the wife of Li Zhanlian; she had a sexual affair with one Li Fuyuan, and her husband caught them and turned them in for prosecution at the county yamen. The magistrate had the adulterers beaten, and he returned Huang Shi to her husband with the option of “keeping her or selling her” (ting qi qu liu), as provided by statute. He sold her to a widower named Zhang Xuemei (51), whose first wife had died childless. Zhang was an orphan who had been raised by his widowed aunt, and now he was the only heir to his father and uncle both. He had neither wife nor sons—a cause of great anxiety to him and his aunt—so they eagerly seized this opportunity to buy a wife for a relatively low price. But Li Zhanlian continued to bear a grudge against the man who had cuckolded him, and three months later he beat up Li Fuyuan, who responded by hanging himself (apparently intending his suicide to implicate Zhanlian). The circumstances of the suicide triggered a criminal investigation, and the sold adulteress Huang Shi was summoned to the county seat to testify. There were various delays, and, before she could testify, Huang Shi gave birth to a baby. In fact, she had already been pregnant when sold to Zhang Xuemei—it is unclear if the father was her first husband or her lover; according to Zhang, “she brought that pregnancy with her, it wasn’t fathered by me; so after the child was born, I gave it away to be adopted by somebody else.” With that problem taken care of, Zhang wanted to take Huang Shi back home, since according to law she would have to wait a month after giving birth to testify in court. But she was terribly upset at the cavalier way Zhang had disposed of her baby, and she refused to return home with him; instead, she insisted that he allow her to visit her former husband, Li Zhanlian, who was being held in jail to await trial. This idea incensed Zhang, who already resented the trouble his new wife had caused. “I scolded her for her shamelessness,” he later testified; she responded by cursing and defying him, so he hit her; but she hit him back and fought and refused to yield—so he ended up beating her to death. The main point of this case for the present discussion is that Huang Shi’s marriage to Zhang Xuemei was legally valid, because her sale had been authorized in advance by the magistrate who judged her adultery. The new marriage’s legitimacy was reconfirmed by Zhang’s sentence according to the statute against “a husband

296

part three

beating his wife to death” (the nominal penalty being strangulation after the assizes, commuted to a beating and fine because Zhang was the sole heir of his deceased father and uncle). In contrast, a crime between partners in an illegal marriage would have been judged as occurring between unrelated parties, so the ordinary homicide statutes would have applied. The magistrate and his superiors offered no comment on what Zhang had done with Huang Shi’s baby, thereby indicating that they deemed his action acceptable. We learn of Huang Shi’s feelings only through the testimony of others, but it is clear that, by giving away her baby, Zhang had wounded her terribly. She responded by demanding to see the former husband who may have been that baby’s father, knowing full well that this demand would provoke her new husband to a jealous rage—a rage matched by her own. It appears that, in her eyes, of the three men who had claimed her, the worst was the one she ended up being stuck with, Zhang. Perhaps she hoped to provoke him into reselling her in marriage to someone else.37 Sale of an Adulteress by the Official Matchmaker Sometimes an adulteress renounced by her husband would not be sold directly by him but instead through the official matchmaker (guan mei) at the county yamen. Official matchmakers played a key role in the way magistrates handled women who came into their custody (including women convicted of prostitution), and they appear to have been a standard fixture of every yamen.38 We find references to the institution in the Qing code and in case records from many parts of China. The novel Jinpingmei cihua (Plum in the Golden Vase) shows that the institution had existed at least since the Ming. In one episode, Sun Xue’e (concubine of the deceased Ximen Qing) is arrested after having an affair and running away with a manservant; she is punished with the finger-press and ordered taken home by the Ximen household. But Ximen Qing’s widow, Wu Yueniang, refuses to take Xue’e back because of the scandal she has caused, so “the magistrate sent for the official matchmaker to sell her off under official auspices.”39 We see the identical procedure in an 1827 case from Ba County. A Chongqing tailor named Gu Yongfa came home early one afternoon and caught his wife, Bai Shi (17), in bed with a neighbor, Peng Rongzun (in his twenties). Peng tried to run, but Gu knocked him down with a piece of firewood and, with the neighbors’ help, proceeded to beat him up, bind him with ropes, and take him and Bai Shi to the county yamen. The magistrate ordered Peng Rongzun held in custody until his injuries healed, at which point he would be beaten and serve a term in the cangue. Bai Shi was beaten, too, and renounced by her husband; Gu testified, “I do not want Bai Shi anymore, and I am willing to divorce her,” and agreed that Bai Shi be “handed over to the official matchmaker to be sold in marriage.” Gu and his mother provided affidavits confirming this and promising no future trouble. After six weeks, the

Chapter 9. Formal Law and Central Court Interpretation

297

official matchmaker (a widow) found a shopkeeper in Chongqing, Liu Zhaoyuan, who paid a “body price” of 4400 cash for Bai Shi. After deducting 2400 cash to cover her commission and expenses, the matchmaker submitted the balance to the magistrate. Liu submitted a “receipt” (ling zhuang) for Bai Shi in which he acknowledged Bai Shi’s record of adultery and promised to take her home as his wife and not to pimp or resell her. The magistrate approved, and Bai Shi was handed over to Liu. The paperwork in this case resembles the documents in illegal wife sales, in that they document the informed consent of both Bai Shi’s husband and her buyer in order to prevent future controversy.40 Authorized Sale of an Adulteress as a Slave The basic statute against “consensual illicit sex” in the Ming and Qing codes specifies that a wife convicted of that offense may be “sold in marriage” (jia mai) by her husband, but such women might also be sold as slaves, and magistrates accepted such transactions. We have already mentioned the episode in Plum in the Golden Vase when Xue’e is sold by the official matchmaker. In fact, Xue’e ends up being sold not as a wife but as a kitchen slave, and she is later resold to a brothel, where she finally hangs herself. Her descent from adulteress to slave to prostitute exemplifies the karmic retribution visited upon all of the novel’s protagonists, but it is also a believable chain of events, given the strong association of unchaste conduct with debased status during the Ming. The sale of adulteresses into slavery may have been unusual in the Qing, but it did happen. In a 1775 case from Ba County, Yang Shi, the wife of a Chongqing man named Chen Youfu, ran off with one Luo Wendeng to nearby Bishan County. The couple was arrested, and the Bishan magistrate had Luo put in the cangue and Yang Shi returned to her husband, who was authorized to keep her or sell her. Chen took her back to Chongqing and sold her as a slave (nu) to a holder of the jiansheng degree named Gong Wenlin, for a body price of 33,000 cash. This matter came to court when Yang Shi ran away from Gong’s home, and he filed suit against her husband in an effort to recover his money. A mediated settlement (approved by the magistrate) canceled the sale, refunded Gong’s money, and returned Yang Shi to her husband. But nothing in the record implies that the magistrate saw anything wrong with Yang Shi’s sale as a slave.41 Unauthorized Sale of an Adulteress As the statute against “consensual illicit sex” makes clear, a husband might sell a wife who had committed adultery only after receiving formal authorization from a magistrate. Many husbands did not bother with such formalities, however, and if these sales came to official attention, they too might be prosecuted as “buying or selling a divorce.” As one magistrate opined in an 1825 case (from Jiajiang County, Sichuan):

298

part three

Hu Zhifu knew full well that his wife Cai Shi was unchaste, but he failed to report her to the authorities so that she could be judged clearly and ordered divorced; instead he rashly accepted cash payment to sell her in marriage. Therefore he should be sentenced to one hundred blows of the heavy bamboo, according to the statute against “selling a divorce.”42

Similarly, in an 1884 case from Nanbu County concerning a wife sale supposedly provoked by the woman’s misbehavior, the magistrate admonished her husband: “The statute concerning ‘the seven offenses that justify a wife’s expulsion’ contains no provision permitting a husband to sell his wife on his own authority (si mai) to another man as concubine, even if she has committed an offense that warrants expulsion.” A husband could take this measure only with the court’s permission.43 The point is further illustrated by an 1801 case from Ba County. A man named Xia Yingguang had discovered that his concubine, Miao Shi, was having a sexual affair with one Zhou Bixiang, so he sold her in marriage to a man surnamed Peng and then filed charges of adultery against Zhou at the Ba County court. By this time, Peng and Miao Shi were long gone and could not be summoned. The magistrate ruled that Zhou and Miao Shi had committed adultery (Zhou admitted it). Nevertheless, since the woman could not be punished, the magistrate decided that her partner in adultery should not be punished either, and he was released on pledge of good behavior. Instead, the magistrate ordered husband Xia Yingguang beaten, for having sold his wife without permission.44 But magistrates did not always enforce this rule, even in central cases. It appears that, for some magistrates, the underlying principle that a wife’s betrayal of her husband entitled him to sell her outweighed the technical violation of failing to obtain advance approval. Thus, for example, in a 1799 case from Gong’an County, Hubei, the magistrate reasoned that: After Zhou Wenshao caught his wife Li Shi in the act of illicit sexual intercourse, the only reason he did not file charges was concern for face, so he should not be punished for failing to do so. Li Shi is “a wife who has committed the crime of illicit sexual intercourse,” so in principle she “may be sold in marriage by her husband.” Since she has already been taken in marriage by Gao Qifeng, she should be returned to him.45

Similarly, in a 1784 case from Qianjiang County, Hubei, the magistrate ruled that: Gan Shi is “a wife who committed the crime of illicit sexual intercourse,” and therefore for her husband to sell her is not the same as “selling a divorce.” Peng Tianyi acted as matchmaker to arrange for her to be remarried. Under the circumstances, it is appropriate to request that neither of them be punished, and that confiscation of Peng Tianyi’s matchmaker fee be waived.46

In my sample of central cases involving the unauthorized sale of an adulteress by her husband, about half of these transactions were prosecuted as “buying or selling

Chapter 9. Formal Law and Central Court Interpretation

299

a divorce,” and the other half ratified retroactively. Moreover, since all of these judgments were approved on review, it is clear that the central judiciary was content to leave this decision to the discretion of local magistrates. Even if a husband did not sell off an adulteress in this unauthorized manner, he could still get in trouble if he failed to report her adultery to the authorities. “Mediating a private settlement” (si he) in such circumstances could be prosecuted according to the following statute: If someone acts as a go-between to facilitate or permit illicit sexual relations, then he shall receive a penalty one degree lower than the offenders themselves. If someone privately mediates a dispute arising from illicit sexual relations [without reporting it to the authorities] (si he jian shi), he shall receive a penalty two degrees lower.47

It should be noted that the Chinese word usually translated as “private”—si—has strongly negative connotations; a more precise translation in this context would be “illicit” or “unauthorized.”48 As far as I can tell, this law was enforced only in major cases such as homicides when “private mediation” of this kind formed part of the background to the principal offense. In routine cases that involved only “minor matters,” magistrates would accept almost any sort of mediated settlement, as long as it took place prior to a formal court hearing (see discussion in Chapter 11). We find an example of “private mediation” being prosecuted in a 1741 case from Tiantai County, Zhejiang, in which Yu Yongshou caught his wife, Xu Shi, having sex with his cousin, Yu Ziren (who already had a wife of his own). Yongshou originally planned to file charges, but his lineage elders imposed a settlement whereby Xu Shi would be divorced and Ziren’s family would pay Yongshou compensation of 22 taels (an exact refund of Xu Shi’s original brideprice) to enable him to take a new wife. Ziren’s brother Yu Yuanxi made a down payment of 11 taels and issued a promissory note for the rest; then Xu Shi was returned to her natal family and remarried to another man (Ziren received no compensation from her new husband and played no role in negotiating her new marriage). But Ziren’s family could not pay the balance owed, preventing Yongshou from taking a new wife. This impasse eventually provoked a fight, when Yongshou’s father tried to drag Ziren to the ancestral hall to be judged by the lineage, and Ziren’s wife was severely injured when she tried to intervene (she later died). When this homicide case came to court, everyone who had participated in mediating the settlement was punished according to the statute cited above, and the compensation money was confiscated (including the unpaid balance). Xu Shi was punished for adultery, but the magistrate ruled her second marriage to be legal, on the grounds that she had “remarried after having already been divorced by Yu Yongshou.” In other words, the mediated settlement violated the law but the divorce and remarriage themselves did not, because they had followed the proper chronology and Yongshou had received no money from Xu Shi’s new husband.49

300

part three

Sale of an Adulteress to Her Partner in Adultery A fundamental restriction on authorized wife-selling was that the betrayed husband could not sell his wife to her partner in adultery. As we have seen, the scenario originally targeted by the Ming statute against “buying or selling a divorce” was precisely a husband’s sale of his wife to her partner in adultery, as an extension of the crime of “abetting or tolerating” that adultery. The Qing amendment of 1646 (the small character commentary discussed above) reinforced this point.50 In addition, the Ming statute against “consensual illicit sex” provided that: A wife who engages in illicit sex may be sold in marriage by her husband; if he is willing to keep her, then he shall be allowed to do so. But if her husband sells her in marriage to the man who has engaged in illicit sex with her, then that man and the husband shall each receive eighty blows of the heavy bamboo; the wife shall be divorced and returned to her natal lineage, and the brideprice shall be confiscated.51

The Qing retained this provision as well; it applied when a husband had had his wife and her lover prosecuted, and then afterward sold her to him anyway.52 Simply put, the purpose of these laws was to punish adultery, not to facilitate or reward it—and therefore, a husband’s prerogative of selling his wife was limited to specific circumstances in which he retained his moral authority. If he acquiesced to her adultery after the fact, even if he had had her prosecuted, then he would lose that authority—as well as any claim to determine his wife’s fate or benefit from her sale. Regardless of what the code said, case records show that, if a husband wanted to sell his wife because of adultery, the path of least resistance might be to sell her to her partner in adultery. After all, selling her to her lover would obviate the sometimes complicated and time-consuming process of finding someone else willing to buy her, and the woman herself would be more likely to cooperate if she were sold to a man with whom she was already intimate. Moreover, if her lover was passionately attached to her, then he might be willing to pay more than another buyer. This is exactly what happened in a case from Yanshi County, Henan, memorialized in 1753, in which peasant Ma Zhongrong caught his wife, Li Shi, having sex with a neighbor, Ma Xian. Zhongrong filed charges at the county yamen, but Ma Xian evaded arrest for several weeks, and the proceedings stalemated. Tired of waiting, Zhongrong decided to sell his wife without bothering with a court hearing, and he engaged his friend Cao Yingshi to act as matchmaker; Ma Xian heard this news, and asked Cao to arrange for him to buy Li Shi. To mollify Zhongrong’s anger and persuade him to agree, Ma Xian offered a huge brideprice of 85 taels of silver, plus 5 taels for the matchmaker (this brideprice is so high that it suggests something like punitive damages). Cao conveyed this offer to Zhongrong, who agreed.

Chapter 9. Formal Law and Central Court Interpretation

301

Eventually, however, the yamen runners did manage to arrest Ma Xian (whose warrant was still outstanding), and even though Zhongrong no longer wished to press charges, a formal court hearing ensued. (The case became far more serious when Ma Xian’s father committed suicide out of shame at his son’s behavior.) The magistrate ruled that the wife sale was criminal, because even though Zhongrong had reported Li Shi’s adultery, he had sold her without waiting for the court’s permission to do so; even worse, he had sold her to her partner in adultery. Citing the 1646 amendment to the statute against “buying or selling a divorce,” the magistrate sentenced Ma Zhongrong, Li Shi, and the matchmaker to beatings, divorced Li Shi from both men and returned her to her natal family, and ordered the confiscation of both the brideprice and the matchmaker’s fee. Ma Xian was sentenced to immediate strangulation for “unfilial conduct” in provoking his father’s suicide.53 When an illegal wife sale was canceled and the wife “divorced and returned to her natal lineage,” the expectation was that she would be permanently separated from both her first husband and her buyer, and be married off to some other man who had not been involved in the illegal transaction. But in this situation, too, the path of least resistance for a natal family trying to find a place for their daughter might be to marry her to the man who had already bought her, as long as he was willing to proceed. Nothing in the code explicitly prohibited such a maneuver, though it certainly violated the spirit of the law against “buying or selling a divorce.” We find confirmation of this point in a case from Luchuan County, Guangxi, memorialized in 1825. Peasant Wu Wansheng had a son, Wu Xingyong, for whom he had acquired an adopted daughter-in-law, Lu Shi. The young couple had not yet consummated their marriage when, in the winter of 1824, Wu Wansheng decided that because of poverty they had no choice but to sell her. Through matchmakers, they arranged to sell her in marriage to one Chen Yigui. But Wu Xingyong, who had bitterly opposed the sale of his bride-to-be, reacted by cutting his own throat on the very night the transaction was finalized; he died, his suicide was reported to the yamen, and as a result Lu Shi’s new marriage was declared to be an illegal act of “selling a divorce.” Wu Wansheng and Chen Yigui were both beaten. Lu Shi’s beating was waived, on the grounds that she had had no choice but to obey her father-in-law, and she was ordered returned to her natal lineage so that her mother could marry her off to someone else. However, Lu Shi’s mother, widow Lu Lin Shi, felt that since her daughter had “already consummated her marriage” with buyer Chen Yigui, it would be best to reunite them. Lu Shi herself agreed, Chen was willing to take her back, and within a month a new match had been arranged in exchange for a new brideprice of 10 taels paid to the natal family; Lu Shi’s uncle drew up a new marriage contract, and she returned to Chen’s household as his wife. Shortly thereafter, however, Lu Shi’s cousin Lu Juming learned that she had been remarried to Chen Yigui, and he warned that this violation of the magistrate’s ruling could lead to trouble. He and

302

part three

another cousin armed themselves with staves and visited Chen to try to make him cancel the marriage. Chen refused, they got into a fight, and Chen killed Juming with a blow to the head. When this homicide came before the court, Lu Shi’s second marriage to Chen was also ruled illegal, and she was ordered “returned to her mother’s custody” again “to be married off to someone else.” Her mother was sentenced to eighty blows of the heavy bamboo for “doing inappropriate things—serious cases,” because “she once again promised her daughter in marriage to the man who had bought her divorce, even though the magistrate had already ordered her daughter divorced from that man.” The new brideprice was confiscated. Once again, Lu Shi herself was spared punishment on the grounds that she had had no choice but to obey her mother.54 C A SE S I N VO LV I N G C O N C U B I N E S A N D SL AV E S

Concubines and slaves represent special cases, because both acquired their status through purchase, and the law acknowledged that fact. In that sense, at least, they could be considered chattel. How, then, would a wife sale involving such persons be judged? Sale of a Concubine by Her Husband A concubine (qie) held a somewhat ambiguous legal status in the polygynous household, in between a main wife (zheng qi or qi) and a bonded maidservant (binü). On the one hand, a concubine was a legal secondary spouse whose children were legitimate and whose sons had full inheritance rights. If widowed, a concubine was entitled to remain single and to enjoy the support of her husband’s household, as long as she remained chaste. In these respects, a concubine resembled a main wife. On the other hand, a concubine’s position was distinctly subordinate and to some degree stigmatized. Like maidservants, concubines would be purchased from their parents, via brokers, and brought in with no fanfare. In fact, one route to concubine status was for a maidservant simply to be “favored” (xing) by her master and promoted once she became pregnant.55 In contrast, the marriage of a woman as main wife (who, unlike concubines, would come from the same social status as her husband) was an occasion for celebration, and in prosperous households a wife would bring dowry to emphasize that she had not been sold—underscoring the distinction between herself and the subordinate women over whom she exercised authority as mistress of the inner quarters.56 Ming and Qing law codified the subordinate status of concubines in a number of ways. For example, a man could have more than one concubine but only one main wife, and it was a criminal offense to demote a wife to concubine status or to displace a wife by promoting a concubine.57 In addition, the relative weighting of

Chapter 9. Formal Law and Central Court Interpretation

303

penalties for various offenses reflected the subordination of concubines as a way to reinforce normative family hierarchy. If a main wife beat a concubine, she would be judged in the same manner as a husband who beat his wife—that is, there would be no penalty at all unless the beating caused broken bones or worse injury. But if a concubine beat a wife, she would be judged far more harshly.58 The statute against “buying or selling a divorce” explicitly includes concubines as well as wives in that prohibition, with the following qualification: “If the woman in question is a concubine rather than a wife, then the penalties shall be lowered by one degree.”59 In other words, a concubine’s subordinate status made her sale less objectionable than sale of a wife. If one turns to central cases, however, the record is mixed. One can certainly find central cases in which magistrates prosecuted a husband’s sale of his concubine under the prohibition against “buying or selling a divorce.”60 But some magistrates seem not to have taken this offense so seriously. The clearest example I have is an 1819 case from Huitong County, Hunan, in which a man named Zhou Lu (33) had purchased a concubine, Liu Shi (19 at the time of purchase), because his main wife had not given him a son. But the concubine and the main wife did not get along, so after two years Zhou’s mother ordered him to sell the concubine. Through matchmakers, Zhou arranged to sell her to one Peng Qishan for a brideprice of 49,000 cash (a high price reflecting her youth and attractiveness). Several days later, Zhou dropped by Peng’s house for a visit and happened to encounter Liu Shi. (The fact that such an encounter was possible shows that Peng was a man of fairly modest means, who did not live in a large enough house to enforce strict separation between inner and outer quarters.) Liu Shi cursed Zhou bitterly for having sold her and for refusing to let her keep all of her clothing. Angered by her effrontery, Zhou beat her, she fought back, and he ended up killing her. This transaction had all the hallmarks of “selling a divorce.” But the magistrate seems to have sympathized with the difficulty of managing a complex household that included three women of different statuses (Zhou’s mother, wife, and concubine). After all, this was the sort of domestic complication that any elite man might encounter. He reasoned that, since Zhou’s concubine did not get along with his main wife, and his mother had ordered him to sell her in marriage, he really had no choice but to do so—and for this reason, the transaction “was not the same as selling a wife.” Therefore, the buyer and matchmakers were not punished. Based on the same logic, however, the magistrate ruled that Zhou and Liu Shi should be treated as “unrelated parties” for purposes of judging the homicide: “He had already sold her in marriage for money, so the moral bond of husbandly benevolence and wifely duty between them had already been broken.”61 This judgment is unique in my sample because the magistrate articulated his reason for ruling the sale of a concubine to be legal. In other cases, magistrates did not even address the transaction, thereby ratifying it by default. It is also

304

part three

noteworthy that all these judgments were confirmed on review. Evidently, the central judiciary was content to leave this matter to the discretion of local magistrates. Wife-Selling Tolerated among Slaves An unusual 1738 case from Beijing documents another context in which wifeselling might be tolerated by the authorities: among slaves. By the mid-eighteenth century, bonded servitude no longer played a significant role in the productive economy of China, and servile laborers no longer constituted a significant proportion of the population, but wealthy households continued to maintain maidservants and other domestic slaves (some of whom were male), and this practice seems to have been especially common among the banner elites concentrated in Beijing.62 This case concerned a man named Shi Tu (45) and his wife, Han Shi (35), both Han Chinese, who were slaves (jia ren) in the household of an officer of the Manchu Yellow Banner named Ganzhou. The couple had “sold” (mai) themselves to him some years before for a body price of 30 taels of silver, in order to pay debts and escape poverty. But Ganzhou had fallen on hard times, so in late 1738, just before the New Year, (in Shi Tu’s words) “my master told me that we should resell (zhuan mai) ourselves to someone else and repay our body price to him.” The couple moved out of Ganzhou’s household and, while looking for a buyer, found temporary lodgings with Shi’s brother-in-law (who was a slave in the household of a Mongol banner officer). At first, Shi tried to find someone who would buy him and his wife together as a couple, but he had no luck, and Ganzhou insisted on being paid within the month. (The consequences of failure are not clear, but Shi Tu felt enormous pressure to come up with the money.) For this reason, “I proposed to sell my wife in marriage so as to use her brideprice to pay back our body price to our master. My mother-in-law and my wife both agreed to this plan.” (Han Shi’s mother was a slave in the household of yet another banner officer.) Shi Tu then enlisted three “brokers in persons” (ren yazi) to find a buyer. The brokers (who were Han Chinese civilians) found two different candidates, each willing to pay a brideprice of 26 taels, but for reasons not explained in the case record, Han Shi vetoed both matches (a good example of the role a woman might play in negotiating her own sale). Then the brokers identified a third buyer, an artillery officer of the Han Martial Yellow Banner named Ma Liu, who offered 20 taels to buy Han Shi to be the wife of his “nephew,” Qian Liu. This time Han Shi agreed, so Shi Tu closed the deal, despite the lower price. According to the brokers, Qian’s family made a living through money-lending, and (as Shi Tu later testified) Han Shi persuaded Shi to accept the lower offer by promising that, once married, “I’ll tell him that my mother is old, and I need to help her buy a coffin. That way I’ll get a few taels for you. Then you can pay the balance we owe our master for our body price.”

Chapter 9. Formal Law and Central Court Interpretation

305

Shi wrote out a contract himself and affixed his hand- and footprints. Of the 20-tael brideprice, 1 tael went to the brokers, he spent 2 taels himself on various expenses, and the remaining 17 taels he gave to Ganzhou as down payment (he still owed Ganzhou 13 taels). Three weeks later, he dropped in at Qian Liu’s home to visit Han Shi, and (according to his later testimony) when he asked for the promised money, she explained that they had been deceived by the brokers. It turned out that Qian Liu was not Ma Liu’s nephew but his slave, and the talk of moneylending had been lies. She ended their conversation brusquely: “I can’t get any money for you, so you should go take care of your own affairs on your own (ni gezi gan ni de shi qu ba).” Shi Tu returned home. But as he thought over the situation, he came to the conclusion that it was Han Shi, rather than the brokers, who had really deceived him: At first there were two other households that offered to pay 26 taels, but Han Shi refused. . . . Later, Qian Liu offered only 20 taels, but she was willing to go with him—and then she broke her promise to get me more money. I was angry! I still owed our master 13 taels, and with the deadline coming up there was no place for me to get that much money. Han Shi already had a secure place (anshen zhi chu), but here I was stuck with no one willing to buy me. I’d been tricked!

After brooding for a time, Shi Tu went back to Qian’s home and killed Han Shi. The key aspect of this case for our discussion is the sentence—or rather the lack of one—imposed by the Board of Punishment, which acted as court of first instance for Beijing cases. Since Shi Tu had sold his wife, “their moral bond of husbandly benevolence and wifely duty was already broken,” and they were ruled to be unrelated parties for purposes of sentencing. For this reason, Shi Tu was sentenced to beheading after the assizes according to the regular statute against “planned murder.” But no one else was punished, and the statute against “buying or selling a divorce” is nowhere mentioned in the case record. The legal term “sell a divorce” never appears, and instead Shi Tu’s sale of Han Shi is referred to simply as “selling for a price” (jia mai). Her new husband, Qian Liu, testified, as did one of the brokers, but neither was sentenced for any offense. Obviously, the board did not consider Shi’s sale of his wife illegal. How is this possible? No doubt, the key factor is that the protagonists were all slaves. But that deceptively simple fact begs a number of questions and leaves unresolved a basic contradiction. One might argue that the sale of Han Shi to become Qian’s wife really amounted to the sale of a slave by one master to another, since her brideprice was actually paid by Qian’s master Ma Liu, and most of the money ended up going to her own master. In other words, this did not really count as a wife sale after all. But the Board of Punishment specifically cited Shi’s sale of his wife to justify treating them as “unrelated parties” in sentencing him for homicide, so there can be no doubt that the board considered the transaction to be a wife sale.

306

part three

One might also argue that, as slaves, the protagonists in this case were persons of debased status to whom the free commoner standard of morality and criminal liability did not apply. Therefore, an act that would have been criminal for free commoners was permitted to them. But if that is true, then why did the board credit the couple with a “moral bond of duty” that could be “broken” by wife sale? Given its context, one cannot dismiss this language as a figure of speech. Before the Yongzheng reforms, Ming and Qing jurists specifically exempted married couples of debased status from being held to this standard—that is why, for example, in prostitute households husbands could pimp their wives with impunity.63 This moral bond of duty was the defining feature of free commoner marriage, and its “breaking” by wife sale was the reason why the code mandated that the wife be returned to her natal lineage. Why would the Board of Punishment invoke this standard for the purpose of homicide sentencing without mentioning the statute against “selling a divorce”? On the one hand, the protagonists in this case were people of debased status who could be bought and sold, and the free commoner standard of morality and criminal liability did not apply to them. On the other hand, whatever the law might say about debased status, Shi Tu’s testimony makes plain his shocked sense of betrayal when he realized that Han Shi had deceived him in order to secure herself an advantageous position at the expense of his own freedom. Obviously he felt that she owed him some sort of moral duty, even after he had sold her, because in negotiating her sale he had placed her interests before his own. But the board’s judgment does not address these contradictions, let alone resolve them. One cannot help concluding that Shi Tu was being held to two standards at once, in an anachronistic maneuver that partook of the age-old discrimination against persons of debased status even as it reflected the extension of free commoner standards that characterized the then still recent Yongzheng reforms.64 The circumstances in which central courts either authorized or tolerated some wife sales recall the age-old association of female chastity with free commoner status, which implied an equally strong association between sexual promiscuity, bonded servitude, and debased (i.e., unfree) legal status. Hence, an unchaste wife had debased herself, both figuratively and (in legal terms) literally, and that is why she deserved to be treated as chattel. A concubine might be chaste, but, if she had provoked her husband and mother-in-law into selling her, then her subordinate status in the household meant that magistrates might tolerate her sale. Despite all that had changed from Ming to High Qing, this web of associations continued to influence the central judiciary’s thinking. T H E C R I M I NA L I Z AT IO N O F C U ST OM

A signal innovation of the Qing judiciary was to expand the criminalization of prostitution and wife-selling far beyond what Ming lawmakers had envisioned.

Chapter 9. Formal Law and Central Court Interpretation

307

Qing jurists achieved this expansion through a combination of new legislation and a reinterpretation of the Ming statute that prohibited husbands from “abetting or tolerating a wife or concubine committing illicit sex offenses” or “selling a divorce.” On the one hand, the Yongzheng reforms eliminated the status-based exceptions to this statute, thereby extending criminal liability for offenses of “illicit sex” to include previously exempt groups. This reform had the effect of prohibiting all prostitution, most of which, in practice, seems to have involved married women. On the other hand, the Qing reinterpretation of the statute extended the prohibition of “selling a divorce” to include wife sales motivated by poverty—which were the vast majority of actual sales. We find scant evidence of wife sales being prosecuted in the Ming or even the early Qing, but the eighteenth century witnessed an explosion of prosecution that was a direct result of the criminalization of widespread survival strategies that had not been criminal under the Ming. That criminalization reflected the well-known fundamentalism toward sexual behavior and gender roles that characterized the High Qing as well as an unprecedented effort to regulate the lives of ordinary people. The expansion of criminal liability under the Qing is a good example of the process by which new institutions of power produce new kinds of information, one result of which is that previously obscure phenomena may become visible in the documentary record. Among other things, it created the vast archives of court cases that made possible the present study: without them, the practices documented here would have remained largely invisible.

10

Absolutism versus Pragmatism in Central Court Treatment of Wife Sales To be frank, we cannot expect foolish peasants in remote regions to know exactly what is forbidden by law. —the guizhou bureau of the board of punishment, 1789 1

The theme of this chapter is a distinct and rising tension between two opposing principles that can be discerned in the discourse and judgments of the central judiciary from the High Qing down through the late nineteenth century. On the one hand, there was the absolutist principle that defined all wife sales as “illicit sexual intercourse” and insisted on prosecuting them all as the crime of “buying or selling a divorce.” On the other hand, there was widespread awareness among officialdom that, in fact, most wife-selling was driven by poverty, and many officials felt that both pragmatism and compassion indicated a less draconian approach. Moreover, the push to extend an absolutist standard of female chastity to all came into conflict with more traditional fundamentalist Confucian priorities, in particular the primacy of filial piety and family hierarchy. The absolutist principle remained empowered discourse throughout the dynasty, although some jurists attempted to reduce its harshness by urging the consideration of mitigating circumstances. The measures they advocated represented a major concession to the social reality of wife-selling both as a survival strategy and as a de facto form of marriage, its ritual impropriety notwithstanding. Despite some gestures of compromise, however, the basic contradiction between these opposing principles was never resolved—except by the collapse of the dynasty. Here we address only central court discourse and judgments—in other words, the highly formal realm of ideology documented in legislation, Board of Punishment rulings, officially published reference works, and major cases reported up the chain of command for central review (for the hierarchy of review, see Appendix E). These texts constitute the sources for the present chapter, and its topic must be 308

Chapter 10. Absolutism versus Pragmatism

309

clearly distinguished from the purely local adjudication of routine cases involving “minor matters” (xi shi) that Chapter 11 shall analyze in detail. We approach this topic from three different angles. First, there is the question of whether a sold wife should really be treated as an adulteress, which arose in several contexts. Should a wife convicted under the statute against “buying or selling a divorce” actually have her mandated beating carried out, as would be the case with a woman convicted of adultery? And if such a woman were kidnapped or raped, how should the offenders be sentenced—that is, should they receive reduced penalties, as if they had kidnapped or raped “a woman who has committed offenses of illicit sex”? Finally, should a wife who died resisting sale be eligible for canonization (jing biao) as a chastity martyr, as would be the case if she had died resisting other challenges to her chastity? Second, there is the question of how wife sale affected the legal standing of family relationships. Jurists spilled a remarkable amount of ink over this question, because it went to the heart of what T’ung-tsu Ch’ü (Qu Tongzu) has called the “particularism” by which imperial law codified normative family hierarchy (see Appendix D.2).2 A classic example of this particularism is the weighted sentencing for offenses committed between husband and wife: a wife would be punished far more severely for an offense against her husband than he would be for the same offense against her. Complications arose when the legal validity of the marriage was in doubt. In the context of an illegal wife sale, should the seller and the sold wife be treated as husband and wife, or as “unrelated parties”? If a wife sale “broke the bond of moral duty” between husband and wife, at precisely what point in the transaction did that irrevocable break take place? But also, what about the sold wife and her buyer? Did the new de facto marriage created by the transaction merit any legal standing, or should this new couple be considered “unrelated parties” no matter how long they had been together? Extending this logic, what about the networks of kinship relations that enmeshed these individuals, such as between the sold wife and her buyer’s parents? How did fraud affect these calculations? For example, if a buyer had acted in good faith, believing that he was marrying a widow, did he deserve any of the deference or authority normally attached to the role of husband? Third, should poverty be considered a mitigating factor when enforcing the statute against “buying or selling a divorce”? No Qing jurist (to my knowledge) argued that wife sales motivated by poverty should be decriminalized. But did it make sense to enforce the statute rigidly in the name of an abstract absolutist ideal, when doing so would only exacerbate the real problems afflicting the individuals involved? With regard to all three questions, we can detect significant ambivalence. The absolutist position dictated a particular approach, but that approach faced repeated challenges generated by the sheer difficulty of translating ideology into practice in

310

part three

the face of stubborn social facts. Given the intractable nature of the underlying contradiction, a contradiction that was never effectively resolved, these questions expose the friction between ideology and practice in Qing law with unprecedented clarity.

T H E S O L D W I F E A S A D U LT E R E S S

The statute on “buying or selling a divorce” mandated that a sold wife receive the same punishment as the seller and buyer: one hundred blows of the heavy bamboo.3 Under the original, narrow interpretation of this statute that prevailed in the Ming dynasty, the rationale for the wife’s punishment is obvious: the offense was defined as a husband’s sale of his wife to her partner in adultery, meaning that she was guilty of adultery and complicit in her own sale. But, during the Qing, the vast majority of wife sales prosecuted under this law included no hint of prior adultery. What was the rationale for continuing to sentence the wife to a beating under this new, expanded application of the statute? The central judiciary of the Qing dynasty consistently held that a sold wife, in principle, should be considered unchaste regardless of whether adultery had preceded her sale. For example, Shen Zhiqi’s 1715 commentary on the statute argues that the wife’s “consent” (he) to being sold was essentially the same as “consent” to adultery: The statute’s phrase “to take in marriage the wife of another with consent” (he qu ren qi) implies that the man buying the divorce first has a consensual agreement (he tong) with the wife; in other words, the wife herself intends to submit happily to her buyer (yue cong). . . . The term “consent” (he) in the phrase “to take in marriage the wife of another with consent” is like the “consent” in “to engage in illicit sex with a woman’s consent” (he jian), so the primary blame must be placed on the wife.4

Shen Zhiqi’s focus on a wife’s consent underscores her criminal culpability, because, in late imperial law, a woman’s consent to sex had no positive connotation whatsoever, existing only as a category of crime: “consensual illicit sex” (he jian) as opposed to “coerced illicit sex” (qiang jian—usually translated “rape”). Moreover, Qing magistrates treated anything short of vigorous, uncompromising resistance as consent by default.5 Seen in this light, Shen’s analysis implies that the proposal to sell a wife constituted a challenge to her chastity that demanded resistance, and a passive attitude was no better than active agreement. He concludes with a rhetorical question: “The man who buys a divorce craves sex (tan se), while the man who sells a divorce craves money (tan cai); if there were no prior consent of the wife, then why would she deserve the same penalty as they?” In short, the wife’s “prior consent” was a prerequisite for the transaction. At heart, therefore, a sold wife could not have been chaste, even before her husband and her buyer closed the sale.6

Chapter 10. Absolutism versus Pragmatism

311

Board of Punishment Rulings on Sold Wives Who Were Abducted or Raped We find this logic applied by the Board of Punishment in response to inquiries from provincial officials about how to apply the code to tricky cases.7 As I have explained elsewhere, the assessment of penalties for male offenses against female chastity required a prior judgment of the female victim: in the case of rape, for example, only if she were ruled innocent of consent could the rapist be found guilty of coercion. Moreover, it was standard practice to reduce the penalties for such offenses if the female victim had a record of unchaste behavior. Specific measures in the Qing code mandated greatly reduced penalties for the rape or abduction of “a woman who has committed offenses of illicit sex” (fan jian funü), crimes that otherwise warranted the death penalty. In other words, the death penalty for these crimes was as much a measure of the woman’s merit as of the offenders’ guilt.8 In this context, the Board of Punishment held that a wife who had been sold by her husband was the legal equivalent of “a woman who has committed offenses of illicit sex.” Take, for example, an 1824 case from Jiangsu, in which a matchmaker helped one Xi Yuan negotiate the sale of his wife, Xu Shi, to a man named Li Gui. Li reneged on the commission he had promised the matchmaker, so the latter abducted Xu Shi. The question was how to punish the matchmaker and his accomplices for the abduction. The Board of Punishment issued the following ruling: Any wife who willingly lets (ganxin tingcong) her husband sell her divorce is a wife who has lost her chastity (shijie zhi fu). . . . In the present case, Xu Shi’s first husband Xi Yuan suffered from paralytic disease, so he sold her to Li Gui to be his wife. Xu Shi’s divorce was sold with her consent (he tong mai xiu), so by the time she was abducted, she had already lost her chastity.

Therefore, it would be inappropriate to impose on her abductors the full penalties mandated by law, because those applied only to the abduction of “a woman of good character.” Instead, Xu Shi’s abductors “should be sentenced to reduced penalties in analogy to the substatute on ‘abducting a woman who has committed offenses of illicit sex.’ . . . In future, any case of abduction involving a wife whose divorce has previously been purchased should be judged by analogy to the same substatute.”9 As this case shows, the husband’s motive for selling his wife (disease and poverty) made no particular difference to how the woman herself was viewed. What mattered was her own attitude about being sold. Until 1824, it was Board of Punishment policy in abduction cases to equate any “woman who has been sold” (xing fan funü) with a “woman who has previously committed offenses of illicit sex” (fan jian funü), and on that basis to reduce the penalties of her abductors. In that year, however, the governor of Shandong complained that this policy failed to

312

part three

distinguish between the different types of females who were sold, thereby insulting some women and letting off some abductors too lightly. Specifically, he sympathized with “young and innocent daughters” who were sold into servitude by their parents. To be sure, these daughters could not qualify as “chaste wives”—interestingly, the governor assumed that sale inevitably meant loss of chastity—but such girls had little choice in the matter, given their parents’ poverty and the filial duty to obey: Although they come from various circumstances, all have suffered from hunger and cold, with no one to appeal to for aid; the principle [followed by the parents] in every instance has been to sacrifice one daughter’s chastity in order to save an entire family. In such cases, being sold is the result of orders from parents or parents-in-law, and simply cannot be helped. If we indiscriminately attach the label of “illicit sex offender” to anyone who has been sold regardless of circumstances, and for that reason spare her abductors the heavy penalties of beheading and strangulation, then we shall have failed to promote public morality and to inflict on criminals the punishments they deserve.

Moreover, unlike wife sales, Qing law usually permitted the sale of daughters by parents. Therefore, the governor urged a small but symbolically important modification of judicial policy: those who abducted a “woman who has been sold” should continue to receive penalties lighter than those for abducting a “wife of good character” but more severe than those for abducting a “woman who has previously committed offenses of illicit sex.” The Board of Punishment accepted his proposal, which was codified in a new substatute.10 But this limited gesture did not extend to wives whose “divorce was sold” by their husbands. In 1828, for example, the Huguang Bureau of the Board of Punishment confirmed the long-standing principle that a wife who “consented” to being sold should be considered an adulteress for purposes of sentencing, if an offense had been committed against her. In the case under consideration, a man named Chen Chaozhen had “bought the divorce” of a woman named Xia Shi from her first husband; a third man, who knew of the situation, organized a gang to abduct her and sell her for a profit. The provincial officials reviewing the case had become confused. Should Xia Shi be judged as a “woman who has been sold,” according to the sympathetic standard proposed by the governor of Shandong in 1824? Or, since her husband had sold her, should she be considered the equivalent of a “woman who has previously committed offenses of illicit sex,” which would imply even lighter penalties for her abductors? Their confusion may have stemmed, in part, from linguistic ambiguity: the term “funü,” often used in legal contexts to refer to women generally, is a compound of the characters for “wife” (fu) and “daughter” (nü). The governor of Shandong’s sympathy had focused on daughters sold by their parents, whereas the present case involved a wife sold by her husband.

Chapter 10. Absolutism versus Pragmatism

313

The board took this opportunity to reaffirm the orthodox standard: “For women, a reputation for chastity is more important than anything else. If a wife is perfectly willing to have her divorce sold—to be remarried in spite of already having a husband—then this is in no way different from engaging in illicit sex.” In abduction cases, it should be a simple matter to discover whether the woman in question “has previously engaged in illicit sex or is a wife purchased by divorce” and, if so, to treat her according to the traditional standard. In the present case, the men who abducted Xia Shi should receive the lightest available penalties, in analogy to the substatute on “abducting a woman who has previously committed offenses of illicit sex.”11 The consistent theme of this hairsplitting is to imagine the sold wife as an active agent with choices; it treats the proposal to sell her as a challenge to her chastity, and her “consent” to be sold as a failure to meet that challenge. This logic ties the judiciary’s theory of wife-selling to its construction of crimes against chastity (such as rape, sexual proposition, and forced remarriage of widows) as well as to adultery “abetted” by the husband. In theory, a wife sold by her husband was an active accomplice in a crime that violated the norms of marriage in essentially the same way as more blatantly sexual offenses. Wife Sale and the Canonization of Chastity Martyrs Aside from the judicial system, the most important institution for promoting female chastity was the canonization system administered by the Board of Rites. Here, if anywhere, one would expect to find enshrined an absolutist, uncompromising standard of female virtue. How, then, was wife-selling treated by the official system for canonizing chastity heroines that expanded so dramatically during the High Qing? As is well known, from the Yongzheng reign on, many thousands of widows were honored for staying single and maintaining an impeccable reputation for chastity for a stipulated number of years following their husbands’ deaths. But the most dramatic honors were reserved for the far smaller number of women who gave their lives as martyrs, including women who had died resisting rape or (in the case of widows) in-laws’ pressure to remarry, or had committed suicide in outraged reaction to a sexual proposition. The theory behind canonization was that these heroines had chosen death over the far easier path of sacrificing their chastity.12 If the proposal to sell a wife constituted a challenge to her chastity that demanded militant resistance, then it would seem to follow that a wife who protested such a proposal and resisted unto death should be canonized, just as wives who died resisting pressure from their husbands or in-laws to engage in prostitution were eligible for this honor.13 We find an example of such canonization in a case from Hengyang County, Hunan, memorialized in 1774, in which a peasant woman, Chang Shi (34), died

314

part three

while fiercely resisting an attempt to sell her in marriage. Her brothers-in-law forced her husband, Nie Yide, to sell her in order to compensate them for their losses after she accidentally started a fire that burned down their homes. But Chang Shi refused to cooperate. She first took refuge at her natal home, and, when fetched back against her will, she jumped in a lake in an apparent suicide attempt. Upon being brought home, she refused to eat and insisted that she would rather starve to death than remarry. Her husband and brothers-in-law had to watch her day and night in order to prevent her from escaping or attempting suicide. Finally, they bound and gagged her so that they could get some sleep, and later they found that she had choked to death. Ruling Chang Shi’s death a homicide, the magistrate placed the blame on Nie Yide’s older brothers (reasoning that he had had no choice but to obey them) and sentenced them for “killing a younger brother’s wife in a fight”; Yide himself received no penalty. The prospective buyer, matchmaker, and others complicit in the transaction were all sentenced to beatings according to the statute against “buying or selling a divorce,” and the brideprice was confiscated from the buyer (even though he had not yet paid it). As to Chang Shi, the magistrate recommended canonization as a chastity martyr: Nie Chang Shi was merely an ignorant village wife (xiangcun yu fu), nevertheless a great righteous duty was clear in her heart (xin ming dayi), and when forced to remarry she refused to obey, so that she was bound and ended up dying. Her death was truly a chaste martyrdom worthy of praise (jie lie ke jia)! Therefore, it is appropriate to request that she be canonized, in order to promote the public morality.

The magistrate’s superiors, including the emperor, approved this request upon review. The formal basis for Chang Shi’s canonization was analogy to the canonization of a widow who died resisting forced remarriage.14 The canonization of Chang Shi makes perfect sense, if one assumes that the proposal to sell a wife constituted a challenge to her chastity comparable to a rape attempt or sexual proposition. Curiously, however, I have found no other example of a woman being canonized for dying while resisting wife sale. On the contrary, by the early nineteenth century, canonization had been explicitly ruled out in this context. An 1807 ruling by the Board of Punishment set a key precedent, which is recorded in The Conspectus of Penal Cases. This ruling concerned a Jiangsu case in which one Chen Shangde had strangled his wife, Li Shi. The background to the murder was that Chen had engaged Liu Biao behind his wife’s back to act as matchmaker and find a buyer for her. Chen told Li Shi that he had found an employer for her (to do domestic work) and escorted her to Liu Biao’s home, where he planned to place her until a buyer could be found. Upon arrival, Liu announced to her their real plan and urged her to cooperate, but when Li Shi realized what the men had in mind, she cursed them both, threatening to report them to the yamen, and fled;

Chapter 10. Absolutism versus Pragmatism

315

Chen and Liu caught her and, because Liu feared prosecution, he persuaded Chen to kill her. In judging this case, the Board of Punishment denied a request for Li Shi’s canonization because, according to the board, “it would be inappropriate to request canonization” for a wife murdered by her husband for resisting sale, because “her death was not the result of sexual proposition or [attempted] rape.” This ruling represents a dramatic change from the policy applied in canonizing Nie Chang Shi in 1774.15 The new policy is confirmed by several later judgments. For example, in an 1830 case from Hunan, Huang Dexiu “used force to sell in marriage” his wife, Chen Shi, provoking her to hang herself. Similarly, in an 1831 case from Anhui, Cao Yushu sold his wife in marriage to Wang Chaofu, who “used force to take her home and consummate the marriage,” after which she committed suicide “out of shame and indignation.” Both husbands were sentenced to one hundred blows of the heavy bamboo and three years of penal servitude, by analogy to the substatute on “a widow committing suicide because her husband’s family forces her to remarry, but she is unwilling to lose her chastity.”16 This same substatute had been cited by analogy in order to justify the canonization of Chang Shi in 1774, but, in these two cases, there is no mention of canonization. The shift in policy appears to reflect what many other sources also show: a growing recognition that it was unrealistic to equate poverty-driven wife-selling with “illicit sex” in the full ideological and ritual implications of that criminal category. Yet this evidence also highlights the impossible double bind that High Qing orthodoxy imposed on women by obliging them to obey their husbands and yet also to conform to an absolute standard of chastity. Women who “consented” to be sold, in obedience to husbands and in-laws, were considered unchaste; this insult was compounded by the imperial state’s refusal to credit as chastity martyrs those women who were murdered for refusing to be sold. Should a Sold Wife’s Beating Actually Be Carried Out? The Qing code provided that, as a matter of routine, female offenders sentenced to beatings should be allowed to “redeem” those penalties by paying cash fines. But the code made an exception for “women who have committed illicit sex”: their beatings should indeed be “carried out” (di jue), although sentences to the cangue might still be redeemed.17 In central case records, sold wives who had committed adultery before being sold always had their beatings carried out. But what about a wife whose sale was motivated solely by poverty? As far as I can tell, the central judiciary never provided any guidance on this question, and central case records show that, in practice, local magistrates had leeway in how they chose to interpret these particular rules. In most cases, magistrates recommended that these women be allowed to redeem their beatings, “as provided

316

part three

by law,” on the grounds that they were women. These magistrates apparently did not consider consenting to be sold to be the legal equivalent of adultery. For example, in a Shanxi case from 1757, the magistrate cited the statute against “buying or selling a divorce” to sentence the seller, matchmaker, scribe, and wife to the mandated beatings (the buyer had been killed), but he allowed the wife to redeem her beating with a fine, explaining that “her crime is different from actually committing illicit sex” (yu shizai fan jian you jian). The men’s beatings were all carried out.18 However, in about 20 percent of my sample of relevant xingke tiben, more doctrinaire magistrates recommended that the beatings be carried out, because in their view any wife who agreed to be sold should be considered “a wife who has lost her chastity” (shijie zhi fu). As the magistrate of Chengbu County, Hunan, explained in an 1825 ruling, “Hu Shi is a wife who lost her chastity because she allowed her divorce to be sold (ting cong mai xiu shijie zhi fu),” and therefore he ordered her beating carried out.19 Similar rhetoric appears in other cases where magistrates ordered sold wives beaten. Of course, these rulings were simply recommendations subject to higher review, but, as far as I can tell, no magistrate was ever overruled on this point, regardless of whether he opted for lenience or severity. The fact that the central judiciary never bothered to clarify whether a sold wife should actually be beaten— in contrast with its constant nitpicking over so many other seemingly trivial matters—suggests the difficulty of resolving the contradiction between absolutist chastity ideology and the social reality of poverty-driven wife-selling. To conclude this section, we should note a striking contrast here with the judicial attitude toward polyandry, marital prostitution, and other survival strategies covered by the crime of a husband “abetting or tolerating” a wife’s “illicit sexual relations”—zong jian for short. As far as I know, no Qing jurist ever expressed any sympathy for couples driven to these strategies by poverty, or any doubt whether such wives constituted genuine adulteresses or deserved to be punished fully as such. As a corollary, there was no question that a wife who died resisting pressure to engage in polyandry or prostitution merited commemoration as a chastity martyr: one can find many cases of such canonization recorded in both xingke tiben and casebooks such as The Conspectus. In the eighteenth century, the Board of Rites amended its rules for canonization specifically to include such cases.20 I M PAC T O F W I F E S A L E O N T H E L E G A L STA N D I N G O F FA M I LY R E L AT IO N S

A basic principle of the law against “buying or selling a divorce” was that for a husband to sell his wife would irrevocably “break the bond of moral duty between them” (yi jue). Therefore, the wife should be “divorced and returned to her natal lineage” (liyi gui zong). An important corollary was that, subsequent to the wife

Chapter 10. Absolutism versus Pragmatism

317

sale, the first husband (i.e., the seller) and his ex-wife should be “treated as unrelated parties” for purposes of law (tong fan lun). This corollary guided sentencing when one ex-spouse committed a violent crime against the other at some point after the sale: the most common scenario found in case records is that the first husband (seller) would kill his ex-wife. A typical example is a case from Ruyang County, Henan, in which a hired laborer named Sun Shui (31) sold his wife, Li Shi (30), because of poverty. The couple lived with Sun’s unmarried brother and widowed father, and Li Shi was the only female in the household. At first she was unwilling to be sold, but he pressured her, and she finally agreed, but she was very bitter. A friend of Sun’s acted as matchmaker (for free) and arranged Li Shi’s sale to a fellow villager, Li Wenyuan (50), for a brideprice of 8000 cash. After the sale, Sun left the village and worked elsewhere for a time, returning home in the summer of 1787. One day he happened to encounter his former wife when she was out in a field drying wheat. He called out to her, but she ignored him, and when he reproached her for her lack of “feeling” (qing), she retorted (according to his testimony), “since I had already sold her, the bond between us was broken, so what ‘feeling’ was left? (haiyou shenme qingfen),” and then she began cursing him. Sun became enraged and stabbed her to death. Li Shi’s assessment of her post-sale relationship with her ex-husband anticipated that of the magistrate who judged this homicide. He sentenced the buyer and matchmaker according to the statute against “buying or selling a divorce” and ordered confiscation of the brideprice. With regard to the homicide itself, “the evidence shows that Sun Shui had sold his wife Li Shi, so their moral bond was already broken, and he should be sentenced as an unrelated party.” Sun Shui was sentenced to beheading after the assizes according to the ordinary statute against “purposeful homicide.” Subsequent review confirmed these sentences. Had Sun still qualified as Li Shi’s legal husband, he would have received the lighter sentence of strangulation after the assizes, according to the statute against “a husband purposely killing his wife.”21 Much of the hairsplitting of the central judiciary traced here involved slight adjustments on a carefully calibrated scale of death penalties (see Appendix D.1). There were three official ways to be put to death: strangulation (jiao), beheading (zhan), and dismemberment (lingchi)—severity being determined by the degree of damage to the body of the condemned, rather than the amount of pain suffered (thus, strangulation was considered least severe). Sentences of strangulation and beheading might be delayed until “after the Autumn Assizes” (qiu hou chujue) or “executed immediately” (li jue). It appears that, in practice, most of the delayed sentences ended up with some sort of reprieve (see Appendix D.3). Dismemberment, however, was always immediate.22 The principle that wife sale broke the moral bond between spouses applied even if the sold wife later returned to her first husband. For example, in an 1819 case from

318

part three

Zhenfeng Department, Guizhou, an immigrant porter named Jiang Rugui (51) sold his wife, Li Shi (23), out of poverty for a “body price” of 10 taels. After a month, however, Jiang had already used up this sum, so he visited the buyer, Li Laoba, and demanded a supplementary payment; when Li refused, Jiang threatened to file charges against him for “taking in marriage a woman with a living husband.” Intimidated, Li opted to burn the wife sale contract and to return Li Shi to Jiang without even demanding a refund of her original body price. Jiang then tried to find a new buyer, but word had gotten around about his treatment of Li Laoba, and no one else was willing to risk similar extortion. All the while, Li Shi deeply resented Jiang—first for having sold her, second for having “cheated to get her back,” and third for abusing her. Eventually, she conspired with a neighbor widow to murder him with rat poison. According to the magistrate’s ruling, “evidence shows that Li Shi first had her divorce sold for money by Jiang Rugui, so according to statute she should have been divorced and returned to her natal lineage. Now she has used poison to murder the husband who sold her divorce, and so according to law she should be sentenced as an unrelated party.” Li Shi was sentenced to beheading after the assizes, according to the ordinary statute on “planned murder” of an unrelated party; had she qualified as Jiang’s legal wife, she would have received the far more severe sentence of death by dismemberment. Subsequent review confirmed this judgment. The buyer and matchmaker both were sentenced according to the statute against “buying or selling a divorce.”23 We find a curious variation on this reasoning in a 1797 case from Yishui County, Shandong. Xin Da (46) had sold his wife, Zhang Shi (33), in a fraudulent transaction to Ma Desheng (44), who believed he was marrying a widow, for a brideprice of 15,000 cash. The couple had agreed to this course of action because of poverty, but they apparently had a good relationship, because after two months Xin secretly contacted Zhang Shi and persuaded her to run away from her new husband and rejoin him (an example of “falconing”). They were caught, however, and Ma Desheng’s neighbors beat Xin to death. When this incident came before the magistrate, he ruled that, since Xin Da had sold Zhang Shi’s divorce, their moral bond was already broken at the time she ran away with him. Therefore, for her to run away with Xin Da counted as a simple case of “abduction,” in which she was complicit. She was sentenced to one hundred blows of the heavy bamboo and three years of penal servitude, according to the statute against “abduction, in which the abductee consents and has knowledge of the situation” (he you zhi qing); the nominal harshness of this sentence was mitigated, however, because she was allowed to redeem both penalties with cash fines.24 At What Point Was the Marital Bond Broken? The same basic principle guided all of these judgments: as the Shandong Bureau of the Board of Punishment confirmed in 1824, if a wife or concubine “has been

Chapter 10. Absolutism versus Pragmatism

319

divorced” (bei chu, literally “expelled”) by her husband, and if she remarries to some other man, then this sequence of events indicates that she and her first husband “both are willing to separate” (liang yuan liyi). Note that this language emphasizes mutuality and the wife’s own complicity. Once that sequence is complete, “her moral bond with her husband is already broken,” and for legal purposes they should be considered unrelated parties.25 But what if that entire sequence had not been completed? At what point in the process of separation and remarriage did this rupture actually take effect? In this respect, was wife-selling different from other scenarios of divorce and remarriage? The Shandong case under consideration in 1824 was typical of the sort found in The Conspectus of Penal Cases, in which tricky ambiguities or contradictions forced central jurists to engage in an exquisite parsing of judicial principles. This case did not involve a wife sale but rather an unusual example of divorce without compensation; nevertheless, the same principles applied. The problem in this case was that the couple in question did not agree, and the wife, though expelled, had not remarried. Du Wenyi had caught his wife, Sun Shi, in adultery and, renouncing her, had sent her back to her natal family without compensation. But she refused to remarry, and her natal family was very poor, so she returned to Du’s home and begged him to relent. He refused, they quarreled, she cursed his ancestors, and so he killed her. The question for the Shandong Bureau was whether, at the time of the homicide, they counted as a married couple or as “unrelated parties.” The bureau reasoned that, since Du had sent his wife back to her natal family, as far as he was concerned “the moral bond was already broken,” but since Sun Shi had refused to remarry, and sought to return to Du, their situation was different from that of a couple who “both are willing to separate.” “In fact, she still had tender regard for her husband, and could not bear this abrupt breaking of their relationship” (shi you juanlian buren jujue zhi xin). Therefore, “her sense of their moral bond had not yet been broken,” her adultery notwithstanding. Therefore, it would be inappropriate to treat Du as an “unrelated party” and sentence him to the full penalty of beheading after the assizes mandated by the basic statute against “purposeful killing.”26 The code mandated that a husband who caught his wife in the act of adultery and immediately killed her on the spot, in the heat of righteous indignation, should be released without punishment.27 But the bureau ruled that, since Du had first expelled his wife and later killed her only after some time had passed, this law should not apply; neither should the substatute that mandated penal servitude for a husband who did not catch his wife in the act of adultery but killed her only after hearing about her adultery “some days later” from a reliable source.28 Since Du’s reaction to Sun Shi’s adultery had been to expel her, he had lost that excuse for killing her. Therefore, what was needed was some way to balance the fact of her expulsion against the fact of her continuing tie to her husband, reflected in her desire to reunite with him.

320

part three

In a flourish of creativity, the Shandong Bureau observed that for Sun Shi to curse Du’s ancestors itself constituted a crime: “For this reason, in fact, she should be considered a criminal.” Therefore, it was possible to sentence Du according to the statute against “unauthorized killing of a criminal already in custody”; the mandated penalty of strangulation after the assizes was lighter than the beheading mandated for “purposeful homicide” of an unrelated party, but far heavier than that for a husband who had killed a wife because of her adultery. In this way, the bureau achieved the desired balance.29 The Sold Wife and Her Buyer as “Unrelated Parties” If a completed wife sale terminated the legal marital relationship between a woman and the husband who sold her, then what about the new relationship created between her and her buyer? Most of the evidence indicates that they, too, were considered “unrelated parties” for purposes of law. A 1777 case from Zhangjiakou Subprefecture, Zhili, illustrates the typical sentencing policy. Wu Jinbao, a landless migrant who had fallen ill, arranged to sell his wife, Shi Shi, through a friend, Guo Chang, who acted as matchmaker. Guo found a single peasant migrant from Shanxi named Zhang Haiyu who was looking for a wife. Zhang, who lived and farmed alone, agreed to pay a body price of 5 taels, plus 4 more taels to be shared by the matchmaker and the scribe who wrote the contract. After the sale, however, Shi Shi was very unhappy. She complained bitterly about Zhang’s poverty, and he in turn would complain that she was lazy and a poor cook, and would beat her. After about a month, these quarrels reached a climax when she attempted to leave him, and he beat her to death. The magistrate found that “Shi Shi was originally the wife of Wu Jinbao, and Zhang Haiyu used money to buy her divorce, so according to statute she should have been returned to her natal lineage. Now that Zhang Haiyu has beaten Shi Shi to death, he should be sentenced as an unrelated party.” Zhang was sentenced according to the ordinary statute against “homicide in a fight”; the other parties all were sentenced according to the statute against “buying or selling a divorce,” and the body price and matchmaker and scribe’s fees were confiscated.30 But marriage united more than just two individuals: it also united their families, so that each spouse acquired a specific legal status relative to each member of the other spouse’s family. The Ming and Qing codes used the official mourning system to map the exact relationship between kin in order to determine the sentences for crimes that occurred within the family (see Appendix D.2). A series of charts at the beginning of each code showed the degrees of mourning that an individual theoretically owed each family member if that person were to die. On each chart, the individual in question appears at the center of a grid made up of a concentric array of squares, with each square representing a hypothetical relative (parents appear one square above, children one square below, siblings one square to

Chapter 10. Absolutism versus Pragmatism

321

the side, etc.). In this way, every dyadic relationship within the extended family could be mapped according to generation, birth order, lateral distance (such as between cousins of different degrees), and gender. Once the relationship between offender and victim was pinpointed, the correct sentence could be determined.31 The main chart maps an individual (presumed male) in relation to his own natal lineage, but subsidiary charts cover all possible variations, such as a wife’s relationship to her husband’s family, a wife’s relationship to her own natal family, a husband’s relationship to his in-laws, and so on. In other words, particularism extended in both directions to cover the entire network of kin created by the marriage of two individuals. As one might expect, therefore, an illegal marriage such as a wife sale affected sentencing not only for offenses between the spouses themselves but also between them and their putative in-laws. How were such cases judged? A key factor was whether the relative in question had been complicit in the wife sale, or had known of the marriage’s illegality. For example, in a 1780 case from Pingli County, Shaanxi, the woman Wang Shi’s father, Wang Chengwan, found out that her husband, Xiong Shiming, had sold her in marriage only after the sale had taken place. (The sale had been transacted without the knowledge of anyone in her natal family.) In anger, Wang attacked his son-in-law, who then killed him. Since Wang had not been complicit in the sale, the magistrate ruled that Xiong should be judged according to their relationship within the mourning system—that is, as Wang’s son-in-law. This decision had the effect of increasing the severity of Xiong’s penalty because of the deference that a son-in-law owed his wife’s father—a result that might be interpreted as an indirect reward for Wang’s righteous anger.32 In contrast, in an 1810 case from Rong County, Guangxi, Huang Anhao had enthusiastically cooperated with the sale of his daughter by her husband, Lai Fengchi, in exchange for a share of her brideprice. After the sale was completed, Huang greedily demanded even more money, provoking a fight in which Lai killed him. In this case, the magistrate ruled that Huang’s complicity in the illegal wife sale demonstrated without a doubt that “the moral bond between father-in-law and son-in-law had already been broken” (wengxu zhi yi yi jue) prior to the killing, and that Lai should therefore be sentenced as an “unrelated party.”33 We find an interesting corollary in an 1825 case from Huan County, Gansu, in which Li Chengfang resented the poverty of his son-in-law, Chang Erhu, and pitied his daughter, Li Shi, for her suffering, so he tried to force Chang to sell her in marriage. But Chang refused to do so, the two men fought, and Chang ended up killing Li. The magistrate ruled that even though no wife sale had actually taken place, the fact that Li Chengfang had pressured Chang to sell Li Shi meant that the “moral bond” between father-in-law and son-in-law was already broken.34 These two cases demonstrate that the “moral bond of duty” between husband and wife extended to relations between the spouses and their respective parents-in-law, and those bonds

322

part three

too could be “broken” by complicity in an illegal wife sale. In both, the decision to sentence the son-in-law as an “unrelated party” had the effect of reducing the severity of his penalty—a result that might be interpreted as indirect punishment of the father-in-law for his criminal behavior. An interesting variation on this principle can be seen in an 1804 case from Lichuan County, Hubei, that involved the sale of an adopted daughter-in-law, Liu Nü. Before her marriage could be consummated, her father-in-law, Lei Wuwen, sold her because of poverty to another man, Li Shuzhao, who wanted an adopted daughter-in-law to marry his wife’s son from a previous marriage. The sale was negotiated without notifying the girl’s natal family. Unfortunately, Liu Nü did not get along with her new in-laws, and Li Shuzhao often beat her for alleged laziness. Finally, one day he beat her to death and secretly buried her body. After a year, however, her father learned she had been sold, tracked down the Li family, found out what had happened, and reported her murder to the authorities. In this case, the magistrate ruled that the sale of an adopted daughter-in-law counted as the crime of “buying or selling a divorce,” although since both “husbands” were minors and neither marriage had been consummated, their fathers bore the criminal liability in their place. The magistrate found that: At first, Liu Nü was the adopted daughter-in-law of Lei Wuwen. Lei Wuwen sold her divorce for cash, and Li Shuzhao knowingly bought her to be his daughter-in-law. According to the relevant statute she should have been divorced [and returned to her natal lineage]. Now, because she failed to obey instructions, she has been beaten to death. Li Shuzhao should be sentenced as an “unrelated party.”

This ruling had the effect of increasing the severity of Li Shuzhao’s penalty; the magistrate sentenced him to strangulation after the assizes, by applying the ordinary statute against “homicide in a fight.” The magistrate also sentenced Lei Wuwen for “selling a divorce,” and ordered confiscation of the brideprice.35

G RU D G I N G R E C O G N I T IO N O F M A R R IAG E S C R E AT E D BY WIFE SALE

Most of the judgments found in central cases conform to the standard illustrated by the above examples, namely that an illegal wife sale would invalidate both the marriage that preceded it and the new one created by it. Nevertheless, we can also find judgments that suggest some difference of opinion among senior jurists—or at least that there was no absolutely binding rule. These judgments reflect an uncomfortable recognition by some jurists that wife sales did create de facto marriages, if not de jure ones. A salient example is a 1753 case from Jinyun County, Zhejiang, in which famine refugee Wang Yongsheng sold his wife, Ge Shi (39), and their infant daughter to

Chapter 10. Absolutism versus Pragmatism

323

peasant Shi Yiren (41), whose first wife had died. The little family was desperate, and Wang, who was seriously ill, agreed to accept the very low price of 2 taels to secure the survival of his wife and daughter. Shortly thereafter, Wang died of his illness. Unfortunately, Ge Shi soon realized that her new husband was nearly as poor as her last had been, and she was bitterly unhappy. They quarreled incessantly, and finally she burned the wife sale contract, announced that she planned to return to her natal home, and threatened to file charges against Shi Yiren if he tried to stop her. This threat provoked Shi to murder her. He tried to disguise her murder as suicide by hanging, but the forensic examination produced evidence of poison, and Shi confessed. The magistrate determined that the transaction constituted an illegal “divorce sale,” and he sentenced the matchmaker according to the relevant statute (the seller was already dead). But in a contradictory move, he decided to sentence Shi Yiren according to the statute against “a husband purposely killing his wife,” reasoning as follows: Shi Yiren “took in marriage the wife of another with consent,” and that is not a legitimate way to marry (fei hunyin zhi zheng). Nevertheless, Shi Yiren had no prior relationship with Wang Yongsheng and his wife, and there was no adultery prompting Shi Yiren to plot to take her in marriage. On the contrary, Wang Yongsheng was poor and ill and he voluntarily sold his wife. They openly drew up a marriage contract, and a matchmaker negotiated the terms. In addition, Ge Shi took her little daughter with her when she entered Shi’s household. Clearly, Ge Shi’s moral bond of marriage with her first husband was already broken, and with Shi Yiren she had already acquired the status of husband and wife. Therefore, now that this crime has been committed [i.e., Shi’s murder of Ge Shi], they should be treated as husband and wife for purposes of sentencing.

Here, the magistrate argued that the formal dictates of codified law must be set against the practical circumstances of a given case. In effect, the impropriety of wife sale was less important than the fact that this couple had been living together as husband and wife, and, for purposes of sentencing, it made no sense to pretend they were not husband and wife. In the magistrate’s words, the couple “had already acquired the status of husband and wife (fufu mingfen)”—and, as we shall see, this de facto status would become a key criterion in future deliberations on this question. This argument represents a major concession to the social reality of wifeselling as a form of marriage. Moreover, the magistrate’s emphasis on the lack of prior adultery, and on the poverty and voluntary consent of the seller, suggest a deep ambivalence about whether to classify this sort of routine wife sale as a form of “illicit sex” or to punish it as “buying or selling a divorce.” Remarkably, the magistrate’s decision to recognize Shi Yiren as Ge Shi’s husband—which guaranteed him a lighter sentence—was confirmed without comment by his superiors.36

324

part three

This report from xingke tiben is the only one I have seen in which this exception was made. Nevertheless, there are indications that other senior jurists shared this magistrate’s point of view to at least some degree, and like him they wrestled with how to square abstract legal principles with the untidy realities of life among the poor. For example, in a 1777 case from Li County, Zhili, Xia Hei’er had sold his wife, Zhao Shi, to a landless peasant named Li Bao’er for a “body price” of 5000 cash; after a couple of years, Li, who made a living as an agricultural laborer, left the village and worked elsewhere for seven months. During Li’s absence, Zhao Shi ran out of resources, fell ill, and became quite desperate. Fortunately, a friend named Xu Erbo (also an agricultural laborer) provided her with food and medicine, and after she recovered she began sleeping with Xu in exchange for his continuing support. When Li returned home, he was ill and broke, and he agreed to allow their sexual relationship to continue in exchange for Xu’s help. Over time, however, Li became abusive, demanding money and beating Zhao Shi and threatening to kill her whenever Xu ran short of funds. Eventually, Zhao Shi and Xu decided they had had enough of this treatment, so they murdered Li. In judging this case, the magistrate acknowledged that, in principle, Li Bao’er and Zhao Shi should be considered unrelated parties for purposes of sentencing. But he bridled at the distortion of social reality that such a rigid application of principle would imply, given that this couple had lived as husband and wife for several years. “Even though Li Bao’er cannot be considered a ‘genuine husband’ for purposes of law (bude yu ben fu tong ke), still, in the end, there is a difference between his status and that of an ‘unrelated party’ (dan jiu yu fanren you bie).” Were Zhao Shi sentenced as a wife who committed the “planned murder” of her husband, she would receive the extreme penalty of death by dismemberment; if, however, she were sentenced as an unrelated party, she would receive the considerably lighter sentence of beheading after the assizes, leaving open the possibility of a reprieve. Seeking a compromise between these two extremes, the magistrate sentenced her to beheading but requested a special edict to authorize immediate execution of the sentence, without review at the Autumn Assizes. This penalty reflected the magistrate’s sense that for Zhao Shi to kill Li was worse than killing an “unrelated person” but not quite as bad as killing a legal husband. This judgment received approval at all levels.37 The 1787 Edict on the Legal Standing of Illegal Marriages Wife-selling was not the only kind of illegal marriage that caused these kinds of headaches for the judiciary. In 1787, the Qianlong emperor attempted to provide guidance for how such cases should be resolved. The Board of Punishment had asked the emperor how it should judge a homicide case involving a Han male and his Mongolian wife (the Qing dynasty prohibited intermarriage across this juridi-

Chapter 10. Absolutism versus Pragmatism

325

cal ethnic boundary), and whether a new substatute should be issued to cover this scenario. The emperor replied that, since the marriage had already taken place some time ago, obviously “their status as husband and wife was already established,” so it would not be appropriate to treat them simply as “unrelated persons” for purposes of sentencing. The emperor added (with apparent exasperation) that in the future “such matters may be judged on a case by case basis,” and that there should be no need to add new substatutes to cover every specific scenario.38 In other words, a couple’s de facto status might outweigh their de jure status, at least for sentencing purposes. This concession to unseemly social practice suggested that an illegal union might confer at least part of the obligations and privileges that attended legally recognized marriage. In 1789, the Guizhou Bureau of the Board of Punishment reflected on the emperor’s edict when considering the case of a widow, Yue Shi, who had remarried during the official three-year mourning period for her first husband (which was prohibited, although very common)39 and subsequently was murdered by her new husband, Zhang Wenmou. Summarizing the edict’s commonsensical position, the board reasoned as follows: To be frank, we cannot expect foolish peasants in remote regions to know exactly what is forbidden by law. There are a great many cases in which the prohibition against remarriage during the mourning system and similar provisions are violated. Now, just because some foolish peasants violate such prohibitions, we cannot ignore codified legal standards, but neither should we casually ignore a couple’s status as husband and wife just because of relatively minor violations of these marriage prohibitions. It is vital for magistrates to act on this edict’s intent by considering each case’s specific circumstances (sui an zhenzhuo), by weighing the facts and degree of seriousness, in order to judge and sentence in an appropriate way.

In the case under consideration, the board clearly sympathized with the widow’s second husband, Zhang Wenmou; he had left her on her own while he worked in another village, and after only a month, when she ran short on resources, she had her uncle arrange a new marriage for her to yet another man (without bothering to inform this third husband about her marriage to Zhang Wenmou). When Zhang found out what she had done, he killed her. The board reasoned that, despite the illegal timing of Yue Shi’s remarriage to Zhang, it had otherwise been properly arranged through a matchmaker, with her dead husband’s brother presiding, and they had been married for over a year before the homicide. Therefore, in fact “their status as husband and wife had already been established,” which was the standard set by the 1787 edict; their situation was different from those couples “who have illicitly joined together in marriage on their own accord, without matchmaker or family elders playing any role, and whose divorce is therefore mandated by law.” On this basis, the board

326

part three

sentenced Zhang according to the statute against “a husband killing his wife,” which mandated a lighter penalty than that for an unrelated party. The sentence was approved by imperial edict.40 The 1787 edict and the 1789 judgment that invoked it seemed to provide a broad mandate for magisterial discretion in weighing the particular circumstances of each case to decide whether the status of husband and wife had actually been “established.” But they specifically mentioned only interethnic marriage and widow remarriage during mourning. What about other prohibited but widespread marriage practices? What about wife sales? The 1808 Substatute In 1808, the Board of Punishment deliberated on this question, at the request of the governor-general of Sichuan, who was wrestling with a homicide case complicated by an illegal marriage between offender and victim. Following the logic of the edict and ruling described above, the board attempted to distinguish between different kinds of illegal marriage that involved greater or lesser violations of ritual propriety. The board’s purpose was to decide when to treat offender and victim as “unrelated persons” and when to treat them as kin within the mourning system for purposes of sentencing: In all cases of illegal marriage, the couple should in principle be divorced, but in fact the circumstances of such marriages vary. For example, in cases of illicit sex followed by marriage, or of knowingly buying a divorce (zhiqing mai xiu), then according to statute the couple should be divorced. Since these kinds of marriage violate ritual propriety, the couple cannot be called “husband and wife” even if matchmaker and marriage contract have been used. Therefore, [in these kinds of illegal marriage,] when homicide or injury occurs between the wife and the husband or his relatives, the offenders should be sentenced as if they are unrelated parties. But if the illegal marriage constitutes a less serious violation of propriety—for example marriage between [unrelated] individuals of the same surname, or between relatives by marriage [i.e., with different surnames] of different generations, or between individuals of commoner and debased statuses; or widow remarriage during the mourning period, or taking a concubine despite having a wife; or a husband selling his wife fraudulently in marriage, so that the man who took her in marriage does not know the facts of her situation [i.e., that she has a living husband]—in less serious violations such as these, even though by law the couple should be divorced, nevertheless as long as the marriage was transacted in the proper way, openly and through a matchmaker (ming mei zheng qu), and the marriage rituals have been fulfilled, so that the status of husband and wife is already established, then it would be inappropriate to ignore their [de facto] status as husband and wife (zhi fufu zhi mingfen yu bu lun) simply because of such relatively minor violations of the laws on marriage. Therefore, in these sorts of cases, if there is some offense between relatives, then the offender should be sentenced according to position in the mourning system

Chapter 10. Absolutism versus Pragmatism

327

relative to the victim, taking into account the marriage relationship that has already been established.41

The kinds of illegal marriage listed in the second clause of the board’s ruling as “less serious violations of ritual propriety” include at least two that were widespread and probably not considered taboo by most people: marriage between unrelated persons of the same surname, and widow remarriage during mourning. Many jurists considered the prohibitions of these practices to be obsolete or, at best, unrealistic, and they were rarely enforced. Fraudulent wife sale was different, in that it remained a serious crime and was indeed punished. The reason it is included here is that the buyer did not know he was marrying “a wife with a living husband.” Therefore, the illegality of the transaction should not be held against him, and he should be credited with the role of husband relative to a wife he had taken in good faith. A substatute issued in the same year formally codified the board’s views: In cases of illegal marriage (jia qu wei lü), in which [according to law] the woman should be divorced, if an offense is committed between her and her husband or her husband’s relatives, then the offender(s) should be sentenced as unrelated parties in the following scenarios: (a) if the husband took her in marriage after first engaging in illicit sex with her; (b) if the couple joined together illicitly, of their own accord; or (c) if the husband knowingly bought her divorce. This rule shall apply even if the illegal marriage was arranged by a matchmaker using a marriage contract. However, if a case involves nothing more serious than marriage between unrelated individuals of the same surname, or between relatives by marriage from different generations, or between individuals of commoner and debased statuses; or marriage during the mourning period, or taking a concubine despite having a wife; or a husband selling his wife fraudulently in marriage, so that the man who took her in marriage does not know the facts of her situation, and he truly took her in marriage in the proper way, openly and through a matchmaker; then in all of these scenarios, even though according to statute the man and woman should be divorced, if there is an offense between them it should be judged according to their relative positions within the mourning system [i.e., as if they were legally married].42

To summarize, with regard to wife-selling, the first clause of this substatute codified what had long been the dominant practice, namely to treat the buyer and wife in an open wife sale as “unrelated parties” for purposes of law. The second clause codified the flexibility advocated by the 1787 edict and the 1789 ruling, which mandated discretion in judging at least some illegal marriages on a case-by-case basis, and, with regard to wife sales, it ruled that a buyer who had acted in good faith (i.e., was unaware that his wife had a living husband) should be credited as the woman’s real husband for sentencing purposes. For some, however, this policy did not go far enough. The late Qing jurist Xue Yunsheng, writing some eighty years later, issued a vituperative critique of this

328

part three

substatute and suggested that it be eliminated as part of a broader rationalization of the Qing code. Xue approved of the substatute’s second clause, which accepted the ignorance of a buyer who acted in good faith (in a fraudulent wife sale) as a mitigating circumstance. But he strongly objected to the first clause’s provision about open wife sales on the grounds that it trivialized the de facto kinship relations that such transactions created: If a wife whose divorce has been purchased beats to death her father-in-law or mother-in-law [i.e., the buyer’s parents], then according to this substatute she should be sentenced as an unrelated party. It follows that if the father-in-law or mother-inlaw kills or injures a daughter-in-law whose divorce has been purchased (or whatever), then he or she too should be sentenced as an unrelated party. But what if the father-in-law or mother-in-law does not know that their son bought her divorce (or had prior adultery with her, etc.)—in that case, would it still be appropriate to treat them as “unrelated parties”?43

Xue certainly did not approve of wife-selling. But the cardinal principle of familial particularism, by which means the judiciary upheld normative Confucian hierarchy, had to take priority. The illegality of a wife sale did not negate the fact that this transaction united individuals as a de facto family in the same household. Within that household, hierarchy had to be upheld, or else human relations would dissolve into chaos. In particular, it would be wrong to treat as an “unrelated party” a parent-in-law who acted in good faith to fulfill that role while ignorant of any impropriety connected to the marriage. Xue’s opinion indicates something more than simple pragmatism. Rather, the Qing criminalization of the widespread practice of wife sale set up a conflict between different priorities within Confucian fundamentalism: female chastity versus family hierarchy and filial piety. Xue was not alone in believing that a rebalance was necessary.44 Application of the 1808 Substatute Xue Yunsheng’s objections notwithstanding, the 1808 substatute provided clearer guidance than had previously existed. It remained in force for the rest of the century; I have many examples of it being applied, and of divergent rulings by lower officials being corrected in line with its provisions upon review.45 Nevertheless, exceptional cases regularly provoked the same sort of doubts that some jurists had always felt about refusing to treat a couple united by wife sale as husband and wife. The Conspectus of Penal Cases offers a striking case from 1831 that the Board of Punishment’s Jiangxi Bureau considered at the request of the governor of that province. The basic question was whether a man who had bought a wife illegally was entitled to any of the authority associated with the role of husband.

Chapter 10. Absolutism versus Pragmatism

329

In this case, Chen Jiazhi had knowingly bought Zheng Qikuan’s wife, Han Shi, to be his own wife. Over the next few years, Han Shi gave birth to three sons for Chen. Then, at some point, she began a secret sexual relationship with their neighbor, Xu Xintuo. One day, Chen caught them in bed having sexual intercourse, and he immediately killed them both with a knife. The stakes in this case could not have been higher. If Chen counted as Han Shi’s legal husband, then he would be released without punishment. Qing law was very clear, however, that only a woman’s husband or other close relatives had the right to “seize adulterers” (zhuo jian), and only the husband could kill them with impunity (but he had to act at once, in the heat of “righteous indignation”).46 Therefore, if Chen counted as an unrelated party, he would be liable for two homicides, which meant the death penalty for sure. Established law ruled out the first alternative (treating Chen as Han Shi’s husband), but the second alternative did not seem right to the magistrate or to his superiors at the provincial level, either. The governor acknowledged that this was a case of “knowingly buying a divorce” and that Chen Jiazhi and Han Shi technically “did not have the status of husband and wife.” Nevertheless, he argued, the fact that this couple had lived together for several years as if married and had had children together meant that Chen “did in fact have the responsibility to supervise and control” Han Shi (shi you guanshu zhi ze), the way a husband would a wife. Since Chen had that practical responsibility, “then he should have been allowed to seize her” in the act of adultery, at least in the manner of a relative, if not a husband. Therefore, the governor asked the board whether he might sentence Chen Jiazhi to penal servitude, by analogy to the substatute against “a relative seizing in the act of illicit sex and immediately killing both the man and woman engaged in illicit sex”; the alternative would be to treat Chen and Han Shi as unrelated parties and sentence Chen to strangulation after the assizes, according to the basic statute against “homicide in a fight.” The Jiangxi Bureau responded by quoting the first clause of the 1808 substatute, to the effect that since a couple married by “knowingly buying a divorce” should, by statute, be divorced, if there is an offense between them, they should be treated as unrelated parties for purposes of sentencing. The underlying principle at stake was that “for a marriage to be legitimate, its beginning must be legitimate” (nannü hunqu dang zheng qi shi). Therefore, it was clear that Chen had to be sentenced as an unrelated party. Nevertheless, the bureau did acknowledge the logic of the governor’s opinion: “This couple had lived together as if married for many years and also had children, and therefore they had in fact acquired the status of husband and wife. Under the circumstances, Chen Jiazhi was not without responsibility for supervising and controlling Han Shi (buwu guanshu zhi ze).” What to do? Fortunately, there existed an apt precedent, a Hubei case judged in 1819 that closely paralleled the facts of Chen Jiazhi’s case: a longstanding illegal marriage

330

part three

produced by open wife sale, in which the buyer caught the wife and her lover in adultery and immediately killed them both. In that case, the buyer was sentenced as an unrelated party, according to the statute against “purposeful homicide,” to beheading after the assizes—but then, at the Autumn Assizes (see Appendix D.3), the Board of Punishment had cited the couple’s longstanding de facto marriage as a mitigating circumstance and had recommended that execution be delayed (which implied that the sentence would almost certainly be commuted in the end). The Jiangxi Bureau recommended this same course of action to the governor: he should sentence Chen correctly, while carefully reporting any mitigating circumstances, and leave clemency to the sovereign. To do otherwise would be “to create a discrepancy between crime and sentence” that implied failure to conform to judicial protocols binding on all ministers.47 The governor’s recommendation of lenience demonstrates a willingness to accept the practical reality of this marriage. In effect, the governor argued, this was a real marriage whether jurists liked it or not, and it would be wrong to pretend otherwise. The board’s response suggests basic agreement with the governor. But the board also had to safeguard protocol, which required the governor to sentence according to the code, especially when the code was unambiguous on the matter in question. In a homicide case, clemency and equity (i.e., deviation from formal law in light of particular circumstances) were the jealously guarded prerogatives of the emperor, who in the Legalist paradigm was both giver and taker of life.48 He exercised these unique, godlike powers at the Autumn Assizes, and they were not to be usurped by his subordinates. Wife-Selling Cases in the Autumn Assizes We have seen how illegal wife sale was taken into account in sentencing, and most relevant cases resulted in sentences of strangulation or beheading “after the assizes” for some act of homicide. These sentences were provisional, and the offenders’ fate would be decided at the annual Autumn Assizes (qiu shen), when the emperor and his senior ministers weighed the mitigating and aggravating factors in each case (see Appendix D.3).49 How, then, did the problematic nature of family relationships in illegal wife-selling cases influence this ultimate decision-making process? The late Qing statesman Gang Yi compiled a reference work of rules and precedents for the Autumn Assizes to guide officials who prepared recommendations for the emperor. Here Gang directly addresses this question: With regard to the mourning system, there are some cases in which the offender has been provisionally sentenced to strangulation after the Assizes, according to the ordinary statute against “homicide in a fight,” on the grounds that he or she is an “unrelated party.” For example, there are cases in which the offender has beaten to death a senior family member, but because one of them has been adopted out of the offender’s lineage, they no longer technically have a relationship within the mourn-

Chapter 10. Absolutism versus Pragmatism

331

ing system (jiang wei wu fu). There are also cases of buying and selling of divorce (or other illegal marriage), in which the sold wife or concubine has beaten her new husband or parent-in-law to death; similarly, there are cases in which the offender has beaten to death his stepmother (jimu), who happens to be a wife whose divorce has been bought by his father. In all such cases, the offenders should be provisionally sentenced to strangulation after the Assizes as “unrelated parties” according to the ordinary statute against “homicide in a fight.” But when deciding their fate at the Autumn Assizes, one may not treat all such offenders in the same [lenient] manner as “unrelated parties” (buke gai yi fan lun). In other words, such offenders should not all be automatically granted a deferral of execution, except in specific cases that are straightforward and include mitigating circumstances (li zhi qing qing).

The scenarios that Gang enumerates here all involve junior family members offending against their seniors; therefore, to treat them as “unrelated parties” instead of according to their relative seniority in the mourning system would imply lenience. For this reason, Gang urges caution: not all offenders with the same provisional sentence deserved the same fate. He adds: There are also cases of a man beating to death a wife whose divorce has been sold. In judging such cases, one must remember that circumstances vary. For example, if a husband beats to death a wife whom he himself has already sold, then the couple previously had the formal status of husband and wife (fuqi mingfen). But if a husband beats to death a wife whose divorce he has purchased, then they now have the practical relationship of husband and wife (fuqi zhi qing). This difference notwithstanding, in either scenario it would seem appropriate to judge such a husband more leniently than if he had beaten to death a woman with whom he had no relationship at all (si ying jiao ou si xun chang funü shao kuan).

Here, Gang considers offenses by husbands against wives, who of course were their juniors in the normative hierarchy. In his view, to treat these men as “unrelated parties” would result in sentences of unwarranted severity. The essence of Gang Yi’s argument is that the absolutist approach to female chastity should be balanced against a realistic appraisal of facts on the ground, as well as against the paramount goal of upholding normative family hierarchy. Although in theory an illegal wife sale should invalidate both marriages, in practice—when these offenders’ lives were at stake—both should be given at least some credit. It might even be appropriate to take the mourning system into account. Gang’s reasoning (like Xue Yunsheng’s opinion cited above) suggests both a pragmatic awareness of the social reality of wife-selling and a reluctance to elevate female chastity over other fundamentalist values.50 Published collections of Autumn Assizes decisions show that Gang’s reasoning reflected actual practice, at least in the late Qing. In a paradigmatic case from Hubei reviewed at the 1868 assizes, one Zhang Chuanyu was on trial for beating his

332

part three

wife to death. She had stolen money from his mother, and when he reprimanded her she defied him and cursed his mother, so he tied her up and beat her to death with a blunt instrument. According to the coroner, Zhang’s beating inflicted ten separate wounds, four of which were in “fatal spots.” He had been provisionally sentenced as an “unrelated party,” because he had acquired his wife from her previous husband in an illegal wife sale. But there were mitigating factors: Zhang had used no edged weapon, “nor were there any broken bones or other aggravating factors,” suggesting to the coroner that he had not actually intended to kill her; and of course the woman’s behavior had been provocative in the extreme. Therefore, Zhang’s crime was found to be “worthy of lenience and a delay of execution, pending future deliberation.” After all, “in the past, in cases before the Autumn Assizes of a husband killing a wife whose divorce he has purchased, such offenders have always been judged relatively leniently, as compared to a man who beats to death an unrelated woman, because [despite the illegality of their marriage] they have the practical relationship of husband and wife (fuqi zhi qing).”51 The Buyer’s Ignorance as a Mitigating Factor in Judging Fraudulent Sales As we have seen, the 1808 substatute drew a distinction between greater and lesser violations of ritual propriety. Greater violations included “knowingly buying a divorce,” whereas lesser violations included marriage that resulted from fraudulent wife sale, in which the unwitting buyer had acted in good faith. According to the substatute, marriages resulting from the lesser violations conferred the legal status of husband and wife at least for purposes of sentencing (even though, by law, such couples should still be divorced). But matters were not always so straightforward, even with fraudulent wife sale. What if the buyer had acted in honest ignorance but later discovered the fraud— that is, found out that his new wife still had a living husband—and then decided to keep her anyway? What would be their relative status before the law under those circumstances? The case record is ambivalent. In the following 1747 example, from Ji Department, Zhili, Liu Wenli took Zhang Shi as his wife, believing that she was a widow; in fact, her husband, Li Da, was alive and had arranged to sell her in a fraudulent transaction. Eventually, Liu learned the truth, but he elected to keep her nevertheless. But Zhang Shi was very unhappy with Liu’s poverty, and she often quarreled with him and with his mother, Chen Shi. Matters came to a head one day when Chen Shi reprimanded Zhang Shi for her incessant complaining, and Zhang Shi responded by cursing Chen Shi and striking her in the face. As a result, Liu decided to sell Zhang Shi, but, before he found a buyer, they had another violent quarrel: she complained bitterly that she lacked adequate food and clothing, accused Liu of “harming” her by taking her as his wife, and again cursed his

Chapter 10. Absolutism versus Pragmatism

333

mother. In response, Liu killed her. The homicide took place not quite two years after the wife sale. The presiding magistrate clearly sympathized with Liu. Everyone else involved in the fraudulent wife sale was sentenced according to the statute against “buying or selling a divorce,” but Liu was sentenced to strangulation after the assizes according to the ordinary statute against “a husband killing his wife.” Evidently, the magistrate believed that the fraud exonerated Liu from blame for the wife sale, despite his subsequent discovery of the truth, and that he should be credited with the authority of a husband. Moreover, the case record includes extensive testimony from family and neighbors about Zhang Shi’s shrewish behavior—for example, “Zhang Shi had no sense of hierarchy or deference! (yuan wu shangxia). She would beat and curse even her own mother-in-law! (lian ta popo dou shi yao da ma de).” This record serves to provide at least partial justification for the homicide: by portraying Liu as an aggrieved husband who had killed a criminally guilty wife, the magistrate enormously improved his chance of lenient treatment at the Autumn Assizes.52 The 1808 substatute (which this judgment preceded by seventy years) required a more unequivocal approach to such cases. Nevertheless, as we see in the following example, a certain amount of confusion persisted over how to judge such a scenario. In an 1819 case from Jiangxi, one Ying Ping bought a woman named Qi Shi as his concubine, believing her to be a widow; in fact her husband, Li Shikui, was alive and had sold her because of poverty. Ying Ping eventually discovered the truth in the following manner: he beat Qi Shi for disobedience, provoking her to run away, and the matchmaker (who helped bring Qi Shi back to him) then told Ying the truth about the fraud and conveyed first husband Li Shikui’s offer to “redeem her body” (shushen) with a partial refund of her brideprice. Ying rejected the offered sum as inadequate. Instead, he watched Qi Shi vigilantly to prevent her from running away and continued to abuse her. A couple of weeks later, after another harsh beating, she hanged herself. Once again, the critical legal question was whether Ying should be judged as Qi Shi’s husband or as an “unrelated party,” because his culpability for her suicide depended entirely on the answer to this question. The stakes were high. The beating had caused no “breaking injury” (zhe shang)—no broken bones or teeth or worse—and therefore, if Qi Shi counted legally as Ying’s concubine, then he was guilty of no crime whatsoever. But if the two counted as “unrelated parties,” then the beating that provoked the suicide would count as a variation of the crime of “forcibly causing another person’s death” (wei bi ren zhi si), with a likely penalty of a beating and three years of penal servitude at very least. This case came before the Board of Punishment in a confused state, because lower officials had failed to clarify the legal relationship between Ying Ping and Qi Shi. In response, the board cited the basic statutory principle that, in cases of

334

part three

illegal marriage where divorce was mandated, the couple “should be treated as unrelated parties” for purposes of sentencing if any offense occurred between them. It further confirmed that selling a wife because of poverty counted as the crime of “selling a divorce,” in which case divorce would be mandated. The board conceded that Ying Ping may have been ignorant of Qi Shi’s true status as “a wife with a living husband” (you fu zhi fu) at the time when he “bought her divorce,” but asserted that the point was moot because even after Yang learned the truth, “he still willingly kept Qi Shi as his concubine” instead of returning her to her husband and reporting their crime to the authorities. His failure to do so was “no different from consenting to buy her divorce” in the first place. Therefore, at the time of Qi Shi’s suicide, she and Yang must have counted as “unrelated parties” for purposes of law.53 Complex scenarios such as these must have driven magistrates to distraction, and there was no shortage of them.54 No matter how often the central authorities tried to clarify how to interpret and enforce the law, some complex new scenario would invariably emerge to stymie them. This frustrating problem resulted from the interplay of three factors. The first was the age-old imperative of particularism in imperial law, which demanded that normative patriarchal hierarchy determine the penalties for offenses between kin. The second was the High Qing campaign to apply an ideal standard of female chastity to all, which included an enormous expansion of the prohibition against “buying or selling a divorce” to include all unauthorized wife sales regardless of motive. The third was the fact that povertydriven wife sales were increasingly common throughout China. SHOU L D P OV E RT Y B E T R E AT E D A S A M I T IG AT I N G FAC T O R ?

Perhaps the most striking example of tension between absolutism and pragmatism is the question of whether poverty should be treated as a mitigating factor in the judgment of wife sales. The Ming dynasty’s policy had been that “whoever sells his wife in marriage out of poverty or illness should not be punished; nor should any second husband be punished if he has used money to buy another man’s wife in marriage, but whose case involves no illicit sex.”55 Under the Qing, however, the vast majority of wife sales prosecuted under the statute against “buying or selling a divorce” were motivated simply by poverty. From the absolutist point of view, poverty was no excuse. As the magistrate of Xinfeng County, Jiangxi, commented in a 1774 ruling: “Chen Yide sold his wife only because he was pressed by both illness and poverty, and the couple agreed to seek separate routes to survival. Be that as it may, to sell off one’s wife for cash constitutes the crime of ‘selling a divorce’ (de qian jia mai ji shu mai xiu).”56 An 1803 handbook for magistrates, Lü Zhitian’s Lü fa xu zhi (Necessary Knowledge of

Chapter 10. Absolutism versus Pragmatism

335

the Law), reiterates the absolutist position on poverty as a corollary to impugning the chastity of a sold wife: When a wife is sold because of poverty, even if her husband proposes the idea, the situation is usually the wife’s fault, because she resents her husband’s poverty and is unwilling to live with him in peace. If a wife is properly diligent at spinning and sewing, and she and her husband get along in harmony, then in spite of poverty it will never occur to him to sell her. There are also some cases in which a lazy, shiftless husband tries to sell his wife, but she refuses to separate from him.57

In short, a truly virtuous wife could not be sold, regardless of poverty or other factors, and therefore the fact that a woman had been sold was proof of her moral failure. We find an echo of this reasoning in an 1881 case from Nanbu County in which the magistrate admonished a man who had complained on behalf of his sister that his brother-in-law was trying to sell her: “If your sister had truly exerted herself to fulfill the way of a wife (ke jin fudao), then her husband would have no reason to sell her divorce!”58 Although the absolutist position remained the dominant and empowered one, the evidence reveals serious misgivings and a lack of consensus within the central judiciary. It was common knowledge that the big story was poverty, not licentiousness, and whether from compassion or pragmatism, many officials were reluctant to enforce the full provisions of the statute against “buying or selling a divorce” when the sole motive for a wife sale had been poverty, because they feared that strict enforcement would only exacerbate the problems that had caused the sale, without clear benefit to anyone. They did not go so far as to advocate decriminalization. But, specifically, they believed it served no useful purpose either to confiscate the brideprice or to force the woman to return to her natal lineage. Most men who sold their wives would spend the entire brideprice immediately, and it was difficult if not futile to try to get it back. In addition, a sold wife who returned to her natal family would just end up being sold once again, causing a great deal of disruption—as well as resentment on the part of the first buyer, who would lose both brideprice and wife. Some officials also felt that the mandated beating with the heavy bamboo was too harsh, and, as we have seen, central cases show that women were often allowed to redeem this penalty with a fine (as long as no prior adultery had been involved). One can find a number of examples of senior officials attempting to moderate policy in this light. For example, in 1761 the governor of Fujian approved a proposal that, in wife sales motivated by poverty or illness of the seller, and involving no prior adultery or “any other factor,” lenience should be shown by allowing the second marriage to stand and the seller to keep the brideprice. In addition, the governor agreed to reduce corporal penalties in such cases from the one hundred blows of the heavy bamboo mandated by the statute against “buying or selling a

336

part three

divorce,” to eighty blows of the heavy bamboo (for the seller) and forty blows of the light bamboo (for the buyer), by applying instead the statute against “doing inappropriate things” (bu ying wei er wei) for “severe” (qing zhong) and “less severe” (qing qing) cases, respectively. Matchmakers would receive a penalty one degree lower than sellers. In recommending this change, the governor’s advisors cited a similar policy that was apparently already in effect in Jiangsu and Zhejiang. The governor ordered that officials throughout the province be notified of the lenient policy, and it was formally incorporated into the provincial regulations (sheng li).59 In a similar vein, Lü Zhitian’s handbook makes the following observation: Recently, cases of wife sale caused by poverty have often been adjudicated by returning the woman to the second husband in acceptance of the marriage contract, without confiscating the brideprice. In this sort of case, there are some couples who according to law (fa) should be ordered divorced, but given their actual circumstances (qing) may be permitted to stay together.60

Lü Zhitian does not specify a region, so his comment suggests a wider application of the lenient policy than just the three provinces mentioned above. (This passage contrasts with the one cited, in which Lü impugns the chastity of sold wives. Lü’s two comments embody the tension between the absolutist and pragmatist positions.) We find the same lenient policy recommended by the governor of Henan in an 1818 case, which is briefly summarized in The Conspectus of Penal Cases: Wang Heigou sold his wife, Hu Shi, to Li Cunjing to be his wife, and interrogation reveals that he did so because of poverty and illness, and that he had no alternative. This is different from selling a divorce for no reason (yu wugu mai xiu zhe you jian). Hu Shi has no relatives on her mother’s side, so if she is divorced as provided in the statute [on “buying or selling a divorce”], then she will just end up losing her chastity (shijie) anyway. Weighing the circumstances of the case (heng qing zhuo duan), it seems that Hu Shi should be handed over to her second husband, Li Cunjing, for him to take home as wife. Also, it is requested that the brideprice money received by Wang Heigou not be confiscated.

Since this judgment appears in an influential casebook, one would expect that the Board of Punishment had approved it. But the fact that it seemed necessary to publish it at all implies that, as late as 1818, there was still no consensus on how such cases should be treated.61 These initiatives suggest a conscientious desire on the part of some senior officials to mitigate the harshness of the code in light of the fact that wife sales nearly always resulted from poverty rather than licentiousness. But I find no evidence of any real impact from these upper-level initiatives. The xingke tiben from Fujian, Jiangsu, and Zhejiang that I have collected show no evidence of the lenient policy prescribed by their provincial regulations. Instead, they follow the usual protocol

Chapter 10. Absolutism versus Pragmatism

337

by rigidly applying the statute against “buying or selling a divorce.”62 In fact, the actual purpose and scope of provincial regulations like these are far from clear. For what it is worth, I have never seen provincial regulations cited or even mentioned in any legal case from the archives, either central or local. Nor is 1818 a watershed in my sample of central cases—in fact, I find only a single judgment (from Jiangxi, dated 1819) that applies the policy approved by the Henan Bureau.63 In contrast, there are many others in which the statute against “buying or selling a divorce” is strictly applied in the usual manner. Moreover, I find no evidence in either central cases or published casebooks that such strict sentences were ever commuted upon review, which shows that local magistrates were safe in opting for strict enforcement over leniency when preparing such major cases. Although it was well known and widely read, The Conspectus of Penal Cases had no official standing as a source of law, and, as this example shows, the cases it summarized did not necessarily count as binding precedents. I have never seen a judgment that actually cites The Conspectus or any other casebook for justification. In his 1844 treatise Qi min si shu (Four Methods for Governing the People), the noted statecraft thinker Bao Shichen decried the fact that “nowadays, whenever there are cases among the common people of a husband selling his wife in marriage because he cannot support her, they are all sentenced according to the statute against ‘selling a divorce’ (jie ke mai xiu).” Bao began his polemic by citing an 1811 case from Shanxi to illustrate the sort of miscarriage of justice that could result from what he saw as an unrealistic policy. In that case, a widower had “purchased in marriage a wife with a living husband” to be his second wife, and later she had murdered both him and his mother. Because of the wife sale, she had been sentenced as an “unrelated party” (as the 1808 substatute required). Bao considered her sentence outrageously lenient: a wife’s penalty for murdering her husband or mother-in-law was dismemberment, but an unrelated party who killed two persons of a single household would receive the lower penalty of immediate beheading. In Bao’s view, the judiciary’s refusal to acknowledge the actual marital relationship in this case had the effect of trivializing a heinous offense against patriarchal hierarchy. In this respect, he agreed with the opinions of Xue Yunsheng and Gang Yi cited above. Even more significantly, Bao Shichen went on to express his deep sympathy for the poverty that was the real cause of the widespread phenomenon of wife-selling: It is those responsible for teaching and nurturing the common people (ren jiao yang si min zhi ze zhe) who should be ashamed if poverty and hunger drive husbands and wives to separate in order to save themselves. Equally, when humble men lack the means to marry according to proper ritual, and buy other men’s wives with the aim of continuing their ancestral sacrifices, their actual circumstances must be taken into consideration. It makes no sense to punish them all in the same way according to this blanket law!

338

part three

Bao argued that the absolutist extension of criminalization to cover all wife sales was based on a misreading of the statute against “buying or selling a divorce.” The statute should properly apply only to cases that involved prior adultery, and all other wife sales should be handled under the broader, more lenient statute against “illegal marriage” (jia qu wei lü). “Ordinary cases of wife-selling because of poverty should not be treated as ‘buying or selling a divorce’—there should be absolutely no doubt about the matter.”64 In effect, Bao advocated a return to the Ming judiciary’s original, narrow application of the statute against “buying or selling a divorce.” The fact that he felt compelled to make this argument in the mid-nineteenth century shows that the absolutist policy of the High Qing remained dominant, despite the lenience that some officials had advocated. The only new feature of Bao’s argument was his pointed suggestion that officials should look to their own failures instead of futilely punishing desperate people who were struggling to survive. In other words, to challenge the absolutist promotion of female chastity, Bao posed the classic Confucian claim that social crisis and a rising crime rate reflected the moral failure of those in power. Once again, we see how the criminalization of wife sales set up a conflict between rival fundamentalist values. In sum, there is a contradiction between the evidence that some senior officials sought to mitigate the harshness of the law, and the lack of evidence that such initiatives had any practical effect. How can we explain this contradiction? One might suppose that the lenient policy advocated by the provincial regulations of Fujian, Jiangsu, and Zhejiang and by the Henan governor’s judgment was intended to apply only to routine “minor cases” handled flexibly at the local level, not to the extraordinary “major cases” reported up the chain of command for central review. If so, then it would not be surprising that we find no evidence of lenience in central cases—and, in the present chapter, we are considering only central cases, not routine cases from local courts. But the evidence from the archives does not support that supposition. Nearly all the wife-selling cases that came before local courts were motivated by poverty, and, as we shall see in the next chapter, poverty alone is not a good predictor of how they were judged. Regardless of what Lü Zhitian wrote in 1803, magistrates did not usually ratify such wife sales by leaving the woman with her buyer, nor did they necessarily refrain from confiscating the brideprice. Moreover, my sample of more than 340 wife-selling cases from local court archives shows no evidence of change over time in how they were adjudicated; there is no evidence in the local archives of the initiatives described above. Therefore, it would seem inaccurate to explain the upper-level initiatives as attempts either to influence or to ratify the way routine “minor cases” were handled by local courts. The evidence shows that the senior judiciary never achieved strong consensus about how to handle wife sales motivated by poverty. It is highly significant that

Chapter 10. Absolutism versus Pragmatism

339

the lenient policy advocated by some was never codified as a substatute. On the contrary, as the magistrate of Nanbu County pointed out in 1840, when condemning a wife sale motivated by poverty: “There is no substatute permitting the sale of wives because of poverty” (li wu yin pin zhun qi mai qi zhi tiao).65 Codification as a substatute would have guaranteed the lenient policy’s application across China, at least in major cases subject to review. The lack of such a substatute reflects the lack of consensus in support of that policy. Under the circumstances, it is no surprise that, when preparing major cases, local magistrates would err on the side of strict enforcement of the code. SHA R P E N I N G C O N T R A D IC T IO N S

The debate over this issue began in the late Ming and persisted over the course of the Qing. Whereas the Ming had prohibited only wife sales by free commoners that constituted “abetting or tolerating illicit sexual relations,” the Qing dramatically expanded prohibition to cover sales motivated solely by poverty. But this expansion provoked misgivings on the part of many officials, and the evidence reveals an undercurrent of disquiet and even dissent that runs right through the dynasty. These misgivings reflected some officials’ pragmatism and compassion but also a concern that the absolutist standard of female chastity threatened to undermine other fundamental values that were at least as important. As noted above, we find no such misgivings with regard to the prosecution of polyandry, marital prostitution, or other survival strategies covered by the criminal category zong jian. There is no question that jurists considered wives who participated in such strategies to be genuine adulteresses who deserved punishment, or that they considered wives who died resisting pressure to do so to be genuine chastity martyrs. But we also find no judicial agonizing about whether to treat the relationships created by polyandry and chosen kinship as “real” marriages or families; no jurist proposed that such relationships might justify sentencing offenders according to the mourning system; no outside male “brought in” by either contract or sworn brotherhood was credited with the authority or prerogatives of a husband. No doubt, the reason for this lack of sympathy is that polyandry and prostitution depended on the unambiguous sexual promiscuity of one woman with more than one man simultaneously, whereas a wife sale resulted in what we might call serial monogamy—not an ideal result, from an orthodox standpoint, but ultimately little different from widow remarriage or authorized divorce. This contrast shows the limits of the imagination and sympathy of even the most broadminded of Qing jurists. A basic purpose of the Qing system of judicial review was to ensure uniformity of standards in judging major cases—that is, to make sure officials at all levels and in all provinces understood how to apply the code to specific scenarios of crime.

340

part three

The review system represented the Legalist dimension of Qing law, which demanded deference to protocols issued by the imperial center in order to reduce local officials’ discretion to an absolute minimum. But this system never worked perfectly, despite constant tinkering. This chapter has focused on major cases reported to the imperial center, and such exceptional cases constituted only a small percentage of those judged at the local level. It would be wrong, however, to imagine that these policy debates had no practical effect, because in specific cases a defendant’s life might hang in the balance. But they seem to have had little bearing on how local magistrates actually handled the vast majority of wife sales that came to court. Rather, the chief significance of these policy debates lies in what they reveal about the sharpening contradictions between the intractable fundamentalism of the dynasty and the mounting social crisis in rural China. As Bao Shichen intimated in his scathing critique of the criminalization of wife sales, High Qing fundamentalism offered no solutions for that crisis.

11

Flexible Adjudication of Routine Cases in the Local Courts In principle, all of the participants in this criminal wife sale should have been severely punished. . . . But when one considers that [the seller] Jingwen was afflicted by both poverty and illness, his offense seems rather forgivable. I ordered [the buyer] Daji to pay him an additional sum of 12,000 cash for his sustenance; and in exchange, I authorized Daji to take Peng Shi home and keep her as his wife. From the standpoint of formal law (fa), my judgment must seem excessively permissive; but given the actual circumstances (qing) of the case, it seems like an acceptable deviation that achieves a higher goal (si ke ququan). To be frank, I based this judgment on nothing more than my sympathy for the weak and vulnerable (ai jin she yu). — magistrate shen yanqing, recounting his judgment of a wife sale 1

In contrast with the two preceding chapters, this one analyzes how routine “minor” cases were handled in the local courts.2 The main source for this chapter is a selection of 291 cases from Ba, Baodi, and Nanbu counties that involved clearly illegal wife sales (chosen from the total sample of 345 local wife sale cases). We focus here on routine adjudication of wife sales, but for comparison we also consider the judgments in fifty-two cases from Ba County that involved husbands’ complicity in their wives’ prostitution or polyandry (the crime abbreviated as “zong jian”). All of the cases in both samples concerned “minor matters” that did not need to be reported up the chain of command for review. For this reason, they present a very different sort of adjudication than that seen in the “major cases” presented in the previous two chapters. Chapter 10 explored a fundamental contradiction between the absolutist chastity ideology of the High Qing, which insisted on interpreting all wife-selling as a variation of “illicit sex,” and the more pragmatic position of many officials who recognized that wife sales were symptoms of a mounting social crisis that could not be understood simply in terms of licentiousness. Despite this contradiction, 341

342

part three

magistrates preparing major cases for review would apply the statute against “buying or selling a divorce” to the letter. But how did they treat wife sales that appeared in routine cases? A basic question is to what extent (if at all) magistrates enforced the Qing code in cases that involved only “minor matters” and therefore were not subject to review.3 The significance of this question extends beyond the specific topic of this book to the broader field of Chinese legal history, which has witnessed intense debate about the nature of local adjudication in the Qing.4 To what extent could the imperial center impose its ideological mandates through the local courts? What this chapter offers is a focused case study of how magistrates, as a matter of routine, dealt with tenacious, widespread social practices that were prohibited. In this confrontation, where did the balance of power lie? What factors guided magistrates’ decisions? The present chapter is divided into four sections. First, we examine out-ofcourt settlements that were approved by magistrates. Perhaps 20 percent of the cases in my sample ended with mediated settlements, most of which were not reported to the yamen, but those that were reported show that magistrates would approve just about any settlement as long as it preceded a formal judgment. Second, I analyze my sample of illegal wife-selling cases to establish why they ended up in court, how they were judged (if they went to formal judgment), and what logic may explain the patterns of adjudication so revealed. The evidence shows that magistrates judged on a case-by-case basis, and the best predictor of how a given case would be judged is the specific reason it ended up in court. Third, I examine fifty-two routine cases involving zong jian (i.e., marital prostitution or polyandry), all of which ended in formal judgments. In some respects, these cases differ from those involving wife sales, but as far as adjudication is concerned, we see the same fundamental pattern: instead of rigidly enforcing the Qing Code, magistrates exercised pragmatic flexibility. Their chief priority was to address, as expediently as possible, the concrete problems that had brought these cases into court. Fourth and finally, I use these findings to appraise past explanations of routine adjudication in the local courts of the Qing dynasty. I focus on the foundational work of Philip Huang and Kishimoto Mio, both of whom have contributed major insights to the field of Chinese legal history, but neither of whom can account adequately for the findings of this chapter. O U T- O F- C O U RT SE T T L E M E N T S A P P R OV E D B Y T H E M AG I ST R AT E

Once charges had been filed, the only sure way to avoid an adverse judgment was to settle out of court, before one’s case reached the final stage of a formal hearing.

Chapter 11. Flexible Adjudication of Routine Cases

343

A significant proportion of legal cases stemming from wife sales ended with outof-court settlements, usually mediated by third parties. Here are the basic outcomes of the 291 cases of illegal wife sales that are the main focus of this chapter: Formal judgment: Ba County: Nanbu County: Baodi County: Contract in file, but no other information: Magistrate accepts petition “to create official record” Out-of-court settlement approved by magistrate: Closed by magistrate without resolution: Unclear:

196 146 41 9 8 2 22 6 57

The last three categories listed are relevant here: at least twenty-two cases ended with out-of-court settlements approved by magistrates; six were closed without resolution because litigants failed to appear in court; and fifty-seven ended without any documented resolution. There is no way to be sure what happened in the last two categories of cases, and no doubt some files in the third category are simply incomplete. But, as Philip Huang has suggested, there is good reason to believe that half or more of such cases actually ended with out-of-court settlements that the litigants simply did not bother to report to the yamen. (On this basis, Huang argues that out-of-court settlement was the single most frequent outcome in his own sample of “civil cases.”)5 If we accept Huang’s reasoning, then we can estimate that perhaps 20 percent of my own sample ended with out-of-court settlements. Here, we focus on the twenty-two mediated settlements that we know received official approval. The evidence shows that magistrates would ratify nearly any settlement, regardless of the terms, as long as it was achieved in the intermediate stage of litigation prior to a formal court hearing. This was true even when magistrates had explicitly denounced those sales as crimes. A 1790 case from Ba County provides a vivid illustration. The case involves a typical open sale, after which the seller filed false charges that the buyer had taken his wife by force. Four neighbors then mediated a settlement and reported it in a petition requesting that the case be closed.6 According to the mediators’ petition, peasant Wang Yugui and wife Dong Shi did not get along with each other and agreed to separate. Two months previously, Wang had voluntarily arranged through a matchmaker to sell Dong Shi in marriage for a brideprice of 16,000 cash to Liu Shijin, giving him a handprint contract to seal the transaction. Later, however, Wang went to Liu’s home and demanded that Dong Shi return two items of clothing that she had taken with her, but she refused, and they had a row. Angry, Wang filed false charges

344

part three

accusing Liu of forcibly taking his wife in marriage. Now, thanks to the mediators’ efforts, the dispute had been resolved: Dong Shi had returned the clothing, Wang had withdrawn his accusations, and all involved begged to end the lawsuit. The petitioners’ account seems disingenuous. It seems unlikely that a peasant would really take the trouble to file a lawsuit at the county seat over two items of clothing, and Wang Yugui almost certainly received a supplementary payment in exchange for withdrawing his charges. That is certainly how the magistrate interpreted their story: If both sides were willing to separate because Dong Shi failed to keep the way of a wife, then she should have been sent back to her natal family to be remarried. For Wang Yugui instead to sell her divorce without authorization, so that he could get her brideprice, was already a serious violation of law. For him to file a false accusation of “forcible marriage” in hopes of extorting more money is even more disgusting! In principle, he should be severely punished. Nevertheless, since you have already mediated the dispute and both sides agree, I shall forgive his peasant stupidity (shu qi xiang yu) and approve your petition to close the case.

The file ends with a pledge from Wang withdrawing his charges, admitting that he sold his wife voluntarily, and promising no more trouble.7 By agreeing to close this case, the magistrate ratified the new marriage, even though he had denounced it as criminal. A second example is an 1864 case from Nanbu County that started out much the same way, with the seller in an open sale filing charges to pressure the buyer to pay supplementary payments. In this case, however, the magistrate made a show of demanding a court hearing before approving the mediated settlement. The file begins with a petition from Wu Shigui (32) asking the magistrate to close his lawsuit. Wu had previously filed charges against Sun Dabin for kidnapping Wu’s wife, Liu Shi. But now Wu confessed that in fact he had sold his wife to Sun voluntarily because of poverty and illness, for thirteen strings of cash; after the sale he spent the entire brideprice and was left utterly destitute, so he asked Sun for a “loan.” Sun refused, and, in desperation to pressure him for more money, Wu Shigui had filed false charges. Now, however, Wu realized the error of his ways: he had submitted to mediation by the matchmaker and had given Sun a written pledge “never again to use the marriage as an excuse to try to extort money.” In conclusion, Wu begged the magistrate to close the lawsuit without a formal hearing. The clear implication is that Wu had been paid off by Sun. But the magistrate refused to be persuaded so easily, and he summoned the parties for a hearing. His findings are summarized in a “pledge of willingness to end the lawsuit” submitted by Wu Shigui: Your Honor granted us the favor of a hearing, in which he has found that Sun Dabin should not have taken in marriage a wife with a living husband, He Tianbi should not

Chapter 11. Flexible Adjudication of Routine Cases

345

have acted as matchmaker to negotiate the marriage, and I should not have sold her divorce and then practiced extortion. All of us deserve criminal penalties. Nevertheless, we have come forward to express our regret for our misdeeds, so Your Honor condescends to waive our punishment. As to Liu Shi, Sun Dabin is ordered to take her home to rejoin him in marriage. We beg that the lawsuit be canceled, and we pledge to submit to Your Honor’s judgment.

In short, despite his insistence on a hearing, the magistrate ended up confirming the mediated settlement without punishing anyone.8 A number of cases show yamen personnel playing a key role in mediating outof-court settlements of lawsuits, and they seem to have done this with the knowledge and approval of magistrates. Such cases illustrate the sometimes intense interaction between what Huang would call the formal and informal realms of “civil justice” in the semi-formal “third realm” that characterized the intermediate phase of a lawsuit between the initial filing of charges and a final judgment.9 Mediation and out-of-court settlements did not occur in a vacuum but rather with awareness of the law and sometimes the active intervention of court personnel. We see an especially vivid example in an 1865 case from Nanbu County involving a typical open wife sale followed by attempted extortion by the seller. In this instance, after the seller filed false charges at the yamen, the runner detailed to the case and at least two clerks helped mediate a settlement. The result was the following submission contract, drawn up in the seller’s name: I, Mei Yinglong, hereby establish this contract to settle a dispute, through the intercession of the yamen, that arose from the trouble I stirred up by filing false charges after having married off my own wife. It all happened because I was matched with my wife Yi Shi in childhood and we were married for several years, but we were very poor and did not get along, and on top of that we had no means of obtaining our daily food, so it was impossible to carry on together. I voluntarily discussed the situation with the lineage and then engaged a matchmaker to arrange for Yi Shi to remarry to Zhao Dazhao as wife. I personally received the entire brideprice, and nothing more was owed me, as is proven by the contract I wrote with my hand and foot prints on it. But then, after having sold my wife, I had no means of getting by, and troublemakers incited me, so I filed charges at the yamen against Zhao Dazhao. Runners were sent out to summon the parties for a hearing. But fortunately, my neighbors, relatives, and friends called a meeting to sit both sides down and reason with us. I, Mei Yinglong, admit that I am responsible for the crime of “buying or selling a divorce” (you mai xiu mai xiu zhi ze), and I voluntarily agree to settle this dispute through the intercession of my family members. In addition, they have persuaded Zhao Dazhao to pay me [an additional] 4400 cash to help me survive while I seek another means of making a living. I will never again cause trouble or difficulty for the family that took Yi Shi in marriage. If I ever repeat this behavior, then the Mei lineage and others (i.e., witnesses named below) will take full responsibility. Now, fearing that people’s hearts are not reliable, I hereby establish this contract for

346

part three

settling this dispute through the intercession of the yamen and give it to Zhao Dazhao to save as future proof [of my promises].

This text twice cites the intercession of yamen personnel (ping an—literally, “relying on the yamen”) as the reason for a settlement being achieved. It attributes the mediation to “family, neighbors, and friends,” but the telling point is that two of the nine witnesses are “yamen branch-office clerks” (fang shu) and one is “the yamen runner detailed to the case” (yuan chai). One of the clerks also acted as scribe to write the document, which is unusually literate for its genre, and the use of the legal term “buying and selling a divorce” also reflects the involvement of official personnel.10 Once again, an out-of-court settlement ratifying a blatantly illegal transaction received official approval—and, in this case, the settlement was actually brokered by yamen personnel. These examples show that magistrates would accept and indeed encourage practically any mediated settlement, as long as it was reached prior to the issuing of a formal judgment.

C A SE S T HAT E N D E D I N F O R M A L J U D G M E N T

Once charges were filed, most wife-selling cases did persist through a court hearing to receive a formal judgment (at least 196 cases out of my sample of 291). How, then, did magistrates judge these cases? The specific reasons that cases came to court tell us a great deal about the perspectives of the main participants in wife sales. But these reasons also help explain the logic behind magistrates’ adjudication of particular cases. We know the reasons that 270 cases (out of 291) came to court:11 A. Extortion by seller or his family: 115 cases (43% of 270)

Extortion of supplementary payments by seller: Extortion of supplementary payments by seller’s family:

105 10

B. Wife and/or her natal family object to sale: 84 cases (31%)

Natal family (but not necessarily wife) objects to sale: Wife (but not necessarily natal family) objects to sale: Both wife and natal family clearly object to sale:

46 28 10

C. Other extortion or money dispute: 30 cases (11%)

Extortion by natal family (usually aimed at getting share of brideprice): Other extortion (by matchmaker, etc.): Other money dispute (over failure to pay, division of brideprice, etc.):

11 5 14

D. Miscellaneous: 48 cases (18%)

Seller and/or buyer petition for permission to sell wife:

10

Chapter 11. Flexible Adjudication of Routine Cases

Runners arrest people on suspicion of trafficking: Seller’s family objects to sale: Buyer in fraudulent sale discovers fraud: Dispute over child custody: Other:

347

9 7 2 2 18

The single most common reason cases came to court was demand for supplementary payments by the seller or his family (115 cases, or 43 percent of the total for which we know the reason). The second most common reason was that the wife and/or her natal family objected to the sale (84 cases, or 31 percent). There is no overlap between these categories, so together these two reasons account for about three-quarters of the cases. The statute against “buying or selling a divorce” mandates that both husbands, the woman, and the matchmaker(s) all be beaten, that the woman be divorced and returned to her natal family, and that the brideprice and any matchmaking fees be confiscated. (For the statute’s text, see Chapter 9.) We can use these three elements—the fate of the wife, the fate of the money, and the infliction of corporal punishment—to assess how strictly magistrates abided by the code when judging illegal wife sales, and how and why they deviated from it. The Fate of the Wife The fate of the wife was the outcome most fundamental to the ideology of female chastity and marital integrity that underlay the statute against “buying or selling a divorce.” Was the new, illicit union allowed to continue? If it were canceled, would the wife be separated from her first husband as well? If the wife was not divorced from both men and returned to her natal family, then we may fairly conclude that both the letter and the spirit of the statute were being ignored. We know the fate of the wife determined by 188 judgments. They can be divided into two groups. The first consists of ninety-six outcomes that are more or less consistent with the statute (although the second item in this group may be a stretch): Returned to natal family: Transaction not yet completed, returned to first husband: No natal family, sold in marriage by official matchmaker: Both marriages canceled, woman becomes a nun:

74 (39% of 188) 11 10 1

The second group consists of ninety-two outcomes that clearly violate the statute’s mandate: Returned to buyer, second marriage allowed to stand: Returned to first husband even though transaction is complete:

61 (32% of 188) 31

348

part three

In sum, nearly half of all judgments violated this aspect of the statute. It is especially striking that about a third of all judgments in effect ratified illegal wife sales by allowing the second marriage to stand. In addition, we know the fate of the wife in all twenty-two cases in which magistrates formally approved out-of-court settlements. Six outcomes (27 percent) are consistent with the statute: Returned to natal family: Transaction not yet completed, returned to first husband:

4 2

But the outcomes in sixteen approved settlements (73 percent) violate the statute: Returned to buyer, second marriage allowed to stand: Returned to first husband even though transaction is complete:

15 1

It is obvious that magistrates did not simply enforce the code—rather, the sheer diversity of outcomes indicates that they determined each woman’s fate on a caseby-case basis. What specific factors influenced a magistrate’s decision about what to do with the wife? In routine cases of this kind, magistrates rarely explained (let alone justified) their decisions in any detail. The best we can do is to try to correlate outcomes with the salient facts of each case and make reasonable guesses about their connections. The single best predictor for a magistrate’s handling of a given case is the reason it came to court. If the wife herself clearly objected to the sale, here is what happened to her (in thirty-one judgments for which we have information): Returned to natal family: Transaction not yet completed, returned to first husband: Sold in marriage by official matchmaker: Returned to first husband even though transaction is complete: Both marriages canceled, woman becomes a nun:

20 5 3 2 1

The most significant point here is that in no instance when the wife herself objected to the sale did the magistrate allow the second marriage to stand. Occasionally a magistrate would explicitly cite the woman’s own wishes in this regard. For example, an 1864 case from Ba County involved an open wife sale followed by the seller’s attempt to extort more money by filing false charges. The magistrate allowed the woman to choose her own fate, as can be seen in the summary of his judgment that concludes her testimony: Today I have been granted the favor of a hearing. My husband Fang Zhengxing should not have sold this humble wife’s divorce to Zhu Sidaye, and Your Honor has granted the favor of having him slapped. This humble wife is not willing to follow Fang Zhengxing as wife, nor am I willing to be married to Zhu Sidaye as wife.

Chapter 11. Flexible Adjudication of Routine Cases

349

Therefore Your Honor has ordered that my father He Qishan take custody of me and find me a different marriage.12

If we examine two Ba County cases that were judged in 1831, within a few weeks of each other, by a single magistrate, we find that each woman’s own actions and preferences seem to have exerted a decisive influence on her fate. The magistrate in question was Gao Xuelian (a holder of the juren degree from Wuwei Department, Anhui), who was highly experienced, having already served in five other Sichuan counties.13 In the first case, peasant Huang Zhongpin had agreed to sell his wife, Lin Shi, for 3500 cash to Wu Zhide because of poverty. (Huang was the youngest of five brothers, and he had fallen into difficulty soon after the division of their father’s household.) But Lin Shi (who was just 16 sui, having consummated her marriage at the unusually young age of 13 sui) objected to being sold and complained to her father, who filed charges. Magistrate Gao ordered punishment for everyone complicit in the abortive sale. He ordered husband Huang Zhongpin, the buyer, and the matchmaker all beaten, and the brideprice confiscated. Since Huang Zhongpin had already spent the money, he was held in custody until his brothers brought the money to court on his behalf. But Magistrate Gao had nothing but praise for Lin Shi, whom he saw as a paragon of wifely virtue: “Even though this young wife is only 16 sui, she refused to be sold, and her chaste resolve is truly meritorious!” He restored her to her first husband, who pledged never again to attempt to sell her, and ordered an older brother to stand guarantee, with the understanding that he too would be punished if Huang violated his pledge.14 If the first case suggests the triumph of orthodox values, the second gives a very different impression. Lu Shihuan (28) sold his wife, Lin Shi (no relation to the wife in the previous case), partly because of poverty but mainly, it seems, because of incompatibility: Lin Shi was very unhappy with her husband, and she repeatedly ran away to her natal home. Lu’s appeals to her father had no effect. Finally, Lu arranged to sell her to widower Liang Wu for the relatively low price of 2400 cash, without informing her natal family; when her father found out, he filed charges. Magistrate Gao ordered that everyone involved in this case be slapped—including Lin Shi and her father, because he blamed her for provoking her own sale and her father for abetting her misbehavior. But Lin Shi testified that she preferred to stay married to the buyer, Liang Wu, so Magistrate Gao ordered the second marriage to stand. Lu Shihuan kept her brideprice, in exchange for his pledge not to bother the couple.15 In neither case did Magistrate Gao follow the code’s mandate that a sold wife be divorced and returned to her natal lineage. Instead, it appears, the key factor determining how he disposed of each wife was her own stated preference. Other sources confirm that magistrates considered the wishes of women when deciding their fate in such cases. For example, the published judgments of Li Jun

350

part three

(who served in Henan in the Daoguang era) include a case of fraudulent wife sale in which the seller tried to extort more money by filing false charges against the buyer. In his account, Li Jun cites the attitude of the wife herself to justify his decision to allow the second marriage to stand: “Yan Shi’s affections had become attached to her second husband (qing lian hou fu), and she repeatedly begged me (zaisan aiken) to let her stay with him. Under the circumstances, it seemed best to make an exception and permit them to remain married.”16 Returning to my case sample of women who objected to their sale: in the seven cases in which the wife was returned to her first husband, she or her natal family had filed suit to prevent or cancel the transaction. Moreover, the three cases that resulted in sale by the official matchmaker each had special circumstances that help explain why magistrates resorted to this solution. In one case, the woman came from a distant county, and neither her natal family nor her first husband was available to take custody; in the second, the woman had committed adultery with a third man after her sale had been transacted, and moreover her natal family had profited from the sale (such complicity usually disqualified a natal family from taking custody of the woman); and in the third case, the magistrate had initially ordered the woman returned to her natal family, but when she failed to cooperate (her father could not control her), the magistrate ordered her taken into custody and sold. In all three cases, the official matchmaker was a last resort when other options seemed untenable. If the natal family (but not necessarily the wife herself) objected to the sale, here is what happened to her (in thirty-six judgments for which we have information): Returned to natal family: Returned to buyer, second marriage allowed to stand: Returned to first husband:

25 6 5

In the five cases in which the magistrate returned the woman to her first husband, that outcome appears to have been the woman’s or her natal family’s preference. The six judgments that ratified the second marriage merit closer scrutiny. First of all, in none of these six cases did the wife herself raise any (recorded) objection. In two of these cases, she and her buyer had left the county and their whereabouts were unknown, making it impossible to summon them for trial; in one case, the sale had taken place seven years before and the woman had already given birth to two sons for the buyer; in one, the marriage had already lasted three years and the woman’s natal family was too poor to support her; in another case, the wife had been sold because of adultery, and the magistrate decided to ratify that sale even though the first husband had failed to seek the necessary authorization before taking such action; and, in the final case, the woman asked to be allowed to remain married to her buyer.

Chapter 11. Flexible Adjudication of Routine Cases

351

We can conclude that, if the natal family objected to the sale, then the magistrate would usually cancel it and return the wife to her natal family. In the exceptions, there were particular circumstances that seemed to justify other measures. Again we see that the attitude of the woman herself often seems to have had an important influence on his decision. What if neither the wife nor her natal family objected to the sale? As we know, the single most common reason wife-selling cases came to court was that the seller or his family demanded supplementary payments after the sale had taken place (43 percent of cases for which we have this information). In most of these cases, I find no evidence that the seller really wanted his wife back. How, then, did magistrates dispose of the wife? We have this information for seventy-four judgments: Returned to buyer, second marriage stands: Returned to natal family: Returned to first husband: Sold by official matchmaker:

34 (46%) 19 18 3

In eleven cases that went to court because of extortion by the seller or his family and ended in mediated settlements approved by the magistrate, we find the wife disposed of in the following manner: Returned to buyer, second marriage stands: Returned to first husband:

10 1

A similar pattern prevails in nine judgments of cases that went to court because of extortion by some other party: Returned to the buyer, second marriage stands: Returned to natal family:

6 3

The correlations are not absolute but nevertheless quite suggestive. It appears that magistrates were most inclined to ratify the marriage that resulted from an illegal wife sale if neither the wife nor her natal family objected to it and if the seller’s main goal was to get more money. In these circumstances, to leave the second marriage intact must have seemed the most expedient course of action. To sum up this section, it is clear that magistrates decided what to do with the sold wife on a case-by-case basis. They weighed the particular circumstances of each case—especially, who filed charges and why—when deciding the woman’s fate. Quite often, it seems, the decisive factor was the woman’s own wishes or those of her natal family, cast in the role of her protector. The Fate of the Money According to the statute against “buying or selling a divorce,” the magistrate was supposed to confiscate both brideprice and any matchmakers’ fees. But what did

352

part three

magistrates actually do? We have information about the fate of the brideprice in 153 cases: Kept by seller: Refunded to buyer: In 18 cases, seller ordered to refund money to buyer In 3 cases, natal family ordered to refund money to buyer in seller’s stead Confiscated: Awarded to natal family: Awarded to wife (to be dowry for remarriage):

121 (79%) 21

8 2 1

As we can see, magistrates almost never confiscated the brideprice, and in over three-quarters of judgments, they allowed sellers to keep the money they had received for their wives. Even in fraudulent wife sales—in which, according to statute, the unwitting buyer should be treated as a victim and have his money refunded—the seller nearly always kept the brideprice: I have eighteen judgments of fraudulent wife sales (that include this information), and in only two did the buyer receive a refund, whereas in sixteen the seller was allowed to keep the money. No doubt this lenience stemmed from pragmatism as much as sympathy, given that most sellers were destitute and had already spent the money. Moreover, in sixteen of the cases where the seller was allowed to keep the brideprice (including three fraudulent sales), the magistrate also ordered the buyer to pay the seller an additional sum of money in exchange for keeping the wife. These sixteen cases include judgments by fourteen different magistrates and more than one example from each of three counties: eight from Ba County, five from Nanbu County, and three from Baodi County. As Table 6 shows, these payments could be substantial, sometimes coming after the seller had already received at least one supplementary payment, and occasionally exceeding the original brideprice. In these cases, magistrates seem to have applied principles normally operating in conditional land sales, with the additional sum paid by the buyer amounting to a zhao tie payment that made the sale absolute and final. In two cases on this list, magistrates also helped sellers to recover shares of brideprice that unscrupulous matchmakers had appropriated. Similarly, in three cases not on this list, magistrates ordered either the woman’s natal family or the seller’s family to pay the seller an additional sum of money to ease his poverty. Separately, in three cases of compensated divorce ruled to be legal (and therefore not included in the table), magistrates also ordered buyers to pay the first husband extra compensation on top of the brideprice. In one of these cases, the first husband had already secured a supplementary payment of 12 taels (on top of the brideprice of 20 taels) before filing false charges to extort even more. The magistrate confirmed the legality of the divorce and the new marriage, refrained from

Chapter 11. Flexible Adjudication of Routine Cases

353

table 6 Final Payments to Sellers by Court Order (Illegal Sales Only) Year

County

Brideprice

Known supp. payments

Final payment

Seller beaten?

1764 1802 1815 1829 1833 1851 1851 1858 1866 1867 1868 1869 1872 1881 1895 1910

Ba Ba Ba Ba Nanbu Ba Ba Nanbu Baodi Ba Ba Nanbu Nanbu Baodi Baodi Nanbu

60 taels 12 taels amount unclear 12,000 cash 11,000 cash amount unclear 8000 cash 20,000 cash 90 strings “east cash” 15 taels & 1 string cash 12,000 cash 6000 cash 10,400 cash 46 strings cash 50 strings cash 15 strings cash

none none 1000 cash none none none 2000 cash none none none none 1200 cash 2600 cash none none 24 strings cash

amount unclear 6 taels 2000 cash 4000 cash 5000 cash 5000 cash 1000 cash 24,000 cash 10 strings “east cash” 4 strings cash 12,000 cash 5000 cash 20 strings cash 40 strings cash 60 strings cash 10 strings cash

unclear yes yes yes no no no no* no yes no no no no no** unclear

* The actual brideprice paid was 34,000 cash, but the matchmakers had given the seller only 20,000; the magistrate ordered the matchmakers to pay the seller the 14,000 cash they had withheld, and the buyer to pay him an additional 10,000 cash. (See discussion of this case in the text below.) ** The actual brideprice paid was 110 strings, but the matchmakers had given the buyer only 50 strings because that is what he had instructed them to charge; the magistrate ordered the matchmakers to pay the seller 40 more strings and the buyer to pay him 20, so that he ended up with a total of 110.

punishing anyone, and ordered the second husband to pay 10 more taels to the first husband in exchange for his pledge to demand no more.17 Such judgments represent extraordinary gestures of sympathy for the seller as loser in a wife sale. Let us take a closer look at one dramatic example of a magistrate ordering extra payment to the seller. In the 1858 case from Nanbu County listed in Table 6, landless peasant Xiang Gaoquan sold his wife, Feng Shi, because of poverty. The couple lived with Xiang’s father and elderly grandmother, and their marriage had been consummated for just one year. Feng Shi’s father agreed to the sale, but, at a meeting of Xiang Gaoquan’s lineage branch, three members of the Xiang lineage pledged in writing to help him financially, in return for his promise not to resort to wife sale. However, these relatives failed to deliver the promised aid, and after several months Gaoquan and his father arranged to sell Feng Shi as wife to widower Liang Xiding for 34,000 cash. Two other members of the Xiang lineage acted as matchmakers, and they insisted on keeping 14,000 cash from the brideprice to buy “wine and pork” to placate the lineage.18 But a quarrel ensued, and the three lineage members who had originally intervened ended up filing charges against the matchmakers over the “unfair” division of the 14,000 cash they had retained.

354

part three

The magistrate who judged this case expressed great sympathy for the seller and his wife, and scorn for everyone else involved. He ruled that the transaction was indeed an illegal “divorce sale” and canceled it. He ordered the buyer and matchmakers slapped; no one else received corporal punishment. But instead of ordering Feng Shi divorced, he returned her to the seller, Xiang Gaoquan; moreover, he awarded Xiang what amounted to punitive damages. To avoid the cangue, the buyer agreed to let Xiang keep the brideprice and to pay him an additional sum of 10,000 cash; the matchmakers agreed to pay Xiang the 14,000 cash they had taken out of the brideprice; and the three lineage members who had filed the lawsuit agreed to give Xiang one dou of husked rice (perhaps a week’s food supply for his family),19 one dry field, and one dry field plus one piece of paddy land, respectively. These creative and entirely extra-legal measures punished those who had exploited Xiang’s poverty but also aimed to relieve that poverty in order to prevent him from again resorting to wife sale. This judgment is a powerful example of the “father and mother official’s” paradigmatic pose as defender of the weak against the powerful.20 Some xingke tiben include records of the previous adjudication of wife sales as routine minor cases, before the major crimes that turned them into central cases had been committed. In two of these cases, magistrates awarded an extra final payment to the seller in exchange for ratifying the second marriage: a 1736 case from Nanpi County, Zhili, in which the magistrate ordered an additional payment of 8 taels, on top of the original brideprice of 4 taels and 1000 cash; and an 1886 case from Zhen’an County, Shaanxi, in which the magistrate ordered an additional payment of six strings of cash, which was the same amount as the original brideprice.21 Significantly, magistrates never ordered such payments in major cases being prepared for review up the chain of command—those cases they always judged strictly by the book. The only reason we catch a glimpse of such informal, flexible adjudication in these two xingke tiben is that the main focus of each is a homicide that took place after the wife sale had been adjudicated as a routine “minor matter.” Similar cases appear in published collections of individual magistrates’ judgments. For example, Li Jun reports a case of fraudulent wife sale in which he ordered the buyer to make a final payment of 20,000 cash, in addition to the brideprice of 90,000 cash and a previous supplementary payment of 4000 cash. Shen Yanqing (who administered three successive counties in Jiangxi during the Daoguang and Xianfeng eras) reports a case of open wife sale in which he ordered the seller to make a final payment of 12,000 cash on top of the brideprice of 26,000 cash (see the epigraph that opens this chapter). Both Li and Shen justified this largess by citing the abject poverty of the men who had sold their wives, and, in both cases, these magistrates allowed the second marriage to stand.22 These examples from diverse sources and regions of China show that it was not unusual for magistrates to mandate an additional, final payment to the seller, even if they did not do so in most cases.

Chapter 11. Flexible Adjudication of Routine Cases

355

If a case went to court because of extortion by the seller or his family, how did the magistrate dispose of the brideprice? Since the purpose of extortion was to obtain more money, how often did it succeed? We have information for sixty-six cases: Kept by seller: In 9 of these cases, the buyer was ordered to pay the seller extra Returned to buyer: In all 7 of these cases, the second marriage was canceled Confiscated:

56 (85%) 7 3

Corporal Punishment How often were the mandated beatings actually imposed? Here are the data for the 188 cases that went to judgment for which we have this information: Seller: Buyer: Matchmaker: Wife: Member of seller’s family: Member of woman’s natal family: Member of buyer’s family: Unrelated party: No one:

108 65 (in 6 other cases, the buyer had not yet been found) 61 28 8 7 4 12 33 (18%)

Thus, in 82 percent of cases that went to judgment, someone did get beaten, but magistrates did not impose punishment uniformly. Of the four protagonists in a wife sale singled out by the statute for punishment (seller, buyer, wife, and matchmaker), the seller was most likely to be beaten, and the wife was least likely to be beaten. In addition, the corporal punishment most often imposed was “slapping” (zhang ze), rather than the heavy bamboo mandated by the code, and the number of blows in practice was often less than the number prescribed by statute.23 Given this lack of uniformity, how common were different combinations of penalties? Here I focus on the four protagonists: Seller only: Seller, buyer, and matchmaker: Seller and matchmaker: Seller and buyer: Buyer only: Matchmaker only: Buyer and matchmaker:

46 cases 16 16 13 12 9 8

356

part three

Seller, buyer, wife, and matchmaker: Buyer and wife: Seller and wife: Wife only: Seller, wife, and matchmaker: Seller, buyer, and wife: Buyer, matchmaker, and wife: Wife and matchmaker:

7 5 4 4 3 3 1 1

As we can see, it was highly unusual for all four of the main parties in a sale to be punished as prescribed by statute. Instead, magistrates seem to have weighed the individual culpability of each party. The frequency of beatings imposed on the seller (and the high number of cases in which he alone was beaten) reflects the fact that so many cases came to court because of extortion and false charges by the seller.

T H E J U D G M E N T S O F G IO R O SI YA N G C I N G

To some extent, the variation seen in this overview may reflect the idiosyncrasies of individual judges—after all, the entire sample includes more than sixty magistrates. No doubt some were rigid and dogmatic while others were relatively flexible and pragmatic. It is striking, however, that the magistrates for whom we have at least a few judgments all conform to the pattern of the larger sample—namely, flexible adjudication on a case-by-case basis. As best as I can tell, therefore, flexible adjudication was the norm. The largest sample I have for a single magistrate is thirty-four cases of illegal wife sales handled by Gioro Siyangcing (Ch. Jueluo Xiangqing), who served two stints in Ba County for a total of six years (1849–50, 1852–56).24 Siyangcing was a Manchu of the Plain Blue Banner who won the presented scholar (jinshi) degree in 1835, and by the time he arrived in Ba County he had already served as magistrate of at least three other counties in Sichuan. Later he would be promoted to prefect of Ningyuan Prefecture in the same province.25 Therefore, Siyangcing arrived in Ba County as both an experienced magistrate and something of a Sichuan expert. (In this respect he was typical, because Ba County was too vital and challenging a post to entrust to a neophyte.) He also enjoyed the prestige of both his top examination degree and his pedigree as a member of a collateral branch of the imperial lineage, as indicated by his surname, “Gioro.” In other words, Siyangcing knew his job at least as well as most magistrates, and he was less likely than most to be intimidated by the local scene. Moreover, we can assume that he performed well, or he would not have returned to Ba County for a second term, much less won promotion.

Chapter 11. Flexible Adjudication of Routine Cases

357

Of Siyangcing’s thirty-four wife-selling cases, poverty motivated the sale in every one for which we know the motive (thirty-one cases). We have Siyangcing’s judgments in twenty-four cases, so let us focus on those. Here is how they break down: Fate of the wife:

Returns to natal family: Second marriage stands: Returned to first husband, even though transaction is complete: Sold by official matchmaker:

14 cases (58% of 24) 6 (25%) 3 1

Disposal of brideprice:

Kept by seller: Seller ordered to refund to buyer: Unclear (probably kept by seller): Confiscated:

20 cases 2 2 0

Who received a beating?

Seller: Buyer: Wife: Matchmaker(s): Other: No one:

13 cases 10 8 6 6 5

Siyangcing’s frequent citation of the statute against “buying or selling a divorce” in rescripts underscores his disapproval of wife sale, but he certainly did not adjudicate strictly by the code. He allowed the second marriage to stand in a quarter of his cases, and he confiscated the brideprice in not a single one (instead usually allowing the seller to keep it). In two cases, he ordered the buyer to make an extra, final payment to the seller in exchange for being allowed to keep the wife. He distributed beatings selectively and, as in the larger sample, the party most likely to be beaten was the seller. If we examine Siyangcing’s twenty-four judgments according to the reason each case came to court, we find a similar pattern. Here I focus on the fate of the wife: Why did cases come to court, and what was the fate of the wife for each category?

Extortion by seller or his family: In 5 cases, the wife is returned to her natal family In 4 cases, the second marriage stands In 1 case, the wife is returned to the seller

11 cases

358

part three

In 1 case, the wife is ordered sold by the official matchmaker Wife is unhappy: In 4 cases, the wife is returned to her natal family In 1 case, the wife is returned to the seller Natal family objects: In both cases, the wife is returned to natal family Extortion by natal family: The wife is returned to natal family Dispute unrelated to wife sale: The second marriage stands Other: In 1 case, the second marriage stands In 1 case, the wife is returned to her natal family In 1 case, the wife is returned to seller Unclear: The wife returned to natal family

5 cases

2 cases 1 case 1 case 3 cases

1 case

Two other cases that came to court because of extortion by the seller ended in mediated settlements that Siyangcing approved, allowing the second marriage to stand. As in the larger sample, the two most common reasons that wife-selling cases came before Siyangcing were extortion by the seller or his family (eleven cases), and objection by the wife or her natal family (seven cases). Generally speaking, Siyangcing handled these two categories of cases differently. He was most likely to allow the second marriage to stand when the case had come to court because of demands for supplementary payments. But he never did so if the wife or her natal family objected; instead, he would cancel the transaction and return her to her natal family or (in one case) to her first husband. In cases that came to court because of extortion by the seller or his family, Siyangcing exercised greater flexibility in disposing of the wife. In four of these cases, he simply let the second marriage stand. Let us take a closer look at two examples. In 1851, Jiang Xuekui and his wife, Pan Shi, found themselves in great difficulty because of poverty, so Jiang engaged a matchmaker to sell her to one Jiang Xingfa (no relation) in a fraudulent sale for 8000 cash. Some weeks later, Xuekui visited the buyer’s home in Chongqing, revealed that he was Pan Shi’s husband, and demanded 6000 cash more. A neighbor who worked as a runner at the circuit intendant’s yamen mediated the dispute, and, after some haggling, Xuekui agreed to accept 2000 cash in exchange for a new wife sale contract to replace the fraudulent original. Several days later, however, Xuekui again dropped by, invited Xingfa out for tea, and demanded 4000 cash (i.e., the balance of the 6000 cash he had previously demanded). This time Xingfa refused to pay, so that same night Xuekui came to Xingfa’s home,

Chapter 11. Flexible Adjudication of Routine Cases

359

caused a noisy scene in the street, and finally slashed his own throat with a knife and collapsed on Xingfa’s doorstep. This commotion aroused the neighbors, and soon the night watch arrived and arrested everyone. Xuekui’s wound proved superficial— his dramatic gesture had been intended mainly to attract attention. Siyangcing judged this transaction an illegal instance of “selling divorce” but opted to punish no one. Instead, he allowed the second marriage to stand and ordered buyer Xingfa to pay seller Xuekui a final sum of 1000 cash in exchange for a pledge to cause no more trouble.26 The second example, in 1853, involved an unemployed laborer in Chongqing named Zhang Zhengming who ran out of money, so he arranged to sell his wife, Li Shi, to another laborer named Zhao Tingfu. A couple of months later, Zhang had spent the entire brideprice, so he persuaded Zhao to pay him an unspecified additional sum in exchange for a submission contract promising to demand nothing more. Regardless of this promise, after just three weeks, Zhang demanded more money, and when Zhao refused, Zhang filed a plaint accusing him of abduction, adultery, and theft. In court, Zhao produced the wife sale contract and submission contract to disprove Zhang’s charges, which Zhang admitted were spurious. Siyangcing had both men slapped, and made them pledge to cause no more trouble. But he allowed the second marriage to stand, without bothering to summon either wife or matchmaker to testify.27 In these two cases, Siyangcing split the difference: no one got everything he wanted, but everyone got something. Both buyers were allowed to keep the brideprice and supplementary payments, while both sellers were allowed to keep the wives they had illegally purchased. In other details, however, the rulings differed. In the first, the magistrate punished no one and awarded what amounted to a zhao tie payment to the buyer, to forestall further demands for more money. But in the second, he ordered the buyer and seller both slapped. There is no certain way to account for these differences, but they do not seem arbitrary. Siyangcing clearly felt sympathy for the first buyer (perhaps because of his dramatic self-inflicted wound) that did not extend to the second (perhaps because of his false accusations). If we set aside the two main causes of wife-selling litigation and examine the three cases that came to court for “other reasons,” we find that Siyangcing ruled in three different ways, depending on the unique facts of each situation. In the first case, the wife and her buyer’s mother quarreled after the sale, prompting the mother to file charges against her son. Siyangcing canceled the transaction and ordered the wife returned to her natal family; he also ordered both buyer and wife slapped for having offended his mother. No one else was punished, and the seller kept the brideprice. In this instance, the buyer had arranged to purchase the wife against his mother’s wishes, and the magistrate obviously considered the new marriage untenable.28

360

part three

In the second case, the buyer made a down payment in exchange for the wife but failed to pay the balance, prompting the seller to file charges accusing him of abduction. Siyangcing canceled the transaction, ordering the wife returned to her first husband (who was required to refund the down payment), and had the buyer, seller, matchmaker, and wife all slapped for being untruthful in their initial testimony.29 In the third case, Siyangcing had previously authorized a poor man with an unhappy wife to return her to her natal family. But, instead of doing so, the husband had engaged a matchmaker to arrange a fraudulent sale. The buyer later resold the woman to a third husband—whereupon the matchmaker from the first, fraudulent transaction filed false charges against this third husband in an extortion attempt. Siyangcing left the wife with her third husband (according to her wishes) and punished no one for either sale. But he did order the matchmaker slapped for extortion and filing false charges.30 In these three decisions, Siyancing set aside the code in favor of solving the concrete problems of the litigants, applying sanctions selectively according to his own sense of different individuals’ degree of blame, and disposing of the wife in whatever manner seemed appropriate to the circumstances. Often, punishment focused less on the wife sale itself than on other misconduct that had brought the sale to his attention. To sum up, Siyangcing’s adjudication reflects what we have seen in the larger sample as a whole. Like other magistrates, he often admonished litigants that “both buying and selling of divorce are strictly prohibited by law” (mai xiu mai xiu jun gan li jin). Nevertheless, in practice he ruled flexibly on a case-by-case basis, apparently because he understood the social realities that framed wife-selling and was more concerned with addressing practical problems than with imposing the ideological dictates of the Qing code. The best way to explain Siyangcing’s judgments is to examine the specific reasons each case came to court. Many of his rulings imply sympathy for men driven by poverty to sell their wives, but also for the women themselves. I have far more judgments by Gioro Siyangcing than by any other individual magistrate (as a result of his unusually long service in Ba County, made possible because the rule of avoidance did not apply to bannermen), so a case study of such detail is possible only for him. However, for several other magistrates I have four or five judgments each, and these smaller samples confirm the pattern seen in Siyangcing’s cases, suggesting that he was typical in this respect.31 R OU T I N E A DJ U D IC AT IO N O F ZON G J IA N C A SE S I N BA C O U N T Y

If we examine the adjudication of routine cases involving either polyandry or the prostitution of a wife with her husband’s complicity, we find a pattern of case-by-

Chapter 11. Flexible Adjudication of Routine Cases

361

case flexibility similar to that seen in routine wife-selling cases. To recapitulate, the relevant statute prohibited a husband from “abetting or tolerating his wife or concubine to engage in illicit sexual relations with another man” (zongrong qiqie yu ren tong jian).32 (This phrase is often abbreviated in legal texts as “zong jian,” and for convenience I shall refer to these cases as “zong jian cases.”) Polyandry, polyamory, and marital prostitution all fell within the purview of this statute, which was cited in the judgment of all “major cases” involving these crimes. There are some important differences between wife-selling cases and zong jian cases, which may reflect differences between the legal standing of the practices they document. Perhaps the most important difference in their legal standing is that Qing law did permit some wife sales: for example, compensated divorce might well be legal, depending on the sequence of events, and the code explicitly authorized husbands to sell unchaste wives, as long as they first obtained a court order. Moreover, there were misgivings within the senior judiciary about how strictly the statute against “buying or selling a divorce” should be enforced with regard to wife sales motivated by poverty. Finally, in routine cases, magistrates would sometimes ratify the second marriage created by an illegal wife sale, as long as the principals themselves did not wish to cancel it. None of these things were true for the crimes covered by the category “zong jian,” which Qing law prohibited without exception after the Yongzheng reforms. In routine cases involving such crimes, magistrates never ratified contracts for prostitution or polyandry; they always made the parties pledge to quit and obey the law. Nevertheless, they did demonstrate great flexibility in how they enforced the law. The statute against zong jian mandated ninety blows of the heavy bamboo for the husband, the wife, and her sexual partner(s); if the wife were “coerced” (yile), then she would not be punished, but the husband and sexual partner(s) would receive one hundred blows and eighty blows, respectively. In either instance, the wife was to be divorced and returned to her natal family. To assess the adjudication of routine zong jian cases, we can repeat our exercise of comparing what the statute mandated with what magistrates actually did. Who, if anyone, received a beating? Was the wife ordered divorced and returned to her natal lineage, as the code required? I summarize below the outcomes in fifty-two routine cases from Ba County (dating from 1768 to 1864, including judgments of sixteen different magistrates), in which married couples were prosecuted for either polyandry or retail prostitution.33 This sample includes twenty-eight cases of retail prostitution outside a brothel (in which the husband acted as tout), fifteen cases of retail prostitution in brothels, and just four cases of stable polyandry with a single partner. (In the five remaining cases, the exact nature of the arrangement is unclear because the files are incomplete.) The four cases of stable polyandry took place in rural settings, whereas all but one of the retail prostitution cases took place in Chongqing. First of all, what became of the wives in these cases?

362

part three

Fate of wives in 52 routine cases:

Divorced: Returned to natal family for remarriage: 24 Pledges to become nun: 3 Handed over to official matchmaker for remarriage: 3 Authorized to live on her own: 1 Stays with husband: Couple ordered to leave town: 7 Returned to natal family’s custody for three years, after which husband may reclaim her if he can support her:

31 cases

21 cases

1

What stands out is that, in twenty-one cases (40 percent of the total), magistrates allowed couples to remain married in defiance of the code’s clear mandate. The following data show the fate of the wife categorized according to how the cases ended up in court: Fate of wives, by reason cases ended up in court (52 cases):

Wife or natal family files charges: Divorced: 27 Stays with husband: 9 Runners on patrol make arrests because of public altercation: Divorced: 1 Stays with husband: 6 Husband files charges: Divorced: 2 Stays with husband: 5 Other: Divorced: 1 Stays with husband: 1

36 cases

7 cases

7 cases

2 cases

More than two-thirds of the cases in this sample (thirty-six out of fifty-two) came to court because a wife or her natal family filed charges against her husband—and it is in these cases that magistrates were most likely to order the wives divorced. The reason is obvious: such charges reflected the woman’s unhappiness and her desire to stop engaging in sex work, and also (in many cases) her desire to escape the husband whom she had been supporting. As in the wife sale cases, magistrates evidently took women’s wishes into account in deciding whether to mandate divorce. It is significant that, in all but one of the thirty-six cases in which women or their natal families filed charges, the women were engaged in (or being pressured to engage in) retail sex work: twelve of these women had been placed in brothels by their husbands, and the other twenty-three worked out of rented lodgings, with

Chapter 11. Flexible Adjudication of Routine Cases

363

their husbands acting as touts. (Just one case out of thirty-six involved a steady polyandrous relationship.) From a woman’s perspective, retail prostitution seems to have been more distasteful and dangerous than a polyandrous relationship with a single partner. Moreover, nearly all of these cases involved poor rural couples who had come to Chongqing and ended up resorting to prostitution because they ran out of options. In such circumstances, women would have been socially isolated and vulnerable to abuse. No wonder they were unhappy. Nevertheless, even in these circumstances, not every wife wanted a divorce. No doubt some of these decisions were excruciatingly difficult. But divorce would entail separation from children as well as husbands, without necessarily guaranteeing a better fate. Moreover, the decision to engage in sex work was nobody’s ideal scenario, and the husband did not necessarily deserve the entire blame for that decision. At any rate, the fact that nine women did return to their husbands shows that magistrates did not rigidly enforce the code when judging these cases but instead considered a variety of factors. In contrast with my sample of routine wife-selling cases, only seven out of fiftytwo cases in the present sample came to court because the husband himself filed charges. As we have seen, the most common reason that wife sales ended up in court was that the first husband (i.e., the seller) would file false charges in an attempt to extort supplementary payments. In most cases involving polyandry or prostitution, that incentive did not exist, so there was no reason for a husband to file charges. The seven zong jian cases in which husbands did file charges are interesting exceptions. In three cases, the husbands had contracted their wives into prostitution for lump-sum payments and then filed false charges in an attempt to extort more money from the brothel-keepers. Here we see the same logic as when husbands demanded supplementary payments after selling their wives.34 In three other cases, the wife had run away with a client, and her husband filed charges against them in an attempt to recover his wife.35 In the seventh case, the wife’s brother had confronted her husband and taken her away from him, provoking the husband to file false charges in an effort to recover her.36 How often did magistrates order beatings in these cases, and who received them? Number of cases in which each party was punished (out of 52 total):

Husband: 42 Wife’s sexual partner(s): 11 Wife: 9 Landlord or brothel-keeper: 8 No one: 8 Once again, we see that magistrates exercised discretion. In eight cases, no one was punished at all. In most cases, the wives’ sexual partners did not appear in court

364

part three

(magistrates made little effort to arrest such men, especially when retail prostitution was involved). When magistrates did order beatings, the husband was far more likely to get one than his wife; as with wife sales, there seems to have been a tendency to put the onus on husbands despite the code’s treatment of wives as equally culpable offenders. In fact, the case files rarely record any explicit judgment as to whether a wife had been complicit or coerced, even though the zong jian statute draws a clear distinction between the two scenarios (a complicit wife was to be beaten). What specific penalties did magistrates actually impose in these cases? Penalties actually imposed:

On husband: Slapping: 13 Beating: 14 Beating plus cangue: 15 On wife’s sexual partner(s): Slapping: 3 Beating: 5 Beating plus cangue: 3 On wife: Slapping: 7 Beating: 2

42 cases

11 cases

9 cases

As we can see, discretion did not necessarily imply lenience. Rather, these tallies show that magistrates resorted to a sliding scale of penalties, including some that were actually more severe than those mandated by the zong jian statute—for example, forty “heavy blows” of the bamboo plus two months in the cangue, imposed on the husband in one case.37

C A SE ST U D I E S : R OU T I N E A DJ U D IC AT IO N O F ZON G J IA N

Judgments by individual magistrates provide a closer look at the kind of discretion exercised in routine cases of this kind. All of the following judgments assumed that prostitution and polyandry were crimes that should be suppressed, and every case ended with the defendants pledging to reform and obey the law. Within those limits, however, magistrates exhibited pragmatic flexibility and considerable creativity.38 Magistrate Dong Chun Shuts Down a Brothel One striking example is an 1813 case judged by Magistrate Dong Chun (a holder of the juren degree from Shandong later promoted to serve as Sichuan provincial treasurer)39 in which yamen runners raided a brothel in Chongqing. It is not clear

Chapter 11. Flexible Adjudication of Routine Cases

365

what prompted the raid, but the two men who ran the brothel confessed that they had “recruited women who came to Chongqing from various places, and had them sell sex as prostitutes.” Runners took twelve women into custody, of whom ten were married and two widowed. Five of the married women had been contracted to the brothel by their husbands; the other five had agreed to work in the brothel in order to make ends meet after losing touch with their husbands, who had left home in search of work and not returned. (Note that, because this case involved multiple couples, I did not include it in the quantitative summary above.) The five women placed in the brothel by their husbands are most pertinent to our inquiry. How did Magistrate Dong treat them? Only three were punished (the record does not specify if they were slapped or beaten), and it is not clear why they were singled out for this treatment. Only two of their five husbands appeared in court: both men were invalids and could not work (one was blind, the other’s legs disabled), and their wives were supporting them and their children. Neither husband was punished, and both were permitted to keep their wives in exchange for pledging to quit prostitution. There is no evidence of any effort to track down the three husbands who failed to appear in court, but failure to appear meant that they forfeited their wives, who were ordered divorced and returned to their natal families. (To execute this order required deporting one woman under escort to Fu Department, some eighty kilometers downriver from Chongqing.) Of the seven women who had entered the brothel on their own initiative (five married women, two widows), five were punished and all seven were ordered returned to their natal families. Presumably the two brothel-keepers were punished more severely, but the incomplete case file does not record their fate.40 To sum up, Magistrate Dong’s priorities in handling this case were to shut down the brothel, to place each woman in a household (by restoring two to their husbands and returning the others to natal families), and to secure pledges from all concerned that they would quit prostitution. He distributed beatings selectively, although there is not enough information to explain how he decided whom to punish. His priority was not to enforce the statute, except in the most basic sense of putting a stop to prostitution. Two Judgments by Yang Pei Two cases judged by Magistrate Yang Pei (a jinshi from the Han Martial YellowBordered Banner)41 illustrate the range of discretion exercised by a single magistrate. In one case, judged in 1835, He Tailin (32) was found to have made both his concubine, Zhou Shi, and his adopted daughter support him through prostitution; He had contracted both women to a brothel in Chongqing in exchange for half their earnings paid as a regular stipend. Eventually, the concubine’s widowed mother found out what was happening and filed charges against He and the brothel-keeper, Wang Gui.

366

part three

Magistrate Yang judged this case harshly: he sentenced both He Tailin and Wang Gui to forty blows of the heavy bamboo and a term in the cangue. He further ordered that the matchmakers for concubine Zhou Shi’s marriage be slapped for their negligence in failing to investigate He’s character. The women were not punished. Zhou Shi was returned to her natal family, and the adopted daughter (who had no natal family) was entrusted to the official matchmaker to be married off. After ten days in the cangue, brothel-keeper Wang Gui fell ill. Magistrate Yang ordered him held in custody while receiving medical treatment, to make sure he did not avoid serving the balance of his sentence once he recovered, but Wang’s prognosis worsened, so the magistrate finally ordered him released under guarantee of good behavior.42 Five years later, Yang Pei judged another case of marital prostitution in a very different way. Peasants Liu Zhengwei and Liu Zhu Shi (30) had been married in 1825, when she was aged 15 sui; in 1836, because of poverty, the couple left their village and moved to Chongqing (about sixty kilometers from their home village), where they lived on Liu Zhu Shi’s earnings from retail prostitution. This situation continued for four years, until Liu Zhu Shi’s brother Zhu Wenyuan tracked the couple down, found out what was happening, and took her to the yamen to file an oral complaint. Magistrate Yang ordered Liu Zhengwei slapped for “abetting his wife to sell sex as a prostitute” (zong qi mai chang) but refrained from punishing her or anyone else. In her testimony, Liu Zhu Shi complained that “my husband does not work at any proper occupation,” and she blamed his laziness for their poverty. Nevertheless, she expressed no desire to separate from him, and, instead of mandating divorce, the magistrate ordered the couple to return to the woman’s home village under the supervision of her brother, who pledged to let them live with him and to help support them. This order amounted to a modified version of return to the natal family, in which the woman was entrusted to their supervision without being divorced. It is not clear how Zhu felt about being assigned this responsibility, but he seems to have brought it upon himself by showing such an interest in his sister’s well-being. Liu pledged never again to make his wife work as a prostitute and agreed “to be beaten to death” if he violated this pledge. All things considered, he had gotten off lightly.43 Had these instances of marital prostitution been prosecuted as part of “major cases” reported up the chain of command for review, then we can be sure that Magistrate Yang would have applied the relevant statute to both in a rigid and uniform manner. But these were only “minor matters,” to be handled locally in the routine manner. The record provides no explanation why Yang Pei chose to handle these two cases so differently, but presumably his judgments rested on his sense of how much victimization was involved in each case and how likely it was that each

Chapter 11. Flexible Adjudication of Routine Cases

367

marriage could be salvaged. The attitude each woman expressed toward her husband probably played a decisive role. Three Routine Cases Judged by Gioro Siyangcing We are fortunate to have several cases judged by Gioro Siyancing that involved husbands’ complicity in their wives’ prostitution or polyandry. The following three are all routine minor cases that illustrate a range of approaches. In an 1850 example, peasant Gu Changsheng and his wife, Gu Liu Shi, had migrated to Chongqing after a year of marriage in search of work, and they ended up relying on retail prostitution to survive. After a year, Gu Liu Shi filed an oral complaint at the yamen against her husband. Apparently the couple had quarreled over some clothing of hers that her husband had sold without her permission, and, when she scolded him, he hit her. In the court hearing, she testified that she and her husband “get along well together” but that she wanted to stop working as a prostitute. Siyangcing ordered both husband and wife slapped: he for “abetting his wife to sell sex as a prostitute” (zong qi mai chang), and she for filing a complaint for “trivial reasons” (xi gu). No one else was arrested. The magistrate made the couple pledge to quit prostitution and released them.44 Two cases that Siyangcing judged the following year both involved brothel prostitution. The first concerned a peasant couple, Zheng Shifang and his wife, Zheng Zeng Shi, who had migrated to Chongqing, where Zheng had placed his wife in a brothel in exchange for a share of her earnings. Eventually, Zheng Zeng Shi’s father found out what was going on, and, after coming to Chongqing to confront her husband, he filed an oral complaint at the yamen. Siyangcing ordered Zheng Shifang slapped, but, in exchange for pledging to take his wife back to his home village and not force her into prostitution again, he was allowed to keep her. There is no evidence that Siyangcing tried to arrest the brothel-keeper or anyone else.45 Siyangcing handled the second case of brothel prostitution very differently. In this case, a Chongqing man named Liu Siyi (36) had a wife, Liu He Shi, as well as a concubine, Liu Cao Shi, both of whom had been supporting him through prostitution for a number of years. Over time, Liu He Shi had saved up a fair amount of jewelry and clothing of her own, separate from the earnings handed over to her husband, and she had also adopted a daughter named Fenglian. This family’s unorthodox domestic situation seems to have remained stable until Fenglian reached her teens, at which point Liu Siyi began pressuring her to do sex work as well. His behavior provoked a quarrel with Liu He Shi, who demanded that Liu “give her a divorce” (gei ta xiu tui) so that she could get out of prostitution and live separately with her adopted daughter. Liu refused, so finally she filed an oral complaint against him at the yamen. Siyangcing ordered Liu Siyi slapped for “forcing his wife to sell sex as a prostitute” but punished no one else. He granted Liu He Shi’s request for a divorce. But she

368

part three

was not returned to her natal family or handed over to the official matchmaker. Instead, Siyangcing awarded her half of the “clothing and gold jewelry that I have saved up on my own” (zi zhi yiwu jin shi), so that she could live independently without resorting to sex work. He further ordered that, when her adoptive daughter Fenglian reached marriage age, half of the girl’s brideprice be awarded to Liu He Shi. It is remarkable that Liu Siyi, despite having pimped his wife and concubine over a period of years, was let off so lightly. In addition, he was allowed to keep custody of his concubine (even though she confirmed in testimony that she was also a prostitute), perhaps because she made no complaint against him, as well as the adoptive daughter Fenglian, whom the magistrate treated as his. With these two young women remaining in Liu’s custody, the risk of recidivism must have been high, his pledge of good conduct notwithstanding. Moreover, in granting Liu half of his wife’s savings and of Fenglian’s future brideprice, the magistrate acted as if these resources constituted a sort of community property to be divided between divorcing spouses—this extraordinary measure had no basis whatsoever in codified law and appears to have been an innovation for this case alone.46 Siyangcing provided no explanation, but clearly this narrow ruling was designed to address Liu He Shi’s specific grievance without meddling in anything else.47 Taken together, these judgments by Dong Chun, Yang Pei, and Gioro Siyangcing suggest that the prohibition of zong jian was enforced only in response to particular complaints, and that magistrates hesitated to order that the women involved be divorced unless they themselves made this specific demand or their husbands failed to appear in court. The relevant statute played only a minor role in their decisions. The case files are short, because almost none of these judgments required more than a single hearing. Magistrates avoided expanding a case beyond the limited scope indicated by the reason it had ended up in court. Moreover, I find no evidence of any effort to follow up on the couples that remained together, in order to check their compliance with the ban on prostitution. A Major Criminal Case that Siyangcing Judged Strictly by the Code This pragmatic flexibility notwithstanding, Ba County magistrates were perfectly capable of ruling strictly by the code when circumstances warranted, as seen in a major criminal case involving polyandry that Siyangcing reported to his superiors for review in 1853. In this case (which I do not include in the sample of fifty-two routine zong jian cases summarized above), peasant Yang Chaoju (39) and his wife, Zhu Shi (22), had two little sons and were very poor. To supplement their subsistence, they came to an agreement with Yang’s friend, Hu Man (22), in which Hu supplied them with rice and cash in exchange for sexual relations with Zhu Shi. According to Zhu Shi’s testimony, she and her husband got along very well with each other and also with Hu—there was no inhibition among them, although they

Chapter 11. Flexible Adjudication of Routine Cases

369

tried to keep their arrangement secret from others. After five months, however, Yang Chaoju’s father found out and confronted him in a row that alerted the couple’s neighbors to what was going on; he threatened to turn the three of them in to the authorities for prosecution. After this, Yang Chaoju stated before witnesses that he “could not live with such shame,” and that night he hanged himself. Yang Chaoju’s death fell within the purview of the Qing code’s laws against “forcibly causing someone’s death” (wei bi ren zhi si), which extended culpability for homicide to include some scenarios of suicide. According to the relevant substatute, if a husband committed suicide out of “shame and indignation” (xiu fen) upon discovering his wife’s adultery, then she and her partner would be liable for his death: she would be sentenced to strangulation after the assizes, and he to one hundred blows of the heavy bamboo and three years of penal servitude. But if the husband had “abetted or tolerated” (zongrong) his wife’s illicit sexual relations, and later “those illicit sexual relations are publicly exposed, provoking him to commit suicide out of shame and embarrassment,” then his wife and her partner “shall be liable only for their illicit sexual relations” and not for his death. The logic here is that, in the second scenario, the husband himself was to blame for the “shame and embarrassment” that had provoked his self-destruction.48 Because this case might involve severe penalties that would require review, Siyangcing handled it strictly by the book. As soon as he learned that the suicide had been provoked by adultery, he followed protocol by alerting his superiors, who ordered a rigorous investigation and adjudication by the code. However, at the inquest, witnesses confirmed that Yang Chaoju had “abetted and tolerated” his wife’s extramarital relationship and then committed suicide out of shame for his own actions. Therefore, citing the second clause of the substatute, Siyangcing provisionally cleared Zhu Shi and Hu Man of culpability for Yang’s suicide and sentenced them only for illicit sex. His formal report conveying these recommendations employs the format and language found in xingke tiben. The prefect, provincial judge, and governor general all approved Siyangcing’s report, and Zhu Shi and Hu Man were punished accordingly, after which she was returned to her natal lineage.49 The contrast between Siyancing’s routine cases and this major criminal case highlights the two very different modes of adjudication practiced by local magistrates, depending on whether they expected their decisions to be scrutinized by their superiors. In routine cases that involved only “minor matters,” codified law was just one of the factors that guided Siyangcing and his colleagues, and not necessarily the most important one. In such cases, their priority was to address specific, concrete problems that had brought cases to court so as to achieve quick and efficient resolutions. But when a “major case” had to be prepared for review, pragmatics yielded to strict and deferential enforcement of the code.

370

part three E X P L A NAT IO N S O F L O C A L A DJ U D IC AT IO N I N T H E Q I N G

What are the implications of my findings for the broader field of Chinese legal history? There has been intense debate among historians about local court adjudication in the Qing, and one purpose of this chapter is to offer new evidence for their consideration. The most controversial question under debate is deceptively simple: on what basis did magistrates decide routine cases? Several hypotheses can be reassessed in light of my evidence. For the sake of space, I shall focus on those of two scholars who have made major contributions in this field, Philip Huang and Kishimoto Mio. Philip Huang’s “Adjudication by the Code” The basic contradiction between codified law and social practice seen in wifeselling cases stands in sharp contrast with most of the subject areas that Philip Huang covers in his path-breaking 1996 study of the Qing “civil justice” system. Huang divides the Qing system into informal and formal realms: in the informal realm outside the courtroom, community-level mediation and compromise dominated, but, once a dispute reached the formal realm of a court hearing, “magisterial adjudication was governed above all by codified law.”50 In other words, adjudication constituted an expression of power, not an act of conciliation: the state imposed its will on litigants, whether they liked it or not. Huang depicts an effective integration of informal and formal realms in a coherent system of “civil justice” that worked remarkably well to solve problems and enforce the law. In most of the cases Huang examines—involving routine disputes related to land, inheritance, marriage, and debt—there were no sharp contradictions between the Qing code and normal social practice. It is not surprising, then, that he can identify a high degree of coherence between implicit principles in the code and implicit principles guiding judgments. On this basis, Huang argues that, when cases reached a formal court hearing, Qing magistrates nearly always “adjudicated unequivocally by the code.”51 This thesis implies a strong imperial center that could impose its will through local courts. But Huang concedes that magistrates almost never cited the code or referred to it in any explicit way, nor did they impose the penalties prescribed by the statutes they were supposedly applying. Hence his need to deduce the implicit basis for these judgments: My ordering [of cases] by statute is based almost wholly on my own interpretation of what laws obtained, not on the texts of the magistrates’ judgments. . . . My argument here is simply that even in the absence of specific citations, a close reading of magisterial judgments in conjunction with the code leaves no doubt about their basis in law. The relevant statutes are implicit but obvious in virtually all these judgments.52

Chapter 11. Flexible Adjudication of Routine Cases

371

Huang has sufficient confidence in his powers of interpretation that he organizes his cases according to the “implicit but obvious” statutes that magistrates supposedly applied when judging them.53 But it is Huang who has drawn the connections between statutes and judgments, not the magistrates who actually judged these cases. Seen in that light, the evidence for Huang’s claim hardly seems “unequivocal.” In fact, it seems to me, by their reticence these magistrates actually left ambiguous exactly what they were enforcing: social norms or the Qing code, or perhaps both at the same time. In other words, the coherence between social norms and the code in the kinds of cases that Huang examines meant that magistrates did not have to make an explicit choice between the two, leaving unanswered the question of how far the state could really impose its will and influence behavior through the local courts. An obvious example is the principle of “equal division between sons” that guided household division and inheritance. This principle was indeed enshrined in the Qing code, but it was also universal customary practice among Han Chinese throughout the late imperial era.54 If a magistrate enforced this principle in a specific legal case without making any reference to codified law, it is by no means obvious that he had adjudicated that case “by the code.” More likely, he had simply enforced a widely shared norm that was taken for granted by everyone in his courtroom, although the fact that this norm did not contradict the code may have made his task easier. The same might be said for judgments that required borrowers to pay debts, tenants to pay rents, sons to care for aging parents, and so on. As this chapter has shown, moreover, different kinds of cases reveal a markedly different picture, casting doubt on the ability of the imperial center to impose its will on society through the local courts. My own sample of routine cases shows adjudication operating according to very different principles than Huang infers from his cases: the code was just one factor in how magistrates judged routine cases involving wife sales, and by no means the most important one. Another important contrast with Huang’s sample is that, when confronted with illegal wife sales, magistrates often did explicitly cite the code—that is, the statute against “buying or selling a divorce”—even when they chose not to enforce it. Magistrates’ frequent invocations of that statute to admonish litigants suggest they could not take for granted that litigants understood such transactions to be criminal, much less that litigants accepted the norms informing criminalization. The magistrates who had to deal with these cases were certainly aware of the sharp contradiction between the ideological mandates of the code and the ubiquitous social phenomenon of poverty-driven wife sales.55 Kishimoto Mio’s “Compromise through Balance” In contrast with Huang, Kishimoto Mio argues that local magistrates aimed above all to achieve social compromise through “balance” and were guided less by the

372

part three

formal mandates of the code than by the imperative to “protect the weak and punish the wicked” while trying to prevent further disputes.56 She developed this thesis in part through her own important article on wife-selling in the Ming-Qing era, which depends on published sources such as magistrates’ handbooks and collected judgments (rather than legal archives)—sources that are several steps removed from the actual cases they discuss. When judging wife-selling cases, she argues, magistrates looked at particular circumstances, especially the feelings and economic condition of the individuals involved, and when wife sales were motivated by poverty, the transaction was usually allowed to stand; in fact, she argues, judicial practice had the effect of decriminalizing wife sales, as long as they were motivated by poverty. To sum up, the guiding principle for local adjudication was to balance “human feelings and circumstances” (renqing) with “the formal law of the state” (guo fa), but, in contrast with Huang, Kishimoto gives far greater weight to the former than to the latter.57 Kishimoto’s concept of “balance” strikes me as a basically accurate characterization of how Qing magistrates responded flexibly to individual cases, at least in my sample of routine wife-selling cases. Magistrates do seem to have weighed formal law against human feeling and specific circumstances, much as she suggests, and often the latter trumped the former. They certainly did not systematically enforce the code, although it is obvious that they were familiar with the relevant statute. Only in “major cases” prepared for review, which always quote statutes verbatim and apply them to the letter, can we find magistrates judging wife sales “unequivocally by the code.” If we want to borrow a concept from the Western legal tradition to approximate magistrates’ flexible approach to these routine cases, it might be “equity.” In England, judges cited the principle of equity in order to soften the harshness of abstract formal law by taking into account the mitigating circumstances found in actual cases. Equity originated with appeals to the king, who had the power to commute sentences as an act of grace. Later, this power was delegated and institutionalized in a separate system of “equity courts” that provided redress against the undue harshness of Common Law court rulings.58 We see something like the English concept of equity at work in routine wife-selling cases, when (for example) a magistrate would let the seller keep the brideprice, even when ordering him slapped. We can also see something like equity in the efforts of some senior officials (described in the previous chapter) to soften the penalties for wife sales motivated by poverty. But Kishimoto’s argument captures only part of the picture that emerges from actual legal archives, perhaps because of the bias of her sources. As Huang has cautioned, published handbooks and case collections tend to reflect Qing officials’ idealized representations of themselves, in marked contrast with what one finds in real case files from the archives.59 In particular, there is no evidence that magis-

Chapter 11. Flexible Adjudication of Routine Cases

373

trates effectively decriminalized wife sales that were motivated by poverty. On the contrary, in the great majority of my routine cases, someone ended up receiving a beating, and, even though poverty motivated the vast majority of these wife sales, the transactions were canceled more often than not.60 The most common decision was to return the wife to her natal family; only a third of judgments allowed the second marriage to stand. The seller’s poverty was not the decisive factor in how these cases were judged, because nearly all sellers were poor, and magistrates took their poverty for granted. Kishimoto sees magistrates’ efforts to achieve compromise through balance as the definitive characteristic of a coherent system of local justice; this is how the system was supposed to work. Here we have a strong contrast with Huang’s 1996 thesis that strict adjudication by the code was the formal system’s definitive characteristic. Since then, however, Huang himself appears to have backed away from the “unequivocal adjudication by the code” formula he once so forcefully espoused: although he nowhere explicitly rejects that formula, his subsequent writings have qualified it to a considerable degree. In his 2001 book, for example, he states: Where there was congruency between code and custom, court actions may be mainly a matter of applying the letter of the law. . . . Where there were continual tensions or outright opposition between code and custom, legal practice could follow a number of different patterns. . . . [The courts] could follow the code in suppressing custom . . . or they could accommodate social practice.61

This strikes me as a dramatic departure from his earlier position: now, apparently, magistrates were not bound by the code at all. Moreover, Huang’s most recent characterization of Qing adjudication begins to approach Kishimoto’s idea of “balance”: [W]hen 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 such cases at court in accordance with the law (without excluding, of course, appropriate considerations of qingli, or human compassion/relations and moral principles).62

If the phrase “in accordance with the law” in this passage is intended to mean the same thing as “adjudicated by the code,” then the qualification added in parentheses opens a big loophole for exceptions to that claim. But Huang does not explain how magistrates would determine the “appropriate” balance between the law and qingli in any given case. An Emerging Consensus? The trajectory of Huang’s thinking suggests a gradual convergence of previously irreconcilable views. The emerging consensus would seem to be something like the following. As Huang insists, magistrates did not themselves directly “mediate”

374

part three

between litigants, although, as we have seen, yamen personnel did act in this capacity, and magistrates would approve just about any mediated settlement that litigants reached prior to a formal hearing. But when magistrates judged routine cases, the code was only one of the factors they took into account. They had to balance the dictates of the code against at least two other factors: considerations of renqing (i.e., human feelings, compassion, and extenuating circumstances), and the concrete social realities that framed a given case, including customary norms and widespread practices that might contradict the formal law of the state. Moreover, expediency appears to have been an important factor, the priority being to clear cases quickly and in a manner calculated to avoid having litigants return to court. A magistrate’s job was easiest when local customs agreed with the norms embodied in the code, because then he could enforce both at the same time without choosing between them. But when tenacious social practice contradicted the code, or there were significant mitigating circumstances, magistrates usually set aside formal law in favor of pragmatic accommodation—and they would typically justify this flexibility in terms of compassion and noblesse oblige, thereby diverting attention from the intractable practical pressures they faced. In effect, magistrates judged flexibly on a case-by-case basis, balancing and adapting these different factors as appropriate to the concrete circumstances of individual cases. As a result, even when magistrates did cite the code (as they often did in wife-selling cases), they rarely enforced it to the letter. This is the picture that emerges from my own study of how magistrates judged routine cases involving illegal wife sales. It is closer to Kishimoto’s position than to the one espoused by Huang’s 1996 book on Qing civil justice, but Huang’s subsequent comments imply movement in this direction. T WO M O D E S O F A DJ U D I C AT I O N

A comparison of routine county cases with the major cases recorded in xingke tiben calls attention to the vast gulf that separated the 1,500 or so courts of first instance from the apex of the judiciary at the imperial capital. But that comparison also highlights the two very different modes of adjudication practiced in those courts of first instance. What stands out is the striking contrast between how local magistrates handled routine “minor matters” that stayed local, and how the same magistrates prepared the exceptional “major cases” destined for central review. In the routine mode of adjudication, the decisive factor was the concrete reason a given case came to court, and, in this mode, magistrates acted flexibly and expediently to solve practical problems. In non-routine adjudication, however, the decisive factor was the Qing code, and magistrates applied its protocols rigidly and with great deference to their superiors.

Chapter 11. Flexible Adjudication of Routine Cases

375

If only a small percentage of wife sales ever came to official attention in the form of a routine court case, then a far, far smaller percentage ended up being reported in xingke tiben as part of a capital case. For example, I have 248 routine cases involving wife sales from Ba County, and it is obvious that the practice was not unusual there, but I have yet to find a single xingke tiben from Ba County that involves a wife sale. In other words, the mode of adjudication seen in xingke tiben was highly exceptional. The vast majority of wife sales that came to court constituted “minor matters” to be handled in the routine, flexible manner seen in this chapter.

Conclusion

T H E P E R SP E C T I V E F R OM T H E M A R G I N S

Through an examination of wife sale, polyandry, and related survival strategies, I have sought to provide a vivid picture of life on the margins and to recast the marriage system of Qing dynasty China from the perspective of the rural poor. From this perspective, I argue, the normative boundaries between marriage and trafficking, and between marriage and sex work, cannot be sustained; instead, these putative boundaries are exposed as ideological constructs that supported the privileges and pretensions of the elite. Moreover, the marriage system as a whole depended on trafficking in order to function and survive—and the practices I have documented here constituted an integral part of that system, their stigma and criminalization notwithstanding. The continuum of practice that linked marriage and the traffic in women shaped household formation for rich and poor alike, albeit in different ways. One aspect of this continuum was the prevalence of brideprice-heavy marriage among the poor, which was sometimes bluntly termed maimai hunyin (“buying/selling marriage”). Granted, if one looks at the marriage system overall, there was some regional variation in the incidence of brideprice-heavy versus dowry-heavy marriage, and a number of factors influenced their distribution.1 But there is no question that class and wealth played a major role in shaping these practices, as becomes obvious if one looks at the extreme ends of the socioeconomic scale. For the wealthy and upwardly mobile, dowry was a valuable status symbol precisely because it served to demarcate rich from poor, and, within elite households, it distinguished a main wife from the subordinate women who were also sexually 376

Conclusion

377

available to her husband. For an elite wife, brideprice-heavy marriage was simply unthinkable, and the same was true of widow remarriage and any other form of marital separation. But concubines and maidservants were purchased on the market. (In fact, as we have seen, the low end of the market for concubines overlapped with the high end of the market for sold wives.) Thus, the elite practice of polygyny involved both extremes of the dowry/brideprice scale: a lavish dowry marked the main wife as unique and also as herself a member of the elite, whereas the lowerranked women of the inner quarters were bought at the going rate in straightforward cash transactions. In effect, polygyny brought class exploitation inside the family (just as polyandry brought sex work inside the family). Moreover, it was standard for a maidservant who had reached a certain age without being “favored” by her master to be sold in marriage, thereby recycling her back into the market.2 For the poor, a dowry of significant value was unaffordable. Instead, their usual form of marriage was to sell a daughter to her in-laws, with her brideprice understood as compensation for the cost of raising her. Marital separation among the poor also took the form of wife sale, either direct or indirect. Similarly, widow remarriage was nearly universal among the poor, for whom it took the form of wife sale (the uxorilocal version being found only in households with significant assets). A closely related theme is the continuum of marriage and sex work among the poor: polyandry, polyamory, marital prostitution, and conditional wife sale all involved a husband sharing his wife with one or more other men in order to help support their family. All of these strategies combined marriage and sex work in different proportions, blurring the ostensibly clear distinction between them. Whereas wife sale was a strategy to survive by breaking up the family (in the process creating a new marriage), these were strategies to survive together as a family by tapping the exchange value of a woman’s sexual and reproductive labor. In this context, sex work was not the antithesis of marriage but rather a practice that complemented and supported marriage by enabling married couples to stay together. Indeed, formal polyandry, which itself should be seen as a form of marriage, turned sex work into domestic labor performed within the family. Wife sale and polyandry were but a subset of a larger field of practice that partook of the same logic, namely commodification of women’s sexual and reproductive labor. This field of practice prevailed among the very poor, who stripped it of all euphemism, but it extended all the way across the socioeconomic scale into the inner quarters of the elite. Elite families benefited by gaining access to women, even as the poor relied upon trafficking to survive. This commodification was inextricably linked to the sex ratio imbalance that continues to distort all age cohorts in China today. Under the circumstances, it is no surprise that human trafficking has staged a comeback, along with a ubiquitous sex industry, a new form of concubinage enjoyed as a status symbol by elite men, and, in at least some rural

378

Conclusion

areas, brideprice-heavy marriage. The persistence of imbalanced sex ratios and the traffic in women rank among the most difficult legacies of the late imperial era for China today. The vital role of the traffic in women in the social and biological reproduction of the elite underscores the double standard of the Qing dynasty’s normative prescriptions. Polyandry and wife sale turn out to be part of the same paradigm as elite polygyny. Why, then, criminalize polyandry, when polygyny was a prestigious status symbol? Why criminalize the sale of a wife, when the sale of daughters (and their purchase to be concubines and maidservants) was perfectly legal and taken for granted? Individual consent was not a value here: the permitted sale of women into polygyny and bonded servitude did not require their consent (only that of their parents). In contrast, polyandry and wife sale generally did require a woman’s consent in order to succeed, but this fact did nothing to mitigate their prohibition. The selective nature of prohibition and stigma upheld elite lifestyle and, as a consequence, was quite arbitrary from the standpoint of ordinary people. In effect, the prohibition of these survival strategies amounted to the criminalization of poverty. To summarize, if we focus on survival strategies among the poor (instead of normative prescriptions of the elite), we find that the binary distinctions between marriage and trafficking, and between marriage and sex work, simply cannot be sustained. The perspective from the margins changes our understanding of the whole because what at first glance may appear to be marginal is revealed as central and fundamental to the Chinese system of marriage and household formation as a whole. G E N D E R A N D T H E L I M I T S O F AG E N C Y

As a practical matter, it was difficult to make these strategies work without at least the grudging cooperation of the wife. Therefore, much of this system depended on the consent and sometimes the active participation of women. As I have shown, some women took the initiative to recruit polyandrous partners, some provoked their own sale, and still others intervened to prevent sales they did not want. The evidence shows that a couple’s choice of which particular strategy to undertake might well depend on the nature of the wife’s relationship with her husband and whether she wished to stay with him or to leave him. Moreover, “being sold” often meant a substantial and desirable improvement in a wife’s standard of living. In the preceding chapters, I have occasionally pointed out very limited parallels between Chinese wife sales and the slave trade in the antebellum American South. But it is worth emphasizing the differences between the two, which are far more significant. Not all sales of people are the same, and not all sold people are slaves. In China, even domestic servile laborers, who were slaves of a sort, lived under conditions dramatically different, and usually far better, than those of African American slaves, who represent possibly the most extreme, absolute, and cruel

Conclusion

379

version of human trafficking ever known. In a wife sale, the husband’s prerogatives and the wife’s duties were transferred from seller to buyer, but the woman became a wife, not a slave, and her children would not become slaves. Most important, with few exceptions wife sales were illegal, which meant that a woman or her natal family who objected could go to court—and, as we have seen, many did. But this is not a happy story, and we should not lose sight of the fundamentally exploitative nature of the practices I have documented. For example, although polyandry and polyamory among the poor represented a sort of mirror image of polygyny among the elite, and there are significant parallels between them, the implications of having more than one spouse were very different for women than for men. With polygyny, one man enjoyed being served by many women, in conditions of luxury, but polyandry and polyamory did not represent a simple reversal of that formula. To be sure, in some cases a woman became the de facto head of her extended household by making herself the indispensable center of a web of relations between men who deferred to her and worked to support her. But probably a more frequent result was for one woman to end up having to serve many men, sexually and otherwise, and certainly not in conditions of luxury. Polyandry and polyamory should be understood as extensions of women’s marketable labor in conditions of agricultural involution, in which people had to work harder and harder to maintain subsistence, at the cost of diminishing returns. This kind of domesticated sex work was one more reflection of how poverty forced the rural poor into the marketplace. There is also the tricky question of stigma and shame: clearly, some women did not care about these things at all, but others invested marriage with values of propriety and respectability that were painful to sacrifice, even if survival required it. Under these circumstances, the exercise of agency often amounted to little more than choosing the least bad option when no good ones were available—“making lemonade,” as it were. Moreover, so many of the women in these legal cases who asserted themselves and their interests nevertheless seem to have ended up as victims, one way or another. We have read case after case in which women were beaten—not necessarily into submission, but beaten and battered all the same. Even if we recognize that the incidence of violence—especially murder—is grossly exaggerated by the sampling bias inherent to legal cases, it is hard to avoid the conclusion that wife beating, at least, was widespread and normalized. Indeed, as is well known, the Qing code explicitly authorized a husband to beat his wife as long as he broke no bones or inflicted worse injury. Wives who resisted being sold did not necessarily love their husbands, who might well be abusive, but they dreaded being forcibly separated from children. The wives who actively sought to be sold were usually those who had no children, or who could take their children with them. This was not a happy story for men, either. In every scenario I have presented, poverty undermined normative masculinity, which for peasants depended above

380

Conclusion

all on having land, a wife, and sons—that is, patriarchy stripped to its essentials. For poor men in a context of inordinately high sex ratios, masculinity proved to be a zero-sum game, in which husbands driven by poverty to share or sell their wives ended up as the losers. The husband who sold his wife would trade places with her buyer, becoming a “bare stick” in his place—an object of pity and contempt. The husband who shared his wife with other men had to surrender the sexual monopoly that, by giving him exclusive access to his wife’s bed and control over her children’s paternity, had defined him as a husband. By making room for other men in his home and in his wife’s bed, a man yielded his fundamental patriarchal roles as husband, father, and head of household. It is obvious that these sacrifices created enormous stress for many of the men who made them. Even if we discount for the sampling bias of legal cases, it is hard to avoid the conclusion that the violence found in these sources is an extreme but accurate reflection of the tensions inherent in such non-patriarchal arrangements. Every system of marriage has its tensions, along with rules and rituals designed to contain them. Chosen kinship, and the contracts and rituals that framed “getting a husband to support a husband,” were strategies to contain the tensions generated by polyandry. In particular, sworn brotherhood offered an alternative vision of masculinity that justified the husband’s sacrifice of patriarchal prerogatives in favor of an idealized fraternal solidarity—a solidarity based on relative contempt for women. As the hero Liu Bei famously declared to his sworn brother, Zhang Fei, “brothers are like hands and feet, whereas a wife is like clothing”—in other words, brothers are fundamental and irreplaceable, whereas a wife is something external that can easily be taken off and changed.3 This ideology of marginalized masculinity was a powerful counter-cultural force that deserves more focused study in its own right. Celebrated in folklore and popular fiction, it provided the organizational basis for Triads, bandits, pirates, and mafia; it also framed long-term same-sex relationships among marginalized males outside the normative family system, as well as polyandry.4 Nevertheless, as many legal cases show, fraternal solidarity did not always suffice to overcome the tensions inherent in polyandry. Unless the first husband swallowed his pride and accepted a subordinate role, it was difficult to sustain these relationships over the long term. T H E I N F O R M A L R E A L M O F I L L IC I T C U S T OM A RY P R AC T IC E

The evidence presented in Chapter 11 about the adjudication of routine wife-selling cases illuminates standard practice in the local courts of the Qing dynasty. But that is not the whole story.5 This evidence also highlights the dysfunctional aspects of the Qing judicial system and the tenuousness of local control. In other words,

Conclusion

381

flexible ad hoc solutions to particular cases reflected the state’s failure to project power, to solve intractable problems, and to reform social practice. The consistent tension between absolutist ideology and the more charitable and pragmatic impulses of some senior jurists also reflected this failure. Wife sale and polyandry represented attempts to solve problems for which the state provided no solution. Under the circumstances, people had to devise their own solutions and avoid official interference as best they could. A number of other kinds of widespread social practice also contradicted the mandates of the Qing code—and, in this respect, the subject matter of this book represents but a subset of a much larger field of illicit but customary practice. Salt “smuggling” (i.e., production and sale outside the state monopoly), illegal migration to the frontiers, and the formation of collective brotherhoods all partook of similar logic, but probably the most widespread field for illicit custom was transactions in land. Several examples come to mind. One is the use of “white contracts” in land sales, in which the parties evaded the transaction tax and so failed to register sales and obtain the official red-inked seal on their contracts (those that bore the seal were known as “red contracts”).6 Another is the sale of banner or noble land in Manchuria to Han immigrants, which required systematic falsification of contracts (similar subterfuges facilitated the illegal alienation of native land in Yunnan, Taiwan, and other frontier zones).7 A third example is the custom of demanding supplementary payments after land sales whenever land values rose or the seller (or his descendants) fell on hard times.8 All of these practices were prohibited by the Qing. For that reason, their efficacy depended on community recognition and enforcement mechanisms that operated apart from and in defiance of the official system. These illicit transactions expose a subterranean field of customary practice that flourished according to its own standards and logic, in defiance of formal law and courts. In contrast with Huang’s schema, this informal realm highlights the dysfunctional aspects of the Qing system: it stood in conflict with the formal realm, instead of complementing it as part of a coherent system with the magistrate at its apex. Its existence implies an alternative set of values and more-or-less conscious resistance to the state. At best, this illicit informal realm might strengthen community solidarity and resistance to what were seen as oppressive intrusions of state power.9 At worst, the resort to illicit practices rendered people vulnerable to the predations of local thugs and mafia. As Diego Gambetta has shown, the classic Sicilian mafia originated as local strongmen who provided private protection for illegal transactions—including very mundane ones, like selling cattle outside the official market to avoid paying taxes (reminiscent of the use of white contracts for land sales in China). Over time, their protection services expanded and coalesced into a parallel system of authority in defiance of the state. We find something similar in, for example, the ways that Triads organized for mutual aid and criminal entrepreneurship in

382

Conclusion

southeastern China.10 Historians have long noted the notorious rise of “local bullies and evil gentry” (tuhao lie shen), along with the proliferation of secret societies and mafias, in the late Qing and Republican eras. Their rise reflects the triumph of the illicit informal realm over the formal one, when imperial control weakened and the dynasty’s institutions broke down.11 Compounding these problems was the Qing policy of keeping its formal bureaucracy at constant size, despite dramatic increases in population and in socioeconomic complexity. Inevitably, it was the local yamen staff (not subject to the rule of avoidance) that expanded instead. As a result, effective administrative power gradually devolved from the imperial center to the local level, and from the degree-holding officials (all of whom were short-term sojourners in their jurisdictions) into the hands of the long-serving clerks and runners. Bradly Reed characterizes the latter as “illicit bureaucrats,” because they were hired in excess of futile statutory limits and operated according to their own self-regulated “customary law.”12 This inexorable devolution of power reached its logical conclusion with the dynasty’s collapse. No one doubts the ability of the Qing state to mobilize vast resources and focus crushing power to meet a specific threat. Even in its last decades, the dynasty demonstrated an astonishing capacity to defeat one challenge after another. But, as a matter of routine, the state had a thin presence on the ground, and its ability to project power, implement policy, and effect change on the local level attenuated over time. By the nineteenth century, local officials in many regions were so completely out of touch with actual conditions in villages that collection of the land tax depended on dunning designated community representatives for agreed quotas that had no connection to specific parcels of land or their owners; in these conditions, tax collection relied on a fragile consensus between taxpayers and the yamen clerks and runners responsible for collection, without which it would fail.13 White contracts became ever more common, because, as the state lost track of actual landholdings, people’s claims to land increasingly had to be secured at the community level without reference to the courts. Under the circumstances, the benefits of registering a contract were no longer worth the expense of doing so. As a result, when land disputes did end up in court, magistrates found they had little choice but to honor white contracts as proof of ownership. Even then, magistrates seem to have had little power to follow up and enforce their judgments in routine land cases, meaning that, in practice, community regulation remained the critical factor.14 The extreme development of this pattern can be seen in Manchuria, where illegal migrants managed to transplant the customary land tenure system of the North China plain, even though that transformation posed a direct threat to the vital interests of the dynasty’s conquest elite; by the mid-nineteenth century, the over-

Conclusion

383

whelming majority of Manchuria’s inhabitants were Han Chinese peasants.15 A similar dynamic occurred in Taiwan, Yunnan, and elsewhere. Supplementary payments after land sales remained common practice well into the twentieth century, and the same is true of wife-selling and other forms of human trafficking. The Qing judiciary was utterly impotent in the face of these challenges. The evidence suggests that, when tenacious social practice sharply contradicted the mandates of the Qing code, the imperial center was not so strong after all. On the contrary, its mandates ended up being compromised through pragmatic flexibility or, in the long run, met with defeat. Among other things, the Qing legal archives expose the dynasty’s inability to find solutions for the fundamental social, economic, and demographic problems it faced—problems of which polyandry, wife sale, and related practices were among the most disturbing symptoms. As I have argued elsewhere, High Qing absolutism with regard to female chastity represented an effort, however misguided, to cope with these problems.16 But its policy expressions—including the scapegoating of “bare sticks” and the prohibition of prostitution and poverty-driven wife sales— amounted to the criminalization of poverty and punishment of the poor. As such, they only exacerbated the underlying problems they were intended to address.

This page intentionally left blank

appendices

A P P E N D I X A : Q I N G DY NA S T Y R E IG N P E R IO D S ( 1 6 4 4 – 1 9 1 2 )

Shunzhi Kangxi Yongzheng Qianlong Jiaqing Daoguang Xianfeng Tongzhi Guangxu Xuantong

1644–61 1662–1722 1723–35 1736–95 1796–1820 1821–50 1851–61 1862–74 1875–1908 1909–12

A P P E N D I X B : P R O F I L E S O F T H E P R O TAG O N I ST S I N W I F E S A L E S

B.1: Data from xingke tiben Occupation/social class of the seller (281 examples):

Peasant: Rural laborer: Beggar: Urban laborer: Peddler: Craftsman/skilled trade: Servant/slave: Other:

194 (69%) 55 (20%) 8 (3%) 4 (1%) 4 (1%) 4 (1%) 3 (1%) 9 (3%) 385

386

appendices

Note: Percentages are of the total number for which I have relevant information. For example, when I say that 69 percent of men who sold wives were peasants, I mean 69 percent of the 281 sellers for whom we have this information. Average age of the seller:

At the time of his first marriage (42 examples): Major marriages (35 examples): Minor marriages (6 examples): At the time of the wife sale (185 examples):

23.3 sui 23.9 sui 18.8 sui 34.6 sui

Note: This sample includes a single uxorilocal marriage, in which the man was 21 sui. For the six minor marriages, I count the man’s age when his marriage was “consummated” (cheng hun or cheng qin). Occupation/social class of the buyer (220 examples):

Peasant: Rural laborer: Merchant/trader: Craftsman/skilled trade: Innkeeper, shopkeeper: Peddler: Exam student, degree holder, or military officer: Other:

173 (79%) 13 (6%) 8 (4%) 6 (3%) 4 (2%) 4 (2%) 4 (2%) 8 (4%)

Marital status of the buyer (out of 172 examples):

Men never previously married: Men who are married, want a concubine: Widowers, want a replacement wife (ji qi):

115 (67%) 32 (19%) 25 (15%)

Average age of the buyer at the time of sale (by marital status):

Men never previously married (85 examples): Widowers (17 examples): Men who are married, want a concubine (20 examples):

35 sui 40.4 sui 45.5 sui

Marital history of the wife prior to sale (293 examples):

Married once (i.e., the seller is her first husband): (Includes 38 adopted daughters-in-law) Remarried widow:

284 (97%) 9 (3%)

Average age of wife at the time of first marriage (42 examples):

Average age at first marriage: (Not including 38 adopted daughters-in-law)

17.5 sui

Appendices

387

Note: This average is two or three years younger than the median ages of women at first (major) marriage in a far larger sample from rural Taiwan (Wolf and Huang, 1980: table 10.2). Adopted daughters-in-law consummated marriage shortly after menarche, in their mid- to late teens. Average age of wife at the time of her sale:

Women sold to be main wives (118 examples): Women sold to be concubines (22 examples):

29.6 sui 27.3 sui

Numbers sold to become main wife (qi), concubine (qie), or prostitute (235 examples):

Wife: Concubine: Prostitute:

195 (83%) 38 (16%) 2 (