University of Virginia Library

Prolegomenon

1 Introduction

It has long been held that throughout Chinese history women occupied
a position subordinate to men, inhabiting a sphere of activity that was
limited by ideology and social custom to serving the needs of a
patriarchal, patrilineal, and patrilocal world. Occasionally a woman
might be able to surmount these constraints, but such instances were
considered aberrations. To early Western observers, the position of
women, like other aspects of China's history and culture, seemed to
change little from ancient times to the beginning of the twentieth
century. And what change there appeared to have been was for the
worse, such as the emergence of the practice of footbinding and the
adoption of Neo-Confucianism as orthodoxy from the Song dynasty
(960-1279) onward.

During the past quarter century, as approaches to the history of
women generally have advanced and become more sensitive and as
the handling of the Chinese sources has been refined, our understanding
of the position of women in Chinese history has grown more sophisticated.
Recent research has made it strikingly clear that the picture
is far more complicated and nuanced than would have been expected
just a few decades ago. Even so, little has been uncovered to suggest
that Chinese women were significantly better off than heretofore
thought, and while broad generalizations can no longer be made
without care and qualification, it does still seem appropriate to conclude
that Chinese women have seen their position—as manifested in social
status, legal protection, economic rights, and ideological valorization—
in continuing decline from earliest times to at least the end of the
nineteenth century.

This decline has not been entirely constant, however, and as one
might expect, changes in the position of women have paralleled changes
in other aspects of China's historical development and have experienced
a number of "turning points." The Song has long been considered one
of the most significant of these turning points, though "point" is
perhaps a misnomer for a span of three hundred years.[1] Chinese and
Western writers alike have pointed to Neo-Confucian orthodoxy, the

importance assigned to widow chastity, and footbinding as developments
during the Song that rationalized and enforced the
inferiority of women.[2] Recent research has shown, however, that the
matter is rather more involved. The idea that widows should remain
chaste and not remarry hardly began with the Song, and the new
emphasis given it was actually a post-Song development.[3] Moreover,
developments were not uniformly to the disadvantage of women. For
example, during the Song women enjoyed much greater property rights
than in earlier periods.[4]

In recent years, scholars have posited a second major turning point
in Chinese women's history: the Ming-Qing transition.[5] One writer has
pointed to the rise of a group of male critics who questioned such
repressive practices as widow chastity and suicide, footbinding and
concubinage. Some of these critics grew out of the new school of Han
Learning that challenged the assumptions of Song Neo-Confucianism,
while others were the product of a culture that sprang from increased
urbanization and commercialization and the spread of literacy among
elite women.[6] Other scholars have pointed to a growing women's
literary movement or to intellectual developments that, though they
might embody a reaffirmation of classical ideals and result in a
solidification of the existing gender system, generally represented
beneficial developments for Chinese women.[7] While these studies
underscore the need for a rethinking of received notions of the situation
of women during the Ming-Qing period, there has also been a recent
reminder that we must be careful in viewing discrete phenomena as
representative of broader and deeper developments. Kathryn Bernhardt
has noted that, when placed in a broader context, many of the beneficial
developments noted by others seem less striking and of diminished
import for the later rise of feminism and growing pressure for the
equality of women. Moreover, Bernhardt shows that, when one
considers the matter from the point of view of law, the important
transition in fact occurred earlier, between the Tang-Song and Ming-Qing
periods,[8] and that, rather than being beneficial, the change
actually brought decline in the legal status of women.[9]

The Song and the Ming-Qing transition are, of course, by no means
the only turning points in the history of Chinese women. During the
Qin-Han period as well, developments occurred that fundamentally
altered the direction of the history of Chinese women. As the examples
of later periods have demonstrated, however, these developments are
best understood in the broader context of the social, economic, and
political changes that were occurring at the time. One would naturally
expect that the creation of the centralized Chinese imperial structure
by the Qin [OMITTED] (221-206 B.C.) and Han [OMITTED] (206 B.C.-A.D. 220) emperors

would have enormous implications for social institutions, and this was
certainly the case where women and the family were concerned. One
of the most significant of these implications stemmed from the desire
of the imperial government to deal directly with the heads of families
rather than through a hierarchy of feudal-like subordinates. As Patricia
Ebrey has pointed out, Chinese patriarchy was to a very large extent
the product not of the classical period but of the early imperial state,
and patriarchal institutions and practices were reinforced by the policies
and laws adopted by the Qin and Han.[10]

The appearance of patriarchy was strongly supported by intellectual
developments. A gradual transformation of yin-yang [OMITTED] thought
took place during the Han that saw the nature of the feminine principle
of yin change from being complementary and equivalent to the male
principle of yang to being subordinate and inferior to yang.[11] The Han
also saw the compilation of the first texts explicitly intended to provide
examples of correct behavior for women. The first among the extant
examples of these, compiled during the Former Han, was Liu Xiang's
[OMITTED] (77-6 B.C.) Lie nü zhuan [OMITTED] [Biographies of Women]. Liu
presented examples of feminine virtues that were hardly intended to
promote expansion of the scope of women's activities outside
traditional roles. A section with a similar title and purpose became a
regular feature of the dynastic histories.[12] The second was the Nü jie
[OMITTED] [Precepts for Women], written during the Later Han by Ban Zhao
[OMITTED] (ca. 49-ca. 120).[13] Both works became extraordinarily influential
in later periods and were models for similar works right up to the
modern era.[14]

With the creation of the centralized imperial structure, the role and
function of royal wives necessarily changed as well. Marriages among
royal families of the pre-Qin states were largely between persons of the
same or nearly the same social standing, and they were contracted to
form political alliances between states or between states and the royal
Zhou house.[15] Once China was governed by a single imperial structure
with an emperor at its head, such marriages were no longer needed
or possible. An imperial wife was chosen from among the emperor's
subjects, and although her family might gain enormous influence
through the marriage, they could never be the equals of the imperial
family. At the same time, the exalted position of the emperor made him
more remote from his ministers and officers, presenting opportunities
for imperial wives and their families to exert extraordinary influence
over the government and the country through manipulation of the
mechanisms of imperial rule or even of the emperor himself. The
potential for mischief contained in this new situation and the threat
it posed to the empire became clear very early in the Former Han, when

Empress Lü [OMITTED] was able for a time to seize control of the government.
Coping with this problem would require a redefinition of the role and
function of imperial wives, to give them a place within the imperial
structure where they were clearly subordinate to the emperor and from
which they could not undermine the imperial family. But a new
definition of the role of the imperial wives had implications beyond the
imperial government, because it established a model for the place of
women generally, whether in the household or at court.

Despite the importance of the early imperial period for the history
of Chinese women, there has been relatively little study of women
during this era and even less that places them in the context of wider
social, economic, and political changes. Earlier studies tended to deal
with outstanding individuals, such as the historian and poet Ban
Zhao,[16] or with particular empresses.[17] Others have examined somewhat
broader issues, such as the influence of the imperial wives and
their relatives on court politics.[18] In most cases these studies have
followed the primarily political concerns of the sources on which they
have been based. More recently, however, historians of the Qin-Han
have undertaken new approaches, not simply looking at women as such
but examining the institutions that shape women's lives, such as
marriage and the family. T'ung-tsu Ch'ü's study of Han social structure,
for example, included chapters on marriage and the position of
women.[19] Although Ch'ü's work advanced our understanding of Han
society significantly and made available in translation a considerable
amount of primary source material, it suffered from treating the Qin-Han
period—a span of more than four hundred years—almost as
though it were a homogeneous block of time. One thus misses a sense
of the developments that occurred over the course of this period.
Subsequent writers have continued to focus on particular aspects of
women in early imperial China, gradually building up a body of
analytical literature and increasing our understanding of the subject.[20]
The fruits of their work not surprisingly reinforce the conclusions
reached by their colleagues studying the history of women in later
periods: Whereas the broad, impressionistic view may show women
to have been subject to social, ideological, and economic constraints,
closer examination reveals that the strength and relative importance of
these forces varied in response to social, economic, and political change.
This should caution us against being too quick to assume that we
understand what life might have been like for any woman—from
empress to peasant—in early imperial China.

The purpose of the present work is not to attempt a history of women
in early imperial China. Although we sketch the history of women from
earliest times through the Later Han—apparently in flagrant violation

of our own caveats against superficial studies—our aim is more modest.
Our hope is to contribute to the growing body of literature and
source material that will one day make possible the writing of that
history. Specifically, we have translated the three fascicles of Chen
Shou's Records of the Three States that are devoted to empresses and
consorts, together with the extensive material found in Pei Songzhi's
commentary to those chapters.

Records of the Three States is the history of the three independent
states of Wei [OMITTED] (220-265), Wu [OMITTED] (222-280), and Shu [OMITTED] (or Shu Han
[OMITTED], 221-263), which were established as a result of the dissolution
of the Han [OMITTED] empire (206 B.C.-A.D. 220). The parts translated here
are the "Hou fei zhuan" [OMITTED] [Biographies of Empresses and Consorts]
from the Wei section, the "Er zhu fei zi zhuan" [OMITTED] [Biographies
of the Consorts and Sons of the Two Sovereigns] from the
Shu section, and the "Fei pin zhuan" [OMITTED] [Biographies of Consorts
and Concubines] from the section devoted to Wu.[21] Because these deal
with the various wives and concubines of the successive heads of state
in the three regions, they form a topically coherent group.[22] The
biographies in this group are intrinsically interesting for what they tell
us about the lives of these women and their relatives, for the attitudes
toward women expressed in them, and for the light they shed on
historians' approaches to writing about women. A comparative study
of these sections can also deepen our understanding of the structure
and composition of the text of Records of the Three States as a whole.

The material in this prolegomenon is meant to give our texts and
their content a historical context. We discuss the treatment of women
by Han writers, how their views were influenced by political developments,
and how attitudes toward imperial spouses in particular
and women in general might have changed over time. Although we
believe the reader will find in our translation and discussion much that
is useful for understanding the history of women of this period, we feel
obliged to caution against assuming that these chapters present a
complete picture of the subject. There is much yet to be gleaned from
a careful study not only of Records of the Three States but of other
writings of the period. Still, taken as a whole, the three fascicles
presented here seem to provide a fuller picture of women in general
than do the chapters on empresses and consorts in the other histories
of the Han period. This has much to do with the differing natures of
the courts of the Three States. The chapter on the Wei comes closest
to the norm for a chapter on empresses and consorts, because the Wei
court was in many ways a continuation of the Han court, with all its
attendant titles and ritual. The Wu court, in contrast, to a very large
degree grew out of a series of marriages among regional elites, perhaps

more akin to those of the pre-Qin era than to those of the Han. And
the Shu Han court and the marriage patterns of Liu Bei [OMITTED] (161-223)
and Liu Shan [OMITTED] (207-271) were even further from the imperial
model of the Han. Thus these three chapters show us a variety of types
of women and possible spheres of activity for them. Moreover, the rich
variety of material found in Pei Songzhi's commentary presents us with
additional perspectives on the women, while allowing us to evaluate
Chen's historiography in the context of his time.

Finally, aside from whatever scholarly merit our modest effort might
have, it is our fervent hope that it will provide the nonspecialist (in
Chinese studies and beyond) an opportunity to appreciate better the
richness of this transitional period and the extraordinary people and
writing it produced.

2 Palace Women in the Early Empire

A fair amount has been written on changes in the status of women
in China over the last fifteen hundred years, but those that occurred
in the preceding millennium and a half up through the end of early
imperial China were no less sweeping or significant.[1] Meager literary
and archeological sources strongly suggest the outlines of a profound
transformation beginning at least with the Shang [OMITTED] (ca. 1700-ca.
1028 B.C.) and ending with the early empire. It began with a time
when women—royal wives in particular—occupied a position of complementarity,
if not equality, in governing. Their position was legitimate
and their acts of governance were recognized. By the end of the early
empire (A.D. mid-third century), however, the situation was quite
different, and though imperial wives and other palace women might
be active in affairs of state, their actions were regularly thought to be
inappropriate and ultimately inimical to the well-being of the empire.
A person alive in the Shang probably would not have predicted that
result.

In the first half of the twentieth century, influenced by Marxist ideas
coming from the West, Chinese historians posited the existence of a
very early period of Chinese history characterized by matrilineal
society.[2] After 1949, this view became orthodoxy, and variations of it
are found in China in general histories and on signs for museum exhibits
about prehistoric times. Ideological content aside, historians in China
have been able to present considerable circumstantial evidence to
support their claims. Early writers such as Chen Dongyuan [OMITTED]
pointed to the supposedly immaculate births of such mythological
figures as Fu Xi [OMITTED], Shen Nong [OMITTED], and Zhuan Xu [OMITTED], whose
mothers were touched by supernatural forces and became pregnant:
a footprint into which Fu Xi's mother stepped; a divine dragon (shen
long [OMITTED]) that quickened Shen Nong's mother; and a rainbow that
affected Zhuan Xu's.[3] A similar myth exists for the birth of Hou Ji [OMITTED]
[OMITTED], or Lord Millet, the founding ancestor the Zhou [OMITTED] dynasty (ca.
1040-256 B.C.), whose mother Jiang Yuan [OMITTED] became pregnant after
she trod in a divine footprint.[4] Other evidence adduced for this

interpretation is that the graph for the word for lineage, xing [OMITTED],
comprises the elements for female and birth. Additionally, several
writers have pointed out that many ancient surnames contain the
element for female.[5] Also frequently cited in support of the evidence
of matrilineality are passages from later texts—mostly fourth and third
century B.C.—that in ancient time "people knew their mothers but not
their fathers."[6] Finally, some have found practices and terminology in
the Shang period that they believe to be artifacts of a pre-Shang
matrilineal society.[7] Having established to their own satisfaction the
existence of matrilineal society, some scholars have made the dubious
inference that matriarchy (rule by women) existed in most ancient
China, a conclusion that is not sustained by the evidence.[8]

PRE-IMPERIAL CHINA

Tantalizing pieces of evidence notwithstanding, the existence of a pre-Shang
matriarchal or matrilineal society remains unproven. Moreover,
even if one grants the existence of matrilineal elements, by Shang times
Chinese society was unquestionably patrilineal and patriarchal.[9] Even
so, royal wives and other women could exercise considerable authority
and have high status. Shang kings performed sacrifices to their female
as well as their male ancestors, and the well-being and health of a royal
consort was often the subject of the king's divinations.[10] Shang kings
seem to have practiced monogamy in the beginning but later adopted
polygyny, probably for political reasons and to address growing
concern about the need for heirs who could continue the royal
sacrifices.[11] According to one Chinese scholar, King Wu Ding [OMITTED] (ca.
1200-ca. 1181 B.C.)[12] had at least sixty-four concubines, not all of
whom lived in the palace. Those he did not favor (maintain as sexual
partners) were given a piece of territory, and some of these were ordered
to perform sacrifices or to conduct military expeditions. They traveled
back and forth between the capital and the outlying regions on the
king's business, and they were for all practical purposes trusted officers
of the king.[13] They also supervised ancestral sacrifices and seem to have
performed other duties at court.[14] The performance of such important
functions appears not to have been limited to Wu Ding's wives or just
to the wives of the Shang king. It has been suggested that the wives
of subject rulers may have presented tribute at the Shang court on behalf
of their husbands, or the women presenting the tribute may in fact have
been subject rulers themselves.[15] The overall impression is that royal
wives, and perhaps upper-class women generally, were respected and
held positions of authority, though Shang women typically occupied
a position inferior to men.[16]

Royal wives continued to perform an active role in governing during

the Western Zhou (ca. 1040-771 B.C.). Bronze inscriptions refer to the
activities of queens, who had their own officers and were persons of
status.[17] H. G. Creel noted that one queen, whom he identified as the
consort of King Cheng [OMITTED] (r. ca. 1035-ca. 1006 B.C.), appears in several
bronze inscriptions performing functions that normally would have
been those of a king.[18] For a later period, a poem in the Shi jing [OMITTED]
[OMITTED] [Classic of Poetry] presents a list of seven of the highest officials of
the government and states that "the beautiful wife splendidly side by
side (with the king) has her place."[19] The "Minor Preface" ("Xiao xu"
[OMITTED]) to this poem says it is a criticism of King You [OMITTED] (r. 781-771
B.C.).[20] The beautiful wife has been understood to be the enchanting
Baosi [OMITTED], with whom King You was so infatuated that he bungled
his rule of the kingdom and allowed it to be overrun by the armies
of a non-Chinese people from the North.[21] As we shall see in Chapter
3, this interpretation may be more a reflection of later thinking than
a description of what actually transpired. Creel could be correct in
saying that Baosi's appearance here (if indeed it is she) might simply
acknowledge the important role this wife played in decision-making
and perhaps even in the conduct of government.[22] In any case, the
impression conveyed by the available sources is that at times during
the Western Zhou, if not throughout the period, royal wives could and
did actively participate in government functions, in some cases acting
as a king might and in others perhaps performing duties analogous to
those of a minister. Our understanding of the situation, however, is
decidedly hampered by limited evidence.

With the Eastern Zhou (720-256 B.C.),[23] source materials become
more plentiful, offering us a more complete and more elaborate picture.
The major texts from the period reveal a set of intricate institutions
and practices involving not only the Zhou kings but the rulers of
subordinate states (zhuhou [OMITTED]) and the aristocracy (qing daifu [OMITTED]
[OMITTED]).[24] Their very complexity suggests that the beginnings of these
institutions would surely be found well back in the Western Zhou, if
only our sources were richer. Particularly interesting are the institutions
of marriage as practiced during the Spring and Autumn period. These
constituted arrangements based on sororal polygyny, whereby the Zhou
king married twelve women at one time, the rulers of subordinate states,
nine women, and aristocrats lesser numbers according to their rank.[25]
Although this practice was in part driven by the desire to ensure an
heir, it was also largely impelled by political motives, especially the need
to establish and sustain alliances among states, as is suggested by the
fact that all primary wives of rulers came from other states. In a process
known as ying [OMITTED], one state would send the primary bride, accompanied
by a younger sister and a niece, while two related states would

each send a secondary bride, also accompanied by a younger sister and
niece, making a total of nine. The primary bride became the primary
wife, and the eight other women (or girls) who accompanied her all
became secondary wives. This ensured that even if the primary wife
failed to produce an heir or was for some reason divorced, there would
be an heir from the lineage of the principal bride or from a related
lineage, thereby preserving the affinal relationships established by the
marriage. These wives were also to be agents of their natal states and
were to protect the short-term interests of their lineages while producing
heirs who would ensure long-term amity between their natal states and
those of their husbands.[26] A practice so involved must have been
difficult to sustain, and it eventually ceased, though vestiges existed
during the Han.[27]

The harem of a ruler also included concubines and maids. Wives and
concubines were ranked within the harem, and we know that there
were at least nine ranks. A woman's ranking determined the status of
her children in the succession, and it could change.[28] There were several
sources for concubines, including rulers sending girls from their lineage
or fathers sending their daughters into concubinage. An abbreviated
marriage rite might be performed for some concubines, particularly
those from other ruling lineages, but they were usually treated as private
property. A ruler could elevate a concubine and make her his wife; this
usually happened only if the concubine had become a special favorite
of the ruler or he wanted to make her son his successor. Such actions
were frowned upon and were apparently made a punishable offense
through an interstate convention; there was a recognition that concubines
were a potential source of disruption.[29] The size of the harems
is unclear, but they could sometimes run into the hundreds.[30]

Although women could still be important in the cementing of
alliances among lineages and states, they do not appear to have
exercised the same sort of authority in the Eastern Zhou as they did
in the Shang and the Western Zhou. The Zhou li [OMITTED] [Rites of Zhou]
lists a number of posts that were to be held by women. The Rites
is a relatively late text,[31] and although many of the positions it describes
are attested to in other, earlier works, this does not seem to be
so much the case with the positions held by women.[32] The Rites does
describe the royal wives as being counterparts of the highest-ranking
ministers in the government, but there is no evidence that they or the
wives of the rulers of the subordinate states exercised any authority
outside the confines of the palace. The separation of the court into inner
and outer domains appears to have been well established by Spring and
Autumn times, and the appropriate realm of the activities of the royal
wives was considered to be limited to the inner court.[33] Because the

main political function of the royal wives was to serve as agents for
their natal lineages, conditions were ripe for them to begin engaging
in the sorts of manipulative, inner-court politics on behalf of their
families that was to characterize the early empire. The situation was
exacerbated by the growing popularity of large numbers of concubines
during the Warring States period,[34] which led to the development of
sizable harems that became hotbeds of competition for the ruler's favor
and fertile ground for the sort of plotting that marked the courts of
the early empire.

THE TRANSITION TO EMPIRE

Multiple consorts and large harems may have been a source of
prestige and gratification for late Eastern Zhou rulers, but in the end
they became simply a part of the spoils of conquest amassed by Qin
as it rolled up the empire. Just as he integrated other aspects of China
to build his empire, so the First Emperor consolidated the harems of
the conquered rulers to form a seraglio worthy of the lord of the
subcelestial realm. He built palaces and pavilions in his capital, where
he assembled the women belonging to the rulers of the states he had
eliminated.[35] One text says, "He demarcated within and without one
hundred forty-five halls and lodges, and the diverse women occupying
the rear apartments numbered more than ten thousand. An emanation
rose and surged to Heaven."[36]

Given the systematizing policies instituted by the First Emperor in
the other spheres of the new empire, it is not surprising to find that
he established an elaborate scale of ranks and titles for the women of
the harem that mirrored those of the civil bureaucracy. The system
differed, at least in titles, from that found in the Eastern Zhou. It
comprised eight ranks, and like many other Qin institutions was
adopted by the Han:

In addition, the emperor's mother was called empress dowager (huang
taihou [OMITTED]) and his paternal grandmother, grand empress dowager
(taihuang taihou [OMITTED]).[38] The titles favorite beauty (jieyu
[OMITTED]), graceful lady (xing'e [OMITTED]), elegant lady (ronghua [OMITTED]), and
compliant lady (chongyi [OMITTED]) were added by Emperor Wu [OMITTED] of
Han (r. 140-87 B.C.), and brilliant companion (zhaoyi [OMITTED]) was added

by Emperor Yuan [OMITTED] (r. 49-33 B.C.).[39] As Hans Bielenstein has
pointed out, three additional ranks were created beyond these. No
earlier than the reign of Emperor Yuan, the sequence of the fourteen
ranks was rearranged, and the individual ranks were correlated with
those of the bureaucracy. Now the harem not only possessed official
rank but also enjoyed the income that came with it.[40]

The reign of Emperor Wu is often associated with grandeur and
excess, and women were part of the display. According to the Sanfu
huangtu [OMITTED] [Yellow Chart of the Capital District]:

Emperor Wu's successors sought to outdo, him, and the increasing
extravagance of emperors in building their harems drew criticism. The
Hou Han shu [OMITTED] [Later Han History] reports, "After Emperors
Wu and Yuan, each generation was increasingly profligate and wasteful,
until the palace women numbered three thousand and their official
ranks grew to fourteen."[42] Following the restoration of the Han,
Emperor Guangwu [OMITTED] (r. 25-57) reduced the size of the harem
and the number of ranks. Besides the empress, there were only
honorable lady (guiren [OMITTED]), beautiful lady (meiren), and chosen lady
(cainü [OMITTED]).[43] The honorable ladies had a small fixed income, but the
beautiful ladies and the chosen ladies did not.[44] Subsequent Later Han
rulers did not feel obliged to emulate Emperor Guangwu's restraint,
and Emperor Huan's [OMITTED] (r. 146-168) harem reached some five to
six thousand women, the vast majority being chosen ladies.[45]

What were the origins of the wives and concubines of the Han
emperors? Unlike the pre-Qin period when the families of the rulers
of the different states married among themselves, thus practicing a sort
of class endogamy where marriages occurred among equals or near
equals (the Zhou king being a special case), once an imperial structure
was established, the ruler had no equals. Moreover, with the founding
of the Han, the matter became a bit more complicated, for the Han
founder Liu Bang [OMITTED] (d. 195 B.C.) and his followers were of plebeian
origins. Consequently, imperial marriage in the Former Han was
relatively free of the strictures that characterized not only Zhou times
but the Later Han as well. A striking example is the case of Lady Wang
[OMITTED], wife to Emperor Jing [OMITTED] (r. 157-141 B.C.) and mother of
Emperor Wu. She had been previously married to a man of rather
modest background. But her mother ended the marriage when a

fortune-teller predicted fame and fortune for her two daughters, and
she arranged to have Lady Wang taken into the harem, where she bore
three daughters and a son to the heir apparent, the future Emperor Jing.
That son ultimately became Emperor Wu.[46] Over time, however, Han
marriage rules grew increasingly rigid, due both to the systematizing
tendencies that characterize the period and to the growth of powerful
lineages.[47] Although in the early years of the dynasty the consorts and
empresses came mainly from humble origins, most of the women in
the "Annals of Empresses" ("Huanghou ji" [OMITTED]) of the Later Han
History were from great families.[48]

Already early in the dynasty there was sometimes a tension between
the Han sovereigns and their more class-conscious officials over the
choice of an empress. To the dismay of officials and historians alike,
some Former Han empresses apparently attained that exalted station
simply because the emperor was fond of them. Such putatively bad
judgment on the part of an emperor might well draw stertorous
objections from officials and was likely to bring out the strong didactic
element that always has been part of Chinese historiography.[49] The
"Wu xing zhi" [OMITTED] [Treatise on the Five Phases] of the Han History
harshly condemns women of low estate who would be empress.[50]
Among those criticized in the "Treatise" and elsewhere are Emperor
Wu's Empress Wei [OMITTED] (appellative Zifu [OMITTED], d. 91 B.C.), who had been
a singer in the retinue of a princess; the same ruler's Lady Li [OMITTED]
and Favorite Beauty Yin [OMITTED], who had been entertainers; Emperor
Cheng's Empress Zhao [OMITTED], better known as Flying Swallow Zhao
(Zhao Feiyan [OMITTED]) for her skills as a dancer and musician; and
Emperor Cheng's Favorite Beauty Wei [OMITTED], who had simply been
a palace maid.[51]

The base origins of such women bothered officials, especially
Confucianists, whose ideology centered on propriety and etiquette. As
the Han ruling house got farther from its own humble antecedents,
similar origins became less acceptable for imperial wives. Shi ji [OMITTED]
[The Grand Scribe's Records] states that only the daughters of princes
and marquises possessing territory were worthy to wed a ruler.[52] By
the time Wang Mang [OMITTED] (45 B.C.-A.D. 23) began maneuvering in
A.D. 2 to have his adolescent daughter made empress of the equally
young Emperor Ping, the principle that imperial wives were to come
from "good families" (liang jia [OMITTED]) was well established. In a memorial
to the throne, Wang said that the difficulties of the state derived
from the lack of an heir and the improper selection of imperial spouses.
He proposed an examination into the Five Classics to fix the ritual for
marriage and correct the duties of the twelve imperial wives as a means
of expanding the succession. He said that a selection should be made
from descendants of the Zhou kings Wen [OMITTED] and Wu, the Duke of

Zhou [OMITTED] and Confucius, and the hereditary full marquises (lie hou
[OMITTED]).[53]

The Confucianists' desire to have consorts from good families was
used by Wang Mang as a pretext for his own daughter's marriage to
the emperor. During the Later Han, however, "good family" (liang jia)
came to mean something different. Under the Former Han, the term
connoted a family that was pure and blameless—that is, one not
engaged in unacceptable occupations such as trade, medicine, or
manufacturing.[54] Such families did not need to have high social status.[55]
During the Later Han, however, although the term retained some of
its earlier sense, it also now clearly referred to large families of some
standing and reputation—in short, to powerful lineages.[56] These
families developed into a powerful force at court and in the government
during the Later Han, and their aggrandizement became a major
factor in weakening the dynasty.[57] Toward the end of the Later Han,
however, such families were themselves greatly weakened in the political
struggles that attended the fall of the Han. Still, their own role in
undermining the dynasty would become a warning to subsequent
rulers, and in the turmoil of the final years of the Han and during the
Three States period, Confucianist concerns with "good families"
would cease to be quite so important in the selection of imperial
wives.[58]

PALACE WOMEN AND PALACE POLITICS

The creation of the imperial structure brought major changes in the
political roles of palace women. Wives could no longer be drawn from
the ruling families of other Chinese states, nor were imperial wives the
means for establishing political alliances among states, as rulers' wives
had been in the pre-imperial period.[59] This meant that imperial wives
did not have the outside source of support and authority that had been
available to pre-Qin rulers' wives, whose natal families were themselves
ruling lineages. Moreover, the formal political participation that had
been available to royal wives in the Western Zhou and before had long
ceased to exist. All activities of the imperial consorts were to be limited
to the inner court, which meant that the only outlet for the political
ambitions of imperial women was through their ability to manipulate
the emperor. Further, because the ruler had now been elevated to an
exalted position over all the empire, he became remote from his
ministers, and the sort of collaborative relationship that had existed
between such men as Guan Zhong [OMITTED] and Duke Huan of Qi [OMITTED]
[OMITTED] ceased to exist. Under such conditions, empresses, empresses dowager,
and concubines became an important means through which
ambitious officials sought to influence and control the emperor.

Although such influence was not always bad, in most cases it worked
to the detriment of the imperial institution, and it was generally railed
against both by honest officials and by those who did not have access
to such influence themselves.[60]

Involvement in state affairs by palace women during the Han
established general patterns for the entire subsequent history of imperial
China and was generally of three kinds. First was the empress who used
her position to seize power in her own right. This was the case with
the first Han empress, Empress Lü. As the wife of Liu Bang, the
founding emperor of the Han, she shared his humble background, and
according to Sima Qian [OMITTED] (145-ca. 86 B.C.), she had "aided him
in pacifying the empire" and was hard and ruthless.[61] Moreover, the
position of the emperor still very much relied on personal abilities and
alliances and was not yet buttressed by the ideology of an imperial
sovereignty that could be violated only with strong justification.[62] Upon
Liu Bang's death in 195 B.C., Empress Lü's son inherited the throne.
Known to history as Emperor Hui [OMITTED] (r. 195-188),[63] this hapless lad
seems to have been unwilling or unable to cope with his domineering
and malevolent mother, who actually ruled during his reign. Upon his
death she placed a succession of two infants on the throne but was
so effectively in control that Sima Qian entitled his chapter covering
the period "Basic Annals of Empress Dowager Lü" ("Lü taihou ben
ji" [OMITTED]). She appointed members of her family to positions of
high authority. Four were named kings, thereby violating an oath taken
by Liu Bang and his followers that only members of the Liu family
could be kings. Others of her kinsmen were made marquises and
generals. Approaching death in 180 B.C., she composed a valedictory
proclamation naming two of her relatives to the most senior positions
in the government, chancellor (xiangguo [OMITTED]) and general of the army
(shang jiangjun [OMITTED]).

The Lü family saw an opportunity to supplant the Liu and seize the
empire for themselves. They were thwarted, however, by kings from
the Liu family and officials who remained loyal to them.[64] Although
Empress Lü failed in her bid to establish her own family, she did leave
a legacy of usurpation of authority by empresses and affinal relatives
that was to bedevil China into the present century. Her case also served
as an object lesson to those later rulers who were willing to heed it.
One who did was Emperor Wu. From his deathbed he ordered the
death of Lady Zhao [OMITTED], mother to the infant heir apparent Fuling
[OMITTED]. When asked why he had the mother killed when he had
established the son, he replied,

The second pattern of interference in the affairs of state by imperial
wives was that in which powerful male relatives used them to exercise
influence or control over the emperor. The Former Han witnessed the
rise of powerful regional families, which was fostered by the
development of the private ownership of land. As these families became
prominent in the bureaucracy and politically active on a national scale,
they maneuvered to have their daughters become the consorts of
emperors in order to improve the position of the family itself or to
strengthen the hand of whatever political faction family members might
represent. As we have seen, representatives of these families sought to
solidify their position and that of their class generally by redefining the
criteria for "good families" so that the term came to encompass only
the powerful. A consort from one of these families was no longer simply
an agent of her family but a pawn whose function was to ensure the
position of her natal family by producing an heir, providing access to
the emperor, and becoming the means for enunciating policy or even
dethroning the emperor once she had become empress dowager.[66]

One of the most important early examples of the manipulation of
an empress to achieve political goals was orchestrated by the powerful
Former Han minister Huo Guang [OMITTED] (d. 68 B.C.). Huo was the
younger half brother of the famous general Huo Qubing [OMITTED], who
brought him to court.[67] He gained the trust and confidence of Emperor
Wu, who promoted him to positions of increasing responsibility. On
the eve of his death, Emperor Wu named Huo one of the three regents
for his successor, the eight-year-old future Emperor Zhao [OMITTED] (r. 8774
B.C.). Huo Guang's granddaughter became consort and then
empress to Emperor Zhao. Following the death of Emperor Zhao in
74 B.C. at the age of fifteen, Liu He [OMITTED], king of Changyi [OMITTED], was
chosen to succeed to the throne. His comportment while he was in
mourning for Emperor Zhao proved so outrageous that Huo Guang
decided he must go. Huo convened a group of ranking officials to
discuss the situation and propose dethronement.[68] After strong initial
reluctance, thirty-six of them were persuaded to sign a memorial containing
a bill of particulars that was then read out to Liu He in the
presence of the fifteen-year-old empress dowager. The empress dowager
was of course Huo's granddaughter, and she was certainly primed on
what was expected of her. She expressed extreme outrage and approved
the measures outlined in the memorial deposing Liu He. Huo was then
free to propose another successor to Emperor Zhao. This time it was

eighteen-year-old Liu Bingyi [OMITTED], who succeeded as Emperor Xuan
[OMITTED] (r. 74-79 B.C.), assuring Huo Guang's dominance.

Huo's action was to reverberate down through the centuries, for he
had created a legitimizing precedent for empresses and empresses
dowager to assume the power of decree. He had thus provided to these
women—and those who controlled them—the means with which to
usurp the emperor's authority and, while perhaps acting ostensibly in
his name, to achieve their own political aims.[69] More specifically, Huo
had provided the model for dethroning an emperor using the authority
of the empress dowager, historical precedent, and the imperial cult. The
case was cited specifically in later dethronements, and it would provide
the model for the dethronement of Cao Fang [OMITTED] (r. 239-254) in
254.[70]

Huo also provided a model for the usurper Wang Mang, who went
beyond him and replaced the Han with his own Xin [OMITTED] dynasty (9-23)
following the death of the juvenile Emperor Ping.[71] Wang was the
nephew of Wang Zhengjun [OMITTED], empress to Emperor Yuan. When
her son acceded to the throne as Emperor Cheng at the age of eighteen,
she named her eldest brother, Wang Feng [OMITTED], regent. Emperor Cheng
was little interested in governing and content to leave affairs of state
to his uncle. Wang Feng died in 22 B.C. and was succeeded by a series
of cousins and brothers until 8 B.C., when Wang Mang, then in his
mid-thirties, became regent. The following year, however, Emperor
Cheng died and was succeeded by his nephew, who became Emperor
Ai [OMITTED] (r. 7-1 B.C.). This emperor was somewhat more interested in
his vocation, and the Wang clan found themselves challenged by the
Ding [OMITTED] clan of Emperor Ai's mother and the Fu [OMITTED] clan of his grandmother.
Wang Mang was forced to withdraw from government, though
Wang Zhengjun remained, since by tradition she was considered the
emperor's adoptive grandmother. When Emperor Ai died in 1 B.C.,
Wang Mang, who had widespread support in the capital, was able to
return to power. Emperor Ai's mother and grandmother had died, and
the emperor himself had succumbed without issue. This left the Grand
Empress Dowager Wang as head of the imperial clan, making it possible
for Wang Mang to engineer the selection of an infant descendant of
Emperor Yuan as successor. This was Emperor Ping, during whose reign
Wang controlled the government. He quickly exacted revenge on the
Fus and the Dings, ordering that the corpses of the Grand Empress
Dowager Fu and Empress Dowager Ding be exhumed, stripped of their
seals, and reburied in wooden coffins as befitted the concubines they
had once been. Empress Dowager Zhao, who had been wife to Emperor
Cheng, was degraded and driven from the imperial palace, as was Ai's
Empress Fu.

Ironically, Wang's actions seem to have been motivated not simply
by a desire to exact revenge but also by a clear understanding of the
threat that affinal relatives posed. He would not allow Emperor Ping's
mother Dame Wei [OMITTED] or her relatives to come near the capital. This
act met with disapproval from several quarters, including from Wang's
own son Wang Yu [OMITTED], who tried to arrange for the Weis to come
to court. For this effort, Wang Mang ordered the execution of his son,
along with members of the Wei clan and others. Wang was left securely
in control, a position he further solidified by orchestrating—over the
opposition of his aunt—the marriage of his daughter to the young
emperor, thereby making himself a relative of the emperor. His carefully
laid plans were dealt a blow, however, when the emperor died in A.D.
6 without having sired a son. Had Ping had a son, Wang would have
been extremely well positioned as father-in-law to Emperor Ping and
grandfather to his successor. Since that was not to be, he apparently
saw assuming the imperial throne himself as the only way to ensure
his continued power. He knew well the difficulties an affinal family
faced in carrying its dominance across generations, for had his aunt
not lived as long as she did and been willing to bring him back, he
might well have remained in the wilderness to which the Dings and
Fus had consigned him.[72]

The third pattern of interference with affairs of state occurred when
an emperor became so taken with one of his harem, especially a lowborn
woman, that he not only took no interest in governing but was
led to excesses that undermined the stability and moral authority of
the imperial institution. Such was the case with Emperor Cheng, who
was smitten by Zhao Feiyan, a slave-entertainer in the service of the
imperial princess of Yang'e [OMITTED]. He took Zhao Feiyan (along
with her sister, known to history as Brilliant Companion Zhao [OMITTED]
[OMITTED]) into his harem, where she became his favorite. When Empress Xu
[OMITTED], losing favor and anxious to produce an heir, was accused by Zhao
Feiyan of performing occult rites, the emperor dismissed Xu and
banished the members of her clan from the capital. Although Emperor
Cheng made Zhao Feiyan empress—over the protests of his mother,
who was offended by her humble background—he gradually lost
interest in her, and she was replaced as his favorite by her sister, the
Brilliant Companion. But neither sister was able to conceive a child by
Cheng. Others were, however, and a slave girl and a certain Beautiful
Lady Xu [OMITTED] each bore him a son. Realizing the threat that direct
male descendants posed to the Zhaos, the Brilliant Companion induced
the compliant emperor to kill both infants. As a consequence, when
Emperor Cheng died in 7 B.C., he left no heir, creating a succession
crisis that was resolved by the selection of a half nephew of the

emperor, a descendant of Emperor Yuan's consort of the Fu clan. There
was some suspicion that Emperor Cheng had not died a natural death,
and the Brilliant Companion committed suicide. Her sister, Empress
Zhao, was protected by Emperor Ai's grandmother, the Empress
Dowager Fu, and remained safe until Wang Mang returned to power
several years later.[73]

The patterns of activity and involvement in court politics by palace
women that developed in the Former Han were repeated and refined
during the Later Han (25-220) and, indeed, on into the present century.
During the Later Han, however, their impact was magnified by
institutional changes adopted by Emperor Guangwu. The power of the
outer court was reduced, and within the inner court the influence and
access of powerful maternal relatives and officials were curtailed. They
were replaced by a palace bureaucracy controlled by eunuchs, who thus
became imperial advisers and were able to control the flow of information
to and from the emperor. Consequently, the emperor was
now raised primarily by palace women and eunuchs. These changes
were to contribute significantly to factional struggles among eunuchs,
affinal relatives, and officials and would result in the dynasty's ruin.[74]

What is particularly striking about Later Han imperial marriages is
the continuing role played by a rather limited group of families until
the final years of the dynasty. The origins of this phenomenon are to
be found in the marriage policy adopted by Liu Xiu [OMITTED] during the
struggles that ended with his becoming the founding emperor, Emperor
Guangwu, of the Later Han. The workings of this policy are redolent
of the system of interstate marriages in the Spring and Autumn period
and presaged the marriage policy of the Suns [OMITTED] at the beginning of
the Three States. In order to construct his power base and build support
in the struggle for dominance in the wake of the fall of Wang Mang,
Liu Xiu concluded alliances with powerful clans from his home
commandery of Nanyang [OMITTED], the Northern Plain, and the Northwest.[75]
These clans were to be the dominant source of imperial wives
until the reign of Emperor Ling [OMITTED] (r. 146-168). For example,
Guangwu's first wife, Guo Shengtong [OMITTED], came from a powerful
family on the Northern Plain, and Guangwu married her in order to
gain needed support against a rival in the region.[76] Once he ascended
the throne in A.D. 25, she became his empress. The support of the
Northern families was no longer needed, however, and the Nanyang
faction increasingly dominated his government and began to press for
the empress to be replaced with a consort from Nanyang, ostensibly
on the grounds that Guangwu's eldest son, born of Yin Lihua [OMITTED]
[OMITTED], should replace Empress Guo's son as heir apparent. Bowing to
pressure, Guangwu divorced Empress Guo in A.D. 37 and replaced her

as empress with Yin Lihua, who was from Nanyang and whom
Guangwu had married in A.D. 23, a year earlier than Guo Shengtong.[77]

The Guo family had been linked with the Ma faction led by the
illustrious general Ma Yuan [OMITTED] (d. A.D. 49). The Yin family were
allied with the Northwestern faction, led by Dou Rong [OMITTED], and their
ascendancy meant that of the Dou as well. From that point on through
the reign of Emperor Ling, most imperial wives came from the Dou
and allied Northwestern families (most notably the Liang [OMITTED]) or from
Nanyang families, such as the Yin and the Deng [OMITTED]. The exceptions
were Emperor Ming's [OMITTED] (r. 57-75) Empress Ma [OMITTED] and Emperor
An's [OMITTED] (r. 106-125) Empress Yan [OMITTED], whose family was from
He'nan [OMITTED].[78] The selection of wives—as well as their dismissal—is
usually described by the dynastic histories as based on very personal
considerations, but in fact the process was clearly driven by factional
concerns, as Hans Bielenstein has cogently demonstrated.[79]

The persistence of this small group of families is quite striking. At
least two of the families, the Mas and the Dous, had been active at
the imperial level during the Former Han, and the Liang family was
already quite wealthy during the reign of Emperor Wu. In part this
persistence was the result of the extreme social stratification that had
occurred by the end of the Former Han and that had resulted in imperial
spouses being taken from a limited group of families. Whereas the
Former Han women could provide entrée to court and a way for the
family to rise (the family of Wang Mang is an example), during the
Later Han marrying a daughter to an emperor became the way to
maintain a family's established position of prominence.[80] This meant,
however, that a family's position might hang by a slim thread, and when
that connection was broken, the family would fall. The most salient
example is the Liang family, who first came to prominence when Liang
Tong [OMITTED] assisted Guangwu in conquering the Northwest. In recognition
of his support, Liang Tong was granted a marquisate, and his
son Liang Song [OMITTED] married an imperial princess, one of Guangwu's
daughters.[81] Although the family's fortunes were dealt a temporary
blow when Liang Song was dismissed in A.D. 59 on charges of
corruption, then jailed and ultimately executed, the family had arrived
at the highest reaches of government. The Liang recovered when Liang
Song's niece entered Emperor Zhang's [OMITTED] (r. 75-88) harem and two
years later bore a son who would become Emperor He [OMITTED] (r. 88-106).
The family subsequently provided empresses for Emperor Shun
[OMITTED] (r. 125-144) and Emperor Huan. A scion of the Liang family,
Liang Ji [OMITTED], dominated the government under Emperor Huan, but
after the empress died in 159, Liang Ji lost a crucial means of control
over the emperor and was unable to replace her. His high-handed

manner had won him the enmity of many, including Emperor Huan,
who turned to the eunuchs for support and drove Liang from power.[82]

Other families fared better. When Emperor Guangwu set aside
Empress Guo in favor of Yin Lihua, for example, Guo's sons were made
kings, and Guangwu continued to honor other members of the family.[83]
In this case, the claims of the author of the Later Han History
notwithstanding, the emperor appears to have understood that he was
setting aside his empress simply for reasons of political expedience and
not as the result of some bitter factional struggle or because she no
longer pleased him.[84] The Mas demonstrated how timely and effective
use of imperial marriages might save a family from destruction. In the
wake of the death of Ma Yuan, who at the time of his passing had
been under attack from the Dou faction, his faction fell from power.
Ma Yuan was posthumously demoted from marquis to commoner, and
the family had to plead with the emperor to be allowed to bury Ma
properly in his ancestral plot. Ma Yuan's nephew Ma Yan [OMITTED] was
distressed by the situation; to fend off disaster, he petitioned to have
Ma Yuan's daughters enter the harem of the heir apparent. His plan
worked. The youngest was accepted, and eventually she became
empress to Emperor He, reviving the fortunes of the family.[85]

Emperor Guangwu was very much aware of the threat that affinal
families could pose to the position of the Lius on the throne. After all,
the object lesson of Wang Mang's usurpation was still vivid. Hence
Guangwu was careful to limit the positions held by the Guos and Yins
so that they did not begin to approach those held by the Wang and
Xu families in the later part of the Former Han.[86] His successor,
Emperor Ming, made an effort to uphold the policies and institutions
of his father. He would not allow relatives of his palace women to be
enfeoffed as marquises or to participate in government.[87] But what
neither he nor his father could foresee was that most of the Later Han
emperors would come to the throne at an early age, providing an
opening for empresses dowager and their families.[88] Because empresses
dowager served as regents for minor emperors—even those who were
not their own sons—and could issue decrees in their names, they were
well positioned to exercise extraordinary authority in the interests of
their own families. Indeed, they could even control the succession, as
was done, for example, by Emperor Shun's Empress Liang. When
Emperor Shun died in 144, he was succeeded by a son by one of his
concubines. The son (Emperor Chong) died a mere five months after
ascending the throne. Empress Dowager Liang then consulted with her
brother and chose another child, though adult candidates were
available. This lad (Emperor Zhi) in turn died under suspicious
circumstances a little more than a year later, and the empress dowager

named an adolescent to succeed him and arranged a marriage with her
younger sister. Because they had supplied the empress dowager, the
dominance of the Liang could in this way be ensured.[89]

The Later Han also saw a large growth in the imperial harem.
Although the First Emperor had created an extensive harem and Former
Han emperors had permitted themselves to be distracted by beautiful
courtesans, the size of Former Han harems seems to have been relatively
controlled. As we have seen, when Emperor Guangwu ascended the
throne, he simplified the harem structure by reducing the number of
ranks from fourteen to three (honorable lady, beautiful lady, and chosen
lady). Growth of the harem under Guangwu's successors, however, was
marked, and by 165 Xun Shuang, who was to become a leading
intellectual and political commentator, was criticizing the enormous
expense and size of the harem, which he had heard contained five to
six thousand chosen ladies.[90] Girls and women between thirteen and
twenty years of age (which could mean between eleven and eighteen
in Western reckoning) were examined each autumn in conjunction with
population registration, and those adjudged suitable were recruited for
the imperial harem. They had to be virgins of good families, and they
were inspected as to beauty, complexion, hair, carriage, elegance,
manners, and respectability, and then graded.[91] If this process were
conducted on an annual basis, it is certainly possible that large numbers
of girls were brought into the palace. Moreover, although a large harem
might be considered the result of imperial extravagance, one should
not rule out the likelihood that people put pressure on the recruiters
to take their daughters in the hope that they might gain imperial favor
or at least be in a position to intervene on the family's behalf. Whatever
the size of his harem, Emperor Huan clearly enjoyed his palace women,
if not his empresses. After he sent his second empress to the Drying
Room and death, he devoted his attention to a group of nine women,
including Chosen Lady Tian Sheng [OMITTED], and although he established
a new empress, he had little to do with her.[92] Regardless of the actual
numbers of women, dedicated officials were right to be concerned,
because the growth of the harem signified a decline in the emperor's
engagement in affairs of state.

CONCLUSION

Had one of Wu Ding's wives been transported through time to the
court of Emperor Huan, she would surely have been astounded at the
condition of her Later Han counterparts and wondered at the changes
that had brought them there. The transformation in the situation of
palace women, particularly the consorts of rulers, in the intervening
period must be considered radical, even allowing for concurrent social,

economic, and political changes. These latter, of course, had much to
do with the former.

The most far-reaching change was the relocation of the sphere of
political activity for palace wives from the outer court (or even beyond
the court) to the inner court. As we have seen, this transfer was already
underway by Spring and Autumn times and thus must have begun
much earlier, perhaps by the middle Western Zhou. Assuming that the
condition of palace wives in some way reflected the situation in the
wider society, the implications of this change are very significant.
Certainly this would have been so for the elite classes, who would have
sought to emulate the court. This shift was evident at the courts of the
subordinate states during the Eastern Zhou, and it set the boundaries
of activity for women at the Han court. Combined with the patriarchal
nature of the imperial structure, this development at the center must
have contributed to the general subordination of women.

The development of the inner court and the creation of the imperial
structure completely altered the nature of political activity. Now such
activity was centered on a single male in an unprecedented way. Political
competition focused on this individual, whether it was competition
among the palace women for favor or among court and government
factions for ascendancy. Even in the latter case, the struggle could be
waged through the women, who were the agents—or pawns—of
particular factions. The possibilities for mischief became legion, and the
ramifications of such mischief were potentially fatal to the imperial
house. With comprehension of this reality came a change in the view
of palace women and, ultimately, of women in general.

Women in Early Imperial History and
Thought

Changes in economic, political, and social structures are inevitably
accompanied by changes in thought and ideology. Thus, in studying
the development of the institutions pertaining to palace women of the
early empire, the question is not whether or not these changes were
reflected in Han writings but how were they manifested and what
impact they had. When we examine the writings of Han thinkers,
historians, literati, and social commentators, the one trend that is
immediately apparent is an increasingly unfavorable view of women
in relation to government. Not only does the position of woman relative
to man decline,[1] but women come to be viewed as requiring constraints
on their behavior and activities to prevent them from causing disruption
and from leading men astray. The evolution of these views was not
simply concurrent with but was, rather, directly influenced by the
activities of the palace women.

WOMEN IN PRE-QIN THOUGHT

In keeping with what we have seen of the position of women in Shang
and Western Zhou times, the Chinese cultural canon often conveys the
impression that women were honored in early China. The "Xu gua"
[OMITTED] [Sequence of the Hexagrams] appendix to the Yi jing [OMITTED] [Classic
of Changes] contains the following theorem:

Although this passage does depict women as a link in the chain from
primordial chaos to government and civilization, it says nothing specific

about their actual position in society. Of course upper-class women,
our main concern here, enjoyed upper-class prerogatives, and we have
seen that individual women might even attain great power or influence.
But the point being made in the text cited here—a point repeated in
other works as well—is that the family is the fundamental unit of
Chinese society, and key in the "Xu gua" passage just quoted are the
social and metaphorical connections between marriage (and
procreation) and the existence and form of the traditional Chinese
polity.[3] Amplifications of this connection are numerous, one of the most
famous being the litany of hierarchical relationships found in the
"Zhong yong" [OMITTED] [Doctrine of the Mean]: "There are five universal
relationships in the subcelestial realm. . . . They are called ruler and
subject, father and son, husband and wife, elder brother and younger
brother, and the association between cohorts and friends."[4] A famous
passage from another canonical text, the "Da xue" [OMITTED] [Great
Learning], reads: "The ancients who wished to illuminate their
enlightened virtue in the subcelestial realm first governed well their
states. Wishing to govern well their states, they first regulated their
families."[5]

Such philosophical pronouncements were serious attempts to
articulate the values of early Chinese society. As we shall see, to a large
extent they continued to inform views of women throughout the Han
and the treatment of women by historians such as Chen Shou, who
subscribed to this vision of society and to the belief in the relationship
between a well-run family and a well-run state.[6] But as will become
apparent, there was a difference. In the texts just cited, the relationship
between man and wife is central, and in the "Xu gua" passage it comes
before the relationships of superior and inferior (father/son, ruler/
subject, etc.), suggesting, if not equality, at least complementarity. This
view was to change.

HAN PHILOSOPHERS AND SOCIAL COMMENTATORS

The change in perceptions of women that occurred during the early
imperial period is perhaps most evident in the evolution of the position
of woman in yin-yang thought. In early expressions of the concept,
the various pairs that embodied the yin-yang duality—sun/moon,
man/woman, Son of Heaven/queen—were complementary, and
the distinction of superior/inferior was muted.[7] A cosmogony
in the Huainanzi [OMITTED] seems to give equal weight to yin and
yang:

In the Book of Rites, for example, yin and yang are linked to the
function of the Son of Heaven and to the queen, without implying that
one occupies a place of honor and the other is debased. Indeed, the
text makes it clear that each requires the other in order to fulfill its
proper functions. The Rites describes separate, complementary realms
of responsibility for the Son of Heaven and his consort and likens their
roles to father and mother of the people:

By the middle part of the Former Han, however, a significant shift
from the complementary view of women was already occurring. The
complementarity evident in such explications of yin-yang theory was
now being replaced by a clear sense that things feminine were inferior
to those masculine and that yang no longer need be balanced by yin.
The salient expressions of this interpretation are found in the Chunqiu
fanlu [OMITTED] [Luxuriant Gems of the Spring and Autumn] of Dong
Zhongshu [OMITTED] (179?-104? B.C.).[10] Dong used yin-yang and
Five Phases (wu xing [OMITTED]) concepts to explain hierarchical and
complementary relationships, such as those between ruler and subject,
man and woman. In his conception, however, yang is superior to yin;
the one is noble, the other base, and by extension man is superior to
woman. Moreover, with Dong it is no longer important if yin and yang
are not in balance. Yang, being good, should prevail, though it should
not destroy yin.[11] Toward the end of the Former Han, the revised
concept is unequivocally expressed in the writings of such people as
the influential scholar Liu Xiang [OMITTED] (79-8 B.C.), as in the following
statement from Liu's Shuo yuan [OMITTED] [Garden of Persuasions]:

As we shall see, similar views were to inform Liu's other writings and
the counsel he gave his emperor.

In the hands of men like Liu, such notions could be powerfully
effective when interpreting the effects that the actions of palace women
might have on the well-being of the empire and the health of the
imperial government. When Liu Xiang and Gu Yong [OMITTED] sought to
have the unfortunate Empress Xu—who had failed to bear an heir—
set aside in 17 B.C., they couched their arguments in terms of the need
to redress an excess of yin. This excess, according to them, was manifest
in a series of events dating from the beginning of the reign, when a
comet had appeared in the lunar mansion House Builder (yingshi [OMITTED]
[OMITTED]) in the first month of spring and the ancestral temple of Emperor
Xuan's father had caught fire.[13] Since then, they asserted, there had been

a series of solar eclipses and natural disasters—floods in particular—
evidence of a serious imbalance between yin and yang. The "Basic
Annals of Emperor Cheng" ("Cheng di ben ji" [OMITTED]) does indeed
record a string of such events,[14] and as they were occurring, the emperor
and some of his officials recognized that there was a serious problem.
In the fourth month of Heping [OMITTED] 1 (28 B.C.), following an eclipse
that had closely followed a major flood the previous month, the
emperor issued a decree criticizing himself for having failed to uphold
the work of his predecessors and quoting the Gongyang zhuan [OMITTED]
[OMITTED] [Gongyang Commentary], "When the male teachings are not
cultivated, affairs pertaining to yang will not succeed, and the sun will
be eclipsed because of it."[15] Five years later, in Yangshuo 2 (23 B.C.),
another decree was issued underscoring the fundamental importance
of working in accordance with the operations of yin and yang and
criticizing officials who did not believe this and consequently failed to
conduct the affairs of Government in accordance with them.[16] Three
years after that, Emperor Cheng issued another mea culpa, declaring
that yin and yang had fallen into disarray because he had failed in his
duties.[17] Thus when Liu and Gu—ostensibly motivated by concern over
the lack of an heir—lay the blame for the excess of yin at the feet of
Empress Xu, the emperor was receptive, and he set her aside.[18]

Throughout the Later Han, yin-yang cosmology continued to
provide a theoretical basis for attacking the influence of women, often
as a means of criticizing the emperor and his appetites. For example,
in his critique of government submitted to the throne in 167, Xun
Shuang railed against the extravagant numbers of women in the rear
apartments of Emperor Huan's court. In part he was distressed by the
enormous expenditure these women entailed, but he also asserted that
their influence produced an inversion of yin and yang that resulted in
natural disasters.[19]

Yin-yang was not the only cosmological principle used to assess the
ramifications of the activities of palace women. Equally important was
the theory of the Five Phases. Of the five phases of Water, Fire, Wood,
Metal, and Earth, the second was that associated with things male, and
disruptions in fire were held to result from the improper behavior of
women.[20] This concept lay behind the references to the fires at the
ancestral temples of Emperor Jing and Emperor Xuan's father
mentioned earlier. Whereas yin-yang cosmology was used most often
in reproaching current practice, references to anomalies arising from
disruptions in the Five Phases were often applied retroactively to explain
events in the past that might serve as cautionary precedents for the
present or the future. The most important examples of this are found
in the treatises on the Five Phases by Ban Gu [OMITTED] (A.D. 32-92) and,

to a lesser extent, in those by Sima Biao, found in the standard histories
of the Former and Later Han.[21]

In explicating the Five Phases, Ban followed the tradition of the
Former Han scholar Fu Sheng [OMITTED], who was active at the beginning
of the Former Han; he also frequently quoted the interpretations of
Dong Zhongshu, Liu Xiang, and Liu Xin [OMITTED] (46 B.C.-A.D. 23). In
discussing Fire, Ban quotes Fu on the actions that lead to disruptions
in this phase: "Disregarding the laws/Dismissing meritorious ministers/
Murdering heirs apparent/Making concubines into principal wives/
These cause Fire not to blaze and ascend."[22] Then Ban cites a number
of instances from the distant past and from the beginning of the Han
that demonstrate the disruptive effects the actions of women might have
on Fire. Referring to an incident in the Chunquiu [OMITTED] [Spring and
Autumn Annals] for the fourteenth year of Duke Huan (698 B.C.),
in which an ancestral temple granary was struck by lightning, Ban
cites Liu Xiang, who attributes the occurrence to the duchess's lewd
behavior, which resulted in the duke's assassination four years later.[23]
And concerning a terrible "disaster" that occurred in Qi [OMITTED] in the
twentieth year of Duke Zhuang [OMITTED] (674 B.C.), Ban quotes Liu again,
who says that it happened because of Duke Huan of Qi's excessive
fondness for women and because he repeatedly made a concubine his
principal wife. Ban also cites Dong Zhongshu, who blames the disaster
on the duchess's licentious behavior and on the fact that seven of the
duke's sisters remained unmarried.[24] Regarding the Han, Ban cites two
cases of lightning striking government buildings during the reigns of
Emperor Hui and Empress Lü, both of which were attributed to
Empress Lü's cruelty.[25] Ban Gu lists a total of twenty-two instances of
portents involving lightning, and in the explanations of each one—by
Ban or an authority quoted by him—women or eunuchs play a role.[26]

Sima Biao's treatises on the Five Phases likewise link portents such
as fires, floods, earthquakes, and spontaneous sex change in a rooster
to imperial consorts' exceeding their proper station or to the ascendancy
and improper behavior of affinal families.[27] Although these examples
did not appear in memorials specifically directed at criticizing the
emperor or his wives, they formed part of the general intellectual
context in which the court operated. Moreover, Ban's inclusion of a
treatise on the Five Phases and his highlighting of episodes involving
the disruption of Fire are surely related to events of his own time.

The interpretation of past events in the light of cosmological theory
as a means of addressing current problems was complemented by
interpretation of the classical canon. During the Han, one of the most
important and effective ways to express views on women—empresses
and consorts in particular—was through commentary on classical texts.

Although the views expressed by a commentator might be his own,
they gained weight by appearing to be explications of the concealed
meanings of canonical texts and by referring back to a classical age—
usually the beginning of the Western Zhou—when China was thought
to have been well governed. A striking example of this genre are the
commentaries to the Classic of Poetry. Poems in the "Guo feng" [OMITTED]
[OMITTED] [Airs of the States] and "Xiao ya" [OMITTED] [Lesser Elegantiae] sections
seem to come out of both upper-class and village culture. These sections
include festal poems, epithalamiums, soldier's complaints, poems
complaining of bad government, hunting poems, and love poems. Since
many of these poems did not seem serious enough for a classic
purportedly edited by Confucius himself, an "apologetic exegesis"
developed.[28] During the Han, moralistic and historicist interpretations
began to be assigned to the poems. Among the most important of these
were the Mao commentary and Mao prefaces.[29] That these interpretations
reflect contemporary concerns is clear from the very first ode,
"Guanju" [OMITTED]. Apparently composed to celebrate the marriage of a
man and woman—not necessarily a ruler and his consort—it is
interpreted by Mao's interlineal commentary so as to describe the
appropriate behavior of a consort:

Largely because of the interpretation expressed here, "Guanju" became
a metonym for the model consort. This interpretation was complemented
by negative examples such as Baosi, and this sort of exegesis
was an important contribution to the growing body of literature
describing the proper place and behavior of a consort.[31]

The dangers posed by the growing influence of women at court
became a recurrent refrain in the writings of Later Han political and
social commentators. It was, for example, an important theme in the
Xin lun [OMITTED] [New Treatise] of Huan Tan [OMITTED] (43 B.C.-A.D. 28), who
had lived through the demise of the Former Han, the Wang Mang
interregnum, and the founding of the Later Han by Emperor Guangwu.
Huan wrote his work as a manual on governing, perhaps intended for
Guangwu.[32] His concern over Emperor Ai's infatuation with Dong
Xian's [OMITTED] younger sister and the threat that it posed to Empress Fu
[OMITTED], the young daughter of his friend Fu Yan [OMITTED] (d. after 1 B.C.), may
have influenced his views. Huan counseled the father on how to conduct
himself and guide his daughter in order to avoid her being replaced
by a new favorite. In doing so, he referred to the cautionary example
of Emperor Wu's Empress Chen [OMITTED], whom the emperor had replaced
with a new favorite, Wei Zifu.[33] The case of Wei Zifu bothered Huan,
and he included it—along with examples of Emperor Wen's favoring
Lady Shen [OMITTED] and Gaozu's [OMITTED] (r. 202-195) excessive reliance on
Empress Lü—in the Xin lun as examples of how otherwise intelligent
rulers had allowed their judgment to be clouded by their consorts.[34]
Unfortunately for Huan, he managed to offend the object of his lessons,
and he died en route to exile after having narrowly escaped execution.

As one might expect, criticism of the influence of palace women—
and of the "feminine" influence of eunuchs—was voiced most strongly
during the reign of Emperor Huan and appeared in the discourse on
affairs of the day submitted to the emperor by candidates for office
who had been recommended to the throne.[35] In their remarks
candidates referred to ancient practice as models of the sort of restraint
that an emperor should be exercising. In 165/166, for example, when
Liu Yu [OMITTED] was recommended to the throne by Grand Commandant
(taiwei [OMITTED]) Yang Bing [OMITTED] as capable and good, sincere and upright,
he submitted comments to the throne on current affairs that offered
strong criticism of the influence of eunuchs and then took aim at the
palace women:

The following year, when Xun Shuang was recommended as
extremely filial by Grand Master of Ceremonies (tai chang [OMITTED]) Zhao
Dian [OMITTED], he similarly criticized the influence of the eunuchs and then
addressed the need to impose propriety (li) on the management of
palace women:

In the tense factional atmosphere of Emperor Huan's reign, such
critiques were little heeded and were likely to result in retribution.[39] Xun
Shuang left office and returned home, and he subsequently suffered in
the partisan (danggu [OMITTED]) persecutions.[40]

PALACE WOMEN AND HAN HISTORIOGRAPHY

Of the writings that deal with Han palace women, the most important
are the histories, in particular Sima Qian's The Grand Scribe's
Records, Ban Gu's Han History, and Fan Ye's Later Han History. Their
importance lies in the variety of material they contain and in the efforts
of their compilers to provide a balanced, if not wholly objective,
account of their subject.[41] All three are of the ji zhuan [OMITTED] (annals
and biographies) form that originated with Sima Qian and evolved
during the Han.[42] This format was adopted for all the standard histories
but Records of the Three States.[43] The material contained in them is

essentially of three types: the annals are rather straightforward
chronicles of the affairs of state; the biographies (or monographs—
some of the zhuan discussed foreign states or minority peoples) present
biographical sketches of persons who had achieved prominence (or
notoriety) in any number of ways, including through politics, military
affairs, scholarship, as worthy officials, and so on; and the specialized
sections present treatises on such matters as political economy, sacrifices
and ceremonial, astronomy, administrative geography, and the
bureaucracy. Any of the three types might touch on the activities of
the palace women, and thus are all important to assembling a
multidimensional view of them.

The Grand Scribe's Records differs from subsequent official histories,
which for the most part take it as a model, in that it contains annals
not only for the Han but for previous dynasties as well. Women appear
infrequently in the pre-Qin annals, usually in references to marriages
and births. But there are also a few cautionary cases, warning against
the bad influence that a woman can have on a ruler.[44] In the annals
for pre-imperial Qin, women appear primarily in connection with
marriages between states,[45] and the annals for the First Qin Emperor
rarely mentions women at all.[46] With the beginning of the Han, however,
we begin to find more frequent, albeit still not very informative,
references to palace women, empresses in particular. At the beginning
of the annals for Gaozu, the founder of the Han, the future
Empress Lü figures prominently in predictions of his coming
eminence.[47] Indeed, in one case his future greatness is obliquely
suggested by predictions of her own nobility.[48] Similar predictions of
greatness are characteristic of the annals of founding emperors in later
dynasties as well.

References to palace women in the annals of subsequent histories
are more perfunctory and are limited to births of emperors, marriages,
deaths of palace women, the punishment of palace women involved
in plotting, and so forth—essentially straightforward records of affairs
of state that involved palace women. A notable exception is the annals
in The Grand Scribe's Records and in the Han History for Empress
Lü, the only empress in the Han period to have her own annals.[49] These
describe her actions in some detail and at times read more like
biographies (liezhuan), narrating her most egregious acts of usurpation
and maliciousness. Although Empress Lü was condemned during the
Han and later for her personal actions, both Sima Qian and Ban Gu
do credit her reign with being a period of peace and prosperity.[50]

In the "Basic Annals" of the Han History, mention of palace women
becomes more frequent from the reign of Emperor Cheng to the end
of the dynasty. This is not particularly surprising, since women were

becoming more involved in court politics. Another factor may have
been that the aunt of Ban Biao, who initiated the Han History, was
a favorite beauty (jieyu) in Emperor Cheng's harem, which perhaps
permitted the Bans to know more about what occurred in the rear
palace and to appreciate better the role of palace women in court
politics.[51] Still, the entries are unembellished, even when they record
what we know to have been bitter factional struggles. Particularly
striking examples are references to events related to the machinations
of the Fu, Ding, Zhao, and Wang families during the reigns of Emperors
Cheng, Ai, and Ping discussed in Chapter 2. For the most part, these
references state simply when a particular empress was established or
deposed.[52]

The same can generally be said of the Later Han History, though
because of changed circumstances—in particular, a greater number of
emperors who ascended the throne in their minority—there is naturally
more frequent mention of political activity on the part of palace
women.[53] The number of references to an empress's assuming authority
on behalf of a minor emperor increase markedly, for example.[54] But
again, events are related in a generally straightforward manner, leaving
any judgments to the reader. Even the unhappy developments surrounding
the accession of Emperor Shun—the death of his mother at
the hands of Empress Yan, the murder of his nursemaid and others—
are narrated without comment, though they were unquestionably
despicable acts.[55] When, in this case, Fan Ye does give his comments
at the close of the chapter, they are brief, offering only muted criticism
of Emperor Shun's failure to make better use of the worthy men at his
disposal and of his allowing the Liang family to establish themselves
as the dominant presence at court.[56]

One might expect the biographical chapters to present more
information and a greater understanding of the lives of palace women,
and to a certain degree they do. But even though these chapters are
ostensibly devoted to them, palace women are not usually the central
focus. For instance, the "Hereditary Houses" section from The Grand
Scribe's Records, covering affinal families, deals with palace women in
rather straightforward fashion, chronicling their rise and fall largely
without comment or insight. Occasionally one comes across a
statement that suggests en passant the pressures and expectations that
faced these women and how they responded to them. An example is
Empress Chen's (wife of Emperor Wu) spending 90 million cash on
physicians in a futile attempt to become pregnant.[57] A significant
exception to this characterization is the brief item by Chu Shaosun [OMITTED]
[OMITTED] (ca. 105-ca. 30 B.C.) tacked on the end of the chapter. Chu lived
during the period of Emperors Yuan and Cheng, and it is unclear when

and how his contributions came to be added to Sima Qian's text. In
this instance, at any rate, his style is gossipier and more judgmental
than Sima Qian's work. He praises Emperor Wu, for example, for
having killed the mother of his son in order to prevent another Empress
Lü.[58]

The section on affinal families in the Han History incorporates much
of the material found in The Grand Scribe's Records for the reigns they
both cover, though the opening paragraph of the former provides more
institutional history on the palace women of the Han than does The
Grand Scribe's Records. And like The Grand Scribe's Records, the Han
History focuses more on the male relatives of the palace women than
on the women themselves. For the later period, however, the History
does occasionally open a wider window on the lives and feelings of
the women of the harem. A poignant example is Empress Xu's defense
against the criticisms by Liu Xiang and Gu Yong of her management
of the rear palace. Described as intelligent and educated, the empress
offered a spirited and cogent vindication of her actions.[59] Another
example is Favorite Beauty Ban, who is portrayed as a woman of
literary accomplishment and political acumen.[60] When Emperor Cheng
asked her to join him in his chariot, she reportedly replied, "If you look
at the ancient paintings, worthy and sage rulers all have famous
ministers at their sides; the last rulers of the Three Dynasties had favorite
women. Now if you want me to join you in the imperial chariot,
wouldn't that be in imitation of them?" The text also extensively quotes
the moving rhapsody she evidently composed upon losing the emperor's
favor to Li Ping and the infamous Zhao sisters.[61] This affair forms a
backdrop for demonstrating the inadvisability of raising low-born
palace women to preeminent positions. The historian's encomium that
closes the two fascicles on the affinal families takes as its theme the
uncertainty of wealth and honor gained through having a daughter
achieve imperial favor. It notes that over the course of the Former Han
more than twenty women had benefited their families through being
favored, but of these only four had been able to preserve the position
of the entire family. The historian pointedly says that those families who
had appreciated the long-standing favor they had received and who had
eschewed excess were able to remain intact. For the others, the great
families were annihilated and the lesser ones, banished.[62]

Unique among the extant histories of the Han, the Han History
includes a separate biography for a single empress, Wang Zhengjun,
or Empress Yuan, consort to Emperor Yuan. However, readers hoping
to find at last the life of a consort described in enough detail to give
deeper insight into the lives of palace women are to be disappointed,
for the purpose of this biography is not to describe the life of an

extraordinary woman whose influence extended through the reign of
four emperors, but to hold up an example of how an empress and
affinal relatives ought to behave when blessed with imperial favor and
the position and wealth that flow from it. The historian highlights
Empress Yuan's loyalty to the Han, which led her to oppose even
her nephew Wang Mang's usurpation of the throne. The point is
underscored by describing how her brother (Emperor Cheng's uncle)
developed good relations with the Liu family following his banishment
to Nanyang and how his sons served the future Emperor Guangwu
in his struggle to restore the Han. In the encomium at the end of the
section, Ban Biao praises the empress for having exercised her influence
over four generations of emperors and over the course of sixty years.
He notes that although her male relatives had been appointed marquises
and generals, she was unwilling to hand over the imperial seal to Wang
Mang, and he contrasts her unstinting loyalty to the Han with the
conduct of earlier empresses such as Empress Lü and Huo Guang's
granddaughter, who put their own and their families' interests first.[63]
Thus, in the final analysis, what we have is less the biography of an
empress than a cautionary piece aimed at present and future emperors
and empresses, one whose placement immediately before the two
fascicles describing Wang Mang's rise and fall underscores its message.

Although Fascicles 10A and 10B of the Later Han History are termed
"Annals of Empresses," they are in reality biographical chapters of the
sort found in The Grand Scribe's Records and the Han History. They
differ from the earlier works, however, in that their purpose and the
criteria for inclusion are unambiguously stated.[64] These fascicles were
explicitly written to caution the emperor on the proper roles of palace
women and affinal families. Those who appear in this section had been
empress or the mother of an emperor, and they are presented so as to
throw into relief the threat to imperial rule that a palace woman and
her relatives might pose. The introductory section draws heavily from
the Rites of Zhou and Liu Xiang's Biographies of Women, emphasizing
the correlative positions of the empress and ranking palace women in
relation to the emperor and his highest ministers. It stresses as well the
role of the empress in supporting the emperor and dwells on the wicked
influence concubines can have on a ruler, citing the cases of Duke Huan
of Qi and Duke Xian of Jin [OMITTED]. It also reprises Huan Tan's criticisms
of Gaozu and Emperor Wen for having failed to observe proper
form in their relations with Empress Lü and Lady Shen. Fan Ye further
laments the growing licentiousness at court from the reigns of Emperors
Wu and Yuan onward.[65] Although Fan notes that Emperors Guangwu
and Ming exercised restraint and put into place regulations to limit the
influence of women and affinal families, it is clear that he believes the

situation went downhill from there. He is particularly concerned about
the influence of empresses dowager, noting that in ancient times it was
worthy and loyal ministers who governed during the minority of a ruler.
It was not until the Qin that an empress dowager governed on behalf
of a minor emperor. During the Later Han, there were four emperors
who came from outside the direct line of succession and six empresses
who ruled on behalf of minor emperors. These empresses, Fan asserts,
all relied on their male relatives and sought to enthrone children in order
to prolong their families' influence.[66]

Fan's examples—positive and negative—are drawn from both
emperors and empresses. He describes the steps taken by the first two
emperors of the Later Han to limit the influence of palace women.
Guangwu, for example, reduced the number of ranks of palace women
and greatly diminished the incomes given them, and Emperor Ming
adopted regulations that curtailed the scope of action of palace women
and affinal families.[67] Fan holds up Empress Ma, daughter of the
illustrious general Ma Yuan, as an example of how an empress ought
to behave. Empress Ma rejected and was critical of proposals to bestow
rank and benefices on male members of her family.[68] Liang Na [OMITTED],
who took charge of affairs of state during the reigns of the child
emperors Chong and Zhi, is also praised by Fan for working hard and
trying to rule well. She was undercut, however, by the machinations
of her brother Liang Ji, who was responsible for the death of Emperor
Zhi and several dedicated officials, causing her to be distrusted and lose
place to the eunuchs.[69] One of the more salient counterexamples offered
by Fan is Empress Deng Sui [OMITTED], who controlled the government
during the reigns of the unfortunate infant Emperor Shang [OMITTED] (r. 106)
and Emperor An. Although her biography records her love of learning
and literary merits and gives examples of the high esteem accorded her
for the competent and compassionate way she governed and comported
herself, Fan is critical of her in his discussion at the end of Fascicle 10A.
He faults her for having clung to power, which resulted in a
deterioration in imperial rule once authority had been turned over to
Emperor An. Fan saw her rule as a significant but pernicious turning
point, following which women remained involved in government and
worthy officials were excluded.[70]

The specialized sections (shu, zhi, biao) of the histories are devoted
to particular topics such as administrative geography, the bureaucracy,
political economy, ritual, religious observances, astronomy, the Five
Phases, and so forth. Scattered throughout these sections are important
comments, anecdotes, and descriptions that contribute significantly to
our understanding of early imperial palace women. The views of palace
women and their influence on rulers as interpreted in the treatises on

the Five Phases have already been discussed. The sections on the
bureaucracy found in the histories of the Former and Later Han list
offices responsible for empresses and other palace women.[71] The "Yu
fu zhi" [OMITTED] [Treatise on Carriages and Robes] of the Later Han
History sets forth the requirements for the chariots used by empresses
and empresses dowager, princesses, and certain members of the harem,
as well as the types of clothing and ornaments to be worn by empresses
and other palace women on specific ceremonial occasions.[72] Treatises
on ceremonial and sacrifices provide tantalizing glimpses of aspects of
Han ritual as it pertained to palace women, such as the sacrifices
marking the posthumous elevation to empress of Emperor He's and
Emperor Shun's mothers and of Emperor An's mother and
grandmother that are found in the "Jisi zhi" [OMITTED] [Treatise on
Sacrifices] of the Later Han History.[73] The "Liyi zhi" [OMITTED] [Treatise
on Ceremonial] of the Later Han History outlines the duties of an
empress upon the death of her husband, describing what is expected
of her with respect to both the funeral and the succession, important
events in the lives and functions of Han empresses.[74] Curiously, though,
the "Treatise on Ceremonial" offers no description of the ceremony of
investiture for an empress, despite the growing political importance of
empresses and empresses dowager during the Later Han.[75]

Although Han historians did recognize the achievements of imperial
consorts, as in the cases of Empresses Lü and Deng, they clearly felt
uncomfortable with the idea of women occupying positions of
authority normally held by men. Perhaps of even greater concern to
them, however, was the possibility that a woman could be the means
by which males from outside the imperial clan might usurp the powers
of the emperor. In their eyes, it was acceptable for an imperial consort
to act as regent for a young emperor as long as she worked in the
interests of the imperial clan. Emperor Yuan's consort Empress Wang
(Wang Zhengjun) is the most obvious case in point. What these writers
feared was that an empress or harem favorite would use her position
in the interests of the males of her own lineage and in the process
weaken the ability of the emperor to act on behalf of the Lius, a problem
that would become more acute with time. Writing history as guides
for their own times, the historians presented the lesson with increasing
directness.

TEXTS DEVOTED TO WOMEN

The criticisms of historians, officials, and social commentators aside,
concern with the influence exercised by palace women also resulted in
works intended to define and present examples of the conduct
appropriate to women in general and to imperial consorts in particular.

Unlike the memorialized injunctions of men like Liu Yu and Xun
Shuang, these works were to have enormous and enduring influence.

Of the two such works that are extant,[76] the first is the Biographies
of Women compiled by Liu Xiang toward the end of the Former Han.
The point of view adopted by this work is entirely in line with the ideas
we have already encountered from Liu. Indeed, the text was written
as a part of Liu's efforts to address what he saw as the deleterious
influence of palace women. Throughout his official career, Liu struggled
against the undue influence at court of palace women and affinal
families. We have already seen the steps he took to have Empress Xu
removed, ostensibly because the empress had not born a son, but more
immediately because of the activities of the Xu family.[77] Although Liu
was ultimately able to thwart the Xus by having the empress dethroned,
the problem returned in even more serious form following the accession
of Emperor Cheng. Liu Xiang now became disturbed by the growing
influence and excesses of the emperor's favorites—Zhao Feiyan and her
sister and Favorite Beauty Wei—and their families.[78] His response was
to compile the Biographies of Women. The structure followed that of
Liu's other works compiled for the edification of the emperor—Shuo
yuan and the Xin xu [OMITTED] [Newly Arranged Anecdotes], collections
of anecdotes drawn from various sources and arranged to provide
guidance to officials and rulers.[79] Biographies of Women contains one
hundred twenty-five biographies organized thematically into seven
chapters. The chapter titles convey some idea of the values that Liu
wished to promote: "Mu yi zhuan" [OMITTED] [Motherly Demeanor],
"Xian ming zhuan" [OMITTED] [Worthy and Brilliant], "Ren zhi zhuan"
[OMITTED] [Compassionate and Wise], "Zhen shun zhuan" [OMITTED] [Virtuous
and Compliant], "Jie yi zhuan" [OMITTED] [Chaste and Righteous],
"Bian tong zhuan" [OMITTED] [Persuasive and Penetrating], and "Nie bi
zhuan" [OMITTED] [Favored Concubines]. The first six of these relate brief
stories of women who exemplify desirable virtues. The last tells of
women whose depraved influence over rulers led their states to ruin,
the first three stories being those of Moxi [OMITTED], Daji [OMITTED], and Baosi,
who are held directly responsible for the demise of the Xia, Shang, and
Western Zhou.[80] The lessons to be drawn from these stories could not
have been clearer. A case in point is the story of Daji. Liu Xiang's
treatment of it offers a good case study of how these stories evolved
in response to the perceived threat of the palace women.

Daji was the consort of Zhou, putative evil last ruler of the Shang.
Zhou was said to have been a ruler of extraordinary strength, cruelty,
and debauchery. He was smitten with Daji and did her bidding,
squandering enormous sums to build her a pleasure terrace and
engaging in orgies. The result, according to The Grand Scribe's Records,

was that the populace became disgruntled and the nobility rebellious.
Zhou responded by making punishments more severe, and he
introduced the punishment of the heated beam. According to The
Grand Scribe's Records, a greased bronze beam was placed across a
pit of coals, and the offending person was forced to walk the beam,
with the usual result being that the unfortunate soul fell into the coals.[81]

This version does not directly link Daji with the heated pole. Rather,
the historian includes it as an example of Zhou's cruelty. When this
is compared with the account in Liu Xiang's Biographies of Women,
however, we find in the latter version that Daji laughed when someone
fell into the coals. Moreover, in the infamous episode in which Zhou
has the heart of the loyal minister Bi Gan [OMITTED] cut out, Liu says it is
because Daji claimed she had heard that the heart of a sage has seven
cavities; in the earlier version recounted in The Grand Scribe's Records,
it is Zhou himself who wants to count the cavities in Bi Gan's heart.
Liu Xiang's purpose is to demonstrate how infatuation with a beautiful
and debauched woman can lead to the downfall of the ruling house,
and he has shifted the balance of responsibility slightly but significantly
toward Daji.[82] Following the death of Empress Zhen [OMITTED] of Wei, we
find the balance tipped even farther in a memorial presented to Emperor
Wen [OMITTED] as he prepared to name a new empress. In Huangchu [OMITTED]
[OMITTED] 3 [222/223], Gentleman-of-the-Household (zhonglang [OMITTED]) Zhan
Qian [OMITTED], in an effort to prevent the selection of Lady Guo [OMITTED] as
empress, memorialized on the importance of choosing a worthy and
enlightened spouse and warned of the dangers of appointing one who
was not: "When Jie [OMITTED] fled to Nanchao [OMITTED], the disaster stemmed
from Moxi. Zhou used the punishment of roasting alive to give joy
and delight to Daji."[83] Here, it is not simply that Daji is amused by
the sufferings of those who slip from the pole: Her amusement has
become the motive for creating the punishment in the first place.

From early on, the influence of Biographies of Women was such that
it inspired some authors to write song [OMITTED] (eulogies) on the biographies.
Furthermore, illustrated editions were prepared, and "texts and illustrations
. . . were frequently painted on Chinese ornamental screens and
on the walls of rooms."[84] Some tomb reliefs even bore such decorations.[85]
Cai Yong, perhaps the greatest scholar of his day, is supposed
to have prepared a set of illustrations, and among the surviving works
of the famous poet Cao Zhi are found fragments of eulogies for the
Biographies of Women.[86] Favorite Beauty Ban may have been referring
to illustrations of stories from Biographies of Women in her admonition
of Emperor Cheng. In fact, she mentions using such paintings as her
"mirror" in the rhapsody lamenting her fall from favor.[87]

The second extant Han work devoted to women is Ban Zhao's

Precepts for Women, which differs from Liu's work in that, rather than
offering examples, it discusses and offers prescriptions for the way a
woman ought to comport herself. Ban was the younger sister of the
historian Ban Gu, and after his death she was summoned to court by
Emperor He to complete the Han History.[88] Her erudition led the
emperor to order the empress and all the honorable ladies (guiren) to
treat her as their teacher. She became a confidant of Empress Deng Sui,
who was herself well read and who apparently consulted Ban on
important matters. In the opening section of the Precepts, Ban states
that she is writing the book for her daughters, who are approaching
the age of marriage and whom she has not been able to instruct properly
in their duties as women. The work comprises seven chapters: "Bei ruo"
[OMITTED] [Lowly and Weak], "Fu fu" [OMITTED] [Husband and Wife], "Jing
shen" [OMITTED] [Respect and Discretion], "Fu xing" [OMITTED] [Womanly
Conduct], "Zhuan xin" [OMITTED] [Single-minded Devotion], "Qu cong" [OMITTED]
[OMITTED] [Yield and Obey], and "He shu mei" [OMITTED] [Harmonize with
Younger Brothers and Sisters-in-law]. The underlying theme of the
work is the importance of a woman's holding to her proper place in
the relationship with her husband and his family. Of particular
significance for later periods was her admonition that a woman should
always remain faithful to her husband and that a widow must not
remarry.[89] Although some of Ban Zhao's injunctions may not have
seemed immediately pertinent to the situation of the imperial inner
apartments, others clearly were, such as this passage from "Respect and
Discretion":

The applicability of this passage to an emperor's relationship with his
consorts would have been obvious, and although the work was
ostensibly written for her daughters, Ban surely shared it with the
empress and other palace women.[91] It was certainly circulated, for the
esteemed scholar Ma Rong [OMITTED] (79-166) praised it highly and
required his wife and daughters to study it.[92]

Liu Xiang's and Ban Zhao's works both inspired imitation. Beginning
with Fan Ye's Later Han History, the standard histories contained a
section entitled "Biographies of Women," as did many local histories

and similar works.[93] The famous second-century literatus Cai Yong
seems to have been inspired by Ban's Precepts to write a similar work.[94]
Both the Biographies and the Precepts were to become basic texts for
the education of women of all classes and have remained influential
into modern times.[95]

CONCLUSION

Even a cursory examination of the literature of the early imperial
period is sufficient to demonstrate that palace women in general, and
the emperor's consorts in particular, had become a subject of major
concern to Han officials and scholars. This development is directly
related to the growth of the centralized imperial state. The role of the
ruler's wives had changed. No longer was a wife the agent of another
state and the means of cementing an alliance, for political activity was
no longer focused on the relations among heads of state. Now there
was but one head of state, and political activity centered on him. This
had the effect of heightening the role of palace women as a route to
political power, and it increased the possibility of their becoming the
locus of conspiracies that could threaten the stability of the empire.

The concern of Han officials (at least those not related to someone
in the harem) and political thinkers was how to limit the influence of
palace women and their relatives. To this end they began to diminish
the position of woman. This was managed to a very large and important
extent through revisions in cosmology that justified a rather different
view of woman than had existed in the pre-imperial period. Granted,
woman had never held a place of dominance over man in ancient
China. Yet, as we have seen, a royal consort was considered to occupy
a position that complemented or even mirrored that of the ruler. But
within a century of the founding of the Han, changes were being made
in cosmological thinking that rendered woman subservient to man and
constrained her scope of activity. The first development is evident in
the writings of Dong Zhongshu, for whom yin was no longer the same
complement to yang that it had once been. The application of cosmology
to history justified limiting the activities of palace women by
offering concrete examples of Heaven's displeasure at their having
stepped beyond their appropriate bounds. Finally, the proper role and
behavior for women were defined through example, as in the models
presented in Liu Xiang's Biographies of Women, and by prescription,
as in Ban Zhao's Precepts for Women. Although Ban's work may not
seem to be aimed particularly at palace women, the values expressed
in it were intended for the education of all women.

At the level of the empress dowager—and, presumably, the widowed
matriarchs of large and extended families—a contradiction presented

itself, for though the empress dowager was a woman and thus a
potential threat to the stability and well-being of the polity, she was
also recognized as having a legitimate role of authority both as senior
member of the family in kinship terms and as regent to a minor emperor.
The problem was to ensure that she carried out these responsibilities
in a way that ensured the prosperity and authority of the imperial family
and that she did not exceed her role in a way that would undermine
them. Empresses who recognized these limits and remained within
them—Empress Yuan, for example—enjoyed the approbation of historians;
those who transgressed were criticized.

The ramifications of the Han scholars' contribution should not be
underestimated. The ideas they developed did not disappear with the
Han but established the framework for attitudes and policies toward
women for the rest of the history of imperial China, as the enduring
influence of the concept of model women developed by Liu Xiang and
Ban Zhao amply demonstrates. The Han scholars defined a subservient
role for woman and presented her as a potential source of instability
that had to be kept within bounds. Although the actualities of women's
lives might depart from these ideals—as was the case with widow
chastity, for example—the ideals remained a standard to be applied
whenever this was deemed appropriate.

4 Empresses and Consorts of the Three States

The Three States were established on the ruins of the Later Han empire,
and they inherited many of its imperial institutions and much of its
imperial ideology. To a greater or lesser extent, each of the three states
of Wei, Shu, and Wu sought to recreate the Han imperium. But
although they possessed the Han blueprint, the materials and context
they had to work with were not the same. The empire was no longer
united, and even within the individual states, central government
control at the local level was nominal at best. In the mid-third century,
the political landscape of China in many ways resembled the multistate
system of the Warring States period, when the various heads of state
were on more or less equal footing and the ability of one to prevail
over the others depended much less on any sort of moral authority than
on simple might and on strategic and tactical acumen. Indeed, in East
Asia the Three States connotes a period of struggle for supremacy
among a group of men known for their cunning, courage, and
commitment to comrades—a period not dissimilar to the Warring
States.

MARRIAGE PATTERNS

Although many of the concerns about palace women that had
occupied the Han continued to be important during the Three States,
the changed political context brought differences in their relative
importance and revived some pre-imperial concerns. Most important
among these latter was the use of marriage to cement alliances. There
were no interstate marriages such as those that had characterized the
pre-imperial period, yet in the struggles preceding the creation of the
Three States, short-lived marriage alliances were concluded among the
men whose families would ultimately rule the three states. These
marriages do not seem to have done much to change the balance among
the three states, but one of them did contribute to the lore of the period,
and it also provides insight into the possible roles for women at a time
when norms had been relaxed. In 209, while Liu Bei was still in Jing
province [OMITTED] and more than a decade before he entered Yi province

[OMITTED] and established Shu Han, he married Sun Quan's [OMITTED] (182-252)
younger sister. This marriage probably took place at the initiative of
Sun Quan, and it followed the general pattern of Sun marriage alliances
discussed below. If Sun hoped for much from this misalliance, he was
surely disappointed. The unconventional Lady Sun [OMITTED] was much
younger than Liu Bei and was considered to be as courageous and
capable as her brothers. She went about accompanied by armed female
retainers, and as relations between Liu Bei and Sun Quan soured, she
was seen by none other than Zhuge Liang [OMITTED] (181-234) to be the
potential source of a coup.[1] She left Liu Bei in 211 and returned to
her brother, attempting unsuccessfully in the process to spirit away the
young heir apparent Liu Shan.[2] Significantly, Chen Shou did not give
Lady Sun a biography, though we cannot be certain whether this was
because of her actions or because of some other selection criterion he
applied.[3]

Cao Cao resorted to marriage as a means of neutralizing the growing
power of Sun Ce [OMITTED] (175-200) at a point when Sun had acquired
Yuan Shu's [OMITTED] (d. 199) troops and gained some victories in the South
and Cao was being pressed by Yuan Shao [OMITTED] (d. 202). Cao's niece
was married to Sun Kuang [OMITTED], Sun Ce's younger brother, and Cao's
son Cao Zhang [OMITTED] (d. 223) was married to the daughter of Sun Ce's
cousin Sun Ben [OMITTED].[4] The marriages were undertaken to address
immediate circumstances, however, and had no long-term effect. The
importance of marriage pacts during this period lay not in concluding
alliances between states to ensure a state's security, but in cementing
ties with other families during struggles for supremacy within a region
or state. This was the case with Cao Cao, who recognized the value
of marriage as a tool for forming alliances with formidable families and
often married off his children accordingly.[5] But he and his immediate
successors also recognized the dangers of a woman who had a power
base and loyalties outside the court, and they were not inclined to turn
to powerful families in selecting their own empresses:[6] Cao Cao's
Empress Bian had been an entertainer, perhaps even a courtesan;[7]
though Empress Zhen was from a family of officials, she was married
to the defeated Yuan Xi [OMITTED] at the time Cao Cao appropriated her
for Cao Pi [OMITTED] (187-226); Cao Pi's future Empress Guo was an orphan
of little status when, as heir apparent, he acquired her; Cao Rui's [OMITTED]
[OMITTED] (206-239) Empress Guo [OMITTED] had been sent to the harem because
her home commandery rebelled, and his Empress Mao [OMITTED] was clearly
of plebian origins.[8] The grossly weakened Cao princes toward the end
of the dynasty did marry relatives of the previous empresses: Empress
Zhen [OMITTED] of Cao Fang, the King of Qi [OMITTED], was grandniece of Cao
Pi's Empress Zhen, while Empress Bian [OMITTED] of the King of Chenliu [OMITTED]

[OMITTED] (Cao Huan [OMITTED]; r. 260-265) and Empress Bian [OMITTED] of the Duke
of Gaogui District [OMITTED] (Cao Mao [OMITTED]; 241-260) were, respectively,
grandniece and great-grandniece of Cao Cao's Empress Bian.[9]
Under the Han, this pattern would have led to a dominant position
at court for one or another of the affinal families. At the end of the
Wei, however, real power lay with the Sima [OMITTED] family.

Of Chen Shou's three fascicles on royal women, that for Shu Han
is probably the least satisfying. The picture it affords of the marriages
of Liu Bei and his son Liu Shan is very sketchy. In contrast to the Caos
and the Suns, the Lius had no apparent marriage policy. Liu Bei was
the ruler of Shu for only two years, and unlike the Caos and Suns, he
had no extended family to draw on or to be concerned about in
planning marriages. Rather, with the exceptions noted below, Liu
marriages followed a pattern that might have been expected for a
military leader in Liu Bei's circumstances. He married women from the
families of his supporters, as did Liu Shan, who successively wed two
daughters of Zhang Fei [OMITTED] (d. 221), one of his father's closest
comrades. In contrast to the Suns, the Lius do not seem to have used
marriage to form the sorts of alliances with local elites that would be
important to assuring the position of a ruling group from outside the
region. But not all the wives of the two Lius have biographies, so it
is difficult to know whether or not there was a marriage policy directed
at local elites. Arguments ex silentio are always risky, but the fact that
no such unions are mentioned suggests that the Lius did not follow
a policy of alliance by marriage.

Interestingly, each of Liu Bei's wives who does have an entry in
Fascicle 34 is a secondary wife. Liu had one, perhaps two, principal
wives before he entered Jing province in 201.[10] During the period when
Liu was actively fighting in eastern China prior to 200, he was three
times forced to abandon his family and suffered the humiliation of
having his wife and children captured. The first time was in 196, when
he was defeated by Lü Bu [OMITTED] (d. 198), an ally of Yuan Shu.
Subsequently, Liu and Lü became allies, and Liu's family was returned
to him.[11] In 198, the two men became enemies again, and Lü once more
captured Liu's wife and children. Lü was captured and killed by Cao
Cao, and Liu was reunited with his family.[12] Finally, Liu was forced
to abandon his family yet again when he was implicated in a plot against
Cao, in whose service he then was.[13]

To console Liu Bei following his defeat at the hands of Lü Bu in 196,
his supporter Mi Zhu [OMITTED], head of a locally prominent and powerful
family, gave his younger sister in marriage to Liu.[14] Presumably once
Liu's first wife returned, he kept Lady Mi, though nothing more is heard
of her. Neither do we know anything further of the wife (or wives) who

had been captured and returned by Lü Bu.[15] We do know that one of
Liu Bei's secondary wives from this period, Lady Gan, accompanied
him to Jing province when he sought the protection of Liu Biao [OMITTED]
(142-208), and there she gave birth to Liu Shan. Following Liu's
assumption of the title of emperor in Yi province, Lady Gan, as mother
of the heir apparent, was made empress.

Liu Bei's Empress Wu [OMITTED] was a widow and the sister of an important
officer who first fought against and then on the side of Liu Bei.[16]
Noteworthy here is Liu Bei's concern that he might have been related
to her deceased husband Liu Mao [OMITTED]. Liu Mao was the son of Liu
Yan [OMITTED] (d. 194), the powerful shepherd of Yi province at the end
of the Han. Both Liu Bei and Liu Yan traced their separate ancestries
to Emperor Jing of the Former Han. Liu Bei claimed descent from
Emperor Jing's son Liu Sheng [OMITTED] (d. 112 B.C.), king of Zhongshan
[OMITTED], and Liu Yan from Liu Yu [OMITTED] (d. 128 B.C.), king of Lu [OMITTED]
[OMITTED]. With the judicious use of historical example, however, Fa Zheng
[OMITTED], a close adviser from Yi province, was able to convince Liu that
under the circumstances the marriage was permissable and advisable.
Still, the Eastern Jin [OMITTED] (317-420) historian Xi Zuochi [OMITTED] (d.
384) thought Liu erred in marrying Liu Mao's widow and criticized
him for it.[17]

Liu Bei had at least two other wives who were apparently with him
in Yi province but who do not have biographies in Records of the Three
States. Their existence is evident from the entries on two of his sons—
Liu Li [OMITTED] (d. 244) and Liu Yong [OMITTED]—who are described as being
Liu Shan's half brothers, each with a different mother.[18] Why Chen
omitted these women is not clear. Chen also seems not to have written
about all of Liu Shan's wives, for Pei Songzhi cites Xi Zuochi's Han
Jin chunqiu [OMITTED] [Han-Jin Spring and Autumn Annals] about a
certain Brilliant Companion Li [OMITTED], who committed suicide rather than
submit to the indignity of being handed over to a Wei general in need
of a wife.[19] Moreover, when Liu Shan tried to build up a harem, he
was told that in ancient times the Son of Heaven had only twelve wives
and that he himself already had enough, suggesting that he had at least
a dozen.[20]

Three of Liu Shan's wives are mentioned by Chen. These include the
two daughters of Zhang Fei mentioned earlier. Little is said about them
other than that they were both made empress, the younger upon the
death of the elder.[21] Also mentioned is Honorable Lady Wang [OMITTED], who
was the mother of Liu Shan's heir apparent. All we are told about her
is that she had been an attendant to the elder Zhang sister. The latter
half of the chapter is largely devoted to Liu Shan's heir apparent, his
half brothers, and their descendants.

Of the three ruling houses of the Three States, marriage alliances were
most important to the Suns. Indeed, they were crucial to the Suns'
establishment of their hegemony over Wu. As a consequence, of the
three fascicles translated here, Fascicle 50 contains more detailed
information about marriages than is found in the other two, and its
description of events in general is more vivid and engrossing. Whereas
the Caos already had a power base and were concerned with
maintaining their control over the imperial and governmental structure
of an existing—albeit weakened—state, the Suns needed first to
consolidate a power base, create an administrative structure, and
establish their right to dominance. To solidify their position, it was
necessary for the Suns to overcome three problems: the resistance of
the Mountain Yue [OMITTED], instability within the ruling group, and lack
of experience in governing a state.[22] The marriage policy they pursued
was important in addressing the second and third of these problems.

The preeminent position of the Suns had been gained through
military accomplishment, primarily north of the Yangtze, where Sun
Jian [OMITTED] (155-192) proved himself a talented commander, first against
the Yellow Turbans and then in the power struggles attending the end
of the Han.[23] Although the Suns were from the South, they were not
prominent there. The family claimed descent from Sun Wu [OMITTED], the
putative author of the famous fifth-century B.C. text Sunzi bingfa [OMITTED]
[OMITTED] [Sunzi's Art of War]—not a particularly impressive pedigree in
the context of the times. Before he made his mark in the wars, Sun
Jian held only minor office in his home area, and even there he
distinguished himself largely through arms.[24] Upon his death, his son
Sun Ce inherited command of a portion of his troops and likewise
embarked on a military career.

Given their background and lack of a solid base in the South, the
Suns needed the acceptance and support of prominent southern
families—something not easily gained. In the South, the closed society
of elite families that would become characteristic of Chinese society in
succeeding centuries was beginning to take shape. Several powerful
lineages had come to dominate the region. The most prominent—the
Zhu [OMITTED], Gu [OMITTED], Lu [OMITTED], and Zhang [OMITTED] families—were known as "the
four lineages" (si xing [OMITTED]), testimony to their importance.[25] But there
were other such families as well.[26] Their social position came through
holding office in the civil government for successive generations, and
they were inclined to look down on military men and their deeds.[27]
This attitude was very likely at the root of Sun Jian's rejection by the
relatives of Lady Wu [OMITTED] described at the beginning of Fascicle 50, for
although her family, the Qiantang Wu [OMITTED], were not among the
most prominent lineages, they nonetheless appear to have been a family

of some standing. The southern elite families were well established and
already dominant in the region, so there was no need for them to seek
power through marriage with the Suns. (Significantly, Lady Wu
overcame her relatives' objections not by arguing the benefits of
concluding such a match but by alluding to the potentially dire
consequences of not doing so.)[28] Moreover, their position vis-à-vis any
interloper bent on establishing control over the region was greatly
strengthened by the close ties among the families through
intermarriage.[29] To have attempted to crush the most powerful of these
families would have been difficult and costly.[30] For most of the
families—especially the four great clans—the Suns employed a series
of measures to win their support, or at least their acquiescence. First,
they assured the political and economic advantages of the elite families
by appointing their members to office in the central and local
governments and by permitting them to control significant bodies of
military manpower. Second, they showed these families respect and
trust through special courtesies and appointments. Finally, they
concluded marriage alliances with the leading families, thereby
effectively linking the fortunes of the two sides.[31]

The Wu section of Records of the Three States on empresses and
consorts offers dramatic examples of intermarriage between the Suns
and Wu elite families. Tables 6 and 7 of Appendix I are illustrative,
showing the Suns' links through marriage with the Lu and Zhu
families.[32] And because of the extensive intermarriage among the Wu
elite, a marital bond with one family brought ties with several others.[33]
Even marriage ties to lesser elite families could bring indirect ties to
greater elites. Sun He [OMITTED] married Zhang Cheng's [OMITTED] (178-244)
daughter, whose sister wed Lu Kang [OMITTED] (226-274). Sun Jian's
younger sister married Xu Zhen [OMITTED], and their granddaughter wed
Lu Shang [OMITTED], and then, following Lu's death, Sun Quan.

The Suns did not take their wives exclusively from the Wu regional
elite families. They also married women from local magnate families.
Sun Jian's marriage to Wu Jing's [OMITTED] (d. 203) sister is an example.
Wu Jing contributed substantially to the consolidation of the power
of the Sun family, and Lady Wu made notable contributions as well.
The Quans [OMITTED], also prominent in Qiantang, are another such family
(see Appendix I, Table 8). Marriage connections with the Quans,
however, proved to be a mixed blessing. Quan Rou [OMITTED] was one of
the first local elites to throw in with Sun Ce. His son Quan Cong [OMITTED]
[OMITTED] (d. 249) was among Sun Quan's ablest commanders and married
Sun Quan's daughters, and his nephew Quan Shang [OMITTED] held high
office in the Wu government and had a daughter who married Sun
Liang [OMITTED] (243-260). But Quan Cong's sons went over to the Caos

in 257 and subsequently became Wei commandery administrators and
marquises.[34]

Sun Quan also concluded marriage alliances with Zhou Yu, a shrewd
adviser and ally of Sun Ce and Sun Quan. Zhou was from a family
of prominent officials in Lujiang [OMITTED] and had two sons, as well as
a daughter who married the heir apparent Sun Deng [OMITTED] (d. 241).
Zhou's eldest son Zhou Xun [OMITTED] married Sun Quan's daughter Sun
Luban [OMITTED]. His younger son Zhou Yin [OMITTED] also married a member
of the royal family. Unfortunately, Yin proved to be something of a
wastrel and was a disappointment to Sun Quan, who banished him.[35]

Besides the Wu local and regional elites, the Suns sought wives from
families that were prominent for reasons other than economic and
military might. Sun Quan's Lady Xie [OMITTED] came from a southern family
with a reputation for producing scholars and worthy officials. Her
father had been a gentleman of the masters of writing and a prefect
under the Han, and her younger brother Cheng [OMITTED] became a commandery
administrator and was well known for writing a history of
the Later Han.[36] Associating themselves with the Xies had obvious
legitimating benefits for the Suns, whose reputation was based on
military exploits and who were considered rather unrefined. Lady Xie's
dismay at having Lady Xu [OMITTED] promoted above her may have been in
part because she considered a military family like the Xus to be her
inferiors.[37]

Still another consideration was probably at work when Yuan Shu's
daughter was taken into Sun Quan's harem in 199. This union was
likely motivated less by a desire to form a connection to a nationally
powerful family than by a sense of obligation and loyalty to a powerful
patron and ally. It served to recognize the role Yuan played in the Sun
family's rise to national prominence. Besides, although the Yuan family
for several generations had held some of the highest offices in the Han
empire, by this point it was fragmented and weakened by the struggles
that were hastening the end of the dynasty. Yuan Shu himself, having
failed in an attempt to establish a new dynasty, was dead. There may,
however, have been residual Yuan assets in the form of military forces
and other support to be gained by forming a marriage alliance with
the Yuans.

Sun men also married for beauty, though such unions could result
in problems. Sun Quan was smitten with Lady Pan [OMITTED] and took her
for his seraglio. She gave birth to Sun Liang, but she was considered
a troublemaker and was killed by several courtiers as Sun Quan lay
on his deathbed.[38] While touring military encampments, Sun Quan was
similarly taken with the beauty of the daughter of one of his
cavalrymen, He Sui [OMITTED], and she was summoned to the palace,

became Sun He's concubine, and bore a son. While she seems to have
been a devoted wife, her family took advantage of the chaos toward
the end of Sun Hao's [OMITTED] (r. 264-280) reign to arrogate power to
themselves, undoubtedly contributing to the fall of the Suns.[39] Sun Hao
himself was enthralled by Zhang Bu's [OMITTED] (d. 264) daughter, whom
he killed in a fit of rage. He reportedly was so heartbroken that he took
Zhang's other daughter from her husband and so occupied himself with
her that he ignored affairs of state.[40] This account is in accord with
what we know about Sun Hao's character, but it also fits the "bad last
emperor" topos emphasized by Liu Xiang in his Biographies of
Women.[41]

Not all of the wives whom the Suns chose for their beauty were so
problematic. Lady Bu [OMITTED], a relative of the chancellor Bu Zhi [OMITTED] (d.
247), was said to have been widely held in fond regard, and Sun Quan
had hoped to designate her as his principal wife.[42] We are told that his
officials opposed such a move, however, and favored Lady Xu for the
position. Since Lady Xu seems not to have been particularly congenial,
support for her may have been based on factional considerations. After
Lady Bu's death, Sun similarly hoped to make Lady Yuan [OMITTED], daughter
of Yuan Shu, principal wife. She was very modest and, because she had
no children, adamantly refused Sun's efforts to promote her.[43]

Finally, there were Sun Quan's two wives surnamed Wang [OMITTED], one
the mother of Sun He and the other the mother of Sun Xiu [OMITTED] (235-264).
Nothing is known about either other than their families' homes
of record and the fact that their brothers were appointed marquises
after the ladies' deaths. That Sun Xiu's mother's home of record was
Langye [OMITTED] may mean she was related to the Langye Wangs, who
were to become so prominent under the Southern Dynasties, but there
is no firm evidence for this.

SOCIAL ORIGINS OF CONSORTS AND POLITICAL POWER

It should be clear from the discussion of Sun marriage policies that
the social origins of prospective consorts remained a concern under the
Three States, though the underlying reasons may have been different
from those during the Han. For the Suns the question was not so much
one of finding spouses who were worthy of them as it was one of
establishing their standing in the region and linking up with the
powerful and wealthy. The Caos, in contrast, at times deliberately
selected women of humble origin, much to the exasperation of some
of the men who served them. Cao Pi and Cao Rui in particular were
criticized for their choices of empresses. Zhan Qian inveighed against
Cao Pi's desire to make Honored Concubine Guo empress, and when
Cao Rui failed to name Madam Yu [OMITTED] as his empress, she did not

hesitate to point out the Caos' proclivity for picking inferior women.[44]
Although powerful and prestigious families did sometimes place their
daughters in the Wei rulers' seraglios, social background was no
guarantee against a woman's ultimately being rejected. Madam Yu and
Madam Ren [OMITTED] are cases in point.[45] The extent to which the Lius
might have been concerned about the social origins of their spouses
is unclear. For the most part, Liu Bei's wives seem to have had
respectable backgrounds, though perhaps not always elite ones. Liu
Shan married the daughters of his father's comrade-in-arms Zhang Fei,
whose own origins were not particularly impressive, though his wife
was supposedly descended from one of Han Gaozu's most important
supporters.[46]

Related to the issue of the social origins of spouses was the problem
of their interference in affairs of state. The lesson of the dangers of such
involvement was part of the legacy that the Later Han bequeathed to
the rulers of the Three States. And it is clear that some, at least,
understood the importance of the lesson. We find, for example,
Zhongchang Tong, writing in his Changyan [OMITTED] [Frankly Speaking]
for the edification of Cao Cao, "You should not conclude marriages
with those to whom you entrust governing; those with whom you
conclude marriages you should not entrust with governing."[47] With the
lesson of the Later Han before him and with ample metaphysical and
cultural support at hand, it is not surprising that Cao Pi, the founder
of the Wei, quickly took steps to remove the potentially baleful influence
of the imperial women and their families. On 25 October 222, just
before he named Honored Concubine Guo [OMITTED] empress, he issued
the following edict:

Cao Pi's decree is reminiscent of the attempts of Emperors Guangwu
and Ming to achieve the same result; his effort was no more successful
than theirs had been.

The involvement of empresses dowager in the affairs of state was
another theme that continued into the Three States from the Han. The
Wei witnessed a reprise of the dismissal of Liu He as successor to
Emperor Zhao when Cao Fang was deposed as emperor by Sima Shi

[OMITTED] in 254. Again the young emperor was accused of being
dissolute, though this time not only with his boon companions but with
female relatives of more than one generation as well. A bill of particulars
was drawn up against him and presented by Sima Shi and more than
forty other officials to Empress Dowager Guo. The empress dowager
was then requested, in accordance with the precedent of Huo Guang's
deposal of Liu He, to take back the seal of the emperor and return
him to be king of Qi.[49] On 17 October 254, the empress dowager issued
a decree saying that although the emperor was now of age (he was
twenty-three), he did not attend to the affairs of state but preferred to
indulge in indecent behavior. Claiming that he was unfit to carry on
the imperial line or serve the imperial ancestral temple, she ordered the
matter reported at the ancestral temple and that he abdicate.[50] It is
certain that Empress Dowager Guo did not support this action, but
troops were posted outside her palace to ensure her compliance.[51] This,
of course, marked the difference between Cao Fang's case and the
deposal of Liu He: In the latter instance, the empress dowager
presumably supported the act (which her grandfather, after all, desired),
whereas Empress Dowager Guo was opposed. It is noteworthy that,
powerful as he was, Sima Shi still felt that he must work through the
empress dowager for the act to have legitimacy.

Shu Han does not appear to have had problems with interference
in the affairs of state by palace women, though the eunuch Huang Hao
[OMITTED] did dominate the later years of Liu Shan's court. Indeed, a
beneficial aspect of Huang's ascendency may have been to prevent
affinal families from gaining influence. The modus operandi of the Shu
court resembled that of the Spring and Autumn and Warring States
periods in that the closest confidants of the ruler were his ministers.
This was, of course, an outgrowth of Liu Bei's relations with his
comrades during the wars for supremacy that attended the breakup of
the Han. Because Zhuge Liang was able to maintain his influence as
Liu Shan's closest adviser for a decade following the death of Liu Bei,
and because the younger Liu was married to Zhang Fei's daughters,
the problems with affinal families experienced by the Han did not arise.

The situation in Wu was quite different from both Wei and Shu Han.
Sun Jian's wife Lady Wu and her brother were perhaps the model of
how a ruler's wife and her family should conduct themselves. Her
brother was a close adviser to Sun Jian, and following Jian's death he
supported the young Sun Ce rather than becoming a rival.[52] His aid
was crucial to Sun's ability to establish himself as his father's successor.
Lady Wu likewise made important contributions to her sons' success,
reportedly giving Sun Ce useful counsel that prevented him from
inadvisedly killing someone and creating resentment, and assisting the

young Sun Quan in administering the army and the state. From her
deathbed, she took steps to ensure that Sun Quan would continue to
receive good advice.[53]

Fascicle 50 contains examples of other women whose behavior could
be considered exemplary. One was Sun Quan's consort Lady Yuan,
who, as we have seen, declined the opportunity to become empress.[54]
Lady Bu, another of Sun Quan's consorts, was widely liked for her
generous nature.[55] Another example of model behavior was Sun Jian's
younger sister, whose son Xu Kun fought in Sun Ce's army. (Xu Kun's
daughter would become one of Sun Quan's consorts.) In a crucial battle,
she gave tactical advice to her son, which he passed on to Sun Ce. The
advice was adopted by Sun, and he was victorious.[56] The involvement
of Sun Jian's sister and Sun Quan's mother in military activities suggests
that the scope of activity for women of the Sun family extended beyond
the normal bounds. This conclusion gains further credence from what
is known of Lady Sun, Sun Quan's younger sister who married Liu Bei.
She is described a having an obdurate personality and going about
accompanied by female armed retainers.[57] Clearly these were
remarkable women, and their activities and personalities perhaps say
something about the nature of the Sun family, suggesting why this
family of relatively humble origins was able to gain the cooperation,
if not the respect, of the more venerable lineages of the Jiangnan region.

Unfortunately for the Suns, these women seem to have been exceptions.
A good portion of Fascicle 50 is taken up with machinations
and backstabbing among palace women. A case in point is Lady Pan,
whom Sun Quan married rather late in life for her beauty. She was
insanely jealous and inclined to speak ill of others behind their backs.
When she was exhausted and unwell from attending to the dying Sun
Quan, a group of courtiers strangled her in her sleep.[58] Throughout
the brief history of Wu, palace women engaged in intrigue, and their
families were not above betraying the Suns to save their own skins.
Members of the family of Lady Quan, wife of Sun Liang, were made
marquises and held numerous offices. In what was surely one of the
more bizarre episodes of the period, several of them, fleeing familial
conflicts and suits, crossed the Yangtze and surrendered to Wei just as
another member, Quan Yì [OMITTED], was supposed to be accepting the
surrender of a Wei general who had rebelled and who wanted to cast
his lot with Wu. The group that had surrendered deceived Quan Yi
into believing that Wu was angry with him and planning to execute
his family. Rather than accepting the surrender of a Wei general on
behalf of Wu, Quan Yi himself went over to Wei. Ultimately, the family
went into decline when a member was discovered hatching a plot, the
purpose of which is not known.[59]

The antics of such people make the Wu section the most interesting
of the sections on empresses and consorts. It differs from the others—
and from those in the three histories of the Han discussed in Chapter
3—in that it deals with more than just the emperors' wives and their
families. Imperial princesses are included as well. The comparatively
titillating flavor of the section is due in part to the personalities of the
people described. But perhaps this characteristic is also a function of
the position of Wu in Chinese history. Only if the Suns could conquer
the rest of the country and lay claim to Heaven's blessing would there
be grounds for considering Wu the legitimate heir to the empire. But
there were grounds for considering either Wei or Shu Han legitimate:
The last emperor of the Han had abdicated to the one, and a member
of the Han imperial clan headed the other. The marginal status of Wu
may have led Chen Shou to feel less need to be discreet in writing
about it, and he may have included the sort of material that he felt
constrained to suppress in his accounts of Wei and Shu. Another
motivation may have been Chen's association with the faction in the
Jin government pushing for an invasion of Wu (see Chapter 5).
Demonstrating the moral bankruptcy of the ruling house of Wu would
have been further proof of the correctness of the pro-invasion party's
position.

HAREMS

Aside from empresses and consorts, as we have seen, the Han harems
had contained large numbers of assorted concubines, handmaidens,
and the like. By the second half of the Later Han dynasty, the harem
was said to have numbered in excess of five thousand women, and there
had been frequent calls for reductions. Besides the enormous
expenditures such a harem entailed, one of the greatest concerns was,
of course, that the ruler would be distracted from his responsibilities.
The problem remained pertinent during the Three States, and the
behavior of both Cao Fang and Sun Hao demonstrated the dangers
to the state if a ruler were permitted to indulge himself. Both Wei and
Wu saw the creation of large harems. Cao Fang's lasciviousness may
have been encouraged by the example of his adoptive father Emperor
Ming, who built up a large harem that occupied his attention at the
expense of the affairs of state.[60] Although it is only obliquely mentioned
in one place in the translation below, Emperor Ming was criticized for
recruiting large numbers of young women for his harem.[61] One passage
from the Wei Epitome reports that he established eight wards in which
to house his ladies of talent (cairen [OMITTED]) according to rank, with those
holding titles of honorable lady (guiren) and lady (furen) or above
occupying the south side.[62] The Wei Epitome goes on to say, "The

emperor often amused himself and feasted there. He chose six women
who were literate and could send correspondence and made them
female masters of writing (nü shangshu [OMITTED])." Finally, the Wei
Epitome observes that "those from the honorable ladies (guiren) down
to shangbao [OMITTED][63] and those who swept the lateral courts [i.e., the
harem] or were versed in entertainment and song, each numbered in
the thousands." Chen Shou described the consequences of Emperor
Ming's excesses: "The emperor built palaces on a large scale and thus
made the people toil; he made extensive levies of girls to fill his harem.
The imperial sons born in the harem died prematurely one after another,
no heir growing up."[64] Emperor Ming's failure to produce an heir was
thus linked—at least in the historian's mind—to his profligacy, and by
extension, so was the fall of the Wei. Profligacy, then, was viewed as
a political problem as well as a moral one, for the growth of a harem
usually was taken to signal a decline in the emperor's engagement in
affairs of state. No doubt similar considerations were operating when
the young Liu Shan wanted to select women to fill out his harem, and,
as we have seen, Dong Yun responded that in ancient times the number
of wives of the Son of Heaven had not exceeded twelve. Dong simply
refused to carry out Liu's instruction, and although Liu acquiesced he
was not pleased.[65]

The Suns were apparently unconcerned about classical precedent. A
large harem developed there, too, and though one cannot be certain
when it began, by the reign of the final ruler, Sun Hao, it was said to
have numbered several thousand.[66] The libidinous activities of the Suns
were specifically cited by Chen Shou as one of the reasons for their
demise, and Sun Quan was likened to Duke Huan of Qi.[67] But the
lessons of Wei and Wu were lost on the Simas. The Jin continued in
their footsteps, taking over some five thousand women from among
Sun Hao's concubines and entertainers. And just prior to that, in the
seventh month of Taishi [OMITTED] 9 (August 273), Emperor Wu widely
selected women of good families to fill the rear palaces. He
accomplished this by first declaring a prohibition on marriage and then
dispatching eunuchs to search throughout the provinces for girls to be
sent to Empress Yang [OMITTED] for final selection. The Jin History relates that,
being of a jealous nature, she chose no real beauties—only those who
were "pure and mature." The families themselves do not seem to have
been keen to have their daughters join the harem, for many young
women are said to have made themselves unattractive to avoid being
selected, apparently as a sweet lady or an even lower rank.[68]

CONCLUSION

The story of the empresses and consorts of the Three States period
is the taste that proves the pudding of George Santayana's time-worn

dictum about what happens to those who forget the past. The examples
of four hundred years of the Han dynasty were manifest and accessible
to all who would benefit from them. And if the import of the lesson
was not immediately apparent, Han political and social critics had
provided clear interpretations. But the rulers of the Three States were
little inclined to master this lesson. Many of the problems associated
with Han consorts, their families, and the harems reappeared in the
Three States. There was nothing inevitable about their recurrence. As
with the Han, the nature of the problems surrounding palace women
grew out of the moral and political culture of the leadership of the
individual states and reflected the rulers' personalities, proclivities, and
weaknesses. Like the men, the women at a dynasty's beginning seem
generally to have been stronger and abler at carrying out their
sanctioned roles. As the quality of rulers declined, along with their
ability to perform properly the duties of their station and to keep their
appetites in check, so did that of the palace women.

In the years prior to the founding of Wei, Cao Cao adopted a policy
of selecting men of ability and attached little importance to birth. This
policy found its analogue in the selection of palace women. Cao Cao's
own pedigree was anything but elite, and he did not feel obligated to
turn to the prominent families for his wives. Although this eliminated
the danger of affinal families coming to dominate the government, it
did not prevent a consort from being able to manipulate an emperor
for her own ends, as the case of Empress Zhen demonstrates.[69] But
whether the palace women might distract a ruler from the affairs of
state was more a function of the inclinations and will of the ruler than
of the women themselves. Emperor Ming and his adoptive son Cao
Fang increasingly directed their interests toward the harem and away
from their responsibilities as heads of state—with predictable results.

Shu Han provided the counterexample. The position of palace
women there was relatively weak, and their families did not exercise
untoward influence on the Lius. One important reason was that the
principal consorts—aside from Liu Bei's wife Lady Wu—did not have
roots in the region, and there were no broad networks of affinal relatives
to try to manipulate the ruler or his consort. Another factor was the
continuing influence of Liu Bei's trusted advisers after his death and
their ability to deter Liu Shan's incipient desire to replicate the examples
of his counterparts in the other two states. Had this not been the case,
Shu Han might well have followed the same path.

In several ways, Wu represented a special case. Marriage alliances
were crucial to the foundation and survival of the Wu state. The Suns
married women who came from families of higher social standing, and
these women may have felt little need to defer to their husbands. The
culture of the Wu court seems to have been comparatively open and

freewheeling, permitting women to play a stronger role than elsewhere.
During the family's struggle for ascendency and in the early years of
the Wu state, Sun women were matches for the early, strong Sun male
leadership. Sun Jian's sister and Sun Quan's mother both seem to have
been capable of offering informed advice on political and military
affairs. Lady Sun had a strong martial bent and was considered as
capable and courageous as her brothers. Whether such women were
common in the Jiangnan region or unique to the Sun family is unknown
but is a question worthy of further study because of its implications
for our understanding of the different roles of women in early imperial
China. Yet as the quality of the leadership declined, so did that of the
palace women, until, like Wei, Wu found its rulers sunk in debauchery
and unable to preserve the state. Thus while writers in early imperial
China might complain about the malignant influence of women, in
truth responsibility lay with the men who set the conditions under
which they lived and acted.

5 Records of the Three States

There is a common tendency to think about China in terms of
comparatively long and relatively stable dynastic periods such as the
Han, the Tang, and the Qing.[1] The Three States period does not follow
this pattern, however, and Records of the Three States differs
significantly from the other standard histories of the early imperial
period. The period it describes was quite unlike those of The Grand
Scribe's Records and the Han History: The subject is three separate and
competing regimes, and the period covered is less than a century. Yet
despite its political fragmentation and brief duration, few historical
periods are as woven into the cultural fabric of a people's existence as
the Three States period is for the Chinese.[2] "If one wishes to understand
China," writes Lyman Van Slyke, "one must have some familiarity with
the history of the Three Kingdoms and with the lore that surrounds
it."[3]

There can be no doubt about the importance of Records of the Three
States to this process of understanding. It is a fascinating work dealing
with an extraordinary period and has long been considered one of the
most important dynastic histories. In its pages are chronicled the ideas
and events of an exciting period in Chinese history—the late second
and third centuries. This was a time of tremendous social, economic,
and political change and an age of great achievements in literature. As
a repository of information and documents concerning political and
military events, people, religion, science, foreign customs, and literature,
Records of the Three States is crucial for an understanding of the
period.[4] Furthermore, the book has indirectly left a lasting impression
on Chinese of all ages through its "pervasive influence in fiction, drama,
and popular religion,"[5] not to mention history and historiography. In
Taiwan alone, there are scores of temples decorated with scenes from
Records of the Three States, as sifted through popular lore and fiction.
The canonization of Guan Yu and the widespread devotions to him
today are just one example of this influence.[6] Similarly, the fact that
the book has long been mentioned in the same breath with Sima Qian's
The Grand Scribe's Records, Ban Gu's Han History, and Fan Ye's Later

Han History as one of the Four Histories (Si shi [OMITTED]) indicates the
high regard in which it has been held.[7]

Although there is no complete Western-language translation of
Records of the Three States, books, articles, and dissertations
sometimes do contain translations of passages or sections. But they
almost never include the relevant parts of Pei Songzhi's commentary,
and the total amount in translation is miniscule. A translator who
sampled broadly from the text was the redoubtable Achilles Fang.[8]
Fang, of course, was translating from Zi zhi tong jian [OMITTED]
[Comprehensive Mirror for Aid in Governing], not Records of the
Three States, and only when the two texts are the same, and in
occasional notes, can he be said to be dealing with Records of the Three
States proper.[9]

CHEN SHOU AND RECORDS OF THE THREE STATES

Chen Shou, author of Records of the Three States, was from Anhan
[OMITTED] prefecture in Baxi [OMITTED] commandery of Shu.[10] In his youth he
studied with an older historian from the same commandery, Qiao Zhou
[OMITTED] (201-270),[11] and learned the Shang shu [OMITTED] [Hallowed
Documents] and the three commentaries to the Spring and Autumn
Annals. But he especially concentrated on The Grand Scribe's Records
and the Han History.[12] Later, Chen was an official in Shu, where
he served as master of records (zhupu [OMITTED]) for the general of the
guards (wei jiangjun [OMITTED]), gentleman librarian in the Eastern
Library (Dongguan mishulang [OMITTED]), and cavalier gentleman-in-attendance
of the Yellow Gates (sanji Huangmen shilang [OMITTED]
[OMITTED]).[13] According to the Jin shu [OMITTED] [Jin History], Chen was removed
from office more than once because he chose to remain aloof from the
eunuch Huang Hao, who controlled Shu at the time.[14]

In 263 Shu was conquered by Wei, and two years later Sima Yan
[OMITTED] (236-290) abolished Wei and became the first ruler of the
Western Jin [OMITTED] (266-317). Chen, however, did not immediately take
office in the Jin government. Sometime before the change of regimes,
he was in mourning for his father, and during the mourning period,
he fell ill and had a maidservant concoct some pills for him to take.
This was apparently a violation of ritual and engendered censure, with
the result that Chen spent some years out of office.[15]

After a time, Chen Shou was recommended as filially pious and
incorrupt[16] by a powerful admirer, the influential official and literatus
Zhang Hua, and so came to serve the Jin. He became an assistant
gentleman editor (zuo zhuzuo lang [OMITTED]), then gentleman editor
(zhuzuo lang), and next chancellor of Pingyang marquisate (Pingyang
hou xiang [OMITTED]).[17] While Chen was serving as gentleman editor,

Inspector of the Palace Writers (zhongshu jian [OMITTED]) Xun Xu [OMITTED]
(d. 289) and Prefect of the Palace Writers (zhongshu ling [OMITTED]) He
Jiao [OMITTED] (d. 292) enlisted him to edit the works of the famous Shu
statesman and strategist Zhuge Liang. The Zhuge Liang ji [OMITTED]
[Collected Works of Zhuge Liang], Chen's edition of Zhuge's works
in twenty-four juan, was submitted to the throne in 274.[18]

With Jin's conquest of Wu in 280, Chen began work on his history
of the three now defunct states.[19] Just when he completed the
manuscript is not known. Chen may have completed portions of the
work as early as sometime prior to the fall of Shu. The Wei section
was certainly well along or complete before the fall of Wu, but the Wu
section itself was probably not begun until after 280, and the complete
work may not have been presented for some years after that.[20] Since
the writing of the work had not been officially sponsored, official copies
were not made until after Chen's death.[21] It was generally well received.
Xiahou Zhan [OMITTED] (243-291), who was compiling his own Wei shu
[OMITTED] [Wei History], was reportedly so impressed with Chen's work that
he quit and destroyed what he had written.[22]

Even so, the praise of Chen's work was not universal, and
assessments of it could be affected by nonhistoriographical
considerations. The Huayang guozhi [OMITTED] [Record of the States
South of Mt. Hua] says:

A few lines later, however, one finds:

Chen's Jin History biography mentions this incident as well but does
not refer to Xun's unhappiness with Chen's history. Rather, it says,
"Xun Xu was jealous of [Zhang] Hua and hated Shou, so he spoke
disparagingly to the Division of Personnel (Li bu [OMITTED]), and Shou was

transferred to be administrator of Changguang commandery. He
declined because of his mother's age and did not go."[25]

The reason for Xun's displeasure over Chen's Record of Wei is not
readily apparent, though given the prominence of the Xun family in
the work and Xun's own role in the events described there, the
possibilities are probably countless. But the problem may have been
a difference over policy between Xun and Zhang, who was Chen's
champion. The two disagreed over plans to conquer Wu, with Xun
opposed and Zhang in favor.[26] Xun may not have wanted Zhang to
have another partisan within the palace, and he may have seen
something in the Record that he felt argued against his position.

In any case, although Chen did not take up the Changguang posting,
later Du Yu [OMITTED] (222-284), who was then general-in-chief who
subdues the South (zhennan da jiangjun [OMITTED]), recommended
Chen to be a cavalier gentleman-in-attendance (sanji shilang [OMITTED]
[OMITTED]). The court actually appointed him attending secretary-preparer of
documents (zhishu shi yushi [OMITTED]), a high office in the
Censorate. By and by, he left office to go into mourning for his mother.[27]

Before she died, Chen's mother had asked to be buried in Luoyang
[OMITTED]. Chen followed her wishes, but once more he met with criticism
on the death of a parent. Some felt that Chen should have returned
her to their home in Shu for burial.[28] At any rate, several years later
Chen was appointed palace cadet of the heir apparent (taizi zhong shuzi
[OMITTED]) but died before he could assume his duties.[29]

Of the sixty-five juan that make up the Records of the Three States,
thirty are devoted to Wei, fifteen to Shu, and twenty to Wu. There are
relatively few textual problems with the work, and there is no doubt
that these are essentially the original work. Although the relative size
of each section may have been partly determined by political
considerations, it is also likely to reflect the amount of material on each
state that Chen had at his disposal. A variety of sources was potentially
available to him.[30] Among them was Wang Chen's Wei History. This
was an official history of Wei that was worked on at various times by
Wei Ji [OMITTED], Miao Xi [OMITTED] (186-245), Wei Dan [OMITTED], Ying Qu [OMITTED]
[OMITTED] (190-252), Xun Yi [OMITTED], Ruan Ji [OMITTED] (210-263), Sun Gai [OMITTED],
and Fu Xuan. The final forty-juan work, however, was completed by
Wang Chen alone on imperial command.[31] Wang's imperially
sanctioned work must be used with care because it avoids or glosses
over matters sensitive to the throne.[32]

Another work Chen must have seen was the Wu shu [OMITTED] [Wu
History]. Again, this project was to some extent collaborative. Zhou
Zhao [OMITTED], Xue Ying [OMITTED] (d. 282), Liang Guang [OMITTED], and Hua He
[OMITTED] all were involved with the Wu History, but the final fifty-five-juan

work is from the hand of Wei Zhao [OMITTED] (204-273).[33] Besides these
official histories of Wei and Wu, Chen Shou would have been able to
draw on at least one privately compiled history—Yu Huan's Wei
Epitome.[34]

Unlike Wei and Wu, Shu seems to have had neither officials charged
with compiling a history nor solid records of court activities. Chen
wrote:

Although he seems to have had much less available to him than in
the cases of Wei and Wu, Chen still managed to complete a section
on Shu.[36] No doubt being from there, having served in its government,
having already authored the Accounts of the Elders of Yi Region, and
having edited Zhuge Liang's works all helped Chen overcome this
difficulty.[37]

Records of the Three States is somewhat different from its famous
predecessors. Lacking treatises and tables, it does not adhere to the
format established by The Grand Scribe's Records and Han History.
Instead, Chen's history consists entirely of annals and biographies.[38]
Also distinctive is the title of the work, for it is alone among the standard
histories in being designated zhi [OMITTED]. Lien-sheng Yang notes, however,
that the three sections of the history have also been called shu [OMITTED] in
the tables of contents and chapter headings of various editions since
at least Song times. This use of shu ("history," "documents") as
opposed to shi [OMITTED] ("history") in the titles of the three major divisions
of the work, Yang suggests, puts it in the line of the Han History, as
opposed to The Grand Scribe's Records; that is to say, it marks it as
a work concerned with a single period, rather than one that overarches
several periods in the manner of Sima Qian and certain later
historians.[39] Although there is no doubt about the coverage of Records
of the Three States, Yang's point about shu is moot.

In his "Shang San guo zhi zhu biao" [OMITTED] [Memorial
Presenting the Commentary to Records of the Three States], Pei Songzhi
refers to Chen's work as guo zhi [OMITTED], or "records of the states."
Whether this was meant to be a title is unclear, although the Zhonghua
shuju editors treat it as such.[40] It may be that "records of the states"
is simply a term that reflects the fact that the three sections were
sometimes treated separately. Evidence for this can be seen in the
bibliographical treatise of the Jiu Tang shu [OMITTED] [Old Tang History],

which places a Wei guo zhi [OMITTED] by Chen Shou (with Pei's commentary)
among the standard histories, while the Wu guo zhi [OMITTED]
(also with Pei's commentary) and Shu guo zhi [OMITTED] (Pei's commentary
not mentioned) are listed, peculiarly enough, in the biannian
[OMITTED]/zawei [OMITTED] (chronologically arranged histories/miscellaneous
unauthorized histories) group.[41] But note that the earlier bibliographical
treatise of the Sui shu [OMITTED] [Sui History] already contains an entry
for Chen's Records of the Three States with Pei's commentary in the
same number of fascicles (sixty-five) as today.[42] The introduction by
the Zhonghua shuju editors in the first edition was ambiguous on this
point, indicating on one page that the three sections were first combined
in a Northern Song edition, while mentioning on the very next page
the Sui History bibliographical entry. In response to research by Miao
Yue proving that the work has had its present title since Western Jin
times, the Zhonghua editors dropped this statement from the revised edition.[43]

The earliest extant block-printed edition of Records of the Three
States is one published by the Directorate of Education (Guozijian [OMITTED]
[OMITTED]) in the sixth year of the Xianping [OMITTED] reign period (1003) in
Northern Song times.[44] There is also a re-engraving dating from the
Shaoxi [OMITTED] reign period (1190-1194), as well as one from the Shaoxing
[OMITTED] (1131-1163) period.[45] The former, which lacks three juan
of the Wei zhi, was supplemented by the latter and photomechanically
reprinted to produce the Bona [OMITTED] edition, one of the four editions
on which the Zhonghua shuju redaction is based.[46] The second of these
four editions, one that has been the basis for various subsequent
typographic and lithographic editions, is the so-called Palace edition
(Wuying dian block-printed edition [OMITTED]) from the Qing
dynasty.[47] This was a redaction of a Ming edition published by the
Directorate of Education at Beijing.[48] The third text used by the
Zhonghua shuju editors is the Jinling movable-type edition (Jinling
huozi ben [OMITTED]), which in turn is a revision of the Feng Mengzhen
[OMITTED] (1546-1605) edition published by the Directorate of
Education at Nanjing.[49] Finally, the Zhonghua editors also consulted
the Jiangnan shuju [OMITTED] redaction, which is a revision of Mao Jin's
[OMITTED] (1599-1659) Jigu ge [OMITTED] edition.[50] Although there are differences
among these and other editions, they are minor, and Chen's text
appears to have been transmitted basically intact.

PEI SONGZHI AND HIS COMMENTARY

Pei Songzhi was born into an important and influential family whose
home of record was Wenxi [OMITTED] prefecture in Hedong [OMITTED].[51] Like
many northerners, the Peis had moved south early in the fourth century

as a result of repeated Xiongnu incursions in the North. Although we
cannot be sure where Pei was born, it would not have been at Wenxi.[52]
By the time he was eight, he is said to have been thoroughly versed
in the Confucian Lun yu [OMITTED] [Conversations] and the Mao version
of the Classic of Poetry, as well as widely read in other works. At twenty
he was made a general of the palace (dian zhong jiangjun [OMITTED]),
and beginning early in the Yixi [OMITTED] reign period (405-418), Pei served
first as prefect (ling [OMITTED]) of Guzhang [OMITTED] in Wuxing [OMITTED] commandery,[53]
then as gentleman of the masters of writing for the Ministry of
Sacrifices (shangshu cibu lang [OMITTED]). Later he served as master
of records of Si province [OMITTED] under Liu Yu [OMITTED], the future founder
of the Liu Song [OMITTED] dynasty (420-479), and was then made viceattendant-clerk
(zhizhong congshi shi [OMITTED]).[54] When Liu
conquered Luoyang, Pei Songzhi was put in charge of the province.
No doubt due to Liu's high regard for him, Pei was transferred back
to Jiankang [OMITTED], where he held a series of offices, including those of
forerunner of the heir apparent (shizi xianma [OMITTED]), administrator
(neishi [OMITTED]) of Lingling [OMITTED],[55] and erudit of the National University
(guo zi boshi [OMITTED]). In 426, Pei was one of sixteen grand commissioners
ordered to tour various parts of the realm. In this capacity
he went to Xiang province [OMITTED]. After returning to court, he served
as a gentleman-in-attendance of the Palace Writers (zhongshu shilang
[OMITTED]) and senior impartial and just of both Si and Ji [OMITTED] provinces,
and was made marquis of Xi district [OMITTED].[56]

As we have observed, Records of the Three States constitutes a major
achievement, especially considering Chen Shou's political and
intellectual environment and the materials with which he had to work.
But there were criticisms of his history, and among these was the
complaint that there were too many omissions. As Leban writes,

This charge of excessive brevity has some merit and is still made
today; one would simply like to know more about many matters.
Examples of great concern to the economic historian are Chen's lack

of detail in describing Cao Cao's creation of the famous system of
agricultural colonies (tun tian [OMITTED]) and the new method of levying
land taxes according to acreage rather than yield. Both were extremely
important administrative changes and the antecedents of major fiscal
institutions in later dynasties, most notably the Tang. But Chen
mentions the first only in passing, and the second not at all.[58] Were
it not for Pei Songzhi's commentary and the Later Han History, we
might completely misunderstand the origins of these two important
institutions.

Fortunately, thanks perhaps to the combination of excitement and
frustration mentioned by Leban, Pei's work exists. Emperor Wen [OMITTED]
(r. 424-453) of the Liu Song dynasty was motivated by the terseness
of Chen's text to order Pei Songzhi to write a commentary to Records
of the Three States. Pei performed his assigned task masterfully. His
contribution lies not only in providing information that helps to clarify
issues in the original history but also in preserving many works that
might otherwise have been lost.[59] Quoting from more than one hundred
fifty works from Wei-Jin times alone, his commentary constitutes a
resource no less important than Chen's. Until recently it was widely
believed that the commentary was, in fact, nearly three times the length
of Chen's original work.[60]

The completed commentary was finished and submitted to the
throne on 8 September 429. The emperor, with considerable foresight,
deemed it an "imperishable" contribution.[61] In 437, Pei Songzhi retired
from office but was then appointed grand palace grandee (tai zhong
dafu [OMITTED]) and concurrent erudit of the National University. He
was also charged with continuing and completing He Chengtian's [OMITTED]
[OMITTED] Guo shi [OMITTED] [National History] but died of illness before it was
done.

The memorial that Pei submitted on the completion of his commentary
reveals a good deal about his approach to writing the
commentary:

Although he does not explicitly mention it here, Pei also provides glosses
and explanations at various points in his commentary.[63] The obvious
importance of Pei's work notwithstanding, some critics have denigrated
it for such real or perceived faults as superfluity (or, alternatively,
inadequacy) and for altering words in quoted material.

CRITICISMS AND CONTEXTS

Despite the praise accorded Records of the Three States, from rather
early on the text has had its critics. One of the oldest criticisms voiced
by Chen's detractors is already mentioned in the Jin History and
concerns the two brothers Ding Yí [OMITTED] and Ding Yì [OMITTED]. The Dings
were important figures affiliated with Cao Zhi at the end of the Han.
This placed them near the very center of power, for although Cao Zhi
is today remembered chiefly as a poet and writer of the first rank, he
very nearly succeeded his father Cao Cao, the single most powerful man
in China, as king of Wei. Ultimately, Cao Zhi's elder brother Cao Pi
was named heir and used his position as a springboard to ending the
Han and becoming the first emperor of the Wei dynasty. Cao Pi's success
in the face of opposition from the Dings cost them their lives.[64] Clearly
they were major players in the political events at the very end of the
Han, yet they have no independent biographies in Records of the Three
States. The account in the Jin History says the reason is that members
of the Ding family refused to pay a bribe to Chen Shou for including
such biographies.[65] Although Liu Zhiji and others accepted this story,
it is assumably spurious. There are more plausible explanations for the
Dings not being accorded their own biographies. For one thing, other
rather well-known figures do not have separate biographies. Several of

the famous writers known as the Seven Masters of the Jian'an Period
(Jian'an qi zi [OMITTED]) have what little is said about them appended
to the biography of Wang Can [OMITTED] (177-217) in juan 21 of the
history.[66] Furthermore, as Rafe de Crespigny writes:

Another story from the Jin History states that Chen Shou's father
had been an adjutant to Ma Su [OMITTED] (190-228) and that when Zhuge
Liang killed Ma Su, Chen Shou's father was treated as a criminal. The
history further reports that Chen Shou himself was slighted by Zhuge
Liang's son Zhuge Zhan [OMITTED] (227-263). Thus, Chen Shou is
supposed to have belittled the two Zhuges' talents in revenge when he
wrote his biography of Zhuge Liang.[68] Miao Yue has marshaled the
arguments of a number of eminent Qing dynasty scholars to refute this
accusation. He concludes by quoting Wang Mingsheng [OMITTED] (1722-1798),
who notes, "The Jin History is fond of quoting diverse accounts
and so is rather rank."[69]

The feature of Records of the Three States that seems to have been
the most controversial is Chen Shou's conferral of legitimacy on the
Wei. By referring to the Wei rulers as emperors and calling his accounts
of them "annals," Chen ordains Wei the legitimate successor to the Han
dynasty and places Shu and Wu, whose rulers are merely accorded
"biographies," in a lesser light. This stance is also reflected in the
amount of space allotted to each of the kingdoms, for Wei gets by far
the preponderance of pages. Wei's legitimacy is conveyed by other
means as well. Chen is silent in the "Wei shu" [OMITTED] section about Liu

Bei and Sun Quan's being proclaimed emperors, and in the "Shu shu"
[OMITTED] [Shu History] and "Wu shu" [OMITTED] [Wu History] sections he gives
their accession dates according to Wei reign years.[70] It is hard to see
how Chen could have done otherwise. He was, after all, a Jin official,
and Jin claimed succession from Wei. Modern historians have usually
sympathized with his predicament.[71] In doing so, they are in part
echoing the bibliographical précis for Records of the Three States in
the monumental Qing dynasty General Bibliography of the Complete
Writings of the Four Treasuries:

It can also be argued that, for all of the attention it pays to Wei,
Records of the Three States does not manifest undiluted allegiance to
that state as the legitimate successor to Han. It has even been maintained
that Chen Shou exhibits a certain favoritism toward his native land of
Shu.[73] The structure of Chen's work clearly concedes the realities of
an immediate post-Han period in which three states existed.[74] Furthermore,
it is Pei Songzhi's commentary, not the history itself, that
preserves the seemingly cynical propaganda and alleged phenomenological
manifestations of approval accompanying Cao Pi's acceptance
of the Han emperor's abdication. This may not exactly demonstrate
partiality toward Shu, but it does suggest an attempt at objectivity or
neutrality on Chen's part.[75]

A final criticism leveled at Records of the Three States is that the
work engages in distortion, especially of events involving the eventual
Western Jin rulers. Perhaps the most influential voices here have been
Liu Zhiji and Zhao Yi [OMITTED] (1727-1814). Zhao, in particular, cites
several cases where Chen's treatment of an event seems at odds with
the facts. He notes, for instance, an inconsistency in how two particular
monarchs' names are handled. In the annals section of the "Wei
History," the last Han ruler is called by his posthumous title Emperor
Xian, even though he became the Duke of Shanyang (Shanyang gong

[OMITTED]) after he abdicated. Cao Huan, however, who sat for five years
on the Wei throne before abdicating to Sima Yan, is never called by
his posthumous title Emperor Yuan [OMITTED].[76] Zhao apparently sees this as
an example of Chen's catering to his Sima superiors.[77] Among the other
examples he gives is the fact that Records of the Three States and the
Wei Epitome are at odds over the role of Empress Dowager Guo in
the dethronement of Cao Fang. The Records has her placing the blame
for his overthrow on the young ruler himself, whereas the Epitome
depicts her as shocked and angry at his being deposed.[78] In another
case, Xi Zuochi's Han-Jin Spring and Autumn Annals recounts Cao
Mao's unsuccessful attempt to resist Sima Zhao and how he died with
a blade through him for his efforts. Records of the Three States,
however, simply says that he died, then goes on to record Empress
Dowager Guo's denunuciation of him.[79] Miao Yue is on the right track
when he writes:

THE DEATH OF EMPRESS ZHEN: A CASE STUDY

The previous section described how the political context in which
Chen Shou wrote may have impinged upon his history. One of the
biographies translated herein provides an excellent example of how
various influences operated on him, while also giving some insight into
the interplay between Chen's work and that of other historians of
the period. These forces and the historiographical tendencies of Chen
and others, which can be appreciated thanks to Pei Songzhi's commentary,
are particularly evident in the biography of Empress Zhen the
Illustrious of the Civilizing Emperor (Wen Zhao Zhen huanghou [OMITTED]
[OMITTED]).[81]

The essential outlines of the story are as follows: Empress Zhen was
initially the wife of Yuan Xi. When her husband went off to govern
You province [OMITTED], she stayed behind in Ji province [OMITTED] to care for
her mother-in-law. When the Cao armies captured Ye [OMITTED], the seat of
Ji province, she was discovered by and subsequently married to Cao

Pi, the future Emperor Wen of Wei. She bore Cao Rui, who would
become Emperor Ming. After Cao Pi acceded to the imperial throne,
Zhen gradually fell out of favor and was replaced by Lady Guo, who
was later named empress in her stead. Empress Zhen was unhappy with
this turn of events and apparently became increasingly difficult.
Emperor Wen grew angry with her and ordered her to commit suicide.
Subsequently, Cao Rui came to the throne. He had her reinterred in
a more lavish tomb and bestowed extensive honors on her family, in
some cases posthumously.

The body of Records of the Three States contains a fairly complete
though terse version of the story, while Pei's commentary cites
sometimes fuller, sometimes alternative versions. The works Pei cites
include two that antedate Chen's work, as well as some written not
long after his death. The first of these is Wang Chen's Wei History,
which deals with the later part of the Zhen story, the events surrounding
her death:

Wang's account of the events surrounding Empress Zhen's death
seems almost perfunctory. From it one gains the impression that,

contrary to being unhappy with Empress Zhen, the emperor wanted
to honor her, and it was only her reluctance that prevented him from
doing so. Even so, he did not entirely give up and and planned to pursue
the matter again later, being thwarted only by the empress's death.
According to Wang, Cao Pi was anguished by the passing of a devoted
spouse and posthumously bestowed the honors of an empress on her.
As we shall see, this version is quite at odds with other accounts of
the empress's death.

The second text that contributes a passage to the story of Empress
Zhen is the Wei Epitome. It describes the initial encounter between Cao
Pi and the future Empress Zhen:

There is another fragment from Yu Huan's Wei Epitome quoted by
Pei Songzhi concerning events following the death of Empress Zhen:

Yu Huan's account here is fairly straightforward, and though clearly
embellished, it is no more so than normal among historical texts of
the period, including the official histories.

Chen Shou's account, written sometime after 265, is the only one
extant that covers Empress Zhen's entire life. It comprises three parts:

The first discusses the empress's family antecedents, tells us that she lost
her father at an early age, and offers a formulaic anecdote that
demonstrates her precocity. The second part, which is of about equal
length to the first, is a terse recounting of her marriage to Yuan Xi and
then to Cao Pi, her giving birth to Cao Rui and the Princess of
Dongxiang (Dongxiang gongzhu [OMITTED]), and then her loss of favor
and death. These latter are described in the following terms:

The third part of her biography, which is three times the combined
length of the first two sections, describes the events after her death,
primarily her reinterment and the honors granted her relatives. In
contrast to the preceding sections, this one is rather detailed.
Considerable space is devoted to the erection of a temple in her memory,
and the memorial proposing it is quoted in full. This is followed by
still further description of the honors bestowed on the empress's
relatives, a number of whom were made marquises. Interestingly,
Emperor Ming made Empress Guo's younger cousin Guo De [OMITTED] the
posthumous son of Empress Zhen's deceased grandnephew Zhen
Huang [OMITTED] and had him take the surname Zhen.

At first glance, Chen's account of all this seems straightforward
enough. But on reexamination, one is struck by the juxtaposition of,
on the one hand, the favor shown Lady Guo and the Ladies Li and
Yin and, on the other hand, Empress Zhen's behavior, which led to
the emperor's dissatisfaction with her and his ordering her to commit
suicide. Add to this the lengthy description of Emperor Ming's
rehabilitation of his mother and her relatives, which clearly gives the
impression of the rectification of an injustice, and it seems clear that
Chen is conveying a discreet message. This becomes even more certain
when he writes in the biography of Empress Guo that "the death of
Empress Zhen resulted from the favor shown Empress Guo."[86] His
assertion is echoed by Xi Zuochi's Han-Jin Spring and Autumn Annals:

Xi wrote during the Eastern Jin, and there is always the possibility
that distance from the actual event encouraged a certain amount of
distortion and elaboration. But we need only recall the passage from
the Wei Epitome cited earlier to realize that the shift in Emperor Wen's
favor to Empress Guo entailed more than a new attraction piqued by
the infusion of fresh blood into the seraglio: Empress Guo had actively
undermined Empress Zhen's position. There can be no question that
Xi's account is close to the truth. But what does that suggest about
Chen's version?

The two quotations from the Wei Epitome provided by Pei Songzhi
are just snippets from what was a complete text, and on reading them,
one is certain that Yu Huan might very well have included an account
of Empress Guo's actions against Empress Zhen. After all, the Epitome
does refer to Empress Zhen's having been slandered, to her having been
improperly coffined, and to her disheveled hair covering her face. As
indicated earlier, the Epitome was one of the sources available at the
time Chen Shou was writing Records of the Three States, and he is
assumed to have consulted it.[88] There can be no doubt, then, that he
intentionally muted his account of Empress Guo's treatment of Empress
Zhen. Unlike Wang Chen, however, he did not suppress it entirely. He
left the alert reader clues to what had actually occurred; if necessary,
his account might be fleshed out by recourse to unofficial works such
as the Wei Epitome.

Even so, the question remains: Why did Chen Shou decide not to
use all that was available to him and write the complete story of the
events surrounding Empress Zhen's death? The Qing scholar He Zhuo
speculates that Chen was reluctant because the Guos were still influential
when Chen was writing.[89] This would probably have made any
reasonable person cautious. But Chen's situation was further complicated
by policy differences in the government. Let us recall that Chen
was a protégé of Zhang Hua. Although Zhang was widely respected

for his abilities, his own origins were relatively humble, and his power
to protect Chen would have been limited. When it came to one of the
most important issues of the day—whether or not to attempt the
conquest of Wu—Zhang was in opposition to both Xun Xu, as noted
earlier, and Jia Chong. In 265, Shu had been defeated and absorbed
by Wei, but Wu was still independent more than ten years later.
Emperor Wu of Jin (Sima Yan) wanted to attack the southern state.
A number of high-ranking officials—including Jia and Xun—were
opposed, while Zhang and others strongly endorsed the plan. In a
memorial to the emperor, Jia and Xun insisted that such an effort was
doomed to failure. The emperor apparently did not agree, for he
ordered Jia to lead the attack. When the campaign experienced logistical
difficulties, Jia and Xun unsuccessfully called for Zhang's execution
because he was responsible for planning grain transport. The Jin
armies were ultimately victorious, however, much to the chagrin of Jia
Chong, who feared that he would now fall into disfavor.[90] Even though
the emperor seems to have been willing to overlook Jia's behavior, this
surely did little to improve Jia or Xun's feelings toward Zhang and
Chen.

In addition to policy differences, there were deeper and more
intensely personal differences that could have affected Chen's work.
Zhang and Chen found themselves embroiled in bitter factional
struggles at court revolving around Jia Chong and his daughter, Jia
Nanfeng [OMITTED] (d. 300), the consort of the future Emperor Hui [OMITTED]
(Sima Zhong [OMITTED]).[91] Xun Xu and his father, Xun Yi, were closely
allied with Jia. Both father and son had proposed that Jia's daughter
be married to the mentally deficient heir apparent Sima Zhong, an idea
for which Xun Yi was widely ridiculed.[92] Zhang Hua, in contrast, later
supported Empress Dowager Yang when Empress Jia sought to have
her executed.[93] As Zhang's protégé, Chen would have found himself
willy-nilly in opposition to the Jia family and its supporters.[94]

Given the context in which Chen wrote his history, then, he could
have expected his work to be carefully scrutinized by people who were
not entirely sympathetic to him.[95] Although according to one account
Xun Xu praised Chen's Record of Ancient States, the same source says
that something in Chen's history of the Wei displeased Xun and led
him to prevent Zhang's gaining Chen an appointment at court.[96] As
noted earlier, considering the prominence of the Xun family in the work,
there were probably numerous opportunities for Xun Xu to take
offense. Even so, what irritated Xun may well have been Chen's treatment
of Empresses Zhen and Guo. Although the details differ, this
episode is in many ways analogous to Empress Jia's treatment of
Empress Yang, and Chen's account could be taken as an implied criticism

of Empress Jia. Moreover, his lengthy description of the honors
bestowed on the Zhen family by Emperor Ming might be seen as a
hint that members of the Yang family deserved similar consideration.[97]
Chen's handling of the Empress Guo affair could also have been
interpreted as being critical of Xun's father, Xun Yi, who had
collaborated with Wang Chen on the Wei History.[98] As we have seen,
the Wei History omitted any mention of Empress Guo's treatment of
Empress Zhen.[99] Chen's circumspection notwithstanding, under such
circumstances his account could not have avoided offending Xun.

Of course, we can never know for certain Chen's reasons for writing
as he did. But the contexts and constraints with which he had to deal
are clear. In hindsight, it would be easy to fault him for not living up
to ideal historiographical standards. But it should be remembered that,
in addition to the political context in which he wrote, history as a field
was still developing and had not come completely into its own.[100] In
any case, Chen did not do all that badly. Despite the pressures and
limitations to which he was subject, he managed to give sufficient
information and leave adequate clues to enable his readers to discern
the truth. This at least seems to have been Pei Songzhi's conclusion when
he compared Chen's version of the Empress Zhen story with that of
Wang Chen:

Fang Xuanling [OMITTED] (578-648) and the other compilers of the Jin
History agreed. They noted that the Wei History by Wang Chen (and
Xun Yi) had covered up a great deal and that it "was not like the true
record of Chen Shou."[102] Fang and his collaborators may indeed have
been thinking in part of Chen's treatment of Empress Zhen when they
wrote in their appreciation of Chen and other Jin historians:

In summary, although Records of the Three States has some
shortcomings—shortcomings that led Pei Songzhi to compile his
commentary—as early as the late Six Dynasties and Tang, Chen Shou
was recognized as an outstanding historian who had succeeded, under
difficult circumstances, in preserving his integrity by producing an artful
history that withheld little from the accomplished reader.

WHO IS INCLUDED?

A word needs to be said about the criteria Chen Shou used in
selecting women to include in his history, for not all palace women or
even all rulers' wives have entries in the Records of the Three States
fascicles on empresses and consorts. Moreover, the treatment of them
differs somewhat state by state. In picking those to be included in the
section on Wei, Chen followed criteria similar to those applied later
by Fan Ye in his Later Han History.[105] All had been empresses and are
listed by their titles, as befit the consorts of the rulers of the legitimate
successor state to the Han. This legitimacy is underscored by the
omission of the term "empress" (hou) from the titles of the chapters
on Shu and Wu. Curiously, however, the section on Shu also lists
empresses by their titles, though Liu Bei and Liu Shan are naturally
not referred to as "emperor." The Shu chapter differs from the other
two in that it combines the biographies of the consorts with those of
the younger male offspring, rather than separating them, as is done for
Wei and Wu. This is, of course, a function of the relatively small number
of persons involved. Fascicle 50 differs from the other two in not listing
any of the women under the title of "empress." With one exception,
all are listed under the title of "lady" (furen), though most had been
declared empress. The exception is Sun He's consort née He, who was
mother of Sun Hao, the last ruler of Wu. Although Sun Hao bestowed

on her the title of empress and then empress dowager, she was listed
as Dame He [OMITTED] to indicate that she was not the wife of a ruler.[106]
Another difference is that Fascicle 50 includes wives of Sun Quan—
Lady Xie and Lady Xu—who had not been declared empresses, either
in their own lifetimes or posthumously. Their inclusion is further
indication that Chen Shou believed there were no grounds for
considering Wu to be a legitimate successor to the Han.

Aside from the empresses who appear in Fascicle 5, the Caos had
other wives who were mothers of younger sons who did not become
emperor. These are mentioned in Fascicle 20 with the entries for their
sons (see Appendix I, Tables 9 and 10). The information is sketchy,
and for most Chen gives little more than a name.[107] Fascicle 59, on
the five sons of Sun Quan who did not rule (Sun Liang and Sun Xiu,
having been rulers, share a separate fascicle with Sun Hao, the final
ruler of Wu), differs somewhat because the mothers of these five are
covered in Fascicle 50. Still, Fascicle 59 contains some additional
information that complements parts of the biographical sketches found
in the section on consorts.

CONCLUSION

There can be no question that Chen Shou was an accomplished
historian for his time, the weaknesses of Records of the Three States
notwithstanding. Those weaknesses were to a large extent the product
of the very difficult circumstances and highly charged and politicized
atmosphere in which he wrote. Chen omitted or only alluded to
important particulars, which led Pei Songzhi to supplement Chen's
work with numerous quotations from contemporary or near-contemporary
sources. But while Pei's additions point up the lacunae
in Chen's work, they also reveal how adroit Chen was in handling
sensitive events and signaling the reader of the need to look more deeply
into a matter. His handling of the death of Empress Zhen is an excellent
example of the context in which he wrote and the constraints under
which he labored. Thanks to Pei Songzhi's selection of material for his
commentary, we are able to appreciate the nature of Chen's
achievement. Small wonder that Chen's work survived rather than
Wang Chen's.

The women whom Chen selected for his history are not, of course,
representative of women at large, nor do they include the totality of
the wives or concubines of any one of the rulers. As we have seen, Chen
applied fairly narrow criteria in making his choices, criteria that resembled
those of historians before him, and ones that were later
followed by Fan Ye. Given what we know about the sources for Fan's

Later Han History, it seems likely that the same criteria were applied
by other historians among Chen's contemporaries, such as Hua Qiao.

We have noted that these fascicles are as much about the male
relatives of the women included as they are about the women
themselves. Wu Jing's biography appears at his sister's entry, and
likewise the entry for Xu Kun, father of Sun Quan's Lady Xu, is
included with hers. Liu Shan's two brothers and his son are given three
quarters the space of the four imperial wives in the Shu section. Thus
these two fascicles each do double duty: The Wu section incorporates
material that in other histories is found in the sections on the imperially
affined families, whereas the Shu section includes the entries normally
found in a section devoted to younger sons of the imperial line. Only
with Fascicle 5, the Wei chapter, do we have something like the typical
standard history section on empresses and consorts.

One wishes that Chen could have broken with the tradition
established by his predecessors and given us a fuller picture of the
women about whom he did write. We are not usually told their given
names, and the descriptions of them are almost entirely from the
standpoint of the impact of their actions on the ruling house. Aside
from a particular woman's being either jealous and scheming or warmhearted
and wise, in most cases the historian tells us little of their
personalities, their lives, and their aspirations. Still, as the reader will
find in the following translation, a careful and sympathetic reading of
the fascicles on palace women—supplemented by Pei's commentary and
by material from other parts of the Records and elsewhere—yields the
outlines of these women and their lives, and though one can never hope
to know them intimately, it is possible at least to have a sense of them
and their world. Clearly there were some extraordinary individuals
among the royal women of the Three States. If the age is rightly known
for having produced men of heroic stature and exceptional ability, the
limited sample of the following chapters is evidence that it also yielded
women who were their match.