The Cambridge Economic History of China: Volume 1, To 1800 [New ed.] 1108425577, 9781108425575

Citation preview

the cambridge economic history of CHINA

China’s rise as the world’s second-largest economy surely is the most dramatic development in the global economy since the year 2000. But China’s prominence in the global economy is hardly new. Since 500 bc e, a dynamic market economy and the establishment of an enduring imperial state fostered precocious economic growth. Yet Chinese society and government featured distinctive institutions that generated unique patterns of economic development. The six chapters of Part I of this volume trace the forms of livelihood, organization of production and exchange, the role of the state in economic development, the evolution of market institutions, and the emergence of trans-Eurasian trade from antiquity to 1000 ce. Part II, in twelve thematic chapters, spans the late imperial period from 1000 to 1800 and surveys diverse fields of economic history, including environment, demography, rural and urban development, factor markets, law, money, finance, philosophy, political economy, foreign trade, human capital, and living standards. D eb in M a is Professor of Economics at Hitotsubashi University, Tokyo, Japan. R ic hard v on G l a h n is Distinguished Professor of History at the University of California, Los Angeles.

Published online by Cambridge University Press

t h e ca m b r i d g e e c o n o m i c h i s t o r y of

CHINA Building on a wide array of recent scholarship, the two volumes of The Cambridge Economic History of China bring together the fruits of pioneering international studies in all dimensions of economic history, past and present. Exploring themes including political economy, agriculture, industry and trade, technology, ecological change, demography, law, urban development, standards of living, consumption, financial institutions, and national income, the two volumes together provide broad temporal coverage across all of Chinese history, including recent developments in contemporary China. VOLUME I To 1800 edited by debin ma and richard von glahn VOLUME II 1800 to the Present edited by debin ma and richard von glahn

Published online by Cambridge University Press

THE CAMBRIDGE ECONOMIC HISTORY OF

CHINA *

VOLUME I

To 1800 *

Edited by

DEBIN MA Hitotsubashi University, Tokyo and RICHARD VON GLAHN University of California, Los Angeles

Published online by Cambridge University Press

University Printing House, Cambridge c b2 8b s, United Kingdom One Liberty Plaza, 20th Floor, New York, n y 10006, USA 477 Williamstown Road, Port Melbourne, vi c 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108425575 d o i: 10.1017/9781108587334 © Cambridge University Press 2022 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2022 Printed in the United Kingdom by TJ Books Limited, Padstow Cornwall A catalogue record for this publication is available from the British Library. Two-Volume Set i s b n 978-1-107-14606-8 Hardback Volume I i s b n 978-1-108-42557-5 Hardback Volume II is bn 978-1-108-42553-7 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

Contents

List of Figures page viii List of Maps ix List of Tables x List of Contributors to Volume I xii Acknowledgments xiv Note on Citations xvi

Introduction to Volume I 1 debin ma and richard von glahn

part i BEFORE 1000 1 . The Economy of Late Pre-imperial China: Archaeological Perspectives 15 lothar von falkenhausen 2 . Agriculture and Its Environmental Impact 52 motoko hara 3 . State and Economy: Production, Extraction, and Distribution richard von glahn 4 . Money, Markets, and Merchants y o¯ h ei ka k i n u m a

131

5 . Economic Philosophy and Political Economy richard von glahn

v

Published online by Cambridge University Press

166

92

Contents

6 . Silk Road Trade and Foreign Economic Influences xinru liu

INTERLUDE

203

241

The Tang–Song Transition in Chinese Economic History r i c h a r d v o n g la h n

p a r t ii 1000 TO 1800 7 . Ecological Change and Resource Constraints david a. bello

259

8 . Population Change 300 s h u j i ca o 9 . Public Finance 340 c h r i s t i a n la m o u r o u x a n d r i c h a r d v o n g l a h n 10 . Political Economy 381 h e l en d u n s t a n 11 . Law and the Market Economy 419 b i l l y k. l . s o an d s u f u m i s o 12 . Property Rights and Factor Markets 448 m i o ki s h i m o t o 13 . The Rural Economy 484 k e nn e t h p o m e r a n z 14 . Cities and the Urban Economy harriet zurndorfer 15 . The Monetary System a k i n o b u ku r o d a

522

560

16 . Merchants and Commercial Networks j o s ep h p . m c d e r m o t t

vi

Published online by Cambridge University Press

597

243

Contents

17 . Foreign Trade 637 angela schottenhammer 18 . Production, Consumption, and Living Standards 676 z h i w u ch e n a n d k a i x i a n g p e n g Bibliography of Primary Works Cited Index 715

vii

Published online by Cambridge University Press

710

Figures

4.1 6.1 6.2 7.1 7.2 7.3 8.1 9.1 10.1 12.1 14.1 14.2 15.1 15.2 18.1 18.2 18.3 18.4 18.5 18.6 18.7

Standard, “clipped,” and “ring” wuzhu coins Lead disks with Greek letters found in China An Jia making a treaty with the Turks Recorded occurrence of flood and drought at twenty-year intervals, 1000–1649 Recorded occurrence of flood and drought at twenty-year intervals, 1644–1800 Reconstructed temperature and flood/drought variability in the Yangzi delta, 1000–2000 ce Correlation between 1391 prefectural population and 1461 numbers of li for ninety-four prefectures Revenues and expenditures of the Taicang Treasury, 1518–1620 Xu Guangqi’s reconstruction of ancient land tenure Land prices in Huizhou and rice prices in China, 1500–1800 Detail from Traveling Upriver at Qingming Festival Panorama of Suzhou in the eighteenth century Frequency of daily transactions of the Tongtaihao firm, fifth–sixth lunar months, 1830 Annual fluctuations of rice prices, Tunxi county, Anhui, 1821–1860 Estimates of China’s historical per capita GDP Population of China, 2 ce –2000 China’s historical crop yield, 1100–1950 Per capita output of processed grain, 1100–1950 Trend in real wages, 971–1920 Silk and rice price indices, 960–1600 Long-term trend in age of death, 600–1870

viii

Published online by Cambridge University Press

page 137 223 228 292 293 294 306 368 400 473 528 551 585 589 679 680 682 683 688 689 697

Maps

1.1 2.1 4.1 4.2 6.1 i.1 7.1 7.2 7.3 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 8.11 8.12 9.1 13.1 14.1 14.2 17.1 17.2

Archaeological sites mentioned in Chapter 1 Geomorphology of China Monetary regions in the Warring States period The Grand Canal in the Sui dynasty Silk Road trade routes, c. 500 Political fragmentation and currency regions, c. 943 Hydroclimate regions Soil regions of China Natural distribution of Cunninghamia (“Chinese fir”) in China Migration routes during the Tang–Song transition, 755–1000 Population densities in Northern Song China, c. 1102 Population densities in Ming China, c. 1391 Population densities in Ming China, c. 1580 Net population losses during the Ming–Qing transition Population densities in Qing China, c. 1776 Urbanization rates, c. 1391 Urbanization rates, c. 1580 Distribution of market towns in the late Ming Urbanization rates, c. 1910 Urbanization rates, c. 1776 Distribution of market towns in 1776 Monetary regions in Southern Song Macro-regional structure of late imperial China Major commercial cities in late imperial China Major cities of Jiangnan Maritime East Asia in Northern Song China Maritime East Asia, c. 1620

ix

Published online by Cambridge University Press

page 16 53 133 157 212 249 266 269 275 312 313 315 316 317 319 326 327 329 334 336 337 351 485 525 530 643 661

Tables

2.1 3.1 3.2 4.1 7.1 8.1 8.2 8.3 8.4 8.5 9.1 9.2 9.3 9.4 9.5 9.6 9.7 13.1 15.1 15.2 15.3 15.4 15.5 15.6 15.7 16.1

Climatic shifts in China, 3000 b ce to 1000 ce page 55 Estimated government revenues of the Han Empire in the first century bce 117 Land allocations under the Northern Wei equal-field system 124 Means of payment in imperial emoluments 141 Hydroclimate by region during the Medieval Warm Period and Little Ice Age 265 China’s population from 200 b ce to 1953 309 Estimated urban population in four Ming provincial capitals, c. 1391 322 Estimated urban population in forty prefectural capitals, c. 1391 322 Estimated county seat urban populations for 289 counties, c. 1391 324 Estimations of Shandong urban population by prefecture, c. 1776 335 Coin revenues in the Northern Song 343 Central government income in 1093 347 Emission of huizi paper currency 353 Ming silver revenues, c. 1580 366 Budgeted military expenditures for the northern frontier commands, 1531–1602 366 Central government revenues, 1766 373 Central government expenditures, 1766 375 Estimates of household income for an average Yangzi delta tenant family 515 Distribution of reign-era names of copper coins from Southern Song hoards 564 Distribution of reign-era names of copper coins from the Aba hoard, Sichuan, 1870s 564 Prices for rice purchases on the occasion of the Qingming festival, Tunxi (Anhui), 1783–1805 579 Receipts and disbursements of silver of Tongtaihao, first day, tenth month, 1841 582 Annual baoyin silver ingot receipts of the Tongtaihao firm, 1821–1848 587 Excerpts from income and expenditure entries of Tongtaihao for the first day of the tenth month, 1841 588 Annual sales and purchases of farm households in Qingyuan county, Hebei, 1930 590 Markets and market towns in twelve provinces in the Ming and Qing 601

x

Published online by Cambridge University Press

List of Tables 16.2 Markets and market towns in Suzhou and Songjiang prefectures in the Ming and Qing 16.3 Markets and market towns in Hubei, Hunan, and Jiangxi provinces in the Ming and Qing 16.4 Social composition of Zhujiajiao market town, Songjiang prefecture, 1814 16.5 Stores (by type) in Linqing market town in the late Ming and mid-Qing 16.6 Routes and volume of grain and soybean trade in the mid-Qing 17.1 State income from foreign trade 17.2 Tribute exchange in the early Ming

xi

Published online by Cambridge University Press

602 604 608 610 613 641 655

Contributors to Volume I

Da vi d A. Bello is E.L. Otey Professor of East Asian Studies, Washington and Lee University. Sh u j i Ca o is Professor of History, Shanghai Jiaotong University. Zh iwu Chen is Chair Professor of Finance and Victor and William Fung Professor in Economics, University of Hong Kong. Helen Dunstan is Professor of Chinese Studies Emeritus, University of Sydney. Lo t h a r v on Fal k e nh aus e n is Distinguished Professor of Chinese Archaeology and Art History, University of California, Los Angeles, and Changjiang Professor, Xibei University. Moto k o Har a is Professor Emeritus, Faculty of Economics, Ryu¯tsu¯ Keizai University. Yo¯ hei K aki nu ma is Associate Professor of Chinese History, Waseda University. Mi o Ki sh im o t o is Professor Emeritus of Chinese History, Ochanomizu University. Akinobu Kurod a is Professor of Chinese History, University of Tokyo. Ch ri s tia n La m o u r o u x is Directeur d’études, École des hautes études en sciences sociales. Xin ru Liu is Professor Emeritus of History, The College of New Jersey, and honorary member of the Centre of Comparative Study between Chinese Civilization and Pre-modern World Civilizations, Chinese Academy of History (Beijing). Jo s e p h P. Mc De r m o t t is Fellow of St. John’s College, University of Cambridge, and Emeritus Reader in Chinese History, University of Cambridge. Kaixian g Peng is Professor of Economics in the School of Economics and Management, Wuhan University.

xii

Published online by Cambridge University Press

List of Contributors to Volume I

Kennet h Po m e r an z is University Professor of Modern Chinese History, University of Chicago. An g ela Schottenhammer is Professor of Chinese Middle Period and Early Modern World History, Shanghai University and KU Leuven. Bil ly K.L. So is Vice Rector and Chair Professor of History, University of Macau. Suf umi So is Visiting Fellow, University of Hong Kong. Richard von Gl ah n is Distinguished Professor of History, University of California, Los Angeles. Ha r r i e t Zurnd orf er is Senior Research Scholar, Leiden Institute for Area Studies, Leiden University.

xiii

Published online by Cambridge University Press

Acknowledgments

We are very grateful in the first place to our nearly fifty contributors, whose collective efforts and dedication made this project possible. The timely progress of this project was greatly facilitated by our ability to hold conferences to assemble the contributors to each volume, exchange ideas, and settle the division of labor among the forty chapters that ultimately were included in the published work. We are deeply indebted to the benefactors who provided financial support for the conferences. We want to thank Bas van Leeuwen for hosting the conference for the contributors to Volume II at the University of Utrecht on December 15–16, 2017, with the generous financial support from the European Research Council under the European Union’s Horizon 2020 Programme/ERC-StG 637695 – HinDI (as part of the Historical Dynamics of Industrialization in Northwestern Europe and China ca. 1800–2010: A Regional Interpretation project). We gratefully acknowledge the organizational assistance from Robin Philips and Zipeng Zhang. The volume benefited from the participation and external discussion of Maarten Prak, Keetie Sluyterman, Jan Luiten van Zanden, and Peer Vries. Debin Ma also acknowledges some editorial assistant support from the School of Economics of Fudan University in China. The conference for contributors to Volume I was held on August 27–28, 2018, at the University of California, Los Angeles (UCLA). We are grateful to the Chiang Ching-Kuo Foundation for International Scholarly Exchange, which provided the main financial support to underwrite the conference expenses. Additional financial support was provided by the Center for Chinese Studies and the Division of Social Sciences at UCLA, and we especially want to acknowledge the UCLA Center for Chinese Studies, its director Yunxiang Yan, and its assistant director Esther Jou for their logistical support. We also want to thank Sunkyu Lee and Kayoko Fujita for their invaluable assistance in facilitating the UCLA conference.

xiv

Published online by Cambridge University Press

Acknowledgments

Lucy Rhymer at Cambridge University Press has instigated, encouraged, and guided this project at every step, and we surely would not have been able to bring it to completion without her unflagging support. We are also grateful to the eagle-eyed Emily Sharp at Cambridge University Press for all of her help in the final production of these volumes.

xv

Published online by Cambridge University Press

Note on Citations

The two volumes of the Cambridge Economic History of China differ in their citation of Chinese and Japanese works in the footnotes and the “Further Reading” bibliographies, reflecting differences in conventions between scholarship on premodern and modern periods. In Volume I, book and article titles for secondary scholarship in Chinese and Japanese are given in the original language and romanization (Pinyin romanization in the case of Chinese). Names of authors are given in full (including first names). In Volume II, Chinese and Japanese book and article titles are given in the original language and English translation without romanization. First names of authors are abbreviated. In Volume I, footnote references to primary sources in Chinese are abbreviated as acronyms. A comprehensive list of Chinese primary sources with full bibliographic information for primary sources appears at the end of the volume. In Volume II, full bibliographic information for references to primary sources is given in the footnote in which they are cited.

xvi

Published online by Cambridge University Press

Introduction to Volume I debin ma and richard von glahn

China’s rise as the world’s second-largest economy surely is the most dramatic development in the global economy since the year 2000. But China’s prominence in the global economy is hardly new. Since 500 B C E, a burgeoning market economy and the establishment of an enduring imperial state has fostered precocious economic growth. Moreover, contrary to the view that China’s economy withered under the dual constraints of Western colonialism and Chinese tradition after 1800, recent scholarship has identified the onset of modern economic growth in response to new incentive structures, investment opportunities, ideas, and technology, laying the foundation for the post-1978 economic miracle. China’s combination of market-led growth under the firm hand of the state has produced a model of economic development that challenges conventional theories of capitalism and economic growth. The spectacular growth of the contemporary Chinese economy has also spurred deeper investigation into the Chinese economy – long a neglected field of study, at least in the Western academy. Scholarship on Chinese economic history has now developed to the stage where a Cambridge History devoted to the subject is appropriate and feasible. These volumes, a collaborative effort by nearly fifty scholars, bring together the fruits of pioneering Western, Japanese, and Chinese scholarship in all dimensions of economic history, past and present. Early studies of the Chinese economy focused on China’s distinctive philosophical and political traditions.1 In his published 1911 Columbia 1

For historiographic surveys of twentieth-century scholarship on Chinese economic and social history, see Timothy Brook, “Capitalism and the Writing of Modern History in China,” in Timothy Brook and Gregory Blue (eds.), China and Historical Capitalism: Genealogies of Sinological Knowledge (Cambridge, Cambridge University Press, 1999), pp. 110–57; Richard von Glahn, “Imagining Pre-modern China,” in Paul Jakov Smith and Richard von Glahn (eds.), The Song–Yuan–Ming Transition in Chinese History (Cambridge, MA, Harvard University Council on East Asian Studies, 2003), pp. 35–70.

1

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

debin ma and richard von glahn

Ph.D. thesis, The Economic Principles of Confucius and His School), Ch’en Huanchang sought to introduce the basic concepts and practices of Chinese political economy, highlighting the pre-eminence of the Confucian tradition.2 Of course, Ch’en was writing at the moment of the dissolution of the Chinese empire and the apogee of the modern world-system defined by industrial capitalism and Western political hegemony. Already in Ch’en’s book – and for long afterwards – study of Chinese economic history was fixated on the question of the apparent lack of economic progress throughout two millennia of imperial history. Many of the answers proposed by Ch’en – Confucian disdain for moneymaking, a rigid and inert social structure, overpopulation, and isolation from the outer world – recurred in subsequent works. The fall of the Qing Empire in 1911 – and with it, the end of China’s imperial past – inspired hope for China’s rapid transformation into a modern, progressive nation. But the failure of republican institutions to thrive in the wake of the empire’s demise gave rise to doubts about China’s ability to “learn from the West.” Some intellectuals advocated wholesale repudiation of Chinese cultural traditions and embrace of Western culture as defined by cosmopolitanism, Enlightenment values, republican government, scientific reasoning, and capitalist economic institutions. Others sought to reinvigorate China’s “national essence” by reviving authentic Chinese values that had been attenuated by Western spiritual pollution and Manchu overlordship. The Japanese scholar Naito¯ Konan, writing in 1914, envisioned a future in which Japan supplanted China as the ascendant center of a reinvigorated “oriental culture” that would eclipse the spiritually vacuous materialism of the West. Naito¯ advanced the novel thesis that East Asia’s “modern age” had actually begun centuries earlier, in the transition from the Tang (618–907) to Song (960–1279) dynasties. This Tang–Song transition had witnessed the triumph of autocratic monarchy over aristocratic rule and engendered a vibrant commoner culture liberated from feudal domination. But China’s incipient modernity proved premature; after the Song dynasty China’s “modern age” degenerated into senility, and in contemporary times the dynamic center of oriental culture had shifted to Japan.3 2

3

Huan-chang Ch’en, The Economic Principles of Confucius and His School (New York, Columbia University Press, 1911). Naito¯ Konan 内藤湖南, Shinaron 支那論 (1914), rpt. in Naito¯ Konan zenshu¯ 内藤湖南全集, vol. 5 (Tokyo, Chikuma shobo¯, 1972), pp. 291–482; see also Hisayuki Miyakawa, “The Naito¯ Hypothesis and Its Effects on Japanese Studies of China,” Far Eastern Quarterly 14.3 (1955), 533–52; Stefan Tanaka, Japan’s Orient: Rendering Pasts into History (Berkeley, University of California Press, 1993).

2

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

Introduction to Volume I

The emergence of history as an academic profession in China and Japan during the 1920s was accompanied by skepticism toward received historical traditions and more critical empiricism in historical methodology. But by the late 1920s historical scholarship in China became enveloped by disputes over the trajectory of Chinese history from antiquity to the present, most notably the so-called Social History Debate, which dwelled primarily on questions of feudalism and capitalism in China. Marxism and the Russian Revolution loomed large over these controversies. Left-wing scholars and writers were keen to demonstrate that China’s historical experience conformed to the universal categories of historical development as defined by Western philosophers. In his Study of China’s Ancient Society (1930), Guo Moruo was the first to apply to Chinese history the five-stage theory of human history (from primitive communism to socialism) formulated by Soviet Marxist scholars.4 Guo argued that the onset of the Iron Age in China in the first millennium B C E inaugurated feudal relations of production that endured even under the centralized bureaucratic empires. The anomalous character and persistence of Chinese “feudalism” – which bore little resemblance to the fragmented political order of medieval Europe – posed a vexing problem. Some historians in China sought to resolve this incongruity by espousing an economic definition of feudalism based on the antagonism between the landowning class and the peasantry, and thus postulated a feudal epoch that stretched from the ancient Zhou era throughout imperial history and continued even during the post-1911 Republican era. In Europe, Max Weber had traced the apparent stagnation of Chinese historical development to the special character of the imperial bureaucratic state, which Marxist scholars reformulated as a distinctive species of “bureaucratic feudalism.” Drawing on the theories of both Marx and Weber, Karl Wittfogel, a leading figure in the Frankfurt school of Marxism, proposed that the Chinese imperial state was founded on an “Asiatic mode of production” which hindered the dynamic forces of class struggle that motivated historical change.5 Although Soviet Marxists firmly repudiated the idea of an Asiatic mode of production, it gained considerable currency among some Chinese and many Japanese scholars in the 1930s. Characterizing China as the 4

5

Guo Moruo 郭末若, Zhongguo gudai shehui yanjiu 中國古代社會研究 (Shanghai, Lianhe shudian, 1930). Karl Wittfogel, “The Foundations and Stages of Chinese Economic History,” Zeitschrift für Sozialforschung 4.1 (1935), 25–58; Wittfogel, “Die Theorie der orientalischen Gesellschaft,” Zeitschrift für Sozialforschung 7.1–2 (1938), 90–122. In the postwar period Wittfogel developed a more elaborate version of this thesis. See his Oriental Despotism: A Comparative Study of Total Power (New Haven, Yale University Press, 1957).

3

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

debin ma and richard von glahn

archetype of an “oriental society” trapped in the Asiatic mode of production provided a succinct explanation for the immobility of Chinese history, and for Japanese historians reinforced Japan’s singular role in both “escaping from Asia” and leading it toward modernity. To be sure, other scholars distanced themselves from these theoretical debates and focused instead on empirical study and honing analytical methods of economic history. In China, this trend coalesced around the journal Food and Money Semi-monthly (Shihuo banyuekan 食貨半月刊), inaugurated in 1934 by Tao Xisheng, which featured contributions by Tao, Ju Qingyuan, He Ziquan, Quan Hansheng, and others. The “Food and Money” group (the name derived from the title given to chapters on fiscal administration in traditional dynastic histories) sought to balance economic theories (both Marxist and non-Marxist) with empirical evidence and rigorous methodology.6 Much of their scholarship centered on the early imperial era, especially the Tang dynasty.7 The 1930s also witnessed a profusion of pathbreaking Japanese scholarship on Chinese economic history, likewise centered on the Tang–Song dynasties, led by Kato¯ Shigeshi,8 Hino Kaisaburo¯,9 Sogabe Shizuo,10 and Miyazaki Ichisada.11 The inspiration for these studies can be traced back to Naito¯’s hypothesis of the Tang–Song transition as the beginning of “East Asia’s modern age,” as Miyazaki – Naito¯’s 6

7

8

9

10

11

For a brief introduction to the Food and Money group and its methodological approaches to economic history, see Su Yongming 苏永明, “‘Shihuo pai’ de jingjishi yanjiu fangfa tantao” ‘食货派’的经济史研究方法探讨, Shixueshi yanjiu 史学史研究 2007.3, 77–83. Most notably, Ju Qingyuan 鞠清遠, Tang Song guansi gongye 唐宋官私工業 (Shanghai, Xin shengming shuju, 1934); Quan Hansheng 全漢昇, Zhongguo hanghui zhidu shi 中國 行會制度史 (Shanghai, Xin shengming shuju, 1935); Tao Xisheng 陶希聖 and Ju Qingyuan 鞠清遠, Tangdai jingji shi 唐代經濟史 (Shanghai, Shangwu yinshuguan, 1936); Ju Qingyuan 鞠清遠, Tangdai caizheng shi 唐代財政史 (Shanghai, Shangwu yinshuguan, 1940). Kato¯’s seminal essays on Tang–Song (and some Qing) economic history were published in a two-volume posthumous work: Kato¯ Shigeshi 加藤繁, Shina keizai shi ko¯sho¯ 支那経済史考証 (Tokyo, To¯yo¯ bunko, 1952). Hino’s prolific research on Tang–Song history has been reprinted in his twenty-volume collected works: Hino Kaisaburo¯ 日野開三郎, Hino Kaisaburo¯ To¯yo¯ shigaku ronshu¯ 日野 開三郎東洋史学論集 (Tokyo, San’ichi shobo¯, 1980–1988). Sogabe’s research primarily focused on Song fiscal institutions and monetary history; see Sogabe Shizuo 曽我部静雄, So¯dai zaisei shi 宋代財政史 (Tokyo, Seikatsusha, 1941); Chu¯goku shakai keizai shi kenkyu¯ 中国社会経済史研究 (Tokyo, Yoshikawa ko¯bunkan, 1976). Miyazaki’s first monograph studied the monetary history of the tenth century: Miyazaki Ichisada 宮崎市定, Godai So¯sho no tsu¯ka mondai 五代宋初の通貨問題 (Kyoto, Hoshino shoten, 1943). His body of work, which extended over the entire breadth of Chinese history, has been reprinted in his collected works, which run to twenty-four volumes: Miyazaki Ichisada zenshu¯ 宮崎市定全集 (Tokyo, Iwanami shoten, 1991).

4

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

Introduction to Volume I

successor as professor of Chinese history at Kyoto University – entitled his influential synthesis.12 The priority of the Tang also can be seen in the first European-language monograph explicitly devoted to Chinese economic history, the Hungarian-born scholar Étienne Balazs’s 1932 doctoral thesis for Berlin University on the economic history of the Tang dynasty.13 Although Balazs championed empirical research over both theoretical formulations and narrowly construed philological study, his scholarship was explicitly couched in the Weberian project of comparative study within universal categories of historical development, and contributed to the conception of China as – in his words – a “permanently bureaucratic state” that obstructed the emergence of an independent merchant class and, by extension, capitalism.14 The prewar generation also pioneered the use of quantitative data for the study of Chinese economic history. Particularly noteworthy in this regard were the contributions of Quan Hansheng, the forerunner in the study of price history, international trade, and national revenue (initially focused on the Song, and later extended to the Ming–Qing periods as well), and Liang Fangzhong, who published a pathbreaking essay on Ming population, land, and taxation statistics in 1935 and his landmark study of the sixteenth-century Single-Whip tax reform in 1936.15 In addition, scholars such as Chen Hansheng and Fei Xiaotong, trained in economics and anthropology in the US and the UK respectively, published monographs on the contemporary rural economy, based on extensive field research, that became foundational

12

13

14

15

Miyazaki Ichisada 宮崎市定, To¯yo¯teki kinsei 東洋的近世 (Osaka, Kyo¯iku taimusu sha, 1950). Published as Étienne Balazs, “Beiträge zur Wirtschaftsgeschichte der T’ang-Zeit (618– 906),” Mitteilungen des Seminars für Orientalische Sprachen 34 (1931), 1–92, 35 (1932), 93–165; 36 (1933), 1–62. See his 1957 essay “China as a Permanently Bureaucratic Society,” in Étienne Balazs, Chinese Civilization and Bureaucracy (New Haven, Yale University Press, 1964), pp. 13–27. For Balazs’s impact on Chinese studies in France, where he spent most of his career, see Harriet T. Zurndorfer, “Not Bound to China: Étienne Balazs, Fernand Braudel and the Politics of the Study of Chinese History in Post-war France,” Past and Present 185 (2004), 189–221. Quan’s studies were published in a two-volume collection, Quan Hansheng 全漢昇, Zhongguo jingji shi luncong 中國經濟史論叢 (Hong Kong, Xinya shuyuan Xinya yanjiusuo, 1972), and a subsequent three-volume collection: Zhongguo jingji shi yanjiu 中國 經濟史研究 (Hong Kong, Xinya shuyuan Xinya yanjiusuo, 1976). A collection of Liang Fangzhong’s writings on economic history was published as Liang Fangzhong 梁方仲, Liang Fangzhong wenji 梁方仲文集 (Guangzhou, Zhongshan daxue chubanshe, 2004). Liang also supervised the compilation of what remains the most authoritative collection of Chinese historical statistics: Liang Fangzhong (ed.), Zhongguo lidai hukou tiandi tianfu tongji 中国历代户口田地田赋统计 (Shanghai, Shanghai renmin chubanshe, 1980).

5

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

debin ma and richard von glahn

studies.16 During the Japanese occupation of China in the 1930s–1940s a legion of Japanese scholars, operating under the auspices of the Investigation Department of the South Manchuria Railway (Mantetsu Cho¯sabu, founded in 1907), compiled hundreds of reports on the Chinese economy that have been extensively mined by later scholars.17 The pronounced influence of Keynesian economics among young Chinese economists also produced the first efforts to compile national income data and measure GDP in the late 1940s, efforts that would be jettisoned along with non-Marxist economics after the founding of the People’s Republic of China in 1949.18 In the postwar era, Japanese scholarship on Chinese economic history was especially robust. The ascendancy of Marxist analysis in the Japanese academy – an explicit repudiation of the prewar imperialist project – established a new paradigm for interpreting Chinese economic history. Rejecting both the “oriental-stagnation” thesis and Naito¯ Konan’s ideas about China’s precocious modernity, economic and legal historians such as Sudo¯ Yoshiyuki and Niida Noboru portrayed the Tang–Song transformation as the formative phase of a feudal society based on the subordination of serfs to a landlord class.19 These servile relations were reproduced by the patriarchal social institutions of family, lineage, and guild that inhibited the emergence of the “rational” legal and economic institutions which the great German sociologists denoted by the term Gesellschaft (impersonal social relations). The density of patriarchal communal social relations (Gemeinschaft, rendered in 16

17

18

19

Chen Han-seng, Landlord and Peasant in China: A Study of the Agrarian Crisis in South China (New York, International Publishers, 1936); Chen, Industrial Capital and Chinese Peasants (Shanghai, Kelly and Walsh, 1939); Fei Hsiao-t’ung, Peasant Life in China: A Field Study of Country Life in the Yangtze Valley (London, Kegan Paul, 1939), Fei, Earthbound China: A Study of Rural Economy in Yunnan (Chicago, The University of Chicago Press, 1945). Foreign researchers also contributed to the compilation of social and economic statistics, most notably John Lossing Buck, Chinese Farm Economy (Chicago, The University of Chicago Press, 1930); Buck, Land Utilization in China (Shanghai, Commercial Press, 1937); Sidney D. Gamble, Peking: A Social Survey (New York, George H. Doran Co., 1921). For overviews of this scholarship, see John Young, The Research Activities of the South Manchuria Railway Company, 1907–1945: A History and Bibliography (New York, Columbia University Press, 1966); Matsumura Takao 松村高夫, Yanagisawa Yu¯ 柳沢遊, and Eda Kenji 江田憲治, Mantetsu no cho¯sa to kenkyu¯ – sono “shinwa” to jitsuzo¯ 満鉄の調査 と研究―その「神話」と実像 (Tokyo, Aoki shoten, 2008). Wu Baosan 巫寶三 (ed.), Zhongguo guomin suode, 1933 中國國民所得 (Shanghai, Zhonghua shuju, 1947); Ta-chung Liu, China’s National Income, 1931–36: An Exploratory Study (Washington, DC, Brookings Institution, 1946). For a synopsis of the voluminous studies by Sudo¯ and Niida (and the prodigious postwar Japanese scholarship on Tang–Song economic history generally), see the stateof-the-field essay by Peter Golas: “Rural China in the Song,” Journal of Asian Studies 39.2 (1980), 291–325.

6

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

Introduction to Volume I

Japanese as kyo¯do¯tai 共同体) in Chinese society precluded the formation of an independent bourgeoisie and the transition to capitalism. More importantly, these studies – and a plethora of non-Marxist scholarship as well – generated a wealth of new empirical research on Chinese economic history, including a growing body of work centered on the Ming–Qing period and the twentieth century. In China, of course, orthodox Marxist historiography prevailed after 1949. However, numerous studies attesting to the vitality of the market economy in the late Ming dynasty gave rise to the idea that the “sprouts of capitalism” (zibenzhuyi mengya 資本主義萌芽) had begun to emerge by the late sixteenth century, if not earlier.20 Shang Yue became the most prominent exponent of the controversial thesis that an incipient bourgeoisie had formed in the late Ming period and that China thus was already beginning the transition to capitalism before it was derailed by the Manchu conquest of the Ming in 1644.21 Japanese historians quickly joined this debate. Scholars who focused on the urban economy and merchants such as Fu Yiling and Tanaka Masatoshi tended to underscore the potential for indigenous capitalist transformation.22 Historians who concentrated on relations of production in the agrarian economy and handicraft industries such as Nishijima Sadao were far more skeptical.23 However, by 1960 Shang Yue’s thesis on the “sprouts of capitalism” was deemed heretical within the Chinese academic establishment. Although Shang Yue’s contention that China was on the verge of a breakthrough to capitalism before the Opium War was repudiated, most scholars assented to the proposition that rising commodity production in the Ming–Qing era attested to an “advanced” form of feudalism in China, challenging the idea that the Western European historical experience exclusively defined the archetype of the feudal economy. According to this line of 20

21

22

23

For a brief and regrettably tendentious introduction, see Albert Feuerwerker, “From ‘Feudalism’ to ‘Capitalism’ in Recent Historical Writing from Mainland China,” Journal of Asian Studies 18.1 (1958), 107–16. Shang Yue 尚鉞, Zhongguo zibenzhuyi guanxi fasheng ji yanbiande chubu yanjiu 中國資本 主义关系發生及演变的初步研究 (Beijing, Sanlian shudian, 1956). Fu Yiling 傅衣凌, Mingdai Jiangnan shimin jingji shitan 明代江南市民經济試探 (Shanghai, Shanghai renmin chubanshe, 1957); Tanaka Masatoshi 田中正俊, “Chu¯goku rekishikai ni okeru ‘shihonshugi no myo¯ga’ kenkyu¯” 中国歴史界におけ る「資本主義の萌芽」研究, in Suzuki Jun 鈴木俊 and Nishijima Sadao 西島定雄 (eds.), Chu¯goku shi no jidai kubun 中国史の時代区分 (Tokyo, To¯kyo¯ daigaku shuppansha, 1957), pp. 219–52. Nishijima Sadao 西島定雄, “Chu¯goku kodai shakai no ko¯zo¯teki tokushitsu ni kan suru mondaiten” 中国古代社会の構造的特質に関する問題点, in Suzuki and Nishijima, Chu¯goku shi no jidai kubun, pp. 175–208.

7

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

debin ma and richard von glahn

thought, the intrusion of foreign imperialism in the nineteenth century and China’s subjugation as a “semi-colony” warped the development of Chinese capitalism, precluding the formation of an autonomous national bourgeoisie and instead fostering state-led “bureaucratic capitalism.” Thus China remained a “semi-feudal, semi-colonial” society until the Communist Party initiated a proletarian revolution beginning in the 1930s. Notwithstanding the ideological bent of PRC scholarship in this era, a vast array of source materials, statistical series, and valuable monographic studies were published that stimulated research both within China and abroad.24 By contrast, in the immediate postwar era economic history was a virtually untouched subject in Western scholarship on China, which was deeply averse to the Marxist cast of the above-mentioned Chinese and Japanese studies. In an essay on the state of the field at the dawn of the 1960s, the American scholar Albert Feuerwerker dourly observed, (1) Monographic studies in Chinese rarely come up to the standards expected of European economic historians. (2) Few detailed investigations have appeared in European languages; and their quality is very uneven. (3) Perhaps both in quantity and quality the most significant body of monographic work has been done in Japan, though here too there are serious limitations growing out of the strong hold of Marxist ideology in Japanese academic circles. (4) There is, to my knowledge, no satisfactory synthetic treatment of Chinese economic history in any language to which a nonspecialist might go for a substantive introduction to this subject.25

Feuerwerker’s citations belie these grim conclusions to some extent; for example, he mentions Peng Xinwei’s Monetary History of China, a magisterial survey that remains unsurpassed to this day.26 Mention also should be made of the numerous publications during these years by Lien-sheng Yang, perhaps the first scholar to write a Ph.D. thesis at a US university (Harvard, 1946) on Chinese economic history.27 Regrettably, the ideological struggles that convulsed China during the Cultural Revolution of the 1960s–1970s shuttered the 24

25

26

27

Albert Feuerwerker, “China’s Modern Economic History in Communist Chinese Historiography,” China Quarterly 22 (1965), 31–61. Albert Feuerwerker, “Materials for the Study of the Economic History of Modern China,” Journal of Economic History 21.1 (1961), 42. Peng Xinwei 彭信威, Zhongguo huobi shi 中國貨幣史 (Shanghai, Shanghai renmin chubanshe, 1958). Yang’s work includes Money and Credit in China: A Short History (Cambridge, MA, Harvard University Press, 1952), Les aspects économiques des travaux publics dans la Chine impériale: Quatre conférences (Paris, Collège de France, 1964), and studies gathered in his essay collections Studies in Chinese Institutional History (Cambridge, MA, Harvard University Press, 1961) and Excursions in Sinology (Cambridge, MA, Harvard University Press, 1969).

8

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

Introduction to Volume I

universities, inflicted enormous personal hardship on many scholars, and effectively suspended serious scholarship. In any event, the decade of the 1960s would prove to be a watershed in Western scholarship on Chinese economic history. The rise of social science research and its application to historical study provided the catalyst for a series of landmark studies that would shape Western scholarship on Chinese economic history for a generation: Ping-ti Ho’s incisive dissection of historical population statistics;28 Albert Feuerwerker’s study of state-led industrialization efforts in the late nineteenth century;29 the meticulous reconstruction of Tang fiscal administration by Denis Twitchett, which amply demonstrated the value of Japanese scholarship;30 Robert Hartwell’s provocative findings on the precocious development of the coal and iron industries in Song China;31 the insights of economic geography applied by G. William Skinner to generate a new paradigm of market structure and marketing systems in China;32 the application of quantitative analysis to agricultural production pioneered by Dwight Perkins;33 Ramon Myers’s revisionist analysis, drawing on quantitative data from the Mantetsu surveys, of economic performance in rural China in the Republican period;34 Shiba Yoshinobu’s empirically rich and analytically sophisticated tour de force on commerce and merchant enterprise in the Song, which became accessible to a wider audience through Mark Elvin’s abbreviated translation;35 crowned by Elvin’s own theoretically innovative paradigm of the course of economic

28

29

30

31

32

33 34

35

Ping-ti Ho, Studies on the Population of China, 1368–1953 (Cambridge, MA, Harvard University Press, 1959). Albert Feuerwerker, China’s Early Industrialization: Sheng Hsuan-huai (1844–1916) and Mandarin Enterprise (Cambridge, MA, Harvard University Press, 1958). Denis Twitchett, Financial Administration under the T’ang Dynasty (Cambridge, Cambridge University Press, 1963) Robert M. Hartwell, “A Revolution in the Chinese Iron and Coal Industries during the Northern Sung, 960–1126 A D,” Journal of Asian Studies 21.1 (1962), 153–62; Hartwell, “Markets, Technology, and the Structure of Enterprise in the Development of the Eleventh-Century Chinese Iron and Steel Industry,” Journal of Economic History 26.1 (1966), 29–58; “A Cycle of Economic Change in Imperial China: Coal and Iron in Northeast China, 750–1350,” Journal of the Economic and Social History of the Orient 10 (1967), 102–59. G. William Skinner, “Marketing and Social Structure in Rural China,” Journal of Asian Studies 24.1 (1964), 3–43; 24.2 (1965), 195–228; 24.3 (1965), 363–99. Dwight H. Perkins, Agricultural Development in China, 1368–1968 (Chicago, Aldine, 1969). Ramon H. Myers, The Chinese Peasant Economy (Cambridge, MA, Harvard University Press, 1970). Shiba Yoshinobu 斯波義信, So¯dai sho¯gyo¯ shi kenkyu¯ 宋代商業史研究 (Tokyo, Kazama shobo¯, 1968), translated as Commerce and Society in Sung China (Ann Arbor, University of Michigan Center for Chinese Studies, 1970).

9

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

debin ma and richard von glahn

development in China across the imperial era, another work deeply informed by Japanese scholarship.36 By the 1970s, then, Western scholarship on Chinese economic history had achieved a new level of maturity. The wealth of new empirical studies since then has fostered vigorous, indeed contentious, debate on issues such as the impact of Western and Japanese imperialism, the nature of the peasant economy, regional systems and market networks, and what has come to be called the “Great Divergence” debate. In particular, the past three decades have seen significant revisionist scholarship on Chinese economic performance in the tumultuous nineteenth and twentieth centuries following the intrusion of Western imperialism in the post-Opium War era. Contrary to the once dominant pessimistic interpretation of a Chinese economy withering under the dual constraints of Western colonialism and Chinese tradition, the new scholarship has identified the onset of modern economic growth in this era – at least in some crucial regions and sectors – as a powerful response to new incentive structures and investment opportunities, as well as the inflow of new ideas and technology, laying the foundation for China’s economic takeoff in the post-1978 reform era. Within China, perhaps the most significant development is the publication of the three edited volumes of History of Chinese Capitalist Development in 1983.37 Under the leadership of one of the co-editors, Wu Chengming of the Chinese Academy of Social Science, these three volumes, although within a Marxist framework, brought together a generation of devoted senior and junior scholars to provide a comprehensive economic history of modern China from the early modern era to 1950. Wu, himself an economist who received a master’s degree from Columbia University in the 1940s but was banished during China’s Cultural Revolution, emerged as an intellectual leader in economic history within China throughout the 1980s and 1990s. Although much of the new scholarship on Chinese economic history has focused on the post-1800 period, for which quantitative data are much more abundant, there have been significant advances in the study of China’s premodern economy, particularly for the ancient period. Important new data generated from archaeological research in China, ranging from textual and artifactual evidence to urban morphology and settlement studies, have yielded fresh insights into social and economic livelihood in ancient China 36 37

Mark Elvin, The Pattern of the Chinese Past (Stanford, Stanford University Press, 1973). Xu Dixin 許滌新 and Wu Chengming 吳承明 (eds.), Zhongguo zibenzhuyi fazhan shi 中國資本主義發展史, 3 vols. (Beijing, Renmin chubanshe, 1985). Volume 1 was translated into English as Xu Dixin and Wu Chengming (eds.), Chinese Capitalism, 1522– 1840 (New York, St. Martin’s Press, 2000).

10

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

Introduction to Volume I

and enable us to trace the course of economic change with much greater temporal and geographic precision. Although Western scholarship on China’s premodern economic history remains modest, quantitatively speaking, compared to the prodigious output of Chinese and Japanese scholars, Western historians have done pioneering work in many aspects of the premodern economy, including environmental history; demography; legal institutions and economic organization; kinship, gender, and the household economy; political economy; and economic sociology. These scholarly developments have coincided with the unfettering of Chinese scholarship from shopworn Marxist–Leninist ideological blinders since the early 1980s and an enormous surge in new scholarship on Chinese economic history within China. With the opening of new archives and improvement of the academic infrastructure in China, new generations of Chinese scholars have begun to make important methodological and theoretical contributions to the study of Chinese economic history ranging from agriculture and demography to finance and law. The rapidly growing presence of Chinese scholars within the global economic history community (as seen, for example, at meetings such as the triennial World Economic History Congress) also testifies to the rising impact of Chinese economic history within the profession internationally. A significant shift in the scholarly landscape has been the gradual rebalancing of the community working on Chinese economic history. Twenty years ago, the few scholars working on economic history were largely based outside China. The last decade has seen a steady growth of a young generation of researchers who returned to China from having gained training in the quantitative and economic approaches from North American and European Ph.D. programs. In conjunction with the rising stock of foreigntrained Ph.D.s in economic history, universities in China and Hong Kong now produce a steady stream of Ph.D.s trained in quantitative economic history, many of whom are making careers in China. The sheer quantity of the new scholarship on Chinese economic history since the 1970s defies adequate summary in this brief essay, but it will be cited copiously throughout these volumes. In addition, synthetic surveys of Chinese economic history have now begun to appear.38 Along with these surveys are two 38

Loren Brandt, Debin Ma, and Thomas G. Rawski, “From Divergence to Convergence: Reevaluating the History behind China’s Economic Boom,” Journal of Economic Literature 52.1 (2014), 45–123; Richard von Glahn, The Economic History of China from Antiquity to the Nineteenth Century (Cambridge, Cambridge University Press, 2016); Okamoto Takashi 岡本隆司 (ed.), Chu¯goku keizai shi 中国経済史 (Nagoya, Nagoya daigaku shuppankai, 2013).

11

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

debin ma and richard von glahn

commissioned special journal issues devoted entirely to Chinese economic history.39 The Cambridge Economic History of China is divided chronologically into two volumes, with the first volume devoted to the period before 1800 and the second volume to the period from 1800 to the present. Each volume is further subdivided into two broad chronological sections, but within these divisions the chapters are organized topically rather than chronologically. Part I of Volume I (in six chapters) covers the period from 1000 B C E to 1000 C E, with Part II (in twelve chapters) devoted to the period from 1000 to 1800. The unequal portions assigned to the pre-1000 and 1000–1800 periods reflect differences in the depth and breadth of the scholarship at this point in time. The usual periodization of Chinese history posits a sharp break between the middle imperial period of 750–1500 (often subsumed under the Tang–Song transition rubric discussed above) and the 1500–1800 era (whether this period should be defined as China’s “early modern” era remains controversial). The scholarship usually reflects this divide as well. However, given the topical structure of the volume, we believe that the 1000–1800 period should be treated as an integral whole. Volume II similarly is divided into two broad chronological parts that are subdivided into thematic chapters. Part I covers the period from 1800 to 1950, encompassing the last century of the Qing dynasty and the Republican era (1911–1949), in fourteen chapters. Part II, in seven chapters, examines the dramatic transformations of the Chinese economy since the founding of the People’s Republic of China in 1949. As the first attempt in the Cambridge History series to focus on Chinese economic history, our two volumes will remedy a large lacuna in the discipline of economic history and respond to the increasing demand from both specialists and the general public for a comprehensive introduction to the subject. These volumes will provide an authoritative survey incorporating up-to-date research at the frontiers of knowledge, including quantitative data that are accessible to a general economic history audience, as well as addressing some of the most important current debates in Chinese and global economic history. We also hope that these volumes will serve both as a standard reference and as a resource for teaching.

39

Debin Ma (ed.), Special Issue: Money, Finance and Commerce in Chinese History, Frontier of Economics in China 13.3 (2018); Kris James Mitchener and Debin Ma (eds.), Special Issue: A New Economic History of China, Explorations in Economic History 63 (2017).

12

https://doi.org/10.1017/9781108587334.001 Published online by Cambridge University Press

part

I

*

BEFORE 1000

Published online by Cambridge University Press

Published online by Cambridge University Press

1

The Economy of Late Pre-imperial China Archaeological Perspectives lothar von falkenhausen

Continental East Asia during the first millennium B C E transitioned from a redistributive “gift-giving economy” (or “prestige-good economy”) to a thriving market economy that was at least partly monetized. This transformation – gradual but all-encompassing and irreversible – led to a veritable “economic miracle” during the Warring States period (c. 450–221 B C E), which brought unprecedented prosperity to large portions of the population. It will here be discussed through its reflections in the material record, spanning the eight centuries from c. 1000 B C E down to approximately the time of the Qin unification in 221 B C E.1 During this period, the Zhou kingdom and its constituent polities formed a relatively homogeneous culture area encompassing the middle and lower Yellow River basin and the middle Yangzi basin. Archaeological discoveries attest that, over time, many of the surrounding smaller and sociopolitically less complex regional cultures – defined by archaeologists on the basis of their material remains – were increasingly drawn into the Zhou orbit.2 The resulting homogenization was caused in part by the spread of Zhou modes of social and political organization, sometimes imposed by military conquest, but it also had an important

I am grateful to Richard von Glahn and Lothar Ledderose for insightful comments on a previous version of this chapter. 1 For coverage of the textual sources on this period, see in this volume von Glahn, Chapter 3, and the chapters by Hara and Kakinuma. 2 For overviews of Zhou archaeology, see Zhongguo shehuikexueyuan kaogu yanjiusuo 中國社會科學院考古研究所, Zhongguo kaoguxue: Liang Zhou juan 中國考古學：兩 周卷 (Beijing, Zhongguo shehui kexue chubanshe, 2004); Lothar von Falkenhausen, Chinese Society in the Age of Confucius (c. 1000–250 B C): The Archaeological Evidence (Los Angeles, Cotsen Institute of Archaeology Press, 2006).

15

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

1 Sanyuanzhuang 2 Shou Xian 3 Qijiacun 4 Houma 5 Doufucun 6 Qiaocun 7 Shouwangfen 8 Linzi 9 Qufu 10 Yan Xiadu 11 Handan 12 Lingshuo 13 Xinzheng 14 Yangcheng 15 Tonglüshan site cluster 16 Zhongba site cluster 17 Jinancheng 18 Xiaotianxi 19 Luoyang

7

46 3 5

10 12 8 11 9 1 13

19

2

14 17 15

16 18

Map 1.1 Archaeological sites mentioned in Chapter 1

economic dimension, reflecting the expansion of the distribution and exchange systems focused upon production centers in the Zhou realm. Below I shall present a string of archaeological case studies that may help us comprehend economic tendencies in late pre-imperial China concretely as they unfolded. (For the locations of sites mentioned in this chapter, see Map 1.1.) Since most of the available archaeological data have been gathered unsystematically, they cannot illuminate the quantitative dimensions of the phenomena observed, but the evidence at hand is sufficient to pinpoint underlying qualitative and psychological factors that influenced economic behavior.

Agriculture As in all ancient societies, agriculture constituted the basis of the early Chinese economy, and one may assume that the vast majority of the inhabitants of the Zhou realm – well over 90 percent of the population – were engaged in fulltime agricultural work. Even so, the rural economy of the Zhou period 16

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

remains an archaeological blind spot. In recent years, the application of advanced scientific methods has yielded many important new insights into the geographical origins and early domestication of various plants and animals in continental East Asia. But virtually all this research has been concerned either with the transition to agriculture during the tenth–sixth millennia B C E, or with the impact of the trans-Asiatic exchange of domestic animals and cultigens that occurred about 2000 B C E, which brought millet and possibly chickens to the West and wheat, barley, cattle, sheep, and horses to East Asia. By contrast, agriculture, animal husbandry, food production, and foodsupply chains during the Bronze Age remain severely understudied. The archaeological evidence available to date is virtually limited to a small and haphazardly gathered body of possible agricultural tools. Interestingly, it seems that bronze was rarely used for agricultural implements: Chinese farmers retained an essentially Neolithic tool kit until the introduction of iron (see below). From the subsequent Han period, we have plentiful pictorial depictions of rural life in tombs, as well as, at Sanyangzhuang (Henan), a full-scale glimpse into a Han rural landscape: a swath of fields interspersed with peasants’ homesteads, preserved Pompeii-like under the sediments deposited by a major flood of the Yellow River. We lack equivalent evidence from Zhou times; and even though Han agriculture was the outcome of innovations during the pre-imperial period – including, in particular, the introduction of iron agricultural implements – it seems risky to use Hanperiod evidence to reconstruct the situation during the preceding centuries. Here lies a challenge to future research.

Urbanism and Commerce The evidence flows somewhat more plentifully for the nonagricultural population of late Bronze Age China. Archaeological discoveries attest significant changes in urban settlement.3 Until approximately the middle of the first millennium B C E, cities – even royal capitals – were the nodes in a network of lineage-based polities; normally, each polity had only one urban center, in which residence was largely kin-based and the economic life, aside from fulfilling subsistence and defense needs, was mainly oriented toward ritual activities. The final centuries of the Zhou period saw the transition to centrally administered territorial states that comprised a tiered 3

Xu Hong 許宏, Xian Qin chengyi kaogu 先秦城邑考古, 2 vols. (Beijing, Jincheng chubanshe, 2017).

17

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

hierarchy of cities within their boundaries, enabling far greater administrative as well as economic integration. Between c. 600 and 250 B C E, the number of cities multiplied by a factor of at least four, and possibly as much as ten (accounting for regional differences); residence within them became far more diverse, increasingly based on occupation rather than kinship; and the role of cities as markets and production centers took on a much greater importance than before. From an archaeological perspective, one can perceive these changes through the growth in the physical size of cities (especially the major capitals, some of which now had populations in the hundreds of thousands). Moreover, changes in the urban form indicate a diversification of functions; in the capitals, for instance, the ruling elite now resided in separately walled palace cities (gongcheng 宮城) that existed side by side with much larger outer cities (guocheng 郭城) for the ordinary population. Arguably, it is these developments that brought about, for the first time in Chinese history, something like an “urban culture” that made cities categorically distinct from the surrounding rural landscape. Concomitantly, Chinese cities became more similar than they had previously been to cities elsewhere in the ancient world. Among urban dwellers, it was the merchants who benefited most from the relative geographic and social mobility of the Warring States period. Their traces have so far been difficult to identify on the ground, but one indication of their activities is coins, which came into increasingly wide circulation about 500 B C E or slightly earlier.4 Current discoveries suggest that the earliest cast bronze coins in East Asia were issued by merchant communities in various north China urban centers which are named in their inscriptions; they were for the most part located in the realms of Jin and its successor states – Wei, Hann, and Zhao – as well as in Zheng, Yan, and Qi. The coins are shaped like miniature spades and knives, probably harking back to an earlier stage in which actual bronze tools of standardized shape and weight had been used as units for reckoning value in transactions. The geographical distribution of these coins, as revealed by archaeological discoveries, can serve as a basis for reconstructing trading patterns; these differ from coin type to coin type as some coins were restricted to local transactions whereas others could be used interregionally. 4

See Kakinuma’s chapter in this volume. The authoritative work on the archaeology of pre-imperial Chinese coinage is Emura Haruki 江村治樹, Shunju¯ Sengoku jidai seido¯ kahei no seisei to tenkai 春秋戰國時代青銅貨幣の生成と展開 (Tokyo, Kyu¯ko Shoin, 2011).

18

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

It was only a matter of time before state governments also issued coins and started to regulate the monetary system; in particular, the use of specially designated coins was found practical in assuring the supply of the huge armies maintained by the Warring States kingdoms. About 300 B C E, the state of Qin, a relative latecomer to coinage, was the first to require the payment of a poll tax in coins, thereby ensuring that its characteristic banliang 半兩 coins – round with a square hole, a shape that was to remain standard throughout imperial Chinese history – came into universal usage within the expanding Qin realm. In the southern kingdom of Chu, bronze coinage took the form of miniature cowrie shells, another erstwhile proto-currency that archaeology has shown to have fallen out of use at just about the time when bronze coins first appeared. These coins existed side by side with a gold currency of rectangular money blocks that could be cut up into smaller units when needed. Silver also was experimented with for monetary purposes in Chu at the very end of the Warring States period, apparently for the first time in China. The other Warring States kingdoms are also known to have used gold bullion as a currency alongside bronze coins, but our only archaeological evidence for a precious-metal currency during pre-imperial times comes from Chu. From the peculiarities of the Chu monetary system, one may infer that Chu intended to define itself from its northern rivals as a separate economic area. By contrast, the currencies of the northern kingdoms, in spite of considerable differences in the shape, weight, and designated usages of their coins, were mutually compatible at least to a certain extent, potentially facilitating cross-boundary trade. Another fascinating glimpse into the activities of Warring States-period merchants is offered by the E Jun Qi tallies 鄂君啟節, collected in 1957 and 1960 in the vicinity of Shouxian (Anhui) and dated to 323 B C E.5 Made of bronze in the shape of plaques of bamboo and featuring gold-inlaid inscriptions in elegant formal script, the five tallies found formed part of two distinct fivepart sets of “boat tallies” and “wagon tallies.” They had been issued at the Chu capital to merchants working under the authority of the Lord of E, a regional administrator within the Chu bureaucracy (the exact location of E is disputed). When shown to local administrators, such tallies were to exempt merchants from road tolls or excise along certain explicitly defined trading routes for the duration of one year. Interestingly, the trade routes – 5

Lothar von Falkenhausen, “The E Jun Qi Metal Tallies: Inscribed Texts and Ritual Contexts,” in Martin Kern (ed.), Text and Ritual in Early China (Seattle, University of Washington Press, 2005), pp. 79–123; q.v. for further references.

19

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

northward overland for the “wagon tallies,” mostly southward by river for the “boat tallies” – seem to have been determined with the aim of tying the recently acquired borderlands of the Chu realm, and the cities within them, into a trading network centered upon the Chu capital, thereby strengthening the kingdom’s internal cohesion. That there was some concern over the loyalty of the northern border populations is evident from the fact that the “wagon tally” inscriptions expressly forbid trading in items of potential military use, such as metal, leather, or bamboo-shafted arrows. As to the volume of the trade, each tally was good for fifty conveyance units, with the understanding that one standard-size “boat” or “wagon” could be replaced, respectively, by specified equivalent numbers of smaller boats or animal or human carriers. It has been estimated that each “boat” could transport up to eighteen metric tons and each “wagon” between 1.5 and 2.7 tons of goods. What was transported remains largely unspecified, though the boat expeditions, at least, seem to have involved livestock. The merchants had to pay applicable excise taxes at the Chu capital at the end of their year-long expeditions, at which point the tally sets were reassembled and reissued. The tally inscriptions further specify that the merchants were ineligible to stay at guest houses for traveling government officials; it appears that they conducted their trade largely at their own risk and responsibility. Fascinating both for their inscribed information and for their materiality, the E Jun Qi tallies remain unique in the archaeological record. Hence we do not know how large a proportion of commerce in Chu took the form described here; when this kind of semi-official, governmentally managed trade originated; and whether the other Warring States-period kingdoms also practiced similar forms of trade. More generally, however, archaeology does attest the use of tallies in administrative as well as economic contexts all over the Zhou realm.

Workshops: Mass Production and Quality Improvements Whereas the archaeological traces left behind by merchants and their places of transaction are fleeting and mostly indirect, the excavation of cities has revealed considerable evidence about the places where the goods that were traded had been produced. Urban workshop sites from first-millennium B C E China attest to a pervasive shift to labor-saving modular production methods. Their implementation brought not only huge increases in output, but also, 20

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

counterintuitive though it may seem, palpable improvements in product quality. In the following, three examples will be discussed at some length.

Qijiacun A relatively early instance of the tendency toward standardization and mass production may be observed at a workshop specializing in the manufacture of stone slit rings excavated at Qijiacun, in Fufeng (Shaanxi province).6 This workshop was part of an industrial zone that also comprised workshops for ceramics, bronzes, and bone artifacts, and which was located in the eastern part of the Western Zhou metropolitan area in the Plain of Zhou (Zhouyuan). Given their location, it seems likely that these workshops served the needs of the royal Zhou court and the various prominent aristocratic lineages who maintained residences in the Zhouyuan area. There has been some debate about the status of the personnel employed here; most likely, in continuity with the preceding Shang period (c. 1600–c. 1046 B C E), they were, at least initially, “attached specialists” working exclusively for their patrons in a redistributive economy. The published archaeological report on the Qijiacun workshop covers 845 square meters of what was originally a somewhat larger compound, excavated in 2002. The site was occupied from predynastic Zhou (i.e., late Shang) times onward and was abandoned at the end of the Western Zhou period; lithic production commenced sometime toward the end of the early Western Zhou period (c. 1046–950 B C E ) and peaked in late Western Zhou (c. 850–771 B C E). Excavated features include five structures (three of which may have been workshop buildings, all dating to middle Western Zhou (c. 950–850 B C E )) and ninety-seven refuse pits, forty-two of which contained lithic-production remains, as well as forty interspersed contemporaneous tombs of different ranks. Twenty-two of these tombs contained stoneworking tools and/or unworked lithic material, justifying their identification as artisans’ tombs. The finds from the refuse pits suggest that, aside from lithic production, a variety of other activities went on at or near the site. Large quantities of household ceramics as well as animal and plant remains attest to food storage and consumption (whether the food was actually prepared at the site while it 6

Shaanxisheng kaogu yanjiuyuan 陜西省考古研究院 et al., Zhouyuan: 2002 niandu Qijia zhijue zuofang he Licun yizhi kaogu fajue baogao 周原：2002 年度齊家制玦作坊和禮村 遺址考古發掘報告, 2 vols. (Beijing, Kexue chubanshe, 2010); Zhouyong Sun, Craft Production in the Western Zhou Dynasty: A Case Study of a Jue-Earrings Workshop at the Predynastic Capital Site, Zhouyuan, China (Oxford, Archaeopress, 2008).

21

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

functioned as a workshop is unclear, as no hearths were found). Spinning whorls and loom weights indicate textile making. Numerous used divination plastrons and bones (one even featuring an inscription) suggest that divination ceremonies took place. And the presence of casting molds seems to indicate the practice of metallurgy at this site. But, for the most part, the Qijiacun workshop seems to have specialized in the manufacture of tiny slit rings (jue 玦) made of stone. These constitute the lion’s share of wasters – unfinished artifacts that had been damaged during manufacture or were considered otherwise imperfect – that were unearthed at Qijiacun, amounting to approximately 30,000 items. A few wasters of other kinds of artifacts made of common stone were also found, and small amounts of unworked jade and shell may indicate that those materials also were processed at the workshop. Surveys have established that the stone material used for the slit rings – marlite (two varieties), schist, and some limestone – was quarried locally on the slopes of Mount Qi or from cobbles recovered in nearby streams. Based on his careful investigation of tens of thousands of wasters and more than 800,000 pieces of production debris, Sun Zhouyong has reconstructed the operational sequence and calculated production statistics. Aside from clarifying the production technology per se, his findings show significant changes over time. In particular, Sun found clear and quantifiable evidence of increasing standardization and optimization of quality: over time, the jue manufactured at Qijiacun became more uniform in shape, as well as thinner and thus less wasteful of material. Moreover, having started out with a diversity of lithic materials, by late Western Zhou times the jue manufacturers had developed an almost exclusive preference for schist. Even more impressive are the period-by-period differences in the volume of production. Even though the six early Western Zhou pits with lithicproduction refuse amount to one-seventh of such pits in the excavated area, the material found in them is paltry, constituting less than 0.5 percent of the total. Since neither of the two early Western Zhou semi-subterranean houses excavated yielded any material related to lithic production, they are believed to have been residential structures rather than workshops. The middle Western Zhou period saw a peak in the number of refuse pits related to lithic production at Qijiacun: two-thirds of such pits date to that period. Since the three buildings dated to middle Western Zhou contained material related to lithic production, they have been identified as workshops. But even though the amounts of debris found soared to more than forty times that of early Western Zhou, each of the twenty-eight middle Western 22

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

Zhou lithic-production refuse pits yielded a relatively small amount of debris (some 5.6 kilograms on average). Still, production had obviously picked up in intensity. During the late Western Zhou period, lithic production at Qijiacun once again experienced explosive growth. This is true even though the number of refuse pits related to lithic production is relatively small, and no building remains dating to that period were found within the excavated area of the workshop. The volume of production, however, was enormous: fully 62 percent of all the lithic-production debris and 85.5 percent of all jue wasters found at the Qijiacun workshop came from two adjacent pits in the northwestern quadrant of the workshop area, H21 and H22. The average amount of refuse is 88.3 kilograms per pit, more than fifteen times the middle Western Zhou figure. Parallel to the increase in jue production, the number of personnel engaged in their manufacture also seems to have increased: six of the ten datable artisans’ tombs (which possibly comprised fourteen out of the altogether twenty-two tombs) date to late Western Zhou, as opposed to three for middle and one for early Western Zhou. Even so, the number of tombs did not rise proportionally with the enormous increase in the volume of production noted above, possibly confirming Sun’s finding that more efficient and presumably labor-saving production methods had been adopted. Two difficulties must be mentioned. First, the excavated area does not comprise the entirety of the Qijiacun lithic workshop, although the report states that the excavations concentrated on the part of the site where production-related deposits were densest. Still, it is possible that the data reported are not altogether representative. Second, embarrassingly, we have no idea of the intended usage of all these innumerable stone jue. Virtually no specimens have been found in contemporaneous archaeological contexts. That none were discovered in the more than 1,100 Western Zhou tombs excavated in the Plain of Zhou and adjacent areas (the only exceptions coming from some of the tombs at the Qijiacun workshop itself) seems to show that they did not serve for bodily adornment, unlike jade jue of identical shape, which were commonly used as earrings. Being small, highly breakable, and made of a cheap material, these mass-produced artifacts may have had an ephemeral use-life. Since Qijiacun was located in the immediate vicinity of a large-scale bonemanufacturing workshop, one is tempted to speculate that the jue might have been joined to bone objects in some sort of composite artifacts. But no such artifacts have yet been discovered. 23

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

One naturally wonders how representative our observations at the Qijiacun lithic workshops are for Western Zhou economic developments as a whole. The evidence for increased standardization, greater efficiency in mass production that enabled economies of scale, and a resultant explosive growth in the volume of production during late Western Zhou accords well with what one may observe, albeit without the support of comparable information from workshop sites, in connection with other types of Western Zhou artifacts, especially with products of the Western Zhou bronze-manufacturing industry. One may therefore cautiously suggest that the late Western Zhou surge in jue manufacture at Qijiacun is a reflection of the general economic developments during that period, prefiguring the even more complex and elaborate systems of mass production and quality control during subsequent centuries. Even though archaeology cannot speak directly to the sociological aspects of these developments, it seems likely that they were accompanied by a change in status of the workshop personnel – from “attached specialists” to, at least sometimes, independent producers – and also in the composition of their clientele, which, rather than being restricted to a small number of steady patrons, came to comprise ever larger segments of the population. This was, in any case, the general long-term trend over the course of the late Bronze Age.

The Houma Foundry Similar developments toward a much greater volume of production, more rationally structured production processes, and concomitant changes in the socioeconomic context of production are well documented at Eastern Zhouperiod bronze foundries. Such foundries have been found at the capitals of several of the major polities of that period. The best-known example so far flourished at Xintian, the capital of the Jin polity at present-day Houma (Shanxi).7 In operation between c. 585 and 376 B C E, it straddles the transition between the Spring and Autumn (770–c. 450 B C E ) and the Warring States periods (c. 450–221 B C E). It was very possibly the largest industrial operation in north China in its time; its products have been found throughout the former Jin territory, as well as in neighboring polities and in the northern steppes. Foundry remains have been discovered in a half-dozen localities all over the Xintian capital. The densest concentration has been identified within 7

Shanxisheng kaogu yanjiusuo 山西省考古研究所, Houma zhutong yizhi 侯馬鑄銅 遺址, 2 vols. (Beijing, Wenwu chubanshe, 1993).

24

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

a survey area of ninety-six hectares directly to the south of the Jin rulers’ palace enclosure. The total extent of that foundry – now known simply as the “Houma foundry” – was even larger, probably also encompassing a foundry site further to the southwest of the survey area that has been separately reported on as the “Baidian foundry.” Based on excavations undertaken in 1959–1964, the archaeological report on the Houma foundry covers only two small portions: Locus I I, of which 2,100 square meters (28 percent of its total area) have been excavated, and the considerably more extensive Locus X X I I, with an excavated area of c. 2,600 square meters (12 percent of the total). Whereas Locus I I (as well as the so-called “Baidian foundry”) yielded evidence of the manufacture of ritual vessels as well as other decorated objects apparently destined for elite users, Locus X X I I was mainly specialized in the production of bronze weapons and tools. Surface probing has revealed that other, as yet unexcavated, parts of the foundry were dedicated to casting spade-shaped coins, among other kinds of objects. The excavated site constitutes a maze of interpenetrating building remains, activity floors, and refuse pits. Through the analysis of their stratigraphic relationships, one can trace the development of the workshops from their establishment shortly after the foundation of the Xintian capital through their decline and abandonment around 400 B C E. Evidently, the foundry workers lived at the site. Intermixed with metalworking remains, artifacts related to their daily-life activities were recovered in considerable quantities. At Locus X X I I, buildings and pits were grouped around a roughly T-shaped activity floor (reported as a “road”) of three to four meters’ width, which was twice paved over after it had been laid out initially, and which, during the final phase of occupation, came to be intruded upon by pits and tombs. That part of the site comprised fifty buildings, five of which were entirely subterranean, the rest semi-subterranean. The vast majority of them were rectangular and comprised a single room of five to nine square meters, sometimes with attached storage spaces dug laterally into the walls. A minority of semi-subterranean structures were round in shape, and there were four structures consisting of more than one room. Some of these buildings served as dwellings, others as workshops or storage facilities. Locus X X I I comprised, furthermore, forty-four subterranean storage pits of different shapes, twenty-one wells, 467 refuse pits, a kiln for the firing of casting molds, and a pottery-making kiln. Locus I I and the “Baidian foundry” comprised similar remains in smaller numbers. Although most of the metal processed at the Houma workshops was probably transported there in the form of ingots, having been smelted near 25

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

the mines, there are indications that a certain amount of smelting (or resmelting) and alloying went on at the site. For the most part, however, the workshops were devoted to bronze casting. The very voluminous casting debris includes raw materials, implements, fragments of crucibles and bellows, remains of casting cores, and the astonishingly well-made ceramic models, pattern blocks, and molds that were used in manufacturing ornamented bronzes. This evidence is of the utmost importance, for it attests that the bronze-manufacturing process at Houma had been divided into many small steps, each assigned to a specialized worker or team of workers. This rationalization of the operational sequence was enabled by a significant technological innovation, which may be described as follows. Stylistically, middle to late Spring and Autumn-period bronzes are characterized by the multiplication of ornamentation bands, consisting of extremely miniaturized versions of interlaced dragons or serpents. Sometime after the turn of the sixth century, these endlessly repeated units of decor came to be produced in a new technique, known as the pattern-block technique, which in time revolutionized both the production process and the visual appearance of the products.8 Pattern blocks – master versions of the decoration units – were carved in clay and then baked; from them the manufacturers could obtain a theoretically unlimited number of negative impressions. Unbaked clay strips with such impressions were fitted into designated spaces in the casting molds, which were then fired in their entirety. This resulted in a very significant economy of labor over the traditional procedures that had been practiced at least since Shang times, in which the decoration had been engraved individually for each object. (It is currently being debated whether this was done in the negative and directly onto the casting molds, or in the positive on a model from which the outer molds were then obtained by impression, whereupon the model would be shaved down to serve as the core during casting.) Although the casting process itself – the time-honored piece-mold casting method – remained unchanged, the use of pattern blocks enabled significant economies of labor: workshops could now make do with but a small number of skilled artisans capable of designing and carving decoration patterns. This in turn facilitated the large-scale production of high-quality vessels with visually striking decoration, opening the way for the commercialization of bronze production and trade. The wide geographical spread of Houma-style 8

Robert W. Bagley, “Replication Techniques in Eastern Zhou Bronze Casting,” in Stephen Lubar and W. David Kingery (eds.), History from Things: Essays on Material Culture (Washington, DC, Smithsonian Institution Press, 1993), pp. 231–41.

26

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

bronzes bespeaks the economic success of this new approach to bronze production. By comparison to the likely subaltern position of the “attached specialists” working at the courts of earlier rulers, the reorganization of production enabled by the new technology must have entailed significant changes in the social and presumably legal status of the artisans involved – including, probably, a growing divide between a small elite of highly trained specialists and a much larger number of ordinary workers. In addition, the complexity of the production sequence likely required ever closer supervision and management by non-worker administrators. The observable status differences among the workers’ tombs associated with the Houma foundry may bespeak this situation. Whether the Jin government was at all involved in the running of the foundry is unknown; the fact that its products circulated far and wide beyond the circles of the Jin ruling elite suggests, at any rate, that at least a certain amount of independent commercial decision making went on. When and where the pattern-block technique was invented remains unclear. Even though it was by no means always used – some bronze objects were still being individually made – it was eventually adopted by major foundries throughout the Zhou culture sphere, not only at Houma, but also at Xinzheng and Luoyang in central Henan, in Chu in the south, Qin in the west, Yan in the north, and Qi in the east. It developed in tandem with other technological innovations in the realm of bronze manufacture. To mention just some of them: (1) the various appendages of vessels, such as legs and handles, were manufactured in standard sizes separately from the vessel bodies and subsequently soldered or mechanically affixed onto the latter, enabling various kinds of combinations in a modular process. (2) Batches of small bronzes (belt hooks, rings, and above all coins) were cast together in a single core-mold assemblage. (3) Some exceptional vessels were cast with an outer layer of openwork decor encasing the actual vessel body. (4) Some vessels now featured pictorial decor of ritual or hunting scenes instead of the traditional ornaments; the decor units were produced in the pattern-block technique and sometimes inlaid. (5) Many bronzes came to be inlaid with elaborate patterns in copper, precious metals, and other substances, exploiting multiple color effects. (6) Multi-media artifacts combining bronze, precious metals, glass, semiprecious stones, and other materials were produced, maximizing luxurious display. All of these departures from the earlier bronze manufacturing tradition exploited the advantages of the reorganization of foundries according to modular procedures. They very likely responded to consumer demand in 27

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

the market, forming and following fashion, and they instantiate a turning away from the earlier ritual contexts of usage and the transformation of bronze objects into luxurious household items. Concomitantly, the increasingly gaudy esthetics of Eastern Zhou-period metalwork attest a society-wide preoccupation with flaunting one’s personal wealth in domestic settings. This trend affected not only the traditional elites, but also the ordinary population, who had earlier been excluded from bronze use (see below).

Doufucun Our third example for the adoption of labor-saving modular processes in Zhou-period craft industries comes from a workshop at Doufucun, Fengxiang (Shaanxi), that specialized in the manufacture of architectural ceramics, notably tiles.9 Doufucun is located within the site of the great city of Yong, which served as the capital of the Qin kingdom between 676 and 383 B C E and continued to flourish thereafter. The architectural-ceramics workshop is part of an industrial zone in the northwest corner of Yong that also apparently contained a number of other workshops (e.g., for woodworking and carpentry, and for casting architectural fittings of bronze), all of which serviced the construction of temples/palaces and government buildings in and around the city. Excavated in 2005–2006, Doufucun is the only large-scale workshop from pre-imperial Qin to have been reported upon so far. The excavated area, some 1,375 square meters, does not by any means constitute the entirety of the workshop, which is likely to have extended farther in all directions. The occupation of the site began in the early phase of the Qin capital, with no earlier cultural strata below; the workshop flourished from the middle Warring States to the early part of the late Warring States period. Workshop-related remains are aligned along two parallel long rectangular pits or ditches, spaced some ten meters apart (a third one, running perpendicularly, was incompletely excavated). The excavators have reconstructed their use-life in three stages. At first they served as clay-extraction pits. After they had been dug to a certain level, they were turned into processing areas where the clay was sifted, mixed with water (blunged), kneaded, and formed into coils and blocks for further processing. In one pit, the excavators found five regularly spaced heaps of finely sifted clay ready for blunging, as well as heaps of coils and blocks of unfired clay. In its third stage of usage, that pit 9

Shaanxisheng kaogu yanjiuyuan 陝西省考古研院 et al., Qin Yongcheng Doufucun Zhanguo zhitao zuofang yizhi 秦雍城豆腐村戰國製陶作坊遺址 (Beijing, Kexue chubanshe, 2013).

28

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

was used as a refuse dump and filled with wastage from nearby kilns, including misfired items, sherds of broken tiles, and ashes. Additional pits of this nature likely exist outside the excavated area; one imagines that, over the lifespan of the workshop, the locus of production shifted from one pit to the next, following the changes in their usage. Other excavated features clustered around the two long rectangular pits included twelve small pouch-shaped round pits that had apparently been used for storing clay; forty small cylindrically shaped round pits, possibly used for water storage; and fifty irregularly shaped semi-subterranean houses that seem to have served both as workshops and as workers’ dwellings. Moreover, there were two cisterns (or wells?), as well as a large rectangular working floor measuring thirty-two by twenty meters, possibly used for drying blocks of clay. Surprisingly, only four small kilns were found. They had been fired with straw and wood, with no indication of charcoal use. Given the quantity and density of the other associated manufacturing remains, additional, probably larger, kilns must have existed nearby in the as yet unexcavated parts of the site. The excavations yielded various production tools, such as ceramic paddles that produced a rough surface pattern, and molds for imprinting eaves tiles. The presence of sundry household goods, such as pottery vessels, spinning whorls, and some objects of apparently ornamental function, attests that, as at Qijiacun and Houma, the workers and their families lived at the site; this seems to have been the common practice at Zhou-period workshops. The products of the Doufucun workshop include the full range of architectural ceramics needed for Eastern Zhou-period temple/palace buildings. Above all, these include roof tiles: convex imbrex tiles (tongwa 筒瓦) of semicircular cross-section, which, on a roof, were placed in alternation with slightly concave tegula tiles (banwa 板瓦), forming channels for the rainwater to flow down. Moreover, the excavators found “clay-nails” with ornamental tops that served to hold the tiles in place, ornamented semicircular facing tiles that could be mounted on a building’s walls, and drainage pipes and a small number of bricks of different kinds (these are, incidentally, the earliest bricks encountered anywhere in China proper). But the most plentiful finds from the workshop were eaves tiles with decorated flat ends (wadang 瓦當), which served to adorn the lower edges of a tiled roof. All of these were imbrex tiles. Their total number is 2,083; thirty-five of these had semicircular ends (banwadang 半瓦當), whilst the remainder – the vast majority – had round ones. Among the latter, the largest number (some 1,576 pieces) featured various types of animal decor: 29

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

dancing phoenixes, badgers, tigers, and stags, sometimes in combination with snakes, leopards, fish, geese, toads, and dogs. The remainder either were plain (316 instances) or featured various kinds of geometric design or, in two cases, auspicious writing. Except for two specimens with incised and thirty with cord-impressed patterns, the decoration on the wadang was mold-impressed; fourteen molds, made of fired clay, were found at the site, facilitating the production of large numbers of identical-looking specimens. The preponderance of eaves tiles among the architectural ceramics from the Doufucun workshop suggests that the excavated part of the site happened to be the workshop’s wadang-producing department. Whereas tiled roofs had begun to replace thatched ones in the Shaanxi area by c. 1100 B C E, ornamented wadang featuring regionally distinctive decoration motifs were a relatively recent addition to the material culture of the Zhou realm, going back no further than the middle of the first millennium B C E. Their manufacturing process was tightly regimented, comprising ten steps from clay extraction to the removal of the finished products from the kiln. In spite of the small size of the kilns found in the excavated area, there can be no doubt that production at the Doufucun workshop as a whole was large-scale and tightly managed, involving a large number of workers and a high degree of division of labor. Over time, moreover, the manufacturing technique underwent a process of rationalization. The excavators reconstructed three ways for producing wadang on a potter’s wheel: (1) putting the flat circular wadang plate with an impressed design facing downward, building a clay cylinder on top of it, and then splitting the entire cylinder in half (this process yielded two eaves tiles with semicircular banwadang; obviously, their impressed design had to consist of two separate units); (2) the same procedure but splitting the cylinder in such a way as to leave the round bottom in its entirety attached to one of the two half-cylinders (this yielded one eaves tile with a round wadang plus one ordinary imbrex tile); and (3) making round wadang plates separately from cylindrical tubes, splitting the tubes in half and attaching a wadang to each of the resulting semi-cylindrical tiles with strips of clay before firing (this potentially yielded two eaves tiles with round wadang per cylinder). The stratigraphy at Doufucun confirms that these three processes emerged successively: imbrex tiles with semicircular wadang preceded those with round wadang (which emerged sometime during the early Warring States period) and the procedure for making the latter subsequently became modularized as described. 30

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

Their weight probably made it impractical for the products of the Doufucun workshop to circulate over long distances. Since the site remains incompletely excavated, it is difficult to assess the scale of production there; nor do we know whether other competing workshops existed nearby. It seems likely that the workshop produced mainly for the Qin government. This is suggested both by its location and by the fact that most of the wadang patterns attested at Doufucun have been identified at official temple/palace buildings in and around Yong. On the other hand, the lack of artisans’ stamps, which are commonly seen on Qin ceramic products with official associations, would suggest that the Doufucun workshop – unlike other, state-run workshops in pre-imperial and imperial Qin – operated without direct involvement of government personnel. Interestingly, moreover, the different wadang molds are not entirely uniform in size: they vary from fifteen to eighteen centimeters in diameter. While this variation is not dramatic, it shows that buildings and their roofs must have been designed individually, rather than following some sort of overarching standard module. But this, too, changed during the later part of the Warring States period; archaeological investigations at Yong and elsewhere in the core of the Qin kingdom have shown that over time, eaves tiles became slightly larger (up to twenty-one centimeters in diameter in imperial Qin and Han times), as well as increasingly standardized. To what extent the Doufucun workshop can serve as a paradigm for Qin craft industries as a whole must remain uncertain in the absence of any directly comparable archaeological material. But the quest for greater efficiency in manufacture evidenced by the adoption of modular processes seems to have been a running theme in Qin economic history just as it was elsewhere in the Zhou culture sphere. Warring States-period Qin bronzes, for instance, exhibit the same tendency. The First Emperor’s now worldfamous terracotta army, as well, is known to have been made in part by workshops – some government-run, some private – that were normally engaged in the manufacture of tiles and water pipes. Lothar Ledderose has highlighted it as a paradigmatic case of the use of modular production techniques in ancient China.10

Discussion Since archaeology cannot determine whether a given workshop functioned as a private enterprise or was managed by government authorities, we have 10

Lothar Ledderose, Ten Thousand Things: Module and Mass Production in Chinese Art (Princeton, Princeton University Press, 2001), pp. 51–73.

31

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

no direct way of knowing whether, in the three cases discussed above, the significant innovations that brought about greater efficiency in production and quality control were imposed by market pressures or by fiscal considerations. The location of all three workshops at major capitals might suggest some connection to the political powers of the time, but such a connection could take many conceivable forms. To understand them better, one must hope that future excavations will yield comprehensive information on workshops located in minor cities, towns, and villages that will permit us to tease out any special characteristics of the metropolitan workshops discussed herein. Perhaps the dichotomy implied in our framing of the issue is a false one; for instance, the E Jun Qi tallies, considered above, seem to attest a combination of both private initiative and government management in connection with economic activity. Nevertheless, the evidence discussed – especially the finds from the Houma foundry – may be said at least to be generally compatible with the current consensus of historians that the later centuries of the Eastern Zhou period, following upon the collapse of the traditional lineage system, were a time of relative socioeconomic fluidity, in which there was more room for market-driven (“private”) initiative than either before or afterward. This reflects, of course, general social developments during the period, in particular the progressive weakening and eventual collapse of the traditional aristocratic order. The great expansion of the bronze industry during that time reflects, on the one hand, an emerging nonelite customer base; at the same time, it suggests changes in the prevailing cultural and religious attitudes toward metal: the removal of an apparent taboo that, before c. 850 B C E, had limited it to items of ritual and military usage restricted to the aristocracy. This process can be traced through several stages. It started with the abandonment of religiously charged decoration motifs on bronze vessels that occurred as part of a major ritual reform at the beginning of the late Western Zhou period (c. 850 B C E).11 Henceforth bronze decoration was transformed into pure ornament, enhancing the saliency of the objects themselves and emphasizing their materiality, rather than drawing the viewers’ attention to the motifs represented. Not accidentally, this was also the time when bronzes with functions unrelated to ritual and warfare made their first appearance in the archaeological record. Over the course of the following centuries, even though ritual vessels, weapons, and horse-and-chariot

11

On this see Falkenhausen, Chinese Society, pp. 29–73 (q.v. for further references).

32

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

furnishings continued to be prominent in the archaeological record, the range and variety of bronze objects greatly expanded. A second ritual restructuring during the middle Spring and Autumn period (early to mid-sixth century B C E ) brought further changes.12 Bronze foundries now produced increasing amounts of small metal items that were within reach even of people of limited economic means. Our best evidence for this situation comes from a commoner cemetery at Qiaocun, Houma (Shanxi).13 Located near the Xintian capital, but post-dating the fall of the Jin polity, Qiaocun served as the resting place of mostly low-ranking inhabitants of the nearby second-tier city of Jiang, known today as the “Fengcheng Ancient City.” Metal objects were found in 513 (54.3 percent) of the 942 tombs excavated, with iron (seen in 325 tombs) taking a slight preponderance over bronze (248 tombs; both metals co-occur in some tombs). These are small items – mirrors, tools, fittings, small bells, and especially belt hooks (the latter sometimes inlaid with gold); in other words, consumer items. They are seen with both male and female tomb occupants, albeit slightly more frequently with males. Even some of the shackled human victims deposited in the ditches surrounding the more prominent tombs at the Qiaocun cemetery retained their own metal belt hooks in death. It may not be too optimistic to interpret this situation as indicating an overall rise in economic prosperity and living standards that was likely caused by the burgeoning market economy of the period; precisely the sort of situation that justifies speaking of a “Warring States economic miracle.” Possibly the best proxy evidence attesting that bronze objects were now produced for the market is mirrors, rarely encountered before c. 500 B C E, but made in large numbers thereafter. They exist in different sizes, and they show a great variety of designs, as well as several levels of luxury in execution (single-plate mirrors with plain backs, single-plate mirrors with relieved ornaments on the back, single-plate mirrors with attached silk backings, and double-plate mirrors with inlaid or lacquered decoration and/or embedded pieces of silk brocade). In their totality, the observable differences among mirrors are probably best explained as evidence of market-driven competition, consumer preferences, shifting fashions, and differences in consumer wealth. The trends noted above in the esthetics of Warring States-period bronzes in general may be interpreted in the same sense. It is no accident that cast bronze coinage also first came into circulation about this time. 12 13

Cf. Falkenhausen, Chinese Society, pp. 326–69. Shanxisheng kaogu yanjiusuo 山西省考古研究所, Houma Qiaocun mudi 侯馬喬村 墓地, 3 vols. (Beijing, Kexue chubanshe, 2004).

33

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

Similar considerations apply to other late Bronze Age craft industries, which for various reasons are less well documented than bronzes: textiles above all, but also lacquer, jade, glass, precious metals, leather, and wood. For some of these, in the absence of workshop sites, archaeologists can trace how they were incorporated into a market-driven consumer economy through the typological development of the manufactured products. In other cases, such an analysis is precluded by a dearth of material, often due to preservationrelated factors. This applies even more to perishable goods such as foodstuffs, which no doubt were also distributed through the same market networks. In connection with the demystification of bronze, I would like to point out a parallel process affecting another technology of great economic relevance: writing. Like bronze, writing in Shang and Western Zhou times had been limited to a narrow range of uses linked to high status and ritual; but, over the course of the first millennium, its use gradually expanded to record-keeping in administrative and commercial contexts. This development also created the possibility for the compilation of China’s earliest books, which covered a variety of subjects ranging from poetry and political philosophy to technology and proto-science. Written on bamboo strips or, more rarely, silk, some such books have been excavated from Warring States-period tombs. From them, Li Ling has been able to trace the rise of professional groups, each with its own written corpus, their written form making relevant professional knowledge potentially accessible beyond the former kin-bound mechanisms of knowledge transmission.14 Without question, the proliferation of writing after the middle of the first millennium B C E was of tremendous importance to the contemporaneous transformation of the Eastern Zhou craft industries: it facilitated bookkeeping as well as administrative control, both within workshops and from the outside (e.g., by government authorities). Since bronze production was the foremost craft industry of the age, one may argue that the simultaneous desacralization of metal and writing cleared the way for market-driven production as well as increasingly sophisticated modes of management and accounting throughout the Zhou economy. It bears pointing out that these developments in economic behavior were triggered by changes in social organization and the religious beliefs that subtended them. More indirectly, the removal of the taboo on the use of bronze may have also cleared the way for the highly consequential adoption of another metal that lacked the erstwhile religious associations of bronze: iron. 14

Li Ling 李零, Zhongguo fangshu zhengkao 中國方術正考 (Beijing, Dongfang chubanshe, 2006); Zhongguo fangshu xukao 中國方術续考 (Beijing, Dongfang chubanshe, 2000).

34

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

The “Iron Revolution” The term “Iron Age” is seldom used in Chinese archaeology, even though iron became known in continental East Asia as early as the mid-second millennium B C E and was in common use by the middle of the first millennium B C E. As in the earlier case of bronze (introduced about 2000 B C E), iron technology came originally from the west, but was fundamentally transformed in China. While bronze remained the prestige material par excellence (as evident, inter alia, from its use in coinage), the availability of cheap iron weapons and implements is often said to have made warfare more deadly and agriculture more productive from approximately the late Spring and Autumn period onward (i.e., after c. 550 B C E).15 In other words, the emergence of a highly developed iron industry must have been a major ingredient in the “Warring States economic miracle.” One would think that this “iron revolution” should be a topic eminently suitable to archaeological investigation and clarification. In fact, however, properly archaeological research into this important topic is still in its infancy. This is due in part to preservation-related factors – iron corrodes easily – and also to cultural attitudes at the time, as a result of which iron objects were rarely deposited in tombs. Since, for better or worse, the lion’s share of archaeological data from the period under consideration comes from funerary contexts, this has created a significant bias. So far, what archaeology can do is to attest the presence of iron in all parts of the Zhou realm, as well as in some surrounding non-Zhou areas, and to register an overall increase in its occurrence after c. 550 B C E. Based on the scrutiny of individual iron objects, archaeology can, furthermore, trace the technological transition from hammered iron to cast iron, which occurred under the influence of China’s long-established bronze-casting tradition (rooted in turn in prehistoric traditions of manufacturing high-quality ceramics). This development was the precondition for the industrial-scale mass production of standardized cast-iron objects during the Warring States period. In their search for ever greater economy, late Warring States siderurgists at the Shouwangfen ironworks in Xinglong (Hebei) switched to casting-molds for iron objects that were themselves cast of iron; though more expensive, they were reusable, unlike the earlier ceramic molds, which were only good for a single casting.

15

Donald B. Wagner, Iron and Steel in Ancient China (Leiden, Brill, 1993).

35

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

Iron workshops, some of them very extensive, have been located in several of the Warring States-period capitals – Linzi of Qi and Qufu of Lu in presentday Shandong; Xiadu of Yan, Handan of Zhao, and Lingshuo of Zhongshan in Hebei; and Xinzheng of Hann in Henan – as well as at several minor urban centers all over north China (including the already-mentioned Shouwangfen ironworks). The best-reported finds are those at Yan Xiadu, where four separate ironworks were excavated. The largest of them extended over seventeen hectares. One workshop specialized in the manufacture of iron weapons; at least two of the others simultaneously produced both bronze objects (in particular, coins) and iron artifacts. At Linzi, radiocarbon dates suggest that ironworking began as early as late Western Zhou times and continued uninterruptedly into the Han dynasty, but so far little can be said about the extent and nature of these workshops, or about their relative importance vis-à-vis those in smaller cities. It is interesting, in this connection, that the largest (twenty-three hectares) individual Warring States-period ironworks so far found – at Yangcheng, Dengfeng (Henan), then in the territory of the Hann kingdom – was in a medium-sized city in a strategically important location some seventy kilometers from the Hann capital. Given the strategic value of iron, it is possible that its production took place under some measure of state control, but none of the Warring States kingdoms, not even Qin, seems to have had anything comparable to the enormous (and, due to the unsustainable felling of north China’s forests in order to fuel them, ecologically permanently damaging) state-run ironworks of the Han dynasty. In spite of the ever-increasing overall number of individual iron objects discovered, archaeology so far can contribute very little to the evaluation of the quantitative claims implied in the notion of an “iron revolution.” In particular, it cannot assess the impact of the iron industry on its alleged two major constituencies – farmers and the military. While iron weapons, iron armor, and iron agricultural tools have been found in some numbers, the archaeological evidence for pre-imperial Chinese warfare is too scarce, and too unevenly distributed, to allow any insights beyond what we already know from texts. Interestingly, on the other hand, many archaeologically excavated iron objects were intended either for household use (e.g., iron cauldrons and iron-footed bronze tripods), or for personal adornment (e.g., iron belt hooks). Some of the latter could be of quite luxurious execution – e.g., inlaid with gold, silver, and semiprecious stones – putting into question the commonly accepted notion that iron was always a cheap material. 36

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

Patterns of Interregional Exchange The overall tendency for the Warring States kingdoms to incorporate surrounding areas was noted at the beginning of this chapter. This incorporation was preceded by many centuries of interregional economic interaction. Since the exchange relationships between the Zhou and their neighbors tended to be asymmetrical, one might attempt to analyze them in terms of a core– periphery model inspired by Braudel’s concept of “world economy” (économie-monde) and Wallerstein’s “world-systems” theory.16 From the standpoint of the dynastic centers in the Zhou cultural sphere, this would result in a picture of resource-consuming “cores” surrounded by resourceproviding “peripheries.” Actually, however, what the archaeological record seems to reflect is, above all, the multidirectional flow of goods among groups of different degrees of sociopolitical complexity all over continental East Asia. Multiple ecological zones with highly diverse populations were linked by a diffuse web of simultaneously operating relationships of different orders. What needs emphasis, above all, is the long duration of these relationships. More interesting than the eventual “acculturation” of “peripheral” areas are the ways in which socioeconomic entities in different regions often maintained their interactions over the long term, with neither exploiting or attempting to absorb the other. We shall now present two archaeologically well-studied examples of such patterns of interregional economic exchange: the copper mine at Tonglüshan, Daye (Hubei) and the saltworks at Zhongba, Zhongxian (Chongqing).

Tonglüshan The hills and mountains to the south of the Yangzi river in present-day Hubei, Jiangxi, and Anhui provinces hold extensive remains of copper mining going back to the late neolithic Post-Shijiahe Culture (c. 2400–1900 B C E). Over time, the scale of production increased, and by the time contemporary with the early part of the Shang dynasty (Erligang period, c. 1500–1300 B C E), the mining techniques already resembled those used in rural China until very recent times: vertical shafts were sunk into the ground, from which horizontal galleries branched out, following the veins of copper ore. Both shafts and galleries were reinforced by wooden frames of standardized size, composed using mortise-and-tenon joinery, sometimes with additional wickerwork or 16

Cf. Li Liu and Xingchan Chen, State Formation in Early China (London, Duckworth, 2003).

37

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

bamboo filling – a modular technology that seems uncannily modern. Typically less than one meter wide on each side, the shafts and galleries were just big enough for one human being to move around in. The copper ore was removed with the help of mining tools made of wood, bamboo, bronze, stone, and (later) iron, and hauled out of the mines in baskets. Lighting was provided by wooden or bamboo torches soaked in oil. During the Erligang period (possibly starting even earlier), the Shang kingdom maintained an outpost at Panlongcheng, near present-day Wuhan, very likely for the purpose of acquiring copper from the mines across the Yangzi. The mines themselves were exploited by people who were beyond the political control of the Shang. They delivered their copper to Shang trader-officials, likely in the form of ingots that were shipped to the foundries in the Shang capital. In exchange, among other things, finished bronze vessels – including some masterpieces of Shang bronze casting – found their way into the area, eventually spurring the rise of distinctive indigenous bronze-manufacturing industries. The cultural significance of these bronzes changed in the process of their transmission. Tonglüshan is the most extensive mining site in this large region to have been explored archaeologically.17 In an area extending over approximately one square kilometer, ten underground mines were found, comprising nearly 500 shafts and galleries and reaching to a depth of fifty-five meters or more below the original ground surface. Tonglüshan was part of a vast “mining landscape” that, aside from the mines, comprised copper-smelting sites as well as settlements of different sizes and functions. This landscape changed considerably over time. Mining at the Tonglüshan site goes back to at least the time contemporaneous with the late Shang period (c. 1200 B C E) and continued for at least a thousand years. About midway through this time span, the metallurgists realized that, rather than transporting the heavy copper ore to their homes in nearby villages for smelting, it would be more efficient to smelt it near the mines. Two copper-smelting sites with more than ten furnaces have been excavated in the immediate vicinity of the mineshafts at Tonglüshan. Since 17

Huangshishi bowuguan 黃石市博物館, Tonglüshan gu kuangye yizhi 銅綠山古礦業 遺址 (Beijing, Wenwu chubanshe, 1999). Many relevant studies, as well as preliminary reports on most of the sites discussed in this section, are collected in Dayeshi Tonglüshan gutongkuang yizhi baohu guanliweiyuanhui 大冶市銅綠山銅礦遺址保 護管理委員會, Tonglüshan gutongkuang yizhi kaogu faxian yu yanjiu 銅綠山古銅礦遺 址考古發現與研究, 2 vols. (Beijing, Kexue chubanshe, 2013). This section has greatly profited from Tao Shi, “Mining, Trade and State Formation in Early China” (Ph.D. dissertation, UCLA, 2018).

38

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

they are buried under large amounts of slag – conservatively estimated at more than 400,000 tons for the Tonglüshan core area alone – it is likely that the smelting area was much more extensive. The two sites excavated so far are said to date to the Spring and Autumn and the Warring States periods respectively. The inception of copper smelting near the mines, sometime during the time contemporary with the Spring and Autumn period, was part and parcel of a comprehensive reorganization of the metallurgical industry that brought about increased professional specialization, hierarchization, and administrative co-ordination, as well as a new concern with security. Concomitantly, finds from nearby settlements indicate that the casting of metal objects shifted away from local workshops, presumably to large foundries at faraway metropolitan centers. One such settlement is Dalupu, in Yangxin (Hubei), which flourished from the late Neolithic through the early Spring and Autumn period.18 It was a small village, some eight hectares in size, located in a small intermountain basin about fifteen kilometers southeast of the Tonglüshan mine and even nearer to several other, more ancient, mines. Among a total of 196 refuse pits excavated at Dalupu, fifty-five (28 percent) yielded pieces of smelting slag; moreover, seven possible smelting-furnace bases have been excavated. Aside from pieces of unsmelted copper ore and furnace debris, the associated refuse pits also yielded molds for metal tools, as well as some finished bronze items. Evidently, the metallurgists at Dalupu not only smelted copper from the nearby mines, but were also able to cast metal objects, presumably after having alloyed the copper with tin that they must have obtained from elsewhere. Even so, the excavated finds in their totality suggest that the mainstay of the Dalupu economy was agriculture, with textile and ceramic production playing a subordinate role. As to metallurgy, in spite of some evidence of increasing intensity over time, its remains only constitute a small proportion of the Dalupu archaeological assemblage: it was no more than a part-time, possibly seasonal, cottage industry, either household-based or undertaken by small groups of community members pooling their resources. A similar situation prevailed at other contemporaneous settlement sites in the area. The Bronze Age ceramics from Dalupu and nearby sites are locally distinctive and differ in both typology and manufacture from those of areas to the north of the Yangzi. The excavators – correctly, in my opinion – assign 18

Hubeisheng wenwu kaogu yanjiusuo 湖北省文物考古研究所 et al., Yangxin Dalupu 陽新大路鋪, 2 vols. (Beijing, Wenwu chubanshe, 2013).

39

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

them to an archaeological culture of their own. The ceramic tradition and basic cultural inventory of this so-called “Dalupu Culture” are broadly similar to those of contemporaneous local cultures in adjacent areas south of the Yangzi, such as Tanheli in Hunan, Wucheng in Jiangxi, and others. They are fundamentally distinct from the Shang and Zhou mainstream; in particular, they differ from the ceramic traditions associated with nearby Zhou-affiliated polities, such as Zeng and Chu on the opposite bank of the Yangzi. Very probably, therefore, the participants in the Dalupu culture saw themselves as distinct from the inhabitants of the Zhou realm. They are sometimes identified as members of the Yangyue 揚越, an obscure tribal group mentioned in later texts. Dalupu was abandoned sometime around the middle of the first millennium B C E, at about the same time as large-scale copper-smelting operations began to be set up right next to the Tonglüshan mines. At this juncture, miners and metallurgists began to dwell permanently near their places of work. At Tonglüshan, at least fifteen settlements surround the central mining area within a perimeter of about 100 to 500 meters. These sites were no longer agricultural in character: they were specialized industrial villages. All were located on hillsides, presumably in order to maximize proximity to the mines, as well as to the mountain forests, which were indispensable as a source of the timber needed to secure the mines and the charcoal needed for smelting. One of these sites is Yanyinshanjiao, Daye (Hubei). Here the Spring and Autumn-period (possibly early Warring States-period) occupation left behind an unfinished mineshaft, a layer of leach tailings, a beneficiation field, a clay pool, and a furnace – in other words, an intact operational sequence of copper smelting. Moreover, the excavators found the remains of one semisubterranean house, identified as a workers’ shed, as well as thirty-five footprints, presumably of miners. Large amounts of copper ore, native copper, smelting slag, furnace debris, splints, and bamboo objects were found, but only a small amount of ceramics and no production tools related to agriculture or household industries. All this indicates that Yanyinshanjiao must have been home to a specialized community of metallurgists. Since no casting molds have been found, the excavators believe that the copper smelted at this site was cast into ingots of standardized size, which were then shipped to faraway foundries, such as Chenjiatai at the Chu capital of Ying at Jinancheng, Jiangling (Hubei).19 Some copper ingots that may date to 19

See Hubeisheng bowuguan 湖北省博物館, “Chu du Jinancheng de kantan yu fajue” 楚都紀南城的勘探與發掘, Kaogu xuebao 1982.3: 325–50; 1982.4: 477–508. Only a small portion of this important site has been excavated so far.

40

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

the Bronze Age, weighing between 1.5 and two kilograms, have been found nearby on the shores of Lake Daye. Aside from villages, archaeologists have found several walled settlements associated with the mining industry. The best-reported to date is Wulijie, Daye (Hubei).20 Wulijie is located about ten kilometers downhill to the southeast from Tonglüshan, along the transportation route linking the Tonglüshan mining area to the Yangzi river, and even closer to two other ancient mines. It is a rectangular moated enclosure with stamped-earth walls measuring 405 by 308 meters (12.5 hectares), aligned with the natural features of the surrounding terrain. Since the walls are incompletely preserved, the total number of gates is uncertain, but the remains of two gates were excavated. Survey and selective excavations revealed traces of the road system inside the settlement, seven raised building foundations and the foundation trenches of two others, one well, twenty-three refuse pits, five refuse-filled ditches, and one former sewage ditch. There were also some remnants of walls that may have enclosed compounds inside the settlement. Of particular interest is a canal, five meters wide and up to two meters deep, piercing the city wall near the southwest corner; it was traced over a length of sixty meters. Linked to the moat – and through it, to the region’s riverine transportation system – it presumably served for the transportation of materials or goods in and out of the city. Wulijie was occupied from approximately the Shang–Zhou transition down to the end of the Spring and Autumn period; its abandonment seems to have coincided with that of Dalupu and similar early metallurgical communities in the vicinity. Approximately 95 percent of the excavated assemblage consisted of Dalupu culture ceramics. Metallurgy seems to have been practiced virtually all over the site. It is likely that Wulijie served as the central node in the exchange networks through which the copper produced at nearby village settlements was supplied to the bronze-manufacturing centers in the Zhou culture sphere. Although it is difficult to quantify the scale of production, it seems to have been considerable by Zhou times. The eighth-century B C E Chu Gong Ni yongzhong inscription – one of the earliest known Chu inscriptions – attests that enormous quantities of copper were brought in, presumably from the mining districts to the south of the Yangzi: 90,000 jun 鈞, amounting to either 900 or 621 metric tons of copper (depending on one’s understanding of the conversion rate) on but a single occasion! 20

Hubeisheng Wenwu kaogu yanjiusuo 湖北省文物考古研究所, Daye Wulijie: Chunqiu chengzhi yu zhouwei yizhi kaogu baogao 大冶五里界: 春秋城址與周圍遺址考古報告 (Beijing, Kexue chubanshe, 2006).

41

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

Surveys discovered twenty-one smaller sites within less than six kilometers of Wulijie, forming part of the settlement pattern dominated by that city. They fall into two size groups: between six and nine and between two and four hectares; some of them were mostly residential, but most yielded some metallurgy-related remains. Two sites were purely late Neolithic, seven ranged from late Neolithic to Zhou (similar to Dalupu), and another six were purely Zhou in date. The remaining six sites – difficult to date due to a lack of ceramics, but probably from the later part of the Zhou period – were accumulations of smelting slag with few or no indications of other daily-life activities; all these were smaller than 1.5 hectares. After the abandonment of Wulijie, two other metallurgy-related walled cities arose in the same area: Ewangcheng, Daye (Hubei), in the hilly country thirty kilometers to the west and Caowangzui, Daye (Hubei), a mere two kilometers to the north of Tonglüshan. Both are less than half the size of Wulijie: Ewangcheng measures 5.4 hectares and Caowangzui 5.5 hectares, and the thin cultural deposits at both sites indicate a relatively brief period of occupation. Evidencing defense considerations, each had thick walls and, unlike Wulijie, only a single entrance gate. Like Wulijie, both were linked to nearby riverine transportation routes; both were directly connected to nearby mining resources; both yielded significant amounts of metallurgyrelated remains; and both were surrounded by a network of smaller, more or less contemporaneous, metallurgical settlements. Even though Ewangcheng dates to the Warring States and Caowangzui mostly to the Western Han period, they represent the same type of settlement as Wulijie: they are fortified early industrial towns that served as a locus of metal production, administration, storage, transportation, and trade. It is interesting that – assuming the reported dates are correct – such sites existed continuously before and after the mid-first-millennium transformation of the mining industry. The ceramic assemblage at Ewangcheng shows few of the characteristics of the earlier Dalupu culture ceramics; instead it corresponds fairly closely to what one sees at contemporary Chu sites. Chu-style roof tiles, absent at Wulijie, are seen here as well. Besides, Ewangcheng has yielded a number of weapons (one of which is inscribed with the name of its owner, Chen Wang), catering to defense needs and intimating that by this period this part of present-day Hubei had become encompassed within the Chu kingdom and thus within the Zhou realm. This situation is confirmed by the discovery of a piece of Chu gold currency, which likewise bears an inscription. 42

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

Also of interest to the understanding of the transformation of the Tonglüshan “mining landscape” during the latter part of the Eastern Zhou period is the Sifangtang cemetery, in Daye (Hubei), which was located immediately to the north of Mine V I I in the northern part of the Tonglüshan site. A total of forty-eight tombs were excavated, of which fortytwo have been dated to the Spring and Autumn period, though I suspect that they actually date from the early to middle Warring States. Their shapes and contents are typical for Chu tombs. Tombs 1 and 34 are much larger and deeper than the others. Tomb 34 is by far the largest, but it seems never to have been occupied and contained neither tomb furniture nor funerary goods. The occupant of Tomb 1 had been put to rest in nested coffins – a common status symbol in the Chinese Bronze Age that is frequently encountered at Chu cemeteries but nowhere else at Sifangtang. The other tombs were all approximately equal in size, ranging from two to three square meters, and contained either single coffins or no coffins at all. Interestingly, the quantity of grave goods in both the large and the ordinary tombs showed no obvious differences in wealth, suggesting that socioeconomic inequality within the Sifangtang “burying community” was not pronounced, even though certain individuals may have commanded exceptional prestige. It stands to reason that Sifangtang served the community that worked in Mine V I I. From the discovery of bronze weapons in thirteen tombs at Sifangtang, one may possibly infer that some of the individuals interred were employed either as guards or as members of a military patrol unit that protected the Tonglüshan mine. The monumental exterior walls of associated walled settlements also hint at a concern with security and defense in connection with the mining industry. The impact of the mid-first-millennium transformation of the Tonglüshan “mining landscape” cannot be overemphasized. From a cottage industry undertaken by an autonomous, mainly agricultural population, copper extraction became a large-scale and full-time specialized metallurgical industry, which may have operated under state control. It is likely that this transformation more or less coincides with the subsumption of the area under the administrative control of the Chu kingdom, which probably had been the major exchange partner of the metallurgists of the Dalupu culture since at least the early centuries of the first millennium B C E. The material culture of Warring States-period sites in the area is recognizably Chu. The exact time when Chu established its control over the mining district south of the Yangzi is as yet unclear, partly due to the unfortunate fact that the archaeological chronologies currently in use remain somewhat inexact. It 43

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

seems likely that the indigenous mining communities of the Dalupu culture continued their activities under the Chu regime, possibly alongside other personnel who came in from the outside, but their separate identity is no longer manifest in the archaeological record. Rather than being the result of a Chu administrative act, the mid-first-millennium B C E transformation of the mining industry seems to a considerable extent to have been propelled by the inner dynamics of technological developments, which in turn occurred in response to increasing external demand for copper. Even the Chu takeover of the mines may have been perceived, in the political-economic environment of the late Spring and Autumn to early Warring States period, as the logical consequence of the centuries-long exchange relationships between Chu and the people of the Dalupu culture. From an economic perspective, even if the mining operations were not directly run by the government, their subsumption under Chu administrative authority was no doubt part and parcel of the ever-tighter integration of the Chu economic area that we may also perceive through the Chu currency system and the E Jun Qi tallies (see above).

Zhongba The salt-production site of Zhongba is now submerged due to the building of the Three Gorges Dam.21 Located for the most part on a large mound in the center of the valley of the Ganjing river, a minor northern tributary of the Yangzi, Zhongba was a multi-period, deeply stratified site with deposits datable from the second millennium B C E through the second millennium C E. The nearby small town of Ganjingzhen was until the mid-twentieth century a center of salt production, exploiting natural brine springs that bubble to the surface in the riverbed. Written sources mention salt making in this area as far back as the third century B C E. For most of the last 2,000 years, the technology of boiling and cleaning the brine – revolving mostly around the use of cast-iron pans placed on oven furnaces – has remained more or less unchanged. The excavation of Zhongba and other sites in the Ganjing river valley for the first time provided evidence of salt making in this area that goes back many centuries before the periods for which there is written documentation. Rather than in iron pans, the early salt makers in this area evaporated the brine in ceramic vessels. Similar pottery-based salt-manufacturing processes 21

The final archaeological report on Zhongba remains unpublished; for preliminary studies, see Li Shuicheng 李水城 and Lothar von Falkenhausen 羅泰 (eds.), Zhongguo yanye kaogu 中國鹽業考古/Salt Archaeology in China, 3 vols. (Beijing, Kexue chubanshe, 2006–13).

44

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

are known from many other parts of the premodern world. What sets the Ganjing river valley sites apart is the astoundingly large scale of salt production. The excavators had to dig their way through the remains of several tens of millions of brine-boiling vessels, all of virtually identical shape, that were densely accumulated to a thickness of several meters with very little else in between. For centuries, the Zhongba salt makers apparently did their work while surrounded on all sides by these innumerable pottery fragments. It seems that the brine boiling went on in makeshift sheds with mud floors, which were moved as heaps of discarded brine-boiling vessel sherds built up around them. The site formed through alternating cycles of leveling the accumulated potsherds, putting down new activity floors, building new workshop sheds and brine-boiling ovens, and dumping disused potsherds around the sheds until the latter became inconvenient to use. Ethnographic evidence suggests that this cycle of leveling and rebuilding may have been tied to seasonal cycles of exploitation. Zhongba witnessed three phases of industrial-style salt manufacturing, datable to the eighteenth to thirteenth (Phase I), twelfth (Phase I I), and ninth to third (Phase I I I ) centuries B C E respectively. Each phase is characterized by its own distinctive type of salt-making pottery, which must have been mass-produced at an as yet unidentified location nearby. During the hiatus between Phases I I and I I I at Zhongba, from the twelfth to ninth centuries B C E, brine boiling apparently occurred at the site of Wazhadi near the confluence of the Ganjing river with the Yangzi, some six kilometers downstream from Zhongba; brine may have been brought to Wazhadi by boat or (in analogy with later practices) by a bamboo pipeline along the Ganjing river. Chemical analysis has demonstrated that the accumulated pottery vessels had indeed been used for salt making. One question of interest to the excavators was whether salt was all that was produced; the discovery of large amounts of animal and fish bones amidst the pottery sherds seems to indicate the production of some salt-based secondary products such as cured meat and fish sauce. It seems that we are dealing with the remains of a sophisticated, large-scale industry involving significant division of labor. Yet even more than the copper mines south of the Yangzi, this industry was located in an area that common prejudice would unquestionably qualify as remote and “peripheral.” The early inhabitants of the Ganjing river valley were linguistically and ethnically distinct from their Shang and Zhou neighbors. Strictly speaking, in fact, the Bronze Age of the Sichuan basin should be regarded as prehistoric down to 316 B C E, when the Qin kingdom conquered 45

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

the area. Whereas the Chengdu plain – ancient Shu – attained state-level sociopolitical complexity as early as the mid-second millennium B C E, the region along the Yangzi in present-day Chongqing municipality seems to have hosted a congeries of unconsolidated tribal entities down to at least the time of the Qin conquest. Some of these so-called Ba 巴 tribes – e.g., those represented by the “royal” tombs at Xiaotianxi, Fuling (Chongqing) (fifth to third centuries B C E) – may have been richer and more powerful than others, but there does not appear to have been a unified Ba territory, and the area does not seem to have been affected by any of the tendencies of administrative, military, and fiscal concentration that characterize contemporaneous Eastern Zhou polities to the north and east. An increase in material prosperity all over the upper Yangzi basin seems apparent from the rise of a local bronze-manufacturing industry during the sixth to third centuries B C E, which produced stylistically distinctive ornamental bronzes. Even though their neighbors in the Shang and Zhou culture sphere would have undoubtedly regarded them as alien, and perhaps as uncouth “barbarians,” the people of Ba were obviously well settled in their native environment of the upper Yangzi basin. What is more, they were able to run complex economic operations such as the salt workshops in the Ganjing river valley. It seems virtually certain that the production of these salt workshops far exceeded local demand; it must have been tied into long-distance trading networks. Even before archaeological evidence became available, the historian Ren Naiqiang ascribed the rise of the Ba bronze industry to the riches obtained from the salt and fish trade.22 Ren suggested – plausibly, I believe – that the market for the salt of Ba was Chu in the middle Yangzi region. In his reconstruction, the salt merchants of Ba would take their wares to the Chu capital to sell, returning with high-quality manufactured goods and cultural know-how. Geographically speaking, it is difficult to imagine that the salt of Ba could have been transported to any area other than Chu, which previous to its conquest of the southeastern coastal areas of China in 334 B C E had no salt resources of its own, and which was Ba’s closest neighbor among the states of the Zhou culture sphere; moreover, while the Yangzi river, although difficult to navigate, provided a direct avenue of transport with Chu, Ba was separated from all other parts of the Zhou realm by forbidding mountain ranges. In partial corroboration of Ren’s scenario, archaeology has confirmed the presence of Chu-manufactured luxury objects in Ba tombs of the sixth to 22

Ren Naiqiang 任乃強, “Shuo yan” 說鹽, Yanyeshi yanjiu 盐业史研究 1988.1, 3–12.

46

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

third centuries B C E, and of objects in the distinctive Ba style in tombs near the Chu capital. For instance, the fifth- to fourth-century B C E cemetery of Yajiao, Zhongxian (Chongqing), in the immediate vicinity of the salt-manufacturing sites at the mouth of the Ganjing river, alongside the simple single-coffin “Batype tombs” typical of the region, contained tombs with manifest Chu characteristics such as nested coffins and Chu-style ritual ceramics. The excavators have interpreted the latter tombs as indicating the presence of Chu individuals in this region – presumably a “trade diaspora” of salt merchants operating outside Chu territory. If so, the Ba–Chu salt trade was perhaps not entirely in the hands of Ba merchants, as Ren Naiqiang hypothesized. But it is also conceivable that Ba merchants traveled to Chu and brought knowledge of Chu burial customs with them to their homeland. In any case, given the distance of some 500 kilometers separating Zhongba from the western borders of Chu, the suggestion of a Chu military takeover of the region sometime during the Warring States period, though repeatedly fielded by scholars, seems implausible. In this respect, the case of Zhongba likely differed from that of Tonglüshan. We still do not know many crucial details about the salt trade between Ba and Chu. For instance, the price of salt is unknown and probably beyond reconstruction. But we know that the salt workshops in the Ganjing valley produced without significant interruptions throughout the Zhou period, attesting that a stable trading relationship was maintained over a long time. The chronological sequence of salt-making pottery vessels at Zhongba continues without a break after the Sichuan basin was conquered by Qin in 316 B C E, and even after Qin had displaced Chu from its traditional political centers along the middle Yangzi in 278 B C E. Since Qin had its own salt resources closer to its political centers, there was probably no reason after the Qin conquest to divert the salt from the Ganjing river valley away from its traditional Chu markets even after the area had come under the centralized Qin administration. The replacement of the traditional pottery-based salt-making technique with iron brine-boiling pans distributed through a government monopoly does not seem to have occurred until sometime in the early imperial era. Before that time, from the fact that the supply routes were maintained regardless of who was in charge, one may perhaps infer that government involvement with salt production and trade was limited, if not absent. At the same time, the fact that salt production continued even after Chu gained access to the sea and thus to sea salt, which was far cheaper than the salt from inland wells, reminds us that purely economic considerations may not suffice 47

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

to understand the nature of the Ba–Chu salt trade. Was the exchange perhaps perceived as a symbolic expression of the sociopolitical relationship between the two regions? Might it have been sanctioned by religious belief? Or did the Chu salt traders merely wish to spread the risk, in case the salt supply from the sea were interrupted for any reason?

Discussion All over China, archaeological finds document similarly long-enduring exchange relationships involving a variety of resources and products. Simply put, the early polities within the Zhou realm (like, very probably, their predecessors during the Shang period) relied on the control of their agricultural core, using whatever surplus they could extract from it to acquire natural resources and local products from their non-Zhou neighbors. In this sense, their techniques of governance – the rudimentary forms of administration and taxation practiced within the framework of a lineage-based society – were ecologically specialized. They made few efforts to exert direct control over the surrounding “mountains and forests, lakes and marshes.” The long-lasting interregional relationships that developed in this way involved some reflux of finished products into the areas that had supplied the raw materials for them. In some cases, such as that of bronze vessels, this eventually triggered the emergence of local industries. To some extent, these economic processes are comparable to those between the First and Third Worlds in today’s global economy. This system broke down after the middle of the first millennium B C E, when the various Warring States kingdoms expanded into areas formerly held by independent tribal populations, absorbing their homelands under an increasingly centralized fiscal administration. The takeover of the Tonglüshan “mining landscape” by Chu is the archaeologically bestdocumented example of this process, which foreshadowed the expansion of urbanism, the emergence of a commercial economy that was at least partly monetized, and the rationalization of manufacturing procedures traced above. All these phenomena played a role in bringing about the rise in general economic prosperity characterizing the “Warring States economic miracle.”

Coda: Transcontinental Contacts One often overlooked ingredient in the economic growth of the Chinese core area during the first millennium B C E is its increasing integration into 48

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China

Eurasia-wide economic networks, long preceding the opening of the “Silk Routes” under the Han dynasty at the end of the second century B C E. The preferred trade route into the western regions ran farther to the north than the later caravan routes; rather than going through the Taklamakan Desert and the Pamirs, it connected central China with Mongolia, northern Xinjiang, the region to the north of the Sayan and Altai mountain ranges, central and western Siberia, and the Black Sea coast. Thanks to many new discoveries in the western regions of China and the better accessibility of finds from formerly Soviet Central Asia, there is now an ever-growing catalogue of “Chinese” cultural features with possible western connections, including even such quintessential Confucian scholars’ props as the qin zither.23 Their number notably increased during the first millennium B C E. On some of these phenomena there is now good archaeological information; on most we are still reduced to speculation. Various models have been proposed to account for how the alleged imports might have been transmitted across Eurasia; tributary exchange and merchant trade are only two possibilities from different extremes of the political-economic spectrum. It seems likely that several mechanisms operated simultaneously. A detailed treatment would go beyond the parameters of the present discussion, but the possible role of long-distance trade as an economic factor should at least be mentioned. At the end of the pre-imperial period, one notes a growing dialectic between, on the one hand, the growing prevalence of east–west contacts, and, on the other hand, the increasingly tightly integrated economy of the Chinese cultural core. The unification of China under the Qin and Han imperial regimes brought a comprehensive assertion of state control over the economy and a turning away from the freewheeling market-driven economic activity that had enabled a relatively high degree of overall prosperity under the “Warring States economic miracle.” Transcontinental trade, as well, arguably became more regimented. While many of the features of economic life that had evolved during the first millennium B C E eventually became an integral part of the fabric of imperial Chinese civilization, the Warring States period stands out as a period of exceptional technological innovation and, by extension, economic prosperity. Moreover, the preceding discussion has highlighted the importance of the ritual reforms c. 850 B C E and during the sixth century 23

Bo Lawergren, “Strings,” in Jenny So (ed.), Music in the Age of Confucius (Washington, DC, Freer Gallery of Art and Arthur M. Sackler Gallery, Smithsonian Institution, 2000), pp. 65–83.

49

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

lothar von falkenhausen

as catalysts of crucial opening processes – as decisive steps toward overcoming the social and religious basis of the earlier “gift-giving economy.” One assumes – though it is currently still difficult to prove conclusively – that the ensuing prosperity reflected a more general rise in living standards. Thanks to archaeological finds, we can now begin to reexperience the “Warring States economic miracle” from the ground up, and to understand ever more clearly the centuries-long processes through which it came to pass.

BCE

Further Reading Amano Motonosuke 天野元之助, Chu¯goku shakai keizaishi: In, Shu¯ no bu 中国社会経済 史：殷、周之部 (Tokyo, Kaimei shoin, 1979). Bagley, Robert W., “Replication Techniques in Eastern Zhou Bronze Casting,” in Stephen Lubar and W. David Kingery (eds.), History from Things: Essays on Material Culture (Washington, Smithsonian Institution Press, 1993), pp. 231–41. Chen Zhenyu 陳振裕, Zhanguo Qin Han qiqiqun yanjiu 戰國秦漢漆器群研究 (Beijing, wenwu chubanshe, 2007). Emura Haruki 江村治樹, Sengoku Shin Kan jidai no toshi to kokka: Ko¯kogaku to bunken shigaku kara no apuro¯chi 戦国秦漢時代の都市と国家：考古学と文献史学から のアプローチ (Tokyo, Hakuteisha, 2005). Falkenhausen, Lothar von, Chinese Society in the Age of Confucius (1000–250 B C): The Archaeological Evidence (Los Angeles, Cotsen Institute of Archaeology, UCLA, 2006). Falkenhausen, Lothar von, “The E Jun Qi Metal Tallies: Inscribed Texts and Ritual Contexts,” in Martin Kern (ed.), Text and Ritual in Early China (Seattle, University of Washington Press, 2005), pp. 79–123. Falkenhausen, Lothar von, “Stages in the Development of ‘Cities’ in Pre-imperial China,” in Joyce Marcus and Jeremy Sabloff (eds.), The Ancient City: New Perspectives on Urbanism in the Old and New World (Santa Fe, School of Advanced Studies Press, 2008), pp. 209–28. Hong Shi 洪石, Zhanguo Qin Han qiqi yanjiu 戰國秦漢漆器研究 (Beijing, Wenwu chubanshe, 2006). Jia Bingqiang 賈兵強, Chu guo nongye keji yu shehui fazhan yanjiu 楚國農業科技與社會發 展研究 (Beijing, Kexue chubanshe, 2012). Ledderose, Lothar, Ten Thousand Things: Module and Mass Production in Chinese Art (Princeton, Princeton University Press, 2001). Li Jiannong 李劍農, Xian Qin liang Han jingji shigao 先秦兩漢經濟史稿 (Taipei, Huashi chubanshe, 1981). Li Ling 李零, Zhongguo fangshu xukao 中國方術续考 (Beijing, Dongfang chubanshe, 2000). Li Ling 李零, Zhongguo fangshu zhengkao 中國方術正考 (Beijing, Dongfang chubanshe, 2006). Li Shuicheng 李水城 and Lothar von Falkenhausen 羅泰 (eds.), Zhongguo yanye kaogu 中國鹽業考古/Salt Archaeology in China, 3 vols. (Beijing, Kexue chubanshe, 2006, 2010, 2013)

50

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

The Economy of Late Pre-imperial China Liu, Li, and Xingchan Chen, State Formation in Early China (London, Duckworth, 2003). Liu Yutang 劉玉堂, Chu guo jingjishi 楚國經濟史 (Wuhan, Hubei Jiaoyu chubanshe, 1996). Rawson, Jessica, Chinese Jade from the Neolithic to the Qing (London, British Museum Press, 1995). Sato¯ Taketoshi 佐藤武敏, Chu¯goku kodai ko¯gyo¯shi no kenkyu¯ 中國古代工業史の研究 (Tokyo, Yoshikawa ko¯bunkan, 1962). Sun, Zhouyong, Craft Production in the Western Zhou Dynasty: A Case Study of a Jue-Earrings Workshop at the Predynastic Capital Site, Zhouyuan, China (Oxford, Archaeopress, 2008). Wagner, David B., Iron and Steel in Ancient China (Leiden, Brill, 1993). Wang Di 王廸, Zhongguo beifang diqu Shang Zhou shiqi zhitao zuofang yanjiu 中国北方地区 商周时期制陶作坊研究 (Beijing, Kexue chubanshe, 2016). Wang Yifeng 王屹峰, Zhongguo nanfang yuanshici yaoye yanjiu 中國南方原始瓷窯業研 究 (Beijing, Zhongguo shudian, 2010). Xu Hong 許宏, Xian Qin chengyi kaogu 先秦城邑考古, 2 vols. (Beijing, Jincheng chubanshe, 2017). Yang Shengnan 楊升南 and Ma Jifan 馬季凡, Shang dai jingji yu keji 商代經濟與科技, in Song Zhenhao 宋镇豪 (ed.), Shangdai shi 商代史, vol. 6 (Beijing, Zhongguo shehui kexue chubanshe, 2010). Zhang Hongyan 張鴻雁, Chunqiu Zhanguo chengshi jingji fazhan shilun 春秋戰國城市經 濟發展史論 (Shenyang, Liaoning daxue chubanshe, 1988). Zhongguo shehuikexueyuan kaogu yanjiusuo 中國社會科學院考古研究所, Zhongguo kaoguxue: Liang Zhou juan 中國考古學：兩周卷 (Beijing, Zhongguo shehuikexue chubanshe, 2004).

51

https://doi.org/10.1017/9781108587334.002 Published online by Cambridge University Press

2

Agriculture and Its Environmental Impact motoko hara translated by richard von glahn

China is one of the seven regions in the world where agriculture developed independently, and among the earliest, with cultivation of cereal crops dating back to at least 7000 B C E. The unique repertoire of crops first cultivated in China – millet, rice, and soybeans – remained staple foods throughout Chinese history, although millet was mostly displaced by wheat, an import from West Asia, by 1000 C E. The long history of the development of agriculture and the rise of cereal cultivation as the mainstay of human livelihood and economic activity dramatically altered the natural landscape, its landforms, soils, and waters as well as its flora and fauna. Domestication of the environment to serve human needs for food, clothing, shelter, and fuel repeatedly realigned the balance between human populations and natural resources. The strain of burgeoning human populations and their demands on these resources necessitated continual technological innovation to sustain agricultural production and conserve natural resources. By the year 1000 the human impact had utterly transformed the ecology of northern China, especially in the watershed of the Yellow River. At that point in time the impact of human settlement and Chinese agricultural practices was far more muted in south China apart from a few regions such as the Chengdu plain and the Yangzi river delta.

Changes in the Natural and Human Environment of Early China The Chinese mainland rests on the Cambrian stratum, which is considered to be the oldest surface layer on Earth, but the various regions that comprise China have been influenced by repeated climatic changes. The formation of

52

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact MONGOLIA

.

aR

ghu

Son

Tian Mts.

.

Tarim Basin

Lu

Ye l

R.

a

R .

Central Plain

w lo Ye l

Wei R.

Jia

lin

g

Hu

ai

R.

Yellow Sea

Ha

R.

Qinling Mts.

n

Yangzi Delta

R. Gan

gR .

R.

Yu an

Xian

Wu

R.

Yangzi. R.

R.

Sichuan Basin

Sea of Japan

A

Guanzhong

ts

M

b ng

RE KO

an

lh Ta

ai

Ch

R

.

M

g

Ordos

Ye

llo w

an

ts.

Tibetan Plateau

BU

East China Sea

Nanling Mts. rl R .

Pea

A

RM

T VIE OS

LA

South China Sea

Over 1800 m 900–1800 m 180–900 m Under 100 m

M

ND

NA

LA

AI

TH

Bay of Bengal

Lia oR . u R .

Gobi

Map 2.1 Geomorphology of China Source: redrawn based on Li Liu and Xingcan Chen, The Archaeology of Ancient China (Cambridge, Cambridge University Press 2012), Figure 2.3, p. 27

the Himalayan mountain range rendered Central Asia and areas adjacent to China increasingly arid, turning grasslands into desert. According to Aiming Lin, roughly 5 million years ago, under the force of northward movement of the Indo-Australian plate, the bend of the Yellow River was formed, creating the loess plateau in the Ordos region.1 The original channel of the Yellow River became today’s Wei river valley, the heart of the Guanzhong region, which would become a main stage for major historical events of ancient China. In addition, large rivers such as the Yellow River, the Yangzi, and the Pearl carried copious quantities of sediment, and the deposition of these sediments in their estuaries resulted in the eastward shift of the coastline, which had a major impact on the agricultural environment. 1

Aiming Lin et al., “How and When Did the Yellow River Develop Its Square Bend?” Geology 29.10 (2001), 951–4.

53

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

The principal soil layer of north China is known as “loess” (or simply “yellow earth” (huangtu 黃土 in Chinese). Since the research of Baron von Richthofen – best known for coining the term “Silk Road” – in the nineteenth century, this loess soil has been considered highly fertile, and even today explanations for the development of Chinese agriculture rely on this assertion. But such a view is mistaken. Richthofen himself observed that salt deposits could be observed on surface soil in the Guanzhong region where cultivation had continued for many years, and the salinity of groundwater increased as the atmosphere became more dry. He speculated that this soil has a special capacity for what might be called “self-fertilization,” and that the fertilizer, like salt, is also supplied from underground.2 However, Richthofen to the end regarded this hypothesis as “speculation,” and he clearly expressed that scientific study was necessary to confirm it. But his observation was commonly regarded as “fact,” giving birth to the myth of “China’s fertile soil.” This misunderstanding also led to other misconceptions, such as the idea that trees cannot grow on loess soil. As a matter of fact, the Yellow River basin was covered with forests around 1000 B C E ; that is, until the end of the Shang dynasty. Soil scientists in China today recognize that loessic soil, now known as “yellow cotton soil” (huangmiantu 黄綿土), does not contain any special nutrients. But there are soils in the Wei river basin that as a consequence of having been treated with fertilizer over many years no longer require the application of fertilizer in each cultivation season. These soils are known as cumulic soils (loutu 塿土). At the time of the Shang dynasty (c. 1550–1045 B C E), before the appearance of cumulic soils resulting from human cultivation, the climate was mild and rainy, and there was also a dense forest spreading over the yellow cotton soil of the Yellow River basin. Archaeological investigation has revealed that paddy rice was also cultivated in the Central Plain, especially around present-day Luoyang and Zhengzhou, in the Neolithic age. These crops were cultivated utilizing nitrogen-rich deposits of humus (decayed organic matter). Conventionally, it has been thought that ancient cultivated lands were river terraces irrigated by spring water. However, in recent years it has become evident that in the Neolithic age rice cultivation was practiced in the Central Plain by taking advantage of natural topography such as bends in small rivers. 2

F.F. von Richthofen, China: Ergebnisse eigener Reisen und darauf gegründeter Studien (Berlin, D. Reiner, 1877).

54

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

Table 2.1 Climatic shifts in China, 3000 B C E to 1000 C E c. 3000–c. 1100 B C E c. 1000–c. 850 B C E c. 700 B C E –c. 1 C E c. 1–c. 600 C E c. 600–c. 1000 C E

Warm Cold Warm Cold Warm

Source: Zhu Kezhen 竺可桢, “Zhongguo jin wuqiannian lai qihou bianqiande chubu yanjiu” 中国近五千年来气候变迁的初 步研究, Kaogu xuebao 考古学报 1972.1, 5–38

By the close of the Shang dynasty, a cooling trend had set in (Table 2.1). In addition, game hunting declined, as revealed by records of hunting carried out as part of the ritual practices of the Shang kings. Naturally, it is thought that yields of agricultural crops, mainly cereals, also decreased. These changes seem to have destabilized the ruling system of the Shang dynasty. The rich yields of the hunter-gatherer economy and the harvests of agricultural produce were reflections of the religious power of the Shang kings, on which their control over subordinate clans depended. As this aura of authority weakened, the Zhou clan, which had begun to emphasize the primacy of cereal production, formed an alliance with other clans and overthrew the Shang kings c. 1045 B C E. The Zhou royal lineage expanded cultivation beyond river terraces to flat, low-lying lands along the riverbanks of the Wei river basin. Furthermore, they made efforts to reclaim wetlands in the Wei river basin and develop agricultural production that could sustain a stable food supply. Pasturing livestock was also practiced, but the vast grasslands where large animals could be raised gradually diminished in areas inhabited by cereal cultivators. Since the Zhou people still relied mainly on stone tools, clearing forests was difficult. Therefore, most agricultural cultivation was practiced on flat ground that had originally been grassland. It was necessary for the Zhou kings to secure the land and labor for raising horses, cattle, sheep, and other livestock to supply the sacrificial rituals on which their power depended. Moreover, in the daily life of commoners, hunting was an indispensable source of animal protein intake. Forests that sustained wild animals such as deer and boar were still abundant at that time. This presumption is supported 55

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

by textual records such as the “Seventh Month” song preserved in the Book of Odes, believed to reflect to some extent the circumstances of the Western Zhou period (1045–771 B C E), which depicts villagers collectively going out to hunt during the slack agricultural season in winter. The spread of iron tools beginning in the Spring and Autumn period (771– 453 B C E) utterly changed this pattern. Iron axes greatly facilitated the cutting of forests. The development of iron metallurgy, especially for the production of weapons, also accelerated the demand for timber as fuel. As iron gradually began to be utilized for the manufacture of agricultural tools, logging further intensified, and with the availability of iron tools the area under crop cultivation enlarged even more. The loss of forests pitted the Zhou royal house and the various lords who owed fealty to the Zhou kings, for whom agriculture was primary, against the commoners whose subsistence depended in part on hunting and gathering in the forests. The idea of converting the entire landscape to agricultural production, which originated with the Qin kingdom in the Warring States period (453–221 B C E), gradually spread to other states as well. When a climatic warming trend resumed at the end of the Spring and Autumn period, yields from cereal production increased. The importance of grain for military provisions in states enmeshed in continuous warfare also rose, and each state competed to increase the land available for cereal cultivation. Eventually, forests disappeared from most of the Central Plain basin and were replaced by cultivated fields, which gradually began to spread over hill slopes as well. Then, depending on the topography, soil alkalization (see below) occurred, which required irrigation to make the land suitable for farming. Still, forests capable of sustaining hunting could be found in some regions, even in the Wei river basin, until the end of the Han dynasty (202 B C E – 220 C E). In the Han, cultivation continued to expand, while the cooling trend resumed. Farmers tended to select varieties of grain that were droughtresistant, such as foxtail and broomcorn millets. Cultivation of millets was supported by the construction of small- and medium-scale irrigation works using ponds and reservoirs, especially in the Huai river basin. In many cases these projects were developed by local officials and landowners who managed their own estates. The Period of Disunion (third to sixth centuries C E), which was marked by one of the coolest climates in the historical era, witnessed the development of farming techniques to cope with the colder climate. The cooling trend had an especially significant impact on the northern grassland zone, where the pastoral livelihood was readily affected by changes in the natural 56

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

environment. In order to preserve their livestock and families, pastoralists were induced to move southward to warmer regions. Migration to the Yellow River basin, whether by peaceful or violent means, increased. Consequently, rulers in the Yellow River basin often proposed policies to encourage cereal production even by peoples who were not habituated to farming. Thus farming practices improved. When nomad leaders took power, some areas which had been converted to cereal cultivation were returned to pasture for raising livestock. Soil nutrients depleted by repeated cereal cultivation recovered with the reversion to livestock grazing. In the Yangzi river basin, where Chinese rule had expanded since the Han period, forests remained largely intact. Cultivation potentially could be expanded in this region by planting rice, which was ill-suited to the arid Yellow River basin. Dry-land cereal cultivation extended over the plains of the Central Plain as far as the eye could see, but cereal production did not advance so easily in the forested Yangzi river basin. People living in thickly forested mountainous areas or areas where lakes, wetlands, and rivers were plentiful – habitats that sustained plentiful fish, fowl, and other wildlife – retained diverse forms of livelihood. A discernible warming trend emerged in the latter half of the Period of Disunion. In the sixth century, population growth shifted to the pastoral regions of the north and west, where the warmer climate restored land productivity. The degree to which the global warming trend induced snowmelt in the Tian Shan and Qilian Shan mountain ranges and increasing wetness in Central Asia remains a subject of debate today. But if we take the policies of the Tang dynasty (618–907) as a guide, it seems undeniable that the peoples inhabiting the Silk Road corridor were motivated to expand the area under cereal cultivation. Cereal cultivation using oasis irrigation emerged in the Tarim Basin and other areas. Around this time the potential for cultivation in the Yangzi river basin rose through the introduction of Champa rice from the Southeast Asian littoral and tea from the Southeast Asian highlands, while the planting of citrus fruits and bamboo revived in the Yellow River basin. Although the warming trend led to greater rainfall, in the Yellow River basin irrigation was used not for paddy rice cultivation but rather for the full development of wheat cultivation. We should also consider the influence of the Xianbei and other peoples who ruled north China during the Period of Disunion and Silk Road merchants from West Asia, such as the Sogdians, who had taken up residence in Chang’an and other Chinese cities, in popularizing wheat consumption in the form of bread. 57

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

The Development and Spread of Agricultural Tools and Cultivation Techniques At present, the oldest remains of cultivated plants so far discovered include rice, barley, wheat, oats, and various species of millet. Remains of domesticated pigs, sheep, goats, cattle, and water buffalo have also been found at various Neolithic sites. The same applies to tools related to agriculture and sericulture, such as spades and spindles. In this section we will address new archaeological finds related to the development of agricultural technology.

Stone and Bone Tools Cereal crops initially were cultivated using stone and bone tools. It is clear from archaeological excavations that stone and bone spades (li 犂) were developed before the appearance of metallurgy. Stone spades and grinding stones have been discovered at the Cishan (Wu’an, Hebei) and Peiligang (Xinzheng, Henan) sites, dated to 8000 B P, and bone spades have been excavated at Hemudu and the Luojiajiao sites in Zhejiang, dated to 7000 B P. Metal Tools Only two bronze tools from a Shang site, Dayangzhou (Xinyu, Jiangxi) have yet been confirmed by archaeological research. Whether these artifacts – from an outpost at the margins of the Shang civilization – were actually used in farming, or instead merely ritual objects with ceremonial purposes, remains in dispute. The use of tools made from meteoric iron is evident from a number of excavated artifacts dated to the Spring and Autumn period. However, iron tools, especially plowshares, begin to appear in archaeological sites across all of China in the Warring States period. Harvesting tools such as sickles, as well as weeding tools, appeared at the same time. However, these early plowshares measure roughly fourteen centimeters in width, and it is estimated that they could till the soil only to a depth of about ten centimeters. The socalled “soil-breaking shares” (zuotiaoli 作條犂) lacked a moldboard. The Origins of Plow Cultivation At present it remains unclear when and where the use of iron plowshares and ox-drawn plows originated. Most scholars have concluded that plow cultivation began c. the fifth century B C E, although some have argued for an earlier date. In recent years, attention also has been drawn to use of human rather than animal power in agriculture through developments in the study of the

58

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

agricultural practices of non-Han peoples. Plows pulled using human labor have been attested for Guizhou, Inner Mongolia, Gansu, Shaanxi, and other provinces. There are numerous reports for the use of traction spades (qiangli 鏹犂) in Shanxi province up until the 1930s. It seems that human-drawn traction spades with a small turning radius were more efficient for cultivating terraced rice fields built on steep slopes.

The Origins and Development of Dry-Land Cultivation Plow cultivation did not supersede the need for human labor in field preparation. The Analects of Confucius make reference to two hermits, Chang Ju and Jie Ni, who led a solitary life in the mountain wilderness and practiced “tilling in pairs” (ougeng 耦耕), in which one person digs a furrow and the other pulverizes soil clods with a maul (you 耰).3 As a consequence of the increasing aridity of the Yellow River basin, the soil overturned by plowing quickly dried out; left alone, it became hardened like sun-dried bricks, hindering crop growth. Therefore the overturned soil was first crushed into small fragments and used to cover the cast seed, thus forming multiple layers with a lightly consolidated surface layer that inhibits water evaporation, which today is accomplished using vinyl sheets. This practice makes it possible to preserve moisture from natural rainfall until the seeds germinate. It was the earliest example in China of adapting agricultural techniques to counteract the drying of the soil. Of course, in later periods various ideas emerged; during the reign of the Han emperor Wu (r. 141–87 B C E), ox-drawn plowing was tried as a means of counteracting soil drying. As will be described later, the “Food and Money Treatise” of the History of the Former Han mentions the “alternate-fields” (daitian 代田) system, in which soil suffering from secondary salinization was restored to cultivation after the construction of large-scale irrigation works. This method of cultivation was similar to the ridge-and-furrow technique described in the “Requirements of the Land” chapter of the Spring and Autumn Annals of Master Lü in the third century B C E.4 But whereas the latter text assumes that seeding as well as weeding was performed by hand, ox-drawn plows are utilized in the daitian system. It was not merely the case that iron tools were more suitable for digging a furrow to a depth of twenty-three centimeters without repeatedly cutting the same groove. It was also necessary to have a team of oxen capable of

3

Analects 18.6.

4

Book 26, Chapter 4, “Rendi pian” 任地篇, in LSCQ, 2, pp. 1730–1.

59

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

pulling the weight of a large moldboard with a wide plowshare. Records of the labor requirements of the daitian method which refer to “paired tilling with two oxen and three men” show that this method satisfied this requirement. The numerous plowshares excavated in Fuping, Lantian, and other places in Shaanxi province generally are too large (thirty centimeters in width) and heavy (7.5–10 kilograms) to be pulled by a single ox. Specimens of a type of seed drill (louli 耬犂) having a smaller share with a width of 9.5–14.2 centimeters, a wide hole at the tip, and an unusual diamond-shaped moldboard have also been excavated. Using this moldboard, the soil can be turned to one side. In other words, by pairing one person guiding the two oxen pulling a large plow with a second person manipulating the seed-drill (“tilling in pairs”), it was possible to practice the ridge-and-furrow method of cultivation. As a result, we can say that dry-land farming relying on natural rainfall was also possible using ox-drawn plow cultivation. However, under the annual rotation system of the ridge-and-furrow method, only one-half of the crop land was under cultivation at a time. In the Northern Wei era (386–534), a moldboard with a round concavity at its center appeared which could remove the turned-over soil to one side. As a result, it was possible to utilize the arable land more fully, and yield per unit of land rose measurably. As will be noted later, the cultivation techniques transmitted in the sixth-century agricultural treatise Essential Techniques for the Common People (Qimin yaoshu 齊民要術) were premised on this. Nor is that all. A method of promoting germination by forming a multiple surface layer by light plowing during the rainy autumn season to preserve soil moisture followed by deep plowing in spring also was developed. Systematic weeding of sown fields, frequently advocated under the rubric “five plowings and five weedings,” can be traced back to the Warring States period. The foundations of the various farming techniques that encouraged appropriate crop rotation according to soil and site conditions presented in the Qimin yaoshu were established a millennium earlier than the Norfolk four-course system of dry-land farming, which appeared in Europe only in the seventeenth century.

Limitations of Irrigated Agriculture Irrigation was indispensable to the creation of powerful states in ancient societies such as Egypt and Mesopotamia, and the hypothesis that the rise of autocratic political power was linked to irrigated agriculture has remained 60

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

popular. However, this hypothesis is inapplicable to the Chinese case. Indeed, although large-scale canal networks had been built repeatedly since ancient times, initially these canals served as drainage channels for the purpose of flood control or routes for transport and transshipment of goods. It is only from Warring States times that we have empirical evidence for the development of water control projects for the purpose of irrigation. In the region of Ye, on the north bank of the Yellow River in the Wei kingdom, water in the bends of the Zhang river tended to stagnate and increased the alkalinity of the soil. Because of the lower productivity of the surrounding land, the tax rate in this region was reduced. It was recorded that water drawn from the Zhang river was flushed to remove the salt content, and the irrigated land was suitable for cultivation of foxtail millet and rice. The Qin state, on the eve of the unification of the empire, constructed the renowned Zheng Guo Canal in the northern part of the Wei river basin using the Jing river as the water source. In Zhou times, this area had been a richly forested marshland known as the Jiaohuo Marshes. However, in the increasingly arid climate, salt accumulated in the soil. As at Ye, the soils here originally were not as parched; rather, given the natural environment and topography of this region, the groundwater level was high, making it a relatively wet area. In north China, a region predominantly covered in very fine soil, the exposure to dry air resulting from climatic change readily causes the groundwater level to rise. At the same time, the salt contained in the groundwater also rises and remains on the surface after the water evaporates. This is the general process of alkalization in north China. Depending on the local environment conditions, this accumulation occurs naturally. The Zheng Guo Canal was a large-scale irrigation project extending roughly 120 kilometers to the Luo river. Grain yields on irrigated arable land in this region witnessed a staggering increase: it was recorded that 100 mu of land (the standard size of a family farm) produced 100 zhong (equivalent to 195 hectoliters) of grain per year.5 Based on such records, it has been argued that state power in ancient China was predicated on large-scale irrigation, and, by extension, became the foundation for theories about the rise of autocratic power. One certain effect of these irrigation policies was that rice – which does not cause alkaline regeneration – could be cultivated on these irrigated lands. Today it is recognized that, as has also been argued for the case of the First 5

SJ, 29, p. 1408.

61

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

Dynasty of Ur in Mesopotamia, dry-land crop cultivation on irrigated lands in arid regions induces alkaline regeneration. The explanation runs as follows. When salt that has accumulated on the surface is inundated by irrigation, the salt naturally flows into drainage ditches or seeps into the groundwater. However, when the air is dry, groundwater rises to the surface. Especially in the case of north China, where the yellow cotton soil is extremely fine and becomes densely packed, capillary-like air gaps easily form. Groundwater rises through these capillary air gaps. If the air is dry, even though the water evaporates, the salt it contained will remain on the surface. The salt then reacts with the air, causing the regeneration of insoluble alkali. Since the regenerated alkali is insoluble, it cannot be removed simply through irrigation, with the result that arable land is gradually reduced to desert. This process was based on complex interactions beyond the knowledge of premodern farmers to resolve. The principal reason for the success of the Zheng Guo Canal was that it allowed for rice cultivation, which did not cause alkali regeneration. In rice farming, water covers the ground surface during the germination and sprouting periods, preventing the rise of groundwater and by extension the possibility of alkali regeneration. The warming trend that made rice cultivation possible was a primary feature of the period from Warring States to Qin. However, in contrast to the general pattern of irrigation, the dredging of the Bai Canal channel proposed in 95 B C E was intended to irrigate dry-land crops such as millet. This can be seen from records of the songs about the benefits the people received upon the completion of this channel: “one picul of Jing river water contains several dou of mud; irrigating and fertilizing, my millet crops flourish.”6 In the Zheng Guo Canal, the fresh water of the Jing river that made rice cultivation possible was transformed into a silt-laden river from the inflow of topsoil carried down from the slopes of the loess plateau. People recognized the fertilizing effects of irrigating their fields with the silt water containing nutrients in the mud derived from forest topsoil. However, this practice also had severe repercussions. In the Guanzhong basin, where the soil became steadily drier as agriculture progressed, the cultivation of millets using irrigation induced alkali regeneration of the soil. This phenomenon is common knowledge in modern agronomy, and remains a problem even today. As China entered the twenty-first century, experiments in the Xinjiang Uygur Autonomous Region achieved some success in preventing alkali regeneration 6

HS, 29, p. 1685.

62

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

by gradually supplying moisture through ceramic pipes perforated with fine holes that are buried under the soil. However, some measures have been taken in various places over the past two millennia. In Manchuria in the 1930s, cultivation was sustained using the ridge-and-furrow method, and the accumulated salinity piled up in the furrows. Applying non-alkaline bean cake during the germination period was intended to reduce the salinity caused by rainfall and weeding activity. A method of harvesting crops without use of irrigation was carried out by taking care to ensure that the roots of the crops were deeply implanted in the soil. This seems to have been successful. As mentioned earlier, the ridge-and-furrow method, which closely resembles this practice, appears in records from the Han dynasty. In any event, in the Yellow River basin, where the soil has been dried out as a result of human activities such as timber cutting, the probability of success with large-scale irrigation – apart from the case of rice cultivation – is considered to be extremely low.

Nonagricultural Production and Techniques Mulberry Cultivation and Silkworm Raising From the evidence of a ceramic spindle, regarded as a mortuary artifact, discovered at the Neolithic site of Hemudu (Yuyao, Zhejiang) and dated to 5000–4500 B C E, it is confirmed that sericulture and silk-weaving were already practiced during the Neolithic era. However, it is not yet possible to determine when cultivation of mulberry trees began, since fabrics made from the cocoons of wild silkworms (shanjian 山繭) raised on wild tree species such as oaks also existed. Cultivation of mulberry trees in Zhou times can be confirmed by the Book of Odes, which is replete with references to mulberry orchards, pruning mulberry trees, and women gathering mulberry leaves. Material evidence shows that reeled silk was used in Shang times and perhaps earlier. By the Warring States period at the latest spindle wheels for reeling silk had replaced spinning whorls, and treadle-operated looms most likely had appeared by this time as well. Sima Qian’s Historical Records states that during the Warring States era the silk goods of Qi were marketed throughout all of China.7 We also find references to sericulture as part of the domestic production of farming

7

SJ, 129, p. 3265.

63

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

families. For example, Mencius wrote that “if all farming households plant mulberries in their gardens the elderly will be spared from freezing.”8

Animal Husbandry and the Decline of Hunting In the Shang dynasty, the royal household bred horses, cattle, and other livestock. According to Okamura Hidenori, the majority of animal remains from sacrificial rituals conducted at the capital at Anyang were cattle: sixteen sacrificial pits have yielded bones from forty oxen, 107 sheep, and ninety-eight dogs.9 Shang oracle bone inscriptions also reveal that horses bred elsewhere were imported to the capital region. In 2005, five chariot burials were excavated at Anyang that confirm the presence of horse breeding. Various interpretations have been proposed. One theory, emphasizing the marked differences between Shang and the livelihood observed at Neolithic sites, maintains that large-scale stock raising began only after the establishment of strong royal authority. Others suggest that the explanation lies in the differences in the ethnic origins of the Shang and the Neolithic peoples. However, hunting – for which the oracle bones contain meticulous records – for the purpose of sacrificial rituals was an important duty of the Shang kings. A decline in the abundance of prey can be discerned by the end of the Shang dynasty, and hunting of wild animals no longer sufficed to meet the need for frequent sacrificial offerings dedicated to the ancestors. The connection between the decline of hunting and the rise of large-scale stock raising should be a subject of further study. Fisheries References to fish breeding begin to appear in the Spring and Autumn period. Given that most of the immense expanse of China was far from the coasts, fish farming was conducted primarily by using freshwater sources such as rivers, lakes, and ponds. Fish were raised almost exclusively as a source of (relatively luxurious) human food, in contrast to the Japanese archipelago, where people had ready access to the sea and fish provided an important source of fertilizer. Zhou bronze inscriptions record the reclamation of wetlands and tidal flats during the initial phase of settled agriculture in the Yellow River basin, which decreased the natural surface-water area. It became necessary to rely on man-made reservoirs for aquaculture. South of the Yangzi river, artificial reservoirs for irrigation developed along with the 8 9

Mencius 7A.22. Okamura Hidenori 岡村秀典, Chu¯goku kodai o¯ken to saiji 中国古代王権と祭祀 (Tokyo, Gakuseisha, 2005).

64

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

spread of rice cultivation. The use of such reservoirs for fish breeding seems to have become universal in Han times and later. The importance of securing and utilizing water resources in the development of fish breeding in China must not be overlooked.

Breweries In 2004, remnants of brewed liquor made from rice, fruit, and honey were detected on ceramic shards unearthed at the Jiahu (Wuyang, Henan) Neolithic site, dated to 7000–5500 B C E. This site represents the earliest known case of brewed liquor. In addition, pots containing traces of liquor and seeds of fruits believed to be ingredients for brewing liquor have been excavated from Shang sites, demonstrating the crucial importance of liquor in Shang sacrificial rituals. Bronze inscriptions and other historical documents from the Zhou era record that excessive consumption of wine was one of the chief political criticisms raised against the Shang kings. Here, too, we can see that liquor brewing has a long history in China.

Evolution of Production Relations and Environmental Change Shang to Spring and Autumn Era Debate persists over what type of agriculture was practiced in the vicinity of Anyang in Henan when it was the capital of the Shang dynasty. Most Chinese scholars believe that agricultural labor was performed by unfree slaves. In contrast, Japanese scholars are inclined to see the Shang royal authority in the context of a clan-based society in which the religious authority of the Shang king is acknowledged by all members of the royal clan, who also participate in the political organization of the kingdom. It is widely believed that agricultural production was carried out by subordinate clans who were subject to Shang rule and paid tribute to the royal clan. Not only did they undertake agricultural tasks as a clan group, but they also collectively engaged in production of textiles, and obtained animal protein through hunting and stock raising. It is also believed that the yield of farming and other activities was provided as tribute to the ruling clans, including the Shang royal clan. Due to such differences in opinion, interpretations of Shang archaeological remains also vary. Among Chinese researchers, the discovery of large numbers of decapitated skeletons from burial sites is proof of the subjugation and subordination of a slave class. But in Japan it is proposed that they were casualties of war between the Shang and its rivals, or captive 65

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

enemies sacrificed as offerings in religious ceremonies. If slaves were used for agricultural work, it would not make sense to slaughter the labor force on such a massive scale. Scholarly opinion regarding the existence of slavery under the Zhou dynasty likewise diverges. In the classical writings of the Zhou themselves we read that members of the defeated Shang ruling class, referred to as “the seven clans of Yin,” were distributed as unfree populations. In addition, large numbers of servile producers were subordinated to the Zhou king. We read in the Book of Odes, Hark! King Cheng Has issued his command. Lead your husbandmen, Sow the hundred crops, Hasten to break ground on your private fields, Thirty li around. With earnest effort till the soil, By tens and thousands, working in pairs.10

Many scholars in China believe that these lines refer to a slave mode of production. But in Japan it is universally agreed that the poem describes agricultural rituals performed by numerous officials and farmers under the leadership of King Cheng, the first Zhou king to rule after the conquest of Shang. On the other hand, recent research on bronze inscriptions has led to new interpretations regarding cultivators. In a number of cases of land transactions, the owner to whom the cultivator is subordinated changes along with transfers of landownership. It appears that personal dependency on the owner of landed property gradually supplanted collective subordination, resulting in recognition of the cultivator as attached to a specific plot of land. In this situation, the farmer who is the object of this “exchange” in reality is detached from his kinfolk and instead is defined socially as the cultivator of land belonging to his master. However, generally speaking, most scholars think that during the Zhou dynasty members of subordinate groups lost their rights as individuals. Descriptions of villagers’ collective production activities in pastoral poetry such as the “Seventh Month” song in the Book of Odes are seen as evidence of the continuity of Shang forms of collective subordination during the Zhou kingdom as well. This suggests that the degree of specialization in their daily work, especially in cereal 10

“Yixi” 噫嘻 (Ode 277), in ShJ, 1, pp. 591–2.

66

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

production, was not particularly high. The need for a collectively performed complex of hunting, gathering, and herding as well as agricultural work remained. Of course, there were kin groups who exclusively engaged in bronze metallurgy as well as those whose principal task was hunting the fowl and mammals whose furs, skins, and feathers provided indispensable tribute goods for the ritual regalia of the Zhou rulers. To be sure, when the Zhou king reclaimed the wetlands of the Wei river valley for cereal cultivation, these kin groups could no longer engage in hunting and fishing in this region. As we can see in bronze inscriptions, the Zhou king compensated these groups for the loss of hunting grounds by allocating to them forest lands in the middle course of the Jing river valley.11 According to the records of classical texts, it was in the Spring and Autumn era that this situation changed dramatically as a consequence of the power struggles which flared among the various Zhou vassals, each of whom sought to strengthen his own state and achieve supremacy within the Zhou ecumene. In addition to pursuing status and power, the now independent lords aimed at increasing cereal production in order to raise revenues and meet the rising costs of warfare, while simultaneously seeking to gain a reputation as “virtuous” rulers who stabilized the livelihoods of their subjects. One product of this quest was the introduction of reforms in the methods of revenue collection. According to the Chronicles of Zuo, it was the state of Lu in 594 B C E that initiated the collection of land taxes in grain.12 In place of the village (yi 邑) as the unit of assessment, as in the past, a new method of calculating the amount of land under cultivation in terms of mu 畝 units and collecting taxes based on a standard rate per mu was instituted. This change encountered strong criticism from intellectuals of the noble class – notably the Confucian philosophers, who adhered to the traditional modes of tribute collection – but was universally adopted during the Warring States period. This innovation has been seen as a watershed event symbolizing a wholesale social transformation whereby the farming household became the basic unit of production and taxation. In some cases, the rulers’ kinsmen and ministers obtained independent power that exceeded the power of the rulers themselves. Such changes in power relationships naturally were often connected to the amount of 11

12

See, for example, the Qiu Wei he and related inscriptions in Constance A. Cook and Paul R. Goldin (eds.), A Source Book of Ancient Chinese Bronze Inscriptions (Berkeley, Society for the Study of Early China, 2016), pp. 84–92. Xuangong 15.8, in ZZ, 2, p. 1888b.

67

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

revenue – that is to say the taxes in grain procured from agricultural producers – that the noble lords could command. The appearance of tax collection methods based on the acreage of arable land is believed to have strengthened the power of rulers who had revenue collection rights to the most productive lands. Against this background, the development and dissemination of knowledge of agricultural techniques and the improvement of farm tools can be confirmed. Mastery of new knowledge and of these new technologies eliminated the need for production techniques utilizing collaborative labor by large numbers of people. A mode of cereal production that generated high yields using only personal or family labor began to proliferate. During the early part of the Spring and Autumn era, the rulers of Zheng played a major role in the relocation of the Zhou court from the old capital of Hao to Luoyang. It is said that the Zheng clan, upon establishing its own capital city at Xinzheng (near Luoyang), opened grasslands to agricultural cultivation. Thus the lands reclaimed for farming in the early Spring and Autumn period were flat grasslands. At this time, iron axes suited for felling large, thick trees were not yet widely used, and the pace of deforestation remained relatively slow. Cultivation of hill slopes was difficult. With the growing frequency of warfare among the various states, felling of timber to supply raw materials for weapons accelerated, while the forest lands set aside for hunting and gathering gradually were brought under cultivation. From the late Spring and Autumn era, the pace of forest clearance accelerated along with the spread of iron tools such as axes. This trend was related to the growing consciousness of “state borders” whereby the rulers of states delineated the territory under their control according to easily recognizable topographic features such as large rivers and high mountains. It was necessary to develop complex strategies for remote forests which posed various military advantages and disadvantages and where defense against attack was more complicated compared to flat cultivated areas.

From Warring States to the Unified Qin Empire In the Warring States period, the hundreds of vassal lords of the early Zhou period were consolidated into seven powerful states, and the military conflicts among them intensified. The chief priority for each state was weaponry. Metallurgical industries had attained a high level of technical achievement in the Shang dynasty, but initially the military application of bronze metalworking was confined to items such as arrowheads. Eventually iron was used for sharp-edged weapons such as knives, swords, and spears as well as to reinforce helmets and armor. With the development of cast iron technology, 68

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

iron metallurgy was also diverted to the manufacture of agricultural tools. The Warring States era witnessed the flourishing of mass production of iron wares for everyday use as tools and implements by ordinary farmers. Iron axes accelerated deforestation. The utilization of iron metallurgy for tools such as plowshares and scythes also quickly became widespread. In virgin forested areas, to say nothing of lands already under cultivation, iron tools made possible deeper and faster cultivation, and also increased the amount of grain harvested per unit of crop area. In the classical literature there are many references indicating that the average amount of land a single adult male could cultivate was 100 mu (roughly 1.9 hectares), which yielded on average one to two hu (roughly 19.5–39.0 liters) per mu of grain. Li Kui, chief minister of Wei c. 400 B C E, calculated the budget for an ordinary farming family and determined that the average annual grain harvest of 1.5 hu per mu would be insufficient to meet the consumption needs of a family of five.13 Labor productivity in agriculture jumped dramatically with the spread of ox-drawn plows. As mentioned above, cast iron plowshares have been excavated in many regions. Scythes and other harvesting tools made from iron also were introduced at this time. As desiccation of the soil progressed, plowing techniques were improved, and increased use was made of iron tools such as the moldboard plow, whose share pushed aside the turned-up soil. By using iron tools farmers were able to remove vegetation and tree roots more easily and plow more deeply. In addition, crushing the soil finely increased the aggregation of soil structure (small particles of soil that form larger clods) as a whole, allowing nutrients to be absorbed by the plant roots. As a result, yield per unit of land increased. Of course, the ox-drawn plow also reduced the demand for physical labor by farmers, which made it possible to expand the amount of land a farmer could cultivate. Utilizing iron harvesting tools further enhanced the efficiency of labor. The impact of the diffusion of iron agricultural tools can be gauged by various references to the size of the family farm. A significant change in size of the mu unit of land can be confirmed from around the time of Emperor Wu of Han. In the past, a mu of 100 square “paces” (bu) – or approximately 1.9 hectares, commonly known as the “old mu” – was considered the average size of a farm worked by one adult male. In Emperor Wu’s time, the “new mu” – measuring 240 square paces or approximately 4.6 hectares – rapidly proliferated. As yet there is no consensus regarding when, where, how, and 13

HS, 24A, pp. 1124–5.

69

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

why this change occurred. A variety of systems of measurement were employed by the various states of the Warring States period, and the units of land measurement varied across space and time. But it is widely agreed that the change to the “new mu” unit accompanied the spread of ox-drawn plow cultivation across north China and reflected the increased productivity realized by reducing the work of reversing the direction of the ox at the edge of the field. During the Spring and Autumn era, the ties that had bound kinship groups in the Western Zhou period gradually atrophied, and individual cultivators established their own economic autonomy. The Warring States era marked the age of the so-called “hundred philosophers,” who proposed policies for economic and social reform to the rulers of the various states. Some plans to adjust tax rates based on the fertility of the soil can be found in their extant writings, which indicates that the spread of cultivation techniques brought about general recognition of natural conditions on land productivity. Furthermore, we can see from these policy recommendations that the conjugal household comprising five to eight persons was universally acknowledged as the social norm. For example, the Confucian scholar Mencius espoused a social ideal of a family of five to eight persons that would have sufficient labor to plant mulberries and raise pigs and chickens and thus would have sufficient resources to take care of the elderly who no longer could provide for their own food and clothing.14 We can see this development in the legal and economic reforms instituted in Qin by the minister Shang Yang in the mid-fourth century B C E (see von Glahn in Chapter 3 of in this volume). In the second round of reforms enacted by Shang Yang in 350 B C E, the inhabitants of conquered states were forcibly relocated to the Yellow River plateau to open lands for cultivation, and arable lands were partitioned into standardized units and allocated to farming households. The intention behind Shang’s measures was to engage as many people as possible in agriculture and convert all land to grain production. Originally, the Qin territory was located in a semi-pastoral region in modern Gansu province. Following the relocation of the Zhou court to Luoyang, the Qin advanced into the Wei river basin, formerly the homeland of the Zhou royal house. Although in the past many of the elite lineages of Qin had raised livestock, they were instructed to devote their energies to agriculture, especially cereal production. The emphasis on the primacy of agriculture meant that farmers were expected to follow the ruler’s policies 14

Mencius 1A.3, 1A.7.

70

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

obediently. They were prohibited from changing their residence between the time of planting and the harvest, nor could they evade requisitions for tax payments or summons to military service. Subsequently other states recognized the success of Shang Yang’s policies to “enrich the state and strengthen the army,” and a growing number of states, such as Wei, mentioned above, adopted similar policies. As part of his strategy for increasing yields from agricultural production, Shang Yang stressed the importance of weeding to preserve soil nutrients and moisture. In The Book of Lord Shang – ascribed to Shang Yang, but regarded as the work of later writers – Shang Yang is said to have issued orders for reclaiming wilderness lands, and many passages throughout this work call for weeding work. The importance of weeding and the use of tools such as hoes for weeding can be seen in many works of political philosophy. For example, we can see in statements such as that of Mencius urging the King of Liang to exhort his people to “plow deeply and weed frequently” the idea that honest toil in agriculture promotes public and private virtue: If Your Majesty practices benevolent government towards the people, reduces punishments and taxation, urges the people to plow deeply and weed frequently, and if the able-bodied men learn, in their spare time, to be good sons and good younger brothers, loyal to their prince and true to their word, they will serve their fathers and elder brothers within the family and serve their elders and superiors outside the family.15

Gradually the activities subsumed under the word “hoeing” (nou 耨) came to include any kind of work that contributes to soil aggregation by pulverizing the soil around plant roots, which makes it easier for the roots to ingest nutrients. Weeding tools emerged in an array of forms, along with knowledge of their proper use according to the type of crop, the growth period, and so on. At the same time, in every state great consideration was given to maintaining a balance between farming and pastoralism. For example, when the state of Qi, originally under the rule of goat-herding pastoralists, occupied the eastern part of the Shandong peninsula in Western Zhou times, its population still seems to have included “Eastern Barbarians” (dong yi 東夷) who were primarily engaged in stock raising. It appears that farmers commonly raised livestock in pens around their homes. But in the latter part of the Warring States period, as grain production became the chief priority of the 15

Mencius 1A.5.

71

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

state, rulers began to exercise ownership rights over livestock and the populace was required to declare how many domestic animals they were breeding. In addition, the custom continued whereby unfree persons belonging to the state (gongmin 公民) were sent to each village during the slack season at the end of the year to take care of the livestock while the villagers went out hunting the game that would be offered in sacrifices at the New Year rituals.16 The increase in yield per unit of land was abetted not only by the development of tools, cropping techniques, and state policies, but also by climatic stability in which rainfall and average temperatures increased compared to the Spring and Autumn era. Although we see records in the transmitted texts of ancient China stating that Chinese agriculture depended on irrigation, in most areas cereal cultivation was carried out using dry-land cultivation techniques, whereas it was general practice to cultivate vegetables in small-scale gardens near dwellings irrigated by well water. We have already mentioned that the Qin state constructed the renowned irrigation channel known as the Zheng Guo Canal on the northern portion of the Wei river basin. Historical narratives celebrate King Zheng of Qin (later the First Emperor) for adopting the proposal by Zheng Guo – originally a spy sent by the rival state of Hann – to invest enormous sums to irrigate these alkaline lands and developing rice farming in an environment otherwise unfavorable to rice cultivation. In reality the rising trend in temperature was a basic condition for cultivating rice in this region, and the success of the Qin state’s “enrich the state and strengthen the army” policies was partly a consequence of the climatic warming trend. Advances in agricultural techniques gave rise to the emergence of a school of “Agriculturalists” among the hundred philosophers, including Xu Xing, who advocated a utopian society in which men and women all lived off the fruits of their own labor.17 However, I would like to draw attention to the detailed knowledge of various cultivation techniques preserved in a set of four essays incorporated in Spring and Autumn Annals of Master Lü.18 These chapters mainly focus on the increases in food production that can be 16

17

18

“Wangfa pian” 王法篇, cited in Yinqueshan hanmu zhujian zhengli xiaozu 银雀山汉 墓竹简整理小组 (ed.), “Yinqueshan zhushu ‘Shoufa,’ ‘Shouling’ deng shisan pian” 银 雀山竹书《守法》《守令》等十三篇, Wenwu 文物 1985.4, 27–38. A.C. Graham, “The ‘Nung-chia’ 農家 or ‘School of the Tillers’ and the Origins of Peasant Utopianism in China,” Bulletin of the School of Oriental and African Studies 42.1 (1979), 66–100. Book 26, LSCQ, 2, pp. 1710–1807. Cf. John Knoblock and Jeffrey Reigel (trans.), The Annals of Lü Buwei (Stanford, Stanford University Press, 2000), pp. 650–67.

72

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

achieved by adopting cereal production as the mainstay of livelihood in place of not only pastoralism and trade but also the hunter-gatherer economy. According to “Requirements of the Land,” which presents the principles of cultivation methods in detail, the method of sowing seeds should vary according to differences in soil type. In addition to general standards for the width (authors’ interpretations differ, but it is generally regarded as one chi or twenty-three centimeters) and depth of the furrow, seed should be sown in furrows on higher-lying fields, but on ridges on lower ground. Meticulous instructions are also given for the use of tools for flattening the soil, the width of the furrows, spacing between plants, weeding, and so on. As a matter of course, it states that in weeding and plowing, “plow five times and hoe five times, making certain to do it carefully, for completeness.”19 The “Discriminating Types of Soil” chapter describes the methods of dry-land cultivation mentioned earlier, such as securing adequate intervals of fallowing appropriate to each crop, the principles of appropriate thinning, the necessity of elaborate weeding, and so on. In addition, “Examining the Seasons” gives advice regarding seasonal agricultural work such as the appropriate times for planting and harvesting each crop. These chapters show that since the reforms of Shang Yang the Qin state sought to train government officials to provide technical guidance to induce people previously unaccustomed to agriculture to engage in cereal cultivation. In addition to agricultural produce, government officials also were involved in the management, procurement, and processing of many flora, fauna, and mineral products from forest and mountain regions, for example wooden utensils used in the eating and drinking vessels of the palace, resins used to make lacquer wares, mercury and other minerals used to make metallic lacquers, mulberry for the manufacture of paper, and charcoal to make ink. Maps drawn on wooden boards excavated at Fangmatan (Tianshui, Gansu province) dated to 239 B C E indicate the types of trees in each of the various river basins near modern Tianshui city and the locations of the mills where the timber was cut into lumber. Already in the Qin period tree species were being managed as natural resources by government officials with the requisite knowledge and skills.

Agricultural Development in the Han Dynasty Beginning with Liu Bang (Emperor Gaozu, r. 202–195 B C E), founder of the Han dynasty, the early Han emperors sought to relieve their subjects from 19

Book 26, Chapter 4, “Rendi pian” 任地篇, in LSCQ, 2, p. 1731; Annals of Lü Buwei, p. 656.

73

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

the miseries of the incessant warfare of the preceding centuries through economic development and stabilization of the people’s livelihood. Manifestations of this goal can be seen in the reduction of the land tax twice during the reign of Emperor Wen (r. 180–157 B C E), and the complete suspension of the land tax during 167–157 B C E. This suggests that the lives of the farming population had been eased to some extent. Nonetheless, family farms engaged in cereal cultivation were still immersed in the market economy. The government collected head taxes paid in coin from the entire population, and, in keeping with the principle of “maximizing the potential of the land,” it had become commonplace since the Warring States era for farmers to sell a portion of their produce in the marketplace. The Qin Empire had abolished the customs stations on the frontiers of the Warring States kingdoms, and interregional commerce was thriving. Merchants engaged in long-distance trade were said to have made great fortunes. The Han central state’s Grand Superintendency of Agriculture (dasinong 大司農) and Superintendency of Waterways and Parks (shuiheng duwei 水衡 都尉) employed expert specialists in farming who might be called “agrarian technocrats,” and many new agricultural innovations should be credited to their achievements. Fan Shengzhi, who compiled the earliest extant agricultural treatise from China, The Book of Fan Shengzhi 氾勝之書, occupied the office of imperial secretary during the reign of Emperor Xuan (r. 74–58 B C E). The “alternate-fields” (daitian) cultivation technique discussed earlier was said to have been developed and popularized by an imperial official, Zhao Guo (see below). A remarkable feature of the early Han was the construction of large-scale canals utilizing techniques that had been developed since the Warring States period. The earliest records refer to transport canals used to secure the food supply in Guanzhong, reflecting the need to transport grain from cerealproducing regions throughout the middle Yellow River valley to provide for the enormous population in the region around the capital, Chang’an, in the heart of the Guanzhong basin. The waterway known as the Cao Canal 漕渠 (meaning “transport channel”) – finished in 129 B C E by the waterworks engineer Xu Bo, based on a proposal by the Grand Superintendent of Agriculture, Zheng Dangshi – was intended for shipment of grain from the south bank of the Wei river to Chang’an. Once it had been completed, however, waters from the Cao Canal were utilized to irrigate farmlands in adjacent areas. Although we cannot confirm definitively which crops were cultivated, at this time the taxes on rice paddies in the capital region were high, and it is possible that the Cao Canal, like the Zheng Guo Canal, was 74

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

used to irrigate rice fields. Still, considering the consumption situation in the vicinity of Chang’an at that time, rather than paying high taxes to cultivate cereals, it is possible that farmers converted their fields to cultivation of more profitable vegetable and orchard crops. Among the various categories of “the rich” listed by Sima Qian in his “Biographies of the Moneymakers” – those whose income was deemed comparable to that of imperial princes and nobles – he included not only great landowners engaged in cereal production, but also cultivators of “a thousand mu of ginger and leeks” along with owners of fruit orchards, those who managed forests, those who raised large herds of stock animals, and usurious moneylenders. Thus it was said that in the time of Emperor Wu those who grew vegetables such as ginger and leeks for which consumption demand in the vicinity of Chang’an was high could become millionaires.20 The necessity of securing the food supply also led to the construction of waterways such as the Six Auxiliary Canals which, like the Zheng Guo Canal, required techniques for drawing water from upland areas. The construction of the Six Auxiliary Canals was proposed in a memorial of 111 B C E from the governor of the eastern capital district, Ni Kuan. The edict issued by Emperor Wu noted the benefits the people reaped from this irrigation project, stating that nowadays the tax on the rice fields in the capital region is much heavier than elsewhere. Let proposals for reducing the tax rate be discussed. Let the officials and the people devote themselves to farming so as to exploit fully the bounty of the land. Let the labor service be equalized and the water made to flow so that they do not miss the proper seasons for sowing and harvesting.21

From such statements we can deduce that in the Han period much land was developed for rice cultivation using such irrigation channels. In addition, in the alkaline desert-soil regions north of the Wei river where there are lands that tend to sink as a result of the action of sodium carbonate, for example in areas adjacent to the Luo river and natural springs, farmers employed the technique – like the qanat underground aqueducts developed in Iran in the early first millennium B C E – of digging wells connected by subterranean channels. Millet remained the principal crop in north China, but the cultivation of wheat and barley expanded significantly during the Han dynasty. It was reported that cultivation of winter wheat was urged for areas subject to 20

SJ, 129, p. 3272.

21

HS, 29, p. 1685.

75

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

repeated flooding in the lower Yellow River valley around the year 120 B C E . Prior to this, the renowned scholar and statesman Dong Zhongshu wrote in a memorial addressed to Emperor Wu, In the Spring and Autumn Annals there are no records of bad harvests of any of the cereal grains except millet and wheat. From this we can see that the Sage [i.e., Confucius] regarded millet and wheat as the most important of the five cereals. At present, the people of the Guanzhong [capital] region by custom are not in the habit of planting wheat . . . May Your Majesty direct the grand superintendent of agriculture to ensure that the Guanzhong populace sows more winter wheat in timely fashion while it is still the proper season for planting.22

In areas prone to flooding, the loss of autumn harvests due to natural disasters could be mitigated if, as soon as the water is drawn off, winter wheat is sown in the fall to be harvested the following summer. From Dong’s statement we can see that cultivation of winter wheat was not yet common in the Guanzhong region. But a century later Fan Shengzhi observed that wheat and barley were widely planted in Guanzhong and produced plentiful harvests. Repeated cropping of cereals naturally led to decline in soil fertility, and the reduction in forests not only entailed a decline in natural fertilizers (forest humus formed by the accumulation of dead floral and faunal matter, excrement, dead trees, and fallen leaves) to supplement soil fertility, but also placed these sources at a farther remove from settlements and cultivated lands, thus adding to the problems of maintaining the productivity of the land. Chinese agriculture was not integrated into pastoralism as was the case in Western Europe. As was often customary for poor farmers with limited availability of land, stock raising was relegated to pastoral peoples such as the Xiongnu living beyond the Great Wall frontier, and sedentary farmers purchased livestock that provided animal power and food supply from the pastoral peoples. For this reason, Chinese farmers lacked animal manure to fertilize cultivated lands, and the decline in soil fertility had significant consequences. The ridge-and-furrow technique – specifically the “alternate-fields” (daitian) method developed by the agriculture official Zhao Guo – was designed to counteract the problem of alkaline regeneration caused by irrigation of dry-land crops. In Zhao’s system, new lands were divided by three furrows twenty-three centimeters in depth and twenty-three centimeters in width 22

HS, 24A, p. 1137.

76

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

(a total of 300 trenches for 100 mu of land). In the first season, seeds were sown in the furrows, and in the following year they were sown in the ridges, hence the name “alternate field.” When the seedlings germinated and began to sprout leaves, weeding was carefully performed and soil on the ridge was pushed down to surround the roots, deepening the roots at the height of the summer heat in order to prevent drought or wind damage. Zhao first experimented with this practice in the capital region, obtaining a yield of 6.4 hu (124 liters) of grain per mu, three or four times greater than average harvests. Gradually this technique also was applied in areas in Guanzhong and then extended as far as Juyan in the western frontier region. Local officials summoned village leaders, supplied them with the necessary tools, and instructed them in the cultivation technique. In places where oxen were unavailable, plows were pulled using human labor. The people found this method advantageous, and it was widely adopted. Although the cultivation system of the “alternate-fields” method recorded in the “Food and Money” chapter of the History of the Former Han is very similar to that described in Spring and Autumn Annals of Master Lü, in contrast to the handwork of weeding and even plowing in the latter the “alternatefields” method used ox-drawn plows. Whereas in later times the plowshare would turn aside the excavated soil, making it possible to use the entire cultivation area, the “alternate-fields” method required the ridge-and-furrow method that overturned soil on both sides. The Grand Superintendency of Agriculture took an active role in developing and manufacturing special tools and disseminating the “alternate-fields” cultivation technique. Excavations at the Sanyuanzhuang site in the lower Yellow River valley show that villagers there practiced ridge-and-furrow cultivation using ox-drawn plows with large iron shares. The Sanyuanzhuang settlement was inundated by flooding and abandoned in the early first century C E, and thus preserves the agrarian landscape of that time.23 Greenhouse cultivation also was practiced in the Han dynasty, although perhaps only on a modest scale. In 33 B C E, when Zhao Xinchen was placed in charge of the Lesser Treasury (shaofu 少府), which acted as the Privy Purse, he exerted himself to tighten imperial finances. An office within the Lesser Treasury responsible for the preparation of palace meals would sow seeds for vegetables such as onions and leeks in winter, roofing over the planted area and stoking fires day and night to keep the young plants warm. Zhao put 23

Tristram R. Kidder, Haiwang Liu, and Minglin Li, “Sanyangzhuang: Early Farming and a Han Settlement Preserved beneath Yellow River Flood Deposits,” Antiquity 86.331 (2012), 30–47.

77

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

a stop to this practice, arguing that consuming such foods out of season was harmful to health. This economizing measure reportedly saved the imperial purse several tens of millions of coins per year, but stymied the development of this method of cultivation.24 Apart from records of state-led agricultural policies and practices, and rare settlement excavations such as at Sanyuanzhuang, there is little direct evidence for the actual situation of Han agriculture. According to the census of 2 C E recorded in the geographical treatise of the History of the Former Han, the entire territory of the Han Empire encompassed 145,136,405 qing, of which lands not under cultivation, such as dwellings, roads, mountains, rivers, forests, and marshes, comprised 102,528,889 qing. Potentially arable but not yet cultivated lands amounted to 32,290,947 qing, but only 8,270,536 qing was under continuous cultivation, suggesting that land was not intensively cultivated. This conclusion is largely supported by the evidence of the “Submitted Account” document excavated at Yiwan (Lianshigang, Jiangsu), which provides information on the total population, households and individuals subject to taxation, land under cultivation, and other data for Donghai commandery in eastern China in 13 B C E. According to the figures in this document – which are not entirely complete, because of broken slips and illegible graphs – the entire registered land of Donghai amounted to 512,092 qing and eighty-five mu. The amount of land sown in winter wheat was 107,300 qing, an increase of 1.8 percent from the previous year. The acreage devoted to silviculture was 656,794 mu, an increase of 7 percent from the previous year. Thus the increase in the portion of land dedicated to silviculture vastly outpaced the increase in land under wheat cultivation. State-managed iron foundries and salterns, which required considerable quantities of fuel, were in operation in Donghai. It appears likely that the demand for fuel, at the time most importantly in the form of firewood, exceeded the demand for food production, and thus necessitated planting of state-owned forests. Sericulture also advanced during the Han dynasty, perhaps stimulated by the proliferation of small family farms and the growth of textile production by women working at home. Mulberry cultivation and remnants of looms in farmhouses at the Sanyuanzhuang site confirm that farming households engaged in domestic silk production. Sericulture naturally required periodic disposal of waste such as cocoon shells, silkworm excrement, remainders of mulberry stems and leaves used to feed silkworms, and bamboo mats used to provide a clean environment for raising silkworms. These waste products piled 24

HS, 89, p. 3642.

78

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

up in vacant spaces near dwellings. Doused by rainwater, this waste produced rich organic matter known as “silkworm dung” (canshi 蠶矢) which could be used as fertilizer. In other words, the production of silk did not simply create beautiful fabrics. Sericulture production was part of a series of interconnected trends that contributed to the displacement of animal husbandry to regions beyond the Great Wall, the conversion of virgin forests to fields for cereal production using the dry-land cultivation techniques, and the transformation of fertilizer sources – what might be called “industrial waste” – into valuable commercial commodities necessary to counteract the loss of soil fertility due to repeated cropping with cereal grains. Despite government policies favoring the development of small family farms, the economic prosperity and commercial growth of the early Han promoted the concentration of wealth and landholdings in the hands of princely households, government officials, and great merchants. Moreover, by the time of Emperor Wu the Han central government was suffering from serious fiscal strains. In order to avoid conflict with the increasingly powerful Xiongnu nomad confederation on the empire’s northern frontier, the early Han rulers favored a policy of appeasement by sending tribute of silk, grain, and other goods to the Xiongnu. By the time of Emperor Wu, the annual tribute amounted to 10,000 piculs of rice and liquor, 5,000 bushels (hu) of grain, and 10,000 bolts of silk. In order to halt this drain of treasure, Emperor Wu abrogated these tribute payments and launched military campaigns against the Xiongnu. But the conscription of farmers to serve in the army and increased requisitions to supply the soldiers with food rations, horses, fodder, weapons, and other equipment only worsened the burden on the people. Emperor Wu enacted measures to raise state income and procure military supplies, imposing government monopolies on the salt and iron industries (see von Glahn, Chapter 3 in this volume), but these policies had a deleterious effect on private commerce. Long-distance trade declined sharply, and rich households turned toward the accumulation of landholdings as the most secure form of investment. Therefore the social stratum known as the magnate clans (haozu 豪族) rose to dominance, and many farming families seeking to escape the burden of taxation and military service subordinated themselves to, and became dependents of, these magnate clans. The condition of these subordinate persons, referred to as “guests” (ke 客) but in reality treated as menial servants by their masters, is cogently described in the “Slave Contract” (tongyue 僮約) satire composed by Wang Bao during the reign of Emperor Xuan. Wang relates that when a man in Chengdu 79

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

(Sichuan) by the name of Wang Ziyuan purchased a slave named Bian, he drew up a contract that detailed all of the kinds of work Bian was expected to perform, including dredging and repairing the irrigation ditches; planting and harvesting ginger, taro, gourds, melons, and onions; cultivating and harvesting fruits such as peaches, pears, apricots, and persimmons; picking chestnuts; raising pigs, oxen, and horses; hunting wildfowl; fishing; husking grain along with processing all of these farm products and buying and selling them in the marketplace; making brooms and woven mats for cleaning the residence; and doing the washing. The slave was expected to eat beans and drink water, forbidden to consume liquor, and subject to a hundred lashes with a staff if he violated any of the terms.25 Wang Mang, the chief minister who usurped the Han throne in 9 C E, attempted to check the rising economic power of the magnate clans. But Wang’s efforts to create a state-directed command economy encountered fierce opposition, and Wang was overthrown and put to death in 23 C E. After Liu Xiu restored Han rule upon ascending to the throne as Emperor Guangwu (r. 25–57 C E), the magnate clans regained their domination over Han government, society, and economy. As the magnate clans extended their landholdings to encompass the diverse environments of forests, marshlands, and orchards, the general farming population gradually lost access to firewood and other resources.

Climatic Cooling and Cropping Changes in the Later Han and Period of Disunion The restoration of Han rule especially increased the power of the magnate clans of the Nanyang region in the Huai river basin, where Liu Xiu’s own estate and those of a number of leading families related to Liu by marriage were located. The Huai river basin became the prototype for constructing irrigation reservoirs planned and implemented by local officials with the collaboration of magnate clans in place of the impressive large-scale irrigation projects of the early Han period. These irrigation reservoirs, connected to each other by water channels lined with large dikes known as po (坡), resulted in a proliferation of small-scale water projects. In his Monthly Ordinances for the Four Peoples (四民月令), Cui Shi (c. 103–c. 170) – a native of the Luoyang region who served as local magistrate in several localities – wrote extensively about everyday life at his own estate and the annual cycle of festivals. In this 25

QHW, 42.11b–12b. The full text of the mock contract is translated in Cho-yun Hsu, Han Agriculture (Seattle, University of Washington Press, 1980), pp. 231–4.

80

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

work Cui gave detailed instructions about all aspects of daily life, from how the family head should lead the junior members of the household in hierarchical order to perform sacrificial rituals at the New Year season to the proper seasons for each type of agricultural task – not only the sowing and cultivation of crops but also when to sell each type of produce and when to purchase building materials and livestock in preparation for the following year – as well as matters such as the malt used to make various types of food; assorted types of pressed oil; preparation of seasonings such as the sauces used to preserve meat and fish; the manufacture of paper, ink, and other instruments for writing; and how to protect books from mold and insect infestation. Although the full text of Cui Shi’s treatise has been lost and we only know of it from the portions quoted in Essential Techniques for the Common People and other works, the book is full of information about life in the Eastern Han era on topics ranging from the hiring of the seasonal agricultural laborers known as “guests” (ke) to the use of mulberries as supplemental food to stave off famine in times of poor harvests. The gradual cooling that set in at the beginning of the Eastern Han period resulted in a decline of paddy rice cultivation in north China in favor of planting millet and wheat. Harvest yield per unit of area also declined, while reports of famine and epidemic increased. During the Three Kingdoms period (third century C E), freezing of the Yellow River and other major waterways in winter became a common occurrence. But climate change was not the only adverse phenomenon that struck agriculture in this era. According to the biography of Deng Ai in the History of the Three Kingdoms, at the close of the decisive battle between the rival kingdoms of Wei and Wu, Deng was charged with the task of securing food supplies for the Wei troops. Deng made a detailed tour of inspection of the Huai river valley and compiled a hydrological treatise entitled “Water Management” (Jihe lun 濟河論). Deng reported that although the quality of soils in the vicinity of Xuchang [Henan, near the headwaters of the Huai river] is poor, the land has been cultivated with paddy rice. These should be abandoned, and channels dug to drain the water toward the east. Farmer-soldier colonies (tuntian 屯田) should be established on the north and south banks of the Huai river . . . If these fields are planted in millet and wheat, we will be able to obtain 5 million bushels (hu) of grain in military provisions per year.26

26

SGZ, 28, pp. 775–6.

81

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

In 241, the Wei completed construction of the Guangcao Canal connecting the Huai and Yangzi rivers, which not only enabled procurement of food supplies, but also facilitated the transfer of troops and reduced the risk of flooding. In the judgment of the authors of the History of the Three Kingdoms, the triumph of Wei over Wu could largely be attributed to Deng’s exploits. In other words, we can say that the production of grain was altered in response to military exigencies. Throughout the Eastern Han and Three Kingdoms periods, the influx of pastoral peoples into north China continued in concert with the steady advance of climatic cooling. Following the sack of Luoyang in 311 and the flight of the Jin dynasty (280–420) court to the Yangzi river delta, China was divided into northern and southern political regimes, with numerous dynasties of diverse ethnic origins established in north China, while south of the Huai river Chinese rule was preserved by a succession of five dynasties – Jin, Qi, Song, Liang, and Chen. With the climatic cooling in the steppe grasslands and the decline in income from stock raising on top of the influx of nonagricultural peoples such as the Xiongnu and the Qiang into north China, the numbers of people fleeing the devastation of war and migrating to the Yangzi river basin soared. Consequently, agricultural production centered on rice cultivation increased in central China. This trend also gave rise to the cultivation of crops that were difficult to grow in the Yellow River basin, such as bamboo shoots and tangerines, which were luxury items rather than necessities of daily life but for which demand was strong. In military terms, the freezing of marshes, rivers, and grasslands during the winter months enabled the swift movement of cavalry, and peoples who were skilled at horse riding achieved dominance. Among them were the Xianbei, whose ruling Tuoba clan became the emperors of the Northern Wei dynasty and succeeded in unifying north China. Initially the Northern Wei maintained the traditional social organization of the Xianbei people and until the reign of Emperor Daowu (r. 398–409) its capital was located at Pingcheng (modern Dazu, Shanxi), on the northern margin of the Chinese world. With the accession of Emperor Taiwu (Tuoba Tao, r. 423–452), the political system began to adopt features of the Chinese-style aristocracy. Subsequently, under Emperor Xiaowen (r. 471–499) – but mostly while Empress Dowager Feng, consort of Emperor Wencheng (r. 452–465), controlled the reins of government, from c. 485 until her death in 490 – the Northern Wei adopted, in addition to cultural policies such as mandating intermarriage among the leading Tuoba and Chinese clans, political reforms that had a major impact on agriculture, such as the equal-field (juntian 均田) tenure system, the 82

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

“three-elders” (sanzhang 三長) village government system, and the zu–diao (租調) tax system. The equal-field system allocated lands for cultivating grain and fiber crops (mulberry or hemp) based on the labor resources of individual households. In addition to a standard land allotment for the male head of household, additional lands were allocated based on the number of adult men and women in the household and the number of slaves and oxen the household possessed (see Chapter 3 in this volume). Under the “three-elders” policy, groups of five families were organized into “neighborhoods” (lin 隣), with five lin constituting one li 里, and five li forming one dang 黨 (125 families). The populace under the control of the magnate clans was attached to the household registers of their masters. Tax payments and labor services (zu– diao) were apportioned to households based on the amount of land they received under the equal-field system in an effort to dissolve traditional tribal allegiances and eliminate the ethnic discrimination that prevailed earlier under Northern Wei rule. Successive dynasties had repeatedly sought to realize such a conception of equality in landownership, but the obstacles to implementing such plans were formidable. The magnate clans possessed vast tracts of land encompassing not only arable land cultivated with cereal crops, but also forests, marshes, and mountains, while typical farm families were barely able to hold onto cereal-producing lands. It is not difficult to imagine the difficulties they encountered in trying to obtain necessities of daily life such as sources of protein, fuel, and raw materials for construction and furniture. It was especially problematic to obtain fuel for cooking purposes given the lack of forests. Once the Xianbei, who were deeply familiar with the steppe lands and the forested regions of the Daxing’an mountains, had occupied the Yellow River basin and observed the decimation of forest lands in that region, they recognized that economic policies solely focused on production of cereal crops were inadequate for sustaining the livelihood of the populace. There was a special feature of the Northern Wei’s equal-field tenure system that has largely escaped attention. At the time when land allocations were first made, when farming families received twenty mu of mulberry lands, they bore responsibility for planting fifty mulberry trees, five jujube trees, and three elms. In areas unsuited for mulberry cultivation, males were allotted one mu of land and held responsible for planting jujube and elms. The same was also true for slaves. The regulations mandated that if the trees had not been planted at the end of three years the allocated land was to be confiscated. 83

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

The state land policies of the Northern Wei dynasty, whose Xianbei rulers originally did not regard cereal production as the mainstay of economic livelihood, thus gave ample consideration to the whole natural environment. Under the Di ruler Fu Jian (r. 357–385) of the Former Qin dynasty, who preceded the Northern Wei in unifying north China, roads were built to connect the capital, Chang’an, to local government offices in each province. Although the First Emperor of Qin and many other subsequent regimes had engaged in large-scale road construction projects, the roads they built were not equipped with rest stations for travelers or stables for providing couriers with fresh horses. Rows of willows and China scholar trees were planted along both sides of the roads built by Fu Jian, providing shade and relief not only for travelers but for the local inhabitants as well. Yuwen Tai (who would be posthumously honored as the founding emperor of the Northern Zhou dynasty) occupied the position of state counselor of the Western Wei for twenty years following the dissolution of the Northern Wei in 534, at the time when the Qimin yaoshu agricultural handbook was composed. Yuwen sought to re-establish the equal-field system of the Northern Wei. In addition, amid the wars against Gao Huan, paramount leader of the Eastern Wei, it is reported that when Yuwen mustered soldiers from the various prefectures into his newly created fubing 府兵 militias, each person was required to plant one tree, and as a result some 7,000 willow trees were planted, each marked with a seal bearing the legend “Military Triumph” (wugong 武功) – further evidence of the extreme shortage of trees in north China by this time.27 The renewed attention devoted to tree cultivation in north China during the Period of Disunion can be glimpsed through the Essential Techniques for the Common People, which is the oldest surviving complete agricultural treatise from China. Its author, Jia Sixie, had been the prefect of Gaoyang commandery (in present-day Shandong province) at the end of the Northern Wei dynasty. The contents of the Essential Techniques cover a wide variety of subjects. It provides a summary of general agricultural techniques and records in detail the techniques for cultivating each variety of cereal crops planted in what is referred to as the “main fields” (datian 大田). In addition, Jia describes fruit tree cultivation; livestock raising; and methods for manufacturing glue, ink, and writing brushes, as well as recording a variety of cooking recipes. His treatise concludes with detailed entries derived from previous writings on crops grown in southern regions outside the Northern Wei territories. 27

BS, 9, p. 332.

84

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

Nishiyama Takeichi and Kumaiyo Yukio have identified the following special features of Jia Sixie’s treatise.28 First, as mentioned above, the crop rotation system known as Norfolk agriculture introduced in Western Europe in the seventeenth to eighteenth centuries had already appeared in China as early as the sixth century. The sophisticated technical level achieved in north Chinese agriculture since the Spring and Autumn era through the accumulation of techniques such as cultivation of unirrigated fields utilizing animal power is clearly demonstrated. Second, although it is recognized that Chinese society of that time was dominated by an aristocracy based on ownership of great landed estates, and commerce had become stagnant, it is noteworthy that small-scale farm management also existed, and trade in agricultural products bought and sold in local markets was not uncommon. Third, it can be seen that specific cultivation techniques – including planting arable land with trees by applying fertilization and irrigation, examples of which are not seen even in other parts of East Asia – were utilized, and profit-oriented activities extended to commodity production not only of fruit trees but of other forms of timber cultivation as well. With regard to the first point raised by Nishiyama and Kumaiyo, Jia Sixie confirms the development of the saddle-shaped moldboard plow with an asymmetrical recessed-cavity share and equipment such as animal-drawn tined harrows (ba 耙) and bush harrows (lao 勞) that could be used not only for spring tilling but also for turning over the soil after the autumn harvest, which served to preserve soil moisture until the beginning of the germination season in the following spring. With regard to the second point, consider the example of safflower cultivation as described by Jia: If there is good land near a town you can obtain an income of 300 bolts of silk by selling it in the market. One qing of land planted in safflower yields a harvest of 200 hu, which sells for the same price as hemp seeds. It can be used for greasing cart axles and candlewax, and of course it can be exchanged for cereals such as millet. Specifically, harvesting one qing of land planted in safflower requires the labor of a hundred persons per day, which is beyond the capacity of a single family. But if you gather together children and bondservants in groups of tens or hundreds to share the work of picking the safflower, as long as one divides the income obtained fairly, giving half to

28

Nishiyama Takeichi 西山武一 and Kumaiyo Yukio 熊代幸雄 (eds.), Ko¯tei yakuchu¯ Seimin yo¯ju¯ 校訂訳注齊民要術 (Tokyo, No¯gyo¯sho¯ no¯gyo¯ so¯go¯ kenkyu¯jo, 1957–1959).

85

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

those who do the labor, in this fashion even a single man or woman can widely plant it.29

Although Jia himself had a career as a local official, his treatise includes a story about his failure when he turned his hand to raising sheep, and it is deeply informed by practical experience in managing a farm. Regarding the third point, Jia lists cultivation methods for numerous tree species, including not only fruit trees but also mulberry, elm, poplar, crab apple, lacquer, willow, scholar tree, Mallotus, catalpa, Pauwlonia, parasol tree (firmania), oak, Xylosma, and bamboo, among others. For example, in the case of the planting method for Mallotus (qiu 楸), Jia writes, since the trees produce no seeds, dig around all four sides of a large tree and collect roots to be transferred. Plant one root in each direction at a distance of two paces [about three meters]. Plant one row for every two mu of ground, with 120 trees in one row, or 600 trees in five rows. Ten years later, in addition to selling each mature tree for 1,000 coins, you will be able to use the timber for making carts, musical instruments, and weiqi [a game best known in the West by its Japanese name, go] boards. As a material for coffins it is superior to pine and cedar.30

Jia was discussing tree cultivation in small-scale plots with the aim of selling forest products in the marketplace. After giving advice on how to cultivate willow trees, Jia calculated their economic value as follows: Saplings grow fast, and in three years they will be stout enough to use for rafter timber. Although the willow is somewhat fragile compared to other tree species, it still can be utilized for this purpose. If you plant 2,160 trees per mu, thirty mu will yield 64,800 trees, which, at a sale price of eight coins per tree, generates an income of 518,400 coins. In addition, since the cut branches from 100 trees will fill one cart with firewood, you can obtain 648 carts of firewood. One cart of firewood sells for 100 coins, so the income from firewood would be 64,800 coins, for a total income of 583,200 coins. If you plant thirty mu per year, or ninety mu over the course of three years, by selling thirty mu of timber in each year, you can ensure a lifetime free from distress.31

As noted earlier, the standard allotment of mulberry fields under the equalfield system entailed that the household was obliged to plant fifty mulberry trees, five jujube trees, and three elms, but in areas where the land was not 29 30 31

Chapter 52, “Zhong honglanhua zhizi” 種紅藍花梔子, in QMYS, 2, pp. 330–1. Chapter 52, “Huai liu qiu zi wu zuo” 槐柳楸梓梧柞, in QMYS, 2, pp. 318–19. Chapter 52, “Huai liu qiu zi wu zuo” 槐柳楸梓梧柞, in QMYS, 2, p. 317.

86

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

suitable for mulberry cultivation the profit-making strategy for silviculture outlined by Jia was still practicable. In contrast to the Warring States philosophers, who celebrated cereal cultivation as a means of bringing moral improvement to the people, in the Northern Wei era such intentions are absent from the thinking of local officials like Jia Sixie. Instead, they assumed that diverse forms of agriculture were necessary to enhance the material livelihood of the people. Furthermore, the climatic cooling that had characterized the early centuries of the Period of Disunion gave way to a gradual warming trend in the fifth to sixth centuries. We can perhaps recognize the Xianbei-ruled Northern Wei dynasty, which promoted agricultural policies such as the equal-field system and during which officials dedicated to agricultural development such as Jia Sixie emerged, as an environment conducive to the restoration of agriculture as the principal economic activity.

Sui–Tang Period: The Spread of Wheat, Rice, and Tea Cultivation The Sui (581–618) and Tang (618–907) dynasties, which reunified the Yellow River basin in the north and the Yangzi river basin in the south, were marked by a return to a mild, temperate environment. Yang Jian (founder of the Sui) and Li Yuan (founder of the Tang) had been powerful officials under the Northern Zhou as well as members of the hybrid Chinese–Toba aristocracy fostered by the Northern Wei. Both the Sui and Tang inherited many of the policies of the previous Northern Dynasties, such as the equal-field land tenure system, albeit modified in significant ways. Gaining the territories south of the Yangzi river brought not only rich rice-growing regions into the national economy but also vast areas of remaining natural forest. The regulations of the early equal-field system mandating tree planting disappeared. In addition, the provisions for allocating cultivated lands based on possession of slaves, oxen, and horses also were eliminated. The Tang version of the equal-field system hewed more closely to the goal of creating a society composed of small farming families with almost equal economic resources that had been idealized by many philosophers since antiquity. It was this Tang dynasty version of the equal-field system that was transmitted to the nascent states of the Korean peninsula and the Japanese archipelago. However, despite the warming climatic trend, rice cultivation did not reappear in the Yellow River valley. Since the Sui–Tang empires inherited the multinational society of the Period of Disunion, many peoples from Central Asia such as the Sogdians now dwelled in the Guanzhong region, 87

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

including the capital, Chang’an. These peoples had a fondness for foods made from wheat, a taste that gradually spread among the Chinese as well. Although rivers in the vicinity of Chang’an had been utilized for irrigated agriculture in the past, water began to figure more importantly as a power source for mills to grind wheat flour than as a resource for irrigation. Water mills were primarily operated by powerful groups such as members of the imperial clan and aristocrats, and farmers in the capital region loudly protested that the mills diminished the volume of water in waterways and caused shortages in the water available for irrigating arable lands. Even though many officials also voiced protests, they had negligible effect. During the Period of Disunion the Yangzi delta, long a sparsely populated frontier, became the economic heartland of the Southern Dynasties. Much of the delta consisted of low-lying swamps, but the region was ideally suited for rice cultivation. Aristocratic families who accompanied the Jin court to the south quickly laid claim to vast expanses of wilderness, creating great landed estates. Yet these estates appear to have been divided into separate plots worked by individuals or small teams of farmhands (including bondservants, indentured laborers, and tenants) rather than under the direct supervision of the landowner. This pattern of small-scale cultivation was especially suited to the labor-intensive demands of wet-rice agriculture. During the Tang dynasty the large estates were steadily replaced by small family farms. Irrigated rice farming was far more labor-intensive than raising dryland crops. But the higher productivity of intensive rice cultivation, including continuous planting without fallowing, yielded at least five times as much food per unit of land as wheat or millet. Many advances in rice cultivation – such as improved tools, transplanting, irrigation techniques, deep tilling, and frequent weeding – were adopted over the course of the Tang dynasty, although their use was mostly confined to upland valleys before the massive settlement and reclamation of the lowland plains during the Song dynasty. As a result of the warming climate and the introduction of Champa rice – which had a shorter growing season – from Southeast Asia, double-cropping of summer rice with winter crops (wheat or legumes) appeared in some areas. The intensification of rice agriculture produced substantial increases in yields. According to Li Bozhong, by the late Tang grain output per mu in the Yangzi delta had increased fourfold compared to the Period of Disunion.32 32

Li Bozhong 李伯重, Tangdai Jiangnan nongye de fazhan 唐代江南农业的发展 (Beijing, Nongye chubanshe, 1990), p. 212, Table 6.4.

88

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact

The humid climate of the Yangzi delta also was ideal for raising silkworms. Since the spring planting season for rice coincided with the peak season for highly labor-intensive sericulture tasks, this combination also encouraged a gendered division of labor, with women performing almost all of the sericulture work. The warming trend induced changes in the kinds of crops that could be cultivated, and also contributed to the proliferation of tea cultivation. Tea plants flourished in the stony, well-drained soils of the hilly interior regions of the Yangzi river basin. In his Classic of Tea, composed around 760, the scholar Lu Yu enumerated China’s main tea-producing areas, beginning with Sichuan and then gradually extending southward, encompassing the entire region south of the Yellow River.33 Originally drinking tea was a leisurely pastime of the elite, but over the course of the Tang dynasty tea became a staple for all social classes. As the popularity of tea drinking rose, tea cultivation spearheaded settlement of the interior regions of the Yangzi river basin.

Conclusion The history of agriculture in China is characterized above all by the primacy of cereal cultivation and the steady replacement of grasslands, marshes, and forests with cultivated fields. This transformation of the natural landscape also caused the displacement of animal husbandry, only temporarily reversed during the Period of Disunion as a result of the influx of steppe nomads and the onset of a colder climate. Clearance of forests and marshes radically reduced the abundance of game, and of course led to the loss of timber and fuel, which were in critically short supply in north China already by the Tang dynasty. Hunting wild fauna – a crucial component of elite social and ritual life in the Bronze Age as well as a supplementary food source for the population more generally – sharply diminished as well. The elimination of alternative food sources intensified the reliance on cereal cultivation and constricted a diet notably poor in protein (only partially offset by legumes such as soybeans). The spread of farming induced other environmental stresses as well. The alkalization and desiccation of the loessic soils of north China compromised soil fertility, and were exacerbated rather than mitigated by artificial irrigation. The crucial problem of maintaining soil fertility prompted 33

Lu Yu, The Classic of Tea (South San Francisco, Long River Press, 2019), pp. 83–9.

89

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

motoko hara

experimentation and development of novel cultivation techniques such as ridge-and-furrow planting and “alternate-field” fallowing. The development of large-scale irrigation works through state investment and management beginning in the Warring States period boosted agricultural production and enabled the spread of rice cultivation in north China, although rice farming disappeared in the north following the onset of a colder climate from the first century C E. The advance of agriculture was powerfully influenced by such longterm oscillations in climate. Expansion of cultivation occurred during periods of climatic warming: in the early Bronze Age, during the era of the Warring States and the first unified empires, and again from the late sixth century C E – in each case, it should be noted, coinciding with trends toward political consolidation. State formation, and especially the rise of centralized bureaucratic government in the Warring States and Qin–Han empires, also was a driving force in the expansion of agriculture. These states aggressively promoted the primacy of cereal cultivation, which exerted a powerful influence on the organization of labor and fiscal systems. Given its vested interest in maximizing grain production, the state also assumed a prominent role in investment in water control and irrigation works and in disseminating innovations in cultivation techniques and improved tools. Market demand also spurred agricultural production, and promoted cultivation of more diverse crops for both food consumption and industrial uses. The human impact on soil, water, vegetation, and animal life had already reached enormous proportions in north China by the advent of the imperial age. In subsequent centuries, Chinese settlers and their farming practices gradually spread to south China, especially in the Yangzi river valley and delta. After the year 1000, the development of wet-rice cultivation and the introduction of major new crops – tea and sugar in the Song era, and New World crops such as maize, sweet potato, and tobacco from the sixteenth century – would profoundly alter the environment and the ecological balance between humanity and nature in south China as well.

Further Reading Amano Motonosuke 天野元之助, Chu¯goku no¯gyo¯shi kenkyu¯ 中国農業史研究, revised ed. (Tokyo, Ochanomizu shobo¯, 1979). Bray, Francesca, “Agricultural Technology and Agrarian Change in Han China,” Early China 5 (1979–1980), 3–13.

90

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

Agriculture and Its Environmental Impact Bray, Francesca, Science and Civilisation in China, vol. 6, Biology and Biological Technology, part 2, Agriculture (ed. Joseph Needham) (Cambridge, Cambridge University Press, 1984). Elvin, Mark, The Retreat of the Elephants: An Environmental History of China (New Haven, Yale University Press, 2004). Hara Motoko 原宗子, Kankyo¯ kara toku kodai Chu¯goku 環境から解く古代中国 (Tokyo, Taishu¯kan, 2009). Hara Motoko 原宗子, Kodai Chu¯goku no kaihatsu to kankyo¯ 古代中国の開発と環境, vol. 1 “Kanshi” chiinhen kenkyu¯ 《管子》地員篇研究, vol. 2 “No¯hon” shugi to “ko¯do” no hassei 《農本》主義と《黄土》の発生 (Tokyo, Kenbun shuppan, 1994, 2005). Hsu, Cho-yun, Han Agriculture: The Formation of Early Chinese Agrarian Economy (206 B . C .– A . D . 220) (Seattle, University of Washington Press, 1980). Kidder, Tristram R., Haiwang Liu, and Minglin Li, “Sanyangzhuang: Early Farming and a Han Settlement Preserved beneath Yellow River Flood Deposits,” Antiquity 86.331 (2012), 30–47. Li Bozhong 李伯重, Tangdai Jiangnan nongye de fazhan 唐代江南农业的发展 (Beijing, Nongye chubanshe, 1990). Liu, Li, and Xingcan Chen, The Archaeology of China from the Late Paleolithic to the Early Bronze Age (Cambridge, Cambridge University Press, 2012). Liu Xinglin 刘兴林, Xian Qin Liang Han nongye yu xiangcun juluo de kaoguxue yanjiu 先秦两 汉农业与乡村聚落的考古学研究 (Beijing, Wenwu chubanshe, 2017). Nishijima Sadao 西嶋定生, Chu¯goku keizai shi kenkyu¯ 中国経済史研究 (Tokyo, To¯kyo¯ daigaku shuppankai, 1966). Okamura Hidenori 岡村秀典, Chu¯goku kodai ¯oken to saiji 中国古代王権と祭祀 (Tokyo, Gakuseisha, 2005). O¯sawa Masaaki 大澤正昭, To¯ So¯ henkakuki no¯gyo¯ shakai shi kenkyu¯ 唐宋変革期農業社会 史研究 (Tokyo: Kyu¯ko shoin, 1996). Shih, Shen-han, On “Fan Sheng-chih shu”: An Agriculturist Book of China (Beijing, Science Press, 1959). Shih, Shen-han, A Preliminary Survey of the Book Chʻi min yao shu: An Agricultural Encyclopaedia of the 6th Century (Beijing, Science Press, 1959). Yoneda Kenjiro¯ 米田賢次郎, Chu¯goku kodai no¯gyo¯ gijutsu shi kenkyu¯ 中國古代農業技術 史研究 (Tokyo, Do¯ho¯sha, 1989).

91

https://doi.org/10.1017/9781108587334.003 Published online by Cambridge University Press

3

State and Economy Production, Extraction, and Distribution richard von glahn

The integrity and durability of the Chinese empire over two millennia rested on the strength of its fiscal capacity. From antiquity, sovereignty was linked to the ruler’s duty to provide for the economic as well as moral welfare of his subjects. Rulers of the Bronze Age manifested and reproduced their authority through distribution of wealth among their kinsmen and noble allies. The autocratic monarchs who rose to dominance after 500 B C E amassed wealth to buttress their military prowess, but they also strove to protect the independence and livelihood of the smallholder family farmers who undergirded their economic prosperity. The institutional apparatus of the fiscal state – centralized planning of taxation and expenditure to satisfy the state’s commitments to good governance – became a defining feature of the first empires, beginning with the Qin dynasty (221–206 B C E). Yet at the same time the rise of a market economy also shaped the relationship between the state and its subjects. Even the command economy instituted by the Qin Empire made accommodations to markets, merchants, and money. During the early imperial era, down to the end of the Tang dynasty (618– 907), a succession of dynasties sought to maximize state control over economic life in pursuit of the twin goals of enhancing state power and securing the welfare of imperial subjects. But these endeavors rarely fulfilled the ambitions of their authors. The transition from the Tang to the Song dynasty (960–1279) marked a crucial turning point in the course of imperial China’s fiscal history. The expansion of the market economy during the Tang–Song transition irrevocably transformed the relationship between the state and its subjects. Subsequently the direct role of the state in the production and distribution of goods diminished. Yet by adopting market mechanisms to

92

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

extract revenues, manage exchange, and invest in production, Chinese rulers and statesmen of the late imperial era in some ways augmented the state’s fiscal capacity and its power to mold economic livelihood.

The Redistributive Economy of the Patrimonial State During its Bronze Age, autarkic domain economies prevailed across China. From the time of the founding of the Zhou dynasty c. 1045 B C E its kings awarded benefices in the form of territory and servile populations to regional rulers – known as “archer lords” (hou 侯) – spread across the former Shang (c. 1550–1045 B C E) realm on north China’s Central Plain. Drawing upon the resources of the royal domain in the Zhou ancestral heartland of the Wei river valley, the Zhou kings made similar dispensations of benefices to royal officers, a practice that steadily reduced its own economic base. A fledgling bureaucracy was created, with the kings appointing ministers for revenue collection, public works, and military affairs and hunting. A large staff of scribes and secretaries – many of them persons of exalted status with close ties to the king – issued royal orders, composed speeches for the king on ceremonial occasions, preserved government records, and acted as his personal emissaries. Little is known, however, about the royal government’s extraction of revenue and tribute. By the beginning of the ninth century B C E – precisely when the Zhou kings were beginning to suffer the constraints of diminished resources – the royal household emerged as a distinct institution within Zhou government. Under the direction of one or more stewards (zai 宰), the royal household managed the ritual, political, and economic affairs of the king and his consorts, who had their own properties and revenues. A staff of ministers and secretaries assisted the stewards in the supervision of the lands, industrial workshops, and ceremonial centers directly attached to the royal household. The archer lords and to a lesser degree the lineages of royal officers who were granted benefices within the royal domain gradually gained independence from the king. In principle, the Zhou state can be defined as a patrimonial form of government in which the monarch shared his sovereign authority with noble lineages established by royal investiture and linked to the royal house through real or fictive kinship ties. Induction into this ruling elite was sanctified by ritual practices centered on ancestor worship and maintained through patrilineal descent and bestowal of ritually prestigious goods. The internal organization of noble lineages and their 93

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

management of wealth, property, and productive assets such as industrial workshops probably emulated, on a smaller scale, the operation of the royal household. Bronze foundries ranked among the most important industrial enterprises of the Bronze Age, and the artisans and overseers who worked in these foundries enjoyed a higher standard of living than the multitude of servile commoners. Gifts of bronze metal ranked as the most highly prized royal prestation. Many of the bronze ritual vessels that were central to the religious and political culture of the Zhou patrimonial state bear inscriptions commemorating the royal gift of the bronze metal from which the vessel was manufactured. The distinctively superior quality of bronze vessels produced by royal workshops attests to the much greater concentration of resources, technical skills, and craftsmanship at the capital compared to the workshops operated by noble houses. Land and agriculture comprised the material base of the Western Zhou economy. In addition to drawing wealth from land revenue, royal prestations, and craft manufacture, noble houses also engaged in exchange, chiefly involving prestige goods but also land. Gift exchanges among noble lineages were governed by exacting ritual procedures and subject to the king’s affirmation and juridical authority. Bronze inscriptions record a small number of transactions involving the exchange of land (and the servile population who worked it) for ritual paraphernalia such as jades, furs, hides, vestments, and chariots. Disputes over claims to land were adjudicated by royally appointed officials, and resolution of these disputes often involved complex ritual protocols such as gift exchanges and feasting. In only very rare cases do we seem to have evidence of a conscious effort to accumulate landholdings. But such cases suggest an incipient movement from a redistributive economy in which wealth circulated through royal largesse and gift giving to a tenurial economy based on rights of permanent possession and exchange validated through contract.

The Domain Economy of the Spring and Autumn City-States In the Western Zhou, the king, the archer lords, and royal officers all drew revenues from lands which they possessed as the common patrimony of their lineage. Like markets, the concept of taxation did not yet exist. The Zhou patrimonial state can be categorized as a domain-state in which the wealth and income of the ruling class derived from personal and familial rights to land and labor. Following the invasion and occupation of the Zhou ancestral 94

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

homeland in 771 B C E by hostile forces, the royal court fled to the east, reestablishing itself at Cheng Zhou (modern Luoyang) on the western edge of the Central Plain. From the onset of the Eastern Zhou period (771–256 B C E) the effective authority of the Zhou kings rapidly faded. Not only was the territory under direct royal control vastly reduced, but over time the kinship bonds that knit together the patrimonial state frayed as genealogical connections became more distant. Social and economic differentiation within aristocratic lineages also weakened kinship solidarity. Although still revered in ritual terms as the sovereign Sons of Heaven, the Zhou kings became politically impotent, and military and economic power devolved to the archer lords, now rulers of their own independent states. During the Spring and Autumn era (conventionally dated to 772–481 B C E), the Zhou polity was fragmented among hundreds of autonomous agrarian city-states, comprising a capital city – the seat of the ruling lineages – and adjoining rural settlements whose inhabitants worked the land under servile conditions. The city-state polities retained the basic features of a domain-state: the ruler – often designated as “lord” (gong 公) – and his subordinates derived their income from personal or lineage rights to the fruits of land and labor. Government affairs usually were entrusted to appointed officials. Ministerial responsibilities were divided among revenue, public works, and military affairs – it became standard practice in the Spring and Autumn period to appoint judicial ministers as well – under the supervision of a prime minister or chancellor. Officials received emoluments in the form of benefices (caiyi 采邑), usually measured in numbers of settlements, and thus obtained their own independent base of territory, subjects, and resources. Noble lineages typically resided at the capital, where they comprised a citizenry (guoren 國人) defined above all in terms of the obligation of military service to the ruler. Over time the citizenry expanded with the extension of liability for military service to include some commoners, but as a rule merchants, artisans, and farmers remained excluded. The capital citizenry became a political force in the Spring and Autumn city-states. In times of crisis, such as military emergencies or succession disputes, rulers or ministers summoned assemblies of citizens to rally support for their actions. The citizenry often played a decisive role in partisan conflicts between the ruler and noble families, or between rival political factions. On occasion, collective action by the citizenry sufficed to depose a tyrannical ruler. The city-state polities proved unstable. The civil strife that afflicted the citystates vitiated the power of hereditary rulers and noble lineages alike – and 95

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

ultimately subverted the citizenry as well. Bureaucratic rank increasingly superseded aristocratic pedigree. In many cases, powerful ministers eclipsed their sovereigns in the conduct of government and diplomacy. Conflict among rival ministers and factions often toppled hereditary rulers or reduced them, like the Zhou kings, to mere figureheads. In Lu, for example, a bloody succession dispute broke out after the death of Lord Huan in 662 B C E that resulted in an abiding rivalry for hegemony among three lineages of Huan’s sons, each with its own army. In Jin, after the assassination of Lord Ling in 607 B C E, conflicts among generals and ministers led to a quasi-partitioning of this, the largest and most powerful state of the time, among six ministerial houses. Endemic conflict among the six houses culminated in the extinction of three of them and the formal division of the Jin territory among the three others in 453 B C E. In Qi – ruled, like Lu and Jin, by an august house closely tied to the Zhou royal lineage – four lords were assassinated by their ministers between 553 and 481 B C E; on the last occasion, the prime minister usurped control of the government, and in 386 B C E his descendants formally displaced Qi’s hereditary lords as the ruling lineage. At the same time, incessant warfare among the rival lords resulted in annihilation of weaker polities and their absorption into expanding territorial states, strengthening the power and resources at the ruler’s command. By the sixth century B C E the patrimonial order had mostly collapsed, giving way to a new political order founded on the centralization of power in the hands of an all-powerful monarch. Fewer than two dozen states survived into the Warring States era (453–221 B C E), of which seven formidable macro-states emerged as contenders for supremacy.

The Rise of the Autocratic State in the Warring States Period The tumultuous warfare of the Spring and Autumn era compelled rulers to seek to centralize military power and control over economic resources in their own hands. Such efforts provoked staunch resistance on the part of the noble lineages, and in many cases the nobility prevailed over their lord. But amid the increasingly intense and larger-scale warfare of the Warring States era, those states where the ruler succeeded in centralizing power gained the advantage. Autocratic monarchies abrogated the privileges of the nobility by asserting direct control over lands and people, revoking hereditary rights to office in favor of appointing trusted and capable subordinates, and mobilizing new infantry armies conscripted from the populace at large and equipped 96

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

with iron weapons – iron metallurgy itself being a transformative technology of this age – in place of the chariot warfare conducted by noble warriors. The growing size of the macro-states required more sophisticated forms of administrative control, communications, and logistics. Rulers recruited professional officeholders (generally designated as shi 士) – still drawn from noble lineages but distinguished by their schooling in the arts of government and experience in practical administration rather than family pedigree – to staff increasingly bureaucratized governments. Hoary principles of warrior fealty and military service were expanded into a generalized notion of taxation imposed on the whole populace. Taxation extended to levies on the produce of the land, but military service remained the core of the subject’s duty to the monarch. The earliest efforts to broaden the basis of taxation and military service – in Lu in the 590s B C E , and the reforms launched by the renowned Zheng statesman Zichan in 543 B C E – appeared in weaker states besieged by powerful adversaries. In both Lu and Zheng officials conducted land surveys and organized the rural population into units liable for military service, although we lack details on how land registration was related to military conscription. The southern state of Chu, the first to field infantry armies in place of chariots, also pioneered the development of bureaucratic tools to mobilize resources to augment its military might. Chu had emerged as a powerful state before its induction into the Zhou ecumene in the mid-seventh century B C E, and its rulers – who had broken with Zhou protocols in arrogating to themselves the title of “king” – exerted unusually firm control over their ministers and noble lineages. Chu conducted the first universal registration of able-bodied adult men for military service in 589 B C E . In 548 B C E, the newly installed chief minister of Chu, Qu Jian, ordered the minister of war to conduct a comprehensive review of state revenues and military strength that entailed recording arable and non-arable lands, measuring the mountains and forests, marshaling the wetlands and marshes, distinguishing higher and lower ground, marking out salt ponds, defining the boundaries of areas prone to flooding, cataloguing reservoirs, dividing lands into small parcels separated by footpaths, setting out pastures for livestock, allocating fertile farmlands, calculating revenues to determine the rates of taxation (liangru xiufu 量入修賦), registering carts and horses, and compiling inventories of the numbers of charioteers, infantry, weapons, and armor.1 1

Xianggong 25.11, in ZZ, 2, pp. 1985b–1986c.

97

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

The salient role of the minister of war, not only in gathering information on military resources and logistics but also in compiling records of economic assets and in setting rates of taxation, reveals the intertwining of fiscal practice and military strategy. Efforts by rulers to lay claim to nonagricultural resources – often expressed as “the profits of the mountains and the marshes” (shanze zhi li 山澤之利) – incited strong opposition from the nobility and commoners alike. On one hand, wilderness lands encompassing hills, forests, wetlands, and meadows were the hunting grounds of the nobility, for whom hunting was a ritualized form of bloodletting meant to affirm their exalted social status through display of their awesome power over life and death. On the other hand, the populace at large depended on wilderness lands to gather game, wild vegetables, firewood, and construction materials to meet their subsistence needs. But rulers regarded the “mountains and marshes” not as public resources but as part of their personal domain, over which they exercised exclusive rights. Such rights also included mineral resources such as iron, copper, and salt, prized as both lucrative revenue sources and vital strategic assets. Throughout the Warring States era, regarded as the “axial age” of Chinese political philosophy, censure of the increasingly autocratic monarchs of the day often focused on the twin issues of their monopolistic claims over “the profits of the mountains and the marshes” and their imposition of excise levies on cross-border commerce. Conversely, those rulers who allowed their subjects access (regulated, of course) to the “mountains and marshes” and lowered customs levies were showered with acclaim.2 As Qu Jian’s decree of 548 B C E suggests, maximizing the state’s economic and military power required careful calculation and documentation of resources. Although population registration was initiated for the purpose of military conscription, it also served as the basis for allocating landholdings at the disposal of the state. The details of early registration systems remain murky prior to the reform program introduced in 356 B C E by Shang Yang, a Wei official recruited by Lord Xiao of Qin (r. 361–338 B C E) to take charge of his government. Lord Xiao’s father had already instituted a census in 375 B C E for conscription of soldiers and laborers. Shang Yang merged the existing Qin household registration with a steeply graded roster of military ranks (seventeen in all) determined by meritorious accomplishment in warfare that entitled individuals to specific legal and economic prerogatives, including 2

Masubuchi Tatsuo 増淵龍夫, “Sen–Shin jidai no sanrin so¯taku to Shin no ko¯den” 先秦 時代の山林藪澤と秦の公田, in Masubuchi, Chu¯goku kodai no shakai to kokka 中國古 代の社會と國家 (Tokyo, Ko¯bundo¯, 1960), pp. 265–327.

98

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

tenurial rights. The Han historian Sima Qian described Shang Yang’s reform as “designating lands and dwellings (mingtianzhai 名田宅) according to the individual’s military service and allowing them to possess slaves and clothing according to family rank.”3 The mingtianzhai land allocations became the foundation of a new social order that established the conjugal household and family farm as the fundamental socioeconomic unit of the Qin state. Shang Yang also abolished the practice of primogeniture, mandating equal inheritance among sons, and made it a crime for fathers and adult sons to live together. The practice of organizing groups of five families into “squads” (wu 伍) who shared mutual legal responsibility for criminal actions of any of their members was said to have been enacted at the time of the census of 375 B C E. Recently excavated household registers and legal statutes from the Qin and early Han empires indicate that Shang Yang’s draconian reforms were not fully implemented, but the principle of equal inheritance subsequently became an enduring feature of Chinese law and social practice. To enact the mingtianzhai allocations, Shang Yang’s government conducted land surveys that divided arable lands into standardized parcels. A fragment of the Qin land statutes dating from 309 B C E stipulates that the land was to be divided into parcels one pace wide and 240 paces long. Two of these parcels constituted a mu unit, and 100 mu (or one qing) became the standard allotment that the Qin state provided to farming households. In this respect Qin followed precedents established in other states, perhaps tracing back as far as the tenurial system introduced by Zichan in Zheng nearly two centuries before. In return for these land allocations the head of the household was expected to submit payments of grain and fodder as well as render labor and military service. Thus civil/military registration laid the foundation for imposing land taxes, which became a crucial source of state income. The extent to which Shang Yang was able to seize lands from noble lineages and reallocate them to ordinary farming households remains unclear. Such reallocations most likely were carried out on newly opened lands and in conquered areas where the local population was forcibly displaced by Qin colonists. Shang Yang’s fiscal reforms in Qin probably differed more in degree than in kind from policies that had been widely enacted across the Zhou ecumene. But in one respect the Qin state clearly distinguished itself from its contemporaries: its single-minded focus on land as the source of all wealth and its abiding hostility toward commerce. The Book of Lord Shang, purported to be 3

SJ, 78, p. 2230.

99

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

a digest of Shang Yang’s economic philosophy, is distinguished from other philosophical works of the late Warring States era by its uncompromising anti-mercantile posture (see von Glahn, Chapter 5 in this volume). The origins of market trade and a nascent merchant class have left only faint traces in the documentary record, but archaeological research conclusively demonstrates a rapid proliferation of bronze currencies among the Central Plain states by the late sixth century B C E. The earliest bronze currencies, which took the form of miniature spades and knives, appear to have been issued by local authorities or private merchants. Current scholarly opinion holds that the knife-shaped currencies that predominated in Yan and Qi (see Map 4.1) originated as a medium of exchange between the Yan population and the non-Chinese Di peoples inhabiting the northeastern frontier of the Zhou ecumene. The spade currencies of the Central Plain initially bore the names of states such as Zheng, Song, Wey, and the Zhou royal domain. By the fifth century B C E, however, a multitude of bronze currencies issued by local cities were in common circulation across state borders. The Zhao city of Jinyang (near modern Taiyuan, Shanxi), for example, issued not only the spade currencies typical of the Central Plain but also knife and round coin currencies. Merchants and other townsmen appear to have accrued sizable wealth in the core states of the Central Plain such as Wei, Hann, and Zhao (the three independent states that emerged from the dissolution of the oncemighty Jin state in 453 B C E), and may have exercised political influence as well. The disputes over the levying of customs excises at state borders mentioned above attest to both the expansion of long-distance trade and the growing importance of commerce as a source of revenue. An anecdote about the aforementioned Zhao city of Jinyang recorded in the Han Fei Zi, a late third-century B C E Legalist tract, provides indirect evidence of the pro-mercantile ethos that emerged in the states of the Central Plain heartland. Shortly after its founding as an independent state Zhao was threatened by a power-mongering rival. When the Zhao ruler, Viscount Xiang (r. 458–425 B C E), asked his adviser Zhang Mengtan which city would serve as the most formidable refuge in the face of a long siege, Zhang recommended Jinyang. Upon arriving at Jinyang, however, Viscount Xiang found the city walls in disrepair and the granaries and treasuries empty. Alarmed, he demanded an explanation from Zhang Mengtan. Zhang answered that “the principle of sagely government is to store wealth among the people, not in the granaries and treasuries.” Since the people of Jinyang possessed ample supplies of grain and money and had not been burdened by demands for labor service, the prince need only issue a plea 100

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

for aid. “Overnight the granaries were filled with grain, the treasuries overflowed with coin, and the arsenals were fully stocked with weapons; within five days the walls were in perfect repair.”4 The adage that “the ruler does not amass goods in his treasuries, but rather stores them among the people” appears in the Guan Zi, an eclectic contemporary treatise on political economy (see von Glahn, Chapter 5 in this volume), and in later Chinese thought denoted a political economy premised on enriching both state and society by minimizing state extraction and allowing the rich to retain their wealth (see Dunstan’s chapter in this volume). Zhang Mengtan’s strategy succeeded in repulsing the attack; the power-monger was killed, and his own state destroyed. For Hanfeizi, this anecdote was a cautionary admonition against “greed and fondness for profit” on the part of rulers. But Zhang’s thinking also can be read as an expression of the salient prominence of private wealth and a prosperous merchant elite in Jinyang and other leading commercial cities. In contrast to the decentralized polities of the Central Plain, rulers of peripheral states such as Yan, Qi, Chu, and Qin managed to maintain greater control over industrial enterprise, markets, and the exchange of goods. Both the knife currencies of Qi and the round banliang coins (with a square center hole) inaugurated by Qin in 336 B C E (see Kakinuma’s chapter in this volume) were highly standardized, and neither circulated much beyond the state’s borders. Nor did foreign coins circulate within their territories, given their rarity in archaeological excavations. Industrial workshops such as iron and bronze foundries and coin mints were concentrated in major cities, and governments everywhere closely supervised the work of artisans in strategic industries such as weapons manufacture. Armories, too, were a ubiquitous feature of the urban landscape of capital cities. Yet there is ample evidence of private industrial enterprises and public marketplaces even in the capitals of the most autocratic states, such as Qin and Qi. By the late Warring States period, two distinctive models of political economy and state building had emerged.5 In the thickly settled heartland of the Central Plain, central government authority weakened after the division of Jin into the three separate states of Wei, Hann, and Zhao in 453 B C E . In this region local cities and their merchant and artisan classes enjoyed considerable autonomy from their royal overlords. Private entrepreneurship 4 5

Chapter 10, “Shiguo” 十過, in HFZ, 1, pp. 177–80. Emura Haruki 江村治樹, Sengoku Shin Kan jidai no toshi to kokka: Ko¯kogaku to bunken shigaku kara no appuroochi 戦国秦漢時代の都市と国家：考古学と文献史学から のアップローチ (Tokyo, Hakuteisha, 2005).

101

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

stimulated industrial and commercial expansion, which in turn generated great private fortunes. In the peripheral states of Qin, Chu, Qi, and Yan, by contrast, autocratic rulers established bureaucratic institutions to control economic resources. In these states industrial production was concentrated in state-managed workshops and trade was subject to tight official regulation. Private trade was by no means suppressed in these states: Chu in particular engaged in cross-border trade with other states, and Linzi, the capital of Qi, was said to be the richest and most populous city of all. The latter pattern of strong state control of the economy was most fully developed in Qin after Shang Yang’s reforms, and this model would prevail after Qin created a unified empire in 221 B C E.

The Command Economy of the Qin Empire Shang Yang’s reforms recast the Qin state – long peripheral to the traditional seats of political hegemony in the Zhou ecumene – into a formidable and aggressive power, bent on conquest and marshaling men and matériel for war. The conquest of Shu (in modern Sichuan) in 316 B C E launched the incipient phase of Qin empire building, which accelerated with the capture of the Chu capital of Ying (in modern Hubei) in 278 B C E. Apart from a crucial defeat inflicted on Zhao in 260 B C E, the conquest of the major Warring States polities and the unification of the old Zhou ecumene under Qin imperial rule was accomplished in a series of lightning advances between 231 and 221 B C E. In very short order the Qin state had to adapt the state-building policies and institutions bequeathed by Shang Yang to the daunting task of uniting an empire over vast and far more populous territories. Rapid imperial expansion seriously strained the capacities of the Qin command economy. Transmitted historical records yield little information about the operation of the Qin government and its economic and fiscal institutions. Even the basic organization of the Qin central government remains poorly understood. But the progress of archaeological research and the recovery of documents excavated from tombs over the past several decades has greatly enriched our understanding of the workings of Qin government at the local level. Among the excavated manuscripts most important to the study of Qin political economy are the Shuihudi documents, which include a corpus of administrative statutes placed in the tomb of a Qin local official who died c. 217 B C E , and the Liye archive, which contains tens of thousands of records pertaining to routine administration in the small county of Qianling in Dongting commandery (western Hunan) from the decade following the 102

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

unification of the empire in 221 B C E. Both of these caches of manuscripts relate to local governance in former Chu territories in southwestern China, and Qianling was in a frontier zone largely populated by non-Chinese. Therefore caution must be exercised in drawing from these materials general conclusions about governance in the Qin homeland and long-settled regions. Apart from the Shuihudi and Liye manuscripts, various smaller collections of documents from unprovenanced sites have been acquired by Chinese academic institutions through the antiquities market. Shang Yang’s reorganization of the Qin government transformed the office of chamberlain (neishi 內史) into that of an exchequer with broad jurisdiction over state finances, second in authority only to the chancellor (xiang 相), who managed diplomacy and military affairs. The rapid territorial expansion of the empire in the mid-third century B C E presented daunting logistical challenges, however. Consequently, the anti-mercantile spirit inculcated by Shang Yang underwent modification, especially after Lü Buwei, a wealthy merchant from the prosperous commercial town of Yangdi in Wey, came to prominence at the Qin court. Lü insinuated himself into the entourage of a Qin prince whose son would become the future King Zheng (r. 247–210 B C E), and later the First Emperor of China. Lü had already ascended to the office of chancellor in 250 B C E before cementing his command of the Qin government upon the accession of the thirteen-year-old King Zheng three years later. Lü patronized merchants and recruited many of them – often in return for munificent donations to the state’s coffers – to serve in his administration. He also established an institution known as the Lesser Treasury (shaofu 少府) to supervise the government workshops that supplied luxury goods to the king and his palace retinue. Although private industry and commerce enjoyed some measure of encouragement under Lü’s regime, the state exercised direct control over most productive assets – including foundries, workshops, mines, forests, and pastures – and also a substantial portion of the agricultural economy through state-run farms. The Qin bureaucracy correspondingly expanded to accommodate these managerial functions. The treatise on statecraft attributed to Lü Buwei enumerated twenty departments believed necessary for the government to carry out its duties: in addition to five departments devoted to calendrical sciences and prognostication, others would assume responsibilities over slaves, clothing, maps, bows, chariots, carts, boats, oxen, palaces, wine, wells, markets, mortars (for grinding grain), physicians, and shamans.6 6

Book 17, “Wugong” 勿躬, in LSCQ, 2, pp. 1077–8.

103

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

After King Zheng took personal control of his government, however, Lü quickly lost favor and was compelled to commit suicide in 237 B C E. Around 227 B C E King Zheng and Lü’s successor as chancellor, Li Si, restructured the Qin central government, replacing the chamberlain with two new offices: the imperial secretary (yushi dafu 御史大夫), who headed the state bureaucracy and in effect served as vice chancellor, and the exchequer of grain tribute (zhisu neishi 治粟內史), who took charge of grain revenues and state-owned farms. The jurisdiction of the Lesser Treasury – under the direct control of Li Si – expanded to include comprehensive authority over fiscal administration and tax collection. The Lesser Treasury became a sprawling complex of treasuries, workshops, and storehouses that collected revenues in gold and coin; managed the state’s parks, pastures, and forests; and supervised public works. In Han times the Lesser Treasury was explicitly designated a Privy Purse responsible for the maintenance of the emperor and the imperial household. Although the Lesser Treasury certainly took charge of the Qin emperor’s personal revenues (for example, income from customs), no clear distinction can be observed between general state revenues and the emperor’s personal domain under the Qin Empire. The Liye documents yield abundant information on revenues, expenditures, and economic management at the local level. In addition to military and judicial offices, Qianling’s magistrate supervised at least four bureaux (cao 曹) dedicated to economic matters: Households, Finance (“Gold and Cloth of the Court,” ting jinbu cao 廷金布曹), Granaries, and Public Works. In addition, Qianling had a treasury for public funds and other sundry assets (such as textiles, livestock, and silkworm cocoons) known as the Lesser Revenue Office (shaonei 少內), which presumably was subordinated to the Greater Revenue Office (danei 大內) at the capital. Neither of these terms appears in the transmitted documentary record, and their functions in relation to other fiscal agencies remain unspecified in the Liye documents. What is clear is that the Qianling county government supervised the production of agricultural and manufactured goods for its own use, for transfer to other government agencies, and for sale on the market. State farms (gongtian 公田), orchards, and pastures worked by convict laborers yielded grain, hemp, chestnuts, bamboo, and livestock, while artisanal workshops (likewise relying on convict labor) produced various types of cloth, lacquer, arrows, sandals, rope, bricks, tiles, and animal hides. Management of this panoply of economic activities necessitated an elaborate system of record keeping, information verification, and communications among numerous units of local government within Qianling and beyond. The Liye documents 104

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

fully testify to the Qin state as, in Robin Yates’s words, an “empire of the scribes.”7 Among the most prevalent documents from the Liye archive are “verification tallies” (jiaoquan 校券), composed by both public officials and private parties, that certified a wide range of economic exchanges, including revenue transfers; disbursements of funds; salary payments; contracts; wills; government loans of grain; equipment such as tools, carts, and oxen; and even convict laborers to private individuals.8 Such documents reveal that a large portion of the population was mired in perpetual indebtedness to the state. Failure to redeem these debts (often denoted as “pledges,” zhi 質) resulted in criminal punishments, calculated as monetary fines but typically discharged through forced labor, including military service. Such indebted persons – including even low-level officials – thus swelled the ranks of servile labor at the state’s disposal. Scholars have noted that under the Qin system of social status the distinction between “convict” and “government slave” often was blurred; both were commonly designated as lichenqie 隸臣妾. Conscription of the commoner population (those with a merit rank of five or lower were obligated to provide one month of labor service annually) also was an important source of labor. But local governments were instructed to first mobilize hard-labor convicts, then slaves (lichenqie), and next government debtors before resorting to general conscription to fulfill their labor needs. Moreover, Qianling local officials routinely purchased slaves (typically those with specialized skills) from private individuals as well as hiring out slaves to private employers. The meticulous record keeping needed to compile and audit the accounts of these various pools of labor (and to track individuals, who might be assigned to work in other jurisdictions), and to calculate the units of labor needed for specific projects – in addition to the logistical and communications infrastructure for circulating this information and optimizing the allocation of state’s resources – again attests to the enormous “scribal capacity” of the Qin state.9 7

8

9

Robin D.S. Yates, “The Empire of the Scribes,” in Yuri Pines, Lothar von Falkenhausen, Gideon Shewlach, and Robin D.S. Yates (eds.), Birth of an Empire: The State of Qin Revisited (Berkeley, University of California Press, 2014), pp. 141–53. Robin D.S. Yates, “The Economic Activities of a Qin Local Administration: Qianling County, Modern Liye, Hunan Province, 222–209 B . C . E.,” in Elisa Levi Sabattini and Christian Schwermann (eds.), Between Command and Market: Economic Thought and Practice in Early China (Leiden, Brill, forthcoming 2021). Maxim Korolkov, “Between Command and Market: Credit, Labour, and Accounting in the Qin Empire (221–207 B . C . E .),” in Sabattini and Schwermann, Between Command and Market; Maxim Korolkov, “Empire-Building and Market-Making at the Qin Frontier: Imperial Expansion and Economic Change, 221–207 B C E” (Ph.D. dissertation, Columbia University, 2019).

105

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

As a frontier outpost Qianling probably was unusual in its heavy reliance on unfree labor, estimated to have comprised one-third of the county’s labor force. But there is little doubt that labor service was the state’s principal form of extraction of value from the general population as well. The Qin land tax was not a fixed levy, but rather appears to have been adjusted annually depending on the state of the harvest; on average, the general rate was roughly 10 percent of the harvest, paid in kind (grain, beans, and hemp). The Qin government also resorted to ad hoc levies to meet urgent needs and extraordinary expenditures, but there is no evidence of the capitation taxes that became the fundamental feature of state taxation under the Han dynasty. The excavated documents contain desultory references to a “household levy” (hufu 戶賦) in the form of modest annual payments in coin, hay (which could be commuted to coin), and silkworm cocoons. Taxes and excises levied on merchants, shops, and trade constituted another source of government income. In-kind payments generally were retained at the county level, but money revenues were forwarded to commanderies or the central government, which often exacerbated liquidity problems for local administrators. Since the time of Shang Yang, as we have seen, the Qin state had allocated mingtianzhai tenurial rights on the basis of military rank. Whether land was alienable remains unclear. There is scarcely any evidence for private land transactions during the Warring States and imperial Qin periods. A Qin ordinance of 216 B C E required that households “self-verify” (zishi 自實) the landholdings in their possession, which some scholars regard as an acknowledgment of private tenurial rights. One Liye document dated to 212 B C E is an affidavit, endorsed by a government scribe, of the transfer (presumably through inheritance) to an adult daughter of property including slaves (seven in all), crops, clothing, and utensils – but not land.10 Other Liye documents likewise make no mention of transfers of lands, nor are landholdings recorded on the Liye household registers. The prominence of slaves in several Liye documents concerning inheritance further attests to the prevalence of convict and slave labor in Qianling and the fungibility of these statuses. Despite the harsh condemnation of merchants as a parasitical class expressed in the Book of Lord Shang and Han Fei Zi (see von Glahn, Chapter 5 in this volume), imperial Qin policies concentrated on regulating the activities of merchants rather than eradicating them entirely. The earliest evidence of legal discrimination against merchants dates from 214 B C E, when 10

Korolkov, “Empire-Building and Market-Making,” pp. 426–30.

106

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

the Qin enacted a deportation law that authorized the impressment of merchants and other kinds of transients into military service on the distant northern and southern frontiers. The Qin state compiled “marketplace registers” (shiji 市籍) that listed merchants and tradesmen separate from the general population. Those currently enrolled on the marketplace registers – or whose parents or grandparents had been so registered – were subject to these deportation orders. Scholars disagree on the question whether marketplace registration was based on the current occupation of individuals or instead inherited from one’s forebears, but certainly the trend was to treat it as a hereditary status. All persons who traveled away from their native place were required to carry passports, and itinerant merchants had to register with local officials to obtain permission to sell their wares. In the Liye documents, marketplaces figure most prominently as sites where the government acquired or disposed of goods, but the collection of fees and excises on private shops and commerce confirms the presence of private merchants as well. Judicial records in the Yuelu Academy collection of Qin excavated manuscripts include several cases dating from the 220s B C E that involved disputes among merchants over property claims to market stalls.11 Sale of goods outside designated marketplaces was forbidden, apart from exceptions such as tiles, bricks, and manure. It appears that private trade in a remote frontier town like Qianling was limited to petty exchange. Apart from selling products manufactured in state workshops and disposing of their surplus stocks, the Qianling authorities relied on markets to procure goods, most importantly textiles and slaves. Local officials were obliged to set prices for government procurement and for leasing or selling public goods based on an annual determination of fair market prices (pingjia 平價, shijia 市價). The Qianling region was rich in metals, and the Liye documents mention in addition to the manufacture of weapons at the county armory a specialized office supervising iron mining and metallurgy. The iron mines and foundries utilized convict labor, which was also employed in tin and copper mining. There are also references to commercial taxes and fees from private trade in metals. It is unlikely that the Qin government instituted a state monopoly on iron mining and manufacturing as the Han government would a century later. Sima Qian’s “Biographies of the Moneymakers” recounts a number of cases in which the Qin relocated owners of ironworks from commercialized

11

Ulrich Lau and Thies Staack, Legal Practice in the Formative Stages of the Chinese Empire (Leiden, Brill, 2016), pp. 145–66, 188–210.

107

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

regions such as Zhao and Wei to more peripheral areas, where they resumed their entrepreneurial activities and acquired even greater fortunes.12 Maxim Korolkov has proposed that the Qin project of empire building also contributed to the development of market networks, albeit within the framework of a command economy.13 State investment in roads, transport, and communications, as well as the establishment of currency regulations and uniform standards for weights and measures, surely yielded benefits in lowering transaction costs for all. The legal system, harsh as it was, also specified an explicit regulatory structure that provided consistency and predictability to exchange transactions. But the dense bureaucratic communications and record keeping of the Qin regulatory regime revealed by the Liye archive and the imperial government’s far-flung networks of production, collection, and transfer of revenues and resources undoubtedly generated high costs as well. The Qianling authorities’ frequent resort to the market for state procurement implicitly conceded the limitations of the command economy. Consequently, state demand stimulated private production and exchange, and the state also played an active role in the development and spread of technological innovations. The influence of a market mentality can be glimpsed in the trend toward measuring the value of goods – notably slaves – in coin. By imperial times, the Qin government had shifted from calculating fines in suits of armor to monetary sums. Awards for meritorious conduct on behalf of the government also were made in money. At the same time, local officials in Qianling mention shortages of coin as a hindrance to government procurement, and scarcity of coin probably was one reason why fines were frequently discharged through labor service rather than money payments. The effects of the expansion of the empire into the populous and more commercialized regions of the Central Plain cannot yet be ascertained. As Qin absorbed the core territories of the Central Plain states into its empire, it is likely that it began to collect a greater portion of revenues in money. The heterogeneity of circulating coin in the early Han period indicates that the Qin imperial government was unable to supplant the array of regional bronze currencies with its own banliang coin. This is hardly surprising; the injunction in Qin legal statutes requiring local authorities to accept substandard coins in payments and the heterogeneity of extant specimens of banliang coin show that the Qin state struggled to supply sufficient stocks of coin even before its major imperial conquests. 12

SJ, 129, pp. 3277–8.

13

Korolkov, “Empire-Building and Market-Making.”

108

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

In the fiscal domain, the Qin Empire can perhaps best be characterized as what I have elsewhere defined as the militarist–physiocratic modality of the Chinese fiscal state.14 The Qin empire builders shared the common predilection of Warring States economic philosophy to exalt agriculture as the fundamental occupation (benye 本業) and the principal source of wealth for the people and the state alike (see von Glahn, Chapter 5 in this volume). To a greater degree than their contemporaries, however, they – like the eighteenth-century French physiocrats – regarded agriculture alone as truly productive, in the sense that it generated surplus value (produit net, in the language of the physiocrats), while deeming all other economic activity inherently sterile. The revenues that support the ruler and his administration thus should be derived from the produce of the land, which enterprising farmers should hold in full possession without any encumbrance. The French physiocrats applauded the monarchal absolutism of eighteenth-century China, and indeed their vision of political economy was inspired by the “despotisme de la Chine,” to quote the title of François Quesnay’s 1767 treatise.15 Above all Quesnay admired the strict obedience to the law, even by the emperor, that he attributed to the Chinese in terms that probably were more applicable to the Qin dynasty than to the Qing dynasty of his own day. Of course, there were crucial differences between Qin political economy and physiocratic ideals, above all regarding the regulation of merchants and commerce. The physiocrats were pre-eminent exponents of laissez-faire (a term coined by Quesnay), but conceived of the freedom of trade in opposition not to an activist state but rather to feudal privilege and monopolies. The physiocratic conception of free trade made allowances for state guidance in ways antithetical to the doctrines of nineteenth-century liberalism. Nor would the French physiocrats, dedicated to the pursuit of the maximum satisfaction of all, embrace the overriding goal of the Qin empire builders: war and conquest. In labeling the Qin modality of political economy “militarist–physiocratic” I borrow the term “physiocratic” in a heuristic sense to emphasize the centrality of agriculture as the source of private and public wealth, without intending any simplistic equation of Chinese political economy with European ideas.

14

15

Richard von Glahn, “Modalities of the Fiscal State in Imperial China,” Journal of Chinese History 4.1 (2020), 1–29. François Quesnay, Despotisme de la Chine, in Christine Théré, Loic Charles, and JeanClaude Perrot (eds.), François Quesnay: Oeuvres économiques complètes et autres textes (Paris, Institut national d’études démographiques, 2005), 2, pp. 1005–1114.

109

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

The fiscal policies of the Qin militarist–physiocratic state proceeded from the cardinal principles of (1) light taxation of the agrarian base of society, (2) strong regulation of commerce and industry with punitive sanctions for profit taking and deficiencies in workmanship, and (3) heavy reliance on conscript labor and military service as the primary obligations of subjects to the imperial state. Despite its enormous territorial and demographic scale (the population of the Qin Empire has been estimated at roughly 25–40 million persons), the Qin state – as the excavated documents vividly demonstrate – was deeply embedded in local society. Intensive governance that penetrated society down to the household level was made possible by building a vast apparatus of civil registration, record keeping, and bureaucratic communication. The “legibility” of society that James Scott sees as strictly the product of the modern state – knowledge of population and resources through the invention of surnames, standardization of weights and measures, freehold land tenure, cadastral surveys and population registration, the standardization of language and laws, and urban spatial design – already was a hallmark of the early Chinese empire.16 This legibility was necessary to the institution of universal conscription for military and statutory labor service that was the foundation of imperial rule.

State and Economy in the Early Han Dynasty The Qin Empire collapsed after the death of the First Emperor in 210 B C E , but its legacy profoundly influenced its long-lived successor, Han (202 B C E – 220 C E), which adopted many Qin institutions. The Han, like the Qin, was committed to the smallholder family farm as the foundation of popular welfare and state power. Many basic elements of Qin rule – including the establishment of the household as the basic social and fiscal unit, the use of merit ranks to award landed property, the legal tenet of mutual responsibility and shared liability among kinfolk and neighbors, and strict regulation of commerce and industry – became cornerstones of Han imperial governance as well. Nonetheless, the early Han rulers also distanced themselves from what they perceived as the excesses of the Qin regime, especially the concentration of power in the hands of an absolute monarch. Professing to return to the benevolent rule of the ancient sage-kings, the founder of 16

James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven, Yale University Press, 1998).

110

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

the Han Dynasty, Liu Bang (Emperor Gaozu, r. 202–195 B C E), revived the principles of shared governance embedded in the patrimonial state. In imitation of the early Zhou kings, Liu awarded sovereign rights over much of his territory to imperial kinsmen. Under this hybrid administrative system, more than two-thirds of the empire’s territory comprised satrapies (wangguo 王國) governed by imperial kinsmen who exercised considerable independence from the court. In addition, the Han emperors frequently awarded nobilities (houguo 侯國) to imperial kinsmen, meritorious officials, and personal favorites. These nobilities entailed rights to a share of the revenue collected from a specified number of households – ranging from several hundred to 10,000 or more – in a local jurisdiction as well as responsibility for law and order among them. In total, 788 nobilities were created between 201 B C E and 5 C E, but the number in existence at any one time was considerably fewer. Although nominally hereditary, nobilities were regularly rescinded for a variety of reasons. For Liu Bang, as for the Qin rulers before him, the overriding priority was to amass resources to sustain control over a vast empire. In 203 B C E, even before the formal declaration of the Han Empire, he inaugurated a capitation tax (suanfu 算賦) of 120 coins assessed on all adults, men and women, as well as an additional twenty coins for each child aged seven to fourteen. The following year, upon the end of hostilities, Liu demobilized his army by awarding mingtianzhai land grants to soldiers and officers based on merit rank (increased from seventeen ranks in Qin to twenty). He also set the land tax levied on households at one-fifteenth of the harvest, probably a reduction from the average rate under Qin. The fledgling Han state reversed the trend of the Qin Empire toward highly centralized fiscal administration. The Lesser Treasury (shaofu), which dominated Qin state finances under Li Si, was reduced to serving as the Privy Purse, providing for the needs of the imperial household. The exchequer (zhisu neishi) regained principal authority over financial administration. Each of these departments had its separate streams of revenue: income from the land tax, the suanfu levies, and state-owned lands was assigned to the exchequer, while the “profits of the mountains and marshes” (i.e., mineral and timber resources), commercial taxes, the suanfu tax on children, foreign tribute, mint seigniorage, and revenues from the emperor’s personal lands were deposited in the Lesser Treasury. When military emergencies or natural disasters demanded an urgent government response, the exchequer frequently had to turn to the Privy Purse for an infusion of funds. 111

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

Labor service and military conscription remained onerous under Han rule. Unfortunately, the standard historical sources offer only inconsistent explanations of how Han labor service and military conscription operated, and interpretation of this fragmentary evidence remains disputed. Recently excavated texts such as the Zhangjiashan manuscripts, which preserve a major portion of the “Statutes and Ordinances” issued by the Han court in 186 B C E , have filled some of the gaps in our knowledge, but some issues remain unresolved.17 It appears that in the early Han adult males generally served two year-long tours of military duty: one year on guard duty at the capital and one year on assignment either at a frontier garrison or within the conscript’s home commandery. The Han raised the age of liability for military service from the Qin standard of seventeen to between twenty and twenty-four years, depending on the person’s merit rank, with the highest ranks exempt from military conscription altogether. In 155 B C E the privileged status of the higher merit ranks was voided, however, and all males were registered for military service at age twenty. Adult women as well as men were subject to labor conscription, but not military service. In an effort to promote population recovery, expectant and nursing mothers were granted a three-year exemption from labor service. By the same token, the state induced young women to marry by imposing a substantial penalty (a fivefold increase in the suanfu levies) on unmarried women of fifteen to thirty years of age. Men and women did not shoulder equal labor service obligations, however. Men were liable for month-long “rotation duty” (gengzu 更卒) assignments – which varied in frequency, from once every four months to once a year, undoubtedly the most onerous part of the suanfu levy – engaged in heavy work such as dike and road building or transport of grain and other government goods. “Rotation duty” – an extension, it seems, of military conscription – often entailed work away from home, while women’s labor assignments were confined to the locality.18 The capitation taxes and labor service levies required detailed documentation of the age and gender of all household members. The “Household Statutes” included in the Zhangjiashan manuscripts reveal that local officials 17

18

For a translation of the Zhangjiashan corpus, see Anthony Barbieri-Low and Robin D. S. Yates, Law, State, and Society in Early Imperial China: Translation and Study of the Zhangjiashan Legal Texts (Leiden, Brill, 2016). This interpretation of the conflicting evidence regarding the nature of gengzu and women’s labor service obligations is based on Washio Yu¯ko 鷲尾祐子, Chu¯goku kodai no sensei kokka to minkan shakai: Kazoku, fu¯zoku, ko¯shi 中国古代の専制国家と 民間社会：家族，風俗，公私 (Kyoto, Ritsumeikan to¯yo¯shi gakkai, 2009), pp. 58–65, 164–5.

112

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

were obliged to conduct annual household surveys and compile at least five types of registers that recorded household members, merit ranks, arable lands, dwellings and gardens, and tax and service obligations. Although merit ranks initially were based on military service, government office quickly eclipsed military accomplishment as the basis of privileged mingtianzhai classification. The higher ranks (ten and above) were restricted to government officials, who received exponentially greater land allotment quotas. In addition, the higher ranks were exempt from inclusion in the wu mutual-responsibility units and by extension from the legal liability of mutual implication in criminal conduct. In many respects, then, the early Han rulers retained the main features of the militarist–physiocratic fiscal regime inherited from Qin. With the ascension of Emperor Wen (r. 180–157 B C E), however, the Han court pivoted toward more liberal economic policies inspired by the Huang–Lao philosophy of government, in which the supreme authority of the universal monarch is subject to the tenets of natural law. Emperor Wen sought to emulate the Huang–Lao ideal of a ruler who ensures the economic well-being of his subjects through personal frugality, minimal taxation, and refraining from disrupting their livelihood. For example, Wen lowered the land tax from onefifteenth to one-thirtieth of the harvest, and in 167 B C E abolished the land tax altogether. The most controversial of Emperor Wen’s policies was his abrogation, in 175 B C E , of the state monopoly on coinage. From the outset the Han state tolerated the circulation of “light coins” – still denominated as banliang – to ensure a sufficient supply of coin. But the empire was soon deluged with debased and underweight coins that triggered price inflation. Wen’s action was intended to restore a sound currency by stipulating uniform standards for the size and weight of coins but allowing private minting. Among those who took advantage of these opportunities were Liu Pi, the Prince of Wu (capital at Yangzhou), whose satrapy encompassed rich copper mines, and Deng Tong, a Chang’an-based merchant who leased mining and minting rights to local merchants in Sichuan. Critics among the emperor’s advisers condemned free coinage for further debasing the money supply, stimulating commerce at the expense of agriculture, and worsening inequality. In fact, numismatic evidence suggests that coins issued during this time were fairly standardized. The subtext of the critique against relinquishing the state monopoly over coinage was that free coinage weakened state power while fanning sedition among the imperial princes. Indeed, in 154 B C E, three years after Emperor Wen’s death, Liu Pi and other princes rose up in

113

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

rebellion against his successor, Emperor Jing (r. 157–141 B C E). But the rebellion failed, and Liu Pi was killed in battle. The miscarriage of the Rebellion of the Seven Princedoms gave Emperor Jing justification both to rescind Wen’s laissez-faire policies and to rein in the refractory princes. In 145 B C E Jing revoked the fiscal autonomy of the satrapies, whose inhabitants henceforth owed their suanfu taxes and labor services to the central government. Free coinage was repealed in 144 B C E, and the land tax reinstated at the original rate of one-fifteenth of the harvest. Nonetheless, the trend toward worsening economic inequality and the concentration of wealth and landholdings in the hands of imperial kinsmen, merchants, and families of government officials continued. The mingtianzhai system of land allocations began to break down, and many farming families lost their lands and independent means of livelihood. In addition, beginning in 133 B C E , the Han court’s tense relations with the Xiongnu confederation of steppe nomads on its northern frontier erupted into prolonged armed conflict. Costly wars against the Xiongnu would lead to even greater government intrusion into economic life than the First Emperor of Qin had envisioned.

The Mercantilist Fiscal State of Emperor Wu and Its Aftermath During the long reign of Emperor Wu (141–187 B C E), the militarist–physiocratic fiscal regime of the Qin–early Han was transformed into what we might describe as a mercantilist fiscal state. The key difference between the two is that the latter concentrated on mobilization of economic resources rather than military manpower. The best expression of the mercantilist fiscal philosophy is found in the “Ratios of Exchange” chapters of the Guan Zi treatise, composed in the early Han period (see von Glahn, Chapter 5 in this volume). In many respects, the principles of political economy expressed in the Guan Zi – maximizing exploitation of agricultural, mining, and manufacturing resources; maintaining a favorable balance of trade; and amassing specie reserves that enabled the state to control prices and consumption – parallel the mercantilist doctrines of early modern Europe. In contrast to European mercantilism, however, the Chinese mercantilist state sought not to strengthen the domestic merchant class but rather to replace it. Emperor Wu enacted these policies to raise the massive revenues needed to defray the costs of his aggressive campaigns of conquest and colonization in Central Asia, Korea, and Vietnam. Apart from boosting state income, Emperor Wu also aimed to arrest the growing wealth and power of the 114

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

seigneurial and mercantile classes. Toward these ends Wu and his advisers imposed new and onerous taxes on merchant income and capital assets, placed the lucrative salt and iron industries under state monopoly, aggressively intervened in commodity exchange, and created a new monetary system. Ironically, the strategies for reviving a state-directed command economy were crafted by a brain trust of merchants whom Emperor Wu recruited as his fiscal advisers. Under their guidance Wu reorganized the central government’s fiscal administration, consolidating fiscal authority in the hands of the Grand Director of Agriculture (danongling 大農令) – as the exchequer had been renamed in 143 B C E, which subsequently was changed to Grand Superintendent of Agriculture (dasinong 大司農) in 104 B C E – and ad hoc agencies, such as the Superintendent of Waterways and Parks (shuiheng duwei 水衡都尉), established in 115 B C E . The first step taken by the new leadership was to broaden a tax originally levied solely on merchants into a universal tax on commercial enterprise. Under the new policy (known as suanmin 算緡), anyone engaged in commerce or moneylending was assessed a tax of 120 coins (one suan, the basic unit of the suanfu poll tax) for each 2,000 coins of assets (i.e., a rate of 6 percent); artisans were taxed at half that rate. Those accused of evading the tax were subject to criminal penalties and confiscation of their property, which could include enslavement of their family members. Moreover, the law forbade merchants from possessing lands allocated under the mingtianzhai system or holding government office. While the suanmin legislation separated merchants into a distinct occupational class subject to legal discrimination, it also was intended to discourage anyone other than professional merchants from engaging in commercial activities. Second, Emperor Wu instituted a state monopoly on the production and sale of iron and salt in order to capture the profits of the most lucrative private industries. The emperor transferred the Privy Purse’s claim to the “profits of the mountains and marshes” to the office of the grand director of agriculture, which in 119 B C E seized iron and salt production facilities and turned them into state enterprises. The Han state already had an elaborate procurement system to obtain iron for its workshops and arsenals; these officials now assumed control of the production and sale of iron in the private market as well. In most cases the former foundry and saltern owners were recruited as managers of the new state operations. The most far-reaching intrusions of the state into commerce resulted from implementation of the “equitable-delivery” (junshu 均輸) and “balancedstandard” (pingzhun 平準) policies. Under the equitable-delivery system, 115

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

local officials used their money revenues to purchase goods (especially textiles) to supply the capital and the frontier armies. In 110 B C E the equitabledelivery system was merged with the salt and iron monopolies and rechristened the balanced-standard policy. Government agents throughout the empire used public funds, including monopoly revenues, to smooth out price fluctuations by buying up goods where prices were low and selling them where prices were high. Although promoted as a means of preventing price gouging by merchants, the balanced-standard system also generated substantial revenues. Later, c. 57 B C E, the Han inaugurated the “ever-normal” (changping 常平) policy, whereby the state sought to curb seasonal oscillations in grain prices, buying up grain after the harvest to ensure favorable prices for producers and selling grain in the lean spring and summer months to benefit consumers. Although the Han ever-normal granary system appears to have been ineffective, this price-control policy became a fixture of fiscal governance and social welfare strategies in later dynasties (see Dunstan’s chapter in this volume). Finally, and perhaps most successfully, Emperor Wu’s government issued a new, heavier bronze currency, the wuzhu coin, to replace the heterogeneous array of banliang currencies in circulation. When first introduced in 118 B C E the new coins were cast in provincial mints and varied significantly in quality. After the minting of wuzhu coins was consolidated at the capital under the aegis of the superintendent of waterways and parks in 113 B C E, the state produced substantial quantities – estimated at more than 400 million coins annually for the remainder of Wu’s reign – of fine wuzhu coins. Although roundly reviled for his short-lived experiments with various fiat currencies – including a deer-hide “bill” valued at 400,000 coins – Wu succeeded in creating a unified currency and a monetary standard that would endure for centuries (see Kakinuma’s chapter in this volume). The fiscal system constructed by Emperor Wu’s advisers came to rely heavily on indirect taxation (Table 3.1). Roughly half of total revenues – especially in-kind receipts of grain and fodder – were retained at the local level. Most of the central government’s income – and nearly half of total revenue – was collected in the form of coin. The salt and iron monopolies – which generated more than half of the central government’s revenue and 20 percent of total revenues – clearly had become vitally important to the Han state. Indirect levies paid in coin also provided most of the Privy Purse’s revenues. This unprecedented intrusion of government into commerce and industry provoked sharp conservative reaction, ultimately leading to the repeal of 116

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

Table 3.1 Estimated government revenues of the Han Empire in the first century B C E (all figures in millions of coins) Revenue source

Central Local Privy government governments Purse

1,000 Land taxa 80 Fodder taxa Suanfu levy 2,071 Salt and iron monopolies 3,800 Other 100 Commercial excises and mining levies Poll tax on minors Mint seigniorage Gold tribute from nobility Imperial clan lands Miscellaneous Total 7,051

6,000 1,200 2,071

9,271

1,300 287 154 19 300 600 2,660

Total 7,000 1,280 4,143 3,800 100 1,300 287 154 19 300 600 18,982

a

In-kind levies. Source: Yamada Katsuyoshi 山田勝芳, Shin Kan zaisei shu¯nyu¯ no kenkyu¯ 秦漢財政収 入の研究 (Tokyo, Kyu¯ko shoin, 1993), pp. 653–8.

much of Wu’s statist agenda. In 81 B C E Wu’s successor convened a formal debate at the court at which Wu’s ministers defended their fiscal policies against harsh criticism by an assembly of “learned men” steeped in Confucian doctrines who argued for a return to the minimalist, laissez-faire policies of Emperor Wen. The outcome of what came to be known as the “Salt and Iron Debates” largely preserved the status quo as far as the state monopolies on salt and iron were concerned (see von Glahn, Chapter 5 in this volume). But the tide had turned. In subsequent decades the court dismantled the rest of Wu’s mercantilist programs, at one point considering whether to ban metallic currency altogether in favor of barter exchange. With the retreat from Wu’s interventionist agenda the concentration of landownership and wealth intensified. An oligarchy of Great Clans (haozu 豪族) comprising landowners and officeholders allied through marriage networks dominated both government and local society. The widening gulf between rich and poor provoked Wang Mang, scion of a distinguished family who became regent to a child emperor in 1 B C E , to seek to curb the rising power of the Great Clans by reviving the mercantilist fiscal state, albeit draping it in Confucian trappings. Wang’s political vision was inspired by the Rituals of Zhou, a canonical text that purports to be a detailed

117

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

blueprint of the bureaucratic organization of the ancient Zhou kingdom but most likely reflects the institutional ethos of imperial Qin. After the child emperor died in 6 C E Wang stalled nomination of a successor, and three years later – in the name of restoring the golden age of the Zhou – he anointed himself emperor of his own Xin (“New”) dynasty. Wang then launched a draconian program of institutional reform, reorganizing the bureaucracy to strengthen central control over local officials and abolishing the Han nobility and its hereditary rights to land revenues. Invoking the Confucian philosopher Mencius’ claim that the ancient sages had distributed land in equal portions to farming families under the so-called “well-field” system (see von Glahn, Chapter 5 in this volume), he nationalized landholdings in the name of redistributing them to the poor. At the time of his usurpation Wang also banned the sale of land and slaves, but he was forced to rescind this prohibition three years later. Wang Mang was equally assertive in expanding state control of commerce and money. While still regent he had introduced two new forms of bronze currency with purely nominal values that circulated concurrently with the existing wuzhu coin. After declaring himself emperor, Wang proscribed the use of the wuzhu coin and created a bewildering array of new fiat currencies by recoining old wuzhu coins. Other measures adopted by Wang Mang exceeded even the ambitions of Emperor Wu, broadening the reach of the state monopolies on commodities, subjecting trade and moneylending to strict regulation, and nationalizing gold stocks. Such policies drove sound coin and gold out of circulation, unleashing rampant inflation and severely disrupting commerce and industrial production. Coins deposited in tombs and hoards during this era consist overwhelmingly of Wang’s new currencies, suggesting that much of the existing stock of wuzhu coins had vanished into the melting pot. Wang Mang’s ambitious agenda directly challenged the political, social, and economic power of the Great Clans, inciting fierce opposition. Government officials resisted implementing policies that threatened their own vested interests. Political chaos bred economic distress, antagonizing all ranks of society. In the end popular insurrections, combined with the determined resistance of the landowning class, toppled Wang’s government. In 23 C E rebel forces seized the imperial palace and put Wang to death. Two years later, Liu Xiu (Emperor Guangwu, r. 25–57 C E ) re-established the Han dynasty with the full support of the Great Clans. Under the Eastern Han regime (25–220 C E), celebrated as a return to sagely Confucian doctrines of

118

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

minimal government interference in economic life, the Great Clans further consolidated their control over land and wealth. Despite the mercantilist bent of Wang Mang’s policies, ultimately they induced a shift to greater reliance on in-kind revenues. The land tax paid in grain was supplemented by a new levy paid in hemp or silk cloth. Given the disruption of trade and industry caused by Wang’s ever-changing fiscal and monetary policies, the proliferation of commercial taxes probably generated little revenue. Ironically, the new excises on commerce and industry extended even to domestic textile production, adding to the burden of farming families. However, the trend toward commuting labor services to payments in money accelerated during the Wang Mang era and after. In 30 C E, Emperor Guangwu dismantled one of the pillars of the militarist–physiocratic state by replacing universal military conscription with money payments to hire substitutes. The principle of service to the state as the foundation of social order had been abrogated, and instead increasing numbers of people became subject to new masters among the rising class of Great Clan landowners. Imperial largesse – in the form of generous mingtianzhai land allocations to officeholders and the award of nobilities to meritorious officials and imperial kinsmen – contributed to the initial formation of great landed estates. By the Eastern Han period the mingtianzhai allocations were defunct, and many indebted smallholder farmers were forced to surrender their independence and work as bondservants in service to the Great Clan landowners. Emperor Guangwu’s reduction of land taxes – in imitation of Emperor Wen – to onethirtieth of the harvest thus chiefly benefited estate owners. Long-distance trade and commercial cities atrophied, yet the manorial estates remained engaged with the market economy. Entrepreneurial estate owners with ample dependent labor at their disposal not only cultivated staple grains and legumes for sale as well as domestic consumption but also produced processed foods and dyestuffs, operated breweries and mills for grinding flour and pressing vegetable oil, and manufactured silk and hemp cloth. The repeal of the iron monopoly in 88 C E and the dismantling of much state-organized manufacturing stimulated private enterprise. Although market circulation was more circumscribed, trade at the local level remained vibrant. The Eastern Han government preserved the main features of early Western Han taxation, including the suanfu capitation taxes, but the contraction of the tax base premised on independent family farms sharply diminished state revenues. Throughout the Eastern Han the state repeatedly had recourse to extraordinary levies (diao 調), so much so that these 119

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

became a routine means of funding government needs. The diao levies, which were assessed on households graded by wealth, marked an important departure in imperial fiscal policy: a trend toward assessing taxes on households rather than individuals, and collecting taxes primarily in kind rather than money. The civil wars that ravaged China after the outbreak of the Yellow Turban Rebellion in 184 enfeebled the Han court, and effective power gravitated to regional powermongers. In 196 the warlord Cao Cao set a puppet emperor on the Han throne while maneuvering to bolster his personal power. Faced with population losses and land abandonment caused by prolonged warfare, Cao assigned tracts of vacant lands to both soldiers and civilian households, who in effect became tenants of Cao’s warlord regime. These agricultural colonies (tuntian 屯田) provided crucial revenues that enabled Cao to solidify his control over north China. Upon Cao’s death in 220 his son formally declared the founding of the Wei dynasty (220–265) as successor to Han. The Wei rulers routinely rewarded generals and officials with allotments of tuntian lands and the households that worked them, which in effect privatized these resources. The Nine Ranks system created by the Wei in 220 formally recognized the Great Clans as an aristocracy with hereditary rights to office and generous legal and economic privileges, including exemptions from labor service. By becoming private clients under the lordship of aristocratic patrons, farming families vanished from the tax registers and sheltered themselves from state exactions. In 198 Cao Cao introduced a new taxation system, the “household levy” (hudiao 戶調), that imposed tax quotas on households in grain and cloth, but not in coin. Guangwu, the first Eastern Han emperor, had resumed minting wuzhu coins, but coinage was largely relegated to local officials, and mint output soon faltered. Circulating coin was debased by clipping and counterfeiting, as caches of coin from this era confirm. The trend toward state payments in high-value silks reflected the volatile value of circulating coin. The Wei kingdom and the succeeding Jin dynasty ceased coinage altogether, and bolts of cloth became the new monetary standard. Sun Quan, ruler of the rival Wu kingdom in the Yangzi river valley, preserved the Han suanfu taxes paid in coin, but Sun too utilized tuntian land allocations to resettle displaced populations and rebuild a firm revenue base. Unlike Wei, however, the Wu state often placed tuntian lands under the jurisdiction of local civil administrators rather than the military. From the evidence of the Zoumalou manuscripts from Linxiang county (modern Changsha) dating to the 230s, many of the households assigned tuntian 120

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

lands were refugees, criminals, prisoners of war, government slaves, and indigents over whom the state wielded considerable coercive power. Indeed, these groups, who probably accounted for at least a quarter of Linxiang’s population, seem to have been forcibly settled on tuntian lands and compelled to pay a much higher land tax in grain – roughly one-half of the harvest – than ordinary landowners. In addition to the grain tax, Linxiang officials levied taxes in cloth and coin (usually converted to additional grain payments) on the tuntian households, as well as a profusion of irregular diao levies.19 In contrast to Cao Cao’s regime, however, Wu’s tuntian households were not under the thumb of officials and generals, but rather maintained their social and economic independence.

Divergent Paths of State Finance during the Period of Disunion In 265 the Sima clan usurped rulership of the Wei kingdom, establishing their own Jin dynasty (265–420). Jin subsequently vanquished the rival Wu and Shu Han kingdoms, re-establishing a unified empire in 280. While retaining Cao Cao’s hudiao system, the Jin sought to restore more uniform taxation by eliminating the distinctions between tuntian and ordinary landholdings and standardizing land taxes and labor services. The Jin attempted to reverse the trend toward private clientage by introducing a new fiscal system (zhantian ketian 占田課田) that adjusted the rates for tax payments in grain and silk according to the household’s landholdings and the age and gender of its members. The new system did not allocate lands to farm families, as often has been supposed, but rather imposed standardized tax rates based on the maximum amount of land that an able-bodied man was deemed capable of cultivating (fifty mu). Its purpose, then, was to generate revenues more than to ensure the economic independence of cultivators. The Jin state did attempt to restrain the growth of latifundia by imposing tax exemption quotas for private clients and landholdings, up to a maximum of fifty client households and 5,000 mu of land for the highest-ranking officials. The extent to which these still-generous limits were enforced remains unclear, but it is unlikely that they had much effect in curtailing private clientage. In 311, steppe nomad invaders sacked the Jin capital, Luoyang, forcing the court to flee to the Yangzi river delta, where it established a new capital at 19

Jiang Fuya 蔣福亚, Zoumalou Wujian jingji wenshu yanjiu 走马楼吴简经济文书研究 (Beijing, Guojia tushuguan chubanshe, 2012).

121

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

Jiankang (modern Nanjing) in 317. Over the next three centuries China was divided between foreign-ruled regimes in the north and a succession of Chinese dynasties, five in all, in the south. The fiscal strategies adopted by the foreign and native dynasties evolved in different directions. In the south, the refugee Jin court and its successors exercised weak control over landholdings, and the concentration of landownership continued unabated. The Great Clans that accompanied the Jin court successfully reestablished manorial estates, especially in the underpopulated marshlands of the Yangzi delta and the Hangzhou Bay littoral. Political privileges enabled them to accumulate large tracts of land, and private clientage provided the labor power to bring them under cultivation. Bondservant clients remained responsible for their own subsistence; they were mostly sharecroppers who divided their harvest with their masters as well as performing various kinds of personal duty in lieu of labor service to the state. In contrast to the north, the monetary economy thrived in the south. Land reclamation and the construction of irrigation dikes and channels in the Yangzi delta created a network of waterways that facilitated transport and commerce throughout the region. Shipping canals connected major cities such as Yangzhou, Suzhou, and Shaoxing with the capital at Jiankang, which became the hub of regional and international trade. The southern courts continued to mint coin, both wuzhu denominations and their own dynastic issues, although the quality was poor and private coinage proliferated. Despite the debasement of the currency, money use expanded, and coin was increasingly used in state payments. According to a 476 report on tax arrears, 39 percent of the unpaid taxes were owed in coin, as opposed to 46 percent in grain and 15 percent in cloth. Apart from the retention of the hudiao system of tax payments in grain, cloth (primarily hemp rather than silk), and coin levied on households, we possess little information about the fiscal administration of the Southern Dynasties. Under the Southern Qi (479–502), taxes were assessed on household wealth (measured by arable lands, mulberry fields, and dwellings), and half of the cloth tax was commuted to coin. This system was continued by the succeeding Liang dynasty (502–551), except that the cloth tax became a capitation tax levied on adult males (ding 丁). The revenue system of the Southern Dynasties can be characterized by a low level of statutory taxation; reliance on volatile sources of revenue, including commercial taxes and ad hoc levies; and a trend toward payments in coin. After 311, northern China was divided into numerous short-lived regional regimes before the rise of the Northern Wei (386–534), ruled by nomadic Xianbei Turks, who gained control of most of the north in the 460s. In 122

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

keeping with nomadic practice, the Xianbei initially awarded large numbers of conquered Chinese as slaves to their generals, nobles, and warriors, while imposing hereditary servile status on skilled craftsmen, clerks, and entertainers. At the same time the Xianbei cultivated the allegiance of the Great Clan magnates, whose support was necessary to consolidate their rule. The drastic reduction of independent family farms narrowed the Northern Wei’s revenue base, however. In the 480s the Xianbei rulers undertook a series of bold initiatives to remake the Northern Wei state into a Chinese-style bureaucratic empire. To foster a unified ruling class, intermarriage between the Xianbei nobility and the elite Chinese Great Clans was encouraged, as well as adoption of Chinese language, dress, and customs. The Northern Wei moved their capital to Luoyang, the Eastern Han capital located on the Central Plain. Most importantly, the Northern Wei sought to rebuild a stable agrarian empire by instituting a new program of state land allocations, the equal-field (juntian 均田) system. Under the equal-field regulations issued in 492, each male adult received an allotment of eighty mu of arable land (lutian 露田) – forty mu (approximately 3.7 hectares) for cereal cultivation along with forty mu of fallow land for cultivation in two-year rotation – as well as twenty mu for planting mulberries (or ten mu in the case of hemp cultivation). Married women were allocated twenty mu of arable land, twenty mu of fallow land, and five mu of hemp land. Additional lands were allotted to households based on the number of their male and female slaves in equal proportion to free subjects (according to the same gender differences). Furthermore, extra allotments – thirty mu of arable land and thirty mu of fallow land – were given for each ox the household possessed. Lands for residences and gardens were apportioned in the amount of one mu for every three free persons in the household and one mu for every five slaves (see Table 3.2). The only direct evidence for the operation of the equal-field system comes from registers compiled at the frontier garrison town of Dunhuang in 547. These documents depict the equal-field system functioning according to its original design, with lands regularly repossessed from deceased householders and redistributed to new ones. But most households received half or less of the amount of land to which they were entitled. These shortfalls are not surprising, however, given that Dunhuang was an oasis town with limited arable land. The deficits were confined almost entirely to the lutian allotments; in contrast, Dunhuang’s inhabitants received nearly all of the statutory allocations of hemp lands to which the household obtained permanent allodial rights, which could be transmitted to heirs. 123

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

Table 3.2 Land allocations under the Northern Wei equal-field system Land type

Arable land Standard (lutian 露田) allocation (zhengtian 正田) Supplemental allocation (beitian 倍田) Mulberry fields (sangtian 桑田) Hemp fields (matian 麻田) Dwelling allocation

Free subjects (liangmin良民) Adult males Adult females

Oxenb Slavesa (nubi 奴婢)

40 mu

20 mu

Same as 30 mu free subjects

40 mu

20 mu

30 mu

20 mu 10 mu 5 mu 1 mu per 3 persons

1 mu per 5 persons

a

Unlimited Up to four oxen Source: WS, 110, pp. 2853–4

b

The zu–diao 租調 tax system, enacted as a corollary to the equal-field land allocations, imposed on each married couple a zu payment of two shi (seventy-four liters) of grain and a diao payment of one bolt (27.9 meters) of silk or hemp cloth. Unmarried males were taxed at a rate one-fourth that of a married couple, with additional assessments of oneeighth for each slave and one-twentieth for each ox. In principle, upon the death of the male head of household – or when he reached age seventy – the lutian land allotments were returned to the state, but fiber crop lands and dwellings were passed on to his heir. The supplemental allocation of lands based on possession of slaves and oxen was highly advantageous to large landowners. Money payments were conspicuously absent from the zu–diao taxes. Coin had largely ceased to circulate in north China during the Period of Disunion, and the meager amount of coin minted by the Northern Wei appears to have circulated only in the capital region. Bolts of cloth largely superseded coin as means of payment. In sum, the Northern Wei revived the basic features of the militarist– physiocratic state of the early empires. The equal-field system supported a smallholder family-farm economy through allocations based on the household’s productive capacities (measured in able-bodied adults, oxen, and

124

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

slaves). The amounts of grain, cloth, and labor service owed to the state were directly proportional to the land allocation the household received. Thus tax rates were relatively uniform; every adult male owed equal tax and labor service obligations to the state. The new fiscal regime also was coupled to efforts to compile accurate population registers and to integrate village society into the state administration through what became known as the “three-elders” (sanzhang 三長) system of local governance, along with a new form of military conscription that restored the principle of universal military service. These institutions became the foundations of fiscal governance for the reunified empire under the Sui (581–618) and Tang (618–907) dynasties.

Continuity and Transformation of Fiscal Administration under the Sui–Tang Empires Internal dissension among the Xianbei elite brought about the dissolution of the Northern Wei regime in 534, but its successors largely retained its fiscal policies, notably the equal-field system. Following the demise of the Northern Wei, Yuwen Tai, the de facto leader of the Western Wei (534– 556), built up a new military organization known as the fubing 府兵 or “garrison militias” into a professional soldiery. Although leadership of the fubing units often was hereditary, over time the allegiance of the soldiers to their commanders weakened and the militias came under direct central control. Groups of six families rotated the duty of providing one conscript for a year-long tour of military service; the other five families supplied the person on active duty with food, weapons, clothing, and a horse. The reunification of China achieved by Yang Jian, founder of the Sui dynasty, in 589 was built on the institutional precedents of the Northern Wei and Western Wei. The Sui retained the equal-field system, but introduced some modifications. Households received allocations divided into eighty mu of “permanent shares” (yongye 永業) and twenty mu of “personal shares” (koufen 口分), which in theory reverted to the state. Separate shares for adult women were eliminated, as well as extra land allotments for households owning slaves and oxen, a change that adversely affected more affluent families. At the same time, to inculcate loyalty among its officials, the Sui awarded substantial land grants based on bureaucratic rank. Most importantly, the Sui sought to merge the fubing militias with the equal-field system. The militias were assigned to local jurisdictions, mostly in the northwest, and received equal-field land allocations. Although the Sui intended to extend the equal-field system to the entire empire, it is unlikely that it was ever applied in 125

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

the territories of the former Southern Dynasties. Rice agriculture required heavy investments of capital and labor to build and maintain irrigation systems, but periodic reallocation of lands would greatly diminish the incentive for making such investments. Moreover, the development of the south’s agricultural potential had largely come at the initiative of powerful families who staunchly resisted any attempts to restrict their landholdings. The short-lived Sui dynasty was soon displaced by the Tang (618–907) imperial house, which also was descended from the hybrid Xianbei–Chinese aristocracy promoted by the Northern Wei. The Tang, too, preserved the equal-field system, but the pressure of population growth on the land hindered its full application. Land and population registers from the frontier oasis towns of Dunhuang and Turfan dating from the 690–769 period reveal that households received their full allotment of permanent-share lands but only a fraction of the statutory amounts of the personal share that reverted to the state upon the death of the holder. Although the personal-share lands were reallocated as required by statute, clearly most land was held as permanent property of the household, indicating a shift toward de facto private ownership. In 624 the Tang introduced a taxation system premised on the equal-field tenurial system known as zu–yong–diao 租庸調 after its three principal components: taxes paid in grain (zu) and cloth (diao) and a requirement of labor service (yong). In contrast to previous practice, in which the conjugal pair of husband and wife (or “bed”) was the unit of taxation, the Tang shifted the basis to the adult male (kekou 課口 or ding). Since in theory each adult male possessed an equal amount of land, the rate of taxation was uniform: two shi (119 liters) of grain, two zhang (6.22 meters) of cloth (increased by 25 percent if paid in hemp rather than silk), and twenty days of statutory labor. Labor service often was commuted to an additional payment of cloth. Although the zu–yong–diao taxes were levied on individual adult men, taxes were paid by the head of the household. Women were exempt from taxation, although presumably they produced most of the cloth that was the principal item of taxation. In addition to the fubing militias, adult males aged twenty-five to fifty were liable for military conscription. The Tang also levied separate land and household taxes. The land tax (dishui 地稅), collected in grain, constituted the primary tax levy in the south, where the equal-field system was not implemented. In addition, the state ranked households according to nine grades of property and collected a household tax (hushui 戶稅) paid in coin earmarked for local 126

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

government expenses. This tax was relatively modest – in the 740s the average assessment was 240 coins per household, generating no more than 5 percent of the revenue collected from the zu–yong–diao taxes – but it was the only tax paid by townsmen who did not own arable land. Over the course of the seventh century, the equal-field landholding system and the fubing militias gradually weakened and collapsed. Restoration of domestic peace unleashed dynamic economic forces that eroded the foundations of the equal-field system, while incessant wars along the empire’s vast frontiers inflicted mounting hardship on the fubing soldiery and military conscripts. The Tang fiscal system relied heavily on taxes paid in grain and cloth by agricultural producers and made little effort to capture revenues from the non-farm sectors of the economy. These problems, already evident in the second half of the seventh century, spiraled out of control during the eighth century. The operation of the equal-field system was encumbered by the growing scarcity of land, especially in long-settled north China. In the densely populated regions of Guanzhong and the Central Plain farm families received only half or less of their nominal equal-field entitlements. The weight of taxation fell heavily on small farmers, who were liable for the uniform zu–yong–diao levies regardless of the actual amount of land they cultivated. This burden forced impoverished families to relinquish or abandon their landholdings, creating a floating population of landless “vagrant households” (futaohu 浮逃戶). Meanwhile, aristocrats, officials, and Buddhist monasteries amassed great estates, notably around the two capitals, Chang’an and Luoyang. By the eighth century, vagrant households had become an acute fiscal as well as social problem. Many no longer supplied tax payments and labor service. Efforts to redress the vagrancy problem, such as the guahu 括戶 resettlement program of allodial rights and tax amnesties for unregistered squatters who occupied abandoned lands enacted in 724–727, achieved some success. The guahu resettlement program was coupled with a new policy for military recruitment, also targeted at vagrant households, that awarded tax exemptions and bonuses to men who served four- to six-year terms in frontier military service. Those willing to settle permanently at frontier garrisons were given lands and dwellings. Decisive action to resolve the deepening contradictions in the fundamental Tang institutions was hindered by factional struggles at the court. From the 720s the central government began to appoint plenipotentiary commissioners to address problems such as vagrancy, counterfeiting, and tax grain transport. The resort to these commissioners reflected the inability of the ministry of 127

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn

revenue to manage fiscal affairs. In 736, the chief minister, Li Linfu, secured dominance – some would say dictatorship – over the Tang government and launched a broad-gauged series of reforms. Seeking to restore the Tang to sound fiscal footing, Li’s administration (1) replaced the hundreds of fubing garrisons with new professional armies organized into ten regional commands; (2) created a new logistical system for grain transport that absolved local officials of the responsibility to muster corvée labor, while relying more heavily on market purchases of foodstuffs to provision the armies; (3) converted most statutory labor obligations to money payments; and (4) streamlined the budget-making process and granted local authorities greater flexibility in meeting their tax quotas. This rationalization of the fiscal administration helped shore up the state’s revenue base: the registered population rose from 7.1 million households in 726 to 8.7 million households in 742, an increase of 23 percent in the span of sixteen years. But these reforms also tacitly acknowledged the stark reality that the equal-field land system and fubing militias were essentially defunct. The catastrophic An Lushan rebellion (755–763) severely crippled the fiscal capacity of the Tang state, and indeed spelled the doom of the institutional heritage of the Northern Wei. The disintegration of the old land, registration, and tax systems compelled the Tang leadership to construct a wholly new paradigm of fiscal governance. With the breakdown of central government control, the court again turned to plenipotentiary commissioners, who were given free rein to raise revenues through emergency taxes. The most important and enduring of these initiatives was the restitution of a salt monopoly under the aegis of Liu Yan, who served as salt commissioner from 760 to 780. By the time Liu was dismissed from office the salt monopoly was generating half of the central government’s revenues, and the salt commission had gained jurisdiction over transport, coinage, and commercial revenues as well. In 780 the chief minister, Yang Yan, dislodged Liu and embarked on a bold effort to restore direct taxation as the basis of the state’s fiscal order. Recognizing that the equal-field system and the zu–yong–diao taxes derived from it were obsolete, Yang transformed the dishui and hushui levies into a new, universal twice-a-year (liangshui 兩稅) tax assessed on households graded into ranks of wealth and property and collected in grain and coin. Although the twice-a-year tax proved successful, the Tang government retained and routinized expedient measures such as the salt monopoly and commercial taxes, which became mainstays of the new fiscal regime. 128

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

State and Economy

With the advent of the post-An Lushan fiscal innovations, the Tang government abandoned the basic principles of the militarist–physiocratic state: uniform direct taxation of smallholder family farms collected in kind and in labor service, universal military service, and subjection of commerce and industry to strict surveillance and regulation. Conceding the state’s inability to rectify the unequal distribution of land, the Tang leadership opted for progressive taxation of property and wealth. Although shortages of currency compelled the Tang to commute most of the coin portion of the twice-a-year tax to payments in textiles, new taxes on consumption and commerce generated substantial money revenues. The transformed fiscal regime accommodated – and in many ways enhanced – the growth of the monetary economy and the freer disposition of land, labor, and capital. These trends accelerated over the following centuries (see von Glahn’s chapter on the Tang–Song transition in this volume), and the fiscal constitution of the militarist–physiocratic state receded ever more deeply into the past.

Further Reading Barbieri-Low, Anthony, and Robin D.S. Yates, Law, State, and Society in Early Imperial China: Translation and Study of the Zhangjiashan Legal Texts (Leiden, Brill, 2016). Chen Mingguang 陈明光, Liuchao caizheng shi 六朝财政史 (Beijing, Zhongguo caizheng jingji chubanshe, 1997). Chen Mingguang 陈明光, Tangdai caizheng shi xinbian 唐代财政史新編 (Beijing, Zhongguo caizheng jingji chubanshe, 1991). Emura Haruki 江村治樹, Sengoku Shin Kan jidai no toshi to kokka: Ko¯kogaku to bunken shigaku kara no appuroochi 戦国秦漢時代の都市と国家：考古学と文献史学から のアップローチ (Tokyo, Hakuteisha, 2005). Fujiie Reinosuke 藤家礼之助, Kan Sangoku Ryo¯ Shin Nanbokucho¯ no densei to zeisei 漢三国 両晋南北朝の田制と税制 (Tokyo, To¯kai daigaku shuppankai, 1989). Hori Toshikazu 堀敏一, Kindensei no kenkyu¯: Chu¯goku kodai kokka no tochi seisaku to tochi shoyu¯sei 均田制の研究：中国古代国家の土地政策と土地所有制 (Tokyo, Iwanami shoten, 1975). Hulsewé, Anthony, “The Influence of the ‘Legalist’ Government of Qin on the Economy as Reflected in the Texts Discovered in Yunmeng County,” in Stuart Schram (ed.), The Scope of State Power in China (London, School of Oriental and African Studies, 1985), pp. 211–35. Jiang Fuya 蔣福亚, Wei Jin Nanbeichao shehui jingji shi 魏晋南北朝社会经济史 (Tianjin, Tianjin guji chubanshe, 2005). Levi Sabattini, Elisa, and Christian Schwermann (eds.), Between Command and Market: Economic Thought and Practice in Early China (Leiden, Brill, 2021). Twitchett, Denis, Financial Administration under the T’ang Dynasty, 2nd ed. (Cambridge, University of Cambridge Press, 1963).

129

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

richard von glahn Twitchett, Denis, “Merchant, Trade, and Government in Late T’ang,” Asia Major, n.s., 14.1 (1968), 63–95. von Glahn, Richard, The Economic History of China from Antiquity to the Nineteenth Century (Cambridge, Cambridge University Press, 2016). von Glahn, Richard, “Modalities of the Fiscal State in Imperial China,” Journal of Chinese History 4.1 (2020), 1–29. Watanabe Shinichiro¯ 渡辺信一郎, Chu¯goku kodai no zaisei to kokka 中国古代の財政と 国家 (Tokyo, Kyu¯ko shoin, 2010). Xiong, Victor Cunrui, “The Land Tenure System of Tang China: A Study of the Equal Field System and the Turfan Documents,” T’oung Pao 85.4–5 (1999), 328–90. Yamada Katsuyoshi 山田勝芳, “O¯ Mo¯ no zaisei” 王莽の財政, Shu¯kan to¯yo¯gaku 集刊東 洋学 33 (1975), 63–85 Yamada Katsuyoshi 山田勝芳, Shin Kan zaisei shu¯nyu¯ no kenkyu¯ 秦漢財政収入の研究 (Tokyo, Kyu¯ko shoin, 1993). Yang Jiping 杨际平, Beichao Sui Tang juntianzhi xintan 北朝隋唐均田制新探 (Changsha, Yuelu shushe, 2003). Yang Zhenhong 杨振红, Chutu jiandu yu Qin Han shehui 出土简牍与秦汉社会 (Guilin, Guangxi shifan daxue chubanshe, 2009). Zang Zhifei 臧知非, Qin Han fuyi yu shehui kongzhi 秦汉赋役与社会控制 (Xi’an, San Qin chubanshe, 2012).

130

https://doi.org/10.1017/9781108587334.004 Published online by Cambridge University Press

4

Money, Markets, and Merchants y o¯ h e i k a k i n u m a

Introduction Metallic coinage, markets, and private merchants appeared in China during the Spring and Autumn era (771–453 B C E ), and they expanded rapidly during the Warring States era (453–221 B C E). These periods were marked not only by rapid economic progress, but also by new conceptualizations of money, markets, and merchants. Both in times of political stability such as the Qin, Han, Sui, and Tang dynasties, and in times of political disunion such as the Warring States, the Three Kingdoms, the Jin, and the Northern and Southern Dynasties eras, money, markets, and merchants performed important economic and social roles. Which kinds of goods, then, served a monetary function from the Warring States to the Tang period? How did people use money, and how and where did they buy and sell commodities? What was the relationship between private merchants and governments? In this chapter, these issues will be examined using transmitted documentary records, archaeological materials, numismatic findings, and recently excavated texts.

Money Coinage in the Warring States Period The origin of money in China dates from the mid-Spring and Autumn era. Cowry shells have been widely considered the oldest Chinese currency, but recent studies cast doubt on this claim. Cowries were highly valued by the ruling classes of the Shang (c. 1550–1045 B C E ) and Western Zhou (1045–771 B C E) dynasties, for whom cowries symbolized life and rebirth. Inscriptions on early Zhou bronze vessels describe kings rewarding their vassals with gifts of

131

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

cowries. Cowries strung into necklaces served as measures of value, although the number of cowries that comprised a necklace was unfixed. But bestowals of cowries took the form of gift giving, a means of strengthening ties with the vassal clans, rather than economic exchange. The idea of monetary exchange (i.e., determining values of daily goods by reference to an abstract medium) did not exist at that time. The word mai 賣 (“sell”) is not attested before the Qin period, and in Western Zhou times the word mai (買), which later came to mean “buy,” denoted redemption or recompense. The first metallic (bronze) currencies, in both spade-shaped bu 布 and knife-shaped dao 刀 forms, emerged in the Central Plain region, probably in the late sixth century B C E. From the fifth century B C E, circulation of both types of currency expanded even as their forms and sizes changed. The circulation of spade-shaped currencies linked the Central Plain to the Zhao and Yan economic zones to the north. These currencies varied in terms of size and weight, and many bore inscriptions denoting the names of cities and of those who cast the currencies. Therefore these currencies appear to have been minted by private merchants in specific cities.1 The knife-shaped currencies that circulated in the northeastern states of Yan and Qi in the Warring States era varied less in terms of form or inscription than spade-shaped currencies did, and thus are regarded as standard currencies issued by the state. Cowry-shaped bronze pellets (yibiqian 蟻鼻錢) began to be used in the Chu state located in the middle basin of the Yangzi river, in the mid-Warring States era, and 94 percent of them had the same mint mark. This suggests that the Chu state was a driving force in the large-scale circulation of this currency. Two types of round coin, with either round or square holes, also circulated in the Wei, Yan, Qi, and Qin states, most of which were deemed state currencies. Only in the state of Qin, which began to mint its banliang 半兩 coin in the 330s B C E, did round coins become the standard form of currency (Map 4.1).

Banliang Coin in the Qin and Han Periods After the Qin created a unified empire in 221 B C E , the banliang coin became the official currency throughout the Chinese world. Essentially, banliang were bronze coins – that is, made of an alloy of copper and tin – but they contained traces of other metals. Privately minted banliang coins sometimes included large amounts of lead. The largest banliang coins had a diameter of 1

Emura Haruki 江村治樹, Shunju Sengoku jidai seido kahei no seisei to tenkai 春秋戦國時 代青銅貨幣の生成と展開 (Tokyo, Kyu¯ko shoin, 2011), pp. 253–312.

132

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

Ri ve r

YAN

Ye

llo w

ZHAO

ow ell

Y

er Riv

QI

WEY QIN

HANN

SONG WEI Yellow Sea

Ya

CHU iver zi R ng

r ive iR z ng

Ya

Spade currency zone Knife currency zone Qin round coin currency zone Chu mixed currency zone

0

100 200 km

Map 4.1 Monetary regions in the Warring States period Source: redrawn from Richard von Glahn, The Economic History of China from Antiquity to the Nineteenth Century (Cambridge, Cambridge University Press, 2016), Map 2.3

ten centimeters, but on average they were in the range of two to three centimeters. Banliang coins were inscribed with two graphs, “half” and “ounce,” which signified their nominal weight, equivalent to approximately eight grams. In reality, the weight of banliang coins varied, and gradually lessened over time. The proliferation of much lighter, privately minted counterfeit coins was one reason, but another was that the government reduced the weight of officially minted coins, and the Han regulations preserved among the Zhangjiashan manuscripts (dating to the beginning of the Han era) accepted lighter coins as valid. Thus the name “banliang” did not signify the actual or legal weight of the coin, but instead functioned as a mark 133

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

of legal tender. The value of coins was measured by count (hereafter called the tale principle) rather than their intrinsic metallic content. In the Warring States period, when each state or city minted coins of various forms and weights, there was no unified unit of account based on the tale principle. After banliang coins began circulating in the late fourth century B C E, the Qin government strongly promoted the unification of currency within its territory. The Qin also prohibited the transport of their coins beyond the state’s borders. As a result, the tale principle began to function. Banliang coin became a nominal money, whose value was based on the signification of the “banliang” inscription. The independence of the value of the banliang coin from its metallic content enabled the Qin to create a monetary system separate from those of other states. From the beginning, banliang coins were minted as currency of the state, by the state, and for the state. Coins such as the banliang facilitated the collection of poll taxes and state expenditures. Banliang coins were useful for officials when buying commodities at markets, for paying salaries to officials and bonuses to soldiers for meritorious service, and for payment of fines and redemption of debts for violations of the law. Banliang coins thus were minted to serve as a means of state payment, which meant they had a relatively strong “internal” character. However, the coins also quickly penetrated the private sector of society. Although the range of circulation of banliang coins was limited to the Qin state and the government intended the coins to serve its own fiscal needs, the Qin populace also employed the coins for their own purposes as a means of exchange. The Han dynasty initially retained banliang coins, but repealed the state monopoly on coinage and permitted the private minting of banliang coins. Han law prescribed minimum standards for privately issued coins; however, the circulation of coins lighter than the nominal weight was still tolerated. There are three possible explanations for this leniency. First, circulation of banliang coins had been largely restricted to the Qin realm during the Warring States period, but after the Qin unification of the Chinese ecumene the demand for these coins in the conquered territories increased sharply, putting pressure on the limited supply of official coins. Second, private coins were minted by powerful local families, and the first Han emperor, Liu Bang (Emperor Gaozu, r. 202–195 B C E), embroiled in civil war against his rival Xiang Yu, might have given them permission to mint coins in order to acquire their political support. Third, the poll taxes levied on male and female adults, introduced in 203 B C E, required an ample supply of coin in circulation. 134

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

Banliang coins spread rapidly in the first half of the Western Han period. However, in the long run allowing private coinage became untenable. Private coiners minted smaller, lighter, and lower-quality coins in order to realize seigniorage profits. Thus heated debate broke out at the Han court over the issues of banning private coinage, managing the weight of coins, and permitting private mining. The government’s policy of accepting lightweight coins based on the tale principle proved difficult to sustain. From the reign of Emperor Gaozu or Emperor Hui (195–188 B C E ) to 175 B C E the government halted private coinage in most regions of the empire, and after 144 B C E a complete interdiction of private coins was instituted.

Private Coinage and Copper Shortages The chaotic monetary situation was ameliorated after the introduction of the new, heavier (c. 3.2 gram) wuzhu 五銖 (“five zhu,” a weight nominally equivalent to five-sixteenths of the banliang “half-ounce”) coin in 119 or 118 B C E. The production of wuzhu coins was confined to mints at the capital, Chang’an, which helped to ensure uniform quality of coinage. Over the next century, down to the reign of Emperor Ping (1 B C E – 6 C E), the Han state officially minted 28 billion coins. Nominally, the wuzhu coin standard was maintained for the next 700 years, until the Sui dynasty (581–618 C E), but debates over whether the coin should be heavier or lighter and whether private coins should be allowed persisted. At the end of the Western Han period, an anomaly occurred when Wang Mang, acting in his capacity as regent, advanced large-scale monetary reforms. In 7 C E, Wang’s administration minted not only wuzhu coins, but also new bronze currencies with arbitrarily ascribed values: the daquan wushi 大泉五十 coin, with a value of fifty wuzhu coins; the qidao wubai 契刀五百 currency, which combined the round coin with the archaic knife shape, valued at 500 coins; and the yidao pingwuqian 一刀平五千 currency, valued officially at 5,000 coins. At the same time the state prohibited private possession of gold. In 9 C E, when Wang Mang deposed the current Han boy emperor and established his own Xin (“New”) dynasty (9–23 C E), the government banned the wuzhu, qidao wubai, and yidao pingwuqian currencies and began to mint a new standard xiaoquan zhiyi 小泉直一 coin, to which the existing daquan wushi coins were linked at a value of fifty to one. These coins were minted both by the central government and by local commanderies and counties. In 10 C E, the government made the xiaoquan zhiyi coin the minimum unit of account and minted coins equivalent to ten, twenty, thirty, forty, and fifty xiaoquan zhiyi. The value of one jin (approximately 240 grams) 135

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

of gold was fixed at 10,000 coins. Silver, tortoiseshells, cowry shells, and bolts of hemp were also officially recognized as currency. However, in reality, with the exception of xiaoquan zhiyi and daquan wushi, these other types of money failed to circulate. In 14 C E, the xiaoquan zhiyi coin was abolished, and Wang Mang introduced two new currencies, the huoquan 貨泉, a round coin, and the huobu 貨布, in the shape of archaic spade currencies. One huobu coin was equivalent to twenty-five huoquan coins. It was also determined that daquan wushi would be allowed to circulate at par with the huoquan for an additional six years, after which it would be demonetized. Wang Mang’s currency reforms had five main features. First, wuzhu coins were abolished. Second, the bronze currency monetary system was tied to a gold standard. Third, concurrent currencies existed based on the principle of nominal (face) value, not weight. Fourth, the government prohibited private coinage, and even private possession of copper. Fifth, coins were minted by local governments as well as in the capital to increase the quantity of coin in circulation. The issue of coins in multiple denominations was based on a monetary theory – metaphorically dubbed “maintaining the balance between mother and child” (zimu xiangquan 子母相權) – whereby the state manipulated the relationship between principal (heavy, or “mother”) and subsidiary (light, or “child”) currencies in order to stabilize their relative values. The theory, expressed in the “Heavy and Light” chapters of the Guan Zi, that the value of money was inversely proportional to the price of goods, also influenced these reforms (see von Glahn, Chapter 5 in this volume). Wang Mang’s drastic reforms and repeated reversals within a short time caused confusion and disruption in the market. After the restoration of the Han dynasty in 25 C E, all of Wang Mang’s novel currencies were abandoned, and the wuzhu coin was restored as the standard currency of the dynasty. It was once thought that the chaos induced by Wang Mang’s reforms resulted in regression to a natural or barter economy in the Eastern Han (25– 220 C E) period. However, wuzhu coins still circulated, and in fact the demand for wuzhu coins was even stronger than during the Western Han period. At the end of the Eastern Han period, the rulers in the Central Plain abolished the poll taxes collected in coin and replaced them with a new “household levy” (hudiao 戶調) tax system that assessed household taxes in bolts of cloth. Nevertheless, wuzhu coins still were employed in market transactions because of their utility as a means of exchange, and the emperor and his officials continued to be concerned about how to control them. In the Three Kingdoms period (220–280), wuzhu coins were used not only in the Wei kingdom in the Central Plain, but also in the other Three Kingdoms states, 136

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

Figure 4.1 Standard (left), “clipped” (center), and “ring” (right) wuzhu coins

Shu Han in Sichuan and Wu in the Yangzi river basin. Furthermore, the latter two governments attempted to defray war expenditures by minting coins with high face values equivalent to 100, 500, or even 1,000 wuzhu coins. After the fall of the Han dynasty, few governments commanded the resources to sustain coinage. Yet the rising demand for coin that accompanied expansion of commerce – especially in the Southern Dynasties of the Period of Disunion (317–589) – intensified the problem of monetary shortages. Many people resorted to filing metal either from the outer edge of wuzhu coins – creating a “clipped” (jianlun剪輪) coin – or from the center square hole, resulting in a “ring” (yanhuan 綖環) shape (Figure 4.1). The circulation of such debased coins was restricted to the private sector, while full-bodied coins circulated among the officials or the upper classes. The Liu Song government (420–479) attempted to reduce the weight standard of coin by issuing sizhu 四銖 (“four zhu,” i.e., four-fifths of the weight of the wuzhu coin) coins. However, the quantity of these new coins remained small, and wuzhu coin remained the main currency in circulation. The short-lived dynasties of north China in the Period of Disunion issued numerous new coins that featured distinctive innovations in nomenclature. The fenghuo 豐貨 (“bountiful currency”) coin minted by the Later Zhao in 319 marked the first time that a coin was inscribed with an auspicious apothegm in place of the units of weight used since the founding of the Qin dynasty. This practice subsequently was revived during the Tang dynasty. The liangzao xinquan 涼造新泉 (“new coin created by Liang”), issued c. 314–320, was named after the state itself, a common practice in the Warring States period that ceased with the issue of the Qin banliang coin. This practice implies that it was the state’s authority that authenticated the credibility of its coin. The hanxing 漢興 coin issued by the Cheng Han state c. 339–343 was the first to be named after the reign era in which it was issued, a practice that became standard from the Song dynasty (960–1279) onward. Privately minted “local

137

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

coins” (tuqian 土錢) also appeared in some regions, as well as blank coins without any inscription. The custom of recognizing 1,000 coins as a standard monetary unit (guan 貫, “string”) became widespread at the end of the fifth century. Stringing together 100 coins as a standard unit of currency (mo 陌) can be traced back to the Han period, and the mo unit was also widely used during the Period of Disunion. Coins varied in size, material, and inscription; thus counting different types of coin according to a standard tale value became necessary.2 The practice of designating fewer than 100 coins a “short string” (duanmo 短陌) equivalent in value to 100 coins also appeared at this time, although scholars disagree about its origins. Despite the profusion of new coinages, the wuzhu coin remained the monetary standard – and Han wuzhu coins or imitations of them comprised the majority of circulating coin – until the Tang dynasty (618–907) established a new standard with the issue of its Kaiyuan tongbao 開元通寶 coin in 621. Dispensing with the former practice of inscribing coins with units of weight, the Tang government intended that its coin should circulate based on the tale principle. The Kaiyuan tongbao was cast in mints scattered across the empire, but the mints maintained uniformity in the size and weight of coins. The government minted 200,000 guan per year on average, rising to 327,000 guan per year in the Tianbao era (742–755).3 Nonetheless, this quantity of coin was insufficient to meet the needs of the burgeoning commercial economy. Private coins were outlawed, but such prohibitions had little effect. An edict of 660 decreed that private coins would be accepted at one-fifth of the value of official Kaiyuan tongbao, but in the market good and bad coins often circulated at equal value. A new Tang coin, the Qianfeng quanbao 乾封泉寶, was introduced in 666 with a face value equivalent to ten Kaiyuan tongbao or privately minted coins. The government intended to replace the Kaiyuan tongbao and privately minted coins with this Qianfeng quanbao coin, but this effort failed and the Qianfeng quanbao was withdrawn six months later. In 723, in an attempt to resolve the problem of shortages of metal needed for minting coin, the government banned its subjects not only from buying and selling copper and tin, but also from making bronze vessels. However, low-quality, privately minted coins continued to circulate, primarily in the Yangzi delta region, but also in the capital, Chang’an.

2

3

Miyazawa Tomoyuki 宮澤知之, “Gishin Nanbokucho¯ jidai no kahei keizai” 魏晉南北 朝時代の貨幣經濟, O¯ryo¯ shigaku 鷹陵史学 26 (2000), 41–82. Denis Twitchett, Financial Administration under the T’ang Dynasty (Cambridge, Cambridge University Press, 1970), pp. 66–83.

138

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

In 780, the Tang government replaced its principal form of taxation, known as zu–yong–diao 租庸調 after its three principal components (payments in grain, cloth, and labor service), with the twice-a-year tax (liangshui 兩税), which was paid primarily in grain and coin. In addition, the demise of its fubing 府兵 militias, which compelled the Tang to recruit mercenary armies, and the outbreak of the An Lushan Rebellion (755–763) caused military expenses to soar to a level six or seven times greater than in the Kaiyuan era (713–741).4 Demand for coin increased, and even though the production of private coin rose as well, shortages of coin resulted in ruinous deflation. Thus the government converted the coin portion of the twicea-year tax to payments in silk and hemp cloth, and allowed the use of silk textiles as well as coin as means of payment for large transactions. Furthermore, the government minted large coins equivalent to ten and fifty Kaiyuan tongbao coins to cover its military expenditures, but found it impossible to enforce the official exchange ratios. In 845, Emperor Wuzong (r. 814–846) abruptly proscribed the practice of the Buddhist religion and ordered the seizure of monastic properties. Bronze statues, bells, and ritual objects confiscated from Buddhist temples were melted and used as raw material for a new issue of coin, the Huichang kaiyuan 會昌開元 coin, named after the current Huichang (844–846) reign era. However, Wuzong died less than a year later, and his anti-Buddhist policies were rescinded. The Huichang kaiyuan coins were quickly withdrawn from circulation as well. Subsequently, the monetary system centered on the Kaiyuan tongbao coin was restored and continued to the end of the Tang period, although the endemic shortage of coins could not be resolved.

Multiple Currencies and Their Relationships From the Warring States to the Tang period, other goods in addition to coins also functioned as a medium of exchange, a store of value, or a unit of account, depending on the situation. These currencies can be thought of as supplementing or complementing bronze currencies. In the Warring States period, gold was frequently used as currency, especially in the Chu state. Chu issued sheets of gold-leaf currency stamped with Ying 郢 and Chen 陳, the names of its capital cities. When necessary, the sheets could be torn into pieces and used as a fractional currency measured by weight. After the Qin conquered Ying in 278 B C E, Chu gold currency 4

Watanabe Shinichiro¯ 渡邊信一郞, Chu¯goku kodai no zaisei to kokka 中國古代の財政と 國家 (Tokyo, Kyu¯ko shoin, 2010), pp. 467–515.

139

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

circulated in the Qin territories as well. Administrative documents from the Qin and Han periods referred to standard bronze coins, bolts of hemp cloth, and gold as “current” monies (xingqian 行錢, xingbu 行布, and xingjin 行金 respectively). The basic unit of weight for gold in the Chu and Qin states originally was designated yi 溢 (approximately 240 grams), but it was renamed jin 斤 at the beginning of the Han period. There was an equation of weights: 1 jin = 16 liang 兩 = 384 zhu 銖. The typical shape of gold currency also was changed from a square plaque (jinban 金版) to a round plaque form (jinbing 金餅). Bolts of textiles also functioned as measures of value and means of exchange in commodity trade. As mentioned above, Qin–Han administrative and legal texts referred to standard bolts of cloth as “current” monies. “Current cloth” was defined as high-quality cloth with a uniform size measuring eight chi (185 centimeters) in length with a width of 2.5 chi (fiftyeight centimeters). According to Qin laws of the Warring States period, merchants in the marketplace and officials were obliged to accept bolts of cloth as well as coin as payments. In the Han, after the reign of Emperor Wu (r. 141–87 B C E), purchases of salt and iron from the state monopoly bureaus were mainly paid for with bolts of hemp textiles. We also find cases of purchasing commodities such as slaves, grain, soybeans, and land with gold and textiles as well as coin.5 Shortages of coin also compelled the government to pay the salaries of its officials in bolts of cloth, at least during the Wang Mang era and the Period of Disunion. However, we cannot ignore the fact that the fungibility and versatility of coin was stronger than that of gold and textiles. Gold was a high-value currency, with one jin of gold equal to several tens of thousands of coins. Thus gold was inconvenient for market exchange. Bolts of hemp textiles also were higher in value than coin, but they were far cheaper than gold. As will be discussed later, Qin legal texts set the price of one bolt of hemp cloth at eleven coins, which suggests that bolts of cloth were not too expensive for buying daily goods, although evidence for such purchases remains elusive. Bolts of cloth also had the disadvantage that they could not be cut into fractional units without destroying their value. The importance of textiles as a means of tax payment rose after the Western Han period. The Eastern Han government often resorted to irregular requisitions (diao 調) to raise funds for military expenditures. In contrast 5

See the examples in Ding Bangyou 丁邦友 and Wei Xiaoming 魏曉明 (eds.), Qin Han wujia shiliao huishi 秦漢物價史料匯釋 (Beijing, Zhongguo shehui kexue chubanshe, 2016).

140

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

Table 4.1 Means of payment in imperial emoluments

Gifts on auspicious occasions Retirement bonuses Get-well gifts Awards for meritorious military service Bounties for capturing criminals Compensations for forced migration Gifts for foreigners Redemptions Gifts to scholars Presents for wedding celebrations Charity for the socially vulnerable Farewell gifts

Western Han

Eastern Han Jin

Song to Chen

Silk or gold

Silk or gold

Silk

Silk

Gold Silk Coin or gold

Coin or silk Coin Coin

Coin or silk Coin or silk Silk

− Coin Coin

Coin or gold

Coin or gold Hemp, silk, gold Coin Coin

Coin Gold or silk

− −

Gold Silk? Coin or gold

Gold or silk Gold or silk Hemp and silk Silk Hemp − Silk? Silk Silk Silk Coin or silk −

Silk

Silk

Silk

Silk

Coin

Coin

Coin

−

to the prevailing Han system of suanfu 算賦 levies, in which a uniform poll tax was assessed on individual adults and paid in coin, the diao levies were assessed on households ranked according to wealth and paid in various types of cloth. By the end of the Han dynasty, a new tax levy based on payments of grain and cloth assessed on households (hudiao 戶調) had replaced the poll tax levied in coin altogether. The obligation to pay taxes in cloth shifted much of the burden of taxation to household production (mainly by women) of textiles. The hudiao tax system, which was retained by the various regimes of the Period of Disunion, diminished the need to market farm produce to obtain coin, while encouraging the formation of conjugal households whose collective labor could meet the state’s demand for grain and textiles. In addition to the growing importance of textiles as a means of state payments in the post-Han period, cloth as well as bronze coins served as means of exchange in private transactions. From the Western Han to the Period of Disunion, gold mostly disappeared from imperial emoluments, while benefactions in textiles rose (Table 4.1).

141

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

Silk textiles and gold also had an important function as international currencies. Sogdian merchants from West Asia dominated the Silk Road trade from the third century C E onward (see Liu’s chapter in this volume). Correspondence from Sogdian merchants who settled at the oasis town of Dunhuang shows that they used gold as currency.6 In fact, it was legally forbidden to carry bronze coins abroad, at least in the Han and Tang periods. Specimens of Chinese coins have been excavated from some oasis town sites, but most of them are relics left by Chinese during times when they ruled those regions. Although some local rulers of Central Asian oasis towns issued imitations of Chinese bronze coins, their circulation was highly circumscribed. Gold and silver coins along with bolts of silk served as the currencies of international trade across the Central Asian trade routes. Silver coins from Western Asia, such as the drachm issued by the Sassanid rulers of Persia, have been found along the northern frontiers of China, and pearls also functioned as means of exchange in the coastal regions of the far south before the Tang period, but neither circulated as currency in the Chinese heartland. During the reign of Emperor Yang of Sui (r. 604–618), the government replaced the hudiao taxes assessed on households with poll taxes paid in grain, labor services, and cloth, which as mentioned earlier became known as the zu–yong–diao tax during the Tang dynasty. Although the government did not collect bronze coin payments through the zu–yong–diao taxes, merchants paid some taxes in coin, which also was disbursed as a portion of official salaries. During the Tianbao era (742–756), half of the silk textiles collected as tax payments were allocated for military expenditures such as soldiers’ salaries and purchase of provisions at the frontier garrisons in Central Asia. The Tang government also allowed the use of silk textiles as well as coins as means of payment in market transactions. Moreover, the prominence of silk textiles as an international currency continued to grow following the incorporation of large parts of Central Asia into the Tang Empire in the early seventh century. The Tang government established large horse ranches in the commanderies of Longxi, Jincheng, Pingliang, and Tianshui northwest of the capital, Chang’an, which altogether maintained herds of approximately 700,000 horses. But these areas were lost after the An Lushan Rebellion, and the Tang were forced to rely on purchases of horses from their Uyghur allies in exchange for silk textiles through “mutual trade” (hushi 互市) at frontier markets. 6

Étienne de la Vaissière, Sogdian Traders: A History (Leiden, Brill, 2005), pp. 13–70.

142

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

In sum, the monetary economy from the Warring States to the Tang periods featured the use of concurrent currencies, including gold and textiles as well as the bronze coins that constituted the official state-issued currency.

Markets The Birth of Walled Markets Many questions remain in terms of the existence of markets before the Spring and Autumn period. Some scholars have interpreted a word that appears in Shang oracle bone and Western Zhou bronze inscriptions as referring to a market, but its interpretations remain controversial. The word shi 市, which becomes the common Chinese word for “market,” appears in the Book of Odes in a context indicating that it was a site located at the entrance gate of a community where men and women gathered for romantic trysts, but there is no specific mention of economic exchange in these lyrics. The word shi appears in a number of Warring States texts, such as the Chronicles of Zuo, Mencius, and Mo Zi. According to these sources, markets existed at least by the seventh century B C E. Markets were located along streets and outside walled cities, and many strangers came to seek business opportunities at such places. Because of its location, the marketplace was also recognized as a liminal space between the community and the outside world. Thus execution of criminals was conducted at the marketplace, which symbolically marked their elimination from the community. Markets were portrayed as places frequented by strangers – not only itinerant merchants but also outcasts, fugitives, and perhaps even monsters – who potentially posed a danger to the community. Thus early Chinese narrative texts such as Biographies of Immortals (Liexian zhuan 列仙傳, first century B C E ?) and In Search of the Sacred (Soushen ji 搜神記, fourth century C E) represented the marketplace as a space where extraordinary phenomena frequently occurred. In the Warring States period, with the growing concentration of both political power and economic activities in urban centers, markets were gradually relocated within walled cities. Permanent markets under the supervision of governmental officials were constructed in metropolises and administrative capitals, and there were also fairs that would open at fixed intervals, such as once a week or a month, in rural villages. The permanent markets and fairs were closely related and formed local market areas. Most villages had already been walled since before the Warring States period, but their walls were not high. The center of a county or commandery consisted of

143

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

several walled neighborhoods (the same term, li 里, was applied both to urban neighborhoods and to rural villages), which were enclosed by a higher wall. Walled villages were also found here and there before the Han period, but these walled villages gradually disappeared after the third to fourth centuries C E. The basic unit of local administration since the Warring States period was the county (xian 縣), a walled city with at least one permanent market which itself was enclosed by walls. According to a speech by a statesman of the Wei kingdom dated to 260 B C E, the Wei “territories east of the capital at Daliang encompassed seventeen large counties, each with walls a thousand zhang in length and a population of ten thousand households, along with thirty-plus small counties with markets.”7 In the Han period, each county had 7,000 to 10,000 households on average, roughly comparable to the size of large counties in the Warring States Wei kingdom. In the Western Han capital of Chang’an, however, there were at least nine permanent markets, and each market was surrounded by a square wall with a perimeter measuring 367 meters. Permanent markets were divided into multiple alleys, or hang 行, in which shops were grouped by trade and merchandise. The term hang thus also became a designation for a specific trade or shop. The “Market Regulations” of the Qi state, dating from the Warring States period, describe the plan for constructing a state-supervised market (guoshi 國市), which according to modern scholars was based on the actual market of the Qi capital of Linzi.8 At the center of the market was a drum tower with a constabulary (ting 亭). Officials struck the drum to announce the opening and closing hours of the market. The size of shopfronts was regulated by the government, with larger spaces allotted to dealers in bulk commodities and smaller ones to shops purveying luxury goods. Under Qin imperial law, shopkeepers in each trade had to be enrolled in “market registers” (shiji 市籍) and were required to pay market taxes (shizu 市租) based on a system of self-assessment and reporting their income to their district offices. Although all permanent shops were listed in the “market registers” and the corresponding tax rolls, it is unclear whether itinerant peddlers were also so registered. According to the market regulations of the early Han dynasty, in addition to grouping shops according to categories of 7 8

Chapter 26, “Jian Tian Bing yu Liangnan zhang” 見田X於梁南章, in ZGZH, 165. Yinqueshan hanmu zhujian zhengli xiaozu 銀雀山漢墓竹簡整理小組, “Yinqueshan zhushu ‘Shoufa,’ ‘Shouling’ deng shisan pian” 銀雀山竹書《守法》《守令》等十三篇, Wenwu 文物 1985.4, 32.

144

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

business, each hang had a “chief of the market lane” (liezhang 列長). The chief of the market lane ensured that each shopowner reported his income and met his tax liabilities; failure to do so would result in fines for the chief as well.9 Governmental officials were charged with maintaining close surveillance over markets. They regularly inspected shops for workmanship and prices to ensure that consumers would not be cheated. Shopowners were obliged to attach price tags to their goods. Itinerant peddlers without permanent shops might simply gather at crossroads or pitch temporary tents in a market. Itinerant peddlers and ordinary citizens selling their products at crossroads catered to daily needs, while market shops traded in luxuries as well as everyday goods.

The Three-Tiered Price System When government officials made purchases in the market, they utilized a three-tiered price system: first was the fixed official price, stipulated by government decree; second was the “adjusted price” (pingjia 平價 or zhengjia 正價), an adjustment of the official price based on monthly or yearly variations of prices in each locality; and third was the actual market price at a given moment in time (shijia 時價). Officials were obliged to investigate market conditions to determine which price would be used to make government purchases. If the actual market price differed substantially from the fixed official price, officials were required to adopt the adjusted price, which corresponded to seasonal fluctuations in market prices. In private trade, of course, the market price prevailed. This three-tiered price system for government purchases, which had been in operation since the Warring States period, endured until the Qing period. In contrast to the fixed official price and the adjusted price, the actual market price fluctuated in markets according to changes in supply and demand. Differences in the quality and quantity of product information among buyers and sellers were an underlying cause of increases or decreases in market prices for commodities. The information asymmetry between buyers and sellers could result in violent fluctuations in prices. However, the value system did not always fall into chaos, in my opinion, because of three features of the market system from the Warring States to the Tang: the organization of the marketplace, customer relations, and price tags. First, the permanent markets were enclosed by walls, and business dealings were 9

Anthony J. Barbieri-Low and Robin D.S. Yates, Law, State, and Society in Early China: A Study with Critical Edition and Translation of the Legal Texts from Zhangjiashan Tomb no. 247 (Leiden, Brill, 2015), 2, pp. 721–3.

145

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

concentrated in a single space. Shops in the same trade were grouped together in alleyways, and this proximity made it easier for customers to compare the quality and price of goods among different shops selling the same wares. Accordingly, biases in information were minimized. Second, buyers and sellers frequently formed habitual relationships and rapport with each other. Because of these relationships, sellers would not cheat buyers, and the product prices would not fluctuate violently. Third, as a general rule, all commodities costing more than one coin were required to have a price tag, so prices were publicly declared to all customers. These three features reduced the problem of asymmetrical information, and they seem to have brought about relatively fair price competition. Incidentally, in cases of high-value transactions such as selling houses or livestock, arguments between buyers and sellers readily arose. For this reason, brokers (kuai 儈) who performed the role of business transaction mediators appeared in the Han period. Brokers were chosen based on their high repute for honesty and business acumen, and when a deal was being brokered they took a commission and ensured that the transaction concluded smoothly and fairly.

Official and Private Markets In the Han period, there were also fairs that would open at fixed intervals, such as once a week or once a month, in rural areas. As mentioned above, there was one permanent market in each large county town, but few villages had their own permanent markets. Periodic fairs suited the minimal need for villagers to exchange goods. Some sources refer to individuals engaging in business even inside villages, but they are few in number, and seem to be exceptional cases. According to Wang Fu (c. 78–163 C E), the number of local markets sharply increased in the Eastern Han period.10 Markets including fairs were closely related to each other, and formed small local market areas. The satirical literary work entitled “Slave Contract” (Tongyue 僮約) written by Wang Bao (?–61 B C E ) illustrates local market activities in the early Han.11 This document purports to be a contract that dictates the terms of service for a newly acquired slave who had been insolent toward his previous master. The contract specified that the slave was obliged to go to a ting 亭 to buy straw mats; as noted above, ting referred to the drum tower constabulary placed at the center of a marketplace, and so signified the 10 11

Chapter 12, “Fuyi” 浮侈, in QFL, 120. QHW, 42.11b–12b. The full text of the mock contract is translated in Cho-yun Hsu, Han Agriculture (Seattle, University of Washington Press, 1980), pp. 231–4.

146

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

market itself. The slave also was ordered to go to Chengdu and Luo (probably Luoxian, near Chengdu) to buy cosmetics for women and sell them in smaller markets. In addition, he was told to return to Chengdu with hemp cloth, and buy and sell dogs and geese on his way. He also was expected to go to Wuyang (fifty kilometers south of Chengdu) to buy tea, and bring swords and arrows to Yizhou (a frontier region far to the southwest of Chengdu) in return for cattle and sheep. Although a work of fiction, the “Slave Contract” sketches a plausible geography of market circulation connecting urban and rural markets in the Chengdu plain region of Sichuan and beyond. There were both walled permanent markets and fairs in the Period of Disunion as well. Urban centers still consisted of several walled neighborhoods (li 里), or wards (fang 坊), and urban markets were also surrounded by walls. Several marketplaces sometimes existed inside the same walled city. For example, in Jiankang (modern Nanjing), the capital of the Southern Dynasties in the fifth to sixth centuries, there were four large walled markets, multiple smaller markets, and specialized markets for fodder, cattle and horses, grain, and raw silk.12 During the Western Jin era (280–317), Luoyang had a “gold market” (jinshi 金市), a “southern market” (nanshi 南市), and a “horse market”(mashi 馬市); after Luoyang became capital of the Northern Wei in 493 it also included a “great market” (dashi 大市), a “lesser market” (xiaoshi 小市), and a “market hub for the four directions” (sidashi 四達市). Smaller cities, such as Wan (Henan) and Xiangping (Liaoning) in the Three Kingdoms period and Pengcheng (Shandong) in the Northern Wei period, also had multiple markets within the walled city. In the Sui and Tang periods, the so-called “ward-and-market” (fangshi 坊市) system of organizing urban space was rigorously enforced. The capital, Chang’an, had at least seven markets within the walled area of the city. Major business activities were mostly concentrated in the two great walled markets, the eastern market and the western market. Foreign traders, notably Sogdians, clustered in the wards around the western market, the terminus of the Silk Road trade route. In contrast, the eastern market lay near the residential quarters of the royal, noble, and powerful in the first half of the Tang era. Many merchants, candidates for the imperial examinations, and agents of the regional military governors took up residence near the posh eastern market.

12

Liu Shufen 劉淑芬, Liuchao de chengshi yu shehui 六朝的城市與社會 (Taipei, Xuesheng shuju, 1992), pp. 135–65.

147

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

After the An Lushan Rebellion, however, the residential wards neighboring the eastern market began to experience obvious business growth.13 Chang’an’s western and eastern markets were each roughly one square kilometer in area and crossed by two sixteen-meter-wide streets running north–south and east–west. Recently archaeologists have excavated large quantities of relics (such as pottery and coins) and construction remains (such as drainage ditches, rutted roads, slab stone bridges, and wells) at the site of the western market.14 People also could buy and sell goods in ordinary wards, and street stalls (lisi 里肆) were set up to provide food and drink, such as bread, fruit, wine, and tea. It was officially forbidden to enter the streets outside the wards after the evening curfew, and people who were late returning to their own ward were obliged to remain within other wards until the next morning. Certain neighborhoods of Chang’an were noted for concentrations of specialized shops. For example, most musical instrument stores were located in Chongren ward, drapers operated shops in Xuanyang ward, and female dancers congregated in the adjacent Pingkang ward. Yanshou ward was famous for shops trading in jade, gold, and silver wares. In contrast to the “ward-and-market” system of organizing urban space, villages and markets in rural areas were not subject to stringent controls. Rural markets were known by terms such as “assembly markets” (jishi 集市), “rustic markets” (yeshi 野市), “village markets” (cunshi 村市), or “small markets” (xiaoshi 小市), depending on colloquial usage. Names of rural markets often reveal their location at fords or riverbanks (heshi 河市, shuishi 水市, bushi 歩市), where traveling merchants could readily unload their wares, or at the foot of hills (shanshi 山市), where products of the plains and the mountains were exchanged. Some designations indicated markets that were held on a periodic basis; “hai-day markets” (haishi 亥市), for example, referred to markets held once in the ten-day Chinese week. Xushi 墟市 (literally “empty market,” referring to the fact that the market was held only on a few days each month, hence usually empty) became a common designation for periodic markets in southern China. The designation caoshi 草市 – originally meaning a market for fodder – became a widely used colloquial term for periodic markets in the Central Plain region from the fourth century onward. In 707, the Tang government prohibited local officials 13

14

Victor Cunrui Xiong, Sui–Tang Chang’an: A Study in the Urban History of Medieval China (Ann Arbor, University of Michigan Center for Chinese Studies, 2000), pp. 165–94. See Tang West Market Museum, “Preservation,” www.dtxsmuseum.com/en_pc/ news_pic_list.aspx?category_id=81 (accessed April 14, 2020).

148

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

from authorizing markets outside prefectural and county cities, but unofficial markets proliferated nonetheless.

Shops and Inns The practice of grouping shops by trade and merchandise in alleyways (hang) inside the permanent urban markets persisted in the post-Han era as well. There were 120 hang encompassing some 3,000 shops and 400 inns at the southern market in Tang Luoyang. According to an eleventh-century gazetteer of Chang’an, there were 220 hang at the Tang capital’s eastern market.15 Ennin, a pilgrim Buddhist priest from Japan, recorded in his diary that the great fire that broke out in Chang’an in July of 843 C E destroyed twelve hang in the capital’s eastern market and some 4,000 buildings.16 Although these figures may be exaggerated, they give some indication of the scale of the eastern market. According to a report from 742, the market at the Central Asian oasis town of Turfan featured separate hang for grain and wheat, rice and noodles, sweets, and vegetables, and a thousand kinds of goods were available for sale.17 Each shopkeeper was obliged to display a flag bearing the hang name. In the capital the imperial treasury (taifusi 太府寺) periodically checked the measures and scales used by merchants. Markets in the Tang period also featured inns (didian 邸店), which were composite businesses that provided lodging and food for traveling merchants, warehouse space to accommodate their goods, and brokerage services for negotiating with local clients. Some innkeepers expanded their business to provide transport services for their customers, or acted as consignment agents, holding the goods left by traveling merchants until buyers could be found. From the late Tang period, innkeepers also began to engage in wholesale business and moneylending, in some cases acting as deposit banks.18 After the An Lushan Rebellion, the urban ward-and-market system gradually declined, and nighttime curfews on movement or conducting business outside the wards were relaxed. As mentioned above, a variety of businesses had already existed outside official markets, but the relaxation of nighttime curfews in the late Tang period accelerated this trend, and many shops 15 16

17

18

CAZ, 8.11b. Ennin, Ennin’s Diary: The Record of a Pilgrimage to China in Search of the Law (trans. Edward O. Reischauer) (New York, Ronald Press, 1955), p. 333. Cited in Ikeda On 池田温, To¯shi ronko¯: Shizokusei to kindensei 唐史論攷:氏族制と均 田制 (Tokyo, Kyu¯ko shoin, 2014), pp. 655–760. Hino Kaizaburo¯ 日野開三郎, To¯dai teiten no kenkyu¯ 唐代邸店の研究, 2 vols. (Fukuoka, San’ichi shobo¯, 1992).

149

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

opened outside the walled markets along the main streets of towns (see Zurndorfer’s chapter in this volume).

Merchants Great Merchants and Multiple Economic Zones It remains unknown how merchants in the Shang and the Western Zhou periods actually worked in detail. The words gu 賈 and shang 商 both were commonly used to designate “merchants” in transmitted texts about the Spring and Autumn period, such as the Chronicles of Zuo, the Spring and Autumn Annals of Master Yan, and the Guan Zi, although these texts were actually composed in and after the Warring States period. For example, the Chronicles of Zuo records that in 627 B C E the Zheng merchant (shang) Xian Gao arrived at a Qin army camp while on his way to conduct trade in Zhou. Xian presented gifts of oxen and tanned hides to the Qin commander, but also secretly sent a messenger to warn Zheng of the impending attack.19 In 597 B C E, another merchant (gu) from Zheng reportedly attempted to help a Jin noble captured by Chu to escape by hiding him in a sack for clothing.20 The same merchant is mentioned as having traveled to Qi, and thus obviously was engaged in long-distance trade among the Zheng, Chu, Jin, and Qi states. In 526 B C E, the Zheng minister Zichan refused to intercede on behalf of a Hann noble who was bargaining with a Zheng merchant (here referred to as both gu and shang) over the purchase of a jade ring.21 Both gu and shang became much more common not only in the transmitted texts but also in the excavated texts of the Warring States period. For example, almanacs (rishu 日書) compiled for purposes of prognostication offer advice on propitious days for making purchases of goods from merchants (gu) in markets. Long-distance trade developed in response to growing demand for goods produced in specific regions. The “Tribute of Yu” chapter in the Book of 19

20

21

Xigong 33rd year, ZZ, p. 1833a. Cf. Stephen Durrant, Wai-yee Li, and David Schaberg (trans.), Zuo Tradition: Zuozhuan Commentary on the “Spring and Autumn Annals” (Seattle, University of Washington Press, 2016), 1, pp. 445–7. This episode is also recorded in the excavated text Xinian 繫年, a chronicle believed to have been composed c. 370 B C E, in the Tsinghua University collection of bamboo-slip texts. See Li Ling 李陵 (ed.), Qinghua daxue cang Zhanguo zhujian 清華大學藏戰國竹簡 (Shanghai, Shanghai wenyi chubanshe, 2011), vol. 2. Chenggong 3rd year, ZZ, 2, p. 1901a. Cf. Durrant, Li, and Schaberg, Zuo Tradition, 2, pp. 744–5. Zhaogong 16th year, ZZ, 2, pp. 2079b–80a. Cf. Durrant, Li, and Schaberg, Zuo Tradition, 3, pp. 1532–7.

150

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

Documents, which was probably compiled in the fourth or third century B C E, divided the Chinese world into nine regions according to the specialized tribute goods submitted from each region. Su Qin, a political strategist active in the late fourth century B C E, explained the specialty of each state as follows: in Yan, grain, horses, jujube, chestnuts, woolen fabrics, leather, dogs, and horses; in Qi, fish and salt; in Chu, citrus fruits; in Hann, arrows and swords; and in Wei, grain. Su Qin also asserted that Zhou people were famous for commercial activities and handicrafts.22 Su Qin’s brother, Su Dai, referred to a special purple dyeing technique for silk textiles in the Qi state.23 Sima Qian’s “Biographies of the Moneymakers,” compiled in the first century B C E, also followed the outline of the “Tribute of Yu” in dividing the Chinese world into economic regions and identified the specialized products of each.24 “Biographies of the Moneymakers” dwells in laudatory fashion on merchants who overcame numerous difficulties to reap great fortunes trading between different economic regions. Many of the merchants listed by Sima Qian developed congenial relationships with government officials in the regions in which they operated, because they could not go on long-distance journeys for trade without the requisite permission from the authorities and protection in times of war, as the evidence of the E Jun Qi tallies shows (see von Falkenhausen’s chapter in this volume). Within the same economic region trade developed more in response to seasonal variations in supplies of and demand for goods rather than comparative advantage and regional specialization. Sima Qian recounts the story of Bai Gui, a Luoyang merchant during the mid-Warring States period, who used the principle of seasonal variations – buying up grain, silk, and lacquer when they were abundantly available in the market, and selling in times of shortage – to amass a great fortune.25 In his Monthly Instructions for the Four Peoples (Simin yueling 四民月令), a manual on managing the household economy addressed to gentleman landowners, Cui Shi (c. 103–170 C E) provided detailed instructions on the proper seasons for selling farm produce such as millet, beans, and sesame and for purchasing goods for domestic consumption, including wheat, bran, glutinous rice (for the brewery Cui operated), hemp cloth, thread, and charcoal.26 Cui’s treatise shows that enterprising landowners sought to profit from seasonal variations in supply and demand. 22 25 26

SJ, 69, pp. 2243, 2245, 2250, 2254. 23 SJ, 69, p. 2270. 24 SJ, 129, pp. 3261–70. SJ 129, pp. 3258–9. Patricia Buckley Ebrey, “Estate and Family Management in the Later Han as Seen in The Monthly Instructions for the Four Classes of People,” Journal of the Economic and Social History of the Orient 17.2 (1974), 173–205.

151

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

Commercial Taxation and Restraints on Merchants In the early Han period the government imposed sumptuary laws regulating dress and vehicles in order to prevent merchants and other nouveaux riches from mimicking the nobility’s privileges of wearing luxurious silk clothing or riding in carriages. Merchants were also prohibited from owning land or houses inside a walled city. These restrictions also reflected the government’s bias against merchants as profit-seeking aggrandizers who exploited the needs of farmers, the true bedrock of the economy. However, these policies had loopholes. The Han government allowed the sale of merit ranks (which entitled the individuals to own certain amounts of land and slaves based on their rank) and even official positions to subjects who made donations of money or valuables to the government. These opportunities were frequently exploited by rich merchants. In addition, many merchants developed close relations with imperial princes who were awarded their own quasiindependent satrapies (guo 國) to govern. Such merchants thus were largely beyond the reach of the central government. In the reign of Emperor Wu, the exorbitant costs of the military expeditions the emperor launched against the Xiongnu steppe nomads caused severe fiscal distress. To raise funds for military expenditures, Emperor Wu authorized the state’s monopolization of the production and sale of salt and iron. Private enterprises in these two key industries were obliged to relinquish their facilities and equipment to the state, while at the same time private entrepreneurs were recruited as salaried managers to operate the state-run foundries and salt works. The state also assumed control of the distribution and sale of salt and iron throughout the dynasty. The Han government’s financial straits worsened after the collapse of the Yellow River dikes in 120 B C E, prompting Emperor Wu to broaden a tax originally imposed exclusively on merchants into a general levy on commercial enterprise. The new levy, known as suanmin 算緡, targeted all people engaging in commercial activities, regardless of whether or not they were enrolled in the “market registers.” The suanmin levy assessed taxes of 6 percent on merchant assets (3 percent in the case of artisans), and failure to pay the tax was severely punished. Those accused of tax evasion were subject to criminal penalties and confiscation of their property, and apparently many officials took advantage of this procedure to expropriate the assets of private merchants. The Yellow River floods also wreaked havoc with the transport system, and this became a further hindrance to commerce.

152

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

As a consequence of Emperor Wu’s policies, the wealth and power of the merchant class was substantially reduced, and long-distance trade atrophied. In the view of some scholars, wealthy families withdrew from commerce and invested in land, amassing large estates at the expense of farming families who were compelled to sell off their lands because of poverty or debt. However, private commerce was by no means extinguished, nor were landed estates wholly self-sufficient. As we can see from Cui Shi’s treatise, the great landowners sold a wide variety of agricultural products and handicraft goods, and obtained other household articles and luxury goods at markets. Independent farming families and tenants and hired laborers working on large estates also routinely engaged in trade with itinerant merchants at rural markets and fairs. Although the scope and scale of commerce had changed, market exchange remained a prominent feature of the economy. Under these circumstances, the ways in which the government supervised merchants also gradually changed. In the early Western Han period, as under Qin rule, professional merchants had to pay market taxes (shizu). Emperor Wu’s suanmin levy extended commercial taxation to cover anyone engaged in trade, but was still restricted to those with substantial commercial assets. In 10 C E, Wang Mang enacted a new “tribute tithe” (gongfa 貢法) of 10 percent of income on various professional classes, including doctors and ritual specialists, as well as shopkeepers and artisans, which was also extended to people who sold products in villages or inns. These initiatives show that the government sought to broaden the range of commercial taxation to cover all forms of exchange. But it is unlikely that they achieved the desired results.

Merchants in the Period of Disunion Despite the political turmoil of the post-Han era, there was no retreat to a wholly self-sufficient economy, and both local and long-distance trade continued. Rulers of the numerous polities spawned by the collapse of the Han Empire depended on commercial sources of revenue. In the wake of the devastating Yellow Turban Rebellion (184–189), the agrarian economy became seriously disrupted as many farming families fled the violence. In 196 the warlord Cao Cao set a puppet emperor on the Han throne and maneuvered to wrest control of fiscal resources. Among other measures, Cao seized control of the salt lake in Hedong district, and assigned “salt officers” (yanguan 鹽官) to supervise the production and sale of salt by merchants. Cao used the revenue from the salt tax from merchants to provide homesteads and cattle to displaced persons and stabilize the agrarian economy. In addition, Cao assigned tracts of abandoned land to both soldier 153

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

and civilian households as tenants (diannong 佃農) of state-managed agricultural colonies (tuntian 屯田). Because the diannong were subject to onerous taxation, the Wei kingdom established in 220 by Cao Cao’s son Cao Pi encouraged them to engage in handicrafts and trade to supplement their income. Subsequently many diannong became merchants, often with close ties to government officials. International trade also flourished during this period (See Liu’s chapter in this volume). During the Three Kingdoms period, many leading aristocrats, merchants, and foreigners gathered at Ye (in modern Hebei), one of the five capital cities of the Wei kingdom, which became a thriving commercial center. The Wei authorities reduced tariffs on commercial goods, opened the frontiers to caravans crossing the Central Asian trade routes, and ensured safe passage through their territories to Luoyang and Ye. The Shu Han kingdom in Sichuan also promoted commercial development and foreign trade. The “Slave Contract” cited above testifies to the development of market networks in the Chengdu basin in Han times, and this situation continued in the post-Han period as well. Zhuge Liang, chancellor in Shu Han, led military campaigns to the southwest (modern Yunnan) to acquire the region’s rich resources (especially minerals), and sent diplomatic embassies to cultivate ties with Central Asian city-states. The Shu Han state also relied heavily on Sichuan’s silk industry: silk and hemp textiles comprised a major portion of its revenues, and Zhuge Liang established statemanaged silk-manufacturing workshops whose products were exported to other regions. Following the nomadic invasions that forced the Jin court to abandon its capital at Luoyang in 311 and reconstitute its regime at Jiankang in the Yangzi delta, northern China descended into political anarchy and economic turmoil until the Tuoba clan of the Xianbei people began to restore order, ruling as the Northern Wei dynasty (386–534). Large numbers of Chinese fled to the south along with the Jin court, and the population of the Chinese heartland in the Central Plain fell sharply. Thus economic development in the north was seriously disrupted. The Tuoba, in keeping with nomadic practices, enslaved large numbers of war captives and imposed hereditary menial status on craftsmen, clerks, entertainers, and other persons with specialized skills. Following the precedent of the Wei and Jin, the Northern Wei collected revenues in the forms of grain and textiles. The Northern Wei belatedly began to mint its own version of the wuzhu coin in 495, but circulation of this coin was confined to the environs of the capital at Luoyang. In outlying areas 154

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

local trade was conducted using old Han coins or sundry coins issued by the Southern Dynasties. Nonetheless, the Tuoba nobility cultivated mutually rewarding relationships with private merchants. Like other Central Asian groups, the Tuoba relied heavily on trade to obtain the goods they desired. Although originally nomads themselves, the Tuoba could not breed sufficient numbers of horses in the conquered territories of north China and continuously imported horses from the steppe grasslands. Prior to the adoption of the equal-field system (juntian 均田) of state-managed land tenure in 484, the Northern Wei did not provide salaries or stipends to government officials, and consequently relied mostly on merchants – notably Sogdians – to act as its fiscal agents. In 487 the Northern Wei court abolished its state-run workshops for silks, jades, and gold and silver wares, and suspended sumptuary regulations on private manufacture of such goods. These examples show that private commerce survived even in north China through the Period of Disunion. The Yangzi river basin also developed economically in the post-Han era. Interregional commerce between the Yangzi delta and the Central Plain and between the upper and lower reaches of the Yangzi river flourished during the Three Kingdoms period despite the hostility and incessant warfare among the rival states. The influx of immigrants from the north after the relocation of the Jin court to Nanjing gave powerful impetus to economic development in the Yangzi delta, heretofore largely a sparsely populated frontier region. Under the rule of the Southern Dynasties (317–589), a series of relatively shortlived regimes, resettled immigrants reclaimed wilderness lands in the Yangzi delta for rice cultivation. Although the plains to the east of Lake Tai mostly remained swampland, rice farming expanded rapidly in areas north and west of the lake, especially around the capital at Jiankang, and in the Shaoxing plain on the southern shore of Hangzhou Bay. Many aristocratic families laid claim to vast estates, notably in Shanyin county (modern Shaoxing) of Guiji commandery. Bronze casting foundries for weapons and mirrors, ceramics workshops, and the manufacture of hemp fabrics all had become regional specialties of Guiji already in the Eastern Han period, and subsequently expanded to all of eastern Zhejiang. Consequently, Shanyin became the largest commercial and industrial city outside Jiankang itself. According to a contemporary report, there were 280,000 households in Jiankang – undoubtedly an exaggeration, but indicative of the capital’s economic prominence. Unlike the fortified cities of the north, at Jiankang only the imperial palaces were enclosed by walls, while the mansions of royal kinsmen and aristocratic families, Buddhist temples, marketplaces, artisan 155

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

workshops, and the humble dwellings of the working population spread across the surrounding countryside.27 Despite the persistence of sumptuary laws that discriminated against merchants, many imperial princes and aristocratic families engaged in commerce and moneylending through intermediaries recruited from among their clients and retainers. The tense military conflicts among rival regimes in north and south throughout the Period of Disunion sometimes interrupted the movements of people and goods. The Northern Dynasties banned commercial traffic across the Huai river until the middle of the Northern Qi (550–577 C E). Most of the time cross-border exchanges were conducted through merchants accompanying diplomatic exchanges, trade at government-supervised border markets (known as hushi, or “mutual trade”), and smuggling. The Northern and the Southern Dynasties sent delegations to each other at least 224 times in 421–588 C E (except during the period of rupture in diplomatic relations in 494–538 C E). These missions often engaged in official exchange of goods, while the emissaries themselves conducted private trade in rare goods. This official trade mostly involved horses and furs from the north in exchange for citrus fruits and handicraft products (such as mother-of-pearl and candlewax) from the south. Private transactions that were permitted as a by-product of diplomatic missions involved goods such as books and gems from the south for dried grapes and ginseng from the north. Private trade typically featured luxury and prestige goods that were highly prized among the upper classes.

Private Merchants and the Credit Economy in the Sui and Tang Periods Older prejudices against merchants were still influential. Professional merchants were prohibited from serving as government officials by the Sui in 596, and likewise excluded from military service by the Tang in 624. High government officials were banned from entering Chang’an’s main permanent markets, i.e., the western and eastern markets. Yet the status of merchants rose appreciably over the course of the Tang dynasty, as did their political presence. Following the reunification of north and south by Yang Jian (Emperor Wen, r. 581–604) under the rule of his Sui dynasty in 589, long-distance trade recovered. The “Treatise on Geography” of the Sui History singles out, in addition to Chang’an, Luoyang, and Ye (all of which had been capitals for the Northern Dynasties), a dozen other cities as major commercial metropolises: 27

Liu, Liuchao de chengshi, pp. 135–66.

156

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

namely (using their modern names) Taiyuan, Chengdu, Nanjing, Zhenjiang, Wuxi, Suzhou, Shaoxing, Hangzhou, Jinhua, Nanchang, Guangzhou, and Hanoi.28 Except for the northern capitals and Taiyuan, all of these cities were located in the southern part of China – a striking affirmation of how much economic and commercial development had already shifted to the southern part of China at this time. A proverb in the Tang period even said, “(When it comes to economic products,) Yang province (i.e., the Yangzi delta) ranks first, and Yi province (Sichuan) second.”29 Under the second Sui ruler (Emperor Yang, r. 604–618), the state constructed multiple canals – collectively known as the Grand Canal – to link together the Yellow, Huai, and Yangzi rivers (Map 4.2). Although notorious for the immense burden of labor

Zhuozhou

Taiyuan

Hejian

Yongji Canal (608 CE) Linji

Yellow River

Weizhou

Heyin Tongguan Chang’an

Luoyang

Guandong Canal (584 CE)

Bianzhou

Songzhou Suzhou

Shanyang Canal Tongji Canal Shanyang (605 CE) (605 CE) Sizhou Huai River

Xuyi

Yangzhou

Danyang

Jingkou

Suzhou Yangzi River

Jiangnan Canal (610 CE) Hangzhou

Map 4.2 The Grand Canal in the Sui dynasty 28

SuiS, 29, pp. 817, 829–30; 30, pp. 843, 860; 31, pp. 887, 897.

29

157

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

ZZTJ, 259, p. 8430.

yō hei kakinuma

conscription imposed on the populace, the Grand Canal enabled the government to transport grain and other southern goods to provision the capital and the armies stationed in the capital region, and also dramatically lowered the costs for merchants shipping commercial goods from the south to the north. The growing ease of transport between south and north expanded the markets for southern goods and fostered new habits of consumption. The most notable addition to Chinese agriculture and social life during the Tang period was the cultivation and consumption of tea. Originating from the Southeast Asian highlands, tea cultivation spread to Sichuan and the Yangzi river valley in the Tang era, and quickly became the vogue in upper-class circles. The Japanese Buddhist pilgrim Ennin repeatedly mentioned in his diary that he was served tea at official meetings and in private houses during his sojourn in China in 838–847. The Classic of Tea, composed c. 780 by the scholar-official Lu Yu, became widely celebrated as a handbook of connoisseurship that identified the best varieties and described the methods of tea preparation. Tea plants flourished on the well-drained slopes in the rugged hills of southern China, and thus tea cultivation both complemented rice farming in the valley bottoms and spurred settlement in the upland areas of the Yangzi river basin. By the end of the Tang period tea had developed from a leisurely pastime to a major article of mass consumption, and teas from south China were not only sold to consumers in the north but also exported to the nomadic peoples of Central Asia. The northern capital cities of Chang’an and Luoyang were above all consumer cities. Tang Chang’an numbered approximately a million inhabitants, including not only officials and soldiers but also large numbers of clerks, laborers, and professionals employed by the state, as well as craftsmen, merchants, and religious clergy. This urban population depended on imports of staple goods such as grain, cloth, salt, and tea, mainly from distant regions. The Tang state collected huge quantities of grain and textiles as tax revenues in various parts of the empire and transferred them to Chang’an. Production and sale of salt and tea were entirely in private hands in the Sui and early Tang periods, and salt and tea merchants became renowned for their great wealth. As Xinru Liu shows in her chapter in this volume, foreign commerce prospered in the Sui–Tang era, and many foreign merchants – notably the Sogdians who had come to dominate the Silk Road trade routes – settled in Chang’an and other northern cities. Chinese governments since the Northern Wei period had conferred the title of sabao 薩寶 (s’rtp’w in the Sogdian language, originally meaning a caravan leader) on the chiefs of the Sogdian 158

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

communities, granting them a certain degree of autonomy. In Chang’an, the homes of Sogdian merchants as well as temples dedicated to Persian religions clustered around the western market, gateway to the Silk Road. Sogdian merchants occupied a prominent place in the cosmopolitan cultural world of Tang China, and the popularity of Persian fashions, music, dance, and sports, such as polo, at the Tang court can be attributed to their influence.30 The Tang Empire also maintained commercial relationships with the states of the Korean peninsula, particularly its ally Silla, based on the tributary system of diplomacy. Following Silla’s defeat – with Tang military assistance – of Paekche in 660 and Koguryŏ in 668, the Silla rulers brought the Korean peninsula under unified rule. The Tang–Silla alliance ruptured after the Silla king resisted Tang efforts to reduce him to a puppet ruler, but the relationship between the two courts was gradually repaired. By the end of the eighth century Silla merchants were frequenting the coasts of China. When complaints arose in the 820s that Chinese often seized and enslaved Koreans, a Korean warlord, Jang Bogo, assembled his own naval force to defend the Silla coast. Jang also commanded a merchant fleet that soon achieved dominance over the Yellow Sea trade between Tang, Silla, and Japan (the Japanese pilgrim Ennin journeyed to China on one of Jang’s ships), and became a potent rival to the Silla king as well. The Yamato court in Japan, erstwhile allies of the Korean state of Paekche, dispatched numerous embassies to cultivate cordial relations with the Tang court. The Japanese envoys returned with rare and precious items (which became known in Japanese as “Tang goods” (karamono 唐物)), especially books and Buddhist religious paraphernalia. The Catalogue of Chinese Books Currently Available in Japan, compiled in the ninth century, records that at that time the court possessed 1,578 volumes, totaling 17,804 scrolls, of Chinese works, most of which had been brought back by Japanese envoys.31 After Jang Bogo was assassinated in 846 C E by an emissary sent by the Silla king, Tang seafarers began to travel directly to Japan, stimulating the development of private commercial exchange between the continent and the archipelago. Numerous Southeast Asian polities, notably the Srivijaya princes in southern Sumatra, sent emissaries to the Tang court, although the Tang only sent one embassy of its own (to Srivijaya in 683) in return. As the numbers of 30

31

Arakawa Masaharu 荒川正晴, Yu¯rajia no ko¯tsu¯, ko¯eki to To¯ teikoku ユーラシアの交 通・交易と唐帝国 (Nagoya, Nagoya daigaku shuppankai, 2010), pp. 336–84. Fujiwara Sukeyo 藤原佐世, Nihonkoku genzaisho mokuroku 日本國見在書目録, in Hanawa Hokinoichi 塙保己一 (ed.), Zoku gunsho ruiju¯ 續群書類從 (Tokyo, Kokusho kanko¯ kai, 1975–2000), 884, pp. 31–50.

159

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

foreign traders arriving at Chinese ports rose, the Tang government established a formal arrangement for conducting private maritime trade under the supervision of a maritime trade superintendency (shibosi 市舶司) and began to collect customs revenues from overseas merchants. Following the establishment of the Abbasid caliphate centered at Baghdad, many Muslin merchants – mostly Arabs and Persians – crossed Asian seas to the coasts of China. According to an Arab historian, thousands of Muslim merchants were massacred by the forces of the Huang Chao insurgency when they seized Guangzhou, Tang China’s major port, in 879. In short, the Tang period witnessed substantial growth in commerce, notably in long-distance trade of goods such as grain, salt, and tea shipped from southern producers to northern consumers and international trade both across the overland Silk Road and increasingly via maritime routes as well. However, the An Lushan Rebellion devastated the Central Plain heartland, and disrupted long-distance trade across north China. Rising military expenditures forced the state to impose new taxes on trade and commerce, notably through the salt monopoly, which by the 770s C E had become one of the government’s primary sources of revenue. As mentioned above, the capital cities of north China were predominantly consumer cities, whereas the production of rice, textiles, salt, and tea was mainly concentrated in the south. After the An Lushan Rebellion, the Hebei and Shandong regions of the Central Plain – the empire’s most densely populated regions, inhabited by approximately 2 million households, or a quarter of the total Tang population – fell under the control of independent military governors who refused to submit tax payments to the capital. Therefore, the central government’s revenues were largely drawn from the provinces in the Huai and Yangzi river basins. To cope with the soaring costs of military expenditure following the outbreak of the An Lushan Rebellion, the central government appointed Diwu Qi as the plenipotentiary tax commissioner for the Huai–Yangzi region in 756 C E. Diwu imposed a variety of emergency levies, the most important of which was the reinstitution of a salt monopoly. In addition to borrowing from merchants, the central government created a special budget and allocated funds to officials to invest in commercial ventures, which were expected to earn income for the government. After Liu Yan took over as salt commissioner in 763 C E he transformed this office into the state’s most powerful fiscal agency, with control over coinage, transport of tax revenues, and commercial taxes as well as the lucrative salt industry. In 780 the chancellor, Yang Yan, succeeded in wresting control of state finances from Liu Yan and enacted the twice-a-year tax system in an effort to restore the 160

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

priority of direct taxation in place of consumption taxes levied on salt and other goods. Although the twice-a-year tax proved a fiscal success, Yang Yan failed in his effort to repeal the salt monopoly. Moreover, the autonomy of provincial military governors in regions such as Shandong and Hebei deprived the central government of a substantial portion of its salt and silk revenues, which instead helped the governors to maintain their independence from Chang’an. The intractable shortages of bronze coin that encumbered both private trade and state finances in the late Tang period also stimulated innovations in money, credit, and finance. Bronze coins minted in Chang’an continuously flowed out to the south. To prevent this drain of coins out of the capital region, the government repeatedly issued prohibitions against carrying coins across provincial and county borders, to little avail. Moreover, low-value bronze coins were cumbersome for merchants to transport over long distances. When private merchants came to Chang’an and received payments in coin for the goods they sold, they could deposit the coin with government agencies or private financiers in the capital in return for a promissory bill that could be redeemed in cash at branch offices of the agency or firm in various provincial cities. Both merchants and government officials utilized these bills, which were known by a variety of names, such as “flying cash” (feiqian 飛錢), “convenient cash” (bianqian 便錢), and “convenient exchange bills” (bianhuan 便換), to avoid the inconvenience and insecurity of transporting cash over long distances. Merchants in the provinces could also deposit cash with local government offices and redeem the bills at the capital. These bills could be transferred to third parties, and gradually came to serve as negotiable instruments of exchange. The promissory bills proved immensely popular with merchants, but in the eyes of government officials they exacerbated the shortage of circulating coin by encouraging both public and private agents to hold large reserves of coin to redeem them. In 812, the government restricted the right to issue promissory bills to the three principal fiscal agencies of the central government, the Board of Finance (hubu 戸部), the Salt Commissioner (yantieshi 鹽鐵使), and the Department of Public Revenues (duzhisi 度支司). These promissory bills were the forerunners of the jiaozi 交子 bills of exchange that appeared in Sichuan in the late tenth century and were transformed into state-issued paper money in the early eleventh century (see Kuroda’s chapter in this volume). The Tang period also witnessed the emergence of safedeposit firms (guifang 櫃房) that provided financing for commercial transactions and loans to the government to meet emergency expenses. 161

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

Conclusion This chapter has traced the development of money, markets, and merchants from the Warring States to the Tang period. During the Warring States era there was a dramatic flowering of the market economy, as the great variety of bronze currencies and their rapid diffusion throughout the Chinese ecumene attests. Already in that period long-distance trade in commodities across different economic zones reveals a complex regional specialization of production. Bronze currencies from the Central Plain states, which appear to have been minted by private merchants and craftspeople in multiple cities, circulated well beyond state boundaries, and gold also was used as an international currency. Despite the diversity of bronze currencies, they were consistently issued in a limited number of forms – knife- and spade-shaped currencies, round coins, and imitation cowries – which suggests that all of these currencies were minted according to government standards. After the Qin unification, its banliang coin became the standard measure of value in both private commerce and state payments. The banliang coins were inscribed with their nominal weight (“half-ounce”), but in reality the weight of banliang coins varied, and gradually decreased over time. The name “banliang” ceased to signify the actual or legal weight of the coin, but instead functioned as a mark of legal tender. In the second century B C E, the Han state abolished the banliang coin, and Emperor Wu replaced the heterogeneous coins in circulation with his new wuzhu coin, which quickly became established as the monetary standard. Throughout the Han period, except for the Wang Mang interregnum, wuzhu coins prevailed in market exchange and tax payments, even though their actual weight varied and at times people discriminated between heavy and light coins. Nevertheless, people generally tolerated variations in the metallic value of wuzhu coins because they functioned as one of the primary means of tax payments. In the Western Han period, textiles and gold also circulated as concurrent currencies without a fixed exchange rate. Output of coin by state mints declined markedly in the Eastern Han period, and coinage often was suspended altogether amid the political turmoil and rapid changes in dynastic fortunes of the Period of Disunion. Nonetheless, the use of coin for market transactions persisted, particularly in the south, and private coinage proliferated to satisfy the market demand. The quality of coins deteriorated throughout the Period of Disunion, and even governments issued cheap coins using lead or iron instead of copper.

162

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants

Only with the founding of the Tang dynasty was a new monetary standard established with the issue of its Kaiyuan tongbao coin beginning in 621. People bought and sold commodities mainly in markets, not only the permanent marketplaces in towns and cities, but also in rural periodic markets and fairs. The permanent markets originally were located inside walled cities in the Qin and Han periods. The permanent markets were also walled off from the rest of the city and operated under surveillance by governmental officials. Shops in a given trade were organized into separate “alleys” to facilitate government surveillance and consumer convenience. The government gradually expanded the range of market taxes after the middle of the Western Han period, which had the consequence of discouraging large-scale commerce and concentrations of commercial capital. At the same time, local markets in rural areas proliferated, and manorial estates remained deeply engaged in the market economy. The monetary economy continued to develop in the Eastern Han and after, but the specter of copper shortages arose. By the end of the Eastern Han, the main form of tax payment had changed from coins to textiles, and the economic and social usages of textiles expanded, although coins still remained the means of everyday transactions. Beginning in the Period of Disunion, the practice of denominating coins by their nominal weight was gradually displaced by auspicious legends or reference to dynastic authority, culminating in the Kaiyuan tongbao inscription on Tang coins. The convention of designating 1,000 coins a standard monetary unit (guan, or “string”), as well as the customary practice whereby a string of fewer than 100 coins (known as a “short string”) would circulate with a value of a full 100 coins, appeared by the fifth century. In this sense, these monetary reforms of the post-Han period were not only complex, but also innovative. In sum, from the Warring States to the Tang period, low-value bronze currencies functioned as the main means of market transactions. Unlike money made of precious metals such as gold and silver, bronze currencies were too cheap for issuers to expect seigniorage profits. However, from the Eastern Han period China suffered from endemic shortages of copper and coins, while war expenditures increased. Textiles often were substituted as money in large-scale transactions, in state payments, and particularly as stores of value. However, textiles did not serve as a perfect substitute for the need for durable, fractional, and small-denomination currencies, and counterfeiting of bronze coins proliferated. Governments frequently attempted to offset war expenditures by in effect debasing their currency through minting coins with face values far in excess of their 163

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

yō hei kakinuma

intrinsic metallic content, but such expediencies encountered stiff opposition in the marketplace and created greater incentives for counterfeiting, and usually were quickly withdrawn. In the absence of an adequate money supply – and to overcome the problem of transporting cumbersome lowvalue coin over long distances – both the Tang state and private merchants began to issue paper bills in exchange for deposits of coin. Such bills gradually circulated as negotiable instruments of exchange, which dispensed with the need for hard currency. Under the subsequent Song dynasty, iron coins and various forms of paper currency were adopted as regional currencies in addition to bronze coin, which helped to alleviate shortages of copper and coin. The late Tang period witnessed the collapse of state-regulated urban markets, the spread of private exchange, and the rising status of the merchant class. Yet the state also intervened more directly in commercial exchange, and the border between private merchants and government officials became ambiguous. The salt monopoly and commercial taxes became the major sources of central government revenue, and the state also took over the issue of “flying cash” and other forms of promissory notes for long-distance transfer of funds, barring private issue of such notes. The autonomous provincial military governors in Hebei and Shandong also relied heavily on salt taxes in addition to revenues collected in silk, which was widely used as a substitute for all-too-scarce bronze coin. The rapid growth of the monetary economy, along with expanding state involvement in market exchange, would continue after the fall of the Tang dynasty in 907. Both trends accelerated after the restoration of unified rule under the Song Empire in the late tenth century.

Further Reading Arakawa Masaharu 荒川正晴, Yu¯rajia no ko¯tsu¯, ko¯eki to To¯ Teikoku ユーラシアの交通・ 交易と唐帝国 (Nagoya, Nagoya daigaku shuppankai, 2010). Emura Haruki 江村治樹, Shunju sengoku jidai seido¯ kahei no seisei to tenkai 春秋戦国時代 青銅貨幣の生成と展開 (Tokyo, Kyu¯ko shoin, 2011). Hino Kaisaburo¯ 日野開三郎, Hino Kaisaburo¯ To¯yo¯ shigaku ronshu¯ 日野開三郎東洋史学 論集, vol. 5, To¯-Godai no kahei to kinyo¯ 唐・五代の貨幣と金融 (Fukuoka, San’ichi shobo¯, 1982) Kakinuma Yo¯hei 柿沼陽平, Chu¯goku kodai kahei keizai no jizoku to tenkan 中国古代貨幣 経済の持続と転換 (Tokyo, Kyu¯ko shoin, 2018). Kakinuma Yo¯hei 柿沼陽平, Chu¯goku kodai kahei keizaishi kenkyu¯ 中国古代貨幣経済史 研究 (Tokyo, Kyu¯ko shoin, 2011).

164

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

Money, Markets, and Merchants Kakinuma, Yo¯hei 柿沼 陽平, “The Emergence and Spread of Coins in China from the Spring and Autumn Period to the Warring States Period,” in Peter Bernholz and Roland Vaubel (eds.), Explaining Monetary and Financial Innovation (Cham, Springer, 2014), pp. 79–126. Kamiya Masakazu 紙屋正和, “Zenkan ko¯hanki iko¯ no kahei keizai ni tsuite” 前漢後半期 以降の貨幣經濟について, in Kawakatsu Mamoru 川勝守 (ed.), Higashi Ajia ni okeru seisan to ryu¯tsu¯ no rekishi shakaigakuteki kenkyu¯ 東アジアにおける生産と流 通の歴史社會學的研究 (Fukuoka, Chu¯goku shoten, 1993), pp. 72–96. Kato¯ Shigeshi 加藤繁, Shina keizaishi ko¯sho¯ 支那經濟史考證, 2 vols. (Tokyo, To¯yo¯ bunko, 1952–1953). Kato¯ Shigeshi 加藤繁, To¯–So¯ jidai ni okeru kin gin no kenkyu¯ 唐宋時代に於ける金銀の研 究, 2 vols. (Tokyo, To¯yo¯ bunko, 1925–1926). Liu Shufen 劉淑芬, Liuchao de chengshi yu shehui 六朝的城市與社會 (Taipei, Xuesheng shuju, 1992). Miyazawa Tomoyuki 宮澤知之, “Gishin Nanbokucho jidai no kahei keizai” 魏晉南北朝 時代の貨幣經濟, O¯ryo¯ shigaku 鷹陵史学 26 (2000), 41–82. Peng Xinwei 彭信威, Zhongguo huobishi 中国貨幣史, 4th ed. (Shanghai, Shanghai renmin chubanshe, 2007). Tada Kensuke 多田狷介, Kan Gi Shin shi no kenkyu¯ 漢魏晉史の研究 (Tokyo: Kyu¯ko shoin, 1999). Tan, Mei Ah, “Monetary Policy as Key to State Authority and Policy in Tang China,” Journal of Chinese Studies 64 (2017), 35–110. Thierry, François, Les monnaies de la Chine ancienne: Des origines à la fin de l’empire (Paris, Belles-Lettres, 2017). Twitchett, Denis, Financial Administration under the T’ang Dynasty, 2nd ed. (Cambridge, Cambridge University Press, 1970). Twitchett, Denis, “Merchant, Trade and Government in Late T’ang,” Asia Major, new series 14.1 (1968), 63–95. Twitchett, Denis, “Provincial Autonomy and Central Finance in Late T’ang,” Asia Major, new series 11.2 (1965), 211–32. Twitchett, Denis, “The T’ang Market System,” Asia Major, new series 12.2 (1966), 202–48. Wang, Helen, Money on the Silk Road: The Evidence from Eastern Central Asia to A D 800 (London, British Museum Press, 2004). Watanabe Shin’ichiro¯ 渡邊信一郎, Chu¯goku kodai no zaisei to kokka 中國古代の財政と 國家 (Tokyo, Kyu¯ko shoin, 2010). Xiong, Victor Cunrui, Sui–Tang Chang’an: A Study in the Urban History of Medieval China (Ann Arbor, University of Michigan Center for Chinese Studies, 2000).

165

https://doi.org/10.1017/9781108587334.005 Published online by Cambridge University Press

5

Economic Philosophy and Political Economy richard von glahn

The genesis of Chinese political economy can be traced to the Warring States era (453–221 B C E), which was marked on one hand by rapid economic progress (the spread of iron metallurgy, advances in agricultural productivity, the invention of coinage, and the emergence of a private merchant class) and on the other hand by the rise of autocratic states (accompanied by the centralization of political power and mass mobilization for war). The economic principles and policies that later shaped the formation of the first unified empires – what I will designate the militarist–physiocratic state – were enunciated by leading ministers of the most successful autocratic states, such as Li Kui in Wei and Shang Yang in Qin, and set down in works such as The Book of Lord Shang and Han Fei Zi. This period also witnessed the emergence of radically different conceptions of state and market as expressed in Mencius’ moral philosophy of the “people’s livelihood,” the economic liberalism espoused by statesmen such as Bai Gui and Fan Li (and later championed by the Han historian Sima Qian), and the diverse – and sometimes contradictory – economic theories gathered together in works such as the Spring and Autumn Annals of Master Lü and the early chapters of the Guan Zi. Under the Qin and early Han empires, the principles of the militarist– physiocratic state and its tight regulation of the economy prevailed and were codified in imperial law. An even more aggressively statist theory of political economy, what I will designate the mercantilist state (exemplified by the later “Light and Heavy” chapters of the Guan Zi), came to the fore during the long reign of the powerful Han emperor Wu (r. 141–87 B C E). The legacy of Emperor Wu’s mercantilist policies and institutional initiatives was the subject of the seminal “Salt and Iron Debates” held at the Han court in 81

166

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy B C E. Ultimately, the Confucian consensus on political economy derived from Mencius’ moral philosophy – what we might call the providential state – triumphed over mercantilist ideas and became enshrined in works such as the “Food and Money” chapters of Ban Gu’s History of the Former Han Dynasty. A partial return to militarist–physiocratic principles can be seen in the institutional reforms, such as the equal-field land tenure system, adopted by the Northern Wei state and its successors, including the Sui and Tang empires. But in the wake of the devastation of the An Lushan Rebellion in the mid-eighth century, a new vision of political economy emerged. The new fiscal order was heralded by Yang Yan’s institutional reforms, such as the inauguration of the twice-a-year tax. This new fiscal order – predicated on an unprecedented accommodation to the market economy – can be described as a synergistic fiscal state, and it would form the basis of fiscal practices by subsequent Chinese states down to the Mongol conquests in the thirteenth century.

The Emergence of the Fiscal State in Warring States China Economic philosophy in China was born amid the demise of the ritually governed order of gift exchange and royal benefactions that had been the basis of the early Zhou patrimonial state and the rise of autocratic rule. This transition was fully underway by the time of Confucius (551–479 B C E), a staunch defender of the old Zhou regime. In the three centuries after Confucius, philosophers and statesmen grappled with the inexorable ascendancy of autocratic states. The political economy of the Warring States era reflected a world in which codified law supplanted ritually governed norms and the patrimonial rights of noble families had been transformed into the public powers of the state. A new political formation – the fiscal state – eclipsed the domainal regimes of the Spring and Autumn city-states.1 It was premised on the subordination of society – constituted by individual conjugal 1

On the use of the concept of the “fiscal state” in scholarship on European and Chinese history, see Richard von Glahn, “Modalities of the Fiscal State in Imperial China,” Journal of Chinese History 4.1 (2020), 1–29. I have adapted key terms of the European lexicon of political economy, such as “physiocracy” and “mercantilism,” to suggest certain parallels between Chinese and Western concepts, but of course these ideas derived from different historical contexts and also diverged in important ways, as I endeavor to show. For example, what I define as the “militarist–physiocratic” tradition in China shared the eighteenth-century French physiocrats’ fundamental conviction that agriculture is the source of all wealth, but in contrast to the laissez-faire principles of the latter, the Chinese militarist–physiocratic tradition was strongly interventionist.

167

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

households – to the autocratic ruler through taxation, labor service, and military conscription. The salient feature of the fiscal state was the unprecedented concentration of wealth and economic resources in the hands of the ruler. In the eyes of its proponents, by centralizing control over economic resources and enhancing his military power the monarch safeguarded his realm against foreign rivals, promoted the economic welfare of his subjects, and ensured that the public good would prevail over private interests. Before the sixth century B C E there is scant evidence of market transactions. Hierarchical relations between rulers and nobles were mediated by gifts of prestige goods and ritual paraphernalia. Private exchange existed, but it too involved personal transactions governed by ritual norms and was subject to the sanction of sovereign authority. Government service was unsalaried; instead, rulers rewarded meritorious conduct by their subordinates with benefactions of territories, towns, lands, and serfs. The warrior ethos of the ruling class defined nobility in terms of military prowess and a code of chivalric conduct. Within the city-states of the Spring and Autumn era (eighth to fifth centuries B C E), the obligation of military service likewise differentiated the privileged citizenry (guoren 國人) from humble commoners (shuren 庶人). From this notion of military duty (fu 賦) would spring the concept of general taxation. By the turn of the sixth century B C E, the waning authority of the Zhou kings and intensified warfare among the Central States compelled rulers to augment their command of resources by imposing taxation on arable land. The earliest reference to direct taxation of land comes from Confucius’ home state of Lu, in 594 B C E. Other states soon followed suit. Rulers began to extend their ability to mobilize war matériel by gathering information regarding both arable land and nonagricultural resources. In 548 B C E, the minister of war in Chu conducted a survey of the realm’s economic resources, including forests, wetlands, salt marshes, and arable fields, as well as calculating revenues and creating tax rolls for assessing levies of carts, horses, soldiers, and weapons. Compilation of written registers enumerating such resources became a crucial tool of government, but at the same time abetted confiscatory exactions. The fourth-century B C E Chronicles of Zuo portrayed Yan Ying (c. 578–500 B C E), a senior minister in Qi much admired by his younger contemporary, Confucius, repeatedly chastising the Qi rulers for imposing heavy taxes on their subjects. In a speech putatively delivered in 522 B C E , Yan condemned Lord Jing of Qi for sending his guards and inspectors to assert royal monopolies over the forests, marshes, and seashores, collect excises on private domestic trade, and draft subjects from 168

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

outlying areas to perform onerous labor services. In Yan’s telling, many had become ensnared in tax debts, and Lord Jing’s favorites seized goods in the marketplaces with impunity. In another speech Yan Ying appears to foretell the future usurpation of the rulership of Qi by its vassals, the Tian, who, in contrast to the burdensome exactions inflicted by the lords of Qi, earned the people’s allegiance by selling salt, fish, and timber at fair prices and charging no interest on loans of grain.2 As Yan Ying’s orations suggest, private commercial exchange at urban marketplaces was commonplace by the late sixth century B C E. Yan does not make specific reference to the use of money, but the invention of bronze currencies – although not yet dated with certainty – had probably occurred by this time (see Kakinuma’s chapter in this volume). The introduction of metallic money and the expansion of both local and long-distance trade fostered the emergence of a class of professional merchants. As in the world of the Greek city-states, scores of authorities – states, cities, and perhaps even merchants themselves – engaged in the manufacture and issue of coined money. The heterogeneous character of these currencies resulted in distinct spheres of circulation. But the currencies issued by the states and cities of the so-called “Three Jin” territories (the states of Wei, Zhao, and Hann) were widely accepted throughout the Central Plain, the most commercially advanced region of China. Much of the controversy over the expanding fiscal reach of the state centered on the assertion of the ruler’s monopoly over nonagricultural resources. As a Jin statesmen had proclaimed already in 585 B C E, “mountains, marshes, forests, and salt are the treasures of the state; if the ruler shares them with his subjects, his house will become impoverished.”3 Initially, the ruler’s arrogation of these resources provoked censure because it restricted rights of the citizenry to engage in ritually significant activities such as hunting. But as the needs of rulers to acquire matériel to wage war grew, “mountains and marshes” became a metaphor for strategic commodities such as minerals, timber, bamboo, hides, and furs. The onset of the Iron Age – which transformed the character of warfare through the mass production of weapons – and the new importance of bronze as a raw material for coinage further enhanced the value of these resources. In the words of the philosopher Mozi (fifth century B C E), the rulers and ministers of the day “desired their kingdoms and families to be rich, their 2 3

Zhaogong 20.6, in ZZ, 2, p. 2093a; Zhaogong 3.3, in ZZ, 2, p. 2031a. Chenggong 6.5, in ZZ, 2, pp. 1902c–3a.

169

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

subjects numerous, and the execution of their laws to be orderly.”4 The autocratic monarchies that dominated the political world of the Warring States era sought above all to establish a firm agricultural base as the foundation of national power. Li Kui, appointed chancellor of Wei in 422 B C E, instituted policies focused on agricultural improvement through more intensive cultivation of staple food and fiber crops. Li is also credited with initiating a form of indirect price control whereby the state purchased grain after abundant harvests when prices were low in order to boost farmers’ income, while protecting urban consumers by selling its stocks in years of dearth. This form of state intervention to smooth out oscillations in food prices became a cardinal feature of the political economy of the Chinese imperial state. Li’s strategic imperatives of “maximizing the potential of the land” (jin dili 盡地利) and protecting farmers from the scourges of drought, flood, and the seasonal vicissitudes of the market became the foundation of the militarist–physiocratic orientation in Chinese political economy. The full flowering of the militarist–physiocratic state occurred in Qin in the fourth century B C E, during the chancellorship of Shang Yang (390–338 B C E). Lord Xian (r. 385–362 B C E), who spent thirty years in exile in Wei before reclaiming rulership of Qin, introduced changes modeled on Wei institutions: assigning market inspectors to collect taxes on trade, establishing household registers for military service, and appointing magistrates in place of nobles as local governors. In 356 B C E Xian’s son, Lord Xiao (r. 361–338 B C E), appointed Shang Yang – formerly an official in Wei – as senior minister with a mandate to undertake sweeping administrative and legal reforms. One of Shang’s first measures was the adoption of a new system of military rank – seventeen grades in all – that became the basis for the reallocation of land. According to the Book of Lord Shang, a digest of Shang Yang’s policies and political theories compiled by later disciples, the rewards for killing one enemy soldier in battle included promotion of one grade in rank and grants of 100 mu of arable land, nine mu of land for a dwelling, and one servant farmer.5 One key goal of Shang Yang’s policies was to establish the conjugal household as the basic unit of social and economic organization. He doubled tax assessments on households with two or more adult males, made it a crime for fathers and adult sons to live together, and abolished the practice of primogeniture. This set of decrees was intended to undermine the economic foundations of the old nobility by breaking up their landholdings and dividing 4 5

Chapter 8, “Shangxian (I )” 尚賢上, in MZ, 1, p. 66. Chapter 19, “Jingnei” 境內, in SJS, p. 119.

170

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

them among numerous sons. Recently excavated household registers and legal statutes suggest that Shang Yang’s draconian reforms were not fully implemented even under the Qin Empire. But the principle of equal inheritance subsequently became enshrined in imperial law for more than twenty centuries and was a cornerstone of China’s family-farm economy. Shang Yang’s division of arable lands into standardized parcels separated by boundary ridges (known as qianmo 阡陌) merely followed precedents set by other states. These standardized land units became the basis of a new system of land tenure and taxation in which the Qin government distributed lands to individual households in return for payments of grain, fodder, and straw, as well as labor and military service. To ensure equity in tax collections Shang Yang promulgated uniform weights and measures. Shang also organized the population into military-style units of five and ten households and established the legal principle of mutual responsibility, whereby the whole group bore equal culpability for crimes committed by any of its members. The Book of Lord Shang begins with the necessity for the ruler to adopt laws and regulations that fit changing historical circumstances. It traces the social conflict that embroiled the world of the Warring States to both egotistical human nature, with its innate propensity to pursue fame and fortune, and the economics of scarcity in a world where population growth continuously outruns the productive capacity of the land. The ruler’s primary task is to create incentives to channel the energies of the people into agriculture and war, the twin pillars of state power. Knowledge conferred power: universal registration of people, land, grain, vehicles, and draft animals enables the ruler to mobilize labor and resources to serve the needs of the state as well as to conduct surveillance and forestall activities inimical to the ruler’s welfare. Men should be richly rewarded for their achievement in warfare, but soldiers were but one of the “three armies”; adult women must be conscripted to support the military by building ramparts, digging traps, constructing bridges, and tearing down houses, and even the elderly and infirm could contribute by tending animal flocks.6 Hostility to commerce pervades the Book of Lord Shang, which urges the ruler to increase customs duties and market levies, impose onerous corvée exactions on merchants and artisans, boost the prices of agricultural goods, and prohibit the export of grain. Not only did the wealth generated by commerce threaten to distract the masses from the toil of farming, but merchants’ mobility allowed them to elude surveillance and control. The 6

Chapter 12, “Bingshou” 兵守, in SJS, pp. 73–5.

171

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

Legalist philosopher Hanfeizi (c. 280–233 B C E) concurred that the concentration of wealth in the hands of merchants undermined social order, above all because others would abandon the arduous work of husbandry to seek the easy profits of trade: When office and rank can be bought, then none will despise merchants and artisans. When ill-gotten wealth and stocks of commodities fetch a good price in the marketplace, there will be no shortage of tradesmen. When jobbers and forestallers earn twice as much as farmers and enjoy greater esteem than soldiers or plowmen, then men of conscience and fortitude will be few and merchants and tradesmen will multiply.7

Thus Hanfeizi deemed merchants to be one of the “five kinds of vermin” that the ruler must eradicate. The outright antipathy toward commerce expressed in the Book of Lord Shang and the Han Fei Zi was virtually unique, but their depiction of agriculture as the “fundamental activity” (benshi 本事, benye 本業) echoed the common habitus of Warring States economic philosophy.8 Emphasizing the need to balance production (land) and consumption (population), the Book of Lord Shang encourages cultivation of virgin lands to expand the resource base of the state, and tax incentives to induce immigration and increase the farming population. The Guan Zi likewise urges the ruler to expand the land under cultivation, ensure that agricultural tasks are not disrupted by untimely demands for labor and military service, and accumulate sufficient stocks of grain to support the people during times of dearth. In the much-repeated (and oft-quoted) words of this text, “the five grains and rice are the Masters of Destiny for the people.”9 Farming also was associated with proper habits of mind, a sedentary lifestyle rather than freedom of movement, and a propensity for domesticity and socioeconomic fixity. It is in the Book of Lord Shang that we first encounter a description of a golden age at the dawn of civilization when “men tilled, and all were fed; women wove, and all were clothed.”10 This gendered division of household labor would endure as a cherished ideal throughout Chinese history.

7 8

9

10

Chapter 49, “Wudu” 五蠹, in HFZ, 2, pp. 1075–6. On the “farming habitus” in pre-Qin political economy, see Roel Sterckx, “Ideologies of the Peasant and the Merchant in Warring States China,” in Yuri Pines, Paul R. Goldin, and Martin Kerr (eds.), Ideology of Power and Power of Ideology in Early China (Leiden, Brill, 2015). Chapter 73, “Guoxu” 國蓄; Chapter 75, “Shanquanshu” 山權數; Chapter 78, “Kuidu” 揆度; Chapter 81, “Qingzhong (yi) 輕重(乙), in GZ, 3, pp. 1259, 1314, 1382, 1451. Chapter 18, “Huace” 畫策, in SJS, p. 107.

172

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

Confucian philosophers, too, esteemed farming and the virtues of fortitude, prudence, and thrift it instilled in the ruler’s subjects. Confucians also recognized that the ruler must first provide for the material welfare of the people before he can train them to be moral subjects. In the words of Mencius, “this is the way of the common people: those with a secure means of livelihood (hengchan 恆產) will be steadfast in their hearts.” Mencius further asserted that “benevolent government must begin with surveying and allocating lands. When boundaries are not drawn properly, neither the division of land according to the well-field system nor the levy of grain reserved for the ruler’s emolument will be equitable.”11 According to Mencius, the “well-field system” (named after the Chinese graph for “water well,” jing 井, which was seen as a visual representation of nine equal-sized plots of land) had been created by the sage-kings of yore to ensure the self-sufficiency of each household. Yet Mencius – perhaps in a direct refutation of Shang Yang – declaimed against expanding the amount of land under cultivation, which he regarded as a ploy for the ruler to amass greater wealth and pursue aggressive warfare.12 Most Warring States philosophers and statesmen recognized the essential contributions of artisans and merchants to the creation of wealth and the maintenance of the state. Even the Qin state sought not to stifle commerce but to regulate it. For example, the Qin chancellor Lü Buwei’s blueprint for imperial rule stipulated that in mid-autumn, when farmers gather their harvest and put it up for sale, customs levies should be lowered to encourage trade and the circulation of commodities.13 Lü’s advice directly contradicted the Book of Lord Shang, but it was widely shared. Guan Zhong, the Machiavellian prime minister whose wily counsel enabled Lord Huan of Qi (r. 685–643 B C E) to wield hegemonic power over the princes of his day, was depicted as a strong advocate of lowering taxes on commerce to encourage the flow of goods into the realm. It was claimed that at Guan Zhong’s insistence Lord Huan reduced customs duties in Qi to 2 percent of the value of goods.14 Mencius lauded Lord Huan for allowing grain to be sold freely across state borders, and chastised his successors for failing to follow his precedent.15 Lowering customs duties and market taxes is a repeated refrain in Mo Zi and Xun Zi as well.

11 14

15

Mencius I I IA.3. 12 Mencius I VA.14. 13 Chapter 8, “Zhongqiu ji” 中秋紀, in LSCQ, 1, p. 422. Chapter 18, “Dakuang” 大匡, in GZ, 1, p. 351. See also Chapter 5, “Chengma” 乘馬; Chapter 10, “Wufu” 五輔; Chapter 20, “Xiaokuang” 小匡, Chapter 48, “Zhiguo” 治國, in GZ, 1, pp. 90, 194–5, 439–40; 2, p. 925 respectively. Mencius V IB.7.

173

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

Contention did arise, however, over the question of state interference in production and trade, and especially over the predatory inclinations of rulers seeking to amass resources to wage war. Mencius, an arch critic of contemporary rulers’ fixation on wealth and power, claimed that in antiquity markets had existed for the simple purpose of exchanging what one has for what one needs, not as a means of accumulating wealth. Government officials supervised marketplaces to keep order, but did not interfere in the buying and selling of goods. Only when scoundrels began to hoard goods in order to manipulate prices did officials begin to tax commerce as a way of discouraging such despicable behavior.16 Yan Ying, as we have seen, harshly rebuked the Qi rulers for monopolizing mineral and forest resources and imposing excise taxes on trade. In the work ascribed to him, the Spring and Autumn Annals of Master Yan, Yan repeatedly accuses the feckless Qi ruler Lord Jing of imposing “exorbitant excise levies” (hou jielian 厚藉斂).17 The Guan Zi, too, warns rulers that “exorbitant excise levies” will arouse the enmity of the populace.18 Perhaps no figure was more critical of rulers’ propensity to exploit commerce as a means of accumulating wealth than Mozi, who railed against the ostentatious consumption of contemporary rulers and excoriated them for “exhausting the people with labor duties and burdening them with excises.”19 Applying strict utilitarian standards to government spending, he also vigorously condemned warfare as a wasteful extravagance which, like gilded palaces and sumptuous banquets, depleted the state’s treasuries and storehouses and impoverished the populace.20 The Guan Zi chapter entitled “On Extravagance in Spending” – a chapter sufficiently idiosyncratic that it must be considered a separate work – directly rebuts Mozi’s arguments in stressing the positive value of state expenditures as a stimulus to economic prosperity. This chapter begins with the premise – a staple of Legalist views – that population growth has the unfortunate consequences of raising land prices and fomenting competition among farmers, bringing ruin and unemployment to many families. But it is precisely the ostentatious lifestyles of the rich – for example, by building magnificent mausolea using the costliest materials and elaborate assemblages of grave goods – that will bring employment to the poor: “let the rich live in

16 19

20

Mencius I IB.10. 17 YZCQ, 1, pp. 163, 236. 18 Chapter 11, “Zhouhe 宙合,” in GZ, 1, p. 231. Chapter 20, “Jieyong (I )” 節用上, in MZ, 1, p. 248; see also Chapter 9, “Shangxian (I I )” 尚賢中, and Chapter 37 “Feiming (I I I )” 非命下, in MZ, 1, pp. 75, 425. Mo’s critique of lavish spending by the ruler on his personal comforts was echoed by Hanfeizi: Chapter 15, “Wangzheng” 亡徵, in HFZ, 1, p. 267.

174

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

luxury and the poor work for them,” the author concluded.21 This text extolls the profit motive underpinning commercial exchange as well as the benefits of conspicuous consumption: Itinerant merchants are no ordinary subjects. They claim no locality as their home, nor do they serve any single prince. They sell goods to earn a profit, but they do not buy simply to acquire possessions. From the mountains and forests of the state they take what they can for profit; in the marketplaces of the capital they double the return on their investment. Thus when those on high spend extravagantly while those below live in luxury, the prince and his ministers each benefit. When those on high and those below treat each other as kinsmen, the prince and his ministers do not hoard their wealth for their private enjoyment. This being so, the poor will have work to do and food to eat.22

This unvarnished embrace of unfettered commerce and consumption remained anomalous among Warring States philosophers, however. Mencius’ conviction that the pursuit of profit inevitably sowed cupidity and deceit marked an important shift in attitudes toward a merchant class that had now become a powerful force in economic affairs, and in political life as well. Mencius’ critique of the profit motive was shared most emphatically by the Confucians’ staunchest adversaries, the Legalists. Statesmen such as Li Kui and Shang Yang deemed the profiteering merchant class a threat to the commonweal embodied by the monarchal state. In their view, the sole means of securing the welfare of the people was to enrich the state, which in turn required eradicating the self-interested aggrandizement epitomized by the entrepreneurial merchant. Most representative, perhaps, of the emerging political economy of the era of the autocratic state are the numerous essays incorporated into the Guan Zi treatise. The Guan Zi purports to record a series of Socratic dialogues between Lord Huan of Qi and his minister, Guan Zhong, but the extant text dates from much later times. The older portions of the Guan Zi probably were composed at the Jixia Academy at the court of Qi during the fourth to third centuries B C E. These essays are eclectic in theme and spirit, representing a wide spectrum of philosophical principles. A more coherent set of economic doctrines – decidedly mercantilist and dirigiste in outlook – appears in the “Ratios of Exchange” chapters, which most likely date from the first century of the Han dynasty. Although surely the work of several hands, the ideas in the “Ratios of Exchange” chapters are sufficiently consistent to 21

Chapter 35, “Chimi” 侈糜, in GZ, 2, p. 652.

22

Chapter 35, “Chimi,” in GZ, 2, p. 730.

175

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

consider them the product of a single author, who for the sake of convenience I will refer to as “Pseudo-Guanzi.” Here I will confine my analysis to the earlier – and in many ways strikingly different – vision of political economy found in the older chapters, whose authors (again for heuristic purposes) I will refer to collectively as “Proto-Guanzi.”23 The cardinal principle of Proto-Guanzi is the necessity for the ruler – as “shepherd of the people,” to quote the title of the book’s first chapter – to accumulate stores of grain and goods in order to provide for his subjects in times of dearth. The ruler must not only have the prudence to lay up stores; he must also have the magnanimity to share his wealth. Proto-Guanzi recognized the tensions and conflicts engendered by the emergence of the autocratic state and the market economy, as expressed in this terse formulation: “the rural areas compete with the marketplaces for inhabitants; private families compete with public storehouses for goods; gold competes with grain for value; the countryside competes with the court for suzerainty.”24 Production and exchange are both essential to the welfare of the state and the people, but the ruler must act to maintain the delicate balance between the two: if the profits of commerce exceed the returns from agriculture, farmers will abandon their lands; if the state imposes exorbitant taxes, the people will have no reserves in times of hardship. Seasonal and annual variations in the supply of and demand for grain can cause ruinous gyrations in the price of foodstuffs, with potentially devastating consequences: cheap grain impoverishes producers, while dear grain imperils consumers. Thus – following in the footsteps of Li Kui – the wise ruler must intervene in markets to maintain price equilibrium. Proto-Guanzi also enunciates what became the canonical conception of Chinese society as comprising four functionally distinct orders: officers (shi 士), farmers, artisans, and merchants. Each order has its proper and necessary role in society, but Proto-Guanzi imagines these occupations as hereditary statuses passed from father to son, and indeed each order should dwell apart from the others. The presumption that the four orders were essentially hereditary appears already in a speech dated to 564 B C E, in which the ruler of Jin is praised because “his husbandmen (shuren) devote themselves to planting and reaping grain, while none of the merchants, artisans, and menial laborers change their occupations.”25 Interestingly, Proto-Guanzi expects 23

24

This distinction between the economic philosophy of the later “Ratios of Exchange” chapters and that of the more eclectic older chapters is based on Kanaya Osamu 金谷 治, Kanshi no kenkyu¯ 管子の研究 (Tokyo, Iwanami shoten, 1987). Chapter 3, “Quanxiu” 權修, in GZ, 1, p. 52. 25 Xianggong 9.4, in ZZ, 2, p. 1942c.

176

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

artisans to reside at government workshops, while merchants occupy the marketplaces – evidence of the prevalence of state-managed industry, in contrast to the more entrepreneurial world of commerce. The Guan Zi also delineates archetypes of the behavior and activities of the four occupational orders. Each is governed by different values and behavioral norms: the officers instruct each other in righteousness, filial duty, respect, and fraternal love; artisans teach their craft; and merchants speak of profit.26 Acting in accord with these norms, each order performs its essential role in the economy. Here is Proto-Guanzi’s depiction of the merchant order: Merchants observe outbreaks of dearth and starvation, scrutinize changes in the fortunes of states, study the patterns of the four seasons, and take notice of what goods are produced in each place. With this knowledge of prices in the marketplace, they gather up their stock of goods, load them on oxcarts and horses, and circulate throughout the four corners of the world. Having reckoned what is abundant and what is scarce and calculated what is precious and what is worthless, they exchange what they possess for what they lack, buying cheap and selling dear . . . Marvelous and fantastic things arrive in timely fashion; rare and unusual goods readily gather. Day and night thus engaged, merchants tutor their sons and brothers, speaking the language of profit, teaching them the virtue of timeliness, and training them how to recognize the value of goods.27

Although this passage is remarkably free of moral judgment, in other places Proto-Guanzi inveighs against the unbridled pursuit of private commercial profit and its damaging effects on primary producers. Ultimately, however, it is up to the ruler to ensure the welfare and prosperity of his people. “If goods are plentiful and productive tasks properly regulated, there will be little need to seek blessings from Heaven,” writes Proto-Guanzi in a chapter devoted to fiscal management. Although ProtoGuanzi repeatedly warns that extravagant and frivolous expenditures on the part of the ruler will impoverish the people and ultimately undermine the state, he advocates restraint, not austerity. The ruler must strike a balance between thrift and indulgence: 26

27

Chapter 20, “Xiaokuang” 小匡, in GZ, 1, pp. 400–2; another version of this passage is recorded in the Warring States-era anthology Discourses of the States: see Chapter 6, “Qiyu” 齊語, in GY, pp. 200–21. Sterckx (“Ideologies,” p. 239) cites a parallel passage in Chapter 30, “Junchen (I )” 君臣上, of the Guan Zi (GZ, 2, p. 550), that gives different norms: “incorrupt” 廉 for officers; “simple” 愚 for farmers; and “honest” 愿 for artisans and merchants. A parallel formation in a pair of nearly identical texts from the second century B C E denotes yet another set of norms: for officers, conduct 行; for farmers, physical effort 力 (or, alternatively, savings 藏); for artisans, skill 巧; and for merchants, calculation 數. See Chapter 11, “Qisu xun” 齊俗訓, in HNZ, 1, p. 368; Chapter 9, “Xiade” 下德, in WZ, p. 260. Chapter 20, “Xiaokuang,” in GZ, 1, p. 402.

177

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

Gold is the standard of expenditures. The prince who discerns the fundamental laws of gold will understand the dangers of parsimony and prodigality. Knowing this, he will exercise moderation in his expenditures. Now, parsimony in expenditure harms husbandry, while prodigality in expenditure harms trade. Parsimony causes the value of gold to fall, and when gold is cheap productive tasks remain uncompleted. For this reason, then, parsimony is harmful to husbandry. Prodigality in expenditure, by contrast, causes the value of gold to rise, and when gold is dear, the value of goods declines. For this reason, prodigality harms trade. Thus goods remain unsold and the peoples suffer privation: such a state of affairs results from ignorance of the standard of expenditures. Production halts and yet the surfeit of goods grows: such a state of affairs results from ignorance of the need to moderate expenditures.28

Proto-Guanzi here denies that goods or money have value in themselves. It is the ruler, through his fiscal policies, that establishes their relative values. Meager revenue extraction and spending will reduce the demand for gold, and the resulting surplus of money in the market causes prices to rise. Conversely, excessive taxation and expenditure, by increasing the demand for gold, induces deflation and discourages production. Thus the inordinately frugal ruler is just as harmful as the extravagantly profligate one. The basic tenets of political economy espoused by Proto-Guanzi were widely shared among contemporary philosophers and statesmen. The opening lines of the Guan Zi treatise – “if the granaries are full, the people will respect propriety and restraint; if food and clothing suffice, the people will understand honor and shame” – expresses the causal relationship between agriculture as the primary occupation and a social order founded on norms of morality that became the hallmark of the farming habitus.29 Proto-Guanzi also is the source for the adage that “the ruler does not amass goods in his treasuries, but rather stores them among the people.”30 Of course, Proto28 29

30

Chapter 5, “Chengma” 乘馬, in GZ, 1, pp. 88–9. Chapter 1, “Mumin” 牧民, in GZ, 1, p. 2. Many Han proponents of the farming habitus cited these lines from Guan Zi, for example Chapter 4, “Wuxu” 無蓄, in XS, p. 163; Chapter 35, “Shoushi” 授時, in YTL, 1, p. 423. The same Guan Zi passage appears prominently in the economic treatises of the major Han histories: SJ, 129, p. 3255; HS, 24A, p. 1128. But none of these later texts repeats the words immediately preceding this passage: “if the state stockpiles wealth, people from afar will flock there; if lands are opened for cultivation, the people will remain there contentedly.” This passage also was routinely excised from later digests of the Guan Zi, for example in the early Tang anthology of writings on governance, Qunshu zhiyao 群書治要. See QSZY, 32, p. 412a. Chapter 3, “Quanxiu,” in GZ, 1, p. 52. As we shall see, Pseudo-Guanzi adopted a very different tone.

178

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

Guanzi also asserts that it is by enriching the people and increasing the agricultural surplus that the ruler can raise an ever-victorious army.31 Although Confucians repudiated the unvarnished goal of enriching the state to bolster its military might, they readily agreed that the ruler bore responsibility for the smooth functioning of markets so that his people would not lack the necessities of life. The third-century B C E Confucian philosopher Xunzi extolled the crucial role of the ruler in managing resources for the benefit of all and ensuring the widest possible circulation of goods, while extracting only modest revenues: These are the laws of a true king: he establishes fair rates of taxation, regulates the people’s livelihood, and disposes of the myriad goods, thereby providing nourishment to all of his subjects. He assesses a tax of one-tenth of the produce of the land; his officials inspect goods at border crossings and marketplaces but levy no excises; he dictates the times when the mountains, forests, marshes, and fishing grounds will be closed or opened for the people’s use without imposing excises on them; and his officials inspect arable lands and grade them according to their productivity. He determines tribute payments after taking into account the distances over which they must be conveyed. He ensures the free movement of goods and grains so that their circulation is not hindered or obstructed. Goods flow to where the need for them arises, and all within the four seas are like one family.32

Like Proto-Guanzi, Xunzi recognized that merchants performed a vital role in the circulation of goods necessary to meet the consumption needs of both the people and the state. But it is up to government officials to ensure that trade is conducted on honest terms and delivers benefits to the people: “the tasks of the overseer of markets are to clear away obstructions, make the roads readily passable, guard against brigands and thieves, and certify fair prices; undertaking these activities in the proper season, he ensures that guests and travelers are secure and that goods and wealth circulate without impediment.”33 The functional specialization and mental habitus of the merchant class precluded any part in governance, however. Xunzi observed that “in circulating goods and wealth, comparing the genuine and the false, and distinguishing what is precious from what is worthless, the gentleman is not the equal of the merchant.”34 But the entrepreneurial skill displayed by 31

32 34

Chapter 48, Zhiguo” 治國, in GZ, 2, pp. 924–7. This chapter more than any other in Guan Zi echoes the perspective of the Book of Lord Shang. Chapter 9, “Wangzhi,” in XZ, p. 175. 33 Chapter 9, “Wangzhi,” in XZ, p. 184. Chapter 8, “Ruxiao” 儒效, in XZ, p. 131.

179

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

merchants in their undertakings was appallingly amoral: “serving the goal of profit, vying for goods and wealth, lacking any deference or compromise, acting with brazen daring, preying with brute greed, single-minded in their lust for gain like a swarm of locusts – such is the courage of merchants and thieves.”35 Even Proto-Guanzi warned that entrusting government to merchants, like allowing women a voice in public affairs, would have catastrophic consequences: If merchants are at court, goods and money will flow upward; if women discuss the affairs of men, then there will be no faith in rewards and punishments. If there are no distinctions between men and women, then the people will have no sense of honor and shame. If goods and money flow upward and the people have no sense of honor and shame, it will be difficult to secure the well-being of the people, and soldiers will not risk their lives for the ruler.36

Like the Book of Lord Shang and Hanfeizi, Proto-Guanzi repeatedly warned that allowing the rich to use their wealth to obtain rank and emolument would ruin “the rule of law” (fazhi 法制).37 The political economy of the Warring States era reflected a world in which codified law supplanted ritually governed norms, and the patrimonial rights of the old Zhou aristocracy had been transformed into the public powers of the state. To be sure, Confucian, Mohist, and Daoist philosophers all repudiated the militarist–physiocratic model of governance that by the end of the fourth century B C E had been adopted by all of the contenders for power. In his famous audience with the king of Wei c. 320 B C E, Mencius excoriated the myopic pursuit of wealth and power that entranced the rulers of the rising autocratic states. In response to the king’s query, “How can you profit my state?” Mencius angrily retorted, “How can you speak of profit? There is only benevolence and righteousness, and nothing else.”38 Xunzi likewise denounced the predilection for rulers and officials “to be exceedingly scrupulous in calculating, recording, taking, and giving goods and wealth, while neglecting ritual and righteousness and allowing them to wither away” as the

35

36 37

38

Chapter 4, “Rongru” 榮辱, in XZ, p. 58. Elsewhere Xun Zi writes, “pursuing wealth at the risk of one’s life, courting disaster while shirking consequences, displaying indifference to truth and falsehood or right and wrong, solely intent on besting others: this is the lowest type of courage.” See Chapter 23, “Xing’e” 性惡, in XZ, p. 557. Chapter 3, “Quanxiu,” in GZ, 1, p. 53. Chapter 13, “Baguan” 八關; Chapter 65, “Lizheng jiubai jie” 立政九敗解; Chapter 67, “Mingfa jie” 明法解, in GZ, 1, p. 266; 2, pp. 1195, 1218 respectively. Mencius 1A.1

180

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

sign of “a shameful state.”39 Rulers should devote themselves to the welfare of their subjects by providing the conditions for family farms to flourish and for the means of livelihood to circulate unhindered by grasping merchants and officials. The mission of the ruler is a providential one, and the ideals of political economy espoused by Mencius and other Confucians can be described as a providential state. This model of political economy shared the universal presumption that agriculture is the foundation of the economy. It also accepted the necessary place of commerce and merchants in the smooth functioning of the economy. But it firmly rejected the intrusion of the values of the marketplace into the conduct of government. “If wastelands lie uncultivated and the ruler does not amass goods and wealth, it is not a calamity for the state,” Mencius wrote, “but if the ruler above neglects the rites and the people below lack learning, lawless people will arise and the days of the state will be numbered.”40 Pursuit of wealth and power undermined rather than enhanced the ruler’s ability to minister to the moral and material needs of the people. Confucius renounced his disciple Ran Qiu for abetting a tyrannical chief minister’s efforts to “amass revenues” (julian 聚斂) and thus make himself “richer than the Duke of Zhou.”41 His followers abhorred “fiscalist ministers” (julian zhi chen 聚斂之臣) focused single-mindedly on maximizing revenues without regard for its effects on the people.42 Xun Zi, like Mencius, predicted that if the ruler devoted himself solely to amassing revenues (julian), his state would surely perish.43 But the Qin and Han empires fully embraced the principles of a fiscal state premised on militarist–physiocratic principles, and even many Confucian statesmen would come to accept it as a pragmatic necessity.

Han Debates over Political Economy and the Emergence of Mercantilism The establishment of a unified empire by Qin in 221 B C E marked the triumph of the military–physiocratic model of political economy and its fusion of social ranking derived from military organization with an agrarian economic base. Qin military success rested on the state’s mobilization of the entire society for warfare. Qin officials enrolled subjects into units of five and 100 39 42

43

Chapter 10, “Fuguo” 富國, in XZ, p. 218. 40 Mencius I VA.1. 41 Analects 11.17. The locus classicus for the term julian zhi chen is the Han-era Confucian treatise The Great Learning. See Chapter 42, “Daxue” 大學, in LJ, 2, p. 1675b. Chapter 9, “Wangzhi,” in XZ, p. 168.

181

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

households for purposes of taxation, military service, and public-works construction. Social hierarchy was based on military rank. Generous rewards and promotions were doled out to meritorious soldiers and farmers. Officials were held to exacting standards in the execution of their duties and faced harsh punishment for even slight shortcomings. In Xun Zi’s words, the rulers of Qin “employ their people harshly, terrorize them with authority, embitter them with hardship, coax them with rewards, and cow them with punishments.”44 The Qin rulers shared the Legalist philosophers’ physiocratic disdain for commerce as inherently sterile. Labor was considered the crucial resource, and the state mobilized slaves, convicts, and ordinary conscripts to work not only in its mines, workshops, and plantations, but also on massive construction projects. The vital importance of labor also can be seen in a system of state extraction focused on labor services equally apportioned among all adults. Mobilization of labor in turn was predicated on the capacity of the state to penetrate local society through meticulous record keeping on a scale scarcely imagined before modern times. The fundamental ethos of the autocratic state was summed up by the guiding principle of “register households in order to make the people equal” (bianhu qimin 編戶齊民). Civil registration became a crucial mechanism for eliminating the special status and privileges of the nobility and reducing the whole free population to common and equal status as subjects of the ruler. Although the Qin Empire barely lasted fifteen years, its legacy profoundly influenced its long-lived successor, Han (202 B C E – 220 C E), which adopted many features of Qin government. The Han, like the Qin, was committed to the smallholder family farm as the foundation of popular welfare. Many elements of Qin political economy – including the establishment of the household as the basic social and fiscal unit, the use of merit ranks to award landholdings, and strict regulation of commerce and industry – became cornerstones of Han imperial governance as well. Qin and Han statesmen alike feared that the concentration of wealth in the hands of powerful landowners and merchants posed the greatest threat to social order and imperial authority. Following its elimination of rival states, the Qin state took an abrupt antimercantile turn, adopting measures to rein in the market economy and curb the wealth and political power of the merchant class of the Central Plain. In 214 B C E the Qin enacted a deportation law that authorized the impressment 44

Chapter 15, “Yibing” 議兵, in XZ, p. 317.

182

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

of merchants and other transients into military service on the distant northern and southern frontiers. The Qin state compiled “marketplace registers” (shiji 市籍) that listed merchants and tradesmen separately from the general population and became the basis of these deportation orders. The marketplace registers thus reinforced the definition of merchant and artisan as hereditary statuses. All persons who traveled away from their native place were required to carry passports, and itinerant merchants had to register with local officials to obtain permission to sell their wares. Social and economic restrictions on merchants became more stringent under the Han, when marketplace registers were used to requisition goods for official use and to impose a broader range of legal discriminations (for example, merchants and artisans at times were forbidden to hold government office, or to own arable land). However, a radical shift in economic policies ensued following the ascension of Emperor Wen (r. 180–157 B C E). In contrast to the militarist–physiocratic principles that prevailed in the Qin and early Han, Wen embraced the Huang–Lao philosophy of government centered on a ruler who ensures the economic well-being of his subjects through personal frugality, minimal taxation, and refraining from disrupting their livelihood. Most notably, Wen reversed the anti-mercantile policies of his predecessors. Wen suspended the state monopoly on minting money, allowing free coinage by private individuals, and sanctioned private development of mineral resources, such as salt and iron, over which Warring States rulers had claimed sovereign proprietary rights. Imperial princes, families of government officeholders, and wealthy landowners rushed to exploit these opportunities to reap profits from industry and commerce. Many statesmen reacted with alarm to the ways in which Emperor Wen’s laissez-faire policies exacerbated economic and social inequality. Two of Wen’s senior advisers, Jia Yi and Chao Cuo, denounced the fragmentation of imperial authority that had abetted the trends toward private concentration of wealth and conspicuous consumption. Jia Yi reaffirmed the primacy of agriculture, claiming that the prosperity of the market economy inevitably impoverished the farming population. Jia reiterated the idea – enunciated by Proto-Guanzi and Xunzi – that the supreme duty of the ruler was to ensure that the people have sufficient stores of grain to endure the inevitable scourges of flood and drought. By giving free reign to the pursuit of mercantile profit, the emperor had encouraged farmers to abandon farming and thus reduced agricultural output. Jia also blamed free coinage for both the neglect of agriculture and the proliferation of debased coins that inflated grain 183

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

prices.45 Chao Cuo, too, warned that many farm families, teetering on the brink of insolvency and starvation, were compelled to sell their land allocations to redeem their debts.46 Chao persuaded the emperor to adopt strategies for increasing agricultural production such as promoting settlement and land reclamation in frontier regions, allowing purchase of merit ranks through donations of grain to the frontier armies, and lowering land taxes – which were eliminated altogether for more than a decade (167–156 B C E ) before being restored at modest rates after Wen’s death. But these actions failed to check the “aggrandizement” (jianbing 兼併) of landholdings by the wealthy and powerful. Even the historian Sima Qian – an admirer of the entrepreneurial spirit – conceded that “at this time the net of the law was slack and people became rich, using their wealth to lord it over the less fortunate, to the point where aggrandizers without title or office brutally imposed their will on the inhabitants of the countryside.”47 Disputes over monetary and fiscal policies at Emperor Wen’s court were embedded in a broader debate on the origins and nature of markets and money. An abiding tradition depicted the sages as inventors of useful technologies; for example, Shun – prominently enshrined among the ancient hero-kings – was acclaimed for teaching his people farming, pottery, and fishing.48 Yu, the mythic hero who quelled the great flood to make the earth habitable for humanity, also figured as a model of the activist ruler. The canonical Book of Documents attributed the origins of trade to Yu, who upon resettling the people after the catastrophic floods “strove to induce them to exchange what they had for what they lacked.”49 According to the “Appended Commentary” of the Book of Changes, probably composed in the third century B C E, markets and exchange were the creation of the most ancient of the sage-kings, the Divine Husbandman, who “held the market at noon and bade the people of the Subcelestial Realm to come hither, thus gathering all the products of his domain. Each exchanged what they possessed for what they desired, and then retired.”50 Advocates of an assertive state presence in economic life went farther, claiming that the sage-kings 45

46

47 48

49

“Wuxu” 無蓄, “Zhuqian” 鑄錢, in XS, 4, pp. 163, 168. Jia Yi’s essays are quoted at length by Ban Gu: HS, 24A, pp. 1127–30; HS 24B, pp. 1153–6. Chao’s memorial to Emperor Wen likewise was given a prominent place in Ban’s history of the Western Han dynasty. See HS, 24A, pp. 1130–4. SJ, 30, p. 1420. This claim, which appears to originate with the Mo Zi – see Chapters 9 and 10, “Shangxian (I I), (I I I),” in MS, 1, pp. 77, 97 – also appears in Mencius (2A.8) and is repeated in Guan Zi, Huai Nan Zi, and even Lie Zi. By contrast, Confucius (Analects 15.5) portrays Shun as “the one who achieved order through non-action.” Chapter 5, “Yiji” 益稷, in ShS, 1, p. 141a. 50 “Xici” 繫辭, I I, in ZY, 1, p. 86c.

184

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

invented money and markets not simply to facilitate the exchange of goods, but for the express purpose of rescuing the people from poverty in times of distress. The Guan Zi traces the origins of coinage to the sage-kings Yu and Tang (legendary founders of the Xia and Shang dynasties respectively): during times of flood and drought, when the starving multitude resorted to selling their children into slavery, “Tang availed himself of the metal of Mount Zhuang, and Yu took the metal of Mount Li, to cast money, which they used to redeem the children from bondage.”51 Other Warring States texts celebrated the crucial interventions of the Zhou kings to adjust the value of money to ensure the unimpeded flow of goods during times of want.52 Proponents of the farming habitus, in contrast, retained the focus on popular welfare but elided both the role of the market and state intervention. Thus Xunzi affirmed that in the face of famine Tang and Yu had ensured that the people set aside sufficient stores of grain to provide for their subsistence, without any mention of money or markets.53 In reaction to Emperor Wen’s laissez-faire policies, critics of nearly all philosophical stripes welcomed the idea that the state should intervene in the market economy to preserve popular welfare. Both Jia Yi and Chao Cuo advocated a strong, providential monarchy that would ensure the security of the state by providing for the subsistence needs of the people (and quell rebellious sentiment). Others embraced an even more expansive role for state management of the economy, not only to redress inequality and ensure the subsistence needs of the people but also to enhance the power of the state. The articulation of the basic principles of what we may call a mercantilist fiscal state is found most cogently in the “Ratios of Exchange” chapters of the Guanzi treatise. These chapters by anonymous authors – here designated collectively Pseudo-Guanzi – probably were composed during the second and first centuries B C E. Chinese mercantilism, like the militarist–physiocratic philosophy, was conceived amid the intense interstate rivalry of the Warring States period. The key difference between the two is that the former concentrated on the mobilization of economic resources rather than military manpower. Initially, the mercantilist type was premised on a world of multiple competing states, long-distance trade, and the asymmetries of comparative advantage. Pseudo-Guanzi urged the ruler to manipulate the terms of trade 51 52

53

Chapter 75, “Shanquanshu,” in GZ, 3, p. 1300. This idea appears in Guoyu and Yizhoushu; see Tamara Chin, Savage Exchange: Han Imperialism, Chinese Literary Style, and the Economic Imagination (Cambridge, MA, Harvard University Asia Center, 2014), pp. 265–7. Chapter 10, “Fuguo,” in XZ, p. 220.

185

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

and especially the money supply to induce an inflow of goods and wealth and prevent a dissipation of his own resources. Control of the money supply enabled the ruler to adjust prices, ensure a favorable balance of trade, and maintain an equilibrium between domestic producers and consumers. In many respects, the principles of political economy expressed in the “Ratios of Exchange” chapters – maximizing exploitation of agricultural, mining, and manufacturing resources; maintaining a favorable balance of trade; and amassing reserves of gold and bronze coin that enable the state to control prices and consumption – parallel the economic strategies encapsulated in the mercantilist doctrines of early modern Europe. In contrast to European mercantilism, however, Chinese mercantilism sought not to strengthen the domestic merchant class but rather to displace it.54 In contrast to the older portions of the Guanzi treatise, Pseudo-Guanzi depicted the ruler as enmeshed in constant struggle not only with rival states, but also with merchants, moneylenders, and noble seigneurs who lived off the revenues of their estates. These groups were demonized as parasites voraciously exploiting the populace for their own private gain through the instrument of the market, and thus Pseudo-Guanzi explicitly rebutted the favor shown to conspicuous consumption by the wealthy found in the idiosyncratic “On Extravagance in Spending” chapter.55 The ruler is expected to take action to protect farming families from the inequities of the marketplace, to equalize the supply of and demand for goods, and to level gross disparities in the distribution of wealth. In contrast to the older sections of the Guan Zi, which stress strategies for accumulating grain reserves and acknowledge the positive contributions of merchants to the smooth functioning of markets, the “Ratios of Exchange” chapters dwell on monetary policy as the key to managing the economy. Pseudo-Guanzi sought to raise revenues while providing tax relief to primary producers. Warning of the debilitating effects of labor and military service – the mainstays of the militarist–physiocratic model – Pseudo-Guanzi advocated extraction of revenue from manufacturing and trade rather than direct taxation. Taxes discouraged production and consumption, but used strategically they could also yield benefits, for example to suppress the usurious practices of moneylenders and grain forestallers. Pseudo-Guanzi 54

55

On the economic philosophy of the “Ratios of Exchange” chapters of the Guan Zi, see Hu Jichuang 胡寄窗, Zhongguo jingji sixiang shi 中國經濟思想史 (Shanghai, Shanghai renmin chubanshe, 1962), 1, pp. 238–377; Kanaya, Kanshi, pp. 152–75; Chin, Savage Exchange, pp. 32–68. Chapter 76, “Shanzhishu” 山至數, in GZ, 3, pp. 1322–51.

186

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

also envisioned a large role for the state in industrial production. The fivepoint outline presented in the “Stabilizing State Finances” chapter stipulates that the ruler must (1) monopolize the resources of “the mountains and marshes”, (2) manufacture consumer goods as well as weapons, (3) generate revenues from production and trade of salt and iron, (4) issue currency, and (5) set aside hills and marshes unsuitable for agriculture as pasture for horses and cattle.56 Here and elsewhere Pseudo-Guanzi asserts that salt and iron are immune from the normal laws of supply and demand: because the demand for these goods is inelastic, the ruler can extract substantial revenues from them that will lessen, if not obviate, the need for direct taxation. But salt and iron are rare exceptions to the principle that supply and demand determine the price of goods. Above all, the price of grain – which Pseudo-Guanzi repeatedly refers to as the “master of destiny” (siming 司命) of the people – is subject to the rule of the market. Grain prices oscillate violently not only because of natural disasters such as drought, flood, and pests, but also because of the actions of forestallers and regraters who accumulate stocks of grain, thus creating shortages that artificially drive up prices. Apart from legislation forbidding hoarding, the ruler must wrest control of the economy by building up his own grain reserves, which can be released into the market when consumers suffer from high prices. The enlightened monarch can manipulate markets to his own advantage, for example by raising the price of grain and other commodities to stimulate production and encourage foreign merchants to import the goods the ruler desires. In addition, the ruler wields a uniquely powerful weapon that he alone possesses: the ability to issue currency. If grain is the master of destiny, currency is the “universal medium” (tongshi 通施) through which the price of grain and all other goods can be controlled. Pseudo-Guanzi shared the common conviction that the function of money was not – as it was for the Greek philosophers – to determine a just price for the exchange of goods, but rather the means whereby the ruler can overcome the inevitable cycles of dearth and plenty to satisfy the subsistence needs of the people. PseudoGuanzi recognized the basic law that the values of money and commodities were related in inverse proportion to their quantities. But Pseudo-Guanzi stressed that the supply of money, unlike grain and other goods, is directly subject to the ruler’s control. Although money possesses no intrinsic value and cannot itself satisfy utilitarian needs for food, clothing, and shelter, 56

Chapter 79, “Guozhun” 國准, in GZ, 3, p. 1394.

187

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

through his monetary and fiscal policies – increasing or decreasing the money supply and intensifying or relaxing taxation – the ruler can manipulate the exchange value of money and by extension commodity prices. This leverage over exchange values enables the state to dictate the terms of trade, and thus manage all economic activities. To accomplish these goals, the ruler must arm himself with both knowledge and tactics. Pseudo-Guanzi especially emphasizes the necessity of compiling accurate data about population, resources, and productive capacities in order to make proper judgments about, for example, the price level or the size of the money supply.57 Through mastery of mathematical laws and their application to fiscal and monetary policies, the ruler acquires the prescience necessary to cope with ever-fluctuating economic circumstances. Amid the constant clash of economic forces, it is the ruler alone who can create the balance (heng 衡) and stability (zhun 准) – two of Pseudo-Guanzi’s favorite watchwords – necessary to enrich the people and the state. Calculation of value in the quantitative terms of the marketplace decoupled the state’s economic policies from the moral norms that governed the farming habitus, yet the power vested in the ruler by control over production, exchange, and consumption was still dedicated to the same goals of popular welfare. The economic logic of Pseudo-Guanzi’s mercantilism pervaded the radical shift in fiscal and monetary policies adopted by the fourth Han ruler, Emperor Wu (r. 141–87 B C E). Aggrieved by the growing wealth and power of the seigneurial and mercantile classes, Wu repudiated Wen’s laissez-faire posture. At the same time Wu deviated from the militarist–physiocratic model by seeking to capture the empire’s commercial and industrial wealth to pay for his costly campaigns of conquest and colonization in Central Asia, Korea, and Vietnam. Toward these ends Wu enlisted a brain trust of fiscal advisers drawn from the merchant class to devise sweeping institutional reforms that empowered the state to seize control of much of private industry and commerce. Wu’s economic philosophy was personified by Sang Hongyang, the son of a Luoyang merchant, who was recruited into the emperor’s inner sanctum of fiscal advisers because of his prodigious facility with figures and accounts and eventually rose to the office of exchequer. Guided by Sang, Wu placed the lucrative salt and iron industries under state monopoly; imposed new and onerous taxes on merchant income and 57

On the crucial importance of mathematical calculation for fiscal and monetary strategy in the Guan Zi, see Chin, Savage Exchange, pp. 34–48.

188

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

capital assets; introduced the so-called “equitable-delivery” (junshu 均輸) and “balanced-standard” (pingzhun 平準) systems whereby government agents stockpiled commodities and aggressively intervened in markets to smooth out price fluctuations; and created a new monetary system that came to rest on a new and remarkably stable bronze coin. Sang Hongyang and his coterie were merchants, not philosophers. But the mercantilist theoretical principles underlying the command economy they created were fundamentally consistent with those found in the “Ratios of Exchange” chapters of the Guan Zi. Cognizant of the value of commerce and industry as sources of wealth, Sang Hongyang sought not to suppress trade but rather to manipulate the terms of exchange and the money supply to ensure an inflow of goods and wealth into the hands of the state. The state assumed monopoly control of the most lucrative industries as well as the production of strategic goods and shifted revenue extraction toward indirect taxes paid in money rather than inkind taxation of agricultural produce and labor. Taxation of landed property remained light. Most of the central government’s income – and nearly half of total revenue – was collected in the form of coin. The salt and iron monopolies – which generated more than half of the central government’s revenue and 20 percent of total revenues – became the mainstay of state finance. Indirect levies paid in coin also provided the vast majority of the revenues of the Privy Purse. Sang Hongyang’s policies initially achieved significant success. “Within a year’s time,” wrote Sima Qian, “the [capital] granaries were full, the frontier camps had surpluses of grain and other goods, and the equitable delivery offices held 5 million bolts of silk. Taxes on the people had not been increased, yet revenues sufficed to meet the imperial government’s expenses.”58 Sang enjoyed the full confidence of the emperor, and continued to direct state fiscal policy for the remainder of Wu’s lifetime and under his successor as well. Many contemporary statesmen and scholars denounced Sang Hongyang’s fiscal policies as blatant expropriation by a tyrannical regime. Sima Qian quoted the imperial counselor Bu Shi fuming that “Sang Hongyang has government officials squatting in the market stalls trading in goods and scrambling for a profit!” Sima himself darkly likened Wu’s agenda to the Qin Empire under the First Emperor, when “the wealth of the whole world was exhausted to serve the ruler, and yet he was not satisfied.”59 In his pungent critique of Emperor Wu’s mercantilist regime, Sima articulated an 58

SJ, 30, p. 1441.

59

SJ, 30, pp. 1442–3.

189

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

alternative economic philosophy that repudiated both state intervention in the marketplace and the moralistic cant of Wu’s critics, and instead sanctioned trade and industry as honorable forms of livelihood. In his grand historical panorama, Sima Qian was preoccupied with two dramatic changes that had been gathering momentum since the Warring States period: on one hand, the immense power of the bureaucratic state; on the other, the pervasive influence of the market economy and the mercantile elite it spawned. Both of these developments upended the old Zhou social order premised on noble rank and hereditary status. Sima did not mourn the demise of the hereditary nobility, but neither was he hopeful about the nascent social order based on wealth. His pithy declaration that “the desire for wealth need not be learned; it is inherent in human nature” merely echoed widely shared views; the Confucian Xunzi, for example, identified “fondness for profit” as the core of human nature. But Sima did not share Xunzi’s conviction that the norms of moral knowledge and ritual propriety could restrain avarice and self-interest. Fear of poverty afflicted kings no less than paupers, and in Sima’s eyes Emperor Wu and his advisers, like the scheming denizens of the marketplace, were driven by naked greed. Sima Qian was duly fascinated not only by the spectacular success of the great merchants of his day, but also by the operation of the market as a spontaneous and autonomous force: Society must have farmers to provide food; woodsmen, fishers, and miners to supply the products of nature; artisans to transform them into the necessities of life; and merchants to distribute them. What need, then, is there for government decrees, requisitions and allocations, or periodic summons to labor? Let each man exercise his own abilities and apply his own effort to obtain what he desires . . . When each person pursues his own occupation and delights in his work, like water running downhill goods will flow ceaselessly night and day without being summoned; the people will produce them without being asked.60

It was in this dynamic interplay of ability, effort, and desire that Sima located the special genius of the entrepreneur. Merchants possessed specialized wisdom; they followed the canonical rules (jing 經) of their trade; their virtues of diligence, discipline, and thrift constituted the “true Way of earning a livelihood” (zhisheng zhi zhengdao 治生之正道). Sima shared the widespread belief – expressed most cogently in the contemporary work The Young Master of Accountancy – that markets operated according to observable and 60

SJ, 129, p. 3254.

190

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

measurable laws, such as laws of supply and demand and the oscillation of business cycles.61 But many merchants also employed uncanny schemes (qisheng 奇勝) and resorted to mean pursuits – as horse doctors, itinerant peddlers, knife sharpeners, dealers in animal fats and dried tripe, even gamblers and grave robbers – to make their fortunes.62 Their success derived precisely from the entrepreneurial but amoral courage derided by Xunzi as the common vice of merchants and thieves. Most radically, Sima Qian erased the sage-kings from economic life and livelihood. His assertion that markets appeared spontaneously without the intervention of a sage-ruler challenged the prevailing assumption that money and exchange were instruments of a moral transactional order. Money, too, was a spontaneous product of exchange among farmers, artisans, and merchants rather than a deliberate intervention by sage-rulers to relieve economic distress.63 Yet Sima acknowledged that control over money not only fostered the concentration of wealth in the hands of aggrandizers but also encouraged challenges to sovereign authority. Sima was unsparing in his condemnation of Emperor Wen – whom he otherwise depicted as a model ruler for his restraint in not interfering in the private economy – for yielding the right to coin money to princes and merchants, most notoriously the prince of Wu, who rebelled against Wen’s successor in 154 B C E. Sima Qian’s economic writings are suffused with a profound ambivalence about the power of money and commerce. For individuals, especially those poor souls without the inherited advantages of rank, office, or landed property, the calling of the merchant offered the opportunity to earn a fortune and “to stand alongside the lord of a city . . . and enjoy the pleasures of a king; do they not deserve to be called the ‘untitled nobility’?”64 Yet the rich inevitably exploit their power. Worse, the triumph of market values reduced all goods, even human beings, to mere commodities; the unfortunate, impoverished by debt, were reduced to slavery. Sima extolled his contemporary Bu Shi, who gave away the riches he made from herding sheep to his kinfolk, his community, and even the state; but conceded that none imitated Bu’s selfless magnanimity, which indeed “was contrary to human nature.” Thus, while condoning the private pursuit of gain, Sima no less than Mencius harshly condemned the intrusion of the values of the marketplace into the conduct of government, especially an administration 61

62

Olivia Milburn, “The Book of the Young Master of Accountancy: An Ancient Chinese Economics Text,” Journal of the Economic and Social History of the Orient 50.1 (2007), 19–40. For Sima Qian’s lengthy extract from this text, see SJ, 129, p. 3256. SJ, 129, p. 3282. 63 SJ, 30, p. 1442; 129, p. 3306. 64 SJ, 129, p. 3307.

191

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

that wielded such enormous power. While the individual properly should focus on enriching his family, the ruler must pursue the very different agenda of enriching society and preventing exploitation of the poor by the rich. In the eyes of other philosophers and statesmen, however, Sima’s formulation of different sets of values for the individual and the state was logically flawed, and few would endorse his positive view of unfettered entrepreneurship.

Triumph of the Farming Habitus The unprecedented intrusion of the autocratic state into economic life during Emperor Wu’s reign in turn provoked sharp conservative reaction, ultimately leading to the repeal of much of Wu’s fiscalist agenda. Overt challenges to Wu’s fiscal policies were rare during the emperor’s lifetime. But soon after Wu’s death in 87 B C E dissident officials sharply questioned the morality of state control over commerce and industry. In 81 B C E the legacy of Wu’s fiscal policies became the subject of a formal debate convened by Wu’s successor, Emperor Zhao (r. 87–74 B C E). Sang Hongyang – still a senior statesman at the Han court – was called upon to defend prevailing policies such as the salt and iron monopolies against an array of sixty-odd “learned men” summoned to the capital to present their grievances. A digest of the debate between Sang and his Confucian detractors has been preserved in dialogue form in the work entitled The Salt and Iron Debates. In the Debates, Sang Hongyang initially defended Wu’s fiscal policies as necessary responses to the menace of steppe nomad invasions and the rising costs of war and defense. But Sang quickly dispelled the idea that these policies were simply emergency expedients, and instead insisted on the ruler’s responsibility to manage the economy for the benefit of the commonwealth in much the same language as Pseudo-Guanzi: The true king should monopolize natural resources, regulate customs barriers and marketplaces, ensure that the tasks appropriate to each season are fulfilled in a timely manner, and govern the people by controlling the ratios of exchange. In good years of bountiful harvests the ruler stores up goods to provide for times of scarcity and want. In bad years of meager yields the ruler disburses money and goods, circulating his accumulated surplus in order to make up for shortfalls.65

Beyond this affirmation of mercantilist principles, Sang justifies his policies by pointing to their achievements: rationalization of imperial finances; 65

Chapter 2, “Ligeng” 力耕, in YTL, 1, p. 28.

192

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

establishment of a sound, unified currency; strengthening the agricultural base of the empire through irrigation and flood-control projects, the diffusion of advanced farming technologies, and sponsorship of agricultural colonies on frontiers; and equalization of the tax burden among different regions. He also upheld the managerial role of the state by extolling the virtues of systematic, long-range planning in contrast to the short-term, survivaloriented decision making of individual households. Moreover, Sang Hongyang underscored the crucial role of the state in achieving economic justice. He repeatedly urges the ruler to take decisive action to “place a yoke on the noble princes,” “expel the rich and great merchants,” and prevent “untitled power mongers” from dominating the trade in vital goods such as salt and iron for their personal profit.66 If the ruler relinquishes the “levers of profit” (quanli 權利) – the power to coin money, the wealth generated by trade in salt and iron – then private self-interest will overwhelm the public good, and aggrandizers will impoverish the people. But if the government acts to create “balance” (heng) and ensure “stability” (zhun) – again, the keys of Pseudo-Guanzi’s political economy – then “even if you send a knee-high child to the market, none would be able to deceive him.”67 Sang drew on the history of Guan Zhong’s home state of Qi to reinforce his argument for the necessity of the ruler to control the sources of wealth. Qi was blessed with natural resources of timber, minerals, and salt, but it was Guan Zhong’s determination to exert sovereign control over the wealth of the land that enabled his liege, Lord Huan, to attain hegemony within the Zhou polity. However, Lord Huan’s successors relinquished control over resources to their vassals, and the rulership of Qi was usurped by the Tian clan. Sang’s interpretation of the fate of the house of Qi was diametrically opposed to Yan Ying’s accusation that the Qi rulers had forfeited their legitimacy by monopolizing the wealth of the state.68 In response, the “learned men” (mostly middle-level officials) conjoined the basic principles of the farming habitus – beginning with the assertion that agriculture is the source of all wealth – with Confucian moral dicta concerning the iniquity of the profit motive, the virtue of frugality, and the dignity of poverty. The learned men interpreted quanli not only as the instruments for amassing wealth and power, but also as a psychological disposition, a “calculus of profit,” in which self-interest and material gain eroded all consideration 66

67 68

Chapter 2, “Ligeng”; Chapter 14, “Qingzhong” 輕重; and Chapter 5, “Jingeng” 禁耕; in YTL, 1, pp. 27, 179, 67 respectively. Chapter 5, “Jingeng” 禁耕, in YTL, 1, p. 68. Chapter 9, “Cequan” 策權, in YTL, 1, p. 120.

193

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

of ethical conduct. Beyond condemning the state’s expropriation of commercial and industrial enterprises, the learned men repudiated the market economy itself as corrupting. They asserted that in antiquity the minimal exchange necessary for meeting subsistence needs was conducted through barter. The introduction of money – and even the specialization of labor – inevitably promoted dishonesty, conflict, and the breakdown of social solidarity. Restoration of the farming habitus must be reinforced by a rededication of government and its agents to the task of the moral regeneration of society: The task at the present moment lies in eradicating the miseries of cold and hunger by abolishing the salt and iron monopolies, repudiating the calculus of profit, dividing and allocating arable lands, urging dedication to the primary occupation, fostering cultivation of mulberries and hemp, and maximizing the productivity of the land. If the ruler restrains his ambitions and curbs his expenditures, the people will enrich themselves.69

Rebuffing mercantilist doctrines, the learned men espoused a physiocratic vision of society founded on self-sufficient farming families shielded from the economic uncertainties and moral depravity of the marketplace. In their view, not only was the imperial state profiting at the expense of its subjects, but the emphasis on increasing revenues distracted officials from their proper duties as guardians of the spiritual and material welfare of the people. But in appealing for a return to the laissez-faire regime of Emperor Wen they abetted the interests of the great landowners and noble lineages. The court debate of 81 B C E did not fundamentally alter the Han state’s fiscal regime; Emperor Zhao was persuaded to retain the salt and iron monopolies. But in the following decades the conservative reaction against Wu’s fiscal strategies swelled. The anti-statist ideology and physiocratic principles enunciated by Sang Hongyang’s critics were widely embraced by the rising class of Confucian statesmen. These officials viewed the emperor essentially as a paragon of moral excellence whose activities should be largely confined to his ritual duties. They rejected not only the strong imperial controls over commerce and industry enacted by Emperor Wu, but also the market economy itself. Under the aegis of Emperor Yuan (r. 49–33 B C E ), the political ideology of this group became deeply insinuated in government policy and institutions. In 44 B C E the Confucian minister Gong Yu advocated the restoration of a barter economy by abolishing metallic money altogether. 69

Chapter 36, “Shuihan” 水旱, in YTL, 1, p. 429.

194

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

Gong’s proposal was deemed impractical, but reforms enacted by Emperor Yuan focused on reducing government expenditures and especially wanton consumption by the imperial household. Yet as the state retreated from the activist agenda of Sang Hongyang, the concentration of landownership and wealth intensified. Over the next several decades an ascendant class of Confucian statesmen fashioned themselves into a new oligarchic elite, the magnate clans (haozu 豪族). The political triumph of the magnate clans also set in motion important economic changes, most notably a reorientation of wealth and investment from commerce to landed estates. Wang Mang, the chief minister who usurped the Han throne in 9 C E, attempted to check the rising power of the magnate clans by reviving – and in some respects exceeding – the mercantilist policies of Emperor Wu. But fierce resistance to Wang’s draconian initiatives – including nationalization of landownership, prohibition of interest, and institution of a bewildering variety of new bronze currencies – stymied his efforts. The restoration of the Han dynasty after Wang was overthrown and put to death in 23 C E heralded the complete triumph of the magnate clans over Han government, society, and economy. The Eastern Han (25–220 C E) court relied almost entirely on in-kind taxation (mainly from direct taxes) while abolishing universal military conscription, thus repudiating both Wang Mang’s mercantilist policies and the militarist–physiocratic fiscal regime of the early Han. The basic principles of this minimalist fiscal regime were enunciated in Ban Gu’s History of the Former Han Dynasty (c. 100 C E), which became the official history of the Western Han era. In his chapters on “Food and Money” and “Biographies of the Moneymakers” Ban Gu rewrote the economic history of the early Han as recorded in Sima Qian’s Historical Records, erasing Sima’s positive evaluation of commercial entrepreneurship and reasserting the primacy of the farming habitus as the foundation of a ritually ordered moral community. Ban reiterated the Mencian principle that the moral transformation of the people proceeds from providing for their subsistence needs: “once the people are well fed and goods circulate unhindered, the state can be sustained, the people enriched, and moral renovation (jiaohua 教化) brought to completion.”70 Ban recentered economic life on the benevolent work of the sage-kings, who fashioned tools, built cities, opened marketplaces, invented money, and founded schools to level the disparities between rich and poor. Ban also urged the rulers of his day to restore the principles of the ancient “well-field” system of equal land allotments to nurture a 70

HS, 24A, p. 1117.

195

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

communal fraternity of mutual solicitude. The pastoral utopia Ban envisioned was cogently summarized in his idealization of women’s work: during the winter months spent indoors, women living in the same neighborhood gather in the evenings to weave and sew, accomplishing forty-five days of labor in the span of a month; they save on expenses for heating and lighting and equalize the level of their handiwork, thereby harmonizing skills and customs.71

Ban Gu fused the principles of the farming habitus to a vision of a strong monarchy drawn essentially from the writings of Jia Yi, which Ban quoted at length. Absent the chronic warfare and economic competition of the Warring States, the emperor need not amass great wealth. But restoration of a tributary order based on the “well-field” system in itself was insufficient for ensuring the people’s welfare. The ruler must take an active role in shaping the social hierarchy, establishing a division of social labor among the four occupational categories of governors (the learned elite), farmers, artisans, and merchants by assigning tasks according to the abilities of each individual. Ban noted with approbation that previous Han emperors had strengthened the agricultural base by disseminating knowledge of farming techniques, distributing tools and oxen, stockpiling grain, and intervening in markets to maintain price stability. Ban shared the apprehensions of Jia and others about the incompatibility of the family-farm economy and the values of the marketplace. In his rewriting – an inversion, really – of Sima Qian’s chapter on the moneymakers, Ban discarded Sima’s celebration of merchants as the “untitled nobility” and instead upheld the traditional status order, depicting merchants as avaricious thieves who impoverish the state and the people alike.72 Yet he did not join Gong Yu and others in repudiating the use of money. Instead, Ban sought to subordinate market exchange to a tributary economy focused on the satisfaction of needs rather than an acquisitive economy motivated by wants. In this way Ban distinguished the moral ethos of the providential state from earlier advocates of the farming habitus such as Li Kui and Shang Yang, who were solely focused on strengthening the power of the state. The vision of the providential state expressed in Ban’s treatise would become the canonical paradigm of political economy for centuries to come.

71

72

HS, 24A, p. 1121. By working at night, women in essence work at a time-and-a-half pace, hence “accomplishing forty-five days of labor in the span of a month.” HS, 91, pp. 3679–94; Chin, Savage Economy, pp. 185–7.

196

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

Post-Han Fiscal Regimes The minimalist fiscal regime established under the Eastern Han was maintained in different forms under the numerous dynastic houses during the Period of Disunion that sprang up after the demise of the Han in 220 C E . But the militarist–physiocratic model was revived in the late fifth century C E in the fiscal reforms adopted by the Northern Wei state in its effort to reestablish a stable agrarian order after two centuries of anarchic warfare and devastation in north China. The centerpiece of the Northern Wei reforms was the equal-field (juntian 均田) system, in which the state allocated arable lands for food and textile production to individual households based on their productive capacities, measured in able-bodied adults, oxen, and slaves (see von Glahn, Chapter 3 in this volume). The enactment of the equal-field system was accompanied by a new tax code in which the amounts of grain, cloth, and labor service owed to the state were directly proportional to the land allocation the household received. Thus tax rates were relatively uniform; in principle, the tax and labor service obligations of every adult male were equal. In a departure from the original militarist–physiocratic model, the Northern Wei and its successors mostly recruited their armies from a separate caste of hereditary soldiers rather than imposing universal military conscription. The equal-field system was intended to resettle the war-torn population, stabilize the state’s revenue base, and re-establish the family farm as the social and economic base of the empire. It also represented a withdrawal from the market economy. In the post-Han era, coin virtually disappeared from the north China territories under the rule of foreign potentates. In contrast to the substantial money revenues extracted by the Western Han state, the equal-field system relied solely on in-kind payments and labor services. A vibrant money economy endured in the Chinese-ruled states of south China despite chronic shortages of coin. Under the Sui and Tang dynasties that reunified China in the sixth to seventh centuries, the equal-field system and in-kind taxation were retained in north China, but no effort was made to extend them to the southern territories. In 624 the Tang introduced a taxation system based on the equal-field system of land tenure known as zu–yong–diao 租庸調 after its three principal components: a land tax paid in grain (zu), a tax levied on households paid in cloth (diao), and a requirement of labor service assessed on adult males (yong). A separate tax system, including some payments in coin, was implemented in the southern provinces. The Tang fiscal system thus relied heavily on taxes in grain and cloth levied on agricultural

197

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

producers, and captured little revenue from the non-farm sectors of the economy. Over the course of the first century of Tang rule the equal-field landholding system gradually weakened and collapsed. The restoration of domestic peace under Sui and Tang rule unleashed dynamic economic forces that strained against the rigid uniformity of the equal-field system. Population growth reduced the availability of land in long-settled north China, and many farm families received only half or less of their nominal allocations. The burden of taxation fell heavily on north China’s small farmers, many of whom were forced to relinquish or abandon their landholdings. Thus the problem of land concentration in the hands of aristocrats, officeholders, and now Buddhist monasteries resurfaced. By the eighth century, “vagrancy” had become an acute fiscal as well as social problem. Many registered households no longer submitted tax payments and labor service, and officials tried to shift their obligations onto the rest of the local community. The eruption in 755 of the An Lushan Rebellion, led by a disgruntled frontier general of Sogdian descent who turned against the imperial court, dealt a shattering blow to the Tang dynasty. The court was forced to flee from the capital, Chang’an, which was seized by An Lushan in 756 and later sacked by a marauding Tibetan army, and seek refuge in the southwest. By the time the rebellion was quelled in 763 the rich agricultural heartland of the Central Plain lay in ruin and hundreds of thousands had perished. The Tang dynasty survived, but it emerged from the rebellion only a shadow of its former self. Effective control over most of the northern provinces was ceded to regional warlords. The equal-field system, already under serious strain, collapsed, compelling the Tang leaders to devise a new fiscal system. After the outbreak of the rebellion the Tang court readily assented to radical measures to shore up state revenues, most notably a proposal to revive the state monopolies on salt and iron that existed during the Han. In 758, the court established a monopoly on the production and sale of salt under the authority of a newly established salt commission. Liu Yan, who occupied the post of salt commissioner from 760 to 779, transformed the salt commission into a supremely powerful organization. Salt revenues rose tenfold over the course of Liu’s tenure as salt commissioner, and by 779 provided the central government with more than half of its annual income. Liu Yan also gathered mining, coinage, and the transport of tax grain from the south to the capital under the purview of the salt commission. Subsequent salt commissioners enacted additional consumption taxes on wine and tea, though these levies generated only modest revenues. 198

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

The virtually unchecked power wielded by Liu Yan inevitably provoked frictions with central government officials. In 779 a bitter rival of Liu, Yang Yan, took office as chancellor, determined to wrest control of the state’s fiscal administration away from the salt commission and return it to the finance ministry. Yang also launched a sweeping reform of the tax system that sought to restore direct taxation as the main source of state revenue. Although Yang’s attempt to abolish the salt commission was aborted, the new tax system he created would endure throughout the rest of the imperial era. Yang Yan’s reform acknowledged that the equal-field system of landownership was defunct, and along with it the zu–yong–diao taxes that had been the centerpiece of the Tang fiscal regime. Yang’s plan formally abolished the zu–yong–diao assessments while incorporating the household and land levies into a new tax structure that came to be known as the twice-a-year tax (liangshui 兩稅). Households were ranked into nine property grades and assessed a tax measured in coin. Given the endemic shortage of coin, however, the household tax generally was commuted to commodities, primarily cloth. Land tax rates became highly variable, with small family farms assessed only modest payments in grain. The household tax was collected in late summer and the land tax after the autumn harvest, hence the name twice-a-year tax. The principle of equity that underlay the equalfield system and earlier measures to restrict the concentration of landholding was abandoned, never to be resurrected in any serious way before the Communist land collectivization policies of the 1950s. The post-An Lushan fiscal system created by Liu Yan and Yang Yan marked a dramatic departure from the militarist–physiocratic and providential models of political economy. The shift from direct taxation of land and population to indirect taxation of commerce and consumption was unprecedented. The twice-a-year tax, although paid by households, was a tax based on wealth; moreover, household rankings were determined by a calculation of all assets (caili 財力), not just arable land, and thus captured revenues from trade, finance, and urban real estate that had remained untaxed under zu– yong–diao. The twice-a-year tax’s steeply graduated rates introduced a strong measure of progressivity into the tax system that recognized the wide disparities in wealth that resulted from the growth of the market economy. Labor services were curtailed, and obligations for military service – both from the hereditary military households and through conscription of the general population – abolished. Instead, the late Tang state and its successors relied on a professional soldiery paid in provisions and money. The shift away from persons and labor to property and wealth as the objects of taxation and from 199

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn

in-kind to money payments freed households and individuals to allocate their labor and capital more efficiently. These changes in the fiscal structure of the state accommodated the market economy and undoubtedly gave further impetus to its growth. At the time, however, this new fiscal system provoked heated protest among leading scholars and statesmen. The most trenchant critique was voiced by Lu Zhi upon his appointment as state councilor in 792.73 Lu echoed militarist–physiocratic principles in his condemnation of the ways in which the twice-a-year tax abetted the growth of the market economy and acquiesced to the widening disparity between rich and poor, but ultimately he adopted a providential posture that implicitly condoned the concentration of landownership in the hands of wealthy families. In antiquity, Lu asserted, taxation essentially was based on the labor power of adult males; rulers did not vary the rate of taxation based on the success or failure to attain wealth. Thus the principle of uniform taxation of people promoted thrift, diligence, and savings among the people; the poor could not evade their obligations to society by virtue of being poor. In contrast, taxation levied on wealth – reduced to an abstract monetary sum – engendered profound inequities. The twice-a-year tax ignored how wealth is used; it discouraged investment (which was visible) and encouraged hoarding (which was not). Moneylenders made significant profits from even the smallest loans, while fixed assets like dwellings yielded no income. To treat all wealth as the same exacerbated the worst features of the monetary economy (usury, forestalling) rather than rewarding the perspicacity and honest toil of the deserving rich. Lu reiterated many staple tenets of the providential orientation: the twice-a-year tax accepted social inequality rather than seeking to rectify it; it focused officials’ attention on maximizing revenues rather than minimizing expenses; and taxation in money compelled farm families to sell their produce and thus exposed them to the vicissitudes of the marketplace. Yet at the same time Lu departed from the communitarian ethos of Mencius’ well-field ideal in favor of using taxation as a means to promote the farming habitus and its moral values. The intellectual luminaries of the day – such as Han Yu, Bai Juyi, and Li Ao – joined with Lu Zhi in condemning the new fiscal regime. Their dissent wrested some concessions, especially the substitution of textile goods in lieu 73

“Junjie fushui xu baixing liutiao” 均節賦稅恤百姓六條, in LXGJ, pp. 243–61.

200

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

Economic Philosophy and Political Economy

of money payments, a necessary compromise because of the state’s inability to mint sufficient coin and the sharp deflation that followed the adoption of Yang Yan’s tax system. But the Tang state retained the twice-a-year tax along with Liu Yan’s consumption taxes, and these institutions became the foundation for fiscal governance during the post-Tang period of disunion (the Five Dynasties era, 907–960) and under the Song dynasty (960–1279) as well. If it was still premised on an agrarian economic base, this new fiscal regime acquiesced to the autonomy of the market and sought to harness, rather than arrest or subvert, market forces to meet the state’s fiscal and economic goals. This new emphasis on a co-operative rather than antagonistic relationship with the market and the merchant class – what we might call the synergistic fiscal state – represented a complete break from the principles of both the militarist–physiocratic and providential models of fiscal governance.

Further Reading Chin, Tamara, Savage Exchange: Han Imperialism, Chinese Literary Style, and the Economic Imagination (Cambridge, MA, Harvard University Asia Center, 2014). Graham, A.C., “The ‘Nung-chia’ 農家 or ‘School of the Tillers’ and the Origins of Peasant Utopianism in China,” Bulletin of the School of Oriental and African Studies 42.1 (1979), 66–100. Hu Jichuang, A Concise History of Chinese Economic Thought (Beijing, Foreign Languages Press, 1988). Hu Jichuang 胡寄窗, Zhongguo jingji sixiang shi 中國經濟思想史, 3 vols. (Shanghai, Shanghai renmin chubanshe, 1962–1988). Kanaya Osamu 金谷治, Kanshi no kenkyu¯ 管子の研究 (Tokyo, Iwanami shoten, 1987). Lévi, Jean (trans), La dispute sur le sel et le fer (Paris, Belles Lettres, 2010). Lewis, Mark, Writing and Authority in Early China (Albany, SUNY Press, 1999). Li Zhixian 李志贤 (Lee Chee Hiang), Yang Yan jiqi liangshuifa yanjiu 杨炎及其两税法研 究 (Beijing, Zhongguo shehui kexue chubanshe, 2002). Lin, Cheng, Terry Peach, and Wang Fang (eds.), The History of Ancient Chinese Economic Thought (London, Routledge, 2014). Lin, Cheng, Terry Peach, and Wang Fang (eds.), The Political Economy of the Han Dynasty and Its Legacy (London, Routledge, 2019). Masubuchi Tatsuo 増淵龍夫, Chu¯goku kodai no shakai to kokka 中國古代の社會と國家 (Tokyo, Ko¯bundo¯, 1960). Nylan, Michael, “Assets Accumulating: Sima Qian’s Perspectives on Moneymaking, Virtue, and History,” in Hans van Ess, Olga Lomová, and Dorothee SchaabHanke (eds.), Views from Within, Views from Beyond: Approaches to the Shiji as an Early Work of Historiography (Wiesbaden, Harrassowitz Verlag, 2015), pp. 131–68. Pines, Yuri, The Book of Lord Shang (New York, Columbia University Press, 2016).

201

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

richard von glahn Sabattini, Elisa, and Christian Schwermann (eds.), Between Command and Market: Economic Thought and Practice in Early China (Leiden, Brill, 2021). Sterckx, Roel, “Ideologies of the Peasant and the Merchant in Warring States China,” in Yuri Pines, Paul R. Goldin, and Martin Kerr (eds.), Ideology of Power and Power of Ideology in Early China (Leiden, Brill, 2015), pp. 211–48. Swann, Nancy Lee, Food and Money in Ancient China: The Earliest Economic History of China to A . D . 25 (Princeton, Princeton University Press, 1950). von Glahn, Richard, “Modalities of the Fiscal State in Imperial China,” Journal of Chinese History 4.1 (2020): 1–29. Wu Song 吴松 et al., Zhongguo nongshang guanxi sixiang shigang 中国农商关系思想史纲 (Kunming, Yunnan daxue chubanshe, 2000).

202

https://doi.org/10.1017/9781108587334.006 Published online by Cambridge University Press

6

Silk Road Trade and Foreign Economic Influences xinru liu

The Silk Road trade, which involved mostly prestige goods, started from the Han dynasty, around the second century B C E, under the protection of Han imperial expansion into Central Asia. The economy of the Han Empire was mainly based on agriculture. Taxes in the form of agricultural products – such as food grains, silk yarn and floss, and bast-weave cloths such as ramie and hemp – in addition to corvée labor provided the major revenue for the state. Although commerce flourished in cities and connected both rural and urban residents into a nationwide market, traders held the lowest status in the social hierarchy. The impetus for trade with foreign countries, therefore, was initiated by the Han ruling elite, who, like aristocrats in ancient regimes around the world, had always been looking for rare and expensive goods to mark their distinguished status. Meanwhile, the Han Empire engaged in warfare with pastoral nomads of the Central Asian steppe grasslands from the founding of the dynasty. The perennial wars with the Xiongnu nomad confederation extended the horizon of the Han rulers, north to the steppe and west to Central and South Asia, reaching as far as the Mediterranean. The curiosities brought to the Han court by generals, envoys, and traders encouraged further expeditions. Silks used as diplomatic gifts to the nomads on the steppe and the agricultural communities of the oases around the Tarim Basin roused the appetite for Chinese goods in Central Asia and eventually created a market for Chinese silks farther afield in South Asia, West Asia, and the Mediterranean. The collapse of the Han Empire in the early third century C E disrupted the established pattern of the trade between agrarian China and lands to the west, however. After warlords divided China into the Three Kingdoms (220–265),

203

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

which were briefly unified by the Western Jin (265–317), nomadic invaders overran north China, sacking the Jin capital of Luoyang in 311 and forcing the Jin court to flee to south China. New rulers from the steppe, the so-called “five barbarian tribes,” established altogether some sixteen regimes scrambling for dominance in the north, which eventually were consolidated by the Tuoba lineage of the Xianbei people into the Northern Wei Empire (386–534). During the fifth and sixth centuries, the Northern Wei built political and economic institutions imitating those of their newly conquered Chinese subjects. Meanwhile, many of the remnants of the Chinese aristocratic households, including the artisans and cultivators who served them, migrated to the south following the relocation of the Jin court. The northerners crossed the Yangzi river and expanded the agrarian economy, adapted to rice cultivation, in the vast river valleys and marshlands of the south. The demographic shifts that accompanied the frequent changes of political regime and rupture of the Confucian patriarchal family system transformed not only the economy but also the cosmology and eschatology of the Chinese people. Buddhist theology and institutions started to take root in China. Buddhism introduced ideas of a cosmos where all creatures rotated among several domains of mortal existence, from the sublime to the purgatorial, through cycles of rebirth. The new cosmology was added to the traditional Chinese universe, conceived as a hierarchy dominated by patriarchal ancestors. In the Buddhist universe, the domain of one’s rebirth – divinity (deva), titan (asura), human, animal, hungry ghost, or denizen of the underworld – was determined by one’s merits (or sins) accumulated over previous lifetimes. Buddhism attracted not only those who left their families to join monastic communities (or those bereft of families altogether), but also groups of lay followers who looked for the meaning of life under religious guidance at a time when frequent migrations tore apart the fabric of family and kinship. In this way, Buddhist monasteries helped repair the broken social structure with a kind of non-governmental, non-family-based community that had not previously existed in Chinese society. The demands of the new rulers and of the new religious institutions sustained streams of commodities along the turbulent Silk Road; at the same time, development of Afro-Eurasian maritime commerce reached ports along the southeast coast of China. When the Tang dynasty, whose rulers also originated from the steppe, consolidated its imperial power in the early seventh century, a new age of foreign trade was inaugurated that pushed the boundary of Chinese cultural influence farther and in all directions. Missions by Buddhists and other religious evangelists accompanied the 204

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

overland caravans and sailed on merchant ships landing in the cities and ports of China. Zoroastrian temples were depots for foreign traders, mostly Sogdians from Central Asia. Nestorian and Manichaean priests sought patronage from the imperial court and provided shrines for their fellow traders to gather and worship. The sheer volume of more and more varieties of goods, both coming into and going out of China, released foreign trade from the grip of the government and the ruling elite. Religious institutions, above all Buddhist ones, legitimated ostentatious expenditure on prestige goods. The spread of Islam throughout the Central Asian trade networks by late Tang times engendered a new revolution in China’s international trade. The new setting of commodity exchanges probably was the last straw for sumptuary laws regulating consumption patterns and the beginning of free markets in China’s foreign trade.

The Western Region of the Han Empire The conflicts between the Han Empire and the Xiongnu nomads paved the path for the empire’s westward trade through Central Asia. When Liu Bang (Emperor Gaozu, r. 202–195 B C E), the founder of the Han Dynasty, ascended the throne in 202 B C E, he took over a turbulent northern frontier amid the post-Qin collapse. The Xiongnu united a group of steppe tribes to form a farflung confederation on the northern frontier of China. The First Emperor of Qin had built the Great Wall to defend agricultural settlements from Xiongnu raids, but the wall was never a secure defense without strong military garrisons. In 200 B C E, Liu Bang was besieged by the Xiongnu at Pingcheng, near modern Datong inside the Great Wall, for seven days. Fortunately for the emperor, the Xiongnu raiders did not intend to occupy Han territory, nor to kill the emperor, but rather sought booty in the forms of silk, grains, and other agricultural products. Liu Bang agreed to a treaty with the Xiongnu acknowledging the Xiongnu chieftain, or shanyu, as his equal. The Han would annually supply the Xiongnu with grain, wine, and silk in addition to the dowry gifts accompanying the princess dispatched to marry the shanyu; in return the Xiongnu should respect the Great Wall as the border between the two states. After decades of humiliation and harassment at the hands of the Xiongnu, Emperor Wu (r. 140–87 B C E) took the offensive. His multi-pronged strategy included military expeditions deep into the steppe, diplomatic efforts to make alliance with adversaries of the Xiongnu, and strengthening and extending the Great Wall both as a military perimeter and as a zone for cross-border 205

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

trade. Wu’s generals conducted successful campaigns to drive the Xiongnu north of the Gobi Desert, but they could not prevent the nomads from returning once the Han armies withdrew inside the Great Wall. Fleeing to the north was just a tactical retreat to avoid costly confrontations, not a military defeat for the Xiongnu. Diplomacy was necessary to deter the Xiongnu, but the princesses married to the Xiongnu chiefs had yet to produce allies of the Han court. The Han court thus tried to get information about the politics on the steppe, and to make allies of the adversaries of the Xiongnu there. As no court officials dared to take up the difficult mission, Emperor Wu recruited Zhang Qian, a petty officer, to make allies on the steppe. The goal of Zhang Qian was to reach the Yuezhi, archenemies of the Xiongnu. With a retinue of a hundred followers, including Ganfu, a native of the steppe, Zhang set out in 139 B C E. But Zhang’s company was apprehended by the Xiongnu while traversing their territory and ultimately detained at the Xiongnu camp for ten years, during which time Zhang took a Xiongnu wife. He eventually managed to escape, together with Ganfu, and reached the Yuezhi camp on the Oxus river in 129 B C E. The Yuezhi, however, had retreated to the banks of the Oxus because the Xiongnu had defeated them and chased them away from the pastures in the foothills of the Qilian mountains. Instead of returning to face the Xiongnu, the Yuezhi crossed the Oxus river into Bactria, a fertile agricultural land dotted with Hellenist cities. Zhang Qian lingered with the Yuezhi for a year. Although he failed to persuade the Yuezhi to fight the Xiongnu, Zhang Qian learned much about Central Asia. On his return to Chang’an, Zhang Qian again fell into the hands of the Xiongnu, but this time he managed to escape only one year later, when the aging shanyu died and civil war broke out among his descendants. Upon arriving at Emperor Wu’s court, Zhang Qian opened the eyes of the monarch to the wonders of the “Western Region” (xiyu 西域) extending all the way to the Pamir mountains, and even beyond. Though the Yuezhi declined to join the Han to fight the Xiongnu, gift exchanges between the two resumed. A bronze mirror of Han manufacture, excavated from Tillya Tepe, site of the tombs of a Yuezhi chief and five court women, verifies the connections the Yuezhi maintained with the Han court after they crossed the Amu river. Tillya Tepe is renowned for the more than 20,000 pieces of gold artifacts excavated from the tombs. If the aristocrats were wearing exquisite silk textiles, as was likely, the fabrics decayed without a trace, although the remaining gold ornaments betray the contours of the pants, shirts, and robes of the tomb occupants. A bronze mirror, placed on the chest of the 206

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

princess of Tomb I I, is a genuine Han mirror inscribed with a Chu-style poem of thirty-four characters.1 The mirror might be one of the official gifts from the Han court to the Yuezhi rulers. While engaged in war against the Xiongnu, the Han Empire consolidated control over the Great Wall frontier zone and extended its defenses to the western end of the Hexi corridor between the Kunlun mountains and the Gobi Desert. The terminal garrison was called Jade Gate, as it was the gateway to Khotan, an oasis on the southern rim of the Taklamakan Desert, in a region rich in jade deposits. The Great Wall was not only a barrier to deter the Xiongnu and other nomads from looting villages and towns on the frontier region; it also facilitated peaceful trade. Garrisons along the wall were checkpoints where travel documents such as passports, or guosuo 過所, issued by various authorities were examined, and also the sites of periodic market fairs. Nomads brought in their horses, cattle, and other steppe products to exchange for grains, silk, wine, and other goods of settled society. The garrisons on the western extension of the wall also were stepping-stones for the import of exotic goods from Central Asia and regions farther west. Maintaining the system for defense as well as trade was expensive. Therefore the Han Empire established farmer-soldier colonies known as tuntian 屯田 to reduce the cost of defending its northern frontier. Providing agricultural tools and technological assistance, the state put soldiers and their families to work reclaiming land for agriculture and building elaborate irrigation systems around the garrisons. The soldiers worked whenever they were not occupied with military action.2 The tuntian system became an institution regularly practiced by Chinese imperial governments down to modern times. It could not make the frontier military camps wholly self-sufficient, but nevertheless it reduced the burden of transporting provisions from interior regions. The agricultural reclamation efforts not only introduced technology from China, but also facilitated innovations to cultivate the arid lands along the Great Wall frontier, which essentially marked the divide between arable lands and desert or semi-desert terrain. For example, the well-canal, an irrigation technique for raising underground water and channeling it to the fields, is frequently mentioned in documents written on wood slips retrieved from archaeological excavations in the 1

2

Li Xueqin 李学勤, Bijiao kaoguxue suibi 比较考古学随笔 (Guilin, Guangxi shifan daxue chubanshe, 1997), p. 68. Xinru Liu, The Silk Roads: A Brief History with Documents (Boston, Bedford/St. Martin’s, 2012), pp. 49–50.

207

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

vicinity of the garrisons. Where the agricultural settlements prospered, they could host more travelers, not only government officials and soldiers, but also traders who ventured beyond the Great Wall to do business. The traders, like other travelers in the region, needed passports issued by local government offices to pass beyond the Great Wall, as did traders from outside seeking to enter Han territories.3 The oases of Central Asia, after all, were not protected by the wall. The Xiongnu and other nomads fought with each other for control of the oases and the trade routes passing through the region. Under Emperor Wu, the Han government established suzerainty over the Tarim Basin, and occasionally dispatched military forces to the large oasis states of the Amu–Syr plain. For better control of the routes west of the Great Wall, Emperor Wu extended the tuntian system to garrisons beyond the Jade Gate around 100 B C E. The oases on the rim of the Taklamakan Desert benefited from the new agricultural techniques and as a result their populations steadily increased. Settled farmers living in the oases tended to ally with the Han instead of the Xiongnu, although at times they were subject to Xiongnu predation when the Han military failed to deter raids. Despite the Xiongnu menace, a commercial network emerged to facilitate the movement of goods across Central Asia. In addition to using bolts of plain silk textiles as media of exchange, oasis authorities issued copper coins. Archaeologists discovered several hundreds of copper coins with Chinese characters on one side, and Kharoshthi inscriptions on the reverse, around the Khotan oasis states. In addition, some coins issued by the kings of Kushan – a nomadruled empire centered in modern Afghanistan–Pakistan during the first to third centuries C E – were recast with Kharoshthi legends. This practice suggests that the oasis authorities issued their own coins to maintain trading activities after both the Kushan and the Han retreated from the region in the late second century C E.4 However, throughout the Han dynasty the market for exotic goods from the west was limited to the ruling elite, which itself underwent momentous transformation. Emperor Wu established Confucianism, which espoused the patriarchal family structure as the basis of political and social hierarchy, as the state ideology. The Confucian doctrine that mandated classical learning as an essential qualification for government office fostered a new ruling elite composed of Confucian-educated officials; by contrast, merchants were deemed the least-respected occupational status. Under the Han imperial 3 4

Liu, Silk Roads, p. 49. Lin Meicun 林梅村, Xiyu wenming 西域文明 (Beijing, Dongfang chubanshe, 1995), pp. 309–11.

208

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

order, traders might attain great wealth, but sumptuary regulations prohibited them from them wearing or displaying prestigious fabrics such as silk tapestry and embroidery, or even brightly colored silks. They could not ride horses or carry weapons. In spite of such restrictions on luxury consumption by merchants, trade flourished under the Han. The initial successes of Wu’s military campaigns against the Xiongnu and the extension of Han military control into the Western Region brought rare goods within sight of the Han ruling elite. Traders set out in the wake of the military expeditions to acquire exotic goods and traveled among the oases under the protection of Han army patrols. The oases around the Tarim Basin, caught between the Han military presence and occasional nomad invasions, more often than not welcomed the Han garrison forces and submitted tribute to the Han court. The Han emperors in turn rewarded the tribute givers with Chinese products, chiefly silk goods. This tributary relationship was the framework for early intercultural trade under the guise of Chinese suzerainty. Archaeological remains of imported luxuries during the Han times are scanty but revealing. A typical Roman blown-glass vase dated to the second century C E was excavated at a suburb of Luoyang city. Certain Roman and Central Asian goods deposited in tombs of post-Han elites probably arrived during the Han times and had been treasured and circulated for several centuries. For example, a fish-shaped glass artifact from the tomb of a Northern Yan aristocrat who died in 415 bears strong similarity with several similar glasswares found in the Begram hoard excavated at the Kushan summer capital in Afghanistan, which is dated several centuries earlier. Gold beads decorated with double-lined granules on a necklace of the Yuezhi princess in Tomb I I of Tillya Tepe were probably Central Asian crafts. A necklace of similar gold beads from an elite tomb of the Eastern Han (25–220 C E) also could have come from Central Asia. Another Eastern Han tomb, probably that of a member of the imperial household, contained a copper vessel decorated with gems, including red coral and lapis lazuli. Both red coral of the Mediterranean and lapis lazuli of Badakhshan of Afghanistan were hot commodities in the Silk Road trade.5 For the Han emperors, the most essential imports from the Western Region were the beautiful horses of Dawan, an oasis state west of the 5

Xu Pingfang 徐苹芳, Sichou zhi lu kaogu lunji 丝绸之路考古论集 (Shanghai: Shanghai guji chubanshe, 2017), pp. 53, 69, 84, 92; Fredrik Hiebert and Pierre Cambon (eds.), Afghanistan: Crossroads of the Ancient World (London, British Museum Press, 2011), p. 172, Plates 52, 53, 54; p. 253, Plate 159.

209

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

Pamir mountain range in modern Uzbekistan. By the second century B C E, the Yuezhi, who established a partnership with the Chinese rulers to supply horses and jade for silk, had settled in Bactria, in present-day northern Afghanistan.6 Although Zhang Qian’s mission to induce collaboration from the Yuezhi to fight against the Xiongnu failed, Zhang brought information of another source of horse supply to the emperor. The lush lands of Ferghana, where Dawan established a powerful state, produced a species, described as “blood-sweating horses,” fed on alfalfa, a much more nutritious fodder than the grass of the arid Mongolian steppe. Dawan was also famous for its viticulture and grape wine. Emperor Wu was eager to obtain the Dawan horses to boost his military power and glorify his court, but the lands west of the Pamir mountains were beyond Han control. When Wu sent a delegate to trade with the Dawan, the Dawan king was so arrogant that he not only declined the request but also killed the envoy. Wu then dispatched his general Li Guangli to conquer Dawan and obtain the “heavenly horses” in 104 B C E. It took two expeditions and a high casualty toll for the Han military forces to overpower the defenses of the Dawan capital. General Li returned to Chang’an with 3,000 horses, as well as introducing alfalfa cultivation to China.7 The one-time success of obtaining horses from the Western Region brought in limited numbers of excellent horses, but it set the precedent for the silk–horse exchanges that would last some 2,000 years.

The Silk–Horse Trade and the Tributary System during the Period of Disunion During the post-Han chaos, the exchange of tribute goods became unstable, rising and falling along with the fortunes of dynastic houses. During the short tenure of Western Jin rule (265–317), the regime tried to revive the tributary exchanges with Central Asian rulers. Emperor Wu of Jin (r. 265–290) claimed to have established tributary relationships with twenty-three countries. Although these relationships collapsed after the flight of the Jin court to the south in 317, silk–horse transactions occasionally went through. Gifts of the famous “blood-sweating horses” of Ferghana and excellent horses from Sogdia arrived at the Jin court during this period.

6

7

Xinru Liu, “Migration and Settlement of the Yuezhi–Kushan: Interaction and Interdependence of Nomadic and Sedentary Societies,” Journal of World History 12.2 (2001), 265, 273–4. Liu, Silk Roads, pp. 17–18.

210

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

After rulers from the steppe settled in north China, the horse–silk exchange continued to be the most important trade with Central Asia. Foreign rulers of north China often cultivated allies among Buddhists to help establish their legitimacy, as both were alien to Chinese civilization, or so they thought. However, once established in China, or even just part of it, foreign rulers typically embraced the idea that they were rulers of the “Middle Kingdom,” i.e., the most civilized land. The Northern Wei emperors changed their Xianbei clan surname Tuoba to Yuan, a Chinese-sounding surname, and other Xianbei noble families followed their example by taking Chinese family names. The Northern Wei rulers also employed Confucian scholars and officials to reconstruct imperial rule within their territories. This role of emperor of the Middle Kingdom meant that they should receive tribute from “barbarian peoples” who lived on its periphery and acknowledged the Chinese emperor as the suzerain of the known world. Upon establishing their rule in the Middle Kingdom, the Northern Wei faced the same conflicts with steppe nomads and made the same compromises as had previous Chinese empires (Map 6.1). The Northern Wei rulers thus established tributary relationships with other nomad groups. During the era of Emperor Taiwu (r. 424–452) a nomadic people called the Tuyuhun, based in the highlands of present-day Qinghai province, approached the Northern Wei court to resume their relations with the Chinese monarchs. The envoy stated that because the Tuyuhun defended the northern border of the Wei Empire and had adopted Chinese-style ceremonial carriages and standards, they deserved a greater allotment of wealth, especially silk, to maintain their cultural eminence. The emperor left discussion of the specifics to his courtiers, who decided the amount of gifts by quoting precedents in Han court records. The Tuyuhun also bred a kind of “heavenly horse” which they submitted as tribute in exchange for Chinese silk robes and hats. In 477, the Northern Wei sent 120 bolts of brocade and colored silk textiles to the Tuyuhun to persuade them to make peace with another group of nomads, the Dangchang, probably for the sake of the Northern Wei’s best interests. The silk–horse exchange in the name of a tributary system thus continued. The Northern Wei also reopened communication with the Western Region after it annexed the Hexi corridor leading to the Central Asian steppe and oases. Initially, the Tuoba – who came from Mongolia in the north – were not that interested in cultivating ties with rulers of the Western Region. After consolidating its rule in north China, however, the Northern Wei regime recognized the value of building a network of tributary allies in Central Asia. Oasis towns of the Tarim Basin and Sogdia sent envoys to the 211

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Map 6.1 Silk Road trade routes, c. 500 Source: used with permission from Richard von Glahn, The Economic History of China from Antiquity to the Nineteenth Century (Cambridge, Cambridge University Press, 2016), Map 5.5, p. 197

Silk Road Trade and Foreign Economic Influences

Northern Wei court. Emperor Taiwu, having decided to conquer the Northern Liang kingdom which occupied the Hexi corridor leading to the Western Region, reciprocated by sending envoys with munificent gifts to resume tributary exchanges with the oasis towns. The envoys provided political intelligence about the Western Region extending all the way from the Tarim Basin to the Iranian plateau. Some distant states resumed submission of tribute, including Ferghana, which continued to send “blood-sweating horses.” The tributary system never regained its apogee under the Han Empire, however, as the Northern Wei Empire was much smaller and shorter-lived than the Han. Meanwhile, private entrepreneurs began to frequent the Central Asian oasis towns and trade routes, carrying commercial commodities in both directions. As the Northern Wei rulers encouraged foreign trade, exotic goods filled the marketplaces as well as the palace treasuries. Empress Dowager Hu (d. 528), the most powerful person at the Northern Wei court after the death of her husband, Emperor Xuanwu (r. 499–515), distributed those fancy goods to court officials, and, around the same time, she sent an envoy, the Buddhist monk Huisheng, to make a pilgrimage to India. The tributary transactions and Buddhist missionary and pilgrimage mingled together to facilitate the flow of cross-country trade. It was only after the Han that the tributary system was extended eastward to Korea and Japan. Chinese immigrants had settled in Korea long before Emperor Wu of the Han imposed colonial rule and established four commanderies in the Korean peninsula in 108 B C E. Subsequently, Chinese political administration – and productive technologies such as farm tools, cultivation methods, and bronze and iron metallurgy – spread to the Korean peninsula. Even after the fall of the Han, colonies of Chinese maintained Chinese culture and introduced their technology to local communities, while a series of indigenous Korean states emerged in different parts of the peninsula. During the fourth century C E, when the steppe nomads invaded north China, three major kingdoms occupied almost the entire Korean peninsula: Koguryŏ in the north, Paekche in the southwest, and Silla in the southeast. All three Korean kingdoms shared in the heritage of Chinese culture and practices and continuously absorbed Chinese immigrants. The Koreans were the first carriers of Chinese culture to Japan. Rice cultivation was introduced to Kyushu, the southernmost major island of the Japanese archipelago, from Korea around 400 B C E and subsequently spread to western Honshu. Agricultural development and population growth 213

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

furthered exchanges between the Korean peninsula and the Japanese islands. The first mention in Chinese records of contacts with island countries in the eastern sea date to the third century C E. In 238, Himiko, said to be a queen ruling over numerous polities in the land of Wa (as Japan was called in Chinese records), sent a mission to the court of Wei, one of the Three Kingdoms, bearing humble gifts, including four male and six female slaves and several bolts of cloth. The Wei emperor Ming (r. 226–239) took pity on the poor country and granted Himiko the title “Queen of Wa and Cousin of Wei” carved on a golden seal as well as generous gifts, including silk tapestries, woolen rugs, pearls, and bronze mirrors. Other local rulers in Japan followed by sending their missions, and received rewards. These exchanges favored Japan, which had not yet fully developed. But information about political and social developments in the islands filtered into China, preparing for the time when the Japanese polities developed sufficiently to make true commercial transactions with China. It seems that by 500 C E a powerful lineage of kings had consolidated various local chiefdoms into the Yamato state, and probably occasionally invaded the Korean peninsula. The coming and going of envoys across the Sea of Japan also paved the way for Buddhist missionaries and pilgrims, who traveled across the sea lanes among China, Korea, and Japan, thus extending the Silk Road to the peninsula and the archipelago. By the time the Sui Dynasty unified China in 589, the news from Japan indicated that a monarchy had emerged there. The Sui emperor Wen (r. 581– 604) frowned on their royal rituals and told the Japanese envoys that their rulers should follow Chinese ways of governance. By then, the Yamato kings in Japan had begun to wear crowns made of colorful silk textiles. It seems that the Japanese rulers already were trying to imitate the Chinese court in their use of silk textiles imported from China. Bronze mirrors, a typical Chinese tributary gift to foreign countries, probably reached Japan through Korea. However, the exquisite bronze mirrors preserved in the Sho¯so¯in imperial storehouse in Nara presumably are treasures received from Chinese rulers. Chinese silk textiles and mirrors were precious prestige goods that Japanese rulers distributed among local chieftains to cement political loyalties. Cut off from the steppe regions to the north and west, the rulers of the Southern Dynasties had to look to the eastern and southern seas to explore communications with the outside world. Sun Quan (r. 225–252), ruler of the Wu kingdom in eastern China, sent envoys to countries in Southeast Asia to explore their landscapes and resources. The envoys reported to Sun Quan that traders from India frequented the land of Funan (along the Gulf of Thailand in the lower Mekong delta). Although the Chinese rulers in the 214

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

south lost direct access to the overland Silk Road, they still admired and desired the goods – such as colorful glassware, woolen tapestries and rugs, spices, fragrances, and gemstones – that flowed across it. Some of these goods reached Jiankang – the capital of Wu and the succeeding Southern Dynasties, located at modern Nanjing – through Jiaozhi, in the Red River delta of modern Vietnam. Both the Wu kingdom and its rival, the Shu kingdom in western China, sought to obtain horses from the Southeast Asian highlands, to which Shu had ready access. Sun Quan therefore seized control of Jiaozhi as a port of trade leading to the highland interior regions where horses could be obtained. The horse trade with the southwestern regions flourished until the Tang period, when the imperial state regained access to horses from the long-established trade routes of Central Asia. In the centuries after the demise of the Han, the variety and quantity of commodities imported to China increased, while China’s exports still consisted mainly of silk products, although made using more sophisticated technology and a greater variety of artistic styles. Archaeological findings in Central Asia display developments in weaving and pigments as well as changes of style in the textiles that were traded among China, India, Persia, and the Mediterranean. Changes in cultural values caused realignments of the markets for textiles in general, and silk textiles in particular, throughout the Afro-Eurasian world. Religious movements – Buddhism, Zoroastrianism, Christianity, and Islam – exerted their values in the artistic designs on textile goods. Chinese silk production had been further refined in the imperial workshops which produced specific kinds of silk clothing to designate ranks within the bureaucratic hierarchy and for gifts to foreign countries and religious institutions. Meanwhile, private enterprises managed to develop sophisticated silk textiles not constrained by the government’s sumptuary rules. They were not allowed to make purple robes and robes with patterns reserved for certain bureaucratic ranks, such as the dragon and phoenix designs reserved for imperial regalia, but they could make textiles with Persian animal motifs, royal or not, and color schemes. Military expansion during the Sui and Tang dynasties accelerated growth in trade with the outside world, but sumptuary laws still sought to constrain the consumption of rare goods by limiting their use to the ruling elite. After centuries of invasions and population movements that rendered borders meaningless, the Tang reimposed border control at the frontiers and seaports and required foreign traders to be registered with the authorities in the cities where they conducted their business. Certain high-value goods such as jade 215

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

from Khotan and all the gifts from foreign countries had to be deposited in the Palace Treasury (zuocangshu 左藏署). In other words, rare goods obtained through tributary exchanges all were deemed the property of the state, which redistributed them according to the sumptuary rules. Meanwhile, there was a long list of goods ranging from silk textiles to iron that the traders were prohibited from carrying out of Tang territory. Private ownership of the best silk textiles, namely brocades (jin 錦) and damask (ling 綾), was forbidden, and thus transporting them over the border would incur harsh punishment. Smugglers would suffer years of penal servitude. Even legal exports had to be channeled through border inspections conducted by government officials. Buddhist pilgrims and missionaries, it seems, were immune from – or evaded – these restrictions, as some of them carried large quantities of silk textiles, incense, and other exotic goods. Despite such constraints, international trade prospered in tandem with the rising political and economic power of the Tang Empire.

Buddhism and the Silk Road Trade The warfare that erupted at the end of the second century and the following collapse of the Han dynasty and political division of China among competing states interrupted, but did not totally destroy, the tributary relationships between Chinese rulers and the Western Region. It was just at that time, however, that Buddhist missionaries began to make use of the travel infrastructure paved by the Han imperial power. At the time of the Eastern Han dynasty, the Yuezhi nomads who had settled and established a sedentary state in Bactria crossed the Hindu Kush mountains to occupy the northwestern portion of the Indian subcontinent, thus creating the Kushan Empire. Buddhism flourished under Kushan patronage. The Mahayana school of Buddhism sanctioned a commercial ethos much more than did the original teachings of Sakyamuni Buddha, who was active in the Ganges plain region during the sixth to fifth centuries B C E.8 From Kushan, Buddhist missionaries accompanied trading caravans to Central Asia and China, reaching the Eastern Han capital of Luoyang and gaining followers among the foreign merchant communities. Yet Buddhist theology and institutions took root in China only after the disruption of the Chinese patriarchal family structure following the collapse of Han imperial power, and especially after nomad 8

Xinru Liu, Ancient India and Ancient China: Trade and Religious Exchanges A D 1–600 (Delhi, Oxford University Press, 1994), pp. 88–102.

216

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

invasions forced the Jin court to flee to the Yangzi delta region in 317. This event set in motion large-scale migrations from north to south that changed the entire demographic landscape of China. Aristocratic families who fled to the south brought with them cultivators and artisans to work on their new estates. Many people who lost their lands and family ties had to regroup among similarly displaced strangers to form new social communities, often centered around Buddhist monasteries. Even during the post-Han Period of Disunion (third to sixth centuries), agriculture remained the economic basis of both foreign-ruled regimes in the north and a succession of Chinese dynasties in the south. Both the nomads from the steppe who now ruled the Central Plain and the Chinese aristocrats newly settled in the frontier regions of the south broadened their horizons thanks to the migrations and exchanges of information that ensued from the time of Zhang Qian. The ruling elites, old and new, were looking in all directions for curiosities and rare goods. The market for exotic goods from the Western Regions made adjustments to the new political and social realities; while experiencing ups and downs, this trade never died out. The horse–silk trade continued to be the major conduit of exchange between agricultural China and Central Asia, including the steppe and the oasis states of the Amu–Syr basin. Rulers of north China, Chinese and foreign alike, made efforts to maintain this trade by promoting silk textile production in the agrarian heartland and exercising control over the oasis states in the Western Region. The newly established Buddhist institutions provided new incentives to trade the luxuries of the Silk Road. Following the flight of the Jin court to the south in 317, Buddhism became the main force promoting trade along the Silk Road and maintaining China’s commercial connections to the Western Region and India. When Lü Guang, a Di warlord, conquered the major oasis state of Kucha on the northwestern edge of the Taklamakan Desert, he proclaimed himself the king of Liang (386–399). Under Lü’s patronage, Kumarajiva, the most famous Buddhist scholar in Kucha, undertook the enterprise of translating Buddhist texts for a Chinese audience. Kumarajiva was born c. 344 to a Kucha princess and an Indian aristocrat, probably a refugee from the defunct Kushan Empire. Kucha at that time was the main hub of Buddhist missionary and cultural activities under the patronage of its strong and stable royal family. The city of Kucha, surrounded by three layers of walls, boasted about a thousand Buddhist stupas and temples. Kucha’s numerous monasteries and nunneries were educational institutions for all the royal families of the oases east of the Pamir. The king’s palace itself, housing statues of the Buddha, looked just like 217

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

a large Buddhist monastery and was said to be as splendid as the abodes of the gods. The many caves of Buddhist monasteries outside the city are famous for beautiful murals, which became the models for Buddhist art along the Silk Road. Kumarajiva began his education in such a monastery in Kucha. His mother left her husband but took her seven-year-old son to the monastery. When Kumarajiva learned everything that he could there in two years, his mother took him to Kashmir. There he engaged in discussions and debates with scholars of various schools, including Brahmanical pundits. This experience probably gave him the opportunity to learn Sanskrit, the language of Brahmans and Mahayana Buddhists. By age twelve, Kumarajiva already had achieved a reputation for learning throughout the Tarim Basin oases. Then two princely Buddhist teachers appeared on the scene and instructed Kumarajiva in the tenets of Mahayana Buddhism. By age twenty, he was ordained as a monk in the royal monastery of Kucha. But the political turbulence in China soon reached Kucha. Sensing the impending troubles, his mother set out for India, leaving her educated son to fend for himself. As a Mahayana Buddhist teacher, the first and foremost important translator and interpreter of Buddhism for a Chinese audience, Kumarajiva was an essential figure in making Buddhist monasteries the facilitators of long-distance trade with China via the Western Region and in furthering the trade with foreign countries to the east and south. In 401 Kumarajiva arrived in Chang’an, at that time the capital of the Later Qin kingdom. Among the many Mahayana Buddhist texts that he translated, the Lotus Sutra and the Amitabha Sutra had the most impact in shaping Chinese religious culture, and also had direct impact on the trade of rare goods via the Western Region. The Lotus Sutra exalted the concept of the bodhisattva, an enlightened figure who stayed in the mundane world instead of crossing the threshold to the state of nirvana in order to assist in the salvation of less fortunate beings. Avalokiteshvara (Guanyin 觀音 in Chinese), the most renowned figure among the bodhisattvas, was believed to search throughout the world to render assistance to anyone who had faith in his salvific power and invoked the bodhisattva’s name in times of peril, including merchants sailing the oceans and caravans trekking across deserts. The Amitabha Buddha vowed to erase all of the accumulated sins of the devout faithful who invoked his name at the hour of their death and to usher them into the Pure Land paradise. The Mahayana tradition fostered the concept that making donations to Buddhist institutions could earn karmic merit that was as valuable as the virtue gained by good behavior. Therefore 218

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

wealth could be transformed into religious merit through donations to Buddhist institutions. As Chinese Buddhists sought to express the sincerity of their faith to Avalokiteshvara, Amitabha, and other Buddhist divinities, their devotion created a demand for the exotic goods obtained from the Western Region. The most commonly traded goods on the Silk Road included items frequently donated to monks and monasteries, such as silk textiles in the forms of apparel, banners, and writing materials, as well as the precious objects called the “Seven Treasures” by Buddhists, namely gold, silver, lapis lazuli, crystal, red coral, pearls, and a kind of striped stone that could be amber or agate.9 All along the Silk Road and in cities in China and South Asia, hundreds and thousands of silk banners were hung on stupas, the reliquaries made of the seven treasures holding relics of the Buddha that marked the sites of Buddhist pilgrimage. Meanwhile, the Chinese produced plenty of silk goods both as commodities to pay for exotic imports as well as donations to make the banners used in Buddhist religious ceremonies and to decorate stupa reliquaries and monasteries. Around the time Kumarajiva arrived at Chang’an, the Chinese Buddhist pilgrim Faxian left the capital for India in search of a complete version of vinaya (monastic regulation) texts. Once in India, Faxian traveled from the upper Indus river valley to the mouth of the Ganges, where he boarded a merchant ship embarking for Sri Lanka, another major center of Buddhist practice and study. After staying in Sri Lanka for two years, Faxian set out for home on a large vessel which could hold more than 200 passengers. This sea voyage proved perilous and was repeatedly delayed by storms. After departing Sumatra for the southern Chinese port of Guangzhou, storms again blew Faxian’s ship off course, causing the merchants to blame Faxian and his Buddhist books and images as the cause of these disasters. Only through the intercession of his patron did Faxian avoid the fate of being thrown overboard. After more than seventy days at sea, the ship finally landed at Laoshan in modern Shandong, far to the north of its original destination, in 413. Laoshan was in the territory of the Jin dynasty, and the local governor welcomed the unexpected merchant ship and Faxian, the courageous Buddhist pilgrim and teacher. But since he had landed in Jin territory, Faxian abandoned the idea of returning to Chang’an and instead traveled to Jiankang, where he devoted the remainder of his life to translating Buddhist texts into Chinese.

9

Liu, Ancient India and Ancient China, pp. 92–102.

219

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

Faxian had gone to India to obtain texts of monastic rules. Although he brought back many different Buddhist texts, his translation of the rules of the Mahasamghika (“Great Congregation”) order was a key catalyst in the development of organized Buddhist institutions in China. Faxian went to India by Central Asian land routes and came back to China by sea. Judging from his autobiography, the land routes from Chang’an to northwest India were relatively well traveled by caravans. The sea routes, however, were full of uncertainty. Sailors had knowledge of the monsoon winds but had few tools to help them ascertain directions and distances. Still, merchants braved the perils of ocean voyaging for the profits of long-distance trade, and more and more Buddhist pilgrims and missionaries joined them. Memoirs such as Faxian’s diary provided crucial information to travelers as well as ruling authorities, spreading knowledge of the routes of passage that could connect China to the outside world. The Buddhist concept of the “Seven Treasures” stimulated demand for rare goods from faraway lands such as Central Asia, India, and the Mediterranean. The commercial ethos in Buddhist theology shaped the aesthetic values of Buddhists in many countries, and consequently influenced the demand and supply of commodities. Buddhist monasteries in Luoyang dominated the Northern Wei economy and wielded considerable clout in court politics. Empress Dowager Hu ordered the building of the Yongning Monastery as a royal foundation outside the southern gate of the imperial palace in 516. In 518 she sent a pilgrimage mission led by the monk Huisheng to India bearing numerous donations from the empress dowager, including a thousand pieces of multicolored silk banners 100 chi (about thirty meters) in length and 500 incense bags of silk brocade. In addition, nobles and court officials also entrusted the pilgrims with 2,000 silk banners that were distributed to monasteries along their travel route. Huisheng noted that the stupas and monasteries in the regions through which they traveled (ending at Gandhara in northwestern Afghanistan) had already received many jewels and silks.10 The Northern Wei emissaries were just another group disseminating exquisite silk products throughout the well-trodden Central Asian trade routes. Buddhist monasteries in Luoyang, the capital of the Northern Wei, hosted Buddhist monks – as well as priests of other religions from the west – and also acted as promoters of trade. Some of the founders of monasteries were 10

Yang Hsüan-chih, A Record of Buddhist Monasteries in Lo-yang (Princeton, Princeton University Press, 1984), pp. 242–5.

220

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

Central Asian in origin, and therefore they could provide services for the traders from their homelands. But the principal host of foreigners of all types was the Yongming Monastery. This monastery, adjacent to the palace and near the capital’s commercial and financial center, housed more than 3,000 monks in its large compound of over a thousand rooms. Clerics from as far away as the Byzantine Empire as well as Southeast Asian lands and islands took up residence there. It seems that the monastery also housed visitors engaged in business other than religious evangelism. A market for foreign goods was located south of Luoyang’s city gate, at the confluence of the Luo and Yi rivers. Foreign traders, including those from the Mediterranean, also took up residence in the four hostels built for them. Those allowed permanent residency in Luoyang would receive a parcel of land to build a house in one of the four designated neighborhoods nearby. Visitors from the west, including those from Byzantium, were said to number over 10,000 households, and their neighborhood was the most affluent and well-maintained in the city.11 The circulation of information among Romans, Persians, Indians, and others corroborated the travel experience of Faxian, and is further verified by the numerous archaeological finds of Byzantine and Sassanid Persian coins along the Silk Road, and goods such as Mediterranean glassware buried in elite tombs in China.

China and the Mediterranean: Connections via the Hellenistic Legacy In the early stage of the Silk Road, coincident with the heyday of the Eastern Han and Roman empires, direct commercial ties between the two major political entities of Eurasia never developed, but the way was paved for connecting China with the Mediterranean world through the Greekspeaking Roman Orient and the Hellenistic legacy in Central Asia. Greekspeaking traders traveled through the Iranian plateau and Central Asia or embarked from Alexandria to reach the west coast of India via the Red and Arabian Seas. In the first century C E Han historians mention “Great Qin” (Da Qin 大秦) as the name of the Roman Empire, which was taken as a synonym for Alexandria (Lijian 犁鞬, 犁靬). During the post-Han period, Chinese naturally used Da Qin and Lijian as names for the Byzantine Empire.12 As rare 11 12

Yang, A Record of Buddhist Monasteries, pp. 13–17, 148–51. Xinru Liu, “Looking towards the West: How the Chinese Viewed the Romans,” in Berit Hildebrandt (ed.), Silk: Trade and Exchange along the Silk Roads between Rome and China in Antiquity (Oxford, Oxbow Books, 2017), pp. 2–3.

221

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

goods from Da Qin became legendary in China, the Da Qin and Lijian (or other variations of Alexandria) place names lingered in official geographies compiled during the Period of Disunion. Several hoards of lead disks with Bactrian Greek inscriptions found inside China indicate the possible presence of Greek-speaking merchants from Hellenistic lands of Central Asia. These hoards include a collection of thirteen lead disks found at the site of the Han capital city, Chang’an; two disks deposited in a Han-era site at Jiangyuan, Fufeng county (Shaanxi); and 274 disks neatly stacked and buried on the bank of the Pu river in Lingtai county (Gansu). There is no context for dating the large hoard, but all of the disks share the similar vague animal motif on the bottom, size (5.2–5.6 centimeters diameter), and weight (110–135 grams) (Figure 6.1).13 As lead is a soft material, the small variations of size and weight could be attributed to wear. The Greek letters form a circle along the rim, although they are hard to read due to wear. The lead disks were not valuable enough to be commodities themselves, and they were not treasured by the owners enough to be buried in their tombs. They were deposited for some reason at sites along the Silk Road. One possibility is that they were weights for trade used by Greekspeaking traders who traveled to China. Silk textiles reached the Roman market through Hellenistic agents in the western segments of the Silk Road. Pliny the Elder, writing in the first century C E, complained that the treasure of the empire was drained away by purchases of luxury imports, including the translucent silks adorning rich women. Erudite as Pliny was, though, he did not know exactly what silk was: Chinese, who are famous for the woolen substance obtained from the forests; after a soaking in water they comb off the white down of the leaves, so supply our women with the double task of unravelling the threads and weaving them together again; so many fold is the labour employed, and so distant is the region of the globe drawn upon, to enable the Roman matron to flaunt transparent raiment in public.14

Pliny evidently confused the mulberry leaves fed to silkworms with the cocoons they spun. Regardless, as an extension of the trade with the Western Region, the Mediterranean world had been linked with north China. The Mediterranean market for silks from China continued and even expanded after the end of the Han dynasty, even as a silk-weaving industry 13 14

Xu, Sichou zhilu kaogu, pp. 10, 22, 26. Pliny, Natural History (trans. H. Rackham) (Cambridge, MA, Harvard University Press, 1962), vol. 10, pp. 171–9.

222

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

Figure 6.1 Lead disks with Greek letters found in China. Excavated at Jiangyuan, Fufeng county (Shaanxi province) Source: Luo Xizhang 罗西章, “Fufeng Jiangyuan faxian Handai waiguo mingwen qianbing” 扶风姜塬发现汉代外国铭文铅饼, Kaogu 考古 (1976.4)

developed in the Byzantine Empire. Chinese writers observed that the Da Qin demand for raw silk material was so high that the weavers unraveled plain silk textiles to obtain silk yarn. The weavers then made silk textiles in styles that appealed to Mediterranean tastes. Analyses of silk samples found in Yingpan, an oasis on the eastern rim of the Taklamakan Desert, dated to the third century C E, reveal that locally woven multicolored silk textiles also were made of yarn unraveled from plain silk cloth.15 At Palmyra in Syria, near the western end of the Silk Road, many heavy tapestry silks woven in China have 15

F. Zhao, “Domestic, Wild or Unraveled? A Study on Tabby, Taquete and Jin with Spin Silk from Yingpan, Xinjiang, Third–Fourth Centuries,” in Hildebrandt, Silk, Trade and Exchange, pp. 95–103.

223

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

been discovered, as well as a piece of gauze made of silk and woolen thread, which could be products of Central Asia.16 The archaeological finds are fragmentary, but they still indicate that there were numerous and resourceful agents engaged in the trade between China and the Mediterranean. The Byzantine Empire made purple silk textiles the symbol of imperial power and ecclesiastical authority, using purple dyes of Mediterranean provenance to decorate silk yarn imported from China. Byzantine silk culture was associated with the propagation of Christianity, and thus its artistic styles were much different from those of China. It seems that the Chinese were aware that the Da Qin market coveted not exquisite weaves with Chinese designs but rather just raw silk yarn. Byzantine and Sassanid Persian buyers competed fiercely for silk materials from China. The developing silk industry in the Levant needed supplies of silk yarn from China, to which it had no direct access. Even plain silks had to be purchased from the Sassanids and others at “sea markets,” the ports on the Persian Gulf or the Red Sea.17 By the reign of Justinian (r. 527–565), the Byzantine Empire had begun to develop its own independent sericulture industry. Many items displaying Hellenistic artistic styles have been found in the tombs of the Northern Dynasties. Byzantine coins were mostly used for decoration instead of currency in China. In the tomb of Princess Ruru of the Eastern Wei in Ci county (Hebei) a gold coin of Anastasius (r. 491–518) and one of Justin I (r. 518–527) were interred upon the princess’s death in 550. One gold coin of Theodosius I I (r. 408–450) and two of Justin I were discovered in the tomb of another Eastern Wei aristocrat, Li Xizong, and his wife (d. 545 and 549 respectively). All three of these coins had clipped edges or drilled holes to allow their use as ornamentation.18 However, it appears that there were few direct commercial exchanges between the Byzantine Empire and the regimes of north China. Artifacts bearing Hellenistic motifs and styles on vessels of Sassanid workmanship were made in Central Asian countries where Greek languages or writing were still in use and Hellenistic art styles remained fashionable. These exquisite artifacts found their way to elite tombs in north China. A gilt ewer retrieved from the tomb of Li Xian (d. 569), a general of the Northern Zhou (557–581), and his wife in Guyuan county (Ningxia) exemplifies this kind of trade. The precise theme of the six figures on the ewer, displaying three pairs of lovers, remains open to debate, but no doubt they illustrate a story drawn from Greek mythology. The gilded silver 16

17

Andreas Schmidt-Colinet, Annemarie Stauffer, and Khaled al-As‘ad, Die Textilien aus Palmyra: Neue und alte Funde (Mainz, Verlag Philipp von Zabern, 2000), p. 183, cat. 490, Plate 74c. SGZ, 30, p. 861. 18 Xu, Sichou zhilu kaogu, pp. 56–7.

224

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

workmanship is typical of Sassanid Persia. The location of the ewer’s site of production also is a subject of contention, but it surely originated in the domain of Hellenistic Central Asia.19 Before adopting the Chinese surname Li, the ancestors of Li Xian were Tuoba, the royal Xianbei lineage of the Northern Dynasties. The ruling elite who carried these remarkable artworks into their tombs certainly appreciated their beauty, but there is little evidence that Chinese artisans reproduced Hellenistic vessels for the domestic market.

Sogdian Caravans and Diasporas The Sogdians, inhabitants of the lands between the Amu and Syr rivers, were the most renowned traders of the Silk Road. The communities in Sogdia never coalesced into a single state, but their oasis city-states shared a common Iranian language, a writing system, a brand of Zoroastrian religion mingled with local cults, and a distinguished commercial culture. The expansion of Achaemenid Persia to the east incorporated Sogdia into the empire by the end of the sixth century B C E. After the region was annexed by the Greeks in the fourth century B C E and the Kushans in the second or first century B C E, Sogdia disappeared from historical records. Probably during this period, the development of agriculture in the oases fostered the formation of city-states ruled by independent clans. Chinese historians eventually named Sogdia the land of the “nine lineages of Zhaowu” (Zhaowu Jiuxing 昭武九姓) after its ruling clans. Sogdians who came to China would bear the surname of their city of origin. For example, a man from Samarkand would use Kang as his surname, or a woman from Bukhara would have the surname An. The number nine was symbolic, as the number of Sogdian city-states changed over time. Samarkand – the principal Sogdian city – is mentioned in a cache of letters written by Sogdian merchants that were found at a watchtower west of Dunhuang by the British explorer Aurel Stein. According to the events described in the Sogdian letters – which never reached their destination in Samarkand – they were composed around 313 C E. These letters show that a postal network carrying private and commercial letters across the Silk Road from various locations in China and the Western Region as far as Samarkand had already formed despite the incessant warfare of the time. One letter, addressed to a certain Varzak at Samarkand by Nanai-vandak, the manager of 19

Zhuo Wu, “Notes on the Silver Ewer from the Tomb of Li Xian,” Bulletin of the Asia Institute 2 (1989), 61–70.

225

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

a trading firm at Gansu, provides firsthand testimony of the sack of Luoyang at the hands of steppe invaders in 311. This letter reported that Luoyang, Ye, and other Chinese cities lay in ruins, thus contact with some of the agents sent there was lost. But the business seemed still ongoing. The letters show that Sogdian traders had become fully ensconced in China as early as the fourth century C E.20 The Sogdian communities that spread across the vast Central Asian tracts of oases, mountains, and steppes not only participated in long-distance trade, but also reworked trade commodities to fit the demands of both local and distant markets. For example, the Zandaniji silk textiles that appeared around the seventh century have been identified as a Sogdian invention. In 1959 the ink letters written on the back of a bale of multicolored silk were deciphered as reading “Zandaniji,” i.e., the product of a town named Zandan near the city of Bukhara in Uzbekistan.21 Zandaniji silk textiles shared several features that distinguished them from Chinese silk and “Persian silk” in color and design. A pair of facing rams, deer, or ducks are the prevalent motifs. The stiff postures of the animals resemble the Persian artistic style but differ from the flowing curves of animals in motion depicted on Chinese silk textiles. However, the colors on Zandaniji silk are warm, in yellow, orange, or green, in contrast to the dark blue hues of the Persian style, and therefore mostly likely were dyed with pigments from China. The textiles were woven with warp-faced patterning, the traditional weaving technique of China.22 By 1981, 102 pieces of Zandaniji silk had been identified, almost all of them found in the treasuries of churches in Europe, with only two pieces found at the oasis town of Dunhuang.23 Since then, as Chinese archaeologists have published more findings, many textiles bearing Zandaniji features have been located in Xinjiang, although they had not been catalogued as such. The Zandaniji silks appear to have been targeted at West Asian markets, where the Zandaniji animal motifs fit the repertoire of Zoroastrian mythology. The weavers in Zandan who produced these special silk textiles combining Persian art styles and Chinese materials sold in far-flung markets both east and west created a famous brand that persisted after the Islamic conquest of Sogdia. 20 21

22

23

Étienne de la Vaissière, Sogdian Traders: A History (Leiden, Brill, 2005), pp. 45–9. D.C. Shepherd and W.B. Henning, “Zandaniji Identified?”, in Richard Ettinghausen (ed.), Aus der Welt der islamischen Kunst: Festschrift für Ernst Kühnel zum 75. Geburtstag am 26. 10. 1957 (Berlin, Verlag Gebrüder Mann, 1959), pp. 15–40. Michael W. Meister, “The Pearl Roundel in Chinese Textile Design,” Ars Orientalis 8 (1970), 255–67. Dorothy C. Shepherd, “Zandanı¯jı¯ Revisited,” in Mechthild Fleury-Lemberg and Karen Stolleis (eds.), Documenta textilia: Festschrift für Sigrid Müller-Christensen (Munich, Deutscher Kunstverlag, 1981), p. 118.

226

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

The dispersion of Sogdian caravaners and traders from their oasis homeland took the form of a trade diaspora, far-flung networks of merchants united by common origin, language, and their Zoroastrian religion. By the sixth century, communities of Sogdian traders had sprung up in various Chinese cities. Several tombs of Sogdian traders found near Xi’an dated to the sixth to early seventh centuries provide much visual evidence of the cultural activities and social organization of the Sogdian communities in China. The burial methods and artwork in the tombs show that the Sogdians remained faithful to their Zoroastrian religion even after they had settled in China for two or three generations. The Sogdians conducted distinctive mortuary practices: all the coffins contain bones which had been cleansed by vultures or dogs, and they were raised above the tomb floor by foot-stands. An Jia, a second-generation immigrant from Bukhara who died in 579, at the time of the Northern Zhou dynasty, was the sabao, meaning the head of a caravan or chief of the Sogdian community, in Tongzhou (Shanxi). The Northern Zhou granted him the title “great governor” (da dudu 大都督), probably in recognition of his leadership of the Sogdian community at Tongzhou. The obituary on An’s tomb stele is brief, but the artwork in the tomb tells us more about the sabao’s life. Above the gate of the tomb entrance is a scene of fire worship, the pre-eminent ritual of the Zoroastrian religion. The fire altar, carried by three camels, occupies the center of the mural, flanked by two priests – depicted with human faces on bird bodies – tending the altar.24 This image of Zoroastrian priests so far has been seen only in tombs of Sogdians in China. Neither the motif nor the artistic style is Sassanid, although in a way the concept of the motif is similar to the simurgh that prevailed in “Persian-style” art. However, the two are substantially different. The simurgh has a bird body with the head of a mammal, such as a dog or a camel. The human-faced birds in the Sogdian tombs were priests performing rituals for the dead. This bird-bodied priest motif could be unique to the religious culture of the Sogdian diaspora in China. Artworks on the stone enclosure of the coffin display scenes of An Jia as the chief of a caravan engaging in trade with Turks. A stone plate displays two scenes of An making treaties with a Turkish chief (Figure 6.2). On the upper half of the plate An rides a horse facing the Turkish chief, also astride a horse. An and the Turkish chief reach out to each other with one arm, a scene that mimics the motif of a Sassanid king receiving investiture from a god. The 24

Rong Xinjiang 荣新江 and Zhang Zhiqing 张志清 (eds.), Cong Samaerhan dao Changan: Suteren zai Zhongguo de wenhua yiji 从撒马尔罕到长安：粟特人在中国的文化遗迹 (Beijing, Beijing tushuguan chubanshe, 2004), pp. 66–9.

227

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

Figure 6.2 An Jia making a treaty with the Turks (stone coffin in collection of Shaanxi Provincial Museum) Source: Rong Xinjiang 荣新江 and Zhang Zhiqing 张志清 (eds.), Cong Samaerhan dao Changan: Suteren zai Zhongguo de wenhua yiji 从撒马尔罕到长安：粟特人在中国的文 化遗迹 (Beijing, Beijing tushuguan chubanshe, 2004), Plate 7, p. 71

228

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

scene on the lower part of the plate depicts treaty negotiations in a more Central Asian style. The two chiefs sit on a rug facing each other, while a man holding a jug, probably wine to celebrate the conclusion of the treaty, stands between them. Another allusion to Sassanid royalty is the depiction of a Sogdian chief on horseback shooting a lion. Although the imagery is majestic and dramatic, it is unlikely that a Sogdian sabao would go on a lion hunt. The Sogdians did like to show themselves as royalty, however. In their homelands, it was common for affluent merchants to decorate the walls of their houses with royal scenes. In China, though, such extravagance had to be hidden from public view, and is found only in their final resting places. Other illustrations probably preserve more realistic images of the commercial and diplomatic activities of the sabao as caravan leaders. In one, the sabao entertains Turks and Persians under a pitched tent with wine and performances of music and whirling dancing, a specialty of the Sogdians.25 Sogdian traders occupied a key position in the geopolitics of Central Asia in the sixth and seventh centuries, often acting as intermediaries in interactions among the north China regimes, the Turks on the steppe, and Sassanid Persia. The rapid ascendancy of the Turkish tribes as the dominant force in the Central Asian steppe in the sixth century posed a new threat to the sedentary foreign rulers of north China, and later on to the Sui and Tang dynasties. Like the Han emperors centuries earlier, the rulers of China needed horses to fight the horse-riding nomads, and thus they adopted the similar strategy of bribing the Turks with silks and other tribute to forestall attacks on border settlements. Meanwhile, the Turks, who themselves had little experience with commerce, employed Sogdian merchants to act as their agents in selling silks to the Persians. The Byzantine historian Menander Protector recorded that during the reign of the Sassanid king Khosro I (r. 531– 579), a Sogdian envoy named Maniakh arrived at the Sassanid court seeking permission to sell silks in Persia. Prompted by an Ephthalite (another steppe people, enemies of the Turks) adviser, Khosro paid for the silks brought by the Sogdians but burned them in front of the delegation to humiliate them. Maniakh then urged the Turks to cultivate the friendship of the Romans (i.e., the Byzantines), who were eager to purchase silk goods.26 Maniakh undoubtedly was acting in his capacity as caravan leader, much like his contemporary An Jia in China. During the heyday of Tang hegemony in East Asia in the seventh and eighth centuries, Sogdian city-states and traders held a key 25 26

Rong and Zhang, Cong Samaerhan dao Chang’an, pp. 70–5. De la Vaissière, Sogdian Traders, pp. 209, 228.

229

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

position in Central Asian politics among the Tang, the Turkish empires, and the Tibetan empires. Guyuan in Yuanzhou (Gansu) was a major waystation of the Sogdian caravan and diaspora community. The tomb of the Northern Zhou general Li Xian, who was buried at Guyuan, is renowned for its Hellenistic gilt silver vase, and also contained a Sassanid green glass bowl and a gold ring with lapis lazuli inlay, both treasures of the Silk Road. Li might have played the double roles of governor under the Northern Zhou and head of the local Sogdian caravanserai. In addition to acting as agents trading in silk and other goods on behalf of the Turks, the Sogdians obtained horses for various regimes in north China. A cemetery near Guyuan contains the graves of several Sogdians who had been in charge of horse management. Byzantine gold coins and Sassanid silver coins as well as glassware were buried with members of Shi family, a lineage from Sogdia.27 Shi Hedan (d. 670) was a second-generation Sogdian immigrant from Kesh in Sogdia. His father, Shi Shewu (d. 610), had served as a sabao under the Northern Zhou and Sui dynasties. Sogdian traders such as the Shi family served the ruling houses in China through successive regime changes. Shi Shewu had joined military campaigns led by the Northern Zhou emperor Yuwen Hu, and subsequently served under the Sui dynasty. Meanwhile, he made a fortune in his Yuanzhou homeland from caravan trade. His eldest son Shi Hedan had a distinguished official career and made a fortune for his family under the Tang dynasty.28 As Yuanzhou was the major pastoral region lying amid the Taklamakan, the Mongolia steppe, and agricultural north China, the Shi family’s prosperity derived from their service in supplying and managing horses for regimes in north China. Shi Hedan achieved prominence as the Director of Military Horses for the Tang dynasty, with responsibility for breeding high-quality horses for the Tang military.29 Shi also was employed in the crucial role of interpreter on behalf of the Tang government in negotiations for horse purchases. Several members of the next generations of the Shi family also assumed responsibility for managing horses for the Tang Empire. Sogdians serving as grooms for horses or camels were such a common sight that tri-colored ceramic figures of a Sogdian groom leading a horse or a camel were commonly included in 27 28 29

Xu, Sichou zhilu kaogu, pp. 29, 114. Rong and Zhang, Cong Samaerhan dao Chang’an, pp. 90–9. Jinxiu Li, “Shi Hedan and the Equine Administration in the Early Tang Dynasty,” in G. Bilavschi and D. Aparaschive (eds.), Studia Mediaevalia Europaea et Orientalia, Miscellanea in Honorem Professoris Emeriti Victor Spinei Oblata (Bucharest, Publishing House of the Romanian Academy, 2018), pp. 127–53.

230

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

elite Tang burials, suggesting that they had performed this service during the lifetime of the tomb occupant. Sogdian communities in China also exerted a cultural impact on Chinese society. Although the Sogdian trade activities and caravans were organized around Zoroastrian temples, not every trader remained a Zoroastrian follower. Some traders shifted their religious devotions, and even joined other religious communities, out of commercial convenience or even sincere faith. When the Sogdians first appeared in China, Sogdia was under the rule of the Kushan, who endorsed Buddhism above other religions. The core of Buddhist missionaries came from the heartland of the Kushan Empire, namely the Gandhara region, but missionaries of different ethnic origins all traveled with caravans on the Silk Roads, by land or by sea, to China. Sogdian merchants followed in their wake. Like Sogdian merchants, preachers and translators of Buddhist texts during the late second and the third centuries often assumed surnames indicating their original homelands: Zhu designated Indians; An was for Arsaces, the ruling family of Parthian Persia, or of Bukhara in Sogdia; Zhi was for Yuezhi and later on the Kushan; Kang was for Kangju, or Samarkand. For instance, during the period from 178 to 189, Zhi Loujiaqian, Zhu Shuofo, and An Xuan were active in translating Buddhist texts at the Eastern Han capital, Luoyang. An Xuan was a merchant who followed the Buddhist faith. Another group of notable Buddhist missionaries in Luoyang during the last decades of the Eastern Han included Zhi Qian, Kang Ju, and Kang Mengxiang. The ancestors of Kang Senghui (d. 280) were merchants from Samarkand who settled in India. Kang Senghui’s father ventured on to trade at Jiaozhi (Vietnam), where Kang was born. After the death of his parents Kang became a Buddhist monk, and he achieved fame as a Buddhist preacher at the court of the Wu kingdom. Buddhist legend records that by miraculously producing a relic of the Buddha Kang convinced the Wu king Sun Quan of the power of the Buddha.30 As Sogdian Zoroastrians were famous for magic performances, Kang Senghui perhaps acquired this skill from his compatriots to promote Buddhism. Some Sogdians became faithful followers of Buddhism both in Sogdia and in the diasporic communities in China. However, as a cultural and ethnic community, Sogdian traders mainly organized their social and commercial networks around Zoroastrian temples. The Sogdian diasporic communities of the Tarim Basin enjoyed autonomy under the suzerainty of Chinese authorities. The Northern Wei government granted the title of sabao to the leaders of the Sogdian communities and incorporated the office of sabao into 30

GSZ, pp. 14–16.

231

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

its bureaucratic system.31 Although the sabao, as in the case of the Shi family, were leaders of their communities even without formal designation by the government, they became part of the ruling elite of China, regardless of which regime was in power, which in turn greatly enhanced their commercial interests and wealth. In Chinese cities, Sogdian communities maintained Zoroastrian temples where magi performed rituals. The sabao exercised leadership in both the religious and administrative affairs of Sogdian communities and handled relations with the Chinese authorities. The religious services at Zoroastrian temples attracted hordes of spectators; music played with pipa (a kind of lute), drum, and flute accompanied drunken singers and dancers. The highlight was often a magic performance, such as a man cutting his own belly without wounding himself. The music and dances were entertaining, but hardly conveyed any religious message to the Chinese audience. Artists, musicians, and dancers from Central Asia (Sogdians and others) who migrated to China also performed for Chinese audiences at the court and in urban spaces. These spectacles were commemorated in poetry and the famous tri-colored ceramic figurines of the Tang era. Although Sogdian music, dance, and artworks contributed to the cosmopolitan culture of the Sui and Tang dynasties, their Zoroastrian activities made little impact on the religious life of the Chinese population. The most significant influence of the Sogdians was conveyed through the exotic commodities introduced by their trading caravans, including glassware, gilt silver vessels, ceramics, and textiles. In the Tang period, silk textiles and materials remained China’s main exports. But these silk products differed significantly from those of the Han era. During the intervening centuries, Chinese weavers developed complex looms to make both warp-faced and weft-faced multicolored silk compound textiles.32 Central Asian traders and immigrants also introduced many new elements into Chinese textile design. As mentioned above, weavers near Bukhara created the Zandaniji silk textiles that combined both Persian and Chinese styles; for example, their animal motifs were similar to the Persian style, but the pearl roundel surrounding the animal originated in China.33 By the eighth century, the pearl roundel 31

32

33

Xu Xuya 许序雅, Tangchao yu Zhongya jiuxinghu guanxi shi yanjiu 唐朝与中亚九姓胡 关系史研究 (Lanzhou, Lanzhou daxue chubanshe, 2012), pp. 151–5. A. Sheng, “Chinese Silks That Circulated among the Peoples North and West: Implications for Technological Exchange in Early Times?”, in Hildebrandt, Silk, Trade and Exchange, pp. 104–23. Meister, “Pearl Roundel,” pp. 255–67.

232

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

patterns on some silk textiles had become more Persian in style, with somber combinations of blue and gray, while the animal motifs often derived from the Zoroastrian artistic repertoire, such as the simurgh creature with the bird body and mammal head. This style most likely developed in Sogdia rather than in Sassanid Persia. The Sogdian traders and immigrants introduced the “Persian style” into Chinese textiles and vessels. He Chou, whose family originally came from Kushaniya in Sogdia, served as director of the royal workshops at the Sui court. He Chou supervised the manufacture of silk robes in the Persian style that were said to be even more beautiful than those presented as gifts by Persian envoys. He Chou’s workshops also imitated Persian green glassware vessels by using Chinese ceramic manufacturing methods. A ceramic bowl imitating Sassanid glassware, similar in style to the green glassware from Li Xian’s tomb, was found at a Northern Wei site in Luoyang.34 This bowl was made before the time of He Chou, thus he was not necessarily the inventor, but a clever craftsman who learned from others. After the demise of the Sui, He Chou continued to serve the Tang court as a director of royal workshops. Several silk textiles from archaeological sites on China’s Central Asian frontier bear the motif of two flying horses facing each other, with or without riders. These motifs certainly represented Persian royal poses, but the weaving technology of these fabrics varies. Such “Persian silks” might be genuine products from West Asia, but they also could be specimens of the “Persian silks” that He Chou introduced to China.35 Persian-style goods also were in high demand among consumers outside the court. He Chou’s uncle, He Tuo, made a fortune by weaving silk brocades with gold thread in the Sichuan region in Sui times. From the seventh century, Yangzhou in south China became a center of making foreign-style textiles. Chinese writings from this time frequently referred to Persian brocades, tapestries, and rugs as well as Persian traders. The meaning of “Persian” could have changed over time, and even in the same time period it could be applied to the products of Sogdia, or even those made in China, that embodied Persian technology and artistic styles. Fashionable Tang people sometimes liked to wear foreign-style clothing, and the Persian style permeated Tang urban culture. Not only was foreignstyle clothing beautiful, it was also not forbidden by the sumptuary laws that served to mark status differences between rulers and the ruled and to rank the 34 35

Xu, Sichou zhilu kaogu, p. 104. Zhao Feng 赵丰, “Tangxi yima wenjin yu He Chou fangzhi Bosi jin” 唐系翼马文锦与 何稠仿制波斯锦, Wenwu 文物 2010.3, 71–83.

233

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

bureaucracy, since foreign-style goods were deemed irrelevant to the political order. To be sure, Tang rulers were annoyed by the great wealth amassed by foreign traders, mostly Sogdians, who emerged as the nouveaux riches in Chinese cities.36 Nevertheless, the immigrants, Sogdians and others, gradually lost their original faith and religious networks by the end of the Tang. No longer were individuals bearing Sogdian surnames buried in the Zoroastrian style as they became assimilated into Chinese society. On the other hand, Chinese urbanites continued to absorb foreign styles into their own clothing fashions.

The Disintegration of the Governmental Monopoly on Foreign Trade As early as the second millennium B C E Chinese rulers had used jade imported from Central Asia as emblems of royal status. Many jade artifacts from the tomb of Lady Hao, consort of King Wuding (r. c. 1250–1192 B C E) of the Shang dynasty, came from Khotan, on the southern rim of the Taklamakan Desert. Chinese governments continually sought to monopolize prestige goods, especially those imported from outside their domains, to mark the status of the ruling elite, but exclusive control proved elusive. From the mid-Tang period, as the magnitude of foreign trade expanded and goods entered China through diverse routes, state control of foreign trade as well as the sumptuary regulations gradually collapsed. The rising scale of imports might have made such restrictions unnecessary. Since the Han dynasty, importing horses for pulling imperial and official carriages and for cavalry mounts had been the major purpose of foreign trade in the eyes of Chinese rulers. During Tang times, however, the court was overwhelmed by an oversupply of horses delivered by the Uyghurs, a troublesome steppe-based ally. Tensions over horse–silk transactions at least partly caused the breakdown of the alliance between the Tang and the Uyghurs, which eventually led to the historic expulsion of foreign religious institutions from Tang territory during the era of Emperor Wuzong (r. 841– 846). The Uyghurs had been crucial allies of the Tang since the mid-eighth century, when they aided the Tang in suppressing the An Lushan Rebellion (755–763). Subsequently, the Tang government treated the Uyghurs in the same way as other steppe allies, including dispatching princesses as brides for Uyghur chiefs and exchanging envoys to facilitate horse–silk transactions. 36

Xu, Tangchao yu jiuxinghu, pp. 127–8.

234

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

However, Tang officials came to regard the horse–silk exchange as disadvantageous to the Tang: Since the Qianyuan era (758–762), the Uyghurs, proud of their contributions to the imperial court, frequently sent envoys with large numbers of horses to exchange for silks, as often as every year. They demanded forty bolts of silk as the price for a single horse, and they often brought tens of thousands of horses. The Uyghur envoys – often more than one – resided at the hostel of the Foreign Affairs Office (honglusi 鴻臚司) for several months awaiting payment. Their appetite for silk was never satiated, while the horses we received were worthless, so the court became frustrated with this exchange. At this time (771), the emperor [Daizong] lavished them with munificent gifts to display his generous patronage, and to embarrass them as well. In the same month, Uyghur envoys brought ten thousand horses to sell to the Tang. Considering that the horses were paid for from taxes submitted by the people, who should not be burdened by excessive levies, Emperor Daizong ordered his officials to restrict purchases within the means of the treasury, and then granted permission to buy no more than six thousand horses.37

As horses were privileged modes of conveyance restricted to court and official use and to certain governmental functions such as postal transmissions, a surfeit of horses in the imperial stables could be a heavy financial burden, especially when the Uyghur horses were not all of high quality. In the following decades the Uyghurs repeatedly pressed the Tang to purchase more horses than were needed. The Tang court accommodated these demands because the Uyghurs were crucial allies in fending off aggression by the Tibetans and the Turks. Uyghur envoys residing in the capital often rode horses, ignoring any disciplinary action by the police. In this atmosphere, sumptuary rules banning traders and other unqualified personnel from riding horses could not be enforced. It was not that the Tang court did not need high-quality military horses. But the Uyghurs blocked alternative sources of good horses from the steppe in order to control the horse–silk trade with the Tang and demand higher prices. The Kirghiz peoples inhabiting regions north of the Uyghur territory traded precious furs such as sable with the Arabs and the Tibetans. The Kirghiz also possessed excellent horses, if not in large numbers, but had no access to the Tang market. In 843, after defeating the Uyghur khanate, the Kirghiz sent envoys to the Tang court seeking a share of the horse–silk tributary trade. The court officials, after a debate, decided that since the Kirghiz, unlike the Uyghurs, did not have a tributary relationship with the 37

JTS, 145, p. 5207.

235

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

Tang, they should just sell their horses on the private market.38 Thus, after the Kirghiz replaced the Uyghurs as the major supplier of good horses, the Tang obtained horses through market purchases. Buddhist pilgrims had long carried silk textiles both to make donations and to use as means of payment for their food and board at monasteries along the Central Asian trade routes. The Tang emperors often bestowed silk robes on Buddhist priests and institutions as gestures of patronage. The Tang court had established a strict sumptuary code for government officials that restricted the privilege of wearing purple robes to the highest rank of officials. This same hierarchy of dress was extended to religious institutions, with the purple robe reserved as the highest honor for priests who served the state. Prestigious figures such as the Buddhist pilgrim Xuanzang and the Nestorian priest Isaac, among others, received purple or tapestry silk robes for conducting their ritual performances.39 Celibate monks lacked heirs, but they often sold or gave away their possessions to members of their social networks, and in this way their silk clothing might circulate beyond the restricted range of designated recipients. As a whole, the development of the silk industry produced more than enough luxury textiles, including those in “Persian” styles, for the needs of a restricted group of consumers, and many of these goods entered market circulation. Some were sold through religious institutions. When Xuanzang finished translating many of the Buddhist texts he had brought from India, he wanted to have the original texts stored in a secure place. In 652 he appealed to Emperor Gaozong (r. 649–683) for assistance to build a stupa at his home monastery. Gaozong bestowed fine garments of deceased palace women on Xuanzang for the project, and the income from selling the clothing enabled Xuanzang to construct a grand monument, the Dayan Pagoda. Devotees who came to the monastery to purchase these garments also gained religious merit, as the money would pay for preserving sacred texts. Since the garments had been purified by Buddhist merit-making, they were exempt from sumptuary rules. The extension of Islamic commercial networks into the Silk Road world introduced new products and broke down the restrictions on the circulation of commodities. The Islamic religion supported commerce, and Muhammad himself was a trader. As the Islamic empires expanded over large portions of Afro-Eurasia, a system of commercial brands, the tiraz, was established to mark textiles produced in various parts of the Islamic world. Tiraz refers to 38

39

Li Jinxiu 李锦绣, “Tang yu Xiajiasi de juanma maoyi” 唐与黠戛斯的绢马贸易, unpublished MS. Liu, Silk Roads, pp. 24, 136–9.

236

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

embroidered or tapestried bands inscribed with the name of the caliph or local governor attached to the hem of the garment. The tiraz inscriptions, composed in various regional calligraphic styles, provided information such as the place and date of manufacture to verify the authenticity of the products. After the Abbasid caliphate had begun to disintegrate in the ninth century, sultans of regional states included their own tiraz messages on locally made textiles. In other words, tiraz was an institution, parallel with coinage, for Islamic rulers to assert their authority.40 The tiraz system expanded the sales of textiles far outside the Islamic territory. The Umayyad caliphate occupied most of the Byzantine lands where sericulture was practiced, but it also intended to extend its conquests in Central Asia to reach the original source of silks in China.41 The Islamic expansion into Central Asia took several centuries, however, and went through a series of stages. The Umayyads annexed Sogdia during the early part of the eighth century, prompting many Sogdians to migrate to the east, including China. Meanwhile, Sogdian cities maneuvered between the Arabs and the Turks to maintain some degree of political and economic autonomy. During the heyday of Tang prosperity, many Central Asian rulers sent envoys to the Tang court to enlist Chinese support against their enemies. The Tang encyclopedia Cefu yuangui recorded long lists of tribute brought by foreign envoys, including at least fifteen embassies from the Umayyad court, who brought gifts including purebred horses, Persian goods such as gold-threaded silk brocade, and camphor and other products from Southeast Asia. In 747, just before the demise of their caliphate, the Umayyads sent six leopards, which might have been the cheetahs treasured by nomads as hunting animals, to the Tang court.42 Three Umayyad gold coins, dated 702, 718– 719, and 746–747, were found in a Tang tomb at Xi’an.43 The Umayyad embassies probably were diplomatic gestures to persuade the Tang to withhold support for Central Asians resisting Islamic expansion, but they also afforded opportunities to survey land and sea routes for trade with China. Through such exchanges, the Tang court gained knowledge about the Arabs and the Islamic empire. The section on “Arabia” (Dashi 大食) in the official

40

41

42

43

Xinru Liu, Silk and Religion: An Exploration of Material Life and the Thought of People, A D 600–1200 (Delhi, Oxford University Press, 1996), pp. 130–57. Ya’qubi, Historiae, ii.346, quoted in W. Bartold, Turkestan Down to the Mongol Invasion, 4th edn (Cambridge, Cambridge University Press, 2007), p. 185. Cefu yuangui, 册府元龟, cited in Zhang Xinlang 张星烺, Zhongxi jiaotong shiliao huibian 中西交通史料汇编 (Beijing, Zhonghua Shuju, 2003), 2, pp. 708–10. Xu, Sichou zhilu kaogu, pp. 14–15.

237

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu

history of the Tang dynasty provides a reliable account of the rise of Islam and correctly locates Arabia to the west of Iran.44 The battle between Tang and Arab armies in 751 at the Talas river, near the borders of modern Kyrgyzstan and Kazakhstan, was the only direct conflict between the Islamic and Tang empires. Although victorious, the Arabs halted their advance eastward because of the Abbasid overthrow of the Umayyad caliphate at that time. But the Arabs returned with 20,000 Chinese and Central Asian prisoners, many of them skilled artisans, who were put to work constructing the Abbasid caliphate’s new capital, Baghdad. One of the prisoners, Du Huan, eventually returned to China and supplied details about the contributions of Chinese and Central Asian workers to the new Islamic regime. Fragments of his story were recorded by his cousin, the renowned historian Du You, in his encyclopedia of government institutions. Subsequently Tang records make reference to envoys from the “BlackRobed Arabs,” i.e., the Abbasid caliphate. As Persians gradually converted to Islam, the new Muslims took advantage of the travel infrastructure of the Silk Road to journey to China. Since these new Muslims brought the same Persian products, the differences between Arabs and Persians blurred in Chinese eyes. Central Asians, including the populations of Bactria and Sogdia, eventually converted to Islam. The Barmakids, the patron family of the Buddhist community of Balkh in Bactria, joined the Islamic cause by serving the Abbasid caliphate as wazirs (chief ministers), while the family members remaining in Balkh rebuilt the great Buddhist monastery Nawbahar, or New Monastery, in the middle of the eighth century, and thus sustained Buddhist activities there for more than a hundred years after the Muslim conquest. Al-Bukhari, a Sogdian Muslim, became one of the most respected authorities on hadith (the sayings of the Prophet Muhammad). The mathematician Al-Khwarismi (c. 780–850), who translated Indian digits to Arabic numbers and developed algebra as an independent branch of mathematics, left his homeland in the Transoxiana region to join the Baghdad community of scholars.45 In other words, Central Asians took a leading role in the intellectual life and commerce of the Islamic world. Sogdian converts to Islam regrouped their commercial networks around mosques and hostels for Muslims. Thanks to the far-ranging Sufi orders, Islamic religion infiltrated the oases of the Tarim Basin. Sufi devotional practices were centered around 44 45

JTS, 148, pp. 5315–16. Xinru Liu, “A Silk Road Legacy: The Spread of Buddhism and Islam,” Journal of World History 22.1 (2011), 55–81.

238

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

Silk Road Trade and Foreign Economic Influences

congregational assemblies conducted by revered spiritual leaders at meeting halls (khanqah) that often served as caravanserais and hospices. The Sufis invented the sama ceremony, where intoxicated Sufis performed whirling dances to the accompaniment of music for hours and even days, reminiscent of the whirling dance of the Sogdians. The Sogdians, now Muslims, continued to act as intermediaries between China, Central Asia, and West Asia, now under the Abbasid caliphate. As Sufis were not that dogmatic in following the aniconic rules of orthodox Islam, the “Persian style,” with all its animal motifs, continued to dominate the material culture of the Silk Road. By the mid-ninth century, when the Tang dynasty was in its twilight, Muslim sailors were regularly voyaging to the southeast coast of China. The weakened Tang government could no longer monopolize luxury goods, whether domestically made or imported, for use in marking bureaucratic ranks or granting imperial honors. The character of Afro-Eurasian trade also changed. China no longer just exported silk, but also porcelain and other goods. The sea routes connecting China to the Indian Ocean and the Islamic world carried much larger cargoes of spices, fragrances, porcelain, and textiles than the caravans on the Central Asian land routes. After the fall of the Tang dynasty in 907, the sumptuary rules that Chinese rulers had tried to impose on their subjects became obsolete. China’s foreign trade increasingly took the form of private commercial enterprise independent of government control.

Further Reading Barfield, Thomas J., The Perilous Frontier: Nomadic Empires and China, 221 B C to A D 1757 (Cambridge, MA, Blackwell, 1989). Boulnois, Luce, Silk Road: Monks, Warriors, and Merchants (Hong Kong, Airphoto International Ltd, 2004). Chen, BuYun, Empire of Style: Silk and Fashion in Tang China (Seattle, University of Washington Press, 2019). Gernet, Jacques, Buddhism in Chinese Society (New York, Columbia University Press, 1995). Hansen, Valerie, Negotiating Daily Life in Traditional China (New Haven, Yale University Press, 1995). Hansen, Valerie, Silk Road: A New History (New York, Oxford University Press, 2012). Hildebrandt, Berit (ed.), Silk: Trade and Exchange along the Silk Roads between Rome and China in Antiquity (Oxford, Oxbow Books, 2017). Lin Wushu 林悟殊, Zhonggu yijiao huahua congkao 中古夷教华化丛考 (Lanzhou, Lanzhou daxue chubanshe, 2011). Liu, Xinru, Ancient India and Ancient China: Trade and Religious Exchanges A D 1–600 (Delhi, Oxford University Press, 1994).

239

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

xinru liu Liu, Xinru, The Silk Roads: A Brief History with Documents (Boston, Bedford/St. Martin’s, 2012). Rong Xinjiang 荣新江 and Zhang Zhiqing 张志清 (eds.), Cong Samaerhan dao Changan: Suteren zai Zhongguo de wenhua yiji 从撒马尔罕到长安：粟特人在中国的文化遗 迹 (Beijing, Beijing tushuguan chubanshe, 2004). Schafer, Edward H., The Golden Peaches of Samarkand: A Study of Tang Exotics (Berkeley, University of California Press, 1985). Schmidt-Colinet, Andreas, Annemarie Stauffer, and Khaled al-As‘ad, Die Textilien aus Palmyra: Neue und alte Funde (Mainz, Verlag Philipp von Zabern, 2000). Sen, Tansen, Buddhism, Diplomacy, and Trade: The Realignment of Sino-Indian Relations, 600– 1400 (Honolulu, University of Hawai‘i Press, 2003). Skaff, Jonathan, Sui–Tang China and Its Turko-Mongol Neighbors (New York, Oxford University Press, 2012). Vaissière, Étienne de la, Sogdian Traders: A History (Leiden, Brill, 2005). Wang, Helen, Money on the Silk Road (London, British Museum Press, 2004). Wang Yongxing 王永兴, Tangdai jingying xibei yanjiu 唐代经营西北研究 (Lanzhou, Lanzhou daxue chubanshe, 2010). Whitfield, Susan, Life along the Silk Road, 2nd ed. (Berkeley, University of California Press, 2015). Whitfield, Susan, Silk, Slaves, and Stupas: Material Culture of the Silk Road (Berkeley, University of California Press, 2018). Whitfield, Susan (ed.), Silk Roads: Peoples, Cultures, Landscapes (London, Thames and Hudson, 2019). Xu Xuya 许序雅, Tangchao yu Zhongya jiuxinghu guanxi shi yanjiu 唐朝与中亚九姓胡关 系史研究 (Lanzhou, Lanzhou daxue chubanshe, 2012). Yang, Bin, Between Wind and Clouds: The Making of Yunnan, Second Century B C E to Twentieth Century C E (New York, Columbia University Press, 2009). Young, Gary K., Rome’s Eastern Trade: International Commerce and Imperial Policy, 31 B C –A D 305 (London, Routledge, 2001). Zhang Xiaogui 张小贵, San yijiao yanjiu 三夷教研究 (Lanzhou, Lanzhou daxue chubanshe, 2014).

240

https://doi.org/10.1017/9781108587334.007 Published online by Cambridge University Press

*

INTERLUDE

Published online by Cambridge University Press

Published online by Cambridge University Press

The Tang–Song Transition in Chinese Economic History richard von glahn

At the turn of the twentieth century, Chinese scholars eager to assimilate the historical sciences of the West and incorporate their history into universal historical narratives readily adopted the tripartite periodization of Western civilization divided into ancient, medieval, and modern epochs. This framework of linear, stadial progression toward modernity offered liberal intellectuals in China the promise of emancipation from China’s stultifying past and rebirth as full citizens in a modern world of equal nation-states. Marxist scholars invoked a parallel tripartite periodization divided into slave, feudal, and capitalist epochs, but adapted to accentuate the defining feature of Chinese history: the rise of a “bureaucratic, centralized feudal state” that fostered “economic stagnation” throughout the longue durée of the imperial era, from the first universal empire of Qin in the third century B C E to the irruption of Western imperialism in the nineteenth century.1 The ideas of “oriental stagnation” and the “Asiatic mode of production” likewise inflected Western historiography on China, and the notion of an unchanging “traditional China” prior to the advent of the West in the post-Opium War era predominated in Western scholarship on Chinese history down to the 1970s. Japanese scholarship, in contrast, was driven by a different ideological agenda: to recuperate a common “oriental” heritage in which an ascendant, modernizing Japan would displace a debilitated, stagnant China as its leading nation. In 1914, Naito¯ Konan, a founding member of the Department of Oriental History at Kyoto University, published his book Thesis on China in which he advanced the proposition that China’s “modern age” began in the transition from the Tang to the Song dynasties.2 For Naito¯, this “Tang–Song transition” marked the collapse 1

2

Fan Wenlan 范文瀾 (ed.), Zhongguo tongshi jianbian 中國通史簡編 (Shanghai, Shenghuo dushu xinzhi Shanghai lianhe faxingsuo, 1949). The translation of the term used by Naito¯ for “modern,” kinsei 近世, requires some clarification. Japanese historians at that time had coined two terms for the modern age:

243

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

richard von glahn

of China’s feudal society, in which autocratic imperial government supplanted the aristocratic order of the early empires. Despite the imposition of autocratic rule, the passing of the aristocratic order also unleashed new productive and social forces and a vibrant commoner culture, enabling previously marginalized groups such as merchants and peasants to achieve some measure of autonomy as “self-governing communities.” Writing in the immediate aftermath of the fall of the Qing dynasty in 1911, Naito¯ believed that a new republican political order would be built on the foundation of these self-governing communities. In subsequent writings he elaborated on the idea of the Tang–Song period as a transition from “middle antiquity” to “modernity,” largely focused on political, intellectual, and cultural changes. In Naito¯’s own time few scholars were persuaded by either his hopes for a democratically based Chinese republic or his interpretation of the trajectory of Chinese history. Although Naito¯ himself devoted only modest attention to the economy, his student and successor at Kyoto University, Miyazaki Ichisada, made economic transformation the centerpiece of his book The Oriental Modern Age (To¯yo¯teki kinsei) (1950). Situating Chinese history within a global scheme of evolution from antiquity to the present, Miyazaki explicitly identified parallels between Song China and the European Renaissance as harbingers of “the modern age.” For Miyazaki, the salient features of “modernity” in both China and Europe were the secularization of society and culture following the rebirth of rational philosophy after ages dominated by Buddhist and Christian religious dogma and the rise of the market economy. The demise of the aristocratic order in China and Europe alike also liberated the rural population from serfdom; land became freely alienable, and economic relations such as tenancy based on formal contract superseded dependency based on ascriptive status. Miyazaki also detailed the wide range of economic changes, such as production for the market, division of labor, regional specialization of production, monetization of exchange and expansion of coinage, relaxation of state controls on commerce, urban growth, and monetization of taxation – in sum, the free disposition of land, labor, and capital – that marked the emergence of (early) modernity in Song China. Although the Naito¯–Miyazaki hypothesis about the “modernity” of Song (1) kinsei, which was equated with the incipient transition from feudalism to modernity, identified in Japanese history as the era of the Tokugawa shogunate (1603–1868) and marked by political centralization and commercial development; and (2) kindai 近代, modernity as defined by Western norms, which the Meiji state adopted as its developmental agenda after 1868. In recent Japanese historical scholarship kinsei is generally equated with “early modern,” and this sense can be applied to Naito¯’s use of the term as well.

244

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

The Tang–Song Transition in Chinese Economic History

China was roundly disputed, it also spurred formidable empirical research by a legion of Japanese scholars – most notably, in the prewar era, Kato¯ Shigeshi, Sogabe Shizuo, and Hino¯ Kaisaburo¯ – who confirmed the far-reaching impact of economic change in this era. In Western scholarship, the idea of the Tang–Song transition as a watershed in Chinese economic history was most forcefully and enduringly expressed by Mark Elvin in The Pattern of the Chinese Past (1973). Harvesting the fruits of Japanese scholarship, Elvin asserted that China experienced a precocious “medieval economic revolution” during the Song driven by technological innovations in agriculture, transport, finance, metallurgy, and textile manufacture. At the same time Elvin ascertained structural and cultural constraints that impeded further technological changes and a breakthrough to self-sustaining growth. Among these constraints were: (1) a knowledge system that favored investment in human capital but not in scientific experimentation or technological expertise, (2) highly integrated domestic markets that nonetheless were isolated from the global economy, and (3) steadily diminishing returns on investments in traditional technologies after the Song advances. More recent scholarship has challenged Elvin’s arguments about the impediments to technological change in post-Song China, and also has departed from Elvin’s singular focus on technology to adumbrate the manifold ways in which the rise of the market economy in the Song was sustained and expanded in the late imperial era. But Elvin’s thesis of a medieval economic revolution in China has become a staple of world-history narratives. During the 1970s and 1980s the Tang–Song transition paradigm became deeply implanted in Western scholarship on Chinese history. The landmark studies of Robert Hartwell – ranging over diverse fields, including coal and iron production and technology, professionalization of the fiscal bureaucracy, social transformation of the ruling elite, and regional economic development – were especially influential in this regard.3 In China, by contrast, the feudal paradigm has remained paramount, and Chinese historians have expressed skepticism about the transformative quality of the Tang–Song transition. Among Chinese scholars who have embraced the concept of the Tang–Song transition, it is the economic dimension – the rise of the market economy and a social order based on wealth rather than status – that has gained the most purchase.4

3

4

Robert M. Hartwell, “Demographic, Political, and Social Transformations of China, 750– 1550,” Harvard Journal of Asiatic Studies 42.2 (1982), 365–442. See, for example, Lin Wenxun 林文勋, Tang Song shehui biange lungang 唐宋社会变革 论纲 (Beijing, Renmin chubanshe, 2011).

245

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

richard von glahn

Naito¯ singled out the An Lushan Rebellion (755–763) as the crucial moment triggering the far-reaching changes encompassed by the idea of a Tang–Song transition. The rebellion of frontier armies led by An Lushan nearly toppled the Tang dynasty (618–907). Although the Tang court was restored to its capital at Chang’an after the rebellion was put down, its authority was badly shaken. Moreover, north China had been devastated by years of war, and after the restoration much of the Central Plain, China’s economic heartland, remained under the control of regional warlords exercising independent rule. The consequences of the rebellion were most immediately felt in the domain of fiscal administration. Even before the rebellion, the equal-field (juntian 均 田) system of state land allocations and the zu–yong–diao revenues – collected in kind (grain, textiles) or labor services – derived from it were plagued by insolvencies, abandonment, and evasion. Following the outbreak of the rebellion the equal-field system collapsed, and along with it the state’s principal source of revenue. In a desperate effort to raise emergency funds for the war effort, in 758 the Tang government instituted a monopoly on the production and sale of salt. Subsequently the newly created Salt Commission gathered mining, coinage, commercial excises, and transport of grain revenues into its portfolio and became the principal fiscal organ of the Tang government. In 780 the Tang leadership sought to re-establish a stable base of direct taxation by jettisoning the zu–yong–diao system in favor of a new “twice-a-year” tax (liangshui 兩稅) levied on household property and assessed in coin as well as grain (in late summer and late autumn respectively, hence the name twice-a-year tax). Households were graded into nine ranks of property ownership – based on landholdings, urban real estate, or business income – for assessment of the new levies. Even after the institution of the twice-a-year tax, however, the Tang government continued to rely heavily on revenues derived from consumption (especially the salt monopoly) and commerce, inaugurating a shift away from direct to indirect taxation. The twice-a-year tax would remain the primary basis of taxation and fiscal administration throughout the rest of China’s imperial era. It also marked a crucial shift from persons to property as the object of direct taxation, which had important implications for the burden of fiscal demands on economic activity. In addition, the new tax system eliminated the universal statutory labor service owed to the central government. Local officials continued to conscript taxpayers for ad hoc public works projects and assigned village officer duties to the highest-ranked – that is to say, the most affluent – households. Coupled with the cessation of universal military conscription in the 730s, the suspension of regular labor service duties had the salutary effect of freeing up 246

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

The Tang–Song Transition in Chinese Economic History

considerable labor resources for private purposes. The new requirement to pay a portion of the twice-a-year tax in coin exacerbated existing shortages of currency, however. By the 820s the government was forced to commute nearly all of the tax collected in money to payments in textiles. The demise of the equal-field land allocations effectively ended the imperial state’s efforts to control land and interdict the concentration of landownership. The Tang government conceded the principle of private ownership of land, facilitating the development of a vibrant market for arable land, and for real estate generally. Although wealthy landowners could accumulate substantial estates, small family farms proved to be more efficient economic enterprises, particularly in the rice-growing regions of the south. Over the long term the small family farm would endure as the fundamental economic unit throughout the late imperial era. Naito¯ proposed that the aristocratic clans, whose power derived from their privileged access to government office and political power at the Tang court, went into permanent decline after the An Lushan Rebellion. The fate of the old aristocracy was sealed during the eleventh century, when the Song dynasty (960–1279) adopted civil service examinations as the principal means of recruitment to government office. Although the Song recruitment system still retained certain privileges for relatives of high officials, over the long run the civil service examinations had a leveling effect that prevented the re-emergence of a tight-knit oligarchy. Recent prosopographic research has shown that the aristocracy retained its coherent social identity and continued to dominate political office both at the capital and in the provinces in the post-An Lushan era. But the prolonged civil wars that erupted in the 880s and culminated in the fragmentation of the Tang empire into a multitude of regional states in 907 dealt a fatal blow to the aristocratic clans and their social and political networks.5 During the Song period, the examination system and the spectacular expansion of schooling and literacy (abetted by similarly dramatic growth in printing and publishing) greatly expanded the potential pool of government officials. At the same time the intensifying competition for examination success compelled elite families to pursue more diverse strategies to maintain their status, power, and wealth. A notable feature of the Song elite was a turn toward “localist” strategies for social success: marriage alliances with locally prominent families, informal leadership in local institutions such as schools and temples, and investment in local 5

Nicolas Tackett, The Destruction of the Medieval Chinese Aristocracy (Cambridge, MA, Harvard University Asia Center, 2014).

247

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

richard von glahn

communities through public works, famine relief, and other charitable activities. Elite lineages no less than humble farming families faced both the opportunities and the pitfalls of a dynamic market economy and were obliged to optimize their human and material resources to ensure economic survival and prosperity for their descendants. The dissolution of the Tang Empire into as many as a dozen regional states in the tenth century spawned intense military and economic competition. In the north, a series of five successive dynasties claiming the mantle of the Tang Empire rose and fell before the establishment of the Song dynasty in 960 (hence this period from 907 to 960 is generally known as the Five Dynasties era). Actual power devolved to dozens of local warlord regimes. In the south, however, a relatively stable multistate system of seven regional kingdoms arose, the last of which submitted to Song rule only in 978. The emergence of these independent regional states stimulated robust economic growth across southern China. Regional specialization in the production of tea, salt, timber, paper, porcelain, copper, silver, and textiles accelerated. Augmenting the trend toward mercantilist fiscal policies adopted in the wake of the An Lushan Rebellion, the southern kingdoms sought to capitalize on their comparative advantages in resources to marshal economic and military power. Although political fragmentation and hostile relations posed obstacles to interregional trade, the rulers of these states depended on commerce to obtain vital supplies – notably iron, salt, sulfur, and alum – as well as now indispensable consumer staples such as tea. The southern states developed cordial trading relations with the remote foreign states on the northern frontier, the Tanguts and the Khitan Liao, principally based on the exchange of tea for war horses. The coastal kingdoms of Southern Tang (937–975), Wu-Yue (902–978), Min (909–946), and Southern Han (909–971) actively promoted overseas trade with the Khitan in Manchuria, the Koryŏ kingdom in Korea, and Japan, as well as Southeast Asia and the Indian Ocean world. Land-poor Min, in modern Fujian province, became especially dependent on maritime trade for resources and revenues. In order to prevent the outflow of its limited supply of bronze currency, in 916 the Min state began to issue low-value lead coins, and subsequently iron coins as well. The neighboring states of Southern Han and Chu (907–951), likewise anxious to husband their bronze currency reserves, quickly followed suit in 918 and 923 respectively (Map i.1). Despite these debasements, cross-border trade boomed; since the lead and iron currencies circulated only within the states that issued them, merchants from elsewhere were obliged to trade for local products for which there was demand in external markets. Landlocked Chu, rich in tea, was wholly 248

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

The Tang–Song Transition in Chinese Economic History

LIAO Sixteen prefectures ceded to Liao in 937

LATER JIN Kaifeng Nanjing

LATER SHU

Chengdu

Hangzhou

JINGNAN

SOUTHERN TANG

Jiangling

WUYUE

Changsha

YIN

CHU

Fuzhou

MIN Zhangzhou Guangzhou

A

N

N

AM

SOUTHERN HAN

0 0

600 km 400 miles

Map I .1 Political fragmentation and currency regions, c. 943 Source: adapted from Denis Twitchett and Paul Jakov Smith (eds.), The Cambridge History of China, vol. 5, part 1, The Sung Dynasty and Its Precursors, 907–1279 (Cambridge, Cambridge University Press, 2009), p. 2

dependent on imported salt. Because Chu did not impose any commercial taxes or customs duties, merchants from other states flocked there to trade salt for tea. Southern Tang, the wealthiest and most powerful of the southern states (it also possessed prolific copper mines), resisted the temptation to debase its currency until 959, after losing its territories north of the Yangzi river to the Northern Zhou and then Song regimes. The mercantilist strategies

249

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

richard von glahn

of the southern kingdoms, focused on preserving domestic stocks of fine coin while promoting international trade, continued to exert influence over fiscal policymaking in the Song period as well (see the chapter by Lamouroux and von Glahn in this volume). Perhaps the most profound long-term consequence of the An Lushan Rebellion, then, was the rise of south China, and especially the lower Yangzi river basin, as the new economic heartland of the Chinese world. The rebellion’s devastation of the Central Plain spurred massive migration from north to south, accelerating a demographic trend that had been building since the fall of the Han dynasty. On the eve of the An Lushan Rebellion, two-thirds of the Tang Empire’s population inhabited the regions north of the Huai river valley, the rough line of demarcation between wheat and rice cultivation. By 1100, two-thirds of the Song Empire’s population lived south of the Huai river (see Cao’s chapter in this volume). This redistribution of population – and the sustained surge in demographic growth over these centuries – stimulated the rapid development of rice cultivation and a profusion of new crop varieties, tools, and cultivation techniques to improve grain yields. Rice agriculture sustained far higher population densities than the wheat and millet staples of north China, and rice cultivation in turn required substantially greater inputs of labor and capital. Rice cultivation also spurred massive public and private investment in water control, including canals that served to facilitate transport and trade as well as flood control and irrigation. The rich natural resources of south China also fostered the emergence of new technologies and industries. Settlement of the hilly inland regions of south China was encouraged by the spread of tea cultivation and exploitation of mines and forests. Tea drinking had become popular among the Tang elite, but in Song times tea became – along with rice, firewood, vegetable oil, salt, soy sauce, and vinegar – one of the “seven items indispensable to the daily life of every household,” according to a thirteenth-century chronicler of the Southern Song capital, Hangzhou.6 The development of true porcelains using highly refractory kaolin clays recentered ceramics industries in southern areas, leading to the eventual emergence of Jingdezhen in Jiangxi as the porcelain capital of the world. The evergreen conifer forests of south China supplied the key raw materials for shipbuilding, one of many industries that underwent significant technological improvement. The adoption of deepkeeled ships, stern-post rudders, and the nautical compass enhanced the capabilities of seafarers to venture overseas. Southern forests also supplied 6

MLL, 16, p. 270.

250

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

The Tang–Song Transition in Chinese Economic History

the main ingredients – paper and ink – for the burgeoning printing industry, with most of the main printing centers, apart from the Song capital, Kaifeng, located in south China as well. Southern mines yielded the massive quantities of copper that enabled the Song government to increase the output of bronze currency more than tenfold compared to the Tang, and supplied ample quantities of silver as well. The humid climate of the Yangzi delta was ideal for raising silkworms, and over the course of the Song dynasty sericulture and silk-weaving steadily relocated there from traditional centers of silk manufacture in the north such as Shandong. The coal and iron industries, which attained levels of output in the eleventh century that China would not again reach before modern times, were perhaps the only key industries still firmly anchored in north China. All of these goods – as well as other southern agricultural produce such as sugar and citrus fruits – became the major commodities traded in regional, national, and even international markets. Growth in economic output was facilitated by rising demand (population growth and urbanization), security of property and labor (tenurial rights and release from labor and military services), and lower transaction costs (transport improvements, reduced state regulation of trade, and development of financial instruments such as contracts, bills of exchange, and paper money). All of these developments will be addressed in subsequent chapters, but I wish to touch on a few of them here to provide more concrete illustration of the economic changes that accompanied the Tang–Song transition. From the onset of the imperial era, the imperial state tightly regulated merchants and commercial activities. Urban trade was restricted to designated marketplaces enclosed by walls and subject to constant government scrutiny. This sequestration of markets from the surrounding city was retained in major Tang cities such as the capital, Chang’an, which featured a chessboard-like grid of walled and gated residential wards that were sealed off during nighttime curfews. Apart from taverns, courtesan brothels, and other places of entertainment, commercial enterprises were restricted to Chang’an’s two great marketplaces: the bustling western market, the terminus of the Silk Road and the hub of international and wholesale trade, and the far more sedate eastern market, nestled among the residential wards of government officials and aristocratic families and catering to their often ostentatious tastes. After the An Lushan rebellion, however, this system of urban control collapsed. Residents pierced the outer walls of wards to open up their residences to the main thoroughfares, which swiftly became lined with shops and market stalls. Curfews were abandoned, and night markets sprang up in the most heavily trafficked neighborhoods. Shopkeepers and 251

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

richard von glahn

artisans were no longer confined to the alleyways assigned to each particular trade in the official marketplaces, but they often congregated in specialized commercial districts – for example, for jewelers, goldsmiths, silk goods shops, and booksellers, as well as butchers, grain merchants, and lumber dealers – scattered throughout the city. By the end of the Tang dynasty the old official self-enclosed marketplaces were defunct everywhere. In Song cities, shops and marketplaces lined the streets and canals within the walled cities, while faubourgs of inns, warehouses, and the shops of wholesale merchants and brokers mushroomed outside the city walls. The new urban morphology encapsulated many of the transformative social and economic changes of the Tang–Song transition (see Zurndorfer’s chapter in this volume). Another emblematic feature of the Tang–Song transition was the revival of the conjugal farm family as the fundamental social and economic unit of Chinese society. The free disposition of landed property after the demise of the equalfield land allocation system necessitated a new institutional framework for the transfer, sale, and inheritance of real property, the foundation of the family’s economic welfare. Although the household (hu 戶) was the basic unit of production, consumption, and taxation, property ownership inhered not in the household but in the patriline (jia 家), the family conceived in terms of its ritual identity, linked to its ancestors in the past and its descendants in the future. The patriline was the crucial unit of biological and economic reproduction across time.7 The long-standing principle of equal division of property among male heirs – each of whom constituted a separate patriline – precluded the preservation of intact landholdings across generations. The head of the household had broad rights to dispose of property as he saw fit as long as he abided by the legal obligation of equal inheritance and the deep-rooted cultural commitment to preserving the patrimony, the material foundation of the patriline’s welfare, for future generations. Individuals could own property – for example, dowries, salaries and wages, and businesses created without using family property – which they could dispose of at will. But real property and businesses built up through investment of family assets were deemed a common patrimony that was subject to strong legal and customary rules. Song law and jurisprudence strongly encouraged the continuity of landownership through the patriline. The sale of arable land – deemed the fundamental means of ensuring the continuity of the patriline – was sanctioned only as a last resort. The most common vehicle for the transfer of property rights in land 7

On the concept of patriline and its role as an economic institution, see Patricia Ebrey, “Conceptions of the Family in the Sung Dynasty,” Journal of Asian Studies 43.2 (1984), 219–45.

252

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

The Tang–Song Transition in Chinese Economic History

was the conditional sale, in which the seller transferred the use of the land to the buyer but retained the right to redeem ownership upon repayment of the original sale price within twenty years (see Kishimoto’s chapter in this volume). Alternatively, the owner would relinquish all claims to the land in an irrevocable sale. Since a conditional sale conveyed – in principle, at least – only usufruct rights, the prices of conditional sales were far lower than irrevocable sales. In practice, redemption of conditional sales appears to have been rare. A third type of land transaction, known didang 抵當, should be understood as a type of securitized loan. Under the didang arrangement, the seller retained possession and use of the land, but forfeited ownership rights if he failed to repay the principal and interest on the loan (through the payment of “rent”) by the specified deadline. In 1027 the Song state outlawed the practice of didang as well as repayment of cash debts by transfer of landownership rights in an effort to protect lands against debt seizures. In its effort to deter didang transactions, the Song state mandated that sellers quit the land and surrender usufruct rights to the buyer. But this deeply engrained customary practice persisted in defiance of the strict letter of the law. Thus Song law favored retention of residual rights to landed property through the institution of conditional sales and by prohibiting forms of mortgage that were regarded as the slippery slope to outright loss of the family’s patrimony. Other restraints on the alienation of land included the right of first refusal in land sales granted to kinfolk and neighboring landowners, subject to a three-year statute of limitations (reduced to one year in 1135). Nonetheless, a highly active market in arable land emerged in the Song, and property rights became vigorously disputed. Land sales were governed principally by written contracts – usually private contracts, in defiance of the rule that contracts be notarized by government officials – but local magistrates intervened forcefully in settling property disputes. Judicial determinations often included testimony by parties to a dispute and witnesses, as well as evidence of tax payments, but written contracts, including private ones, and adherence to legal statutes were the pre-eminent considerations. The ideological commitment to preserve the patrimony and ensure the continuity of the patriline also induced the Song government to grant unprecedented recognition to the decision-making authority of widows. Although widows could not inherit the property of their deceased husbands, in the absence of majority-age male heirs widows were recognized as the best representatives of the patriline’s interests and caretakers of its patrimony. Magistrates entrusted widows with authority over the adoption of heirs in cases where there were no male offspring and with management of the family’s property and commercial 253

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

richard von glahn

businesses as long as they did not divest real assets. Despite their fealty to patriarchal principles, Song authorities recognized that male kinsmen tended to favor the interests of their own patriline, while widows were supremely devoted to the welfare of their own children and their husband’s legacy. It should be noted, however, that the prominent role of widows as substitutes for their husbands in the management of family affairs did not endure. By the Ming dynasty widows were deprived of such authority, which instead was vested in the husband’s male kinsmen. A further consequence of the institution of conditional sales was the emergence of multitiered landownership rights. Conditional sales in effect divided property rights in two: the seller (and his heirs) retained residual rights of redemption for a specified period of time, while the proprietary rights obtained by the buyer (including usufruct right of cultivation) could be transferred to another party without the consent of the original seller. The frequent – however illegal – practice of non-possessory sale (didang) resulted in the seller becoming a tenant who leased the land from the buyer. Over time these practices resulted in customary tenurial rights on the part of tenant cultivators and the ability of the nominal landowner to transfer ownership rights to other parties at will. The state recognized only one property owner – defined as whoever paid the twice-ayear tax – but tolerated the customary practice of alienable, multitiered property rights. The concession of strong tenurial rights to cultivators in turn reinforced the stability of the small family farm as the fundamental feature of the late imperial agrarian economy (see Pomeranz’s chapter in this volume). These examples illustrate a few of the new economic conditions and institutional innovations engendered by the rise of the market economy, the signal feature of the Tang–Song transition. Mention also should be made of the profound intellectual and cultural changes wrought by the demise of the aristocracy, the enhancement of state power, the institution of civil service examinations, the expansion of schooling, and efforts by elite families to diversify their strategies for social success. Confucian social and political values, reformed and reinvigorated by a renaissance of Confucian learning, examination and school curricula, and novel thinking about metaphysics, ethics, and politics, prevailed in political discourse, philosophical investigation, and social ideology. This “Neo-Confucian” movement still honored court-centered ritual practice, but opened up new vistas of moral action and social leadership. A novel selfimage of the ruling elite as “literati” (shi 士, or shidafu 士大夫) – steeped in Confucian learning, trained in the arts of government, but above all pursuing self-cultivation in all aspects of life, from personal conduct to family management, public leadership, and aesthetic accomplishment – took root. Literati 254

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

The Tang–Song Transition in Chinese Economic History

identity ultimately inhered in this pursuit of self-cultivation rather than examination success, careers in government service, or social pre-eminence. Despite the state’s efforts to instill intellectual orthodoxy through the civil service examinations and state schools, fervent philosophical debate flourished, largely in a domain of civic discourse increasingly autonomous from the imperial state and its institutions. Nor was status within the intellectual elite congruent with wealth and social power. Indeed, the blurring of social boundaries and economic standing became a cardinal feature of the late imperial era.

Further Reading Clark, Hugh, “Why Does the Tang–Song Interregnum Matter?”, Journal of Song–Yuan Studies 46 (2016), 1–28; 47 (2017–2018), 1–31; 49 (2020), 1–44. Elvin, Mark, The Pattern of the Chinese Past (Stanford, Stanford University Press, 1973). Fogel, Joshua A., Politics and Sinology: The Case of Naito¯ Konan (1866–1934) (Cambridge, MA, Harvard University Council on East Asian Studies, 1984). Hartwell, Robert M., “Demographic, Political, and Social Transformations of China, 750– 1550,” Harvard Journal of Asiatic Studies 42.2 (1982), 365–442. Lau Nap-yin 柳立言, “Hewei Tang Song biange?” 何謂唐宋變革, Zhonghua wenshi luncong 中華文史論叢 81 (2006), 125–71. Li Huarui 李华瑞, “‘Tang Song biange’ lunde youlai yu fazhan” 『唐宋变革』论的由来 与发展, Hebei xuekan 河北学刊 2010.4, 57–65; 2010.5, 67–77. Luo, Yinan, “A Study of the Changes in the Tang–Song Transition Model,” Journal of Song– Yuan Studies 35 (2005), 99–127. Miyakawa, Hisayuki, “The Naito¯ Hypothesis and Its Effects on Japanese Studies of China,” Far Eastern Quarterly 14.3 (1955), 533–52. Miyazaki Ichisada 宮崎市定, To¯yo¯teki kinsei 東洋的近世 (Kyoto, Kyo¯iku taimusu sha, 1950). Naito¯ Torajiro¯ (Konan) 內藤虎次郎(湖南), Shina ron 支那論 (Tokyo, Bunkaido¯ shoten, 1914). Satake Yasuhiko 佐竹靖彦, “Zo¯setsu” 総説, in Satake Yasuhiko et al. (eds.), So¯ Gen jidaishi no kihon mondai 宋元時代史の基本問題 (Tokyo, Kyu¯ko shoin, 1996), pp. 3–42. Tackett, Nicolas, “A Tang–Song Turning Point,” in Michael Szonyi (ed.), A Companion to Chinese History (Hoboken, Wiley Blackwell, 2017), pp. 118–28. von Glahn, Richard, “Imagining Pre-modern China,” in Paul J. Smith and Richard von Glahn (eds.), The Song–Yuan–Ming Transition in Chinese History (Cambridge, MA, Harvard University Council on East Asian Studies, 2003), pp. 35–70.

255

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

https://doi.org/10.1017/9781108587334.008 Published online by Cambridge University Press

part

II

*

1000

Published online by Cambridge University Press

TO

1800

Published online by Cambridge University Press

7

Ecological Change and Resource Constraints david a. bello

Intricate relations between people, animals, and plants were the basis of the entire imperial Chinese order no matter what dynasty was ostensibly in charge. Such relations were environmental in the sense that they formed interdependencies between species under diverse ecological conditions of climate and topography. The most significant environmental historical result of these relations for the eight centuries under study here was agriculture, the main source of China’s human-induced (or “anthropogenic”) ecological change. An extended, instructive example of the intricacies of farming’s requisite resource management comes from the Ming (1368–1644) town of Pingwang in the Yangzi delta, likely the most developed area of contemporary imperial agrarian practice: To search for fertilizer, Pingwang is the place to go. The town produces ox dung called “ground earth” and barnyard manure from pigs. Those are the best fertilizers for the paddy field . . . As for nightsoil, Hangzhou is the best place to go . . . Fresh nightsoil is most effective . . . (during the height of the mulberry season) . . . one can only go to nearby towns to purchase “commode nightsoil.” If one can purchase it in the morning and apply it to the field in the afternoon, it will be very effective.1

This account from a late Ming agricultural treatise illustrates the interdependency of human, vegetable, and animal waste in the production of manure that, in turn, produced staple grains like rice in quantities sufficient to keep an empire fed. Pingwang’s manure was a by-product of the town’s rice1

Quoted from Yong Xue, “‘Treasure Nightsoil as if It Were Gold’: Economic and Ecological Links between Urban and Rural Areas in Late Imperial Jiangnan,” Late Imperial China 26.1 (2005), 44–5.

259

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

processing network, powered by oxen-driven mills. Other by-products of this human-orchestrated animal–plant production included pig feed from crushed grains that ultimately ended up as manure and nitrogen-rich straw and soil spread on the floor to catch the oxen’s excrement and urine that was ground into molu as the livestock turned the millstone. Even a human by-product, nightsoil, was included in the network as townspeople in densely populated places like nearby Hangzhou ate not just rice, but unusual quantities of readily available pork and seafood likewise high in nitrogen. Moreover, Chinese urban areas like Hangzhou probably ended up much cleaner than their European counterparts as a result. In these ways, China’s urban population, and its attendant rural domesticates, stimulated agricultural production not just by consumption, but by digestion. Pingwang was an exemplary realization of the sophisticated organization of environmental relations achieved by trans-dynastic effort, which nevertheless depended on many nonhuman ecological factors. Imperial China can thus be considered the product of a network of ecological forces which human culture adapted to and, then, strove to orchestrate within the resulting environmental parameters that also encompassed people. Agriculture was the main human adaptation during the imperial period, which made arable land the main resource constraint and motive for ecological change. Although this view is reductive, agriculture was unquestionably the single most important environmental relationship conditioning, and even defining, the Song, Yuan, Ming, and Qing dynasties that ruled over the varying expanses of China between the year 1000 and 1800 C E. Even by the highly commercialized eighteenth century, the state was still deriving almost 75 percent of its revenue from the land tax. Agricultural development was the primary concern of signature state environmental interventions like water control and disaster management. Deforestation was one of the most transformative effects of agricultural expansion, which also began to reach Inner Asian grasslands and woodlands. More arable land also produced more people, conventionally seen as the leading environmental actors. In sum, arable land can be directly connected to nearly all other major imperial ecological changes and resource constraints over this period of 800 years.

General Patterns General studies of China’s environmental relations have tended to proceed from the unusual scale of transformation wrought by a commensurately 260

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

extraordinary number of humans working in a highly systematic fashion over an extended period of time. The eight centuries covered in this chapter saw the peak development of China’s preindustrial agriculture, which can be traced back thousands of years into the Neolithic period. Consequently, an account of ecological change and resource constraints from the year 1000 to 1800 C E begins toward the end of a long-gestating process. Already, by 1000 C E, north China’s Central Plain had been effectively deforested for at least a millennium. The plain’s roughly France-sized loess plateau – as the cradle of Chinese civilization – had possessed one of the largest tracts of fertile soil in the world, which both ecological and anthropogenic change had altered even before 1000 C E. Enhanced exploitation of south China’s rich resources, water chief among them, had begun to move beyond its initial stages in response to these northern changes. Agrarian development’s accelerating pace was largely due to the timely introduction of new, quick-ripening Champa rice strains that would promote the conversion of millions of acres into paddies and the diversion of long stretches of the Yangzi river and its tributaries into irrigation canals. The Yellow River, despite its already extensive system of dikes, catastrophically experienced its third major shift that would divide north China’s main waterway into three courses by the mid-eleventh century. Many of these changes were related to the steadily rising population of the Song dynasty (960–1279), which appears to have doubled that of the preceding Tang dynasty (618–907) to rise to roughly 100 million people by 1100. This date falls approximately in the middle of the Medieval Warm Period (c. 1000–1300 C E) when temperature in many places in the northern hemisphere, including China, may have reached its warmest points prior to the twentieth century. A recent study of the vegetation history in the middle Yangzi during the last 1,500 years outlines how some of these environmental shifts wrought by people and ecology may have interacted to create change in different ways. Using data derived from pollen and charcoal sediments in Hubei province’s Taibai Lake, the study found that human activity had had limited effects on the regional vegetation up to 1320. There was, nevertheless, substantial expansion of evergreen broad-leaved forests at the expense of mixed-conifer woods between c. 1050 and 1320 that was apparently driven by climate change from the Medieval Warm Period, which was more favorable to the former forest type. From 1320 to 1650, the study found that humans replaced climate as the main agents of change that reduced the overall area of primary forest that included synchronous reductions of both evergreen broad-leaved and Pinus (pine) conifer forests, which the existing range of 261

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

contemporary climate change could not have accomplished. Overall reduction in forest cover is generally affirmed in Ming regional records of logging of evergreen conifers like Cunninghamia lanceolata (the Chinese “fir,” shanmu) and Pinus trees. This trend continued at a roughly similar rate until c. 1740, when a new set of dynamics took over. Environmental synergy between warmer climate and limited human forest clearance expanded evergreen broad-leaf cover from 1050 to 1320. However, there was a reversal during the much cooler Little Ice Age (c. 1400–1900) conditions of the late imperial period that, combined with much more extensive human clearance – including evidence of extensive swidden (slash-andburn) cultivation associated with highland in-migration and New World crops – radically reduced regional forest cover and increased erosion. Although both climate cooling and human agrarian clearance reduced forests and related biomass, the effects were still relatively moderate compared to conditions from the late eighteenth century onwards, when human cultivation and its consequent population increase were less and less constrained as the Little Ice Age gradually declined. The late imperial period was one of transition in middle Yangzi forest cover from change mainly driven by climate in the early fourteenth century to change mainly driven by humans after the mid-eighteenth.2 In broader terms, the combination of climate changes, involving both warming and cooling trends at different times and places, and the unprecedented progression of southward human expansion constituted the initial conditions for Chinese civilization’s access to critical resources over this period of eight centuries. Although inputs external to Chinese imperial territory were certainly vital – especially New World cultivars and minerals – it would have been difficult for Chinese to exploit these inputs to sustain and expand their resilience throughout the period if there had not already been a substantially developed domestic resource base with a commensurately elaborate structure of control to manage large-scale changes. The most basic ecological inputs, some of which were more subject to human intervention than others, were climate, soil, and water. The following three sections will take up some general issues pertinent to each for the period as a whole.

2

Xiayun Xiao et al., “Vegetation History and Dynamics in the Middle Reach of the Yangtze River during the Last 1500 Years Revealed by Sedimentary Records from Taibai Lake, China,” The Holocene 23.1 (2012), 65.

262

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

Climate The implications of warmer conditions that mark the beginning of the period seem relatively straightforward for a society whose environmental relations were rooted in cereal plants. Double-cropping, for example, was common to Jiangsu’s Lake Tai basin from the Southern Song (1127–1279), then became so constricted under subsequent cooling conditions that single-cropping eventually became the norm in the eighteenth century. “Warmer” is nevertheless an imprecise term with counterintuitive implications when applied to the diverse and fluctuating expanses of imperial China. In broad terms, China’s overall climate patterns, aside from other ecological factors, divide the country’s main staples into wheat and millet north of the Yangzi river and rice south of it. A recent international study has estimated that a one-degree Celsius increase in temperature – a change equivalent to that some studies find to have occurred in China during the Medieval Warm Period – could lead to a maximum decline in Chinese wheat yields of over 5.5 percent. In other words, warmer temperatures do not necessarily produce higher yields.3 The warmer conditions throughout China during the Medieval Warm Period actually manifested themselves quite differently in agricultural terms in different regions of China. There were two main warming episodes (peaking in 1000 C E and 1250 C E), separated by a comparatively cooler one. During the first warm interval, annual mean precipitation generally increased along the Yangzi river, while precipitation decreased in north China and in the far south. At the level of seasonal variation, there were different patterns. In winter, precipitation generally increased except for the northwest, which experienced a decrease. Furthermore, the magnitude of precipitation changes in summer was larger than those in winter over northeastern and northwestern China, as well as in the middle Yangzi. These precipitation variants were, in turn, caused by wind variations that formed southwesterly anomalies over south China, dumping more rain on the Yangzi river valley in summer. Easterly anomalies prevented a similar extra bounty of moisture from the Bay of Bengal and the Arabian Sea from falling on south China. An analogous northwesterly anomaly similarly deflected rainfall from the northwestern Pacific Ocean away from north China.

3

Center for the Study of Carbon Dioxide and Global Change, “Medieval Warm Period in China,” CO2 Science 16.31 (July 2013), 12; Chuang Zhao et al., “Temperature Increase Reduces Global Yields of Major Crops in Four Independent Estimates,” PNAS 114.35 (2017), 9327.

263

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

One way to sum up these complex dynamics, which, like the very existence of a Medieval Warm Period, are still debatable, is that the Yangzi river valley was the prime agrarian beneficiary of the first interval’s climate changes, which brought both warmer and wetter air to the region. North China agriculture, however, was far less fortunate in terms of the resulting moisture, if not in terms of temperature. Climate change from the Song to the Yuan dynasties tended to favor southern over northern agriculture, which coincides with the dramatic expansion – especially in the Yangzi delta – of paddy acreage, although there was also upland terracing. One problem with this rather neat correlation is the centennial timescale upon which it is based. If a multi-centennial timescale is adopted, several critical differences emerge. A recent study reveals more detailed variation to distinguish two main regions. One major regional division, west China, was influenced climatically by unobstructed westerly winds from the distant Atlantic Ocean and experienced a dry to moderately dry climate. The other major division, east China, which was subject to the summer East Asian Monsoon from the Pacific Ocean, experienced a comparatively wet climate, but with significant differences in subregional precipitation that distinguish a moderately wet north China zone and a moderately dry south China zone roughly divided along the Huai river. The remaining major region, southwest China, exhibited no consistent pattern – possibly because its hydroclimate is mainly driven by a mechanism distinct from both the westerlies and the East Asian Monsoon. This third mechanism – the South Asian Monsoon – pervades the southwest China region, including some of southern Tibet, unobstructed by the rain shadow of the Himalayas shielding northwestern China and its Inner Asian hinterlands. In multi-centennial perspective, during the Medieval Warm Period, west China was dry and east China was wet. Moreover, within east China, its northern subsection was wetter relative to its southern subsection.4 Whatever the precise nature of the warm period, it did not last, and with the onset of the Little Ice Age, a period of prevailing cold climate coincided with the dynasties of late imperial China, the Ming (1368–1644) and Qing (1644–1912). During this period, conditions for China’s major regional divisions were basically the inverse of the Medieval Warm Period at a multicentennial scale, with westerly areas of west China wetter and East Asian monsoon areas of east China drier. East China’s northern subsection was 4

Jianhui Chen et al., “Hydroclimatic Changes in China and Surroundings during the Medieval Climate Anomaly and Little Ice Age: Spatial Patterns and Possible Mechanisms,” Quaternary Science Reviews 107.1 (2015), 99, 103–4.

264

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

Table 7.1 Hydroclimate by region during the Medieval Warm Period and Little Ice Age Region West Chinaa East Chinab Northern subregionc Southern subregiond Southwest China

Medieval Warm Period (c. 1000–c. 1300 C E )

Little Ice Age (c. 1400–c. 1900 C E )

Dry

Wet

Wet Dry No pattern

Dry Wet No pattern

a

Regions of China under influence of westerlies (northwest China, Central Asia, northern Tibetan plateau and western Mongolia), roughly north of 30° N. b Regions of China under summer East Asian monsoon influence (including Manchuria), roughly east of 105° E. c Subregion of east China, roughly north of the Huai river at 34° N. d Subregion of east China, roughly south of the Huai river at 34° N. Source: Adapted from Jianhui Chen et al., “Hydroclimatic Changes in China and Surroundings during the Medieval Climate Anomaly and Little Ice Age: Spatial Patterns and Possible Mechanisms,” Quaternary Science Reviews 107.1 (2015), 98–111

drier compared to relatively wet south China. Again, however, southwest China manifested no clear pattern. If China’s Medieval Warm Period and Little Ice Age climates are compared, it is possible to gain some idea of general climate trends at a multi-centennial scale from Song to Qing up to 1800. Table 7.1 reflects one conceptualization of the extreme variation in regional precipitation during periods of both climate warming and climate cooling at multi-centennial scales that nevertheless exhibits a broadly consistent inverse pattern alternating between west and east China and between the north and south subregions of China (Map 7.1).5 Current research suggests that these variations are substantially driven, in turn, by El Niño/La Niña (ENSO) alterations in the Pacific Ocean, which manifest differently in its eastern and western sections. In El Niño years, when the eastern tropical Pacific is relatively warm, the main monsoon precipitation belt is held stationary in south China centered over the Yangzi valley, which then tends to flood, and is kept out of north China, which becomes drought-prone. In La Niña years, when the eastern tropical Pacific is relatively cold, regional 5

At least three different versions of China’s historical climate shifts are current in the sinophone literature alone; see Liu Ts’ui-jung, “A Retrospection of Climate Changes and Their Impacts in Chinese History,” in Carmen Meinert (ed.), Nature, Environment and Culture in East Asia: The Challenge of Climate Change (Leiden, Brill, 2013), 107–36.

265

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

105° E

East (north)

West

33° N

30° N

Yellow River

er

Huai river Yan g z

v i ri

East (South)

Southwest

Map 7.1 Hydroclimate regions Source: created by the author

conditions are reversed. Even interior regions can be affected by ENSO, although to a lesser degree than in east China. In terms of the implications for the management of its agricultural resources, imperial China during most of this time not only did not experience hydroclimate trending in a single direction as temperature tended to do, but actually experienced inverse regional hydroclimates simultaneously. The domain of the Northern Song (960–1126), for example, included all of east China and parts of west and southwest China, as defined in Table 7.1. During the contemporary Medieval Warm Period, this meant that dynastic authorities would have had to contend with inverse conditions between the empire’s western peripheries, relatively prone to drought, and its eastern core, relatively more prone to flood, but with the complicating proviso that its northern subregion was more likely to flood in comparison with its southern region, which was more likely to see drought. In climatic and spatial contrast, the Qing realm, more than twice as large at its mid-eighteenth-century peak and about 1 °C cooler than the Northern Song, was experiencing inverse conditions in generally expanded versions of these regions within 800 years.

266

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

Regional variation makes it more difficult to link climate change with dynastic change, a perennial historical theme that has been substantially augmented by climate science research. A number of recent studies have found significant correlations between cold periods of reduced agricultural production and increased social conflict. One study has even found evidence for a thirteenth-century pluvial period of prolonged, above-average moisture that uniquely raised the net primary productivity of steppe resources critical for the rise of Chinggis Khan’s empire.6 Climate, however, was far from the only variable for which imperial Chinese state and society had to account. Indeed, it was the only one of the three ecological inputs immune to conscious human intervention. Fortunately, the other two inputs, soil and water, were susceptible to manipulation by imperial cultivators and administrators.

Soil Globally, only about 11 percent of soils can be farmed without some sort of additional input, such as irrigation or manure. Even in these comparatively fertile regions, without careful maintenance to replenish soil nutrients and prevent erosion soil quality will rapidly decline as a result of farming. These two tasks have traditionally been the main challenges to sustainable agriculture, which inherently degrades fertile soil. Crops not only extract nutrients, but also do not hold soil in place nearly as well as more diverse natural vegetation does. Plowing reduces soil-dwelling organisms that promote fertility. Irrigation can bring subterranean herbicidal salts to the surface. These are some of the main dynamics that render soil a kind of mineral resource that is mined by farming. Even considerable twentieth-century technological advances, led by inorganic nitrogen-based fertilizers, have not permanently changed these dynamics, only temporarily arrested them.7 Chinese farmers adapted their techniques in critical ways from the Song to Qing periods to achieve an unprecedented intensification of farming through “unremitting attention to the conservation of soil productivity.”8 One key practice was to keep the soil in continuous production without fallowing 6

7

8

David D. Zhang et al., “Climate Change and War Frequency in Eastern China over the Last Millennium,” Human Ecology 35.4 (2007), 403–14; Neil Pederson et al., “Pluvials, Droughts, the Mongol Empire, and Modern Mongolia,” PNAS 111.12 (2014), 4375–9. Stephen Mosely, The Environment in World History (London, Routledge, 2015), pp. 56–7, 60, 74, 80. Mark Elvin, “Why Intensify? The Outline of a Theory of the Institutional Causes Driving Long-Term Changes in Chinese Farming and the Consequent Modifications

267

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

through complex schedules of crop rotation and a broad battery of fertilizers that ranged through vegetable, animal (including human “nightsoil”), silt, and lime. This practice effectively sped up the rate at which natural systems would have replaced soil nutrients over time. Soil fertility was one of imperial China’s most fundamental resource constraints, to which the practice of manuring, including inundation, without fallowing was the main response. Of course, the idea of fertility was relative across such vast expanses of imperial territory. Map 7.2 depicts the general distribution of China’s major soil regions as they currently exist. Soils classified as region 1 types, ferralsols, prevail in the extreme south and also allow tropical crops, as well as rice paddy triple-cropping. The acrisols of region 2 are characteristic of China proper south of the Yangzi, where triple-cropping (usually two crops of rice and one winter crop) can also be practiced in plains areas. Region 3’s cambisols are spread over the agricultural cores of warm, temperate northern China and northern subtropical central China. Three crops in two years can be harvested north of the Qinling mountains and two crops annually south of them. Good agricultural luvisols also lie across region 4 in northeastern China, although in its northernmost areas podzoluvisols best suited mainly to forestry, or secondarily to grazing, predominate. The most fertile agricultural soils are found in the pheozems–chernozems zone of region 5 in temperate, subhumid northeastern China. The most important pastoral soils, the kastanozems of region 6, comprise the western part of northeastern China into the eastern areas of the Inner Mongolian plateau. Desert–steppe zone xerosols– yermosols pervade region 7, which generally requires extensive and sustained irrigation for oasis arability. Finally, the soils of region 8, alpine cambisols–xerosols, are mainly distributed across the Qinghai–Tibet plateau and are largely suitable only for pasture. During the Song–Qing period, the core of imperial agriculture primarily lay in the eastern half of region 3 in southern China, which contained watersaturated gleysols that had to be drained before they could be cultivated, and throughout region 4. Most of the remaining arable spots scattered across regions 5–10 did not come under anything approaching comparable sustained cultivation until later in the eighteenth century, and even considerable parts of region 3 – such as the Guizhou plateau in its west-central subregion – had very limited agricultural potential because of high elevation and related to the Environment,” in Sverker Sörlin and Paul Warde (eds.), Nature’s End: History and Environment (Houndmills, Basingstoke, Palgrave Macmillan, 2009), p. 275.

268

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Region 1: ferralsols Region 2: acrisols Region 3: cambisols Region 4: luvisols

90

400 800 km 100

2

3

110

1

Guangzhou

Chengdu

Xi’an

Beijing

6

Region 5: pheozems–chernozems Region 6: kastanozems Region 7: xerosols–yermosols Region 8: alpine cambisols–xerosols

8

7

1

2

4

Shanghai

5 Harbin

Map 7.2 Soil regions of China Source: adapted from Songqiao Zhao, Physical Geography of China (New York, Wiley, 1986), pp. 66–7

0

Lhasa

Urumqi

david a. bello

factors of climate. Elsewhere in parts of region 4 – for example, the famous loess plateau in its northwestern subregion – productivity was also limited by comparatively cold and arid conditions, exacerbated by high rates of erosion, that affected soil quality as well as crop development. Such subregional variants are basic examples of the concept of “patchiness” in the environmental sciences. In simplest terms, patches are localized areas within wider landscapes that exhibit a different set of ecological dynamics from their surroundings to promote greater localized diversity. These differences inevitably influence the range of responses of humans who inhabit these patchy landscapes. Chinese agriculturalists took patchiness of soil and climate into detailed account. In the Song, Chen Fu’s 1149 Agricultural Manual (Nongshu 農書) made distinctions between dry, colder highlands and warmer, wetter lowland cultivation and gave instructions on manuring for soil management. Wang Zhen’s 1313 Agricultural Manual (Nongshu 農書) divided soil types according to relief and moisture regimes at a highly articulated range of elevations that included highland, mountain, upland, lowland, wetland, and marsh. Soil color and texture were also considered, as were field types ranging from mountain terraces to dry furrowed fields to garden fields to wet lowland and coastal polders. The Ming author Xu Guangqi compiled soil typologies from the works of various predecessors in the opening pages of his magisterial Complete Works on Agricultural Administration (Nongzheng quanshu 農政全書). The monumental 1742 work the Authorized Imperial Compendium on Agricultural Practice (Qinding shoushi tongkao 欽定授時通考), a trans-dynastic compilation of agricultural experience, emphasized that one of agriculture’s prerequisites is to ensure proper relations between a region’s soil and its corresponding agriculture. A 1741 report on Gansu province’s constrained agricultural circumstances conveys an idea of the dynamics between people and their surroundings under patchy conditions of soil quality. Governor Yuan Zhancheng had “gradually come to establish a system” in the process of “putting field taxation accounts in order.” He concluded that Gansu (extending roughly from the northwestern part of the region 4 “patch” through the southeastern part of the region 9 “patch” on Map 7.2) had “numerous” mountains and “few” fields and that “all genuinely long-term arable areas” were already being taxed. He ranked these areas as “gold, silver, copper, and iron” in accordance with their soil quality. The “gold and silver” areas were gradations of “top-quality alluvial” zones, followed by highland flatlands, with “copper and iron foothills” at the lowest ranks. In some of these “copper and iron” areas, Yuan had conducted test cultivation of “coarse grains” in 270

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

“corners of mountain peaks that are steep and narrow with much sand and gravel.” After an initial “fortuitous” harvest, these areas became “suffused with alkali” that terminated further cultivation. From Yuan’s official perspective, these regions were especially troublesome because “as soon as their taxes are fixed, problems follow.”9 By the time Yuan wrote his memorial, a combination of various environmental factors, including population pressure throughout agrarian China and soil exhaustion in many of its more marginal plots, were driving imperial subjects to mine literally every patch of soil of any quality for nutrients, an image that is only reinforced by Yuan’s metallic categorization of soil indexing. Indeed, problems of soil exhaustion were so acute in some of Yuan’s jurisdiction that he stated that people were clearing land that “was never fertile and relatively uncultivable.” The soil in these areas was “basically fragile; some can be sown annually; others every two–three years.” He also ranked this type of land as “copper and iron,” but it remained attractive to some because it was not subject to encroachment by local elites. Responding to the great potential for the unpredictable occurrence of drought, flood, and famine, Yuan urged that “copper and iron” lands never be taxed and proposed some regulation for other “empty areas of the plains originally not cultivated due to lack of water” then being opened to irrigation. Such areas were to be given six years to become viable for taxation, which would be voided if the land became saline in the process. Yuan’s metallic classification is a relatively late example of China’s pioneering concept of soil indexing, which can be dated back in written form to the text of the “Tribute of Yu” (Yugong 禹貢). This classical account included an evaluation of soil fertility, texture, moisture, and vegetation across the “nine provinces” (jiuzhou 九州) traditionally held to make up Shang China (c. 1600–1045 B C E), but was more likely composed in the Warring States period. Variations on these classical concepts remained influential throughout the imperial period. Zhongshan Park in Beijing’s Forbidden City still preserves a five-colored agricultural altar, the Altar of Land and Grain set up in 1421, whose color scheme corresponds to the general distribution of soil types in China proper. Northern China soils tend to be black from accumulations of organic carbon. Eastern China soils are often waterlogged, giving them a bluish tinge from lack of oxygen. Southern China has red soils from its more humid, warmer climate. Arid western China retains more salt in its 9

“Gansu xunfu Yuan Zhancheng zhengyue ershierri zou” 甘肃巡抚元展成正月二十二 日奏, in Ge Quansheng 葛全胜 (ed.), Qingdai zouzhe huibian: Nongye, huanjing 清代奏 折汇编：农业，环境 (Beijing: Shangwu yinshuguan, 2005), p. 47.

271

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

soils, which often appear white as a result. The wind-deposited loess plateau soils in the northwest are famously yellow and give the heavily silted Yellow River its name.

Water In some respects, similar to soil, China’s water resources were sufficiently diverse to cause corresponding regional adaptations by humans working to exploit them. The most fundamental difference lay between the relatively abundant water resources of south China and the increasingly scarce and unstable resources of north, and especially northwest, China. These conditions generally obtained throughout the period. In gross terms of “water control” (shuili 水利), there was not enough water in north China, so irrigation was its main regional concern, and too much water in south China, making drainage its main regional concern. Moreover, other than scant precipitation, what water there was in north China primarily flowed through the more than 5,000 volatile kilometers of the Yellow River and its tributaries. In the Song period, new agricultural priorities of both state and society that favored expansive poldering altered the southern landscape. North China water exploitation became increasingly complicated, a trend accelerated by a range of human and ecological factors best represented by Inner Asian conquest and catastrophic instabilities in the flow of the Yellow River. Dynastic transitions seriously disrupted central and local environmental management regimes like water control. Water control prior to the Song was centered on the Yellow River and its tributaries, which were diverted primarily to transfer their rich silt to improve the comparatively alkaline soils of north China. Environmental conditions had changed significantly with the consolidation of the dynasty, which oversaw an unprecedented expansion of agriculture even before it was driven south of the Yangzi by the Jurchen conquest of north China in 1127. Among its other effects good and bad, this expansion probably increased the amount of silt carried by the Yellow River in a dramatic and unprecedented fashion. In 1101, the Northern Song official Ren Boyu (1047–1119) disparagingly observed that the river was “half silt.” More than 450 years later, the famous Ming hydro-engineer, Pan Jixun (1521–1595), estimated that 60 percent of the Yellow River’s flow was silt in non-flood seasons and 80 percent during floods.10 10

SS, 93.2310; MS, 223.5870.

272

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

Before Pan Jixun’s intervention to resolve the silting problem, the Yellow River had not been systematically managed in a sustained fashion and main efforts were usually devoted to diversion of high waters, rather than, as Pan advocated, the reduction of silt. Consequently, the river’s instability had expanded from a relatively local problem to an empire-wide one as an unsustainable torrent of silt flowed into the adjoining Grand Canal. There is, perhaps, no more dramatic demonstration of the Chinese empire’s power to effect environmental change in response to resource constraints than the Grand Canal. By the Song, this 1,200-kilometer artificial waterway was replenishing the resource-depleted Central Plain through the resource-rich lower Yangzi by connecting the world’s third-longest river, the Yangzi, with its sixth, the Yellow River. This monumental system of resource transfer, which has been called “the world’s largest and most extensive civil engineering project prior to the Industrial Revolution,” was not, at bottom, however, powered by imperial institutions, but by water. As one 1314 Yuan report somewhat overstated it, the “shallow and uneven” Grand Canal lacked its own natural groundwater source and so could “only look up to the heavens for rain.”11 The institutions that supervised the empire’s main waterways were constrained by water in terms of its seasonal fluctuations and silt content. These chronic hydrological instabilities limited the canal’s utility and almost certainly required far more resources to maintain it than to construct it in the first place. One particularly difficult stretch, the western spurs of Mount Tai in Shandong, is illustrative. The area’s usual, sediment-laden summer floods provided water in the right quantities, but of unusable quality. This required storage of clearer water gathered in less abundant quantities during the previous winter and spring in a reservoir. Uneven water quality, in turn, required the construction both of an additional feeder canal to bring the clear water in and of a diversion canal to move silted waters into the sea. In addition, the diversion canal was isolated by a temporary dike built when grain transports were passing through, then demolished once they had passed. Finally, the feeder canal’s course was unstable and required regular intervention to keep it on track. Despite these complicated and ingenious human adaptations to the region’s hydrological constraints, it remained impossible to keep this stretch of the canal filled with water year-round.

11

Cited in Pan Qing 潘清, “Yuandai Jiang-Huai liuyu shuili jianshe shulun” 元代江淮流 域水利建设述论, Xueshu yanjiu 学术研究 12 (2014), 106–7.

273

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

That such ambitious convolutions were even possible was because Pan had conceived of and initiated an elaborate infrastructure of hydrological control designed to squeeze the Yellow River into a flow powerful enough to scour it out (subsequently known as shushui gongsha 束水攻沙 – “restrict the current to attack the silt”). By 1579, this hydro-powered system of double-pair levees was basically completed, even though critical annual maintenance and other refinements continued under the Qing. It was so effective that the Yellow River did not undergo another avulsion (a major shift from an old to a new channel) until the control system’s collapse in 1855, amid the midnineteenth-century political crises. Pan Jixun’s system certainly functioned to concentrate water as intended, but it also inadvertently concentrated silt, which was the system’s main source of disruption. Human efforts that lead to a successful overconcentration of a given resource can often end up disrupting the underlying network of extended environmental relationships that make that resource available in the first place. While the people of imperial China could not devise a practical means of promoting the right sort of climate, they certainly did find ways to enrich the fertility of soil and increase the amount of water far beyond what would have been possible if everything had been left up to their surrounding ecologies. Of course, these enhancements were ultimately intended to maximize the output of the empire’s most fundamental resource, grain, which was likewise subject to the contradictions of overconcentration. The next section will focus on agriculture, broadly defined. As the ultimate site of imperial resource production, agricultural development drove both the intensive exploitation of soil and water in the process of cultivation and the intensive pressure on woodlands and grasslands in the process of reclamation. The scale and continuity of these transformations define agriculture as the primary form of anthropogenic ecological change from Song to Qing.

Agriculture and Related Practices From a modern perspective, it is quite easy to read the environmental history of China in conventional declensionist terms that exclusively stress the unsustainable aspects of anthropogenic change over spans of time that can sometimes be numbered in mere decades. There is certainly no question that during the eight centuries covered in this chapter China experienced serious declines in biodiversity, especially in forest biospheres south of the Yangzi. However, environmental change was complex and often resilient, rather than simply a precipitous decline. 274

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

kilometers 600 300

0

600

Map 7.3 Natural distribution of Cunninghamia (“Chinese fir”) in China Source: drawn by the editor

The transformation of natural Chinese fir (Cunninghamia lanceolata) habitat, as related in an entry of the 1873 edition of the local gazetteer from Yushan district in Jiangxi, is representative. Exploitation of abundant Cunninghamia stands had begun in the early eighteenth century, and there remained considerable numbers of deer, pheasants, hares, ducks, egrets, boars, and bears until the 1790s. Subsequently, however, “bamboos and trees have been widely cleared, and human settlements grown denser. The wild beasts have not waited to be driven away, but have departed of their own accord.”12 The environmental relationship that most likely caused Yushan’s deforestation and decline in wildlife was intensive agricultural development, which was also the material basis for the ultimately disruptive expansion of the human population that primarily lived off of it. This development’s effects on flora and fauna were apparently devastating, but what replaced them was not 12

Cited in Mark Elvin, “Three Thousand Years of Unsustainable Growth: China’s Environment from Archaic Times to the Present,” East Asian History 6 (1993), 36–7.

275

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

a lifeless wasteland, but a different habitat that supported a different set of species, not just humans. The gazetteer entry goes on to elaborate how Yushan’s highland environment had been gradually altered, apparently unhindered by the soil exhaustion previously noted: “now, sweet potatoes from the West flourish, and people from Fujian, Guangdong and Guangxi come to this mountain to sow them . . . It has already been over a century, and most of the mountain’s southern face is suitable for maize and its north face suitable for sweet potato.” Yushan also produced cotton, ramie, and fruit trees. Although its entries are somewhat ambiguous on the subject, Yushan’s record of soil fertility suggests that highland cultivation was possible on a century’s timescale. The sustainable dimension of imperial agriculture must also be reckoned when accounting for ecological change and resource exploitation, which Chinese were able to continue in varying degrees of intensity for 800 years. There are even contracts extant from the late imperial period that required hillsides of southern China cleared for agriculture to be replanted with trees like Cunninghamia and Pinus massoniana (Chinese red pine). Other methods of more sustainable cultivation were of long establishment. The iconic “mulberry dike– fishpond” complex, centered in the lower Yangzi and Pearl river deltas, is one such anthropogenic ecosystem of considerable, even classical, resilience that became highly developed by the seventeenth century. While technically not a completely enclosed ecosystem that is, ideally, independent of external inputs, the mulberry dike–fishpond complex recycled exchanges between mulberry trees, silkworms, fish, and pond mud with remarkably little else added except solar energy and human labor. These exchanges are concisely summarized in the 1894 edition of the Gaoming district gazetteer, Guangdong province: “Plant mulberry on the dike and raise fish in the pond; the mulberry leaves feed the silkworms; silk worm excrement feeds the fish. This is entirely to the benefit of both, and ten times more profitable than crops.”13 The complex is actually even more interlinked than it appears in this quotation because fish excrement, along with other organic materials, enriches the pond bottom, which is dredged up periodically to fertilize the mulberry trees. In more technical terms, the mulberry tree is the first energy transfer (trophic) level in the mulberry dike–fishpond complex.14 Its leaves grow through photosynthesis and are then conveyed by human labor to the 13 14

GMXZ, 2:30a. The following discussion is based on Gongfu Zhong, “The Mulberry Dike–Fish Pond Complex: A Chinese Ecosystem of Land–Water Interaction on the Pearl River Delta,” Human Ecology 10.2 (1982), 191–202.

276

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

silkworms (Bombyx mori), which are the primary consumers at the next trophic level. The worms have a unique metabolic system that produces large amounts of cocoon protein through the efficient digestive conversion of about two-thirds of the nitrogen in mulberry leaves. What the worms do not digest feeds the secondary consumers, the fish, in the form of excreta and chrysalises. Then, aquatic organisms in the water decompose the fish excreta and any other organic materials present to enrich the pond with nitrogen, phosphorus, and potassium. This mud fertilizer is then transferred, again by human labor, back to the mulberry trees to renew the cycle. Of course, the complex is still not as simple as it appears because trees, insects, and fish all have different ecological requirements at different stages of their development that are mutually conditioning. For example, sunlight, temperature, and soil fertility together determine how many times mulberry trees can be harvested annually. Sunlight is especially critical to produce highprotein leaves. For the trees to grow at all temperatures must remain above 10 °C, but they cannot survive in drought conditions if temperatures exceed 38 °C. Any significant stress the trees experience will be transferred to the silkworms in the form of nutrient-deficient leaves. Climate also affects how many generations silkworms may produce annually – a rate of generation known as voltinism – through its influence on both silkworm and mulberry growth. In subtropical and tropical climates that allow mulberry tree leaf production year round, “polyvoltine” worms can lay eggs several times annually. In more temperate areas of the lower Yangzi worms were bivoltine, laying twice a year, once in spring and once in autumn. Some colder regional climes in provinces like Shandong still allowed “univoltine worms” to produce one clutch annually. The famous seventeenth-century work The Exploitation of the Works of Nature (Tiangong kaiwu 天工開物) generally understood silkworm voltinism in terms of an “early” univoltine type and a “late” bivoltine type, which enabled multiple harvests within a single season. Like their mulberry tree hosts, the worms are also sensitive to temperature and humidity and stop growing at ranges outside 15 °C to 32 °C. They will succumb to a drought at temperatures of 40 °C. They thrive when humidity is between 70 and 80 percent.15 15

Translation modified from Sung Ying-hsing, Chinese Technology in the Seventeenth Century: T’ien-kung K’ai-wu (trans. E-tu Zen and Shou-chuan Sun) (Mineola, NY, Dover Publications, 1966), p. 37.

277

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

Such ranges for mulberry and silkworm habitat mean that in places like the Pearl river delta, high temperature and humidity lie beyond ranges the worms can tolerate from May to September. At this time, humans intervene to shelter the worms with bamboo frames that reduce heat and allow the right amount of space for cocoons to form. Humans must also intervene to catch fish fry of several subspecies of carp to be released into the ponds when they grow to the proper size. Fish growth is likewise influenced by the temperature of the water, as well as oxygen. Fish prefer a range of 5 °C to 35 °C, and no fish in the pond can survive temperatures below 3 °C. Ideally, they also require five milligrams of oxygen per liter of water; at less than 0.2 milligrams per liter they die. Once a stable habitat within these ranges is produced, however, the fishpond will contain a number of niches for different fish species that further expand the complex. Bighead and silver carp occupy the pond’s surface, mud and black carp root around the bottom, and the grass carp inhabit the level in between. Grass carp, now the main species, feed quite economically on silkworm excreta and produce their own excrement, which in turn nourishes the phytoplankton that are the main food source of the silver carp, as well as the zooplankton that sustain the bighead carp. Whatever is not eaten is vacuumed up by the mud and black carp. These complicated interdependencies, however, can be much simplified by keeping the keystone grass carp fed. Fish need not live by excrement alone, but can subsist on the various types of leaves, including mulberry, that are introduced into the pond. Modern studies have found that a mu of mulberry can sustain a pond of carp of equal size. They have also found that silkworm excreta has a higher nitrogen content than livestock excreta. Calculations have shown that worms can convert an average mu of mulberry tree leaves into 1,200 kilograms of excreta, which then feeds 160 kilograms of carp. As intricate and absorbing as it is, the mulberry dike–fishpond complex of course was only one component of a larger agrarian complex that sustained China proper, and often beyond, from Song to Qing. Several scholars have gone so far as to argue for the unmistakable appearance of an “organic agriculture” or “ecological agriculture” by the late imperial period. Two of them, Wen Dazhong and David Pimentel, have attempted to measure this agrarian complex’s energy flows and sustainabilty in Jiaxing prefecture in Zhejiang in the seventeenth century.16 16

The following section is based on Wen Dazhong and David Pimentel, “Seventeenth Century Organic Agriculture in China: I I. Energy Flows through an Agroecosystem in Jiaxing Region,” Human Ecology 14.1 (1986), 15–28.

278

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

Wen and Pimentel found that despite the area’s reputation as a major sericulture area, mulberry cultivation took up no more than 10 percent of Jiaxing’s fields, most of which were devoted to dry-land and paddy crops. Seventy percent of its human labor was concentrated on these food and mulberry crops, with 12 percent expended on other forms of silviculture and the remaining 18 percent on animal husbandry. Recycling about 18 percent of biomass from residues of the previous crop, farmers were able to cover about 62 percent of the inputs required to produce the new crop. The remaining 38 percent came from human labor and manure, technically another recycled product. The resulting harvest roughly tripled the energy, measured in kilocalories, put into it. Once the paddy rice was harvested, then two-thirds of Jiaxing’s total arable area of 287,418 hectares was planted in winter crops, such as wheat and barley, while one-third was planted with a green manure, Chinese milk vetch. Mulberry silviculture generally took place in uplands less suitable for cereal cultivation. There are many varieties of mulberry tree, but these can be broadly divided into “wild” and “domestic” species that all belong to the same family, Moraceae. The domestic varieties were the product of grafts from wild varieties to increase leaf production, which became common practice from the Yuan onward. Although it took about three years for the trees to produce leaves that could feed silkworms, maintaining the trees took relatively minimal effort, possibly half the labor of cereal cultivation. Jiaxing seems to have been able to produce about 9,200 kilograms of leaves per hectare, which would produce 600 kilograms of cocoons, or about 50 kilograms of silk. Cocoons represented about 14 percent of the total animal product, as measured in kilocalories, that Jiaxing agriculture generated, but this does not include an estimated 6,000 kilograms of worm excrement, usable as fish feed or fertilizer. Manure was also a vital Jiaxing product. Sixty-eight percent of the 930 kilograms of manure annually produced there ended up as crop fertilizer while 32 percent was plowed back into the mulberry complex. About 19 percent of the total quantity of fertilizer came from silkworms, 34 percent was collected from humans, and the rest came from livestock, mainly from pigs, which produced about 43 percent; sheep produced well under 1 percent. Pigs, about 85 percent of Jiaxing’s animal product, were obviously critical for manure production and were likely kept mainly for this purpose since they were fed at a considerable deficit. The ratio of feed expended to pork produced ran at a deficit of about five to one, although part of the pigs’ diet was by-product from distilled grain. 279

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

“By-product” can misleadingly imply something extraneous, but little if anything integrated into Jiaxing’s agricultural complex appears as less than essential. Once 55 percent of Jiaxing’s main product of grain was eaten, 33 percent of it distilled and only 2 percent of it exported out of Jiaxing, twothirds of the estimated 4,680 kilograms of harvest crop residue was used as fuel to offset what was, arguably, the complex’s most unsustainable wasteful by-product, deforestation. Another vital resource that went largely to waste was solar radiation, which, along with human labor, was the complex’s main power source. Of an estimated 8,000 million kilocalories of solar radiation in a growing season, Jiaxing only exploited about 0.4 percent, a figure comparable with that used as recently as the 1980s. Depending on which resource is considered at what timescale, China proper’s agriculture as represented by seventeenth-century Jiaxing can appear as either sustainable or unsustainable. Taken as a complex within the timeframe of this chapter, however, it appears remarkably resilient, orchestrating for production people, plants, livestock, and even insects whose by-products are largely recycled. Some of this recycling, however, is actually a drain on the complex, particularly the burning of agricultural residue for fuel, which permanently removed nutrients and organic material from the soil. Responses to this problem would have to center on devoting more acreage to silviculture, with reductions in grain ultimately appearing as fewer pigs and their feed. Extension of silviculture at the expense of grain would have been unusual, even in highlands that were that were not very arable. Under these conditions, swiddening like that described in Sichuan mountains by the Southern Song literatus Fan Chengda (1126–1193) is more representative. Fan described clear-cutting of mountains, which were then burned off just before sowing. Burning was critical because “the mountains are mostly gravel and the soils have little strength, so there is repeated cutting and burning; only then can they be cultivated.”17 Writing centuries later and thousands of kilometers to the east, the Ming literatus Wang Shimao gave much the same description of the Fujian highlands: Mountain plots are of poor quality without manure. The farmer burns the mountain grasses and waits for the rain to arrive to wash them into the fields as manure. This is why when one enters them in the spring, each mountain is

17

“Shetian” 畬田, in FSJ, p. 217.

280

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

aflame. Seen from afar from a boat at night, the burning mountains are wondrous; encountered on foot, they are all bald and black.18

Both authors attest to the chronological continuity and geographical scope of highland swiddening that was used not simply to clear ground, but to make it arable, at least for a time, depending on whether and for how long the land is subsequently fallowed after an initial period of cultivation. Both burning and fallowing are critically needed to build up nutrients in soil that would otherwise have little, if any, fertility. Burning can even benefit micro-organisms, whose mortality is initially high. Subsequently, however, numbers may recover beyond their original population because of more favorable habitat conditions of increased temperatures, enhanced nutrients, and even, under some circumstance, increased soil moisture. Some of these effects account for frequent increased rates of soil mineralization that provide crops more nitrogen and phosphorus. Any negative effects, such as nutrient loss through burning, may also be offset by soil type, climate, and management practices. Some current research affirms that swiddening is not necessarily more degrading to soil than other forms of more intensive agriculture, depending on a range of existing human and ecological interactions in a given area. At least one region put trees at the center of its own centuries-old agricultural complex to create its distinctive forms of sustainable swiddening. The indigenous highland swiddening practiced in Yunnan before, during, and after the period under study here represents a sustainable alternative combination of grain cultivation and silviculture. Many of Yunnan’s indigenous peoples carefully conserved forests through crop rotation and forest regeneration by keeping stumps alive or replanting saplings to facilitate soil regeneration and prevent serious soil erosion. A single village stand of trees managed in this way could sustainably provide for the inhabitants for around a century without recourse to clear-cutting. Regional Han officials seem to have been generally critical of indigenous southwestern agriculture, less for reasons of subsistence than for reasons of empire. In 1726, Yunnan–Guizhou provincial administration commissioner Chang Deshou articulated the official view that the agricultural potential afforded by southwestern environmental conditions was grossly underexploited because of local passivity. Chang complained that the main problem was not a lack of fertile soil in the region, but that “60 to 70 percent of it” was under indigenous supervision. He asserted that “the local barbarians do not understand the sowing and reaping of agriculture” 18

MBS, pp. 40–1.

281

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

and unproductively believed that “only when thunder rumbles and rain meets the fields can there be plowing; if there is no rain, then the fields are abandoned.”19 Chang’s critique exemplifies some of the human tensions over resource exploitation that conditioned relations between borderlands and China proper. As Chang’s memorial nevertheless demonstrates, these tensions cannot be reduced to “steppe–sown” polarities between Inner Asian herders and Han Chinese farmers all along the north and west China ecotones (transition zones from one type of ecosystem to another) with Mongolia, Manchuria, Tibet, and Xinjiang. There were also significant differences emerging within agricultural practices that were the results of intersections between cultural and natural diversity. For officials like Chang, the main problem was the wrong human adaptive response by indigenous swiddeners (although he was also critical of some regular state officials as well) to southwestern ecological conditions of terrain and precipitation. Unlike the region’s rich copper and tin deposits, its soil minerals were not being properly mined because of the indigenous population’s passive and exclusive dependency on rain-fed agriculture. Of course, Chang’s convictions, which may stand as a summary of Chinese imperial agricultural ideology, were the historical products of centuries of Han farming experience in the easterly alluvial plains, not in westerly highlands. Even other Han streaming into the southwest unquestionably saw its potential differently from official perspectives on the superiority of intensive agriculture regardless of environmental circumstances. In the rugged mountainous terrain of southeastern Guizhou’s lower Qingshui river valley, for example, silviculture had become a primary form of regional commercial agriculture no later than the 1750s in response to the local ecology’s poor prospects for conventional agricultural development.20 Even before this point, timber – sometimes known as “Miao wood” (miaomu 苗木), in reference to the indigenous Miao population – was being cut for Chinese consumption in increasingly large quantities in the Hunan– Guizhou border region. In the wake of this exploitation, tree plantations, apparently begun by Han entrepreneurs and emulated by their Miao counterparts, arose on lands that were not suitable for cereal cultivation. The 19

20

Chang Deshou 常德壽, “Chenzhuo zhi bufa tusi guanjian zhe” 陳酌治不法土司管見 摺, in YZZP, vol. 7, p. 776. Meng Zhang, “Financing Market-Oriented Reforestation: Securitization of Timberlands and Shareholding Practices in Southwest China, 1750–1900,” Late Imperial China 38.2 (2017), 109–51.

282

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

overwhelmingly preferred cultivar was Cunninghamia lanceolata, which was used for general purposes ranging from coffins to roofing. A silviculture crop of southern China from no later than the Tang dynasty, Cunninghamia cuttings from either new branches or stumps were relatively easy to grow and could, unlike most other cultivated trees, be intercropped with grains. Agricultural manuals routinely discuss how to farm Cunninghamia from the Yuan period on. Its utility derived in large measure from its fifty-meter height, the lower 40 percent of which was usually free of branches, and its firm, straight trunk up to two meters in circumference. Cunninghamia’s main problem from the farming perspective was that these attractive natural attributes took about thirty years to mature. This ecological barrier was overcome by the economic means of securitizing the trees into independently tradable shares. This is an example of how human institutions, here the elaborate contractual relations to diversify investments in different Cunninghamia plots at various stages of growth over their twenty- to thirtyyear period of development, were substantially conditioned by ecological constraints. Eighteenth-century Guizhou silviculture resembled its Huizhou prefecture predecessor in southern Anhui. Huizhou’s inhabitants sustainably managed their woodlands to remain a major supplier of lumber to the urbanizing lower Yangzi delta from the Song through the Ming. Regional silviculture helped to build the Southern Song capital, Hangzhou, and sustain Huizhou’s economy. By the late twelfth century, it was already recorded that the Huizhou county of Xiuning made its living off Cunninghamia rather than rice farming; by the sixteenth century, in Qimen county, “where the mountains are many and the fields few,” it was said that “the people’s daily expenses all rely on mountain trees.” By the latter half of the fifteenth century, when Jiangnan urban areas began an extended period of growth, commercial practices that would later be adopted in eighteenth-century Guizhou became widespread in Huizhou. Saplings were carefully spaced, then weeded until they reached a height of about three to four Chinese feet (chi). Summer millet and winter wheat were planted among the growing trees where possible and some mixed stands, including fruit trees and bamboo, were mentioned as well. This pattern of multi-cropping was part of an overall diversification strategy to deal with environmental uncertainty, which could include arson and theft as well as flood and wildfire.21 21

All references to Huizhou are from Joseph P. McDermott, The Making of a New Rural Order in South China, vol. 1, Village, Land, and Lineage in Hiuizhou, 900–1600 (Cambridge, Cambridge University Press, 2013), pp. 369–72, 395–6.

283

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

Cultivation choices were determined by a number of ecological as well as cultural factors. Length of cultivar growth period was a fundamental issue, as recognized by one Ming-period Qimen resident who advised that “if you have a one-year plan, plant grain. But if you have a plan for several years, plant trees – this is really the correct way to make a living.” Human choice was likewise constrained by terrain, which is why forestry has been called “the dominant form of land use on steep mountain slopes,” where Cunninghamia plantations “proved to be a more valuable and sustainable form of land use than agriculture.”22 In many parts of southern China, environmental conditions seem to have promoted not simply agriculture, but agro-forestry. It is possible to intercrop Cunninghamia stands not only with wheat and millet, but also with Old World vegetables and New World ones like maize and sweet potatoes, depending on the tree canopy development and crop shade tolerance. Joseph McDermott’s formulation that “we are dealing here less with the impact of Huizhou’s particular terrain and customs . . . than with forestry practices and a commercial logic” might be revised as an “environmental logic” that is not exclusively dependent on human agency, but instead on human adaptation to ecological constraints in Anhui and Guizhou.23 Cunninghamia’s botanical attributes, most especially a habitat range effectively restricted to south China and comparatively fast growth rates, were instrumental for the regional diversification of human agriculture, both Han and indigenous, to encompass a viable silviculture. The same tribute could be paid both to New World crops, which became comparatively widespread in China by the sixteenth century, and to the Old World variants of early-ripening and drought-resistant Champa rice that had begun to establish themselves by the eleventh century. One of Song agriculture’s major and most enduring agricultural achievements – reclamation of Yangzi delta marshlands – was enabled by Champa rice, whose poldered cultivation nevertheless effected an ecological transformation of the region that made it more vulnerable to flooding. Like Cunninghamia and many other farmed trees, all such crops, developed over millennia by humans from various wild grass and tuber species, dramatically reduced China’s chief resource constraint, arable land, because they could be grown on hillsides in relatively saline soils and on other types of “wasteland” where ecological 22

23

Cited in Christian Daniels and Nicholas K. Menzies, Science & Civilization in China, vol. 6, part 3, Agro-industry and Forestry (Cambridge, Cambridge University Press, 1996), pp. 629, 666n. McDermott, The Making of a New Rural Order in South China, vol. 1, p. 372.

284

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

conditions did not fall within the parameters of the traditional Chinese crop suite. Such transformations, however, took a long time to spread throughout the whole of the empire and were often accompanied by serious interlinked environmental problems like erosion and flooding. Even in the mid-eighteenth century, a venerable agricultural locale like Anhui was still able to expand its cultivation space through the import of new strains from other provinces. In 1740, Anhui governor Chen Dashou reported that all the fertile land and flatlands in his jurisdiction had been planted in rice and “various grains.” Land on “high hills and steep slopes” alone remained unplowed because these areas were not arable for purposes of conventional grain cultivation. Upon observing extensive patches of overgrowth during a survey of a river earlier in the year, Chen decided to have local peasants work these unpromising areas. The effort proved fruitless until Chen heard about a dry-paddy early-ripening drought-resistant rice from Fujian called shesu 畬粟 that had been brought to Jiangnan by former Liangjiang governorgeneral Hao Yulin. The name of the rice may possibly refer to “grain” (su) raised in fields “swiddened” (she) by the eponymous She people indigenous to highland areas of Fujian and Zhejiang. Chen described shesu’s affinity for “the high and dry” locales “requiring no irrigation.” Test cultivation in Anhui had proved effective, and Chen hoped that soon “like silviculture, [seshu’s] cultivation, and so its benefit, will daily increase over a few years’ time and steadily reach other provinces, until every unsown highland and slope can be converted into usefulness.”24 Over time, utility became a relative concept where some field-expanding cultivars were concerned. Maize, the earliest recorded reference to which dates to a 1511 gazetteer also from Anhui, for example, uses water and sunlight very efficiently. Its nutritional energy in terms of kilocalories is equivalent to that of polished rice and it is actually higher in protein. These characteristics, in combination with the relatively high tolerance for cold of its highland varieties, make maize in many respects an ideal hillside crop, which explains its spread across late imperial China in the hands of poor, land-hungry migrants, known as “shack people.” In a 1772 report, the governor of Hubei, Chen Huizu, affirmed that “deep in the steep peaks of the mountains west of Yichang, many people reside scattered throughout the crags and cliffs and plant maize for a living.” The area’s relatively sparse population had drawn migrants from the adjoining provinces of Sichuan and Hunan, who 24

“Anhui xunfu Chen Dashou jiuyue ershisiri” 安徽巡撫陳大受九月二十四日, in Ge, Qingdai zouzhe, pp. 43–4.

285

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

also found “the mountain soil most suitable for maize.” It was easy to plant and harvest in abundance “without concern for drought or flood . . . which is why they hasten here from far and near.” All mountaintop space had thus been cleared for maize “without any remaining,” and livelihoods were “quite prosperous.”25 Growing human dependency on maize was reciprocal. Long before it reached China, the crop had become so domesticated that its seeds could not be naturally separated from the cob, but required people to disseminate them. Altered by human hands over thousands of years, maize eventually lost the ability to propagate itself, but it was also spread much farther beyond its original Mesoamerican homeland by those same hands than would likely have been possible if left to itself. Within the first decades of the nineteenth century, maize even rivaled rice as the primary staple in at least one province, Guizhou. Guizhou around the same time also began to experience the distinctive sort of desertification to which its limestone karst highlands were prone. The potential for rocky desertification is rooted in the carbonate rock that composes karst formations that extend across south China. Silicate-poor carbonate rocks are unusually water-soluble and more vulnerable to weathering without producing much soil. Rock that is easily dissolved has extreme difficulty maintaining stable water levels and ground storage capacity, so that even with 1,000 millimeters to 2,000 millimeters of rain annually, Guizhou’s soil’s water storage capacity is critically deficient for vegetation. Karst soil also often lacks a layer, the C-horizon, that secures higher layers to the bedrock. Overall soil coverage, in consequence, is thin and fragile. Indeed, karst soil is naturally irreplaceable at human timescales. There are estimates that karst soil formation rates were roughly a hundred times less than those from non-carbonate rock in the Sichuan basin and that it would take 250,000 to 750,000 years for one meter of soil to form through the dissolution of twenty-five meters of Guizhou karst, or about one centimeter every 4,000 to 8,500 years. The region’s steep-slope topography naturally works to promote karstification and erosion, which casual cultivation would only enhance. Karst’s erosive inclinations are largely inhibited by thick soil and vegetation, the sort of coverage that would be eliminated by intensive swidden cultivation of crops like maize; nor were the deleterious effects of rocky desertification confined to karst areas. It has been estimated that 25

“Hubei xunfu Chen Huizu sanyue shierri” 湖北巡撫陳輝祖三月十二日, in Ge, Qingdai zouzhe, p. 247.

286

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

southwest China sends about five times more eroded material into the Pearl and Yangzi rivers than east China does.26 Even under more stable cultivation conditions on non-carbonate rock, the spacing required for proper plant development may leave large areas of soil exposed to erosion throughout the growing season, exacerbating such tendencies already inherent in highlands. Such dynamics, both ecological and historical, have led one study to conclude, on the basis of both Qing and Republican records, that there is a significant correlation between rocky desertification in Guizhou – still expanding today – and late imperial highland maize cultivation.27 Qing authorities across a number of south China provinces certainly linked serious erosion to unbridled highland cultivation in general and to maize in particular. Maize cultivation was officially banned in some Zhejiang and Anhui locales. An 1807 edict which approved a general expulsion of shack people from the Huizhou prefecture highlands allowed those who had already established productive fields to remain on the condition that they switch from maize cultivation to planting tea, fir, and pine to “supplement firewood while avoiding collapse and runoff.”28 The late imperial record and modern scientific studies on maize cultivation demonstrate that ecological change, effected by New World crops in this instance, is highly variable at different timescales. No apparent indication in the historical record has come to light regarding any serious ecological problems caused by maize cultivation during the first three centuries after its introduction into China in the early sixteenth century. With the onset of the nineteenth century, however, problems proliferated. Maize, a crop that originally expanded arable land in Guizhou to rival rice as the main staple, ended up reducing arable land not only on mountains through mechanisms like stony desertification, but also through excessive siltage dumped on lowland fields. As an 1877 gazetteer from Guizhou’s Zheng’an prefecture put it: “Once developed, the mountain plots are still barren gravel. They promptly collapse during a long rain and revert to wasteland as before. The fields even become piles of sand that are then washed away to the stony bone.”29

26

27

28

Zhongcheng Jiang, Yanqing Lian, and Xiaoqun Qin, “Rocky Desertification in Southwest China: Impacts, Causes, and Restoration,” Earth-Science Reviews 132 (2014), 2, 5–7. Yan Qiyan 严奇岩, “Qingdai yumi de yinjin yu tuiguang dui Guizhou shimohua de yingxiang” 清代玉米的引进与推广对贵州石漠化的影响, Guizhou shifan daxue xuebao (shehui kexue ban) 贵州师范大学学报 (社会科学板) 164.3 (2010), 48–53. TZQJ, 4, pp. 226–8. 29 ZAZZ, 5.6b.

287

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

The erosion problems created by New World crops like maize were by no means inevitable, as demonstrated by traditional Mesoamerican cultivation of maize in combination with beans and squash – traditionally called “the three sisters.” The taller maize plants support bean trailers while the bean roots fix nitrogen to the soil, over which the squash leaves grow to provide ground cover and retain moisture. Even rice could have been more sustainably accommodated on less hospitable southwestern Chinese ground if the imperial state and society had found indigenous glutinous varieties – which were better adapted to the wide fluctuations in temperature and rainfall characteristic of the region’s highlands and easier to cultivate without largescale deforestation – more to their tastes. Chinese consumers, however, generally preferred to eat non-glutinous varieties, and the Chinese state found these varieties easier to process, transport, and store. By the early eighteenth century, the state ranked the three southwest Chinese provinces of Sichuan, Yunnan, and Guizhou among the five chronically deficit-ridden provinces where administrative expenditures exceeded revenues, thus requiring transfer payments from the nine surplus provinces that were nearly all located on the lower or middle Yangzi. There was, consequently, constant pressure on authorities in the southwestern provinces to raise their revenues at least to the level of self-sufficiency at which Fujian, Guangdong, and Guangxi were ranked. Aside from basic subsistence requirements, fiscal ranking may help to explain why provincial authorities in revenue-surplus provinces like Anhui, with established agricultural infrastructure, wanted to expel migrant shack people cultivators from local hills, while authorities in revenue-deficit provinces like Guizhou, with comparatively few lowland fields at risk from highland erosion, actively welcomed them. Shack people, as a problem or as a solution, would probably not have existed at all without New World crops. For reasons of both nature and culture, Chinese agriculture overall, like maize in particular, steadily became committed to a more limited range of possible adaptations over time. Advantages gained were substantial and measurable in the centuries of agricultural development and population increase that were secular trends throughout the period covered here. The mechanisms used to establish, maintain, and – most importantly – increase these advantages incurred substantial environmental costs, however, measurable both in net reductions of biodiversity and in systemic disturbances, which could easily develop into disasters through a synergy of natural and anthropogenic forces. Costs could also run into deficits that were not – and could not be in cases like that of rocky desertification – fully made good. By 288

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

1000 C E, state and society had inherited a complex system of not simply agricultural but also environmental relations because of farming’s existential interdependency with the ecologies in which it was embedded. By 1800 C E, environmental complexity, in terms of change and constraints, seems to have increased, along with population and arable acreage, to approach unsustainable levels. The mutual trajectories of these trends may be considered through an examination of disturbance.

Disturbance/Disasters Within the enormous complexity of terrestrial environmental dynamics, the term “disaster,” like the implicitly related terms “stability” and “steady state,” is relative and insufficiently nuanced. In historical contexts, disasters tend to be defined exclusively by their serious, large-scale, and infrequent effects on humans. In more environmental terms, however, it may be more precise to speak of varying scales of “disturbance” that are endemic to ecosystems, of which humans are certainly, but not exclusively, members. Thus disturbances are as naturally inevitable as the droughts, floods, and locust infestations (which were not disastrous from the insects’ perspective) that the Ming agricultural encyclopedist Xu Guangqi called the three causes of “serious famine.”30 To the extent that these scourges were intrinsic to the natural disturbance regime, human action was only one additional factor, albeit a critical one, in bringing about changes that could disrupt existing environmental relationships. Furthermore, and regardless of historical period, the degree of disruptive alteration remained fundamentally dependent on ecosystemic and institutional “inertia” and “resilience,” technical terms that respectively refer to their capacities to resist and recover from disturbance. Much state effort was devoted to enhancing the inertia and resilience of mainly agrarian ecosystems and institutions against often intertwined natural and anthropogenic disturbances to maintain the existing environmental equilibrium or “steady state” that was the material basis of dynastic power. Steady-state strategies, which have been globally inherent, even definitive, in the maintenance of state power up to the present, tend to solve resource problems in the short term, like declining yields, but success in controlling one variable, that often fluctuates, leads to changes in variables that operate at other temporal and spatial scales, like nutrients or 30

NZQS, 3, p. 1299.

289

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

food web dynamics. Such management creates landscapes that become spatially homogenized and vulnerable to disturbances that previously could be absorbed.31

One of the main disturbing consequences of imperial China’s agricultural success between 1000 and 1800 C E was the relative homogenization of what had been a much more diverse landscape, especially south of the Yangzi, with a commensurate, steady reduction in diversity-dependent resilience. During this period, a greater expanse of acreage was put under the plow and much more food produced for a growing population, but this outstanding preindustrial progress was accompanied by commensurate rates of topsoil degradation, deforestation, desertification, and reductions in biodiversity. One representative example of these costs of imperial agricultural success can be seen in disturbances that contracted the tiger habitat during the late imperial period. Prior to around the mid-sixteenth century, tigers seem to have been relatively plentiful; records suggest, for example, that attacks on humans by the south China tiger (Panthera tigris amoyensis) remained relatively common in Fujian, Guangdong, and Jiangxi since the Song, and archaeological evidence shows tiger residence in highlands south of the Yangzi no later than the Han dynasty. Toward the mid-Ming, however, significant declines in tiger subspecies and habitats in southern and northwestern China as well as southern Manchuria seem to have begun, possibly linked to increases in human population, arable land, and deforestation. This general decline could be reversed during times of upheaval that depopulated places like coastal Fujian during the Ming–Qing transition. As the region was reoccupied, however, human–tiger encounters increased to the point where “the menace of tigers frequently emerged everywhere . . . with tigers even daring to come into town to devour people.”32 The more northerly and sparsely populated reaches of Manchuria remained effectively undisturbed. This is at least partly because intensive cereal cultivation as practiced in China proper was climatically impossible in these much colder and more arid northern reaches, which preserved habitat for both predators like the Amur, or “Siberian,” tiger (Panthera tigris altaica) and prey, especially deer. The vast and restricted imperial hunting reserves north of the Great Wall, of which Muran in Chengde was the most 31

32

Carl Folke, “Resilience: The Emergence of a Perspective for Social–Ecological Systems Analyses,” Global Environmental Change 16 (2006), 255–6. Luke Hambleton, “An Ambush of Tigers: A Socio-ecological History of the Ming–Qing Fujian Tiger Menace,” in Jeff Kyong-McClain and Yongtao Du (eds.), Chinese History in Geographical Perspective (Plymouth, Lexington Books, 2013), p. 115.

290

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

important, generally maintained rich habitat for wildlife much longer than elsewhere in China proper. Here, the multiethnic pastoral and foraging diversity of Inner Asian conquest dynasties like the Qing was critical for the (somewhat unintentional) preservation of biodiversity. Certainly, the Qing state never sponsored tiger eradication operations in these areas as it did in Fujian and in southern Shaanxi, where it was recorded that “during the Yongzheng (1722–1735) reign, the walls of the district seat of Lüeyang tumbled down, tigers then frequently entered the town, and residents did not dare go out of doors.” Things got so bad in Shaanxi’s Xixiang county that the magistrate hired several dozen archer-hunters to go into the tigers’ lairs, and they shot sixty-four dead between 1713 and 1715.33 Tigers and grain fields had great difficulty coexisting in southern China, but tigers and forests or grasslands seem to have been reasonably compatible in Inner Asia. This is probably part of the reason why there are still about 540 Amur tigers left in the wild, while the last confirmed sighting of a freeroaming south China tiger was in the 1970s. Of course, the eradication of tigers may be a disaster only from the human perspective in a regional and historical sense. For southern Chinese, a proliferation of tigers would have been a disaster; for their Inner Asian overlords, the disaster would have been the total eradication of tigers, which were being imported southward from Manchuria into Beijing hunting reserves for banner military training until 1822. Despite undeniable similarities across substantial expanses of China proper and Inner Asia, ecological patchiness, along with subspecies and ethnic diversity, should qualify overgeneralizations or anachronisms about “tigers in China” over the eight centuries examined here. Other forms of disturbance, however, were more universally pervasive and so more conventional, if still subject to regional variations. Flood and drought were the most persistent and catastrophic disturbances between 1000 and 1800 C E that forged in response institutional and social continuities which generally enhanced agricultural inertia and resilience. Extant sources, however, do not exhibit the same degree of continuity, making it difficult to discern a complete pattern of disturbance for the period. Figure 7.1 covers the Song through the Ming, for which sources are generally limited, especially by a paucity of local gazetteers before 1470. From 1470 on, the relative abundance of local gazetteers greatly augments the data points, 33

Cao Zhihong 曹志红 and Wang Xiaoxia 王晓霞, “Ming Qing Shaannan yimin kaifa zhuangtai xia de ren hu chongtu” 明清陕南移民开发状态下的人虎冲突, Shilin 史林 5 (2008), 55–6.

291

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello 70 60 50 40 30 20 10 10 00 – 10 19 40 – 10 59 80 – 11 99 20 – 11 39 60 – 12 79 00 – 12 19 40 – 12 59 80 – 13 99 20 – 13 39 60 – 14 79 00 – 14 19 40 – 14 59 70 – 15 89 10 – 15 29 50 – 16 79 00 – 16 19 40 –4 9

0

Floods

Droughts

Figure 7.1 Recorded occurrence of flood and drought at twenty-year intervals, 1000–1649 Data do not necessarily include all flood and drought events recorded for all dynastic periods within the time frame, but only those disasters at scales rated significant by the compilers from whose works table data have been adapted. Sources: 1000–1469 data adapted from Zhang Binglun 张秉伦 et al., Zhongguo gudai ziran zaiyi zhengtixing yanjiu 中国古代自然灾异整体性研究 (Hefei, Anhui jiaoyu chubanshe, 2002), pp. 120, 176; 1470–1649 data adapted from Gao Wenxue 高文学 (ed.), Zhongguo ziran zaihai shi 中国自然灾害史 (Beijing, Dizhen chubanshe, 1997), p. 252

which number no more than around 1,000 for the entire historical record before 1470, but several tens of thousands from 1470 to 1912. This transformation includes an extension of imperial- or provincial-level data down to district-level data. District-level data points also enable greater refinement in the scaling of flood and drought in terms of degree of severity, geographical extent, and so on. Some sense of the disparity created by the preservation of local gazetteers may be gained from a comparison with Figure 7.2, which provides flood and drought statistics for the Qing period up to 1800. Except for a forty-year period from 1684 to 1723, any twenty-year interval of Qing floods appears to exceed that of the total floods recorded for the Song through Ming in Figure 7.1. Moreover, every twenty-year interval for drought during the Qing apparently exceeds the corresponding Figure 7.1 total by many orders of magnitude. The enormous divergences in sources, as well as different disaster-scaling systems of data analysis, make meaningful comparisons between the late imperial period and the preceding Song–Yuan era problematic. Nevertheless, at least one striking trend is immediately apparent within Figure 7.1, which exhibits a pronounced and sustained peak in flooding over a 120-year period from 1280 to 1399. This substantial interval encompasses

292

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints 3000 2500 2000 1500 1000 500

Flood

0 80

3

–1 84 17

64

–8

3 17

17

44

–6

3 17

24

–2 04 17

–1 84 16

–4

3

3 70

3 –8 64 16

16

44

–6

3

0

Drought

Figure 7.2 Recorded occurrence of flood and drought at twenty-year intervals, 1644–1800 Source: adapted from Li Xiangjun 李向军, Qingdai huangzheng yanjiu 清代荒政研究 (Beijing, Zhongguo nongye chubanshe, 1995), pp. 113–93

both the Song–Yuan and Yuan–Ming dynastic transitions to effectively cover the entire tenure of the Yuan dynasty. It is often difficult to separate anthropogenic from ecogenic causes of disturbance and disaster, mainly because both can operate synergistically. However, the admittedly far from complete statistics in Figure 7.1 suggest that the Yuan state faced distinctively persistent difficulties, not necessarily entirely of its own making, in water control at the ecological transition point between the Medieval Warming Period and the Little Ice Age. A more legible perspective on flood and drought patterns can be discerned on a regional scale for the Yangzi delta over the past millennium in Figure 7.3. Here, the onset and development of the Little Ice Age seem to have had a marked effect on the frequencies of both flood and drought from roughly 1500 to 1700. The third dynastic transition of the 1000–1800 time period, that of the Ming–Qing, occurs within this period. Figure 7.3 shows that frequencies of both flood and drought were somewhat extraordinary, in the context of the Qing as a whole, from the time of the Manchu conquest in 1644 until its consolidation in the 1680s, which may suggest that the social chaos of the Ming–Qing transition had some pronounced effect on the frequency of these types of disaster.

293

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

Figure 7.3 Reconstructed temperature and flood/drought variability in the Yangzi delta, 1000–2000 C E. Flood and drought index: 2: heavy flood; 1: flood; 0: normal; –1: drought; –2: heavy drought Source: reprinted from Quaternary International 176–7 (2008); Qiang Zhang et al., “Climate Changes and Flood/Drought Risk in the Yangtze Delta, China, during the Past Millenium,” 62–9, copyright 2006, with permission from Elsevier and INQUA

One pattern that was fairly consistent was the problem of Yellow River management, a chronic challenge for all dynastic regimes. Nevertheless, human interventions to control the river were comparatively limited before the Song, which expanded operations under the pressure of military exigency rather than agrarian production. Consequently, major changes in the river’s course were dominated by climate until the Song, and increasingly subject to human interference thereafter. In the transitional period between these two secular trends, a combination of ecological and anthropogenic circumstances, including persistent, heavy rains and Song tactical breaching of local riverbanks, transformed the strategic frontier zone of Hebei into a swampy barrier against the cavalry of Inner Asian enemies to the north. By the 1030s, much of the area was inundated,

294

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

and great efforts were required to maintain about 260 kilometers of water obstacles. In 1048, a sudden northeastern shift of over 700 kilometers left “about twenty percent of Hebei’s population dead in its wake.” Over the next eighty years, as the river continued to move north and east through Hebei until it shifted out of the province in 1128, the Song state became locked into a “hydraulic mode of consumption” that wasted political capital, labor, and other resources in a futile attempt to maintain its river barrier while restoring the region’s economy.34 Indeed, the fact that the succeeding Jin state inherited the Song’s hydraulic mode of consumption with little alteration in terms of policy and practice furthermore suggests an ecological continuity that may transcend social change and even shape it, without absolutely determining it. Subsequently, the Yuan dynasty, following decades of conflict, certainly did move beyond the mere restoration of the empire’s water control infrastructure to expand it through major undertakings like the shortening of the Grand Canal and substantial irrigation projects. Yellow River floods, course shifts, sandification, and salinization nevertheless persisted as serious problems over which Yuan officials deliberated, but ultimately failed to formulate a practical course of action.35 This result was not too dissimilar from the Song experience. In contrast, the late imperial states were able to break this pattern, at least before the mid-nineteenth century. Instability in Yellow River hydraulics between 1128 and 1855 was mainly caused by siltage, which was probably exacerbated by human intervention in the river’s natural dynamics. Prior to 1565, this was handled mainly by the diversion of flood waters into the Huai river. Pan Jixun had, as previously noted, stabilized the river, but in the process of his system’s suppression of natural drastic shifts in course (or “avulsions”) and silting, the riverbed became “super-elevated” to about ten meters above its natural level, rendering it prone to more frequent levee breaches downstream. Moreover, if breaches were not quickly repaired before the next flood, a feedback loop was created in which a breach increased siltation, due to the drop in scouring water pressure, which would then trigger more breaches upstream. This dynamic was confirmed in 1689 by one of Pan’s successors, Yellow River commissioner Jin Fu: “when the river breaches in a given upper reach, the channel downstream would certainly silt up. When the channel downstream 34

35

Ling Zhang, The River, the Plain, and the State: An Environmental Drama in Northern Song China, 1048–1128 (Cambridge, Cambridge University Press, 2016), pp. 5–6, 69–73, 282–4. Zhao Yanfeng 赵彦风, Yuandai huanjing shi zhuanti yanjiu 元代环境史专题研究 (Ph.D. diss., Shaanxi Normal University, 2016), Chapters 2, 3.

295

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

was silted up, the channel upstream would certainly be breached. This has been a regular pattern.”36 The real problem was that maintenance was not simply a matter of repairing breaches but of preventing them in order to avoid triggering the feedback loop. Subsequent costs to deal with the consequences of Pan’s solution began to outweigh the benefits from riverine transport during the Qing. Water control efforts to promote agriculture were intended to curtail two of the three main causes of famine, namely flood and drought, but could simultaneously sustain the third, locusts. A particular alternation of flood and drought may work synergistically to increase locust breeding habitat by leaving the ground just damp enough for ideal reproduction that can otherwise be disrupted by too much or too little water. Human policies to preserve wetlands as overflow catchments or as sources of reeds for dike construction inadvertently protected locust breeding habitats. Agricultural “niches” afforded unnaturally high concentrations of nutritious grasses and encouraged the swarming behavior that turns some species of grasshopper, including China’s main scourge, Locusta migratoria manilensis, into locusts. This transformation, technically termed density-determined phase polymorphism, occurs as normally solitary grasshoppers come into unusually close contact that releases the neurotransmitter serotonin which not only encourages swarming, but also causes locusts to sprout wings. Imperial agriculture even recognized density-determined phase polymorphism to the extent that peasants were exhorted to find and eradicate locusts at their more vulnerable preflight stages as eggs or juveniles. Chen Chongdi’s 1874 Manual of Locust Control warned that “to control [winged] locusts is not as easy as controlling immature locusts, which, in turn, is not as easy as controlling the eggs. Likewise, controlling them after a drought has already developed is not as easy as controlling them when floods recede.”37 Complex interplay between the three main agricultural disasters of flood, drought, and locusts could produce unintended outcomes like the expansion of locust habitat and the very encouragement of locust swarms even as people worked to maximize the output of their fields through the manipulation of ecological systems intended to reduce the disastrous effects of all three. There is probably no better demonstration of the equivocal nature of

36

37

Cited in Yunzhen Chen, James P.M. Syvitski, Shu Gao, Irina Overeem, and Albert J. Kettner, “Socio-economic Impacts on Flooding: A 4000-Year History of the Yellow River, China,” AMBIO 41 (2012), 693. ZHS, 2a–b.

296

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints

preindustrial environmental relations between culture and nature that inadvertently created changes and constraints.

Conclusion The ecological changes and constraints over the eight centuries discussed in this chapter were products of relationships between humans and their surroundings whose dynamics are not completely understood even today. Under preindustrial conditions, it may only become apparent with the passing of years or of centuries, or of warm periods and little ice ages, that practices such as highland maize cultivation or Yellow River avulsion control have become unsustainable. Nevertheless, the long-term trend of imperial Chinese resource exploitation underlying all environmental relations may be understood as inexorably moving “towards maximal ‘arablization,’ that is . . . cereal cultivation, as opposed to rearing stock . . . or reliance on non-farm sources of food from forests and wetlands.”38 This protracted, fundamental commitment to specific forms of agriculture at intensive scales similarly constrained all dynasties – Han and Inner Asian – in terms of basic resources technologically obtainable from climate, soil, and groundwater. Of these three, soil was the resource base both most easily depleted and replenished, within limits that seem to have been very nearly exceeded in many areas around the turn of the nineteenth century. China’s imperial state and society could not control the temperature or make more water, but they could, and did, deploy a formidable range of techniques to maintain soil quality. Yet, they could also indulge an array of erosive inclinations that degraded soils, many of which were quite fragile. That the exploitation of such marginal prospects was even possible is largely due to the fortuitous arrival of higher-yielding upland crops from the New World, which certainly expanded the otherwise rigidly fixed resource of farmland. The agricultural horizons of territorial expansion were limited. The vast majority of this new territory, with some notable exceptions in southern Manchuria, southwestern China, and Taiwan, was arid with poor soils and scarce groundwater that left it more suited for pastoral exploitation or less intensive forms of agriculture. It is an indication of the overall adaptability of Chinese environmental relations from 1000 to 1800 C E that the population roughly quadrupled to between 300 and 400 million, although biodiversity, especially in China 38

Elvin, “Three Thousand Years of Unsustainable Growth,” 10.

297

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

david a. bello

proper, was almost certainly stressed to the point where, in Robert Marks’s words, the “Chinese empire had reached its ecological limits.” Signs that these limits were being exceeded were visible to many individuals at different times in different places, but no viable, systematic solutions were forthcoming. As expressed by Hong Liangji, “China’s Malthus,” in the late eighteenth century, living space and farmland “have only doubled, or at most, increased three to five times, while the population has grown ten to twenty times . . . Nature’s way of making adjustments” to this disparity lay in “flood, drought and plagues,” which would reduce excess population. Hong concluded that during an extended span of peaceful rule like the eighteenth century, “Nature cannot stop the people from reproducing. Yet the resources with which Nature nourishes the people are finite.”39 Enormous population growth has long been the primary explanation for both changes and constraints in China’s ecology. Yet, from an environmental perspective, it is perhaps more precise to say that synergy between people and their chosen plant and animal domesticates – supported by adequate water within the right temperature ranges to fructify sufficient expanses of the right kind of soil – “successfully” generated enough change to overload China’s preindustrial carrying capacity by the end of the eighteenth century.

Further Reading Bao, Maohong, “Environmental History in China,” Environment and History 10.4 (2004), 475–99. Bello, David A., “Environmental Issues in Pre-modern China,” in Oxford Bibliographies in Chinese Studies (Oxford, Oxford University Press, 2014), https://www.oxfordbibliogra phies.com/view/document/obo-9780199920082/obo-9780199920082-0087.xml? rskey=RJOZ1s&result=46. Bray, Francesca, Science and Civilisation in China, vol. 6, part 2, Agriculture (London, Cambridge University Press, 1984). Cui, Jianxin, Hong Chang, Kaiyue Cheng, and George S. Burr, “Climate Change, Desertification, and Societal Responses along the Mu Us Desert Margin during the Ming Dynasty,” Weather, Climate and Society 9.1 (2017), 81–94. Daniels, Christian, and Nicholas K. Menzies, Science and Civilisation in China, vol. 6, part 3, Agro-industries and Forestry (London, Cambridge University Press, 1996). Di Yu 翟禹, “Yuandai huozai shi yanjiu (yi) (er)” 元代灾害史研究述评 (一), (二), Chifeng xueyuan xuebao (Hanwen zhexue shehui kexue ban) 赤峰学院学报(汉文哲学社会科学 版) 7 (2018), 10–15; 8 (2018), 32–38

39

Cited in Robert B. Marks, China: Its Environment and History (New York, Rowman and Littlefield, 2012), pp. 222, 221.

298

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

Ecological Change and Resource Constraints Elvin, Mark, The Retreat of the Elephants: An Environmental History of China (New Haven, Yale University Press, 2004). Elvin, Mark, and Liu Ts’ui-jung (eds.), Sediments of Time: Environment and Society in Chinese History (Cambridge, Cambridge University Press, 1998). Elvin, Mark, Hiraoka Nishioka, Keiko Tamura, and Joan Kwek (eds.), Japanese Studies on the History of Water Control in China: A Selected Bibliography (Canberra, The Institute of Advanced Studies, Australian National University, 1994). Gao Wenxue 高文学 (ed.), Zhongguo ziran zaihai shi 中国自然灾害史 (Beijing, Dizhen chubanshe, 1997). Guo Tao 郭涛, Zhongguo gudai shuili kexue jishu shi 水利科学技术史 (Beijing, Zhongguo jianzhu gongye chubanshe, 2012). Marks, Robert B., China: Its Environment and History (New York, Rowman and Littlefield, 2012). Miller, Ian Matthew, Fir and Empire: The Transformation of Forests in Early Modern China (Seattle, University of Washington Press, 2020). Mostern, Ruth, “Sediment and State in Imperial China: The Yellow River Watershed as an Earth System and a World System,” Nature and Culture 11.2 (2016), 121–47. Muramatsu Ko¯ichi 村松弘, Chu¯goku kodai kankyo¯ shi no kenkyu¯ 中国古代環境史の研究 (Tokyo, Kyu¯ko Shoin, 2016). Sterckx, Roel, Martina Siebert, and Dagmar Schäfer (eds.), Animals through Chinese History: Earliest Times to 1911 (Cambridge, Cambridge University Press, 2019). Wang Yude 王玉德 and Zhang Quanming 张全明, Zhonghua wuqiannian shengtai wenhua 中华五千年生态文化, 2 vols. (Wuchang, Huazhong shifan daxue chubanshe, 1999). Xiao Ruiling 肖瑞玲, Cao Yongnian 曹永年, Zhao Zhiheng 赵之恒, and Yu Yong 于永, Ming–Qing Neimenggu xibu diqu kaifa yu tudi shahua 明清内蒙古西部地区开发与土 地沙化 (Beijing, Zhonghua shuju, 2006). Xu, Jiongxin, “A Study of Long-Term Environmental Effects of River Regulation on the Yellow River of China in Historical Perspective,” Geografiska Annaler, Series A, Physical Geography 75.3 (1993), 61–72. Yin, Shaoting, People and Forests: Yunnan Swidden Agriculture in Human–Ecological Perspective (Kunming, Yunnan Education Publishing House, 2001). Zhang, Ling, The River, the Plain, and the State: An Environmental Drama in Northern Song China, 1048–1128 (Cambridge, Cambridge University Press, 2016). Zhang Tao 张涛, Xiang Yongqin 项永琴, and Tan Jing 檀晶, Zhongguo chuantong jiuzai sixiang yanjiu 中国传统救灾思想研究 (Beijing: Shehui kexue wenxian chubanshe, 2009). Zhao Zhen 赵珍, Ziyuan, huanjing yu guojia quanli: Qingdai weichang yanjiu 资源，环境 与国家权力：清代围场研究 (Beijing: Zhongguo renmin daxue chubanshe, 2012).

299

https://doi.org/10.1017/9781108587334.009 Published online by Cambridge University Press

8

Population Change shuji cao translated by dong yan and richard von glahn

Introduction: The Micro and Macro in Population History In the preface to his seminal Studies on the Population of China, 1368–1953, Ping-ti Ho called his book “basically an essay in economic history” that “is not intended to be a demographic analysis, which must be undertaken by experts differently equipped than I.” Ho’s approach was endorsed by John K. Fairbank in his foreword to the book, because “statistics of the modern or would-be-modern type – census data and government statistical reports designed for the purpose – are unavailable for China in the Ming and Ch’ing periods.”1 For Ho and Fairbank, the research approach Ho took in his work belonged to population history, not historical demography. Whereas population history is a subfield of historical studies, using historical research methods to investigate population patterns in history, historical demography is a subfield of demography, mainly using research methods in modern demographic studies, in particular statistical analysis of data such as marriage, reproduction, death, and family structures, and investigates the relationship between fluctuations in these indicators and their social and environmental settings. Population history belongs to the domain of macro-level research, while modern demographic research is focused on micro-level patterns. Macrolevel population history takes regions as the basic unit of analysis, while micro-level demographic history takes families or kinship groups as units of analysis. Although the evidence utilized in this chapter is derived from 1

Ping-ti Ho, Studies on the Population of China, 1368–1953 (Cambridge, MA, Harvard University Press, 1959), pp. xi–xii, x.

300

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

aggregate data from population surveys or censuses, in an age of rapidly improving research methods of statistical analysis there remains ample opportunity to apply quantitative analysis to these data. Especially on data platforms using spatial analysis such as GIS as a foundation, regional population data with prefecture or county units of analysis can be matched with economic or other statistical data, opening up new vistas for research in economic history, or history generally. Rather than viewing war, pestilence, and famine as simply stemming from poverty (an excess of consumption demand over subsistence resources), as a Malthusian analysis does, it is important to recognize the crucial role of the state in initiating warfare, preventing pestilence, and supplying famine relief. Although micro-level changes in population were determined by households, state behavior had a substantial effect on macro-level changes. Any blurring of distinctions between the two, including using micro-level data to explain macro-level phenomena (or vice versa), would be inadvisable. Much of the data and analysis of changes in China’s population between the year 1000 and 1800 presented in this chapter is derived from the multivolume series Population History of China and the History of Chinese Migration, both published under the general editorship of Ge Jianxiong.2 In addition, I have made extensive revisions and additions to the Ming and Qing volumes of Population History of China that will be published in a new, two-volume study entitled Demographic History of China: 1368–1953.3 This chapter comprises four sections: (1) population records and demographic data, (2) national population figures, (3) regional distribution and population movements, and (4) urban population in late imperial China. Reconstruction of the regional distribution of population, migration, and urbanization is based on secondary administrative areas formally established since the Song dynasty as the units of analysis, including provinces, prefectures, and counties, as well as market towns. In terms of methodology, the chapter is not a study of historical demography using the household as the unit of analysis, but rather a population history based on regional units of analysis.

2

3

Ge Jianxiong 葛剑雄 (ed.), Zhongguo renkou shi 中国人口史, 6 vols. (Shanghai, Fudan daxue chubanshe, 2000–2005); Ge Jianxiong 葛剑雄 (ed.), Zhongguo yimin shi 中国移民 史, 6 vols. (Fuzhou, Fujian renmin chubanshe, 1997). Cao Shuji 曹树基, Zhongguo renkou shi, 1368–1953 中国人口史, 1368–1953 (Shanghai, Shanghai Jiaotong daxue chubanshe, forthcoming).

301

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

Population Records and Demographic Data Submission of reports on tax revenues and population estimates by local officials to state governments (known as “submitted accounts,” shangji 上計) became a regular practice as early as the Spring and Autumn era (771–453 B C E). The oldest extant household registers (huji 戶籍) date from the dynastic Qin state c. 220 B C E (see von Glahn, Chapter 3 in this volume). Under the Qin Empire (221–206 B C E), the population was organized into units of five families that bore mutual responsibility for tax payments and legal infractions. Adult sons were obliged to separate their households from that of their parents. Thus the household, usually comprising a conjugal couple, their children, and perhaps an elderly parent, became the basic unit of state regulation, taxation, and social control, a pattern that largely held true throughout the imperial era. From rosters of military garrisons stationed at Juyan, on the northern frontier of the Han Empire, dating from the first century B C E, we can see that Han household registers typically included details on the head of the household and his dependents, such as their place of registration; their place of residence; their merit rank; the names, genders, and ages of all members of the household; and their possessions, such as dwellings, livestock, and slaves. Despite the political turmoil of the Period of Disunion (third to sixth centuries C E), ruling states in both north and south China continued to compile population surveys for the purposes of levying taxation and recruiting subjects for labor and military service. The extant registers from the frontier oasis town of Dunhuang dating to 416 C E contain similar information as Han registers, except that the system of merit ranks had long been defunct, and the status of household heads was recorded in terms of official rank, soldiery, and a category (san 散), whose meaning remains unclear. Under the Tang dynasty (618–907), heads of households were required to file annual affidavits (shoushi 手實) that recorded the details of household members, but also the landholdings in their possession under the equal-field system of state land allocations. The shoushi records were the basis of the triennial household registration records submitted by local magistrates to the central government. A number of actual shoushi documents and household registers from the frontier regions of Turfan and Dunhuang have been preserved. These documents indicate the pervasive and stable nature of the institution of household registration. Household registers often were not reflective of actual population figures and their changes over time. Since the main purpose of registration was to

302

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

determine tax liabilities, there was a strong incentive to underreport household membership, falsify ages, or claim special dispensations for the elderly and lame. Concealment of full knowledge of household membership became part of a process of long-term contention between the state and its subjects. Under the Song dynasty (960–1279), which retained the twice-a-year tax (liangshui 兩稅) tax system adopted by the Tang government in 780, households were ranked into grades based on household assets (originally nine, reduced to five in 1033). Wu Songdi has divided household registration surveys in Song into three broad categories of records: (1) adult-male service registers (dingzhang 丁賬), (2) five-tiered household wealth ledgers (wudeng dingchan bu 五等丁產簿), and (3) tax payment receipts (shuizhang 税賬).4 The Song government relied on the adult-male service registers to mobilize individuals for labor service duties. The household ranking system was used to determine tax liabilities based on the value of household property (arable land in rural areas, shops and business revenue for urban residents). The population was divided into resident (zhuhu 主戶) and guest (kehu 客戶) households; resident households were divided into five ranks, with the burden of taxation falling heaviest on the top three ranks (who comprised 10–20 percent of the total). The meaning of “guest household” has been the subject of much debate, but initially it referred to landless households or immigrant settlers who were granted amnesty from land taxes for a certain period of time. In principle, guest households who acquired lands would become liable for taxation, and over time the number of guest households dropped precipitously, at least in the most economically developed regions. During the Song dynasty, local authorities were required to submit tax collection estimates for summer and autumn taxes every third year, in years with intercalary months (hence the designation runniantu 閏年圖, “leap year charts”), based on the adult-male service registers. These surveys were designed for the fulfillment of taxation and labor service quotas, and thus were not actually demographic surveys, since they only counted males liable for such obligations. The Song government regularly compiled comprehensive geographic gazettes that incorporated the national population data derived from the “leap year charts.” Thirteen comprehensive geographic gazettes were compiled during the Song era, of which two – Taiping huanyu ji 太平寰宇記 (c. 980) and Yuanfeng jiuyu zhi 元豐九域志 (c. 1080) – survive.

4

Wu Songdi 吴松弟, Zhongguo renkou shi 中国人口史, vol. 4, Song Liao Jin Yuan shiqi 宋 辽金元时期 (Shanghai, Fudan daxue chubanshe, 2000), pp. 24–49.

303

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

The system of household registration surveys during the Mongol Yuan period was complex and multifaceted, proceeding as the Mongols conquered Chinese territories in stages. After the conquest of Jin and the occupation of north China in 1235, the Mongols registered their new subjects under various occupation categories, including – apart from ordinary subjects (min 民) – Buddhist and Daoist clergy, stage post caretakers, and artisans. In 1252, in addition to these occupational categories, new “ethnic” classifications were instituted, including Huihui 回回 (Muslims), Uyghur 畏吾兒, and ortoq 斡脫 (merchants, mostly of Central Asian and Muslim origin, enjoying certain privileges and acting as commercial agents for the Mongol nobility), as well as a new occupational category of “Confucian households” (ruhu 儒戶). All of these population categories were subject to household registration. In 1370, Zhu Yuanzhang (r. 1368–1398), founder of the Ming dynasty (1368– 1644), ordered the ministry of revenue to conduct a general household registration survey, with each household given confirmatory affidavits (hutie 戶帖). Local officials were expected to register changes in household membership every year. Extant copies of hutie issued in 1371 list, in addition to the head of a household, resident male and female members, including children, as well as arable land, dwellings, and livestock. In the early Ming period hutie not only included information on the male population, but also that of females, even female children. The Ming also followed the Yuan practice of designating hereditary occupational castes, primarily military, artisan, and saltern households, who were registered separately from civilian households. In 1381, Zhu Yuanzhang ordered the compilation of “yellow ledgers” (huangce 黃冊) for taxation and labor service purposes. Sub-county-level yellow ledgers recorded the names of heads of household (huahu 花戶), but at the county level and above these ledgers only contained aggregate data. At the same time the population was organized into groups of 110 households known as li 里, of which ten rotated the office of jia 甲 headman, a system that became generally known as lijia. Among other duties, jia headmen were assigned responsibility for collection and delivery of tax payments from individual households in their li. Households headed by persons ineligible for labor service (e.g., women, widowers, orphans) were attached to regular li units as “supplemental” (daiguan 带管 or jiling 畸零) households. Different standards for “supplemental” units meant that the actual number of households in a li might vary from 110 to 120 to as many as 150 to 160. Although under-registration was common in peripheral areas, 304

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

many of the population and li data compiled in 1381 (and subsequently revised in 1391, which are the data extant now) are considered reliable. In the Hongwu era, primary households (huahu) corresponded to actual families. As families increased in size over generations, multiple subsidiary households (zihu 子戶) appeared. Due to the government’s strict control over division of households, many individual or subsidiary households continued to share a pre-existing primary-household register. There were also instances where newly formed households were not registered separately from existing primary households. Since the state’s interest in data collection was mainly based on its need to ensure a stable source of revenue (rather than track actual population fluctuations), compilation of yellow ledgers took on a formulaic quality in later periods, as primary households became little more than tax accounts. This remained the case into the early Qing. Thus, after the Hongwu era, the household registration system no longer kept accurate records of changes in household composition, resulting in static population records that did not correspond to social reality. However, until the Tianshun era (1457–1464), the Ming government still compiled aggregate information on lijia units under procedures that appear to have been uniform across provinces. The Universal Gazetteer of the Great Ming (Da Ming yitong zhi 大明一統志), compiled under imperial auspices and promulgated in 1461, preserves data on li units for prefectures and counties. Thus it might be possible to extrapolate population figures from the number of li administrative units recorded in the Universal Gazetteer. To determine whether there existed an inherent relationship between population and li units at the level of prefectures, we first extracted prefectural-level population figures in 1391 for twelve provinces, and the number of li listed according to prefectures in these provinces in the Universal Gazetteer of the Great Ming. Then we calculated the ratio of prefectural-level population to provincial population, as well as the ratio of li numbers within provincial li figures, and arrived at 101 pairs of effective records. Data for a few areas that underwent substantial migration between 1391 and 1461 (such as large-scale migration from Zhending to Dongchang, and the population surge in Beiping following the relocation of the court there in the 1420s) were removed. The correlation between population and numbers of li in the remaining ninetyfour pairs of records is quite clear, with R2 = 0.9221 (Figure 8.1). Under the subsequent Qing dynasty (1644–1911) the imperial state eventually halted household registration altogether. In 1668, the Qing government replaced the compilation of yellow registers, which had served as the basis of the Ming land and labor taxes, with quinquennial surveys known as bianshen 305

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

Figure 8.1 Correlation between 1391 prefectural population and 1461 numbers of li for ninety-four prefectures

編審. Unlike the yellow ledgers, which were supposed to include data on all members of the household, the bianshen surveys merely gathered information on adult males who were liable for the ding 丁 labor service levy. In 1712, the Qing established permanent quotas for the ding levy, and in 1729 the Yongzheng Emperor (r. 1722–1735) formally merged the ding levy into the land tax on an empire-wide scale. Henceforth individual landowners paid a single lump-sum tax in silver, and the household itself no longer was a unit of taxation. The merging of the ding levy into the land tax rendered the bianshen surveys obsolete, although they were formally ended only in 1772. Since the population registers no longer had any fiscal purpose, compiling population data became an empty bureaucratic exercise. As part of its belated effort to modernize its administrative and fiscal institutions, in 1910 the Qing government conducted its first empire-wide population census. Comparing the 1910 Qing census with prefectural data from the 1953 census, the first national census undertaken by the People’s Republic of China, one can see that only figures from Sichuan and Shandong in the 1910 census can be regarded as reliable. Therefore, in this work, we prefer to use population data from 1953 as the benchmarks for reconstructing population figures from earlier periods, supplementing these data with the Shandong and Sichuan data for 1910. This approach was adopted by Ho Pingti in his study. Despite a long tradition of household registration and aggregation of population data in imperial China, the quality of these population surveys across time was uneven. For example, household figures for various commanderies and principalities in 2 C E recorded in Ban Gu’s History of the Former

306

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

Han Dynasty are considered to be not only complete but of good quality, while the figures for the Eastern Han censuses preserved in Sima Biao’s Continuation of the Han History are much less satisfactory. From late Eastern Han to the reunification under Sui, China for most of the time was politically divided, with widespread omissions and concealment of household figures. Hence population estimates for this period remain speculative. Numerous historical and geographic works preserve empire-wide population figures for the Sui and subsequent dynasties. But even the best of these data present problems. For example, the “Geographical Treatise” of the History of the Song Dynasty lists the number of households and “mouths” (kou 口) for each of the empire’s 234 prefectures in 1102 (some figures actually date to 1080). But the average number of mouths per household is only 2.34 for the empire as a whole; in forty-eight prefectures the average is below 1.8, while in eighteen prefectures the average exceeds 4.0. Scholarly consensus holds that the “mouths” figures in Song sources reflect the number of males – not only adult males (ding) liable for labor services, but minor and elderly males as well. But these figures probably undercount even the male population; other sources such as famine-relief records and local gazetteers suggest average household sizes in the range of 5.0–5.4 persons. Thus, while household data preserved in Song records are considered reliable, data on individuals are clearly incomplete. To correct for this underestimation, scholars have adopted an average head count of five persons per household to generate total population estimates. For regions lacking household survey records or for which we are unable to reconstruct the original figures, we can arrive at estimates by comparison with adjacent regions with similar environmental features. This was the method also applied to estimate populations for some regions during the Hongwu era in the early Ming. Omissions in data must, however regrettably, be accepted. Through various methods of reconstruction, Wu Songdi has compiled prefecture-level household figures for 980, 1078, and 1102, which were then multiplied by an average head count of five to generate national population estimates for these three time periods.5 Empire-wide household figures for the Hongwu era of the early Ming can be found in official publications such as the 1587 edition of Statutes of the Great Ming (Da Ming huidian 大明會典), but these data were not categorized by prefecture. Prefectural-level population figures were mostly obtained 5

Wu, Song Liao Jin Yuan, pp. 122–37, Table 4-2.

307

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

through various local gazetteers, including full data for the capital region, Henan, Shanxi, Jiangxi, and Guangdong provinces in provincial gazetteers; prefectural data for Zhejiang in various prefectural gazetteers; and some prefectural data from other provinces. In addition, extrapolations were needed to obtain figures for twenty-nine prefectures and sixteen militaryadministered borderland zones, and prefectural-level figures in Yunnan and Guizhou. Take Shandong as an example. Local gazetteers provide population figures for three (Dongchang, Qingzhou, and Laizhou) of Shandong’s six prefectures for 1391, as well as for ten counties of Yanzhou, from which figures for the remaining thirteen counties of Yanzhou were extrapolated. Similarly, local gazetteers provide data for four of Dengzhou’s eight counties, which formed the basis for extrapolating population figures for the remaining four. By subtracting the population figures for these five prefectures from the provincial total, we arrive at a population of approximately 986,000 for the sixth prefecture, Ji’nan. According to the sixteenth-century edition of the provincial gazetteer, the population for Ji’nan prefecture in 1526 was 2,103,000 persons, which implies an annual population growth rate of 1.6 percent between 1391 and 1526. This growth rate is broadly similar to that of most other regions in north China, which in turn confirms the accuracy of estimates for Ji’nan in 1391. For Shaanxi and Sichuan, for which only provincial-level population data are available, prefecture-level population figures have been derived from the number of li units in each prefecture, a method that is validated by the results presented in Figure 8.1. From the example of Ji’nan prefecture mentioned above, various regional population growth rates in the Ming could also be calculated, resulting in a national prefectural-level population estimate for 1580. From these estimates, Hou Yangfang has constructed a map of prefectural-level population densities in 1580 using ArcGis. Based on the maps and data kindly supplied by Professor Hou, I have constructed a prefectural-level map of population densities in 1391 so that population distribution and density levels for these two dates could be compared. Household data for the Qing came from a multitude of sources. Empirewide prefectural data were extracted from the Universal Gazetteer of the Great Qing (Da Qing yitongzhi 大清一统志) (1812), then corroborated with household data gleaned from various local gazetteers, resulting in estimates for prefectural-level population in 1820. Multi-year data from gazetteers were also used to generate population estimates for 1680 and 1851. Select individual samples were combined with prefectural-level survey data from 1953 to 308

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

extrapolate prefectural-level figures for 1910, which were used to determine urbanization rates for 1776. To use Ji’nan prefecture again as an example, the Universal Gazetteer of the Great Qing recorded figures of 736,000 households and 4,015,000 individuals, an average of 5.5 persons per household. Ji’nan’s population in 1953 was 6,120,000 persons, which indicates an annual population growth rate since 1820 of 0.39 percent. The Ji’nan Prefectural Gazetteer of 1838 recorded population data for counties and garrisons (wei 衛) in 1838, with totals of 743,000 households and 4,202,000 individuals, or an average of 5.7 individuals per household. Based on these sources, the annual population growth rate in Ji’nan prefecture from 1820 to 1838 was 0.26 percent. Assuming an annual growth rate of 0.32 percent, we can infer a population estimate of 3,519,000 for Ji’nan in 1776.

National Population Figures Based on the procedures outlined above, we obtained twenty-four data points for the national population up to the first national census in 1953 (Table 8.1). Working backward from the earliest empire-wide census of 2 C E , Ge Jianxiong has estimated China’s population in the early Han dynasty (202 B C E –220 C E) at 15 million to 18 million, a figure whose reliability has been confirmed through exhaustive corroboration from other sources.6 The average household size of Table 8.1 China’s population from 200 B C E to 1953 (figures in millions)

6

Year

Population

Year

Population

200 B C E 2 CE 24 140 609 755 980 1110 1210 1290 1330

17 60 35 60 60 70 35 99 124 75 85

1341 1391 1580 1630 1680 1776 1820 1851 1880 1910 1953

90 75 173 221 185 312 383 437 364 436 590

Ge Jianxiong 葛剑雄, “Guanyu Qindai renkou shuliangde xin guji” 关于秦代人口数量 的新估计, in Ge Jianxiong, Ge Jianxiong zixuanji 葛剑雄自选集 (Guilin, Guangxi shifan daxue chubanshe, 1999), pp. 16–25.

309

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

4.2 to 5.0 persons, according to Ge’s estimate, fits well with general descriptions of a household size of five individuals. The average population growth rate of 1.25 percent between 200 B C E and 2 C E also corresponded with the average population growth rate assumed under normal conditions for China. Census figures often show sharp declines from previous peaks at the outset of a new dynasty, due to both actual declines (from war, starvation, displacement, etc.) and the limited capacity of new dynasties to compile complete registration records. Thus the recorded population in 980, shortly after the founding of the Song dynasty, was estimated at only 35 million, around half of the previous population peak in the mid-eighth century. After this period, the Chinese population began to grow at a rapid pace, reaching 124 million in the twelfth to thirteenth centuries. After the upheavals of the Song–Yuan and Yuan–Ming transitions, China’s population dipped to 75 million in 1391, in the early years of the Ming. By the end of the Ming, the population had risen to 221 million, before war, famine, and pestilence in the mid-seventeenth century reduced the population to around 185 million in 1680. Subsequently Qing China enjoyed a long period of domestic peace and prosperity that raised the population to over 312 million by 1776 and 437 million by 1851, on the eve of the outbreak of the devastating Taiping Rebellion (1850–1864). In terms of the rate of population growth, the highest rates of growth occurred during the Tang–Song transition between 980 and 1110 (annual growth rate of 1.4 percent, inflated by the significant undercount in 980), under the Ming dynasty between 1391 and 1630 (annual growth rate of 0.8 percent, although the figure for 1391 also surely undercounts the actual population), and during the long eighteenth century from 1683 to 1851 (annual growth rate of 0.8 percent). But large-scale population decimation also took place before these three periods of rapid population growth. Clearly war, famine, and pestilence were key factors in influencing Chinese population fluctuations. According to Malthus, Western societies controlled their population growth mainly through restrictions on marriage, which was termed “preventive checks,” whereas the Chinese population grew uncontrolled, until death rates also increased as a result of worsening poverty, otherwise known as “positive checks.” If we consider the fundamental nature of war, famine, and pestilence to be one of poverty, then Malthusian theories would best describe the phenomenon of cyclical fluctuations in Chinese population. In periods without major war, pestilence, and famine, the Chinese population maintained annual average growth rates as high as anywhere in the premodern world, a result of adjustments made by Chinese households in their marriage and reproductive patterns due to their economic conditions. 310

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

The high rate of population growth in China after 1949 was more a product of government policies that promoted public health care, reduced infant mortality, and encouraged reproduction. State intervention in the post-1949 period provides another reading of historical population fluctuations in China; if war, pestilence, and famine are not seen as fundamentally manifestations of poverty, but as the result of state behavior, and we regard ending warfare, preventing pestilence, and relieving famine as state actions, then it can be said that micro-level increases in Chinese population were determined by household behavior, but macro-level increases were determined by the Chinese state.7 Chinese population growth in the past two millennia is the result of repeated interaction between these two forces. Existing individual studies show that in Chinese households, household wealth can be positively correlated with the number of children, although wealthy households with few children also exist.8 So at the micro level, preventive and positive checks can coexist, which in turn require further individual case studies.

Regional Distribution and Population Movements In antiquity and throughout the early imperial period, the Yellow River valley, especially its lower course across the Central Plain of north China, was the economic heartland of China. In periods of political disturbance and warfare, people seeking asylum mainly tended to migrate from the Yellow River valley to the Yangzi river basin in south China. The steppe nomad invasions of north China beginning in the late third century C E and the relocation of the court to the south after the sack of the Jin dynasty’s capital, Luoyang, in 311 prompted the first large-scale southward migration by northern Chinese. Most of these refugees settled in the Wu and Shu kingdoms, with their capitals in the Yangzi delta and Sichuan respectively. If one were to view this as a precursor to large-scale southward migration by northern Chinese, then the following three southward migrations formed

7

8

For evidence of use of preventive checks by a premodern Chinese population – bannermen households in Liaoning province in the late Qing period – see James Z. Lee and Cameron Campbell, Fate and Fortune in Rural China: Social Organization and Population Behavior in Liaoning, 1774–1873 (Cambridge, Cambridge University Press, 1997). See, for example, Stevan Harrell, “The Rich Get Children: Segmentation, Stratification, and Population in Three Chekiang Lineages,” in Susan B. Hanley and Arthur P. Wolf (eds.), Family and Population in East Asian History (Stanford, Stanford University Press, 1985), pp. 81–132.

311

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

Luoyang Chang’an Xiangyang Chengdu Jiangzhou

0

250

Yangzhou

Hangzhou

500 km

Direction of migration

Map 8.1 Migration routes during the Tang–Song transition, 755–1000

the most important migratory events in Chinese history, as well as the basic framework of post-Song population growth in China. The outbreak of the An Lushan Rebellion in 755–763 had a devastating impact on the Central Plain, although the duration of warfare was brief enough for most of the refugees to return in due course. In contrast, the Huang Chao Rebellion (874–884) and the ensuing internecine fighting among warlords endured for more than thirty years, resulting in the fall of the Tang dynasty in 907 and the fragmentation of China into numerous regional states. This prolonged period of warfare triggered large-scale migration from north to south (Map 8.1). The thriving economic development of south China during this period also acted as a magnet for immigration from the north. The spectacular advance of rice cultivation in the Yangzi river basin, especially in the southern part of the Yangzi delta (the Jiangnan region), and the emergence of new patterns of mass consumption stimulated by southern products such as tea, sugar, porcelain, paper, and timber (along with transplanted northern industries such as silk), sustained economic and demographic expansion in the south down to the Mongol conquests of the thirteenth century. Prefectural-level data for c. 1102 show that the Yangzi delta, the Chengdu plain in Sichuan, coastal Fujian, and even parts of the middle Yangzi river valley had eclipsed the Central Plain of north China as

312

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

Kaifeng Nanjing Suzhou Hangzhou

Chengdu

Quanzhou Guangzhou

Population density (p/km2) 0–34 34–79 79–157 157–318

0

250

500 km

Map 8.2 Population densities in Northern Song China, c. 1102

the most dynamic centers of population growth (Map 8.2). In northern China, Hebei was the most densely populated region in 1102, with average population densities of sixty to a hundred persons per square kilometer, followed by Henan and Shandong. In southern China, population density was highest in Jiangnan, at seventy to 110 persons per square kilometer, followed by Jiangxi and Fujian. Using the Huai river valley – located midway between the Yellow River and Yangzi river basins, and roughly speaking the northern limit of rice cultivation – to demarcate between north and south China, south China comprised only 19 percent of the empire’s population in 2 C E, but grew inexorably to 28 percent in 609 and 45 percent in 742, and eventually overtook north China with 57 percent of the total population in 980. From c. 1100, twothirds of China’s population lived south of the Huai river and one-third in the north, ratios that have remained largely constant down to modern times. In the late imperial period, the demographic structure of the Chinese empire had become inverted compared to the early imperial era. In 1126, the Jurchen kingdom based in Manchuria overran the Song capital, Kaifeng, and brought most of north China under the rule of its Jin dynasty (1115–1235). The Song court continued to rule south China,

313

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

including Sichuan, from its refugee capital at Hangzhou. The Jurchen conquest of north China triggered another wave of southward migration, and the economic prosperity of the south was maintained during the Southern Song dynasty (1127–1279). It is estimated that roughly 5 million people, or around 15 percent of the entire northern population, fled to the south during this time. The economic and demographic centrality of the Jiangnan region became even more pronounced during the Southern Song period, when the imperial capital was located in this region, at Hangzhou. Population growth also accelerated in Jiangxi and Fujian. The Kaifeng area remained the mostly densely populated region in north China under Jin rule. By contrast, the borderland regions between the Song and Jin states in Henan and the Huai river valley suffered population losses on both sides of the military frontier. Although there is strong genetic evidence that the plague bacillus that precipitated the Black Death pandemic in Western Asia, the Mediterranean, and Europe beginning in 1347 originated on the western margins of China, the impact of bubonic plague on China remains unclear. Chinese sources make no mention of the distinctive symptoms of bubonic plague. Some scholars have argued for outbreaks of plague in northern and northwestern China as early as the first quarter of the thirteenth century.9 Although China’s population fell in the thirteenth century and again in the mid-fourteenth century (Table 8.1), China seems to have been spared the massive mortality that the Black Death caused in the Mediterranean and Europe. From the Jurchen invasion of north China in 1126 down to the founding of the Ming, wars among the Song, the Jurchen, and the Mongols, as well as famine, pestilence, and waves of migration, resulted in severe depopulation in most parts of northern China and the central-western portions of southern China. In response, the Ming government organized a massive effort of population relocation to redress these imbalances and revive regional economies. For example, the central government transferred inhabitants of Shanxi to other parts of north China, with migrants from Zhending prefecture (Hebei) moving to northwest Shandong, population from north of the Yan mountains (northwest of

9

For arguments in favor of this view, see Cao Shuji 曹树基, “Dili huanjing yu Song–Yuan shidai de chuanranbing” 地理环境与宋元时代的传染病, Lishi dili 历史地理1995.12, 183–92; Robert Hymes, “Epilogue: A Hypothesis on the East Asian Beginnings of the Yersinia pestis Polytomy,” in Monica H. Green (ed.), Pandemic Disease in the Medieval World: Rethinking the Black Death (Kalamazoo, Arc Medieval Press, 2014), pp. 285–308. For a skeptical view, see George Sussman, “Was the Black Death in India and China?”, Bulletin of the History of Medicine 85.3 (2011), 319–55.

314

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

modern Beijing) relocating to the south of the mountain range, and migrants from the Shandong peninsula moving to western Shandong. Migrants from Jiangxi moved westward into Hunan and Hubei. In addition, the government forcibly relocated large numbers of people from the Jiangnan region to populate the new capital at Nanjing. The creation of a new system of military garrisons (weisuo 衛所) spread across the empire was another major vehicle of population transfer. By 1391, the combined number of migrants recorded on civilian and military registers reached 11 million, or nearly 15 percent of the total population. Of these 11 million migrants, military households comprised 4 million, or more than one-third of the total. Garrison deployments in northern and southwest borderlands deserve most of our attention, since many parts of today’s Liaoning, Hebei, Inner Mongolia, Shanxi, Gansu, Yunnan, and Guizhou provinces are populated by descendants of Ming military households. The 1391 census vividly demonstrates the continued centrality of the Jiangnan region, where population density exceeded 300 persons per square kilometer, as the demographic and economic heartland of the early Ming empire (Map 8.3). All regions of the empire enjoyed steady population growth throughout the fifteenth and sixteenth centuries. Although the rate of growth was highest in northern regions such as Shanxi, Shandong, and Beijing (which became the Ming capital in the 1420s), in the late Ming era the

Beijing Linqing Nanjing Suzhou Hangzhou

Chengdu

Population density (p/km2) 0.4–21.2 21.2–56.1 56.1–117.7 117.7–228.4 228.4–497

Guangzhou 0

250

500 km

Map 8.3 Population densities in Ming China, c. 1391

315

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

Beijing Linqing Nanjing Suzhou Hangzhou

Chengdu

Population density (p/km2) 0.3–35.6 36.6–91.4 91.4–188.4 188.4–464.2 464.2–938.4

Guangzhou 0

250

500 km

Map 8.4 Population densities in Ming China, c. 1580

densest populations were still found in Jiangnan, the central Yangzi river basin, and Fujian, as the data from 1580 show (Map 8.4). The western regions, even Sichuan, remained relatively sparsely populated. The breakdown in social order following the first Manchu invasion in 1618 culminated in widespread rebellion and civil war in the 1630s to 1640s, followed by the outbreak of devastating epidemics, which coincided with the poor harvests and famine conditions fostered by the seventeenth-century Little Ice Age (see Bello’s chapter in this volume). The Manchu conquest of China in 1644–1645 was relatively bloodless, but Manchu soldiers may have been the agents for the spread of epidemic disease to the southernmost regions of the empire. Between 1630 and 1680, China’s population declined by approximately 15 percent (Table 8.1). Among the regions most harshly affected by population losses during the Ming–Qing transition were Sichuan, the Han river valley in Hubei, and upland regions of south China in Jiangxi, Hunan, Zhejiang, and Fujian (Map 8.5). Following the consolidation of Qing rule after the suppression of the Rebellion of the Three Feudatories (1681) and the defeat of the renegade Zheng regime in Taiwan (1683), central and local governments made strenuous efforts to induce migrants to settle in depopulated areas and along the empire’s expanded frontiers. In addition, the stability and prosperity China

316

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

Beijing Linqing Nanjing Suzhou Hangzhou

Chengdu

Net reduction rate of population 0–0.164 0.164–0.308 0.308–0.493 0.493–0.752 0.702–0.98

Guangzhou 0

250

500 km

Map 8.5 Net population losses during the Ming–Qing transition

enjoyed throughout the eighteenth century encouraged migration from densely populated areas to frontier regions in the west. Large numbers of people from Hunan and Jiangxi moved westward to settle in Sichuan and the Han river basin. Farther west, migrants from Hunan, Guangdong, Sichuan, and central/northern Jiangxi also settled in Guizhou, Guangxi, and Yunnan. Migrants who took advantage of New World food crops such as maize and sweet potato to settle in the agriculturally marginal highland areas of the interior such as Jiangxi, Hunan, and Hubei became known as “shed people” (pengmin 棚民). The development of borderland areas also generated further waves of migration, with Hakka populations in the highland areas of northeast Guangdong and southwest Fujian migrating to Jiangxi, Zhejiang, and Sichuan. Not all of these newcomers were welcome. Tensions and at times violent conflicts arose between Han settlers and the indigenous inhabitants of western Hunan, Guizhou, Guangxi, and Yunnan, and between Hakka immigrants and long-settled populations in the southeastern provinces. Taiwan was another new frontier for Chinese migration and settlement. In the seventeenth century, Taiwan became a contested prize among competing maritime trade empires, including the Dutch, the Spanish, and the Zheng clan from Fujian. Under Dutch and – after expelling the Dutch from Taiwan in 1661 – Zheng rule, Chinese settlers in Taiwan began to develop an 317

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

agricultural base despite hostile conflict with the island’s indigenous peoples. In the eighteenth century, immigration to Taiwan from Fujian swelled, driving the indigenous peoples into the rugged mountains in the eastern half of the island. One region that was largely exempt from immigration was Manchuria. Initially, the Qing court compelled convicts, including many Ming loyalists and soldiers, to settle in relatively developed areas of Manchuria such as the Liaodong peninsula. After 1668, however, Manchuria was entirely closed to Han settlement, in order to preserve the region as the sacred homeland of the Manchus. Much of the region was designated as a vast hunting ground, although it also produced valuable commodities such as ginseng, which was placed under government monopoly. Still, the ban on immigration was never fully enforced, and immigrant Han were recruited to farm the estates assigned to Manchu banner units. Mongolian regions under Qing rule also straddled the fine line between legal outlawing of migration and de facto encouragement of settlement. It was not until 1860 that the Qing government relaxed its restrictions on settlement in northwestern Jilin and areas north of Harbin. Although the Qing government stationed garrisons of Manchu banner troops in major cities throughout the empire, apart from in Beijing their numbers were small relative to the surrounding population. As noted earlier, the population of Qing China grew at a rapid pace throughout the eighteenth century and into the early nineteenth (Table 8.1). In contrast to the trend since the late Tang dynasty, when population growth was concentrated in Jiangnan as well as in south China generally, during the eighteenth century the main regions of population growth were in the north (central and southern Hebei, western Shandong, and eastern Henan), the far south (the Pearl river delta in Guangdong and coastal areas in southern Fujian), and above all the west (Sichuan, northern Hubei, and some parts of the far southwestern provinces of Yunnan and Guizhou). Nonetheless, Jiangnan remained the most densely populated region in the empire (Map 8.6).

Urban Population in Late Imperial China Figures for urban populations before the Song mostly come from literary sources, and thus are prone to be scattered, vague, or exaggerated. For example, the ninth-century poet Han Yu in one poem recited the line (frequently repeated at the time) mentioning “Chang’an’s million families,” but in another essay declared that “the number of people in the capital today 318

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

Beijing Linqing

Chengdu

Nanjing Suzhou Hangzhou

Hankou

Population density (p/km2) 0.1–42.6 42.6–97.9 97.9–179.8 179.8–357.2 357.2–719.4

Guangzhou 0

250

500 km

Map 8.6 Population densities in Qing China, c. 1776

is no less than a million.”10 These two statements conflate “families” and “people.” Most likely the so-called “Chang’an’s million families” actually meant a million people, but this number too can only be considered a rhetorical flourish. According to contemporary literary accounts, Luoyang, the Tang dynasty’s eastern capital, possessed a population second only to Chang’an’s. Other major cities with substantial populations included Jinling (modern Nanjing), Bianzhou (Kaifeng), Yangzhou, Jiangling, Suzhou, Chengdu, Hongzhou (Nanchang), Hangzhou, Guangzhou, and Taiyuan.11 Generally speaking, higher-level administrative cities tended to have populations that correlated with their administrative level. Provincial capitals were usually more populous than prefectural capitals, which in turn were larger than county seats. But special prominence as commercial hubs or strategic military sites might elevate a city population beyond its administrative rank. By the Song period, statistics on urban population began to appear in historical records, facilitating more thorough study of urban populations. Administrative records often distinguished urban residential districts – 10

11

Cited in Dong Guodong 冻国栋, Zhongguo renkou shi 中国人口史, vol. 2, Sui Tang Wudai shiqi 隋唐五代时期 (Shanghai, Fudan daxue chubanshe, 2002), pp. 216–18. Dong, Sui Tang Wudai, pp. 502–7.

319

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

designated “urban precincts” (chengxiang 城厢), “within the walls” (chengnei 城内), or “urban wards” (fang 坊) – from rural ones, yielding concrete data on urban populations in individual cities. For example, an edict of 1021 states that 97,750 households lived in the 120 urban wards (fang) of the old and new portions of the Northern Song capital, Kaifeng; in addition, there were fourteen wards outside the city walls. If the ex-urban wards had the same average population as the intramural wards, the total number of households would be around 110,000, or – based on an average household size of five persons – approximately 550,000 people. Wu Songdi estimates that the urban population for Kaifeng in 1102 was around 137,000 households, and if the imperial household, sojourning bureaucrats, and soldiers were included, then the city’s total population would be around a million. Wu also calculates the urban population of the Southern Song capital, Hangzhou, c. 1265 at slightly over 174,000 households, or – adding the imperial entourage and families of soldiers stationed there – a total of 1.2 million persons. Using similar methods, Wu calculated the population for Beijing in 930 (when it was the Liao “Southern Capital”) and 1207 (then the Jin “Central Capital”) as around 200,000 and 400,000 inhabitants respectively. By 1327, the population of Beijing – which the Mongol Great Khan Khubilai established as his capital, Dadu/Khanbaliq, in 1260 – could have reached 1,060,000 inhabitants.12 Wu further classified Song cities into tiers. The first-tier cities (with around 50,000 or more households, or 250,000-plus inhabitants) included Suzhou, Jiangning (Nanjing), and Shaoxing in the Jiangnan region; Hongzhou (Nanchang) in Jiangxi; Fuzhou and Quanzhou in Fujian; Ezhou (Hankou) in Hubei, Tanzhou (Changsha) in Hunan; and Chengdu in Sichuan. He further ranked six cities in the second tier (20,000–50,000 households, or 100,000–250,000 inhabitants), and thirteen cities in the third tier (5,000– 20,000 households, or 25,000–100,000 inhabitants).13 Although some scholars have postulated that the proportion of town dwellers in the Southern Song reached as high as 20 percent of the total population, a more likely estimate would be in the range of 12 percent.14 For the Ming, local gazetteers provide copious references to the numbers of urban wards (chengli 城里 or fangxiang 坊厢), although they rarely report actual data on urban populations. If we assume that from early to late Ming, urban wards and rural li grew or were incorporated at roughly similar rates, then the ratio between urban wards and rural li in mid- to late Ming could be 12 14

Wu, Song Liao Jin Yuan, pp. 573–4, 583–8. Wu, Song Liao Jin Yuan, pp. 614–19.

13

Wu, Song Liao Jin Yuan, p. 600.

320

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

viewed as the same as for the Hongwu era. This hypothesis is borne out through many local gazetteers. Given prefectural-level population figures, with the number of li per county, we can arrive at estimates for population per county, and also for urban populations in cases where the li figures are broken down into urban and rural components. Consider the example of Hangzhou. Hangzhou prefecture had two metropolitan counties, Renhe and Qiantang, which together comprised 41 percent of the prefecture’s total number of 1,400 li in 1391, when the total civilian population of the prefecture was 1,081,000. Thus we can infer that civilian population in these two counties was around 442,000 in 1391. The 1475 Hangzhou prefectural gazetteer provides current urban and rural li figures for both counties, but only Renhe’s li figures for 1391. Renhe’s li decreased between 1391 and 1475, from forty to thirty-two urban li (20 percent) and 354 to 333 rural li (6 percent). If we apply the same ratios of decline to Qiantang, we can infer that in 1391 Qiantang had fifty-eight urban li and 121 rural li. This yields a total of ninety-eight urban li for the city of Hangzhou in 1391 (17.1 percent of the total number of li in the two counties), and thus an estimated urban civilian population of 76,000. If we include military garrisons and their families, the total population for the city of Hangzhou would be around 88,000.15 Similar calculations were applied to arrive at estimates of urban populations for the cities of Nanchang, Fuzhou (Fujian), and Wuchang (Hankou) (Table 8.2). Applying the method used in the Hangzhou case to data for forty prefectural capitals in various provinces yields the results presented in Table 8.3. To highlight regional variations, we separated the Southern Metropolitan province into three parts – southern Jiangsu, northern Jiangsu, and southern Anhui. Regions are listed in order of population density. Zhejiang had the highest population density, and the largest urban populations – averaging 23,000 inhabitants – for prefectural cities. Southern Jiangsu ranked second in population density and average size of prefectural cities (15,000). In Jiangxi, southern Anhui, and Fujian, the average size of prefectural cities ranged between 10,000 and 13,000. The smallest prefectural city populations were found in Huguang and Beiping. We must remember that at this time Beiping was not the imperial capital, but merely the seat of a depopulated province 15

Using Ming-era regulations on military garrisons, each chiliad or battalion (qianhusuo 千 户所) should include 1,120 soldiers, and each garrison (wei 衛) should comprise five chiliads, or a total of 5,600 soldiers. Assuming that each soldier had two dependents, the entire garrison should reach a population of around 16,500. Occasional adjustments in these garrison figures throughout the Ming are not included here.

321

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

Table 8.2 Estimated urban population in four Ming provincial capitals, c. 1391 Hangzhou Metropolitan counties Total no. of li Urban li Percentage urban li Population of metropolitan counties (thousands) Estimated urban population (thousands) Urban military population (thousands) Estimated total urban population (thousands)

Nanchang

Renhe

Qiantang

394 40 10.2

Fuzhou

Wuchang

Min

Houguan

Jiangxia

179 58 32.4

Nanchang, Xinjian 738 57 7.7

126 16 12.7

68 9 13.2

61 15 21.6

343

156

473

90

49

70

35

51

37

11

6

17

37

0

5

19

0

11

123

42

36

28

Sources: HZFZ, juan 2; BMTZ, juan 15–16; NCFZ, juan 5; HGZS, juan 1

Table 8.3 Estimated urban population in forty prefectural capitals, c. 1391

Prefectural capitals Zhejiang 2 Southern Jiangsu 2 Jiangxi 7 Shandong 2 Southern Anhui 3 Fujian 5 Northern Jiangsu 1 Beiping 5 Huguang 13

Percentage of urban li

Average registered civilian population (thousands)

(Civilian) population density (persons/km2)

3.7 5.9 6.8 7.4 10.6 6.9 12.1 9.5 9.7

23 15 12 13 13 10 11 2 5

295 127 49 40 39 32 29 22 11

322

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

that was beginning to attract immigrants. Outside Beiping itself, the largest prefecture was Baoding, a town of 4,000, and the prefectural seat of Yongping had a civilian population of only 400, although the presence of a local garrison raised its total population to around 6,000. We might reach more accurate conclusions if we adopt a method of analysis that is not based on administrative divisions such as provinces. In some cases there is a significant disparity between the ratio of prefectural population to provincial population and the ratio of prefectural li numbers to provincial li numbers, so we have decided to exclude figures from four prefectures that veered too far off the mean. Analysis of the remaining thirty-six prefectures reveals that the denser the population of a region, the more populated its prefectural cities will be, and vice versa. The regression here is Y = 5.145x – 7.656, R2 = 0.5843, meaning that over 58 percent of a prefectural population’s variance could be explained through population density. In addition, within these cases, the larger the population of the prefectural city, the smaller the ratio of the prefectural township to the county it was situated in.16 Figure 8.1 shows that there is a high correlation between the ratio of prefectural to provincial li numbers and the ratio of prefectural to provincial population. This conclusion can be extended to the following: there is a high correlation between the ratio of county to prefectural li numbers and the ratio of county to prefectural population. If we classify prefectural cities with civilian populations of 20,000 or more as “large” and those smaller than 20,000 as “small,” then Ji’nan, Taiyuan, Kaifeng, Xi’an, Wuchang, Chengdu, and Kunming were all small prefectural cities in 1391. Guizhou was not yet a province at the time, and the future provincial capital, Guiyang, was only a barracks. We are still uncertain about why Guilin appears as a “large prefectural city” in 1391. If garrison population and their dependents are included in a prefectural city’s population tally – that is, with one-third of the garrison on active duty, two-thirds of the garrison engaged in farming, and each soldier with two dependents – then for every wei, the number of soldiers with their dependents is precisely the head count for each wei – 5,600 people. If we include these military populations in our estimate, then the list of “large prefectural cities” also would include Kaifeng, Henan, Taiyuan, Huizhou, Fengyang, Raozhou, Jianchang, Ganzhou, Wuchang, Jinzhou, Huangzhou, Ningbo, Zhaoqing, Chengdu, and Yunnan.

16

For further details on the estimation techniques, see Cao, Renkou shi.

323

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

As with prefectural capitals, we assume that from early to mid-Ming, urban and rural li of counties were added or merged at the same rate. For example, according to the 1475 gazetteer for Hangzhou, Lin’an county originally had four urban li, of which three remained; the county’s sixteen rural districts (xiang 鄉) originally had sixty-two li, of which forty-seven remained. The ratios of urban li to total county li in Lin’an were 6.1 percent in 1391 and 6.4 percent in 1475. According to the same gazetteer, the county seats of Haining, Yuhang, Fuyang, and Lin’an possessed eleven, seven, five, and four urban li respectively, while there were no records of urban li for Yuqian, Xincheng, and Changhua counties. Thus not all county seats can be considered urban centers. Applying the method used in the case of the Hangzhou prefectural capital to data for 289 county seats in the same provinces analyzed in Table 8.3 yields the results presented in Table 8.4. Again, the Southern Metropolitan region is separated into three parts – southern Jiangsu, northern Jiangsu, and southern Anhui – and listed in order of population density. In Shandong, northern Jiangsu, and Beiping, the county seat’s population was highly correlated with population density. in Southern Jiangsu, Zhejiang, Fujian, and Jiangxi, the correlation between population density and county population was less significant. For both north and south, county seat population was not related to the ratio of urban li to total county li. Based on the data shown in Table 8.4, we modeled for the relationship between county seat population density and county seat urbanization for 115 counties in northern China, which yielded an exponential Table 8.4 Estimated county seat urban populations for 289 counties, c. 1391

Southern Jiangsu Zhejiang Jiangxi Shandong Southern Anhui Fujian Northern Jiangsu Beiping Huguang

Prefectural capitals

Percentage of urban li

Average registered civilian population (thousands)

8 17 29 34 7 30 8 73 83

2.4 5.0 5.3 6.0 7.5 12.0 13.4 7.5 7.9

6 7 6 3 4 7 2 1 2

324

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

(Civilian) population density (persons/km2) 183 152 49 42 39 32 32 22 11

Population Change

function of y = 0.0534e1.0246x；R2 = 0.5827; that is, over 58 percent of the variance in county seat population could be explained through population density. In the case of the 154 counties in south China, an even more robust exponential function of y = 0.6055e0.5302x；R2 = 0.6523 was obtained. This analysis greatly improved the correlation between population density and county seat population in northern China, and increased the accuracy of estimates for county seat populations. In southeast China, in addition to population density, other factors also affected the scale of county centers. For example, in Shaoxing prefecture, where average population density was 142 persons per square kilometer, the six outlying county seats had an average estimated population of 14,000: for Xiaoshan and Yuyao, the estimated urban population is in the range of 27,000–29,000; for Zhuji, Shangyu, Shengxian, and Xinchang, it is in the range of 7,000–10,000. However, in neighboring Huzhou, which had a higher average population density of 191 persons per square kilometer, the average population of its outlying county seats was only 3,000. Adjacent Yuqian county in Hangzhou prefecture had no urban li at all. Closer study of the local economy of these areas is needed to evaluate these discrepancies. Our calculations generate high rates of urbanization for Fujian in 1393. Tingzhou, Yanping, and Jianning prefectures in the rugged interior of Fujian, despite low population densities of fifteen, thirty-two, and thirty-eight persons per square kilometer respectively, had estimated county seat populations of 8,600, 11,300, and 10,800 respectively. In the Song–Yuan period, Tingzhou was located on a key salt trade route into southern Jiangxi. We estimate that 20 percent of Tingzhou’s population resided in its county seats, probably because of the central place of the salt trade in the local economy. The large populations of county seats in Yanping and Jianning prefectures probably were related to their proximity to Quanzhou. The prosperity of Quanzhou as an international port stimulated local industries such as production of export porcelains throughout the city’s hinterland. These areas also absorbed goods imported into Quanzhou.17 Commerce and commercial routes seem to explain the higher rates of urbanization found in Fujian. Applying these methods to estimate urbanization rates for 1391 yields the results shown in Map 8.7. Urbanization rates were significantly higher in the south than the north, and higher across all regions of the south, not only in more densely populated regions such as Jiangnan. Apart from the 17

Billy K.L. So, Prosperity, Region, and Institutions in Maritime China: The South Fukien Pattern, 946–1368 (Cambridge, MA, Harvard University Asia Center, 2000).

325

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

Beijing Linqing Nanjing Suzhou Hangzhou

Chengdu

Urbanization rate 0–0.048 0.048–0.088 0.088–0.176 0.176–0.33 0.33–0.588

Guangzhou 0

250

500 km

Map 8.7 Urbanization rates, c. 1391

concentration of urban population in the new capital, Nanjing (and the anomaly of Fengyang, Zhu Yuanzhang’s hometown, which he sought to build up as a secondary capital), as we have already mentioned urbanization rates are also noticeably higher in Fujian, especially around the port of Quanzhou. For the early Ming Empire as a whole, the civilian urbanization rate in 1391 is estimated at 7.5 percent, and if garrison populations are included, 8.3 percent. Using population in 1391 as the base, the urban population in 1580 can be extrapolated by applying average population growth rates for each region; the resulting data include the military garrison populations. Individual calculations were made for new counties formed in late Ming; we also took into account small townships that had significant concentrations of commercial activity, and assigned an average of 1,000 for each township. The results are shown in Map 8.8. Overall, China’s urbanization rate in 1580 was 9.5 percent, a substantial increase from the 7.1 percent we calculate for 1391. The regions with the highest urbanization rates in 1580 were Beijing, Hejian (Hebei), and Linqing (Shandong) in the north and Yangzhou, Suzhou, and Hangzhou in Jiangnan, along with Guangzhou and Chengdu. Urban populations were most highly concentrated in the imperial capital and cities along the Grand Canal, which was crucial for delivery of resources to Beijing, as well as the major regional

326

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

Beijing Linqing Nanjing Suzhou Hangzhou

Chengdu

Urbanization rate 0–0.068 0.068–0.11 0.11–0.16 0.16–0.233 0.233–0.435

Guangzhou 0

250

250 km

Map 8.8 Urbanization rates, c. 1580

commercial centers in Jiangnan and other parts of south China. It is unclear, however, why some areas, such as Pingle prefecture in the southwestern province of Guangxi, registered such high rates of urbanization. Periodic rural markets appeared at least by the Han dynasty, and proliferated during the Tang period. Many of these markets were situated at crossroads and ferry landings along waterways. They arose spontaneously out of local marketing needs rather than government initiative, and they lacked the walls and gates that enclosed urban markets at that time. Stimulated by the commercial expansion of the Song period, rural periodic markets evolved into more permanent trading centers, with shops, inns, and warehouses that attracted regular business from traveling merchants (see Zurndorfer’s chapter in this volume). The Song government began to collect taxes from commercial activity, and as marketplaces grew the state began to recognize them as townships (zhen 鎮) and assigned constables and tax collectors to monitor the town population. According to Wu Songdi, townships generally had a minimum population of roughly 100 households (500 persons), but many of these market towns had populations ranging from 1,000 to 5,000 or more residents, equivalent to the typical populations of county seats.18 18

Wu, Song Liao Jin Yuan, pp. 607–8.

327

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

In some cases, the township population exceeded that of the county seat. Such was the case for the prosperous market town of Huangchi in Dangtu county (Anhui). The commercial tax revenues of the market towns of Hufu and Henglin in Changzhou prefecture (Jiangsu) greatly exceeded those of Changzhou’s county seats of Yixing and Wuxi.19 Thriving market towns often were upgraded to serve as administrative centers; many county seats in the Jiangnan region originated as market towns. Nonetheless, most townships in the Song–Yuan era remained small. Due to an absence of systematic documentation, especially outside the Jiangnan region, it is difficult to conduct research on Song-era township population. Records concerning townships, especially in local gazetteers, become both plentiful and systematic for the Ming period. Xu Panqing has compiled a database of townships derived from the encyclopedic Conspectus of Historical Geography (Dushi fangyu jiyao 讀史方輿紀要), compiled in the late seventeenth century by Gu Zuyu on the basis of late Ming records.20 Excluding those with uncertain sources and county seats, his database lists a total of 1,014 townships. Combing other sources such as local gazetteers yields an additional thirty-nine townships, bringing the total up to 1,053. But can we determine whether all of these townships actually functioned as commercial centers? For the Qing era, scholars have concluded that the townships listed in the “Gates and Passes” (guanai 關隘) section of the Universal Gazetteer of the Great Qing should be regarded as commercial towns. Comparing Xu Panqing’s database with the 1744 edition of the Universal Gazetteer of the Great Qing, we classify Ming-era townships that were still designated as townships in the Qing as market towns. Those that no longer appear in the Universal Gazetteer of the Great Qing or in contemporary local gazetteers have been excluded from our sample. Although this method might leave out some actual market towns, it should give a close approximation to reality. For example, Gu Zuyu’s work recorded twenty-four market towns in Songjiang prefecture (Jiangsu), seventeen of which reappear in the Universal Gazetteer of the Great Qing. Four of the other seven can be verified as market towns based on local gazetteers. Fan Shuzhi identified twenty-seven market towns based on local gazetteer records, although some of these did not correspond to those in Gu Zuyu’s work.21 Since the number of Songjiang market towns listed in Ming and Qing sources remained roughly constant, 19 20 21

Wu, Song Liao Jin Yuan, p. 612. I am indebted to Professor Xu for sharing his database with me. Fan Shuzhi 樊树志, Jiangnan shizhen: Chuantongde biange 江南市镇：传统的变革 (Shanghai, Fudan daxue chubanshe, 2005), pp. 108–13.

328

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

Beijing Linqing Nanjing Suzhou

Chengdu

Hangzhou

Guangzhou 0

1 point is 1 town

250

500 km

Map 8.9 Distribution of market towns in the late Ming

we can assert that the twenty-four townships listed in Gu were all market towns, of which six were major commercial centers. In addition to population figures that are recorded for major market towns, we uniformly assigned population estimates for average market towns based on conventional estimates. For Jiangnan and Fujian, each average market town was assigned a population of 2,000, and for other regions a population of 1,000. The distribution of market towns for the late Ming era is shown in Map 8.9 (areas for which we have no data are indicated in white). The prominent concentration of market towns in the Jiangnan region, with additional clusters in the middle Yangzi river basin, coastal Fujian, the Chengdu plain, and the Grand Canal corridor, is consistent with our general knowledge of the commercial geography of the late Ming period. For the Qing period, there are no records equivalent to the Ming-era urban wards, so we were unable to estimate average populations for provincial, prefectural, or county seats as we did for the Ming. Instead, we have used a unique set of sources to estimate urban population in Shandong for 1910, from which we extrapolated urban population in other provinces for 1910. Using these estimates as a base, combined with empirical research and modeling, we arrive at estimates for the urban population in 1776.

329

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

For eight commercial ports and major cities in Shandong (Yantai, Qingdao, Ji’nan, Weixian, Zhoucun, Longkou, Zaozhuang, Boshan), we conducted case-by-case empirical research to estimate their population. We have omitted that process here.22 We modeled for estimates of average urban population in Shandong based on Lin Xiuzhu’s Survey Records of Localities in Shandong (hereafter referred to as Survey Records), compiled in 1919, which recorded the population not only for each Shandong county, but also for the population of major market towns as well as the number of major commercial firms (shanghao 商號) in each county.23 Lin’s systematic compilation of commercial firms lends itself to the following hypothesis: by modeling the relationship between the number of commercial firms and urban population we can generate population estimates for each market town. The Survey Records covered eighteen categories of data (e.g., county area, population, arable land, taxation quotas), four of which (population, industry, commerce, and transportation) were related to urban population. Population figures were compiled and submitted by each county, and were more detailed than the 1910 census.24 As for commercial firms, we defined them as trading firms or shops that possessed their own firm names and were of a certain operational size, as distinct from peddlers. The Survey Records did not specify whether the commercial firms it listed operated only in county seats or throughout the county (including market towns). In only one case, for Yishui county, was it specified that the tabulated commercial firms included “over 70 commercial firms dealing in general merchandise and money exchange in the county seat, and another 150 or so in the rural areas. Others are generally small-scale enterprises.”25 Assuming that the seven market towns throughout Yishui county had 155 firms, with Yuezhuang, a larger town, having thirty-five firms, we can estimate that each of the other six towns had an average of twenty commercial firms. But for some counties, if we use this average to estimate the number of commercial firms in market towns, the aggregate figure would surpass the number for the entire county, which means that commercial firms listed in the Survey Records generally did not include market towns outside the county seat. We mainly consulted two sources to cross-check data from the Survey Records; the first is Agricultural and Commercial Statistics, an annual digest of 22 23

24

25

For details, see Cao, Renkou shi. Lin Xiuzhu 林修竹, Shandong gexian xiangtu diaocha lu 山東各縣鄉土調查 (Ji’nan, Shandong shengzhang gongshu jiaoyuke, 1919). Wang Shida 王士達, Minzhengbu hukou diaocha ji gejia guji: Fulu 民政部户口調查及各 家估計: 附錄 (Shehui kexue zazhi hedingben, 1932–1933). Lin, Shandong diaocha, vol. 2, p. 130.

330

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

reports concerning industry, commerce, mining, agriculture, and forestry that county-level officials submitted to provincial authorities and were published by the Ministry of Agriculture and Commerce’s bureau of statistics. Ten of these surveys were conducted between 1912 and 1921.26 This section mostly uses supplementary data from their second, fourth, and fifth surveys to cross-check data from the Survey Records. The second source for crosschecking data is the Gazetteer of the Three Provinces of Shandong, Shanxi, and Henan, published in 1925 (hereafter Gazetteer of Three Provinces).27 The Survey Records listed 1,993 commercial firms in Ji’nan, which included four townships, among which Luokou, the terminus of the Qingdao–Ji’nan railway, was the largest. Assuming that there were 100 commercial firms in Luokou, and on average twenty for the other three townships, then there would have been 1,833 firms in the city of Ji’nan.28 In addition, there were 148 artisanal workshops and forty manufacturers. Each artisanal workshop generally employed around twenty people on average. The manufacturers were established after 1910, with around forty employees per firm.29 Thus forty manufacturers would be the equivalent of eighty commercial firms, yielding a total of 1,913 “commercial firms” in Ji’nan. Given that the city’s population in 1919 was around 180,000, there would have been one commercial firm for every ninety-four people. Using the same method, we found that in Qingdao there was one commercial firm for every ninety-five people, and in Yantai (including Fushan) one for every 109 people. Therefore, to calculate the urban population for Shandong as a whole, we estimate an average of ninety urban inhabitants for each commercial firm. For example, the Survey Records listed 302 commercial firms in Wendeng, a county of 378,000 people. If one firm corresponds to ninety people, then there would be an urban population of 27,000. In 1932, there were 277 large-scale commercial firms in the county, a figure that tallied with those in the Survey Records.30 It should be pointed out that this referred to commercial firms 26

27

28

29

30

Nongshangbu zongwuting tongjike 農商部總務廳統計科, Zhonghua minguo nongshang tongji biao 中華民國農商統計表 (Beijing, Nongshangbu zongwuting tongjike bianji, 1912–21). Bai Meichu 白眉初, Zhonghua minguo shengqu quanzhi 中華民國省區全誌, vol. 3, Lu Yu Jin sansheng zhi 鲁豫晋三省志 (Beijing, Beijing shifan daxue shidixi, 1925). Agricultural and Commercial Statistics 1912 listed only 264 firms affiliated with the Chamber of Commerce in Ji’nan by 1911; as the discrepancy between the two figures is too large, we have omitted the data from Agricultural and Commercial Statistics 1912. Ji’nan shizhi biancuan weiyuanhui 济南市志编篡委员会 (ed.), Ji’nan shizhi ziliao 济 南市志资料 (Ji’nan, Ji’nan shizhi biancuan weiyuanhui, 1982), vol. 3, pp. 1–15. Shandongsheng Wendeng shi difangzhi biancuan weiyuanhui 山东省文登市地方志 编篡委员会 (ed.), Wendeng shizhi 文登市志 (Beijing, Zhongguo chengshi chubanshe, 1996), p. 347.

331

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

throughout the county, not simply those in the county seat or the port of Weihaiwei. In 1898, the Qing government was compelled to lease Weihaiwei to the British, who transformed it from a military base to a major commercial center. Weihaiwei was divided into Weihai Wharf, where the Royal Navy China Fleet was stationed, and the walled city of Weihaiwei, which remained under Chinese administration. In 1909, the Weihai Wharf Chamber of Commerce was founded with eighty commercial firms.31 Agricultural and Commercial Statistics 1912 recorded seventy-six firms in the Weihai Chamber of Commerce, and twenty for the Wendeng Chamber of Commerce. As their definition of commercial firms departed significantly from that of the Survey Records, however, we have decided to omit these figures. The population of Weihaiwei was almost 50,000 in 1930. The urban population in 1919 was listed at 19,000, a figure that on the whole corresponds to the Survey Records.32 In the case of Ye county, the Survey Records recorded a population of 440,000 persons, 618 commercial firms, and five towns, which by our calculation corresponded to an urban population of around 55,600. The Gazetteer of Three Provinces referred to the town of Shahe as “vibrant in commerce, and far above that of others in Ye county”; Hutouya township “is a famous small seaport”; and Zhuqiao township was “commercially prosperous, ranking second to Shahe.” From these reports, we estimated that commercial firms based in the county seat were probably one-sixth of the county’s total figure; that is, 103 commercial firms, suggesting an urban population of around 9,000. Wendeng and Ye counties, which both benefited from well-developed transportation infrastructure, boasted vibrant commercial sectors and a high degree of urbanization. But most Shandong counties lacked good transportation facilities and thus probably had significantly different levels of urbanization. For example, Xiajin county, according to the Survey Records, had a population of 188,000 persons, fifty-five commercial firms in the county seat, and four townships. If we apply the assumption of twenty commercial firms for each township, then the entire county would have had 135 commercial firms,

31

32

Liu Bensen 刘本森, “Jindai zhimin zujiedi shangye zuzhi de dianxing ge’an – yi Weihaiwei de shangbu shanghui (1916–1930 nian) wei li” 近代殖民租借地商业组织 的典型个案 – 以威海卫的商埠商会 (1916–1930年) 为例, Jianghan xueshu 江汉学术 2014.3, 111. Zhonggong Weihai shiwei yanjiushi 中共威海市委研究室 (ed.), Weihai shiqing: 1949– 1989 威海市情: 1949–1989 (Ji’nan, Shandong renmin chubanshe, 1990), p. 397.

332

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

with 5,000 people in the county seat, and a total urban population of 12,000. In 1949, the county seat of Xiajin had a population of 6,000, with a nonagricultural population of 4,000, which corroborates our estimated population of 5,000 for Xiajin in 1919.33 This “Shandong model” for estimating township population in 1910 was tested against data for other regions taken from Agricultural and Commerce Statistics, the Gazetteer of Three Provinces, Comprehensive Provincial Gazetteers of the Republic of China (Zhonghua minguo shengqu quanzhi 中華民國省區全誌), local gazetteers, and other sources, with satisfactory results. For an example, we have chosen Chengdu prefecture in Sichuan, because of the high degree of correlation between its county-based population data in 1910 and 1953, with an R2 value of 0.96. This means that if we know the county population in 1953, we can estimate its population in 1910. For the city of Chengdu, Agricultural and Commerce Statistics 1912 listed 5,238 commercial firms. If each commercial firm corresponded to ninety people, then Chengdu’s urban population would be 477,000, which broadly matches the population figure for Chengdu of 500,000 from the Gazetteer of the Four Provinces of Shaanxi, Gansu, Qinghai, and Sichuan.34 For nearby Guan county, Agricultural and Commercial Statistics 1913 listed 126 commercial firms, corresponding to a population of 11,000. The Gazetteer of the Four Provinces noted a population of 8,000 for the county seat of Guan. The county’s nonagricultural population was estimated at 11,000 in 1949, so all three sets of data matched each other.35 The number of commercial firms given in Agricultural and Commercial Statistics 1913 should be that of the entire county. Similar procedures were followed for other Sichuan counties. Outside Sichuan, we must use empirical methods to look for records one by one. In some provinces, such as Zhejiang, Agricultural and Commercial Statistics provides detailed data on the number of commercial firms – in Zhejiang’s case, data from fifty counties, forty-three of which have been verified as accurate. In contrast, Agricultural and Commercial Statistics provides only unreliable data or none at all for provinces such as Guangdong, Guangxi, Yunnan, and Guizhou. Comparisons of the available prefecture-level township data for 1910 with prefecture-level population for 1953 indicate a strong 33

34

35

Shandongsheng Xiajin xianzhi biancuan weiyuanhui 山东省夏津县志编纂委员会 (ed.), Xiajin xianzhi 夏津县志 (Ji’nan, Shandong renmin chubanshe, 1991), pp. 108–9. Bai Meichu 白眉初 (ed.), Zhonghua minguo shengqu quanzhi 中華民國省區全誌, vol. 4, Qin Long Qiang Shu sisheng zhi 秦隴羌蜀四省志 (Beijing, Beijing shifan daxue shidixi, 1926). Sichuansheng Guan xianzhi biancuan weiyuanhui 四川省灌县志编纂委员会 (ed.), Guan xianzhi 灌县志 (Chengdu, Sichuan renmin chubanshe, 1991), p. 121.

333

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

Beijing Ji’nan

Xi’an Chengdu Hankou

Urbanization rate 0–0.049 0.049–0.088 0.088–0.148 0.148–0.274 0.274–0.432

Shanghai Hangzhou

Guangzhou 0

250 250 km

Map 8.10 Urbanization rates, c. 1910

correlation: the larger the population for each prefecture, the larger its townships’ population. Applying this model, we obtained the results for urbanization rates in 1910 shown in Map 8.10. Assuming that prefecture-level population in 1776 displayed similar patterns, the model used to determine prefecture-level township population in 1910 could also be used to extrapolate prefecture-based township population in 1776. Although many parts of China experienced population losses during the Taiping Rebellion (1850–1864) and other large-scale rebellions in the mid-nineteenth century, by 1910 the level of urbanization in most regions had generally returned to that of 1776, albeit with significantly lower levels of urbanization in the middle reaches of the Yangzi river valley. Due to declines in the status of Chengde and overland trade with Russia, urbanization rates for areas north of the Great Wall rapidly declined in the late nineteenth century, while urbanization rates for Jiangnan rebounded. Given available archival materials on population for some prefectures and counties, we can adopt a process of analogy; that is, using criteria such as total population, transportation infrastructure, and commercial development, we estimate for other prefectural and county seats that are analogous or adjacent to the places with reliable data. We term the prefecture-level township

334

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

Table 8.5 Estimations of Shandong urban population by prefecture, c. 1776 (population figures in thousands) Empirical populaPrefecture tion

Modeled population

Adjusted population

Empirical populaPrefecture tion

Modeled population

Adjusted population

Xiezhou Jiangzhou Puzhoufu

42 47 54

42 50 60

42 47 54

110 129 36

111 120 34

Pingyangfu 84 Xianzhou 5 Huozhou 17 Zezhou 42 Lu’anzhou 66 Qinzhou 10 Liaozhou 10

86 3 17 50 53 11 9

84 5 17 42 66 10 10

Fenzhoufu 111 Taiyuanfu 120 Pingding34 zhou Xinzhou 18 Baodezhou 5 Daizhou 28 Ningwufu 11 Datongfu 48 Shuopingfu 15 Guisui 22

18 4 28 10 46 30 3

18 5 28 11 48 15 22

population estimates obtained through this method the “empirical population.” Applying this process, the number of townships was obtained through cross-referencing the 1744 and 1812 editions of the Universal Gazetteer of the Great Qing along with local gazetteers. Apart from major townships, population in most townships was assigned according to their level of importance in their own regions. The results of such extrapolation are termed “modeled population.” Comparing the results of “empirical population” with the “modeled population,” we discovered potential flaws in both research methods. We then select the more probable estimate from the two sets of results to form a new “adjusted population” estimate. Table 8.5 shows the results of this estimation for Shanxi province.36 The urbanization estimates by prefecture for the Qing Empire as a whole in 1776 are shown in Map 8.11. Our model estimates Chinese urbanization in 1776 at only 7.3 percent of the total population, lower than the urbanization rates for 1391 (7.5 percent) and 1580 (9.5 percent). Despite the vibrant mercantile economy and flourishing market towns of the eighteenth century, the empirewide urbanization rate regressed compared to the Ming dynasty. Nor did urbanization advance in the late Qing period. The urbanization rate 36

For further details on the estimation techniques, see Cao, Renkou shi.

335

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

Beijing Linqing Nanjing Suzhou Hangzhou

Chengdu Hankou

Urbanization rate 0–0.021 0.021–0.053 0.053–0.09 0.09–0.16 0.16–0.326

Guangzhou 0

250 250 km

Map 8.11 Urbanization rates, c. 1776

in 1910 was only 7.8 percent, not including the Manchurian provinces of Liaoning, Jilin, and Heilongjiang, for which we have no 1910 data. If we extrapolate urbanization for Manchuria for 1910 from the 1953 figures, the overall urbanization rate in 1910 might have reached 9 percent, a level still lower than during the late Ming. Reconstructing the distribution of market towns in 1776 was reached through cross-referencing the 1744 and 1812 editions of the Universal Gazetteer of the Great Qing with various local gazetteers. With the exception of large townships, the populations of average market towns in various provinces were assigned based on different levels of importance within their own regions (Map 8.12). Comparing Map 8.8 with Map 8.11, it is apparent that the number of market towns increased from the late Ming to the mid-Qing, and they were more widely distributed as well. Although the densest concentration of market towns continued to be found in Jiangnan, the most substantial growth in market towns was found in north China, the middle Yangzi river basin, and Sichuan. The broad distribution of market towns in interior regions signified the westward shift of population and growth of commerce in hinterland areas in the mid-Qing period.

336

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change

Beijing Linqing Nanjing Chengdu

Suzhou Hangzhou

Hankou

Guangzhou 0

1 point is 1 town

250 250 km

Map 8.12 Distribution of market towns in 1776

Conclusion The establishment of stable political regimes under the Song, Ming, and Qing dynasties inaugurated sustained phases of population growth lasting 150 years or more. Demographic increases were driven by economic expansion during these periods, but also influenced by state policies, institutional changes, and micro-demographic behavior, all subjects deserving further research. Expansion of cultivated land and adoption of new crops and farm technologies accompanied these spurts of population growth. Most significant, surely, was the shift of population from north to south and the settlement of the Yangzi river basin during the Song period. The adoption of rice farming and the development of southern industries such as tea, porcelain, sericulture, sugar, paper, and timber vastly increased economic productivity and fostered population growth. During the Ming dynasty, population growth remained concentrated in the Yangzi delta region and other core areas of south China – the middle Yangzi river basin, coastal Fujian, and the Pearl river delta. The flourishing cotton and silk industries of the Yangzi delta stimulated urban growth and the proliferation of market towns in rural areas, resulting in a far higher population density than in other regions. During the first half of the Qing dynasty, the dynamics of population growth shifted. Migration to the

337

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

shuji cao

sparsely populated interior regions in central and western China was spurred by state-supported homesteading policies and the expansion of arable land made possible by the introduction of New World crops such as maize, sweet potato, and tobacco. Improved farming methods, rising crop yields, the adoption of non-food commercial crops, and state-supported famine relief institutions bolstered population growth in long-settled regions as well. Northern regions such as the Central Plain of the lower Yellow River valley achieved a resurgence of population growth not seen since pre-Song times. The doubling of China’s population from 185 million in 1683 to 383 million in 1820 was a milestone in global population history. Nonetheless, the economic and commercial prosperity of Ming–Qing China did not propel urban growth on the scale seen in Song times, or in early modern Europe. The urbanization rate of 7.3 percent we have calculated for China as a whole in 1776 was below the levels achieved in the Ming dynasty (8.3 percent in 1391 and 9.5 percent in 1580), and well below the estimate of 12 percent for the Southern Song suggested by Wu Songdi. The growing commercial integration of city and countryside in the eighteenth century and the spread of market towns in most regions of the empire did not raise urban population growth above rural levels. The low rate of urbanization in 1776 (and even in 1910) confirms that the predominance of agriculture in the Chinese economy remained unchanged in the late imperial era.

Further Reading Cao Shuji 曹树基, Zhongguo renkou shi, 1368–1953 中国人口史, 1368–1953 (Shanghai, Shanghai Jiaotong daxue chubanshe, forthcoming). Cao Shuji 曹树基 and Chen Yixin 陈意新, “Maersasi lilun yu Qingdai yilaide Zhongguo renkou: Dui Meiguo xuezhe jinnianlai xiangguan yanjiude piping” 马尔萨斯理论与 清代以来的中国人口 – 对美国学者近年来相关研究的批评, Lishi yanjiu 历史研 究 2002.1, 41–54. Cao Shuji 曹树基 and Li Yushang 李玉尚, Shuyi: Zhanzheng yu heping: 1320–1960 nianzhong Zhongguode huanjing yu shehui bianqian 鼠疫：战争与和平: 1320–1960 年中中国的 环境与社会变迁 (Ji’nan, Shandong huabao chubanshe, 2006). Deng, Kent, “Unveiling China’s True Population Statistics for the Pre-modern Era with Official Census Data,” Population Review 43.2 (2004), 1–38. Deng, Kent, and Shengmin Sun, “China’s Extraordinary Population Expansion and Its Determinants during the Qing Period, 1644–1911,” Population Review 58.1 (2019), 20–77. Deng, Kent, and Lucy Zheng, “Economic Restructuring and Demographic Growth: Demystifying Growth and Development in Northern Song China, 960–1127,” Economic History Review 68.4 (2015), 1107–31.

338

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

Population Change Dunstan, Helen, “The Late Ming Epidemics: A Preliminary Survey,” Ch’ing-shih wen-t’i 3.3 (1975), 1–59. Ge Jianxiong 葛剑雄 (ed.), Zhongguo renkou shi 中国人口史, 6 vols. (Shanghai, Fudan daxue chubanshe, 2000–2005). Ge Jianxiong 葛剑雄 (ed.), Zhongguo yimin shi 中国移民史, 6 vols. (Fuzhou, Fujian renmin chubanshe, 1997). Harrell, Stevan (ed.), Chinese Historical Microdemography (Berkeley, University of California Press, 1995). Harrell, Stevan, “The Rich Get Children: Segmentation, Stratification, and Population in Three Chekiang Lineages,” in Susan B. Hanley and Arthur P. Wolf (eds.), Family and Population in East Asian History (Stanford, Stanford University Press, 1985), pp. 81–132. Hartwell, Robert M., “Demographic, Political, and Social Transformations of China, 750– 1550,” Harvard Journal of Asiatic Studies 42.2 (1982), 365–442. Ho, Ping-ti, Studies on the Population of China, 1368–1953 (Cambridge, MA, Harvard University Press, 1959). Jiang Tao 姜涛, Renkou yu lishi: Zhongguo chuantong renkou jiegou yanjiu 人口与历史： 中国 传统人口结构研究 (Beijing, Renmin chubanshe, 1998). Lavely, William, and R. Bin Wong, “Revising the Malthusian Narrative: The Comparative Study of Population Dynamics in Late Imperial China,” Journal of Asian Studies 57.3 (1998), 714–48. Lee, James, “Food Supply and Population Growth in Southwest China, 250–1850,” Journal of Asian Studies 41.4 (1982), 711–46. Lee, James Z., and Cameron Campbell, Fate and Fortune in Rural China: Social Organization and Population Behavior in Liaoning, 1774–1873 (Cambridge, Cambridge University Press, 1997). Lee, James, and Wang Feng, One Quarter of Humanity: Malthusian Mythology and Chinese Realities, 1700–2000 (Cambridge, MA, Harvard University Press, 1999). Song Changbin 宋昌斌, Zhongguo huji zhidu shi 中国户籍制度史, revised ed. (Xi’an, Sanlian chubanshe, 2016).

339

https://doi.org/10.1017/9781108587334.010 Published online by Cambridge University Press

9

Public Finance christian lamouroux and richard von glahn

The long transition – often marred by violence – between Tang and Song discerned by Naito¯ Konan marked the advent of a new world. The An Lushan Rebellion (755–763) triggered profound political and military crises that shattered the institutions of the Tang dynasty, but also set in motion the slow progression of the market economy, which the Tang leadership began to see as the necessary means to restore its fiscal authority. With the collapse of the equal-field system of state land allocations in the wake of the rebellion, the Tang abandoned the principle of uniform, in-kind taxation of farming households as the basis of its fiscal and military systems. Urgent necessity prompted the adoption of new and more flexible fiscal strategies to secure revenues from commerce and consumption. The expansion of the market economy mitigated the Confucian elite’s traditional hostility toward commerce and acted as a key catalyst for the mercantilist policies pursued by the southern kingdoms during the first half of the tenth century. The reconstruction of fiscal institutions by the Song in 960 drew upon this experience to reestablish, thanks to the development of multiple information systems, an equilibrium between the centralization of power and new forms of delegation of the center’s authority. This reconstruction relied upon the evolution of the monetary system, whose tensions led to the appearance of the first paper money, and especially to the emergence of a major feature that would endure to the end of the empire: silver. The new fiscal regime that emerged during the Tang–Song transition shifted away from taxing people to taxing wealth and eliminated military conscription and most labor services, thus liberating land, labor, and capital for more productive uses. State initiatives to capture the wealth generated by

340

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

the market economy reached their apogee with the deeply interventionist New Policies inaugurated by the chief councilor Wang Anshi in the late eleventh century. Although most of the New Policies were dismantled in the Southern Song period, public finance still relied on taxation of commerce and consumption. The Mongol conquest disrupted the Song fiscal state and subjected China to a predatory form of tribute exaction. The definitive reversal of the Song fiscal trends occurred after the restoration of Chinese rule under the Ming dynasty (1368–1644). The Ming founder, Hongwu, adopted a profoundly antimercantile agenda, including reversion to direct, in-kind taxation of land and people, in pursuing his atavistic dream of restoring a family-farm economy unencumbered by money and markets. Hongwu’s policies ultimately failed, and subsequent leaders of the Ming and Qing (1644–1911) empires adopted a more accommodative posture toward the market economy. But they retained the Confucian ideological commitments to minimal taxation and the primacy of the family farm. Commerce flourished under the Ming–Qing fiscal regime, but the state reaped little revenue from the growth in private wealth.

The Recentralization of Finances (960–1040) The first task of the general Zhao Kuangyin, who became Emperor Taizu of Song (r. 960–976) after his coup to overthrow the Later Zhou dynasty (950– 960), was to weaken the forty regional governors who remained in office. Reprising the methods employed under the Later Jin dynasty (936–946), he entrusted administration of the conquered regions to fiscal intendants (zhuanyunshi 轉運使) whose mission was to control and limit the resources allocated to the armies. At the same time, the prefectural and sub-prefectural administrations were restructured in such a fashion as to ensure effective central control over local officials. Each prefect henceforth was accompanied by a vice prefect specifically assigned control of fiscal affairs. The prefect’s authority could be further reduced as a result of the appointment of a central government official in a sub-prefecture to handle delicate or sensitive situations. The integration that derived from this redefinition of powers at the prefectural level permitted Taizu to accumulate surpluses in precious metals and silks in a special palace treasury, the Sealed Reserve Treasury (fengzhuangku 封樁庫). His goal was to redeem the sixteen northern prefectures which had passed into the control of the Liao Empire in 936, or to finance their reconquest without having to tap regular state income. 341

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

To centralize fiscal authority, Taizu depended on two institutions placed under his direct authority: the Secretariat of Military Affairs (shumiyuan 樞密 院) and the Finance Commission (sansi 三司), which, in the tradition of the Five Dynasties, often was managed by Taizu’s close comrades. Not until 982, under Emperor Taizong (r. 976–997), was the commission restructured, after the dismissal of numerous officials accused of corruption. Between 983 and 993 the Finance Commission fragmented and then recomposed itself by compartmentalizing its operations and multiplying its agencies of financial control. With regard to personnel, the emperor did not hesitate to promote, out of concern for specialization and professionalization, subaltern agents about whose merits he boasted in 995 in these words: “Regarding the strengths and weaknesses of state finances, these men have lived and breathed in this universe since their youth, and certainly know it down to its foundations.”1 The management of public finance became a major issue after the two defeats inflicted on the Song by the Liao in 979 and 986. Taizong was obliged to adopt a costly defensive policy to reinforce the frontier fortresses and garrisons. After a renewed conflict, a truce was duly signed by his son Zhenzong (r. 997–1022) in 1005. Although it is difficult to know what portion of the budget military expenses comprised, the testimonies of court debates cite figures of 60–70 percent following the outbreak of war against the Xi Xia in 1039 and 80 percent during the 1060s, rising to 90 percent during the conflict with the Jin Empire in the twelfth century. Beyond the rhetoric, the military and political situation had a concrete impact on the structure of public finance. In addition to its debt to the emperor’s Privy Purse (neizangku 內 藏庫), the Finance Commission was obliged to discharge debts owed to merchants who purchased monopoly privileges. The financial autonomy of the Privy Purse increased thanks to the income it drew from the regular transfer of the gold and silver paid by merchants to obtain licenses for the salt and tea monopoly revenues from the Finance Commission’s Office of Public Monopolies (quehuowu 榷貨務) to the Privy Purse beginning in 1004. In 1005, when the profits from the tea monopoly hitherto absorbed by the war effort were reassigned to the capital, the Finance Commission was forced to deliver all of its surplus receipts in currency to the Privy Purse. Progressively incapable of maintaining its own resources, the commission depended on annual loans granted by the Privy Purse: between 970 and 1020 approximately a quarter of the annual income of 1

XCB, 37, p. 813.

342

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

Table 9.1 Coin revenues in the Northern Song (in millions of guan (1 guan = 1,000 bronze coins))

Salt Tea Liquor Subtotal for state monopolies Commercial taxes Coinage Twice-a-year tax New policies Total coin revenues Total revenues (coin equivalent) Percentage of total revenues in coin

997

1021

c. 1044

1064

c. 1077a

3.00 2.86 3.26 9.12 4.00 ? ? − 16.93c 35.59 48

3.00 3.30 11.59 17.89 12.04 ? ? − 29.93 57.23 52

7.15 1.50b 17.10 25.75 19.75 ? ? − 45.50 ? −

11.23 1.18 12.86 25.27 8.46 ? 4.93 − 36.82 60.00 61

22.30 0.77 12.93 34.70 8.05 5.95 5.59 18.00 72.29 89.33 81

a

Figures variously from 1076, 1077, and 1078; does not include revenues from Green Sprouts loans and maritime customs (because revenue data for these categories is not divided into coin and non-coin portions). b Figure from 1034; 1054 figure was 1.28 million. c Total coin revenue also includes additional sources apart from the above categories. Source: Richard von Glahn, The Economic History of China from Antiquity to the Nineteenth Century (Cambridge, Cambridge University Press, 2016), Table 6.6, p. 231

the emperor was allocated for such loans. In 1007, the accounts of funds borrowed from the palace treasury since the beginning of the dynasty were annulled, since “year after year it had proven impossible to make restitution.”2 These decisions confirmed a basic fact: income from the twice-a-year tax, whose revenues – calculated on the basis of landed wealth – remained stable in the absence of new cadastral surveys, no longer furnished the bulk of tax receipts. Instead, government revenues increasingly depended on commercial taxes paid by merchants and the income generated by state monopolies. Commercial taxes and monopoly revenues rose from 37 percent of total revenues at the end of the tenth century to over 80 percent by 1077 (Table 9.1). As for the operation of the monopolies, they added to government indebtedness, as a consequence of the so-called ruzhong 入中 method for provisioning the frontier garrisons. Under the ruzhong system, adopted at the end of the tenth century, merchants supplying military provisions were issued bearer certificates, payable at the capital in cash or exchanged for rights 2

XCB, 67, p. 1497.

343

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

to trade in commodities (salt and tea) subject to state monopoly. By discounting these certificates or bills of exchange the merchants created a credit currency. These certificates, which might be redeemed at the capital months or even more than a year later, passed all the more readily under the control of the great commercial firms upon whom the authorities depended to validate the demands of the bearers. This financial arrangement sealed a partnership between the government and the great merchants, while also generating a structural rivalry. Despite frequent modifications of the regulations, this system remained a favored mode of financing, which helped maintain the cohesion of the links between distant regions and the capital as well as more coherence among different means of payment. After its conquest of Sichuan in 965, the Song maintained the iron currency of the previous Shu kingdom, as much to appease the fears of the population, long deprived of bronze coin, as to obtain silk and tea at minimal cost. Absent the circulation of bronze currency and in the face of the depreciation of iron coins, which provoked a large-scale insurrection in the mid-990s, sixteen rich merchant houses in Chengdu prefecture began to issue “exchange bills” (jiaozi 交子) to facilitate commercial transactions. In 1024, the government imposed a monopoly on the issue of these bills, strictly limiting the volume of notes in circulation (initially set at 1.26 million guan) and requiring their redemption every two years, subject to a service charge of 3 percent. Until 1072 these rules ensured the stability of the world’s first paper currency, which sustained commerce between Sichuan and Shaanxi, and also partially financed the war against the Xi Xia in 1040–1044. New credit monies, issued beginning in 1048 in the form of licenses for the salt monopoly in Shaanxi, contributed to stabilizing the exchange rate of the iron currencies that the authorities had introduced in tandem with the bronze currencies of Shanxi and Shaanxi to provide emergency financing for this new conflict.

The Budgetary Crisis and the Bureaucratization of Finances (1040–1127) A growing budgetary crisis set in during the long reign of Emperor Renzong (r. 1023–1063). Contemporaries traced its origins to excessive and permanent expenditures: the maintenance of the armies, then the new war; the increase in the annual payments to preserve peace with the Khitan Empire and the Xi Xia kingdom; and the salaries paid to growing numbers of government officials. The civil service examination system guaranteed emoluments to almost every “presented scholar” (jinshi 進士), whether he occupied an 344

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

actual government post or was awaiting assignment. The personnel budget exploded as a result of the doubling of the number of civil officeholders from 10,000 at the beginning of the eleventh century to 20,000 by the 1050s, before climbing to 24,000 a decade later. The budgetary crisis was exacerbated by a paradoxical phenomenon: the onset of endemic “money famines” (qianhuang 錢荒), even though the volume of metallic currency had steadily increased, with nearly 200 million guan in circulation in the 1080s. The frequent use of “short strings” in the marketplaces, alongside the official reduction of the mo unit of account (a string of 100 coins) to seventy-seven coins, was a symptom of the scarcity of money of good quality. There were multiple causes of the “money famine”: hoarding by urban elites, melting of the best coins, and the adoption of Song bronze currency as the international money for Asian commerce. More fundamentally, the value of coins continued to be defined by current prices, since the number of coins in a string was set by market fluctuations or the decisions of trade associations, and these regional and seasonal fluctuations made money much more a sign of wealth than of productive capital. The hoarding of good coins and the tensions provoked by the increasing fiscal demands of the state, combined with the irregularity of market production, led to recurrent imbalances in the money supply. The response to these challenges came with the reformist “New Policies” inaugurated by Emperor Shenzong (r. 1067–1085), who was determined to follow the proposals of his chief councilor, Wang Anshi. These initiatives, resumed by his sons, Zhezong (r. 1085–1100) and Huizong (r. 1100–1125), created terrible and lasting divisions within the court, to the point of being accused as early as the twelfth century of having precipitated the disaster of 1127, with the loss of the north China heartland to the Jin and the retreat of the dynasty to the south. In fact, from Wang Anshi to Cai Jing, chief councilor to Huizong, the reformist camp pursued a consistent policy agenda: through the monetarization of fiscal channels, they endowed the Kaifeng capital administration and then the empire as a whole with the fiscal means capable of developing a systematic commercial activism. From 1069, with the adoption of the “balanced-supply regulations” (junshufa 均輸法), the transport intendancies (fayunsi 發運司) could arbitrage between price differentials to make grain purchases at low cost in the markets of the southeast while profiting from sales at higher prices when grain was dear. In 1072, Wang Anshi established a Bureau of Market Transactions (shiyiwu 市易務) at Kaifeng through which he financed brokers and merchant guilds, making massive purchases to slash commodity prices in the 345

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

capital by placing the government in a situation of quasi-monopoly over wholesale trade. Elevated to a superintendency (dutiju shiyisi 都提舉市易司) in 1073, this agency supervised seven regional branches and acquired control of about thirty branches in the mid-1080s, especially in the most commercialized regions. Through these massive purchases, the Superintendency of Market Transactions turned local officials into active commercial brokers, including engaging in retail trade when inventories proved too large, as happened in 1078. The pursuit of revenues also dictated the adoption of the “service-exemption” (mianyi 免役) policy, which granted farmers exemptions from public service obligations in exchange for a tax paid in coin. However, this measure raised revenue 20 percent higher than the wages paid to hired substitutes, and households previously not subject to the service-exemption policy were also obliged to pay this tax. Public credit also was inaugurated in 1069 under the “Green Sprouts” (qingmiao 青苗) policy, which encouraged the prefectures to mobilize their ever-normal granary (changpingcang 常平倉) resources to make interestbearing loans to the rural populace, and subsequently to town dwellers. The “land reclamation and irrigation” (nongtian shuili 農田水利) policy similarly offered interest-bearing loans using the resources of the same prefectural granaries, in this case to support investment in agricultural production. In addition, merchants and brokers who engaged in state procurement on behalf of the superintendency were encouraged to conduct their buying and selling on credit, and all of these forms of lending often became mandatory, as attested especially in the case of the “Green Sprouts” loans. These measures were completed in 1072 by the “land survey and tax equalization” (fangtian junshui 方田均稅) policy aimed at conducting a new cadastral survey to improve the revenue yield from the land tax by curbing tax evasion. In 1082, as part of a wholesale reorganization of the central government, the reformist leadership abolished the Finance Commission and replaced it with a revived Ministry of Revenue (hubu 戶部) that reported directly to the prime minister of state. Yet more than half of central government revenues were controlled by various independent agencies and deposited in the new Yuanfeng Treasury (元豐庫), over which – like the Privy Purse – the ministry of revenue exercised little control (Table 9.2). By expanding state moneylending and commercial operations, the New Policies gave unprecedented importance to money in state finance. On one hand, the government, which increased annual mint output to a peak level of 6 billion coins in 1080, refilled its coffers, and even was in a position to provide salaries to government clerks, who in lieu of fixed remuneration had been 346

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

Table 9.2 Central government income in 1093 (converted to thousands of silver kilos) Percentage Revenues of total Privy Purse (Neizang Treasury) Precious metals Domestic tribute Coinage Maritime customs Foreign tribute Hemai (silk requisitions) Other Subtotal Yuanfeng Treasury Alum monopoly Salt monopoly Wine monopoly Hired service levy Breweries and ferries Subtotal Tea Marketing Agency Tea monopoly Ministry of Revenue (Zuozang Treasury) Twice-a-year tax Commercial taxes Head tax Subtotal Total

15.6 6.4 223.5 10.3 2.3 92.6 67.7 421.8

0.5 0.2 6.4 0.3 0.1 2.7 2.0 12.1

22.0 386.0 503.9 395.9 143.7 1,431.4

0.6 8.7 14.4 11.3 4.1 41.0

192.7

5.5

1,107.6 328.0 5.9 1,441.4 3,490.2

31.7 9.4 0.2 41.3

N.B.: Not all sources of state revenue are accounted for. Source: Robert Hartwell, “The Imperial Treasuries: Finance and Power in Song China,” Bulletin of Sung–Yuan Studies 20 (1988), Table 8, 71

collecting fees from litigants and taxpayers; on the other hand, families had to use money to meet tax obligations or repay their loans. In serving as a universal unit of account, bronze coin had placed state power at the heart of the market economy, and the swift collapse of public finances following the abolition of the reforms after Shenzong’s death in 1085 clearly demonstrated that the monetization of taxation compelled the tapping of resources other than land taxes or monopoly revenues. From the early 1090s, the quest for the liquidity that the government needed destabilized the monetary system: between 1093 and 1105, the financing of wars in the west caused the volume of jiaozi notes, for which the fixed 347

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

quota had been abandoned since 1072, to surpass 13 million guan through the retention of old notes and the increase in new issues. The devalued jiaozi were retired in 1107 without redemption to allow the circulation of new “cash vouchers” (qianyin 錢引), which circulated without any metallic reserve and thus were inconvertible with iron coins. By 1128 the market value of a qianyin note with a face value of one guan had fallen to no more than a dozen iron coins. Within the space of two years the Advisory Council (jiangyisi 講議司), established by Cai Jing in 1102 to devise policy proposals outside the regular bureaucratic channels, had decided to create four financial circuits that would secure revenues for the emperor and the central government, but also for the palace agencies (zhujusuo 諸局所) responsible for managing various imperial projects. These measures were supposed to enable a moral reform of the empire directly inspired by the Rituals of Zhou. This moralization of government was reflected in the revival of the “land survey and tax equalization” policy, which had been intended to apportion the land tax more fairly and reduce evasion. The policy proved ineffective, however, and was rescinded in 1120. The service-exemption policy had been re-established without any maximum ceiling, and the system of advanced loans secured by silk payments (heyumai 和預買), originally intended to provide relief for rural households during the lean season but transformed by the state into compulsory payments, was made a general practice. In addition, the court depleted the granary stores of the southeast, which had ensured the integrity of the grain tribute quotas, and demanded payments in money in lieu of in-kind tribute grain before returning, from 1109, to deliveries of grain financed by the regional governments. Land taxes were further increased in 1116 with the creation of the Bureau of Public Lands (gongtiansuo 公田所) controlled by the eunuchs, which forced the population in the northern regions to pay rents to the imperial treasury for uncultivated lands. This quest for revenues was pursued even more fiercely with respect to the commercial sector. The government introduced in 1100 a Bureau for Price Stabilization (pingzhunwu 平准務), successor to the Superintendency of Market Transactions, and Cai Jing extended state monopolies to salt and tea in the southeastern regions. The profits derived from this credit money put the finishing touch to the series of currency manipulations that the state had conducted since 1102. The emission of large coins of low bronze metal content, fixed at values of five and then ten cash, had destabilized prices and provoked rampant counterfeiting. Each level of government had to find the means to fulfill its quota, even though the funds were siphoned 348

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

off for the benefit of the court or the palace. In other words, local governments were pushed to finance their operations by creating nonstatutory taxes.

The Fragmentation of Finances (1127–1205) From a fiscal and monetary perspective, the sack of Kaifeng in 1127 entailed the loss of hundreds of tons of gold and silver and 100 million guan of bronze coins deposited in the vaults of the imperial treasury at the time. It thus became necessary to resort to expedients to finance the war against the Jin, which would last sixteen years at a cost of 70 million guan per year. The main response to this challenge rested on the massive issuance of paper money and bills of exchange secured by monopoly commodities as well as temporary tax measures that eventually became permanent levies: “taxes for supply and regional finances” (jingzong zhiqian 經總制錢); the “surtax on commutation of tribute silk” (zheboqian 折帛錢); and “monthly reserve funds” (yuezhuangqian 月樁錢), which were gradually extended from the southeastern prefectures to the entire empire. The documentary sources allow us to follow the policies employed in two strategic regions: in the lower Yangzi basin, where a new capital was established at Hangzhou in 1138, and in Sichuan. In Sichuan, between 1128 and 1136, the fiscal intendant, Zhao Kai, succeeded in augmenting revenues by means of the sale of licenses for liquor and tea as well as taxing textile production and all commercial transactions. To facilitate transactions, he issued more than 600,000 guan of qianyin notes in 1128, secured by bronze currency minted especially to supply the reserve funds that would allow the withdrawal of excess notes. Zhao had to resign his post, however, without having succeeded in reining in the issue of qianyin notes needed to provision the armies. Twenty-five million guan were in circulation in 1136, and nearly 42 million in 1142 when the peace treaty with the Jin was concluded. The value of these qianyin notes nonetheless remained fairly stable, with one guan of paper money worth 800 iron coins or 400 bronze coins until 1157. The regional authorities succeeded in instilling confidence in paper money by accepting qianyin notes in their fiscal operations, particularly in the context of the state monopolies, and by utilizing them for government purchases and to pay official salaries. In the southeast, as in Sichuan, indirect taxes on consumption constituted a growing part of state revenues. Despite the precarious state of public finances 349

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

and faced with steep inflation resulting from a sharp decline in state coinage and the proliferation of counterfeiting, officials determined that without guarantee of convertibility it would be impossible to issue paper money in this region. In 1131 the government made recourse to issuing promissory notes known as guanzi 關子, which were distributed to merchants who traded them at a discount, the bearer being paid in coin by the office of state monopolies in the capital. These expedients attest to the scarcity of cash, and the end of the war in 1142 marked a new period of “monetary famine,” resulting in deflation that persisted throughout the next two decades. In sum, by forcing massive emission of paper monies, the war reinforced the differences in means of payment in the two major strategic regions and highlighted the specific articulation between the creation of money and credit in each region. The war also aggravated a structural feature of the fiscal crises that had begun a century before: the inadequacy and deficiency of tax data. The confidentiality of the Privy Purse’s fiscal accounts limited the information available to the Ministry of Revenue, which, now paralyzed by the war, no longer was in a position to halt the dizzying growth in the regional agencies’ expenditures. Under the control of Qin Gui, the early Southern Song state redefined the basis of the land tax by expanding the cadastral survey (jingjie 經界), initially applied in Suzhou, to the whole territory between 1142 and 1149. The new census enacted a public registration of landholdings based on “fish-scale registers” (yulince 魚鱗冊), which recorded details of the size of plots and quality of land in every administrative unit. It was hoped that this survey would restore a proper return in tax revenues based on the realities of land tenure, but the ability of landowners to suborn local officials rendered the measure ineffective. The peace of 1142 made possible a restoration of civilian control over the military. In 1141 three general commands (zonglingsuo 總領所) were created in the Huaidong, Huaixi, and Jing–Hu (more commonly rendered as Hu– Guang) regions, and in 1145 a fourth one in Sichuan. Charged with provisioning the armies, the chiefs of the general commands – appointed from the ranks of the central government – were invested with powers intended to prevent the rise of new military potentates. But their authority, which aside from fiscal ledgers and finances extended to control of local officials, stockpiling weapons, and recruiting soldiers, reinforced regional autonomy. This autonomy became even more formidable from the 1160s, when the central government entrusted the general commands with management of their own currencies. 350

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

The outbreak of renewed warfare with the Jin, which coincided with the accession of Emperor Xiaozong (r. 1162–1189), led to the issue beginning in 1161 of the huizi 會子 paper currency in the southeastern region in order to resolve the monetary famine and to finance the new military conflict. With the return of peace in 1165 Xiaozong undertook two measures to stabilize state finances: institutional reforms at the center and the adoption of a composite monetary regime defined by the concurrent circulation of paper monies throughout the empire through the creation of four monetary zones distinguished by different means of payment from one general command to another (Map 9.1). From 1166 the allocation of fiscal resources was entrusted to a Budget Department (guoyongsi 國用司). Independent of the ministry of revenue, this office compiled estimates not only for expenditures by the armies and the civil offices at the capital, but also for all tax receipts raised by the circuits and prefectures and paid to the Ministry of Revenue, as well as the fiscal

Zhongdu (Beijing)

I: Capital and Southeast: huizi paper money and bronze currency II: Sichuan: qianyin paper money and regional iron currency III: Huainan: huaijiao and huizi paper monies and regional iron currency IV: Hu–Guang: huhui paper money and regional iron currency

XIXIA JIN

w llo Ye

R.

Kaifeng Huai R.

TIBET

IV

Chengdu

III Hangzhou

II

Yangzi R.

I DALI

Map 9.1 Monetary regions in Southern Song Source: Richard von Glahn, “Origins of Paper Money in China,” in K. Geert Rouwenhorst and William N. Goetzmann (eds.), Origins of Value: The Financial Innovations That Created Modern Capital Markets (New York, Oxford University Press, 2005), p. 74

351

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

requirements of the general commands. The role played by the Ministry of Revenue in the fiscal transfers among central government agencies was further weakened by the fact that Xiaozong had increased the number of imperial treasuries while making the chief treasury of the Ministry of Revenue, the Treasury of the Left (zuozangku 左藏庫), systematically dependent on borrowing from the Privy Purse. Xiaozong’s deep immersion in financial affairs resulted in a fortuitous stabilization of the value of huizi notes. In 1166–1167 Xiaozong disbursed more than 112 tons of silver to redeem excess notes and withdraw them from circulation. The value of the 6 million guan of huizi remaining in circulation was sustained by the introduction of a policy of “equilibrium maintenance” (chengti 稱提) whereby the government would routinely repurchase excess notes. Moreover, from 1168 new regulations rendered huizi notes inconvertible during their three-year term of expiry while simultaneously requiring their acceptance in tax payments. Public revenues and expenditures, including salaries and fiscal transfers, had to be paid half in coin and half in paper currency from 1170 onwards. Public confidence kept the notes close to their nominal value until the 1190s. Meanwhile, the “monetary war” that culminated in the flight of bronze coins toward the Jin empire, where they were more highly valued, had prompted the Song authorities to create several monetary buffer zones from 1165. The Huai river basin, the principal Song–Jin frontier, was provided with huaijiao 淮交 notes – denominated initially in bronze coin in 1166, then in iron coin from 1169 – along with huizi currency from the capital region (Map 9.1, zone I I I). In 1163 the middle Yangzi river basin, poor and depopulated, was supplied with iron currency and huhui 湖會 notes, which became an official paper currency at the end of the 1170s (zone I V). To stabilize the system, the Ministry of Revenue once again did not hesitate to disburse 4 million ounces of gold and silver in 1175 to revive the value of huizi notes in circulation, thus demonstrating that silver had become the preferred reserve instrument for paper currency.

Budgetary Crises in the Thirteenth Century (1205–1276) Renewed war with the Jin empire in 1205–1207 inaugurated a period of budget crises linked to a semipermanent war economy that lasted until the end of the dynasty. In the southeast, prices soared due to a new monetary depreciation induced by the circulation of some 140 million guan of huizi notes of varying 352

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

Table 9.3 Emission of huizi paper currency Issue Term of Quantity of notes number circulation (millions of guan) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

1168–1171 1170–1173 1171–1180 1173–1183 1180–1186 1183–1189 1186–1195 1189–1198 1195–1204 1195–1204 1204–1211 1204–1211 1207–1211 1211–1234 1211–1234 1231–1240 1234–1264 1240–1276

Total volume of notes in circulation at time of issue (millions of guan)

10 10 10 10 10 18 23.23 ? 30 ? 36.33 47.58 55.48 112.63 119.80 133.55 139.86 ?

10 10 10 10 20 28 [41.2]a [46.5] [53.2] [60.0] 83.9 83.9 139.4 232.4b 232.4b 365.9 273.4 ?

a Numbers in brackets are estimates based on the assumption that the quantities of issues 8 and 10 were the same as the previous issues, 7 and 9 respectively. b Beginning with issue 14 the Song government abandoned the principle of fixed terms of expiry and instead repeatedly issued the same notes over an extended period of time. Thus the figures in this column for issues 14–15 indicate the total volume of notes issued over their lifetime (1211–1234), not when they were first issued. Source: von Glahn, “Origins of Paper Money in China,” Table 4.1, p. 77

terms of expiry, a surfeit of currency that could not be withdrawn after the cessation of the war (Table 9.3). The authorities in Sichuan declared a deficit of 7 million guan, for which they expected recompense from the central government. In the late 1210s, as the deficits deepened further following another conflict with the Jin in 1215, the debate among policymakers focused on raising revenues. The administration then decided to increase tax surcharges and to amplify commutation operations and revenue transfers among regions. The demise of the Jin Empire in 1234 – which came about through an illconsidered alliance of the Song with the new rising steppe power, the Mongols – swiftly turned into a clash with the armies of the Mongol khan Ögödei. To compensate for “annual revenues which no longer covered more

353

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

than ten months’ expenditures,”3 the issue of paper money expanded more than ever, to as many as 100,000 guan of notes printed daily. In 1234 more than 300 million guan of notes were in circulation, rising to more than 500 million guan in the early 1240s. The depreciation of paper currency, which in effect now circulated without expiration, particularly upset the soldiers, who feared that they could not divest themselves of notes that constituted half of their pay. The huizi notes became officially inconvertible in 1247. In 1238, to defray outlays for strengthening the defensive lines in the Jing– Hu region, 9 million guan of huhui notes were added to the 2.5 million guan already in circulation since 1212. The controller-general of Hu–Guang, Jia Sidao, nonetheless was able to redeem some of the surplus notes in circulation in 1241 and thus maintain price stability throughout the region. But in 1246, in the aftermath of the conflict, the general command affirmed that it lacked sufficient funds for public purchases of grain, and thus was incapable of provisioning the troops. The revenue problem was general, as in the same year the central government declared that its income could cover no more than half of its expenditure obligations. The answer was partly technical. The government reinstalled the budget department (guoyongsi), which had been abolished in 1207. The lack of information was not, however, resolved, since local authorities, confronted with unpredictable requisitions and allocations from the central government, were forced to improvise their tax collection policies. The flood of depreciated paper money, which had made it possible to cope with budgetary difficulties, triggered galloping inflation. The solutions were radical. In 1247, beset by military deficits, the central government forced local officials to resort to collecting advance payments of regular taxes under the name of “advance loans” (yujie 預借), with the consequence that as soon as the following year some levies were collected on taxes due in 1254. Another measure made “harmonious purchases” (hedi 和糴) – procurement of grain for military provisions from the private market – compulsory, even though local officials often ignored the regulation requiring abiding by market prices in making purchases. The policies adopted by the state monopolies bureaus, which found themselves unable to compensate the salt producers, exacerbated the discontent in the salt-producing regions: surplus salt was declared contraband, and purchase of monopoly salt was made compulsory in the form of household quotas.

3

“Yiwei guanzhi ce” 乙未館職策,QXJ, 1:24a–b.

354

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

In addition, at least since the reign of Ningzong (r. 1194–1224), the concentration of landholdings had become a persistent preoccupation. In 1234, Liu Kezhuang, observing that “gobbling up the resources of hundreds of families, relentlessly aggrandizing the cultivated lands of several circuits, benefiting from incomes on the order of a million bushels of grain per year – such things are absolutely unprecedented,” asked that only the great landowners be subject to the obligation of “harmonious purchases.”4 The following year Wang Mai proposed the idea of state confiscation of lands. Yet the most powerful landowners, who were in a position to aggrandize arable lands, effectively evaded the twice-a-year tax. The intensification of military operations and the prolonged loss of rich territories to the Mongols, including Lizhou prefecture in Sichuan in 1253, compounded the lack of revenues. The deficit in land taxes had a direct impact on the supply of grain and fodder to the armies, which were increasingly dependent on “harmonious-purchase” procurements. Emperor Lizong (r. 1224–1264) finally realized the extent of the problem in 1247, when he entrusted Shi Zhaizhi with the task of developing wastelands or public lands (gongtian 公田) in the southeastern regions. Alluvial lands, polders, and confiscated properties were placed under the authority of a Land Office (tianshisuo 田事所). But it was the rise to power of Jia Sidao, hero of the brief war unleashed in 1258 by the Mongol khan Möngke in Sichuan and the middle Yangzi region, as chief councilor in 1260 which allowed the court to tackle the dysfunctional fiscal system. Jia conducted two major reforms. The first concerned the information system, which was strengthened by auditing procedures for the budgets managed by officials in charge of the armies. In fact, while military spending represented nine-tenths of state expenditures, some of the plenipotentiary generals and governors did not hesitate to overstate their numbers and to divert supplies into private commerce and moneylending. The other, better-known, measure aimed to reassert central control over the land tax by implementing a proposal to redeem lands held by official households (guanhu 官戶) according to a scale calculated on the basis of their household rank. In 1263 the chief councilor seized the opportunity to revise the cadastral survey as part of the “public-land” (gongtian 公田) initiative, whose principles had been formulated even before he came to power. First tested in several prefectures of the rich region of Zhejiang, where Jia Sidao set an example for his fellow officials by donating 10,000 mu of his own property, 4

HCQJ, 51:6b.

355

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

the reform was soon extended to around 3 million mu (more than 230,000 hectares) placed under the authority of a State Lands Bureau (guantiansuo 官 田所). Beyond its geographical extension to Jiangdong and Jiangxi, the policy of forced land sales soon encompassed ordinary landowners as thresholds were gradually reduced; in the end, anyone owning more than 100 mu was obliged to relinquish their “surplus” lands. The reform was accompanied by an “inventory calculation” (tuipai 推排) intended as a systematic census of unregistered properties on the model of the “surveying” (jingjie) policy conducted a century earlier. Unpopular because they were judged too severe, these reforms nonetheless partly attained their objectives: “harmonious purchases” diminished and the issue of money was significantly reduced between 1263 and 1268. The public-land expansion sustained the provisioning of the armies until 1275, at the height of war with the Mongols, which had renewed in 1264, the year of Lizong’s death. From the beginning of the 1270s, when the enemy had seized some forty prefectures in Sichuan, 10 million guan were devoted to consolidating the twenty or so prefectures still under Song control in order to protect the regions from Chongqing to Luzhou farther west, even though these regions “had been totally devastated, such that only one sub-prefecture survived in a prefecture, or only one canton in a sub-prefecture.”5 Four million guan were also allocated to consolidate defenses in the Huainan region, facing the Mongol strongholds. But public finances were eroded by the reduction in the tax base and the constant depreciation of the currency. In January of 1268, eight years before the fall of Hangzhou, the price of rice in the capital’s markets was already double that which the residents of Kaifeng had had to pay at the end of the terrible siege in January 1127.

The Emergence of a Mongol Fiscal System (1230–1271) The Mongol khans had to confront a major challenge to assert the universality of their imperial power in China: moving from an organization founded on predation and the redistribution of booty to one based on routine taxation. The conquest of the Jin Empire, completed in 1234, marked a major stage in this process. Joining the Mongol court shortly after the capture of the Jin Central Capital by Temüjin (Chinggis; d. 1227), the Khitan 5

XCYS, xia, 7a.

356

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

minister Yelü Chucai convinced Chinggis’s successor, Ögödei (r. 1229–1241), to retain in part the old Jurchen model of fiscal intendancies to make north China’s Central Plain the base of operations for his conquests. The “Han territories” were divided into ten administrative circuits (lu 路), each controlled by an office of tax levies (zhengshou keshuisuo 徵收課稅所), placed under the authority of a regional branch of the Department of State Affairs (xing shangshusheng 行尚書省) based in Beijing and directed by Yelü and two “companions” (nököd) of the emperor.6 This organization, which ensured the centralization of revenues and control of officials, was put into effect only after 1236, once a household census (kuohu 括戶) was completed. Of the 1,736,000 households registered at the time, 900,000 were enrolled under the appanages (touxia 投下) of royal princes and the military nobility through personal subordination rather than attachment to the land, in keeping with Mongol practice. Consequently, much of the population was enslaved to the leaders of the conquering armies and escaped control by the civil administration subordinated to the central government. The census of 1236 divided the “Han” population into taxation categories corresponding to their professional occupations (civil servants, craftsmen and merchants, soldiers, salt workers) or religious affiliations (Buddhist monks, Confucian scholars, Nestorian and Muslim communities). The religious groups, often great landowners engaged in commercial or artisanal activities, owned inns, mills, or pawnshops and, like military and artisan households, had to pay a property tax (dishui 地稅). But this tax yielded far less revenue than the capitation tax (dingshui 丁稅) which, at least until the reign of Khubilai (Emperor Shizu, r. 1260–1294), was owed by every adult male (probably beginning at age fifteen), while the nomad populations contributed 1 percent of their livestock. In addition, the population owed regular labor services (chaiyi 差役) to the Mongols for maintenance of the postal relay stations, provisioning the military, and supporting Mongolian emissaries. They also had to pay irregular fees according to the needs of their Mongol masters, the qubchir levy owed to the khan and the nobles, which became the baoyin 包銀 levy assessed in silver beginning with the reign of the khan Möngke (r. 1251–1259). The return to an active land market following the unification by Khubilai resulted in fiscal confusion: many landowners became responsible for paying two kinds of levies, the capitation tax and the land tax. From 1236 the property tax rate of two to four shi of grain per household had 6

In this context “Han” refers to the Chinese inhabitants of the former Jin territories, as opposed to the Mongols and Central Asian peoples (semuren 色目人) and also distinct from the “southerners” (nanren 南人) who lived under Southern Song rule until 1279.

357

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

been raised to two shi (equivalent to 190 liters of rice, or half the annual consumption of an adult male) per adult male, with the servile population taxed at less than half that rate. Twenty percent of state income was used to pay for administrative costs in China, while the court at Qaraqorum placed the remainder in the hands of the ortoq (“partners”), Central Asian merchants who acted as commercial agents of the Mongol nobility. The ortoq carried off much of this silver to West Asia, so that by 1260 north China was suffering from a chronic shortage of silver. In addition to the complexity and instability of the tax regime in the north – which was inseparable from the constant tension between the central government and the semi-autonomous appanages – the Mongols decided, following the conquest of the Southern Song in 1279, to retain the twicea-year tax in that region. Established on the old basis of land tenure, it included payments in kind (silk or other fabrics) and money in the summer, and then grain in the autumn. Although for some time the southern regions had not paid the summer tax, the land tax was regularly subject to tax commutations by which the authorities offset the effects of inflation caused by demanding payments in silver. Again, regional disparities in tax collection rates were the rule. Rather than a symptom of disorder, however, the organization of the Yuan dynasty into prefectures and “branch executive departments” (xingsheng 行省) of the central government – the future provinces – was the result of a lasting compromise between the centralized control of resources needed for military power, the pillar of the Eurasian empire, and the “de-concentration” of fiscal power at the regional level, entrusted to ten military commands (xuanfusi 宣撫司).7

Political Economy under the Yuan Close ties between the government and the world of commerce were a main feature of this entire period. From 1239 Ögödei had entrusted tax collection in north China to a Muslim merchant, Abd al-Rahman, who persuaded Ögödei to adopt a tax-farming system by promising to deliver 44,000 silver ingots to the government in the very first year. In 1253 the authorities decided to regulate the activities of the ortoq. These merchants, often Muslims valued 7

Li Zhian distinguishes between “decentralization,” synonymous with strong regional autonomy, and “de-concentration,” allowing the central government to strengthen itself by allocating sufficient resources to regional authorities. See Li Zhian 李治安 et al., Tang Song Yuan Ming Qing zhongyang yu difang guanxi yanjiu 唐宋元明清中央與地方關 係研究 (Tianjin, Nankai daxue chubanshe, 1996).

358

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

for their knowledge of local laws and customs, henceforth enjoyed the privileges of official intermediaries, responsible for reaping profits – especially through moneylending – from the capital entrusted to them by the appanage holders of the Mongol elite. Even more radical was the decision of Khubilai to place state finances under the control of a Persian merchant, Ahmad, from 1262 to 1282. To finance the emperor’s conquests and great public works – including building the new capital of Dadu (Beijing), beginning in 1264, and dredging the Grand Canal, completed in 1293 – Ahmad increased the number of households subject to taxation by 500,000 and achieved a considerable leap in revenues from commerce and public monopolies, which he placed under the control of regional inspectors (tijusi 提舉司). Commercial taxes rose tenfold, and revenues from the salt monopoly more than sixfold, between 1271 and 1286. Despite the execution of Ahmad on charges of corruption in 1282, his policies were largely continued by Lu Shirong and then the Tibetan Sangha, who fell from power in 1291. Confronted by the first inflationary spiral due to the massive issue of paper money that followed the conquest of the south in 1279, they too looked for new sources of revenue: increasing monopoly income by manipulating the price of licenses, strengthening the monopoly on liquor, and imposing surtaxes on maritime commerce under the monopoly control of the ortoq. By 1285 the salt monopoly and commercial taxes comprised 85 percent of total monetary revenues. The government’s internal exchanges were organized mainly around the transfer of resources from the south to the north by utilizing the Grand Canal, which was capable of transporting some 380,000 tons of tribute grain annually, and through the development of maritime shipping. From 1283 maritime traffic was conducted principally between the ports of Liujia (located near present-day Shanghai) and Zhigu (modern Tianjin). After 1292, maritime routes became reoriented to ports along the southeast coast from Wenzhou to Fujian to take better advantage of ocean currents and to increase the volume of grain shipments to the north, which reached 270,000 tons in 1329. Anxious to increase tax revenues as cheaply as possible as well as to facilitate commercial and financial transactions, the Mongol authorities decided, even before the founding of the Yuan, to monetize taxation by adopting the paper-money systems already in wide use under the Jin and the Song. While paper monies had circulated in certain regions since 1236, the new Zhongtong yuanbao 中統元寶 notes issued by Khubilai from 1260 359

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

became the sole legal tender and were accepted in state payments. According to the official exchange rate, two 1,000-coin notes of Zhongtong yuanbao were worth one ounce of silver, which circulated in the form of ingots. The use of paper money was reinforced by the ban on employing silver in private commerce enacted in 1263 (eventually rescinded in 1311). Between 1260 and 1269 the emission of paper currency was kept within a limited annual quota of 700,000 ding 錠 (a unit nominally equivalent to 100,000 bronze coins) and was backed by reserve funds that guaranteed the convertibility of notes. But the conquest of the south was followed by the injection of more than 5 million ding of paper notes into the money supply between 1276 and 1280 without any corresponding increase in hard-currency reserves. Rather than considering these decisions as facile and inconsistent responses of a “conquest dynasty” confronted by the problems posed by perennial budget deficits, current historical scholarship offers a much more convincing explanation that situates Mongol imperial institutions in a broader Eurasian perspective.8 To maintain the military capacity of the appanages and the nobility in each of the empire’s four khanates, the Mongols would have to seek the financial integration of the empire on a continent-wide scale using the only instrument capable of creating monetary equivalents: silver. But this “Eurasian silver century,” which ended precisely at the time of the fall of the Yuan, depended not only on China’s wealth of silver, but also on its paper currencies, on which the tax system was based. Paper currency freed up silver, which – as archaeological excavations have shown – circulated among all the aristocratic appanages. The reforms of Khaishan (Emperor Wuzong, r. 1308–1311), which introduced a new paper currency, the Zhida yinchao 至大 銀鈔, denominated in weights of silver, appear to have been an attempt to stabilize the monetary system by basing it on the value of silver metal, as was the case in the rest of Eurasia. The failure of this reform led to the resumption of excessive issue of paper notes and the permanent weakening of the dynasty’s public finances. From 1344, after the great floods of the Yellow River, China was beset by rampant inflation, and this fragility posed a grave handicap to efforts to suppress the rebellions that erupted during the 1350s. Yet although these rebellions brought about the demise of the Yuan dynasty, they could not destroy Mongol military power, which continued to threaten the Ming Empire.

8

See Akinobu Kuroda, “The Eurasian Silver Century, 1276–1359: Commensurability and Multiplicity,” Journal of Global History 4.2 (2009), 245–69.

360

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

The Early Ming Reversal The Mongol conquest disrupted the fiscal balance established in the Southern Song and subjected China to predatory tribute exaction. But the epochal reversal of Song fiscal trends and the rupture of the synergy between the state and the market occurred after the restitution of Chinese rule in the late fourteenth century. Hongwu (r. 1368–1398), first emperor of the Ming dynasty, was determined to eradicate what he regarded as the pernicious influence of Mongol rule and to restore the institutions and values of the agrarian society enshrined in the Confucian classics. In so doing Hongwu repudiated the market economy that had developed during the Tang–Song transition. Hongwu’s policies ultimately foundered in the face of ineluctable economic forces that rekindled private enterprise and the revival of the market economy. But the early Ming marked a crucial historical disjuncture that paved the way toward a new, alternative conception of the fiscal role of the state. In spirit, if not precisely in institutions, the early Ming regime harked back to the fiscal paradigm – predicated on direct taxation of an agrarian population – of the early empires. Hongwu established a tax system based on in-kind payments and labor services, introduced a fiat paper currency in lieu of metallic money, and shifted most of the administrative burden of local government directly onto the shoulders of landowners. Hongwu did make certain concessions to the changed socioeconomic realities of his day. Apart from confiscating properties of the Yangzi delta (Jiangnan) region’s great landowning families, he did not attempt an empire-wide nationalization of land. In addition, rather than imposing universal military conscription, Hongwu designated a separate caste of self-supporting military households (junhu 軍戶) to lessen the burden of military expenditures. Many of Hongwu’s policies failed abysmally, and his successors abandoned his attempt to re-create a command economy directly subordinate to the ruler’s will. Apart from his ideological convictions, Hongwu’s policies also were motivated by a practical objective: commandeering the wealth of Jiangnan to rebuild an empire ravaged by decades of war and destruction. China’s population fell by at least 15 percent, and perhaps as much as a third, between 1340 and 1370 (see Cao’s chapter in this volume). But the Jiangnan region emerged from the late Yuan civil wars with its population largely unscathed and its pre-eminent position in the national economy intact. Hongwu’s decision to establish his capital at Nanjing reflected the economic reality

361

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

that the resources needed to consolidate his rule could only come from the Jiangnan region. Initially Hongwu sought the co-operation of the local elite of Jiangnan in his imperial project. The revenue system he enacted in 1371 designated the wealthiest landowners in each county as tax captains (liangzhang 糧長) with responsibility for collecting and delivering tax grain to the capital. By 1380, however, Hongwu became convinced that the Jiangnan elite, both as government officials and as private landowners, was the real obstacle to his goals. He then reversed course, purging thousands of officials and seizing their landholdings. Hongwu revised his vision of social reform: the greatest priority, as he now saw it, was to free family farmers from exploitation by wealthy landowners and corrupt officials as well as the vicissitudes of the money economy. Many of the tax captains had been swept up in Hongwu’s campaigns against corruption. Although the tax captain system was retained, in 1381 Hongwu introduced a new institutional framework of rural control known as lijia 里甲, which organized rural society into groups of 110 households (li 里) under the leadership of ten headmen who served in rotation over a ten-year period. The li headmen were entrusted with broad responsibility for local governance, including tax collection and assignment of labor services. Despite Hongwu’s antipathy for the Mongols, his government also followed the Yuan precedent of registering the population into hereditary occupational groups. Civilian households, the great majority of the population, were liable for taxes levied in grain and other goods as well as statutory labor service; military households (about one-fifth of the total population in the early fifteenth century) were stationed at garrisons and given lands to farm when not engaged in military duties; artisan and saltern households were obliged to provide either labor services or goods to the government. Land and population surveys carried out in 1387 and 1393 provided the benchmarks for refining the lijia system and establishing permanent tax quotas. Lijia registration and tax quotas remained unchanged after 1393, and thus increasingly diverged from social and economic reality. In keeping with Hongwu’s intention to reduce the logistical and administrative costs of government, the Ming fiscal system generated a low level of income compared to the Song. The land tax, collected in grain, amounted to no more than 5–10 percent of yields. No revenue was earmarked for military expenditures, since the military households were expected to be self-supporting. After a brief attempt to restore a sound bronze currency, Hongwu instituted a new type of inconvertible paper currency (baochao 寶鈔) while 362

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

simultaneously banning the use of gold and silver as money. But from the outset the paper currency suffered from steep depreciation; by 1394 baochao were discounted by nearly 80 percent. The Ming state also throttled the flourishing maritime trade that had developed during the Song and Yuan periods. In 1374 Hongwu prohibited private merchants from engaging in overseas commerce, while allowing some trade under a highly regulated form of tributary diplomacy with foreign rulers. Although the early Ming government succeeded in reducing the burden of administrative expenses, it reverted to a heavy reliance on agrarian sources of state income. Fixed land-tax quotas severely limited the state’s ability to capture new revenue as the agricultural economy recovered and expanded. Unlike earlier dynasties, the Ming failed to separate state revenues from the personal income of the emperor, resulting in rampant abuses, especially at the hands of the eunuch cadre that dominated the palace administration. The extravagant imperial ambitions of Hongwu’s son, Emperor Yongle (r. 1402–1424), exacerbated the state’s fiscal troubles. In the course of building a new capital at Beijing, conducting prolonged campaigns of conquest against Vietnam and the Mongols, and launching immense naval expeditions across the Asian seas under the command of Admiral Zheng He, Yongle saddled the Ming state with enormous deficits. Moreover, the initial success with the military farm system – which generated almost as much grain revenue as the regular land tax in 1403 – swiftly melted away under the burden of onerous taxation and rampant evasion, forcing the government to find other means to provision its armies. Efforts to defray these costs through profligate issue of the inconvertible baochao paper notes eviscerated their value. By the 1430s baochao were accepted at only 2 percent of their face value, and the government resorted to paying civil servants in silk and cotton cloth and exotic imported goods such as pepper and sappanwood. The state’s inability to coin money further complicated the monetary predicament caused by the failure of the baochao paper money. Shortages of copper forced the government to shutter its mints by the mid-1430s, and for nearly a century afterward the Ming issued no new bronze coin. Confronted with this intractable monetary crisis, the Ming court grudgingly shifted from payments in kind and baochao to a new fiscal regime based on silver.

Silverization of the Ming Fiscal Administration Although the state refrained from issuing silver currency, it acceded to the prevalence of silver in private exchange and converted most state exactions to 363

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

payments in silver. Initially, the commutation of taxes to silver, which circulated in the form of ingots that varied widely in weight and fineness, proceeded in piecemeal fashion. But after 1550 – when imports of foreign silver began to flood into China – the silverization of state finance rapidly accelerated. The adoption of a silver standard in taxation, which extended to labor services as well as goods, represented a complete reversal of the Ming founder’s fiscal principles. In 1433 the government – pressed by landowners protesting proliferating tax surcharges – agreed to convert the grain tribute of Jiangnan to a fixed payment in silver. In 1436 this practice (later designated “gold-floral silver,” jinhuayin 金 花銀) was extended to the other southern provinces. In 1442 the government established the Taicang Treasury 太倉 as the depository for the Ministry of Revenue’s silver revenues, which included income from the salt monopoly, customs duties, and various excises and fees, as well as a share of the gold-floral silver. Most gold-floral silver revenues were reserved for the maintenance of the imperial household, but the court regularly allocated 20–40 percent of this income to military expenditures. A portion of military salaries was disbursed in silver beginning in 1447, and by the 1480s it became general practice to pay soldiers stipends in silver in addition to their grain and cloth rations. Labor services, too, were commuted to payments in silver. By the end of the fifteenth century Hongwu’s key institutional innovations in local governance – the lijia village household system and the liangzhang tax headmen – were essentially defunct. Scattered efforts to relieve the burden of lijia service by transferring the assessment of service duties from individuals to property led to the creation of an empire-wide program of “Equalized Service” (junyao 均徭) in 1488. This trend culminated in the “Single-Whip” (yitiaobian 一條鞭) reforms of the late sixteenth century, which converted most labor services into a unified silver payment assessed on landholdings. The hereditary occupational households inherited from the Yuan had atrophied as well. Merchants and artisans warmly welcomed the conversion of requisitions for goods and labor services to payments in silver. But when the central government imposed new requisitions on the commercial classes (the so-called “Mines and Taxes” levies) in 1599–1604, violent protests erupted in many cities, forcing the government to rescind them. Although Ming statesmen abandoned the stridently antimarket vision of Emperor Hongwu, they remained distrustful of private enterprise and refrained from imposing commercial or consumption taxes whose costs would ultimately be borne by farming families. The salt monopoly was the main exception. From the outset the Ming state resurrected the defunct Song ruzhong policy (renamed kaizhong 開中), rewarding merchants who delivered 364

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

supplies to frontier armies with salt certificates. But already by the 1420s the number of salt certificates issued by the state greatly exceeded the available supplies of salt, forcing merchants to wait years or even decades to redeem their certificates. Various measures to expedite the redemption of salt licenses failed, and periodic attempts to force holders of older licenses to accept baochao in payment instead further compromised their value. By the end of the fifteenth century the frontier delivery system had broken down, to be replaced by new procedures under which frontier merchants sold their licenses to merchants in Yangzhou and other major salt-trading centers. The Yangzhou salt merchants, mostly Huizhou men, had the ample capital resources needed to survive the vicissitudes of what had become a highly uncertain business. The final resolution to the problems of erratic supplies and endemic mismanagement came in 1617, when the government instituted a franchise system that awarded exclusive and permanent salt trading rights to syndicates of Yangzhou merchants. Subsequently the Yangzhou salt merchants became the wealthiest men in the empire, yet the state reaped only modest income from the salt monopoly. Although the salt monopoly became, along with the gold-floral silver, the most significant source of revenues for the Taicang Treasury (each delivered roughly one-third – about 1.3 million taels per year – of Taicang receipts c. 1580), salt revenues and other indirect taxes generated only a fraction of the revenue obtained from the land tax (Table 9.4). The conversion of land-tax revenues to silver payments proceeded unevenly, with considerable regional variation. Tax records, which usually retained the original in-kind units of account, provide only incomplete information on commutations to silver. In 1527, when bad harvests induced the government to commute many grain tax payments to silver, silver revenues amounted to 27 percent of total land-tax income. In 1549, total Taicang silver revenues amounted to 3.95 million taels, of which only one-quarter was derived from land taxes. But conversion of land taxes to silver payments accelerated rapidly after 1550, notably under the stewardship of Zhang Juzheng, chief civil minister at the Ming court from 1572 until his death in 1582. In addition to being the key proponent of the Single-Whip reforms, Zhang also conducted an empire-wide cadastral survey of landholdings and tax assessments in 1580–1583. Ray Huang proposed that by 1600 roughly 80 percent of land taxes (and the service levy connected to the land tax) had been converted to silver payments.9 Although Huang’s estimate is highly conjectural, we also find a corresponding shift 9

Ray Huang, Taxation and Governmental Finance in Sixteenth-Century Ming China (New York, Cambridge University Press, 1974), p. 175. But Huang’s estimate of silver revenues is twice as high as the highest reported figure for total silver revenues,

365

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

Table 9.4 Ming silver revenues, c. 1580 Revenue source

Silver revenues (millions of taels)

Land tax and service levy commutation Salt revenues Commercial taxes and excises Commutations of government requisitions Administrative income

25–30a 1.29 0.94 1.69 1.15

Estimate of magnitude; no comprehensive figures. Source: Ray Huang, Taxation and Governmental Finance in Sixteenth-Century Ming China (New York, Cambridge University Press, 1974), pp. 175, 214, 263 a

Table 9.5 Budgeted military expenditures for the northern frontier commands, 1531–1602 Number of frontier commands 1531 1549 1582 1593 1602

7 9 13 13 13

Number of military personnel 371,374 459,180 686,523 651,665 645,911

Silver expenditures (taels)

Grain allocations (shi)

3,362,386 5,677,282 8,279,165 7,154,630 6,676,271

3,004,523 841,863 1,983,646 1,900,000 2,593,300

Source: Lai Jiancheng 赖建诚, Bianzhen liangxiang: Mingdai zhonghouqide bianfang jingfei yu guojia caizheng weiji, 1531–1602 边镇粮饷：明代中后期的边防经费与国家 财政危机 (Hangzhou, Zhejiang daxue chubanshe, 2010), Tables 12.1–12.6, pp. 275–9

beginning in the 1550s from supplying frontier armies primarily with grain rations and other in-kind payments to allocations of silver funds. In the late sixteenth century, at least three-quarters of government allocations for military garrisons on the northern frontier were disbursed in silver (Table 9.5). The expansion of silver payments in lieu of in-kind taxation of goods and services after 1550 was made possible by massive imports of foreign silver. By the early sixteenth century, growth in agricultural production 14.61 million taels, cited in the national accounts registry compiled by Zhang Juzheng’s administration in 1578. See Taniguchi Kikuo 谷口規矩雄, “Minmatsu no kinkagin ni tsuite” 明末の金花銀について, in Umehara Kaoru 梅原郁 (ed.), Chu¯goku kinsei no ho¯sei to shakai 中國近世の法制と社會 (Kyoto, Kyo¯to daigaku jinbun kagaku kenkyu¯jo, 1988), p. 489.

366

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

and regional specialization in the manufacture of goods such as cotton, silk, porcelain, iron wares, and books had stimulated market development and long-distance trade across the empire – and beyond. Although Hongwu’s ban on private maritime trade remained in effect, Fujian seafarers easily evaded the lax net of enforcement to embark on trading ventures to Japan, Ryukyu, and Southeast Asia. The allure of overseas trade abruptly rose with the boom in Japanese silver mining in the 1530s, followed by the arrival of Portuguese trading vessels laden with silver from Spain’s American colonies. The voracious demand for silver undermined the Ming court’s interdiction of private overseas trade. Friction between the Ming state and clandestine seaborne traders (dubbed Wokou or “Japanese pirates,” although most were Chinese) erupted into outright warfare in the 1540s–1550s. Devastating Wokou attacks on coastal areas and pressure from merchant and official constituencies in the southeastern provinces finally persuaded the Ming court to relax the trade ban in 1567. The founding of the Spanish colony at Manila in 1571 opened a more direct conduit for the export of American silver to China across the Pacific Ocean. In the second half of the sixteenth century, China imported on average almost fifty tons of silver per year. Estimates of China’s annual silver imports in the first half of the seventeenth century range from ninety to 125 tons. Approximately 60 percent of this silver came from Japan, and the rest originated in Peru and Mexico. We lack comprehensive national accounts for government revenue in the Ming. Most revenue collected at the local and provincial levels was reallocated without being deposited in the treasuries of the central government. Silver revenues deposited in the capital or frontier treasuries averaged slightly more than 5 million taels annually during the reign of Emperor Wanli (r. 1572–1620), only a fraction of Huang’s estimate of total silver income. Originally conceived as a reserve fund that also defrayed most palace expenditures, over the course of the sixteenth century the Taicang Treasury became the mainstay of financing frontier defense. New military crises – the Wokou pirate attacks on coastal cities and Mongol incursions under Altan Khan, which penetrated to the outskirts of Beijing in 1550 – quickly drained the reserve funds, and in the 1550s–1570s the Taicang Treasury incurred substantial deficits (Figure 9.1). Zhang Juzheng succeeded in curbing profligate spending during his decade as chief minister, but chronic deficits returned after 1583. The costly military campaigns of the 1590s – including the expeditionary armies sent to thwart the Japanese invasion of Korea – raised 367

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn 7

Millions of Taels

6 5 4 3 2 1 0 1520

1530

1540

1550

1560

1570

Revenues

1580

1590

1600

1610

1620

Expenditures

Figure 9.1 Revenues and expenditures of the Taicang Treasury, 1518–1620 Source: Quan Hansheng 全漢昇 and Li Longhua 李龍華, “Mingdai zhongyehou Taicang suichu yinliangde yanjiu” 明代中葉後太倉歲出銀兩的研究, in Quan Hansheng, Zhongguo jindai jingji shi luncong 中國近代經濟史論叢 (Taipei, Daoxiang chubanshe, 1996), Table 7, p. 306

Taicang disbursements to twice their previous level. After 1590 military expenditures accounted for at least 85 percent of total Taicang outlays. But it was the Manchu invasions beginning in 1618 that sent military costs spiraling out of control. During the next two decades the Ming suffered repeated military reversals. Desperate to stave off utter collapse of its defenses, the court enacted a rapidly escalating series of emergency taxes that quintupled the Taicang revenues, which rose from 4 million taels in 1618 to 21.5 million taels in 1641. War expenditures still outran state income, however, while the tax increases fanned popular discontent. Poor harvests in the mid-1630s sparked large-scale rebellions in north China. Devastating floods, famines, and outbreaks of epidemic disease during 1638–1642 crippled both the national economy and the Ming state. In 1644 a peasant rebel army seized Beijing, followed several months later by the Manchu occupation of the capital and the founding of a new conquest dynasty, the Qing.

Fiscal Governance under Manchu Rule Departing from the anti-market principles of the Hongwu Emperor, the late Ming leadership gradually shifted from in-kind taxation to silver payments, a transition accelerated by the influx of foreign silver from the sixteenth century onward. At the same time the principle of direct taxation of land and labor was preserved. Apart from a few episodes of

368

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

predatory attacks on commercial sources of wealth – notably the notorious Mines and Taxes exactions of 1599–1604 – the government generally favored minimal interference in industrial and commercial activities. The paradigm of fiscal governance that emerged in the latter half of the Ming – which was maintained by the subsequent Qing dynasty as well – was grounded in the Neo-Confucian ideological commitment to the well-being of the family farm under the aegis of a paternalist agrarian state.10 Fiscal policies of the paternalist agrarian state aimed to ease the burden of state exactions by maintaining a low level of direct taxes on land and labor, with minimal taxation of commerce, industry, or consumption (with the signal exception of the salt monopoly). In principle, the state also was dedicated to improving the people’s livelihood through, for example, investments in famine relief and flood control. Yet in contrast to the activist conception of the state that prevailed in the Song, much of the actual responsibility for popular welfare was ceded to local public and private initiative. The laissez-faire attitude toward commerce adopted by the Qing rulers encouraged Smithian dynamics of market expansion, division of labor, and regional specialization. But at the same time the limited fiscal capacity of the Qing state deterred investments in public goods that would help sustain economic growth. Similar to earlier Chinese dynasties, the first priority of the Qing state was to restore a stable agricultural base. By 1683 the Manchus had secured control of the former Ming territories and Taiwan, although the Manchu Empire would be greatly enlarged during the eighteenth century through conquests of Mongolia, Tibet, and the “New Frontier” carved out of Uyghur- and Mongol-populated Central Asia. The Qing emperors largely preserved the Ming administrative structure both at the capital and in the provinces, but they also introduced several new military and fiscal institutions. The Manchus originally organized their armies into units known as “banner armies” (eight altogether, separated into Manchu, Mongol, and Han Chinese divisions). Following the completion of the conquest, banner units were stationed at strategic points throughout the empire to maintain domestic control. Tracts of land were set aside to support the banner soldiers, who became a hereditary military caste. The Manchus also created a new agency, the Imperial Household Department (neiwufu 內務府), to manage the affairs 10

On the characterization of the Ming–Qing model as an “agrarian paternalist state,” see R. Bin Wong, China Transformed: Historical Change and the Limits of European Experience (Ithaca, Cornell University Press, 1997).

369

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

and finances of the emperor and the imperial family. The Neiwufu’s activities ranged from managing the banner lands and various monopolies (such as salt, Manchurian ginseng, copper imports from Japan, and the imperial silk and porcelain procurement agencies) to printing official editions of scholarly works. Under Qing rule the Ministry of Revenue and its provincial-level intendants strengthened central control over fiscal resources. The Single-Whip tax reforms in the late Ming resulted in the compilation of provincial “comprehensive accounts of taxes and service levies” (fuyi quanshu 賦役全書) that not only established quotas for tax payments but also created minutely detailed budgets for local government expenditures ranging from staff salaries to sundry supplies such as candles and paper. Under the Qing the central government claimed more than 80 percent of revenues collected at the local level. Locally collected revenues were remitted to provincial fiscal authorities, but it was the central government that determined allocations to local governments and transfers of revenues to the capital or other provinces. For example, in 1713 Shanxi province delivered 33 percent of its income to the capital and transferred 30 percent for provisioning military units in other provinces, while retaining 37 percent for intra-provincial expenditures.11 Generally only 5–15 percent of locally generated revenues were devoted to local government expenses. Thus the chronic underfunding of local government that had become a vexing problem in the late Ming persisted under Qing rule, forcing local officials to resort to extra-statutory fees and exactions (huohao 火耗) as well as customary gifts from subordinates (lougui 陋規) to make ends meet. Emperor Yongzheng (r. 1722–1735) sponsored fiscal reforms that regularized the huohao exactions by turning them into fixed surcharges (haoxian 耗羡) on the statutory land tax. The proceeds from these revenues were assigned to provincial and local officials and divided among retirement of outstanding deficits; supplemental allocations (known as “nourishing virtue” (yanglian 養廉) funds) to cover the costs of local and provincial administration; and expenditures on public works, famine relief, and other emergency needs (gongfei 公費). But under Yongzheng’s successor, Qianlong (r. 1735–1795), the central government assumed control of haoxian revenues, depriving local governments of the flexibility to utilize these funds efficiently in response to changing needs.

11

Iwai Shigeki 岩井茂樹, Chu¯goku kinsei zaisei shi no kenkyu¯ 中國近世財政史の研究 (Kyoto, Kyo¯to daigaku gakujutsu shuppankai, 2004), p. 91.

370

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

In comparison to earlier dynasties, the Qing state’s presence at the village level was minimal. Rather than undertaking its own land surveys, the Qing government relied on the Ming cadastral registers compiled in 1580–1583 to allocate tax quotas. In 1712 the Qing established permanent quotas for silver payments in lieu of labor service (the ding levy) based on the 1711 census returns. In the following year the Kangxi Emperor (r. 1661–1721), in an act of imperial magnanimity, decreed that the land tax quotas would be permanently frozen at the level of the 1711 assessments. In 1729, Yongzheng formally merged the ding levy into the land tax on an empire-wide scale. Henceforth individual landowners paid a single aggregated tax in silver. This step completed the process of shifting taxation from persons to land. The Qing tax reforms obviated the need to compile detailed information about the composition of households, rendering household surveys obsolete, since the family no longer was a unit of taxation. By 1772 the Qing state had formally halted household registration. The conjugal family remained the fundamental socioeconomic unit in Chinese society, but it had attained unprecedented autonomy from the attenuated reach of the imperial state. Consequently, the Qing imperial government wielded at best indirect control over local society, relying on mediation by local elites. Since the final decades of the Ming dynasty, local patricians had steadily expanded their responsibilities in managing many aspects of local governance and social welfare, including irrigation projects, public security, famine and poor relief, schools, and temples. This trend accelerated during the eighteenth century. The Qing period also witnessed the proliferation of a wide range of new intermediate social institutions – including corporate lineages; merchant and artisan guilds; native-place associations; communal water control leagues; and an array of religious, fraternal, and philanthropic societies – to which the imperial state entrusted managerial responsibilities within local society. State officials and local elites shared essentially similar agendas of social control, popular welfare, and moral leadership. Still, the managerial roles assumed by local elites allowed freer rein to the pursuit of private interests in the guise of public welfare. One arena of public welfare in which the Qing government did intervene vigorously was food supply and famine relief. The Qing established a comprehensive network of “ever-normal granaries” (changpingcang 常平倉) in each county whose reserves were used not only to provide famine relief but also to stabilize grain prices. The state’s target for ever-normal granary stockpiles reached as high as 58 million shi (enough to feed 85 million adults for a month) in 1740. State purchases and stockpiling directed the flow of grain from surplus to deficit areas, but officials recognized the limits of bureaucratic intervention and 371

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

generally were disinclined to interfere with the private grain trade. Evidence from the mid-eighteenth century suggests that the state’s famine relief policies achieved success in moderating the volatility of grain prices and containing subsistence crises during years of dearth. The efficient operation of the state granary system and famine relief policies undoubtedly contributed to China’s sustained prosperity and demographic growth during the eighteenth century.

Revenue and Expenditure in Eighteenth-Century China At the time Kangxi established permanent land-tax quotas in 1713 the Qing enjoyed substantial budget surpluses. The land tax – now fixed, regardless of the actual productivity of the land – supplied roughly three-quarters of central government revenues, with a mere 16 percent coming from the salt monopoly and customs excises (Table 9.6).12 Despite the shift from in-kind taxation to payments in silver, the real per capita tax burden (measured in grain) fell to a mere 30 percent of the fifteenth-century level.13 The state’s presence in the national economy diminished as well. Although estimating total national income for any period of imperial Chinese history is highly speculative, it seems likely that during the Qing period the proportion of government revenue as a share of the national economy was far lower than in the Song, which relied heavily on taxation of commerce and consumption, and perhaps below the Ming level as well. Apart from the salt monopoly, operated under the franchise system inherited from the Ming, the Qing government interfered little in domestic commerce and industry. In the case of the salt monopoly, the several hundred merchants holding monopoly privileges amassed enormous fortunes from the transport and sale of salt. Still, the government garnered the lion’s share of profit from the salt monopoly. According to Wang Chongyun’s detailed study, operating costs consumed 43 percent of gross 12

13

Yeh-chien Wang estimated that in 1753 surcharges on salt, customs, and miscellaneous revenues amounted to 7.187 million taels. Wang also gives a far higher figure for haoxian tax surcharges in 1753 (10.439 million taels) than the 1766 figures show (3.5 million taels). Wang’s data for 1753 indicate that the land tax generated 73.4 percent of total Ministry of Revenue income, which he estimated at 73.8 million taels. Wang’s figures suggest that even though the 1766 data in Table 9.6 may underestimate surcharge revenues (and thus total revenues), the proportion generated by land taxes was fairly constant. See Yeh-chien Wang, Land Taxation in Imperial China, 1750–1911 (Cambridge, MA, Harvard University Press, 1973). Wu Hui 吴慧, “Ming Qing (qianqi) caizheng jiegouxing bianhuade jiliang fenxi” 明清 (前期)财政结构性变化的计量分析, Zhongguo shehui jingji shi yanjiu 中国社会经济 史研究 1990.3, 45.

372

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

Table 9.6 Central government revenues, 1766 (all revenue figures in millions of silver taels)

Revenue source Land taxes, rents, and grain tribute Haoxian land tax surcharges Salt gabelle Ginseng monopoly Internal and maritime customs Pawnbroking Regular contributions Miscellaneous local taxes Stamp taxes Brokerage and pawnshop licenses Marshes and fishing excises Mining excises Tea excise Total

Imperial Percentage Household Ministry of Total of total Departmenta Revenue 0.15

0.40 0.30 0.50 0.25

1.60

31.06 20.31b

51.52

72.4

3.50 5.75

3.50 6.15 0.30 5.92 0.25 2.00 0.86 0.19 0.16 0.15 0.08 0.07 71.15

4.9 8.6 0.4 8.3 0.4 2.8 1.2 0.3 0.2 0.2 0.1 0.1

5.42 2.00 0.86 0.19 0.16 0.15 0.08 0.07 69.55

a

Averages from 1760 to 1770. Portion of land tax collected in grain, converted to silver taels based on 1753 rates in Yeh-chien Wang, Land Taxation in Imperial China, 1750–1911 (Cambridge, MA, Harvard University Press, 1973), p. 70. Sources: Chen Feng 陈锋, Qingdai caizheng zhengce yu huobi zhengce yanjiu 清代财政 政策与货币政策研究 (Wuhan, Wuhan daxue chubanshe, 2008), Table 6-3, p. 369; Lai Huimin 赖惠敏, Qianlong huangdide hebao 乾隆皇帝的荷包 (Beijing, Zhonghua shuju, 2016), Figure 1-1, p. 66, Figure 2-1, p. 125, Figure 2-2, p. 129, Table 4-2, p. 199 b

salt sales while 44 percent went to various state organs, leaving 13 percent net profit for merchants.14 The salt cartel also was tightly intertwined with state finance through loans contracted from the Imperial Household Department and periodic compulsory contributions to the state treasury. But the vast majority of merchants and artisans inhabited a highly competitive market economy, and the profits generated by industry and commerce were almost wholly privatized. Nominally voluntary – but in fact compulsory – “contributions” (juanna 捐納) were a unique feature of the Qing revenue system. These funds derived from 14

Wang Chongyun 汪崇筼, “Qianlongqi Huishang zai Huai yanye jingyingzhongde huoli gusuan” 乾隆期徽商在淮盐业经营中的获利估算, Yanye shi yanjiu 盐业史研 究 2000.1, 13–22.

373

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

three sources: officials who were obliged to remit a portion of their salaries to provincial governments; individuals who purchased examination degrees and titles (and, on occasion, actual offices) to secure higher social status; and wealthy merchants who received special favors bestowed by the state, primarily the cartels of salt merchants and the Cohong merchants granted monopoly privileges to trade with European merchants at Guangzhou. In the 1740s juanna contributions from officials became regularized at fixed levels. Contributions from wealthy merchants oscillated, rising sharply at times of fiscal emergency – mostly in wartime, but also to defray extraordinary expenditures for famine relief, flood control, and unbudgeted palace spending such as the imperial tours by Kangxi and Qianlong. A report of 1768 claimed that the Yangzhou salt merchants alone submitted contributions amounting to nearly 1.2 million taels each year, but archival records indicate that contributions from all salt merchants averaged around 650,000 taels annually during the Qianlong Emperor’s sixty-year reign. In contrast to the Ministry of Revenue’s reliance on land taxes, nearly all of the income of the Imperial Household Department (neiwufu) was derived from commercial and financial sources. Banner lands were entrusted to estate managers, who served essentially as tax farmers and delivered fixed revenues – often as advanced payments – to the Neiwufu. Rents on banner estates were frozen along with general land taxes in 1713, and thereafter lagged well behind market levels. Land revenues generated only 100,000– 200,000 taels per year for the Neiwufu, while surpluses from maritime and inland customs and income from the ginseng and salt monopolies on average yielded 1.2 million taels annually in the latter half of the eighteenth century. The Neiwufu also utilized its surplus stocks of furs, ginseng, and tribute products as capital for moneylending. In the mid-eighteenth century, the Neiwufu operated twenty-six pawnbrokerages with combined capital assets of 727,000 taels, which generated annual returns of roughly 10 percent.15 But Qianlong, disappointed by the modest returns of the pawnbrokerages and ever suspicious of direct state involvement in profit-oriented enterprise, preferred to make loans to favored merchants, such as the cartels granted monopoly privileges in the copper trade with Japan, the fur trade with Russia, and the domestic salt industry, as well as the 200-odd merchants employed as purveyors of goods to the imperial household. In Qianlong’s day the Neiwufu lent enormous sums of money to these privileged merchants – as much as a million taels to a single individual – on long-term contracts of ten 15

Lai Huimin 赖惠敏, Qianlong huangdide hebao 乾隆皇帝的荷包 (Beijing, Zhonghua shuju, 2016), pp. 144–6.

374

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

Table 9.7 Central government expenditures, 1766

Expenditure (millions of taels) Percentage of total

Yellow River Civil adminiswater Military tration Palace conservancy

Postal stations Total

19.14

5.08

0.95

3.80

2.00

62.00

16.50

3.00

12.00

6.50

30.97

Source: Ni Yuping 倪玉平, Qingchao Jia-Dao caizheng yu shehui 清朝嘉道财政与社会 (Beijing, Shangwu yinshuguan, 2013), pp. 79–80

or more years. By the 1790s, salt merchants alone were remitting nearly 1.5 million taels annually in interest payments to the Neiwufu. The Fan family, which made its fortune through military procurement and importing Japanese copper, relied on generous credit from the Neiwufu to diversify into the fur, salt, and timber trades (including supplying timber for the construction of Qianlong’s Summer Palace) and real estate. But in 1783, after the Fans’ outstanding debts had swelled to 1.5 million taels, the government confiscated the family’s assets to recoup some of its losses. By this time many of the great salt merchants, beset by falling retail prices and growing state demands for contributions, had likewise suffered reversals of fortune and bankruptcies. Information on government expenditures is less complete than for revenues, but the available data suggest that in the mid-eighteenth century ordinary military expenses consumed at least half (and probably closer to two-thirds) of total central government expenditures (Table 9.7). The frequent military campaigns waged by Qianlong raised military expenditures to even higher levels. By the eighteenth century, the Green Standard corps – professional soldiers recruited from the Chinese populace – constituted the mainstay of the Qing armies. While the banners numbered roughly 200,000 soldiers (half of whom were stationed in Beijing), the Green Standard armies generally had 600,000 troops. Mindful of the debilitated state of the Ming armies at the time of the conquest, the Qing emperors were determined to ensure adequate incomes for their troops and officers in times of peace, and granted substantial bonuses for deployment on campaigns. Both the bannermen and Green Standard troops received stipends in silver as well as grain rations. As in the case of the Imperial Household Department, Yongzheng sought to capitalize on commercial opportunities to mitigate the state’s

375

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

financial burden. In 1729, under the rubric of “profit-generating enterprise” (yingyun shengxi 營運生息), Yongzheng allocated 1.2 million taels (roughly four to five taels per soldier) to the military as capital for commercial ventures. Military units – not individual soldiers – typically invested these funds in moneylending. The income from such ventures was intended to pay for soldiers’ marriage and funeral expenses, which often caused great hardship to military families, as they did to the civilian population. The policy produced mixed results; enterprises in the commercial heartland prospered, but those in remote regions often failed. In any event, Qianlong strongly disapproved of the entanglement of government personnel in the unseemly hustle for commercial profit, and in 1781 he discontinued this practice. Logistical and financial management of military campaigns was conducted through the civil bureaucracy. For each campaign the court established ad hoc commissions that drew up budgets, allocated revenues, and audited expenditures. Funding for campaigns was disbursed mostly through interregional transfers from provincial treasuries and salt administrations rather than paid out of central government reserves. The Second Jinchuan War of 1771–1776 – whose total cost of nearly 63 million taels dwarfed the costs of Qianlong’s previous campaigns – was largely financed from the regular tax income and granary stocks of Sichuan and adjacent provinces; in addition, juanna contributions from salt merchants provided about 10 percent of the war chest.16 The rugged terrain of the Tibetan borderlands where the Jinchuan campaign was fought was the chief reason for its exorbitant expenditures: transport of food and supplies accounted for nearly three-quarters of total outlays, in contrast to the mere 17 percent spent on salaries, provisions, bonuses, and rewards. Military equipment – including soldiers’ clothing and shoes – comprised only 8 percent of expenditures. Only minor sums were spent on gunpowder weapons: 400 to 600 cannon, 3,500 muskets, and 2,550 tons of gunpowder for an army numbering 114,000 troops. The ability of the Qing state to conduct lengthy, large-scale military campaigns without resort to deficit financing in part reflected its minimal capital outlays. Virtually no money was spent on a navy (by far the greatest military expense of European nations), and only modest sums were devoted to fortifications and firearms. Nonetheless, Qianlong was appalled by the cost of the Jinchuan campaign and ordered the compilation of a Wartime Expenditure Code (junxu zeli 軍需 則例) to tighten oversight and rein in military spending. 16

Ulrich Theobald, Finance and Logistics in Late Imperial China: A Study of the Second Jinchuan Campaign (1771–1776) (Leiden, Brill, 2013).

376

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

Still, the Qing court was unprepared for the staggering financial costs it would incur in suppressing the White Lotus and Miao rebellions that erupted in 1794–1795 and lasted for a decade. Outlays for the White Lotus campaigns have been estimated, conservatively perhaps, at 150 million taels, and contributions from merchants soared to 18 million taels, three times the total raised during the Second Jinchuan War. Military expenditures drained – but did not wholly exhaust – the substantial treasury surpluses (reaching a peak of 70 million taels in 1781, on the eve of the Second Jinchuan War) accumulated over the course of the eighteenth century. Although the White Lotus Rebellion escalated into a dire military crisis, the Qing managed to weather the turbulence without departing from its familiar repertoire of fiscal expediencies. The palpable sense of military competition that suffused early modern European political consciousness was absent, at least before the outbreak of the Opium Wars in the mid-nineteenth century. A conspicuous feature of Qing fiscal policy – as was true of earlier periods – was the absence of public debt. Indeed, rather than borrowing from financiers to raise funds for wartime expenditures, the Qing government through the Imperial Household Department became an important source of credit for private merchants. After the defeat of the White Lotus rebels the government resumed its focus on generating annual fiscal surpluses and accumulating savings in the form of treasury reserves as the means to cope with future emergencies. Yet the Qing state’s aversion to debt and borrowing, like its commitment to minimal taxation based on fixed revenue quotas, constrained its ability to respond to military and economic crises. Moreover, the Qing tax reforms, by making land rather than people the object of taxation, had rendered civil registration meaningless. After the cessation of population registration in the mid-eighteenth century, the Qing state no longer had the capacity to identify, let alone mobilize, its population. The legibility of society vital to the operation of the fiscal state vanished. The Qing state thus was left with an immobile fiscal system with diminished capacity to capture or generate new revenues. Formal quotas for revenues and expenditures were mandated by the central government, but actual tax collection at the local and regional levels relied on informal and customary practices, especially to meet the largely unbudgeted costs of local government. At times – chiefly at the instigation of the Yongzheng Emperor – the Qing state aggressively pursued an activist agenda of tax reform, public investment, and expansion of the state’s presence at the local level. But Qianlong rescinded or watered down many of his father’s initiatives. Not only was the idea of continuous growth in output and productivity inconceivable within the intellectual milieu of Qing political economy, but Qing officials remained resolutely 377

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn

focused on the efficient use of land and labor resources, the elimination of waste (especially in the form of luxury consumption), and largely rhetorical efforts to promote the diffusion of best practices in agriculture. The noninterventionist stance that prevailed at Qianlong’s court left little scope for the state to promote economic development. Despite the progressive monetization of taxation in the late imperial era, which surely improved the efficiency of tax collection, the fiscal powers of the state had been sharply curtailed compared to earlier periods. The Ming–Qing state retreated from the high levels of revenue generated – especially from indirect taxes – under the Song in favor of minimal taxation derived overwhelmingly from landowners. Although military costs consumed a markedly smaller percentage of total economic product in the early Qing compared to earlier periods, the self-imposed limits on the state’s fiscal capacity hindered investment in growth-oriented public goods such as education, infrastructure, technological improvement, and resource development.

Conclusion The late imperial Chinese state exercised only superficial control of fiscal resources. The reversion to direct taxation (focused on land, not people) and an ideological commitment to low and equitable taxation (probably the lowest and most equitable of any major Eurasian state) generated positive inducements for private utilization of land, labor, and capital, but at the same time greatly diminished the state’s extractive capacity. This fiscal regime abetted private economic growth as long as the empire enjoyed domestic peace and external security. But when confronted by the urgent need to cope with the military crises of the nineteenth century, the Qing state lacked the infrastructural capacity to mobilize and control fiscal resources. In these respects, Ming–Qing China contrasted sharply with the concentration of fiscal power in the hands of centralizing states in early modern Europe. European rulers’ efforts to expand the state’s fiscal capacity can be traced to the rise of large-scale, capital-intensive warfare triggered by the pursuit of power both within Europe and around the globe. The rise of the fiscal state in Europe was premised on a fundamental shift away from direct taxation of people and property and toward indirect taxation of commerce and consumption – not unlike the pattern we observe in Song China. The fiscal structures and policies of European states assumed diverse forms, but everywhere rulers adopted permanent forms of taxation. Even the new taxes and excises were insufficient to meet the exorbitant costs of military 378

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

Public Finance

expenditure, compelling recourse to massive increases in public debt. The broad trend toward centralization of state power was abetted by mercantilist strategies of political and economic competition and increasing state intervention in the economy through supervision of markets, control of the food supply, and the creation of monopolies. The role of competition among European states in promoting convergence in fiscal strategies and mercantilist trade policies intended to strengthen state power and national economies found no parallel among contemporary Eurasian empires such as the Ottomans, the Mughals, and Ming–Qing China.17 Certainly, the Qing rulers – who faced no serious threat between 1683 and 1850 – did not develop the macroeconomic institutions to penetrate local society and extract revenues on a scale comparable to Europe’s mercantilist governments. Peer Vries has calculated that per capita revenues in China amounted to only one-eighth of the level of Britain’s c. 1750. Vries concludes that the weakness of the Qing state’s “infrastructural power” – which resulted from its frugal rather than predatory character – inhibited its capacity to foster economic growth.18 The Qing state’s commitment to agrarian paternalism may have been more conducive to a Smithian dynamic of market expansion and division of labor than was European mercantilism. But the argument that it was not Smithian dynamics but rather a Schumpeterian logic of asymmetric markets, infrastructural development, investment in strategic industries, promotion of new knowledge and technologies, and demand-driven development that promoted economic growth in the early modern world appears increasingly compelling.

References Bao Weimin 包偉民, Songdai difang caizheng shi yanjiu 宋代地方財政史研究 (Shanghai, Shanghai guji chubanshe, 2001). Chen Feng 陈锋, Qingdai caizheng zhengce yu huobi zhengce yanjiu 清代财政政策与货币 政策研究 (Wuhan, Wuhan daxue chubanshe, 2008). Chen Feng, Qingdai yanzheng yu yanshui 清代盐政与盐税, 2nd ed. (Wuhan, Wuhan daxue chubanshe, 2013). Chen Gaohua 陳高華, “Yuandai shuiliang zhidu chutan” 元代稅糧制度初探, in Nanjing daxue lishixi yuanshi yanjiushi 南京大学历史系元史研究室 (ed.), Yuan shi lunji 元史論集 (Beijing, Renmin chubanshe, 1984), 340–71.

17

18

Patrick O’Brien, “Fiscal and Financial Preconditions for the Formation of Developmental States in the West and the East from the Conquest of Ceuta (1415) to the Opium War (1839),” Journal of World History 23.3 (2012), 513–53. Peer Vries, State, Economy, and the Great Divergence: Great Britain and China, 1680s–1850s (London, Bloomsbury Academic, 2015).

379

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

christian lamouroux and richard von glahn Dai, Yingcong, “Military Finance of the High Qing Period: An Overview,” in Nicola di Cosmo (ed.), Military Culture in Imperial China (Cambridge, MA, Harvard University Press, 2009), pp. 296–316. Gao Congming 高聰明, Songdai huobi yu huobi liutong yanjiu 宋代貨幣與貨幣流通研究 (Baoding, Hebei daxue chubanshe, 2000). Guo Zhengzhong 郭正忠, Songdai yanye jingji shi 宋代鹽業經濟史 (Beijing, Renmin chubanshe, 1991). Hartwell, Robert, “The Imperial Treasuries: Finance and Power in Song China,” Bulletin of Sung–Yuan Studies 20 (1988), 18–89. He Ping 何平, Qingdai fushui zhengce yanjiu: 1644–1840 nian 清代赋税政策研究: 1644–1840 年, 2nd ed. (Beijing, Gugong chubanshe, 2012). Huang, Ray, Taxation and Governmental Finance in Sixteenth-Century Ming China (New York, Cambridge University Press, 1974). Iwai Shigeki 岩井茂樹, Chu¯goku kinsei zaisei shi no kenkyu¯ 中國近世財政史の研究 (Kyoto, Kyo¯to daigaku gakujutsu shuppankai, 2004). Lai Huimin 赖惠敏, Qianlong huangdide hebao 乾隆皇帝的荷包 (Beijing, Zhonghua shuju, 2016). Lai Jiancheng 赖建诚, Bianzhen liangxiang: Mingdai zhonghouqide bianfang jingfei yu guojia caizheng weiji, 1531–1602 边镇粮饷：明代中后期的边防经费与国家财政危机 (Hangzhou, Zhejiang daxue chubanshe, 2010). Lamouroux, Christian, Fiscalité, comptes publics et politiques financières dans la Chine des Song: Le chapitre 179 du Songshi (Paris, Institut des hautes études chinoises, 2003). Liu, William Guanglin, “The Making of a Fiscal State in Song China, 960–1279,” Economic History Review 68.1 (2015), 48–78. Miyazawa Tomoyuki 宫泽知之, So¯dai Chu¯goku no kokka to keizai 宋代中國の国家と經 濟 (Tokyo, Sho¯bunsha, 1998). Qi Xia 漆俠, Songdai jingji shi 宋代經濟史 (Shanghai, Shanghai renmin chubanshe, 1988). Qi Xia, Wang Anshi bianfa 王安石變法 (Shanghai, Shanghai renmin chubanshe, 1979). Schurmann, Franz, Economic Structure of the Yüan Dynasty (Cambridge, MA, HarvardYenching Institute, 1956). Tang Wenji 唐文基, Mingdai fuyi zhidu shi 明代赋役制度史 (Beijing, Zhongguo shehui kexue chubanshe, 1991). Umehara Kaoru 梅原郁, “So¯dai no naizo¯ to sazo¯ – kunshu dokusaisei no zaiko” 宋代の內 藏と左藏–君主獨裁制の財庫, To¯ho¯ gakuho¯ 東方学報 42.1 (1971), 127–75. Wang Shengduo 汪聖鐸, Liang Song caizheng shi 兩宋財政史 (Beijing, Zhonghua shuju, 1995). Wang, Yeh-chien, Land Taxation in Imperial China, 1750–1911 (Cambridge, MA, Harvard University Press, 1973). Wong, R. Bin, “Taxation and Good Governance in China, 1500–1914,” in Bartolomé YunCasalilla and Patrick K. O’Brien (eds.), The Rise of Fiscal States: A Global History, 1500– 1914 (Cambridge, Cambridge University Press, 2012), pp. 353–77. Zelin, Madeleine, The Magistrate’s Tael: Rationalizing Fiscal Reform in Eighteenth-Century Ch’ing China (Berkeley, University of California Press, 1984). Zhang Jinling 張金岭, Wan Song shiqi caizheng weiji yanjiu 晚宋時期財政危機研究 (Chengdu, Sichuan daxue chubanshe, 2001).

380

https://doi.org/10.1017/9781108587334.011 Published online by Cambridge University Press

10

Political Economy helen dunstan

“Political economy” is a Western term that carries its own, evolving ideological baggage. For John Stuart Mill, political economy was a science – that which “traces the laws of such of the phenomena of society as arise from the combined operations of mankind for the production of wealth, in so far as those phenomena are not modified by the pursuit of any other object.”1 Adam Smith used the word “science,” but meant what Mill would have called “art”: for him, “political oeconomy” could be “considered as a branch of the science of a statesman or legislator” and had as its objectives “enabl[ing]” the people to prosper through their own efforts and “supply[ing] the state or commonwealth” with means of payment for “the public services.”2 Smith takes us closer than Mill to what the authors mentioned in this chapter understood as their mission. It is not that Chinese writers were incapable of identifying infallibly observed regularities, but construction of a disciplinary edifice through the systematic “tracing” of such regularities in economic behavior was not a premodern Chinese project. Best-practice guidance to the rulers of “All under Heaven” was, by contrast, the core business of the Confucian professional.3 The notion that one recognize a “branch” of the statesman’s “science” which addresses the material resources of both populace and state would not have seemed alien in so-called “Confucian China,” although there was no name for such a field 1

2

3

John Stuart Mill, Essays on Some Unsettled Questions of Political Economy (London, J.W. Parker, 1844), p. 140. Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, 5th ed., 1789, rpt. (Chicago, University of Chicago Press, 1976), vol. 1, p. 449. Cf. John W. Dardess, Confucianism and Autocracy: Professional Elites in the Founding of the Ming Dynasty (Berkeley, University of California Press, 1983), Chapter 1.

381

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

of statecraft. While jingji 經濟, which means “economy” today, was common in premodern texts, this pair of verbs (“to set to rights” and “to rescue”) referred to governance in general. It was the shi 世 (literally, [present] generation), that was to be put to rights, and the economic approach to doing this was one of several. Economic morality was a key component of the governing-class ethical edifice that was Confucianism, and Confuciantrained officials inevitably confronted economic practicalities as they assisted the Son of Heaven (the emperor) in governing. With these clarifications, the phrase “economic statecraft” is a serviceable sinological gloss for the English term “political economy.” Smith’s statement of the “objects” of political economy may sound reminiscent of standard Chinese encapsulations such as guoji minsheng 國計民生 (to translate freely, “the state’s budget and the population’s livelihood”). However, Smith’s context in the world of competing European nationstates differed utterly from that assumed by mainstream Chinese political economists during most of the long era addressed in this chapter. It was shortly before the dawn of the Industrial Revolution and in refutation of the conventional doctrines of mercantilism that Smith published his seminal investigation of “the nature and causes of the wealth of nations.” Premodern Chinese political economy, by contrast, was generally pursued as if the Chinese realm were self-contained and the one legitimate focus were the flourishing of the domestic population, which was susceptible to either benefit or harm from the state apparatus that its taxes supported. Arguments about policy were formulated in an intellectual milieu in which homage to sage founders was practiced in tension with recognition of ongoing historical change. As similar issues came to be addressed in different ages, discursive traditions developed. These traditions were variously fed by norms from Confucian scriptures compiled at least a millennium earlier in another, radically different, sociopolitical context, and/or by anecdotes and arguments from authors addressing the same issue at an earlier stage in the tradition’s history. Some authors wrote without reference to their precursors; some contributed variations on pre-existing arguments; a few contributed apparent innovations. This chapter opens with an exploration of premodern Chinese positiveresource language; that is, the words for wealth and gain, their semantic and conceptual reach, and the contexts in which writers used them. This material is presented as an antidote to the too-easy attribution of an anti-commercial, anti-economic bias to Confucian mentality. Beyond this examination of key linguistic tools, this chapter is informed by three key assumptions. First, 382

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

premodern Chinese political economists were not unanimous on values, objectives, and techniques. We come closer to their world when we identify the issues they debated. Second, failure to recognize that an author is taking a position within a tradition, or that he is employing an established methodology, may lead to flawed assessment of his work’s significance. Third, the quest for true significance often lies in the close reading that enables the researcher to pinpoint what differentiates a given author’s contribution from previous contributions, and to relate that difference to a specific historical context. This chapter stakes the claim of fine-brushstrokes historiography. Apart from the little fables that have a long history as tools of argumentation in China, the study of premodern Chinese political economy requires familiarity with two main kinds of source material. First are the ancient Confucian scriptures that created a cognitive field, biased heavily toward notions of moral governance, in which most recorded discourse on political economy took place (not to the exclusion of all reference to the classic texts of Daoism and Legalism). As sources of allusions with which to expedite an argument, the ancient texts provided occasion to discuss questions of governmental principle, especially in the extreme case of the controversial Officers of Zhou (Zhouguan 周官), commonly called the Rituals of Zhou (Zhouli 周禮). Problematically claimed for Confucianism from its likely origins in the state-making project of the first imperial dynasty (Qin), the Officers of Zhou could serve as a focus for debate on the merits or otherwise of centralized government, densely staffed bureaucracy, enrichment of the state exchequer, and particular forms of engagement with the natural environment.4 The scholarly essays, treatises, and addresses to the throne that built on the canonical tradition are sampled in the middle section of this chapter; they have their value as reflections of normative assumptions about the well-ordered economy and the state’s role in it, but they are not enlightening as to the theoretical grounds on which one practical technique could be argued to be more effective than another. It is discussions – for example, those in policy recommendations or critiques submitted to the emperor – of the practicalities of implementing specific policies that yield the richer insights into the ability of Confuciantrained bureaucrats to formulate economic theory, or at least identify theoretical assumptions. This is one reason why, although some famous names 4

David Schaberg, “The Zhouli as Constitutional Text,” in Benjamin Elman and Martin Kern (eds.), Statecraft and Classical Learning: The Rituals of Zhou in East Asian History (Leiden, Brill, 2010), pp. 33–63; Jaeyoon Song, Traces of Grand Peace: Classics and State Activism in Imperial China (Cambridge, MA, Harvard University Asia Center, 2015).

383

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

appear in this chapter, covering the writings of the “great” is not the purpose. It is the interest of the arguments that they propounded that earns obscure officials a place in these pages. In the last part of this chapter, re-examination of a dispute that spawned a new cliché for discussing the distribution of resources between state and society introduces a survey of continuity and innovation in discussions of state intervention in the grain trade over nine centuries. The latter topic is chosen primarily because the corpus of relevant writings affords unique insights into the market consciousness that underpinned arguments for state restraint in seeking to assure popular subsistence. Linking the main sections of the chapter are reflections on the deceptive allure of the words “conservative,” “progressive,” and “modern,” easily applied but not always entirely justified, as the concluding comparison of an eleventh-century and an eighteenth-century moment of radicalism illustrates.

The Linguistic Foundations: Concepts of Wealth and Gain Classical Chinese has two words that primarily refer to wealth, cai 財 and fu 富. The difference between them is perhaps best explored through etymology. Fu is basically a stative verb; it means “to be rich” and connotes wealth that is held. As a stative verb, it can modify a noun or couple with a noun in a verb–object phrase; thus fumin 富民 means either “rich commoners” or “to enrich the people.” It exists also as an abstract noun, as in the phrase cangfu 藏 富, “to store riches,” or in that protean cliché of Confucian economic statecraft, cangfu yu min 藏富於民 (“to store riches within the population”). Inspired by a notion that can be traced at least as far back as the Guan Zi (third to second centuries B C E), the cliché could be taken to refer to letting society be wealthy so that it can support government activity, particularly in time of need (see von Glahn, Chapter 5 in this volume). In practice, it was readily interpreted as an injunction to preserve wealth where it existed – with the wealthy. This perpetuated the assumption that economic inequality is a social fact, with the rich providing for the poor as well as assisting government. Cai is very different, despite some apparent semantic overlap. Fundamentally a noun, it is cognate, and was once interchangeable, with the words cai 才 (“endowment,” hence “talent”) and cai 材 (“timber,” but also more generally “material,” hence also “treasures” and “natural qualities”). It seems that “resources” is indeed the common underlying meaning 384

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

of all three words, and so when used as a verb, cai 財 arguably meant “to treat as resources,” hence “to exploit.”5 There is thus a dynamism in the word cai 財 (the resources for doing something, as in the later compound caili 財力, literally “resources-power”) that is absent from the word fu. It is, however, the cowrie radical (貝) that differentiates cai 財-resources from resources of other kinds, and as cowrie shells functioned as proto-money in ancient China, cai 財 acquired the connotation of “monetary wealth,” with the concrete referent at any given time depending on the currency system. As Xu Guangqi (1562–1633) put it c. 1630, “What was called wealth in the Tang and Song was nothing but strung-together coins, and what is called wealth in the present age is nothing but silver.”6 Cai in the monetary sense often referred to wealth as something held in government or private treasuries, and so licai 理財, “to manage wealth,” typically referred to fiscal management; that is, management of the monetary resources in the state’s exchequer. The phrase shengcai 生財 (literally, “to generate wealth”), however, was more versatile. It was often used in fiscal contexts to designate the income-generation aspect of wealth management. However, statements such as Bao Shichen’s early nineteenth-century “those who generate wealth are farmers” invite us to a deeper exploration of premodern Chinese concepts of wealth.7 Did such statements merely claim that all monetary wealth is traceable to agricultural toil, or did they draw upon a broader understanding of wealth as resources of any kind, with staple (that is, cereal) foodstuffs occupying pride of place? The famous Han dynasty historian Sima Qian had bequeathed to the Chinese language the phrases benfu 本富, mofu 末富, and jianfu 姦富 – basic, derivative, and ill-gotten riches.8 In later ages, that benfu referred to wealth of agricultural provenance went without saying; Sima’s own historical magnum opus helped to establish the conventional opposition, which had originated in Legalist writings, between benye 本業 or benshi 本事, the valorized basic occupation of tilling the soil, on one hand, and mozuo 末作 or moye 末業, the derivative occupations of manufacture and commerce, on the other. While it is tempting to translate benfu as “true wealth,” the connotation of ben, literally a tree’s rootstock and lower trunk, was rather of solidity and dependability. In fiscal contexts, one stressed that which was ben in order to assert the common sense of a large agrarian polity: the taxes 5

6 8

For the above glosses, see Bernhard Karlgren, Grammata Serica Recensa, rpt. (Göteborg, Elanders Boktryckeri Aktiebolag, 1964), p. 247; Piet van der Loon, “Notes on Xunzi 9, “Wang zhi” 王制 (unpublished ms, n.d.). Xu Guangqi 徐光啟, “Tuntian shu” 屯田疏, in HMJSWB, 490:27a. 7 SC, 1b. SJ, 129, p. 3272.

385

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

levied on households that owned arable land were the bedrock of state finance, whether those taxes were paid in kind or not. A thriving population of farming families was indeed a powerful governing-class ideal. Xu Guangqi invoked it thus in his quixotic elaboration of a proposal, probably made in 1630, that the state try to attract superfluous, derivatively engaged households from crowded areas, such as the commercialized, sophisticated southeastern littoral, to reclaim wasteland in adjacent regions: If, now, the method of evening the population out is implemented, and southerners move gradually northward, so that the folk who strive for riches derivative and ill-gotten are all turned into folk who strive for basic riches, the people’s strength will be daily less burdened, and their customs grow daily more magnanimous. The propagation (shengxi 生息) will be daily broader, and resources (caiyong 財用) daily ampler.9

Writing at an advanced stage of China’s transition to a silver-dominated bimetallic monetary system, with uncoined silver used as a common currency, Xu authored an articulate repudiation of the commonplace equation of wealth (cai) with money. He claimed this equation to be disproved by the relationship between the quantity of “silver and/or coin” and the price of essential products (literally, “grain and silk”): the greater the former, the higher the latter, and the worse the resulting hardship. It was the essential products themselves – “the grain that feeds people and the silk that clothes them” – that were what “the sage-kings of old” had meant by “wealth.” The emergence, over time, of shortages of wealth resulted from the inability of a growing population to increase grain production correspondingly.10 That Xu Guangqi saw no need to specify for whom rising prices spelled hardship possibly reflects the onset of the serious inflation that would accompany the coming change of dynasties. Two major episodes of deflation, c. 1570–c. 1586 and c. 1660–c. 1690, by contrast, were marked by discourse that differentiated between the two forms of money, the usurping uncoined silver and real money, the base-metal, nominally “copper,” coin that had been properly minted by the Chinese state and was the pathway to prosperity. The perceived economic problem was that the capriciousness of silver imports conspired with the attractiveness of silver as an object of possession to leave society short of means of exchange, with serious consequences for producer prices. However, there was also a perceived political dimension: it was as if silver flouted the emperor’s sovereignty and made a mockery of his 9

Xu, “Tuntian shu,” in HMJSWB, 490:16a.

10

Xu, “Tuntian shu,” in HMJSWB, 490:27b.

386

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

authority.11 Samples from this discourse will both substantiate claims made above about the meaning of cai in fiscal contexts and introduce premodern Chinese understandings of what was involved in the generation of cai, a key theme of this chapter. Such was the animosity of Jin Xueyan against the white metal that in 1570 he levelled against silver alone the charge that it can serve as neither food nor clothing. Effective policy to restore coin to its rightful place would signify, he suggested, recovery of the one important “method of wealth generation from previous ages” that was now conspicuously lacking. In Jin’s context, “wealth generation” referred primarily to the exercise of an array of sovereign powers to manufacture spending power for the state at minimal monetary cost and without expenditure of silver. However, from his remark that “If coin is used, the people’s livelihood becomes daily more prosperous; if coin is minted, the state’s means wax in abundance” we may infer that he also anticipated a reversal of the debilitating impact of deflation if his proposals were adopted.12 Nine decades later, Zhao Tingchen used the phrase “one aspect of wealth generation” in reference to the advantages to be reaped from a program of provincial-level minting to produce a single currency that would be demanded from taxpayers in partial satisfaction of their obligations. Translated literally, Zhao’s declaration of desired outcomes reads as follows: If the copper of which there is surplus is now used to supplement the silver that is in short supply, coin will be able to relieve silver, while silver will be interchangeable with coin. Families that have silver will use silver; families that lack silver will use coin. Silver will be coin, coin silver; there will therefore be an annual increase of a myriad silver [ounces] of [base-metal] coin. The more [base-metal coin] is minted, the greater the increase of silver. This true way of storing riches will supplement the state’s means while relieving the population’s indigence . . .13

Confused as this sounds, it echoes, and perhaps draws on, words written by Tan Lun in 1569 to defend his proposed plan for increased mint output against the objection that the high value of silver would make the venture 11

12

13

Cf. Richard von Glahn, Fountain of Fortune: Money and Monetary Policy in China, 1000–1700 (Berkeley, University of California Press, 1996), pp. 156–9, 211–15, 218–24; Mio KishimotoNakayama, “The Kangxi Depression and Early Qing Local Markets,” Modern China 10.2 (1984), 229–38. Jin Xueyan 靳學顏, “Jiangqiu caiyong shu” 講求財用疏, in HMJSWB, 299:11a–b, 12b, 14a–15a. Zhao Tingchen 趙廷臣, “Shu qianfa yi ji minyong shu” 疏錢法以濟民用疏, in HCJSWB, 53:44b.

387

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

unprofitable. Tan had opened his discussion by asserting that enriching the people was a precondition for sufficing the state and entailed “giving weight to cloth, silk, pulse, and grain” while “cheapening silver.”14 Zhao’s outcomes statement is preceded by his testimony – one of several extant from the late seventeenth century – as to the “desolate” aspect both of markets and of silver-starved rural communities, and his essay begins by affirming the importance of an adequate money supply for the well-being of primary producers. While his suggestion that the outcomes perhaps constitute “one aspect of wealth generation” may seem to imply a crude equation of wealth with money in any form, it is more likely that “wealth generation” here refers to the whole process of societal enrichment to be unleashed by the provision of ample means of exchange. By the seventeenth century, there was a long history of understanding wealth generation as something deeper than creating fiscal spending power. The process was not even necessarily human. “Heaven generates seasons and Earth generates wealth,” remarked an official to the second Song emperor in 997. An oft-repeated caveat had it that Heaven and Earth generate wealth – unfortunately, in limited amounts. In 1071, Emperor Shenzong (r. 1067–1085) declared the removal of people to sparsely populated territories to be “the way of a true king in generating wealth and opening up soil”; in 1095, under his successor, the minister of revenue noted the immutable principle that “the true way of wealth generation brings increase to the state and makes the population prosperous.” A late Ming official who preceded Xu Guangqi in advocating wasteland reclamation in Shandong asserted that “in the true way of wealth generation, what counts is assembling the people.” After blaming underpopulation for the uncultivated state of fertile Shandong land, he remarked – adapting a passage from the ancient Book of Rites (Liji 禮記) – that “if there are people, there will be territory and there will be wealth.” If this sounds “physiocratic,” with the word translated “territory” basically meaning “soil,” in another context the same author claimed that privatesector merchants “are those who generate wealth on the state’s behalf.”15 In short, in different contexts different authors might mention the creative role of the Heaven-led cosmos or the productive role of human activity, or both, as in the remark of Yi Xueshi (fl. 1671) that “there are seasons for Heaven’s generation of wealth and times for humankind’s deployment of effort,” 14 15

Tan Lun 譚綸, “Lun licai shu” 論理財疏, in HMJSWB, 322:13a. XCB, 41, p. 872, 114, p. 2676, 248, p. 6049; SHY, “shihuo” 5:16; Shen Yiguan 沈一貫, “Kentian dongsheng shu” 墾田東省疏 and “Qian shi lun” 遣使論, in HMJSWB, 435:8a– b, 435:17a.

388

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

which voiced the common sense of an agrarian economy. That a resourcerich polity could be achieved only through human responsiveness to celestial and/or terrestrial generativity was not a controversial proposition, even among the many who casually used the phrase shengcai of revenue generation for the state. That the enterprise of so responding was not confined to the agrarian sector was obvious to the Confucian scholar Guan Xianyao (fl. 1742), who wrote, “that farmers, craftsmen, and merchants generate wealth is known to people.”16 We turn finally to the word most often used for material gain, li 利 (“benefit,” “advantage,” “profit”). The fact that Confucians perennially dichotomized li and yi 義 (“righteousness”) as, respectively, unworthy and worthy motives, while the label yanli zhi chen 言利之臣 (“ministers who speak of profit”) was invariably censorious, has fostered a biased understanding both of the word li itself and of “Confucian” attitudes toward capitalistic activities such as commerce. The inadequacies of the conventional image of an economically illiterate “Confucian” bureaucracy were exposed by Thomas Metzger, who claimed perceptively that, far from being fixated on the vision of thrifty lifestyles in a peasant-based society, “Confucians” “had the more complex and profound goal of a society seeking to exploit a varied range of economic possibilities without tearing apart the social fabric.”17 Profits and interest rates that were perceived as immiserating or endangering society’s more vulnerable members were unacceptable, and a “Confucian” state’s agents might attempt various actions against them. However, li as “economic possibility” was intrinsically good, and the touchstone of any policy involving it was whether li was to be harnessed to the common good, with benefits outweighing harms. In actual usage, li was the benefit inherent in a given situation, activity, life form, body of water, or, most basically, stretch of land. The form of the character – grain and a knife – suggests a bias toward reaped – that is, realized – benefit; the rule-of-thumb translation “profit” is justified by the fact that li regularly referred to monetary reward on investment. As, however, the juxtaposition of grain with knife could also suggest reapable produce, it is no surprise to find li regularly used also of potential benefit, as in the phrase yili 遺利, “benefit left unexploited.” It was yili that the official Gui E, writing in the late 1520s, discerned in the uncultivated land flanking 16

17

Yi Xueshi 易學實, “Licai” 理財, in HCJSWB, 26:3b; Guan Xianyao 官獻瑤, “Jingyan jiangyi er pian” 經筵講義二篇, in HCJSWB 10:45a. T.A. Metzger, “The State and Commerce in Imperial China,” Asian and African Studies 6 (1970), 38.

389

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

Beijing’s city walls and those of other cities; to recruit sturdy beggars and vagabonds to reclaim for the production of grain and straw these squalid areas polluted by rubbish and dumped corpses would fall, he claimed, within “the true way of wealth generation of the men of old.”18 Although the phrase “exploiting Earth’s benefits to the full” (jin dili 盡地 利) often referred to reclaiming land to realize whatever potential it might have for cereal production, it could equally be used of intensifying the exploitation by applying shuili 水利 (the benefit inherent in water; that is, irrigation), and it was sometimes used of exploiting the potential of marginal land for non-cereal production. Thus in c. 1749 the prefect Qiao Guanglie argued that while “Earth’s benefits” must indeed be “exploited to the full,” that phrase did not mandate full exploitation exclusively through agriculture, for: Those who, of old, were good at planning on the folk’s behalf brought the fertile, cultivable plains and lowlands under cultivation, but where the land was barren, stony and unsuitable for grain of the five species, they planted it with fruit-bearing and timber trees and vines, with the result that enough mulberry was grown for the supplying of silk yarn, and they also planted it with dates, chestnuts, and tubers enough for supplying goods to sell and being prepared for famine.19

For a fuller illustration of Metzger’s point, and of the essential legitimacy of commercial li in the Confucian worldview, it is worth examining a lecture text that Guan Xianyao purportedly prepared for a court seminar, probably in the 1740s. Guan used explication of the “nine offices” (jiu zhi 九職) paragraph in the “chief minister” section of the canonical Officers of Zhou as a vehicle for adumbration of an integrated rehabilitation plan for the underutilized landscapes of his day. It was, he argued, because of the interconnected toils of those performing all nine “offices” that, in the model polity orchestrated by the hallowed Duke of Zhou, full exploitation of the realm’s terrestrial benefits had been perfectly matched with the full exploitation of its human labor. Instead of the bare, denuded mountainsides of the contemporary far west and deep south, the lacustrine wastes of the Huai and lower Yellow River valleys, and the orchardless north China countryside, there had once been variegated, wooded landscapes teeming with birds, beasts, waterfowl,

18 19

Gui E 桂萼, “Yingzhi tiaochen shi shi shu” 應制條陳十事疏, in HMJSWB, 179:14a. Qiao Guanglie 喬光烈, “Tongzhoufu huangdi ququan yi” 同州府荒地渠泉議, in HCJSWB, 38:9b.

390

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

fish, turtles, and fruit-bearing trees and vines. Such landscapes had sustained the ancient model economy at whose restoration policy today should aim. The ancient blueprint assumed nine “offices” – that is, roles – roles 1 to 4 were the primary production roles that made possible roles 5 to 9. The contribution of the Officers of Zhou had been to ordain a place for gardeners, including fruit-tree cultivators (role 2), wilderness-resource extractors (of two kinds, role 3), and livestock raisers (role 4) within a primary-production sector that the linguistic convention of subsequent ages was to represent as comprising only grain growers (nong 農, or “farmers,” role 1) and their silkwormrearing spouses (role 7, differently imagined). The Officers of Zhou model provided for three classes of grain grower; Guan followed tradition in assuming these to be defined by the kind of terrain each cultivated. In his interpretation, once the “hill” and the “marsh” grain growers had been allocated such of the “upland forest,” “highland,” and “wetland” terrain as could be tilled, the gardeners, livestock raisers, moor men (role 3A) and swamp men (role 3B) had been assigned responsibility for, respectively, “plantation matters,” “domestic livestock matters,” “hill matters,” and “marsh matters” over the vast uncultivated territories that remained. Not only did these arrangements enable the residents of these territories to “enjoy the benefits” inherent in the three types of terrain, but the “‘goods’ ‘amassed and circulated’ by the ‘merchants’” (role 6 – the inner quotation marks indicate the canonical words that Guan was rearranging), the “‘silk and hemp yarn’ ‘processed’ by the ‘wives’” (role 7), “the ‘eight kinds of material’ ‘formed into their due shapes’ by the ‘hundred kinds of craftsman’” (role 5), “‘the roots and berries’ ‘gathered’ by the ‘bondsmen and bondswomen’” (role 8), and “the ‘migrations’ of the ‘casual population [that is, labor force]’” (role 9) all derived from or depended on this non-cereal primary production, or one or more specific sectors of it.20 Whatever Guan may have intended by positioning the merchants in the middle of the sequence, not in their original place after the craftsmen, and by prefacing the rearranged canonical words about them with other venerable phrases attesting to the normality of commerce, the effect is arguably to emphasize the crucial role of merchants, shopkeepers, and other entrepreneurs and middlemen in mediating between the primary- and secondaryproduction sectors, whether by channeling raw and semi-processed materials to artisan workshops and cottage workspaces, or by undertaking the foodstuff redistribution needed to sustain division of labor in the increasingly 20

Guan Xianyao, “Jingyan jiangyi,” in HCJSWB, 10:45b–46a.

391

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

sophisticated economy of Guan’s day. For the word fu 阜 of the Officers of Zhou phrase futong huohui 阜通貨賄 (“amass goods and circulate them”) early acquired a dynamic character that “amass,” a translation justified by fu’s basic meaning, “earthen mound,” fails to capture. Glossed as sheng 盛 (“ample”) by the canonical work’s most influential early commentator, fu was used transitively to mean “make [object] abundant.” Fu is not the antisocial hoarding of the profiteer but the facilitative role of commerce in the expansion and diversification of production. Guan Xianyao’s text left much to the imagination as a paradigm applicable to the economy of late imperial China. There is something ludicrous about the notion that the highly statist Officers of Zhou provided valid lessons about rendering landscapes productive in an age in which small-scale silviculturalists in remote Guizhou would soon independently develop a contractregulated system of tradeable shares in order to make commercial forestry a realistic proposition for households with the normal need for steady income.21 However, as exploitation of the “benefits [inherent in] the hills and marshes” presumably included mining and salt production, Guan need not be accused of advocating reversion to a primitive economy of natural abundance. Rather, his ideal was use of “all the world’s [inherent] benefits” (tianxia zhi li 天下之利) to nourish “all the world’s people.”22 Just as for Song Yingxing, author of the treatise on matters technological The Works of Heaven and the Inception of Things (Tiangong kaiwu 天工開物, 1636), the word cai referred to the whole gamut of products, so li meant realizable benefits of every kind.23 While Metzger surely did not intend “economic possibility” as a translation of li, it actually fits rather well.

Conservatism and Tradition in Chinese Political Economy A scriptural passage facilitated readiness to understand “wealth” as something more basic than money. This was an oft-quoted prescription in the Great Learning, a chapter of the Book of Rites that acquired canonical status as a separate text from the twelfth century onwards:

21

22 23

Meng Zhang, “Financing Market-Oriented Reforestation: Securitization of Timberlands and Shareholding Practices in Southwest China, 1750–1900,” Late Imperial China 38.2 (2017), 120–37. Guan Xianyao, “Jingyan jiangyi,” in HCJSWB, 10:45a. Song Yingxing 宋應星, “Mincai yi” 民財議, cited in von Glahn, Fountain, p. 200.

392

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

There is a great way for generating wealth. If those generating it are many but those consuming it few, and if there is energy in the making of it but ease in the use of it, wealth will ever be sufficient.24

Economic moralists down the ages preached to this passage, invoking it when arguing for lower dependency ratios, greater fiscal prudence, and/or restraints on popular extravagance. Combined with other maxims and nostrums from the hallowed ancient texts, it formed the conceptual scaffolding of the discourse that gives “Confucian orthodoxy” its reputation for austerity. These maxims included the principles, both derived from the “Kingly Institutions” chapter of the Book of Rites, that the state should live within its means (liang ru wei chu 量入爲出, “determine outgoings by measuring income,” to quote the maxim in its conventional, abridged form), and that every three years of agricultural production should yield grain reserves sufficient for one year.25 It is tempting to label as “conservative” any tradition that prioritizes food security, emphasizes budgetary restraint, and recycles timeworn clichés and modes of argumentation in support of these penchants. Take, for example, the use made down through the ages of the lamentation by Han Yu, in his renowned essay of 805 on “Essentials of the Way,” that the multiplication of the unproductive Buddhist and Daoist monks and nuns had bloated to six the “four orders of the people” (simin 四民: scholars, farmers, craftsmen, and merchants). Han had argued that in the six-order society, the farming, manufacturing, and trading professions each had six professions living off their productive or distributive labour.26 In a letter of 1027, Fan Zhongyan included soldiers among the two supernumerary orders and complained that idleness, frivolity, and fecklessness within all six orders were putting stress upon the food supply grown by the farming order.27 His contemporary, Chen Shunyu (d. 1074), recognized two extra supernumerary orders and claimed that a hundred grain-consumers were now feeding on the toil of every farmer. Chen’s orders 7 and 8 were soldiers and vagrants (youshou 游手, literally “wandering hands”), a category that was on the Song official mind.28 It may have been the extraordinary scale on which, from about the time of Chen Shunyu’s death onwards, the Song state sold ordination certificates that

24 25 27 28

Chapter 42, “Daxue” 大學, in LJ, 2, p. 1675b. Cf. Chapter 5, “Wangzhi” 王制, in LJ, 1, p. 1334a. 26 “Yuandao” 原道, in HCLQJ, pp. 171–5. “Shang zhizheng shu” 上執政書, in FWZJ, 8:7b–8a. Zhihong Liang Oberst, “Chinese Economic Statecraft and Economic Ideas in the Song Period (960–1279)” (Ph.D. diss., Columbia University, 1996), pp. 63, 277.

393

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

assured the survival of the two kinds of religious clergy as the supernumerary orders par excellence.29 In 1229, Zhen Dexiu concluded an explication of Han’s argument by criticizing the court’s reliance on such sales as a supplementary means of “generating wealth”; he was not the first to claim that depleting the farmers’ ranks by licensing monks undermined that endeavor.30 Zhen, who was commenting on the Great Learning’s doctrine on creating wealth, made the supposed attrition of the farming order by the popularity of ordination a concrete example of what it meant to have “those generating” wealth “few” and “those consuming” it “many.” So did the early Ming official Wang Shuying, in a memorial to the ill-fated Jianwen Emperor (r. 1398– 1402).31 In the early Qing, Jin Fu elaborated magisterially on the theme, claiming that ancient experience had shown that food would be ample if the farming order comprised 70 percent of the population, expatiating on the evils of Buddhism and Daoism, and vividly describing the “limitless” flock of “vagabonds and loafers,” prominent among them aggressive beggars, who now constituted a virtual third supernumerary order.32 The idlers, themselves part of a long textual tradition, displace the monks and nuns entirely in the two essays whose composition in 1793 has earned their author, Hong Liangji, the questionable soubriquet of “China’s Malthus.” Neither of these essays mentions the religious orders. The second, having argued that population increase is leaving the four legitimate orders crushed between falling wages and rising prices, concludes with a veiled warning of the risk of violence in times of natural disaster from “vagrants and lovers of idleness,” whose numbers have swelled to “tens of times” their “previous” level.33 Four-orders discourse, with its agrarian bias, did not provide a framework for informed discussion of the most dynamic and sophisticated sectors of the Chinese economy in the post-Tang era. Nor, despite the positive role that it 29

30

31 32

33

Robert Hymes, “Sung Society and Social Change,” in John W. Chaffee and Denis Twitchett (eds.), The Cambridge History of China, vol. 5, part 2, Sung China, 960–1279 (Cambridge, Cambridge University Press, 2015), pp. 531–2. Oberst, “Chinese Economic Statecraft,” p. 311; see also the discussion by Li Chunnian quoted in XNYL, 89:13b. Wang Shuying 王叔英, “Zizhi ce shu” 資治策疏, in HMJSWB, 12:1b, 3b. Jin Fu 靳輔, “Shengcai yuxiang diyishu” 生財裕餉第一疏, in JWGZ 7:43a–48a; partial translation in Helen Dunstan, Conflicting Counsels to Confuse the Age: A Documentary Study of Political Economy in Qing China, 1644–1840 (Ann Arbor, University of Michigan Center for Chinese Studies, 1996), pp. 172–83. Hong Liangji 洪亮吉, “Shengji pian” 生計篇, in HBJYJ, 1, pp. 179–80. For a translation of the first essay, see K.C. Liu, “Hong Liangji: On Imperial Malfeasance and China’s Population Problem,” in William Theodore de Bary and Richard Lufrano (eds.), Sources of Chinese Tradition, 2nd ed. (New York, Columbia University Press, 1999–2000), vol. 2, pp. 174–6.

394

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

accorded commerce, was it hospitable to advocacy of the spirit of adventure that helps fuel economic innovation. In that sense, it can be called “conservative,” as can the wealth-creation doctrine of the Great Learning when invoked within it. Perhaps, however, “tradition” is a more useful concept than “conservatism” for engaging with such material on its own terms. Authorial skill lay in using the frameworks offered by tradition to deliver a pointed analysis of, or withering polemic on, contemporary issues. We have seen Guan Xianyao do this with the canonical tradition of “nine-roles” discourse; Wang Shuying did it in his systematic assessment of the economic and fiscal arrangements of the close of the fourteenth century in light of the Great Learning’s four conditions for sufficiency of wealth. Wang’s memorial is adorned with fragments of canonical Confucian economic doctrine. An opening sleight of hand adapts Confucius’ supposed pronouncement on how to improve on a situation in which a state is already populous: first enrich the people, and then teach them.34 Wang represented the people’s enrichment as the key to remedying a situation in which the realm’s as yet unteachable populace was “not particularly numerous,” as was true in the late fourteenth century. Before embarking on his exposition of how the four conditions for sufficiency of wealth were not being met, Wang invoked the Mencian concept of hengchan 恆產 (constant means of livelihood) and the ancient ideal of uniformity in land allocation and taxation in arguing that the very “basis of [wealth] generation” was being undermined by inequality and a tax regime so oppressive that arable land was left untilled. In explaining how the energy available for wealth production was being sapped by the corvée-levying authorities’ disregard for seasonal fluctuations in the intensity of agricultural work, Wang accused those authorities of “robbing [the people] of their time,” a Mencian phrase that every reader would have recognized. In outlining the first of the ways in which “ease” was lacking in the “use” of wealth, he directly quoted from Mencius a pair of affirmations of the wisdom of harvesting natural resources sparingly and with due respect for season.35 Close paraphrase of the Book of Rites passage about each triennium of cultivation providing food enough for a fourth year introduces discussion of the lack of institutions, such as those of earlier dynasties, for saving grain in case of natural disasters.36 However quaint a modern mind may think Wang Shuying’s use of scripture to structure and support his politico-economic analysis, implicit in 34 36

Analects 13.9. 35 Mencius 1A.3, 1A.5, 1A.7; 3A.3. Wang Shuying, “Zizhi ce shu,” in HMJSWB, 12:1a–2b, 5a–6a.

395

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

his text are important issues such as the relationship between social justice and productivity, and the state’s potential role in advancing the former by “instituting” “constant means of livelihood,” as had been done by the designers of uniform land-allocation regimes in ancient and medieval times. Whether a contemporary state should or could “institute” a uniform, egalitarian system of landholding had long been regarded as problematic, but Wang’s hint that this was a thing regrettably left undone was timely. The building of a new dynastic order was the context par excellence in which utopian schemes for micromanaging rural society could seem worth trying, as the Ming founder’s own endeavors illustrate.37 More convincingly attuned to contemporary realities was Wang’s claim that a second factor draining the energy available for producing wealth was the overburdening of the sparse population caused by failure to amalgamate depopulated administrative units. There were more county-level establishments needing to be staffed through compulsory service than the taxpaying populace of underpopulated regions could comfortably bear.38 If the frameworks offered by tradition were valuable tools, it is vital that modern analysts be aware of the sub-traditions in which authors were writing. Scrutiny of both text and context is required to distinguish material present because tradition supplied it from material that authentically reflects an author’s understanding of the issues of his day. One must correct for possible biases induced by the traditional templates in authors’ cognition, be aware of the value as historical evidence of material that refracts reality rather than reflecting it, and watch for variations in the ways in which the authors used the templates. It is these variations that alert us to newness and possible contemporary relevance of content. For example, Hong Liangji’s remarks about “vagrants and lovers of idleness” occur in the same essay as a statement about the means of livelihood of the four legitimate orders, but they are not part of a jeremiad about supernumerary orders. Rather, their warning of the threat to civil order posed by the vagabonds provides an effective conclusion to a pair of essays (themselves part of a set of miscellaneous exercises in refutation), the first and more “Malthusian” of which purports to dispel the notion that it is a happy thing to live in a well-ordered age. That in the late eighteenth century a Chinese scholar-official sounded the alarm about population increase and the disruptive potential of those who lacked four-orders 37

38

See, e.g., the Ming founder’s “Placard Instructing the People,” translated in Edward Farmer, Zhu Yuanzhang and Early Ming Legislation: The Reordering of Chinese Society Following the Era of Mongol Rule (Leiden, Brill, 1995), pp. 195–209. Wang Shuying, “Zizhi ce shu,” in HMJSWB, 12:5a–b.

396

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

livelihoods fits with modern scholarly consensus on late imperial China’s demographic history and the role of marginalized elements in fomenting the famous late Qing rebellions, the first of which – the sectarian White Lotus Rebellion – began in 1796. Despite the secular content and step-by-step reasoning of Hong’s first essay, however, of the two anachronistic models of prophet and social scientist, it is arguably the former that provides the more effective simile for Hong as “China’s Malthus.” He read the signs of his times and gave a partially accurate warning, drawing on the resources of the long tradition in which he was steeped. His reliance on the value-laden social taxonomy of that tradition obscured the actual marginalized, who included dispossessed aboriginal peoples, under a pall of disapproving vagueness and maladroit characterization. As new research suggests that the White Lotus Rebellion was in fact fomented mostly by people who had surplus income and a stable foothold in their local economy, commitment to the traditional taxonomy arguably misdirected Hong’s prophetic gaze.39 Awareness of the sub-traditions helps us distinguish reincarnations from new births and to recognize novelty in reincarnations. For example, when in 1801 Bao Shichen presented a calculation of the realm’s total acreage in support of an argument that prosperity was attainable even with the present population level, he was contributing to a sub-tradition whose calculation formula could be expressed in either of two ways, depending on one’s purpose and the scripture from which one began. One must read from within the sub-tradition to understand what Bao was about and notice where he made mistakes.40 A review of prior episodes within the sub-tradition will reveal Bao’s political arithmetic from a culturally relevant perspective while illustrating the uses to which others had applied the formula. The formula in the Book of Rites goes as follows. One first notes the number of mu 畝 (unit of area) per block of land one li 里 (unit of distance) square – in the Rites, 900. One then sequentially calculates the number of mu per block ten li square, 100 li square, and 1,000 li square, arriving at a total of 900 million mu. One next determines how many blocks 1,000 li square there are by estimating the number of approximately 1,000-li stretches on each of a roughly stated north–south and east–west axis traversing the territory that is under the king’s suzerainty. By assuming that the stretches that exceed 1,000 li offset those that fall short of it, one arrives at an estimate of the area of 39

40

Yingcong Dai, The White Lotus War: Rebellion and Suppression in Late Imperial China (Seattle, University of Washington Press, 2019), pp. 21–7. Cf. William T. Rowe, Speaking of Profit: Bao Shichen and Reform in Nineteenth-Century China (Cambridge, MA, Harvard University Asia Center, 2018), pp. 75–7.

397

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

the whole – in the Rites, imagined as a block of land 3,000 li square, implying 8.1 billion mu. Finally, one makes a one-third reduction to allow for natural and man-made features on which crops cannot be grown.41 The alternative was to start from the Officers of Zhou tradition that provides a rationale for the 900-mu area of the base-level unit and then identifies the higher-level squares with names; for example, the block 1,000 li square is called a ji 畿. The base-level unit is a block of nine 100-mu squares, eight of which are assigned to individual families, the central one being the ‘public’ allotment from which the food-energy needs of the governing power are met. Called a jing 井 (well) because this graph is reminiscent of the block’s internal boundaries, the unit was the basis of the mythical uniform land allocation system (the so-called “well-field system”) that Wang Shuying invoked in pride of place when hinting that the young Ming dynasty might “institute” an equitable tenurial regime. Neither of the two formulae was necessarily born Confucian. Even the Book of Rites one shares arithmetical assumptions with the “doctrine of fully exploiting Earth’s strength” attributed to Li Kui, the minister of the Wei kingdom in the Warring States period often regarded as a harbinger of Legalism. This doctrine featured calculations intended to quantify the difference that uniform diligence or uniform laziness would make to the yield of the 6 million mu of arable within a block of land 100 li square.42 In the reign of the Yuan emperor Khubilai (r. 1260–1294), Zhao Tianlin used a calculation of the total acreage of the ji-sized block in support of an argument for land reform to address the agrarian inequality of his day. After a description of the idealized grid system in wording much of which comes from the Mencius and another ancient text, he noted that a ji would have contained a million jing, from which 360,000 jing should be deducted to allow for “mountains, rivers, and cities.” This left 512 million mu for family landholdings, and 51.2 million mu for public allotments once one had deducted 12.8 million for the twenty mu of housing land per jing for which another early tradition allowed. While Zhao’s case for land reform was largely a matter of social justice, he hinted that the Yuan state stood to benefit fiscally from reversion to ancient-style agrarian egalitarianism. If one assumed an average yield of 1.5 shi (Chinese bushel) per mu, the 51.2 million mu of public land per ji would annually yield the state 76.8 million shi of income. Such a ratio of revenue to land was unachievable, he indicated, under an exploitative

41

Chapter 5, “Wangzhi” in LJ, 1, pp. 1347b–1348a.

42

HS, 24A, p. 1124.

398

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

landholding system such as that which he described – with some of the latifundia even being used as pasturage.43 In his seventeenth-century magnum opus on agriculture, Xu Guangqi devoted one of two chapters on field systems to reconstructing the ancient grid system from scattered passages in the Officers of Zhou and elsewhere. The challenges were major, for the Officers of Zhou not only differs from the Book of Rites in basing its hierarchy of nested square units on multiplication by four but also envisages the grid’s squares at all levels from the 100-mu block up as separated by the channels of a rectilinear irrigation network. Undaunted, Xu calculated the acreage of the squares at all levels up to the block 100 li square, adding the acreage of the associated paths, roadways, and water channels, and reconciling the two hierarchies of nested squares (Figure 10.1).44 Although Xu performed this herculean computational labor as an antiquarian exercise, his diagrams convey the grandeur of the vision discernible in ancient scripture for the inspiration of latter-day systematic thinkers such as Jin Fu. Jin Fu followed the Book of Rites tradition in seeking to convince the Kangxi Emperor (r. 1661–1722) that, with more effective management of both the realm’s land and its human resources, there should be no question of a revenue shortage. He used the Book of Rites formula with some substantive adaptations. That his block 1,000 li square contained only 270 million mu of cereal-growing arable land is explained partly by his having started with a one-li block of only 540 mu (a figure that reflected the contemporary definition of the standard mu and li), and partly by the fact that after applying the reduction by one-third to allow for non-arable acreage, he deducted a further quarter to allow for land under non-cereal crops. In his calculation of the number of blocks 1,000 li square, however, conservative estimation took Jin back to the Book of Rites standard. The combined area of the Chinese provinces was at least 5,000 to 6,000 li square, he said, but using the maximal vertical and horizontal axes reduced that figure to 4,000 li square, and if one substituted 3,000 even for that, one was left with nine blocks 1,000 li square, the unstated implication of the Book of Rites figures. Jin then assumed a minimum yield of one shi per mu, the long-sanctioned ideal of a 10 percent taxation rate, and a specified average commutation rate to calculate 24.3 million shi of grain and 87.48 million taels of silver as the ample revenue that should be realizable from the land alone on the most conservative assumptions.45 43

44 45

Zhao Tianlin 趙天麟, “Shu bashi yi feng tianxia zhi shihuo: Xian tianchan” 樹八事以 豐天下之食貨：限田產, in YZYJL, 1, pp. 291–2. NZQS, Chapter 4. Jin Fu 靳輔, “Shengcai yuxiang diyishu” 生財裕餉第一疏, in JWGZ 7:43a–48a; translated in Dunstan, Conflicting Counsels, pp. 172–3.

399

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

Figure 10.1 Xu Guangqi’s reconstruction of ancient land tenure. From the 1639 edition of Xu Guangqi 徐光啟, Nongzheng quanshu 農政全書, juan 4, p. 24b in the Chinese Collection, Harvard-Yenching Library. Copyright President & Fellows of Harvard College Xu’s reconstruction of the block 100 li square (tong 同). Its taxable core, which comprised a set of four du 都 each comprising four xian 縣, had a diameter of eighty li, to which was added ten li on either side to allow for a broad water channel. Detail in the top right-hand corner shows that each xian comprised four dian 甸. There was one more intervening unit before the four-based schema connected with a three-based one accounting for the nine allotments of the jing unit. The tong’s total area (9,173,629.44 mu) comprised the 9 million mu of arable yielded by the calculation method in the Rites, plus 173,629.44 mu of irrigation channels, paths, and roadways.

Jin Fu was writing near the end of a large-scale rebellion that had drained the central government treasury. As annual land-tax monetary revenue even in the four post-rebellion decades of the Kangxi era probably fluctuated around a mere 28 million to 29 million taels, including service-levy proceeds, he was indeed limning an alluring

400

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

prospect for the ruler of the day.46 His was in part a back-to-work agenda and in part a call for large-scale state irrigation and reclamation projects in parts of northern and east-central China. The irrigation infrastructure having been put in place, soil fertility would be assured; the land tax should then be set at a full 10 percent of agricultural yield to prevent laziness among the rural populace; and the members of the parasitic orders, from bonzes to beggars, should be given land and expected to till it. A retrospective third-party account indicates that the prototype for Jin’s reclamation scheme drew on the vision of a rectilinear irrigation grid dividing “well-field” squares; if so, choosing the Book of Rites formula for one’s macro-calculations did not necessarily imply rejection of the Officers of Zhou micro-model.47 The ideal animating Jin’s plan for realizing the whole realm’s productive potential was that of a uniformly toiling populace in a well-regulated landscape. If this was social justice, it took a harsh form. Readers can be spared the flawed calculations that Bao Shichen used in 1801 to argue that, whatever others might say, the aggregate ratio of land to population was such as to support agrarian prosperity, with 40 percent grain surpluses at the household level. Suffice it to note that, while his reckoning was more explicitly modernized than that of Jin Fu – he shows awareness of the broadened imperial borders of his day and measures long distances in “degrees” – the parentage of the Book of Rites is unmistakable. Bao retains the tedious progression of the Book of Rites prototype through orders of magnitude, and there are small but telling continuities of wording.48 His neglect of the convention that one enhanced an argument’s authority by naming canonical sources creates more difficulty for the modern scholar than it would have for his scripture-steeped contemporaries. Bao’s authorial agency can be assessed only once his indebtedness is clear. The ancient computational framework did not conduce to accuracy in political arithmetic, but it could nourish the bureaucratic imagining that flourished in quasi-utopian system design – the ludic counterpart of rational administration. Fortunately, however, most premodern Chinese political economy was shaped not by archaic ratiocination but by engagement with

46

47

Cf. Shi Zhihong 史志宏, Qingdai hubu yinku shouzhi he kucun tongji 清代戶部銀庫收 支和庫存統計 (Fuzhou, Fujian renmin chubanshe, 2008), pp. 40–1. HFSY, 57a. 48 SC, 43b–44b; cf. Rowe, Speaking of Profit, pp. 76–80.

401

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

the practicalities of government. We next explore how traditions that outlasted dynasties emerged when their creators left the realm of values and ideals to engage with distribution (how resources and responsibilities should be apportioned), which begged the question of mechanics (how forces work).

Resource Distribution and Visions of the State in the Economy The following discussion takes as starting point a cliché of eleventh-century origin, a dictum bearing on the proper distribution of resources between society and state. Generally thought symptomatic of a conservative politicoeconomic standpoint, the cliché is more precisely viewed as a protean conceptual resource – a form of words available for manipulation and remolding to suit a variety of purposes and contexts. Exposure of a myth about its originator’s assumptions will clarify wherein he differed from his main antagonist in his position about fiscal and economic growth, and the state’s proper role. The discussion recontextualizes a well-known eleventh-century episode of pressure for more state control in the economy, thereby preparing the way for comparison with an eighteenth-century episode of pressure for less state intervention. It is hoped that this comparison will undermine some facile assumptions about labeling. The finance commissioner, Cheng Lin, probably expected to be understood as stating the obvious when he wrote in 1034 that “there is a limit to the generation of wealth by Heaven and Earth” (tiandi shengcai you xian 天地生 財有限). His purpose was to contrast the ultimately limited output of cosmic generativity with the seemingly endless escalation of military expenditures.49 Similarly, the caveat that “Heaven and Earth generate only so much wealth” (tiandi shengcai zhi you cishu 天地生財止有此數) was usually entered not as a substantive thesis, but as a self-evident truth justifying an ensuing claim – a practical implication or policy position. For the late Ming statesman Zhang Juzheng, the point was that the state had better economize, rather than scheming to extract more wealth from a private sector that could not simply produce more wealth and so absorb the loss.50 One common rhetorical maneuver was to follow the caveat with some version of the truism that

49 50

XCB, 114, p. 2676. Zhang Juzheng 張居正, “Chen liushi shu” 陳六事疏 and Zhang Juzheng et al., “Kanxiang Hubu jincheng jietie shu” 看詳戶部進呈揭帖疏, in HMJSWB, 324:10a–b, 325:19b respectively.

402

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

such wealth as existed was necessarily held within either the private or the public sector, before arguing that, in the matter under discussion, a proper distribution of the realm’s wealth would favor the private sector. The cliché that “Heaven and Earth generate only so much wealth, and if it is not with the authorities, it will be with the people” (buzai guan ze zai min 不 在官則在民) is traceable to an account by the Song statesman Sima Guang of the discussion at two audiences that he represented as having occurred a few days before Emperor Shenzong (r. 1067–1085) appointed the fiscal innovator Wang Anshi to the State Council position from which Wang launched his New Policies. As the account is usually taken at face value, it is vital that we read it critically. Sima remembered Wang as having claimed, in words easily recognizable as paraphrased from Sima Qian, that the hallmark of a good financial administrator was that he provided amply for the state without inflicting extra taxes on the people. He remembered himself as having riposted with an innuendo that Wang was trying to deceive the young emperor just as the financial wizard Sang Hongyang had deceived Emperor Wu of the Han dynasty, before delivering himself of the remark: “The valuables and [other] goods that Heaven and Earth generate amount only to so much, and they are either among the people or else with the state.” There was no source other than the people from which to draw the wherewithal to provide amply for the state.51 It is even harder to know what Sima Guang believed about cosmic generativity than it is to know what he said. An alternative, almost certainly eleventh-century, account of Sima’s utterance quotes him as having claimed – by way of an analogy to clinch his point – that “in the case of rainfall, if there is summer flooding then there will be autumn drought.” Su Shi, who recorded this inanity, was in the capital soon enough after the second audience to have heard gossip or eyewitness accounts.52 One might have expected the version of Sima Guang’s account in the “manuscript records” associated with what purports to be his diary to establish what he really said, but while this text is largely identical with that preserved in his collected works, it must have been finalized after Wang’s elevation in late February 1069 (presumably long enough afterward for Sima to have realized that event’s significance), whereas the discussions of which it purports to give verbatim transcripts occurred in mid-September 1068.53 The text, in short, is

51 52

Sima Guang 司馬光, “Erying zoudui” 邇英奏對, in SMCJJ, 42:18a–21a. DPQJ, houji, 11:6b. 53 SMGRJ, pp. 128–33.

403

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

suspect, as are others (introduced below) that we receive through Sima’s artifice. Zhihong Liang Oberst is right to question the received interpretation of Sima’s argument as he recorded it – that he took cosmic generativity to be immutable and limited. Sima’s wording has straightforward logic as a refutation of Wang’s claim about what good fiscal administrators can accomplish. Attention to context protects us from misreading other words of Sima’s that seem to reinforce the conventional interpretation. Sima wrote, “There is constancy in the productive activity of Heaven and Earth, but humankind becomes daily more abounding, while tillers grow gradually fewer and vagrants daily more numerous. Desires are limitless, and customs become daily more extravagant. What hope for a wish that resources be exempt from running out?” The problem lay not with cosmic generativity, which was ultimately reliable, but with population growth, a rising dependency ratio, and court behavior that encouraged folly in society at large.54 Wang Anshi was orthodox in focusing on human agency in the generation of wealth; however, he was controversial in the unusual degree to which he envisaged state activity as able to promote economic growth, solve the state’s fiscal problems, and benefit not only consumers but also primary producers and legitimate trading interests by substituting state programs for (perceived) unfettered profiteering. Scholarly consensus portrays Wang as a doctrinaire enforcer of militantly activist state programs whose purported justification was bound up with the authority of selected Confucian scriptures (see the Lamouroux and von Glahn chapter in this volume). Wang enlarged the bureaucracy and recruited men who were attuned to the ways of the countinghouse and marketplace for an approach to economic statecraft that Paul J. Smith has dubbed “bureaucratic entrepreneurship.”55 The treacherous ambiguity of the phrase shengcai came into its own as never before in Chinese history. An early expression of Wang’s views on the state’s role in wealth generation, translated literally (save for the contraction of “[all] under Heaven” to “world”) to avoid distortion via unlabeled interpretation, warns of his bias toward fiscal considerations as the purpose of state sponsorship of wealth creation: Although your servant has never studied wealth and gain, I take the liberty of having surveyed the broad outline of former generations’ governance of 54

55

Sima Guang, “Lun caili shu” 論財利疏, in SMCJJ, 25:14a; Oberst, “Chinese Economic Statecraft,” p. 298. Paul Jakov Smith, Taxing Heaven’s Storehouse: Horses, Bureaucrats, and the Destruction of the Sichuan Tea Industry, 1074–1224 (Cambridge, MA, Harvard University, Council on East Asian Studies, 1991), pp. 111–18.

404

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

[= policy on] wealth. It is reasonable to say that they relied on the world’s strength [= the strength of the world’s people] to bring forth (sheng 生) the world’s wealth and took [= drew on] the world’s wealth to provide for the world’s expenditures.

The point was not to deplore the state’s high payroll costs but to expand the state’s revenue base in order to be able to afford them.56 Thus wrote Wang in 1058, ten years before he came to power. During Wang’s ascendancy, a project intended to strengthen the economic position of peasant farmers and so enhance agricultural productivity soon started to return a handsome profit to the state. The “Green Sprouts” (qingmiao 青苗) policy, a scheme inaugurated in 1069 to rescue peasants from private loan sharks by lending them money raised by selling grain that had been accumulated in pursuance of the time-honored technique for stabilizing grain prices, became a byword for irresponsible and ultimately self-interested state meddling in the agrarian economy. The metropolitan Bureau of Market Transactions (shiyiwu 市易務) was hatched in 1072 from a proposal to apply the principles governing the said price-stabilization technique to the management of all branches of trade entering the capital, Kaifeng, thereby preventing “the power of opening and closing [trade], and gathering and dispersing [commodities]” from “veering to the rich,” to the detriment of ordinary merchants and the public. The experiment, intended to keep trade moving, the market supplied, and the state’s needs satisfied, progressively metamorphosed into an almost realm-wide network of allegedly monopolistic state wholesaling and moneylending operations that, being highly lucrative, epitomized that deadly sin of “Confucian” political economy, “competing with subjects for profit” (yu min zheng li 與民爭 利).57 Wang’s New Policies thus gave the fiscal moralists of later generations a rhetorical weapon in the battle for the boundary between legitimate exploitation of resources rightfully belonging to the state and unconscionable usurpation of commercial opportunities. The late eleventh-century episode of innovative finance came to be treated as both an object lesson in the moral dangers of fiscal self-aggrandizement and an aid in clarifying which means of raising revenue were beyond the pale of righteous governance. 56

57

Wang Anshi 王安石, “Shang Renzong Huangdi yan shi shu” 上仁宗皇帝言事書, in LCWJ, 39, p. 417. XCB, 231, p. 5622; Paul Jakov Smith, “Shen-tsung’s Reign and the New Policies of Wang An-shih, 1067–1085,” in Denis Twitchett and Paul Jakov Smith (eds.), The Cambridge History of China, vol. 5, part 1, The Sung Dynasty and Its Precursors, 907–1279 (Cambridge, Cambridge University Press, 2009), pp. 394–7, 403–7, 414–19, 429–34, 443–5.

405

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

In 1062, Sima Guang had taken the simpler position that a prosperous tax base was important as a prerequisite to an appropriately high state income. If the first half of his advice to “nurture the basis and draw on it sedately” sounds potentially interventionist, his intent is clarified by his metaphor of the woodcutter ensuring a lasting source of firewood by harvesting the twigs and branches while nurturing the rootstock – as opposed to digging it out. Prosperity would be the natural state of things if farmers stayed on the farm and devoted all their strength to farming, if craftsmen possessed and deployed the plain skills that were essential to their calling, and if merchants did their work of “circulation” (liutong 流通 – the commercial circulation of commodities) unimpeded. Sima’s program for positive nurturance was minimal. When good harvests glutted grain markets, the state should serve as buyer, and when harvest failures necessitated relief operations, farmers should have priority. The court and state authorities should lead society by insisting on sound workmanship and plain, functional wares. Policies toward business should proceed from the recognition that merchants have their minds set on profit alone and will desert their calling if no profit can be made. Sima wrote not in terms of boosting productivity but rather of ensuring the surpluses on which the state could amply draw. Sound fiscal governance prevailed when “other parties [i.e., the private sector] have surpluses and are not harmed even when we [i.e., the authorities] draw on them in large quantities.”58 While this is not a vision of economic growth, Sima could, without inconsistency, have joined Wang in affirming that the wisdom of “former generations” – presumably of statesmen – had been to rely on the world’s muscle power to bring forth its wealth, and draw on its wealth to provide for its expenditures. The habit of quoting out of context simplified versions of Sima Guang’s remark about the limitations of cosmic generativity can be traced back at least to the Yuan dynasty. In the original context, it made no difference whether the subject population or the state was mentioned first as holder of part of the wealth. Out of context, however, “if [the wealth] is not among the people it will be with the state” (to translate Sima’s wording literally) did not sound meaningful. The order therefore came to be inverted, as it is in a memorial by Zhang Fujing to the Jiajing Emperor (r. 1521–1566): if not with the state, or “authorities,” then with the people. Thus modified, the cliché intimated that any part of the cosmos’s ultimately limited output that was not claimed by the state would be available for holding by its subjects, 58

Sima Guang, “Lun caili shu,” in SMCJJ, 25:4b, 7b–10a.

406

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

exactly as was prescribed by that much older cliché, “store riches within the population.” The irony that this was not at all what Sima had advocated in 1062 went unheeded. Sima had alluded to the ancient cliché, but only to point out that while storing riches within the population was best and storing them in granaries and treasuries second-best, to have one’s granaries and storehouses empty and one’s population poor was perilous.59 After the newer cliché entered the rhetorical toolkit of Chinese political economy, many variants emerged. One could make the message even more explicit, as did a quixotic edict in the name of the Qianlong Emperor (r. 1735– 1796). Qianlong observed that if the world’s limited wealth were not “amassed above” it would be “dispersed below” – obviously a desirable state of affairs, for he proceeded to announce a one-year suspension of the land tax, the Qing state’s leading source of public revenue.60 One could change the emphasis, as did Zhang Juzheng with his “whether the wealth that Heaven generates is with the authorities or with the people, there is only so much of it.” One could, like Zhao Shiqing, embellish an argument about the impact of extortionate taxation with a nonspecific version: increasing one party’s share in the limited total output of wealth would inevitably be at the expense of the other party’s share – “the more there is of mine, the less there is of yours,” in other words.61 One could omit all reference to dichotomous shares, as in the edict of 1749 in which Qianlong expressed concern about a recent surge in the output of the state-run copper mines potentially leading to exhaustion of the ores.62 One could retain the reference to the public-sector–private-sector dichotomy but bide one’s time before invoking it, as Grand Secretary Shen Shixing did in 1586, in a “productivist” response to the limits of cosmic generativity: just as the tiny size of a patch of ground would not stop a rich family from growing fruit or vegetables on it, so a reasonable response to the gap between the present level and the ultimate ceiling of production would be to “realize the Earth’s [inherent but unexploited] benefits in order to augment the people’s wealth.” If, as now, vast tracts of arable land were left barren, those benefits were “neither with the state nor with the people.”63 59

60

61

62 63

Zhang Fujing 張孚敬, “Ying zhi chen yan” 應制陳言, in HMJSWB, 177:9b; Sima Guang, “Lun caili shu,” in SMCJJ, 25:3b. Cited in Helen Dunstan, “The Finance of Imperial Munificence: How Simple Quantitative Work Can Help Us Rethink High-Qing History,” T’oung Pao 100 (2014), 167. Zhang, “Chen liu shi shu,” and Zhao Shiqing 趙世卿, “Guanshui kuijian shu” 關稅虧 減疏, in HMJSWB, 324:10a, 411:6a. QSL, Gaozong shilu 高宗實錄, 345:11b. MSL, Shenzong shilu 神宗實錄, 172:4b–5a; cf. Pierre-Étienne Will, “Clear Waters versus Muddy Waters: The Zheng-Bai Irrigation System of Shaanxi Province in the Late-

407

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

The malleability of the form of words inherited from Sima Guang warns us against taking the idea of dichotomous shares of an ultimately fixed economic pie as an axiom of post-Tang Chinese political economy. Even the fixed-pie premise was not uncontestable. In assuming that copper was subject to the rule of limited cosmic generativity, Qianlong had logic and the empirical observation of responsible officials on his side, but there was an alternative to the view that Heaven and Earth had laid down finite quantities of ore. In 1734 a Manchu governor-general had advocated extracting different metals by turns because metals came into being “by congelation of Earth’s energy,” and their cycles of flourishing and declining production were asynchronous. Extraction policies that were adapted to these cycles would not induce exhaustion of the ores. Although the vision of metals as self-regenerating resources was too distant for practical Qing statecraft, the doctrine of their origin in congelation of Earth’s qi 氣 (energy, ether) was consistent with the cosmology of Zhu Xi, the most influential post-Tang Confucian thinker, and drew upon much earlier notions.64 Did a common penchant for conservatism link all those who noted that, resources being ultimately limited, less in state hands meant more available within the private sector? Sima Guang, who did not make such an observation, opposed radical fiscal and political innovations that spelled state selfaggrandizement and, potentially, the molding of a sociopolitical elite indoctrinated for subservience in an ideologically streamlined bureaucracy. After the manner of conservatives, he preferred that what he took to be the basic structures of society and the economy be left alone, that trends which undermined those structures be discouraged, and that inequalities of wealth be accepted as natural and functional.65 Those who adapted his words to imply that, as far as possible, resources were better placed with the people than with the state staked a claim that was logically consistent with Sima’s trust in society. So, arguably, did the Qing court in 1748 when insisting that only so much grain was produced, and that stockpiling in the state’s pricestabilizing granaries therefore necessarily reduced availability for the

64

65

Imperial Period,” in Mark Elvin and Liu Ts’ui-jung (eds.), Sediments of Time: Environment and Society in Chinese History (New York, Cambridge University Press, 1998), pp. 309–10. Helen Dunstan, “Official Thinking on Environmental Issues and the State’s Environmental Roles in Eighteenth-Century China,” in Elvin and Liu, Sediments of Time, p. 588; A.C. Graham, Disputers of the Tao: Philosophical Argument in Ancient China (La Salle, Open Court, 1989), p. 101; Fung Yu-lan, A History of Chinese Philosophy (Princeton, Princeton University Press, 1952–3), vol. 2, pp. 542–3. Peter K. Bol, “Government, Society, and State: On the Political Visions of Ssu-ma Kuang and Wang An-shih,” in Robert P. Hymes and Conrad Schirokauer (eds.), Ordering the World: Approaches to State and Society in Sung Dynasty China (Berkeley, University of California Press, 1993), pp. 177–80, 185.

408

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

public.66 The difference was that, this time, the court was adjusting its popular-subsistence policy in response to radical critiques of the timehonored “ever-normal” system, which embodied the one positively interventionist principle of government that Sima had endorsed in 1062. The course of action advocated by the radicals – making way for market forces – may sound “modern” and “progressive.” Let us look more closely.

Market Forces and Officialdom’s Parental Duties Despite variations in governmental capacity and motivation over four dynasties, the Chinese tradition regarding the management of the agricultural surplus that a universally toiling peasantry was expected to produce was interventionist. Central to that tradition was faith in the “ever-normal” technique, explained thus in a famine-relief manual compiled by Dong Wei (d. 1217): The ever-normal method is for the sole purpose of making relief sales in times of harvest failure. When grain is cheap, one buys at an increased price, so that the cheapness does not harm the farmers, and when it is dear, one sells at a reduced price, so that the dearness does not afflict the people – hence the name “ever-normal.”67

Whether the predominant use of the stockpiles in “ever-normal granaries” in any given period was for smoothing price fluctuations or as precautionary reserves for famine-relief operations, there was usually consensus that storing grain to sustain the lives of vulnerable subjects was an appropriate responsibility to give the “parent officials” (county magistrates) charged with transmitting monarchical benevolence to a childlike populace.68 Less central in the Chinese tradition than direct governmental management of a fraction of the agricultural surplus was a range of coercive, regulatory, and hortatory practices intended to quell antisocial profiteering in the handling of the larger portion of the surplus that was left for privatesector management. These practices were far from constituting a rigid system of controls, and it is difficult to assess their actual prevalence, given the widespread availability of arguments against them, not to mention a certain ambivalence at the central government level. Generally speaking, the 66 67 68

QSL, Gaozong shilu 高宗實錄, 313:34a, 318:44a, 319:6a, 320:23b, 28b, 322:4b. JHHMS, 2:1b. Cf. Pierre-Étienne Will, Bureaucracy and Famine in Eighteenth-Century China (Stanford, Stanford University Press, 1990), pp. 182–3, 186–7, 277–9.

409

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

underlying assumption was that government action was often necessary to keep grain flowing to hungry consumers. It was those interventionist practices that blocked or sabotaged free circulation which attracted the most criticism, provoking already in the Song dynasty what William Rowe has called “an alternative economic discourse” that advocated respect for market forces.69 The eleventh-century harbingers of the market-conscious discourse were little stories extolling wise officials who rejected calls to depress the price of grain by fiat or attracted grain cargoes to their jurisdictions by assuring merchants of the high prices obtaining there. One story, possibly apocryphal, depicts a famous statesman artificially raising prices to lure grain merchants to his famine-threatened prefecture.70 Two others simplified the historical record to the point of distortion. It was Sima Guang who reduced the somewhat diverse record of famine-prevention efforts made by the early ninth-century official Lu Tan into a single-stranded fable hinting that subsistence crises could be halted by the mere convergence of importers on jurisdictions in which grain prices had been allowed to rise.71 Sima also contributed to manufacturing the legend of his contemporary, Zhao Bian, an official whose real-life approach to famine relief was the epitome of careful planning.72 Zhao was probably best known to readers of famine-relief manuals down the ages through the story, recorded by Sima, of the savior who attracted rice by being the only prefect in a dearth-stricken region to post wayside notices authorizing the sale of grain “at an increased price.” His counterparts were diligently publicizing bans on raising rice prices.73 All three stories were reproduced repeatedly in famine-relief manuals and more general compilations from the twelfth century to the eighteenth. Given the economic changes discussed elsewhere in this volume, it is surprising neither that the stories, all of which are set south of the Yangzi, emerged during the Song (drawing, in the Lu Tan case, on late Tang source material), nor that they were soon complemented by more abstract advocacy of what, to exclude culturally inappropriate baggage, we may call not a laissez-faire approach but a ting qi ziran 聽其自然 one. Ting qi ziran means “let matters 69

70 71

72

William T. Rowe, Saving the World: Chen Hongmou and Elite Consciousness in EighteenthCentury China (Stanford, Stanford University Press, 2001), p. 183. On the interventionist tradition, see Helen Dunstan, State or Merchant? Political Economy and Political Process in 1740s China (Cambridge, MA, Harvard University Asia Center, 2006), Chapters 1–2. NGZML, 3:20. Compare ZZTJ, 237, p. 7653 with Li Ao 李翱, “Gu Dongchuan jiedushi Lu gong zhuan” 故東川節度使盧公傳, in WYYH, 792, p. 4189. NFLG, 19:12a–13b. 73 SSJW, p. 285.

410

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

take their [natural] course.” Dong Wei’s contemporary Ni Si wrote that as merchants’ profit-seeking nature drew them to markets on which prices were elevated, the best plan was to “let matters take their course,” for prices would stabilize spontaneously once the imports had arrived.74 Dong Wei himself adumbrated a notion of what Robert Hymes has called a “self-regulating process.” The emerging elite understanding of price mechanisms may, however, have seemed commonplace to those frequenters of Southern Song marketplaces who, according to Ouyang Shoudao, were wont to say, “With prices, no sooner has one heard that they are cheap than they are dear, and no sooner has one heard that they are dear than they are cheap.”75 Dong Wei’s most interesting contribution to China’s indigenous tradition of free-market discourse was the dual nature of his substantive argument against “forcing down the price.” In warning that such action deterred importation he did not surpass the anecdotal tradition, but he further argued that attracting imports was crucial for incentivizing local grain holders to sell their grain pre-emptively, before the increase in supply made prices tumble, whereas the food shortage attendant on a lack of imports would decrease the likelihood of sales.76 Dong thus paid linked attention to both the spatial and the temporal aspects of the implicit model that underpinned post-Tang understandings of the workings of the grain market. In the spatial aspect, the key question was whether geographic unit A could compete successfully for commercial grain supplies with geographic units B, C, and D. In the simple version that informed both the anecdotal and the abstract discourse against price controls, the model represented ability to attract or retain lifesaving grain as a function of relative price levels. In the temporal aspect, the question was how the controller of substantial stocks of grain would time his sales. The timing of sales was expected to be determined by responsiveness to local price signals.77 The lurking danger, from the local consumer’s perspective, was that the most alluring signals would, on the contrary, be those of the interregional grain market. Zhou Kongjiao (fl. 1580), for example, warned that the inevitable response of local wealthy households to price controls – “barring the door to purchasing of grain” – would actually be a smokescreen 74

75

76 77

Essay by Ni Si 倪思, quoted in Wang Deyi 王德毅, Songdai zaihuang de jiuji zhengce 宋 代灾荒的救济政策 (Taipei, Taiwan shangwu yinshuguan, 1970), pp. 157–8. Robert P. Hymes, “Moral Duty and Self-Regulating Process in Southern Sung Views of Famine Relief,” in Hymes and Schirokauer, Ordering the World, pp. 295–302; Ouyang Shoudao 歐陽守道, “Yu Wang Jizhou lun junzheng shu” 與王吉州論郡政書, in XZWJ, 4:5a. JHHMS, 2:15a–16a (translated in Hymes, “Moral Duty,” pp. 295–6). Dunstan, State or Merchant?, pp. 140–3.

411

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

for something more dangerous for the surrounding community – surreptitious sales to “merchants from afar.”78 Officials might write as though the local hoards were destined for the local market, but in the larger picture the host community was but a unit A with many counterparts, including distant urban markets. What “progress,” if any, is discernible over the six centuries after Ni Si and Dong Wei introduced some simple theory to the anecdote-based free-market discourse? Research thus far has uncovered a few more-developed expressions of belief in market forces, but what prompted these was not any inherent evolutionary dynamic but rather the short-term stimulus of perceived interventionist excesses. Pierre-Étienne Will distills from a memorial of 1763 a model that incorporates the role of competition among brokers into its implicit formula for a well-functioning, self-regulating grain trade: so far from manipulating prices, the state-licensed brokers were forced to operate fairly, with due regard for grain importers’ interest in a satisfactory return on capital, by the freedom of those importers to choose the broker through whom they would sell their cargo. The 1763 memorial might never have been written but for the requirement placed on its author to explain his abolition, several years earlier, of a layer of control that had been imposed under his predecessor, with serious consequences for the Guangdong consumers who depended on rice from Guangxi.79 Other research emphasizes the context in which the classicist Fang Bao suggested, as he did in 1738, that price incentives assure the optimal spatial distribution of commercialized grain surpluses. It was certainly an advance in sophistication for Fang to argue that the highest prices naturally led the merchants to the areas of greatest need, and that once the shortage in those areas had been relieved by imports, merchant interest adjusted to the new configuration of price levels. However, the argument was shaped as a refutation of the assumptions underlying an imbecilic control procedure mandated by the central government’s Ministry of Revenue for channeling the cargoes of grain merchants to disaster-afflicted regions.80 In 78

79

80

Treatise by Zhou Kongjiao 周孔教, quoted in Zhou Zhiyuan 周致远, Mingdai huangzheng wenxian yanjiu 明代荒政文献研究 (Hefei, Anhui daxue chubanshe, 2007), p. 212. Pierre-Éitenne Will, “Discussions about the Market-Place and the Market Principle in Eighteenth-Century Guangdong,” in Zhongguo haiyang fazhan shi lunwen ji 中國海洋發 展史論文集 7 (Taipei, Academia Sinica, Sun Yat-sen Institute of Humanities and Social Sciences, 1999), pp. 335–46. For Will’s partial translation of the 1763 memorial, see “A Memorial on Grain Prices, the Grain Trade, and Government-Controlled Brokerages,” in de Bary and Lufrano, Sources of Chinese Tradition, 2, pp. 169–72. Dunstan, Conflicting Counsels, pp. 304–7; FBJ, 2, pp. 554–5 (translated in Conflicting Counsels, pp. 324–6).

412

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

the history of Chinese economic thought, Fang’s intervention may have been a breakthrough, but it was not a turning point. The same could be said of some interesting snippets of argument advanced in 1743 to justify the notion that there were positive advantages in distributing famine relief in monetary form. Although at the time this practice was officially regarded as an unfortunate expedient, to be adopted when the precautionary grain stockpiles did not suffice, there was a long pedigree of arguments for substituting money, at least in relatively mild disasters. These arguments, however, were couched in terms of the physical and other properties of grain and money, and/or of the satisfaction of needs. Money, it was argued, was cheaper to transport overland than grain, while the risk of pilfering was easier to control than in the case of grain. Monetary doles enabled famine victims to meet non-food needs, to feed their families for longer periods using cheaper foodstuffs, or to redeem pawned farming tools or seed grain and resume their livelihoods. They could empower the resourceful for self-help and spare self-respecting peasants the indignity of being herded off to feeding stations. While claims that it was precisely for remote areas that monetary distributions were particularly appropriate would not have made sense without an assumption that the doles could be exchanged for food there, in the absence even of references to the processes through which this would be possible, one cannot cite these claims as evidence of belief in market forces.81 There is a qualitative difference between these arguments and that with which the central government official Li Qingzhi sought to clinch the case in 1743. After noting that money made possible the purchase of coarse foodstuffs and the pursuit of victim-determined pathways to survival, Li claimed that “if famine victims have money, merchants are bound to come plying their trade,” and that commercial imports would thus save the situation. The supervising censor Wu Wei alluded more explicitly to the merchant perspective when he claimed two months later that “once the folk have money, it will be possible to sell them rice, on which merchants from near and far will all come one after the other.”82 The predictions seem casual to the point of arbitrariness, but they were distinctive among claims for the efficacy of issuing relief in monetary form because they signified an increase in the 81

82

JHHMS, 1:31b, 2:9a–b, “shiyi” supplement, 11a–b; the Ming sources cited in Zhou, Mingdai huangzheng, pp. 173–4; Helen Dunstan, “A Different Trajectory: MarketConsciousness in Chinese Political Economy, 800–1800,” Journal of Chinese History 4.1 (2020), 73–7. Dunstan, State or Merchant?, pp. 181–3, 238–40 (translations slightly modified).

413

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

supply of staple cereal food to the afflicted market. They also implied a refinement of the conventional assumption that grain gravitates to wherever the price level suggests the largest profit margin. This will happen only if the market can bear (that is, if enough people have the means to pay) merchant-attracting prices. Only if the demand is effectual will the gravitational pull be effectual. The context in which these remarks were made goes far to explain their limited impact on premodern Chinese economic theory. Two factors in combination – an intensification of an inflationary trend in grain prices, particularly in the Yangzi valley, and a quixotic and generally unsuccessful attempt vastly to increase the anti-famine reserves held in the ever-normal granaries – had provoked a major intragovernmental debate on the value of the ever-normal system. By 1743, this debate was in its fifth year, and rice prices at the entrepôt of Suzhou were surging. Although recent research casts doubt on the widespread accusations that procurement for the government stockpiles was causing the price rises, there was some plausibility to claims that public-sector buying caused adverse market reactions that private-sector purchasing did not. There was power in the contrasting images of, on the one hand, the conspicuous, commercially insensitive official procurement mission, constrained by quotas, deadlines, and fear of bureaucratic auditors, and, on the other, the astute merchant who did his buying while prices were low and slipped away before increases jeopardized his profit margins. In the later 1740s, some thoughtful defenses of privatesector stockpiling reinforced a growing suspicion that the ever-normal system did not entirely deserve its conventional recognition as the socially desirable alternative to merchant management of the annual grain surplus.83 Wu Wei was exceptionally radical in advocating, as he did in 1739 and 1743, that the present ever-normal granary reserves be sold in their entirety. In the last years of this fifteen-year debate, however, a proposal that was akin to Wu’s but less extreme was one of three initiatives by senior officials to advocate that, at least in designated regions with good water transport, the state keep only vestiges of its time-honored role of entering the grain market in order to adjust it and, in the event of famine, partially replace it. Would not the efficacious waterborne grain trade function all the better minus the pressure of

83

Dunstan, State or Merchant?, pp. 97–8, 132–8, 164–87, 194–211, 292–306.

414

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

official competition?84 However, to treat the profit motive as a panacea was not a real option for mid-eighteenth-century officialdom. Scriptural teaching, raison d’état, and the living witness of experience combined to forge consensus that a governmental apparatus accountable to Heaven and vulnerable on Earth must not leave famine-stricken subjects to the fate that would be theirs but for active measures against mass starvation. For Wu Wei and others in the 1740s, to advocate the distribution of relief in monetary form was to argue that the state could take such measures without maintaining vast grain stockpiles. However, the elevation of the tolerated expedient into a positive famine-relief policy was too radical for the central government decision makers who dismissed the idea in 1743. Although, by 1750, the unachievable storage targets for the ever-normal granaries had twice been drastically cut, the caution of the emperor’s wellinformed advisers ensured that maintenance of significant grain reserves remained imperial policy.85 The debate’s settlement removed the stimulus for paradigm-shifting argumentation in official discourse on state intervention in the grain trade.

Conclusion The years 1069 and 1743 saw two contrasting moments in the history of thought and policy about the state’s role in that central arena of an agrarian economy, the grain market. In both years, liquidation of ever-normal stockpiles was envisaged by a radical. In 1069 the purpose was intensification of the state’s role in the economy through an agricultural credit scheme that might have contributed to growth through gains in productivity; in 1743 the purpose was to relax the pressure of state intervention buying on the grain market while encouraging territorial officials to use monetary fuel to power market forces. Wang Anshi’s initiative, as implemented, was perennially open to the criticism that it had wrought a shift in the distribution of wealth toward the authorities and away from the people. The Qianlong Emperor sought to justify his radical repudiation of the vision of the state as stockpiler extraordinaire by alleging maldistribution of cereal resources in favor of official storage, to the detriment of market availability. In the eleventh century, 84 85

Dunstan, State or Merchant?, pp. 181–3, 234–5, 343–4, 406–11. Dunstan, State or Merchant?, pp. 241–4, 350–6.

415

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan

state self-aggrandizement that surpassed the fiscal realm challenged complex vested interests and evoked a conservative response; in 1749 and the preceding years, by contrast, the conservative position lay in defending the moderate levels of state stockpiling that could be achieved in practice, while the most extreme opposition to state stockpiling was positively avantgarde. Of the two contrasting radical initiatives, those of 1069 and 1743, which was the more progressive, if not modern? It is not obvious that the answer is “the latter.” Tempting as it may be to hail the likes of Wu Wei as apostles of a liberal enlightenment, they contributed little that was durable to the longestablished tradition of respect for market forces. The avant-garde welfare economics of the 1740s was encapsulated in the following remark, by the investigating censor Sun Hao: Although there may be a famine year, there will certainly not be exhaustion of all grain supplies. There is such a thing as having grain and fearing the want of silver, but if only one has silver, how will one fear the want of grain?86

In terms both of the consciousness that underlay such observations and of the inherited discursive tradition, Sun Hao, Wu Wei, and Li Qingzhi were well placed for launching a critique of what Amartya Sen has called the “food availability decline” understanding of famine, and for theorizing the role of money in empowering vulnerable consumers in the competition to attract inflows of grain.87 It remains to be demonstrated, however, that the later normalization of the practice of issuing relief in monetary form was underpinned by any such conceptual sophistication.88 The problem, if there was one, lay with state dominance of discourse, not resources. Traditions of didacticism entrenched the view that the state had a role to play in keeping the populace industrious; traditions of “rescuing” sustained the image of the populace as infantile and in need of state protection. Behind the institutional forces that curbed radical thought – through dismissive rejections by central government committees, for example – was the need for ideological bolstering of the 86 87

88

Dunstan, State or Merchant?, p. 237 (translation slightly modified). Amartya Sen, Poverty and Famines: An Essay on Entitlement and Deprivation (Oxford, Clarendon Press, 1982), pp. 7–8, 57. Will, Bureaucracy and Famine, pp. 276–7, 295–7; Lillian M. Li, Fighting Famine in North China: State, Market, and Environmental Decline, 1690s–1990s (Stanford, Stanford University Press, 2007), pp. 226–7.

416

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

Political Economy

imperial edifice. Wang Anshi would probably have understood a call for that “investment in growth-oriented public goods” the capacity for which, von Glahn suggests, the Qing state willfully denied itself (see the chapter by Lamouroux and von Glahn in this volume). But Wang would the have imposed a stranglehold on discourse.

Further Reading Dunstan, Helen, Conflicting Counsels to Confuse the Age: A Documentary Study of Political Economy in Qing China, 1644–1840 (Ann Arbor, University of Michigan, Center for Chinese Studies, 1996). Dunstan, Helen, “A Different Trajectory: Market-Consciousness in Chinese Political Economy, 800–1800,” Journal of Chinese History 4.1 (2020), 55–83. Dunstan, Helen, State or Merchant? Political Economy and Political Process in 1740s China (Cambridge, MA, Harvard University Asia Center, 2006). Elman, Benjamin and Martin Kern (eds.), Statecraft and Classical Learning: The Rituals of Zhou in East Asian History (Leiden, Brill, 2010). Elvin, Mark and Liu Ts’ui-jung (eds.), Sediments of Time: Environment and Society in Chinese History (New York, Cambridge University Press, 1998). Gao Wangling 高王凌, Huozhe de chuantong: Shiba shiji Zhongguo de jingji fazhan he zhengfu zhengce 活着的傳統: 十八世紀中國的經濟發展和政府政策 (Beijing, Zhongguo shehui kexue chubanshe, 1995). Hartwell, Robert M., “Historical Analogism, Public Policy, and Social Science in Eleventhand Twelfth-Century China,” American Historical Review 76.3 (1971), 690–727. Hu Jichuang 胡寄窗, A Concise History of Chinese Economic Thought (Beijing, Foreign Languages Press, 2009). Hu Jichuang 胡寄窗, Zhongguo jingji sixiang shi 中國經濟思想史, 3 vols. (Shanghai, Shanghai renmin chubanshe, 1962–1981). Hu Jichuang 胡寄窗 and Tan Min 談敏, Zhongguo caizheng sixiang shi 中國財政思想史 (Beijing, Zhongguo caizheng jingji chubanshe, 2016). Hymes, Robert P. and Conrad Schirokauer (eds.), Ordering the World: Approaches to State and Society in Sung Dynasty China (Berkeley, University of California Press, 1993). Kishimoto Mio 岸本美緒, Shindai Chu¯goku no bukka to keizai hendo¯ 清代中国の物価と 経済変動 (Tokyo, Kenbun shuppan, 1997). Li Huarui 李華瑞, Songdai jiuhuang shi gao 宋代救荒史稿, 2 vols. (Tianjin, Tianjin guji chubanshe, 2014). Metzger, T.A., “The State and Commerce in Imperial China,” Asian and African Studies 6 (1970), 23–46. Rowe, William T., Saving the World: Chen Hongmou and Elite Consciousness in EighteenthCentury China (Stanford, Stanford University Press, 2001). Smith, Paul Jakov, Taxing Heaven’s Storehouse: Horses, Bureaucrats, and the Destruction of the Sichuan Tea Industry, 1074–1224 (Cambridge, MA, Harvard University, Council on East Asian Studies, 1991).

417

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

helen dunstan Song, Jaeyoon, Traces of Grand Peace: Classics and State Activism in Imperial China (Cambridge, MA, Harvard University Asia Center, 2015). von Glahn, Richard, Fountain of Fortune: Money and Monetary Policy in China, 1000–1700 (Berkeley, University of California Press, 1996). Will, Pierre-Étienne, Bureaucracy and Famine in Eighteenth-Century China, trans. E. Forster (Stanford, Stanford University Press, 1990). Will, Pierre-Étienne, “Développement quantitatif et développement qualitatif en Chine à la fin de l’époque impériale,” Annales: Histoire, sciences sociales 49.4 (1994), 863–902. Will, Pierre-Étienne and R. Bin Wong, with James Lee, Nourish the People: The State Civilian Granary System in China, 1650–1850 (Ann Arbor, University of Michigan, Center for Chinese Studies, 1991).

418

https://doi.org/10.1017/9781108587334.012 Published online by Cambridge University Press

11

Law and the Market Economy billy k.l. so and sufumi so

As research over the past few decades, including the studies in this volume, has revealed, there is ample evidence indicating that a vibrant market economy, including highly specialized cash-crop farming, was practiced for sustained periods of time in certain parts of China from the northeastern and southeastern coasts to the Yangzi river delta and valley to the north China plain and the Shanxi plateau between the year 1000 and 1800. This chapter is concerned with the question whether the legal framework of imperial China facilitated or hindered the development of China’s market economy during this period. Imperial China was largely an agrarian society. Alongside the widespread agricultural activity came the long and winding history of landownership that eventually saw the legal protection of private property rights in land during the Song dynasty (960–1279), which helped regulate economic activity more effectively. The law was highly relevant to the lives of the farming population, as it required the legal establishment of landownership as well as land transfers to family members and non-family members in order to sustain continuity and development in agriculture. Evidence for legal disputes in relation to inheritance, mortgage, and sales transactions is found even in preSong historical documents, but becomes considerably more abundant in the Song and subsequent dynasties. Discussions of law and custom pertaining to land and the agrarian economy can also be found elsewhere in this volume, in the chapters on the rural economy by Kenneth Pomeranz and property rights and factor markets by Kishimoto Mio. This chapter is mainly focused on the effects of commercial law on economic activity in China. To ground our discussion, we first provide an overview of the concept of law and a legal

419

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

system in China from the year 1000 to 1800, followed by an examination of the role of law in China’s market economy during the same period. We will then discuss some selected cases of the application of commercial law to the regulation of commercial activities during the period.

Traditional Chinese Law: An Overview The Concepts of Tianli (Heavenly Reason), Guofa (Imperial Law), and Renqing (Human Feelings) In the twentieth century, the prevailing view was that law in traditional Chinese society was no more than an instrument to instill fear in the population and maintain social control by the arbitrary decisions of rulers. There was no sort of moral or transcendental justification but imperial authority itself to guide or constrain the use of law. Recent research, however, suggests that the vitality of traditional Chinese law was maintained through a set of customary rules and practices and that these norms operated holistically, mirroring the values permeating Chinese society. The subject of law, as it turned out, is far more complex than was previously thought. The newer scholarship has extended the scope of analysis in both ideological and practical dimensions and presents a tripartite approach encompassing the seemingly unrelated notions of “heavenly reason” (tianli 天理), “imperial law” (guofa 國法), and “human feelings” (renqing 人情) to clarify the concept of imperial Chinese law. “Heavenly reason” constituted the philosophical core of Song dynasty Neo-Confucianism. Many thinkers of the time focused on the cosmological and ontological underpinnings of li, “reason” or “principle.” Despite disputes over its precise definition, there seems to have been general agreement that li should refer to the underlying reason and order of nature (cosmology) as reflected in its organic forms (ontology). This view posits the fundamental connection between the concept of heavenly reason on the one hand and morality and justice in human society on the other. Neo-Confucian cosmological views placed particular importance on the simultaneously transcendent and immanent character of heavenly reason. Hence, one can comprehend a creative universe, because the creativity embodied in the universe is the same as the moral creativity that emanates from human beings. It should also be noted that since as far back as the Han dynasty (202 B C E –220 C E ) the doctrine of the Mandate of Heaven (tianming 天命), which is closely connected to the idea of heavenly reason, had been invoked

420

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

as the moral justification for the power and the sanctity of the emperor and imperial rule. Both heavenly reason and the Mandate of Heaven, it was believed, determine every phenomenon of reality and serve as the basis and the point of departure for moral knowledge and action. China’s first laws originated from customs, traditions, and declarations by regional rulers in the pre-imperial era. Over the centuries, divergent legal principles merged in the laws approved by each emperor, which culminated in the promulgation of the earliest extant comprehensive law code, the Tang Code, in 653 C E. Each subsequent ruling dynasty revised and added to the code of the previous dynasty. In other words, the emperor was the center of the legal system in imperial China and the emperor’s right to rule was believed to have been bestowed by Heaven (tian 天). Herein lies the intimate relationship between heavenly principle and imperial law (guofa). Considering this, it can be postulated that imperial law could disregard secular interests or worldly politics to serve justice as understood in terms of heavenly reason. In fact, recent studies of Chinese legal history have identified such an orientation in legal documents and judicial practice from the Song to the Qing. Proper legal procedures meant following the cosmological order of heavenly reason, and procedural justice referred to the fact that the rules based on the cosmological principle of heavenly reason were observed in legal proceedings. “Human feelings” (renqing) connoted, within the legal context of the period in question, a set of social norms, including emotions and intuitions as well as judgmental sentiments, by which one has to abide in order to get along with other people in society. As such, the notion of human feelings was instrumental in forming judgments about the rightness or wrongness of actions under certain circumstances. Any resulting orders or judgments had significant social and cultural consequences. Further, human feelings were organically linked to the heavenly reason that represented the transcendent source of knowledge and ideas in making judgments on judicial matters. For instance, if a civil or commercial dispute was handled extrajudicially, the result still had to follow the cosmological principle of heavenly reason. The notion of human feelings was readily applied in the implementation of imperial law. If it was used properly, no conflict should arise out of the proceedings simply because both human feelings and imperial law merged in heavenly reason.

The Evolving Body of Written Law The Tang Code (Tang lü 唐律), the oldest surviving comprehensive legal code, is considered to be the paramount summation of imperial China’s 421

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

legal achievement. It became the basis on which the codes of the later dynasties were developed, albeit using different organizational structures, with the last dynastic code of imperial China being the Statutes and Sub-statutes of Great Qing (Da Qing lüli 大清律例). Law codes in imperial China comprised statutes (lü 律, tiao 條, or lütiao 律條), ordinances (ling 令), imperial instructions (chi 敕) on judicial principles, legal precedents (li 例) that would guide judges in deciding similar future cases, supplementary regulations (ge 格), and legal standards (shi 式) promulgated by central and local government authorities, among which statutes received the highest status of written law. Whenever new problems emerged or a new agenda came to the forefront, emperors issued new edicts, which resulted in the multiplication of statutes, rules, regulations, and the like and transformed the legal code into a hierarchically structured compilation that had binding legal effect.1 As to commercial law, it is hard to understand fully how it was practiced during the Song and Yuan dynasties given the limited documentation relating to legal cases available, while a clearer understanding of its practice may be reached in the subsequent Ming and Qing dynasties. Regardless, there are historical sources that can inform us of certain aspects of commercial laws in each of those periods. The promulgation of legal rules concerning commercial matters occurred at different times, taking varied forms. Commercial law in Ming and Qing China was largely based on Song precedents, which retained most of the provisions of the Tang Code, but it also incorporated some elements of the Yuan legal code and included new laws and revised statutes. The commercial statutes and sub-statutes of the Ming and Qing codes were grouped into the two new sections of “monetary obligations” (qianzhai 錢債) and “markets” (shichan 市廛) under the category of “household law” (hulü 戶律), and the section on “violence and robbery” (zeidao 賊 盜) under “penal affairs” (xinglü 刑律).2

1

2

Umehara Kaoru 梅原郁, So¯dai shiho¯ seido kenkyu¯ 宋代司法制度研究 (Tokyo, So¯bunsha, 2006), pp. 757–848; Jinfan Zhang, The Tradition and Modern Transition of Chinese Law, 3rd ed. (Heidelberg, Springer, 2014), pp. 280–1; Dai Jianguo 戴建國 and Guo Dongxu 郭東旭, Nan Song fazhi shi 南宋法制史 (Beijing, Renmin chubanshe, 2011), pp. 1–37. DML, pp. 81–4, 143; DQLL, pp. 269–77, 412–15; cf. Junjian Jing, “Legislation Related to the Civil Economy in the Qing Dynasty,” in Kathryn Bernhardt and Philip C.C. Huang (eds.), Civil Law in Qing and Republican China (Stanford, Stanford University Press, 1994), pp. 42–84.

422

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

The Judicial System Recent research points to the prevalence of civil and commercial litigation cases involving commoners during the Song dynasty. Apparently the central authorities actively encouraged commoner subjects to consider amicable settlement to resolve their disputes before resorting to adjudication. Such efforts, however, did not help reduce the growing number of people who would turn to judicial resolution. In Song China, commercial litigation cases were handled mainly at the level of the county (xian 縣) – numbering roughly 1,200 at any given time – which served as the lowest unit of the empire’s field administration. The chief officer of the county was the magistrate, who administered the law with the assistance of his subordinates and presided over all cases, including both civil and commercial litigation matters, that occurred within the county. All judicial rulings made at the county level were regularly reviewed by the next higher level of the administrative hierarchy, the prefecture, variously designated as “superior” (fu 府), “standard” (zhou 州), “military” (jun 軍), and “industrial” (jian 監) prefectures, which numbered more than 300 in Song times. The next higher level of Song government, the “route” or circuit (lu 路), which varied in number from fifteen to more than twenty at different times while undergoing spatial modifications, supervised all prefecture-level administrative affairs, including judicial decisions, recommendations, and appeal petitions in commercial and all other cases. Over time, prefectureand circuit-level administrative work was separated into a number of specialized tasks. Such trends led to the emergence of officers well versed in the law and experienced in handling legal matters who would provide their superiors with technical and legal assistance in the adjudication process and the postadjudication review. While serious cases that could result in punishments such as death, penal servitude, and life exile eventually had to be adjudicated at the highest court at the central government level, civil and commercial litigation cases seldom went beyond the circuit level. The flow of history is always affected by time, continuity, and change. The dynastic regime transition from Song to Yuan was no exception. The Mongolruled Yuan dynasty (1271–1368) became the first non-Chinese regime to rule over all of China as a result of the Mongol conquest of the Southern Song in 1279. One notable new feature of the Yuan dynasty was the introduction of a legally mandated caste system based on an ethnic hierarchy that itself was defined in terms of perceived allegiance to the Mongol rulers. Yuan subjects were divided into four classes with the Mongols on top, followed by “affiliated

423

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

peoples” (semuren 色目人, mostly merchants and professional groups – many of them Muslims – originally from West and Central Asia), “Han people” (hanren 漢人, Han Chinese subjects of the Jin dynasty that ruled northern China before the Mongol conquest of 1235), and finally “Southerners” (nanren 南人, former subjects of the Southern Song dynasty). Under this caste system, the Mongols and the “affiliated peoples” enjoyed a variety of legal privileges, while the “Han people” and “Southerners” suffered from manifold legal discriminations. The hierarchy required that people in each class should be judged and sentenced according to their ascribed “ethnic” identity. Cases involving members of more than one class were decided by a mixed judicial board consisting of representatives from the relevant classes. Regardless, the harsher treatment of Chinese in the bottom two classes was systematically reinforced by a law code. Their penalties were far more severe than for the top two classes for committing the same offenses in all legal disputes, not to mention civil and commercial litigation cases. Notwithstanding the discriminatory social and judicial policies under Yuan rule, it was likely that judicial officials routinely adhered strictly to the letter of the law as they were better trained and developed a strong sense of professionalism as legal experts in comparison to Song magistrates. Although the current knowledge of legal practices and institutions in Yuan China is too fragmentary to confirm the accuracy of such a comparison, there is no doubt that the legal system of the Yuan dynasty was created, as in previous dynasties, to maintain justice and achieve good governance. Furthermore, despite the four-class system, Yuan practice of commercial law bore striking similarities to its Song counterpart. The Ming and Qing dynasties witnessed remarkable population growth and an expanding market economy, which in turn engendered a surge in commercial disputes and in the number of individuals taking their disputes to courts of law. Still, commercial litigation cases were mostly handled at the county level and seldom moved beyond the provincial level (sheng 省) of government, which was an outgrowth of the regional branch secretariat (xingsheng 行省) system instituted by the Mongol rulers during the Yuan dynasty. Magistrates at the county level were often assisted by prefectural or provincial judicial officials in adjudicating cases and conducting postadjudication reviews. In Qing China, offices were created in certain localities for circuit intendants (daotai 道台) to function as specialized intermediaries at the prefectural and county levels of administration, including judicial responsibilities. Intendants were assigned by the provincial government and they remained under the control of the provincial governor. 424

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

Law and Economic Development The Role of Law in Economic Development Did law matter for the vibrant market economy that developed in China beginning in the Song period? Did it provide a regulatory framework to facilitate economic activity? Or was it irrelevant to market-driven economic phenomena that sprang up in some regions before 1800? To answer these questions, we shall first look at the insights offered by what has become known as the New Institutional Economics (NIE). We will then discuss two opposing views on the effects of the regulatory framework on economic vibrancy in imperial China. The NIE is a multidisciplinary approach, drawing on such fields as economics, history, sociology, political science, business organization, and law, that ascribes a crucial role to law in the development of market economies. According to Douglass North, regarded as the progenitor of the NIE, the underlying determinant of economic performance consists of the political, social, and economic institutions that form the incentive structure of a society. For North, institutions are “the humanly devised constraints that shape human interaction,” including both formal rules such as laws and constitutions and informal norms of behavior (e.g., customs, taboos, cultural myths).3 The institutional environment or the “rules of the game” in which organizational forms and transactions are embedded influences and shapes human psychology and behavior. Such effects have an impact on everyday economic activity and, by extension, on economic performance. In North’s view, institutions that act to lower transaction costs and facilitate exchange are the key to market-based economic growth. Furthermore, according to North, institutional change plays a crucial role in fostering overall economic growth and changes in relative prices and technological innovations. When these changes occur, one or both parties in a transaction may consider changing the rules of their agreement or contract to benefit one or both sides, thereby introducing a new institutional arrangement. North’s groundbreaking work and a great many subsequent studies, which have contributed to an increased understanding of institutions in the economic, political, and social realms of action, draw mainly on extensive research in and knowledge of European development. Eurocentric as it 3

Douglass C. North, Institutions, Institutional Change, and Economic Performance (Cambridge, Cambridge University Press, 1990), p. 3. See also Billy K.L. So, Prosperity, Region, and Institutions in Maritime China: The South Fukien Pattern, 946–1368 (Cambridge, MA, Harvard University Asia Center, 2000).

425

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

may have been in its origins, the NIE approach nonetheless bears broader implications and can shed light on the mechanisms underlying development (or its absence) in premodern economies in other historical contexts such as China. It should be noted, however, that a China–Europe comparison (e.g., the question of why Europe, but not China, developed an Industrial Revolution) is not our intention in this chapter. Rather, our concern is with the role of law in the market economy of late imperial China that witnessed the rapid development of economic forces. For instance, was there a legal and regulatory framework that helped market production continue and spread far and wide for extended periods of time? Did it also facilitate interregional and interprovincial trade? Concerning the role of law in China’s market economy in premodern times, scholars are generally split into two camps. In one camp, which prevailed through most of the twentieth century, are scholars who argue that imperial Chinese law stretching back two millennia was predominantly a system of penal law focused on cases pertaining to criminal justice, and thus the law rarely played a role in fostering economic development. According to these scholars, a central concern of the Chinese empire was the effective administration of criminal justice in order to consolidate and maintain imperial power. The empire’s legal system was not meant to protect the rights of individuals, such as securing property rights. These scholars also emphasize the general antipathy toward litigation in Chinese culture. Commercial disputes pitting the claims of one party against another were usually resolved extrajudicially through, for instance, guild mediation, not through the courts or the application of legal rules enforced by the state. One common extrajudicial mechanism for dispute settlement was through advisory and mediation services offered by the disputing parties’ relatives, village elders, local scholars, and people acting as middlemen, as well as nongovernmental organizations such as community compacts (xiangyue 鄉約), kinship networks, and guilds. In other words, commercial and business behavior was largely constrained by extrajudicial settlement measures. It can thus be postulated that the development of a market economy in premodern China was carried out in a legal vacuum. Essentially, scholars of this camp hold the view that law was irrelevant to the Chinese market economy until the imperialist European powers imposed a series of unequal treaties on the Qing government in the nineteenth century. For scholars of the other camp, there is a puzzle yet to be solved, especially in light of the NIE’s foundational insights. The puzzle is why China’s market economy remained vibrant (albeit with some ups and downs) without 426

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

a proper legal framework over periods spanning several centuries. Over the past three decades, unlike for much of the twentieth century, premodern Chinese civil and commercial law has attracted scholarly attention to an astonishing degree because of the increased amount of primary source materials for legal history readily available to scholars, and also because of the phenomenal expansion of legal study in China. As a result, new empirical evidence has been presented on the role of law in premodern China’s market economy, which enables us to put to the test the above-mentioned view of Chinese law as overwhelmingly penal in nature.4 While researchers are still far from arriving at any definitive conclusions, the subject matter is gaining new perspectives. In this chapter we summarize the current knowledge about Chinese commercial law as well as the impact of regulatory dynamics on market behavior during the Song–Qing era, before the reality and severity of the European penetration into China, economic and otherwise, began in the nineteenth century.

The Purposes and Functions of Commercial Law Supposing that law indeed played a role in how the market economy developed in premodern China, did Chinese society or the courts of law take law and legal institutions seriously in civil and commercial cases? What were the important considerations of magistrates or judicial officers for adjudicating cases involving business interests? What were the courts’ preferred outcomes in those cases? In civil litigation, what might have most influenced the judicial decisionmaking process: Confucian values or the normative power of positive law (i.e., legal rules enacted by imperial authorities)? In Qing China, for instance, local magistrates, who were not necessarily trained as legal professionals but were well versed in the Confucian classics, would still rely on the law in deciding a case. However, they were not restricted to the rigid provisions of codified law. They were allowed to exercise their discretion to consider Confucian values or “human feelings” in their interpretation of a statutory rule.5 On the other hand, under the foreign rule of the Mongols during the Yuan dynasty, court decisions in civil cases were generally guided by existing 4

5

For examples, see So, Prosperity, Region, and Institutions; Madeleine Zelin, The Merchants of Zigong: Industrial Entrepreneurship in Early Modern China (New York, Columbia University Press, 2005); Qiu Pengsheng 邱澎生, Dang falü yushang jingji: Ming Qing Zhongguo de shangye falü 當法律遇上經濟: 明清中國的商業法律 (Taipei, Wunan chuban gongsi, 2008). Shiga Shu¯zo¯ 滋賀秀三, Shindai Chu¯goku no ho¯ to saiban 淸代中囯の法と裁判 (Tokyo, So¯bunsha, 1984), pp. 263–328.

427

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

laws rather than Confucian values, which is in line with the idea of positive law. While they might be tinged with elements of Confucian rhetoric, judicial decisions were actually reached by following the letter of the law and little effort was made to override the imperial statutory law with ethical principles of Confucianism.6 It is true that a Confucian moral tone is pervasive in most extant verdicts. However, it is clear that legal principles were upheld, especially in the appeals process and retrials or when judicial review was exercised. The majority of noncriminal cases heard in county courts were commercial litigation (albeit depending on the nature of the economy of a specific locale within the county), and a county magistrate had to adjudicate a large number of cases within a limited amount of time. The ability to resolve litigated cases swiftly and decisively became an important yardstick for assessing the degree of local government efficiency – which in turn affected the magistrate’s prospects for promotion. Thus there was sufficient incentive for local officials to work hard toward realizing the Confucian ideal of a society free of litigation (wusong 無 訟). Magistrates consciously sought to prevent litigation by promoting extrajudicial means of settling disputes rather than resorting to the courts. They also tried to terminate court cases with certainty and finality in order to avoid protracted litigation. To do so, disputes had to be resolved to the satisfaction of both parties. Figuring out how to curb litigation most effectively was a matter of urgent practical concern for county governments.7 Recent studies have found that the primary purpose of implementation of commercial laws was simply to maintain stability in the social order. It was an important agenda for imperial rulers who were given the Mandate of Heaven to rule society and therefore they assumed full responsibility for following the laws of Heaven to preserve harmony between the cosmic order and human society. Its unintended positive consequences were experienced by individuals and entities engaged in trade or business activities. The legal 6

7

Hu Xingdong 胡興東, Yuandai minshi falü zhidu yanjiu 元代民事法律制度研究 (Beijing, Zhongguo shehui kexue chubanshe, 2007), pp. 243–8. See also Bettine Birge, Women, Property, and Confucian Reaction in Sung and Yüan China (960–1368) (Cambridge, Cambridge University Press, 2002); Bettine Birge, Marriage and the Law in the Age of Khubilai Khan: Cases from the Yuan dianzhang (Cambridge, MA, Harvard University Press, 2017). Philip C.C. Huang, Civil Justice in China: Representation and Practice in the Qing (Stanford, Stanford University Press, 1996), pp. 51–109; Linxia Liang, Delivering Justice in Qing China: Civil Trials in the Magistrate’s Court (Oxford, Oxford University Press, 2007), pp. 248–58; So, Prosperity, Region, and Institutions, pp. 249–51. For a comprehensive collection of Ming and Qing commercial disputes, some of which are discussed in this chapter, see Fan Jinmin 范金民, Ming Qing shangye jiufen yu shangye susong 明清商業糾紛與商業 訴訟 (Nanjing, Nanjing daxue chubanshe, 2007).

428

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

framework served to shape many aspects of business operations and promoted a functional regulatory environment through, for example, the efficient resolution of disputes arising from the intensification and expansion of commercial activity.8 In other words, commercial law in late imperial China ultimately enabled the development of a prosperous market economy sustained over long periods of time, even if it was never meant to drive economic growth.

Laws and Cases Contract Law during the Song and Yuan Dynasties In Song China, as market activity became progressively widespread and intense, the number of commercial disputes increased significantly. There was a basic legal framework to deal with such disputes and govern business behavior. At the heart of this framework was contract law. Up until the 1980s, the notion prevailed that contract law did not play much of a role in economic or social development in premodern China. Recent research, however, has challenged this view. Today it is more commonly understood that during the Song and Yuan dynasties, contract law served an important function in creating obligations that were binding on all parties involved. Contracts (qi 契) were made legally binding and enforceable by a set series of conventional procedures, including fees for registering a contract with the government. Proper taxation for contract registration constituted an important factor in contract litigation disputes. In fact, the majority of the commercial-law disputes found in the extant records involved contract violations. It should be noted, however, that neither the Song nor the Yuan code laid out a set of laws regulating the making of contracts. Rather, the principles of contract law were embedded in different statutory provisions. Commercial contracts mentioned in the historical records of the Song and Yuan dynasties can be broadly divided into the three categories of transaction, commission, and rental agreements. There were at least three types of transaction agreements: (1) sales contracts for buying or selling goods, livestock, movable property such as boats and carts, and domestic servants of both sexes; (2) commodity futures contracts to facilitate advance payments for delivery of goods such as cash crops at a specified time in the future; and (3) credit contracts for sales in which the goods were shipped and delivered 8

So, Prosperity, Region, and Institutions, pp. 228–35; Qiu, Dang falü yushang jingji.

429

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

before payment was due. Sales contracts provided the best protection when ownership disputes or fraudulent schemes such as double selling arose. On the other hand, futures and credit contracts, the validity of which depended on trust between the parties involved, were more vulnerable to disputes and additional complications if they lacked effective regulatory systems. Commission agreements comprised at least five types: (1) contracts by which a principal entrusted the sale of goods or services to a person working as the principal’s full- or part-time agent; (2) contracts between a capital investor known as the “master household” (zhujia 主家) and a borrower who was not the household’s employee, by which the master household loaned money to the borrower with interest on mutually agreed terms and conditions for operating specified businesses such as pawnshops and warehouses;9 (3) contracts in which one party agreed to provide warehousing services for the other party’s goods; (4) commission contracts whereby agents collected taxes in cash or in kind on behalf of the government, or conducted business for merchants such as handling transactions and collecting their long-distance trade goods; and (5) brokerage contracts whereby a licensed local broker agreed to sell or buy goods for traders, often nonlocals, and received commission-based compensation in return. While most of the existing records of rental agreements are land tenancy contracts, two other types are particularly relevant for commercial activity: (1) commercial lease agreements made when an individual rented a commercial property (e.g., warehouse, retail space) from another individual,10 and (2) transportation rental agreements made when an individual paid for a transportation service, with shipping included, for domestic or overseas freight, be it a partial space in a ship, an entire ship, an overland vehicle, or a pack animal. The principles and practices of contract law from the Song and Yuan dynasties, as outlined above, were apparently carried into the following Ming and Qing periods without significant changes. However, regarding business partnerships formed among merchants, as well as between merchants and the state, by willingly pooling their funds in the business and sharing in its profits (liancai heben 連財合本), which began to appear during 9

10

See the cases entitled “Ling kubenqian ren jipin zhenzhuo jianhuan” 領庫本錢人既貧 斟酌監還, “Zhiku lixi yu sizhai butong” 質庫利息與私債不同, and “Beizhu lai kubenqian” 背主賴庫本錢 in QMJ, juan 9, 1, pp. 335–8. See the cases entitled “Buken huanlin tuiwu” 不肯還賃退屋, “Zhan linfang” 占賃房, “Linzhe xiwu” 賃者析屋, and “Lin renwu er zi qizao” 賃人屋而自起造 in QMJ, juan 6, 1, pp. 195–7; juan 9, 1, pp. 334–5.

430

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

the Song dynasty, we lack sufficient evidence about laws governing partnership to understand how partnership disputes were adjudicated in courts of law under the Song or Yuan. But records relating to the Ming and Qing laws regulating the formation and conduct of business partnerships are more readily available for historical research.

Partnership Disputes during the Ming and Qing Dynasties China experienced significant and expansive economic growth in the late Ming dynasty, notably following the formation of global trade networks beginning in the mid-sixteenth century. Riding this wave of commercial prosperity and looking to magnify the scale of their businesses, a growing number of merchants formed partnerships, which quickly became one of the most prevalent business forms. Alongside this trend emerged a large volume of partnership disputes. Like the Song and Yuan legal systems, neither the Ming Code nor the Qing Code contained a set of specific provisions regarding business partnerships. Studies focused on partnership litigation cases in the late eighteenth century have found that most of these cases were settled in court but in accordance with local business customs.11 In the absence of a formal regulatory framework, such settlement practices remained the norm through the late nineteenth century. Three cases are presented below to show some of the complex legal and contractual issues that showed up in business partnerships during the Ming and Qing dynasties. The first case shows that the judiciary took the partnership agreement as an enforceable contract. It also reveals that even people with power and influence were not always allowed to do things their way and run away from their responsibilities. It is the case of Feng Shunyu v. Hu Shengjian, coowners of an unspecified shop, brought to the judicial office of Guangzhou prefecture in the late 1620s. When the relationship between the two partners soured after they had been running the shop together for many years, Hu converted the partnership business into a sole proprietorship of his own through his contacts in government, without Feng’s consent. Feng filed his grievance against Hu in the prefectural court. The case, in which the statutes concerning fraud or business debt collection were not deemed applicable, 11

Zelin, Merchants of Zigong; Madeleine Zelin, “The Firm in Early Modern China,” Journal of Economic Behavior & Organization 71 (2009), 623–37; Robert Gardella, “Contracting Business Partnerships in Late Qing and Republican China: Paradigms and Patterns,” in Madeleine Zelin, Jonathan K. Ocko, and Robert Gardella (eds.), Contract and Property in Early Modern China (Stanford, Stanford University Press, 2004), pp. 327–47; Lin Lin, “The Evolution of Partnerships in China from the Perspective of Asset Partitioning,” Stanford Journal of Law, Business, & Finance 18.2 (2013), 215–49.

431

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

was handled by a judicial officer who effectively mediated and resolved the conflict, apparently to the satisfaction of both parties, by splitting the shop into two entities in accordance with their capital contributions and division of profits based on the terms specified in the partnership agreement. The same officer concluded that both men deserved criminal punishment for having caused much trouble in mishandling their business matters and recommended that his superior punish them by a beating with the heavy baton. Provincial judicial authorities took heed of his suggestion but considered Hu to be the only one who deserved the punishment of twenty blows for the “violent” (as labeled in the court record) and wrongful dissolution of the partnership resulting in a breach of contract.12 From the second case it can be inferred that a deceased man’s share in a business venture normally passed in full to his surviving wife and that even his concubines were entitled to receive a portion. It appears that there was a tacit understanding that certain unwritten obligations had to be voluntarily performed. This case happened around the same time as the first one. It involved Huang Zheng, a sojourner merchant, and Chen Chengwu, a local merchant, who had maintained a business partnership in Guangzhou for many years until Chen died unexpectedly while trading in other places in China. Chen’s surviving concubine, Madam Feng, and her son were immediately put in dire financial straits. The situation was so serious that Feng could not afford to transport Chen’s body back home for a proper burial. Feng thus asked Huang to bear the expenses, but her request was met with a flat refusal. Huang insisted that all of Chen’s partnership shares remained the property of the partnership business until Chen’s family closed his account. Feng took legal proceedings against Huang. An acting magistrate determined that Huang acted improperly and deserved the punishment of a beating with the heavy baton. The decision was fully endorsed by prefectural and provincial authorities. Furthermore, Huang was ordered to provide Feng with a minimum of fifty taels, approximately 10 percent of Chen’s stake in the partnership business, but he could hold the remaining 90 percent until Chen’s family resolved the matter.13 The third case suggests that there was a legal framework for partnerships to indemnify or protect partners who had left the business partnership. The case, included in the Baxian (Chongqing prefecture, Sichuan) archives, is about potential legal liabilities arising out of partnership activities. In 1780, Li Chengrang, a sojourner merchant, partnered with Chen Taiji, a Shaanxi 12

MSZ, p. 541.

13

MSZ, p. 419.

432

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

native, to open a grain mill in Chongqing. Chen put up an investment of 100 taels to purchase furniture, equipment, and horses for the business, of which Li acted as the sole manager for two years before the partnership was dissolved. At the time they dissolved their partnership, Li owed sixty taels to Chen, for which a debt agreement was drawn up. Afterwards, Chen returned to Shaanxi, while Li formed a new partnership with Feng Tinghui and Du Yuanzhen to run the same Chongqing factory. In the new partnership agreement, Li contributed capital assets in the form of the building, furniture, equipment, and horses estimated to be worth twenty-six taels, and Feng and Du invested eighty taels in cash to operate the business. Li thereafter entrusted the business operations entirely to Feng and Du and returned to his native place. As their business soon faced dwindling profits, Feng and Du closed the mill and rented out the factory building for twelve taels per year. A dispute occurred when Chen returned to Chongqing and claimed that Feng and Du, as Li’s business partners, were liable for repayment of Li’s loan from him.14 While its outcome is not known, this litigation case shows layers of complexity of liability issues as subsequent modifications were made to the original partnership agreement over the years.

Laws on Debt and Default The debt law enforced by the Song government played an important role in regulating market activities as it dealt with contractually binding obligations. Debts associated with landed property, which are far more complex than commercial debts and deserve fuller treatment, are discussed in Kishimoto’s chapter in this volume. Here we only examine commercial debt cases. The most useful provisions relating to the payment of debts arising under commercial contracts were in the “Misuse of Entrusted Property for Nonentrusted Purposes” section of the Song Penal Code (Song xingtong 宋刑統), promulgated in 963.15 This section included an article stipulating that a debtor who failed to make payment according to the terms of the contract agreement should receive the punishment of a beating of twenty blows with the light baton if the value of the debt exceeded one bolt (pi 匹) of cloth and payment was made within twenty days after the repayment date, with an additional degree of punishment if payment was further delayed for another twenty days, and that punishment should not exceed sixty blows with the 14 15

BXDA, 1, p. 386. SXT, pp. 467–8. See the same text under a different statute in TLSY, 2, p. 1803, 1807; cf. Wallace Johnson (trans.), The T’ang Code (Princeton, Princeton University Press, 1997), vol. 2, p. 464.

433

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

heavy baton regardless of the length of the delayed payment period beyond forty days. If the debt’s value was equivalent to between thirty and 200 bolts of cloth, an additional two degrees of punishment were imposed. If the debt’s value exceeded 200 or more bolts of cloth, two higher degrees of punishment were added. Also, the debt had to be fully repaid. A codicil appended to the article noted that according to the law, debt meant money or property owed by a borrower, with the exception of property pledged as collateral for a loan. As stipulated in another article, creditors were forbidden from using force to compel debtors to make restitution. Only government officials were allowed to enforce debt payments. Creditors using force to collect debts were charged with a criminal offense. Like many other parts of the Song Penal Code, the language used in these articles was no doubt inherited from the Tang Code. However, new provisions, which could have been based on an ordinance added later to the Tang Code, were included in the Song Penal Code to further regulate debt payments. In one such example, it was stipulated that both private and public lending of movable property would require drawing up a private loan contract signed by both parties, and that there should be no interference by public authorities with the process. It was further specified that the interest rate should not exceed 6 percent per month, and compound interest was prohibited. Moreover, it was required that regardless of the length of the loan, the total amount of interest penalties should not exceed the outstanding principal debt. Another stipulation obliged debtors who had lost their household property to pay their outstanding debts via labor services equivalent in value to the outstanding debt plus interest, and only men were allowed to undertake labor services to pay off the household’s debts. Disputes arising from such cases as interest accruing beyond the legal maximum, involuntary repayment exceeding an amount specified in the contract, and repayment of interest-free loans might be taken to government agencies for adjudication. It was stipulated that if the borrower was unable to repay the debt, his guarantor should be liable to pay off the debt on behalf of the borrower. There was yet another statutory requirement that parties involved in loan-with-interest transactions should maintain amicable relations. Furthermore, any party involved could notify the government of private contracts whereby an interest rate above the legal limit was applied to the principal. Such notification could result in the government confiscating the unpaid balance of principal and excess interest to punish lenders for extortionate demands. Moreover, those who informed the authorities about such cases of 434

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

extortion would be rewarded with the confiscated funds. In addition, according to a decree issued by the ministry of revenue (hubu 戶部), the maximum monthly interest rate for all collateralized private loans was capped at 4 percent rather than the 5 percent interest rate charged by the government for collateralized loans. As discussed above, the legal provisions in the Song Penal Code contain certain minute regulations for various debt issues such as interest rate ceilings, alternative debt payment methods, the resolution of disputes in court, the guarantor’s responsibilities, and rewards for those who divulged information about lawbreakers.16 The debt law explained thus far was implemented during the Southern Song period under the Song Penal Code, but with modifications resulting in the imposition of severe penalties. For instance, the fragments of the 1202 compilation of standing laws contain statutory provisions for loans and debts under which people could be punished by up to 100 blows with the heavy baton for failure to pay their debts. As mentioned earlier, compound interest was prohibited under the 963 code, but no stipulated penalties were included. Offenders against this law now received punishment of sixty blows with the heavy baton. Furthermore, the government would no longer handle the adjudication of disputes in connection with loans contracted five or more years earlier, in-kind loans converted into cash payments, or cash interest payments changed to payments with livestock. In addition, as paper money began to circulate widely and the pawn business proliferated, new laws were introduced in 1168 and 1187 to make it illegal to refuse to accept payment in paper money and to exempt pawnshops from complying with interest rate caps on ordinary loans, thus allowing pawnbrokers to earn the amount of interest greater than the principal.17 Some of these provisions remained in force during the Yuan dynasty as well. The extant sources contain decrees issued in 1266 that limited the total amount of interest payable by the borrower and prohibited compound interest and conversion of debts contracted in cash to payments in livestock. There is also a 1282 decree capping monthly interest rates at 3 percent. Many more provisions similar to the ones in the Song Penal Code were adopted into the Comprehensive Institutions of the Great

16 17

SXT, pp. 468–9; TSL, 2, p. 371; Dai and Guo, Nan Song fazhi shi, p. 120. Chapter 80, “Zamen: Boxi caiwu” 雜門：博戲財物, in QYTF, pp. 599–601; see also Chapter 32, “Caiyongmen: Liqian” 財用門: 理欠, in QYTF, pp. 347–8.

435

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

Yuan (Da Yuan tongzhi 大元通制), a compilation of imperial edicts and judicial decisions issued in 1323.18 In Ming and Qing times there was a surge in debt-related litigation. The four cases explained below illustrate how legal principles were applied in the judicial processes. The first case, which involved commodity futures, took place in Guangzhou sometime between 1628 and 1630, and resulted in the court issuing a judgment for debt payment. An itinerant merchant named Liao Shuwu and a few other merchants worked together buying lychee fruit in Guangzhou and selling them in other parts of China. Liao sued the father and son brokers Feng Jingya and Feng Xizhi and fruit shop owner Li Zhanran for failing to deliver the fruit, for which he had paid in advance a total of 182 taels. During the trial, the court found that one of the defendants, Feng Xizhi, had signed at least two contracts pertaining to the business deal in question and that he falsely claimed to have reimbursed the full amount. In the end, a guilty verdict was delivered and all three defendants were sentenced to a beating with the heavy baton for their deceptive conduct and false claims as well as having to reimburse Liao Shuwu and the other merchants for the advance payment.19 The second case, in which a loan default occurred, is concerned with who – other than the original creditor – was legally allowed to collect debts. In this particular case, brought against a debtor by a man who asked to exercise the right to receive loan payments on behalf of his fellow creditor, the court ended up siding with the plaintiff. Yang Xueming, a Fujian paper merchant, was engaged in the business of shipping paper to Hua county in Henan province, more than 1,000 kilometers north of Fujian, where retailers bought the paper on credit to sell it in their stores. One of the retailers was Li Shihua, who over time accumulated an outstanding debt to Yang amounting to 31.6 taels. When Yang was about to return to Fujian, he had Li sign a contract in which the unpaid amount was converted into an interestbearing loan. Eighteen years later, Yang’s brother Yang Mingxue went to Hua county and demanded that Li honor his debt obligation. Upon Li’s refusal to comply with the brother’s request, Yang Mingxue filed a lawsuit against Li in Hua county in the late 1620s or early 1630s. The court concluded that the case was simple and straightforward and ordered Li to pay his debt to plaintiff Yang Mingxue. Li agreed to the repayment obligations imposed 18

19

TZTG, p. 319. See also Hu, Yuandai minshi falü, pp. 102–6. For an overview of the Yuan legal system, see Paul Ch’en, Chinese Legal Tradition under the Mongols: The Code of 1291 as Reconstructed (Princeton, Princeton University Press, 1979); Birge, Marriage and the Law. MSZ, p. 154; cf. Fan, Shangye jiufen, p. 60.

436

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

upon him by the court but insisted that his payment be made to Yang Xueming, with whom the loan was originally contracted, not his brother. The court, however, rejected Li’s argument as immaterial and irrelevant in delivering its verdict.20 The third case, dating to 1675 and involving a commission agreement between a venture capital investor and a pawnbroker, illustrates how the court applied the debt law to uphold the investor’s contractual right. A dispute occurred between the investor, Liu Guangxia, and Zhu Yizhou, who had borrowed 5,300 taels as a capital investment from Liu through middlemen Zhu Xiaoyou and Zhu Jike, both Zhu’s nephews, to open a pawnshop. When the accounts were audited after the first year in the business, it was discovered that Zhu Yizhou had misappropriated part of the loan, totalling 3,880 taels, for his personal use. On the (ill-advised) counsel of a “litigation master” (songshi 訟師; private solicitors whose activities were officially proscribed) named Hong Zhiqin, in order to cover up his misappropriation of business funds Zhu sued Liu for fraud and embezzlement in Jiaxing (Zhejiang) by falsifying the pawnshop’s accounting records. Zhu’s allegation of embezzlement against Liu fell apart in the prefectural court, which found that the financial documents submitted by Zhu as evidence bore neither Liu’s personal signature nor witness signatures and concluded that they had been fabricated. Also, Zhu Xiaoyou and Zhu Jike, the nephews who had served as middlemen for Zhu, testified in court that Zhu indeed had surreptitiously withdrawn funds that Liu had invested in the pawnshop. In the end, because coincidentally the Qing court declared a general amnesty at that time, the court did not impose any punishments on Zhu or Hong for making false accusations. However, the court did order Zhu to make full restitution to Liu.21 The fourth and last example is a case of credit contract in which a magistrate effectively enforced a contract agreement through mediation rather than imposing the punishment on the offender stipulated in the law code. A business entity with the name of Lu Dayou Hao was a supplier in Nanjing city. In 1718 the supplier’s agent, Xi Chenghou, was sent to Chaoxian (Huizhou, Anhui), about 300 kilometers south of Nanjing, to file a lawsuit in 20

21

XC, juan 6, rpt. in Yang Yifan 楊一凡 and Xu Lizhi 徐立志 (eds.), Lidai panli pandu 歷代 判例判牘, vol. 4 (Beijing, Zhongguo shehui kexue chubanshe, 2005), p. 370. See two other similar cases on debt obligations arising from credit sales in XC, juan 2 (Li Jian’s case in Shandong province), in ibid., p. 299; and in XC, juan 8 (Feng Sanwai’s case in Henan province), in ibid., p. 390. SHRJ, juan 5, rpt. in Yang and Xu, Lidai panli, 9, p. 443.

437

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

the magistrate’s court against the textile shop owner Wang Qianchu for defaulting on payment for textiles purchased on credit. Wang was in debt to the supplier for eighty-five taels of silver. Through mediation under the supervision of the magistrate, Wang agreed to honor his payment obligations. To pay off his debt, he sold and mortgaged most of his property and personal effects and borrowed money from his family members to come up with the full amount owed to the supplier. Once informed that after mediation Wang had repaid his debt, the magistrate closed the case.22 It appears that once the debt was paid in full and the accompanying lawsuit dismissed, the court in practice was likely to mitigate a defendant’s punishment. It seems that statutory penalties were instituted more to reduce and deter malfeasance than to compel magistrates to stick to the letter of the law and treat violators as criminals. In brief, three legal principles underpin debt and default disputes from the Song through the high Qing period: (1) capping monthly interest rates at 3 percent, (2) prohibiting compound interest, and (3) restricting the total amount of accrued interest in arrears to an amount equal to the outstanding principal debt. In Ming and Qing times, defaulting on a debt or loan carried a punishment of ten blows with the light baton if the debt or loan amount was 5,000 copper coins in cash or equivalent and if the repayment was three months or more behind schedule. In general, loan defaulters received heavier punishment during the Ming–Qing period than in previous eras, possibly with the intention of preventing litigation and reducing the court caseload.

Fraud in Trade As stipulated in provisions under Song law, which was modeled after the Tang Code, fraudulent behavior such as misappropriation of assets and deceptive financial reporting constituted criminal offenses. For example, one of the major provisions of the Song Penal Code stated that those who intentionally deceive people to obtain their money or movable property in public or private dealings committed fraud and should be punished in accordance with the law of theft. Specific fraud statutes instituted under another article stipulated that punishments equal to the penalties for crimes of theft should be imposed on those committing fraud by forging official or unofficial documents such as contracts, bills, and accounting records or by altering 22

The full text of this case documented in the Wang family records held at the Institute of Economics, Chinese Academy of Social Science (file no. 075) is cited in Zhang Youyi 章有義, Ming Qing ji jindai nongye lunji 明清及近代農業論集 (Beijing, Zhongguo nongye chubanshe, 1997), pp. 434–40.

438

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

such documents in order to reap monetary benefits or to escape confiscation or loan payment obligations.23 Although these laws hardly deterred fraudulent activity in reality, the criminalization of fraud with applicable laws and regulations in place enhanced the government’s ability to sanction individuals who committed fraud or engaged in corruption. Such wrongful practices could have a detrimental effect on an expanding market economy, as postulated in the body of scholarship on trust in business history.24 Under Yuan law, more severe punishments than those that prevailed in the Song dynasty were imposed on those who knowingly misrepresented themselves as holding certain official positions or forged official documents such as tax returns and commodity monopoly certificates.25 However, it is unclear whether such antifraud regulations were in place for private dealings, since Yuan laws survive only in fragments containing insufficient information. Similar fraud laws, rules, and regulations applicable to private-sector business entities and activities were, however, incorporated into the Ming and Qing legal codes. Another type of fraud in trade involved faulty units of measurement, standards, and measuring instruments. Without an established set of reliable measurement standards, market transactions would incur unnecessary costs and cause confusion and chaos. Following the Tang Code, Song dynasty law criminalized fraudulent measurement practices in commercial transactions, and the regulatory framework laid out in some statutory articles included severe punishments for violations of measurement. For instance, the offense of failure to comply with the official standards of measurement for length, weight, and capacity entailed a punishment of seventy blows with the heavy baton. Those who engaged in the production and sale of tableware, utensils, and silk and other fabrics that failed to meet the official standards in terms of sturdiness, length, and so forth were subject to such punishment. A punishment of fifty blows with the light baton was imposed on those who made substandard instruments to measure capacity, weight, and length of products sold on the market. If merchants gained undue profits as a result of their intentional mismeasurement, they would be charged under theft laws.26 Officials would 23

24

25 26

SXT, pp. 448–9. The same provisions can be found in the Tang Code; see TLSY, 2, pp. 1857, 1728. See, e.g., Christopher Kobrak, “The Concept of Reputation in Business History,” Business History Review 87 (2013), 763–86. YDZ, 52:3a–9b. SXT, pp. 482–4. This was the same under the Tang Code; see TLSY, 2, pp. 1728, 1859, 1866. On transaction costs and measurement of goods in maritime trade, see So, Prosperity, Region, and Institutions, p. 231.

439

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

receive a punishment one degree less than that for fraud if they overlooked fraudulent practices during an annual inspection of measurement standardization. Individuals who knew of acts of noncompliance with laws and regulations concerning measurement and failed to inform relevant authorities would face the same punishment as offenders. The Yuan dynasty inherited the practices of its predecessor and promulgated laws that required compliance with official standards in measuring commercial products and prohibited manufacture of nonstandard measuring instruments. One statute made it unlawful to use substandard instruments for measuring weight, volume, and length of commercial goods, while another forbade the use of not only faulty measuring tools but also private brokers.27 The above statutory framework to criminalize fraudulent and deceptive business practices essentially remained the same during the Ming and Qing dynasties, but the operational aspects of fraud law and litigation were further elaborated to meet needs arising from the rapid increase of market transactions and long-distance trade. What stood out during this period was the increasingly sophisticated brokerage system as an extra-bureaucratic arm of commercial regulation and the multifarious duties of licensed brokers in supervising commercial transactions, particularly with regard to curbing fraud in measuring merchandise items. The following legal cases from the Ming–Qing period illustrate how enforcement of measurement fraud laws contributed to regulating commercial activities. In the late 1620s, two lawsuits were brought to the magistrate’s court in Guangzhou against the local silver assayer Mai Xunnan. One was filed by a sojourner merchant, Zheng Xiang, who traded in silver and commissioned Mai to recast silver he received in payments into standard ingots. Zheng claimed that Mai substituted silver of inferior quality. After Xiao Jiyu, another silver shop owner who had a business relationship with Zheng, testified in support of Zheng’s claim, the dispute was referred to the local constable (dibao 地保), Ou Zhenfan, who resolved the dispute through mediation with the help of neighbors. Afterward Zheng returned home. Two months after the dispute with Zheng was settled, Mai was sued again by a local merchant for fraudulent and dishonest conduct of a similar kind. Mai testified in his own defense, maintaining his innocence and blaming the silver shop owner Xiao for the purported misconduct in this particular case. After a trial in the magistrate’s court, Mai was found guilty of falsely accusing Xiao 27

YDZ, 57:50b–51b, 51b–3b.

440

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

and of committing fraud regarding silver deals. The magistrate recommended punishment of a specified number of blows of beating with the heavy baton. Judicial officials at the provincial level, however, ruled that Mai should be punished more severely with an additional twenty blows and public display in the cangue for ten days.28 Another case of fraud, involving complicity in the fraudulent measurement of merchandise and the role of law in preventing such fraud, was recorded in the Baxian Archives in Chongqing. In 1763 Liu Qilong, a tinsmith who purchased some tin from the shop of Liu Yu, filed a lawsuit with the sheriff’s office claiming that Liu Yu used a substandard scale to weigh the tin he sold to Liu Qilong. At the initiative of the sheriff, the dispute was referred to the headman of the Eight Provinces Merchants Association (basheng kezhang 八省客長), an association of sojourner merchants in Chongqing formed with assistance from the local government and tasked with mediating commercial disputes.29 The association reached a consensus that the standard unit for tin transactions should be 18.5 liang (0.694 kilos). Dissatisfied with this decision, Liu Qilong pursued the case further. This time the magistrate’s court handled the litigation and issued a ruling that eighteen liang should be the standard unit for tin. Liu Qilong disagreed with the court’s decision and appealed to the prefectural government for adjudication.30 Unfortunately the final outcome of the dispute is unknown. Nonetheless, the case sheds light on the vital role played by the use of standardized units of measurement for goods traded on the market. It illustrates further that legal and regulatory frameworks were important tools to prevent and discover measurement fraud and facilitate its punishment, and that the judicial and extrajudicial channels functioned in tandem, with their complementary relationship providing an essential component in the dispute resolution process.

Brokerage Regulations The brokerage system was well established under the Song dynasty. Brokers, who were variously designated ya 牙, yaren 牙人, yahang 牙行, or yabao 牙保, supported the commercial expansion of the Song era through their 28 29

30

MSZ, p. 154. The cangue was a wooden collar worn by criminals. On this merchant association, see Zhou Lin 周琳, “Chengshi shangren tuanti yu shangye zhixu: Qingdai Chongqing basheng kezhang tiaochu shangye jiufen huodong wei zhongxin” 城市商人團體與商業秩序: 清代重慶八省客長調處商業糾紛活動 為中心, Nanjing daxue xuebao (Zhexue renwen kexue shehui kexue ban) 南京大学学报 (哲学人文科学社会科学版) 2011.2, 80–99. BXDA, 1, pp. 310–11.

441

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

involvement in regulating market transactions. Generally speaking, the government required officially licensed brokers called guanya 官牙 to supervise all transactions pertaining to immovable property such as land and buildings and certain specified movable commodities such as livestock and human beings. Transactions of most other commodities, such as rice, salt, tea, charcoal, aromatics, medicine, and ceramics, were handled by private brokers (siya 私牙). In essence, brokers acted as middlemen in many business and financial transactions and their duties ranged from providing buyer–seller business matchmaking services to negotiating prices, verifying transaction agreements as guarantors, and assisting with storage of goods.31 Although the terms for brokers mentioned earlier frequently appeared in official and private records of economic and social life in the Song period, the words “broker” or “brokerage” were not used in the titles of any statutory provisions in the Song Penal Code (nor had they appeared in the Tang Code). There was only a passing mention of brokers in some provisions relating to, for instance, the illicit sale of property by a junior member of a household whose head was still alive. The Comprehensive Institutions of the Great Yuan, however, used the word “broker” in the titles of three statutory articles, including one that prohibited engaging private brokers (yaren) in addition to using substandard measuring equipment (which is mentioned in the section on fraud in trade in this chapter). The other two articles, included in the section on “Customs and Markets” (guanshi 關市), are devoted to general rules on brokerage houses (yahang) and broker misconduct (yabao qibi 牙保 欺蔽).32 One thing that can be gleaned from these laws is that some brokers, sheep and lamb brokers in particular, manipulated prices and gained undue profit by deliberately preventing buyers and sellers from meeting in person and citing different prices to buyers and sellers to ensure a wide difference between purchase and sale prices. These cunning brokers were profiteering in the name of collecting a broker’s fee. Such practices caused supply shortages and inflated the prices of goods. As a result, the government took action to criminalize such conduct. Unlike the brokerage law of the Yuan dynasty that allowed brokers to operate only in certain trades, laws in Ming–Qing China permitted licensed brokers to handle a wide array of commodities. The brokers were regulated and controlled through the licensing system. Individuals who acted as unlicensed brokers were liable to the punishment of sixty blows with the 31

32

Yoshinobu Shiba, Commerce and Society in Sung China (Ann Arbor, University of Michigan Center for Chinese Studies, 1970), pp. 165–80. Chapter 18, “Yabao qibi” 牙保欺蔽 and “Yahang” 牙行, TZTG, pp. 233–4, 245–6.

442

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

heavy baton and the forfeiture of all brokerage fees earned. In 1758, a law was enacted dividing offenses relating to brokering activity into the two categories of default and fraud, with heavier punishments inflicted for the latter. While default could occur when the broker failed to pay for goods entrusted to him according to the contract provisions, fraud law applied to brokers who used deceit to obtain a financial or property advantage in business transactions.33 The following two legal cases illustrate how brokerage laws were implemented in Ming China. The first case, which is included in a litigation manual as an illustrative example, shows how complex brokerage regulations could be in cases relating to sojourner merchants. A dispute arose when a sojourner wholesale merchant entered into consignment and subcontractor agreements with a local broker in Haining (Zhejiang), entrusting him with the task of engaging local retailers to sell a shipment of cotton cloth valued at fifty-eight taels. The local broker then made contractual arrangements with several fabric shops in the local area to leave the merchant’s cotton cloth in their possession to sell. In the end, the shops failed to fulfil their contractual obligations to pay the broker, who then defaulted on his obligations to the sojourner merchant. The merchant then sued the broker for breach of contract in the magistrate’s court. In the end, the adjudication decision required the broker to pay his debt to the merchant plaintiff before he could file lawsuits against the defaulting shops in court.34 The case makes the point that the broker’s initial contractual commitments superseded any consignment or subcontractor agreement he subsequently made with another party. Another example is a lawsuit brought to court by the sojourner merchant Ni Yuan at an unknown location in 1627 or 1628. Two unlicensed brokers, Fang Shaowu and Wu Da, lured Ni into believing that the licensed broker Chen Chao had manipulated the price for Ni’s soybeans upward, overcharging the buyer by three taels and pocketing the extra profit. Believing what Fang and Wu told him, Ni sued Chen for price manipulation. During the trial, however, Ni was unable to identify the buyer in question and the court found that the soybean deal in fact was never completed. The court quashed the case due to a lack of evidence and delivered its decision that Ni as well as Fang and Wu should be charged with conspiracy to fabricate the accusation against 33 34

Qiu, Dang falü yushang jingji, pp. 217–25. FLZTZ, 3:8a–9b; Billy K.L. So, “Institutions in Market Economies of Premodern Maritime China,” in So (ed.), The Economy of Lower Yangzi Delta in Late Imperial China: Connecting Money, Markets, and Institutions (London, Routledge, 2012), pp. 222–3.

443

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

Chen. Although there is no mention in the record as to what punishments were inflicted upon the three offenders, the case informs us that a fabricated allegation against an innocent person was a punishable crime.35 Furthermore, considering the small percentage of profit on the alleged transaction in this case, it is apparent that brokerage regulations – including capping broker fees, prohibiting price manipulation, and banning the use of private brokers – were strictly enforced. Government regulations indeed served as the vehicle for resolving brokerage-related disputes.

Conclusion The above accounts of commercial and business litigation cases and applicable laws paint a positive picture of functional (but not necessarily causal) interrelationships between the legal and economic processes in late imperial China. A solid foundation for the legal framework of commercial litigation had already been laid during the Tang, even before the commercial expansion and economic prosperity of the Song era. Simply put, the Tang Code, building upon the previous legal codes, consisted of a set of exacting and complicated laws, which were written in a clear and comprehensible manner to ensure the enforcement of the values to be promoted in society and the punishment of deviant behavior. Commercial law, like other types of law, was based on timehonored principles of justice and morality. It was intended to establish effective and accountable governance mechanisms that would allow entrepreneurs to earn a just profit from business ventures while curbing profit maximization. It was not the intention of China’s imperial government to institute commercial law for the purpose of promoting dynamic markets and economic growth. Nonetheless, the law, enforced by the courts, incidentally played a role in the expansion of market activities and its impact on economic success by providing ways to resolve commercial disputes and ensure the execution of judicial decisions, as we see in cases concerning contracts, debt and default, fraud in trade, and brokerage regulations. While non-litigation approaches such as mediation were often used, particularly in the early stages of commercial dispute resolution, a good number of cases actually ended up in court. That litigation was an accessible resolution method to which disputing parties could resort provided an incentive for the parties to 35

LMBJ, juan 8, in Yang and Xu, Lidai panli, 4, p. 241.

444

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy

comply with their obligations as established by judicial and extrajudicial institutions. Sun Lijuan has argued that customary commercial law played a part in economic activity during Qing times just as lex mercatoria or the “law merchant” governed commerce in medieval Europe before the emergence of Western commercial law.36 Given our current understanding of how disputes were resolved through the use of a variety of available judicial and extrajudicial means in late imperial China, it is fair to say that multiple conflict resolution arrangements existed that tended to be more fluid, diverse, and complex than they might appear. They were dynamic rather than identifiably separate entities, which may be better characterized as a continuum as perceived in people’s lived experience. These dispute resolution arrangements coexisted, mutually supplemented one another, and fostered organic relationships in which they each carried out their own functions sometimes independently and sometimes interdependently to serve the Chinese legal ideal of just and efficient conflict resolution. After all, there were long-established general principles to be upheld consistently and coherently in any dispute resolution procedure, through mediation or litigation, under the doctrine of heavenly reason as prescribed in Confucian texts. To summarize the main points and implications, the rise of China’s market economy from the year 1000 to 1800 certainly benefited from the consistent and pertinacious implementation of the legal framework interwoven with and supplemented by community-based extrajudicial dispute resolution mechanisms. The legal framework provided regulations for the conduct of market actors that positively influenced market structure and behavior. The framework worked well enough to maintain social order as it gave the impression of delivering justice. As economic activity expanded in all directions and became more complex (e.g., an increase of transactions on credit, a rise in business partnerships with non-family members, the expansion of long-distance trade, and the spread of moneylending institutions), the number of commercial disputes grew and so did the number of disputing parties choosing 36

Sun Lijuan 孫麗娟, Qingdai shangye shehui de guize yu zhixu 清代商業社會的規則與 秩序 (Beijing, Zhongguo shehui kexue chubanshe, 2005). Zhang Taisu has similarly observed that customary precedents prevailed over the divergent legal rules stipulated in Qing law in property disputes. See Zhang Taisu, The Laws and Economics of Confucianism: Kinship and Property in Preindustrial China and England (Cambridge, Cambridge University Press, 2017), pp. 35–63.

445

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

billy k.l. so and sufumi so

litigation. With time and growth, new layers of complexity were added to disputes. For instance, the increasingly common practice of investors forming partnerships was accompanied by the rise in conflicts between business partners. As a result, an incremental adjustment of the judicial system was necessary in response to changing conditions. In short, while it was not designed to promote market economy in late imperial China, imperial law on commerce, alongside broader social and economic factors, undoubtedly facilitated the remarkable market-driven economic growth over sustained periods of time. Moreover, we should be cognizant of the impact of the internal dynamics inherited from the interplay between law and market economy down to 1800 on the drive to align existing legal systems with China’s market-based economic aspirations and dynamism in the late nineteenth and the twentieth centuries.

Further Reading Ch’ü, T’ung-tsu, Law and Society in Traditional China (Paris: Mouton, 1961). Dai Jianguo 戴建國 and Guo Dongxu 郭東旭, Nan Song fazhi shi 南宋法制史 (Beijing, Renmin chubanshe, 2011). Fan Jinmin 范金民, Ming Qing shangye jiufen yu shangye susong 明清商業糾紛與商業訴 訟 (Nanjing, Nanjing daxue chubanshe, 2007). Gao Nan 高楠, Songdai minjian caichan jiufen yu susong wenti chutan 宋代民間財產糾紛 與訴訟問題初探 (Kunming, Yunnan daxue chubanshe, 2009). Gardella, Robert, “Contracting Business Partnerships in Late Qing and Republican China: Paradigms and Patterns,” in Madeleine Zelin, Jonathan K. Ocko, and Robert Gardella (eds.), Contract and Property in Early Modern China (Stanford, Stanford University Press, 2004), pp. 327–47. Guo Dongxu 郭東旭 et al., Songdai minjian falü shenghuo yanjiu 宋代民間法律生活研究 (Beijing, Renmin chubanshe, 2012). Hansen, Valerie, Negotiating Daily Life in Traditional China: How Ordinary People Used Contracts (New Haven, Yale University Press, 1995). Hu Xingdong 胡興東, Yuandai minshi falü zhidu yanjiu 元代民事法律制度研究 (Beijing, Zhongguo shehuikexue chubanshe, 2007). Huang, Philip C.C., Civil Justice in China: Representation and Practice in the Qing (Stanford, Stanford University Press, 1996). Ji Huaiyin 季懷銀, Zhongguo chuantong minshangfa xingshuai zhijian 中國傳統民商法興 衰之鑒 (Beijing, Zhongguo minzhu fazhi chubanshe, 2003). Jing, Junjian, “Legislation Related to the Civil Economy in the Qing Dynasty,” in Kathryn Bernhardt and Philip C.C. Huang (eds.), Civil Law in Qing and Republican China (Stanford, Stanford University Press, 1994), pp. 42–84. Liang, Linxia, Delivering Justice in Qing China: Civil Trials in the Magistrate’s Court (Oxford, Oxford University Press, 2007).

446

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

Law and the Market Economy Lin, Lin, “The Evolution of Partnerships in China from the Perspective of Asset Partitioning,” Stanford Journal of Law, Business, & Finance 18.2 (2013), 215–49. McKnight, Brian E., Law and Order in Sung China (Cambridge, Cambridge University Press, 1992). Qiu Pengsheng 邱澎生, Dang falü yushang jingji: Ming Qing Zhongguo di shangyefalü 當法律 遇上經濟: 明清中國的商業法律 (Taipei, Wunan chuban gongsi, 2008). Shiga Shu¯zo¯ 滋賀秀三, Shindai Chu¯goku no ho¯ to saiban 淸代中国の法と裁判 (Tokyo, So¯bunsha, 1984). So, Billy K.L., “Institutions in Market Economies of Premodern Maritime China,” in So (ed.), The Economy of Lower Yangzi Delta in Late Imperial China: Connecting Money, Markets, and Institutions (London, Routledge, 2012), pp. 208–32. So, Billy K.L., Prosperity, Region, and Institutions in Maritime China: The South Fukien Pattern, 946–1368 (Cambridge, MA, Harvard University Asia Center, 2000). Sun Lijuan 孫麗娟, Qingdai shangye shehui de guize yu zhixu 清代商業社會的規則與秩序 (Beijing, Zhongguo shehui kexue chubanshe, 2005). Umehara Kaoru 梅原郁, So¯dai shiho¯ seido kenkyu¯ 宋代司法制度研究 (Tokyo, So¯bunsha, 2006). Wang Xiaolong 王曉龍 and Guo Dongxu 郭東旭, Songdai falü wenming yanjiu 宋代法律 文明研究 (Beijing, Renmin chubanshe, 2016). Wu Peilin 吳佩林, Qingdai xianyu minshi jiufen yu falü zhixu kaocha 清代縣域民事糾紛與 法律秩序考察 (Beijing, Zhonghua shueju, 2013). Zelin, Madeleine, The Merchants of Zigong: Industrial Entrepreneurship in Early Modern China (New York, Columbia University Press, 2005). Zelin, Madeleine, “The Firm in Early Modern China,” Journal of Economic Behavior & Organization 71 (2009), 623–37. Zhang, Taisu, The Laws and Economics of Confucianism: Kinship and Property in Preindustrial China and England (Cambridge, Cambridge University Press, 2017).

447

https://doi.org/10.1017/9781108587334.013 Published online by Cambridge University Press

12

Property Rights and Factor Markets mio kishimoto

Since the late nineteenth century, some scholars have emphasized the free character of the traditional Chinese economy, while others have regarded it as a feudal or Asiatic one that prevented the development of a market economy. The question of property rights and factor markets, which are the themes of this chapter, is closely related to this subject. The first section of this chapter will give a brief survey of the institutions governing the markets for land, labor, and capital from the perspective of law and policy. In the second section, several concrete illustrations will be presented to describe the functioning of factor markets in the Song–Qing economy. In the last section, an outline of the short-, middle-, and long-term changes that occurred in factor markets will be described. As will be discussed in detail in the first section, after the equal-field (juntian 均田) system collapsed in the mid-Tang period, land could be sold and purchased almost freely, with little governmental interference. People could directly manage or rent their lands at will and sons inherited their father’s land according to the rule of equal distribution. Nevertheless, this does not necessarily mean that the free character of Song–Qing factor markets was based on the notion of “property rights” in the Western sense. An overhasty attempt to find “modern ownership” in imperial China might result in overlooking the basic principles by which Chinese people reasoned the legitimacy of ownership. If we can safely say that the logical basis of modern ownership was the axiom of self-ownership – that is, John Locke’s claim that “every man has a property in his own person”1 – then the logical origin of 1

John Locke, Two Treatises of Civil Government (London, Everyman’s Library, 1970), Second Treatise, Section 27.

448

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

Chinese ownership was the idea of “king’s land, king’s people” (wangtu wangmin 王土王民), namely that all lands and all people are possessed by the sovereign, which appeared as early as the Book of Odes, dating to the sixth century B C E or earlier. It might be said that the notion of the ruler’s possession of land and people had already become an empty theory by the Song dynasty, when lands came to be traded freely, but this principle remained the logical starting point of ownership throughout the imperial period. The conventional thinking of Confucian intellectuals is displayed in an essay written in the mid-seventeenth century that proposed the following: an ancient sage-king allotted a portion of his land to every household (the so-called jingtian or “well-field” system; see von Glahn’s and Dunstan’s chapters in this volume) so that everyone could enjoy a peaceful life, but after the well-field system was abandoned during the Warring States period, people no longer had a constant means of livelihood; consequently, while the rich became as powerful as kings and nobles, the poor were forced to sell not only their land but also their own selves and their children.2 According to this thesis, free trade in land and people was a phenomenon that deviated from the ideal norm. However, as free trade had already become an established practice, most intellectuals, including government officials, thought that forcible intervention to abolish this practice would be unrealistic. Free trade in land and people was not encouraged, but it was tolerated as long as it did not bring about social disorder. The logic of this relationship between the ruler and ownership provides an interesting contrast to that of modern Europe. According to the logic of Lockean liberalism, private ownership preceded the formation of the state. Owners established the state in order to protect their ownership. In contrast, in traditional Chinese discourse, sovereign ownership was the original condition. Private ownership was generated through the allocation of lands by the sovereign, and gradually became an accepted reality. Sovereign ownership remained the latent, logical foundation of free trade and private ownership. It might be said that what people sold and bought was not the land itself but the rights to use it, which were originally granted by the sovereign.

2

YYXQ, p. 575.

449

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

Institutions Land Markets As agriculture was the chief economic activity in China before the midtwentieth century, land was the most important factor of production during more than 2,000 years of China’s imperial regime. Under imperial rule, land sales generally were not prohibited, except during the period from the midfifth century to the mid-eighth, when the government adopted interventionist policies, the most famous of which was the equal-field system. Adhering to the basic land policies of preceding dynasties, the early Tang government enforced the equal-field system, under which land would be distributed to each household according to the number of male members. Upon the death of the property holder, his allotment had to be returned to the government for reallocation. Selling and buying land were prohibited. However, after the mid-Tang period, the equal-field system broke down, and the government no longer intervened in land sales. A new tax system called the “twice-a-year tax” (liangshui 兩稅) was inaugurated that levied taxes on household wealth, which in most cases was based on landholdings. Nonintervention did not necessarily imply positive protection of property rights, but the Tang government was forced to adjust laws on landownership to cope with increasing conflicts and lawsuits over land sales. Instead of formally abolishing the laws prohibiting land sales, the government issued a new law that placed time limits on filing lawsuits, stipulating that the courts should not hear cases concerning land sales that had occurred twenty or more years earlier.3 This law indirectly protected buyers by limiting the rights of sellers (or often their heirs) to recover their former landholdings. As early Tang land contracts from Dunhuang and Turfan show, land sales did not cease even under the legal prohibition. After the demise of the equalfield system, however, lands came to be traded more openly and vigorously, and a new landed class emerged during the Five Dynasties and Song periods. Unlike the great landowners in the Tang, who held large manorial estates gained through their political privileges, this new seigneurial class usually accumulated their lands through piecemeal purchases of dispersed plots. Although there was no fixed pattern in the management of these lands, tenant farming seems to have been more common than direct management, especially in south China, where intensive rice farming was widespread. It 3

Mio Kishimoto, “Property Rights, Land, and Law in Imperial China”, in Debin Ma and J. L. van Zanden (eds.), Law and Long-Term Economic Change (Stanford, Stanford University Press, 2011), pp. 74–5.

450

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

would be safe to say that frequent land sales, the accumulation of land through market exchange, and tenant farming were the main features that characterized the Song–Qing agricultural system. The general procedure for land transactions was as follows. People who wished to sell or buy a piece of land would look for buyers or sellers through their personal network or land brokers. Although there were some customary restrictions on the free will of sellers and buyers, such as the lineage members’ right of pre-emption, there was usually little official or communal control of contract terms. When the buyer and seller agreed on the terms of a contract, the seller would meet with the buyer, a mediator, and witnesses, where the seller – along with the mediator and witnesses – signed the contract and delivered it to the buyer. Signatories in principle were males, except in cases such as a household consisting of a widow and young children. Contracts normally included phrases such as “this sale is based on the seller’s free will,” although distressed sales were not uncommon. There were no notary publics who officially authenticated the contracts; rather, relatives and friends of the seller and buyer, acting as mediators and witnesses, confirmed its validity. After the land sale, the buyer had to pay a contract fee at the local government office and obtain an official vermilion seal affixed to the contract (thus such sealed contracts were known as “red contracts”), but the official seal was not an essential condition for the contract’s validity. Land sales took two forms: irrevocable sale (juemai 絕 賣) and conditional sale (dianmai 典賣). The latter allowed the seller to redeem the land at the original price after a certain period, while the former did not. The lack of a clear distinction between the two often was the cause of land disputes, as is discussed in detail below. The Luminous Collection of Judgments by Illustrious Figures (Minggong shupan qingming ji 名公書判清明集), a casebook of actual judicial decisions by famous officials in the Southern Song period, includes many rulings on land disputes. It is obvious that these officials made decisions according to the principle that the buyers’ ownership rights should be protected as long as there was clear evidence of a sale, such as written contracts or the oral testimonies of witnesses. However, the Song laws they cited rarely included positive statements on proprietary rights. Instead, the laws most often invoked concerned the time limits on filing lawsuits.4

4

Atsushi Aoki, So¯dai minjiho¯ no sekai 宋代民事法の世界 (Tokyo, Keio¯ gijuku daigaku shuppankai, 2014), Chapter 5.

451

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

The Ming Code (and the Qing Code, which fundamentally followed the Ming Code) included a few statutes on land sales, namely “illegal sales of real estate” (daomai tianzhai 盜賣田宅) and “the conditional sale and purchase of real estate” (dianmai tianzhai 典賣田宅). The former statute stipulated punishments for those who sold or disposed of real estate not belonging to them, while the latter prescribed punishment for crimes concerning conditional sales, such as the evasion of contract fees, double selling, and so on. Although numerous sub-statutes on land sales were enacted during the Ming–Qing period, a consistent system of civil law was not established until the Republican period. Throughout the Song to Qing period, laws relating to landownership were no more than an accumulation of ad hoc penal regulations to cope with conspicuous cases of wrongdoing. Despite the prevalence of free trade in land, there existed in Song–Qing China various customs such as lineage members’ rights of pre-emption, conditional sale, and topsoil (tianmian 田面) ownership that apparently contradicted the inviolability of property rights. It should be noted that these customs were not vestiges of the ancient nonmarket regime, but rather measures developed to cope with the risks of the unpredictable land market. The right of pre-emption was intended to prevent the loss of land to others outside the kin group. A conditional sale, as noted above, included the stipulation that the seller could repurchase the land at the original price after a certain number of years. Although the price in a conditional sale was lower than that of an irrevocable sale, most sellers preferred a conditional sale to avoid the risk of permanently losing the land. Topsoil ownership was a custom seen widely in central and south China. A topsoil owner typically was a tenant cultivator who paid rent to the subsoil owner, but unlike ordinary tenants, he had security of tenure and could freely sell or rent his cultivation right to other farmers without the permission of the subsoil owner. This situation was called “two owners to a field” (yitian liangzhu 一田兩主). The price of topsoil rights was sometimes as high as that of subsoil ownership, but farmers were willing to pay in order to secure security of tenure and a subsistence foundation for their families. Government policies on these customs were inconsistent throughout the Song–Qing period. As the government’s main concerns were tax collection and dispute prevention, these customs sometimes were banned because of their detrimental effects on revenues and public order, but at other times they were tolerated as harmless. During China’s imperial era, private ownership was not necessarily supported positively by the government, but rather tolerated passively as long as 452

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

it did not interfere with dynastic authority or social order. In the same way, various customs were also tolerated as long as they did not disrupt the public peace. There were active factor markets, but the strict notion of “property rights” and systematic civil (or commercial) codes to protect them did not exist. This situation provided people’s economic activities with free but rather risky environments.

Labor Markets In the contemporary economy, the term “labor market” usually refers to supply-and-demand relations between capitalist employers who seek wageworkers and free workers who seek employment. If we regard this term as a concept related only to free wageworkers, the size of the “labor market” must have been very small in late imperial China. In this chapter, the term “labor market” will be interpreted in a broader sense, including the availability of the labor of tenants and bondservants. From this point of view, it is necessary to review the status system of this period. The chief characteristic of work in this era was the variety of relations of production. Land, capital, and labor could be combined in various ways through agreements between the concerned parties, while the government, as a general rule, would not (or could not) intervene in such private agreements. From the viewpoint of sharing profits and risks, large landowners mainly had the following choices: (1) leasing the land to tenants at a fixed rent, (2) sharing a fixed proportion of the produce of the land with tenant cultivators, (3) employing laborers at a fixed wage, or (4) exploiting domestic servants’ labor. Each choice included more minute options that varied according to the degrees of the landowner’s managerial engagement and the laborers’ personal subordination to landowners. Some landowners exerted lord-like influence over their tenants, while others did not. Some long-term, live-in workers were not so different from house servants, while short-term hired laborers lived on their own. As a result, slavery, serfdom, sharecropping, contract farming, and proto-capitalist managerial farming coexisted during this period. It is difficult to estimate the proportion among these various types of labor arrangements, but it might be safely said that large-scale farms such as latifundia and modern plantations, which employed a large number of laborers, were rare. Most agricultural production was in the hands of smallscale farming households, the majority of whom were tenants, especially in the central and southern parts of China, but the actual relationship between the landowners and tenants varied from case to case. 453

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

Although the imperial government adopted a laissez-faire policy concerning the relations of production formed among the people, it enacted rules to define the legal status of bound labor. Since the Tang dynasty, people were legally divided into the two major categories of “honorable people” (liangmin 良民) and “mean people” (jianmin 賤民). People who engaged in certain pursuits regarded as dishonorable, such as bondservants and entertainers, were categorized as “mean people” and were subject to discriminatory treatment; for example, they were forbidden from taking the civil service examinations or marrying “honorable” people. The most common group within the “mean” category were “slaves.” The term “slave” (nubi 奴婢) generally meant hereditary servants who could be bought and sold by their masters, but as a legal term its meaning changed over time. Because the Song and Ming states prohibited ordinary families (i.e., families other than high-ranking officials) from possessing slaves, the legal term “slave” was applied only to hereditary servants belonging to privileged families. Actually, ordinary families also possessed hereditary servants who could be sold, but these servants were legally defined not as “slaves” but as “indentured servants” (guyongren 雇傭人) or “adopted sons” (yinan 義男), in order to keep formal consistency with laws. After the mid-Qing period, the legal definition of “slaves” was broadened to include hereditary servants held by ordinary families. Although the number of these legally discriminated people was small compared with the total population, the government regarded it as essential from the viewpoint of maintaining social order to classify people properly according to their actual degree of subordination. Unlike slaves, indentured servants and adopted children were no longer regarded as “mean people,” but legally they were treated not as independent freemen, but as the unequal dependents of their masters.5 It should be noted that the distinction between subordinate labor and free labor was not as clear in late imperial China as people today might expect. The Chinese population enjoyed the “freedom” to make various kinds of contracts (including selling their children and themselves), but it did not necessarily mean that they were always “free” laborers. There was a spectrum of subordination, ranging from slave to freeman.

5

Yoshiro¯ Takahashi, So¯–Shin mibunho¯ no kenkyu¯ 宋―清身分法の研究 (Sapporo, Hokkaido¯ daigaku tosho kanko¯kai, 2001).

454

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

Capital Markets Needless to say, there was no institution like modern equity markets in late imperial China, but people did not run their businesses solely relying on their own capital resources. There were various ways to raise funds for running a business: for example, borrowing and jointstock partnerships. It was a common practice for peasants to resort to short-term borrowing from moneylenders or pawnshops. There were various types of loans. When farmers borrowed funds for planting from landowners or moneylenders on the condition of repayment after harvest, the interest rates were invariably high. Selling farm produce at a discount for future delivery was a widespread practice. A famous phrase from a Tang poem, “selling new silk yarn in the second month, and selling new grain in the fifth month” (eryue mai xinsi, wuyue tiao xingu 二月賣新絲, 五月糶新 榖), described the common situation in which farm families were forced to contract future sales of their produce several months in advance in order to obtain funds to tide them over during the lean spring and summer months.6 The government took some measures to ease the predicaments of family farms. One of the New Policies enacted by the chief councilor Wang Anshi in the late eleventh century was the “Green Sprouts” rural credit program (see the chapter by Lamouroux and von Glahn in this volume). Under this program, the government provided farming families with low-cost loans to spare them the usurious terms demanded by private lenders. However, the Green Sprouts program was highly controversial, and it was abolished after Wang Anshi’s death. Imperial law consistently fixed limits on interest rates for private loans. Although the upper limit varied, it was usually within the range of 3 to 6 percent per month. It was also the general rule that the total amount of interest could not exceed the sum of the original loan. For smallscale farming families, there was no clear distinction between subsistence loans and borrowing for investment purposes. Preparing for the next year’s work was no less crucial for the household economy than surviving an immediate subsistence crisis. The purpose of their investment was not necessarily making profit, but in most cases aimed at maintaining their subsistence. The joint-stock partnership (hehuo 合夥) was mainly employed in commerce and manufacture run by richer merchants and industrialists. There is 6

Nie Yizhong 聶夷中, “Shang tianjia shi” 傷田家詩, quoted in JWDS, 126, p. 1658.

455

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

mention of joint-stock partnerships adopted by merchants engaged in maritime trade as early as the Song period. The simplest type of joint-stock partnership consisted of several partners who each contributed equal sums of money, worked together, and shared the profit equally. During Ming– Qing times, more complex types of joint-stock partnership developed in which partners had different roles: for example, the commenda type in which one partner invested the capital and the other undertook the burden of traveling to distant markets and buying and selling goods, and the agency type in which a group of investors hired professional managers. Although their contributions were not of the same kind, each partner was assigned a certain number of shares in proportion to their equity investment or work responsibilities. Shares in partnerships could be sold or transferred through inheritance. Joint-stock partnerships in China were usually unlimited partnerships organized among kinfolk or friends, and consequently remained smallscale. Although Chinese partnership contracts often included a phrase emphasizing permanent solidarity, such as “blood pledge” (shaxue dingmeng 歃血訂盟), these joint-stock partnerships generally were short-lived and smaller in scale compared with those in early modern Europe, perhaps because active but risky capital and commodity markets encouraged investors to seek short-term profits rather than long-term returns. 　 The above survey of the institutions governing the markets for land, labor, and capital shows that factor markets were generally free in nature. Nevertheless, there existed some features that contradicted our images of a “modern” market – for example, the notion of sovereign ownership and the commodification of people. This situation has long perplexed historians who have tried to interpret Chinese economic history according to the European model of development. In Japan, historians engaged in heated debate during the 1950s–1960s on the problem whether the Song–Qing Chinese society was “feudal” or “early modern.” Just at that time, there arose a controversy on the “sprouts of capitalism” (zibenzhuyi mengya 資本 主義萌芽) among historians in the People’s Republic of China. Although definitions of the “sprouts of capitalism” varied, most of these scholars endeavored to discover evidence of a “capitalist” mode of production employing free wage laborers in the Ming–Qing economy. Surely, such examples existed, but their significance should be understood within the concrete situation of those cases. In the next section, we shall see, through more concrete illustrations, how factor markets worked and asserted influence on economic behavior. 456

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

The Functioning of Factor Markets in the Chinese Economy As noted above, the late imperial economy was characterized by the coexistence of various types of relations of production. In this section, I will provide several concrete examples of these relations to draw a clearer picture of how factor markets worked. However, this does not mean that these examples can be generalized as common situations throughout China across this entire time period. Nevertheless, these examples will shed light on some important features of Chinese factor markets.

Rural Households Let us draw a picture of a rural household based on previous scholarship on the agrarian economy in Jiangnan (the Yangzi delta) in the Ming–Qing period, which has been the main focus of many studies of Chinese economic history. In Jiangnan, the predominant form of land tenure was tenant farming with fixed rent, generally in kind, with the rent payment amounting to almost a half of the average yield of the main product (usually rice). In addition to rice farming, cultivation of secondary crops (wheat, colza, beans, vegetables, etc.) and cottage industries (silk reeling, cotton weaving, etc.) were carried out, the income from which usually entirely redounded to the cultivator. When either the landlord or tenant changed, a new contract (written or oral) had to be made, but the conditions of tenancy seem to have been stable. Jiangnan agriculture was the most intensive type of farming in premodern China. It is said that a typical farmer could cultivate ten mu (0.61 hectares) of wet-rice fields. If a farm family of five members cultivated ten mu of rice paddy, the total yield would be around twenty shi (one shi = 103.5 liters) on average. After paying half of the yield in rent, ten shi of rice would remain. As the average amount of rice consumption for an adult male was said to be four shi, it must have been difficult for peasants to support their family only through rice cultivation. Moreover, many farm families cultivated fewer than ten mu. A quantitative study based on the land registers from seventeenth-century Suzhou showed that almost 60 percent of rural households cultivated 2.5 mu or less.7 As a result, most peasant households had to depend on subsidiary sources of income, such as handicrafts, to survive. These cottage industries were 7

Keiji Adachi, Min Shin Chu¯goku no keizai ko¯zo¯ 明清中国の経済構造 (Tokyo, Kyu¯ko shoin, 2012), pp. 295–6.

457

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

carried out by mobilizing all possible labor among family members, especially that of women. The most famous cottage industry products of Ming– Qing Jiangnan were cotton cloth and raw silk, which were marketed not only empire-wide but also abroad. Many observers, however, described these industries in a pessimistic tone. A late Ming magistrate, Jin Yipai, wrote an essay on the sericulture economy in northwest Zhejiang in which he attributed the predicament of rural households to the shortage of capital: Although silk is the more profitable of the two [rice and silk], it requires equipment. All the cost of equipment for sericulture must be borrowed [from moneylenders] on mortgage. As the months go by, bitter work dims the eyes and turns the hair white; finally, the silk is reeled and people feel happy with surviving the hardships of this year. [Nevertheless] public and private debts have to be settled at the silk market . . . Before the payment of taxes to the officials, the moneylenders, whose expenses have been light but whose income has been great, confront the peasants: they squeeze out the last drop of blood, and the farmers are left, as before, with empty pockets.8

These rural sericulturalists were not wage laborers but independent proprietors who had to obtain the necessary equipment for sericulture themselves. Under favorable economic conditions, they might gain large profits, while in an unfavorable situation, they would suffer from tight capital markets. Although Jin Yipai emphasized that household handicrafts did not improve the living standards of rural households, their situation was not unchangeable. Rather, they struggled amid downward and upward mobility. We can find examples of upward mobility in family genealogies compiled by wealthy families almost all over China during the Song–Qing period, especially after the late Ming. Records on the early generations who immigrated into an area are usually unclear or fictional, which suggests that they were obscure people with insignificant assets and reputation. Through several generations’ endeavor, the family could accumulate sufficient fortune to publish a genealogy. In this sense, although the records they include are not always reliable, family genealogies are clear evidence of upward mobility. In most cases, however, the downward pressure seems to have been heavier. Even if a household owned some land, upon the death of the household head it had to be divided among his sons according to China’s

8

SMXZ, 3:115b–116b. Translation modified from Masatoshi Tanaka, “Rural Handicraft in Jiangnan in the Sixteenth and Seventeenth Centuries,” in Linda Grove and Christian Daniels (eds.), State and Society in China: Japanese Perspectives on Ming–Qing Social and Economic History (Tokyo, University of Tokyo Press, 1984), p. 82.

458

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

common rule of equal inheritance. The division of land made a new household vulnerable to market fluctuations and natural disasters. Poor landowners often relinquished their lands first through conditional sales and finally through changing the conditional sale to an outright sale by receiving additional payments. Landless persons might become tenant farmers, but if the lands they leased were insufficient to support their families, some family members had to seek work as agricultural laborers, bondservants, street vendors, and so on. Some fortunate people might succeed in making some money through these endeavors, but in general it was difficult for manual laborers to save enough money to afford wives and children. That was the end point of the downward mobility. The lineage mutual-aid system might have worked as a safety net for impoverished members, but not all of the poor could share in the benefits of lineage membership. Topsoil ownership was an important method by which tenants secured their rights to the use of land. The custom of topsoil ownership prevailed widely in central and south China, for example in Jiangsu, Jiangxi, and Fujian provinces. Scholars have not yet reached a consensus concerning the origin of this practice. Some scholars trace the germination of topsoil ownership to Northern Song documents concerning state-owned lands, which indicate that tenants who improved land at their own expense were permitted to sell these leaseholds in order to recover their investments. Others argue that the practice of topsoil ownership did not develop until tenants’ cultivation rights became securely established in the late Ming period. A folktale recorded in a late Yuan anthology has attracted historians’ attention in this regard. In this episode, a tenant in Yangzhou tried to mortgage his leasehold to another family without first obtaining the landowner’s approval. But a neighbor bribed the subsoil owner’s son to obtain the tenancy for himself. The subsoil owner seized the land and transferred it to the neighbor, apparently leaving the original tenant with no recourse to recover rights to the land.9 This episode suggests that the practice of topsoil ownership had become widespread, but security of tenure was not yet fully established in late Yuan Yangzhou. The establishment of this practice can be illustrated in two ways. The first way was land reclamation for cultivation. Topsoil ownership was often seen in newly settled areas where tenant farmers invested their own resources to reclaim lands, paying only nominal rents to the landowners. These tenants enjoyed topsoil ownership rights that 9

CGL, 13, pp. 162–3.

459

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

reflected the value of their investment, and could freely dispose of their cultivation rights. The second way derived from rent deposits. In areas like Jiangnan, where there was bitter competition among tenants for lands, landlords would require their tenants to pay substantial deposits as a protection against default. When a farmer relinquished his tenancy, the landlord was supposed to refund the deposit (after subtraction of the amount of any unpaid rent) and seek a new tenant who would pay the same (or a larger) deposit. However, the landlord’s intervention in this process was gradually omitted, because absentee landlords became indifferent to changes of tenant as long as the rent was paid on time. Instead, transferring tenancies directly to other tenants in exchange for the value of the original deposit without the landlord’s consent became an established practice, and the tenants came to be regarded as the topsoil owners. Thus two kinds of ownership were formed – subsoil ownership and topsoil ownership – on the same land, and consequently the two types of land market came to work separately. In Song–Qing China, when lands were leased through mutual agreement, landlords and tenants were of equal legal status unless unequal relationships were formed through agreement. It was natural, however, that patron–client relationships were formed between powerful landlords and poor tenants in certain circumstances. Some evidence shows that the subordination of tenants to landlords was severe in newly developed inland areas such as Hunan and Sichuan during the Song–Yuan period. These tenants were attached to the land; they could not move without the landlord’s permission; and when selling land the landlord transferred the tenants as well. Even in the Qing period, “servile tenants” (dianpu 佃僕), who were customarily distinguished from free tenants, were found in some localities such as Huizhou in southern Anhui. Although they earned their own living cultivating leaseholds, they could not sever the relationship with their landlord at will, and they were obliged to perform various duties for the landlord’s family – for example, services during funerals and weddings, repairing and constructing houses, and so on.10 The Yongzheng Emperor issued an edict in 1727 to emancipate servile tenants from hereditary servitude, but this practice continued until the late Qing period.

10

H. Zurndorfer, “Contracts, Property, and Litigation: Intermediation and Adjudication in the Huizhou Region (Anhui) in Sixteenth-Century China,” in Ma and Van Zanden, Law and Economic Change, pp. 102–4.

460

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

Asset Management by Landlords Just as farming households were maintained by their earnings from various types of employment, such as farming, handicrafts, and commerce, wealthy families also invested their money in diverse ways. An early Qing scholarofficial, Zhang Ying, once quoted his friend Lu Yulin’s words on the most advisable investment: I’ve had much experience with practical matters. Pawnbroking, commerce, and currency speculation all force one to engage in unethical practices. Though one might quickly reap huge profits, in the end your return will melt away to nothing. Only arable land and commercial real estate [i.e., buildings] yield a reliable income over the long term. But comparing the two, real estate is still inferior to cultivated land.

The reason why Lu preferred investment in arable land was that tenant farmers were more honest than commercial tenants, whom he depicted as quarrelsome city dwellers who resisted paying their rent, compelling the landlord to go to the magistrate’s court. Tenant farmers, in contrast, were simple and humble folk who “will surely pay the rent due to the landlord before they settle their private debts to others.” Zhang Ying expressed his approval of these words based on his own experience.11 According to Zhang’s observation, it was a noticeable trend of the time that the scions of wealthy families often sold their lands and engaged in commerce. He admitted that the profit from land was only about 30–40 percent of that from commerce. Certainly, it was a common notion at the time that the interest return on land (i.e., the proportion of annual rent income to the price of land) was less than that of commerce or moneylending. Scattered data show that the annual interest return on land was around 5–20 percent.12 Compared with the maximum legal interest rate on loans (3–6 percent per month), the return on land was low. Nevertheless, Zhang insisted on the advantages of landownership, including the relative safety of real property in contrast to the risks of commerce. Moreover, the degree of safety increased for landowners who lived in the countryside, not only because rural life was less expensive than dwelling in cities, but also because the landowner could keep close watch on their tenants. Zhang severely criticized absentee 11

12

Cited in Hilary J. Beattie, Land and Lineage in China: A Study of T’ung-Ch’eng County, Anhwei, in the Ming and Ch’ing Dynasties (Cambridge, Cambridge University Press, 1979), pp. 141–2. Mio Kishimoto, Shindai Chu¯goku no bukka to keizai hendo¯ 清代中国の物価と経済変 動 (Tokyo, Kenbun shuppan, 1997), pp. 50–1, 419–20.

461

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

landlords who were ignorant of the borders of their properties, whether their tenants were diligent or indolent, the condition of nonarable resources such as water ponds and forests, or the current price of rice.13 The fact that the rate of return on land was lower than that of commerce and moneylending in itself shows that most people at the time agreed with Zhang, as the low rate on land indicates that people were willing to pay high prices for land because of the relative safety of the investment. While absentee landlordism was widespread, especially in central and south China, there also existed many examples of managerial landlords. Master Shen’s Treatise on Agriculture (Shenshi nongshu 沈氏農書), which was written during the last years of the Ming period, illustrates the situation of a managerial farm based on the author’s own experience in Huzhou, in the heart of Jiangnan’s sericulture region. The author, whose real name is unknown, described the management of his farm, paying special attention to the balance between income and expenditure. According to an estimate by Adachi Keiji, the farm consisted of thirty mu of wet-rice field and ten mu of mulberry fields. The main products were rice, silk cloth, livestock (pigs, sheep, fowls, etc.), and liquor. Some monthly or daily workers were hired in addition to a few long-term laborers. The sericulture and silk-weaving tasks seem to have been the work of female family members and female servants.14 The author of this book was eager to increase not only land productivity but also labor productivity. As a means of saving the labor of carrying manure from distant places, the author proposed raising livestock on the farm. He also tried to enhance the laborers’ motivation by improving the quality and quantity of meals. According to the author, the wages of laborers were increasing at that time. The labor market in late Ming Huzhou seems to have become a sellers’ market. Compared with Jiangnan, managerial farms were more widely seen in north China because dry-field agriculture was more suitable for large-scale farming than wet-rice cultivation. There are few detailed descriptions that illustrate the situations of the managerial farms in north China before 1800. According to the comprehensive study by Jing Su and Luo Lun on managerial farms in Shandong around 1900, 80 percent (105 cases out of 131) of the managerial landlords farmed fewer than 500 mu using wage labor, while 7 percent farmed over 1,000 mu. Wealthy managerial landlords accumulated

13

Beattie, Land and Lineage, p. 149.

14

Adachi, Keizai ko¯zo¯, Chapter 3.

462

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

their lands piece by piece over several decades.15 The notable characteristics of managerial farms in north China were their relatively large size and use of plow oxen. According to some agricultural handbooks written by literati landlords from north China during the Qing period, the main products were grain (wheat, millet, sorghum) and soybeans. Long-term laborers and day workers were employed flexibly to adjust to seasonal fluctuations in the demand for labor. There are no extant written contracts concerning employment of agricultural laborers before the eighteenth century, but samples of long-term labor contracts were included in some daily encyclopedias of the Ming–Qing period. A sample one-year labor contract from a Ming encyclopedia, after stating the arranged sum of wages to be paid quarterly in silver, expressed the laborer’s words of guarantee as follows: “(I will) work on the land diligently from morning till evening; I dare not evade nor neglect my duties, I will not carelessly handle the various tools in the master’s house . . . If an unforeseen accident should happen, it would be my own fate, and the employer need not take any responsibility.”16 These wordings suggest that the power relationship between the employer and the laborer was not equal. During the busy season, markets for day laborers emerged. For example, an entry in the 1752 gazetteer of Linxian county (Henan province) reported, Unemployed people bringing farming tools gather in marketplaces in the morning, and were hired as short-term laborers. They call it the “human market” (renshi 人市). In a busy season, they frequently move from one farm to another. Masters can gain labor, and workers can earn a wage. This practice has been regarded as convenient and has long continued.17

The existence of active markets for short-term labor suggests that many poor households had to supplement their family budget with short-term wage labor employment. Large managerial farms were supported not only by longterm hired labor but also by recourse to the market for short-term labor.

15

16

17

S. Jing and L. Luo, Landlord and Labor in Late Imperial China: Case Studies from Shandong (Cambridge, MA, Harvard University Press, 1978), Chapters 4, 6. Cited in Niida Noboru 仁井田陞, (Hotei) Chu¯goku ho¯seishi kenkyu¯: Dorei no¯do ho¯, kazoku sonraku ho¯ (補訂)中国法制史研究: 奴隷農奴法・家族村落法 (Tokyo, To¯kyo¯ daigaku shuppankai, 1980), pp. 769–70. Cited in Wei Jinyu 魏金玉, “Ming–Qing shidai nongyezhong dengjixing guyong laodong xiang feidengjixing guyong laodong de guodu” 明清时代农业中等级性雇 佣劳动向非等级性雇佣劳动的过度, in W. Li et al., Ming–Qing shidaide nongye zibenzhuyi mengya wenti 明清时代的农业资本主义萌芽问题 (Beijing, Zhongguo shehui kexue chubanshe, 1983), p. 332.

463

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

The status of wage laborers has attracted historians’ attention because the absence of personal subordination was regarded as a criterion for finding the “sprouts of capitalism” in managerial farms of the Ming–Qing period. In reality, the relationship between landowner and laborer varied. The imperial government tried to distinguish the subordinate type of laborers from the equal type by applying the legal status of “indentured servant” to the former. If a laborer was identified as an “indentured servant,” he would be punished more severely for offenses against his master than in a case between equals. The distinction was not easy to maintain, however. The Qing government finally issued a law in 1788 that stipulated that a person should be classified as an “indentured servant” based on social conditions, such as that he dared not sit together with his master on everyday occasions (i.e., apart from ritual ceremonies), dared not drink or eat with his master, dared not talk with his master using the familiar terms “you” (er 爾) and “I” (wo 我), and, in short, had long abided by a master–servant relationship (zhupu mingfen 主僕 名分).18 It can be safely said that wealthy families in late imperial China invested their money in various ways upon careful consideration of the potential profits and risks. In this sense, landowners, as well as tenant farmers, were rational players in the active market economy of this period. Nevertheless, “capitalist” development does not seem to have emerged from the rural sector of China. In the following subsection, we will turn to urban industry.

Workshops in Urban Centers and Hinterland Regions Although rural peasants produced a much greater portion of industrial goods in total volume than did urban manufacturers, some cities enjoyed empirewide fame for their special industrial products, such as the silk industry in Suzhou and Hangzhou and porcelain production in Jingdezhen (see Zurndorfer’s chapter in this volume). As the technologies and recruitment procedures of laborers differed among industries even in the same city, labor markets were not closely integrated. The damasks and brocades produced in Jiangnan cities were already famous in the Song period. An anecdote written by a lower-level official, Xu Yikui, in the late Yuan period described a silk workshop near his home in Hangzhou as follows: Every night I could hear cheerful singing voices till midnight. It seemed that the workers were singing. I sighed and said: “How joyful!” When I passed 18

Takahashi, So¯–Shin mibunho¯ no kenkyu¯, pp. 317–24.

464

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

nearby one morning, I found an old tumbledown building with four or five looms set in the north–south direction, with ten or more workers laboring hard. As they were pale and dazed, I asked them: “Why are you so cheerful though your work looks so toilsome?” A worker answered: “It depends on our thinking. If one is not greedy, he can find joy even in poverty . . . My job is a humble one, but I earn 200 cash a day and I am provided meals and clothing by the master, so I can support my parents and family. Though meals are not so good, I am free from hunger and cold. As I think this situation is normal, I have no other ambition. Our cloth is woven skillfully and valued by people, so our master can sell it easily to earn the money to pay our wages. It is the reason why we express our feeling with one voice and forget our exhaustion.”19

The weaver continued by recounting how a fellow worker who had changed his job in search of a better income finally ruined himself. Although this anecdote apparently was a fictional account created by Xu to argue in favor of being content with one’s lot in life, it suggests that silk-weaving laborers at that time could freely choose their employers. Government-managed silk factories had been established in Jiangnan’s industrial centers such as Suzhou, Hangzhou, and Nanjing since the Southern Song period. The workers in these factories were not always free laborers. In the early Ming period, for example, the skilled workers in the government-operated factories belonged to hereditary artisan households (jianghu 匠戶). This system of household registration based on hereditary occupations was first developed during the Yuan dynasty. The Ming government retained this system in more simplified form by dividing the population into four major occupational categories: commoner, soldier, artisan, and salt worker. In this system, people could not freely choose their occupation, but rather were compelled to provide labor services to the government according to their assigned (or inherited) occupational category. For example, members of artisan households were obligated to work at governmental factories for a certain period of time each year. This system broke down after the midMing period, however, and the government silk factories in Jiangnan gradually replaced compulsory labor with free wageworkers. A gazetteer published in 1684 described silk production in Suzhou city as follows: The residents in the east part of the walled city are mainly engaged in weaving . . . Artisans employed full-time by a master are paid a daily wage, 19

“Zhigong dui” 織工對, in SFG, 1.3a–4a.

465

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

but when additional laborers are needed, the master will hire independent artisans temporarily . . . Artisans without regular employment gather every morning at the foot of bridges and wait to be hired . . . They flock together like refugees by the tens and hundreds, longing to be called. After the morning mealtime, they break up and go home. If jobs in the weaving houses become scarce, these artisans will find it hard to make a living.20

According to this author, the gathering places of laborers varied depending on their particular skill at weaving different kinds of silk textiles. Li Bozhong has estimated that the number of silk-weaving workers in Suzhou city increased from several thousand in the late Ming period to 36,000 in the midQing period. Including other industries such as dyeing, papermaking, and calendering of cotton cloth (a finishing process to smooth and thin the fabric), the total number of industrial workers working in and around Suzhou city during the mid-Qing period amounted to at least 150,000, comprising 15 percent of the adult population.21 It seems that most of these laborers were not Suzhou natives. Hu Fenghui, a mid-Qing Suzhou official, reported in 1723 that most of the dyeing and calendering laborers were originally from Nanjing, Taiping, or Ningguo (all located several hundred kilometers west of Suzhou), and had no home or family in Suzhou. Unlike the weaving of high-grade silk textiles, which required considerable skill, dyeing and calendering workers were simple manual laborers. These laborers, who came from remote rural areas and had no assets other than their own brawny bodies, were regarded as latent criminals by officials. According to Hu, “every time a crime is reported, a calenderer is always implicated.”22 The managers of calendering workshops, known as baotou 包頭, provided the workers with housing, food, and clothing as well as the implements for calendering, and supervised the workers closely. Usually the managers did not own the workshops, but rather leased them from landlords. The managers received contracts for calendering from wholesale cotton merchants and made piecework payments to the workers, who had to repay the managers for their living expenses. As Hu Fenghui’s comments attest, the calendering workers had a notorious reputation for being rowdy and violent, and they often organized demonstrations and strikes demanding higher pay. 20 21

22

CZSZ, 3, fengsu, 4b–5a. B. Li, Duoshijiao kan Jiangnan jingji shi, 1250–1850 多视角看江南经济史, 1250–1850 (Beijing, Sanlian shudian, 2003), Chapter 10. Hu Fenghui 胡鳳翬, “Wei zouwen difang qingxing shi” 為奏聞地方情形事, in GZD: YZ, 1, p. 163.

466

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

Local authorities erected numerous stone steles during the Qing period to proclaim new arrangements for payments to calenderers and to impose restrictions on the workers’ behavior. Although urban silk and cotton workshops operated on a larger scale than rural cottage industries, they did not necessarily displace the latter. The urban workshops specialized in high-quality textile production or finishing processes that were technically difficult for domestic handicraft workers. For ordinary products, rural industries could compete very effectively with urban workshops, owing to the low costs made possible by exploiting the surplus labor of family members. The rural industries coexisted with urban workshops, providing the latter with materials (such as cotton cloth or raw silk) and laborers working away from home. It should be noted that workshops employing large numbers of wage laborers developed not only in the cities in core areas like Jiangnan, but also in inland mountainous regions. Fu Yiling, one of the pioneering socioeconomic historians of China who participated in the “sprouts of capitalism” debates of the 1950s–1960s, argued that in China the sprouts of capitalism spread “from the mountains to the plains”; that is, he believed that the mountainous regions – usually regarded as underdeveloped – were actually the cradles of Chinese capitalism.23 Certainly, the economies of newly developed mountainous regions after the Song were notable for their market orientation. Nevertheless, the market economies of the mountainous regions differed significantly from those of densely populated regions such as Jiangnan, as we can see in the highlands of the upper Han river basin (at the boundaries of Hubei, Sichuan, and Shaanxi provinces). Migration to the Han valley highlands began as early as the Song–Yuan period, but it was not until the Qing period that this area was fully opened for settlement. According to a memorial cited in A Survey for Border Defense of the Three Provinces, an authoritative work on the hinterland regions of Hubei, Shaanxi, and Sichuan in western China compiled in the early nineteenth century, the abundance of land and variety of products in these forested regions attracted millions of migrants: As the mountains and rivers are steep and lands are barren, taxes and corvée labor duties are very light. Upon payment of only several thousand copper cash to the landowners, migrants can lease a number of valleys and hills. Millions of landless people come from Jiangxi, Guangdong, Guizhou, Hubei, 23

Yiling Fu 傅衣凌, Ming–Qing shehui jingji bianqian lun 明清社会经济变迁论 (Beijing, Renmin chubanshe, 1989), pp. 166–8.

467

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

Sichuan, and Shaanxi to reside in [this hinterland region] . . . They are known as “shed people” (pengmin 棚民). Besides cultivating lands, they make their living through employment in lumber mills, salt wells, iron mines, paper mills, and coal mines.24

According to Yan Ruyi, the compiler of the book, a large lumber mill in this region might employ as many as 3,000–5,000 laborers who worked under strict discipline like soldiers. The main investors in the timber industry were Shaanxi merchants.25 The rapid increase in the population of mountainous areas was supported by newly introduced American crops like maize and sweet potatoes, which were suitable for mountain soils. However, the instability of agriculture in mountainous regions made the farming livelihood risky. Most migrant laborers in mountainous areas lacked social safety nets like the kinship and lineage networks that existed in the agricultural areas in the plains. According to Yan Ruyi, migrant laborers “with neither lineage bonds nor ethical checks” assembled friends and fellows to form sworn brotherhoods, and fictitious kinships were formed easily. “They associate and reside indiscriminately with non-family members, therefore adultery and abduction occur every day. Naturally, morals are debased and troubles arise frequently among them.”26 The mountainous areas were breeding grounds for religious associations created as mutual-aid organizations by migrant laborers. When environmental problems such as soil erosion caused by unbridled land development grew more serious, religious societies became more active, which incited officials’ suspicion. It was in this mountainous region on the borders of Hubei, Shaanxi, and Sichuan that the White Lotus Rebellion broke out in 1796.

Changes in Factor Markets Secular Trends in Chinese Factor Markets There are no reliable time-series data for the prices of land or labor over the 800-year period from the year 1000 to 1800. Some scholars who have tried to trace long-term changes in real wages have found declining trends from the Song period to the Qing, but the data are scattered and some Song figures are exceptionally high.27 It is also difficult to construct long-term, time-series data 24

25 27

Zhuo Bingtian卓秉恬, “Chuan–Shaan–Chu laolin qingxing jiyi quchu” 川陝楚老林情 形亟宜區處, in SSBF, 14:16b–18b. SSBF, 9:2a. 26 SSBF, 11:21a–b. For examples, see Kang Chao, Man and Land in Chinese History: An Economic Analysis (Stanford, Stanford University Press, 1986), pp. 217–20; William

468

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

of land prices because the units of area differed according to time and place. Therefore, instead of trying to estimate the long-term changes in factor markets based on quantitative data, I will overview the main changes in factor markets by focusing on the state’s land policies. According to recent studies on Chinese demographic history, the population of China is estimated to have increased from 100 million in the late Northern Song period to 383 million by 1820 (see Cao’s chapter in this volume). It is difficult to determine whether overpopulation was a consistent problem during this period, as the definition of “overpopulation” is not so clear. Nevertheless, it can be safely said that in most regions there was a keen demand for arable land. The imbalance between land and labor was mainly resolved through migration and bringing virgin lands under cultivation. Migration policies varied depending on the dynasty, but in general, migration was not prohibited, and in some cases was encouraged. The early Ming government even coerced people in Jiangnan and Shanxi to migrate to devastated areas in north China. However, this marked the last case of coercive migration in Chinese history. The Qing government sponsored people’s migration to Sichuan, but no longer organized forced migrations – perhaps because population pressure made such coercion unnecessary. Migration policies seem to have been guided by two different considerations. One was the state’s interest in increasing arable lands, easing population pressure, and raising tax revenue. According to the principle of “king’s land, king’s people,” lands without private owners (for example, virgin soil, abandoned lands, newly emerged alluvial lands, etc.) were regarded as the state’s property. The government usually granted ownership to those who reclaimed new lands (or contracted to reclaim lands) on the condition that they paid the requisite taxes. Although the Ming and Qing codes stipulated that those who covertly reclaimed lands without paying taxes were to be punished and the land should be confiscated, in practice officials often sanctioned the cultivator’s ownership rights if he agreed to pay the taxes owed. The other consideration was to avoid social unrest and ensure the empire’s security. For example, while there was hardly any interference with largescale movement of people across the vast territory of the Mongol Empire in the Yuan period, the newly established Ming dynasty prohibited coastal Guanglin Liu, The Chinese Market Economy, 1000–1500 (Albany, SUNY Press, 2015), Appendix E.

469

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

residents from seafaring. Moreover, the Manchu government of the Qing dynasty forbade Han Chinese to migrate to Manchuria, which, as the Manchu homeland, was deemed sacred territory. Furthermore, emigration abroad was banned. Despite these prohibitions, population pressure ensured that clandestine migration would occur. As for migration to the mountainous areas in the interior, the government’s attitude was ambivalent. Ming–Qing rulers had not been positive about developing mines in mountainous regions before the mid-Qing period, when the Qing rapidly expanded development of the Yunnan copper mines to resolve the money shortage. The mingling of Han Chinese with ethnic minorities in the southwestern regions was also restricted by the government for fear that conflict between Han and non-Han peoples could easily cause social disturbances. Next, let us examine the government policies on the expropriation of lands. Although the late imperial states did not restore direct governmental control of land, such as the equal-field system, they sometimes passed laws to expropriate landed property. In 1263 the Southern Song prime minister Jia Sidao enacted a “public-lands” (gongtian 公田) law that forced Jiangnan landowners to sell one-third of their lands in excess of 200 mu to the government at prices much lower than their current value. This law was intended to increase state revenue to cope with the Mongol invasion, as the rents collected from government-owned lands were much greater than the tax revenue collected from the same lands (see the chapter by Lamouroux and von Glahn in this volume). Although this policy was short-lived and subsequently roundly condemned by Confucian scholars, it was not without legitimacy, because in theory all land belonged to the monarch. In the initial stage of the establishment of a dynasty, governments did not hesitate to deprive people of their lands. Zhu Yuanzhang, the founder of the Ming dynasty, expropriated lands from landowners and created a large amount of state-owned land (guantian 官田) in Jiangnan, which had been the base of his most powerful rival. Following this expropriation, government lands comprised 45 percent of the total registered land in this region. Tenants on these government lands had to pay rents that were much higher than the land tax on private lands. In principle, they could not sell their tenancies freely, but over time such land sales came to be overlooked. Land reforms in the fifteenth and sixteenth centuries erased the distinction between government lands and private lands, essentially privatizing the government lands and equalizing the burden of state payments levied on both. A similar process of privatization can be seen in other cases of governmentowned land, for example the “enclosed lands” (quandi 圈地) in the Qing 470

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

period. After Manchu forces occupied north China in 1644, they confiscated properties near Beijing, the capital city, to provide lands for princes and soldiers of the Eight Banners. These banner lands were state-owned lands that were not supposed to be sold, but after the mid-Qing period, impoverished bannermen began to mortgage or sell them to Han Chinese, and the system of banner lands broke down. The land tax remained one of the basic sources of government revenue throughout the Song–Qing period. Changes in the land tax had great influence on land markets. The main trends in state taxation during this period were as follows. (1) Shift from in-kind taxation to money taxes. When the twice-a-year tax system was introduced in the late eighth century, it was stipulated that the tax should be paid in coin. As this method exacerbated the serious shortage of money in the private economy, the government soon permitted landowners to pay the tax in kind instead. During the Song period, actual methods of payment of the twice-a-year tax varied, as the twice-a-year tax payment could be converted into silk, money, and so on. In the early Ming, payment in kind became the rule again, but from the mid-Ming period the land tax was gradually monetized, and after the Single-Whip reform (yitiao bianfa 一條 鞭法) in the sixteenth century most of the land taxes were paid in silver. (2) Simplification of tax assessments. Before the mid-Ming period, the burden levied on each household was decided not only by the size of the landed property but also by other factors such as the number of its male members or the total value of its assets. The assessment of these factors was a hotbed of official corruption. One of the purposes of the Single-Whip reform was to unify the complicated methods of taxation into a simple assessment based on landownership. In the mid-Qing period, the remaining poll tax for labor services was also incorporated into the land tax. These tax reforms enabled governments to collect taxes from landowners more efficiently. (3) The disparity between the statutory tax quota and the real burden of taxation. It was a characteristic of the late imperial tax system that the formal quota of land taxes assessed by the central government was fixed early on and remained relatively static thereafter. Nevertheless, actual administrative costs would increase along with population growth and price inflation. Local governments resorted to informal surtaxes or compulsory services on landowners to compensate for fiscal shortages. As these informal surtaxes and services were beyond the control of the central government, the actual

471

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

burden arbitrarily levied by local officials sometimes became several times as high as the formal tax. This dual structure of taxation could not be eradicated despite repeated attempts at reform. During the eight centuries from the year 1000 to 1800, the alienability of land and mobility of people generally increased, and the imperial state tried to adapt its institutions to these trends. It should be noted, however, that officials were always sensitive to the risks accompanying such high mobility. They were not as optimistic about market forces as some of today’s historians, who seem to believe that the market would naturally bring peace and prosperity to society.

Booms and Depressions Although China’s factor markets gradually expanded their sphere of influence during the Song–Qing period, it was not a linear process of evolution. Cycles of booms and depressions exerted great influence on people’s behavior in factor markets. The simplest indicator of economic cycles was the price of land. When people had plenty of money to invest and expected agricultural products to be sold at good prices, then people would buy land, and the price of land would increase. On the other hand, when impoverished landowners were forced to sell their lands because of poor sales of agricultural products or heavy land taxes, the price of land would fall. As there are no time-series data for land prices in the Song–Yuan period, I will use data from 1500 to 1800 (i.e., from the mid-Ming to the mid-Qing period) to illustrate the relationship between economic cycles and land markets. In Figure 12.1, land price (A) is a time-series of prices of paddy lands in Huizhou and neighboring prefectures compiled by Chao Kang based on documents housed at Nanjing University,28 while land price (B) shows the price data of paddy lands in Huizhou based on contracts stored at Anhui Provincial Museum and the Institute of History of the Chinese Academy of Social Sciences.29 The rice-price data are a time series made by Peng Xinwei of scattered data collected from various materials all over China. Although there are more reliable time-series data for rice prices during the Qing period,

28 29

Chao, Man and Land, p. 130. Anhuibowuguan 安徽博物馆 et al. (eds.), Ming–Qing Huizhou shehui jingji ziliao congbian 明清徽州社会经济丛编, 2 vols. (Beijing, Zhongguo shehui kexue chubanshe, 1988, 1990); Liu Hehui 刘和惠 and Zhang Aiqin 张爱琴, “Mingdai Huizhou tianqi yanjiu” 明代徽州田契研究, Lishi yanjiu 历史研究 1983.5, 125–39.

472

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets 400 350 300 250 200 150 100

0

1501–1510 1511–1520 1521–1530 1531–1540 1541–1550 1551–1560 1561–1570 1571–1580 1581–1590 1591–1600 1601–1610 1611–1620 1621–1630 1631–1640 1641–1650 1651–1660 1661–1670 1671–1680 1681–1690 1691–1700 1701–1710 1711–1720 1721–1730 1731–1740 1741–1750 1751–1760 1761–1770 1771–1780 1781–1790 1791–1800

50

Price per mu (A)

Price per mu (B)

Rice prices

Figure 12.1 Land prices in Huizhou and rice prices in China, 1500–1800 (ten-year average; index: 100 = 1541–1550)

Peng Xinwei’s data are the only time-series data available so far that cover the entire 1500–1800 period.30 We can distinguish several phases in the economic trends shown in this figure. Phase 1 is the sixteenth century, during which rice prices were stable or rose slightly but land prices fell. Phase 2, during the first half of the seventeenth century, was characterized by increases in the prices of both rice and land. During Phase 3, in the latter half of the seventeenth century, the prices of rice and land fell and remained at a low level. In Phase 4, after 1700, the prices of rice and land rose in tandem, at an especially rapid pace during the middle of the century. Phase 1

Narrative descriptions written by contemporary literati of the economic situation in the sixteenth century unanimously attribute the fall in land prices to heavy taxes and labor service duties. The essayist Yu Bian reported the cautionary story of the ruin of a Jiangnan landowner who had greedily purchased considerable land during the Hongzhi era (1488–1505), and added his general observation: 30

Peng Xinwei 彭信威, Zhongguo huobishi 中国货币史, 3rd ed. (Shanghai, Shanghai renmin chubanshe, 1965), pp. 705, 859.

473

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

In recent years (the 1520s), large landowners categorized as “superior [i.e., wealthy] households” (shanghu 上戶) have been obliged to perform labor service as tax captains (liangzhang 糧長). Because of the burden of this service, only a few large landowners did not lose their property within a few years. Learning from this lesson, people would not buy land, and the price of land has fallen sharply. Land formerly valued at ten liang per mu cannot be sold now even at one or two liang . . . Even in Huizhou, where land is more expensive than anywhere else south of the Yangzi, land that once sold for twenty to thirty liang cannot find any buyer even at five or six liang.31

Although the fall of land prices shown in Figure 12.1 is less drastic than this essay describes, the downward trend across the sixteenth century is obvious. The main reason for the increase in the taxes and labor services levied on land was the continual warfare against the Mongols in the northern border regions. Military expenditures for defense of the northern border rose continuously throughout the sixteenth century. For example, the silver disbursed from the central treasury for current military expenditures on the northern border (jingyun nianli yin 京運年例銀) increased from less than a million liang annually in the 1530s to 4 million liang around 1600 (see the chapter by Lamouroux and von Glahn, especially Table 9.2, in this volume).32 This amount did not include emergency military expenditures or funds delivered directly from other regions. Curiously, in spite of the increase in military expenditure, the formal quota of land taxes stayed at a low level. As mentioned above, however, one of the characteristics of the late imperial tax system was the disparity between the statutory tax quota and the real burden. Under fiscal pressure, local officials arbitrarily levied surtaxes and compulsory services beyond the formal quota that increased the actual burden on taxpayers. The increase in these informal levies induced other serious problems, including corruption and inequities in tax assessment and collection. One of the purposes of the Single-Whip reform was to resolve these problems by standardizing surtaxes and compulsory services, but its effect was limited. Taxes collected in silver were transported every year to the border regions, and most agricultural areas suffered from a scarcity of silver. Landowners were forced to sell more products to obtain scarce silver to pay their taxes, which increased the real tax burden and in turn reduced income from landownership. The demand for land declined, and land prices consequently fell. The beneficiaries of this flow of silver were officials, military officers, and privileged merchants who supplied military stores. Among the merchants 31

SQXY, 8:2a–b.

32

Kishimoto, Keizai hendo¯, Chapter 6.

474

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

who played an essential role in military logistics were the Huizhou merchants, who built up enormous fortunes during the late Ming period. They invested their abundant capital not in land, which they regarded as yielding little profit, but in commerce and moneylending (see McDermott’s chapter in this volume), so the land market remained slack. The scarcity of silver strengthened the power of merchants in capital markets. The case of silk producers cited above shows how rural households endeavored to earn their living in a buyers’ market dominated by merchants and moneylenders who had silver. Some members of the landowning classes also enjoyed privileges. Government officials (including ex-officials), as well as holders of higher civil service examination degrees, enjoyed the privilege of exemption from compulsory services, and thus they could accumulate lands without worrying about bankruptcy caused by the rise in informal surtaxes and services. They preferred to reside in cities where they could lead a luxurious lifestyle and enjoy a rich social life. In the late Ming period, there was a sharp contrast between the prosperous cities and the impoverished rural areas. Phase 2

The turning point in price trends varied among regions. According to local gazetteers and essayists, land prices seem to have increased in lower Yangzi localities such as Nanjing and Songjiang, as well as in coastal Fujian after the Single-Whip reform of the 1570s. However, the empire-wide rise in grain prices did not begin until the 1620s. The 1620s to 1650s was a turbulent era, when Chinese people suffered from famines, rebellions, and the wars of the Ming–Qing transition. However, in areas such as Jiangnan that were not heavily affected by these calamities people had opportunities to benefit from surging grain prices and the abrupt increase in the state’s needs for provisions, clothing, and other military supplies as the war fronts shifted from remote northern border regions to the interior of the empire. According to Ye Mengzhu, who wrote vivid descriptions of Shanghai local society during the Ming–Qing transition period, already in the Chongzhen reign (1628–1644) “there were no lowpriced lands nor large-scale properties reaching 10,000 mu.” In the early years of the Qing dynasty, competition for land rose further: In the early Shunzhi era (1644–1661), rice prices soared and people scrambled for lands. They contended and sued, claiming redemption or additional payments for the lands they had already sold, as well as trying to defraud

475

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

neighbors of adjacent fields. As a result, some were bankrupted, while others made fortunes . . . The dearness of land reached its peak in this period.33

Landowners earned high profits, but at the same time they faced the danger of intensifying land disputes. Poor ex-owners often would bring a lawsuit against the buyers, claiming that they had rights to redeem the lands, or at least – since the value of the land had increased – that they should be paid additional money even for lands relinquished through outright sale. The plaintiffs usually could squeeze some additional payment from the buyers. Ming and Qing local judges often ordered a wealthy buyer to make a concession to a poor seller, even if the buyer was legally faultless. Judges prioritized harmonious relationships among rich and poor over the strict protection of property rights.34 Rising prices for agricultural products encouraged landowners to invest more effort in farm management. Master Shen’s Treatise on Agriculture, mentioned above, was one illustration of how a managerial farm was operated at this time. It is difficult to show the trends in real wages because the data are so sparse, but some observers pointed out a rise in the wages of agricultural laborers. This rise in wages seems to have begun as early as the 1620s, when a Huzhou official complained that “recently the costs for agricultural laborers have been rising day by day.”35 In the Shunzhi era, a scholar in Jiaxing prefecture in northern Zhejiang also remarked that laborers were lazy despite their high wages.36 Phase 3

The boom in the mid-seventeenth century did not last long. Ye Mengzhu described the collapse of the Shanghai land market in the 1660s as follows: [After the boom in the early Shunzhi period], the price of rice gradually declined, yet taxes and labor services became more burdensome day by day. Consequently, the land prices immediately began to fall . . . Ordinary lands were not worth a single copper cash, and even the most fertile lands were valued at a mere 0.3–0.5 liang per mu . . . Some prudent families took this opportunity to accumulate large tracts of land; consequently, the biggest landowners amassed several tens of thousands of mu of rice fields, and the second-class ones 10,000 to 50,000 mu. This was a drastic change in the situation of landholding.37

33 36

YSB, 1, pp. 22–3. YYXQ, p. 1414.

Kishimoto, “Property Rights,” pp. 79–82. YSB, 1, p. 23.

34 37

476

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

35

YCXP, 2, p. 46.

Property Rights and Factor Markets

Ye Mengzhu attributed the collapse of the land market to the arbitrary imposition of compulsory services and the fall in rice prices. Although the statutory tax quotas of the early Qing period were lower than in the late Ming, collection of taxes became more rigorous, and the retrenchment policy of the central government made it inevitable for local governments to levy informal surtaxes and services to cover their administrative costs. In addition, the Qing government enforced a strict maritime ban from the late 1650s to the mid-1680s to prevent the Zheng family’s anti-Qing regime on Taiwan from trading with people in the coastal areas. Owing to this blockade policy, the quantity of silver flowing into China decreased, and the “scarcity of silver” again became the object of popular resentment. The warning by Zhang Ying against commercial investment mentioned above was made in this context. It was precisely because people hastily sold their lands as profitless (or even harmful) property that Zhang argued in favor of the advantages of landholding in the long run. The state of the labor market also changed with deflationary economic trends. In an essay written around 1671, Wei Jirui – a native of southern Jiangxi – described the poverty spreading all over China. According to Wei, even in the most prosperous areas such as Suzhou, Hangzhou, Jiangning (Nanjing), Fujian, and Guangdong, not only poor households but also upperclass families complained of economic difficulties: For the last few years, the price of rice in Jiangxi was at its nadir . . . Even so, many people died of starvation. Peddlers and day laborers usually said that they wanted to eat expensive grain. For, when grain is expensive, the rich have money and can employ laborers and buy goods, so that the petty folk can share in their good fortune. Now, with all the rich impoverished, the poor are even less able to procure a livelihood.38

The situation of poor people wanting “to eat expensive grain” suggests that their livelihood had become deeply dependent on the labor market. Unlike Phase 1, when at least cities enjoyed prosperity, the depression of Phase 3 affected not only rural areas but cities and market towns as well. Phase 4

In the 1680s, after the Qing government lifted the maritime ban and eased the retrenchment policy, rice and land prices recovered, but they did not show

38

“Zongkuo dayi” 總括大意, in SCTG, 10.64b.

477

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

a definitive rising trend before the 1740s. Huang Ang, a local scholar from Changzhou prefecture, wrote around 1752, Recently, the price of land has become several times as high compared to the Yongzheng reign (1723–1735). Perhaps this is because in former years the tax collection was very severe, and people only considered the disadvantages of landowning, with the result that families sold lands while few bought them. At the beginning of the Qianlong era (1736–1795), taxes in arrears were forgiven, and villagers were freed from the scourge of clerks dunning them for tax payments. What is more, rice prices rose, so people became more and more aware of the benefits of landowning. Accordingly, nowadays many families buy lands, while few sell them.39

Huang Ang also made the interesting observation that at that time most lands were sold by absentee landlords dwelling in towns to tenant farmers. According to Huang, the rise in commodity prices impoverished city dwellers who lived extravagant lives, while higher prices for agricultural goods gave cultivators incentive to increase their savings through frugality and diligence. As in Phase 2, land disputes over redemption and additional payments occurred frequently during this period. The rise in rice prices in the late 1740s attracted the attention of many contemporaries, including the Qianlong Emperor. The emperor and high officials conducted a spirited debate over the causes of the inflation of rice prices, which was mainly attributed to the increase in population and the government’s purchase of grain for the ever-normal granaries – that is, the relative scarcity of grain in the marketplace.40 If overpopulation was the cause of the increase in rice prices, it is plausible that real incomes were decreasing and people were becoming impoverished. However, we should note that some observers welcomed high prices as a symbol of prosperity. Wang Huizu wrote in 1794, “Nowadays rice is always expensive but people enjoy their lives. Formerly only rice was expensive, but today fish, shrimp, vegetables and fruits are all high-priced, so peddlers and farmers can earn their living.”41 It is difficult to estimate the trends in the labor market in the eighteenth century. Some scholars have argued that the income of agricultural laborers increased with the improvement of their social status,42 while others have 39 40

41

XJSXL, 1.19a. For detailed analysis of the “grand discussion” at that time on the cause of rising rice prices, see Helen Dunstan, State or Merchant? Political Economy and Political Process in 1740s China (Cambridge, MA, Harvard University Asia Center, 2006). BTMHL, 1, p. 532. 42 For example, Wei Jinyu, “Ming–Qing guyong laodong,” pp. 439–40.

478

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

argued that “a remarkable decline in real wages” of agricultural laborers occurred between 1744 and 1819 based on the data collected from the judicial records in the archives of the central government.43 The data collected recently by Robert Allen and his collaborators, whose main purpose was to compare wage rates in Europe and China, indicated that the real wages in Beijing, Jiangnan, and Guangzhou showed no significant upward or downward trend during the period from the 1740s to 1800.44 In any event, it can be safely said that population pressure and the increase in effective demand worked together to push people toward more remote inland and highland areas, where market-oriented production spread together with environmental disruption (see Bello’s and Pomeranz’s chapters in this volume).

Famines and Factor Markets In long- or middle-term trends, land prices usually moved in tandem with grain prices, because high grain prices encouraged land investment. On the contrary, in regions struck by famines, grain prices and land prices moved in opposite directions. Many descriptions of famines make reference to a sharp decline in land prices. The case of the 1784–1786 drought in north China provides an illuminating example of the land market situation during a period of famine. A local gazetteer of Yancheng county, Henan province, recorded that in the spring of 1786 “the price of one-tenth of a shi of wheat rose as high as 1,400 copper cash, while the price of land was no more than 1,000 cash per mu”; that is, the price of a hectare of arable land had tumbled to less than 160 liters of wheat.45 It is difficult to estimate the average harvest yield in eighteenth-century Henan, but we can make a rough guess based on a case in early twentieth-century Shandong, adjacent to Henan, where the yield of wheat in ordinary years was approximately 1,300 to 1,700 liters per hectare.46 Though the record from Yancheng county might be exaggerated, we can readily imagine how drastic the fall of land prices in Henan was in 1786. The governor-general of Henan at the time was Bi Yuan, a famous scholarpoet of the day. In his memorials to the emperor, Bi proposed that people 43

44

45 46

For example, Liu Yongcheng 刘永成 et al., “Shiba shijiu shiji Zhongguo nongye gugong de shizhi gongzi biandong qushi” 十八、十九世纪中国农业雇工的实质工资变动趋 势, in Zhongguo diyi lishi dang’anguan 中国第一历史档案馆 (ed.), Ming–Qing dang’an yu lishi yanjiu 明清档案与历史研究, vol. 2 (Beijing, Zhonghua shuju, 1988). Robert C. Allen, Jean-Pascal Bassino, Debin Ma, Christine Moll-Murata, and Jan Luiten van Zanden, “Wages, Prices, and Living Standards in China, 1738–1925: In Comparison with Europe, Japan, and India,” Economic History Review 64 (suppl. 1) (2011), 8–38. YCXJ, 5:37a–b. Jing and Luo, Landlord and Labor, p. 22. The units used in the original data were jin (unit of weight) and mu.

479

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

who had sold their lands at low prices during the famine should be permitted to redeem their lands, even if the lands had been sold outright. He wrote, The powerful rich families of Shanxi and other provinces who heard the news [of the distress of the Henan people] came to Henan and seized people’s lands as a substitute for the repayment of loans . . . The rich who make profits taking advantage of others’ misery are indeed detestable. Unless the government enforces precise regulations to permit sellers to redeem their lands, Henan people will become more impoverished day by day while those in other provinces will become more enriched day by day. How can the distressed people recover their vigor?47

The Qianlong Emperor approved Bi’s proposal and 563,600 mu of lands were redeemed by the original owners. It should be noted, however, that this policy of allowing the redemption of lands after a famine was an isolated case. Similar proposals were often made during the Qing period but rarely approved, because most officials, although sympathetic to the predicament of the distressed, worried about the confusion caused by the cancellation of contracts once concluded. These different attitudes of officials to the distressed sale of lands suggest that there was no consensus about the absolute validity of contracts. Just as the distressed sale of land in famine years brought about the collapse in the land market, natural calamities exerted a powerful influence on the labor market, too. In famine years, people could not afford to buy commodities other than food, nor could they employ workers, so laborers lost their source of income. This double assault by scarcity of food and loss of income brought about serious subsistence crises. The policy of “providing work as famine relief” (yigong daizhen 以工代賑), which was famously championed by the renowned statesman Fan Zhongyan in the Northern Song period, was one method to increase employment in the period of famine. Most famine administration handbooks published during the Ming–Qing period advocated this policy, whereby officials would recruit indigent people to work in water control projects, constructing bridges, repairing city walls, and so on. Not only local officials but also local elites would take the initiative in “providing work as famine relief.” Whether funded by the state or by nongovernmental groups, this policy can be understood as an effort to create employment using a nonmarket method in a period of crisis, when the collapse of the labor market threatened people’s subsistence. 47

Bi Yuan 畢沅, “Hefu Bi jinzou wei zouming shi (Qianlong wushiyinian xiaji)” 河撫畢 謹奏為奏明事(乾隆五十一年夏季), in TLML, ce 冊 271, 19a–b.

480

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

When we try to draw a vivid picture of the factor markets from the year 1000 to 1800, not only long-term trends but also middle-term and short-term fluctuations should be taken into consideration. From the viewpoint of individual economic actors, middle-term or short-term changes might have been the most important factors to guide their behavior. In the early 1930s, R.H. Tawney compared the position of the rural population in Republican China to “that of a man standing permanently up to the neck in water, so that even a ripple is sufficient to drown him.”48 If this comparison has some validity for Song–Qing China, too, we should study not only the depth of water, but also the “ripples.” Although this section has focused on economic fluctuations during the Ming–Qing period because of the relative abundance of related materials for this era, the economic situation of the Song–Yuan period likely was not that different from that of later periods. During the first half of the thirteenth century, when the excessive issue of paper money caused inflation, lawsuits related to conditional sales were filed frequently.49 In 1297, in the midst of high inflation, a Yuan official reported that “commodity prices are climbing rapidly, and the price of real estate has risen severalfold, so that the speculative spirits in people are stirred up, and this has caused an increase in lawsuits.”50 This observation suggests that the market for real estate fluctuated according to commodity prices in the Song–Yuan period as well.

Conclusion It seems that the old image of China’s economy during the late imperial period as a static and restrictive one with inactive factor markets has already been replaced by a new view that emphasizes its free, vigorous, and marketoriented characteristics. Kenneth Pomeranz contends that “eighteenthcentury China . . . actually came closer to resembling the neoclassical ideal of a market economy than did western Europe.” He argued that hereditary rights restricting the sale and lease of lands in fact were weaker in eighteenthcentury China than in Western Europe in the same period. Moreover, bound labor had become inconsequential in China by the eighteenth century at the latest, while villeinage and seigneurial restrictions still endured in eighteenthcentury Europe. Migration to frontier regions was less restricted in China than in Europe in the sixteenth through eighteenth centuries. Although 48 49

R.H. Tawney, Land and Labor in China, rpt (Boston, Beacon Press, 1966), p. 77. Kishimoto, “Property Rights,” pp. 77–8. 50 YDZ, 19:7a–b.

481

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

mio kishimoto

migration toward the core areas with plentiful capital was easier in Europe, nonetheless China was closer to the model of a smoothly functioning neoclassical labor market.51 This argument compels us to re-examine the meaning of “the neoclassical ideal of a market economy.” Probably, it can safely be said that factor markets in late imperial China were less restrictive than those of most European societies during the same period. People in China could even sell themselves and their family members rather freely, which shocked the Europeans who visited China after the late Ming period. The imperial government in the Song–Qing period rarely interfered with various local practices that might encroach on the strict rights of landownership, as long as they did not affect tax collection or social stability. Free trade was not necessarily guaranteed, but was widely tolerated. This type of economic freedom brought about many risks as well as opportunities. We can find in the economic behavior of Chinese in the late imperial era both haste to seize opportunities and prudence in avoiding risk. They relied more on personal networks than on formal institutions to make profit and to protect themselves in a highly competitive market economy. How can we best understand China’s economic performance from the Song to mid-Qing period? This is a controversial question that has puzzled historians of Chinese economy for more than a century. The controversy surrounding this question is caused not only by the lack of quantitative data but also by the difficulties in characterizing Chinese economic institutions according to the models constructed out of European historical experiences. On one hand, the late imperial Chinese economy shows ostensibly “modern” features in the sense that, for example, people could freely trade land, choose jobs, and move to other regions. On the other hand, Chinese society seems to have lacked precise concepts of proprietorship and human rights, which are the cornerstones of modern European society. The answer to that question would not be to emphasize its “modern” character, nor to categorize it as feudal or Asiatic, but rather to seek to analyze Chinese economic behavior from an emic perspective within the specific economic and cultural environment of late imperial China.

51

Kenneth Pomeranz, The Great Divergence: China, Europe, and the Making of the Modern World Economy (Princeton, Princeton University Press, 2000), pp. 70–85.

482

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

Property Rights and Factor Markets

Further Reading Adachi Keiji 足立啓二, Min Shin Chu¯goku no keizai ko¯zo¯ 明清中国の経済構造 (Tokyo, Kyu¯ko shoin, 2012). Allen, Robert C., Jean-Pascal Bassino, Debin Ma, Christine Moll-Murata, and Jan Luiten van Zanden, “Wages, Prices, and Living Standards in China, 1738–1925: In Comparison with Europe, Japan, and India,” Economic History Review 64 (suppl. 1) (2011), 8–38. Aoki Atsushi 青木敦, So¯dai minjiho¯ no sekai 宋代民事法の世界 (Tokyo, Keio¯ gijuku daigaku shuppankai, 2014). Beattie, Hilary J., Land and Lineage in China: A Study of T’ung-Ch’eng County, Anhwei, in the Ming and Ch’ing Dynasties (Cambridge, Cambridge University Press, 1979). Chao, Kang, Man and Land in Chinese History: An Economic Analysis (Stanford, Stanford University Press, 1986). Dunstan, Helen, State or Merchant? Political Economy and Political Process in 1740s China (Cambridge, MA, Harvard University Asia Center, 2006). Fu Yiling 傅衣凌, Ming–Qing shehui jingji bianqian lun 明清社会经济变迁论 (Beijing, Renmin chubanshe, 1989). Grove, Linda and Christian Daniels (eds.), State and Society in China: Japanese Perspectives on Ming–Qing Social and Economic History (Tokyo, University of Tokyo Press, 1984). Hansen, Valerie, Negotiating Daily Life in Traditional China: How Ordinary People Used Contracts, 600–1400 (New Haven, Yale University Press, 1995). Jing, Su and Lun Luo, Landlord and Labor in Late Imperial China: Case Studies from Shandong (Cambridge, MA, Harvard University Press, 1978). Iwai Shigeki 岩井茂樹, Chu¯goku kinsei zaiseishi no kenkyu¯ 中国近世財政史の研究 (Kyoto, Kyo¯to daigaku gakujutsu shuppankai, 2004). Kishimoto Mio 岸本美緒, Shindai Chu¯goku no bukka to keizai hendo¯ 清代中国の物価と 経済変動 (Tokyo, Kenbun shuppan, 1997). Li Bozhong 李伯重, Duoshijiao kan Jiangnan jingjishi, 1250–1850 多视角看江南经济 史，1250–1850 (Beijing, Sanlian shudian, 2003). Li Wenzhi 李文治, Wei Jinyu 魏金玉, and Jing Junjian 经君健, Ming–Qing shidaide nongye zibenzhuyi mengya wenti 明清时代的农业资本主义萌芽问题 (Beijing, Zhongguo shehui kexue chubanshe, 1983). Liu, William Guanglin, The Chinese Market Economy, 1000–1500 (Albany, SUNY Press, 2015). Ma, Debin and Jan Luiten van Zanden (eds.), Law and Long-Term Economic Change (Stanford, Stanford University Press, 2011). Pomeranz, Kenneth, The Great Divergence: China, Europe, and the Making of the Modern World Economy (Princeton, Princeton University Press, 2000). Takahashi Yoshiro¯ 高橋芳郎, So¯–Shin mibunho¯ no kenkyu¯ 宋―清身分法の研究 (Sapporo, Hokkaido¯ daigaku tosho kanko¯kai, 2001). Zelin, Madeleine, Jonathan K. Ocko, and Robert Gardella (eds.), Contract and Property in Early Modern China (Stanford, Stanford University Press, 2004).

483

https://doi.org/10.1017/9781108587334.014 Published online by Cambridge University Press

13

The Rural Economy kenneth pomeranz

The rural economy will predominate in almost any preindustrial society – perhaps particularly so in China. No barriers comparable to medieval Europe’s guild rules made large sectors into urban monopolies; and though China was probably the world’s most urbanized large society c. 1200, and perhaps still as urban as Europe in the late 1600s, much of its elite lived in the countryside rather than in cities or fortified castles (especially between roughly 1100 and 1550). Moreover, the property systems prevailing in China’s most commercialized areas created incentives for most nonelite families to remain in the countryside, transferring labor not needed for farming to handicrafts without moving to town. The result was a highly diversified rural economy and cities that, though often quite large, were much smaller than the rural surplus could have supported. Histories of China’s rural economy tend to follow one of two paths. Some are histories of what Marxists call the forces of production: crop varieties, irrigation systems, and other technologies; climate; and resources, especially land/labor ratios. Others emphasize relations of production, broadly construed: landholding, labor and tenancy arrangements, the extent of markets, relations with the state, and so on. In reality, of course, the two intersect, jointly determining the welfare of producers, consumers, and communities. This chapter follows a rough arc from more physical to more social topics, while insisting that the two are never really separate. It ends with a brief discussion of living standards near the end of our period, and looming impediments to further rural economic growth. As we shall see, regional differences were profound. To explore these regional differences, this chapter will refer to the physio-geographic “macro-region” definitions developed by G. William Skinner (see Map 13.1).

484

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

Map 13.1 Macro-regional structure of late imperial China Source: map from G.W. Skinner (ed.), The City in Late Imperial China. Copyright © 1977 by the Board of Trustees of the Leland Stanford Jr. University. All rights reserved. Used by permission of the publisher, Stanford University Press, sup.org

“Natural” Conditions and Resources: A Baseline Climate Farmers worry constantly about weather. Climatically, our period divides into three broad subperiods. During 930–1320, temperatures were relatively

485

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

warm (which increases rice yields), and the East Asian monsoon generally strong; these were beneficial conditions for agriculture in the most densely populated parts of China (see Bello’s chapter in this volume). The monsoon generally makes landfall on China’s southeast coast; strong monsoon winds carry the clouds considerable distances to the west and north before they have released all their moisture, so inland areas get enough rain, and the coast does not get too much. Beginning around 1320, the climate cooled, with a slight warming from the 1560s to 1610s, followed by an especially cold period lasting until roughly 1710. That cooler climate corresponded, as it generally does, with frequent weak monsoons, bringing more prolonged rains (and thus flooding) to the southeast coast, and droughts to northern and central China. Thus the 1600s were difficult for Chinese agriculture, and for much of the northern hemisphere; epidemics and sociopolitical instability compounded the problems. However, the turmoil also fostered important institutional changes. Temperatures rose and monsoons became generally stronger during the eighteenth century, though conditions were still less optimal than during 930–1320.1 The Qing had restored political stability by the 1680s. An impressive economic upturn began by 1700. By the mid-1700s, however, other signs of stress had begun to appear, loosely linked to the period’s unprecedented population growth. By the late 1700s, there were signs of both environmental and institutional fragility, only partially offset by continuing diffusion of improved agricultural practices and continuing growth of regional marketing networks. (Trans-regional networks were probably deteriorating after midcentury, however.) After about 1820, more unfavorable climatic conditions returned, while governance became less effective. These trends combined with the above-mentioned stresses and new pressures from abroad to produce major economic, social, environmental, and political crises. But we cannot measure the weight of climate per se. Overall, climate trends certainly mattered, but the correlation between shifts in the weather and socioeconomic outcomes is too loose to give these forces a dominant role. The very best climate period (960s–1020s) featured a buoyant economy, and the worst climate periods (1350s–1380s and 1580s–1640s) were very bad, but in general, climate was just one of many factors.

1

Pingzhong Zhang et al., “A Test of Climate, Sun, and Culture Relationships from an 1810-Year Chinese Cave Record,” Science 322 (2008), 940–2.

486

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

Land and Water Circa the year 1000, north China’s population was still far lower than before the An Lushan Rebellion (755–763). South China’s population had grown considerably, but the incorporation of mostly well-watered (but also overgrown and disease-ridden) southern regions was still at an early stage. Consequently, land– labor ratios were near their most favorable levels, in all regions. Even in the very densely populated and highly productive Yangzi delta, an average farm family probably cultivated about forty mu (2.7 hectares) during the Song period (960–1279), versus roughly ten mu in the eighteenth century.2 Vast amounts of additional land came under cultivation over the next eight centuries. Not surprisingly, this proceeded most rapidly in periods of significant population growth; but the correlation was by no means perfect, and the vectors of causation are often unclear. Crucially, the addition of farmland could not match long-run population growth: population went from probably fewer than 70 million in the year 1000 to perhaps 350 million in 1800 (at least a 400 percent increase), while cultivated acreage grew by perhaps 178 percent, from roughly 360 million to 1 billion mu.3 Increasing yields were thus essential. Other resources, such as timber, were also relatively plentiful in the year 1000 if judged on an empire-wide basis, but were already becoming scarce in some densely populated areas. While we have very little evidence with respect to animals, it seems likely that the relatively high land–labor ratio of the Song would have gone along with at least slightly higher per capita supplies of draft animals than in the more crowded late empire. While it is at least possible that most Song dynasty rice-farming households owned an ox, this was probably no longer true in Ming times. However, as Li Bozhong notes, sharing or renting oxen was common for Qing farmers, allowing oxen to be more widely used in Qing rice fields than in Ming ones despite greater population pressure.4 If there were problems created by decreased per capita 2

3

4

Bozhong Li, “Was There a ‘Fourteenth-Century Turning Point’?”, in Paul Jakov Smith and Richard von Glahn (eds.), The Song–Yuan–Ming Transition in Chinese History (Cambridge, MA, Harvard University Asia Center, 2003), pp. 163–4, 170. Eighteenthcentury estimates are discussed in Kenneth Pomeranz, “Facts Are Stubborn Things: A Response to Philip Huang,” Journal of Asian Studies 62.1 (2003), 172. For acreage, see Dwight Perkins, Agricultural Development in China, 1368–1968 (Chicago, Aldine Publishing, 1969), pp.229, 234. A population of 70 million in the year 1000 implies a relatively low rate of growth in the following century, and so is likely an overestimate; for 1078, see Wu Songdi 吴松弟, Zhongguo renkou shi 中国人口史, vol. 3, Liao Song Jin Yuan shiqi 辽宋金元时期 (Shanghai, Fudan daxue chubanshe, 2000), pp. 135, 162. Bozhong Li, Agricultural Development in Jiangnan, 1620–1850 (New York, St. Martin’s Press, 1998), pp. 45–6, 77–9.

487

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

supplies of animals, they would have been particularly significant in the north, where animal labor was more needed, than south of the Yangzi. Northern farms required deep ploughing more than southern rice paddies did; grinding wheat, barley, and sorghum into flour also benefited from animal traction not needed in preparing rice for cooking. The north’s relative lack of navigable waterways also raised demand for work animals. Meat animals present an even cloudier picture. Dwight Perkins has speculated that the per capita supply of hogs – the most important meat animal – was probably unchanged from early Ming to the twentieth century; there are also grounds for thinking that the number may have declined slightly. Again, any changes would have mattered most in the north, since fish were more plentiful in southern diets. We know even less about pre-twentieth-century poultry supplies. Water was also a critical resource. Here, however, the label “natural resources” becomes unhelpful, as the crucial issue was often not the amount of water (especially in the south) but the degree to which technology and social organization allowed people to manipulate it. In this case, then, resource availability per capita could grow along with population if the right investments were made, and often did. Consequently, we will consider water later, along with products of human activity.

Forests and Other Nonagricultural Resources Since forests are dealt with in David Bello’s chapter in this volume, I mention them only briefly here. It is, however, noteworthy that even before our period’s great expansion of cultivated acreage, forest resources were already scarce in much of north and northwest China, and in the Yangzi delta. Thus even in Song times – and much more so by the Qing – Chinese farmers depended to an unusual degree on their own production and/or purchases for animal feed, fertilizer, and fuel. Fertilizer in particular became increasingly commercialized, whether that meant buying urban nightsoil or (especially in Qing times) purchasing the dregs produced in extracting peanut, soybean, and other vegetable oils.

Agricultural Transformations I: New Crops New crops transformed Chinese agriculture during our period. The first and most important was early-ripening rice, which arrived from Southeast Asia (as its other name, Champa rice, indicates) via Fujian ports; its arrival probably pre-dates our period, but its first certain use in China dates to 488

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

1012. It spread through the Yangzi delta during the eleventh century, although diffusion to the rest of southern China took much longer. Second, both chronologically and in probable importance, came cotton. It, too, arrived via Fujian, in this case from India, probably in the 1200s. The Mongol period (1279–1368) brought new Central and West Asian plants which entered Chinese cuisine – carrots, a type of lime, chickpeas, watermelon – but without great effect; new medicinal plants had greater impact, but still took up little acreage. None, certainly, was remotely as important as Champa rice or cotton. The next great wave of botanical change began in the 1500s, as American crops began arriving – both via maritime Southeast Asia and overland from Burma – and crested two centuries later, along with migration to various highland regions. These crops included tobacco, maize, potatoes, Mexican chili peppers, tomatoes, various squashes, sweet potatoes, and peanuts. (Sweet potatoes are not exclusively American in origin, but reached China together with New World cultivars.) Each new crop offered particular advantages, and each made particular demands on land, labor, and other resources. Early-ripening rice was more drought-resistant than previously available varieties, and would grow on soils where indigenous varieties would not. However, it did not store as well as indigenous varieties, and, having a lower gluten content, was considered less appetizing. Over the course of the Song dynasty, farmers created many hybrid rice varieties, suited to their particular farms, markets, and planting schedules. Planting different rice varieties, or rice and other crops – wheat, strictly a northern crop before about 800, was grown on many lower Yangzi farms by about 1200 – spread out a farmer’s work. This reduced seasonal unemployment and enabled a farmer to manage more land than if all his land had demanded peak labor simultaneously. But initially most farmers still grew only one crop per plot per year, planting different cultivars on different parcels. Crucially, even single-cropped irrigated rice offered far higher yields (in calories per acre) than dry-land crops: over five times greater than wheat, for instance. True multi-cropping, with a given parcel harvested more than once per year, spread more slowly, although it was present in the Yangzi delta by the 1200s, making plausible some reports of extraordinarily high yields per acre. Double-cropping seems to have become widespread in the far south, with its long growing season, before this occurred in the Yangzi valley outside the delta; by modern times some Guangdong farms were triplecropped. 489

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

True multi-cropping allowed for higher yields per acre, and could thereby support denser populations and/or yield larger surpluses. But it also demanded more labor per acre, and more fertilizer to replenish land being so intensely cultivated. Often it also demanded more sophisticated water management; for instance, the rice–wheat double-cropping that became nearly universal in many Yangzi valley communities during the Qing required rapid draining of fields after the rice harvest so wheat could be planted on time. As we shall see, this often became possible only because new ways of organizing irrigation communities allowed for pumping water quickly out of all plots within a polder, rather than just the most favorably located ones. Lastly, on land that was farmed by free tenants – which was most land in particularly commercialized, densely populated, and productive regions, such as the Ming–Qing Yangzi delta – it was customary, from as soon as rice–wheat double-cropping emerged, that rent was based on the principal crop only. Thus double-cropping boosted the tenant’s income more than it increased overall output; a reasonable estimate for the Yangzi delta in the eighteenth century is that adding a wheat crop to a rice field increased a tenant’s output by only one-third but increased his net income by over 60 percent, with income per labor day unchanged.5 It is not surprising, then, that true double-cropping increased as fewer landlords managed their farms directly, leaving cultivators to make most production decisions. Where landlords were more active, they sometimes resisted double-cropping, worrying both about soil depletion and about tenants taking less care with the main crop. Other yield-maximizing practices also spread slowly, making it hard to estimate average yields for any time between late Tang (when they are first recorded) and mid-Qing (when they had become nearly universal). For instance, the maximum yields reported in Wujiang (Jiangsu) as early as the thirteenth century would not be surpassed until modern times, yet average yields there rose perhaps 40 percent between 1300 and 1800. More generally, growth after the Song was overwhelmingly a matter of bringing more land under cultivation and of diffusing best practices, rather than of inventing new practices that raised maximum possible yields on the best plots of land. This was, moreover, by no means unusual: Overton estimates about a 40 percent increase in English yields per acre between 1300 and 1800, also based mostly 5

Kenneth Pomeranz, “Beyond the East–West Binary: Resituating Development Paths in the Eighteenth-Century World,” Journal of Asian Studies 61.2 (2002), 542–3.

490

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

on bringing other farms up to the yields that had already been achieved on a few farms in the thirteenth century.6 Wheat had been grown in China since at least 2000 B C E, though not in great quantities. It expanded significantly beginning sometime around the Tang–Song transition, spreading in ways that paralleled early-ripening rice. In the north, it typically displaced millet and sorghum: it produced higher yields, and was a preferred food. However, it used more labor, and carried more risk. Most Chinese wheat is grown over the winter, when northern China gets very little precipitation. Increased wheat cultivation thus required irrigation, whether using wells or surface water; because this was limited in the north until the 1950s, the advance of wheat was uneven. Wheat may have spread more in the south than the north during the Song, although it never came close to becoming the main cereal there. At first, most northerners growing wheat probably did so on part of their land, while keeping other plots in sorghum and millet; this reduced meteorological risk and spread work across the year, much as the first uses of early-ripening rice did. This strategy would have made particularly good sense during periods when the north was recovering from demographic catastrophes and land–labor ratios were relatively generous: the early Song, the Yuan (after another population decline), and the first half of the Ming. What became the northern form of multi-cropping – in which wheat, sorghum, and beans were rotated on the same plots to yield three crops in two years in most places, and two crops per year in a few – was implemented in the Huai river region (the southernmost part of north China) in the Song, but did not spread all the way north to present-day Hebei until the mid-Qing, when population soared and investment in irrigation grew. Cotton presents a very different story. Philip Huang puts it succinctly: “In 1350, almost no one wore cotton; by 1850, almost every peasant did.”7 Hemp clothing, once extremely common, became rare, though hemp was still grown for rope, bags, and other uses; ramie, which became linen, was still common even in the early 1900s, but it had also retreated in the face of cotton. For roughly its first two centuries, Chinese cotton cultivation was mostly localized in the Yangzi delta, especially on slightly higher ground that was not 6

7

E.C. Ellis and S.M. Wong, “Sustainable Traditional Agriculture in the Tai Lake Region of China,” Agriculture, Ecosystems and Environment 61 (1997), 185; Mark Overton, Agricultural Revolution in England: The Transformation of the Agrarian Economy, 1500–1850 (New York, Cambridge University Press, 1996), p. 77. Philip Huang, The Peasant Family and Rural Development in the Lower Yangzi Region, 1350– 1988 (Stanford, Stanford University Press, 1990), p. 44.

491

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

optimal for rice paddies. But within these areas, its takeover was remarkable; if the early seventeenth-century polymath Xu Guangqi’s estimate is even close, Songjiang prefecture had as much cotton acreage in 1600 as in the 1930s. Most of this cotton was spun and woven locally; the cloth was sold across the empire. But estimates of output are very uncertain. By roughly 1600 another major cotton-growing region had developed in north China, mostly along the Grand Canal in Shandong and Hebei. At first, north China cotton farmers did not spin or weave, instead exporting their raw cotton to the lower Yangzi, and purchasing cloth for their own use. The extremely dry air characteristic of the region during winter (when there was most time for spinning) often caused threads to snap. By 1700, however, people learned that cellars which were partially underground trapped enough humidity for spinning; before long, north China produced most of its own cotton cloth, and exported some to Manchuria, Mongolia, and other markets opened up by the Qing conquests. Northern cotton growing also expanded, however, so raw cotton continued to be sold to the lower Yangzi region. If Yangzi delta cotton output is hard to estimate, that of eighteenth-century north China is harder still. Acreage estimates made by contemporaries – including Fang Guancheng, a governor who wrote the best-known account of north China cotton growing – imply output figures far higher than those for the early twentieth century.8 That is possible, given the intensifying competition between grain and non-grain crops for north China land: the region’s population probably rose about 40 percent between 1750 and 1900, while cultivated acreage, yields, and per capita food supply all seem to have changed little. However, taking Fang Guancheng and other observers literally yields consumption levels that seem implausibly high. A compromise guess is that north China cotton output in 1750 may have been close to that of 1900 (a particularly bad year). Even this would give us an empire-wide average of roughly three kilograms of cotton per person per year, based on just the figures for these two (admittedly dominant) areas; combined with other kinds of cloth, it would place Chinese cloth consumption per capita somewhat above Europe’s at the same time. Production figures for the Yangzi delta would approach those for the United Kingdom (with a large share exported in both cases). Many scholars think this is too high, and prefer to assume that eighteenth-century consumption per capita was at roughly early twentieth-century levels.

8

MHT.

492

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

Meanwhile, cotton production was increasing elsewhere in eighteenthcentury China, and would grow faster in the nineteenth and early twentieth centuries (another reason why lower Yangzi and north China output c. 1750 might have rivaled that of 1900). Eventually, this produced a repetitive pattern of import substitution, in which (for instance), the middle Yangzi began producing its own cloth, reducing imports from the Yangzi delta, and also exporting some cloth further upriver to Sichuan; later, Sichuan repeated the process, displacing middle Yangzi cloth from its markets, and developing (small) export markets still farther west. A few areas began producing cotton cloth without growing the fiber: eighteenth-century Guangdong, for instance, began exchanging sugar for lower Yangzi raw cotton, with which rural women in Guangdong produced cloth both for local use and for export to Vietnam (in exchange for rice imports). Craig Dietrich has estimated that by late Qing between 60 and 80 percent of the empire’s counties produced some cotton cloth.9 Cotton spread for many reasons. It is more comfortable than hemp, and warmer; it is much less expensive than silk. It is also relatively easy to produce: an acre of cotton can produce up to ten times as much usable fiber as an acre of hemp, and it takes much less labor to turn it into cloth. Cotton did require lots of water – such that its cultivation often depended on developing irrigation – but far less than rice paddies. It required more labor per acre than the wheat, beans, and sorghum with which it competed for land in north China, but much of this was labor well suited to women or even children, who were much less productive than men in grain farming. Weaving was skilled work, but it was generally done by women (as was common in the Yangzi delta, where small farms needed little female labor in the fields, except during brief peak seasons) in the winter. At least in the mideighteenth century, the returns per labor day were better than women could earn in lower Yangzi farming, and comparable to what men could earn in north China farming. If spinning – by far the least lucrative part of the production process – could be assigned to children or the elderly, the economics of family cotton production became even more compelling. Diffusion was also facilitated by Confucian social norms – under which it was preferable for women to earn money by indoor cloth making rather than 9

Craig Dietrich, “Cotton Culture and Manufacture in Early Ch’ing China,” in W.E. Wilmott (ed.), Economic Organization in Chinese Society (Stanford, Stanford University Press, 1972), p.111. See also Li, Agricultural Development in Jiangnan, pp. 107–8, and Robert Marks, Tigers, Rice, Silk, and Silt: Environment and Economy in Guangdong, 1250– 1850 (New York, Cambridge University Press, 1997), pp. 171–3, for regional examples.

493

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

working in the fields, visible to non-kin – and by Qing dynasty campaigns (influenced by these norms) that promoted cotton cloth production as the means both to reduce poverty and to cultivate female virtue. Other new plants arrived under the Mongols, including many used in Persian and other Muslim medical traditions. These did not cover much acreage, however. The medical changes themselves affected both human and animal care. Among veterinary treatments, those for treating horses are best documented, since the court took a special interest in them, but there may have been improvements in animal care more generally. The final wave of new crops added to Chinese agriculture in our period came from the Americas, beginning in the sixteenth century. Within this group, tobacco stands out for the speed at which it spread, despite its lack of nutritional value – a story hardly unique to China. Tobacco reached China sometime after 1550, by at least three routes: overland from India and Burma, by sea from Southeast Asia to Fujian and Guangdong, and to Liaodong and north China via Japan and Korea. By 1600, it was extensively grown in Fujian and Guangdong. Over the next few decades it spread through north China – like cotton, it was particularly common near the Grand Canal – and along the commercially navigable Wei river further west. After about 1680, migrants carried the crop into highlands throughout southern China, a pattern which was shared with many other American crops. By the mid-1700s, China probably produced between 1.5 and 2.0 kilograms of tobacco per capita, a figure slightly higher than the amount consumed in England and Wales at that time, and – as with our estimates for cotton – considerably higher than per capita estimates for the early twentieth century. In both highlands and lowlands, the crop was grown on small farms; but because processing the leaf was relatively capital-intensive, many of the farmers were either tenants who depended on their landlords for financing, or smallholders who mortgaged their crop to merchants. The other major American crops were foods, complexly related to both market-oriented and subsistence production. Many of them would grow on land – especially at high elevations – where previously available food crops did poorly. Some cash crops, such as tea and indigo, had powered earlier expansions of highland cultivation, but the cultivators’ food generally had come from lower altitudes. This made the spread of American food crops revolutionary. Among these food crops, the big stories were maize, sweet potatoes, and “Irish” potatoes, probably in that order. All not only grew on some previously uncultivated land, but provided many calories per acre and required 494

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

relatively little labor. Thus they were eagerly adopted by entrepreneurs looking to cheaply feed highland laborers (such as loggers and miners in the Han, Gan, Jinsha, and Yangzi river highlands), by marginalized people seeking new subsistence niches (such as indigenous populations in the southwest forced farther uphill by Han immigrants), and by people hedging their reliance on potentially lucrative but risky cash crops (such as Pearl river delta farmers turning rice paddies into sugarcane fields). The Qing population, very roughly, tripled between 1700 and 1850. Data for cultivated acreage are even rougher, but doubling would represent an optimistic guess. Much of the population growth, and an even larger share of the new farmland, was in interior, upland regions, and this became increasingly true with time: while much early Qing population growth occurred in plains being repopulated after the seventeenth-century disasters (particularly in the middle Yangzi and Sichuan), more and more of it, as time went on, accompanied settlement of highland regions, particularly in the southwest. There, western hemisphere crops played a key role. The sweet potato was present in Fujian by the 1560s, and perhaps earlier. During a famine in 1594, the governor encouraged planting it as an auxiliary food, and it spread rapidly thereafter. It also reached Yunnan by the 1560s, along trade routes from Burma. While it was almost never the principal crop for lowlanders – except among the very poor – the sweet potato made an excellent secondary crop: it required little labor, grew on marginal soils, yielded many calories per acre, stored easily, and was relatively droughttolerant. Sweet potatoes were rarely marketed in these settings, but they were a valuable addition to subsistence security for many cash-crop farmers. Moreover, some people who also grew higher-priced foods sold them, and lived on sweet potatoes instead. Irish potatoes followed a roughly similar trajectory. Peanuts probably arrived in southeast China even earlier, between 1510 and 1550. While never a staple, they made a nutritious dietary supplement, and could also be pressed for oil (leaving dregs that made excellent fertilizer or animal feed). In addition, peanut vines fed hogs and peanut shells fed fires. Peanuts, too, required little labor, and they grew well on sandy soils not otherwise in demand; indeed, by fixing nitrogen, peanut cultivation enriched such soils. Maize probably arrived first overland, reaching Yunnan by about 1530; it spread rapidly through the southwestern highlands, adopted by both indigenous peoples and Han immigrants. (The two groups tended to grow it differently, with immigrant methods increasing yields but proving less 495

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

sustainable.) It also reached the southeast coast no later than 1570, and eventually spread almost everywhere. It was in the highlands that these three crops became central to the food supply. The reason was simple: wheat, rice, and millet usually did not do very well at high elevations. These crops were marketed locally, if at all, but were often tied to commercialization nonetheless. While highland settlement partly reflected crowding in lowland areas, this was not a purely Malthusian story, in which people were pushed uphill by declining living standards. In many cases, it was lowland prosperity that fueled upland settlement, as growing demand for highland products – from construction timber to paper, charcoal, metals (smelted with charcoal), medicinal mushrooms, and exotica – created commercial opportunities. Logging and mining camps could have thousands of people, most of whom had to purchase their food; writing early in the 1800s, Yan Ruyi noted that these enterprises often stopped when food prices rose. Thus having foodstuffs that could be grown locally by men who also produced marketable goods greatly accelerated the exploitation of highland resources. Yan estimated that half the people in the Han river highlands relied on maize or potatoes for food; in Guizhou, the figure may have been 70 percent.10 In the lowlands, these crops had less impact, but still mattered: in Hunan and Hubei – provinces that produced massive rice exports c. 1550–1750, but much less thereafter – maize and white potatoes fed about 6 percent of the nineteenth-century population. Meanwhile, in parts of north and northwest China, maize replaced sorghum, as it fared better on increasingly waterlogged soils.

Agricultural Transformations I I: Water Control The spread of new crops was mostly driven by private individuals, although officials sometimes introduced one to a specific place; probably only in the case of cotton was sustained government effort significant. Changes in fertilizer use, which we will consider later, were almost exclusively private initiatives. The story is very different for water control, the other great motor of agricultural change during our period. Though historians emphatically (and rightly) reject narratives of a state-dominated “hydraulic society,” noting the centrality of private interests and efforts, the state was nonetheless an essential actor in the evolution of water management. 10

SSBF, 8:4b; Robert Jenks, Insurgency and Social Disorder in Guizhou: The “Miao” Rebellion 1854–1873 (Honolulu, University of Hawai‘i Press, 1994), p. 16.

496

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

It may be useful to begin at the end of our story, describing water control c. 1800. Dwight Perkins, assembling mostly early twentieth-century data, estimated that 57–75 percent of land in southern provinces was irrigated (except for 41 percent in very hilly Anhui); most figures are around 65 percent. His southwestern figures are around 33 percent; northern figures are all under 10 percent, except for 14 percent in Henan.11 Given the enormous destruction in the mid-nineteenth-century civil wars, it seems unlikely that these highly approximate figures represent big increases over 1800 levels; they are probably not great underestimations, either. A bigger problem is that we know little about the quality of irrigation that these numbers represent. Still, irrigation was clearly extensive, at least in the south. North and especially northwest China get limited rainfall, heavily concentrated in June through August. Irrigation was thus critical, and became more so as more winter crops were grown. But surface water is also relatively scarce in these regions. Wells were often the best option, but their costs varied wildly with the local water table and terrain: estimates from eighteenth-century officials range from one or two taels (equivalent to about six weeks’ worth of grain for a family of five) to twenty (or even forty) taels for deeper, brick-lined wells.12 Thus families often dug smaller wells, while larger wells usually required a local association, typically based around a temple. Tapping surface water also often involved such groups. The state played little role in north China irrigation during our period. While earlier dynasties had made significant investments in north China irrigation, they had refocused their efforts well before the year 1000. Prior to 750, two-thirds of state-sponsored irrigation projects had been in the north; afterwards, two-thirds were in the south, with almost half in the Jiangnan region alone.13 The Yuan and Ming also left north China irrigation up to local actors. The Qing made serious but sporadic efforts to subsidize and encourage well digging, especially in Shanxi, plus an abortive attempt to create rice

11 12

13

Perkins, Agricultural Development, p. 67. Chen Shuping 陈树平, “Ming Qing shiqi de jingguan” 明清时期的井灌, Zhongguo shehui jingji shi yanjiu 中国社会经济史研究 4.1 (1983), 36; for the figure of forty taels, see Wang Xinjing 王心敬, “Jing li shuo” 井利说, in HCJSWB, 38.4a–5b. During the Tang and Yuan dynasties Jiangnan was an official administrative unit, referred to here. Later “Jiangnan” became a more informal term, referring to roughly the part of the Yangzi delta south of the river – with different writers defining its boundaries slightly differently. All subsequent uses in this chapter refer to the latter sense of the term. “Lower Yangzi” takes in all of the delta plus some adjacent territory with which the delta was functionally integrated, but which was farther from the river and generally less prosperous.

497

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

paddies in some Hebei marshes, but there were few lasting successes until after 1949. In more water-abundant areas, histories of local settlement often reflected the difficulty of controlling water. Sites next to reliable and nonthreatening water sources – for instance, small streams running off gentle hills – were generally farmed first, followed by the rest of the plain. Sites adjacent to larger, more unruly, rivers came later, along with low-lying marshes, portions of lake beds, and other sites requiring more work to drain and protect them. Similarly, farming tended to move gradually up hillsides as terracing, the digging of ponds and canals, and other labors created patches of more or less level and irrigable land. A dramatic case of gradually expanding well-watered fields was the reclamation of “sand fields” (shatian 沙田) along the shores of the Pearl river delta. Lineage organizations placed rocks offshore, trapping river sediment and keeping out saltwater. Once enough silt had been trapped, reeds were planted; they had little economic value, but helped hold the accumulated soil until it sufficed to support actual crops. The process – which could take twenty years – would then be repeated farther beyond the original shoreline. These efforts slowly added vast amounts of prime farmland. In 1300, the city of Guangzhou was on the coast; 600 years later, it was seventy miles inland. Similar efforts transformed Fujian’s coastline, particularly around Xinghua Bay. There, the state played an important early (ninth-century) role, but lineages took over thereafter, and the process essentially reached its limits in the late Ming.14 The Jianghan plain, where the Han river flows into the Yangzi, was also greatly expanded by reclaiming submerged land. In the year 1000, it was a checkerboard of small lakes, rivers, islands, and marshes. The first challenge was to wall off some areas and keep them dry enough to farm. Enclosures, or “polders,” called yuan 垸, are first recorded there in the 1200s, built by soldiers to reduce their dependence on imported food. In Ming times these enclosures were largely taken over by civilian immigrants. Many more yuan were built between roughly 1500 and 1800. Small yuan might enclose a few dozen families’ farms; the largest ones, arable land for over a thousand families. Some yuan were “official” (guan 官) enclosures, but Ming–Qing officials generally supervised them loosely, and contributed money only during crises. “People’s” (min 民) enclosures were registered with the government, 14

Marks, Tigers, Rice, Silk, and Silt, pp. 66–83; Kenneth Dean and Zheng Zhenman, Ritual Alliances of the Putian Plain (Leiden, Brill, 2010), 1, pp. 62–3 and Maps 5–9.

498

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

but almost entirely civilian-managed. There were also dikes along the rivers. At first, these were mostly short sections of embankment, designed to protect a particular enclosure nearby. Over time, however, as more and more land needed protection, dikes came to extend all along the banks of local rivers, and along much of the Han and Yangzi rivers, too. Local government took an increasingly active role in building and maintaining them. Still, overall co-ordination remained elusive. Fundamentally, the Jianghan plain had more water than was consistent with its ever-increasing land surface; protecting one enclosure from that water inevitably increased pressure elsewhere. Lawsuits proliferated, especially in the Qing, as did complaints by officials about residents of neighboring jurisdictions building dikes (with or without official connivance). Eighteenth-century officials particularly complained about “private” (si 私) dikes, which were not officially registered, often because they would not have been approved; such dikes were illegal, but hard to eliminate. Some people took more direct action, sabotaging others’ dikes (whether legal or not) that placed pressure on their own. Over the long haul, then, reclamation there was undermined by its own success. Population density in the Jianghan plain increased 1,200 percent from 1391 to 1820, and during the seventeenth and eighteenth centuries the region was also a major rice exporter; but environmental problems were increasing, and flooding became incessant from roughly 1800 until after 1949. Meanwhile, early Qing officials were horrified to find people in poor, hilly, western Hunan depending solely on rainfall. They encouraged the building of reservoirs, dams, and ditches that would capture water before it all flowed to the plains below, and the installation of pumps (powered by either oxen or people) to raise water where needed. They cut taxes to free resources for irrigation, and partially funded some larger projects; for the most part, though, they hoped that landlords would contribute capital and the tenant cultivators labor, creating a self-sustaining system. While large gains were made in the 1700s, these irrigation systems proved difficult to maintain – less for technical reasons than because local elites often appropriated them, either diverting more than their share of the water or turning reservoir beds into farmland. Cultivators often sued, but rarely prevailed.15 Yangzi delta settlement accelerated in the late Tang, as people fled chaos in the north; the military assisted with much of the early dike construction. 15

Jiayan Zhang, Coping with Calamity: Environmental Change and Peasant Response in Central China (Honolulu, University of Hawai‘i Press, 2014), pp. 33–44, 54–74 (population data on p. 35); Peter Perdue, Exhausting the Earth: State and Peasant in Hunan (Cambridge, MA, Harvard University Asia Center, 1987), pp. 127–30.

499

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

During the Five Dynasties interregnum, a regional state, Wu-Yue, invested heavily in water control facilities. Scholarly opinions vary, however, about whether the biggest gains in reclamation were made under Wu-Yue or under their Song successors – i.e., after the year 1000. On paper, Northern Song achievements were enormous – one study estimates a total area reclaimed in Zhejiang and some adjoining areas that would exceed total farmland in the province today – but such figures probably overstate the actual gains.16 Lake bed reclamation in particular increased flooding elsewhere, and eliminated reservoirs that higher-lying fields could tap during droughts. Thus the region’s maximum possible output soared, but so did the incidence of disasters. Frequent flooding also means that the 35 million mu probably includes significant double counting – with land being reclaimed, submerged by a flood, and then reclaimed again. Major state investments resumed after 1127, when the Southern Song government, having lost the north, needed to maximize lower Yangzi output; they included a series of seawalls, which significantly increased the delta’s hydraulic stability. Yet efforts to rationalize delta water control again often foundered on officials’ inability to control private reclamation. Eventually, the state retreated. By Ming times, delta water control was largely in private hands, with loose government supervision. This was feasible partly because by this time little scope remained for creating additional land here. Settlement was dense, and the shoreline stable – after an 1172 seawall was completed, it barely changed for eight centuries.17 Lake bed reclamation continued, but much more slowly than before, the process having already reached rapidly diminishing returns. Thus, by Ming–Qing times, the Yangzi delta had less need for the large amounts of patient capital and access to coercive power needed for big reclamation projects. Nonetheless, elites – including many degree holders – still played crucial roles in delta water management until at least the 1500s. But during the late Ming, absentee landlordism greatly increased, propelled both by commercial and cultural opportunities in town and by elites seeking security amidst considerable rural violence, including uprisings that largely eliminated the remaining estates which had been farmed by bondservants. By 16

17

Mira Ann Miheilich, “Polders and the Politics of Land Reclamation during the Northern Song Dynasty (960–1126)” (Ph.D. diss., Cornell University, 1979), p. 193, gives figures totaling roughly 35 million acres. This would be over 200 million late imperial mu – an impossibly large number. Even 35 million mu, if that was the intended number, would be larger than the farmed area in Zhejiang today. Huang, Peasant Family, pp. 25–30.

500

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

the time the Qing conquest was completed, few of the town-dwelling degree holders (who at this time still owned a substantial fraction of delta land18) knew who their tenants were; this dominance of arm’s-length relationships was, however, still new enough to be considered anomalous. Commoner landlords were more likely to live in villages, although the richest among them were also often absentees. Along with producing an alarming (to elites) decline in tenant deference and landlord control, the urbanization of landlords might have caused irrigation management crises in some Yangzi delta locations. The Ming system – which in many ways dated back to the Song – was based on polders: low-lying fields enclosed within raised embankments, allowing water to be pumped in or out as needed. Most polders enclosed numerous plots, some farmed by smallholders and some owned by landlords and farmed by their tenants. The fields around the edges, higher than those in the center, not only received irrigation water first; they were also much easier to drain when necessary. They were therefore more valuable than plots nearer the center of the enclosure, and were typically owned by wealthy landlords, who managed and funded most of the infrastructure of which they were the prime beneficiaries. As these landlords withdrew from the countryside, they did less to manage these systems, had less leverage over others, and found that wages alone were not always sufficient to motivate small farmers to maintain polders which did not serve them well. But by mid-Qing times at the latest, the landscape was reconfigured, with more small polders and more internal embankments. This made it possible for co-ordinated efforts by smaller households to manage the system adequately; and, by democratizing access to water (now nobody’s land was hopelessly far from the water source), it both gave everyone an incentive to participate in irrigation work and raised overall production. Among other things, drainage improved significantly. Land at the center of polders, which used to be left under reeds so water could be dumped there when necessary, could now grow rice. Fallowing within the polder – widespread in Song times – was thus largely eliminated by the mid-Qing. With this shift accomplished, villages or clusters of villages, usually grouped around temples, took over most water control, with all full members of the village participating. (This generally included both smallholders and secure tenants.) Landlords, on the other hand, became less essential to production; consequently, there 18

Li Wenzhi 李文治 and Jiang Taixin 江太新, Zhongguo dizhuzhi jingji lun: Fengjian tudi guanxi fazhan yu bianhua 中国地主制经济论 : 封建土地关系发展与变化 (Beijing, Zhongguo shehui kexue chubanshe, 2005), pp. 256–69, 295, 372–83.

501

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

was even less reason for them to remain in the countryside. A mideighteenth-century Suzhou source recorded that only 10–20 percent of landlords lived in villages.19 This is not the place to pursue the full implications of this shift for rural society – or even for property rights (see Kishimoto’s chapter in this volume). It should be noted, however, that the increasingly attenuated role of landlords in agricultural production – or, put otherwise, the growing degree to which cultivators managed production themselves – made it increasingly important for landlords that their tenants have a sufficient stake in the land’s long-run productivity to take good care of it. This gave further impetus to the consolidation of secure cultivation rights in this region; similar patterns are evident in some other relatively rich, commercialized, urbanized, and irrigated areas, from the Pearl river delta to parts of the Fujian coast and (after our period) even some land around Tianjin. And those property rights, as we shall see, profoundly influenced the rural economy more generally.

Agricultural Transformations I I I: Soil Management, Crop Rotations, and Fertilizer Productivity-enhancing developments often intensified demands on the soil. In many regions, the gradual disappearance of unfarmed land – forests, marshes, and so on – eliminated free goods that could either help replenish soils or feed animals whose wastes did so. In some ecologically fragile areas, intensifying farming had disastrous effects: the switch from swiddening to annual cultivation in parts of Yunnan, for instance. But more often, innovative and intensive fertilization protected China’s existing cultivated lands, even as the soil was worked harder; we have few indications of the soil exhaustion and declining yields that plagued some other early modern locales. Most Qing environmental disasters stemmed instead from beginning to cultivate land not previously farmed at all, as on some steep slopes and on thin tropical soils in Lingnan. North and northwest China experienced significant land degradation, though nutrient depletion was rarely the issue. The loess soil layer in northwest China is extremely thick, and so survived considerable erosion; the cropping regimes there were China’s least intensive. North China was somewhat more vulnerable to soil depletion, especially under cotton and 19

Cited in Xue Yong, “Agrarian Urbanization: Social and Economic Changes in Jiangnan from the Eighth to the Nineteenth Centuries” (Ph.D. diss., Yale University, 2006), p. 217.

502

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

tobacco, but intensive fertilization seems to have handled the problems adequately. Various effects of flooding were much more serious problems: waterlogging, salinization and eventually sand deposition along the lower Yellow River, and gullying from flash floods in the severely deforested northwest. Remediation for these problems required interventions well beyond the capacities of individual farmers – or, arguably, even of preindustrial states. A wide variety of fertilizers and techniques for preparing them had been known since ancient times – including pelleting, the labor-intensive yet highly effective technique of packing seeds into balls of nutrient-rich material before planting, which also provided some protection from pests. Unfortunately, we do not know how widely particular techniques were used at particular dates. We can, however, identify some changes during our period, many of them involving new markets for fertilizer. The expansion of irrigated rice required new fertilization techniques – some of which probably arrived with the new rice varieties themselves. Azolla – an aquatic fern which, when it grows floating on the surface of paddies, both fixes nitrogen and suppresses weeds – was known in sixthcentury central China, but probably little used. It was apparently more common in eleventh-century Vietnam, and was likely introduced again to Fujian with Champa rice. By roughly 1600, it was widely used in Fujian and Zhejiang, though not in other rice regions. Applying blue-green algae – which also grow in paddies – to postharvest dried fields is another restorative practice, which seems to have become common by the eighteenth century. Sixth-century authors already knew that waterborne nutrients matter more for irrigated rice than the soil itself; minerals from irrigation water seep into the soil, so that fields often become more productive over time. This works best, however, when stimulated by proper manuring. A variety of fertilizers became available for purchase between roughly the year 1000 and 1600: mollusk shells, river mud, silkworm waste, animal bones, and, eventually, nightsoil and the dregs produced by extracting vegetable oils. Soybean cake in particular was already shipped long distances in the late Ming – from Hunan to Jiangsu, for instance. However, the arrival of Manchurian bean cake in eastern and southeastern China after 1683 represented a significant enlargement of the fertilizer trade. Some Song farmers were already using urban nightsoil as fertilizer, but large-scale trade in urban waste probably began in the late Ming Yangzi delta. An estate owner’s agricultural manual from 1603 discusses traveling significant distances in order to procure top-grade fertilizer from cities with the 503

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

richest diets, rice-milling centers with many oxen, and so on.20 Ordinary family farmers could not be so picky, but the proliferation of towns across the lower Yangzi in late Ming and Qing would have given most farmers there access to some urban nightsoil; merchant boats carried wastes considerable distances from major lower Yangzi cities. Other regions had less-developed networks; a 1737 source criticizes northerners for failing to use urban wastes, but many cities did eventually duplicate this practice. Perhaps the best we can say is that nightsoil distribution networks were growing during the Qing, although Jiangnan’s remained particularly impressive. The use of crushed bean cake was also most pronounced in the Yangzi delta, joined later by the Pearl river delta. It is on this trade, above all, that Li Bozhong bases his idea of a Ming–Qing “fertilizer revolution.”21 That may overstate the transformation, but there was clearly a significant quantitative and qualitative shift. In the Ming, the lower Yangzi’s bean cake purchases mostly came from the middle Yangzi, but after 1683, when the Qing lifted their ban on coastal shipping, massive bean cake shipments began coming from southern Manchuria, over 1,100 kilometers away. By the mid-1700s, Manchurian bean cake was traveling twice that far to fertilize Guangdong’s rice and sugar cane fields. Ecologically speaking, this meant that China’s most intensively farmed regions were importing nutrients from distant frontiers, no longer depending on a closed system that recycled the area’s own biomass. The amount of this relief was substantial, since bean cake was much more nutrient-dense than any manure. Unfortunately, our best source for the size of this trade uses an ambiguous measurement; but even if we assume the smaller unit, and further assume that only 20 percent of imported bean cake was used for fertilizer, it would have provided 20–30 percent as much fertilizer as all the nightsoil generated by the 31 million inhabitants of the lower Yangzi’s core prefectures.22 These developments allowed the Yangzi and Pearl river deltas to sustain high cropping ratios and rising yields without soil depletion.

20

21

22

Xue Yong, “Treasure Nightsoil as if It Were Gold: Economic and Ecological Links between Urban and Rural Areas in Late Imperial Jiangnan,” Late Imperial China 26.1 (2005), 44–52. Li Bozhong 李伯重, Fazhan yu zhiyue: Ming Qing Jiangnan shengchanli yanjiu發展與制 約 : 明淸江南生産力研究 (Taipei, Lianjing chubanshe, 2002), pp. 287–362. For calculations and explanation, see Pomeranz, “Beyond the East–West Binary,” 583– 4. Xue Yong, however, argues for a much lower figure. See Xue Yong, “A ‘Fertilizer Revolution’? A Critical Response to Pomeranz’s Theory of ‘Geographic Luck’,” Modern China 33.1 (2007), 203.

504

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

Moreover, some bean cake clearly substituted for other fertilizers, rather than simply being added to them – which saved large amounts of labor. Bean cake provided as much nutrition as manure weighing roughly 30–50 times more – making it far easier to apply to scattered fields – and involved much less work than properly preparing manures. Its use therefore indicated that many peasants in these regions would spend money to save labor, an important fact to remember in light of debates about returns to rural labor. But by the same token, it is important to note that most other regions used little bean cake. Evidence about north China farmers – who often grew soybeans, but could not afford to use them as fertilizer – is particularly telling. Twentieth-century data from six Shandong counties suggest that farmers there applied very little bean cake, relying overwhelmingly on animal wastes and compost that they gathered and prepared themselves; in one intensively studied county, the few farmers who did use bean cake were ones that had significant incomes from nonagricultural sidelines.23 This reinforces anecdotal data from scattered north China locales suggesting that commercial and handicraft earnings also often provided essential capital for irrigation, in both Qing and Republican times. For those involved, then, nonagricultural income, slightly better-capitalized farming, and higher yields could form a (mildly) virtuous circle, raising incomes even as average farm size shrank; but participation in these incremental gains was sharply limited beyond China’s most commercially developed regions.

The State and the Rural Economy The water control story told above captures important aspects of the state’s role more generally. State involvement in irrigation probably declined after the Song, albeit with occasional reversals. Initially, massive migration to the south created enormous opportunities, but often required more organization and resources than the new arrivals could muster. Second, the intense military competition during the Five Dynasties and Song periods encouraged an activist state, which undertook major developmental efforts and charged exceptionally high taxes in order to fund a massive military. (The Song army was roughly as large in 1040 as were the Qing armies in 1840, with a population perhaps only one-fourth that of 1840.) Moreover, the initial construction of waterworks required large 23

Tang Zhiqing 唐致卿, Jindai Shandong nongcun shehui jingji yanjiu 近代山东农村社会 经济研究 (Beijing, Renmin chubanshe, 2004), pp. 428–33.

505

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

capital investments; ongoing maintenance and operation was more laborintensive and less capital-intensive than original construction, and so potentially could be handled by the expanded local population. In moments of particular vigor, such as during Wang Anshi’s reforms, the government tried to force water control to serve agriculture in general, rather than powerful families’ narrower interests, but it rarely succeeded for long. After all, even at its most vigorous, the state needed local elite cooperation, and when it successfully controlled the biggest flood dangers, the state aided precisely the private interests that would later frustrate it. It is unsurprising, then, that the long-term trend was toward devolution of responsibility to local organizations under the aegis of lineages, temples, and villages. But there is another story of state involvement in water control. If we look at larger-scale projects, especially in poorer areas, we can see both a cyclical pattern within each dynasty – with early ambition and relative success giving way to later failures and frustration at the hands of local interests – and longer-run trends of increasing state efforts, centralization, and a higher priority accorded to subsistence security. This is most evident with respect to Yellow River flood control, where the exceptional scale of the engineering problems and the relative poverty of most communities near the river made state activism essential for even temporary stability, but it is also evident on the potentially wealthy but ecologically vulnerable Jianghan plain. Let us consider the better-known case first. The Northern Song tried to maintain the increasingly silt-laden Yellow River’s southern banks, which protected core imperial interests; they were content, however, to allow flooding of Hebei to the north, which was also claimed by the Liao, and considered of uncertain loyalty. Beginning in the 1030s, the river indeed shifted northwards, culminating in the huge 1048 floods. Most Song officials accepted this as inevitable, although Wang Anshi’s activist regime briefly (and unsuccessfully) attempted more comprehensive flood control. The river was gradually allowed to spread across many channels. Hebei became a waterlogged zone, sacrificed to geopolitical priorities. This situation persisted until 1128, when retreating Song armies deliberately cut the dikes, inducing major floods to slow down (temporarily) the Jurchen cavalry – and ultimately shifted the river’s mouth hundreds of miles southwards.24 24

Ling Zhang, The River, the Plain, and the State: An Environmental Drama in Northern Song China, 1048–1128 (Cambridge, Cambridge University Press, 2016), pp. 122–61.

506

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

Two noteworthy premises guided these policies. First, military considerations dominated Yellow River policy; agriculture mattered mostly where it was fiscally important. Second, the river was considered ultimately uncontrollable – and certainly impossible to keep within narrow banks. Two other dramatic riverbed shifts within the next two centuries (in 1194 and 1288) seemingly confirmed this. Meanwhile, relocating China’s capital to Beijing – in a region unable to feed such a large city reliably, plus defenders posted further north – changed the state’s calculations. Stability demanded massive grain imports from the south, which the Ming decided (after a brief experiment with seaborne transport) should be shipped on the Grand Canal. Given how the canal and the Yellow River now intersected, this required intensified efforts at Yellow River control. Moreover, if north China experienced too much flooding, southern rice shipments alone would not guarantee security. Beginning in the late 1400s, central government officials strove to confine the river to a single channel; in 1565, Yellow River commissioner Pan Jixun began further systematizing these efforts. Gradually, a larger river control system, with more stable staffing, took shape, reflecting Pan’s vision. The early Qing further strengthened and centralized the Yellow River Administration; it became larger, more able to impose co-ordination on provincial and local officials, and increasingly staffed by men with technical expertise. Not coincidentally, the eighteenth century was probably the most flood-free in the river’s entire history prior to the application of modern engineering after 1949. North China peasants also benefited from a Grand Canal increasingly available to private traders; this facilitated agricultural commercialization, such as increased planting of cotton and tobacco. While maintaining the system became increasingly expensive as the Yellow River’s bed rose, those costs were mostly passed along to taxpayers in the more prosperous Yangzi valley. The Huai river valley suffered hydrologically, but this damage was not enough to offset the gains experienced by residents of the Yellow River floodplain. In the middle Yangzi region, too, the state gradually became more consistently involved in river control, and more focused on protecting agriculture. Its story thereby roughly paralleled both the cyclical and linear dynamics evident in the history of Yellow River management, although with many specific differences. The Jianghan dikes built during the Southern Song, under military direction, aimed above all at creating a watery barrier to invading cavalry. Although they were also supposed to facilitate farming to feed Song troops, 507

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

the overriding purpose was military, not economic. The resulting farms were large, were claimed by powerful elites, and were not farmed intensively, producing very low yields. The Ming no longer needed defense works there. They instead made a vigorous effort, utilizing corvée labor, to create farmland. However, the farmers who settled behind large government-built river dikes quickly began building enclosures that undermined overall regional flood control. Beijing periodically sent flood control commissioners; however, they often prioritized nonagricultural interests, such as the geomancy of tombs important to capital elites. Undermined by private overdevelopment and by sporadic, often counterproductive, state intervention, Jianghan water control collapsed shortly before the dynasty itself. A new cycle began under the Qing; again the state built the basic infrastructure, and private development followed. This time, however, central funding was more generous, and officials provided more sustained and systematic supervision for the dikes. Responsibility was assigned to the regular bureaucracy, rather than occasional special commissioners, and agriculture was consistently prioritized. When possible, the state wanted local communities to provide most of the funding and labor for water control, and ceded everyday management authority to them. But the government was certainly more involved here than in the lower Yangzi, where state supervision remained quite loose. The achievements – agriculture that fed a vastly expanded population, and also yielded a large tradable surplus – should not be minimized. Nonetheless, private overdevelopment once again came to undermine flood control. By 1800, the stage was set for frequent, and serious, flooding, which plagued the region until after 1949.25 The Ming–Qing flood control story sketched above – with the Qing doing more than the Ming, especially in poorer regions – parallels Qing efforts (though those were sporadic, as we saw) to foster irrigation in the north. It was also echoed in the late imperial state’s other major “safety net” initiative: granaries. The Song had maintained government-run granaries, and for a few decades used them for as-needed famine relief. However, their stocks were more often diverted to soldiers, with responsibility for relief being delegated to 25

Pierre-Étienne Will, “State Intervention in the Administration of a Hydraulic Infrastructure: The Example of Hubei Province in Late Imperial Times,” in Stuart Schram (ed.), The Scope of State Power in China (New York, St. Martin’s Press, 1985), pp. 308–44; Pierre-Étienne Will, “Une cycle hydraulique en Chine: La province du Hubei du 16e au 19e siècles,” Bulletin de l’École française d’extrême orient 68.1 (1980), 280–8.

508

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

local elites. The early Ming rulers, who hoped to freeze a stable rural society in place, certainly cared about rural subsistence, and were hardly averse to state intervention. But by the mid-fifteenth century the market was returning, both in fact and (more slowly) as a statecraft ideal; despite some centralized efforts, grain storage was mostly left in local, private hands. The Qing granary system was far more ambitious. It included government-run “ever-normal granaries” in every county seat, charity granaries in major towns, and rural community granaries. Officials ran only the evernormal granaries, but were supposed to oversee all three kinds. Ever-normal granaries not only held reserves for emergency use, but acquired and released grain regularly, to smooth out price fluctuations across the annual harvest cycle and from year to year. Grain could be disbursed through loans, sales at below-market prices, or (when necessary) grants. Granaries were encouraged to stockpile food, but many held a significant percentage of their reserves in silver (which does not spoil) – particularly in downstream, highly commercialized areas that could easily import grain. Stocks were assembled though voluntary contributions (often made in return for examination degrees), purchases during cyclical periods of low prices (such as immediately after harvests), and diversions of tribute grain shipped up the Grand Canal toward Beijing. Annual tribute, when fully collected, was far more than the capital needed. While acquisitions through all these mechanisms fluctuated, diversions generally increased during 1660–1760, making the central government’s role more and more critical. Community granaries were increasingly emphasized, relative to ever-normal granaries, after about 1760, but officials supervised them quite actively until a 1799 edict curtailed this oversight. State involvement in subsistence security thus remained strong throughout the eighteenth century. State expenditures on granaries – probably slightly over 5 percent of all non-military spending not dedicated to basic administrative costs – were significant by premodern standards. So were the effects. During disasters, the Qing could mobilize remarkable quantities of grain, greater than many twentieth-century governments; thus even the worst harvest failures led to much smaller grain price increases than in, say, early modern France. Beyond such emergency activity, an active year of price-smoothing efforts might involve from 2 to 10 percent of a provincial population living for about two months on cheap food loaned or sold by granaries. Quite deliberately, the system did the most in less commercialized, relatively inaccessible regions, where markets were least able to relieve harvest fluctuations; it remained effective in Yunnan and Guizhou, which had many such locales, long after it had declined greatly in wealthier regions. 509

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

Moreover, the granaries focused on rural areas at a time when European relief systems were largely focused on containing urban crises. Overall, the Qing arguably did less to develop the rural economy than their Song predecessors, but they had become fairly adept at stabilizing it.26

Rural Handicrafts The countryside produced more than crops, providing most of its own cloth, wood, and metalworking, and most other goods and services. These trades continued expanding and diversifying, though unevenly in space and time, right up to 1800 (and in some cases beyond). They provided the entire livelihood of some rural people, but it was far more common for families to supplement farming with market-oriented sidelines. Unfortunately, we know very little about most of these activities; research has focused heavily on textiles. Rural industry was predominant in many parts of the world before factorybased industrialization, but perhaps especially so in China. China did not have the government-backed guilds that claimed urban monopolies on various European crafts; relative peace after 1400 may also have given Chinese industry less reason to locate behind urban walls. Probably more importantly, China’s property system incentivized rural proto-industrialization. Most farmers in less commercialized areas owned land (although they might owe money against it); this became even more true with time, since bound labor continued declining, estates using wage labor generally failed to take hold, and renting out land was not especially lucrative in poorer regions. Richer areas were more evenly divided between smallholding and small-scale tenancy. Irrigated rice offers few economies of scale; it rewards careful labor, for which neither wage earners nor bound laborers have much incentive. Most Song-era tenants had very little security. Beginning in the Yuan, however, we see southeastern and lower Yangzi tenants beginning to gain secure cultivation rights. Such rights became very common by the Wanli (1572–1620) period, and by Qing times were themselves a form of heritable, tradable property. In more commercialized areas of south China, most tenants probably had some form of long-run guarantee.27 26

27

Pierre-Étienne Will and R. Bin Wong, with James Lee, Nourish the People: The State Civilian Granary System in China, 1650–1850 (Ann Arbor, University of Michigan Press, 1991); Pierre-Étienne Will, Bureaucracy and Famine in Eighteenth Century China (Stanford, Stanford University Press, 1990). Zhou Yuanlian 周远廉 and Xie Zhaohua 谢肇华, Qingdai zudianzhi yanjiu 清代租佃制研 究 (Shenyang, Liaoning renmin chubanshe, 1986), pp. 290–2, 299–301; Evelyn Rawski, Agricultural Change and the Peasant Economy of South China (Cambridge, MA, Harvard

510

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

Tenants possessing such rights earned much more than the marginal return on labor that either a landless laborer or a tenant at will was likely to earn. Robert Allen has estimated that the earnings of a secure tenant in the Yangzi delta with an average-size (ten-mu) farm c. 1820 were roughly triple what a year-round agricultural laborer earned; I have obtained similar results using slightly different data and assumptions.28 And since urban unskilled wages only slightly exceeded rural ones – as we would expect without strong guilds – even a tenant farming a below-average plot had no reason to leave in search of urban wages. For rural people who did migrate, land frontiers made better destinations than cities; average incomes there might be lower, but reclaiming land often resulted in gaining long-term tenurial rights, and thus in incomes approaching the regional average, rather than a mere fraction of some city’s higher mean. Urban population growth continued, but the urban population percentage fell (see Cao’s chapter in this volume). High-end crafts, and many services, concentrated in towns. But urban growth was far slower than a large rural surplus and shrinking average farm sizes would predict; and in the richest region of all, Jiangnan, urban growth after the 1200s mostly reflected the proliferation of small towns, with the biggest cities barely growing. Thus a great deal of handicraft labor remained in the countryside, embedded in households that also continued farming while enjoying relatively easy access to markets. It is possible that the predominantly rural, part-time nature of much craft production inhibited certain kinds of technical breakthrough. A rural setting meant somewhat lower labor costs and higher capital costs than in cities, and part-time producers had less incentive to pursue innovations in general. It does appear that technical innovation in textiles slowed down after about 1300. But this is highly speculative. Unfortunately, output levels in most handicrafts are unknowable; gazetteers list local products, but few tell us much more. Only for cotton and silk textiles can we make tentative estimates, and even in those cases scholars are far apart. One of the few things that all agree on is that the percentage of nonluxury handicrafts produced in the countryside remained high.

28

University Press, 1972), pp. 19–24, 86–7; Huang, Peasant Family, pp. 25–30; Li and Jiang, Dizhu zhi, pp. 291–4. See also Kishimoto’s chapter in this volume. Robert C. Allen, “Agricultural Productivity and Rural Incomes in England and the Yangzi Delta, c. 1620–1820,” Economic History Review 62.3 (2009), 525–50; Kenneth Pomeranz, “Standards of Living in Rural and Urban China” (paper delivered at World Economic History Congress, Helsinki, 2006), available at www.helsinki.fi/ieh c2006/papers2/Pomeranz.pdf.

511

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

Earnings estimates are also controversial, although they have recently converged significantly. Part of the controversy has stemmed from the fact that cotton weaving earned much more than spinning, but could not occur without it. Spinning, meanwhile, could be done by children or elderly people with few other income-earning possibilities, while most weavers were women of prime working age. Moreover, sixteenth- and seventeenthcentury cloth prices fluctuated wildly relative to grain. But the eighteenth century was more stable, and we can generate conservative estimates by assuming that weavers also did their own spinning and preparatory work. Under these assumptions, a Yangzi delta textile producer would have earned fifty to sixty cash a day, which matches figures from two biographies of mid-century “virtuous widows” who supported themselves by such labor.29 This amount of income was less than delta men with secure cultivation rights earned from farming, but more than women (with much lower agricultural labor productivity) would have earned, and well above unskilled wages for either sex. In north China, where (as we will see) farmers earned much less than in the Yangzi delta, textile earnings per day were competitive with male earning power – and men in this region often did weave, especially in the off-season. Data for sericulture and rural silk weaving – almost exclusively women’s work – are even spottier than for cotton, but nineteenth- and early twentiethcentury data suggest similar dynamics: women in sericulture earned more than they would have in farming, although generally less than men.30 Rural handicrafts, then, generally represented a positive development for those involved, and for the economy. The problem was not that too many people were “driven” into non-farm work, but that too few people were drawn into it.

Living Standards, Subsistence Security, and Economic Development We can assemble the various trends above into a rough picture of rural Chinese living standards over the eight centuries covered here, but that 29

30

Kenneth Pomeranz, The Great Divergence: China, Europe, and the Making of a Modern World Economy (Princeton, Princeton University Press, 2000), p. 319; Pomeranz, “Beyond the East–West Binary,” 558–61. Zhang Li 张丽, Feipinghenghua yu bupingheng: Cong Wuxi jindai nongcun jingji fazhan kan Zhongguo jindai nongcun jingjide zhuanxing (1840–1949) 非平衡化与不平衡 : 从无锡近 代农村经济发展看中国近代农村经济的转型 (Beijing, Zhonghua shuju, 2010), pp. 38–40, 62–154.

512

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

picture is incomplete and very tentative. A broad consensus holds that living standards in the Song were relatively high, especially in the lower Yangzi, but later declined significantly – though opinion varies about whether that decline stemmed from thirteenth-century nomadic invasions, the antimarket policies of the early Ming, or other factors. Recovery began sometime in the Ming; was interrupted by the wars, epidemics, and harvest failures of the mid1600s; and then resumed in the 1700s, though scholars disagree on whether eighteenth-century per capita incomes surpassed peak Song levels. Opinions also differ about whether the rural economy was or was not facing a sustainability crisis by the late 1700s. Breaking these questions down by region adds both clarity and complexity. In their contribution to this volume, Zhiwu Chen and Kaixiang Peng argue that mid-Qing living standards were probably not radically different from those of the Northern Song, seven centuries earlier. They note that while per capita grain output probably declined over this period, it remained well above subsistence levels; they also point to a proliferation of institutions that arguably improved subsistence security, and to modest increases in consumption of other goods. My own estimates are broadly consistent with these conclusions, although they are based on different evidence and perhaps are slightly more positive about overall trends. But if there is some emerging consensus on living standards at the end points of our period, there is less agreement on fluctuations during the centuries in between, and on how mid-Qing living standards might compare with those elsewhere in the eighteenth-century world. William Guanglin Liu has argued for an especially sharp drop in per capita GDP and income between c. 1100 and c. 1500. He relies heavily on soldiers’ wages, a tricky source given both the above-mentioned problems with inferring general income levels from wages and additional questions about what government wages reflect. Skipping over the late Ming and early Qing (and thus the last part of our period), he finds his next comparable data point at c. 1880. By then, he argues, per capita income had roughly recovered to Southern Song levels, but no more than that.31 If, however, living standards declined somewhat between the 1750s and 1880s (as suggested by declining per capita availability of cotton, tobacco, sugar, and so on), it would suggest that living standards during the eighteenth-century peak of prosperity probably exceeded the Southern Song peak, although perhaps not by much. 31

William Guanglin Liu, The Chinese Market Economy 1000–1500 (Albany, SUNY Press, 2015), pp. 123–7, 134.

513

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

Alternative approaches involve either estimating output of key commodities and dividing by population, doing a full GDP estimate for some particularly well-documented time and place, or reconstructing costs and income on “typical” farms to derive “average” incomes for some time and place. All these approaches have problems, but they all yield estimates not too distant from those already cited for the mid-Qing, increasing our confidence. For the Yangzi delta region, there is general – though not unanimous – agreement on average farm size in late Ming and mid-Qing. There is also a reasonable degree of consensus on production costs, yields, and prices in the mid-eighteenth and early nineteenth centuries, and a manageable range of disagreement on labor requirements; the most strongly contested variables are the extent of double-cropping at particular dates and some details of fertilizer use. For the late Ming, there is more uncertainty about relative prices (which fluctuated much more than in the Qing) and about how typical certain intensive cultivation practices were, but, given certain assumptions, we can generate tentative income estimates comparable to other periods. Kent Deng and Patrick O’Brien have assembled several of these estimates, and adjusted them to be comparable to each other, and to estimates of British farm incomes at the same dates. These results are helpful, although they have almost certainly understated eighteenth-century farm size; and, even once that has been adjusted, their typical mid-eighteenth-century household works 23 percent fewer days of labor per year than either seventeenth- or nineteenth-century ones.32 This seems unlikely, but I will leave the work-days figure unadjusted here, rather than making guesses about how much income the missing labor days would have yielded. The results for Jiangnan household income, then, are given in Table 13.1. These figures assume a tenant paying rent but not land taxes, and bearing most production expenses. A good guess is that an owner–farmer with the same-sized plot would have earned about 40 percent more than if he were a tenant, and his household (including, in both cases, a wife producing textiles, with the same output whether the family land was owned or rented) slightly over 25 percent more. Overall, there were probably about as many 32

Kent Deng and Patrick O’Brien, “How Well Did Facts Travel to Support Protracted Debate on the History of the Great Divergence between Western Europe and Imperial China?” (MPRA paper no. 77276), posted 5 March 2017, at https://mpra.ub.unimuenchen.de/77276; Kenneth Pomeranz, “The Data We Have vs. the Data We Want: A Comment on the Date of the Divergence Debate,” parts 1 and 2, New Economics Papers, posted 6 June 2017, at https://nephist.wordpress.com/2017/06/06/ the-data-we-have-vs-the-data-we-need-a-comment-on-the-state-of-the-divergencedebate-part-ii.

514

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

Table 13.1 Estimates of household income for an average Yangzi delta tenant family

1620 1750 1820

Li

Allen

17.6

18.5

18.2

15.1/21.0a

Pomeranz 19.8

a

Based on two different sets of assumptions. Sources: Robert C. Allen, “Agricultural Productivity and Rural Incomes in England and the Yangzi Delta, c. 1620– 1820,” Economic History Review 62.3 (2009), 535–9; Bozhong Li, Agricultural Development in Jiangnan: 1620–1850 (New York, St. Martin’s Press, 1998), pp. 23, 111, 124–53; Kenneth Pomeranz, The Great Divergence: China, Europe, and the Making of the Modern World Economy (Princeton, Princeton University Press, 2000), pp. 316–26

tenants as smallholders in southern China as a whole, although tenants were far more numerous in the most commercialized districts. The incomes of tenant farmers would, of course, tend to be lower than GDP estimates for the same region. Li Bozhong, however, has estimated GDP for two wealthy Jiangnan counties in the 1820s.33 These were relatively urban counties, but interestingly Li’s data do not show significantly higher incomes per capita for urban people. In fact, he estimates that urban workers – averaging skilled and unskilled laborers together – earned less than farmers, although a small number of town-dwellers made much higher incomes than any rural residents. Li’s GDP estimate converts to slightly under 10 shi of rice per capita, which is about 33 percent above Liu’s (probably somewhat high) empire-wide estimate for the Song. If, as Debin Ma has (very roughly) estimated, lower Yangzi income per capita was around 50 percent higher than the empire-wide average prior to industrialization,34 this would suggest that income per capita in the Yangzi delta in the 1820s was not too far from where it had been seven centuries earlier, a finding consistent with the Peng–Chen survey of other welfare indicators. 33

34

Li Bozhong 李伯重, Zhongguo de zaoqi jindai jingji: 1820 niandai Huating– Louxian diqu GDP yanjiu 中国的早期近代经济: 1820 年代华亭–娄县地区 GDP 研究 (Beijing, Zhonghua shuju, 2010), pp. 223, 232, 247–51, 261, 278. Debin Ma, “Modern Economic Growth in the Lower Yangzi in 1911–1937: A Quantitative, Historical, and Institutional Analysis” (Discussion Paper 2004-06-002, Foundation for Advanced Studies on International Development, Tokyo), 6.

515

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

However, Li also points out that the 1820s were a particularly bad decade in the lower Yangzi region: freakishly cold weather, plus the worst floods in many decades, caused grain output to plummet, while the ongoing squeeze on Jiangnan’s cotton textile exports (driven by the import substitution in other regions discussed above) caused cloth prices to fall sharply. Overall, Li suggests that per capita incomes in his counties had probably been 30 percent higher in the eighteenth century than they were in the 1820s. This would suggest – though based on what are by now a lot of assumptions – that the mid-Qing income peak had been around 30 percent above the Song one. Moreover, Li has estimated annual per capita food consumption for these two Jiangnan counties at 2,664 kilocalories and seventy-eight grams of protein per day in the 1820s. These figures compare favorably with Chinese and global average figures for the second half of the twentieth century. In sum, it is clear that rural people in the lower Yangzi were relatively well off by preindustrial standards, but their situation was not improving much over time, and by 1800 at the latest it was probably deteriorating. It seems likely that at least per capita GDP figures fell behind the wealthiest parts of Europe (Britain and the Netherlands) during the 1700s, although exactly when and how far is less clear. Importantly, this gap seems to have been entirely the result of differences in the urban sector. Robert Allen estimates that on a scale where the English agricultural heartland is given a labor productivity value of 100 (and England as a whole between 5 and 10 percent lower), Yangzi delta agricultural labor productivity c. 1820 would be 85, and output per acre almost 800; income for a typical tenant family with a wife producing textiles comes out within 5 percent of that for an English farm laborer whose earnings were supplemented by those of a wife and child. This almost certainly indicates a lead for the delta in 1750 even vis-à-vis England, and a huge lead over most of Europe. However, it is also noteworthy that Allen sees a small decline in the Yangzi delta’s agricultural labor productivity relative to 1620, and a sharp decline in income (partly due to smaller farms).35 North China presents a considerably more grim picture. It suffered huge population losses during the thirteenth and fourteenth centuries; even in the late sixteenth century, the population of north China was probably still well below where it had been c. 1200. Average farm size therefore increased, and cultivation techniques did not undergo anything like the intensification found in the south, except in a few localities. It was only in the Qing, as we saw 35

Allen, “Agricultural Productivity,” 528, 541, 546, 548.

516

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

earlier, that more intensive crop rotations can be confirmed for most of north China. In terms of “development,” then, the north fell further and further behind southern and eastern China; in terms of per capita income, the gap was smaller, but probably considerable nonetheless. Working backwards from a variety of early twentieth-century data, I have elsewhere estimated the income of an average north China tenant at only slightly above that of an agricultural wage laborer: a logical outcome, since, unlike in the south, most tenants were tenants at will, with very little bargaining power.36 With rents averaging around 50 percent of the crop, northern owner–cultivators would have earned somewhat less than double what a tenant on the same-sized plot earned (since owners paid taxes and all of their own production expenses) – in other words, perhaps two-thirds of the agricultural income of a Jiangnan tenant with secure cultivation rights. (The variation around the mean of farm sizes seems to have been much larger in north China than in the lower Yangzi, further vitiating the value of these already tentative comparisons, but given space limitations here, averages will have to do.) This is roughly consistent with what we know about differences between these regions in the better-documented twentieth century. Most of the rest of the south was also probably better off than the empire-wide average – if less so than the Yangzi delta – leaving north and especially northwest China below average, along with the far southwest. Average rural incomes from nonagricultural sources in eighteenth-century north China also would also have lagged far behind those in the lower Yangzi, and probably most other southern regions; they certainly did in the early twentieth century. Evidence from eighteenth-century gazetteers in Hebei suggests a very frugal diet, composed mostly of coarse grains with very little meat, and often only two meals a day; the exceptions were some areas with significant handicraft income and areas near Beijing and Tianjin (where late eighteenth-century foreign visitors reported a much more varied diet, including plenty of meat).37 Shandong was, on average, probably a bit more prosperous than most of Hebei, but not much. Moreover, mid-eighteenth-century north China incomes, low as they were, were probably higher than incomes in much of the same region during the early 1900s. By the 1930s, north China’s population was perhaps 80 percent above 1750 levels, while land under cultivation increased very little (if we are to believe official data, in fact, not at all), and there is little evidence of rising 36 37

Pomeranz, “Standards of Living.” Lillian Li, Fighting Famine in North China: State, Market, and Environmental Decline, 1690s– 1990s (Stanford, Stanford University Press, 2007), pp. 106–9.

517

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

crop yields. Cropping intensities continued to increase, but not enough to make up for these other factors. Evidence for the late nineteenth and early twentieth centuries suggests that life in north China was becoming more precarious, except for those who obtained rising nonagricultural incomes; this seems likely to have also been true over the longer haul since the midQing. J.L. Buck’s survey from c. 1930 suggests that a typical north China wheat farm provided “labor returns” that were only 40 percent of those in his east China region – only 34 percent if we omit Anhui from East China (to which it doesn’t fully belong).38 The way that Buck calculated these labor returns makes it impossible to compare them directly with our earnings estimates here, and of course the early twentieth century was different from the eighteenth. Still, there is no question that there was a huge regional gap even at the earlier date; it may have narrowed in the nineteenth century, when the Taiping Rebellion hit the lower Yangzi especially hard, but would have widened again afterwards. Moreover, in parts of north China where farmers adapting to increasingly waterlogged soil replaced sorghum or millet with maize, the new crop provided more calories, but less protein. (This depletion of protein might not have mattered to highland swiddeners who were among the first adopters of maize, since many of them also hunted, but was more serious for north China’s full-time farmers.) Nonagricultural incomes for rural northerners also lagged far behind those in more accessible, commercialized areas.

Was a Malthusian Crisis Looming? Thus the rural economy had both dynamic, prosperous regions and much more troubled ones in the eighteenth century. In frontier regions, especially in the southwest, population growth was very rapid, and the cultivation of new lands often had serious environmental consequences, increasing erosion and downstream flooding. Unsurprisingly, much less new land was being brought under cultivation in long-settled lowlands; but where this was happening, it often involved reclaiming lake beds or marshes, eliminating outlets for floodwaters, and thus also raising environmental risk. In a few places, such pressures were already causing acute problems by the late eighteenth century; in many others, this would not happen until well into the nineteenth. That the Qing bureaucracy did not grow significantly, while 38

John L. Buck, Chinese Farm Economy: A Study of 2866 Farms in Seven Localities and Seven Provinces in China (Chicago, The University of Chicago Press, 1930), p. 114.

518

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy

population soared and the cultivated area grew substantially, worsened the already poor prospects for successfully managing these problems. As we have seen, rural communities in richer areas sometimes could take over successfully from a retreating state; but in poorer regions, where stabilization was most urgent, such success was unlikely. That said, average farm yields continued to rise, and there is no sign of a classic Malthusian crisis except, perhaps, much later. The evidence on life expectancy is mixed, but in general it is comparable to figures for Europe at the same time, and there is no obvious downward trend.39 Average per capita food supply remained adequate; and for much of the eighteenth century, the government and other institutions made impressive efforts (by premodern standards) to prevent crises from causing actual starvation. Xia Mingfang, for instance, has shown that famine deaths were relatively rare in the first half of the Qing, before soaring after c. 1865.40 Long-distance grain markets were by far the largest in the world (by volume), and up until the middle of the 1700s, they were also impressively well integrated, matching or exceeding any in continental Europe. In the most densely populated areas, a combination of slowly but steadily rising yields and grain accessed via these markets helped people not only survive, but remain prosperous, by both Chinese and global standards. But the fragility of these achievements is, in retrospect, evident. We have already seen that the textile exports and primary-product imports that sustained high living standards in the Yangzi delta had begun to shrink as interior regions filled up, developed their own textile production, and exported less grain. A recent study suggests that while China’s long-distance markets were indeed well integrated in the early and middle eighteenth century, they were becoming less so in the second half of the century – first in the north, but also in the south by the century’s end.41 This is to be expected if dry-farming areas – where raising per-acre yields to match population growth was much harder than in paddy rice regions – were increasingly concentrating on food production, 39

40

41

See the chapter by Chen and Peng in this volume; also William Lavely and R. Bin Wong, “Revising the Malthusian Narrative: The Comparative Study of Population Dynamics in Late Imperial China,” Journal of Asian Studies, 57.3 (1998), 714–48. Xia Mingfang 夏明方, Minguo shiqi ziran zaihai yu xiangcun shehui 民国时期自然灾害 与乡村社会 (Beijing, Zhonghua shuju, 2000), pp. 78–9, 400–2. Daniel Bernhofen, Markus Eberhardt, Jianan Lin, and Stephen Morgan, “Assessing Market (Dis)integration in Early Modern China and Europe” (Center for Economic Policy Research discussion paper, 2016), at https://cepr.org/active/publications/dis cussion_papers/dp.php?dpno=11288Discussion paper. For a more positive assessment, see Carol Shiue and Wolfgang Keller, “Markets in China and Europe on the Eve of the Industrial Revolution,” American Economic Review 97.4 (2007), 1189–1216.

519

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

kenneth pomeranz

reversing their early Qing extension of cash-cropping. (The well-documented deterioration of the Grand Canal must have exacerbated these trends.) In areas such as the middle Yangzi, where (import-substituting) handicrafts were increasing, local and regional markets might have continued to grow, but longdistance trade probably shrank. And this, as we have seen, also put downward pressure on earnings from handicrafts in China’s richest – and most tradedependent – areas, contributing to stagnation, if not yet creating a crisis. Without major breakthroughs in the urban sector that would have allowed it to absorb labor far more productively – much less provide truly transformative new technologies – the redistribution of rural industry across China’s landscape could do only so much to compensate for increasing demographic pressure – pressure which was, in many ways, an unsurprising result of the rural economy’s long-running dynamism. Overall, the combination of a narrower environmental margin of safety, a less effective state, and deteriorating markets made for an increasingly perilous situation, in spite of centuries of impressive growth in the rural economy. With the addition of a negative climatic turn in the nineteenth century, plus foreign battering of an already weakening state, these crises became particularly acute.

Further Reading Anderson, Eugene, Food and Environment in Early and Medieval China (Philadelphia, University of Pennsylvania Press, 2014). Bray, Francesca, Science and Civilization in China, vol. 6, part 2, Agriculture (Cambridge, Cambridge University Press, 1984). Elvin, Mark, The Pattern of the Chinese Past (Stanford, Stanford University Press, 1973). Hamashima Atsutoshi 浜島敦俊, Mindai Ko¯nan no¯son shakai no kenkyu¯ 明代江南農村社 会の 研究 (Tokyo, To¯kyo¯ daigaku shuppankai, 1982). Ho, Ping-ti, “The Introduction of American Food Plants into China,” American Anthropologist, new series 57.2 (1955), 191–201. Huang, Philip, The Peasant Economy and Social Change in North China (Stanford, Stanford University Press, 1985). Huang, Philip, The Peasant Family and Rural Development in the Lower Yangzi Region, 1350– 1988 (Stanford, Stanford University Press, 1990). Li, Bozhong, Agricultural Development in Jiangnan: 1620–1850 (New York, St. Martin’s Press, 1998). Li Bozhong 李伯重, Fazhan yu zhiyue: Ming Qing Jiangnan shengchanli yanjiu 發展與制約 : 明淸江南生産力研究 (Taipei, Lianjing chubanshe, 2002). Li, Lillian, Fighting Famine in North China: State, Market, and Environmental Decline, 1690s– 1990s (Stanford, Stanford University Press, 2007).

520

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

The Rural Economy Li Wenzhi 李文治 and Jiang Taixin 江太新, Zhongguo dizhu zhi jingji lun: Fengjian tudi guanxi fazhan yu bianhua 中国地主制经济论 : 封建土地关系发展与变化 (Beijing, Zhongguo shehui kexue chubanshe, 2005). Marks, Robert, Tigers, Rice, Silk, and Silt: Environment and Economy in Guangdong, 1250–1850 (New York, Cambridge University Press, 1997). O¯sawa Masaaki 大沢正昭, To¯ So¯ henkakuki no¯gyo¯ shakai shi kenkyu¯ 唐宋變革期農業社會 史研究 (Tokyo, Kyu¯ko shoin, 1996). Perdue, Peter, Exhausting the Earth: State and Peasant in Hunan 1500–1850 (Cambridge, MA, Harvard University Asia Center, 1987). Perkins, Dwight H., Agricultural Development in China, 1368–1968 (Chicago, Aldine Publishing, 1969). Pomeranz, Kenneth, “Beyond the East–West Dichotomy: Resituating Development Paths in the Eighteenth-Century World,” Journal of Asian Studies 61.2 (2002), 539–90. Pomeranz, Kenneth, The Great Divergence: China, Europe, and the Making of the Modern World Economy (Princeton, Princeton University Press, 2000). Sudo¯ Yoshiyuki 周藤吉之, So¯dai keizai shi kenkyu¯ 宋代経済史研究 (Tokyo, To¯kyo¯ daigaku shuppankai, 1962). von Glahn, Richard, The Economic History of China: From Antiquity to the Nineteenth Century (Cambridge, Cambridge University Press, 2016). Watabe Tadayo 渡部忠世 and Sakurai Yumio 桜井由躬雄, Chu¯goku Ko¯nan no inasaku bunka: Sono gakusaiteki kenkyu¯ 中国江南の稲作文化 : その学際的研究 (Tokyo, Nihon ho¯so¯ shuppan kyo¯kai, 1984). Will, Pierre-Étienne, “State Intervention in the Administration of a Hydraulic Infrastructure: The Example of Hubei Province in Late Imperial Times,” in Stuart Schram (ed.), The Scope of State Power in China (New York, St. Martin’s Press, 1985), pp. 295–347. Will, Pierre-Étienne and R. Bin Wong, with James Lee, Nourish the People: The State Civilian Granary System in China, 1650–1850 (Ann Arbor, University of Michigan Press, 1984). Xia Mingfang 夏明方, Minguo shiqi ziran zaihai yu xiangcun shehui 民国时期自然灾害与 乡村社会 (Beijing, Zhonghua shuju, 2000). Zhou Yuanlian 周远廉 and Xie Zhaohua 谢肇华, Qingdai zudianzhi yanjiu 清代租佃制 研究 (Shenyang, Liaoning renmin chubanshe, 1986).

521

https://doi.org/10.1017/9781108587334.015 Published online by Cambridge University Press

14

Cities and the Urban Economy harriet zurndorfer

Introduction While China’s rural economy predominated during the imperial era, some of the world’s largest cities were part of the Chinese landscape. From the Song dynasty (960–1279) onward, the number of cities and towns rose, the urban population expanded, and the urban sector of the economy became a significant indication of the wealth and prosperity of the Chinese empire. Even though large cities such as Chang’an and Luoyang had existed both before and during the Tang dynasty (618–907) and featured sites of production and services, they were founded and functioned primarily as political capitals. In the Song era an extensive array of types of cities besides capitals – maritime ports, provincial transport hubs, manufacturing and commercial centers – flourished as trade and cultural metropoles. Chinese cities took on a different configuration during the Song – one may speak of a “new urban paradigm”: in contrast to the cities of the Tang era with their enclosed wards, gridiron streets, tightly controlled markets, and sharp hierarchical social structure, the Song-era city was shaped by mercantile society and managed by pragmatic bureaucrats.1 The expansion of the urban sector during the Song and later eras drew upon the rural sector’s mobility, which helped reinforce and boost urban-based commerce and consumption. From the Song onward the number of small market towns below the rank of county seat mushroomed while the formal distinctions between cheng 城 (administrative city), zhen 鎮 (market or commercial town), and shizhen 市鎮 (market town) became more complex. A location labeled zhen (or sometimes 1

Chye Kiang Heng, Cities of Aristocrats and Bureaucrats: The Development of Medieval Chinese Cityscapes (Honolulu, University of Hawai‘i Press, 1999).

522

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

shi) might have as many as 100,000 inhabitants enjoying all kinds of urban amenities, but it lacked recognition as a government administration center. A cheng was an official permanent seat of government bounded by walls and a site of state temples, while a market town – originating in commercial transactions or manufacturing – was subject to economic fluctuation and could retrogress into a village, or progress into a recognized xiangzhen 鄉鎮 (county market).2 In other words, administration and commerce did not always converge: while some towns were merchant enclaves and beacons of commercial prosperity, they were not grid points of the empire-wide administrative network, and often fell under the authority of local forces such as lineages and religious organizations. Exact population levels of cities and large towns are difficult to gauge accurately because these larger agglomerations, including major cities, were commonly under the jurisdiction of two or three county administrations which formed the basis of population assessment of both urban and rural areas. In general, cities had no regular form of taxation that was the functional equivalent of the land tax. Such insights into urban and market development were not always evident to China historians until the late twentieth century, when the model of “the Chinese city” bequeathed by Max Weber finally lost its paradigmatic allure. Weber’s vision of Chinese cities – which in contrast to European cities engendered neither autonomy nor the civic institutions that had been so important to the development of occidental modernity – was very influential. In the first decades after World War I I , the eminent French sinologist Étienne Balazs reiterated Weber’s assertions about Chinese cities lacking an urban elite autonomous from the state. Balazs wrote in his essay “Chinese Towns,” The town was unable to fill the role of social catalyst. Unlike our towns it could not become the center of attraction because its life remained dominated . . . by the omnipresent and omnipotent state – that is the uncontested, absolute, and despotic power of a class of scholar-officials who could not tolerate any form of private enterprise, or who seized any private undertaking . . . in order to stifle it.3

2

3

Yoshinobu Shiba, “Song Urbanism Revisited,” in Shiba, The Diversity of the Socio-economy in Song China, 960–1279 (Tokyo, To¯yo¯ Bunko, 2011), pp. 55–88. Étienne Balazs, “Chinese Towns,” in Balazs, Chinese Civilization and Bureaucracy: Variations on a Theme (New Haven, Yale University Press, 1964), p. 78. See also Harriet Zurndorfer, “Not Bound to China: Étienne Balazs, Fernand Braudel and the Politics of the Study of Chinese History in Post-war France,” Past and Present 185 (2004), 189–221.

523

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

By the late 1970s this static view of Chinese urbanization had been challenged by G. William Skinner, whose regional marketing approach posited a division of China into eight essentially autonomous macro-regions, not including Manchuria (see Map 13.1).4 With its set of socioeconomic criteria focused on a nested hierarchy of central places and how these locations served the administrative, religious, and cultural functions of their hinterland, Skinner’s methodology theorized that the exchange of goods and services within each macro-region was more significant than trade across neighboring territories. In each macroregion one would find, at the low end of the scale, sparsely populated areas with little commercialization, and at the opposite extreme, the central metropolis which was not only densely populated, but also highly commercialized. He also postulated that these regional urban systems coincided closely with physiographic units, defined in terms of drainage basins or watersheds. Skinner considered market towns important to his schema. In locations where the products of agriculture diversified and local economies prospered, marketing took on new forms, from periodic to intermediate marketing settlements to standard market towns. In other words, the agricultural system was integrated into China’s economic development, and the growth of urban conglomerations was not dependent on imperial government initiative. Skinner was not without his critics, who thought he minimized the importance of interregional trade.5 And, as a number of case studies discussed in this chapter demonstrate, his marketing schema is not always applicable. Yet his conclusion that the urban commercial system of late imperial China was devolving into regional units may be confirmed in statistical evidence of the declining number of large and medium-size cities and the rising numbers of market towns after the Song dynasty. In any event, whatever the flaws in Skinner’s systematic approach to marketing and cities, he drew attention away from the Cold War-era developmental theory that had given voice to Weber’s rigidities of premodern versus modern. This chapter approaches Chinese cities and the urban economy from the year 1000 to 1800 in three parts. First, it discusses the two Song capitals, Kaifeng and Hangzhou, and the dual Ming capitals, Nanjing and Beijing – the latter was also the Qing dynasty capital – to investigate the relationship between state economic policies and commercial development in these cities. The second part of the chapter focuses on the economies of a number of provincial cities serving as ports, manufacturing and distribution centers, and 4

5

G. William Skinner, “Introduction: Urban Development in Imperial China,” in Skinner (ed.), The City in Late Imperial China (Stanford, Stanford University Press, 1977), pp. 3–31. Carolyn Cartier, “Origins and Evolution of a Geographical Idea: The Macroregion in China,” Modern China 28.1 (2002), 34–73.

524

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

commercialized administrative metropoles, including an overview of market towns and urbanization rates. The third part considers Chinese cities in comparative perspective, in relation to Eurasian urbanization as well as within China, and then reflects on Weber’s city model in view of current assessments of the administrative history of the Chinese urban economy.

Beijing

Linqing llow

er Riv

Ye

Jining

Kaifeng

Grand Canal

Chang’an

Luoyang

Zhuxianzhen Huai river

Yangzhou Zhenjiang Shanghai Suzhou

Nanjing Chengdu Yangzi

river

Hankou-Wuhan Lake Dongting

Lake Tai

Songjiang Hangzhou

Jiujiang Lake Poyang Jingdezhen

Meiling Pass

Zhangzhou

Ningbo

Quanzhou Xiamen

Guangzhou Major trade routes

Foshan

Map 14.1 Major commercial cities in late imperial China

525

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

Four Capital Cities With the demise of the Tang dynasty and its once-celebrated capital Chang’an (modern Xi’an), Chinese capital cities no longer conformed to the classic model of urban planning featuring highly regulated marketplaces enclosed in separate wards. What is striking about Chinese capitals throughout the imperial era is that they were among the largest cities in the world, with upwards of a million or more residents. Their viability and capacity to absorb hundreds of thousands of people were due in part to the expansion of agricultural productivity, especially in Jiangnan (the lower Yangzi basin), as well as the efficacy of water transportation that brought foodstuffs and other goods northwards. Both Kaifeng and Beijing relied on the Grand Canal for the delivery of grain to feed the massive armies assigned to defend these locations as well as the cities’ urban residents.

Kaifeng in the Northern Song Among the nine cities that served as capitals in imperial Chinese history, Kaifeng was the first to be chosen in large part for economic reasons. Because of its strategic location on the Central Plain of north China and its access to water transport via the Yellow River and the Bian river connection to the Grand Canal, the town of Kaifeng grew in importance as the Tang dynasty ebbed. During the Five Dynasties era (907–960), Kaifeng – also known as Bianjing, or “Bian Capital” – was the capital for four of these regimes. The first Song emperor, Taizu (r. 960–976), saw the advantages of Kaifeng as the logistical hub of a system of waterways and roads linking the city to the far corners of the realm. Kaifeng’s connection to the Grand Canal, which could easily replenish supplies for the armies stationed in and around the city, outweighed the location’s potential shortcomings, including the threat of Yellow River flooding and its susceptibility to attack by militant steppe rivals. In his effort to centralize his regime, Taizu placed the army directly under his control in the capital, which meant that the shipment of grain from elsewhere was integral to the well-being of the city – contemporaries reckoned that the government dispatched yearly 6 million piculs of grain from the southeast to Kaifeng. The concentration of military forces, estimated at around 300,000 men in 1048, also stimulated commerce in the city – the need for food and other comestibles gave business opportunities for merchants to trade and to expand their financial interests in city property, both shops and housing. In comparison to Chang’an’s extensive administrative center dominated by high walls, Kaifeng grew as a relatively open city; it lacked Chang’an’s

526

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

exact grid-like organization and precise alignment of wide thoroughfares. Kaifeng’s streets were irregular and narrow, and its “old palace city” was much smaller than the Tang equivalent. Kaifeng even lacked a strict boundary defined by a city wall – the city continuously expanded from an “old city” to a “new city” and into the rural suburbs over the course of the tenth century. The population of the city and its environs rose quickly from the time of its founding as a capital. Estimates of Kaifeng’s population vary around 890,000 between 976 and 984, increasing to 1.2 to 1.5 million a century later, which counts the military, registered civilian households (numbering approximately 1,050,000, based on 180,000 households averaging 5.8 persons), and some 150,000 unregistered persons.6 Kaifeng was not only the political capital of China, it was also the largest commercial center of the country. According to data for commercial tax quotas in 1077 for over 2,000 places ranging in size and population from large cities to market towns, Kaifeng had the highest commercial tax revenues in the empire – five times greater than the next-largest city, Hangzhou. A metropolis with round-the-clock business and entertainment establishments, Kaifeng hosted varieties of people, from street vendors and peddlers to the cultural elite who frequented the city’s many theatres, restaurants, and courtesan quarters. It was a city of palaces, altars, and temples as well as a mish-mash of commercial enterprises from petty street hawkers to elegant shops catering to the rich and well-connected. Such urban pursuits were reproduced in the panoramic scroll Traveling Upriver at Qingming Festival attributed to Zhang Zeduan (1085–1145), which over the centuries became the prototype of a “cityscape” genre representing a harmonious and prosperous world that blended urban life with natural beauty. The scroll begins with a passage through the countryside along the Bian river, illustrating a country village set against the wind-blown loess soil of northern China, barren weathered trees, and time-worn thatched-roof cottages. The busy urban scene depicted in the scroll (Figure 14.1), showing people in conversation, eating in restaurants, purchasing goods in shops, consulting fortune tellers, or engaging in everyday work routines, such as transporting goods or selling wares, against the background of an elaborate city gate and multistoried buildings, appears in the latter half of the scroll. Although controversy exists over the painting’s artist and date of 6

Wu Songdi 吳松弟, Zhongguo renkou shi 中国人口史, vol. 3, Liao Song Jin Yuan shiqi 辽宋金元时期 (Shanghai, Fudan daxue chubanshe, 2000), p. 574; Kang Chao, Man and Land in Chinese History: An Economic Analysis (Stanford, Stanford University Press, 1986), p. 50.

527

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Figure 14.1 Detail from Traveling Upriver at Qingming Festival Source: photograph by Werner Forman/Universal Images Group via Getty Images

Cities and the Urban Economy

composition, Traveling Upriver at Qingming Festival does convey the significance of artistic portrayal of the scenic contrast of urban life and natural landscape, elements integral to the painting of “cityscapes.” Kaifeng’s infrastructure as a commercial center was boosted by a change in the law enabling persons to buy and sell property within urban boundaries, and property prices depended on commercial attractiveness; the usefulness of real estate altered the traditional perception of land in China. One may consider this new right to buy and sell property as a commodity within urban boundaries a sea change in Chinese economic development because it allowed a given area of urban land to accommodate more people than before. Moreover, as the Song government aimed to keep urban taxes low, it unintentionally encouraged rural residents to move to cities, which in turn had a positive impact on the quickening pace of commercial development. Kaifeng was also an industrial center. As a metropolitan area, Kaifeng’s demand for construction materials and fuel fostered the expansion of the iron, steel, and coal industries, which generated increased local specialist production of agricultural and handicraft implements as well as weapons. Kaifeng had several arsenals, employing some 13,000 workers, who made a variety of military goods, including armor, spears, swords, and mechanized bows, as well as saddles, ropes, and stoves. Such industrial production on a large scale, however, proved regional and arguably cyclical – with the invasion and occupation of Kaifeng by forces of the Jurchen Jin kingdom in 1127, the city’s economic dynamism atrophied and it rapidly became a political and cultural backwater. Its heyday was remembered by many writers who wrote nostalgically about its vitality and sights, as exemplified by Meng Yuanlao’s (1090–1150) memoir Dreams of the Splendor of the Eastern Capital. Meng’s work expresses in unprecedented detail the glories and daily rhythms of Kaifeng: the opening of gates and markets, the traffic, the rapid increase in urban noise, and, not least, the tremendous varieties of food on offer from lowly dumpling shops to highclass wine houses. But because of its fragmentary nature, one should not read this text as “proof” of specific urban conditions. Rather, Dreams of the Splendor of the Eastern Capital may be considered the first example of a “capital-journal” text by a local observer, a genre of literary writing that continued to gain adherents during the late imperial era.7 7

Stephen West, “The Interpretation of a Dream: The Sources, Evaluation, and Influence of the Dong Meng Hua Lu,” T’oung Pao 71.1 (1985), 63–108.

529

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

Hangzhou in the Southern Song The Jin conquest of north China divided the empire into two regions, with the productive southern localities eventually ruled from the new Southern Song capital at Hangzhou. Hangzhou had become wealthy and prosperous already in the Sui dynasty thanks to the building of the Grand Canal, which terminated there. With the demographic shift southwards in the Tang, Hangzhou rose from secondary standing to the primary economic and cultural center of the Yangzi delta region (Jiangnan). Rice farming, sericulture, and tea cultivation, as well as porcelain manufacture, textile production, and papermaking, made Hangzhou and its environs ever more rich and affluent. Hangzhou had another geographical advantage: its coastal location allowed it ready access, via canal networks, to nearby deep-water harbors such as Zhapu, Ganpu, and Ningbo that had shipping links with southern seaports such as Quanzhou and Guangzhou serving the overseas trade with Southeast Asia as well as direct trade with Japan and Korea.

YANGZHOU

Yangzi R. ZHENJIANG NANJING

Gr an d

Jiangyin

Ca na l

CHANGZHOU

Liujiagang

Wuxi Lou R.

WUHU

Luodian

SUZHOU Lake Tai

Shanghai Huangpu R.

Wusong R.

Zhujiajiao SONGJIANG JIAXING

HUZHOU Wuqing

Puyuan

Zhapu Ganpu

HANGZHOU

SHAOXING Prefecture County Seat Market Town

Map 14.2 Major cities of Jiangnan

530

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

NINGBO

Cities and the Urban Economy

Although the choice of Hangzhou as the capital was disadvantageous because of its layout – it lacked any traditional rectangular form, thus making the city unsuitable as a capital from a ritual standpoint – it did possess strategic benefits: a natural barrier of waterways, lakes, rice fields, and hills that would make difficult any attacks from the Jin. From an economic standpoint, Hangzhou’s advantages seemed infinite. It was effectively connected by waterways to other prominent economic centers: along the Grand Canal to Suzhou and Zhenjiang; and via the Zhedong canal to the coastal port of Ningbo. In the region surrounding Hangzhou, craftsmen manufactured porcelain, lacquer, leather goods, and bamboo utensils, as well as paper for the growing publishing industry. City markets also supplied a growing luxury consumer demand for jewelry, fine textiles, and hats. Farmers in surrounding villages brought rice to the many market towns near Hangzhou in exchange for food items such as flour, vinegar, soy sauce, salt, pepper, and ginger. The bustling economy also served peoples of all different social backgrounds who enjoyed popular entertainment, including street storytellers and stage performances in theaters capable of holding several thousand people. The city’s premier pleasure attraction was West Lake, whose magnificent natural setting invited tourists to enjoy the locale’s many wine houses, tea pavilions, entertainment quarters, and souvenir shops. And thanks to the commercial taxes on making and selling wine, water plants (bamboo and water chestnuts), and aquaculture, the government also gained income from tourism, which it used in part for the upkeep of West Lake, a site that had to be maintained regularly in order to provide water for Hangzhou. In its prime as the capital of China, Chao Kang has estimated, Hangzhou had a million people living within the city walls, but another 1.5 million inhabiting the city’s suburbs, for a total metropolitan population of 2.5 million persons.8 The officially registered population of Qiantang county, the seat of the prefecture, grew at a rate of 1.2 percent between 1170 and 1225, while the communities outside the city’s walls expanded even more quickly – 2 percent per year. The significance of these figures reflects the trend of urban and market development in mid- to late imperial China whereby some major cities were not production sites but commercial emporia selling goods along with grain and other foodstuffs grown elsewhere specifically for urban consumption.9 In the 8

9

Chao, Man and Land, p. 54. Wu Songdi more modestly calculates the population within the walls of Hangzhou c. 1265 at around 800,000–900,000 persons and the suburban population at around 400,000 persons. See Wu, Song Liao Jin Yuan, p. 584. Yoshinobu Shiba, “The Business Nucleus of the Southern Song Capital of Hangzhou,” in Shiba, Diversity of the Socio-economy, pp. 89–128.

531

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

Hangzhou region, rural residents began to depend on supplying raw materials for urban silk manufacture as their main source of income, which they used to pay for food and other household necessities. Shipbuilding for both inland and overseas vessels was also an important industry. A sophisticated transport and marketing system delivered vast amounts of grain, mainly from nearby Suzhou, to feed the inhabitants of Hangzhou, while a commercialized distribution system involving large and small wholesalers provided for the needs of rural residents and supported a growing number of market towns. The city’s urban economy was scrutinized in Former Matters of Wulin (c. 1280) by Zhou Mi. Zhou’s encyclopedic publication, following the capital-journal genre of Dreams of the Splendor of the Eastern Capital, details the goods and movements of Hangzhou’s abundant markets, the confusions of space along the city’s dense alleyways, and the sumptuary displays of the rich and famous, but like its predecessor it is not a systematic account of the capital’s economy. It is, however, a certain indication of the intricacies of urban living in the world’s most densely populated city of that time. The Mongol capture of Hangzhou in 1276 did not curtail the prosperity of the city or the surrounding Jiangnan region. The Mongol Yuan government did not restrict commerce, nor the further development of market towns; in fact, it “enhanced the position of the Jiangnan region as the economic center of the Chinese world.”10 In the interest of gaining tax revenues, the Yuan authorities encouraged commercial activity. An important new agricultural product promoted by the Yuan government was cotton, which transformed the sandy coastal region between Hangzhou Bay and the Yangzi river and boosted this once agriculturally poor location to rising prosperity.11 Another indication that the Yuan era witnessed a continuation of earlier market affluence is the record of Jiangnan temple building, which also demonstrates the connection between commercial wealth and religious piety. As affluent merchants and rich landowners endowed temples, monasteries, and godworshipping societies, religious organizations began to attract rural inhabitants living in small hamlets and drew them into larger catchment areas that helped extend marketing networks into their locales. Hangzhou’s demise is connected to the wars occurring in the last decades of the Yuan, when chaos and military rivalries, preceded by natural disasters 10

11

Richard von Glahn, The Economic History of China: From Antiquity to the Nineteenth Century (Cambridge, Cambridge University Press, 2016), p. 278. Harriet Zurndorfer, “Cotton Textile Manufacture and Marketing in Late Imperial China and the ‘Great Divergence’,” Journal of the Economic and Social History of the Orient 54.5 (2011), 701–38.

532

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

and widespread epidemics, brought an end to earlier peace and opulence. While the city was not destroyed during the Yuan–Ming transition, it never regained the central political authority of the Southern Song era nor the prodigious levels of prosperity of the Yuan dynasty. Nevertheless, Hangzhou did continue to thrive as a mercantile and industrial center for silk textiles throughout the Ming, and rivaled Suzhou in terms of commercial wealth, despite the removal of the capital. And it remained a beautiful site where generations of people visited and enjoyed the material goods produced in its surroundings, as well as the sensuous pleasures of West Lake.

Two Ming Capitals: Nanjing and Beijing Nanjing was the capital of the Ming dynasty from its founding in 1368 until 1421, when Beijing replaced it as the primary seat of government. The relocation may be explained in part by governmental concern with northern border defense and the Yongle Emperor’s (r. 1403–1424) policy of military expansion. Yongle’s father, Zhu Yuanzhang, the first Ming emperor (Hongwu, r. 1368–1398), had chosen the original capital site known as Jinling (Golden Hills), a city with a rich cultural and religious history and myriad economic advantages because of its location on the Yangzi river. Nanjing’s economic prosperity was already important in the Southern Dynasties era (317–589) when, it is claimed, it was the most densely inhabited city in the world, with a population of around a million persons (280,000 households).12 Then known as Jiankang, the city was a booming entrepôt for domestic and imported goods, and a center for Buddhists (both Chinese and foreign). Reputed for the manufacture of bronze mirrors, lacquer, ceramics, and silk goods established by northern migrants, the city thrived on its commerce and easy access to local food resources. Although Jiankang was razed during the Sui dynasty, it recovered and retained some economic and strategic importance thereafter. But it was Zhu Yuanzhang – envisioning the site as a tribute to its earlier capital-city-building traditions – who refashioned Nanjing into the magnificent symbol of Ming political grandeur. With its three high and wide concentric city walls, largely built out of brick, and thirteen gates, Nanjing conveyed allure as the political center of China. Featuring broad avenues, grand palaces, and vast rows of state-sponsored residential quarters, Nanjing 12

Liu Shufen, “Jiankang and the Commercial Empire of the Southern Dynasties: Change and Continuity in Medieval Chinese Economic History,” in Scott Pearce, Audrey Spiro, and Patricia Ebrey (eds.), Culture and Power in the Reconstitution of the Chinese Realm, 200–600 (Cambridge, MA, Harvard University Asia Center, 2001), pp. 35–52.

533

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

had an entirely different appearance and ambience than the two Song-era capitals. In contrast to Kaifeng and Hangzhou, Nanjing in its earliest decades was almost entirely populated by soldiers, state-employed artisans, and civil servants – it “lacked any semblance of a private sector.”13 Local merchants, as part of their corvée obligations, were expected to provide government agencies with goods and services on various occasions such as the civil service examinations, ceremonies for Ming princes, or imperial weddings. After 1421, Nanjing remained the secondary capital, with civil and military agencies as well as eunuch households duplicating the Beijing government, and attracted an array of officials who extended their social networks beyond the city. As the city’s population grew, its cultural prominence also expanded: it became the leading center for book publishing, while its silk industry, including both state and private workshops, also enlarged. Over the course of the dynasty, Nanjing’s economic growth developed in spite of the Hongwu Emperor’s initial attempts to limit its private commerce and industrial capacities. The choice of Beijing as the primary capital of the Ming dynasty was entirely political – it lacked natural resources and was not located on a coast or major river. Earlier, three regimes – the Khitan Liao (916–1125), the Jurchen Jin (1115–1234), and the Mongol Yuan – had designated the same site as a capital city (known as Dadu during the Yuan era) and constructed channels for water supply from three nearby rivers which eventually became a chain of picturesque lakes west of the city center. In the first decades of the Yuan era, the Grand Canal, whose northern portions had been abandoned during the Southern Song, was renovated with a new arm dug from Shandong directly to Dadu, creating a direct northward link from Hangzhou. With a population of nearly half a million or more, Dadu was one of the world’s greatest cities at the time – no wonder that the visitor Marco Polo was impressed, since the population of his own hometown Venice, then one of Europe’s largest cities, numbered around 90,000. One of the first tasks of the new Beijing-based Ming government was to renovate the Grand Canal that had fallen into disuse during the last decades of the Yuan. Other undertakings at the time included the reforestation of north China, the rebuilding of the Great Wall, the construction of new palaces and ritual sites, repairs to the city’s walls and moats, and the resettlement of populations from the poverty-stricken provinces of north China. Beijing’s newly transferred military residents included 250,000 men, which 13

Richard von Glahn, “Municipal Reform and Urban Social Conflict in Late Ming China,” Journal of Asian Studies 50.2 (1991), 281.

534

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

meant that from early on the city’s population would increase substantially and urban space outside the inner palace area would become densely crowded. By the second quarter of the fifteenth century, Beijing was a magnet for commerce, which the government encouraged. Merchants were granted langfang 廊房 (dwelling compounds with porches) and allowed to deal in surplus government-requisitioned porcelain and textile wares in the growing number of city markets. The demand for coal and building materials tied the locale to north China’s wider regional economy. While the crucial strategic position of Beijing may have been its only obvious advantage, by mid-century the capital had become a flourishing metropolis of 700,000 residents with access to both ordinary and extraordinary goods and services. With its brick houses, tiled roofs, and symmetric north–south and east–west perimeters, Beijing seemed a beacon of prosperity in an austere terrain, and despite its notoriety because of the thousands of beggars living on its streets. From the mid-fifteenth century, Beijing benefited from improved local farming and better irrigation systems that helped boost local food production and supplies, but these enhancements were insufficient for feeding the large armies stationed along the northwestern and northern frontiers. To solve this problem, the Ming court commissioned merchants to deliver large quantities of grain and war materiel from central China to the northern borders, and paid them with highly lucrative franchises in the interregional salt distribution network. Such linkages received further stimulus during the sixteenth century, when Japanese silver and then New World silver lubricated the already burgeoning Chinese economy. As empire-wide commerce and trade proliferated, the Ming state provided huiguan 會館 (native-place lodges) for traveling merchants in Beijing, and the institution began to spread all over the empire. Despite its wealth and riches, the Ming regime faltered and succumbed to the Manchu conquest of the empire. The Qing dynasty made Beijing the sole capital, and the city remained a military fortress with massive walls and gates; a political center; and a sophisticated, commercialized, and attractive metropole. The three high Qing (1680–1796) emperors Kangxi, Yongzheng, and Qianlong built a series of new palaces north of the city, adding princely mansions, gardens, temples, and – under the influence of the Jesuits – European-style buildings and fountains. Beijing’s Liulichang book district, famous wine houses, and exclusive shops catered to the civil servant elite, while its boisterous and busy southern district hosted increasing numbers of huiguan and entertainment establishments. The growing city population reached more than 750,000 in the eighteenth century, and peaked at a million by 1800. By then Beijing had expanded in size, with more residential 535

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

and commercial buildings constructed in the dense southern outer city. Analysis of 498 extant Qing house-sale contracts indicates that real-estate prices grew steadily from 1650, declined after 1720, and remained constant thereafter, trends that are consistent with rice price indices for the same period.14

Secondary Cities The extraordinary population figures of China’s capital cities were due initially to the high numbers of military troops stationed at these locales, while the expansion of their urban economies relied on effective access to food, goods, and services. But there were other urban centers that also achieved significant population levels through trade, manufacturing, and distribution services. Their histories are noteworthy for what they reveal about patterns of market integration and the Chinese urbanization process over the centuries, but also for what they conceal in the long term: that the proportion of smaller cities or market towns, with a threshold of 2,000 or more inhabitants, rose as the number of very large cities declined from the Ming period onward. One reason for this phenomenon relates to the relative blurriness of the rural–city dichotomy – small Chinese cities and market towns had close links with the countryside and could be integrated into both agricultural and nonagricultural economic sectors. Such inter-sectoral mobility indicates the flexibility of commercially oriented smallholder farmers, and, in some cases, a profit-driven local elite keen to recoup losses from the relative meager returns of farming by tapping investment in trade and handicrafts. Although there were numbers of Chinese secondary cities before the Song era that functioned as seaports, transport hubs, manufacturing sites, or commercialized administrative centers, these places enjoyed ever greater wealth and prosperity from the year 1000 onward. As maritime links between China and other parts of Asia intensified, coastal cities benefited from both foreign trade and the strengthening of commercial ties with their hinterland areas, not only for the supply and delivery of trading goods but also for their own sustenance. Within China, urbanization could also be driven by transportation: several towns sprang up in northern China as a result of the rebuilding of the Grand Canal in the fifteenth century and its expansion to 14

Zhang Xiaolin 张小林, Qingdai Beijing chengqu fangqi yanjiu 清代北京城区房契硏究 (Beijing, Zhongguo shehui kexue chubanshe, 2000).

536

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

cities, such as Shandong province’s Jining and Linqing, which were distribution points for grain, textiles, and bricks destined for the Great Wall and Beijing’s Forbidden City. Elsewhere, centers for craft production of fabrics, ironwork, and ceramics grew in size and population, becoming major manufacturing cities but not officially recognized cheng, while other locations with formal urban status achieved wealth and prominence through commerce, as discussed below in the cases of Yangzhou, with its salt industry, and Suzhou, through silk manufacture. By the end of the eighteenth century, however, secondary cities – whatever their significance to the expansion and diversity of the imperial economy – no longer were growing as fast as the smaller and mid-sized towns that mediated market exchange between metropole and hinterland. In other words, urbanization in late imperial China was taking place at the lower levels of the urban hierarchy, a phenomenon directly the inverse of the pattern in early modern Western Europe.

Port Cities Along China’s extensive littoral were a number of well-known port cities, some of which had lengthy histories and far-reaching trade connections with other parts of maritime Asia (see Schottenhammer’s chapter in this volume). Their economic well-being was tied not only to their coastal position but also to regional networking within China. The best-known port of China was Guangzhou, which had long served as a trading post and the gateway for maritime tribute embassies. Guangzhou’s position as southern China’s premier city was consolidated by the Sui dynasty, which designated it the port of entry for overseas embassies and the launching station for punitive naval actions against coastal raiders. Guangzhou’s links with the Yangzi river were established via a series of rivers and canals that passed through the modern provinces of Guangxi and Hunan as well as those connecting the rich agricultural region of Lingnan (Guangdong and Guangxi provinces) to Jiangxi and then on to the Yangzi river valley. Goods – including pearls, rhinoceros horn, glass, nutmeg, cloves, and pepper – carried by foreign merchants from as far away as the Persian Gulf were exchanged for Chinese silk at Guangzhou. The amount of traffic and the high value of the trade goods prompted the Tang government to create a maritime superintendency (shibosi 市舶司) whose function, in principle, was to oversee the shipment of tribute items to the capital but also to prevent local officials from exploiting the overseas traders in their own interest. In 879, when the rebel leader Huang Chao captured Guangzhou, he reportedly massacred large numbers of Arab, Persian, Christian, and Jewish merchant residents, and it 537

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

would be a long time before these seafaring traders would visit Guangzhou again.15 That Huang Chao was reputed to have destroyed the region’s mulberry trees and thereby wrecked silk production, at least temporarily, also played a role in the temporary demise of Guangzhou at the end of the Tang period. Another important Chinese port city was Ningbo, which from the eighth century flourished as a center both of coastal trade with Japan and Korea, and of long-range commerce with other regions of China. The city was linked to Hangzhou via canals and thus effectively to the Grand Canal itself. Large oceangoing junks from southeastern China transferred their cargoes at Ningbo, from where they were shipped to Hangzhou or trade posts along the Yangzi river. Ningbo, too, gained a maritime superintendency in 992, which until 1523 controlled the maritime tribute of Korea and Japan, except for brief periods in the Yuan era. Water conservancy, the construction of dams and floodgates, and canal expansion were essential for the protection and commercial expansion of Ningbo, which was connected to the rich agricultural and craft production of Jiangnan and supported by powerful local families during the Song era and thereafter.16 The history of Quanzhou is instructive, because unlike Guangzhou and Ningbo – which continued to function as important port cities, even after a number of cyclical economic downturns – the prosperity of this Fujian maritime enclave did not endure beyond the last years of the Yuan dynasty. The foundations of Quanzhou’s commercial economy were laid in the tenth century, when monastic institutions there acquired large landholdings, thereby forcing local warlords to turn to taxing maritime trade as a source of income. As a transshipment area for goods imported from insular southeast Asia and the Indian Ocean and destined for the markets of central and northern China as well as Japan and Korea, Quanzhou profited from the growth of foreign trade in the Song dynasty and overtook Guangzhou as the primary Chinese port of trade.17 By the fourteenth century, however, Quanzhou’s efflorescence dimmed: the city declined and very much unraveled by the early Ming, when trade restrictions came into force. But in the late

15

16

17

John W. Chaffee, The Muslim Merchants of Premodern China: The History of a Maritime Asian Trade Diaspora, 750–1400 (Cambridge, Cambridge University Press, 2018), pp. 47– 9, 62. Yoshinobu Shiba, “Ningbo and Its Hinterland,” in Shiba, Diversity of the Socio-economy, pp. 129–80. Hugh R. Clark, Community, Trade, and Networks: Southern Fujian Province from the Third to the Thirteenth Century (Cambridge, Cambridge University Press, 1991).

538

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

Ming other southern Fujian port cities such as Xiamen and Zhangzhou burgeoned and became major trading emporia. During the first 100 years of the Song dynasty, there was no maritime superintendency at Quanzhou, which meant neither native nor foreign traders had to pay a customs duty (usually 15 percent of the goods’ value). As such, the port attracted more and more business, and ever higher numbers of foreign merchants coming from afar, as evidenced by the increasing numbers of mosques and Tamil shrines in the city. Despite the lack of a customs office, local Quanzhou officials during the 1060s traded with these merchants and collected their own levies as a 10 percent duty on the trade goods. But in 1087 the Song government, under pressure to finance wars on its northern frontiers, finally took the step of establishing a maritime superintendency at Quanzhou that directly collected customs duties on imports. This levy became a rich source of income for the dynasty. While the government’s maritime tariff revenues lingered around a yearly halfmillion guan through most of the eleventh century, the amount increased to an average of 1.1 million guan in the years 1102–1110, and then soared to around 2 million in the mid-twelfth century.18 Such income helped Quanzhou displace Guangzhou as the largest port in the empire. In its twelfth-century heyday, Quanzhou’s urban population was around 200,000 to 250,000 persons.19 While native Chinese traders from Quanzhou expanded their overseas business networks, more and more foreigners took up residence in the city. Quanzhou’s Ashab (Shengyou) and Qingjing mosques, Tamil Hindu Shiva monuments, and Christian and Manichaean churches – as well as China’s largest Buddhist temple, the Kaiyuansi, whose twin pagodas so dazzled Marco Polo – demonstrated the city’s cosmopolitanism. Quanzhou’s thriving economy was also reflected in the relatively extraordinary numbers of local men who succeeded in the civil service examinations and attained the highest degree of jinshi. During the Southern Song Quanzhou produced four vice councilors of state, three of whom became chief councilors, which was quite an achievement for a prefecture with a relatively modest administrative status. That status was also enhanced during the Southern Song era, when the southern branch of the imperial clan took up residence in Quanzhou. Some clan members also served as supervisors of maritime trade. 18 19

Clark, Community, Trade, and Networks, p. 132. Billy Kee-long So, Prosperity, Region, and Institutions in Maritime China: The South Fukien Pattern, 946–1368 (Cambridge, MA, Harvard University Asia Center, 2000), pp. 178–83.

539

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

Quanzhou’s economy continued to boom during the Southern Song as more of its hinterland farmers pursued commercial cropping of cotton, sugarcane, and lychees instead of grain. The importation of rice from Guangdong and Zhejiang became normalized. At the same time, most of Quanzhou prefecture’s counties became involved in ceramic production, which was strongly supported by local traders investing in manufacture and distribution of pottery in both domestic and international markets. Scholars have offered various reasons for Quanzhou’s decline before its final demise as an international port at the end of the Yuan. These explanations include structural changes in the local economy involving the depletion of coins carried overseas, thereby leaving the local population with not enough cash to pay for food imports; disruptions in overseas markets, in particular in Srivijaya and Champa; and the reorientation of trade whereby Ningbo overtook Quanzhou in terms of trade volume. An alternative clarification for Quanzhou’s weakening position during the Yuan focuses on the control of trade. Unlike the Song government, the Yuan authorities gave monopolistic privileges to foreign traders, mainly Muslims who now were organized into ortoq or merchant associations. The ortoq were part of the controlling patronage networks headed by Mongol royalty and generals at the court in Dadu that extended downwards through powerful Muslim officials and then to local Quanzhou Muslim families.20 The ortoq had a monopoly on the most valuable items of trade (gold, silver, copper, iron, and slaves), from which private traders were prohibited. Tensions between the Chinese trading elite and foreigners in Quanzhou began to mount, and in 1357, at a time when the rebellions spreading throughout central China reached the southeast, the city fell to the so-called Persian Garrison, a militant group notorious for plunder and exploitation of maritime trade. Many rich local Chinese and foreign merchants at this point moved away from Quanzhou. With the establishment of the Ming state, Quanzhou’s trade became tied to the lucrative tributary missions from the Ryukyu islands, but Quanzhou never regained the status or wealth it had once enjoyed as a great port city of maritime Asia.

Manufacturing Cities Jingdezhen, located in northeast Jiangxi province, has been associated with the production of ceramics, both pottery and porcelain, throughout imperial history. The first pottery manufactured in this locale was in the Han period, 20

Chaffee, Muslim Merchants, pp. 126–9.

540

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

and during the Tang the site supplied porcelain to the imperial court on several occasions. In the Song and Yuan dynasties, high-quality porcelain and coarser pottery were made in workshops and kilns scattered in the surrounding hills, but little, if any, was crafted in Jingdezhen itself, which functioned as a government control point for official orders and a distribution center.21 But in the Ming, and in particular during the sixteenth century, the city of Jingdezhen was transformed from a market hub into an industrial center. Several factors account for this change: increasing demand by the imperial court for Jingdezhen wares, the availability of raw materials, the technical expertise of the potters, growing labor supply, expanding transport and marketing networks, and increasing commercial opportunities. In the late Ming the imperial factory shifted from employing hereditary artisans to hired labor, and contracted out orders to private kilns. Jingdezhen wares were now produced in the town, and most of the outlying kilns closed down. Jingdezhen became “an island of urban industry in a sea of agriculture and small-scale handicrafts.”22 At no point in its history, however, did Jingdezhen ever attain the administrative status of a county seat. Jingdezhen had many advantages: local agriculture produced rice and tea, and other crops including wheat, millet, buckwheat, and sorghum, and there was nearby lake and river fishing. The surrounding forested mountains supplied the necessary manufacturing materials: kaolin (China clay) and bedunzi (China stone), as well as wood, both brushwood and pine, for firing the kilns. The surrounding extensive river system connected Jingdezhen to Lake Poyang and allowed the ceramics to be shipped northward to the port of Jiujiang on the Yangzi, the route for wares transshipped to Nanjing and eventually Beijing, and southward (partly by land via the Meiling Pass) to Guangzhou for both domestic and foreign markets. The commercial boom of the mid-Ming, from 1550 onward, fueled the Jingdezhen ceramic industry to new heights of production. Court records indicate that the demand for “official wares” intensified. These items were made both under the supervision of court-appointed officials and under contract to private kiln operators who manufactured “private wares,” consisting of the whole range of goods from fine porcelain treasured by collectors to tableware for everyday use. Within Jingdezhen there were no large-scale 21

22

Liang Miaotai 梁淼泰, Ming Qing Jingdezhen chengshi jingji yanjiu 明清景德鎮城市經 濟硏究 (Nanchang, Jiangxi renmin chubanshe, 2004). Michael Dillon, “Transport and Marketing in the Development of the Jingdezhen Porcelain Industry during the Ming and Qing Dynasties,” Journal of the Economic and Social History of the Orient 35.3 (1992), 278.

541

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

commercial enterprises to market the private wares, and very little of what was produced was transported out of town that was not already sold via the broker system. Licensed brokers (yahang 牙行) dealt with potential buyers, offering samples to view and arranging the purchase and delivery once the buyer had made his choice. All sales had to be conducted through the brokers, and once a deal was made, the brokers would arrange the details of the sale, quantities, prices, packing, and transport for a fee of 25 percent of the cost. Under such circumstances, with the yahang in total control of the sale and distribution process, there was no opportunity for buyers to deal directly with the potters. In the mid-Ming, Huizhou merchants began buying Jingdezhen wares via these brokers, and through their extensive business networks were able to distribute them all over the empire (see McDermott’s chapter in this volume). Before the end of the Ming, their involvement in Jingdezhen became more intense: they invested in private kilns and imported textiles (which local people could not produce). But the decentralized system of marketing persisted, and even in the early Qing – when Jingdezhen porcelain was highly prized, especially among Europeans – little had changed within the city of Jingdezhen. Some retail businesses did take root in the town, and on Porcelain Street one could find porcelain and pottery of varying quality for direct sale, but usually these were small family businesses which could not afford to display and sell highly prized items. By the mid-nineteenth century, both the industry and town of Jingdezhen began to decline in terms of production output and profits. Although ceramic manufacture in the locale did experience something of a revival in state-controlled factories during the first decades of the PRC, the recent era of economic reform has not boosted either the quality or the quantity of Jingdezhen ceramics, and the city is now better known for its tourism than for the value of its products. Jingdezhen and three other towns – Foshan, Zhuxianzhen, and Hankou – were known for their spectacular industrial or commercial development in the late imperial era, but they were not integrated into the administrative hierarchy of Chinese urban areas.23 Foshan (Guangdong province), located some twenty-five kilometers from Guangzhou, had been a Buddhist temple town since the Tang dynasty, famous for its ceramics and metalwork. Like Jingdezhen, Foshan fostered a huge population, estimated at 200,000 inhabitants in the early Qing, and thrived because of its natural resources and 23

He Yimin, “Prosperity and Decline: A Comparison of the Fate of Jingdezhen, Zhuxianzhen, Foshan and Hankou in Modern Times,” Frontiers of History in China 5.1 (2010), 52–85.

542

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

geographical advantages encompassing connections to southern China’s major inland waterways and the Pearl river delta. In the Ming and Qing dynasties the manufactures of Foshan’s iron-smelting and casting workshops, as well as its textile industry, were shipped all over the empire and overseas. Nevertheless, like Jingdezhen, Foshan never reached a higher administrative status than zhen. But in contrast to the porcelain center whose commerce was tied to brokers and sojourning Huizhou merchants, this southern city’s wellbeing was monopolized by four powerful lineage trusts that controlled the production and distribution of its industries. They owned and operated foundries, kilns, wharves, ferries, and warehouses. Although it may be argued that such lineage organizations engaged in commerce and aimed at profit in order to support their members’ financial well-being, the promotion of the city of Foshan into some kind of proto-capitalist bulwark was far beyond the capacity of these kinship organizations.24

Commercialized Administrative Cities A number of cities in Ming and Qing China, including Xiamen, Guangzhou, Chengdu, Nanjing, and the Wuhan conurbation of Hankou, Wuchang, and Hanyang, were among the largest in the world. These metropoles, recognized within the administrative hierarchy as cheng, did not seek autonomy from central political authority, but they did possess a certain amount of practical communal self-management. Their prosperity was linked to the growth of regional market towns that intensified local and interregional trade, which ultimately fed urban consumption. These cities attended to the administrative needs of imperial governance – including tax collection, the maintenance of local infrastructure such as canals and city walls, and crime prevention – while their economies expanded and attracted ever more trade. Among China’s best-known examples of commercialized administrative cities were Yangzhou and Suzhou, both of which possessed vibrant economies as well as rank within the empire’s administrative system. Both these metropoles had extensive histories of trade and commerce before the Song, and one may argue that their economic success reflected the role of government agency in their long-term development. Both Yangzhou and Suzhou in their heydays were opulent, and attracted cosmopolitan and sophisticated elites as well as visiting Qing emperors. But there were also marked 24

David Faure, Emperor and Ancestor: State and Lineage in South China (Stanford, Stanford University Press, 2007), pp. 218–32.

543

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

differences between the two cities. While Yangzhou was probably wealthier than Suzhou in the eighteenth century, it was situated in a poorer, less urbanized region than Suzhou and lacked economic resources other than salt production and distribution. Suzhou, located in the center of Jiangnan, profited from its hinterland where market towns and well-managed transport networks linked the city to a diversity of agricultural and craft production centers. Yangzhou

In the history of Chinese cities, Yangzhou is exceptional because it achieved riches and cultural fame in two different eras, the Sui–Tang period and the Ming–Qing epoch. In the eighteenth century, with an urban population of 100,000, it was one of the empire’s wealthiest cities. As the administrative center of the Qing government’s richest monopoly, the Lianghuai salt trade, Yangzhou attracted Huizhou merchants who lavishly patronized the city’s art and architecture, and gave generously to philanthropic activities. Yangzhou’s gardens, night markets, and teahouses were famous all over the empire. But in the early nineteenth century, with the waning of the salt trade, the city went into decline. Located north of the Yangzi river 250 kilometers from the coast, Yangzhou did not have a favorable hinterland. The low-lying terrain was prone to frequent flooding, and the soil was too salty for extensive agriculture. Thus “Yangzhou became a prosperous city in a poor region.”25 The Eastern Han (25–220 C E) saw the construction of the first canals connecting the area to the Yangzi, and by the time of the Sui dynasty both the region and the city of Yangzhou were flourishing. The Sui and Tang governments built a series of canals forming an extensive network of waterways connecting the Yangzi, Huai, and Yellow river systems to one another. Yangzhou’s location at the intersection of the Yangzi river and the Grand Canal made it a hub of both domestic and international trade, even hosting merchants from Central and Western Asia. The city also possessed shipbuilding, textile, leather-making, and iron and bronze metallurgy industries, as well as the largest salt market in the empire. In the Northern Song, Yangzhou’s international-port status was eclipsed by Ningbo, and once Hangzhou became the capital in the Southern Song, the transport of grain and goods to the north also ended. The city’s fortunes took 25

Antonia Finnane, Speaking of Yangzhou: A Chinese City, 1550–1850 (Cambridge, MA, Harvard University Asia Center, 2004), p. 11.

544

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

an upward turn during the Yuan when the government made Yangzhou the center of the salt administration, but the wars at the end of the dynasty decimated the city’s population, and even as late as the mid-sixteenth century both agriculture and handicraft industries in Yangzhou prefecture were hardly developed. In the late Ming, however, Yangzhou’s demography changed. While the prefecture’s population level remained relatively low, the number of people living in Yangzhou city began to increase. Shanxi merchants and then Huizhou merchants came to the city to trade in salt, and subsequently took up residence there. Yangzhou at that point began to gain a reputation as a city of decadence, offering countless leisure venues where sojourning merchants and “extravagant” women were known to engage in dubious behavior. The salt monopoly offered merchants highly lucrative franchises in interregional salt distribution networks (see the chapter by Lamouroux and von Glahn in this volume). By the late Ming, Huizhou merchants dominated the Yangzhou salt industry and became fabulously wealthy. The Qing government’s restoration of the salt franchise system advanced the position of Yangzhou’s Huizhou merchants even further. They enjoyed the legal protection of the imperial government in exchange for guaranteeing that Yangzhou’s relatively meagre supply of grain would not be a problem: the Huizhou salt merchants established charitable granaries for famine and flood relief, a task normally undertaken by local gentry elsewhere. A certain proportion of the fortunes that salt merchants made between 1750 and 1800 – estimated by one account to be 250 million taels26 – was poured into making the city of Yangzhou beautiful, mirroring the tastes of Emperor Qianlong (r. 1736–1796), who was known to shield the traders against bureaucrats denouncing their practices. For Qianlong’s 1768 visit to Yangzhou, the salt merchants were reported to have spent some 4,679,000 silver taels cleaning and grooming the city. During their southern tours both Qianlong and his grandfather Kangxi marveled at the architectural face of Yangzhou, where a number of villas emulated the appearance of imperial palaces. Yangzhou residents also had an appreciation of what we would call global consumer items: euroiserie in the form of clocks and watches, tobacco containers, and even English woolens and cotton broadcloth came into routine use among the elite. Yangzhou architecture also indulged in European styles and architectural technologies that included the construction 26

Ho Ping-ti, “The Salt Merchants of Yang-chou: A Study of Commercial Capitalism in Eighteenth-Century China,” Harvard Journal of Asiatic Studies 17.1 (1954), 130–68.

545

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

of verandas, the decoration of buildings with clockworks and mirrors, and the creation of Western-style windows made of glass. Yangzhou in the eighteenth century saw a vibrant theater culture, with rich salt merchants sponsoring opera troupes from all over the empire performing at banquets as well as at temples, where they drew huge crowds of urban commoners. The splendors of eighteenth-century Yangzhou urban life, and in particular its gardens, were the subject of the most famous guidebook to the city, Li Dou’s Chronicle of the Painted Barques of Yangzhou, first compiled in 1765. In his guide Li stressed the elite’s demand for high-quality goods and materials from distant reaches of the empire and from foreign lands as far away as Burma, Arabia, and Europe, with which they built their mansions, adorned their interior walls and external facades, and turned their properties into a variety of landscapes, including some with Islamic and European artistic influence. He recorded how “marble from Sichuan and Qinghai, jade from Yunnan and Burma, stone from the bed of Lake Tai, and tropical woods from southeast Asia” were integral to Yangzhou’s luxurious building culture, and how such exotic elements added spectacle to the city’s allure.27 Some scholars attribute Yangzhou’s decline at the end of the eighteenth century to conspicuous consumption. In his classic study of commercial capitalism in China, Ho Ping-ti posited that the spending habits of Yangzhou merchant families not only cast them as “socially vulgar” in the minds of many local scholars, but also became their economic downfall – they had invested the bulk of their profits in their lineages and in nonbusiness uses, which in the long run depleted their financial resources.28 Buying luxurious items for Qianlong’s palaces might have brought political clout, but ultimately it also helped empty one’s purse. Alternatively, Huizhou merchants might have found rising land values in their home community a more attractive investment and therefore abandoned Yangzhou.29 But it is also important to consider Yangzhou’s demise in the context of the failing salt administration at the end of the eighteenth century, when the government was having to divert sums originally destined for waterway maintenance and the salt fields to military campaigns of suppression against the Miao and White Lotus rebellions. The floods in the early nineteenth century that extensively damaged the Grand Canal, which had not been properly maintained for several decades, forced the government to ship grain and salt via 27 29

YZHFL, pp. 162–3, 401. 28 Ho Ping-ti, “Salt Merchants of Yang-chou.” Harriet Zurndorfer, Change and Continuity in Chinese Local History: The Development of Hui-chou Prefecture 800 to 1800 (Leiden, Brill, 1989), p. 165.

546

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

other means. Ultimately these circumstances undermined the economic base of Yangzhou. Yangzhou’s particular development defies Skinner’s urbanization paradigm. Its hinterland acted as a barrier. The city was not nested in a series of hierarchical marketing regions, and it was the imperial state that directed the flow of goods along the Grand Canal and the salt commerce; Yangzhou’s connections with other cities and market towns followed a dendritic schema. Yangzhou’s wealthy merchants did not reinvest their profits into the poorly developed countryside, nor did they, arguably, relate to Yangzhou’s native elite or “local gentry,” who celebrated their affinity with the locale in local histories and literature after the Huizhou merchants had departed. While these merchants in their heyday did support artistic, literary, and philanthropic activities in Yangzhou, in the long term they did not fuse with the local city elite such as the families of the distinguished scholars Wang Zhong and Ruan Yuan, who married among the prefectural local gentry. As such, Yangzhou’s stratified society reflected the economic development of a city dependent on a sojourning merchant elite endorsed by the central government that in the long term was the principal agent behind the city’s rise and fall. Suzhou

Around the time Yangzhou was beginning to experience a sense of ennui, in the early nineteenth century, Suzhou was at the height of its economic prosperity. No city in Jiangnan or the rest of the empire boasted as much wealth and population.30 In contrast to the Ming era, when dyeing and calendering of cotton cloth and the manufacture of fans, lacquerware, copperwork, brushes, inkstones, ceramics, and jade carving were diffused throughout Jiangnan, in the Qing these industries were concentrated in Suzhou. Facets of cotton production shifted from Songjiang to Suzhou and its environs, while silk manufacture, a portion of which fell under imperial monopoly, remained a primary Suzhou industry. Other features of Suzhou during the Ming continued into the Qing, such as its onerous grain tax burden, its masses of poor people, and at the other extreme its passion for luxury consumption and high living – with regard to fashion, the city was a trendsetter and quality producer.

30

Michael Marmé, “Survival through Transformation: How China’s Suzhou-Centred World Economy Weathered the General Crisis of the Seventeenth Century,” Social History 32.2 (2007), 144–65.

547

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

Like Yangzhou, Suzhou has a lengthy history. Located on the lower reaches of the Yangzi river near Lake Tai, Suzhou was founded in 514 B C E as the capital of the Wu kingdom during the Eastern Zhou’s multistate conflicts, and remained ever after a seat of southern Chinese culture, celebrated for its wealth, luxury, beautiful women, gardens, and bridges. During the imperial era, it was a prefectural capital surrounded by seven counties, of which two had administrative status within the city – Wu in the western half and Changzhou in the east – and renowned for its silk industry. Ever since the Tang period, Suzhou prefecture’s rice yield was the highest in the empire, and taxed accordingly. These financial tolls also reflect the productivity of Suzhou’s agriculture, which by Song times had benefited from such land reclamation innovations as embankments, channels, and polders, allowing labor-intensive wet-rice agriculture and multiple cropping managed by private landlords.31 During the Northern Song, rice continued to be Suzhou’s most famous product and thus its commercial orientation at that point was directed toward the agricultural sector. But in the aftermath of the Jin conquest of the north, Suzhou experienced a migration wave, and with the city’s population now reaching 300,000 persons, more land reclamation and water control renovations were instituted, and the inhabitants also turned to handicrafts as a source of income.32 During the Yuan period Suzhou’s economic strength grew steadily; in particular, it benefited from the Beijing-based Mongol rulers’ decision to send southern grain north by sea through the nearby port of Taicang. The social unrest of the late Yuan did not affect Suzhou much. As Marmé argues, the one-time salt smuggler and rebel leader Zhang Shicheng, who in 1356 took over Suzhou and its environs, in effect continued the benign policies of the previous regime: “he encouraged handicraft production, expanded trade, undertook water-control projects, reclaimed land, supported existing social arrangements, and courted the traditional cultural elites.”33 But his rival, Zhu Yuanzhang, the first Ming emperor, who considered the Suzhou elite a potential source of trouble, uprooted the wealthy and educated elite families (about 92,000 households) in order to disrupt their networks, and he forced many of these people to resettle in the new capital at

31

32

33

Michel Cartier, “Aux origines de l’agriculture intensive du Bas Yangzi,” Annales: Économies, sociétés, civilisations 46.5 (1991), 1009–19. Michael Marmé, Suzhou: Where All the Goods of All the Provinces Converge (Stanford, Stanford University Press, 2005), pp. 49–53. Marmé, Suzhou, p. 58.

548

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

Nanjing or in more rural locations. He also taxed punitively the remaining residents. With Beijing’s installation as the new capital and the reconstruction of the Grand Canal, Suzhou in the early fifteenth century began to play a key role in the integration of China’s now separate political and economic centers. While the wholesale grain trade remained its principal source of wealth, the spread of textile production, and in particular the production of highquality silk fabrics, raised Suzhou’s prosperity to new levels. Sericulture involving the cultivation of mulberry trees, rearing silkworms, maintaining the cocoons, reeling yarn, warping, and weaving silk fabrics became integral to Suzhou’s reputation. Initially, during the early Ming, silk weaving had been confined to the city’s male artisans because of the expense of the looms and the need for specialized training. But as demand for superior-quality silk skyrocketed in the 1470s–1480s, production shifted to the countryside under the supervision of the local elite, who employed inexpensive and semiskilled labor to take up silk weaving. Rural people were familiar with cotton weaving, and merchants in Suzhou’s immediate hinterland organized their learning to weave (simple) silk textiles. Local merchants were in charge of a putting-out system that became increasingly specialized in terms of weaves, weights, and dye colors. This shift also stimulated the spread of rural markets that eventually became satellite marketing towns, extending Suzhou’s commercialization beyond the city’s walls. As for Suzhou’s elite, Huang Xingzeng wrote in his Collection of Jottings on Wu [Suzhou] that the local gentry (shenshi 紳士) “often take increasing their wealth to be an urgent matter: [their] shops in the six suburbs open for business in moneylending and pawnbroking and do a brisk trade in salt and alcoholic beverages.”34 Controlling local markets, moneylending, and reaping rural land rents were the material basis that enabled Suzhou’s local elite to build exquisite city mansions and create beautiful gardens.35 In the course of the Ming, as the economic interdependence of city and countryside became more pronounced, Suzhou’s economic well-being relied more on the profits of nonagricultural pursuits. Now rice from the middle and upper reaches of the Yangzi river valley was imported to feed the local population, as well as being re-exported to other rice markets in eastern China, making Suzhou the hub of the national grain market. While Suzhou remained the most densely populated region in China from the mid-Ming to 34 35

WFL, 5–6. Yinong Xu, The Chinese City in Space and Time: The Development of Urban Form in Suzhou (Honolulu, University of Hawai‘i Press, 2000).

549

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

the end of the eighteenth century – with an urban population of around half a million in the late Ming, rising to more than 1.4 million in the eighteenth century – its fortunes did experience downturns, first in the second half of the sixteenth century, and then during the Ming–Qing transition. But the city recovered relatively quickly and became ever more wealthy in the eighteenth century. This economic rise may be attributed to the expansion of city-based handicraft industries producing paper, fans, lacquerware, brushes, and inkstones, as well as the amplification of the silk industry and cotton processing in the surrounding area. In the environs of Suzhou there was also a substantial industry for building oceangoing ships that Fujian- and Guangzhou-based merchants ordered. Moreover, Suzhou had easy access to the empire’s three main transportation routes: the Grand Canal (north– south), the Yangzi river and its tributaries (east–west), and the coastal sea routes extending to Manchuria in the north and Guangzhou in the south. Suzhou in the eighteenth century, now the political capital of Jiangsu province, was at the core of Jiangnan’s urban regionalization and China’s wealthiest city. Its economic dynamism was depicted in prints and paintings of its cityscapes, showing a lively mercantile culture. Qing artworks, such as Xu Yang’s well-known painting Emperor Qianlong’s Southern Tour (1770; Figure 14.2), are distinguished by their portrayals of city people (shimin 市民), in contrast to the Suzhou-based Ming literati painting repertoire of scenic suburbs visited by men of letters engaging in cultural exchange. This does not mean that the Suzhou scholar-literati disavowed the bucolic learned life, but such artworks are testimony to the reverence for commerce and the importance of the market to the city’s well-being. What is striking in this illustration is how orderly and how natural the contours of the densely built city of Suzhou conform to the surrounding countryside. While the activities of individual persons in the painting indicate the frenetic and commercial side of bustling Suzhou, the painting’s panoramic view represents the symbolic significance of the work’s homage to Emperor Qianlong, for whom this picture was composed, in commemoration of his 1751 visit to the city. In other words, although the scene here depicts urban activities, including shops, boats, a local canal, and the general hustle and bustle of city life, the figures and buildings are repetitive and stiff, and thus less representative of Suzhou as a lively metropolis and more indicative of the city in a grand performance in honor of the emperor’s visit there. In the eighteenth century, Suzhou depended not only on rice imports to feed its population and pay its taxes, but also on imports of timber to supply its shipbuilding industry, bean cake to fertilize its soil, and raw cotton from 550

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Figure 14.2 Panorama of Suzhou in the eighteenth century Xu Yang, Qianlong nanxun tu 乾隆南巡圖 (Emperor Qianlong’s Southern Tour) (1770), Metropolitan Museum of Art, New York, Purchase, The Dillon Fund Gift, 1988

harriet zurndorfer

north China to augment its own supply.36 While Suzhou continued to prosper into the early nineteenth century, the health of its economy began to unravel as the general stresses of the 1830s, including natural disasters and poor governmental management, took their toll on the city

Market Towns Ever since the Northern Song dynasty, with the shift of China’s economic heartland from the north to the southeast, commercialized agriculture helped create a dense network of small cities and towns in Jiangnan that dominated Chinese development down to the nineteenth century. Although the commercial tax data for 1077 indicate that, besides Kaifeng, other northern and some western (Sichuan province) cities and towns also contributed significant revenue, it was in the southern regions, especially along the coast where agricultural and handicraft production thrived, that market settlements expanded in both number and size. One estimate puts the Southern Song urbanization rate, i.e., the proportion of the total population living in towns and cities, at around 12 percent, but the scale was uneven: even in places featuring commercialized agriculture such as Huizhou (Anhui), with its tea and tree plantations, only 3 percent of the households lived in the prefectural city. One region which experienced vigorous market town growth, especially during the Yuan period, was the Yangzi delta.37 Here, where sericulture was the mainstay of the economy, marketplaces expanded into towns as they developed from collecting hubs for silk-making supplies from rural producers into trade emporia that sold silk textiles. For example, Wuqing – originally a Tang military garrison, located along a watercourse parallel to the Grand Canal – emerged during the early Southern Song as a prosperous market town, famous for its bridges and roads, mansions and gardens, and the diversity of goods in its shops. In the early thirteenth century, the town was divided into two separate administrative units (Wuzhen and Qingzhen) under the jurisdiction of different counties. Although razed during the late Yuan civil wars, Wuqing recovered its prosperity in the late Ming, when the town was said to house 5,000 households, thanks to its advantageous access 36

37

Michael Marmé, “From Suzhou to Shanghai: A Tale of Two Systems,” Journal of Chinese History 2.1 (2018), 79–107. Richard von Glahn, “Towns and Temples: Urban Growth and Decline in the Yangzi Delta, 1100–1400,” in Paul Jakov Smith and Richard von Glahn (eds.), The Song–Yuan– Ming Transition in Chinese History (Cambridge, MA, Harvard University Asia Center, 2003), pp. 176–211.

552

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

by waterways to the region’s major towns and cities. In the Qing period, silk weaving became concentrated in other towns such as Shuanglin and Shengze, while industry in Wuqing became more rural. Wuqing’s inhabitants supplied silk yarn to the weaving centers, and also were renowned for producing light, soft cotton textiles. Also indicative of this development is the history of Puyuan, situated fifteen kilometers south of the Jiaxing prefecture capital, which became an important silk production and market center in the Yuan period. This town owed its origins to the Pu lineage, originally from Shandong province, which had fled the Jin invasion and settled in Jiangnan. In the early fourteenth century the family expanded its landholdings that it had financed via sericulture, including the manufacture of silk cloth, and brokerage in rice and other agricultural items. In 1307 the Pu lineage built the Yongle marketplace near their residential compound, complete with streets and shops; they also used their wealth to endow local religious institutions.38 Puyuan and the Pu clan experienced a terrible downturn in their fortunes, however, with the late Yuan civil wars and the Ming dynasty founder’s deliberate policy to strip Jiangnan’s wealthy families of their wealth and status. For nearly a century, the Pu lineage, like other local enterprising wealthy kin groups, endured economic loss, but in the sixteenth century the delta’s economy, reinvigorated by expanding empire-wide markets for silk (as well as cotton) textiles, saw Puyuan’s silk industry make a tremendous comeback. Innovative weaves like figured gauzes elevated Puyuan’s silk manufacture to new heights of prosperity and the town’s population and size both doubled. One may reflect that while late Ming urban growth was indicative of yet another pathbreaking stage in Chinese economic development, known as the “second commercial revolution,” the initial stages of intensive rural production and commerce were crucial to setting in motion the foundation of Ming market towns. From the mid-Ming, as the empire’s population grew, so did the number of market towns, both large and small, increase all over the country (see Cao’s chapter in this volume). But these commercial conglomerations were not a sign of growing urbanization – rather, as Chao suggests, the proliferation of small towns was due to the ruralization of nonagricultural production.39 In Jiangnan, as population pressure forced rural households to reallocate surplus 38

39

Richard von Glahn, “The Sociology of Local Religion in the Lake Tai Basin,” in John Lagerway (ed.), Religion and Chinese Society, vol. 2, Taoism and Local Religion in Modern China (Hong Kong and Paris, Chinese University Press and École française d’extrême-orient, 2004), pp. 773–815. Chao, Man and Land, p. 57.

553

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

labor no longer absorbed by farming, they produced yarns and textiles (cotton and silk) and other handicraft products which merchants collected at rural sites such as landings, crossroads, and bridges, close to the producers’ villages. These collection points eventually developed into market towns that connected with surrounding villages that had easy access to water transport.40 Local gazetteer records chart the growth of a number of market towns in Jiangnan. For example, the number of market towns in Jiaxing prefecture rose from twenty to forty-nine between the Southern Song and the late Ming, while those in Changzhou increased from nineteen to eightythree over the same period.41 From the early sixteenth century to the end of the eighteenth, the number of towns in the Jiangnan prefectures of Songjiang, Suzhou, and Hangzhou nearly doubled.42 With regard to similar developments in the rest of China, it is known that in the Xiangtan river valley (Hunan), the number of recorded market towns rose from sixteen in 1591 to 117 in 1747. Comparable figures may be found for the upper Yangzi (Sichuan) and Lingnan regions.43 The rise of small towns reduced transportation costs and thereby stimulated trade and commerce. Interestingly, as China experienced its greatest population rise in its history, from 185 million in 1680 to 312 million by 1776, the urbanization rate (based on a threshold of 2,000 persons) decreased to around 7 percent for the entire country, and for Jiangnan from 23 to 19 percent.44 Moreover, the largest cities in 1800 – Beijing, Suzhou, Guangzhou, Chengdu, Nanjing, and the Wuhan conurbation – were not as large in terms of population and geographical size as the two Song capitals Kaifeng and Hangzhou had been. By 1820, when China’s urbanization rate was around 5 percent and the population 383 million, the total number of urban dwellers was not much larger than the roughly 17 million urban residents during the Southern Song, when the population was only 124 million (see Cao’s chapter in this volume). The enormous demographic increase in the Qing had been 40

41

42

43

44

Kawakatsu Mamoru 川勝守, Min Shin Ko¯nan shichin shakai shi kenkyu¯: Ku¯kan to shakai keisei no rekishigaku 明清江南市鎮社会史研究: 空間と社会形成の歴史学 (Tokyo, Kyu¯ko shoin, 1999). Fan Shuzhi 樊树志, Ming Qing Jiangnan shizhen tanwei 明清江南市镇探微 (Shanghai, Fudan daxue chubanshe, 1990), pp. 487–94. Liu Shiji 刘石吉, Ming Qing shidai Jiangnan shizhen yanjiu 明清时代江南市镇研究 (Beijing, Zhongguo shehui kexue chubanshe, 1987), pp. 141–56. William Rowe, “Social Stability and Social Change,” in Willard Peterson (ed.), The Cambridge History of China, vol. 9, part 1, The Ch’ing Dynasty to 1800 (Cambridge, Cambridge University Press, 2002), pp. 538–9. Li Bozhong 李伯重, Jiangnan de zaoqi gongyehua 江南的早期工业化, 1550–1850 (Beijing, Shehui kexue wenxian chubanshe, 2000), p. 417.

554

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

primarily absorbed by the rural sector. What began as China’s second commercial revolution in the late Ming, with empire-wide commerce in low-value-per-bulk staples of cotton, cotton cloth, rice, and grains, matured into a systematic graded network of long-distance intra- and interregional trade, both rural–urban and rural–rural – supported by a greatly expanded transport labor force of sojourning porters, carters, animal drivers, and boatmen – that sustained China’s tremendous population expansion.

Chinese Cities and the Urban Economy in Comparative Perspective While the Chinese economy remained vastly rural in the late eighteenth century, so was much of Europe. According to one calculation, only around 13 percent of Europeans lived in an urban community of 10,000 or more inhabitants in 1800, but in some countries, including England, the Low Countries, and Italy, the percentage was higher.45 Urban population growth in China continued until the sixteenth century, when the population in rural market towns began to rise faster, while in Europe the number of larger cities, as opposed to small towns, began to increase. Nevertheless, by 1750 only thirteen European cities – topped by London, Paris, Amsterdam, Naples, and Vienna – had more than 100,000 inhabitants. Several decades later both Beijing and Suzhou had more than a million residents. Looking further afield, in Japan, Edo’s population size also reached a million by 1800, and a century earlier, some cities of the Safavid, Mughal, and Ottoman empires, including Isfahan, Shahjahanabad (Old Delhi), and Constantinople, along with the capital cities Cairo and Damascus, counted populations of hundreds of thousands. Beyond these Eurasian comparisons, one may also assess specific locations with regard to urban development in a restricted framework. A case in point is late medieval northern Italy and Jiangnan around the year 1300, which demonstrates the relationship between rural production, marketing, and urbanization.46 In that year, Milan – with a population of 200,000 – was a finishing center for cloth (woolens and cotton–linen blends) that was 45

46

Jan de Vries, European Urbanization, 1500–1800 (Cambridge, MA, Harvard University Press, 1984), pp. 72–3. Chris Wickham, “Jiangnan Style: Doing Global Economic History in the Medieval Period,” in John H. Arnold, Matthew Hilton, and Jan Rüger (eds.), History after Hobsbawm: Writing the Past for the Twenty-First Century (Oxford, Oxford University Press, 2018), pp. 121–39.

555

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer

initially processed in small satellite towns by wage-labor artisans. In contrast, the raw materials for the silk textiles manufactured in Hangzhou originated in rural peasant households in exchange for cash payment intended for farmland rent and the purchase of food. But in northern Italy, peasants were required to pay their rent in kind, which landlords marketed to their own advantage in these towns. What is striking here is how limited peasant commercialization in northern Italy was in comparison to Jiangnan at that time. Northern Italy’s urbanization depended on a division of labor by which peasants and skilled workers labored in separate spheres, and which constrained commercial opportunities for the former and allowed local elites to develop urban economies more easily for their own profit, while Chinese farmers prospered through inter-sectoral economic mobility. Looking at the cities discussed in this chapter, one may observe a number of comparative points in their histories. Both Quanzhou and Yangzhou in their halcyon days were cosmopolitan emporia. Quanzhou hosted a wide range of people from afar and provided space for their religious practices, while the late imperial Yangzhou elite delighted in the material culture of foreign places, including the accoutrements of European architecture. Also, neither Quanzhou nor Yangzhou had an immediate hinterland which could feed the residents sufficiently, which meant that their well-being was tied to commercial alternatives. While Quanzhou turned to the production and distribution of ceramic and iron goods to raise cash for food imports, Yangzhou relied on the Huizhou salt merchants to ensure an adequate supply of grain and to establish charitable granaries. Moreover, the prosperity both of Quanzhou during the Yuan period and of Ming–Qing Yangzhou was tied to the practice of state-issued licenses (for foreign trade in the former case, and salt in the latter) whereby the government employed a merchant stratum on a large scale to provide logistical services. Another comparison concerns the pre-Tang economic development of the two cities of Suzhou and Nanjing (Jiankang) which were both highly developed in terms of resources, trade, and agricultural and handicraft production before the Tang–Song transition. The early success stories of these cities – in the case of Suzhou based upon well-managed agriculture, and in Jiankang on trade and manufacture – indicate that the ideal of urban prosperity in the Jiangnan region existed long before the economic changes fostered by the Northern Song regime. Common to both locations was the role of magnate landlords who in effect became wealthy “commoner” landowners and merchants beginning in the fifth century. Interestingly, neither location fared well under strong dynastic regimes: 556

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy

Suzhou was intermittently burdened with high tax rates, and Jiankang was annihilated by the Sui dynasty and “de-privatized” during the early Ming. Finally, let us consider Weber’s stance on Chinese cities, which he believed were not seedbeds of modern capitalistic development because they lacked autonomy and self-government and remained subject to the dominance of imperial authorities. In contrast, this survey of China’s urban economy indicates that cities and towns experienced different degrees of governmental supervision: from close association (capital cities, Yangzhou) to self-sufficiency (manufacturing cities and market towns). In county seats and walled cities, the magistrate and his functionaries superintended commercial networks and public services such as the maintenance of water supply and sewage, crime and fire prevention, and city garrisons, while in market towns such duties fell to “urban gentry” and retired scholar-officials native to the locale. In smaller towns, local powerholders (landowning gentry, local lineage groups, rich merchants, and religious organizations) took over these tasks, and aimed to ensure incorporation of the lowest units of habitation, i.e., village and polder, under their suzerainty. In the long term, whatever the scale of regulatory supervision, the economies of city and countryside alike drew upon the agricultural production and craft industries of rural producers geared to commercial profit, which operated within a framework intended not to celebrate urban life per se, but to ensure the prosperity of the empire.

Further Reading Clark, Hugh R., Community, Trade, and Networks: Southern Fujian Province from the Third to the Thirteenth Century (Cambridge, Cambridge University Press, 1991). De Pee, Christian, “Purchase on Power: Imperial Space and Commercial Space in Song-Dynasty Kaifeng,” Journal of the Economic and Social History of the Orient 53.1–2 (2010), 149–84. Dillon, Michael, “Transport and Marketing in the Development of the Jingdezhen Porcelain Industry during the Ming and Qing Dynasties,” Journal of the Economic and Social History of the Orient 35.3 (1992), 278–90. Fan Shuzhi 樊树志, Jiangnan shizhen: Chuantong de biange 江南市镇 : 传统的变革 (Shanghai, Fudan daxue chubanshe, 2005). Fan Yijun 范毅軍, Chuantong shizhen yu quyu fazhan: Ming Qing Taihu yidong diqu weili, 1551–1861 傳統市鎮與區域發展 : 明清太湖以東地區為例, 1551–1861 (Taipei, Zhongyang yanjiu yuan, Lianjing chuban gongsi, 2005).

557

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

harriet zurndorfer Farmer, Edward, “The Hierarchy of Ming City Walls,” in James Tracy (ed.), City Walls: The Urban Enceinte in Global Perspective (Cambridge, Cambridge University Press, 2000), pp. 461–87. Faure, David, “What Weber Did Not Know: Towns and Economic Development in Ming and Qing China,” in David Faure and Tao Tao Liu (eds.), Town and Country in China: Identity and Perception (Basingstoke, Palgrave, 2002), pp. 58–84. Fei, Si-yen, Negotiating Urban Space: Urbanization and Late Ming Nanjing (Cambridge, MA, Harvard University Asia Center, 2009). Finnane, Antonia, “Chinese Domestic Interiors and ‘Consumer Constraint’ in Qing China: Evidence from Yangzhou,” Journal of the Economic and Social History of the Orient 57.1 (2014), 112–44. Finnane, Antonia, Speaking of Yangzhou: A Chinese City, 1550–1850 (Cambridge, MA, Harvard University Asia Center, 2004). Gabbiani, Luca (ed.), Urban Life in China, 15th–20th Centuries: Communities, Institutions, Representations (Paris, École française d’extrême-orient, 2016). Gernet, Jacques, Daily Life in China on the Eve of the Mongol Invasion, 1250–1276 (Stanford, Stanford University Press, 1970). Gerritsen, Anne, The City of Blue and White: Chinese Porcelain and the Early Modern World (Cambridge, Cambridge University Press, 2020). Gillette, Maris Boyd, China’s Porcelain Capital: The Rise, Fall, and Reinvention of Ceramics in Jingdezhen (London, Bloomsbury, 2016). He, Yimin. “Prosperity and Decline: A Comparison of the Fate of Jingdezhen, Zhuxianzhen, Foshan and Hankou in Modern Times,” Frontiers of History in China 5.1 (2010), 52–85. Heng, Chye Kiang, Cities of Aristocrats and Bureaucrats: The Development of Medieval Chinese Cityscapes (Honolulu, University of Hawai‘i Press, 1999). Johnson, Linda Cooke (ed.), Cities of Jiangnan in Late Imperial China (Albany, SUNY Press, 1993). Johnson, Linda Cooke, Shanghai: From Market Town to Treaty Port, 1074–1858 (Stanford, Stanford University Press, 1995). Kawakatsu Mamoru 川勝守, Min Shin Ko¯nan shichin shakai shi kenkyu¯: Ku¯kan to shakai keisei no rekishigaku 明清江南市鎮社会史研究 : 空間と社会形成の歴史学 (Tokyo, Kyu¯ko shoin, 1999). Lam, Joseph, Shuen-fu Lin, Christian de Pee, and Martin Powers (eds.), Senses of the City: Perceptions of Hangzhou & Southern Song China 1127–1279 (Hong Kong, Chinese University of Hong Kong Press, 2017). Li, Lillian, Alison J. Dray-Novey, and Haili Kong, Beijing: From Imperial Capital to Olympic City (Basingstoke, Palgrave Macmillan, 2007). Liang Miaotai 梁淼泰, Ming Qing Jingdezhen chengshi jingji yanjiu 明清景德鎮城市經濟 硏究 (Nanchang, Jiangxi renmin chubanshe, 2004). Marmé, Michael, Suzhou: Where All the Goods of All the Provinces Converge (Stanford, Stanford University Press, 2005). Naquin, Susan, Peking: Temples and City Life, 1400–1900 (Berkeley, University of California Press, 2000). Prazniak, Roxann, Sudden Appearances: The Mongol Turn in Commerce, Belief, and Art (Honolulu, University of Hawai‘i Press, 2019).

558

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

Cities and the Urban Economy Shiba, Yoshinobu, The Diversity of the Socio-economy in Song China, 960–1279 (Tokyo, To¯yo¯ Bunko, 2011). Skinner, G. William (ed.), The City in Imperial China (Stanford, Stanford University Press, 1977). So, Billy Kee-long, Prosperity, Region, and Institutions in Maritime China: The South Fukien Pattern, 946–1368 (Cambridge, MA, Harvard University Asia Center, 2000). Tian Yinsheng 田银生, Zouxiang kaifang de chengshi: Songdai Dongjing jieshi yanjiu 走向开 放的城市 : 宋代东京街市研究 (Beijing, Sanlian shudian, 2011). von Glahn, Richard, “Towns and Temples: Urban Growth and Decline in the Yangzi Delta, 1200–1500,” in Paul Jakov Smith and Richard von Glahn (eds.), The Song–Yuan– Ming Transition in Chinese History (Cambridge, MA, Harvard University Asia Center, 2003), pp. 176–211. Xiong, Victor Cunrui, Sui–Tang Chang’an: A Study of Urban History of Medieval China (Ann Arbor, University of Michigan Center for Chinese Studies, 2000). Xu, Yinong, The Chinese City in Space and Time: The Development of Urban Form in Suzhou (Honolulu, University of Hawai‘i Press, 2000). Zurndorfer, Harriet, Change and Continuity in Chinese Local History: The Development of Huichou Prefecture 800 to 1800 (Leiden, Brill, 1989).

559

https://doi.org/10.1017/9781108587334.016 Published online by Cambridge University Press

15

The Monetary System akinobu kuroda

Introduction: Characteristics of the Chinese Monetary System Money serves as a means of exchange in any society and in any period. However, depending on place and time, money exhibits distinct characteristics. Chinese imperial monetary history spotlights a blind side about money that mainstream thinking overlooked. Chinese historical experience is replete with examples of money use transcending the theoretical frameworks of those who attribute the acceptability of money to its intrinsic value as well as those who emphasize enforcement by public authorities. Chinese monetary history may appear highly irregular from the viewpoints of contemporary foreign observers and modern social science, but a coherent system surely existed in an endogenous way behind the apparent disorder. Chinese imperial monetary history had three unique features: first, dependence on base metals (copper alloys) rather than gold or silver for coinage; second, widespread circulation of state-issued paper money far earlier than elsewhere in the world; and third, the circulation of silver money measured not by tale (count) but by weight (yinliang 銀兩, often referred to as sycee silver). The issue of non-precious-metal coinages by public authorities and their actual usages by end users show that the metallic content of coins was not the decisive factor determining their acceptability in exchange. The success of some dynasties in creating a viable paper currency and the failure of others indicates the inapplicability of the concept of fiat money for understanding paper money in the historical past. Neither can the concept of fiduciary money be applied to coin-denominated notes, which circulated at the local level in quantities far larger than the assets of local issuers. Moreover, the significance of transactions

560

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

through book transfers using an imaginary unit of account reveals the limitations of the popular concept of a trinity of monetary functions: means of exchange, measure of value, and means of storing wealth. The first characteristic appeared even before the establishment of the Chinese empire in the third century B C E. The production of copper coins rose substantially during the period this chapter covers, from the eleventh to the eighteenth centuries, but varied over time. Mint output of copper coin reached its highest peak already in the eleventh century, and attained its second-highest level in the eighteenth century. In addition, state-issued iron coins played a significant role in the eleventh and twelfth centuries. The inconvenience of having multiple iron currencies – which were too bulky and heavy to transport over long distances – circulating in different provinces was the precondition for the emergence of state-issued paper money. The Song and Yuan governments sought to establish the acceptability of their paper monies by limiting their circulation to specified regions and by creating fixed terms of expiry, after which old notes had to be exchanged for new issues. In contrast, paper monies with no spatial limit and no terms of expiry issued by the Jin and Ming states failed to gain acceptability among the populace. Generally speaking, Chinese imperial governments did not issue currencies made of precious metals. Silver ingots measured by weight and fineness became a popular device for exchange, especially after the sixteenth century. The Ming and Qing states collected taxes denominated in weights of silver, but they did not mint silver coins for circulation. This practice resembled the use of silk as a device for long-distance monetary transfers without issuing actual silk currencies by earlier dynasties from the Han to the Tang (see Kakinuma’s chapter in this volume). Thus copper alloy coin was virtually the sole device whereby people conducted market exchange, since even when paper money was viable the denominations were much too large for everyday transactions.1 The importance of coins with fractional value correlated with the predominance of one-time transactions in local markets. The strong autonomy of proximate exchange 1

Government records stipulate that the Song and Yuan dynasties printed paper notes whose denominations were smaller than one guan (nominally 1,000 wen [coins]), such as 100 wen, etc. Excavations from tombs have discovered specimens of low-denomination Yuan paper notes. For example, a ten-wen Zhongtong jiaochao note was found in Baida, Inner Mongolia, and twenty-wen, thirty-wen, and other denominations of Zhiyuan chao 至元鈔 (first issued in 1287) were recovered from a tomb in Yuanling, Hunan. However, the fact that certain denominations of paper money were issued does not mean that they actually circulated. Wu Chouzhong 吴筹中, Zhongguo zhibi yanjiu 中国纸币研究 (Shanghai, Shanghai guji chubanshe, 1998), pp. 17–20.

561

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

induced merchants to separate the means of exchange in local transactions from those used in long-distance trade. Merchants sometimes used paper instruments such as state-issued salt certificates or privately issued bills of exchange (for example, bills issued by the piaohao 票號 banks in the late Qing period) in making interregional settlements. Thus the simultaneous use of multiple currencies was a cardinal feature of the monetary system of late imperial China. This chapter examines not only official state-issued currencies but also local currencies created by private agents, which appear to have predominated in exchange at the ground level in local markets. The case of privately issued notes such as the coin-denominated qianpiao 錢票, which appeared from the eighteenth century, demonstrates the inadequacy of the conventional understanding that the acceptability of money must be validated by either public authority or intrinsic value. The latter two conditions can enhance the acceptability of money, but they are not absolutely necessary. In addition, no single monetary medium could mediate all spheres of exchange at any time throughout the history of the Chinese empire. The concept of a single, sovereign currency within a territorial state is the historical product of the recent past.

Persistence amidst Change: Official Copper Coinages The Chinese economy always featured multiple currencies in concurrent circulation, and the forms of money differed from period to period. However, one thing that remained unchanged was the persistence of the standard weight and fineness of official copper coin. After the reinstitution of copper coinage following its suspension during the fifteenth century, the alloy content was changed from bronze (copper–tin) to brass (copper–zinc), but the weight of coins (around four grams) and their copper content (50–90 percent) remained fairly constant not only throughout the period covered in this chapter, but even dating back to the creation of the wuzhu 五銖 coin by the Han emperor Wu in the second century B C E. Considering the frequent degradation of the monetary unit of account over time in other parts of the world, it is remarkable that the same standard for the monetary unit was maintained in China for such a long period. For example, the weight of the silver penny, or denarius, in England decreased from 20.3 grains in 1344 to 12.0 grains in 1464 and to 7.74 grains in 1601.2 Moreover, England undoubtedly maintained a higher standard in coinage than did the 2

Angela Redish, Bimetallism: An Economic and Historical Analysis (Cambridge, Cambridge University Press, 2000), pp. 89–90.

562

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

various polities of continental Europe. Deterioration of coinage for seigniorage also was an abiding feature of the history of minting in Tokugawa Japan. Put simply, in contrast to other regimes that minted coins from precious metals, the profits from seigniorage rarely attracted the attention of Chinese rulers. A similar phenomenon is found in the medieval Mediterranean world, in which a gold coin, called zecca in Venice, kept the same weight and fineness for a millennium.3 However, the maintenance of a copper coin standard in China and the gold standard of the Mediterranean world derived from differences rather than similarities in the major types of exchange. In the Mediterranean, the predominance of transnational long-distance trade supported a common gold standard, while in China the copper coin standard was based on everyday market exchange at the local level. Unlike precious metal coins, fractional and bulky base-metal coins tend to stay with end users after being distributed among them. In contrast to the common practice whereby rulers demonetized coins issued by their predecessors and required circulating gold and silver coins to be returned to the mint for recoinage (and reaping seigniorage revenue in doing so), Chinese governments rarely invalidated copper coins minted by earlier dynasties. In practical terms, returning old coins of fractional value for reminting was far harder than recalling precious-metal coinage, and generated no seigniorage revenue. Table 15.1 enumerates coins from two Southern Song hoards – one large and one small, both excavated in modern Hunan province – according to the reign-era dates of their issue. The large number of coins in the Yankeshan hoard suggests that they were intentionally saved over a period of time; in other words, this hoard is an example of a stock of coins. In contrast, the smaller cache of coins at Wenqiao yidui perhaps indicates an emergency deposit of coins held at a single moment in time, and therefore can be taken as an example of flow.4 Regardless of whether the quantity of coins was large or small – in other words, whether the hoards represent the stock or flow of currency – the range of dates of issue is similar. During the Southern Song, business transactions were conducted with coins that had been minted centuries before, including during the preceding Tang dynasty.

3

4

Alan M. Stahl, Zecca: The Mint of Venice in the Middle Ages (Baltimore, Johns Hopkins University Press, 2000). The 726 coins from Wenqiao yidui might seem to be a large quantity in other contexts, but, considering the low value of copper coins, this quantity is still too small to be considered an example of a stock of money saved over time.

563

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

Table 15.1 Distribution of reign-era names of copper coins from Southern Song hoards Yankeshan (Hunan) Percentage Wenqiao yidui (Hunan) Percentage Before 618 618–907 907–960 960–1004 1004–1054 1054–1101 1101–1126 1126–1201 1201–1260 Total

27 866 12 472 2,706 7,156 2,018 212 0 13,469

0.2 6.4 0.1 3.5 20.1 53.1 15.0 1.6 0.0

0 57 3 32 213 277 95 45 4 726

0.0 7.9 0.4 4.4 29.3 38.2 13.1 6.2 0.5

Source: Miyake Toshihiko 三宅俊彦, Chu¯goku no umerareta senka 中国の埋められた 銭貨 (Tokyo, Do¯seisha, 2005), pp. 104–5

Table 15.2 Distribution of reign-era names of copper coins from the Aba hoard, Sichuan, 1870s All periods

Number of coins

Tang (618–907) Song (960–1279) Ming (1368–1644) Qing (1644–1911) Foreign

7 30 11 16,508 13

Total

16,569

Qing dynasty

Number of coins

Shunzhi (1644–61) Kangxi (1662–1722) Yongzheng (1723–35) Qianlong (1736–95) Jiaqing (1796–1819) Daoguang (1820–50) Xianfeng (1851–61) Tongzhi (1862–74) Total

5 176 17 5,400 2,776 4,787 1,945 1,402 16,508

Source: Li Jun 李俊 and Chen Yajun 陈亚军, “Sichuan Aba lüchang Qingdai qianbi jiaocang qingku jianbao” 四川阿坝铝厂清代钱币 窖藏清库简报, Zhongguo qianbi 中国钱币 2015.1, 33–9

What about the situation in later periods? Table 15.2 enumerates the copper coins excavated from the Aba hoard in Sichuan, deposited in the late 1870s. Eighteenth-century Qianlong coins comprise the largest portion of the hoard. Although the predominant coins in the Aba hoard span a shorter period of time than those from the Southern Song hoards, people in the 1870s

564

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

still continued to use coins that had been minted more than a century earlier. Such hoards reveal that throughout China’s imperial history large quantities of copper coins tended to remain with end users for a long period of time. Since old coins comprised a significant portion of the stocks of coin held by the population, it was difficult for a new regime to debase the currency by issuing coins of lower intrinsic value to substitute for fine old coins that had accumulated at the local level. This tendency toward “stagnation” of local supplies of copper coins in China becomes obvious when we make comparisons with hoards of silver pennies in medieval England. Excavations of English hoards dating from the eleventh century reveal that pennies were held only for within twenty years of their issuance.5 In both China and medieval England (and Western Europe generally) mono-unit currency systems – one-wen copper coins in the former and one-pence silver coins in the latter – prevailed in market exchange. We may infer that the intrinsic value of the material of a coin was inversely related to the length of time in which it circulated. The lower intrinsic value of copper coins did not mean that issuers incurred lower costs for minting them. On the contrary, the maintenance of a full-bodied standard for copper coinage sometimes meant that the costs of coinage exceeded the coins’ face value. Here take two examples. An essay written in 1125 states that although, according to the official government price of copper, the intrinsic copper value of 1,000 one-wen coins was 600 wen, after adding the costs of transport, fuel, labor, and so on, the total cost to mint 1,000 copper coins amounted to 1,400–1,500 wen. This situation prevailed even in the period shortly after the Northern Song had issued the largest quantity of copper coins in history. A memorial written in the late sixteenth century reveals a debate between the minister of works and another government official over the question of resuming the minting of coinage. The minister opposed the resumption of coinage since minting five one-wen copper coins cost one fen of silver, which at the time was equivalent to seven wen; thus the government would incur a loss of 40 percent on each coin it minted. The proponent of resuming minting argued that despite the loss to the government, expanding the money supply by resuming coinage would increase the wealth of society as a whole. In defiance of the minister’s logic and modern common sense, the government accepted the latter’s reasoning and decided to go ahead and mint the new Wanli coins based on a high copper metal content. The result – a frequent occurrence throughout 5

Philip Grierson, Numismatics (London, Oxford University Press, 1975), p. 134.

565

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

Chinese imperial history – was that people melted down the fine state-issued copper coins either to recast them as cheaper private coin or to hold them as stores of value in the form of copperwares such as religious statuary. At times, Chinese authorities persevered in minting full-bodied copper coins despite the encouragement they gave to counterfeiting by private coiners, who remained undeterred by the legal threat of capital punishment. Often, however, the authorities ceased minting fine coin. Table 15.2 shows that in the late nineteenth century fewer Ming coins remained in circulation than Song coins that were minted much earlier. In addition, no Yuan coins were found. Even if the Song coins in this hoard were imitations cast in later periods rather than genuine coins (a common practice, as we shall see), the temporal distribution of coins reflects significant changes in monetary policy over time. It is impossible to estimate the quantities of copper coins minted by Chinese governments before the Qing dynasty, for which archives have been preserved. However, it is safe to conclude that state-minted output of coin reached its peak in the Northern Song period. The Southern Song court continued minting coin, but in far lower quantities than the Northern Song. The Mongol-ruled Yuan dynasty, except for several brief periods, not only ceased to mint copper coins but banned their usage in the marketplace. Despite this prohibition, the use of copper coins persisted in local exchange, especially in the southern regions previously under Southern Song rule. Initially the first Ming emperor, Hongwu (r. 1368–1398) issued his own Hongwu coins, but before long he sought to replace copper coins with a new paper currency known as baochao 寶鈔, and even went as far as to ban the use of copper coins, including his own Hongwu issues. Beginning in the 1430s the Ming court suspended coinage altogether for over half a century, until the Hongzhi (1488–1505) period. As we shall see, Song coins, both genuine coins and private imitations, comprised the majority of current coins during the Ming period. When official copper coinage was resumed in the sixteenth century, the government mints adopted a new alloy, substituting zinc for tin, and thus creating brass coins in place of the traditional bronze currency.6 The fullbodied Wanli coinage mentioned above, inaugurated in 1576, could not last in the face of meltage and private recoinage, and in the early seventeenth century the Ming reverted to a debased standard. The Qing followed in the 6

Chemical analysis proves that Ming coins from the Jiajing era (1522–1566) and afterward were made of brass, in contrast to earlier Ming issues of bronze coin (including the Hongzhi coin). See Zhou Weirong 周卫荣, Zhongguo gudai qianbi hejin chengfen yanjiu 中国古代钱币合金成分研究 (Beijing, Zhonghua shuju, 2004), pp. 90–3.

566

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

footsteps of the Ming in minting brass coins. In the eighteenth century, coins issued under the reign of the Qianlong Emperor (r. 1736–1795) predominated in local markets thanks to rising mint output (in tandem with the growth of copper mining in Yunnan in the southwest) and the wide distribution of official copper coins in rural areas through the proliferation of government famine relief agencies (which often distributed loans and welfare payments in the form of coin). As we shall see in the next section, even in large-scale transactions such as sales of arable land and real estate, copper coins often were used as means of payment at that time. At the end of the eighteenth century, however, unauthorized inferior coins from inland provincial mints such as in Sichuan began to permeate coastal regions, undermining the unified standard of official copper coins. In the first half of the nineteenth century the output of government mints gradually decreased. With the outbreak of civil war between the Qing and the insurgent Taiping Heavenly Kingdom rebels in 1850, the court abandoned its policy of minting only mono-unit one-wen copper coins, and tried – in vain – to introduce largedenomination coins with values ranging from five to as much as 1,000 wen. Because these coins were highly overvalued in terms of their metallic content, they quickly depreciated, and the Qing soon withdrew them from circulation. The Qing resumed minting mono-unit copper coins, but finally replaced the traditional one-wen copper coins with machine-made 10-wen copper coins (without the square hole in the center) beginning in 1902. Thus, although the governments in late imperial China largely retained a single standard for the weight and fineness of copper coins, the quantity of coins issued by state mints varied drastically depending on economic conditions and the availability of monetary metals. As the next section will show, these variations in coin output were primarily related to fiscal policies. In fact, periods in which the rulers could provide sufficient quantities of coin to meet demand were far more rare than periods of coin shortages. As we will see later in more detail, the Chinese populace had no hesitation in turning to private currencies when necessary. Although the unity of the official coinage standard endured for centuries, at the same time the current coin in market circulation was constantly undergoing adjustment at the local level.

Money in Tax Collection and in Land Sales Transactions Obviously, when a government depends on labor services and taxation in kind for its revenues, it devotes less attention to currency. By extension, 567

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

when the government collects taxes in money, it cannot ignore the circulation of currencies. Moreover, the form of money designated for tax collection is not always the same as the currency that the government itself has issued. The Tang state’s shift in 780 from its zu–yong–diao taxation system (collected in grain, cloth, and labor services) to the new twice-a-year tax (liangshui 兩稅) made later Chinese authorities dependent on the collection of land tax in terms of copper coins. During the Northern Song, when the state was able to issue copper coins in enormous quantities, taxation in money (copper coins) became deeply rooted. However, a mechanism for collecting taxes does not always work as a mechanism for transferring revenues from local agencies to the central government or between local jurisdictions. It was impractical for the government to transfer copper coins over long distances, and, needless to say, impossible to do so for even bulkier iron coins. Besides silk and silver, state-issued paper currencies acted as a tool for transferring funds over long distances. During the Southern Song, the state retained the copper coin standard as the unit of account in its fiscal accounts, but it adopted paper currencies as a major form of tax collection. Silver also became more significant as a mechanism for transfers of funds. However, it is important to note that although the role of copper coins in actual tax collection diminished, the values of all the currencies employed in public finance (paper monies, silver, and silk) were basically calculated in terms of copper coin units. The popularity and importance of official paper currencies increased under the Yuan, whose revenue structure was distinctly different from Chineseruled dynasties. In contrast to the usual dependence of Chinese governments on land taxes as their chief source of revenue, the Yuan instead derived most of its revenues from commercial sources, most notably the salt monopoly. These revenues were mainly collected in the form of paper currencies, especially the Zhongtong jiaochao 中統交鈔 currency first issued by Khubilai Khan (r. 1260–1294) in 1260. Although the Yuan continued to denominate its paper currencies in copper coins, government accounts used weights of silver as the unit of account. In practice, from the late 1270s, Yuan paper currencies no longer could be converted to either copper coins or silver ingots. By calculating the value of paper currencies with a face value designated in copper coins in terms of weights of silver, the Yuan established a connection between a mercantile system favored in Eurasianwide monetary exchanges (silver measured by weight) and the prevailing practice of its former Southern Song subjects to keep accounts in copper coins measured by tale. 568

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

The Ming tried to revive a revenue structure based on labor services and inkind taxation. In keeping with the Hongwu Emperor’s hostility to the market economy, the Ming adopted the most conspicuously negative attitude toward issuing copper coins among all the dynasties in Chinese history. Hongwu’s intention was to control state revenues and expenditures strictly in terms of stateissued paper currency. Although the Ming continued the precedent of its predecessors in denominating the face value of its paper money in copper coins, as we have seen for much of its history the Ming court was reluctant to mint copper coins. But the Ming effort to rely solely on fiat paper currency was doomed to failure. Consequently, the Ming became the first dynasty to adopt a fiscal system that collected revenues in the form of a currency – silver – that was not officially issued by the state itself. Burdened by the inefficiency of a revenue system predicated on labor services and in-kind taxation, in the early fifteenth century the Ming court resorted to the expediency of using uncoined silver (whose value was calculated by weight and fineness) as its principal means of transferring wealth from the lower Yangzi region to the new national capital at Beijing and to military bases across the empire’s northern frontier. Over the course of the fifteenth century the Ming gradually replaced a revenue system based on labor services and in-kind taxation with taxes denominated in standardized weights of silver, yet without issuing its own silver currency. The adoption of silver as the new monetary standard in state fiscal accounts was constrained by the limited supply of silver within the Ming borders before the abrupt rise of foreign trade in the sixteenth century. Silk and porcelain – and later tea – produced in southern China began to be traded abroad in return for large quantities of silver, initially from Japan and later from Spain’s American colonies. The Qing rulers basically followed the administrative system established by the Ming. However, the Qing had to bridge the gap between land taxes uniformly measured in weights of silver and a mechanism by which peasants across the empire could actually pay their taxes. The dramatic increase in copper coinage – to a level surpassed only in the Northern Song period – and the creation of an empire-wide network of local granaries on an unprecedented scale in the eighteenth century fostered the proliferation of copper coins as means of payment in local exchange. Although in principle the Qing dynasty collected nearly all of its tax revenues in the form of silver, the majority of peasants paid taxes in copper coins, which were converted into weights of silver by local agencies. Exchange rates between copper coins and silver taels varied from region to region, and fluctuated over time. Thus a complementary structure between copper coins for local exchange and weights of silver for interregional settlement took root. 569

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

So far, we have traced the transformation of money from the perspective of tax collectors. What impact did changes in the taxation system have on money use in actual exchange? The trend in monetary units used to draw up land contracts provides important clues. The determination of land prices must have had a strong connection to the way in which land was taxed. During the Northern Song, land sales were contracted in terms of copper coins. By the thirteenth century, however, the majority of land contracts specified payment in state-issued paper money. This trend continued during the Yuan period, when Zhongtong jiaochao prevailed as the chief form of payment in land contracts. Unlike previous periods for which information is patchy, we have more ample and continuous data on land sales for the Ming period for specific regions such as Huizhou in Anhui province. The Huizhou contracts reveal that after the early Ming, when land sales were negotiated in the official Ming currency (initially copper coin and subsequently paper money), from the mid-1430s onward land sales were uniformly contracted in terms of weights of silver. Land sales contracts from Nanping, Fujian, show the same tendency. Three contracts dated between 1401 and 1431 specified payment in paper money, while five contracts from 1501 to 1638 designated payments in weights of silver.7 Silver payments predominated in land sales in this region until the end of the eighteenth century, when copper coins began to replace silver, a trend that became dominant in the nineteenth century. At the end of the nineteenth century, land sales contracted in terms of the silver dollar appeared, and the use of ten-penny silver coins, called xiaoyang 小洋, also became popular. Although units of grain and cloth occasionally were used to specify land prices before the fifteenth century, weights of silver became almost universal in land contracts after the mid-sixteenth century. However, from the mideighteenth century, land sales denominated in copper coin reappeared, and during the nineteenth century, copper coin prevailed as the predominant means of payment for land sales in many provinces. The extent to which copper coin eclipsed silver in land contracts varied from region to region. In large cities such as Beijing and particular regions such as Huizhou, silver continued to prevail in land sales.8 Meanwhile, land sales from Qinghe, Gansu, were predominantly contracted with payments in small copper 7

8

Fujiansheng qianbi xuehui 福建省钱币学会 (ed.), Fujian huobi shilue 福建货币史略 (Beijing, Zhonghua shuju, 2000), p. 56. Li Hongmei 李紅梅, “Shindai ni okeru Fukkensho¯ no kahei shiyo¯ jittai: Tochi baikenrui o chu¯shin to shite” 清代における福建省の貨幣使用実態: 土地買券類を中心と して, Matsuyama daigaku ronshu¯ 松山大学論集 18.3 (2006), 131–73.

570

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

coins, and contracts from Qingshuijiang, Guizhou, reveal that, as late as the 1850s, transactions in terms of copper coin dominated.9 Although land sales contracted in terms of foreign silver coins (chiefly Spanish pesos) began to appear in southern coastal areas such as Fujian from the end of the eighteenth century, the majority of land sales in China changed from payments in weights of silver in the mid-eighteenth century to copper coins by the midnineteenth century. Thus the use of copper coins for land sales that had prevailed in the Northern Song period eventually was revived and came to dominate transactions of real property in the last century of the Qing dynasty. A key difference between the two periods is that the Song rulers collected taxes in copper coin, but the Qing rulers calculated tax obligations in terms of weights of silver. Prices for land sales were linked to the means of state payments in the Song, but in the Qing, the measure for calculating land prices became separated from the means of state payments. This separation can be traced back to the failure of state-issued paper money in the early Ming and the absence of a state-issued silver currency. Below we will consider why taxation in weights of silver was not accompanied by issuance of silver currency. Obviously, land sales are large-scale transactions. Thus largedenomination paper money and precious-metal currencies such as silver are more convenient for this type of transaction. As for everyday trade in local markets, the absence or shortage of official currencies suited to the scale and frequency of small transactions caused local dealers to resort to counterdevices to ensure smooth exchange operations. Threats of capital punishment for forgery and counterfeiting had little effect as long as demand for such exigencies persisted in local markets. In the next section we will trace the transformation of official currency supplies and the responses by end users at the local level across the eight centuries from the year 1000 to 1800.

Fluctuating Issue of Official Currencies and Local Use of Non-official Currencies The division of the Chinese empire into a number of competing states for more than a half-century after the demise of the Tang dynasty in 907 conditioned the choice of monetary policies adopted after reunification by the Song in 960. Throughout the two millennia of Chinese imperial 9

Guizhou minzu wenhuagong 贵州民族文化宫, Guizhou Qingshuijiang liuyu Ming Qing tusi qiyue wenshu 贵州清水江流域明清土司契约文书 (Beijing, Minzu chubanshe, 2013–14), Jiunan, pp. 130–5; Liangzhai, pp. 192–6.

571

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

history, it is only in this period that we find multiple sovereign states competing to issue their own currencies. Importantly, some states distinguished between one currency for local markets and another for interregional exchange. In order to prevent their own currencies from flowing out of their territories, rulers of rival states – especially in the southern regions – sought ways to discourage transport of coins over long distances. For example, in 918 the Southern Han kingdom adopted a lead coin standard to discourage the removal of its currency outside its borders: due to the fiscal deficit, lead coins were issued. Ten lead coins were deemed equivalent in value to one copper coin . . . Trade within the city should be conducted using lead coins, while outside copper coins should be used. Both import and export of them were forbidden. Violators were sentenced to death.10

For the first time in Chinese history, state authorities minted significant quantities of coins from base metals such as tin, lead, and iron and condoned their use in local trade. This dual structure of official coinage shaped the course of Chinese monetary history for the remainder of the imperial period. We might expect that the primary task for the Song to create a unified monetary system would have been to incorporate the existing base currencies into their own currency system. Within the core regions of the empire, the Song was able to substitute new coins of good quality for old coins issued by previous states. However, after the conquest of Sichuan in 965 the Song found it difficult to replace the existing iron currency in use in this region with new copper coins because of high transportation costs. Later, confrontations with the steppe kingdoms of Xi Xia and the Liao compelled the Song to adopt policies similar to those of the southern kingdoms of the Five Dynasties period. Iron coins were introduced in Shaanxi along the northwestern frontier in order to prevent high-quality coins from flowing out of Song territory into the Xi Xia domains. It surely is no coincidence that at least six hoards of iron coins weighing more than one ton have been discovered in Shaanxi. Thus, despite the re-establishment of unified sovereignty under the Song, the dual structure of universal coins and provincial coins was retained. Roughly speaking, iron coins were ten times bulkier than their equivalent value in copper coins. In this regard, it is not surprising that bills of exchange appeared in Sichuan at this time in order to avoid the significant cost of 10

QZ, p. 195.

572

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

transporting these low-value iron coins. However, bills of exchange for settling payments between buyers and sellers at a distance were still far removed from paper money that circulated as a general means of payment in market transactions. Bills of exchange in the narrow sense were just one type of paper certificate used mostly for interregional settlements. Hostile relations with its steppe nomad rivals required the Song to maintain large standing armies along its northern frontiers and to develop logistical systems to supply its troops. One mechanism adopted by the Song was to reward merchants who delivered food, clothing, and other supplies to the frontier garrisons with payment in certificates for monopoly goods such as tea and salt, which could be sold in designated regions at substantial profit. In addition, these certificates were negotiable instruments that could be transferred to third parties through brokers in the capital, and eventually they served as an important mechanism for interregional settlements. When the Song state began to issue its own paper currency these bills won popular acceptance as a substitute for settling interregional payments. In the background, astronomical quantities of copper coin issuance worked to back the acceptability of paper monies in terms of the copper coins. During the Northern Song period, state-issued paper instruments were dominant in interregional settlement, while copper coins dominated local exchanges. To facilitate large transactions and long-distance settlements, private dealers adopted customary “short-string” units that differed according to commodity and place. These private short-string units varied in the number of coins they contained, but they all shared the feature of containing fewer coins than the state’s official short-string standard (called shengmo 省 陌), whereby an actual string of seventy-seven coins was deemed equivalent to a full string (mo) of 100 coins (wen). Meng Yuanlao’ s memoir of city life in the Northern Song capital, Kaifeng, records the best-known statement of this practice: Officials were using a short-string standard of seventy-seven coins per mo, while on the street a seventy-five-coin standard was generally used. In transactions for fish, meat, and vegetables, a seventy-two-coin standard was used; gold and silver shops used a seventy-four-coin standard; while for purchases of pearls, hiring female slaves, and buying insects (used for sport) the standard was sixty-eight coins.11

11

“Dushi qianmo” 都市錢陌, in DJMHL, 3, p. 22.

573

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

As we shall see, such variations in local customs in money use steadily persisted until the early twentieth century. Throughout the Southern Song, the state created distinct types of official paper currency for different regions, such as the Hubei huizi 湖北會子 in the middle Yangzi region and the southeastern huizi 東南會子 that circulated in the capital of Hangzhou and the southeastern provinces, rather than a single, empire-wide currency. These paper currencies had fixed terms of expiry (usually three years), after which they had to be exchanged for the subsequent issue of the same type of note. The Southern Song authorities succeeded in maintaining the viability of these fiat paper currencies by backing them with a range of instruments possessing monetary value. If the value of the paper currencies depreciated in the private market, the government would seek to call them back in exchange for copper coins, silver, silk, salt and tea certificates, and even ordination licences for Buddhist priests (which entitled the bearer to certain tax exemptions). There was no single reserve currency that backed the paper monies, nor was it the utility of paper money as a means of state payments (which in fact it was) that sustained their viability in private circulation. The Song officials understood that it was crucial to maintain a balance between supply and demand for means of exchange (known as “leveraging balance,” chengti 稱提) in order for people to accept them. During the Southern Song, the limited quantity of coins minted by the state and large-scale printing of paper currencies resulted in a reversal of the relationship between coin and paper compared to the Northern Song. People tended to hoard fine copper coins; thus we might go so far as to say that paper currencies predominated in local trade, while copper coins were used to settle interregional transactions. Yuan paper currency policies largely followed Song precedents. The Yuan divided paper money circulation into several circuits and set annual quotas for issuing new notes. Importantly, each paper note bore a serial number. When worn notes were replaced, the old note was exchanged for a new one after checking the serial number. In addition, confirmation of the serial number was crucial to settling disputes over the genuineness of the bill.12 By maintaining control of the quantity of paper currencies in circulation, the Yuan state was able to avoid disastrous depreciation and corresponding price inflation until the final decades of the dynasty. 12

Funada Yoshiyuki 船田善之, “Gendai shiryo¯ toshite no kyu¯hon Ro¯kitsudai: Sho¯ to bukka no kisai o chu¯shin toshite” 元代史料としての旧本『老乞大』: 鈔と物価の 記載を中心として, To¯yo¯ gakuho¯ 東洋学報 83 (2001), 122–3.

574

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

The Yuan prohibition against the use of metallic currencies (copper coin, silver, and gold) and its very rare and short-lived issuance of coins meant that copper coins became increasingly scarce in market transactions. At the same time, paper monies could not reach the level of wage goods transactions. Observations by contemporary Korean visitors to Yuan China reveal that paper money of one-guan denomination was equivalent to 500 grams of pork. Government edicts mention that people used salt packets and lacquered tokens as substitutes for small denominations of money. A well-known passage by a late thirteenth-century Chinese observer relates that in the countryside near Weitangzhen, in the Yangzi delta, farmers brought sacks of rice to local markets to exchange for salt or oil – implying that such transactions were conducted without using either metal or paper currencies.13 It may be safe to assume that in the situation of scarce supply of currencies, people conducted exchange at the local level using readily marketable commodities such as grain. The official ban of copper coin during the Yuan, coupled with the rarity of state-issued coins, would appear to diminish the possibility of conducting business in terms of copper coins. However, actual trade at the ground level was always different from the descriptions found in formal sources. As the paper currency system collapsed in the late Yuan period, the use of copper coins that previously had been prohibited or restrained was revived. Because of the scarcity of coin, however, various short-string units were adopted in the major cities of the Yangzi delta, as this observation from the 1350s shows: The government set the unit of one guan [in paper currency] at 100 coins, but the people used various guan units ranging from forty to eighty coins. Even the units that prevailed in Wu [southern Jiangsu] differed from those in Yue [northern Zhejiang]. In Jiaxing and Huzhou, the official exchange rate of 100 coins per guan was maintained, while in Pingjiang [Suzhou] fifty-four coins were equivalent to one guan, and in Hangzhou a mere twenty coins were equivalent to one guan.14

The custom in each county was not based on the authority of the state, but rather on an unofficial consensus among traders. In other words, the circulation of coins was maintained through private, locally determined rules, as was the case for the different short-string units adopted in different commodity trades in Northern Song Kaifeng mentioned above. The Ming paper money policy differed markedly from that of the Yuan. The Ming baochao notes had no geographical limits in circulation and no 13

XGJK, 18.142a–b.

14

ZZZJ, 1, pp. 25–6.

575

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

expiry date. Nor did the baochao have serial numbers. Theoretically, the single type of baochao notes could circulate throughout the empire indefinitely. In one sense, for the first time since the fall of the Tang, the Chinese government sought to eliminate the multiple currencies circulating in separate regions of the empire. However, it also meant that the Ming government had no effective control over the circulation of its paper currency. Although state accounts continued to denominate prices in baochao notes until the early sixteenth century, in private exchange the baochao had become worthless and ceased to circulate already in the mid-fifteenth century. Consequently, provincial variations in money use – relying on unofficial currencies – emerged. A memorial from 1503 states that cowries were used as money in Yunnan in the southwest; rice, silver, and cloth in the middle Yangzi region; and animal hides in the northwestern provinces of Shanxi and Shaanxi.15 Although the Qing dynasty again made a short-lived effort to revive paper currency during the Taiping Rebellion, no state-issued paper currency won acceptance in private markets before the beginning of the twentieth century. After the Tang, no Chinese dynasty allowed free coinage, but private minting occurred all the time. Subject to the availability of raw materials, local people had no hesitation in casting private coins. The mid-Ming period in particular was noteworthy for the proliferation of private coinage. Lu Rong (1436–1494), a scholar residing in the Yangzi delta region, reported that although Ming coins still circulated in his youth, they disappeared in the latter part of the fifteenth century, and all the coins used in the marketplace bore Song dynasty reign titles.16 Contraband trade between Fujian and Japan provided private mints in coastal southern China with copper produced in Japan. “Old coins” – both genuine Tang and Song coins and freshly minted private imitations – circulated side by side in the marketplaces. Shroffing – discriminating among good and bad coins and selecting the former as acceptable means of payment – was widely practiced, not only in China but in foreign countries such as Japan that imported coin from China. Initially, shroffing was based on distinctions between standard and private coins. In the context of chronic monetary shortages, the latter often circulated as the actual mono-unit currency, while the former served as a multipleunit currency. However, as the gap in values between the two types of coins widened in the marketplace, other intermediate coins between the two classes might emerge to create a third level of monetary circulation, and as the most convenient coins depreciated even fourth and fifth levels of coins 15

MSL, Xiaozong shilu 孝宗實錄, 197.8a.

16

SYZJ, 10, pp. 122–3.

576

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

appeared. Variations in regional preferences resulted in a proliferation of local monetary standards. For example, the Tang Kaiyuan tongbao 開元通寶 coin was popularly used in the vicinity of Nanjing, but some counties favored a “small” Kaiyuan coin, while others preferred a “large” Kaiyuan coin. The distinction between small and large Kaiyuan coins suggests that these were not genuine Kaiyuan coins minted in the Tang, but rather contemporary imitations. Local customary usages preferring particular “old coins” persisted even in the early Qing period. For example, wealthy households in Putian county, Fujian, accumulated stores of Song coins, which were deemed three times more valuable than the current Qing coins.17 Under such conditions, in which the value of copper coins varied according to local customary practices, weights of silver had an advantage against even fine old coins in interregional settlements and as a means of accumulating savings. From the early sixteenth century, the use of silver as a means of exchange spread, beginning with the southern coastal areas. In 1468, a Japanese ambassador en route to Beijing recorded the prices of goods purchased upon his arrival at the port of Dinghai, in the Zhoushan archipelago near Ningbo, in terms of copper coin. However, the Japanese envoys who landed at Dinghai in 1546 recorded prices in terms of weights of silver, even for petty purchases, such as three fen of silver for baked cakes and four fen for oranges.18 While some may doubt whether they really made purchases using such small amounts (roughly one gram) of silver, in all probability they did. It was no coincidence that drastic changes in money use occurred in the Ningbo region between the mid-fifteenth century and the mid-sixteenth. From the 1540s silver produced at the Iwami mines in western Japan began to pour into the coastal regions of southeastern China in exchange for silk and other goods. The Zhoushan Islands served as the main hub of both official tributary trade and contraband smuggling between China and Japan. Later, the mainstream of silver imports shifted southward to Zhangzhou and other Fujian ports, which had convenient access to the silver brought by the Spanish galleons from the Americas to Spain’s colony of Manila in the Philippines. In the late sixteenth and early seventeenth centuries, Lima, the capital of Spanish Peru, annually imported silk and porcelain valued at 17

18

Akinobu Kuroda, “Copper-Coins Chosen and Silver Differentiated: Another Aspect of the ‘Silver Century’ in East Asia,” Acta Asiatica 88 (2005), 69, 71–2. Kuroda Akinobu 黒田明伸, “Higashi Ajia kaheishi no naka no chu¯sei ko¯ki Nihon” 東 アジア貨幣史の中の中世後期日本, in Suzuki Kimio 鈴木公雄 (ed.), Kahei no chiikishi 貨幣の地域史 (Tokyo, Iwanami shoten, 2007), pp. 31–2.

577

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

2 million silver pesos. Japanese silver also continued to flow into China in substantial quantities. According to one estimate, China imported 115 tons of silver annually from abroad during the 1600–1645 period.19 This exchange of Japanese and New World silver for Chinese silk and other goods opened a new era in global trade. A Swiss mercenary employed by the Dutch East India Company who visited China in the early seventeenth century observed that local people cut and weighed thin plates of silver to exchange for goods.20 Around the same time, a local gazetteer of Fujian recorded that in the past none ever saw silver, but now every household possessed scissors for cutting silver plate and a scale for weighing pieces of silver.21 This situation continued in the early Qing. In mid-eighteenth-century Yangzhou and Nantong, transactions in terms of weights of silver were so popular that people cut silver plate with scissors minutely enough for payments of one-thousandth of a tael.22 Bao Shichen recalled that around 1780 in Huizhou no copper coins circulated, and people paid pieces of silver even for vegetables. He described vegetable dealers who carried a scale to weigh silver, and even small shops were equipped with crucibles for melting the pieces of silver they obtained from sales of merchandise.23 Thus, to different degrees, depending on time and place, the practice of payment in weights of silver spread from the mid-sixteenth century to the mid-eighteenth to cover even everyday market transactions in many parts of China. Yet just as unofficial standards of “old” coins varied from one locality to the next, so did local units of silver. Table 15.3 shows the rice prices in Tunxi, a rural part of Huizhou prefecture (Anhui), at the turn of the nineteenth century. In this area many clans held an annual ritual to pray to their ancestors in the early spring, at the time of the Qingming festival (the third day of the third lunar month). Prices of items purchased for this ritual were recorded in account books. In one case a clan recorded prices of rice in terms of weights of silver until 1791, and subsequently switched to copper coins. Another account book compiled by the 19

20

21 22

23

Richard von Glahn, “Money Use in China and Changing Patterns of Global Trade in Monetary Metals, 1500–1800,” in Dennis O. Flynn, Arturo Giráldez, and Richard von Glahn (eds.), Global Connections and Monetary History, 1470–1800 (Aldershot, Ashgate, 2003), p. 190. Yves Giraud, Voyages et aventures aux Grandes Indes (1617–1627) (Paris, Les Éditions de Paris, 1997), p. 185. JGLBS, p. 1308. Cited in Kuroda Akinobu 黒田明伸, Chu¯ka teikoku no ko¯zo¯ to sekai keizai 中華帝国の 構造と世界 経済 (Nagoya, Nagoya daigaku shuppankai, 1994), p. 37. Cited in Kishimoto Mio 岸本美緒, Shindai Chu¯goku no bukka to keizai hendo¯ 清代中国 の物価と経済変動 (Tokyo, Kenbun shuppan, 1997), p. 359.

578

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

Table 15.3 Prices for rice purchases on the occasion of the Qingming festival, Tunxi (Anhui), 1783–1805 (prices for three dou [approximately thirty-one liters] of rice) Price in silver taels 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794

Price in copper coins

0.77 1.25 0.86 1.50 1.458 0.90 1.071 1.115 1.20 750 900 960

Price in silver taels

Price in copper coins

1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805

930 930

720 720 870

870 960

Source: Jinshan sihui shouzhi bu 金山祀會收支簿, in the archive of the Institute of Economics, Chinese Academy of Social Sciences, No. Si36 Han549

Cheng clan, which we shall analyze later, reveals that this switch was made earlier, in 1780. It might not be wrong to say that in the last quarter of the eighteenth century the use of silver as a means of payment in everyday commerce had disappeared from the Huizhou region. The information from account books confirms the retrospective statement by Bao Shichen mentioned above. We can infer that the substitution of copper coins for silver in land sale contracts, shown in previous section, followed the trend in the local grain market with some time lag. Note that Huizhou was among the last regions in which the custom of cutting and weighing silver for daily transactions persisted. Through the third quarter of the eighteenth century, copper coins replaced silver as the predominant means of payment in local markets throughout the lower Yangzi region, of which Huizhou was a part. The preference for copper coins as a means of exchange was so strong that even land sales were contracted in terms of copper coin payments. The abundance of copper coin supply at this time – as mentioned earlier, the Qianlong reign marked the second-highest peak output of state-issued coin in Chinese history – was the primary reason for the displacement of silver by copper coins in both daily transactions and large-scale purchases.

579

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

The apparent unity of a monetary system based on copper coin did not last long. Even in the last decade of the Qianlong era, debased “small coins” (i.e., private issues as well as unauthorized provincial issues in the upper Yangzi region) proliferated in coastal regions, as mentioned in the previous section. The deterioration of copper currency made Spanish silver pesos – minted according to exacting standards of weight and fineness – attractive as an alternative means of exchange. Foreign silver coins had the advantage, compared to sycee silver, of being valued simply by count (tale) – and hence were denominated as “silver dollars” (yinyuan 銀圓) – without requiring cumbersome assaying and weighing. In addition, chronic shortages of copper coins prompted the use of coin-denominated paper notes (qianpiao 錢票) issued by local merchants. Although not endorsed by public authorities nor accepted as means of state payments, foreign silver coins (confined mostly to the southern provinces) and qianpiao notes were widely used in the nineteenth century, as we shall see in detail below.

The Silver Tael as an Imaginary Unit of Account Regrettably, few scholars have focused on why the Chinese did not develop the practice of calculating the value of silver by tale rather than by weight before the nineteenth century. First of all, we should note that historically, silver, like silk, had been used as a supplement to copper coin. From the Eastern Han period, standardized bolts of silk were commonly used as means of payment for taxes and long-distance payments as well as for interregional revenue transfers (see Kakinuma’s chapter in this volume). Silver was not used for such purposes before the ninth century. During the Song dynasty, silver was increasingly used as means of payment in long-distance exchange and for large transactions. However, neither silver or silk displaced copper coin as the measure of value for setting market prices during the Song period. The Jin dynasty was the first state in Chinese history to use silver as a unit of account. However, the introduction of the silver unit of account did not mean that silver was actually used to collect taxes or make purchases. The Jin issued a silver coin in 1197, but it failed to win acceptance and was discontinued in 1200. A recently excavated metal plate for printing a local paper currency denominated in terms of silk bears a date corresponding to 1250; it presumably was issued by the Mongol rulers of the region (Shandong) before Khubilai Khan introduced the Zhongtong jiaochao paper money in 580

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

1260.24 Thus we can see that silk retained its function as money in China into the thirteenth century, yet the monetary use of silk disappeared thereafter. Although the Yuan state adopted silver as the monetary unit in government accounts, the Zhongtong jiaochao notes and subsequent Mongol paper currencies, like Song paper money, bore face values denominated in copper coins. Apart from the short-lived Jin experiment, Chinese governments never coined silver before the late nineteenth century. As we saw above, the use of silver as means of payment became so widespread by the sixteenth century that it was commonly employed in everyday exchange. Yet it was often the case that transactions were conducted using not actual weights of silver (shiyin liang 實銀兩), but rather an “imaginary” silver tael unit (xuyin liang 虛銀兩). The value of the silver tael unit was determined by three elements – weight (ping 平), fineness (se 色), and a divisor (dui 兌, e.g., 0.98 in the case of the Shanghai tael), in contrast to actual silver ingots, which were measured only by weight and fineness. People used a variety of weights, degrees of fineness, and divisors according to current circumstances. For example, Zhan Yuanxiang, a scholar living in Huizhou in 1700, recorded in his diary that he made a payment of 0.7 taels based on the chang 常 weight, 0.94 fineness and 0.9 divisor.25 Using the 0.9 divisor means that he used 0.63 liang in actual silver to make a payment of 0.7 tael (0.9 × 0.7 = 0.63). Unlike actual pieces of silver, whose value varied with the demand and supply of silver ingots, the value of an imaginary tael could be independent of the actual movement of silver ingots. Agents of the English East India Company recorded in 1820 that Chinese merchants at Canton (Guangzhou) valued silver ingots of ninety-eight liang as 100 hundred taels. H.B. Morse traced the origin of Shanghai tael, which used a 0.98 divisor, to this practice.26 We should note that the same custom of calculating monetary values in imaginary silver units (taels) had already appeared among merchants in the Tang capital, Chang’an, in the early ninth century. In this case, the silver tael unit was linked to a “short string” (duanmo 短陌) of copper coins, likewise an imaginary unit. Thus we can historically identify the silver tael as a money of account as a variation on the “short-string” custom. Since they were not coined, silver ingots and pieces were never fungible, as we can see from the account books of the Tongtaihao, a grocery business 24

25 26

Gao Congming 高聪明, “Zizhou jiaohui kao” 淄州交会考, Zhongguo qianbi 中国钱币 2016.1, 9–12. WZRJ, p. 207. Hosea Ballou Morse, The Chronicles of the East India Company Trading to China, 1635–1834 (Oxford, Clarendon Press, 1929), vol. 3, p. 374.

581

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

Table 15.4 Receipts and disbursements of silver of Tongtaihao, first day, tenth month, 1841 From

Ingot

Baoyin 393 Baoyin Bitongxing 394 Baoyin 395 Tianxianghao Baoyin 396 Longxianghao

Liang 50 50

}

50

50

Value in coins

To

Ingot

Baoyin 394 Baoyin 149,788 393 Baoyin 395 297,560 Tianxianghao Baoyin 396 146,262

Huang Xueguang

Value in Liang coins

} }

50 50 50

303,060

302,560

50

Source: Tongtaihao Daoguang ershiyinian yin churu 統泰號道光二十一年銀出入, archive of the Institute of Economics, Chinese Academy of Social Sciences, No. Qitadiqu han2 c024

active in the town of Daliuzhen, Ningjin county (Hebei), in the early nineteenth century. The account books include volumes entitled “registers of receipts and payments of silver” (churu yin liushui zhang 出入銀流水賬). Table 15.4 shows all items in this register recorded on the first day of the tenth lunar month of 1841. The entries list receipts and payments of several types of silver ingots (baoyin 寳銀, yankeyin 鹽課銀, ganbaiyin 干白銀, and baiyin 白 銀), but the majority consist of baoyin ingots weighing fifty liang. Moreover, unlike other kinds of silver, each baoyin ingot was given a serial number upon receipt, beginning with “1” for the first ingot received that year. Thus the baoyin ingot received from the Longxianghao shop was the 393rd baoyin ingot entered in the Tongtaihao account book in that year. On the same day, that same baoyin ingot was paid to another merchant, Huang Xueguang. The baoyin ingot commonly weighed fifty liang, but variations in the fineness and weight of the ingots led traders to value them at either a premium or a discount. Baoyin ingots were a popular means of payment in business transactions, but they were not all fungible – in other words, one baoyin ingot could not necessarily be exchanged for another. The silver ingot no. 393 thus moved from Longxianghao to Huang Xueguang via Tongtaihao. We can assume that, accompanying the ingot’s movement, Tongtaihao sold some goods to Longxianghao and purchased other goods from Huang Xueguang. The total value of the two silver ingots numbered 393 and 394 at the time of receipt was 297,560 wen of copper coins.

582

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

At the time of payment, however, their combined value was recorded as 303,060 wen, with the additional note “two months.” Another account book, meanwhile, specifies that there was a two-month interval between the receipt of the ingot from Longxianghao and payment to Huang Xueguang. The Tongtaihao firm separately recorded all of its transactions in terms of jingqian 京錢 (a short-string custom popular in northern China at the time, which counted forty-nine coins as 100 wen). As we will see in the next section, merchants in this town made mutual clearance through bookkeeping in terms of this short-string unit. We can take the liang units recorded in Table 15.4 to be actual weights of silver (shiyin liang). However, unlike Tongtaihao, merchants keeping accounts in terms of silver liang often used a divisor, such as the Shanghai tael unit mentioned above. An account book for expenses of seasonal ritual ceremonies by the association of fur merchants in Shanghai records all income and expense in various monetary units from 1868 to 1881.27 The ceremonies were held at the Ningbo Guild Hall, and we can assume that they reflected customary practices in Ningbo.28 On each occasion, a certain number of Mexican Republican silver dollars (then the standard market currency in Shanghai) was exchanged for copper coins, for example two dollars for 2,200 coins. The donations by merchants to the fund in many cases were recorded in terms of the Shanghai tael (based on the 0.98 divisor), yet there are no entries in the account book for exchanging liang/taels with either copper coins or silver dollars. But there are entries for converting other monies, including two cases of conversion from the Mexican dollar to the Shanghai tael. The absence of conversion from the liang reveals few actual movements of ingots in terms of the liang. The imaginary silver unit tael worked only conceptually through transfers between the accounts on books kept by associated merchants. A monetary unit decoupled from actual metallic currency facilitated book transfers among merchants. In Shanghai, the Shanghai tael mediated business transactions in this fashion, and in Chongqing (Sichuan) the Chongqing tael performed the same function. Consequently, the exchange rate between the Shanghai tael and the Chongqing tael did not always correspond to the actual flow of silver specie between the two cities. For example, if we include the transportation cost of 1,000 liang of silver – seventeen taels – between Shanghai and Chongqing, under the parity in weights of silver that 1,000 Shanghai liang was equivalent to 950 Chongqing liang, the exchange rate 27 28

Keaitang qiuye gongsuo 可愛堂裘業公所, Shanghai Municipal Archives S247-2-6-13. Transactions through book transfers already were popular among Ningbo financiers by the middle of the nineteenth century. See JHNP, p. 122.

583

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

should have fluctuated between 933 and 967 Chongqing taels. However, it often fell under 933; for example, on 1 October 1919 the exchange rate dropped to 908 taels.29 According to our conventional understanding, the exchange rate between two cities must fluctuate according to the cost of transporting specie. However, as the case mentioned above reveals, it was not uncommon that the exchange rate between different tael units sometimes fluctuated independently of the transportation costs of silver ingots between the two cities. The exchange rate in imaginary silver taels represented the different degree of transaction demand for money between two cities rather than the degree of specie supply. We should note the significant difference between actual weights of silver and imaginary silver taels.

Autonomous Local Markets and Native Monies The end users of currency in China were highly sensitive to the quality of each currency with which they dealt. Wang Huizu, who grew up in Zhejiang in the eighteenth century, heard from his mother that his father used to keep “big coins” (daqian 大錢) for the fees paid to his tutor. “It would be our shame if your teacher was rejected when redeeming his clothes at the pawnshop because of a mixture of ‘small coin’ (xiaoqian 小錢),” his father explained. Wang’s mother would select 300 “big coins” out of every 1,000 coins from her earnings weaving cotton cloth to pay his tutor’s fees.30 Wang’s memoir concerns a time in the eighteenth century when good state-minted copper coins (“big coins”) were in abundant supply. It is easy to imagine that in periods when good-quality coin was scarce the relationship between good and debased coins more seriously affected market transactions. In the late nineteenth century, the British missionary Joseph Edkins complained about the mixing of good and bad coins in the marketplaces of north China: The condition of the cash becomes worse and worse until, as in some parts of the Province of [Henan], everyone goes to market with two entirely distinct sets of cash, one of which is the ordinary mixture of good with bad, and the other is composed exclusively of counterfeit pieces.31 29

30 31

“Ju¯kei saikin kin’yu¯ jijo¯” 重慶最近金融事情, Tsu¯sho¯ ko¯ho¯ 通商公報 588 (1919), pp. 219–20. Wang Huizu 汪輝祖, Shuangjietang yongxun 雙節堂庸訓, in WLZYS, 2, pp. 777, 785. North China Herald, 1889, vol. 411, cited in Eduard Kann, The Currencies of China (Shanghai, Kelly & Walsh, 1927), p. 416.

584

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System 120 100 80 60 40 20

25 28

22

19

16

7 10 13

4

th on

ay

6t

h

m

25 28

19

16

7 10 13

4

22

td 1s

1s

td

ay

5t

h

m

on

th

0

Figure 15.1 Frequency of daily transactions of the Tongtaihao firm, fifth–sixth lunar months, 1830 Source: Tongtaihao Daoguang shinian siyue churu liushui zhang 統泰號道光十年四月出入 流水賬 and Daoguang shinian liuyue churu qian liushui zhang 道光十年六月出入流水賬 (National Library of China), Wenjinge, nos. 49120–44, 145

Around the same time, Arthur Smith vividly described how traders in Chinese villages spent several hours after closing their shops every evening counting all of the copper coins in their possession.32 Merchants closely examined the quality of each coin and separated the good ones from the bad. Much trade in rural areas was conducted at periodic markets. In the Tongtaihao account books, the frequency of transactions surged on the second and seventh days of each ten-day Chinese week (Figure 15.1). The firm’s “register of receipts and payments of coin” (churu qian liushuizhang 出入錢流水賬) account book (hereafter “cash book”) appears to have been used to record all Tongtaihao transactions except for miscellaneous daily expenditures. Most of the customers in the cash book entries on the second and seventh days are individuals who rarely appear twice in the same year. In contrast, most of the merchant names that appear on other than the second and seventh days were from the town, Daliuzhen, in which the firm was located. The differences by day strongly suggest that the town had a five-day market held on the second and seventh days and the market attracted a large number of individual peddlers. We should note that peddling was an 32

Arthur H. Smith, Village Life in China: A Study in Sociology (New York, F.H. Revell Co., 1889), p. 51.

585

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

important side business for peasants. The individual persons appearing in the cash book of the Tongtaihao must be the dealers known as xiaofan 小販. Other studies of rural markets confirm that payments in coin dominated local retail trade. Sometimes Tongtaihao sold on credit and purchased on debt with the peddlers. Even in those cases, clearly, there was a tendency for both sides to settle as late as the next market day, or five days later. No interest appears to have been charged on such deferred settlements. However, even if sales credit existed between the grocery firm and the peddlers, deferred payments comprised only a small portion of total transactions. For example, during the month between the eighteenth day of the fifth month of 1830 and the seventeenth day of the sixth month of 1830 only 918,554 wen of credit transactions with peddlers were recorded out of a total of 23,616,946 wen in transactions. In addition, among fifty-seven persons who made deferred payments to Tongtaihao during this period, only three reappear in the account book for deferred payments in the same period of the next year. Between the twelfth day of the first month of 1830 and the twelfth day of the seventh month (seven months that included an intercalary lunar month), 420 persons appeared only once among the 483 deferred payments recorded in the account book.33 Thus we can see that cash transactions were so dominant that sales credit rarely was repeated between the same buyer and seller. The preference for one-time transactions was particularly strong in the case of transactions in periodic markets. This finding coincides with observations through modern rural investigations that peasants (and peddlers) typically made no subsequent transactions with particular merchants. The prominence of cash in retail trade suggests that a sufficient supply of cash in denominations suitable for everyday purchases was indispensable for local exchange. This strong preference for cash created a discrepancy between close and distant exchanges that correlated to the two levels of marketing associated with silver and copper coins. Table 15.5 shows the number of silver ingots received by Tongtaihao in nine different years from 1821 to 1848. After the peak in 1833, when 968 silver ingots were received, the number dropped to 168 in 1846, less than one-fifth of the 1833 level. Tongtaihao recorded the silver receipts in terms of jingqian (the local unit of copper coins), and the registers reveal that the conversion rate 33

Tongtaihao 通泰號, Daoguang shinian zhengyue zanfuji zongzhang 道光十年正月暫浮 記總賬 and Daoguang shiyinian zhengyue fuji zongzhang 道光十一年正月浮記總賬 (Chinese National Library), Wenjinge, no. 49120–66, 168.

586

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

Table 15.5 Annual baoyin silver ingot receipts of the Tongtaihao firm, 1821–1848 Year

Number of ingots

1821 1823 1828 1830 1831 1833 1841 1846 1848

255 133 509 361 528 968 504 168 176

Source: Tongtaihao zhangbu 統泰號帳簿 (National Library of China), Wenjinge, no. 131001–1, 2, 4, 5, no. 49120–42, 148, 153; (Institute of Economics, Chinese Academy of Social Sciences), no. Qitadiqu han2 c024, c029

from silver to copper coin changed day by day, and sometimes even within the same day. One liang of silver exchanged for 2,590 wen in jingqian in 1821, 2,820 wen in 1830, 2,970 wen in 1841, 3,580 wen in 1844, and 4,800 wen in 1848. (All figures are taken from the rate obtaining in the early tenth month of each year. According to the jingqian custom, one copper coin was counted as two wen.) Over this period of nearly thirty years, when silver was flowing out of China in significant quantities, the value of copper coins in terms of silver declined by nearly half. However, the contraction of silver circulation did not always directly affect the business between local merchants. Let us take another look at the Tongtaihao cash book. Table 15.6 shows thirteen transactions among a total of twenty-one entered on the income side and eleven transactions among twenty-one entered on the expenditure side in the cash book for the first day of the tenth month of 1841. Note that in the first nine entries equal sums appear in both the income and expenditure columns, and also that the total sum of receipts from Chengli Fashenghao, Jushenghao and Huanghesheng in the last three transactions on the income side is equal to the amount paid to Tianxianghao in the last entry on the expenditure side. These entries suggest that these transactions may have been made not through the exchange of actual currency, but through transfers recorded in

587

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

Table 15.6 Excerpts from income and expenditure entries of Tongtaihao for the first day of the tenth month, 1841 From Zengtaidian From Simeidang From Yandian From Hufang From Longtaihao From Tianxianghao From Shuangchengdia From Wangfasheng From Huanghesheng From Jushenghao From Chengli Fashenghao From Jushenghao From Huanghesheng

2,000 1,480 4,000 1,28000 5,500 150,000 40,000 8,000 9,000 40,000 28,000 50,000 72,000

To Tongxiyoufan To Liubingchun To Tangji To Yuanshuzhai To Yuanqingao To Yongshunhao To Jiangtongsheng To Sun To Tongdian youfang To Longxianghao

2,000 1,480 4,000 128,000 5,500 150,000 40,000 8,000 9,000 40,000

To Tianxianghao

150,000

Source: Daoguang ershiyinian bayue churu liushuizhang 統泰號道光二十一年八月出 入流水帳 (National Library of China), Wenjinge, no. 131001–10

the merchants’ account books. For example, Tongtaihao received a commodity valued at 2,000 wen from the Zengtai shop, and on that same day paid the same amount to the Tongxi oil producer. Probably, Zengtai likewise recorded 2,000 wen in the expenditure side of their account books, while Tongxi recorded 2,000 wen in the income side of their books. It was not necessary to use currency as long as the amounts of the two transactions were exactly the same. Importantly, these operations worked not just bilaterally but multilaterally. This is the custom called guozhang 過賬 among Chinese merchants. Most of the merchants appearing in the Tongtaihao account books presumably conducted business in the same town of Daliuzhen. In cases in which the grocery firm engaged in transactions with traders from outside the town, the place name is recorded in front of the name of the merchant shop, such as the Fashenghao shop in Chengli. Notably, in business dealings with outside merchants, silver was mostly, though not always, used to settle the transaction. Thus we can assume that the contraction of silver flows at this time had less effect on business among merchants who utilized book transfers to settle their accounts multilaterally than on those engaged in interregional trade, who were inclined to bilateral settlements in cash. Figure 15.2 shows the movement of rice prices in the rural county of Tunxi (Anhui) during the period from 1821 to 1860, based on the account books of 588

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System 300 250 200 150 100 50 0 1821

1825

1830

1835

1840

1845

1850

1855

1860

Figure 15.2 Annual fluctuations of rice prices, Tunxi county, Anhui, 1821–1860 (wen per five sheng [approximately five liters]) Source: Chengshi qingming jisi bu 程氏清明祭祀簿 (Institute of Economics, Chinese Academy of Social Sciences), no. Si39, B105

a local clan for purchases for the Qingming festival. The rice prices are given in copper coins. As the Tongtaihao account books show, copper coins suffered serious depreciation relative to silver in this period. However, in Tunxi, despite the depreciation of coin, coin prices for rice remained surprisingly stable for almost half a century. Basically, interregional transactions were conducted in terms of silver. Thus the appreciation of silver must have hampered long-distance trade. Moreover, since tax obligations were calculated in terms of silver, the shortage of silver caused difficulties for taxpayers who were obliged to remit tax payments in increasingly dear silver. However, considering that rural households mostly made purchases in local markets, as shown in the 1930 data from Qingyuan county in Hebei (Table 15.7), they would have benefited from price stability in local trade. In this case rural households sold grain mostly for export out of the district, but other agricultural products were traded entirely within the local area. The total amount of both sales and purchases within the local market was greater than those outside the locality. In other words, rural dwellers traded primarily with other local inhabitants rather than outside merchants, and thus required means of payment of relatively small value that would be appropriate to the small scale of these transactions. What constituted a “local” market for money use in China? Obviously, even the county was not the minimum unit for adjusting local monetary supply and demand. Wang Huizu wrote in 1799 that in the far eastern part of his native county of Xiaoshan (Zhejiang), people counted 100 private coins as

589

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

Table 15.7 Annual sales and purchases of farm households in Qingyuan county, Hebei, 1930 (figures in yuan) Sales

Grain Other agricultural products Other Total

Purchases

Within district

Outside district

Within district

Outside district

11.22 43.28

52.90 4.80

11.22 43.28

0 0

28.22 87.72

0 57.70

28.22 87.72

35.55 35.55

Source: Cui Xiaoli 崔晓黎, “Jiating, shichang, shequ: Wuxi Qingyuan nongcun shehui jingji bianqian de bijiao yanjiu (1929–1949)” 家庭·市场·社区: 无锡清苑农村社 会经济变迁的比较研究, Zhongguo jingjishi yanjiu 中国经济史研究 1990.1, 57

ninety official coins, while people in the rest of the county considered 100 private coins as equivalent to ninety-five official coins. Given this regional disparity, people in the eastern section would have suffered a loss if they brought official coins to the county seat.34 The situation in Tongshan county (Jiangsu) in the lower Yangzi region around 1930 reveals the complexity of local money use caused by the demand for small-denomination currencies and the circulation of qianpiao notes.35 Here as elsewhere a wide variety of one-yuan currencies – including Yuan Shikai silver dollars (issued briefly by the central government in 1914), Mexican silver dollars, Bank of China notes, and Bank of Communication notes, among others – circulated in all parts of the county. However, there were also a great variety of smaller-denomination currencies of different kinds in circulation within the county. Qianpiao notes issued by local merchants in denominations of 50, 100, 200, 300, 500, and 1,000 wen accelerated the multiplicity of currency circulation. There were two kinds of qianpiao: one type was issued by moneychangers and reputable firms with substantial assets located in the county seat, and another by shops in rural areas. The former circulated across the entire county, while the sphere of circulation of the latter was more limited, approximately forty to fifty square kilometers. Tongshan was by no means exceptional, but rather broadly representative of local monetary systems throughout China. Government officials had 34 35

Wang Huizu 汪輝祖, Menghen yulu 夢痕錄餘, in WLZYS, 2, pp. 642–3. Jin Weijian 金維堅, Tongshan nongcun jingji diaocha 銅山農村經濟調查 (Zhenjiang: Peasant Bank of Jiangsu Province, 1931), pp. 23–4.

590

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

already recognized the popularity of qianpiao currencies already in the early nineteenth century, but we must look to later observations for details on how they circulated. The town of Nanzhaoji in a rural part of Yingshang county (Anhui) was flooded by privately issued qianpiao in 1936. More than forty shops in the town issued several hundred dollars’ worth of their own notes. In many cases, shops issued notes whose total value was several times greater than their assets. The issuers claimed that their notes were convertible with hard currencies or legal tender, but in fact this was not true. Locals complained that when bearers brought notes issued by shop A to shop A, they would not be redeemed in cash but rather in notes issued by shop B, and if the bearer brought the notes issued by B to shop B, shop B would give the bearer notes issued by shop C. Despite the inconvertibility of these paper notes, they circulated as far as twenty to thirty kilometers away from the town.36 We may consider the walkable distance from a rural market as the minimum range of circulation for a local currency in traditional China.

Conclusion: Varieties of Monetary Economy The exchanges mediated by money did not always share the same attributes. It is not easy to mediate commerce among long-distance traders and everyday exchange in local markets with the same monetary medium. One-time transactions in open markets require cash for settlement, while repeated transactions between familiar parties tend to obviate the need to exchange cash. Monetary systems also vary according to different types of exchange. The monetary system of late imperial China primarily facilitated one-time transactions among locals in open markets. The predominance of mono-unit currencies made of non-precious metals demonstrates that the major portion of transactions involved sales and purchases of everyday items of consumption among ordinary people. As far as local exchange was concerned, it might not be wrong to say that in China one-time transactions in open markets were more prevalent than repeated transactions between familiar parties. Spot transactions among ordinary people required currencies whose denominations were sufficiently small to facilitate petty exchange. Unlike currencies made of precious metal, base-metal coins cost too much for issuers to expect seigniorage profits. If Chinese governments were determined to maintain a full-bodied currency standard, they often were unable to 36

Anhui qudi ge difang shanghao sifa zhibi 安徽取締各地方商號私法紙幣 (Academia Historica 國史館), no. 018000027642A.

591

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

mint sufficient quantities of coin, and the scale of minted output fluctuated significantly over time. In addition, low-value coins tend to stagnate in local markets; in other words, they have a high degree of viscosity in comparison with high-value coins, and did not readily flow in response to shifts in demand. Thus, at the local level, Chinese people suffered from endemic shortages of currency. Consequently, end users did not hesitate to establish their own currencies for local uses. People often separated monies for spot exchanges from those intended as stores of value or for long-distance exchange. They carefully sifted through stocks of coin and selected different coins for different uses, as Wang Huizu’s memoir cited above shows. Professional merchants conducted their business in a more sophisticated way. Some merchants in the same town shared account books to transfer assets. Through multilateral clearance among a cohort of merchants, they could eliminate the need to make transfers in cash. Regardless of whether the account books were kept in units of copper coins or weights of silver, the units of account were often separated from the intrinsic metal content of actual circulating currencies. The units of account were linked to circulating currencies by devices such as the “short-string” unit used for copper coins or a divisor in the case of weights of silver. Book transfers were conducted without resort to cash payments, yet this system was not credit-oriented but rather clearance-oriented, since merchants had a tendency to leave no unpaid balance. Merchants engaging in long-distance trade of a particular commodity shared a common currency for settlements. For example, rice merchants in Wuhu, Anhui – a major grain emporium on the Yangzi river – continued to make transactions in long-defunct Spanish silver pesos as late as the 1910s. Throughout the nineteenth century the lower Yangzi region exported large quantities of rice to Guangdong in exchange for Spanish pesos that were in common circulation there. Even after the disappearance of Spanish pesos from market exchange, merchants engaged in rice exports to Guangdong kept their accounts in Spanish dollars. At the Yangzi river port city of Jiujiang (Jiangxi), porcelains from Jingdezhen were traded in Mexican dollars, while cotton textiles from Ji’an (in western Jiangxi) were traded in Japanese silver yen. Thus various combinations of particular commodities and particular currencies – which I have designated “currency circuits” – could be found in many parts of China. Although currency circuits were a common feature of monetary systems throughout world history, they were an especially distinctive feature of the Chinese economy in the nineteenth century. 592

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

The official story related in Chinese dynastic histories depicted the monetary system in terms of a superior, state-issued currency versus inferior private currencies. Only state-issued currencies had the sanction of legal tender. In reality, numerous forms of money were created from the bottom up. Stateissued currencies at best established a standard that private currencies must emulate but not necessarily imitate. As we have seen, in the Song period, when the official short-string (shengmo) consisted of seventy-seven copper coins, merchants established their own short-string conventions that contained fewer than seventy-seven coins. Chinese monetary history was a different world from one in which the state could force subjects to accept an inferior currency to extract profit from seigniorage, or one in which people accepted the official currency because of its utility in state payments. “Giving convenience to the people” (bianmin 便民) was frequently invoked by government officials to justify minting coin even at the risk of incurring substantial losses. Throughout Chinese history, resort to minting inferior and/or underrated coins often signaled the impending collapse of the dynasty. In contrast to the conventional image of the imperial Chinese state as possessing absolute power, monetary circulation was actually far beyond the government’s control. Coins basically flowed in one direction, from the mints to end users. Only a portion of the coins in circulation returned to governmental coffers as tax payments. The paper monies issued by the Song and Yuan states returned to the government after circulation in greater proportion than did coins. However, this tendency to return paper money to the government was only possible in tandem with other local currencies that met the needs of end users. At various times Chinese governments collected cloth and silver in tax payments, but they were not issued in any specific form for circulation designed by governments as were copper coins and paper monies. We should distinguish between a currency designed to circulate as a means of exchange and one that served as a measure of value. Chinese governments issued copper coins with the intention that they would function as means of exchange, while designating (in the Ming–Qing periods) particular units of silver for collecting revenues and making disbursements. In the late imperial period, silver inherited the functions of measure of value and means of tax payment performed by silk in the early imperial period. Foreign silver coins that began to circulate within China from the late eighteenth century substituted for copper coins rather than sycee silver. Because of the difference in functions, the 593

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda

exchange rate of silver dollars against sycee silver fluctuated from day to day and place to place. Sycee silver and silver dollars were recognized as made from the same material, but as monetary instruments they worked quite differently. In Shanghai the exchange rate of silver dollars to the Shanghai tael rose every year in early summer because the dollars flowed out to the sericulture regions. It might sound strange to say that one form of silver appreciated against another form of silver. In effect, the demand for means of payment (silver dollars) for cocoon producers pushed down the value of the monetary unit (the Shanghai tael) used by merchants in Shanghai to settle their accounts. Let us return to the three characteristics of the Chinese monetary system identified in the introduction. The prevalence of low-value fractional currencies reflected the primacy of the demand for money in local exchange. From the founding of the first empires, bulky, lowvalue copper currency required complementary monetary media such as silk for long-distance settlements. Under the Northern Song, the combination of regional circulation of even more bulky iron coins and the use of state-issued commodity certificates for interregional settlements laid the foundations for the creation of state-issued paper currencies. By the fourteenth century, weights of silver had replaced bolts of silk in long-distance settlements, and silver became indispensable after the collapse of the official Ming paper money. However, weights of silver were unsuited for everyday market transactions, although imaginary units of silver worked well as units of account for bookkeeping transfers among merchants. At the local level, people usually made one-time transactions for which they required low-value fractional currency. When the supply of full-bodied state currencies proved insufficient, they frequently resorted to locally manufactured private currencies. End users also shared customary practices within the sphere of exchange in which they negotiated transactions in person. The multiplicity of local means of exchange throughout Chinese imperial history shows that such customary monies could fulfill the needs of exchange within narrow local spheres of trade, but their acceptability could not be extended beyond the locality. Thus circuits characterized by distinct but complementary currencies flourished as trade developed. Although the operation of this monetary system appeared chaotic from a modern perspective, an inherently logical order was at work endogenously behind the scenes. 594

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

The Monetary System

Further Reading Chen Zhaonan 陳昭南, Yongzheng Qianlong nianjiande yinqian bijia biandong 雍正乾隆年 間的銀錢比價變動 (Taipei, Zhongguo xueshu zhuzuo jiangzhu weiyuanhui, 1966). Dai Jianbing 戴建兵, Zhongguo jindai yinliang shi 中國近代銀兩史 (Beijing, Zhongguo shehui kexue chubanshe, 2007). Gao Congming 高聰明, Songdai huobi yu huobi liutong yanjiu 宋代貨幣與貨幣流通研究 (Baoding, Hebei daxue chubanshe, 1999). Ichiko Sho¯zo¯ 市古尚三, Mindai kahei shi ko¯ 明代貨幣史考 (Tokyo, O¯tori shobo¯, 1977). Kato¯ Shigeshi 加藤繁, To¯–So¯ jidai ni okeru kingin no kenkyu¯ 唐宋時代に於ける金銀 の研 究 (Tokyo, To¯yo¯ bunko, 1943). Kishimoto Mio 岸本美緒, Shindai Chu¯goku no bukka to keizai hendo¯ 清代中国の物価と 経済変動 (Tokyo, Kenbun shuppan, 1997). Kuroda, Akinobu, “Anonymous Currencies or Named Debts? Comparison of Currencies, Local Credits and Units of Account between China, Japan and England in the Pre-industrial Era,” Socio-economic Review 11.1 (2013), 57–80. Kuroda, Akinobu, “The Collapse of the Chinese Imperial Monetary System,” in Kaoru Sugihara (ed.), Japan, China and the Growth of the Asian International Economy, 1850–1949 (Oxford, Oxford University Press, 2005), pp. 103–26. Kuroda, Akinobu, “Concurrent but Non-integrable Currency Circuits: Complementary Relationship among Monies in Modern China and Other Regions,” Financial History Review 15.1 (2008), 17–36. Kuroda, Akinobu, “The Eurasian Silver Century, 1276–1359: Commensurability and Multiplicity,” Journal of Global History 4.2 (2009), 245–69. Kuroda, Akinobu, A Global History of Money (London, Routledge, 2020). Kuroda, Akinobu, “Strategic Peasant and Autonomous Local Market: Revisiting the Rural Economy in Modern China,” International Journal of Asian Studies 15.2 (2018), 195–227. Maeda Naonori 前田直典, Gencho¯ shi no kenkyu¯ 元朝史の研究 (Tokyo, To¯kyo¯ daigaku shuppankai, 1973). Miyashita Tadao 宮下忠雄, Chu¯goku heisei no tokushu kenkyu¯: Kindai chu¯goku ginryo¯ seido no kenkyu¯ 中国幣制の特殊研究: 近代中国銀両制度の研究 (Tokyo, Nihon gakujutsu shinko¯kai, 1952). Miyazaki Ichisada 宮崎市定, Godai–So¯sho no tsu¯ka mondai 五代宋初の通貨問題 (Kyoto, Hoshino shoten, 1943). Miyazawa Tomoyuki 宮澤知之, “Gencho¯ no zaisei to sho¯” 元朝の財政と鈔, Bukkyo¯ daigaku rekishigaku ronshu¯ 仏教大学歴史学論集 2 (2012), 43–64. Miyazawa Tomoyuki 宮澤知之, So¯dai Chu¯goku no kokka to keizai: Zaisei, shijo¯, kahei 宋代 中国の国家と経済: 財政,・市場,・貨幣 (Tokyo, So¯bunsha, 1998). Peng Xinwei 彭信威, Zhongguo huobi shi 中國貨幣史, 2nd ed. (Shanghai: Shanghai renmin chuabanshe, 1965). Takahashi Hiroomi 高橋弘臣, Gencho¯ kahei seisaku seiritsu katei no kenkyu¯ 元朝貨幣政策 成立過程の研究 (Tokyo, To¯yo¯ shoyin, 2000). Thierry, François. Les monnaies de la Chine ancienne, des origines à la fin de l’empire (Paris, Belles-Lettres, 2017).

595

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

akinobu kuroda Vogel, Hans Ulrich, Marco Polo Was in China: New Evidence from Currencies, Salt, and Revenues (Leiden, Brill, 2013). von Glahn, Richard, “Foreign Silver Coins in the Market Culture of Nineteenth-Century China,” International Journal of Asian Studies 4.1 (2007), 51–78. von Glahn, Richard, Fountain of Fortune: Money and Monetary Policy in China, 1000–1700 (Berkeley, University of California Press, 1996). von Glahn, Richard, “Monies of Account and Monetary Transition in China, Twelfth to Fourteenth Centuries,” Journal of the Economic and Social History of the Orient 53.3 (2010), 463–505. von Glahn, Richard, “Origins of Paper Money in China” in K. Geert Rouwenhorst and William N. Goetzmann (eds.), Origins of Value: The Financial Innovations That Created Modern Capital Markets (New York, Oxford University Press, 2005), pp. 65–89. Wang Shengduo 汪聖鐸, Liang Song huobishi 兩宋貨幣史 (Beijing, Shehui kexue wenxian chubanshe, 2003). Wang Wencheng 王文成, Songdai baiyin huobihua yanjiu 宋代白銀貨幣化研究 (Kunming, Yunnan daxue chubanshe, 2001).

596

https://doi.org/10.1017/9781108587334.017 Published online by Cambridge University Press

16

Merchants and Commercial Networks joseph p. mcdermott

Over the past two generations a fundamental change has taken place in the scholarly understanding of the commercial world of late imperial China. Lasting from the Song (960–1279) to the end of the Qing dynasty (1644–1911), this millennium of Chinese history had long been judged a period of decline, its initial economic breakthroughs never fulfilling their promise. The commercial and technological innovations of the eleventh and twelfth centuries were thought to have given way to economic stagnation and cultural conservatism, as the enterprising peasantry and merchants of south China lost out to the prerogatives of Confucian scholar-officials and their statesponsored culture in the Ming (1368–1644) and Qing dynasties. Tested by a highly competitive examination regime and thereafter sheltered by a host of privileges, these scholar-officials acquired and retained an unrivaled hegemony that was cultural, political, and, some would add, economic. When China suffered a severe economic downturn during the nineteenth and twentieth centuries, the once-admired stability of the Qing regime was criticized for its backwardness, and the late imperial economy of these scholar-officials’ rule was condemned for its stagnation. In recent decades, however, this long-standing wisdom has come under sharp attack. Historians inside and outside China have discovered an expansion of the late imperial economy so strong, if not continuous, that it is hard to believe it was ever overlooked. China’s population is understood to have quadrupled from roughly 100 million c. 1077 to 383 million in 1820, thanks largely to an expansion of arable land in the north and especially the southwest and more intensive cultivation of paddy fields in the southeast. With more grain crops produced for both local and distant consumption,

597

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

long-distance transport flourished, particularly in the great river valleys of central and south China. State monopolies such as the salt trade and silk weaving factories for the court helped to kick-start a revival of Song-style economic expansion from the mid-Ming. But it was the private market in grain, cotton, natural resources, and handicraft production that constituted the bulk of Chinese transactions from 1500 to 1800. As scholars discovered how new commercial and financial institutions grew up to handle this expanding trade, it is little wonder that they began to divert some of the attention traditionally devoted to scholar-officials onto late imperial merchants, especially the long-distance “sojourning merchants” (keshang 客商) who oversaw this great expansion in the commercial and financial sectors. Disparaged since antiquity as a self-interested pursuer of pleasure and profit, the merchant occupied the bottom of a four-rank occupational order of scholar, farmer, artisan, and merchant that most Confucians assumed was moral, natural, and essential to social and economic stability. The merchant’s recent rise to greater historiographical interest marks a sharp reversal of the Neo-Confucian status system and a belated recognition of his long-term significance in the development of late imperial Chinese society and economy. While not yet revered as was the scholar-official, he is nonetheless now recognized to have made key contributions to three extended periods of Chinese economic expansion in the last millennium, namely the early and mid-Song, the latter half of the Ming, and the mid-Qing. In surveying the eight-century stretch of these merchant contributions, this chapter risks reducing a complex story of various regional groups and economic changes into a one-dimensional narrative. To be sure, certain concerns will predominate – the persistent drive toward commercialization, the seemingly permanent predominance of the southern over the northern economy, and merchant agility in seizing trading opportunities and establishing both human and institutional networks. But underlying this account of increasing integration of China’s national economy are three simple questions that will require a diverse set of answers: (1) to what extent and how did these merchants establish a network of markets for their transactions throughout the empire? (2) how did they practice business in these market networks? and (3) what kind of working relationship did they establish with the scholar-official rule supposedly so antagonistic to their commercial practices and values? In providing a historically informed response to these questions, this chapter will discuss in turn the long-term expansion of the Chinese domestic market, merchant solutions to the problems encountered in operating in these markets, the types of commercial organization they 598

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

formed and adopted in line with their family organizations, and their evolving relation to the late imperial state, particularly at the local level. As these merchants emerge from beneath the shadow of the scholar-officials, they will appear in this account as an unexpected mixture of avid trader, cautious investor, and cunning official; that is, a shrewd manipulator of markets and men. Having penetrated the vastness of the Chinese interior and established not just markets but institutions capable of regularly carrying out both small and large transactions, by 1800 they had emerged as individual actors better prepared for international competition than Chinese historians have usually acknowledged. In fact, they were far better prepared than the scholar-officials who ruled over them and all too often misruled their country. The first section will describe the growth of the market as an institution in late imperial China, describing its quantitative and qualitative changes, and the importance of trading networks and of transport nodes such as Linqing on the Grand Canal and the cotton market towns in the Yangzi delta. The second section will consider how this market growth affected longterm changes in the distribution pattern of three notable commodities – grain, cotton textiles, and books. The third section will then discuss several organizational and economic problems merchants faced – security, market access, affordable capital, weak credit institutions, and the operation of commercial partnerships – and the solutions they devised in a commercial world full of risks. The focus will fall on long-distance sojourning merchants, who took advantage of the commercial booms of the late Ming and mid-Qing periods to expand their trade and establish markets and other commercial institutions throughout China. The fourth and final section will deal directly with an ongoing issue of Chinese history that is implicitly addressed in the three preceding sections: the merchants’ relationship to the imperial state, or rather the range of relationships that the state and merchants entertained with one another. If the merchant’s relationship to Chinese society and economy at this time was complex, then his relationship to the state was doubly so. This relationship is difficult to study, not simply because of the great variety of merchants but also because the transactions between these two parties were seldom recorded and made public. Virtually all our records come from the official side. This galling absence of merchant archives such as can be found for Italian or Dutch merchants throughout most of this period was no accident. It was the consequence of political decisions that for nearly the full length of this period restricted the institutional autonomy of merchants as an occupational 599

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

group and that even afterwards shaped their unequal interactions with the government. Their dealings thus were seldom part of the public record, and even though the social divide between officials and rich merchants during this period narrowed as more and more officials came from rich merchant families, their relationship would be marked repeatedly by silent collaboration, collusion, and corruption, and only sometimes by opposition.1

Market Growth and Product Distribution From the eleventh through the eighteenth century, the number and distribution of markets steadily expanded, as revealed in four surveys dating from c. 1077, 1429, c. 1580, and 1792.2 These surveys, the first two governmental and the second two private, all aimed at locating the sites of significant commercial activity and their distribution within the empire. Despite self-evident differences and serious inadequacies – some are primarily concerned not with markets per se but with commercial tax stations and thus administrative rather than economic units – they provide rough figures indicative of chronological and geographical shifts in overall market activity. Although they say little of market volume, together they constitute the best series of statistical overviews we have of late imperial market activity, especially when they are combined with more detailed figures recently calculated for markets and market towns in the middle and lower Yangzi valley during the Ming and Qing. Thus, whereas in c. 1077 the Song dynasty drew up a list of 2,060 commercial tax stations (plus 20,606 additional sites for collecting salt and wine levies), recently compiled figures from the late Ming–early Qing periods show 7,682 markets from just twelve provinces, and an 80 per cent increase over that figure by the late Qing to 13,993 (Table 16.1). This explicitly incomplete late Qing figure for fewer than three-fifths of Qing China’s twenty-one provinces gives support to Fang Xing’s empire-wide estimate of no fewer than 22,000 markets in the late Qing.3 The reports for markets in intervening 1

2

3

Ping-ti Ho, The Ladder of Success in Imperial China (New York, Columbia University Press, 1965), pp. 41–2, 77–8; Lien-sheng Yang, “Government Control of Merchants in Traditional China,” in Yang, Sinological Studies and Reviews (Taipei, Shih-huo, 1982), pp. 27–47. Joseph P. McDermott, The Making of a New Rural Order in South China, vol. 2, Merchant, Market, and Lineage in Huizhou, 1500–1700 (Cambridge: Cambridge University Press, 2020), Chapter 1; SGBL, 3.32a–33b. Fang Xing 方行, Jing Junxian 经君键, and Wei Jinyu 魏金玉, Zhongguo jingji tongshi – Qing 中国经济通史–清, vol. 2 (Beijing: Zhongguo shehui kexue chubanshe, 2007), p. 778.

600

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

Table 16.1 Markets and market towns in twelve provinces in the Ming and Qing Ming and early Qing

Late Qing

Province

Markets and Percentage Markets and market towns of total market towns Change

Hubei Hunan Jiangxi Jiangsu Zhejiang Sichuan Anhui Guangdong Fujian Zhili Shandong Shaanxi Total

700 384 672 1,112 605 135 708 1,270 205 527 1,126 238 7,682

9.1 5.0 8.7 14.4 7.9 1.8 9.2 16.5 13.7 10.1 14.7 3.1

1,918 433 1,305 1,351 734 2,655 671 1,969 225 826 1,555 351 13,993

+274.0% +12.8% +94.2% +21.4% +21.3% +19.7% –5.23% +55.0% +9.75% +52.7% +38.1% +47.5% +82.2%

Percentage of total 13.7 3.1 9.3 9.6 5.2 19.0 4.8 14.1 1.6 5.9 11.1 2.5

Note: table omits nine provinces, three in the north (Shanxi, Gansu, and Xinjiang), three in the south (Guizhou, Yunnan, and Guangxi), and the three Manchurian provinces of Fengtian, Jilin, and Heilongjiang. Source: Ren Fang 任放, “Ming Qing shizhen de fazhan zhuangguang ji qi pingjia zhipiao tixi” 明清市镇的发展状光及其评价指标体系, in Chen Feng 陈锋 (ed.), Ming Qing yilai Chang Jiang liuyu shehui fazhan shilun 明清以来长江流域社会发展 史论 (Wuchang, Wuhan daxue, 2006), pp. 187–228, esp. 201

centuries – that is, the reports for 1429, c. 1580, and 1792 – are comparatively impressionistic and far less complete. The first lists just the top thirty-three “places where merchants converge,” and those for 1429 and 1792 provide an empire-wide list of the major markets with no attempt to be comprehensive. Nonetheless, when interpreted with care, they too point to an overall growth in the empire’s number of marketing sites between 1429 and 1580 as well as between 1580 and 1792. The only pair of these reports that together bookmark a period of overall decline in market activity date from c. 1077 and 1429, a decline most evident in Sichuan and the Huai river valley, in the virtual disappearance of coastal ports, and in the sparseness of markets in north China as a whole. Elsewhere in the 1429 list, two key features of the c. 1077 survey persist: the concentration of major markets along the Grand Canal and the Yangzi river and their predominance in the lower Yangzi valley.

601

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

From 1429 to 1580, by contrast, the signs of market expansion throughout south China are dramatic, with many more markets reported from the middle Yangzi valley (especially Hubei and Jiangxi), Guangdong in the far south, and the southwestern provinces of Yunnan, Guangxi, and Sichuan. Arguably more significant, however, were structural changes to some markets and market towns in the delta. The trading centers of Suzhou and Songjiang prefectures, in addition to the considerable increase in their numbers (Table 16.2), were transformed into “clusters” of neighboring market towns involved in different stages of the production process of cotton and silk textiles. In these sites, merchants, brokers, and their representatives met with home-based producers of yarn and cloth. Most merchants made little Table 16.2 Markets and market towns in Suzhou and Songjiang prefectures in the Ming and Qing

Suzhou counties Changshu Wu Wujiang Kunshan Taicangzhou Jiading Baoshan Suzhou total Songjiang counties Qingpu Shanghai Nanhui Chuansha Fengxian Songjiang Chuanshan Songjiang total Overall total

Pre-1551 Ming

1551–1722

1723–1861

24 18 16 7 20 12 5 102

43 21 15 14 26 20 9 148 (+45%)

46 26 17 19 25 22 12 157 (+9%)

15 12 6 1 8 10 7 59

20 26 13 1 14 20 9 113 (+92%)

30 27 33 5 23 33 16 167 (+48%)

161

261 (+62%)

334 (+28%)

Source: Fan Shuzhi 樊树志, Ming Qing Jiangnan shizhen chuantong de biange 明清江南 市镇传统的变革 (Shanghai, Fudan daxue chubanshe, 2005), Tables 14, 15, pp. 150–1

602

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

investment in the production process other than to pay workers (or loom owners) for the cloth they wove or dyed. But some, if only as pawnbrokers, acquired ownership of looms that they then lent to weavers obliged to work off their debt through cloth production. For indications of the spread of the market between the surveys of c. 1077 and c. 1580 we have much additional quantitative and nonquantitative regional evidence. During the interval between c. 1077 and c. 1580, the number of markets operating in the middle Yangzi valley increased fivefold and, depending on one’s choice of modern estimates, between twofold and fivefold in the Yangzi delta. According to Shiba Yoshinobu, the increase in delta towns and markets from c. 1400 to c. 1600 would be 220 percent for Suzhou, 179 percent for Songjiang, and 209 percent for Hangzhou.4 Within just the relatively agricultural Suzhou county of Wujiang, the change was more dramatic. Its three markets and four market towns in c. 1510 had become ten markets and four market towns by 1561 and then ten markets and seven market towns by c. 1600. By that time, four delta market towns had populations over 50,000, another had 35,000, and seven more between 10,000 and 20,000. Comparing the c. 1580 survey with that of 1792 (plus a recent compilation of late Ming and late Qing figures for markets and market towns in the middle and lower stretches of the Yangzi valley), three significant changes are discernible (Table 16.3). First, by the end of the eighteenth century, when the number of Chinese markets had surpassed 22,000, production for the market had spread throughout the empire and was no longer concentrated in the southeast. In fact, certain interior provinces such as Yunnan, Guizhou, and Guangxi in the southwest and Shaanxi and Gansu in the northwest are mentioned in detail in our lists for the first time since the Song. Second, the provinces of Sichuan, Jiangxi, Guangdong, and modern Hebei show rates of increase in numbers of markets far greater than those of the eastern coastal provinces of Jiangsu, Zhejiang, and Fujian (the noticeably slower market expansion in deeply rural Hunan is an exception). Other provinces record a market expansion of about 50 per cent; Anhui alone records a decline. The figures in Table 16.3 indicate the uneven distribution of this market expansion in the interior: generally, the closer

4

Yoshinobu Shiba, “On the Emergence and Intensification of the Pattern of Rural–Urban Continuum in Late Imperial Jiangnan Society,” in Billy K.L. So (ed.), The Economy of Lower Yangzi Delta in Late Imperial China: Connecting Money, Markets, and Institutions (London, Routledge, 2013), p. 176. See Table 16.2 for their growth over an even longer time period.

603

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

1918

Changsha Hengzhou Yongzhou Baoqing Yuezhou Changde Chenzhou Yuanzhou Yongshun Lizhou Chenzhou Jingzhou Guiyang Qianzhou Yongsui Fenghuang Huangzhou

Prefecture

Overall totals: Ming/early Qing: 1,756; Late Qing: 3,656 Source: Ren, “Ming Qing shizhen,” p. 199

700

Total

Late Qing 128 137 106 222 260 252 165 68 171 179 221 9

Ming/early Qing

Wuchang 74 Hanyang 71 Huangzhou 75 De’an 143 Anlu 66 Xiangyang 38 Yunyang 21 Yichang 19 Jingzhou 152 Shi’nan 7 Jingmen 30 Zuifeng 4

Prefecture

Hubei

118 49 56 21 48 18 14 2 1 23 16 1 14 – – 3 – 384

140 54 57 22 55 19 12 8 3 23 16 1 14 – – – 9 433

Ming/early Qing Late Qing

Hunan

672

Nanchang 74 Raozhou 40 Guangxin 10 Nankang 6 Jiujiang 14 Jianchang 33 Fuzhou 109 Jianjiang 59 Ruizhou 43 Yuanzhou 41 Ji’an 31 Ganzhou 144 Nan’an 27 Yudu 41

1305

207 69 75 82 24 14 147 124 29 77 138 285 34 –

Prefecture Ming/early Qing Late Qing

Jiangxi

Table 16.3 Markets and market towns in Hubei, Hunan, and Jiangxi provinces in the Ming and Qing

Merchants and Commercial Networks

a prefecture was to the Yangzi river or a major tributary, the more likely an increase in its market numbers. Third, three sites are highlighted in the 1792 list as particularly important: Suzhou for being “extremely resplendent”; the market town of Hankou for being “the number one, largest wharf for the assemblage and sale of goods in the world”; and the market town of Foshan in Guangdong for having wharves “that assemble every sort of goods from overseas.”5 Indeed, in striking contrast to the 1429 and 1580 surveys, the 1792 list pays attention to places engaged in foreign trade and the import of foreign goods (in addition to Foshan, six other such sites are mentioned, two of them in northern Shanxi near the border). The economic geographer G. William Skinner focused his research into China’s market systems on one overriding question: how, if at all, were these 20,000 or so markets integrated, when they were so widely dispersed over the Qing imperium? His answer was complex: late Qing China had an eight-tier marketing hierarchy of economic central places (plus periodic markets), which provided an institutional infrastructure for the circulation of goods within no fewer than eight articulated “macro-regions” (see Map 13.1 in Pomeranz’s chapter in this volume).6 Locating these macro-regions within key physiographic areas, Skinner then described how each macro-region’s products circulated largely within its own confines and not across its borders. He downplayed the integrative role of interregional trade and, by implication, long-distance “sojourning merchants,” although he recognized that by the Song dynasty interregional trade was overwhelmingly private rather than public and goods were transported, whenever possible, on water rather than on land for reasons of cost, speed, and comfort. Skinner also acknowledged that his principles of market distribution did not fit the passage of goods along the greatest trade route in late imperial times; that is, between the middle and the lower reaches of the Yangzi river. Moreover, he failed to compare his schema with historical evidence on Grand Canal traffic and many other merchant trade routes that are detailed in merchant route manuals published in the late Ming and the Qing. The trade routes laid out in these books indicate the crucial role of “nodes”; that is, major market sites where trading routes and their traders converge, for long-distance “sojourning merchants” as well as these same merchants’ 5 6

SGBL, 3.32a–33b. G. William Skinner (ed.), The City in Late Imperial China (Stanford, Stanford University Press, 1977).

605

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

less significant role in the activities of smaller regional markets and market towns.7 These market towns nonetheless provided a strong foundation for the late imperial commercial economy. In more than half of the Northern Song’s twenty-four circuits (or provinces), the total commercial tax revenues and presumably the level of commercial activity for a county’s market towns were higher than for the county seat. In some counties, the county seat paid no commercial tax, while its market towns or even periodic markets shouldered its entire commercial tax quota. While no empire-wide lists of commercial tax figures survive for markets and market towns in later centuries, the surge in market town development in the Yangzi valley and especially the delta during the Ming and Qing dynasties (Table 16.2) would seem to confirm that during the late imperial period market towns in the delta collectively replaced their county seat as the focal point of commercial activity closest to villages. Their continued importance in the Qing is underlined by their accounting for one-fifth of all marketing and production sites mentioned in the 1792 survey. Few sites provide clearer evidence for understanding how these market towns and large trading nodes actually worked in the Ming and Qing marketing network than the Yangzi delta market town of Luodian and the town of Linqing in Shandong province, the first exemplifying the transformation of market towns and their surrounding area into cotton trade centers and the second the growth of a north China market town into a key trading node for the collection, storage, and distribution of commodities. Whereas the privately developed market towns of the delta over time formed the basic units of the trading and manufacturing clusters that would prove essential to China’s industrialization in the twentieth century, Linqing during the Ming and Qing dynasties served as a state-supported linchpin in the integration, however incomplete, of the markets of north China with those of the lower and middle stretches of the Yangzi valley. Situated near the south bank of the Yangzi in the Suzhou county of Jiading, Luodian was privately established as a rural market by a shop owner surnamed Luo toward the end of the Yuan dynasty on the hunch that his rural site could support a regular market for fish and rice.8 Little did he foresee that within 150 years his anticipated village shop would be 7

8

Skinner, The City in Late Imperial China, pp. 217–18, 235, Table 5, 242, 248; Antonia Finnane, Speaking of Yangzhou: A Chinese City, 1550–1850 (Cambridge, MA, Harvard University Asia Center, 2004), pp. 36–8. Fan, Jiangnan shizhen, pp. 573–92.

606

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

transformed into a bustling market town dealing overwhelmingly in cotton. By the early sixteenth century, the surrounding countryside of Jiading had switched from the monocrop rice culture that had prevailed in earlier times to mainly cotton cultivation, with three-quarters of its fields devoted to cotton and just one-quarter to rice. The town’s residents and nearby villagers were thereby drawn into highly commercialized networks of cotton production and distribution. Luodian’s populace planted cotton, traded raw cotton or spun it into yarn, wove cotton cloth (including at least four special kinds of cotton textiles), and dyed and sold the cloth. By the early seventeenth century, Luodian ranked as Jiading’s largest market town, stretching one mile east to west and two-thirds of a mile north to south along its main waterways. Its cotton and fish markets regularly attracted Huizhou merchants, and despite repeated reports of currency and price manipulation by local brokers it retained a dominant position in Jiading’s cotton trade up to the late nineteenth century. By then, inhabitants of the surrounding villages concentrated on spinning yarn, while town residents worked mainly at weaving such yarn into cotton cloth.The social composition of towns in this cotton production region can be inferred from a rare set of occupation statistics for the nearby market town of Zhujiajiao. Deeply engaged in cotton textile production from the sixteenth century, by 1814 Zhujiajiao had a registered population of nearly 6,000, two-thirds of them males (Table 16.4). By the end of the eighteenth century its economic base had broadened, but raw cotton, cotton yarn, and cotton cloth, along with fish and rice, remained its principal lines of business. Its streets boasted butcher shops, teahouses, taverns, drug shops, cloth peddlers and wholesale cotton goods shops, and that universal sign of debt and inequality, pawnbrokerages. Registered shop staff as well as hired laborers and male servants accounted for less than 5 percent of the total population, suggesting that most of its businesses, including cotton firms, were small-scale; they largely drew on family members (with an average of fewer than three male members per household, much of the work in these family businesses was presumably performed by the sons) for their management and workforce; and these businesses also drew on surrounding villages (not included in the 1814 census) for part of their labor force. The north China port of Linqing reveals a very different type of commercial development, one that drew strength initially from government decisions (for example, the shift of the Ming’s principal capital to Beijing in 1419, and subsequent government investments in its infrastructure) and consistently 607

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

Table 16.4 Social composition of Zhujiajiao market town, Songjiang prefecture, 1814 Households

1,502

Total population Male adults Female adults Young boys Young girlsa Shop staff Hired laborers Male bondservants Female bondservants Buddhist monks Daoist priests Nuns

5,937 2,279 1,950 1,025 187 267 99 21 48 23 20 9

a

Clearly the number of young girls was vastly underreported, perhaps reflecting their marginal significance to administrators and tax collectors Source: Mori Masao 森正夫, “Shukakakuchin ryakushi” 朱家角鎮略史, in Mori Masao (ed.), Ko¯nan deruta shichin kenkyu¯: Rekishigaku to chirigaku kara no sekkin 江南デルタ市鎮研 究: 歴史学と地理学からの接近 (Nagoya, Nagoya daigaku shuppankai, 1992), p. 80

from its location along the state-maintained Grand Canal.9 In the early Ming, Linqing had a population of 5,000 registered households (perhaps 25,000 persons). By 1580 its population had reached 150,000 persons, and by the end of the Ming totaled 200,000, an eightfold increase that saw the city’s area concurrently expand several miles in all directions, and the wharves and shops along the Grand Canal stretched for more than ten miles on each bank. Linqing’s rapid economic growth in the Ming as the major transshipment node between the northern and southern economies won it close attention from the state, which laid claims to its success by promoting what in the early 9

Xu Tan 许檀, “Ming Qing shiqi de Linqing shangye” 明请时期的临清商业, Zhongguo jingji shi yanjiu 中国经济史研究 1986.2, 135–57; Xin Shufang 邢淑芳, “Gudai yunhe yu Linqing jingji” 古代运河与临清经济, Liaocheng shifan xueyuan xuebao (zhexue shehui ke xueban) 聊城师范学院学报(哲学社会科学版) 1994.2, 68–75; Xu Tan 许檀, Ming Qing shiqi Shandong shangpin jingji de fazhan 明清时期山东商品经的发展 (Beijing, Zhongguo shehui kexue chubanshe, 1998).

608

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

Ming had been a mere tax station to a county seat in 1449 and to a prefectural seat in 1484. Located at the intersection of the Grand Canal and the Wen and Wei rivers in the northwest corner of Shandong province, Linqing attracted merchants and goods from Beijing, Shanxi, Shaanxi, Shandong, and Liaodong in the north and from Jiangsu, Zhejiang, Anhui, and Jiangxi provinces in the south. Most of its resident merchants were migrants from Huizhou. Its merchandise for sale consequently ranged from iron pots made in Foshan in Guangdong and iron farm tools from Lüzhou in Shanxi to paper from Jiangxi and Fujian and tea from Anhui and Fujian. The two products most commonly traded in this emporium, both essential to the lives of millions of northern Chinese, were grain and cotton. In addition to its government granaries, used to store the annual grain tax shipments of over 2,220,000 bushels of rice from ten southern prefectures, Linqing handled private trade in wheat and barley carted and shipped in from Henan, Liaodong, and elsewhere in Shandong, as well as rice from the Huai and Yangzi river valleys. By the late sixteenth century, this private grain trade fluctuated annually between 5 million and 10 million bushels. Linqing’s dealings in cotton cloth are estimated to have consisted of 1 million to 2 million bolts a year in the Ming; its transactions in silk yarn were far more modest. As a result of all this trade plus the constant passage of boats through its port during the ice-free months, the Linqing customs station had the empire’s top commercial tax quota in the Wanli era (1573–1620), twice that of Hushu station in Suzhou’s suburbs and even 20 percent more than that of Chongwen Gate station, the Grand Canal’s terminus in Beijing. A large increase in its commercial tax quota later in the Wanli era reportedly forced the closure of many Linqing stores, reducing by 1625 its actual commercial tax payments to just three-quarters of the new (and higher) quota. Yet this 1625 tax payment still ranked as the second-highest among all tax stations in the empire (the top position that year fell to Hushu station).10 In other words, by the end of the Ming this former market town along the Grand Canal had become, apart from the capital, Beijing, the most thriving market center in north China, thanks to its place as the crucial node in the integration of long-distance trade between north and south China. It was to north China in the Ming what Hankou would become to central and west China in the Qing.

10

WXTK (Wang), 2, p. 937.

609

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

Table 16.5 Stores (by type) in Linqing market town in the late Ming and mid-Qing Store type

Late sixteenth century

Cloth stores

73

Silk cloth stores Miscellaneous goods Grain shops Ceramics shops Paper shops Leather shops Liaodong wares Tea leaf shops

32 65 Not specified 20+ 24 Not specified 13 large shops Not specified

Salt traders (hang)

In addition to government shops (gongdian), every street has a shop owned by one of 13 families Pawnbrokers 100+ Traveling mer- Several hundred, big and small chant inns (kedian)

1749

Change

“on every street” 7 or 8 Not specified

?

100+ “cut to half” 5 or 6 7 or 8 Now none 28 big shops, uncounted small shops “recently not even half”

? 50% ~75% ? –100% ?

16 or 17 “cut by half”

~75 % ?

~50% ~85% ~50%

Source: Tang Wenji 唐文基 (ed.), 16–18 shiji Zhongguo shangye geming 16–18 世纪中国 商业革命 (Beijing, Shehui kexue wenxian, 2008), pp. 160–6

During the early and middle Qing, however, Linqing’s trade declined, as indicated by the reduction in the types and numbers of its shops from the late Ming to the mid-eighteenth century (Table 16.5). Although figures are unreported for five of the twelve categories for 1749, the available data show a clear decline in store numbers compared to the late sixteenth century. Two of the sharpest drops – five-sixths of the town’s pawnshops and half of its inns – suggest a serious weakening of market activity and facilities for long-distance trade. In 1776 the Qianlong Emperor passed through Linqing on one of his southern tours and observed that it was reviving from damage inflicted during the Wang Lun uprising in the area just two years earlier. Yet a more lasting affliction of the city and a sign of its dim future as an emporium of north China was the rising siltage level in the canal, a problem that, along with persistent social and military unrest in the surrounding countryside, would lead to the canal’s closure and Linqing’s permanent decline in the nineteenth century.

610

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

Generally speaking, the commercial network of the late Ming and early Qing consisted of four interlinked regional market groups: the northern Grand Canal group, the Yangzi river group, the southern Grand Canal (or Yangzi delta) group, and the southeast coastal group. Collectively these four groups indicate the extent of China’s commercial expansion and urban transformation by the early seventeenth century. By 1800 further growth of markets in the west and south, encompassing new centers such as Foshan in Guangdong and Kunming in Yunnan, thus completed the framework of the emergent national market. To these sites were drawn large numbers of migrants who invigorated these nodes’ transport and commercial life (the great majority of Ming dynasty Linqing residents, like those registered in Hankou during the Qing, were merchants from elsewhere, especially Huizhou). The convergence of great numbers of “sojourning merchants” at these places made them significant centers of interregional trade. Purchasing the Yangzi valley’s rice, timber, ceramics, silk goods, cotton textiles, and books for shipment to the north, they brought south such northern goods as furs, metals, raw cotton, and dried fish; or they brought salt, silk thread and textiles, cotton textiles, and other manufactured goods up the Yangzi, while down the Yangzi they carried raw cotton, timber, tea, and, of course, rice. Such was the interregional vibrancy of this trade that by the late Ming the once minor market of Linqing had been transformed into a linchpin in north–south trade, and an upland Jiangxi county seat like Yanshan attracted an astonishing variety of textile types and other goods from throughout the Yangzi valley and the southeast coast.

Long-Distance Transport of Three Commodities: Grain, Cotton Goods, and Books The scale of long-distance trade, the dominance of private over public transactions, and the spread of the market throughout the empire are all evident in changes to the production and distribution of three important Chinese commodities – grain (overwhelmingly rice), cotton textiles, and books – from the eleventh century to the late eighteenth. A review of their production and distribution will underline how market institutions, stimulated by Huizhou merchants and other long-distance traders and reliant on China’s extensive river system, became crucial to commoners and scholarofficials, villagers and urban residents, as well as merchants and consumers. Whereas, in the first millennium of imperial rule, grain distribution in China had been dominated by tax grain transport in north China, usually 611

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

heading to the capitals from the Central Plain and points east, the second millennium saw a shift to increasing government and private dependence in north China, especially its capitals, on shipments of central and southern rice surpluses. From the start of the eleventh century, the southern circuits were expected to ship over 6.2 million bushels annually as their basic annual grain tax to the northern capital, 2.6 million more bushels of “harmonious purchases” (hedi 和糴 – grain purchased by the government on the private market, nominally at market rates) to the capital and other parts of the empire, and yet millions more as the staple foodstuff for people in both south and north China. The Yangzi delta prefectures reportedly contributed nearly a third of annual grain tax shipments, although disappointing harvest results sometimes prevented their full payment. In addition, the total volume of private shipments of rice along the Yangzi river and the southeast coast during the Song came to surpass these figures for government grain transport on the Grand Canal. In the twelfth and early thirteenth centuries, the Yangzi delta would reign as the most productive rice-producing area, its bumper harvests supposedly feeding the entire empire. But the delta’s population growth far outstripped available arable land, and Hangzhou’s elevation as the capital in the 1130s required more and more rice shipments from throughout south China to feed its population of over a million. By the mid-thirteenth century, most of the prefectures in this erstwhile grain basket were in some years obliged to import rice from the middle Yangzi region. The Yuan conquest greatly upset these arrangements. Many areas of the upper and middle Yangzi valley produced far less grain for export thanks to severe damage and depopulation inflicted by the advancing Mongol troops, while the delta’s prefectures (which suffered much less from the Mongol conquest) had less need for grain imports. Hangzhou ceded its position as capital to Beijing, and the delta’s rice surplus areas once again exported large amounts of rice, estimated at 3.5 million bushels, to the north by sea. The founding of the Ming dynasty in 1368 led to the capital’s relocation first to Nanjing and then from 1419 back to Beijing, where 4 million bushels of tax grain were to be sent annually from the 1470s. After 1500 the private grain market in the Yangzi valley expanded, and the upstream provinces of Hunan, Hubei, Jiangxi, and southern Anhui regularly shipped rice to the now rice-deficient prefectures of southern Jiangsu and Zhejiang. Whereas, according to a common saying of the twelfth and thirteenth centuries, a bumper crop in just two delta prefectures – Suzhou and Huzhou (or Changzhou) – could feed the entire empire, the mid-Ming 612

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

saw a long-term shift of the empire’s grain basket some 700 miles upstream. Increasingly, middle Yangzi rice production fed the delta, where much land was converted to cotton and mulberry tree cultivation to provide local farmers with raw materials for the more lucrative, market-oriented cotton textile and sericulture industries. By the late sixteenth century, the highquality rice produced by the delta’s rural inhabitants was sold in urban markets, while local farmers bought and consumed cheaper, less tasty varieties harvested and shipped in from the middle Yangzi region. According to the estimates of Xu Dixin and Wu Chengming, annual longdistance private rice trade in the late Ming totaled about 10 million bushels, a big increase over the Song but less than a third of what these same scholars have found for the peak volume of the grain trade three centuries later in the mid-eighteenth century (Table 16.6). Not only had there been a three-and -a-half-fold increase in the private grain trade, but also the great westward expansion of the grain trade up the Yangzi river valley encompassed many areas of the empire previously no more than tenuously engaged in long-

Table 16.6 Routes and volume of grain and soybean trade in the mid-Qing

Exporting areas

Importing areas

Foodstuff

Amount (millions of bushels)

Jiangsu, Zhejiang, Jiangxi, Anhui, Hubei, Hunan Fengtian and the northeast Fengtian Henan, Tianjin De’an, Xianyang, Anlu (via Hankou) Anhui, Jiangxi Hunan, Sichuan Jiangsu, Zhejiang Taiwan Jiangxi, Hunan Total

Beijing, Shanxi, Shaanxi

Rice

6.0

Tianjin, Shandong

Wheat, soybeans 1.0

Shanghai Linqing Shaanxi

Soybeans, wheat 10.0+ Wheat, sorghum 0.2+ Wheat 0.06

Jiangsu, Zhejiang Jiangsu, Zhejiang, Fujian Fujian Fujian Guangdong

Rice Rice Rice Rice Rice

5.0 10.0 0.1+ 1.0+ 2.0 35.4+

Source: Dixin Xu and Chengming Wu (eds.), Chinese Capitalism, 1522–1840 (London, St. Martin’s Press, 2000), pp. 167–70

613

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

distance transactions. In 1753 the Yangzi valley route would have accounted for as much as 40 to 60 percent of the private grain trade. Hunan alone exported 6 to 10 million bushels in normal harvest years. In the mid-eighteenth century (c. 1741–1755), an estimated total of at least 30 million bushels of grain was regularly held in state granaries, principally for famine relief and price stabilization. Although we still understand little about the operation of this vast marketing hierarchy, by the 1690s some commentators considered Hankou the main market of central China. A century later a merchant manual even ranked it the premier market for all of China. South China’s other major nodal grain markets included Jiujiang, Suzhou, Yangzhou, Chang’an market town (in Haining, Zhejiang), and various prefectures in Sichuan and Hunan. Throughout the Ming and Qing eras, the surge in the trade of cotton goods was even steeper than that of grain. Introduced to the Yangzi delta in the late thirteenth century, cotton farming rapidly gained favor among delta cultivators for its reliable harvests and high returns, especially on dry, hilly soil unsuited to rice farming, and among consumers for making a cloth lighter and warmer than linen and far more affordable than silk. In fact, so popular did the delta’s cotton cloth become that, by the end of the fifteenth century, Songjiang was already said to “clothe the entire empire,” an exaggeration perhaps, but in terms of its share of the commercial textile market not far from the truth. During the fifteenth and early sixteenth centuries, cotton planting spread north of the Yangzi into the Huai river valley, Henan province, and even parts of Shandong and Shanxi provinces. Less skilled at spinning cotton yarn and weaving cotton cloth, northern farmers became distant suppliers of raw cotton via the Grand Canal to the textile industry of the Yangzi delta, where a better-skilled workforce spun the cotton into yarn, which then was woven into cloth. (The estimate by the Portuguese Jesuit Alvaro Semedo of over 200,000 looms operating in just the Shanghai area in 1621 strikes me as inflated, however.)11 By 1600 cotton cloth was being shipped from the delta in all directions: north to the capital, Beijing, westward up the Yangzi as far as Sichuan, and southward along the coast to Guangzhou and Southeast Asia. By then, northerners had started spinning yarn and weaving cloth for markets elsewhere in the north. And so in the late Ming the delta’s cotton cloth exporters opened up markets in the upper reaches of the Yangzi valley, 11

Alvaro Semedo (trans.), The History of the Great and Renowned Monarchy of China (London, E. Tyler, 1655), p. 14.

614

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

connecting their distribution network with the trading distribution pattern for other delta products shipped westward up the Yangzi for the increasingly larger supplies of rice needed from the interior. During the Qing the delta remained the center of cotton textile production, but its share of the national market for cotton cloth suffered a decline as more and more areas in the western and northern interior planted cotton, spun yarn, and wove cloth. Songjiang’s strong advantages in this trade were its dyeing practices, specialty fabrics, and higher skills, but its high prices cost it budget-minded customers. Moreover, at the advent of the nineteenth century, Songjiang’s weaving practices and technology remained largely those used since the seventeenth century. With hindsight, the industry was facing a structural crisis that would be resolved only by the introduction of radically new forms of production in the late nineteenth century. Our third commodity, woodblock-printed books, has been chosen to underline the broad significance of changes already seen in the Ming and Qing grain and textile sectors, such as the spread of production centers to the west, south, and even north of the Yangzi delta, and their structural significance for wider distribution patterns. Whereas in the Song empire of over 100 million people just four sites – Kaifeng, Hangzhou, Chengdu, and Jianyang – served as major book production centers, in the Ming their number grew to six, with two of the major Song publishing centers, Jianyang and Hangzhou, supplemented by Nanjing, Suzhou, Changshu, and Huizhou (Chengdu’s publishing industry disappeared after the Mongol conquest, to recover only in the eighteenth century). All these major Ming publishing sites except Jianyang and Huizhou were located in the Yangzi delta, resulting in a highly unbalanced concentration of commercial publishing activity that was only partially alleviated by production in eleven or so relatively minor locales scattered about the empire, usually in provincial seats. Over the course of the eighteenth century, the Yangzi delta’s supremacy in the volume of book production and sales was challenged by the success of no fewer than twelve major and thirty-three minor production sites elsewhere. Large cities, especially Suzhou, continued to dominate quality publishing and the rare-book market in the south. Suzhou bibliophiles’ family collections made even the Qianlong Emperor (r. 1736–1796) green with envy. But Suzhou relinquished its primacy in north China’s rare-book market to Beijing’s Liulichang quarter, and arguably in the new-imprint market as well by the early nineteenth century. The delta’s overall share in south China’s book production and trade also faced serious challenges from the many flourishing 615

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

provincial sites producing standard titles with low-quality carving and paper, under no legal copyright restraints, and at lower prices. Given the simple technology and low setup costs for woodblock publishing, it may seem odd that this proliferation of book production and distribution centers – occurring in most of the empire only during the Qing – took so long to develop. Was the demand – that is, the classical or general literacy rate – so low? Nonetheless, publishing eventually became one of late imperial China’s major empire-wide industries, not simply because its production sites were by the nineteenth century scattered extensively throughout the country, but also because, according to research by Cynthia Brokaw, they were then part of a complex set of production and distribution networks.12 Overall, book production in the north lagged far behind that of the south, so that some books printed in new southern and even southwestern centers were shipped thousands of miles via the Yangzi and Grand Canal to peddlers, retailers, and readers in northern provinces. Upper-level markets in metropolitan centers such as Beijing were linked not just to the traditional major publishing centers of Suzhou, Hangzhou, and Nanjing, but also to intermediate-level market towns such as Xuwan in Jiangxi (from the mid-eighteenth century to the early twentieth many Xuwan book merchants headed bookstores and publishing houses in Beijing’s Liulichang quarter), and even to lower-level hinterland communities like Sibao in Fujian. So successful was this expansion in the interior that some new publishing sites assumed distinct niches in an older publishing site’s book production. For example, women carvers in Magang in Guangdong cut woodblocks for publishing houses in Suzhou, as did male carvers in Yuechi in Sichuan for publishers in Chengdu and Chongqing. Carvers and other artisans also routinely migrated to meet the expanded demand for their skills in both new and old publishing sites. The expanded role in book publishing and distribution of the three market towns of Sibao, Yuechi, and Xuwan during the middle and late Qing merits particular notice, since these market towns were located both far from the delta (where no market town ever became a major publishing site) and well down the marketing hierarchy. Their emergence suggests the spread of the practice of reading throughout the country and among various social classes (an impression strongly confirmed by their books’ often low printing and paper quality, crude and simple titles, crowded page layout, and relatively 12

Cynthia J. Brokaw, “Empire of Texts: Book Production, Book Distribution, and Book Culture in Late Imperial China,” in Joseph P. McDermott and Peter Burke (eds.), The Book Worlds of East Asia and Europe, 1450–1850 (Hong Kong, University of Hong Kong Press, 2015), pp. 187–235.

616

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

cheap prices), as many literate commoners could now afford a few simple books. Although reading, for so long in Chinese history the supposed passport to higher status, wealth, and power, would fail to confer its vaunted security and status on its new practitioners, the late Qing emergence of a truly national book market stretching from woodblock carving shops in the Sichuan countryside to booksellers in Beijing would facilitate the wider and deeper dissemination of ideas and information. Up to the Taiping Rebellion (1850–1864), the Qing government operated a censorship regime far more rigorous than those of earlier dynasties; it even enlisted the services of a supposedly private teaching academy in Suzhou to check and approve all woodblocks headed for local print shops.13 Nonetheless, as this enlarged book world was no longer confined mainly to the Yangzi delta and northern Fujian, by the dynasty’s close its distribution network helped to disseminate heterodox political and religious views to literati and other readers in interior cities and towns previously quiescent at times of national crisis.

Merchant Problems, Merchant Solutions The next two sections will focus on how late imperial merchants practiced trade. Here we suffer from a severe shortage of primary source records, as these merchants were highly adept at concealing their activities. Nonetheless, enough scraps of evidence survive, particularly on the Huizhou merchants, to reveal how these traders prospered through resort to practices that rarely captured the attention of historians, Chinese as well as Western. To be specific, I would like to show how Ming and Qing merchants often (but not always) found ways to sidestep or overcome commonly noted obstacles to their work, such as capital shortages, poor security for themselves and their goods, access to markets, and state attacks on merchants through taxes, expropriation of their wealth, and unpredictable interference in the market. More often than not, merchants proved adept at turning weaknesses into strengths, as they used the two most legitimate institutions of their society, the family and the state, to pursue their own commercial interests. This path to success would be most evident in the advance of long-distance “sojourning merchants” in Chinese economic life. Having first emerged in some numbers in the late Tang dynasty (618–907), they attained unrivaled prominence in commodity markets and financial services from the sixteenth 13

Joseph P. McDermott, “Rare Book Collections in Qing Dynasty Suzhou: Owners, Dealers, and Uses,” in Lü Miaofeng 呂妙芬 (ed.), Jinshi Zhongguode ruxue yu shuji 近 世中國的儒學與書籍 (Taipei, Zhongyang yanjiuyuan, 2013), pp. 199–249.

617

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

century. Although these men were commonly regarded as outsiders, they were insiders enough to become the most important component in each of the ten regional merchant groups of the late imperial period. How, then, did these merchants run their businesses in this market economy? In the past, historians have tended to stress the obstacles facing Chinese businesses, beginning with China’s lack of a commercial code and thus its limited legal guarantee for the surety of businesses, their assets, and their transactions. The government provided minimal physical protection to private transport not just on its rivers and the ocean but even on its patchily maintained roads. It arbitrarily levied a host of taxes on merchants and their goods for both court and local uses. It lost control of its currency from the fifteenth century, initially through the excessive issue of paper money and then through a heavy reliance on silver imports from Japan and the New World to fuel the country’s economic expansion. Moreover, historians have emphasized that the imperial state failed to establish an effective regime for its private financial sector. Banking facilities remained poorly co-ordinated and grossly inadequate for the amount of capital circulating and needed by merchants. Up to the mid-eighteenth century its money market was barely nascent, and, to make matters worse, loans made by money shops, pawnshops, and private parties commonly bore very high interest rates. Moreover, a government run by Confucian-educated officials frequently treated its merchants with distrust and sometimes with disregard, so that even in strictly economic transactions the merchants’ institutional autonomy was circumscribed. No wonder that some modern scholars concluded that the government was these merchants’ most resolute foe. Over the past three decades much of this conventional wisdom has rightly fallen under question. In continuing this reassessment I want to show not so much that the original criticisms were wrong (or the countercritique invalid), but simply that they greatly exaggerated the damage from these impediments. Despite these real obstacles, many late imperial merchants attained and retained considerable wealth, as they found institutional and noninstitutional means to thwart or dull the potential damage. Chinese merchants naturally preferred to keep these solutions to themselves, and as they were able to organize publicly into political groups only from the late nineteenth century, much of their earlier, unpublicized success escaped scholarly attention and analysis until recently. Here I will present some of that story by focusing on institutions, such as ancestral halls, pawnshops, and commercial partnerships, that late imperial 618

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

merchants found useful in solving problems that they (like merchants elsewhere) commonly faced in plying their trade: security for themselves and their goods, access to markets and to capital, and business organization. The fourth and final section separately considers the very complicated and often contradictory history of the merchants’ relationship to the state and its agents. Chinese merchants usually refrained from pursuing government reform or installing new administrative institutions. They far preferred low-key remedies to handle these risks, small practical steps that called for adjustments in human arrangements rather than government institutional involvement. They seem to have instinctively recognized that their wealth would usually escape official notice so long as they did not flaunt it, paid most of their taxes, concealed their bribes, and – despite the increasing overlap of their background and activities with those of officialdom – kept away from official politics, especially court factions. On a key problem, access to markets, the legal codes of all late imperial dynasties made explicit promises to assure proper orderly trade. A considerable number of imperial pronouncements and official decrees strengthened this stance with bans on coercive takeovers of a market, price manipulation, unauthorized work by private brokers, sale of substandard goods, the making and use of private measures for market transactions, and merchant collusion (see the So and So chapter in this volume). Protectionist practices at markets within the empire were not to be condoned. Yet the Ming state, particularly at the county and prefectural levels, often half-heartedly implemented decrees intended to assure unimpeded trade. It sometimes stationed a government (usually a military) official in markets or market towns to assure orderly transactions, but in the Yangzi delta such officials seldom enforced bans against local interests. Some of the most flagrant offenders were local laborers’ groups, which sought to restrict competition for unskilled jobs in transport, port administration, and other urban transactions. In delta cities and towns stevedores formed gangs that claimed monopoly rights to the handling of cargo at specific wharves; transport within these urban sites was regulated by unwritten “turf” rules that obliged shops to hire specific porters or boats for passage into specific neighborhoods and over or under certain bridges. Qing government bans on such abuses were routinely ignored. Similar territorial restrictions held true for the arrangement of funeral and marriage processions and the hiring of sedan chair bearers, ritual practitioners, and even Buddhist monks for their services. 619

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

However inconvenient these protectionist charges may have been (their overall effect on transaction costs would have been small), they surely drew from late imperial commentators and officials far less bile than did market brokers (yaren 牙人). Distrusted by commentators during the Song and Yuan, these local merchants were positively reviled during the Ming and Qing by outsiders, including officials and sojourning merchants. In fact, in 1369 the first Ming emperor arbitrarily banned them from all markets of his empire, and instead ordered local yamen employees to handle all private trade transactions. Only in 1397 did officials see reason to reverse this ruling and accept the need for officially registered brokers. In late imperial legal codes these brokers were expected to assure the smooth operation of markets. Registered as a local man of property and purchaser of an official broker’s licence, a broker was called upon to match local and outsider parties interested in trading, and then report monthly to the local government on each transaction he had handled (plus hand over the commercial transaction charges he collected for it). The range of his services to merchants, especially sojourning merchants from other parts of the country, was extensive, as Susan Mann reminds us for large cities in the Qing: Brokers proliferated rapidly because they could operate at many different levels of the marketing system. They could negotiate lateral agreements between two peasants for a few coppers; they could hire out as agents to major wholesaling firms and solicit business from primary producers and retailers; they could invest in hostels, stables, and warehouses and operate wholesale businesses of their own; they could invest in boat and ferry rentals and contract to carry waterborne or overland trade from place to place. There was, in short, a brokerage for every pocketbook.14

Brokers might be called upon to find suitable accommodation for itinerant merchants; collect and dispense trade information; match potential buyers and sellers; extract market-entry fees from outsiders, including peasants peddling vegetables; arrange shipping, docking, and warehousing services; manage warehouses, hostels, and even shops; negotiate prices and convey payments; and pressure parties to a transaction to fulfill their promises and complete contracts (undoubtedly, many agreements were oral, and so a broker’s “word” was crucial for their implementation). In often opaque markets rife with misinformation and merchant distrust, the broker had the 14

Susan Mann, Local Merchants and the Chinese Bureaucracy, 1750–1950 (Stanford, Stanford University Press, 1987), pp. 64–5.

620

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

opportunity – which he appears to have often seized – to represent naturally opposing commercial interests, at a cost to others. Unsurprisingly, he ended up as the favorite object of contempt of all the other parties. These criticisms, however justified, missed the point that the government and its officials did not want to become directly involved in commerce and so sought to co-opt powerful local figures, who inevitably became the target of outsiders’ anger. The broker was expected to handle naturally conflicting and contentious parties in a rapidly commercializing economy, in which merchants were regularly testing the outer range of acceptable practice. Since Ming officials appear to have avoided upsetting local commercial arrangements lest they impinge on their own extraction practices, they showed minimal interest in establishing fair commercial practices and institutions. And so the broker was called upon to handle numerous disputes as negotiator or even as an unofficial judge. This mode of operating clearly opened the door for retired officials and local gentry to be called in to negotiate controversial decisions and handle complicated transactions. As a result of the increased complexity and weight of these tasks, brokers gradually evolved into brokerage houses. The single figures aided by their sons in the delta’s larger cities during the pre-1600 era were thereafter increasingly replaced by brokerages run as multiparty firms or partnerships. The officially registered individual broker, in effect, became a business that he ran either with partners or with others acting as his extended staff in order to handle the greatly increased volume of work. Some brokerages continued to constitute a family firm operated by the family head; upon his death the sons might continue the business along strict partnership terms, with some earning more due to their greater intake of business. Or a brokerage house might receive others’ investments, take on paid staff, and thus be expected to pay out annual dividends to private investors in its operations. Predictably, the investors in such brokerages were often local merchants concerned more with using these brokerages to preserve their share of the local market than to increase the dividends from their brokerage shareholdings. This local market was further skewed when these brokers not only invested in local businesses anxious to trade with outsiders but also ran their own businesses in local products that these outsiders wanted to acquire. The first Ming emperor’s ideal of the broker as a disinterested party bridging locals and outsiders, buyers and sellers, became over time a thoroughly disabused fiction. To overcome or rather circumvent such structural obstacles, sojourning merchants sought simple, personal solutions. If at all possible, they or their sons signed up as brokers, or they hired others as their private brokers. They 621

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

purchased their own wharves, rented shops and warehouses, and found accommodation in inns, temples, and even government facilities. Over time, they might set up their own homes in these large cities and, as we shall see in detail below, find accommodation, security, and commercial advice in collective institutions, known from the Ming as native-place associations (huiguan 會 館), that helped them resist exploitation by local brokers and other local groups. Access to capital was the third and probably most persistent economic problem for late imperial merchants, as many necessarily lived off credit through advance purchases, delayed payments, and various forms of borrowing and collateral. In the late Ming a merchant’s venture might rely on others’ investments or loans for reportedly as much as 60 percent of its operating capital. A common shortage of funds was sometimes worsened by an unreliable supply of currency (especially before the inrush of foreign silver from the mid-sixteenth century to the mid-seventeenth and then the mining of Yunnan’s copper deposits in the eighteenth century), by an inefficient and limited banking system, and by high interest rates for loans. In place of merchant-run banking institutions, the Chinese merchant had recourse to moneylenders, pawnbrokers, and kinsmen or close friends, sometimes organized into credit associations. The earliest examples of native banks (qianzhuang 錢莊) appeared as late as the mid-eighteenth century. From no later than the Song, however, one common traditional solution to this capital shortage was to form small-scale revolving credit associations (hui 會) that allowed each member in turn to gain access, at an affordable rate of interest, to a relatively large amount of capital that members were regularly obliged to donate. Though such credit practices have been found throughout the world, the Chinese schemes from the delta during the Qing and Republican periods are distinguished by their financial sophistication. Often set up for a year or less, they worked best for those merchants who knew of easy ways to land quick profits. Also, to reduce risks, they were conducted almost invariably among close kinsmen and family friends. Another common credit institution in late imperial China was the pawnshop. Originating in the fifth century as a service offered by Buddhist temples, this religious credit institution eventually spawned four additional kinds of loan institution: the private secular pawnshop run as a financial business; government loan shops; shrine-based credit associations; and, in lineage-dominated parts of south China, ancestral-hallbased credit associations run for lineage members. The first type of these credit agencies, the secular pawnshop, was from no later than the mid622

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

Ming the main credit institution (excluding the individual moneylender). Operated commonly throughout the eastern third of the country under Huizhou or Shaanxi merchant ownership and management, these shops provided small bits of capital to locals, including villagers desperate for an infusion of cash. Run by outsiders, they aroused fierce criticism for their extortionate interest rates, arbitrary fees, and high-handed demands for payment of first the interest and only then the principal. In relatively prosperous areas of the southeast, pawnshop interest rates regularly surpassed the legal limit of 30 per cent per annum in the early Ming, but by the late Ming, at least in the Yangzi delta, they fell often to 20–30 percent depending upon the debtor’s borrowing record, the valuation of his collateral, the promised date of repayment, the size of the loan, and each party’s bargaining skills. Shifting economic conditions in local markets undoubtedly mattered, yet the annual interest rate in surviving loan contracts from Ming and Qing Huizhou and some other parts of China up to the midnineteenth century falls consistently within a range of 20 to 30 percent. Some scholars have argued that actual lending rates were often lower. They claim that these high interest rates reflect personal transactions and not the separate and presumably cheaper commercial rate available to merchants borrowing from other merchants. They also point to studies of several late Ming pawnshop account books that indicate overall annual net profits of 8 to 12 percent. But this analysis is problematic. Not only do these studies fail to indicate to what extent the favored clients in these account books were merchants, but also the profit rates they cite are not general commercial interest rates; they have been calculated after deducting client defaults, absconding, and bankruptcies, and they overlook the likelihood that such shops restricted loans to the most reliable borrowers. Furthermore, recent research on the wider historical record from the late Ming up to the mid-nineteenth century effectively counters claims of cheaper commercial rates. The only known explicit mention of a commercial rate before the late seventeenth century points, paradoxically, to there being no distinction between commercial and personal rates. In his book Merchants Come to Their Senses at Just One Glance (1635), apparently written for Huizhou merchants, the outspoken Fujian author Li Jinde refers to these two loan types as distinct categories. Whatever his grounds for this distinction, he nonetheless fixes their interest rate at the same or roughly the same level: For business transactions (jingying maoyi 經營貿易) and for making private loans (fang sizhai 放私債), only take interest of 20 or 30 percent. This is the usual take, for which there is no resentment. If you want 70 or 80 percent

623

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

interest, then even though it approaches the [purchase] cost of the piece, it is all right. [Yet,] it is hard to regard it as a permanent [rate].15

A “usual” trade interest rate of 20 or 30 percent, let alone an “acceptable” rate of 70 or 80 percent, suggests that such loans were few and mainly short-term. More significantly, the most comprehensive research ever undertaken on the history of Chinese interest rates has recently demonstrated that between 1660 and 1840 the average empire-wide annual commercial interest rate was 39.2 percent, the median 30 percent, and the standard deviation 27.65 percent.16 Another comprehensive review of all kinds of surviving Huizhou loan contracts from 1645 to 1950 showed some regional and temporal variations, but reached a similar conclusion about annual rates there only rarely falling below 20 percent.17 If high interest rates persisted for so long, where and how did capital-hungry merchants find affordable capital? For small sums and thus for villagers trying to become traders, two other institutional sources of credit come to mind, both arising from village institutional arrangements. Village worship associations (she 社) and their related shrines (miao 廟), mainly in north China, provided members with short-term loans at relatively low interest rates, while ancestral halls, mainly in the south, similarly made loans available to their members through their credit associations at relatively cheap rates. Funds collected for eventually constructing an ancestral hall would be loaned out on an interim basis either to non-lineage members at 30 percent or higher rates of interest or to lineage members at 12 to 15 percent interest. When shrewd management increased these funds enough to cover a hall’s construction cost, lineage heads sometimes balked at seeing all their hard-earned capital “dissolve” into tiles and pillars. In addition to retaining these funds by delaying construction for as long as possible, these men concocted ways to recover and expand them. They levied “candle charges” on worshiping members, and, more importantly, required further donations before male ancestors’ spirit tablets were permanently admitted into the hall’s ancestral worship services. Such donations emerged not just from pressure on lineage members to demonstrate their filial piety but also from the donors’ wish to continue qualifying for the loans at low 15 16

17

KSXM, p. 311. Chen Zhiwu 陈志武, Peng Kaixiang 彭凯翔, and Yuan Weipeng 袁为鹏, “Qingchu zhi ershi shiji qianqi Zhongguo lilüshi chutan – ji yu Zhongguo lilüshi shuzhuku (1660– 2000) de kaocha” 清初至二十世纪前期中国利率史初探 – 基于中国利率史数据库 (1660–2000) 的考察, Qingshi yanjiu 清史研究 2016.4, 36–52. Wu Bingkun 吴秉坤, Qing zhi Minguo Huizhou minjian jiedai lilü ziliao huibian ji yanjiu 清至民国徽州民间借贷利率资料汇编及研究 (Shanghai, Shanghai Jiaotong daxue, 2016).

624

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

interest rates that the hall reserved for its paying members. These low-cost loans, usually small, would have eased their borrowers’ initial forays into peddling local products or into making small investments in a kinsman’s promising business venture. For larger sums, we are forced to consider the help from kinsmen in the establishment of family businesses and ultimately in more long-lasting “house firms.” These family-based commercial organizations assumed numerous forms, differing greatly in the size of their membership and the scale of their capital. The simplest form of family business might consist of a two- or three-generation family, whose members pooled their resources and revenues under the guidance of the family head, in a manner of operation akin to that of a Chinese farming household. In other words, the key male relationships of Chinese peasant families – those of father–son, grandfather– grandson, uncle–nephew, and brother–brother – were also those of these merchant families, as was the farmer’s overriding concern to preserve, if not expand, his family’s wealth or property. This adoption of farm family concerns by small family businesses was also evident in their shared fear of the breakup of the family upon the death of its head. Upon his decease the simple kinship structure of a three-generation business most likely collapsed, and the surviving sons commonly went off to set up independent families and businesses. And so, just as landholding families sought to reduce the impact of such family divisions on family assets by forming larger corporate kinship groups with collectively owned property, such as a large communal family or a landed trust lineage, so did commercial families often seek a flexible organizational structure that allowed them to practice commerce both independently and collaboratively with kinsmen for the long run (essentially, they sought a commercial variant of the normal charitable landed estate arrangements validated for lineages since 1050). These lineages, barring those based solely on trusts of commercial real estate, were themselves seldom businesses. But within their larger, looser, and more inclusive kinship and property-owning framework they tolerated a considerable number of commercial alliances, or businesses, formed by lineage members often along their branch lines of descent. As seen in the expansion of some Huizhou merchant lineages into the Yangzi delta and the middle Yangzi valley during the second half of the Ming dynasty, different branches and their segments explored distinct regional markets for themselves. By establishing long-distance trading networks, they linked their investments in one area’s natural resources and agricultural products with their access to or control of skilled handicraft labor elsewhere. As the 625

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

distributor and purveyor of these grain, salt, and handicraft products, traveling merchants rarely invested in the means of production. Perhaps that is why they acquired great profit from their long-distance transport and trade of these products, even while retaining a base of operations back in their Huizhou village. Two other strikingly different forms of family business merit mention. First, segments within these large branches might over several successive generations build up highly successful family businesses, which as the segment’s own membership expanded came to encompass a variety of distinct business ventures run by its members but funded by a wider group of fellow branch or lineage members. Cross-investment and crossshareholding among segments and even branches would have been the norm, but the segment retained a strong sense of its distinct identity through insistence on the importance of its members’ close descent ties and collective ancestral worship. Successful family firms evolved over three or more generations into “house firms,” which acquired considerable influence in major markets such as the salt trade in Yangzhou, the grain trade in Hankou, and the silk business in Suzhou. Second, at the opposite extreme were those family businesses which as segments or branches sought to expand their membership and capital holdings through samesurname alliances. These “family” businesses were formed not on the basis of ascertainable descent lines, but rather solely on the basis of a shared surname and the accompanying presumption of a common blood tie (in remote antiquity if necessary). Hunting up an undervalued ancestor from the historical record (or from collective memory) whom all interested parties could claim as a common ancestor, they redrew descent lines not merely to update their genealogy or to justify the construction of an ancestral hall, but, more importantly, to lay a kinship basis for plans of shared commercial investments and activities. From the late Ming this creative blend of kinship and business was commonly adopted by enterprising families throughout China, even in those areas where families ordinarily insisted on maintaining a strictly recorded line of descent for ancestral worship.18 Needless to say, such strategies seldom won universal approval from other branch or lineage members. All these types of family business enabled the managers to rely on fellow segment, branch, or lineage members to carry out numerous in-house transactions, and thereby, as Ronald Coase famously argued for firms in 18

McDermott, New Rural Order, vol. 2, Chapter 6.

626

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

general, reduce transaction costs and establish smooth relations with regular clients.19 Of course, these same practices could and did inflict unexpected losses in an economic crisis. If bankruptcy occurred, debtors could be charged and imprisoned. But in general that was not the response of lineage and branch members to this kind of financial crisis portrayed in late Ming and mid-Qing records. Normally, repayment was negotiated privately, outside the court and the prison. Given the variety of these family business arrangements and their selective openness to outside investments, it is not surprising that the involvement of kinsmen in commerce often took the form of commercial partnerships. Originating no later than the second century B C E, the practice of establishing commercial partnerships was first described in some detail in the eleventh and twelfth centuries. But only from the sixteenth century was it discussed in a wide variety of extant sources, such as contract form-books, descriptions of local merchant practice, exercises in mathematics treatises, and some official reports. (Actual partnership contracts are few, and directly relevant laws and legal case decisions virtually nil.) Men and women of virtually all social strata, not just full-time merchants, appear to have been keenly interested in forming, joining, and investing in commercial partnerships from at least the twelfth century. In this diverse record three general types of commercial partnership can be discerned: “commenda-like” (weituo 委托, fuben 副本), agency (jingshang 經 商), and joint-share (heben 合本, hegu 合股).20 Commenda-like partnerships, first mentioned in eleventh-century Kaifeng, required the participation of two kinds of investor, one (or several) who provided capital, and the other, a manager, who provided his time, labor, skills, and daring to search for trading profits, often in distant places. This partnership usually lasted for just a single trip (itself usually less than a year), for which the stay-at-home investor might have set the overall travel and commercial plan but seldom fixed the specifics of any journey. He let the manager handle decisions on prices, cargo volume, crew, and travel details. Clearly, the choice of manager – who had to be experienced, clever, and above all honest – was crucial to a venture’s success (the manager’s remuneration ranged from 30 to 50 percent of the net profits in the known cases; the rest went to the other investor(s). In the Yuan dynasty, the Mongol government sent forth nonChinese merchants into Central Asia to acquire luxury goods on such terms, 19

20

Ronald H. Coase, The Firm, the Market, and the Law (Chicago, The University of Chicago Press, 1988), p. 7. McDermott, New Rural Order, vol. 2, Chapter 5.

627

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

but under other dynasties the practice proved popular among private Chinese merchants with surplus cash and trustworthy assistants. Overseas traders also commonly relied on commenda-like partnerships in the Song and Yuan, and the practice appears to have continued through the Ming and presumably Qing, as would appear from M.A.P. Meilink-Roelofsz’s exhaustive research in the Dutch East India Trading Company’s archives: The entire overseas trade of the Chinese [in the Indonesian archipelago] was carried on almost exclusively in the form of commenda and only very few of the merchants on the Chinese junks were travelling to the Indonesian Archipelago with their [own] goods and capital. The money-lenders and suppliers of goods who remained behind in China derived huge profits from this commenda trade.21

Every year usually eight large Chinese vessels holding 300 or more men arrived at Melaka on the Malay peninsula, along with some smaller boats with crews of sixty to a hundred men (in 1626 overseas Chinese trade to all of Southeast Asia, including the Philippines, was estimated to involve over 300 ships a year). These merchants sold or bartered Chinese silk, porcelain, tea, and manufactures for pepper, drugs, ivory, tin, and valuable woods like sandalwood. On their return north, as on their voyage south, they tended to drop into various Southeast Asian and southeast Chinese coastal ports, usually in Guangdong or Fujian. Commenda-like arrangements were also popular and suitable for some long-distance domestic trade. Regional long-distance merchant groups in both the north and south, from Shanxi, Shaanxi, Dongtingshan (near Suzhou), and Jinhua prefecture (Zhejiang), practiced it regularly, with the principal variant being the relative share of the profits designated for the stayat-home investor and the traveling manager. In south China, where longdistance travel was commonly by boat, the managers in such partnerships might use their own vessel, and so arguably their overall share in the south tended to be higher than their counterparts’ in the north, where travel was more often by cart. Success in commenda partnerships clearly led some partners to renew their agreement and launch subsequent voyages. Others came to a natural end, or they evolved into agency partnerships, whereby the manager, having determined where he could find profits in a distant market, returned and settled there as the agent of one or more merchants, selling their goods at a commission rate. 21

M.A.P. Meilink-Roelofsz, Asian Trade and European Influence in the Indonesian Archipelago between 1500 and about 1630 (The Hague, Martinus Nijhoff, 1962), p. 265.

628

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

An agency partnership had certain clear advantages over a commenda-like arrangement for both parties. It required minimal capital outlay and expenditure, demanded less commitment from both parties, posed fewer risks, and could easily last longer, thus requiring less time for rearranging and managing its terms. If access to a local market was restricted or too costly, reliance on a local agent was a relatively safe way to test the waters, acquire a firm foothold, and then, on the basis of several years’ business returns, determine one’s future partnership policy. Meanwhile, these agents could also work easily as full-time innkeepers, shop managers, brokers, or warehouse managers, and so have other income sources. Indeed, both investors and agents could operate somewhat independently, to the extent of simultaneously having other business partnerships under different arrangements, even commenda, with other merchants. Problems would arise, however, in economic conflicts or crises, as neither the agent nor his paymaster was necessarily committed to the well-being of the other, and one partner could easily terminate the arrangement if he felt it was not to his advantage or if he had found a more suitable partner even under another agency contract. It is the third type of partnership, joint-share partnership, that proved most popular for long-distance trade and other kinds of family business ventures in Ming and early Qing commercial circles overall, especially among the Huizhou merchants. This form of partnership essentially had all partners share risks and losses, and profits and dividends, according to the amount of capital each had invested. Its key attraction was its toleration of varying oral understandings or written-contract terms to suit the particular needs and wishes of its participants. It had no intrinsic restriction on the number of potential investors and managers (the largest example from the late Ming had several tens of partners), on the amount of capital invested, on the duration of its partnerships (the longest duration reported from the Song was about ten years, that from the Ming at least three generations, and possibly one over 200 years), or the terms of profit distribution (some allowed a reduction for the managers, most did not). Crucially, the agency partnership did not fix the degree of partners’ liability, which varied depending on the degree of their involvement in management decisions: in general, partners who had no voice in management decisions apparently were liable only for their investments and not for their other private assets. In short, investors could put as much money as they wished into as many partnerships as their capital allowed, so long as they could persuade others to join their venture, or be allowed to join others’ ventures (the role of widows or wives investing family capital or their own dowry awaits investigation). As a rule, commercial partners were 629

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

expected to inform each other of their additional financial commitments, a pledge that undoubtedly was honored largely in the breach. The durability of these commercial partnership arrangements throughout the late imperial period argues strongly for their adaptability and acceptability to most merchants. How successful they were in persuading hoarders to invest, establishing long-lasting trading networks, or achieving profit levels that matched those to be found in well-run state monopolies remains to be determined. But, at the very least, they were successful enough to support up to modern times the efflorescence of kinship-based institutions – the lineage, the ancestral hall, and the family business – whose long-term functioning made an indispensable contribution to the premodern Chinese practice of commerce. This analysis, however, also needs to consider how Chinese merchants made use of the Chinese state and its capital resources.

Government, Merchants, and the Late Imperial Chinese Economy The retreat of the late imperial Chinese state’s direct control over the economy is commonly traced to the protracted civil wars and foreign invasions that in the mid-eighth century destroyed the Tang dynasty’s “equalfield” land distribution system. Landownership fell into the hands of a multitude of private parties obliged henceforth to pay the state not rent but land taxes. Likewise, Tang commercial policies were fundamentally transformed, so that officials no longer determined the price, location, and time for the transaction of goods. By the eleventh century such powers lay overwhelmingly in the hands of private merchants and their associations. The failure of the late eleventh-century state-centered reforms espoused by the chief councilor, Wang Anshi, to reverse these trends and thereby strengthen the Song government’s management of the economy and its ability to resist foreign invasion confirmed in many officials’ minds the wisdom of withdrawing from strict economic management (see the Lamouroux and von Glahn chapter in this volume). Despite a few periods of revived central government interest in economic matters, such as the Hongwu reign (1368–1398), the years of Zhang Juzheng’s administration (1572–1582), and the Yongzheng reign (1721–1735), Ming and Qing governments were more concerned with supervising and taxing production rather than directing or determining it. Predictably, their vague rules on market transactions harped on abuses less in production than in circulation, such as price manipulation, private weights and measures, and lack of market access. 630

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

In other words, the state’s retreat from direct control of production and exchange signaled its disinterest in strictly laissez-faire policies and a corresponding interest in hybrid public–private activities, mainly in the commercial sector, that its officials judged politically and militarily most sensitive, convenient to supervise, and rewarding to squeeze and tax: the highly productive paddy fields of the Yangzi delta; the salt fields of coastal Huainan north of the Yangzi delta; the northern border exchange of tea for horses; the copper, iron, and coal mines in Yunnan, Beijing, and Shanxi; the porcelain kilns of Jingdezhen; the large ports along the Yangzi river and the Grand Canal; and the major coastal ports engaged in the lucrative overseas trade. Consequently, state policy toward merchants and commerce could be described as concurrently a mixture of neglect, exploitation, stimulation, and collusion, depending on the time, place, size, and type of transaction. For example, in sharp contrast to their predecessors, officials in the seventeenth century and later showed a keen interest in pawnshops and other financial services. They recorded the names of pawnbrokerages, they calculated their number and capital assets in the delta and elsewhere, and from 1623 they levied an annual tax on each shop’s estimated assets. Over time, the mode of the post-Tang state’s engagement with its monopolies like salt, tea, and iron would evolve, from assignment to supervision, to loose delegation, to farming out, and then to annual payment of a fixed amount of revenue from the economic activity concerned. And so, not surprisingly, the nonagricultural sector’s share of government tax revenue fell considerably from its Song levels. Throughout the Ming and Qing dynasties agriculture remained the principal source of government revenue. This understanding of the late imperial state’s relationship to the economy has much merit. Yet it needs qualification by considering three concurrent changes that indicate a much more complicated story about late imperial state–merchant relations and highlight another important source of merchant capital. First, during these centuries the number and variety of informal state supervisory personnel increased considerably. State personnel consisted of much more than the central government and its prefectural and county government officials and offices with their local clerks and other yamen support, including those hired above the official quota. Particularly at lower levels, the regulatory and clerk staffs expanded, as did the crowds of authorized and unauthorized brokers who privately sought to mediate deals with and between merchants. In theory, the state could have acknowledged the increase in merchant power and influence brought on by commercial expansion and investments in the private market. But, as we have seen in the 631

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

state’s handling of the local brokers, while all levels of government might have distrusted one another, they shared a fundamental distrust of merchants and a deep reluctance to grant them independent institutional powers. As a result, merchants’ relations with the state grew far more complicated, opaque, and unpredictable. Official policy might seek to stimulate trade, but did so often within a counterproductive administrative framework. Second, some wealthy merchants were just as interested in growing rich off the state as the state sometimes was in feeding off them and their private riches. This mutual engagement is best seen in the operation of a state monopoly like salt, as when in the Ming dynasty salt merchants acquired great fortunes not just by transporting and selling government salt but also by illicitly cornering the market in the salt certificates that the government required them to buy in order to acquire and transport salt for profit. Their success at milking this source of capital at the government’s expense eventually obliged the government from 1617 to give certain salt merchant families permanent franchises in the lucrative salt monopoly, and thus easy and by extension regular access to large amounts of cheap capital for use in their commercial ventures. Even European traders along the China coast heard of their financial clout and observed how just a close association with salt merchants assured another merchant easy access to capital. And, third, great commercial fortunes hastened crucial changes within the ranks of Chinese merchants: a proliferation in their number, functions, and places of origin. Their ranks came to include countless peddlers, wholesalers, retailers, agents, warehouse operators, shopkeepers, long-distance traders, salt merchants, grain speculators, brokers, pawnbrokers, moneylenders, accountants, bookkeepers, and managers, each with a separate task in the operation of different markets. Men holding these jobs came increasingly from a non-commercial background, as commerce became far more widely practiced full-time and part-time in some parts of the southeast, to the detriment of agriculture. The most often mentioned of these new commercial parties was the sojourning merchant, who transported goods, many of them daily commodities, over long distances on major roads and waterways. As their number and significance grew from the mid-Ming, merchants came to be characterized not just by their function but also by the type of goods they traded, the size of their capital assets (up to a million taels of silver by 1600 and more than ten times that amount by 1800), and their place of origin. As a rule, native places commonly served as popular, informal categories for distinguishing groups of sojourning merchants, none of whom were cohesively organized by common investments or common 632

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

strategies. In fact, these regional categories were coined by nonlocals, who most likely observed the tendency of merchants from one particular area to do business together and to rely on distinctive local customs of transaction, worship, cuisine, and dialect. Such local ties mattered, but apart from Shanxi and Shaanxi merchants a kinship tie, or even common surname, usually mattered more. The two earliest and best-known of these regional groups of merchants, those from Shanxi province and Huizhou prefecture in Anhui province, dominated major markets in separate areas of the empire, Huizhou in the south (mainly the Yangzi valley and the Grand Canal corridor) and Shanxi in the north (the Yellow River basin and, in the Qing, Beijing). Both Shanxi and Huizhou merchants first rose to prominence in delivering and selling these separate areas’ salt as well as their own regional specialty goods (such as tea and timber from Huizhou). The local branches of their house firms almost certainly co-ordinated prices among themselves, as they did their overall marketing and distribution strategies. Yet, crucially, there is no evidence that the firms in any regional (as opposed to kinship) grouping co-ordinated among themselves any single coherent and overriding strategy for the multiple markets they may have collectively dominated. Their perspective was primarily family-centered, and as outsiders their involvement in local politics was severely restricted. The Huizhou and Shanxi merchant groups also expanded into financial activities and services. From the late fifteenth century onwards Huizhou merchants ran pawnshops in towns and even some villages throughout east China from Beijing to Guangzhou, as did Shanxi merchants in the north. Thereafter, the Shanxi merchants spread their financial businesses in the west and south during the early Qing. By the early eighteenth century they had replaced Huizhou financiers as the dominant operators of pawnshops in north China, and from mid-century they were opening up native banks and remittance agencies throughout the country. Their prolonged expansion up to the late nineteenth century thus contrasts with the commercial contraction suffered by Huizhou merchants. From the late eighteenth century, the Huizhou merchants’ market share and profits faded, even in the Yangzi valley salt and timber trades that they had long dominated. And, in central and south China – for the first time in at least three centuries – they found it hard to fend off competition from Shanxi merchants in financial services as well as they did in commodity markets from smaller southern merchant groups in Guangdong, Zhejiang, Jiangxi, and Hubei. 633

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott

Over time, the size of all these regional groups grew. By the late Ming most would have consisted of thousands of individuals (in hundreds of family units), while the Huizhou and Shanxi merchant groups would have already by then included tens if not hundreds of thousands (reportedly 70 percent of late Ming Huizhou’s adult males worked as itinerant traders). But in most places they frequented, they rarely numbered in the hundreds (clear exceptions include, once again, the Huizhou and Shanxi merchants at particularly large trading nodes like Yangzhou, the center of the Yangzi valley salt trade, Linqing along the Grand Canal in the Ming, and Hankou along the Yangzi in the Qing. In the large cities where sojourning merchants congregated, merchant associations gradually emerged to serve their needs. The pace of this change was glacial. Since at least the eighth century, governments had organized merchants into groups by their line of trade (known as hang 行, or “alleyways”). This policy, whereby city streets were lined with shops in specific trades (e.g., jewellers, butchers, etc.), arguably suited traders and customers in their native town. But it ill-suited outsiders like sojourning merchants, who initially resorted to approved religious establishments like shrines and temples for their accommodation, storage, and meetings with other “outsider” merchants. From the mid-Ming, however, as the number of sojourning merchants grew, travelers from certain areas set up their own establishments in specific cities far from home. Probably to dispel official fears about trading abuses like price fixing, these native-place associations initially accommodated men of all occupations from a specific area (though many associations in Beijing continued to host only visiting officials and examination candidates). Only from the early and mid-Qing did native-place associations commonly acquire an exclusively commercial character, in that they served all merchants from a specific place or, as in many European guilds, from a particular trade (as in Hankou from 1688). The heads of these associations were normally chosen through consultation between their leading members and local officials, with preference given to rich and experienced merchants already heavily committed to the trade (inheritance of the post was common). Though our knowledge of the private deliberations of these associations on local prices, market conditions, and trading practices remains painfully unclear, most likely merchant members used these institutions formally or informally to co-ordinate (i.e., fix) prices, acquire capital, and help one another penetrate local markets without being exploited by brokers and other locals. 634

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

Merchants and Commercial Networks

Thus, instead of confronting the imperial state, rich merchants throughout late imperial times would have pursued their interests less through these associations than through their contacts within the corridors of power. Just as they had bested Ming officials and eunuchs in conflicts over salt certificates and served, as noted above, as unacknowledged Qing government censors in Suzhou’s private book trade, so would they have sought work and assignments within relevant state institutions to pursue their own commercial interests. This strategy would be evident, I suggest, in the operation of government monopolies (not just salt), but also in the management of military-related industries, the workings of the financial world (including government investments in pawnshops), mineral extraction, and tax decisions on commerce. But these are just the economic dimensions of the late imperial political economy. Once we factor regional merchant interests into the analysis of noneconomic issues, our study of merchants in the world of Ming and Qing politics and society will expand far beyond the confines of the late imperial economy, and will need to explore merchant impact, or otherwise, on internal government decisions regarding economic policy and practice. Responsibility for the nineteenth-century breakdown of the Chinese economy, I suggest, lies in this opaque world replete with private arrangements which Chinese officials and merchants did their best to prevent erupting into public scandals, or even being studied by historians.

Further Reading Brook, Timothy, The Confusions of Pleasure: Commerce and Culture in Ming China (Berkeley, University of California Press, 1998). Fan Jinmin 范金民, Ming Qing Jiangnan shangye de fazhan 明清江南商业的发展 (Nanjing, Nanjing daxue, 1998). Faure, David, China and Capitalism (Hong Kong, Hong Kong University Press, 2005). Fu Yiling 傅衣凌, Ming Qing shidai shangren ji shangye ziben 明淸時代商人及商業資本, 2nd ed. (Beijing, Renmin chubanshe, 1980). Fujii Hiroshi 藤井宏, “Shin’an sho¯nin no kenkyu¯” 新安商人の研究, To¯yo¯ gakuho¯ 東洋学 報 36.1 (1953), 1–44; 36.2 (1953), 32–60; 36.3 (1953), 65–118; 36.4 (1953), 115–45. Grove, Linda, and Christian Daniels (eds.), State and Society in China: Japanese Perspectives on Ming–Qing Social and Economic History (Tokyo, University of Tokyo Press, 1984). McDermott, Joseph P., The Making of a New Rural Order in South China, vol. 2, Merchant, Market, and Lineage in Huizhou, 1500–1700 (Cambridge, Cambridge University Press, 2020). Qiu Pengsheng 邱澎生, Dang falü yushang jingji: Ming Qing Zhongguo de shangye falü 當法 律遇上經濟: 明清中國的商業法律 (Taipei, Wunan tushu gongsi, 2008).

635

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

joseph p. mcdermott Rawski, Thomas, and Lillian Li (eds.), Chinese History in Economic Perspective (Berkeley, University of California Press, 1992). Shiba Yoshinobu 斯波義信, So¯dai sho¯gyo¯shi kenkyu¯ 宋代商業史研究 (Tokyo, Kazama shobo¯, 1968). Skinner, G. William (ed.), The City in Late Imperial China (Stanford, Stanford University Press, 1977). So, Billy K.L. (ed.), The Economy of Lower Yangzi Delta in Late Imperial China (London, Routledge, 2013). Usui Sachiko 臼井佐智子, Kishu¯ sho¯nin no kenkyu¯ 徽州商人の研究 (Tokyo, Kyu¯ko shoin, 2005). Terada Takanobu 寺田隆信, Sansei sho¯nin no kenkyu¯: Mindai ni okeru sho¯nin oyobi sho¯gyo¯ shihon 山西商人の研究: 明代における商人および商業資本 (Kyoto, Do¯ho¯sha, 1972). Wang Yuming 王裕明, Ming Qing Huizhou dianshang yanjiu 明清徽州典商研究 (Beijing, Renmin chubanshe, 2014). Xu Dixin and Wu Chengming (eds.), Chinese Capitalism, 1522–1840 (London, St. Martin’s Press, 2000). Zelin, Madeleine, “Chinese Business Practice in the Late Imperial Period,” Enterprise and Society 14.4 (2013), 769–93. Zhang Haipeng 张海鹏 and Zhang Haiying 张海瀛 (eds.), Zhongguo shida shangbang 中国 十大商帮 (Hefei, Huangshan shushe, 1993).

636

https://doi.org/10.1017/9781108587334.018 Published online by Cambridge University Press

17

Foreign Trade angela schottenhammer

Characteristics of China’s Foreign Trade over the Longue Durée Foreign trade always mattered in imperial China. Especially during the middle period and early modern times, China experienced enormous growth and expansion of foreign trade. According to Confucian concepts, merchants belonged to the lowest echelon of society; agriculture, not trade, was considered the basis of a stable state and society. Most Chinese governments indeed sought to maintain more or less strict control over foreign commerce and those who were responsible for it. But one has to emphasize the co-operative rather than antagonistic relationship with markets and with the merchant class during most of China’s imperial history. In addition, we can observe certain characteristics and qualitative changes throughout the centuries. As will be shown in this chapter, the wish for control first of all may not be frivolously identified with an interest in obstructing trade. Second, state measures to promote and expand trade (both domestic and foreign) also should not be taken as an interest in or development toward modern capitalist or market economy structures. Third, one should distinguish between official, tribute, and private (legal or illegal) trade, although clear-cut distinctions are not always easy. Envoys of official diplomatic missions, for example, were permitted to conduct private trade under special regulations, and at times they were involved in contraband trade. Finally, it is important to keep in mind that China was a wealthy country with a high level of agricultural and manufactural output, in other words with a solid economic foundation. The basic official mechanisms for foreign trade were the tributary system, border markets, and offices that were first temporarily established and later

637

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

institutionalized. The Chinese court time and again attempted to squeeze trade into its ideological framework, resulting in a tributary system that was governed by ritual formalities. Under the tributary system, permission to trade was deemed a magnanimous gesture to foreigners, expressed as “being gracious to those from afar” (rouyuan 柔逺). The Chinese sought to incorporate foreign countries into their imaginary world-empire, their “one family within the four seas” (sihai yijia 四海一家), and ranked them according to the supposed closeness of kinship ties. This practice was particularly cherished during early Ming times, when private maritime trade was interdicted (1371–1567). But actually the commercial dimensions of foreign trade played an ever-increasing role also for Chinese governments. Moreover, the magnitude, nature, and composition of foreign trade and its impact on local economies changed over time, as did China’s borders. Although an extremely rich – in terms of agricultural and industrial output – country, or better “national economy,” China until early Ming times was actually becoming poorer in terms of monetary values in the state coffers with increasing integration into supra-regional, “global” economic structures.1 Chinese merchants and government authorities to a great extent exchanged valuable manufactured goods, products that possessed high added value, for aromatics and spices that eventually, due to consumption, disappeared from the internal circulation sphere. Huge quantities of manufactured goods and monies, especially copper coins, were sold or smuggled abroad. From the late eleventh century to the early fourteenth, large amounts of copper coins and silver ingots flowed out to Japan, Korea, Southeast Asia, Central Asia, the Arab countries and beyond as far as Europe.2 Yet, in the late sixteenth century, China became the recipient of a massive inflow of silver bullion from Japan, Europe, and Spanish America. This trend in turn was reversed in the early to mid-nineteenth century when British merchants started to import Indian opium – the (medicinal) drug of early modernity, one could say – into China.3

1

2

3

Kangying Li, “A Study on the Song, Yuan and Ming Monetary Policies within the Context of Worldwide Hard Currency Flows during the 11th–16th Centuries and Their Impact on Ming Institutions,” in Angela Schottenhammer (ed.), The East Asian Maritime World 1400–1800: Its Fabrics of Power and Dynamics of Exchange (Wiesbaden, Otto Harrassowitz, 2007), pp. 99–136. Akinobu Kuroda, “The Eurasian Silver Century, 1276–1359: Commensurability and Multiplicity,” Journal of Global History 4.2 (2009), 245–69. The impact of opium importation as the primary cause of the outflow of silver from China has recently been critically reassessed; see Richard von Glahn, “Cycles of Silver in Chinese Monetary History,” in Billy K.L. So (ed.), The Economy of Lower Yangzi Delta in

638

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Song Dynasty China: The Economic “Motor” of Asia After the An Lushan rebellion in the mid-eighth century, the heyday of the traditional overland Silk Roads had passed and we observe a shift from overland to maritime routes as the principal avenues of China’s foreign trade (see the chapter by Liu in this volume). With the final collapse of the Tang dynasty in 907, China became divided into ten independent kingdoms that coexisted for more than half a century before it was reunified under the Song dynasty (960–1279). The Song empire was smaller than the Tang, but it had ready access to the sea, and the focus of economic activities in general moved from north to south China. The shift from overland to maritime routes was also caused by the contemporary geopolitical multistate order, by military conflicts across the overland routes, and by the establishment of hostile regimes – the Khitan Liao (907–1125), the Tangut Xi Xia (1038–1227), and later the Jurchen Jin (1115–1234) dynasties – on the Song’s northern frontiers. It also was accompanied by active official sponsorship of commerce, which was assessed increasingly positively as a source to be tapped rather than a malign parasite. The introduction of a new state ideology that reconsidered the relationship of the state to its economic resources spurred the expansion of both domestic and foreign trade (see the chapter by Lamouroux and von Glahn in this volume). It resulted in an intensification of economic relations with other Asian countries, but also in qualitative changes, with coastal regions experiencing the highest degree of commercialization. Nevertheless, we have to be careful about not marginalizing trade with Inner Asia. Horses from Tibet, Mongolia, and Transoxania, as well as frankincense from the Islamic world, were exchanged for Chinese silks and Sichuan tea via overland routes. In 1077, for example, a tribute mission from Khotan arrived in Kaifeng with over twenty tons of aromatics, and another embassy transported sixty-five tons of incense from Khotan to Xizhou.4 Emperor Shenzong (r. 1068–1085) in particular explicitly intended to facilitate trade and commerce in order to fill state coffers in the face of increasing military expenditures. He promoted a kind of state-controlled exchange by first permitting and then requiring taxpayers to pay an increasing share of their taxes in the form of money, basically copper coins and silver. To be able

4

Late Imperial China: Connecting Money, Markets, & Institutions (London, Routledge, 2013), pp. 45–58. But it is a fact that China paid huge amounts of silver for opium. Robert Hartwell, “Foreign Trade, Monetary Policy, and Chinese ‘Mercantilism’,” in Kinugawa Tsuyoshi 衣川強 (ed.), Ryu¯ Shiken hakase sho¯ju kinen So¯shi kenkyu¯ ronshu¯ 劉子 健博士頌壽紀念宋史研究論集 (Kyoto, Doho¯sha, 1988), pp. 454–88.

639

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

to pay taxes in money, taxpayers had to sell something first. In this way production for market exchange steadily increased, a development that was especially promoted during the reform period under state councillor Wang Anshi. This trend was accompanied by improvements in transportation networks and technological advances in agriculture, waterworks, mining and metallurgy, porcelain, textile manufacture, printing, geography and mapping, navigation, and shipbuilding. A chief medium of exchange in foreign trade was bullion, especially silver. Nevertheless, much foreign trade can be characterized as a form of barter, for example, the exchange of Chinese tea for Central Asian horses. Foreign trade was also taxed at special customs offices located at the borders and ports. The customs levy varied over time; rates ranging between 10 and 20 percent (depending on the commodity) were the rule. Customs officials usually took a certain percentage of imported goods in kind and forwarded them to the central government (a practice known as choujie 抽解). Merchants could then sell the remaining goods in local markets. Governments also applied pre-emptive rights and used public funds to purchase foreign goods, or simply monopolized the resale of certain products that could be sold profitably, such as incense, ivory, and tortoiseshells. As early as 977, the director of the Aromatics and Drugs Warehouse, Zhang Sun, suggested that a Bureau of Licensed Trade (queyishu 榷 易署) be established. The government would sell aromatics, drugs, and other foreign merchandise in exchange for bullion or textiles. Zhang reckoned that these sales would yield an annual profit of 500,000 guan (strings of coin) to supplement the government’s revenue. The emperor approved the proposal, which initially generated profits of 300,000 guan a year and increased steadily to 500,000 a year. This amount was equal to approximately 2–3 percent of the Northern Song’s total revenue (see Table 17.1). Tribute lists, customs reports, and other sources reveal that the most important Song imports were horses, Japanese sulfur, Vietnamese gharuwood, Arabian frankincense, Javanese sandalwood, Malayan sappanwood, spices such as black pepper from Java, cloves and nutmegs from the Moluccas, ivory and camphor from Borneo and Sumatra, cinnabar, and Korean ginseng. Exports mostly consisted of silks, tea, salt, manufactured goods such as porcelain and ironwares, and re-exported monopolized imports. By the late eleventh century, it has been estimated, the total volume of international commerce at Chinese seaports reached the equivalent of more than 520,000 kilograms of silver annually, or approximately 1.7 percent of GNP.5 5

Hartwell, “Foreign Trade and Chinese ‘Mercantilism’,” p. 453.

640

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Table 17.1 State income from foreign trade (figures in guan) c. 977–90

1077 c. 1087–98 1128 c. 1128–1134 1137 1147 1159

Maritime trade 300,000 to 500,000 Northern trade (e.g., with Khitan) 400,000 (silver, coins, fabrics, sheep, horses, camels) 540,173 in cash, grain, silver, aromatics, spices, and medicinal drugs c. 400,000 annually c. 2,000,000 annually 980,000 (Quanzhou only) c. 1,000,000 annually c. 2,000,000 annually c. 2,000,000 annually

Source: Hugh R. Clark, “The Politics of Trade and the Establishment of the Quanzhou Trade Superintendency,” in Organisation Committee of the International Seminar on China and the Maritime Silk Road (ed.), China and the Maritime Silk Route (Fuzhou, Fujian renmin chubanshe, 1994), p. 387

Upswing of Maritime Trade Starting already in the tenth century, rulers increasingly began to treat maritime trade as a way to underpin the economic foundations of their regimes. They sought to use the income derived from this trade not only for their own private consumption needs, but also to defray the political and economic maintenance of the state. Governments undertook many efforts to expand and promote maritime trade. Coastal kingdoms such as Wu-Yue, Min, and Southern Han established maritime trade relations with the Khitan Liao, the Koryŏ kingdom in Korea, Japan, and Southeast Asia. The active involvement of the Southern Han in maritime trade is attested by a shipwreck, the Intan wreck, that sank in the early tenth century fully loaded with Chinese and Southeast Asian ceramics, incense, copper and tin ingots, gold jewelry, Southern Han lead coins, and above all almost 190 kilograms of silver ingots that had been submitted to the Southern Han treasury as tax payments.6 The early Song government established a series of maritime trade offices (shibosi 市舶司) at major ports. In 971, the first such office was opened in Guangzhou. Two years later a ban on Chinese merchants venturing overseas signaled the government’s intention to control foreign trade. Then, in 987, Emperor Taizong (r. 976–997) sent four missions into the “South Seas” to invite traders to come to China. Two years later a second maritime trade 6

Denis Twitchett and Janice Stargardt, “Chinese Silver Bullion in a Tenth-Century Indonesian Wreck,” Asia Major, third series 15.1 (2002), 23–72.

641

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

office was opened in Hangzhou, followed by additional offices at Ningbo in 992 and at Quanzhou in 1087 (Map 17.1). Much of this maritime trade took place under the aegis of tribute trade. In the eleventh century – especially in concert with the Wang Anshi reforms – the state steadily tried to increase its control of foreign trade and also established monopoly offices on trade in domestic goods, for example the sale of salt and tea. It also monopolized the purchase and distribution of “aromatics and medicines” (xiangyao 香藥) – a general term that included not only products such as incense, scented woods, perfumes, and medicines, but also pearls, ivory, rhinoceros horn, ebony, and sappanwood – imported from abroad. Their importance can also be seen in the fact that the government determined the promotion prospects of officials serving in the maritime trade offices by the amount of frankincense that arrived in their jurisdiction. Yet government authorities encountered various problems in their attempts to control foreign trade more strictly. As far as the administration of maritime trade is concerned, we observe a steady back-and-forth between central control and local autonomy. As time went by, the early attempts at strict centralized control gave way to a more decentralized system involving a host of maritime trade offices or, after 1080, of maritime trade superintendencies (shibo tijusi 市舶提擧司); at one time or another these could be found in up to ten ports. Another important characteristic of China’s maritime trade is the fact that initially it was not Chinese who ventured abroad but foreigners who came to China. While maritime exchange with Southeast Asia and parts of India had existed for centuries, it was conducted basically by Persian Gulf traders. “Iranian” (bosi 波斯) and later “Arab” (dashi 大食) merchants – terms that should probably be understood as generic references to ships from the Persian Gulf area, probably crewed by people of diverse ethnicities and religions – were the ones who initiated more routinized, long-distance maritime commerce with China. It was only in the late eleventh century that Chinese themselves increasingly sailed to Southeast Asia in their own ships. This development was propelled by a government decision in 1090 to liberalize maritime trade by no longer obliging merchants to register at ports with an official maritime trade office. Still, state regulations forced them to concentrate on destinations in East and Southeast Asia. During the eleventh century, Quanzhou overtook Guangzhou as the major port of call for foreign merchants from the “southern seas” (nanyang 南洋) – a term referring to Southeast Asia and beyond, including the Indian Ocean. A report on foreign countries whose ships frequently called at 642

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

LIAO

JAPAN KORYŎ

XIA Hakata

Mizhou

Sinan

Kaifeng

Suzhou Hangzhou

TUFAN (TIBET)

Ningbo Wenzhou

SONG

Okinawa

Fuzhou Guangzhou

DALI

Quanzhou

Nanhai no. 1

DAI VIET PAGAN

0

500

1000 km

Maritime trade office Shipwreck

A M PA

LUZON Huaguang Reef no. 1

CH

ANGKOR

MA LA

Y

LA SU

ŚR ĪV IJA YA

PE NI N

BORNEO

Melaka Belitung

Palembang

Intan

Map 17.1 Maritime East Asia in Northern Song China Source: adapted from J.W. Chaffee and D. Twitchett (eds.), The Cambridge History of China, vol. 5, part 2, Sung China, 960–1279 (Cambridge, Cambridge University Press, 2015); Benjamin Z. Kedar and Merry E. Weisner-Hanks (eds.), The Cambridge World History, vol. 5, Expanding Webs of Exchange and Conflict, 500 C E –1500 C E (Cambridge: Cambridge University Press, 2015), Map 19.2, p. 526; and John Chaffee, The Muslim Merchants of Premodern China: The History of a Maritime Asian Trade Diaspora, 750–1400 (Cambridge, Cambridge University Press, 2018), Map 2

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

Quanzhou lists more than thirty locations overseas, including the Islamic world, the S´rı¯vijaya kingdom in Sumatra, Cambodia, Pagan, Borneo, Java, the Malay peninsula, Angkor, and chiefdoms across the Philippine islands.7 Japan also figured prominently in Song China’s foreign trade relations. Imports from China included great quantities of copper coins, books, ceramics, and a whole range of luxury goods known in Japan as karamono 唐物 (that is, “Chinese goods,” regardless of their original source). Massive export of Chinese coins to Japan began as early as the 1170s and peaked in the century from 1250 to 1350, when paper money displaced coin for many monetary functions in China. In Japan, Chinese merchants exchanged copper coins for gold and handicrafts such as fans and swords, but primarily for sulfur, timber, and mercury; that is, for bulk commodities that were essential for military and construction purposes in China. The predominant position of sulfur in Sino-Japanese trade at this time has prompted one scholar to refer to this trade as the “sulfur road.”8 During this period China exported large quantities of manufactures, such as copper coins, ceramics, silks, books, and ironwares, durable goods that possessed high added value, while mainly importing exotic natural products and raw materials, especially aromatics, medicines, spices, scented woods, and the like. Song officials believed that the drain of copper coins abroad severely depleted the domestic money supply, but restrictions prohibiting the exportation of coin remained fruitless. In 1215, Chen Qiulu reported to the court that “large foreign ships fully loaded with our coins are sailing overseas. The coins are the pillars of our economy, but we exchange them for useless and alien exotic things. We are suffering tremendous losses.”9 By late Song times, the pattern of imported goods had changed, and luxury goods were increasingly replaced by bulk commodities. Export trade changed as well. Silk – China’s principal export in the heyday of the Silk Road – was supplanted by ceramics as China’s main export commodity. The invention of true porcelain in the tenth century spurred the growth and sophistication of Chinese ceramic production. In the Southern Song period the porcelain industry in south China became increasingly oriented to export markets. Song ceramics have been excavated all over the shores of the Indian Ocean as far as the Persian Gulf, the Red Sea, and the eastern coast of Africa.

7 8

9

YLMC, 5:19b–20a. Yamauchi Shinji 山內晉次 coined the term “sulfur road” to emphasize the central place of sulfur in Sino-Japanese trade; see his Ni–So¯ bo¯eki to “io¯ no michi” 日宋貿易と『硫黄 の道』(Tokyo, Yamakawa shuppansha, 2009). SS, 180, p. 4399.

644

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Archaeological finds of shipwrecks provide us with an increasingly better picture of the quantity and quality of maritime commerce. The Quanzhou wreck, dated to 1277, is well known: the ship was carrying a multitude of goods labeled as the property of the “Southern Family” (nanjia 南家), a reference to the branch of the imperial clan that resided in Quanzhou, indicating its involvement in the venture and possibly even ownership of the vessel. Other important finds include the Nanhai I wreck (late twelfth century), found off the coast of Guangdong with a large cargo of ceramics, mostly from Fujian but including some Jingdezhen and Longquan wares; the Huaguang Reef no. 1 wreck (late Southern Song) in the South China Sea, laden mostly with Jingdezhen porcelains; and the Sinan wreck (dated c. 1323), a vessel en route on a return voyage from Ningbo to Japan that carried more than 8 million Chinese coins (totaling twenty-eight tons) in addition to ceramics and other wares (Map 17.1). All of these wrecks attest to the prominent role of Chinese ceramics and coins as export goods. Huge quantities of Song coins have been excavated in Japan. Some Song coins exported to Japan and Southeast Asia were used not as money but for religious purposes. Recent isotope analysis of religious bronze objects from Japan, such as Buddhist statutes and bells, has substantiated that they must have been cast using melted Chinese copper coins as raw material.10

Foreign Trade along the Continental Frontiers Loss of easy access to the traditional overland Silk Road due to political instability and territorial fragmentation acclerated the shift to the sea, but it did not mean that foreign trade with continental neighbors was entirely interrupted. Various commodities still reached China through transcontinental routes, and goods were exchanged at special trading posts along the borders. Part of the trade was barter or took the form of tributary exchange, with foreign envoys receiving Chinese gifts in return. Northern and northwestern neighbors regularly brought furs and livestock, such as horses, sheep, and camels. Of particular importance for China was the trade in horses. Because the Song did not control any premium horse-breeding areas, it never possessed a sufficient supply and was forced to purchase virtually all of its warhorses. The Song court tried to “pacify” its Central Asian neighbors by permitting them to exchange horses for tea (which remained fresh longer than other beverages and was consumed as 10

O¯ta Yukio 大田由紀夫, “12–15 seiki shoto¯ Higashi Ajia ni okeru do¯sen no ryu¯fu: Nihon, Chu¯goku o chu¯shin toshite” 12–15 世紀初頭東アジアにおける銅銭の流布: 日本中 国を中心として, Shakai keizai shigaku 社会経済史学 61.2 (1995), 161–2.

645

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

a stimulant after long exposure to cold climates) at special border markets. The tea and horse trade was centered in Sichuan and Shaanxi. Approximately 2,600–3,000 tons of tea annually were exported and around 10,000 horses purchased. But the sale of horses was time and again interdicted by both the Liao and the Xi Xia. As a rule, official, government, and private trade were strictly demarcated. Official trade was a government monopoly with officials sent from the capital, Kaifeng, to supervise trade at the border trading posts. The Song annually sent 100,000 liang of silver and 200,000 bolts of silk as tribute to the Khitan, an amount that was later raised to 200,000 liang of silver and 300,000 bolts of silk. Much of this silver, however, returned to Song territory through purchases of Chinese goods. On average, it has been estimated, China’s trade with the Liao showed a favorable balance of 800,000 guan, of which the share of official trade constituted about 400,000–500,000 guan.11 The officials responsible for the supervision of border trade also settled disputes over prices and had to keep their eyes open for espionage and smuggling activities. Contraband trade in forbidden items such as books, maps, weapons, coins, and salt could be especially lucrative to smugglers. Trade with the Khitan Liao was carried out through both official and private channels. Officials in charge of cross-border trade were provided with money or goods to carry out the exchanges. Merchants engaged in private trade had to pay the government a fixed commercial tax and a brokerage fee. The trade mainly consisted of an exchange of pastoral products, such as livestock (horses, sheep), furs (white marten, fox), woollen cloth, carpets, brocade, metal items (silver and gold ornaments, iron armor suits), slaves, and timber from the Liao for silks and silk brocades, tea, gold, silver, and a variety of local products or re-exports of exotic goods from the southern seas. Fragments of Song porcelain discovered at Liao sites attest to the Liao elite’s appreciation of this sophisticated Chinese commodity, but the most prized import from the Song seems to have been silk. Under the terms of the Treaty of Shanyuan (1005), the Song agreed to open five permanent trading posts for trade with the Liao, while the Liao opened three of its own markets. Unfortunately, we do not possess much information about these frontier trade activities, but a few records provide us with some insights. Archaeological excavations such as the Xuanhua tomb complex in Hebei show that Buddhism played a major role in Liao society and at 11

Yoshinobu Shiba, “Sung Foreign Trade: Its Scope and Organization,” in Morris Rossabi (ed.), China among Equals: The Middle Kingdom and Its Neighbors, 10th–14th Centuries (Berkeley, University of California Press, 1983), p. 98.

646

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

least some Buddhist texts may have entered the Liao sphere of control from China as well as from Xi Xia territory. The tomb inscription for Han Shixun (d. 1110), a wealthy merchant engaged in cross-border trade living in the Liao realm, observed that “enduring the hardships of travel, he bought cheap and sold dear, his ambitions entirely concentrated on business and trade, and before ten years had passed, his family had become wealthy. He transformed them from poor to rich, and turned frugality to abundance.”12 Trade with the Tanguts remained restricted, as the latter did not possess many resources, but partly also because of the repeated hostilities between Song and Xi Xia. The peace treaty signed at the conclusion of the 1042–1044 Song–Xi Xia war forced the Song to pay the Xi Xia annual tribute of 50,000 liang of silver, 130,000 bolts of silk, and 30,000 jin of tea; an additional 22,000 liang of silver, 23,000 bolts of silk, and 10,000 jin of tea were nominally designated as gifts to the Xi Xia. Official trade was conducted at designated border markets, and also horse-purchasing marts were established. Contraband trade also flourished along the Song–Xi Xia borders. The Tanguts were primarily interested in obtaining Song weapons, printed books, and copper and iron coins. For the Song’s supply of horses, trade with the Qingtang Tibetans also was essential, because the Liao and Xi Xia repeatedly imposed embargoes on the exportation of their own horses. The Song government purchased 15,000 to 20,000 horses annually from the Qingtang, primarily in exchange for tea. Some 5 million jin of tea produced in Sichuan (approximately 17 percent of total Sichuan production) was bartered for horses with the Qingtang. Qingtang merchants also offered mercury, musk, and furs, and served as middlemen for trade in “western” goods from Central and Western Asia. The Jurchen conquest of north China in 1127 and the gradual advance of the Mongols in the early thirteenth century blocked overland routes to Central Asia, bringing about a further pivot of the Song toward maritime trade. Despite the Jin conquest of north China, the Song maintained regular trade relations with them. In a treaty of 1142, both sides agreed to re-establish licensed trading markets. In both states cross-border trade officially was a sort of government monopoly, although much exchange was carried out privately or through clandestine channels. Official regulations, such as permits to enter Song territory or sale certificates on the Jin side, constituted a significant source of income for both governments. Regular trade relations 12

Cited in Hebeisheng wenwu yanjiusuo 河北省文物研究所 (ed.), Xuanhua Liaomu 宣 化遼墓 (Beijing, Wenwu chubanshe, 2001), p. 304.

647

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

were only interrupted during renewed outbreaks of war, such as in 1161–1165 and 1206–1208. To summarize, Song China mainly exported manufactured goods and metallic monies, items with high added value, in exchange for a great variety of aromatics, medicinal drugs, incenses, and other exotic luxury items that served the personal and religious demands of the social elite and eventually, due to consumption, disappeared from the internal circulation sphere. Probably with the exception of horses needed for military purposes, the state was not dependent on any of the imported items. We have also noted the gradual trend in imports from luxuries to bulk commodities, although we should definitely not overstate their importance, as the shift to large quantities of bulk commodities only occurred during the subsequent Yuan dynasty. The development of large-scale ceramics industries is one example of this trend.

China as Part of the Mongol Empire Commanding an empire encompassing multiple world regions, the Mongol rulers had a strong interest in strengthening their role throughout Eurasia and maintained a positive attitude toward foreign trade. The Mongol rulers consequently revived and extended the overland trading networks throughout Central Asia. Cities like Samarkand, Karakorum, and Bukhara, located along the traditional overland Silk Roads, became major entrepôts in transEurasian trade. After the conquest of China and the founding of the Yuan dynasty (1271–1368), the Mongols encouraged both overland and maritime trade, especially with the Islamic and Turkic worlds. Not only textiles but also large numbers of West Asian weavers and other artisans and craftsmen, as well as physicians, were sent to China during the Mongol era. The Mongols’ thirst for luxury goods, such as the best pearls, has been clearly demonstrated by Thomas Allsen.13 The Mongols were intent on building a trade empire, and one can go so far as to claim that commerce was used as a weapon for the subjugation of peoples and polities. In this connection, the Mongols also wanted to profit from private trade. This becomes evident from Mongol support of private merchants, and the close relationship between official purposes and private trade.

13

Thomas T. Allsen, The Steppe and the Sea: Pearls in the Mongol Empire (Philadelphia, University of Pennsylvania Press, 2019).

648

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Military conflicts among the feuding Mongol khanates in West and Central Asia prompted a yet further reorientation of China’s foreign trade toward maritime routes. Written and archaeological evidence attests to flourishing overseas trade relations. Khubilai Khan (r. 1260–1294) explicitly stated that “most of the foreign countries overseas produce rare and precious goods, which should be treated with more attention by China.”14 The Yuan government also re-established maritime trade offices, designated Quanzhou as the primary port for the south seas trade, and sent numerous diplomatic missions abroad as far afield as India inviting tribute and trade. Overseas expeditions to Vietnam, Champa, Java, and Japan were intended to extend Mongol authority abroad and to impose control over important trading hubs throughout the Indian Ocean. Troops were supposed to implant Mongol hegemony in these regions, following which trade could commence. During the attack of Java in 1294, for example, Khubilai forbade any private merchants to sail to the south seas and the Indian Ocean. At the same time, diplomats and officials accompanied the military expeditions to encourage neighboring countries to pay tribute to the Mongol court in China and to enter into trade relations with them. Even when Mongol military attacks failed, as was the case in Java, diplomatic and trade relations developed in the aftermath. The Mongols provided a special group of merchants, called ortoq (“partnership” in Turkish), comprising mainly Uyghurs and Central Asian Muslims, with a kind of trading monopoly. The ortoq served as agents of foreign trade for the Yuan state and individual Mongol nobles, who provided them with ships and capital (mostly silver) to engage in commerce. For these services the ortoq agents received commissions of 30 percent of the cargo. From 1280 Khubilai also entrusted supervision of the imperial treasury to the ortoq merchants, who used these funds for moneylending, financing overland trade caravans to Western Asia, and maritime commerce. The selection of ortoq for such purposes was actually an attempt to strengthen government control over foreign trade. Muslim ortoq merchants operated side by side with Chinese traders. Merchants from Quanzhou figured prominently in the development of these merchant networks. On the eve of his departure for home in 1294, the Venetian merchant Marco Polo was overwhelmed at the sight of all the ships he observed in the port of Quanzhou (“Zaytón,” in his account), which he described as “one of the greatest havens in the world for commerce.”15 14 15

YS, 210, p. 4664. Marco Polo, The Book of Ser Marco Polo, vol. 2 (ed. Sir Henry Yule and Henri Cordier), 3rd ed. (London, John Murray, 1929), p. 351.

649

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

The Yuan government strictly controlled foreign trade through the Money and Tax Supervision Bureau (quanfusi 泉府司) and its regional branch offices (xing quanfusi 行泉府司) as well as local salt administration offices. The Quanfusi, located in Hangzhou, supervised the maritime trade offices and the privileged ortoq merchants. Customs exactions were mostly collected using the “percentage levy” (choufen 抽分) system that the Mongols, following Song practice, instituted in 1281, whereby the government took a certain percentage of the trade commodities in kind. Marco Polo mentions that the percentage levy amounted to 30 percent for fine goods and about 40 percent for coarse goods.16 On four occasions the Yuan court suspended maritime trade, prohibiting merchants from sailing abroad and abolishing the maritime trade offices, but these bans were short-lived. In comparison to Song times, more bulk commodities were imported during the Yuan era, for example many kinds of cotton fabrics and other textiles from South and Southeast Asia. But luxury items were “necessities” in the Mongols’ political culture and consequently remained essential. The redistribution of luxuries from the imperial treasury, especially garments and valuable textiles, gems, pearls, and precious metals, was part and parcel of their policy of imperial rule. Marco Polo expressed wonder at the great amounts of pepper imported into China. He speaks of forty-three loads daily brought into the city of Kinsay – that is, Hangzhou – with each load equal to 223 libras (around 100 kilograms).17 Somewhat later, Ibn Battuta claimed that Chinese merchants dominated the sea routes from the pepper-growing regions in India to China.18 During early Ming times pepper even came to serve as commodity money and as payment for officials.19 Apart from Quanzhou, Ningbo and Taicang (which connected Suzhou to the Yangzi river) developed into a crossroads of overseas trade where foreign merchants mingled. Records of the maritime trade superintendency at Ningbo enumerate 224 different kinds of imported goods.20 Recently discovered Yuan ceramics, which were clearly designed for export overseas, at Taicang corroborate contemporary descriptions of flourishing trade in this area.

16 18

19

20

Book of Ser Marco Polo, p. 235. 17 Book of Ser Marco Polo, p. 204. Ibn Battuta, Travels in Asia and Africa, 1325–1354 (London, Routledge and Kegan Paul, 1929), p. 235. Ju-kang T’ien, “Cheng Ho’s Voyages and the Distribution of Pepper in China,” Journal of the Royal Asiatic Society of Great Britain and Ireland 2 (1981), 186–97. SMXZ, 5:1a–5a.

650

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

The Mongol rulers entrusted their silver revenues to the ortoq merchants to finance their trade in Western Asia and the Islamic world in general. While during Song times much of the silver the Song paid to both the Khitan Liao and the Tanguts as tribute flowed back into China as a result of trade, the outflow of silver reached a peak under the Yuan dynasty, leaving the early Ming rulers with insufficient silver and copper to maintain these metals as legal tender. The Mongol rulers, it should be mentioned, were the first regime in China to adopt paper money as the exclusive currency of domestic trade. They also tried, if not successfully, to induce their trading partners to accept their paper money as well. Marco Polo observed that this paper money is circulated in every part of the Great Khan’s dominions . . . All of his subjects receive it without hesitation, because, wherever their business may call them, they can dispose of it again in the purchase of merchandise they may require; such as pearls, jewels, gold, or silver.21

To pay for all the incense, aromatics, spices, and scented woods that entered China via maritime trade, great quantities of silver continued to be exported to Southeast Asia, India, and as far west as Arabia. Some of the silver shipped to the Islamic world was exchanged for horses. Merchants from the island of Kish in the Persian Gulf played an initiating role in the overseas horse trade by establishing trading bases across the Indian Ocean. The Mongol-ruled Ilkhan regime in Iran (1258–1335) depended on the networks and the know-how of these Kish merchants. In addition to silver, Chinese silks, cotton, and ceramics were traded for spices and aromatics by Muslim merchants active in the Indian Ocean. Against this background, recent research has revised the formerly prevailing attitude that Mongol rule was not conducive to economic development and foreign commerce. Sino-Islamic contacts especially experienced a significant upswing during the Mongol period. Commercial exchange went hand in hand with the transfer of technologies, for example in the fields of medicine, navigation, and geography. Ceramics production and trade, too, provide ample evidence of intensive mutual exchange. The Mongol rulers of China facilitated the importation from Iran of the cobalt blue (known in China as bosilan 波斯藍, or “Persian blue”) used to decorate China’s famous blue-and-white porcelains. At the same time, Chinese motifs such as the dragon and the phoenix were imitated by Iranian potters. 21

Book of Ser Marco Polo, pp. 160–1.

651

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

Ming Dynasty Official (Tribute) Trade versus Private (“Illegal”) Trade: Tension or Complement? With the establishment of Ming rule, centuries of active promotion of foreign trade came to an end. But that did not mean that trade relations no longer existed, nor that China was suddenly isolated. The Ming initiated a tribute system that linked all overseas trade to diplomatic relationships which subordinated tributary subjects to the nominal suzerainty of the Ming emperor. The first Ming ruler, Emperor Hongwu (r. 1368–1398), was convinced that “if he did not prohibit private trade, the people could be lured by profit from trade and eventually be punished by law.”22 Against this background he permitted foreign trade only as part of the official tribute system, while imposing a ban on private maritime trade, a political measure that had far-reaching consequences for the entire maritime region. For nearly 200 years, from the enactment of the maritime trade prohibition in 1371 until its abolition in 1567, Ming subjects were prohibited from going to sea or contacting foreigners, or they faced capital punishment. Facilities for oceangoing voyages were deliberately destroyed. Harbors at trading ports were blocked with stone and wooden stakes. Over 100,000 coastal inhabitants were ordered to abandon their homes and withdraw to the interior to prevent them from sailing abroad. In the words of a later Ming statesman, Hongwu’s policy would ensure that “not a single plank of wood was allowed to float in the sea.”23 The maritime ban did not, however, mean that no trade took place. Trade, including exchange among private merchants, continued as part of the official tribute system. The logic behind this prohibition policy is difficult to understand at first glance. Maritime trade, encouraged by authorities in China for centuries, suddenly came to a halt. Seagoing merchants were now deemed “pirates” who violated the laws by continuing to trade, or – in the eyes of the authorities – to smuggle. The effort to fight against these “pirates” also strained the state’s fiscal resources. Similarly, it is hard to grasp why the Ming government deliberately sealed up a channel that had enabled its predecessors to reap considerable revenues. The response to this riddle has to be sought in the Confucian political ideology of the first Ming emperor. Hongwu was committed to remedying social conflict and economic inequality by restoring the self-sufficient agrarian economy idealized in the Confucian classics. Farmers were expected to provide for their own 22

MSL, Taizong shilu 太宗實錄, 70.8b.

23

MS, 205, p. 5403.

652

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

necessities, from food to utensils, without resort to the market, while the state would obtain the goods it required by recruiting artisans to perform labor service at state-run workshops. Foreign trade for profit, for purely commercial purposes, in other words, was incompatible with this political ideology. Trade across the northern frontiers with Central Asia also was curtailed. China’s northern neighbors, the Mongols, persistently demanded the reopening of border trading stations and continuation of the trade in horses and livestock for various Chinese manufactured goods and raw materials. But numerous officials and eunuchs who were responsible for the administration of foreign trade – many of whom greatly profited from their monopoly of managing border trade relations – argued fiercely against more liberal commercial relations. Their reluctance to liberalize foreign trade eventually led to a costly and inefficient, even destructive, defense policy along China’s continental and maritime borders: the reconstruction of the Great Wall to ward off the nomadic peoples in the north and an expensive coastal defense system to interdict clandestine trade along the seacoasts. The “response to trade at sea,” Frederick Mote noted, “exactly paralleled the response to Mongol demands for overland trade on the northern borders. In both cases the unrealistic Chinese response failed to control the system.”24 The Yongle Emperor (r. 1403–1424) adopted a markedly different attitude toward foreign trade. For him, trade was a tool that should be strictly controlled by the government in pursuit of its political goals. Consequently, while Yongle continued the ban on private maritime trade, because of his deep concern about the persistent military threat from the Mongol confederations he established government-supervised markets along the northern frontier. Yongle concluded that it is not because we need their horses and cattle that we must trade with [the steppe nomads], but rather because everything they need comes from China. Therefore, a prohibition on trade [like the private maritime trade ban] will incur their enmity, while allowing them to trade will demonstrate our solicitude and will help win their obedience.25

Tribute trade under Yongle basically served to mitigate the Mongol threat on the Ming’s northern frontier.

24

25

Frederick Mote, Imperial China, 900–1800 (Cambridge, MA, Harvard University Press, 2000), p. 720. XWXTK, 26:31a.

653

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

At the same time Yongle also dispatched envoys to foreign countries to elicit tribute embassies from various Central Asian rulers, including the Jurchen of Manchuria, the Tibetans, and Chosŏn Korea. Special mention should be made of the trade between Ming China and the Timurid (1370– 1507) rulers in Central Asia. Temür or Tamerlane (1336–1405), who sought to re-establish a Mongol empire at his capital of Samarkand, originally maintained good political and economic relations with Ming China. After Hongwu’s death in 1398, however, Temür made plans to invade China before he himself passed away in early 1405. Temür’s successor, Sultan Shahrukh (r. 1409–1447), and Yongle both were eager to resume peaceful relations and trade. At least eleven embassies from Samarkand to China are recorded during Temür’s reign, and 111 missions altogether from Samarkand in Ming sources. The Timurids were especially interested in Chinese manufactured goods, such as silks and porcelain. Tribute from the Timurid regimes in Central Asia included horses, camels, jade, pearls, coral, glasswares, knives, and swords. Strict restrictions and rules were adopted regarding the reception of foreign missions and permissions to trade. The principal agency for receiving tributary envoys was the Interpreters’ Institute (huitongguan 會同館), an institution originally established by the Mongols in 1276 that provided lodgings for foreign embassies. With the exception of envoys and merchants from Korea and Ryukyu, who were not subjected to any restrictions, tribute missions and merchants from other foreign tributary countries were permitted to conduct private trade at the Interpreters’ Institute for a brief period of three to five days, a practice that was later continued in Qing times. Although the Ming retained maritime trade offices (shibosi), in contrast to the Song and Yuan their duties were largely confined to hosting foreign embassies, in addition to supervising trade conducted under the auspices of the tributary system. In 1403 Yongle ordered palace eunuchs to assume supervision of the maritime trade offices. The emperor obviously took this measure to increase his personal control over a sensitive sphere of imperial government. As servants in the imperial household, eunuchs became valuable minions of the Ming emperors, who distrusted the scholar-official elite that dominated the civil bureaucracy. In Fuzhou (Fujian), the Ming established a Reception Bureau (rouyuanyi 柔 逺驛, lit. “relay station for being gracious to those afar”) which handled the visits of Ryukyu envoys, collected their tribute goods, and issued gifts for presentation at the Ryukyu court. From China, Ryukyu imported a variety of products, including re-exports from Southeast Asia: raw cotton and cotton 654

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Table 17.2 Tribute exchange in the early Ming Year

Country

Import

Export

Korea 9,800 horses Cambodia 60,000 jin of incense (35.8 tons) Siam 10,000 jin of pepper (6 tons) 100,000 of sappanwood (60 tons) 1403–1570 Mongolia 4,000 horses annually 100,000 hides annually 1406–1407 Japan 1411 Melaka 1417 Sulu 1393 1394

19,000 rolls of silk cloth

21.5 million copper coins 2.6 million copper coins 3 million copper coins

Source: Gang Deng, “The Foreign Staple Trade of China in the Pre-modern Era,” International History Review 19.2 (1997), Table 2, 260

yarn, silks, and medicinal products and drugs such as rhubarb, as well as luxury goods such as tortoiseshells or ginseng. Ryukyu’s major exports for both tribute and trade were sulfur, horses, local textiles, and maritime products such as seaweed and fish. Korean tribute items included goods such as gold, silver, woven mats, leopard and sea otter skins, silk, cloth, white paper, brushes, and ginseng. In addition, the Koreans exchanged products such as horses, ginseng, and paper for silks, copper, tung oil (used for caulking), and a great variety of handicraft goods (see Table 17.2). Whereas Korea and the Ryukyu kingdoms were allowed to send tribute missions once or twice every two years, Annam and Siam were allowed to do so only once every three years, and in the case of Japan only once every ten years. Nonetheless, as Wang Gungwu has argued, at least for various Southeast Asian polities such as Siam inclusion in the Ming “tributary system was useful because, although slightly restrictive, it provided Siam with close trading contacts with the Southern Chinese coasts.”26 Private merchants involved in cross-border trade along the northern frontier included Chinese and Koreans as well as smugglers, so that it constituted a kind of partly official, partly private, and partly illegal exchange. Commodities traded included textiles, handicrafts, furs, and ginseng (the last two items originated primarily from Manchuria), as well as some prohibited military provisions, such as saltpeter, sulfur, and weapons. Korean court 26

Gungwu Wang, “China and Southeast Asia, 1402,” in Jerome Ch’en and Nicholas Tarling (eds.), Studies in the Social History of China and South-East Asia: Essays in Memory of Victor Purcell (Cambridge, Cambridge University Press, 1970), p. 396.

655

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

chronicles from the sixteenth century to the nineteenth make frequent reference to “clandestine merchants” (qianshang 潛商) engaged in unauthorized cross-border trade. Mention of clandestine merchants surged in the seventeenth century, during the Ming–Qing transition period. After the Manchu armies advanced on Ming territories from 1618 onward the Chosŏn court feared becoming embroiled in the conflict and suspended trade with the Chinese. But Korean merchants conducted clandestine trade with the Ming generals who had retreated to Pidao Island, off the Korean coast, and provided them with provisions. It was also Yongle who, beginning in 1405, dispatched the Muslim eunuch Zheng He on naval expeditions into the Indian Ocean and, for the first time in history, attempted to bring large parts of the Indian Ocean within the orbit of Chinese state-controlled trade. The chief aim of the expeditions was to assert – if necessary with force – China’s claim of sovereignty across the Asian maritime region that would oblige foreign countries to pay tribute to the Ming and thus acknowledge China’s political–military and cultural– ideological superiority. These overtures were widely welcomed by local rulers throughout the Indian Ocean world, who were eager to gain trading privileges at the trifling cost of acknowledging the nominal suzerainty of the Ming emperor. Although Zheng He’s expeditions were primarily political in nature, they also served economic goals. Pepper may be mentioned as a case in point to demonstrate that trade mattered, too. Calicut on India’s Malabar Coast, destination of the first three of Zheng He’s expeditions, was a leading commercial center in the Indian Ocean world and a chief exporter of pepper. Reports about Zheng He’s voyages contain detailed descriptions about the availability of pepper. For example, the Overall Survey of the Ocean’s Shores compiled in 1433 by Ma Huan, who accompanied Zheng He on three voyages as a translator of Arabic, states that “Semudera . . . had much pepper, camphor, and gold,” “Quilon . . . produced . . . pepper,” and “Calicut . . . had a great quantity of pepper.”27 After the Zheng He voyages, both foreign and private Chinese merchants as well as tributary embassies brought cargoes of pepper to China. China had long been an important market for pepper, but pepper imports surged in the early fifteenth century, to an estimated one million kilograms on average annually.28 Price differences 27

28

Ma Huan, Ying-yai Sheng-lan: The Overall Survey of the Ocean’s Shores (trans. J.V.G. Mills) (Cambridge, Cambridge University Press, 1970), pp. 64–5. T’ien, “Cheng Ho’s Voyages,” p. 187.

656

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

were enormous, and pepper could be sold in China at a profit of 300 percent or more. Mention should also be made of the diffusion of military technology that followed in the wake of Yongle’s overseas overtures. Warships were equipped with various gunpowder weapons, including firearms and probably cannon.29 In the early fifteenth century, illicit trade in weapon-making materials between Vietnam and the southern Chinese provinces flourished, as copper and other materials used for weapons production were exported to Vietnam (Đại Việt). At the same time Ming forces also acquired superior weapons from Vietnam, such as a “magic fire-lance arrow” that, due to its heavy ironwood wad fixed behind the arrow in order to increase pressure within the barrel, could be shot as far as 300 paces. Vietnamese craftsmen captured in battle during Yongle’s campaigns were also employed in manufacturing cannon, fire lances, and gunpowder.30 China also exported or reexported gunpowder, saltpeter, sulfur, and copper to various neighboring countries, including the Spanish Philippines after the founding of Manila in 1571. Although Zheng He’s voyages projected Ming power into the Indian Ocean world, following Yongle’s death in 1424 this more open and also partly aggressive foreign policy came again to an end. Although Zheng He undertook one more voyage to the Indian Ocean in 1431–1432, he died during the return voyage. The Ming retreat from the sea was upended in the sixteenth century, however, when the revival of domestic commercial growth and increasingly lucrative opportunities for foreign trade incited growing defiance of the Ming ban on overseas trade.

The Role of Smuggling and “Piracy” Smugglers and so-called “pirates” were active throughout the centuries of China’s foreign trade. But probably contraband trade and piracy never played such an important role as during the approximately 200 years of the Ming maritime trade proscription. By government decree, thousands of seagoing merchants whose families had relied on trade for centuries were arbitrarily 29 30

DMHD, 123:3b; XCSL, p. 30. Sun Laichen, “Chinese Military Technology and Dai Viet: c. 1390–1497,” Asia Research Institute Working Paper Series 11 (2003), 14–15; Sun Laichen, “Chinese Gunpowder Technology and Đại Việt, ca. 1390–1497,” in Nhung Tuyet Tran and Anthony Reid (eds.), Viet Nam: Borderless Histories (Madison, University of Wisconsin Press, 2006), p. 95.

657

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

labeled “illegal” elements conducting contraband trade as “pirates” – a term that consequently has to be treated with caution in the Chinese and East Asian context. While private maritime trade was officially prohibited, in reality, according to a sixteenth-century observer, “thousands of people embarked and hundreds of pirate ships were sailing away fully loaded with people and cargo.”31 The more strictly the prohibition was applied, the more violations of it occurred, and the more destructive the piracy became. The strict Ming regulations that permitted trade only as a part of tribute missions also led to yet another phenomenon – imposter missions and envoys (weishi 偽使). In an attempt to circumvent the maritime trade proscription, many merchants sought to disguise themselves as foreign embassies bringing tribute to China. The number of imposter envoys proliferated in the sixteenth century, when commercial expansion in China’s domestic economy whetted Chinese appetite for silver imports. Whereas China had been a great exporter of monetary metals – especially copper coins but also silver – during Song and Yuan times, the situation changed dramatically in the late Ming period, when China started to import large quantities of money metals, both silver and copper. The Ming state ceased minting copper coin for nearly a century between the 1430s and 1520s, and domestic silver mines were largely played out by 1450 as well. The attempt to reintroduce an empire-wide paper currency system had proved a complete failure, leaving both private markets and the state’s treasury bereft of viable monetary media. It was, thus, a fortuitious coincidence for China that this monetary crisis coincided with the discovery of silver mines in Japan and with the arrival of European traders and their insatiable demand for Chinese products, for which they paid primarily in silver. This was a turning point in China’s economic relationship to the outside world: whereas until early Ming times ever greater quantities of monetary metals (both coins and silver) leaked out of China, causing a deficiency in the state coffers, from the sixteenth century onwards increasingly large quantities of silver started to flow into China. Although the Ming court suspended all official trade with Japan in 1523 after conflicts between representatives of two rival Japanese daimyo¯ clans sent to Ningbo as tribute missions erupted into violence that killed some Chinese, the discovery of rich silver mines in Iwami in western Japan around the same time triggered rampant smuggling.

31

QTWZ, 1:7b.

658

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

In Chinese sources, these smugglers were, as a rule, simply designated “Japanese pirates” (wokou 倭寇). Although we know that pirate raids in China and Korea in the thirteenth and fourteenth centuries were in fact mainly carried out by inhabitants of Japan or the islands lying between Japan and Korea, during Ming times most of the Wokou were actually Chinese. The term “Wokou” referred to Japan as the geographical base of these “merchantpirates,” not to their ethnicity. In fact, the Wokou were mostly multinational bands, later including even European traders, such as some Portuguese. When the first Europeans, the Portuguese, appeared in East Asian waters in the early sixteenth century while the Ming maritime proscription was still in effect, they joined these Asian smugglers to gain access to highly desired Chinese commodities, especially porcelains and silks. Moreover, many of the so-called “pirate raids” along the Chinese coast were in reality more or less desperate undertakings to force local authorities and the Chinese court to reopen China’s ports for private trade. Tang Shu, in a letter of c. 1557 to the military commander charged with suppressing Wokou activities, observed that the Ming trade proscription had turned merchants into pirates: “pirates and traders are the same people; when trade flourishes, pirates become traders, and when trade is banned, traders become pirates.”32 Wang Zhi, the so-called “king of the Wokou,” himself avowed to a Ming envoy that “piracy will disappear once the trade prohibition is lifted.”33 Nonetheless, the court adamantly opposed proposals to rescind the ban, and instead adopted aggressive tactics to eradicate smuggling by force. In 1557 Wang Zhi was lured to Ningbo from Japan by Ming officials under the false promise of receiving trading privileges for laying down his arms, and then arrested and executed. But his statement was vindicated by what happened after 1567, when the Ming government, weary of the mounting costs of suppressing the Wokou bands and under pressure from local elites in coastal regions who were eager to share in the profits of foreign trade, repealed the maritime ban policy: “the Wokou have disappeared and cease to cause trouble as people are enjoying doing business.”34

The Import of Silver and Foreign Trade in the Late Ming The influx of silver provided the liquidity necessary to facilitate the burgeoning boom in domestic commerce. In addition to the use of silver as the principal medium of market exchange, the Ming state also adopted silver as 32

33

Tang Shu 唐樞, “Fu Hu Hailin lun chu Wang Zhi” 復胡海林論處王直, in HMJSWB, 270:4a. CHTB, 9:62a. 34 GQ, 12:80b.

659

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

the basis of its fiscal system (see the chapter by Lamouroux and von Glahn in this volume). By the late sixteenth century, the Chinese government had converted most land taxes, labor services, and other miscellaneous levies into a single silver tax payment in what came to be known as the “Single-Whip” tax reforms. Thus both the market economy and the fiscal system became dependent on a steady infusion of foreign silver. Following the repeal of the maritime ban in 1567, the Ming government designated Yuegang in Fujian as the home port for Chinese merchants who received licenses to embark on overseas trading expeditions. Chinese seafarers were still forbidden to venture to Japan (and vice versa), compelling Chinese and Japanese merchants to conduct trade with each other at neutral intermediary ports such as Manila or Hoi An in central Vietnam. Japanese overseas trade also took off dramatically in the early seventeenth century after the Tokugawa shogun Ieyasu, having established a unified national government in Japan, began to issue “vermillion seal” (shu¯in 朱印) licenses to favored merchants and daimyo¯ lords. Exports of Japanese silver to China rose steadily down to the late 1630s, when the Tokugawa shogunate reversed course and sharply restricted overseas contacts, including abolishing the vermillion seal licenses and forbidding subjects from venturing abroad. Initially, Japanese mines were the source of silver imports to China, but in the second half of the sixteenth century silver extracted from Spanish colonial mines in the Americas began to flow to China as well. After seizing Melaka on the Malay Peninsula – the linchpin of trade routes connecting the Indian Ocean with insular Southeast Asia – in 1511, the Portuguese began to venture into East Asian waters. In an effort to sever the Portuguese from the Wokou bands, in 1557 the Ming court granted the Portuguese permission to establish a trading colony at Macau. Portuguese traders were thus able to import Japanese silver into China as well as bringing silver from Europe to purchase Chinese goods. The founding of the Spanish colony at Manila in 1571 opened up another, more direct, channel for imports of silver from the mines of Spanish America. Through the Spanish transpacific trade China became linked to Mexico, Peru, and their hinterlands, a significant further step in the historical development of a global economy spanning the Atlantic, Indian, and Pacific Oceans (Map 17.2). Although the Spanish court intended that Manila serve as the beachhead for gaining access to Chinese goods, the drain of silver from its New World colonies to China soon prompted second thoughts. In 1593 Spain imposed, on the value of the cargoes of what became known as the “Manila galleons,” limits that were intended to curb the export of silver across the Pacific. The Spanish galleons brought mainly silver ingots, coins, and some American 660

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Map 17.2 Maritime East Asia, c. 1620 Source: Richard von Glahn, “The Maritime Trading World of East Asia from the Thirteenth to the Seventeenth Centuries,” in Tamara H. Bentley (ed.), Picturing Commerce in and from the East Asian Maritime Circuits, 1550 to 1800 (Amsterdam, Amsterdam University Press, 2019), Figure 2.1, p. 64. Reproduced by permission of T. Bentley

products such as wines, sweet potatoes, chocolate, cacao, and tobacco from Acapulco to Manila, whence the silver was shipped to China (especially Fujian), in exchange for Chinese goods, above all silks and porcelains, but also a great variety of other items ranging from handicraft goods such as ivory figurines and writing desks to agricultural products such as fruits and flour. Spanish galleons sailing from Manila to Acapulco in the 1630s on average carried between 300 and 500 boxes (approximately thirty to fifty-

661

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

two metric tons) of Chinese silk; by the early eighteenth century the galleon trade brought from 4,000 to as many as 12,000 bales (239 to 716 tons) of raw silk to supply Mexico’s silk-weaving industry.35 The number of Chinese ship arrivals registered at the port of Manila rose from six in 1574 to fifty-four in 1637. In 1659, the Casa de Contratación in Seville announced that Peru alone annually shipped a quantity of 500,000 pesos of silver via Acapulco to China in exchange for Chinese and Asian goods. The foreign trade boom that followed the lifting of the maritime trade ban in 1567 induced ever-increasing quantities of silver to flow into China. The total amount of silver imported into China between 1550 and 1600 has been estimated at approximately 2,150 to 2,350 metric tons (mostly from Japan); in the first half of the seventeenth century, according to conservative estimates, a total of 5,000 tons of silver was imported to China between 1601 and 1645, roughly half from Japan and half from the New World.36 An estimated 33 to 40 percent of the silver output of Mexico and Peru, which combined accounted for approximately 85 percent of global production between 1500 and 1800, ultimately flowed to China. In the seventeenth century, the maritime powers of Europe engaged in violent struggle for access to the Chinese market. After gaining independence from Spanish rule, the Dutch Republic chartered its seafarers, gathered under the umbrella of the Dutch East Indies Company (Vereenigde Oostindische Compagnie, or VOC), with a monopoly on trade in Asian seas. Initially the Dutch mostly preyed on Portuguese and Spanish ships in East Asian waters, but in 1609 Tokugawa Ieyasu granted the Dutch “vermillion seal” licenses to trade at the Japanese port of Hirado (later extended to Nagasaki as well). In 1619 the VOC established Fort Batavia on the island of Java as the headquarters for its military and trading operations in Asia. Seeking to circumvent the Portuguese Nagasaki–Macau trade route and create their own access to the Chinese market, the VOC founded Fort Zeelandia on the southern coast of Taiwan in 1624. But the Dutch also had to compete with Chinese rivals, most notably the Zheng clan of merchants headed by Zheng Zhilong. In 1628 Zheng obtained formal recognition from the Ming court to defend the coastal region of Fujian against pirate bands, 35

36

Gang Deng, “The Foreign Staple Trade of China in the Pre-modern Era,” International History Review 19.2 (1997), 275. Gang Deng, Chinese Maritime Activities and Socio-economic Consequences, c. 2100 B . C . –1900 A . D. (New York, Greenwood Press, 1997), p. 111; Richard von Glahn, Fountain of Fortune: Money and Monetary Policy in China, 1000–1700 (Berkeley, University of California Press, 1996), p. 140.

662

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

and over the next fifteen years he built up a formidable navy as well as farflung commercial operations (including violating the Ming prohibition against sailing directly to Japan). After the Manchu armies arrived in Fujian in 1646, Zheng Zhilong sought to negotiate a similar arrangement with the fledgling Qing government, but he was seized and imprisoned. His son Zheng Chenggong pledged his fleet in service to the Ming resistance, and ultimately succeeded in controlling the coastal regions of Fujian in defiance of the Qing. Under Zheng Chenggong’s leadership, the Zheng clan created a trading empire that encompassed Japan (where the Tokugawa had restricted trade to Chinese and Dutch merchants at the single port of Nagasaki), Vietnam, the Philippines, Cambodia, and Siam. The Zheng fleet voyaging to Japan alone comprised fifty ships a year. The Zheng family’s annual profit from Japanese trade has been estimated at 550,000 liang, and their total profit from overseas trade at 2.3 million to 2.7 million liang a year.37 In the late 1650s, Zheng Chenggong launched attacks on cities in the Yangzi delta with perhaps as many as a thousand ships and about 130,000 soldiers before he was finally repulsed by Qing forces in 1659. Fierce competition between the Zheng regime and the Dutch for control over intra-Asian trade led to the Zheng invasion and expulsion of the Dutch from Taiwan in 1661. Subsequently the Zheng moved their base of operations to Taiwan. At the same time, in an effort to blockade Zheng commercial activities the Qing government reintroduced a ban on overseas trade and ordered coastal communities to withdraw thirty li (seventeen kilometers) inland. The steep upswing of maritime trade after the lifting of the trade ban in 1567 and the Qing invasion of China stimulated emigration from the coastal regions of south China to trading outposts throughout maritime East Asia. Chinese settlers often formed their own enclaves (or “Chinatowns”) and as a rule maintained close commercial relations with their home districts in China. Chinese migration to Manila and other European colonial cities in Asia fostered informal trading networks that played a major role in cultural as well as economic exchange between Chinese and Europeans. The maritime ban enacted by the Qing in 1662 disrupted movements of Chinese abroad, but after the ban was repealed in 1683 overseas migration, especially to Southeast Asia, swelled.

37

Deng, “The Foreign Staple Trade,” 255.

663

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

The Revival of Foreign Trade in the Early and High Qing During the Qing dynasty, major changes in the organization and structure of foreign trade were introduced. In various respects the Manchu emperors were more pragmatic in addressing economic and diplomatic problems than their Ming predecessors. The early Qing court, especially under the Kangxi Emperor (r. 1662–1722), generally possessed a more positive attitude toward and active interest in certain forms of foreign trade. The maritime trade offices that had managed overseas commerce since early Song times were discontinued under the Manchus. At the outset, the Qing government was engaged in protracted efforts to consolidate its rule against both domestic insurrection (the Rebellion of the Three Feudatories, 1673–1681) and the Zheng merchant-princes in Taiwan, who nominally swore allegiance to the defunct Ming dynasty. In 1683, after defeating the Zheng and occupying Taiwan, Kangxi repealed the maritime embargo enacted to isolate the Zheng and reopened Chinese ports to foreign trade. Kangxi was also convinced that trade with foreigners would benefit China. The complete reorganization and restructuring of the institutions and personnel responsible for the administration of maritime trade in the Kangxi era attests to the changed attitude toward foreign trade. Following the repeal of the maritime ban, the Qing established customs houses in the four most important coastal regions of Guangdong, Fujian, Zhejiang, and Jiangsu provinces. Numerous subordinate customs stations were added to the head offices in each of these provinces, which in Fujian alone totalled thirty-three in 1728. At the same time, new regulations were set up for the treatment of official tribute and private trade. Goods from up to three ships accompanying foreign tribute missions were exempt from customs duties. Private traders had to pay customs duties at a uniform rate of 20 percent. In 1685, foreign traders received permission to trade in Chinese ports, and the authority for levying taxes was transferred from the Ministry of Rites (libu 禮部) – which supervised tribute embassies – to the Ministry of Revenue (hubu 戶部). In 1686, the governor of Guangdong, Li Shizhen, authorized the issue of licenses for overseas trading ventures to “foreign goods firms” (yanghuo hang 洋貨行). Annual license fees of fifty to 1,000 liang were levied for these licenses. Data from 1752–1754 indicate that the average import levy for overseas trading vessels ranged between 470 and 610 liang.38 38

Deng, Chinese Maritime Activities, pp. 111–12.

664

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Furthermore, in contrast to earlier periods, Chinese merchants dominated trade with mainland and insular Southeast Asia during the Qing and played a major role in the establishment of intra-Asian trading networks. Trade with Southeast Asia boomed after the reopening of maritime trade in 1683. A dense coastal merchant network emerged, notably at the port of Xiamen (Amoy), which had eclipsed Quanzhou as the main entrepôt of Southeast Asian trade. Emerging first in the late fifteenth century as a base for smugglers – particularly those engaged in contraband trade with the Portuguese and Japanese – Xiamen was officially designated as a seaport with an administrative seat when the maritime trade proscription was lifted in 1567. A new feature of these trade relations was the increasing importation of bulk products such as rice from Siam, the Mekong delta, and Luzon. Close commercial relations also instigated Chinese emigration to many parts of Southeast Asia. In Qing times we also observe an increasing importance of overland commercial routes from Yunnan in southwest China to polities and cities in Thailand, Cambodia, and Bengal, regions with direct access to the Indian Ocean. Nevertheless, due to security calculations and concerns about domestic stability, in 1716 Kangxi imposed a second maritime prohibition primarily directed against Chinese ships sailing to Southeast Asia, where renegade Ming loyalists were believed to have taken refuge. But trade with Japan, Ryukyu, and Annam – which provided China with great quantities of rice – as well as with Europeans continued. Trade with Southeast Asia was disrupted – whereas eight Chinese vessels had made port at Batavia in 1716, none arrived in 1723 – but commercial interests and the desire to raise revenues eventually won out. Kangxi’s successor, Yongzheng (r. 1722–1735), rescinded the ban on Southeast Asian trade for Fujian in 1728 and for the remaining provinces in 1729. Subsequently trade with Southeast Asia – accompanied by mass emigration of Chinese to the region – reached unprecedented heights. From their Ming predecessors the Manchu rulers inherited the institutions of the Interpreters’ Institute (huitongguan) and the Translators’ Institute (siyiguan 四譯館), which the Qianlong Emperor (r. 1735–1795) merged into a single institution known as the Huitong siyiguan in 1748. The main function of this office lay in the reception and lodging of tributary envoys. In addition, a Court of Colonial Affairs (lifanyuan 理藩院) was established to manage affairs with Mongols, Central Asian Muslims, and Russians. With the eastward expansion of the Russian Empire over the course of the seventeenth century, Russia became a new neighbor of China. The Qing concluded the first Chinese treaty with Russia (the Treaty of Nerchinsk) in 1689, followed by the Treaty of Kyakhta in 1727. The Treaty of Nerchinsk 665

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

loosely defined a border between China and Russia and established the principle of diplomatic parity. The Treaty of Kyakhta moved the border lines farther west and specified two places for cross-border trade, Kyakhta in northern Mongolia and Tsuruchaitu on the Manchurian frontier, near Nerchinsk. On the Chinese side, Shanxi merchants dominated the Russia trade. Russian traders mostly sold pelts, hides, and furs in return for Chinese tea, silks, cotton fabrics, and rhubarb. While trade with the Europeans was concentrated in Chinese port cities, the Central Asian and Russian trade should not be underestimated. The Russian government sponsored new markets and customs stations at Siberian outposts as early as the seventeenth century, and the trade in rhubarb, prized for its medicinal value, was particularly lucrative. Tsar Aleksei Mikhailovich (r. 1645–1676) imposed a monopoly on the rhubarb trade in 1652, and in 1657 prohibited the private purchase and sale of rhubarb by Russians of all classes as well as Siberians, Bukharans, and Tatars in order to expand and regularize the import of rhubarb from China and Korea. Chinese rhubarb reached Europe almost exclusively via Russia until the late nineteenth century. Commercial intercourse with various Asian neighbors shows that it is difficult to draw a clear-cut distinction between official and private trade. Both private and official brokers and even “guild houses” (huiguan 會館) were involved in crossborder trade, but this trade was strictly supervised by the Qing goverment. A local gazetteer from the Zhejiang port of Zhapu includes a detailed description of how the inspection and taxation of goods was implemented and which documents merchants had to carry. Government-licensed brokers (yahang 牙 行) prepared the application forms to pass goods through customs (baodan 報單) and to obtain a passport (xianzhao 縣照), which were presented at the subprefectural office for coastal defense (haifang tongzhi shu 海防同知署). The name, age, and appearance, and native place of the ship owner, the helmsmen, and each sailor had to be clearly listed on the passport. After receiving a department warrant (bupai 部牌) and red certificate (hongdan 紅單), the merchant went to the customs office to declare the goods for examination. The red certificate was a declaration of the types and quantities of the commodities the merchant was importing, the port of entry, and the amount paid in customs duties. In the case of exports of domestic goods, merchants paid duties at the customs stations and obtained a sealed document with an official stamp (yongyin 用印) to receive their export permit (lingchu 領出).39 39

ZPBZ, 6:3b–4b.

666

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Although the Qing government generally adopted a laissez-faire attitude toward foreign trade, it imposed controls on a few strategically important commodities, most notably imports of Japanese copper. While, as we have seen above, silver played the major if not essential role in China’s foreign trade in the sixteenth and seventeenth centuries, in the eighteenth century the importation of copper from Japan gained in importance. Japanese copper (in addition to domestic copper mined primarily in Yunnan) was used to cast brass (copper–zinc alloy) coins. The rising demand for currency has to be seen in direct relationship to expanding local markets, which required a reliable means of exchange for petty-commodity circulation, for which high-value silver was inadequate (see the chapter by Kuroda in this volume). Japan – long an importer of Chinese copper coins – now exported raw copper to China. The Qing state’s copper procurement system was a complicated structure that underwent various changes over the course of the dynasty. Generally, we can observe a tendency to assert strict control over the import of Japanese copper while at the same time trying to shift the risks to private merchants. After 1700 the procurement of copper was transferred to a group of nine magnate contractors. But these merchants continuously incurred losses, prompting the government to hand over all responsibility for copper imports to a group of roughly a dozen “quota merchants” (eshang 額商) in 1737 and (from 1744) a single magnate contractor. The quota merchants were granted special licenses to trade in Japan (wozhao 倭照) and assigned specific quotas of copper to be delivered to the ministry of revenue. The customs superintendent selected the quota merchants from wealthy families in the Yangzi delta region who had sufficient capital resources to engage in the overseas copper trade. Official trade with Ryukyu likewise was assigned to a special group of merchants, the so-called “Ryukyu merchants” (qiushang 球商). Their identity is still open to discussion. The term “Ryukyu merchant” originally may have been a designation for all merchants engaged in this trade, but eventually the Qing government designated a group of ten merchants to act as official brokers (yahang) for the tribute missions arriving at Fuzhou from Ryukyu. In contrast to the official brokers who supervised the Ryukyu trade in the Ming, the Qing-era “Ryukyu merchants” maintained relative independence from the government and established their own coastal and domestic networks, engaging smaller merchants to provide them with the required commodities. The Ryukyu delegation would provide the “Ryukyu merchants” with lists of the Chinese goods they sought to buy and prices they 667

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

were willing to pay, and the latter would send agents to Suzhou and Guangzhou to procure the desired items. Qing sources also speak of commissioned merchants known as keshang 客商 (lit. “guest merchants”). In the beginning, at least, both the “Ryukyu merchants” and “guest merchants” seem to have pursued similar business activities and fulfilled common functions. The change in designation from Ming “brokers” (yahang) to Qing “commissioned merchants” (keshang) may be an indication of the development from brokers as middlemen in Sino-Ryukyu tribute trade during Ming times to a more independent status, as private merchants directly trading with the Ryukyuans but still under government supervision in Qing times. One should bear in mind that both official and private merchants, as well as government brokers, were active in foreign trade and that they frequently also co-operated with each other. Trade with Korea was increasingly carried out by private merchants. Korean merchants from Ǔiju, near the mouth of the Yalu river (the boundary dividing Korea from Qing territory) traded with their Chinese counterparts at mid-river islands. Major products involved were Korean ginseng, Chinese silver, and Japanese copper. Trade between China and European merchants slumped in the midseventeenth century during the Ming–Qing dynastic transition and the Qing maritime ban, and remained modest before 1700. But the eighteenth century witnessed significant growth in European demand for Chinese goods such as porcelains and silks, and increasingly for tea. During the late seventeenth and early eighteenth centuries, China sold at least a million pieces of porcelain to the British East India Company alone. After an incident in 1757, European traders were restricted to the single port of Guangzhou (or Canton, as Europeans referred to it). The so-called “Canton system,” which governed European trade with China from 1760 until the outbreak of the Opium War in 1839, confined all maritime trade with Western countries to Guangzhou/ Canton. European merchants – most of whom had been obliged to join “East India companies,” which acted as national cartels for Asian trade – were required to conduct their business with a consortium of twenty-odd Chinese merchant firms (hang, or hong in Cantonese), who became known as the Cohong 公行 merchants. The major purpose of the Cohong was not, as previously thought, to eliminate competition and disadvantage foreign commerce, but on the contrary to preserve competition so that trade with the Europeans could continue and foreigners would not be discouraged from returning to China. The Cohong was founded at the initiative of the merchants Swequa and 668

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

Poankeequa as a reaction to the “triple alliance” formed by the merchants Cai Hunqua, Swetia, and Chetqua, who had become so dominant in the trade with the Europeans that they had gained control of one-third of the export market in Guangzhou. Paul van Dyke has concluded that the Cohong “provided the means whereby [a select group of Chinese] partnerships could gain more control and channel foreign investments to China’s production areas to finance expansion, and all without sacrificing competition and growth.” Prices remained competitive, and increasing numbers of foreigners were attracted. Van Dyke thus considers the Cohong a “market managing group,” not a monopoly.40 But when the triple alliance began to fall apart after Cai Hunqua’s death in 1770, the Cohong had outlived its purpose and was abolished the following year. Contrary to the traditional idea that the Chinese government intended to restrict foreign trade and to this end established all kinds of obstacles for foreign merchants, we know that both the Qing court and local authorities were in fact strongly interested in expanding trade with the Europeans. Although foreign trade was not a leading source of income for the imperial government, customs revenue generated approximately 2 to 3 percent of central government income in the late eighteenth century. The positive attitude of the government toward foreign trade and the efforts by Chinese merchants to meet the increasing European demand for Chinese products resulted in a steep rise of commercial activity at Guangzhou and Macau. But meeting this demand caused much trouble for many Chinese merchants, including the Cohong merchants, as they had to pay high taxes and charges to government authorities. Domestic producers of tea, porcelain, and other products frequently could not meet the market demand. Profits from foreign trade consequently could not be reinvested on a large scale, and many Cohong merchants became dependent on advances and credit from the foreign companies, although this was prohibited, at least officially. The major export commodities of China in the trade with the European East India companies were, above all, tea, ceramics, and silks, but included, of course, a wide variety of other products from lacquers and paintings to exotic fruits. The Europeans paid for these goods with silver until the English East India Company started to import opium and cotton cultivated in India into China in the late eighteenth century. An important structural change that occurred in the mid-eighteenth century was the increasing role of primary 40

Paul A. van Dyke, Merchants of Canton and Macao, vol. 2, Success and Failure in EighteenthCentury Chinese Trade (Hong Kong, Hong Kong University Press, 2016), pp. 4, 6.

669

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

goods as export commodities, above all tea but also raw silk and other agricultural products. Soaring demand for tea in Europe (most notably in Britain, the Netherlands, and Russia) prompted the emergence of new varieties of tea produced and exported mainly from Fujian, Jiangxi, Hubei, Hunan, and Guangdong provinces. On average tea comprised 78 percent of the total value of British imports from China between 1722 and 1833 and 72 percent of Dutch imports between 1729 and 1793.41 At the same time, the importance of Chinese manufactured goods such as porcelain gradually declined as Europeans developed their own porcelain industries and substituted domestic manufactures for imports from China. To be sure, the politico-economic basis for Chinese merchants to carry out their business was and remained substantially and qualitatively different from the environment that dominated the world of the rising capitalist economies in Europe. But the Chinese government did intend to increase commercial activities with the Europeans, and to attract more foreign ships to China, in order to increase imperial revenues. Earnings from exports were in fact invested to expand inland production to meet export demand. But the revenue gained from this trade was basically designated for court and state consumption, and not reinvested to produce commodities that could profitably compete in world markets. Nor did China ever develop a system of public credit – i.e., of national debt – as it first took root in the Netherlands, where merchant capital ruled over production. In contrast to the early modern European mercantilist countries, where public debt became one of the most powerful levers of private accumulation, the Qing government never intended to transform its economy into one focused on the increase and accumulation of capital in the hands of private producers; in other words, it never intended to introduce a capitalist mode of production. Moreover, the Chinese court and government officials remained suspicious of their own merchants. The latter were forbidden to communicate with foreign authorities or to establish direct trade with foreign countries. And the official policy of attracting as many foreign ships as possible even resulted in practices favoring foreigners to the detriment of Chinese merchants. Chinese, for example, paid higher interest rates than foreigners and they were officially prohibited from borrowing money from foreigners. If foreign merchants had a dispute with Chinese merchants, they could appeal to local officials for resolution. Chinese, on the contrary,

41

Deng, Chinese Maritime Activities, p. 119.

670

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

received absolutely no support from the government when foreigners failed to fulfill agreements and did not pay them. Chinese merchants did their best to satisfy foreign demand because they saw it as an opportunity to earn money (silver), but they relied on a type of production that otherwise had nothing to do with capitalism. In this respect, foreign demand did, of course, have a certain impact on China’s economy (in terms of quantity of production and the varieties of goods produced to meet foreign demand), but it did not transform China’s commercial economy into a capitalist one. Rudimentary credit systems had appeared in China, but one has to distinguish between the accumulation of wealth in the form of money (silver) and money advances required to bridge the seasonal and local divergences between purchases and sales on the one hand, and capitalist wealth in the form of credit money or capital with an entitlement for growth on the other. This latter form did not exist in imperial China.

Conclusion China’s increasing commercial integration into the global economy, including the transpacific connection to the Americas, went along with increasing activities of private merchants abroad, especially in the maritime realm. Throughout the middle and the late imperial era, except perhaps for the early Ming, Chinese governments were interested in foreign trade, but – drawing on different premises and methods – they sought to regulate it and place it under some form of government control. Despite their interest in trade and exchange with foreign countries, many Chinese courts remained concerned about the possibility that private wealth could be converted into political power and thus constitute a potential threat to imperial authority. But production for sale on markets and the subsequent accumulation of money assets never came to be the sole criterion of China’s economy. Consequently, private commerce never received government support comparable to that of early modern European states, and private merchants active in foreign trade remained “merchants without empire,” as Wang Gungwu has put it.42 Some general characteristics can be discerned in China’s foreign trade over the longue durée. (1) Foreign trade developed from a means to satisfy the personal interests of the ruling elites into a source to be tapped for state 42

Gungwu Wang, “Merchants without Empires: The Hokkien Sojourning Communities,” in Wang (ed.), Greater China and the Chinese Overseas (Singapore, Marshall Cavendish Times Academic Press, 1991), pp. 79–102.

671

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer

finance. (2) Foreign trade experienced increasing institutionalization and bureaucratization. (3) While the first Ming rulers attempted to treat foreign trade as part of an official tribute system and subordinated it to the Ministry of Rites, in general the ritual dimensions of foreign trade were less important during the time under investigation here. The Song, Yuan, late Ming, and Qing governments primarily focused on the commercial dimensions of foreign trade. This priority can also be seen in the bureaucratic organizational structure of foreign trade. During Song times, for example, the maritime trade offices were subordinate to the central state’s fiscal institutions, such as the regional fiscal commission (zhuanyunsi 轉運司) or the Ministry of Revenue. (4) The major actors in early foreign trade were foreigners. In antiquity and the early middle period Chinese merchants remained relatively passive as far as long-distance journeys to far-away countries were concerned. The initiators of China’s long-distance trade were West Asian peoples, such as Sogdians, Iranians, or Arabs. On the overland routes Chinese merchants were never as active as these foreigners. Chinese mariners generally began to sail to Southeast Asia and the Indian Ocean in their own ships only from the late eleventh century onwards. (5) Foreign trade activities prior to the Qing dynasty primarily focused on exchange between China and other Asian societies, while trade with Europeans only gained in importance from the sixteenth century. (6) A great percentage of pre-Qing foreign trade was carried out on the principle of barter. (7) The increasing integration of China into global trading networks before Ming times greatly fostered commercial growth and enabled private merchants to amass huge fortunes, but at the same time it made China’s “national” economy, calculated in terms of monetary metals in state coffers, in fact poorer, because from the Song dynasty to the early Ming large quantities of money metals and valuable manufactured goods were exchanged for a wide range of consumer goods, such as spices, medicinal drugs, scented woods, and so on that would eventually fall out of the circulation sphere due to consumption. (8) Despite their interest in and promotion of foreign trade, rulers of many dynasties, perhaps except for the Mongols, maintained a fair degree of suspicion toward private commerce. (9) The period studied in this chapter, from the tenth century until the late eighteenth, witnessed an ever-increasing role of private business and commerce and (10) an ever-increasing orientation of foreign trade toward the maritime world. Certainly the long-maintained view that imperial Chinese governments were always obstructing foreign and domestic trade is no longer tenable. The influx of foreign silver beginning in the mid-sixteenth century was particularly crucial for domestic 672

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade

commercial growth. (11) We can also observe a structural change in China’s maritime trade patterns from the mid-eighteenth century onwards: China, which traditionally had been an exporter of manufactured goods, such as silks, ceramics, and metal products, became an exporter of primary products – above all tea but also raw silk – while the importance of manufactures such as porcelain and silk textiles began to fade. From the Song dynasty onward, foreign trade had an ever-growing impact on local societies, especially in the coastal regions, an impact that is also reflected in the expansion of local markets and the rising demand for copper coin as a medium of exchange (see Kuroda’s chapter in this volume). In addition, the consumption of products that originally arrived in China as luxury goods, such as pepper and cotton, became more widely diffused to the medium strata of society. After the lifting of the Ming trade ban, taxation in the form of money (silver), too, became more and more important. A basic change and a steep increase in foreign – especially maritime – trade occurred when Europeans began to purchase huge quantities of Chinese goods in exchange mainly for silver, silver that ended up not only in the purses of private merchants but in the domestic economy in general, including state coffers. In this context, the sixteenth century can be considered a decisive turning point in the history of China’s foreign trade development. Foreign trade became a positive force in the Chinese economy, including state finances, above all because the nature of foreign trade, of exports and imports, changed. Yet China’s foreign trade remained the result and expression of a political economy whose society produced enormous wealth but never instilled or wanted private profit maximization in the form of money as its sole production criterion. Following the Opium Wars, Britain tried to force open the Chinese market for its manufactured goods, such as cotton textiles, but with little success. The reason for this failure has to be sought exactly in China’s particular mode of production, which combined small-scale agriculture with domestic household craftsmanship that produced its own cloth and other basic necessities. Consequently, the most highly advanced, technologically speaking, production system in the world was unable to dislodge textiles produced by hand with mostly primitive looms.

Further Reading Atwell, William S., “International Bullion Flows and the Chinese Economy, circa 1530– 1650,” Past and Present 95.1 (1982), 68–90.

673

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

angela schottenhammer Chaffee, John W., The Muslim Merchants of Premodern China: The History of a Maritime Asian Trade Diaspora, 750–1400 (Cambridge, Cambridge University Press, 2018). Chang, Pin-tsun, “The Rise of Chinese Mercantile Power in Maritime Southeast Asia, ca. 1400–1700,” Crossroads: Studies on the History of Exchange Relations in the East Asian World 6 (2012), 205–30. Deng, Gang, Chinese Maritime Activities and Socio-economic Consequences, c. 2100 B . C . –1900 A . D. (New York, Greenwood Press, 1997), Deng, Gang, “The Foreign Staple Trade of China in the Pre-modern Era,” International History Review 19.2 (1997), 253–85. Giersch, Pat, “Cotton, Copper and Caravans: Trade and the Transformation of Southwest China,” in Eric Tagliacozzo and Wen-chin Chang (eds.), Chinese Circulations: Capital, Commodities, and Networks in Southeast Asia (Durham, NC, Duke University Press 2011), pp. 37–61. Haneda, Masashi, and Mihoko Oka (eds.), A Maritime History of East Asia (Kyoto, Kyoto University Press, 2019). Heng, Derek T.S., Sino-Malay Trade and Diplomacy from the Tenth through the Fourteenth Century (Athens, Center for International Studies, Ohio University, 2009). Li, Kangying, The Ming Maritime Trade Policy in Transition, 1368–1567 (Wiesbaden, Otto Harrassowitz, 2010). Lin Tianwei 林天尉, Songdai xiangyao maoyi shigao 宋代香藥貿易史稿 (Hong Kong, Zhongguo xueshe, 1930). Lo, Jung-pang, China as a Sea Power, 1127–1368: A Preliminary Survey of the Maritime Expansion and Naval Exploits of the Chinese People during the Southern Song and Yuan Periods (Singapore, National University of Singapore Press, 2011). Mancall, Mark, “The Ch’ing Tribute System: An Interpretive Essay,” in John K. Fairbank (ed.), The Chinese World Order: Traditional China’s Foreign Relations (Cambridge, MA, Harvard University Press, 1968), pp. 63–89. Mukai, Masaki, “New Approaches to Pre-modern Maritime Networks,” Asian Review of World Histories 4.2 (2016), 179–89. Ng Chin-keong, Trade and Society: The Amoy Network on the China Coast, 1683–1735 (Singapore, Singapore University Press, 1983). Schottenhammer, Angela, “Brokers and ‘Guild’ (Huiguan 會館) Organizations in China’s Maritime Trade with Her Eastern Neighbours during the Ming and Qing Dynasties,” Crossroads: Studies on the History of Exchange Relations in the East Asian World 1/2 (2010), 99–150. Schottenhammer, Angela, “Characteristics of Qing China’s Maritime Trade Politics, Shunzhi through Early Qianlong Reigns,” in Schottenhammer (ed.), Trading Networks in Early Modern East Asia (Wiesbaden, Otto Harrassowitz, 2010), pp. 101–54. Schottenhammer, Angela (ed.), Early Global Interconnectivity across the Indian Ocean World, vol. 1, Commercial Structures and Exchanges (London, Palgrave Macmillan, 2019). Sen, Tansen, Buddhism, Diplomacy, and Trade: The Realignment of Sino-Indian Relations, 600– 1400 (Honolulu, University of Hawai‘i Press, 2003). Shiba, Yoshinobu, “Sung Foreign Trade: Its Scope and Organization,” in Morris Rossabi (ed.), China among Equals: The Middle Kingdom and Its Neighbors, 10th–14th Centuries (Berkeley, University of California Press 1983), pp. 89–115.

674

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

Foreign Trade So, Billy K.L., Prosperity, Region, and Institutions in Maritime China: The South Fukien Pattern, 946–1368 (Cambridge, MA, Harvard University Area Center, 2000). Van Dyke, Paul A., Merchants of Canton and Macao, vol. 1, Politics and Strategies in Eighteenth-Century Chinese Trade (Hong Kong, Hong Kong University Press, 2011); vol. 2, Success and Failure in Eighteenth-Century Chinese Trade (Hong Kong, Hong Kong University Press, 2016). Vogel, Hans Ulrich, Marco Polo Was in China: New Evidence from Currencies, Salts and Revenues (Leiden, Brill, 2013). von Glahn, Richard, “The Ningbo–Hakata Merchant Network and the Reorientation of East Asian Maritime Trade, 1150–1350,” Harvard Journal of Asiatic Studies 74.2 (2014), 249–79. Wang, Gungwu, The Nanhai Trade: The Early History of Chinese Trade in the South China Sea (Singapore, Marshall Cavendish International, 1998). Wheatley, Paul, “Geographical Notes on Some Commodities Involved in Sung Maritime Trade,” Journal of the Malayan Branch of the Royal Asiatic Society 32.2 (1959), 1–139. Xing, Hang, Conflict and Commerce in Maritime East Asia: The Zheng Family and the Shaping of the Modern World, c. 1620–1720 (Cambridge, Cambridge University Press, 2015). Zhao, Gang, The Qing Opening to the Ocean: Chinese Maritime Policies, 1684–1757 (Honolulu, University of Hawai‘i Press, 2013).

675

https://doi.org/10.1017/9781108587334.019 Published online by Cambridge University Press

18

Production, Consumption, and Living Standards zhiwu chen and kaixiang peng

The Chinese economy demonstrated significant vigor from the eleventh century to the nineteenth. This period of nine centuries achieved remarkable progress in implementing the imperial examination system, improving literacy, establishing private landownership, developing market institutions, adopting new crops and improved farming technology, strengthening the lineage order, and lifting ordinary people’s capacity to deal with risks. According to the optimistic view, this is also the period during which two economic revolutions, the Tang–Song and Ming–Qing transitions, took place, marked by continuous economic growth and improvement in living standards for the population.1 To other scholars, however, this long period of growth was largely achieved through population increase, preventing China from escaping the “Malthusian trap.” That is, while the total size of the Chinese economy may have grown due to the rising population, per capita output failed to rise above historical norms and might have even declined during the long period. This period’s achievement and impact were nothing comparable to those of the Industrial Revolution which started in eighteenth-century Britain. To shed light on this ongoing debate, we first need quantitative measures of economic progress for each historical period. To this end, recent attempts have been made to estimate historical levels of gross domestic product (GDP) in China. Associated with these estimations, however, are theoretical and methodological challenges, requiring cross-validations to ensure accuracy. Besides, ways of living have evolved significantly as new goods and services were invented or imported from overseas, with the result that aggregate GDP 1

Richard von Glahn, The Economic History of China: From Antiquity to the Nineteenth Century (Cambridge, Cambridge University Press, 2016), pp. 295–6.

676

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

estimates do not fully capture the changes in actual living standards. Thus, while important, GDP estimates alone do not necessarily reflect the full picture of the evolution of living standards over the centuries, especially given that any GDP measure is only one of many dimensions that define what is meant by living standards. Micro-level investigations of consumption are thus helpful for studying China’s history of economic activity and well-being. To that end, we will use the Malthusian “subsistence-economy” framework to examine what consumers were consuming and assess the extent to which China was still in, or close to being out of, the Malthusian trap. In applying “subsistence economy” as a reference framework for assessing living standards, economists have conventionally taken it to mean an economy in which producers make just enough to survive on and are normally on the verge of dying from hunger. But the actual situation for a subsistence economy does not have to be that grim. Had there been no risk events such as wars and natural disasters, a subsistence level of income would have worked fine for almost all households and misery would have been infrequent if not rare. However, such risks were always present and frequent in reality. Therefore, focusing on income alone, especially aggregated per capita income, is not enough, and we should pay as much attention to the evolution of risk-mitigation capabilities in order to assess China’s history of living standards. The evolution of income (as measured by GDP, per capita GDP, and/or other metrics) reflects only a partial picture of how living standards have changed. The ability to mitigate risk affects individual welfare as much as income does, which is especially true for an agrarian economy where the average income is not high. In this chapter, we will first review estimates of historical GDP, wages, and consumption, followed by an assessment of how personal security and risk mitigation capacity evolved during the different dynasties. In particular, we will examine four main approaches to mitigating risk: the lineage or kinship network, social networks (religious and otherwise), financial markets, and government welfare. Our goal is to present a more complete picture of longterm changes in both productivity and living standards in historical China.

Estimates of Production Output Attempts to estimate historical GDP for China and other countries were initiated by Angus Maddison in his series of publications.2 According to 2

See Angus Maddison, Chinese Economic Performance in the Long Run, 960–2030 A D, 2nd ed. (Paris, OECD Publishing, 2007).

677

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

Maddison, China’s economy was the largest in the world until the early nineteenth century; hence the recent run-up in Chinese economic activity could be regarded as a restoration of its earlier position. His work was introduced to China about twenty years ago, stimulating much debate among scholars. As Maddison’s estimates of China’s historical GDP are quite rough, it is not surprising that new efforts have been launched to improve estimation techniques. Maddison’s GDP estimates for the twentieth century are more accurate, as there are official contemporary estimates for some time points, e.g., 1933, based on modern concepts and statistical materials that were known at the time, and forward/backward extrapolations can be applied to nearby decades. However, his estimates for pre-nineteenth-century GDP rely heavily on two key assumptions: (1) constant agricultural product per capita from 1368 to 1968 as established by Dwight Perkins’s work (discussed below), and (2) a constant urbanization ratio lasting from the Tang to the early Qing dynasty, derived from Gilbert Rozman.3 These assumptions are oversimplifying, of course, prompting Chinese economic historians to seek to refine production estimates for each period in order to assess China’s total and per capita GDP diachronically. Figure 18.1 shows some of the representative results from these efforts. As shown in Figure 18.1, the per capita GDP series by Broadberry et al. and Xu et al. are 30 to 50 percent higher than that of Maddison for periods before the mid-eighteenth century, while their gaps become smaller for the late nineteenth and early twentieth centuries, when the estimation of Maddison is more difficult to challenge. That means that there had to have been large declines in per capita GDP from around the beginning of the eighteenth century in the estimation series by Broadberry et al. and Xu et al., which is contrary to the assumptions made in Maddison’s qualitative historical studies. Another important difference is that there is significant growth in the Maddison series from the tenth century to the thirteenth (corresponding to the Tang–Song transition), which does not exist in Broadberry et al. The most important source of these differences lies in the population estimates used. China’s population, as shown in Figure 18.2, remained relatively stable at about 60 million to 70 million during the first millennium, before rising to more than 100 million in the Song and starting a new phase of exponential 3

Gilbert Rozman, Urban Networks in Ch’ing China and Tokugawa Japan (Princeton, Princeton University Press, 1973).

678

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards 1200

Maddison (2007) Xu et al. (2016)

Broadberry et al. (2018) Li and Van Zanden (2012)

Per capita GDP (1990$)

1100 Yangzi delta

1000 900 800 700 600 500 400 900

1100

1300

1500

1700

1900

Year

Figure 18.1 Estimates of China’s historical per capita GDP Note: Maddison (2007), Broadberry et al. (2018), and Xu et al. (2016) are estimations for the Chinese empire as a whole, while Li and Van Zanden (2012) is a regional estimation for the Yangzi delta. Sources: Angus Maddison, Chinese Economic Performance in the Long Run, 960–2030 A D, 2nd ed., revised and updated (Paris, OECD Publishing, 2007); Stephen P. Broadberry, Hanhui Guan, and David Daokui Li, “China, Europe, and the Great Divergence: A Study in Historical National Accounting, 980–1850,” Journal of Economic History 78.4 (2018), 955–1000; Yi Xu et al., “Chinese National Income, ca. 1661–1933,” Australian Economic History Review 57.3 (2016), 368–93; Bozhong Li and Jan Luiten van Zanden, “Before the Great Divergence? Comparing the Yangzi Delta and the Netherlands at the Beginning of the Nineteenth Century,” Journal of Economic History 72.4 (2012), 956–89

growth, which is quite unusual in pre-eighteenth-century global population history (see also Cao’s chapter in this volume). On the one hand, this population growth resulted from economic development since the early Song; on the other hand, once the economy developed onto a Malthusian exponential population growth path, it would soon work to cut per capita GDP and limit further population growth. According to the Malthusian logic, population losses due to wars from the fourteenth century to midseventeenth helped offset the decline in GDP (with temporary fluctuations in between), just enough to keep per capita GDP close to the historical norm. However, one cannot accept this Malthusian interpretation without reservation. For example, civil wars and the Manchu attacks on the Ming in the early to mid-seventeenth century not only caused dramatic population losses but 679

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng 140000

Population (10,000)

120000 100000

Modern era

80000 60000

Qing

40000

Song Sui Tang

Han

20000 0

Yuan

Ming

Jin

0

200

400

600

800

1000 1200 1400 1600 1800 2000 Year

Figure 18.2 Population of China, 2 C E – 2000 Source: Cao Shuji 曹树基, Zhongguo renkou shi 中国人口史, vol. 5, Qing shiqi 清时期 (Shanghai, Fudan daxue chubanshe, 2001)

also destroyed many urban and rural areas. Historical documents suggest that China was indeed in eonomic recession and social turmoil during the period, consistent with Maddison’s analysis but contrary to what the estimates by Broadberry et al. and Xu et al. show. Thus we need to examine more evidence before accepting the Malthusian explanation of economic change since the tenth century. Besides the inter-temporal variation of per capita GDP, Li and Van Zanden show significant interregional differences, with the per capita GDP for the Yangzi delta region being 60 to 70 percent higher than the national average in the early nineteenth century. But the gap between leading and lagging regions was wider in earlier periods, due to lower levels of market integration at earlier times (discussed below). Large regional differences mean that how development spreads across regions must be important for understanding the temporal changes in per capita GDP, as the dynamics of regional convergence complicate the relation between population growth and aggregate productivity. Although commercial and handicraft industries grew during the Song and thereafter, the Chinese economy was largely dominated by agriculture (mainly grain farming). Therefore farming productivity was decisive for the relation between per capita GDP and population. Agricultural production, such as food output, is key to our analysis. Fortunately, there are relatively 680

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

more historical documents on agricultural productivity available to researchers today, because, as an agrarian society, China’s imperial bureaucracy always made particular efforts to track and document agriculturerelated output and other information. Next, we turn to reviewing such evidence against the GDP estimates discussed above.

Per Capita Food Output Apart from the population factor, determinants of per capita grain output include yield per mu (a unit of land area of approximately 667 square meters) and total arable land. Given that China’s traditional statistical work focused on tax collection data and lacked systematic national accounting methods, it is impossible to find historical documents that provide direct figures such as aggregate arable land and GDP for each year or decade of every past dynasty. Thus scholars have to rely on indirect information, such as tax revenues, to infer arable land and other aggregates. Yield per mu is somewhat easier to estimate as one can use both macro and micro historical data to calculate and cross-check, and one can also track changes in grain yield. Although traditional methods for improving grain yield are not comparable to today’s biotech and chemical approaches in terms of effectiveness, farmers were still able to raise grain yield by increasing the number of crops per year, adding fertilizers, diversifying across grain types, changing farming devices and processes, and so on (see Pomeranz’s chapter in this volume). That is how grain yield improved from the Song to the Qing dynasty. Figure 18.3 collects several studies that demonstrate a clear long-term upward trend in crop yield, both in the Jiangnan region and nationally, until the turn of the nineteenth century, although all the studies except for Perkins conclude that, after about 1800, Chinese crop yields experienced a sustained downturn. It is no surprise that the estimate by Perkins for the earlier period is lower than the others, since Perkins focuses on the food yield needed to cope with population growth while the others derive their estimates from the production side. In fact, crop yield per mu reached its peak around the early nineteenth century (above 300 jin per mu), a level that was not exceeded until the 1970s. The continuous rise in crop yield during the post-Song years nevertheless worked to lower the pressure from population growth. But it is also true that the per capita grain output could not grow as fast as crop yield. Figure 18.4 illustrates the evolution of per capita processed grain per year based on 681

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng Perkins (1969) Wu (1985, 1998) Zhao et al. (1995) & Zhao (2001) Guo (2001) Li (2003, 2010) Shi (2015)

Yield (Raw grain, jin per mu)

800

Yangzi delta

400

200

100 1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

2000

Year

Figure 18.3 China’s historical crop yield, 1100–1950 Notes: (1) Crop yield per mu in this figure refers to yield averaged by agricultural acreage, while sown acreage, equal to multiplying agricultural acreage by multiple-crop index, is more used in China’s current statistics. (2) Li (2003, 2012) is a regional estimation about the Yangzi delta, while all other series are national estimations. (3) Chao et al. (1995) and Chao (2001) are a combination of estimates for the Song dynasty and some rough estimates for the declining trend for the 1750–1900 period. Sources: Dwight H. Perkins, Agricultural Development in China, 1368–1968 (Chicago, Aldine Publishing, 1969); Wu Hui 吴慧, Zhongguo lidai liangshi muchan yanjiu 中国历代粮食亩产研 究 (Beijing, Zhongguo nongye chubanshe, 1985); Wu Hui, “Lishi shang liangshi shangpinlü shangpinliang cegu: Yi Song Ming Qing wei li” 历史上粮食商品率商品量测估: 以宋明清 为例, Zhongguo jingjishi yanjiu 中国经济史研究 1998.4, 16–36; Zhao Gang 赵冈 et al., Qingdai liangshi muchanliang yanjiu 清代粮食亩产量研究 (Beijing, Zhongguo nongye chubanshe, 1995); Zhao Gang, Nongye jingji shilun ji: Chanquan, renkou yu nongye shengchan 农业经济史论 集: 产权、人口与农业生产 (Beijing, Zhongguo nongye chubanshe, 2001), p. 173; Guo Songyi 郭松义, “Ming–Qing shiqide liangshi shengchan yu nongmin shenghuo shuiping” 明 清时期的粮食生产与农民生活水平, Zhongguo shehui kexueyuan lishi yanjiusuo xuekan 中国 社会科学院历史研究所学刊 2001.1, 373–96; Li Bozhong 李伯重, Duoshijiao kan Jiangnan jingjishi 多视角看江南经济史 (1250–1850) (Beijing, Sanlian shudian, 2003), pp. 83, 128, 326–7; Li Bozhong, Zhongguo de zaoqi jindai jingji: 1820 niandai Huating–Louxian diqu GDP yanjiu 中国的 早期近代经济: 1820年代华亭–娄县地区GDP研究 (Beijing, Zhonghua shuju, 2010), pp. 395–8; Shi Zhihong 史志宏, “Qingdai nongye shengchan zhibiaode guji” 清代农业生产 指标的估计, Zhongguo jingjishi yanjiu 中国经济史研究 2015.3, 5–30

estimates of crop yield and land area. First, although crop yield increased steadily until the mid-Qing, the rise in per capita grain output was more modest and even began to drop in the eighteenth century, while the

682

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards 950 Per capita grain (finished product, jin)

Perkins (1969)

850

Wu (1985, 1998)

750

Shi (2015)

Guo (2001)

650 550 450 350 250 150 1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

Year

Figure 18.4 Per capita output of processed grain, 1100–1950 (in jin) Sources: as for Figure 18.3

population grew rapidly even with limited availability of new arable land. Second, per capita grain output declined significantly over the nineteenth century, due to both rapid population growth and declining crop yield. This downward trend continued well into the first half of the twentieth century, although subsistence need could still be met as measured by Perkins, whose estimated per capita processed-grain output was even on the rise after the turn of the twentieth century (partially due to the introduction of new and more efficient food-processing machinery). Although the method differs between Perkins and others, the large disparity in their estimates still deserves more discussion. One startling point is that the peak of per capita grain output reached in the early eighteenth century estimated by Shi and Guo was not only more than 250 percent higher than that of Perkins, but also higher than the peak value reached in China at the end of the twentieth century. This might be partially due to war-induced population loss in the mid-seventeenth century. In their calculations, the gross yield is obtained by multiplying the normal average yield by cultivated land area. However, during wars, much land may have been idle and labor or capital may have been scarce, making normal average yields unobtainable, which implies that gross and per capita yields may be overestimated, inflating the peak values.

683

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

Another issue is that most average-yield records are based on rent data for school-owned or official lands, which typically were in better agricultural condition. One feature of farming in late imperial China was intensive cultivation, but this only developed in response to population growth. Furthermore, it is well known that the cultivated land was substantially underreported in China’s official historical documents, and the area of cultivated land is believed to have increased greatly during the eighteenth and nineteenth centuries. Researchers have to estimate the unreported area of cultivated land and multiply it by the yield per mu to get the amount of total output. However, the intensity of cultivation of the newly cultivated land could not be at the same level as the land cultivated for a long time, and it is difficult to apply the average yield records to all cultivated land. In fact, the cultivation rate of hilly areas in south China was low until the early twentieth century, as the ratio of rice farming was much lower in the mid-eighteenth century than in the early twentieth century and, as a result, there was still much room for the application of intensive cultivation in many parts of China in the eighteenth century.4 Similarly, for areas unfit for intensive rice cultivation but suitable for the New World crops such as maize and sweet potato, their contribution to productivity improvement did not occur until the nineteenth and twentieth centuries. The proportion of the New World crops in food production was less than 1 percent in the eighteenth century,5 and their contribution to average crop yield could not be close to 5 percent.6 Given these facts, the yield records for officially held lands could not represent the average productivity for all types of land, although the problem of overestimation would be less severe with the growth of population, the development of intensive cultivation, or the spread of New World crops. Therefore, while Perkins gives a “subsistence-level” estimate for per capita output, other scholars’ estimates reflect the possibly achievable production capacity of China’s arable land. It is fair to say that the two groups capture the lower and upper bounds of grain output respectively. The actual per capita 4

5

6

Han Maoli 韩茂莉, Zhongguo lishi nongye dili 中国历史农业地理 (Beijing, Beijing daxue chubanshe, 2012), pp. 47, 493–510. On the evolution of cultivation of New World crops from the eighteenth century onward, see Cao Shuji 曹树基, “Qingdai yumi fanshu fenbu di dili tezheng” 清代玉 米番薯分布的地理特征, Lishi dili yanjiu 历史地理研究 1990.2, 287–303; Li Xinsheng 李昕升 and Wang Siming 王思明, “Qing zhi minguo meizhou zuowu shengchan zhibiao guji” 清至民国美洲作物生产指标估计, Qingshi yanjiu 清史研究 2017.3, 126–39. This figure underlies Shi Zhihong’s estimation of crop yield shown in Figure 18.3. See Shi Zhihong 史志宏, Qingdai nongye de fazhan he bufazhan (1661–1911 nian) 清代农业的 发展和不发展 (1661–1911年) (Beijing, Shehui kexue wenxian chubanshe, 2017).

684

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

output should be between these two bounds, and its variation from the eighteenth century to the nineteenth should be much milder than the high estimates show. But as to the precise level and direction of historical trends for per capita output, estimations from production are difficult to prove decisively and more evidence is needed.

Cross-Validation and Explanation Let us now turn back to long-term per capita GDP estimates for China. In the preceding section, we saw that per capita grain output did not suffer any significant long-term decline, and might even have risen, prior to the eighteenth century. As for the periods around and after 1800, for which most scholars cited in the preceding section have estimated a dramatic decline in per capita grain output (Figure 18.4), the impact of such a decline in grain output on per capita GDP should have been partially offset in the long run by gains due to development in the commercial, agricultural, handicrafts, and modern industrial sectors, especially after the mid-nineteenth century when the treaty ports were opened. Moreover, we just discussed that the decline in per capita grain output around the nineteenth century might have been exaggerated. Thus the decline in per capita GDP in the late Qing should have been milder, or there might even have been a rise. Based on these considerations, the actual trajectory of per capita GDP from the Song to the Qing may lie somewhere between the estimates made by Maddison and those of more recent scholars cited above. We can also verify which way per capita GDP has changed by crosschecking with ordinary people’s experience, such as how their real wages evolved. In theory, if the decline in per capita GDP was due to a fast-growing population and its resulting diminishing marginal returns to labor, real wages should decline. Ideally, we should use annual labor income data for a typical rural household or family, as that was the standard unit of comsumption and production, but most historical records refer to daily wage rates of urban labor. However, labor did move between rural and urban sectors. Although there were frictions to changing one’s occupation or moving between places, it is still reasonable to assume that the trend should be similar in the long run between annual wages for urban labor and for rural labor. Furthermore, as seasonality for urban wages was much lower than for rural wages, it is more suitable to estimate the wage trend based on the annual wage for urban labor. Before discussing the long-run wage trend, we will first introduce the wage picture by dynasty. 685

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

Wage records from the Song dynasty that are used in existing studies are largely confined to government workers. According to Cheng Minsheng and William Guanglin Liu, craftsmen working for the Song government often received wages in both coin and rice, and a typical pay per day consisted of two to three sheng of rice and a cash amount between tens and a few hundred wen of copper coin.7 If the cash part is converted to rice equivalents, the total wage rate ranges from less than three sheng to above ten sheng of rice. However, total wage rates above ten sheng of rice were mostly for skilled work or laborious tasks that involved long-distance travel or war-related dangers. When these exceptional high wage observations are excluded, the wage gap between the Northern and the Southern Song periods becomes much smaller, with typical day wages for both periods in the range of four to nine sheng, which translates into three to six sheng based on the Ming and Qing weight standards, or about 2.52 to 5.04 kilograms.8 Now let us turn to the Ming period. Unskilled laborers in the mid-sixteenth century were often paid three fen of silver per day, equivalent to five to six sheng of rice at the time.9 However, as the price of rice rose in the early seventeenth century, Jiangnan laborers earning this silver wage could only afford to buy fewer than three sheng of rice. While such a real-wage decline was caused by nominal wage stickiness, the actual wage depreciation was not so dramatic, since a substantial part of the day wage was usually paid in kind. Gao Shouxian uses official records to examine price levels in Beijing during the Wanli reign (1572–1620). Translated into real wages at then current prices, even unskilled laborers could earn about five sheng of rice per day.10 In summary, from the Song to the Ming, we do not find significant change of real wages in units of rice.

7

8

9

10

Cheng Minsheng 程民生, Songdai wujia yanjiu 宋代物价研究 (Beijing, Renmin chubanshe, 2008), pp. 339–68. William Guanglin Liu, The Chinese Market Economy, 1000–1500 (Albany, SUNY Press, 2015), pp. 245–57. The measurement units, sheng and dou, were smaller during the Song than in the later times, as one dou or ten sheng in the Song was equivalent to about 6.2 sheng of the Ming and Qing times. See Wu Luo 吳落, Zhongguo duliangheng shi 中国度量衡史 (Shanghai, Shangwu chubanshe, 1937). Meanwhile, one sheng of rice in Qing times was equivalent to about 0.84 kilogram, according to Peng Kaixiang 彭凯翔, Qingdai yilai de liangjia: Lishixue de jieshi yu zaijieshi 清代以来的粮价: 历史学的解释与再解释 (Shanghai, Shanghai renmin chubanshe, 2006), p. 25. Hu Tieqiu 胡铁球, “Mingdai guanfeng goucheng biandong yu junyaofade qidong” 明 代官俸构成变动与均徭法的启动, Shixue yuekan 史学月刊 2012.11, 22–42. Gao Shouxian 高寿仙, “Ming Wanli nianjian Beijing de wujia he gongzi” 明万历年间 北京的物价和工资, Qinghua daxue xuebao (zhexue shehui kexue ban) 清华大学学报 (哲学社会科学版) 2008.3, 45–62.

686

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

For the first half of the Qing dynasty, more systematic archival data are available. Skilled kiln workers in Jingdezhen, the global capital of the porcelain industry, earned around 6.5 sheng per day during the Jiajing (1521–1566) and Wanli (1571–1620) reigns of the Ming dynasty. At the other end of the skill spectrum, the elderly, children, and disabled persons who worked at the task of grinding pigments (rubo 乳钵) in Jingdezhen during the Qianlong reign (1735–1795) were making 0.6 silver taels per month (excluding meals), or 2.5 sheng of rice per day; by the early twentieth century, the day wage for these grinder workers rose to 2.7 sheng a day. We can also look at ordinary laborers in the early twentieth century working as craftsmen and at unskilled jobs, such as coolies and handymen, who were paid seven sheng (meals included) and five sheng of rice per day respectively.11 Thus, at least in the industrial city of Jingdezhen, there was no noticeable decline in the rice wage from the Ming to the early twentieth century. As another example, wages for carpenters in Huizhou also did not show any decline during the Qing period. From records in various accounting books, it can be roughly estimated that in the early and mid-eighteenth century, low-skilled workers (with almost no skills) were paid eight fen of silver per day (or more than four sheng of rice, with meals included), while high-skilled workers were paid around ten to twenty fen a day. By the late nineteenth century, even low-income carpenters were earning around eleven fen of silver a day, or approximately five sheng of rice. The Huizhou experience confirms the pattern found in Jingdezhen. Robert Allen and his collaborators have studied welfare ratios for several cities of China from the early eighteenth century to the early twentieth century according to wage data from different sources.12 Their findings show some decline in welfare ratios around the early nineteenth century, when there was a switch in their data source, but it is difficult to find significant change for the whole period. Research on wage series of different occupations in Beijing also shows that

11

12

Liang Miaotai 梁淼泰, “Mingdai houqi Jingdezhen yuqichang jiangyi zhidu de bianhua” 明代后期景德镇御器厂匠役制度的变化, Zhongguo shehui jingji shi yanjiu 中国社会 经济史研究 1982.1, 83–93; Liang Miaotai 梁淼泰, “Qingdai Jingdezhen yichu lucun yaohaode shouzhi yingli” 清代景德镇一处炉寸窑号的收支营利, Zhongguo shehui jingji shi yanjiu 中国社会经济史研究 1984.4, 1–16. Robert C. Allen, Christine Moll-Murata, Debin Ma, and Jan Luiten van Zanden, “Wage, Price and Living Standards in China 1738–1925: In Comparison with Europe, Japan, and India,” Economic History Review 64.S1 (2011), 8–38.

687

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

Figure 18.5 Trend in real wages, 971–1920 Notes: (1) Each bubble represents one observation; the solid lines are obtained by LOESS (locally weighted smoothing) estimation; the shadow areas are confidence intervals of the smooth estimations with confidence level of 95 percent. (2) Abnormal wage rates above 10 sheng of rice were dropped.

there was no significant trend, up or down, for real wages in the nineteenth century.13 Thus the Jingdezhen and Huizhou experiences were not outliers. To provide a general view about wage trends for the second millennium, we combine the different estimates introduced above in Figure 18.5. The smoothed trend line shows that the average real wage from Song to Qing stayed around four to five sheng of rice per day and there was no significant long-run trend tracking the persistent population growth. If the development of handicraft and intensive cultivation is considered, there would be an increase in annual working days and the trend of the annual wage would be more positive in the long run. Meanwhile, there were shorter-run fluctuations, but it is difficult to make statistically significant judgments when the number of observations is limited and the confidence interval is wide. One exception is the mild increase in real wages from the low around late Ming to the peak of the late eighteenth century 13

Peng Kaixiang 彭凯翔, Cong jiaoyi dao shichang: Chuantong Zhongguo minjian jingji mailuo shitan 从交易到市场: 传统中国民间经济脉络试探 (Hangzhou, Zhejiang daxue chubanshe, 2015), pp. 301–12.

688

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards 1000

Price index, 960–1276 = 100

Rice price index Silk cloth price index

100

1590

1560

1530

1500

1470

1440

1410

1380

1350

1320

1290

1260

1230

1200

1170

1140

1110

1080

1050

990

1020

960

10

Year

Figure 18.6 Silk and rice price indices, 960–1600 Note: Prices are transformed to indices with the average price for the whole Song (960–1276) as 100. Source: Peng Xinwei 彭信威, Zhongguo huobi shi 中国货币史, 2nd ed. (Shanghai, Shanghai renmin chubanshe, 1965), pp. 505–7, 704–11

and the early nineteenth. That is contrary to estimates that show a large decline in per capita grain yield or per capita GDP around the eighteenth century, when the population pressure was most intense. As wages are affected by factors such as employer type (government versus private sector) and job category, we may use the relative prices of different product types as an additional metric to measure changes in labor price over time. Take textile products as an example. Compared to grain crops, silk and cotton production were generally more labor-intensive before the twentieth century. If population growth diminished marginal returns to labor, real wages would decline and more labor would shift to labor-intensive industries, leading to declines in relative price for labor-intensive products as opposed to grain crops. Therefore changes in these relative prices should reflect changes in real wages. Figure 18.6 treats the average price of each product for the whole Song period (960–1276) as 100 and displays the price indices of both rice and silk between the tenth and sixteenth centuries. Relative to their Song price levels,

689

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

both indices went down by half during the first half of the Ming. The drop in rice price index was slightly more than that of silk. During the Ming period, the cotton textile industry experienced rapid growth, while the relative position of silk products weakened. Silk prices therefore were affected not only by an increase in workforce supply but also by an inward shift of the demand curve. Given this background, the fact that silk prices declined by less than rice prices further demonstrates that real wages could not have declined from the Song to the Ming. For the eighteenth century, during which wage data and grain output estimations diverged greatly, there also was no significant change in relative prices between textile products and grain crops, although inflation was observed.14 This pattern could be consistent with a mild increase in real wages if the postwar economic recovery from the mid-seventeenth century onward and the continuing development of the textile industry led to improvement in labor productivity. However, a great decline in per capita grain output or per capita GDP cannot be supported by the evidence on relative prices.

Consumption and Welfare So far we have focused on the production side to assess the evolution of China’s living standards. Since production and consumption must be equal in the long run, we can also address the question from the consumption side. Production data are available only for a few products such as crops and minerals, and information for other products and services can only be estimated by analyzing consumption. Moreover, focusing on consumption can help mitigate certain aggregation errors, allowing us to understand changes in people’s welfare more systematically. As the old Chinese saying goes, “food is what matters for ordinary people.” Thus food figures extremely high in ordinary people’s consumption basket in agrarian societies. This partially explains why there are abundant historical documents and sources on crop production, and economic historians have devoted much attention to the subject, particularly food output per capita. As Wu Hui points out, a daily ration of two sheng of rice, roughly one kilogram, was distributed to each soldier in the Song. This relatively high ration standard possibly included non-staple foods. On the other hand, a soldier in the Ming and Qing generally received a daily ration of one sheng of rice (0.63 14

Peng, Qingdai yilai de liangjia, pp. 32–46.

690

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

kilogram). However, it should be taken into account that the overall civilian food consumption must have been lower than the official ration at the time. Wu also estimates that in the Ming and Qing, the farming and non-farming population consumed on average 17.5 kilograms and 16.4 kilograms of processed rice per person per month respectively. Assuming that 70 percent of the total population engaged in farming, the aggregated average consumption per capita would be 17.2 kilograms per month (0.575 kilograms per day for thirty days per month), or 206.5 kilograms per year. In addition to staple food consumption, some grain output was used to make snacks and other non-staple food products, such as seeds and animal feeds, for which Wu Hui suggests a ratio of 7:3 between staple food and other grain usage. Applying this ratio in our calculation gives an estimated total consumption of grain per capita of 295 kilograms per year.15 The Ming and Qing populations clearly consumed far more than needed for subsistence: the standard for famine relief programs during the Song normally distributed a daily ration of one sheng of rice (0.5 kilogram) and 0.5 sheng (0.25 kilogram) to each adult and child respectively, and occasionally a reduced daily ration in the form of rice porridge.16 Even when compared to modern standards, this level of consumption was still high. Statistics since 1950 show that only after 1986 – as a result of agricultural reforms – did the per capita consumption of processed grain in China reach sixteen kilograms per month (0.53 kilogram per day), the so-called basic subsistence level in the Song, while the rural farming population reached this consumption level as early as 1979, since they still were engaged in heavy manual labor just as farmers in traditional China were.17 In any case, an annual ration of 206.5 kilograms of processed rice per person, or total annual consumption of 295 kilograms of processed grain, by no means represented the minimum subsistence standard of living. So it provides further evidence to support Perkins’s estimates on grain production as a reasonable lower bound. On the other hand, estimates by Guo Songyi and Shi Zhihong, both much higher than 295 kilograms per capita per year before the nineteenth century, can be regarded as frontiers of production possibilities. Wu Hui considers grain output surplus as a precondition for enhancing commercial development. In other words, only the existence of a substantial grain surplus would make it possible for grain to be allocated to nonbasic food consumption and to promoting nonagricultural industries. 15 16 17

Wu, Liangshi muchan, pp. 64, 68, 78–81; Wu, “Liangshi shangpinlü,” 16–36. SHY, ruiyi 瑞異 2:22a, shihuo 食貨 60:1a–b, 68:146b, 68: 147a, etc. Zhongguo nongye tongji ziliao huibian, 1949–2004 中国农业统计资料汇编 1949–2004 (Beijing, Zhongguo tongji chubanshe, 2005).

691

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

Although available consumption data are inadequate for completely recovering the actual output at the time, they suffice to illustrate that grain production was more than enough to keep the society above subsistence during normal times. Certainly the Jiangnan consumption standards of the Qing can hardly be regarded as a “survival economy,” and it is quite plausible to suggest that luxury consumption did happen in the Song and late Ming. Even for inland regions in the Qing, research about Ba county finds that durable luxury goods were not rare for middle-income people.18 Although it is still arguable whether there was a “consumer revolution” in daily consumption and material life during the Song or Ming–Qing periods, it is hard to reject the claim that Chinese society experienced many profound changes in consumption habits during the second millenium. Some of these changes were related to production, among which breakthroughs in the cotton textile industry were the most important examples, as they directly affected people’s day-to-day lives. One particular kind of cotton, Gossypium herbaceum, was introduced into northwest China as early as the second century, but its low yield and short fibers, as well as the limited textile technology at the time, constrained its spread. In fact, the textile industry was dominated by silk and hemp manufacture up to the Southern Song, after which time the shift to cotton textiles occurred as cotton’s popularity grew and spread rapidly from the southeastern region (see Pomeranz’s chapter in this volume). By early Ming, when cotton textile production had become a national phenomenon, cotton overtook hemp as the main clothing material worn by the general public. An eighteenthcentury prefect in Fuzhou (Fujian) observed that the much cheaper and softer cotton had become overwhelmingly popular: “there are none who do not wear cotton, nor is there anyplace not suited to cultivating cotton.”19 This long-run development of the cotton textile industry, together with 18

19

For the Song, see Shiba Yoshinobu 斯波義信, Songdai shangyeshi yanjiu 宋代商业史研 究 (Taipei, Daohe chubanshe, 1997), pp. 473–9. For the late Ming, see Wu Renshu 巫仁 恕, Pinwei shehua: Wanming de xiaofei shehui yu shidafu 品味奢华: 晚明的消费社会与 士大夫 (Beijing, Zhonghua shuju, 2008). For the Qing, the Yangzi delta situation is explored by Huang Jingbin 黄敬斌, Minsheng yu jiaji: Qingchu zhi minguo shiqi Jiangnan jumin de xiaofei 民生与家计: 清初至民国时期江南居民的消费 (Shanghai, Fudan daxue chubanshe, 2009). Consumption in an inland district, Ba county in Sichuan, is discussed in Wu Renshu 巫仁恕 and Wang Dagang 王大纲, “Qianlongchao difang wupin xiaofei yu shoucang de chubu yanjiu: Yi Sichuansheng Baxian wei li” 乾隆朝地 方物品消费与收藏的初步研究: 以四川省巴县为例, Zhongyang yanjiuyuan jindai shi yanjiusuo jikan 中央研究院近代史研究所集刊 89 (2015), 1–41. Li Ba 李拔, Zhongmian shuo 種棉說, quoted in Zhao Chengze 赵承泽 (ed.), Zhongguo kexue jishu shi: Fangzhi juan 中国科学技术史: 纺织卷 (Beijing, Kexue chubanshe, 2002), p. 22.

692

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

changes in crop variety, technology, and factor endowment, served to further enhance the industry’s relative importance. For example, the processing-cost component in cotton cloth prices fell from more than 95 percent in northwest China during the Tang to just 25 percent by the early Ming, and to 20 percent by the early Qing.20 Cotton clothing was not only made much cheaper, but cotton clothes were also smoother, warmer, and more amenable to dyes than hemp cloth. Mentions of women fulling cloth (daoyi 搗衣) – pounding hemp and ramie with a mallet to soften the stiff fibers – which frequently appear in Tang and Song poetry disappear in Ming and Qing poems. This evolution of the textile industry led to gradual improvements in people’s standards of living. Porcelain is another example illustrating how technical advancement changed people’s lives. Jingdezhen rose as the most important center for porcelain production during the Song and after. It experienced two major technical advancements in the following centuries.21 The first was the binary blending method, introduced in the Yuan, which significantly reduced rawmaterial expenditure by blending clays and stones. The second was an innovation that raised the highest kiln temperature from under 1,200 °C in the Five Dynasties and the Song to over 1,300 °C in the Qing. These two innovations improved both the quality and quantity of porcelain production and allowed the large-scale industry to expand rapidly in the Ming and Qing. There were also changes in regulations governing the industry. During the Wanli reign of the Ming dynasty, there was a severe shortage of the government-monopolized Macang clay, resulting in a major decline in the quality of porcelain produced by state-owned porcelain workshops, while private producers discovered and later controlled the much cheaper, yet still highquality, kaolin clay from Gaoling Mountain.22 Since then the private sector of Jingdezhen’s porcelain industry flourished, producing better-quality porcelain than state-owned workshops. As a result, ordinary people could purchase high-quality porcelain products that previously were only available to privileged government officials and elite consumers. Private production played the most vital role in expanding the porcelain industry. It is estimated that, during the 200 years from the late Ming to the mid-Qing (Qianlong’s reign), Jingdezhen produced at least 6.52 million tons 20

21

22

Kang Chao, The Development of Cotton Textile Production in China (Cambridge, MA, East Asian Research Center, Harvard University, 1977), p. 73. Liu Xinyuan 刘新园 and Bai Kun 白焜, “Gaolingtu shikao” 高岭土史考, Zhongguo taoci 中国陶瓷 7 (supplement) (1982), 141–82; Lu Jiaxi 卢嘉锡 and Li Jiazhi 李家治, Zhongguo kexue jishu shi: Taoci juan 中国科学技术史: 陶瓷卷 (Beijing, Kexue chubanshe, 1998), p. 349. FLXZ, 8.8b–9a.

693

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

of porcelain of diverse quality, or tens of billions of pieces. The life span of each porcelain piece was estimated to be around ten years, and therefore hundreds of millions of porcelain items were in use each year, but the imperial court possessed only a small portion of them (about 100,000 pieces).23 This estimated level of porcelain usage implies that porcelain was no longer a product of luxury consumption, and almost every household possessed several pieces. It should be noted that besides Jingdezhen, there were many other porcelain centers in Jiangxi, Zhejiang, Jiangsu, Fujian, and Guangdong provinces. Similarly, the development of woodblock printing using cheap bamboo paper facilitated the growth of the printing industry and enhanced cultural consumption. Although not as strong or durable as hemp or bast paper, bamboo paper was considered suitable for commercial publications such as textbooks and popular literature. Along with reduced paper cost, a standardized printing font known as the “Song style” (songti 宋體) was developed in the Ming, making books even more affordable and bringing increasing prosperity to the publishing industry.24 A popular folklore saying at the time reveals something of the popularity of writing and reading: In recent decades, a student who has scored well in at least one round of exams will surely write a book; anyone who has done well materially will leave a tombstone with richly inscripted essays; most such writings must be lost because, should they all survive, there would not be enough space to store them even if all of the land were used as bookshelves . . . Today’s printed books are so voluminous that if they were burned, it would cause wood and coal [fuel] prices to crash.25

Although these observations were all focused on private editions, there was an impressive quantity of publications, including popular novels and general books, during the Ming–Qing era (see McDermott’s chapter in this volume). These developments in cultural consumption not only fostered a vibrant urban popular culture, but also created important industries in the Ming– Qing economy and contributed to the spread of secular knowledge about agriculture, medicine, business, and so on. 23 24

25

Liu and Bai, “Gaolingtu shikao,” pp. 364–9. J.P. McDermott, Social History of the Chinese Book: Books and Literati Culture in Late Imperial China (Hong Kong, Hong Kong University Press, 2006), pp. 28–32. However, note that this mistakenly asserts that the wage for woodblock carvers decreased from thirty-five wen for 100 graphs in 1250 to 3.5 wen in 1600. The wage in 1600 should be 3.5 fen of silver, much higher than thirty-five wen, so the wage actually rose. SLQH, 7.13b–14a.

694

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

Besides advancement of technology, market development and population movements contributed to growth in trade, promoted the division of labor, and diversified consumption patterns. As noted earlier, the most influential new crops introduced into China in the Ming–Qing period were maize and sweet potato, both of which came to China via trade routes from Spain’s American colonies. We have already discussed their impact on food output, and will shortly address their contribution to Chinese people’s ability to cope with famine and other natural disasters. Although not as important for human survival, other imported crops helped to diversify the diet of ordinary people: carrots (brought from northern Europe via Iran during the Yuan); watermelon (introduced in the Liao and Song); chili peppers, tomatoes, pumpkins, zucchini, beans, potatoes, peanuts, pineapples, tobacco, strawberries, and sunflowers (all brought from the New World to China after the late Ming); and apples (transplanted from Europe during the nineteenth century).26 All these transplanted foreign fruits and crops surely have enhanced the well-being of the Chinese population since the Song. Even indigenous vegetable and fruit species spread more widely as a gradual result of population movement, the circulation of commodities, and urbanization. For example, it was only during the Song that more delicious varieties of Chinese cabbage were cultivated. After its introduction to Beijing from southern China in the mid- to late Ming, the yellow cabbage became a major part of Beijing residents’ winter diet, while in some developed regions such as Yuyao in Zhejiang province it was not successfully adopted until the eighteenth century.27 Similarly, the quality of radish was constantly improved, becoming an essential part of the Chinese food portfolio, overtaking okra and turnip during the Ming and Qing. Rape, originally used as a vegetable, was found to be an important source of cooking oil in the Song and Yuan. After its introduction to southern China, rape complemented rice cultivation to create a double-crop rotation system (planting rice in the spring–summer and rape in the fall–winter), significantly enriching the supply

26

27

See Wang Siming 王思明, “Meizhou yuanchan zuowude yinzhong caipei jiqi dui Zhongguo nongye shengchan jiegoude yingxiang” 美洲原产作物的引种栽培及其 对中国农业生产结构的影响, Zhongguo nongshi 中国农史 2004.2, 16–27; Sun Ji 孙机, Zhongguo gudai wuzhi wenhua 中国古代物质文化 (Beijing, Zhonghua shuju, 2014), pp. 17–24. See Wang Yuquan 王毓铨 (ed.), Zhongguo jingji tongshi: Ming 中国经济通史: 明 (Beijing, Jingji ribao chubanshe, 2000), vol. 1, p. 397; Zhu Wenzhi 朱文治, “Xiaohan zhuzhici” 销寒竹枝词, in Lei Mengshui 雷梦水 et al. (eds.), Zhonghua zhuzhici 中华竹 枝词 (Beijing, Beijing guji chubanshe, 1997), pp. 2114–15.

695

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

of cooking oil. The value of such developments for lifting living standards should not be underestimated. Transfer of vegetables and fruits across different climatic and ecological regions was generally more difficult and slower than that of cereals, so even in Ming–Qing times there was still room for their wider adoption. Liang Jiamian provides a survey of this process. His comparison shows that the dietary structure for Chinese people was certainly substantially different, and much improved, in the Qing than in pre-Song times, following the formation of the typical modern Chinese summer vegetable diet composed of melons, eggplants, leafy vegetables, beans, and meats.28 Accordingly, vegetables and fruits and their processed varieties became the second most important source of energy and nutrition, surpassed only by grains.29 Apart from the significant developments discussed above, other “luxury” goods discussed in the scholarship on material culture, such as furniture, tourism, and carriage services, also achieved remarkable progress from the late Ming onwards, raising living standards in various ways. Consumption beyond subsistence needs was driven not only by improvement on the supply side, but also by changes from the demand side as a result of changes in real income, culture, and institutions such as the civil service examination system. These factors affected individuals’ welfare, but it is difficult to measure such changes directly. The welfare ratio studied by Robert Allen et al. is a measure of real wages, as discussed above. Novel methods, including height measurement and age-heaping (a measure of literacy), are sometimes used to quantify welfare changes in Chinese history, but they are mainly limited to periods after the eighteenth century, when relevant data become available.30 However, there are numerous biographical records that provide information on life span, a commonly used index of welfare, for earlier periods. Figure 18.7 shows the life spans, measured by death age, of more than 28,000 Chinese males born between the seventh and nineteenth centuries, based on records in the China Biographical Database (CBDB). Although this database includes mainly the elite and officials, the sample still can represent the situation of a much larger population. There was great annual variation in age of death that reflected the impact of dynastic turnovers and wars, but the 28

29 30

Liang Jiamian 梁家勉, Zhongguo nongye kexue jishu shigao 中国农业科学技术史稿 (Beijing, Nongye chubanshe, 1989), p. 532. Huang, Minsheng yu jiaji, pp. 103–6. Joerg Baten, Debin Ma, Stephen Morgan, and Qing Wang, “Evolution of Living Standards and Human Capital in China in the 18th–20th Centuries: Evidences from Real Wages, Age-Heaping, and Anthropometrics,” Explorations in Economic History 47.3 (2010), 347–59.

696

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

Figure 18.7 Long-term trend in age of death, 600–1870 Note: For the period before 1700, the LOESS smoothing line is plotted to show the trend; for the period after 1700, the dashed line shows the trend adjusted for the effect of truncations at the end of the sample period. For details on the adjustment, see Peng Kaixiang 彭凯翔, “Qingren nianshou zai tan: Jiyu zhuanji shujuku de ruogan kaocha” 清 人年寿再探: 基于传记数据库的若干考察, Xin shixue 新史学 12 (2020), 153–76.

trend shows that Chinese life spans tended to increase in the long run. The most important improvement occurred between the eleventh and twelfth centuries. Although the rise in life span flattened after 1200, it did not decline as the population grew. Combined with advances in the consumption portfolio, the life span trend does not support the Malthusian view of declining living standards in the post-Song period.

Capacity to Deal with Risks to Survival The Book of Rites declares that “the state should budget its expenses for a thirty-year time frame,” while the Guan Zi proclaims that “it is in the nature of farming that there will be months of dearth and years of surplus.”31 These venerable homilies point out a fundamental fact of agrarian societies:

31

Chapter 5, “Wangzhi” 王制, in LJ, 1, p. 1334a; Chapter 48, “Zhiguo” 治國, in GZ, 2, p. 925.

697

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

although aggregate output may be adequate in the long run, shortage of subsistence goods due to short-term climatic events, natural disasters, or human disruptions can easily threaten people’s survival. Such risks adversely affected the expected utility or livelihood of farmers. For this reason, when measured in annual aggregates such as “average annual output per capita” over a long period such as a decade or more, grain production might have been adequate, but this does not mean that farming families were able to handle short-term output shocks at the individual household level. The inability to deal with short-term deviations can be particularly serious for households who were constantly living on the verge of subsistence and had no accumulated surplus buffer. Thus, when measuring human developmental progress, we should not just use average per capita GDP per year as the sole metric of living standards or welfare. We must also pay attention to progress in human capacity to deal with risks such as drought, flood, earthquake, epidemic, war, and other types of famine events that threaten people’s livelihood. Historical records and a growing body of physical evidence illustrate that there have been several significant climatic swings since the tenth century, including the Little Ice Age in the sixteenth and seventeenth centuries and the abnormally cold period around the 1820s (see Bello’s chapter in this volume). Since extreme cold periods are usually accompanied with drought, famine tends to happen at the same time. The great famine in the late Ming and the “Daoguang recession” in the early and mid-nineteenth century are examples of risk events that correlated with climate changes. While much historical research has focused on the effects of natural disasters on the rise and fall of dynasties and on the macro economy, we should also give attention to how they affected individual lives. Grain price fluctuations may be the most informative evidence for the severity of survival risks faced by an agricultural society. On the one hand a significant share of volatility in grain prices can be attributed to such shocks, but on the other hand it is nothing compared to the immediate price surge right after a natural disaster, and it should not be surprising that unpredictable risk events cause severe distress. For example, it was commonly said that after a major disaster some farmers would be forced to sell their wives or daughters in order to use the proceeds to buy food. A recent study indicates that during the eighteenth and nineteenth centuries a rise of 10 percent in the average rice price would cause a wife’s selling price to drop by 30 percent, because a major famine created by a disaster would force more husbands to sell wives or daughters, while the demand for wives would be low at such 698

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

a time, resulting in oversupply on the wife-selling market. Distress-led wife selling was especially prevalent in years of extreme and long-lasting drought.32 The capacity to deal with risk was even worse in previous eras. The general public suffered more when risk events hit. Although archival data on grain prices for the Song are not available as systematically as for the Qing, there are enough to reveal the general state of affairs. For example, in early 1074 the price of rice in the Song capital, Kaifeng, rose to 150 wen per dou, but fell to ninety wen in March after the government allocated more than 2 million shi of grain for sale in Kaifeng’s markets, while a normal price at that time was more like seventy to eighty wen. Clearly, grain price could easily double in the short run even in the capital, where government oversight was tightest. In Guangzhou, grain prices in 1171 were four times the normal levels after several years of continuous bad harvests, while Hangzhou was hit with a major crop failure in the year 1000, raising rice prices to nine to ten times greater than in normal years.33 Other evidence also indicates that the magnitude and frequency of price oscillations, as well as regional differences in such experiences, were all much higher in the Song than in the Qing. Therefore, in terms of survival distress imposed by risk events, there seems to have been noticeable improvement from the Song to the Qing. To examine how Chinese society innovated to improve ordinary people’s capacity to deal with risk, we divide risk mitigation methods into five types: technological innovations (e.g., new crops), lineage or clan institutions, nonkinship social networks (e.g., the church), financial markets, and government welfare.34 Next, we turn to assessing developments in each of the five risk mitigation solutions from the Song onwards.

Introduction of Insurance Crops First, let us look at how new technologies affected Chinese people’s capacity to deal with risk. Earlier, we discussed the effects of new farming techniques and the New World crops on productivity and overall food output. While the contribution by the new crops to gross production was limited before the eighteenth century, their contribution to improving well-being through their 32

33 34

Chen Zhiwu 陈志武, He Shijun 何石军, Lin Zhan 林展, and Peng Kaixiang 彭凯翔, “Qingdai qiqie jiage yanjiu: Chuantong shehui li nüxing ruhe bei yongzuo bixian zichan?” 清代妻妾价格研究: 传统社会里女性如何被用作避险资产, Jingjixue jikan 经济学季刊 2019.1, 253–80. Cheng, Songdai wujia, pp. 125–6, 138–40. See Chen Zhiwu 陈志武, Jinrong de luoji 金融的逻辑 (Beijing, Guoji wenhua chuban gongsi, 2009).

699

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

insurance or risk mitigation value was much more important. The key here lies in the fact that, relative to traditional crops such as rice and wheat, maize and sweet potato are more drought-resistant and less dependent on the quality of soil and fertilizer, with more flexible sowing times and longer growing seasons during the year. Thus farmers could plant these new crops anytime they anticipated an upcoming drought or encountered a poor harvest for the main crops. These advantages of the New World crops, especially sweet potato, were widely acknowledged by people and officials in southeast China already in the early seventeenth century, and they were advocated nationally by the government around the mid-eighteenth century. As insurance crops, maize and sweet potato might not have been noticeable as a percentage of overall food production during years of normal harvests, but their distinctive significance became obvious during drought and other crop-failure years when they were crucial for saving lives. For this reason, their relatively low share in total food output in normal times did not diminish their high marginal contribution to meeting subsistence needs in times of crisis. It is therefore not surprising that the New World crops were found to be significant contributors to China’s population growth and to long-term decline in the frequency of peasant revolts.35 Meanwhile, the new crops, especially maize, were double-edged swords. In the short run, they helped relieve the population from the impact of disasters, reducing social unrest. But in the long run cultivation of these crops, especially maize, damaged mountain forests and caused soil erosion. During the Jiaqing (1795–1820) and Daoguang (1820–1850) reigns, officials took notice of these ecological problems and repeatedly forbade planting maize. However, it is undeniable that the new crops lifted the population’s capacity to deal with risk and made positive contributions to mitigating subsistence risks. As a result, by the 1930s maize had become the fourth most important crop, surpassed only by rice, wheat, and millet.

The Rise of Lineages and Other Social Organizations Since antiquity, kinship networks have had a long history in China, but their initial development was largely limited to the aristocracy. Before the Song dynasty, Confucian classics and institutions such as ancestral halls and family temples were preoccupations of the elite, not of the common people. Then 35

Shuo Chen and James Kai-sing Kung, “Of Maize and Men: The Effect of a New World Crop on Population and Economic Growth in China,” Journal of Economic Growth 21.1 (2016), 71–99; Ruixue Jia, “Weather Shocks, Sweet Potatoes and Peasant Revolts in Historical China,” Economic Journal 124 (2014), 92–118.

700

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

the Song launched a long sequence of efforts to bring the lineage infrastructure to commoners at the grassroots level. Neo-Confucian philosophers of the Dao Learning tradition, for example Cheng Yi and Zhu Xi, revised and redesigned the rites and rituals of ancestor worship to render them more appropriate for common people. In particular, Zhu Xi made family-temple construction and ancestor worship no longer a privilege reserved for aristocrats, but rather a universal obligation required of all. Also during the Song, commoners were made eligible to take the civil service examinations, which provided an extra incentive for children of commoners to study the Confucian classics, closing the gap between Confucian book learning and common people’s daily practice. These reforms led to the construction of ancestral halls, family temples, and lineage genealogies among commoners all over the country, especially from the early sixteenth century, when laws that had restricted the construction of ancestral temples to government officials were lifted. Two institutional factors in the Ming dynasty contributed to the development of lineage formation at the local level, especially in southeast China.36 First, difficulties in tax collection prompted local governments to accept lineage organizations both as intermediaries in tax collection and as basic administrative units. Second, as more members of local society climbed up the social ladder through the civil service examinations and other methods, they founded lineage organizations to demonstrate their integration into the elite class of the empire, by which means they also became middlemen between government and local society. Ancestral halls and temples, together with periodic family rituals, made the lineage the dominant organization in the local social landscape. Although the origins of the lineage trace back to institutions created to provide relief for clansmen during the Song, it really became the most important organization for interpersonal and intertemporal risk sharing for the Chinese people after lineages became basic administrative units with significant property estates from the mid-Ming onward. Reciprocity between families or kinsmen was a basic method for dealing with risk shocks. Frequent and regular lineage assemblies helped members address adverse-selection and moral-hazard problems, as such gatherings reduced informational asymmetry among kinsmen. Furthermore, collectively owned lineage property, whose rental yields almost always exceeded what was needed to pay for 36

See Zheng Zhenman 郑振满, Ming Qing Fujian jiazu zuzhi yu shehui bianqian 明清福建 家族组织与社会变迁 (Beijing, Zhongguo renmin daxue chubanshe, 2009), pp. 183–95; David Faure, Emperor and Ancestor: State and Lineage in South China (Stanford, Stanford University Press, 2007), pp. 125–48.

701

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

lineage sacrifices and taxes, acted as the key endowment or trust for the lineage. Besides operations to increase the trust fund’s value, such as providing credit to merchants within the lineage, the more basic functions of the lineage trust fund were to provide relief for the elderly or poor kinfolk and widows; to subsidize lineage members when they met with unexpected hardship and emergency expenses, or faced the high costs of attending the imperial civil service examinations or marriages and funerals; and to share resources during times of famine, banditry, and other general crises. In this way, the Chinese lineage could be regarded as an “internal financial market” for risk sharing. Note that as lineage institutions developed significantly after the midMing, other nongovernment organizations similarly proliferated at the local level, and charitable organizations were an important part in this process. These organizations provided aid to the elderly, widows, and orphans; took part in poor and famine relief; distributed medicine to the poor; and provided burial services for the indigent.37 Some of these charitable organizations were sponsored by trade associations or guilds which, as basic administrative units similar to lineages in rural areas, provided some relief for members directly. Their functions were diverse, but these organizations offered social welfare services and helped reduce the impact of survival-risk events on people’s livelihood. While not necessarily raising productivity or per capita income, the development of lineage institutions and social organizations from the Song onward significantly improved the Chinese people’s living standards as they enhanced risk mitigation capacity.

The Development of Financial Markets Financial markets can, of course, directly address risk mitigation challenges both before and after a natural calamity occurs. In a society without modern insurance and other financial products, lending and borrowing were the most important financial tools to soften the blow of temporal shocks to people’s lives. Since detailed historical credit transaction data are not available, it is difficult to track the evolution of credit volume since the early dynasties. We can only rely on interest rates to infer whether there was any progress. According to the description of Yuan Cai in his Precepts for Social Life, in the twelfth century, interest rates for lending money were normally 36–60 percent per annum, and 24–48 percent per annum for secured loans from 37

Fuma Susumu 夫馬進, Chu¯goku zenkai zendo¯ shi kenkyu¯ 中国善会善堂史研究 (Kyoto, Do¯ho¯sha, 1997); Liang Qizi 梁其姿, Shishan yu jiaohua: Ming Qingde cishan zuzhi 施善 與教化: 明清的慈善組織 (Taipei, Lianjing chuban shiye gongsi, 1997).

702

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

pawnshops.38 These rates were considered acceptable at the time. Interest rates charged by the government for the “Green Sprouts” loans (qingmiao qian 青苗錢) during the Wang Anshi reform period – issued at the time of spring planting (the season when food shortage was most severe) and paid off at the fall harvest – were about 20 percent for a six-month period, corresponding to an annualized rate of 40 percent; the Bureau of Market Transactions (shiyiwu 市易務) established by Wang Anshi at Kaifeng in 1072 charged merchants who borrowed for trade and business investment an interest rate of 20 percent per annum.39 These figures provide a general picture of the credit market in the Song. In contrast, interest rates in the Qing were considerably lower: normal interest rates were 20–30 percent per annum for lending money and 20 percent for collateralized loans from pawnshops after the early Qing. Official rates were about 10 percent per annum charged to merchants, and less than 15 percent for charitable loans by government-run pawnshops in Beijing. The lowest interest rates reached prior to the twentieth century, at or below 5 percent per annum, were offered by remittance banks (piaohao 票號) in the early nineteenth century. Broadly speaking, for the period from the late Ming to the early twentieth century, there was a gradual declining trend for interest rates in China, indicating general improvement over time.40 Interest rates of 20 percent per annum or higher for common private lending were not low, of course, and have even been condemned as “usurious” when compared to those of modern banks. However, it should be noted that when risk events occur, the impact on survival is usually short-lived, so a “bridge loan” for a relatively small amount may be enough to prevent a family from starving, selling family members, or committing violence. The nature of consumption credit can explain why the interest rate was typically high for consumption loans during the Qing: the interest rate had to cover the transaction costs, which were much higher, on a per-dollar-borrowed basis, for small loans. Thus, although the absolute level of interest rates was high for emergency loans, the long-term 38 39

40

YSSF, pp. 62–3. Huang Xiangyang 黄向阳, “Guanyu Tang Song jiedai lilüde jisuan wenti” 关于唐宋借 贷利率的计算问题, Zhongguo shehui jingji shi yanjiu 中国社会经济史研究 1994.4, 33–45; Wei Tianan 魏天安, “Songdai qingmiaoqian lilü kaoshi” 宋代青苗钱利率考实, Zhongguo jingji shi yanjiu 中国经济史研究 2006.1, 155–60. Liu Qiugen 刘秋根, Ming Qing gaolidai ziben 明清高利贷资本 (Beijing, Shehui kexue wenxian chubanshe, 2000), pp. 202–7; Chen Zhiwu 陈志武, Peng Kaixiang 彭凯翔, and Yuan Weipeng 袁为鹏, “Qingchu zhi ershi shiji qianqi Zhongguo lilü shi chutan: Jiyu Zhongguo lilü shi shujuku (1660–2000) de kaocha” 清初至二十世纪前期中国利率史 初探: 基于中国利率史数据库 (1660–2000) 的考察, Qingshi yanjiu 清史研究 2016.4, 36–51.

703

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

decline in interest rates observed above should have had welfare-improving effects for ordinary people, as the falling trend meant both lower borrowing costs and possibly increased availability of funds at times of natural calamity. One important factor that contributed to the improvement of credit availability was the increasing security of private property rights in land since the Song (see Kishimoto’s chapter in this volume). Private property rights in land can be traced back to the fourth century B C E, but they were curtailed by recurring state redistributions of land and the privileges of the aristocracy in the early imperial era. However, such interventions largely disappeared after the Song. By the eighteenth century, it had become common practice for a landowner to sell a portion of his property rights or take out a mortgage on his property when needing to raise cash, which would not have been possible without security of land tenure and a functioning market for trading property rights in land.

The Role of Religious Organizations Religious organizations have played an important role in mitigating disaster risks and other adverse events for ordinary people. By advocating mutual help and non-monetary-based risk sharing, religious organizations act as friendly societies whose members help each other and as charitable organizations taking donations from members and providing relief to people in need or in distress. Bound together through common belief, members of a religious community enjoy greater mutual trust, which serves to increase the credibility and reliability of intertemporal commitments among them. In other words, religion provides “club goods” that are shared among members in the resulting social network. Some of the earliest forms of organized religion in China corresponded to this model. For example, the Way of Celestial Masters, which attracted a devoted following in western China in the second century C E, was centered around collective worship and communal social relations. They formed utopian communities that pooled resources and shared food and other subsistence necessities. After the fifth century, Buddhist institutions assumed an important social role in dispensing charity and disaster relief. Monasteries received ample donations from the Buddhist faithful, had extensive outreach, and won patronage and material support from the government. In some circumstances, the government even authorized Buddhist organizations to manage relief services. However, in the mid-ninth century, the Tang court issued a proscription against Buddhism and expropriated the property of Buddhist institutions, whose accumulation of wealth was believed to be draining the state of tax 704

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

revenue. Although the proscription of Buddhism was short-lived, from then on the government took over relief services formerly delegated to Buddhist and other social organizations. This practice continued in later dynasties.41 Although religious organizations lost their official status as enablers of social governance, they were still active in the larger society. Various Daoist and Buddhist organizations regained popularity in the late imperial era. Some of them followed the tradition of the Way of Celestial Masters and other utopian movements in developing mutual-aid infrastructure, while others fostered social identity and solidarity centered on worship of local gods, including trade- and other profession-based deities.42 In fact, it would be difficult to imagine the development of social organizations such as the lineage and mutual-aid organizations without the influence of religious institutions and practices. One important change that contributed to the revival of religious organizations was the transformation of popular culture through adopting elements of Buddhism and Daoism into Confucianism. Ideas such as the Buddhist notion of karmic retribution were widely accepted by the Confucian scholarofficial elite, who then popularized them within the larger society. This trend led to improvement in the official standing of religious activities. After its introduction in the late Ming period, Christianity also made a positive impact on Confucian scholar-officials and charitable activities, especially in the nineteenth and early twentieth centuries.

Government Welfare Chinese political mythology traced back government efforts to combat natural risk events to high antiquity. The formation of the Chinese state was attributed to the legendary sage-king Yu, who is said to have quelled the Great Flood and rendered the world habitable for humanity, and consequently founded the Xia dynasty. Such legends show how much weight Chinese society has placed on government interventions to overcome subsistence threats from famine and other calamities. Thus famine prevention and relief efforts were, not surprisingly, viewed as a key responsibility of the ruler. Risk mitigation policies such as state-operated granaries, trans-regional grain distribution networks, relief grants, and investment in preventive measures such as water control projects were enshrined in the idealized 41 42

Liang, Shishan yu jiaohua, pp. 32–4. C.K. Yang, Religion in Chinese Society: A Study of Contemporary Social Functions of Religion and Some of Their Historical Factors (Berkeley, University of California Press, 1961), pp. 28–126.

705

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

system of government outlined in the Rituals of Zhou, which became a classic authority on political economy (see Dunstan’s chapter in this volume), and were widely practiced throughout the imperial era. However, the political turmoil and frequent dynastic changes between the Han and Tang periods destabilized the fiscal and social capacities of the imperial state, leaving the larger society to depend more on the family and religious organizations for famine relief and risk sharing. After the restoration of stable central government in the Song dynasty, however, the state’s infrastructural capacity rebounded. For example, the “ever-normal granaries” (changpingcang 常平倉) designed to stabilize market prices and food supplies during periods of excess demand or surplus were established on a more regular and continuous basis. Although this system was weakened during the Yuan, it was revived and expanded under the Ming and especially the Qing regimes. The Qing government’s performance in famine relief, during the eighteenth century in particular, was outstanding when compared to previous dynasities. Apart from the state-managed granaries, Qing local officials were instructed to follow elaborate procedures to report grain prices and weather and harvest conditions to the central government on a routine basis so that government agencies could better mobilize resources and address natural disasters, both before and after risk events. All of these measures increased the state’s capacity to help offer famine relief and mitigate subsistence risks.43 Another important feature of the Qing government’s social welfare policy in the eighteenth century was the greater attention paid by the state to mobilizing social groups. This trend developed since the late Ming, when state–society relations changed as a result of the taxation reforms and grassroots governance. As official support for the lineage and other gentry-led social organizations increased, these grassroots institutions were also burdened with new duties, including providing charity and disaster relief that went beyond the interests of their members. In this process, the government pushed the local gentry to become involved in public works and famine relief. Thus they set up nongovernmental charitable granaries (yicang 義倉) and community granaries (shecang 社倉) to complement the state-run evernormal granaries. At times of natural disaster, the government adopted various strategies to raise funds, including awarding scholarly titles and official ranks in return for “contributions” for famine relief or poor relief. 43

Pierre-Étienne Will, Bureaucracy and Famine in Eighteenth-Century China (Stanford, Stanford University Press, 1990).

706

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards

One financing method, known as “entrusting capital to merchants to produce profit” (fashang shengxi 發商生息), developed extensively in the Qing, was generally used in charitable works as well. With this method, the government or local social welfare organizations invested funds with reliable merchants, and used the investment returns to pay for operational costs. In sum, by mobilizing social resources, the government increased its capacity for both social governance and helping society to mitigate risks better.

Conclusion In this chapter, our review of the available data has shown that per capita output did not experience significant growth from the Song to the Qing, but neither did it suffer any dramatic decline as claimed by some recent studies. Upon closer examination, we can actually find significant improvement in living standards over the second millennium, including ordinary people’s consumption bundles, which expanded in quality, quantity, and diversity. Furthermore, the institutionalization of the civil service examination system led to increased lineage solidity, the emergence of a growing gentry class, and a rise in the sophistication and popularity of cultural consumption and social life at the local level. Although these developments did not change the agrarian nature of Chinese society or the frequency of natural calamities, they did enhance ordinary people’s well-being. In particular, the introduction of insurance crops from the New World, the development of the lineage and other social organizations, improvements in private credit markets, the spread of religious organizations, and innovations in government disaster relief programs all provided more ways for the Chinese population to cope more effectively with risk events and live a more robust and secure life. Such improvement was especially pronounced after the Tang–Song transition. The “Malthusian trap” has been the main analytical framework used to evaluate the economic history of each and every society. Hence economic historians have been led to focus on per capita income or productivity as the only way to measure progress in living standards. However, in this chapter we have taken the view that a society’s capacity to deal with risks is almost as important as its level of per capita income. Living standards and well-being should be considered as having substantially improved so long as there has been progress in the society’s capacity to deal with risks, even if its productivity has stagnated. The progress in China since the Song that we have reviewed is not comparable to the dramatic changes brought about by the Industrial 707

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

zhiwu chen and kaixiang peng

Revolution. China’s past changes up to the late imperial period were relatively minor when compared to the transformation wrought by the Industrial Revolution in the nineteenth century. Still, it is remarkable that China was able to achieve so many innovations and maintain relative stability even against the backdrop of population growth unprecedented in human history. At least we can say that China evaded the Malthusian population trap successfully over the millennium from the late Tang to 1800. In the preIndustrial Revolution world, no other society managed to deal with the challenges of sustaining a population of several hundred million as well as China did in the late Ming and the Qing periods. In this sense, China’s development experience from the eleventh century to the eighteenth provides as rich a sample as Europe and other regions to study economics and beyond.

Further Reading Allen, Robert C., Christine Moll-Murata, Debin Ma, and Jan Luiten van Zanden, “Wage, Price and Living Standards in China 1738–1925: In Comparison with Europe, Japan, and India,” Economic History Review 64.S1 (2011), 8–38. Broadberry, Stephen, Hanhui Guan, and David Daokun Li, “China, Europe and the Great Divergence: A Study in Historical National Accounting, 980–1850,” Journal of Economic History, 78.4 (2018), 955–1000. Chao, Kang, Man and Land in Chinese History: An Economic Analysis (Stanford, Stanford University Press, 1986). Deng Yunte 鄧雲特, Zhongguo jiuhuang shi 中国救荒史 (Beijing, Shangwu chubanshe, 2011). Han Maoli 韩茂莉, Zhongguo lishi nongye dili 中国历史农业地理 (Beijing, Beijing daxue chubanshe, 2012). Huang Jingbin 黄敬斌, Minsheng yu jiaji: Qingchu zhi Minguo shiqi Jiangnan jumin de xiaofei 民生与家计: 清初至民国时期江南居民的消费 (Shanghai, Fudan daxue chubanshe, 2009). Li Bozhong 李伯重, Zhongguode zaoqi jindai jingji: 1820 niandai Huating – Louxian diqu GDP yanjiu 中国的早期近代经济: 1820年代华亭–娄县地区GDP研究 (Beijing, Zhonghua shuju, 2010). Liang Gengyao 梁庚尧, Zhongguo shehui shi 中国社会史 (Beijing, Dongfang chuban zhongxin, 2016). Liang Jiamian 梁家勉, Zhongguo nongye kexue jishu shigao 中国农业科学技术史稿 (Beijing, Nongye chubanshe, 1989). Liu, William Guanglin, The Chinese Market Economy, 1000–1500 (Albany, SUNY Press, 2015). McDermott, Joseph P., The Making of a New Rural Order in South China, vol. 1, Village, Land, and Lineage in Huizhou, 900–1800 (Cambridge, Cambridge University Press, 2013). Maddison, Angus, Chinese Economic Performance in the Long Run, 960–2030 A D, 2nd ed., revised and updated (Paris, OECD Publishing, 2007).

708

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Production, Consumption, and Living Standards Peng Kaixiang 彭凯翔, “Renkou zengzhangxiade liangshi shengchan yu jingji fazhan: You Shi Zhihong yanjiuyuan de Qingdai nongye chanchu cesuan tanqi” 人口增长下的粮 食生产与经济发展: 由史志宏研究员的清代农业产出测算谈起, Zhongguo jingjishi yanjiu 中国经济史研究 2015.7, 38–49. Perkins, Dwight H., Agricultural Development in China, 1368–1968 (Chicago, Aldine Publishing, 1969). Shi Zhihong 史志宏, Qingdai nongye de fazhan he bufazhan (1661–1911 nian) 清代农业的发 展和不发展 (1661–1911年) (Beijing, Shehui kexue wenxian chubanshe, 2017). Smith, Paul Jakov, and Richard von Glahn (eds.), The Song–Yuan–Ming Transition in Chinese History (Cambridge, MA, Harvard University Asia Center, 2003). Sun Ji 孙机, Zhongguo gudai wuzhi wenhua 中国古代物质文化 (Beijing, Zhonghua shuju, 2014). Will, Pierre-Étienne, Bureaucracy and Famine in Eighteenth-Century China (Stanford, Stanford University Press, 1990). Wu Renshu 巫仁恕, Pinwei shehua: Wan Ming de xiaofei shehui yu shidafu 品味奢华: 晚明 的消费社会与士大夫 (Beijing, Zhonghua shuju, 2008). Yang, C.K., Religion in Chinese Society (Berkeley, University of California Press, 1961). Zheng, Zhenman, Family Lineage Organization and Social Change in Ming and Qing Fujian (Honolulu, University of Hawai‘i Press, 2001). Zhou Rong 周荣, Ming Qing shehui baozhang zhidu yu Liang Hu jiceng shehui 明清社会保 障制度与两湖基层社会 (Wuchang, Wuhan daxue chubanshe, 2006).

709

https://doi.org/10.1017/9781108587334.020 Published online by Cambridge University Press

Bibliography of Primary Works Cited

Abbreviations for Collectanea SBCK: Sibu congkan 四部叢刊 (Shanghai, Shangwu yinshuguan, 1929). SKQS: Wenyuange Siku quanshu 文淵閣四庫全書 (Taipei, Taiwan shangwu yinshuguan, 1983). SSJZS: Ruan Yuan 阮元 (comp.), Shisanjing zhushu 十三經注疏, rpt (Beijing, Zhonghua shuju, 1980). XX SKQS: Xuxiu Siku quanshu 續修四庫全書 (Shanghai, Shanghai guji chubanshe, 1995).

Titles BMTZ: Bamin tongzhi 八閩通志 (1491). BS: Bei shi 北史 (Zhonghua shuju ed.). BTMHL: Wang Huizu 汪輝祖, Bingta menghen lu 病榻夢痕錄, in Wang Longzhuang yishu 汪龍莊遺書, rpt (Taipei, Huawen shuju, 1970). BXDA: Sichuansheng dang’anguan 四川省檔案館 and Sichuan daxue lishixi 四川大學歷 史系 (eds.), Qingdai Qian Jia Dao Baxian dang’an xuanbian 清代乾嘉道巴縣檔案選編 (Chengdu, Sichuan daxue chubanshe, 1989–1996). CAZ: Song Minqiu 宋敏求, Chang’an zhi 長安志, rpt in Song Yuan difangzhi congshu 宋元 地方志叢書 (Taipei, Dahua shuju, 1987). CGL: Tao Zongyi 陶宗儀, Nancun chuogeng lu 南村輟耕錄 (Beijing, Zhonghua shuju, 1959). CHTB: Zheng Ruozeng 鄭若曾, Chouhai tubian 籌海圖編 (SKQS ed.). CZXZ: Changzhou xianzhi 長洲縣志 (1684). DJMHL: Meng Yuanlao 孟元老, Dongjing menghua lu 東京夢華錄, in Zhonghua shuju Shanghai bianjisuo 中華書局上海編輯所 (ed.), Dongjing menghua lu (wai sizhong) 東京 夢華錄 (外四種) (Beijing, Zhonghua shuju, 1962). DMHD: Xu Pu 徐溥 et al. (comps), Da Ming huidian 大明會典 (SKQS ed.). DML: Huai Xiaofeng 懷效鋒 (ed.), Da Ming lü 大明律 (Shenyang, Liao Shen shushe, 1990). DPQJ: Su Shi 蘇軾, Dongpo qiji 東波七集 (Sibu beiyao 四部備要 ed.). DQLL: Zhang Rongzheng 張榮錚, Liu Yongqiang 劉勇強, and Jin Maochu 金懋初 (eds.), Da Qing lüli 大清律例 (Tianjin, Tianjin guji chubanshe, 1993). FBJ: Fang Bao 方苞, Fang Bao ji 方苞集 (Shanghai, Shanghai guji chubanshe, 1983). FLXZ: Fuliang xianzhi 浮梁縣志 (1823).

710

https://doi.org/10.1017/9781108587334.021 Published online by Cambridge University Press

Bibliography of Primary Works Cited FLZTZ: Jianghu zuizhong langsou 江湖醉中浪叟 (pseud.), Falin zhaotianzhe 法林照天 燭, undated Ming dynasty MS, Fu Sinian Library, Institute of History and Philology, Academia Sinica, Taipei. FSJ: Fan Chengda 范成大, Fan shihu ji 范石湖集 (Shanghai, Shanghai guji chubanshe, 2006). FWZJ: Fan Zhongyan 范忠淹, Fan Wenzheng ji 范文正集 (SBCK ed.). GMXZ: Gaoming xianzhi 高明縣志 (1894). GQ: Tan Qian 談遷, Guoque 國榷 (SKQS ed.). GSZ: Huijiao 慧皎, Gaoseng zhuan 高僧傳 (ed. Tang Yongtong 湯用彤) (Beijing, Zhonghua shuju, 1992). GY: Guoyu jijie 國語集解 (ed. Xu Yuangao 徐元誥), rev. ed. (Beijing, Zhonghua shuju, 2002). GZ: Guan Zi jiaozhu 管子校注 (ed. Li Xiangfeng 黎翔風) (Beijing, Zhonghua shuju, 2004). GZD: YZ: Gongzhongdang Yongzheng chao zouzhe 宮中檔雍正朝奏摺 (Taipei, Guoli gugong bowuyuan, 1977). HBJYJ: Hong Yongqin 洪用懃 (comp.), Hong Beijiang xiansheng yiji 洪北江先生遺集 (1877), rpt (Taipei, Huawen shuju, 1969). HCJSWB: He Changling 賀長齡 and Wei Yuan 魏源 (comps), Huangchao jingshi wenbian 皇朝經世文編 (1827), rpt (Taipei: Wenhai chubanshe, 1966). HCLQJ: Han Yu 韓愈, Han Changli quanji 韩昌黎全集 (Beijing, Zhongguo shudian, 1991). HCQJ: Liu Kezhuang 劉克莊, Houcun xiansheng da quanji 後村先生大全集 (SBCK ed.). HFSY: Zhang Aisheng 張靄生 (comp.), Hefang shuyan 河防述言 (Siku quanshu zhenben ed.). HFZ: Han Fei Zi jishi 韓非子集釋 (ed. Chen Qiyou 陳奇猷) (Taipei, Hanjing wenhua shiye gongsi, 1983). HGZS: Huguang tujing zhishu 湖廣圖經志書 (1522). HMJSWB: Chen Zilong 陳子龍 et al. (comps), Huang Ming jingshi wenbian 皇明經世文編, rpt (Beijing, Zhonghua shuju, 1962). HNZ: Huainan honglie jishi 淮南鴻烈集釋 (ed. Liu Wendian 劉文典) (Beijing, Zhonghua shuju, 1989). HS: Ban Gu 班固, Han shu 漢書 (Zhonghua shuju ed.). HZFZ: Hangzhou fuzhi 杭州府志 (1475). JGLBS: Gu Yanwu 顧炎武, Tianxia junguo libing shu 天下郡國利病書, rpt (Kyoto, Chu¯bun shuppan, 1975). JHHMS: Dong Wei 董煟, Jiuhuang huomin shu 救荒活民書 (SKQS ed.). JHNP: Duan Guangqing 段光淸, Jinghu zizhuan nianpu 鏡湖自撰年譜 (Beijing, Xinhua shudian, 1960). JTS: Jiu Tang shu 舊唐書 (Zhonghua shuju ed.). JWDS: Jiu Wudai shi 舊五代史 (Zhonghua shuju ed.). JWGZ: Jin Wenxiang Gong zoushu 靳文襄公奏疏 (comp. Jin Zhiyu 靳治豫), rpt (Taipei, Wenhai chubanshe, 1966–1973). KSXM: Li Jinde 李晉德, Keshang yilan xingmi 客商一覽醒迷 (Taiyuan, Shanxi renmin chubanshe, 1992). LCWJ: Wang Anshi 王安石, Linchuan xiansheng wenji 臨川先生文集 (Hong Kong, Zhonghua shuju, 1971). LJ: Li ji 禮記 (SSJZS ed.). LMBJ: Su Maoxiang 蘇茂相, Lüli linmin baojing 律例臨民寶鏡 (1632).

711

https://doi.org/10.1017/9781108587334.021 Published online by Cambridge University Press

bibliography of primary works cited LSCQ: Lüshi chunqiu jiaoshi 呂氏春秋校釋 (ed. Chen Qiyou 陳奇猷) (Shanghai, Xuelin chubanshe, 1994). LXGJ: Lu Zhi 陸贄, Lu Xuangong ji 陸宣公集 (Hangzhou, Zhejiang guji chubanshe, 1988). MBS: Wang Shimao 王世懋, Minbu shu 閩部疏 (Taipei, Chengwen chubanshe, 1975). MHT: Fang Guancheng 方觀承, Mianhua tu 棉華圖 (Beijing, Airusheng shuzihua jishu yanjiu zhongxin, 2009). MLL: Wu Zimu 吳自牧, Mengliang lu 夢粱錄, in Dongjing menghua lu (wai sizhong) 東京夢 華錄 (外四種) (Beijing, Zhonghua shuju, 1962). MS: Ming shi 明史 (Zhonghua shuju ed.). MSL: Ming shilu 明實錄 (Taipei, Zhongyang yanjiuyuan, 1962–1966). MSZ: Yan Junyan 顏俊彥, Mengshuizhai cundu 盟水齋存牘 (Beijing, Zhongguo zhengfa daxue falü guji zhengli yanjiusuo, 2002). MZ: Mo Zi jiaozhu 墨子校注 (ed. Wu Yujiang 吳毓江) (Beijing, Zhonghua shuju, 1993). NCFZ: Nanchang fuzhi 南昌府志 (1588). NGZML: Wu Zeng 吳曾, Nenggaizhai manlu 能改齋漫錄 (Shanghai, Shanghai guji chubanshe, 1979). NZQS: Xu Guangqi 徐光啟, Nongzheng quanshu jiaozhu 農政全書校注 (Taipei: Mingwen shuju, 1981). QFL: Qianfulun jian jiaozheng 潛夫論箋校正 (ed. Peng Duo 彭鐸) (Beijing, Zhonghua shuju, 1985). QHW: Quan Qian Han wen 全前漢文, in Yan Kejun 嚴可均 (ed.), Quan shanggu sandai Qin Han sanguo liuchao wen 全上古三代秦漢三國六朝文 (Guangzhou, Guangya shuju, 1887). QMJ: Zhongguo shehui kexueyuan lishi yanjiusuo 中國社會科學院歷史研究所 and Song Liao Jin Yuan shi yanjiushi 宋遼金元史研究室 (comp.), Minggong shupan qingmingji 名公書判清明集 (Beijing, Zhonghua shuju, 1987). QMYS: Jia Sixie 賈思勰, Qimin yaoshu jinshi 齊民要述今釋 (ed. Shi Shenghan 石聲漢) (Beijing, Kexue chubanshe, 1957–1958). QSL: Qing shilu 清實錄, rpt (Taipei, Huawen shuju, 1970). QSW: Quan Song wen 全宋文 (Shanghai, Shanghai cishu chubanshe, 2006). QSZY: Wei Zheng 魏徵 et al. (comp.), Qunshu zhiyao 群書治要 (SBCK ed.). QTWZ: Xie Jie 謝傑, Qiantai wo zuan 虔臺倭纂 (Xuanlantang congshu xuji ed.). QXJ: Wang Mai 王邁, Quxuan ji 臞軒集 (SKQS ed.). QYTF: Xie Shenfu 謝深甫 et al. (comp.), Qingyuan tiaofa shilei 慶元條法事類 (Tokyo: Koten kenkyu¯kai, 1968). QZ: Hong Zun 洪遵, Quanzhi 泉志, rpt in Zhongguo qianbi wenxian congshu 中國錢幣文 獻叢書 (Shanghai, Shanghai guji chubanshe, 1994). SC: Bao Shichen 包世臣, Shuo chu 說儲, rpt in Guocui congshu 國粹叢書, series 1 (Shanghai, Guoxue baocun hui, 1906). SCTG: Wei Jirui 魏際瑞, Sicitang gao 四此堂稿 (1675). SFG: Xu Yikui 徐一夔, Shifeng gao 始豐藁, rpt in Wulin wangzhe yizhu 武林往哲遺著 (Hangzhou, Dingshi Jiahuitang, 1894). SGBL: Wu Zhongfu 吳仲孚, Shanggu bianlan 商賈便覽 (1792). SGZ: Chen Shou 陳壽, Sanguo zhi 三國志 (Zhonghua shuju ed.). ShJ: Shijing 詩經 (SSJZS ed.). SHY: Xu Song 徐松 (comp.), Song huiyao jigao 宋會要輯稿, rpt (Beijing, Zhonghua shuju, 1957).

712

https://doi.org/10.1017/9781108587334.021 Published online by Cambridge University Press

Bibliography of Primary Works Cited SJ: Sima Qian 司馬遷, Shi ji 史記 (Zhonghua shuju ed.). SJS: Shangjun shu zhuizhi 商君書錐指 (ed. Jiang Lihong 蔣禮鴻) (Beijing, Zhonghua shuju, 1986). SLQH: Ye Dehui 葉德輝, Shulin qinghua 書林清話, rpt (Shanghai, Shanghai shudian, 1989). SMCJJ: Sima Guang 司馬光, Sima Wenzheng Gong chuanjia ji 司馬文正公傳家集 (SKQS ed.). SMGRJ: Sima Guang 司馬光, Sima Guang riji jiaozhu 司馬光日記校注 (ed. Li Yumin 李 裕民) (Beijing, Zhongguo shehui kexue chubanshe, 1994). SMXZ: Zhizheng Siming xuzhi 至正四明續志, rpt in Song Yuan difangzhi congshu 宋元地方 志叢書 (Taipei, Dahua shuju, 1987). SQXY: Yu Bian 俞弁, Shanqiao xiayu 山樵暇語 (Hanfenlou miji 涵芬樓秘笈 ed.). SHRJ: Lu Chongxing 盧崇興, Shou He riji 守禾日記 (1678). ShS: Shangshu 尚書 (SSJZS ed.). SS: Song shi 宋史 (Zhonghua shuju ed.). SSBF: Yan Ruyi 嚴如熤, Sansheng bianfang beilan 三省邊防備覽 (XX SKQS ed.). SSJW: Sima Guang 司馬光, Sushui jiwen 速水記聞 (Beijing, Zhonghua shuju, 1989). SuiS: Sui shu 隋書 (Beijing, Zhonghua shuju ed.). SYZJ: Lu Rong 陸容, Shuyuan zaji 菽園雜記 (Beijing: Zhonghua shuju, 1983). SXT: Dou Yi 竇儀 et al. (comp.), Song xing tong 宋刑統 (Beijing: Falü chubanshe, 1999). TLML: anon., Tiaoli mulu 條例目錄 (1734–1794; collection of To¯yo¯ Bunko). TLSY: Zhangsun Wuji 長孫無忌 et al. (comp.), Tanglü shuyi jianjie 唐律疏議箋解 (ed. Liu Junwen 劉俊文) (Beijing, Zhonghua shuju, 1996). TSL: Tianyige bowuguan 天一閣博物館 and Zhongguo shehui kexueyuan lishi yanjiusuo Tiansheng ling zhengli ketizu 中國社會科學院歷史研究所天聖令整理 課題組 (eds.), Tianyige cang Ming chaoben Tiansheng ling jiaozheng 天一閣藏明鈔本 天聖令校證 (Beijing, Zhonghua shuju, 2006). TZQJ: Tao Zhu 陶澍, Tao Zhu quanji 陶澍全集 (Changsha, Yuelu shushe, 2010). TZTG: Bozhulu Chong 孛朮魯翀 (comp.), Tongzhi tiaoge jiaozhu 通制條格校注 (ed. Li Linggui 李齡貴) (Beijing, Zhonghua shuju, 2011). WFL: Huang Xingzeng 黃省曾, Wufeng lu 吳風錄, rpt (Taipei, Taiwan wenyuan shuju, 1964). WLZYS: Wang Huizu 汪輝祖, Wang Longzhuang yishu 汪龍莊遺書, rpt (Taipei, Huawen shuju, 1970). WS: Wei shu 魏書 (Zhonghua shuju ed.). WYYH: Li Fang 李昉 et al. (comps), Wenyuan yinghua 文苑英華 (Beijing, Zhonghua shuju, 1966). WZ: Li Dingsheng 李定生 and Xu Huijun 許惠君 (eds.), Wen Zi jiaoshi 文子校釋 (Shanghai, Shanghai guji chubanshe, 2016). WZRJ: Zhan Yuanxiang 詹元相, Weizhai riji 畏齋日記, in Zhongguo shehui kexueyuan lishi yanjiusuo Qingshi yanjiushi 中国社会科学院历史研究所清史研究室 (ed.), Qingshi ziliao 清史資料, vol. 4 (Beijing, Zhonghua shuju, 1983). XC: Zhang Kentang 張肯堂, Xunci 㽦辭 (c. 1628–1644). XCB: Li Tao 李燾, Xu Zizhi tongjian changbian 續資治通鑑長編 (Beijing, Zhonghua shuju, 2004). XCSL: Fei Xin 費信, Xingcha shenglan jiaozhu 星槎勝覽校注 (ed. Feng Chengjun 馮承鈞), (Beijing, Zhonghua shuju, 1954). XCYS: anon., Xianchun yishi 咸淳遺事 (SKQS ed.). XGJK: Fang Hui 方回, Xu gujin kao 續古今考 (SKQS ed.).

713

https://doi.org/10.1017/9781108587334.021 Published online by Cambridge University Press

bibliography of primary works cited XJSXL: Huang Ang 黃卬, Xijin shixiao lu 錫金識小錄 (1752). XNYL: Li Xinchuan 李心傳, Jianyan yilai xinian yaolu 建炎以來繫年要錄 (SKQS ed.). XS: Jia Yi 賈誼, Xinshu jiaozhu 新書校注 (ed. Yan Zhenyi 閻振益 and Zhong Xia 鐘夏), (Beijing, Zhonghua shuju, 2000). XWXTK: Cao Renhu 曹仁虎, Qinding xu wenxian tongkao 欽定續文獻通考 (SKQS ed.). XWXTK (Wang): Wang Qi 王圻 (comp.), Xu Wenxian tongkao 續文獻通考, rpt (Hangzhou, Zhejiang guji chubanshe, 1988). XZ: Xun Zi jishi 荀子集釋 (ed. Li Disheng 李滌生) (Taipei, Xuesheng shuju, 1979). XZWJ: Ouyang Shoudao 歐陽守道, Xunzhai wenji 巽齋文集 (SKQS ed.). YCXP: Zhu Guozhen 朱國禎, Yongchuang xiaopin 湧幢小品 (Beijing, Wenhua yishu chubanshe, 1998). YCXJ: Yancheng xianji 郾城縣記 (1934). YDZ: Da Yuan shengzheng guochao dianzhang 大元聖政国朝典章 (Taipei, Guoli gugong bowuyuan, 1972). YFLG: Zeng Gong 曾鞏, Nanfeng xiansheng Yuanfeng leigao 南豐先生元豐類藁 (SBCK ed.). YLMC: Zhao Yanwei 趙彥衛, Yunlu manchao 雲麓漫鈔, rpt (Beijing, Zhonghua shuju, 1996). YS: Yuanshi 元史 (Zhonghua shuju ed.). YSB: Ye Mengzhu 葉夢珠, Yueshi bian 閱世編 (Shanghai, Shanghai guji chubanshe, 1981). YSSF: Yuan Cai 袁采, Yuanshi shifan 袁氏世范 (Shanghai, Shangwu yinshuguan, 1937). YTL: Huan Kuan 桓寬, Yantie lun jiaozhu 鹽鐵論校注 (ed. Wang Liqi 王利器) (Beijing, Zhonghua shuju, 1992). YYXQ: Zhang Lüxiang 張履祥, Yangyuan xiansheng quanji 楊園先生全集 (Beijing, Zhonghua shuju, 2002). YZHFL: Li Dou 李斗, Yangzhou huafang lu 揚州畫舫錄 (Yangzhou, Jiangsu Guangling guji keyinshe, 1984). YZYJL: Chen Dezhi 陳得芝, Qiu Shulin 邱樹林, and He Zhaoji 何兆吉 (eds.), Yuandai zouyi jilu 元代奏議集錄 (Hangzhou, Zhejiang guji chubanshe, 1998). YZZP: Zhongguo diyi lishi dang’anguan 中國第一歷史檔案館 (ed.), Yongzheng hanwen zhupi zouzhe huibian 雍正漢文硃批奏摺匯編 (Nanjing, Nanjing guji shudian, 1989– 1991). ZAZZ: Zheng’an zhouzhi 正安州志 (1877). ZGZH: Zhanguo zonghengjia shu 戰國縱横家書 (ed. Mawangdui Hanmu boshu zhengli xiaozu 馬王堆漢墓帛書整理小組) (Beijing, Wenwu chubanshe, 1976). ZHS: Chen Chongdi 陳崇砥, Zhihuang shu 治蝗書 (1874). ZPBZ: Zhapu beizhi 乍浦偹志 (1828). ZY: Zhouyi zhijie 周易直解 (SSJZS ed.). ZZ: Zuozhuan 左傳 (SSJZS ed.). ZZTJ: Sima Guang 司馬光, Zizhi tongjian 資治通鑒 (Beijing, Guji chubanshe, 1956). ZZZJ: Kong Qi 孔齊, Zhizheng zhiji 至正直記 (Shanghai, Shanghai guji chubanshe, 1987).

714

https://doi.org/10.1017/9781108587334.021 Published online by Cambridge University Press

Index

Page numbers in bold refer to content in tables; page numbers in italics refer to content in figures and maps. Aba hoard, Sichuan, 564 Abbasid caliphate, 160, 237, 238–9 Acapulco, Mexico, 661 Adachi Keiji, 462 administrative cities. See commercialized administrative cities agency partnerships, 629 Agricultural and Commercial Statistics, 330, 332, 333 Agricultural Manual, 270 agriculture, 52, 259–60 ancient China, 16–17 changes in the post-1000 C E period, 274–89 development of tools and techniques, 58–60 environmental changes of early China, 52–7, 53 estimates of food output, 681–5, 682, 683 Han dynasty developments, 17, 73–80 insurance crops, 699–700 Jiangnan type of farming, 457 later Han and Period of Disunion, 80–7, 88 limitations of irrigation, 60–3 new crops, 488–96 nonagricultural production, 63–5 Qin period, 70–3, 109 Shang to Spring and Autumn period, 65–8 soil fertility, 54, 59–60, 267–72, 269, 276, 502–5 Sui–Tang period, 87–9 Warring States period, 68–73, 172–3 water control, 272–4, 488, 496–502, 505–8 Ahmad (Persian merchant), 359 Al-Bukhari, 238 alkalization, 61–3, 76

Al-Khwarismi, 238 Allen, Robert, 479, 511, 516, 687, 696 Allsen, Thomas, 648 Al-Rahman, Abd, 358 Altan Khan, 367 Altar of Land and Grain, 271 Amitabha Sutra, 218 Amu–Syr plain, 208 An Jia 安伽, 227, 228 An Lushan 安祿山 Rebellion, 128, 139, 142, 148, 149, 160, 198, 234, 246, 247, 250, 312, 340 An Xuan 安玄, 231 Analects, 59 ancestral halls, 624, 700–1 Anhui province, 283–5 animal husbandry, 64, 71, 79, 487–8 Annam, 655 Anyang, Henan, 64, 65 archaeological perspectives. See late preimperial China archer lords, 93 artworks, 550, 551 Authorized Imperial Compendium on Agricultural Practice, 270 autocratic state, 96–102, 244 Ba tribes, 46–8 Bactria, 216, 238 Bai Canal, 62 Bai Gui 白圭, 151 Bai Juyi 白居易, 200 Baidian foundry, Shanxi, 25 balanced-standard system, 115, 189 balanced-supply regulations (junshufa), 345

715

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Cao Cao 曹操, 120, 153 Cao Pi 曹丕, 154 Caowangzui, Hubei, 42 capital citizenry, 95 capital markets, 455–6 capitation tax, 111, 112, 119, 122 cash vouchers (qianyin), 348, 349 Catalogue of Chinese Books Currently Available in Japan, The, 159 Cefu yuangui, 237 census records, 98, 306, 310, 315, 357, 371 Central Asian steppe, 203, 205, 211, 229 Central Plain, 100, 101, 108, 127, 132, 136, 160, 261, 311, 357 ceramics production, 28–31, 39, 42, 540–2, 644 Ch’en Huan-chang 陳煥章, 2 Champa rice, 261, 284, 488–9 Chang Deshou 常德壽, 281–2 Chang’an, Shaanxi, 74–5, 84, 88, 144, 147–8, 149, 158, 161, 198, 218, 251 Changzhou, Jiangsu, 328, 554 Chao Cuo 晁錯, 184, 185 Chao Kang 趙岡, 472, 531 charitable organizations, 702 Chen Chao 陳朝, 443 Chen Chengwu 陳誠吾, 432 Chen Chongdi 陳崇砥, 296 Chen Dashou 陳大受, 285 Chen Fu 陳旉, 270 Chen Hansheng 陳翰笙, 5 Chen Huizu 陳輝祖, 285 Chen Qiulu 陳求魯, 644 Chen Shunyu 陳舜俞, 393 Chen Taiji 陳太極, 432 Chen Zhiwu 陳志武, 513 Cheng Han (state), 137 Cheng Lin 程琳, 402 Cheng Minsheng 程民生, 686 Cheng Yi 程頤, 701 Cheng, King of Zhou 周成王, 66 Chengdu, Sichuan, 46, 154, 312, 333 Chengzhou, Henan, 95 China Biographical Database (CBDB), 696 Chinggis Khan, 267 Chongqing, 583 Chongzhen, Emperor of Ming 明崇禎, 475 Christianity, 705 Chronicle of the Painted Barques of Yangzhou, 546 Chronicles of Zuo, 67, 143, 150, 168 Chu (state), 19–20, 43, 46–8, 97, 102, 132, 139, 168, 248 Chu Gong Ni-yongzhong inscription, 41 Ci county, Hebei, 224

Balazs, Étienne, 5, 523 Balkh, Bactria, 238 Ban Gu 班固, 195–6, 306 banliang coin, 108, 113, 132–5 Bao Shichen 包世臣, 385, 397, 401, 578 baochao notes, 362, 363, 575 Baoding, Hebei, 323 Barmakid family, 238 Baxian archives, 432, 441 beancake fertilizer, 504–5 Begram hoard, 209 Beijing, 320, 321, 357, 507, 534–6, 615, 686, 687, 695 Bi Yuan 畢沅, 479 bianshen surveys, 306 bills of exchange (jiaozi), 344, 347, 572 Biographies of Immortals, 143 Biographies of the Moneymakers, 107, 151, 195 Black Death, 314 bone tools, 58 Book of Changes, The, 184 Book of Documents, The, 151, 184 Book of Fan Shengzhi, The, 74 Book of Lord Shang, The, 71, 99, 106, 615–17 Book of Odes, The, 56, 63, 66, 143 Book of Rites, The, 388, 392, 395, 397, 399–401, 697 book production, 34, 615–17, 694 booms, market, 472–9, 473 Braudel, Fernand, 37 breweries, 65 British East India Company, 581, 668, 669 Broadberry, Stephen, 678 Brokaw, Cynthia, 616 brokers, 146, 620–2 regulations, 441–4 Bronze Age. See late pre-imperial China bronze production, 26–7, 32–3, 94 Bu Shi 卜式, 189, 191 bubonic plague, 314 Buck, J.L., 518 Buddhism, 139, 204, 238, 646, 704–5 Silk Road trade, 216–21, 236 Sogdian followers, 231 burning, agricultural, 280 Byzantine Empire, 224 cadastral surveys, 350, 355, 365, 371 Cai Hunqua 蔡??, 669 Cai Jing 蔡京, 345, 348 calendering workshop, 466–7 Cambrian stratum, 52 Cao Canal, 74

716

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Confucius, 59, 167, 173, 181 conservatism, 392–402, 408 Conspectus of Historical Geography, 328 consumption, 677, 690–7 Continuation of the Han History, 307 contract law brokerage regulations, 441–4 debt and default, 433–8 fraud in trade, 438–41 land sales, 451 Ming and Qing partnership disputes, 431–3 Song and Yuan dynasties, 429–31 contributed funds, 373, 376 convicts, 105 copper, 37–44, 208, 251, 562–7, 564, 667 cosmic generativity. See wealth cottage industries, 457 cotton, 489, 491–4, 607, 614–15, 692–3 cowry shells, 131 cowry-shaped bronze pellets, 132 credit associations, 622, 624 Cui Shi 崔寔, 80, 151, 153 cultivation. See agriculture Cultural Revolution, 8 cumulic soils, 54 Cunninghamia (Chinese fir), 275, 275, 283–4 currency, 560–2 banliang coin in the Qin–Han periods, 132–5 copper coinage, 208, 562–7, 564 Five Dynasties period, 248, 249 Han dynasty, 113–14, 116, 120, 187–9 imaginary silver tael, 580–4, 582 late pre-imperial China, 20, 100, 101, 131–2, 133 local markets and money, 584–91, 585, 587, 588, 589, 590 Ming dynasty, 362, 363, 566, 569, 570, 575–6, 658 money in tax collection and land sales, 567–71 multiple currencies and their relationships, 139–43, 141 official versus non-official, 571–80 Period of Disunion, 122, 124 private coinage and copper shortages, 135–9, 137 Qin dynasty, 108, 132–4 Qing dynasty, 566, 569, 571, 578–80, 579 Song dynasty, 344, 345, 347, 349–50, 351–4, 353, 563, 566, 568, 570, 571–4, 645 Tang dynasty, 161, 577, 581 varieties of monetary economies, 591–4 wealth concepts, 386–8

Cishan site, Hebei, 58 city economies, 522–5, 525 commercialized administrative cities, 543–52, 551 comparative perspective, 555–7 Hangzhou in the Southern Song, 530–3 Kaifeng in the Northern Song, 526–9, 528 manufacturing cities, 537, 540–3 market towns, 552–5 Nanjing and Beijing, 533–6 port cities, 537–40 Classic of Tea, 89, 158 climatic changes, 52, 55, 56–7, 72, 81–2, 87, 88, 261, 263–7, 265, 266, 485–6 cloth payments, 126, 140–1 clothing, 233, 236 Coase, Ronald, 626 Cohong merchants, 668 coins, decorative, 224 Collection of Jottings on Wu [Suzhou], 549 command economy, 102–10, 115 commenda partnerships, 627–8 commerce, post-1000 C E, 597–600. See also international trade long-distance commodities distribution, 611–17, 613 market growth and product distribution, 600–11, 601, 604, 606, 608, 610 merchant problems in the late-imperial period, 617–30 merchants and late-imperial government, 630–5 commerce, pre-1000 C E. See also international trade; markets; merchants; Silk Road trade late pre-imperial China, 18–20 mercantilist fiscal state, 114–18 Qin dynasty, 110 Warring States period, 171–2, 173–5 commercial law. See law commercial partnerships, 627–30 commercial taxation, 152–3 commercialized administrative cities, 522, 543–52 commission agreements, 430 Communist Party, 8 Complete Works on Agricultural Administration, 270 Comprehensive Institutions of the Great Yuan, 436, 442 Confucian Classics, 361, 652, 700 Confucianism, 2, 179, 195, 208, 254, 381, 383, 389, 395, 427–8, 701, 705

717

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index E Jun Qi tallies, 19–20, 32 ecological change. See agriculture; environmental changes economic philosophy, 166–7 fiscal state in the Warring States period, 167–81 Han debates and mercantilism, 181–92 post-Han fiscal regimes, 197–201 triumph of the farming habitus, 192–6 Economic Principles of Confucius and His School, The, 2 Edkins, Joseph, 584 El Niño/La Niña (ENSO), 265 Elvin, Mark, 9, 245 England, 516, 562, 565 Ennin 圓仁 (Buddhist priest), 149, 158, 159 environmental changes, 52, 259–60. See also agriculture; climatic changes disturbances/disasters, 289–97, 292, 293, 294 early China, 52–7, 53 post-1000 C E patterns, 260–2 soil, 54, 59–60, 267–72, 269, 276, 502–5 water, 272–4, 488, 496–502, 505–8 equal-field system, 83–4, 87, 123, 124, 125–7, 128, 197–8, 246 equitable-delivery system, 115, 189 Erligang period (c. 1500–1300 B C E ), 37 Essential Techniques for the Common People, 60, 81, 84–7 Europe, 555–6, 668–70 ever-normal method, 116, 409 Ewangcheng, Hubei, 42 Exploitation of the Works of Nature, The, 277

currency (cont.) Xin dynasty, 118 Yuan dynasty, 359–60, 566, 568, 570, 574–5, 581 customs excises, 100 Da Qin, 221, 223 daitian (alternate-fields) system, 59, 74, 76–7 Daizong, Emperor of Tang 唐代宗, 235 Daliuzhen, Hebei, 582, 585 Dalupu, Hubei, 39–40 Dangchang people, 211 Dao Learning tradition, 701 Daowu, Emperor of Northern Wei 魏道武 帝, 82 Daquan wushi coin, 135 Dawan (oasis state), 209 Dayan Pagoda, 236 Dayangzhou site, Jiangxi, 58 Daye county, Hubei, 40–3 debt laws, 433–8 deflation, 386 demographics. See population changes Deng Ai 鄧艾, 81 Deng Tong 鄧通, 113 Deng, Kent, 514 Dengzhou district, Shandong, 308 depressions, 472–9, 473 Di people, 84, 100 diao levies, 120, 121, 141 didang transactions, 253 Dietrich, Craig, 493 ding levy, 306, 371 Dinghai, Zhejiang, 577 disasters, environmental, 289–97, 292, 293, 294 disturbances, environmental, 289–97, 292, 293, 294 Diwu Qi 第五琦, 160 domain economy, 94–6 Dong Wei 董煟, 409, 411 Dong Zhongshu 董仲舒, 76 Donghai commandery, 78 Dongting commandery, 102 Doufucun workshop, Shaanxi, 28–31 Dreams of the Splendor of the Eastern Capital, 529 drought, 291–7, 292, 293, 294, 700 dry-land cultivation, 59–60, 72 Du Huan 杜環, 238 Du You 杜佑, 238 Du Yuanzhen 杜元珍, 433 Dunhuang, Gansu, 123, 226 Dutch East India Company, 578, 628, 662 Dutch traders, 662–3

factor markets, 448–9 asset management by landlords, 461–4 booms and depressions, 472–9, 473 capital markets, 455–6 famine periods, 479–81 labor markets, 453–4, 462–4 land markets, 450–3 rural households, 457–60 secular trends, 468–72 urban and hinterland workshops, 464–8 Fairbank, John K., 300 fairs, 146–7 fallowing, 281 family businesses, 625–7 family households, 70, 83, 110, 170, 252–4, 457–60 family temples, 700–1 famine, 479–81, 698

718

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index relief measures, 371, 410, 413–15, 480, 705, 706 Fan Chengda 範成大, 280 Fan Shengzhi 氾勝之, 74, 76 Fan Shuzhi 樊树志, 328 Fan Zhongyan 范仲淹, 393, 480 Fang Bao 方苞, 412 Fang Guancheng 方觀承, 492 Fang Shaowu 方少吾, 443 Fang Xing 方行, 600 Fangmatan site, Gansu, 73 farm sizes, 69 farmer-soldier colonies, 207 farming habitus, 192–6 Faxian 法顯 (Buddhist pilgrim), 219–20 Fei Xiaotong 費孝通, 5 Feng Jingya 馮敬涯, 436 Feng Shunyu 馮順宇, 431 Feng Tinghui 馮廷惠, 433 Feng Xizhi 馮禧之, 436 Feng, Empress of Northern Wei 魏馮皇后, 82 Fenghuo coin, 137 fertilizer, 502–5 feudalism, 3, 6, 7 Feuerwerker, Albert, 8, 9 First Emperor of Qin (Qin Shihuangdi 秦始 皇帝; see also Zheng, King of Qin), 31, 72, 84, 110, 114, 189, 205 fiscal state, pre-1000 C E, Han dynasty, 195–6 post-Han regimes, 197–201 Warring States period, 167–81 fish, 64, 276–8 fish-tail registers, 350 Five Barbarians, 204 Five Dynasties period, 248–50, 249, 450, 500, 526 floods, 291–7, 292, 293, 294, 506–8 Food and Money Semi-monthly, 4 food output, estimates of, 681–5, 682, 683, 690–1 foreign influences, pre-1000 C E, 203–5 Buddhism, 216–21 disintegration of trade monopoly, 234–9 Han Empire’s Western Region, 205–10 Mediterranean connections, 221–5 Period of Disunion’s silk–horse trade, 210–16 Sogdian caravans and diasporas, 225–34, 228 foreign trade. See foreign influences; international trade forests, 40, 54, 55–6, 68, 261–2, 488

Former Matters of Wulin, 532 Foshan, Guangdong, 542, 605 foundries, 24–8, 94, 107 Frankfurt school of Marxism, 3 fraud, 438–41 free coinage, 113 French physiocrats, 109 fruit trade, 695–6 Fu Jian 苻堅, 84 Fu Yiling 傅衣凌, 7, 467 fubing militias, 125, 127, 128 Fufeng county, Shaanxi, 222 Fujian province, 290, 312, 318, 325, 498, 664 Fuzhou, Fujian, 654, 692 gain. See wealth Ganfu 甘父, 206 Ganjing river, 44, 45, 47 Ganjingzhen, Chongqing, 44 Gansu province, 270 Gao Huan 高歡, 84 Gao Shouxian 高寿仙, 686 Gaoyang commandery, 84 Gaozong, Emperor of Tang 唐高宗, 236 Gaozu, Emperor 漢高祖 (Liu Bang 劉邦), 73, 111, 134, 205 Gazetteer of the Four Provinces, 333 Gazetteer of the Three Provinces, 331, 332 Ge Jianxiong 葛剑雄, 301, 309 general commands (zonglingsuo), 350 geomorphology, 52–4, 53 Gesellschaft (impersonal social relations), 6 gift-giving economy, 15 gold currency, 139 Gong Yu 貢禹, 196 grain tax, 121, 612 grain trade, 187, 611–14, 613 market forces, 409–16 granaries, 508–10, 706 Grand Canal, 157, 157, 273, 295, 359, 492, 494, 507, 526, 611 Great Clans, 117, 118–19, 120, 122, 123 Great Divergence debate, 10 Great Learning, 394, 395 Great Wall, 205–6, 207, 653 Greek legacy. See Hellenistic legacy Green Sprouts (qingmiao) policy, 346, 405, 455, 703 Green Standard corps, 375 greenhouse cultivation, 77 gross domestic product (GDP) levels, 676, 677–81, 679, 685–90 Gu Zuyu 顧祖禹, 328

719

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index guahu resettlement program, 127 Guan county, Sichuan, 333 Guan Xianyao 官獻瑤, 389, 390–2, 395 Guan Zhong 管仲, 173, 175, 193 Guan Zi, 101, 114, 136, 172, 174–80, 185–9, 384, 697 Guangcao Canal, 82 Guangdong province, 493 Guangwu, Emperor of Han 漢光武帝, 80, 118–19 Guangzhou, Guangdong, 160, 498, 537, 668 Guanzhong basin, 62, 74, 76, 127 Gui E 桂萼, 389 Guiji commandery, 155 Guizhou province, 283, 286–7, 288 Guo Moruo 郭末若, 3 Guo Songyi 郭松义, 683, 691 Guyuan, Ningxia, 224, 230

Hemudu site, Zhejiang, 58, 63 Henan province, 36, 479 Hexi Corridor, 207, 211 Himiko 卑彌呼, Japanese queen, 214 Hino Kaisaburo¯, 4 Historical Records, 63 History of Chinese Capitalist Development, 10 History of Chinese Migration, 301 History of the Former Han, 59, 77, 78, 195, 307 History of the Song Dynasty, 307 History of the Three Kingdoms, 81 Ho Ping-ti 何炳棣, 9, 300, 546 hoeing, 71, 73 Hong Liangji 洪亮吉, 298, 394, 396–7 Hongwu, Emperor of Ming 明洪武 (Zhu Yuanzhang 朱元璋), 304, 361–3, 470, 533, 566, 569, 652 Hongzhi, Emperor of Ming 明弘治, 473, 566 horse trade Han dynasty, 209–10 Period of Disunion, 210–16, 217 Song dynasty, 645 Tang dynasty, 234–6 Yuan dynasty, 651 Houma foundry, Shanxi, 24–8 house firms, 626, 633 household registers, 112, 126, 182, 302–5, 371, 465 Hu Fenghui 胡鳳翬, 466 Hu Shengjian 胡聖健, 431 Hu, Empress of Northern Wei 魏胡太后, 213, 220 Huai river, 80, 81, 156, 160, 295, 352, 507, 601, 614 huaijiao notes, 352 Huan, Lord of Lu 魯桓公, 96 Huan, Lord of Qi 齊桓公, 173, 175, 193 Huang Ang 黃卬, 478 Huang Chao 黃巢, 160, 312, 537 Huang Xingzeng 黃省曾, 549 Huang Xueguang 黃學廣, 582–3 Huang Zhen 黃貞, 432 Huang, Ray 黃仁宇, 365 Huangchi, Anhui, 328 Huang-Lao philosophy, 113, 183 Hubei province, 261, 496 hudiao tax system, 120, 122, 136, 141 huhui notes, 352, 354 Hui, Emperor of Han 漢惠帝, 135 Huichang kaiyuan coin, 139 Huisheng 惠生 (Buddhist monk), 213 Huizhou merchants, 475, 544, 545, 546–7, 625, 633–4

Hakka people, 36, 317 Han dynasty agriculture, 17, 56, 59, 73–80 climatic cooling and cropping changes, 80–7 currency, 134–7, 140–1, 141, 187–9 debates over political economy, 182–5 early state–economy relationship, 110–14 international trade, 203, 205–10, 216 markets, 144, 146–7 mercantilist fiscal state, 114–21, 185–92 merchants, 152–3 population changes, 302, 309 revenues, 117 triumph of the farming habitus, 192–6 Han Fei Zi, 100, 106, 172 Han Feizi 韓非子, 101, 172 Han river, 467–8 Han Shixun 韓師訓, 647 Han Yu 韓愈, 200, 318, 393 handicrafts, 154, 457, 510–12, 520 Hangzhou, Zhejiang, 260, 283, 320, 321, 324, 356, 464, 530–3, 612, 642, 650 Hann (state), 100 Hanxing coin, 137 Hao Yulin 郝玉麟, 285 harmonious purchases, 354 Hartwell, Robert, 9, 245 He Chou 何稠, 233 He Tuo 何妥, 233 He Ziquan 何茲全, 4 heavenly reason, concept of, 420 Hebei province, 36, 294, 313, 517 Hellenistic legacy, 221–5, 223

720

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Huizhou, Anhui, 283, 287, 473, 552, 570, 578–9, 623, 687 huizi notes, 352–4, 353, 574 Huizong, Emperor of Song 宋徽宗, 345 human feelings, concept of, 421 Hunan province, 496, 563, 614 Hundred Schools of Thought, 70, 72 hunting, 64, 98 Huobu currency, 136 Huoquan coin, 136 hutie registration system, 304 Hymes, Robert, 411

Japan, 2, 159, 213–14, 367, 538, 576, 644, 658, 660, 667 scholarship, 4, 243–5, 456 Japanese occupation (1930s–1940s), 6 Ji’nan, Shandong, 308, 309, 331 Jia Sidao 賈似道, 354, 355–6, 470 Jia Sixie 賈思勰, 84–7 Jia Yi 賈誼, 183, 185, 196 Jiading county, Suzhou, 606 Jiahu site, Henan, 65 Jiajing, Emperor of Ming 明嘉靖, 406 Jianghan Plain, 498–9, 506, 507 Jiangnan region, 313, 314, 315, 361–2, 457, 460, 470, 475, 515, 530, 553 Jiangyuan, Shaanxi, 222, 223 Jiankang (modern Nanjing), 147, 155–6, 215 Jianwen, Emperor of Ming 明建文, 394 Jiaohuo Marshes, 61 Jiaozhi, Red River delta, 215 Jiaxing, Zhejiang, 278–80, 554 Jin (state), 96 Jin dynasty (Great Jin), 295, 313, 342, 351, 352, 356, 580, 647 Jin dynasty (Sima Jin), 82, 121–2, 141, 147, 154, 204, 210, 219 Jin Fu 靳輔, 295, 394, 399–401 Jin Xueyan 靳學顏, 387 Jin Yipai 靳一派, 458 Jing river, 61, 62, 67 Jing Su 景甦, 462 Jing, Duke of Qi 齊景公, 168 Jing, Emperor of Han 漢景帝, 114 Jingdezhen, Jiangxi, 250, 540–2, 687, 693–4 Jinyang, Shanxi, 100–1 Jiujiang, Jiangxi, 592 Jixia Academy, 175 joint-stock partnerships, 455–6, 629–30 Ju Qingyuan 鞠清遠, 4 jue manufacture, 22–4 Jurchen (Jin) conquest, 272, 313, 529, 647 Justinian, Byzantine Emperor, 224

Ibn Battuta, 650 Ilkhan regime, Iran, 651 imperial law, concept of, 421 In Search of the Sacred, 143 India, 219–20 Indian Ocean, 239, 642, 649, 651, 656 inflation, 386, 481 inns, 149 insurance crops, 699–700 international trade, 637–8. See also Silk Road trade commenda partnerships, 628 Five Dynasties period, 248–50 official versus private Ming trade, 367, 638, 652–7, 655 port cities, 537–40 pre-imperial China, 48–9 Qing dynasty, 664–71 silver imports in the late Ming, 577, 659–63, 661 smuggling and piracy, 657–9 Song dynasty, 639–48, 641, 643 Tang dynasty, 158–60 Yuan dynasty, 648–51 interregional exchange, pre-imperial Tonglüshan, Hebei, 37–44 Zhongba, Chongqing, 44–8 iron Han dynasty, 115, 152, 188 late pre-imperial China, 35–6, 56, 58, 68–9 Qin dynasty, 107 irrigated agriculture, 60–3, 72, 75, 401, 497–8, 503 Isaac (Nestorian priest), 236 Islamic influences, 236–9 Italy, 555

Kaifeng, Henan, 314, 320, 349, 526–9, 528, 573, 646, 699 Kaiyuan tongbao coin, 138–9, 577 Kang Ju 康居, 231 Kang Mengxiang 康孟詳, 231 Kang Senghui 康僧會, 231 Kangxi, Emperor of Qing 清康熙, 372, 399, 664 karst landscapes, 286 Kashmir, 218 Kato¯ Shigeshi, 4

Jade Gate, 207 Jang Bogo 張寶高, 159

721

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Keynesian economics, 6 Khitan people. See Liao Empire Khosro I, Sassanid king, 229 Khotan, 208, 216 Khubilai Khan, 320, 357, 359, 398, 568, 580, 649 Kirghiz people, 235 Kish island, 651 knife-shaped currencies, 132 Korea, 159, 213–14, 538, 655, 668 Korolkov, Maxim, 108 Kucha (oasis state), 217–18 Kumaiyo Yukio, 85 Kumarajiva (Buddhist scholar), 217–19 Kushan Empire, 216, 231

debt and default laws, 433–8 economic development, 425–9 fraud in trade, 438–41 Ming and Qing partnership disputes, 431–3 Song and Yuan contract law, 429–31 traditional law overview, 420–4 lead disks, 222, 223 Ledderose, Lothar, 31 Legalists, 175, 182 Lesser Treasury, 77, 103–4, 111 Levant, 224 Li Ao 李翱, 200 Li Bozhong 李伯重, 88, 466, 487, 504, 515–16, 680 Li Chengrang 李承讓, 432 Li Dou 李斗, 546 Li Guangli 李廣利, 210 Li Jinde 李晉德, 623 Li Kui 李愧, 69, 170, 176, 398 Li Linfu 李林甫, 128 Li Ling 李零, 34 Li Qingzhi 李清植, 413 Li Shihua 李時化, 436 Li Shizhen 李士楨, 664 Li Si 李斯, 104 Li Xian 李賢, 224, 230 Li Xizong 李希宗, 224 Li Yuan 李淵, 87. See Taizong, Emperor of Tang Li Zhanran 李湛然, 436 Liang dynasty, 122 Liang Fangzhong 梁方仲, 5 liangzao xinquan coin, 137 Liao Empire, 341, 572, 639, 646–7 Liao Shuwu 廖淑吾, 436 lifespan trends, 696, 697 lijia system, 362, 364 Lin, Aiming 林愛明, 53 Lin Xiuzhu 林修竹, 330 Lin’an county, Zhejiang, 324 lineage institutions, 700–2 Ling, Lord of Jin 晉靈公, 96 Lingtai county, Gansu, 222 Linqing, Shandong, 606, 607–10, 610 Linxian county, Henan, 463 Linxiang county, Hunan, 120 Linzi, Qi state, 36, 102 lithic production, 21–3 Little Ice Age, 262, 264–6, 265 Liu Bang 劉邦. See Gaozu, Emperor of Han Liu Guangxia 劉光夏, 437 Liu Kezhuang 劉克莊, 355 Liu Pi 劉濞, 113

labor markets, 453–4, 462–4 labor service, 105–6, 112 Ming dynasty, 362, 364 Northern Wei dynasty, 197 Qin dynasty, 110, 182 Tang dynasty, 126, 246 laissez-faire policies, 183, 369, 454 Lake Tai, 155, 263, 548 land allocation, 98–9, 123–4, 124, 170, 195, 340. See also equal-field system landownership Han dynasty, 117, 119 Jin dynasty (Sima Jin), 121 Period of Disunion, 83, 122 post-1000 C E, 458, 461 Qin dynasty, 106 Tang dynasty, 247, 252–4 land prices, 472–8, 473, 570, 571 land sales, 252–3, 254, 355, 450–3, 570–1 landlord–tenant relationships, 460 asset management, 461–4 Laoshan, Shandong, 219 late pre-imperial China, 15–16. See also Warring States period agriculture, 16–17 Doufucun workshop, Shaanxi, 28–31 Houma foundry, Shanxi, 24–8 international trade, 48–9 interregional exchange in Tonglüshan, 37–44 interregional exchange in Zhongba, 44–8 iron revolution, 35–6 Qijiacun workshop, Shaanxi, 21–4 urbanism and commerce, 17–20 workshop developments, 31–4 Later Zhao (state), 137 law, 419–20 brokerage regulations, 441–4

722

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Liu Qilong 劉起龍, 441 Liu Song dynasty, 137 Liu Xiu 劉秀. See Guangwu, Emperor of Han Liu, Xinru, 158 Liu Yan 劉晏, 128, 160, 198–9, 201 Liu Yu 劉域, 441 Liu, William Guanglin 劉光臨, 513, 686 living standards, 512–18, 515, 676–7 capacity to withstand survival risks, 697–707 consumption and welfare, 690–7, 697 cross-validation of per capita GDP, 685–90, 688 estimates of production output, 677–81, 679 per capita food output, 681–5, 682, 683 Liye documents, 102, 104–8 Lizong, Emperor of Song 宋理宗, 355 loans, 455, 622–5, 702–4 local market transactions, 584–91 local society, 110, 371 Locke, John, 448 locust, 296–7 loess soil, 54, 502 Lotus Sutra, 218 Lu (state), 36, 67, 96, 97, 168 Lü Buwei 呂不韋, 103–4, 173 Lu Dayou 陸大有, 437 Lü Guang 呂光, King of Liang, 217 Lu Rong 陸容, 576 Lu Shirong 盧世榮, 359 Lu Tan 盧坦, 410 Lu Yu 陸羽, 89, 158 Lu Yulin 陸遇霖, 461 Lu Zhi 陸贄, 200 Luminous Collection of Judgments by Illustrious Figures, The, 451 Luo Lun 罗崙, 462 Luo river, 75 Luodian (market town), 606–7 Luojiajiao site, Zhejiang, 58 Luoyang, Henan, 147, 158, 220–1 luxury goods, 215, 217, 219, 220, 546, 644, 648, 650, 692, 696

maize, 285–6, 287, 468, 495, 695, 700 Malay Peninsula, 660 Malthusian theory, 310, 677, 679, 707 managerial farms, 462–4 Manchu conquest, 7, 316, 368–72, 663 Manchuria, 63, 290, 318 Maniakh (Sogdian envoy), 229 Manichaeans, 205 Manila (Spanish province), 367, 577, 660–2 Mann, Susan, 620 Manual of Locust Control, 296 manufacturing cities, 537, 540–3 manure production, 279 maritime trade Ming trade ban, 367, 652, 653 Qing dynasty, 664–71 silver imports in the late Ming, 659–63, 661 smuggling and piracy, 657–9 Song dynasty, 639, 641–5, 643 Yuan dynasty, 359, 649 market economy law. See law market forces, 409 markets birth of walled markets, 143–5 growth in post-1000 C E period, 327–9, 329, 336, 337, 522, 552–5, 600–11, 601, 604, 606, 608, 610 official and private markets, 146–9 Qin dynasty, 107–8 shops and inns, 149–50 Tang–Song transition, 251–2 three-tiered price system, 145–6 Warring States period, 169 Marks, Robert, 298 Marmé, Michael, 548 Marxism, 3–4, 243 Master Shen’s Treatise on Agriculture, 462, 476 medical treatments, 494 Medieval Warm Period, 261, 263–6, 265 Mediterranean trade, 221–5, 563 Meilink-Roelofsz, M.A.P., 628 Melaka, Malay peninsula, 660 Menander Protector, 229 Mencius, 64, 70, 71, 118, 143, 173, 175, 180–1 Meng Yuanlao 孟元老, 529, 573 mercantilist fiscal state, 114–21, 117, 185–92 merchants, 598–9. See also markets; mercantilist fiscal state currency use, 592 European trade, 668–70 government and the late-imperial economy, 630–5

Ma Huan 馬歡, 656 Ma, Debin 马德斌, 515 Macau, 669 McDermott, Joseph, 284 macroregional structure, 485, 524, 605 Maddison, Angus, 677–8 Magang, Guangdong, 616 magnate clans, 79–80, 83, 195 Mai Xunnan 麥巽南, 440

723

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index silverization, 363–8, 366, 569, 658, 659–63 smuggling and piracy, 657–9 trade, 612–13, 614–15, 638, 652–7, 655, 659–63 urban economy, 465, 533–6, 541–2, 545, 548–50, 553–4 water control, 272–4, 498, 500, 507, 508 mingtianzhai land allocation, 106, 111, 114, 115, 119, 272–4 Miyazaki Ichisada, 4, 244 Mo Zi, 143, 173 Monetary History of China, 8 monetary systems. See currency Möngke Khan, 355, 357 Mongol conquest period, 353, 612 emergence of a fiscal system, 356–8 monographic studies, 8 Monthly Instructions for the Four Peoples, 80, 151 Morse, H.B., 581 Mote, Frederick, 653 Mount Tai, 273 Mozi 墨子, 169, 174 mulberry cultivation, 63, 78, 83, 86, 276–8, 279 mulberry dike–fish pond complex, 276–8 multi-cropping, 283, 489–90 Muran, Chengde, 290 Muslims, 160 Myers, Ramon, 9

merchants (cont.) great merchants and multiple economic zones, 150–1, 344 Huizhou merchants, 475, 544, 545, 546–7, 625, 633–4 late-imperial problems, 617–30 Period of Disunion, 153–6 Qin dynasty, 106–7, 182 restraints and commercial taxation, 152–3 salt monopoly, 364, 372, 374 Sui and Tang periods, 156–61 Warring States period, 171, 173, 177, 179–80 Merchants Come to Their Senses at Just One Glance, 623 metal tools, 58 metallurgy, 39, 42, 68, 97, 107 Metzger, Thomas, 389, 392 Mexican dollars, 583, 592 Miao Rebellion, 377 Miao wood, 282 migration, 311–18, 312, 313, 469–70, 481. See also population changes Mikhailovich, Aleksei, tsar of Russia, 666 militarist–physiocratic state. See fiscal state, pre-1000 C E military expenditure, 375, 375–7 military garrisons, 315, 366 military service Han dynasty, 112, 119 Ming dynasty, 361 Period of Disunion, 125 Qin dynasty, 110 Spring and Autumn period, 95 Warring States period, 97–8, 170 Mill, John Stuart, 381 Min (state), 248 Ming Code, 452 Ming dynasty consumption levels, 691, 693 currency, 566, 569, 570, 575–6, 658 law, 424, 431–3, 436–8, 440–1, 442–4 living standards, 686, 701–2, 706 market towns, 600–5, 601, 604, 606, 608, 610 merchant problems, 617–30 merchants and the state, 630–5 population changes, 304–5, 306, 307–8, 310, 314–16, 315, 316, 317, 320–7, 322, 324, 326, 327, 328–9, 329, 469–70 property rights and factor markets, 452, 470, 475 public finance, 341, 361–3 rural economy, 259–60, 504, 509, 514

Naito¯ Konan, 2, 243, 246, 247, 340 Nanjing, Jiangsu, 315, 533–4, 556 Nanping, Fujian, 570 native-place associations, 634–5 Neiwufu, 369, 374 Nestorians, 205 New Institutional Economics (NIE), 425–6 New Monastery, 238 New Policies (Song dynasty), 345–7, 403, 405, 455 Ni Kuan 倪寬, 75 Ni Si 倪思, 411 Ni Yuan 倪元, 443 Niida Noboru, 6 Nine Ranks system, 120 Ningbo, Zhejiang, 538, 577, 583, 642, 650 Ningzong, Emperor of Song 宋寧宗, 355 Nishijima Sadao, 7 Nishiyama Takeichi, 85 nobility, 96, 98, 111, 118, 168, 182 nonagricultural resources, 63–5, 169 North, Douglass, 425 Northern Qi (state), 156 Northern Wei dynasty agriculture, 60, 82–4, 87

724

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index land allocation, 122–5, 124, 197 markets, 147 merchants, 154–5 trade, 204, 211–13, 220–1, 231

Pingcheng, Shanxi, 82 Pingwang (Yangzi Delta), 259–60 Pinus massoniana (Chinese red pine), 276 piracy, 652, 657–9 plagues, 314 Pliny the Elder, 222 plow cultivation, 58, 69, 77 political economy, 166–7, 381–4 conservatism and tradition, 392–402, 400 fiscal state in the Warring States period, 167–81 Han debates and mercantilism, 181–92 market forces and parent officials, 409–15 post-Han fiscal regimes, 197–201 resource distribution and visions of the state, 402–9 triumph of the farming habitus, 192–6 wealth and gain concepts, 384–92 Yuan dynasty, 358–60 Polo, Marco, 534, 539, 649, 650, 651 Pomeranz, Kenneth, 481 population changes, 297, 300–1 national population figures, 309, 309–11, 680 production output levels, 678, 681 records and demographic data, 302–9, 306 regional distribution and migration, 311–18, 312, 313, 315, 316, 317, 319, 495 Tang and Song periods, 128, 250, 261, 487 urban economy, 536, 553–5 urban population post-1000 C E, 318–36, 322, 324, 326, 327, 329, 334, 335, 336, 337 Population History of China, 301 porcelain, 250, 540–2, 644, 668, 693–4 port cities, 537–40 Portuguese traders, 660 Post-Shijiahe Culture (c. 2400–1900 B C E ), 37 Precepts for Social Life, 702 primogeniture, 99 printing industry, 251, 694 privy purse, 77, 111, 115, 342 production output, 676–7 consumption and welfare, 690–7 cross-validation of per capita GDP, 685–90, 688 estimates of, 677–81, 679 per capita food output, 681–5, 682, 683 promissory bills, 161, 350 property rights, 448–9 labor markets, 453–4 land markets, 450–3, 704 prosperity. See wealth providential state, 181 Pu 濮 lineage, 553

O’Brien, Patrick, 514 Oberst, Zhihong Liang, 404 Officers (Rites) of Zhou, 117, 348, 383, 390–2, 398, 399, 706 Ögödei Khan, 353, 357, 358 Okamura Hidenori, 64 Opium Wars, 7, 668, 673 Oriental culture, 2 Oriental Modern Age, The, 244 Origins of the Way, The, 393 ortoq merchants, 649, 651 Ou Zhenfan 歐振凡, 440 Ouyang Shoudao 歐陽守道, 411 Overall Survey of the Ocean’s Shores, 656 Overton, Mark, 490 Palmyra, Syria, 223 Pan Jixun 潘季馴, 272–4, 295 partnership disputes, 431–3 pastoralism, 71, 76 patriline, 252, 253 patrimonial state, 93–4, 111 Pattern of the Chinese Past, The, 245 pattern-block casting, 26–7 pawnshops, 622–3, 633, 703 peanuts, 495 Pearl river, 53, 278, 498, 504 Peiligang site, Henan, 58 Peng Kaixiang 彭凯翔, 513 Peng Xinwei 彭信威, 8 People’s Republic of China, 6 pepper imports, 650, 656 Period of Disunion agriculture, 56–7, 81–7, 88 currency, 137–8, 141, 141 international trade, 210–16, 217–20 merchants, 153–6 Nanjing (Jiankang), 147, 533 population changes, 302 state–economy relationship, 121–5, 124 Perkins, Dwight, 9, 488, 497, 681, 683 Persian Garrison militants, 540 Persian Gulf, 642, 651 Persian-style goods, 232–4 Philippines, 657 physiocratic philosophy, 109, 194 piece-mold casting, 26 Pimentel, David, 278

725

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index public finance, 340–1 budgetary crises (1205–1276 C E ), 352–6, 353 budgetary crisis and bureaucratization (1040–1127 C E ), 344–9, 347 early Ming developments, 361–3 eighteenth-century revenue and expenditure, 372–8, 373, 375 fiscal governance under Manchu rule, 368–72 fragmentation of finances (1127–1205 C E ), 349–52, 352 Mongol fiscal system (1230–1271 C E ), 356–8 recentralization of finances (960–1040 C E ), 341–4, 343 silverization in the Ming dynasty, 363–8, 366, 368 Yuan dynasty’s political economy, 358–60 public-land (gongtian) initiative, 355 Putian county, Fujian, 577 Puyuan (market town), 553

factor markets, 464, 470, 475–9, 480 fiscal governance, 368–72 law, 424, 427, 431–3, 436–8, 440–1, 442 living standards, 513–18, 515, 687–8, 706–7 market towns, 600–6, 601, 604, 606, 610 merchant problems, 617–30 merchants and the state, 630–5 population changes, 305–6, 308–9, 310, 316–18, 317, 319, 328–36, 334, 335, 336, 337, 469–70, 495 public finance, 341 rural economy, 487, 490, 504–5, 509–10, 518 Suzhou, Jiangsu, 466, 550–2, 551 trade, 613, 614–17, 664–71 urban economy, 535, 542 water control, 274, 497, 499, 501, 507, 508 Yangzhou, Jiangsu, 545–7 Qingdao, Shandong, 331 Qinghe, Gansu, 570 Qingming festival, 578, 579, 589 Qingshuijiang, Guizhou, 571 Qingtang Tibetans, 647 Qingyuan county, Hebei, 590 Qu Jian 屈建, 97, 98 Quan Hansheng 全漢昇, 4, 5 Quanzhou, Fujian, 325, 538, 556, 642, 645, 649 Quesnay, François, 109

Qi (state), 36, 96, 101, 102, 132, 144, 168, 173, 193 Qianfeng quanbao coin, 138 Qiang people, 82 Qianling, Hunan, 102, 104–8 Qianlong coins, 564 Qianlong, Emperor of Qing 清乾隆, 370, 374–5, 377, 407–8, 415, 478, 480, 545, 567, 610, 615, 665 qianpiao notes, 580, 590–1 Qiao Guanglie 喬光烈, 390 Qiaocun, Shanxi, 33 Qidao wubai currency, 135 Qijiacun workshop, Shaanxi, 21–4 Qin (state). See also Qin dynasty agriculture, 61, 70–3 commerce, 47, 173 currency, 19, 101, 132, 139 Doufucun workshop, 28–31 militarist–physiocratic state, 170–1 Qin dynasty. See also Qin (state) command economy, 92, 102–10, 181–3 currency, 132–4 markets, 144 population records, 302 Qin Gui 秦檜, 350 qin zither, 49 Qing dynasty consumption levels, 691–2 currency, 566, 569, 571, 578–80, 579 eighteenth-century revenue and expenditure, 372–8, 373, 375 environmental changes, 266, 291, 292

Ran Qiu 冉求, 181 Rebellion of the Seven Princedoms, 114 Rebellion of the Three Feudatories, 316 redistributive economy, 93–4 registration systems, 98, 113, 182. See also household registers religious organizations, 704–5 Ren Boyu 任伯雨, 272 Ren Naiqiang 任乃强, 46 rental agreements, 430 Renzong, Emperor of Song 宋仁宗, 344 resource distribution, 402–9 revenue figures Han dynasty, 117 Ming dynasty, 366, 368 Qing dynasty, 370, 372–8, 373 Song dynasty, 343, 343, 347 Yuan dynasty, 359 rice cultivation, 54, 62, 75, 88, 250, 312 rice prices, 472–3, 473, 476, 477–8, 579, 589, 689, 698 Richthofen, Baron von, 54 right of pre-emption, 452 Roman Empire, 221 Rowe, William, 410

726

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Shi Shewu 史射勿, 230 Shi Zhaizhi 史宅之, 355 Shi Zhihong 史志宏, 683, 691 Shiba Yoshinobu, 9, 603 shipbuilding, 250, 550 shops, 149 Sho¯so¯in repository, 214 shoushi records, 302 Shouwangfen ironworks, Hebei, 35 Shu (state), 102 Shu Han (state), 137, 154 Shuihudi documents, 102 Shun 舜, legendary emperor, 184 Shunzhi, Emperor of Qing 清順治, 475 Siam, 655 Sichuan basin, 45, 47, 286 Sichuan province, 288, 308, 333, 349, 356, 467–8, 493, 564, 572, 601 Sifangtang cemetery, Hubei, 43 silk as currency, 142 horse–silk trade, 210–16, 217, 234–5 Mediterranean market, 222–4 price of, 689 production, 154, 465–6, 549, 552–3 Zandaniji silk, 226, 232 Silk Road trade, 203–5 and Buddhism, 216–21 disintegration of governmental monopoly, 234–9 Han Empire’s Western Region, 205–10 Mediterranean connections, 221–5 Period of Disunion, 210–16, 212 Sogdian caravans and diasporas, 142, 225–34, 228 silkworm-raising, 63, 89, 251, 276–7 Silla, Korea, 159 silver, 360, 371, 386, 387, 570, 589, 594 imaginary tael unit, 580–4, 582 import routes, 577 Ming dynasty silverization, 363–8, 366, 569, 658, 659–63 silviculture, 78, 87, 279, 280–3 Sima Biao 司馬彪, 307 Sima Guang 司馬光, 403–4, 406–7, 408, 410 Sima Qian 司馬遷, 63, 75, 99, 107, 151, 184, 189–92, 196, 385 Single-Whip tax reform, 5, 364, 370, 471, 474 Six Auxiliary Canals, 75 Skinner, G. William, 9, 484, 524, 547, 605 Slave Contract, 146, 154 slavery, 65–6, 80, 105, 146, 454 Smith, Adam, 381–2

royal household institution (Zhou dynasty), 93 Rozman, Gilbert, 678 Ruan Yuan 阮元, 547 rural economy, post-1000 C E, 484 handicrafts, 510–12 living standards, 512–18, 515 natural conditions, 485–8 new crops, 488–96 soil and fertilizer, 502–5 state’s role, 505–10 water control, 496–502 rural households. See family households Russia, 665–6 ruzhong system, 343, 364 Ryukyu kingdom, 367, 540, 654, 667–8 safe-deposit firms, 161 safflower cultivation, 85 Salt and Iron Debates, 117, 192 salt monopolies, 115, 128, 152, 160, 188, 198, 342, 348, 359, 364–5, 372, 545, 632 salt production, 44–8, 153, 158 Samarkand, Uzbekistan, 225, 654 Sang Hongyang 桑弘羊, 188–9, 192–3, 403 Sangha 桑哥 (Yuan fiscal officer), 359 Sanyuanzhuang site, Jiangsu, 77, 78 Sassanids, 224–5, 229 satrapies, 111, 114, 152 scholar-officials, 597–9, 705 Scott, James, 110 Second Jinchuan War, 376 secularization, 244 Sen, Amartya, 416 sericulture, 78, 89, 458, 512 Seven Treasures concept, 219, 220 Shaanxi province, 308, 467–8, 572 Shahrukh, Timurid ruler, 654 Shandong province, 36, 308, 314, 330–3, 462, 479, 505, 517 Shang dynasty, 21, 38, 54–5, 64, 65–6, 131 Shang Yang 商鞅, 70–1, 98–100, 102, 103, 170–1 Shang Yue 尚鉞, 7 Shanghai, 475, 583–4, 594 Shanxi merchants, 633–4 Shanxi province, 59, 370, 467–8 Shaoxing, Zhejiang, 155, 325 shed people, 317 Shen Shixing 申時行, 407 Shenzong, Emperor of Song 宋神宗, 345, 388, 403, 639 shesu rice, 285 Shi Hedan 史訶耽, 230

727

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index S´rı¯vijaya Empire, 159 standardized land units, 171 state–economy relationship, 92–3 autocratic state in the Warring States period, 96–102 command economy of Qin, 102–10 domain economy of the Spring and Autumn era, 94–6 early Han dynasty, 110–14 mercantilist fiscal state of Emperor Wu, 114–21, 117 Ming–Qing dynasties, 630–5 Period of Disunion, 121–5, 124 redistributive economy of the Zhou dynasty, 93–4 Sui–Tang dynasties, 125–9 Statutes and Sub-statutes of the Great Qing, 422 Statutes of the Great Ming, 307 Stein, Aurel, 225 stone tools, 58 Studies on the Population of China, 300 Study of China’s Ancient Society, 3 Su Dai 蘇代, 151 Su Qin 蘇秦, 151 suanfu levy, 111, 112, 114 suanmin levy, 152–3 Sudo¯ Yoshiyuki, 6 Sufism, 238 Sui dynasty agriculture, 87–9 equal-field system, 125–6, 197 markets, 147 trade, 156–8, 214, 215, 233 Sui History, 156 Sumatra, 159 Sun Hao 孫灝, 416 Sun Lijuan 孫麗娟, 445 Sun Quan 孫權, 120, 214 Sun Zhouyong 孙周勇, 22 supply and demand, 187, 191, 574 Survey for Border Defense of the Three Provinces, A, 467 Survey Records of Localities in Shandong, 356 Suzhou, Jiangsu, 465–6, 551, 556, 606, 615 sweet potatoes, 495, 695, 700 swiddening, 281–2

Smith, Arthur, 585 Smith, Paul J., 404 smuggling, 655, 657–9 Social History Debate, 3 social welfare institutions, 700–2 Sogabe Shizuo, 4 Sogdians, 57, 87, 142, 147, 155, 158, 211, 225–34, 228, 237, 238 soil, 54, 59–60, 267–72, 269, 276, 502–5 solar radiation, 280 Song dynasty budgetary crises (1205–1276 C E ), 352–6, 353 budgetary crisis and bureaucratization (1040–1127 C E ), 344–9 currency, 563, 566, 568, 570, 571–4, 645 fragmentation of finances (1127–1205 C E ), 349–52, 352 Hangzhou in the Southern Song, 530–3 Kaifeng in the Northern Song, 526–9, 528 law, 423, 429–31, 433–5, 438, 439, 441 living standards, 686, 701 markets, 606 population changes, 303, 307, 310, 313, 319–20, 327–8 production and consumption, 689, 689, 690, 693 property rights, 451, 704 recentralization of finances (960–1040 C E ), 340–4 recruitment system, 247–8 revenue figures, 343, 347 rural economy, 487, 508, 513 trade, 639–48, 641, 643 urban economy, 522, 538–40, 544, 548, 552 water control, 272, 294, 500, 506–7 Song Penal Code, 433–5, 438 Song Yingxing 宋應星, 392 Songjiang, Shanghai, 606, 615 South Manchuria Railway, 6 Southeast Asian trade, 367, 488, 489, 642, 655, 665 Southern Han (state), 248, 572, 641 Southern Tang (state), 248, 249 spade-shaped currencies, 132 Spanish pesos, 592 Spanish traders, 660–2 Spring and Autumn Annals of Master Lü, 59, 72, 77 Spring and Autumn Annals of Master Yan, 174 Spring and Autumn period, 67–8, 94–6, 131, 150, 168, 302 sprouts of capitalism, 7, 456, 464, 467 Sri Lanka, 219

Taibai Lake, 261 Taicang Treasury, 364, 365, 367, 368 Taiping Rebellion, 334, 567, 576 Taiwan, 317, 477, 664 Taiwu, Emperor of Northern Wei 魏太武帝, 82, 211, 213

728

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Taizong, Emperor of Song 宋太宗, 342, 641 Taizong, Emperor of Tang 唐太宗 (Li Yuan 李淵), 526 Taizu, Emperor of Song 宋太祖, 341–2 Taklamakan Desert, 208, 217, 223 Tan Lun 譚綸, 387 Tanaka Masatoshi, 7 Tang Code, 421, 434, 438, 444 Tang dynasty, 246–7, 450 agriculture, 57, 87–9 currency, 138–9, 142, 161, 577, 581 fiscal system, 92, 126–9, 197–201, 340 international trade, 204, 215–16, 234–9, 537 markets, 147–9 merchants, 156–61, 630 population records, 302, 319 Sogdian influence, 230, 232, 233–4 Tang Shu 唐樞, 659 Tang 唐, legendary emperor, 185 Tang–Song transition aftermath of the An Lushan Rebellion, 246–7 decline of the aristocracy, 247–8 economic growth, 251–5 Five Dynasties period, 248–50, 249 Japanese scholarship overview, 2, 6, 243–5 population changes, 312 rise of South China, 250–1 state–economy relationship, 92 Western scholarship overview, 245 Tangut people. See Xi Xia Tao Xisheng 陶希聖, 4 Tarim Basin, 57, 203, 208, 209, 211, 218, 231, 238 Tawney, R.H., 481 taxation, 471–2, 474, 567–71 Han dynasty, 74, 111, 113, 115, 116, 119, 121, 152–3, 186, 189 Jin dynasty (Sima Jin), 121 Ming dynasty, 361–2, 364, 365, 368 Mongol conquest period, 357–8 Northern Wei dynasty, 83, 197 Period of Disunion, 122, 124, 154 Qin dynasty, 106 Qing dynasty, 369, 370–1, 378 Song dynasty, 343, 347, 348, 349, 354, 639 Sui dynasty, 142 Tang dynasty, 126–7, 128, 160, 197, 199–201, 246–7 Warring States period, 97–8, 173 Yuan dynasty, 358–9 Zhou dynasty, 67, 168–9

tea, 89, 158, 250, 342, 348, 670 Temür, Timurid ruler, 654 tenurial rights. See land ownership textile payments, 140–3 Thesis on China, 243 three-elders system, 83, 125 Three Gorges Dam, 44 Three Kingdoms period, 81–2, 136, 147, 154, 155, 203 three-tiered price system, 145–6 Tian 田 clan, 193 tiger populations, 290–1 Tillya Tepe site, Bactria, 206, 209 Timurid Empire, 654 ting qi ziran policies, 410 tiraz system, 236 tobacco, 494 Tokugawa shogunate, 660, 662 Tonglüshan, Hebei, 37–44 Tongshan county, Jiangsu, 590 Tongtaihao accounts, 581–3, 582, 585, 585–8, 587, 588 Tongzhou, Shanxi, 227 tools, agricultural, 58 topsoil ownership, 459–60 trade. See international trade; merchants; Silk Road trade tradition, 392–402, 400 transaction agreements, 429 transport of commodities, 611–17, 613 Traveling Upriver at Qingming Festival, 527, 528 Treaty of Kyakhta (1727), 665 Treaty of Nerchinsk (1689), 665 Treaty of Shanyuan (1005), 646 tree cultivation, 84, 86–7 tributary system, 210–16, 237, 637, 652, 653–5, 655 Tribute of Yu, 150, 271 tuntian land allocation, 120–1, 207–8 Tunxi, Anhui, 578, 579, 589 Tuoba 拓跋 clan, 82, 154–5, 204, 211 Turkish tribes, 229 Tuyuhun people, 211 twice-a-year tax system, 128, 139, 160, 199, 246–7, 303, 343, 450, 471, 568 Twitchett, Denis, 9 Umayyad caliphate, 237–8 Universal Gazetteer of the Great Ming, 305 Universal Gazetteer of the Great Qing, 308, 328, 335, 336

729

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index merchants, 150–1, 171, 173, 177, 179–80 rise of the autocratic state, 96–102 water, 272–4, 488, 496–502, 505–8. See also floods Way of Celestial Masters, 704 Wazhadi site, Chongqing, 45 wealth concepts of, 384–92 conservatism and tradition, 392–402, 400 resource distribution and visions of the state, 402–9 weapon production, 657 Weber, Max, 3, 523, 557 Wei (Cao Wei), 81, 120, 154, 214 Wei (state), 100, 132, 144, 170 Wei Jirui 魏際瑞, 477 Wei river, 53, 54, 61, 67, 70, 74, 494 Weihaiwei, Shandong, 332 well-field system, 118, 173, 195, 449 Wen Dazhong, 278 Wen, Emperor of Han 漢文帝, 74, 113–14, 183 Wen, Emperor of Sui 隋文帝 (Yang Jian 楊 堅), 214 Wencheng, Emperor of Northern Wei 魏文 成帝, 82 Wendeng county, Shandong, 331 Wenqiao yidui hoard, Hunan, 564 Western Wei dynasty, 84, 125 wheat, 491 White Lotus Rebellion, 377, 397, 468 widows, 253 Will, Pierre-Étienne, 412 Wittfogel, Karl, 3 Wokou (pirates), 367, 659 women, 112, 196, 253, 493, 512 Works of Heaven and the Inception of Things, The, 392 workshops developments in pre-imperial China, 31–4, 94 Doufucun workshop, Shaanxi, 28–31 He Chou’s workshops, 233 Houma foundry, Shanxi, 24–8 iron production, 36 Qijiacun workshop, Shaanxi, 21–4 urban industries, 464–8 world-economy theory, 37 world-systems theory, 37 writing, 34 Wu (state), 120, 137 Wu (Sun Wu), 81, 214 Wu Chengming 吴承明, 10, 613 Wu Da 吳達, 443 Wu Hui 吴慧, 690, 691

urban economy, post-1000 C E, 522–5, 525 commercialized administrative cities, 543–52, 551 comparative perspective, 555–7 Hangzhou in the Southern Song, 530–3 Kaifeng in the Northern Song, 526–9, 528 manufacturing cities, 537, 540–3 market towns, 552–5 Nanjing and Beijing, 533–6 port cities, 537–40 urban and hinterland workshops, 464–8 urban economy, pre-imperial, 17–18 Uyghurs, 234–6 vagrancy, 127, 198 van Dyke, Paul, 669 van Zanden, Jan Luiten, 680 vegetable trade, 695–6 verification tallies, 105 veterinary treatments, 494 Vietnam, 493, 657 village worship associations, 624 Vries, Peer, 379 wadang patterns, 29–31 wage trends, 685–90, 688 walled markets, 143–5 Wallerstein, Immanuel, 37 Wang Anshi 王安石, 341, 345, 403, 404–5, 415, 417, 455, 506, 630, 640, 703 Wang Bao 王褒, 79, 146 Wang Chongyun 汪崇筼, 372 Wang Fu 王甫, 146 Wang Gungwu 王賡武, 655, 671 Wang Huizu 汪輝祖, 478, 584, 589 Wang Mai 王邁, 355 Wang Mang 王莽, 80, 117–19, 135–6, 153, 195 Wang Qianchu 汪乾初, 438 Wang Shimao 王世懋, 280 Wang Shuying 王叔英, 394, 395–6, 398 Wang Zhen 王禎, 270 Wang Zhi 王直, 659 Wang Zhong 汪中, 547 Wanli coins, 565, 566 Wanli, Emperor of Ming 明萬曆, 367, 510, 609 ward-and-market system, 147 Warring States period. See also late pre-imperial China agriculture, 68–73 commerce, 18–20, 171–2, 173–5 currency, 131–2, 133, 139 emergence of the fiscal state, 167–81 markets, 143–4

730

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Wu Songdi 吴松弟, 303, 307, 320, 327 Wu Wei 吳煒, 413–15, 416 Wu, Emperor of Han 漢武帝, 59, 75, 79, 140, 188–9, 192, 205, 208, 210, 403 mercantilist fiscal state of, 114–21, 152–3 Wu, Emperor of Jin 晉武帝, 210 Wuhu, Anhui, 592 Wujiang county, Suzhou, 603 Wujiang, Jiangsu, 490 Wulijie, Hubei, 41–2 Wuqing (market town), 552 Wu-Yue (state), 248, 500 wuzhu coin, 116, 118, 122, 135, 136–7, 137, 154, 562 Wuzong, Emperor of Tang 唐武宗, 139, 234 Wuzong, Emperor of Yuan 元武宗, 360

Xuanzang 玄奘 (Buddhist pilgrim), 236 Xun Zi, 173, 181, 182 Xunzi 荀子, 179–80, 185, 190 Xuwan (market town), 616 Yajiao, Chongqing, 47 Yan (state), 36, 100, 132 Yan Ruyi 嚴如熤, 468, 496 Yan Ying 晏嬰, 168, 174, 193 Yancheng county, Henan, 479 Yang Jian 楊堅, 87, 125, 156 (see Wen, Emperor of Sui) Yang Lien-sheng 楊聯陞, 8 Yang Mingxue 楊明學, 436 Yang Xueming 楊學明, 436 Yang Yan 楊炎, 128, 160, 167, 199, 201 Yang, Emperor of Sui 隋煬帝, 142, 157 Yangcheng, Henan, 36 Yangyue tribe, 40 Yangzhou, Jiangsu, 233, 365, 374, 544–7, 556 Yangzi delta agriculture, 88–9, 251, 259, 504, 612 currency, 575 environmental changes, 293, 294 markets, 552 Period of Disunion, 121, 155 rural economy, 490, 514–15, 515 textile production, 491, 512, 614 water control, 499, 501 Yangzi river basin cultivation, 57, 82, 87, 158 currency, 352 migration, 311–13 Period of Disunion, 155 Tang dynasty, 160, 250 Yankeshan hoard, Hunan, 564 Yanyinshanjiao, Hubei, 40 Yanzhou district, Shandong, 308 Yates, Robin D.S., 105 Ye county, Shandong, 332 Ye Mengzhu 葉夢珠, 475, 476 Ye, Hebei, 61 yellow ledgers, 304 Yellow River cultivation, 59, 63, 70, 74, 76, 77 environmental changes, 53, 54, 81, 83 migration, 57, 311 water control, 261, 272–4, 294–6, 506–7 Yellow Turban Rebellion, 120, 153 Yelü Chucai 耶律楚材, 357 Yi Xueshi 易學實, 388 Yidao pingwuqian currency, 135 Ying, Hubei, 102

Xi Chenghou 席成候, 437 Xia Mingfang 夏明方, 519 Xiajin county, Shandong, 332 Xiamen, Fujian, 665 Xian Gao 弦高, 150 Xian, Lord of Qin 秦獻公, 170 Xianbei people, 57, 82, 83, 122, 125, 154, 204, 211 Xiang Yu 項羽, 134 Xiang, Viscount of Zhao 趙襄子, 100 Xiangtan river valley, 554 Xiao Jiyu 蕭濟宇, 440 Xiao, Lord of Qin 秦孝公, 98, 170 Xiaoquan zhiyi coin, 135 Xiaoshan county, Zhejiang, 589 Xiaowen, Emperor of Northern Wei 魏孝文 帝, 82 Xiaozong, Emperor of Song 宋孝宗, 351–2 Xin dynasty, 118, 135 Xinjiang Uygur Autonomous Region, 62 Xintian. See Houma foundry, Shanxi Xiongnu people, 76, 79, 82, 114, 152, 203, 205–7, 208 Xi Xia, 342, 344, 572, 639, 647 Xixiang county, Shaanxi, 291 Xu Bo 徐伯, 74 Xu Dixin 许涤新, 613 Xu Guangqi 徐光啟, 270, 289, 385, 386, 399, 400, 492 Xu Panqing 许盘清, 328 Xu Xing 許行, 72 Xu Yang 徐揚, 550 Xu Yi 徐毅, 678 Xu Yikui 徐一夔, 464 Xuan, Emperor of Han 漢宣帝, 74 Xuanhua, Hebei, 646 Xuanwu, Emperor of Northern Wei 魏宣武 帝, 213

731

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press

Index Zhao Kai 趙開, 349 Zhao Shiqing 趙世卿, 407 Zhao Tianlin 趙天麟, 398 Zhao Tingchen 趙廷臣, 387–8 Zhao Xinchen 召信臣, 77 Zhao, Emperor of Han 漢昭帝, 192, 194 Zhejiang province, 155, 321, 333. See also Hangzhou, Zhejiang; Ningbo, Zhejiang Zhen Dexiu 真德秀, 394 Zhending county, Hebei, 314 Zheng (state), 68, 97, 150 Zheng Chenggong 鄭成功, 663 Zheng Dangshi 鄭當時, 74 Zheng Guo Canal, 61–3, 72, 74 Zheng He 鄭和, 363, 656–7 Zheng Xiang 鄭祥, 440 Zheng Zhilong 鄭芝龍, 662 Zheng regime, 316, 663 Zheng, King of Qin 秦鄭王, 72, 103 Zhenzong, Emperor of Song 宋真宗, 342 Zhezong, Emperor of Song 宋哲宗, 345 Zhi Loujiaqian 支婁迦讖, 231 Zhi Qian 支謙, 231 Zhida yinchao notes, 360 Zhongba, Chongqing, 44–8 Zhongshan (state), 36 Zhongshan Park, Beijing, 271 Zhongtong yuanbao notes, 359, 568, 570, 580 Zhou dynasty, 15. See also late pre-imperial China; Warring States period agriculture, 55–6, 66–8 cowry shells, 131 domain economy of the city-states, 94–6 redistributive economy, 93–4 Zhou Kongjiao 周孔教, 411 Zhou Mi 周密, 532 Zhoushan Islands, 577 Zhu Shuofo 竺朔佛, 231 Zhu Xi 朱熹, 408, 701 Zhu Yizhou 朱亦周, 437 Zhu Yuanzhang 朱元璋. See Hongwu, Emperor of Ming (Zhu Yuanzhang) Zhuge Liang 諸葛亮, 154 Zhujiajiao (market town), 607, 608 Zichan 子產, 99, 150 Zoroastrianism, 205, 227, 231–2 Zoumalou manuscripts, 120 zu–diao tax system, 83, 124 zu–yong–diao tax system, 126, 128, 142, 197, 199, 246

Yingpan (oasis state), 223 Yingshang county, Anhui, 591 Yishui county, Shandong, 330 Yiwan site, Jiangsu, 78 Yong, Shaanxi, 28 Yongle, Emperor of Ming 明永樂, 363, 533, 653–4, 656–7 Yongning Monastery, 220–1 Yongzheng, Emperor of Qing 清雍正, 291, 370, 375, 377, 460, 478, 665 Young Master of Accountancy, The, 190 Yu Bian 俞弁, 473 Yu the Great 大禹, 184, 705 Yuan Cai 袁采, 702 Yuan dynasty currency, 566, 568, 570, 574–5, 581 environmental changes, 293, 295 law, 423–4, 427, 429–31, 435, 439, 440, 442 political economy, 358–60 population changes, 304 topsoil ownership, 459 trade, 627, 648–51 urban economy, 532, 534, 540, 545, 548, 552 Yuan Zhancheng 元展成, 270–1 Yuan, Emperor of Han 漢元帝, 195 Yuanzhou, Guyuan, 230 Yuechi, Sichuan, 616 Yuegang, Fujian, 660 Yuelu Academy manuscripts, 107 Yuezhi people, 206, 210, 216 Yunnan province, 281, 288, 502, 665 Yushan, Jiangxi, 275–6 Yuwen Hu 宇文護, Emperor of Northern Zhou, 230 Yuwen Tai 宇文泰, 84, 125 Zandaniji silk, 226, 232 Zhan Yuanxiang 詹元相, 581 Zhang Fujing 張孚敬, 406 Zhang Juzheng 張居正, 365, 402 Zhang Mengtan 張孟談, 100 Zhang Qian 張騫, 206, 210 Zhang Shicheng 張士誠, 548 Zhang Sun 張遜, 640 Zhang Ying 張英, 461–2, 477 Zhang Zeduan 張擇端, 527 Zhang river, 61 Zhangjiashan manuscripts, 112 Zhao (state), 36, 100–1, 102, 132 Zhao Bian 趙抃, 410 Zhao Guo 趙過, 74, 76

732

https://doi.org/10.1017/9781108587334.022 Published online by Cambridge University Press