The Making of the Modern

Chinese State

The Making of the Modern Chinese State: 1600–1950 offers an historical analysis
of the formation of the modern Chinese state from the seventeenth to the
mid-twentieth centuries, providing refreshing and provocative interpretations
on almost every major issue regarding the rise of modern China.
This book explores the question of why today’s China is unlike any other
nation-state in size and structure. It inquires into the reasons behind the striking
continuity in China’s territorial and ethnic compositions over the past centuries,
and explicates the genesis and tenacity of the Chinese state as a highly cen-
tralized and unified regime that has been able to survive into the twenty-
first century. Its analysis centers on three key variables, namely geopolitical
strategy, fiscal constitution, and identity building, and it demonstrates how
they worked together to shape the outcome of state transformation in
modern China.
Enhanced by a selection of informative tables and illustrations, The Making
of the Modern Chinese State: 1600–1950 is ideal for undergraduates and gradu-
ates studying East Asian history, Chinese history, empires in Asia, and state
formation.

Huaiyin Li is Professor of History at the University of Texas at Austin. He is

the author of Village Governance in North China, 1875–1936; Village China
under Socialism and Reform: A Microhistory, 1948–2008; and Reinventing
Modern China: Imagination and Authenticity in Chinese Historical Writing.
The Making of the Modern
Chinese State
1600–1950

Huaiyin Li
First published 2020
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
52 Vanderbilt Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2020 Huaiyin Li
The right of Huaiyin Li to be identified as author of this work has
been asserted by him in accordance with Sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic, mechanical,
or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or
retrieval system, without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
Names: Li, Huaiyin, author.
Title: The making of the modern Chinese state 1600-1950 /
Huaiyin Li.
Description: Abingdon, Oxon ; New York, NY : Routledge, 2019. |
Includes bibliographical references and index. |
Identifiers: LCCN 2019014830 (print) | LCCN 2019019540
(ebook) | ISBN 9780429432071 (eBook) | ISBN 9781138362444
(hardback : alk. paper) | ISBN 9781138362451 (alk. paper) | ISBN
9780429432071 (ebk.)
Subjects: LCSH: China--Politics and government--1644-1912. |
China--Politics and government--1912-1949.
Classification: LCC DS754.17 (ebook) | LCC DS754.17 .L49 2019
(print) | DDC 951/.03--dc23
LC record available at https://lccn.loc.gov/2019014830

ISBN: 978-1-138-36244-4 (hbk)

ISBN: 978-1-138-36245-1 (pbk)
ISBN: 978-0-429-43207-1 (ebk)

Typeset in Galliard
by Swales & Willis, Exeter, Devon, UK
Contents

List of figures vii

List of tables viii
List of maps ix
Preface x

1 Introduction 1

PART I
The formation of the Qing state 21

2 The rise of an early modern territorial state: China in the

early to mid-Qing period 23

3 Limits to territorial expansion: fiscal constitution and

war-making under the Qing 52

PART II
The transition to a sovereign state 83

4 Regionalized centralism: the resilience and fragility of the

late Qing state 85

5 Between the frontier and the coast: geopolitical strategy

reoriented 112

6 A nation-state in the making: Manchu–Han relations under

the New Policies 132
vi Contents
PART III
The making of a unified and centralized state 161

7 Centralized regionalism: the rise of regional fiscal-military

states 163

8 In search of national unity: frontier rebuilding under the

Republic 192

9 The fate of semi-centralism: the nationalist state succeeded

and failed 216

10 Total centralism at work: the confluence of breakthroughs

in state-making 242

11 Conclusion 276

List of characters 289

References 298
Index 326
Figures

3.1 The Qing state’s cash reserves, war expenses, and tax
exemptions, 1644–1862 55
3.2 The low-level equilibrium in the fiscal constitution of the
Qing state 64
Tables

2.1 Land tax as a percentage of land yield, 1652–1820 38

2.2 All taxes collected as a percentage of total economic output,
1652–1820 39
3.1 The Qing state’s official revenue and expenditure, 1653–1840 54
3.2 Agricultural productivity during the Qing, 1600–1887 65
3.3 Agricultural production and population during the Qing,
1600–1887 66
4.1 The revenue and expenditure of the Qing state, 1841–1911 91
7.1 Revenue of the central government of the Republic, 1912–1945 166
7.2 Budgeted revenue and expenditure of individual provinces, 1925 171
Maps

I.1 China under the Qing, circa 1800 22

III.1 China under the Republic, circa 1920 162
Preface

This book is about how the Chinese state has become what it is today – or at
least what it was by 1949, if we do indeed believe that no significant changes
have occurred to it since then.
There have been two grand narratives about the making of the modern Chin-
ese state. One can be subsumed under the rubric of revolution. It is assumed
that the encroachment of foreign imperialism since the mid-nineteenth century,
coupled with the escalating tax burden imposed by various authorities through-
out the late Qing and Republican periods, resulted in the impoverishment and
dislocation of the rural population, and the rise of local bullies and ruffians in
village leadership; all these paved the way for recurrent peasant rebellions and
ultimately the rise of the Communist Revolution. The Japanese full-scale aggres-
sion after 1937 and the Chinese Communist Party (CCP)’s ability to penetrate
rural society only accelerated this process.
The other body of literature on modern China can be put under the category
of reform or modernization. Its arguments can be summarized as follows – at
the risk of oversimplification. The introduction of foreign culture, technology,
and capital and the expansion of foreign trade stimulated China’s economic
growth and enabled its progress in defense, industry, transportation, education,
urban development, and so on. Meanwhile, China also made headway in state
building, as seen in the experiments of representative democracy, the institution-
alization of government bodies, the internationalization of diplomatic practices,
and the recovery of its sovereignty by the end of World War II. The Communist
Revolution and the founding of a communist state, seen in this light, was an
anomaly rather than a necessary outcome of long-term structural developments
in late Qing and Republican China.
History-writing, as I argue elsewhere, is inseparable from the temporal con-
text in which historians identify the issues that concern them and offer interpret-
ations that reflect more of their ideological inclination and intellectual tradition
than the historical realities they attempt to reconstruct (H. Li 2013). Different
versions of the “revolution” narrative thus prevailed in China in the 1950s
through the early 1980s when legitimizing the CCP’s “mandate of heaven” to
rule China was historians’ chief duty, and to some degree it also influenced the
understanding of “Red China” in the West, where researchers attempted to
 Preface xi
build connections between China’s past experiences and the present realities.
The “modernization” narrative, for its part, came to reshape Chinese historiog-
raphy in the late 1980s and 1990s when China abandoned Maoism and
embraced the world with the programs of “Reform and Opening-Up,” and
when Chinese historians became eager to find historical precedents for the coun-
try’s ongoing modernizing endeavors. In the West (and also Taiwan), this narra-
tive emerged as early as the 1950s at the height of the Cold War and resurged
in the 1990s when neoliberalism came to dominate the thinking of global
geopolitics after the collapse of the Soviet Union. Both narratives, however,
are flawed by an ideology-driven teleology. While the revolution thesis sees
China’s transition to socialism as the inevitable end of its modern history,
the modernization construct interprets Western-style representative democracy
and free-market capitalism as the ultimate destination of China’s modern
development.
This book offers an alternative, open-ended account of the making of the
modern Chinese state. I reinterpret the rise of modern China as a process of
state transformation that is longer in time span and wider in scope than both
revolution and modernization, and which remains unfinished at the present
time. To put it briefly, the Chinese state as it was by the mid-twentieth cen-
tury was the accumulative result of state-rebuilding that had lasted for three
centuries and consisted of three key breakthroughs: the formation of an early
modern territorial state by the mid-eighteenth century that had laid
a territorial and ethnic foundation for the making of modern China; its transi-
tion to a sovereign state in the late nineteenth century to become integrated
with the global system of nation-states; and its gradual transformation into
a centralized and unified state in the first half of the twentieth century.
China redefined and rebuilt itself in each of these three stages; developments
in each stage shaped the direction and dynamics of state-rebuilding in the
next stage. In two important ways, the making of the modern Chinese state
differed from the experiences of Western countries and many other non-
Western societies.
First, instead of the “normative” path of transition “from empire to nation-
state” commonly found in the literature on state-building in the modern world
(i.e., the emergence of multiple nation-states on the ruins of a former military
or colonial empire following its prolonged decline and eventual loss or splitting
of territories), China witnessed striking continuities in its territorial, demo-
graphic, and administrative patterns throughout the three-century process of
state transformation. The key to understanding this puzzle, I shall argue, lies in
the fact that the formation of the Qing state after inaugurating its rule in China
departed fundamentally from empire-building in other parts of the early modern
world, as evidenced in the motivation behind its expansion, the fiscal mechanism
that bolstered its military operations, and the way it governed the frontiers, as
the individual chapters of this book will document. The Qing, in other words,
was not an empire, but instead an “early-modern territorial state” as I would
term it. State-making in China means first of all its painstaking yet successful
xii Preface
transformation from a territorial state into a sovereign state, rather than the dis-
ruptive transition from an empire to a nation-state. Behind the territorial con-
tinuity were the impressive records of the Qing in maintaining the frontiers and
dealing with the unprecedented crises at home and abroad by rebuilding its
fiscal, military, and administrative systems, for all the frustrations and failures in
the process.
Second, state transformation in late Qing and Republican China took the
form of a protracted struggle between regionalism and centralism in the
restructuring of power relations between central and local/regional author-
ities, culminating in the creation of a highly centralized, authoritarian state
by the mid-twentieth century. The historical root of the high-level centralism
that features the Chinese state today can be traced to the unique mode of
geopolitical and fiscal constitutions of the Qing, termed “low-level equilib-
rium” in this book. This denotes the Qing state’s unparalleled geopolitical
security and fiscal sufficiency. Temporary and conditional as it was, this equi-
librium prevailed throughout most of the eighteenth century, thanks to the
presence of a peculiar combination of international, demographic, and finan-
cial factors. The vulnerability of this equilibrium and the Qing state’s subse-
quent unpreparedness for the advent of domestic and foreign crises in the
nineteenth century, however, gave rise to a new form of power relations, dubbed
“regionalized centralism,” which in turn paved the way for the emergence of “cen-
tralized regionalism” in the early Republican years, “semi-centralism” of the
Nationalist state, and “total centralism” of the Communist state, to be explicated in
each of the chapters that follow. Each of these prevailed over its predecessor
because of its higher level of centralization and greater competitiveness in the
rivalry for dominance in China. This dynamism again distinguishes China from
the historical paths of many nation-states in the West that ended in creating
a government by the principles of representative democracy that ideally
manifests popular sovereignty, and from the experiences of many other non-
Western countries whose state-building centered on a nationalist movement
against colonialism and imperialism.
Finally, it is worth emphasizing that the making of the modern Chinese state,
a process beginning in the 1640s, remains unfinished by the early twenty-first
century. While the contemporary Chinese state has shown impressive stability
and vitality since its founding in the mid-twentieth century, the legacies and
deficiencies it inherited from the aforementioned three breakthroughs continue
to empower as well as challenge China and will shape its future trajectory of
evolution. Unlike the revolution or the modernization narrative of the twentieth
century that has a definite, closed ending, the thesis of state transformation pre-
sented in this book offers us a different angle to comprehend the issues of
China’s uncertainties and potentialities as a modern state in the twenty-first cen-
tury, an age of accelerated globalization in which its impact on the rest of the
world is bigger than ever before.
My writing of this book has benefited from two bodies of scholarly work. On
the theoretical or explanatory level, this project has been inspired by the
 Preface xiii
literature on state-making and, more recently, on the fiscal-military states in
early modern and modern Europe; I found discussions on the roles of geopolit-
ics, war-making, and fiscal institutions in state formation particularly relevant to
this project. On the empirical level, while much of my research is based on
my own use of original source material, it is impossible for me to delve into
such sources for every topic examined in this book, given the breadth of the
entire project. Therefore, I have relied heavily on the findings that historians
of different sub-fields in late imperial and modern China have presented in
their publications over the past few decades, which I cite throughout this
book. In the past few years, I have been fortunate to have direct exchanges
with the most prominent scholars in China in each of these sub-fields and
to give talks on my research at their institutions. Among them are Professors
Chen Feng and Zuo Songtao of Wuhan University; Professors Lu Xiaobo,
Li Lifeng, Sun Jiang, Kong Fanbin, and Zhang Sheng of Nanjing University;
Professor Jin Guangyao of Fudan University; Professors Wang Qisheng,
Gao Shiyu, and Niu Ke of Peking University; Professor Wang Xutian of
Renmin University of China; Professors Zhong Weimin and Ni Yuping
of Tsinghua University; Professors Wang Xianming, Yu Xinzhong, and Jiang
Pei of Nankai University; Professor Liu Jiafeng of Shandong University; Pro-
fessor Li Deying of Sichuan University; Professors Liu Zenghe and Zeng
Guangguang of Jinan University; Professors Liu Yonghua, Lu Xiqi, Zhang
Kai, and Zheng Zhenman of Xiamen University; Professors Xie Shi and
Zhao Libing of Sun Yat-Sen University; Professor Liu Chang of East China
Normal University; Professors Zhu Ying, Xu Xiaoqing, and Jiang Manqing
of Central China Normal University; Professors Yang Songtao and Zhang
Xiangdong of Henan University; Professors Xing Long, Hu Yingze, and
Chang Libing of Shanxi University; and Professors Zhao Xiaoyang, Li Changli,
Ma Yong, and Zuo Yuhe of the Institute of Modern Chinese, Chinese Acad-
emy of Social Sciences. My sincere thanks to all of them for their hospitality
and the many good questions and insights on the topics we discussed. This
book originated from two papers I read at the conferences of the Social Sci-
ence History Association in 2010 and the Association for Asian Studies in
2011. Part of the book manuscript was also presented at the conference on
“The Inner and Outside of Historic China” at Beijing Normal University in
2016. I thank the organizers and discussants of these conferences and panels
for their invitations and comments on my papers; among them are Professors
Mark Elliot, Siyen Fei, Wenkai He, James Lee, and R. Bin Wong. For their
comments and suggestions on the book manuscript, I would like to thank Pro-
fessors Marc Blecher, Vivienne Shue, and my colleagues who worked in the
History Department at the University of Texas at Austin, including Professors
Anthony Hopkins and Mark Metzler, as well as my former and current PhD stu-
dents John Harney (Centre College), James Hudson (UNC), Zhaojin Zeng (Pitt),
Jing Zhai, Guangji Hu, He Tan, Benjamin Yeager, and Fei Guo. Comments by
the anonymous reviewers for Routledge helped me sharpen my arguments and
refine the structure of the entire project.
xiv Preface
Finally, and most importantly, I thank my wife Gwen Zhao and our two chil-
dren, Cathy and Daniel, whose many questions about China in their journeys
from middle or high schoolers to longhorns of UT Austin motivated my writing
of this book. This book is for all of you.
Huaiyin Li
Austin, Texas
August 2019
1 Introduction

The modern Chinese state as it exists today, namely the People’s Republic of
China (PRC), is unlike any other nation-state in the world in size and struc-
ture. Physically, it is big. China has not only one of the largest territories,
crossing from the northeastern provinces to central Asia and from Inner Mon-
golia down to the tropical Hainan islands, but also the very largest population
with diverse ethnic backgrounds, encompassing the Han people as well as the
Mongols, Uighurs, Tibetans, and other ethnic groups. More importantly,
the composition of its land and peoples is directly traceable to the legacies of
the Qing that conquered China in the 1640s; China, therefore, is believed to
be the only modern state today that has inherited its territorial and ethnic pat-
terns from a former “empire” (Esherick 2006: 229), if that term applies to its
last dynasty.
Equally important about the Chinese state is that its power structure is
highly centralized and proves to be exceptionally durable. Since it won the
Civil War against the Nationalist regime in 1949, the Chinese Communist
Party (CCP) has exclusively and continuously governed the country, and its
functionality explains in large part China’s emergence as a global power by
the beginning of the twenty-first century. There are no signs indicating the
state’s demise in the foreseeable future, for all the challenges and turbulences
confronting it over the past decades. This reality again distinguishes China
from the vast majority of nation-states that have developed or at least consti-
tutionally adopted a democracy in the Western sense, from many other Third
World countries that have been plagued with a “failed” state, and from all
other communist states that are either short-lived (the Soviet Union and its
satellite states in East Europe) or small in size (North Korea, Vietnam, Laos,
and Cuba).
Why, then, did the centuries of political transformation in China end in the
rise of an enormous state under an authoritarian government that remains as
robust as it appears in the twenty-first century? To what extent did the Chinese
experiences in state-building resemble or deviate from the trajectories of other
nation-states? Has the modern Chinese state taken a definite shape, or will it
continue to evolve in the future? These are the central issues to be pursued
throughout this study. But before turning to these questions, let us first briefly
2 Introduction
look at how the Chinese state is interpreted in the existing literature in the
Chinese and English languages.

The modern Chinese state: why “big” and “strong”?

Contrasting interpretations
The CCP’s historiography offers a straightforward explanation of the origins of
the state it inherited and rebuilt. Its basic assumptions are as follows. First,
China is a “unified, multiethnic” state because, for more than two thousand
years since the Qin (221–206 BC) and the Han (206 BC–AD 220), unified
dynasties have predominated in its history; its historical territory covered areas
inhabited by both the Han and non-Han peoples under such dynasties; and the
Qing, as the last unified, multiethnic dynasty, was the culmination of the
long tradition of China’s territorial unification and ethnic integration (e.g., Tan
Qixiang 1982; Fan Wenlan et al. 2008). Second, the CCP’s leadership in China
was the necessary outcome of a century-long revolution by the Chinese people
against feudalism and imperialism, which began with their resistance against
British aggression during the Opium War of 1840, continued in the Taiping
Rebellion, the Revolution of 1911, and the Boxer Rebellion, but failed one
after another until the CCP came into being in 1921 and took over the leader-
ship of the revolution and made it a success in 1949, after overthrowing the
corrupt and incompetent Nationalist (Guomindang or GMD) regime through
a civil war (e.g., Hu Sheng 1981). Unification and revolution thus constitute
the two overarching themes in the CCP’s narratives of China’s imperial and
more recent histories; these narratives, well formulated in standard history text-
books in today’s China, well serve the purpose of legitimizing the political
system built and maintained by the Party (H. Li 2013).
Yet, in the eyes of many observers in the West, it is precisely the combination
of the two basic characteristics (i.e., being simultaneously “big” and “strong”)
of the contemporary Chinese state that make its legitimacy and viability prob-
lematic. For decades, interpretations on the making of modern states in the
English-language literature have tended to juxtapose empire and nation-state as
the two contrasting forms of polity. An empire is typically associated with war-
like propensities and territorial expansion; it is therefore demographically multi-
ethnic and culturally heterogeneous; ideologically it is cosmopolitan, claiming
the universality of its ideas and institutions throughout the world; and politically
it maintains indirect rule over colonies, dependencies, and tributaries.1 In sharp
contrast, a nation-state, according to interpretations based mainly on the histor-
ical experiences of state-making in Western Europe, is supposed to have the fol-
lowing three traits. First, it is characterized by the identification of its people
with the land they inhabit; ideally, the territory of a nation-state corresponds to
the geographic area of the people who have shared cultural and political tradi-
tions and/or ethnic origins. Second, its ideology, nationalist in nature, empha-
sizes the particularities of their ethnic origins and/or the glories of their shared
 Introduction 3
experiences; as a sovereign state, it claims equality with all other nations under
the international laws and exclusive rights on its territory that has fixed and
clearly demarcated borders with other states. Third and no less important, the
sovereignty of a nation-state is vested in its people rather than the monarch, and
the ideal form of government is a rights-based state as the liberal thinkers of
seventeenth and eighteenth-centuries Europe envisioned or a representative
democracy that prevailed increasingly in both the Western and non-Western
worlds.2 Modern state-making, therefore, is largely equated with a linear pro-
gression from empire to nation-state.3
The historical realities of state formation in the modern world, including the
past of European states per se, are much more complex than the teleology
above suggests.4 China’s experiences in building a modern state, in particular,
contradict the above assumptions in two important ways. First, instead of the
breakdown of a former empire and the emergence of multiple nation-states on
its ruins as commonly seen after the fall of all other empires, what characterized
China is the striking continuity between the Qing and the Republic found in
1912 (as well as the PRC) in territorial and ethnic compositions (Zhang Yongle
2016). Second, instead of establishing a democracy that presumably best exem-
plifies popular sovereignty, what has prevailed in China since 1949 is
a government system that is constitutionally different from the polities in many
other countries. Given these conspicuous “anomalies” of the Chinese state, it is
not surprising that many have predicted the inevitable splitting of China’s terri-
tory, the collapse of the communist state, and its eventual transition to
a Western-style democracy in the future.5

Three major issues

To answer why the Chinese state is unusually “big” and “strong” or whether or
not it is historically legitimate, we should first of all avoid quick judgment based
on any theoretical presumptions derived from the experiences of European
countries. Nor should we limit our attention to the period after 1840, a turning
point signaling the coming of foreign influences as a driver behind a wide spec-
trum of changes in late Qing and Republican China. Instead, we should think
of state formation in China as a longer process that started much earlier than
the mid-nineteenth century.6 Compared to the past studies that have addressed
aspects of the modern Chinese state, this work offers a systematic account of the
making of modern China from the seventeenth to the twentieth centuries. Central
to my concern is the formation of the three key features of China as a modern
state, namely its territory, sovereignty, and government system. More specifically,
this study centers on the following three issues.
(1) The territorial expansion of the Qing and the nature of its rule in China
in the seventeenth and eighteenth centuries.
The origins of the twentieth-century Chinese state can be traced to the
Qin and Han dynasties in term of its territorial coverage and governing insti-
tutions. From the Qin down to the Ming dynasty (1368–1644), much of the
4 Introduction
central features of the Chinese state, such as centralized bureaucracy and the
geographic scope of its rule, persisted by and large for all the developments
throughout the dynasties in between. But the Chinese state today is unlike
any of its predecessors prior to the Qing in that its territory encompasses not
only the interior provinces inhabited by predominantly the Han people but
also the frontiers of the Manchus, Mongols, Uighurs, Tibetans, and so on. It
was during the Qing that China underwent a transition from a largely mono-
ethnic state of the Han people to a multiethnic state comprising both the
Han provinces and the non-Han frontiers. The first set of questions that we
have to explain in this book thus is: why did the Qing expand? What is the
peculiar fiscal-military mechanism that enabled as well as constrained its
expansion? How did the Qing rule China? Was the Qing an empire or some-
thing else? These questions are important because the Qing laid the territorial
and demographic foundations on which the modern Chinese state was estab-
lished in the twentieth century. An exploration of these questions (Chapters
2–3) is essential to our understanding of the origins of China as a modern
state and the nature of its territoriality.
(2) China’s transformation into a sovereign state and its integration with the
global system of states in the nineteenth century.
Given its conventional role as the only dominant power claiming cultural and
political superiority over all the other states around it, China’s involvement in
the Europe-centered system of states was particularly difficult and prolonged.
For China, the most challenging task in this transition was to renounce its self-
claimed centrality in the world order, to accept all other states as its equals, and
to give up its suzerainty over the surrounding tributary states and acknowledge
their independence. Needless to say, China would not do so until it came under
irresistible pressure from outside, which took the form of its repeated defeat by
foreign powers in the nineteenth century. Thus, before it became a sovereign
state, China had to concede to the demands of the powers that had subjugated
it militarily, such as extraterritoriality, fixed tariff rates, cession of land, and the
one-sided most-favored-nation status granted to foreign powers. Nevertheless,
China was among the few (others include Japan, Siam/Thailand, Persia/Iran,
and Abyssinia/Ethiopia) that avoided conquest by colonial powers and survived
imperialism in its modern era. Even more surprisingly, unlike all other empires
that invariably ended in territorial splitting-up into multiple independent states,
the Qing succeeded in preserving most of its territory, including the frontiers,
until the very end, which it then passed on in entirety to its heir, the Republic
of China. Why was late Qing China able to resist the threats of foreign powers
and keep its land largely intact? How did the Qing redefine its geopolitical strat-
egies and rebuild its fiscal-military institutions in response to external and
internal challenges? And how did changes in the relationship between the ruling
court and provincial authorities explain the viability as well as vulnerability of
the Qing state in the late nineteenth and early twentieth centuries? We cannot
understand the trends and outcome of state rebuilding after the fall of the Qing
without a thorough analysis of all these questions (Chapters 4–6).
 Introduction 5
(3) The making of a unified and centralized state in the Republican period.
Influenced by the various doctrines of nationalism and, in particular, the idea of
popular sovereignty, and inspired by the precedents of American and French
Revolutions, almost all national revolutionaries of different ethnicities in the
modern world aspired to establish a government on their land that not only had
complete sovereignty over its territory but also took the form of representative
democracy. The Chinese state-makers were no exception. The revolutionaries
of the late Qing period and later both the GMD and the CCP were all committed
to establishing a republic in China. But the democracy prescribed in the constitu-
tions of the Republic of China never became truly functional; it was frequently
compromised by dictatorship under the Republican government in Beijing
(1912–1927) and yielded to one-party rule under the Nationalist regime
(1927–1949). The Communists who defeated the Nationalists in 1949 defined
the newly established PRC as a socialist state under “people’s democratic dictator-
ship,” a new form of government influenced by the Soviet Union and character-
ized by high-level centralization of power under the sole control of the CCP.
What concerns us here is not the question why the Communist revolution suc-
ceeded, but rather why the centuries of state rebuilding ended in the formation of
a highly centralized regime instead of Western-style democracy. While an in-depth
inquiry into these questions (Chapters 7, 9, and 10) will shed light on the dynam-
ics behind the rise of a unified and centralized state in the first half of the twentieth
century, a discussion on frontier rebuilding under the Republic (Chapter 8) is
necessary for understanding the striking continuity in China’s territoriality.
State transformation in modern China, in a nutshell, is a coherent process consist-
ing of three key links, namely (1) territorial expansion in the early-to-mid Qing
period, (2) adaptation to the world system of states in the late Qing period, and (3)
consolidation of state power in the Republican period. These three major break-
throughs in state rebuilding were historically and logically linked. The successive
expansions in the late seventeenth and early eighteenth centuries resulted in the
enlargement and redefinition of China, which generated a geographic and demo-
graphic basis on which the modern Chinese state defined its territoriality and sover-
eignty. In its subsequent integration into the global system of states in the
nineteenth century, China suffered damages to its territorial integrity and sovereignty
but nevertheless succeeded in preserving its territorial and demographic settings, des-
pite the tilting of the power structure of the state towards the regional level. State
rebuilding in twentieth-century China, therefore, meant the consolidation of its ter-
ritorial integrity and the recentralization of the state apparatus. It is, therefore, the
combined effects of the three developments over a three-century process of state for-
mation and transformation that account for the physical size and structural strength
that the Chinese leviathan had developed by the mid-twentieth century.

Geopolitics, finance, and identity: a macrohistorical analysis

This study employs a macrohistorical approach to examine China’s three-century
process of state rebuilding. Macrohistory refers to the analysis of a long-term
6 Introduction
historical process that has a wide and profound impact on society or a human
group. However, unlike the general history of a nation or society that covers all
aspects of human activity, macrohistory seeks to identify the patterns or “laws”
of historical trends or a particular process of historical significance, and its ana-
lysis centers on a limited number of variables that shape the dynamics and pat-
terns of that process. A macrohistorical study, therefore, tends to adopt
a comparative or world-historical perspective to inquire into the causes of
change and paths of development characteristic of the society or process under
investigation. Furthermore, macrohistory does not respect borders in time,
especially periodization based on regime change; instead, it looks for connec-
tions in human activities over a long time range that form a coherent process
with its own dynamics.7 In this study, my analysis of state transformation in
China centers on three variables; namely geopolitical setting, fiscal constitu-
tion, and identity building.

Geopolitical setting
Geopolitics is first of all about a state’s strength and position in relation to other
powers and the impact of its foreign relations on domestic politics. In early
modern and modern Europe, a horizontal pattern of interstate relations pre-
vailed after the 1648 Peace of Westphalia, where the individual states saw one
another as equals legally and diplomatically, and they “always appear in competi-
tion with each other, and gain their identities by contrast with rival states”
(Tilly 1990: 23). Countries with a greater degree of geographic exposure and
hence facing greater threat of land warfare had the greater urgency to build
a standing army and a bulky bureaucracy under an absolutist state to withstand
that threat, whereas a maritime power such as England, free of the sustained
military pressure that confronted continental powers, was able to do away with
a standing army and develop parliamentarism and self-government (Roberts
1967; Hintze 1975). The fierce competition among the various forms of polit-
ical entity resulted in recurrent wars between individual states or groups of
states, causing them to decrease in number, from roughly 200 in 1490 to about
30 in 1890, and to enlarge in size, with the average size of European countries
growing from 9,500 to 63,000 square miles during the same period (Tilly
1990: 42–47). It is true indeed that “war made the state, and the state made
war” in European history (Tilly 1975: 42).
War-making, and the lack of it, is also key to understanding the dynamics
of state transformation in China. From the seventeenth to the nineteenth cen-
turies, China’s geopolitical relations with its neighbors and rivals underwent
drastic changes, involving (1) the Qing state’s competition with the Zunghar
Mongols of Central Asia that led to the expansion of its Inner Asian frontiers
by the mid-eighteenth century; (2) its dominance within a tributary system it
maintained with neighboring states before the mid-nineteenth century; and
(3) rebuilding of its defenses and restructuring of territoriality in overcoming
the crises in foreign relations in the late Qing period. Challenges and
 Introduction 7
opportunities in China’s geopolitical settings also fundamentally drove state
rebuilding in the first half of the twentieth century, which took the form of
fierce competition among the regional forces that rivaled for national domin-
ance. Exactly how the imperatives in China’s geopolitical strategy triggered the
major events in state rebuilding and where China resembled or differed from
other states and empires in geopolitical strategizing will be one of the focal
points in this study.

Fiscal constitution
Whether or not a state is able to achieve its strategic objectives, however,
depends in large measure on its fiscal and military capabilities, which in turn
reflect the extractability of its economy and the state’s ability to turn the
extracted resources into real military strength. Among the various methods of
extraction, the most important is taxation, which takes the form of collecting
direct taxes on the land and indirect taxes on goods and service. The state can
also use financial tools for extra revenue, including currency inflation, issuance
of government bonds, or loan borrowing. Additionally, the state generates
income by selling government positions or honorary titles. Finally, military
expansion and conquest for new territories, hence taxation of the conquered
land and demand for tributes from the dependencies, are the most effective
means for empires to increase wealth.
The different levels and means of extraction have different impacts on the for-
mation of the state apparatus. It is widely assumed that taxation on the land is
highly difficult and costly; rulers of early modern European states, therefore,
preferred the convenience of farming out tax collection to entrepreneurial
agents, whose predatory and exorbitant methods in collecting taxes often
incurred resistance from taxpayers. Rural resistance in turn necessitated the
state’s use of coercion to generate the needed revenues, hence the rise of abso-
lutist regimes or the coercion-intensive path to state formation. On the other
hand, taxing commerce and the wealth of landed elites or borrowing from finan-
ciers with their consent are relatively easy in a commercialized economy, but
bargaining with the wealthy also meant the rulers dispensing with an extensive
bureaucratic system and coercive means of exaction, which provided a basis for
the urban elites to participate in government and hence the growth of
a constitutional state, or the capital-intensive path to state formation (Tilly
1985; M. Mann 1980, 1986: 450–499; Downing 1992: 3–17).8
Generating enough revenue for strategic purposes is also a central task for
builders and rulers of the Chinese state as well as their challengers during the
centuries under consideration. The sources of revenues for each of them varied
widely, ranging from predominantly the land tax under the Qing before the
mid-nineteenth century and in the Communist base areas in the 1930s and
1940s, to taxes on trade and industry, which became increasingly important in
the late Qing and Republican periods, and banking and financing, which were
critical to the Nationalist state. How much revenue they generated and how
8 Introduction
they obtained it directly shaped their approaches to governance and strength in
competing with their adversaries.

Identity building
Identity building here means not only the legitimization of a regime through
the propagation of its ideology or the forging of consensus on its goals among
the social and political groups under its rule, but more importantly the opti-
mization of relationships between the state’s power center and its regional agen-
cies, among regions of different interests under its rule, and among people of
different ethnic origins and religious or political persuasions on its territories.
How efficient the fiscal-military resources are redistributed at different levels and
in different regions and how effective they are used for the state’s strategic
goals, therefore, depend in large part on the level of identity among the individ-
uals or groups involved in the redistribution.
In the process of state rebuilding in Qing and Republican China, we can
identify three redistribution patterns. In the first, as seen in the Qing before
the mid-nineteenth century, the central government monopolized almost all
resources, leaving local authorities devoid of any substantial means at their
disposal and therefore completely submissive to the center. In the second, the
reverse is true: the redistribution was highly decentralized, resulting in a weak
central government and strong regional forces; the latter controlled the rev-
enues and the military within their areas and contended among themselves or
with the central government for more power and wealth, a situation prevail-
ing in early Republican years. In the third, the distribution of resources fell
between the two extremes; both the center and its local contenders controlled
a certain amount of revenue and armed forces, and the balance between the
two sides shifted constantly, as seen in the late Qing period and under the
Nationalist regime.
Much of the redistribution of fiscal and military capabilities in different
periods of modern China reflected the level of identity building among the dif-
ferent political forces at play; the higher the level of political identity was
achieved, the more likely the fiscal and military power could be centralized,
and the more powerful the central government would be in rivalry with its
challengers inside and outside its territory, and vice versa. On the other hand,
a regional force could also be a formidable challenger to the authority above
it, if it not only controlled the resourses within the region or group but also
built strong identity among group members, as we will find in the early history
of the Nationalist force in the 1920s and in the rise of the Communist force in
the 1940s.
To recapitulate, while external or internal imperatives conditioned the formu-
lation of geopolitical priorities of any political entity (the state or a regional
competitor) in Qing and Republican China, whether or not it was able to
achieve its strategic goals, hinged on the fiscal resources it generated (hence its
military strength) as well as the level of political identity it built among its
 Introduction 9
supporters. These three key variables – geopolitics, finance, and identity – inter-
played to propel state transformation in modern China.

Preliminary arguments

The formation of an early modern territorial state

This book has three parts, each covering one of the three major breakthroughs
in the making of modern China. Part I examines the formation of the Qing
state in the seventeen and eighteenth centuries. It was the Qing, after all, that
laid the territorial, ethnic and even administrative foundations by the mid-
eighteenth century on which the modern Chinese state was established. Given
that the Qing originated as an “alien” dynasty of the Manchus, establishing its
rule in China after conquering the Ming, and that it further extended its control
over Outer Mongolia, Tibet, and Xinjiang after a series of wars on the frontiers,
it comes as no surprise that the Qing is often equated with a conquest dynasty
arguably showing parallels in state formation to expansionist empires in other
parts of the world. Chapter 2 challenges this received wisdom. It shows that, in
several important ways, the Qing is fundamentally different from other major
empires in the Eurasian continent. First, after establishing its rule in China, the
Qing limited its expansion of Inner Asian frontiers (i.e., Outer Mongolia, Tibet,
and Xinjiang) only to the period from the 1690s to the 1750s. There was no
military operation to seek frontier expansion in the five decades before the
1690s or the one and a half centuries after the 1750s, though wars of varying
scales for frontier consolidations occurred from time to time before and after
that interim. In other words, military conquest and territorial expansion were
the exceptions rather than the norm in the history of the Qing, unlike other
empires in the early modern and modern world that constantly engaged in wars
throughout their existence.
Second, unlike its counterparts in premodern or early modern Eurasia who
fought wars for the purposes of either increasing wealth or spreading a religion,
the Qing established its Inner Asian frontiers for none of these reasons. Instead,
its frontier expansions were the unexpected results of its resistance to the
repeated fatal threats from the Zunghar Khanate. The Qing launched a series of
expeditions into the Gobi Desert in the 1690s only after the Zunghars invaded
Outer and Inner Mongolia, thus constituting an immediate threat to Beijing,
the capital city of the Qing, and to its alliance with the Mongols south of the
Gobi Desert, which had enabled the Qing conquest of China in the 1640s and
remained central to the Qing’s geopolitical security afterwards. Later, the Qing
launched expeditions into Tibet and Xinjiang in the early to mid-eighteenth
century only after the Zunghars invaded Tibet, threatening to sever the trad-
itional political and religious ties between Lhasa and Beijing and thus undermine
the Qing court’s patronage of Lamaism, which, together with intermarriage
between Manchu and Mongol nobles, was essential for maintaining the
Manchu–Mongol alliance. Therefore, the primary goal for the Qing to rule the
10 Introduction
newly established frontiers was to safeguard its strategic security; not surpris-
ingly, instead of demanding payment of taxes or tributes from the frontiers, it
was content with establishing direct military control of the frontiers by station-
ing a standing army at the key points of the frontiers, allowing the religious and
secular elites of the frontiers full autonomy to manage their internal governing
affairs, and patronizing Lamaism yet at the same time containing the influences
of the separate Lama leaders in Tibet and Mongolia by practicing the “divide
and rule” policy. All these explain why the Qing was able to maintain its stable
control of the frontiers all the way until its demise in 1911, which contrasted
sharply with other contemporary empires whose frontiers were in constant flux.
Third and most important, the Qing cannot be likened to an expansionist
empire chiefly because it was fiscally self-sufficient. Three factors contributed to
the Qing state’s financial sufficiency, namely (1) the huge population of the
interior provinces, which constituted an immense basis from which it generated
a revenue more than it needed during times of peace; (2) the geopolitical secur-
ity that kept its military spending at a stable and low level; and (3) the informal
mechanisms of local governance that minimized the cost of government. The
goal of its geopolitical strategy, therefore, was inward-oriented, that is, to per-
petuate its rule in China by maintaining its legitimacy in the interior provinces
of the Han people and, to that end, demonstrating itself as a dynasty embracing
the institutions and ideology of the preceding dynasties of the Han Chinese, as
best seen in its unfailing adherence to the policy of light taxation on the land
throughout its history in accordance with the Confucian tradition of “benevo-
lence” in governance. The interior provinces and the frontiers were very differ-
ent in terms of their importance to the Qing: the former was where the dynasty
maintained its status as a legitimate regime ruling Zhongguo or China and where
it obtained almost all of its fiscal resources, and the latter only served as instru-
ments to enhance its strategic security and control of China. These historical
realities contradict the assumption that the Qing was an “Inner Asian empire”
under a “universalistic” rulership which treated its Inner Asian constituencies on
equal terms with the interior provinces.
Nor can we liken the Qing to the preceding Chinese dynasties and simply
equate it with yet another dynasty in the long series of dynastic cycles in China’s
imperial past. After all, the Manchu conquest resulted in the enlargement and
redefinition of Zhongguo from an essentially monoethnic state of the Han people
that was largely limited to the land south of the Great Wall to a larger state
encompassing both the interior of the Han population and the frontiers of
Inner Asian nomads. Therefore, the conventional Sinocentric view that sees
non-Han peoples as “ethnic minorities” and marginalizes their roles in the for-
mation of the Qing state downplays the most significant and defining part of
Qing history. The Qing did indeed inherit from the preceding Ming a highly
centralized bureaucracy that was unmistakably “Chinese” in terms of the overall
administrative structure and the political culture that undergirded it; yet what
made the Qing unique was its hybridity, most evident in the Manchu-Han
diarchy in the higher levels of the bureaucratic hierarchy and the complementary
 Introduction 11
relationship between the interior and frontiers unseen in the previous Chinese
dynasties.
Chapter 3 takes a deeper look into the formation of the Qing state by
focusing on its fiscal constitution in order to understand how the peculiar
mechanism in its revenue generation and expending empowered as well as
constrained the dynasty’s frontier expansion. Central to my analysis is the
concept of “low-level equilibrium,” which distinguishes the Qing from all
other states in the early modern world and explains its fate in the nineteenth
century. Unlike the “fiscal-military states” in early modern Europe whose rev-
enue structure was highly dynamic and full of the potentials to expand (partly
because of the ever-growing industry and commerce that formed the major
sources of tax revenues and partly because of the ever-escalating demands for
revenue for war efforts and an enlarged bureaucracy), the fiscal system of
Qing China before the mid-nineteenth century was largely static, with a rigid
structure of regular revenues primarily from the land and regular expenditures
that remained relatively stable over time. Two factors made this inelastic
structure possible. The first is the Qing’s geopolitical security that allowed for
a stable and low military spending, and the second is an optimal ratio of
population to the land that allowed for a huge source of revenue from the
large number of taxpayers, even though the tax rate was kept at a very low
level. The coexistence of these two preconditions resulted in the prevalence of
a delicate equilibrium in its fiscal structure (i.e., a fixed regular revenue that
was slightly highly than a fixed regular spending). When this equilibrium
existed, the Qing state had the advantage of an enormous amount of cash
reserves accumulated year after year because of its fixed spending on the one
hand and an annual surplus on the other in the times of peace; it was this cash
reserve that enabled the Qing to expand in wartime. However, neither the Qing
state’s geopolitical supremacy nor the size of its population was a given, and the
equilibrium that was based on them was vulnerable and would be lost whenever
any of the two preconditions disappeared. It was, therefore, the low-level and
eventual loss of this equilibrium that ultimately shaped China’s course of trans-
formation in the nineteenth century when both preconditions vanished.
The Qing state, in the final analysis, is neither a typical expansionist empire
nor an emerging fiscal-military state resembling its counterparts in early
modern Europe and beyond. Instead, it can be best described as an early
modern territorial state. The Qing was a “territorial state” because it had
stable frontiers and effectively controlled its territory that had fixed borders
clearly demarcated with the neighboring states, though it was yet to be
incorporated into the system of sovereign states that had prevailed in Europe
since the seventeenth century. The Qing state was “early modern” because it
possessed full-fledged state apparatuses such as a standing army, a centralized
and specialized bureaucracy, and an institutionalized tax-collecting system,
which the national states in Europe also gradually developed in the seven-
teenth and eighteenth centuries.
12 Introduction
The transition to a sovereign state
Part II scrutinizes the second breakthrough in the making of the modern Chin-
ese state, namely, the transformation of the Qing from an early modern territor-
ial state into a sovereign state in the late nineteenth and early twentieth
centuries. Conventional historiography typically depicts this period as a time of
the Qing’s decline. The late Qing state did indeed suffer repeated defeats at the
hands of foreign powers, the subsequent loss of part of its territory, and dam-
ages to its sovereignty. Yet, unlike any other empires whose decline invariably
ended in territorial splitting, and unlike most non-Western states whose encoun-
ters with colonialism all ended in their loss of independence and the establish-
ment of colonial rule, the Qing was among the few in the non-Western world
and the only non-Western “empire” (so to put it) that successfully defended
most of its territory and maintained its independence until its sudden demise in
1911. By that point, China had reinvented itself as a sovereign state, kept its
Inner Asian frontiers largely intact, and achieved a substantial level of military
and government modernization.
Thus, unlike past studies on late-nineteenth-century China that have tended
to center on the issue of why the late Qing state “failed,” Chapter 4 explicates
its impressive resilience by first inquiring into its fiscal constitution. Central to
my analysis is the transformation of the late Qing fiscal constitution from one of
low-level equilibrium, which lacked elasticity and expandability on both the
demand and supply sides, to one of high-level disequilibrium, in which ever-
growing demand propelled the expansion of the supply and in which non-
agricultural revenues (indirect taxes, borrowing, and other means of financing)
replaced the land tax to be the most important sources of government income.
This took place in the 1850s through the 1870s, after the two factors bolstering
the original low-level equilibrium disappeared because of the fatal threats of for-
eign powers that entailed the Qing state’s huge expenses on defense and
because of the internal crisis of the population explosion which, coupled with
the outflow of silver from the domestic market, had exhausted much of China’s
economic surpluses and thus curtailed its population’s tax-paying ability. This
transformation accounted in large measure for the Qing state’s success in pacify-
ing internal revolts, modernizing its military, and defending its frontiers against
foreign aggression, hence the three decades of “restoration” (zhongxing) from
the mid-1860s to the mid-1890s. But the transformation was achieved at the
expense of the imperial court’s centralized control of regular and irregular rev-
enues on the one hand and the rise of provincial elites on the other hand, who
played a key role in the restoration by taking advantage of regional financial,
military, and administrative resources at their disposal. However, instead of
interpreting this development as the root cause of the late Qing state’s decline,
as often found in past studies, this chapter underscores the fact that, for all the
growing autonomy of regional leaders in military and civil administration, the
central government never lost its control of provincial authorities in the late
nineteenth century; it retained and exercised its power in decision-making over
 Introduction 13
the allocation of the officially reported resources of all kinds (financial, military,
and administrative) at the regional level, a relationship that can be best termed
as “regionalized centralism,” which was key to the late Qing state’s abilities to
survive domestic and foreign crises.
Chapter 5 examines how regionalized centralism functioned, focusing on the
Qing state’s adjustment of its geopolitical strategy and the efficacy and limits of
this adjustment in actual implementation. The Qing adjusted its strategies in
response to geopolitical crises on two fronts: the traditional threat to its north-
western frontier (i.e., the invasion of a Kokand-based force into Xinjiang) in the
mid-1860s, and the new threat from Japan to the southeast and east coasts after
the mid-1870s. Overall, its strategy underwent a transition from solely concen-
trating on the security of Inner Asian frontiers to paying attention to both fron-
tier and maritime defenses, though priority was persistently given to the
northwestern frontiers until the 1880s, and the defense of the east and southeast
coasts, for all its growing importance in the redefined strategy, was never a top
priority for the imperial court until its defeat in the Sino-Japanese War of 1894.
Regionalized centralism explains in large part the Qing state’s success in defend-
ing its northwestern frontier: Governor-General Zuo Zongtang was able to
carry out the highly expensive operations in Xinjiang partly because of the cen-
tral government’s generous funding and effective coordination of fiscal and mili-
tary resources at both national and regional levels, and partly because of Zuo’s
utilization of his personal networks and his access to the newly available means
of financing. In sharp contrast, the secondary importance of maritime defense in
the Qing court’s new strategy and hence its limited spending on this front,
more than any other factor, accounted for Governor-General Li Hongzhang’s
eventual loss in competing with the Japanese, despite his initial success in indus-
trial and military modernization because of the limited availability of regional-
ized financial and administrative resources.
Regionalized centralism, while contributing to the Qing state’s longevity,
nevertheless enhanced the fiscal, military, and administrative leverages for
regional leaders to bargain with the imperial court and gave rise to “conditional
loyalty” among them, to distinguish them from the unconditional loyalty of
their predecessors before the mid-nineteenth century. This was especially true in
the last decade of the Qing, when a series of “New Policies” (xinzheng) were
vigorously pursued at both national and local levels as a comprehensive scheme
of modernization. Central to the implementation of the New Policies, Chapter
6 shows, was the Qing court’s efforts to recentralize its control of fiscal and
military institutions, hence an infringement of the interests of regional author-
ities, which in turn led to their weakened identity with the Qing court; the
Manchu elite’s monopolistic control of the cabinet and exclusion of Han bur-
eaucrats only worsened the identity crises. The fall of the Qing, in a word, was
chiefly the immediate consequence of the Manchu elites’ radical measures of
recentralization and the widened Manchu–Han schism during the New Policies
decade, rather than an inevitable result of regionalized centralism of the late
nineteenth century.
14 Introduction
Along with the obvious failures of the Qing state in its last decade, Chapter 6
also highlights other developments that worked together to keep China from
falling apart in 1911, namely, the Qing court’s effective management of its fron-
tiers through religious patronage and preferential treatment of frontier elites,
the Sinification of the Manchus, the migration of the Han people to Manchuria,
Inner Mongolia, and Xinjiang, and the frontiers’ administrative integration with
the interior provinces in the last decades of the Qing. All these contributed to
the forging of a shared statehood among the elites and ordinary people of differ-
ent ethnic origins. Although rebels and revolutionaries in different parts of
China turned to anti-Manchu racism to win popular support in the early stage
of the anti-Qing movement around the turn of the twentieth century, they
quickly renounced it and embraced the formulation of a “Republic of Five
Races” (Manchus, Mongols, Tibetans, the Hui Muslims, and the Han), as did
the Manchu and other non-Han elites of the frontiers. All these enabled the
Republic’s retention of the frontiers of the former Qing upon its inauguration
in 1912.

The making of a centralized and unified state

Part III explores the third breakthrough in the modern transformation of the
Chinese state that ended the long-term regionalization of fiscal, military, and
administrative power. It was in their quest for a unified and centralized state
that we find parallels between Republican China and other latecomers to state-
making in both the West and the non-Western world. For the first time, war
became the driver of state-building in China, and developing the capacity to
generate the revenues for war became the most decisive factor in state formation
and in shaping the outcome of competition among regional contenders.
Chapter 7 focuses on the years under the Republican government in Beijing
(1912–1928). Instead of viewing this period as a time of warlordism and equating
the latter with political decay or failure in state-rebuilding, this chapter highlights
the rise of regional fiscal-military states as a critical link in China’s long-term pro-
cess of state transformation and a prelude to the making of a unified and central-
ized state on the national level. The warlords or warlord cliques of individual
provinces, I argue in this chapter, were not just concerned with military build-
up and war making; the most successful of them also took vigorous steps
to strengthen themselves by centralizing the bureaucratic system within the
territories under their jurisdiction, unifying and standardizing the fiscal and
financial systems, constructing highways, railroads, and other infrastructural
projects, promoting public education and health, encouraging the develop-
ment of industry and commerce, advocating self-government at the grassroots
level, and allowing the existence of representative and deliberative assemblies
at the provincial or county level to build consensus among the elites.
Together, these efforts fell under the rubric of “centralized regionalism,”
a new development in the age of warlordism that enabled the most ambitious
provincial leaders to turn their areas into regional “fiscal-military states.” By
 Introduction 15
the mid-1920s, two powers emerged as the most successful and influential
among all of the regional states, namely, the Fengtian clique under Zhang
Zuolin in Manchuria, and the Guomindang force under Sun Yat-sen and later
Chiang Kai-shek in Guangdong. It was eventually the Guomindang state that
triumphed over all of its competitors and proceeded the furthest in the making
of a unified national state in China by the mid-1930s. State-making in modern
China, therefore, did not follow the top-down path experienced by European
first-comers such as France and Britain, where the centralization and bureau-
cratization of state power proceeded from the central to local levels; instead,
after the failure of a similar top-down experiment under the New Policies in
the late decade of the Qing, China took on the bottom-up path, following the
precedents of later-comers such as Germany, Italy, and Japan, where the most
powerful regional states took the lead in unifying the country and establishing
a government on the national level.
Chapter 8 explores how the frontiers were rebuilt during the Republican
period through the triangular interactions of three factors: the influences of for-
eign powers, the attitudes of frontier elites, and domestic politics in China. Brit-
ain played a key role in Tibet; its strategy to turn the highland into a buffer
between British India and China and its persistent acknowledgment of Beijing’s
suzerainty over Lhasa explained Tibet’s capricious relationship with Beijing and
its failure to seek official independence and international recognition. Regime
changes in Russia and Moscow’s shifting strategies in the Far East formed the
geopolitical background in which Outer Mongolia exited, rejoined, and finally
separated from China. Finally, China’s alliance with the United States was key
to its triumph in the war with Japan and the recovery of its sovereignty over the
lands that had been ceded to Japan or ruled under its puppet state. Equally
important in shaping the continuity in China’s territory in the twentieth century
were the survival of the historical and religious ties that the Qing court had
built with frontier elites as well as the nationalism of political leaders under the
Republic. A more decisive factor, however, had to do with the resilience of
the Nationalist state under Chiang Kai-shek during its most difficult years in the
late 1930s and early 1940s, when China suffered Japanese invasion and occupa-
tion. Without the Nationalist state’s resistance and contribution to the Allies’
war against Japan, China’s recovery of its lost territories from Japan would have
been significantly delayed and discounted.
Chapter 9 examines the Nationalist regime’s successes and failures in state
rebuilding. Guomindang achievements in state-rebuilding were impressive:
it eliminated or subjugated all other regional competitors and established
a national government in China that was at least nominally unified by the late
1920s; as the only ally fighting Japan in the Far East before the outbreak of the
Pacific War in December 1941, it survived Japan’s full-scale invasion in the most
difficult years of the War of Resistance; and it fully recovered China’s sover-
eignty over the lands lost to Japan by the end of World War II, and turned
China into one of the five permanent members of the Security Council of the
newly founded United Nations, thus paving the way for China to act as a major
16 Introduction
player in world politics in the decades to come. Yet, the Nationalist state’s fail-
ures were equally striking. Instead of underscoring the rampancy of corruption
and moral deterioration as conventional wisdom suggests, this chapter shows
that the GMD regime’s vicissitudes in state-making lay chiefly in the peculiar
nature of its fiscal-military formation and identity building, which can be sub-
sumed under the rubric of “semi-centralism.” In sharp contrast with the high
level of centralization and unification that the Guomindang force once achieved
in its early years in Guangdong, after its unification of China in 1928 Nanjing
never succeeded in incorporating the original warlord forces into a unified and
centralized state system, not to mention its abilities to penetrate rural society
and establish an infrastructure for social control and mobilization. Fiscally, it
was increasingly constrained, not only because it prioritized military spending
for power consolidation and national defense, but also because, as a national
regime, it shouldered the huge burden of supporting the entire civilian person-
nel. Its failure to establish an efficient, centralized bureaucracy greatly curtailed
its abilities at revenue generation, hence the worsening disequilibrium in its
fiscal regime that ultimately resulted in bankruptcy. Politically, Chiang Kai-shek
based his leadership more on compromises with his rivals than on ideological
persuasion and personal charisma. Thus, when the War of Resistance was over
and the Civil War broke out, the Guomindang regime was crippled by the defi-
ance of regional Nationalist leaders and the defection of Nationalist soldiers,
which, coupled with the bankruptcy of its fiscal system and loss of popular sup-
port, led to its collapse in 1949.
In sharp contrast, the CCP outcompeted the GMD not only because of the
Communists’ abilities at mass mobilization or the incompetence of its enemy, as
past studies have shown, but rather because of the convergence of its break-
throughs in identity building, geopolitical setting, and fiscal-military constitution
in the 1940s. Chapter 10 demonstrates that the Communist force overcame the
problems of regional autonomy (shantouzhuyi or “mountaintop-ism”) that had
long been embedded in separate Communist base areas and factionalism (zong-
paizhuyi) within the top leadership of the Party, after years of the Rectification
Campaign in the early 1940s. As a result, by the end of the War of Resistance,
Mao Zedong had established himself as the undisputed leader of the Party, the
supreme commander of its military, and the sole ideological authority. All these
allowed the CCP to achieve a high level of organizational solidarity and political
identity among its members throughout the country and prepared it well to
prevail over the Nationalists on the battleground. Geopolitically, the CCP’s situ-
ation fundamentally improved after World War II, when Japan was no longer
a barrier to China’s state-building and the Soviet Union stepped in as the big-
gest geopolitical factor to influence China’s domestic politics. Its support turned
out to be critical to the CCP, making possible its swift infiltration and eventual
occupation of Manchuria, the strategically most decisive area in the Party’s
rivalry with the Nationalists. Controlling Manchuria further fundamentally
reshaped the CCP’s fiscal and military constitution, from being based on the
isolated, poverty-stricken rural areas in northwestern China to fully utilizing the
 Introduction 17
resources of the northeast, one of the most productive regions both agricultur-
ally and industrially in the nation. For the first time, the Communists were able
to compete with the Nationalists on an equal footing. No less important to the
CCP, however, was its mobilizing abilities that it had developed through its
incessant penetration into rural communities, from which the CCP generated
the manpower and logistic support on a massive scale for its military offensives.
What characterized the Communist force, therefore, was a peculiar yet supple-
mentary combination of its newly obtained strength through centralized control
of urban manufacturing and financial resources on the one hand and its trad-
itional ability in the decentralized mobilization of rural manpower and labor ser-
vice on the other. It generated an equilibrium in the CCP’s fiscal constitution
that was self-sustainable and highly expandable. The strength of the Communist
force in the 1940s, in the final analysis, lay in its “total centralism” or all-round
centralization in party organization, fiscal-military institutions, and political iden-
tity. It was on the basis of total centralism that the CCP defeated the GMD and
built a highly unified and centralized state in 1949.

State transformation unfinished

The modern Chinese state that took shape in the mid-twentieth century is best
seen as the accumulative result of the three breakthroughs outlined above; namely
the birth of an early modern territorial state, its transition to a modern sovereign
state, and the making of a unified and centralized state. Each of these three break-
throughs has its own logic and dynamic; together, they made the Chinese path to
a modern state distinct from the “normal” transition from empire to nation-state
commonly seen in other parts of the world. Instead of building an expansionist
empire, what emerged under the Qing by the mid-eighteenth century was an early
modern state that laid the territorial and ethnic foundations for the rise of
a modern Chinese state. Instead of its colonization by Western powers or splitting
into multiple entities upon its demise, China’s transition from an early modern ter-
ritorial state to a sovereign state in the nineteenth century was characterized by the
striking continuities in its territorial and ethnic structures. Finally, instead of
embracing Western-style democracy, state-making in Republican China ended in
the rise of a unified and centralized state, a result of both regionalized centralism
of the late Qing period and centralized regionalism of the early Republican period.
The modern state thus made in China by the mid-twentieth century appeared to
be at once “big” in size and “strong” in structure.
However, despite the establishment of the PRC in 1949, and despite the sur-
prising stability and vitality it has displayed throughout its existence in the past
seven decades, state transformation in modern China, a process that began in
the mid-seventeenth century, remains unfinished. Chapter 11 identifies three
challenges that confront the Chinese state in the twenty-first century:

(1) the cultivation of national identity among Han and non-Han peoples, by
doing away with the Han-centered construct of national history, cultivating
18 Introduction
the awareness of a shared nationhood on the basis of mutual cognition of
their respective cultural heritage and distinctiveness, and developing a shared
set of interests, values, and national goals on the basis of economic
integration;
(2) the reconfiguring of China’s relations with the rest of the world, by doing
away the victim mindset of the revolutionary age, participating or creating
mechanisms of regional integration and global governance to benefit all par-
ticipating nations, and promoting in the rest of the world its “soft power” or
cultural, intellectual, and political creativeness and commitment to the whole
of humanity;
(3) the rebuilding of its relationship with the people it governs at home, by priori-
tizing the wellbeing of the people in policy-making and opening the process
of governance at all levels to participation by all citizens regardless of their
stances.

China cannot be a great nation-state with the global standing it deserves

unless it tackles these challenges with success in the decades to come.

Notes
1 For an overall analysis of empires and imperial ideologies in world history, see, e.g.,
Eisenstadt (1963); Doyle (1986); Hobsbawn (1987); Scammell (1989); Pagden
(1995, 2003); Howe (2002); Burbank and Cooper (2010).
2 For literature on nationalism and popular sovereignty, see Morgan (1988); Greenfeld
(1992); Alter (1994); Brubaker (1996); Gellner (1997, 2006); Hechter (2000); Yack
(2001); Anderson (2006); Bourke and Skinner (2015); and Lee (2016).
3 For discussions on the rise of modern nation-states, see Emerson (1960); Mehta (1999);
Muthu (2003); Opello (2004); Roeder (2007); Hobsbawn (2012); and Tuck (2015).
4 Recent studies have shown that conquest and colonization were also central to the forma-
tion of the major states and kingdoms in medieval and early modern Europe, including
England/Britain, France, and Spain, a process not too different from empire-building,
although these countries are often invoked as well-defined early nation-states in the litera-
ture of nationalism. The differences between empire and nation-state blur when one looks
at how nationalist rivalries among the great powers of Europe such as Britain, France, and
Germany developed into imperialism in the period from the 1870s to World War I, when
they competed for overseas territories and turned each of them into a global empire.
Nationalism of this period, therefore, was inherently imperialist, and “imperialism and
nationalism were part of the same phenomenon,” as historian Christopher Bayly aptly puts
it (Bayly 2004: 230). Despite their expansive look and colonial rule over different parts of
the world, all those European powers treated one another as nation-states. “If nation-
states can be seen as empires,” Krishan Kumar thus writes, “empires, especially modern
empires, can seem no more than nation-states writ large” (Kumar 2010: 133).
5 Not surprisingly, whether “the Chinese communist regime [is] legitimate” constitutes
the number one issue when specialists in the field address the various “China ques-
tions” (Rudolph and Szonyi 2018).
6 There have been, to be sure, a number of studies on the modern Chinese state from the
perspective of long-term developments throughout the Qing and the Republican periods,
and each of these works focuses on a particular aspect of state-making, such as political
participation and competition (Kuhn 2002), the rise of nationalism as an ideology (Harrison
2001), continuity and transformation in legal institutions (P. Huang 2001, 2010),
 Introduction 19
territoriality and frontier building (X. Liu 2010), moral discipline and corruption control
(Thornton 2007), and so on. Other scholarly works emphasize a global, comparative
approach to understand the patterns of the Chinese experiences in state building, as seen in,
for instance, state formation and social resistance (Wong 1997), transformation in fiscal
institutions (W. He 2013), or renovations in statecraft (Halsey 2015). There are even more
studies on the Chinese state of a given period, such as the high Qing (1644–1840),
the late Qing (1840–1911), or the Republican (1912–1949) period, and each of these
studies centers on a specific issue, such as ethnic relations, ideology, finance, military
systems, and institutional building (to be cited throughout the following chapters).
7 For discussions on macrohistory and its differences from “long-range history” of
a given nation or a “total world history,” see Galtung (1997). For an overall review of
the literature on macrohistory, see Collins (1999).
8 This generalization about the relationship between socioeconomic conditions and the
mechanism of recourse extraction and its consequence on state formation, while helping
us to make sense of the historical trajectories of the English state and the major powers
in continental Europe, does not fully reflect the actual experiences of individual European
states in developing their respective political institutions (Brewer 1989: xiii–xxii; Storrs
2009; Glete 2002: 42–51), nor does the assumption about the correlation between an
agrarian or commercialized economy and the different levels of difficulty in resource
extraction stand the test of the histories of specific states (Ertman 1997: 10–19).
Part I

The formation of the

Qing state
Map I.1 China under the Qing, circa 1800
2 The rise of an early modern
territorial state
China in the early to
mid-Qing period

On the surface of it, the Manchu rulers’ conquests in the seventeenth and the
first half of the eighteenth centuries appeared to be no different from the similar
processes of empire-building in the rest of the world. This began with the rise
of a tribal people of Jurchen stock in the area along the northern edge of the
Korean border in the Changbaishan (Long White Mountain) region in southern
Manchuria, their competition with other tribal forces and the unification of the
entire Manchuria region under the Later Jin (founded in 1616, renamed
“Qing” in 1636), and their subjugation of the 24 Mongol tribes (mainly the
Chahar Mongols) south of the Gobi Desert in the 1630s. The next and most
important step was their conquest of the Ming and the inauguration of their
rule in China in 1644 after relocating their capital city to Beijing. The conquest
resumed decades later through a series of expeditions that ended in the incorp-
oration into their territory of Outer Mongolia in the 1690s, Xinjiang in the
1750s, and Tibet from the 1720s through the early 1750s.
Not surprisingly, students of Qing history have tended to compare the expan-
sion of frontiers by the Manchus to empire-building in other parts of the world.
Evelyn Rawski, for instance, posits that “[t]he Qing conquests in Inner and
Central Asia were comparable to the colonizing activities pursued by European
nations” and that the so-called Qing formation “bears many of the hallmarks of
the early-modern paradigm used to characterize European history in the seven-
teenth and early eighteenth centuries” (2004: 217, 220). James Millward
(1998), too, describes the Qing conquest of Xinjiang as a sort of imperialism,
although he distinguishes it from European colonialism that was associated with
economic motives and a missionary impulse. Peter Perdue is most explicit in
comparing the Qing expansion to state-building in other parts of the Eurasian
continent. Like Rawski, he challenges the conventional wisdom that groups
China, India, and the Ottomans as “agrarian empires” and distinguishes them
from European “states” that experienced “state building” and led in the making
of the early modern and modern world. Central to his argument is the import-
ance of warfare in shaping the structure of the Qing state and thus making
China analogous to Western Europe. He writes, “[i]n the early seventeenth cen-
tury, the Manchus constructed a state apparatus designed for military conquest.
Expansion of their state’s territory remained the primary task of the dynasty’s
24 The formation of the Qing state
rulers until the mid-eighteenth century” (Perdue 2005: 518). Military mobiliza-
tion “transformed the fiscal system, commercial networks, communication technol-
ogy, and local agrarian society.” The Qing state, therefore, “was not an isolated,
stable, united ‘Oriental empire’ but an evolving state structure engaged in mobil-
ization for expansionist warfare”; and the empire “did not diverge from Europe”
until its territorial expansion came to an end (2005: 527).
Was the Qing truly a warlike, expansionist empire comparable to the Euro-
pean military-fiscal states or colonial empires in terms of state formation? If yes,
why was the Qing able to keep almost all of its territory until its very end,
unlike other empires? If no, what can we draw from this historical process for
understanding the nature of the Qing and its implication for the rise of the
modern Chinese state? To address these questions, this chapter first looks at the
dynamics and motivations of the Qing operations leading to territorial expan-
sion. The rest of the chapter examines the different approaches adopted by the
Manchus to govern the frontiers and the interior, and explicates the uniqueness
of Qing formation against political traditions in Chinese history.

Frontier expansion
To understand the changing dynamics of state formation under the Qing, it is
important to distinguish between two different waves of its territorial expansion.
During the first wave that ended with the Qing conquest of the Ming dynasty
by the 1650s, the driving force behind the war efforts of Manchu warriors was
their thirst for land, people, and resources, which made the Qing expansion
during this period essentially no different from the empire-building activities in
the rest of the Eurasian continent. However, after its control of the interior
provinces in the mid-seventeenth century, the Qing lost the momentum to
expand. For nearly half a century from the 1640s, its Inner Asian frontiers
remained largely unchanged, when its goals were to preserve what it had con-
quered and to revive the suzerain-tributary relationships that had existed
between the Ming and Inner Asian tribal states. There was no sign that the
Qing rulers wanted to expand its frontiers in the north, northwest, or southwest
any further. It was not until the late 1690s that another wave of expansion
started, resulting in the incorporation of Outer Mongolia, Tibet, and Xinjiang
into its territory by the 1750s.

Geopolitical strategy and military campaigns

The key to understanding the Manchus’ engagement in the second wave of
expansions after 1690 lies in the Qing dynasty’s new geopolitical strategy, which
the Manchus formed after conquering the interior provinces in the 1640s. At
the core of this strategy was their alliance with the southern Mongols, who had
enabled their joint conquest of the Ming and remained critical to the safety of
the Capital City (Beijing) after the conquest because of Inner Mongolia’s geo-
graphic proximity to it. Next to the Manchu–Mongol alliance in strategic
 The rise of an early modern territorial state 25
importance was the Qing court’s close ties with Tibet, which permitted its role
as the supreme patron of Lamaism or Tibetan Buddhism, a religion that was
central to the spiritual life of both Tibetans and Mongols and therefore funda-
mental to the Manchu–Mongol alliance as well. The Qing launched a series of
military expeditions in the late 1690s and afterwards precisely because it faced
a new threat, the invasion by Zunghar Mongols west of the Gobi Desert into
Outer Mongolia, Inner Mongolia, and Tibet in succession, which challenged
the core strategic interest of the Qing. Therefore, unlike its conquests before
the mid-seventeenth century that were offensive and expansionist in nature, mili-
tary operations in the second phase were defensive at the beginning and increas-
ingly preemptive later; its inclusion of Xinjiang, Outer Mongolia, and Tibet into
its territory was an unexpected outcome of its expeditions against the Zunghars,
rather than the original objective of its war efforts.
The second wave of the Qing expansion involved several steps. The first was
incorporating into its territory the areas inhabited by the Khalkha Mongols
north of the Gobi Desert in 1691. Right after the inauguration of his reign
in Beijing, the Shunzhi emperor (r. 1644–1661) had shown his intention to
unite with the Khalkha Mongols on the ground: that both the Mongols and the
Manchus had “long belonged to one family” (su wei yi jia) or that both had
“long belonged to one single entity” (yi ti zhi guo) (Qimudedaoerji 1998),
repeating the same appeal that had been used by his grandfather Nurhaci
(1559–1626) in 1620 to win over the Khalkhas in resisting the Ming (Elliott
2001: 69). But the early Qing rulers were generally satisfied with maintaining
their nominal suzerainty over the Khalkhas despite the repeated defiance by the
latter. The Khalkhas saw no necessity to seek protection from the Qing until
1688, when they suffered fatal attacks from the Zunghars, the strongest of the
four major Oirat Mongol tribes that dominated the vast area west of the Gobi
Desert and who had for long claimed their tributary status from the Qing. Led
by Galdan, the Zunghars’ invasion also constituted an immediate threat to the
Qing. After defeating the Khalkha Mongols, the Zunghars moved southward,
invading the land of the southern Mongols until they suffered defeat by the
Qing force at Ulan Butong in 1690 (Wang Sizhi 2003, 1: 350–360; Perdue
2005: 155–157). In 1691, the Khalkha Mongols formally joined the Qing, to
be grouped with its new leagues and banners, as were the southern Mongols;
Outer Mongolia thus was fully integrated into the Qing. To completely elimin-
ate the Zunghars from the steppes south and north of the Gobi Desert, Kangxi
personally led three campaigns in 1696 and 1697, forcing the enemy to with-
draw from these areas.
The next step was the Qing’s control of Tibet after defeating the occupying
Zunghars in 1718 and 1720. For decades since its rule in Beijing, the Qing
court had only maintained a nominal relationship with the highland, content
with its role in endorsing the spiritual supremacy of the Dalai Lama and the
administrative power of the Khoshut khans of the Oirat Mongols in Tibet. Con-
flict developed, however, between the ruling Mongol khans (especially the last
one, Lhazang Khan, who had been backed by the Qing) on the one hand and
26 The formation of the Qing state
the Tibetan regent (Sanggye Gyatso), who acted on behalf of the Dalai Lama,
and local Gelug abbots on the other. The Qing court did not interfere until
1717 when Zunghars invaded Tibet at the appeal of the Tibetan abbots, killed
Lhazang Khan, and put the Dalai Lama into custody. In response, Emperor
Kangxi launched two campaigns in 1717. After expelling the Zunghars from
Tibet in 1720, the Qing officially established its protectorate over Tibet by
sending a garrison to Lhasa, setting up a government there consisting of
Tibetan ministers appointed by the Qing court, and in 1728 sending a royal
commissioner to directly supervise the government and mediate between the
Tibetan ministers (Zhao Yuntian 1995: 40–46; Goldstein 1997: 13–15; van
Schaik 2011: 138–141).
The most important step was of course the Qing’s eventual subjugation of
the Zunghars. Believing that “its land, if annexed, is not suitable for cultivation,
and its people, if obtained, is not fit for exploitation” (bing qi di buzu yi gen-
zhong, de qi ren buzu yi qushi), Emperor Kangxi treated the Oirats, as well as
the Khalkhas north of the Gobi Desert, as peoples different from the Southern
Mongols. He intended them to keep their status as tributaries of the Qing, as
long as they stayed where they were and did not challenge the Qing or its allies,
primarily the southern Mongols and Tibetans (QSL, Kangxi 227: 45–10-yiji).
He even tolerated the claims from the Oirats and the Khalkhas for being equal
to the Qing, instructing, for instance, in 1682 his envoys to tolerate Mongol
etiquette when received by the Oirat and Khalkha leaders, without insisting on
the humbling ritual on the part of the latter appropriate for their tributary status
(QSL, Kangxi 102: 21–7-yimao). Later, Kangxi decided to launch several cam-
paigns between 1690 and 1697 against the Zunghars only when the latter
marched eastward, invading the lands of the Mongols north and south of the
Gobi Desert and thus posing a direct threat to the Qing as mentioned above.
He launched a second round of campaigns against the Zunghars only after the
latter invaded Tibet in 1716 and thus threatened the Qing’s rights over the
highland. Afterwards, the Qing sought to restore the traditional relationship
with the Zunghars, accepting the Zunghar Khanate as a tributary (zhigong zhi
guo or nagong zhi guo) to the Qing while maintaining trade relations and
a peaceful border between the two sides (QSL, Kangxi 143: 28–12-xinwei). In
other words, Kangxi’s actions again the Zunghars were largely reactive and
defensive; he showed no intention of attacking the homeland of the Zunghars
and completely rooting them out.
However, the Qing did change its strategy from a passive one into
a preemptive type during the reigns of Yongzheng and Qianlong. Other than
Galdan Tseren’s (head of the Zunghar khanate) refusal to return Lobzang
Danjin, a Khoshut prince who had rebelled against the Qing in 1723, Emperor
Yongzheng experienced no severe provocation from the Zunghars, and his view
of Zungharia was no different from his father’s: “its land, even if obtained, is
not up for cultivation, and its people, even if conquered, is not up for employ-
ment” (de qi tu buzu yi gengyun, de qi min buzu yi qushi) (QSL, Yongzheng 4:
7–2-kuiji). Nevertheless, he insisted on an expedition against them, which did
 The rise of an early modern territorial state 27
happen in 1729, believing that, if not eliminated, the Zunghars “would consti-
tute a huge harm to all Mongols and could become a potential threat to the
state” (ibid.). Eliminating the potential threat from the Zunghars was especially
important for the Qing to secure its control of Tibet. For Emperor Yongzheng,

the question of Tibet would not be settled as long as the Zunghars remain
to be eliminated. If the Tibetan affairs cannot be properly handled, all of
the Mongols would remain undecided and skeptical. This is really
a potential worry of the state, which matters to the well-being of the state
and the people. Shenzu [Emperor Kangxi] knew well the whole issue and
its implications, and was determined to stabilize Tibet by eliminating the
Zunghars. How insightful Shenzu was! Without any other options, he
could only take this action in response [to the Zunghars’ threats] (budeyi,
biying juzhao yi).
(XZDA, 2: 395–396)

Unfortunately, Yongzheng’s campaign resulted in an unexpected loss at

Hoton Nor in 1731; the emperor saved face only after a Khalkha cavalry annihi-
lated more than ten thousand Zunghar troops in a single battle at Guangxiansi
(or Erdeni Zu) in 1732 (Feng Erkang 2003: 254–263; Perdue 2005:
252–255). Emperor Qianlong continued the preemptive strategy. Taking advan-
tage of the recurrent internal strife among the Zunghar nobles contending for
the khanship after the death of Galdan Tseren, Qianlong launched a campaign
in 1755 that defeated the forces of Dawaachi, the most powerful contender,
captured him, and further defeated the forces of Amursanaa in 1756–1757,
another contender who once sided with but quickly rebelled against the Qing
(Zhou Yuanlian 2003: 131–186; Perdue 2005: 274–289). To rationalize his
invasion into the homeland of the Zunghars, Emperor Qianlong refuted
a shared assumption among the “mediocre folk” (yongzhong) – including his
grandfather in fact – who had thought the Zunghars to be a people “who
showed no fear when confronted with the might [of the Qing] and no grati-
tude when granted favor, and whose land is uncultivable and whose people is
unruly” (wei zhi buzhi wei, hui zhi buzhi huai, di buke geng, min buke chen);
for him, the Zunghars had to be subjugated because, in his words, “our coun-
try encompasses all of the Mongols; how could then the Zunghars alone be
left outside the kingly civilization (wanghua)?” (cf. Zhang Yuxin 1995: 369).
Not surprisingly, after defeating the Zunghars, Qianlong incorporated the
entire area inhabited by the Oirats into the formal territory of the Qing, deem-
ing this area to be “no different from the interior provinces in all institutions
and regulations” (yiqie zhidu zhangcheng yu neidi shengfen wuyi) (QSL, Qianlong
722: 29–11-wushen).
In addition to the military expeditions outlined above, the Qing also engaged
in campaigns against the Uyhgur Muslims, the Tibetans, the Miao, and Burma
to consolidate its frontiers in the northwest and southwest. But the expeditions
against the Zunghar Mongols turned out to be the most decisive in expanding
28 The formation of the Qing state
the territory of the Qing. One and a quarter centuries later, when justifying his
proposal of defending Xinjiang against Yaqub Bek’s invasion (see Chapter 4),
Governor General Zuo Zongtang well explained the significance of the cam-
paigns against the Zunghars and the strategic importance of Xinjiang to the
Qing’s safety:

The reason to emphasize Xinjiang is to safeguard Mongolia; to safeguard

Mongolia is to defend the capital. If the northwestern frontiers remain com-
plete like the fingers connected with the arm, then no cracks will be left
[for enemies] to take a chance. If Xinjiang is not stabilized, Mongolia will
be insecure, the borders of Shaanxi, Gansu, and Shanxi will be exposed to
frequent intrusions, causing endless threats to our defense, and the posts
and mountains directly north [of the capital city] will not be at rest.

The reason why “our dynasty, with Beijing as the capital and Mongolia to sur-
round it in the north, has no alarms and threats for a century and several dec-
ades” was precisely because of the feats of “our ancestral reigns” that “destroyed
the Zunghars, pacified the Muslims, explored Xinjiang, and established military
administrative authorities there” (ZZT, 6: 701–702).

Limits to Qing expansion

It is worth emphasizing, however, that in the course of military expeditions into
the northwestern frontiers, the Qing court showed no intention of expanding
its territory any further than the land of the Zunghars. The Manchu rulers gen-
erally limited their expansion to areas inhabited by the Mongols and Tibetans
who practiced Lamaism as well as areas of different ethnic and religious back-
grounds that had been controlled by the Mongols, and they showed no interest
in further expanding the territory of the Qing into the land beyond these areas.
From Emperor Qianlong’s point of view, the Zunghars had been part of the
Mongol people (yi Menggu tonglei) and subject to the rule of the Yuan Dynasty
(1206–1368) (ben you Yuan zhi chenpu); it was only during the Ming that the
Zunghars withdrew from China and constituted a constant threat to the Ming
frontiers. Therefore, it was right for him to terminate the Zunghars’ state of
“staying lonely outside [China] and resisting civilization for generations” (he zi
waixie, shushi genhua) (QSL, Qianlong 496: 20–10-wuwu). In sharp contrast,
his attitude towards the Kazakhs was totally different. Having once sheltered
Amursana, the head of the rebelling Zunghar force, and afraid of the Qing
force’s retaliation in its pursuit of the Zunghar chief, the Kazakh Khan Ablai
proposed submitting the entire land of the Kazakhs to the Qing in
August 1757. Qianlong rejected this. For him, the Kazakhs “had never been in
communication with China since ancient times” and were different from the
Zunghars who had been part of the Mongols and once ruled by the Yuan. In
other words, the Kazakhs had always been “outsiders” to China, whereas the
Zunghars had once been insiders and therefore they had to be incorporated into
 The rise of an early modern territorial state 29
the dynasty of China once again. The emperor, therefore, instructed the Kazakhs
to remain a tributary state of China, in the manner of states such as Annam,
Liuqiu, and Siam, to which the Qing court would “neither grant official
titles nor demand taxes” (bu shou guanjue, bu ze gongfu) (QSL, Qianlong
555: 23–1-bingchen), instead of being incorporated into the Qing’s territory
and turned into provinces and counties of interior China or into the banners
of the Qing frontiers such as the land of the Khalkha Mongols, to which the
emperor often referred as “our Khalkha tribes” (wo Kaerka zhubu) (e.g., QSL,
Qianlong 526: 21–11-genshen) (see also Guo Chengkang 2005).
To recapitulate, unlike the territorial expansion of other empires driven pri-
marily by a religious cause or the greed for wealth, the Qing’s wars against the
Zunghars were largely defensive; a religious war or the thirst for more people
and land were never part of its plans. After all, the Manchu rulers of China did
not accept and promote any religion to legitimate their regime, and the 18
inland provinces were large and affluent enough to generate the revenues
needed by the Qing rulers to maintain the state apparatus. Thus, instead of
demanding payment of taxes or tributes in substantial amounts from the newly
established frontier, the Qing government subsidized Xinjiang as much as
1 million taels of silver each year in the eighteenth century (Millward 1998:
58–61). The ultimate goal of state formation under the Qing, in the final ana-
lysis, was not unlimited expansion of territory and subjugation of the population
for the purpose of enlarging the sources of state revenue; instead, it was driven
primarily by the imperial rulers’ pursuit of geopolitical security.

China enlarged and redefined

In the course of frontier expansion, the Qing rulers redefined the concept of
“Middle Kingdom” (Zhongguo, Huaxia, or Zhongxia) and its relationship with
the peoples outside it. Since the early days of Chinese civilization, the rulers of
Zhongguo, or the land inhabited by the Han Chinese, had assumed their cul-
tural superiority over the surrounding barbarians (yi or di, or yidi). The area
inhabited by the “civilized” (hua or xia, or huaxia), mostly the Han people,
was divided into commanderies (later provinces) and counties under the junxian
system, and they were subject to the direct control of a bureaucratic hierarchy,
primarily through tax collection and lawsuit adjudication. The surrounding
tributary states, for their part, were only expected to acknowledge the suzerainty
of the emperor of the Middle Kingdom through their periodical tributary visits
to his court and the subsequent performance of the ritual of kowtow to the
emperor. The emperor, therefore, was supposed to be the ruler of “All under
Heaven” (tianxia), whose relationship with the surrounding tributary states was
likened to that between the sovereign and the subject. It is also important to
note that, for Confucian adherents dating back to Mencius (372–289 BC) who
underscored cultural identities in defining different peoples, the boundary
between the civilized and the uncivilized was never fixed, and it was by no
means defined merely by ethnicity; whether a people were civilized or not
30 The formation of the Qing state
depended on their acceptance and practice of the cultural norms and political
institutions prevailing in the Middle Kingdom, a process known in Confucian
discourse as “kingly acculturation” (wanghua). Therefore, the yi (barbarians)
could become the xia (the civilized) if they were assimilated, and, as a result,
their land could be part of the Middle Kingdom, and the xia could degenerate
into the yi if they lost the cultural identities of the civilized (Zhang Qixiong
2010: 133).
After conquering the Ming, the Manchu rulers inherited from their predeces-
sors the Chinese notion of All under Heaven, believing that

the All under Heaven is the All under Heaven of all those who are under
Haven and it is not to be privately controlled by those from any part of it,
be they from the north or south or from the interior or exterior of it (fu
tianxia zhe, tianxiaren zhi tianxia ye, fei nanbei zhongwai suo de er si).
(QSL, Qianlong 1225: 50–2-xinchou)

In other words, the Manchus as a people from the exterior had equal rights to
claim their ruling status as the Han Chinese from the interior. Without having
to deny their origins as the yi, the Manchu rulers repeatedly justified their legit-
imacy to reign in Zhongguo by stressing the interchangeability between xia and
yi, and they were more than eager to prove their legitimacy before the Han
elites by showing their commitment to “kingly acculturation” (wanghua) or
Sinicization and therefore their attainment of the identity as a new member of
the xia, which did not have to be achieved at the expense of their original
ethnic identities. It was precisely for the purpose of legitimating its role as the
ruling dynasty of All under Heaven that the Qing inherited the tributary system
from the preceding Ming dynasty and demanded the former tributaries of the
Ming perform exactly the same rituals to the Qing court as they had done to
the Ming, as evidenced by its dealing with the Koreans after subjugating the
latter in 1637 (Rawski 2010: 80).
Where the Qing rulers truly departed from the traditional Sino-centric notion,
however, is their redefinition of Zhongguo (or Zhongtu) in the course of the
empire’s westward expansion. As Emperor Yongzheng stated clearly in 1729
when launching an expedition against the Zunghars:

Since our dynasty came to dominate the Zhongtu (the Middle Kingdom)
and to be the lord of All under Heaven, even the Mongol tribes of extreme
distance were all incorporated into the domain. The territory of Zhongguo
thus has been expanded widely and much farther, which is indeed a huge
blessing for the people of Zhongguo.
(QSL, Yongzheng 86: 7–9-kuiwei)

Although Yongzheng’s expedition failed, his redefinition of Zhongguo neverthe-

less influenced the perception of China by the following rulers. It is in this con-
text that a statement made by Emperor Qianlong in 1755 can be properly
 The rise of an early modern territorial state 31
understood. To justify his military campaign that ultimately defeated and elimin-
ated the Zunghars, the emperor claimed that “the Zhongxia of the imperial
Qing” was not the same as “the Zhongxia of the Han, the Tang, the Song, and
the Ming” (QSL, Qianlong 496: 20–10-wuwu). For him, the territory of the
Middle Kingdom encompassed not only the 18 provinces of the Han Chinese,
but also the lands inhabited by the Manchus and the Mongols, which had been
grouped into the various “banners” (qi) and under the effective administration
of the Qing court. The old divide between the “inner” (nei) and “outer” (wai)
by the Great Wall made no more sense and yielded to a united China (Elliott
2001: 359).

Governing the frontiers

The formation of the Qing state, therefore, is more than military conquest and
territorial expansion. Equally challenging to the Qing rulers was how to govern
the vast land comprising different populations, cultures, and religions. Whether
or not the imperial state could keep the different parts together and perpetuate
itself depended chiefly on the rulers’ abilities to legitimize their rule among the
diverse groups of people and cultivating their loyalty to the imperial court,
rather than the use of coercion and violence over the subjugated peoples.

From equals to subordinates: the frontiers under the Qing

In the course of frontier expansion, the Manchu rulers changed their relation-
ship with the Mongols and Tibetans from one of equals in hostility against, or
in alliance with, each other to one of inequality, in which the Mongols and
Tibetans were submissive to the Qing court. Take the Mongols north and west
of the desert, for example. Before it conquered the Mongols of these areas, the
Qing court treated them as outside its territory and separate from the southern
Mongols; it was generally content with the relationship with the Mongol tribes
that it had inherited from the Ming, in which the latter claimed themselves as
tributaries of the Qing court. Having been a tribal state itself before conquering
China, the Qing at times even tolerated the Mongol tribes’ claim of being equal
to it and the subsequent treatment of the envoys of the Qing court with “Mon-
golian etiquette” (Mengguli) or “Mongolian practices” (Menggu zhi li) that
bespoke of equality between the two sides, instead of the ritual appropriate for
their tributary status (QSL 102: Kangxi 21–7-yimao; Kangxi 22–7-wuxu). The
same relationship existed between the Qing and Tibet before the 1720s, in
which the former claimed its political supremacy over the latter but tolerated
a de facto equality between the two (Zhang Yongjiang 2001: 89, 92). After the
conquest was completed, however, the Manchu rulers transformed their rela-
tionship with the Mongols and Tibetans from a horizontal one into a vertical
one that entailed a political and moral hierarchy between the center and the per-
iphery. Epitomizing this transition was the Khalkha Mongol nobles’ perform-
ance of “three prostrations and nine kowtows” (san bai jiu kou) before the
32 The formation of the Qing state
Qing emperor or his envoys after 1691 (QSL, Kangxi 151: 30–5-dingkai). This
transition was also seen in the relationship between the Qing court and Tibetan
leaders. While Manchu nobles suggested Emperor Shunzhi travel to the border
area to personally welcome the visiting Dalai Lama in 1652, the Han ministers
objected on the ground that this proposal contradicted the emperor’s role as
“the lord of all states under heaven” (tianxia guojia zhi zhu) (QSL, Shunzhi 68:
9–9-renshen).
Although the Qing court imposed on the frontiers a hierarchy that had its
origins in the Confucian moral order and ritualistic traditions unique to the
Chinese political universe, it avoided extending to the frontiers the administra-
tive system of interior provinces; nor did it propagate Confucianism among the
peoples of the frontiers (Rawski 2004: 229). Instead, it created a separate
administrative system that allowed the Mongol and Tibetan leaders a degree of
autonomy in managing domestic administrative affairs while asserting its political
supremacy over the latter by stationing its troops at the frontiers and appointing
or authenticating their spiritual and administrative chiefs. Upon their submission
to the Qing one after another, the Mongol tribes were grouped into banners
(qi) in the Manchu style in the south (49 banners), north (86 banners), and
west (16 banners) of the Gobi Desert and in Qinghai, with the leaders of indi-
vidual tribes being officially appointed as the heads of the banners (known as
the jasagh or zhasake in Chinese), directly under the jurisdiction of the Court of
Frontier Affairs (Lifanyuan, traditionally translated as Court of Colonial Affairs)
in the Capital City (Beijing). Above the banners were the larger regional leagues
(meng). The Court not only appointed and supervised the heads of the banners
and leagues, but also demarcated the boundaries between different banners and
installed border inspection stations (karan, or kalun, and ebo) staffed with patrol
soldiers to prevent the tribes and herds of each banner from transgressing into
the pastures of neighboring banners.1 An important purpose behind the Qing
court’s division of the vast areas of the Mongols into separate banners and
leagues was “to dissipate their power through multiple partitioning” (zhong jian
er fen qi shi) (Liu Yuewu 2009: 65). The Mongol nobles who were appointed
as the heads of the banners and leagues were independent from one another
and each answered directly to the Court of Frontier Affairs; therefore, it was dif-
ficult for them to act as a group to challenge Beijing. To safeguard the border-
lands and ensure the Mongol elites’ subordination, the Qing government
installed garrison troops under the commanders of different ranks in the land
inhabited by the Mongols, including the Uliyasutai (Uliastai) General as the
supreme military and administrative leader as well as two subsidiary ministers
stationed in Khobdo (Kebuduo) and Kuren (Kulun or Ulan Bator) to oversee
Outer Mongolia, and the Ili (Yili) General to oversee the Mongols west of the
Gobi Desert.
The Qing court’s treatment of the Tibetans was somewhat different. After
expelling the Zunghar invaders from Tibet in 1720 through two military exped-
itions, the Qing put Tibet under its direct control by stationing a permanent
garrison in Lhasa and appointing an Imperial Commissioner in Tibet to
 The rise of an early modern territorial state 33
supervise the newly organized government, known as the Kashag (gaxia). In
1793, the Imperial Commissioner was further granted the power to supervise
the religious ceremony of selecting the new “living Buddha” (the Dalai Lama
and the Panchen Lama). But the Qing court allowed greater autonomy in Tibet
than in Mongolia and Xinjiang, as best seen in the fact that it never attempted
to divide the highland into administrative subunits under the direct supervision
of the Court of Frontier Affairs as it did to Mongolia (Zhang Yongjiang 2001:
129); instead, a unitary secular Tibetan government was retained to govern the
entire highland. The different treatments of Mongolia and Tibet reflected the
Qing court’s different purposes on the two frontiers. It had to put Mongolia,
especially Inner Mongolia, under its close and effective control because of its
geographic proximity to the Capital City and therefore its strategic importance
to the Qing’s security. Tibet, on the other hand, was distant from the political
center; fragmenting the highland into smaller units and putting them under the
Qing court’s direct control was strategically unnecessary, geographically incon-
venient, and fiscally uneconomical. Therefore, the imperial court allowed the
unitary Tibetan government to retain a high degree of autonomy in ruling the
entire highland.
In addition, after suppressing the Muslim rebels in 1759, the Qing govern-
ment put the Uighurs in different localities in southern Xinjiang under the lead-
ership of the hereditary beg (boke) of different ranks (328 in total), who were
subject to the jurisdiction of the Court of Frontier Affairs as well (Wang Dongp-
ing 2005). To prevent the Muslims from becoming too autonomous and
powerful, the Qing court prohibited the Islamic priests (Akhund) from interfer-
ing with the administrative tasks of the begs, a policy that contrasted sharply
with its allowing Lamaist leaders to have administrative power in Tibet and
Mongolia. It also enforced the policies of segregating the Han immigrants from
the Muslims and prohibiting their intermarriage to avoid potential conflicts
between the two populations (Yang Shu and Cai Wei 2008).
All of the Mongol leaders of the banners and leagues as well as the Uighur
begs had to pay a mandatory visit to the imperial court in December annually
(known as chaojin or nianban); failure to personally report to the court would
subject them to the penalty of stipend suspension. The Dalai and Panchen
Lamas in Tibet had the obligation of sending their envoys to Beijing every two
years (see Zhang Shuangzhi 2010).

Lamaism as an Inner Asian denominator

For the Qing rulers, Tibet was important not only because it served as
a geographic buffer between the interior provinces and the outside world
beyond Tibet but more importantly because it was home to Lamaism; patroniz-
ing and promoting Lamaism turned out to be critical for the Manchus to con-
trol Inner and Outer Mongolia. But the Manchu elites’ attitude towards
Lamaism was complex and changed over time. During the Later Jin and the
early days of the Qing, the Jurchen/Manchu leaders were actually disinterested
34 The formation of the Qing state
in, and even repelled by, Lamaism, believing it to be responsible for the decline
of the Mongols because of their indulgence in the “wasteful” and “deceptive”
practices of the religion (Shang Hongkui 1982: 109). Later, Hong Taiji
(1592–1643) became enthusiastic in building close ties with the Dalai Lama
and Panchen Lama by sending his emissaries to Tibet and inviting the Tibetan
Lamaist delegates to Shengjing (today’s Shenyang, capital city of the Qing
before 1644) only after he saw Lamaism as an effective tool to win support
from the Mongols and form an alliance for his objective of conquering the
Ming. After the conquest, Emperor Shunzhi continued the policy by inviting
the fifth Dalai Lama to Beijing and treating him in a lavish manner in 1652, as
mentioned above, for the purpose of stabilizing the alliance with the southern
Mongols and winning over the defiant Khalkha Mongols north of the desert.
The next year, the emperor granted the Dalai Lama an elaborate honorific title
upon his return to Tibet, which established the precedent of the Dalai Lama of
following generations to receive their post officially from the Qing court.
The emperor’s gesture as “the Lord Patronizing the Dharma” (hu fa zhi zhu)
turned out to be very effective. The Khalkha nobles, as devout believers in
Lamaism, soon established their tributary relationship with the Qing and com-
mitted to providing the Qing court with the “tribute of nine whites” (jiu bai
zhi gong, including one white camel and eight white horses). Later, emperor
Kangxi adopted the tactic of “divide and rule” towards the spiritual leaders of
the Tibetans and Mongols to prevent Lamaism from developing into a unitary
religious organization so as to compete with Beijing for dominance in Tibet and
Mongolia. He deliberately elevated the status of Zanabazar (1635–1723), the
first Jebtsundamba Khutuktu or spiritual head of the Gelug lineage of Tibetan
Buddhism among the Khalkha Mongols, to counterbalance the influence of the
Dalai Lama, an action that led to resentment from the Zunghar Mongols under
Galdan, who was backed by the fifth Dalai Lama, hence their invasion of the
land of the Khalkas in 1688; and Zanabazar turned out to be a key figure in
leading the Khalkha Mongols to join the Qing in 1691 (Shang Hongkui 1982:
110–111).2 It is also worth noting that, despite their prolonged confrontation
with the Qing, the Zunghars, too, distinguished clearly between Russia and the
Qing on a religious basis; they tenaciously resisted Russian encroachment, while
intermittently maintaining a tributary relationship with the Qing. Galdan thus
claimed in 1688 and 1690, even when fighting with the Qing forces, that
“Zhonghua and us share the same tradition” (Zhonghua yu wo yidao tonggui)
(QSL, Kangxi 146: 29–6-jiashen) and that “I have no intention to alienate
myself from the emperor of China and the Dharma of the Dalai Lama” (QSL,
Kangxi 137: 27–11-jiashen; see also Guo Chengkang 2005: 8). In fact, part of the
reason behind the military confrontation between Galdan and Kangxi, as indicated
earlier, had to do with their competition for influencing Lamaism through different
religious leaders and thereby building their political dominance.
Lamaism, therefore, played a key role in the formation of the Qing state from
the late seventeen to the mid-eighteenth centuries. For the Qing court, revering
the spiritual leaders of Lamaism and maintaining close ties with them was the
 The rise of an early modern territorial state 35
most effective way to win over the Mongols and Tibetans. But the Manchu
rulers’ approach to Lamaism was pragmatic; they needed it only when they
found it useful in frontier-building. They did not believe in Lamaism themselves
and did not promote the same religion among the Manchus; later, they lost
much of their interest in patronizing Lamaism and instead became increasingly
restrictive of it after the dynasty was consolidated. The Qing court deliberately
encouraged the different traditions of Tibetan Buddhism to grow in different
regions and thereby to weaken the influence of the Dalai Lama, who had once
been the supreme spiritual leader over all of Tibet and Mongolia in the first few
decades of the Qing (Liu Yuewu 2009: 65–66; Zhao Yuntian 1984, 1995:
55–56). Hence the emergence of four major tulkus or “Living Buddhas” in the
eighteenth century, namely the Dalai Lama over the Lhasa-centered Front
Tibet, the Panchen Lama over the Shigatse-centered Rear Tibet, the Jebtsun-
damba Khutuktu in Khalkha (Outer) Mongolia, and the Cang-skya Khutuktu in
Inner Mongolia (Yuan Senpo 1996; Xing and Chen 2010).
To sum up, not all of the Inner Asian frontiers were equally important to the
Qing court. For the Manchu rulers, their alliance with the southern Mongols
and therefore the security of Inner Mongolia were most important. Next to
Inner Mongolia was Tibet, whose security and submission to the imperial court
were critical to its control of the southern Mongols because of Tibetan Bud-
dhism being the spiritual tie between the Manchu and Mongol elites. Finally,
Xinjiang and Outer Mongolia were important only to the extent that their sta-
bility or instability directly affected the security of Tibet and Inner Mongolia. At
any rate, all of these frontiers served one purpose for the Qing, that is, providing
geopolitical security for the dynasty. In relation to the interior, their roles were
instrumental and indeed peripheral; the core of the Qing dynasty remained the
interior provinces, which mattered the most to its legitimacy, wealth, and lon-
gevity. How to govern the land of the Han Chinese, therefore, was the most
challenging task to the Qing rulers.

Governing the interior provinces

The ideological, geopolitical, and social contexts

The Qing differed from any prior Chinese dynasties of non-Han origins in that,
before relocating its capital city to Beijing, it had been under the influence of
Chinese culture, as seen in their veneration of Confucius (Shang Hongkui
1982: 107–108), acceptance of the Chinese notion of All under Heaven, and
definition of itself as a periphery in relation to the dynasty that ruled the interior
provinces (Guo Chengkang 2005: 2–3). Nevertheless, the biggest challenge to
the Qing rulers after conquering the interior was that they were to rule
a population far outnumbering the Manchus and having shown an unusual abil-
ity of assimilation in history, partly because of the sheer size of the population
and partly because of the superiority of the civilization it had created. They thus
endeavored to maintain the ethnic, linguistic, and cultural distinctions of the
36 The formation of the Qing state
Manchus as a privileged ruling group, while making great efforts to legitimate
and perpetuate its rule in China by proving its authenticity as the successor to
the Ming, with the Mandate of Heaven to rule the country. To that end, it
inherited from the Ming a centralized bureaucracy to rule the Han people and
embraced the Confucian ideology that undergirded the government system. As
gestures to confirm this legitimacy, the inaugural Emperor Shunzhi was enthroned
in typical Chinese style by personally presiding over a ceremony to offer sacrifices to
Heaven and Earth, a mausoleum was constructed to duly rebury the last emperor
of the Ming, and an honorary title was granted to the descendant of Confucius.
Later, to prove himself as an authentic ruler of China, Emperor Kangxi paid
a personal visit in 1684 to the Temple of Confucius in Qufu and performed the
ritual of “three prostrations and nine kowtows” to the statue of Confucius, rather
than merely “two prostrations and six kowtows” as Han Chinese emperors had
done prior to the Qing, and he performed the same ritual to the founding emperor
of the Ming during five out of his six southern tours to show his proper respect
to, and inheritance of his rulership in China from, the preceding dynasty (Shang
Hongkui 1982: 109).
But the Qing did not simply duplicate the Ming. Aside from a greatly
expanded territory, a changed geopolitical setting, and many administrative
innovations, the biggest difference between the two dynasties lay in the Qing’s
non-Chinese origin and the antagonism between the conquerors and the con-
quered throughout the history of the dynasty, which constituted the biggest
threat to the Qing court. Indeed, most of the rebellions in the interior provinces
appeared as anti-Manchu revolts by the Han Chinese, including the rebellions
by the Three Feudatories in the 1670s, the White Lotus Sect in the 1790s, the
Triads throughout the Qing, more recently the Taipings in the mid-nineteenth
century, and, finally, the revolutionaries in the early twentieth century. But in
most of the eighteenth century, the Qing rulers successfully handled Manchu–
Han relations, established the legitimacy of the Manchu rule, and minimized
the cost of maintaining peace in China. The key to their success lay in the
Qing court’s policy towards the gentry elites in Chinese society. It ruthlessly
persecuted disloyal literati through the so-called “literary cases” (wenziyu).
Meanwhile, to win over the literati, it revived the civil service examination,
offered them high-ranking positions in the government, and entrusted them
with the task of compiling encyclopedic works. More importantly, to enhance
the Qing’s legitimacy in Chinese society, the Manchu elites not only embraced
Chinese culture and therefore turned out to be much more Sinicized than any
preceding non-Chinese regimes in China (Ho 1998), but they also appeared
to be more enthusiastic and sincere than preceding Han Chinese dynasties in
promoting Confucianism as the ideology of the state. For Confucian elites as
well as Qing rulers, the best way to practice Confucianism was to implement
administrative policies that best manifested the Confucian tenet of “benevolent
government” (renzheng); light taxation, therefore, became a sacred principle
that the Qing rulers faithfully adhered to throughout the eighteenth and nine-
teenth centuries.
 The rise of an early modern territorial state 37
But the Qing’s ability to implement this policy rested on the following two
preconditions. The first is a favorable geopolitical setting. After subduing the Three
Feudatories in 1681 and incorporating Mongolia, Xinjiang, and Tibet into its terri-
tory, the Qing emerged as a dominant power in the Far East, where no surrounding
state could challenge its supremacy. In the absence of severe threats outside the
empire and content with its supremacy over the neighboring states under the tributary
system, the Qing rulers saw no reason to enlarge or upgrade its standing army. In
most of the eighteenth century, the size of the standing army was limited to a total of
approximately 800,000 to 850,000 men, including 600,000 of the “Green Stand-
ards” and 200,000 to 250,000 of the “Eight Banners” (Peng Zeyi 1990: 55; Chen
Feng 1992: 21–23, 97), which only accounted for an insignificant fraction of the
total population in China (0.4% around 1700 and 0.19% around 1800) (Ho 1959).
The government’s regular expenditures on the military, ranging from 13 million taels
per year in the 1720s to 17 million in the 1790s (Chen Feng 1992: 194), again
accounted for a negligible percentage of China’s total economic output (1.24% in the
1720s and 0.85% in the 1790s). Although the annual expense of maintaining the
regular forces accounted for 58–65% of the Qing government’s total expenditure, it
was well within affordability. Staying at the level of about 17 million taels, they
remained largely unchanged from the 1730s through the 1840s.
The second precondition is ethnic homogeneity – hence the overall absence
of social disorder engendered by ethnic or religious differences – within the
interior that minimized the cost of government, and the large population of
interior provinces that provided a huge tax base. All these made it possible for
the Qing to tax the land at a very low rate; they also made the Qing different
from any of the preceding Chinese dynasties; the latter, despite their embrace of
the Confucian principle of benevolent government, did not hesitate to increase
the tax burden on the ruled to sustain huge military spending when confronted
with threats from the neighboring tribal states, as seen in the late Ming dynasty.

Low taxation
The Qing declared its commitment to the policy of “light taxation” (qingyao
bofu) from the very beginning of its rule in China.3 By and large, the burden of
land taxes under the Qing from the late seventeenth to the early nineteenth cen-
turies was light and decreased in relation to land yield. Table 2.1 shows the
total amount of land taxes that were actually collected (including the illegal or
informal surcharges that accounted for about 25% of the legal tax liabilities, see
Yeh-chien Wang 1973, Table 4.1, note d). As shown, after the Qing suppressed
internal revolts, the tax burden quickly decreased to less than 10% of the gross
amount of land yields in 1685, about 5% in 1724, less than 3.5% in the late
eighteenth, and below 3% in the early nineteenth century.
Another way to gauge the level of tax burden is to compare the total amount
of all taxes that were actually collected (including both the direct taxes on land
and adult males and the indirect taxes on commodities and transactions) to the
total agricultural and industrial output. Again, we find in Table 2.2 a similar
38 The formation of the Qing state
Table 2.1 Land tax as a percentage of land yield, 1652–1820 (in 1,000
silver taels)

Year Land taxes collecteda Land yieldb Tax burden

(%)

1652 49,795 315,991 15.75

1685 38,809 415,359 9.35
1724 44,550 890,648 5.01
1766 57,360 1,783,965 3.21
1784 54,348 1,646,449 3.30
1820 63,478 2,313,710 2.74

Notes:
a Land taxes in 1652 include (1) diding: 21,261,383 taels in silver and 5,739,424 dan
in grain (He Ping 1998: 84) (the payment in grain equals 18,574,746 taels, con-
verted at 3.30 taels per dan), totaling 39,836,129 taels; and (2) informal surcharges:
9,959,032 taels (25% of diding; the same rate is applied to the following years). Land
taxes in 1685 include (1) diding: 24,449,724 taels and 7,331,131 dan (including
3,000,000 dan as caoliang) or 6,598,018 taels at 0.90 tael per dan, totaling
31,047,742 taels; (2) informal surcharges: 7,761,936 taels. Land taxes in 1724
include (1) diding: 26,362,541 taels and 7,731,400 dan (including 3,000,000 dan
as caoliang) or 9,277,680 taels at 1.20 taels per dan, totaling 35,640,221 taels; and
(2) informal surcharges: 8,910,055 taels. Land taxes in 1766 include (1) diding:
29,917,761 taels and 8,317,735 dan (or 15,970,051 taels at 1.92 taels per dan),
totaling 45,887,812; and (2) informal surcharges: 11,471,953 taels. Land taxes in
1784 include (1) diding: 29,637,014 taels and 7,820,067 dan (including 3,000,000
as caoliang), equivalent to 13,841,519 taels at 1.77 taels per dan, totaling
43,478,533 taels; and (2) surcharges: 10,869,633 taels. Land taxes in 1820 include
(1) diding: 30,228,897 taels and 8,821,183 dan or 20,553,356 taels (at 2.33 taels
per dan), totaling 50,782,253 taels; and (2) surcharges: 12,695,563 taels.
b The total size of arable land in 1652 was 403,392,500 mu (Chen Zhiping 1986: 89).
Suppose that grain yield per mu declined to a third of the late Ming level, i.e., 0.627
dan per mu, due to war devastation, then the total land yield would be 252,792,633
dan. Further assume that two-thirds of the total land yield, i.e., 168,528,422 dan, was
unhusked rice or wheat, thus worth 252,792,633 taels at 1.5 taels per dan. The remain-
ing one-third (84,264,211 dan), as miscellaneous grains, was worth 63,198,158 taels at
0.75 tael per dan (i.e., a quarter of the price of husked rice; the same for the following
years). The value of the total land yield, then, was 315,990,791 taels, assuming that the
land devoted to cash crops had the same return as land for grain production. The value
of total land yield in other years is calculated using the same method.

level of tax burden (though slightly higher) in those years and a similar long-
term decline.
The tax burden in Qing China was also light when compared to the level of
taxation in major European countries in the eighteenth century. In England, for
instance, taxes accounted for nearly 12% of its GNP in 1700 and 18.61% in
1789 (Goldstone 1991: 204–206). In France, the tax burden increased from
5.06% of its GNP in 1700 to 6.87% in 1751 and 8.71% in 1789. It should be
noted, however, that in both England and France, industry and trade were
much more important to their economies than in China. In France, industry
and trade accounted for 25% of its GNP in 1700 and 31% in 1789. In England,
 The rise of an early modern territorial state 39
Table 2.2 All taxes collected as a percentage of total economic output,
1652–1820 (in 1,000 silver taels)

Year Total amount of Total economic Tax burden

taxes collecteda outputb (%)

1652 55,328 351,100 15.76

1685 45,658 461,510 9.89
1724 52,412 1,047,821 5.00
1766 76,480 2,098,782 3.64
1784 72,464 1,936,999 3.74
1820 84,639 2,722,012 3.11

Notes:
a Including the actual amount of land taxes from Table 2.1 and all other kinds of taxes
that are estimated to be 10% of the total amount of taxes actually collected in 1652,
15% in 1685 and 1724, and 25% in 1766, 1784, and 1820 (for the actual share of
land tax in the state’s official revenue, see Table 3.3).
b Including the amount of land yields from Table 2.1 and non-agricultural output that
is estimated to be 10% of the total economic output in 1652 and 1685 when the
economy had not fully recovered, and 15% in the following years (for a gross estimate
of non-agricultural output as a percentage of total economic output during the
Qing, see also Zhou Zhichu 2002: 40).

they made up 45% of its GNP in 1700 and 55% in 1789 (ibid.: 204–206). By
contrast, in China, industry and trade contributed only about 10% to its total eco-
nomic output in the late seventeenth and early eighteenth centuries and 15% in
the second half of the eighteenth century. As a result, tax income from industry and
trade also had a much higher percentage in the total tax revenues of the two Euro-
pean countries (in France, 54% in 1700 and 50% in 1789; in England, 66% in 1700
and 82% in 1789) (ibid.: 204–206), whereas in China, they accounted for only about
10% of the government’s revenues in the early eighteenth century and about 25% in
the second half of that century. Yet, despite its smaller share in the total tax revenues
in France and England in most of the eighteenth century, agricultural taxes in both
countries remained heavy when measured against their agricultural output and when
compared to the situation in China. In the 1780s, for example, agricultural tax
accounted for nearly 6.34% of land yield in France, and 7.36% in England (ibid.:
204–206), whereas in China, it accounted for only 3.3% of land yield, or approxi-
mately half of the level in the two European countries. The tax burden was especially
light in China when compared to that in smaller European states such as Prussia,
where cultivators paid 16% of their produce as military taxes and an additional 18% as
seigneurial dues in the second half of the eighteenth century (Wilson 2009: 118).

Fiscal sufficiency and tax exemptions

To further understand why the Qing rulers adopted the light taxation policy, we
need to inquire into the reasons behind the state’s recurrent efforts to limit and
40 The formation of the Qing state
reduce the tax burden throughout the eighteenth century. A frequently prac-
ticed measure was regional or nationwide exemption of land taxes for a
given year. From the beginning of the Shunzhi reign (1644–1661) to the first
two decades of the Kangxi reign (1662–1722), when the dynasty was not yet
stabilized and the economy not fully recovered, the Qing rulers had already
repeatedly exempted the burden of land taxes in certain areas that suffered war
devastation, excessive logistic service to the military, or natural calamities. Tax
exemption in the early Qing, in other words, was used only as a means of relief
for victims of artificial or natural disasters. After 1700, such disaster-related
exemptions continued and became even more frequent and more generous than
before. What was new to the exemptions in the eighteenth century, however,
was the universal waiver that applied to the entire country no matter whether or
not a disaster had happened. The countrywide tax waiver first occurred in 1711,
when the Kangxi emperor announced that the 19 provinces of the empire
would be grouped into three batches to benefit from the waiver in turns over
three years (1711–1713) and which would apply to not only the land and adult
male taxes due in one of those years but also to all of the arrears owed by the
provinces. As a result, more than 38 million taels were exempted (Zhang Jie
1999: 58). The universal tax waiver took place five times during the reign of
Qianlong (1736–1795): in 1746, 1770, 1777, 1790, and 1795, respectively,
and a total of more than 27 million taels was waived each time (SQYJ: 12–17;
see also He Ping 1998: 41).
What made the universal waiver possible, of course, was economic prosperity
in most of the eighteenth century and hence the unprecedented level of surplus
of cash reserves at the Board of Revenue. When the Kangxi emperor announced
the first countrywide tax waiver in 1710, for example, the surplus had reached
45 million taels, which were more than the state’s total amount of annual rev-
enue (about 35 million taels). During the Qianlong reign, the surplus almost
reached or surpassed the mark of 70 million taels each time when the empire
proclaimed a countrywide tax exemption. But the unprecedented affluence was
not the only reason behind the Qing state’s repeated announcements of univer-
sal exemption. In his decrees to announce the universal tax waiver, the Qianlong
emperor repeatedly (at least three times in the 10th, 35th, and 37th year of his
reign) justified his decision by asserting the following two points. First, the
wealth of “All under Heaven” was a fixed amount (tianxia zhi cai zhi you ci
shu), which could “either gather together at the above level or disperse at the
lower level” (bu ju yu shang, ji san yu xia) (QSL, Qianlong 242: 10–6-dingwei).
In other words, the distribution of wealth between the government and the
people was a zero-sum game: it could either concentrate in the hands of
the central government or be shared by the people; to have too much reserve at the
state’s control would mean a reduced share of wealth by the latter. Second, the
government’s expenditures were a regular amount (guoyong yuan you changjing)
and now it had “abundant reserves” (fangjing lu cang cong ying) that could satisfy
all of its financial needs (Cf. Dai Yi 1992: 28). Therefore, “instead of hoarding too
much wealth at the state’s treasury, it is preferable to let it circulate among the
 The rise of an early modern territorial state 41
ordinary people” (yuqi duo ju zuo chang, wuning shi maoyan buwu ziwei liutong)
(QSL, Qianlong 850: 35–1-jimao; see also Qianlong 920: 37–11-guimao). “The
sufficiency of the people” (zu min), the emperor admonished, was the fundamental
way to “maintain abundance and secure peace” (chi ying bao tai) (QSL, Qianlong
242: 10–6-dingwei). The goal of the Qing state’s fiscal policy, as these statements
indicate, was not to maximize the government’s tax revenue or to endlessly increase
its reserves but instead to seek a balance between the state and the people in wealth
distribution; once the state gathered enough for its needs, to keep increasing its
reserve would threaten the livelihood of the people, which in turn would hurt the
state’s revenue and social stability.

Limits to taxation
Among the many measures taken by the Qing rulers to limit the level of tax-
ation, the most important one was no doubt the 1712 announcement to freeze
the tax on adult males and therefore permanently exempt males born afterwards
from paying that tax, which led to the merging of the adult male tax into the
land tax in most of China in the following decades and consequently the cancel-
ation of the tax burden on the landless.4
Another important measure to lessen the tax burden was the so-called huohao-
guigong implemented in the 1720s. The “melting fee” (huohao) was an unoffi-
cial surcharge collected together with the formal land tax under the excuse of
offsetting the loss of silver paid by individual households in its melting into bul-
lions for shipment to the treasury of the central government. The rate of the
surcharges ranged from 10% to as high as 30 or 40% (as in Shanxi province) or
even up to 50% (in Shaanxi and Gansu) of the land tax, most of which were
controlled by the county or prefectural yamen and used to augment yamen per-
sonnel’s salaries (Zelin 1984: 73, 88, 329 n44). The new measure promoted by
the Yongzheng emperor legalized and regularized huohao but limited its collec-
tion rate to 10–20% of the land tax. While most of the huohao income was used
as supplementary salary (yanglianyin) for local government officials, the regular-
ization of its collection at a fixed rate effectively reduced the burden of taxpayers
and improved the local government’s fiscal situation (ibid.: 130).
The Qing court repeatedly adjusted its tax policies on newly reclaimed land
for the same purpose. During the initial Shunzhi reign (1644–1661) when the
state’s military expenditures surpassed its limited tax revenue, the ruler imposed
a tax on this type of land after a period of tax exemption for three to six years
(Chen Feng 2008: 134–145). By the 1710s, as the economy recovered and the
state’s fiscal condition improved, the Kangxi emperor discouraged taxation on
the new land.5 Shortly after his enthronement, the Qianlong emperor took
a further step in dealing with the newly reclaimed land. In his view, local admin-
istrators were interested in the registration of reclaimed land and reporting this
type of land to the imperial court only for the purpose of increasing the govern-
ment’s tax revenues; such activities, however, “do not benefit the local [govern-
ment] at all and can only do harm to the people” (feitu wuyi yu defang, erbing
42 The formation of the Qing state
yihai yu baixing ye) (QSL, Yongzheng 4: 13–9-yihai). Therefore, at his instruc-
tion, the court enacted a new law that permanently terminated the survey of
fields and the compulsory registration of newly reclaimed land in the country.
Obviously, the Qing state was satisfied with the existing quantity of revenue
generated from the land. To increase its revenue by taxing the newly reclaimed
land made no more sense to the state, and would only increase the burden of
landowners. The best option for the Qing rulers, therefore, was to tolerate the
wide existence of unregistered “black land” or “hidden land” and the subse-
quent evasion of land taxes. By adopting the policy against land surveying and
the registration of newly reclaimed land, the Qing government actually put
a freeze on its revenue from land taxes. Indeed, throughout the reigns of Qian-
long, Jiaqing, and Daoguang, the state’s revenues from the land remained
largely stable, fluctuating around 30 million taels per year, which was only
slightly higher than the land taxes of the Ming during the Wanli reign. The
result was an increasing gap between the taxable amount of land that remained
a largely fixed quantity throughout the eighteenth century (varying from
720 million mu in the 1720s to 750 million mu in the 1850s, see Guo and
Zheng 1995: 101–102) and the actual amount of arable land that increased
steadily during the same period because of the incessant efforts of land reclam-
ation. The Qing state had no interest in investigating the actual amount of
taxable land in China as long as the predetermined quantity of tax revenues
was secured.
To recapitulate, the Qing state’s low taxation policy was based on two expli-
citly stated assumptions: (1) its annual expenditures were a fixed and institution-
alized amount, and the tax revenues that had been established by the Qing
statute were sufficient to cover all of its regular expenditures; (2) the increased
population had caused an unprecedented pressure on land resources and, there-
fore, priority should be given to the “sufficiency of the people” rather than the
augmentation of state revenues. These two assumptions further gave rise to two
features of the fiscal system under the Qing. The first is the overall inelasticity of
the state’s revenues, which resulted from three measures: (1) fixing the tax on
adult males in 1712; (2) merging the fixed male taxes into the land tax and the
regularization of melting fees during the Yongzheng reign, mostly in the 1720s;
and (3) fixing the land taxes by placing a ban on land surveying and the taxation
of newly reclaimed land in the 1730s. The second feature is the state’s substan-
tive approach to the perceived population pressure on the land: it chose to put
a ceiling on the overall level of the tax burden by terminating the taxation of
unregistered and newly reclaimed land, rather than conduct a thorough and univer-
sal land survey and thereby impose a reduced flat tax rate on both the registered
and the unregistered or newly reclaimed land.

Local governance
The low level of government spending under the Qing also had to do with the
pragmatic method it used in ruling local communities. The formal, centralized
 The rise of an early modern territorial state 43
bureaucracy of the Qing dynasty was limited to the county level and above;
below the county were various indigenous organizations and practices, which
aimed to serve the interests of the local population while collaborating with
state authorities in taxation and security control. By the Qing state’s design, all
of the rural and urban households were to be incorporated into a system called
baojia (every ten households formed a jia and every ten jia formed a bao) for
mutual surveillance and collective responsibility in tax payments. In reality, how-
ever, few communities were able to organize their households into the decimal
unit from the very beginning of its implementation, and few baojia units thus
created were able to operate and survive. Instead, in most localities, the baojia
was incorporated into, or built on the basis of, the preexisting community
organizations and therefore took different forms in different regions. In parts of
South China where the village community was built on lineage ties, for example,
the clan organization often superseded the baojia to perform the duty of tax
collection on behalf of the member households of the clan. In the lower Yangzi
region, in another instance, where land ownership was highly differentiated and
the wealthy households, especially those who had members graduating from the
civil service examination or retiring from government service, predominated, the
landed elites often engaged in proxy remittance (baolan) in tax payments, that
is, to pay taxes on behalf of the households under their patronage at a reduced
tax rate that was granted only to the privileged degree holders, a practice that
benefitted both the elites and their clients but caused the loss of the state’s tax
revenues (Bernhardt 1992).
In most parts of North China where social differentiation was low and owner-
cultivators prevailed and where the influential gentry elites were relatively weak,
ordinary villagers often took the lead to form various voluntary organizations
within or across village communities to sponsor temple fairs, the ceremonies of
rain prayer or worshipping the Dragon King, crop watch, flood control, defense
against banditry, and so on. In taxation, various communities developed differ-
ent practices. In communities where the livelihood was harsh, moving into or
out of the village was common, and therefore lineage ties were weak – as best
seen in the existence of multiple surnames in a single village – and the residents
found it difficult to cooperate in paying taxes to the government; instead, it was
usually up to the most influential individuals in the community to nominate an
ordinary person to serve as a semi-official agent, known locally as xiangbao,
dibao, or difang, who was responsible for urging the households in his commu-
nity or the larger area of multiple villages to pay the land tax before a time limit
and assisting the county government in carrying out other kinds of administra-
tive duties (P. Huang 1988: 225–232; Duara 1991: 50–53, 130).
In sharp contrast, in communities where the lineage ties were strong, cooper-
ation in tax payments prevailed, in the form of rotation among members of the
same lineage to serve as the agent of the community, called xiangdi, who paid
in advance the dues on behalf of all households within the community, thus
saving the member households the cost and time in paying taxes to the county
seat individually. Likewise, the xiangdi also served as an intermediary between
44 The formation of the Qing state
the county government and his community to carry out all kinds of administra-
tive duties. Without salary payment and formal appointment by the government,
the xiangdi position was unofficial, filled by villagers according to local customs
or “village regulations” (cungui). These regulations established the order of
rotation among the member households for the xiangdi service and made it
mandatory for the rest of the member households to compensate him for his
service by paying him a “commission” whenever they sold their properties or
goods, no matter whether or not the xiangdi served as a broker in the transac-
tion. Whenever the villagers disputed over the selection of the xiangdi, either
because of their attempts to avoid the service due to its burdensome duties or
because of their competition for this position when the commissions that the
xiangdi received from his fellow villagers outweighed the cost, mediation was
usually conducted within the community according to the villagers’ regulations.
If mediation failed and the litigation went to the county court, the magistrate
again would base his ruling on the village regulation, in the absence of any state
laws guiding the selection and service of the xiangdi (H. Li 2005: 66–91).
Given the preponderance of informal community arrangements and quasi-
official agents in local taxation and administration at the village or trans-village
level, the formal bureaucracy of the Qing state at the county level and above
was relatively small. Until the end of the nineteenth century, its size was limited
to roughly 23,000 salaried officials (Chang 1962: 42), or one official per
17,000 people. Its spending on civil administration was also very limited. In
1766, for instance, its spending on the salaries and stipends for all of the civilian
bureaucrats and the nobility was only 4.97 million taels, and the state’s entire
spending was only 42.21 million taels (Chen Feng 2008: 408–409), which
accounted for only about 2% of the total economic output at the same period.
The government system in Qing China was truly small in relation to the size of
its population, though it was cost-efficient in relation to the size of its economy.
In the final analysis, the Qing state that operated in the interior provinces was
a peculiar combination of a highly centralized formal bureaucracy, which existed
at the county level and above, with the informal or semi-official mechanisms that
performed the government’s everyday administrative functions at the grassroots
level. Rather than a consequence of its incompetence or shortage of revenues, the
Qing state’s limited penetration into local society resulted primarily from its
limited demand for revenues, which in turn reflected the state’s low-level spend-
ing on defense within a secured geopolitical setting as well as the limited size of
the government system in a highly homogeneous society. The Qing state, in
a word, felt no need to enlarge its formal apparatus, extend it beyond the county
level, and penetrate local communities as long as the latter fulfilled their duties to
the central government in paying taxes and maintaining stability.

The uniqueness of the Qing in Chinese history

Now we can put together the two separate systems of government that the
Qing employed to rule the frontiers and the interior and rethink the nature of
 The rise of an early modern territorial state 45
the Qing state and the reasons behind its unusual longevity and stability as
a dynasty of alien origin.
For long, historians have tended to interpret the conquest dynasties in Chinese
history from a Han-centered perspective, assuming a shared sense of superiority
among the Han people in their relations with those of different ethnic origins,
and underscoring their abilities to assimilate the conquerors that were inferior to
the Han economically and culturally. John K. Fairbank described this dominant
attitude among the ruling elites of Chinese dynasties as a form of culturalism, to
distinguish it from nationalism that is premised on competition and equality
among all nation-states in the modern world. It was this assumption of cultural
superiority that undergirded the Qing dynasty’s regulation of foreign trade under
the tributary system (Fairbank 1969; Hamashita 2008: 29–32). When explaining
the Qing’s striking longevity in relation to other “alien” dynasties in Chinese his-
tory and its success in ruling a vast, multiethnic land, a prevailing view up to the
1980s was to emphasize its policies of Sinicization: the Manchus who founded
the dynasty and conquered China not only inherited from the preceding Ming
the entire administrative system at all levels as well as the Confucian ideology that
accompanied the system, but they also embraced Chinese culture and language
to such an extent that by the early nineteenth century few Manchus were able to
speak their own language. The Qing was able to rule China for more than two
and a half centuries, according to this thesis, in large part because of its institu-
tional adaptation (Fairbank and Goldman 2006: 147) or Sinicization (Ho 1967,
1998) that allowed the Confucian elites of the Han people to develop identity
and loyalty with the conquest dynasty.
More recent studies since the 1990s have challenged this interpretation in
two important ways. Instead of viewing the Qing as another Chinese dynasty in
China’s long series of “dynastic cycles,” they present a new image of the
emperors of the Qing at the peak of its influence in the eighteenth century as
cosmopolitan rulers seeking to embody in their rulership the religious author-
ities and cultural values of different subject peoples on their territory (Rawski
1996). A Qing ruler arguably acted as a sagely, benevolent emperor, a khan,
and a cakravartin king at the same time, or aspired to be a “transcendent, uni-
versal ruler” as did the Qianlong emperor (Crossley 1992, 2002: 224). The
imperial ideology, accordingly, constructed a “universal empire” made of the
Manchu, Mongol, Tibetan, Uighur, and Han Chinese “constituencies.” Rather
than the center, China was reduced in this ideology to “a province of the
empire” (Crossley 2002: 341) or regarded “as only a part, albert a very import-
ant part, of a much wider dominion that extended far into the Inner Asian terri-
tories” (Waley-Cohen 2004: 195; see also Crossley 1992: 221; Rawski 1998: 2;
Millward 1998: 201). Another way to challenge the thesis of Sinicization is to
highlight the differences between the Manchu and Han peoples. Throughout
the history of the Qing dynasty, as Mark Elliott argues, the Manchu rulers
endeavored to defend the identities and privileges of the Manchus by preserving
their language and customs in dealing with an “identity crisis” that loomed
in the eighteenth century, safeguarding the Eight Banners as an institutional
46 The formation of the Qing state
framework to preserve the integrity of the Manchus as the conquest people, and
promoting the use of the Manchu language and the cultural heritages of
nomadic origins, which altogether formed the characteristic “ethnic sovereignty”
of the Manchu elites (Elliott 2001, 2006).

Between the Inner Asian and Chinese traditions

Throughout this chapter I have emphasized the distinctiveness of state forma-
tion in China under Manchu rule. The Qing’s origins from Manchuria, its con-
quest of the Ming, and its incorporation of Manchus, Mongols, Muslims, and
Tibetans into its territory made the new dynasty very different from the Ming; it
can thus be termed a truly multiethnic state. To treat the Inner Asian popula-
tions of the frontiers as mere “minorities” playing a marginal role in the Qing
formation, as does the traditional Han-centered historiography, downplays the
most significant part of the history of the Qing. Instead, this chapter emphasizes
the characteristic hybridity of the Qing state, which distinguishes it from all of
the preceding Chinese dynasties. As shown above, Qing China consisted of two
distinctive parts territorially, ethnically, culturally, and administratively: the
inland provinces of the Han people and the frontiers of the Manchus, Mongols,
Tibetans, and other non-Han populations. The state relied on the provinces for
wealth and manpower while safeguarding its geopolitical security with the estab-
lishment of frontiers. To ensure its effective extraction of taxes and maintain its
legitimacy as the rightful dynasty ruling Zhongguo, the Qing court put the Han
population within the provinces under its direct control while embracing Confu-
cianism and practicing the Confucian principle of benevolence in local adminis-
tration. To ensure the stability of the frontiers and cultivate their identity with
the imperial court, it allowed the elites of the non-Han races outside the interior
a degree of autonomy, patronized Lamaism among the Mongols and Tibetans,
and separated the Uighur Muslims and other tribal peoples from the Han popu-
lation. Despite these distinctions, however, the Qing succeeded in cultivating
a shared consciousness of a single statehood among the elites of all populations
on its territory, due to its emphasis on the use of religious ties, cultural assimila-
tion, and administrative adaptation, rather than military control and administra-
tive coercion, in ruling the different parts of the country.
Nevertheless, the so-called Inner Asian characteristics of the Qing state
cannot be overstated so as to equate the Qing with an Inner Asian empire and
downplay the interior provinces to only a part of it. A basic fact about the Qing
was that, after relocating its capital to Beijing and conquering the interior, it
changed itself from a frontier regime of the Manchus to a dynasty of the Middle
Kingdom, redefined Zhongguo by enlarging it to include the newly established
frontiers, and thereby equated the Qing with Zhongguo (see also G. Zhao
2006; Huang Xingtao 2011a). Accordingly, the Qing rulers transformed their
relationship with frontier elites from the traditional one bespeaking a degree of
equality between tribal states to a new type inherited from the preceding Chin-
ese dynasty and based on the Confucian principle of appropriate relationship
 The rise of an early modern territorial state 47
between the ruler and the subject. This was precisely the reason why Emperor
Kangxi rejected the use of Mongol etiquette by the leaders of Khalkha Mongols
and instead required their performance of “three prostrations and nine kow-
tows” after incorporating Outer Mongolia into his territory. Therefore, despite
the administrative and religious autonomy granted to the non-Han elites of the
frontiers, the relationship between the imperial court and the frontiers remained
characteristically the one continued from the Chinese past and embedded in the
Confucian vision of political hierarchy. Safeguarding the stability of the frontiers
remained critical to the Qing’s geopolitical security throughout the history of
the dynasty. But the importance of the frontiers stopped there; the Qing rulers
were neither attracted by the people and the wealth of Xinjiang, nor did they
truly believe in Lamaism. These frontiers were important only because they
worked to enhance the Qing dynasty’s geopolitical security and nothing else.
For the Qing rulers, the most important part of Zhongguo remained unmis-
takably the interior, which provided the imperial court with not only the rev-
enue and talent to run the state apparatus but also the political and cultural
basis to legitimize their authority over the entire Middle Kingdom, including its
newly established frontiers. The frontiers, for their part, only served as the
“fences” to enhance the security of the Qing state and safeguard its rule in the
interior provinces rather than to increase its revenues. The goal for the Qing to
rule the frontiers, therefore, was primarily to keep them in order and in submis-
sion. Its approach to ruling the frontiers thus was highly pragmatic: it patron-
ized Lamaism in Tibet and Mongolia but enforced measures to restrict the
influence of religious leaders in both regions while discouraging its spread
among the Manchus (Zhao Yuntian 1984). Its concerns with Islam as
a potential threat to the Qing emperor’s ultimate authority and with the volatil-
ity of the Muslim population accounted for its policy to encourage the immigra-
tion and settlement of Han farmers and merchants first in the northern part and
later in the southern part of Xinjiang (Fletcher 1978; Millward 1998). Its exped-
ition against the Zunghar Mongols ended in massacring almost the entire Zunghar
population at the instruction of the Qianlong emperor (Perdue 2005: 282–285).
All these contrast sharply with the Qing rulers’ earnest promotion of Confucianism
and steadfast adherence to the tenet of benevolent government in the interior prov-
inces. The Qing ruler’s boast of equal treatment of all peoples on his territory was
a high-sounding rhetoric rather than a historical reality.
All in all, the Qing is best seen as a hybrid polity that combined the traditions
of a conquest dynasty of nomadic origins with the cultural and political heritages
of the imperial Chinese state. Compared to the dynasties of alien origins that
ruled North China or the entire Chinese land before the Qing, including most
notably the Northern Wei (386–534) of the Tuoba clan of the Xianbei, the
Liao (907–1125) of the Khitans, the Jin (1115–1234) of the Jurchens, and the
Yuan (1206–1368) of the Mongols, the Qing had the longest history in ruling
China (1644–1911). Central to its exceptional durability was the supplementary
combination of two features: (1) its formal incorporation of the Inner Asian
frontiers into its territory and hence elimination of the threat from tribal states;
48 The formation of the Qing state
and (2) its transformation from a conquest dynasty into a hybrid dynasty inherit-
ing its legitimacy from the preceding Ming. Therefore, it won the loyalty of
gentry elite from inland provinces by respecting their privileges, adhering to the
Confucian ideology, and governing the populace accordingly; it thus was able to
put the Han population under its direct control yet keep the government to
a minimal size by delegating most of the everyday administrative duties to the
elite that dominated local communities while preventing them from becoming
autonomous from the central government. It also won the support of the reli-
gious and secular leaders of the frontiers by promoting Lamaism and granting
them autonomy in local administrations; it thus was able to maintain its stable
control over the frontiers without having to station a large number of troops
and use coercion to ensure the submission of the border regions. The unity that
the Qing court maintained with the frontiers through religious and political ties,
instead of coercion, also contributed to the growth of identity of the frontiers
with the imperial court, who would remain within the territory of China even
after the fall of the Qing. The two-century process of frontier-building and con-
solidation resulted in not only China’s territorial expansion but also the culti-
vation of a shared awareness among the Han and non-Han elites of belonging
to a redefined state, a consciousness that turned out to be more lasting than
the state itself. The Qing, in the final analysis, was undoubtedly the last in the
genealogy of Chinese dynasties, but the Chineseness of the Qing was not the
same as the one before the Qing; it was redefined and regenerated, embodying
the traits essential to the cultural and political legacies from the Chinese past
while showing a diversity and vibrancy unseen before. In fact, the Chinese civ-
ilization itself has been so enduring precisely because of its long tradition of
openness to the cultural and ethnic influences of the aliens; it was the unusual
power of the Han Chinese to assimilate the aliens and at the same time to
incorporate the elements of cultural legacies of the latter that accounted for its
resilience and continuity over the past thousands of years; the Qing was only
the last episode of that long story.

Between empire and sovereign state

What, then, is the nature of the Qing state? First of all, the Qing was not an
expansionist empire. Despite its early history of vigorous territorial expansion,
which resembled empire-building in other parts of the world, the nature of the
Qing changed after occupying China, from a conquest dynasty aiming to
enlarge its territory and wealth to the ruling dynasty of the Middle Kingdom
inheriting its legitimacy from the preceding Ming. In other words, it lost the
incentive to continue the conquest. Committed to turning itself into the legit-
imate dynasty in China, the Qing continued most of the institutions of the
Ming in governing the interior and regulating its relations with the tributaries;
it did not seek to conquer Korea, Vietnam, or any other neighbors of China,
nor did it demand the payment of tributes in large quantities from its tributar-
ies. For the first half-century after its inauguration in Beijing, the Qing was
 The rise of an early modern territorial state 49
content with the territory it had inherited from the Ming; territorially, its only
difference from the Ming was the adding of Manchuria and Inner Mongolia,
which it had controlled before occupying the interior provinces, in addition to
Taiwan, which was turned into part of Fujian province in 1683 after defeating
the Ming loyalists on the island. Later its establishment of frontiers in the north,
northwest, and southwest from the 1690s to the 1750s resulted from its reac-
tions to the threats from the Zunghars, and it showed no intension to push its
frontiers any further into Central Asia, although it was certainly able to do so at
the peak of its fiscal and military capabilities in the 1750s. Unlike the empires in
the rest of the Eurasian continent, whose limited size of their heartland often
compelled them to expand their territories incessantly for more revenues and, to
that end, to turn the subjugated territories into their tributaries or provinces,
the Qing saw no reason to do so, given the vast size of the Chinese heartland
and the abundant revenues obtainable from it. Unless its security was threat-
ened, the Qing court felt no need to turn its tributaries into frontiers for the
sole purpose of augmenting its revenues. Thus, although the Qing began as
a conquest dynasty of nomadic origins, it eventually developed into a hybrid
state that combined its tradition as an Inner Asian nomadic regime and Chinese
political legacies.
Nor was the Qing state a sovereign state in the European sense because it
firmly rejected the idea of legal equality among all sovereign polities that pre-
vailed within the Westphalian system of states. Nevertheless, it also showed
unmistakable signs of evolving towards the sovereign state before it was encap-
sulated in the Europe-centered system of states in the nineteenth century.
Although the Qing inherited most of its tributaries from the Ming and such
tributary states remained important for sustaining its legitimacy, it showed much
less enthusiasm for increasing the number of the tributaries or the frequency of
their visits to the imperial court, having established itself on the basis of alliance
with other forces outside China proper and felt much more confident and secure
than the Ming rulers. To the merchants and emissaries from Europe and Russia,
the Qing showed more flexibility in accommodating their needs and requests and
did not insist on regulating its relationship with them strictly according to the
conventions long established under the tributary system. More important, the
Qing state showed a growing awareness of its territorial sovereignty through its
confrontation with Russia and contacts with other Asian states; it demarcated its
boundaries with the neighboring states by negotiating treaties with the latter,
mapping its territories with the cartographic technologies borrowed from Europe,
and routinely patrolling its borders (Li Zhiqin 1995; Yu Fengchun 2006; Li Zhit-
ing 2011; Sun Hongnian 2011; Esherick 2010: 23), thus departing from the
traditional Chinese political principle that assumed the Chinese ruler’s universal
sovereignty over All under Heaven.
But the most essential characteristic of the Qing state, aside from its hybridity,
was its fiscal sufficiency before the nineteenth century. The establishment of the
Inner Asian frontiers and the disappearance of external threats by the mid-
eighteenth century allowed the Qing to limit its military spending to a largely
50 The formation of the Qing state
fixed, low level in the following century, while the prevalence of informal prac-
tices in the administration of local communities allowed the government to keep
its size to a minimal level. At the same time, the huge population of tax payers
generated a revenue for the government that was much more than the latter
actually spent during years of peace, even when the rates of the land taxes were
deliberately kept at a very low level. It was ultimately the financial solvency, epit-
omized by its repeated universal exemptions of land taxes, that explained the
loss of momentum for the Qing state to conquer any further, hence the stability
of its frontiers after the 1750s and the stabilization of its borders with the
neighboring states. All these distinguished Qing China from an expansionist
empire and qualified it as a self-sufficient, territorial state. It was not yet
a modern sovereign state before it was incorporated into the Westphalian system
of states, but it had developed some of the most essential features of an “early
modern” state found in Europe after the seventeenth century, in particular
a centralized bureaucracy, a merit-based officialdom, and a standing army.
Taking all these into account, we may term early to mid-Qing China as an early
modern territorial state.

Notes
1 The karan stations were also installed along the boundary between the Qing frontiers
and Russia and other neighboring states to patrol the borders (including 28 such sta-
tions along the border between Russia and the land of the Khalkha Mongols or Outer
Mongolia) (Baoyinchaoketu 2007).
2 The rationale for the first Jebtsundamba to propose the Khalkas’ annexation by the
Qing, rather than Russia, was religious and cultural: “to the north of us, there is the
state of Russia, which is governed fairly, but its land is unsuitable for preaching Bud-
dhism, and its people, with different customs, is not communicable with us; to the
south of us, there is China, which is governed fairly too, and the country is peaceful
and stable, where Buddhism is worshipped, and people wear fantastic costumes. So
wealthy is the land that embroidered silk, brocade, and fabric of fine hair abound. If
we join them, we will have good fortune for ten thousand years” (Miao Zhou
1988: 5.17).
3 In 1644, Prince Regent Dorgon (1612–1650) abolished the notorious “three military
surcharges” (san xiang) that had been imposed on taxpayers by the preceding Ming in
its last few decades and that totaled 169.5 million silver taels, or 67% of its regular
annual tax revenues (Yang Tao 1985). The Qing court reasserted its light taxation
policy in 1646 by compiling Fuyi quanshu (the complete compendium of taxes and
labor service) (completed in 1657). As the only official guidebook for tax collection,
its stipulations on the kinds and rates of taxes and the methods of tax collection were
all based on the original standards and practices established during the Wanli reign
(1573–1619) of the Ming, and it eliminated all of the surcharges later added during
the rest of the preceding dynasty (He Ping 1998: 4–5). But the huge amount of mili-
tary expenses involved in its conquest of China throughout the Shunzhi reign
(1644–1661) and in maintaining the regular forces, which was more than 30 million taels
per year, far surpassed its affordability (its annual revenue hovered around 24 to
25 million taels in the 1650s), making it impossible to carry out the light tax policy in
reality. In 1646, the Qing government restarted the collection of the “Liao surtax” (Liao
xiang) that totaled 5.2 million taels a year. In 1661, it further revived the collection of the
“Lian surtax” (Lian xiang) that amounted to 7.3 million taels a year (ibid.: 12). Both
 The rise of an early modern territorial state 51
surtaxes had been created by the late-Ming rulers and officially abolished by the Qing in
1644. The actual burden of land taxes under the Qing in its first few decades thus could
be as heavy as 15% of land yield or above (see Table 2.1). It was only after the suppression
of the Three Feudatories in 1681 that the Qing rulers were able to seriously implement
that policy by terminating the wartime revenue-generating measures and observing the
rules and regulations in tax collection as stipulated in Fuyi quanshu.
4 An obvious reason leading to this policy, as the Kangxi emperor explained in his 1712
decree, was that “the treasury of the state abounds nowadays. Despite the repeated
exemptions (of taxes) that amounted to more than tens of millions, there is no worry
of insufficiency in meeting the needs of the state’s expenditures” (QSL, Kangxi 249:
51–2-renwu). But the emperor also linked the policy with his understanding of the
increasing population pressure on the land: “Having enjoyed a long period of peace
within the country,” wrote the emperor, “mouths and households multiply. It has
become infeasible to increase the taxes according to the current numbers of adult
males: although the population increases, the size of arable land does not.” “Because
of the gradual increase of the people since the Pacifications [of the Three Feudator-
ies],” the emperor further explained, “there is no more land that is not reclaimed, and
there is no wasteland even in mountainous and hilly areas. Viewed in this light, the
living beings have become really too numerous” (ibid.: 51–2-renwu). The Kangxi
emperor’s reasoning was plain enough: just like the wealth of China that was a fixed
quantity, there was also a limit to the arable land in the country, whereas population
growth was unlimited. To continue taxation on the increased number of mouths made
no sense not only because the state had already enough revenue to satisfy all of its
needs but also because doing so would jeopardize the subsistence of the people.
5 Commenting on a report of land reclamation in the southern provinces, the emperor
wrote in 1713: “My idea is that the state has already had enough for its expenditures,
and there is no need to impose new taxes [on the reclaimed land]” (zhen yi guoyong yi zu,
bu shi jiazheng). Again he justified this decision by referring to the population pressure on
the land: “In earlier years, one mu of land was only worth a few cash coins because of the
abundance of land and the dearth of people. Nowadays, owing to the numerous people,
the land has become as expensive as several taels per mu. In the capital city and its vicinity,
residential and business houses have increased so much that no land is left vacant.”
“This year,” he further noted, “not only was there a bumper harvest of staple crops, but
crops such as sesame and cotton also had a good yield. Despite a year of bumper harvest,
however, rice and millet remain expensive. The reason is nothing less than the abundance
of people and the shortage of land” (QSL, Kangxi 256: 52–10-bingzi).
3 Limits to territorial expansion
Fiscal constitution and war-making
under the Qing

As shown in the preceding chapter, the Qing state’s expansion and establish-
ment of its Inner Asian frontiers in the late seventeenth and the first half of the
eighteenth centuries, as the most important steps in the formation of the Qing
state, were primarily a result of the imperial court’s reactions towards the con-
stant threat from Zunghar Mongols. The Qing rulers’ strategies in dealing with
this threat changed over time, from being reactive and defensive during the
reign of the Kangxi emperor (r. 1662–1722), to being preemptive under the
Yongzheng (r. 1723–1735) and Qianlong (r. 1736–1795) emperors, and finally,
as seen in this chapter, to becoming once again conciliatory in the 1830s. After
defeating the Zunghar Mongols in the areas north and south of the Gobi
Desert in the 1690s, the Kangxi emperor showed no willingness to continue his
operation into the vast area west of the desert where the Zunghars came from
and instead allowed the latter to maintain a tributary relationship with his court.
Decades later, the Qianlong emperor reversed Kangxi’s strategy and succeeded
in an expedition to completely wipe out the Zunghars in the 1750s, but the
Qing stopped its expansion afterwards at the peak of its fiscal and military
strength. The Qing rulers justified their varying strategies rhetorically over time:
whereas Kangxi legitimated his conservatism by dismissing the land and people
of the Zunghars as useless and unruly, Qianlong couched his offensives in the
lofty language of “kingly civilization.”
Neither geopolitics nor the rulers’ justifications, however, can sufficiently
explain why the Qing expansion into Inner Asian frontiers took more than half
a century, nor can it explain why the Qing rulers’ strategies changed over time.
What, then, was the mechanism that sustained and constrained its war efforts?
And what does this dynamism tell about the nature of the Qing state, especially
when viewed from the perspective of empire-building or state-making in the
early modern and modern world? In his studies of the Qing expeditions against
the Zunghars, Peter Perdue underscores logistics as a key factor in shaping the
decisions on military operations: while the logistic barrier prevented the Kangxi
emperor from launching a prolonged campaign against his enemies, the con-
struction of a supply route leading through Gansu into Xinjiang allowed the
Qianlong emperor to eventually defeat the Zunghars. The improved transporta-
tion of logistic supplies, he argues, was in turn sustained by “growing market
 Limits to territorial expansion 53
integration” in China (Perdue 1996: 780). “Only the commercialization of the
eighteenth-century economy as a whole allowed Qing officials to purchase large
supplies on the markets of northwest China and ship them out to Xinjiang”
(Perdue 2005: 523). But he admits the limited level of market integration in
Gansu, as evidenced in the fact that grain price increased by three times in this
area during the campaigns of 1755–1760 because of the army’s purchase of
grain from local markets (Perdue 1996: 781, 2005: 523).
This chapter shows, however, that aside from the Qing court’s calculation of
its geopolitical interests, a more fundamental reason behind its adjustments of
military strategies had to do with changes in its financial condition from the late
seventeenth through the early nineteenth centuries; it was its fiscal strength and
weakness, which was more fundamental than logistic service per se, that ultim-
ately explain the Qing state’s varying strategies toward the frontiers. It further
demonstrates that the key to understanding the state’s fiscal constitution is an
equilibrium prevailing in most of the eighteenth century, in which the supply of
revenues was largely commensurate with, or moderately higher than, the
demand for routine military and administrative expenditures. To explicate this
equilibrium and its implication for the Qing state, in this chapter I first examine
the overall relationship between the Qing state’s military operations and its fiscal
regime, as seen in a series of “fiscal cycles.” I then scrutinize the changing avail-
ability of land resources to the ever-growing population, which was key to
understanding the government’s financial abilities or, more specifically, the
prevalence and vulnerability of the low-level equilibrium in the Qing state’s
fiscal constitution. A thorough analysis of its fiscal–military relations in turn
sheds light on the unique nature of the Qing dynasty, which is compared with
early modern European states in the last section of this chapter.

War and finance under the Qing

Fiscal cycles
A good way to gauge the fiscal condition of the Qing state is to look at the size
of the surplus of cash reserves at the imperial court’s Board of Revenue, or the
accumulative amount of the remainder of the board’s annual revenues after
deducting its annual expenditures (see Table 3.1 for the state’s annual revenue
and expenditure). In most of the period under study, the annual revenue of the
Board, which increased slowly yet steadily from, for example, 31.32 million taels
in 1685 to 48.54 million taels in 1766 (see Figure 3.1 for details), was larger
than its regular expenditures that remained relatively stable, ranging from
27 million taels in 1687 to 30.77 million taels in 1766, thus yielding an annual
surplus of several (up to more than 10) million taels. But the cost of suppressing
domestic rebellions, conquering tribal peoples in border areas, or fighting an
international war, which were in addition to the regular expenditures, could
drastically reduce the reserves and even cause a deficit in the Board’s annual
expending allowances, hence undermining the solvency of the state. By and
54 The formation of the Qing state
Table 3.1 The Qing state’s official revenue and expenditure, 1653–1840 (in 1,000
silver taels)

Year Total Diding Salt Tariff and Misc. Total Balance

revenue taxes tax transit incomes expenditure

1652 24,380 21,260 2,120 1,000

1654 17,824 14,804 2,720 300 15,219 2,605

1657 25,486
1682 31,100
1685 34,240 28,230 3,880 1,220 910 29,207 5,033

1724 36,490 30,280 3,870 1,350 990

1725 35,850 30,070 4,430 1,350
1748 42,660 29,640 7,010 4,590 1,420
1766 42,540 29,910 5,740 5,400 1,490 34,510 8,030

1766 62,497 42,210 20,288

1791 43,590 31,770
1812 40,140 28,020 5,800 4,810 1,510 31,500

1838 41,273 36,209 5,063

1839 40,307 34,788 5,520
1840 39,035 35,805 3,230

Sources: for revenues in the years 1652, 1685, 1724, 1753, 1766, and 1812, see Xu and Jing
1990; SQYJ: 113–115. For data for 1654, see Liu Cuirong 1967; for data for 1652, 1682, 1766,
and 1791, see QSG, 125: 3703–3704. For data for 1838–1840, see HBS, 1.1: 172; Peng Zeyi
1983: 38.

large, we can identify five cycles, each spanning approximately 40 to 50 years, in

the fiscal condition of the Qing from the mid-1600s to the mid-1800s. Typic-
ally, a cycle started when the fiscal health improved, as seen in increases in the
reserve, owing largely to the restoration of peace and order in society; the cycle
reached its peak when the reserve increased to the highest level as a result of
economic recovery or expansion during prolonged peace; and it came to an end
when the reserve declined to the lowest level because of war expenses. What fol-
lows is a brief description of the five cycles.
Cycle I, 1644–1681. Throughout the Shunzhi reign (1644–1661), the Qing
government spent approximately 100 million taels of silver on conquering the
interior provinces, averaging nearly 6 million taels a year, which were in addition
to the regular expenses in maintaining the military (about 13 million taels
 Limits to territorial expansion 55

Figure 3.1 The Qing state’s cash reserves, war expenses, and tax exemptions,
1644–1862

a year) (Peng Zeyi 1990: 55). For example, the total amount of annual military
expenses was 20 million taels in 1656 and 24 million taels in the following few
years, whereas the Qing government’s total revenue was only about 20 million
taels a year, causing an annual deficit of about 4 million in the late 1650s (He
Ping 1998: 6). But the financial situation of the Qing soon improved after
the conquest. The Board of Revenue saw a surplus in its cash reserves during
the following nine years (1664–1673) when the Qing state enjoyed peace and
stability by and large. The surplus reached its highest level (21.36 million
taels of silver) in 1673, right before the outbreak of the rebellion of three
Han feudatories in Yunnan, Guangdong, and Fujian. Under Emperor Kangxi,
the eight-year campaign (1674–1681) cost the Qing at least 100 million taels
in total (Chen Feng 1992: 247), averaging 12.5 million per year, and drained
most of the Board of Revenue’s surplus, reducing it to as low as 3.32 million
taels in 1678, when the Qing army was fighting the most costly battles with
its enemies.
Cycle II, 1682–1722. After defeating the Three Feudatories in 1681, the situ-
ation under the Qing stabilized. Military campaigns occurred intermittently in
border regions afterwards, including: a two-year operation (1682–1683) to take
Taiwan from the Zheng regime, which cost about 4 million taels in total or
2 million a year; the war with Russians on the northeast border in 1695 and
1696, which cost less than 1 million taels in total; and the eight-year expeditions
(1690–1697) against the Zunghar tribes under Galdan in the region across the
Gobi desert, which culminated in the three decisive campaigns led by Emperor
Kangxi in person in 1696 and 1697 and cost a total of about 10 million taels,
or 1.25 million a year. Well affordable to the Qing, these operations did not
56 The formation of the Qing state
affect the fiscal condition negatively; quite the reverse, the surplus of cash
reserves at the Board of Revenue saw a steady growth, from less than 10 million
taels in the late 1670s to 26 million in 1686, nearly 32 million in 1691, and
40 million in 1694. After 1697, the Qing state enjoyed peace in its border
regions for 17 years; in most of this period, the board had a surplus of more
than 40 million taels (peaking at the level of 47 million taels in 1708),
which allowed Emperor Kangxi to repeatedly announce the policy of land-tax
exemption in selected regions during this period. But the peace on the bor-
ders came to an end when Emperor Kangxi launched a prolonged campaign
(1715–1726, especially 1720–1722) against the Zunghars who had invaded
and occupied Tibet, in which the Qing army spent a total of approximately
50 million taels, or about 5.5 million a year (Chen Feng 1992: 251–252).
The rebellion led by Zhu Yigui in Taiwan worsened the fiscal condition of
the Qing; suppressing the rebels cost 9 million taels in 1721. As a result, the
surplus of cash reserves at the Board of Revenue increased to 27 million taels
in 1722 (Shi Zhihong 2009: 104).
Cycle III, 1723–1761. After 1723, battles with the Zungars lingered for a few
more years into the Yongzheng reign but no longer constituted a financial
burden to the Qing. As a result, the Board of Revenue’s reserves quickly
bounced back to 47 million taels (the highest level in the previous cycle) in
1726 and skyrocketed during the next few years when peace prevailed in the
border areas, peaking at the level of 62 million taels in 1730. New threats, how-
ever, soon came from the Zungars under Galdan Tseren, forcing Emperor
Yongzheng to launch a six-year expedition (1729–1734) into the northwest,
which cost about 54 million taels, or 9 million per year. In 1734–1735, the
campaign to suppress the Miao people in Guizhou province in the south further
consumed 4 million taels. As a result, the Board of Revenue’s surplus plum-
meted during those years, to as low as 32.5 million taels in 1734, or only about
half of its level in 1730. Battles with tribal peoples took place sporadically in the
rest of this phase, including, most importantly, the revived expedition to the
northwest that ultimately subjugated the Zunghars (1755–1757) and the subse-
quent campaign against the Uighur Muslims (1758–1761), which cost the Qing
state a total of 33 million taels, averaging more than 4 million a year.1 Because
of these war expenses and other expenditures, the surplus of the Board of Revenue’s
cash reserves fluctuated largely between 30 and 40 million taels in most of the period
from 1734 to 1761 (Shi Zhihong 2009: 104; Chen Feng 1992: 258–259).
Cycle IV, 1762–1804. After stabilizing the northwestern borders, the Qing
enjoyed relative peace in the following three decades. As a result of unprece-
dented security and economic expansion (to be discussed later), the surplus of the
cash reserve at the Board of Revenue soon hit the mark of 60 million taels (the
highest level in the previous cycle) in 1765, further surpassed the 70 million taels
level in 1768, and reached nearly 82 million taels in 1777, which was the highest
record throughout Qing history. To be sure, sporadic wars took place in the
southern boarder regions during the three decades. The biggest was the revived
campaign (1771–1776) against the Tibetans in Jinchuan, which consumed a total
 Limits to territorial expansion 57
of 70 million taels, or 11.66 million a year, thus bringing the Board’s surplus
from nearly 79 million down to 74.6 million, though the overall health of the Qing
fiscal condition remained intact.2 Nevertheless, the state’s affluence was short-lived.
A turning point in the fiscal history of the Qing was the outbreak of the White
Lotus rebellion at the beginning of the Jiaqing reign, which lasted for nine years
(1796–1804) and swept five provinces in central and northwest China. As a result
of the Qing government’s massive spending (totaling 150 million taels, or
16.66 million a year) on subduing the rebels, the Board’s surplus precipitated from
nearly 70 million taels on the eve of the rebellion to less than 17 million taels in
1801, a very low level unseen in the preceding century (Chen Feng 1992: 268,
270, 275; Shi Zhihong 2009: 104).3
Cycle V, 1805–1840. Peace prevailed again in the Qing after it had suppressed
the White Lotus rebellion in 1804. For the next 16 years, no major warfare
took place within China or along its borders.4 Surprisingly, the surplus of the
Board’s cash reserves did not recover as quickly as it had happened at the beginning
of each of the previous cycles. In most of those years, the surplus was limited to
a range between 20 and 30 million taels, owing to reasons to be explained shortly.
The intrusion of the Kokand-based Jahangir Khoja and his followers into the north-
western border region in 1820 incurred the Qing state’s nine-year campaign
(1820–1828) to completely defeat him and an additional two-year (1830–1831)
operation to subjugate his brother, Yusuf Khoja. The Qing spent a total of
12 million taels (or 1.33 million per year) for the first campaign (causing a drop of
the Board’s surplus from 31 million taels in 1820 to 17.6 million in 1826) and
9 million taels (or 4.5 million per year) for the second (hence a drop of the surplus
from 33.4 million taels in 1829 to 25.7 million taels in 1832; no data for 1831 is
available). Throughout the 1830s, the Board’s surplus remained at a low level, ran-
ging between 20 and 30 million taels, despite the fact that no major warfare took
place during the decade (Chen Feng 1992: 273–275; Shi Zhihong 2009: 104).5

Financing the war

It is obvious from the above account that war expenses had an immediate
impact on the fiscal health of the Qing state. From the 1690s to the 1830s, the
recurrent campaigns against the Mongols, Tibetans, Gurkhas, Uighurs, and the
Miao involved a large sum of spending, totaling approximately 250 million
taels. Unlike the expenses on the normal operation of the government and the
military, which were usually a predictable and fixed quantity, those on wars were
not included in the state’s “budget” or regularly planned expenditures, because
warfare was unpredictable in most cases. Therefore, once a war broke out, the
surplus of cash reserves from the Board of Revenue would be the major source
of funding, which could be directly allocated to the military for a campaign or
be used to offset “Coordinate Remittances” (xiexiang) (see Chapter 4) from
individual provinces for war purposes when the campaign was over.6 The scale
of the war and hence the expenses involved thus directly affected the balance of
the Board’s surplus; this is evident in the fiscal cycles in the cash surplus
58 The formation of the Qing state
between the 1640s and the early 1830s as shown above. Each time a major
campaign took place, the level of the cash surplus declined; and when the cam-
paign was over, it bounced back. By and large, the expenditure on military
expeditions on the border, which rarely exceeded 10 million taels a year, was
within the scope of the Board of Revenue’s cash surplus, which ranged between
30 and 50 million taels of silver each year in most of the period from the 1690s
to the early 1760s, and between 60 and 80 million taels each year during the
period from the late 1760s to the early 1790s, and therefore was well affordable
to the Qing.
Of course, the cash surplus of the Board of Revenue was not the only source
of funding for the military expenses. A supplementary source, for instance, was
the voluntary donations from wealthy merchants.7 From the 1670s to the
1830s, salt merchants contributed a total of 42.75 million taels of silver to sup-
port the state’s wartime military expenditures, which accounted for about 17%
of the government’s total expenses on border wars during the same period.8
The salt merchants were enthusiastic in donating money to the court because
that was the most convenient and indeed the noblest means to show their loy-
alty to it and improve their standing in society; each time a merchant donated
a huge sum of silver to the court, he would be rewarded by the latter an hon-
orary high-ranking official title or be invited to a meal or entertainment with
the emperor. In fact, during the reigns of Qianlong and Jiaqing (from the
1730s to the 1810s), when the salt trade was most profitable, the merchants
were so active in willingly donating silver to the court during wartime that the
Qian long emperor had to reject many such contributions, claiming that “the
state’s treasury is abundant enough that there is no need to rely on merchants’
contributions” (guojia kufu chongying, wuji shangren juanshu) (cf. Chen Feng
1992: 334).
Military expeditions against the non-Han tribal peoples in border regions that
led to territorial expansion, therefore, did not necessitate the Qing government’s
measures of creating new taxes or increasing the rate of existing taxes. The
Yongzheng emperor thus proudly claimed in 1735 that “for the campaigns in
the western frontiers, all military expenditures were obtained from the imperial
court’s treasury, and no slightest burden was added to the people” (xichui
yongbing yilai, yiying junxu jie qugi yu gongnu, sihao buyi leimin) (QSL, Yongz-
heng 156: 13–5-jiachen); virtually the same words were repeated by Emperor
Qianlong in his comments on the campaign against Burma in 1769 (QSL, Qian-
long 840: 34–8-genshen). This was in sharp contrast to the Qing government’s
fiscal reactions to the massive campaigns against domestic rebellions by the
mainly Han population. During the period under study, there were two major
domestic rebellions by the Han people, one by the Three Feudatories
(1674–1681) and the other by the White Lotus Sect (1796–1804). Sweeping
many provinces and lasting for years, both rebellions undermined the Qing fiscal
condition by causing a huge loss of tax incomes from the areas damaged or
occupied by the rebels and incurring an enormous amount of expenses on
quenching the revolts. The Three Feudatories’ rebellion, therefore, wiped out
 Limits to territorial expansion 59
85% of the Board of Revenue’s surplus in a short span of six years, reducing it
to as low as 2.6 million taels in 1678 and even lower in the following few
years. The White Lotus rebellion, likewise, reduced the Board’s surplus by 76%
over the same number of years, to about 17 million taels by 1801. In fact, so
huge were the costs of the two wars that the Qing state, as to be explained
soon, had to reverse or revise many of its domestic policies to increase its tax
incomes. And the damage caused by the war to the Qing economy and the
state’s solvency was so immense and profound that it actually brought to an
end the unprecedented era of prosperity that lasted during the reigns from
Kangxi to Qianlong (the so-called “Kang-Qian shengshi”); in the following
decades after 1800, the Board of Revenue’s surplus would never recover to
a level higher than 33.5 million taels, or only 40% of its highest level in the
late 1770s.
In sharp contrast, the wars against the non-Han peoples in the border areas
only had limited effects on the fiscal conditions of the Qing. Instead of
a decline in its cash reserves during the eight-year campaign (1690–1697)
against the Zunghars, the Board of Revenue actually saw a steady increase in
the surplus (from about 31 million taels at the beginning to more than
40 million at the end), owing to the limited amount of war expenses
(1.25 million taels a year on average) and a rapidly recovering economy after
suppressing the Three Feudatories. The following two major operations
against the Zunghars did cause a significant decline in the Board’s surplus (by
approximately 40% between 1714 and 1723 and by 46% between 1729 and
1734, when the wars were most costly), but they were nevertheless within
the range of the Qing state’s financial affordability, as evidenced in the fact
that the state felt no need to make any change to its domestic policies regard-
ing revenue generation (to be discussed in detail soon). The last war against
the Zungars (1755–1757) had almost no effect on the Board’s surplus, which
actually increased during the first two years of the war and remained largely
equal to its level before the war. During the following four-year battle against
the Uighur Muslims (1758–1761), the Board’s surplus remained basically
unchanged, suggesting no effect of the war on the Qing state’s finance. The
most expensive war that the Qing ever fought with the non-Han people in its
border regions was on the Tibetan rebels in Jinchuan during 1771–1776.
Despite the hefty expenditure of more than 11 million taels per year and
a total cost of 70 million, the Board’s surplus surprisingly showed little
change during the six years; it stayed at the high level of nearly 65 million
taels even during the year (1775) when the war was most costly, thanks to
the unprecedented affluence that the Qing state enjoyed during the peak of
its fiscal strength and a prolonged economic prosperity throughout the entire
country. The abundance of the imperial court’s revenue, in turn, accounted
for its generous funding on the campaigns over the frontiers (Zhang Xiaotang
1990: 26–27). In fact, so affluent was the Qing court that the Yongzheng
emperor ordered in 1729 the exemption of the five peripheral provinces of
Gansu, Guangxi, Guizhou, Sichuan, and Yunnan from the following year’s
60 The formation of the Qing state
land taxes (Dai Yi 2009: 192) at the beginning of the six-year campaign
against the Zunghars under Galdan Tseren.
The contrast between the campaigns against internal rebellions and those on
the frontiers is especially stark when we look at their different impacts on the
Qing state’s fiscal policies. To put it briefly, the huge cost of wars in suppressing
the Han rebels within China proper and the subsequent strains on Qing finances
necessitated various measures for the state to increase its revenues, whereas
during the rest of the High Qing period, the state consistently adopted the
policy of “light taxation” that aimed to limit, reduce, and even exempt the tax
burden on landowners and cultivators as shown in the previous chapter, irre-
spective of the recurrent wars against nomadic and tribal peoples in the border
regions.
The impact of warfare on the finance of the Qing was especially evident
during the campaign against the Three Feudatories. This rebellion erupted
at a time when the economy under the Qing was far from a full recovery
due to the Manchu conquest that persisted throughout the Shunzhi reign
and involved the massive slaughter of the Han people, the destruction
of their dwellings, and the devastation of farmland. Thus, instead of
a surplus in the Board of Revenue’s cash reserve, there was a net deficit of
4 million taels at the beginning of the Kangxi reign (1662–1722) (He Ping
1998: 17). In the following 12 years, the improved stability of society per-
mitted a slow rehabilitation of the population and economy, and hence
a surplus of more than 20 million taels at the Board of Revenue on the eve
of the rebellion. The outbreak of the rebellion, however, interrupted the
recovery and greatly curtailed the tax revenue of the Qing when the rebels
occupied most of the southern provinces. The massive spending on sup-
pressing the rebels, therefore, soon exhausted the surplus and forced the
state to take various fiscal measures to increase its revenue and satisfy its
military needs:

1. increasing the tax rate by 30% on the land owned by government officials
and degree holders in prefectures of the historically prosperous lower Yangzi
region that was not affected by the war;
2. increasing the rate of salt taxes by 7.8 to 39.0% in different salt production
areas;
3. temporary collection (in 1676 and 1681 only) of the tax on houses for com-
mercial purposes at the rate of 0.2 to 0.6 tael per room throughout the
entire country;
4. the creation (in 1677) of new taxes on the sales of land and houses, which
were imposed as a total quota ranging from 100 to 600 taels on each
county;
5. shortening the period of tax exemption on newly reclaimed land from the
original ten years to three years;
6. investigating, informing against, and self-reporting of unregistered and
untaxed land;
 Limits to territorial expansion 61
7. selling government positions that were immediately available (between 1674
and 1677, the selling of more than 500 positions of county magistrates gen-
erated a total income of more than 2 million taels);
8. cutting the salaries of government officials at all levels from 50 to 100%
(Chen Feng 1992: 302–331).

During the campaign against the White Lotus sectarians in 1796–1804, the
Qing court once again turned to extraordinary means to generate enough rev-
enue for its unprecedented war needs. Unlike the fiscal situation at the begin-
ning of the war against the Three Feudatories, however, the Board of Revenue
had a huge surplus (nearly 70 million taels, or more than three times the figure
in 1673) before the outbreak of the rebellion. Thus, the Qing government did
not have to increase the rates of land and salt taxes or any other means of rev-
enue generation that would increase the burden of landowners as it did during
the war against the feudatories. Instead, it relied on the following two measures:

1. selling government positions in larger number and more frequently than

before, which generated a total income of more than 70 million taels during
the war and the following few years of rehabilitation (Zheng Tianting
1999: 319);
2. contributions from salt merchants, which resulted in a total of 15.8 million
taels between 1799 and 1803 (Chen Feng 1992: 333).

It thus can be concluded that, while the wars against large-scale internal
revolts, which were limited only to the short periods in the second half of the
1670s and at the turn of the nineteenth century, did incur a large amount of
military expenses and hence significant changes to the Qing state’s fiscal policies,
the recurrent wars in border regions in the late seventeenth and eighteenth cen-
turies had little effect on its fiscal policies. Why, then, was there such
a difference?
To answer this question, we have to first look at the Qing state’s different
strategies towards the challenges from within the interior and on the border-
land. For the Manchu rulers, the interior provinces were not just another piece
of land their ancestors or they themselves had conquered; it formed the core of
the Qing territory and the very basis on which they claimed their supremacy
over other Inner Asian peoples and all the tributary states to Zhongguo. The
Han people’s revolts, especially the large-scale rebellions sweeping several prov-
inces, therefore, constituted a fundamental challenge to the Qing court, and it
had to eliminate them at all cost and, to that end, meet the needs of military
expenditures by all possible methods, including creating new taxes or increasing
the rates of existing taxes. By contrast, the Qing court’s strategy toward threats
on the borderland displayed a striking degree of flexibility and varied over time,
depending on the geopolitical importance of the frontier in trouble regarding
the security of the dynasty. By and large, as demonstrated earlier, the court
underwent a transition from the conservative one of the Kangxi reign to an
62 The formation of the Qing state
increasingly aggressive one during the Yongzheng and especially the Qianlong
periods, which in essence also reflected the improved fiscal condition of the cen-
tral government over time. The fiscal abundance of the Qing state in the eight-
eenth century, to be sure, was not the sole factor leading to its offensives
against the border peoples, but it definitely encouraged the rulers to take actions
against them and to eliminate the threats once and for all. Not surprisingly, the
two most expensive offensives (in terms of average annual expenditure), namely
Yongzheng’s operation against Galdan Tseren in 1729–1734, which cost
9 million taels a year, and Qianlong’s operation against the Jinchuan Tibetans in
1771–1776, which cost nearly 12 million taels a year(!), coincided with the two
most remarkable surges of the Board of Revenue’s cash reserves, first in the late
1720s and again in the late 1760s and early 1770s (see Figure 3.1). It was, in
a word, the Qing state’s exceptional fiscal strength during those years that
enabled its most expensive military operations.
In sharp contrast, by the early nineteenth century, when the Board’s cash rev-
enue had plummeted to somewhere between 20 and 30 million taels (compared
to its peak at around 80 million taels in the 1770s), the Qing court’s strategies
toward trouble in the border areas again became conservative and conciliatory,
as best seen in the case of the invasion of Kokand forces into southern Xinjiang
in 1830, to which Emperor Daoguang responded by instructing his diplomats
to “concede to all of the requests” (yiqie ru qi suo qing) from Kokand for reli-
gious and consular privileges in Xinjiang (WY: 195), thus signaling its comprise
with Britain and other European powers in 1840 and afterwards.

The low-level equilibrium in Qing finance

The equilibrium defined

To further understand the Qing state’s fiscal constitution, we should take into
account both the demand and supply sides of its revenues. The demand side is
readily clear. It was argued in the preceding chapter that, in most of the eight-
eenth century, the Qing state enjoyed a peculiar combination of two factors,
namely (1) the high degree of security in its relationship with neighboring states,
and (2) the low cost of maintaining domestic order among a homogeneous popu-
lation which led to the state’s low spending on the military and government. It
was the coincidence of these two factors that enabled the state to keep its military
spending and hence its demand for revenues to a low level in relation to the size
of its economy and population. Indeed, as Table 3.1 shows, the Qing state’s
annual expenditure was relatively stable from the 1760s through the 1840s, ran-
ging between 34 and 38 million taels in most of those years.
Now let us focus on the supply side. Here again two key factors worked
together to determine the availability of fiscal resources to the Qing state. One
was its reliance on the land tax as the main source of revenues, which was no
different from the fiscal condition of the empires in other parts of Eurasia and
some of the continental European states in the seventeenth and eighteenth
 Limits to territorial expansion 63
centuries. Given the low productivity of the land under traditional farming tech-
nologies and hence the limited size of economic surplus from agriculture, this
excessive reliance on the land tax seemed to be unfavorable to the Qing. How-
ever, this disadvantage was greatly offset by the other factor, namely the huge
size of its taxable land. The Qing state was able to generate enough revenue
to satisfy its regular needs, even when the tax rate was set at a low level in
relation to the yield of the land. Given the Qing rulers’ persistent adherence
to the light taxation policy, as outlined in the preceding chapter, the annual
revenue from land taxes remained largely a constant figure during most of the
eighteenth century. Throughout the eighteenth and the first half of the nine-
teenth centuries, the state’s income from land taxes was limited to approxi-
mately 30 million taels of silver a year, while its total annual revenues
increased from about 35 million taels around 1700 to more than 40 million in
the second half of the eighteenth century, thanks to an escalation in indirect
taxes (on salt and other goods) (see Table 3.1). Therefore, throughout the
eighteenth century, the Qing state was able to maintain a favorable balance
between its revenue and expenditure, with the former somewhat higher than
the latter, hence an accumulative sum of surpluses sufficient enough to cover
unexpected expenses (on war and flood-control projects). Hence the existence
of an equilibrium in the Qing state’s fiscal constitution.
But this equilibrium was conditional and fragile. It existed only when the fol-
lowing two conditions prevailed. First, there was no severe challenge to the geo-
political settings and internal socio-political orders that had allowed the Qing
state to keep its expenditures at a low level; once a large-scale internal revolt or
a devastating challenge from outside took place that entailed a huge spending
on the military, the balance between demand and supply could be quickly
disrupted. Second, there were no strains on the generation of tax revenues,
especially the ability of landowners to fulfill their tax duties; once the population
grew to a point that would exhaust much of the economic surplus and thereby
handicap taxpayers’ ability to fulfill their duties to the state, causing the dis-
appearance of the state’s cash reserves and shortage in the state’s revenue, the
equilibrium would disappear.
In Figure 3.2, which illustrates the equilibrium, the U-shaped curve indicates
the conditions of demand (y-axis) and supply (x-axis) in Qing finance. From the
inauguration of the Qing in China in 1644 to the pacification of the country in
1681 (point “b” on the curve), the demand (mostly from military expenditures)
surpassed the supply in most of those years, and the gap was the widest at point
“a” when the deficit in the Qing budget reached the highest level. The equilib-
rium came into being only between points “b” and “d” (from the 1680s to the
1830s) when the supply surpassed the demand; once the difference between
supply and demand reached a high level (at point “c”), the state would reduce
the supply by the measures of regional or universal tax exemptions. But the
equilibrium in Qing finance was temporary and at a low level: it would disappear
once there was a change on either the demand or the supply side. In actuality,
what brought that equilibrium to an end were precisely the following two
64 The formation of the Qing state

Figure 3.2 The low-level equilibrium in the fiscal constitution of the Qing state

factors: first, a change on the supply side, or a strain on revenue generation in

the late eighteenth century because of the increased population pressure and
a reduced size of economic surplus available for the state’s extraction; and second,
a sudden increase in demand caused by large-scale internal revolts and unprece-
dented external threats in the nineteenth century which finally toppled the balance
between supply and demand in the Qing fiscal system.
In fact, except for the first four decades of the Qing when the suppression of
domestic resistance, especially the rebellion of the Three Feudatories
(1673–1681), incurred a huge amount of spending on military campaigns, the
Qing court was able to maintain an equilibrium in its fiscal constitution until
the mid-nineteenth century. Its revenues were slightly higher than its routine
expenditures, and the Board of Revenue thus was able to generate a cash
reserve, that is, the balance of the state’s total revenue after deducting its regular
expenditures. Theoretically, the cash reserve would increase indefinitely when
the state’s annual income was higher than its spending, thus producing a margin
to add to the cash reserve year after year. The cash reserve, which peaked at
a range between 70 and 80 million silver taels in the 1770s, or nearly twice of
the Qing state’s total annual revenue, was critically important for the state to
meet its extra needs of spending on unexpected events, such as disaster relief,
water-control projects, or military operations. For the Qing court, whether or
not to launch a military expedition on its frontier depended not only on its per-
ception of the severity of the threat to its geopolitical security but also on its
affordability, which depended in large measure on the size of the cash reserve.
 Limits to territorial expansion 65
Not surprisingly, the level of the Qing court’s cash reserve rose and fell in
a number of cycles from the late seventeenth throughout the first half of the
nineteenth century before the equilibrium was destroyed; these cycles are key to
understanding the Qing state’s recurrent efforts at frontier-building.

The affordability of the Qing economy

It was noted earlier that until the early nineteenth century the tax burden in Qing
China accounted for only a small fraction (3–5%) of its total economic output.
That does not mean, however, that the tax duties were always easily affordable to
taxpayers. To understand their affordability, we need to consider several factors:
the total economic output, especially the yield of agricultural production that con-
stituted the major source of government revenues through the collection of land
taxes; the amount of economic output needed to maintain the subsistence of the
population; and the remainder of the total economic output after deducting the
portion for consumption by the population, that is, the amount of economic sur-
plus potentially available for extraction through taxation.
The studies by economic historian Guo Songyi (1994, 1995, 2001) offer so
far the most exhaustive and reliable estimate of agricultural output in Qing
China. Based on more than 1,000 data items on crop yields per land unit in
different parts of China and different periods of the Qing as well as more than
3,000 items of data on the yield of individual crops in more than 400 counties
and prefectures, Guo assesses the total grain output, the amounts of grain yield
per mu (1 mu equals 0.16 acre) and per cultivator, and the number of mouths
that a cultivator could feed in different periods of the Qing, as shown in
Table 3.2.
To further determine the amount of surplus grain or the remainder of grain
after self-consumption, Guo first arrives at the gross amount of grain per person

Table 3.2 Agricultural productivity during the Qing, 1600–1887

Year Total amount of Total size of Grain Total Grain yield per farmer
grain yield (in grain land yield number of
1,000 catties) (1,000 mu) per mu farmers (catty) Number of
(1,000) mouths fed

1600 171,601,741 669,946 256 26,359 6,510 8.3

1766 289,074,380 932,498 310 41,081 7,037 8.9

1790 286,151,985 908,419 315 60,251 4,749 6.0
1812 301,298,820 944,695 319 70,293 4,286 5.4

1887 290,835,468 1,013,364 287 81,138 3,584 4.6

Note: 1 catty equals 1.1 pounds.

66 The formation of the Qing state
in rural households by dividing the gross amount of grain in China by the
number of people involved in agriculture; he then determines the per capita
amount of processed grain ready for consumption by deducting 20% as
a production cost from the gross amount and further deducting 58% for husks
left after grain processing that were usually not for consumption. The average
amount of surplus grain per capita is the remainder of the net amount of pro-
cessed grain after deducting 350 catties for self-consumption (see Table 3.3).
Guo’s estimates, to be sure, do not cover the agricultural population’s income
from cash crops (such as cotton, tea, sugar cane, hemp, mulberry trees) and
family sideline activities (handicraft, fishing, husbandry, etc.). But the foremost
importance of grain production in agriculture was obvious: it occupied about 90%
of all cultivated land in Qing China (Xu and Wu 2003a: 221) and about 80% of
peasant households’ income from the land.9 Since our purpose here is to look at
the long-term changes in the levels of agricultural productivity and economic sur-
plus from agriculture, rather than an accurate estimate of total economic output
in a given period, Guo’s figures, incomplete as they are, are particularly relevant
here, for they clearly show such changes.
First of all, it is clear from Tables 3.2 and 3.3 that the Qing economy showed
an expanding momentum until the 1760s, and the increase in grain production
(1.69 times in 1760 that of the 1600s) was accompanied by a largely compar-
able expansion in the population (1.66 times) and cultivated land (1.43 times)
in China. In other words, most of the increase in grain yield came from the
enlarged arable land, and the vast majority of the increased population was
absorbed by the newly reclaimed land or sparsely inhabited areas through
migration.10 As a result, while the total size of the rural population increased
remarkably during the Qing before the 1760s, the average size of arable land
per farmer remained largely constant, hovering around 25 to 27 mu per capita.
This means that population growth did not cause a severe strain on the land or

Table 3.3 Agricultural production and population during the Qing, 1600–1887

Year Total Agricultural Arable land Grain shared by agricultural

population population population (catties per capita)
(1,000) (1,000) Total size Per
(1,000 farmer Gross Net Surplus
mu) (mu)

1600 120,000 97,200 725,464 27.52 1,765 819 469

1766 200,000 170,000 1,036,109 25.22 1,700 789 439
1790 300,000 255,000 1,009,354 16.75 1,122 521 171

1812 350,000 297,500 1,050,436 14.94 1,012 470 120

1887 400,000 340,000 1,125,960 13.88 855 397 47
 Limits to territorial expansion 67
a significant increase in the population–land ratio. In other words, there was no
grave population pressure in China before the 1760s. Although there was no
significant improvement in technological or capital inputs, the expanded farm-
land and the use of preexisting techniques of intensive farming that had been
developed in the preceding centuries allowed farmers to produce enough to sup-
port the enlarged population without lowering its living standard. In the
absence of a revolution in farming techniques and inputs, to be sure, there
could not be any significant increase in labor productivity. However, land rec-
lamation and migration effectively alleviated the possible pressure of the
increased population on the land and avoided the process of labor intensification
in agriculture and the subsequent decrease in labor productivity. This is evident
in the fact that, by as late as the 1760s, the amount of grain produced by an
average farmer could still support as many as nearly nine individuals, and each
individual of the agricultural population had an additional 439 catties of grain as
surplus after self-consumption. Altogether, the enlarged arable land, the stable
level of labor productivity in agriculture, and an increase in the sheer size of the
agricultural surplus explain in large part not only the unprecedented prosperity
of the Chinese economy in the mid-eighteenth century but also the affordability
of the tax burden to the population and the large surplus in the Qing state’s
cash reserves during the same period.
But the situation changed a lot in the late eighteenth century, as population
pressure became increasingly palpable. While the population grew from roughly
200 million in 1766 to 350 million in 1812, the land available for reclamation
became scarce during the same period (in fact, there was almost no increase in the
total acreage of arable land during this period), hence a steady decline in the average
size of arable land per farmer to 16.75 mu in 1790 and 14.94 mu in 1812 (less than
60% of its 1766 level). To secure the subsistence of the increased population, several
developments occurred after the mid-eighteenth century or became more observable
than before, which in turn evidenced the existence of a mounting population pressure
on the land in the late eighteenth century and afterwards. These include:

1 the wide cultivation of maize and potato, which had been introduced from
America as early as the mid-to-late Ming period but which did not see their
widespread use in different parts of the country (especially the hilly areas
that had been newly inhabited by immigrants) until the period from the
mid-eighteenth century through the early nineteenth century (Guo Songyi
2001: 384);
2 the practice of double cropping (wheat in winter and rice in summer, or
growing rice twice a year) in the lower Yangzi region and triple cropping in
the southeastern region, which had been rare in the Ming but increasingly
observable after the 1750s and more rapidly promoted in the early nine-
teenth century (Min Zongdian 2003);11
3 the Qing rulers’ reassertions of the policy of permanent exemption of taxes
and strict prohibition of the registration of newly reclaimed land in the mid-
to-late eighteenth century;12
68 The formation of the Qing state
4 most importantly, the sharp decline in labor productivity in agriculture after
the 1760s, as evidenced in the decrease in grain yield per farmer (from
7,037 catties in 1766 to 4,749 in 1790 and 4,286 in 1812) and grain sur-
plus per capita (from 439 catties in 1766 to 171 in 1790 and 120 in 1812)
(see Tables 3.2 and 3.3).

All these signs suggest the effects of a mounting population pressure on agri-
culture in the last few decades of the eighteenth century and their implications
for the Qing state’s fiscal conditions. As the reclaimable land was exhausted and
emigration slowed down, most of the increased population had to be absorbed
locally, hence the intensification of labor input in farming and greater involve-
ment in off-farm income-making activities. It was likely, therefore, that eco-
nomic growth in the Qing was experiencing a transition from the traditional
type driven primarily by the booming market and, behind it, the specialization
and commodification of production in agriculture and other sectors (known as
the Smithian model of economic growth) before the 1760s to a new type pro-
pelled mainly by labor intensification under the mounting population pressure
(the Malthusian model) afterwards. The result of this transition was an enlarged
size of the rural economy and the total economic output in China but a decreased
size of the economic surplus, much of which had been exhausted by the increased
population, as clearly evidenced in the sharp decline in the per capita amount of
surplus grain in the late eighteenth and early nineteenth centuries. Thus, although
the government’s tax revenues, which remained largely unchanged, constituted
a smaller proportion than before when compared to the expanded economy, the
decrease in economic surplus greatly curtailed the ability of the population to fulfill
its tax duties to the government as well as the government’s commitment to “ben-
evolent governance.”13
After the 1790s, this greatly reduced size of economic surplus was coupled by
two new developments that further undermined landowners’ tax-paying ability.
One was the drainage of silver from the Chinese market by the rapidly develop-
ing opium trade and hence the escalation of the price of silver and the actual
increase in the tax duties that had to be paid in silver (Peng Xinwei 2007:
629–645). Price inflation, coupled with the increasing expenditure of the gov-
ernment in response to rapid population growth, made it difficult for the
income from the regularized and fixed surcharge of the “melting fee” after the
enforcement of huohaoguigong (see Chapter 2) to satisfy the actual needs of
county governments, which gave rise to the collection of various customary fees
and gifts (lougui) from taxpayers or subordinates as a solution to the worsening
deficit in government spending and salary payment (Zelin 1984: 298–301). The
other development was the government’s turning to the illegal practice of tax
farming by profiteering merchants to ensure its timely completion of tax collection,
which was widespread in the collection of commercial levies (S. Mann 1987). Tax-
payers, for their part, also turned to the privileged local elites for the illegal business
of proxy remittance (baolan) to reduce their tax burden (Bernhardt 1992). All
these led to the growing tension between the government and local communities,
 Limits to territorial expansion 69
which surfaced first in the mid-nineteenth century as clearly seen in the revolt in
Leiyang county of Hunan province in 1843 (Kuhn 2002: 80–91) and heightened
in the early twentieth century when tax resistance became increasingly frequent
(Prazniak 1999; Thornton 2007: 85).
To conclude, while the Qing state’s peculiar geopolitical setting explains the
necessity of its military operations on the frontiers, the low-level equilibrium in
its fiscal constitution determines both the possibility and limits of its war efforts.
The immense size of taxable land and the taxpaying population within China
proper brought the Qing court revenues that not only satisfied its routine
expenditures but also generated an ever-increasing surplus during times of
peace, making possible the unprecedented military prowess and the territorial
expansion that happened by the mid-eighteenth century. But the huge quantity
of its cash reserves quickly dwindled as the three preconditions that sustained
the equilibrium (optimal geopolitical setting, population-to-land ratio, and
stable silver price) disappeared in the late eighteenth and early nineteenth cen-
turies due to the phenomenal population growth, the drainage of silver from
the domestic market, and, most fatally, the arrival of European powers. This
reversal in its fiscal condition, in turn, explains much of the Qing court’s concili-
atory strategies in handling geopolitical crises in the decades to come.

Qing China in world historical perspective

The above discussion on the effects of the Qing state’s fiscal mechanism on its
war efforts and the socioeconomic factors behind the functioning of this mech-
anism offers us another vantage point to further think about the nature of the
Qing state by comparing it to the Ottoman empire and the emerging nation-
states in early modern Europe.

The Qing state and the Ottoman empire

Both the Qing and the Ottoman empire emerged as conquest dynasties,
expanding their territories to encompass the vast areas inhabited by peoples of
different ethnic and religious backgrounds. But the motivations behind their
conquests were different. The Ottoman warriors embarked on incessant offen-
sives primarily out of their zeal for Islamic conversion; spreading their faith by
waging the Holy War motivated the sultan and his cavalry more than any other
factor (Baer 2008). There were, however, also economic and social reasons
behind their war efforts. The practice of dividing the newly conquered areas
into sections and granting them as non-hereditary or hereditary revenue assign-
ments (timars) to the ruling elites and fighting men for military service, for
instance, generated a strong incentive for expansion (Streusand 2010: 81, 208);
the constant need for more land forced the rulers to undertake conquests con-
tinuously. Therefore, the entire state apparatus of the Ottoman empire, includ-
ing its military organizations, civil administration, landholding, and tax systems,
was geared to the needs of military expansion and colonization into the land of
70 The formation of the Qing state
the infidel. For the Ottomans, war was essential to the operation of their state,
and the concept of perpetual war to defend and spread their faith was integral
to their world view. Their territorial expansion would not stop until military con-
quests reached a limit imposed by climate, geography, and the transportation
technologies available to them. Once the expansion came to a standstill and no
more resources could be obtained, however, the oversized bureaucratic and mili-
tary systems that had been built for war and were reliant on military operations
would contract and deteriorate (Guilmartin 1988; see also Lewis 1958). In sharp
contrast, religion played no role in motivating the Manchus’ conquest of China
or the incorporation of the Inner Asian frontiers into their territory; nor was there
a major economic reason to incentivize their war efforts after the 1640s. Instead,
their primary goal was to replace the Ming as the legitimate dynasty ruling China
proper. Therefore, their conquest was limited to the land that had been ruled by
the Ming. Once this objective was achieved, the Manchus lost the incentive to
conquer any further. Their campaigns against the Zunghars were largely reactive
and preemptive, taking place only after the latter repeatedly invaded Outer and
Inner Mongolia and Tibet, constituting a constant threat to the geopolitical
security of the Qing, as shown in the preceding chapter.
A starker contrast between the Qing and the Ottomans lay in their different
ways of generating revenues. Like the Qing, the Ottoman empire had a vast ter-
ritory holding a population of diverse ethnic and religious origins. While allow-
ing Egypt, North Africa, and most of the Arab world a high degree of
administrative and fiscal autonomy, and content with the reception of an annual
tribute or a fixed quantity of payment from local tax-farm holders (Streusand
2010: 102; Shaw 1976: 121–122), the imperial government established in the
late fifteenth and sixteenth centuries a centralized timar system for tax collection
and local control in most of its European and Anatolian provinces, which consti-
tuted the core parts of the empire. Here about 85% of the arable land was
worked by individual farming households, thus called miri land, who owed
a tax to the central government. About half of the miri land, which was the
most fertile part, was under the government’s direct control, and the cultivators
on it had to pay a tax either to the imperial treasury or through tax farmers
from the capital or local provinces. Another half of the miri land was assigned
to the empire’s cavalrymen, who were allowed to collect taxes from the farmers
on the fief to compensate their military service, including the cost of raising
horses, and the government had the right to confiscate the fief from
a cavalryman who had failed to perform his military duties for seven years. But
this centralized system of taxation deteriorated in the seventeenth and eight-
eenth centuries despite the empire’s intermittent efforts to rebuild it, due in
part to miri farmers’ abandonment of their land under the ever-growing
burden of taxes, and in part to fief holders’ inability to provide military ser-
vices, thus giving rise to provincial notables turning the miri land into their
own estates, engaging in tax farming, and arrogating to themselves as much as
about two-thirds of the empire’s net tax revenues in the eighteenth century
(Kıvanç and Pamuk 2010: 609).
 Limits to territorial expansion 71
Unlike the diverse Ottoman populations who had to be grouped into mil-
lets of different sizes and structures based on their religions for administrative
and taxation purposes even in the capital city Istanbul and the core provinces
of the Balkans and Anatolia, the homogeneity of the Han population allowed
the Qing to implement a uniform bureaucratic hierarchy in the interior prov-
inces. The state channeled its power from the center all the way down to
every community through provincial and prefectural/county yamen, who
interacted with individual households through the quasi-official baojia organ-
ization or its local variants (Hsiao 1960; Ch’ü 1962; H. Li 2005). Further-
more, despite the huge size of the Han population, which was more than five
times the entire Ottoman population in the early seventeenth century and
more than 13 times the latter by the late eighteenth century, the Qing went
much further than the Ottoman empire in building a centralized fiscal system.
As a result, the central government in China controlled most of the land
taxes collected from landowners throughout the provinces; and it regulated
the collection and remittance of tax funds by curbing illegal practices, such as
charging taxpayers an extra-statutory “melting fee,” and converting it into
part of the regular land tax but limiting it to 10 to 30% of the total tax dues
(Zelin 1984: 130). Tax farming, a dominant form of tax collection in the
Ottoman empire, was banned in land taxation under the Qing before the
nineteenth century (Wang 1973; Bernhardt 1992).
However, the higher level of centralization that the Qing state achieved did
not translate into its longevity any more than that of the Ottoman empire.
A fundamental weakness of the Qing lay in the low-level equilibrium trap that
characterized its fiscal system. The Qing state’s fiscal system in the eighteenth
and the first half of the nineteenth centuries operated on a relatively fixed
budget for routine expenditures, which was slightly highly than the revenues
derived primarily from land taxes that were set at a very low and fixed level; the
stable and low levels of revenues and spending were in turn based on two pre-
conditions: its military supremacy and geopolitical security in relation to other
states, which allowed the Qing to keep its military spending to a fixed level; and
an optimal, if not unchanging, population-to-land ratio in China, which allowed
the Qing rulers to set the rate of land taxes at a low level and to guarantee the
ability of taxpayers to pay. Both preconditions prevailed in the early and high
Qing periods but disappeared in the late eighteenth and early nineteenth centur-
ies, when the ever-expanding population, together with the outflow of silver
from China to overseas markets, drained much of the economic resources avail-
able for the state’s extraction and crippled small landowners’ ability to pay tax,
while the arrival of Western powers caused a fundamental challenge to the Qing
state’s geopolitical security.
The Ottoman empire’s fiscal system showed a totally different dynamic.
Unlike the Qing that claimed its military supremacy over the neighboring states
and had a long period of peace in much of the eighteenth and early nineteenth
centuries, the Ottomans engaged in interstate wars one after another through-
out the history of their empire, fighting repeatedly with the Habsburgs,
72 The formation of the Qing state
Russians, Safavids, and other competitors for new territories or suzerainty over
the neighboring minor political entities. With much of its miri land already
assigned to the existing cavalry force, the empire had to keep expanding its terri-
tory; its revenues from the conquered land thus always lagged behind its
demands for more sources of revenues to sustain the military, which in turn
forced the imperial ruler to maximize its revenues by farming out land taxes to
local notables in the absence of an effective administrative system able to pene-
trate local communities. The equilibrium between supply and demand that pre-
vailed in the Qing state’s fiscal system in the eighteenth century rarely existed
under the Ottoman empire. Nevertheless, unlike Qing China where the steady
growth of the population by roughly three times from the early seventeenth to
the early nineteenth centuries resulted in the eruption of large-scale rebellions
by the impoverished peasants and thus destroyed the equilibrium in the end,
revolts of this magnitude were absent in the Ottoman empire, because its popu-
lation was not only small (less than one tenth of China’s population by 1800)
but also largely unchanging throughout the seventeenth and eighteenth centur-
ies. The Ottoman rulers were certainly nagged by the unrest among the rural
dwellers who lost their farmland in Anatolia, especially in the late sixteenth
century after a long period of population growth (Shaw 1976: 156, 174), but
the stable population-to-land ratio afterwards made them free of large-scale
peasant revolts that confronted the Qing rulers; instead, the domestic threats
to the Ottoman sultan came primarily from the rebelling soldiers when they
failed to receive the expected salaries from the government (Shaw 1976: 193,
196, 206, 211).

Qing China and early modern Europe

At the beginning of the sixteenth century, Europe was extremely fragmented,
encompassing up to 500 political entities of different sizes and with varying
degrees of autonomy and a monopoly of coercive power within their territories;
there were large empires and sovereign states as well as duchies, principalities,
bishoprics, city-states, and other smaller entities. The sovereigns of large states
tended to rule their populations indirectly, via the privileged and autonomous
intermediaries, such as priests, feudal lords, and urban oligarchies, who acted
independently and often resisted state demands that infringed on their interests.
Nor did the states have their own armed force on a permanent and regular
basis. Instead, military organizations were often heterogeneous, uncoordinated,
and in many cases dominated by mercenaries, usually belonging to local lords,
bishops, cities, guilds, and other communities, whose loyalty to the state was
conditional, often depending on the degree of their interest in the success of
the war. In sharp contrast, by the late seventeenth century, the armed forces in
different parts of Europe had become standardized and permanently controlled
by the states through a hierarchy of professional officers. To raise the enlarged
and upgraded forces, the state broadened its financial bases by increasing taxes
and customs duties. What made this possible was in turn a process of political
 Limits to territorial expansion 73
transformation, in which the state achieved a high level of uniformity in jurisdic-
tion, tax collection, and economic activities by weakening the political and mili-
tary roles of local interests and integrating many autonomous cities and
territories into the larger areas under its direct control (Tilly 1990: 38–47).
A key force driving this transformation in early modern and modern Europe,
as widely observed, was war. Warring with neighboring states forced the sover-
eigns to build effective military machines, which in turn necessitated the state’s
increased abilities in taxation, conscription, and resource mobilization, hence the
proliferation and differentiation of government apparatuses. Because of the cen-
tral importance of the use of fiscal resources for strengthening and monopoliz-
ing the means of coercion, historians have tended to describe the emerging
national states in Britain and continental Europe from the sixteenth to the
eighteenth centuries as “fiscal-military states” (Brewer 1989). Military revolution
played a key role in the rise and survival of such states, in which not only did
weapons, tactics, and strategies improve dramatically but the armies also became
larger, more sophisticated, and more permanent; raising the military and waging
a war also became more expensive, which in turn forced the state to enlarge and
elaborate the fiscal system as well as the entire administrative apparatus in order
to extract more taxes and exploit other sources of revenue, hence the transition
from a demesne state that generated its income primarily from its landholdings
to a tax state that based its revenue mainly on the taxable wealth of its subjects
(Brewer 1989; see also Glete 2002: 10–15; M. Mann 1986; Rasler and Thomp-
son 1989; Downing 1992; Storrs 2009).14
To some extent, early to mid-Qing China resembled the fiscal-military states
in early modern Europe. It had a centralized administrative apparatus that con-
trolled a well-defined territory through a professional bureaucracy; it had an
effective tax-extracting machine, and more than half of its revenues were used to
support the military; the state possessed a standing army that was larger than
any of its European counterparts; and it repeatedly engaged in campaigns to
expand its territory and consolidate its borders. It thus makes sense to categor-
ize the Qing as an “early modern” state that paralleled its counterparts in
Europe from the sixteenth century onward in several ways (Rawski 2004; Lie-
berman 2008). In her study of fiscal reform during the Yongzheng reign
(1723–1735), for instance, Madeleine Zelin depicts the Qing as “a dynamic
state struggling to devise its own formula for rational and efficient bureaucratic
rule” (1984: xv), which made eighteenth-century China comparable to its coun-
terparts in early modern Europe, because both of them shared the need for
fiscal stability when confronted with resource competition from within or out-
side the government. Likewise, Kent Guy (2010) finds parallels between, on the
one hand, the institutionalization of provincial governors in Qing China as
a means to enhance the ruler’s prerogatives by making preferential appointments
and, on the other hand, the formation of absolutist monarchy in Europe from
the mid-seventeenth to the eighteenth centuries. Nevertheless, a comparison in
the geopolitical, economic, and historical contexts of state formation also reveals
fundamental differences between the Qing and early modern European states.
74 The formation of the Qing state
Qing China differed from the European states chiefly in its geopolitical set-
ting. Unlike the European states who were members within the same interstate
system and who maintained a horizontal, if not equal, relationship with
one another, China appeared as the only dominant power within its geopolitical
universe. Instead of mutual competition or alliance on an equal footing, the
interstate relations here were vertical, at least symbolically and ideologically,
with the Qing at the top and almost all of the surrounding states as its tributar-
ies. The Qing warred with neighboring states only when the latter challenged its
suzerainty or threatened its geopolitical security. The lack of rivalry between
equals within the China-centered interstate setting had a profound impact on
the inner workings of the Chinese state. Unlike the situation in early modern
Europe where the constant competition and warfare forced the rulers of individ-
ual states to keep expanding and upgrading their armed forces and hence
increasing their expenses on the military, Qing China in the eighteenth century
saw a largely static picture of its military spending, organization, and training.
From the late seventeenth to the mid-nineteenth century, the number of regular
armies remained limited to roughly 800,000 to 850,000 soldiers, as mentioned
earlier. The same was largely true in the Qing state’s regular expenditure on the
military, which was fixed at the amount of approximately 17 million taels of
silver from the 1730s down to the mid-nineteenth century. The methods to
recruit, raise, and train the soldiers saw no significant changes throughout the
Qing until the late nineteenth century. Without imminent threats from outside
or inside China, the ruling elites were content with the preexisting military insti-
tutions and showed no interest in improving them. The rigid regulations of the
cost and standards of manufacturing weaponry and long-term price inflation fur-
ther made it impossible to upgrade their armaments (Mao Haijian 2005:
33–88). As a result of the poor training of soldiers during the extended period
of peace, insufficient funding, and deterioration of military equipment, the over-
all combating abilities of the Qing armed forces declined throughout the eight-
eenth and the first half of the nineteenth centuries.
The lack of substantial competition and challenges from outside China not
only led to the deterioration of the military capabilities of the Qing but also
accounted to a large extent for the overall stagnation in the development of its
administrative and fiscal apparatuses. Except for the measure taken by the
Yongzheng emperor (r. 1723–1735) to restructure the decision-making organs
at the top level and thereby enhance his own power, the bureaucratic system of
the Qing experienced no significant changes until the late nineteenth century.
The Qing rulers saw no reason to expand and upgrade the fiscal machine when
the existing amount of tax revenue was sufficient to cover its regular expend-
itures that remained largely constant and unchangeable in principle. What char-
acterized the fiscal regime of the Qing thus were its reliance on direct tax on
the land as its most important source of revenues, its highly centralized system
of tax collection and remittances, its exceptionally low rate of land taxes, its pro-
hibition on tax farming and other illegal activities in taxation, and therefore the
overall absence of peasant riots against taxation before the nineteenth century.
 Limits to territorial expansion 75
All these contrasted strikingly with the chaotic situation in early modern contin-
ental Europe, where the biggest challenge for any government was “how to
mobilize the resources of the state in time of war without arousing excessive dis-
content among the population at large” (Bonney 1988: 1). Constrained by
a limited tax base and a decentralized tax system, the French kings, for instance,
had to rely on borrowing from financiers, farming out the collection of indirect
taxes, and burdening the poorer peasants with an exorbitant direct tax (the taille)
for their financial needs, which inevitably resulted in the growth of the king’s
debt, rampant corruption in the government, huge losses of state revenues, and
recurrent peasant rebellions (Bonney 1981, 1988).

Socioeconomic structure and state formation

Economic and social structures were as important as geopolitics in shaping the
process of state formation and the subsequent relationship between the state
and society. Both Charles Tilly and Michael Mann find a close link between the
availability of resources for extraction and the form of government in European
countries. Those with a large rural population and less commercialized econ-
omy, for example, would find it difficult to increase their revenues (mainly
through the collection of land taxes) for war and other government activities
and therefore were likely to develop an extensive fiscal apparatus and an absolut-
ist regime in order to “mobilize” in a coercive way monetary and manpower
resources from mainly rural areas. On the other hand, those with a highly devel-
oped commerce and a large pool of resources would find it relatively easy to
generate enough revenues by taxing commerce and the wealth of landed elites;
therefore, they were likely to avoid building a centralized bureaucracy and
instead develop a constitutional government (Tilly 1985: 172–182; M. Mann
1986: 456, 476, 479; see also Downing 1992: 9; Ertman 1997: 13). In his
study of European states, Tilly further distinguished three distinctive patterns in
which the economy conditioned state activities. First, in “coercion-intensive”
regions where agriculture predominated, the rulers turned primarily to head
taxes and land taxes for war-making and other activities and, to that end, cre-
ated large fiscal machines and left a wide array of power to local elites. Second,
in “capital-intensive” regions where the economy was commercialized, the
state turned to customs and excise duties that were convenient to collect as
well as readily available credit as sources of state revenue, which hence resulted
in limited and segmented central apparatuses. Between the two ideal types was
the third pattern in regions of “capitalized coercion,” where the state extracted
resources from both land and trade and thus created dual state structures in
which the landed elites confronted as well as collaborated with financiers (Tilly
1990: 99).
To what extent, then, is Tilly’s model of the three trajectories of state forma-
tion relevant to our understanding of the nature of the Qing state? Obviously,
both the capital-intensive path and the capitalized-coercive path do not apply to
eighteenth-century China, a predominantly agricultural country. This is clear
76 The formation of the Qing state
when it is compared to England, a country that took the path of capitalized
coercion. Although China’s total economic size was 7.7 times that of England
in 1700 and 6.3 times in 1820 (Maddison 2001: Table B-18), its industry and
trade accounted for only about 30% of its total economic output, whereas in
England, industry and trade contributed 45% to its GNP in 1700 and 55% in
1789 (Goldstone 1991: 206).15 The contrast between the two countries is even
starker when we look at the share of the taxes on industry and trade in their
respective government revenues. In China, it accounted for only about 17% of
the Qing government’s total income around 1700 and 30% around 1800 (Xu
and Jing 1990), while in England, it accounted for 66% in 1700 and 82% in
1789 (Goldstone 1991: 206). The relative unimportance of the taxes on indus-
try and trade in Qing China is obvious even when it is compared to eighteenth-
century France: although France’s economic structure was quite similar to
China’s (75% of its GNP was from agriculture in 1700, which remained as high
as 69% in 1789), its taxes from industry and trade accounted for 54% of its total
income in 1700 and 50% in 1789 (Goldstone 1991: 204–205).16
The varying importance of industry and trade to government revenues meant
a lot to the rulers in Europe and China. In the capital-intensive regions of
Europe, the rocketing cost of war drove the rulers to increasingly rely on capit-
alists (merchants, bankers, and manufacturers) for revenues through borrowing,
taxation, and purchase, without creating a bulky and durable state apparatus,
and the capitalists in turn took advantage of their economic and financial dom-
inance to shape the policies of the states that prioritized the protection and
expansion of commercial enterprise (Tilly 1990: 150–151). In countries of cap-
italized coercion, the expensive warfare and the growing needs for taxes, credits,
and payment of debts forced the rulers to bargain with the major classes within
the population over issues ranging from the right of election to the use of vio-
lence, which led to the creation or strengthening of representative institutions in
the form of Estates, Cortes, and eventually national legislatures (ibid.: 188).
None of these developments occurred in eighteenth-century China. Despite
the growing share of the taxes on industry and trade in the Qing state’s rev-
enues (from 13% or less at the beginning of the Qing to about 30% by 1800,
see above), their importance in its revenue structure remained marginal, and its
primary purpose to collect the mercantile-based taxes was to regulate the market
places rather than to increase its revenues (S. Mann 1987: 18–25). Land taxes
remained the major source of government income until the late nineteenth cen-
tury (Zhou Yumin 2000: 238–239). Although contributions from wealthy salt
merchants were an important supplement to the government’s extra expend-
itures during times of war and other exigencies, the Qing rulers never needed to
increase the taxes on industry and trade or to borrow from merchants and finan-
ciers for funds on war or catastrophe relief; they relied chiefly on their own cash
reserves for most of the unexpected expenses. Therefore, Chinese merchants
lacked any leverage to bargain with the state for political and economic privil-
eges, and the state indeed took no measures to promote the merchants’ stand-
ing in society or to encourage their expansion of commercial enterprise, except
 Limits to territorial expansion 77
for bestowing them an honorary title upon receiving a generous donation.
Quite the contrary, in the state’s ideology, agriculture was the very “root” (ben)
of the economy that had to be nurtured and protected, while commerce and
industry were only the “branches” (mo) that competed with the root for human
and material resources and therefore had to be restricted, as a famous saying of
Emperor Yongzheng evidenced, although in reality the Qing government also
implemented a series of regulations to ensure the normal operation of trading
and manufacturing businesses and the livelihood of ordinary merchants (Deng
Yibing 2004; Wang Rigen 2000).
Nor should China be equated with a coercion-intensive state. In coercion-
intensive regions such as Russia where the economy was uncommercialized and
the resources extractable from capitalists were limited, the state had to turn to
the means of coercion rather than negotiation and contracts to generate enough
revenues for war and state-making. The result was their enhanced exploitation
of cultivators by enforcing the system of serfdom and the construction of
a centralized bureaucracy to its full, ponderous form (Tilly 1990: 140–141). In
sharp contrast, the large taxpaying population and a relatively small size of the
military allowed the Qing state to keep the land taxes at a level that was easily
affordable to most of the landowners before the increase in population
exhausted much of the economic surplus in the late eighteenth and early nine-
teenth centuries. Therefore, the rulers in China felt no need to maximize their
extraction of rural resources by turning cultivators into serfs and even slaves or
building a bulky administrative machine. Instead, the Chinese state saw the
existence of a large taxpaying population of mainly small owner-cultivators and
their economic security as the fiscal foundation of the dynasty and, therefore,
took various measures, such as tax exemption, land reclamation, rent limitation,
and famine relief, to ensure their subsistence. And the state saw no reason to
extend its administrative arms beyond the county level as long as the existing fiscal
machine generated enough revenues to satisfy its regular and irregular needs. In the
absence of a formal state apparatus below the county level, local communities in
rural and urban China from the Song down to the late Qing period displayed
a degree of “autonomy” in self-governance. A variety of village regulations or com-
munity covenants was developed to define the duties and rights of community
members in tax collection, security control, local defense, and public benefits
(Hsiao 1960; Rowe 2001: 388–405). These functioned to benefit members of the
community and at the same time also satisfy the state’s needs for revenue extraction
and social control, rather than to exclude the state’s influences. The state did not
intervene with such activities unless disputes occurred with villagers that could not
be solved by the community itself and which would threaten their fulfillment of
their duties to the government (H. Li 2005).

The low-level equilibrium jeopardized

All in all, geopolitical relations and socioeconomic conditions in eighteenth-
century China gave rise to its characteristic form of government that was unlike
78 The formation of the Qing state
any of the three paths of state formation found in early modern Europe. The
constant competition for territory and military superiority among the European
powers explained the overall absence of equilibrium between their ever-
increasing demand for revenues and the sluggish growth of income from direct
taxes (primarily on the land), which in turn led to their quest for more income
through enhanced exploitation of the rural population and tightened the control
of society or taxation on commodities and borrowing from the rich, and hence
their development into coercion-intensive or capital-intensive states or some-
thing between. In sharp contrast, the lack of outside competitors of comparable
strength explained the largely unchanged size and armaments of the military in
early to mid-Qing China, and therefore its limited and fixed demand for rev-
enues, which in turn led to its limitation on land taxation – the major source of
its income. Despite its fixed amount and shrinking size in relation to the
enlarged economy throughout the eighteenth century, however, the tax revenue
of the Qing government not only satisfied its needs for regular expenditures but
also generated a sizeable surplus for unexpected spending before the ever-
growing population exhausted the extractable economic resources, a situation at
least true until the 1760s. Central to understanding the characteristics of the
Qing state, in a nutshell, was the prevalence of the peculiar equilibrium in its
fiscal constitution as well as in the relationship between the state and society,
which in essence reflected an optimal geopolitical and demographic situation
prevailing in China by the late eighteenth century.
That equilibrium, however, came into jeopardy in the last two or three dec-
ades of the eighteenth century when the ratio between the population and
economic resources in China became increasingly unfavorable, causing
a steady decrease in the government’s fiscal surpluses and consequently its
defensive strategies in handling border troubles. Meanwhile, it was during the
same period that the most advanced countries in Europe underwent the
Industrial Revolution. Before the Industrial Revolution, the European powers
could never become a real threat to China when their economies were based
on manual labor and organic energy; although the division of labor and com-
modification as well as the degree of population pressure drove their econ-
omies to expand, the growth was slow and incremental. None of the
European powers before the Industrial Revolution was able to rival China in
terms of both total economic output and the overall competitiveness of
manufactured goods. However, by the 1820s and 1830s when the Industrial
Revolution was first complete in Britain, the manufacturing capacities of the
island had exploded; what propelled its economic growth, which was truly
exponential, was primarily the use in production and transportation of
machines propelled by fossil energy, rather than the traditional dynamics in
the Smithian or Malthusian model.17 For the first time, the European capital-
ist economy became a truly global force: “the need of a constantly expanding
market for its products chases the bourgeoisie over the entire surface of the
globe. It must nestle everywhere, settle everywhere, establish connexions
everywhere” (Marx and Engels [1848] 1969). China was inevitably to
 Limits to territorial expansion 79
become involved in the global wave of capitalist expansion that was accom-
panied with the advent of military threat from the European powers. Unfor-
tunately, the Qing state was ill prepared. Despite its success in handling the
crises on its northwestern frontiers, until the mid-nineteenth century the
Qing court never perceived its geopolitical issues from a global perspective
and never developed an integrated foreign policy that transcended its nar-
rowly defined, fragmented, frontier policies (Mosca 2013: 1–18). Even worse
than its failure to develop a grand strategy on a global scale, the Qing court
was faced with a fiscal crisis from within, which constrained its capacity to
tackle the new challenge from outside, as shown in the next chapter.

Notes
1 Other military operations included: the campaigns against the Miao in Hunan prov-
ince in 1737 and in Guizhou in 1741, which cost about 100,000 taels in total; the
war against the Tibetan tribes in Zhandui in 1745–1746, which cost more than
1 million taels; and the war against the Tibetan tribes in Jinchuan in 1747–1749,
which cost 3.3 million taels, or 1.1 million a year on average.
2 Other military expenditures include: 13 million taels on the operation in Burma
(1767–1769, or 4.33 million a year); 10 million on the campaign in Taiwan
(1787–1788, or 5 million a year); 1.3 million on the war against Vietnam
(1788–1789); 1 million on fighting the Gurkhas of Nepal (1788–1789) and another
11 million on the same enemy (1791–1792); and 15 million on suppressing the
Miao in Hunan and Guizhou provinces (Chen Feng 1992: 275).
3 In addition to the spending on suppressing the White Lotus rebels (averaging
16.66 million taels per year), there was an additional 15 million taels spent on defeat-
ing the Miao in Hunan and Guizhou in 1795–1797 (averaging 5 million), thus the
average expenses on wars in 1796–1797 were hypothetically 21.66 million taels
per year.
4 There were only two noteworthy military operations during those years. The first is
the nine-year (1802–1810) annihilation of the so-called “oceanic bandits” (yangfei)
led by Cai Qian, who frequently harassed southeastern coastal provinces; suppressing
the rebels cost a total of 7 million taels, or about 780,000 taels a year. The second
was the suppression of the rebellion by the Tianli Sect in 1813, which cost 110,000
taels in Anhui province (Chen Feng 1992: 274) and probably up to 1 million in the
whole country.
5 The only noticeable campaign in this decade was the suppression of the Yao people
in Hunan and Guangdong in 1832, which used 1.53 million taels (Chen Feng
1992: 275).
6 Ulrich Theobald notes that the funds for the first Jinchuan war “was financed by the
province of Sichuan” and that “the burden of 35 million liang spent for the western
campaigns was likewise distributed on the shoulders of various provinces and those of
the Ministry” (2013: 103), but he fails to point out that all these funds from individ-
ual provinces were actually the Coordinate Remittances that the provinces owed to
the Board.
7 Other sources included provisional impositions on residents in the regions where the
war took place, which, however, could be reimbursed sometimes in part or in full
after the war.
8 During the campaign against the Tibetans in Small Jinchuan in 1773, for example,
Jiang Chun, a merchant from the Huai River region, alone contributed 4 million
taels. In another instance, during the second campaign against the Gurkhas of Nepal
80 The formation of the Qing state
in 1791–1792, the contribution from salt merchants amounted to 6.1 million taels,
or more than half of the total expenditure (11 million taels) on this operation (Chen
Feng 1988, 1992: 334). These generous donations were possible because the mer-
chants had amassed a huge sum of wealth through their monopoly of salt trading;
each of the more than 100 large merchants in the Huai River region, for example,
was believed to have assets from a few million to 10 million and even as much as 70
to 80 million taels (Song Liangxi 1998).
9 During the period from 1914 to 1918, grain production accounted for 79.78% of
the total volume of agricultural output and 83.11% of the total market value of agri-
cultural output (Xu and Wu 2003b: 1098). The role of grain production in agricul-
ture in the Qing should not be too different from this estimate.
10 Multiple factors accounted for the phenomenal economic growth in the eighteenth
century. Aside from the long period of peace, the reclamation of the land, the settling
of new territories, the expansion of foreign trade, and the introduction of new crops
from America, historians have also underscored the growing number of free, inde-
pendent peasant households in south China following the disappearance of large
estates in the course of the Qing conquest and the marketization and specialization
of production as important reasons behind economic expansion (B. Li 1998; Gold-
stone 2004).
11 Of the 159 instances of double cropping of rice found in local gazetteers during the
Ming and the Qing period, 113 of them (or 71%) occurred after the 1750s (Min
Zongdian 1999).
12 The ban on the registration of newly reclaimed land and hence its taxation was first
included in the Qing code in 1740. In the same year, Emperor Qianlong reasserted
the policy of allowing people to freely reclaim the scattered pieces of wasteland with-
out tax obligations. In 1773, Emperor Qianlong once again stressed the prohibition
of taxing the newly reclaimed land by denying the existence of uncultivated land in
China proper; in his view, it was only in the Urumqi region of Xinjiang province
where land was still available for reclamation (Zhang Yan 2008: 123–124).
13 For a discussion on the decline in the Qing ruler’s commitment to “nourishing the
people” after the mid-eighteenth century, see Dunstan (2006).
14 War-making, to be sure, was not necessarily the most important factor accounting for
state-building in early modern Europe, and it was not always conducive to the rise of
“modern” state forms; war could also lead to the destruction of the local economy,
crises in state finances, and the breakdown of a centralized state. The growth of mili-
tary forces, likewise, was often accompanied by corruption, nepotism, and decentral-
ization – not to mention pillage and plunder in wartime – instead of their
rationalization, the expansion of the power of the state, and provision of protection
to the civilians within their territory. And the development of the fiscal systems to
sustain states at war frequently involved negotiation, compromise, and distortion
(Gunn 2010). Other factors such as jurisdiction, religion, and ideology were no less
important than the development of fiscal-military institutions to the rise of centralized
national states. Thomas Ertman (1997), for his part, traces the origins of different
political regimes (absolutist or constitutional) at the end of the early modern period
to their very starting conditions, especially the various patterns of local government in
different parts of Europe in the centuries following the demise of the western Roman
Empire; he stresses, in particular, the timing of the onset of sustaining geopolitical
competition as an important factor explaining the varying nature of state infrastruc-
tures across the continent by the late eighteenth century. Others underscore the role
of internal judicial, administrative, and ideological developments in the process of
state-building (e.g., Strayer 1970: 35–89; van Creveld 1999: 126–184; A. Harding
2002: 295–335). For legal historians, the development of a “law state” in late medi-
eval England, as seen in the proliferation of litigation and legislation, for example,
 Limits to territorial expansion 81
was no less important than the development of a “war state” (Kaeuper 1988:
134–184; Harriss 1993; Gunn et al. 2008).
15 According to Angus Maddison (1998), industry and trade accounted for 31.5% of
China’s GDP in 1890, which was essentially no different from its economic structure
back in the late eighteenth and early nineteenth centuries. Liu Ruizhong (1987) esti-
mated that the share of industry and trade in the Chinese economy was 30% in 1700,
33% in 1750, and 36% in 1800. As for the Ming dynasty (1368–1644), it is estimated
that industry and trade accounted for 10% of its economy in the early period of its
history and about 20% in the rest of the dynasty (Li and Guan, n.d.).
16 There were, to be sure, enormous differences between England and continental
Europe. The highly centralized tax system, its increasing reliance on indirect taxes, its
abolition of farming indirect taxes, and hence the rapid increase in the English state’s
revenues that far outpaced its economic growth, all these lead Patrick K. O’Brien to
dub the fiscal regime in England from the late seventeenth to early nineteenth cen-
turies as “fiscal exceptionalism,” which contrasted with the decentralized, corrupt,
and inefficient tax systems based on inflexible regional quotas and the prevalence of
tax farming in continental Europe during the same period (O’Brien 2002).
17 For further discussions on the different types of economic growth and their implica-
tions for understanding late imperial Chinese economy, see Feuerwerker (1992);
Wong (1997: 43–52); Goldstone (2004).
Part II

The transition to
a sovereign state
4 Regionalized centralism
The resilience and fragility of the late
Qing state

Beginning in the 1790s, the Qing state encountered two fatal challenges in
succession. One was the expansion of the Europe-centered system of states,
driven by the unmatched military prowess and industrial productivity of Western
countries, which were in turn bolstered by the proliferation of capitalism in the
economy and the application of modern scientific knowledge to the military and
production. China’s collision with this system in the nineteenth century inevit-
ably led to the breakdown of the interstate order in East Asia, the transform-
ation of the Chinese dynasty into an ordinary sovereign state, and its integration
into the European interstate order on the basis of legal equality among sover-
eign states. China, however, would have had a better chance to absorb the
impact of the collision, to accommodate the new interstate relations, and to
strengthen itself by actively and selectively borrowing from Europe, had the
clash taken place before the late eighteenth century when the Qing was eco-
nomically dynamic, fiscally solvent, and militarily well-built. Unfortunately, the
advent of the European challenge coincided with the occurrence of another
crisis, that is, the explosion of the population, which exhausted most of the eco-
nomic surplus and undermined the Qing state’s fiscal health as well as domestic
stability as seen in the Taiping Rebellion (1851–1864). Nevertheless, the Qing
survived the fiscal and geopolitical crisis; even more surprisingly, it witnessed
three decades of “Restoration” from the mid-1860s to the mid-1890s and
another wave of modernization in the last decade (1901–1911). This and the
following two chapters, therefore, will discuss how the Qing survived these
two crises and, more importantly, how China became a sovereign state for all
the setbacks and pitfalls it encountered in this period.
Earlier studies of late Qing China generally underscored various forms of
incompetence, corruption, and frustration in its modernizing endeavors.1 This
narrative of failure mirrored the modernization paradigm in social sciences in
the 1950s and 1960s, central to which was a shared assumption about the fun-
damental incompatibility between the cultural values of a “traditional” society
and the requirements of modernization. Joseph Levenson’s interpretation of Con-
fucianism (1968) best exemplified this paradigm in the historiography of modern
China (see also Cohen 2010). The prevalence of the failure narrative also had
to do with the comparative perspective characteristic of the modernization
86 The transition to a sovereign state
paradigm; scholars frequently highlight the “failures” of late Qing China by
comparing it to the “successes” of Meiji Japan.2
More recent scholarship on late Qing China has challenged the failure narra-
tive. The reason why the Self-Strengthening movement has long been seen as
a failure, according to Benjamin Elman, was because of China’s defeat by Japan
in 1894 that abruptly ended the program. However, he argues, China had taken
the lead for years since the 1860s in East Asia in importing Western science and
technology through translations, manufacturing, and military modernization. It
was the disastrous result of the Sino-Japanese war, which few in and outside
China could predict, that caused contemporary observers as well as historians to
retrospectively interpret China as a loser in the competition with Japan during the
Self-Strengthening period (Elman 2004, 2006). Rather than a failure, Stephen
Halsey (2015) shows the impressive accomplishments of a famous steamship navi-
gation company and other government-sponsored enterprises in the 1870s and
1880s in breaking the Western monopoly on modern shipping and developing
industries in China.3
But the most important achievement of the late Qing state is no doubt its
success to defend most of its territories and its transition to a sovereign state.
This chapter, therefore, focuses on an examination of the institutional context in
which those achievements took place. At the core of this context is the trans-
formation of the Qing state’s fiscal constitution, which had a profound impact
on the relationship between the central government and provincial authorities.
It was the changes in this relationship that accounted for both the phenomenal
achievements of the modern initiatives as well as the emergence of new prob-
lems that were fatal to the Qing in the end.

Fiscal regionalization

The low-level equilibrium deteriorated

The best indicator of the Qing fiscal condition before the 1850s, as explained
in Chapter 3, was the size of the surplus of the cash reserve at the Board of
Revenue. A turning point in this regard was the White Lotus Rebellion
(1796–1804), which exhausted most of the reserve, reducing it from above
70 million taels before the rebellion to less than 20 million by the end of the
rebellion because of the massive military spending incurred. The Qing state saw
relative peace both domestically and on the border in the next four decades
until the outbreak of the Opium War in 1840, except for a few small-scale mili-
tary operations that cost 4.5 million taels a year on average (see Chapter 3). Yet,
the surplus level of the cash reserve remained limited to a low level, ranging
largely between 20 and 30 million taels in most of the four decades; the only
exception was from 1828 to 1830 when it reached 32 or 33 million taels. This
was in sharp contrast with the rapid increase in the surplus level during the
times of peace in the late seventeenth and eighteenth centuries. The situation
worsened steadily in the 1830s on the eve of the Opium War when the surplus
 Regionalized centralism 87
decreased from 32 million taels in 1830 to 22 million in 1833 and 10 million in
1840 (see Figure 3.1); this was surprising, given the fact that China enjoyed
almost complete peace without major military operations in that decade. Even
more surprising was that this decline in the cash surplus was only the figure on
paper; the actual amount was significantly lower than the official record.4
Two basic factors explain why the fiscal condition of the Qing state deterior-
ated in the first half of the nineteenth century, despite the fact that the country
saw no major troubles at home or on the border. The first is population
growth, which increased from about 120 million in the 1640s to 200 million in
the 1760s and 300 million in the 1790s, hence a steady decrease in the size of
arable land per farmer, from 27.52 mu in 1600 and 25.22 mu in 1766, to
16.75 mu in 1790, 14.94 mu in 1812, and 13.88 mu in 1887 (see Table 3.3).5
As a result, the amount of “surplus grain,” or the total output of grain after
deducting the amount for self-consumption by the farming population, also
decreased from 439 catties per capita in 1766 to 171 catties in 1790, 120 catties
in 1812, and only 47 catties in 1887 (Table 3.3), which crippled land owners’
abilities to fulfill their tax duties to the state. The second factor is the drainage
of silver from the domestic market because of opium smuggling on the south-
east coast and the subsequent deficit of foreign trade on the Chinese side. It is
estimated that to offset the deficit that increased steadily in the early nineteenth
century, the silver that left China totaled more than 11 million taels between
1800 and 1809 (Li Qiang 2008: 261), 20 million taels between 1823 and
1831, another 20 million taels between 1831 and 1834, and more than
30 million taels between 1834 and 1838, thus amounting to more than
100 million taels altogether during the 15 years from 1823 to 1838, or 2.5
times the Qing state’s annual revenue (roughly 40 million taels). An immediate
result of the continuous outflow of silver from China was the growing value of
silver tael, which was used mandatorily in tax payment, in relation to copper
cash, which was used by ordinary people in everyday transactions. The price of
each silver tael almost doubled in the first four decades of the nineteenth cen-
tury, increasing from 700 or 800 wen in copper cash in the mid-1790s to 1,243
wen in 1816–1826, 1,234 wen in 1827, 1,420 wen in 1835, and 1,643 wen in
1840 (Deng Shaohui 1998: 38–39). Small owner-cultivators, therefore, had to
pay twice what they used to as tax to the government (Luo Ergang 1947: 16).
Combined, the reduced grain surplus under population pressure and the
doubled value of silver taels explain the increasing difficulties that confronted tax
payers and hence the escalating deficit that the Qing state faced in tax collection.
Compared to the official quota of land taxes (33.35 million taels) to be collected
annually in the entire country, which the Qing government had no difficulty in
achieving at its full amount in most years before the 1790s, it collected only
30.43 million taels or 8.75% less than its target in 1841, 11.27% less in 1842,
9.40% in 1845, and 21.07% less in 1849 (Liang Fangzhong 2008: 573).
By and large, however, the low-level equilibrium continued into the 1840s,
with the traditional fiscal system remaining functional on the eve of the Taiping
Rebellion. Thanks to the huge amount of cash reserves in surplus that had
88 The transition to a sovereign state
accumulated over the Qianlong reign, the Qing was able to survive the White
Lotus Rebellion in the late 1790s and early 1800s without having to overhaul
the tax system; by turning in part to its already decreased cash surplus and in
part to the treasuries of the southeast provinces, the Qing again survived the
Opium War without the destruction of its fiscal equilibrium.

The regionalization of fiscal power

What brought the low-level equilibrium to an end was the Taiping Rebellion.
Multiple reasons explain why the rebellion swept through South and Central
China and became an unprecedented threat to the Qing dynasty. Aside from
the lingering anti-Manchu feeling among the Han Chinese, to which the
rebels appealed in agitating the populace, and the heightened tax burden that
threatened the livelihood of peasants, the ultimate reason behind the wide-
spread rebellion, as Luo Ergang (1958) argues, lay in the enlarged population
and its pressure on the available resources, which led to the drainage of the
economic surplus, the impoverishment of the rural population, and the subse-
quent emergence of uprooted people in large numbers (see also Ho 1959:
183, 274).
Given the rampancy and persistence of the rebellion and the fact that it broke
out at a moment when the surplus of the Board of Revenue’s cash reserve had
decreased to its lowest level in history (merely 8.96 million taels in 1850 and
7.63 million taels in 1851 according to official records; the actual amount of
surplus in cash was only 1.87 million taels by the end of 1850, see Deng Shao-
hui 1998: 45), the challenge for the Qing government to meet the demand for
military expenditure to suppress the rebels was unprecedented; the cost of the
war was far beyond its affordability. The Qing court’s military expenses during
1850–1852 in Guangxi province alone, where the rebellion originated, exceeded
11.24 million taels; altogether, the expenses on suppressing the rebels during
the first three years of the rebellion in affected provinces amounted to
29.63 million taels, nearly 10 million a year on average (Peng Zeyi 1983: 127).
As a result, the Board of Revenue’s cash reserve decreased to as little as
0.22 million taels in June 1853. Likewise, the treasuries of the provinces
affected by the rebellion were either completely drained (such as Hunan in
August 1853) or left with only a few thousand taels (such as Hubei and Anhui
in late 1853) (Shi and Xu 2008: 58–60). The Qing court’s fiscal condition
reached its worst point when the Taiping rebels occupied the lower Yangzi
region, the most important source of the Qing state’s revenues. Subsequently,
the Board of Revenue’s official figure of accumulative surplus plummeted from
5.72 million taels in 1852 to 1.69 million in 1853 when the rebels occupied
Nanjing, and to 1.17 million in 1860 when the rebels captured Suzhou (see
Figure 3.1); the actual amount of the surplus was only 119,000 taels in 1853
and 69,000 taels in 1860 (Shi Zhihong 2009: 111). The Qing state’s treasury
was virtually exhausted, and the low-level equilibrium that characterized the
Qing government’s fiscal system before the 1850s no longer existed.
 Regionalized centralism 89
How then did the Qing court manage to meet the demands for massive mili-
tary spending during the years of the Taiping Rebellion and other revolts?
According to an incomplete estimate, the Qing spent a total of 170.6 million
taels as direct reimbursements for expenses on combating the Taipings as
reported by military and civilian authorities from different provinces (Peng Zeyi
1983: 127, 136). The actual expenses on the war, however, could have been
more than 290 million taels during the 14 years (1851–64), averaging
20.71 million taels per year (Zhou Yumin 2000: 153). In addition, the Qing
spent 31.73 million taels on the Nian Rebellion (1853–1868), 118.89 million
taels on the rebellion of the Hui people in the northwestern provinces
(1862–1873), and 101.07 million taels on small-scale revolts in the southwes-
tern and southeastern provinces (1851–1873). Altogether, the Qing govern-
ment’s spending on the revolts during the 23 years from 1851 to 1873 totaled
more than 541 million taels, or 23.55 million taels per year, which was roughly
60% of the Qing state’s annual revenue in the first half of the nineteenth century
(approximately 40 million taels). In addition, the expedition into Xinjiang in
1875–1878 cost another 70–80 million taels; the war with France cost
30 million taels; and the spending on the Northern Fleet alone totaled about
23 million taels for two decades by 1894 (ibid.: 267–277). To pay the war indem-
nities after the major defeats in 1840, 1894, and 1900 as well as other forms of
compensation to foreign powers, the Qing spent no less than 900 million taels in
total (Xiang Ruihua 1999).
In the absence of a minimal amount of cash reserve at the Board of Revenue,
the Qing court’s initial response to the huge spending at the beginning of the
Taiping Rebellion was to turn to two traditional methods without revamping
the tax system per se. One was selling official titles (juanli or juanna), which
started in 1851, originally intended for a period of one year but repeatedly
extended until 1879. This was effective in the first two years, generating 1 to
3 million taels a year in the early 1850s. The other was minting copper cash and
printing paper notes with a large face value. From 1851 to 1861, the large-
value copper cash and paper notes or paper cash issued by the Board equaled
60,249,000 taels in total, or 69.5% of the Board’s total revenues during the
same period, which greatly made up the imperial court’s deficit during the Taiping
Rebellion (Peng Zeyi 1983: 87–115, 146–149; Deng Shaohui 1998: 48–51; Shi
Zhihong and Xu Yi 2008: 70–72).
More fundamental than these two traditional solutions to the fiscal mire of
the Qing state after the outbreak of the Taiping Rebellion are measures associ-
ated with the devolution of power from the Board of Revenue to provincial
governors and governors general to raise funds on themselves by whatever
methods that were workable in the local area as the Xianfeng emperor ordered
in July 1853 (QSL, Xianfeng 97: 3–6-yichou). It was these new measures that
led to the restructuring of the tax system and the transition of the fiscal system
of the Qing state from the conventional type characteristic of its fixed sources,
and the rigid level of regular revenue and its fixed regular expenditures, to the
new one that was open, elastic, and dynamic.
90 The transition to a sovereign state
One of the measures of fiscal decentralization was collecting surcharges on
the preexisting land taxes, thus breaking the rule, and indeed a taboo, of “never
increase the land tax” (yong bu jia fu) faithfully observed by the Qing rulers
since 1712.6 Borrowing from domestic or foreign debtors was another import-
ant source for additional revenue during wartime.7 A third and the most import-
ant measure was the collection of a new tax, lijin, on goods when sold locally
or shipped to a new locality. First created in the vicinity of Yangzhou of north-
ern Jiangsu in 1853 and promoted countrywide in 1855, this transit tax was col-
lected in different ways in different provinces. Typically, however, a lijin was
levied on commodities at the locality of their origins and then recollected each
time they were transported to a different place. Since all of the lijin income
went to provincial treasuries in theory, rather than the Board of Revenue of the
central government, local governors or governors general and military author-
ities had a strong incentive to maximize their revenue from this source by creat-
ing as many lijin collection stations as possible along important transportation
routes within the area under their jurisdiction. In Hubei province, for instance,
more than 480 collection stations were set up before they were reduced to 86
in 1868 (Peng Zeyi 1983: 157). The rate of lijin varied from 5% to as high as
10% of the value of goods. Thus, after being transported through several collec-
tion stations within a local area, the total payment of lijin before the goods
were finally sold to consumers could be 20–30% or as high as 40–50% of the
value of the goods per se, most of which were embezzled by collecting clerks
and government officials (Peng Zeyi 1983: 159; Shi Zhihong 2009: 123). The
total amount of lijin collected throughout the country and officially reported
ranged from 14 to 15 million taels in the 1870s and 1880s (Luo Yudong 1970:
469). It is estimated that, because of the collection of lijin, the Qing state’s
mercantile-based taxes increased tenfold by 1908 (S. Mann 1987: 7, 45).

The fiscal constitution transformed

As a result, the fiscal condition of the Qing improved steadily in the three dec-
ades following the Taiping Rebellion. Compared to the Qing government’s
total revenue in the first half of the nineteenth century, which fluctuated by
around 40 million taels per year, the total amount of officially reported revenues
had more than doubled three decades later, reaching 80.76 million taels in
1889 and 89.68 million taels in 1891, the highest level before the outbreak of
the Sino-Japanese War in 1894 (see Table 4.1). It should be kept in mind, how-
ever, that the provincial authorities rarely reported to the Board the full amount
of revenues they had obtained; before the nineteenth century, the unreported
portion usually accounted for 20–30% of its actual income. During the three
decades following the Taiping Rebellion, this hidden income could be as high
as 40%. If we take into account the unreported portion, the total amount of the
Qing government’s annual revenue could be as high as 140 to 150 million taels
in the late 1880s and early 1890s (Shi and Xu 2008: 279; see also Table 4.1).
Table 4.1 The revenue and expenditure of the Qing state, 1841–1911 (in 1,000 silver taels)

Yeara Land tax Lijin Salt tax Maritime custom Other incomes Revenue Expenditure Balance

Official Realb

1841 29,432 4,958 4,208 38,598 55,000 37,340

1842 29,576 4,982 4,130 38,688 55,000 37,140
1845 30,214 5,074 5,511 40,799 58,000 38,816 1,796
1846 39,223 56,000 36,287 2,935
1847 39,387 56,000 35,584 3,803
1848 37,940 54,000 35,890 2,050
1849 32,813 4,986 4,705 42,504 61,000 36,444 556
1874 60,800 111,000
1881 82,349 150,000
1885 32,357 14,250 7,394 13,528 9,558 77,086 140,000 72,866 4,221
1886 32,805 15,693 6,735 14,366 11,669 81,270 148,000 78,552 2,718
1887 32,790 16,747 6,998 19,319 8,360 84,217 153,000 81,281 2,936
1888 33,224 15,565 7,507 17,754 13,742 87,793 160,000 81,968 6,423
1889 32,083 14,930 7,716 16,767 9,265 80,762 147,000 73,070 7,682
1890 33,736 15,324 7,428 16,710 13,609 86,808 158,000 79,411 7,397
1891 33,587 16,326 7,172 18,207 14,432 89,685 163,000 79,355 10,330
1892 33,281 15,315 7,403 17,623 10,742 84,364 153,000 79,645 7,719
1893 33,268 14,277 7,680 16,801 11,083 83,110 151,000 73,433 9,677
1894 32,669 14,216 6,737 10,674 16,737 81,034 147,000 80,276 758

(Continued )
Table 4.1 (Cont.)

Yeara Land tax Lijin Salt tax Maritime custom Other incomes Revenue Expenditure Balance

Official Realb

1899 33,934 12,160 13,050 22,052 27,132 108,328 181,000 101,566 6,762
1903 35,460 16,000 12,500 31,500 9,460 104,920 175,000 134,920 –30,000
1906 35,000 12,000 13,000 40,000 27,000 130,000 217,000
1908 21,000 234,820 276,000 237,000 –2,180
1909 263,219 310,000 269,876 –6,657
1911 48,101 43,188 46,312 42,139 117,221 296,962 349,000 338,652 –41,690

Sources: QCXW, 66: 8225, 8227–8229; 67: 8231–8232; 68: 8245–8249; SQYJ: 144–148; HBS, 1.1: 172–173; QSG, 125: 3704–3709.
Notes: aThe data for 1911 are budgeted estimates rather than the actual amounts of government revenues and expenditures (QCXW, 68: 8245–8246).
b
Estimates of the Qing state’s real revenues are based on its official revenues, assuming that the latter is 25% less than the former before the Taiping Rebellion,
40% less than the former between 1851 and 1907 (Shi and Xu 2008: 277, 279, 289), and 15% less than the former between 1908 and 1911 because of the
Qing court’s investigation of local fiscal conditions in 1908 that greatly narrowed the gap between reported and actual revenues (QCXW, 68: 8249).
 Regionalized centralism 93
What made possible the rapid growth of the state’s total revenue was primar-
ily income from the following two sources. One is the collection of lijin dis-
cussed above, which officially generated 14 million taels or more each year in
the 1880s and early 1890s, or 15–17% of the Qing state’s official revenue. The
other is the income from maritime customs, which increased from 4 or 5 million
taels a year in the 1840s to 14.47 million taels in 1885 and 23.51 million taels
(26% of the state’s total income) in 1891 (Deng Shaohui 1998: 99). As a result,
the structure of the Qing government’s revenues underwent a meaningful change
in the second half of the nineteenth century. Before the 1850s, the land tax
constituted the most important source of its revenue, accounting for 73–77%
of its total income; by the early 1890s, its share had reduced to about 40% or
less. On the other hand, the revenue from commercial sources (salt taxes, lijin,
maritime customs, etc.) increased from less than 30% before 1850 to about 60%
before 1894 (Table 4.1).
What needs to be further explained is how the revenues from lijin and mari-
time customs were allocated and used. As noted earlier, lijin was collected and
controlled by provincial authorities, but most of its officially reported revenue
was used for the Qing court’s countrywide purposes (the so-called guoyong) in
the form of various remittance quotas that went to the Board of Revenue,
whereas the portion of the reported lijin revenue for local provincial usages
(shengyong) rarely exceeded 10% (Luo Yudong 1970: 222–229). The same is
true for the revenue from maritime customs. The amount for “countrywide
usages” (guoyong) accounted for roughly 70% of the total revenue of maritime
customs in the 1870s through the early 1890s, whereas the portion for provin-
cial usages was only 13–19% (Shi and Xu 2008: 186–187). But there was
a fundamental difference between the two sources. Unlike the lijin income,
most of which (70–80%, see ibid.: 123) was unreported to the Board of Rev-
enue, the reported income from maritime customs, which were efficiently man-
aged by Westerners and largely free of the corruption and embezzlement of
Qing bureaucrats, was close to its actual level, and most of it was indeed used
for countrywide purposes, although provincial authorities continued to detain
most of the portion for “countrywide usages” on the grounds of revenue short-
ages or that they were in charge of those countrywide programs that were
ongoing in their own provinces, and the actual amount eventually remitted to
the Board of Revenue accounted for 20–37% of the total income from maritime
customs in the 1870s, and only 7–20% in the 1880s and early 1890s (ibid.:
182–183).
At any rate, the Qing state’s fiscal constitution experienced a transition from
relying on traditional sources of revenue, primarily the land taxes that remained
largely a fixed quantity, to increasingly drawing on indirect taxes. Compared to
its revenue before the transition, such as in 1849, when the land tax account for
77% of total government income, by 1885, the land tax only contributed 42%
to its total revenue, and the rest came from the taxes on commodities, primarily
the lijin, salt tax, and maritime customs. By the end of the Qing, the land tax
further declined to only 16% of the government’s total revenue, while the
94 The transition to a sovereign state
revenues from the lijin, salt tax, and maritime customs were 2.73 times the rev-
enue from the land tax.
The fiscal condition of the Qing state, therefore, significantly improved from
the 1860s to the early 1890s. This was true for the entire government system,
as evidenced in the doubling of its annual revenue during this period, as well as
for the imperial court in Beijing, as best seen in the increases in the cash reserve
at the Board of Revenue, which reached 8.06 million taels by the end of 1882,
9.85 million taels by 1883, and 10.38 million taels by 1891 (Shi and Xu 2008:
276), thus in sharp contrast with the emptiness of the Board’s treasury in most
of the 1850s. Indeed, it is no exaggeration to term the ten years from 1885 to
1894 before the outbreak of the Sino-Japanese War as a “golden decade” for
the late Qing state in terms of its fiscal solvency. This was a decade when the
country enjoyed domestic order and peace on the border, when its economy
recovered from war devastation and expanded, and when government revenues
not only grew steadily but also generated a sizable surplus. In each year of this
decade, the official revenues of the Qing government, averaging 83.57 million
taels, exceeded its official expenditures, averaging 77.59 million taels, by
a margin of more than 7%, and generated a surplus of 5.98 million taels
per year. As mentioned earlier, the actual income of the Qing government,
including the reported and the unreported, could be as high as 140–150 million
taels a year, almost twice its annual expenditures.

The fiscal autonomy of provincial authorities

But the transition in the sources of government revenues from the land to
mainly commerce and borrowing was not driven by the expansion of trade or
the growth of the non-agricultural economy, nor did it come with the establish-
ment of a centralized and bureaucratized state for the efficient extraction and
use of revenues, as often seen in the most successful fiscal-military states in early
modern Europe. Instead, the transition was a result of provincial governors’
response to the pressing tasks of dealing with external and internal exigencies;
the fiscal measures leading to the transition were typically introduced in
a passive, piecemeal, and corrective style, often initiated and implemented by
local administrators for extra revenues to fund the urgent tasks that confronted
them. Therefore, this transition inevitably led to the decentralization or region-
alization of the fiscal system. Unlike the situation before the mid-nineteenth
century when the imperial court was able to put most of the revenues under its
control, in the late nineteenth century most of the new sources of revenue were
in the hands of provincial leaders and not subject to the central government’s
effective surveillance and reallocation. What made this possible was the formal-
ization of the temporary measures that had been practiced during wartime as
a solution to the plight of the budget deficit, a development leading to the
growing fiscal autonomy of provincial authorities.
A hallmark of the Qing state’s highly centralized control of tax revenues
before the mid-nineteenth century was the remittance system. Under this
 Regionalized centralism 95
system, the remainder of every province’s tax revenues, after deducting the
statutory portion retained for local administrative expenses, had to be com-
pletely remitted to the Board of Revenue (hence named jingxiang or “Metro-
politan Remittance”) and, if necessary, to other provinces as instructed by the
Board (xiexiang or “Coordinate Remittance”) (Liu Zenghe 2014: 49–54; Ni
Yuping 2017: 15). The total amount of Metropolitan Remittance from indi-
vidual provinces was roughly 10 million taels each year in the late eighteenth
century, or a quarter of the state’s total annual revenue (Shi Zhihong 2009:
11). During the early years of the Taiping Rebellion, however, the provinces
affected by the revolt petitioned one after another for delaying or retaining
the Metropolitan Remittance for local military expenses and even detained
the remittance from neighboring provinces being transported to the capital
city or other needy provinces. So chaotic and paralyzed was the remittance
system that the imperial court, without sufficient funds to allocate, eventu-
ally decided in 1853 to allow provincial authorities to raise funds on their
own for local administrative and military expenses, as mentioned earlier.
Nevertheless, to ensure that the Board had the necessary minimal income to
sustain the central government and regular military forces, the court ordered
in 1856 individual provinces to remit a total of 4 million taels a year as
a “fixed” amount of Metropolitan Remittance, to be apportioned among the
provinces. This “fixed” quota was later increased to 5 million taels in 1860,
7 million in 1861, and 8 million in 1867, which would remain unchanged
until the end of the Qing (Deng Shaohui 1998: 54–60; Shi and Xu
2008: 136).
But the Metropolitan Remittance was not the only obligation of provincial
authorities to the central government in the late nineteenth century. From the
1860s through the 1880s, the Qing court imposed on individual provinces and
maritime customs additional annual quotas of mandatory remittance to the
Board of Revenue for specific purposes, including:

• enhancing the defense of the capital city (660,000 taels from 1863);
• increasing the budget for the Imperial Household Department (neiwufu,
600,000 taels from 1869);
• building coastal fortresses and naval forces (4 million taels from five mari-
time customs and the lijin of coastal provinces from 1875);
• sending diplomats abroad (more than 1 million taels from 1876);
• strengthening the garrison troops in Manchuria (2 million taels from
1880);
• relieving famines (120,000 taels from 1883);
• recovering the salaries of bureaucrats in Beijing to full levels (260,000 tales
from 1883);
• compensating bureaucrats in Beijing whose salaries had been reduced from
1851 (1 million taels from 1885);
• training the Banner troops in Beijing (1.33 million taels from 1885);
• constructing railroads (800,000 taels from 1889);
96 The transition to a sovereign state
• strengthening the defense in the vicinity of the capital city (2 million taels
from 1892) (Luo Yudong 1970: 196–207; Shi and Xu 2008: 137–138; Ni
Yuping 2017: 248–255).

Altogether, the mandatory recurring revenue contributions (the Metropolitan

Remittance and other obligatory remittances) shared by all provinces and maritime
customs to the central government increased to 21.77 million taels annually by
1892. The fiscal system of the Qing thus underwent a transition from the imperial
court’s centralized and complete control of government revenues at all levels before
1851 to relatively fixed revenue-sharing between the central and provincial author-
ities in the second half of the nineteenth century. What, then, does this transition
mean for the Qing state and its fiscal abilities?
An obvious result of this transition was the growing autonomy of provincial
leaders, who were able to control the remainder of the revenues generated
within their territories after fulfilling their obligations to central government.
Because of this, they had a strong incentive to increase the revenues of provin-
cial treasuries by whatever means available to them, primarily maximizing the
collection of lijin and surcharges on land taxes and domestic or foreign debts,
as mentioned earlier, for new projects they initiated and sponsored in their own
provinces. And they set up their own financial bureaus under various names to
collect and manage the non-conventional revenues; these agencies answered dir-
ectly to the provincial leaders. On the other hand, the Board of Revenue’s
power to supervise the fiscal activities at the provincial level deteriorated; the
Financial Commissioner (buzhengshi) to every province, who had conventionally
been an agent of the Board in charge of tax revenues in the province and acted
independently of the provincial governor and governor general, was gradually
turned into a subordinate of the latter.8
The financial autonomy of provincial authorities was also seen in their ability
to bargain with the central government over their sharing of various remittance
quotas imposed by the latter. The governors or governors general rarely remit-
ted the full amount of required quotas to the Board, and they would use what-
ever excuse to delay or reduce their remittances, despite the Board’s repeated
prompts. Overall, the provincial authorities fulfilled only 44.4% of their quotas
in 1897, 45.8% in 1898, and 60% in 1899 (Shi Zhihong 2009: 268–272; also
see Shi and Xu 2008: 140). In fact, the same situation existed before 1894. In
general, they showed greater commitment to fulfilling their duties directly
linked with the security of the imperial capital and Manchuria – the homeland
of the Manchus, because doing so helped prove their loyalty to the court – than
to projects that only worked to strengthen the position of certain provincial
leaders, especially the funds on coastal defense and naval forces, which were
often perceived as a personal enterprise of Governor General Li Hongzhang
from the mid-1870s to the mid-1890s, as shown in the following discussion on
maritime defense. At any rate, fiscal regionalization (i.e., the growth of the
financial independence of provincial authorities) was an unmistakable trend
during and after the Taiping Rebellion.
 Regionalized centralism 97
But the autonomy of late Qing provincial authorities should not be so exag-
gerated as to conclude that the central government had lost control of fiscal
resources at the provincial level. In fact, the Board of Revenue remained decisive
in reallocating between the central and provincial governments all of the rev-
enues reported to it; what the Board could not control was only the portion
hidden from the reports of provincial or lower-level governments. By and large,
the unreported revenues accounted for up to about 40% of the actual amount
of revenues generated by governments at different levels in the second half of
the nineteenth century (see Table 4.1). Among all the different kinds of taxes,
lijin had the biggest problem with underreporting; nevertheless, the Board of
Revenue was able to spend most of the officially reported lijin income on pro-
jects at the national level (guoyong), as discussed earlier. We will see in Chapter
6 that, by implementing a nationwide fiscal investigation, the central govern-
ment was able to put under its control most of the lijin income that had been
previously unreported, thus resulting in the skyrocketing of the lijin revenue
to 43 million taels in 1911, three times its level in the 1890s (ranging from 12
to 16 million taels annually); and the unreported incomes were accordingly
reduced to about 15% of the total revenues that were actually generated in
individual provinces (see Table 4.1).9

The rise of conditional loyalty

The devolution of power to Han bureaucrats

The regionalization of the fiscal system came with the restructuring of power
relations during and after the Taiping Rebellion in the 1850s and early 1860s.
Before the rebellion, what characterized the government system at the top level
and within the 18 interior provinces was not only a highly centralized bureau-
cratic system, much of which had been inherited from the Ming, but also the
Qing court’s firm control of provincial authorities through its appointment of
Manchu nobles and Banner elites to almost all of the positions of governors
general, who were in charge of the military and local defense, while allowing
only half of the positions of provincial governors to be filled by Han bureaucrats
to manage local civilian affairs, as typically seen during the reign of Emperor
Qianlong (1736–1795) (Kong Lingji et al. 1993: 327). The situation improved
during the Jiaqing (1796–1820) and Daoguang (1821–1850), when the Han
elites occupied half of the governor-general positions (Long Shengyun 1990:
298). In 1850, for instance, six out of the ten governors general were the Han
Chinese, and among the 15 provincial governors, only one was a Manchu, and
the rest were all Han bureaucrats (Wang Kaixi 2006). However, whenever
a crisis emerged, the emperor invariably appointed a Manchu he trusted as his
commissioner (jinglue dachen or canzan dachen and later qinchai dachen) to
supervise military campaigns, and the provincial administrators (governors and
governors general) were only to carry out the orders of the commissioner and
play an auxiliary role in military operations (cheng haoling, bei ceying), a fact that
98 The transition to a sovereign state
remained true during the Opium War (QSG: 3264). During normal times, the
governors and governors general were also subject to an intricate mechanism of
checks and balances in their relationship with the two key agents of the central
government at the provincial level, the Financial Commissioner (buzhengshi),
who supervised civil administration and taxation in his province, and the Judicial
Commissioner (anchashi), who oversaw litigation and adjudication of the
province.
The outbreak of the Taiping Rebellion reversed this tradition. During the
first two years of the rebellion, the Qing court failed to curb the rapidly
spreading rebellion with its regular army, the Green Standards, which had
long lost its fighting abilities due to the lack of training, poor organization,
and corruption among military officials. As a result, the rebels marched swiftly
up to central China to capture Wuchang, the capital city of Hubei, in Decem-
ber 1852, thus they were in a position to attack and occupy the key cities of
the lower Yangzi region, the most prosperous area. It was at that point that
the Qing court ordered government officials in retirement to recruit adult
males from their communities to form a military corps (tuanlian) as an auxil-
iary force to guard their communities and fight the rebels, a practice that had
been tried in the 1790s in suppressing the White Lotus Rebellion. Zeng
Guofan of Hunan province was one of the 43 “ministers of the limitary
corps” (tuanlian dachen) in ten provinces appointed by Emperor Xianfeng
in early 1853. Zeng soon proved his military talents by leading a force of
more than ten thousand men from his province to retake Wuchang in
July 1854. After fighting in different parts of the middle and lower Yangzi
region over the next few years, he was eventually appointed as the Governor
General of Liangjiang (Jiangsu and Anhui provinces) and Imperial Commis-
sioner supervising military affairs in Jiangnan or the lower Yangzi region in
1859; in late 1861, he was further authorized to supervise all civilian and
military officials in the provinces of Jiangsu, Jiangxi, Anhui, and Zhejiang.
At his recommendation, a number of his subordinates from Hunan province
were appointed as governors of the provinces in the middle and lower
Yangzi region, hence the rise of the Hunan faction that would influence
Qing politics at the provincial level for decades in the second half of the
nineteenth century. Wenqing (1796–1856), Grand Secretary and Grand
Councilor of State, explained well why the Qing court had to rely on the
Han elites instead of the Manchus and put them to the key positions of
civilian and military leadership:

the Han people have to be relied on for handling tasks of the first magni-
tude under Heaven. These men are all from the fields of the countryside,
familiar with the weal and woe of the people and knowing well what is true
and what is false. How can we [Manchu nobles] compare to them, having
never stepped out of the capital city and having no idea about grand
strategies?
(Xue Fucheng 1987: 250)
 Regionalized centralism 99
It is estimated that from 1851 to 1912, Manchus held only 34.6% and 22.2%
of all governor and governor-general positions, respectively, during this period,
which contrasts with 57% and 48.4% during the entire Qing dynasty (Rhoads
2000: 47–48).
What bolstered the persistent dominance of the Hunan faction of provincial
leaders in late Qing politics, however, was first of all their tight control of the
personalized militia as the main force in place of the regular army to suppress
the Taiping rebels. The Hunan corps differed fundamentally from the regular
army in that strong personal ties bound together different ranks of command-
ers and soldiers. When creating a corps, the commander at each level was
responsible for recruiting his subordinate officers at the next level all the way
down to the head of the basic unit of ten soldiers, and the head was to see
that he recruited his solders from his home community, that he knew well the
soldiers’ home addresses, parents, personalities, and abilities, and that the sol-
diers signed a pledge to ensure their obedience to the disciplinary rules of the
corps (Liu Wei 2003: 119; Kuhn 1970: 122–148). This method of organiza-
tion precluded the soldiers’ defection and delinquency on the battleground
that were common among the soldiers of the Green Standards. It also turned
the corps into essentially a private army of the provincial governors or gover-
nors general who established it.
The corps was in essence the private army of the provincial leaders also
because it was self-sustained and completely independent of central govern-
ment’s support. The provincial leaders had to support their respective corps,
using whatever resources locally available to them (jiudi chou xiang), such as
collecting the lijin, selling government titles for contributions, detaining the
remittance of tax monies and maritime customs, levying surcharges on land
taxes, and so on, as discussed earlier in this chapter. To ensure that they gener-
ated enough revenue and managed it efficiently, each province established its
own fiscal agency (liangtai or zhong liangtai) under the direct control of the
governor or governor general, to disburse funds and reimburse expenses without
being supervised by the Financial Commissioner of the province.
The provincial leaders not only controlled the military and finances of their
provinces and took them largely out of the central state’s reach, they further
controlled the personnel of local governments by appointing their own men to
the key positions of the provinces. It had long been a tradition in the history of
the Qing for the provincial leaders to recommend candidates to the imperial
court for appointments of administrators at the circuit, prefecture, and county
levels, but their recommendations were limited (three candidates a year for
a governor general and two a year for a governor) (Liu Ziyang 1988: 35). After
being appointed as the Governor General of Liangjiang in charge of military
affairs of four lower Yangzi provinces, Zeng Guofan went far beyond the limit
and made frequent recommendations for a wide array of positions ranging from
provincial governors to financial and judicial commissioners as well as officials at
lower levels. To satisfy the needs of providing logistic support to the corps,
postwar rehabilitation, and other administrative tasks, a governor or governor
100 The transition to a sovereign state
general was free to establish as many “bureaus” (ju), “stations” (tai), or
“offices” (suo) as he saw fit, each in charge of a specific task. All of them were
outside the regular government system and responsible only to the provincial
leaders.
In a word, in the course of suppressing the Taiping Rebellion and post-war
rehabilitation, provincial governors and governors general, mostly of Han
origin, emerged as strong regional leaders by arrogating to themselves the
military, fiscal, and personnel power to a level unseen before the mid-
nineteenth century and by weaving themselves into factions through personal
ties to rival the Manchu elites, who had dominated Qing politics before the
rebellion. The end result of the rebellion, then, was the unmistakable trend of
power devolution from the imperial court to provincial leaders, in which Han
bureaucrats came to dominate late Qing politics at the expense of Manchu and
Banner elites.

Loyalty redefined
The Qing court, to be sure, was alert to the rapidly growing power of the
Han elites. Zeng Guofan’s retaking of Wuchang in 1854, for instance, made
Emperor Xianfeng at once excited and worried, after being reminded by
a grand secretary of the potential danger from an “ordinary person” (pifu)
such as Zeng who had been able to “rally more than ten thousand followers
with a single shout,” a fact that signaled “a bad omen for the state” (kong fei
guojia zhi fu) (cf. Wang Kaixi 2006: 170). Emperor Xianfeng, therefore, gave
up his original decision to award Zeng with the position of acting governor
of Hubei province, and would remain vigilant regarding Zeng for the next six
years. With only the nominal title of “Assistant Minister of the Board of
War,” Zeng was repeatedly frustrated when he failed to gain support for his
troops from local military and civilian officials while combatting rebels in
neighboring provinces. While using the informal corps to fight the rebels in
the middle and lower Yangzi provinces, the Qing court concentrated its regu-
lar Green Standard forces on two major camps stationed on the two sides of
the Yangzi river on the outskirts of Nanjing, the capital of the Taiping rebels,
to besiege the city, but it would not attack it until the informal corps had
annihilated most of the rebel forces in the surrounding provinces. The pri-
mary purpose of the two camps led by the Manchu commissioners, in other
words, was to prevent the informal corps led by the Han elites from claiming
their decisive role in defeating the enemy. Not surprisingly, when the two
camps were completely destroyed by the Taiping rebels in 1860, Zeng
Guofan and his subordinates reacted with unconstrained joy and relaxation in
their private communications (Fan Wenlan 1949: 144–145). Afterwards, the
Qing court had to rely entirely on the informal corps under Zeng and other
Han elites. He was eventually appointed as a governor general in charge of
military affairs in four provinces that enabled him to command all military
and civilian officials in this region.
 Regionalized centralism 101
But tension between the court and the Han elites continued in the following
years and became exceptionally keen at certain moments.10 The alliance between
the Manchu and Han elites was essentially conditional; it was based on the Con-
fucian notion of a proper relationship between the ruler and the minister: “the
ruler employs his ministers by the rule of propriety, and the minister serves the
ruler with loyalty” (jun shi chen yi li, chen shi jun yi zhong) (ibid.). In other
words, the relationship between the two sides was reciprocal; the ruler’s proper
treatment of a minister was the precondition of the latter’s loyalty to him. At
the critical moments testing their relationship, a crisis could erupt into uncon-
trollable chaos if either of the two sides refused to concede.
Fortunately, for the Qing court, Zeng turned out to be a savvy statesman,
imbued deeply in the tradition of Confucian morality and statecraft and able
to navigate the recurrent crises with his vigilance and forbearance. To some
degree, his loyalty to the Qing, conditional as it was, can be seen as a reward
to the latter for its persistent efforts in the preceding two centuries of embra-
cing neo-Confucianism as its orthodox ideology and of narrowing the gap
between the Manchus and the Han in filling government positions. In return,
the vast majority of the Han elites had identified with the Qing in the eight-
eenth and early nineteenth centuries, accepting it as a legitimate regime in
China, despite its alien origins, and defined their relationship with the court
on the basis of the Confucian tenet of morality. Added to Zeng’s loyalty to
the Qing was also the fact that the two sides faced the same enemy, the Taip-
ing rebels. For Zeng and many other Han elites, to combat the rebels was not
only to save the Qing, but more importantly to defend the “Chinese-ness” of
his country that had been threatened with the rebels’ embrace of an alien reli-
gion (pseudo-Christianity), which was in many aspects subversive to the Con-
fucian values and Chinese cultural traditions, as he articulated eloquently in his
famous “Proclamation of an Expedition against the Bandits from Guangdong”
(Tao Yue fei wen) (ZGF, shiwen: 232–233).
Therefore, what made the Han elites of the 1850s and 1860s unique and
a powerful force in late Qing politics was primarily a leverage that none of their
predecessors had ever possessed, that is, their control of the military, finance,
and personnel of their provinces and, more importantly, their formation of
a tightly knit faction to defend themselves as a group and bargain with the Qing
court.11 The relationship between the Qing court and Han bureaucrats thus
underwent a substantial change during and after the Taiping Rebellion. Before
the 1850s, for more than two centuries since its inception in Beijing, the Qing
state experienced power concentration to a level unseen in the preceding Chin-
ese dynasties.12 The Manchu rulers defined their relationship with the Han
people – including high-ranking bureaucrats of the Han origin – as essentially
the one between the conquers and the conquered, demanding unconditional
submission and loyalty from the latter (Guo Chengkang 2000), thus departing
from the Confucian doctrines of statecraft that assumed a reciprocal relationship
between the emperor and his ministers and saw the loyalty of the latter as con-
tingent on the proper treatment of them by the former. This one-way
102 The transition to a sovereign state
relationship, however, came to an end in the second half of the nineteenth cen-
tury. The indispensable role of the Han bureaucrats in suppressing the rebels
and handling the foreign crisis, the devolution of military, fiscal, and administra-
tive power to them, and the subsequent growth of their autonomy in relations
to the imperial court made their loyalty to the latter increasingly conditional.

The “eighteen small countries”

What needs to be clarified here is whether or not the trend of regionalization
during the Taiping Rebellion was reversed in the post-Taiping decades when
the Qing court attempted to recover the military and fiscal powers that it had
lost in the 1850s and early 1860s. The strongest evidence that supports this
assumption of recentralization was the dissolution of most of the Hunan army
right after its capture of Nanjing in 1864. Indeed, in a matter of a few months,
a division of the Hunan army, roughly 50,000 men, that sacked Nanjing under
the leadership of Zeng Guoquan (Zeng Guofan’s younger brother) were com-
pletely disbanded, and Guoquan himself renounced all of his positions to return
to his home community for a “sick leave.” A year later, most of the Hunan
army, which once had more than 200,000 men during its height, was dispersed,
except for a few battalions of about 10,000 that survived for local security and
a traditional naval force to patrol the Yangzi River.
Disbanding the Hunan Army, however, was only one side of the story. The
other side was the survival and expansion of the Huai Army that had originated
from the former under the leadership of Zeng’s disciple, Li Hongzhang. Li was
appointed as the Governor of Jiangsu province in 1862 for his key role in
defending Shanghai against the Taiping rebels with his newly created army,
known later as the Huai Army, mostly recruited from his home province. After
cutting down a few thousand aged, feeble, or wounded soldiers, the Huai Army
kept more than 50,000 men, which turned out to be indispensable for the Qing
government to suppress the Nian Rebellion (1853–1868), especially after the
Nian rebels decisively defeated the Qing’s best regular force and killed its com-
mander, the Mongol General Senggelinqin in 1865. Li Hongzhang’s success in
completely and swiftly annihilating the Nian rebels as well as his proven abilities
in handling the thorny foreign affairs led to his appointment as the Governor
General of Zhili and the Commissioner for Trade in North China for fully 25
years (1870–1895), which made him the most influential figure among the Han
bureaucrats in the late nineteenth century. His Huai Army also expanded to 95
battalions in 1875 and as many as 146 battalions over the next two decades and
which were stationed in more than ten provinces, which in effect replaced the
antiquated and incompetent Eight Banners and Green Standards to become the
best equipped and the most important defense force of the Qing, before it was
finally replaced by Yuan Shikai’s New Army during the decade following the
Boxer Rebellion (Dong Conglin 1994: 28; Liu Wei 2003: 277–278). Li’s
unparalleled influence allowed him to appoint or recommend his trusted subor-
dinates to the key positions of the Huai Army, the newly established Northern
 Regionalized centralism 103
Fleet, and provincial authorities; they thus formed a faction on the basis of
family and kinship ties, fellow townsmanship, and patron–client relations.
The Qing court not only failed to reestablish its centralized control of the
military, but it also found it difficult to curb the growing autonomy of provin-
cial leaders in civil administration. Throughout the Tongzhi and Guangxu
reigns, the court took measures repeatedly to limit the power of provincial gov-
ernors or governors general. When making appointments at the provincial or
lower levels, for instance, it only permitted the governors to make recommenda-
tions, while keeping the power of decision-making on appointments to itself,
a policy reasserted at the beginning of the Tongzhi reign. To stem provincial
leaders’ nepotism in making recommendations, it announced in 1894 a new
policy of rejecting all of the candidates in the same package of recommendations
if any of them was found to be unqualified. To enhance its supervision of pro-
vincial governments, the court encouraged the financial commissioners and judi-
cial commissioners of each province to impeach their governors and governors
general for any kind of wrongdoings they had uncovered, realizing that these
commissioners had not submitted any substantial memorials to the court for sev-
eral decades since the 1860s other than routinely reporting their commence-
ment of, or retirement from, their positions (Liu Wei 2003: 362–363). These
measures, however, had little effect on local governors or governors general.
The central government had no control of the large number of informal posi-
tions (titled zongban, huiban, tidiao, weiyuan, siyuan, etc.) of the irregular agen-
cies (the various bureaus or offices) created and appointed by the provincial
authorities, which multiplied in the 1870s through the 1890s to accommodate
the new needs of modernizing projects.
The Qing court’s biggest failure in recentralizing state power, however, lay in
its inability to control local finance in the late nineteenth century. As noted earl-
ier, provincial administrators only reported to the court 30–40% of their revenues
from lijin, and the actual revenue of the Qing government at all levels in 1894,
for example, could be as high as 146 million taels of silver, rather than the official
figure of merely 81 million taels (Shi and Xu 2008: 133, 275, 279, 289). Equally
difficult for the imperial court to know was how much the local governments
actually spent on new projects that had never been part of their conventional
expenditures, primarily those associated with the Self-Strengthening movement.
When the provincial authorities fell short of funds, they were free to make loans
from foreign banks or domestic debtors without the court’s approval. Because of
the underreported revenue and unknown expenses, the auditing system that the
Qing court had used to supervise the financial activities of the provincial govern-
ment before the Taiping Rebellion was largely ineffective and perfunctory, despite
its recovery from wartime paralysis. The result was the fiscal independence and
self-sufficiency of the provincial authorities in the late Qing period, which con-
trasted sharply with the central government’s effective control of local finance
before the nineteenth century.
Given the growing autonomy of provincial leaders in managing the military,
personnel, and finance of local governments in the late nineteenth century, it is
104 The transition to a sovereign state
no wonder that contemporaries tended to describe the 18 interior provinces as
“eighteen countries” (shiba guo) (LQC, 2: 649). The Qing court’s dependence
on the governors and governors general for defense and revenues also accounted
for their rise as a decisive force in shaping the central government’s decision-
making process, which distinguished them from their predecessors before the
nineteenth century who had usually played a passive and subservient role in rela-
tion to the imperial court. Li Hongzhang, for example, was so powerful during
his tenure as the Governor General of Zhili and the Commissioner for Trade in
North China that “the court has to consult [him] on whatever issues of import-
ance before a decision was made, and it invariably approves whatever the North-
ern Commissioner has requested,” as a contemporary observed (Wang Ermin
1987: 393). In addition to Li, the most prominent provincial leaders also
include: Zuo Zongtang (1812–1885), Li Hongzhang’s long-time rival and the
Governor General of the southeastern, northwestern, and lower Yangzi prov-
inces in sequence from the 1860s to the early 1880s; Liu Kunyi (1830–1902),
serving repeatedly as the Governor General of the lower Yangzi and southern
coastal provinces from the 1870s to the 1890s; and later Zhang Zhidong
(1837–1909), the Governor General of three different regions in the 1880s
through the 1900s but mostly of Hubei and Hunan, the two middle Yangzi
provinces. Each of them played a key role in making policy suggestions or dir-
ectly shaping the Qing court’s decisions on domestic and foreign affairs.

“Self-strengthening” and incipient nationalism

The relative autonomy of provincial authorities in fiscal constitution and civil
administration made it possible for some of the governors and governors general
to launch a series of projects during the Self-Strengthening movement. Begin-
ning with the training of the self-recruited militia in Western methods and
equipping it with modern weaponry in the early 1860s in the course of sup-
pressing the Taiping rebels, the movement focused on establishing three major
arsenals in Shanghai, Nanjing, and Tianjin, a modern shipyard in Fuzhou, as
well as a number of small-scale arsenals in other provinces. The modernization
of defense culminated in the establishment of three fleets to patrol the waters
along the northern, eastern, and southeastern coasts, respectively, with the
Northern Fleet fully established in 1888 as the largest and most advanced in
China and the Far East as well. Affiliated with these military establishments were
the institutions to translate books in Western languages into Chinese and the
schools to train students in foreign languages and modern sciences. In addition,
the movement also expanded in the 1870s to construct a series of civilian pro-
jects ranging from steam ship navigation and railroad transportation to mining,
iron smelting, telegraph communication, and textile manufacturing, as well as
a program to send students to the United States for education (Kuo and Liu
1978; Elman 2005: 355–395; Halsey 2015). Except for the fleets that were
funded mostly by the central government, all other projects were initiated and
fully funded by the governors or governors general using their own financial
 Regionalized centralism 105
resources. Altogether, the total spending on establishing and running the
arsenals and shipyard during the period from 1866 to 1895 ranged from 50 to
60 million taels. The methods of funding for the civilian enterprises varied;
some of them were fully financed by the provincial government, and others
were either partially funded by the government and partially by private investors
or fully invested by merchants but supervised by the government. The total
amount of funding by provincial authorities on the civilian projects during the
same period could be as high as 15 million taels (Zhou Yumin 2000: 303–304).
The immediate reason leading to the provincial leaders’ establishment of the
arsenals was, of course, to equip their own armies with the advanced weapons
and thereby to elevate their standing within the bureaucratic system. But this
selfish motivation alone cannot explain why the Self-Strengthening movement
persisted for more than three decades and developed into a comprehensive pro-
gram in modernizing not just the military but also manufacturing, education, and
the infrastructure. When justifying their initiatives for a new project, Li Hon-
gzhang and other leaders invariably linked it with China’s “self-strengthening” to
survive the challenges from European powers and rival Japan, a country that also
made efforts to modernize its military and become a new threat to China. It is
worth stressing that Li Hongzhang realized this potential danger from Japan as
early as 1864, four years before the Meiji Restoration. As he insightfully stated in
a memorial to the imperial court:

today’s Japan is where the dwarf pirates of the Ming came from, being dis-
tant from the West but close to us. If we have the means to remain inde-
pendent, then it will depend on us and thereby get to know the strength
and weakness of the West; if we have no means to strengthen ourselves,
then it will imitate the West and join the West to take advantage of us.

He then went on to argue that:

if China wishes to strengthen itself, nothing is better than learning from

foreign countries their advanced equipment; to learn their advanced equip-
ment, nothing is better than studying their technologies to produce the
advanced equipment, that is, to study their methods but not to have to rely
on foreigners.
(LHZ, 29: 313)

It is also worth noting that whenever Li referred to China in relation to for-

eign powers in his memorials, he invariably used the word “Zhongguo” (the
Central State), an informal term that had long been used by the Han Chinese
to describe their land, rather than “Da Qing” (the Great Qing), huangchao (the
imperial dynasty), or tianchao (the celestial dynasty), terms that had been widely
used in official discourse and documents by the Qing rulers. This frequent use of
Zhongguo in lieu of Da Qing reflected a subtle transition among the Han bur-
eaucrats in their consciousness of identifying themselves and their modernizing
106 The transition to a sovereign state
efforts with the benefits of the land of the Chinese people rather than with the
interests of the Qing dynasty. This consciousness foreshadowed the rise of incipi-
ent nationalism among the Han elites in response to the recurrent foreign threats
in the late nineteenth century, no matter how ambiguously defined and how
entangled it was with their commitment to the imperial state. Unlike the national-
ism of twentieth-century China that pervaded the society, incipient nationalism
was largely limited to the Han Chinese ruling elites who had contacts with, or
knowledge of, the West and Japan before the Sino-Japanese War of 1894. And its
roots can be traced to the pragmatism in Confucian statecraft thinking, which
Kwang-ching Liu (1970) terms “Confucian patriotism.”
But the provincial leaders’ incipient nationalism was curtailed by their priori-
tization of local and personal interests over national interests. With a large and
ever-increasing amount of informal and unreported revenues at their disposal,
the provincial authorities decided where and how much they spent the monies
primarily according to their own stance and interests. They were most enthusias-
tic in building and strengthening their own military and civilian enterprises with
the funds available to them. They were also willing to release their funds as
Coordinate Remittance for the projects of governors or governors general who
were their friends or relatives or members of the same faction, and for military
operations that they proposed or supported. However, they were reluctant to
remit the mandatory quotas imposed by the court to a destination that did not
relate to themselves, and they would use whatever excuses they had to delay or
reduce their remittance. The success or failure of the Self-Strengthening move-
ment or any military operation during the three decades of the movement thus
depended in large measure on the attitudes of the provincial leaders and the
amount of funds they contributed, a fact that was most evident in the processes
of frontier defense and maritime defense, the subjects of the next chapter.

Regionalized centralism
To conclude, most of the changes in the relationship between the central and
local authorities of the Qing state in the second half of the nineteenth century
can be subsumed under the category of “regionalized centralism,” a development
that had positive as well as negative impacts on China’s transition to a modern
sovereign state. First of all, the diversification and regionalization of government
revenues under the initiative of provincial governors and governors general
resulted in a transformation of the Qing state’s fiscal constitution, in which the
ever-growing demand for revenues drove the expansion of the supply, and in
which non-agricultural revenues (indirect taxes, borrowing, and other means of
financing) replaced the land tax to be the most important sources of government
income, thus showing a striking level of expandability. This transformation, taking
place from the 1850s to the 1870s, accounted in large measure for the Qing
state’s success in pacifying internal revolts, modernizing its military, and defending
its frontiers against foreign aggression, hence a “restoration” (zhongxing) from
the mid-1860s to the mid-1890s. But the transformation was achieved at the
 Regionalized centralism 107
expense of the imperial court’s centralized control of regular and irregular rev-
enues on the one hand and the rise of provincial elites on the other hand, who
played a key role in the restoration by putting regional financial, military, and
administrative resources at their disposal. Hence the subtle yet critical changes in
the relationship between Han bureaucrats and the imperial court, which account
for both the striking resilience of the late Qing state and its eventual demise. By
and large, before the 1850s, this relationship was characterized by the Han bur-
eaucrats’ unconditional, one-way loyalty to the Qing court, which can be attrib-
uted to two basic factors. One is the Qing court’s inheritance from the preceding
Ming of neo-Confucianism as a state ideology and its time-honored adherence to
Confucian tenets in governing the interior provinces, which won the Han elites’
endorsement of its legitimacy. The other is the Qing court’s inheritance of the
patrimonial tradition from its own past as a nomadic state before 1644, which
assumed a master–slave type of relationship between the ruler and his subjects
and demanded absolute submission from the latter, thus departing from the Con-
fucian principle of reciprocity between the monarch and his subjects, by which
the latter were committed to the well-being of “All under Heaven,” rather than
the ruler alone, and whose loyalty was premised on the former’s respect and trust
of them. After the mid-nineteenth century, however, the rise of Han bureaucrats
at the provincial level and their arrogation to themselves of regional fiscal, admin-
istrative, and military power to an unprecedented level led to a return to their
reciprocal relationship with the imperial court: they served the latter with loyalty
as long as they were trusted by the court and as long as they were bound by the
Confucian way of thinking; much of the success of the “restoration” in the 1850s
and 1860s thus can be attributed to these realities. However, they would turn
their backs on the imperial court if they were distrusted and excluded by the
latter from the power center, which was exactly what happened to them during
the last years of the Qing dynasty.
It is worth emphasizing, however, that for all the enhanced control of fiscal,
military, and administrative resources by provincial leaders within the areas
under their jurisdiction and their shared assumption of conditional loyalty to the
imperial court, the central government never lost its effective control over the
appointment, promotion, demotion, or removal of provincial governors or gov-
ernors general, and it firmly maintained the ultimate decision-making power
over the redistribution of all officially reported revenues between the central and
provincial governments. The centralized bureaucracy that the Qing state
inherited from the Ming in its early history and developed in the eighteenth
century remained largely in shape and functioning by the end of the nineteenth
century. The regionalization of financial and military power in individual prov-
inces after the Taiping rebellion did not necessarily mean the subvention or mal-
functioning of the Qing state’s centralized bureaucracy; it took place within the
centralized bureaucratic system. By and large, the regionalization process was
a win–win situation for both the Qing court and local authorities. Its positive
effects, as seen in the thriving of regional modernizing enterprises and the over-
all improvement of the Qing state’s capabilities, outweighed its negative
108 The transition to a sovereign state
consequences on the state’s coordination and control of local resources during
the reigns of the Tongzhi and Guangxu emperors (1862–1908). The secret of
the late Qing state’s exceptional resilience lay precisely in the regionalization of
government power on the premise of conditional allegiance.
But the late Qing state was also fragile, because its legitimacy rested ultimately
upon the identification with it by local political elites, especially the provincial
leaders of the Han origin. As shown in Chapter 6, once they were distrusted by,
or ousted from, the power center, and the premise of conditional allegiance no
longer existed, the Han bureaucrats and social elites would openly confront
the Qing court, counting on the strength that they had built in the preceding
decades; and the Qing state would become extremely vulnerable when their
challenge to the central authorities from within the government system was
joined by the assaults from the anti-Qing revolutionaries and anti-Manchu rad-
icals outside the system.

Notes
1 In her classic study of the “Tongzhi Restoration” in the 1860s and early 1870s, Mary
Wright, for instance, contends that the Qing failed to modernize its defense, education,
and foreign relations, because of the conservatism among gentry scholars and bureau-
crats, which had its roots in the Confucian values and practices that bolstered the existing
social and political orders (Wright 1957). In a similar fashion, Albert Feuerwerker
(1958) examined China’s “early industrialization” in the late Qing period, through the
examples of a steamship navigation company and some other enterprises. He found the
roots of their failures in the institutional and ideological obstacles inherent in Chinese
society, as evidenced by the notorious mold of “official supervision and merchant man-
agement” (guandu shangban) in running the enterprises, which militated against imper-
sonal, rationalized, and specialized organizations to make the modern projects a success.
Likewise, John Rawlinson blamed traditional institutions based on classical ideology for
the Qing’s inability to develop a modern navy (Rawlinson 1967).
2 Later studies questioned the failure narrative. Instead of interpreting the Self-
Strengthening program as one ending in total disaster, they found “marked success”
of some of the modern projects in expanding their businesses, competing with for-
eign firms, and transferring modern technology from the West to China (Lai 1994;
K. Liu 1994). And instead of seeing Confucianism as an obstacle to modernization,
these studies reinterpret it as a conducive factor. Kwang-Ching Liu, for instance,
attributed the success of Zeng Guofan (1811–1872) in the “Restoration” to his Con-
fucian statecraft thinking; Liu also traced the leading role of Li Hongzhang
(1823–1901) in the Self-Strengthening movement to what he called “Confucian pat-
riotism” or commitment to China’s security that had its roots in “a strain of realism
in Confucianism” (K. Liu 1970: 43). What constrained the modernizing projects to
develop and compete successfully was not the ideological and cultural values inherent
to Chinese society, but primarily the lack of sufficient funding from the central gov-
ernment (Pong 1994).
3 Wenkai He (2013) finds the emergence of institutional elements that enabled the late
Qing state to make long-term borrowing, which would have allowed the Qing to
build a modern fiscal state, had there been the necessary economic conditions and
a credit crisis to force its transition in this direction.
4 As shown later in 1843, while the Board of Revenue had an official figure of
12.18 million taels as cash surplus, the actual amount of the surplus was only
 Regionalized centralism 109
2.92 million as a thorough investigation revealed (Zhou Yumin 2000: 69; Shi Zhi-
hong 2009: 110); in other words, there was a difference of 9.25 million taels that
was missing. As it turned out, the actual amount of the cash reserves at the Board of
Revenue had never been verified for more than 60 years since Heshen (1750–1799)
headed the board in 1780; therefore, at the order of the Daoguang emperor, the
families of all the board officials since 1780 were held responsible for collectively
making up the deficit (Zhou Yumin 2000: 110).
5 The size of arable land per capita was 7.87 mu in 1736, 4.98 mu in 1766, 3.35 mu
in 1790, 2.89 mu in 1812, 2.78 mu in 1840, and 2.78 in 1851. China’s population
was 125 million, 208.1 million, 301.49 million, 363.7 million, 412.81 million, and
450 million in 1736, 1766, 1790, 1812, 1840, and 1851 respectively (Jiang Tao
1990). The size of arable land was 984 million mu in 1736, 1,147 million mu in
1840, and 1,254 million mu in 1851 (Shi Zhihong 1989, 2011).
6 The surcharges were collected in different ways in different provinces. In Sichuan,
a province known for its lightest land tax in history, beginning in 1854, all land-
owners had to officially pay one additional tael (known as “land tax subsidy” or an
liang jintie) to each tael of their land tax quotas; in addition, they had to pay
a “contribution” (juanshu) as high as 2 to 4 taels for each tael of land tax quota.
Their total land tax burden thus was increased by three to five times. In other prov-
inces, the surcharge was usually based on the size of landholdings; in Guizhou, for
instance, landowners paid 10 to 20% of their harvest as a surcharge (ligu), whereas
in reality, it could be 40 or 50% of their harvest. In Jiangsu and Anhui, the sur-
charge (known as mujuan) was collected at the rate of 20 to 80 wen of copper
cashes per mu of land, which could be as high as 400 wen per mu in some localities.
In Jiangsu and Zhejiang where the taxpayers paid grain taxes (caoliang), each shi of
the tax quota was often discounted again and again, so that taxpayers had to actu-
ally pay two or more shi of grain to fulfill each shi of the tax quota (Peng Zeyi
1983: 160–166).
7 The domestic debtors were invariably wealthy families or “gentry merchants” (shen-
shang) who made “both contributions and loans” (juan jie jian xing) to the local
government. The amount of a loan from these people varied from a few thousand to
hundreds of thousands of taels, depending on the needs of local authorities and the
affordability of the loaners. Beginning with a loan borrowed by the intendant of the
So-Song-Tai circuit in 1853 from a foreign firm in Shanghai, foreign debt became
a more important and regular source of extra revenue than domestic loans for local
authorities in the 1860s and afterwards to satisfy their military and civilian needs,
mostly to be repaid from the income of local maritime customs. During the few years
from 1861 to 1865, the administrators in Jiangsu, Fujian, and Guangdong, for
instance, made 12 loans from English and American companies, which totaled
1,878,620 taels, with an interest rate at 8–15% a year. In another instance, from
1867 to 1868, Shaan-Gan Governor General and Commissioner Zuo Zongtang bor-
rowed a total of 2.2 million taels from British firms in Shanghai for his campaigns
against Nian and Hui rebels in the northwestern provinces (Deng Shaohui 1998: 57;
Peng Zeyi 1983: 152–153). Obviously, most of the loans were short-term debts; in
comparison, long-term loans worked better to enhance the state’s fiscal ability and to
contribute to its transition to a modern fiscal state (He 2013).
8 To put the provincial fiscal activities under its control, the Qing court attempted to
audit in 1864 (the end of the Taiping Rebellion) the military spending by provincial
leaders during the rebellion but soon gave up because of its impracticability and pos-
sible recalcitrance by the governors; later the court did revive the traditional practice
of annual auditing (zouxiao) over the collection and management of regular taxes in
individual provinces, but it became more of a perfunctory formality than an effective
tool to check provincial finance, because the official figure of revenues reported by
110 The transition to a sovereign state
provincial authorities rarely reflected the actual amount of income they had generated
through non-conventional venues.
9 Susan Mann’s analysis of the reallocation of lijin revenues in Guangdong and Jiangsu
provinces adds evidence to the overall improvement in the central government’s
ability to control local revenues, as seen in the decline of the portion retained for
provincial usages to less than 50% of the total lijin revenue in Guangdong and 15
to 20% in Jiangsu in the early twentieth century (S. Mann 1987: 105–110). Chal-
lenging the conventional wisdom that equates the history of late Qing China with
“decentralization, disintegration, or weak central control,” her findings underscore
improvements in the state’s governing capabilities, especially its penetration into
local political and economic activities for augmentation of government revenues; its
dependence on tax farmers in collecting commercial taxes and levies, which has
been seen as a sign of the state’s deteriorating abilities, is interpreted as “useful
compromises” conducive to the building of a modern Chinese state (ibid.: 6–7).
10 One such moment, for example, was when the corps led by Zeng Guoquan (Zeng
Guofan’s younger brother) captured Nanjing and thus decisively defeated the Taiping
rebels in July 1864. To prevent the Zengs from claiming too much for their unparal-
leled feats, the Qing court threatened to investigate the whereabouts of the tons of
treasures hoarded by the Taiping leaders in Nanjing that had been plundered by
Hunan soldiers as well as the whereabouts of the missing Young Heavenly King of
the Taiping regime. Zeng conceded by promising to disband the militia, cut down
military spending, and send his younger brother back to his home province; as
a result, the court stopped the investigation. In another instance, Prince Gong was
repeatedly impeached in 1865, right after the suppression of the Taiping Rebellion,
for several offenses, one of which was his excessive reliance on the Han bureaucrats
such as Zeng. Empress Dowager Cixi, resentful of the prince for long, deprived him
of all positions. Zeng responded with “bitter disappointment and fear to the
extreme” (hanxin zhuili zhi zhi). When discussing a proper reaction to this move in
a secret conversation with one of his trusted subordinates on a small boat, the two
“sighed again and again” (xixu jiu zhi), as Zeng described in his diary (cf. Zhu
Dong’an 2007: 37). Fortunately, Prince Gong was soon rehabilitated, and the crisis
for Zeng was over. In each of the many critical moments like these, the relationship
between the Qing court and Zeng verged on a breakup.
11 At the core of the Hunan faction was of course Zeng Guofan until his death in
1872; so influential was Zeng during the peak of his career that “seven or eight prov-
inces were under his control, and state affairs ranging from government appointments
to taxation and military logistics were all subject to his dictation; it was as if the
entire country listened to one single person” (Rong Hong 1985: 107). Surrounding
Zeng were his numerous subordinates who were soon elevated to the key positions
of local governments. By 1872, 11 of them had been appointed to the positions of
governors or governors general (five served as governors general and nine as provin-
cial governors in 1863). The most prominent among Zeng’s disciples was of course
Li Hongzhang, who later started his own Anhui faction, named after his home prov-
ince. Li and his elder brother served simultaneously as governor general for four
years, and four of his key subordinates served as provincial governors during the
period from 1865 to 1877 (Long Shengyun 1990: 482). In sharp contrast, during
the same period, apart from only one or at most two Manchus serving as governors
general, and for two years only, there was no Manchu governor general at all. Also,
there was only one Manchu governor each year during most of the period; there was
only one year when two Manchu governors existed, and another year when no
Manchu governor existed at all (ibid.).
12 After conquering China, the Manchu rulers not only inherited from the Ming the
whole set of government apparatus that had been already highly centralized, they
 Regionalized centralism 111
further rebuilt it by combining the Chinese system with the practices originating
from their earlier history as a tribal state in which the cooperation in military oper-
ations and hence the rise of a mechanism of power-sharing among the chiefs of
nomadic groups under the emerging confederacy of tribes was coupled with the uni-
fying khan’s absolute power over those he subjugated and enslaved (Crossley 1992).
It was through their incessant struggles against the institution of collective deliber-
ation at the top level, known in the early Qing years as “the council of princely min-
isters on state affairs” (yizheng wangdachen huiyi), that the Qing emperors of the
eighteenth century restructured the central organ of decision-making to concentrate
power in their own hands, while reducing, if not completely eliminating, the roles of
the counseling princes.
5 Between the frontier and the coast
Geopolitical strategy reoriented

The geopolitical crisis that the Qing encountered in the nineteenth century
was twofold, involving a traditional challenge from Inner Asian forces to its
northwestern frontier and a new threat from the oceanic powers to its coast
and heartland. Given its incremental adaptation to the new geopolitical set-
ting and its sluggish yet solid steps in rebuilding its fiscal-military appar-
atuses, it was not unlikely that the Qing state would have slowly but steadily
transformed itself into a modern power and reasserted its influence in East
Asia following the temporary frustrations wrought by the European states
around 1840 and 1860. Unfortunately, a new development in East Asia in
the mid-1870s added uncertainty to the geopolitical realities in this region,
and that was the rise of Japan, which saw a modernizing China as its chief
geopolitical rival and the biggest barrier to its becoming a dominant power
in the region. Therefore, from the 1870s, conflicts with Japan gradually
replaced the tensions with European powers to be the primary challenge to
the Qing’s foreign relations, which in turn had a profound impact on its
domestic politics. It was in this reshuffled geopolitical setting that China had
to redefine its geopolitical strategy while experiencing the transition to a modern
sovereign state.

The ending of the traditional geopolitical order

Just like the demographic crisis that loomed in the late eighteenth century and
led to the ending of the low-level equilibrium in the fiscal constitution of the
Qing state, signs of a geopolitical challenge emerged in the same period. After
establishing its colonial rule in parts of the South Asian subcontinent, the British
East India Company expanded its trade to China while seeking to establish
normal diplomatic relations with the Qing court, as seen in its sponsorship of
the mission of Lord McCartney to Beijing in 1793, which failed for the obvious
reason of conflict over the etiquette by which the English envoy was received by
Emperor Qianlong but more fundamentally because of China’s lack of interest
in foreign trade, given the self-sufficiency and huge domestic market, and
because of the Qing ruler’s fear of potential damage to the existing system gov-
erning China’s relationship with other states.
 Between the frontier and the coast 113
The growing tension between the English merchants trying to expand their
trade in, and smuggling opium to, China on the one hand, and the Qing
government’s restriction on foreign trade under the monopoly of authorized
Cantonese guilds (the Cohongs) and the confiscation of opium from the English
merchants on the other hand, led to the Opium War from 1840 to 1841, fol-
lowed by the Second Opium War from 1856 to 1860 in which China suffered
the joint attacks of British and French forces. No less important for the Euro-
pean powers was to establish formal diplomatic relations with China under the
conventions that had been widely practiced in the West since the Truce of West-
phalia, central to which was the legal equality of all sovereign states in their rela-
tions to one another. For the Qing court, however, preserving the existing
interstate order and defending its self-assumed supremacy over all other states
were key to maintaining the political order and its legitimacy at home. For the
Manchus originating from outside China and ruling the Middle Kingdom as
conquerors and with dubious legitimacy, safeguarding the supremacy of the
imperial court in its foreign relations was especially important for the court to
legitimate itself as the authentic regime in China. To accept any other countries
as China’s equals or any foreign rulers as parallel to the Qing was to deny the
latter as the supreme ruler of the universe within the Chinese political imagin-
ation and thus constituted a fundamental challenge to the traditional political
order and the legitimacy of the imperial regime.
Not surprisingly, the European powers’ demand for equality in diplomatic
relations with China turned out to be the most difficult part for the Qing court
to accept. Although the Treaty of Nanjing of 1841 provided that communica-
tion between the subordinates of both countries be conducted “on a footing of
perfect equality” (TN: 11), there was no explicit statement of equality between
China and Britain at the state-to-state level, and the tributary system under the
Qing as well as all of its traditional diplomatic practices continued after
the Opium War, except for the Cohong system that was officially ended by the
treaty. The treaty itself, in fact, was kept a secret by the Qing court and never
made known to the public in China. It was only after it was decisively defeated
that the Qing conceded to the Europeans and allowed for equality in China’s
foreign relations under the pressure of military occupation. The Treaty of Tien-
tsin (Tianjin) signed between the Qing and Britain in June 1858 thus prescribed
that the ambassador, minister, or other diplomatic agent from one side would
be sent to the capital city of the other side “in accordance with the universal
practice of great and friendly nations,” and more important, the English diplo-
mat in Beijing “shall not be called upon to perform any ceremony derogatory to
him as representing the Sovereign of an independent nation on a footing of
equality with that of China,” and at the same time “he shall use the same forms
of ceremony and respect to His Majesty the Emperor as are employed by the
Ambassadors, Ministers, or Diplomatic Agents of Her Majesty towards the Sov-
ereigns of independent and equal European nations.” The treaty further
demanded that “the character ‘I’ (barbarian) shall not be applied to the Govern-
ment or subjects of Her Britannic Majesty in any Chinese official document
114 The transition to a sovereign state
issued by the Chinese authorities, either in the capital or in the provinces”
(Hertslet 1908: 33).
Needless to say, none of these requests from the British side was excessive
and unjust in the light of modern international relations. To exchange ambassa-
dors and to officially communicate between two countries on an equal footing
were indeed “universal practices” of the sovereign states within the Westphalia
system of nations, and they by no means caused any damage to China’s sover-
eignty. Chinese historians of the twentieth century, keenly aware of the import-
ance of sovereign rights to a modern nation-state, therefore were puzzled by the
Qing court’s stubborn but futile resistance to these requests of equality in state-
to-state relations and at the same time its easy concession to the British demands
for territory (Hong Kong) and privileges, such as extraterritoriality and most-
favored-nation status, which truly damaged China’s sovereignty (Jiang Tingfu
1939; Xiao Yishan 1967, 3: 708). In fact, the right of the extraterritoriality of
foreign nationals, as a convenient solution for the Qing court concerning dis-
putes involving foreigners, had its historical roots in the “legal pluralism” in
Chinese administrative and jurisdictional traditions (Cassel 2012). At any rate,
the signing of the Treaty of Tientsin symbolized the official ending of the Qing
dynasty’s self-assumed superiority in its traditional geopolitical order and the
beginning of its incorporation into the Europe-centered system of international
relations. In 1861, the Zongli Geguo Shiwu Yamen (or the Tsungli Yamen) was
established as a department of the imperial court to handle foreign affairs as well
as domestic programs in connection with foreign countries; part of its task was
to investigate and selectively adopt the diplomatic conventions of Western coun-
tries, primarily through the translation and publication of foreign treaties on
international laws. As it turned out, the translation of Western-language writings
on international law and their growing popularity among intellectuals and bur-
eaucrats after 1895 was one of the key catalysts behind a transition in the Chin-
ese discourse on international questions from that of Sinocentrism to
“unambiguously that of seeing China as a sovereign nation in the large family of
nations defined by international law” (Svarverud 2011).
Practicing diplomacy with the European powers “on a footing of perfect
equality” turned out to be no less difficult than negotiating and signing the
treaty per se. It was not until 1873 when the young Emperor Tongzhi offi-
cially began his attendance of state affairs at the age of 18 that his reception
of the ministers of Britain, France, Russia, the USA, and the Netherlands
and the ambassador of Japan was held, after several months of negotiation
in which the Qing court finally gave up its insistence on the foreign diplo-
mats’ performance of full prostration before the emperor; as a compromise,
the ministers and ambassador removed their hats and bowed to the emperor
five times instead of only three times as they originally proposed. Earlier in
1868, the Qing court had sent its first envoy, Zhigang, to the USA and
Europe. In 1870, it sent Zhonghou as a minister to France, who was also
the first minister of the Qing to a specific foreign country (Xiao Yishan 1967,
3: 861–862).
 Between the frontier and the coast 115
Frontier Defense versus Maritime Defense
After defeating the Taiping rebels in 1864, especially after suppressing the Nian
and Hui rebels in the late 1860s and early 1870s, the Qing entered an era of
relative order and peace domestically. Meanwhile, the threats from European
powers receded; after having two Opium Wars with the Qing, Britain and
France appeared to be content with what they had secured in China through
the signing of several treaties. By 1866, when the Qing court paid off the war
indemnities to the two European countries, the humiliation inflicted on China
by Western powers seemed to be over. Nevertheless, the Qing soon encoun-
tered new threats from neighboring countries. First in Xinjiang, the widespread
Muslim revolts in the 1860s incurred the invasion of a Kokand-based force led
by Muhammad Yaqub Bek (1820–1877), who soon defeated his competitors
and occupied southern Xinjiang and part of northern Xinjiang by 1870, causing
Russia to occupy the city of Ili in 1871 to safeguard its interests in central Asia.
Trouble also came from Japan, which sent a fleet of more than 2,000 men to
land in southern Taiwan in 1874, which ended in the signing of an agreement
whereby the Qing government offered “compensation” of 500,000 taels to
Japan to avoid having to wage war with it.
These two threats were different in nature. The invasion of the Kokand force
and its creation of an independent state in Xinjiang repeated the geopolitical
crisis that had confronted the Qing in the late seventeenth and eighteenth cen-
turies; subjugating the conventional central Asian enemies entailed massive
expenses on the military, especially on its logistic support, given the long dis-
tance between Xinjiang and the interior provinces, without having to upgrade
its weaponry, because its enemy was not superior to the Qing forces in terms of
armaments. The Japanese invasion of Taiwan, on the other hand, renewed the
maritime threats that had started after the Opium War in 1840 and would con-
tinue to challenge the Qing in the late nineteenth century. This was totally dif-
ferent in nature. The enemy never outnumbered the Qing forces in manpower
but it was equipped with modern weaponry. Surviving the challenge from mari-
time enemies entailed the modernization of the Qing defense, which in turn
meant long-term massive investment in the construction of manufacturing and
infrastructural facilities. It is interesting, therefore, to look at how the Qing state
responded to the different types of geopolitical challenges, one traditional and
one new, with different strategies, and how effective the newly transformed
fiscal system was in meeting the state’s needs to overcome the crisis.
The concurrence of the incidents in Xinjiang and Taiwan in the early 1870s
triggered a debate among provincial leaders over the strategic priority to be
given to either “Frontier Defense” (saifeng, i.e., military campaigns against the
Kokand-based enemy in Xinjiang) or “Maritime Defense” (haifang, i.e., con-
struction of coastal defenses and a modern naval force). Supporters on each side
offered different rationales to justify their preferred strategy.
Zuo Zongtang, Governor General of Shaaxi and Gansu and Imperial Com-
missioner for military affairs in the northwest, was the most ardent in advocating
116 The transition to a sovereign state
Frontier Defense, though he never discounted the importance of Maritime
Defense. He prioritized Frontier Defense by resorting to history and the ration-
ale that the earlier Qing rulers had used to justify their military campaigns in the
northwest, that is, the critical importance of Xinjiang to the safety of Mongolia
and the key role of Mongolia to the safety of the capital city of the dynasty. For
Zuo, the only difference between the times of Emperors Kangxi and Qianlong
and his own times was that Russia had replaced the Zunghars to be the greatest
threat to China in the long run, which shared with the Qing the longest border
from the northwest to the northeast and had already occupied Yili. If the Qing
forces gave up the defense in the northwest and “retreat by an inch,” the enemy
would then “move in by a foot” (ZZT, 6: 191).
Zuo was not alone in speaking for Frontier Defense; he was joined by a few
other provincial leaders, such as the governors of Hunan, Shandong, and
Jiangsu, who shared with Zuo the same concern with the potential threat from
Russia. “The troubles from all other countries,” in the words of Ding Baoz-
hen, Governor of Shandong, “are like the diseases on the four limbs, which
are distant and light, but the trouble from Russians is like the disease of the
heart, which is close and severe” (cf. Jia Shuocun 2004: 91). The most
decisive support to Zuo came from Wenxiang (1818–1876), Minister of Per-
sonnel and Grand Councilor of State, who was one of the two key statesmen
of the Qing court during the reign of Tongzhi, the other being Yixin
(1833–1898) or Prince Gong. Wenxiang concurred with Zuo by distinguish-
ing the Qing from the preceding Ming that had limited its territory to interior
provinces and by emphasizing the strategic importance of Xinjiang to the
safety of Mongolia and the security of the capital city. Wenxiang’s opinion
turned out to be pivotal for the two empress dowagers (Ci’an and Cixi) of the
court to make the decision regarding an expedition to be led by Zuo against
the forces of Yaqub Bek in Xinjiang.
Wenxiang was in fact also the first to propose Maritime Defense. In Novem-
ber 1874, in response to the Japanese extortion for a compensation following
the Taiwan incident, he submitted together with Yixin a memorial to the
emperor, in which they officially suggested a comprehensive program of Maritime
Defense. The target of Maritime Defense, for Wenxiang and Yixin, was no longer
Britain and France that had invaded the southeast coast decades earlier but Japan,
“a tiny state in the eastern ocean who dared to make troubles under an excuse,
having recently practiced Western military skills and bought two ironclads”
(CBYW, Tongzhi, 98: 41). It would be too late, they warned, if no measure was
taken to prepare for further threats from Japan. The court responded by asking
provincial leaders to deliberate on the six concrete measures suggested by the
memorial. Among the replies from the governors and governors general, Li
Hongzhang (1823–1901), Governor General of Zhili and Imperial Commis-
sioner for Trade in the North, was the most ardent in supporting Maritime
Defense, and he did so by downplaying the importance of Frontier Defense.
Li first pointed out a fundamental change in China’s geopolitical situation
in recent decades. Compared to the situation of border defense in the previous
 Between the frontier and the coast 117
dynasties that had focused on the northwest, Li found that contemporary China
was faced with “a totally different situation that has never been seen in the past
several thousands of years” (shu qiannian lai weiyou zhi bianju), in which

the more than ten-thousand li of the southeastern coast is open to the

trade and missionary activities of all countries; [foreigners] travel freely,
flocking to the capital and interior provinces. While pretending to be at
peace [with China] in public, they all harbor the conspiracy of eating up
[China] in secret; whenever a country makes trouble, all others provoke
together.

Meanwhile, China was confronted with “formidable enemies unseen in the

past several thousands of years” (shu qiannian lai weiyou zhi qiangdi):

Their steam boats and telegraphs are as swift as lightning; their military
weapons and machines are a hundred times more efficient than human
labor; their canons and guns destroy all and no passes and posts on the land
or in the water can stop them.
(LHZ, 24: 825)

China, in a word, entered a totally different age.

Li Hongzhang further pinpointed two coastal areas central to China’s
strategic interests: the most important was the area around Dagu, Beitang,
and Shanghaiguan of Zhili, which served as the front gate of metropolitan
Beijing, “the very foundation of All under Heaven”; the other was the area
from Wusong to Jiangyin of Jiangsu, the mouth of the Yangzi River, the
most important source of China’s revenues. In sharp contrast, Li found that
Xinjiang, which had been included in the territory of the Qing as late as
the Qianlong reign, cost as much as “three million taels of silver a year to
keep it even at times of peace”; “it is not worth it to keep the wasteland as
wide as thousands of li and add a leaking wine vessel for hundreds and
even thousands of years.” Surrounded by growing neighbors such as Russia,
Turkey, Persia, and British India, Li observed, “Xinjiang can never be kept
forever in the future, even if it is temporarily restored.” Therefore, he pro-
posed terminating the military operations in the northwest, withdrawing the
force from the frontier, and diverting the spending there to Maritime
Defense. The best solution to the crisis in the northwestern frontiers, Li
suggested, was to allow the Muslim tribes there to govern themselves and
accept the Qing’s suzerainty as did the Vietnamese, Koreans, and the vari-
ous ethnic groups of the southwestern provinces. For Li Hongzhang, “the
failure to recover Xinjiang does no harm to the fundamental health of the
body; leaving the maritime frontier unguarded, however, would complicate
the severe disease of the heart” (LHZ, 24: 825). Li’s view, in a word,
was to allow the vast area of Xinjiang to return to its original state before
the 1750s.
118 The transition to a sovereign state
Most of the governors and governors general of coastal provinces as well as
a few Manchu courtiers supported Li Hongzhang’s opinion. But proponents of
Frontier Defense eventually prevailed during the debate in late 1874 and early
1875. The reasons behind the triumph of Frontier Defense supporters were
multiple. First, despite Li’s exceptionally keen insight on China’s entry into
a new era when its geopolitics was completely redefined by its enemies who
were equipped with modern weaponry, and despite his emphasis on moderniz-
ing China’s coastal defense and naval forces, which turned out to be a key factor
shaping the country’s destiny in the late nineteenth century, the peaceful hand-
ling of Japan’s brief intrusion into Taiwan with an insignificant sum paid to
Japan did not alert most of the Qing high-ranking officials to the severity of the
unprecedented challenge to China as Li Hongzhang had correctly perceived.
For most of the ruling elites, the trouble was over, and a tiny neighbor such as
Japan was far from becoming a real and imminent threat to China; Maritime
Defense was indeed important to China’s survival in the long run, but it was by
no means as urgent as Frontier Defense at the time of the debate when the vast
territory of Xinjiang had already fallen into the hands of the enemy. Second, the
Qing court and most of the ruling elites inevitably continued to define China’s
strategic priorities from the perspective of their traditional geopolitical thinking,
irrespective of the new challenges that confronted the country. For them, Xin-
jiang was important not only because it was critical to the safety of Mongolia
and the security of the imperial capital but more important because it was
a legacy of the preceding reigns of the Qing. Safeguarding the ancestors’
bestowal was central to their legitimacy as the rightful ruler of the dynasty.
Finally, having recently survived the domestic upheavals and foreign humili-
ations, the Qing state witnessed the improvement of its fiscal conditions,
which, compounded with the pacification of the interior provinces and restor-
ation of a peaceful relationship with European powers, added to the court’s
determination to solve the problem in Xinjiang, hence Zuo Zongtang’s
legendary Western Expedition of 1875–1878.
It should be noted, however, that the Qing court did not ignore the needs of
Maritime Defense. In fact, in May 1875, the same month when the decision on
the Western Expedition was made, the court also decided to establish two
modern fleets in northern and southern China, with Li Hongzhang and Shen
Baozhen (1820–1879, Governor General of Liangjiang) appointed as the imper-
ial commissioners responsible for coastal defense in the north and the south and
building the two fleets respectively. This development was strategically signifi-
cant. For more than three decades since the conclusion of the Opium War of
1840, the Qing had made no substantial steps towards the strengthening of its
Maritime Defense. The Second Opium War, which culminated in foreign troops
occupying Beijing, had far greater impact on the Qing ruling elites than the Jap-
anese momentary intrusion into Taiwan, though it did not cause the imperial
court to start the construction of modern naval forces. A widely shared opinion
among the ruling elites since 1840, beginning with Lin Zexu and Emperor
Daoguang, had been that the Europeans from afar did not intend to occupy
 Between the frontier and the coast 119
China’s territory but only aimed at expanding their commercial interests in
China; therefore, it was unnecessary and impractical for China to have a war at
sea with European powers; the only way to deal with them was waiting until the
enemy landed on the coast and then annihilating it (CBYW, 13: 23). But the
intrusion of Japan into Taiwan in 1874 was of a different sort and an omen of
greater trouble for China in the future. In the view of Li Hongzhang, Japan
“emulates Britain and the U.S. for everything and is bound to be an immedi-
ate and imminent threat (zhouye zhi huan, literally ‘the decease of the upper
arm and armpit’) for China” (LHZ, 30: 439–440), and it “goes even further
than the West in craftily bullying others”; to establish the naval force and
strengthen Maritime Defense, therefore, was primarily to deal with Japan, not
the West (LHZ, 9: 261).
The Qing court’s decision to build two modern fleets in 1875 thus signaled
a transition from its conventional geopolitical strategy that prioritized the
defense of Inner Asian frontiers to the new one that stressed both Frontier
Defense and Maritime Defense. The concept of Frontier Defense essentially con-
tinued the traditional Inner Asian strategy that had preoccupied the Qing rulers
since the late seventeenth century; the only change was that Russia increasingly
replaced Central Asian tribal forces to become China’s greatest threat in the
northwestern, northern, and northeastern frontiers. In sharp contrast, the con-
cept of Maritime Defense was brand new; it reflected the Qing rulers becoming
alert to the new threat from Japan. As it turned out in the rest of the nineteenth
and first half of the twentieth centuries, it was primarily the threat from Japan,
rather than any other foreign power, that would redefine China’s geopolitical
setting in the modern era and have a profound impact on its domestic politics.
It is thus interesting to examine how the Qing simultaneously handled the two
different kinds of challenges in the 1870s and before the eventual outbreak of
the Sino-Japanese War of 1894.

Centralized regionalism at work

Financing the Western Expedition

The single decisive factor leading to the stark success of Zuo’s expedition was
the Qing state’s generous funding of this project, which amounted to
26.7 million taels of silver during the three years from 1875 to 1877 and
another 25.6 million taels in 1878–1881, thus totaling 52.3 million taels during
the seven years (Liu and Smith 1980: 239). In fact, the actual expenses on the
Western Expedition (1875–1877) could be as high as 42.71 million taels. The
total spending on Frontier Defense (from 1875, when the Western Expedition
started, to 1884, when the Province of Xinjiang was established) was even
more, ranging from 70 to 80 million taels (Zhou Yumin 2000: 266–267). Why,
then, was the Qing state able to pool so much money on this project?
The most important source of the funding for the Western Expedition was
Coordinate Remittance from different provinces, which totaled 20.49 million
120 The transition to a sovereign state
taels, or about 48% of the funds that Zuo received during 1874–1877. The
annual amount of Coordinate Remittance assigned by the court to each
province varied, ranging from 240,000 taels for Anhui, 360,000 taels for
Jiangsu, and 480,000 taels for Fujian, Jiangxi, and Hubei, to 840,000 taels
for Guangdong and 1,440,000 taels for Zhejiang, most of which came from
the lijin revenues of the provinces. Altogether, the annual Coordinate Remit-
tances from all provinces for Zuo’s Western Expedition should have been
“more than seven million taels” officially (GXCD, 1: 167). But few provinces
remitted their respective quotas in full. The actual amount that Zuo received
in 1874, for instance, was only 4.69 million taels (Jiang Zhijie 1988);
according to Zuo’s memorial in October 1875, the total amount of delin-
quent remittances owed by individual provinces had reached 27.4 million
taels since Zuo was in charge of military affairs in the northwestern provinces
in 1866, and Fujian province alone had more than 3 million taels in arrears
(GXCD, 1: 167; Shi and Xu 2008: 141).
The situation did not improve in the following years. To make up the def-
icit in remittances, Zuo resorted to a second source, borrowing foreign and
domestic debts. During the three years of the Western Expedition in Xinjiang,
Zuo obtained three loans from foreign banks, including two in 1875 (1 million
and 2 million taels, respectively) and one in 1877 (5 million taels), with the
assistance of the famous financier Hu Guangyong (1823–1885) as his agent in
Shanghai. In addition, Zuo borrowed from individual Chinese merchants a total
of 5.6 million taels. All of these debts would be repaid by individual provinces
and maritime customs, to offset the delinquent Coordinate Remittances they
had owed to Zuo; and the provincial and custom administrators were required
by the imperial court to affix their seals to the transaction documents to guaran-
tee their timely repayment of the debts (Ma Jinhua 1911: 71). Therefore, these
debts, 13.6 million taels in total, can be counted as part of their Coordinate
Remittances as well. Other sources include the direct allocation of funds from
the Board of Revenue’s cash reserve; a total of 4.5 million taels was sent to Zuo
in 1874–1877, accounting for 10.5% of all funds Zuo received during those
years’ donations and contributions (1.67 million taels), and other miscellaneous
sources (Jiang Zhijie 1988).
Thus, despite the common delinquency among provincial leaders in fulfilling
their financial commitments to the Western Expedition, the actual funds Zuo
received, totaling more than 42 million taels during the three years of military
campaigns, or more than 70 million taels for the decade-long (1875–1884)
endeavor of Frontier Defense, were sufficient to warrant his success on the
battleground, while the exceptionally high morale of his forces, boosted by
Zuo’s personal determination and the generous stipends on his soldiers, was no
less important in explaining the recovery of Xinjiang, a shining accomplishment
of Tong-Guang Restoration.
The mechanism that enabled the largely successful funding of this project
can be termed “regionalized centralism,” as discussed in the preceding chap-
ter. Nearly 80% of the funds that Zuo received for his military operations
 Between the frontier and the coast 121
came from individual provinces in the form of either Coordinate Remittance
or the repayment of Zuo’s debts by the delinquent provinces, and the most
important source of provincial funding was lijin, which was generated and
administered locally by provincial authorities and independently of the central
government’s direct control. The provincial governors or governors general
managed to delay and reduce their remittance or to bargain with the central
government over the quota of their remittance, or tried their best to support
Zuo, depending on their personal relationship with the latter. The Western
Expedition thus witnessed the continuation of the kind of regionalism that
had emerged during the Taiping Rebellion. Nevertheless, unlike Zeng Guofan
and other provincial leaders back in the 1850s who had to raise funds on
their own for their militia corps without substantial support from the central
government at all during the most difficult years of the campaign against the
rebels, Zuo received critical and consistent support from the court. Having
reestablished the remittance system, the Board of Revenue was able to par-
tially fund Zuo’s campaign with money directly from its own cash surplus,
a fact that contrasted sharply with its dearth of funds in the 1850s, and it
was also able to impose on individual provinces, especially those in the south-
east with the largest revenue from lijin, mandatory quotas and force those in
delinquency to pay off the foreign or domestic debts on Zuo’s behalf. It pun-
ished the few provinces and maritime customs that had failed to remit at least
80% of their quotas (Liu Zenghe 2017). What accounted for the relative suc-
cess of the funding of the Western Expedition, in a word, was the combin-
ation of the regionalized means of revenue generation with the centralized
means of revenue reallocation.
An immediate result of Zuo Zongtang’s stunning victory over the Kokand-
based forces during the three-year Western Expedition in 1875–1877 was the
full integration of Xinjiang into the administrative system governing the interior
provinces. Except for a few prefectures and counties in northeastern Xinjiang
populated by the Han Chinese, for more than a century since it defeated the
Zunghars and Uighurs in the late 1750s and early 1760s, the Qing had ruled
the rest of Xinjiang through the military government (junfu) system in which
the General of Ili (Yili jiangjun) was at the top of a hierarchy of military com-
manders at different levels to garrison separate areas of Xinjiang, leaving the rou-
tine administration of the civilian affairs of local populations to the chieftains of
different ethnic groups under the beg (boke in Chinese) system for Uighurs in Ili
and Southern Xinjiang and under the jasak (zhasake in Chinese) system for the
Mongols, Kazaks, as well as the Uighurs in the rest of Xinjiang (Ma and Ma
1994: 333–356; Perdue 2005: 338–342; Cassel 2012: 19). All of those who
filled the key positions of the military units in Xinjiang, ranging from the General
of Ili to his subordinates at different localities, had been mostly the Manchus
since the 1760s.
With the military government largely destroyed by the rebellions of local
populations and the invasion of the Kokand forces, the Western Expedition and
the subsequent control of Xinjiang by Zuo Zongtang’s forces fundamentally
122 The transition to a sovereign state
changed the political landscape; most of the key military and civilian positions in
Xinjiang came into the hands of the Han Chinese who had been Zuo’s subor-
dinates (the so-called “Hu Xiang dizi man Tianshan”) (Wang 2010: 15). Zuo
thus was in a position to call for turning Xinjiang into a regular province and
thereby substituting the junfu system with the standard bureaucratic system
(junxian). His official rationale was of course to enhance the Qing govern-
ment’s control of Xinjiang, but a more practical reason was to open opportun-
ities for the Han Chinese, mostly his followers, to perpetuate their control of
local government positions. The provincialization of Xinjiang thus was in
essence the continuation of the strengthening of the position of the Han bur-
eaucrats at the expense of Manchu elites in the interior provinces during and
after the Taiping Rebellion. Not surprisingly, the Qing court hesitated initially
in responding to Zuo’s proposal. The final solution was a compromise: Xin-
jiang officially became a province in 1884, and the position of the General of
Ili was also kept, but it was reduced from the military commander-in-chief for
the entire Xinjiang to the leader of Banner troops in a limited area (Ili and
Tarbagatai), while the rest of Xinjiang was reorganized into prefectures and
counties under the civilian Governor of Xinjiang. The beg and jasak systems
throughout Xinjiang were abolished, and a large portion of native dwellers,
who had been serfs to local nobles, now became landowners and taxpayers
(Ma and Ma 1994: 359–360).
As it turned out, the Sinicization of Xinjiang in the 1880s was a critical step
leading to its full integration with China after the fall of the Qing. With most of
the government positions in the hands of Han Chinese after the founding of
Xinjiang province, the Republican decades saw their continual control of Xin-
jiang and their strong commitment to staying within China as an integral prov-
ince. In sharp contrast, the failure of the Qing court to establish two proposed
provinces in Outer Mongolia and subsequently the limited penetration of the
Han Chinese into the administrative authorities and the limited migration of the
Han Chinese into this area accounted for its loss of identity with the Chinese
state after the fall of the Qing and its eventual separation from China in the
Republican years.

Financing Maritime Defense

As mentioned earlier, the Qing court was also committed to funding Maritime
Defense as a result of the debate in 1875. But Maritime Defense was less urgent
and important than Frontier Defense to the Qing court because Xinjiang was
already in the hands of an invading force and losing Xinjiang meant a grave and
imminent threat to the stability of the Inner Asian frontiers in the Manchu
rulers’ traditional geopolitical thinking. Therefore, despite Li Hongzhang’s
request of “at least a budget of more than ten million taels to begin with” for
the three Maritime Defense tasks of “purchasing vessels, training soldiers, and
procuring weapons,” the imperial court only came up with an allocation of
4 million taels per year, to be distributed equally to Li Hongzhang and Shen
 Between the frontier and the coast 123
Baozhen, the two commissioners responsible for building two fleets in north
and south China. Half of the 4 million taels would be derived from maritime
customs and another half from the lijin revenues of six southeastern provinces
(Jiangsu, Zhejiang, Fujian, Jiangxi, Hubei, and Guangdong) (Zhang Xia et al.
1982: 615–617).
Li Hongzhang realized from the outset that the budget of 4 million taels for
Maritime Defense was more of a promise than a reality; he knew well that the
maritime customs did not have much left after fulfilling their obligation of
Coordinate Remittance while the six provinces had committed most of their
lijin income to other expenditures. Not surprisingly, during the first three years
(1875–1877), the actual amount of funds spent on Maritime Defense was only
about 2 million taels, or roughly 670,000 taels a year on average. As Li com-
plained in 1877, the six provinces had remitted either half of their Maritime
Defense quotas or none at all, while the maritime customs had to divert half of
their quotas to repay the debt for the Western Expedition. To ensure that the
Northern Fleet, which was strategically more important for the security of the
capital city, had the necessary funds to start up, Shen Baozhen willingly gave up
his share of half of the Maritime Defense funding for the first three years so that
Li had more to spend on the Northern Fleet. The situation worsened in the
next three years (1878–1880), when the funds had to be evenly divided
between the two fleets while the six provinces remitted even less than before.1
The exceptional difficulties Li and Shen encountered in obtaining the funds
promised by the imperial court accounted for the sluggish progress in Maritime
Defense during the decade from 1875 to 1884.2 Thus, by the outbreak of the
Sino-French War in early 1884, China’s Maritime Defense forces remained rudi-
mentary in nature. The Northern Fleet was equipped with two cruisers, six steel
gunboats, four wooden gunboats, and a few other vessels for logistic and train-
ing purposes, totaling 12,296 tons and 10,655 horse power (Fan Baichuan
2003: 1017). The Southern Fleet was smaller, with 14 small-size vessels, mostly
manufactured by the Kiangnan Arsenal in Shanghai and the Foochow Arsenal in
Fuzhou. In addition, two other even smaller naval forces were established in
Fujian and Guangdong, each having more than ten vessels of varying sizes. Not
surprisingly, when the Sino-French War broke out, the southeastern coast was
poorly guarded, and the French forces easily destroyed the 11 vessels of the
Fuzhou naval force as well as almost the entirety of the Foochow Arsenal in
August 1884, although they later suffered a sound defeat at the hands of the
Qing army on the border between China and Vietnam in February 1885.
The Sino-French War was a turning point in the course of Maritime Defense.
Right after signing the truce with France, the Qing court announced a decree in
June 1885 to declare its determination on “investing in the naval forces on
a massive scale” (da zhi shuishi) (LHZ, 11: 99). This announcement signaled
a shift in the Qing state’s geopolitical strategy from prioritizing Frontier Defense
over Maritime Defense during 1875–1884 to focusing solely on Maritime
Defense afterwards. As an initial step, the imperial court ordered the establish-
ment of the Admiralty (haijun yamen) in October 1885, which was responsible
124 The transition to a sovereign state
for the unified management of the funding of Maritime Defense for all fleets. At
the Admiralty’s request, the court repeatedly urged individual provinces and
maritime customs to prioritize the full remittance of their respective Maritime
Defense quotas, totaling 4 million taels a year, over all other commitments, thus
making it more difficult for the provinces to default on their payments, though
delinquency never ceased. The court further generated an additional revenue of
4.5 million taels through a three-year program (1885–1887) of selling govern-
ment titles under the name of “contributions to maritime defense” (haifang
juanshu) and another 4 to 5 million taels through a second round of contribu-
tions (1890–1894); it also ordered the Board of Revenue to commit an extra
1 million taels to Maritime Defense funding from the source of lijin on the
opium trade (Fan Baichuan 2003: 1139–1143). Altogether, the Qing govern-
ment spent a total of 41.248 million taels on Maritime Defense during the
decade from 1885 to 1894 (ibid.: 1171), averaging more than 4 million taels
a year; the Northern Fleet alone received 23.18 million taels, more than 56% of
the Qing government’s total spending on Maritime Defense (ibid.: 1228).
Compared to the preceding decade (1875–1884) when the Northern Fleet
received less than 1 million taels per year on average (ibid.: 1103), its funding
during this decade more than doubled, thanks to the Qing state’s stronger commit-
ment to Maritime Defense and, more importantly, to the overall improvement of
its fiscal condition during this decade.

Why Maritime Defense failed

However, despite the large amount of funding on Maritime Defense during the
two decades from 1875 to 1894, about 69 million taels all told (including
about 28 million taels in 1875–1884 and 41 million taels in 1885–1894, see
Fan Baichuan 2003: 1173, also 1033–1034, 1171), the Qing failed to achieve
its original goal, that is, to fend off the threats from Japan and safeguard
China’s territory. Beginning with a debate triggered by Japan’s invasion of
Taiwan in 1874, Maritime Defense ended in the complete destruction of the
Northern Fleet during the Sino-Japanese War in 1894 and the eventual cession
of Taiwan to the Japanese.
To understand this outcome, let us first look at how the funds for Maritime
Defense were spent in 1885–1894. Regrettably, the doubling of the funding for
the Northern Fleet did not result in corresponding progress in the building of
its combating capabilities. This is evident in its spending on new warships,
which totaled 4.44 million or more than 34% of the fleet’s total spending in
1875–1884 (including the cost of two ironclads that remained the most import-
ant battleships of the fleet in 1894) but only 2.8 million taels or 12% of the
fleet’s total spending in 1885–1894 (mostly the cost of four cruisers and one
torpedo boat); the same was true about its spending on imported weapons,
which cost 2.7 million taels or nearly 21% of its total spending in 1875–1884
but only 2.18 million taels or 9.4% of its funding in the following decade. On
the other hand, the Northern Fleet spent nearly 9 million taels or more than
 Between the frontier and the coast 125
37% of its funding as salaries to its Chinese and foreign crews in 1885–1894,
compared to only 2 million taels or 15% in the preceding decade (Fan Baichuan
2003: 1227–1228). In sharp contrast with the first decade when the fleet’s
spending on its personnel was only 28% of its expenses on vessels and weapons,
in the second decade its payment to the former was 1.78 times the cost of the
latter. Obviously, the fleet should have been able to spend more on expanding
its battle capabilities than on its personnel.
It should be noted that the establishment of the Admiralty, while enabling its
centralized management of the funds for Maritime Defense and thereby making
possible the increased funding of the Northern Fleet, also discouraged provincial
leaders of the southeast coast from developing their own naval forces, because it
allowed the Northern Fleet to use and even requisition their vessels under the
pretext of coordinated exercises, and because it kept its allocation of funds to
the naval forces of the southeastern provinces to the minimal level, for
example, only 6.5 million taels of the Southern Fleet in 1885–1894. Another
important purpose of the Admiralty, as it turned out, was to facilitate the
Qing court’s extraction of funds from provincial sources for its own projects,
most notoriously the reconstruction of the Summer Palace (Yiheyuan), under
the pretext of raising funds for Maritime Defense. It is estimated that the
Qing court obtained a total of 14 to 16 million taels in this way, which would
have permitted the full establishment of the Southern Fleet (Fan Baichuan
2003: 1146–1157). In fact, as a result of the Qing court’s diversion of funds
from Maritime Defense to its own luxurious expenses, the development of the
Southern Fleet and the naval forces in Guangdong and Fujian slowed down or
came to a standstill; the original grand plan of Maritime Defense, that is, to
establish three full-fledged fleets to patrol China’s waters in the north, the
east, and the south, was never realized.
To what extent, then, did the Northern Fleet, the only one that was fully
established, satisfy China’s needs for Maritime Defense? The first and foremost
goal of the Northern Fleet, as Li Hongzhang and other proponents of Maritime
Defense clearly stated, was to deal with Japan. Indeed, since Japan showed
off its ironclad when invading Taiwan in 1874 and later again when annexing
Liuchiu in 1879, there had been a naval armaments race between the two coun-
tries, especially in purchasing ironclads. For Li Hongzhang and Shen Baozhen,
the two commissioners responsible for establishing the Northern and Southern
Fleets, possessing the largest and the best equipped ironclads was the only way
to outcompete Japan and wipe out its threat to China’s coastal waters. The pur-
chases of two battleships in 1880 and 1881, named Dingyuan and Zhenyuang,
respectively, partially fulfilled their goals, both being the largest (7,000 tons
each) and the most advanced in the Far East. By 1888 when the Northern Fleet
was fully established, it stood out as the biggest in the Far East, far surpassing
the Japanese fleet in tonnage and battle capacities (Fan Baichuan 2003: 1113;
for a complete comparison between China and Japan of their naval forces from
1875 to 1894, see ibid.: 1175–1176). Therefore, Li Hongzhang had a good
reason to claim with confidence that “as far as the defense of the Bohai Bay is
126 The transition to a sovereign state
concerned, a state of solidity and invincibility have been achieved” (jiu Bohai
menhu er lun, yiyou shengubuyao zhi shi). For him, the visit of the Northern
Fleet to several Japanese cities in the summer of 1891 was indeed a long-
awaited occasion to show off the latest development of China’s naval force.
Unfortunately, after its establishment in 1888, the Northern Fleet stopped
buying any new vessels and upgrading its weaponry, partly because of its inflated
spending on personnel, and partly because of the Qing court’s diversion of its
revenue from Maritime Defense to its own recreational projects. As a result,
only three vessels made at the Foochow Arsenal were added to the fleet during
the following years until 1894 (Fan Baichuan 2003: 1095–1096). By contrast,
the Japanese government, stimulated by the visiting Chinese fleet, was deter-
mined to outcompete China in naval armaments; it drastically increased its
spending on the navy, from about 5 million yen a year in the late 1880s to
10 million yen in 1891 and even more in each of the following years, account-
ing for about one-tenth of the Japanese government’s annual expenditure, and
thereby quickly narrowing its gap with the Qing. Thus, by the outbreak of the
Sino-Japanese War, the naval forces of the two countries were largely compar-
able to each other in terms of the number of vessels, total tonnage, and battle
capacities; in general, however, the Japanese warships, many of which were
newly made in the early 1890s, were speedier than the Chinese ones. Li Hon-
gzhang realized the danger of his fleet’s lagging behind Japan’s. He noted on the
eve of the Sino-Japanese War:

a tiny country such as Japan can still save enough to add large vessels annu-
ally. China, however, has not added a single vessel since establishing the
Northern Fleet in 1888. What can be done is only to diligently train the
twenty-plus vessels. Whether this can continue is really worrisome.
(LHZ, 15: 335)

Shortly after the outbreak of the war, Li requested an additional allocation of

2 million taels to buy new cruisers, but it was too late. Not surprisingly, when
the war was over, Li lamented:

Had the naval budget been fully allocated every year, the vessels and canons
of the Northern Navy would have been on top of the world in less than ten
years. How could there be such a huge defeat? I would not take the blame
for the failure.3
(Xiao Yishan 1967: 942)

In the final analysis, then, insufficient funding, more than any other factor,
led to the failure of Maritime Defense. Compared to its spending on Frontier
Defense, which totaled 42 million taels during 1875–1877, averaging
14 million taels a year, the monies on the Northern Fleet during the key decade
of 1885–1894 were less than 23 million taels, or 2.29 million taels a year, only
one-sixth of its annual spending on Frontier Defense. In fact, Maritime Defense
 Between the frontier and the coast 127
only accounted for a small fraction of the Qing state’s total spending on the
military. The largest portion of its annual budget (36 million taels!) went to the
traditional but increasingly useless military forces (18 million taels a year on the
Eight Banners and Green Standards and another 18 million a year on local
corps or yong camps) (Shi and Xu 2008: 227). The Qing state spent so little on
Maritime Defense not because it was fiscally impossible to do so, but because of
its lack of determination to maximize its spending on the Northern Fleet and
other naval forces, which in turn had to do with its perception of China’s geo-
political situation. For most of the decision-makers in the Qing court in the
1880s and early 1890s, the threat from Japan was only potential in nature; it
was not as urgent as the Kokand forces’ occupation of Xinjiang back in the early
1870s that was real and fatal to the Qing state’s conventional geopolitical inter-
ests. This was especially true in the few years after 1888. The three decades of
the “Tong-Guang Restoration” had made solid headways by that time; the
Qing state had succeeded in quelling domestic revolts, maintaining a peaceful
relationship with foreign powers, rapidly increasing its revenues and generating
a sizeable surplus, and modernizing its defense, which culminated in the full
establishment of the Northern Fleet that far surpassed the Japanese naval force
at that point. All these developments, truly impressive as they were, caused the
Qing elites to form an illusion that the decades of humiliation at the hands of
foreign powers was over and that China had reemerged as a dominant, modern-
izing power in the Far East (YBSWJ, 2: 688). China only needed to observe its
treaties with the European powers to avoid further troubles; Japan, on the other
hand, did have the potential to make trouble along China’s coast with its
quickly modernizing fleet but remained a secondary power that was too small in
size to be a serious and imminent threat to China.
Thus, despite China’s repeated defeats by European powers during the two
Opium Wars and its more recent confrontation with French forces in 1884, the
ruling elites adhered to their faith in the validity and legitimacy of the basic
institutions underlying the Qing state. Their goal of statecraft was to preserve
the status quo by reestablishing the institutions that had deteriorated during
the times of disorder, adjusting the state’s defense strategy, or borrowing the
advanced technologies from the West to enhance and protect those time-
honored institutions. They wanted no more changes or improvements as
long as the status quo in domestic politics or diplomatic relations could be
maintained. It was this mentality among the Qing elites that explained why
the court failed to keep increasing its funding on Maritime Defense after the
goal of restoring domestic order and geopolitical security was temporarily
achieved in the late 1880s and early 1890s. It was not until 1894 when the
Qing suffered a fatal defeat at the hands of the Japanese, resulting in the loss
of Taiwan and the payment of a huge sum of war indemnity, that the Qing
ruling elites were eventually awakened from the illusion of a restoration. It
was also because of this defeat that the Qing state eventually lost its feature
as a suzerain dynasty in East Asia and experienced a metamorphosis into
a modern sovereign state.
128 The transition to a sovereign state
The transition to a sovereign state
For more than two centuries since establishing its rule in China, the Qing had
managed its relationship with neighboring states under the tributary system.
Central to this system was the suzerainty of the Qing court over the latter. As
an extension of China’s domestic political order, the proper relationship
between the court and the tributary states was designed to be the one
between the ruler (jun) and his subjects (chen). The first and foremost pur-
pose of the tributary system, therefore, was to evince China as the center of
the universe and the emperor of China as the universal ruler of “All under
Heaven,” which in turn served to confirm and enhance the political order at
home and, above all, the standing of the emperor as the legitimate ruler of
China. Another and a more practical purpose of this system was using the
tributaries as the surrounding “fence” to enhance China’s security. The secur-
ity strategy of the Qing thus was using “Liuchiu to guard the southeast,
Korea to guard the northeast, Mongolia to guard the northwest, and Vietnam
to guard the southwest” (cf. Sun Hongnian 2011: 20). The European con-
cept of sovereignty was alien to both the Qing and the surrounding tributary
states in managing their mutual relations. While they did have customary bor-
ders to separate their respective territories and rights from each other, the
tributary states saw the Qing court as the ultimate source of legitimacy for
their own power, while the Qing was expected to show its generosity by
granting its territory in dispute to the former if they appeared to be truly sub-
missive and respectful (Sun Hongnian 2006: 24).
Unfortunately, as Western powers came to expand their commercial and colo-
nial interests in the Far East, and as Japan quickly upgraded its military to chal-
lenge the existing geopolitical order in East Asia, the tributary states of the
Qing became the first targets of their aggression and colonization. Obligated to
protect the tributary states but driven primarily by security reasons, the Qing
court was inevitably involved in conflicts with the imperialist powers, and its fail-
ures in this regard resulted in not only the ending of the tributary system after
the neighboring states became “independent” by the treaties between the Qing
and the colonial powers but also the unprecedented disasters that befell the
Qing dynasty itself. Depending on how strategically important the tributary
states were, the Qing court’s involvement in protecting each of them varied.
Liuchiu was the very first among the tributary states to cease its traditional ties
with the Qing.4 Having been a tributary kingdom to China from the beginning
of the Ming but controlled in actuality by the Satsuma han of Japan since the
early seventeenth century, the Liuchiu islands were officially turned into
the Okinawa Prefecture by the Meiji government of Japan in 1879. In response,
the Qing court intervened, attempting to restore Liuchiu’s traditional relation-
ship with itself, but it soon gave up, on the grounds that Liuchiu was, in the
words of Li Hongzhang, “only a tiny island in isolation on the sea, far from
China but close to Japan” or, as the governor and governor general of the
southeastern provinces stated, “has never been a place of strategic importance;
 Between the frontier and the coast 129
without the benefit of defending the border, it only causes the worry of incur-
ring provocation by neighboring countries” (cf. Wang Yunsheng 2005, 1: 153).
Vietnam’s attitude toward its tributary status in relation to the Qing was hesi-
tant and opportunist. Since the early nineteenth century, it had showed reluc-
tance to acknowledge the Qing’s suzerainty and aimed to achieve its full
autonomy in managing diplomatic issues, especially after witnessing China’s
repeated defeat by European powers; it thus did not notify the Qing court at all
when ceding its three southern provinces to France in 1862 and even tried to
terminate its tributary relationship with the Qing by turning to France for pro-
tection. As a result, the Qing court acted only as an onlooker of this event – as
long as the French encroachment in Vietnam did not constitute an immediate
threat to China’s security. However, when France threatened to invade the
northern part of its territory in the early 1880s, the Vietnamese court had to
seek protection from the Qing by appealing to its traditional tributary relations
with the latter. Responses from the ruling elites of the Qing varied. Li Hon-
gzhang, Governor General of Zhili, insisted on a non-intervention stance, on
the grounds that China did not have to risk war with a European power for
a disloyal neighbor and that its naval force was ill-prepared for a possible sea
battle. However, having recently won the Western Expedition and upgraded the
military with modern weaponry for two decades, the war party prevailed and
argued for the “protect[ion of] the fence to consolidate the territory” (bao fan
gu yu) on the grounds of France’s possible encroachment on China’s southern
border after its complete control of Vietnam (Zheng and Zheng 1993). The
Qing won the decisive battles in the border area but suffered the French
fleet’s fatal attack on the fleet and shipyard of Fujian province as mentioned
earlier. Without having to pay a reparation to France, the war nevertheless
cost the Qing somewhere from 20 to 30 million taels, and China lost Vietnam
as its “fence” forever by recognizing the French protectorate over Annam and
Tonkin.
Even more costly to the Qing was its involvement in the crisis of Korea and the
subsequent Sino-Japanese War of 1894. The leading decision-makers of the Qing
unanimously acknowledged the extreme importance of Korea to China’s strategic
security: neighboring Manchuria and close to the metropolitan area of the capital
city, Korea’s relationship with China was likened to lips and teeth: “losing the lips
will cause the chilliness of the teeth” (chunwangchihan). Therefore, while allowing
the Korean court full autonomy in managing its domestic affairs, the Qing was
committed to protecting its immediate and loyal “fence,” which paid tributary
visits to Beijing more often than any other tributary state. Not surprisingly, when
a military coup erupted in 1882 and later a peasant rebellion swept Korea in 1894,
the Qing government actively intervened by sending troops to suppress the rebels
and playing a key role in reshaping its domestic and diplomatic policies, and
thereby turning the tributary relationship from the one of formality during normal
times to its actual dominance of the Korean court during the times of crisis in the
1880s and early 1890s, hence its inevitable confrontation with Japan, which had
long coveted the peninsula. The resultant Sino-Japanese War was devastating to
130 The transition to a sovereign state
the Qing: it not only lost Korea as its most important tributary state but further
ceded Taiwan to Japan and paid to the latter 250 million taels as war indemnity,
which was more than three times the Qing government’s total spending on Mari-
time Defense in the preceding two decades.
Nevertheless, the late Qing state succeeded by and large in keeping its frontiers,
thanks in part to the historical, cultural, and religious bonds that the Manchu
rulers had maintained with the Tibetans, Mongols, and other frontier peoples, and
in part to the renewed military and fiscal strength of the dynasty that accounted
for its effective defense of the frontier in the late nineteenth century. The last
decade of the Qing further saw the imperial court’s systematic efforts to incorpor-
ate the frontiers into provinces. As part of the “New Policies” (see next chapter),
the Qing government encouraged the migration of Han Chinese into the frontiers,
legalized their intermarriage with non-Han peoples, and established three prov-
inces in Manchuria, following the precedents of Xinjiang and Taiwan in 1884 and
1885 respectively; Inner Mongolia almost became a province on the eve of the fall
of the Qing and eventually did so in 1914; serious proposals were also made to
turn Tibet and Outer Mongolia into provinces and thereby fully integrate them
into the centralized bureaucratic system under the Qing (Jiang Bingming 1990:
58–60; Sun Hongnian 2004; Li Yongjun 2011).
To recapitulate. In response to the rampant unrest under the mounting
population pressure in interior provinces, crises on its northwestern frontier,
and the growing threat from Japan, the Qing state experienced a series of
fundamental shifts in statecraft in the late nineteenth century. The most
important among such shifts were: (1) the reorientation of its geopolitical
strategy from stressing the security of its Inner Asian frontiers to safeguarding
its east and southeast coasts; (2) the transformation of its fiscal-military con-
stitution from the imperial court’s centralized control to the growing auton-
omy of provincial authorities; and (3) the rebuilding of its foreign relations
by giving up its claim of suzerainty over the surrounding tributaries and
accepting all other states as its equals. It was in the process of defending its
Inner Asian frontier against foreign invasion, building a modern defense to
fend off the Japanese threats, establishing formal diplomatic relations and
handling disputes with foreign countries under international law (wanguo
gongfa) that late Qing China gradually shook off its old, self-assumed image
as a “celestial dynasty and superior state” (tianchao shangguo) and became an
ordinary member of the world system of states at the expense of its territorial
entity and sovereign rights.

Notes
1 Three provinces (Jiangsu, Guangdong, and Fujian) never remitted at all, while the
other three and the maritime customs altogether remitted only about 300,000 taels
a year on average, or 15% of what Li Hongzhang should have received. According to
Li, the reason that the aforementioned three provinces failed to remit was because
“these provinces simply kept [their lijin income] for self use” or because they “have
been squeezed by the Metropolitan Remittance and various Coordinate Remittances”
 Between the frontier and the coast 131
(LHZ, 34: 10). Whenever these provinces were confronted with conflicting commit-
ments over the limited funds available to them, they would prioritize other more
urgent expenditures over Maritime Defense. Thus, the part of lijin income that should
have been reserved as remittance for Maritime Defense was diverted for use as either the
repayment of foreign debt or the compensation to foreigners involved in the “religious
incident” in Yunnan, or the relief funds in northern provinces, or expenses on various con-
struction projects. All these provinces, as Li lamented in 1880, “see Maritime Defense as
of little importance” (shi haifang wei wuzuqingzhong) (YWYD, 2: 424).
2 Li’s original vision of Maritime Defense was ambitious. He expected that the building
of China’s naval force, beginning with the creation of the northern fleet, would result
in three fleets to guard China’s northern, eastern, and southern waters. After trying
a few cheap, small-size gunboats (dubbed “mosquito boats”) only suitable for short-
distance coastal defense during the first few years after 1875, especially after Japan’s
annexation of Liuchiu (China’s long-time tributary) in 1879, counting on its arma-
ment with three ironclads, both Li and Shen were determined to purchase the most
advanced, high-tonnage ironclads from Europe in order to develop their fleets’ ocean-
going, combating capabilities (SBZ: 351). In Li’s view, China needed at least four such
ironclads, two in the north and two in the south, in order to outcompete Japan
(Zhang Limin 1982, 1: 21–24). With the improvement of funding for Maritime
Defense, Li did put his plan into action by ordering two ironclads from a German
shipbuilding enterprise (Aktien-Gesellschaft Vulcan Stettin) in 1880 and 1881, each
having more than 7,000 tons, which, however, did not arrive in China and join the
northern fleet until late 1885.
3 To attribute the failure merely to the Qing court’s insufficient funding was not com-
pletely convincing; a more immediate reason behind the defeat of the Northern Fleet,
as pointed out by the court, had to do with the incompetence of its commanders and
their poor performance during the war and with its lack of sufficient and proper train-
ing before the war (Fan Baichuan 2003: 1114–1128). However, in the face of
a rapidly expanding naval force in Japan, the Qing court’s failure to keep increasing its
budget for the naval forces and the subsequent standstill in the Northern Fleet’s build-
ing of battle capacities was decisive in shaping Li Hongzhang’s strategy for dealing
with the enemy and the morale of the commanders when confronted with the Japanese
fleet. Li lacked confidence; he adhered to a defensive strategy for his fleet throughout
the war, and his goal was to avoid a battle with the enemy and preserve his forces.
4 Liuchiu was spelt, following Chinese pronunciation, as Loo-choo or Lew Chew in
English literature in the late nineteenth century, and more popularly as Liuchiu in the
first half of the twentieth century. It was only after World War II that its spelling in
English was changed to Ryukyu according to its pronunciation in Japanese following
the founding of the “United States Civil Administration of the Ryukyu Islands” in
December 1950. Throughout this book, the islands are spelt as Liuchiu to respect the
historical reality.
6 A nation-state in the making
Manchu–Han relations under the
New Policies

The Qing encountered unprecedented geopolitical and fiscal crises after 1894.
Its defeat in the Sino-Japanese War resulted in the cession of Taiwan to Japan,
the loss of Korea as its most important tributary, and the subsequent scramble
for concessions and spheres of influence by foreign powers in the late 1890s;
China’s image as a modernizing power in East Asia was gone. No less damaging
to the Qing court was the huge burden of war indemnity to Japan, which
totaled 231.5 million taels, or more than ten times the Qing government’s total
spending on the Northern Fleet during the preceding decade. Six years later,
the raging of the Boxer Rebellion in North China resulted in the joint occupation
by eight foreign powers of Beijing and the Qing court’s payment of war indem-
nities that would total 982 million taels (including principal and interest at the
rate of 4% per year).1 These calamities greatly weakened the Qing state, both fis-
cally and politically. Given the fact that the Qing collapsed in 1911, only ten
years after the signing of the Boxer Protocol, it is not surprising that the literature
on the last decade of the dynasty has tended to interpret the events after 1900 as
a series of failures leading to the eventual demise of the Qing.
Yet, the post-Boxer decade also turned out to be the most critical moment in
China’s transition to a nation-state. Despite the crushing burden of war repar-
ations, the Qing state embarked on the audacious program of “New Policies”
(xinzheng, or “new administrative measures”), which involved the transform-
ation of the state’s infrastructural institutions, ranging from defense and public
security to education, jurisdiction, revenue generation and expenditure, and eco-
nomic planning; all these entailed massive spending by the governments at both
national and local levels and a huge amount of investment by private parties.
Nevertheless, solid progress was made during this decade in revamping the pol-
itical, economic, legal, and educational systems, which paved the way for
China’s further development in the rest of the twentieth century.
Even more surprising was the fact that, although the Qing dynasty collapsed,
China’s vast territory remained largely intact and continued into the Republican
era. The country faced three possible scenarios before and after the fall of the
Qing. The worst was of course its splitting by foreign powers and their estab-
lishment of colonial governments in the divided land to replace the Qing,
a possibility that did exist during the Eight Powers’ occupation of Beijing as
 A nation-state in the making 133
well as the northern and northeastern provinces in 1900 and 1901. Less disas-
trous would be the separation of the areas inhabited by the Manchus, Mongols,
Tibetans, Uighurs, and other Muslim groups from the interior provinces and
the return of the country to a land of the Han Chinese, which was indeed the
goal of some of the anti-Manchu revolutionaries on the eve of the Revolution
of 1911. A third possibility, which was least disruptive and also appeared to be
in the best interests of China from the viewpoint of the majority of Chinese
elites, was the survival and maintenance of China as a multiethnic state. China
in fact achieved the best result after the fall of the Qing: it survived the peril of
foreign invasion, maintained territorial integrity, and retained the framework of
the Chinese state as a political entity of multiple ethnicities. The abrupt demise
of the Qing in 1911 reflected more the failure of the Manchu elites in monop-
olizing their control of the central government than the efforts in rebuilding the
infrastructure of the state during the prior decade.

Modernization and fiscal expansion

Financing the Xinzheng program

Right after the signing of the Boxer Protocol in 1901, the Qing court issued
a series of decrees to implement the New Policies, including:

• The establishment of a centrally controlled and standardized New Army,

with the goal of creating 36 divisions, each having 12,500 men, to replace
the traditional regular forces of the Eight Banners and Green Standards as
well as various irregular regional forces; a related task was the creation of
a police force, which was first tried out in the key cities of Zhili in 1902
and soon expanded to other parts of the country.
• The reorganization of the government system, including the establishment
of consultative and legislative bodies at the national and provincial levels in
1909 and 1910, the restructuring of the central government from the trad-
itional six boards to eleven ministries, and, most important, the 1908
announcement of creating a constitution and a congress within nine years.
• The modernization of the financial and fiscal system, beginning with the
founding of a central bank in 1904 (The Bank of the Board of Revenue,
renamed the Bank of China in 1911), a commitment made in 1910 to cre-
ating a standardized currency, and the introduction of a fiscal budget of
government revenues and spending in 1911.
• The promotion and protection of private investments in industry, com-
merce, and railroad construction through the promulgation of a series of
laws, most importantly the Corporate Law of 1904, and the wide establish-
ment of Chambers of Commerce in towns and cities.
• The introduction of new-style schools at different levels to offer Western-
ized curricula in place of the traditional education under private tutoring
schools that solely focused on the rote learning of Confucian texts.
134 The transition to a sovereign state
Beginning with the promulgation of an imperial regulation of the new
school system in 1902, the reform culminated in the abolition of the civil
service examination system in 1905.
• The reform of the legal system, including the promulgation of laws on civil
rights and criminal and civil litigation, and most importantly a penal code
that replaced the traditional set of punishments with a new one borrowed
from the West, as well as reforms pertaining to jury, attorneys, and jails.

These reforms were indeed a very ambitious and comprehensive package to

modernize the state and the entire country, far exceeding the scope of the pro-
posals and policies made during the Hundred-Day Reform of 1898. Despite the
imperial court’s hesitation in implementing a constitutional government and the
interruption of the movement by the Revolution of 1911, the xinzheng program
succeeded in establishing the framework of a modern legal and justice system,
a modern educational system, modern military institutions, as well as legislative
organs and self-government bodies at local levels; all these would continue in
modified forms after the fall of the Qing and have a profound impact on state-
building in twentieth-century China.
The central government was responsible for funding the newly created mili-
tary forces, which were the most costly among all xinzheng projects; the planned
36 divisions and additional subdivisions would have required an annual budget
of 120 million taels, or roughly 3 million for each division (Deng Shaohui
1998: 211). Local governments’ burden was even more cumbersome; it was
estimated that each province would have to spend 2 to 3 million taels on main-
taining a police force, 1 million on the new justice system, and another million
on the new school system. A more serious budget on the preparation of
a constitutional government in nine provinces showed that the total spending
on education, jurisdiction, policing, self-government, and industrial growth
during the planned period of 1910–1916 would be 418.67 million taels (Zhou
Yumin 2000: 399–400); if the other nine provinces had similar budgets, then
the expenses on the xinzheng projects in the entire country would mount to
837 million taels.
Given the Qing government’s huge burden of war reparation in the post-
Boxer years, the xinzheng program seemed to be a mission impossible, because
the total cost of the program would be several times the principal of war indem-
nities to be paid to the Eight Powers. Nevertheless, this program proceeded
steadily during the post-Boxer decade. Six divisions of the New Army were first
created in 1904 under the commandership of Yuan Shikai, who had succeeded
Li Hongzhang as the Governor General of Zhili and the Northern Commis-
sioner for Trade since 1901. Stationed separately in the northern provinces
around Beijing, it was the largest (more than 70,000 men) and best equipped
among the many divisions of the New Army in different provinces. By the end
of the Qing, the New Army had about 161,800 men, consisting mainly of 14
divisions (zhen) and 18 additional “mixed sub-divisions” (hunchengxie) (Deng
Shaohui 1998: 213). The educational reform was also fruitful. New-style schools
 A nation-state in the making 135
increased from 769 in 1903 to 4,476 in 1904, 8,277 in 1905, 23,862 in 1906,
and 59,117 in 1909; students enrolled in these schools increased from 6,912 in
1902 to 1.6 million in 1909 (Shang Bing 2007: 147). Equally impressive was
industrial growth, which accelerated after the promulgation of the Corporate
Law and the wide establishment of Chambers of Commerce in 1904. In com-
parison to the nine years from 1895 to 1903 when less than 23 enterprises were
created each year on average, with a total investment of 4.14 million taels
per year, the eight years from 1904 to 1911 saw the founding of more than 74
companies with a total investment of 16 million silver yuan each year. New
investments in Chinese industrial enterprises totaled more than 138 million
silver yuan during the post-Boxer decade (1902–1911). In addition, Chinese
investors pooled 87.6 million silver yuan in total to fund railroad construction
during the 1905–1911 period (Wang Jingyu 2000: 79). Altogether, progress in
the realms of education, jurisdiction, finance, police, and the military contrib-
uted to the renewal of the infrastructure of the Chinese state, although their
effects on state-building were less visible than the transformation of its super-
structure or the government framework at the top level.
As a result, the Qing state’s spending in the post-Boxer decade skyrocketed,
from roughly 80 million taels before the Sino-Japanese War of 1894, to nearly
135 million taels in 1903, 237 million in 1908 and 270 million in 1909. By
1911, the last year of the Qing, its annual spending had reached 338 million
taels, or more than four times its 1894 level (see Table 4.1). As the Qing gov-
ernment’s 1911 budget suggested, most spending was on the military and
police (27.16%), war indemnities (16.79%), transportation (15.36%), and gov-
ernment administration (8.48%); the rest was on finance (5.82%), administration
of justice (2.15%), civil affairs (1.43%), education (0.83%), industry (0.52%), and
other items such as foreign affairs, custom administration, construction, public
debts, and border defense (QCXW: 8245).
Behind the dramatic escalation in the Qing state’s spending was the expan-
sion of its revenues, which grew from 81 million taels in 1894 to nearly
105 million in 1903, 235 million in 1908, and 297 million in 1911. Among
the major sources of government revenues, the land tax, which generated 32
to 33 million taels annually during the decade prior to 1894, saw an increase
to more than 37 million taels in 1903 and 48 million taels in 1911. But its
relative importance declined, accounting for 88% of the government’s total
revenue in 1849, 40% in 1894, and only 16% in 1911 (Table 3.3). Adhering
to the old principle of light taxation, the Qing court refrained from increasing
the official collection rate of land taxes per se, despite the imposition of tem-
porary surcharges on taxpayers by local administrators during wartime or
postwar rehabilitation and despite the manipulation by tax collectors in con-
verting copper cashes into silver taels in tax payment. Therefore, the Qing
state’s official revenue from the land tax did not increase too much before
1900 (fluctuating around 30 million taels from the late eighteenth century)
and its increase in the last few years of the Qing was primarily a result of the
formalization of local revenues, as to be discussed shortly. In fact, because of
136 The transition to a sovereign state
the long-term decline in the price of silver in the last six decades of the Qing
and the subsequent rise of the price of agricultural crops by more than two-
fold, the actual income of the Qing government from land taxes, or more
exactly the purchasing power of its revenue from land taxes, dwindled (Xu
Daofu 1983: 89–90; Wang and Yan 2007: 36, 43–46). In other words, des-
pite the creation of various surcharges on the land tax in the late Qing
period, the actual tax burden of landowners did not increase as much as the
numbers in silver taels of the land taxes suggested.
Instead of maximizing its extraction of revenues directly from landowners, the
Qing court turned primarily to the following two methods to increase its rev-
enues. One was the imperial court’s efforts to investigate and consolidate the
collection and management of tax revenues at the provincial level, and thereby
to turn much of the unreported or underreported revenues from the land taxes
and lijin into the formal revenue of the state, subjecting them to its unified
budgeting and allocation. As mentioned in Chapter 4, the informal income of
governments at the provincial and lower levels that had never been reported to
the imperial court could be as much as 40% of the reported revenue. Most of
the increases in government revenues after 1908, therefore, was a result of the
Qing court’s investigation of local finances in the last four years of the dynasty
(QCXW: 8249).
More important in augmenting the Qing state’s revenue was its growing reli-
ance on the non-traditional, commercial sources. One such source was foreign
and public debts, which the central and local governments frequently borrowed
to make up deficits. From 1900 to 1911, for example, the Qing court borrowed
a total of 340 million taels as foreign debt (this was in addition to the debt for
paying war reparations), mostly (76.86%) for the construction of railroads (Xu
Yisheng 1962: 90). The biggest source of revenue, however, was commercial
taxes, including primarily salt tax, lijin, and maritime custom duties, which
altogether generated a total of 42.55 million taels (or 52% of the government’s
total revenue) in 1894 and 131.64 million (44%) in 1911. Among the three
sources, maritime customs was the largest, increasing steadily from 11 million
taels in 1871 to 22 million in 1890, 30 million in 1902, and 36 million by
1911, which reflected the expansion of the domestic market for imported goods
and of the size of China’s economy on the whole. Next to this was the revenue
from lijin, which contributed 13 to 17 million taels annually until 1908. The
third was salt taxes, generating 6 to 7 million taels a year in most of the nine-
teenth century but surging quickly to 13 million in 1900 and as high as
46 million by the end of the Qing, thus becoming the largest commercial
source, as a result of the Qing court’s centralization and consolidation of the
production and marketing of salt (see Table 4.1).
The revenue structure of the Qing state, therefore, finished a revolutionary
transition in the last decades of its history, from the traditional one that was
based primarily on agricultural sources (the land taxes) and remained largely
stagnant and fixed, to the new one that was based on commercial sources (taxes
on industrial and commercial goods as well as foreign and public debts) and
 A nation-state in the making 137
open to expansion, which made the late Qing government analogous to modern
nation-states in other parts of the world in terms of revenue composition.

The rise of high-level disequilibrium

As shown in the preceding chapter, the first two decades of the Guangxu reign
saw a rapid increase in the Qing state’s revenues and the reestablishment of
a low-level equilibrium in its fiscal constitution, in which its revenue was
slightly higher than its expenditures. The Qing government revenue continued
to grow, at a higher rate, after 1894, especially in the last decade of the Qing.
By 1911, its total revenue (297 million taels) had been three times its 1894
level (81 million taels), or a growth of 7.94% a year. But the expenditures of
the Qing state grew even faster, from 80 million taels in 1894 to more than
376 million by 1911, at 9.53% a year! The result was a widening deficit,
almost 13 million taels in 1899 (or 14% of the Qing state’s revenue),
30 million in 1903, 41 million in 1910, and 79 million (or 26% of its revenue)
in 1911 (see Table 4.1).
A more fundamental difference in the Qing state’s fiscal condition before and
after 1894 lay in the dynamics behind the changes in the fiscal structure.
During the first two Guangxu decades, growth in government revenues came
primarily from the rapid expansion of maritime custom duties; free of war
indemnities, the state was able to use the increased revenue to finance its pro-
jects in military modernization, despite all the constraints discussed in the last
chapter. In other words, it was the increases in government revenue that permit-
ted increases in government spending. In sharp contrast, after 1894, it was the
rapid increases in government spending, primarily the payment of war indem-
nities to foreign powers and the funding of the New Army and other xinzheng
projects, that drove the Qing state to enlarge the sources of government rev-
enues; but the increases in revenues lagged increasingly behind the expenditures,
hence a disequilibrium in the Qing state’s fiscal structure. The disequilibrium,
to be sure, had occurred during wartime, first in the 1670s and again during
the Taiping Rebellion, as an anomaly in the long history of the Qing state’s
finances since the 1680s. After 1900, however, it became a new norm of
China’s fiscal constitution in the decades to follow.
Did the recurrence and perpetuation of the disequilibrium mean a crisis in the
Qing fiscal system, as the fall of the dynasty in 1911 seemed to suggest? To
judge whether or not the Qing state’s extraction of resources from society
through taxation and other means of revenue generation was so excessive as to
hurt its legitimacy, we have to take into account two factors. One was the long-
term decline in the price of silver in the world market in the second half of the
nineteenth and early twentieth centuries and the subsequent increases in the
conversion rate between copper cashes, used by ordinary people in daily transac-
tions, and silver tael, used in collecting and paying taxes. During the last 17
years of the Qing (1894–1911), because of the depreciation of silver in the
world market and the increase in the value of copper cashes, the purchasing
138 The transition to a sovereign state
power of silver tael in China declined by 40.35% (Wang and Yan 2007:
272–273). The actual value of the 297 million taels of silver that the Qing state
generated in 1911 was only 2.18 times the value of the 81 million taels of silver
generated in 1894, rather than 3.66 times, as the face values of its revenues sug-
gest, and the annual growth rate of its actual revenue was only 4.69%, rather
than 7.94%, during the same period.
The other factor was the Qing state’s revenue in relation to the potential of
taxable resources. According to the estimate made in 1899 by Robert Hart,
Inspector-General of the Imperial Maritime Custom Service, for example, China
had at least 4 billion mu of arable land, which had the potential to generate at
least 400 million taels as taxes (ZGHG: 49), or more than ten times the land
taxes actually collected around 1899. Hart’s estimate, to be sure, was a wild
assumption, for the actual size of the arable land in China was only about
1.25 billion mu as late as 1952 and roughly 2 billion mu in the 1980s (Shi
Zhihong 2011). However, even if we take 1.2 billion mu as the actual size of
the arable land in late Qing China, there could have been an additional
amount of 24.9 million taels as land taxes to be generated, or 52% of the land
taxes actually collected in 1911.
We can also gauge the potential of revenue generation by looking at the size
of China’s economy. Historians have offered different estimates of China’s gross
national product in the late Qing period, ranging from 2 billion taels in 1800
(Liu Ruizhong 1987) and 3.3 billion taels in 1888 (Fairbank and Liu 1980: 2),
to 4.2 billion in 1894, 5.8 billion in 1903, and 6.9 billion in 1908 (Zhou
Zhichu 2002: 259). If these estimates are plausible, then the Qing government’s
official amount of tax revenues only accounted for 1.5% of China’s economic
output in 1800, 1.92% in 1894, 1.81% in 1903, and 3.4% in 1908. If we further
accept 6.9 billion taels as a conservative estimate of China’s economic output in
1911, then the state extracted only 4.3% of it as taxes. But the official records
of revenues were far below the actual amount of taxes and fees collected by
local governments and their agents. Liang Qichao, for instance, believed that
the taxes and fees collected by government officials and runners could be as
high as “three or four times” the official amount reported to the central govern-
ment and that the total tax burden of Chinese people could be “above four
hundred million taels” (YBS, 8: 26). Despite the rapid increase in government
revenues and the huge quantity of tax incomes unreported by local administra-
tors and pocketed by tax collectors, the real tax burden in the late Qing period
was not as exorbitant as the conventional wisdom has suggested. It accounted
for only 5.79% of China’s total economic output, if we accept Liang’s estimate
(400 million taels).
However, this burden had different meanings for people in different economic
sectors. For the rural population, the land tax remained relatively light, account-
ing for roughly 2–4% of agricultural output in the last few years of the Qing.2
For urban dwellers, especially those living on small trades, however, the newly
created taxes and fees, which quickly increased after 1894, especially during the
xinzheng period, could be truly extortionate. The Qing court’s 1911 budget
 A nation-state in the making 139
shows that the revenue from miscellaneous taxes had replaced maritime customs
to become the most important source of its revenue, accounting for as much as
40% of its total revenue, which contrasts sharply with only 12–14% before 1894
(see Table 4.1). It was small wonder that unrest took place primarily among the
people in non-agricultural sectors rather than the taxpaying cultivators during
the xinzheng years, especially in the last few years of the Qing (see, e.g., Praz-
niak 1999). But these riots were by no means fatal to the Qing dynasty; they
were on a small scale, limited to local areas, and were largely manageable to
local authorities. The increased tax burden alone cannot explain why the Qing
dynasty fell abruptly during the height of the xinzheng movement; to under-
stand the demise of the Qing, we need to further inquire how the xinzheng pro-
jects and the subsequent rise of high-level disequilibrium affected the
relationship between the Qing court and local authorities.

Recentralization and alienation

Recentralizing fiscal power

One of the central agendas of the xinzheng program was lixian, or creating
a constitutional government. However, people of different backgrounds
embraced it for different purposes. For its advocates among the reform-minded
intellectuals and the emerging bourgeoisie, imposing a constitution on the gov-
ernment was believed to the best way to make the latter accountable and to
have their own voice heard in the policy-making process. For provincial bureau-
crats, especially those of Han origin, the constitutional government appeared to
be a better choice that hopefully worked to curb the power of the inexperi-
enced, capricious Manchu nobles, who came to dominate the court after the
death of Empress Dowager Cixi in 1908. The Manchu elites also welcomed the
proposal of a constitutional government, anticipating their enhanced control of
power under a constitutional government. Tieliang (1863–1939), Grand Coun-
cilor of State, stated unequivocally in 1906, right after the announcement of
“preparation for a constitutional government” (yubei lixian): “government
power has to be centralized before the establishment of a constitutional govern-
ment; and to centralize government power, the military and fiscal power of gov-
ernors and governor-generals have to be taken away and returned to the central
government” (cf. Hou Yijie 1993: 79).
A key step in fiscal centralization was the Qing court’s decision to “investigate
and consolidate the fiscal system” (qingli caizheng) in 1908. It thus sent out
two fiscal superintendents to each province, whose duties were to oversee the
operation of the newly established Fiscal Clearance and Consolidation Bureau in
the province and its compilation of a detailed report on the revenues and
expenses of provincial government institutions, especially the portion that had
never been reported to the central government (Su Quanyou 2010). Provincial
governors and governors general were held responsible for submitting their
budgets to the central government within a time limit to avoid the penalty of
140 The transition to a sovereign state
demotion and salary suspension for up to a year. Meanwhile, all of the irregular
fiscal bureaus and offices originally set up by provincial authorities had to be
abolished, and the Financial Commissioner’s Department had to be completely
overhauled or replaced with a new organ called the Financial Department (caiz-
heng ju) or the Financial Institute (caizheng gongsuo). And the provincial gover-
nors and governors general were prohibited from borrowing foreign loans and
issuing paper currency on their own (Liu Zenghe 2014; Deng Shaohui 1998:
264–269). Another important step in recentralizing fiscal power to the imperial
court was to revamp the administrative system in salt production and marketing.
The Division for Supervising Salt Administration was set up and headed by
a Manchu prince in 1909 to curb the rampant practices of smuggling illegal
salt, embezzling licensing fees, and taking bribes from salt merchants; and the
most important measure in this regard was taking away the power of personnel
appointment and income management from provincial authorities and putting
the salt administrative system under its direct control.
These measures produced some obvious effects. The amount of lijin reported
to the court increased dramatically from 12–13 million taels a year in the early
1900s to 21 million in 1908 and more than 43 million in 1911, as reflected in
the Fiscal Department’s 1911 budget. Even more surprising was the increase in
salt taxes, from around 13 million taels around 1900 to more than 46 million
taels in 1911, which surpassed lijin and maritime customs to become
the second largest source of government revenue, next to the land taxes (see
Table 4.1). Provincial governors and governors general resisted the financial
investigation and consolidation program from the beginning; they refused to
“disclose the whole” (hepantuochu) of unreported revenues to the central gov-
ernment, give up their power in managing salt taxation, and include the spend-
ing on local public welfare programs into their administrative expenses; they
were particularly resentful of the preliminary budget for 1910 that allocated
most of the administrative expenditures to the individual departments of the
central government while cutting down the budgeted expenses of local govern-
ments (Liu Wei 2003: 376). After the 1911 budget plan was finalized and pub-
licized, they ignored it. The Financial Superintendents complained in early 1911
that “the individual provinces remain as wasteful in spending as before, and they
discard the budget plan altogether; savings of funds have been barely heard
of” (DQXFL, 11: 8). The fiscal reform, in a word, succeeded in uncovering
much of the hidden revenues from provincial administrations, but failed in
centralizing the actual management and spending of the public funds from
provincial administrators.

Recentralizing military power

Even more challenging to the Qing court was centralizing its military power
through the establishment of the New Army. The official reason for creating the
New Army right after the Boxer Rebellion, when the state was burdened with
a huge amount of war reparation, was to protect the capital city against potential
 A nation-state in the making 141
unrest from surrounding provinces where the rebellion had just subsided and to
safeguard the interior provinces against the impending outbreak of the Russo-
Japanese war in Manchuria. As an initial step, the Division of Military Training
(lianbingchu) was established in October 1903 to establish 36 divisions of
450,000 soldiers under its unified command. Conflicts between the imperial
court and provincial administrators became inevitable when the latter, especially
the prominent governors general of the southern provinces such as Zhang Zhi-
dong of Hubei and Wei Guangtao of Jiangsu, resisted the court’s demand for
more funds to support the New Army and openly opposed the unification and
standardization of the military units that would threaten their control of local
forces; but they soon gave up under pressure from the court. In sharp contrast
with these provincial administrators as losers in the centralization of military
power, Yuan Shikai, Governor General of Zhili and the Minister-Superintendent
of Trade in the North, emerged as the biggest winner. Trusted by Empress
Dowager Cixi for his proven loyalty that led to Emperor Guangxu’s failure in
the Hundred-Day Reform back in 1898, Yuan was appointed as Assistant Minis-
ter of the Division of Military Training. With Prince Yikuang (1838–1919) as
the head of the division, who was widely believed to be ignorant of military
affairs, Yuan dominated the planning of the New Army. Most of the funds con-
tributed by individual provinces for this purpose thus was spent on the six divi-
sions (70,000 soldiers altogether) of the New Army created by Yuan himself
and stationed in the key localities of Zhili and Shandong provinces. These divi-
sions became the best equipped and best trained in the country. Yuan tightly
controlled the six divisions through his selection and appointment of high-
ranking officers from his own subordinates, who owed their personal loyalty to
him for his long-time patronage.
Yuan, however, did not limit his influence to the military; he dominated the
planning of the bureaucratic reform (guanzhi gaige) as a key member of a board
(guanzhi bianzhi ju) appointed in 1906 to oversee the reform. He was particu-
larly interested in the idea of a “responsibility cabinet” (zeren neige) and gener-
ally supportive to the ongoing constitutional movement, partly to show his
alleged commitment to political progress in China, but more importantly to
safeguard himself against the potential revenge from the Guangxu emperor after
the death of the Empress Dowager; to limit the ruler’s power with a constitution
and an independent cabinet, in other words, was the best way to protect himself.
But Yuan’s unparalleled influence in the military and his unveiled political ambi-
tion also made the Empress Dowager increasingly cautious of him. To appoint
him as a Grand Councilor of State and thereby remove him from the position of
Governor General of Zhili in September 1907 thus was at once a promotion for
him and a weakening of his power base.
The death of Empress Dowager Cixi in November 1908 signaled a turning
point for both Yuan’s political career and the Qing dynasty. While Cixi was
alive, a bond existed between the imperial court and senior Han bureaucrats; for
decades, the Empress Dowager had trusted the latter with the most important
tasks in defense and counted on them to make key decisions that helped the
142 The transition to a sovereign state
Qing survive the recurrent crises, and the Han bureaucrats remained loyal to the
imperial court as long as they received proper treatment and respect from it.
Added to this bond was the existence of a small group of Manchu elites such as
Ronglu and Duanfang who kept a close relationship with the Han bureaucrats
and mediated between the Manchu and Han elites. The death of the Empress
Dowager coincided with the passing of this generation of experienced and
trusted Han bureaucrats, most prominently Li Hongzhang who died in 1901,
Liu Kunyi in 1902, Wang Wenshao in 1908, and Zhang Zhidong in 1909.
After 1908, the new generation of Manchu princes who dominated the court
had isolated themselves to the circle of the Manchu elite itself, had no personal
contact with Han bureaucrats at provincial and central levels, and lacked suffi-
cient experience in managing state affairs. They did not readily trust the latter
and in fact had no sense of security when dealing with them, especially with
powerful Han leaders such as Yuan Shikai, who appeared to be self-assured and
aggressive, lacking the prudence and integrity found in the old generation of
Han bureaucrats. Therefore, the old divide between the Manchus and the Han,
which had been taboo in the public discourse of the ruling elites, came back to
shape their consciousness when faced with the initiatives of political reforms,
which the Manchu nobles readily interpreted as steps to limit their privilege and
aggrandize the Han people.
Not surprisingly, a major step taken by the Manchu princes in enhancing
their power was ousting Yuan from the court on the excuse of his need to
recover from a foot illness in January 1909. But Yuan never lost his control of
the six divisions of the New Army through his trusted subordinates, who
remained loyal to him. The death of Cixi and his subsequent expulsion only
made Yuan no longer obligated to support the Qing court; the conditional loy-
alty that the Han bureaucrats had maintained since the 1850s eventually came
to an end for Yuan and his followers. To reappoint Yuan to the key positions of
the government and the military after the outbreak of the Uprising of Wuchang
in October 1911 merely offered him the power to bargain with the revolution-
aries regarding the terms of his role in the abdication of the Qing ruler and the
newly born Republic. The individual governors and governors general of the dif-
ferent provinces, who had been discontent with the Qing court’s recurrent and
heightened measures to limit their power and extract local revenues, saw no
reason to remain loyal to the court when the Revolution swept their provinces
without much violence involved.

The gentry elite

A generational transition was also visible in the composition of the gentry elites
and their relations with the Qing state in the 1900s. Before 1900, the gentry
class, consisting mainly of graduates from civil service examinations at county
through metropolitan levels but also including a small portion of purchased-
degree holders, served as the very foundation of the imperial state (Chang
1955). Trained in Confucian classics, the gentry class was the primary source
 A nation-state in the making 143
from which the state recruited government officials. As informal leaders of local
communities, the gentry members who did not enter the government or retired
from officialdom also acted as intermediaries between the government and soci-
ety. The social and political standing of the gentry class, therefore, was directly
linked with the state’s sponsorship of the examination and recognition of their
privileges as degree holders. The government, for its part, was able to keep itself
to a minimal size and yet effectively ruled its people through the gentry elites
without having to enlarge the formal bureaucracy and penetrate society.
All this came to an end after the Qing court abolished the civil service exam-
ination in 1905 as part of the xinzheng package. This it did for a good reason:
as long as the examination continued, parents would be reluctant to send their
children to new-style schools for education in modern knowledge. But terminat-
ing the examination also meant shutting off the source that generated the trad-
itional gentry class and severing its ties with the state. Students therefore had to
either receive new-style education, mostly in the mushrooming “schools of law
and administration” (fazheng xuetang) or military schools in the cities, or travel
to Japan and other countries for higher education. As graduates from the new
schools or returned students, their standing in society and career development
no longer depended upon the Qing state’s recognition; therefore, they no
longer owed their chief loyalty to the imperial court. Many of them shifted their
interest from serving the government to investing in commerce and industry,
especially in modern businesses such as railroad construction and transportation
or textile manufacturing. Known as gentry-merchants (shenshang), they consti-
tuted the main body of the reformist elite and came to dominate urban society
(Esherick 1976: 66–105; Bergère 1986: 37–62). Those who graduated from
military schools at home or abroad would later became the backbones of the
New Army. Together, these civilian and military elites came to dominate polit-
ical transformation in the last decade of the Qing and early Republican years.
Thanks to the introduction of the self-government program at the local level
as part of the xinzheng program, the gentry-merchants organized themselves
into the chambers of commerce (shanghui) that were bourgeoning in cities and
towns. The most active of the gentry elites were elected into the local Delibera-
tive Councils (yishihui) and Board of Directors (dongshihui), which had been
established in each county as self-government organs to collaborate with county
magistrates in managing public affairs, especially in determining the kinds and
rates of surcharges and the fees for local xinzheng projects; they also headed the
newly established schools or other self-government organs such as the School-
Promotion Office (quanxuesuo), the Police Office (xunjinshu), or the Financial
Clearing Office (qingli caizheng chu). At the provincial and national levels,
a consultative assembly (ziyiju in each province and zizhengyuan in Beijing) was
established in 1909 and 1910, which had the right to make policy recommenda-
tions and the power to approve or veto government budget plans. All these
newly created institutions offered a channel for the gentry elites to formalize
their leading role in society and their influence upon the government. Conflicts
between the gentry-controlled deliberative councils or assemblies on the one
144 The transition to a sovereign state
hand and government leaders on the other took place at each level, when the
deliberative organs questioned the administrators’ budget or tax plans, or when
the administrators rejected the assembly’s proposal; frequently the elected repre-
sentatives in those organs accused the administrators of being conservatives
resisting reforms while the latter condemned the assemblies as unruly elements
infringing on government prerogatives (Rankin 1986: 202–299; Liu Wei 2003:
186–187; H. Li 2005: 196–197).
Despite such recurrent and sometimes bitterly vented schisms, however, the
two sides more often than not found grounds to collaborate in dealing with
the imperial court. This was especially true at the provincial level, where the
governors or governors general often took the initiative in promoting self-
government programs within their provinces. For instance, the provincial
administrators were generally sympathetic to the “rights recovery” (shouhui
liquan) movement in which local gentry-merchants raised funds on their own
to buy back the rights of mine production and railroad construction from for-
eign investors (X. Zheng 2018). The governors were particularly interested in
promoting the construction of railroads, believing it to be the most effective
way to boost the local economy and government revenues.3 Therefore, when
the Qing court tried to take over the construction of two major railroads in
South and Southwest China by signing a loan contract with European debtors
in April 1909, the governors of concerned provinces joined gentry-merchants
to oppose this move. The Qing court’s decision in May 1911 to “nationalize”
railroads in the country only caused gentry-merchants to agitate widely and
confront the central government. Some governors asked to resign after they
failed to appeal to the imperial court or found that their fiscal and military
power was greatly curtailed. It was no wonder that, when the Uprising of
Wuchang broke out and the Consultative Assembly of Hubei declared the
province’s independence from the court, governors in other provinces either
joined this action or adopted a wait-and-see attitude, or simply abandoned
their job and fled; few remained loyal to the Qing.
To conclude, the end result of the Qing court’s measures of fiscal and mili-
tary recentralization in its last years was its alienation of two key groups,
namely, the gentry elites that played a leading role in local communities and
the bureaucrats of the Han who ran the government in most provinces.
Having served as the most important social basis of imperial states in Chinese
history since the Tang, the gentry class eventually severed its ties with the
government during the last decade of the Qing. The most fundamental chal-
lenge to the Qing dynasty, however, came from the Han bureaucrats at the
provincial level. Their loyalty to the imperial court hinged on (1) the court’s
respect of their interests, especially their control of local financial resources
and hence the survival of xinzheng projects they founded and maintained, and
(2) the Qing court’s proper treatment of them in line with the Confucian
tenets on the relationship between the ruler and ministers. What undermined
this relationship in the last decades of the Qing were precisely (1) the imper-
ial court’s infringement upon their interests through the incessant efforts of
 A nation-state in the making 145
recentralization that aimed to deprive the provincial administrators of their
fiscal and military power, and (2) the loss of mutual trust between the two
sides, and the subsequent worsening of the Manchu–Han divide in their pol-
itical thinking and action.
Nevertheless, the fall of the Qing in 1911 should not be equated with the
failure of state-rebuilding in nineteenth and early twentieth-century China.
Many of the xinzheng programs implemented in the last decade of the dynasty
continued after 1911 and spearheaded China’s modernization on almost every
front. Most importantly, the collapse of the Qing did not result in the splitting
of its territory that the early Qing ruler had put together through military con-
quest and consolidation: the Republic after 1911 continued to be a multiethnic
state with its territorial boundary largely identical to the Qing’s. Despite the
popularity of the slogan of “expelling the barbarians and restoring a Chinese
state” (quchu dalu huifu Zhonghua) among the anti-Qing activists before and
during the Revolution of 1911, the fall of the dynasty did not lead to the cre-
ation of a Han Chinese state as the rebels had intended or the separation of the
frontiers of the former Qing from the Republic. We thus need to further explore
why the Republic was able to inherit the frontiers from the Qing.

The making of a new nation

Homogenizing the Manchus and the Han

For mainstream historians in post-1949 China, the territorial and administrative
continuities between the Qing and the Republic were a given. The Revolution
of 1911, therefore, was interpreted as a revolution by an emerging Chinese
bourgeois class against the autocratic imperial regime (e.g., Hu Sheng 1981).
The anti-Manchu mobilization and the tensions between Han revolutionaries
and the elites of non-Han populations under the Qing, which had been central
to the narrative of the Revolution in the nationalist historiography before 1949,
were obscured. In the writings on modern China in the West, preoccupation
with the thesis of modernization vis-à-vis cultural tradition in the 1950s and
1960s also led scholars to dismiss ethnic tensions between Manchus and Han in
the late Qing period as insignificant or irrelevant. Based on an 1865 edict of the
Qing court that allowed the Manchus to exit the banner system, make their
own living, and be treated the same way as the Han people in litigation, Mary
Wright concludes that “[m]ost of the last restrictions separating the Manchus
from the [Han] Chinese were removed” thereafter (Wright 1957: 53–54) and
that “[t]he Sino-Manchu amalgam that had been developing since the eight-
eenth century had reached its full maturity in the mid-nineteenth century”
(Wright 1968: 21). Indeed, by the last decade of the Qing, the Manchus had
been Sinicized to such a degree that, according to a contemporary, “all of the
Manchus speak the language of the Han people, and there is not a single
Manchu who can speak the Manchu language” and “those of the Manchus who
are literate are only able to read the Chinese rather than the Manchu texts”
146 The transition to a sovereign state
(YDJ, 1: 273); the early Qing rulers’ efforts to encourage the use of the
Manchu language among the imperial elites of Manchu origins failed.
More recent historiography has questioned this proposition. Instead of the
melting of Manchu–Han cleavages, Pamela Crossley argues that the rise of
a racial consciousness among the Manchus was partly in response to the hostility
of the Han people during the Taiping Rebellion and partly a result of “court
sponsored ideals of Manchu identity which had made racial thinking not only
acceptable but required” (1990: 228). This “sharpening” of ethnic conscious-
ness among the Manchus, Mongols, and other non-Han peoples, together with
inspiration from Chinese nationalism, she suggests, gave rise to the secessionist
movements in Manchuria, Mongolia, Xinjiang, and Tibet in the early twentieth
century (Crossley 2002: 343–344). Edward Rhoads, too, challenges Wright by
demonstrating the ineffectualness of the 1865 edict, the continuation of institu-
tions and practices that segregated the Manchus from the Han, and gross
inequality between the two populations, as seen in the prohibition against inter-
marriage between them, the quotas of government positions for the Manchus,
the banner people as a paralytic population living entirely on government sti-
pends and in isolation from Han society, the severer punishment of the Han
than the Manchus for similar offenses, and the persistence of the Eight Banners
as a garrison force stationed in cities throughout the country to supervise the
Han people despite its declined role in national defense (Rhoads 2000:
35–63).4 Mark Elliott offers a more nuanced analysis of ethnic relations in late
Qing China. He underscores the success of the Qing in maintaining what he
calls the “ethnic sovereignty” of the Manchus, or the special professional, legal,
and residential arrangements of the banner people as a conquest force, through
the Eight Banners system that remained in place until the early twentieth cen-
tury, although he does not deny the fading of most of the cultural and linguistic
traits that distinguished the Manchus from the Han by the nineteenth century.
More importantly, while he cautions against equating the “Great Qing” with
Zhongguo or a historical China, given the contradictions associated with the
making of the Qing and a modern Chinese state, Elliott nevertheless identifies
a historical link between the Qing court’s efforts to unify the “inner” (China
proper) and the “outer” (Inner Asian frontiers) territorially and the redefinition
of the modern Chinese nation in the twentieth century, hence a paradox in
which the centuries of Manchu rule that was “so un-Chinese” yielded a territorial
and ethnic legacy that accounts for the myth of a “unified China” in the modern
era (Elliott 2001: 359–361).
At any rate, ethnic and institutional divides continued to shape the mutual
perceptions and animosity between the Manchu and Han peoples, which indeed
worsened at the beginning of the twentieth century as an increasing number of
revolutionaries built their strength by agitating the Han people with the slogan
of “expelling the Manchus” (pai Man). Both the Han and Manchu elites real-
ized the severity of the Manchu–Han divide and its threat to the legitimacy of
the Qing state at a time when Chinese society, especially the Han elites, was
exposed to new ideas of popular rights and nationalism imported from the
 A nation-state in the making 147
West and Japan. Zhang Zhidong, the Governor General of Huguang, thus
noted in 1901 that “ironing out the Manchu–Han differences” (hua Man Han
zhenyu) was “an idea of utmost importance” (shangshang zuiyao zhi yi) among
the Qing state’s many political proposals of post-Boxer rehabilitation and was
key to its survival and the elimination of revolutionary rebels (ZWXG, 175:
14). Duanfang, the acting Governor General of Liangjiang, too, saw the
Manchu–Han cleavage as the “the biggest barrier to the xinzheng of China
and the severest danger to the future of the imperial dynasty” (XHGM, 4:
39). Both governors, therefore, proposed the equal treatment of the Man-
chus and the Han and the elimination of Manchu privileges as a fundamental
solution to the crisis of the Qing and the precondition for successful imple-
mentation of xinzheng programs.
The Qing court acted accordingly. Beginning in 1901, the imperial court no
longer applied the rule of dyarchy between the Manchus and the Han to the
four newly created ministries (of foreign affairs, commerce, police, and educa-
tion), a practice that was adopted in 1906 during the “reform of bureaucracy”
(guanzhi gaige) for other government offices. At the same time, the court also
allowed Han bureaucrats to fill positions that had been monopolized by the
Manchus, such as the positions of jiangjun (the supreme military commander
and civil administrator of a large border area) and dutong (the supreme military
commander and civil administrator of each of the Eight Banners). In 1907, the
court further turned Manchuria into three provinces, replacing the original
jiangjun positions with the new positions of a single governor general and three
provincial governors, to be filled by candidates regardless of their Manchu or
Han origins. In fact, by the very end of the Qing, of the nine governors general,
four were Han, two were banner men of Han origin, and only three were Man-
chus; and of the fourteen provincial governors, only two were Manchus
(QMLX, 1: 559–565).
The Qing court also took steps to narrow the gap between ordinary Manchu
and Han people. In early 1902, it revoked the ban on intermarriage between
them. In 1908, it prescribed the same punishment for the Manchus and the
Han for the same offense. The criminal code of 1909 further prescribed the
adjudication of cases involving the Manchu or Han people by the same organ
(i.e., the newly established court system at the local level that was independent
of county and prefectural administrators). Most importantly, in 1906, the Qing
court allowed banner people to seek their own vocations and to purchase real
estate under their own names; next year, it further ordered them to be governed
by local administrative authorities and for banner soldiers to be demobilized
with a ten-year stipend plan and to make a living on their own afterwards (Chi
Yunfei 2001).
But the effects of these measures on eliminating the differences between the
two populations were limited. Intermarriage between ordinary Manchu and Han
residents remained rare in the following years because of their different customs.
The equal treatments of the two peoples in adjudication and civil administration
would undoubtedly have helped integrate the Manchus or the entire banner
148 The transition to a sovereign state
people into a society dominated by the Han, had the Qing stayed longer to
allow enough time for these new policies to take effect. Unfortunately, all these
developments were curtailed by developments in court politics after 1908 and
interrupted by the outbreak of the Revolution of 1911. To concentrate the
fiscal and military power of the government into the hands of the few Manchu
nobles only aroused the ill-feelings of the Han bureaucrats against them, alien-
ated the gentry-merchant class who had harbored the hope of a constitutional
government, and offered the anti-Qing revolutionaries a new reason to agitate
the ordinary Han people.
We can thus distinguish two parallel trends in Manchu–Han relations in the
late Qing period, which competed with each other during the last decade of the
Qing: one was the long-term trend of incorporating the Manchus into Chinese
society and the rise of the Han bureaucrats at the provincial and central levels;
the other was the Manchu rulers’ recurrent efforts to maintain the ethnic char-
acteristics of the Manchu people and safeguard the privileges of Manchu elites.
By and large, however, the first trend prevailed, resulting in the systematic inte-
gration of the former into Han society before and during the xinzheng decade;
the Manchu elites’ attempts to reassert their prerogatives should be seen as
a temporary and last stand against the trend of integration. In the final analysis,
the homogenization of the Manchus and the Han in language and culture, the
dependence of the Manchu elites on Han bureaucrats in government, the insep-
arability of the Manchus from Han Chinese society, and, no less important, the
integration of Manchuria with interior provinces, altogether these developments
contributed to the growth of a shared nationhood between the Manchu and
Han populations.

Imagining “the Chinese nation”

To understand why the Republic of China was able to inherit the entirety of the
Qing state’s territory upon its inauguration in 1912, we need to look further at
how the revolutionaries who founded the Republic, as well as their enemies,
defined and redefined China as a new nation that they wanted to create before
and during the Revolution of 1911.
To begin with, it is important to note the growing awareness among the
ruling elites of the Qing, including both the Manchus and the Han, of China or
Zhongguo as a country that was more extensive in scope and carried greater
weight than the Qing state itself. While they generally referred to the Qing state
as Da Qing (“the Great Qing”), Huang Qing (“the Imperial Qing”), or Huang
Chao (“the Imperial Dynasty”) in official discourse, especially in discussions
concerning the relationship between the Qing and preceding Chinese dynasties
or on matters between the imperial court and its subjects, the Qing rulers and
bureaucrats switched to the word Zhongguo when discussing issues pertaining
to the entire country. The word Zhongguo, to be sure, had been used com-
monly in the seventeenth and eighteenth centuries by the Qing elites in discus-
sion concerning China’s relations with neighboring tributary states and other
 A nation-state in the making 149
countries. What was new in the discourse of the late Qing period, especially
during the xinzheng decade, was that Zhongguo was increasingly used also in
discussions on domestic issues arising from China’s contact with the outside.
They frequently juxtaposed Zhongguo with waiguo or foreign countries when
describing domestic changes resulting from foreign influences and when ration-
alizing a proposal or a new policy that departed from traditional practices and
aimed to catch up with foreign countries. This “Zhong-wai” (Zhongguo versus
foreign countries) dyad permeated and indeed undergirded their discourse on
reforms and new projects during the Self-Strengthening movement, the Hun-
dred-Day Reform, and the xinzheng movement. To embark on such new
reforms and modernizing projects, in other words, was important not only for
the viability of the Qing dynasty alone but also for the well-being of the entire
country that was Zhongguo. This shared awareness among the Manchu and
Han elites of belonging to one country grew in the second half of the nine-
teenth and early twentieth centuries, when they were confronted with the fatal
challenges coming from outside China and were committed to updating its own
military and civil institutions for the survival of the country; it increasingly
superseded the old Manchu–Han dyad to become the most decisive factor
redefining their self-identity and political consciousness, allowing them to build
consensus on major issues concerning China’s survival and progress, regardless
of their Han or Manchu origins, and to transcend the narrow notion of
a Manchu–Han rift that tended to divide them.
Not surprisingly, those who were most enthusiastic for the Self-Strengthening
projects in the late nineteenth century or for the constitutional movement in the
early twentieth century were not limited to the reform-minded Han bureaucrats;
the enlightened Manchu elites were equally active and even playing a more
important role in advancing the modernizing projects. One such Manchu nobles
was Ruicheng (1863–1915), the Governor General of Huguang, who joined
Xiliang (1853–1917), the Governor General of the three Manchurian provinces,
in October 1910 to lead the third petition by 19 provincial governors and gov-
ernors general for the early opening of congress, forcing the court to announce
its commitment to opening it in 1913. A contemporary thus observed that “not
all Manchus are evil, for there is no lack of those who have revolutionary ideas
and seek social progress” and, on the other hand, “not all Han people are good,
for many of those in power today are helpers to the evildoers, and they should
not be condoned simply because they are the Han people” (SLXJ, 2: 1005–1006).
Being a Manchu or a Han alone was not a sufficient criterion in judging one’s polit-
ical attitude.
The necessity of a shared national identity among the Han and other ethnic
groups was most clearly articulated in the early twentieth century by a new gen-
eration of literati who had been exposed to political doctrines imported from
the West and concerned with modernizing reforms in China. Liang Qichao, for
example, saw eliminating the differences between the Han and Manchu as the
very beginning of future reforms in China.5 For Liang, the Han people or the
“Chinese nation” (Zhonghua minzu) itself “has never been a single ethnic group
150 The transition to a sovereign state
from its very beginning but a blend of multiple ethnic groups in actuality”
(ibid., 3: 1679–1680). Yang Du (1875–1931), another young activist,
expounded on the new concept of “Chinese nation” and expanded it to include
other ethnic groups than the Han.6 Similar ideas were also found among the
young generation of Manchus and Mongols studying in Japan in the 1900s. For
them, the Manchus and the Han, as well as the entire “people of China”
(Zhongguo zhi renmin), had currently become “people of the same nation but
different ethnic origins” (tong minzu yi zongzu zhi guomin), and they “belong
to the same nation in historical reality and suffer the same exploitation by for-
eign powers” (cf. Huang Xingtao 2011b: 77). But these young Manchus and
Mongols also admitted that, while the Manchus and the Han had been so inte-
grated to make them “inseparable” from each other, it would take time for the
Mongols, the Hui, and the Tibetans to be fully and necessarily assimilated, or
China would soon face the danger of losing its land (ibid.: 78).
At any rate, a growing consensus among the new generation of Han and
non-Han elites in the early twentieth century was that the making of a new
nation was as important as the making of a new state and that the new nation,
based on shared memory of the past and shared concern with the vicissitudes of
the land where they lived, would be possible only when the state was rebuilt to
fully represent the interests of the people regardless of their ethnic origins, so
that China could be turned from a “nationless state,” to borrow from Fitzgerald
(1995), into a true nation-state.
In addition to the forging of a national consensus as outlined above, the
implementation of a series of administrative measures under the late Qing state
also substantially facilitated the integration of the frontiers into interior prov-
inces and worked against the centrifugal inclination among frontier elites. The
first was the Qing state’s policy of allowing Han people to migrate into its fron-
tiers in Manchuria, Inner Mongolia, and Xinjiang in the nineteenth century,
thus spawning economic and cultural ties between the frontiers and the interior,
and driving the gradual integration of the trading and production activities on
the frontiers into the larger markets and economy of interior provinces, which
in turn cultivated the loyalty of non-Han elites and landowners to the Qing
court, given their growing dependence on inland markets (see, e.g., Reardon-
Anderson 2000; Kim 2016). The second measure was merging the frontiers
into the interior administrative system, which began with turning Xinjiang into
a province in 1884 and culminated in establishing three provinces in Manchuria
in 1907. As a result, most of the key administrative and even military positions
in these provinces fell into the hands of Han bureaucrats, which accounted in
large measure for these provinces’ allegiance to the newly established Republic
after the fall of the Qing. The third measure was the Qing court’s religious pol-
icies on Tibetan Buddhism. While patronizing the Lamaist priests in different
parts of Tibet and Mongolia to win their loyalty, the Qing court also made
efforts to prevent the Dalai Lama in Tibet and the Jebtsundamba Khutughtu in
Mongolia from becoming too influential by retaining its power in sanctioning
the procedures of their reincarnation and by elevating the status of the Panchen
 A nation-state in the making 151
Lama in western Tibet and the Icang-skya Khutughtu in Inner Mongolia. It was
primarily through its prerogative in conferring the official title to the Lamaist
leaders that the Republican government renewed and maintained its claim of
sovereignty over the two frontiers; the persistent allegiance of the Panchen
Lama and Icang-skya Khutughtu to the Republican state further evinced and
enhanced its claim. Therefore, in the absence of a shared sense of national
belonging among the Han and non-Han peoples, the heritages of Qing admin-
istrative and religious institutions, more than any other factors, explain the
Qing–Republican continuity in China’s territoriality.

From “eighteen stars” to “five colors”

It is interesting here to examine how the revolutionaries, who were committed
to overthrowing the Qing dynasty, thought about the issue of ethnic division
and integration in China. At the beginning of his revolutionary career, Sun Yat-
sen defined the Revolution as primarily an “anti-Manchu” (pai Man) struggle
with the goal of “expelling the barbarians and recovering Zhonghua” (quchu
dalu, huifu Zhonghua). For him, expelling the Manchus, or “nationalism” as he
defined it, meant overthrowing the Manchu court and reestablishing
a government of the Han people, rather than “completely eliminating the
Manchu people”; he reminded his followers that “we have no reason to take
revenge on the Manchus if they do not oppose us during the revolution” (SZS,
1: 325). Not all of the revolutionaries agreed with him. The most radical of
them embraced a narrowly defined nationalism, or more exactly a unique version
of racism, that aimed at establishing an exclusive state of the Han people and
expelling the non-Han populations from Zhongguo, which was again narrowly
defined as consisting merely of the 18 interior provinces. Zou Rong
(1885–1905), a young radical propagandist from Sichuan, thus called for “kill-
ing all of the five-millions-odd animal-like Manchus and washing away the
shame of two-hundred sixty years of brutal persecution and harsh exploitation”
and thereby establishing “a China of the Chinese China” (ZRWJ: 55). Zhang
Taiyan (1869–1936), a philologist trained in the Han-centric tradition, also tar-
geted the entire Manchu population when authoring his influential anti-Manchu
essays in the early 1900s; for him, “hatred against the Manchus necessarily
means hatred against all of them” (ZTY, 1: 197).
However, the radical revolutionaries soon realized the problem of the exclu-
sive nationalism they upheld, admitting that “expelling the Manchus” was
merely a tactic to mobilize the people, rather than the ultimate goal of the
Revolution. Chen Tianhua (1875–1905), another radical propagandist, thus
wrote before his suicide that the anti-Qing Revolution was essentially a political
rather than a racist one and that the Manchus should be granted equal status
after the Revolution since “modern-day civilization will never allow vindictive
killing [of the Manchus]” (SLXJ, 1: 155). Zhang Taiyan later also clarified on
the eve of the Revolution of 1911 that the “nationalist revolution” (minzu
geming) only meant “recovering our own sovereignty and preventing its
152 The transition to a sovereign state
arrogation by others” and that it should never be equated with “killing all of
the Manchus” or “discriminating against the Manchus” (ZTY, 1: 519).7 Publi-
cations by the revolutionaries in Japan in the late 1900s generally promised
their equal treatment of all ethnic groups in China after the fall of the Qing and
defined the Revolution as the one targeting only the Manchu court rather than
the entire Manchu population. Thus, despite their different goals, the anti-Qing
revolutionaries and those who embraced a constitutional government under the
Qing had a shared cognition of integrating the Han and other major ethnic
groups into a new nation and respecting their political equality under the new
government they endeavored to create.
But the power of the racist anti-Manchu mobilization should not be under-
estimated. After all, the discourse on national integration and the equality of all
ethnic groups in China was largely limited to the political elites. To agitate the
common people who were largely illiterate and had never been exposed to the
imported political ideas, appealing to the history of the Manchus’ atrocities back
in the mid-seventeenth century and to the realities of Manchu discrimination
against the Han remained the most convenient and powerful tool. In fact, the
revolutionaries had to recruit their members primarily from the underground
anti-Qing societies that had long been active in South China, who were com-
mitted to expelling the Manchus and recovering the preceding Ming dynasty.
Not surprisingly, when the Revolution of 1911 broke out first in the city of
Wuchang on October 10, the revolutionaries declared their goal to be “restor-
ing the Han and expelling the Manchu slaves,” and they appealed to the slogan
of “killing all of the barbarians” and “reviving the Han and eliminating the
Manchus.” The flag they invented was the so-called Eighteen-Star Flag that rep-
resented the 18 provinces of the Han people, thus excluding the territories
populated by the non-Han peoples. The same flag was also used by the rebels in
many other provinces during the Revolution. The announcement of Li Yuan-
hong, commander of the revolutionary force in Wuchang, also treated all of the
Manchus as well as the Mongols, the Hui, and the Tibetans as potential
enemies, to be separated from the Han people (Wang Xi’en 2011: 23).
Despite this extreme form of expression of Han-centered nationalism and the
subsequent violence against the Manchus or banner people during the rebellions
in several localities, which resulted in the killing of Manchu civilians in hundreds
or as many as thousands (Rhoads 2000: 187–205), large-scale atrocities against
the Manchus did not occur widely in the interior provinces (Jin Chongji 2001:
17). Nevertheless, the Han-centered mobilization aroused wide concern among
the experienced revolutionaries, who saw the danger of the separation of Man-
churia, together with the lands of the Mongols, the Uighurs, and the Tibetans,
from China, which had served as the “border fence” (pingfan) of the interior
provinces (Esherick 2006: 247). After he returned to China, Sun Yat-sen
worked with his colleagues to redesign the flag through repeated discussions.
The new Five-Color Flag, consisting of five horizontal stripes in red, yellow,
blue, white, and black, and representing the unity of the Han, Manchu,
Mongol, Hui, and Tibetan peoples, was used by the rebels in Jiangsu, Zhejiang,
 A nation-state in the making 153
and other places during the Revolution. In his declaration on the New Year of
1912, Sun Yat-sen, the provisional president of the newly founded Republic of
China, stated: “The foundation of the nation lies in its people. Integrating the
Han, the Manchus, the Mongols, the Hui, and the Tibetans into one people is
the gist of the unification of the nation” (SZS, 2: 2).
Nevertheless, whether or not the newly inaugurated Republic was able to
legally and actually establish its rule over the whole territory of the Qing dynasty
depended on not only the revolutionaries’ own willingness and announcements,
but more importantly on the attitudes of the Qing court and the elites of its
frontiers. After all, the revolutionaries, loosely organized as they were, only con-
trolled the provinces in South and Central China and a few northern provinces
during the first few months after the Wuchang Rebellion. Most of the provinces
in North China as well as all of the frontiers remained loyal to the Qing court.
A prolonged negotiation between the Qing court and the revolutionaries thus
took place to end the confrontation between the north and the south, which
offered an opportunity for the court to bargain for the terms of its abdication.
As a result, Sun agreed to renounce his provisional presidency upon the Qing
emperor’s abdication. After the “Emperor-chaired Conference” (yuqian huiyi)
held on January 17–23, 1912 and participated in by the key members of the
royal family as well as the representatives of Mongol princes and the cabinet
headed by Yuan Shikai, who all agreed on the terms of the abdication, the Emp-
ress Dowager Longyu officially announced the “Decree of Abdication” (xunwei
zhaoshu) on February 12, 1912, wishing “the integration of the whole territory
of the five peoples of the Manchus, the Mongols, the Han, the Hui, and the
Tibetans into one single greater Republic of China” (MGDA, 1: 217). The idea
of “a five-race republic” (wuzu gonghe) was later fully articulated in the Provin-
cial Constitution of the Republic of China publicized on March 11, 1912,
which defined the territory of the Republic as “twenty-two provinces, Inner and
Outer Mongolia, Tibet, and Qinghai,” allowing each of the provinces, Inner
Mongolia, Outer Mongolia, and Tibet to provide five members to form the
senate of the congress and Qinghai to provide one. The fifth article of
the second chapter of the Constitution further stated: “the people of the
Republic of China are all equal regardless of their differences in ethnicity, class,
and religion” (MGDA, 2: 123).
The Qing court’s Decree of Abdication was critical for the Republic to legally
exercise its sovereignty over the entire territory it had inherited from the Qing
and for the frontiers of the former Qing to stay within the Republic (Zhang
Yongle 2016). Right on the eve of the Wuchang Rebellion, witnessing the ram-
pant anti-Qing agitations and foreseeing the inevitable collapse of the dynasty,
Liang Qichao expressed his deep concern over the potential separation of the
frontiers from China after the fall of the Qing; for him, these frontiers had been
“attached to the interior” (neifu) only because of “this dynasty’s influence and
might” (benchao zhi shengwei) (YBSWJ, 3: 1618.). During the north–south
negotiation following the outbreak of the Wuchang Rebellion, the Mongol
princes showed strong aversion to the revolutionaries from the south who had
154 The transition to a sovereign state
built popular support by appealing to the racist slogan of expelling from China
all “barbarians” (quchu dalu), including both the Manchus and the Mongols. In
their collective letter to Yuan Shikai on November 26, 1911, therefore, the 24
Mongol princes affirmed their loyalty to the Qing emperor only: they could
only “lead the land and the people to accept the unity under the Great Emperor
and acknowledge nothing else” (yi wei shi shuai qi tudi renmin yi shou tongyi yu
dahuangdi, buzhi qita ye) (XHGM, 7: 298–299). At the “Emperor-chaired
Conference,” a few Mongol princes stated that:

Mongolia has for long been a subservient part of China only because of its
blood ties with the Qing court that has kept it as a dependency and their
mutual relations in intimacy and fraternity for two centuries and decades.
Once the Qing court is abolished, the relationship between Mongolia and
China would come to an end.
(Bohai shouchen 1969: 899)

But they were quick to assert that their attitudes depended on the result of the
conference: “[we] all depend on the Emperor-chaired conference for
a resolution. If it decides on a republic, we Mongols will all join the great
republican state” (ibid.: 906). The announcement of the Decree of Abdication,
therefore, invalidated the legal ground for the Mongol elites to seek separation
from China, because it ordained the transfer of its rule over the entire country
to the new Republic by explicitly stating that the Empress Dowager, in conjunc-
tion with the emperor, “return the right of government to the entire country,
of which the state system shall be a constitutionalist republic” and that “the five
races of the Manchus, the Han, the Mongols, the Hui, and the Tibetans shall
be united with their entire territories to form a single great republic of China”
(MGDA, 1: 217). For the Mongol elites, especially the princes from Inner Mon-
golia, who had been loyal to the Qing, the reason to stay outside the Republic
thus no longer existed; for those in Outer Mongolia, who had been discon-
tented with the Qing court’s “New Policies” during the last decade of the dyn-
asty, to be discussed in Chapter 7, the legal ground for separation from China
also vanished.
All this suggests the complexity of the consciousness and discourse of the
Han and non-Han elites in the late Qing and early Republican periods. Among
the Han intellectuals, there were radical nationalists such as Zhang Taiyan who
fervently advocated expelling the Manchus from China to create a purely Han
Chinese state, and there were also cosmopolitan liberals such as Liang Qichao
who called for integrating all ethnic groups in China into a single Chinese
nation (YBSWJ, 1: 454; see also Crossley 2002: 344–361); and among the non-
Han elites there were those who sought the separation of their frontier land
from China and those who endorsed the idea of a “Republic of Five Races”
(Huang Xingtao 2010). Regardless of their different ideas about a future state,
a basic fact about late Qing and early Republican politics is that a unified and
centralized nation-state based on the identity of all peoples on the territory of
 A nation-state in the making 155
China was far from a reality. Equally important, however, are the facts that the
transition from the Qing to the Republic involved only a limited amount of
violence and that the fall of the Qing did not cause the separation of its
frontiers from China. These facts should be best seen as the results of cul-
tural homogenization among the Manchu and Han peoples from the mid-
seventeenth century as well as the administrative and economic integration
of the frontiers with interior provinces that accelerated in the late nine-
teenth and early twentieth centuries. All these developments undoubtedly
contributed to the growth of shared identities among the Han and non-
Han elites as well as ordinary people in the last decades of the Qing.

A comparison with the Ottoman empire

The Qing state’s abilities to defend most of its frontiers stood in sharp contrast
with the Ottoman empire’s loss of territories in the Balkans and North Africa in
the eighteenth and nineteenth centuries. More surprisingly, after the collapse of
the Qing, all of its frontiers remained under the Chinese state’s control or suzer-
ainty, which again contrasted with the modern Turkish republic, whose territory
was limited to the heartland of the original Ottoman state. To understand why
the Qing and the Ottoman empire took different paths in their transition to
modern nation-states, let us begin with a look at their respective fiscal and geo-
political conditions in the nineteenth century.
Before the nineteenth century, both the Qing and Ottoman states derived
their revenues primarily from the land, and both witnessed a very slow growth
of annual revenues throughout the eighteenth century, reaching about
80 million taels or roughly 3,000 tons of silver for the Qing around 1800, and
about 150 tons of cash revenue for the Ottoman government or a total of 250
tons of silver if the receipts kept by cavalrymen under the timar system were
also included (the total revenue of the Ottoman empire, therefore, fluctuated
between 200 and 300 tons from the sixteenth through the eighteenth century,
without a significant upward trend) (see Kıvanç and Pamuk 2010: 604). In the
nineteenth century, especially after the 1850s, both states saw a rapid increase in
revenue, reaching more than 3,000 tons by the 1910s, or a growth factor of 15,
for the Ottoman government (ibid.: 594, 619–621), and more than 11,000
tons by 1911, or a growth factor of about 4, for the Qing court. And both gov-
ernments made possible the phenomenal growth in revenue by turning to the
creation of surcharges, taxation of goods, debasements, and borrowing from the
financial market. This huge increase in revenue explains in large measure why
both the Qing and the Ottoman empire were able to update their military, sur-
vive foreign pressures, and defend their sovereignties with varying degrees of
success in the nineteenth centuries. Both survived as sovereign states by the turn
of the twentieth century also because of their unique role in the regional geo-
political order. Both were huge in size, playing a pivotal role in sustaining the
balance of power among the major contenders in the region; and none of the
156 The transition to a sovereign state
great powers allowed a single member among them to dominate or conquer the
entire country.
What accounted for the different fates of the Qing and Ottoman empire in
keeping their land was primarily the peculiar pattern of geopolitical relations
within their territory. A fundamental weakness of the Ottomans was the geo-
graphic location of their capital city, Istanbul, which was too close to Europe,
thus forcing them to defend their European possessions with excessive atten-
tion and resources until the Balkans were lost. They failed to keep the latter
partly because of revolts by the Christian populations and more importantly
because of intervention by European powers in this region under the pretext
of protecting the Christian minority; the latter had been content with their
autonomy under the Ottoman policy of tolerance for centuries when the
empire remained powerful, but sought independence in the nineteenth and
early twentieth centuries after the empire suffered repeated defeat by Russia
and other great powers (Braude 2014). The Arab and African provinces,
having contributed little or nothing to the empire fiscally and militarily,
received much less attention from the center throughout history than did the
Turkish core and they became increasingly dissatisfied with the empire in the
modern era for its weakness in dealing with foreign powers on the one hand
and incompetence and arbitrariness in managing the provinces outside the
Turkish heartland on the other. Egypt, which had never been effectively
incorporated into the empire, became independent as early as the end of the
seventeenth century. The most fatal step toward the loss of its Arabian prov-
inces was, of course, the Ottoman government’s decision to ally with Ger-
many and Austria–Hungary in World War I; their defeat in 1914 led to the
collapse of the Ottoman empire and the loss of its territories other than
Anatolia and a small area in Europe, on which the Republic of Turkey was
established in 1923.
The Qing dynasty’s advantages in relation to the Ottomans lay in, first of all,
the overwhelming dominance of the Han population in its demographic, eco-
nomic, fiscal, and military formations, which contrasted sharply with the Otto-
man empire which had excessively relied on its European possessions
throughout history and whose population of 12 million in the Turkish heartland
accounted for only 57% of its entire population (roughly 21 million) as late as
the end of the nineteenth century. Thus, whereas the independence of individ-
ual Balkan provinces from the empire from the late eighteenth through the early
twentieth centuries constituted a huge loss to the Ottomans demographically,
economically, and fiscally, the homogeneity of the Han population in all of the
provinces under the Qing precluded the possibility of the independence of any
of these provinces. And while the Qing allowed its frontiers outside the prov-
inces a high degree of autonomy in internal administration, as did the Ottomans
towards its provinces outside the Turkish core, the close political and religious
ties that the Qing court had established with the frontiers, and more import-
antly its decisive role in selecting the chief administrators or religious leaders of
the frontiers, and its stationing of military forces at key points of the frontiers,
 A nation-state in the making 157
also worked to prevent the latter from seeking independence from the imperial
court.
Ideology and identity also mattered in explaining the different consequences
between China and Turkey in their transition to modern nation-states. In
response to the growing threats from foreign powers and under the influence of
patriotic and nationalist ideas imported from Europe, the political and intellec-
tual elites of the Ottoman empire were attracted to various schemes to forge
political identity among them and their supporters, including:

1 Ottomanism, which called for equality of all religious groups (millets) within
the empire and once prevailed among the Young Turks but failed to appeal
to the Balkan Christians who were reluctant to identify themselves as
Ottomans;
2 Pan-Islamism, which advocated unity of all Muslims throughout the empire
and beyond for the goal of creating a single state under a legitimate caliph,
which interested the Ottoman government but failed to attract the young
generation of intellectuals educated in the West or influenced by Western
ideas; and
3 Pan-Turkism, or the unification of all the Turkish-speaking peoples in
Turkey, Russia, Persia, Afghanistan, and China into a single state, an idea
that found its supporters mainly among the Turkish exiles and immigrants
from Russia.

What eventually prevailed and worked in building a new state on the ruins of
the Ottoman empire, however, was the idea of creating a territorial state of the
Turkish people in Turkey, as formulated by Mustafa Kemal (1881–1938), who
limited the scope of his revolution only to the land inhabited by the Turks,
which he termed “our natural and legitimate limits” (Lewis 2002: 353–354).
The Qing shared with the Ottoman empire its emphasis on loyalty to the
monarch in identity building among its subjects, as did all other dynastic states
in world history. However, in three important ways, the Qing was fundamen-
tally different from the Ottomans when it came to the definition of its rulership
and the state per se. First, unlike the Ottoman sultan who defined himself as
not only the ruler of his empire but more importantly the caliph of the entire
Islamic community committed to unifying Muslim lands and conquering the
non-Islamic world, and who relied on religious ties more than any other means
to keep the empire together, the Qing emperors defined themselves primarily as
the ruler of a secular state and played no role in any religions in China, nor did
they make efforts to promote any religion throughout the land they ruled,
although they frequently acted as the patron of the Lamaist communities in
Tibet and Mongolia for the pragmatic purpose of keeping them in submission.
The Chinese state, in a word, was a secular state; it was never defined in reli-
gious terms. Second, unlike the Ottoman empire that had no definite geo-
graphic borders because of its frequent expansion or loss of frontiers, the border
of the Qing state after the 1750s was stable and saw no significant change until
158 The transition to a sovereign state
the second half of the nineteenth century when foreign powers encroached; it
defined its borders with neighboring states by signing treaties, agreements, sta-
tioning patrolling forces, or customary arrangements, and the term Zhongguo
was widely used in official documents to refer to the entire territory under its
rule, including both the interior provinces and the newly established frontiers
outside them. And third, unlike the Turks that had ancestral origins outside the
Turkish core of the Ottoman empire and had their related populations scattered
from central and eastern Europe to central Asia, the Han people that constituted
the dominant population in China were concentrated in the interior provinces,
although the growing number of emigrants from coastal provinces to southeast
Asia gave rise to Chinese diasporas there by the early twentieth century. Before
the Manchu conquest in the 1640s, the geographic distribution of the Han
people was largely identical with the territory of the state they had established;
the Qing incorporation of the Inner Asian frontiers into its territory turned the
state into a multiethnic one in theory but the demographic, economic, and cul-
tural predominance of the Han people remarkably distinguished the Qing from
the Ottomans where the population outside the Turkish core accounted for
more than 40% of the empire’s total population by the turn of the twentieth
century.
Qing China, in a nutshell, was a secular, territorial state, whose borders
could be easily defined by its written or customary demarcations with neigh-
boring states and by its people and their shared historical traditions. Unlike
the Turks in the nineteenth and early twentieth centuries who had difficulties
in identifying or redefining the political entities to which they belonged or
aspired to build, hence struggling with ambiguous and often illusory religious,
imperial, or ethnic boundaries, and even having a hard time finding a proper
name for the new state they wanted to build (Lewis 2002: 354–355), for the
Chinese state-makers in the early twentieth century, the state they wanted to
remake was clearly defined and already there; it was the Zhongguo that the
Qing had inherited from the Ming and expanded by the 1750s, a state that
included both the interior provinces and the newly established frontiers. To
be sure, the most radical among them once called for the establishment of an
exclusive republic of the Han people. But this scheme was more of
a propaganda device intended to arouse the Han population’s anti-Manchu
feelings by recalling the history of the Manchu conquest; it soon yielded to
a consensus among them to build a “Republic of Five Races,” which came to
guide the organization of the new Republican state after the Revolution of
1911. The contrast between the makers of modern China and Turkey thus is
stark. While for Mustafa Kemal, Ottomanism, Pan-Islamism, and Pan-Turkism
were nothing less than “great fantasies” or “ideas which we did not and
could not realize” (Lewis 2002: 353–354), for Sun Yat-sen and his comrades,
to establish a new Republic on the territory of the former Qing was not only
feasible but also mandatory for the legitimacy of the new state being built –
after all, the state called Zhongguo had already been there for centuries; they
only needed to rebuild it, rather than invent it.
 A nation-state in the making 159
Notes
1 In actuality, China paid 652 million taels in total from 1902 through 1938 (see Wang
Nianyong 1994: 177).
2 Official revenue from the land tax was 32.79 million taels in 1887, but the tax monies
and extra charges pocketed by local tax collectors and the portion remitted to the local
government but unreported by government authorities at the provincial and lower
levels were more than the officially reported figure, while agricultural output was
2,230 million taels in the same year. By the end of the Qing, the officially reported
land tax income increased to more than 48 million taels as a result of fiscal consolida-
tion in the preceding three years, though the amount of taxes and surcharges actually
paid by landowners remained around 60 to 70 million taels while agricultural output
did not change too much from its 1908 level.
3 They helped the merchants raise funds by using government influence or directly col-
lecting special levies on the land and other commodities, and they generally favored
railroad companies when mediating disputes over the purchase of land for railroad con-
struction and provided protection whenever the construction encountered local resist-
ance. Sixteen railroad companies were established in 1903–1907 through the “joint
efforts between the government and the people” (guan min heli). Contemporaries
described this relationship as “the government takes advantage of the gentry’s
strength, and the gentry relies on the government’s power” (guan yong shen li, shen ji
guan wei) (ZGTL, 2: 1012).
4 The Qing court had long tolerated since the early eighteenth century intermarriage
between the Han banner people and ordinary Han residents found in Liaodong and
other places, but it strictly banned intermarriage between the banner people of
Manchu and Mongol origins and the Han residents (Ding Yizhuang et al. 2004:
272–273).
5 Influenced by the works of Johann Kaspar Bluntschli (1808–1881), Liang proposed in
1903 the concept of “greater nationalism” (da minzuzhuyi) or “merging the Han, the
Manchus, the Hui, the Miao, and the Tibetans into one large nation” in China, to
replace the outdated “petty nationalism” (xiao minzuzhuyi) or the ethnocentrism of
the Han people, following the precedents of the ancient Roman Empire and the
United States, where “the major ethnic groups within a country were united and
melted into a new nation” (YBSWJ, 1: 454).
6 For him, the Manchus had long been assimilated into the “Chinese nation” while the
Mongols, the Hui (namely the Uighurs and other non-Han Muslims), and the
Tibetans were “yet to be fully integrated into it.” Central to the task of “national inte-
gration” (guomin tongyi), therefore, was “promoting the equality between the Man-
chus and the Han while assimilating the Mongols, the Hui, and the Tibetans,” so that
no ethnic distinctions would ever exist in the “Chinese nation” that had already
“mixed myriads of ethnic groups in the past thousands of years.” Yang further elabor-
ated the idea of Zhongguo as a territorial and national entity. “Situated in today’s
world,” he wrote, “Zhongguo cannot afford to lose any of the lands of the Hans, the
Manchus, the Mongols, the Hui, and the Tibetans, and it cannot afford to lose any of
the peoples of the Hans, the Manchus, the Mongols, the Hui, and the Tibetans.” To
“merge the five into one” (he wu wei yi), he suggested, should be “a policy of supreme
importance” for China in the future (YDJ, 1: 303).
7 For a discussion on the different attitudes between Zhang Taiyan and Liang Qichao
towards ethnic relations in China, see Crossley (2002: 345–361). Note that Crossley
fails to point out Zhang’s adjustment of his opinion on this issue as shown here.
Part III

The making of a unified

and centralized state
Map III. 1 China under the Republic, circa 1920
7 Centralized regionalism
The rise of regional fiscal-military
states

Known as the age of warlordism, the years following the death of President
Yuan Shikai in 1916 witnessed the rise and fall of political factions and military
groups vying for dominance under the Republican government in Beijing, prior
to the establishment of the Nationalist government in Nanjing as a national
regime in 1928.1 To explain the success or failure of the warlords, scholars have
typically emphasized the role of personal ties or factionalism, especially when
discussing the old-fashioned warlords in North China.2 Others pay more atten-
tion to the importance of ideological propagation and party organization when
analyzing the rise of the Nationalist (or Guomindang) forces in the southern
provinces (e.g., Wilbur 1968; Luo Zhitian 2000).3 This chapter instead reinter-
prets the dynamics of military rivalry among the warlord cliques from the per-
spective of modern state-building. It shows that the warlords or warlord cliques
of individual provinces were not just concerned with military buildup and war
making; the most successful of them also took vigorous steps to strengthen
themselves by centralizing the bureaucratic system within their territories, uni-
fying and standardizing the fiscal and financial systems, constructing highways,
railroads, and other infrastructural projects, promoting public education and
health, encouraging the development of industry and commerce, advocating self-
government at the grassroots level, and allowing the existence of autonomous
representative and deliberative assemblies at the provincial or county level to build
consensus among the elites. Altogether, these efforts allowed the most ambitious
contenders to turn their provinces into regional “fiscal-military states,”
a development quite comparable to state-making in early modern Europe. Which
warlord or regional force would survive the competition and prevail in the end
depended ultimately on how they organized their fiscal and military institutions
and how they generated their revenues. Those who centralized and bureaucrat-
ized the fiscal and military apparatus would eventually outcompete those who did
not. Geopolitical and socioeconomic conditions also mattered. Those who
enjoyed a relatively secured geopolitical position and at the same time established
a centralized apparatus for resource extraction, especially the taxation of com-
merce and urban economy, and had access to the modern means of financing,
would be able to build a powerful fiscal-military regime and prevail in the
competition for national dominance. The most competitive force, therefore,
164 The making of a unified and centralized state
was invariably the one that possessed not only a well-endowed and secured
region but also a highly centralized machine of control and extraction. In con-
trast, those who failed to establish their stable control of a region and
a centralized regime would inevitably lose the competition and be wiped out
or subjugated by the winner.

Warlordism

The central government crippled

The emergence of warring regional cliques in early Republican China can be
traced to the decentralization of fiscal and military power since the late 1850s,
when the Metropolitan and Coordinate Remittance systems, a device used by
the central government to centrally control the revenues generated by local
authorities, collapsed because of the Taiping Rebellion. Consequently, provincial
governments were allowed to keep the remainder of whatever they had collected
and controlled after fulfilling two basic categories of obligations: (1) zhuanxiang
jingfei, or fixed quotas of annual remittance to the central government for spe-
cific civilian and military purposes, and (2) tanpai, or the compulsory sharing of
foreign debt and war indemnities, which first occurred in 1895 after the Sino-
Japanese War and dramatically increased after 1901 (Zhou Yumin 2000: 242–244,
389–396). The Qing court, to be sure, attempted repeatedly to formalize and
recentralize the provincial governments’ collection and management of taxes, cus-
toms, and fees in the late nineteenth and early twentieth centuries. Especially note-
worthy was its systematic efforts in the last few years of the dynasty to investigate
the fiscal condition of provincial authorities, which resulted in dramatic increases in
the formally reported quantity of local revenues, to centralize its control of the salt
tax, which soon surpassed the land tax and maritime customs to become the largest
source of revenue, and to establish a modern budget system (Liu Zenghe 2014:
80–198). The outbreak of the Revolution of 1911, however, interrupted all this
progress towards fiscal centralization.
The Republican government in Beijing, therefore, had only a few million
yuan annually at its disposal, mostly from the salt taxes and domestic customs
from adjacent provinces in 1912 and 1913, whereas its administrative and mili-
tary expenses amounted to 4 to 5 million yuan a month (MGDA, 3.1: 87;
Jia Shiyi 1934: 45–46, 170). The only way to meet the state’s fiscal needs was
to incur foreign debt and issue government bonds, the largest of which was
the so-called “grand debt for postwar rehabilitation” (shanhou da jiekuan) of
£25 million (equivalent to 248.3 million yuan), made by the banking groups of
five foreign countries to the central government under President Yuan Shikai in
April 1913.4 In the following years, as Yuan Shikai gradually consolidated his
power over individual provinces, he was able to reestablish the revenue sharing
system, and provincial governments had to submit to Beijing a fixed quantity of
local tax funds. The remittance of the shared tax funds (zhongyang jiekuan)
increased from 5.6 million yuan in 1913 to 14 million in 1914 and 19.02 million
 Centralized regionalism 165
in 1915 when Yuan was at the height of his political influence (see Table 7.1). In
addition, the Yuan government was also able to obtain from individual provinces
a growing amount of tax income that was designated as revenue belonging exclu-
sively to the central government (known as zhongyang zhuankuan), including the
taxes on deeds, stamps, salt and tobacco sales and licenses, and sales commissions;
the Beijing government’s revenue from this source increased to more than
18.7 million yuan in 1915 and 24.4 million in 1916 (Table 7.1). Combined, the
revenues that the Yuan government obtained from the provinces reached nearly
38 million yuan in 1915 and more than 43 million yuan (equivalent to roughly
30 million taels) in 1916, which were more than the special funds that the prov-
inces owed to the Qing court in the late nineteenth century (22 million taels, see
Zhou Yumin 2000: 371). This was undoubtedly a considerable achievement
toward the goal of fiscal and military recentralization, to which the Yuan adminis-
tration was committed. In the absence of serious challenges after suppressing the
rebellion in the southern provinces, Yuan indeed appeared to be the most compe-
tent leader to reestablish political order in China.
Unfortunately, Yuan’s failure to establish an imperial regime and his subse-
quent death in June 1916 opened the door for the recurrence of political chaos
and military rivalry among the different cliques of power contenders in the next
decade. A central government continued to exist, but it was fiscally weak, due
to the disappearance of provincial contributions (jiekuan and zhongyang zhuan-
kuan) after 1922, which had been the major source of the government’s regular
revenue during Yuan’s presidency, and due to the ending of its revenues from
the “remainder of maritime customs” (guanyu) (most of the maritime customs
had been retained by the Maritime Customs Service as repayment of foreign
debts and war indemnities) after 1921. The “remainder of salt taxes” (yanyu),
the largest source of domestic revenue in the 1910s (again most of the salt taxes
had been retained as payment of foreign debts since 1913), also diminished in
the 1920s, to less than 9 million yuan by 1926 (Yang Yinfu 1985: 5–9). Thus,
the Beijing government had to increasingly rely on foreign loans and govern-
ment bonds, which brought its total revenue to as much as 350 million yuan in
1918. As the central government’s borrowing abilities dwindled in the following
years, however, its annual revenue also decreased to 126 million yuan in 1922
and to only 29 million yuan by 1926 (Table 7.1).
In sharp contrast with the poor and feeble government in Beijing was the
steady buildup of fiscal and military strength of provincial warlords in the 1910s
and 1920s. Three major cliques of warlords prevailed in North China. One was
the Anhui clique, which controlled eight provinces (Shanxi, Shaanxi, Shandong,
Anhui, Zhejiang, Fujian, Gansu, and Xinjiang as well as two special administra-
tive zones, Rehe and Chahar). The second was the Zhili clique, which occupied
five provinces (Zhili, Jiangsu, Henan, Hubei, and Jiangxi as well as Suiyuan and
Ningxia). The third was the Fengtian clique, which ruled three northeastern
provinces (Fengtian, Jilin, and Heilongjiang). All these reflected the situation in
June 1920 prior to the outbreak of the war between the Anhui and Zhili cliques
(Lai Xinxia et al. 2000: 594). In addition, there were minor warlord factions
Table 7.1 Revenue of the central government of the Republic, 1912–1945 (in 1,000 yuan)

Yeara Maritime Domestic/ Salt taxb Central Central Stamp/ Tobacco / Foreign Public Treasury Bank Real
customsb unified remittance special direct alcohol loanse debt bonds loans revenue
d
taxc funds funds tax tax

1912 111,700 6,200 0

1913 5,600 331,900 6,800 2,200
1914 14,000 33,500 25,000 10,100
1915 6,000 31,389 19,018 18,748 1,600 25,800 400 1,160 130,678
1916 6,000 18,875 24,400 35,100 8,800 1,800 0
1917 10,800 5,000 15,472 10,360 2,000 6,000 68,800 10,500 200 8,000 137,000
1918 2,700 4,200 52,800 6,043 5,755 2,000 4,000 126,200 139,400 7,000 0 350,000
1919 21,700 4,800 43,300 5,553 4,245 2,000 2,673 34,000 28,400 5,300 669 153,000
1920 17,800 3,500 36,100 4,917 4,245 1,500 2,230 36,200 122,000 24,700 2,736 256,000
1921 0 3,600 29,500 2,959 4,245 800 1,784 27,400 115,400 29,000 45,163 260,000
1922 0 690 21,000 0 0 750 1,450 9,700 83,200 2,200 6,885 126,000
1923 0 718 20,000 0 0 720 1,400 31,000 5,000 3,500 5,858 68,000
1924 0 668 15,000 0 0 720 1,400 16,200 5,200 100 185 39,000
1925 0 668 12,000 0 0 720 1,400 125,900 23,000 0 16,000
1926 0 668 8,900 720 1,400 1,900 15,400 29,000
1927 12,500 600 20,800 70,000 148,256
1928 179,142 27,691 29,542 14,544 3,034 3,549 44,506 24,048 28,078 434,440
1929 275,545 36,567 122,146 11,385 5,427 6,831 90,511 539,005
1930 312,987 53,330 150,484 3,548 6,111 8,617 192,816 185,458 714,468
1931 369,742 88,681 144,223 175 4,799 7,626 125,456 108,111 682,991
1932 325,500 79,600 158,100 673,300

(Continued )
Table 7.1 (Cont.)

Yeara Maritime Domestic/ Salt taxb Central Central Stamp/ Tobacco / Foreign Public Treasury Bank Real
customsb unified remittance special direct alcohol loanse debt bonds loans revenue
taxc funds funds taxd tax

1933 352,400 105,000 177,400 124,000 801,600

1934 71,200 115,300 206,700 124,000 1,207,000
1935 24,200 152,400 184,700 560,000 1,328,500
1936 635,900 131,300 247,400 455,000 1,972,600
1937 239,000 30,000 141,000 256,000 2,010,000
1938 78,041 9,755 29,265 $100,000 10,975 712,739
1939 98,173 6,188 17,159 $175,668 7,033 858,246
1940 2,975 5,715 6,264 5,950 $45,000 626 403,949
1941 0 6,898 6,059 $70,000 4,635 392,557
1942 0 12,864 15,104 11,046 $500,000 1,996 329,126
1943 0 12,779 14,217 17,860 18,193 287,926
1944 0 11,898 24,697 11,014 $200,000 2,799 303,103
1945 0 12,094 26,753 7,205 628,070

Sources:
Yang Yinpu (1985: 7-8, 10, 12, 15-16, 22, 45, 47, 64, 104, 107, 109, 112, 150); Jia Shiyi (1932: 55-63, 158-159); Jia Shiyi (1933: 160-170, 199-206, 296-
299, 409-410, 556-559); Jia Shiyi (1934: 60-61); MGDA 3.1: 197-200, 212-216, MGDA 5.1.1: 547-549; 551-553, 565-567, 572-581, 599-600; Young
(1971: 38, 52, 56 and 73).
Notes:
(a)
Data from 1912 to 1926 are in silver dollars (yin yuan). Statistics of the years from 1938 to 1945 are recalculated into the amounts in the fixed value of the
fabi (legal tender) in 1937 (see Yang Yinpu 1985: 159);
(b)
Maritime customs and salt taxes before 1928 were the remainders of the custom dues (guanyu) and remainder of salt taxes (yanyu);
(c)
The data under “domestic/general tax” before 1928 are the amounts of domestic regular taxes (changguanshui), and the data after 1928 are the amounts of
the unified tax (tongshui);
(d)
The numbers under “stamp/direct taxes” refer to stamp tax before 1940 and various direct taxes (including stamp tax, income tax, and sales tax, etc.) (see
Yang Yinpu 1985: 112);
(e)
Foreign loans in 1938-1942 and 1944 were in U.S. dollars (Yang Yinpu 1985: 153).
168 The making of a unified and centralized state
including the Guangxi clique that occupied Guangxi and Guangdong, and the
Yunnan clique that dominated Yunnan and Guizhou.

The weakness of the Anhui and Zhili cliques

In terms of their fiscal strength, the first two cliques were quite comparable to
each other: the revenues of the eight provinces and two zones of the Anhui
clique totaled about 54 million yuan, while the five provinces and two zones of
the Zhili clique generated a total of nearly 51 million yuan (Jia Shiyi 1932:
138–139). The resemblance between the Anhui and Zhili cliques in total
amounts of revenues explained at least in part their rough equivalence in military
forces on the eve of the war between them in June 1920: 55,000 men on the
Anhui side and 56,000 men on the Zhili side (Lai Xinxia et al. 2000: 618–619).
On the other hand, the Fengtian clique appeared to be much weaker, having
only 26 million yuan as its total revenue. But the Anhui and Zhili cliques had
their own weakness. Duan Qirui and Feng Guozhang, the respective leaders of
the two cliques, had been Yuan’s two most capable and trusted subordinates,
but Duan and Feng had competed with each other while Yuan was alive. After
Yuan died, Duan controlled the central government as the premier of the state
council and much of North China through the military governors loyal to
him; Feng, for his part, prevailed in South China by serving as the military
governor of the most prosperous Jiangsu province and allying with other gov-
ernors in the Yangzi River region while also serving as the vice president and
the acting president of the Beijing government until October 1918. But the
military governors of individual provinces formed a clique under Duan or
under Feng (or Cao Kuan after Feng’s death in late 1919) only on the basis of
their personal relationship with either of the two clique leaders (Lai Xinxia
et al. 2000: 26). In other words, Duan or Feng (or Cao) built their respective
factions only by taking advantage of their position as leaders of the Beijing
government to appoint or recommend their trusted followers or friends as the
military governors to different provinces within their reach of power; when
they had conflicts in appointing the governors, Duan and Feng had to bargain
with each other to achieve a gross balance between the appointments each of
them proposed.
Such personal networks and loyalty to the factional leaders did matter when
the two factions had a war; for the military governors, to join the war was the
best way to protect their own positions and military forces. Nevertheless, neither
Duan nor Feng (and other Zhili leaders) had succeeded in turning the individ-
ual provinces of their respective cliques into a solid fiscal-military entity. Each of
the military governors completely controlled their own armies and were respon-
sible for raising enough monies to feed their soldiers; they also completely con-
trolled the tax revenues generated within their own provinces, and refused to
contribute any part of their revenues to the central government regardless of
their personal loyalty to the clique leader who held a key position in the central
government. Both the Anhui and Zhili cliques, in a word, were essentially
 Centralized regionalism 169
a confederation of warlords who were loosely brought together by their personal
loyalty to the clique leader; there was no centralized administrative or military
mechanisms to keep them together as members of a political and military body
with a high level of group solidarity. Later in 1926 and 1927, the Zhili clique
lost their control of the provinces in the Yangzi River region precisely because
the key leaders of the group in different provinces (most notably, Wu Peifu in
Hubei and Sun Chuanfang in Jiangsu) failed to act together and help each
other to fight their shared enemy, the Guomindang force from the south; the
latter had no difficulty defeating each of them one after another (Lai Xinxia
et al. 2000: 990–991).

The rise of the Fengtian clique

The lack of group solidarity on the part of the Anhui and Zhili groups was
in sharp contrast with the centralized administrative and military organiza-
tions that undergirded the Fengtian clique in Manchuria. In fact, one of the
reasons that prevented the Anhui or Zhili clique from building a centralized
fiscal and military entity of their own had to do with the fact that the
member provinces of these cliques were geographically scattered in different
areas of the country and were interwoven with the member provinces of the
opposite clique. This scatteredness not only exposed each province to military
attacks from its enemy, but also prevented the member provinces of each
clique from mobilizing their resources to build a centralized political and
military body of their own.
The situation of the Fengtian clique was very different. Zhang Zuolin
(1875–1928), the head of this clique, started building his power base with his
control of the 27th division of the new army that garrisoned Fengtian province
after the Revolution of 1911, which enabled him to further seize the positions
of the military governor and the provincial governor of Fengtian, and to create
the new 29th division and take control of the 28th division of the army after
1916. After establishing his complete control of Fengtian, Zhang extended his
reach into the neighboring Heilongjiang province in 1917 by recommending
Bao Guiqian, his relative by marriage, to be the new military governor of the
province. Zhang himself also received the Beijing government’s appointment
as the Inspecting Commissioner of the Three Northeastern Provinces (Dong-
shansheng xunyueshi) in 1918, which made him the official administrative and
military ruler of the three provinces (Hao Bingrang 2001: 174; McCormack
1977: 48). To the extent that Zhang controlled the neighboring two prov-
inces by appointing his trusted men as their military governors, he was no dif-
ferent from the leaders of the Anhui and Zhili cliques in building their
respective factional groups. What made Zhang distinctive and successful in the
end in the rivalry among the warlords under the Beijing government was that
he took advantage of the geographical isolation of Manchuria to turn the
three northeastern provinces into an independent and highly centralized polit-
ical, fiscal, and military entity.
170 The making of a unified and centralized state
To achieve this goal, Zhang first cut off the political ties of his three provinces
with the Zhili-clique controlled central government by declaring his independence
from Beijing, after he lost the war against the Zhili clique in April 1922 and failed
to expand his influence into the interior provinces. He further made efforts to mod-
ernize the administrative system in Manchuria by introducing the civil service exam-
ination system in September 1922 to recruit government officials at different levels
on the basis of their merits. Zhang also rebuilt his troops, replacing most of the
lower and middle-ranking military officers of bandit backgrounds with graduates
from the military academies in Japan or Beijing; he also expanded the size of the
Northeastern Military Academy to train his own military officers and held training
sessions for high-school graduates to prepare them for military service. To root out
the personal networks (mostly in the form of sworn-brotherhoods among the high-
ranking military commanders) from the military, Zhang reorganized the three divi-
sions (the 27th, 28th, and 29th) and an additional new division he had established
into 27 infantry brigades and other units, and appointed his trusted subordinates to
lead them (Hao Bingrang 2001: 39–42; McCormack 1977: 100–106). The admin-
istrative and military unification in Manchuria meant a lot for the Fengtian clique:
it prevented internal strife among the three provinces, ensured political stability and
social order, freed Manchuria from the recurrent warfare that plagued the interior
provinces, and enabled Zhang to mobilize the resources from all of the three prov-
inces for his strategic goals.
To enlarge and maintain his military forces, Zhang generated revenues through sev-
eral channels. In addition to the collection of taxes from the land, industry, and com-
merce, the Fengtian clique made government investments on a large scale, which
ranged from mining and lumbering to textile manufacturing, power generation, sugar
production and the defense industry, and, most importantly, its construction of
a railroad network. It also developed financial tools to generate additional revenues,
such as issuance of public debt and the printing of paper notes, known as fengpiao,
which had been widely accepted and circulated in Manchuria because of its stable
value from 1917 to early 1924 (Suleski 2002: 47–49; Kong and Fu 1989: 41–42).
The strength of the Fengtian clique, in a word, lay in its ability to integrate
the three northeastern provinces into a single administrative, fiscal, and military
entity that was centralized under one single strongman. Individually, none of
the three provinces was the most prosperous in the entire country. In 1925, for
instance, Jiangsu province had a budgeted revenue of 16.6 million yuan, much
higher than the revenue of Fengtian, the richest of the three northeastern prov-
inces; another two interior provinces, Sichuan and Guangdong, again had
an annual revenue higher than or close to Fengtian’s (see Table 7.2). But the
problem for the warlord cliques who controlled a number of the interior prov-
inces was that the fiscal resources in each of the provinces fell into the hands of
the military governor who controlled that province; as a result, none of the
clique leaders was able to single-handedly control the resources from all of the
provinces of his clique. In sharp contrast, by establishing his tight control over
Manchuria, Zhang Zuolin was able to extract fiscal resources from all of the
three provinces for his military buildup; combined, the annual revenues from
Table 7.2 Budgeted revenue and expenditure of individual provinces, 1925 (in silver yuan)

Provinces Land tax Taxes on Local taxes Local fees Gov’t enterprise Misc. Total Military Total
goods income incomes revenue spending expenditure

Zhili 5,809,139 926,790 1,592,809 270,384 460,436 282,605 9,342,163 6,692,844 10,961,692
Fengtian 4,086,999 5,049,778 1,359,044 300,644 1,597,089 12,393,554 6,918,538 10,131,248
Jilin 2,157,052 3,770,359 2,041,420 117 255,876 8,224,824 8,686,404 11,930,995
Heilongjiang 1,483,047 3,270,218 1,153,793 43,606 656,815 60,253 6,667,732 5,641,262 7,818,573
Shandong 8,135,171 797,364 1,281,933 3,500 2,860 185,441 10,406,269 13,800,000 17,306,301
Henan 5,471,148 846,000 1,186,134 680,000 10,203 133,788 9,327,273 16,817,253 19,824,134
Shanxi 5,929,289 743,980 622,423 40,000 7,335,692 5,636,044 8,021,263
Shaanxi 3,643,281 1,010,000 195,943 4,849,224 2,468,226 4,328,662
Gansu 1,467,451 997,067 349,561 129,800 3,776 10,850 2,958,505 3,051,569 5,258,741
Xinjiang 1,590,412 472,401 377,129 34,551 86,062 2,560,555 3,331,116 4,882,701
Jiangsu 8,496,046 6,428,507 1,620,000 64,612 168,150 16,777,315 6,122,374 14,892,393
Anhui 3,822,137 1,538,700 1,370,800 30,380 6,762,017 3,800,305 6,472,491
Jiangxi 4,355,234 2,572,511 889,407 20,000 376,964 8,214,116 9,528,914 12,156,409
Hubei 2,659,757 3,223,227 1,605,265 568,465 31,250 8,087,964 8,130,415 10,974,811
Hunan 2,801,952 2,352,456 702,990 60,000 5,917,398 3,564,014 6,989,338
Sichuan 6,861,394 819,402 4,466,484 132,287 265,000 12,544,567 26,296,358 30,061,790
Zhejiang 5,928,980 1,819,822 869,000 2,240,475 319,591 11,177,868 9,876,625 14,371,463
Fujian 3,235,290 1,430,000 1,167,054 240,974 6,073,318 10,624,000 13,204,829
Guangdong 3,889,585 4,562,179 3,302,274 447,285 135,860 12,337,183 15,959,398 19,662,056

(Continued )
Table 7.2 (Cont.)

Provinces Land tax Taxes on Local taxes Local fees Gov’t enterprise Misc. Total Military Total
goods income incomes revenue spending expenditure

Guangxi 2,324,800 1,435,441 283,100 10,800 50,000 4,104,141 5,673,435 7,469,452

Yunnan 1,153,377 642,015 263,169 145,815 13,692 2,218,068 2,131,416 4,260,138
Guizhou 739,313 461,289 295,481 1,496,102 2,814,300 4,489,078
Rehe 157,204 364,148 295,986 18,000 52,000 71,761 959,099 997,208 1,575,683
Chahar 445,572 72,140 47,734 64,103 629,549 1,245,538 1,796,439
Suiyuan 94,338 66,290 58,921 262386 4,284 44,865 531,084 709,864 1,072,115
Total 87,515,719 45,428,798 28,942,549 4,768,718 1,873,283 4,561,630 173,333,992 182,418,613 253,797,479

Source: Jia Shiyi (1932: 146–152).

 Centralized regionalism 173
the three provinces totaled 27.3 million yuan (Table 7.2), which was in addition
to the revenues he generated through issuing public debts and printing paper
currencies. His fiscal power, in a word, surpassed any of the military governors
of the interior provinces.
The geographical isolation of Manchuria from the rest of China also contrib-
uted a lot to Zhang’s success in competing with the warlord cliques of the inter-
ior provinces. After establishing his control of the three northeastern provinces,
Zhang was able to join the rivalry among the warlords in North and East China
and thereby to extend his military and political reach there. He started his mili-
tary operations in the interior when the warlord cliques there were fighting with
each other. When he failed in this venture, he was able to easily withdraw from
the interior and keep his home base in Manchuria intact. He was then able to
concentrate on building his fiscal and military muscle in Manchuria and wait for
another chance to venture into the interior, which he then succeeded at after
the interior warlords had exhausted their resources after years of battles. This
geopolitical advantage, together with the differences between the Fengtian
clique and the major cliques of the interior provinces (Anhui and later Zhili) in
their respective fiscal constitutions, were more important than any other factors
(political or military) in explaining why the Fengtian clique ultimately won the
rivalry and came to dominate the Beijing government.

Small provinces, powerful contenders

The Fengtian clique, in fact, was not the only one that took advantage of its
geographical isolation to build a highly centralized fiscal and military entity
and complete successfully with other cliques. Another example is Shanxi prov-
ince, which was small in territory but strategically important in national politics
throughout the Republican era. Located in northwest China, Shanxi was
a plateau fenced by the Taihang Mountains in the east and the Yellow River in
the south, while its northern border backed against the Gobi desert, and its
western border was lined with mountains. Historically, the province was
known for being “easy to defend and difficult (for external enemies) to attack”
(yi shou nan gong). For the entire Republican period, it was under the consist-
ent control of warlord Yan Xishan (1883–1960). Yan began his rule in Shanxi
as its military governor right after the Revolution of 1911, and controlled all
of the military forces within the province in 1917. Like all other warlords, Yan
built his power base by vigorously expanding his army, which grew from
merely 7,000 men before 1916 to roughly 100,000 around 1923; he thus
became one of the few local strongmen in North China that had nationwide
influences (Gillin 1967: 19–25).
Key to Yan’s survival and longevity in Shanxi was his ability to establish and
maintain an efficient and highly centralized political and military system across
the entire province, which made it largely free of instability and warfare before
the Japanese invasion. He filled key positions in the military with his trusted fol-
lowers, most originating from Wutai, his home county, or from Shanxi, to
174 The making of a unified and centralized state
a larger extent. But he also paid attention to the merits of government and mili-
tary officials and thus recruited qualified candidates regardless of their geograph-
ical origins; many of them were thus promoted from the rank and file to higher
positions and owed their lifetime indebtedness to Yan. A contemporary thus
described that “the military and administrative circles throughout Shanxi prov-
ince look like a big family, with Mr. Yan as the patriarch and all of the cadres as
his disciples” (Wang Xutian 2000: 61). To put the rural area under his effective
control, Yan promoted the so-called “village-based governance” (cunben zhengzhi),
by which the rural communities were reorganized into administrative villages (bian-
cun), each consisting of roughly 300 houses, who were divided into a number of
neighborhoods or lü, with each lü further divided into five lin and each lin having
five households. Leaders of these organizations shouldered the duties of policing
local residents, collecting taxes and levies, and assisting the government to recruit
soldiers. The goal of the village-based governance was the so-called “combining
farming with warring” (bing nong heyi) or militarizing the rural society in prepar-
ation for war (He Yuan 1998).
Equally important for Yan’s building of an independent administrative and military
system in Shanxi was his efforts to develop the local economy and ensure its sustain-
ability and self-sufficiency. Part of his goal in rural Shanxi was the so-called “six meas-
ures” (liu zheng, namely, water control, tree planting, sericulture, a ban on opium
smoking, a ban on footbinding for women, and pigtail-cutting for men) and “three
matters” (san shi, namely, cotton cultivation, foresting, and husbandry), which aimed
to promote prosperity in the village communities (He Yuan 1998: 243). His govern-
ment invested massively in modern industry and transportation, which later formed
the basis of the Northwestern Industrial Incorporation, a conglomerate founded in
the early 1930s that encompassed a wide range of manufacturers, including mines,
smelters, power plants, and factories producing machinery, chemicals, construction
materials, textile and leather products, and consumer goods. But the most important
and successful project was the famous Taiyuan Arsenal, which was one of the three
major arsenals in China in the 1920s and 1930s (the other two were located in
Hanyang of Wubei province and Shenyang of Fengtian province) and which pro-
duced a wide variety of guns and canons and a large quantity of ammunition. Given
his ability to mobilize the rural population and the capacity of his arsenal, Yan’s army
stood out as one of the three major forces in North China by the end of the 1920s,
the other two being the forces under Zhang Zuolin and Feng Yuxiang, which will be
discussed shortly.
The importance of geopolitical security and centralized control of territory
for the warlords’ survival and strengthening is also seen in yet another case in
Guangxi province. Located at the southwestern border of China and domin-
ated by mountains, Guangxi was never a target for warlords in other parts of
China to compete for, yet its remoteness and relative isolation also offered
the necessary conditions for ambitious local strongmen to build their inde-
pendent domain and use it as a power base to vie for national influences
when they became strong enough. The first military strongman in Republican
Guangxi was Lu Rongting (1859–1928), who ruled the province as its
 Centralized regionalism 175
military governor right after 1911 for ten years and was later remembered by
local people for his ability to weed out bandits and keep the land at peace. At
the peak of his influence, Lu was able to defeat his rival in Guangdong in 1917
and dominate the neighboring province for three years (Mo and Chen 1991, 1:
11–27; Lary 1974: 27–32). Although the limited resources from the poverty-
stricken Guangxi constrained his military buildup and explained in large part – in
addition to the fatal defection by a key commander of his army – his loss in con-
fronting the newly rising forces in Guangdong, the new warlords who replaced him
to rule Guangxi after 1924 turned out to be more successful.
Unlike the warlords in other provinces who tended to fight each other for
exclusive control of a given area, the new strongmen in Guangxi, namely Li
Zongren (1891–1969), Bai Chongxi (1893–1966), and Huang Shaohong
(1895–1966), proved to be exceptionally collaborative with one another in
building a unitary political and military force – in fact, the limited territory and
resources of Guangxi made it unaffordable for any of them to fight with each other;
the best strategy for them to survive, therefore, was to work together as a single,
unified entity.5 Knowing well the limited revenues of the poor province that greatly
curtained their military potentials, the New Guangxi clique adopted an approach to
build their fiscal and military strength that was essentially no different from Yan
Xishan’s methods in Shanxi. Summarized as the “Three-Self Policies” (san zi
zhengce), this approach aimed at three goals. The first was “self-defense” (ziwei) or
military buildup. Lacking enough tax revenues to raise a sizeable standing army,
they chose to militarize society under the so-called “Three Build-in Policies” (san
yu zhengce), namely to build the military by establishing local militia throughout
the province (yu bing yu tuan), to train military officers directly among school stu-
dents (yu jiang yu xue), and to recruit soldiers directly from the registered male
adults of the reorganized villages (yu zheng yu mu). Not surprisingly, after the out-
break of the Marco Polo Bridge Incident in 1937 that signaled the beginning of
a full-scale war of resistance against Japan, Guangxi province turned out to be the
fastest among all the provinces at mobilizing; it recruited enough soldiers to create
four army corps and 48 regiments in a matter of two months (Mo and Chen 1991:
372). The second goal was “self-government” (zizhi), which aimed to establish
a clean and efficient government by purging corrupt officials, training qualified
cadres, and, most importantly, restructuring rural society under the baojia system,
in which the head of a township or a village also served as the school principal and
the militia leader at the same level. To achieve the third goal of “self-sufficiency”
(ziji), the Guangxi clique was committed to investing in a wide range of manufac-
turing, mining, and transportation projects, promoting compulsory and higher
education, and developing forestry and agriculture (Tan Zhaoyi 2009, 2010;
Levich 1993: 65–98).

Winners and losers

The success of the Fengtian, Shanxi, and Guangxi cliques, therefore, lay primar-
ily in their fiscal strength; they used all kinds of methods to generate as much
176 The making of a unified and centralized state
revenue as they could. In addition to the traditional methods of taxing the land
and commercial goods, they all made systematic efforts to invest in modern
industry and a transportation system, and they all utilized modern financial tools
for additional incomes. But the sharp contrasts between Fengtian and the other
two cliques in geographic and economic sizes meant that there were significant
differences between them regarding their respective approaches to state and
military building. The rapid development of modern industry and transportation
and the large amount of industrial output in the vast area of Manchuria made it
possible for the Fengtian clique to rely mainly on the taxes from commercial
goods and commerce rather than the taxes on the land. As a result, its budgeted
revenue from the taxes on goods and other non-agricultural sectors amounted
to 19.6 million yuan in 1925, or more than 2.5 times its revenue from land tax-
ation (Table 7.2). In sharp contrast, despite their efforts to industrialize, the
economies in Shanxi and Guangxi remained largely agricultural. The revenue
from goods and other non-agricultural sectors in Guangxi, therefore, was only
about 73% of its land taxes; and in Shanxi, only 23% (Table 7.2). Even more
striking was the contrast between them in the absolute amount of annual rev-
enue. The three provinces of Fengtian clique had a total of 27.3 million yuan in
1925, whereas the Shanxi clique had only about 7.3 million yuan and the
Guangxi clique had even less at 4.1 million yuan (Table 7.2).6 Not surprisingly,
the Fengtian clique was able to spend more on its military, and its total budget
in this regard amounted to more than 21 million yuan in 1925, which was
roughly three times the military spending in Shanxi and Guangxi (Table 7.2).
For the two smaller cliques, in the absence of sufficient funding of the military,
a better, more cost-saving approach to military building, therefore, was to mili-
tarize society by reorganizing the rural population into strictly controlled units,
establishing militia widely, and making it readily available for mobilization and
recruitment.
In sharp contrast with the winners are the warlord cliques or provinces that
failed because of their lack of advantages that the former had. The first is war-
lord Feng Yuxiang (1882–1948), whose forces grew to more than 400,000 men
and dominated large areas of North and Northwest China at the peak of his
career in the late 1920s. Feng’s biggest weakness, however, was his lack of
a stable and solvent base area to support his army. He thus faced exceptional
and constant difficulties feeding the soldiers, despite the various efforts tried,
such as retaining the salt taxes, raising the rates of railroad shipments of cargo,
and selling public debts by force in its occupied areas (Liu Jingzhong 2004:
79–82, 104–105, 275, 316, 356–357). The reason Feng was able to expand his
forces and survive for more than a decade until the eventual collapse of his
clique in 1930 had to do in part with his highly speculative strategies that led to
his frequent defection, alliance, breakup, and realignment with other competing
cliques, and in part with the generous support from the Soviet Union after
1925.7 After the Soviet Union cut off its supplies, Feng’s competitiveness
quickly deteriorated. In addition to its fiscal insolvency, equally fatal to his
clique was that Feng never made serious efforts to centralize his control of the
 Centralized regionalism 177
various corps, divisions, and brigades he created, due to the scatteredness of
their occupied areas where the individual forces had grown (Liu Jingzhong
2004: 342, 397–400).
A second example that contrasted with the successful cliques is the warlords
in Sichuan province. Sichuan, in fact, had the perfect conditions to grow
a powerful warlord competing nationally. Located at the southwest frontier, it is
relatively isolated from other provinces by the mountains surrounding it, while
the vast basin area within it was one of the most fertile in China, giving rise to
a highly developed agriculture and a high density of population. Therefore,
among all of the provinces in China, Sichuan had the largest area of cultivated
land (151 million mu) and the largest population (47 million). The govern-
ment’s budgeted annual revenue in this province was consistently the second
highest in the entire country, around 12.5 million taels in the late 1910s and
1920s, after Jiangsu province (Table 7.2; Jia Shiyi 1932: 139).
Precisely because it was so large in size and so important as a huge source of
revenue and soldiers, however, warlords inside and outside the province com-
peted for a piece of it, yet no one was able to control the entire province alone.
Thus, after years of rivalry and chaos following the Revolution of 1911, by 1918
and 1919 the province had become extremely fragmented as a result of the rise of
the so-called “garrison-area system” (fangqu zhi) in the province. Under this
system, the province was divided into 15 or more garrison areas, each containing
a number of counties (varying from nine or ten to as many as 33 counties), and
the military force in each garrison area was responsible for supporting itself with
the revenue from within the area. Most of the garrison areas in southern Sichuan
and around the city of Chengdu fell into the hands of warlord forces from the
neighboring Yunnan and Guizhou provinces, while the rest of the garrison areas
belonged to local Sichuan forces. Despite the existence of a provincial govern-
ment, the warlord force of each garrison area turned the territory it occupied into
an independent domain subject to its exclusive control by appointing its own gov-
ernment officials, collecting taxes, and retaining the revenues that should have
been remitted to the provincial or central government (Kuang and Yang 1991:
92–99; Kapp 1973: 34–38). To enlarge or protect their garrison areas, the war-
lord forces fought with each other year after year; therefore, military spending in
Sichuan was also the highest in the entire country, reaching more than 26 million
yuan in 1925, or more than twice its budgeted revenues for the same year (Table
7.2). The division and chaos in Sichuan persisted until 1935, when the province
came under the unified control of the Nationalist government (Kuang and Yang
1991: 451–457; Kapp 1973: 136–141). Before that point, administrative and mili-
tary fragmentation excluded the possibility of any warlord from Sichuan playing
a role in national politics commensurate with its wealth and population.
Finally, let us look at Jiangsu, the richest province in imperial and Republican
China. In the late 1910s and early 1920s, its budgeted revenue was more than
16 million yuan, well above all other provinces, and its real revenue was not too
different from the budget, at about 15 million yuan a year, after the maritime
customs and salt taxes, totaling more than 20 million yuan, were diverted from
178 The making of a unified and centralized state
the province (Shen Jiarong 1993: 292). But two factors prevented the province
from becoming a powerful military contender in the country. The first was geo-
political. Located in the lower Yangzi region and dominated by plains, the prov-
ince is open to all other provinces without any geographic barrier to separate
and protect it. As the most prosperous area in China, it was the target for exter-
nal warlords to complete for control, but there was no single warlord who
could rule Jiangsu for a prolonged time because of the fierce competition
among the warlords of different cliques for this area.
A second factor that made the warlords’ centralized control of Jiangsu diffi-
cult was the existence of a strong gentry-merchant class in this province and
its resistance to the warlords’ excessive extraction of revenues from the prov-
ince by controlling the provincial assembly. Throughout the early Republican
years to 1927, the warlords who ruled Jiangsu consistently paid particular
attention to the public opinion of the local elites in order to maintain their
legitimacy and to yield from time to time to the demand from the latter “the
separation of the military from the civilian” (junmin fenzhi) and “government
of Jiangsu by the people of Jiangsu” (Su ren zhi Su), which meant preventing
the military governor from interfering in the administrative affairs of the prov-
ince, especially the appointment of key positions of the provincial govern-
ment, including the governor and the head of the financial department,
which they insisted should be filled by natives of Jiangsu, rather than by the
warlords’ men (Jiangsu sheng zhengxie wenshi ziliao weiyuanhui 1993: 102;
Chen and Shen 2009). The strong resistance from the gentry merchant class
in Jiangsu greatly limited the room for the warlords to extract local resources
and expand their forces; it also accounted for the fact that, despite its highest
level of revenue among all provinces in China, budgeted military spending
was limited to only 3.9 million yuan or 23% of its revenue in 1919 and
6.1 million yuan or 36% of its revenue in 1925 (or 23% of the military spending
in Sichuan in the same year), whereas in the country as a whole, military spend-
ing was 46% and 105% of government revenue in 1919 and 1925, respectively
(Table 7.2; Jia Shiyi 1932: 140).
All these instances suggest that each of the three factors (namely geopolitical
setting, socioeconomic conditions, and fiscal-military organization) is indispens-
able for any regional force to build its competitive advantage. While stable and
secured control of a given region served as the very precondition for long-term
military buildup, socioeconomic conditions of the region determined the nature
and scope of economic and financial resources available for extraction. Whether
or not such resources could be turned from potential into real competitive
strength, however, ultimately depended on the regional force’s ability to build
a centralized and unified fiscal apparatus.

Why did the Nationalist force win?

Now let us look at the situation in Guangdong, a southern province that
was only next to Jiangsu in terms of economic prosperity and government
 Centralized regionalism 179
revenue. Until his death in March 1925, Sun Yat-sen, the Grand Marshall
of the Military Government from August 1917 and later the “extraordinary
president” of the Republic after April 1921 that ruled only Guangdong and
Guangxi, struggled to build an army strong enough to eliminate the warlords
and unify the country; but he failed repeatedly (Wilbur 1976: 29–35; Bergère
1998: 247, 296). He never expected, and indeed no one else could anticipate,
that only a year after his death the Nationalist Revolutionary Army’s Northern
Expedition would begin, and, even more surprisingly, defeat the warlords in
the middle and lower Yangzi regions in less than ten months, wipe out the
Fengtian-clique forces from the interior provinces, and overthrow the Beijing
government in June 1928. The unification of China, which Sun could only
dream of, came true in December 1928 when Zhang Xueliang (1901–2001),
the warlord of Manchuria, declared his allegiance to the newly established
Nationalist government in Nanjing.
The Nationalist army was unlike any of the warlord forces in that it emphasized
ideological indoctrination among the soldiers and that military officers at different
levels were under the control of the Nationalist Party, hence known as the “party
army” (dangjun). It thus exhibited an unusual degree of solidarity and high
morale when compared to the warlord forces (Wilbur 1976: 220–221, 231–232,
244; McCord 1993: 313–314). But the role of nationalist propaganda in the
building up of the Nationalists’ military should not be overemphasized. There
were, in fact, significant continuities between the Beijing government and the
Nationalist state in their commitments to nationalism and the subsequent foreign
policies (Waldron 1995: 273–275). The most important reason behind the
Nationalists’ triumph over the warlords, as shown below, is its unparalleled fiscal
strength, which was developed through three key steps. The first was of course the
making of a solid fiscal-military state in Guangdong, which enabled the Guomin-
dang force to launch the Northern Expedition and occupy most of South China
and the middle and lower Yangzi regions in a few months. The second was its alli-
ance with the financiers in Shanghai after occupying the city, which made it pos-
sible for the Guomindang force to dramatically increase its revenues by borrowing
from the bankers and selling government bonds. The revenues thus generated
fueled the Guomindang’s continuous Northern Expedition in North China, which
finished in June 1928 when its troops occupied Beijing. The third step was its
decision to restore China’s autonomy in maritime customs in December 1928
after unifying the country, which quickly made such customs the most importance
source of the central government’s revenue (accounting for about 60% of its total
income in the late 1920s and early 1930s; see Table 7.1).

Guangdong and the Northern Expedition

The reasons behind Sun’s frustrations lay partly in his reliance on warlord forces
in the absence of a military force under his own control. He first sought support
from the warlords of Guangxi, who however came to control the military gov-
ernment and turned Sun into a figurehead; later he turned to warlord Chen
180 The making of a unified and centralized state
Jiongming (1878–1933), who nevertheless firmly opposed Sun’s idea of
a Northern Expedition and only wanted to build his stronghold locally. Sun also
attempted in late 1922 to ally with the Fengtian clique to attack the Zhili
clique, and later with Feng Yuxiang, after Feng defeated warlord Wu Peifu in
October 1924 and terminated the Zhili clique’s control of the Beijing govern-
ment. It was on his trip to Beijing at the invitation of Feng and other strong-
men while seeking the unification of China that Sun died, with a reminder to
his followers that “the revolution has not been successfully concluded” (Wilbur
1976: 278).
A more fundamental reason leading to Sun’s failures, however, was the inabil-
ity of his government to generate enough revenues before 1925, which in turn
had to do with the fragmented political map of Guangdong in the early 1920s.
When he came back to Guangzhou in February 1923 after defeating Chen
Jiongming with the help of the Guangxi and Yunan armies, Sun found that the
only place where his orders were effective was Guangzhou itself, and the rest of
Guangdong remained in the hands of either Chen’s forces or the troops from
Yunnan, Guangxi, Hunan, or as far as Henan. As a result, his government only
collected 4.6 million yuan as its revenue in the first half of 1924 mostly by farm-
ing out tax collection to merchants, and the financial ministers of his govern-
ment stepped down one after another because of the huge difficulties in raising
enough revenue (Qin Qingjun 1982; Wilbur 1968: 227–229).
A turning point in Sun’s military buildup was the financial aid and military
supplies that he received from Soviet Russia after May 1923, which included
a loan of 2 million rubles in 1924, various aid to the amount of 2.8 million
rubles in 1925, and at least 2.84 million rubles in 1926 (Zhu Hong 2007).
The Soviet aid made it possible for Sun to establish the Whampoa Military
Academy in May 1924 and to deploy his newly established forces, including
students from the academy, for two “Eastern Expeditions” in February and
October 1925, which eventually subjugated the forces of Chen Jiongming and
other local warlords and brought the entire province under the Nationalist
government.
The two years following the unification of Guangdong saw the explosion of
the Nationalist government’s revenue. Its tax incomes in 1925, for example,
increased from about 4 million yuan in the first half of the year to 12.2 million
yuan in the second half, when the Nationalist government expanded its control
of the province quickly, thus bringing its annual tax revenue in the entire year
to more than 16 million yuan, which nearly doubled its 1924 level (8.6 million
yuan). The growth of the Nationalist government’s tax revenues in Guangdong
was even more dramatic in the following two years, to 69 million yuan in 1926
and 122.6 million yuan in 1927, which was more than 14 times its 1924 level
and 3.3 times Guangdong province’s annual revenue (37.4 million taels) in the
last years of the Qing (Qin Qingjun 1982: 169, 171).8 Therefore, the province’s
unification alone cannot explain the skyrocketing of its revenue. More important
to Guangdong’s fiscal strength were the measures taken by Song Ziwen (Tse-ven
Soong or T. V. Soong) (1894–1971), a Ph.D. in economics from Columbia
 Centralized regionalism 181
University, who began his service as the financial minister of the Nationalist
government and the head of the financial department of Guangdong province
in September 1925 (Wu Jingping 1992: 5–34). The measures he implemented
fell into the following categories:

(1) Centralizing the collection and expending of government revenues, such as:
• taking over the power of tax collection from military forces in the local-
ities where they were stationed;
• terminating the practice of farming out tax collection to merchants;
• prohibiting the retaining of tax monies by the military or any civilian
institutions; and
• dispatching personnel to the newly occupied areas to establish local
financial offices directly answering to the provincial government.

(2) Bureaucratizing the tax collecting agencies, such as:

• incorporating the preexisting separate institutions for collecting taxes on
stamps, gambling, and opium prohibition into a unified system under the
financial ministry;
• creating the statistics department within the ministry to enhance its
accounting and auditing abilities;
• building a province-wide force against the smuggling of commodities;
and most importantly
• recruiting civil servants of the ministry through examinations open to
the public and punishing their involvement in corruption.

(3) Consolidating the major sources of tax revenues, such as:

• reorganizing the channels for stamp sales, extending the collection of
stamp taxes to alcohol and firecrackers, standardizing the use of stamps
on all commodities (together, these measures brought the annual income
from stamp taxes from 0.6 million to 3.04 million yuan in 1926);
• collecting a special tax on kerosene (which generated more than
2 million yuan in the second half of 1926);
• surveying and taxing the sand fields along the coast (which resulted in
an increase by more than a million yuan annually);
• monopolizing the sales of opium through the prohibition of its smug-
gling (as a result, the six-month income from this source increased
from 2 million yuan in 1925 to more than 9 million yuan in the
following year);
• surveying and auditing the collection of lijin on all domestic commod-
ities and increasing its collection rate by 20% in January 1926 and
an additional 30% in the following month (the annual revenue from
this source alone increased by a factor of 2 to nearly 16 million
yuan in 1926);
182 The making of a unified and centralized state
• collecting a “domestic tax” on all goods at the rate of 2.5% on ordinary
items and 5% on luxury items (which began in late 1926 and generated
about 5 million yuan annually); and
• selling public debts (totaling 24.28 million yuan by September 1926)
instead of issuing extra amounts of paper currency by the Central Bank
to safeguard its creditability and avoid runaway price inflation (MGDA,
4: 1400–1404; Qin Qingjun 1982).

This is an impressive list. Together, these measures produced a centralized,

efficient bureaucracy able to fully tap the financial resources of a province that
was only next to Jiangsu in terms of economic prosperity and even more developed
in domestic and foreign trading than the latter. The results of these measures were
surprising. In a matter of only two years since Song’s tenure as the financial minis-
ter, Guangdong province’s annual tax revenue multiplied to more than 122 million
yuan by 1927 as noted earlier, which made the Nationalist force the financially
most solvent among all competing military forces throughout the country.
For years, the Fengtian clique had been the most affluent force, with an
annual revenue of more than 27 million yuan from its three Manchurian prov-
inces in the mid-1920s, which militarily made the clique the most competitive
among all warlord forces, which had no challengers in North China after defeat-
ing the Zhili clique and controlling the Beijing government in 1924. But the
rise of the Nationalist force in Guangdong changed this situation, and what
backed the Nationalist force’s meteoric rise was the building of a centralized
and efficient fiscal machine engineered by Song Ziwen that mobilized the prov-
ince’s financial resources to its fullest extent.
Song’s financial policies in Guangdong were exorbitant. The lijin and “mis-
cellaneous taxes,” which accounted for the largest portion of the province’s rev-
enue, covered almost all kinds of goods and services at a collection rate unseen
before. Li Zongren, who ruled Guangxi after 1925 and later joined the North-
ern Expedition, thus honestly described Song’s measures in Guangdong as
“draining the pond to catch all the fish” (jiezeeryu) and “ruthless taxation and
extortionate collection” (hengzhengbaolian) (LZR: 244). Nevertheless, Song’s
policies worked. The ample revenue they generated provided an indispensable
fiscal base on which the Nationalist force started the Northern Expedition in
June 1926 and succeeded swiftly on the battleground. According to Song
Ziwen’s report on the Nationalist government’s fiscal condition during the year
from October 1925 to September 1926, the total revenue of the Nationalist
government in Guangdong reached 80.2 million yuan, and its expenditure also
expanded to 78.3 million yuan, of which 61.3 million yuan (78.3%) was spent
on the military (MGDA 4: 1400). Until as late as November 1926 when the
Nationalist force had fully occupied Hunan, Jiangxi, Fujian, and Hubei province
and were ready to march up to Henan province and down to the lower Yangzi
region, Guangdong province was the only source from which the Nationalist
force derived its revenue. And the monthly spending on the military and war
increased 7.3 million yuan as the Northern Expedition extended into Henan,
 Centralized regionalism 183
Zhejiang, and Jiangsu provinces (MGDA 1.1: 518). As Song proudly stated, “as
the revolutionary base, Guangdong province alone shouldered the expenses of
a countrywide revolution in China. The recent military campaigns across central
China as well as the cost of the Northern Expedition have all relied on Guang-
dong” (Wu Jingping 1992: 43). The success of the Northern Expedition, to be
sure, had to do with multiple factors, including the high morale and strict dis-
cipline of the Nationalist forces and popular support to them, as well as the col-
laboration of the forces from Guangxi and Feng Yuxiang’s army from the
northwest and, on the other hand, the lack of coordination among the warlords
in the middle and lower Yangzi regions. But key to the Nationalist force’s com-
bating capabilities and triumph on the battleground were its quick expansion
from 130,000 soldiers at the beginning of the expedition to 550,000 men by
early 1927, the ample supply of weapons and ammunition for it, and generous
stipends for its soldiers (Zeng Xianlin et al. 1991: 73, 197) – all these would
not happen without the unparalleled revenue from Guangdong province that
sustained the Nationalist force until early 1927.

The plight of the Nationalist government in Wuhan

The Nationalist forces’ control of Hubei in late 1926 and, more importantly,
their occupation of Shanghai in March 1927 changed the way the Northern
Expedition was fueled. Shortly after the relocation of the Nationalist govern-
ment from Guangzhou to Wuhan in January 1927, Hubei replaced Guangdong
as the primary source of its revenue. In the view of Financial Minister Song
Ziwen, Guangdong had been “extremely vulnerable and completely exhausted”
(kongxu ji yi, luojue ju qiong) after supporting the Northern Expedition for
a year; it was imperative, therefore, for other provinces that had come under the
Nationalist government to share the financial burden with Guangdong (Wu
Jingping 1992: 43, 61). With Guangdong no longer a viable major source of its
revenue and other provinces reluctant to contribute their revenues to it before
financial unification in the occupied areas, the Nationalist government in Wuhan
had to rely on Hubei province as the major source of its tax incomes, and Song
tried some of the measures in Hubei that he had successfully employed in
Guangdong to maximize his government’s revenues. Unfortunately, Hubei was
very different. Having endured successive wars between different forces for more
than a decade, the economy in the province was yet to be recovered from pro-
longed recession; thus, despite the same measures enforced in Hubei as in
Guangdong, the Nationalist government in Wuhan was only able to collect
about 19 million yuan as its revenue from various taxes from September 1926
to September 1927, which was only about 34% of the annual revenue that the
Guangdong government had generated through taxation prior to Septem-
ber 1926 (Jia Shiyi 1932: 114–116). As a result, the Wuhan government had to
rely on issuing public debt, borrowing from banks, and printing paper currency
as the major sources of its revenue (accounting for 84.5% of its total revenue
during the same period) (Wu Jingping 1992: 43–47).
184 The making of a unified and centralized state
Having not firmly controlled the province, however, the Nationalist govern-
ment in Wuhan found it difficult to obtain collaboration from local business and
financial elites. Its growing tension with the Nationalist force in Nanjing and
the subsequent embargo by the latter only worsened its insolvency. The Wuhan
government thus had to introduce the radical policy of freezing the outflow of
cashes from Wuhan and banning the use of currencies other than the paper
notes it printed, which only alienated local business elites and caused runaway
price inflation and further economic recession (MGDA 4: 1485–1486; Yang
Tianshi 1996: 420–427). Having lost its financial creditability, however, the
Wuhan government found it increasingly difficult to raise funds through public
borrowing or printing paper notes, and it would soon give up in its competition
with both the warlords in North China and the Nationalist forces in the lower
Yangzi region.

Reenergizing the Northern Expedition

The schism between the left and right wings among the Nationalists started well
before the Northern Expedition. By and large, the leftists assembled around
Wang Jingwei (1883–1944), head of the Nationalist government in Guangdong
from July 1925 and later the president of the Nationalist Government in Wuhan
after his trip from France in April 1927, who insisted on collaborating with the
Chinese Communist Party (CCP) in his government and the military, while
Chiang Kai-shek, the generalissimo of the Nationalist army, led the rightist
group that opposed the CCP’s radicalism and the Guomindang’s united front
with it. The tension between the two sides mounted in the course of the North-
ern Expedition, especially after the occupation of Shanghai in March 1927, and
developed into open confrontation between the Nationalist government in
Wuhan and another Nationalist government started by Chiang in Nanjing on
April 18, 1927, who ruthlessly purged the Communists from the cities he con-
trolled. Chiang eventually prevailed over the left-wing Nationalists, after a few
months of power consolidation within the Guomindang in late 1927 that
allowed him to resume his position as the generalissimo in January 1928 and to
restart the Northern Expedition three months later (Taylor 2009: 77–78).
The most important reason behind Chiang’s rise as the new leader of the
Nationalists and his success in competing with the left-wing opponents within
the Guomindang and defeating the warlords in North China during the second
phase of the Northern Expedition was his control of Shanghai and the lower
Yangzi region, the most prosperous area of the country, which allowed him to
take advantage of the abundance of financial resources in this area.
Chiang’s personal background as a native of Zhejiang and his earlier career in
1920–1921 as a broker for Shanghai Securities and Commodity Exchange run
by Yu Qiaqing (1867–1945), once a leader of the Shanghai Chamber of Com-
merce, were key to his success in connecting himself with the financiers in
Shanghai, who mostly originated from his home province. To collaborate with
Chiang, Yu soon created the Shanghai Association of Commerce in March 1927
 Centralized regionalism 185
to replace the disorganized Shanghai Chamber of Commerce that had sided
with warlord Sun Chuanfang and to show his open support to Chiang’s Nation-
alist force. After his arrival in Shanghai, Chiang took action to purge the Com-
munists by arresting and killing hundreds of them. Two days after establishing
the Nationalist government in Nanjing in April 1927, Chiang officially set up
the Jiangsu and Shanghai Financial Committee, with 10 of its 15 members
appointed from the business and financial leaders in Shanghai and Jiangsu, and
headed by Chen Guangfu (K. P. Chen, 1880–1976), president of the Shanghai
Banking Association, whose primary duty was to raise funds for the Nationalist
government. In exchange, Chiang granted the committee the complete
authority in appointing and managing all of the financial institutions of the
Nationalist government (Wu Jingping 1992: 58–59). The committee’s first
action was to make a short-term loan of 3 million yuan and another loan of
3 million yuan for Chiang in April 1927 (SSL: 57–59). A bigger step it took
was to assist the Nationalist government in issuing 30 million yuan of govern-
ment bonds in May 1927, to be guaranteed by the government’s additional
revenue in the future through a rising of maritime customs by 2.5% (MGDA
5.1.1: 4–5); to that end, another committee consisting of 14 members, mostly
business leaders from Jiangsu, Zhejiang, and Shanghai, was created to manage
the funds raised through the bonds (SSL: 74–75). In October 1927, the
Nationalist government issued another 24 million yuan of government bonds,
again guaranteed by the government’s revenue from increased maritime customs
(MGDA 5.1.1: 521–522).
Selling the large amounts of government bonds, however, turned out to be
a huge challenge to both Chiang and buyers. For Chiang, raising enough
money in a timely manner through the sales of government bonds was “where
the lifeblood of all the affairs of the army, the government, and the party lie”;
so desperate was his need of funds for the newly established government in
Nanjing and the consolidation of his military control in Jiangsu and Zhejiang
that he pleaded for help in his letter to Chen Guangfu in May 1927 with these
words: “the life and death of the party and the state, as well as the glory and
shame of the nation, all depend on this [the sales of government bonds]” (SSL:
110). The bankers and business owners, for their part, tried to reduce or delay
under various excuses their purchase of the bonds that had been apportioned
among them. Chiang, therefore, repeatedly turned to compulsory measures.9
Such measures could be temporarily effective in obtaining the funds he wanted,
but could not be counted as a regular means to generate the required money.
Therefore, the sales of another 40 million yuan of bonds in October 1927 was
exceptionally sluggish when the purchase of them was intended to be voluntary.
Trapped in his prolonged confrontation with the Nationalist government in
Wuhan, his defeat by the northern warlord force in Xuzhou, and the retreat of
the Nationalist army back to the south of the Yangzi River, Chiang stepped
down from his generalissimo position in August 1927, to allow time for
a reconciliation between Wuhan and Nanjing and for his own rebuilding of per-
sonal connections with financial leaders in Shanghai (Ibid.: 32–35).
186 The making of a unified and centralized state
As anticipated, Chiang came back to power in January 1928 when he faced
a new situation: the Nationalist government in Wuhan had been relocated to
Nanjing to join the Nationalist government there, Chiang himself had married
Soong Meiling in December 1927 and thus become Song Ziwen’s brother-in-
law, and most importantly, Song Ziwen had accepted the position of the Minis-
ter of Finance in the Nanjing government right before Chiang’s resumption of
the generalissimo position. With Song’s help, the Nanjing government’s fiscal
situation quickly improved in 1928. In a matter of six months since he began
serving as the financial minister of the Nanjing government, Song raised a total
of 190 million yuan(!) through acquiring bank loans, sales of government
bonds, and taxation (MGDA, 5.1.1: 520–522). The generous funding made it
possible for Chiang to restart the Northern Expedition and launch a full-scale
campaign against the Fengtian-clique forces in the northern provinces in
April 1928. During the height of the war, Song was required to provide
1.6 million yuan every five days. He did the job and, in fact, generated much
more than the required amount (Jia Shiyi 1932: 216).
The Northern Expedition culminated in the Nationalist force’s occupation of
Beijing in June. The Nationalist government in Nanjing thus formally declared
the unification of China on June 15, 1928. This, however, reflected more of the
Nationalists’ political determination than the reality in the country, because the
Fengtian clique, which had withdrawn from Beijing, still controlled the three
northeastern provinces. To march into the vast area of Manchuria and defeat
the military forces of the Fengtian clique would be the most challenging task for
the Nationalists. Fortunately, after Zhang Zuolin died of an explosion planned
by the authority of the Japanese Kwantung Army on June 4, 1928, and who
had resisted the Japanese pressing for further control of Manchuria, his son
Zhang Xueliang, as the new leader of the Fengtian clique, accepted the leader-
ship of the Nanjing government on December 29, 1928 and thus made all of
the Chinese provinces officially unified (Taylor 2009: 84).

China as a latecomer: from regional to national in state-making

To sum up, what propelled the rise of regional fiscal-military states in Repub-
lican China was the dynamics of centralized regionalism. This originated from
the regionalized centralism of the late Qing period; one can easily identify the
various direct or indirect links between the regionalization of military, fiscal, and
administrative power after the 1850s on the one hand and the rise of provincial
strongmen and warlord cliques in the early Republican years on the other. How-
ever, the centralized regionalism of the early Republican era was essentially dif-
ferent from the regionalized centralism of the late Qing decades. First, the
regional powers of the early Republican years were highly independent of the
central government’s military, fiscal, and even administrative systems, and their
independence was often openly justified; the military strongmen of different
regions used the rhetoric or concepts they invented to legitimize the various
civil and military programs in areas under their exclusive control, unlike the
 Centralized regionalism 187
provincial governors or governors general of the late Qing period who had to
keep their informal or non-statutory practices in generating and spending extra
revenues in a hidden or disguised manner, and whose regionalized measures
usually appeared as expedient solutions to the crises that confronted them.
Second and more importantly, the relationship between the central govern-
ment and regional forces in early Republican years was fundamentally different
from that in the late Qing period. Under regionalized centralism, the Qing
court, though unable to put regional resources and civil or military positions
(especially those of the newly created institutions) under its complete control,
effectively dominated individual provinces by making or revoking appoint-
ments to key positions at the provincial level. It also had the ultimate power to
reallocate the revenues generated and officially reported by provincial author-
ities and, most importantly, to deploy all of the military units stationed
throughout the country, a situation that remained true until the last years of
the Qing. The provincial civil and military leaders could only exercise their dis-
cretion and autonomy within the scope allowed by the center; by no means
could they openly challenge the authority and legality of the imperial court.
Centralism, in a word, was the most defining characteristic of the relationship
between the central and regional authorities in the late Qing period; regional
autonomy existed and functioned within, rather than outside, the framework
of a highly centralized government system.
In sharp contrast, the regional strongmen and warlord leaders of the early
Republican years had complete control of the military and fiscal resources as
well as the administrative system within the areas they occupied. Instead of sub-
mitting to the authority of the central government, they often openly challenged
the latter and even used violence to oust their opponents from the top positions
of the government; they were essentially a force of regionalism in nature.
Finally, and most importantly, late Qing regionalized centralism, while allowing
the dynasty to survive domestic turmoil and foreign intrusions and even witness
a three-decade restoration, also undermined the central government’s control of
regional resources and institutions; once the Han elites who generated and man-
aged such resources lost their allegiance to the imperial house, the latter would
become extremely fragile, as seen in the last few months of the Qing when the
provincial authorities declared their independence from the imperial court one
after another. The regional leaders of the early Republic, for their part, did not
seek the independence of their provinces from the central government, despite
their defiance of the authority of both Beijing before 1927 or Nanjing after-
wards. Quite the reverse; whenever a warlord launched a military campaign
against his enemy, his excuse was precisely to defend China’s political unity and
national interests (MGDA, 3.2: 4, 6, 63, 65, 74, 79, 180, 276), or to maintain
peace and order in the country (MGDA, 3.2: 6, 74, 181, 276) (see also Kuhn
2002: 134). As a matter of fact, national unification was not merely a rhetoric
disguising the warlords’ abuse of violence for self-aggrandizement; to engage in
war and defeat their competitors was also the best way for the most ambitious
warlords to survive the competition and strengthen themselves.
188 The making of a unified and centralized state
To make sense of the Chinese path to a modern nation-state under the
Republic, let us put it in the larger context of state-making in world history. By
and large, we can distinguish between two different paths to the rise of
a centralized and unified national state. One is found among the firstcomers,
such as England, France, and some other Western and Northwestern European
states. These countries first achieved the goal of expanding and consolidating
their territories as early as the tenth through the thirteenth centuries through
conquest and thereby forming a centralized national monarchy; in the following
centuries, they further consolidated state power by turning the monarch’s indir-
ect rule of the country into direct rule, that is, to eliminate the autonomy of
intermediary forces such as priests, landed aristocrats, urban oligarchs, or privil-
eged warriors while establishing a standing army to replace the mercenaries and
enlarging government bureaucracy to enhance its abilities of tax extraction
and hence military buildup. The other path prevailed among the latecomers to
modern national states, such as Germany and Italy as well as the Asian states
that survived Western colonialism in the nineteenth century. The biggest chal-
lenge for these countries in the course of modern state-making was the existence
of multiple regional states in a fragmented territory and the predominance of, or
direct occupation by, foreign powers over parts of their territories. State-making
for the latecomers, therefore, meant first of all eliminating foreign dominance
and building the territorial foundation for a future, unified state. Thus, instead
of the penetration of a national state down to the regional level as seen among
the firstcomers, here state-making typically unfolded from the regional to the
national level, beginning with the competition among regional fiscal-military
states, each of which aimed to centralize state power and modernize the mili-
tary, and culminating in the rise of the most powerful regional state that estab-
lished its national dominance after defeating all other regional competitors and/
or recovering its territory from foreign occupation. By accident, both Germany
and Italy finished the task of territorial unification in 1871; in the same year, all
of the regional daimyos in Japan returned their territories to the newly estab-
lished central government. For the latecomers, the regional powers (Prussia for
Germany, Piedmont for Italy, and Satsuma-Choshu for Japan) played the most
decisive role in the making of a modern national state.
By and large, China took the bottom-up path of state-making, but it was dif-
ferent from other latecomers in at least three ways. First, it was huge geograph-
ically, which meant that there were necessarily more domestic competitors for
national hegemony than in any other country of small size; defeating the rivals
and achieving the goals of national unification and power centralization was
more difficult for the Chinese state-makers. Second, China was confronted
with a geopolitical environment that was much harsher than the other late-
comers, because of the threats from foreign powers that all appeared to be
stronger and wealthier than China, and, in particular, because of the existence
of a militarily aggressive neighbor (Japan) that turned out to be the biggest
barrier to China’s quest for a centralized, unitary state. Third, unlike Germany
or Italy that had built a strong and centralized regional state before it
 Centralized regionalism 189
embarked on the course of territorial expansion and national unification, the
regional states in early Republican China, for all their efforts at military
buildup, economic reconstruction, and power consolidation at home, remained
in the rudimentary stage of state-making; there was still a long way for each of
them to go before building a strong economy, an efficient bureaucracy, and
a powerful military machine. For the Guomindang state that established its
national dominance in the late 1920s, how to consolidate its rule and turn China
into a truly unified and centralized nation-state remained the most challenging
task in the decades to come.
Nevertheless, the key factors driving the rise of nation-states in early
modern and modern Europe also worked to varying degrees in China.
Among such factors are geopolitical setting, fiscal constitution, and socioeco-
nomic structure. The regional powers from Manchuria and Guangdong, and
to a lesser degree from Shanxi and Guangxi, prevailed over their opponents,
or survived the fierce competition, precisely because of the favorable location
of the region they controlled, the centralized apparatuses they built for mobil-
ization and extraction, and ultimately the socioeconomic conditions that
determine the availability of resources for extraction. In his study of the for-
mation of modern European states, Charles Tilly distinguishes three patterns
in which the economy conditioned state activities, namely, coercion-intensive,
capital-intensive, and capitalized coercion (Tilly 1990; see Chapter 3 for more
details). To some degree, this typology is illuminating for understanding the
regional states in warlord China. In provinces where subsistence agriculture
predominated, the local warlord regime had to rely excessively on coercion
to extract economic resources for military buildup or self-profiteering. In
Sichuan, for instance, the individual warlords of garrison districts competed
for taxing the cultivators by increasing the frequency of collection to as many
as ten to fourteen times a year or collecting in advance the land taxes of
future years to as far as 2008 (Ch’en 1979: 141), but none of them made
serious efforts at state-making at the regional or national level. In sharp con-
trast, the military leaders of Guangdong and Manchuria, by fully tapping local
economic resources, succeeded in turning themselves into the most formid-
able contenders for national dominance. In addition to taxing the land, they
both relied primarily on the revenues from trade, industry, and modern finan-
cing. Their approach to regional state-building, therefore, resembled more or
less the pattern of capitalized coercion in Tilly’s terminology. In the final ana-
lysis, geopolitical and socioeconomic conditions critically shaped the competi-
tiveness of regional warlord forces, but the most decisive factor that
determined the result of their competitions remained the methods whereby
they mobilized the economic and financial resources available from the terri-
tories they controlled and turned them into revenues and hence fiscal-military
strength. Those who enjoyed geopolitical and socioeconomic advantages and
at the same time built highly centralized and unified states were the most
competitive. The Nationalist regime prevailed over all others precisely because
of a combination of all these factors in its state-building efforts by 1928.
190 The making of a unified and centralized state
All in all, the rise of regional fiscal-military states in Manchuria, Guangdong,
and other provinces was the most important breakthrough in the century-long
process of state rebuilding in China; it reversed the trend of regionalization and
fragmentation that had persisted since the mid-nineteenth century, leading China
to the recentralization of state power and the reunification of its political landscape.
The state-making efforts by those regional forces were reminiscent of the most
exciting moments in the courses of state-making in other parts of the world. Man-
churia under the Fengtian clique, for instance, would have likely played a role in
the unification of China similar to Prussia in Germany, had its developments not
been impeded and impaired by Japanese penetration throughout the early Repub-
lican years. The Nationalists’ efforts to establish a strong and healthy fiscal regime
in Guangdong under Financial Minister Song Ziwen, which were central to the
Guomindang’s rise as the most competitive force in South China, echoed in many
ways the measures taken by Cavour (1810–1861), Minister of Finance (and later
the Prime Minister) of Piedmont-Sardinia, which enabled the northern Italian king-
dom to take the lead in unifying Italy. Thus, contrary to the old image of “warlord-
ism” involving incessant warfare, dislocation, chaos, corruption, and all other
symptoms of political decay, the late 1910s and the 1920s was a key link in the
making of the modern Chinese state. It was during this period that the long-term
trend of decentralization and disintegration came to a halt, yielding to the rise of
multiple centralized regional states, which paved the way for further steps of cen-
tralization and unification taken by the Nationalists, and later the Communists, in
the two decades after 1928.

Notes
1 The activities of individual warlords in different parts of the country have been the
subject of many studies in the past (e.g., Sheridan 1966; Gillin 1967; Kapp 1973; Lary
1974; McCormack 1977; Wou 1978; McCord 1993, 2014; Suleski 2002). Positive
signs of development undoubtedly emerged under the warlord regimes, such as indus-
trial growth, the freedom of the press, the flourishing of cultural institutions, the intro-
duction of parliamentary rule despite its overall dysfunctionality, and the impressive
quality of cabinet members, as some studies have documented (Pye 1971; Waldron
1995). Nevertheless, the dominant image in the writings on this period remains the
one of extreme fragmentation and chaos in domestic politics, recurring war and social
disorder, countless taxes on the people, and national humiliations at the hands of for-
eign powers.
2 See, e.g., Ch’i (1976); Nathan (1976); Ch’en (1979); Li Xin and Li Zongyi (1987:
212–246); Lai Xinxia et al. (2000: 596–602). For studies that question the conven-
tional wisdom on decentralization and personalization in the late Qing and early
Republican periods, see Kwang-ching Liu (1974); MacKinnon (1973).
3 But this juxtaposition between the northern and southern forces should be taken with
caution. As it turns out, some of the warlord leaders or regimes in North China dis-
played a clear nationalist stance when handling foreign conflicts for the sake of national
interests or their personal fame (e.g., McCormack 1977: 250; Wou 1978: 29, 39,
262). By contrast, Sun Yat-sen, as the early leader of the Guomindang, actually went
further than the northern warlords in seeking support from Japan at the cost of
China’s sovereignty regardless of his anti-imperialist rhetoric; his successor, Chiang
 Centralized regionalism 191
Kai-shek, was equally dependent on old-style, private relations in building his leadership
and personal control of the Nationalist state (Yang Kuisong 2003; Jin Yilin 2005).
4 For a description of the financial crisis of the Beijing government, see van de Ven
(1996).
5 To some degree, the three strongmen were able to work together also because of the
perfect complementing of their respective personalities. While Li was said to be kind-
hearted, tolerant, and generous, which made him a welcome political leader, Bai was
eloquent, resourceful, and resolute, which made him a superior military commander,
and Huang’s detail-minded and hands-on inclinations made him a good administrator
(Wang Yugui 1996: 76).
6 All these, to be sure, were budgeted figures, but they nevertheless suggest the gap
between them.
7 By August 1926, the Soviet Union provided to Feng’s army 31,500 rifles, 51 million
bullets, 272 machine guns, 60 cannons, 58,000 shells, and 10 aircraft (Liu Jingzhong
2004: 369).
8 One silver tael equaled 1.09 silver yuan in 1910.
9 For instance, to compel the Shanghai branch of the Bank of China to advance
10 million yuan to satisfy the urgent needs of military spending when the sales of the
bonds proceeded slowly, Chiang threatened the branch’s manager, who had financed the
Nationalist government in Wuhan, with the accusation of “blockading the revolution”
(zuai gemeng) and “intending to side with the rebels” (youyi funi) (Wang Zhenghua
2002: 105). He ordered the arrest of the businessmen or their family members and confis-
cated their property under the excuse of their counter-revolutionary activities in assisting
warlords previously, when the latter failed to buy the required amount of bonds (Coble
1986: 32–35).
8 In search of national unity
Frontier rebuilding under the
Republic

Upon the inauguration of the Republic of China on January 1, 1912, provisional

president Sun Yat-sen affirmed in his proclamation the principle of “national
unity” (minzu zhi tongyi):

the foundation of the state lies in uniting the lands of the Han, the Manchus,
the Mongols, the Muslims, and the Tibetans into one country, which also
means uniting the ethnic groups of the Han, the Manchus, the Mongols, the
Muslims, and the Tibetans into one people.

In the same announcement, Sun stated the principle of “territorial unity”

(lingtu zhi tongyi) in response to the announcement by individual provinces of
their “independence” from the Qing court during the Revolution of 1911: “the
so-called independence meant at once separation from the Qing court and unifi-
cation among the provinces, and the same was true for Mongolia and Tibet”
(XZDA, 6: 2346). A few months later, Yuan Shikai, as the formal president of
the Republic, reasserted the idea of “a Republic of Five Races” (MGWJ, 1: 84).
In his communication with the Jebtsundamba Khutuktu, who had declared the
independence of the Mongolian state from the Qing in Kulun on December 1,
1911, President Yuan opposed the independence by appealing to the Qing
court’s Decree of Abdication (MGWJ, 1: 99); Yuan’s stance was backed by his
understanding of the history of Mongolia’s integration with China:

Outer Mongolia belongs to the same Chinese nation and has been of the
same family for hundreds of years. There is absolutely no ground for separ-
ation at this moment when state affairs are in danger and the frontier issues
become increasingly thorny.
(MGWJ, 1: 86)

This was one of the very first official uses of the broadly defined term of “the
Chinese nation” (Zhonghua minzu), which soon came to dominate the discourse
on national unity in twentieth-century China.
The biggest challenge to Beijing in its efforts to maintain China’s territorial
unity came from three powers: (1) Britain, which strove to establish its political
 In search of national unity 193
and military supremacy in Tibet and use Tibet as a buffer to fend off the poten-
tial threat to British India from a unified and modernized China that was likely
to emerge after the Revolution of 1911; (2) Russia, which aimed to turn Outer
Mongolia into an independent state and use it as a buffer to enhance its geopol-
itical security in the Far East; and (3) Japan, which targeted Manchuria and the
eastern part of Inner Mongolia in its ambition for territorial expansion after
annexing Korea in 1910.
The frontier crises took place at a time when China was politically frag-
mented. The preoccupation of the central government with recurrent rebellions
when Yuan was in power and, after his death, the military competition among
provincial warlords for the control of the central government and self-
aggrandizement provided the opportune condition for the thriving of centrifugal
forces in the frontiers. But there also existed favorable conditions for China to
prevent the loss of its frontiers. First of all, as a result of both the increasing
awareness of national survival among the elites in Chinese society and the two
centuries of integrating the frontiers with the interior, the Chinese elites no
longer equated their country with only the 18 provinces by the end of the
Qing; instead, the public generally accepted the broadened concept of China as
a land encompassing both the interior of the Han Chinese and the frontiers of
the non-Han populations (Esherick 2006). Whether or not the government was
able to defend the frontiers and maintain China’s enlarged territory thus directly
affected its legitimacy. The central government as well as regional competitors
invariably came under immense pressure from the public to take action and
defend the frontiers whenever the latter were in jeopardy. No less important was
the fact that the ruling elites of the frontiers were rarely a monolithic entity in
favor of secession from China; they were often divided between those who had
vested interests in the preexisting arrangements between the central government
of China and local religious and administrative institutions and those who chal-
lenged the existing patterns of interests and power relations. Even more
decisive was the geopolitical setting that confronted the frontiers; the changing
strength and strategic goals of the neighboring powers directly affected the
frontier leaders’ decisions on seeking separation and protection from the for-
eign powers or staying within China. The results of China’s efforts to defend
and rebuild its frontiers during the Republican era thus depended on the inter-
play of three factors: the will and ability of China’s domestic leadership, the
attitudes and strength of the ruling elites of the frontiers, and the strategies of
the surrounding foreign powers.
The situation in the frontiers varied, depending on the degree to which they
had been administratively integrated with the interior before the fall of the
Qing. After incorporating Xinjiang into its territory in the 1750s, the Qing
encouraged Han merchants and farmers to migrate to the eastern and northern
parts of this region; it also relied on Han businessmen in southern Xinjiang by
the early nineteenth century as a result of its growing distrust toward native resi-
dents (Millward 1998: 114–117, 225–226). As a province after 1884, Xinjiang
was least troublesome to Beijing, where the government system was in the
194 The making of a unified and centralized state
hands of the Han people after Zuo Zongtang’s expedition in 1875–1878,
which allowed his subordinates, mostly from his home province of Hunan, to
take over local authorities in most parts of the region that later formed the
province. During the rest of the Qing, the governor’s position in Xinjiang was
invariably filled by Han bureaucrats, while the General of Yili, a separate
authority controlled by the Manchus and responsible for local defense and for-
eign affairs, limited its influence only to the areas of Ili and Tarbahatai along
the northwestern border of the province. After 1911, military governor Yang
Zengxin (1864–1928) ruled Xinjiang for 17 years until his death by assassin-
ation. He centralized the administrative system in the province by putting the
military authorities in Yili, Tacheng (Tarbahatai), and Altay under his direct
control between 1916 and 1919 (Gu Bao 2005, 5: 52; Chen and Chen 2007:
71–73). Like warlords in other provinces, Yang dictated the administrative
and military affairs in Xinjiang and made his province highly autonomous from
Beijing, but he was also committed to safeguarding China’s sovereignty over
Xinjiang and never intended to separate his province from the Republic. The
challenge to Beijing’s claim of China’s territorial integrity, therefore, came pri-
marily from Tibet and Outer Mongolia, the two areas that had been relatively
autonomous by and large under the Qing and became even more centrifugal
to the Chinese Republic after 1912.

Under the “suzerainty” of China: Tibet in 1912–1950

Compared to Xinjiang and Mongolia, Tibet saw the least penetration by the
Han population and experienced no substantial integration with the adminis-
trative system of the interior during the Qing. This fact, together with its geo-
graphical isolation from the interior, made it most likely for Tibet to seek
independence from China. Surprisingly, Tibet never officially declared the
founding of an independent state of its own, and throughout the Republican
era, the spiritual and secular leaders of Tibet, as well as the foreign powers
behind them, never officially denied China’s “suzerainty,” as they called it,
over the highland. All these provided a historical ground for the People’s
Republic of China to take action to turn Tibet into one of its “autonomous
regions” and thereby put it under Beijing’s complete administrative and military
control in the 1950s.

The centrifugal tendency

Dissatisfaction among the Tibetan elites with the Qing began with its signing of
two treaties with Britain in 1890 and 1893, which conceded part of Tibet’s land
to Dremojong (Sikkim), Britain’s protectorate, and opened Yadong, a Tibetan
locality next to Dremojong, to British subjects for trade. But the growing Brit-
ish influence also incurred resistance from Tibetan elites under the leadership of
the 13th Dalai Lama, which culminated in an armed defense at Gyangze in
July 1904. The conflict ended in the Tibetan force’s complete defeat, the British
 In search of national unity 195
force’s subsequent occupation of Lhasa, and the Dalai Lama’s escape to Outer
Mongolia. The treaty signed in the same year between Tibet’s acting regent and
British representatives stipulated that Tibet opened two more ports for trade,
dismantled defense facilities along the route to Lhasa, paid Britain a war repar-
ation, and most importantly, denied any foreign countries the right to interfere
with Tibet’s domestic affairs, use its land, or do business unless the same business
rights were granted to Britain. Without the signatures of the Dalai Lama and the
Qing court’s commissioner in Tibet, however, the treaty lacked enough validity.
Another treaty thus was negotiated and signed on April 27 1906 between Britain
and Beijing, which included the 1904 treaty as an appendix but acknowledged
China’s sovereignty over Tibet (Hertslet 1908: 203).
The Qing court, however, failed to win over the Tibetan leaders throughout
the process. Based on a misleading report from Youtai, the Royal Commissioner
in Tibet, who had discouraged the Tibetans’ armed resistance against Britain
but later blamed the Dalai Lama for the military failure, the Qing court deprived
Thubten Gyatso of his Dalai Lama title in 1904. In 1907, the Qing court
reinstalled the Dalai Lama title on Thubten Gyatso upon his visit to Beijing
but rejected his request for the right to directly report to the court and
insisted on the role of the Royal Commissioner in Tibet as the intermediary
between the two sides. The situation worsened after the Dalai Lama’s return
to Tibet in 1909. The entry into Tibet of 2000 poorly trained New Army sol-
diers from Sichuan at the request of Lianyu, the new Royal Commissioner in
Tibet, regardless of the Dalai Lama’s opposition, and their random shooting in
Lhasa, caused the Dalai Lama to flee to India and the Qing court’s termin-
ation of his title again in February 1910.
To redefine Tibet’s status, a conference was held in Simla in northern India
from October 1913 to July 1914, following President Yuan Shikai’s decision to
send more troops to Tibet and “reinstall” the Dalai Lama title on Thubten
Gyatso in October 1912. Each of the three parties at the conference had its
own goals. The Tibetan delegate’s initial and primary purposes were to seek
Tibet’s independence and demarcate its boundary with China’s interior that
would include the entire Qinghai province and the border areas of Sichuan into
its territory. Britain did not openly propose Tibet’s independence from China
but sought to turn Tibet into a buffer state between India and China and estab-
lish its dominance in Tibet by forcing China to retreat from its positions as pre-
scribed in the 1906 treaty and reducing China’s role in Tibet to a nominal
suzerain. Beijing, for its part, wanted to defend its positions by asserting Tibet
to be part of China, maintaining its garrison force of 2,600 men in Lhasa, and
having the ultimate authority over Tibet’s foreign and domestic affairs. Never-
theless, confronted with the more urgent tasks of suppressing the rebelling prov-
inces in South China and in need of recognition by Western powers, especially
Britain, Beijing had to concede to Britain’s requests except for its proposal of
dividing Tibet into the so-called “outer” and “inner” areas. The conference
resulted in the so-called Simla Accord officially signed between the British and
Tibetan delegates, which recognized China’s “suzerainty” over Tibet as well as
196 The making of a unified and centralized state
“the autonomy of Outer Tibet.” The treaty further stipulated that China
“engages not to convert Tibet into a Chinese province” while Britain “engages
not to annex Tibet or any portion of it” (Goldstein 1989: 833). Beijing, how-
ever, refused to ratify this treaty because of strong opposition from within the
government after its details were disclosed, despite the Chinese delegate’s pre-
liminary signing of a draft of the treaty at the end of the conference.1

“Turning inward”
Therefore, Tibet remained in a state of “autonomy” from the Chinese govern-
ment in the following decades without officially declaring its “independence” as
a sovereign state. The actual relationship between Lhasa and Beijing (and Nan-
jing after 1927) varied under different Tibetan leaders. From time to time, the
13th Dalai Lama expressed his willingness to “turn inward” (xiangnei) and
improve his relationship with the Republican government in Beijing and Nan-
jing while distancing himself from Britain and punishing the pro-British elites in
Lhasa. This change in his attitude had to do in part with the splitting of the
Tibetan elites into two factions. The old-generation faction consisted primarily
of the priests and leaders of major monasteries in Lhasa, who favored maintain-
ing the status quo, including Tibet’s traditional ties with Beijing, in order to
preserve their preexisting status and privileges, while opposing the Westernizing
reforms that greatly increased their tax burden. The new-generation faction
comprised secular government leaders, especially those who controlled the
newly created or upgraded military and police forces, and therefore benefitted
from their ties with their British counterparts, and subsequently advocated pro-
British policies and Tibet’s independence from China. The Dalai Lama himself
felt an immediate threat from the young, pro-British military commanders who
planned in vain a coup against him in 1923. To curb the influence of the new
elites, he removed the commanders from their military positions, disbanded the
military training center, shut down the English-language school in Lhasa, and
banned the wearing of Western suits among the Tibetans.
Another factor contributing to the Dalai Lama’s inclination to move closer to
Beijing and later Nanjing was the worsened economic situation in Tibet. The
constant tension and recurrent wars along the border between Tibet and the
interior provinces of Sichuan and Qinghai in the 1910s through the early 1930s
impeded trading activities between the two sides, hence the skyrocketing of the
price of goods imported from the interior. According to Konchok Jungne, who
visited Nanjing in 1929 as the Dalai Lama’s delegate and explained in person to
Chiang Kai-shek the Dalai Lama’s willingness to “turn inward,” the price of tea
rose up to ten times as a result of Tibet’s troubled relationship with the interior
(XZDA, 6: 2481). Unable to procure grain from the neighboring provinces, the
Tibetan troops along the border had to turn to local Tibetan residents for food
supplies and thus threatened their livelihood. As an alternative to trading with
the interior, Tibet turned to Britain and India for imported goods and markets
of its domestic products, but the unfair exchange rate between the British and
 In search of national unity 197
Tibetan currencies caused huge losses on the Tibetan side. The Dalai Lama
pushed for a “turning inward” policy also because of his rivalry with the ninth
Panchen Lama, who fled to the interior in 1924 after he felt that his personal
security was threatened by the Dalai Lama’s hostile actions. To counterbalance
the Panchen Lama’s popularity in the interior, where he received a warm wel-
come from the central and provincial authorities with majestic ceremonies wher-
ever he traveled, it was necessary for the Dalai Lama to promote his profile in
the interior by ameliorating his relationship with the Republican government.
The Dalai Lama’s gestures of “turning inward” began with his interaction
with the delegates sent by the military governor of Gansu Province to Lhasa in
November 1919. The Dalai Lama reportedly told his guests at the farewell ban-
quet that his earlier “pro-British” actions did not reflect his “true intension” but
were responses to the Qing commissioner’s “excessive compulsions.” It was said
that the Dalai Lama also expressed his determination to “sincerely turn inward
and join the effort for the happiness of the Five Races” (qingxin xiangnei, tong-
mou wuzu xingfu) (XZDF: 353). After the establishment of the Nationalist gov-
ernment in Nanjing, the Dalai Lama ordered his aforementioned delegate to
visit Shanxi and Nanjing, where the delegate explained the Dalai Lama’s three
points, namely, “not to befriend the British, not to betray the central govern-
ment, and wish to welcome the Panchen Lama back to Tibet” (XZDA, 6: 2475).
According to Konchok Jungne’s explanations, the Dalai Lama had never sought
to ally with Britain, and he promised that the administrative, military, and diplo-
matic affairs of Tibet would be subject to the central government’s management
in the future and a standing commissioner of the central government would be
allowed to be stationed in Tibet while Tibet should also be allowed sufficient
autonomy. In his conversation with Liu Manqing, an emissary of the Nanjing
government who visited Lhasa from February to May 1930, the Dalai Lama
reportedly made the following remark:

The British indeed intended to seduce me, but I know that our sovereignty
cannot be compromised. The characters and natures [of the British and
Tibetan peoples] are incompatible with each other; therefore, when they
came here, I dealt with them only perfunctorily and did not concede to
them the slightest degree of rights. It is not difficult to solve the question
of Tibet and Xikang peacefully as long as China achieves domestic unity.

Having witnessed “the unspeakable suffering of the Indian people in their

recent resistance against Britain’s extreme exploitation,” the Dalai Lama told
Liu, “What I wish the most is China’s true peaceful unification” and that
border disputes between Tibet and the interior provinces should never be solved
by violence because they “both belong to China’s territory” (XZDF: 357–358).
In response to the invitation by the Nationalist government, the Dalai Lama
sent six formal delegates and three non-voting delegates to attend the National
Congress in Nanjing in May 1931; the Panchen Lama, too, sent four formal
delegates and five non-voting delegates to the same conference.
198 The making of a unified and centralized state
The Panchen Lama, for his part, was in fact the most enthusiastic Tibetan
leader advocating a pro-China policy. In a statement made upon the inaugur-
ation of “the Front Office of the Panchen Lama in the Capital City” in Nanjing
in January 1929, he described the relationship between Tibet and China as
follows:

Tibet’s relations with China have deepened over time since the Han and
Tang dynasties; furthermore, government officials and a garrison troop
[from China] were installed [in Tibet] in the late Qing period. Given their
historic and geographic relations, it is practically impossible for Tibet to
separate from China in search of autonomy; on the other hand, for China
to lose Tibet is just like a vehicle losing its wheels. Therefore, the relation-
ship between China and Tibet is subject to a definite law: united, they both
profit; divided, they both suffer.
(XZDA, 7: 3089)

Eager to go back to Tibet, he accepted the appointment by the Nanjing gov-

ernment as “Commissioner for Propagation and Education in the Western Fron-
tier” (Xichui xuanhua shi) in December 1932, and began his preparation for
the return trip. He also received 1 million yuan from Nanjing and a body guard
of 300 men for that purpose. The 13th Dalai Lama also welcomed his return.
Unfortunately, the return trip never succeeded after the death of the Dalai
Lama in December 1933.2 After the outbreak of a full-scale war of resistance
against Japan in July 1937 and in desperate need of help from Western powers,
the Nationalist government retreated from its earlier position and instructed the
delay of the Panchen’s return trip. The Panchen Lama died in frustration in
December 1937.

The ebb and flow of Chinese influence after 1934

After the death of the 13th Dalai Lama, the fifth Reting Rinpoche (Jamphel
Yeshe Gyaltsen, 1910–1947) was elected as the regent of Tibet in January 1934;
he would remain in this position until his retirement in 1941. The Nanjing gov-
ernment’s influence in Tibet climaxed during his regency. Upon his assumption
as the regent, Reting informed Nanjing of his new job, which was the very first
of such actions by Tibetan leaders ever since the founding of the Republic in
1912, obviously because he sought confirmation of his authority in Tibet from
the Nationalist government, which had recently subjugated all of the regional
warlords within the interior and made the emergence of a unified and strong
China a promising future. Later in August, the Tibetan government received the
delegation of the Nationalist government led by Huang Musong to attend the
funeral of the deceased Dalai Lama with a ceremony comparable to the welcome
event for the Royal Commissioner of the Qing court before 1911. In their con-
versations with Huang, the Tibetan leaders objected to the stationing of Chinese
troops in Lhasa and on their position over Tibet’s border with Xikang, but they
 In search of national unity 199
agreed that “externally speaking, Tibet is part of China’s territory,” that “the
authorities and regulations on Tibet’s domestic and foreign affairs should follow
the instructions of the Chinese government as long as these instructions do not
obstruct the administrative and religious activities [in Tibet],” and that “the
Chinese and Tibetans should work together over the major issues regarding
Tibet’s negotiation and signing of treaties with foreign countries” (XZDA, 6:
2684). The Tibetan government, in other words, continued to defend its auton-
omy but did not deny China’s sovereignty over Tibet.
Another important event during Reting’s regency that involved a delegation
from the Nationalist government was the selection and installation of the 14th
Dalai Lama. After identifying three child candidates, Reting reported to the
Nationalist government and invited a delegate from the latter to participate
in the ceremony to select the new Dalai Lama by drawing lots in accordance
with the established practice during the Qing. In response, the Nationalist
government appointed Wu Zhongxin, head of the Council for Mongolian
and Tibetan Affairs, as its delegate in December 1938. Wu arrived in Lhasa
on January 15, 1940, where he received a warm welcome by the Tibetan
government. By that time, however, Reting had determined the candidate
from Qinghai to be the true reincarnated Dalai Lama. Wu insisted on per-
sonally inspecting the child before permitting the exemption of drawing lots.
All of his activities went on smoothly afterwards until the issue of his seating
at the ceremony of the new Dalai Lama’s reinstallation arose on February 22.
Upon his insistence and with Reting’s support, Wu was seated next to the
young Dalai Lama on his left side, facing the south, rather than seated
together with other guests in front of the Dalai Lama and facing the north.
The British delegate, who would have been seated together with other
guests, did not attend the ceremony in the end. Also because of Reting’s
support, a permanent Office of the Council of Mongolian and Tibetan Affairs
(CMTA) of the Nationalist government was established in Lhasa on April 1
before Wu’s departure two weeks later (XZZL: 297–305).
Tibet’s relationship with the Nationalist government deteriorated after Reting
stepped down and the third Taktra Rinpoche (Ngawang Sungrab Thutob,
1874–1952) became the regent (1941–1950). To push for Tibet’s independ-
ence, Taktra announced the establishment of the “Bureau of Foreign Affairs” in
July 1942, stopped his government’s provisions to the CMTA Office in Tibet,
and informed the office to directly contact his bureau for whatever issues per-
tained to the Han and Tibetan peoples, though the CMTA Office firmly refused
to deal with the bureau, causing the Tibetan government to create a separate
organ in 1943 to communicate with it. A second attempt to claim Tibet’s inde-
pendent status was Taktra’s decision to send a delegation to the Asian Relations
Conference held in New Delhi in March 1947, at the advice of Hugh Edward
Richardson, a representative of the British government in Lhasa. After the Chin-
ese delegation protested, the conference organizer removed the “snow lion flag”
that had been used by the Tibetan Army from the array of flags of participating
countries at the conference and modified the huge map of Asia behind the
200 The making of a unified and centralized state
rostrum that had designated Tibet as a separate country from China (Goldstein
1989: 563–564).
One more effort to show Tibet’s independent gesture internationally was to
send a “business delegation” to travel abroad from December 1947 to
March 1949. What concerned the Tibetan as well as the Nationalist govern-
ments was how the delegation was treated by the destination countries and,
most importantly, how their travel documents were handled. For the Tibetan
delegation, carrying a Tibetan passport accepted by the destination countries
was an unmistakable sign of their acknowledgement of Tibet as an independent
state. Unfortunately, none of the countries the delegation visited officially
endorsed their passports. In response to the Chinese ambassador’s inquiry, the
Indian government replied that by convention it did not require a passport or
any formal procedure for Tibetan visitors to legally enter India. George Mar-
shall, Secretary of State of the United States, assured the Chinese ambassador
that his government fully respected the stance of the Nationalist government
and that its general consulate in Hong Kong, therefore, only issued a permit
of entry to the Tibetan delegation, rather than the standard visa on the pass-
ports of its members. The British government explained to the Chinese
embassy that it treated the delegation’s visit as only “a private commercial
affair, not in any official capacity” (Grunfeld 1996: 89) and that it only
required the use of an affidavit instead of a passport to issue visas to the Tibetan
visitors (Zhou Gu 1990).

An independent Tibet?
The changing relationship between the Tibetan and Republican governments
reflected to some extent the rivalry between Britain and China for domin-
ance of this highland. As it turned out, support from the British government
or its agents in Tibet was key to the pro-British Tibetan elites’ activities in
seeking Tibet’s independence from China. But Britain consistently avoided
advocating Tibet’s formal independence; instead, it was only interested in
turning Tibet into a buffer between India and China and establishing its
dominance there while allowing China nominal suzerainty over Tibet,
a strategy that was in the best interests of Britain’s global objectives. For the
British government, to preserve its political influence and commercial inter-
ests in China and Hong Kong far outweighed the potential interests it could
materialize from an independent Tibetan state. The United States did not
want to offend China by endorsing Tibet’s independence either. It needed
China’s collaboration as an ally during World War II and so continued to
support Nationalist China after the war to safeguard its interests in East Asia
concerning the oncoming conflict with the Soviet Union. The growing
importance of China in global affairs as one of the five standing members of
the Security Council of the United Nations after 1945 also made other
nations, including India, refrain from openly supporting Tibet’s separation
from China.
 In search of national unity 201
The Chinese government, for its part, had not sought actual control of
Tibet’s domestic affairs or to incorporate it into the administrative system of the
interior following the founding of the Republic in 1912. Preoccupied with the
burning tasks of fighting domestic and foreign enemies, the Republican govern-
ment in Beijing and later in Nanjing had neither the willingness nor the ability
to put Tibet under its administrative and military control. Its objective in Tibet,
therefore, was only to prevent the highland from being officially separated from
China and thereby to preserve the latter’s territorial unity, which was central to
its legitimacy as the central government. Understandably, the essential guideline
for the Republican government to manage its relationship with Tibet was to
maintain the status quo or continue the preexisting arrangements it had
inherited from the Qing, as best illustrated by the CMTA’s secret instruction
to its office in Lhasa in 1946: “the biggest task is to avoid the rise of any
tasks, and the biggest achievement is that nothing is achieved” (yi wushi wei
dashi, wugong wei dagong) (WSZL, 79: 130).
As a result, both the Republican government’s efforts to defend its rights in
Tibet and the pro-British Tibetans’ attempts to assert Tibet’s independence
were limited to the symbolic level, as seen in the former’s insistence on its dele-
gate’s seating at the 14th Dalai Lama’s reinstallation and the latter’s use of their
own passports and flags outside Tibet. All in all, three basic facts combined to
form the geopolitical realities in which Tibet defined its relations with China or
other nations in the first half of the twentieth century.
First, despite its consistent efforts to defend China’s sovereignty over Tibet,
the Nationalist government had little control of Tibet’s domestic affairs, includ-
ing the selection and reinstallation of the Dalai Lama; compared to the Qing
court, its presence in Lhasa greatly weakened in the absence of an institutional-
ized authority representing it before 1934, and the CMTA’s office in Lhasa
after 1934 was far less influential than the Qing court’s Royal Commissioner in
Tibet, which was above the secular authority of the Tibetan government before
1911. The Republican state even failed to live up to its role as a “suzerain” as
acknowledged by Britain, given the facts that it did not play a positive role in
shaping Tibet’s external relations and sanctioning its political leadership, as the
word “suzerainty” suggests; its control of Tibet usually took a passive form:
externally, it only aimed to prevent Tibet from developing formal diplomatic
relationship with other nations; domestically, it only wanted to ensure that its
rights inherited from the Qing were respected in Tibet.
Second, despite its substantial autonomy from China, the Tibetan authorities
never succeeded or even seriously attempted to create an independent Tibetan
state during the first half of the twentieth century. To be sure, the 13th Dalai
Lama did announce a proclamation in January 1913 upon returning from India
to Lhasa, which blamed the Chinese provincial authorities for their attempts “to
colonize our territory” (referring to the Sichuan troops’ entrance into Lhasa in
1909), and claimed that “the Chinese intention of colonizing Tibet under the
patron-priest relationship has faded like a rainbow in the sky” (referring to the
Tibetan rebellion and subsequent expulsion of the Chinese troops from Tibet in
202 The making of a unified and centralized state
April 1912 in response to the anti-Qing uprising in Wuchang in October 1911)
and that “we must defend our country” (Shakabpa 1967: 246–247). The Dalai
Lama made this statement in a particular circumstance: the Qing court had
deprived him of the Dalai Lama title twice, forcing him into exile in India; he
was so discontented that he rejected President Yuan Shikai’s proposition to
“restore” the title to him after the fall of the Qing, saying that he “intended to
exercise both temporal and ecclesiastic rule in Tibet” (Goldstein 1989: 59–60).
It is equally clear, however, that the Dalai Lama did not say in this statement
that the “patron-priest relationship” itself had ended (or “faded like a rainbow”);
he referred to the “Chinese intention of colonizing Tibet” or more exactly the
stationing of Chinese troops in Lhasa. Instead, as a Tibet historian notes, he “was
quite happy for the patron-priest ideal to continue” (van Schaik 2011: 285). It is
also important to note that another Tibetan Buddhist leader, the Panchen Lama,
whose influences prevailed over the Tsang region centered in Shigatse and whose
standing was no less important than the Dalai Lama in Tibetan Buddhism,
refused to cooperate with the latter in expelling the Han people (Goldstein 1989:
62–63). The Dalai Lama’s statement, in other words, was not fully representative
of the will of all Tibetan leaders. Most importantly, the Dalai Lama himself later
modified his stance and at least verbally respected the Nanjing government’s
rights over Tibet as shown earlier.
Third, there was no sovereign state in the world that officially acknowledged
Tibet’s independence and established formal diplomatic relations with it, nor
was it accepted as a member of the United Nations after 1945. On the other
hand, China’s rights on Tibet were widely recognized. Not only did Britain offi-
cially and consistently acknowledge China’s “suzerainty” over Tibet, but the
Tibetan leaders, including the 13th Dalai Lama, the ninth Panchen Lama, and
the regent Reting, all acknowledged on specific occasions that Tibet was part of
China’s territory and that the Republican government in Beijing or Nanjing was
the “central government” to Tibet. It was against this background that the rela-
tionship between Tibet and the People’s Republic of China (PRC) evolved after
1949. The only difference before and after 1949 was that the PRC leaders,
equally committed to China’s territorial unity as their Republican predecessors,
not only felt it rightful to defend China’s sovereignty over Tibet, but more
importantly, they had the will and strength to do so. After all, turning Tibet
into an “autonomous region” of the PRC and sending Chinese troops to Lhasa,
for the Communist leaders in the 1950s, was only to restore and continue the
relationship between China and Tibet that had prevailed before 1911 but deteri-
orated afterwards.

From autonomy to independence: Outer Mongolia, 1911–1946

Unrecognized independence
Compared to Tibet, which was geographically isolated and remote from the pol-
itical center of China, Outer Mongolia was closer to Beijing and, therefore,
 In search of national unity 203
subject to the greater impact of the “New Policies” of the Qing during its last
decade. The stronger resentment among the Mongolian religious and political
elites against the New Policies that had impaired their privileges accounted for
their greater inclination to separate from China than their counterparts in Tibet;
this, together with tsarist Russia’s support, explained why the Mongol elites
went further than the Tibetans by formally declaring the independence of Outer
Mongolia from China in late 1911.
Outer Mongolia’s independence was in part a result of local elites’ resistance
to the Qing court’s New Policies. Unlike its policy in Tibet, the Qing court
actively engaged in ruling Outer Mongolia through its Administrative Commis-
sioner (banshi dachen) in Kulun (Niislel Khuree, or Ulan Bator today), who
supervised the activities of the Jebtsundamba Khutughtu (head of the Gelug lin-
eage of Tibetan Buddhism in Mongolia), and its General of Uliastai (Uliasutai)
(Wuliyasutai jiangjun) and the Deputy Commissioner of Kobdo (Kebuduo
canzan dachen), who were in charge of local defense affairs. Among the New
Policies introduced in the 1900s by Sanduo, the last Administrative Commis-
sioner, were the creation of a variety of new government offices and military
units as well as modern schools, police, clinics, goods exhibition halls, post
offices, and telegraph offices. Local people opposed these projects because they
saw little benefit from them while their tax burden greatly increased. Those who
lived on husbandry opposed the policy that encouraged Han people’s migration
and reclamation of pastures into farmland, which only increased local govern-
ments’ income from land taxes and rents (Wang Bingming 1990). Among the
Mongol elites, the most resentful were the nobles whose stipends from the gov-
ernment were canceled or reduced and who were forced to make contributions
to fund the “modernizing” projects. The eighth Jebtsundamba alone, for
instance, bought 60,000 taels of bonds for this purpose (Hai Chunliang 2009).
The Mongols knew well that their quest for independence could not succeed
without support from Russia. But Russia’s strategy in Outer Mongolia was not
the same as the goal of the Mongols. Russia only intended to turn Outer Mon-
golia into a buffer between itself on the one hand and China and Japan on the
other; from its point of view, an independent Mongolia would worsen its rela-
tionship with Japan, incur rivalry among the powers that could threaten its
dominance in this area, and constitute a potential troublesome neighbor to it in
the future (Paine 1996: 291–291). It thus was in the best interests of Russia to
make Outer Mongolia its protectorate with full autonomy from China while
accepting China’s suzerainty, a strategy that was essentially the same as the Brit-
ish approach to Tibet. Not surprisingly, when a Mongol delegation led by
Khanddorj (1871–1915) visited Russia in July 1911 to seek their support for
Outer Mongolia’s independence, Russia only responded by acting as a mediator
between China and Mongolia, suggesting the Qing court put a stop to the New
Policies in Outer Mongolia. To enhance its influence there, however, Russia
nevertheless provided the Mongols with a loan of 2 million rubles and a large
quantity of weaponry and helped train Mongol soldiers while sending a cavalry
of more than 800 men to Kulun to enhance its presence there. Later the
204 The making of a unified and centralized state
Mongol leaders saw the outbreak of the Revolution of 1911 in Wuchang as
a perfect opportunity to push for independence. They gathered together the
Mongol forces in Kulun, disarmed the Qing garrison force there, forced Sanduo
and his retinue to leave, and declared the founding of an independent Mongo-
lian state on December 1, 1911, with the eighth Jebtsundamba enthroned as
a theocratic ruler, called Bogd Khan.
A treaty signed between Russia and Mongolia in November 1912 reflected
Russia’s unchanged strategy in this area: it supported Mongolia in maintaining
its autonomy but refused to recognize it as an independent state; both sides
agreed in the treaty, however, that no Chinese troops would be allowed to enter
the territory of Mongolia and no Chinese would be granted the rights to migrate
and colonize it either. The Republican government in Beijing responded by refus-
ing to acknowledge the treaty and insisting that Mongolia was part of China and
that any treaty pertaining to Mongolia had to be made with the central govern-
ment of China instead of Kulun. However, pressed with the urgent tasks of fight-
ing the pro-Guomindang provinces, seeking the recognition of the newly
founded Republic by foreign powers, and obtaining a large foreign loan that
involved Russian banks, President Yuan Shikai retreated from his original position
and started the negotiation with Russia over Mongolia’s status. The result was
the signing in May 1913 of a six-article treaty over Mongolia, by which Russia
recognized it as part of China’s territory and agreed not to send troops to
Outer Mongolia, while China agreed not to alter the preexisting autonomous
institutions in Mongolia, allowed it the rights to organize its own troops and
police and to refuse the immigration of non-Mongolian peoples, and recog-
nized Russian business privileges in Outer Mongolia as specified by the 1912
treaty between Russia and Mongolia (MGWJ, 1: 120).
The treaty satisfied neither side, however; it was soon negated by the Guo-
mindang-controlled Senate of the Beijing government and abolished unilaterally
by the Russian government. Negotiation resumed in Beijing afterwards, result-
ing in the signing of a five-article treaty in November 1913. By this treaty,
Russia acknowledged China’s “suzerainty” over Outer Mongolia and agreed not
to send its troops or migrants there, while China acknowledged Outer Mongo-
lia’s autonomy and its rights to handle domestic affairs but retained Beijing’s
rights to send its commissioners together with armed guards to be stationed in
Kulun and other localities there. In a four-article statement attached to the
treaty, Russia further recognized Outer Mongolia as part of China’s territory,
while the Chinese government agreed to consult the Russian government for
whatever political and territorial issues pertaining to Outer Mongolia arose in
the future (YZHB, 2: 947–948). Another treaty signed by China, Russia, and
Outer Mongolia in June 1915 largely reaffirmed these terms (YZHB, 2:
1116–1120). As a result, Outer Mongolia failed to be an independent state.
Beijing continued to send its commissioners and garrison forces to Kulun and
its other parts of Outer Mongolia; most importantly, Beijing retained the right
of conferring the Jebtsundamba Khutughtu title to the de facto leader of the
Mongolian government, as it actually did in July 1916.
 In search of national unity 205
From the cancellation of autonomy to eventual independence
The outbreak of the October Revolution in Russia in 1917 had an immediate
impact on Outer Mongolia. Unable to seek protection from Russia anymore,
the Mongols were confronted with uncertainties in and outside their land, as
seen in the rapid growth of White Russian forces, the invasion of the Red Army
from Soviet Russia under the pretext of suppressing the Whites, and the pene-
tration of the Japanese, who attempted to create an independent Mongolian
government in collaboration with the White Russians and thereby to put the
whole of Mongolia under its control. Division and rivalry also developed among
the Mongol elites. While high-ranking priests (the so-called Yellow Faction)
around Bogd Khan continued their control of government positions after their
declaration of independence in late 1911, the Mongol nobility (the Black Fac-
tion) who had ruled Khalkha Mongol tribes before 1911 attempted to take back
government power from the priests, and the factional struggle intensified after
1917 when the Yellow Faction lost Russian support. To achieve their goal,
members of the Black Faction expressed to Beijing’s commissioner in Kulun
their willingness to end Outer Mongolia’s autonomous status in exchange for
Beijing’s support, regardless of the opposition from the Yellow Faction. As
a result of negotiation between the Mongol nobles and the Beijing government,
and under pressure from Xu Shuzheng (1880–1925), the newly appointed
Commissioner Overseeing the Northwestern Frontiers (Xibei choubianshi), the
Mongol leaders petitioned Beijing to end their government’s autonomous status
on November 18, 1919. Four days later, Beijing officially announced the cancel-
lation of Outer Mongolia’s autonomy and the restoration of their relationship
that had existed during the Qing (MGWJ, 1: 513–514).
Nevertheless, the recovery of the pre-1911 relationship between Beijing and
Outer Mongolia was short-lived. In February 1921, a tsarist cavalry unit under
Baron Ungern occupied Kulun and drove out the Chinese forces. At Ungern’s
machination, the Bogd Khan once again declared Mongolia’s independence.
Five months later, a Mongolian resistance force led by Damdiny Sükhbaatar,
together with units of the Soviet Red Army, captured Kulun, and an independ-
ent “People’s Government” was established on July 11, 1921, with the Bogd
Khan as nominal head of state. In 1924, following the mysterious death of the
Bogd Khan and the subsequent abolition of the monarchy, the state was
renamed the Mongolian People’s Republic.
Soviet support was key to the success of Mongolia’s independence. In its dec-
laration on August 3, 1919, the Soviet government claimed Mongolia to be “a
free state” whose power belonged to the Mongolian people. The entrance of
the Red Army (more than ten thousand soldiers of infantry, cavalry, and artil-
lery) into Outer Mongolia and its defeat of the Whites occupying Kulun in
July 1921 directly led to the establishment of the Mongolian People’s Govern-
ment. In its agreement with the new Mongolian authority signed on Novem-
ber 5, the Soviet government recognized it as the “only legitimate government”
of Mongolia. A secret treaty between the Soviet and Mongolian governments
206 The making of a unified and centralized state
signed on February 20, 1923, further stipulated the use of Soviet advisors in the
Mongolian government and the stationing of the Soviet Army in Mongolia (Ao
Guangxu 2007: 60). This happened less than a month after Adolph Joffe,
a Soviet diplomat in China, assured Sun Yat-sen, leader of the revolutionary
government in Guangzhou, that the Soviet government “will never, nor have
the intention to, carry out an imperialist policy in Outer Mongolia or to separ-
ate it from China” (SZS, 7: 52). To improve its relationship with China, Lev
Mikhailovich Karakhan, the Soviet ambassador to Beijing, signed an agreement
with Willington Koo, the Chinese foreign minister, on May 31, 1924, in which
the Soviet Union “recognizes Outer Mongolia as a part of the complete Republic
of China and respects China’s sovereignty on this territory” (YZHB, 3: 423).
Only five months later, however, the Mongolia People’s Republic was established,
with Soviet advisors playing a key role in its government and the Soviet Army
controlling its territory. No other countries, however, recognized it as an inde-
pendent state, except for the Soviet Union.3
Mongolia as an independent state won diplomatic recognition only after
World War II as the result of a secret deal among Soviet, U.S., and British lead-
ers and the negotiation between the Chinese and Soviet governments. Accord-
ing to the secret agreement regarding Japan signed by Joseph Stalin, Franklin
Roosevelt, and Winston Churchill at the Yalta Conference on February 11,
1945, the Soviet Union would enter the war against Japan two or three months
after German surrender on the conditions that “the status quo in Outer Mongo-
lia (The Mongolian People’s Republic) shall be preserved” and that the Russian
rights in Manchuria and beyond “violated by the treacherous attack of Japan in
1904” should be restored. But different parties had different interpretations of
“the status quo” in Outer Mongolia. For Stalin, it meant the Soviet Union’s
negation of its earlier stance, which was stated in its 1924 agreement with
China, and China’s recognition of Outer Mongolia’s independence. For the
United States, however, it meant the continuation of China’s legal suzerainty
over Outer Mongolia despite its inability to exercise it in actuality. For the Nanjing
government, it only meant China’s sovereignty over Outer Mongolia as the 1924
Sino-Soviet agreement affirmed (Wu Jingping 1998: 465).
The Nanjing government was informed of this agreement as late as June 14,
1945, and it did not have to accept the conditions demanded by Stalin in order
for the Soviet Union to enter the war against Japan because, after the atomic
bombing of Hiroshima on August 6, Japan was doomed to fail even without
the Soviet involvement to end the war; and, indeed, to ensure the Soviet
Union’s bargaining power after Japanese surrender, Stalin hastily declared war
on Japan on August 9 without the signing of a Sino-Soviet treaty to satisfy his
terms. The real leverage that Stalin had in pressing Nanjing to accept his terms,
in fact, was his attitude toward the imminent civil war in China and the Soviet
occupation of Manchuria after the Japanese surrender. During the negotiation,
Stalin threatened Song Ziwen (T. V. Soong), the foreign minister of the Nan-
jing government, that the Chinese Communist force would be allowed to enter
Manchuria if no treaty was signed between the two governments (Harriman
 In search of national unity 207
1975: 496). For Chiang, the Soviet siding was decisive to his rivalry with the
communist forces. He eventually conceded to Stalin’s pressure after the latter
assured him that the Soviet Union would not provide moral or material assist-
ance to the Communists, that all Chinese military forces should be under the
control of the Nationalist government, and that the Soviet Union would do
its best to push for China’s unification under Chiang Kai-shek’s leadership
(Truman 1986: 269). The Nationalist government officially recognized Outer
Mongolia as an independent state on January 5, 1946, after a referendum was
conducted on October 20, 1945, in favor of Outer Mongolia’s independence
from China. But Stalin’s assurances did not come true in the following civil war
in China, and the latter lost Outer Mongolia forever.

Recovering China’s territorial entity

The Cairo conference of 1943

A defining moment in China’s quest for the recovery of its territorial entity was
the Cairo Conference on November 22 to 26, 1943, in which the leaders of
China, Britain, and the United States, foreseeing the Allies’ imminent victory
over Germany and Japan after starting their offensives in Europe and the West
Pacific, discussed their coordinated operations to end the war and arrangements
for postwar international relations and settlements of the territories that had
been seized by Japan before the war. For Chiang Kai-shek, the top priority in
China’s struggle for national survival and the search for territorial entity would
be the recovery of the lost territories that had been integral parts of the country,
including primarily Taiwan, which had been turned into a province in 1885 but
ceded to Japan in 1895, and Manchuria, which had formed three provinces in
1907 but was occupied by the Japanese Kwantung Army in 1931 and ruled by
the puppet Manchukuo afterwards. Another important goal for Chiang partici-
pating in the Cairo Conference was to free China’s former tributary states from
European or Japanese colonial rule and make them independent states; for
Chiang and the Nationalist government, this was both China’s moral obligation
as an ex-suzerain state towards its former tributary states, of which Korea had
been the most important, and a strategic choice to ensure China’s geopolitical
security in the postwar era. Thus, among the “biggest issues” (zuida zhi wenti)
that Chiang planned to discuss with President Franklin Roosevelt of the United
States and Prime Minister Winston Churchill of Great Britain, the two pertain-
ing to territory were (1) “the Northeast and Taiwan should be returned to our
country” and (2) “the independence of Korea,” as he noted in his diary on
November 17 (JJSR: 11/17/1943).
Chiang did discuss these issues in Cairo. In his conversation with Roosevelt
on November 23 at the dinner hosted by the latter, Chiang raised the territorial
issue and insisted on the return of the “four Northeastern provinces, Taiwan,
and the Penghu Islands” to China, as noted in his diary. Roosevelt agreed on
Chiang’s proposition. This conversation was confirmed by another source, the
208 The making of a unified and centralized state
English translation of an official “Chinese summary record,” which was provided
by Hollington Tong, the Chiang regime’s ambassador in Washington, in 1956
and included in Foreign Relations of the United States Diplomatic Papers.
According to this source,

Generalissimo Chiang and President Roosevelt agreed that the four North-
eastern provinces of China, Taiwan, and the Penghu Islands [Pescadores]
which Japan had taken from China by force must be restored to China after
the war, it being understood that the Liaotung Peninsula and its two ports,
Lushun (Port of Arthur) and Dairen, must be included.
(USFR: 324)

Thus, despite Churchill’s suggestion of changing the wording of the settlement

pertaining to the aforementioned three Chinese territories from “to be restored to
China” to “to be renounced by Japan,” which Chiang opposed, the final draft of
the joint statement of the three leaders nevertheless reflected Chiang’s purpose:

The three great Allies are fighting this war to restrain and punish the
aggression of Japan. They covet no gain for themselves and have no
thought of territorial expansion. It is their purpose that … all the territories
Japan has stolen from the Chinese, such as Manchuria, Formosa, and the
Pescadores, shall be restored to the Republic of China.
(ibid.: 448)

The Soviet leader, Joseph Stalin, did not attend the conference, but he fully
endorsed the above statement at the conference he held with Roosevelt and Church-
ill in Tehran on November 28 to December 1, 1943. According to the minutes of
the conference, Stalin “thoroughly approved the communiqué and all its contents.”
He further stated that “it was right that Korea should be independent, and that Man-
churia, Formosa and the Pescadores Islands should be returned to China” (ibid.:
566). A consensus on the restoration of Manchuria, Taiwan, and the Penghu Islands
to China thus was clearly and fully achieved among the leaders of the Allied Powers,
including China, the United States, Great Britain, and the Soviet Union.
Chiang, therefore, could not hide his excitement and pride by writing the fol-
lowing words shortly after the conclusion of the Cairo conference:

The three eastern provinces and Taiwan as well as the Penghu Islands, as
territories having been lost for fifty or more than twelve years, will be
returned to our country by the joint statement of the U.S. and Britain, and
it is also recognized that Korea will be independent and free after the war.
How immeasurable is the importance of such an event, such a proposition,
and such a hope! All of them are included in the publicized joint statement
by the three Powers. This is indeed a diplomatic success unprecedented in
China and the world.
(JJSR: 11/26/1943)
 In search of national unity 209
The Liuchiu issue
It is worthwhile at this juncture to probe into the exchanges among the major
leaders of the Allies at the Cairo and Tehran conferences over the postwar status
of the Liuchiu Islands. According to the aforementioned record provided by
Tong, in his conversation with Chiang Kai-shek at the dinner on November 23,
Roosevelt “enquired more than once whether China would want the Liuchius,”
and Chiang replied that “China would be agreeable to joint occupation of the
Liuchius by China and the United States and, eventually, joint administration
by the two countries under the trusteeship of an international organization”
(ibid.: 324). Chiang’s response to Roosevelt’s inquiry here was not improvisa-
tional and unprepared. A week earlier, on November 15, Chiang wrote in his
diary: “Liuchiu is different from Taiwan in their respective historical standings
in our country. Liuchiu formed a kingdom, whose standing was equivalent to
Korea, therefore, the Liuchiu issue should not be raised in this round of pro-
posals” (JJSR: 11/15/1943).
Two questions thus remain. First, was Roosevelt sincere in proposing the
return of the Liuchiu Islands to China? Two pieces of evidence suggest
a positive answer to this question. One was the fact that Roosevelt raised this
issue to Chiang “more than once” as quoted above, indicating that this inquiry
was not a mere casual remark, but a serious issue that he wanted to discuss with
Chiang. More telling evidence comes from the minutes of a meeting of the
Pacific War Council held in Washington on January 12, 1944, and attended by
President Roosevelt, the ambassadors of the Netherlands, China, and Canada,
the representative of the British ambassador, the vice president of the Philip-
pines, and the ministers of New Zealand and Australia. According to this
record, Roosevelt informed the Council his “highly satisfactory” discussions
with Chiang Kai-shek and with Stalin at the Cairo and Tehran conferences
respectively. “President Roosevelt also recalled,” the minutes continue, “that
Stalin is familiar with the history of the Liuchiu Islands and that he is in com-
plete agreement that they belong to China and should be returned to her”
(USFR: 869). This record suggests that Roosevelt not only discussed the Liu-
chiu issue with Chiang at the Cairo conference but further discussed the same
issue with Stalin at the Tehran conference and that both the U.S. and Soviet
leaders agreed on China’s right over the Liuchiu Islands. It should be noted
that Roosevelt’s exchange and agreement with Stalin over the Liuchiu issue
took place after Chiang Kai-shek had explicitly and repeatedly expressed his idea
of joint military occupation and administration by China and the United States.
The reasons behind Roosevelt’s insistence on returning Liuchiu to China
remain unclear. But three obvious facts certainly worked together to influence
Roosevelt’s attitude. First, while historically both China and Japan claimed
their suzerainty over Liuchiu, for him China was the only one that maintained
the historical right over the islands after Japan was defeated. Second, Chiang
Kai-shek had insisted on the joint occupation and administration of the islands
by China and the United States, which would actually leave the latter as the
210 The making of a unified and centralized state
dominant power to occupy and govern the islands after the war, given China’s
limited military capabilities to partner the United States in this regard. Third,
by acknowledging China’s sovereignty over Liuchiu, Roosevelt wanted to secure
China’s collaboration with the United States in the military operations in the China
and Southeast Asia theaters and ultimately to turn China into the United States’
ally in the West Pacific to ensure American dominance in the postwar Far East.
The second question is why Chiang rejected Roosevelt’s proposal of returning
Liuchiu to China and insisting on the two countries’ joint control of it. In his
essay titled “China’s Destiny,” published in early 1943, Chiang put Liuchiu at
the top of the list of territories central to China’s security.4 The strategic
importance of the Liuchiu islands to Chiang was obvious. In his diary
on November 23, Chiang offered three reasons why he did not accept Roose-
velt’s proposition. The first was “to assuage the U.S.” (an Meiguo zhi xin).
The second was that “Liuchiu had belonged to Japan before 1895”; here again
Chiang limited his goal to restoring the Chinese territories that were seized by
Japan in and after 1895, primarily Taiwan, the Penghu islands, and Manchuria,
disregarding the fact that Japan annexed Liuchiu as late as 1879. The third was
that “putting this area under the joint administration with the U.S. is better
than under our exclusive possession” (JJSR: 11/23/1943); Chiang knew well
that China was militarily too weak to exclusively occupy the Liuchiu Islands and
that, with the United States as China’s ally in the postwar era, it was also
unnecessary and unwise for China to shoulder all of the duties and costs of
occupying and governing the islands.

China’s territorial entity: lost and rescued

Republican China encountered the biggest crisis in the 1930s and the early
1940s in safeguarding the territory it had inherited from the Qing. Compared
to the Qing at its height in the mid-eighteenth century, China had lost not only
all of the tributary states outside its immediate territory, but also Taiwan to
Japan and parts of its northeastern and northwestern frontiers to Russia during
the late Qing period; after the founding of the Mongolian People’s Republic in
1924 and Manchukuo in 1932, despite their lack of recognition by the inter-
national community, China further lost Outer Mongolia and its four northeast-
ern provinces in actuality. Xijiang and Tibet never officially declared their
independence from China, but the Nationalist government’s influence in those
areas were very limited. China, in a word, returned in essence to where it had
been before the Qing in terms of territoriality. The dynamics leading to these
results involved an intricate interaction among three sets of players: foreign
powers that had a stake in Chinese territory, the ruling elites of the frontiers,
and the domestic politics of inland China.
Each of the three major powers, namely Britain, Russia, and Japan, which
played a decisive role in shaping China’s territory in the Republican era, had its
own goals and strategies towards China. Britain had much more direct invest-
ment in the Chinese provinces than in Tibet; to protect its economic interests in
 In search of national unity 211
China were no less important than maintaining its influence in Tibet. Therefore,
it avoided overt support to Tibet’s separation from China at the risk of offend-
ing the latter and sacrificing its interests there. Its basic policy towards Tibet was
to support the latter’s autonomy and at the same time acknowledge China’s
suzerainty over the highland (Rubin 1968).
In sharp contrast, the Soviet Union’s direct investments in China were negli-
gible, but it was a direct neighbor on three frontiers: Xinjiang, Outer Mongolia,
and Manchuria, which mattered a lot to China’s security, hence its strategies
towards each of the frontiers varied. Russia found it most difficult to penetrate
Manchuria, where it encountered a hostile Chinese warlord regime under
Zhang Zuolin and a strong Japanese presence in Manchuria; instead, it adopted
a defensive strategy in this area and avoided direct confrontation with Japan, its
most formidable rival in the Far East. Russia felt most secure in its relationship
with Xinjiang, where it faced no competition from any other major powers; on
the other hand, the Soviet leaders also found it neither viable to penetrate the
Chinese frontier when it was under the firm control of a Chinese governor,
Yang Zhengxin (1864–1928), who remained loyal to Beijing until his assassin-
ation; nor did they find it wise to alienate Nationalist China by overtly support-
ing Shen Shicai (1897–1970), the next Chinese warlord in Xinjiang, regardless
of his overt pro-Soviet stance, because keeping China as a partner or ally was
important to Russia for it to thwart the Japanese threat to it before and during
World War II. By contrast, Russia had its biggest stake in Outer Mongolia, and
turning this Chinese frontier into an independent state best served its strategic
interest, not only because a weak Chinese presence in Outer Mongolia made it
possible and relatively easy to separate this area from China, but also because an
independent Outer Mongolia could be a buffer in its relationship with Japan,
which had dominated Manchuria since the Russo-Japanese War of 1904–1905.
The Soviet Union thus encouraged the separationist movement in Outer Mon-
golia and played a key role in its independence from China.
For Japan, Manchuria was critically important not only because of its massive
investment and its dependence on the resources and markets in this area, but also
because of its expansionist strategy in which Manchuria served as a springboard
for its further aggression with China. Putting Manchuria under its direct occupa-
tion or under its control through a puppet state best served its economic and stra-
tegic interests. But Japan had to remove two barriers before it was able to achieve
these goals: the Fengtian clique under Zhang Zuolin that appeared to be accom-
modative to the Japanese but at the same time fiercely resisted any moves that
threatened its own interest; and behind the Fengtian clique the Nationalist gov-
ernment in Nanjing that was to incorporate Manchuria into it and thereby build
a unified and strong Chinese state, a prospect that would ultimately shatter the
Japanese dream of expanding their empire to the entire East Asian region. There-
fore, eliminating the Fengtian forces from Manchuria, militarily occupying this
area, and preventing the rise of a unified China under the Nanjing government
through a full-scale war against China were the necessary steps that Japan had to
take to achieve its strategic goals.
212 The making of a unified and centralized state
Each of the frontiers acted differently in dealing with the foreign powers at
stake. The Fengtian clique, though among the strongest regional forces in
China, was disproportional to Japan militarily. It gave up Manchuria to the Jap-
anese Kwangtung Army in 1931 in conformity to the order of the Nationalist
government in Nanjing, which wanted to avoid a full-scale confrontation with
Japan at all cost in the early 1930s. Xinjiang, for its part, remained under the
Nationalist government throughout the 1930s and 1940s because of Stalin’s
lack of interest in annexing the Chinese province, which in turn had to do with
the absence of serious threats from this area to Russia, and, more importantly,
because of China’s growing importance to the Allies in their coordinated efforts
to defeat Japan in the last years of World War II, which explained Shen Shicai’s
reversal of his pro-Soviet stance and submission to the Nationalist government
after 1943. In Tibet and Outer Mongolia, the schism between religious and
secular elites or among the spiritual leaders there accounted for the complexity
and capriciousness of their relationship with the Republican government in Bei-
jing or Nanjing. In the end, despite the weak presence of the Beijing or Nanjing
government in Lhasa, the Tibetan elites never overtly challenged China’s suzer-
ainty over the highland. Britain’s unwillingness to openly stand behind the
separationists was no doubt an important reason. Even more important in
explaining this reality, however, was the lingering impact of the religious and
political bonds between Tibet and the Chinese state that had existed during the
Qing and that survived into the Republican years, as well as the divisions among
the Tibetan spiritual and secular leaders, which prevented them from reaching
a consensus over their land’s definite relationship with China. In Outer Mongo-
lia, likewise, the division between the secular and spiritual leaders explained the
vicissitudes in their relationship with Beijing in the 1910s; it was only in the
early 1920s when the pro-Soviet Mongols came to dominate the government
that Outer Mongolia’s separation from China became irreversible, partly because
of the Soviet Union’s overt support and machinations behind the scenes, and
partly because of the Mongol revolutionaries’ lack of any religious or historical
ties with Beijing.
The attitudes of the Republican state and regional forces towards the frontiers
were equally important in explaining the status of the latter in relation to Beijing
or Nanjing. It was true that divisions among the domestic forces and the over-
whelming pressure of foreign threats made it impossible for Beijing or Nanjing
to play an aggressive role in its frontiers as did the Qing in the eighteenth cen-
tury. However, when confronted with foreign threats or a crisis on the frontiers,
the political elites at the national and regional levels were all compelled to
commit themselves to defending China’s national interest and safeguarding the
security of its frontiers, partly out of the need to protect their own regimes, and
partly because of their nationalist impulse. President Yuan Shikai’s handling of
China’s foreign relations, therefore, appeared to be conciliatory and at the same
time driven by his strong nationalist commitment (Su and Jing 2004). The same
was true about Duan Qirui, leader of the Anhui clique, and Wu Peifu, leader of the
Zhili clique, both appearing to be at once realistic and uncompromising in dealing
 In search of national unity 213
with Western or Japanese imperialism. The growth of the Fengtian clique under
Zhang Zuolin had much to do with his accommodation of Japanese interests in
Manchuria, and he was therefore often known as a pro-Japanese warlord; yet,
Zhang turned out to be firm in defending Chinese sovereignty and rejecting
Japanese excessive claims in the area under his jurisdiction, which led to his
assassination by the Japanese in 1928. All in all, despite the fierce confrontation
over domestic politics under the Beijing and later the Nanjing governments, the
shared need for legitimacy among the national and regional leaders in the age of
nationalism and their subsequent commitment to defending China’s sovereignty
contributed to the survival of most of the frontiers that the Republican govern-
ment had inherited from the Qing.
In the end, it was the interplay between the vicissitudes in China’s geopolitical
setting and the dynamics of its domestic politics that determined what the coun-
try would become. The victory of the Allies over Japan in the Pacific War in
1945 directly led to China’s recovery of its sovereignty over Manchuria, Taiwan,
and the Pescadores. For a while in 1945 shortly after the Japanese surrender,
the Nationalist state was able to claim its sovereignty over a territory that was
not too different from the territory of the Qing during its height in the mid-
eighteenth century, ranging from Manchuria in the northeast, through Outer
and Inner Mongolia in the north and Xinjiang in the northwest, to Tibet in the
southwest and Taiwan in the southeast. China would also have been in
a position to establish its sovereignty over the Liuchiu Islands, had Chiang Kai-
shek been more confident and positive in accepting President Roosevelt’s
repeated propositions. But the Nationalist government’s weakness in dealing
with the Soviet Union and Chiang Kai-shek’s miscalculation nevertheless led to
China’s recognition of Outer Mongolia’s independence at a moment when
China had survived the eight-year Japanese occupation and recovered most of
its lost territories that had been seized by Japan. In terms of sheer size, what the
Republican state lost outweighed what it regained at the end of World War II.
Despite these setbacks, the Nationalists’ state-rebuilding efforts should not be
underestimated. China emerged as one of the Four Powers, together with the
United States, the Soviet Union, and Great Britain, by the end of World War
II, and therefore was in a position to play an important role in shaping the post-
war international order in the Far East. China’s elevated standing later qualified
it to be one of the five standing members of the Security Council of the United
Nations, which prepared the country to play a growing role in world politics
and shake off its image as a weak nation subject to the bullying of foreign
powers. Most importantly, Republican China succeeded in keeping or restoring
most of the frontiers it had inherited from the Qing, which was indeed “the
greatest accomplishment of Republican diplomacy” (Kirby 1997: 437). Much of
the continuity in China’s territory in the twentieth century should be credited
to the historical and religious ties that the Qing court had maintained with the
Mongol and Tibetan elites as well as to the nationalism of the national and
regional leaders under the Beijing and Nanjing governments. A more decisive
factor, however, had to do with the amazing tenacity and viability of the
214 The making of a unified and centralized state
Nationalist state under Chiang Kai-shek during its most difficult years in the late
1930s and early 1940s when China suffered Japanese invasion and occupation.
Without the Nationalist state’s resilience, resistance, and contribution to the
Allies’ war against Japan, China’s recovery of its lost territories from Japan
would have been significantly delayed and discounted. Exactly how the Guomin-
dang state built itself and managed to survive the crises of domestic division and
foreign aggression is to be scrutinized in the next chapter.

Notes
1 A secret agreement between the British and Tibetan delegates at the conference
redefined the border between Tibet and India, which was neither officially reported to
the Dalai Lama by the Tibetan delegate in the absence of the Tibetan government’s
authorization in advance for him to negotiate the border, nor publicized by the British
government for years after the conference (Goldstein 1989: 75–76, 299–309,
412–419). By this boundary (known as the “McMahon Line” named after the British
delegate), more than 70,000 square kilometers of land that had been under the juris-
diction of the Tibetan government would belong to India. Later it became the focal
point in the border dispute between China and India in the early 1960s.
2 One reason had to do with opposition from the secular authorities in Tibet, who saw
his return as a threat to their political standing and economic interests, because the
Panchen Lama would be the only surviving spiritual leader and enjoy much higher
standing in Tibet if he returned, and they would no longer be able to collect taxes
from Western Tibet, which had been ruled by the Panchen Lama before his flight in
1924; those who had been involved in the activities causing the Panchen to flee also
worried about his potential revenge once he came back. A more important reason,
however, was opposition from Britain, which saw the return of a pro-China leader as
a clear threat to its influences in Tibet. Their dispute with the Panchen and the Nan-
jing government focused on the use of the 300 Chinese armed escorts accompanying
his return, which the Nanjing government did intend to be a means to enhance its
presence in Tibet while the British and secular Tibetan authorities interpreted it as the
beginning of the revival of Chinese control there.
3 It is worth noting here the varying attitudes of political forces within China towards
Mongolia’s independence and the Soviet role in it. For a few years after 1917, the Bei-
jing government found itself in an advantageous position to deal with Soviet Russia,
when the latter was beset by the joint military operation of multiple nations in
1918–1919 and the Japanese threat in the Russian Far East, which continued until
1922. After surviving the allied intervention, Soviet Russia was in desperate need of
diplomatic recognition to end its isolation from the international community. To win
over China and overcome food shortages at home by reestablishing trade relations
with China, the Soviet government announced its willingness to renounce all of the
prerogatives, including the Boxer Indemnity, that the tsarist government had procured
in China. Skeptical of the Soviet government’s intension and fearful of the influence of
the Russian revolution in China, however, the Beijing government ignored this
announcement, adopted a hostile attitude toward Soviet Russia, and, as a member of
the Allies in World War I, joined the allied intervention against it. To defend China’s
sovereignty in Outer Mongolia, the Beijing government planned a military expedition
into this region in April 1921, in response to Ungern’s occupation of Kulun. After the
Soviet forces defeated the Whites there, Beijing further demanded the return of Kulun
and other localities occupied by the Red Army and requested the withdrawal of Soviet
forces from Outer Mongolia. It also protested the Soviet government’s signing of
a secret agreement with the new Mongolian government in 1922, and succeeded in
 In search of national unity 215
insisting on the Soviet government’s recognition of Outer Mongolia as part of China
and its commitment to withdrawing its troops from there, as stipulated in the 1924
agreement mentioned above. Thus, for all the conflicts among the warlords that con-
trolled the central government in turn, Beijing was persistent in defending China’s
interests and rights in the northern frontiers, partly because of its need of legitimacy-
building, and partly because of the professionalism of a new generation of Chinese dip-
lomats who had received their training in the West. In contrast, the regional military
cliques adopted a more pragmatic approach toward the border issues. To obtain mili-
tary and financial aid from Soviet Russia, Sun Yat-sen wrote Adolph Joffe in
August 1922, for instance, that the Soviet forces did not have to withdraw from Outer
Mongolia because Soviet Russia had no intention of occupying that area. In return,
Joffe proposed to provide Sun with a loan of 20 million U.S. dollars for ten months,
on the condition that Sun would collaborate with the liberal warlord Wu Peifu of the
Zhili clique to fight the Fengtian clique, which was openly hostile to Soviet Russia,
and form a pro-Soviet government in China (GCGJ, 1: 130–146). Not surprisingly, in
the joint Sun–Joffe manifesto signed in January 1923, Sun concurred with Joffe that
the Soviet forces did not need to withdraw from Outer Mongolia immediately and
that the Russian management of the railroad network in Manchuria, which had been
invested in by tsarist Russia, would remain unchanged (SZS, 7: 52). Zhang Zuolin,
head of the Fengtian clique, turned out to be more opportunistic. At the Tianjin
Meeting attended by the warlords of North China in April 1921, Zhang appeared to
be the most ardent in leading an expeditionary force to take back Outer Mongolia
from the Russian Whites. After receiving a funding of 3 million yuan from Beijing, he
dispatched his troops to Rehe and put the province under his control but refused to
order his forces to march further to the north; later, after the Soviet force captured
Kulun, he stopped the operation and asked Beijing to deal with Moscow for the return
of the city (Hu and Li 2005: 260).
4 Chiang wrote: “Liuchiu, Taiwan and Penghu, the Northeast, Inner and Outer Mongolia,
Xinjiang, and Tibet – all these areas constitute the fortress defending national survival.
Ceding or splitting any of these areas would mean the dismantling of China’s national
defense” (Jiang Jieshi 1946).
9 The fate of semi-centralism
The nationalist state succeeded
and failed

From the 1920s to the 1940s, China experienced the biggest challenge and set-
back but also the most important breakthrough in state rebuilding. Aside from
the persistence of provincial military forces that continued to foil the state’s unify-
ing efforts, the threat to China came primarily from Japan, whose ambition to
establish its commercial and military dominance in East Asia drove its occupation
of Manchuria in 1931 and its full-scale invasion of China in 1937. Had the GMD
force failed to build a national regime or had it lost the war with Japan, the result
of World War II in the Far East as well as the historical course of modern China
would have been different. As it turned out, however, the Nationalist state not
only withstood the Japanese aggression, but also emerged triumphant in 1945
after the eight-year War of Resistance. Even more surprisingly, it recovered all
of the lands that had been lost to Japan since the Sino-Japanese War of 1894,
and it further joined the United States, the Soviet Union, Britain, and France
to form the “Five Powers” to inaugurate the United Nations. At home, the
Nationalist leader Chiang Kai-shek reached the peak of his influence and repu-
tation after the Japanese surrender, and a negotiation between his regime and
the CCP took place to create a democratic coalition government. For
a moment, China appeared to be on the verge of becoming not only an inde-
pendent nation with its territorial integrity largely salvaged, but also a rising
power basing its government on the principles of political freedom and repre-
sentative democracy. Unfortunately, that key step of state-making failed
abruptly in mid-1946, ushering in another round of civil war between the
Nationalists and the Communists, in which the CCP eventually prevailed in
1949. The prospect of building a new state embracing capitalism and democ-
racy yielded to the reality of a highly centralized state under the rule of the
Communist Party afterwards. Why, then, did the GMD succeed in resisting
Japan but fail to fight the CCP?
One cannot fully understand the strength and weakness of the Nationalist
government in Nanjing without first looking at the Northern Expedition that
led to its founding in 1927; it was in this two-year campaign that the National-
ist force arose from a regional competitor, subjugated all other warlords in dif-
ferent parts of China, and turned itself into a national regime. Past studies on
the origins of the GMD government have pinpointed different factors behind its
 The fate of semi-centralism 217
success in the Northern Expedition, such as the GMD’s nationalist ideology and
popular support (Isaacs 1951; Ch’i 1976) or the formation of a “party army”
based on the Soviet model to ensure the army’s organizational cohesion
(McCord 1993: 313–314). Other observers dismiss the importance of mass
mobilization and instead underscore Chiang Kai-shek’s compromise with indi-
vidual warlords (see, e.g., Jordan 1976). Despite the different interpretations on
the rise of the GMD, almost all historians agree on the detrimental effect of
Chiang’s policy of warlord conversion into the party-state he established. They
find that assimilating the large number of conservative, self-interested warlords
or former bureaucrats of warlord governments soon resulted in the rampant
problems of corruption, nepotism, and inefficiency in the GMD’s party and gov-
ernment organizations, in spite of its tremendous efforts of state-building that
achieved breakthroughs on many fronts.1
No matter whether the Nationalist state was merely another warlord regime
beset with failures or a modernizing force, one cannot deny two basic facts
about it. The first is its ability to withstand the Japanese attack and survive
throughout the eight years of the War of Resistance; as Albert Wedemeyer,
Chief of Staff to Chiang Kai-shek commanding all U.S. forces in China,
observed, the Nationalist forces displayed “amazing tenacity and endurance in
resisting Japan,” which contrasted sharply with France in World War II that col-
lapsed after only six weeks of fighting with Germany (Wedemeyer 1958:
277–278). Thus, even Lloyd Eastman, who is among the most dismissive of the
Nationalist regime, praises Nationalist China’s resistance as “a marvel of deter-
mination and self-sufficiency,” which “contributed significantly to the total
Allied war effort against the Axis powers” by tying up “approximately a million
Japanese troops on the continent of Asia – troops that might otherwise have
been used to combat the island-hopping armies of the Western allies in the
Pacific” (Eastman 1984: 130–131). The other fact about the Guomindang state,
of course, is its collapse in the Civil War, which testified to the vulnerabilities
inherent in the same regime.
To offer a balanced account of the GMD’s state-making efforts, therefore,
this chapter looks into the reasons behind the remarkable competence and resili-
ence it demonstrated before and during the War of Resistance as well as its fra-
gility when confronted with the Communists in the decisive Civil War years.
Chapter 7 has discussed the origins of the Nationalist state in Guangdong up to
1927; this chapter, therefore, will look at how the GMD rebuilt a centralized
and unified state on military, ideological, and political fronts after 1927, which
is key to understanding its successes and failures in the 1930s and 1940s.

Toward a unified state

Military unification
The completion of national unification that the Nationalist government declared
at the end of 1928 was nominal and temporary. To achieve the goal of unifying
218 The making of a unified and centralized state
China during the Northern Expedition, the Nationalist force did eliminate its
immediate enemies, the warlord forces of the Zhili clique, from the middle and
lower Yangzi regions, and it also defeated the forces of the Fengtian clique in
North China. In doing so, however, the Nationalists also compromised with the
rest of the warlords, allowing them to keep the territories, armies, and financial
resources that they had possessed before, and expanded during, the Northern
Expedition, as long as they allied with the Nationalists and acknowledged the pol-
itical leadership of Chiang Kai-shek. Thus, China remained divided after 1928.
The military forces that occupied the different parts of the interior provinces fell
into the following four groups:

(1) The First Army, or the “Central Army” (zhongyangjun), of roughly 500,000
men under the direct control of generalissimo Chiang Kai-shek, which occu-
pied Jiangsu, Anhui, Jiangxi, Zhejiang, and Fujian provinces;
(2) The Second Army, or the “Northwestern Army” (xibeijun), of 420,000 men
under Feng Yuxiang, which controlled Shaanxi, Gansu, Ningxia, Henan, and
Shandong provinces;
(3) The Third Army of more than 200,000 men under Yan Xishan of the
Shanxi clique (Jingxi), stationed in Shanxi, Hebei, Chahar, and Suiyuan
provinces as well as the cities of Beijing and Tianjin;
(4) The Fourth Army of over 200,000 men under Li Zongren of the Guangxi
clique (Guixi), garrisoned in Guangxi, Hunan, and Hubei provinces.

Open conflicts between Chiang Kai-shek’s central authority and the three
regional forces, however, soon culminated into wars because of their disagree-
ments over the program of “reorganizing and discharging the Nationalist
Army” (guojun bianqian), which Chiang initiated at the beginning of 1929. For
Chiang, military unification and centralization were indispensable for the
making of a modern state in China.2 Using the story of the returning of terri-
tories and armies from individual domains to the Meiji government after their
collaboration in overthrowing the Tokugawa shogunate, Chiang tried to per-
suade the regional military leaders who had worked together with him in over-
throwing the Beijing government to cooperate again in implementing the
program. Unfortunately, the Nanjing government had neither the authority nor
the resources that the Meiji government had arrogated. For Feng, Yan, and Li,
Chiang was only the first among equals, rather than the one above them. None of
them was willing to sacrifice his own army for the reorganization of the military
that only worked to the advantage of Chiang. True unification was possible only
after the regional forces were militarily subjugated.
The war first broke out in March 1929 between Chiang’s Central Army and
Li Zongren’s Guangxi Army, when Li attempted to enhance his control of
Hunan province by removing its governor without Chiang’s approval, and
quickly ended next month when Li suffered a defeat and retreated back to
Guangxi. Feng, who had been discontented with Chiang’s limitations on his
military force and expansion into Hebei and Shandong, launched another war
 The fate of semi-centralism 219
against Chiang in October 1929, but quickly failed next month. Chiang thereby
expanded his controlled area to Hunan, Guangdong, and parts of Henan and
Anhui after defeating Li and Feng. The third and the largest war erupted in
May 1930 between Chiang’s army of more than 900,000 people on the one
hand and the alliance of the forces of Yan, Feng, and Li on the other. Chiang
once again won the war, thanks to Zhang Xueliang’s siding with Chiang and
sending his Northeast Army to the interior provinces in September, which
decisively changed the balance of military strength in favor of Chiang (LZR:
463). As a result of the war, Feng’s Northwest Army was completely dissolved,
while the remnants of the Guangxi and Shanxi cliques retreated to their respect-
ive home provinces and would never have the strength to challenge the central
government again.
An obvious reason behind the repeated failures of regional military leaders
was the lack of coordination and the frequent occurrence of defection among
themselves when combating Chiang’s Central Army, which in turn had to do
with the conflicting interests and goals of the allied forces and the complex
composition and poor integration of each of the regional forces. But the most
important reason leading to Chiang’s successes on the battleground was the
striking disparity between the Nanjing government and its challengers in fiscal
and military strength. Among the regional forces, Feng’s was the largest, but it
was based on the poorest northwestern provinces; expanding his area to the
richer Hebei and Shandong provinces and occupying the major cities (Beijing
and Tianjin) of North China thus were the primary goal of his wars against
Chiang. Without much resources to support the large size of his army, he kept
the stipends and living standards of his soldiers as well the high-ranking officials
to a very low level, which in turn affected the overall morale of the Northwest
Army. Yan Xishan’s army was best equipped among the regional forces, but the
scale of the war he led against Chiang in 1930 was too large for the limited
financial resource from his own province to sustain; he tried to raise more funds
to support the war by issuing 6 million yuan of “wartime universal notes”
(zhanshi tongyong piao), which, however, helped little in relation to the huge
expenses on the war. Having been defeated in 1929 and based on the poor and
remote province of Guangxi, Li Zongren’s army was the smallest among the
three regional forces, having only about 50,000 soldiers to join the war against
Chiang in 1930 (Li Jingzhi 1984).

The renewed strength of the Nanjing government

In sharp contrast with the limited fiscal resources available to the anti-Chiang
alliance, the strength of the Nationalist government in Nanjing lay precisely in
its access to the largest sources of tax revenues and public debts in China,
thanks to its control of the richest provinces (including Jiangsu, Zhejiang, and
Jiangxi before 1929 and Guangdong and Hunan thereafter) and the city of
Shanghai, where the wealthiest financiers of China concentrated. Even more
important to the Nationalist government’s fiscal health was the phenomenal
220 The making of a unified and centralized state
increases in its income from maritime customs and salt taxes. What made this
possible was first of all the recovery of China’s autonomy on maritime customs,
which had been one of the central goals of the Nationalist Revolution. For
seven decades since the signing of the Treaty of Tianjin (Tien-tsin) in 1858, the
tariff rate on imported goods collected by the Qing and, after 1911, the Repub-
lican governments had been limited to 5% of the value of imported goods in
principle but only about 3% in reality. In the late 1910s and early 1920s, the
Republican government in Beijing made repeated efforts toward the recovery of
China’s tariff autonomy and almost succeeded at the international tariff confer-
ence in Beijing in 1925, where the 13 participating countries agreed on granting
China the autonomy on maritime customs to begin in 1929. The Nationalist
government in Nanjing, therefore, resumed the same effort and signed new
treaties with 12 Western countries in 1928, and with Japan in 1930, to nullify
the preexisting agreements on maritime customs and affirm China’s complete
tariff autonomy. Under the new tariff regulations that became effective on Feb-
ruary 1, 1929, the maritime customs on imported goods varied from 7.5 to
27.5%, which were later adjusted to 5 to 50 % in 1931 and as high as 80% on certain
goods after 1933 (Wu Jingping 1992: 106–107). The result was the skyrocketing in
maritime customs from about 134 million yuan in 1928 to 245 million yuan in
1929, 292 million yuan in 1930, and 387 million in 1931 (decreasing to around
300 million yuan per year in the following years because of the Japanese occupation
of Manchuria in late 1931) (MGDA, 5.1.1: 547–548).
Equally important to the Nanjing government’s fiscal strength was streamlin-
ing the collection of salt taxes. As a condition for the Yuan Shikai government’s
borrowing of the so-called “postwar rehabilitation loan” from five countries in
1913, salt taxes in China were designated as a guaranteed source for repaying
the debt, and, to that end, salt taxation was reorganized under a new hierarchy
in which a foreigner and a Chinese were appointed as co-administrators at each
level, and all of the income from salt taxes were deposited in designated foreign
banks to fulfill China’s obligation of loan repayment; the remainder was then
remitted to the Chinese government. To reassert China’s sovereignty in salt tax-
ation, the Nationalist government at once abolished this co-administrative
system and established a new set of systems headquartered in Shanghai after
1927, which, however, only resulted in a deterioration in administrative effi-
ciency and a decline in salt tax income. To correct the problem, financial minis-
ter Song Ziwen reappointed most of the original Chinese and foreign
administrators to the salt tax agencies because of their proven professionalism,
but delinked the collection of salt taxes with the repayment of foreign debts,
while his ministry was committed to fulfilling the central government’s duties
on foreign debts. All of the income from salt taxes, therefore, was deposited in
Chinese banks and subject to the complete control of the Nationalist govern-
ment. Meanwhile, Song’s ministry made systematic efforts to rationalize the
rates of salt taxes in different areas, reorganized the administrative and auditing
agencies in salt taxation, rebuilt the policing force against salt smuggling
throughout the country, and introduced in 1931 the policy of taxation at salt
 The fate of semi-centralism 221
production sites only and the free trade of salt instead of the monopoly system
in selected areas; by 1937, this new policy had been applied to as many as 60%
of the counties and cities in China. As a result of these measures, government
revenue from salt taxes escalated from about 29.5 million yuan in 1928 to
122.1 million yuan in 1929, 150.5 million yuan in 1930, and 247.4 million
yuan in 1936 (see Table 7.1; listed as 54 million yuan in 1928, 85 million yuan
in 1929, 130 million in 1930, and 218 million yuan in 1936 in MGDA, 5.1.1:
551; see also Young 1971: 56).
The quick increases in the Nanjing government’s revenues after 1928 and the
growing imbalance between Chiang and his regional challengers in fiscal capaci-
ties explain in large measure his successes on the battleground. The generous
budget on the military, accounting for about 44 to 45% of the government’s
total expenditure in 1929–1931 (Yang Yinfu 1985: 70), allowed Chiang to
import from Germany advanced weapons in large quantity, which made his
Central Army much better equipped than the forces of regional leaders. His
offering of financial aid and military supplies to Feng won over the latter’s sup-
port during his war against Li Zongren in early 1929, and his generous cash
rewards to the high-ranking commanders of Li’s forces accelerated their defec-
tion and Li’s failures in the war. Later, to prepare for the war against Feng in
late 1929, Chiang successfully bought over Han Fuqu and Shi Yousan, the two
key commanders of Feng’s forces, who led more than 100,000 troops away and
thus fundamentally undermined the combating ability of Feng’s Northwest
Army. Finally, during his confrontation against the allied regional forces in
1930, Chiang outcompeted his enemies both militarily and fiscally. Altogether,
the allied regional forces of Yan, Feng, and Li spent only about 10 million yuan
per month during the war that lasted for six months, whereas Chiang’s forces
spent about 30 million yuan a month (cf. Zhang Hao 2008: 116). Chiang,
therefore, was able to offer large sums of rewards to his commanders and sol-
diers to boost their morale on the battleground (he offered on August 24,
1930, for instance, a reward of 1 million yuan to any of his troops who first
occupied Luoyang and Zhengzhou, and a reward of 200,000 yuan to those
who first occupied the neighboring Gong county, see Li Jingzhi 1984: 226). The
most important step leading to Chiang’s victory in the war of 1930, however, was
Zhang Xueliang’s decision to side with him, which was possible only after Chiang
agreed to offer Zhang 5 million yuan as the expenses of the Northeast Army, in
addition to 10 million yuan as public debt to be raised by Song Ziwen for Zhang
(Chen Jinjin 2000: 12).
Chiang’s superior fiscal strength thus played a key role in his subjugation of
the regional forces and the military unification of China. Chiang’s own observa-
tions best illustrate the stark contrast in the political situations in China before
and after the three consecutive wars in 1929 and 1930. In March 1929, on the
eve of the war against Li Zongren, Chiang remarked: “Has China been truly
unified? Just take a look at the actual political condition, and we can conclude
that China is still not unified in reality” (SSJY, 1929, Jan.-Apr.: 727). On
November 12, 1930, right after the war against the allied regional forces,
222 The making of a unified and centralized state
Chiang stated proudly: “it is my deep conviction that a situation has formed for
our Party to unify China; it is absolutely impossible for any traitors to the Party
and the nation to rise again” (Guo Xuyin 1992: 193). To say that China had
become a politically and militarily unified state by the end of 1930 was far from
true; the bordering provinces ranging from Guangxi, Yunnan, and Sichuan in
the south and southwest to Shanxi in the north and the three northeastern prov-
inces were yet to be fully integrated into the administrative and military systems
under the central government, not to mention the numerous base areas of the
communist forces that had expanded quickly during the wars against Li, Feng,
and Yan in the late 1920s and the early 1930s. Nevertheless, Chiang did indeed
firmly establish his political and military supremacy throughout the provinces
after 1930. There was no longer any regional cliques and any provincial leaders
under the Republic that could openly challenge the authority of the central govern-
ment. The long-term trend of political fragmentation and military decentralization
since the 1850s eventually came to a halt by 1930 and was substantially reversed
afterwards, and China definitely moved toward the making of a centralized modern
state under Chiang Kai-shek’s leadership.

Fiscal unification
The unification of fiscal resources was as important as the centralization of mili-
tary power to the making of a modern state. Back in Guangdong in the mid-
1920s, Song Ziwen’s stunning success in centralizing the financial resources
within the province was key to the Nationalist force’s meteoric rise in the course
of the Northern Expedition. Before its complete subjugation of regional military
forces, however, the Republican government in Nanjing faced huge barriers in
establishing a unified revenue system throughout the country. With only three
provinces (Jiangsu, Zhejiang, and Anhui) under its direct control, the Nanjing
government during the first few years of its existence received only about
4 million yuan per month as regular tax revenues from these provinces. In the
rest of the country, tax collection was in the hands of provincial or lower-level
authorities, who retained all of the revenues they generated and resisted the cen-
tral government’s appointments and policies pertaining to fiscal affairs that did
not serve their interests.
The most audacious and successful step in fiscal unification taken by the Nan-
jing government after 1930 was the abolition of lijin and, in its stead, the intro-
duction of “unified taxes” (tongshui).3 Soon after its relocation to Nanjing, the
Nationalist government announced in July 1927 its plan to abolish lijin in six
lower Yangzi and southeast provinces on September 1 of the same year but
delayed the schedule over and again because of warfare and instability through-
out the country. As a preliminary step, however, the Nanjing government never-
theless first enforced in February 1928 the abolition of lijin on cigarettes and
instead collected a one-time “unified tax” of 27.5% at the factory (for domestic
cigarettes) or maritime customs (for imported cigarettes) to qualify their free
trade without taxation afterwards (Wu Jingping 1992: 116); the same tax policy
 The fate of semi-centralism 223
was also applied to the trading of flour in 1928. It was not until the Central
Army decisively defeated the regional forces in late 1930 that the Nanjing gov-
ernment announced the nationwide abolition of lijin to begin on January 1,
1931. Local governments who had relied on lijin for their normal functioning
turned to sales tax and subsidies from the central government as the new
sources of revenue. To offset the loss of tax revenue after the abolition of lijin,
the collection of the unified tax was extended to cotton yarns, matches, and
concrete in 1931. The number of provinces where the collection of unified
taxes was enforced also expanded from five in 1928 to 15 in 1933 and 20 in
1936. Government revenue from this source also increased from nearly
27.69 million yuan in 1928 to 88.68 million yuan in 1931, 105 million yuan in
1933, and 152 million yuan in 1935 (see Table 7.1). It thus joined maritime
customs and salt taxes to be the three major sources of the central government’s
revenues in the 1930s prior to the Japanese invasion in 1937.

The making of a new orthodoxy

The Nationalist state also made headway in identity building. Unlike any pre-
ceding governments in Chinese history, the GMD advocated a unique set of
ideology, known as Sanminzhuyi or “the Three Principles of the People” (the
principles of nationalism, democracy, and people’s livelihood, or minzu, min-
quan, and minsheng), and established its orthodoxy and supremacy in political
discourse in areas it controlled. First formulated as a guideline for the Tong-
menghui established in 1905, Sanminzhuyi was vigorously propagated by Sun
Yat-sen after 1923 as a tool to build political consensus among the GMD mem-
bers and supporters. Having suffered repeated failures in the Second Revolution
in 1913, the Pro-Constitution Movement in 1917–1918, and especially the first
round of the Northern Expedition that was discontinued because of the treach-
erous rebellion by Chen Jiongming (1878–1933), one of his most-trusted fol-
lowers and governor of Guangdong province, in 1922, Sun believed that the
biggest reason behind those failures was that the Guomindang had performed
“too much struggle by violence but too little struggle by propaganda” (SZS, 8:
568); in other words, the revolution failed because “the vast majority of the
people in the country do not understand the reason of the revolution; and
people do not understand the reason of the revolution because no propaganda
has been widely conducted” (SZS, 8: 322). Therefore, Sun asserted in Decem-
ber 1923 that another task of the GMD afterwards was to propagate Sanminz-
huyi and win people’s support, or “to conquer with the ism” (yi zhuyi zhengfu)
(SZS, 8: 432), a task “more important than armed struggle” (zhuyi shengguo
wuli) for achieving the GMD’s goal of state-building (SZS, 9: 107).
After the death of Sun on March 12, 1925, the GMD leadership was even
more enthusiastic in propagandizing Sanminzhuyi. Public lectures on San-
minzhuyi by Wang Jingwei and other elites of the Guomindang were fre-
quently held on various occasions in late 1925 and early 1926 before the
resumption of the Northern Expedition. The Nationalist government, newly
224 The making of a unified and centralized state
established in Guangzhou on July 1, 1925, forcefully promoted “the Party-
ization education” (danghua jiaoyu) in public schools, as reflected in its reso-
lution that “all educational measures have to be in line with the spirits of
Sanminzhuyi, and education at all levels have to be incorporated with the
indoctrination of the Party’s teachings as much as possible” (Chen Jinjin
1997: 116). Subsequently, a mandatory class on Sanminzhuyi, at least 50
minutes per week, was added to the curriculum of schools in the GMD-
control areas, and textbooks were audited and revised to ensure the centrality
of the Party’s doctrine and policies in education. A weekly meeting commem-
orating Sun Yat-sen was routinely held in schools, and schoolchildren were
organized into boy scouts by the GMD to receive systematic and ritualized
ideological training (He and Wen 2008: 10).
After establishing the Nationalist government in Nanjing in April 1927 and
officially finishing the task of political unification in June 1928, the GMD made
further efforts to establish the orthodoxy of its ideology in order to legitimize
its rule in China. Sun Yat-sen was venerated as the “Father of the Republic”
(guofu) and, to some degree, deified through the rituals and ceremonies in
honor of him. His portrait was required to be displayed at the central site in all
official buildings, with couplets on two sides of the portrait that bore his famous
saying in two sentences: “The revolution has not succeeded yet; comrades
should continue the endeavor” (geming shang wei chenggong; tongzhi reng xu
nuli). Beginning in March 1928, government officials, and later all civil servants
of the government as well as male teachers of all schools, were required to wear the
uniform named after Sun Yat-sen, which was associated with the revolutionary spir-
its of discipline, rigidity, and conformity and designed to symbolize Sun’s ideas
(Chen Yunxi 2007). The history of the Republic of China was also reinvented so
that Sun Yat-sen occupied a central place in it. According to the GMD’s new narra-
tive, the Nationalist revolution began with Sun’s founding of Xingzhonghui in
1894 and then Tongmenghui in 1905, culminated in the Revolution of 1911 in
which Sun served as a spiritual leader despite his complete absence from it, failed
afterwards because of the revolutionaries’ deviation from Sun’s correct path, and
resurged in the early 1920s after Sun fully took the leadership of the Party and the
military (Ouyang Junxi 2011). All these contrasted sharply with the Beijing govern-
ment’s deliberate downplaying of the Revolution of 1911 and the overall negative
image of Sun Yat-sen in the media and public discourse in the 1910s and early
1920s in areas outside the GMD’s influence.
Given the newly established orthodoxy of Sanminzhuyi in China’s political
life under the Nationalist rule, it was not surprising that a precondition for the
warlord forces to be accepted by the Nanjing government and incorporated into
its system in the course of political unification in the late 1920s and 1930s was
their acknowledgment of the supremacy of Sanminzhuyi as an ideology, an
action that was equated with their recognition of the legitimacy of the National-
ist government.4 Accepting Sanminzhuyi, in other words, was the very basis on
which different political forces manifested their identity with the newly estab-
lished Nanjing government after 1927.
 The fate of semi-centralism 225
Despite its supremacy as an orthodox ideology and central role in maintaining
the Guomindang government’s legitimacy, however, Sanminzhuyi failed to serve
as an ideological basis for building political identity in Nationalist China. Quite the
contrary, because of its lack of theoretical coherency and its all-encompassing
framework, Sanminzhuyi was subject to different interpretations and therefore in
actuality conducive to political division and confrontation. In the early 1940s, both
the pro-Japanese forces under Wang Jingwei (WWZQ, 1: 104, 217–218) and the
communist forces under Mao Zedong (ZGZY, 11: 633–634) thus were able to
interpret the same set of ideology differently to legitimate their existence and
enhance their rivalry with the Nationalist force under Chiang Kai-shek. Within the
GMD itself, Chiang also failed to use Sanminzhuyi to build solidarity among the
party members and loyalty toward himself, because of the lack of sufficient persua-
siveness of Sanminzhuyi as a doctrine.5 The weekly commemoration of Sun Yat-sen
attended by all public servants in the military, educational, and government institu-
tions on every Monday became purely ritualistic, lacking any substance in indoctrin-
ation and identity building. In the absence of ideological commitment and political
identity, what kept the GMD elites together were merely their shared interests and
power, which would inevitably lead to conflicts and corruption among them. Fore-
seeing the GMD’s fundamental weakness in this regard and its inevitable failure in
identity building, Hu Hanmin (1879–1936), one of the senior Nationalist leaders,
made the following remark right after the inauguration of the Nationalist govern-
ment in Nanjing in 1927: “Short of unwavering faith and unable to generate
enough strength to resist temptations and threats because of the fundamental
reason of lacking a thorough understanding of its ism, how could such a party
avoid its collapse?” (SSJY, 1927, Jan.-Jun.: 883).

Political identity under the party-state

Establishing the party rule

The Nationalist government departed from the preceding regimes in China not
only in its prioritization of the party’s ideology in political life, but also in its
establishment of the supremacy of the party (i.e., the GMD) in the entire
system of state apparatus. The idea of making a party-state in China can be
traced back to Sun Yat-sen’s designation of the unique role of the Chinese
Revolutionary Party (Zhonghua geming dang) in government after the Revolu-
tion. As he conceived it in 1914, the Party would “shoulder full responsibility
for administrating all military and civilian affairs of the state” during the entire
period of the Revolution, which began with “military rule” (junzheng) to clear
by violence all barriers on the way to a true republic, continued through the
phase of “political tutelage” (xunzheng) to promote self-government at the local
level, and culminated in the phase of “constitutional government” (xianzheng)
that aimed to promulgate a constitution (SZS, 3: 97). During the revolutionary
period, however, only party members had the rights to elect or be elected, and
their rights varied depending on the inception of their membership at different
226 The making of a unified and centralized state
points of the Revolution, while non-party members were not allowed to join
the government and denied rights as citizens before the promulgation of the
constitution (ibid.: 98, 104). Furthermore, all of the party members had to
give up their individual freedom and show unconditional submission to the
single will of the party leader, or Sun himself, which he believed to be essential
to overcoming the problem of political division among the revolutionaries
(ibid.: 92, 105, 184).
Sun’s idea of building a party-state in China crystalized in the early 1920s
when he endeavored to transform the GMD by borrowing from Soviet Russia.
For him, the Soviet one-party system was superior to the parliamentary democ-
racy in Europe and America. As he put it, “both the French and American
republics are old-styled, and only the Russian [government] today is of new
style; we should therefore build the newest republic [in China]” (SZS, 6: 56).
“If we want the revolution to succeed,” he said, “we have to learn from the
Russian methods, organizations, and trainings to make the success possible”
(SZS, 8: 437). The Russian methods, in his view, were nothing else but the
highly centralized organization, its strict disciplinary measures, and, most
importantly, the party’s tight control of the military (yi dang ling jun) and sol-
diers’ indoctrination with the Party’s ideology (zhuyi jianjun). Subject to the
GMD’s supervision through its Central Executive Committee and especially the
Political Council (zhengzhi weiyuanhui) within the committee, the Nationalist
government established in Guangzhou in July 1925 formed a prototype of the
party-state envisaged by Sun. With Sun as its chairman, the Political Council
served as the highest organ of decision-making for all institutions of the Party,
the government, and the military; members of the Political Council headed each
of the government’s departments, which only carried out the decisions made by
the council (Chen and Yu 1991: 337–338, 347).
The Nationalist party-state took its definite form in the late 1920s under the
slogan of “Let the Party govern the state” (yi dang zhiguo), when the Revolu-
tion entered the phase of political tutelage, following the Nanjing government’s
unification of the country. According to the “Outline of Political Tutelage”
(xunzheng gangling) announced by the GMD leadership in October 1928,
the Party’s Congress exercised governmental power on behalf of the National
Congress (guomin dahui) during the period of political tutelage, and the Party’s
Congress, when not in session, trusted governmental power to the Party’s Polit-
ical Council. The National Congress was eventually held in May 1931, but the
Provisional Constitution of the Political Tutelage Period (xunzheng shiqi yuefa)
which it passed only reaffirmed the principles of the “Outline” and went even
further by elevating the power of the president of the national government, who
appointed the heads of the Five Yuan and the ministries and served as the Gen-
eral Commander of the military forces. Chiang Kai-shek was again elected as the
president of the national government, who further named himself as the head of
the Administrative Yuan, thus officially becoming the leader of the Party, the
government, and the military. Institutional building under the Nationalist state
was characterized by the trend of militarization, in which the military leader
 The fate of semi-centralism 227
came to control the government, and the central government controlled individ-
ual provinces by military means (Tien 1972).

Chiang’s quest for leadership

But Chiang’s efforts to concentrate the power of the Nationalist state into
his own hands was far from smooth in the 1920s and 1930s. The biggest
challenge to Chiang as a native of Zhejiang province and as military com-
mander in building his leadership within the GMD came from the fact that,
from its very beginning, the Party had been dominated by those from
Guangdong province who had a strong sense of representing the authentic
force of the Nationalist Revolution and the orthodox leadership of the
Party, a shared feeling that was further enhanced by their keen awareness of
belonging to an exclusive, Cantonese-speaking ethnic group.6 Over the course
of the Northern Expedition, when the Revolution expanded from Guangdong
to central and eastern China after the death of Sun, the Cantonese members of
the GMD continued to see Guangdong as the very basis of the Revolution, not
only because they occupied the key positions of the Party, but also because
their home province remained the most important fiscal source shouldering the
expenses on military operations well until Chiang Kai-shek turned to the Shang-
hai-based financiers for help, as discussed in Chapter 6. Not surprisingly, the
senior Cantonese leaders of the GMD proudly identified themselves as the true
revolutionary force representing the orthodoxy of the leadership of the Party;
they had no problem accepting Chiang Kai-shek as the general commander of
the Nationalist Revolutionary Army (NRA), but they tenaciously resisted
Chiang’s attempts to assume the GMD’s supreme political leadership in place of
the late charismatic leader Sun Yat-sen.
Chiang, for his part, had to turn to the people from Zhejiang, his home prov-
ince, for help in his rise to the top leadership of the Nationalist state. Among
the key Zhejiang supporters of Chiang were Zhang Jingjiang (1876–1950), Sun
Yat-sen’s time-honored friend and generous donor and Chiang Kai-shek’s
unfailing patron, who recommended Chiang to Sun and named Chiang as the
general commander of the NRA; Dai Jitao (1891–1949), once a secretary of
Sun and a believer in Marxism and, after Sun’s death, a leading right-wing
propagandist standing behind Chiang’s anti-communist maneuvers; and Yu
Qiaqing (1867–1945), co-founder of the Shanghai Securities and Commodities
Exchange and president of the Shanghai Chamber of Commerce in the early
1920s, who played a key role in making the Shanghai-based financiers’ support
available to Chiang.7 Chiang’s biggest barriers to his rise to the GMD leadership
in the 1920s and early 1930s were primarily the natives of Guangdong, most
prominently Hu Hanmin (1879–1936) and Wang Jingwei (1883–1944), both
having been Sun’s faithful followers since the founding of the Tongmenghui
and, after Sun’s death, key GMD leaders of the highest seniority, and later also
Sun Ke (or Sun Fo) (1891–1973), Sun Yat-sen’s son and a member of the
GMD’s central executive committee.
228 The making of a unified and centralized state
Chiang’s rivalry with the GMD’s Cantonese elites began with his involvement
in the handling of the case of Liao Zhongkai (1877–1925), the most famous
leftist among the senior GMD leaders and financial minister of the newly estab-
lished Nationalist government in Guangzhou, who was assassinated on
August 20, 1925. As the leading rightist and the most likely suspect, Hu
Hanmin was detained and later sent to the Soviet Union for a tour. Xu
Chongzhi (1886–1965), a native of Guangzhou and another leading rightist
serving as the military minister of the Nationalist government plus the chairman
of Guangdong provincial government, lost his positions for protecting his sub-
ordinate who was also suspected, thus clearing the way for Chiang’s control
of the military. After the Zhongshan Warship Incident in March 1926, in
which the captain of the warship, a CCP member, was accused of rebelling
against the Nationalist government, thus causing all CCP members to with-
draw from the Nationalist army, Wang Jingwei also quit his positions as the
president of the Nationalist government and chairman of the GMD’s military
committee and went to France because of his obvious pro-CCP inclinations.
Two months later, at the Second Session of the GMD’s Second Congress,
Chiang not only proposed a resolution to limit the CCP members’ positions
within the GMD party but also enhanced his influence by nominating his
patron, Zhang Jingjiang, as the chairman of the standing committee of the
Party’s central executive committee.
Chiang suffered his first setback in March 1927, when the Third Session of
the GMD’s Second Congress, held in Wuhan and dominated by left-wing
GMD elites and CCP members, elected Wang Jingwei in absentia as the chair-
man of the standing committee of the Guomindang central executive committee
and the head of the GMD central’s organization department, and at the same
time abolished the position of the chairman of the GMD central’s military com-
mittee, which had been held by Chiang Kai-shek, while allowing Chiang to
keep his position as the General Commander of the NRA. Chiang reacted by
establishing a separate Nationalist government in Nanjing to rival Wuhan, which
only resulted in the GMD central’s decision to terminate Chiang’s GMD mem-
bership at Sun Ke’s proposal. To allow for a reconciliation between Wuhan and
Nanjing, Chiang Kai-shek stepped down from his position as the supreme mili-
tary commander in August 1927 in the wake of a devastating military defeat in
northern Jiangsu and under pressure from Li Zhongren of the Guangxi clique,
who collaborated with Wang Jingwei of Wuhan, after the latter purged the
CCP, to constitute a fatal threat to Nanjing (Huang Daoxuan 1999). When he
resumed his position as the General Commander of the NRA in January 1928,
Chiang had to allow the Guangdong elites to hold key positions in the Nanjing
government, as a way to enhance his legitimacy as the new leader of the party-
state, but his rivalry with the Guangdong elites was far from conclusive; the
latter accepted Chiang as a military leader but refused to recognize him as
a political leader (Dong Xianguang 1952: 108).
A second round of his rivalry with the Cantonese elites broke out in Febru-
ary 1931, when Chiang put Hu Hanmin under custody, ostensibly because of
 The fate of semi-centralism 229
Hu’s opposition to Chiang’s insistence on promulgating the Provisional Consti-
tution of the Political Tutelage Period and also because of Hu’s insulting defi-
ance of Chiang’s leadership; by that time, Chiang had established his status as
the undisputed leader of China after decisively subjugating his most deadly
adversaries, including Li Zhongren, Feng Yuxiang, and Yan Xishan, who had
allied against him in 1930. In protest against Chiang’s detainment of Hu,
almost all of the Cantonese elites left Nanjing and assembled in Guangzhou,
where they established the Extraordinary Council of the GMD Central
Executive Committee on May 27, 1931, and a separate Nationalist govern-
ment in Guangzhou the next day. Chiang found himself in a nightmare over
the following months, when he had to fight on three fronts simultaneously:
the rapidly growing CCP forces in south-central China, the GMD opposition
force in Guangzhou, and, more deadly, the Japanese military force that occu-
pied Manchuria and which was taking advantage of the domestic chaos in
China. So frustrated was Chiang that he complained of the Cantonese who
made excessive requests in negotiating with Nanjing as intending to turn
Guangdong into a separate “Cantonese state allying with Japan to besiege
and attack our country” (SLGB, 12: 196). In his view, the Cantonese worked
together to oppose the Nanjing government “only because of the reason of being
Cantonese” and they only showed their “narrow-mindedness” by “confronting the
entire country with a few Cantonese comrades” (Jin Yilin 2005: 123–124). Under
pressure from Guangzhou and from within the Nanjing government, he stepped
down again on December 15, 1931.
To prevent a dictatorship within the Party, the First Session of the GMD
Fourth Congress held in Nanjing in late December, 1931 passed a resolution
that the Administrative Yuan, instead of the president of the Nationalist govern-
ment, would be trusted with the exercise of government power. Chiang Kai-
shek, Hu Hanmin, and Wang Jingwei were elected at the same session as the
three standing members of the Party’s Central Political Council and who would
serve as the chairmen of the Council in rotation. A new cabinet was subse-
quently formed in the New Year of 1932. Of the 14 members of the cabinet
(including the head and ministers of the Administrative Yuan), nine were from
Guangdong, including Sun Ke as the head. The Cantonese thus appeared to be
triumphant in their rivalry with Chiang. Without the necessary experience and
political resources, however, Sun Ke soon faced huge difficulties running the
government and, as a result, quit his job in less than a month, despite the moral
support he received from Hu Hanmin, who refused to go to Nanjing and
stayed in Guangzhou until his death in 1936, where he was officially in charge
of the GMD and the Nationalist government’s administrative affairs in South-
west China, thus remaining semi-independent from Nanjing in actuality. After
Sun Ke’s resignation, Wang Jingwei and Chiang Kai-shek reached a compromise,
by which the two would be in charge of the government and the military,
respectively. Consequently, Wang became the new head of the Administrative
Yuan on January 28, 1932, while Chiang was elected as the chairman of the
GMD Central’s Military Committee on March 6. Seven months later, Wang quit
230 The making of a unified and centralized state
his job and went to France again, realizing that his power as the government
head had been greatly curtailed by Chiang and that Yuan had been only allowed
to handle trivial administrative affairs. Chiang reestablished his control of the
Nationalist party-state afterwards. There would be no more substantial challenge
to his leadership from within the GMD’s power center before the fall of his gov-
ernment in 1949.

Factionalism
Chiang’s establishment of his uncontested leadership at the top level of the
GMD by 1932 should not be equated with his valid control of the government
system and the military in the entire country. To be sure, since its completion
of the Northern Expedition and announcement of China’s unification in
June 1928, the Nationalist government in Nanjing had established its status as
the only legitimate central government to be unanimously accepted by the
authorities of all provinces. After defeating the allied regional forces in 1930,
the NRA under Chiang’s direct leadership, known as the Central Army (zhon-
gyangjun), further emerged as the dominant military force in China that no
other regional forces could compete with. While accepting Sanminzhuyi as the
only orthodox ideology, claiming their loyalty to the Nanjing government, and
acknowledging Chiang’s supreme leadership, the regional military authorities
remained largely semi-independent and resisted Nanjing’s attempts to penetrate
their provinces administratively and militarily. In the early to mid-1930s, these
semi-independent forces included:

(1) Chen Jitang of Guangdong province;

(2) Li Zongren and Bai Chongxi of Guangxi province;
(3) Long Yun of Yunnan province;
(4) Wang Jialie of Guizhou province;
(5) Liu Xiang of Sichuan province;
(6) Liu Wenhui of Sichuan before 1934 and Chuanbian (Xikang) afterwards;
(7) Ma Bufang of Qinghai province;
(8) Yan Xishan of Shanxi province;
(9) Song Zheyuan of Beiping (Beijing), Tianjin, Hebei, and Chahar; and
(10) Han Fuqu of Shandong province.

For the Nanjing government, however, the biggest threat to its rule in China
came from two sources, the bourgeoning communist base areas in south-central
China in the early 1930s, and Japanese aggression beginning with its occupation
of Manchuria in 1931. Of the two, Chiang perceived the former as the fatal
“disease of the heart” (xinfu zhi huan) (MGSL, 1.3: 35) and therefore carried
out the strategy of prioritizing the suppression of the Communists or, in his
words, “pacifying the internal before fighting the external” (rangwai bi xian
annei) (Furuya 1977, 9: 115). How the regional forces responded to Chiang’s
actions in suppressing the Communist forces and resisting the Japanese thus
 The fate of semi-centralism 231
reflected the actual relationship between the central government and regional
military factions.
Both Nanjing and the regional semi-independent forces saw the Communists
as their enemies but fought the Red Army with different strategies. In his oper-
ations against the communist forces, Chiang typically let the regional forces
combat the enemy whenever possible, a strategy that worked to his advantage
no matter what the result was, because the weakening of either the Communists
or the regional forces or both that he distrusted would always enhance the pos-
ition of his own army and help expand his control into the areas under regional
forces. When chasing the Red Army, he sometimes allowed the enemy to enter
the area controlled by regional forces so that he would have good reason to send
his troops into the area and thereby establish his control there. The regional
forces, for their part, fought the communist forces only to prevent both the
enemy and Chiang’s Central Army from entering their area. For them, the very
existence of a powerful communist force would keep Chiang preoccupied with
the operations against it and thus improve their own opportunities to survive.
Therefore, whenever possible, they would avoid a direct conflict with the Red
Army to save their own forces. The Red Army thus easily survived the Nationalist
forces’ first and second blockade lines along the border areas between Guangdong
and Jiangxi provinces in October and November 1935 after its defeat in Chiang’s
fifth annihilation campaign and withdrawal from its base area in Jiangxi, thanks to
a secret agreement between the CCP and Chen Jitang, by which the former
agreed not to proceed deep into Guangdong province and not to station or build
a base area there, and the latter promised not to combat the Red Army on its way
passing through his territory, though in actuality he provided the former with
a large quantity of salt and hundreds of chests of ammunitions (ZHWS, 3: 303).8
Li Zongren and Bai Chongxi, likewise, deliberately allowed the Red Army to
peacefully pass through northern Guangxi province in late November 1934 with-
out blocking it head on as Chiang Kai-shek had ordered and only made a gesture
of chasing it after most of the communist troops had passed their territory. Later
in his memoir, Li thus justified his action:

Chiang stationed his army in northern Hunan, allowing the Red Army to
march westward, and then ordered the Central Army to move south slowly
and force the Communist troops to enter Guangxi, … expecting that both
our force and the Communist force suffer from a mutual fight and give the
Central Army an excuse to enter and occupy Guangxi. How sinister this is!
(LZR: 478)

Long Yun, too, ordered his troops to allow the Red Army to pass through
Yunnan province peacefully and then pretended to chase it without real combat
while making every effort to prevent Chiang’s Central Army from entering Kun-
ming, the capital city of Yunnan; to please Chiang, Long nevertheless removed
a few county heads from their positions for “incompetence in annihilating the
Communists” (WSZL, 62: 14, 130). Later, the Red Army successfully crossed
232 The making of a unified and centralized state
the Dadu River in May 1935, thanks to the minimal resistance of Liu Wenhui’s
troops, who were there just for show. The Red Army would have had no
chance of surviving, had Liu fought a serious battle with his well-built fortresses
on the northern side of the river or had completely destroyed the iron-chained
bridge on the river. The only exception among the semi-independent regional
GMD leaders was Wang Jialie, who wanted to please Chiang and therefore
fought the Red Army hard, only resulting in repeated defeats and hence the
Central Army’s entry and occupation of Guizhou province in January 1935;
Wang subsequently lost his positions as the commander of his troops and gov-
ernor of Guizhou (WSZL, 93: 51–59).

National crisis and political unity

Japanese aggression was another fatal threat to Nationalist China.9 For Japan,
the rise of a centralized, strong China necessarily meant the end of its territorial
ambitions in the Asian continent; therefore, it had to do whatever it could to
prevent the Nationalist state from establishing its control of Manchuria and to
block Chiang’s unification of the entire country. A major step it took thus was
to completely occupy Manchuria in 1931 and turn this area into a puppet state
under its military control, which constituted the biggest blow to Chiang’s quest
for unification. But Japan’s ambition did not stop there; in the following years it
kept pushing its military presence down to the North China provinces neighbor-
ing Manchuria, which inevitably led to the direct confrontation in July 1937,
hence the outbreak of Japan’s full-scale invasion of China’s interior provinces
and its subsequent occupation of much of East and Central China over the next
eight years (1937–1945), which was the severest crisis of national survival in
China’s modern history and the biggest frustration in its century-long course of
state-rebuilding.
Here again a huge schism existed between Chiang and the regional semi-
independent forces. As an ambitious political and military leader, Chiang’s
nationalist feeling and commitment to China’s independence were strong, but
when it came to the making of his government’s diplomatic policies, Chiang
turned out to be exceptionally cautious, and he was particularly pessimistic of
China’s abilities to resist the Japanese aggression. For years before and after the
Incident of September 18, 1931, he had insisted on a non-resistance policy
regarding the Japanese military offensives, claiming that “they [the Japanese]
can occupy all of the coastal areas of our country, no matter which locality it is,
in a matter of three days rather than ten days” (GMWX, 72: 361). This policy
led to the Japanese occupation of Manchuria in four months by February 1932.
The Tanggu Truce signed between the Nationalist government and the Japanese
military authority in May 1933 further allowed the Japanese to occupy the areas
north of the Great Wall. But the leaders of regional military forces thought dif-
ferently. Those who controlled the northern provinces directly faced the Japan-
ese threats and therefore were the most ardent in resisting the Japanese
intrusion; Song Zheyuan’s 29th Army and Zhang Xueliang’s Northeastern Army
 The fate of semi-centralism 233
thus were among the first to combat the invading Japanese forces in the areas
north of the Great Wall in early 1933. When the conflicts with the Japanese
subsided in the following three years, however, the regional leaders of the
northern provinces, including Song Zheyuan, Yan Xishan, and Han Fuqu,
tended to manipulate both the Japanese and the Nanjing government; they
sought compromise with the former while bargaining with the latter for greater
control of the administrative, fiscal, and military power within their provinces.10
The biggest challenge to Chiang Kai-shek, to be sure, came from Zhang Xue-
liang of the Northeast Army and Yang Hucheng of the 17th Army. Unwilling
to carry out the offensive against the Red Army, Zhang and Yang acted together
to detain Chiang Kai-shek in Xi’an on December 12, 1936 (henceforth known
as the Xi’an Incident), though they later released him after a two-week negoti-
ation in which Chiang verbally conceded to their demands for terminating the
civil war against the Red Army. Because of his unexpected detention that put
the country on the verge of fragmentation and civil war at a time when it was
already vulnerable enough to the imminent threat of a full-scale Japanese inva-
sion, Chiang won almost unanimous sympathy and support from the entire
nation. His return to Nanjing was a nationwide celebration.
After the outbreak of the Lugouqiao (Marco Polo Bridge) Incident on July 7,
1937, which signaled the Japanese full-scale invasion of China, Chiang Kai-shek
eventually established himself as the undisputed leader of the nation when he
showed his determination to lead the country in the War of Resistance and
when all of the semi-independent regional forces, including the communists, ral-
lied under the Nationalist government for the concerted effort of fighting the
Japanese. Emblematic of Chiang’s unprecedented influence within the GMD as
a national leader, he was elected the Party’s Zongcai (general director) at the
GMD’s provisional congress in Hankou in April 1938, a newly created position
that enabled Chiang to firmly control the party and yield greater power than
Sun Yat-sen had had as the Zongli of the party. In July 1940, the seventh ses-
sion of the GMD’s Fifth Congress further designated July 7 as the date on
which all members of the GMD swore their loyalty to the Zongcai. Chiang con-
gratulated himself by saying: “Having fought for thirty years for our party and
state, I eventually receive recognition today; having vacillated for fifteen years,
our party eventually stabilizes today” (GMDD, 2: 647).
During the early years of the War of Resistance, China achieved an excep-
tional degree of political and military unity for the first time in its modern his-
tory. The regional forces actively participated in the war and, in fact, played
a key role in the major operations against the Japanese after the battles in
Shanghai, in which Chiang Kai-shek put more than 700,000 of his best-
equipped troops into the war, though about half of them were lost in the three
months from August 13 to November 12, 1937. Among the regional forces,
more than 600,000 from Sichuan province were killed, injured, or missing
during the eight years of the war, accounting for about one-fifth of all Chinese
casualties in the entire country (Ma and Wen 1986: 276); in Guangxi, more
than 800,000 were mobilized and about 120,000 suffered casualties (Gao
234 The making of a unified and centralized state
Xiaolin 1999); Long Yun dispatched about 200,000 from Yunnan to the fronts
and more than half of them suffered casualties (Xie and Niu 1990: 199).
Because of their heavy loss on the battleground and Chiang’s military restructur-
ing, most of the formerly semi-independent regional forces were eliminated or
greatly weakened by the end of the war. Zhang Xueliang’s Northeast Army was
divided and dispatched to different parts of the country after the Xi’an Incident
and Zhang himself was put under custody for the rest of his life; Song Zheyuan
died in 1940 after his 29th Army suffered huge casualties in the wars in North
China and was reorganized by Chiang; Han Fuqu was shot because of his defi-
ance of Chiang’s order and loss of Shandong; Liu Xiang died mysteriously in
1938 and his province came under Chiang Kai-shek’s complete control after the
Nationalist government was relocated to Chongqing in December 1937; Long
Yun unexpectedly lost his control of Yunnan in October 1945 when his army
was dispatched to Vietnam to accept the Japanese surrender. The exceptions
were mainly the forces from Guangxi under Li Zongren and Bai Chongxi and
from Shanxi and Suiyuan under Yan Xishan and Fu Zuoyi, who would plague
Chiang during the upcoming civil war after the Japanese surrender.

The semi-centralism of the nationalist state

The GMD made headway in building a unified and centralized state before and
during the War of Resistance. It recovered China’s territorial entity and sover-
eign rights. It made solid steps in curbing and eliminating the centrifugal forces
within the country. By the outbreak of Japan’s full-scale invasion in July 1937,
Chiang Kai-shek had established his firm control of the Nationalist party by out-
competing his opponents within the top leadership of the Party; the central
apparatus of the party-state expanded its control from the few provinces in East
China into other regions, in particular, the southern and southwestern prov-
inces. A centralized framework thus was largely established throughout the prov-
inces in administrative, military, fiscal, and educational realms. Fiscally, it
controlled the indirect taxes nationwide through the vigorous measures of cen-
tralization and standardization, and the greatly widened basis of tax revenues
fueled the Nationalist state’s efforts to seek military and administrative central-
ization throughout the country. These breakthroughs made it possible for
Chiang and his government to mobilize the fiscal, military, and political
resources on a national level in resisting Japan’s aggression, to implement the
strategic relocation of its political center and military forces from East China to
southwest China after losing the battle in Shanghai in November 1937, and to
act as the most important force among the Allies against Japan in the Far East
until the outbreak of the Pacific War in December 1941. Afterwards, the
Nationalist force, roughly 4 million troops, worked “to tie down some half
a million or more Japanese soldiers, who could otherwise have been transferred
elsewhere” (Mitter 2013: 379). Had the Japanese full-scale invasion occurred
before Chiang consolidated his national leadership, China’s chances of surviving
would have been truly slim.
 The fate of semi-centralism 235
In fact, the Nationalist state not only survived and finally triumphed over
Japan at the end of World War II, but miraculously joined America, Russia, and
Britain to become a member of the Big Four at the Cairo Conference in 1943,
dictating the formation of the postwar international order in the Far East, and
later to become one of the five permanent members of the UN Security Council
in 1945.11 While the offensives by the Allied forces directly accounted for
Japan’s unconditional surrender, China’s survival during the eight years of Jap-
anese aggression and its enormous contribution to the Allied forces’ defeat of
Japan in the last few years of World War II had much to do with the National-
ists’ state-building efforts before and during the war.12
It is equally clear, however, that the Nationalists’ state-making efforts were far
from a solid success. The key to understanding the progress and setbacks in the
Nationalists’ state-making efforts is “semi-centralism” in the Nationalist state’s
fiscal constitution, government organization, and political identity. This is best
seen as the result of the ramification, dilution, and distortion of the centralized
regionalism that had once made the Guangdong-based GMD force the most
powerful regional state before 1927. The highly centralized and unified fiscal
system that Song Ziwen established in Guangdong generated as much as 80 to
90 million silver yuan annually, or nearly 150 million silver yuan annually by
1928 if the revenue generated by government bonds were included, thus turn-
ing GMD into the fiscally strongest regime among all regional contenders and
making possible its military unification of China. After its establishment of
a national regime in 1927, however, Nanjing was unable to apply the Guang-
dong model to the rest of the country, nor could it establish a highly centralized
and unified fiscal system at the national level. Its enlargement of government
revenue relied primarily on maritime customs, the unified tax, salt tax, and other
indirect taxes as well as the sales of government bonds. The growth of govern-
ment revenues lagged far behind the skyrocketing expenses of a national regime
that was burdened with a rapidly expanding government body and non-military
expenditures (see Chapter 10 for details), which in turn curtailed government
spending on the military. By 1936, for instance, Nanjing’s military spending was
only about 555 million silver yuan, or 4.23 times its level in 1927, while non-
military spending grew 68 times during the same period, reaching 1.338 billion
yuan in 1936 (Yang Yinfu 1985: 70). The constraints on military spending
made it difficult for Chiang Kai-shek to completely subjugate the regional forces
and achieve real unification of the country.
The contrast is also stark in the GMD’s organizational solidarity and political
identity before and after 1927. In its early years in Guangdong, the GMD lead-
ership was relatively unified, which, coupled with its alliance with the CCP and
engagement in political propaganda against imperialism and warlords, made the
Nationalist Army a high-morale combat force in the Northern Expedition. After
1927, however, Chiang Kai-shek faced huge difficulties in his quest for political
unity within the Nationalist Party, especially in dealing with the defiant Canton-
ese elites. Beyond the power center, Chiang had to battle the regional forces
that had been incorporated into the Nationalist regime but remained alienated
236 The making of a unified and centralized state
from Chiang and autonomous to Nanjing before and during the Japanese inva-
sion, despite their alleged loyalty to the Nationalist state and acceptance of
Chiang’s national leadership. Unable to put the party-state apparatus under his
personal control and use it effectively for national goals, Chiang had to turn to
statism (guojiazhuyi, a doctrine calling for the prioritization of national goals
and the enlargement of state power at the expense of individual rights) and, for
a moment, imported fascism to rally support for his personal dictatorship and at
the same time relied on the intelligence and secret service agencies (under the
central government’s military council and the Party’s central organ, known as jun-
tong and zhongtong) as well as fascist organizations (most notably, the Blue Shirts,
or lanyishe) to enhance his power (Eastman 1974: 31–84). In the absence of pol-
itical identity based on shared ideology and commitment, Chiang’s leadership was
increasingly based on the use of violence and terror against dissidents and challen-
gers on the one hand and reliance on traditional personal bonds and small-group
solidarity on the other. It lacked a well-developed social and political basis. Zhang
Zhizhong (1890–1969), one of Chiang’s most-trusted generals, observed that,
after 1927,

factional consciousness dominated the [Nationalist] Party’s organizational

relations, thus turning the Party from the one that had centered on its
ideology and revolutionary commitment into the one that centered on fac-
tionalism and even individuals. Party members joined the Party only for per-
sonal power, rather than the revolution, thus forcing the motivated
revolutionaries to quit.
(ZZZ : 252–253)

Finally, on the local level, the Nationalist state failed to effectively penetrate
society. Despite the GMD state’s reorganization of government organs and
implementation of the self-government program at the county level and below,
malfeasance and corruption by local officials in revenue generation and manage-
ment, which had long been the central authority’s concern throughout the
Qing, remained rampant and worked to undermine the power and legitimacy of
the regime. “Local bullies and evil gentry” (tuhao lieshen), which had been the
target of the Nationalist Revolution from its very beginning, remained active at
the grassroots level, often in collaboration with government officials for self-
aggrandizement at the expense of local communities (Wang Qisheng 2003;
Thornton 2007). To be sure, compared to the Qing and early Republican
states, the Nationalist regime undoubtedly extended its administrative arm fur-
ther into the village communities, especially in provinces under its direct con-
trol, as seen, for instance, in the villagers’ participation in the election of local
government offices and the investigation of unregistered “black land,” which
resulted in the uncovering of much more illegal holdings to be officially sur-
veyed and taxed by the state than before. The state’s nationalist propaganda also
came to influence rural communities and shape the discourse of local elites (H.
Li 2005). Nevertheless, traditional community ties based on kinship and
 The fate of semi-centralism 237
neighborhood or supra-village networks based on collective undertakings in
defense, water control, and religious activities continued to dictate the param-
eters of the peasants’ social space and the self-governing mechanisms of village
communities, a phenomenon that was especially true in places where the state’s
efforts of bureaucratization failed and traditional protective elites yielded to the
rise of profiteering “entrepreneurial brokers” in taxation (Duara 1991). The vast
majority of ordinary rural dwellers continued to identify themselves primarily with
their clan, village community, or the immediate area surrounding their village,
rather than with the entire Chinese nation or the state above them. Mobilizing the
peasant population for state-making projects remained a task beyond the GMD
regime’s capability. Preoccupied with defeating warlord and CCP forces in the late
1920s and early 1930s, the Nationalist state lacked the time and energy to pene-
trate rural society by establishing an effective administrative network reaching every
household, thus leaving the collection and use of the land tax, which could have
been the largest source of revenue for the central government, to provincial and
local authorities. Much of the failure of the GMD in competing with the CCP after
1945, therefore, had to do with the bankruptcy of its fiscal system, which was
plagued with runaway inflation when its limited revenues from the indirect taxes
and borrowing failed to sustain the rapidly escalating spending on the military and
feed the huge size of civil and military personnel, hence the low morale of its troops
and the quick erosion of the state’s legitimacy among the urban elites.
To recapitulate, both the breakthroughs in Guomindang’s quest for a modern
nation-state and the nationalism of the contending regional forces accounted for
the Nationalist regime’s resilience in withstanding Japanese aggression in the 1930s
and the first half of the 1940s. However, Chiang’s failure to build a party-state on
ideological and organizational fronts undermined the formation of political identity
among his followers and supporters. Chiang’s national leadership and the tenacity
of the Nationalist regime during the eight years of Japanese aggression, in other
words, was a result of the shared commitment among the different regional forces
to China’s survival at the time of an unprecedented national crisis rather than the
fruition of Chiang’s state-building efforts. Political unity under Chiang was fragile
during the War of Resistance. Severe divisions already existed within the GMD
party-state between Chiang and Wang Jingwei, ending in the latter’s creation of
another, pro-Japanese Nationalist state in Nanjing in 1940, and between the GMD
and the CCP in the early 1940s. Once the national crisis was gone and the
common ground disappeared for the regional forces to stay together after the Jap-
anese surrender in 1945, Chiang’s national leadership would be subject to chal-
lenges from the regional forces within the Nationalists and, more fatally, from the
CCP, and the Nationalist state’s viability would be in jeopardy.

Notes
1 The inability of the central government to penetrate individual provinces, it is
argued, forced Chiang to build his leadership by focusing on the use of the military,
the cultivation of personal loyalty to himself among party and military elites, and
238 The making of a unified and centralized state
various forms of political repression during the decade of the Nationalist rule in Nan-
jing (1927–1937) (Tien 1972; Eastman 1974). Historians further find that, after
losing the best of the armies under his direct control at the beginning of the War of
Resistance in 1937 and again in the Japanese offensives in 1944, Chiang’s party-state
apparatuses verged on a paralysis because of his inability to control the resistant
regional militarists and because of the demoralization of the Nationalist government
itself (Ch’i 1982; Eastman 1984). More recent studies offer a different image of the
Guomindang state. The Nationalists, Julia Strauss argues, were “surprisingly success-
ful” in institutional building, as best seen in the organization and operation of
a number of central government institutions, especially in the bureaucratization and
professionalization of the personnel of such institutions, which, she suggests, signified
the government’s continuation of the “top-down, centralizing, state-building” initia-
tives of its late Qing and early Republican predecessors (Strauss 1998: 8, 25). At the
local level, state capacities also grew, as evidenced in Elizabeth Remick’s study of two
counties in Guangdong province, where the government enhanced its penetration
into village communities, thus resulting in increases in both its revenues and expend-
itures, without having to weaken the existence of local elites (Remick 2004). William
Kirby finds in the Republican period breakthroughs in China’s international cooper-
ation on all fronts and impressive records in improving its international standing
(Kirby 1984, 1997; see also Taylor 2009: 589–590). He further describes the Guo-
mindang government during the Nanjing decade as a “developmental state” for its
use of trained bureaucrats and technocrats to formulate plans for state-led industrial-
ization and for its commitment to integrated economic and industrial developments,
which were later inherited by the Communist state after 1949 (Kirby 2000). Along
this line, Morris Bian argues for institutional rationalization in China’s ordnance
industry and related heavy industries in the 1930s and 1940s in response to the Jap-
anese aggression (Bian 2005). In his study of wartime Wuhan, Stephen MacKinnon
depicts the ten-month defense of the city against the Japanese assault in 1938 as
a legendary moment of shared responsibility and cooperation among the Nationalist
central government, military commanders of non-Whampoa origins, and the Com-
munists, which permitted China to survive the war against Japan; the ordered with-
drawal from the city stood in sharp contrast to the collapse that had occurred with
the fall of Shanghai and Nanjing (MacKinnon 2009).
2 At the inaugural ceremony of the Committee for Reorganizing and Discharging the
Nationalist Army on January 1, 1929, Chiang made the following remark: “What is
the condition for the making of a modern state? The first is unification, and
the second is centralization” (GMWX, 24: 4–6).
3 Together with the low and fixed tariff rate under the Qing government’s treaties with
foreign powers that made the prices of imported goods competitive to domestic
products, lijin has long been seen as another major hurdle to the growth of domestic
industry and commerce because of the repeated collections of dues on native goods
at different localities in their transportation to destinations. The heavy burden of lijin
was also the very excuse by which foreign powers had insisted on the fixed tariff rates
on imported goods and refused to agree with China’s tariff autonomy. Since its
inception in the 1850s, lijin had also constituted the major source of revenue for
local military and civilian authorities and been out of the central government’s direct
control; it thus formed the fertile soil on which regional autonomy developed in the
late Qing and early Republican years. Where the access to the capital market was
limited or nonexistent, the collection of lijin became the most important source of
income for warlords in the interior provinces in the 1910s and 1920s. Abolishing
lijin, therefore, was not only necessary for the healthy development of domestic trade
and industry but also essential for recovering China’s tariff autonomy and weakening
the financial base of regional warlord forces.
 The fate of semi-centralism 239
4 Feng Yuxiang, for instance, formally announced his acceptance of Sanminzhuyi
upon pledging his loyalty to the GMD in September 1926 and leading his army to
join the National Revolutionary Army under Chiang Kai-shek’s leadership. Earlier,
to convert Feng into a pro-GMD revolutionary, Sun had sent him 6,000 copies of
pamphlets on Sanminzhuyi and an additional 1,000 copies of his own works when
visiting Beijing in early 1925 (Feng and Luo 2009: 6). Yan Xishan, too,
announced his acceptance of Sanminzhuyi in June 1927 upon being appointed by
Chiang as “the General Commander of the National Revolutionary Army in the
North” (SSJY, 1928, Jan.-Jun.: 4). Zhang Zuolin’s Fengtian clique was once the
leading force in “punishing the reds” (taochi), that is, warring with the GMD’s
Northern Expedition forces, in 1926 and 1927, allegedly because of his hostility
to the “radical” ideology of the Guomindang when it was under the influence of
the Chinese Communist Party (CCP) and Soviet advisors. Therefore, Zhang
Zuolin never officially accepted Sanminzhuyi. Nevertheless, to seek compromise
and collaboration with the Nationalist force after Chiang Kai-shek purged the CCP
members from the Nationalist forces in April 1927, Zhang Zuolin expressed his
willingness to “carry out Mr. Sun Yat-sen’s unfinished will, not to betray the prin-
ciples of Sanminzhuyi, and work together with Mr. Sun Yat-sen’s followers to
fight the Communist Party” while blaming the “radicals” for “misunderstanding
Mr. Sun’s ism.” After his father’s death, Zhang Xueliang lifted the ban on pro-
GMD magazines, newspapers, and books in Manchuria when negotiating with the
Nanjing government over the terms of switching his loyalty to the latter. He finally
announced his obedience to the Nationalist government on December 29, 1928
by hanging the flag of Blue Sky and White Sun and acknowledging his “adherence
to Sanminzhuyi” (zunshou Sanminzhuyi) (BYJF, 5: 895).
5 Wang Zizhuang, a secretary of the Guomindang central department, thus remarked
in the late 1930s: “For more than ten years since the Northern Expedition, books on
the Party’s doctrines are rare. Not only did [the Party’s doctrines] find no place in
social sciences, but studies that interpret them are also few” (WZZ, 5: 102).
6 The Tongmenghui, the predecessor to the GMD, to be sure, was much more open
because of its establishment in 1905 in Tokyo on the basis of several anti-Manchu
organizations, including the Revive China Society (Xingzhonghui) of mostly overseas
Cantonese, the China Revival Society (Huaxinghui) of the revolutionaries from
Hunan, and the Renaissance Society (Fuxinghui) of the students from Zhejiang prov-
ince. However, after Sun Yat-sen established the Chinese Revolutionary Party in
1914 and later engaged in the Pro-Constitution Movement and the Northern Exped-
ition, Sun turned primarily to the Cantonese for membership of his party (renamed
Zhongguo Guomindang, the Chinese Nationalist Party, in 1919) and based his mili-
tary activities mostly in Guangzhou, which inevitably resulted in the dominance of
the Cantonese in the leadership of the party (Jin Yilin 2005: 116). Not surprisingly,
at the first two national congresses of the GMD, Cantonese was used as the official
language, and all speeches and proposals by the Mandarin-speaking delegates had to
be translated into Cantonese.
7 Later, Chiang continued to rely on his fellow natives of Zhejiang in maintaining his
control of the party-state, among whom were the brothers of Chen Lifu
(1900–2001) and Chen Guofu (1892–1951), who stewarded Guomindang’s organ-
izational growth; Chen Cheng (1898–1965), Tang Enbo (1898–1954), and Hu
Zongnan (1896–1962), Chiang’s most-trusted military commanders; Dai Li
(1897–1946), who oversaw Guomindang’s intelligence agency; Xu Entong
(1896–1985), head of the Guomindang government’s secret service; and Chen Bulei
(1890–1948), Chiang’s ghostwriter.
8 A loyal follower of Sun Yat-sen and a commander of the Nationalist force in Guang-
dong in constant rivalry with Li Zongren’s Guangxi clique in the late 1920s, Chen
240 The making of a unified and centralized state
Jitang (1890–1954) allied with Li and other forces in Guangdong and Guangxi to
openly challenge Chiang after the latter detained Hu Hanmin; after 1931 he
remained the legal ruler of Guangdong in his capacity as a key member of Guomin-
dang’s Southwestern Executive Department and the Nationalist Government’s South-
western Administrative Council in Guangzhou, making the province largely
independent of Nanjing until 1936 when he failed in his collaboration with Li on
a military campaign against Chiang.
9 As mentioned in Chapter 4, since the 1870s Japan had replaced the European
powers to be the major and immediate threat to China. As it turned out, Japan’s
annexation of Liuqiu in 1879 and Taiwan in 1895 were only the first steps in its long
quest for expansion. Defeating Russia and turning southern Manchuria into its sphere
of influence in 1905 marked the beginning of Japan’s aggression against China’s
northeast provinces, which naturally became its next target of expansion after annex-
ing Korea in 1910. In the following years, Japan not only secured its exclusive privil-
eges in southern Manchuria, but repeatedly attempted to expand its influence to
eastern Inner Mongolia despite the Yuan Shikai regime’s firm resistance. Japan’s
aggression became truly fatal to Nationalist China in the late 1920s and early 1930s,
when the latter finally ended the century-long trend of disintegration and, for the
first time, showed signs of the emergence of a unified, modernizing state under the
leadership of Chiang Kai-shek.
10 Song, for instance, was appointed as the chairman of the Hebei and Chahar
Administrative Council (Ji-Cha zhengwu weiyuanhui) under the Nanjing govern-
ment in December 1935 to legitimate his control of the vast areas of North
China. Indicative of his semi-independence in relation to Nanjing, Song told his
subordinates: “We won’t talk about separation from the central government, but
we would never be a tool for Chiang Kai-shek’s personal manipulation” (cf. Chen
and Chang 1997: 204). Unlike the northern regional leaders who had at once to
resist and conciliate with the Japanese, the regional leaders of the southern and
southwestern provinces appeared to be consistent in articulating their resistance to
Japan, partly because of their nationalism and partly for the purpose of building
their own legitimacy as patriotic and democratic forces in rivalry with Chiang Kai-
shek. On two important occasions, therefore, they openly challenged Chiang
because of their disagreement with Chiang’s non-resistance policy as well as dicta-
torship. The first was the action taken by the commanders of the 19th Army and
their supporters to establish the People’s Revolutionary Government of the Chin-
ese Republic in Fuzhou of Fujian province in November 1933 (known as the
Fujian Incident), who had been known nationwide for their heroic combat against
the Japanese troops in Shanghai in the famous battle of January 28, 1932.
The second was the joint military offensives by Chen Jitang of Guangdong and Li
Zongren and Bai Chongxi of Guangxi against Chiang’s Central Army in
June 1936 (known as the Guangdong-Guangxi Incident) in resisting Nanjing’s
attempt at unifying the administrative systems in the two southern provinces after
the death of Hu Hanmin. These anti-government actions, however, received little
sympathy from the general public at the times of worsened crisis of national sur-
vival. Counting on the superiority of his military forces and the support of public
opinion, Chiang suppressed the rebellion in Fujian. The allied forces of Chen, Li,
and Bai were also aborted because of the defections from within Guangdong;
Chen left Guangzhou and completely lost his influence there while the Guangxi
leaders remained largely intact.
11 For a discussion of the Nationalist state’s successes in rebuilding China’s international
relations, see Kirby (1997).
12 It is also important to note, however, that some of the regional forces who played
a prominent role in the War of Resistance, in particular those from Guangxi, Yunnan,
 The fate of semi-centralism 241
and Sichuan, accepted Chiang’s leadership not because they had been fully incorpor-
ated into the Nationalist party-state; far from that, they had been consistently treated
by Chiang with suspicion and among the first to be deployed to the front combatting
the Japanese and often suffering more casualties than Chiang’s own forces. Neverthe-
less, these regional leaders voluntarily dispatched the best of their forces from the
remote southern or southwestern provinces to East and Central China and actively
participated in the most important campaigns in Xuzhou, Wuhan, and Changsha.
The motivation behind their self-sacrifice had more to do with their commitment to
China’s survival as an entire nation, which outweighed their calculation of personal or
regional interest. Nationalism, in other words, served as the common ground uniting
the central and regional forces that had been in conflict for years. The formation of
a united front among all political and military forces and the establishment of Chiang
Kai-shek’s national leadership, in a nutshell, were the result of both the Nationalists’
state-making efforts and the regional leaders’ prioritization of national survival at the
expense of self-interests, which in turn had to do with their need for legitimization in
twentieth-century China in which nationalism reigned supreme in political discourse.
10 Total centralism at work
The confluence of breakthroughs in
state-making

To understand the rise and eventual success of the Communist Revolution, it is

important to distinguish between three interrelated yet substantially different
phases of the event. The first is the nine years from the CCP’s break with the
GMD in 1927 to its defeat by the Nationalists and subsequent relocation to
Shaanxi province in 1936. This was when the communist force grew in the
mushrooming base areas in south-central China by taking advantage of the Nan-
jing government’s preoccupation with war against regional forces in the late
1920s and early 1930s, but suffered huge losses as a result of the Nationalists’
annihilation campaigns. The second is the eight years of the War of Resistance
from 1937 to 1945, when the communist forces thrived in North China. Com-
pared to its size in November 1935 when the central force of the CCP arrived in
northern Shaanxi, with only about ten thousand soldiers, by the end of this period
the communist regular forces had expanded to more than 910,000 soldiers occupy-
ing 19 base areas, with a total population of nearly one hundred million (Pang and
Jin 2011, 1: 372, 2: 713–714), thus constituting the most formidable challenge to
the Nationalist regime.1
Nevertheless, no matter how phenomenal its growth was during the War of
Resistance, the communist forces by 1945 were far from definitive in shaping
the result of the Communist Revolution in 1949. For decades, the CCP was
disadvantaged in relation to the GMD regime. Geopolitically, its guerrilla
forces were confined to the poverty-stricken rural areas of the southern and,
after 1935, northwestern provinces. Without access to the goods, funding, and
advisors provided by the Soviet Union, which had been critical to its rise in
the 1920s, the communist forces remained poorly equipped and small in size,
despite their penetration into the “rear areas” under Japanese occupation. Politic-
ally, the Party struggled with factionalism at the top level, between the leaders
who had returned from Moscow and the indigenous leaders led by Mao, and
with the division between the communist forces of different base areas. In spite of
its quick expansion during the War of Resistance, the communist forces were
inferior to the Nationalists in almost all regards during the interim from Japan’s
surrender in September 1945 to the outbreak of the Civil War in June 1946.
There were no clear signs that the CCP was bound to defeat the GMD on the
eve of the Civil War.
 Total centralism at work 243
What was decisive to the CCP’s fate, then, were the developments during the
Civil War years (1946–1949), the third phase of the Communist Revolution.
The conventional wisdom typically explains the CCP’s success in the Civil War
as a result of its mobilization of the rural population through land reform and
its recruitment of a large number of peasant soldiers and laborers for logistic ser-
vice (e.g., Pepper 1978), while attributing the GMD’s failure to the entrenched
corruption of its bureaucrats dating back to the late 1920s and 1930s which
only worsened after its takeover of the cities from the Japanese in 1945, the
unpopularity of Chiang’s insistence on a war to eliminate the CCP, and most
fatally, the lack of coordination among its troops and runaway price inflation in
the cities (Pepper 1978; Eastman 1984). More recent interpretations accentuate
the military factor by focusing on the key role of the battles in Manchuria in
shaping the result of the Civil War without having to downplay the significance
of land reform in strengthening the communist forces (Levine 1987; Tanner
2013, 2015), or to focus on the social impact of the war (Lary 2015), or to
underscore Mao’s superior leadership and strategic thinking as well as his gen-
erals’ combat experience (Cheng 2005; Lew 2009), or to pinpoint the GMD’s
mistakes in military strategies and tactics in fighting the CCP in Manchuria and
East China while discounting the importance of land reform to the CCP and
the damage of the GMD regime’s failures in economic and financial policies to
its legitimacy (Westad 2003).
To inquire into the reasons behind the CCP’s surprising success, this chapter
looks at three key factors, namely the geopolitical settings in which the Revolu-
tion unfolded and redefined its goals, the fiscal constitutions of the communist
regime that ultimately shaped its military strength or weakness, and identity
building that determined the CCP’s ability to tap and employ resources for its
strategic goals. We will find that the CCP achieved breakthroughs in each of the
three phases of the Communist Revolution, and it was the confluence of these
breakthroughs during the last phase that accounted for its success in 1949. The
CCP saw its first breakthrough in the late 1920s and early 1930s in penetrating
peasant society and building a fiscal regime on the basis of a subsistence econ-
omy; it further witnessed success in identity building afterwards, which resulted
in the firm establishment of Mao’s leadership of the Party and the unity of its
political and military forces which existed in different base areas by 1943. How-
ever, the most important breakthrough took place after 1945, in which the
communist force capitalized on a totally different geopolitical setting in its favor
and therefore transformed its fiscal regime from relying on poverty-stricken rural
areas to being centered on Manchuria, the economically most productive and
strategically most important region in China. It was the CCP’s combination of
its strength built before 1945, that is, its ability to mobilize rural support and
maintain high-level identity, with its rebuilt fiscal-military strength after 1945,
that enabled the communist force to fight the Nationalist forces on an equal
footing and to experience a shift in its military strategy from prioritizing guer-
rilla warfare, which was defensive most of the time prior to 1945, to regular
warfare, which became increasingly offensive by 1947.
244 The making of a unified and centralized state
Political identity under Mao

Mao’s quest for leadership: from military to political

Both Chiang Kai-shek and Mao Zedong emerged as the supreme leaders of the
Nationalist and communist forces, respectively, by the late 1930s. But the ways of
their quests for dominance were different. Chiang bolstered his leadership invari-
ably with his control of the military; it was the unparalleled fiscal and military
strength of the Guomindang forces that enabled him to subjugate the warlords one
after another by the early 1930s. It was also on the basis of his military power that
Chiang triumphed over his opponents within the Guomindang. Nevertheless,
Chiang’s leadership was fragile and incomplete. Instead of eliminating his rivals
within and outside the Guomindang, he established his dominance primarily by
compromising with them and incorporating them into the military, government,
and party organizations under his control, an approach that minimized his efforts at
state building but also rendered his leadership unstable and ineffective. Despite the
momentary reputation as a national leader that he gained during the War of Resist-
ance against Japan and shortly after the Japanese surrender, Chiang exercised his
power primarily as a military strongman, rather than a charismatic leader with
unchallenged legitimacy and influence within the Nationalist party-state. Ideologic-
ally, he was a poor interpreter of Sun Yat-sen’s Sanminzhuyi; he never developed
a set of ideas in his name, not to mention a consensus within the Party on the
orthodoxy of his thoughts. In a word, Chiang lacked soft power; he based his
rule mainly on military strength, rather than ideological persuasion and spiritual
inspiration.
Mao Zedong’s leadership of the CCP, too, began with his control of military
power within the Party. As one of the founders of the Red Army, Mao lost his
position as the “general political commissar” and a key commander of the army
at a meeting of Party leaders in Ningdou in October 1932, and was subse-
quently ousted from the Party’s leadership (Pang and Jin, 1: 297–300). It was
after the Red Army suffered repeated defeats, resulting in huge losses from
87,000 to about 30,000 soldiers during the first few months of its retreat from
the Jiangxi base area, that Mao was able to make suggestions redirecting the
Army’s route of retreat in his capacity as a member of the CCP’s political
bureau in late 1934. At the famous meeting at Zunyi of Guizhou province on
January 15–17, 1935, Mao was elected as a standing member of the Party’s pol-
itical bureau and named as an assistant to Zhou Enlai, the supreme commander
of the Red Army. After the meeting, Zhou, Mao, and Wang Jiaxiang soon
formed the Trio (sanrentuan), with Zhou as the head, to command the army,
and Mao to play a key role when making decisions on the army’s operations
(ibid.: 353). Later in August 1935, at a meeting of the political bureau, Mao
was officially put in charge of the Party’s military affairs (ibid.: 364). His military
leadership was further confirmed by his appointment as the chairman of the
CCP’s Revolutionary Military Committee in December 1936 after the Red
Army migrated to Shaanxi province. The official leader of the Party, however,
 Total centralism at work 245
remained Zhang Wentian, who had been the general secretary since Febru-
ary 1935, shortly after the Zunyi meeting (Cheng Zhongyuan 2006: 140).
The biggest challenge to Mao in his rise to the Party’s top leadership came
from Wang Ming, who had been working for the Comintern in Moscow since
1931 and was sent back to Yan’an in December 1937; by that time the CCP
and the GMD had formed the United Front for the War of Resistance (HQM:
44). He was soon appointed as the secretary of the CCP’s Changjiang Bureau
in Wuhan, with Zhou Enlai as the vice secretary, to represent the CCP in com-
munication with the Nationalist government that had retreated from Nanjing to
Wuhan. Mao continued to oversee the Party’s military affairs. Contrary to
Mao’s insistence on the CCP’s autonomy, however, Wang adhered to the
Comintern’s instructions and emphasized the supreme importance of the
United Front and the CCP’s unreserved acceptance of Chiang Kai-shek’s leader-
ship, including his employment of the communist force in the regular “mobile
warfare” against Japanese troops. Behind Wang’s stance was the Comintern’s
concern with Chiang Kai-shek’s possible compromise with Japan and, subse-
quently, Japan’s attack on the Soviet Union after consolidating its control of
China. Mao, therefore, had good reason to complain later that Wang “paid too
little attention to [China’s] own affairs and cared too much about others’”
(Pang and Jin 2011, 2: 518). Backed by the Comintern, Wang saw himself as
above all other CCP leaders and acted assertively, a situation that did not
change until September 1938, when Wang Jiaxiang came back from Moscow
with the Comintern’s new message: the CCP leadership should be “headed by
Mao Zedong” and should build an “atmosphere of intimacy and unity” (ibid.:
519). Wang thus lost his standing as the Comintern’s representative within the
CCP and the influence associated with it. Zhang Wentian also proposed to hand
over his position as the Party’s general secretary to Mao, but Mao refused
(Cheng Zhongyuan 2006: 140). With the Comintern’s endorsement, Mao def-
initely established his leadership of the CCP at its Sixth Session of the Sixth
Congress from September 29 to November 6, 1938, in which he refuted
Wang’s slogan of “conformity with the United Front in everything” and stressed
the importance of the CCP’s independence under the United Front; he further
justified the necessity of “concrete application of Marxism to China” or the “com-
bination of Marxism with our country’s specific characteristics” as the CCP’s
approach to ideology building (Pang and Jin 2011, 2: 522–523). By this point,
Mao had established himself as not only the supreme commander of the CCP’s
military forces but also the de facto political leader of the Party.

Toward the hegemony of Maoism

Mao’s quest for leadership, however, did not stop there. To legitimate his
power and base it on a solid ideological base, he took further steps to create
a new set of ideas, known later as “Mao Zedong Thought” (Mao Zedong six-
iang). It was in this arena where Mao differed from, and indeed surpassed,
Chiang Kai-shek. Mao’s ability to turn his political thinking into the only
246 The making of a unified and centralized state
orthodoxy in the CCP’s ideology, coupled with his relentless efforts at organiza-
tional consolidation, accounted for the exceptional level of political identity and
conformity among the Chinese communists, which was critical to the CCP’s
success in its rivalry with the Nationalists in the years to come.
Before establishing himself as the sole ideological authority within the CCP,
Mao was only recognized as a military talent, and his influence was largely
limited to military affairs; although he was later also accepted as the Party’s de
facto leader after September 1938, his speeches and writings were not duly
respected and accepted by party elites.2 To build his ideological influence, Mao
authored a series of essays in late 1939 and early 1940. His writings had two
foils. One was Wang Ming, who saw the Chinese Revolution as consisting of
three phases, with the struggle against imperialism as the only task of the
ongoing first phase, which should be distinguished with its future tasks in
the second and third phases, namely the struggle against feudalism and the tran-
sition to socialism, respectively, and therefore stressed the importance of the
United Front and the CCP’s acceptance of the Guomindang’s leadership for the
current task of resisting Japanese aggression (Hu Qiaomu 2003: 198). The
other target was the Guomindang’s ideologists, who questioned the legitimacy
of the CCP and its Marxist ideology by advocating “one party” and “one ism.”
To refute the two enemies inside and outside the Party, Mao formulated his
own thesis of “New Democracy” (xin minzhuzhuyi). For him, the Chinese
Revolution consisted of two steps, the “New Democratic Revolution” and the
socialist revolution. The New Democratic Revolution, Mao wrote, was at once
a national revolution against the oppression of external imperialism and
a democratic revolution against the oppression of domestic feudalism, and the
struggle against imperialism was the most important task (Pang and Jin 2011, 2:
568). To finish the New Democratic Revolution, however, the Chinese proletar-
iat (namely the CCP, as the vanguards of the proletariat) should take the leader-
ship and work together with peasants, the petty bourgeoisie, and the national
bourgeoisie as a united front; the Chinese bourgeoisie, for its part, had proven
to be liable to compromise with the two enemies and therefore ineligible to be
the leader of the Revolution.
The state to be established after the Revolution, Mao further explicated,
should be neither a bourgeois republic nor a proletarian dictatorship; instead, it
should be the dictatorship by a coalition of all classes committed to the struggle
against imperialism and feudalism and under the leadership of the proletariat
(the CCP). And the government itself should take the form of “democratic cen-
tralism” (minzhu jizhongzhi). The state would permit the ownership of private
properties and not forbid capitalist production, but state-owned, socialist enter-
prises should be the leading sector of the national economy. In rural areas, the
property of feudal landlords would be confiscated and redistributed to landless
and land-poor peasants, while the rich-peasant economy would be permitted
and all kinds of cooperative economy would be encouraged. The New Demo-
cratic Revolution, in a word, would clear the way for the development of capit-
alism and at the same time brewed elements of socialism; the ultimate goal of
 Total centralism at work 247
the Chinese Revolution in the future, Mao claimed, was the transition to social-
ism (MXJ, 2: 602–614, 621–656, 662–711).
Clearly defined and coherently structured, Mao’s theory of New Democracy
immediately distinguished itself from Sanminzhuyi, which was all-encompassing
and open to highly subjective and flexible interpretations for different purposes. It
was also different from the classic Marxist theory that called for a straightforward
transition from capitalism to socialism via a class struggle between capitalists and
the proletariat. As a key step in his quest for the CCP’s ideological autonomy,
Mao’s theory of New Democracy provided the Party with an effective tool to
legitimate its existence when it was ready to break with the GMD and compete
for dominance in domestic politics in the years to come.
Before he was able to turn his theory into the ideology of the entire Party,
however, Mao had to first discredit the existing ideological authorities within
the Party. For that purpose, Mao Zedong and his assistants put together all of
the relevant documents of the Party of the preceding 13 years and compiled
them into a volume, entitled Since the Sixth Congress (Liuda yilai), published in
two versions (complete and abridged), and made them available to all of the
high-ranking cadres of the Party. Dubbed the “party book” (dangshu), this
volume systematically showed for the first time the struggles within the Party
between two different lines, the “leftist line” pursued by Wang Ming and his
followers that led to the CCP’s repeated disasters, and Mao’s “correct line” that
explained the Red Army’s earlier successes. The book produced immediate
effects. As Mao described it, “once the party book was published, many com-
rades put down their weapons” (Pang and Jin 2011, 2: 641). Those who had
sided or sympathized with Wang Ming changed their stance and supported
Mao. It was against this background that the CCP’s political bureau held its
“enlarged conference” in September 1941 (known as the “September Meeting”),
in which a consensus was achieved among the Party elites on Mao’s ideological
supremacy. The meeting began with Mao’s attack on the “subjectivism” and
“sectarianism” characteristic of the political line of Wang and Bo. He won
overwhelming support from the high-ranking party cadres, including those
who had followed or once sided with Wang Ming, such as Bo Gu, Zhang
Wentian, and Wang Jiaxiang, who made “self-criticism” regarding their mis-
takes (HQM: 194–195; Pang and Jin 2011, 2: 641–643). Mao thus finally
established his status as not only the supreme military commander and political
leader but also the sole ideological authority of the Party (HQM: 48).

Identity building through indoctrination and compulsion

Mao, however, was not content with a consensus on his political leadership and
ideological supremacy merely among the top-level party elites, which was obvi-
ously achieved at the September Meeting. To ensure the unreserved loyalty to
Mao in the entire Party from the top down to the grassroots level, to turn the
CCP into an organization with a high level of solidarity, and to free the Party of
factionalism, corruption, and lack of discipline, Mao faced two more tasks: to
248 The making of a unified and centralized state
propagandize his doctrine among all members of the Party; and, more important,
to purge from the ranks of party cadres those who had sided with his political
opponents within the Party or who were suspected of being disloyal to the CCP.
For these two ends, Mao launched the Rectification Campaign in February 1942.
The origin of the campaign can be traced to the Comintern’s resolution on
March 11, 1940, that urged the CCP to take measures to purify the ranks of
party cadres. Mao welcomed the resolution as well as Stalin’s earlier talk on the
Bolshevisation of Communist Parties in individual countries (Zhang Xide 2009),
using them as the rationale to launch the Rectification Campaign that served his
own political agenda. The campaign, therefore, first targeted more than ten
thousand party cadres and pro-CCP intellectuals in Yanan where the Party’s
headquarters was located. They were required to study Mao’s writings and other
documents of the Party and examine their own “non-proletarian thinking.”
After several months of intensive indoctrination and self-examination, Mao
believed that the “extremely chaotic and confusing ideas” that had existed
“among the large number of young as well as senior cadres and among the
intellectuals” were gone, and their thinking was “unified” (Pang and Jin 2011,
2: 657). After June 1942, therefore, the campaign was further extended to CCP
members in all areas throughout the country, targeting primarily the high and
middle-ranking cadres of regional party organizations, and the focus of the cam-
paign was on the party cadres’ examination of their problems with “liberalism”
(ziyouzhuyi, i.e., noncompliance with the Party’s discipline and directives) and
“demanding independence from the Party” (dui dang nao dulixing wenti). To
ensure the “unified leadership” (yiyuanhua lingdao) of the Party, Mao replaced
Zhang Wentian as the official leader of the CCP after he took the position as
chairman of both the political bureau and the secretariat of the Party in
March 1943 (ibid.: 659).
The campaign entered its second phase after April 1943 with a new task: to
investigate the history of each and every cadre’s political career and to expunge
the “impure” elements from the rank of party cadres, in response to penetration
and sabotage by GMD forces. But the investigation soon developed into unre-
strained persecution involving the use of torture, unverified accusation, and false
confession. In Yan’an alone, more than 1,400 individuals were identified as the
GMD’s secret agents in about two weeks (Pang and Jin, 2: 663). Mao later real-
ized the severe consequences of the indiscriminate persecution that targeted
mainly intellectuals, forbade the death penalty for any of the persecuted, and
rehabilitated most of them in early 1944. By that point, Mao had definitely
achieved what he had wanted from the campaign: a high level of political iden-
tity and organizational solidarity within the Party on the basis of a consensus of
his leadership among the entirety of the Party cadres at different levels and
ordinary party members, to be bolstered by their faith in the correctness of the
Maoist ideology as well as their fear of removal from the Party.
As if all these were not enough, Mao took two more actions to firmly anchor
his leadership on an unchallengeable ground. One was to chair a meeting of the
CCP’s political bureau on September 7 to October 6, 1943, in which he
 Total centralism at work 249
castigated the two major groups of “sectarians,” namely the “dogmatist” group
and the “empiricist” group. Since Wang Ming and Bo Gu, as the leaders of the
former group which had been completely discredited, Mao’s real target this time
was the empiricist group, and Zhou Enlai was number one among them (Shi
Zhongquan 2012: 17). To please Mao, attendants at the meeting claimed that
since the dogmatist leaders, including Wang, Bo, Zhang Wentian, and Wang Jiax-
iang, no longer constituted a threat to the Party, the empiricists, primarily Zhou,
would be the most dangerous in the future (HQM: 295). Zhou was the Party
leader with the longest service and of the highest reputation before Mao’s rise to
the top. Being partly responsible for removing Mao from the Party’s leadership at
the Ningdou meeting back in 1932, and as Mao’s superior and the supreme mili-
tary commander of the Red Army in place of Mao for years afterwards, Zhou was
fully aware of his delicate situation at that moment; he made the lengthiest and
most thorough “self-reflection” and “historical examination” among all of the
speakers at the meeting. Mao was ambivalent towards Zhou. He was resentful of
Zhou’s tolerance of, and compromise with, the dogmatist leaders on many occa-
sions, but he was also reliant on Zhou for his ability to handle the thorniest chal-
lenges from within and outside the Party. Most importantly, Zhou had never
taken the lead to openly challenge Mao and, after the Zunyi Meeting, played
a key role in supporting Mao’s leadership. Mao, therefore, did not take further
action against Zhou, but he would be on alert against any signs of his unruliness
for the rest of his life.
The last step Mao took to seal his supremacy within the CCP was to have
“the Resolution regarding Several Historical Issues” unanimously passed at the
Seventh Session of the Party’s Sixth Congress on April 20, 1945. Central to
the resolution was a new narrative of the Party’s history that highlighted the
repeated struggles between the correct political line represented by Mao and the
wrong line pursued by dogmatist Party leaders. Nevertheless, the Resolution
should not be merely interpreted as a result of Mao’s manipulation of the Party
leadership for his personal power. Nor should the Rectification Campaign be
simply seen as another round of the power struggle between Mao and his
opponents within the Party. In fact, no one had been able to challenge Mao’s
standing within the CCP since he received the Comintern’s endorsement of his
leadership in September 1938, as mentioned earlier. It was likely, to be sure,
that seeking revenge on his old-time adversaries and competitors within the
Party (including dogmatists such as Wang Ming as well as empiricists such as
Zhou Enlai) could be part of the reasons behind Mao’s launching the Rectifica-
tion Campaign. For Mao, however, the campaign as well as the drafting and
passing of the Resolution were necessary not only because they worked to turn
himself into a full-fledged leader of the Party and to gloss his leadership with
a layer of charisma, but more importantly because they were indispensable for
the CCP to achieve a high degree of political identity and organizational solidar-
ity on the basis of a party-wide consensus on his political correctness and ideo-
logical supremacy. Likewise, the party elites’ acceptance of Mao’s leadership
should not be simply interpreted as a result of their failure in rivalry with him or
250 The making of a unified and centralized state
their submission to his growing dictatorship within the Party; it more or less
reflected their consensus on his outstanding ability to lead the Red Army and
the entire Party as well as the persuasiveness of his theory on the Chinese Revolu-
tion. Therefore, the description of Mao in the opening paragraph of the Resolution
was not so much a flattery as a shared recognition among the party elites of Mao’s
competence as the leader of the Party:

Ever since its birth in 1921, the Communist Party of China has made the
integration of the universal truth of Marxism–Leninism with the concrete
practice of the Chinese revolution the guiding principle in all its work, and
Comrade Mao Zedong’s theory and practice of the Chinese revolution rep-
resent this integration. … In the course of its struggle the Party has pro-
duced its own leader, Comrade Mao Zedong. Representing the Chinese
proletariat and the Chinese people, Comrade Mao Zedong has creatively
applied the scientific theory of Marxism–Leninism, the acme of human
wisdom, to China, a large semi-feudal and semi-colonial country in which
the peasantry constitutes the bulk of the masses and the immediate task is
to fight against imperialism and feudalism, a country with a vast area and
a huge population, where the situation is extremely complicated and the
struggle extremely hard, and he has brilliantly developed the theories of
Lenin and Stalin on the colonial and semi-colonial question as well as
Stalin’s theory concerning the Chinese revolution. It is only because the
Party has firmly adhered to the correct Marxist–Leninist line and waged
a victorious struggle against all erroneous ideas opposed to this line that
it has scored great achievements.
(MXJ, 3: 952–953)

It is equally undeniable, however, that Mao became increasingly intolerable of

different voices and open challenges to his authority in any form, which was
true during the Rectification Campaign and particularly so during and after the
CCP political bureau’s meeting in November 1943 when Bo Gu, Zhang Wen-
tian, and Zhou Enlai had to make lengthy confessions on their mistakes on the
Party’s history. From that point on, free, honest, and collegial exchange of
opinions yielded to a mix of respect, fear, and unprincipled flattery among the
high-ranking party elites in their attitudes towards Mao. Mao increasingly
turned his leadership into a form of dictatorship that necessitated the use of
compulsion to exercise his power. But his dictatorship was different from
Chiang Kai-shek’s. Chiang, as shown in the preceding chapter, based his lead-
ership mainly on his control of the military, and he was never able to stifle dif-
ferent voices from his challengers within the GMD and to completely
eliminate the autonomy of regional military forces. In sharp contrast, Mao not
only put the communist forces under his effective and complete control but
further established his leadership on a solid ideological ground and wiped out
all kinds of “sectarianism” or opposition within the Party and among the mili-
tary through the Rectification Campaign. The “unprecedented ideological,
 Total centralism at work 251
political and organizational solidarity and unity” (MXJ, 3: 953) that the CCP
and its military had achieved nationwide by the end of the War of Resistance
prepared them well for the Civil War with the Nationalist forces after 1946.

Manchuria and the Civil War

For decades, the CCP’s military strategies had been defensive by and large
before the outbreak of the Civil War in 1946. This was true during the Jiangxi
Soviet period from 1927 to 1934, when the Red Army and its base areas had to
survive the Nationalists’ blockade, military sieges, and repeated annihilation
campaigns. It was also true during the War of Resistance from 1937 to 1945,
when the communist forces and their base areas expanded quickly but suffered
the Japanese army’s repeated mopping operations and later the GMD’s blockade
and localized attacks. The biggest disadvantage to the communist forces in
northern Shaanxi and other areas of North China was that these areas were
among the poorest parts of China; they had huge difficulties obtaining enough
food and other goods to support the quickly expanding forces.3 Predictably, as
the War of Resistance came close to an end and the tension between the com-
munist and Nationalist forces escalated, the CCP leaders’ strategy for the expan-
sion of their base areas was to prioritize their development in the relatively
prosperous southern provinces; Mao Zedong and the Party central issued direct-
ives one after another to dispatch the communist forces down to the south and
create or enlarge base areas in Hunan, Guangdong, Fujian, Zhejiang, and
Jiangsu provinces in late 1944 and early 1945 (TDGG 1981, 15: 32–36,
145–147, 181–187).
It is also worth stressing that, despite the quick expansion of communist
forces during the War of Resistance, the CCP was inferior to the GMD in
almost all regards during the interim from Japan’s surrender in Septem-
ber 1945 to the outbreak of the Civil War in June 1946. Having just
emerged as a victor in the war against Japan, the Nationalist government and
its leader Chiang Kai-shek enjoyed unprecedented popularity among the
people. Backed with the military and monetary aid from the United States,
the Nationalist troops far surpassed the communist forces in equipment and
size. Mao, therefore, succumbed to pressure from Chiang, Joseph Stalin, as
well as the general public in China for a negotiation over the formation of
a coalition government in August 1945, despite his unwillingness and confi-
dence in the viability of the communist forces. It was widely observed that,
even if the negotiation went well, the reorganized government would remain
under the GMD’s control and the best result for the CCP was to have a few
seats, together with other political parties, at the central level and retain part
of the communist troops that would be significantly smaller than the GMD’s.
Indeed, when the negotiation was done and an agreement was signed
between the two parties in January 1946, the CCP leadership was serious in
thinking about ceasing military struggle, embracing “the new phase of peace
and democratic reconstruction” (heping minzhu jianshe de xin jieduan)
252 The making of a unified and centralized state
(ZGZY, 16: 62), relocating its headquarters from Yan’an to Qingjiang in
northern Jiangsu, and continuing its rivalry with the GMD by “using ballots
instead of bullets” (Hu Sheng 2001: 13). Later, when the military confronta-
tion between the communist and Nationalist forces in North China and Man-
churia escalated and a civil war became imminent, observers such as Liang
Shuming, general secretary of the China Democratic League, all excluded the
possibility of the GMD’s defeat by the Communists, given its superior mili-
tary force and control of the national government; the only possible result
that the CCP could achieve at most, he believed, was the splitting of China
into two halves, with the CCP controlling the northern part and the GMD
the southern part (Liang Shuming 2006: 318).

Why was Manchuria so important?

But the unexpected events in August 1945 – the atomic bombing of Hiroshima
on August 6, the Soviet Union’s declaration of war against Japan on August 8,
and its subsequent invasion and complete occupation of Manchuria by Septem-
ber 2 – caused a fundamental change to the CCP’s strategy for post-war develop-
ment. The strategic advantages of Manchuria and its extreme importance were
obvious to the CCP. First, Manchuria neighbored the Soviet Union to its north,
Mongolia to its west, and North Korea to its east – all of these neighbors were
communist states or areas and friendly to the CCP; once it occupied this area,
the communist forces would have a secure and stable base area, which would be
relatively isolated from the interior provinces and immune to the Nationalist
threat, and, from there, it could take an offensive strategy by launching large-scale
operations against the Nationalist forces in the interior provinces. Second, unlike
the CCP’s preexisting base areas, which were small and scattered, Manchuria was
vast, covering about 1.3 million square kilometers, making it impossible for the
Guomindang forces to occupy the entire area right after the Japanese surrender
while allowing the CCP enough room to establish its own base areas, to retreat
when faced with the Nationalist troops’ attack in competition for control of this
area, and to display a high level of mobility in planning large-scale offensives to
eventually expel and annihilate the Nationalist forces from the area. Third and
most important, Manchuria was rich. Agriculturally, the immense and fertile soil
in this area accounted for its high level of productivity, which, coupled with the
relatively low density of population, yielded a large quantity of surplus in food
supply compared to other parts of China, making it a net exporter of grain. Fur-
thermore, Manchuria had a well-developed industry, especially the heavy industry
of machinery and energy production, which accounted for about 90% of the total
heavy industrial output in the country in the late 1940s; its arsenals were among
the best and the largest in China.4 Manchuria also had a highly developed trans-
portation network; its railroads reached 14,000 kilometers in total, or about half
of the total mileage of railroads in the country (Zhu Jianhua 1987a: 140). Once
occupied, Manchuria would be a solid logistic base for the communist forces to
conquer the rest of the country.
 Total centralism at work 253
The strategic importance of the Northeast was also clear to the CCP. If they
targeted East China first, where the political and economic centers of Nationalist
China were located, the Communists had little chance of winning because the
Guomindang government, backed by the generous support of the United
States, had their best equipped forces stationed in this area, which could easily
besiege and annihilate the communist forces that were poorly equipped. Indeed,
as it turned out, with about a third of its forces concentrating in Jiangsu and
Shandong, the Nationalists had no difficulty wiping out the preexisting com-
munist base areas in central and northern Jiangsu in the second half of 1947
and pushing the communist forces back to southern Shandong and later further
back to central Shandong by May 1947 (Zhang Xianwen et al. 2005, 4: 80–83,
103–104). Alternatively, if the communist forces prioritized North China as its
target, they would have faced the Nationalists’ attacks from both the Northeast
and East China. The best option, therefore, was to control Manchuria first,
taking advantage of the Soviet occupation of this area, “seal in” the Nationalist
forces in it which had recently entered the area, and completely annihilate them.
Only after its complete control of Manchuria, could the communist forces then
be concentrated in East China to combat the Nationalists’ main forces, taking
advantage of the abundant supply of military and logistic resources from Man-
churia (Ye Jianying 1982).
Anticipating substantial support from the neighboring communist states, espe-
cially the Soviet Union, which continued to occupy Manchuria until April 1946,
Mao and the CCP leadership soon gave up the original strategy of southward
expansion and instead formulated the new strategy of “develop in the north,
defend in the south” (xiang bei fazhan, xiang nan fangyu) in September 1945
(LSQ, 1: 371–372). Mao justified this strategy by saying: “The four northeast-
ern provinces are an extremely important area, having lots of industrial equip-
ment, large factories, and large cities; if they are put under our leadership, we
will have a basis to win.” He continued,

From our Party’s point of view, and in the light of the nearest future of the
Chinese revolution, the Northeast is particularly important. As long as we
have the Northeast, the Chinese revolution will have a solid basis, even if
we have lost all other bases areas now. Of course, the basis of the Chinese
revolution will be even more solid if we have the Northeast while keeping
other base areas.
(MWJ, 3: 410–411, 426)

Liu Shaoqi, who was the Party’s acting chairman of the CCP central committee
when Mao was in Chongqing and next only to Mao in the CCP’s new leader-
ship after the Seventh Congress, first suggested the idea of “develop in the
north” and held a similar view:

As long as we are able to control the Northeast as well as the two provinces
of Chahar and Rehe, and as long as we have the collaboration from all of
254 The making of a unified and centralized state
the liberated areas in the country and from all of the people in the struggle,
the victory of the Chinese people can be guaranteed.
(LSQ, 1: 372)

The Soviet support

The Soviet occupation force’s collaboration was key to the CCP’s establishment
of base areas and successful control of the whole of Manchuria, despite the
former’s commitment to the terms of the treaty between the Soviet Union and
Nationalist China, which forced the CCP to give up its original plan of “exclu-
sively occupying the Northeast” (du zhan dongbei) and switch to the new plan
of “giving up the main route and occupying the two sides” (rangkai dalu,
zhanling liangxiang) and establishing base areas in the countryside (TDGG, 15:
433–436; Jin Congji 2006: 14–15). The Soviet troops welcomed the arrival of
a CCP force at Shanhaiguan and allowed the latter to take over local govern-
ment authority in early September 1945. Later, they allowed all of the commun-
ist troops in different parts of Manchuria to act freely as long as they did not
bear the official title of the CCP’s force, a situation that was true especially in
the first two months of their entry into Manchuria (Li Yunchang 1988). Their
generous provision of weapons made it possible for the preexisting communist
army in Manchuria, known as “the Allied Resistance Force” (kanglian), to form
a “self-defense force” of 48,500 men within a month. The Russian occupying
authority also handed over the arsenal of the former Japanese Kwantung Army,
which was not too far from Shenyang, to a communist force under the com-
mand of Zeng Kelin, making it possible for Zeng to expand his force from
4,000 to 60,000 men. In early October, it further informed the CCP authority
in Manchuria that it was ready to hand over all of the military equipment of the
former Kwantung Army in Manchuria that could arm several hundreds of thou-
sands of soldiers. So huge was the quantity, however, that the communist
troops in actuality were only able to take over about 10,000 rifles, 3,400
machine guns, 100 canons, and 20 million bullets initially. In late October, the
Russians handed over to the communist forces all of the arsenals and ammuni-
tion depots, as well as some heavy weapons and even airplanes, in southern
Manchuria. Before their withdrawal from Manchuria in April 1946, the Soviet
troops further turned over to the communist forces the Japanese arsenals in
northern Manchuria, including, among others, more than 10,000 machine guns
and 100 canons. Altogether, according to an unverified source, the Communists
received from the Russians Japanese weapons that amounted to about 700,000
rifles, 13,000 machine guns, 4,000 canons, 600 tanks, 2,000 vehicles, 679
ammunition depots, 800 airplanes, and some gunboats (Yang Kuisong 1999:
262; see also Liu Tong 2000 for a different estimate). The communist force
thus established its definite superiority over the Nationalist troops in Manchuria
in terms of armament as well as manpower. In early 1946, to ensure that the
communist force had enough time to prepare for its quick occupation of the
smaller cities and rural areas in Manchuria after the withdrawal of the Soviet
 Total centralism at work 255
occupying forces, the latter used various excuses to deliberately delay their with-
drawal and prevent the Nationalist troops from landing in Dalian and taking
over the cities as scheduled (Du Yuming 1985: 519–520, 536–545).

Manchuria as a powerhouse
Manchuria, therefore, became the theater where the first of the three decisive
military campaigns by the communist forces took place (known as the Liaoxi-
Shenyang Campaign), in which they completely defeated the Nationalists and
occupied the region by early November 1948 after more than seven weeks of
battles. As Mao Zedong anticipated, Manchuria became the CCP’s largest and
most important base area. Bolstered by its highly developed modern industry
and transportation as well as a productive agriculture, this region soon func-
tioned as a powerhouse to provide the CCP with manpower, weaponry, and
logistic support that enabled the Communists’ success in the next two major
campaigns against the Nationalists, namely the Huai-Hai Campaign in East
China (from November 6, 1948 to January 10, 1949) and the Beijing-Tianjin
Campaign in North China (from November 29, 1948 to January 31, 1949),
respectively.
Manchuria was first of all the CCP’s most important source of soldiers during
the Civil War. As a result of its vigorous efforts of recruitment and ample supply
of weapons, the communist troops in Manchuria expanded quickly, from about
200,000 soldiers at the end of 1945 to 380,000 soldiers a year later, which
almost doubled to more than 700,000 by the end of 1947 (Zhu Jianhua
1987b: 602, 604); combined, the communist forces in the northwestern, north-
ern, eastern, and north-central provinces increased only by 300,000 by late
1947. By August 1948, on the eve of its major offensive against the Nationalist
troops, the CCP forces in Manchuria further increased to 1.03 million, far out-
numbering its enemy of about 500,000 men (Wang Miaosheng 1997: 94).
They were not only the largest among the communist regional forces, accounting
for nearly 37% of the entire communist army, but also the best equipped.
Altogether, from 1945 to July 1948, the CCP recruited 1.2 million soldiers from
Manchuria, accounting for more than 60% of the increased soldiers of the entire
communist force during the same period (Zhu Jianhua 1987a: 286). After
winning the Liaoxi-Shenyang Campaign in Manchuria, the communist army of
the Northeast sent more than 800,000 of its troops, in addition to 150,000
peasants as logistic workers, down to the interior provinces, who constituted
the main force to defeat the Guomindang troops in the Beijing-Tianjin Campaign
(Zhu Jianhua 1987b: 69).
Equally important was Manchuria’s role in weaponry production and its sup-
port for the Civil War in the interior provinces. Unlike the small-scale guerrilla
warfare the CCP troops fought with the Nationalist or Japanese forces before
1945, in which they had no or little use of heavy weapons, the three major cam-
paigns during the Civil War took the form of large-scale mobile and positional
warfare, each involving the deployment of millions of soldiers, intensive use of
256 The making of a unified and centralized state
artilleries, and the consumption of ammunition in large quantities. The Soviet
assistance in handing over the arsenals of the former Japanese Kwantung Army
to them only partially satisfied their needs in Manchuria. Taking advantage of
the preexisting equipment and Japanese technical personnel who remained in
service, the CCP forces quickly recovered and expanded weaponry production
after they entered and occupied Manchuria. Thus, by the summer of 1948, for
instance, they had owned 55 military factories of different sizes, able to annually
produce about 17 million bullets, 1.5 million grenades, 500,000 shells, and
2,000 60-millimeter cannons (Huang Yao et al. 1993: 436). In 1949, their pro-
duction capacity further increased to 2.3 million shells and 21.7 million bullets
a year, in addition to different kinds of artillery, employing more than 43,000
workers (Zhu Jianhua 1987b: 70). The ammunition made in Manchuria was
indispensable to the communist forces to defeat the Nationalists in the major
campaigns in the interior provinces.
Finally, the Northeast contributed to the Civil War in the interior provinces
by providing the CCP forces with large quantities of logistical goods. The prod-
uctivity of agriculture in Manchuria, which ranged between 900 and 1,000 kilo-
grams of grain per shang (roughly a hectare) or 12–13 million tons of grain
annually in total in 1948 and 1949 (Zhu Jianhua 1987b: 141–143), allowed
the communist authorities in this area to collect an agricultural tax totaling 2.3
to 2.4 million tons a year (at the rate of 21% in 1947 and about 18% in 1948
and 1949) (ibid.: 446). During the entire Civil War period from 1946 to 1949,
the total amount of agricultural taxes (gongliang) collected from Manchurian
farmers reached 6.86 million tons of grain; in addition, they procured from the
farmers 1.8 million tons of grain and 7,488 tons of cotton as well as other kinds
of agricultural products (DBCJ: 210). By exporting these products in large
quantities to the Soviet Union, the Communists purchased from the latter the
industrial, medical, and military goods they needed. The fiscal revenues from
agricultural taxes and other revenues allowed the Northeastern Communist
authorities to spend an equivalent of 3.8 million tons of grain on the military in
1949, 45% of which was on their troops operating in the interior provinces. In
addition, the Northeastern authorities provided the interior provinces with more
than 3 million tons of goods, including 800,000 tons of grain, 200,000 tons of
iron and steel, and 1.5 million cubic meters of lumber (Zhu Jianhua 1987a:
384, 1987b: 71).

Fueling the war: the making of a communist fiscal regime

From decentralization to centralization: pre-1945 experiences

It is apt at this juncture to examine how the communist forces generated rev-
enues to fuel their military operations and expansions. By and large, the com-
munist forces experienced three cycles of expansion and contraction from the
late 1920s to 1949, each beginning with a decentralized approach of revenue
generation and evolving toward the establishment of a centralized fiscal regime.
 Total centralism at work 257
The first cycle took place in the Jiangxi Soviet period, when the newly estab-
lished Red Army supported itself initially by taking over the possessions of the
Nationalist troops or bandits they defeated, and, more importantly, by “attack-
ing local bullies” (da tuhao), confiscating the land and housing of landlords and
powerful gentry elites (dizhu haoshen), directly taking away their movable prop-
erties such as jewelry, furniture, animals, and tools, or forcing them to pay
a hefty fine or amount of cash to redeem the properties sealed by the Red
Army; these methods were supplemented by demanding rich peasants and busi-
ness owners to make “donations” in different amounts commensurate with their
economic status instead of confiscating their properties (Xu Yi 1982: 414–417,
458–466).5
Later, the Red Army gradually switched to a more regulated and centralized
method of revenue generation, as the “local bullies” available for confiscation
vanished and as its base areas expanded.6 The CCP authorities collected the
land tax from all of the landowners on a progressive basis, with rich peasants
shouldering the highest rate of tax payment. They also collected business
income taxes from shop owners and transit taxes on goods in transportation
through the 24 customs points within the Central Base Area. Indicative of this
transition in revenue generation, the Red Army was officially exempted from the
task of “raising funds” after July 1932 (Xu Yi 1982: 434–438, 467–486); it was
up to the Financial Department of the Central Government in the base area to
provide support to the Red Army.
Despite the stable increase in its revenues along with the phenomenal expan-
sion of base areas in Jiangxi and the neighboring provinces, the Red Army’s
fiscal situation was very different from the GMD’s during its early days in
Guangdong. Basing itself on the commercial center in South China, the Nation-
alist government in Guangzhou was able to turn itself into a true fiscal-military
machine that generated enough revenue to sustain the rapidly growing army
and fuel the Northern Expedition through the various modern means of tax-
ation and financing under the leadership of Minister Song Ziwen (T. V. Soong),
as was shown in Chapter 7. In sharp contrast, commerce and modern industry
were poorly developed in the mountainous areas controlled by the Communists;
farmland in this area was also much less fertile than the lower Pearl River and
lower Yangzi deltas where the Guomindang government was located before and
after 1927, respectively. Without enough revenue from the land or commercial
taxes to sustain the Red Army that had grown to more than 100,000 soldiers by
late 1933 in the Central Base Area (in a single year from September 1933 to
September 1934 alone, more than 100,000 new soldiers were recruited into the
Red Army, see Yang Jing 2002: 53), the CCP had to turn to compulsory meas-
ures to augment its income, including forcing people to buy government bonds
with grain or cash and later with grain only, sending “shock teams” to different
localities to collect land taxes in kind, demanding that peasant households
“donate” or “lend” grain to the Red Army (Xu Yi 1982: 486–497; ZGZY, 9:
263–268, 10: 82–86, 323–326, 351–354). By early 1934, the communist gov-
ernment in the Central Base Area had encountered a bottleneck in generating
258 The making of a unified and centralized state
their revenue; after nearly three months of collection, they obtained less than
one-tenth of the planned amount of revenue from the land tax by late Jan-
uary 1934, mostly paid in cash rather than in kind; and after a campaign of
more than five months, they had sold less than half of the government bonds.
Food shortage thus became a severe problem challenging the Red Army’s
sustainability, not to mention its expansion (ZGZY, 10: 82). The fiscal and
hence military weakness of the Red Army, more than any other factor,
explained its failure in resisting the GMD’s fifth and last campaign of annihi-
lation that employed 1 million soldiers, 200 airplanes, and various kinds of
artillery in October 1934.
The communist forces’ fiscal situation during the War of Resistance was no
better than before. “Attacking local bullies” was no longer an option under the
United Front with the GMD, in which the CCP had to incorporate the
“enlightened gentry” (i.e., pro-CCP and anti-Japanese landlords and bureau-
cratic elites) into government organizations in its base areas. Instead, the com-
munist authorities relied on confiscating the properties of the pro-Japanese
“collaborators” (hanjian), voluntary donations of cash and grain (you qian chu
qian, you liang chu liang) by wealthy and ordinary households, and allocations
of funds and goods from the Nationalist government during the initial years of
the United Front (ZGZY, 11: 327–330). A general principle guiding the indi-
vidual communist forces’ activities in revenue generation in separate base areas
was “self-raising for self-use” (zichou ziyong) (Lu Shichuan 1987: 235) or
“expend what was raised” (shui chou shui zhi) without a budget (Wei Hongyun
et al. 1984, 4: 53). Later, the CCP forces increasingly turned to the progressive
taxation of landowners, issuance of “government bonds for national salvation”
(jiuguo gongzhai), collection of “contributions for national salvation” (jiuguo
juan) from wealthy households, imposition of military levies on the CCP-controlled
village offices, or the direct “borrowing” of grain and straw from peasant households
where the CCP-controlled offices did not exist (ZGZY, 11: 615, 842). The village
governments under communist control had to supply the communist troops
with “agricultural taxes for national salvation” (jiuguo gongliang) whereon the
latter provided them with the “military grain-coupon” (junyong liangpiao)
issued by the communist base-area authorities (Wei Hongyun et al. 1984, 4:
180–182, 188–189). Whether imposed by the communist or other military
forces, the multiplying military levies as well as the miscellaneous expenses
exhausted the villages of the base areas by the early 1940s.
After the Nationalist government cut off its supplies to the CCP forces and
blockaded the communist base areas because of the mounting tensions between
them, the latter encountered severe shortages of food and military supplies, and
they could only turn to two options for their survival: downsizing the govern-
ment and the military (jing bing jian zheng) (MXJ, 3: 880–883), and engaging
in production of food, clothing, and goods for self-sufficiency, hence the so-
called Great Production Movement (dashengchan yundong). These efforts
helped the CCP forces to make ends meet. But the striking poverty of the rural
population together with the limited ability of the base areas in the production
 Total centralism at work 259
of weapons nevertheless constrained the CCP’s warring capacities. The best
option for them to survive and prosper was to avoid large-scale war with the
Japanese occupation forces and an escalation of their conflicts with the National-
ists and, at the same time, to take advantage of the weak presence of both the
Japanese and the Nationalists in the vast rural areas of North China and pene-
trate the “rear areas” by widely establishing CCP grassroots organizations in vil-
lages, recruiting as many villagers as possible into the Party and its guerrilla
forces, putting them under the Party’s centralized control while leaving them
self-reliant in logistic support. The CCP followed this strategy and thrived. Its
base areas mushroomed in different parts of the country during the eight years
of the War of Resistance, covering approximately 880,000 square kilometers in
total and a population of more than 100 million by 1945; its membership sky-
rocketed from about 40,000 in 1937 to 1.2 million in 1945 (MXJ, 3: 1108);
and its armed forces similarly exploded by a factor of 28, from roughly about
45,000 men at the beginning of the war to 1.27 million by the time of the Jap-
anese surrender in 1945 (Pang and Jin 2011, 2: 713–715).

Toward fiscal unification

After the full-scale outbreak of the Civil War in July 1946, the CCP, with only
1.3 million poorly equipped soldiers, faced tough challenges. First, its enemy,
the GMD forces, had a total of 4.3 million men, or 3.3 times the size of the
communist forces (Zhang Xianwen et al. 2005, 4: 199). After taking over the
military equipment from the Japanese troops of more than 1 million and receiv-
ing military supplies from the United States, the GMD troops were far better
armed, of which 22 divisions were fully or partially equipped with American
weapons. The resources available to the two sides to sustain the war were also
disproportional: the Nationalist government controlled three-fourths of the ter-
ritory of the country and two-thirds of China’s population as well as the major
industrial cities and transportation networks. It also received generous economic
and military aid from the United States; in the first half of 1946 alone, it
obtained $1.33 billion worth of American goods in addition to a large quantity
of weaponry sold, leased, or donated (ibid.: 66).
Second, the CCP had huge difficulty in sustaining its own rapidly expanding
forces, which reached 2 million by late 1947, 3 million by late 1948, and
4 million by mid-1949. The cost of supporting the solders was huge. At the
North China Financial and Economic Conference attended by delegates from
the CCP’s different base areas in April 1947, it was estimated that feeding one
soldier required the equivalent of 16 dan (1 dan equals roughly 2.75 bushels)
of millet annually (including 1.5 to 2 catties of grain daily, 1 to 2 catties of
meat monthly, and so on), which equaled the total amount of agricultural taxes
paid by 60 to 80 peasants, given that each peasant could produce an average of
2.5 dan or 400 catties of millet per year and 15 to 20% of which could be paid
as agricultural taxes. In other words, the total size of the communist forces
should be limited to 1 to 1.5% of the population under their control; the
260 The making of a unified and centralized state
130 million people in the “liberated areas” by that point could only sustain an
army of up to 2 million soldiers (Feng and Li 2006; FDS, 1990: 589).
Waging a war was also expensive. During the Civil War, each battle often
involved the deployment of hundreds of thousands of soldiers and the consump-
tion of thousands to tens of thousands of shells and millions of bullets, and each
shell alone cost a middle peasant his income for one or two years (Li Weiguang
2000: 248–249; Feng and Li, 2006: 193). The input of manpower in logistic
service was also enormous; typically, one peasant worker was needed for one sol-
dier during the war and for three soldiers during peace time (HBJ: 296). Liu
Bocheng (1892–1986), a key commander of the communist forces in North
China, thus said: “providing for a large-size corps is no different from providing
for a mobile city” (cf. Li Weiguang 2000: 249).
To support the millions of soldiers and to afford the huge expenses of the
war, the CCP leadership came up with three solutions: unifying the resources
from all liberated areas through the establishment of a centralized fiscal regime;
launching the land reform to enlarge the base of taxpayers; establishing its
mobilizing mechanism at the village level.
As in the first two cycles, the communist forces used a decentralized method
of logistic support at the beginning of the Civil War, especially when they
entered a newly liberated area or fought outside the base area. In the Northeast,
for example, “decentralized self-sufficiency” (fensan ziji) was the principle guid-
ing their logistic support in 1946. The properties confiscated from the Japanese
and the puppet Manchukuo accounted for 36.7% of the revenue of the CCP
authorities in this region for the entire year (Zhu Jianhua 1987b: 427). When-
ever they fell short of food and other kinds of supplies when arriving in a newly
occupied area, the CCP troops invariably “borrowed” from landlords and rich
peasants, and to a lesser degree, also from middle-earning and even poor peas-
ants (Feng and Li 2006: 197). As one of the ten military principles formulated
by Mao Zedong in December 1947, the communist forces had to “supply
themselves with the entirety of the weapons and most of the personnel captured
from the enemy. The sources of our troops’ manpower and supplies lie primarily
at the front” (MXJ, 4: 1248). As late as October 1948, Mao was still ordering
that “when the troops entered the GMD-controlled areas and thus fought with-
out support from the rear or with only partial support, all military needs have to
be solved completely or in most part on their own” (MXJ, 4: 1348).
Coordination among different base areas began with agreements between
them being made at the North China Financial and Economic Conference in
April 1947, by which the trade barriers between the different areas were
removed, and mutual aid became possible. The government of the Shanxi-
Hebei-Shandong-Henan Border Area, for instance, was committed to providing
for the East China Field Army’s operations in the northern provinces in Septem-
ber 1947, while the CCP’s East China bureau was responsible for providing the
Border Area with material compensation (Feng and Li 2006: 176). A more
important step towards fiscal unification took place in July 1948, when the CCP
Central Committee established a Financial and Economic Department, with
 Total centralism at work 261
Dong Biwu as its head. Under its leadership, the North China Financial and
Economic Council was created to unify the currencies and policies on trade and
finance in the three liberated areas of North China, the Northwest, and Shan-
dong. Beginning in January 1949, the three areas merged their banks and trade
organizations, used the standard new currency called renminbi, and unified their
revenue and budget systems (Feng and Li 2006: 180). Meanwhile, the CCP’s
central military committee established a General Logistics Department in
April 1948 to ensure the establishment of a multi-tiered system of logistics in
different regions and, after the outbreak of the Huai-Hai Campaign, to coordin-
ate the logistics involving the three strategic zones of North China (Huabei),
North-central China (Zhongyuan), and East China (Huadong) and five provinces
(Jiangsu, Anhui, Shandong, Henan, and Hebei) (Feng and Li 2006: 197).
Fiscal consolidation and unification also took place within each of the liber-
ated areas before they joined the countrywide efforts of centralization. The
Northeast was particularly successful in this regard under the stewardship of
Chen Yun (1905–1995), one of the key CCP leaders in this area since late 1945
and head of the Northeast Financial and Economic Council after 1948. Begin-
ning in 1947, the consolidation and unification of the fiscal system in the liber-
ated areas involved the following steps: establishing a budget system that required
monthly updating of the budgets on, and the monthly reporting of, recurring
expenditures; establishing an auditing system with auditors installed in every
public institution for monthly examination; establishing a unified, multi-tiered
treasury system with strict regulations on the allocation of funds; separating the
sources of revenue for the central government (including all of the agricultural
taxes, profits of state-owned enterprises, and most commercial taxes) from the
revenues for provincial governments (income of provincial enterprises and specific
portions of tax revenues); separating the expenditure of the central government
(mostly on the main forces of the CCP) from the expenditure of provincial
authorities (mainly on local security, construction, education, etc.); standardizing
the fiscal system, including its management of personnel and accounting proced-
ures; unifying the collection of taxes on different kinds of commodities and ser-
vices; and more. After establishing its full control of the entire Northeast in
November 1948, to maximize government revenues, the CCP authority further
increased the rates of agricultural taxes, enforced the monopolistic sales of tobacco
and alcohol, issued government bonds, promoted foreign trade, and stabilized
prices (Zhu Jianhua 1987b: 432–438, 487–496).
All these measures were reminiscent of the similar steps taken by Song Ziwen
in Guangdong in the late 1920s. Thanks in part to the relatively high level of
industrialization of its economy and in part to Chen Yun’s exceptional compe-
tence in overseeing its financial affairs, the Northeast emerged as the most solv-
ent and affluent among all of the liberated areas; it was closer than all others to
a modern “fiscal-military state” in terms of the composition of its revenue.
Unlike other liberated areas where the traditional agricultural taxes or gongliang
accounted for 70 to 80% of their revenues (FDS 1990: 585), in the Northeast,
gongliang (totaling 1.34 million tons) only accounted for 37.04% of its annual
262 The making of a unified and centralized state
revenue in 1948, while the profits from foreign trade and textile manufacturing
contributed 23.32% to its revenue in the same year, and commercial taxes con-
tributed another 17.15%. In 1949, the income from gongliang (2.48 million
tons) further declined to 23.32%, whereas industrial profits accounted for the
largest portion (30.41%) of its revenue and commercial taxes occupied third
place (17.33%) (Zhu Jianhua 1987b: 440; for different figures, see FDS 1990:
795). As the economically most advanced and the first liberated area, the North-
east was as important to the CCP for defeating the Nationalists and conquering
the entire country in the late 1940s as Guangdong had been for the Guomin-
dang forces to defeat the warlords and to unify the country in the late 1920s.

Mobilization and penetration: the dual purpose of the land reform

By and large, the villages in North China witnessed a relatively low level of dif-
ferentiation in land ownership and wealth distribution. Random investigations
showed that landlord and rich peasant households, accounting for 10 to 30% of
rural households, possessed roughly 30 to 50% of farmland in the surveyed
localities (Li Lifeng 2008: 69). Poor peasants in the communist base areas in
North China certainly benefited from the CCP’s Double Reductions (jianzu
jianxi) policy (namely, for landlords to reduce rent by 25%, and for moneylend-
ers to reduce interest rates to 15%) and a progressive tax policy (higher rates for
landlords and rich peasants and low rates for the poor) during the War of
Resistance. Many localities saw the increase of middle-earning peasants and the
decrease of landless and poor peasants as well as landlord and rich peasant
households (Luo Pinghan 2005: 49). In some areas, landlords and rich peasants
became rare or nonexistent at all by the mid-1940s (e.g., Hinton 1966: 209;
Crook and Crook 1979: 62; Friedman et al. 1991: 42–43; Zhang Peiguo 2000:
70: 152). So why did the CCP launch the land reform campaign in 1946 and
push for a radical program of land redistribution in 1947?
The land reform began with CCP Central’s announcement of a directive on
May 4, 1946. Known as the “May Fourth Directive,” it declared “land to til-
lers” as its goal and encouraged peasant masses to “obtain the land from the
hands of landlords” in the course of the preexisting struggles against pro-
Japanese “traitors” and for the reduction of rents and interests. At the same
time, the directive warned against attempts to hurt rich peasants or the landed
gentry elites who had been active in the anti-Japanese struggles and cooperative
with the CCP; it also urged local CCP authorities to “give proper consideration
to the livelihood of middle and petty landlords” and distinguish them from the
“traitors, despotic gentry, and native bullies” (Zhang Xianwen 4: 237–238).
The moderate nature of this directive is not difficult to understand. Having not
yet completely broken up with the GMD, having relied on support from the
“enlightened gentry” to maintain its legitimacy in the base areas, and, more
importantly, to win sympathy from the vast array of social groups that adopted
a neutral stance between the CCP and the GMD, the communist leadership had
to adhere to its preexisting United Front policy and avoid isolating itself by
 Total centralism at work 263
launching an indiscriminate program against all of the landed elites. This was
especially important for the CCP headquarters that was still based in the
Shaanxi-Gansu-Ningxia Border Area, where it had developed an amiable rela-
tionship with the “enlightened gentry” and absorbed much of it into its govern-
ment system.
Foreseeing an upcoming full-scale confrontation with the Nationalists, how-
ever, Mao saw the land reform as the most effective means to win support from
the peasants and mobilize them to join the communist forces. Two weeks after
the issuance of the Directive, therefore, he ordered local communist authorities
to shift the focus of the movement from confiscating the properties of the
former Japanese and pro-Japanese puppet authorities and individual traitors to
“turning the large quantity of landlords’ land into the hands of peasants,” in
response to the escalating conflicts between the communist and Nationalist
forces in this area (ZGZY, 16: 164). After the full-scale outbreak of the Civil
War in July 1946, Mao overtly urged regional CCP authorities to implement
the land reform program as the fundamental way to mobilize the peasants, as he
stated in October 1946:

The experience of the past three months has shown that, whereas the Cen-
tral’s Directive of May Fourth was resolutely and quickly implemented and
wherever the land problem is deeply and thoroughly solved, the peasants
would side with our Party and our troops, to oppose the attacks of Chiang
Kai-shek’s troops. The peasants would take a wait-and-see attitude where the
Directive of May Fourth was carried out without resolution or implemented
too late or in a number of phases mechanically and where the land reform
was ignored completely on the excuse of preoccupation with the war.
(Zhu Jianhua 1987b: 152)

As a result, the CCP abandoned the moderate land reform policy and called
for a radical campaign for land redistribution in July 1947 when the communist
forces fought hard with the overwhelming Guomindang forces on all of its
fronts in the Northeast as well as the interior provinces. It continued the policy
of confiscating landlords’ properties until the summer of 1949, when the com-
munist troops grew to 1.95 million, thus surpassing the Nationalist force in
size, and started its strategic offensive against the latter afterwards. As the pres-
sure for the CCP to mobilize peasants into the army was greatly mitigated, the
Party reintroduced the moderate policy of Double Reductions in all of the
“newly liberated areas” (xin jiefangqu) that came under the CCP’s control after
July 1947. This reversal, as Mao honestly explained in May 1948, was to ensure
that “the wealth of the society will not be dissolved, social order will be relatively
stable, so that all of the forces can be concentrated to annihilate the Guomindang
reactionaries” (MXJ, 4: 1326). The same policy continued into the next year
when Mao emphasized the importance of “maintaining the current level of agri-
cultural production and preventing its decline” as the goal of the CCP’s agrarian
policy in South China (MXJ, 4: 1429).
264 The making of a unified and centralized state
Penetrating the village
Mobilizing the peasantry for the CCP’s war efforts, however, was not the only
motivation behind its implementation of land reform. Even more important for
the CCP was to use land reform as a means for its penetration into the village.
Before the arrival of the communist forces, peasant communities in the base
areas, as in the rest of rural China, had been subject to the dominance of lineage
organizations or other forms of native networks for the purposes of collective
defense, flood control, popular festivals, and religious activities, in which the
rural elites played a leading role, be they lineage heads, gentry members, retired
bureaucrats, landlords, or simply the ordinary villagers with exceptional abilities
and good reputation. When the CCP forces established their base areas in differ-
ent parts of rural China during the War of Resistance, many of the village elites
in those areas continued to occupy the positions of village or township govern-
ment after showing their willingness to cooperate with the Party for the war
effort. After the Japanese surrender in 1945, however, with many of them being
landlords or rich peasants, these village or township leaders soon became the
barrier to the Communists’ efforts to mobilize the villagers and expand their
base areas in preparation for the war with the Nationalists. To effectively carry
out land reform in the villages, therefore, the CCP directly appealed to the poor
in rural societies by sending “work teams” to every village for propaganda and
mobilization activities. With the “poor peasant group” (pinnongtuan) as its
core, the newly established “peasant association” (nonghui, later renamed “peas-
ant council” or nongmin daibiaohui) consisted of poor peasants and hired labor-
ers as well as work team members from outside the village. Under the slogan of
“all power to the peasant association” (yiqie quanli gui nonghui), it replaced the
preexisting CCP branch secretary and village leaders to function as the govern-
ing body in every village. It was from the leaders of the peasant association that
the head and vice head of the new village government, the secretary of the
Party’s branch in the village, and the head of the village’s militia, were selected
(TDGG: 89; Hinton 1966: 306–311; DeMare 2019).
In addition to enlisting young villagers, the most urgent task in the first year
of the Civil War, village authorities in the liberated areas increasingly focused
their energies on two key tasks in the last two years of the Civil War: collecting
taxes and providing logistics support for the communist troops. The effect of
land reform on the Communists’ ability to collect tax was obvious. Before land
reform was introduced, the rates of agricultural taxes for individual households
in the liberated areas were progressive, with landlords and rich peasants paying
40 to 50% of their incomes and the poor paying at a much lower rate. Overall,
each individual in the liberated areas paid 9 to 20% of their agricultural income
as agricultural taxes in 1946 (FDS 1990: 601). After land reform was fully car-
ried out in late 1947, the progressive tax system gave way to the proportional
tax system that required all households to pay their agricultural taxes at a fixed,
universal rate or proportion of their income with or without a deductible (tax-
ation in the “newly liberated areas” remained progressive, where landlords and
 Total centralism at work 265
rich peasants were subject to the Double Reduction policy; these people con-
tinued to exist until the full completion of land reform in early 1953). The over-
all tax burden of the rural population escalated as well, from 16% of their annual
income in 1946 to 20% in 1947 in Shandong; from 20 to 24% in the Northeast;
and from 9 to 27% in the Shan-Gan-Ning Border Area (ibid.: 601–602). In
another two areas, the Shanxi-Charhar-Hebei Border Area and the Shanxi-
Hebei-Shandong-Henan Border Area, the annual tax burden increased by 28%
and 38% in 1947, respectively, and by a further 24% and 9% in 1948 (ibid.:
639). These increases reflected two basic facts: the first was the widened base of
taxpayers from primarily landlords and rich peasants before land reform to all of
the rural households subject to a universal tax rate that increased year after year;
the second and more important was the enhanced ability of the CCP authorities
in controlling the rural population and extracting agricultural resources after
establishing its cells at the bottom of society.
The communist penetration of village communities also enabled the Party’s
mobilization of the peasants for wartime logistics support. It was estimated, for
example, that 3.13 million peasants from the Northeast were mobilized to serve
the communist troops with more than 300,000 stretchers, the same amount of
carts, and more than 900,000 draft animals during the three years of the Civil
War in Manchuria (FDS 1990: 800–801). In the Hebei-Shandong-Henan Area,
more than 2.9 million peasants, 100,000 stretchers, and 397,000 carts were
mobilized for war service during the same period (ibid.: 651). In the Huai-Hai
Campaign, the largest of the three campaigns, more than 5.4 million peasant
workers were recruited, who rescued more than 110,000 wounded soldiers from
the front. During the later stage of the campaign, for every one soldier, nine
peasant workers were organized to provide logistics support (Feng and Li 2006:
194). On average, each adult farmer in these areas spent about nine to ten
workdays a year on logistics tasks; in the war zone or areas along the transportation
lines, the enlisted villagers could spend as many as 20 to 28 days on logistics tasks
(FDS 1990: 603, 801).

A comparison

Geopolitics and military strategy

Between 1927 and 1949, the geopolitical centers of the communist move-
ment shifted from south-central China to the northwest, and further from the
northwest to the northeast, which resulted in due changes in its military strat-
egies and the outcomes of its competition with the Nationalists. Throughout
the first two phases of its existence in south-central and northwestern China,
the communist force was agrarian in nature, not only because its soldiers
were almost entirely recruited from the peasantry, but its supplies also came
chiefly from the rural base areas, which in turn explained mass mobilization as
the primary method for military building, the inferiority of its armaments,
and the use of a defensive strategy in dealing with its enemies. It was only
266 The making of a unified and centralized state
after shifting its geopolitical center to the northeast after 1946, where the per
capita level of agricultural productivity was higher than the rest of the country
and where modern industry and transportation were more developed than any
other regions, that the communist force grew quickly in size and quality; for
the first time, it competed with the Nationalists on an equal footing, changed
its strategy to the offensive one, and eventually defeated the Nationalist forces in
large-scale, decisive military campaigns. Needless to say, the Soviet occupation of
Manchuria at the end of World War II and its assistance given to the commun-
ist forces in the course of its withdrawal from this area were key to the latter’s
success in controlling the northeastern provinces and the fundamental reversal in
its military strategy and relationship with the Nationalist forces.
Chiang Kai-shek, to be fair, well realized the strategic importance of the
Northeast for his government and military; he was willing to renounce China’s
sovereignty over Outer Mongolia in exchange for the Soviets’ commitment to
timely withdrawal from the Northeast and handing it over to his forces. But the
economic and political centers of his government were located in the lower
Yangzi region, which meant that he had to prioritize the defense of the areas
surrounding the region, namely the provinces in East, North, and Central
China; the Northeast was less important in his military strategizing, not only
because it was geographically distant from the economic and political centers
and shipping enough soldiers to the area to formally take the northeastern prov-
inces over from the Russians was a huge challenge to his army, but the com-
munist force’s rapid entry into the area despite the Sino-Soviet agreement was
also beyond Chiang’s expectation. After the outbreak of the Civil War, Chiang
launched a “full-scale attack” against the communist forces on all fronts in
northern Jiangsu and the provinces in North and Northeastern China, hoping
to defeat the enemy in “eight to ten months” (Zhang Xianwen et al. 2005, 4:
223). Short of a clear and well-planned strategy, however, Chiang as well as his
high-ranking commanders had no idea whether they should prioritize their mili-
tary operations in Manchuria or the interior and whether they should shift the
strategic emphasis from the South to the North or otherwise (Wang Chaoguang
2005: 100). In sharp contrast, the Communists’ strategies were clearly and wisely
defined: they prioritized the control of Manchuria; once that goal was achieved,
they quickly shifted their strategy to emphasize annihilating the Nationalist forces
in East China before moving back to the north to wipe out the remnants of the
enemy in the Beijing-Tianjin area and the rest of the northern and northwestern
provinces. Chiang, therefore, was unprepared for the Communists’ campaign
in East China, and his maneuvering of Nationalist forces in response to the
campaign was largely a disaster.
Added to Chiang’s military failure was the attitude of the United States
toward the Civil War in China. Just like the Soviet Union, whose occupation of
Manchuria and assistance to the Communists were key to the military success of
the latter in this area, the United States provided substantial support to the
Nationalists. In the first half of 1946 alone, the United States provided the
Nationalist government with various goods worth $1.33 billion and equipped
 Total centralism at work 267
22 divisions of Nationalist troops with American weapons (Zhang Xianwen
et al. 2005, 4: 66). In April 1948, an act was approved to offer the Nationalist
government another $463 million as economic aid (ibid.: 225). But the American
aid was conditional. The U.S. forces shipped Nationalist troops to Manchuria in
early 1946, for instance, on the condition that Chiang would accept General
George Marshall’s mediation and compromise with the Communists, and threat-
ened to stop the shipment or arms sales if Chiang refused to do so (JJSR: 4/22/
1946, 8/30/1946). Later, as the Nationalists’ failures occurred one after another
in the Civil War and as the incurable corruption in the Nationalist government
disappointed American observers, the U.S. government rejected Nanjing’s
repeated pleas for support and even suspended the 1948 act, causing a dramatic
decline in the goods shipped to China in the first half of 1949 and accelerating
the collapse of the Nationalist regime.

Identity and morale

The contrast was even starker between the Nationalist and communist forces in
the realms of political identity and the morale of the military. As noted in the
preceding chapter, Chiang Kai-shek brought various regional forces under his
government in the late 1920s and early 1930s, but he never succeeded in
extending his real influence into the provinces controlled by the former war-
lords. After 1937, the national crisis caused by the Japanese full-scale invasion
enabled Chiang to mobilize the regional forces and thereby consolidate his lead-
ership throughout the War of Resistance, and his influence reached its peak
when Japan surrendered. But the schism between Chiang and his regional com-
petitors soon surfaced in the Civil War and worsened as Chiang suffered
repeated defeats by the Communists that greatly weakened the forces under his
direct command.7
Equally fatal to the GMD was the low morale of its troops, which was espe-
cially true after the Nationalist forces lost the wars in Northeast and East China
in late 1948 and the beginning of 1949. In January 1949, the CCP leaders,
having won the three major campaigns, anticipated fighting several major wars
before occupying the major cities such as Nanjing, Wuhan, and Xi’an, and
planned to expand their control to only nine provinces in Central, East, and
Northwest China by the winter of the same year. To their surprise, the com-
munist forces captured all of these cities without much effort and sweepingly
occupied almost all of the provinces in the mainland (except for Tibet) by the
end of the year. The Nationalists failed to organize effective resistance through-
out this time. Defections occurred everywhere. Throughout the Civil War,
nearly 500 organized defections took place among the Nationalist troops,
involving 153 complete divisions, more than 1,000 high-ranking commanders,
and 1.77 million soldiers (Cai and Sun 1987, 1: 1). Changchun (one of the
major cities in Manchuria) and Beiping (Beijing) thus fell into the hands of the
communist forces without a major battle. When combating the communist
forces or withdrawing from the battlegrounds, the different divisions of the
268 The making of a unified and centralized state
Nationalist forces were concerned only with their own survival and made little
effort to coordinate with each other. So frustrated was Chiang by the poor per-
formance of his subordinates at the Battle of Menglianggu in May 1947, for
instance, that he complained that “the high-ranking commanders have become
warlords, corrupt, degenerating, concerned only with preserving their own
strength, and unable to assist each other in time of either emergency or peace”
(Wang Chaoguang 2005: 103). Reviewing the Nationalists’ failure in the Civil
War, Chiang Kai-shek admitted in late 1949: “We failed this time not because
we were defeated by the Communist bandits but actually because we were
defeated by ourselves” (JGSX, 23: 133). His diary on the last day of 1949 well
summarized the reasons why the Nationalists had failed:

Only two-tenths of the troops were wiped out because of combating; two-
tenths surrendered out of speculations; five-tenths perished because of flee-
ing and escaping the war; only one-tenth was transported to Taiwan and
other islands for preservation and consolidation.
(JJSR: 12/31/1949)

In contrast with the absence of integration among the regional forces under
the Guomindang regime and the low morale of its military at all levels, the
strength of the CCP lay precisely in its all-round centralization, or “total cen-
tralism” as I would term it, evidenced in: Mao’s establishment of his supreme
authority within the Party through organizational cleansing and ideological
rebuilding; the centralization and unification of Party organizations, military
units, and administrative systems at all levels; the formation of an integrated
fiscal system for coordination and control of all resources from different base
areas; and the functioning of an effective system of logistic support based on the
massive mobilization of the rural population.
More specifically, the CCP outperformed the GMD first of all in the excep-
tional degree of political unity it achieved in the early 1940s, a result of Mao
Zedong’s efforts to enhance his leadership of the Party by building his supreme
authority in the ideological field, completely discrediting the Party leaders
before him, and eliminating “sectarianism” or the various factions within the
Party that did not obey him. All this made it possible for Mao to introduce the
practice of the Party’s “unified leadership” (yiyuanhua lingdao) of the military
in September 1942, which aimed to put the Party organizations and military
units that were scattered across different base areas throughout the country
under the centralized control of the top leadership through its central and
regional bureaus (ZGZY, 13: 426–436). In June 1945, the Party decided to
reestablish its commissions at every level of the military. After the outbreak of
the Civil War, all of the communist forces were grouped into a number of “field
armies” under the direct command of the CCP’s Central Military Commission;
when in operation, each field army was required to create a “front commission”
of the Party, and a “general front commission” of the Party had to be created
when two or more armies joined the same operation. Both types of commissions
 Total centralism at work 269
were subject to the directives of the Party’s Central Committee (Zhang Yutao
et al. 1991: 4).
Beginning in January 1948, the Party further enforced the practice of “report-
ing and requesting instruction” (baogao he qingshi) to curb the tendencies of
“lacking discipline and being anarchic” (wu zuzhi wu zhengfu) that were aggra-
vated during war. It was required that the secretary of each regional bureau of
the Party had to report every two months to the Party’s Central Committee and
Chairman on the military and all other affairs of the region and that the chief
commander of each field army, too, had to report to the Party’s Military Com-
mission every month on combat results, casualties, and the strength of the army
and every two months on the overall state of the army (MXJ, 4: 1264–1266,
1332–1333). All these measures, while ensuring the Party leadership’s command
of its field armies and coordination of their operations, did not preclude the ini-
tiatives from the commanders of the field armies. In fact, Mao repeatedly
encouraged the commanders to make expedient decisions on their own in times
of emergency without having to seek instruction from above on every issue.
Not only did the commanders of the field armies show unanimous obedience
to the Party Central’s leadership, the morale of the soldiers of the communist
troops was also high. The CCP was faced with two challenges in boosting the
soldiers’ morale at the beginning of the Civil War. The first came from the fact
that a large proportion of the soldiers had joined the army during the War of
Resistance out of patriotism. Thus, when the war was over and when the peace
talks between the CCP and the GMD continued for months following the Jap-
anese surrender, the mentality of retiring from the military and going home
for peace-time life grew widely among them. Not surprisingly, when the Civil
War broke out, many were reluctant to fight the battle seriously, adhering to
the notion of “Chinese do not fight Chinese,” and fleeing was not rare on the
battleground. The other challenge was that a growing number of the soldiers
were converted from the captured Nationalist troops, who continued their
habits of loose discipline, beating, and robbery after joining the communist
forces.8
To correct these problems, the CCP employed two measures. The first was
widely conducting meetings of “speaking bitterness” (suku) among the soldiers
to cultivate their political loyalty from the winter of 1947 to the summer of
1948. Speakers at these meetings were carefully selected to complain about the
unusual hardship of their lives before joining the army and the harshness and
brutality of evil landlords and local bullies they had endured. These personal
witnesses were often supplemented with stage performances of revolutionary
dramas on the same theme. So moving were the speeches and performances
that, according to the accounts of official sources, the gatherings often culmin-
ated in loud and unrestrained weeping by the entire audience of soldiers, who
swore to fight the war hard to end the “old society” and rescue the people from
the “sea of bitterness” (Zhang Yong 2010: 76). These occasions of “speaking
bitterness” were believed to be most effective in growing the soldiers’ identity
with the CCP and arousing their hatred toward the Nationalists. Another
270 The making of a unified and centralized state
measure was the so-called “three investigations” (sancha), namely investigating
the soldiers’ class status, activity, and morale. It served several purposes: to test
the soldiers’ loyalty to the Party by looking at their attitudes toward the land
reform and the landlord class; to prevent defection and sabotage among the sol-
diers by searching for enemy agents, especially the former officers of the Nation-
alist troops who had disguised themselves as ordinary prisoners of war and
joined the PLA; and to curb the various forms of corruption among military
officers at different levels and promote equality between commanders and sol-
diers. All these worked to keep the soldiers in discipline and maintain high
morale (Wang Chaoguang 2005: 104).

Fiscal regime
How the Nationalists and the communists fed the personnel of their respective
party-state apparatuses and financed the military was as important as geopolitics
and political identity in shaping the outcome of the Civil War. The Guomin-
dang state had two major financial burdens: to maintain the enlarged rank of
civil servants and to meet the rapidly escalating demands for military expenses.
Supporting the ever-increasing number of civil servants was a huge burden for
the Nationalist state during the few years after 1945.9 Administrative expenses
increased from 18.1% of the Nationalist government’s total expenditure in 1944
(He Jiawei 2009: 147), to 28.5% in 1946 and 29.7% in 1947 (Zhang and Yang
1986: 127), much of which were spent on salary payments. The largest part of
Guomindang state’s spending, to be sure, was on the military, which accounted
for nearly 60% of its total expenditure in 1946, 55% in 1947, and 69% in the
first seven months of 1948 (Yang Yinfu 1985: 173). Both of these burdens
drove the Nationalist government’s spending to rise quickly in relation to its
revenue during the Civil War years, causing widening deficits, which accounted
for 47% of its expenditure in 1945, 62% in 1946, and 68% in 1947 (Zhang and
Yang 1986: 244). Printing more paper currency became the most convenient
means for the government to offset the deficits, which in turn triggered price
inflation; prices increased 1.57 million times from January 1946 to July 1948,
and another 112,490 times from August 1948 to April 1949 (Zhang and Yang
1986: 243).
The social cost of this financing approach was huge. Before the War of Resist-
ance, for instance, university faculty, as a prestigious social group, received
a salary (200–600 yuan per month) that was typically more than twenty times
the wage of an ordinary worker (10–20 yuan per month), which allowed them
to live a very comfortable life. As a result of price inflation, by the end of the
war, their living standard had steadily deteriorated to less than one-tenth of the
prewar level. After the outbreak of the Civil War, the runaway inflation quickly
impoverished school teachers and public servants as well as all those living on
salaries to the subsistence level. A large number of them ran into debt and suf-
fered hunger; some committed suicide (Ci Hongfei 1994; Wang Yinhuan
2005). Corruption was rampant among the public servants as a necessary choice
 Total centralism at work 271
to secure their survival during times of scarcity; military commanders cut down
and pocketed their soldiers’ pay and provisions, which dampened the morale of
their troops. All these aggravated the anti-government feeling among the gen-
eral public. Chiang admitted in November 1948 in his diary that “the general
circle of the intellectuals, especially the leftist professors and press editorials,
slandered and defamed the government in every possible way. The wavering and
hostility of the public opinion has never been as terrible as it is today.” He was
vexed that such “slanders and rumors” had “unexpectedly penetrated deep into
the cadres of the Party, the government, and the military” and that “this type of
venom was more damaging than any kinds of weapons” (JJSR: 11/5/1948).
The communist force’s fiscal constitution differed from the Nationalist
regime’s in several ways. First of all, its rank of cadres was relatively small,
having only about 200,000 to 300,000 individuals on the eve of the Civil War,
or one-fifth to one-fourth the number of civil servants under the Nationalist
government.10 More importantly, all of the cadres were not salaried; they were
maintained under the so-called “supply system” (gongji zhi), which provided
them and their family members with different quantities and kinds of free food,
housing, and other necessities of life according to their ranks, thus freeing them
of the threat to subsistence caused by price inflation as seen in the Guomin-
dang-controlled areas. The supply system did not cause a financial burden to the
central authority of the communist forces because it was decentralized; each
base area under the CCP’s control was responsible for its own provisions to its
cadres, and the standards of provisions varied over time, depending on their
availability, which tended to be more egalitarian in times of scarcity and differ-
entiated when the supplies were relatively ample. This decentralized supply
system, egalitarian in nature most of the time, was effective in ensuring the sub-
sistence of the communist forces, keeping the cadres’ morale high, and minimiz-
ing the problem of corruption among them. Zhu De thus commented on the
role of this system at the end of 1948 when the Communist Revolution was
bound to succeed:

We survived under the condition of the supply system; no pay was needed
for fighting a battle, for cooking, or for anything. The revolution succeeded
depending on this system; the construction of the new state in the future
will also depend on this system.
(Yang Kuisong 2007: 116)

The CCP also financed its war in a different way. Unlike the Nationalist army
that relied entirely on the central government’s budget, which was severely com-
promised by price inflation and military officials’ corruption, the communist
forces used an approach that combined their regional armies’ decentralized
financing efforts with the central authority’s unified coordination. Decentraliza-
tion was the primary method used by the regional forces to fight a war outside
their base area, to build a new base area, or to maintain themselves where no
funding was available from the CCP central authority. Under these circumstances,
272 The making of a unified and centralized state
the troops supported themselves and renewed their combating capacities by pri-
marily capturing weapons and provisions from the defeated enemies, confiscating
the properties of wealthy households, or levying ordinary residents. As the base
area stabilized and expanded, and especially as the communist forces launched
a large-scale offensive that deployed armies in hundreds of thousands which
affected several provinces, centralized planning became essential in supplying the
combating forces with massive amounts of manpower, weapons, and provisions.
To meet this goal, the communist forces again employed two approaches. One
was the traditional method of bottom-up mobilization of manpower and pro-
visions supplied by the millions of peasants that minimized the CCP central
authority’s spending, as seen in their enlistment into the army, participating in
logistics service, and providing the army with the required amount of grain
and supplies. The other was the CCP authorities’ newly gained ability of cen-
tralized planning in managing production in industrial enterprises, using finan-
cial leverages, and procuring resources from the market after their base was
shifted from the countryside to the cities. Su Yu (1907–1984), a key com-
mander of the CCP troops in the Huai-Hai Campaign, the largest of the three
major campaigns in the Civil War, thus said: “The liberation of East China,
especially the victory of the Huai-Hai Campaign, could not have been con-
ducted without the small pushing carts of the peasant logistic workers from
East China and the large-size shells manufactured in Dalian” (LSJZ: 587).
To conclude, the CCP outcompeted the GMD because of the confluence of
its breakthroughs in geopolitical strategizing, its fiscal-military constitution, and
identity building. Geopolitically, the CCP found itself in an optimal situation
after World War II, when Japan, once the biggest barrier to China’s state-
building, was finally removed, while Russia stepped in as a key player in favor of
the CCP. The Soviet support turned out to be critical to the Communists,
making possible their swift infiltration and eventual occupation of Manchuria,
the strategically most decisive area in their competition with the Nationalists.
Occupying Manchuria permitted a fundamental transformation of the CCP’s
fiscal constitution, from being based on the isolated, poverty-stricken rural areas
in the northwestern provinces and other parts of China to fully utilizing the
resources of Manchuria, one of the most productive regions both agriculturally and
industrially in the nation. Equally critical to the CCP was its mobilizing ability that
it had developed through its incessant penetration into rural communities, from
which it generated the manpower and logistic support on a massive scale for its
military offensives. What characterized the communist forces, therefore, was
a peculiar yet supplementary combination of its newly obtained strength through
centralized control of urban manufacturing and financial resources on the one hand
and its traditional ability in the decentralized mobilization of rural manpower and
labor service on the other, which resulted in an equilibrium in its fiscal constitution
that was self-sustainable and highly expandable.
Most importantly, the communist forces achieved a high level of political
identity by overcoming factionalism within the top leadership and centrifugal
tendencies among its regional forces, and by establishing Mao’s political and
 Total centralism at work 273
ideological supremacy in the early 1940s. These developments, together with
the CCP’s obvious disadvantages in the size and equipment of its military at the
beginning of the Civil War and therefore the unprecedented survival crisis,
caused the communist forces in different areas to minimize their differences,
submit to Mao’s leadership, and coordinate with one another to fight the
Nationalists. In the meantime, the CCP’s stress on ideological propaganda and
tight control of the military at the bottom level, coupled with the material
incentives generated by the land reform, kept the morale of its solders at a high
level. The CCP’s organization and entire military force, from commanders to
foot soldiers, thus developed a strong motivation for victory; all these contrasted
sharply with the lack of conformity and confidence among the Guomindang’s
military commanders and of the discipline among the soldiers.
What accounts for the CCP’s success in state-rebuilding, in a nutshell, is
a metamorphosis it underwent before and during the Civil War in the 1940s,
in which the Party turned itself from an agrarian regime struggling for survival
into a formidable fiscal-military apparatus drawing on both rural and urban
resources, and from an organization ridden with factionalism into one of high-
level centralism.

Notes
1 Not surprisingly, studies on the Communist Revolution have concentrated on these
two phases, especially the second. Depending on the sources and localities they inves-
tigated, these studies offer different interpretations of the dynamics of the Revolution.
The fact that the communist forces boomed during the Japanese occupation in
North China caused Chalmers Johnson (1962), drawing on sources from Japanese
agencies on the CCP’s activities behind the lines of the Japanese forces, to emphasize
the rise of mass nationalism among the peasantry in response to the brutality of Jap-
anese operations, which paved the way for the Communists to step in and win over
the awakening peasants. Likewise, Ralph Thaxton (1983, 1997) underscores the cen-
tral role of disgruntled peasants against extortionate warlord forces and landlords or
against the Nationalist state whose monopolistic policies threatened their livelihood.
Here again the CCP only capitalized on the preexisting situation of popular grievance
to advance its cause by appealing to the moral norms or subsistence needs of the
rural population; its role in peasant mobilization was indirect. Contrary to these inter-
pretations of the Communist Revolution as a reactive movement, others emphasized
the CCP’s initiatives and organizing abilities in the making of the Revolution. Mark
Selden (1971) thus attributes the Party’s success to its flexibility and audacious
experiments in social, economic, and political transformations in the most destitute
area in northwest China; his analysis centers on mass mobilization, cooperative pro-
duction, self-reliance, and the purity of the Party ranks, known altogether as the
“Yan’an way” (Yan’an daolu). Pauline Keating (1997) discounts the applicability of
the CCP’s flexible and gradualist practices in the area around its headquarters in
Yan’an to other parts of rural China with different ecological and socioeconomic con-
ditions, where she finds a different legacy of the Communist Revolution, namely the
Party’s top-down compulsion and authoritarianism, which persisted in response to
resistance from entrenched local power relations. In the base areas in Jiangsu and
Anhui, likewise, Yung-fa Chen (1986) shows the success of the CCP in developing
a sophisticated set of strategies to win support from upper elites by appealing to
274 The making of a unified and centralized state
nationalism while engaging in moderate reforms and elections to rebuild economic
and political relations at the grassroots level, in the wake of the Japanese occupation
and the Nationalist regime’s withdrawal from these areas. Odoric Wou (1994), too,
underscores the CCP’s active mobilization of peasants in Henan and Anhui through
its implementation of a variety of measures to benefit its supporters. The net effect of
the communist mobilization of peasant masses, as these studies show, was that, when
the Nationalists came back to North and Central China after the Japanese surrender
in 1945, they found a predominance of communist influences in rural areas.
2 The ideological authorities within the Party remained those who had received
“authentic” Marxist training in Moscow, primarily Wang Ming and Zhang Wentian.
Wang had his 1931 essay, “Struggle for the CCP’s Greater Bolshevik-ness,” reprinted
for the third time in Yanan in March 1940 and widely circulated among Party mem-
bers (Pang and Jin 2011, 2: 634). As the Party’s general secretary, Zhang authored
in 1937 a textbook, The History of Revolutionary Movements in Modern China, as the
CCP’s most authoritative account of modern Chinese history (H. Li 2013: 95–102).
Indicative of his ideological authority, Zhang also headed the Marx-Lenin Institute
(later renamed the Central Academy), founded in May 1938, to specifically train the
CCP cadres in Marxist–Leninist theories (Cheng Zhongyuan 2006: 275–285). In
sharp contrast, Mao had little influence in the ideological field. In early 1941, for
instance, he wanted to publish his book, Rural Surveys, authored back in 1930
through 1933, in which he argued for the importance of first-hand investigation, but
he found it difficult to get his book published. In May 1941, he delivered a lecture,
“Reform Our Study,” at the Yanan Cadres’ Meeting, to emphasize the principle of
“seeking truth from facts” (shishiqiushi), but this lecture “had no influence at all,” as
Mao complained later (Pang and Jin 2011, 2: 641). Without enough ideological
creditability, Mao even found it difficult to lecture at the CCP’s Party School (HQM:
287; Pang and Jin 2011, 2: 645).
3 It was precisely the utmost poverty, the sparse population, and the difficulty in
recruiting soldiers in Northern Shaanxi province that drove the Red Army to launch
an “Eastern Expedition” into Shanxi province in January 1936 (Pang and Jin 2011,
1: 383). For the same reason, the Red Army planned a Western Expedition into
Ningxia province in May 1936 to have access to the supplies from the Soviet Union
(ibid.: 383, 389, 402). Later, after the outbreak of the Civil War, the severe shortage
of food and supplies in the CCP-controlled Shaanxi–Gansu–Ningxia border area pre-
vented the communist forces from other regions to be relocated into this area to
fight the Nationalist forces that outnumbered the communist forces in this area by
a factor of eight (250,000 versus less than 30,000), leaving Mao and the CCP’s
headquarters in an extremely precarious situation (Pang and Jin 2011, 2: 803).
4 It was estimated that in 1943 Manchuria produced 88%, 93%, 66%, 78%, and 50% of
pig iron, steel, cement, electric power, and coal in China, respectively (Zhu Jianhua
1987a: 11).
5 These fund-raising activities lacked planning from the top of the CCP. It was up to the
individual Red Army troops to identify the targets for confiscation or the demanding of
a donation, and they expended the revenues thus generated without a budget; in typical
fashion, they “spend what was just obtained” (lai yidian yong yidian) (Wang and Li
1981: 29). Though backed by the communist ideology of class struggle, which was
indeed necessary at the beginning of the Red Army’s existence when they lacked a base
area stable and large enough to sustain itself, these activities failed to generate stable
and adequate revenues and they were essentially no different from the way the trad-
itional bandits and rebels supported themselves; the Red Army thus was vilified as “red
bandits” (chifei) in GMD propaganda, not without reason.
6 By November 1931, when the Provisional Central Government of the Chinese
Soviet Republic was inaugurated at Ruijin of Jiangxi province, for instance, the
 Total centralism at work 275
Central Revolutionary Base Area (the largest among the many communist base
areas in south-central China) had expanded to cover 50,000 square kilometers,
including 21 county seats and a population of 2.5 million (Yang Jing 2002: 49).
7 As before the War of Resistance, the biggest challenge to Chiang’s leadership con-
tinued to be Li Zongren and Bai Chongxi of the Guangxi clique. Thus, while
Chiang appointed Bai as the commander of the Nationalist forces in Central China
in June 1948 to block the southward deployment of the communist forces under
Liu Bocheng and Deng Xiaoping while weakening Bai’s forces at the hands of Liu
and Deng, Bai later rejected Chiang’s orders to dispatch his troops to East China
during the Huaihai Campaign in December 1948, anticipating the elimination of
Chiang’s direct forces by the Communists and the subsequent rise of the Guangxi
clique in place of Chiang. Indeed, as it turned out, under pressure from Li Zong-
ren and Bai Chongxi who, with a force of more than 300,000 soldiers, called for
peace talks with the Communists after the disastrous failure of the Nationalist
forces in the Huaihai Campaign, Chiang stepped down in January 1949, and Li
subsequently assumed the position of Acting President (Taylor 2009: 400).
Chiang, however, never gave up his actual control of the government as well as
the military, and the lack of coordination between his own forces and the troops
commanded by Bai Chongxi explained in large part the failure of the Nationalists’
defense along the Yangzi River and the Communists’ rapid occupation of South
China.
8 During the three years of the Civil War, 2.8 million prisoners of war from the
Nationalist troops joined the communist forces, accounting for 65 to 70% of the
People’s Liberation Army (PLA) in 1949 (Zhang Yong 2010: 72).
9 No reliable data on the size of civil servants of the Republican state before 1945
exist. A 1931 survey shows that the personnel of the central government and its affili-
ated institutions in Nanjing amounted to 46,266 individuals. It was estimated that
the number of government employees at the provincial level could be more than ten
times the size, or roughly about 500,000 individuals (Li Lifeng 2004: 103). If we
take 300 as the average number of civil servants in each county, the 1,800-odd coun-
ties in the entire country (Tan Qixiang 1991: 70–71) could have 540,000 civil ser-
vants. Altogether, the total number of civil servants under the Nationalist
government could be around 1 million individuals before the War of Resistance.
During the war, the number shrank as a result of the loss of the territory under the
Nationalist government, but it could be at least half a million before the Japanese
surrender. The number expanded quickly after the Nationalist government moved
back to Nanjing and retook the provinces and cities that had been occupied by the
Japanese and pro-Japanese puppet forces. By 1947, the total number of civil servants
at the central, provincial, and municipal level had reached 678,472 (including offi-
cials, clerks, military officers, policemen, and all other employees in the administrative,
judicial, law-enforcement, and defense services) (He Jiawei 2009: 146). If the salaried
government personnel at the county and lower levels were included, the total size
could be as high as 1.2 to 1.3 million. If the rank of civil servants was expanded to
include teachers and clerks in all public schools as well as all of the employees living
on government-paid salaries, its total number was well above 11 million (He and
Luo 2011: 42).
10 This estimate is based on the assumption that the cadres accounted for one-fourth or
less of the members of the CCP. In 1945, the Party’s membership was 1.21 million,
which increased to nearly 4.49 million in 1949, while the number of the communist
cadres also reached nearly 1 million.
11 Conclusion

China’s transition to a modern sovereign state deviated from the “normal”

path of state-making typically found in the literature on nation-states, which
assumes their emergence as a process directly driven by geopolitical competi-
tion and interstate war but ultimately shaped by the complex interaction of
multiple factors involving commercialization, capital formation, power config-
uration, legal tradition, religious and political heritages, and so on, as seen in
the history of early modern and modern Europe; or as a process of decoloniza-
tion and nationalism, as happened to the non-Western world in the nineteenth
and twentieth centuries. In both cases, the rise of modern sovereign states has
been typically interpreted as the aftermath of the collapse of medieval or
modern empires on the Eurasian continent or the demise of European colonial
empires in the non-Western world. And, ideally, a nation-state, Western or
non-Western, should comprise people of shared ethnic or cultural traditions,
claim its exclusive sovereignty over the land and water within its boundary,
and have a government in the form of representative democracy that manifests
popular sovereignty; all these characteristics of a modern nation-state stand in
sharp contrast with a continental or colonial empire that typically encompassed
peoples of diverse ethnic and religious backgrounds, had unstable or hierarch-
ical territories, and deprived its domestic and/or overseas subjects the right
of popular participation. State-making is thus conceived as a grand transition
from empire to nation-state.1
This transition never happened to China. The making of the modern Chinese
state, as shown throughout the foregoing chapters, involved three key develop-
ments: China’s transformation from a prototypic, monoethnic state to an early
modern territorial state by the 1750s; from a territorial state to a sovereign state
in the late nineteenth and early twentieth centuries; and from a fragmented and
decentralized to a unified and centralized state in the first half of the twentieth
century. This three-step transformation, however, should not be equated with,
or simplified as, a transition from empire to nation-state, not only because
a military or colonial empire never existed in China right before or during the
transformation, but because the end result of this transformation is unlike any
nation-state in today’s world: compared to all other polities, the Chinese state
as it stands today is exceptionally large in physical size and centralized in its
 Conclusion 277
power structure. This peculiar form of state can be properly understood only
when we take all of its three historical developments into consideration. While
the first two account for the vastness of China’s territory and the diversity of
its population today, the third explains the exceptional cohesiveness and
strength of the Chinese state that remains largely unchanged after nearly seven
decades since its inauguration in the mid-twentieth century.

Making sense of the Chinese leviathan

Why it is so “big”
China’s physical immensity and diversity can be attributed to the combined
effects of the following heritages. The first is territorial expansion and adminis-
trative integration under the Qing. The Qing differed from the preceding Ming
dynasty chiefly in that, in addition to the interior provinces taken over from the
latter, it added Manchuria, Mongolia, Xinjiang, and Tibet as its frontiers; China,
therefore, transitioned from a monoethnic state of the Han Chinese into
a multiethnic state consisting of the provinces of the Han people and the fron-
tiers of the non-Han populations. It is important to note, however, that territor-
ial expansion under the Qing was fundamentally different from empire-building
in other parts of the premodern or early modern world. Unlike the military
empires in world history that constantly engaged in war for territorial expansion,
interstate war and expansion were the exception rather than the norm in the his-
tory of the Qing from 1644 to 1911. After relocating its capital city to Beijing
and conquering the former Ming provinces, the Qing showed no intention of
expanding its Inner Asian frontiers any further for about half a century. The
Manchu rulers carried out a series of military expeditions from the 1690s to the
1750s primarily because of the recurrent and fatal threats from the Zunghar
Mongols. Its incorporation of the land once occupied by the Zunghars into its
territory during the expeditions was primarily for the purpose of defending its
security rather than for fiscal or religious reasons as typically seen in the histories
of empires. Furthermore, unlike the typical course of empire-building in world
history that sees recurrent expansion or contraction of territories, with the fron-
tier serving only as a springboard leading to further conquest, the Qing saw no
reason for further expansion after eliminating the Zunghars in the 1750s,
although its fiscal and military strength had reached a peak by that point and
could easily sustain further expeditions. Territorial expansion was the exception
rather than the rule in the history of the Qing dynasty in China. Content with
its status as the legitimate ruling dynasty in China and the huge quantity of rev-
enue from the 18 provinces, its overall strategy after 1644 was defensive and
conservative, aimed at preserving its preexisting territory, rather than endless
conquest and expansion, which made no sense to the Qing rulers.
Second and more importantly, the Qing succeeded by and large in keeping its
land in its entirety until the very end, which further made it different from
a conquest dynasty that saw frequent adding or losing of territories. Behind the
278 The making of a unified and centralized state
striking stability and continuity of its territorial and demographic basis was the
Qing court’s sophisticated and effective methods to govern the Han and non-
Han populations. As a dynasty of alien origin, the Qing governed the 18 inter-
ior provinces of the Han population by embracing the Confucian principle of
“benevolent government” with greater sincerity than its predecessor, in order to
establish and maintain its status as the rightful dynasty to rule China. Its unfail-
ing adherence to the policy of light taxation yet tight political and intellectual
control of the Han elites resulted in the impressive stability of Han society. The
huge amount of revenue that the Qing court obtained from the interior prov-
inces fueled its military operations to safeguard its security, and the geopolitical
security it established by building frontiers in turn allowed the Qing court to
keep its taxation of the Han people at a low level. Meanwhile, the Qing also
demonstrated its striking ability to govern and preserve its frontiers. The close
ties between the imperial court and frontier elites through religious patronage
and/or intermarriage, the migration of the Han people to the frontiers, the
growing economic ties between frontier and interior areas, the administrative
integration of the frontiers into interior provinces, and finally a growing aware-
ness among the Han and non-Han elites of a shared statehood at times of crises
in the late Qing period – all these factors worked together against the centrifu-
gal inclinations of the frontiers and accounted for China’s territorial continuity
after the fall of the Qing.
Therefore, rather than an empire with frontiers constantly expanding or
shrinking as seen in the history of other parts of the Eurasian continent, the
Qing that ruled China is best termed an early-modern territorial state. It was
“early modern” because it had fixed and stable-territory; it maintained a large
standing army without engaging in frequent interstate war for territorial expan-
sion; and it had a highly centralized and bureaucratized administrative system.
All these made Qing China closer to a modern sovereign state than any of its
counterparts outside Europe; and it was much better prepared than most of the
non-European states for integration into the global system of sovereign states in
the modern era.
Third, and no less important, China’s huge territory and diverse population
also resulted from its fruitful efforts to defend against colonialism in its integra-
tion into the Europe-centered system of states after the 1840s, which in turn
can be attributed to the following three factors:

(1) the working of the mechanism of “regionalized centralism” that enabled the
late Qing state to mobilize fiscal and military resources for modernizing
defense, education, transportation, and manufacturing and later reforming
the entire state apparatus, a development that prevented China from being
conquered and divided up by foreign powers;
(2) the growth of nationalism among the Chinese political and intellectual elites
in the early twentieth century that prevented warlordism in the interior prov-
inces from developing into perpetual fragmentation and splitting of China
into separate states;
 Conclusion 279
(3) the powerful legacies of frontier building under the Qing that remained
functional after the fall of the dynasty to prevent Manchuria, Inner Mongo-
lia, Xinjiang, and Tibet from seeking formal independence even when China was
plagued with political division and instability under the Republican government.

All these factors contributed to the continuities in the Qing transition from an
early modern territorial state to a modern sovereign state. Therefore, the overall
structure of China’s territoriality remained largely unchanged from the peak of
Qing expansion in the 1750s to the founding of the People’s Republic and its
consolidation in the 1950s; today’s China is still a land consisting of the interior
provinces of the Han people and the frontiers of Mongols, Uighurs, Chinese
Muslims, and Tibetans.
Thus, rather than a deviation from the “normal” path to the nation-state,
China’s regeneration as a modern sovereign state in the mid-twentieth century
was the result of a three-century process of state transformation; what happened
to China was not a transition from empire to nation-state, but the building of
a territorial state by the mid-eighteenth century and its arduous yet successful
integration into the system of sovereign states after the mid-nineteenth century.

Why it is so “strong”
State transformation in China since the mid-seventeenth century not only resulted
in an enlarged territory and a diverse population that showed striking continuity in
spite of recurrent regime changes, but also culminated in the making of a highly
centralized, authoritarian party-state, rather than a democracy, by the mid-
twentieth century. Once again China’s experiences deviated from the “normal”
course of state-building implied by the paradigm of empire to nation-state that
links the modern nation-state with Western-style democracy and further equates
the latter with the only legitimate form of government manifesting the ideal of
popular sovereignty.
China actually experimented with democracy twice in the Republican era, first
in 1912–1913 under the government in Beijing, and again in 1946 under the
government in Nanjing. Both experiments, however, were short-lived because of
intensified struggles between competing political forces. The failure of parliamen-
tary democracy in 1913 ushered in president Yuan Shikai’s dictatorship and the
rampancy of warlordism after his death; and the outbreak of a civil war between
the GMD and the CCP in 1946 nullified the reintroduced democracy from its
very beginning. The most fundamental reason behind the chronic political dis-
order and subsequent failure of democracies in Republican China, however, lay in
institutionalized decentralization of power within the government system, the
growth of autonomous provincial or regional forces, and their competition with
the central government for control of military, fiscal, and administrative resources,
as shown throughout the chapters on the late Qing and Republican periods.
The rise of a highly centralized “strong” state in twentieth-century China, for
its part, was the result of a dialectic process of interaction between the central
280 The making of a unified and centralized state
and regional forces in seeking solutions to the problems of the regionalized dis-
tribution of fiscal and military resources. The rise of “regionalized centralism” in
the wake of the Taiping Rebellion proved to be an effective and indeed the only
viable solution for the Qing state to end domestic disorder and engage in mod-
ernizing projects in the late nineteenth and early twentieth centuries. However,
under the condition of a widening rift between the Manchu and Han elites and
the vanishing loyalty of the latter toward the royal court in the last few years of
the Qing, the regionalization of state power turned out to be fatal to the dyn-
asty, as clearly seen on the eve and during the Revolution of 1911. After the fall
of the Qing, especially following the death of President Yuan Shikai, political
fragmentation under warlordism worsened; ambitious provincial leaders all
engaged in building a highly centralized regional regime to mobilize local fiscal,
military, and administrative resources. The predominance of “centralized region-
alism,” therefore, aggravated political division and military competition among
the regional contenders, yet at the same time it also prepared the basis for the
founding of a centralized and unified state at the national level. The reason why
the GMD emerged triumphant in the fierce competition among the regional
forces is precisely because it had built a fiscal-military organization that was
more unified and centralized than all other competitors. Nevertheless, after
establishing a national regime in 1927, the GMD was far from successful in
incorporating the preexisting regional forces into the national military and gov-
ernment systems and thereby establishing a truly centralized and unified state
apparatus, not to mention its failed attempt to penetrate urban and rural society
all the way to the bottom level so as to build an effective infrastructure for rev-
enue extraction and social control. Compared to the GMD’s “semi-centralism,”
the CCP proved to be strikingly successful in forging an apparatus that was
highly unified in party organization and ideology and that characteristically com-
bined a highly centralized fiscal-military system with a decentralized mode of
resource mobilization. It was this higher level of centralization or “total central-
ism,” coupled with a reversal in its geopolitical setting after World War II, that
enabled the CCP to outcompete the GMD in 1949.
The birth of a strong state in twentieth-century China, in the final analysis,
had its roots in the geopolitical setting and fiscal constitution that prevailed in
the eighteenth century. It was Qing China’s supremacy in relations with neigh-
boring states and its superiority in military strength that accounted for the
dynasty’s loss of incentive to keep upgrading its military and the prevalence of
the low-level equilibrium in the dynasty’s fiscal constitution. The stagnation and
decline in the size and quality of its regular forces (the Eight Banners and the
Green Standards), coupled with a rapidly growing population, the outflow of
silver from the domestic market, and hence the deterioration of the fiscal equi-
librium, made the Qing poorly prepared in the 1850s when confronted with
large-scale domestic disorder; the imperial court had no choice but to count on
local initiatives when its regular forces failed to suppress the rebels and when its
regular revenues were inadequate to meet the mounting spending on domestic
and foreign crises. Had China been surrounded by competitors of equal
 Conclusion 281
strength in the eighteenth century, the Qing state would have likely engaged in
a military revolution, and its revenue structure would have accordingly changed,
shifting from relying mainly on the land tax to taxing goods and borrowing, to
satisfy its growing military and security needs, as happened to contemporary
European states.
Developments in state rebuilding during the Republican period were deeply
rooted in the heritage of the late Qing years. Regionalized centralism in
the second half of the nineteenth and the first decade of the twentieth centuries
foreshadowed the rise of centralized regionalism, or the building of centralized
and unified fiscal-military regimes at the regional level, in the early Republican
years. The GMD state, originally one such regional competitor, outcompeted all
other regional contenders because it built the most centralized and unified
fiscal-military apparatus. The CCP further defeated the GMD in 1949 because it
built a fiscal-military force that was even more unified and centralized than its
enemy. The sequence of the key links in state rebuilding in China over the three
centuries from the 1640s to the 1940s thus is clear: high Qing supremacy in geo-
political relations established by the 1750s led to low-level equilibrium in fiscal
constitution and military stagnation in the following century, which in turn
caused the regionalization of military and fiscal resources after the 1850s and the
rise of regional fiscal-military states in the late 1910s and the 1920s, and all these
paved the way for the rise of the GMD and later the CCP states. For any regional
contender committed to national dominance, the only way to survive the fierce
competition and realize the objective of establishing a countrywide regime is to
build a fiscal-military apparatus that was more centralized and unified than its pre-
decessors and competitors. State transformation in the first half of twentieth-
century China, therefore, took the form of growing levels of centralization and
unification in fiscal-military institutions and political identity; the end result of this
process was inevitably the birth of a huge, powerful party-state that aimed to
penetrate the entire society and control all resources within its reach for extraction
and self-strengthening.

China as a historically multi-layered state

The rise of the modern Chinese state, then, should be best seen as the result
of the combined effect of different layers of historical heritages in the long
process of state transformation, and we can identify four such layers that
work together to define what China is today. On the top level, China appears
as a party-state, with a highly centralized power structure under the leadership
of the CCP, which resulted immediately from the Communist Revolution
that ended in 1949. On the next level, China exists as a sovereign state,
which claims its exclusive rights on the land and water territories within its
officially defined boundaries and acknowledges its equality with all other
states under international law. This resulted from China’s incorporation into
the world system of sovereign states in the second half of the nineteenth cen-
tury and early twentieth century at the expense of part of its territory and
282 The making of a unified and centralized state
sovereign rights. The GMD’s state-building efforts from the 1920s to the
1940s were most decisive in shaping the territorial range of China’s sover-
eignty. On the third level, China represents itself as a unified multiethnic
state, comprising provinces of a dominantly Han population and “autono-
mous regions” or frontiers of minority populations, an administrative struc-
ture that has its roots in the Qing expansion that ended in the 1750s and
that also owes in large measure to the Qing state’s effective governance of
a diverse population and successful maintenance of its territory throughout
the rest of its history. Finally, on the bottom level lies a “proto-China” that
has grown to cover the 18 provinces of the Qing; this is where the Chinese
people of Han origin live and where the Chinese civilization has flourished,
nourishing a rich culture which forms Chinese identity. In the twentieth cen-
tury, this shared sense of belonging fueled the growth of Chinese nationalism
and propelled the rise of a modern Chinese state more than any other heri-
tages. State transformation in China, in a nutshell, took place as the continu-
ous buildup of different layers of historical heritages from the bottom to the
top levels outlined above, rather than the disruptive transition from empire to
nation-state. The form and content of the Chinese state was renewed and
redefined each time a new layer was added. With its origins in the remote
history of an ancient and undisrupted civilization, the modern Chinese state is
ultimately the cumulative result of a long-term transformation that began in
the 1640s.
China’s experiences in state transformation, then, contradict in many ways the
thesis of empire to nation-state commonly found in the literature of nationalism
and state-making, including some writings on late imperial and modern China.
Instead of the rise of an expansionist empire engaging in incessant conquest,
state transformation in China began with the formation of a territorial state that
sought no expansion throughout most of its history, and its intermittent exped-
itions for a relatively short period, which did indeed lead to territorial expansion
in response to the Zunghar offensives in Mongolia, Xinjiang, and Tibet, took
only about 20 years in total (eight years under Emperor Kangxi: 1690–1697,
1715–1722; eight years under Emperor Yongzheng: 1723–1724, 1729–1734;
and five years under Emperor Qianlong: 1755–1759), or less than one-
thirteenth of the Qing rule in Beijing. Instead of the constant addition or loss
of frontier characteristics of the history of a military empire, the Qing frontiers
showed striking stability once they joined its territory; instead of the breakup of
territory and the establishment of independent states on former frontiers,
dependencies, or colonies as happened to all empires in world history, the tran-
sition from the Qing to the Republic witnessed striking continuity in China’s
territory, which remained largely unchanged in the further transition from the
Republic of China to the People’s Republic of China in 1949. And finally,
instead of creating a Western-style democracy, what prevailed in China by the
mid-twentieth century was a highly centralized and unified state in response to
the prolonged fragmentation and disunity that had hindered China’s growth as
a modern nation-state.
 Conclusion 283
All in all, state transformation in China was a long-term process driven ultim-
ately by the forces internal to the land; it was the geopolitical setting, fiscal
constitution, and political identities unique to China that combined to dictate
the dynamics and define the course of its transition from a proto-, monoethnic
state through a multi-ethnic, territorial state to a centralized and unified modern
sovereign state. External forces were critical in shaping China’s geopolitical rela-
tions and its subsequent strategic priorities, but the country was so immense in
size and deeply entrenched in its own traditions that any attempts to transplant
a foreign model onto the land would have to yield to the dynamic and logic of
state-making intrinsic to the country itself.

State rebuilding in China: an unfinished task

Huge changes have taken place in China since the founding of the PRC, the most
salient being the transformation of the nation from an agrarian society, with most
of its population living on subsistence agriculture, to the largest manufacturer and
contributor to world trade than any other nation by the late 2010s. With its
industrialization accelerating since the 1980s, the country has also experienced the
largest scale of urbanization in world history, turning hundreds of millions of rural
dwellers into urban residents in a matter of a generation. After more than six dec-
ades of vigorous growth, the Chinese economy continues to expand in the 2010s,
and is likely to overtake the United States to become the largest in the world
before long. Despite these developments, however, the formation of the Chinese
state remains largely unchanged. Territorially, other than its recovery of sover-
eignty over Hong Kong in 1997 and Macau in 2000, China remains identical to
what it was in 1951 after establishing its control over Tibet. Even more striking is
the stability and continuity of the state apparatus itself. The government system
that was established in 1949 and its key features, such as the CCP’s leadership,
the composition of, and relationships among, the major branches of the central
government, and the relationship between the state and its people, all remained
largely the same over the next seven decades, which, of course, testified to the
state’s unusual ability to cope with the myriad of challenges from outside and
within while pursuing its strategic goals with impressive records. Yet, the stability
that the Chinese state has shown in the past does not suggest that it has taken
a definite shape or that the task of state-making is done. Far from that, as China is
experiencing economic modernization and social transformation on a scale unseen
in human history and redefining its position in relation to other nations, the state
will continue to evolve in response to the challenges that will confront it in the
decades to come.

Redefining national identity

It was true that in the course of territorial expansion the Qing court made delib-
erate and consistent efforts to build its ties with the ruling elites of the frontiers
through religious patronage, intermarriage, and a degree of autonomy allowed
284 The making of a unified and centralized state
to the latter in internal governance; these measures indeed succeeded in cultivat-
ing the allegiance of frontier elites toward the Qing. And it was true that under
the influence of the printed media, and as a result of modern education they
received in China or abroad, an awareness of shared destiny grew among the
intellectuals of Han and non-Han origins in the last years of the Qing. It was
also true that the last decades of the dynasty witnessed the accelerating integra-
tion of the frontiers with the interior through the massive migration of the Han
people into Manchuria, Inner Mongolia, and Northern and Eastern Xinjiang,
and through the establishment of new provinces in parts of the frontiers. All
these developments were critical to the Qing in maintaining its claims over the
frontiers and preventing the splitting of its territory in its decline in the nine-
teenth century; they also contributed to the rise of a consensus among the
Han and non-Han elites on the idea of a “Republic of Five Races” after the fall
of the Qing. The newly established Republic thus was able to inherit from the
preceding dynasty the entirety of its territory, including all of the frontiers.
Nevertheless, these developments should not be simply equated with signs of
the emergence of a shared nationality among the Han and non-Han peoples.
After all, the so-called Republic of Five Races was only in part a result of the
lingering loyalty of frontier elites toward the ruling house of the Qing and in
part a result of the Han elites’ redefinition of “nationalism” after giving up the
idea of establishing a republic of the Han people only. The formation of
a “Chinese nationality” transcending the ethnic differences between the Han
and non-Han populations remained an ideal limited largely to the intellectual
and political elites of Han origin; it was far from a reality in the absence of
substantial integration among the peoples of different origins in their everyday
economic, social, and cultural life.
After 1949, huge government efforts were made to promote “national unity”
among the 56 officially identified ethnic groups by migrating the Han people to
the frontiers, adopting the Beijing-based dialect of Mandarin as the “universal
language” throughout the country, cultivating the elites of ethnic minorities
through opportunities in education and political careers, improving the living
conditions of the ethnic populations through a series of economic and social
policies, and integrating the frontiers with the interior provinces through the
massive construction of infrastructural projects. National integration certainly
made headway in the Mao era from the 1950s through the 1970s, but much of
it was achieved by the top-down implementation of the state’s policies and
therefore was unsustainable when government intervention yielded to market
forces. Since the 1980s, the introduction of economic reforms and the relax-
ation of some of the policies regulating ethnic relations resulted in the massive
inflow of the Han people into the frontiers for business opportunities as well
as the migration of the minority people into interior provinces; a spontaneous,
bottom-up, and two-way process of integration eventually set in, but the
inequality between the Han and minority populations in education and access
to resources in the market economy also resulted in widespread tensions and
conflicts among them.
 Conclusion 285
Adjusting its policies toward the minorities to help them succeed in realms
where they had been disadvantaged could only provide a temporary solution to
the immediate challenges that confronted the state. A more fundamental solu-
tion to the strained ethnic relations entails the state’s redefinition and rebuilding
of the basis on which a true nationhood shared by both the Han and non-Han
populations can develop. Since the founding of the Republican state in 1912,
the cultivation of national identity has centered on the glorification of the his-
tory and culture of the Han people. The official version of the “national history”
(guoshi) of China has little to do with the past of non-Han peoples; when they
were mentioned, the nomadic states were invariably depicted as backward and
marginal to the civilized dynasties of the Han people; the ethnic and cultural
divides between the Han and the surrounding nomadic peoples, which had been
central to the Han-centric view prevailing in imperial China, remain deeply
ingrained in the minds of the Han Chinese. China remains largely an “ethnic
nation” of the Han in which the historical experiences of the non-Han peoples are
obscured. Therefore, a more challenging task for the Chinese state is to cultivate
the awareness among its entire population of the rich and diverse heritages of all
ethnic groups. The making of a “Chinese nationality” should not depend merely
on the top-down implementation of compulsory measures for cultural, economic,
and political integrations or the imposition of a Han-centered collective memory
on the entire population; equally important is respect for the tradition and rights of
all other ethnic groups and their voluntary identification with the state.

Repositioning China in the world

Aside from rediscovering the history and culture of the non-Han populations,
which are essential to the making of a shared nationhood, reassessing China’s
recent past in a global context is equally important in shaping its new identity.
Since the Nationalist Revolution in the 1920s, the history of China after the
Opium War of 1840 has been reinvented into a series of “national humiliations”
(guochi) inflicted by foreign imperialism. Two basic reasons explain why this ver-
sion of history has prevailed in China. One has to do with the history itself.
Having long been the only dominant power in East Asia, China found it
extremely difficult to give up its self-claimed superiority over all other states and
embrace the new system of states in which it would be treated as only an equal,
if not inferior, partner to all other members of the system. Its integration into
the world system of states, therefore, turned out to be particularly sluggish and
painstaking, involving repeated defeats by foreign powers and its loss of all
tributaries as well as damages to its territorial entity and sovereign rights. Unlike
many other non-Western countries that were quickly subjugated after the arrival
of European powers and soon took on the path of colonization or Westerniza-
tion, it took China more than half a century to adapt to the new order of the
world; the mental suffering inflicted upon the Chinese elites over a long period
was enduring, shaping their subjective perception of China’s past experiences
and relations with the outside world.
286 The making of a unified and centralized state
The other reason has to do with the needs of the Nationalist and later Com-
munist state-makers in the twentieth century, who defined their task as primarily
a Nationalist Revolution against imperialism; central to the making of their
ideologies was the writing of modern Chinese history as a series of conflicts
between China and foreign powers. The repeated indoctrination of the anti-
imperialist ideology and history, compounded with the collective memory of
China’s losses in the past century, resulted in a deep-seated subconsciousness
about China as a victim of foreign imperialism among the elites as well as the
ordinary recipients of Nationalist propaganda. Gone is the old Han-centric
view of a civilized Middle Kingdom surrounded by savages; in its stead is
a new vision of the world, in which China as a victim is juxtaposed against for-
eign powers as victimizers.
This vision of a polarized world, instrumental as it was in formulating the
ideology of the Nationalists and the Communists, worked well for both in their
struggle for the recovery of China’s territorial integrity and sovereignty; it also
worked well when the state was isolated from the system of states dominated by
Western powers, as happened to the PRC throughout most of the Mao era. But
it does not help with China’s reentry into the world system and its resurgence
as a global power. A lesson that China can learn from the Cold War is that
there is no nation in the world which can truly escape the ever-expanding global
capitalist system and the states system that accompanies it. Only after it becomes
a full member of the world capitalist system and a full member of the system of
states can China develop itself by taking full advantage of the opportunities that
can be found from within the system or tackle the challenges that confront it.
What China needs in preparation for its new role in the world, therefore, is to
jettison the victim mentality from among the political elites as well as the popu-
lace, which could easily lead to China’s isolation from, and hostility toward, the
outside world, and rebuild its self-confidence on the basis of an informed and
balanced comprehension of its own past. As shown in the preceding chapters,
China’s experiences in the nineteenth and early twentieth centuries were not
just a record of failures and humiliations; more significant than that record were
a series of breakthroughs that accounted for China’s successes in defending its
territory and independence in the nineteenth century, its quick transition from
fragmentation to centralization and unification in the early twentieth century, and
its emergence as a triumphant, unified nation with its territorial integrity and sov-
ereignty largely restored by the end of World War II; all these developments in
the late Qing and Republican periods provided China with the necessary precon-
ditions for its reemergence as a major power in the global age. Instead of repeat-
ing the old story of expansion and subjugation that had characterized the rise of
empires and major powers in past centuries, China’s engagement with the world
in the twenty-first century should focus on building a geopolitical environment
friendly to its development, beginning with the formation of transnational mech-
anisms for regional integration in manufacturing, trade, and finance, which would
benefit both China as a powerhouse of the world economy and the surrounding
countries as participants in globalization.
 Conclusion 287
Rebuilding political identity
A third and the most challenging task that remains unfinished in China’s quest
for a modern nation-state is to redefine the relationship between the state and
its people. Much of this has to do with pre-1949 legacies in seeking solutions to
the problem of an excessively decentralized power structure in the late Qing
decades, and to the worsened situation in the early Republican years. Despite
his making of a party-state that placed the Guomindang above the civilian and
military institutions and further put the party-state apparatus under his personal
dictatorship, Chiang Kai-shek never succeeded in eliminating the centrifugal
forces at the regional level and establishing the organizations of the party-state
at the local level for effective mobilization of manpower and other resources.
The Communist Party succeeded in rivalry with the Nationalists precisely
because it built a party-state that was more centralized and unified, effectively
penetrating the peasant communities in areas it occupied. However, neither the
GMD nor the CCP allowed room for the protection of individual rights or the
promotion of the awareness of them among the public. Quite the contrary, indi-
viduals under the party-state had to sacrifice personal interest for the cause of
the Party. This Party culture, coupled with the survival of the “feudal remnants”
of neo-Confucian values on loyalty to authorities and the priority of group
objectives over individuals, explained Chiang Kai-shek’s dictatorship and reckless
persecution of liberal intellectuals; it also accounts for the cult of Mao Zedong,
the brutality of the Red Guards during the height of the Cultural Revolution
(1966–1976), and the tragedies that occurred repeatedly afterwards.
The introduction of the mechanisms of market and competition since the 1980s
allowed a degree of freedom for people to make a choice in economic realms, but
the civil rights of individual citizens remain to be protected, despite the recent for-
mation but malfunctioning of a legal framework for that end; the arbitrary exercise
of power by government officials at all levels remains largely unchecked, in the
absence of transparency in policy-making, an independent media, and an informed
public. Individuals and private corporations remain powerless and helpless when
confronted with the abuse of state agents. In sharp contrast to the strong state is
the underdevelopment of a civic culture in society, despite the emergence of
a sizeable middle class in society and the rapid increase of the wealth in its posses-
sion; non-governmental organizations and self-governing bodies remain limited in
number and subject to the state’s surveillance. In the absence of independent
venues to express themselves, it is difficult for citizens to have a free debate, and
build consensus, over the issues that concern them. The state, in other words, is
yet to be based on the solid foundation of a citizenry capable of performing their
rights and duties defined and protected by law. The strength and viability of the
state, in the final analysis, do not come from its monopoly of violence or the
increased revenues that enable its enlargement and strengthening of the apparatus
of coercion, nor do they come from the state’s indoctrination of a dated ideology
that has lost much of its appeal to the people. More fundamental than all these for
the state to rebuild its vitality and legitimacy is the consent of the people, which
288 The making of a unified and centralized state
can only be cultivated through the development of civil organizations for citizens
to articulate themselves and, more importantly, through the opening of the policy-
making process to the supervision and participation of the informed public.

Note
1 Underlying this construct, needless to say, is a Eurocentric perspective that reflects
more or less the historical experiences of Western European states in their early
modern and modern ages, as critics have pointed out (see, e.g., Torgovnick 1990;
Balibar 1991; Said 1979; Hobson 2006, 2012).
List of characters

an liang jintie 按粮津貼

an Meiguo zhi xin 安美國之心
anchashi 按察使

bade 八德
banshi dachen 辦事大臣
bao fan gu yu 保藩固圉
baogao he qingshi 報告和請示
baojia 保甲
baolan 包攬
ben you Yuan zhi chenpu 本有元之臣僕
benchao zhi shengwei 本朝之聲威
biancun 編村
bing qi di buzu yi genzhong, de qi ren buzu yi qushi 併其地不足以耕種，得其人
不足以驅使
boke 伯克
bu ju yu shang, ji san yu xia 不聚於上，即散於下
budeyi, biying juzhao yi 不得已，必應舉著也
buzhengshi 布政使

caizheng gongsuo 財政公所

caizheng ju 財政局
canzan dachen 參贊大臣
caoliang 漕粮
chaojin 朝覲
cheng haoling, bei ceying 承號令，備策應
chi ying bao tai 持盈保泰
chifei 赤匪
chu Jiang 除蔣
chunwangchihan 唇亡齒寒
cunben zhengzhi 村本政治
cungui 村規
290 List of characters
da minzuzhuyi 大民族主義
Da Qing 大清
da tuhao 打土豪
da Yazhouzhuyi 大亞洲主義
da zhi shuishi 大治水師
danghua jiaoyu 黨化教育
dangshu 黨書
dashengchan yundong 大生產運動
de qi tu buzu yi gengyun, de qi min buzu yi qushi 得其土不足以耕耘，得其民不
足以驅使
di 狄
dibao 地保
difang 地方
dizhu haoshen 地主豪紳
Dongshansheng xunyueshi 東三省巡閱使
Dongshihui 董事會
du zhan dongbei 獨佔東北
dui dang nao dulixing wenti 對黨鬧獨立性問題
dutong 都統

ebo 鄂博

fangjing lu cang cong ying 方今帑藏充盈

fangqu zhi 防區制
fazheng xuetang 法政學堂
feitu wuyi yu defang, erbing yihai yu baixing ye 非徒無益於地方，而并貽害於百
姓也
fengpiao 奉票
fu tianxia zhe, tianxiaren zhi tianxia ye, fei nanbei zhongwai suo de er si 夫天下
者，天下人之天下也，非南北中外所得而私
Fuxinghui 復興會

Fuyi quanshu 賦役全書

gaxia 噶廈
geming shang wei chenggong, tongzhi reng xu nuli 革命尚未成功，同志仍需努力
genjudi 根據地
geren ziyou 個人自由
gongji zhi 供給制
guan min heli 官民合力
guan yong shen li, shen ji guan wei 官用紳力，紳藉官威
guanyu 關餘
guanzhi bianzhi ju 官制編制局
guanzhi gaige 官制改革
Guixi 桂系
guofu 國父
 List of characters 291
guojia kufu chongying, wuji shangren juanshu 國家庫府充盈，無藉商人捐輸
guojiazhuyi 國家主義
guojun bianqian 國軍編遣
guomin dahui 國民大會
guomin tongyi 國民統一
guoyong 國用
guoyong yuan you changjing 國用原有常經

haifang 海防
haifang juanshu 海防捐輸
haijun yamen 海軍衙門
hanjian 漢奸
hanxin zhuili zhi zhi 寒心惴栗之至
he wu wei yi 合五為一
he zi waixie, shushi genhua 何自外攜，數世梗化
hengzhengbaolian 橫徵暴斂
Heningwang 和寧王
hepantuochu 和盤托出
heping minzhu jianshe de xin jieduan 和平民主建設的新階段
Hongjun jiangling 紅軍將領
hu fa zhi zhu 護法之主
Hu Xiang dizi man Tianshan 湖湘弟子滿天山
hua Man Han zhenyu 化滿漢畛域
Huabei 華北
Huadong 華東
Huaxinghui 華興會
Huang Chao 皇朝
Huang Qing 皇清
Huaxia 華夏
huiban 會辦
hunchengxie 混成協

Ji-Cha zhengwu weiyuanhui 冀察政務委員會

jiangjun 將軍
Jiangguo fanglue 建國方略,
jianzu jianxi 減租減息
jiezeeryu 竭澤而漁
jing bing jian zheng 精兵簡政
jinglue dachen 經略大臣
Jingxi 晉系
jiu Bohai menhu er lun, yiyou shengubuyao zhi shi 就渤海門戶而論，已有深固不
搖之勢
jiu bai zhi gong 九白之貢
jiudi chou xiang 就地籌餉
jiuguo gongliang 救國公糧
292 List of characters
jiuguo gongzhai 救國公債
jiuguo juan 救國捐
ju 局
juan jie jian xing 捐借兼行
juanli 捐例
juanna 捐納
juanshu 捐輸
jun shi chen yi li, chen shi jun yi zhong 君使臣以禮，臣事君以忠
junfu 軍府
junmin fenzhi 軍民分治
juntong 軍統
junxian 郡縣
junyong liangpiao 軍用糧票
junzheng 軍政

kalun 卡倫
kanglian 抗聯
Kebuduo 科布多
Kebuduo canzan dachen 科布多參贊大臣
kong fei guojia zhi fu 恐非國家之福
kongxu ji yi, luojue ju qiong 空虛已極，羅掘俱窮
Kulun 庫倫

lai yidian yong yidian 來一點用一點

lanyishe 藍衣社
lianbingchu 練兵處
liangtai 糧臺
Liao xiang 遼餉
Lifanyuan 理藩院
ligu 釐谷
lijin 釐金
lin 鄰
lingtu zhi tongyi 領土之統一
liu zheng 六政
Liuda yilai 六大以來
lixian 立憲
liyilianchi 禮義廉恥
lü 閭

Mao Zedong sixiang 毛澤東思想

Meng 盟
Menggu zhi li 蒙古之例
Mengguli 蒙古禮
Minben 民本
Minquan 民權
 List of characters 293
Minsheng 民生
minzhu jizhongzhi 民主集中制
minzu 民族
minzu geming 民族革命
minzu zhi tongyi 民族之統一
mujuan 畝捐

nagong zhi guo 納貢之國

neifu 內附
nianban 年班
Nian xiang 練餉
Nonghui 農會
nongmin daibiaohui 農民代表會

pai Man 排滿
pifu 匹夫
pinnongtuan 貧農團

qi 旗
qinchai dachen 欽差大臣
qingli caizheng 清理財政
qingli caizheng chu 清理財政所
qingxin xiangnei, tongmou wuzu xingfu 傾心嚮內，同謀五族幸福
quanti ziyou 全體自由
quanxuesuo 勸學所
quchu dalu 驅除韃虜
quchu dalu, huifu Zhonghua 驅除韃虜，恢復中華

rangkai dalu, zhanling liangxiang 讓開大路，佔領兩廂

rangwai bi xian annei 攘外必先安內
renai sixiang 仁愛思想
renminbi 人民幣
renzheng 仁政

Shunningwang 順寧王
Shunyiwang 順義王
saifeng 塞防
san bai jiu ke 三拜九磕
san shi 三事
san xiang 三餉
san yu zhengce 三寓政策
san zi zhengce 三自政策
sancha 三查
Sanminzhuyi 三民主義
sanrentuan 三人團
294 List of characters
shanhou da jiekuan 善後大借款
shanghui 商會
shangshang zuiyao zhi yi 上上最要之義
shen Jiang 審蔣
shengyong 省用
shenshang 紳商
shi haifang wei wuzuqingzhong 視海防為無足輕重
shiba guo 十八國
shishiqiushi 實事求是
shouhui liquan 收回利權
shu qiannian lai weiyou zhi bianju 數千年來未有之變局
shu qiannian lai weiyou zhi qiangdi 數千年來未有之強敵
shui chou shui zhi 隨籌隨支
siwei 四維
siyi 四夷
siyuan 司員
Su ren zhi Su 蘇人治蘇
su wei yi jia 素為一家
suku 訴苦
suo 所

tai 臺
tanpai 攤派
Tao Yue fei wen 討粵匪檄
taochi 討赤
tianxia 天下
tianxia zhi cai zhi you ci shu 天下之財止有此數
tianxia guojia zhi zhu 天下國家之主
Tianzi you dao, shou zai si yi 天子有道，守在四夷
tidiao 提調
tong minzu yi zongzu zhi guomin 同民族異種族之國民
tongshui 統稅
tuanlian 團練
tuanlian dachen 團練大臣

wei zhi buzhi wei, hui zhi buzhi huai, di buke geng, min buke chen 威之不知畏，
惠之不知懷，地不可耕，民不可臣
wanghua 王化
weiyuan 委員
wo Kaerka zhubu 我喀爾喀諸部
wu zuzhi wu zhengfu 無組織無政府
Wuliyasutai jiangjun 烏里雅蘇臺將軍
wuquan fenli 五權分立
wuzhuang baowei Sulian 武裝保衛蘇聯
wuzu gonghe 五族共和
 List of characters 295
xia 夏
xiang bei fazhan, xiang nan fangyu 向北發展，向南防禦
xiangbao 鄉保
xiangdi 鄉地
xiangnei 嚮內
xianzheng 憲政
xiao minzuzhuyi 小民族主義
Xibei choubianshi 西北籌邊使
xibeijun 西北軍
xichui yongbing yilai, yiying junxu jie qugi yu gongnu, sihao buyi leimin 西陲用
兵以來，一應均需皆取給於公帑，絲毫不以累民
Xichui xuanhua shi 西陲宣化使
xiexiang 協餉
xin jiefangqu 新解放區
xin minzhuzhuyi 新民主主義
xinfu zhi huan 心腹之患
Xingzhonghui 興中會
xixu jiu zhi 欷歔久之
xunjinshu 巡警署
xunwei zhaoshu 遜位詔書
xunzheng 訓政
xunzheng gangling 訓政綱領
xunzheng shiqi yuefa 訓政時期約法

Yan’an daolu 延安道路

yangwu 洋務
yanyu 鹽餘
yi 夷
yi dang ling jun 以黨領軍
yi dang zhiguo 以黨治國
yi Menggu tonglei 亦蒙古同類
yi shou nan gong 易守難攻
yi wei shi shuai qi tudi renmin yi shou tongyi yu dahuangdi, buzhi qita ye 亦惟是
率其土地人民以受統一於大皇帝，不知其他也
yi wushi wei dashi, wugong wei dagong 以無事為大事，無功為大功
yi ti zhi guo 一體之國
yi zhuyi zhengfu 以主義征服
yici geming lun 一次革命論
yidi 夷狄
yiqie quanli gui nonghui 一切權力歸農會
yiqie zhidu zhangcheng yu neidi shengfen wuyi 一切制度章程與內地省份無異
Yili jiangjun 伊犁將軍
yishihui 議事會
yiyuanhua lingdao 一元化領導
yong bu jia fu 永不加賦
296 List of characters
you qian chu qian, you liang chu liang 有錢出錢，有糧出糧
yongzhong 庸眾
youyi funi 有意附逆
yubei lixian 預備立憲
yu bing yu tuan 寓兵於團
yu jiang yu xue 寓將於學
yu zheng yu mu 寓徵於募
yuqi duo ju zuo chang, wuning shi maoyan buwu ziwei liutong 與其多聚左藏，
無寧使茅檐部屋自為流通
yuqian huiyi 御前會議

zeren neige 責任內閣

zhanshi tongyong piao 戰時通用票
zhasake 札薩克
zhen 鎮
zhen yi guoyong yi zu, bu shi jiazheng 朕意國用已足，不事加徵
zhengzhi weiyuanhui 政治委員會
zhigong zhi guo 職貢之國
zhong jian er fen qi shi 眾建而分其勢
Zhongdong tielu 中東鐵路
Zhongguo 中國
Zhongguo benbu 中國本部
Zhongguo Guomindang 中國國民黨
Zhongguo zhi mingyun 中國之命運
Zhongguo zhi renmin 中國之人民
Zhonghua minzu 中華民族
Zhonghua yu wo yidao tonggui 中華與我一道同軌
zhong liangtai 總糧臺
Zhonghua geming dang 中華革命黨
zhongtong 中統
zhongxiao renai xinyi heping 忠孝仁愛信義和平
zhongyang jiekuan 中央解款
zhongyang zhuankuan 中央專款
zhongyangjun 中央軍
Zhongyuan 中原
zhouye zhi huan 肘腋之患
zhuanxiang jingfei 專項經費
zhuyi jianjun 主義建軍
zhuyi shengguo wuli 主義勝過武力
zichou ziyong 自籌自用
ziji 自給
ziqiang 自強
ziwei 自衛
ziyiju 諮議局
ziyouzhuyi 自由主義
 List of characters 297
zizhengyuan 資政院
zizhi 自治
zongban 總辦
Zongcai 總裁
Zongli 總理
Zongli Geguo Shiwu Yamen 總理各國事務衙門
zouxiao 奏銷
zu min 足民
zuai gemeng 阻礙革命
zuida zhi wenti 最大之問題
zunshou Sanminzhuyi 遵守三民主義
References

Original sources
BYJF. Beiyang junfa 北洋軍閥, 6 vols., edited by Zhongguo shixuehui 中國史學會
and Zhongguo shehui kexueyuan jindaishi yanjiusuo 中國社會科學院近代史研究所.
1990. Wuhan: Wuhan chubanshe.
CBYW. Chouban yiwu shimo 籌辦夷務始末 (Complete records on managing foreign
affairs), 6 vols., edited by Wenqing 文慶, Jia Zhen 賈禎 and Baoyun 寶鋆, in Xuxiu
siku quanshu 續修四庫全書, 414–421. 1995. Shanghai: Shanghai guji chubanshe.
DBCJ. Dongbei jiefangqu caizheng jingji shi ziliao xuanbian 東北解放區財政經濟史
資料選編 (Selected source material on the financial and economic history of the north-
eastern liberated area), vol. 3., edited by Zhu Jianhua 朱建華. 1988. Harbin: Hei-
longjiang renmin chubanshe.
GCGJ. Gongchanguoji, Liangong (bu) yu Zhongguo geming wenxian ziliao xuanji 共
產國際、聯共(布)與中國革命檔案資料叢書 (Series of archival material on the Com-
munist International, the Communist Party of the Soviet Union (Bolsheviks), and the
Chinese revolution), 21 vols., edited by Zhonggong zhongyan dangshi yanjiushi diyi
yanjiu bu 中共中央黨史研究室第一研究部. 1997–2012. Beijing: Beijing tushuguan
chubanshe, Zhongyang wenxian chubanshe.
GMDD. Zhongguo guomindang lici daibiao dahui ji zhongyang quanhui ziliao 中國國
民黨歷次代表大會及中央全會資料 (Source material on the national congresses and
plenary sessions of the Chinese Nationalist Party), 2 vols., edited by Rong Mengyuan
榮孟源. Beijing: Guangming ribao chubanshe.
GMWX. Geming wenxian 革命文獻 (Literature on the Revolution), edited by Zhong-
guo guomindang zhongyang dangshi shiliao bianzuan weiyuanhui 中國國民黨中央
黨史史料編纂委員會. 1958–1989.
GXCD. Guangxu chao donghualu 光緒朝東華錄 (Donghualu, guangxu reign). 1958.
Beijing: Zhonghua shuju.
HBJ. Huabei jiefangqu caizheng jingji shi ziliao xuanbian 華北解放區財政經濟史資
料選編 (Selected source material on the financial and economic history of the North
China liberated area), vol. 1, edited by Huabei jiefangqu caizheng jingji shi ziliao
xuanbian bianjizu 華北解放區財政經濟史資料選編編輯組. 1996. Beijing: Zhong-
guo caizheng jingji chubanshe.
HBS. Zhongguo jindai huobishi ziliao 中國近代貨幣史資料 (Source material on the
history of currencies in modern China), edited by Zhongguo renmin yinhang
canshishi jinrongshiliao zu 中國人民銀行參事室金融史料組. 1974. Taipei: Wenhai
chubanshe.
 References 299
HQM. Hu Qiaomu huiyi Mao Zedong 胡喬木回憶毛澤東 (Hu Qiaomu’s recollection of
Mao Zedong), edited by Hu Qiaomu huiyi Mao Zedong bianxiezu 胡喬木回憶毛澤
東編寫組. 2003. Beijing: Renmin chubanshe.
JGSX. Xian zongtong Jiang gong sixiang yanlun zongji 先總統蔣公思想言論總集, 10
vols., edited by Qin Xiaoyi 秦孝儀. 1984. Taipei: Zhongguo guomindang zhon-
gyang weiyuanhui dangshi weiyuanhui.
JJSR. Inventory of the Chiang Kai-shek Diaries, Hoover Institution Archives. Stanford,
CA: Stanford University Press.
LHZ. Li Hongzhang quanji 李鴻章全集 (Complete works of Li Hongzhang), edited by
Gu Tinglong 顧廷龍 and Dai Yi 戴逸. 2008. Hefei: Anhui jiaoyu chubanshe.
LQC. Liang Qichao quanji 梁啓超全集 (The complete works of Liang Qichao), edited
by Liang Qichao 梁啓超. 1999. Beijing: Beijing chubanshe.
LSJZ. Liaoshen juezhan 遼瀋決戰 (The decisive battle of Liao-Shen), vol. 1, edited by
Zhonggong zhongyang dangshi ziliao zhengji weiyuanhui 中共中央黨史資料徵集
委員會, et al. 1988. Beijing: Renmin chubanshe.
LSQ. Liu Shaoqi xuanji 劉少奇選集 (Selected Works of Liu Shaoqi), 2 vols., edited by
Zhonggong zhongyang wenxian bianji weiyuanhui 中共中央文獻編輯委員會.
1981. Beijing: Renmin chubanshe.
LZR. Li Zongren huiyilu 李宗仁回憶錄 (A memoir of Li Zongren), 2 vols., edited by
Tang Degang 唐德剛. 1995. Shanghai: Huadong shifan daxue chubanshe.
MGDA. Zhonghua minguo shi dang’an ziliao huibian 中華民國史檔案資料彙編
(Compilation of archival material on the history of the Republic of China), edited by
Zhongguo dier lishi dang’an guan 中國第二歷史檔案館, five series. 1979–2000.
Nanjing: Jiangsu renmin chubanshe.
MGSL. Zhonghua minguo zhongyao shiliao chubian: Dui Ri kangzhan shiqi 中華民國
重要史料初編──對日抗戰時期 (A preliminary compilation of the important histor-
ical sources of the Republic of China: The period of War of Resistance against Japan),
7 series, edited by Qin Xiaoyi 秦孝儀. 1981. Taipei: Zhongguo guomindang zhon-
gyang weiyuanhui dangshi weiyuanhui.
MGWJ. Zhonghua minguo waijiaoshi ziliao xuanbian 中華民國外交史資料選編
(Selected source material on the diplomatic history of the Republic of China), edited
by Cheng Daode 程道德 et al. 2 vols. 1988. Beijing: Beijing daxue chubanshe.
MWJ. Mao Zedong wenji 毛澤東文集 (Collection of the works of Mao Zedong), 8 vols.,
edited by Zhonggong zhongyang wenxian yanjiushi 中共中央文獻研究室.
1993–1999. Beijing: Renmin chubanshe.
MXJ. Mao Zedong xuanji 毛澤東選集 (Selected works of Mao Zedong), 4 vols., edited
by Zhonggong zhongyang wenxian bianji weiyuanhui 中共中央文獻編輯委員會.
1991. Beijing: Renmin chubanshe.
QCXW. Qingchao xu wenxian tongkao 清朝續文獻通考 (Continuation of the compre-
hensive investigations of records and documents of the Qing dynasty), 4 vols., edited
by Liu Jinzhao 劉錦藻. 2000. Hangzhou: Zhejiang guji chubanshe.
QSG. Qingshigao 清史稿 (A draft history of the Qing). 1979. Beijing: Zhonghua shuju.
QSL. Qingshilu 清實錄 (Veritable records of the Qing). 2008. Beijing: Zhonghua
shuju.
SBZ. Shen Wensu gongdu 沈文肅公牘 (Official documents of Shen Baozhen), edited by
Shen Baozhen 沈葆楨. 2008. Fuzhou: Fujian renmin chubanshe.
SLGB. Jiang Zhongzheng zongtong dang’an shilue gaoben 蔣中正總統檔案 事略稿本
(Archival documents about President Chiang Kai-shek: Drafted outline of activities),
70 vols., edited by Wang Zhenghua 王正華, et al. 2003–2012. Taipei: Guoshiguan.
300 References
SQYJ. Shiqu yuji 石渠餘記 (Additional records of Shiqu), edited by Wang Qingyun 王
慶雲. 1985. Beijing: Beijing guji chubanshe.
SSJY. Zhonghua minguo shishi jiyao chugao 中華民國史事紀要（初稿）(Draft of an
outline of historical events of the Republic of China), edited by Zhonghua minguo
shishi jiyao bianji weiyuanhui 中華民國史事紀要編輯委員會.
SSL. Yi jiu er qi nian de Shanghai shengye lianhehui 一九二七年的上海商業聯合會
(The Shanghai association of commerce in 1927), edited by Shanghaishi dang’anguan
上海市檔案館. 1983. Shanghai: Shanghai renmin chubanshe.
SZS. Sun Zhongshan quanji 孫中山全集 (Complete works of Sun Zhongshan), edited by
Zhongguo shehui kexueyuan jindaishi yanjiushuo 中國社會科學院近代史研究所,
Guangdongsheng shehui kexueyua lishi yanjiusuo 廣東省社會科學院歷史研究所,
and Zhongshan daxue lixishi 中山大學歷史系, 11 vols. 1981. Beijing: Zhonghua
shuju.
TDGG. Jiefang zhanzheng shiqi tudi gaige wenjian xuanbian, yijiusiwu – Yijiusijiu解
放戰爭時期土地改革文件選編 （一九四五──一九四九） (Selected documents on
land reform during the Liberation War period, 1945–1949), edited by Zhongyang
dang’anguan 中央檔案館. 1981. Beijing: Zhonggong zhongyang dangxiao
chubanshe.
TN. The Treaty of Nanking, Signed August 29th, 1842. And the Supplementary Treaty,
Signed October 8th, 1843. 1844. Macao: The Office of the Chinese Repository.
USFR. Foreign Relations of the United States Diplomatic Papers, The Conferences at
Cairo and Tehran, 1943, Department of State Publication 7187, Historical Office,
Bureau of Public Affairs. 1961. Washington: United States Government Printing
Office.
WSZL. Wenshi ziliao xuanji 文史資料選輯 (Selected collection of cultural and historical
source material), edited by Zhongguo renmin zhengzhi xieshanghuiyi quanguo
weiyuanhui wenshi ziliao yanjiu weiyuanhui 中國人民政治協商會議全國委員會文
史資料研究委員會. 1960–2013. 163 vols. Beijing:Zhonghua shuju, Wenshi ziliao
chubanshe, Zhongguo wenshi chubanshe.
WWZQ. Wang wei zhengquan ziliao xuanbian 汪偽政權資料選編 (Selected source
material on Wang Jingwei’s puppet government), vol. 1, Wang Jingwei guomin
zhengfu chengli 汪精衛國民政府成立 (The establishment of Wang Jingwei’s Nation-
alist government), edited by Huang Meiyun 黃美真 and Zhang Yun 張雲. 1984.
Shanghai: Shanghai renmin chubanshe.
WY. Shengwuji 聖武記 (Records of the military feats of great sages), edited by
Wei Yuan 魏源. 1984. Beijing: Zhonghua shuju.
WZZ. Wang Zizhuang riji 王子壯日記 (Diaries of Wang Zizhuang), 10 vols. 2001.
Taipei: Zhongyang yanjiuyuan jindaishi yanjiusuo.
XZDA. Yuan yilai Xizang difang yu zhongyang zhengfu guanxi dang’an shiliao hui-
bian 元以來西藏地方與中央政府關係檔案史料彙編 (Compilation of historical
sources on the relationship between local Tibet authorities and the central government
sing the Yuan dynasty), 7 vols., edited by Zhongguo zangxue yanjiu zhongxin 中國
藏學研究中心, et al. 1994. Beijing: Zhongguo zangxue chubanshe.
XZDF. Xizang difang jindaishi ziliao xuanji 西藏地方近代史資料選集 (Selected col-
lection of source material on the local modern history of Tibet), edited by
Xu Guangzhi 許廣智and Dawa 達瓦. 2007. Lhasa: Xizang renmin chubanshe.
XZZL. Qingdai yilai zhongyang zhengfu dui Xizang de zhili yu huofo zhuanshi zhidu
shiliao huiji 清代以來中央政府對西藏的治理與活佛轉世制度史料彙集 (Compil-
ation of historical sources on the central authorities’ governance of Tibet and the
 References 301
practice of the Living Buddha’s reincarnation), edited by Zhao Xueyi 趙學毅,
Chang Weimin 常為民 and Ou Shengmeng 歐聲明. 1996. Beijing: Huawen
chubanshe.
YBSWJ. Yingbingshi wenji dianjiao 飲冰室文集點校 (The collected essays of Liang
Qichao, revised and updated), edited by Liang Qichao 梁啓超. 2001. Kunming:
Yunnan jiaoyu chubanshe
YWYD. Yangwu yundong 洋務運動 (The Westernizing Movement), 8 vols., edited by
Zhongguo kexueyuan jindaishi yanjiusuo shiliao bianjishi 中國科學院近代史研究所
史料編輯室and Zhongyang dang’anguan Ming-Qing dang’anbu bianjizu中央檔案
館明清檔案部編輯組. 1951. Shanghai: Shanghai renmin chubanshe.
YZHB. Zhongwei jiu yuezhang huibian 中外舊約章彙編 (Compilation of historical
treaties and agreements between China and foreign countries), 3 vols., edited by
Wang Tieya 王鐵崖. 1957. Beijing: Sanlian shudian.
ZGF. Zeng Guofan quanji 曾國藩全集 (Complete works of Zeng Fuofan), edited by
Luo Zhenyue 羅鎮岳 and Han Changgeng 韓長耕 1985. Changsha: Yuelu shushe.
ZGZY. Zhonggong zhongyang wenjian xuanji 中共中央文件選集 (Selected documents
of the CCP Central Committee), vol. 1–18, edited by Zhongyang dang’anguan 中央
檔案館. 1991. Beijing: Zhonggong zhangyang dangxiao chubanshe.
ZHWS. Zhonghua wenshi ziliao wenku 中華文史資料文庫 (Compendium of essays on
Chinese culture and history), 20 vols., edited by Quanguo Zhengxie wenshi ziliao
huiyuanhui 全國政協文史資料委員會. 1996. Beijing: Zhongguo wenshi
chubanshe.
ZZT. Zuo Zongtang quanji 左宗棠全集 (Complete works of Zong Zongtang), edited by
Liu Yangyang 劉泱泱. 1992. Changsha: Yuelu shushe.
ZZZ. Zhang Zhizhong huiyilu 張治中回憶錄 (Zhang Zhizhong’s memoire), edited by
Zhang Zhizhong張治中. 2007. Beijing: Huawen chubanshe.

Secondary works
Alter, Peter. 1994. Nationalism, 2nd ed. London: Edward Arnold.
Anderson, Benedict. 2006. Imagined Communities: Reflections on the Origin and
Spread of Nationalism, revised ed. London: Verso.
Ao Guangxu 敖光旭. 2007. “1920 niandai guonei Menggu wenti zhi zheng:
Yi Zhong-E jiaoshe zuihou jieduan zhi lunzheng wei zhongxin” 1920 年代國
內蒙古問題之爭──以中俄交涉最後階段之論爭為中心 (Domestic debate on the
Mongolian issue in the 1920s: The case of the controversy during the
last phase of the Sino-Russian negotiation), Jindaishih yanjiu 近代史研究, 4:
55–73.
Baer, Marc. 2008. Honored by the Glory of Islam: Conversion and Conquest in Ottoman
Europe. Oxford: Oxford University Press.
Bain, Attwood. 2003. Rights for Aborigines. Sydney: Allen & Unwin.
Balibar, Etienne. 1991. “Is There a ‘Neo-Racism’?” In Eienne Balibar and
Immanuel Wallerstein, eds., Race, Nation, Class: Ambiguous Identities (London:
Verso), 17–28.
Baoyinchaoketu 寶音朝克圖. 2007. “Qingdai Menggu diqu kalun shezhi shijian kao:
Yi Mobei diqu wei zhongxin” 清代蒙古地區卡倫設置時間考──以漠北地區為中心
(An inquiry into the timing of Karan installation in the Mongolian areas under the
Qing: The case of Mobei region). Hebei shifan daxue xuebao 河北師範大學學報,
30.2: 135–138.
302 References
Bayly, C. A. 2004. The Birth of the Modern World, 1780–1914: Global Connections and
Comparisons. Malden, MA: Wiley-Blackwell.
Bergère, Marie-Claire. 1986. The Golden Age of the Chinese Bourgeoisie, 1911–1939.
Cambridge, UK: Cambridge University Press.
Bergère, Marie-Claire. 1998. Sun Yat-sen. Stanford, CA: Stanford University Press.
Bernhardt, Kathryn. 1992. Rents, Taxes, and Peasant Resistance: The Lower Yangzi
Region, 1840–1950. Stanford, CA: Stanford University Press.
Bian, Morris L. 2005. The Making of the State Enterprise System in Modern China: The
Dynamics of Institutional Change. Cambridge, MA: Harvard University Press.
Bohai, shouchen 渤海壽臣. 1969. “Xinhai geming shimo ji” 辛亥革命始末記 (A his-
tory of the 1911 Revolution). In Shen Yunlong 沈雲龍ed., Jindai Zhongguo shiliao
congkai 近代中國史料叢刊, 1.42. Taipei: Wenhai chubanshe.
Bonney, Richard. 1981. The King’s Debts: Finance and Politics in France, 1589–1661.
New York: Oxford University Press.
Bonney, Richard. 1988. Society and Government in France under Richelieu and Maza-
rin, 1624–61. New York: St. Martin’s Press.
Bourke, Richard and Skinner, Quentin, eds. 2015. Popular Sovereignty in Historical
Perspective. Cambridge: Cambridge University Press.
Braude, Benjamin. 2014. Christians and Jews in the Ottoman Empire. Boulder, CO:
Lynne Rienner.
Brewer, John. 1989. The Sinews of Power: War, Money and the English State, 1688–
1783. London: Routledge.
Brubaker, Rogers. 1996. Nationalism Reframed: Nationhood and the National Ques-
tion in the New Europe. Cambridge: Cambridge University Press.
Burbank, Jane and Cooper,Frederick. 2010. Empress in World History: Power and the
Politics of Difference. Princeton, NJ: Princeton University Press.
Cai Huilin 蔡惠霖 and Sun Weihou 孫維吼. 1987. Guangrong de jueze: Yuan Guo-
mindang qiyi jiangling huiyilu 光榮的抉擇——原國民黨起義將領回憶錄 (Glorious
choice: Memoirs of former Guomindang defecting commanders), part 1. Beijing:
Zhongguo renmin jiefangjun guofang daxue chubanshe.
Cassel, Par Kristoffer. 2012. Grounds of Judgment: Extraterritoriality and Imperial
Power in Nineteenth-Century China and Japan. New York: Oxford University
Press.
Ch’en, Jerome. 1979. The Military-Gentry Coalition: China under the Warlords.
Toronto: University of Toronto-York University Joint Center on Modern East Asia.
Ch’i, Hsi-sheng. 1976. Warlord Politics in China, 1916–1928. Stanford, CA: Stanford
University Press.
Ch’i, Hsi-sheng. 1982. Nationalist China at War: Military Defeats and Political Col-
lapse, 1937–45. Ann Arbor, MI: University of Michigan Press.
Ch’ü, Tung-tsu. 1962. Local Government in China under the Ch’ing. Cambridge,
MA: Harvard University Press.
Chang, Chung-li. 1955. The Chinese Gentry: Studies on Their Role in Nineteenth-
Century Chinese Society. Seattle, WI: University of Washington Press.
Chang, Chung-li. 1962. The Income of the Chinese Gentry: A Sequel to the Chinese
Gentry: Studies on Their Role in Nineteenth-Century Chinese Society. Seattle, WI:
University of Washington Press.
Chen Feng 陳鋒. 1988. Qingdai yanzheng yu yanshui 清代鹽政與鹽稅 (Salt
administration and taxation under the Qing). Zhengzhou: Zhongzhou guji
chubanshe.
 References 303
Chen Feng 陳鋒. 1992. Qingdai junfei yanjiu 清代軍費研究 (A study of military
expenses under the Qing). Wuhan: Wuhan daxue chubanshe.
Chen Feng 陳鋒. 2008. Qingdai caizheng zhengce yu huobi zhengce yanjiu 清代財政政
策與貨幣政策研究 (A study of fiscal and monetary policies under the Qing). Wuhan:
Wuhan daxue chubanshe.
Chen Fulin 陳福霖 and Yu Yanguang 余炎光. 1991. Liao Zhongkai nianpu 廖仲愷年
譜 (A chronology of Liao Zhongkai). Changsha: Hunan Chubanshe.
Chen Huisheng 陳慧生 and Chen Chao 陳超. 2007. Minguo Xinjiang shi 民國新疆史
(A history of Xinjiang during the Republican period). Urumqi: Xinjiang renmin
chubanshe.
Chen Jinjin 陳進金. 1997. Kangzhan qian jiaoyu zhengce zhi yanjiu 抗戰前教育政策
之研究 (A study of the educational policy before the War of Resistance). Taipei: Jindai
Zhongguo chubanshe.
Chen Jinjin 陳進金. 2000. “Dongbeijun yu Zhongyuan dazhan” 東北軍與中原大戰
(The Northeaster Army and the war in North China), Jindaishi yanjiu 近代史研究,
5: 1–34.
Chen Mingsheng 陳明勝 and Shen Xiaoyun申曉雲. 2009. “Jiangsu sheng fei du yun-
dong yu sheng zizhi 江蘇省廢督運動與省自治” (The movement for abolishing the
governor and provincial self-government in Jiangsu), Jiangsu shehui kexue 江蘇社會
科學, 6: 179–183.
Chen Shuwei 程舒偉 and Chang Jiashu 常家樹. 1997. Kang Ri zhanzheng zhongyao
wenti yanjiu 抗日戰爭重要問題研究 (A study of important issues regarding the War
of Resistance against Japan). Shenyang: Dongbei daxue chubanshe.
Chen, Yung-fa. 1986. Making Revolution: The Communist Movement in Eastern and
Central China, 1937–1945. Berkeley, CA: University of California Press.
Chen Yunxi 陳蘊西. 2007. “Shenti zhengzhi: Guojia quanli yu minguo zhongshanz-
huang de liuxing” 身體政治：國家權力與民國中山裝的流行 (Body politics: State
power and the popularization of the Sun Yat-sen-style tunic in the Republican
years), Xueshu yuekan 學術月刊, 39.9: 139–147.
Cheng, Victor S. 2005. “Imagining China’s Madrid in Manchuria: The Communist
Military Strategy at the Onset of the Chinese Civil War, 1945–1946,” Modern
China, 31.1: 72–114.
Cheng Zhongyuan 程中原. 2006. Zhang Wentian zhuan 張聞天傳 (Biography of
Zhang Wentian). Beijing: Dangdai Zhongguo chubanshe.
Ci hongfei 慈鴻飛. 1994. “Ersanshi niandai jiaoshi, gongwuyuan gongzi ji shenghuo
zhuangkuang kao” 二三十年代教師、公務員工資及生活狀況考 (An inquiry into
the salaries and living condition of teachers and civil servants in the 1920s and
1930s), Jindaishi yanjiu 近代史研究, 3: 285–291.
Ci Yunfei 遲雲飛. 2001. “Qingmo zuihou shinian de ping Man-Han zhenyu wenti”
清末最後十年的平滿漢畛域問題 (The issue of narrowing the gap between the
Manchus and Han in the last decade of the Qing), Jindaishi yanjiu 近代史研究, 5:
21–44.
Coble, Parks. 1986. The Shanghai Capitalists and the Nationalist Government, 1927–
1937. Cambridge, MA: Harvard University Press.
Cohen, Paul A. 2010. Discovering History in China: American Historical Writing on
the Recent Chinese Past. New York: Columbia University Press.
Collins, Randall. 1999. Macrohistory: Essays in Sociology of the Long Run. Stanford,
CA: Stanford University Press.
304 References
Crook, Isabel and Crook, David. 1979. Ten Mile Inn: Mass Movement in a Chinese
Village. New York: Pantheon Books.
Crossley, Pamela Kyle. 1990. Orphan Warriors: Three Manchu Generations and the
End of the Qing World. Princeton, NJ: Princeton University Press.
Crossley, Pamela Kyle. 1992. “The Rulerships of China,” American Historical Review,
97.5: 1468–1483.
Crossley, Pamela Kyle. 2002. A Translucent Mirror: History and Identity in Qing
Imperial Ideology. Berkeley: University of California Press.
Dai Yi 戴逸. 1992. Qianlong di jiqi shidai 乾隆帝及其時代 (Emperor Qianlong and
his times). Beijing: Zhongguo renmin daxue chubanshe.
Dai, Yingcong. 2009. The Sichuan Frontier and Tibet: Imperial Strategy in the Early
Qing. Seattle, WI: University of Washington Press.
DeMare, Brian. 2019. Land Wars: The Story of China’s Agrarian Revolution. Stanford,
CA: Stanford University Press.
Deng Shaohui 鄧紹輝. 1998. Wanqing caizheng yu Zhongguo jindaihua 晚清財政與
中國近代化 (Finance and modernization in late-Qing China). Chengdu: Sichuan
renmin chubanshe.
Deng Yibing 鄧亦兵. 2004. “Qingdai qianqi yishang wenti xintan” 清代前期抑商問題
新探 (A new inquiry into the issue of limitations on commerce in the first half of the
Qing), Shoudu shifan daxue xuebao 首都師範大學學報, 4: 1–8.
Ding, Yizhuang, Guo, Songyi, Li, Zhongqing (James Z. Lee), Kang, Wenlin (Cam-
eron Campbell). 2004. Liaodong yimin Zhong de qiren shehui: Lishi wenxian,
renkou tongji yu tianye diaocha 遼東移民中的旗人社會：歷史文獻、人口統計與田
野調查 (Immigration and Eight Banner Society in Liaodong). Shanghai: Shanghai
sheui kexueyuan chubanshe.
Dong Conglin 董叢林. 1994. “Lingxiu daoxiang yu xianghuai xi shili de ‘yiliu’” 領袖導
向與湘淮系勢力的“異流”(The inclinations of leaders and the “divergence” between
the Hunan and Anhui cliques), Jindaishi yanjiu 近代史研究, 2: 14–36.
Dong Xianguang 董顯光. 1952. Jiang zongtong zhuan 蔣總統傳 (A biography of Presi-
dent Chiang). Taipei: Zhonghua wenhua chuban shiye weihuanhui.
Downing, Brian. 1992. The Military Revolution and Political Change: Origins of Dem-
ocracy and Autocracy in Early Modern Europe. Princeton, NJ: Princeton University
Press.
Doyle, Michael W. 1986. Empires. Ithaca, NY: Cornell University Press.
Du Yuming 杜聿明. 1985. “Guomindang pohuai heping jingong Dongbei
shimo” 國民黨破壞和平進攻東北始末 (The process of the Guomindang’s
undermining of peace and attack on the Northeast). In Zhongguo renmin
zhengzhi xieshang huiyi quanguo weiyuanhui wenshi ziliao yanjiu weiyuanhui
Liaoshen zhanyi qinli ji bianshenzu 中國人民政治協商會議全國委員會文史資料
研究委員會《遼瀋戰役親歷記》編審組, ed., Liaoshen zhanyi qinliji: Yuan Guo-
mindang jiangling de huiyi 遼瀋戰役親歷記——原國民黨將領的回憶 (Beijing:
Zhongguo wenshi chubanshe): 514–564.
Duara, Prasenjit. 1991. Culture, Power, and the State: Rural North China, 1900–
1942. Stanford, CA: Stanford University Press.
Dunstan, Helen. 2006. State or Merchant? Political Economy and Political Process in
1740s China. Cambridge, MA: Harvard University Press.
Eastman, Lloyd E. 1974. The Abortive Revolution: China under Nationalist Rule,
1927–1937. Cambridge, MA: Harvard University Press.
 References 305
Eastman, Lloyd E. 1984. Seeds of Destruction: Nationalist China in War and Revolu-
tion, 1937–1949. Stanford, CA: Stanford University Press.
Eisenstadt, Samuel N. 1963. The Political Systems of Empires: The Rise and Fall of the
Historical Bureaucratic Societies. New York: The Free Press of Glencoe.
Elliott, Mark C. 2001. The Manchu Way: The Eight Banners and Ethnic Identity in
Late Imperial China. Stanford, CA: Stanford University Press.
Elliott, Mark C. 2006. “Ethnicity in the Qing Eight Banners.” In Pamela
Kyle Crossley, Helen F. Siu and Donald S. Sutton, eds., Empire at the Margins: Cul-
ture, Ethnicity, and Frontier in Early Modern China. Berkeley, CA: University of
California Press, 27–57.
Elman, Benjamin A. 2004. “Naval Warfare and the Refraction of China’s Self-
Strengthening Reforms into Scientific and Technological Failure, 1865–1895,”
Modern Asian Studies, 38.2: 283–326.
Elman, Benjamin A. 2005. On Their Own Terms: Science in China, 1550–1900. Cam-
bridge, MA: Harvard University Press.
Elman, Benjamin A. 2006. A Cultural History of Modern Science in China. Cam-
bridge, MA: Harvard University Press.
Emerson, Rupert. 1960. From Empire to Nation: The Rise to Self-Assertion of Asian
and African Peoples. Cambridge, MA: Harvard University Press.
Ertman, Thomas. 1997. Birth of the Leviathan: Building States and Regimes in Medi-
eval and Early Modern Europe. Cambridge: Cambridge University Press.
Esherick, Joseph W. 1976. Reform and Revolution in China: 1911 Revolution in
Hunan and Hubei. Ann Arbor, MI: University of Michigan Press.
Esherick, Joseph W. 2006. “How the Qing Became China.” In Joseph
W. Esherick, Hasan Kayali and Eric Van Young, eds., Empire to Nation: Histor-
ical Perspectives on the Making of the Modern World (Lanham, MD: Rowman &
Littlefield), 229–259.
Esherick, Joseph W. 2010. “China and the World: From Tribute to Treaties to Popu-
lar Nationalism.” In Brantly Womack, ed., China’s Rise in Historical Perspective
(Lanham, MD: Rowman & Littlefield), 19–38.
Fairbank, John K. 1969. Trade and Diplomacy on the China Coast: The Opening of the
Treaty Ports, 1842–1854. Cambridge, MA: Harvard University Press.
Fairbank, John K. and Goldman, Merle. 2006. China: A New History. Cambridge,
MA: Harvard University Press.
Fairbank, John K and Liu, Kwang-ching, eds. 1980. The Cambridge History of China,
vol. 11, Cambridge: Cambridge University Press.
Fan Baichuan 樊百川. 2003. Qingji de yangwu xinzheng 清季的洋務新政 (The
modern enterprises in late Qing China). Shanghai: Shanghai shudian chubanshe.
Fan Wenlan 范文瀾. 1949. Zhongguo jindaishi 中國近代史 (A history of modern
China). Shanghai: Shenghuo dushu xinzhi.
Fan Wenlan 范文瀾, Cai Meibiao蔡美彪 et al. 2008. Zhongguo tongshi中國通史 (A
general history of China), 12 vols. Beijing: Renmin chubanshe.
FDS (Zhonghua renmin gongheguo caizhengbu Zhongguo nongmin fudan shi bianji
weiyuanhui 中華人民共和國財政部《中國農民負擔史》編輯委員會). 1990. Zhong-
guo nongmin fudan shi中國農民負擔史 (A history of the fiscal burden of Chinese peas-
antry), vol. 3. Beijing: Zhongguo caizheng jingji chubanshe.
Feng Erkang 馮爾康, ed. 2003. Qingchao tongshi: Yongzheng chao fenjuan 清朝通史,
雍正朝分卷 (A general history of the Qing dynasty: Volume on the Yongzheng reign).
Beijing: Zijincheng chubanshe.
306 References
Feng Lida 馮理達, Luo Yuanzheng 羅元錚. 2009. “Feng Yuxiang zai Meiguo de rizi
li” 馮玉祥在民國的日子裡 (Feng Yuxiang’s years in the United States). In
Renwu bianjibu, ed., Zuoye changfeng yi zhiqin 昨夜長風憶至親 (Recollections on
parents’ lives) (Beijing: Dongfang chubanshe), 1–11
Feng Tianfu 馮田夫 and Li Weiguang 李煒光. 2006. Zhongguo caizheng tongshi,
geming genjudi juan 中國財政通史 革命根據地卷 (A general history of finance in
China, the volume on the revolutionary base areas). Beijing: Zhongguo caizheng
jingji chubanshe.
Feuerwerker, Albert. 1958. China’s Early Industrialization: Sheng Hsuan-huai (1844–
1916) and Mandarin Enterprise. Cambridge, MA: Harvard University Press.
Feuerwerker, Albert. 1992. “Presidential Address: Questions about China’s Early
Modern Economic History That I Wish I Could Answer,” The Journal of Asian
Studies 51.4: 757–769.
Fitzgerald, John. 1995. “The Nationless State: The Search for a Nation in Modern
Chinese Nationalist,” The Australian Journal of Chinese Affairs, 33: 75–104.
Fletcher, Joseph. F. 1978. “Ch’ing Inner Asia c. 1800.” In John King Fairbank ed.,
The Cambridge History of China, vol. 10, Late Ch’ing 1800–1911, Part I. Cam-
bridge: Cambridge University Press.
Friedman, Edward, Paul G. Pickowicz and Mark Selden. 1991. Chinese Village, Social-
ist State. New Haven, CT: Yale University Press.
Furuya, Keiji 古屋奎二. 1977. Jiang Zongtong milu 蔣總統秘錄 (Secret records of
President Chiang). Taipei: Zhongyang ribao she.
Galtung, Johan. 1997. “Macrohistory and Macrohistorians: A Theoretical Frame-
work.” In Johan Galtung, ed., Macrohistory and Macrohistorians: Perspectives on
Individual, Social, and Civilizational Change (Westport, CT: Praeger), 1–10.
Gao Xiaolin 高曉林. 1999. “Guixi defang shilipai dui kang Ri zhanzheng de gong-
xian” 桂系地方實力派對抗日戰爭的貢獻 (The contribution of the Guangxi-based
local forces to the War of Resistance against Japan), Zhonggong dangshi yanjiu 中共
黨史研究, 2: 87–92.
Gellner, Ernest. 1997. Nationalism. New York: New York University Press.
Gellner, Ernest. 2006. Nations and Nationalism, 2nd ed. Oxford: Blackwell.
Gillin, Donald. 1967. Warlord: Yen Hsi-Shan in Shansi Province, 1911–1949. Prince-
ton, NJ: Princeton University Press.
Glete, Jan. 2002. War and the State in Early Modern Europe: Spain, the Dutch Repub-
lic and Sweden as Fiscal-Military States. New York: Routledge.
Goldstein, Melvyn C. 1989. A History of Modern Tibet, vol. I, The Demise of the Lama-
ist State. Berkeley, CA: University of California Press.
Goldstein, Melvyn C. 1997. The Snow Lion and the Dragon: China, Tibet, and the
Dalai Lama. Berkeley, CA: University of California Press.
Goldstone, Jack. 1991. Revolution and Rebellion in the Early Modern World. Berkeley,
CA: University of California Press.
Goldstone, Jack. 2004. “Neither Late Imperial nor Early Modern: Efflorescences and
the Qing Formation in World History.” In Lynn Struve, ed., The Qing Formation in
World-Historical Time. Cambridge, MA: Harvard University Press, 242–302.
Greenfeld, Liah. 1992. Nationalism: Five Roads to Modernity. Cambridge, MA: Har-
vard University Press.
Grunfeld, A. Tom. 1996. The Making of Modern Tibet. Armonk, NY: East Gate Book.
Gu Bao 谷苞. 2005. Xibei tongshi 西北通史 (A general history of the Northwest), 5
vols. Lanzhou: Lanzhou daxue chubanshe.
 References 307
Guilmartin, Jr., John F. 1988. “Ideology and Conflict: The Wars of the Ottoman
Empire, 1453–1606,” Journal of Interdisciplinary History, 18.4: 721–747.
Gunn, Steven. 2010. “War and the Emergence of the State: Western Europe,
1350–1600.” In Frank Tallett and D. J. B. Trim, eds., European Warfare, 1350–
1750 (Cambridge: Cambridge University Press), 50–73.
Gunn, Steven, Grummitt, David and Cools, Hans. 2008. “War and the State in Early
Modern Europe: Widening the Debate,” War in History, 15.4: 371–388.
Guo Chengkang 郭成康. 2000. “Ye tan Manzu hanhua” 也談滿族漢化 (The issue of
the Manchus’ Sinification revisited), Qingshi yanjiu 清史研究, 2: 24–35.
Guo Chengkang 郭成康. 2005. “Qingchao huangdi de Zhongguo guan” 清朝皇帝的
中國觀 (The Qing emperors’ views of China), Qingshi yanjiu 清史研究, 4: 1–18.
Guo Chengkang 郭成康 and Zheng Baofeng 鄭寶鳳. 1995. “Lun Qingdai ‘bu jiafu’
jiqi dui shehui jingji de yingxing” 論清代“不加賦”及其對社會經濟的影響 (The ‘no
tax escalation’ policy and its impact on social economy during the Qing), Shehui
kexue jikan 社會科學輯刊, 2: 100–107.
Guo Songyi 郭松義. 1994. “Qing qianqi nanfang daozuoqu de liangshi shengchan”
清前期南方稻作區的糧食生產 (Grain production in the southern rice-growing
areas in the first half of the Qing), Zhongguo jingjishi yanjiu 中國經濟史研究, 1:
1–30.
Guo Songyi 郭松義. 1995. “Qingdai beifang hanzuoqu de liangshi shengchan” 清代
北方旱作區的糧食生產 (Grain production in the northern dry-crop areas in the
Qing), Zhongguo jingjishi yanjiu中國經濟史研究, 1: 22–44.
Guo Songyi 郭松義. 2001. “Ming Qing shiqi de liangshi shengchan yu nongmin shen-
ghuo shuiping” (Grain production and peasants’ living condition in the Ming and
the Qing) 明清時期糧食生產與農民生活水平. In Zhongguo shehui kexueyuan lishi
yanjiusuo xuekan bianweihui 中國社會科學院歷史研究所編委會, ed. Zhongguo
shehui kexueyuan lishi yanjiusuo xuekan 中國社會科學院歷史研究所學刊, vol. 1.
Beijing: Shehui kexue wenxian chubanshe, 373–396.
Guo Xuyin 郭緒印. 1992. Guomindang paixi douzheng shi 國民黨派系鬥爭史 (A his-
tory of factional struggle within the Guomindang). Shanghai: Shanghai renmin
chubanshe.
Guy, Kent R. 2010. Qing Governors and Their Provinces: The Evolution of Territorial
Administration in China, 1644–1796. Seattle, WI: University of Washington Press.
Hai Chunliang 海純良. 2009. “Qingmo xinzheng yu Waimenggu duli” 清末新政與外
蒙古獨立 (The late-Qing “new policies” and the independence of Outer Mongolia),
Neimenggu minzu daxue xuebao 內蒙古民族大學學報, 35.1: 35–38.
Halsey, Stephen. 2015. Quest for Power: European Imperialism and the Making of
Chinese Statecraft. Cambridge, MA: Harvard University Press.
Hamashita, Takeshi. 2008. China, East Asia and the Global Economy: Regional and
Historical Perspectives. New York: Routledge.
Hao Bingrang 郝秉讓. 2001. Fengxi junshi 奉系軍事 (The military of the Fengtian
clique). Shenyang: Liaohai chubanshe.
Harding, Alan. 2002. Medieval Law and the Foundations of the State. New York:
Oxford University Press.
Harriman, W. Averell. 1975. Special Envoy to Churchill and Stalin, 1941–1946.
New York: Random House.
Harrison, Henrietta. 2001. Inventing the Nation: China. London: Hodder Education.
Harriss, Gerald. 1993. “Political Society and the Growth of Government in Late
Medieval England,” Past & Present, 138: 28–57.
308 References
He Jiawei 何家偉. 2009. “Nanjing guomin zhengfu gongwuyuan shuliang de
pengzhang jiqi kuibai zhi kaichao” 南京國民政府公務員數量的膨脹及其潰敗之
考察 (An examination of the expansion of the number of civil servants of the
Nationalist government in Nanjing and its collapse), Renwen zazhi 人文雜誌,
2: 145–152.
He Jiawei 何家偉 and Luo Jun 駱軍. 2011. “Guomin zhengfu gongwuyuan fengei yu qi
jingji diwei guanxi shanbian shulun” 國民政府公務員俸給與其經濟地位關係嬗變述
論 (The civil servants of the Republican government: Changing relationship between
their salaries and economic conditions), Bianjiang jingji yu wenhua 邊疆經濟與文化,
10: 39–43.
He Ping 何平. 1998. Qingdai fushui zhengce yanjiu: 1644–1840 nian 清代賦稅政策研
究：1644–1840 年 (A study of the tax policies under the Qing, 1644–1840). Beijing:
Zhongguo shehui kexue chubanshe.
He, Wenkai. 2013. Paths toward the Modern Fiscal State: England, Japan, and China.
Cambridge, MA: Harvard University Press.
He Yuan 賀淵. 1998. “1912–1927 nian Yan Xishan zhi Jin sixiang chutan” 1912–1927年
閻錫山治晉思想初探 (a preliminary study on Yan Xishan’s thoughts on governing
Shanxi during the period from 1912 to 1927), Jindaishi yanjiu 近代史研究, 1: 232–
245.
He Zhaowu 何兆武 and Wen Jing 文靖. 2008. Shangxue ji 上學記 (Recollections on
the school years). Beijing: Sanlian shudian.
Hechter, Michael. 2000. Containing Nationalism. Oxford: Oxford University Press.
Hertslet, Godfrey E. 1908. Hertslet’s China Treaties. London: Harrison & Sons.
Hinton, William. 1966. Fanshen: A Documentary of Revolution in a Chinese Village.
New York: Vintage Books.
Hintze, Otto. 1975. “Military Organization and the Organization of the State.” In
Felix Gilbert, ed., The Historical Essays of Otto Hintze (New York: Oxford University
Press), 178–215.
Ho, Ping-ti. 1959. Studies on the Population of China, 1368–1953. Cambridge, MA:
Harvard University Press.
Ho, Ping-ti. 1967. “The Significance of the Ch’ing Period in Chinese History,” The
Journal of Asian Studies, 26.2: 189–195.
Ho, Ping-ti. 1998. “In Defense of Sinicization: A Rebuttal of Evelyn Rawski’s ‘Reen-
visioning the Qing’,” The Journal of Asian Studies, 57.1: 123–155.
Hobsbawn, E. J. 1987. The Age of Empire, 1875–1914. New York: Vintage Books.
Hobsbawn, E. J. 2012. Nations and Nationalism since 1780: Programme, Myth and
Reality, 2nd ed. Cambridge: Cambridge University Press.
Hobson, John M. 2006. The Eastern Origins of Western Civilization. Cambridge:
Cambridge University Press.
Hobson, John M. 2012. The Eurocentric Conception of World Politics: Western Inter-
national Theory, 1760–2010. Cambridge: Cambridge University Press.
Hou Yijie 候宜杰. 1993. Ershi shiji chu Zhongguo zhengzhi gaige fengchao 二十世紀初
中國政治改革風潮 (The turmoil of political reforms in early twentieth-century
China). Beijing: Renmin chubanshe.
Howe, Stephen. 2002. Empire: A Very Short Introduction. Oxford: Oxford University
Press.
Hsiao, Kung-chuan. 1960. Rural China: Imperial Control in the Nineteenth Century.
Seattle, WI: University of Washington Press.
 References 309
Hu Sheng 胡繩. 1981. Cong Yapian zhanzhan dao Wusi yundong 從鴉片戰爭到五四運
動 (From the Opium War to the May Fourth Movement). Beijing: Renmin chubanshe.
Hu Sheng 胡繩. 2001. “Hu Sheng guanyu zhuanxie Cong wusiyundong dao renmin
gongheguo chengli yishu de tanhua” 胡繩關於撰寫《從五四運動到人民共和國成
立》一書的談話 (Hu Sheng’s talks on the writing of From the May Fourth Move-
ment to the Establishment of the People’s Republic). In Cong wusiyundong dao
renmin gongheguo chengli ketizu “從五四運動到人民共和國成立”課題組, Hu Sheng
lun “Cong wusiyundong dao renmin gongheguo chengli” 胡繩論“從五四運動到人民共
和國成立” (Hu Sheng on From the May Fourth Movement to the Establishment of the
People’s Republic). Beijing: Shehui kexue wenxian chubanshe, 3–51.
Hu Yuhai 胡玉海 and Li Rong 里蓉. 2005. Fengxi junfa dashiji 奉系軍閥大事記 (The
Fengtian clique: a chronology). Shenyang: Liaoning minzu chubanshe.
Huang Daoxuan 黃道炫. 1999. “Guanyu Jiang Jieshi diyici xiaye de jige wenti” 關於
蔣介石第一次下野的幾個問題 (Several issues regarding Chiang Kai-shek’s initial
resignation), Jindaishi yanjiu 近代史研究, 4: 142–165.
Huang, Philip. 1988. The Peasant Economy and Social Change in North China. Stan-
ford, CA: Stanford University Press.
Huang, Philip. 2001. Code, Custom, and Legal Practice in China: The Qing and the
Republic Compared. Stanford, CA: Stanford University Press.
Huang, Philip. 2010. Chinese Civil Justice, Past and Present. Lanham, MD: Rowman
& Littlefield.
Huang Xingtao 黃興濤. 2010. “Xaidai Zhonghua minzu guannian xingcheng de
lishi kaochao: Jianlun Xinhai geming yu Zhonghua minzu rentong zhi guanxi
現代“中華民族”觀念形成的歷史考察——兼論辛亥革命與中華民族認同之關係”
(A historical survey of the formation of the modern concept of the Chinese
nation: With a supplementary discussion on the relationship between the Revo-
lution of 1911 and the identification with the Chinese nation). In Liu Fengyun 劉鳳
雲 and Liu Wenpeng劉文鵬, eds., Qingchao de guojia rentong清朝的國家認同 (Iden-
tities under the Qing state). Beijing: Zhongguo renmind daxue chubanshe.
Huang Xingtao 黃興濤. 2011a. “Qingdai Manren de ‘Zhongguo rentong’” (清代滿
人的‘中國認同), Qingshi yanjiu a清史研究, 1: 1–12.
Huang Xingtao 黃興濤. 2011b. Wenhuashi de zhuixun: Yi jinshi Zhongguo wei shiyu
文化史的追尋：以近世中國為視域 (An inquiry into cultural history, with a focus on
modern China). Beijing: Zhongguo renmin daxue chubanshe.
Huang Yao 黃瑤, Li Weimin 李維民, Pan Tianjia 潘天嘉, Yang Guoqing 楊國慶,
Bai Ren 白刃 and Li Zhijing 李志經. 1993. Luo Ronghuan zhuan 羅榮桓傳 (Bibli-
ography of Luo Ronghuan). Beijing: Dangdai Zhongguo chubanshe.
Isaacs, Harold. 1951. The Tragedy of the Chinese Revolution. Stanford, CA: Stanford
University Press.
Jia Shiyi 賈士毅. 1932. Minguo xu caizheng shi (1) 民國續財政史 （一） (Supple-
ment to the fiscal history of the Republican period), vol. 1. Shanghai: Shangwu
yinshuguan.
Jia Shiyi 賈士毅. 1933. Minguo xu caizheng shi (2) 民國續財政史 （二） (Supple-
ment to the fiscal history of the Republican period), vol. 2. Shanghai: Shangwu
yinshuguan.
Jia Shiyi 賈士毅. 1934. Minguo caizheng shi 民國財政史 (Fiscal history of the Repub-
lican period). Shanghai: Shangwu yinshuguan.
310 References
Jia Shuocun 賈熟村. 2004. “Ding Baozhen biannian shilue” 丁寶楨編年事略 (A
chronological outline of Ding Baozhen’s activities), Linyi shifan xueyuan xuebao 臨
沂師範學院學報, 26.1: 90–94.
Jiang Jieshi 蔣介石. 1946. Zhongguo zhi mingyun 中國之命運 (China’s Destiny).
Shanghai: Zhengzhong shuju.
Jiang Tao 姜濤. 1990. “Qingdai renkou tongji zhidu yu 1741–1851 nianjian de
Zhongguo renkou” 清代人口統計制度與1741–1851年間的中國人口 (The census
system of the Qing and the population of China in 1741–1851), Jindaishi yanjiu 近
代史研究, 5: 26–50.
Jiang Tingfu 蒋廷黻. 1939. Zhongguo jindaishi 中國近代史 (Modern Chinese history).
Shanghai: Shangwu yinshuguan.
Jiang Zhijie 蔣致潔. 1988. “Zuo Zongtang shoufu Xinjiang zhanyi junxiang wenti
tantao” 左宗棠收復新疆戰役軍餉問題探討 (An inquiry into the issue of the military
spending on Zuo Zongtang’s campaign to recover Xinjiang), Zhongguo shehui jing-
jishi yanjiu 中國社會經濟史研究, 2: 26–31.
Jiangsu sheng zhengxie wenshi ziliao weiyuanhui 江蘇省政協文史資料委員會 ed. 1993.
Minguo Jiangsu de dujun he shengzhang 民國江蘇的督軍和省長 (The military governor
and the provincial governor in Republican Jiangsu). Nanjing: Jiangsu wenshi ziliao
bianjibu.
Jin Chongji 金衝及. 2001. “Xinhai geming he Zhongguo jindai minzuzhuyi” 辛亥革
命和中國近代民族主義 (The Revolution of 1911 and nationalism in modern
China), Jindaishi yanjiu 近代史研究, 5: 1–20.
Jin Chongji 金衝及 2006. “Jiaoliang: Dongbei jiefang zhanzheng de zuichu jieduan”
較量：東北解放戰爭的最初階段 (A showdown: The initial phase of the Liberation
War in the Northeast), Jindaishi yanjiu 近代史研究, 4: 1–28.
Jin Yilin 金以林. 2005. “Diyu guannian yu paixi chongtu: Yi ersanshi niandai
Guomindang Yue ji lingxiu wei zhongxin de kaocha” 地域觀念與派系衝突──以
二三十年代國民黨粵籍領袖為中心的考察 (Native-townsmanship and factional
strife: An examination focusing on the Guangdong-originated Guomindang
leaders in the 1920s and 1930s), Lishi yanjiu 歷史研究, 3: 115–128.
Johnson, Chalmers. 1962. Peasant Nationalism and Communist Power: The Emer-
gence of Revolutionary China, 1937–1945. Stanford, CA: Stanford University Press.
Jordan, Donald. 1976. The Northern Expedition: China’s National Revolution of
1926–1928. Honolulu: University of Press of Hawaii.
Kaeuper, Richard W. 1988. War, Justice, and Public Order: England and France in the
Later Middle Ages. New York: Oxford University Press.
Kapp, Robert. 1973. Szechwan and the Chinese Republic. New Haven, CT: Yale University
Press.
Keating, Pauline. 1997. Two Revolutions: Village Reconstruction and the Coopera-
tive Movement in Northern Shaanxi, 1934–1945. Stanford, CA: Stanford Uni-
versity Press.
Kim, Kwangmin. 2016. Borderland Capitalism: Turkestan Produce, Qing Silver, and
the Birth of an Eastern Market. Stanford, CA: Stanford University Press.
Kirby, William. 1984. Germany and Republican China. Stanford, CA: Stanford Uni-
versity Press.
Kirby, William. 1997. “The Internationalization of China: Foreign Relations at
Home and Abroad in the Republican Era,” The China Quarterly, 150: 433–
458.
 References 311
Kirby, William. 2000. “Engineering China: Birth of the Developmental State,
1928–1937.” In Wen-Hsin Yeh ed., Becoming Chinese: Passages to Modernity and
Beyond (Berkeley, CA: University of California Press), 137–160.
Kıvanç, Karaman K. and Pamuk, Şevket. 2010. “Ottoman State Finances in Com-
parative European Perspective, 1500–1914,” The Journal of Economic History,
70: 593–627.
Kong Jingwei 孔經緯 and Fu Xiaofeng 傅笑楓. 1989. Fengxi junfa guanliao ziben 奉
系軍閥官僚資本 (The bureaucratic capital of the Fentian clique of warlords). Chang-
chun: Jilin daxue chubanshe.
Kong Lingji 孔令紀, Qu Wanfa 曲萬法, Liu Yunzhen 劉運珍 and Liu Jinxing 劉錦星,
eds. 1993. Zhongguo lidai guanzhi (Government positions of the Chinese dynasties)
中國歷代官制. Jinan: Qilu shushe.
Kuang Shanji 匡珊吉 and Yang Guangyan 楊光彥. 1991. Sichuan junfa shi 四川軍閥史
(History of warlords in Sichuan). Chengdu: Sichuan renmin chubanshe.
Kuhn, Philip. 1970. Rebellion and Its Enemies in Late Imperial China: Militarization
and Social Structure, 1796–1864. Cambridge, MA: Harvard University Press.
Kuhn, Philip. 2002. Origins of the Modern Chinese State. Stanford, CA: Stanford Uni-
versity Press.
Kumar, Krishan. 2010. “Nation-states as Empires, Empires as Nation-states: Two
Principles, One Practice?” Theory and Society, 39.2: 119–143.
Kuo, Ting-yee and Liu,Kwang-Ching. 1978. “Self-Strengthening: The Pursuit of
Western Technology.” In Denis Twitchett and John K. Fairbank, eds., The Cam-
bridge History of China, vol. 10 (Cambridge: Cambridge University Press), 491–
542.
Lai, Chi-Kong. 1994. “Li Hung-chang and Modern Enterprise: The China Mer-
chants’ Company, 1872–1885.” In Samuel Chu and Kwang-ching Liu, eds., Li
Hung-chang and China’s Early Modernization (Armonk, NY: M.E. Sharpe),
216–247.
Lai Xinxia 來新夏, Jiao Jingyi 焦靜宜, Mo Jianlai 莫建來, Zhang Shuyong 張樹勇 and
Liu Benjun 劉本軍. 2000. Beiyang junfa shi 北洋軍閥史 (a history of the northern
warlords). Tianjin: Nankai daxue chubanshe.
Lary, Diana. 1974. Region and Nation: The Kwangsi Clique in Chinese Politics, 1925–
1937. Cambridge: Cambridge University Press.
Lary, Diana. 2015. China’s Civil War: A Social History, 1945–1949. Cambridge: Cam-
bridge University Press.
Lee, Daniel. 2016. Popular Sovereignty in Early Modern Constitutional Thought.
Oxford: Oxford University Press.
Levich, Eugene W. 1993. The Kwangsi Way in Kuomintang China, 1931–1939.
Armonk, NY: Sharpe.
Levine, Steven I. 1987. Anvil of Victory: The Communist Revolution in Manchuria,
1945–1948. New York: Columbia University Press.
Lew, Christopher. 2009. The Third Chiense Revolutionary Civil War, 1945–49: An
Analysis of Communist Strategy and Leadership. New York: Routledge.
Lewis, Bernard. 1958. “Some Reflections on the Decline of the Ottoman Empire,”
Studia Islamica, 9: 111–127.
Lewis, Bernard. 2002. The Emergence of Modern Turkey. 3rd ed. New York: Oxford
University Press.
Li, Bozhong. 1998. Agricultural Development in Jiangnan, 1620–1850. New York:
St. Martin’s Press.
312 References
Li Daokui 李稻葵 and Guan Hanhui 管漢暉. n.d. Mingdai GDP shitan 明代 GDP
試探 (A tentative estimate of China’s GDP under the Ming). Unpublished.
Li, Huaiyin. 2005. Village Governance in North China: 1875–1936. Stanford, CA:
Stanford University Press.
Li, Huaiyin. 2013. Reinventing Modern China: Imagination and Authenticity in
Chinese Historical Writing. Honolulu: University of Hawaii Press.
Li Jingzhi 李靜之. 1984. “Shi lun Jiang-Feng-Yan zhongyuan dazhan” 試論蔣馮
閻中原大戰 (A preliminary study on the war among Jiang Jieshi, Feng Yux-
iang, and Yan Xishan in North China), Jindaishi yanjiu 近代史研究, 1: 219–
247.
Li Lifeng 李里峰. 2004. “Minguo wenguan kaoshi zhidu de yunzuo chengxiao”
民國文官考試制度的運作成效 (The operational effectiveness of the civil service
examination system during the Republican period), Lishi dang’an 歷史檔案, 1:
101–107.
Li Lifeng 李里峰. 2008. “Jingji de ‘tugai’ yu zhengzhi de ‘tugai’: Guanyu tudi
gaige lishi yiyi de zaisikao” 經濟的“土改”與政治的“土改”——關於土地改革歷
史意義的再思考 (The economic “land reform” and the political “land reform”:
Reflections on the historical significance of the land reform), Anhui shixue 安徽
史學, 2: 68–75.
Li Qiang 李強. 2008. Jinrong shijiao xia de “Kang-Qian shengshi”: Yi zhiqian tixi wei
kexin 金融視角下得“康乾盛世”──以制錢體系為核心 (The “Great prosperity of
Kangxi-Qianlong reigns” in fiscal perspective: A study centering on the statutory cur-
rency of copper cash). Hefei: Huangshan shushe.
Li Wang 王力. 2010. “Wanqing Man-Han guanxi yu Xin-Yi fenzhi” 晚清滿漢關係與
新伊分治 (The Manchu-Han relations and the administrative separation between
Xinjiang and Yili), Xiyu yanjiu 西域研究, 2: 13–19.
Li Weiguang 李煒光. 2000. Zhongguo caizheng shi lungao 中國財政史述論稿 (A
drafted treatise on the fiscal history of China). Beijing: Zhongguo caizheng jingji
chubanshe.
Li Xin 李新 and Li Zongyi 李宗一, eds. 1987. Zhonghua minguoshi, di er bian,
Beiyang junfa tongzhi shiqi 中華民國史, 第二編, 北洋軍閥統治時期 (A history of the
Republic of China, part 2, the period of the Northern Warlords regime), vol. 1.
Beijing: Zhonghua shuju.
Li Yongjun 李勇軍. 2011. “Qingmo minchu de Xizang jiansheng lun” 清末民初
的西藏建省論 (The debate on establishing Tibet as a province in the late Qing
and early Republican years), Zhongnan minzu daxue xuebao 中南民族大學學報,
31.5: 70–74.
Li Yunchang 李運昌. 1988. “Yi Ji-Re-Liang budui tingjin Dongbei” 憶冀熱遼部
隊挺進東北 (A recollection on the march of the troops from Hebei, Rehe, and
Liaoning into the Northeast). In Zhonggong zhongyang dangshi ziliao zhengji
weiyuanhui 中共中央黨史資料徵集委員會and Zhongguo renmin jiefangjun
Liaoshen zhanyi jinianguan jianguan weiyuanhui 中國人民解放軍遼瀋戰役紀念
館建館委員會, eds. Liaoshen juezhan 遼瀋決戰, part. 1 (Beijing: Renmin chu-
banshe), 167–183.
Li Zhiqin 李之勤. 1995. “18 shiji zhongye zhi 19 shiji zhongye Qingdai xibei
bianjiang de xunbian zhidu he xunbian luxian” 18 世紀中葉至 19 世紀中葉清
代西北邊疆的巡邊制度和巡邊路線 (The system and routes of border patrol
along the northwestern frontiers under the Qing from the mid eighteenth to
 References 313
the mid-nineteenth centuries), Zhongguo lishi dili luncong 中國歷史地理論叢,
3: 197–214.
Li Zhiting 李治亭. 2011. “Lun Qingdai bianjiang wenti yu guojia ‘dayitong’” 論清代
邊疆問題與國家“大一統”(The frontier issues of the Qing and the state’s “complete
unification”), Yunnan shifan daxue xuebao 雲南師範大學學報, 43.1: 1–11.
Liang Fangzhong 粱方仲. 2008. Zhongguo lidai hukou, tiandi, tianfu tongji 中國歷代
戶口、田地、田賦統計 (Statistics of households, population, land, and land taxes in
the Chinese dynasties). Beijing: Zhonghua shuju.
Liang Shuming 梁漱溟. 2006. Yiwang tanjiu lu 憶往談舊錄 (Recalling the past and
telling the old). Beijing: Jincheng chubanshe.
Lieberman, Victor. 2008. “The Qing Dynasty and Its Neighbors: Early Modern China
in World History,” Social Science History, 32.2: 281–304
Liu Cuirong 劉翠溶. 1967. “Shunzhi Kangxi nianjian de caizheng pingheng wenti
(1644–1722)” 順治康熙年間的財政平衡問題 (The issue of fiscal balance during
the reigns of Shunzhi and Kangxi), unpublished.
Liu Jingzhong 劉敬忠. 2004. Guominjun shigang 國民軍史綱 (An outline of the his-
tory of the Guomin Army). Beijing: Renmin chubanshe.
Liu, Kwang-Ching. 1970. “The Confucian as Patriot and Pragmatist: Li Hung-chang’s
Formative Years, 1823–1866,” Harvard Journal of Asiatic Studies, 30: 5–45.
Liu, Kwang-Ching. 1974. “The Limits of Regional Power in the Late Ch’ing Period:
A Reappraisal,” The Tsing Hua Journal of Chinese Studies, 10.2: 207–223.
Liu, Kwang-Ching. 1994. “The Beginnings of China’s Modernization.” In
Samuel Chua and Kwang-ching Liu, eds., Li Hung-Chang and China’s Early Mod-
ernization (New York: M. E. Sharpe), 3–14.
Liu, Kwang-Ching. and Smith, Richard. 1980. “The Military Challenge: The North-West
and the Coast.” In John K. Fairbank and Kwang-Ching Liu, eds., The Cambridge His-
tory of China (Cambridge: Cambridge University Press), vol. 11, 202–273.
Liu Ruizhong 劉瑞中. 1987. “Shiba shiji Zhongguo renjun guomin shouru guji jiqi
yu Yingguo de bijiao” 十八世紀中國人均國民收入估計及其與英國的比較 (An esti-
mate of national income per capita in eighteenth-century China and its comparison
with Britain), Zhongguo jingjishi yanjiu 中國經濟史研究, 3: 105–120.
Liu Tong 劉統. 2000. “Jiefang zhanzheng zhong Dongbei yezhanjun wuqi laiyuan
tantao: Jian yu Yang Kuisong xiansheng shangque” 解放戰爭中東北野戰軍武器來
源探討——兼與楊奎松先生商榷 (An inquiry into the sources of the weaponry of
the Northeastern field army in the Liberation War: An exchange with Mr. Yang Kui-
song), Dang de wenxian 黨的文獻, 4: 76–81.
Liu Wei 劉偉. 2003. Wanqing dufu zhengzhi: Zhongyang yu difang guanxi yanjiu
晚清督撫政治：中央與地方關係研究 (The politics of governor-generals and pro-
vincial governors in the late Qing period: A study of the relationship between cen-
tral and local authorities). Wuhan: Hubei jiaoyu chubanshe.
Liu, Xiaoyuan. 2010. Recast All Under Heaven: Revolution, War, Diplomacy, and
Frontier China in the 20th Century. New York: Continuum.
Liu Yuewu 柳岳武. 2009. Chuantong yu bianqian: Kong-Yong-Qian zhi Qingting yu
fanbu shuguo guanxi yanjiu 傳統與變遷：康雍乾之清廷與藩部屬國關係研究 (A
study on the relationship between the Qing court and its frontiers and tributary states
during the reigns of Kangxi, Yongzheng, and Qianglong). Chongqing: Bashu shushe.
Liu Zenghe 劉增合. 2014. “Guang-Xuan zhijiao qingli caizheng qianxi de sheju yu pai-
guan” 光宣之交清理財政前夕的設局與派官 (Office installation and official
314 References
appointment on the eve of financial investigation in the transitional years between the
Guangxu and Xuantong reigns), Guangdong shehui kexue 廣東社會科學, 2: 111–120.
Liu Zenghe 劉增合. 2017. “Zuo Zongtang xinzheng chouxiang yu Qing ting zhanshi
caizheng tiaokong” 左宗棠西征籌餉與清廷戰事財政調控 (Zuo Zongtang’s efforts
to raise military funds for the Western Expedition and Qing court’s wartime fiscal
control), Jindaishi yanjiu 近代史研究, 2: 84–99.
Liu Ziyang 劉子揚. 1988. Qingdai difang guanzhi kao 清代地方官制考 (A study of
local government positions under the Qing). Beijing: Zijincheng chubanshe.
Long Shengyun 龍盛運. 1990. Xiangjun shigao 湘軍史稿 (History of the Hunan
Army). Chengdu: Sichuan renmin chubanshe.
Lu Shichuan 盧世川. 1987. “Shandong kangri genjudi caizheng de chuangjian he
fazhan” 山東抗日根據地財政的創建和發展 (The establishment and development of
the fiscal system in the Shandong base area for the War of Resistance against Japan).
In Caizhengbu caizheng kexue yanjiusuo 財政部財政科學研究所 ed., Kangri gen-
judi de caizheng jingji 抗日根據地的財政經濟 (Finance and economy in the base areas
for the War of Resistance against Japan) (Beijing: Zhongguo caizheng jingji chu-
banshe), 233–246.
Luo Ergang 羅爾綱. 1947. Taiping tianguo shigang 太平天國史綱 (An outline of the
history of Taiping Heavenly Kingdom). Shanghai: Shangwu yinshuguan.
Luo Ergang 羅爾綱. 1958. “Taiping tianguo geming qian de renkou yapo wenti” (The
issue of population pressure before the Taiping revolution) 太平天國革命前的人口壓
迫問題. In Bao Zhunpeng 包遵彭, Wu Xiangxiang 吳相湘 and Li Dingyi 李定一, eds.,
Zhongguo jindaishi luncong 中國近代史論叢, 2.2 (Taipei: Zhengzhong shuju), 16–88.
Luo Pinghan 羅平漢. 2005. “yijiusiqi nian xiabannian jiefangqu tugai yundong zhong de
‘zuo’ qing cuowu jiqi jiuzheng” 一九四七年下半年解放區土改運動中的“左”傾錯誤
及其糾正 (The leftist mistakes and their correction in the land reform movement in the
liberated areas in the second half of 1947), Zhonggong dangshi yanjiu 中共黨史研究, 2:
44–54.
Luo Yudong 羅玉東. 1970. Zhongguo lijin shi 中國釐金史 (A history of lijin in
China). Taipei: Wenhai chubanshe.
Luo Zhitian 羅志田. 2000. “Nanbei xinjiu yu Beifa chenggong de zai quanshi” 南北新舊
與北伐成功的再詮釋 (The new and the old in the south and the north:
A reinterpretation of the success of the Northern Expedition), Kaifang shidai 開放時
代, 9: 47–62.
Ma Jinhua 馬金華. 1911. Waizhai yu wanqing zhengju 外債與晚清政局 (Foreign debt
and late Qing politics). Beijing: Shehui kexue wenxian chubanshe.
Ma Ruheng 馬汝珩 and Ma, Dazheng 馬大正 eds. 1994. Qingdai de bianjiang
zhengce 清代的邊疆政策 (The frontier policies of the Qing). Beijing: Zhongguo
shehui kexue chubanshe.
Ma Xuanwei 馬宣偉 and Wen Xianmei 溫賢美. 1986. Chuanjun chu Chuan kangzhan
jishi 川軍出川抗戰紀事 (A documentary on the Sichuan Army’s engagement in the
War of Resistance outside the Sichuan province). Chengdu: Sichuansheng shehui kex-
ueyuan chubanshe.
MacKinnon, Stephen R. 1973. “The Peiyang Army, Yuan Shih-k’ai, and the Origins
of Modern Chinese Warlordism,” Journal of Asian Studies, 32.3: 405–423.
MacKinnon, Stephen R. 2009. Wuhan, 1938: War, Refugees, and the Making of
Modern China. Berkeley, CA: University of California Press.
Maddison, Angus. 1998. Chinese Economic Performance in the Long Run. Paris:
OECD
 References 315
Maddison, Angus. 2001. The World Economy: A Millennial Perspective. Paris:
OECD.
Mann, Michael. 1980. “State and Society, 1130–1815: An Analysis of English State
Finances.” In Maurice Zeitlin, ed., Political Power and Society Theory: A Research
Annual, vol. I. Greenwich, CT: JAI Press.
Mann, Michael. 1986. The Sources of Social Power, Vol. I: A History of Power from the
beginning to A.D. 1760. Cambridge: Cambridge University Press.
Mann, Susan. 1987. Local Merchants and the Chinese Bureaucracy, 1750–1950. Stan-
ford, CA: Stanford University Press.
Mao Haijian 茅海建. 2005. Tianchao de bengkui: Yapian zhanzheng de zai yanjiu
天朝的崩潰：鴉片戰爭的再研究 (The collapse of the celestial dynasty: The Opium
War revisited). Beijing: Sanlian shudian.
Marx, Karl and Frederick Engels. 1969. “Manifesto of the Communist Party.” In
Karl Marx and Frederick Engels eds., Selected Works (Moscow: Progress Publishers), vol.
1, 98–137.
McCord, Edward. 1993. The Power of the Gun: The Emergence of Modern Chinese
Warlordism. Berkeley, CA: University of California Press.
McCord, Edward. 2014. Military Force and Elite Power in the Formation of Modern
China. New York: Routledge.
McCormack, Gavan. 1977. Chang Tso-Lin in Northeast China, 1911–1928. Stanford,
CA: Stanford University Press.
Mehta, U. Singh. 1999. Liberalism and Empire: A Study in Nineteenth-Century Brit-
ish Liberal Thought. Chicago, IL: Chicago University Press.
Miao Zhou 妙舟. 1988. Mengzang fojiao shi 蒙藏佛教史 (A history of Mogolian and
Tibetan Buddhism). Taipei: Wenhai chubanshe.
Millward, James P. 1998. Beyond the Pass: Economy, Ethnicity, and Empire in Qing
Central Asia, 1759–1864. Stanford, CA: Stanford University Press.
Min Zongdian 閔宗殿. 1999. “Cong fangzhi jizai kan Ming Qing shiqi woguo shui-
dao de fenbu” 從方誌記載看明清時期我國水稻的分佈 (Distribution of rice cultiva-
tion in our country in the Ming and the Qing as documented in local gazetteers).
Gujin nongye 古今農業, 1: 35–48.
Min Zongdian 閔宗殿. 2003. “Ming Qing shiqi Zhongguo nanfang daotian duoshu
zhongzhi de fazhan” 明清時期中國南方稻田多熟種植的發展 (The evolution of the
multi-cropping of rice in South China during the Ming and the Qing). Zhongguo
nongshi 中國農史, 3: 10–14.
Mitter, Rana. 2013. Forgotten Ally: China’s World War II, 1937–1945. Boston, MA:
Mariner Books.
Mo Jijie 莫濟傑and Chen Fulin 陳福霖, eds. 1991. Xin Guixi shi 新桂系史 (History of
the New Guixi Clique). Guilin: Guangxi renmin chubanshe.
Morgan, Edmunds. 1988. Inventing the People: The Rise of Popular Sovereignty in Eng-
land and America. New York: W. W. Norton & Co.
Mosca, Matthew. 2013. From Frontier Policy to Foreign Policy: The Question of India and
the Transformation of Geopolitics in Qing China. Stanford, CA: Stanford University
Press.
Muthu, Sankar. 2003. Enlightenment against Empire. Princeton, NJ: Princeton Uni-
versity Press.
Nathan, Andrew J. 1976. Beijing Politics, 1918–1923: Factionalism and the Failure of
Constitutionalism. Berkeley, CA: University of California Press.
316 References
Ni Yuping 倪玉平. 2017. Cong guojia caizheng dao caizheng guojia: Qingchao Xian-
Tong nianjian de caizheng yu shehui 從國家財政到財政國家——清朝咸同年間的財
政與社會 (From state finance to financial state: Finance and society during the Xian-
feng and Tongzhi reigns of the Qing dynasty). Beijing: Kexue chubanshe.
O’Brien, Patrick K. 2002. “Fiscal Exceptionalism: Great Britain and its European
Rivals from Civil War to Triumph at Trafalgar and Waterloo.” In Donald Winch
and Patrick K. O’Brien, eds., The Political Economy of British Historical Experience,
1688–1914 (New York: Oxford University Press), 245–265.
Opello, Jr, Walter C. 2004. The Nation-State and Global Order: A Historical Introduc-
tion to Contemporary Politics, 2nd ed. Boulder, CO: Lynne Rienner.
Ouyang Junxi 歐陽軍喜. 2011. “Jiyi yu lishi: Sun Zhongshan dui Xinhai geming shi
de jiangou jiqi yingxiang” 記憶與歷史：孫中山對辛亥革命史的建構及其影響
(Memory and history: Sun Yat-sen’s inventing of the history of the 1911 Revolution
and its consequences), Renwen zazhi 人文雜誌, 5: 123–128.
Pagden, Anthony. 1995. Lords of All the World: Ideologies of Empire in Spain, Britain
and France, c. 1500 – C. 1800. New Haven, CT: Yale University Press.
Pagden, Anthony. 2003. Peoples and Empires. New York: The Modern Library.
Paine, S. C. M. 1996. Imperial Rivals: China, Russia, and Their Disputed Frontier.
Armonk, NY: M. E. Sharpe.
Pang Xianzhi 逄先知 and Jin Congji 金衝及, eds. 2011. Mao Zedong zhuan 毛澤東傳,
vol. 1–6. Beijing: Zhongyang wenxian chubanshe.
Peng Xinwei 彭信威. 2007. Zhongguo huobi shi 中國貨幣史 (A history of currencies in
China). Shanghai: Shanghai renmin chubanshe.
Peng Zeyi 彭澤益. 1983. Shijiu shiji houbanqi de Zhongguo caizheng yu jingji 十九世
紀後半期的中國財政與經濟 (Finance and economy in late nineteenth-century
China). Beijing: Renmin chubanshe.
Peng Zeyi 彭澤益. 1990. “Qingdai caizheng guanli tizhi yu shouzhi jiegou” 清代財政
管理體制與收支結構 (Fiscal administration and the structure of revenues and
expenditures in the Qing). Zhongguo shehui kexueyuan yanjiushengyuan xuebao 中國
社會科學院研究生院學報, 2: 48–59.
Pepper, Suzanne. 1978. Civil War in China: The Political Struggle, 1945–1949. Los
Angeles, CA: University of California Press.
Perdue, Peter. 1996. “Military Mobilization in Seventeenth and Eighteenth-Century
China, Russia, and Mongolia,” Modern Asian Studies, 30.4: 757–793.
Perdue, Peter. 2005. China Marches West: The Qing Conquest of Central Eurasia.
Cambridge, MA: Harvard University Press.
Pong, David. 1994. Shen Pao chen and China’s Modernization in the Nineteenth Cen-
tury. Cambridge: Cambridge University Press.
Prazniak, Roxann. 1999. Of Camel Kings and Other Things: Rural Rebels against
Modernity in Late Imperial China. Lanham, MD: Rowman & Littlefield.
Pye, Lucian. 1971. Warlord Politics: Conflict and Coalition in the Modernization of
Republican China. New York: Praeger.
Qimudedaoerji 奇木德道爾吉. 1998. “1640 nian yihou de Qingchao yu Kaerka guanxi”
1640 年以後的清朝與喀爾喀關係 (The relationship between the post–1640 Qing dyn-
asty and the Khalkha Mongols), Neimenggu daxue xuebao 內蒙古大學學報, 4: 12–20.
Qin Qingjun 秦慶鈞. 1982. “Beifa zhanzheng shiqi de Guangdong sheng caizheng” 北伐
戰爭時期的廣東省財政 (Finance in Guangdong province during the Northern Exped-
ition period). In Zhongguo renmin zhengzhi xieshang huiyi Guangdongsheng Guangz-
houshi weiyuanhui wenshi ziliao yanjiu weiyuanhui 中國人民政治協商會議廣東省廣州
 References 317
市委員會文史資料研究委員會, ed., Guangzhou wenshi ziliao 廣州文史資料, vol. 27:
161–191.
Rankin, Mary B. 1986. Elite Activism and Political Transformation in China: Zhejiang
Province, 1865–1911. Stanford: Stanford University Press.
Rasler, Karen and Thompson, William. 1989. War and State Making: The Shaping of
the Global Powers. Boston, MA: Unwin Hyman.
Rawlinson, John. 1967. China’s Struggle for Naval Development, 1839–1895. Cam-
bridge, MA: Harvard University Press.
Rawski, Evelyn S. 1996. “Reenvisioning the Qing: The Significance of the Qing
Period in Chinese History,” The Journal of Asian Studies, 55.4: 829–850.
Rawski, Evelyn S. 1998. The Last Emperors: A Social History of Qing Imperial Institu-
tions. Berkeley, CA: University of California Press.
Rawski, Evelyn S. 2004. “The Qing Formation and the Early-Modern Period.” In
Lynn Struve, ed., The Qing Formation in World-Historical Time. Cambridge, MA:
Harvard University Press, 207–241.
Rawski, Evelyn S. 2010. “Chinese Strategy and Security Issues in Historical Perspec-
tive.” In Brantly Womack, ed., China’s Rise in Historical Perspective Lanham, MD:
Rowman & Littlefield, 63–87.
Reardon-Anderson, James. 2000. “Land Use and Society in Manchuria and Inner
Mongolia during the Qing Dynasty,” Environmental History, 5.4: 503–530.
Remick, Elizabeth J. 2004. Building Local States: China During the Republican and
Post-Mao Eras. Cambridge, MA: Harvard University Asia Center.
Rhoads, Edward J. 2000. Manchus and Han: Ethnic Relations and Political Power in
Late Qing and Early Republican China, 1861–1928. Seattle, WI: University of
Washington Press.
Roberts, Michael. 1967. “The Military Revolution, 1560–1660.” In Michael Roberts
ed., Essays in Swedish History (London: Weidenfeld & Nicolson), 195–225.
Roeder, Philip G. 2007. Where Nation-States Come From: Institutional Change in the
Age of Nationalism. Princeton, NJ: Princeton University Press.
Rong Hong 容閎. 1985. Xixue dong jian ji 西學東漸記 (The introduction of Western
learnings to the East). Changsha: Hunan renmin chubanshe.
Rowe, William T. 2001. Saving the World: Chen Hongmou and Elite Consciousness in
Eighteenth-Century China. Stanford: Stanford University Press.
Rubin, Alfred. 1968. “The Position of Tibet in International Law,” The China Quar-
terly, 35: 110–154.
Rudolph, Jennifer and Szonyi, Michael, eds. 2018. The China Questions: Critical
Insights into a Rising Power. Cambridge, MA: Harvard University Press.
Said, Edward. 1979. Orientalism. New York: Random House.
Scammell, G. V. 1989. The First Imperial Age: European Overseas Expansion, c. 1400–
1715. London: Routledge.
Selden, Mark. 1971. The Yenan Way in Revolutionary China. Cambridge, MA: Har-
vard University Press.
Shakabpa, Tsepon. 1967. Tibet: A Political History. New Haven, CT: Yale University
Press.
Shang Bing 桑兵. 2007. Wanqing xuetang xuesheng yu shehui bianqian 晚清學堂學生
與社會變遷 (New-style school students and social change in the late Qing period).
Guilin: Guangxi shifan daxue chubanshe.
318 References
Shang Hongkui 商鴻逵. 1982. “Lun Qingdai de zun Kong he chongfeng Lamajiao”
論清代的尊孔和崇奉喇嘛教 (On the elevation of Confucius and the promotion of
Lamaism during the Qing). Shehui kexue jikan 社會科學輯刊, 5: 107–113.
Shaw, Stanford. 1976. History of the Ottoman Empire and Modern Turkey. Vol. I:
Empire of the Gazis: The Rise and Decline of the Ottoman Empire, 1280–1808. Cam-
bridge: Cambridge University Press.
Shen Jiarong 沈嘉榮, ed. 1993. Jiangsu shigang, jindai juan 江蘇史綱, 近代卷 (An
outline of the history of Jiangsu, the volume of the modern period). Nanjing: Jiangsu
guji chubanshe.
Sheridan, James. 1966. Chinese Warlord: The Career of Feng Yu-hsiang. Stanford:
Stanford University Press.
Shi Zhihong 史志宏. 1989. “Qingdai qianqi de gendi mianji ji liangshi chanliang guji” 清
代前期的耕地面積及糧食產量估計 (An estimate of arable land and food production
in the first half of the Qing), Zhongguo jingjishi yanjiu 中國經濟史研究, 2: 47–62.
Shi Zhihong 史志宏. 2009. Qingdai hubu yinku shouzhi he kucun tongji 清代戶部銀庫
收支和庫存統計 (Statistics of the revenues and expenditure and cash reserves at the
Board of Revenue of the Qing dynasty). Fuzhou: Fujian renmin chubanshe.
Shi Zhihong 史志宏. 2011. “Shijiu shiji shangbanqi de Zhongguo gengdi mianji zai guji”
十九世紀上半期的中國耕地面積再估計 (A second estimate of China’s arable land in
the first half of the nineteenth century), Zhongguo jingjishi yanjiu 中國經濟史研究, 4:
85–97.
Shi Zhihong 史志宏 and Xu Yi 徐毅. 2008. Wanqing caizheng: 1851–1894晚清財政:
1851–1894 (Late Qing finance: 1851–1894). Shanghai: Shanghai caijing daxue
chubanshe.
Shi Zhongquan 石仲泉. 2012. “Yi Qiao Mu tongzhi tan Yan’an zhengfeng: Jian-
lun Yan’an zhengfeng yundong” 憶喬木談延安整風——兼論延安整風運動
(Recollection on Comrade Qian Mu’s conversations on Yan’an Rectification,
plus an analysis of the Rectification Campaign in Yan’an), Zhonggong dangshi
yanjiu 中共黨史研究, 5: 10–23.
Song Liangxi 宋良曦. 1998. “Qingdai Zhongguo yanshang de shehui dingwei” 清代
中國鹽商的社會定位 (The social standing of Chinese salt merchants during the
Qing), Yanye shi yanjiu 鹽業史研究, 4: 24–33.
Storrs, Christopher. 2009. “Introduction: The Fiscal-Military State in the ‘Long’ Eight-
eenth Century.” In Christopher Storrs, ed., The Fiscal-Military State in Eighteenth-
Century Europe: Essays in Honor of P. G. M. Dickson (Burlington, VT: Ashgate Publish-
ing Co.): 1–22.
Strauss, Julia. 1998. Strong Institutions in Weak Politics: State Building in Republican
China, 1927–1940. Oxford: Clarendon Press.
Strayer, Joseph R. 1970. On the Medieval Origins of the Modern State. Princeton, NJ:
Princeton University Press.
Streusand, Douglas E. 2010. Islamic Gunpowder Empires: Ottomans, Safavids, and
Mughals. Boulder, CO: Westview Press.
Su Quanyou 蘇全有. 2010. “Youguan Qingmo caizheng wenti de liangdian sikao”
有關清末財政問題的兩點思考 (A few thoughts on the financial issues in the last
years of the Qing dynasty), Anhui shixue 安徽史學, 4: 11–18.
Su Quanyou 蘇全有. and Jing Dongsheng 景東昇. 2004. “Lun Yuan Shikai de chou
Ri zhengce ji shijian” 論袁世凱的仇日政策及實踐 (OnYuan Shikai’s policy of hos-
tility against Japan and its implementation), Lishi jiaoxue 歷史教學, 5: 22–26.
 References 319
Suleski, Ronald. 2002. Civil Government in Warlord China: Tradition, Modernization
and Manchuria. New York: Peter Lang.
Sun Hongnian 孫宏年. 2004. “20 shiji chu Yingguo dui Zhongguo Xizang de qinlue
yu Xizang jiansheng wenti yanjiu” 20 世紀初英國對中國西藏的侵略與西藏建省問題
研究 (A study on the issues of Britain’s aggression of Tibet and the establishment of
Tibet as a province in the early twentieth century), Xizang yanjiu 西藏研究, 3: 15–20.
Sun Hongnian 孫宏年. 2006. “Qingdai fanshu guannian de bianhua yu Zhongguo
jiangtu de bianqian” 清代藩屬觀念的變化與中國疆土的變遷 (Adjustments in the
Qing state’s views of frontiers and tributary states and changes in China’s territor-
ies), Qingshi yanjiu 清史研究, 4: 17–27.
Sun Hongnian 孫宏年. 2011. “Qingdai Zhongguo yu linguo ‘jiangjie guan’ de pengz-
huang, jiaorong chuyi” 清代中國與鄰國“疆界觀”的碰撞、交融芻議 (A preliminary
study on the conflict and mutual penetration between Qing China and the neigh-
boring states in their respective views on territories and boundaries), Zhongguo bian-
jiang shidi yanjiu 中國邊疆史地研究, 21.4: 12–22.
Svarverud, Rune. 2011. “Re-constructing East Asia: International Law as Inter-Cultural
Process in Late Qing China,” Inter-Asia Cultural Studies, 12.2: 306–318.
Tan Qixiang 譚其驤, ed. 1982. Zhongguo lishi ditu ji 中國歷史地圖集 (Historical
atlas of Chinese history), vols. 1–8. Beijing: Zhongguo ditu chubanshe.
Tan Qixiang 譚其驤. 1991. Jianming Zhongguo lishi dituji簡明中國歷史地圖集 (A
concise atlas of Chinese history). Beijing: Zhongguo ditu chubanshe.
Tan Zhaoyi 譚肇毅. 2009. “Xin Guixi lun” 新桂系論 (A new interpretation of the
new Guangxi clique), Guangxi shehui kexue 廣西社會科學, 6: 14–19.
Tan Zhaoyi 譚肇毅. 2010. “Xin Guixi de ‘sanzi zhengce’” 新桂系的“三自政策”(The
new Guangxi clique’s “three-self policy”), Guangxi difangzhi 廣西地方誌, 1: 47–51.
Tanner, Harold. 2013. The Battle for Manchuria and the Fate of China: Siping, 1946.
Bloomington, IN: Indiana University Press.
Tanner, Harold. 2015. Where Chiang Kai-shek Lost China: The Liao-Shen Campaign,
1948. Bloomington, IN: Indiana University Press.
Taylor, Jay. 2009. The Generalissimo: Chiang Kai-shek and the Struggle for Modern
China. Cambridge, MA: Harvard University Press.
Thaxton, Ralph. 1983. China Turned Rightside Up: Revolutionary Legitimacy in the
Peasant World. New Haven, CT: Yale University Press.
Thaxton, Ralph. 1997. Salt of the Earth: The Political Origins of Peasant Protest and
Communist Revolution in China. Berkeley, CA: University of California Press.
Theobald, Ulrich. 2013. War Finance and Logistics in Late Imperial China: A Study
of the Second Jinchuan Campaign (1771–1776). Leiden: Brill.
Thornton, Patricia. 2007. Disciplining the State: Virtue, Violence, and State-Making in
Modern China. Cambridge, MA: Harvard University Asia Center.
Tien, Hung-Mao. 1972. Government and Politics in Kuomintang China, 1927–1937.
Stanford: Stanford University Press.
Tilly, Charles. 1975. “Reflections on the History of European State-Making.” In
Charles Tilly, ed., The Formation of National States in Western Europe (Princeton,
NJ: Princeton University Press), 3–83.
Tilly, Charles. 1985. “War Making and State Making as Organized Crime.” In
Peter Evans, Dietrich Rueschemeyer and Theda Skocpol, eds., Bringing the State
Back In (Cambridge: Cambridge University Press), 169–191.
Tilly, Charles. 1990. Coercion, Capital, and European States, AD 990–1992. Malden,
MA: Blackwell Publishers Inc.
320 References
Torgovnick, Marianna. 1990. Gone Primitive: Savage Intellects, Modern Lives. Chi-
cago, IL: University of Chicago Press.
Truman, Harry S. 1986. Memoirs of Harry S. Truman, vol. 1, “1945, Year of Deci-
sions.” Cambridge, MA: Da Capo Press.
Tuck, Richard. 2015. The Sleeping Sovereign: The Invention of Modern Democracy.
Cambridge: Cambridge University Press.
van Creveld, Martin. 1999. The Rise and Decline of the State. Cambridge: Cambridge
University Press.
van de Ven, Hans J. 1996. “Public Finance and the Rise of Warlordism,” Modern
Asian Studies, 30.4: 829–868.
van Schaik, Sam. 2011. Tibet: A History. New Haven, CT: Yale University Press.
Waldron, Arthur. 1995. From War to Nationalism: China’s Turning Point, 1924–
1925. Berkeley, CA: University of California Press.
Waley-Cohen, Joanna. 2004. “The New Qing History,” Radical History Review, 88:
193–206.
Wang Bingming 汪炳明. 1990. “Qingmo xinzheng yu beibu bianjiang kaifa” 清末新政
與北部邊疆開發 (The New Policies and the development of the northern borders in
the last years of the Qing), in Ma Ruheng 馬汝珩 and Ma Dazheng 馬大正, eds.,
Qingdai bianjiang kaifa yanjiu 清代邊疆開發研究, Beijing: Zhongguo shehui kexue
chubanshe.
Wang Chaoguang 汪朝光. 2005. “Quanmian neizhan chuqi Guomindang junshi shili
yuanyin zhi bianxi” 全面內戰初期國民黨軍事失利原因之辨析 (An analysis of the
reasons behind the Guomindang’s military losses in the initial stage of the full-scale
civil war), Minguo dang’an 民國檔案, 1: 97–105.
Wang Dongping 王東平. 2005. Guanyu Qingdai Huijiang boke zhidu de jige wenti
關於清代回疆伯克制度的幾個問題 (Several issues concerning the Beg system in the
Muslim areas under the Qing), Minzu yanjiu 民族研究, 1: 72–79.
Wang Ermin 王爾敏. 1987. Huaijun zhi 淮軍志 (A study of the Huai army). Beijing:
Zhonghua shuju.
Wang Jingyu 汪敬虞, ed. 2000. Zhongguo jindai jingjishi 中國近代經濟史 (Economic
history of modern China), 1895–1927, vols. 1–3. Beijing: Renmin chubanshe.
Wang Kaixi 王開璽. 2006. “Qingmo Man-Han guanliao yu Man-Han minzu yishi
jianlun” 清末滿漢官僚與滿漢民族意識簡論 (A brief discussion on the Manchu and
Han bureaucrats and their ethnic consciousness during the late Qing period), Shehui
kexue jikan 社會科學輯刊, 6: 168–174.
Wang Liqi 王禮琦 and Li Bingjun 李炳俊. 1981. “Tudi geming shiqi geming genjudi
de caizheng (zhong)” 土地革命時期革命根據地的財政（中）(Finance of the revolu-
tionary base areas during the agrarian revolution, part 2), Caizheng 財政, 1: 29–31.
Wang Miaosheng 王淼生, ed. 1997. Zhongguo renmin jiefangjun quanguo jiefang
zhanzheng shi 中國人民解放軍全國解放戰爭史 (A history of the nationwide Lib-
eration War of the PLA), vol. 4. Beijing: Junshi kexue chubanshe.
Wang Nianyong 王年詠. 1994. “Jindai Zhongguo de zhanzheng peikuan zhongzhi”
近代中國的戰爭賠款總值 (The total amount of China’s war reparations in the
modern era), Lishi yanjiu 歷史研究, 5: 175–177.
Wang Qisheng 王奇生. 2003. Dangyuan, dangquan yu dangzheng: 1924–1949 nian
Zhongguo guomindang de zuzhi xingtai 黨員、黨權與黨爭──1924–1949 年中國國
民黨的組織形態 (Party members, party authority, and internal strife within the
party: The organizational formation of the Chinese Nationalist Party from 1924 to
1949). Shanghai: Shanghai shudian chubanshe.
 References 321
Wang Rigen 王日根. 2000. “Lun Kangxi de xushang sixiang yu shijian” 論康熙的恤商
思想與實踐 (On Kangxi’s thinking and practice of cultivating commerce), Yunnan
caimao xueyuan xuebao 雲南財貿學院學報, 16.2: 78–85.
Wang Sizhi 王思治, ed. 2003. Qingchao tongshi: Kangxi chao fenjuan 清朝通史
——康熙朝分卷 (A general history of the Qing dynasty: Volume on the Kangxi
reign), 2 vols. Beijing: Zijincheng chubanshe.
Wang Xi’en 王希恩. 2011. “Xinhai geming zhong de Man-Han maodun jiqi yingxiang”
辛亥革命中得滿漢矛盾及其影響 (The Manchu-Han conflict and its impact during the
Revolution of 1911), Xinan minzu daxue xuebao 西南民族大學學報, 10: 20–25.
Wang Xutian 王續添. 2000. “Difangzhuyi yu minguo shehui” 地方主義與民國社會
(Localism and society during the Republican period), Jiaoxue yu yanjiu 教學與研究,
2: 57–63.
Wang, Yeh-chien. 1973. Land Taxation in Imperial China, 1750–1911. Cambridge,
MA: Harvard University Press.
Wang Yinhuan 王印煥. 2005. “Minguo zhengfu gongjiao renyuan shenghuo zhuang-
kuang de yanbian” 民國政府公教人員生活狀況的演變 (Changes in the living con-
dition of the personnel of public service and education under the Nationalist
government), Beijing keji daxue xuebao (shehui kexue ban) 北京科技大學學報（社
會科學版）, 21.1: 66–82.
Wang Yugui 王玉貴. 1996. “Shi xi xin Guixi jiuju minguo zhengtan de yuanyin” 試析
新桂系久據民國政壇的原因 (A preliminary analysis of the reasons behind the pro-
longed activism of the new Guangxi clique in Republican politics), Guangxi shehui
kexue 廣西社會科學, 3: 75–80.
Wang Yunsheng 王芸生. 2005. Liushi nian lai Zhongguo yu Riben 六十年來中國與日
本 (China and Japan in the past sixty years). Beijing: Sanlian shudian.
Wang Yuru 王玉茹 and Yan Hongzhong 燕紅忠. 2007. Shijie shichang jiage biandong
yu jindai Zhongguo chanye jiegou moshi yanjiu 世界市場價格變動與近代中國產業結
構模式研究 (A study on changes in world market prices and economic structural pat-
terns in modern China). Beijing: Renmin chubanshe.
Wedemeyer, Albert C. 1958. Wedemeyer Reports! New York: Henry Holt & Co.
Wei Hongyun 魏宏運, Xing Guang 星光 and Fu Shangwen 傅尚文. 1984. Kangri
zhanzheng shiqi Jin-Cha-Ji bianqu caizheng jingjishi ziliao xuanbian 抗日戰爭時期
晉察冀邊區財政經濟史資料選編 (Selected Source Material on the financial and eco-
nomic history of the Shanxi-Chahar-Hebei boarder area during the War of Resistance
against Japan), vol. 4. Tianjin: Nankai daxue chubanshe.
Westad, Odd A. 2003. Decisive Encounters: The Chinese Civil War, 1946–1950. Stan-
ford, CA: Stanford University Press.
Wilbur, C. Martin. 1968. “Military Separatism and the Process of Reunification under
the Nationalist Regime, 1922–1937.” In Ping-ti Ho and Tang Tsou, eds., China in
Crisis: China’s Heritage and the Communist Political System (Chicago, IL: Univer-
sity of Chicago Press, 1 (book one)), 203–263.
Wilbur, C. Martin. 1976. Sun Yat-sen: Frustrated Patriot. New York: Columbia Univer-
sity Press.
Wilson, Peter. 2009. “Prussia as a Fiscal-Military State, 1640–1806.” In
Christopher Storrs, ed., The Fiscal-Military State in Eighteenth-Century Europe: Essays
in Honor of P. G. M. Dickson (Burlington, VT: Ashgate Publishing Co.): 95–124.
Wong, R. Bin. 1997. China Transformed: Historical Change and the Limits of Euro-
pean Experience. Ithaca, NY: Cornell University Press.
322 References
Wou, Odoric Y. K. 1978. Militarism in Modern China: The Career of Wu Pei-fu,
1916–39 (Folkestone: Dawson, 1978).
Wou, Odoric Y. K. 1994. Mobilizing the Masses: Building Revolution in Henan. Stan-
ford, CA: Stanford University Press.
Wright, Mary C. 1957. The Last Stand of Chinese Conservatism: The T’ung-chih Restor-
ation, 1862–1874. Stanford: Stanford University Press.
Wright, Mary C. ed. 1968. China in Revolution: The First Phase, 1900–1913. New
Haven, CT: Yale University Press.
Wu Jingping 吳景平. 1992. Song Ziwen zhuan 宋子文傳 (Biography of Song Ziwen).
Fuzhou: Fujian renmin chubanshe.
Wu Jingping 吳景平. 1998. Song Ziwen zhengzhi shengya biannian 宋子文政治生涯編
年 (A chronology of Song Ziwen’s political career). Fuzhou: Fujian renmin chubanshe.
Xiang Ruihua 相瑞花. 1999. “Shixi jindai Zhongguo de zhanzheng peikuan” 試析近
代中國的戰爭賠款(A preliminary analysis of war indemnities in modern China),
Qinghai shifan daxue xuebao 青海師範大學學報, 1: 76–82.
Xiao Yishan 蕭一山. 1967. Qingdai tongshi 清代通史 (A general history of the Qing
dynasty). Taipei: Shangwu yinshuguan.
Xie Benshu 謝本書 and Niu Hongbin 牛鴻賓. 1990. Jiang Jieshi he xinan defang shili-
pai 蔣介石和西南地方實力派 (Chiang Kai-shek and the southwestern local forces).
Zhengzhou: Henan renmin chubanshe.
Xing Quancheng 星全成 and Chen Beiping 陳柏萍. 2010. Zangchuan fojiao sida huoho
xitong yu Qingchao zhili Meng Zang fanglue 藏傳佛教四大活佛系統與清朝治理蒙藏
方略 (The four “Living Buddha” systems of Tibetan Buddhism and the Qing dynasty’s
strategies to govern Mongolia and Tibet). Xining: Qinghai renmin chubanshe.
Xu Daofu 許道夫. 1983. Zhongguo jindai nongye shengchan ji maoyi tongji ziliao 中國
近代農業生產及貿易統計資料 (Statistics on agricultural production and marketing
in modern China). Shanghai: Shanghai renmin chubanshe.
Xu Dixin 許滌新 and Wu Chengming 吳承明. 2003a. Zhongguo zibenzhuyi fazhan shi:
Diyijuan, Zhongguo zibenzhuyi de mengya 中國資本主義發展史：第一卷, 中國資本
主義的萌芽 (History of the growth of capitalism in China, vol. 1, the germination of
Chinese capitalism). Beijing: Renmin chubanshe.
Xu Dixin 許滌新 and Wu Chengming 吳承明 2003b. Zhongguo zibenzhuyi fazhan shi:
Dierjuan, Jiu minzhuzhuyi geming shiqi de Zhongguo zibenzhuyi 中國資本主義發展
史：第二卷, 舊民主主義革命時期的中國資本主義 (History of the growth of capital-
ism in China, vol. 2, Chinese capitalism during the period of Old Democratic Revolu-
tion). Beijing: Renmin chubanshe.
Xu Tan 許檀 and Jing Junjian 經君健. 1990. “Qingdai qianqi shangshui wenti xintan”
清代前期商稅問題新探 (An reexamination of commercial taxes in the first half of
the Qing). Zhongguo jingjishi yanjiu 中國經濟史研究, 2: 87–100.
Xu Yi 許毅. 1982. Zhongyang geming genjudi caizheng jingji shi changbian 中央革命
根據地財政經濟史長編 (A history of finance and economy in the Central Revolution-
ary Base Area, the extended edition), vol. 2. Beijing: Renmin chubanshe.
Xu Yisheng 徐義生. 1962. Zhongguo jindai waizhai shi tongji ziliao 中國近代外債史統計
資料 (Statistics on the history of foreign debt in modern China). Beijing: Zhonghua
shuju.
Xue Fucheng 薛福成. 1987. Xue Fucheng xuanji 薛福成選集 (Selected works of Xue
Fucheng). Shanghai: Shanghai renmin chubanshe.
Yack, Bernard. 2001. “Popular Sovereignty and Nationalism,” Political Theory, 29.4:
517–536.
 References 323
Yang Jing 楊菁. 2002. “Shixi zhongyang geming genjudi de caizheng shouru
zhengce” 試析中央革命根據地的財政收入政策 (A preliminary analysis of the finan-
cial and revenue policy of the Central Revolutionary Base Area), Dangshi yanjiu yu
jiaoxue 黨史研究與教學, 4: 47–56.
Yang Kuisong 楊奎松. 1999. Mao Zedong yu Mosike de Enenyuanyuan 毛澤東與莫斯
科的恩恩怨怨 (Gratitude and Resentment in the relationship between Mao Zedong
and Moscow). Nanchang: Jiangxi renmin chubanshe.
Yang Kuisong 楊奎松. 2003. “Sun Zhongshen churang Man-Meng quanyi wenti de zai
tantao” 孫中山出讓滿蒙權益問題的再探討(An reexamination of Sun Yat-sen’s com-
promise on the issues of China’s rights and interests in Manchuria and Mongolia), Zhon-
gyang yanjiuyuan jindaishi yanjiusuo jikai 中央研究院近代史研究所集刊, 40: 249–
262
Yang Kuisong 楊奎松. 2007. “Cong gongjizhi dao zhiwu dengji gongzizhi: Xin
Zhongguo jianli qianhou dangzheng renyuan shouru fenpei zhidu de yanbian” 從供
給制到職務等級工資制──新中國建立前後黨政人員收入分配制度的演變 (From
free supply to the scaled salary system based on appointments: Changes in the
income distribution system of party and government personnel before and after the
founding of New China), Lishi yanjiu 歷史研究, 4: 111–137.
Yang Shu 楊恕 and Cao Wei 曹偉. 2008. “Ping Qingchao zai Xinjiang de geli zhidu”
評清朝在新疆的隔離制度 (The segregation system in Xinjiang during the Qing),
Zhongguo bianjiang shidi yanjiu 中國邊疆史地研究, 18.2: 40–48.
Yang Tao 楊濤. 1985. “Ming mo caizheng weiji yu sanxiang jiapai” 明末財政危機與
三餉加派 (The fiscal crisis and the imposition of three military levies in the late
Ming). Yunnan shifan daxue xuebao 雲南師範大學學報, 2: 9–15.
Yang Tianshi 楊天石, ed. 1996. Zhonghua minguo shi, di er bian, di wu juan, Beifa
zhanzheng yu Beiyang junfa de fumie 中華民國史 第二編 第五卷 北伐戰爭與北洋軍
閥的覆滅 (History of the Republic of China, part 2, volume 5, the Northern Exped-
ition and the demise of the Northern Warlords). Beijing: Zhonghua shuju.
Yang Yinfu 楊蔭溥. 1985. Minguo caizheng shi 民國財政史 (A history of finance in the
Republican period). Beijing: Zhongguo caizheng jingji chubanshe.
Ye Jianying 葉劍英. 1982. “Weida de zhanlue juece” (Great strategizing 偉大的戰略
決策). In Zhongguo renmin jiefangjun zhanshi chubanshe 中國人民解放軍戰士出
版社 ed., Xinghuo liaoyuan xuanbian zhi shi 星火燎原選編之十 (Beijing: Zhong-
guo renmin jiefangjun zhanshi chubanshe), 1–21.
Young, Arthur N. 1971. China’s Nation-Building Effort, 1927–1937: The Financial
and Economic Record. Stanford: Hoover Institution Press.
Yu Fengchun 于逢春. 2006. “Lun Zhongguo jiangyu zuizhong dianding de shikong
zuobiao” 論中國相遇最終奠定的時空坐標 (On the temporal and special contexts
for the final formation of China’s territoriality), Zhongguo bianjiang shidi yanjiu 中
國邊疆史地研究, 16.1: 1–13.
Yuan Senpo 袁森坡. 1996. “Qingchao zhili Meng Zang fanglue” 清朝治理蒙藏方略
(The Qing dynasty’s strategies to govern Mongolia and Tibet), Zhongguo bianjiang
shidi yanjiu 中國邊疆史地研究, 4: 43–49.
Zelin, Madeleine. 1984. The Magistrate’s Tael: Rationalizing Fiscal Reform in Eight-
eenth-Century Ch’ing China. Berkeley, CA: University of California Press.
Zeng Xianlin 曾憲林, Zeng Chenggui 曾成貴, and Xia Jiang 江峽. 1991. Beiyang
junfa shi 北洋軍閥史 (A history of the Northern Warlords). Chengdu: Sichuan
renmin chubanshe.
324 References
Zhang Gongquan 張公權 and Yang Zhixin 楊志信. 1986. Zhongguo tonghuo pengz-
hang shi: 1937–1949 中國通貨膨脹史（一九三七——一九四九）(A history of price
inflation in China, 1937–1949). Beijing: Wenshi ziliao chubanshe.
Zhang Hao 張皓 ed. 2008. Zhongguo xiandai shi 中國現代史 (Modern Chinese his-
tory). Beijing: Beijing shifan daxue chubanshe.
Zhang Jie 張杰. 1999. “Qingdai Kangxi chao juanmian zhengce qianxi” 清代康熙朝
蠲免政策淺析 (A tentative analysis of the tax exemption policy during the Kangxi
reign of the Qing), Gujin nongye 古今農業, 1: 54–59.
Zhang Peiguo 張佩國. 2000. Diquan fenpei, nongjia jingji, cunluo shequ: 1900–1945
nian de Shandong nongcun 地權分配, 農家經濟, 村落社區——1900–1945 年的山東
農村 (Distribution of land properties, peasant economy, and rural communities:
Rural Shandong in 1900–1945). Jinan: Qilu shushe.
Zhang Qixiong 張啟雄. 2010. “Zhonghua shijie zhixu yuanli de yuanqi: Jindai
Zhongguo waijiao fenzheng zhong de gudian wenhua jiazhi” 中華世界秩序原理的
源起：近代中國外交紛爭中的古典文化價值 (Origins of the principles of the Chin-
ese world order: The classic cultural values in diplomatic disputes in modern China).
In Wu Zhipan and Li Yu 李玉, eds., Dongya de jiazhi 東亞的價值 (The values of
East Asia) (Beijing: Beijing daxue chubanshe), 105–146.
Zhang Shuangzhi 張雙智. 2010. “Qingchao waifan tizhi nei de chaojin nianban yu
chaogong zhidu” 清朝外藩體制內的朝覲年班與朝貢制度 (The annual hearing and
the tributary systems in the Qing state’s government systems regarding the frontiers
and tributary states), Qingshi yanjiu 清史研究, 3: 106–115.
Zhang Xia 張俠, Yang Zhiben 楊志本, Luo Shuwei 羅澍偉, Wang Subo 王蘇波, and
Zhang Limin 張利民. 1982. Qingmo haijun shiliao 清末海軍史料 (Source material
on the naval force in the late Qing period). Beijing: Haiyang chubanshe.
Zhang Xianwen 張憲文, et al. 2005. Zhonghua minguo shi 中華民國史 (A history of the
Republic of China), 4 vols. Nanjing: Nanjing daxue chubanshe.
Zhang Xiaotang 張曉堂. 1990. “Qianlong nianjian Qing zhengfu pingheng caizheng
zhi yanjiu (A study of the Qing government’s fiscal balancing during the Qianlong
reign) 乾隆年間清政府平衡財政之研究. In Zhongguo renmin daxue Qingshi yan-
jiusuo 中國人民大學清史研究所, ed., Qingshi yanjiuji 清史研究集, 7: 26–60.
Zhang Xide 張喜德. 2009. “Gongchanguoji yu Yan’an zhengfeng yundong” 共產國
際與延安整風運動 (The Comintern and the Rectification Campaign in Yan’an),
Zhongguo dangshi yanjiu 中國黨史研究, 4: 74–83.
Zhang Yan 張研. 2008. 17–19 shiji Zhongguo de renkou yu shengcun huanjing 17–19世
紀中國的人口與生存環境 (Population and living environment in China from the
seventeenth to the nineteenth centuries). Hefei: Huangshan shushe.
Zhang Yong 張永. 2010. “Jiefang zhanzheng zhong yi sukuhui wei zhongxin de xinshi
zhengjun yundong” 解放戰爭中以訴苦會為中心的新式整軍運動 (The new-style
military rectification movement that centered at the meetings of speaking bitterness
during the War of Liberation), Zhonggong dangshi yanjiu 中共黨史研究, 6: 72–80.
Zhang Yongjiang 張永江. 2001. Qingchao fanbu yanjiu: Yi zhengzhi bianqian wei
zhongxin 清朝藩部研究：以政治變遷為中心 (A study of the Qing frontiers: Aspects
of political changes). Harbin: Heilongjiang jiaoyu chubanshe.
Zhang Yongle 章永楽. 2016. Jiubang xinzao 舊邦新造 (an old state rebuilt), 1911–
1917. Beijing: Beijing daxue chubanshe.
Zhang Yutao 張馭濤, Xu Zhanquan 徐占權 and Jiang Ying 江英. 1991. “Lun Zhong-
guo gongchandang dui jundui shixing juedui lingdao de jiben jingyan” 論中國共產
 References 325
黨對軍隊實行絕對領導的基本經驗 (On the basic lessons about the CCP’s absolute
command of the military), Junshi lishi yanjiu 軍事歷史研究, 2: 1–9.
Zhang Yuxin 張羽新. 1995. Qingdai qianqi xibu bianzheng shilun 清代前期西部邊政
史論 (On the history of the administration of western frontiers in the early Qing dyn-
asty). Harbin: Heilongjiang jiaoyu chubanshe.
Zhao, Gang. 2006. “Reinventing China: Imperial Qing Ideology and the Rise of
Modern Chinese National Identity in the Early Twentieth Century,” Modern
China, 32.1: 3–30.
Zhao Yuntian 趙雲田. 1984. “Qingdai qianqi liyong Lamajiao zhengce de xingcheng
he yanbian 清代前期利用喇嘛教政策的形成和演變” (The formation and evolution
of the policy to utilize Lamaism in the first half of the Qing period), Xizang minzu
xueyuan xuebao, 1: 64–76.
Zhao Yuntian 趙雲田. 1995. Qingdai zhili bianchui de shuniu Lifanyuan 清代治理邊
陲的樞紐──理藩院 (Lifanyuan as the central institution for the Qing state’s frontier
administration). Urumqi: Xinjiang renmin chubanshe.
Zheng Tianting 鄭天挺. 1999. Qingshi tanwei 清史探微 (A study of the Qing history).
Beijing: Beijing daxue chubanshe.
Zheng, Xiaowei. 2018. The Politics of Rights and the 1911 Revolution in China. Stan-
ford: Stanford University Press.
Wang Zhenghua 王正華. 2002. “1927 nian Jiang Jieshi yu Shanghai jinrong jie de
guanxi” 1927 年蔣介石與上海金融界的關係 (Jiang Jieshi’s relationship with the
financial circle in Shanghai in 1927). Jindaishi yanjiu 近代史研究, 4: 76–112.
Zhongguo shixuehui 中國史學會. eds. 1961. Yangwu yundong 洋務運動 (The West-
ernizing movement), vol. 1–8. Shanghai: Shanghai renmin chubanshe.
Zhou Gu 周谷. 1990. “Xizang maoyituan Mei Ying zhixing shimo” “西藏貿易團”美
英之行始末 (The Tibetan business delegation and its trips to the US and Britain),”
Zhongguo Xizang 中國西藏, Spring issue.
Zhou Yuanlian 周遠廉. 2003. Qingchao tongshi: Qianlong chao fenjuan 清朝通史——
乾隆朝分卷 (A general history of the Qing dynasty: Volume on the Qianlong reign).
Beijing: Zijincheng chubanshe.
Zhou Yumin 周育民. 2000. Wan Qing caizheng yu shehui bianqian 晚清財政與社會
變遷 (Finance and social change in the late Qing period). Shanghai: Shanghai
renmin chubanshe.
Zhou Zhichu 周志初. 2002. Wan Qing caizheng jingji yanjiu 晚清財政經濟研究 (A
study of the finance and economy in the late Qing period). Jinan: Qilu shushe.
Zhu Dong’an 朱東安. 2007. “Wanqing Man-Han guanxi yu Xinhai geming” 晚清滿
漢關係與辛亥革命 (The Manchu-Han relations in the late Qing period and the
1911 Revolution), Jindaishi yanjiu 近代史研究, 1: 36–43.
Zhu Hong 朱洪. 2007. “Dageming shiqi Sulian he Gongchanguoji dui Guo-Gong
liang dang jingji yuanzhu zhi bijiao” 大革命時期蘇聯和共產國際對國共兩黨經濟援
助之比較研究 (a comparison between the aids to the Guomindang and the CCP
provided by the Soviet Union and the Comintern during the Great Revolution
period), Dang de wenxian 黨的文獻, 2: 52–57.
Zhu Jianhua 朱建華. 1987a. Dongbei jiefang zhanzheng shi 東北解放戰爭史 (A history
of the Liberation War in the northeast). Harbin: Heilongjiang renmin chubanshe.
Zhu Jianhua 朱建華. 1987b. Dongbei jiefangqu caizheng jingji shigao 東北解放區財政
經濟史稿 (The financial and economic history of the northeastern liberated area).
Harbin: Heilongjiang renmin chubanshe.
Index

Admiralty 123 Board of Revenue 40, 53; cash reserves of

agricultural output 65, 87 57, 60, 64, 86–88, 108–109, 120
All under Heaven 29–30, 35, 40, 49, 107, Bogd Khan 205
117, 128 Bonney, R. 75
Alter, P. 18 Bourke, R. 18
Amursanaa 27 Boxer Rebellion 2, 132
Anderson, B. 18 Braude, B. 156
Anhui Clique 165, 168 Brewer, J. 19, 73
Annam 29, 129 Britain 15, 78, 200; see also England
Asia Relations Conference 199 Brubaker, R. 18
Austria-Hungary 156 Burbank, J. 18
Burma 27, 58, 79
Baer, M. 69
Bai Chongxi 175, 231, 234, Cai Huilin 267
240, 275 Cai Qian 79
Balibar, E. 288 Cai Wei 33
Bank of China 133 Cairo Conference 207–208
baojia 43, 71 Cao Kun 168
baolan 43, 68 Cassel, P. 114, 121
Baoyinchaoketu 50 CCP 1, 5, 10, 16–17; base areas of 230;
Battle of Menglianggu 268 historiography of 2; expansion of military
Bayly, C.A. 18 forces 242; and factionalism 242; fiscal
beg/boke 33 constitution of 271; and Nationalists
Beijing Government 14, 164; budgeted 184, 216, 228, 231, 237; and revenue
revenue and expenditure of individual generation 256–262; success in the Civil
provinces under 171–172; and Outer War 243–273; see also Mao Zedong;
Mongolia 214; revenue of 164–167 Red Army
Beijing-Tianjin Campaign 255 centralized regionalism 14–15, 186, 280
benevolent governance 68, 278 Chahar Mongols 23
Bergère, M. 143 chamber of commerce 133, 143
Bernhardt, K. 43, 68, 71 Chang, C. 44, 142
Bian, M. 238 Chang Jiashu 240
Blue Shirts 236 Changbaishan 23
Bluntschli, J. 159 Ch’en, J. 189
Bo Gu 250 Chen Beiping 35
 Index 327
Chen Bulei 239 neo-Confucianism 107; and Tongzhi
Chen Chao 194 Restoration 108
Chen Cheng 239 Confucius 35; Temple of 36
Chen Feng 37, 41, 44, 56, 61, 79, 80 Cooper, F. 18
Chen Fulin 175, 226 Coordinate Remittance 57, 79, 95,
Chen Guangfu 185 119–120, 121, 123, 130, 164
Chen Guofu 239 Corporate Law 133
Chen Huisheng 194 Court of Frontier Affairs see Lifanyuan
Chen Jinjin 221, 224 Crook, D. 262
Chen Jiongming 179–180, 223 Crook, I. 262
Chen Jitang 230, 231, 239–240 Crossley, P.K. 45, 146, 154
Chen Lifu 239 Cuba 1
Chen Mingsheng 178
Chen Shuwei 240 Dai Jitao 227
Chen Tianhua 151 Dai Li 239
Chen, Y. 273 Dai Yi 40
Chen Yun 261 Dalai Lama 25–26, 32, 33–35, 150,
Chen Yunxi 224 194–199, 201–202, 214
Chen Zhiping 38 Dawaachi 27
Cheng, V. 243 Decree of Abdication 153–154, 192
Cheng Zhongyuan 245, 274 DeMare 2019
Ch’i, H. 190, 217, 238 Deng Shaohui 88, 89, 95, 109, 134, 140
Chi Yunfei 147 Deng Xiaoping 275
Chiang Kai-shek 15, 184, 191, 214, 215; Deng Yibing 77
and Cairo conference 207–208, 213; as Ding Baozhen 116
chairman of GMD Central’s Military Ding Yizhuang 159
Committee 229; as GMD’s zhongcai Division of Military Training 141
233;as General Commander 228; Dong Biwu 261
popularity of 251; as president of the Dong Conglin 102
Nationalist government 226; on regime Dong Xianguang 228
failures 271; and regional forces 231, Dorgon 50
240–241, 267, 287 double cropping 67, 80
Chinese Communist Party see CCP Double Reduction 265
Ch’ü, T. 71 Downing, B. 7, 73, 75
Churchill, Winston 206, 207 Doyle, M. 18
Ci Hongfei 270 Du Yuming 255
Ci’an, Empress Dowager 116 Duan Qirui 168, 212
civil service examination 143 Duanfang 142, 147
Cixi, Empress Dowager 110, 116, 141 Duara, P. 43, 237
Coble P. 191
Cohen, P. 85 Eastman, L. 217, 236, 238, 243
cohong 113 Egypt 70, 156
Collins, R. 19 Eight Banners 37, 102, 127, 133, 147
Comintern 245 Eighteen-Star Flag 152
Communist revolution 5, 242–243, 271, Eisenstadt, S.N. 18
273, 281; see also CCP, Mao Zedong, Elliott, M. 25, 45–46, 146
Red Army Elman, B. 86, 104
Confucianism 29, 32, 85, 101; and Emerson, R. 18
Confucian patriotism 106; and Emperor Daoguang 62, 118
328 Index
Emperor Kangxi 25, 26, 27, 36, 40–41, 47, Friedman, E. 262
51, 52, 56, 282 from empire to nation-state 17, 276,
Emperor Qianlong 26; ban on taxation of 279, 282
newly reclaimed land 80; and Lord Frontier Defense 115–119
McCartney 112; on military expenditure Fu Xiaofeng 170
58; and tax waiver 40; and Zunghars 28, Fu Zuoyi 234
31, 282 Fujian Incident 240
Emperor Xianfeng 100 Furuya, K. 230
Emperor Yongzheng 26, 74, 77, 282; on
military spending 58; tax exemption Galdan Tseren 25, 26, 27, 34, 55, 60, 62
under 59; on war against Zunghars 30 Galtung, J. 19
empire 1–3; see also from empire to Gao Xianlin 233–234
nation-state, Inner Asian empire Gellner, E. 18
Engels, F. 78 gentry elites 142
England 38–39, 75, 80 gentry merchants 143
Ertman, T. 19, 75, 80 Geopolitics: defined 6
Esherick, J. 1, 49, 143, 152, 193 Germany 15, 156, 188
Ethiopia 4 Gillin, D. 173, 190
extraterritoriality 114 Glete, J. 19, 73
GMD 2, 169
factionalism: in CCP leadership 16; of Goldman, M. 45
GMD regional forces 230–232; see also Goldstein, M. 26, 196, 200, 214
zongpaizhuyi Goldstone, J. 38, 76, 80, 81
Fairbank, J.K. 45, 138 Great Production Movement 258
Fan Baichuan 123–126, 131 Green Standards 102, 127, 133
Fan Wenlan 100 Greenfield, L. 18
fascism 236 Grunfeld, A.T. 200
Feng Erkang 27 Gu Bao 194
Feng Guozhang 168 Guan Hanhui 80
Feng Lida 239 Guangdong province: in Northern
Feng Tianfu 260, 265 Expedition 179; revenue of 180
Feng Yuxiang 174, 176, 183, 191, 218, Guangxianshi 27
229, 239 guanyu 165
fengpiao 170 Guilmartin, J. 70
Fengtian Clique 165, 169–173, 182, 190, Gunn, S. 80, 81
212, 218 Guo Chengkang 29, 34, 35, 101
Feuerwerker, A. 80, 108 Guo Songyi 65–67
Financial Commissioner 96, 98 Guo Xuyin 222
Fiscal Clearance and Consolidation Guomindang see GMD
Bureau 139 Gurkhas 79
fiscal-military state 73; in early modern Guy, K. 73
Europe 11; in early republican
China 14, 163 Hai Chunliang 203
Fitzgerald, J. 150 Halsey, S. 19, 86, 104
Five-Color Flag 152 Hamashita, T. 45
Fletcher, J. 47 Han dynasty 2, 3
Foochow Arsenal 123, 126 Han Fuqu 221, 230, 233–234
France 15, 38–39, 75, 76, 114, 129, 217 Hao Bingrang 169, 170
 Index 329
Harding, A. 80 Isaacs, H. 217
Harriman, W.A. 206 Italy 15, 188
Harrison, H. 18
Harriss, G. 80 Jahangir Khoja 57
Hart, Robert 138 Japan 4, 15, 86, 188, 193
He Jiawei 275 jasagh 32
He Ping 40, 50 Jebtsundamba Khutuktu 34, 50, 150, 192,
He, W. 19, 108 203, 204
He Yuan 174 Jia Shiyi 164, 167, 168, 172, 178, 183, 186
He Zhaowu 224 Jia Shuocun 116
Hechter, M. 18 Jiang Bingming 130
Hertslet, G. 114, 195 Jiang Chun 79
Heshen 109 Jiang Jieshi 215
high-level disequilibrium 12, 137 Jiang Tao 109
Hinton, W. 262, 264 Jiang Tingfu 114
Hintze, O. 6 Jiang Zhijie 120
Ho, P. 36, 37, 45, 88 Jiangsu province 178
Hobsbawn, E.J. 18 Jiangxi Soviet 251
Hobson, J. 288 Jin Congji 152, 242, 254
Hong Taiji 34 Jin dynasty 47
Hoton Nor 27 Jin Yilin 229, 239
Hou Yijie 139 Jinchuan 56; Tibetan rebellions in
Hsiao, K. 71, 77 59, 62, 79
Hu Guangyong 120 Jing Junjian 54, 76
Hu Hanmin 225, 227–229, 240 Johnson, C. 273
Hu Qiaomu 246 Jordan, D. 217
Hu Sheng 2, 145, 252 Judicial Commissioner 98
Hu Yuhai 215 Jurchens 33, 47; see also Manchus
Hu Zongnan 239
Huai-Hai Campaign 255, Kaeuper, R. 80
265, 272 Kapp, R. 177, 190
Huang Daoxuan 228 karan 32
Huang Musong 198 Kashag 33
Huang, P. 18, 43 Kazakhs 28
Huang Shaohong 175 Keating, P. 273
Huang Xingtao 46, 150, 154 Kemal, Mustafa 157, 158
Huang Yao 256 Khalkha Mongols 25, 29, 31, 34, 47,
Hunan Army 99, 102 50, 205
Hundred-Day Reform 134, 141 Khobdo 32
Kiangnan Arsenal 123
Icang-skya Khutughtu 151 Kim, K. 150
identity building 8, 17 Kirby, W. 213, 238, 240
India 193 Kivanç, K. 70, 155
Inner Asian empire 10, 46 Kokand 13, 57, 62, 121
Inner Mongolia 1, 9, 14, 24–25, 33, 35, Konchok Jungne 196
49, 70, 130, 150–151, 153–154, 193, Kong Jingwei 170
213, 240, 279, 284 Korea 48, 128, 129, 193, 207
Iran 4 Kuang Shanji 177
330 Index
Kuhn, P. 18, 69, 99, 187 Liang Shuming 252
Kumar, K. 18 Liao dynasty 47
Kuo, T. 104 Liao Zhongkai 228
Kuren/Kulun/Ulan Bator 32 Liaoxi-Shenyang Campaign 255
Kwantung Army 186, 207, 212, 254, 256 Lieberman, V. 73
Lifanyuan 32, 33
Lai, C. 108 light taxation 37, 40, 60, 63, 135
Lai Xinxia 165, 168, 190 lijin 90, 93, 103, 121, 131, 136, 140, 181,
Lamaism 9–10, 25, 33–35, 46, 47 182; abolition of 222–223, 238
land reform 262–263 Lin Zexu 118
Land tax 7, 38, 60, 87, 93, 135, 159; see literary cases 36
also light taxation Liu Bocheng 260, 275
Laos 1 Liu Cuirong 54
Lary, D. 175, 190, 243 Liu Jingzhong 176, 191
Late-Qing state: integration with world Liu, K.C. 104, 106, 108, 119, 138, 190
system of states 5 Liu Kunyi 104, 142
Later Jin 23, 33 Liu Ruizhong 80, 138
Lee, D. 18 Liu Shaoqi 253
Levenson, J. 85 Liu Tong 254
Levich, E. 175 Liu Wei 99, 102, 103, 144
Levine, S. 243 Liu Wenhui 230
Lew, C. 243 Liu, X. 19
Lewis, B. 70, 157, 158 Liu Xiang 230
Lhazang Khan 25 Liu Yuewu 32, 35
Li, B. 80 Liu Zenghe 95, 121, 140, 164
Li Bingjun 274 Liu Ziyang 99
Li Daokui 80 Liuchiu/Liuqiu 29, 128, 131, 209–210,
Li, H. 2, 44, 71, 77, 144, 236, 274 213, 215, 240
Li Hongzhang 13, 96, 102, 142; and Lobzang Danjin 26
Maritime Defense 116–117, 122–126, Long Yun 230, 234
130, 131; and the Qing court 104; and Longyu, Empress Dowager 153
Self-Strengthening 105, 108; and Zeng Lord McCartney 112
Guofan 110 low-level equilibrium 11, 12, 62–65,
Li Jingzhi 219, 221 77–79, 86–88, 112
Li Lifeng 262, 275 Lu Rongting 174
Li Qiang 87 Lu Shichuan 258
Li Rong 215 Lugouqiao (Marco Polo Bridge) Incident
Li Weiguang 260 175, 233
Li Xin 190 Luo Ergang 87, 88
Li Yongjun 130 Luo Jun 275
Li Yuanhong 152 Li Yunchang 254 Luo Pinghan 262
Li Zhiqin 49 Luo Yuanzheng 239
Li Zhiting 2011 Luo Yudong 93, 96
Li Zongren 175, 182, 218, 219, 221, Luo Zhitian 163
228–231, 234, 239, 240, 275
Li Zongyi 190 Ma Bufang 230
Liang Qichao 138, 149, 153, 154; on greater Ma Dazheng 121
nationalism and petty nationalism 159 Ma Jinhua 120
 Index 331
Ma Ruheng 121 Mo Jijie 175
Ma Xuanwei 233 Mongolia 128, 214; see also Inner
MacKinnon, S. 190, 238 Mongolia, Outer Mongolia
macrohistory 5–6 Mongolian People’s Republic 205
Maddison, A. 76, 80 Mongols see Chahar Mongols, Khalkha
Manchuria: and CCP 16, 252–254; Mongols, Oirat Mongols, Zunghar
industrial production in 274; Japanese Mongols
occupation of 232; under Zhang Zuolin Morgan, E. 18
15, 169, 211 Mosca, M. 79
Manchus: banners 32; identity of 30; mountaintop-ism see shantouzhuyi
relations with Han elites 97, 101, Muthu, S. 18
110–111, 122, 139, 142, 147–148;
Sinification of 14, 45 Nanjing Government 163, 219; military
Mandate of Heaven 36 spending of 235, 270; price inflation
Mann, M. 73 under 270; revenues of 219–220
Mann, S. 68, 76, 110 Nathan, A. 190
Mann, M. 7, 75 nation-state 1–3, 17, 18, 45, 114, 132,
Mao Zedong 225; and CCP leadership 16, 150, 189, 276, 279, 282
225, 244–245, 250, 268; cult of 287; nationalism 5, 15, 18, 45, 278; in early
ideological dominance 247; on New twentieth-century China 146; GMD and
Democracy 246; and relationship with 179; greater 159; Han-centered
Zhou Enlai 249; strategizing in the Civil 151–152; incipient 104–106; petty 159;
War 251, 255; and Ten Military popular 273–274; redefinition of 284; in
Principles 260; see also Mao Zedong republican China 213, 240–241; Sun
Thought Yat-sen and 223
Mao Zedong Thought 245 Nationalist Government: in Guangdong
maritime customs 91–92, 93–95, 99, 109, 179–183; in Wuhan 183–184; see also
120–121, 123–124, 130, 136, 137, Nanjing Government
139–140, 164–165, 166–167, 177, 179, Nationalist Party see GMD
185, 220, 222–223, 235 nationless state 150
Maritime Defense 115–119 Nepal 79
Marshall, George 200, 267 New Army 133, 140, 143
Marx, K. 78 New Policies 13, 130, 132–134
May Fourth Directive 262 Ni Yuping 96
McCord, E. 179, 190, 217 Nian Rebellion 89, 102
McCormack, G. 170, 190 Niu Hongbin 234
McMahon Line 214 North Africa 70
Mehta, S. 19 North Korea 1
melting fee 41, 68 Northern Expedition 179, 182–183,
Mencius 29 184–186, 216, 223
Metropolitan Remittance 95, Northern Fleet 104, 123, 124
130, 164 Northern Wei 47
Miao people 27, 79
Miao Zhou 50 O’Brien, P. 80
millets 71 Oirat Mongols 25
Millward, J. 23, 29, 45, 47, 193 Opello, W. 18
Min Zongdian 67, 80 Opium War 2, 86, 113, 118
Ming dynasty 3, 50, 80 Ottoman empire 69–72, 155–158
miri land 70 Ottomanism 157
332 Index
Outer Mongolia 9, 15, 23, 24–25, 28, 32, Reardon-Anderson, J. 150
33, 35, 47, 50, 122, 130, 153–154, Rectification Campaign 248
192–195, 202–207, 210–215, 266 Red Army: Eastern Expedition of 274; in
Ouyang Junxi 224 GMD propaganda 274; in Jiangxi 251;
Mao’s leadership of 244, 247, 249–250;
Pagden, A. 18 methods to generate revenue 257; and
Pamuk, S. 70 Nationalist forces 231–232; Western
Pan-Islamism 157 Expedition of 274; Zhang Xueliang
Pan-Turkism 157 and 233
Panchen Lama 33–34, 150–151, regionalized centralism 13, 106–108, 120,
197–198, 214 186, 187, 278, 280, 281
Pang Xianzhi 242, 244, 247, 259, 274 Remick, E. 238
Peng Xinwei 68 renminbi 261
Peng Zeyi 54, 55, 88, 89, 109 Republic government see Beijing
People’s democratic dictatorship 5 Government, Nanjing Government
People’s Republic of China see PRC Republic of Five Races 14, 284
Pepper, S. 243 Reting Rinpoche 198
Perdue, P. 23–24, 25, 27, 47, 52–53, 121 Revolution of 1911 2, 134, 152, 204
Persia see Iran Rhoads, E. 99, 146, 152
Pong, D. 108 Roberts, M. 6
population: of Qing China 66–67 Roeder, P. 18
Prazniak, R. 69, 139 Rong Hong 110
Prince Gong 110, 116 Ronglu 142
Prince Yikuang 141 Roosevelt, Franklin 206, 208, 209
PRC 1, 3 Rowe, W. 77
provincial governors and governors-general: Rudolph, J. 18
control of the military and local Ruicheng 149
government 99–100; financial autonomy Russia 116, 193, 203; and Mongolia
of 96; and unreported tax revenues 136, 203–204
139–140; and self-government 144
Pye, L. 190 Safavids 72
Said, E. 288
Qimudedaoerji 25 salt merchants 58
Qin dynasty 2, 3 salt tax 60, 93, 136, 140, 164, 220, 235
Qin Qingjun 180, 182 Sanggye Gyatso 25
Qing state: domestic governance 31–44, Sanminzhuyi 223–225, 244, 247
278; expenditure of 37, 54; fiscal cycles Scammell, G.V. 18
53–57; formal bureaucracy of 44; Second Opium War 113, 118
geopolitical strategy 24–28, 123; military Selden, M. 273
forces of 74; military spending 88–89, Self-Strengthening movement 86, 103,
124; revenue of 54, 63, 90–94, 135, 104–106
137; standing army of 37; territorial semi-centralism 16, 234–237
expansion 5, 24, 277 Shakabpa, T. 202
Shang Bing 135
Rankin, M. 144 Shang Hongkui 34, 35, 36
Rasler, K. 73 shantouzhuyi 16
Rawlinson, J. 108 Shaw, S. 70, 72
Rawski, E. 23, 30, 32, 45, 73 Shen Baozhen 118, 122–123, 125
 Index 333
Shen Shicai 211 Svarverud, R. 114
Shen Xiaoyun 178 Szonyi, M. 18
Sherridan, J. 190
Shi Yousan 221 Taiping Rebellion 2, 85, 87, 88, 89, 90, 92,
Shi Zhihong 56, 57, 88, 89, 93–94, 95–96, 95–98, 100–103, 107, 109–110,
103, 109, 119, 127, 138 121–122, 137, 146, 164, 280
Shi Zhongquan 249 Taiwan 49, 56, 116, 118–119, 124–125,
Shunzhi emperor 25, 32, 36, 40, 50 213, 215, 240, 268; cession to Japan
Siam 29; see also Thailand 127, 130, 132, 240; and Cairo
Sichuan province: garrison-area system conference 207–210; Japan’s invasion of
in 177 115; Qing campaign on 79; under the
Simla Accord 195–196 Zheng regime 55
Sinicization 45 Taktra Rinpoche 199
Sino-French War 123 Tan Qixiang 2, 275
Sino-Japanese War of 1894 13, 106, 124, Tan Zhaoyi 175
129, 132, 216 Tang Enbo 239
Skinner, Q. 18 Tanner, H. 243
Smith, R. 119 tariff autonomy 220
Song Lianxi 80 tax farming 74
Song Zheyuan 230, 232–233, 240 tax waivers 40
Song Ziwen 180, 183, 186, 206, 220, 222 Taylor, J. 186, 238, 275
Soong Meiling 186 territorial state 11
Southern Fleet 123 Thailand 4
Soviet Union 1, 176, 180, 191, Thaxton, R. 273
211, 214 Theobald, U. 79
speaking bitterness 269 Thompson, W. 73
Stalin, Joseph 206, 208 Thornton, P. 19, 69, 236
state-making: defined 3 Three Feudatories 36, 37, 55, 58,
statism 236 60, 61
Storr, C. 19, 73 Tianli Sect 79
Strauss, J. 238 Tibet 1, 4, 9, 10, 14, 15, 23, 25–28,
Strayer, J. 80 31–35, 37, 45–47, 56–57, 59, 62, 70,
Streusand, D. 69, 70 79, 130, 133, 146, 150–154, 157, 159,
Su Quanyou 139 192–202, 210–214, 277, 279, 282, 283
Su Yu 272 Tieliang 139
Suleski, R. 170, 190 Tien, H. 238
Summer Palace (Yiheyuan) 125 Tilly, C. 6, 7, 73, 75, 76, 77, 189
Sun Chuanfang 169 timar system 69, 70, 155
Sun Fo see Sun Ke Tong, Hollington 208
Sun Hongnian 49, 128, 130 Tongmenghui 239
Sun Ke 227–229 Tongzhi emperor 114
Sun Weihou 267 Tongzhi Restoration 108
Sun Yat-sen 15, 152–153, 158; and Adolph Tonkin 129
Joffe 206, 215; and Chinese Torgovnick, M. 288
Revolutionary Party 225; and GMD 239; total centralism: defined 17
as Grand Marshall 179; and Guomindang Treaty of Nanjing 113
190; as provisional president 192; on Treaty of Tianjin (Tien-tsin) 113, 220
Sanminzhuyi 223 tributary system 113, 128
Supply System 271 Truman, H. 207
334 Index
Tsungli Yamen 114 245, 251, 258–259, 262, 264, 267, 269,
Tuck, R. 18 270, 275
warlordism 14, 164–178
Ulan Butong 25 Wedemeyer, Albert 217
Uliastai/Uliyasutai 32 Wei Guangtao 141
unified tax 222, 235 Wei Hongyun 258
United Front 245, 262 Wen Jing 224
United Nations 15 Wen Xianmei 233
United States 266 Wenqing 98
Urumqi 80 Wenxiang 116
Uyhgurs 27, 33, 59 Westad, O. 243
White Lotus Rebellion 57, 79, 86, 98
van Creveld, M. 80 White Lotus Sect 36, 58, 61
van de Ven, H. 191 Wilbur, M. 163, 179, 180
van Schaik, S. 26, 202 Wong, B. 19, 80
Vietnam 1, 48, 79, 128, 129; see also World War II 15
Annam Wou, O. 190, 274
Wright, M. 108, 145
Waldron, A. 179, 190 Wu Chengming 66, 80
Waley-Cohen, J. 45 Wu Jingping 181, 183, 185, 206, 220
Wang Bingming 203 Wu Peifu 169, 180, 212, 215
Wang Chaoguang 266, 268, 270 Wu Zhongxin 199
Wang Jialie 230
Wang Jiaxiang 244, 245 Xi’an Incident 233
Wang Jingwei 184, 223, 225, xiangbao 43
227–229, 237 xiangdi 43
Wang Jingyu 135 Xiao Yishan 114
Wang Kaixi 97, 100 Xie Benshu 234
Wang Liqi 274 Xiliang 149
Wang Miaosheng 255 Xing Quancheng 35
Wang Ming 245, 246, 274 Xinjiang 9, 13, 14, 23–25, 28–29, 33, 35,
Wang Nianyong 159 37, 47, 52, 53, 62, 80, 89, 115–116,
Wang Qisheng 236 127, 130, 146, 150, 165, 171, 193–194,
Wang Rigen 77 211–213, 215, 277, 279, 282, 284; beg
Wang Sizhi 25 system in 121–122; incorporated into the
Wang Wenshao 142 Qing 193; jasak system in 121–122;
Wang Xi’en 152 junfu system in 121–122; Qing court’s
Wang Xutian 174 annual expenditure in 117; Zuo
Wang, Y. 37, 71 Zongtang’s Western Expedition to 118
Wang Yinhuan 270 Xu Chongzhi 228
Wang Yugui 191 Xu Daofu 136
Wang Yuru 136 Xu Dixin 66, 80
Wang Yunsheng 129 Xu Entong 239
Wang Zhenghua 191 Xu Shuzheng 205
Wang Zizhuang 239 Xu Tan 54, 76
War of Resistance 15, 16, 175, 198, 216, Xu Yi 88, 257
217, 233, 234, 237, 238, 240, 242, 244, Xu Yisheng 136
 Index 335
Yack, B. 18 Zhang Hao 221
Yalta Conference 206 Zhang Jie 40
Yan Hongzhong 136 Zhang Jingjiang 227, 228
Yan Xishan 173, 218, 219, 229, 230, Zhang Limin 131
233–234 Zhang Peiguo 262
Yang Du 150 Zhang Shuangzhi 33
Yang Guangyan 177 Zhang Taiyan 151, 154, 159
Yang Hucheng 233 Zhang Wentian 245, 248, 250, 274
Yang Jing 257, 275 Zhang Xia 123
Yang Kuisong 254, 271 Zhang Xianwen 253, 259, 262, 266, 267
Yang Shu 33 Zhang Xiaotang 59
Yang Tao 50 Zhang Xide 248
Yang Tianshi 184 Zhang Xueliang 179, 186, 219, 221,
Yang Yinfu 165, 167, 221, 235, 270 232–234, 239
Yang Zengxin 194, 211 Zhang Yan 80
Yang Zhixin 270 Zhang Yong 275
yanyu 165 Zhang Yongjiang 31, 33
Yao people 79 Zhang Yongle 3, 153
Yaqub Bek 28, 115 Zhang Yuxin 27
Ye Jianying 253 Zhang Zhidong 104, 141, 142, 147
Yixin see Prince Gong Zhang Zhizhong 236
Young, A. 167 Zhang Zuolin 15, 169–170, 174,
Youtai 195 186, 211, 213, 215, 239
Yu Fengchun 49 Zhao, G. 46
Yu Qiaqing 184, 227 Zhao Yuntian 26, 35, 47
Yuan dynasty 47 Zhigang 114
Yuan Senpo 35 Zhili Clique 165, 218
Yuan Shikai: and the abdication of the Qing Zhongguo 10, 20, 46, 105, 148,
emperor 153; and the Dalai Lama 202; 151, 158
and debt borrowing 164, 220; Zhonghou 114
dictatorship of 279; handling of foreign Zhonghua minzu 192, 284
affairs 212; and Mongol princes 154; and Zhou Enlai 244, 249, 250
New Army 102, 134, 141; ousted from Zhou Gu 200
the court 142; and Outer Mongolia 204; Zhou Yuanlian 27
as president 192; relationship with Zhou Yumin 76, 89, 105, 109, 119,
Cixi 141 134, 164
Yusuf Khoja 57 Zhou Zhichu 39, 138
Zhu Dong’an 110
Zanabazar 34 Zhu Hong 180
Zelin, M. 41, 68, 71, 73 Zhu Jianhua 252, 255, 256, 260, 262
Zeng Guofan: and administrative autonomy zongpaizhuyi 16
99–100; and the Qing court 110; and Zou Rong 151
Taiping Rebellion 98, 121 Zunghars 6, 9, 25, 52, 70, 116, 282
Zeng Kelin 254 Zuo Zongtang 13, 28, 104, 109; on
Zhandui 79 Frontier Defense 115; and Western
Zhang Gongquan 270 Expedition 120, 194