Governing

China
150 1850
John W. Dardess
Governing China
1501850
Governing China
1501850

John W. Dardess

Hackett Publishing Company, Inc.

Indianapolis/Cambridge
Copyright 2010 by Hackett Publishing Company, Inc.

All rights reserved

Printed in the United States of America

14 13 12 11 10 1 2 3 4 5 6 7

For further information, please address

Hackett Publishing Company, Inc.
P.O. Box 44937
Indianapolis, Indiana 46244-0937

www.hackettpublishing.com

Cover design by Abigail Coyle

Text design by Mary Vasquez
Maps by William Nelson
Composition by Cohographics
Printed at Sheridan Books, Inc.

Library of Congress Cataloging-in-Publication Data

Dardess, John W., 1937
Governing China : 1501850 / John W. Dardess.
p. cm.
Includes bibliographical references and index.
ISBN 978-1-60384-311-9 (pbk.) ISBN 978-1-60384-312-6 (cloth)
1. ChinaPolitics and government. 2. ChinaSocial conditions.
3. ChinaHistoryHan dynasty, 202 B.C.220 A.D. 4. ChinaHistory
Qing dynasty, 16441912. 5. Political cultureChinaHistory. 6. Social
institutionsChinaHistory. 7. EducationChinaHistory. I. Title.

DS740.2.D37 2010
951dc22 2010015241

The paper used in this publication meets the minimum requirements

of American National Standard for Information Sciences
Permanence of paper for Printed Library Materials, ANSI z39.481984.
 Contents

Preface vii
Introduction: Comparing China in 150 and China in 1850 x
Timelines xxiii
Maps xxvii

Part 1. From Fragmentation to Reunification, 150589 1

The Unraveling of the Later Han, 150220 3
The Three Kingdoms, 221264 5
The Western Jin, 266311 6
A Fractured Age, 311450 8
Unity in the North: The Northern Wei, 398534 12
Not by Blood Alone: Steps to Reunification, 534589 16
Part 2. Imperial Grandeur and Its Aftermath, 589907 19
Unification by the Short-lived Sui, 581618 22
The Long-lived Tang, 618907 23
The Tang as Empire 26
After Empire: Reconstructing China 28
Confucianism and Buddhism in the Late Tang 30
The Shattering of the Tang, 868907 32
The Five Dynasties and Ten Kingdoms, 907960 34
Part 3. The Tripartition of China, 9601279 36
The Liao (9271125), Northern Song (9601126), and
Xi Xia (9821227) 38
The Jin (11151234), Xi Xia (9821227), and
Southern Song (11271279) 45
New Developments in the Southern Song 47
The End of Tripartition 49
Part 4. Permanent Unity Largely Achieved: Yuan, Ming,
and Qing, 1271 to 1850 52
Yuan China, 12711368 61
The Creation of Ming China, from 1368 66


vi Contents

The Ming Refounded: Jianwen (r. 13981402) and

Yongle (r. 14021424) 69
The Middle Years of the Ming 71
The Reign of Jiajing, 15221566 73
Late Ming Foreign Relations 76
The Late Ming: Internal Developments 78
The Ming Collapse, from 1627 80
Qing China: The Founding, from 1644 82
The Conquest of South China, 16441683 84
Beyond China: Building the Qing Empire 86
Refocusing on China: Yongzheng, r. 17221735 88
The Reign of Qianlong, 17351799 89
Social Institutions in Political Context 91

Further Reading 103

Index 107
 Preface

M y excuse for writing this short book is just a guessthat it

might prove useful for students of China and the occasional
general reader to have at hand a concise and accessible account of
what happened among, and to, that huge agglomeration of people
we call Chinese at the east edge of the Eurasian landmass over a very
long segment of their recorded history. Originally, it was the Italian
editors of a multiauthored set of books about China who invited me
to contribute a long section on that topic within the 1501850 time
frame. The idea made immediate sense to me, and I accepted the
invitation. The present offering revises and expands a bit on what I
wrote for an Italian readership.

Why focus on governing, and why adopt this particular time frame?
I take governance to be both the ultimate barometer as well as the
ultimate determinant of the well-being of human society and its pan-
oply of activities. The institutions of government and the men and
women who assume the governing positions reach us individually
only occasionally, but they always affect us generally. This is because
governments must try above all else to enforce order and ensure
physical security to all within their borders. Political history is the
ongoing story of how these things are done, or not done, over such
and such a landscape over a given stretch of time. This basic matter
is crude and easily stated, but the detail is infinite in its variety and
complexity. The long story line ever fascinates with its triumphs and
disasters, its marches to victory, its collapses in ignominy, its what-ifs,
its who-is-to-blames, its factional struggles, its great stars, its cruel
and corrupt villains, its good administrators, its feeble dupes, and on
and on.
I found myself enjoying the writing of this little book. I liked
reading up on a very large body of recent scholarship on Chinese
history. I also found meaningful resonances with my own sense of
autobiographical place in the larger story of the governance of the
United States in the mid-to-late twentieth century: the mobilization
of family members into the war against the Third Reich; the Cold

vii
viii Preface

War that followed; Georgetown University; the U.S. Armys demand

for Chinese-language specialists; the Army Language School; the
Army Security Agency and its listening post on Taiwan; Columbia
University; the University of Kansasall large and powerful institu-
tions set in a larger context of national needs and security, whose
various missions gave definite form and direction to my own lifes
course. I have witnessed from afar three different Chinas in my life-
time: first, the miserable, war-torn, and starving China as shown in
current events films to us grade school children in the late 1940s;
then, the frightening Red China with its human-sea tactics, nuclear
bombs, Great Leap Forward, and such horrors; and now, since the
late 1970s, the gradual emergence of an authoritarian and repressive
China as a great economic power and sober global actor. Both gover-
nance in general, and Chinas governance in particular, intertwine
with my own biography.
As for the time frame, the years 150 to 1850, the choice appears
practical to me. It would be physically nearly impossible to try to
cover all of Chinas past in one short book. Besides, both early China
and modern China enjoy strong and compact communities of inter-
ested scholars, students, and readers. The intervening 1,700 years are
by contrast so divided among specialists in the various dynasties and
topical fields that the whole span seems to consistmisleadingly
of completely discrete and unrelated bits. Tang experts seldom have
much to say to Ming specialists; religious historians rarely venture
into military affairs. So there is an opportunity here to address the
whole long story and tease out the main lines of coherence and devel-
opment. And the years 150 and 1850 serve fairly well as bookends in
a narrower sense. In 150, the long-lived Han realm, for centuries
the dominant power in East Asia, reached a point of deflection and
downspin into civil war, ruin, and, seventy years later, collapse. By
1850 the Qing, also the dominant East Asian power, reached a simi-
lar point and it too collapsedsixty years later. Foreign monks and
the originally Indian religion of Buddhism rose to places of promi-
nence in China after the Han fall; Marxism, Comintern personnel,
and Western ideas and advisers of all sorts entered China soon after
the fall of the Qing. In both cases, Chinas educated people sought
answers to terrible domestic problems. The falls of the Han and
Qing produced two very different developments, then, that in some
 Preface ix

ways trace similar trajectories. What else comes up if one compares

the years 150 and 1850?
It is intriguing to note that in 150 China was not far from the
eve of great apocalyptic uprisings, including especially the Yellow
Turban outbreak of 184. In 1850, Qing China was also about to
explode in rebellions, among which the largest was the millenar-
ian Taiping rebellion of 18501864. Both rebellions came to base
themselves in the more thickly settled parts of the China of their
time; significantly, both shared the goal of taiping (great peace or
great leveling); and while both were eventually suppressed, the
suppression was made possible only by an eventually fatal devolu-
tion of power downward from the imperial center to the regions and
provinces. Between 150 and 1850 only one rebellion rivaled those
outbreaks in scale and ferocity, and that was the Red Turban affair of
13511354, also millenarian, which engulfed all of China and led,
indirectly, to the collapse of the Mongol Yuan dynasty. Many other
popular risings, to which central governments were extremely sen-
sitive, erupted during times in between. But broader comparisons
than this between the late Han and the late Qing can be drawn as
well and to those this book now turns.
 Introduction
Comparing China in 150
and China in 1850

M odern scholarship seldom addresses itself to the question of com-

paring China at two such remote moments in time as 150 and
1850, so one is forced to wade a bit into unexplored territory to perform
such a comparison. That said, one cannot but be impressed by the
durability of Chinas political, social, and educational institutions over
the 1,700 years from 150 to 1850, from the time of Antoninus Pius,
say, to that of Queen Victoria. Durability, but not rigidity. If one were
to compare the institutions of the Later Han dynasty of 150 to those
of the Qing of 1850, the similarities are of gross features, not of detail.
These similarities can be explained in part by the persistence of envi-
ronmental constraints (which were very different for Antoninus Pius
and the queen), plus the continuity of the written language and the
perennial emphasis in Chinas educational curriculum on the study of
history and the Confucian classics. The compilation and preservation,
dynasty after dynasty, of institutional monographs, encyclopedias, and
other written guides made it no difficult matter to discover in great
detail how governments of the past had been structured.
As for the environment, it may be said that China long occupied
the best land in all of Eurasia. Once upon a time, thousands of years
ago, East Asia could boast one of the richest and most varied endow-
ments of plants and animals to be found anywhere on earth. Unlike
species-poor Europe and North America, China was never defaced
by glaciers. First to be thickly settled, farmed, and made to sustain
a civilized polity was north China, with its rich soils, from roughly
the second millennium BCE. A long and slow process of immigra-
tion from the north added south China to this core. By about 1000
CE, the souths population rose to parity with that of the north. In
1000 CE, Chinamost of it under the rule of the Northern Song
dynastywas beyond question the largest and wealthiest organized
society in the world.


 Introduction xi

But what nature generously endowed with one hand, it flogged

with another. China proper was, and still is, susceptible to prolonged
and devastating droughts, and to catastrophic floods in the several
great river systems that flow west to east across its territory. From very
early on, it became incumbent upon the rulers of this rich land to
defend the realm against encroachment from without and also to
devise means to alleviate the effects of drought and flood within.
By 150 CE, when our story begins, China had already developed
a sophisticated and organized bureaucracy at the central and local
levels, had assembled a copious repertory of ideas and techniques
for governing the people, and had long and detailed written records
available showing how these ideas and techniques had worked, or
failed to work, in the past.
By 150 CE, the main principles of Chinas general strategy had
been laid; and when our story ends, in 1850 CE, for all the changes
that had occurred over those 1,700 years, the principles are recog-
nizably the same. What were they? First, that the main business
of China was China. Empire building, that is, the occupation of
Inner Asian and other territory not inhabited by Chinese people, or
Chinese speakers, was an occasional policy grounded mainly in secu-
rity concerns. Empire was costly and simply never paid economic
dividends. China was rich; territory beyond China was, compara-
tively speaking, poor. The pursuit of global commercial dominance
by sea, as with the famous Ming flotillas of the fifteenth century,
was also inordinately expensive. The Ming ventures were cancelled
after a quarter century of operations and were never resuscitated. So
China might pursue empire, but it might also withdrawas it did
in the Later Han (25220 CE), the Tang (after 755), and the Ming
(13681644)without major damage. Thus the focus of China was,
preeminently, China.
This leads directly to the second principle: that the purpose of any
Chinese government was to protect the people of China from the
vagaries of nature, from external enemies, and from themselves; to
provide peace and tranquility and good order, so that happy and pro-
ductive lives might be enjoyed by all, and the future made secure for
everyones children and grandchildren. Policies, and the institutions
necessary to their realization, were designed accordingly. Often they
worked well. Sometimes they worked badly, or not at all. There were
xii Introduction

intense disagreements over basic policy, as there was often more

than one way to achieve a desired end, with momentous institutional
consequences attendant upon the choice, as in the eleventh-century
Northern Song.
There is a third set of principles observable in the story of China
from 150 to 1850. Whether controlling all of China, or only a frag-
ment of it, the government was always a monarchy. In addition to a
military component, such monarchies always featured an ordered
and ranked bureaucracy. These bureaucracies could range from
small and docile groups of advisers and functionaries in the service
of a warlord to large (2050,000) meritocracies recruited through the
national Confucian civil service examination system, minutely regu-
lated, riddled with factionssome self-absorbed, some corruptyet
often with a strong sense of rectitude and even moral superiority that
monarchs often came to fear and resent. It might be argued that the
most fundamental constant of all was the Confucian extended fam-
ily system, which proved capable of adapting to and surviving almost
anything: flood, drought, barbarian invasion and rule, murderous
internal banditry and warfare, migration, opulence, poverty, agrarian-
ism, commercialism, anarchy, new religions, and much else. Closely
connected with the strength and durability of the family system was
educationthe desire of families and extended families to transmit
literary skills and the knowledge of classical Confucian texts to their
junior members (predominantly but not exclusively the boys). From
this educated pool came the bureaucrats who helped rulers govern
China. (I should like to defer a further discussion of family institu-
tions to a later point in this book.)
Looking back from 1850 to 150, two simple facts disturb the
image of a unitary China ruled by ethnic Chinese. For 60 per-
cent of those 1,700 years, China was ruled in whole or in part by an
avowedly non-Chinese ruling dynasty. For 40 percent of that time,
regional dynastic states shared rule over Chinese-inhabited territory.
Only for 40 percent of the time did a Chinese ruling house rule
all of a unitary China (the end of the Later Han, the Jin, the Sui,
the Tang, and the Ming). So if there were a typical year between 150
and 1850, it featured either a non-Chinese ruler, a divided political
landscape, or both. But toward the end of the span, that is, from 1279
to 1912, China was more typically united.
 Introduction xiii

The institutional profile of China as compared at those two moments,

1,700 years apart, suggests that there was always something slippery
in the interface between government and people. No amount of fine
tuning in the internal ordering of government in the interim man-
aged fully to surmount this problem.
Confucian theory laid down that legitimate government enjoyed
a mandate from Heaven, a mandate that was contingent upon the
rulers ability to instruct the people and to protect and preserve them
from human and natural harm; Heaven would punish serious fail-
ures in these respects by licensing the people to overthrow ruler
and government and replace them with a new regime with a new
mandate. Although the leaders of the great rebellions of 184, 1351,
and 1850 espoused apocalyptic doctrines of Daoist, Buddhist, or
Christian origin, respectively, they still gathered their forces within
the standard Confucian paradigm of celestially mandated dynastic
change and hoped to found dynasties. Whereas such giant outbursts
were rare, smaller ones were not uncommon, and the leaders of
many of them also found it expedient to announce an intention to
found new dynasties.
What of the interface between government and society? Over
the very long term, there were gross similarities and some subtle but
important differences in the ways in which the Han and the Qing
met the people. In the Later Han, and in every subsequent dynasty
down to and including the Qing (founded in 1644), the lowest unit
in the control of the central government was the xian (commonly
translated as county)of which there were 1,179 in 140 CE, serv-
ing a population of some fifty million, and some 1,360 in 1850, rul-
ing a population in the 300400,000,000 range. Qing magistrates
thus had five times the burden of their Later Han counterparts.
County magistrates were the bottom tier of central appointees in
both the Han and Qing. In the Han, the magistrate controlled some
hundreds of subordinate officials and clerks, organized into around
a half dozen specialized bureaus for litigation, markets, and the like;
below the counties lay approximately 36,000 districts, and under
them 1229,000 communes, each with officials in charge of polic-
ing, tax collection, and other such tasks. All these officials and func-
tionaries appear to have been salaried. In 1850, we find, once again,
some hundreds of subordinate officials and clerks, an even larger
xiv Introduction

number of messengers and runners, and roughly a dozen specialized

county bureaus. Yet despite the greatly increased workload, the size
of the county magistrates staff and subordinates changed hardly at
all. What this implied for the nature of Chinas ruling systems will
be noted in due course.
Nor were those all the changes in governing from the Han to the
Qing. In 150, the Han magistrate could reach downward via salaried
functionaries into the subcounty units of district and commune. In
1850 the Qing magistrate had no such reach, insofar as unsalaried
local conscripts staffed the subcounty units. Thus Han local govern-
ment would appear to have been much the denser and the more
closely controlled, with tax requirements that were commensurately
heavy. Qing government was, by comparison, very cheap. The break-
down of the Later Han and the inception of four centuries of dis-
union are strongly linked to antagonisms of long standing between
local government and locally powerful landed magnates controlling
armed serfs. The central government feared and distrusted these
people, yet it was they who provided the means to quell the Yellow
Turban rebels, and it was they who a few years later helped bring
about the Han downfall. In 1850, by contrast, locally powerful lin-
eages never openly challenged Qing hegemony as they mobilized
soldiers and successfully rallied to the dynastys defense against the
Taiping and other rebels.
In both the Han and Qing, the interface of government with society
was powerfully affected (in a positive way, usually) by official recruit-
ment. The recruitment of new men into government, and the means
by which it was done, was an absolutely central issue in 150 and 1850
and at all times in between. Recruitment was no simple matter of hir-
ing and staffing. The government was everyones employer of choice.
Government office was a young mans only sure portal to fame and
fortune. Even a weak central government was in a position to control
levers determining which men, out of a huge press of candidates, were
qualified to hold official positions. Down through all the centuries of
Chinas history, various recruitment channels were made available,
from purchase and inheritance, to the more prestigious and competi-
tive ones: recommendation, and from Song times especially, written
examinations. In 150, the Han central government relied mostly
upon an elaborate system of recommendations, with mainly regional
 Introduction xv

(commandery) officials responsible for recommending a given num-

ber of local men annually (the number depending upon the size of
the commandery population), using as criteria good character and
administrative ability, the latter usually tested by way of probation-
ary assignment. Recommendation appears to have worked very well,
but by around 150, intense rivalries at the highest levels in the capital
introduced factionalism and corruption into the system.
In 1850, it was of course the national written examination system
that had long eclipsed (in prestige, not numbers) all other modes of
bureaucratic recruitment. A colossal industry produced, as reliably as
clockwork, every third year a harvest of some 1,300 provincial degree
winners (juren) and 300 metropolitan and palace degree winners
(jinshi) out of a massive pool of around 90,000 classically educated
eligible men who were themselves but a tiny fraction of the millions
of local (county and prefectural) licentiates. What Han recruitment
by recommendation did not do well, but the Qing examination
system did, was to serve not just as a mechanism for filling official
positions but also as a means for creating and sustaining a national
society of educated local gentry without office but with fiscal and
legal privileges and elite status.
Both the Han and the Qing, as central governments sensitive to
the need to sustain unity over a very large territory, devised regional
quotas to ensure that men from all parts of that territory had roughly
equal chances to participate in the recruitment processes. Society
itself would not have been so fair-minded. That central imperial
authority played the vital role in ensuring distributive fairness is evi-
dent at many junctures, perhaps most dramatically in early Ming: In
1397, the examiners who were southerners awarded jinshi degrees
to fifty-two men, all of them southerners. On strict grounds of merit,
such a result might have been justified, because southerners had
economic, cultural, and educational advantages that the northern-
ers lacked. But the Ming founder reacted in fury, executing several
of the graders. He ordered another exam. This one awarded degrees
to sixty-one northerners. Thereafter, quotas guaranteed that, despite
deficiencies, northerners would never again be excluded.

Pondering China and its ruling profile over a period of 1,700 years
inevitably raises the question of the ideal and often the reality of
xvi Introduction

political unity, and of just how the imperial state acted to maintain
that unity. Political unity was not solely, or even mainly, seen as
imposed by forceby the army and police, essential as those were.
That there should be only one overarching governing structure
encompassing all territory inhabited by people who spoke some form
of Chinese, and were culturally Chinese in other ways, was early on
anchored in the cosmic theory developed in that culture. Theory
posited a known universe consisting of a triad of Heaven, Earth, and
Humankind, with the emperor (Son of Heaven) answerable to
Heaven for the well-being of all the people, and Heaven responding
with favorable or unfavorable portents. With the help of first one,
then another, of the five cyclic powers (fire, metal, water, wood, and
earth), Heaven delivered its mandate to good rulers and withdrew it
from bad ones. This idea was more powerful in the Han than in the
Qing. Over the very long term, the cosmic idea came to be doubted
and partly discredited, though it was too useful ever to fade from
political discourse completely.
The cosmic idea was not restricted to process. It permeated peo-
ples thoughts of structure as well. Indeed, hierarchy, the implicit
framework upon which almost all complex structures hang, was
according to inherited Confucian theory written into both the vis-
ible cosmos and human society. The subordination of moon to sun,
of foreigners to Chinese, of child to parent, of younger brother to
older, of wife to husband, of higher officials to lower, of all officials
to ruler, and of ruler to Heaven was no arbitrary arrangement, but
an invariable principle of nature. People of course often behave in
unfilial or rebellious ways (government existed to encourage good
behavior and repress bad), but seldom was it possible seriously to
challenge the idea of hierarchy. In the absence or collapse of hier-
archy in human systems, chaos (luan) was the usual awful result.
The founder of the Ming dynasty, who early in his life had been
involved with the antinomian Red Turban rebellion of the 1350s,
later recalled that he saw how people of all classes, rich and poor, at
first joined that rebellion against the Mongol Yuan dynasty in hope
of raising their fame and status, or simply from an irresponsible love
of disorder, only to discover that even rebel movements need elites
of talent and leadership, and thus found themselves much worse off
than they were before.
 Introduction xvii

Society needed hierarchy for its own survival, and China pre-
ferred political unity for related reasons. Why unity was desirable was
explicitly discussed by several south China literati in the Yuan-Ming
interregnum of the mid-fourteenth century. Liu Ji, for one, argued
that disunity was contagious because it was a license for every sort of
corrupt evil; a disunited China, even a decentralized China, was a
China ruled by the worst kinds of unsupervised local adventurers and
thieves. But the whole period from 150 to 1850 shows that China in
fact muddled through long periods when political disunion prevailed
and prolonged interior warfare created repeated scenes of bloodshed
and destruction. At such times, educated literati preserved the uni-
fying ideals in written history, classical commentary, or literature,
offered advice based on those ideals to warlords and local rulers, and
perhaps put the ideals into practice at the levels of locality or family.
So one finds the development of a near religion of Confucian filial
piety, replete with stories of martyrs and miracles, during the post-Han
centuries of political fragmentation, when effective ways were sought
to strengthen family hierarchy, even at a time when it was Buddhism
that best explained the universe and, together with Daoism, provided
institutional support and psychic comfort in almost every part of the
once-united realm. Or, in the politically broken world of twelfth-cen-
tury China, the fathers of the Neo-Confucian (daoxue) movement of
south China, frustrated with court politics, created their own hierar-
chies of masters and disciples and, largely circumventing the central
state, experimented with autonomous local institutions to provide
education, famine relief, and security.
It was only reasonable that theory should require that a unified
Chinese state should come under a ruling house of ethnic Chinese
origin, and yet a unified China managed somehow to carry on
even under Mongol and later Manchu rule. The Chinese state car-
ried on because it was possible to argue that virtue and capability
always trumped ethnicitythat Heaven, casting about impartially
for a worthy candidate to rule China, might well discover its man out-
side China altogether, in Mongolia or Manchuria. So the Chinese
founder of the Ming dynasty reasoned about the Mongol Yuan that he
had just destroyed: No doubt Heaven had given Chinggis Khan and
his successors its mandate to rule China, but at the same time it was
also clear that it should never have been Mongols ruling Chinese.
xviii Introduction

That was like wearing shoes on ones head and hats on ones feet.
Worse, they never truly understood Chinas Confucian ideals, and in
the end they proved incompetent to handle the tasks they had taken
on. The Manchus rested their right to rule China on their virtues
and capabilities, and they severely repressed any among the Chinese
who called their ethnic credentials into question. The Yongzheng
emperor went so far as to bring a Chinese critic and would-be orga-
nizer of an anti-Manchu rebellion named Zeng Jing to Beijing, had
him observe for himself how orthodox and well-run Qing govern-
ment was, and then published for general distribution a detailed
refutation of Zengs now recanted and discredited earlier views. (In
1736, however, the young Qianlong emperor thought his father had
been scandalously lenient and had the hapless Zeng executed).
Thus China managed to survive many centuries of political
anomaly. Yet the facts showed that having ethnic Chinese for ones
rulers was not necessarily a blessing either. The last century and a
half of Ming rule features five successive emperors whose pattern of
behavior ranged from the capricious and bizarre to the monstrously
extravagant and cruel. Officials criticized those emperors, but they did
so at the risk of their careers and even their lives. The burst of great
fiction writing in late Ming times centered upon rulers from history
or other figures in positions of authority who, with their relentless stu-
pidity, arbitrary brutality, and heedless self-absorption seem to have
served as surrogates for the actual rulers of late Ming. Yet the power
and influence of the Ming center was such that by their behavior the
emperors unintentionally set models and examples for the rest of the
realm. If emperors could act as they pleased, so might other people.
Breaking the bonds of Confucian orthodoxy, ignoring master-disciple
hierarchies, granting a measure of legitimacy to commonplace greed,
and endorsing the passionate pursuit of ones heartfelt desires, what-
ever those might be, became the new direction taken by some leading
scholar-officials and literati, including L Kun (15361618) and such
seventeenth-century esthetes as Qi Biaojia (16021645), Zhang Dai
(1597at least 1684), and Wen Zhenheng (15851645).

Between the imperial court above and the local officials minister-
ing to millions of people below, there was sandwiched the intricate
machinery of central and regional government, staffed mainly by
 Introduction xix

Confucian-trained mandarins, who were the custodians and supervi-

sors of routine administration and simultaneously articulate spokes-
men for what was sometimes called the national right (guo shi),
which is to say they aimed to be constructive critics of erring rulers
and of ongoing high policy. Those were two very different roles, dif-
ficult for single individuals or government departments to keep in
balance. The one role stemmed from Legalist theory, the other from
Confucian.
The routine of government, as embodied in its array of central
offices, did not change fundamentally over the years 150 to 1850. The
topmost (prime ministerial) organs showed some change, however.
At times the prime ministerial function lay in the hands of one indi-
vidual (in 1380, Ming Taizu abolished that arrangement for all time);
at other times, the function was divided between two or more men. In
150 there was a grand commandant with a large array of supervisory
bureaus under his control, but he was also but one of a troika of high
officials, with whom he shared censorial and advisory duties. Below
the troika sat the nine ministers, each with special functions and
each with large staffs, not controlled by the troika, but subject to its
oversight. One main difference between the Nine Ministries of the
Later Han and the analogous Six Ministries of the Qing was that the
former included palace functions and the latter did not. But Rites,
Justice, Revenue, and War were ministries common to both. The
Qing also had ministries for personnel and public works.
A torrent of documents coursed through both governments:
reports, impeachments, and advice rising upward, and imperial
edicts, judicial decisions, directives, and other pronouncements
flowing downward. In Later Han, a large Imperial Secretariat han-
dled most of the documentary traffic. In Qing, which is known in
extraordinary detail, a sophisticated system of paper flow after 1736
divided official business into routine and nonroutine, with a new
body, the Grand Council, handling the latter with remarkable effi-
ciency and speed.
Later Han central government consisted of a large and confus-
ing agglomeration of offices; by contrast, Qing central government
appears rational and streamlined, with palace functions clearly sepa-
rated from the rest. Both governments established ranking systems
for officialsthe Han based on salary, the Qing on a numerical scale
xx Introduction

of eighteen grades. Rivalrous imperial distaff families and palace

eunuchs participated in Later Han government; in the Qing, these
elements had been all but disenfranchised. Both governments can
be described as meritocracies, as officials were regularly supervised,
assessed, and reported upon, and were promoted or punished accord-
ingly. The censorial functions of remonstrance, impeachment, and
surveillance (anotable part of almost all of Chinas historical govern-
ments) were diffuse and decentralized in the Later Han, with the
result that late in that period, provincial-level officials had little dif-
ficulty turning themselves into autonomous warlords. In the Qing,
the system of provincial governors, governors-general, and circuit
intendants, with divided and overlapping duties, helped protect cen-
tral authority from that sort of breakdown. As concerns military insti-
tutions, retrospectively one can see that the years 150 and 1850 were
late in the history of both regimes, and their once-powerful military
systems were in decline. The national draft of the Han had eroded,
northern border defense had been subcontracted to various Inner
Asian tribes, and local landed magnates were more and more left
to provide for their own security. Likewise, the Qing Banner sys-
tem of hereditary Manchu, Mongol, and Chinese forces had weak-
ened; and as the Han had had to rely on new armies to deal with
the Yellow Turbans, so too the Qing had to allow the recruitment of
new regional armies to deal with the Taiping and other rebellions.
Whereas in both instances the rebels were defeated, neither dynasty
was able fully to recoup central authority in the aftermath of the
suppression. The Han fell soon after, but the Qing managed much
better to retain the loyalty of the commanders of the new regional
armies.
On balance, then, Qing China was culturally as well as institu-
tionally not all that far removed from Later Han China. Han history
was full of useful examples, and students, scholars, and scholar-offi-
cials in Qing times regularly studied it. Indeed, it was Han historians
who created the model for organizing and writing dynastic history, a
model faithfully followed in later times, down to the Qing publica-
tion of the Ming dynastic history in 1739, and of the Qing dynastic
history by the Republic in Taipei in 1962. Between 150 and 1850,
no challenge was threatening enough to prompt a complete insti-
tutional remodeling of the country. Gunpowder weapons, invented
 Introduction xxi

in China in the thirteenth century, and reintroduced in improved

form by the Portuguese in the sixteenth and seventeenth centuries,
failed to have the dramatic impact in China that they had had in
late medieval Europe. At the very same time, unprecedented quan-
tities of Japanese and South American silver poured into China in
exchange for huge shipments of silk and porcelain, and this did cause
some significant economic and fiscal changes, but never enough to
call the whole inherited institutional apparatus into question. The
structural soundness of it all is striking. Chinas was a simple system,
really: small governments, usually dedicated to downsizing and tax
reduction; grounded in Confucian and Legalist doctrines that were,
at least on the surface, easy to understand and hard to challenge; and
reasonably effective in providing such foundational public goods as
security (an) and good order (zhi).
The 1,700 years of recorded history that lie between 150 and 1850
show that as the centuries wore on, slow but visible progress was
made in refining the tools of a unified political system. There took
place a continual re-engineering of the dynastic system, such that the
Ming and Qing, as compared to earlier dynasties, achieved an unusu-
ally high level of internal stability and unity. Changes occurred in
response to serious challenges. As Liu Ji remarked in the fourteenth
century, new dynasties may be institutionally tailored to overcome
a current set of problems, only to encounter a set of new and unan-
ticipated problems that require new policies and approaches for their
solution. It is this feature of history that lends compelling interest to
all of Chinas past, not just to the more recent parts of it.

It seems convenient to divide this very long historical record into

several smaller segments of time during which the wielders of power
over Chinas landscape dealt, successfully and unsuccessfully, with
the big problems of the times. I have singled out four such segments
in this book. First is the long age of fragmentation and disorder that
begins with the Han collapse in 221 and extends to the Sui reunifi-
cation in 589. The second deals with the golden age of empire that
encompasses the Sui and Tang down to the An Lushan rebellion of
755 and continues into the long postimperial period of increasing
disorder and disunion that lasted to the formal end of Tang rule in
907. The third segment covers the years 907 to 1279, nearly four
xxii Introduction

centuries during which the government of China was divided

among the Five Dynasties and Ten Kingdoms (907975), then fell
short of complete reunification, as first the Liao then the Jin ruled
in the north, the Xi Xia ruled in the extreme west, and the Northern
and then the Southern Song ruled in the center and south. This
appeared to be a permanent tripartition of mainland East Asia. In
something of a surprise move, an external force, the Mongols, reuni-
fied China in 1279. Like the period 221589, the period 9071279
featured innovation in many fields: technological, intellectual, and
political. The final segment takes us from 1279 to 1850, a coherent
unit of time insofar as the Mongol Yuan unification of China estab-
lished a legacy of unity and stability that with a few relatively brief
interruptions has continued down to modern times. The book ends
in 1850, on the eve of the great rebellions that, in conjunction with
the increasing pressures of the industrializing West, eventually led to
the ruin of the Qing, as well as the whole inherited dynastic tradition
along with it, in the early part of the twentieth century.
The governing of China had two very different aspects. One was
structural, bureaucratic, rational, usually orderly and predictable, and
slow to change. The other was emotional and volatile, a playground
for rivalry, ambition, fear, greed, shortsightedness, stupidity, faction-
alism, partisan zealotry, and corruption. There was always dynamic
interplay between these two aspects of governance. Both rode on a
substratum of society at large as well as the productive resources that
the people of China wrung from the landscape. People and resources
grew and shrank and changed over time, with consequences for both
aspects of governance. It all makes for an absorbing story.
 Timelines

1. An Age of Disunion
The fragile, short-lived, and mainly regional northern dynasties of this
era are difficult to plot on timelines or maps. Many of their names,
however, are ancient names for regional states during Chinas Zhou
era (1100256 BCE), and thus it may be noted that the various Yan
dynasties originated in the northeast, the various Jin, Zhao, and Wei
dynasties in north-central China, the Qin dynasties and the Zhou
in the west, and the various Shu regimes in Sichuan. The term Six
Dynasties (not used in this book) refers to south China, from the
Wu dynasty down to the Chen.

Later Han, 25220

Failed massacre of palace eunuchs, 168
Yellow Turban rebellion, 184
Rule of Cao Cao, 196220
Cao Caos defeat at Red Cliffs on the Yangzi, 208
The Three Kingdoms, 221264
Cao-Wei, 221266 Shu-Han, 221264 Wu, 229280
Western Jin, 266311
War of the Eight Princes, 301307
Liu Cong destroys Luoyang, 311
Latter Zhao, 330349 Eastern Jin, 318420
Shi Le, r. 330333
Shi Hu, r. 333349
Ran Min (Wei dynasty), r. 349352
Former Yan, 353370
Former Qin, 370383
Former Qin defeated by Eastern Jin at Battle
of the Fei River, 383
Northern Wei, 398534 Liu-Song, 420477
North China reunified, 450 Qi, 479501
Revolt of the Six Garrisons, 523534 Liang, 502556
Western Wei, 534550; Eastern Wei, 534557
Chen, 557587

xxiii
xxiv Timelines

Northern Zhou, 557581; Northern Qi, 550577

Northern Zhou reunifies north China and Sichuan, 577
Sui founded, 581
2. Imperial Grandeur
Sui, 581618
Wendi, r. 581604
China fully reunified, 589
Yangdi, r. 604618
Tang, 618907
Taizu (Li Yuan), r. 618626
Taizong (Li Shimin), r. 626649
Empress Wu, in power 655705
Xuanzong, r. 712756
An Lushan rebellion, 755763
Wuzong, r. 840846
Suppression of Buddhism, 845846
Huang Chao rebellion, 875884
Five Dynasties and Ten Kingdoms, 907960 (975)
3. The Tripartition of China
Liao (9271125) Northern Song (9601126) Xi Xia (9821227)
Liao acquires the sixteen north China prefectures, 937
Zhao Kuangyin (also called Song Taizu), r. 960976
Treaty of Shanyuan, 1004
Wang Anshi in power, 10671085
Huizong, r. 11001125
Jin dynasty declared, 1115 (to 1234)
Jin conquest of Kaifeng, 1127
Southern Song founded, 1127
Treaties between Jin and Song, 1142, 1165, 1208
Zhu Xi Neo-Confucianism made state orthodoxy, 1240
Jia Sidao as Southern Song chief minister, 12591275
Fall of Hangzhou, Southern Song court surrenders
to the Mongols, 1276
Last remnants of the Southern Song destroyed, 1279
4. Permanent Unity Achieved
Yuan, 12711368
Khubilai, r. 12601294
 Timelines xxv

Confucian examination system restored, 1315

Yellow River rerouted, 1351
Red Turban rebellions, 13511354
Chancellor Toghto dismissed, 1355
Yuan breakup and civil war, 13551368
Ming, 13681644
Taizu (Zhu Yuanzhang), r. 13681398
Prime ministers position abolished, 1380
Yongle (Zhu Di), r. 14021424
Voyages of Zheng He, 14051433
Invasions of Mongolia, 14101424
Ming capital moved to Beijing, 1421
Vietnam occupation, 14061427
Zhengtong emperor captured by Mongols, 1449
Jiajing reign, 15221566
Wanli reign, 15721620
Grand Secretary Zhang Juzheng in power, 15721582
Matteo Ricci in China, 15831610
Wars in Korea, 1593, 1597
Eunuch Wei Zhongxian in power, 16241627
Revolts of Li Zicheng and Zhang Xianzhong, from 1628
Li (Shun dynasty) occupies Beijing, 1644
Suicide of Chongzhen, last Ming emperor, 1644
Qing China (1644 to 1850)
Shunzhi, r. 16431661 Southern Ming, 16441659
Kangxi, r. 16611722
War of the Three Feudatories, 16731681
Conquest of Taiwan, 1683
Treaty of Nerchinsk, 1689
Defeat of the Zunghars, 1697
Conquest of Tibet, 1720
Yongzheng, r. 17221735
Qianlong, r. 17351799
Creation of Grand Council, 1736
Extermination of the Zunghars, 1757
Annexation of Eastern Turkestan (Xinjiang), 1765
Heen, in power 17751799
Jiaqing, r. 1796 (1799)1820
Daoguang, r. 18201850
 N

MONGOLIA

MANCHURIA

XINJIANG
(EASTERN TURKESTAN)
KOREA
HEBEI

SHANXI
SHANDONG
QINGHAI GANSU
NINGXIA
HENAN JIANGSU
SHAANXI
TIBET
ANHUI
HUBEI
SICHUAN
ZHEJIANG
JIANGXI

Map 1. Provinces and regions of present-day China.

HUNAN

GUIZHOU FUJIAN

0 100 200 300 400 500 mi YUNNAN TAIWAN

GUANGDONG
0 200 400 600 800 km GUANGXI
 Ningyuan
Pingcheng Beijing

Ha
N

iR
ORDOS er

iv
Liangzhou

Xiangguo

Ye

Yellow River
Luoyang Kaifeng
Changan Xu
Hu Fei River
ai R
ive
r Gaoyou
Xiangyang Jiankang Yangzhou
(Nanjing)
gzi River
Chengdu Yan
Jiangling Red Cliff Hangzhou
Wuchang

Fuzhou

Quanzhou

Canton

Macao

0 100 200 mi

0 100 200 300 km

Map 2. Place-names mentioned in the text.

 Part 1

From Fragmentation
to Reunification
150589

U nprecedented turmoil troubled the four and a half centuries

from 150 on. There was the breakdown of the Han; then the
three-way split of power that followed; then a short-lived reassem-
bly of the broken pieces under the Western Jin (266311); then an
extreme splintering, featuring many competing regimes; and finally
the piecemeal construction of a new central system that was once
again capable of reuniting the country and providing it with security
and order. There is no exaggerating the suffering that all of this polit-
ical turmoil spread in its wake: the roving armies and militias, with
their burning, looting, and massacres; the famines; the enserfments,
enslavements, and mass migrations both voluntary and forced. It is
no accident that the legend of Peach Blossom Spring, where for cen-
turies a happy society of peasants were said to have lived in seclusion,
shielded from all the post-Han upheavals, took final shape under the
pen of Tao Qian, whose life spanned the years 365 to 427, right in
the middle of those disordered times.
The problems that had to be solved before a unified realm could
be reconstituted were too many and too difficult to be solved all at
once. In north China, the mainly Inner Asian rulers and their advis-
ers needed to find ways to get Inner Asian and Chinese elites to coop-
erate; to integrate Inner Asian and Chinese armies; and to create a
resource base sufficient to support central armies powerful enough
to overcome Chinese local magnates with their private estates and
militias as well as hostile Inner Asian frontier tribes. It complicated
matters that there had grown and spread the Daoist and Buddhist


 From Fragmentation to Reunification, 150589

religions, which could either oppose dynastic rule or collaborate

and help strengthen it. South Chinas dynastic states meanwhile did
little to lay groundwork for reunification. North Chinas regimes did
all the work. However, for centuries, stability in the north seemed
beyond reach. But from about 450, the Northern Wei dynasty began
to lay important institutional foundations, which their successors
retained and improved upon. From then on, the advance toward
empire began to gather speed, such that the reunification, when
it came about in 589 with the Sui extinction of the last southern
regime, seems in retrospect almost unaccountably sudden.
The era of fragmentation, aside from the destructiveness of all
its wars, also featured court-centered political violence of extraor-
dinary proportions. In both north and south China, royal murders
and palace massacres were almost a commonplace component of
factional infighting at the highest levels. Away from court, many of
the educated elites of south China (there were few of them in the
north) tried to avoid the dangers of government and instead occu-
pied themselves with literature, religion, and family matters. Elites
insisted on drawing a hard distinction between themselves as an
upper class with deep genealogical credentials and the mass of ordi-
nary people who lacked them. That distinction would blur in future
times. What is notable about this era is the blurring of what would in
the future become rigidly differentiated. The age of fragmentation
featured strong intermingling at the top of the social, political, and
military hierarchies. Members of the best families were among those
who founded dynasties. Socially high-ranking families married their
daughters to the emperors. The emperors relatives held many of
the highest military and political positions. Men often held civil and
military posts concurrently. Likewise there was no clear dividing line
between palace positions and positions in the outer bureaucracy. All
of this blurring would end in later times.
The years from 150 to 589 were not unrelievedly dismal, however.
This was a time for innovation. This was a time when Buddhism
spread through China north and south, bringing with it new ways of
thinking about life and death, as well as temples endowed with land
and other resources that helped promote economic redevelopment
in the countryside and pioneered such new institutions as orphan-
ages, pawnshops, auctions, as well as roads, bridges, inns, and other
 The Unraveling of the Later Han, 150220 

facilities for travelers. In south China, elite migrs from the north
produced a formidable body of new literature that would come to
constitute a major cultural legacy in Tang and later times. But let us
now look at this age of disunion and its resolution in greater detail.

The Unraveling of the Later Han

150220

F or more than four hundred years, the principal East Asian power
was, by every relevant index, Chinas Han dynasty (206bce
220ce). It ruled some fifty million people living mainly in the
northern parts of what is now commonly called China proper.
The capital was Luoyang, which encompassed an area slightly less
than half the size of Manhattan, and at some five hundred thousand
inhabitants may have been the largest city in the world of its time. In
fact, of the fourteen largest cities in the world, nine were Chinese.
Led by hereditary monarchs of the Liu family, the realm was admin-
istered by a ranked and salaried bureaucracy, which through a hier-
archy of prefectures and counties exerted a high degree of control
over the country at large.
But by the latter part of the second century, a long era of prosper-
ity had ended and the regime was clearly in trouble. At the cen-
ter, in Luoyang, a complex power struggle among the civil officials,
the imperial in-law families, and the large corps of palace eunuchs
led in 168 to the exposure of a plot to massacre the eunuchs, fol-
lowed at once by the eunuchs retaliatory purge of everyone even
remotely involved. This purge had serious consequences. Those
purged prided themselves as a pure group, bonded together by
ties of ethical learning and altruistic motives. In exile they took up
pure conversation as free intellectuals, making pure moral judg-
ments (qingyi) of men and affairs. The effect of this on the fate of the
Han is hard to gauge, but in the end it can be said that the best and
brightest of Chinas intelligentsia were not motivated to rally to the
dynastys support.
In the countryside at large, a Han government that earlier was
adept at extending disaster relief gave up on it after about 100CE
 From Fragmentation to Reunification, 150589

By the 150s, famine had created large bodies of starving drifters in

the north China plains. This set the stage for a complete breakdown
of the existing Han order. In the 170s, Daoist sectarians in thickly
populated Shandong developed a vision of a post-Han utopia of
Great Peace (taiping), while in Sichuan a deified Laozi revealed
himself to an alchemist named Zhang Daoling, warned him of the
approaching end of the world, and empowered him to recruit and
lead a community of chosen people. This movement soon turned
itself into an effective and self-regulated paragovernment. Distressed
peasants turned to these sects, or to local magnates, for aid and pro-
tection, or turned to violence on their own. For several months in
the spring and summer of 184, a cataclysm known as the Yellow
Turban rebellion, led by the sectarian Zhang brothers (no relation to
Zhang Daoling), raged across much of north China. It was crushed
by the end of that year not by central forces, but by armies under
the command of local and regional warlords only nominally loyal
to the Han. Further struggle among the warlords brought about the
complete massacre of the palace eunuchs in the same year, 184,
and the capture by first one warlord, then another, of first one, then
another young scion of the Han imperial family. Luoyang was looted
and burned. By 196, the dominant northern warlord, the highly lit-
erate and capable Cao Cao, abandoned Luoyang for Xu, a prefec-
ture some distance east. He placed the last Han emperor there and
through him tried to rule China. Cao used a different city, Ye, as his
own military headquarters.
Cao sought to create a new regime for China. He attracted to his
cause literati who had long been alienated from the Han. He built
a new government using a new recruitment mechanism, the Nine
Ranks and Impartial Judges system, which put the literati and their
pure opinions to use weighing the talents and virtues of the recruits.
He set up military and civilian plantations for defeated troops and
refugees, allocating them lands, and putting their labor and taxable
crops to use by the central state. He brought the Gongsun warlords
of southern Manchuria to heel, and, in Liangzhou in the far west,
he ended a sixty-year rising by frontier generals of the Qiang people.
Trade connections with Central Asia and India were then reopened.
In 216, an independent Daoist state in northeast Sichuan peacefully
surrendered to him. However, these efforts at reunification were
 The Three Kingdoms, 221264 

perhaps too far ahead of their time, and Cao too advanced in years to
sustain them. In 208, in the famous Red Cliffs encounter, the war-
lords Liu Bei in Sichuan and Sun Quan on the lower Yangzi River
joined naval forces and foiled Caos attack. He never attempted
another.
A few months after Cao Caos death, his eldest son and successor,
Cao Pi, under the guidance of the best ritual experts of the time,
arranged for the peaceful abdication of the last Han emperor and
the transfer of the heavenly mandate to a new dynasty, the Wei, with
himself as its first emperor, wrapped in the aura of a Daoist sage. A
partially rebuilt Luoyang was his capital. Warlord Liu (who asserted
kinship with the Han imperial house) broke away and declared his
own legitimist Shu-Han dynasty in Sichuan in 221. The warlord Sun
founded a Wu dynasty in 229, with its capital first at Wuchang (now
Echeng county, Hubei) on the mid-Yangzi and later downstream at
Jiankang (now Nanjing). Thus began the so-called Three Kingdoms
era, immortalized in the famous Ming novel Romance of the Three
Kingdoms, with its intriguing opening thesisThey say that the
realm has a strong tendency to unite after a long period of disunion,
and after a long period of unity, to divide.

The Three Kingdoms

221264

S hu-Han was the smallest and weakest component of tripartite

China, despite the charismatic presence and administrative tal-
ents of its chief official, Zhuge Liang (d. 234). In 264, Shu-Han was
conquered and annexed by Wei. The Shu-Han regimes most famous
scholar and man of letters, Qiao Zhou, helped negotiate favorable
treatment for both his ruler and himself.
The Wu empire was territorially as large as the Wei, but it had a
much smaller population, as the lands south of the Huai and Yangzi
rivers were filling up only slowly with Chinese-speaking settlers. The
Sun family, founders of the Wu state, were native southerners of
undistinguished pedigree. Though administratively weak, and never
able to carry out a proper census, Wu was geographically coherent
 From Fragmentation to Reunification, 150589

and economically well-off, and it successfully defended itself until

conquered from the north in 280.
The Cao-Wei regime, by far the largest of the Three Kingdoms
in population and resources, was gradually subverted from within by
succeeding generations of the exceptionally powerful Sima family.
Sima Yi (179251), his sons Shi and Zhao, and his grandson Yan
(236290) reduced the Cao emperors to ciphers, much as Cao Cao
had earlier done with the Han. The realm was also vexed along its
entire northern frontier by non-Chinese peoples, some nomadic and
some not, over whom the Cao-Wei struggled to impose a degree of
hegemony. Such groups as the Yuwen and Murong in the northeast,
the Tuoba directly north, the Xiongnu in southern Shanxi, and the
Tuyuhun in the far west seem to have been Turco-Mongolian speak-
ers. Near neighbors of the Tuyuhun were the Di and the Qiang,
speakers of languages apparently related to Tibetan. All these border
peoples would play major military and political roles in north China
in the coming years.

The Western Jin

266311

I n the same way that Cao Pi deposed the last Han emperor in 221,
so in turn, on February 8, 266, Sima Yan deposed the last Cao-Wei
ruler and made himself founding emperor of a new dynasty called the
Jin. (Later, it was referred to as the Western Jin.)
The Jin introduced some important policy changes, including
especially a strong tendency to disfavor obscure men of talent (whom
Cao Cao had tried to recruit through the Impartial Judges system),
and instead to cultivate the support and encourage the participa-
tion in government of men from powerful families of established
pedigree. Sima Yan favored his own powerful family as well. Some
fifty-seven princedoms were set up for Sima family members. Each
controlled armies. Some five hundred smaller fiefdoms, carefully
graded, were created as hereditary holdings for other powerful sup-
porters of the Sima. The main goal for the new regime was reuni-
fication. A well-planned campaign against the Wu realm featured
 The Western Jin, 266311 

a six-pronged assault, with five armies attacking from the north

while a flotilla organized up the Yangzi in Sichuan attacked from
the west. Weakened by internal strife, horrific palace murders, and
major defections to the enemy, the last Wu emperor surrendered on
May31, 280, and his realm was annexed to the Jin. China was reuni-
fied. But not for long.
Hardly was China reunified when it began to descend into a
maelstrom of violence and even worse disintegration than that occa-
sioned by the fracture of the Han into the Three Kingdoms eighty
years before.
Reunified China held together under the controlling hand of the
first Jin emperor, Sima Yan. When he died in 290, however, his suc-
cessor was his eldest surviving son, Sima Zhong (259306), unfortu-
nately mentally disabled. Fierce rivalry broke out between the Yang
family of the new emperors mother and the Jia family of his wife.
Their struggle attracted the intervention of the Sima princes, hold-
ers of military and civil governorships as well as fiefs. In 291, a Sima
prince conducted a slaughter of the Yang and their adherents. In
301, prompted by the murder of the heir apparent, another prince,
Sima Lun, massacred the Jia family and declared himself emperor.
That sparked the infamous War of the Eight Princes (301307),
which laid much of north China to waste, and indeed ended Jin rule
there.
This fratricidal war focused first on Luoyang; but after 305, massa-
cre, famine, and disease afflicted other cities, such as Changan, and
much of the north China countryside. Refugees fled to the north-
east, the northwest, and especially to the south.
In part of the northwest, the Li family, of non-Chinese (Ba)
descent, led a mass migration of some two hundred thousand Daoist
communitarians away from local violence and famine south to
Sichuan, where in 302 Li Te founded a Cheng (later called Cheng-
Han) dynasty. Elsewhere in the northwest, a local Chinese magnate
named Zhang Gui (254314) founded a (Former) Liang dynasty
and established a haven for refugees from wars further east. In the
far northeast, meanwhile, Murong Hui (d. 335) gradually built a
Chinese-style administration, welcomed Chinese refugees, set up a
Chinese school to educate his own family and select members of the
tribal elite, and laid groundwork for future expansion. The formal
 From Fragmentation to Reunification, 150589

declaration of a (Former) Yan empire would come about under a

successor in 353.
Thus bits first broke away from the Jin in its eastern and western
extremities and in Sichuan. Worse followed. Xiongnu tribespeople,
long settled in Shanxi as semisedentary horse breeders, found them-
selves drawn into the vortex of the War of the Eight Princes. The
hereditary Xiongnu leader Liu Yuan had had a good Chinese educa-
tion and had spent time in Luoyang in better days. One of the war-
ring Sima princes enlisted his aid in the struggle. But Liu Yuan felt
himself and his people disrespected and mistreated by the Jin. In 304,
at his fort in central Shanxi, he declared his own Han dynasty. In 307
he conquered southern Shanxi. Two large bandit forces roving about
the north China plain joined him the same year. One, consisting of
Chinese refugees and displaced tribespeople, was commanded by
Shi Le, a non-Chinese warlord with a major future role to play. Liu
Yuan died of illness in the summer of 310, whereupon his brother
Liu Cong, also well educated, murdered Liu Yuans designated
successor and seized power for himself. Liu Cong then destroyed
Luoyang in a frightful bloodbath in 311. The Jin emperor fled, but
was caught and put to death. So ended the so-called Western Jin.

A Fractured Age
311450

T he grand leitmotifs that shaped events in the foregoing, and in

what follows, seem to be three in number. One is the very dura-
ble idea of the dynastic state, grounded in Legalist and Confucian
theory, and best exemplified in the great Qin and Han regimes of
the past. Knowledge of the ideas and the history was transmitted gen-
eration after generation through education, often family centered,
and at times also religion based or state supported. The second is
the importance of family and lineage for Chinese and non-Chinese
alike. The potent combination of state and elite lineage, each need-
ing the help of the other, seems to have prevented China from evolv-
ing into either a centralized despotism or, at the other extreme, a
Polish-style feudalism. The third leitmotif is the high value everyone
 A Fractured Age, 311450 

placed on Chinese education, Chinese texts, and Chinese literary

mastery. Chinese and non-Chinese elites alike shared a rough agree-
ment on all these important matters. Thus nothing like an all-out
racial war pitting Chinese against non-Chinese ever took place, even
though language differences, differences in clothing and customs,
differences in hair color and facial characteristics, and perceived
unequal treatment could kindle savage animosities from time to
time, as noted previously, and as will be noted again later.
The period from 304 to 439 in north China, from Liu Yuans
declaring a Han dynasty down to the reunification of the north by
the Northern Wei, is traditionally known as the period of the Five
Barbarians and Sixteen Kingdoms. Eighteen kingdoms would be
more exact. The so-called Five Barbarians were the Xiongnu (under
the rule of the Liu, Juqu, and Helian families); the Jie (the Shi
family); the Xianbi (Murong, Tufa, Qifu, and later the Yuwen and
Tuoba); the Di (Fu, L, and Yang families); and Qiang (Yao family).
In the south, the Chinese Jin dynasty was restored and relocated by
a regional prince, Sima Rui, at Jiankang on the Yangzi in 318. It is
known as the Eastern Jin.
The last Xiongnu ruler of Liu Yuans new Han dynasty, Liu Cong,
died in 318. His cousin Liu Yao, who had conquered Guanzhong
(the Land within the Passes, now southern Shaanxi), thereupon
declared a (Former) Zhao dynasty at Changan. Shi Le, warlord
of the Jie people (their ethnic identities are not clear), earlier an
adherent of Liu Yuan, then declared his own (Latter) Zhao dynasty
at Xiangguo (near Ye, in southern Hebei). The two raced to seize
Luoyang. Shi won and executed Liu in 329. In the year following,
Shi assumed the title of emperor. Foreign warlord though he may
have been, Shi favored schools, liked to have famous old Chinese
texts read aloud to him, and had a few Chinese literati (such as
Zhang Bin) as advisers. But of equal importance in advancing his
career was his association with the non-Chinese Buddhist monk
Fotudeng (also rendered as Fotucheng, d. 348 or 349) from Kucha
in present-day Xinjiang. As a Buddhist convert and church patron,
Shi helped introduce a new element to Chinaan officially spon-
sored and text-based higher religion that was neither Chinese nor
barbarian and was thus able to minister equally to both. Fotudeng
was also useful to Shi as a rainmaker, prognosticator, and political
10 From Fragmentation to Reunification, 150589

and military strategist. Additionally, he gathered about him many

famous Central Asian and Chinese disciples who would carry on the
work of evangelizing Buddhism in China.
Although Shis Latter Zhao unified much of north China, the
Murong clan continued to build its power in the extreme northeast,
while in the extreme west, the Zhang family ruled autonomously. In
Chouchi, a statelet squeezed between the Latter Zhao in Shaanxi
and the Cheng-Han state in Sichuan, Yang Nandi (a White Horse
Di) also ruled autonomously. All three regimes accepted the token
suzerainty of the Eastern Jin at Jiankang. Early in 333, Shi himself
sent a peace envoy to Jiankang.
But none of this presaged an imminent revival of peace, order,
and unity. Shi died on August 17, 333. His nephew Shi Hu seized
power and ruled violently but effectively until he died in 349. Three
years of utter hell followed. Ran Min, a Chinese who was Shi Hus
adopted son and commander of his forces at Ye (the capital), slaugh-
tered the entire Shi family and changed the name of the state to
Wei. Then, calling on ethnic Chinese soldiers, he began a campaign
of genocide against the Jie and other barbarians in and around Ye.
Some two hundred thousand barbarians of both sexes and all ages
are said to have been killed. Into this pit of horror advanced the
armies of Murong Jun from the northeast, their way prepared by pro-
paganda of saving the people from Ran. Murongs cavalry defeated
Rans infantry, and on June 1, 352, Ran himself was captured and
executed. A siege of Ye followed. It fell on September 8. Exactly
what prompted Rans personal and racial hatreds is unfortunately
obscure.
The Murong invasion of north-central China seemed to promise
peace under a stable and relatively well-ordered regime. Urged on by
Chinese advisers, Murong declared himself emperor of the (Former)
Yan dynasty on January 4, 353. Five years later, he moved his capital
southwest to Ye. Preparations were afoot to conduct a major cam-
paign into the far west and so to unify north China under Yan rule.
But Murong died in 360, leaving a very young successor and uncer-
tainties in top leadership. Then came severe famines, forcing a post-
ponement of the campaign. In 370, the Former Yan was destroyed
by an invasion from the far west. The invader, Fu Jian, would try his
own hand at unifying not just north China, but the south as well.
 A Fractured Age, 311450 11

Why were these early hybrid regimes, composed of barbarians

and Chinese, so fragile? Institutionally, military power dominated
everything else, with the partial exception of an emerging Buddhist
church. A major defeat of an army often led at once to the destruc-
tion of the state that sponsored it. Both Chinese and non-Chinese
were enrolled into armies, but as Ran Mins acts showed, mutual
hostilities were not hard to provoke. Nor were the various barbar-
ian armies necessarily friendly with one another. An army led by
the ruler, his family members, and close followers tended (when not
on campaign) to be concentrated around the rulers fort or capital,
walled towns such as Xiangguo or Ye, and there was little permanent
deployment of forces beyond there.
Precious little is known of civil administration, tax-assessing, or tax
collection. Trade was precarious, agriculture insecure. The Chinese-
settled parts of north China seem to have been peppered with small
defensive forts in which rural inhabitants might from time to time find
protection under the hereditary local strongmen who built them. For
the rulers of states, manpower shortage was clearly a problem. Wars
were fought as much to acquire people as to occupy land. One regu-
larly reads of Chinese and barbarian captives being forcibly moved
in the tens of thousands, usually to the rulers capital region. Then
when the state collapsed, many of these captives would migrate back
to whence they came. Commanders and their armies in the field,
Chinese and barbarian, occasionally defected en masse to the Eastern
Jin regime in the south. Disorders propelled many Chinese, perhaps
two million in all, to emigrate from north China altogether and seek
new lives in the south. Northern regimes could scarcely cope with all
the difficulties.

After the Former Yan failure, the next would-be unifier was Fu Jian,
leader of the Di people of the far west and founder of the (Former)
Qin empire. Fu managed to press together a huge but loose coalition
of people and their commanders and articulate a clear purposeto
bring Buddhist salvation to all and rescue humankind from the trou-
bles. In 383, a combined Former Qin force said to number two hun-
dred seventy thousand horsemen and six hundred thousand infantry
began a southeastward assault across the Huai. Incredibly, and for
reasons that are now obscure, it was there stopped by an Eastern
12 From Fragmentation to Reunification, 150589

Jin counterattack. (This was the famous Battle of the Fei River, a
tributary of the Huai, at a site no longer identifiable, due to the
hydrographic rearrangements made by man and nature since that
time.) Almost at once, the Former Qin state broke apart. Murong
Chui declared independence under a (Later) Yan dynasty (384
408). Another Murong declared a (Western) Yan in central Shanxi
and Guanzhong (384396). Yao Chang, leader of the Qiang people,
put Fu Jian to death near Changan in 385 and founded the (Later)
Qin dynasty (384417). L Guang, a Di commander earlier sent by
Fu to conquer Xinjiang (and capture the famous Buddhist monk
and translator Kuma-rajva from Kucha) set up his own (Later) Liang
dynasty (386403) in Gansu. But by far the most important piece of
the Former Qin to declare independence was the Dai regime, later
the Northern Wei, of the Tuoba tribe of Xianbi, based close to the
steppes in northern Shanxi.

Unity in the North: The Northern Wei

398534

T he Northern Wei rulers conquered bit by bit the fragments of

the collapsed Former Qin and by the year 450 achieved the
military unification of north China. By force, luck, and some con-
structive achievements in administration and policy, the Northern
Wei managed to postpone its own collapse until the year 534. This
represented a great leap upward in the strengthening of the northern
system of rule. Important groundwork was laid for the Sui and Tang
reunifications many years later.
The Northern Wei is famous for its colossal Buddhist statuary, first
carved at Yungang in the vicinity of Pingcheng, an early capital in
far northern Shanxi. The city of Pingcheng itself was a monument
to what sheer compulsion could accomplish. Conquered peoples
by the tens of thousands were, over many years, forced to relocate
there in order to raise crops, tend cattle and horses, and live within
the walls of the city, which was planned, built by forced labor, and
divided into a palace quarter, residential wards, markets, and slave-
manned workshops for weapons, woolen and silk manufacture,
 Unity in the North: The Northern Wei, 398534 13

brewing, and the like. The whole was heavily policed. Pingcheng
was like an overgrown military or prison camp.
Eager to harness spiritual power to the dynastic cause and facili-
tate a shift from tribalism to autocracy, the Northern Wei emperor
Taiwu interrupted his predecessors preoccupations with Buddhism
to take up during the years 425 to 450 a flirtation with the Chinese
Daoist church, a descendant of the theocracy created in Sichuan in
the second century, as noted earlier. The churchs celestial master,
Kou Qianzhi, received a divine revelation that Taiwu was destined
to become a perfect ruler of Great Peace. Kous chief disciple, Cui
Hao (381450), served Taiwu as Fotudeng had earlier served Shi Le,
but did much more: besides prognosticating and strategizing as the
Kuchean Fotudeng had done, the Chinese Cui also became prime
minister and virtual ruler, directing the growth of an official Daoist
church and to some extent remodeling Northern Wei government
so as to bring it in line with a long-standing Daoist dream (based on
vague information about the Roman Empire) of a utopia anchored
in peace, justice, and ritual order.
A major rebellion of peoplesChinese, Di, Qiang, and Xiongnu
erupted in Guanzhong (southern Shaanxi) in 448. Emperor Taiwu
personally led the forces that put this down. Discovering that Buddhist
monks had been heavily involved in the rising, Taiwu began an
empire-wide suppression of that faith. The official Daoist church was
text-based and elitist, whereas Buddhism had begun to make converts
among the common people of the Chinese villages and the tribal
settlements. Dissident monks had been involved in not just this, but
also several other eschatological movements against the Northern Wei
government. The suppression was short-lived. The Daoist churchs
extraordinary hegemony ended with Kous death in 448, the execu-
tion of Cui and his adherents in 450, and the assassination of Taiwu
himself in 452. Buddhism then returned to favor, and the Northern
Wei drive toward autocracy was dampened a bit until 490.
From 465 to 490, Dowager Empress Feng presided over an astound-
ingly active phase of Northern Wei history. She was the part-Chinese,
part-Xianbi granddaughter of the last emperor of the brief Northern
Yan dynasty of Liaodong. Her father defected with his family to the
Northern Wei; after his execution, she was put in the harem, and from
there she rose to power on the basis of her own vigor, passion, and
14 From Fragmentation to Reunification, 150589

manipulative skills. She resumed the carving of the Buddha statues

at Yungang, with the Buddhas made to resemble the earlier Northern
Wei rulers, who were now considered reincarnated Buddhas them-
selves. She built monasteries and pagodas in Pingcheng, which by
476 had some hundred structures housing some two thousand monks
and nuns. Most important, the urge to further conquest was post-
poned while Empress Feng encouraged the growth and development
of government. She had aristocratic Chinese officials play a key role
in crafting policies and institutions that significantly enhanced state
power. These were the so-called Taihe Reforms of 472492. They
included procedures for evaluating and promoting regional and local
officials; graded official salaries; a law code; an equal field system of
land allotments; and a system of organizing households for taxation
and control purposes based on multiples of five, each with a respon-
sible leader. Not since the time of Cao Cao, 250 years earlier, or even
then, had anything like this been done anywhere in China.
The Northern Wei did not survive the repercussions of its next big
move, which was the abandonment of Pingcheng, and the rebuild-
ing de novo of Luoyang far to the south in the heart of China proper.
This was another totally planned capital city, a kind of Pingcheng, but
on a grander scale. Recurrent food crises and a desire to be closer to
the southern battlefronts seem to have contributed to this move, but
perhaps the main motive was Emperor Xiaowens desire to escape the
tribal milieu of the far north and place his capital in a Chinese envi-
ronment more hospitable to his autocratic impulses. Forced labor,
and the forced relocation of many tens of thousands of people, again
became the order of the day. So was compulsory social engineering.
Emperor Xiaowen was determined to create stability at the center, and
to that end, he envisioned a world of ranked, fixed, and hereditary
social classes, all living in designated urban wards, or in the countryside
under official control, with officialdom recruited strictly from among
the prominent lineages. There was a Chinese aristocracy of lineages.
There were neither lineages nor aristocracy among the Xianbi, and so
by an edict of 495, Xiaowen created an aristocracy and commanded
that it be integrated with the Chinese into a single pool. Ethnic inter-
marriage at the higher ranks was encouraged. At court, the Xianbi lan-
guage, clothing, and tribal religion were forbidden. Surnames had to
conform to Chinese usages. In line with that requirement, the ruling
 Unity in the North: The Northern Wei, 398534 15

Tuoba tribe changed its own name from Tuoba to Yuan. (The model
for much of this was the aristocratic Western Jin.)
The move to Luoyang was fiercely opposed by many forced evacu-
ees, but especially by the tribes left behind in the north. This problem
was never remedied. The Northern Wei survived barely a quarter cen-
tury in Luoyang before it collapsed in the frontier-wide Revolt of the
Six Garrisons. Until the outbreak of that disaster, sheer central power
achieved wondrous results. Under the direction of another dowager
empress, this one ne Hu (a Chinese aristocrat from Gansu), there
were built in Luoyang over five hundred lavish Buddhist temples, a
spectacular pagoda (the Yongning), and, not far from Luoyang, the stu-
pendous cave statues of Buddhas at Longmen (somewhat replicating
those carved earlier at Yungang). Royalty and high aristocracy enjoyed
huge mansions, fantastic wealth, and native and imported luxuries of
every kind. But not for long. After five years of frontier insurrections,
on May 17, 528, the army of the Jie warlord and erstwhile Shanxi
ranching chief Erzhu Rong marched on Luoyang, seized Dowager
Empress Hu and the baby emperor, and drowned them in the Yellow
River. His men slaughtered upward of a thousand officials and high
elites. Yet again, the horrors of civil war and chaos blanketed north
China. By 534, Luoyang was a total ruin. It would have been hard to
guess then that the reunification of all China lay just a half century in
the future, and that the force behind it would issue from the north.

Another China, meanwhile, lay south of the Huai river system.

Nominally, it occupied a huge territory, reaching as far south as pres-
ent-day Vietnam. In population and resources, however, it was heavily
overshadowed by north China. The south was ruled by a succes-
sion of rather weak governments whose capitals were placed mainly
at Jiankang, but also at times up the Yangzi River at Jiangling. The
Eastern Jin (318420) was the longest lasting of the southern regimes,
although it was never able effectively to register and tax its population.
It suffered under a succession of juvenile and dysfunctional emperors,
as well as from factional struggles among the leading families. Many
of its talented intellectuals and writers were disinclined to lend the
regime their best efforts. Nevertheless, a series of military dictators
managed to make the Eastern Jin a serious player in the all-China
power game. In fact, there was more southern aggression against the
16 From Fragmentation to Reunification, 150589

north than northern aggression against the south. Dictator Huan Wen
conquered Sichuan in 347. There were several short-lived recoveries
of the old capital of Luoyang and of Changan as well. Dictator Liu Yu
captured two northern rulers, brought them as prisoners to Jiankang,
and there had them decapitated. And earlier, in 383, dictator Xie An
foiled Fu Jians attempt to conquer the south. But more usual than war
were diplomacy, alliance making, and peaceful defections back and
forth from one side to the other. The north never developed the river-
borne forces necessary for the conquest of the south, while the south
failed to organize the materiel and manpower needed for siege warfare
in the north. Neither side had bureaucracies capable of permanently
occupying and administering such territory as was temporarily seized
from the other.
The southern warlords and military dictators predicated the legiti-
macy of the Eastern Jin upon its determination to recover the lost north.
But for many, perhaps most, of the several million upper-class northern
migrs who took over southern society and its civil government, the
legitimacy of the Jin and its successors lay not in reconquest but in its
symbolic defense of Chinas literati culture (wen). The migr poets,
philosophers, critics, historians, and anthologists indeed far outnum-
bered and outweighed in importance their counterparts who stayed in
the north. Whereas the literati disdained political ambition, a swollen
roster of official positions served their need for status. They were a ser-
vice aristocracy, their relative social ranking a vital concern that only
a dynastic state could supervise for them. But the dynastic states they
servedafter the Eastern Jin came the Liu-Song, the (Southern) Qi,
the (Southern) Liang, and the Chenweakened as time went on.

Not by Blood Alone:

Steps to Reunification
534589

W hen the Northern Wei collapsed in frontier revolts, two regimes

emerged: one in the east under Gao Huan (d. 547), a frontiers-
man of mixed Chinese and Xianbi heritage and a former follower of
 Not by Blood Alone: Steps to Reunification, 534589 17

Erzhu Rong; the other in the west (Guanzhong) under Yuwen Tai
(d. 556), also a frontiersman, of Xianbi and Xiongnu descent, and
also Erzhu Rongs former officer. Both men ruled behind Northern
Wei puppet emperors. Gaos realm was the larger in population and
resources, so when his son Gao Yang founded his own Northern Qi
dynasty in 550, he had definite advantages over all rivals. But things
went very badly. A Jie generalHou Jingdefected with his entire
army to the Southern Liang (where he was murdered in 552, after his
failed attempt to set up his own dynasty at Jiankang). Far from easing
ethnic tensions, Gao Yang exacerbated them by favoring Xianbi over
Chinese. He suffered from progressive insanity, flew into murderous
rages, and died in 559 at the age of 31. Northern Qi attacks on the
troubled Southern Liang came to nothing.
In the west, meanwhile, Yuwen Tai and a collegium of Xianbi
and Chinese generals ruling behind another puppet emperor fared
rather better. They captured the cruel and violent Southern Liang
emperor at Jiangling and executed him on January 25, 555. They
accommodated a major defector, Xiao Cha, and his whole army by
making him ruler of a Later Liang dynasty at Jiangling. Yuwens gen-
erals returned to Changan with a large number of captives whom
they enslaved. Their main enemy was, however, the Eastern Wei/
Northern Qi. Yuwen deflected two major invasions of Guanzhong
by Gao Huan in 537. In 538 and 543, Gao in turn defeated major
thrusts eastward by Yuwen. Things then settled into a long stale-
mate. Gao held a distinct edge in cavalry and in overall numbers.
Yuwen had to make do with a mixed fighting force of Di, Qiang,
and Chinese, plus a small Xianbi cavalry, but his superior organiz-
ing overcame his disadvantages. Many details of his military reforms
are unclear, but there was a heavy reliance on Chinese peasant
recruits, who were well treated, given Xianbi surnames, and granted
tax exemptions. Their commanders were Yuwens associates, who
had no personal ties to them, a feature that helped ensure the loyalty
of everyone, commanders as well as troops, to the center. The small
Western Wei regime grew with the conquest of Gansu in the 540s
and Sichuan in 553.
The Western Wei ended with the murder of its puppet emperor
and the founding of a new dynasty, the Northern Zhou, on
February15, 557. The nominal founding ruler was Yuwen Tais very
18 From Fragmentation to Reunification, 150589

young son, whose cousin Yuwen Hu, a self-styled latter-day Duke

of Zhou (semilegendary founder of the ancient Confucian political
order), sponsored the reorganization of the new government along
Confucian lines. Of this, the principal architect was a Chinese
scholar-official, Su Chuo (498546), a stern Confucianist whose
program of personal and political morality, economic sufficiency, tax
fairness, and careful official recruitment certainly helped guide the
process of institutionalization that provided a major key to political
stabilization and, eventually, the reunification of China.
The unhappy Northern Qi regime was severely weakened by a
Chen attack from the south in 573, and it was finally destroyed by
its archrival the Northern Zhou in 577. The leader who successfully
pressed that assault despite strong advice to the contrary was Emperor
Wu, who personally murdered regent Yuwen Hu in 572 and was
surprisingly, perhapsthe only effective sovereign the dynasty ever
had. By 577, he controlled all of China except the Chen-held lower
Yangzi. Emperor Wu intended to take it, but in 578 death cheated
him of the opportunity.
Wus son and successor died in 580. In 581, the dead successors
father-in-law, Yang Jian, a Chinese high official and general, hav-
ing killed fifty-nine members of the Yuwen imperial family, deposed
the last Northern Zhou emperor, a child, and founded his own Sui
dynasty (581618). In 588, six Sui armies attacked the Chen from
the north while a navy sailed downstream from Sichuan (essentially,
a repeat of the Western Jin strategy of 280). Early in 589 the Chen
capital, Jiankang, was breached. The city was then completely razed
and made into farmland. The Chen emperor and his court were
transported to Changan. Some four centuries of disunion were over.
China was reunified.
 Part 2

Imperial Grandeur and

Its Aftermath
589907

S ui and early Tang China shared the Eurasian landmass with several
other empires of great size: the Umayyad, the Byzantine, and, a little
later, the Carolingian. Most featured glittering capital cities (Changan,
Baghdad, and Constantinople), cultural and artistic brilliance, and
dominating imperial personalities. All were overstretched, and all
shrank over time. In China, the Sui and early Tang dynasties featured
a top-heavy, centralized infrastructure of military, political, and fiscal
institutions that had been pioneered in the later stages of the preced-
ing age of disunion. These institutions, described in elaborate written
statutes and directives, and staffed by trained bureaucrats, served the Sui
and especially the Tang well for about a century and a half.
The suddenness with which the Sui reunified China in 589 and
immediately went on to copy the early Han and create an extra-
China empire, which is to say a zone of security surrounding China
to the north and west, only to collapse in 618, reminds one of earlier
short-lived unifying dynasties: the Qin (221206 BCE), and of course
the Western Jin (266311). Whereas there are interesting parallels
between the Qin and the Sui (as well as between the Former Han
and the Tang), the Sui and Western Jin do not compare very well,
and that is in some major respects a measure of the institutional and
other changes that had taken place over the intervening centuries.
The Sui rulers were in a position to bring about some spectacular
accomplishments, which the Jin rulers were not, because, thanks
to the panoply of earlier state-centered measures developed by the
Northern Wei and the Northern Zhou, the Sui founder, Yang Jian,

19
20 Imperial Grandeur and Its Aftermath, 589907

was in effective control of the economic and human resources of

the entirety of a China whose population and production were on
the rebound. The puzzling question, then, is why the Sui enjoyed
so short a mandate (thirty-seven years). The answer surely has to be
searched for in Sui leadership at the top: Given immense resources,
what did the Sui emperors do with those resources, once reunifica-
tion was achieved? The institutions of the time, by themselves, could
not save a regime confused as to its goals and bent upon squandering
its resources, immense as those may have been. The Tang founders
resumed command of much the same set of centralizing institutions
as the Sui had commanded, but used them to rather better effect,
and the result was by many measurescultural and artistic certainly,
but also political, religious, military, and administrativeone of the
high points in the whole history of imperial China.
The central machinery of early Tang government seems to have
brought Northern Wei, Northern Zhou, and Sui precedents to some-
thing of a pinnacle of rationality and efficiency. The busy center of
it all featured the emperor, with such palace organs and assistants
as he (or shethe empress Wu) might choose. Outside the palace,
the emperor personally supervised the academicians, the Censorate,
and the three highest executive organs (handling policy formulation
and recommendation; policy review and revision; and policy imple-
mentation, mainly through the Six MinistriesPersonnel, Revenue,
Rites, War, Justice, and Works). There were in addition many other
bureaus. One of them was the Directorate of Education, which had
charge of several central schools, including an upper-tier School for
Sons of State (guozi xue), whose student body numbered some three
hundred young men from the highest social and official classes, and
a lower-tier National University (taixue) for some five hundred stu-
dents of more modest pedigree. The expectation was that they would
all one day be recruited into appropriate positions in bureaucracy.
There were a number of entryways for new recruits into early Tang
bureaucracy, which was divided into nine ranks and grew in size from
fourteen thousand in 657 to around nineteen thousand in 737. In 737,
there were some 137 eligible sons of high officials, guardsmen, clerks,
and capital and prefectural students competing for entry, and in any
given year, about two thousand could be found vying for appointment
to some five to six hundred vacancies. The most prestigious route of
 Imperial Grandeur and Its Aftermath, 589907 21

entry was by way of passing the competitive written examinations, of

which there were originally several kinds, but clearly the most sought
after was the jinshi degree, awarded for demonstrated literary talent.
Examinations at first emphasized administrative competence, but in
736 the management of the system was shifted from the Ministry of
Personnel to the Ministry of Rites (where it remained, until the aboli-
tion of the examination system in 1905). The shift was significant; mere
administrative competence had become too commonplace a require-
ment, and the dynasty was persuaded to recognize and reward the
rarer talents of the realm. The jinshi examination was held annually in
the capital, where about 400 aspirants vied for some 20 to 30degrees.
But the Tang examination system, for all the attention lavished upon it
then and later, had a rather modest effect upon government. Mainly,
the system was a mechanism for giving merit awards to the wellborn
and well educated. More than identifying a political elite, it identified
an intellectual elite.
The good and glorious times of empire lasted for about a cen-
tury and a half. It all ended with the horrific An Lushan rebellion of
755763, after which the Tang was restored, although on a quite differ-
ent institutional footing. Hindsight suggests that the Sui and early Tang
were Chinas answer to a major and prolonged crisisthat is, to a set
of problems dating as far back as the end of the Later Han. The Tang
empire was the handiwork of several generations of statesmen, intel-
lectuals, and rulers, and it succeeded in what it aimed to do: restore
the dynastic state to a position of comprehensive predominance over
every major aspect of existence (economic, social, religious, intellec-
tual, military, etc.) in every part of the country. And from that position,
indeed, the early Tang radiated influences well beyond China. Early
Tang models also exerted influence long after the end of the Tang
itself, thanks to the high quality of the early Tang compilation and
codification of its legal, ritual, governmental, and other systems.
There were weak spots in the institutional tapestry, however.
After the An Lushan rebellion, the rips and tears could no longer
be patched or hidden away. The biggest problem was military.
The key question, not solved until the Song founding in 960, was
how to organize and manage what had become a very large mili-
tary establishment, now that the central state had lost much of the
control it had earlier exercised over its people and national resources.
22 Imperial Grandeur and Its Aftermath, 589907

During the two hundred years leading up to the Song, there were a
series of attempts to settle the matter, and for students of institutional
history, these centuries are of great interest, especially because other
changescultural, intellectual, religious, and administrativewere
involved as well. But it is time now for a closer look at this period of
union, starting with the Sui.

Unification by the Short-lived Sui

581618

T he emperor-dominated Sui dynasty, commanding the resources

of a newly unified realm, embarked on a gigantic program of
internal construction and foreign aggression unparalleled since the
heyday of the Qin in the third century BCE. Changan, Luoyang,
and Jiankang, wrecked in war or razed, were rebuilt as completely
new cities on new foundations. The Grand Canal was dug at great
speed, using massive hordes of conscript laborers, to link the rice
of the lower Yangzi (Jiangnan) with Luoyang to the northwest and,
from there, northeast to northern Hebei. A great wall was built, or
rebuilt, using the fortifications already put in place by the Northern
Qi and the Northern Zhou regimes.
The Sui founder, Yang Jian (Wendi, r. 581604) exhibited a furi-
ous energy. He also sought some kind of superordinate doctrine for a
dynasty that, for all its power, did not command much love or loyalty
on its own merits. He gave an early nod to the Daoist church. For
a while he favored a national system of Confucian schools; then,
disappointed in the results, he abolished it in 601. Finally, he turned
to Buddhism, specifically southern Chinas Tiantai sect, as a source
of Sui legitimacy. Wendi became Buddhisms great defender and
foremost patron. His model was nothing Chinese, but Indias King
Asoka. At the practical end of things, as Su Chuo served the Northern
Zhou, so Gao Jiong (555607) served the Suias strategist and chief
architect of its social, legal, and fiscal systems. In this Su Chuos son,
Su Wei (540621), gave him important help. The national census of
the early seventh century showed a healthy 8.9 million households,
46 million individuals, still heavily concentrated in the north.
 The Long-lived Tang, 618907 23

The Sui regime, though not wholly sure of itself, was institution-
ally well grounded and should have lasted for centuries, as did its
successor, the Tang. However, the Sui fell apart during the reign of
just its second emperor, Yangdi (r. 604618). Traditional history, with
some justice, places the blame on him personally. Yangdi was a head-
strong autocrat in sole command of an autocratic system. He became
obsessed with the misbehavior and defiance of the small Korean
state of Koguryo, which sat astride the Yalu River and, as Yangdi saw
it, threatened the security of northeastern China through its alliance
with a new power in the northern steppes, the Tujue (Turks). There
was an unstable situation in the China-Korea-Manchuria triangle, to
be sure. Wendi had made careful plans to invade Koguryo in 598, but
after he lost 80 percent of his three hundred thousand men to storms,
flood, famine, and disease, he was forced to cancel the attack. Yangdi
resumed the offensive four times by land and sea. He led in person
some six hundred thousand men in 612. There were further thrusts
in 613, 614, and finally 615. Logistical breakdowns, food shortages,
and massive desertions forced each expedition to a halt. Yangdis
relentless determinationor madnesseventually turned his own
class of people, the Sino-barbarian elite of northwest China, against
him in rebellion. His own military commanders strangled him in his
headquarters in south China in 618. That ended the Sui.

The Long-lived Tang

618907

B etween the outbreak of the first anti-Sui rebellions in 612 and

the Tang destruction of its last major rival in 621, there elapsed
but nine years of chaos and war. Why was this period of disunion
so short, as compared to the situation from the third to the sixth
centuries? There is no easy answer. At the outset, the splintering of
military power was extreme, with hundreds of independent roving
bandit gangs, local self-defense forces, and autonomous local offi-
cials. By 618, strong leaders had consolidated all these into nine
empire-seeking warlord organizations. Then, in what has been styled
a bandwagon effect, it turned out that the armies led by Li Yuan
24 Imperial Grandeur and Its Aftermath, 589907

and his intrepid younger son Li Shimin held the clear upper hand in
leadership quality, overall strategy, and battlefield tactics. There was
no die-hard resistance to them. In 621, it was all over.
Of course, the rise of the Latter Zhao and the Former Qin earlier
on could also be described as rapid. What was different about the
Tang was its adoption and further development of all the institu-
tional groundwork laid earlier. That, and the happy circumstance
that five of its first seven rulers were capable, and of them, two (Li
Shimin and Empress Wu) were exceptionally capable. Tang power
and its durability owe much to geopolitical context, domestic institu-
tions, and the accident of effective leadership.
Until the Communist Revolution of 1949, no state exercised so
direct a control over the agrarian and commercial life of China as did
the early Tang. It micromanaged much of north China landholding
through its equal field system, first instituted in the Northern Wei,
which aimed to assign abandoned farmland on an equal basis to culti-
vators and their families; to periodically redistribute the land as families
grew larger or smaller; to extract equal amounts of grain, cloth, and
labor services from each unit of land; to prevent the rise of a private
landlord class (the central state was sole landlord); and to ensure the
state a steady income. The great damage the Sui-Tang civil wars had
inflicted upon north China made idle land plentiful and labor scarce,
thus making state management of national agriculture feasible. (Still,
it took decades for the Tang to reestablish the same degree of control
over the realm that the Sui had imposed.) Running this system required
a veritable army of clerks and accountants, and tons of paper reports,
some remains of which have been recovered in the desert climate of
Dunhuang and Turfan in Xinjiang. Central offices maintained con-
stant surveillance over local officials in the carrying out of this system
(monasteries, high officials, and probably all of south China lay outside
the equal fields domain). In addition, Tang officials directly operated
nearly a thousand state farms along the western and northern frontiers
to supply some six hundred thousand garrison soldiers.
The Tang government also controlled nearly all dimensions of
trade: local, interregional, international. Merchants were officially
discriminated against, and most private trade was either discouraged
or forbidden. Local officials ran local markets, setting prices and con-
ditions of sale. The central government monopolized interregional
 The Long-lived Tang, 618907 25

trade, itself transporting huge quantities of grain and cloth over the
Grand Canal every year from south China to Luoyang and Changan.
A maritime trade office at Canton dominated foreign commerce
there, while state-run frontier markets monopolized cloth-for-horses
transactions, mainly with the Tujue Turks.
The main Tang capital was the gigantic planned city of Changan,
begun by the Sui and completed early in the Tang. Though artificial
and located at the edge of a reunified China, it became the worlds
largest city with a population of about one million. There were many
foreign residents living in their separate wards in Changan, and the
city enjoyed trade connections with Tibet, India, and the greater
Middle East. However, the citys main function was as the center of
government. The dominant people included the very large imperial
establishment in its own palace quarter, some two thousand ranked
central officials, plus larger numbers of clerks, soldiers, police, and
students.
The Tang government managed not only the nations agriculture
and commerce, but social, religious, and intellectual life as well.
As Wudi, emperor of the Southern Liang (r. 502549), had done,
so too the early Tang authorized, compiled, and published in 638
and again in 659 a national register of elite lineages whose members
were certified as socially qualified for office. The 638 register recog-
nized 1,651 lineages (sorted into nine grades), and the 659 register
2,287 (recognition more liberally granted on the basis of recent offi-
cial service). This Tang aristocracy lacked legal privileges other
than those granted actual holders of office, however.
Changan was the center of the Buddhist church, although most
early Tang rulers did not embrace Buddhism as fervently as had the
Sui rulers, or Liang Wudi, or indeed Shi Le or Fu Jian. The Tang
founders Li lineage claimed descent from Laozi, which made the
imperial house rather more favorable toward the Daoist church than
the Buddhist. Nevertheless, the Tang state closely supervised the
Buddhist religious establishment and patronized learned sects such as
the Tiantai and the Faxiang. When the monk Xuanzang, founder of
the Faxiang sect, returned from India in 645, for example, Li Shimin
funded his enormous plan to translate 1,347 Sanskrit holy texts.
Secular or Confucian learning was also overwhelmingly concentrated
in Changan under imperial support and patronage. Informal groups,
26 Imperial Grandeur and Its Aftermath, 589907

advisory colleges, and bureaucratic scholarly agencies turned out a

ritual code; a standard version of the Five Classics; dynastic histories
of the post-Han age of disunion (in which Li was much interested per-
sonally; he even wrote some portions); and much else. The early Tang
rulers made no effort to identify and enforce an orthodoxy, however,
and open debate over scholarly issues was not discouraged.
In the city of Changan, in China, and indeed in greater East Asia
as a whole, the one dominating figure was Emperor Li Shimin, known
posthumously as Taizong. In 626 he killed two of his brothers, forced
his imperial father into retirement, and seized direct rule for himself.
The Li family were members of the same Sino-Xianbi aristocracy
that had produced the founders of the Northern Zhou and Sui dynas-
ties (Taizongs paternal grandmother was a Xianbi; his mother was a
niece of Emperor Wu of the Northern Zhou). The ethos of that aris-
tocracy featured hunting and war; to it Taizong, who was highly liter-
ate, added an intellectual bent. For the first dozen years of his reign,
he was accessible and collegial, in contrast to the remote despotism
practiced by the failed Sui Yangdi. After 636 or so, and down to the
end of his life in 649, Taizong became autocratic, often sidestepping
the regular administrative organs in favor of special operations of his
own. Taizongs son and successor Gaozong (r. 649683) was troubled
by lifelong health problems, but his consort, Empress Wu (d. 705),
proved extremely capable. She disfavored Changan and ruled mostly
from Luoyang, which was rebuilt yet again, and at great expense.
Changan, and Tang China, reached their apogee of power, prestige,
and cultural brilliance under Xuanzong (r. 712756). A long succes-
sion of effective rulers, plus strong institutional support, explain the
early Tang success and the high visibility of the Tang in the world of
its time and in history to this day.

The Tang as Empire

T he early Tang rulers, being history minded, looked back to the

Former Han as a yardstick for their own endeavors in empire
building beyond China. The Former Han, however, had waited
several generations after its founding in 206 BCE before it began
pursuing empire. Tang empire building began almost immediately.
 The Tang as Empire 27

Power in the steppes was held by the Eastern Turks, or Tujue. To win
their temporary military support, the Tang founder, Li Yuan (later
called Taizu), had gone so far as to accept their suzerainty. But his
son, Taizong, personally confronted the Tujue and in 630 inflicted a
heavy defeat upon them. Large numbers of surrendered Turks were
made to settle inside China and in Changan as a military reserve.
In an impressive ceremony held that same year, Taizong assumed an
additional titleCelestial Khaghan. (This had a precedent of sorts in
the preceding era, when several non-Chinese rulers bore simultane-
ously the titles emperor for the Chinese and shanyu for the Inner
Asians). There followed the destruction of the Tuyuhun in Qinghai
in 634; the conquest of the Xinjiang oases in the 630s and 640s; and
three land-and-sea attacks on Koguryo (644649). As in the case of
the bigger but failed Sui campaigns earlier, the strategic reasons for
the Tang assaults on Koguryo are not entirely clear. Likewise, despite
one great tactical victory, Taizongs campaigns in Korea ultimately
failed. Taizongs justification for his expansionism was ethical: to
spread righteousness and crush disorder worldwide.
Although large gains were no longer to be had, the Tang success-
fully defended its empire for a century after Taizongs death. Imperial
expansion had been achieved using small numbers of highly trained
offensive fighting forces plus Turk auxiliaries. But defending the
empire was different: defense demanded large armies placed in
permanent garrisons along the frontiers. Permanent frontier garri-
sons had ended Northern Wei rule in 534, and, in due course, they
would vex Tang China as well. Other institutional changes were
afoot, including especially the partial breakdown of the equal field
system. State-managed land allotments became a victim of their own
success, as land available for redistribution in north China became
scarce, and people in great numbers migrated south, beyond the sys-
tems reach. Heroic efforts were made to save the system, particularly
in 723 and 724 by Yuwen Rong. (The system collapsed completely
in the wake of the An Lushan rebellion of 755763).
Over the long term, the greater geopolitical environment more
and more disfavored the Tang empire. In the seventh century,
except for Koguryo, only loosely organized tribal entities such as
the Tujue and Tuyuhun ringed China. In the eighth century, the
Tang faced formidably organized states in the newly risen empires
28 Imperial Grandeur and Its Aftermath, 589907

of Tibet and Nanzhao. The Tibetan threat was so serious, and so

close to Changan, that in 676 the Tang evacuated Korea and south
Manchuria for good in order to face it. A complex infrastructure of
forts, garrisons, and military farms was built up along the Tibetan
frontier. Then, during the years 736 to 752, under the direction of
the aristocrat Li Linfu, a new system of permanent frontier defense
commands was set up, stretching from the northeast all the way out
into Xinjiang in the west. In 749, Li abolished the last remnant of
the inherited self-supporting militia system that had won Tang the
empire. It was now obsolete. Because the commanders of the new
permanent garrisons would necessarily be powerful figures with per-
sonal ties to their troops, Li persuaded Xuanzong in 748 to order
that those commanders should be ethnically non-Chinese since
they made better fighters and lacked potentially dangerous con-
nections in high government circles. The result of this ill-advised
policy change was almost immediately catastrophic. The central
government did not closely supervise the new appointees. In 755,
the commander of several northeastern garrisons, a general of mixed
Sogdian and Turkish descent named An Lushan, rebelled with his
army and declared a new dynasty. For eight years, war raged across
north China. Xuanzong fled Changan. A new Tang ruler, Suzong,
supported by loyalist military governors, defectors from rebel ranks,
and four thousand Uighur horsemen, finally destroyed the last of
the rebels in 763. Changan was damaged. Luoyang was plundered
and burned. Peace was restored, but autonomous military governors,
especially in Ans old bailiwick in the northeast (present-day Hebei),
remained as a problem not settled until the founding of the Song
dynasty two centuries in the future.

After Empire: Reconstructing China

W hen the dust cleared after the An Lushan rebellion, Tang

China was a smaller and a much-changed place. It was no
longer an empire. Xinjiang, Gansu, and Qinghai fell to Tibet.
The Uighurs, who followed the Tujue in dominating the northern
steppes, placed Tang China under their military protection and prof-
ited greatly from the exchange of their horses for Tang silk. Tibet
 After Empire: Reconstructing China 29

allied itself with Nanzhao and looked culturally to India rather

than China. The Uighurs accepted the Manichean religion of their
Sogdian merchant trading partners. Korea and south Manchuria
came under the strong states of Silla and Parhae, respectively, and
were no longer targets of Tang aggression.
But as in the earlier case of the Later Han, so the Tang after its
loss of empire experienced a very likely increase in overall domestic
prosperity. The years after An Lushan saw profound changes in the
way China was administered. No longer did the central government
manage society directly. Instead, it did so indirectly. Some forty new
territorial units called provinces came to be placed between the cen-
tral government and the many preexisting smaller units, the prefec-
tures and counties. Among the rural population there emerged a
new class of private landlords. From 780, instead of the old equal
field assessments, a new imposition called the twice-a-year tax was
laid on property, not people. Amounts were determined by nego-
tiationbetween the provinces and lower units, and between the
center and the provinces. In south China from 760, a different fis-
cal apparatus prevailed, imposed by a special commission sent by
the central government to Yangzhou on the Grand Canal, where it
seized control of salt production, sold the salt to private merchants
at very high prices, and used the proceeds to fund tax shipments and
to pay directly into the imperial coffers. In both north and south,
government control receded, and private landlords and merchants
advanced. National productivity began an upward leap (as it did after
the abandonment of collective farming in 1976).
The Tang central government, despite its reduced role, was still
active after 763. The throne became for a while a rallying point for
an emerging Confucian intelligentsia that was taking advantage
of the competitive civil service examination system both to make
of itself a meritocracy and, as such, to champion political reform.
Several of the later Tang emperors strove to reduce the autonomy of
the provincial military governors and recentralize the realm. Dezong
(d.805) tried to use military force to eliminate the Hebei governors,
but supply problems and a mutiny ended the attempt. An exami-
nation degree-holder, the southerner Lu Zhi successfully steered
Dezong through the crisis and gained undying fame in so doing.
Dezongs sickly son and successor reigned scarcely a year. Palace
30 Imperial Grandeur and Its Aftermath, 589907

eunuchs then enthroned Xianzong (r. 805820), who enjoyed more

success. He recentralized all but Hebei and placed all the central
military forces under the supervision of the palace eunuchs; but his
reign ended on a sour note, as he suffered from elixir poisoning in an
attempt to achieve immortality and became less able to govern.
Eunuch power emerged after the An Lushan rebellion and was in
part a consequence of it. Recruited largely from among non-Chinese
aborigines of the extreme south, the eunuchs were castrated as chil-
dren and came to be schooled in the palace and appointed mili-
tary supervisors, or commanders of the Shence (Divine Strategy)
Army. This army was originally organized to fight An Lushan, and it
was subsequently centralized, well fed, well trained, and freed of all
legal restraints. By 820, there were 4,018 palace eunuchs.
The eunuchs had enemies. In 805, a cabal of inner court officials
tried but failed to overpower them. In 835, a violent attempt by the
officials to remove the eunuchs ended in a horrific bloodbath, as the
eunuchs retaliated in fury and massacred some three thousand offi-
cials in Changan and elsewhere. That decisively ended all lingering
hope that the Tang government could be reformed at the center by
the direct action of scholar-officials.

Confucianism and Buddhism

in the Late Tang

I n the century following the An Lushan rebellion, a handful of

literati, graduates of the examination system, pioneered a revival
of Confucianism. The leading figures were Han Yu (768819), LiAo
(774836), and Liu Zongyuan (773819). Early in the Tang, the
Confucians had been court-centered officials, patronized by rulers.
The new breed were no longer court centered and were no longer
content with the tepid and circumscribed Confucianism of earlier
times. Liu conceded religious truth to Buddhism but aimed to rees-
tablish the primacy of Confucianism as political philosophy. Li and
Han, more ambitious, aimed also to recapture and reassert the inner
core of Confucianism as a supreme guide to ethical self-renewal. All
three, but especially Han, were devotees of what was called ancient
 Confucianism and Buddhism in the Late Tang 31

prose (guwen), an expressive and refreshing change from the by-

now stale standards set in south China in the centuries preceding
the Sui unification and still required in the Tang examinations.
Despite their great celebrity, the late Tang Confucian revivalists did
not dominate the world of their time, at least not politically, save for
occasional spectacular political protests. As noted earlier, most posi-
tions were assigned not to the successful examinations takers, but to
guardsmen, senior clerks, and the sons of high officials. Corruption
was rife and occasionally spectacular. Weak emperors, who often died
young of elixir poisoning, let power gravitate into the hands of finance
experts, eunuchs, or Daoist holy men. In the provinces, many of them
autonomous or semiautonomous, Chan Buddhism prevailed, as the
Chan patriarchs came to merge Buddhism with the Chinese family
order and Confucian ritual.
Authorities in the separatist provinces were able to offer some pro-
tection to the Buddhists in the furious throne-directed destruction of
Buddhism in the years 845 and 846. Led into this wreckage by Daoist
pleadings, by envy of the churchs wealth, and by the long-standing
complaint that Buddhism was a foreign religion, the emperor Wuzong
(r. 840846) was soon able to boast of closing over 4,600 monaster-
ies, confiscating enormous tracts of church land, laicizing two hun-
dred sixty thousand monks and nuns, and reselling or redistributing
one hundred fifty thousand church slaves. All foreign monks were
deported from China. All other foreign religions were closed down as
well: Nestorian churches, Zoroastrian temples, andthanks to the col-
lapse of the sponsoring Uighur khanateManichean temples, whose
scriptures were burned and whose priests were executed. Shortly after
Wuzongs death in 846, the bans were lifted, but irreparable damage
had been done.
In 840, during the reign of this same Wuzong, the steppe-based
Uighur empire had collapsed, sending large numbers of refugees
together with their leaders west into Xinjiang and southeast to the
Tang frontier, where they asked the Tang court for refuge. The Tang,
by now a weak and insecure state, feeling threatened by the Nanzhao
kingdom in the south and Tibet just to the west, hoped to persuade the

1. Chan is sometimes given its Japanese pronounciation, Zen.

32 Imperial Grandeur and Its Aftermath, 589907

Uighurs to leave the border region and return home. But negotiations
proved unavailing, and so in February 843, the Tang, using a force of
Tuyuhun, Shatuo Turks, and other allies, attacked the Uighurs main
camp and massacred them. All Uighurs living in China were ordered
to wear Chinese-style clothing. Uighur shops and houses were confis-
cated, along with the Manichean temples. No power dominated the
northern steppes again until the rise of Chinggis Khans Mongols in
the early 1200s.

The Shattering of the Tang

868907

A n increasingly destructive series of internal revolts that began in

the south in 868 brought about the end of the Tang in 907. The
problem mainly stemmed from the increasing militarization of inte-
rior China after the An Lushan rebellion of 755763. In An Lushans
time, six hundred thousand troops were posted on the frontiers. A
half century later, nearly twice that number occupied the northern
provinces, mainly in the interior. No longer the serflike dependents
of commanders as in the third and fourth centuries, the soldiers of
the late Tang were volunteers, free agents given to mutiny when
their rewards and bonuses were delayed or cancelled. The Pang Xun
mutiny of 868 broke out among Henan troops and their officers sent
far to the southwest to fight the Nanzhao kingdom. Unchecked, the
rebels marched back north to Henan, adding peasants and bandits
to their numbers as they went. The Tang had to call upon four thou-
sand Shatuo Turk horsemen to suppress this affair in 869.
The infamous Huang Chao rebellion of 875884 was not a mili-
tary mutiny, however. It began as a consolidation of gangs in the fam-
ine-wracked area of present-day Henan and Shandong. Huang Chao,
who rose to lead this force, was a civilian, one of many thousands who
unsuccessfully took the civil service examination. Like a swarm of
locusts, the rebels went on a murderous long march through weakly
defended south China to Canton, then back north to Luoyang and
Changan, which they occupied in 881. The eunuch-led Shence
Army had by this time become corrupt and useless, and the emperor
 The Shattering of the Tang, 868907 33

Xizong and his court fled south to Sichuan. In Changan, Huang

announced a new dynasty and held civil service examinations to
recruit officials for it; but his army, lacking a tax base or source of sup-
ply, turned to plunder and massacre, while elsewhere in north China
the military governors reconsidered the allegiance they had earlier
pledged to him. The Tang court obtained the help of Shatuo Turk
leader Li Keyong and his cavalry and secured the defection from
Huangs forces of the commander Zhu Wen and his army. Together,
Li and Zhu drove off Huang Chao, who committed suicide, or was
killed, in the summer of 884. Xizong retuned to Changan. Remnants
of Huangs armies continued the rebellion until 889.
Institutional decline reached a nadir in the late ninth century.
Nominally, the Tang still ruled, but it was at the mercy of Li and Zhu,
who fought each other for control of the court. Autonomous military
governors ruled much of north China. In Sichuan, a former bandit
by the name of Wang Jian set up an independent regime with the aid
of former Tang officials including the Daoist scholar Du Guangting
and the litterateur Wei Zhuang. Elsewhere in the south, former reb-
els against and local defenders of the empire took power: in Fujian, it
was the outlaw Wang brothers from Henan; in Zhejiang, a salt smug-
gler and local defender named Qian Liu. In north China, Huangs
former follower Zhu (originally from a family of classics teachers that
had fallen on hard times) applied the coup de grce to the Tang.
He murdered the last Tang emperor, massacred the eunuchs, and
abandoned Changan to establish his own regime (the first of the so-
called Five Dynasties of north China) at a new locationBianzhou
(Kaifeng) in Henan. Thus began the era of the Five Dynasties and
Ten Kingdoms (907960).
In fact, political breakdown affected not just China at this time,
but all of greater East Asia as well. The Uighur khanate collapsed
in 840, as noted previously. The once-powerful Tibetan empire fell
apart in the 840s and 850s, never to return. The state of Nanzhao in
what is now Yunnan province broke up after 902. Annam slipped out
of Chinas control and from 939 became independent. In the north-
east, the state of Parhae (its rulers of Koguryo stock imposing a Tang-
style administration over the subject Malgal people) disappeared in
926. On the Korean peninsula, a unified realm, Silla, broke into
three warring parts after 889.
34 Imperial Grandeur and Its Aftermath, 589907

The Five Dynasties and Ten Kingdoms

907960

T he Five Dynasties era is poorly served by source material and

is still understudied, and it is not clear how Chinas inherited
unifying values and systems managed to survive all the stress and
turmoil. There is no obvious answer to the question of just what is
(or was) China, or to what extent after a breakdown it naturally
comes together as one unified and coherent civilization. It helps,
however, to compare the late Tang and Five Dynasties period of 907
to 960 with the early medieval scene of 221 to 589.
The south China regional governments of the Ten Kingdoms period
were rivals, yet they were less bent on war than were the Southern
Dynasties of 221 to 589, preferring to use diplomacy in an intricate
game of power balancing. The richest, longest lasting, and perhaps
strongest of the southern group was the Wu-Yue state founded by Qian
Liu in Zhejiang province, which lasted from 907 to 978. Uninterested
in achieving national power, the Wu-Yue rulers were content to
acknowledge the overlordship of the militarized northern regimes,
which accorded Wu-Yue a status just short of imperial. In return, Wu-
Yue funneled lavish tribute to, and facilitated economic relations with,
the resource-starved north. On its own, Wu-Yue conducted diplomatic
and trade relations with the Khitans of the northeast, the three Korean
states, and Japan. But Wu-Yues western neighbor, the Southern Tang,
had even closer relations with the Khitans, hoping thereby to counter-
balance Wu-Yues ties with various dynasties of north China.
The Canton-based Southern Han regime, meanwhile, carried on
a very large trade with Indonesia, as a recent discovery of a shipwreck
makes clear. The ship was Southeast Asian and when it sank it was
carrying Chinese silver ingots, lead coins, ceramics, and other com-
modities now difficult to identify.
Within months of Zhu Wens deposition of the last Tang emperor
in June 907, Wang Jian down in Sichuan declared himself emperor of
a fully independent Great Shu dynasty, with its capital in Chengdu.
He welcomed literati fleeing the disorders in the north and initi-
ated a major cultural program involving the construction of palaces,
 The Five Dynasties and Ten Kingdoms, 907960 35

bridges, and pleasure parks, the inauguration of musical ensembles,

the encouragement of painting, and the sponsorship and publication
of a new poetic form, the song-lyric (ci). Conquered and annexed by
the Latter Tang (one of the Five Dynasties) in 926, the Great Shu
regime rose again when its provincial governor, Meng Zhixiang,
declared his independence in 934. The earlier cultural program was
revived, and the dynasty flourished until it was conquered yet again,
this time by the Northern Song, in 965.
The rulers of the Southern Tang, who laid a legitimist claim to
kinship with the imperial Li family, presented much the same profile
as the Great Shu. Based in Nanjing, the Southern Tang court also
patronized artists and literati, but its administrative efficiency and
military power fell to neglect, and it was invaded and annexed by the
Northern Song in 975. Had the Ten Kingdoms shown a better ability
to entrench and defend themselves, China might have undertaken a
devolved pattern of multistate development, culturally vibrant, and
much different from the long-term drive toward uniformity and cen-
tralization that in fact took place.
In early medieval times, south China was not divided as it was
in the tenth century, but politically united for the most part, and its
dictators harbored desires to conquer the north, which was often in
a state of political fragmentation. South China was commercially
more active than the north, but the north held the great prepon-
derance of population and resources. Centuries later, in the Ten
Kingdoms era, many regional governments divided up south China.
All of them fearfully expected an eventual takeover by one of the
northern powers. In the earlier era, south China witnessed social
conflicts among northern migrs, a southern Chinese population
established earlier, and a large presence of non-Chinese aborigines.
Conflicts also divided Chinese aristocratic lineages from plebeian
cold gates (families of obscure social origin). Then, assimilation
quietly did its work. Centuries later, all these differences seem to
have disappeared. Built up by steady immigration during the Tang
era, south Chinas population grew to equal that of the north, and
its resources became many times greater. Paradoxically, perhaps, a
bigger and richer south China did not at this time translate its new
advantages into political and military power.
 Part 3

The Tripartition of China

9601279

O ne is always tempted to attach the label Song dynasty to the

period from 960 to 1279. The sheer size and cultural weight
of the Northern Song (and later the Southern Song) indeed lend
it importance of the first magnitude. With a landscape the size of
modern France six times over, and as many as one hundred million
people in its eleventh-century heyday, Northern Song China was
unquestionably the largest and richest organized society on the face
of the earth at the time. But it must not be forgotten (the Song rul-
ers certainly did not forget) that there were two other Chinas. The
XiXia in the northwest may have numbered as many as three million
people, half of them Han Chinese, the rest Tangut, Tibetan, Qiang,
or Uighur. The Liao population, hard to estimate, must have been
several times larger, with a dense Han Chinese population in the
sixteen prefectures it acquired in China during the Five Dynasties
era (most of Liao territory lay in Mongolia and Manchuria). (In
the twelfth century, the Jin destroyed the Liao and conquered all
of north China and a Chinese population of some forty million.) A
constant source of irritation for the Song government was its inability
to destroy those competitors and gather all the Sinophone peoples
under one roof as in the Han and the Tang. The Northern Song,
in order to enhance its military power, undertook some extraordi-
narily comprehensive efforts at institutional reform. But the irony
is that the Song labored in vain. Its creativity meant it owed little to
Han or Tang institutional precedents (Song reformers preferred to
bypass the Han and Tang and engage with the remote Golden Age
of Antiquity), and it left little in the way of a concrete institutional
legacy for the future. The Liao and Jin had deep institutional roots

36
 The Tripartition of China, 9601279 37

in the Tang, and it was Liao, Jin, and Xi Xia institutions, not those
of the Song, that influenced the deus ex machina that did reunify
China: the Mongol Yuan dynasty of the thirteenth century.
The big problem that was never solved in the late Tang was
regional military autonomy. With the removal of the Tang from
the scene in 907, it was regional military rulers themselves in north
China who, free to found dynasties of their own, worked out the
institutional mechanisms that effectively brought the military back
under central control. The Song dynasty, founded by way of a mili-
tary coup in north China in 960, was an important contributor to, as
well as beneficiary of, this process.
Then, a new set of problems emerged. With the conquest of south
China, the Song came to rule a population that had approximately
doubled between the sixth and the eleventh centuries, from fifty to as
many as a hundred million. The per-acre productivity of Chinas agri-
culture had expanded as well, indeed creating a large surplus beyond
subsistence that could flow into trade and taxation. There were so
many people and so large a resource base that it made no sense to
try to restore the micromanagerial apparatus of the Sui and early
Tangstate control and periodic repartition of farmland, the state-
run commercial economy, and so on. So, during the first century of
Northern Song rule, its government, about the same size as that of
the early Tang, could create in its part of China (by far the largest
part) a general climate conducive to prosperity and further economic
growth, while requiring but a modest tax income to support its work.
After a century of rule, however, it had become clear to many at
court and among some of the educated elite that the Song was, in
some essential respects, a failed state. Despite its size and wealth,
it was militarily weak. Its inability to overpower the Xi Xia and the
Liao put its legitimacy as a dynasty in question. What to do about
this problem prompted an intense and ever-shifting partisan struggle
that continued right down to the end of the Song altogether in 1279.
But for three centuries, the power situation in East Asia avoided both
the ruinous fragmentation of the post-Han era, as well as the often-
stifling constraints of empire, and resembled closest the competitive
multistate arrangement of Western Europe.
Song China has long had a rather misleading reputation as a wholly
civilian, antimilitarist society, devoted to the Confucian classics and
38 The Tripartition of China, 9601279

the peaceful arts overall. This picture has some truth to it, and the
generally poor record of the Song in its wars with the Liao, Jin, and
Xi Xia certainly reinforces it. However, it is well to remember that
the Song founders were military men; that it was through warfare that
most of China proper was physically reunited by 979; and that the pro-
fessional Song army of some half-million soldiers became by far the
largest armed force on the face of the globe in the eleventh century.
It is also well to note that, especially in the north of Song-held China
proper, there survived a long-standing tradition among the peasantry
and some landowners of violence, martial arts, and a love of the
military life. Marriage ties between the Song imperial house and the
daughters of military officers was sustained throughout the Northern
Song. It was also Song Chinese who pioneered the development of
bombards and firearms. And irredentismthe hope of recapturing
lost territorywas always a compelling issue in both the Northern
and Southern Song. What the early Northern Song rulers did was not
so much civilianize China so as to render it militarily weak; rather, in
order to break the cycle of coups, they fragmented the military insti-
tutionally, using spies, hostages, and imperial directives to divide the
command structure against itself and frequently transferring officers
to discourage the growth of loyalties between commanders and troops.
Whereas it is likely that these measures dampened Northern Song
military effectiveness, bad strategic decisions on the part of top leader-
ship were more to blame for Song defeats, and it is a fact that no coup
or attempted coup by the military ever troubled the Song.
It is time now for a closer look at the long era of tripartition.

The Liao (9271125), Northern Song

(9601126), and Xi Xia (9821227)

T he Five Dynasties of north China followed one after another, in

a long spate of violence and turmoil, from 907 until the Song
founding in 960. Through this violence there was gradually forged a
political-military system of a new type that put an end to the provin-
cial autonomy that, in mild or severe form, characterized China, the
northeast especially, ever since the An Lushan rebellion of 755763.
 The Liao, Northern Song, and Xi Xia 39

In essence, what happened was that as one provincial military gov-

ernors personal army and administrative machine kept expanding at
the expense of his rivals, it became the nucleus of a wholly central-
ized new order, crystallized in the Song founding of 960.
That development, though important, does not paint the whole pic-
ture. If there were any paramount power in the fractured world of tenth-
century East Asia, then it was none of the Five Dynasties of north China,
but the confederation led by the Khitan (Qidan) people of Manchuria,
whose leader, Yel Deguang, announced the founding of the Liao
dynasty in 927. This confederacy, in which the Khitans were an ethnic
minority, was highly complex institutionally, with no exact parallel in
the past. It controlled several different non-Khitan tribes in the steppes
as far west as the Altai, as well as the former subjects of the Parhae state,
Malgals or Jurchens for the most part, in south Manchuria. The Liao
raided China on many occasions, bringing north captives whom it set-
tled in some forty walled Chinatowns (Hancheng). Gradually, there
evolved parallel systems of rule: one Liao government, mainly military,
that ruled the tribal populations of Inner Asia and another Liao govern-
ment, principally civilian, that replicated much of the old Tang order
and ruled the settled Chinese population.
The chiefs of this confederacybeginning with Yel Abaoji
(872929) and continuing to his son Deguang (902947), Deguangs
nephew Yuan (918951), and son Jing (931969)were caught in a
cultural dilemma: How far to lean toward the Chinese model, how
far to retain steppe traditions? The chiefs chose to intermarry exclu-
sively with the Xiao clan of Uighur Turkish origin. They chose not
to settle in any of the several capital cities they built, but rather to
keep on the move, nomad style, in an elaborate tent complex. Abaoji
tried, unsuccessfully, to end the consultative tribal system of gover-
nance and replace it with the patrilineal dynastic system of China.
But many Khitans considered his son and designated heir Bei as too
culturally Chinese and not enough of a fighting leader, and so his
mother and other Khitan dignitaries denied him the succession.
(Eventually, Bei fled to China, where he was murdered in 937).
The Liao, despite its difficult bipartite structure, was strong enough
to meddle for years in the unsettled power situation on its southern
frontier in China. This meddling reached an apex in the period from
936 to 938, when Shi Jingtang, the Shatuo Turk founder of the Latter
40 The Tripartition of China, 9601279

Jin dynasty, handed over sixteen frontier prefectures to the Liao in

exchange for Liao aid in his overthrow of the Latter Tang. A few years
later, however, Shis successor repudiated the Liao. A four-year war
followed, and in 947 the Liao army, led by the ruler Yel Deguang,
triumphantly occupied the Latter Jin capital at Kaifeng. But after five
months occupation, the Liao had to evacuate north China altogether,
never to try to occupy it again as long as the dynasty lasted (to 1125).
The Liao forces were harassed by armed resistance, and lacked the
personnel, resources, and carefully drawn strategy to bring about a
permanent occupation of all of north China. The sixteen prefectures
were as much of China as they cared to try to control.

The Northern Song founder, a general by the name of Zhao

Kuangyin, conducted what turned out to be the last successful
military coup in Chinas dynastic history when he extinguished the
Latter Zhou in Kaifeng in 960. With that act, all autonomous mili-
tary governorships and commands were at last extinguished. From
960 to 979, the Song postponed a confrontation with the Liao over
the sixteen prefectures, while it conducted an intensive series of
diplomatic and military offensives to bring about the annexation of
nearly all the rest of China.
The founding of the Khitan Liao state in the northeast preceded
in date the Song in the center and south, and the Song preceded
the founding of yet a third principal East Asian power, the Tangut
Xi Xia state (9821227) in the northwest. Like the Liao, the Xi Xia
dynasty and state were Chinese in general format, but polyethnic in
composition. The Xi Xia, however, did not adopt the Liao system
of two governments. Founded by Li Jiqian, the Xi Xia expanded to
include northern Shaanxi, Ningxia, and the Gansu corridor, at the
west end of which it adjoined Uighur territory. Thus, whereas the
extreme fragmentation of the Five Dynasties and Ten Kingdoms era
was largely resolved by the late tenth century, the net result was no
new unitary Han or Tang, but rather a three-way division of East
Asia, with large Chinese populations inhabiting not just the Song,
but Liao and Xi Xia territory as well. Institutionally, the Northern
Song dynasty constitutes a peculiar phase in Chinas history, without
either a distinct ancestry or a compelling institutional legacy for
the future. In the eyes of many of its inhabitants, it should not have
 The Liao, Northern Song, and Xi Xia 41

fallen so short of the Han and Tang standard. For all its wealth and
power, the Song did not control an empire, did not govern all of East
Asias Chinese-speaking populations, and, at the outset, even within
its own boundaries, did not reach effectively into society at large.
The three East Asian powers of the tenth to twelfth centuries were
each compact, stable, and well organized as they faced each other in
a kind of prolonged stalemate. Intermittent war between the Song and
Liao over control of the Chinese-inhabited sixteen prefectures ended
with the famous Treaty of Shanyuan of 1004, one term of which stipu-
lated that the Song court would provide generous annual subsidies
in silver and silk to the Liao court. Despite controversy, peace on the
frontier prevailed for nearly a century thereafter. The Song disdained
the Xi Xia and refused to accord it equal treatment with the Liao.
The great wealth and stability of Northern Song society has been
attributed to the managerial expertise of its founding generations of
high officials, men from sixty families from all parts of China who
made their homes in Kaifeng, the capital, and who constituted an
intermarried hereditary elite, quite similar to the mid-Tang aristoc-
racy. But these men from privileged families were subjected to the
rigors of an official examination system designed to identify early
on men with legal, financial, and other special talents and direct
them into the appropriate career channels. A friendly and predict-
able national climate for economic growth was the result. Moderate
tax rates, a huge and gradually increasing supply of copper currency,
safe conditions for interregional and international maritime trade,
and government encouragement of and investment in industry
(coal mining and iron production especially)policies based upon
administrative ideas first developed in the fourth century BCEall
combined to produce an astounding surge in national prosperity.
However, within a century of the Northern Song founding, official
complacency and corruption began to set in. Also, newer generations
of men, particularly from the south, beneficiaries of the educational
opportunities that prosperity and government policy had created,
began passing the civil service examinations in large numbers and also
began challenging the honesty and competence of top officialdom.
The first wave of these men was born around the year 1000. The means
through which they sought to exert themselves was not management
expertise, but rather classical learning, moral refinement, and critical
42 The Tripartition of China, 9601279

judgment. They were the champions of a Confucian revival, com-

monly called Neo-Confucianism, that originated in the late Tang (as
noted earlier) and was now resumed, to deepen and spread through
the whole remainder of the Song period and beyond.
The new men, many of them southern Confucian revivalists,
demanded reforms, and their reform demands escalated over the
years as a sense of unresolved crisis in national security mounted
with the Songs chronic inability to overpower the Liao and Xi Xia.
This led to the astounding remaking of Song government conducted
by the southern degree-holder and chief minister Wang Anshi during
the years from 1067 to 1085. Enrich the state, strengthen the army
was the slogan. With the warm support of Emperor Shenzong, Wang
justified on Confucian grounds a holistic vision of an all-powerful
state by developing through commentary an update of the classic
Rites of Zhou, a text last seen put to use for reform purposes by Su
Chuo in the sixth century. Wangs idea was that government should
have a direct hand in fostering economic growth nationwide, and
that the revenues derived from enhanced growth should then sustain
a greatly improved military capability. Much of this New Policies
program was accomplished using special task forces and new revenue
sources that bypassed a largely hostile regular bureaucracy. By 1082,
Wangs Council of State had charge of finance, personnel, military
planning and operations; mining and metallurgy; salt and tea pro-
duction; rural credit; and wholesale interregional and international
sea trade. A land resurvey was ordered for north China. A national
militia was developed to replace eventually the huge, expensive, and
ineffective professional army inherited from the Five Dynasties. The
examination system was completely retooled to include a national
school system whose curriculum was based in Wangs New Studies.
This curriculum heavily discouraged the study of history, and criti-
cism and debate were not welcome. Wangs critics grounded their
dislike of the New Policies in their comprehensive and detailed
study of Chinas history. Many conservative opponents of Wang were
dismissed, or resigned, or retired in exile to Luoyang.
Uppermost in Shenzongs mind was the purpose of all this
reform: enhanced Song military leverage over the Liao and Xi Xia
and indeed everywhere else along the Song land frontier, includ-
ing Sichuan, the Dali kingdom in Yunnan, and a now independent
 The Liao, Northern Song, and Xi Xia 43

Annam (Vietnam). Fights with the Liao and Xi Xia in the 1040s
had gone badly and had ended in the payment of increased Song
subsidies to both. Only occasionally allied, the Liao and Xi Xia more
often fought each other, which helped to create opportunities for
Song aggression against the weaker party, the Xi Xia. Song-Liao hos-
tility over border demarcation flared up from 1074 to 1076 but was
settled by negotiation. The Song and Xi Xia had never agreed on a
common border, and Song annexationist wars over border territory,
which broke out three times in the 1080s and 1090s, escalated into
disastrous attempts to destroy the Xi Xia state altogether. Wang was
in favor of deferring war until his reforms were further advanced.
It was Shenzong, in combination with palace eunuchs, hereditary
military generals, and hawkish civilian officials all looking for career
advancement, who agitated for war.
Shenzong died in 1085. A regency, headed by the dowager empress
from 1085 until her death in 1093, sponsored the removal of the par-
tisans of the New Policies and their replacement by a loose grouping
of conservative opponents, among them such famous names as Su
Shi, Sima Guang, and Cheng Yi. Many of the New Policies were
rescinded. Aggression against the Liao and Xi Xia was toned down.
Then, during the reigns of Zhezong (to 1100) and Huizong (1100
1125), yet a final partisan turnover brought on the restoration of the
New Policies faction, the blacklisting of some three hundred officials
of the conservative opposition, and the suppression of all internal pro-
test and policy debate. Chief minister Cai Jing ran the government
for many years. When old age diminished the chief ministers vigor,
Huizong relied more and more on palace eunuch Tong Guan. The
reformists not only revived much of Wangs centrally directed program,
but went even further. They expanded a costly nationwide system of
schools that came to enroll some two hundred thousand students, all
following a New Policies curriculum that dropped poetry and history
to focus on Wangs Confucian commentaries, Huizongs commen-
taries on the Daoist classics, and the Song legal code. The reformists
also put into effect programs that made the central state something
of a national welfare agency: poorhouses, clinics, and free cemeter-
ies absorbed unknown but possibly large outlays of revenue. The late
Northern Song was thus a kind of brief golden age, and the emperor
who presided over it was a cultural impresario, with deep personal
44 The Tripartition of China, 9601279

interests in architecture, garden\ building, painting, calligraphy, music,

medicine, and Daoism.
But there was another side to the Northern Song. From 1116,
when the eunuch Tong was appointed to the Council of State, the
dynasty resumed the aggressive military stance it had assumed ear-
lier under Shenzong. The Liao confederation in the north began
unraveling in the early 1100s, wracked both by tribal rebellions
and by murderous intrigues at court. From 1112, and over the next
decade, the Jurchen tribes of northern Manchuria, led by Aguda,
conducted a series of fierce wars against the Liao, conquered more
and more Liao territory, and declared itself the Jin (Gold) dynasty
in 1115. The Song court, thanks to its many envoys to the Liao and
Jin, was well aware of these developments, and, prodded by Tong,
unwisely decided to turn the Liao distress to its own advantage and
negotiate an agreement with Aguda, so that the Song might then
recapture the sixteen prefectures lost to the Liao in 937. However,
the Song army, partly because it was distracted at this time by war
with the XiXia and by the dangerous Fang La rebellion in Zhejiang,
performed wretchedly in its attack on the Liao. Aguda destroyed the
Liao on his own and captured the sixteen prefectures, a fait accompli
the Song had to accede to in 1123. That such a blunder could hap-
pen has been attributed to the silencing of all internal debate that
the Northern Song reform faction had long imposed.
What came next was fatal for the Northern Song. In the autumn
of 1125, for reasons that are not clear, the Jin invaded north China in
force and besieged Kaifeng. The siege was lifted diplomatically, but
only after the Song court agreed to ruinous terms. There followed a
complete turnabout in central government. Under the leadership of
chief minister Li Gang (10831140), the officials persuaded Huizong
to abdicate in favor of his son, known posthumously as Qinzong. A
dramatic demonstration by patriotic students of the national univer-
sity helped bring about the execution of Tong Guan and the removal
of other influential eunuchs as well. The reform faction, discredited
by its yielding to Jin demands, was turned out of office. The Song,
hoping to rely on newly recruited militias, rejected the agreement
with the Jin and vowed to stand and fight. They had no time to pre-
pare, however, and the Jin invaded again. Kaifeng fell in January
1127. The Jin army thoroughly plundered it. Huizong, Qinzong,
 The Jin, Xi Xia, and Southern Song 45

and nearly the entire imperial household, numbering in the thou-

sands, were marched off in captivity to Manchuria. A Song imperial
prince, Gaozong, managed to head a new government that eventually
decided, for safety reasons, to relocate south, not to Nanjing, but to
Hangzhou, as a temporary capital. Thus was founded the Southern
Song dynasty (11271279). Inconclusive wars with the Jin over con-
trol of north and central China continued until 1142. Tripartite East
Asia consisted now of the Jin (with a much larger presence in north
China than the Liao had had), Southern Song, and Xi Xia.

The Jin (11151234),

Xi Xia (9821227), and
Southern Song (11271279)

T he Song dynasty lost a third of its territory with its relocation to

south China. The central government had to be built again from
the beginning. The military likewise had to be completely recon-
stituted, mainly by recruiting former bandit armies. Fundamental
national security policywhether to try to retake north China or
remain quietly in the south as a Jin vassal statewas never conclu-
sively determined but shifted this way and that until the very end, the
Mongol conquest of the 1270s.
The Jurchen Jin, too, were divided over fundamental policy.
Divisive issues of cultural identity also troubled the regime. The Jin
was heavily committed to north China, and partly for that reason it
was less able than the Liao to control Inner Asia. The Tangut Xi Xia
state found itself cut off from the Southern Song, owing to the Jin
conquest of north China. In this circumstance, the Xi Xia became
less warlike and focused its efforts on internal development. Like
the Jin, it was a multiethnic state, but with a more effective pol-
icy of managing interethnic relations. With a nod to the Tibetans,
the XiXia became a Tantric Buddhist theocracy; with a nod to the
Chinese, it simultaneously developed Confucian institutions and
civil service examinations. Great efforts were made to translate
Buddhist scripture and classic Chinese texts into Tangut. A massive
46 The Tripartition of China, 9601279

Tangut-language law code combined Tang law with Tangut tribal

tradition. But despite this cultural efflorescence, Xi Xia isolation and
withdrawal more and more left the Jin and Southern Song in a two-
way power standoff in East Asia.
The Jin made deep but unsuccessful probes south across the
Yangzi in pursuit of Gaozong and other key surviving members of
the Song imperial family, but the conquest of the south, and the
unification of China under Jin rule, was never the Jins driving ambi-
tion. It was difficult enough for the Jin to consolidate its hold on the
forty million Chinese in the north. Eventually, Jurchen and other
fighters and their families, perhaps some six million people, were
moved from Manchuria and resettled in special military colonies in
north China.
In 1142 the fighting ended, and a peace agreement between the
Jin and Song was reached. This came about after the Jin had seri-
ously mishandled the many defections of Song generals, armies, and
civil officials, and after Gaozong had disposed of the irredentists on
his side, especially the loyalist general Yue Fei and the leading coun-
cilor, Zhao Ding. The agreement heavily favored the Jin. The two
sides recognized the Huai River as their common border. The Song
agreed to send annual subventions in silk and silver. It also agreed to
accept an inferior posture of vassalage to the Jin, a concession it did
not advertise at home. This was a shaky peace. The resumption of
war always remained open as an option for political factions in both
empires.
In the Jin case, the war issue was connected in a complex way with
a problem that had earlier vexed the Liao: how far to preserve tribal
ways, and how far to change or end them in the interest of Chinese-
style autocratic and bureaucratic state building? The tensions here
led to appallingly bloody acts. In 1147, EmperorXizong (not to be
confused with the Tang emperor also called Xizong)murdered 48 for-
mer Liao Chinese officials; in 1150, Xizongs cousin, usurper Wanyan
Liang (Prince Hailing), killed Xizong, 72 male members of the impe-
rial family, and large numbers of the Jurchen elite. Wanyan Liang
stood on the Chinese side of the issue. From 1150 to 1161, the new
ruler moved most of the Jin ruling apparatus out of Manchuria into
north China and built a new capital on the site of present-day Beijing
as well as a southern capital at Kaifeng. Then, he amassed a huge
 New Developments in the Southern Song 47

army and navy, and in 1161 he took personal command of an all-

out invasion of the Southern Song. This was an act not of considered
policy, but of megalomania. The Khitans and others refused to join in.
In December 1161, just after his forces reached but failed to cross the
Yangzi, officers acting on behalf of imperial clan moderates murdered
Wanyan Liang. That ended the attempted conquest and annexation.
Jin China reached its apogee during the long and generally peace-
ful reign of Shizong (Wanyan Wulu, r. 11611189), who cancelled
the attack on the Southern Song and in 1165 agreed to a new treaty,
this one less favorable to the Jin than the treaty of 1142 had been. By
1207, a very large, perhaps bloated government of some forty-seven
thousand officials ruled north China (under Wang Anshi, Northern
Song bureaucracy had expanded to thirty-four thousand officials in
1086). Jurchen tribal aristocrats dominated the top positions, while
Chinese, recruited through a much-expanded national Confucian
school system and civil service examinations, filled most of the rest
of the positions. Former Liao Chinese were given advantages over
former Song Chinese. Jurchen customs, language, and script were
heavily promoted. Tribal consultative procedures, featuring votes
and formal head counts, were reintroduced at court. During the
reigns of Zhangzong (11901208) and Xuanzong (12131224) there
was extended debate and votes taken on the troublesome question of
Jin dynastic identity. Just where did the Jin fit in Chinas historical
sequence? Did the sequence go from Tang to Liao to Jin, or Tang via
a gap to Jin? What about the Northern Song? Was it the Jin mission,
then, to carry on Tang traditions and to deny cultural legitimacy
either to the Liao or the Song? The debates took place off and on
from 1194 to 1214. No definitive conclusion was ever reached. The
Jin never gained a sure corporate sense of itself.

New Developments
in the Southern Song

L ike the An Lushan rebellion of 755763, the Jurchen invasion of

1126 and the Song evacuation of north China was a watershed
event, a great breaking point in history. The traumatic reduction of
48 The Tripartition of China, 9601279

Song China to a southern rump changed many things, some con-

cretely, some symbolically. Politically, it discredited the centrally
directed New Policies reform programs of Wang Anshi and Cai Jing.
While in the north under Jin rule, the Chinese literati were on the
whole content to follow the traditional literary standards of the Tang,
in south China under the Southern Song there developed an intense
and many-sided debate about the whole future direction of Chinas
civilization. The examination system in both north and south turned
out many more graduates than the bureaucracy needed. The Jin
responded by creating more bureaucratic positions. The Southern
Song did not, and so there emerged in south China an elite society
that was strongly localist and unconnected to Hangzhou, as well as
a sizable Confucian intelligentsia that was deeply concerned for the
future of China but only loosely involved with the central state. One
of the leading figures was Zhu Xi (11301200). Zhu won his jinshi
degree (the highest degree) in 1148 at the very young age of eighteen,
although he spent most of his life not in office, but pursuing his own
agenda of reforming China. This was an ambition as all-encompass-
ing as Wang Anshis, but now no longer involving the central state
but rather the Confucian intelligentsia, in office or out, as the lead-
ing engine of change. In the field of philosophy, Zhu singled out
and annotated the so-called Four Books (in preference to the Five
Classics) as the core curriculum for education; identified an orthodox
line of correct transmitters of the Confucian doctrines down through
time; and elaborated a metaphysics that at last displaced Buddhism
as the ultimate explainer of all cosmic and human existence. In the
area of institutions, he helped shape the intelligentsia into a national
Confucian fellowship focused upon ethical self-cultivation and lead-
ership over local institutions: the privately funded academy, the com-
munity compact, the community granary, and the local militia. The
functions of these institutions had earlier been included in the central
agenda of the Song state. As for the central state, Zhu would make it
the creature of the Confucian fellowshipthe overall guardian of the
new Confucian order and a passive recipient of moral and political
advice, rather than a free agent. This complex of ideas was commonly
referred to as daoxuethe learning of the way. Its opponents called
it specialized learning, private theories, and even false learning.
 The End of Tripartition 49

The Southern Song state, meanwhile, wavered, as did the Jin,

between war policies and peace policies. In the summer of 1206 it
formally declared war and launched a northern invasion that soon
bogged down in wet weather, logistical mishaps, and incompetent
command. A hoped-for revolt of Jurchen-ruled Chinese never mate-
rialized. Peace negotiations in 1208 forced the Song to raise its annual
payments. The war faction, thoroughly discredited, was thrown out
of office. The severed head of its leader was sent in a lacquer box to
the Jin. Neither contestant could foresee in the Mongol confedera-
tion of Chinggis Khan, just then emerging, its own imminent demise
and the rise of a wholly new East Asian system of power.

The End of Tripartition

B attered by unremitting natural disasters (including the cata-

strophic 1194 flood of the Yellow River), and beset by lethal
power struggles at court, the Jin dynasty began slowly to fall apart.
From 1206, Chinese warlords created separatist regimes in Shandong.
Around the same time, a revived Khitan Liao state came to power in
central Manchuria. A separate Jurchen state, its founder committed
to the Daoist religion, formed in northern and eastern Manchuria in
1215. In addition to all this, annual Mongol raids into north China
began in 1211. In 1215, the Jin court felt compelled for safety reasons
to move its capital south to Kaifeng. The Mongols then captured the
Jin central capital (Zhongdu, present-day Beijing).
Until almost the very end, the three powers sharing the Chinese
cultural landscape were more preoccupied with one another than
with the Mongols. There was a Xi Xia attack on a weakened Jin
in 1217, along with a series of Jin attacks on the Southern Song.
Finally, the last Jin ruler, Aizong (r. 12231234), his territory hav-
ing shrunk to Henan, Shaanxi, and parts of Shandong and Shanxi,
concluded peace on equal terms with both the Song in 1224 and the
Xi Xia in 1225. But in 1227 the Mongols destroyed and annexed the
Xi Xia. A yearlong Mongol siege of Kaifeng (12321233) ended in
Aizongs flight south and suicide. Early in 1234, the Jin dynasty was
no more, and the Mongols controlled all of north China.
50 The Tripartition of China, 9601279

The Southern Song court for a long time saw the Mongol attacks
less as a threat than as an opportunity to defeat the Jin and recover
north China for themselves. The court made alliances with the sepa-
ratist warlords in Shandong. In 1233, Song emperor Lizong agreed to
contribute twenty thousand troops and a large quantity of provisions
to the final Mongol assault on the Jin. Then, unpersuaded by argu-
ments that Henan was too depopulated and economically ruined
by years of war to sustain a Song annexation, Lizong sent his armies
north. After eight weeks, the Mongols routed them at Luoyang. So
ended the long, sad story of Southern Song irredentism.
Still, for over forty years, the Southern Song maintained a defense
against Mongol-ruled north China. During that time, the court took
a series of steps, ending in 1241, to make the Zhu Xi school of daoxue
official state orthodoxy. This was an effort at moral rearmament, a
Song attempt to universalize on spiritual grounds what could no
longer be unified militarily. The decision was also prompted by the
Mongols official sponsorship of Confucian study and by their one-
off revival of the civil service examinations.
The standoff with the Mongols began to go badly for the Song in
the 1250s and 1260s, as the Mongols came to understand that siege
machinery and naval forces were going to be the keys to military
success on the battlefronts of China. The Mongols six-year siege
of Xiangyang on the Han River ended in 1273 with the defection
of the defending Song generals and their armies. In the Southern
Song capital at Hangzhou, a controversial finance expert of military
origins, Jia Sidao, was appointed chief minister of state in 1259, a
position he held until dismissed in March 1275. His policy of pay-
ing for defense by ruthlessly confiscating large landholdings in the
Yangzi delta area kindled such anger among high officials and many
Confucian literati that they blamed him, rather unfairly, for the
Song collapse.
Unlike the Mongols conquest of north China, the advance of
Mongol-led armies into south China involved minimal bloodshed
and destruction. Hangzhou was entered peacefully. The Song court
rejected the urgings of some to mount an all-out defense. The
empress dowager and the child emperor surrendered early in 1276.
They were transported to the Mongol capital, Dadu (now Beijing),
and were treated well there. That was not quite the end, however.
 The End of Tripartition 51

Two infant princes and their supporters fled south by sea while the
greatest celebrity of the day, the number-one jinshi degree-holder
Wen Tianxiang, mounted a hopeless but symbolically important
loyalist act of resistance against the Mongols in Jiangxi. Wen was
defeated and captured in 1278; and because he refused to capitulate,
he was executed in Dadu in 1283. Meanwhile, a Mongol-comman-
deered fleet destroyed the last of the fleeing Song court off Canton in
1279. The Song mandate was thus extinguished. The Song legacy,
not of institutionalized power, but of Neo-Confucianism and loyal-
ism, lived on.
 Part 4

Permanent Unity
Largely Achieved:
Yuan, Ming, and Qing

T he last span of history to be covered in this book consists of the

Yuan, Ming, and Qing to 1850, a period of nearly six hundred
years during which only three dynasties, ruling one after the other,
and despite their great differences, built a durable institutional
framework for the governance of China sufficient to hold the coun-
try together under central control most of the time, and, in the Yuan
and Qing, added considerable extra-Chinese territory to it. The
question is, how did these regimes, which were by no means carbon
copies, manage to achieve such an impressive record of stability?
One symptom of the new stability was the choice of a permanent
site for the capital city, built (and rebuilt) at an eccentric spot in the
northeast, where it has remained to the present day under the name
Beijing. Changan, Luoyang, Kaifeng, Hangzhou, and Nanjing,
capitals of past regimes, including long-lived dynasties such as the
Han and Tang, yielded to Beijing, a place without deep historical
resonances, located on the fringe of Inner Asia, and first built up by
non-Chinese regimes, the Liao and Jin. The Ming founder chose
Nanjing, but his son, the Yongle emperor, rebuilt Beijing and moved
the capital back there in 1421.
The ruling housesMongol, Chinese, and Manchuall came
to power after first waging all-out war to win the mandate of Heaven,
sweeping every competitor from the field. Many of the preceding
dynasties had been founded by coups dtat: the Song by a mili-
tary mutiny; the Tang similarly. A tradition of dynastic abdication

52
 Permanent Unity Largely Achieved 53

(shanrang) made possible in some cases a bloodless turnover of

ruling houses. Founding a dynasty on the back of victory in all-out
war seems to have lent a formidable aura of legitimacy to the Yuan,
Ming, and Qing. (That the Yuan collapsed as soon as it did came as
a bit of a surprise, but dynasties such as the Qin and Sui that reuni-
fied China after long periods of division seem for various reasons to
have experienced short life spans. Still, the Qin lasted fifteen years,
the Sui twenty-nine, and the Yuan eighty-eight.)
The Mongols and their Chinese advisers at the outset ignored
Song institutions and drew heavily upon Liao, Jin, and ultimately
Tang precedents to construct a governing system for a reunited
China. The main Yuan effect upon the complexion of Ming and
Qing institutions appears to have been the idea, not originated but
certainly emphasized by the Mongols, that control of such a huge
empire required that distinct lines of separation needed to be drawn
between palace and bureaucracy, between civil society and a hered-
itary military, and between and among ethnic, occupational, and
regional groups in bureaucracy and society at large. The Yuan also
carved China up into large provinces, much of whose geographical
shape survives to the present day. In the Yuan, the Censorate became
a distinct and occasionally powerful organ of political protest as well
as of surveillance over the rest of the bureaucracy. A few Southern
Song arrangements were added later, at the expense of northern tra-
dition; most important, Zhu Xis Neo-Confucianism was singled out
as Chinas premier unifying ideology and as the basis for a revived
national system of competitive written examinations qualifying can-
didates for office holding. Restricted quotas ensured that the poten-
tially powerful influence of degree-winners was counterbalanced
by officials with other kinds of qualifications; tight quotas were also
responsible for extending from Song times the continued fostering
of a very large community of students and educated men, literati
who had little hope of entering government and were therefore nor-
mally engaged in family, teaching, and other local matters, but who
in times of crisis served as an articulate national intelligentsia. The
Ming founders, despite their anti-Mongol ethnic posturing, clearly
built their dynasty on these Yuan foundations.
The most pressing problem the Yuan bequeathed to its Ming suc-
cessor was the evident shakiness of its central structure of control.
54 Permanent Unity Largely Achieved

Was the problem the structure, or the officials who staffed it, or both?
The founding emperor of the Ming and the literati who advised him
constantly talked about why the Yuan had collapsed and caused sev-
enteen years of anarchy and horror for all the people of China. For
them, understanding the Yuan collapse was a prerequisite to build-
ing safeguards sufficient to prevent a similar collapse of the new
Ming order. It is this intellectual edge to the Ming founding that
differentiates it from its predecessors in the Chinese past that have
been discussed thus far. Major dynastic founders, men such as those
who founded the Tang and Northern Song, were by and large prag-
matic actors who had their advisers, to be sure, but certainly were not
surrounded, as the Ming founder was, by a militant and mobilized
community of scholarly ideologues from south Chinas Zhejiang
province, men who were versed in Neo-Confucianism, history, and
current events and who were able and eager to lay out a comprehen-
sive agenda for the new Ming state.
The Ming founders analysis of the Yuan collapse was subtle and
complex, but also impassioned and driven by their own political
needs. At the outset, they found no single issue adequate to explain
it, but rather a tissue of causes having something to do with Yuan
institutions, more to do with the dominant positions of non-Chinese
elites, but most of all to do with ethical anomie, administrative lax-
ity, and corruption on the part of nearly everyone holding official
positions high and low. Further, it was especially Chinese officials
in Yuan government whom they found guilty of corruption and
malfeasance on such a scale as to give the masses of the people of
China little choice but to rise up in nationwide rebellion. All of the
Ming builders had witnessed one or more aspects of this corruption.
Zhu Yuanzhang, the ex-peasant and founding emperor (known post-
humously as Ming Taizu), made some vivid statements of his own
about what he as a youth had seen of corruption and incompetence
in Yuan local government. Things had gotten to such a sorry state
not because the Mongols were cruel, but because, in Taizus view,
they were negligent. At court, the emperors had allowed power to
drift downward into the hands of prime ministers who then denied
the emperor any real role in governance and on their own ran the
great Yuan empire by means of favoritism and bribery.
 Permanent Unity Largely Achieved 55

The Zhejiang Confucian intelligentsia, welcomed by the Ming

founder into his developing regime as early as 1360, also had direct
personal experiences of the evils of Yuan local government, and they
drew from these experiences conclusions that were quite compatible
with Taizus. They were older than the unschooled Taizu, and, as
educated men, they had earlier tried to reform from below the whole
Yuan system. They had enlisted the support of sympathetic Yuan offi-
cials (some of them non-Chinese) who were serving in Zhejiang, in
the cause of achieving distributive fairness in the imposition of service
obligations on large and small households, a difficult task that required
breaking the corrupt linkages between large and rich households and
greedy officials and clerks in local government. Such reforms were
indeed launched, with great intensity, in a cloud of Confucian moral
propaganda, but in the end they failed because the regional cen-
sors could not be persuaded to endorse them. And so, things drifted.
Instead of mobilizing forces to suppress great outlaws such as the
pirate Fang Guozhen, Yuan central government offered them titles
and rewards to buy their loyalty. The Ming founders were determined
to put a permanent end to this pattern and, generally speaking, they
did just that.
No earlier dynasty was ever founded quite as the Ming was, and
there is no doubt but that Taizu and his advisers seized upon the
eradication of local corruption and replaced it with honest govern-
ment for the sake of all the people as the energizing and legitimiz-
ing aim of their many efforts. As for overthrowing the Yuan, the
Ming founders considered it a legitimate dynasty, and they never
claimed responsibility for its destruction. No, they argued; the Yuan
forfeited the empire on its own initiative. What the Ming conquered
and destroyed was not the Yuan, but the various Chinese and other
warlords who had come to power in the wake of the Yuan downfall.
But establishing clean and honest government in early Ming China
proved to be a most frustrating task. It required nothing less than
huge and bloody purges of everyone suspected of corruption or trea-
son in officialdom, and the reeducation of all the people of China in
basic Confucian morality. It also required some institutional recon-
figuring of the inherited Yuan machinery of government, extin-
guishing all prime ministerial functions at the provincial and central
56 Permanent Unity Largely Achieved

levels, with a view to giving the emperor unrestricted personal access

to every available lever of powerand all this because the moral
rot prevalent in Yuan times had eaten so deeply into the minds of
even Chinas educated elites and highest officials that they could
not be trusted with unchecked and unsupervised responsibilities of
any kind.
There exists much debate in China and abroad over whether Taizu
was a good and effective rulera protonationalist who expelled the
Mongols, unified China, and ruled it wellor an evil influence
whose chief legacy was a machinery of centralized despotism that
placed such a straitjacket on China that centuries later it could barely
respond to the challenges of the modern world, and who reinforced an
authoritarianism whose effects linger to the present day. The answer
appears to be that he was both, or neither. The power of his legacy is
hard to gauge, for two reasons. One, every new emperor in the Ming,
beginning with Taizus own grandson and immediate successor, the
Jianwen emperor, felt free to shake off the supposed straitjacket and
make changes in the inherited institutions and lines of policy. And
two, Taizu ruled so long (thirty years), issued so many edicts and direc-
tives, and penned so much himself that his written legacy was so rid-
dled with contradictions that much of it was simply ignored, or it was
selectively invoked by political actors in later times to support almost
any position they might choose. At the same time, it is also the case
that the gross institutional features of Ming government were adopted
by its successor, the Qing, and the Qing rulers were always respectful
of the Ming founder and protective of his reputation. But it might be
better said that the Qing institutions for the governing of China were
based upon a much deeper inheritance reaching back to Yuan, Tang,
and Han times and earlier. Perhaps Taizus principal contribution was
his removal of permanent, regularly appointed prime ministerial posi-
tions at the central and provincial levels.
A leading example of the ambiguities in Taizus legacy is his posi-
tion on the role of eunuchs. He is on record as having sternly advised
that palace eunuchs be kept few in number, that they be forbidden
to learn to read and write, and that their duties be restricted to the
menial. But in fact, he ignored his own advice and found eunuchs
useful in military supervision and in the conduct of border trade and
foreign relations. From the time of the Yongle emperors usurpation
 Permanent Unity Largely Achieved 57

in 1402, palace eunuchs began to proliferate and to develop their

own recruitment modes, schools, and system of internal ranking and
controls. By the late Ming, the eunuch corps had become much
larger than the civil bureaucracy, numbering a hundred thousand
or more in Beijing, Nanjing, and many other parts of China, where
eunuchs handled military and police functions, managed imperial
farms, collected special taxes, and managed salt production, min-
ing, and silk manufacture in addition to their earlier roles in foreign
trade and diplomacy. In Beijing, they acted as the emperors per-
sonal office staff (in addition to performing their culinary and many
other services); they also substituted for the emperor in the highest-
level review of legal appeals and court cases. In some cases eunuchs
became very rich, patronized other eunuchs, formed links with the
north China criminal underworld, and headed families of their own
by adoption. It is noteworthy that they never managed to seize con-
trol of the throne in the Ming as they did in the late Tang. To the
end, Ming eunuch power remained fragile, and emperors could and
did punish and kill them almost at will. Just as there was no execu-
tive prime minister in charge of the regular bureaucracy, there was
no regular chief eunuchs office either. Unless found in the general
principle of the primacy of imperial authority in any conceivable
context, Taizus supposed legacy regarding restrictions on eunuchs
was nowhere to be found in this situation.
In this connection, it is sometimes overlooked that both the Ming
and Qing had what amounts to a third branch of government that
was associated personally with the emperor and functioned sepa-
rately from the civil and military hierarchies. The Ming-style palace
eunuch corps was continued in the very early Qing, but it came to be
argued that eunuchs had had a major role in causing the deteriora-
tion and collapse of the Ming. Eunuch factions and factions in the
regular bureaucracy had repeatedly made cooperative arrangements
with a view to destroying their respective rivals. So, in 1661 the deci-
sion was taken to drastically reduce the size and curtail the func-
tions of the eunuchs, and to supplant them with the many imperial
bondservants (booi) that had accompanied the Manchu conquest
machine into China. Thus the Imperial Household Department
(Neiwufu) came into being, with the remaining small corps of
eunuchs placed under its supervision. From an original staff of 402,
58 Permanent Unity Largely Achieved

by 1796 the department numbered 1,623. It was a self-contained

bureaucracy, with its own schools and testing procedures, with cen-
sors and bureaus that resembled those of the Six Ministries of the
civil bureaucracy. Revenue was the most important bureau; it con-
trolled a kind of conglomerate imperial enterprise encompassing
commerce and industry, especially involving ginseng, copper, salt,
silk, textiles, and foreign trade, much like its Ming eunuch predeces-
sor. By the eighteenth century, the Imperial Household Department
had grown extremely corrupt. However, it never controlled the flow
of documents between emperor and bureaucracy, nor did it have the
military and police capabilities of the Ming eunuchs.
The Yuan situation appears to have been different. There were
many revenue-producing agencies based in the Yuan capital that
were under nominal imperial control, but the later Yuan emperors
were mainly creatures of high-level Mongol or Turkish grandee fac-
tions. The leaders of these factions controlled military-guard units,
and they routinely seized the revenue-producing agencies for per-
sonal or factional use whenever they came to power. In this respect,
the Ming and Qing represent a distinct upward leap in imperial
command.
The early Ming profile, at court and in the country at large, was
very different from the situation in the early Qing. A list of com-
parisons can show this: Manchus from the northeast, outside China
proper, founded the Qing; Chinese from the center and south cre-
ated the Ming. Ming military unification was accomplished relatively
swiftlyover about eight years, from 1360, when Taizu established his
capital in Nanjing, to 1368, when Ming forces entered the Yuan capi-
tal, Dadu , and forced the Yuan imperial court to retreat to Mongolia;
the Qing required some forty years, from 1644 to 1683, to defeat all its
former military allies and unify China proper. But the Mings Taizu
had an easier time of itthe Yuan loyalism he faced was weak, and his
warlord opponents put up little in the way of fierce, diehard resistance
to him. It was much harder going for the Manchus. While defections
and capitulations to the Manchus were commonplace, there were
also major urban pockets of suicidal Ming loyalism and local pride,
led by Chinese literati, especially in the south, the like of which was
last seen in Wen Tianxiangs late thirteenth-century resistance to the
Mongol conquest. Ming Taizu never had to face such obstacles; the
 Permanent Unity Largely Achieved 59

Qing takeover was, for that and other reasons, much more violent.
Ming Taizu was alert to the need to discipline and punish unruly
troops, and the wars of Ming unification evidenced fairly little in the
way of massacres, plunder, and wanton wreckage; there were no early
Ming counterparts to the Manchu massacres of whole populations in
a half-dozen big south China cities, including the infamous slaughter
and destruction at Yangzhou in May 1645 (where much of the actual
killing was done not by Manchu, but by Chinese troops), and even
the Mongol conquest of south China was peaceful by contrast. Nor
did the Yuan or Ming founders, Khubilai or Taizu, find it necessary
to conduct a national headcount of all supporters and resisters, as the
Manchus did, when in June 1645 they imposed the requirement that
all Chinese males show their new loyalty by adopting the Manchu
coiffure of a shaved forehead and rear-hanging pigtail. Taizus violent
streak manifested itself after he unified China; the Manchus, by con-
trast, eased up after the unification was complete.
It is interesting, too, to note the very different things leading intel-
lectuals were saying about the condition and the future of China
in the seventeenth century as compared to the fourteenth. In the
fourteenth century, the Zhejiang intellectuals who gathered around
Taizu and advised him had already thought hard about the future of
China and, generally speaking, found solutions to the Yuan break-
down and subsequent disorders in the building up of emperor-cen-
tered power, institutionally untrammeled and guided only by ethical
constraints, strong enough to impose by diktat whatever remedies
penal, educational, bureaucraticwere required to cleanse and
reorient the minds of all the people of China and return the civiliza-
tion to something approaching its Golden Age state of grace. The
leading seventeenth-century intellectuals were quite different. Gu
Yanwu (16131682), Huang Zongxi (16101695), and others were
personally involved not in the Qing conquest, but in the failed south-
ern Ming resistance. The fourteenth-century intellectuals located
the source of the trouble in a nationwide epidemic of self-centered
greed and corruption; and indeed, this approach was endorsed in
the seventeenth-century courts of the Shunzhi and Kangxi emper-
ors, with the first focused on suppressing corruption of all sorts, and
the second on educating all the people of China in fundamental
Confucian values. But the seventeenth-century intellectuals, who
60 Permanent Unity Largely Achieved

thought long and hard, as their fourteenth-century colleagues had,

about what had gone wrong with China, did not trace the troubles
to corruption, and corruption to mental or spiritual failures. Both
Huang and Gu were chary of Song Neo-Confucian orthodoxy, but
Gu went further to emphasize the value of evidentiary research
and resurrect Han dynasty thinkers as sources closer to the origi-
nal Confucianism. Neither Huang nor Gu liked centering the
discussion of Chinas troubles on minds and mentalities, and they
looked instead to the desirability of reworking institutions. (Huang
and Gu knew each other, read each others work, and were agreed
upon many issues.) Huang thought the Ming emperors had had too
much authority and had abused it, and so he proposed that imperial
authority be shared with a restored prime minister; and since the
Ming system of creating intrabureaucratic jurisdictional conflict as
a means of sustaining central control appeared counterproductive,
Huang would further apportion discretionary authority all down the
line to include the county magistrates. Both Huang and Gu wanted
to abolish the apparatus of provincial and circuit supervision that
every postAn Lushan dynasty, including the Ming, had placed
between the center and the localities. Still, the case remains that
however much admired their reform ideas were in literati circles,
they had virtually no impact on the Qing rulers, who were on the
whole content to rule China by working within the inherited and
familiar Ming institutional format.
The Ming was, then, exceptional in making common cause with
a militant and visionary wing of the Confucian intelligentsia during
its founding phase. And what of the Mongols? The Mongols had
important commitments in other parts of the world, and it was not
until the 1250s and 1260s that Khubilai, while still a young regional
prince, called in northern Chinese Confucians to assist him in creat-
ing appropriate machinery for ruling China and to cultivate support
in educated circles throughout China for his drive toward supreme
power. Northern Confucians such as Xu Heng (12091281) also
played an important part in familiarizing young Mongols and Central
Asians at court in the psychological (mind-and-heart) basics of
Confucian doctrine. The Manchus had less need of all that, being
all along more familiar with Ming civilization than the Mongols
were with the Jin or Southern Song, and having consistently had
 Yuan China, 12711368 61

their attention much more heavily focused on China. Indeed, Ming

defectors and educated Chinese natives of Manchuria were from the
beginning prominent participants in the rise of the Qing, whereas
ethnic Chinese had scarcely any role at all in the critical early phases
of the creation of the Mongol empire. Although none of the early
Qing rulers bothered to search out a Confucian ideological van-
guard and make common cause with it in the style of Ming Taizu,
they certainly made practical use of advisers such as the Manchurian
Chinese Fan Wencheng (15971666) in reconstituting Ming-style
institutions for the ruling of China.
It is time now to take a closer look at the Yuan, Ming, and Qing,
providers of nearly seven centuries of unity and relatively stable rule,
the longest such span China ever experienced.

Yuan China
12711368

T he Mongol empire was unprecedented in its geographical scale.

China had never seen anything quite like it. Earlier barbarian
regimes, such as those founded by the Khitans, the Jurchens, or,
earlier, the Tuobas, had all been heavily involved with north China
and its cultural and administrative heritage. This was much less the
case with the Mongols. From Chinggis Khans time down to the rise
of his grandson Khubilai in the 1260s, the Mongols, their power cen-
ter still located in the steppes, were also attracted to Transoxania,
Iran, the Middle East, and the Volga and Caucasus regions. For fifty
years, the Mongols did not even bother to declare a Chinese-style
dynasty in north Chinauntil, in 1271, Khubilais Chinese advisers
persuaded him to declare one, the Yuan.
The effect of Mongol raiding and warfare in north China appears
to have been catastrophic. In 1207, the Jin could count some 50
million people; from 1234 to 1236, the Mongol authorities found
but 8.5 million. Early Mongol rule in heavily battered north China
featured at the local level a complicated mix of Chinese warlords,
former Jin officials, and Buddhist and Daoist priests serving as com-
munity leaders. At higher levels, governance reflected the Mongols
62 Permanent Unity Largely Achieved

stunning ability to commandeer administrative and technical exper-

tise from all parts of their Eurasian empire. Thus, Muslim finance
and tax experts were fetched from Transoxania and put to work in
north China, where they imposed their own techniques and often
ignored Chinas administrative traditions (Hanfa). Similarly, admin-
istrators from China were put to work in Transoxania. North China
was squeezed as a resource base for Mongol operations worldwide.
Other than that, it held no special importance for the Mongols.
It was Khubilai who changed things. A different approach was
shown in the north China appanage of young prince Khubilai (1215
1294), where his mother, a Nestorian Christian, proved to be a good
manager. Her and Khubilais princely court became a magnet for
Confucian intellectuals and others in north China who were desper-
ately seeking someone among the Mongol elite whom they could
support and, as advisers and operatives, influence in such a way as
to restore some semblance of traditional normality to the dreadful
state of affairs in their once-flourishing homeland. As it turned out,
Khubilai was ambitious to succeed his elder brother Mngke as great
khan of the Mongol empire; if Chinese Confucians could help him
in his drive for power, well and good. Khubilai flattered them, and his
Chinese adherents did everything they could to rally support behind
Khubilais seizure of power from 1260 to 1264. In the 1270s, they
worked hard to convince the southern Chinese of the legitimacy of
the khan-emperors attempt at the conquest and annexation of the
Southern Song.
The tripartite power competition that for three centuries had
commanded the scene in East Asia was at last over, thanks to
Khubilais successful plan to secede from the overstretched Mongol
world empire and build a China-centered Mongol empire of his
own. The destruction of the Xi Xia and Dali states and the various
Manchurian entities, and the incorporation of their territories into
the Yuan realm, created an enlarged lebensraum for China proper,
as the Yuan government opened the annexed territories to Chinese
immigration.
But after the 1279 conquest of the Southern Song, Khubilai
ceased his close cooperation with his Confucian officials and advis-
ers, leaned instead upon his mainly Muslim fiscal experts, and
made of Yuan China a gigantic resource base for further expansion
 Yuan China, 12711368 63

in Asia: Burma (1277, 1287), Vietnam (1281, 1285, 1287), Japan

(1274, 1281), and Java (1292). There were also long and expen-
sive wars against other Mongol khanates to the west over control of
Xinjiang and Mongolia (in the end, the Yuan kept Mongolia and
ceded Xinjiang). All these thrusts were costly and unproductive, and
Khubilais grandson and successor, Temr, after about 1300, put an
end to all further expansion.
Given that there were not many Mongols (perhaps seven hundred
thousand warriors), and given that most of China is not good terri-
tory for mounted warfare in the nomad style, there is no accounting
for the astounding Yuan success in unifying China unless one recog-
nizes the Mongols aptitude, perhaps derived from animal herding,
for commandeering the efforts and skills of others. In addition, the
Mongols had a willingness to tolerate, and indeed patronize, the phi-
losophies and religions of the people they conquered. The conquest
of south China called not for cavalry maneuvers, but for naval forces
and sieges; and as for sieges, the Mongols brought into China from
the Muslim world experts in the new ballistic machines, counter-
weighted trebuchets, and put them to use in the siege of Xiangyang.
For navies, the Mongols commandeered Korean and Song fleets for
maritime operations, among other measures. The military manpower
used in the conquest of south China consisted mainly of Chinese
armies under Mongol command. Defectors were, on the whole, wel-
comed and well treated. Although wary of Chinese cultural power,
the early Mongols were not afraid of it and used imported Turks,
Muslims, and Tibetan Buddhists as alternative sources of adminis-
trative talent.
Yuan rule over China became exceedingly complex, a confusing
amalgam of religious, occupational, and ethnic groups under the con-
trol of an apparatus that, like the Liao and Jin, mixed the consensual
but often bloody decision-making traditions of the steppes with the
hierarchical and authoritarian approach of China. There developed
under the Yuan a four-class system for allocating legal and fiscal privi-
leges, whose basis was the degree of historical closeness a given group
had to the house of Chinggis Khan and the timing of the groups
addition to the empire. Thus, first in order came the Mongols, who
were themselves differentiated by inherited rank. Second came other
foreigners, including especially Tanguts and Turks. They were called
64 Permanent Unity Largely Achieved

semu. (Marco Polo seems to have been treated by his hero Khubilai
as a semu and used by him as a special emissary to report on military
operations in Yunnan and Burma.) Third came the Hanren, which
took in all former subjects of the Jin empire, including Chinese,
Khitans, and Jurchens. Last came the Nanren, former subjects of the
Southern Song. Wealth and population were overwhelmingly concen-
trated at the bottom of the ethnic hierarchy, but south Chinas legal
and political disadvantages were tacitly compensated for by its much
lighter administrative and tax burden, as compared to north China.
As emperor of China, Khubilai can be credited with the accom-
plishment of at least two major projectsthe creation of a national
system of paper money backed by precious metal reserves, and the
reextension of the Grand Canal north to Dadu, a portion defunct
since China was last unified in the late Tang.
After Khubilais death in 1294, the Yuan dynasty gradually evolved
into a less exploitative regime that more and more focused its efforts
on addressing the needs and wants of its huge Chinese population.
Chinese demands for the institution of a civil service examina-
tion system based in the daoxue texts and doctrines of ZhuXi were
answered affirmatively by the Yuan authorities and made reality
starting in 1315. Young men from all four of the officially recog-
nized ethnic classes were invited to compete in the examinations. A
quota of twenty-five jinshi every three years was set for each of the
four. Thus, Mongol and semu youths, domiciled in the provinces
without access to the keshig (the largely hereditary imperial cadet
corps of fourteen thousand or so stationed in Dadu), were given the
opportunity to study daoxue, become Chinese-style degree-holders,
and enter bureaucracy in that fashion. Bureaucrats who entered gov-
ernment by way of hereditary privilege or clerical training always
outnumbered degree-holders, even in the Tang and Song. Although
never more than a tiny fraction of Yuan bureaucracy, degree-holders
tended to occupy positions of low rank but high visibility in the vari-
ous literary academies or in the Censorate, where their intellectual
credentials and ability to articulate at times overwhelmed the great
majority of their colleagues.
The forty-year period from 1294 to 1333 saw nine emperors come
and go. Far from being autocrats, the later Yuan emperors were crea-
tures of factional struggles among various alignments of Mongol and
 Yuan China, 12711368 65

Turkish (semu) grandees. By 1328, one faction went so far as to enlist

the support of Confucianist degree-holding officials, many of them
Chinese. By the late 1330s, there developed a split between those
high elites who favored the Confucian style and those who insisted
upon Mongol traditions. The Mongolists won in 1333, but lost in
1340. The 1340s featured a very different split, as a series of natural
disasters demanded that the government formulate and apply rem-
edies. One faction, consisting of officials of various ethnic classes,
took the minimalist Confucian line that the disasters were best han-
dled by giving local officials on the scene all the authority necessary
to bring relief. The other faction, also polyethnic in composition,
echoed the centralist, world-transforming approach of Wang Anshi.
Catastrophic floods had destroyed the north China portion of the
Grand Canal, wrecked the Hejian salt works (a major source of gov-
ernment income), and effected a shift in the lower course of the
Yellow River, so that it now entered the sea north of the Shandong
peninsula instead of debouching into the Huai River system south
of the peninsula as before. The energetic young Mongol chancellor
Toghto argued that fixing the canal and the salt works required that
the Yellow River first be rerouted to its former natural channel,
south of the Shandong peninsula. And so, in what was the great-
est hydraulic project ever undertaken in China to that point, the
Yuan central government successfully engineered the rerouting of
the Yellow River over the period May to December 1351.
Meanwhile, in actions that were prompted by the natural disasters,
but only tangentially related to Toghtos rechanneling project, all of
China erupted in an orgy of violence. Inspired and loosely directed by
sectarian holy figures, first in the north from 1351, then in the south
from 1352, gangs of young idlers and drifters donned red headbands
and went from county seat to county seat, killing officials and rich
landlords and declaring an end to Yuan rule. There had been nothing
quite like this in China since the Yellow Turban revolt of 184.
The Red Turbans, as they were called, at first caught the Yuan mil-
itary and police off guard and sent them reeling in shock. However,
Chancellor Toghto thought suppressing these riots to be a task that
lay within the competence of the central government to manage, and
he had the emperor appoint him supreme director of the whole crack-
down. His method was traditionally and effectively Mongol. Just as the
66 Permanent Unity Largely Achieved

Mongols had earlier built Dadu and conquered south China in the
1270s, so too now suppressing the Red Turbans nationwide required
commandeering both local and foreign manpower and expertise and
putting these together in temporary task-oriented groupings in such a
way that no generals or other leaders were ever in a position to arro-
gate power and authority to themselves. Toghto personally took com-
mand of a heterogeneous force whose job was to clear the route of
the yet to be rebuilt Grand Canal. Thus, he recovered Xuzhou from
rebel control in October 1352. Then, he turned to a costly project to
convert north Chinas farmland into rice paddies. Late in 1354, he
assembled a huge multinational force to besiege the Grand Canal
city of Gaoyou. The taking of Gaoyou would have broken the back of
the great riots and restored Yuan central authority nationwide.
It was not to be. Toghto had issued huge sums of unbacked paper
currency to fund all of his activities, and runaway inflation was
about to destroy the Yuan monetary system altogether. In his years
in office, Toghto had made enemies, many of them of the conserva-
tive Confucian persuasion, who did not like the chancellors cen-
tralizing methods. In the end, the emperor was persuaded to issue
an edict dismissing him from office. Toghto received that edict in
January 1355 while with his army besieging Gaoyou. Though urged
not to, he obeyed the edict. The siege thereupon collapsed. The
besieging army units melted away. The rebels inside Gaoyou could
hardly believe their good fortune. Other rebel remnants, chased into
hiding in the back country (among them the future Ming founder,
Zhu Yuanzhang), found themselves with a sudden new lease on life.
China was headed for thirteen more years of destructive internal
warfare.

The Creation of Ming China

T he Ming founding of 1368 was historically unique in several

respects. The founder, Zhu Yuanzhang (13281398), was not
at the outset a military commander, nor could he boast of a high
social pedigree, or even an educated background. He came from
a central China peasant family, almost all of whose members died
in the diseases, famine, and wars of the 1340s and 1350s. Not even
 The Creation of Ming China 67

the founders of the Han in 202 BCE or the Latter Liang in 907
were burdened by origins so precariously low. A grim and suspicious
autodidact, the Ming founder (posthumously named Taizu) inspired
fear and respect, but no love or devotion. Yet until he accomplished
the feat, no dynasty had ever built up strength in the south first, then
marched north to reunify China. (All earlier attempts to do that had
failed: the Eastern Jin in the fourth century, the Northern Song late
in the eleventh century, and the Southern Song early in the thir-
teenth). And no earlier founder ever developed, and to so high a
degree imposed, a visionary program aimed at a total transformation
of China. Only Mao Zedong is comparable.
The Ming unification and subsequent remaking of China was only
to a limited degree a product of revanchist ardor. Rather, it was the
product of the cold, relentless institutional engineering that Taizu
conducted from his earliest moments as leader of a contingent of
former Red Turban rebels that emerged from hiding after Toghtos
dismissal, migrated south across the Yangzi, and captured Nanjing in
1356. Taizus institutional engineering was centered in a profound dis-
trust of human nature, a belief that no one could be trusted for long
with discretionary authority of any sort, that all functionariescivil,
military, or princelyalways needed to have their powers checked
and divided, and to be under scrutiny, both open and covert, at all
times. Taizus penal repressions were frequent and savage. There is no
doubt that all this allowed, indeed required, a personal despotism.
But it was a despotism girded by a Confucian rationale. Unlike
earlier dynastic founders, Ming Taizu actively sought ideological
direction. By the early 1360s, he had recruited to his cause several
Confucian intellectuals of national reputation from Zhejiang prov-
ince, including especially Liu Ji (13111375) and Song Lian (1310
1381). These intellectuals, who had been thinking and writing about
the future of China, impressed upon their patron a totalizing vision of
national sociomoral transformation, a return to the utopian norms of
Golden Age antiquity. This was their remedy for the laxity, careless-
ness, and corruption of Yuan rule that, in their view, allowed the Red
Turban horrors to happen and caused the Yuan collapse. So, the Ming
founder was advised to take up the duties not just of ruler, but also of
teacher to the people of China. He was a ruler-teacher (junshi). The
guiding slogan was repossessing antiquity (fugu). No founder, since
68 Permanent Unity Largely Achieved

the semimythical golden age itself, two millennia earlier, had assumed
so much responsibility.
A burst of administrative energy was not untypical of major
dynastic foundations, but the Ming was exceptionally energetic. It
inherited a China battered by seventeen years of war and defenseless
against whatever the Ming founder now wished to impose upon it.
Teaching the people ethical fundamentals was believed to be essen-
tial to the reinstitution of order. Ming propaganda was produced and
disseminated in great quantities. Reams of rules, regulations, and
exhortations, written in part by the founder himself, let the people of
China know what was expected of them in the way of good behavior.
Resolute efforts were made to penetrate society down to the village
level and even out into the fields. Blind men led by young boys went
about sounding a clapper and shouting so that all the working peas-
ants could hear Taizus Six Maxims demanding good behavior.
Minds everywhere had to be remolded so that the horrors of the late
Yuan collapse might never be repeated. For those who declined to
heed the warnings and teachings, the penalties were merciless.
For thirty years Taizu ruled China with an iron hand. His aim
to remold ethically the people of China sat uneasily alongside his
fundamental distrust of them. He soon grew dissatisfied with inher-
ited institutions. He found that he could not rely on the civil service
examination system to deliver honest officials, and so he scrapped it
in 1373 in favor of a system of recommendations (later he restored
the exams). He found he could not trust his bureaucracy, and he
purged it heavily and bloodily on several occasions. He found he
could not trust his prime minister, so in 1380 he abolished the office
altogether, leaving the emperor himself as sole supervisor of some
twelve different organs of government. He found he could not trust
the generals who had conquered China for him; at first he made
them a hereditary nobility, but in 1393 he destroyed them in a huge
purge and assigned their high-level military responsibilities to fifteen
of his sons, enfeoffed mainly in the north as regional princes.
The Ming founders policy was deliberately aimed at maintain-
ing national unity and order at minimal fiscal cost. The huge Ming
army was made self-supporting and hereditary, which allowed tax
rates on everyone else to be set very low. The people themselves were
required to perform low-level judicial, police, and tax-collecting
 The Ming Refounded: Jianwen and Yongle 69

functions for free. Civil service was cut to minimal size and was not
generously paid. Uninterested in empire, early Ming China con-
fined itself to a China proper that the Yuan had enlarged to a size
eight times that of present-day France.

The Ming Refounded:

Jianwen (r. 13981402) and
Yongle (r. 14021424)

T he Ming founders Ancestral Instructions mandated throne suc-

cession strictly by primogeniture. Taizus eldest son died in
1392. Arguably, the next in line would be the next eldest son. But
instead, when the founder died in 1398, the throne passed, surely
on his orders, to the eldest surviving son in the next generation, his
grandson, the twenty-one-year-old Jianwen emperor. In a fashion that
would later become a pattern in Ming successions, the new emperor
sponsored a major policy reversal. Influenced by a younger generation
of Confucian intellectuals, Fang Xiaoru (13571402) in particular,
Jianwen proceeded to reintroduce trust into social and institutional
relationships, to concede much to local initiative, and to dismantle
the despotic machinery of rectification that had become counterpro-
ductive. He also began to remove the armed princes, a troublesome
remaining legacy of the founders radical mistrust. All went well, until
the strongest of the young successors princely uncles, Zhu Di (Prince
of Yan), was confronted at his fief in present-day Beijing. Jianwen held
an overwhelming edge in manpower and resources. Victory ought
to have been his. But strategic errors, acts of treachery, and bad luck
fatally afflicted the imperial side. Zhu Dis forces prevailed. On July
13, 1402, the rebel prince entered Nanjing. The imperial palace was
set ablaze, and no sure sign of Jianwen was ever seen again. The Ming
dynasty was, in effect, refounded. Zhu Di was emperor. The reign title
he chose was, inexplicably almost, Yongle, or perpetual joy.
The Yongle reign (14021424) consisted less of joy than in an orgy
of activity. It ranks with the reign of another usurper, Tang Taizong,
as one of the most spectacular in Chinas long history.
70 Permanent Unity Largely Achieved

On the home front, the examination system, based in Zhu Xi

Neo-Confucianism, became permanently established as definitely
the most prestigious avenue to office for the educated youth of the
realm. To ensure standardization and Neo-Confucian orthodoxy, the
Yongle emperors government edited and issued to every school in the
realm three large compendia as study guides. Also edited and pub-
lished were a giant revision of the Buddhist Tripitaka and a colossal
collectaneum of nearly all writing surviving in China from earliest
times down to Yongles reign. Construction having gone on for many
years, a completely new city of Beijing was declared the official capi-
tal of China in 1421. The Grand Canal, rebuilt yet again, connected
the city to the grain surpluses of the lower Yangzi.
Abroad, the inward focus of the Ming founder was completely
reversed. Palace eunuchs were sent on diplomatic missions to the
tribes of the Amur River region of northern Manchuria, to Xinjiang,
and as far west as Samarkand and Herat. The purpose was to proclaim
the Ming and to encourage submissions to, or at least tributary rela-
tions with, the Ming court. Agreements were made with Japan and
with local authorities in Tibet and Nepal. Most extraordinary of all,
of course, were the five fantastically expensive sea voyages led by the
eunuch Zheng He to Southeast Asia, the Persian Gulf, and even as far
as the east African coast, from 1405 until 1433.
The relocation of the Ming capital from Nanjing to Beijing sig-
naled a shift from what might have evolved into a regime eager to
foster economic growth like the Song to a more warlike regime like
the Han or Tang, whose main preoccupation was military security.
Beijing was a garrison city. The Mongol court, so recently expelled
from China, was perceived to be the principal enemy. Yongles rela-
tions with the Mongols were exceedingly hostile. His father, Taizu,
had believed that the Mongol court was unreconciled to its loss of
China, and in a series of assaults into the steppes culminating in 1388
he all but destroyed the Yuan court in exile. Yongle as prince partici-
pated personally in much of this military activity. Later, as emperor,
he faced two new and competing Mongol confederations: the Oyirads
in the west and Arughtais people in the east. Yongles strategy was to
protect China from their raids by supporting whichever confederation
was the weaker at any given moment. He personally led four military
expeditions into the steppes: 1410 (against Arughtai), 1414 (against the
 The Middle Years of the Ming 71

Oyirads), 1423 (against Arughtai), and 1424 (also against Arughtai).

Yongle died on August 12, 1424, perhaps of elixir poisoning, while
returning from the last one. No Chinese emperor had ever before per-
sonally commanded war expeditions into Mongolia. Yongles doing
so contributed little to Chinas security in the long run. Perhaps for
reasons of economy, he pulled his forward defense posts southward,
out of the steppe margins. He also declined to subsidize Mongolia, as
the Yuan dynasty had done. For as long as the Ming lasted, a legacy of
violence and bad planning regularly vitiated efforts to achieve a peace-
ful resolution of differences. (From 1472, the Ming began building
what Europeans later called the Great Wall of China, arguably a fall-
back measure that signaled the failure of all better policy alternatives.
Portions of such a wall had been built in early times, but what the
Ming built was a wholly new construction.)
There was also the curious Vietnam invasion of 1406, an action
that was not very bound up with Chinas perceived security needs.
Ming China under Yongle was asserting a global moral hegemony,
arrogating to itself the authority to confer recognition on foreign rul-
ers, but also accepting an obligation to intervene forcefully, when
called upon, on those rulers behalf. Zheng Hes fleet intervened in
civil war in Java and issued threats elsewhere, in support of rulers
granted recognition by the Ming, but under challenge from internal
rebels. This was also the case in Vietnam, where the deposed Trn
ruling house appealed to the Ming court to help restore it to power.
The 1406 invasion was a success, but the Trn were in no condition
to resume their rule. The Yongle emperor then made an ill-advised
decision to annex Vietnam outright as a province of China. The
consequences were unpleasant. Yongle soon lost interest; Vietnam
was wretchedly administered; and rebellions against Ming rule were
unrelenting. In 1427, after Yongles death, the Ming decided to swal-
low its pride, withdraw, and officially recognize the Latter L dynasty
as the legitimate rulers of the country.

The Middle Years of the Ming

Y ongles eldest son and successor, the Hongxi emperor, was on the
verge of reversing many of Yongles policies and moving the capital
72 Permanent Unity Largely Achieved

back to Nanjing when he died in 1425 at the age of forty-six, after a

reign of barely eight months. His own eldest son, the Xuande emperor,
aged twenty-six, kept the capital at Beijing, but he reversed many other
policies of Yongle (the maritime expeditions, the Vietnam annexa-
tion) and ruled until his own death in 1435. These rulersYongle,
Hongxi, and Xuandeall had administrative experience before com-
ing to power, were personally involved in governance, and together
saw Ming China come to something of a pinnacle of power and pres-
tige. After 1435, things changed. Emperors were badly schooled, often
came to power as children, and their later involvement in administra-
tion tended to be intermittent or capricious. Xuandes elder son and
successor came to the throne as the Zhengtong emperor at the age of
nine in 1435. In 1449, he was persuaded by his eunuchs to command
in person a military expedition into the steppes, where unfortunately
he was ambushed, captured, and held hostage by Esen, khan of the
Oyirad Mongols. The atmosphere in Beijing turned frantic as high
officials debated whether to evacuate north China altogether. Minister
of War Yu Qian was most forceful, and his opinion carried the dayto
depose Zhengtong in absentia, ignore his baby son, and enthrone his
older half brother, aged twenty-one, as the Jingtai emperor. That act
deprived Zhengtong of all further value, and Esen returned his now
useless hostage to Beijing in 1450. The deposed emperor was put
under house arrest. In 1457, however, he returned to power (as the
Tianshun emperor) when a coup on his behalf put an end to the life
of his brother. (Another anomaly in the imperial succession took place
in 1521, when the Zhengde emperor died childless at the age of thirty,
and, again, it was high officials who decided that his successor should
be a cousin of the same generation, the fourteen-year-old Prince of
Xing, who ruled as the Jiajing emperor, from 1521 to 1567.)
As these later Ming emperors came to show a lack of interest, or
competence, in governing the realm, there arose a sort of free-floating
authority that became a bone of contention among palace eunuchs,
officials of the Censorate and offices of scrutiny, and the grand sec-
retaries (constitutionally improvised successors to the prime ministe-
rial function that was eliminated in 1380; the grand secretaries were
advisers only, but occasionally they reached for executive author-
ity as well). Partisan organization among bureaucrats was a statu-
tory crime, but during the sixteenth century there developed in the
 The Reign of Jiajing, 15221566 73

bureaucracy informal but wide-ranging networks often inspired by

ideas of moral rededication and Confucian renewal. The networks
heavily influenced education, examination criteria, and official per-
sonnel placement. Members at times became vehement critics of
government.
Wang Yangming (14721529) and his many ardent followers
challenged the official Zhu Xi orthodoxy. Wangs career was that of a
highly effective provincial official. His doctrines stressed ethical self-
realization and interpersonal ethical activism. His powerfully attrac-
tive reinterpretation of the Confucian way bypassed such standing
institutions as the national school system and, rather like Zhu Xis
original daoxue movement in the Southern Song, was propagated
mainly in informal discussion circles and in privately funded acad-
emies. The movement was less confrontational than Zhus. Still,
from 1570 to 1579, long after Wangs death, the reforming Grand
Secretary Zhang Juzheng (15251582) closed down all the acade-
mies for criticizing national policy, wasting resources, and distracting
officials and students from their proper duties. That action effectively
put an end to the prevalence of Wang Yangming studies in China
(his doctrines would later find welcome soil in Japan). Zhu Xi ortho-
doxy then underwent a major revival, but not under official auspices.
Ironically, it was revived in private discussion circles (attended by
off-duty officials and students) and in private academies, including
especially the famous Donglin Academy in Wuxi (Jiangsu province),
founded in 1604. The Donglin doctrines were aimed not inwardly,
as Wangs had been, but outwardly, and strove to bring about the
complete ethical cleansing and remaking of Ming government. But
that is to jump ahead of the story.

The Reign of Jiajing

15221566

T he forty-four years of the Jiajing emperors reign, coming as it

did late in the middle of the Ming, is as good a moment as any to
gauge the often paradoxical qualities of the mature imperial Chinese
institutional system.
74 Permanent Unity Largely Achieved

A teenage prince living far to the south in Hubei province, Jiajing

was called to the throne in 1522 by Chief Grand Secretary Yang
Tinghe, de facto dictator of China upon the death of the childless
Zhengde emperor, the young princes cousin. Immediately there
developed a profound struggle between Jiajing and Yangs fac-
tion in the Grand Secretariat and Hanlin Academy over the highly
technical and deceptively trivial question of whether Jiajing should
posthumously promote his deceased princely father to the status of
emperor, or whether he should, in effect, disown his own father and
accept Zhengdes father, the Hongzhi emperor (r. 14881505), as
his adopted father. This was the so-called Great Rites Controversy,
which stretched over many years, ending only in 1545 with Jiajings
triumphal emplacement of his fathers spirit tablet in the imperial
ancestral temple. Jiajing won every earlier round in the battle, includ-
ing especially the great showdown of 1524, when over two hundred
high-ranking central officials conducted a mass protest in front of
the Forbidden City, demanding that Jiajing agree to make himself
Hongzhis son by adoption and so maintain the integrity of an unbro-
ken Ming imperial descent line. The demonstrators were utterly seri-
ous and totally sincere. They knew very well they were risking their
careers and probably their lives as well. But in their view, the issue
went to the very heart of Chinas civilization. The demands of public
imperial ritual had to trump the young emperors personal and self-
ish concern for private family ritual. Otherwise, all of China would
imitate Jiajings selfishness and collapse in an orgy of hedonism.
Jiajing resolutely faced this demonstration down. He ordered the
arrest and cruel flogging of the demonstrators. Seventeen men died
of the beating. In 1528, Jiajing ordered published an official text
that authoritatively stated his own side of the issue. To the end of
the Ming, his position was never again challenged. Yet in a way,
the demonstrators were right: Jiajings stance did affect family order
nationwide. It seems to have opened the floodgates that had long
held back pressures to expand the family system. Orthodox Neo-
Confucianism, in its official Zhu Xi version as sanctioned by the
Ming state, did not permit limitless lineage expansion. But Jiajings
victory in the Great Rites Controversy was a victory of personal feel-
ing over text-based literalism and rigidity. By and large, the people
of China were on Jiajings side. As will be noted later, the formation
 The Reign of Jiajing, 15221566 75

of large endowed lineages with ancestral halls proceeded apace over

the course of the sixteenth century. But here lay an unexpected para-
dox: Wang Yangming and his philosophical followers in bureaucracy
sided with Jiajing; it was Zhu Xi adherents who opposed him. Yet
Jiajing disdained Wang and his school. The rites issue notwithstand-
ing, the Zhu Xi orthodoxy continued to enjoy the emperors support.
Jiajing was no revolutionary by intention.
Nor was he a loved and admired leader of his realm. He toler-
ated no criticism. He categorically distrusted high officials and sub-
jected many of them to terribly brutal treatment. Several attempts
were made on his life, none so telling, perhaps, as his palace ladies
near success at killing him by strangulation in 1542. From that point,
Jiajing moved out of the Forbidden City to a detached palace in a
nearby imperial park, where he occupied himself with Daoist adepts
and slowly (and accidentally) poisoned himself with their elixirs of
immortality. Legal and personnel matters he often decided by con-
sulting the spirits with a planchette. Bouts of depression alternated
with fits of rage.
During Jiajings long reign, the dynasty temporarily lost control
of both its northern and its eastern (coastal) borders. The Mongols
demanded the opening of trade relations, and when Jiajing refused,
they conducted devastating raids deep into Ming territory. From the
1540s, Japanese traders and their coastal Chinese partners (including
members of powerful local lineages) likewise turned to horrific plun-
dering expeditions inland when Jiajing denied their trade demands.
The Portuguese tried to open foreign trade as early as the 1520s but
were flatly denied. Jiajing clung steadfastly to an unimaginative pol-
icy of unrelenting hostility against pressures from both outside and
inside Ming China to liberalize economic relations with the rest of
the world.
The emperor was fortunate that none of his enemies sought to chal-
lenge the legitimacy of Ming rule. The Mongols wanted trade rela-
tions only and harbored no plans to conquer China again. The traders
and pirates of the coastal regions built large fleets and fortified bases
onshore, but they developed no dynastic ambitions. High officials and
their factions competed over the question of how best to satisfy Jiajings
foreign policy demands. Chief Grand Secretary Xia Yans idea was to
take on the Mongols aggressively, chiefly by invading and annexing
76 Permanent Unity Largely Achieved

the Ordos desert region, which was Altan Khans main staging area
for raids. That idea failed for logistic and other reasons, and Xia was
executed in 1548. His rival and successor, Yan Song, preferred wall
building and a passive defense, but this failed to stop Mongol raid-
ing; Jiajing, who never wholly trusted Yan, reduced him to commoner
status in 1565, confiscating his possessions, and executing his son,
mainly on corruption charges lodged by Yans bureaucratic enemies.
Similarly, a campaign of extreme violence against the coastal traders
and pirates during the years from 1547 to 1550 backfired and had to
be dropped. A subsequent policy of coaxing surrenders worked rather
better. A military genius, Qi Jiguang, who enjoyed the protection of
high officials in Beijing, organized effective antipirate defenses in
the 1560s. Unknown to Jiajing, it was regional officials who ceded
Macao to the Portuguese as a permanent trading base in the 1550s.
Then, upon Jiajings death on January 23, 1567, the Ming government
rescinded the ban on private maritime trade and opened peaceful trib-
ute and trade relations with Altan Khans Mongols.
The Jiajing reign illustrates the overall strength and resiliency
of the Ming system. Forty-four years of a misguided foreign policy,
and forty-four years of internal military revolts, famines, and revenue
shortfalls under an arguably dysfunctional emperor, would have
been more than enough to cause the collapse of one of Chinas
top-heavy dynasties of the sixth century. A thousand years later, the
sixteenth-century Ming, for all its inadequacies, was never in danger
of falling. Its legitimacy went without serious challenge. Beneath the
turbulence of high politics, routine administration continued as ever.
Despite trade bans, foreign (Japanese and American) silver poured
into the country in payment for Ming silk and porcelain. People on
their own undertook to organize large and powerful lineages. They
took control of local education. The Ming state either ignored all
this or could do little about it.

Late Ming Foreign Relations

Z hang Juzheng was chief grand secretary during the minority

of the Wanli emperor (r. 15721620), a child of nine at acces-
sion. Zhang was a practical reformer, not a visionary. By ruthlessly
 Late Ming Foreign Relations 77

cutting expenditures, he accumulated very large silver reserves. He

patronized a team of talented officials: Wang Chonggu (who negoti-
ated a peace agreement with Altan Khan of the Tmet Mongols);
Pan Jixun (who engineered a reconfiguration of the lower reaches
of the Yellow River); and Qi Jiguang (who trained and led effec-
tive defense forces against pirates along the coast and Mongols to
the north). Zhangs nationwide land tax reform was cut short by his
untimely death in office. His arrogance and harsh repression of crit-
ics ensured his posthumous condemnation and the undoing of most
of his reforms.
Ming China lasted 276 years. During that long era of unified
imperial rule, Chinas population may have tripled to nearly two
hundred million, with at least a commensurate quantitative growth
in the national economyless centered on large urban nodes as in
the Song, but quasi-rural, decentralized, and heavily competitive. The
Ming state was too understaffed to be able to tax this growth effec-
tively, nor, as rising quantities of silver bullion from Japan and South
America entered the country through illegal trade channels during
the sixteenth century, could it control the national monetary sup-
plyas the Song state had been able to do half a millennium earlier.
What did happen was that local officials made piecemeal adjustments
to the increasing availability of bulk silver, and gradually the dynastys
whole fiscal base shifted from labor services and taxes in kind to a sim-
plified lump-sum silver payment. China exported over three million
pieces of Jingdezhen porcelain westward to Europe, mainly in Dutch
ships, between 1602 and 1657. Ming silk and porcelain went eastward
too, in return for silver, after the establishment of the Spanish galleon
routes from Manila to Acapulco in 1564 and 1565. It has been argued
that Ming China was probably not an outpost, but the central focus of
an emerging global economy. It has also been argued that price infla-
tion in China brought on by the uncontrolled influx of foreign silver
helped occasion the collapse of the Ming in the 1640s.

Ming Chinas posture in the larger world of its time was, thus, riddled
with contradictions. Foreign relations theory gave China pride of place
as the ethical center and moral arbiter of the universe, symbolized in
the so-called tribute system, with its extremely strict controls on all
relations with foreign states and foreigners. It was never possible to
78 Permanent Unity Largely Achieved

bring Japan comfortably into this system, and so all official links with
it were cut in 1549. That helped boost predatory Japanese piracy along
the China coast. Despite official interdicts on private foreign trade,
there grew during the course of the Ming an international maritime
community, featuring Japanese, Dutch, Spanish, Portuguese, as well
as Chinese sea traders and rovers, and a trading network extending
from the coast of China to Japan, Southeast Asia, and from there to
Europe and the Americas. Likewise with the Mongols: the Ming was
seldom able to do better than to encourage tribal anarchy, but anarchy
encouraged continued Mongol raiding along the increasingly walled
frontier. The frontier area was one of great intermixing, with many
Chinese peasants illegally settled in Mongol territory under Mongol
protection; large numbers of Mongols, many of them soldiers in Ming
service, settled inside China; and Ming border troops carrying on a
lively trade in contraband with the enemy.
The Ming made some grudging exceptions to the official trib-
utary model, partially lifting the ban on private trade and quietly
granting a trading base to the Portuguese at Macao in 1557. In 1583,
the Jesuit missionary Matteo Ricci (15521610) was given permis-
sion to live in China, and in 1598 he was allowed to move to Beijing.
The Jesuits, despite their small numbers, made many thousands of
converts in the late Ming, and not just among the educated elite. In
Beijing, they had good success introducing mathematical cartogra-
phy, Euclidean geometry, calendrical astronomy, mining technol-
ogy, and advanced gunpowder weaponry to the Ming government.
(There was a parallel burst of interest among the late Ming literati in
Chinese technologies, including especially agriculture, and in a dis-
passionate and thorough review of all problems of national admin-
istration). Jesuit books and annual letters, describing the fall of the
Ming dynasty and the rise of the Qing, brought current events in
East Asia to public attention in Europe for the first time.

The Late Ming: Internal Developments

W hile late Ming China developed economic, cultural, and tech-

nological ties to the outside world to a degree perhaps not seen
since Tang Chinas links to Central Asia, Iran, India, and Buddhism
 The Late Ming: Internal Developments 79

nearly a millennium earlier, those ties, though significant, did not

override in importance developments specific to China and East Asia.
Earlier under the tutelage of Zhang Juzheng, the Wanli emperor, as
he came of age, became interested in military affairs. The civil bureau-
cracy would have none of it. So, Wanli simply ignored the bureaucracy
and developed a parallel systema corps of eunuch tax collectors
and supervisors of mines whose job was to provide revenues sufficient
(when combined with the savings amassed by Zhang) to fund war mak-
ing. This made possible several large campaigns, including especially
Wanlis military interventions against the Japanese in Korea in 1593
and 1597, operations that enjoyed some success and ended when the
Japanese leader Hideyoshi died in 1598 and the Japanese withdrew.
In 1619, a special nationwide tax surcharge was imposed to finance a
major punitive expedition into Manchuria to dispose of a pugnacious
Jurchen leader named Nurhaci. This time, despite Ming superiority
in numbers and in advanced European firearms, the expedition was
poorly planned and badly led and was never able to use its accustomed
siege tactics; the Ming armies were decisively routed by Nurhaci in a
series of open-field encounters.
The summer of 1620 witnessed the death of Wanli, the accession
and sudden death of his son and successor, the Taichang emperor,
and then the succession of Wanlis fourteen-year-old grandson, the
disastrous Tianqi emperor (r. 16201627). Rather like the Hongxi
emperor in his short reign in 1424, Taichang sponsored a major policy
reversal during his single month of rule in the summer of 1620, put-
ting an end to militarism and eunuch paragovernment and restoring
to primacy the civil bureaucracy, now dominated by the moralizing
Donglin faction. Under Tianqi, policy shifted back again. In 1624,
the Donglin party staged an all-out protest. Led by palace eunuch
Wei Zhongxian, the anti-Donglin officials in the bureaucracy and
central police arrested and tortured to death some thirteen leading
lights in the Donglin faction and purged hundreds of others from
office. Meanwhile, the Jurchen khan Nurhaci inflicted another ugly
defeat on Ming forces in Manchuria. But in 1626, under the aegis of
Wei Zhongxian, the Ming defenders of the city of Ningyuan, thanks
in large part to the accurate fire of their Portuguese cannon, at last
put Nurhaci to rout. Nurhaci died later that same year. For a brief
moment, things looked promising for the Ming.
80 Permanent Unity Largely Achieved

However, Tianqi died without an heir in 1627 (his five children

had all died in infancy). He named and was succeeded by his younger
half brother, the Chongzhen emperor, aged sixteen. Chongzhen
was conscientious. He sponsored yet another top-level policy shift.
Wei Zhongxian and his adherents were purged. Wei committed sui-
cide. But so degraded had political life become as a result of the
extreme violence provoked by the Donglin, and so unremitting the
thirst for revenge, that the petulant Chongzhen, conscientious as
he may have been, could never harness the bureaucracy effectively
and get it to face the grave problems that now threatened the realm
on all sides. Meanwhile, the Confucian moralists regrouped as the
Restoration Society (Fu she), a nationwide league of discussion
groups that argued, much as the Donglin had done, that a national
spiritual awakening and renewal must precede any practical reform
effort.

The Ming Collapse

T he Chongzhen reign (16271644), the last of an intact Ming

dynasty, was not a happy time. A corrupt provincial and local
bureaucracy; a monetary system in which the state repeatedly debased
the copper coinage; and subsistence crises in many parts of the coun-
try, but especially in north Chinathese factors helped prompt wide-
spread banditry that, with the rise of Li Zicheng and Zhang Xianzhong
in Shaanxi from 1628, began to assume threatening proportions. Li
and Zhang were two among many military deserters from the north-
ern frontier who gathered distressed peasant youths and began mobile
raiding operations, first in Shaanxi, and soon spreading elsewhere. If
these uprisings were the only problem facing the Ming court, they
could have been managed. A civil official, Hong Chengchou, in
charge of military operations, inflicted a series of severe defeats on Li
and by 1638 had all but captured him. But then Hong was transferred
to the Manchurian front, where, by 1639, Nurhacis successor Hung
Taiji had renamed his people the Manchus, declared a Chinese-style
Qing dynasty, subdued Korea, and absorbed the eastern Mongols.
He became patron of the Dalai Lama and the Yellow Hat sect of
Tibetan Buddhism, welcomed Chinese civil and military defectors
 The Ming Collapse 81

(including Hong himself in 1642), and made several deep raids into
Ming territory. Thanks largely to this Manchu activity, Li gained a
reprieve. A horrible drought in Henan in 1639 and 1640 prompted
Lis reemergence in that province, where he raised havoc, conquered
walled towns, and was conducting a long siege of Kaifeng in 1642
when a catastrophic flood of the Yellow River, fed by heavy rains,
completely destroyed the city. Li withdrew west to Shaanxi in 1643,
where (shades of Huang Chao in 881) he declared a dynasty, con-
ducted civil service examinations, and at last made a serious attempt
to shift from banditry to orderly administration in a bid to succeed the
Ming. A well-executed two-pronged military assault seized Beijing on
April 25, 1644, with a minimum of disorder. Rather than negotiate
an abdication, the Chongzhen emperor hanged himself. Nominally,
Lis Shun dynasty now ruled China, but only for forty daysin early
June it collapsed.
The Shun did show some populist features. Rather than tax the
poor, the Shun regime funded itself by plundering the rich. Lis men
seized the imperial Ming treasury. They forced imperial princes and
in-laws to disgorge huge sums. They crudely graded eunuchs, mer-
chants, great landlords, and high officials by income level and took
some 20 percent to 30 percent of their riches. The Shun dynasty,
within weeks, is reckoned to have collected some seventy million taels
of silver altogether. The regime revived Tang official nomenclature.
Positive references were made to the early Tang equal field system.
Much of north China accepted Shun-appointed magistrates. Many
officials and literati deserted the Ming cause and accepted the Shun.
As emperor, the ex-bandit Li lived modestly and seemed to want to
rule responsibly. What, then, went wrong? Why did the Shun not suc-
ceed as successor to the Ming?
There were two sets of difficulties, one internal to the Shun, the
other relating to Lis relationships to external forces and their lead-
ers. Internally, Lis military elite, of bandit origin like himself, was
loath to accept civilian control. Very shortly after their orderly entry
into Beijing, they cast off all restraint to engage in arbitrary arrests,
torture, murder, and looting. Externally, there were too many failed
negotiations, stemming from Lis visceral dislike for defectors. In
1641, for example, Li did not take advantage of an opportunity to
bring aboard fellow rebel Zhang and his forces. In February 1644,
82 Permanent Unity Largely Achieved

while Li and his army were still on the march to Beijing, he received
but ignored a letter from the Manchu regent Dorgon proposing joint
operations. In April and May, Li fumbled away any hope of joining
forces with Wu Sangui, the leading Ming commander still serving
on the Manchurian frontier. Wus family, living in Beijing, had been
mistreated by Lis men. Wus father was made a hostage. In May
1644, Wu proposed and Dorgon agreed to join forces against Lis
attack on Wus position. Lis army, gorged with Beijing loot, seemed
to lack interest in the fight. It was routed. The Shun collapse was
immediate and catastrophic. The joined forces of Dorgon and Wu
entered Beijing on June 6, 1644. The Shun fled in disarray back to
Shaanxi. Sometime in 1645, Li disappeared, at the approximate age
of thirty-nine. He was never seen again. The Manchu conquest of
China had begun. Like the Mongol conquest four centuries earlier,
the Manchu takeover would take half a century to complete.

Qing China: The Founding

from 1644

T he Manchu Qing conquest machine featured so many weak-

nesses that, far from being inevitable, its eventual success seems
hard to account for. Among these weaknesses was the poverty of their
Manchurian base; the perennial outsiders problem of reconciling
tribal tradition with Chinese institutional norms; frictions among
Chinese, Manchu, and Mongol ethnic groups; and murderous rival-
ries among the members of the ruling Aisin Gioro clan. These weak-
nesses would be enough to explain a Qing failure, had it occurred.
But of course, the Qing in the end created the largest of all the
China-centered empires, larger than even the Yuan or the Tang. And
the Qing dynasty, from its declaration in 1639, somehow survived
every kind of challenge until finally it abdicated to Yuan Shikai on
February 12, 1912.
The nominal founding emperor of the Qing dynasty in China,
known by the reign title Shunzhi (r. 16431661), was only six years
old in 1644 when the Qing forces took Beijing. He suffered from
what appears to have been tuberculosis and died in 1661 at the
young age of twenty-three. Although volatile in temperament, he
 Qing China: The Founding, from 1644 83

was also exceptionally studious and a committed devotee of Chan

Buddhismtraits perhaps more to be expected in a sovereign of a
mature or failing dynasty than in the ruler of a rising one. But while
he lived, Shunzhi ruled with vigor. He seized control of the reins
of state not long after the the death of the dictatorial prince-regent
Dorgon in 1650. He then went so far as to reformulate Manchu pol-
icy and restyle its institutions in a pro-Chinese direction and to spon-
sor severe crackdowns on bureaucratic corruption and factionalism.
Shunzhi was succeeded by another minor, his seven-year-old son
Kangxi (r. 16611722). But Kangxi and the next four emperors, who
ruled into the middle of the nineteenth century, were each blessed
with long or very long reigns, and each showed an active interest in
administration. On balance, their successes outweighed their many
mistakes and failings. Securing good rulers was in large part a prod-
uct of a Qing innovation: instead of Ming-style automatic primogen-
iture, with not even regencies permitted for child emperors, Qing
rulers secretly chose their successors from among their many eligible
sons. Not even the heir apparent knew who the choice was until his
fathers death, when the successors name was at last revealed.
Key to the success of the early Qing enterprise was the Manchus
creative construction or reconstruction of Mongol and Chinese mili-
tary and political institutions and the all-inclusive patronage they
accorded to religious and intellectual traditions of various kinds.
From very early on, the Manchus adopted the alphabetic script and
the military organization of their Mongol neighbors, intermarried
with the leading eastern Mongol clans, and absorbed Mongol fight-
ers into their emerging conquest machine. Also from early on, while
still confined to Manchuria, they absorbed local Chinese fighters
and administrators and encouraged defections from the Ming side,
especially by frontier generals such as Hong Chengchou, together
with their armies. Without the Manchus welcoming of Hong and
other defectors, the coming Qing conquest of China would have
been impossible. Also, starting from the 1637 visit of the fifth Dalai
Lama to Mukden (Shenyang, the Manchurian capital), the Qing rul-
ers were taken into the Yellow Hat Buddhist pantheon as incarnations
of the bodhisattva Majusr, that is, as living objects of worship, and
this went far to make the later Qing conquests of Mongolia and Tibet
acceptable to the faithful of those vast regions. (The emperors them-
selves, however, remained patrons, not converts; and they made no
84 Permanent Unity Largely Achieved

attempt to convert the Manchu people, who were encouraged to stay

with their inherited shamanic religious traditions.) The Jesuits, mean-
while, purveyed the arts and technologies of contemporary Europe to
the Manchu rulers, who were eager to make use of them in the inter-
est of empire building. These included gunnery, cartography, calen-
drical astronomy, and even the diplomatic skills that were used at the
negotiation of the 1689 Nerchinsk agreement with Russia. And, once
permanently based inside China after 1644, the Manchu rulers had
to place themselves and their government convincingly into the for-
midable historical, institutional, and literary framework bequeathed
to them by the civilization of China.

The Conquest of South China

T he Qing conquest of south China was accomplished with the

indispensable help of Chinese armies commanded by the turncoat
Ming generals Wu Sangui (16121678), Geng Jingzhong (executed
1682), and Shang Kexi (d. 1676). These so-called Three Feudatories
were granted a wide share of autonomy as reward for their successful
assault on the Southern Ming (established at Nanjing in June 1644
and chased from there a year later, with several courts in flight until the
demise of the last of them in Burma in 1659, its desperate conversion
to Christianity and its plea for help from Pope Innocent X in Rome
notwithstanding). Then in 1673, the Kangxi emperor made the rash
decision not only to accept the aged Shangs request to retire, but also
to try to abolish his Guangdong-based princedom. Afraid of becom-
ing the next targets, Wu and Geng, based in southwestern China and
in Fujian, respectively, decided to rebel. Eight years of civil war fol-
lowed. Geng surrendered in 1676. Shangs son surrendered in 1677,
then rebelled, but was captured and forced to commit suicide in 1680.
The strongest feudatory, Wu, declared his own Zhou dynasty late in
1673 and commenced a military reconquest of China that came very
close to success, but Wu died of illness in 1678. Qing forces besieged
and finally destroyed his grandson and successor in Yunnan in 1681.
The Kangxi emperors victory over the feudatories had less to do with
military superiority than with political skill, especially the coaxing and
handling of defections from the rebel side.
 The Conquest of South China 85

Why did the so-called Southern Ming not regroup like the Eastern
Jin (318420) or the Southern Song (11271279) and create a new
regime, stable but reduced in size, in south China? Or follow Ming
Taizus example and build power by organizing the intellectuals, offi-
cials, bandits, sectarians, rioters, local militias, and all the other detritus
of dynastic collapse? Whereas Eastern Jin governance is known only
dimly, it is clear that the founding of the dynasty involved mass immi-
gration from the north, semi-independent warlords, very few bureau-
crats, a reasonably competent founding emperor in Sima Rui, and a
general consensus that the reconquest of the north was the regimes
main and overriding purpose (as it was for Chiang Kai-sheks national-
ists on Taiwan). Parts of the Eastern Jin profile fit the Southern Song
as well. Even though the Qing and Southern Ming forces seldom
came directly to blows, the Southern Ming, like its Jin and Song pre-
decessors, had to rebuild institutions completely and deal somehow
with semiautonomous armies, despite a shortage of officials capable
of managing the task. Where the Southern Ming faltered was in badly
handling not just a few things, but almost everything. There were not
one but five rivalrous Southern Ming courts. All the Ming princes who
would be emperors were incompetent and deeply unpopular. Lethal
hatreds among the high officials, and an inability of any of the five
courts to make common cause with roving armies, doomed the whole
enterprise. The Eastern Jin lasted a century; the Southern Song lasted
a century and a half; the so-called Southern Ming barely survived
for eighteen years. But it could pride itself on its ethical posture: the
Southern Ming boasted thousands of anti-Qing suicides and willing
martyrs, men and women, officials and common people, far beyond
the Wen Tianxiang affair of the late Southern Song, far beyond any-
thing known to history thus far.
The leading nominal defender of the Southern Ming cause was
the Zheng family, who for four generations raided Qing-held territory
from their fleets and fortresses, located principally along the Fujian
coast. Theirs was not the traditional continental power. The Zheng
familys power was based, rather, in the greater East Asian maritime
world, centered on the China coast, with lucrative trade links to Japan,
Macao, the Philippines, and Southeast Asia, and connections and
rivalries with such competitors as the Japanese, Portuguese, Spanish,
Dutch, and others. The most famous of the Zhengs, Zheng Chenggong
86 Permanent Unity Largely Achieved

(Koxinga, 16241662), whose Fujianese father was a Christian con-

vert, was actually born near Nagasaki of a Japanese mother. Koxinga
was able to gather huge fleets and very large fighting forces. Although
he had once been a Confucian student, attending the national col-
lege in Nanjing in 1645, he never courted the literati, nor was he
ever influenced in any important way by the traditional Confucian
moral and political teachings. (His openly hostile relationship with
his father, who defected to the Qing in 1646, was un-Confucian in the
extreme.) Koxingas power base was, by choice, commercial and mili-
tary. His loyalism on behalf of the Southern Ming was for a remote
cause that could in no way interfere with his freedom of action. The
eventual failure of the Zheng phenomenon has roots in effective Qing
countermeasures; in a downturn in the international maritime econ-
omy; and in Koxingas own cruel and occasionally crazed behavior
(and untimely death at the age of thirty-seven). The Qing was able
to bring about key defections from the Zheng regime, particularly
that of Admiral Shi Lang in 1646. Then during the 1660s, the Qing
authorities forced the evacuation ten or so miles inland of the entire
southeastern coastal population, creating a dead zone, thus depriving
Koxinga of food and other resources. The Qing thwarted Koxingas
1659 attack on Nanjing; and so he and his fleet withdrew to Taiwan,
where in 1660 he compelled the Dutch to yield to him. In 1683, Shi
invaded Taiwan on Qing behalf and finally put an end to the Zheng
enterprise. The Kangxi emperor treated the last Zheng chieftain well.
Taiwan he considered an insignificant acquisition, a useless little ball
of mud. Thus as things turned out, neither Fujian nor Taiwan would
harbor any independent maritime power, any Venice. Both ended up
wholly swallowed into the Qing-administered continental system.

Beyond China:
Building the Qing Empire

A s soon as the China domain was cleared of all serious competi-

tion (1683), the Kangxi emperor directed his attention to estab-
lishing the Qing as the dominant power in greater East and Central
Asia. The Qing was not simply a Chinese-style dynasty. From the
 Beyond China: Building the Qing Empire 87

time of Nurhaci, it was also an Inner Asian khanate, with a differ-

ent, mainly non-Chinese repertoire of policies and a different set of
military, political, and religious institutions. In this respect, it bore
resemblances to the Liao and Yuan.
The Amur River region of outer Manchuria, thinly settled by
hunting and fishing tribes speaking Tungus dialects distantly akin
to Manchu, had since the 1640s attracted increasing numbers of
Russian Cossack pioneers eager to impose iasak (demands for sable
and other furs) on the natives and build forts and settlements for
themselves. Qing military intervention was sporadic, until, in a tri-
umph of superior numbers, artillery, and well-organized logistics,
the Cossacks were driven out by 1686. With the help of Jesuit nego-
tiators, the famous and far-reaching Treaty of Nerchinsk was drawn
up in 1689, which delimited clearly the frontier between the Qing
and the Romanov empires in the Far East, established mutual secu-
rity that lasted until the 1860s, and was the first step in a truly grand
strategy in which the expanding Qing and Russian empires agreed
to maintain peaceful relations and not use Mongols or other Inner
Asian tribes against each other. (In 1727, Russia and China signed a
similar agreement delimiting the Mongolia-Siberia frontier.)
Kangxi next directed his attention to the Zunghar Mongols, based
in the Ili valley south of Lake Balkhash, where Galdan was building
a khanate (devotedas was the Qingto Yellow Hat Buddhism)
and aiming to annex Outer Mongolia, thus challenging the Qing
for supremacy all across Inner Asia. In a series of battles fought
in the steppes, two of them commanded by Kangxi in person, the
Qing by 1697 forced the Zunghars back west to the Ili. In 1720,
two Qing armies entered Tibet and Lhasa, ended the Zunghar occu-
pation there, and installed Kangxis candidate as the seventh Dalai
Lama, thus making the Qing court the defender and protector of the
Yellow Hats, a step that in time would go far to cement Qing power
in Buddhist Inner Asia.
An exceptionally vigorous monarch, Kangxi was a contemporary
of two other active rulers: Peter the Great (r. 16891725) and Louis
XIV (r. 16641715). Aside from Inner Asia, Kangxi also managed to
pay close attention to China. He made six inspection tours down the
Grand Canal to south China. As though to surpass the Ming founders
Six Maxims, he published in 1670 sixteen moral maxims to be read
88 Permanent Unity Largely Achieved

out periodically to the common people of China. He sponsored large

literary projects. He patronized the arts. He interested himself in the
Western sciences as propagated by the Jesuits. He kept Manchu tradi-
tions alive by taking part in seasonal hunts. He ruled as autocrat for
fifty-five years and died in 1722 at the age of sixty-eight. In a disputed
but probably legitimate succession, his fourth son followed him and
ruled as the Yongzheng emperor.

Refocusing on China: Yongzheng

r. 17221735

D uring Yongzhengs reign, the Zunghars slowed the pace of Qing

expansion in Inner Asia (in 1738, a Qing-Zunghar boundary
agreement was reached). Yongzheng devoted most of his attention
to China proper, developing a system of secret memorials through
which he bypassed the regular channels of official communication
and imposed his own personal scrutiny and detailed control over fis-
cal and personnel matters. Rather in the style of the Ming founder,
Yongzheng distrusted the degree-holding elite and insisted upon a
tight centralism under his own personal guidance. Through this
means, he almost single-handedly conducted complex tax and fis-
cal reforms and developed state-managed grain storage as part of a
national system of famine insurance, surely one of the most signifi-
cant acts of state intervention in the economy since the days of Wang
Anshi in the Northern Song, seven centuries earlier.
Yongzheng died suddenly at the age of fifty-six, possibly of elixir poi-
soning. His heir, secretly designated as such early on, became perhaps
the most powerful ruler in the whole history of ChinaYongzhengs
fourth son, Hungli, aged twenty-four, who would rule for most of the
eighteenth century as the Qianlong emperor (r. 17351799). In a
way, it is appropriate that he was a contemporary of such enlightened
European despots as Catherine II the Great (r. 17621796), Frederick
II the Great (r. 17401786), and Maria Theresa (r. 17401780). Like
them, Qianlong played many roles: in his case, autocrat, investigator,
generalissimo, art collector, litterateur, traveler, Confucian moralist,
Manchu traditionalist, historian, and holy man.
 The Reign of Qianlong, 17351799 89

The Reign of Qianlong

17351799

T he long and copiously documented reign of Qianlong can be

divided into two nearly equal halves. The first half ended with
the extermination of the last steppe empire, that of the Zunghars, in
1757, and the annexation of Eastern Turkestan (Xinjiang) in 1765. At
that point, a tacit but momentous decision was taken to expand the
Qing realm no further into Inner Asia. A century-long mission was
accomplished: eastern Inner Asia was wholly pacified with the incor-
poration of Manchuria, Mongolia, Xinjiang, and Tibet, and security
for China all along the troublesome northern frontiers was achieved.
No big collective goal remained undone. And so into the second
half of the Qianlong reign there gradually crept sloth, corruption,
aimlessness, and demoralization. The once-predominating ruler
allowed extraordinary powers to gather into the hands of a personal
favorite, the smooth and spectacularly corrupt Manchu guardsman
Heen. Suspicious of nearly everyone besides Heen, Qianlong still
unpredictably intervened from time to time in affairs, such as in the
notorious national witch hunt, the soul-stealing case of 1768.
Until about 1765, an argument can be made for considering the
Qing a member in good standing of a company of expanding Eurasian
empiresthe Russian and perhaps even the British included. The
Qing possessed the latest in gunpowder weaponry, mapmaking, dip-
lomatic techniques, and statistics collection. Central command,
communications, and logistics were already sophisticated when
Qianlong set up the machinery for the so-called Grand Council in
1736. The Grand Council was a kind of extrastatutory supergovern-
ment laid atop the regular bureaucracy, featuring its own person-
nel, archives, research groups, rapid communications networks, and
effective modes of command and control. Thanks largely to the work
of the Grand Council, Qianlong overcame a logistics breakdown
that, in Yongzhengs time, contributed to the Qing stalemate with
the Zunghars. It now became possible to tap the grain and other
resources of China proper to support commerce, colonization, and
military provisioning deep into Inner Asia. It was in Qianlongs time
that the Qing became the largest of all the China-based empires
90 Permanent Unity Largely Achieved

(the Peoples Republic is some 20 percent smaller). But from 1775,

because the Grand Council fell under Heens control (until 1799),
it could do nothing to stop the horrendous corruption, the profligate
costs, and the dubious results of Chinas late eighteenth-century wars
along the western and southern frontiers.
Yet even in decline, Qing central authority continued to exert
significant powers. It still tried to manage the national food supply,
albeit with decreasing effectiveness. It still managed the national
hydraulic system of dykes, channels, and polders along the disas-
ter-prone Hai, Yellow, Huai, and Yangzi river basins. From 1824 to
1826, the Daoguang emperor (18201850) was still able to direct
personally the huge and technically demanding job of rebuilding
the whole Grand Canal artery. For the most part, however, one finds
the central authority muddling along with too few personnel and too
small a tax base for such a large empire. Along the coast and in the
core parts of China proper, the late Qing approach became increas-
ingly accommodationist, with ever wider grants of authority given to
merchant guilds (such as the famous cohong, which from 1760 had
charge of all foreign trade in Canton). Judicial, educational, social
welfare, and a range of other functions were more and more con-
ceded to local lineages, self-organized along Confucian lines.
The Qing governments approach was otherwise along the fron-
tiers, heavily populated by non-Chinese ethnic groups or by Chinese-
speaking Muslims (Hui). Here, the late Qing regime preferred violent
repression to negotiation and accommodation. There was a special
difficulty with Qing rule over Muslims in Yunnan and Xinjiang, both
Hui as well as Turkic-speaking Muslims, the people later known as
Uighurs (not directly related to the Uighur Turks prominent in Tang
times). Acquired for mainly strategic reasons by Qianlong, Xinjiang
was for many decades an albatross, requiring economic and financial
assistance from China proper. And whereas the Qing emperors could
convincingly assert leadership over Confucians, Chinese and Tibetan
Buddhists, Catholic Christians, Manchus, Mongols, Tibetans, and
Chinese, they could never pose as spiritual leaders of Muslims. They
had to rule Muslims indirectly, through local begs and khwa-jas. In
Yunnan, Qing forces and ethnic Chinese militias cooperated in massa-
cres of the Hui, which eventually provoked the bloody Panthay rebel-
lion of 18561873. In Xinjiang, beginning in the 1820s, Han Chinese
officials pushed a powerful new agenda of encouraging massive Han
 Social Institutions in Political Context 91

(ethnic Chinese) immigration so as to transform the region and bring

it fully within the orbit of Chinese civilization. That was not accom-
plished without challenge. A series of jiha-d-style attacks on Xinjiang
from the Muslim Khoqand khanate west of the Pamirs led in 1832 to
an unequal treaty, whereby the Qing ceded to Khoqand commercial
and other rights in western Xinjiang. (This action bore some resem-
blance to the extraterritorial concessions the Qing made to Britain in
1842, after the Opium War). Violence in and over Xinjiang continued
for another half centuryand, indeed, up to the present day.

In 1850 there lay in Chinas immediate future internal rebellions of

historically unprecedented scale (especially the Taiping and Nian
rebellions of 18501868). Regional authorities, using traditional meth-
ods, eventually suppressed these, even while China was under his-
torically unprecedented pressure from the newly industrialized West.
The Opium War of 18391842 was just the first in a series of wars and
demands designed to force the Qing to open its ports and its markets to
foreign commerce and to compel the empire to deal with the West in
accordance with Western legal and diplomatic norms. During the first
half of the twentieth century, the old system collapsed completely. The
dynastic state was trashed. Revolutionary elites targeted the old-style
extended Confucian family for destruction. Confucian doctrines were
utterly discredited. But now, in the early part of the twenty-first century,
there has been taking place from the bottom a surprising recrudescence
of parts of the traditional institutional repertoire, including a revival of
the extended family system and of the teaching of Confucian ethics. It
remains to be seen whether a hereditary leadership (as in North Korea)
will eventually emerge or, much more likely, a continuation of some-
thing vaguely like the old examination-based meritocracy, a new man-
darinate of degree-holders in engineering and business management
that would hold the dominant role in the governance of the country.

Social Institutions in Political Context

T he question of the ways in which the people of China engaged

with the state, or withdrew from it, or rebelled against it, is exceed-
ingly complex. The further back in time one probes, the harder it
becomes to fix a confident social perspective on it. Generally speaking,
92 Permanent Unity Largely Achieved

the earlier the time period, the more surviving sources consist of cen-
trally focused histories and other compendia, and often only archaeol-
ogy can help provide a window into life at ground level. By the Ming
and especially the Qing, however, the rise of population, the spread of
literacy, and the growth of book printing and publishing created such
an overwhelming mass of local and family records, much of which still
survives, that no one person can ever hope to read it all. But early or
late, the imperial state, which at times seemed to exist only for its own
sake, had to help shape, or at least accommodate itself to, the broad
institutional contours of the people that it ruled. It was probably the
early Ming state that went the furthest of all in its forceful attempt to
remake Chinese society institutionally.
When one looks at the whole period from 150 to 1850, a trend
that stands out is the rise to prominence and then the slow decline of
religious institutions. Through early Tang times, Buddhist temples
and temple estates in the countryside acted as little central points
for economic redevelopment and as schools and welfare agencies
for the people nearby. Then from at least the ninth century, and
extending well into the Song and beyond, the Buddhist and Daoist
churches, with their physical infrastructure, their sizable communi-
ties of monks and nuns, their state patronage, their holy sites small
and large, and the private fervor and devotion they attracted, under-
went a long process of decline. Yet they never disappeared altogether.
There were occasional revivals, such as the late Ming Buddhist
revival. Both Buddhist and Daoist temples certainly remained part
of the urban landscape, even as their physical presence in the coun-
tryside shrank, often by forcible dispossession, their place taken by
Confucian schools and corporate lineage estates.
Until Ming times, roughly speaking, the Buddhist establishment
had been able to ingratiate itself with Chinas age-old preoccupa-
tions with ancestors, tomb rituals, family, and filial piety. But the
leaders of the Neo-Confucian movement of Song times and later
laid the intellectual groundwork for what turned out to be the suc-
cessful rescue of Chinas familial institutions from the embrace of
the churches. This involved the development of rules and formats
for the composition of genealogies; the development of authoritative
guidebooks for the conduct of family rituals (coming of age, mar-
riage, death, and posthumous worship); the creation of regulations
 Social Institutions in Political Context 93

for the rearing and education of children; and rules for the carrying
out of daily family governance.
Until the Yuan period, there was a limitation, reemphasized by no
less an authority than Zhu Xi, on how far a family could legitimately
expand the size and scale of its formal organization. The Confucian
classics (and Tang aristocratic precedent) placed limits on the
number of generations back in time a family could reach in identify-
ing and worshipping an original ancestor. Peasants could go no fur-
ther back than a grandfather. Educated men could go further back,
how far depending upon how high they had risen in state service. Of
course, the further back one was allowed to go to identify an ances-
tor, the larger the group of descendants presently living that could
be gathered together as a coherent lineage. The only way for peas-
ants or merchants legitimately to circumvent that severe size restric-
tion was somehow to found a family all of whose male descendants
agreed not to divide their inheritance, but to keep living together
communally on an undivided estate (thus the famous Zheng family,
commoners of Pujiang county in Zhejiang; and the Lu family, drug
merchants in Jinxi in Jiangxi, kinsmen of Zhu Xis philosophical
rival, Lu Xiangshan). But these were fairly rare occurrences.
In the Yuan, curiously, it was a Mongol law relating to widows
together with a Neo-Confucian desideratumthat broke these long-
standing limitations on family-based organization in China and
opened the way to the formation of the large lineages that became,
not rarities, but quite typical features of Chinese society in the Ming
and Qing. Mongol law compelled the widow to remain in the fam-
ily of her dead husband, where her labor and assets remained at her
husbands familys disposal. She herself could be forced to marry
one of her husbands relatives. Applied to China, the law was soon
adjusted along Confucian lines to allow the widow not to remarry
a relative of her dead husbands if she so chose; but in 1330, the
Ministry of Rites further ordained that if a wife or widow left her
husbands family, she had to leave her assets behind, and that only
her dead husbands family could arrange a remarriage for her. This
put an end to the Song rule that gave women personal control of
their dowries and other assets and, if widowed, permitted them to
return to their natal families and remarry whomever they chose. This
ruling had a direct impact on litigation in the courts of the county
94 Permanent Unity Largely Achieved

magistrates and so led by degrees to the creation of stronger lineages

nationwide. By 1350, in parts of south China (especially southern
Anhui), one could find local elites writing and updating lineage
genealogies, building Confucian-style ancestral temples, and fund-
ing ritual gatherings and elementary schooling from the proceeds of
endowed corporate estates. (In 1000 CE, hardly any of that would
have existed. Northern Song society, in its intense focus on the cen-
tral state, was more closely comparable to early Tang society than to
the much more heavily localized Ming or Qing.)
The spread was gradual. Only from the mid-Ming (the sixteenth
century) did lineages fully form in Fujian, Guangdong, Jiangxi, the
lower Yangzi region, Shaanxi, and elsewhere. First came arduous
research efforts. Often it was the aspiring lineages junior literati
who volunteered, or were asked, to locate and identify all knowable
ancestors, account for descendants, compile and publish the results,
and persuade senior officials who were also well-known literati (but
usually not members of the lineage itself) to validate the completed
genealogy by writing an approving preface for it.
That was early on. After a couple of generations, there emerged
among the high literati a willingness to overlook problems of poor
or missing evidence that had inhibited the earlier genealogists and
to agree to endorse the construction of large associations of kinsmen
sharing the same surname whose exact genealogical relationships
were often murky. Individual landowners were persuaded to donate
land or money to the building of main and branch ancestral temples,
whose endowments might be used to fund any or all of a range of pur-
poses: periodic rituals; schooling for children and juniors; food relief
and other assistance to the poor; help in emigrating to other parts of
China; land reclamation; business investments; litigation on behalf
of members; local antibandit defense; and so on. The extent and scale
of lineage formation, and the purposes to which lineage corporations
were put, of course varied from one part of China to anotherfrom
the very highly organized and powerful lineages of Guangdong and
Fujian to the smaller and often unendowed common-descent groups
of north China, and all sorts of variants in between.
Aside from the change in the law that boosted patrilineal author-
ity, there was another and quite different development that, inadver-
tently perhaps, helped local elites establish an identity and later form
 Social Institutions in Political Context 95

lineages. Following the horrific battering the rebellions and wars of

the 1350s and 1360s inflicted upon China north and south, the early
Ming dynasty, in its utopian drive to remake the civilization, intruded
more deeply into society at ground level than any authority since the
early Tang. This included the creation of a nationwide apparatus for
the imposition of labor services in person and the collection of taxes
in kind. County units were divided into cantons, cantons into town-
ships, and, most important, townships into wards (li). Each ward,
in theory, was made up of 110 landowning households, occupying
a few square miles of land. The heads of the ten richest families
shared management of the tax collections, for which they were per-
sonally liable in case of default. This was rural social organization as
imposed from the early Ming center, at Nanjing. But this so-called
lijia system, coercive as it may have been at the outset, soon estab-
lished itself as a source of local social legitimacy: the descendants of
the original leaders came to base their social status and honor, their
very identity as local elites, upon their lijia membership as estab-
lished early in Ming. This remained the case even after the labor-
based lijia system itself yielded to the property-based single whip
tax system during the sixteenth century.
Lijia institutionalized social status in another way. Taxes and
services involved two forms of participation. People could provide
physical labor, or they could provide supervision and management.
The elite were, on the whole, willing to perform managerial duties,
despite the risks. Uneducated men, smallholders or tenant farmers,
did the actual hauling of the grain tax and performed required ser-
vices such as policing, public works, running messages, and the like.
The power to identify these workers, assign them their tasks, and
direct their work fell to the heads of the richer landowning families.
Thus, national policy contributed heavily to the solidification and
perpetuation of a rather rigid rural class system across the length
and breadth of China. The upper class of comparatively affluent
and educated families typically gave themselves identifying choro-
nyms, usually based upon the name of the ward they lived in (thus
the Peach Spring Xiao lineage of Taihe county, Jiangxi, and many
thousands more like it). By the sixteenth century, the enterprising
individual landlords and estate builders of early Ming gave way to
corporate boards of lineage managers. Informal organization yielded
96 Permanent Unity Largely Achieved

more and more to rule-bound procedures. As local economies came

to be based heavily on silver, and as taxes in silver assessed upon
land came to replace taxes in kind, the lineages developed corporate
estates that could not easily be divided or sold and so escaped the
leveling effects of Chinas system of partible family inheritance.
There was a dark side to the emerging lineage order. It was not
poor people who were responsible for Chinas demographic growth
(approximately a tripling during the Ming, and nearly again dur-
ing the Qing); it was the educated and the affluent. They practiced
polygyny and also female infanticide. Known population growth rates
among the Ming upper (managerial) class show that it was capable
of doubling its size about every seventy years. This should have led to
a nationwide population approaching Chinas current population of
1.3 billion by the end of the Ming. Emigration from densely settled
parts of China to new frontiers within China was what allowed the
total Ming population to triple as it did. But the reason why popula-
tion rose to nowhere near a billion was because the lineages regularly
consigned some major portion of each new generation to downward
social mobility. The scarcity of women created during each genera-
tion a large class of young men who could never hope to marry.
Their usual fate was to become drifters, join gangs, turn to banditry,
be sold, or sell themselves into bondage. Additionally, even in a prov-
ince like Guangdong, heavily populated by powerful lineages, per-
haps some 70 percent of the men of the province belonged to small
or weak families that lacked special privileges or protection. Late
imperial China came to feature a massive and unhappy reservoir
of victims of debt, downward mobility, and bondage; a lower class
of boatmen and porters, of drifters and bandits; of criminal gangs
and sectarian adherents; of despised and legally disfranchised groups
such as musicians, beggars, and the like that the Yongzheng emperor
tried to emancipate in the 1720s and 1730s. There were more than
sufficient labor reserves at hand to fuel the widespread bond servant
rebellions and other disorders of the late Ming, as well as the huge
Taiping and other upheavals that engulfed China in the middle of
the nineteenth century.
The lineages of Ming and Qing times were extremely durable
at the local level. They have been styled an aristogeny. How did
they differ from the so-called aristocracy of the age of disunion, Sui,
 Social Institutions in Political Context 97

and early Tang? For one, they were much more formally institution-
alized. For another, their ties to central government, though still
important, were much thinner. Chinas population growth over the
preceding thousand years explains why: from fifty million or less to
two or three hundred million and more, without a commensurate
increase in the size of government. No lineage, or small group of
lineages, could hereditarily dominate government in the Ming or
Qing, as was sometimes possible in earlier times, especially given the
hurdle of the open examination system and the constant personnel
turnover it ensured.
Chinas social great leap from the restricted lineage group of ear-
lier times to the open system of the Ming and Qing was mirrored,
indeed anticipated, by the behavior of the imperial lineages of Song
times and later. The Li imperial lineage of the Tang, which was man-
aged by a special institution, the Court of the Imperial Clan, followed
the classical models and reduced all distantly related kinsmen to com-
moner status. The Northern Song, ruled by the Zhao family of military
origin, created a new pattern. Taking advantage of the ongoing debate
in Confucian circles over the appropriate size of lineages, the Song
court under Zhezong (r. 10861100) decided that the Zhao lineage
should include all descendants and no longer observe the Tang-style
restrictions. But also unlike the Tang, the Northern Song court from
the outset barred all kinsmen from military or civil office and forced
them to live in what has been called functionless luxury in the capi-
tal, Kaifeng. They were expensive to maintain; by the late eleventh
century their numbers had swelled to some five thousand, and two
satellite settlements had to be created for them elsewhere in north
China. The Zhao kinsmen suffered grievously in the Jurchen invasion
of the early twelfth century. The Southern Song made major changes
in policy regarding the now much-reduced Zhao lineage. Members
were afforded special access to the body of civil service examination
candidates and, if successful in the competition, they were permitted
to take office. Many did. Many more became military officers. Special
lineage communities, funded by local resources, were placed in the
wealthy Fujian port cities of Quanzhou and Fuzhou. By the end of
the Southern Song, the imperial lineage had grown very large indeed,
but it fell well short of the estimated eighty thousand males the Zhu
lineage, descendants of Ming Taizu, reached by the early seventeenth
98 Permanent Unity Largely Achieved

century. Like their Northern Song counterparts, the Zhu were forbid-
den to take examinations or enter government by other means until as
late as 1595. Despite enormous total annual allotments of grain, many
distant Zhu relatives became poor. The imperial Aisin Gioro lineage
of the Qing reached some twenty-nine thousand by the end of the
dynasty. Members could take exams and had some access to power
positions, unlike their Ming counterparts, and more in the style of the
much smaller group of descendants of Chinggis Khan in Yuan China.
The demographic behavior of the Ming emperors was very like that
of the Chinese upper class generallyheavily lopsided sex ratios in
favor of males; polygyny; and population growth rates approaching
2percent per annum. All these imperial lineages seem to have been,
on the whole, a useless drain, forced into parasitism and contributing
little or nothing to the economy, the security, or the political leader-
ship of the country.

Lineagesthese private, kin-based corporations based in Confucian

doctrinedid not exist in defiance of the imperial state. Indeed, they
owed their existence to the states protective embrace. Lineage lead-
ership, almost without exception, consisted of men who had secured
honors, exemptions, titles, offices of many kinds, and academic degrees
from the imperial state. The rarest plum in the recognition-granting
cornucopia that the state controlled was of course the metropolitan
examination degree (jinshi), earned in an intense series of competi-
tion rounds and almost a guarantee of high office, fame, and wealth.
And there were countless other, less distinguished tokens that the
state might confer. Each such token guaranteed the recipient a cer-
tain level of privileged relationship to the world of government. Since
Tang times, the examination system produced many more educated
aspirants than could ever hope to obtain actual bureaucratic posi-
tions. It was this great surplus that made possible the strongly localist
turns that China took in the Southern Song, the Yuan, and again in
the mid-Ming and the early Qing. The central states reaction to these
localist turns was to offer friendly support, but also to train a suspi-
cious and wary eye on them. Lineages were not incorruptible. Their
leaders were always tempted to disown the poor and the weak, sell
genealogical credentials to rich interlopers, and fight other lineages
over such desirable resources as tomb sites, fields, markets, and water
 Social Institutions in Political Context 99

rights. Like little empires, lineages were also prone to keep expand-
ing in size and power until forcibly stopped. Yet, it was by recruit-
ing Hunan lineages that Zeng Guofan was able to create forces large
and powerful enough to suppress the Taiping rebellion in the 1860s.
Clearly, the existence of lineages was more an advantage than a dis-
advantage to the late imperial state system.

The civil service examination system of the Ming and Qing was a
colossal national enterprise conducted according to a rigid schedule
with utterly dependable regularity. For the most part it was diligently
policed, so major fraud, corruption, and scandal were rather rare.
It may be useful to an understanding of the system to follow the
career of a hypothetical child and later adult as he negotiated all the
rungs of the ladder. Perhaps the childs immediate family was poor,
but the elders of the larger lineage of which his family was a compo-
nent noticed the boys academic talent and so invited him to join his
young cousins in the lineage elementary school. This prepared him
for his first hurdle, a test to qualify as a licentiate, a local government
student, when he reached the approximate age of sixteen. There were
in all some 1,500 prefectures and counties with government schools,
with paid instructors who early in the Ming actually taught the stu-
dents; but the schools soon evolved into registration and testing cen-
ters only, leaving the students to seek instruction in private Confucian
academies or in their own informal study circles. The status of these
licentiates was probationarythey had to keep requalifying by fre-
quent testing. At the beginning of the Ming, all licentiates (twenty in
each county, forty in each prefecture) received stipends. During the
course of the fifteenth century, the student body quota was repeat-
edly expanded, eventually to permit the registration of all qualified
aspirants, but without stipends. Nationwide, the local student body
grew from some fifty thousand in 1500 to some five hundred thou-
sand in 1700. The licentiates often led culturally and intellectually
rich lives in their home localities. The Ming and Qing capital cities
(Nanjing, then Beijing) exerted much less attraction than Changan
in early Tang or Kaifeng in the Northern Song. And by the late Ming,
lowly licentiate status was no longer a bar to literary fame: Chen Jiru
(15581639), Pu Songling (16401715), and Wu Jingzi (17011754)
are but three outstanding examples. The vast majority of licentiates,
100 Permanent Unity Largely Achieved

however, had to turn to humbler sources of income, such as teaching,

estate management, tax engrossment, or litigation.
The goal of every licentiate was to clear the next hurdle, the award
of a provincial degree. Every three years, several thousand licenti-
ates were certified to gather in their provincial capital for a competi-
tive test that took three days and two nights. Here, a strict quota of
around a hundred passers ensured a failure rate of about 95 per-
cent. Licentiates might try these exams many times. The chance was
extremely small that any one licentiate would ever achieve the degree,
but winning it meant a great deal: permanent elite status, eligibility
for official position, and important tax and legal exemptions.
That was not the top of the ladder. In the year following each pro-
vincial examination, all of those who had passed gathered in Beijing
for the metropolitan round. At this level the failure rate was still
upwards of 90 percent, with some four thousand provincial degree-
winners competing for three hundred available metropolitan degrees.
(The competition was not totally meritocratic; in order to ensure
national integration, the quota system was configured in such a way
as to provide opportunities for less well-prepared candidates from the
north China and border provinces, which lagged behind the south
economically and culturally.) Nor was this the top. There immediately
followed a palace examination nominally conducted by the emperor
himself, according to the results of which a final ranking of 1 to 300
was established. The top finishers were taken into that great literary
think tank, the Hanlin Academy, a training ground for later appoint-
ment to the best positions in the central bureaucracy. Those toward
the bottom of the list typically were sent out to the counties of the
realm to begin their careers as local magistrates. (The appointments
system placed these men anywhere but in their native provinces and
usually put new men in small and quiet counties, not in busy and
difficult ones.) The average age of metropolitan degree-winners was
about thirty-one.
However, it is a fact that the majority of the positions in this rather
exclusive mandarinate consisted of very low posts such as vice and
assistant magistrates, fishery monopoly directors, instructors in the
government school system, and the like; typically it was not pro-
vincial or metropolitan degree-holders who filled these posts, but
rather men promoted up from the clerical service, or degree-seeking
 Social Institutions in Political Context 101

licentiates who repeatedly failed, were then sent to (or purchased

status in) the national Confucian college, and, after a very long wait,
were finally offered appointments. These men were at times forbid-
den promotion to the elite ranks. Bureaucracy thus featured a kind of
internal caste system as an outgrowth of its meritocratic principles.
The higher officials of the Ming and Qing bureaucracy usually had
no direct contact with the public but spent much of their time study-
ing, supervising, criticizing, and periodically rating the work of other
officials lower in the hierarchy. It was county magistrates, and officials
below that rank, who regularly met the public in connection with such
functions as tax assessment and collection, disaster relief, bandit sup-
pression, litigation, and education. The system was laced with fines
and other penalties for administrative misdemeanors and mistakes,
so much so that by the middle of the eighteenth century office hold-
ing tended to become such an unpleasant experience as to prompt
early resignations and truncated careers. After the completion of Qing
empire building in the 1760s, it tended more and more to be the case
that high-ranking civil officials, Chinese graduates of the examination
system, were given the personally risky and less-than-glamorous tasks
of organizing and commanding the military suppression of rebellions
(such as the White Lotus rebellion of 17961805) and troubles along
the frontiers in the southwest, and in Sichuan, Xinjiang, and Tibet.

Much of history entails paradox. Certainly, late imperial China is a

case in point, an overstretched Manchu-run regime, with too small
a government, army, and tax base for its huge territory and society,
strong enough to do impressive things and survive extraordinary
shocks yet also observably shaky and vulnerable even in the best
of times. Some European visitors noted this. In the 1550s, Galeote
Pereira thought Ming China conquerable by a good European army.
European onlookers were puzzled as to how a small and cultur-
ally primitive folk like the Manchus were able to seize control of
so large and sophisticated a place as late Ming China. In the early
1700s, when the Kangxi emperor was still ruling, John Bell thought
China could be conquered by Russia. In the 1740s, English sea
captain Anson was struck by Chinas palpable military weakness.
Macartney, leader of the famous but failed 1793 British embassy to
China, made several remarkable comments about the institutional
102 Permanent Unity Largely Achieved

complexion of the Qing empire. He marveled that, unlike the case

of the Hanoverian dynasty that ruled Britain and became British, the
few Manchus that ruled three hundred million Chinese, from the
Qianlong emperor on down to the lowliest soldiers, managed to pre-
serve an ethnic distinctiveness. He was amazed that a dynasty and a
government, unloved by the people it ruled, could yet perform such
an intricate and demanding task as to keep the huge realm under
some semblance of oversight and control. But, he stated, I often
perceived the ground to be hollow under a vast superstructure.
He thought the empire might well collapse in his own lifetime.
Frequent insurrections in distant provinces were clear warning
signs of that. Macartney was wrong only by a century. He died in
1806. The Qing ended in 1912.

. J. L. Cranmer-Byng, ed. An Embassy to China: Being the Journal Kept by

Lord Macartney during His Embassy to the Emperor Chien-lung 17931794
(London: Longmans, Green and Co. Ltd., 1962), 239.
. Ibid., 238.
 Further Reading

A proper bibliography of books, articles, and chapter-length stud-

ies in English would turn out to be impossibly long, and longer
by a factor of three or four if one were to include all the work avail-
able in modern Chinese, Japanese, and the principal languages of
Europe. As a search of the Internet will turn up several book lists,
maps, and timelines prepared by teachers of Chinese history, let me
just comment here on a limited number of contributions, some old
but most new, that have struck me as interesting, well written, not
too specialized or theorized, and reliably grounded in the original
sources. Inevitably I will sin by omission, and I must apologize to
the many authors to whose work I am indebted but cannot begin to
list here.
For the whole period from 150 to 1850, the multivolume
Cambridge History of China, originally edited by Denis Twitchett
and John K. Fairbank, is slowly nearing completion, and contains
many very readable chapters on dynasties, rulers, and events. F. W.
Motes Imperial China, 9001800 (Cambridge: Harvard University
Press, 1999) is, despite its great bulk of over a thousand pages, a
comprehensive account that keeps the general reader in mind.
Interesting and thought provoking is S. A. M. Adsheads China in
World History (3rd ed., New York: St. Martins Press, 2000). Almost
forty years after publication, the same can still be said of Mark Elvin,
The Pattern of the Chinese Past (Stanford: Stanford University Press,
1973). Charles Horner attempts to link present-day China back into
its Yuan, Ming, and Qing past in his interesting Rising China and
Its Postmodern Fate: Memories of Empire in a New Global Context
(Athens: University of Georgia Press, 2009). Unexpectedly readable
is Endymion Wilkinson, Chinese History: A Manual, Revised and
Enlarged (Harvard University Press, 2000).
Most of the available publications dealing with the years from 150
to 589, an age of flux and disunion, focus either on religion or on lit-
erature. However, Albert E. Diens Six Dynasties Civilization (New
Haven: Yale University Press, 2007) gives prominence to archeol-
ogy and material culture. The work of Arthur F. Wright, especially

103
104 Further Reading

Buddhism in Chinese History (Stanford University Press, 1959), and

The Sui Dynasty (New York: Alfred A. Knopf, 1978), is first rate. W.
J. F. Jenners Memories of Loyang: Yang Hsan-chih and the Lost
Capital, 493534 (Oxford: Clarendon Press, 1981) provides excel-
lent perspective and detail on north China in the turbulent fourth
through sixth centuries.
No survey of the Tang can fail to mention the absorbing works
of Edward H. Schafer: The Golden Peaches of Samarkand: A Study
of Tang Exotics (Berkeley: University of California Press, 1963),
and The Vermilion Bird: Tang Images of the South (University of
California Press, 1967). David A. Graffs Medieval Chinese Warfare,
200900 (London and New York: Routledge, 2002) presents its sub-
ject with clarity. Tang history finds itself effectively intertwined with
biography in the lives of some leading literary figures, for instance in
James J. Y. Liu, The Poetry of Li Shang-yin: A Ninth-Century Baroque
Chinese Poet (Chicago: University of Chicago Press, 1969); Charles
Hartman, Han Y and the Tang Search for Unity (Princeton:
Princeton University Press, 1986); and Robin D. S. Yates, Washing
Silk: The Life and Selected Poetry of Wei Chuang, 834?910 (Harvard
University Press, 1988).
Despite a growing body of scholarship on the Liao, Song, Jin, and
Xi Xia, most books about this era are specialized, and few pass the
stern tests of significance and readability. James T. C. Lius China
Turning Inward: Intellectual-Political Changes in the Twelfth Century
(Harvard University Press, 1988) comes close. There is also Ronald
C. Egan, Word, Image, and Deed in the Life of Su Shi (Harvard
University Press, 1994).
The leading work on the Mongols in China is Morris Rossabi,
Khubilai Khan: His Life and Times (University of California Press,
1988). The underwater archaeology of Khubilais failed naval attacks
on Japan has been well described by James P. Delgado, Khubilai
Khans Lost Fleet: In Search of a Legendary Armada (University of
California Press, 2003). Recently, Stephen G. Haw has intelligently
addressed a never-ending interest in his Marco Polos China: A
Venetian in the Realm of Khubilai Khan (Routledge, 2006).
Several good books help illuminate the long history of Ming China.
Sarah Schneewind, A Tale of Two Melons: Emperor and Subject in
Ming China (Indianapolis and Cambridge: Hackett Publishing
 Further Reading 105

Company, 2006) deals with the slipperiness of such intimate inter-

changes as took place been the ruling center and people in the local-
ities of the early Ming. The Ming naval voyages to Southeast Asia
and Africa have been elegantly discussed in Louise Levathes, When
China Ruled the Seas: The Treasure Fleet of the Dragon Throne,
14051433 (New York and Oxford: Oxford University Press, 1994).
Also readable, and better from a technical viewpoint, is Edward L.
Dreyer, Zheng He: China and the Oceans in the Early Ming Dynasty
(New York: Pearson Longman, 2006). Well written and still inter-
esting is Ray Huang, 1587: A Year of No SignificanceThe Ming
Dynasty in Decline (Yale University Press, 1981). Not to be missed
is Arthur Waldron, The Great Wall of China: From History to Myth
(New York: Cambridge University Press, 1990).
The seventeenth century and the Ming-Qing transition have
been extensively written about, thanks to copious sources, the pres-
ence in China of Europeans, mainly Jesuits, and a deepening global
dimension to economic and other history. Frederick Wakeman
Jr., The Great Enterprise: The Manchu Reconstruction of Order in
Seventeenth-Century China (2 vols., University of California Press,
1985) looks forbidding but is illustrated and quite readable. Lynn A.
Struve, The Southern Ming: 16441662 (Yale University Press, 1984)
is a vivid narrative of its subject. There are several good books on
the Jesuits in China; most recent and interesting is Liam Matthew
Brockey, Journey to the East: The Jesuit Mission to China, 15791724
(Harvard University Press, 2007). Jonathan D. Spence is the talented
author or editor of no fewer than six works on this era, including The
Death of Woman Wang (New York: Viking Press, 1978), featuring
local governance in a poverty-stricken part of seventeenth-century
north China, and Emperor of China: A Self-Portrait of Kang-hsi
(Alfred A. Knopf, 1974).
The high Qingmeaning, for the most part, the Qianlong reign,
17351799has been illuminated by some exceptionally interest-
ing offerings. Let me simply list some in chronological order of their
publication: Harold L. Kahn, Monarchy in the Emperors Eyes: Image
and Reality in the Chien-lung Reign (Harvard University Press,
1971); Philip A. Kuhn, Soulstealers: The Chinese Sorcery Scare of
1768 (Harvard University Press, 1990); Alain Peyrefitte, The Immobile
Empire (Alfred A. Knopf, 1992); James A. Millward, Beyond the Pass:
106 Further Reading

Economy, Ethnicity, and Empire in Qing Central Asia, 17591864

(Stanford University Press, 1998); Jonathan D. Spence, Treason by
the Book (Viking, 2001); and Peter C. Perdue, China Marches West:
The Qing Conquest of Central Eurasia (Harvard University Press,
2005).
The Opium War of 18391842 has traditionally been taken
to be the starting point for the history of modern China, and the
many good books on the topic will not be covered here. Although
the earlier years of the nineteenth century have not been intensively
investigated, there does exist one superb study: Jane Kate Leonard,
Controlling from Afar: The Daoguang Emperors Management of the
Grand Canal Crisis, 18241826 (Ann Arbor: Center for Chinese
Studies, University of Michigan, 1996).
 Index

Aguda, 44 Dezong (Tang ruler), 29

Aizong (Jurchen Jin ruler), 49 Di (tribe), 6, 9, 10, 11, 12, 13, 17
Altan Khan, 76, 77 Donglin, 73, 79
An Lushan rebellion, 21, 27, 28, Dorgon, 82
32, 47, 60 Du Guangting, 33
Annam, 33, 43; see also Vietnam Dutch, 77, 78, 85
Arughtai, 7071
Eastern Jin dynasty, 9, 1516, 85
Beijing, 49, 50, 52, 57, 70, 72, 82 Eight Princes, War of the, 78
Bianzhou, 33; see also Kaifeng Erzhu Rong, 15, 17
Buddhism/Buddhist church, 9, 11, Esen, 72
13, 15, 22, 25, 31, 45, 61, 70, 92 eunuchs, 3, 30, 33, 5657, 72, 79
bureaucratic recruitment, xivxv, examination system, see bureau-
2021, 53, 64, 99101 cratic recruitment

Cai Jing, 43 Fan Wencheng, 61

Cao Cao, 4, 6, 14 Fang Guozhen, 55
Cao Pi, 5, 6 Fang La, 44
Chan Buddhism, 31 Fang Xiaoru, 69
Changan, 19, 22, 25, 26, 27, 28, Fei River battle, 12
30, 32, 33 Feng (Northern Wei empress), 13
Chen dynasty, 16, 18 Five Dynasties and Ten
Chen Jiru, 99 Kingdoms, 3335, 3839
Cheng (Cheng-Han) dynasty, 7 Former Liang dynasty, 7
Cheng Yi, 43 Former Qin dynasty, 11
Chongzhen (Ming ruler), 80 Former Yan dynasty, 7, 10
Chouchi, 10 Former Zhao dynasty, 9
Confucianism, xvii, 3031, 45, Fotudeng, 9, 13
48; see also Neo-Confucianism Fu Jian (Former Qin ruler), 10,
Cui Hao, 13 12, 16, 25

Dai (regime), 12 Galdan, 87

Dalai lamas, 80, 83, 87 Gao Huan, 16, 17
Dali, 42, 62 Gao Jiong, 22
Daoguang (Qing ruler), 90 Gao Yang, 17
Daoist church, 13, 22, 25, 49, Gaozong (Tang ruler), 26
61, 92 Gaozong (Southern Song ruler),
daoxue, see Neo-Confucianism 45, 46

107
108 Index

Geng Jingzhong, 84 Kou Qianzhi, 13

Gongsun warlords, 4 Koxinga, see Zheng Chenggong
Grand Canal, 22, 25, 29, 64, 65, Kumarajiva, 12
66, 70, 90
Grand Council, 89 Later Han dynasty, 34, 5, 29
Great Rites Controversy, 74 Later Jin dynasty, 40
Great Shu dynasty, 3435 Later Liang dynasty (386403), 12
Great Wall, 71 Later Liang dynasty (sixth
Gu Yanwu, 59 century), 17
Later Liang dynasty (tenth
Han dynasty, see Later Han century), 67
dynasty Later Qin dynasty, 12
Han Yu, 30 Later Tang dynasty, 40
Heen, 89 Later Yan dynasty, 12
Hong Chengchou, 80, 83 Later Zhao dynasty, 9
Hongxi (Ming ruler), 7172 Later Zhou dynasty, 40
Hou Jing, 17 Li Ao, 30
Hu (Northern Wei empress), 15 Li Gang, 44
Huan Wen, 16 Li Jiqian, 40
Huang Chao rebellion, 3233 Li Keyong, 33
Huang Zongxi, 59 Li Linfu, 28
Hui (Chinese Muslims), 9091 Li Shimin (Tang ruler), 24, 26, 69
Huizong (Northern Song ruler), Li Te (Cheng ruler), 7
43, 44 Li Yuan (Tang ruler), 23, 27
Hung Taiji (Qing ruler), 80 Li Zicheng, 8081
Liao dynasty, 36, 37, 3940, 43,
Japan/Japanese, 75, 76, 79, 85 44, 49
Jesuits, 78, 84, 87, 88 lijia system, 95
Jia Sidao, 50 lineages, Ming and Qing, 9697
Jiajing (Ming ruler), 72, 7376 Liu Bei, 5
Jianwen (Ming ruler), 69 Liu Cong, 8, 9
Jie (tribe), 9, 10, 15, 17 Liu Ji, xvii, 67
Jin (Jurchen) dynasty, 36, 44, Liu Yao, 9
4547, 48, 49, 61, 64 Liu Yu, 16
Jingtai (Ming ruler), 72 Liu Zongyuan, 30
Lizong (Southern Song ruler), 50
Kaifeng, 40, 41, 44, 46, 49, 81 Lu Xiangshan, 93
Kangxi (Qing ruler), 83, 8488 Lu Zhi, 29
Khitans, 47, 64; see also Liao L Guang, 12
dynasty L Kun, xviii
Khubilai (Yuan ruler), 60, 6163 Luoyang, 4, 7, 9, 14, 15, 22, 26,
Koguryo, 23, 27 28, 32, 42, 50
 Index 109

Macartney, Lord, 101102 Qinzong (Northern Song ruler),

Manchus, 5861, 83 44
Meng Zhixiang, 35
Ming dynasty, xviii, 6682 Ran Min, 10
Ming Taizu, see Zhu Yuanzhang Red Cliffs battle, 5
Murong (Xianbi clan), 6, 7, 9, Red Turban rebellion, xvi, 6566,
10, 12 67
Restoration Society, 80
Nanzhao, 28, 29, 31, 32, 33 Ricci, Matteo, 78
Neiwufu, 5758 Russia, 87
New Policies, 42, 43, 48
Neo-Confucianism (daoxue), xvii, semu, 6364
42, 48, 50, 53, 60, 64, 70, 73 Shang Kexi, 84
Nerchinsk treaty, 84, 87 shanyu, 27
non-Chinese rule, xviixviii, 12 Shanyuan treaty, 41
Northern Qi dynasty, 17 Shatuo (tribe), 32, 33, 39
Northern Song dynasty, 35, 37, Shence Army, 30, 32
38, 4045, 47 Shenzong (Northern Song ruler),
Northern Wei dynasty, 1215, 42, 43
19, 24 Shi Hu, 10
Northern Yan dynasty, 13 Shi Jingtang, 39
Northern Zhou dynasty, 17, 18, Shi Lang, 86
19 Shi Le, 8, 9, 10, 13, 25
Nurhaci, 79 Shizong (Jurchen Jin ruler), 47
Shu-Han dynasty, 5
Oyirads, 7071, 72 Shun dynasty, 8182
Shunzhi (Qing ruler), 8283
Pan Jixun, 77 Silla, 33
Pang Xun, 32 Sima family, 68
Parhae, 33 Sima Guang, 43
Peach Blossom Spring, 1 Six Garrisons, Revolt of the, 15,
Pingcheng, 12, 14 27
Portuguese, 75, 76, 78, 79, 85 Song dynasty, 21; see also
Pu Songling, 99 Northern Song dynasty,
Southern Song dynasty
Qi Biaojia, xviii Song Lian, 67
Qi Jiguang, 76, 77 Song Taizu, see Zhao Kuangyin
Qian Liu, 33, 34 Southern Han dynasty, 34
Qiang (tribe), 6, 9, 12, 13, 17, 36 Southern Liang dynasty, 16, 17
Qianlong (Qing ruler), 88, 8991 Southern Ming (era), 8486
Qiao Zhou, 5 Southern Song dynasty, 4551, 85
Qing dynasty, 8291 Southern Tang dynasty, 34, 35
110 Index

Spanish, 77, 78, 85 Western Wei dynasty, 17

Su Chuo, 18, 42 Western Yan dynasty, 12
Su Shi, 43 Wu dynasty, 5, 7
Su Wei, 22 Wu (Northern Zhou ruler), 18, 26
Sui dynasty, 18, 1923 Wu (Tang empress), 24, 26
Sun Quan, 5 Wu Jingzi, 99
Suzong (Tang ruler), 28 Wu Sangui, 82, 84
Wu-Yue, 34
Taichang (Ming ruler), 79 Wudi (Southern Liang ruler), 25
Taihe Reforms, 14 Wuzong (Tang ruler), 31
Taiwu (Northern Wei ruler), 13
Taizong (Tang ruler), see Li Xi Xia dynasty, 36, 37, 40, 41, 43,
Shimin 45, 49, 62
Tang dynasty, 2021, 2434 xian (county), xiii
Tangut (tribe, language), 36, 46, Xia Yan, 7576
63 Xianbi (tribe), 12, 13, 14, 17, 26
Tao Qian, 1 Xianzong (Tang ruler), 30
Three Kingdoms (era), 56 Xiao Cha, 17
Tianqi (Ming ruler), 7980 Xiao (Uighur clan), 39
Tibet, 28, 31, 33, 45,70, 87 Xiaowen (Northern Wei ruler), 14
Toghto, 6566 Xie An, 16
Tong Guan, 43, 44 Xiongnu (tribe), 6, 8, 13
Tujue (tribe), 23, 25, 27, 28 Xizong (Tang ruler), 33
Tuoba (Xianbi clan), 6, 9, 12, 15 Xu Heng, 60
Tuyuhun (tribe), 6, 27, 32 Xuande (Ming ruler), 72
Xuanzang, 25
Uighurs, 28, 29, 3132, 33, 36, Xuanzong (Tang ruler), 26, 28
40, 90 Xuanzong (Jurchen Jin ruler), 47

Vietnam, 71; see also Annam Yan Song, 76

Yang Jian (Sui ruler), 18, 19, 22,
Wang Anshi, 42, 47 23
Wang Chonggu, 77 Yang Tinghe, 74
Wang Jian, 33, 34 Yangdi (Sui ruler), 23, 26
Wang Yangming, 73, 75 Yao Chang (Later Qin ruler), 12
Wanli (Ming ruler), 79 Yellow Turban rebellion, ix, xiv,
Wanyan Liang, 46 xx, 4, 65
Wei (Cao-Wei) dynasty, 56 Yel Abaoji (Liao ruler), 39
Wei Zhuang, 33 Yel Bei, 39
Wei Zhongxian, 7980 Yongle (Ming ruler), 6971
Wen Tianxiang, 51, 58, 85 Yongzheng (Qing ruler), 88
Wendi (Sui ruler), see Yang Jian Yu Qian, 72
Western Jin dynasty, 1, 68, 19 Yuan dynasty, 6166
 Index 111

Yue Fei, 46 Zhao Kuangyin (Northern Song

Yuwen (Xianbi clan), 6, 9, 18 ruler), 40
Yuwen Hu, 18 Zheng He, 70
Yuwen Tai, 17 Zheng Chenggong, 8586
Yuwen Rong, 27 Zhengtong (Ming ruler), 72
Zhezong (Northern Song ruler),
Zhang Bin, 9 43, 97
Zhang Dai, xviii Zhu Di, see Yongle
Zhang Daoling, 4 Zhu Wen, 33, 34
Zhang Gui (Former Liang ruler), Zhu Xi, 48, 50, 53, 64, 73, 7475,
7 93
Zhang Juzheng, 73, 7677, 79 Zhu Yuanzhang (Ming ruler),
Zhang Xianzhong, 8081 5456, 5859, 6669
Zhangzong (Jurchen Jin ruler), 47 Zhuge Liang, 5
Zhao Ding, 46 Zunghars, 87, 88, 89
This compact narrative history of government institutions and their
dialectical relation to society makes a perfect introduction to traditional China
for political science, modern history, and comparative politics classes. The
thesis, upheld by both specifics in lively prose and thought-provoking cross-
period comparisons, is that unity, however valorized, always required hard
work: military, political, and cultural creativity amidst ever-changing ethnic,
class, and religious formations. Dardess also washes out old libels on non-Han,
female, and eunuch power holders simply by recounting the facts.
S. Schneewind, University of California, San Diego

John Dardess has provided a concise yet rich description of the evolution of
the governing institutions of imperial China through periods of unity and
disunity from the late Han (150 CE) to the late Qing (1850) dynasties. His
account of change and continuity in governance over a two-thousand-year span
of Chinas history is an excellent and accessible introduction to a remarkably
durable political tradition that resonates even today in the workings of the
modern Chinese state and contemporary state-society relations.
William Joseph, Wellesley College

Includes timelines, maps, suggested further readings, and an index.

John W. Dardess is Professor Emeritus, Department of History, University of

Kansas.

ISBN-13: 978-1-60384-311-9
90000

9 781603 843119
FnL1 00 0000