The Eternal Palaces of the Western Han: A New View of the Universe

Author(s): Jessica Rawson

Source: Artibus Asiae , 1999, Vol. 59, No. 1/2 (1999), pp. 5-58
Published by: Artibus Asiae Publishers

Stable URL: https://www.jstor.org/stable/3249861

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Artibus Asiae Publishers is collaborating with JSTOR to digitize, preserve and extend access to
Artibus Asiae

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 JESSICA RAWSON

THE ETERNAL PALACES OF THE WESTERN HAN:

A NEW VIEW OF THE UNIVERSE

Two recent archaeological finds, the terracotta warriors of the First Emperor of Qin (reigned
zzI-Io B.C.) and the jade suits of the Imperial family of the Han dynasty (zo6 B.C.-A.D. 2.0),
have been widely seen in the great capitals of the world as center pieces of several major exhibitions.
But although we have all marveled at these phenomena, on reflection it seems that we have asked
ourselves too few questions about them. We have not considered whether these extraordinary finds
are really consistent with our usual picture of ancient China. Indeed I shall argue that they are not.
In this paper, I claim that the terracotta warriors and the jade suits show that a fundamental
transformation of ancient Chinese burial practices must have happened during the latter part of the
third century B.C. To assess the scope of this change, the warriors and the jade suits have to be
measured against assemblies of bronze ritual vessels, bells and weapons, placed within wooden coffins
in shaft tombs that had delineated the afterlife, and the ritual practices of that afterlife, during the
preceding Zhou period. Although from the third century B.C., and indeed earlier, many new im-
pulses are evident in archaeological finds, political unification under the Qin and the Han seems to
have been the critical moment when a major change was consolidated.
Four particular features of Qin and Han period tombs make this present re-examination necessary.
The first two I have already mentioned: the terracotta warriors, installed in pits to the east of the
tomb mound of the First Emperor, Qin Shi Huangdi, were completely unprecedented when first
discovered. Today we know that armies of ceramic figures on a smaller scale were buried near Han
Imperial tombs at Xi'an and at the entrances to some of the rock-cut tombs to be described below.
All such ceramic armies differed fundamentally from the Shang and Zhou presentations of military
might in typical Yellow River area tombs.
Jade suits first became known to present-day archaeologists and scholars from those belonging to
a king of the Han period State of Zhongshan, Liu Sheng (died 113 B.C.), and his consort Dou Wan.
Now it transpires that these are merely two examples of what was a widely used burial garment
among the kings of the Liu family (who formed the Han dynasty), their consorts, and also, very
occasionally, lower ranking members of the /lite.I Although in many periods before the Han, jades
had been used to cover the bodies of the dead, such carefully made suits without a chink of space
between the plaques, had not been customary before the ascendency of the Liu family. What is more,
as these suits are extravagant of material and complex in workmanship, their similarity one to an-
other, not to say uniformity, indicates both some sort of shared belief system and ideology and also a
shared workshop tradition.

I For a survey of some of the known suits, see Kaogu 1981. I, 51-58 and Wenwu I989.IO, 60-67. For a list of the identities of the
members of the Liu family buried in excavated tombs, see Wenwu 1992. 2, 37-43. One striking exception to the proposition that
such jade suits were confined to members of the Liu family and their immediate associates is the tomb of the King of Nan Yue in
which Zhao Hu was interred in a suit. For all references to this tomb, see Xi Han Nan Yue Wang mu, 2 vols. (Beijing: Wenwu
chubanshe, 991i), hereafter referred to as Nan Yue Wang report.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 A yet more surprising development was the way in which the funerary officials of minor Han
kingdoms sought out tomb sites on small mountains into which they cut several linked rooms on a
horizontal rather than a vertical axis. Such tombs were a radical departure from the standard, late
Zhou period plan with a vertical shaft and nested coffins in wood at the base. Shaft tombs had,
indeed, been standard from the neolithic period. Given that it would have been perfectly possible to
create cave-like tombs with many separate rooms within the loess,2 it is all the more remarkable that
tombs with a horizontal arrangement seem to have been especially favored in areas with small rocky
mountains.

The fourth change is illustrated most clearly by the repertory of bronze vessels of Western Han
tombs, as seen, for example, in those of Liu Sheng (fig.I). The substantial, carefully cast bronze food
and wine containers used in the rituals addressed to the ancestors during the Warring States period,
(and earlier), seem to have been relegated to an inferior position, and even, in many instances, com-
pletely abandoned. Instead, Han tombs contained wine flasks, basins, lamps, incense burners and
braziers, often made especially attractive by fine decoration in inlay or gilding on bronze, and quite
unlike the pieces previously used for ancestral rites. It is possible that offerings in substantial highly
decorated bronzes continued and that these were just not buried. But this is in itself would have been
a major change, as in earlier Warring States tombs, the traditional vessels, especially ding and hu,
were interred. Ding and hu do appear in Han tombs, but seem no more and often less important than
other vessel categories. If food and wine were to be offered by the tomb occupant (and they probably
were), vessels of lacquer or clay, whose character we cannot at present clearly discern among the
surviving pieces in major tombs, must have been used.3
These four innovations indicate a radical departure from previous ritual and burial practices. If the
practices had altered, it is likely that salient attitudes to the ancestors and spirits had also been
transformed. Indeed, I shall argue that Western Han tombs suggest that some sort of implosion had
taken place, contributing elements from many areas, but leaving us with a situation in which we are
forced to conclude that concern for or knowledge of the previous Zhou culture was absent. The atti-
tude to the life after death illustrated by the tombs was notably preoccupied with creating a
microcosm of the universe within the confines of the tomb for the benefit of the tomb's occupant, as
well as asserting the tomb occupant's claims to a prominent place in the afterlife hierarchy.
In this paper I shall address this development through an account of the Western Han tombs of
the kings of the Liu family and through finds of the immediately preceding period of the late Zhou,
at which time some new developments had begun to appear. But although antecedents are, indeed,
to be found earlier than the third century B.C., I will argue that the unification under the Qin and

2 Shaft tombs within which an additional cave-like compartment was dug out at the foot were employed in Shaanxi province. For
Qin state tombs of this form, see Kaogu yu wenwu I98I.I, I2-38. While the form of these graves may have contributed to the
development of large, many-roomed tombs on a horizontal axis, the fact that they did not initially stimulate such a development in
Shaanxi indicates the priority of the mountain tombs in eastern and central China in providing models for Han tombs on a
horizontal axis lined with stone slabs or fired clay bricks.
It is beyond the scope of this paper to consider further the nature of the offerings to the ancestors. Buildings at or near the tumuli of
the Warring States, and the tomb of Qin Shi Huangdi indicate that offerings to the tomb occupant were made near the tombs. But
this tells us nothing about the offerings, if any, made by the tomb occupant. Developments in the Eastern Han seem to indicate
that forms of respect and offerings had to be thought up and put into practice, for example see comments on practices in the later
Han, B.J. Mansvelt Beck, The Treatises of Later Han, Their Author, Sources, Contents and Place in Chinese Historiography (Leiden: E.J.
Brill, 1990), Io5-o8.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
Han was responsible for a rupture with, rather than simply a continuing development from, past
practice. Political unification gave the Qin and the Han emperors an opportunity to assimilate
systematically practices and beliefs from many parts of the area we today call China, but which, in
the fourth and third centuries B.C., had been parts of independent political entities. What resulted
was not only a new political, social and economic order, but also, it now seems, a new religious or
ideological outlook.
The spirits revered included many awesome figures additional to, or instead of, the ancestors to
whom the Shang and Zhou ritual banquets had been addressed.4 To describe such spirits and to
enlist their support, objects and decorative elements were adopted from the south, from the east, and
even from the areas beyond the Chinese-speaking world, giving physical existence to a new view of
the universe. These imports or additions to traditional Yellow River material culture seem to have
been viewed not simply as dazzling new luxuries, but as potent pieces whose value may have
stemmed from their links with distant and powerful spirit realms.
Fundamental to this account is the view that the landmass we now designate as China was far
from unified during the periods known by the names of the ancient dynasties, the Shang (circa
Ioo-circa Io5o B.C.) and the Zhou (circa Io5o-22I B.C.). These two dynasties held territories along
the Yellow River and its tributaries, especially the River Wei. Throughout the period prior to the
flight of the Zhou from their capital near Xi'an in 771 B.C., many other groups and probably even
some states flourished in other areas. The Yangzi River, in particular, supported highly developed
societies. These did not necessarily seek hegemony, but remained centered in Sichuan, Hubei and
Jiangxi, where agricultural land was abundant. During the Spring and Autumn (770-465 B.C.) and
Warring States (465-22I B.C.) periods, regional diversity contributed to distinctive cultures in many
of the states into which the landmass was now divided.

In this paper, the cultural characteristics of the Yellow River area will be treated as the basis on
which all earlier accounts of ancient China have been founded. The argument that I shall put forward
describes many of the features of the Qin and Han tombs as borrowed from areas outside these
Yellow River centers, especially from the state of Chu and from the east, and it depends upon the
view that these areas had earlier fostered cultures that were in many respects distinct from those of
the Yellow River.

During the Warring States period there had, of course, been extensive contact between the states,
and practices of one area had been borrowed by another. Such interchange was, above all, evidence of
diversity of practice and of material culture as between the different regions. However, Western Han
tombs and the picture of the afterlife presented in them are remarkable for their consistency, between
tombs of individuals, (especially between the kings of the eastern kingdoms), between tombs of
people at different levels of the extensive social hierarchy and between groups in different regions.
Such uniformity in description of the universe implies a unified ideology. So, not only will the paper
describe the ways in which features of a new religious and material assemblage were drawn together

4 The balance between offerings to the ancestors and the concern with other spirits has been insufficiently considered. A significant
contribution to this debate has been made by Mu-chou Poo, In Search of Personal Welfare, A View of Ancient Chinese Religion
(Albany: State University of New York, 1998). Poo makes a strong case for the large role of the beliefs in spirits and deities (other
than the ancestors) in ancient China. However, he seems to remain anxious to demonstrate that these beliefs were in some way
"popular," while having to admit that they were espoused by the highest ranking of the elite. In addition, he is inclined to see
more continuity between the Shang and early Zhou and the later Zhou and Han period than can easily be sustained on
archaeological evidence.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
from many areas, it will also emphasize the ideological coherence or uniformity which could only
have been achieved as an aftermath of political unification. Indeed, a factor that was undoubtedly
important, but which cannot be extensively pursued here, is the need that the First Emperor and his
Han successors felt to reinforce their political control with religious rites and other practices.
The paper will start by describing the principal features of the major Han tombs that are the basis
of this account, with references to Qin and earlier material where relevant. It will then offer some
information on the intellectual background as seen from excavated and transmitted texts. In the
second part of the paper, the sources of the new object categories and tombs structures in the south,
the east and the northern borderlands will be described.

MAJOR FEATURES OF IMPERIAL QIN AND WESTERN HAN TOMBS

We have insufficient evidence of the shape and plan of the tomb of the Qin Emperor to consider it in
any detail. Although described by Sima Qian (circa 145-86 B.C.) in the Shi ji, we cannot be sure
whether his description of this tomb as depicting the constellations in the heavens and the rivers of
the world is correct. It would, of course, be significant if this account described the actual tomb
(rather than merely Han conventions), because if this were so, it would indicate that an attempt to
create a tomb as a microcosm of the universe had begun before the Han period (it did of course have
precedents as we shall see). We can imagine that the Qin Emperor's tomb is likely to have been as
least as massive and as complex as the well-known, late fifth-century tomb of the Marquis Yi of Zeng
in Hubei province.5 It may have been even more elaborate. However, because the Qin tomb was
located in the center of Shaanxi province, and was thus physically and chronologically sited near
vertical shaft tombs of a very ancient type, it may or may not have been a fairly conventional shaft
tomb. Thus, we do not know whether it anticipated the horizontal arrangement that became almost
universal during the Han period. While we have no archaeological evidence of the orientation of Han
period Imperial tombs,6 interestingly, Sima Qian describes the tomb of Emperor Wen (18o-57 B.C.)
as cut into a cliff face, near Xi'an.7
We know that rock-cut tombs were a high priority from the significant numbers of immense
tombs for members of the Liu family found in the hills of north-eastern and eastern China.8 The
principal and most impressive examples are the tombs of the Han kings of Chu, Liang, and Lu at
Xuzhou, Yongcheng and Qufu, all within a relatively confined region in the northern part ofJiangsu
province, the east of Henan province and the south of present-day Shandong province.9 The best
known, are those of the King of Zhongshan, Liu Sheng, and his consort, Dou Wan, at Mancheng in

5 For all references to the tomb of the Marquis Yi of Zeng, see Zeng Hou Yi mu, 2 vols. (Beijing: Wenwu chubanshe, 1989), hereafter
referred to as Zeng Hou Yi report.
6 For a discussion of the Han Imperial tombs based upon textual evidence, see Michael Loewe, "The Imperial Tombs of the Former
Han Dynasty and Their Shrines," T'oung Pao, 78 (1992), 302-340; Michael Loewe "The Imperial Way of Death in Han China," in
Court and State Ritual, I. J. McDermott, ed., (Cambridge: Cambridge University Press, 1999), 81-III.
7 Wu Hung, Monumentality in Early Chinese Art and Architecture (Stanford: Stanford University Press, 1995), 166.
8 While some scholars have paid considerable attention to the rock-cut tombs, see Wu Hung, Monumentality, 127-42, the exact
importance of these tombs as providing an impetus for a entirely new tomb structure at all levels of society has been overlooked.
This extraordinary development was to have an indelible effect on Chinese burial practice, most especially in encouraging a
disposition of rooms along a horizontal passage.
SFor the excavations at Xuzhou see note II below; for those at Yongcheng, see Yongcheng Xi Han Liangguo wangling yu ginyuan
(Zhengzhou: Zhengzhou guji chubanshe, 1996), hereafter referred to as Liangguo report; for the tombs at Qufu, see Wenwu 197.-5,
39-44, 54, and back cover.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
Hebei province (fig. z).Io Although they are on a less impressive scale than those of other kingdoms,
these two tombs are important for this discussion because their contents were completely
undisturbed. The tomb of the King of Nan Yue, Zhao Hu (I37-zz B.C), at Guangzhou, Guangdong
province, a tomb that seems to mimic the tombs of the kings in very many respects, can also tell us a
lot about the range of contents in such burials.
The main tombs at Xuzhou are those at Beidongshan, Guishan and Shizishan, small limestone
hills within the greater area of the present city (fig. 3).n All of the tombs have long, neatly squared off
access passages, with small chambers either side. The coffin chambers were hollowed out at the rear,
with a reception room, dance hall and other rooms, including kitchens, granaries and storerooms, to
make up complete palaces. The tomb at Beidongshan (dated by the excavators to the time of Wendi)
had a complex of subsidiary chambers halfway along the principal access passage. The foreman or
planner of the tomb had obviously not realized that the small hill was not all solid rock, but had a
large fall of earth on one side into which some of the rooms branched horizontally (fig. 3a). Here the
rooms were lined with stone slabs. In the tomb at Guishan, belonging to King Xiang, the title taken
by Liu Zhu the sixth Western Han King of Chu (128-I6 B.C.), the separate accommodations for the
king and his consort were joined by a passage between two of the rear rooms (fig. 3b). The better
known tombs at Mancheng in Hebei province have a simplified arrangement of rooms (fig. 2).
Within the innermost chamber of Liu Sheng's cliff palace, his private retiring room was lined with
vertical stone slabs to make a room in a larger cave." The patterns of the Liang kingdom tombs at
Yongcheng are much more complex, with rows of small chambers along the extensive passageways
(fig. 3c).I3
These magnificent tombs set their imprint on the many lesser burials in this and other parts of the
empire. In Jiangsu tombs of several chambers were constructed from stone slabs set vertically, as in
Liu Sheng's private chamber. Many such slabs carry carved scenes of banqueting, excursions in cha-
riots, dancing, music and processions, buildings and deity figures (fig. 4). Less elaborate, but closely
related in form, are tombs of brick in Henan, some made with large slab-like bricks, others of small
ones. These latter tombs were often vaulted. Such constructions also were decorated with incised and
painted scenes or with stamped motifs, covering the same range of activities.4 While it is necessary
to recognize regional and chronological differences, the fundamental role of the mountain tomb in
setting the model for lesser burials is essential to our understanding of the ways in which tomb

0o For all references to Mancheng, see Mancheng Han mu fajue baogao, 2 vols. (Beijing: Wenwu chubanshe, 1980), hereafter referred to
as Mancheng report.
ii For Guishan, see Kaogu xuebao 1985.I, 119-37; and KaoguI997.2, 36-46; for Beidongshan, see Wenwu 1988.2, 2-18, 68; for Shizishan,
see Kaogu 1986.I2, I-16, Connoisseur (Autumn 1995), I6-37, Wenwu 1998.8, 4-33.
12 As Wu Hung has noted in connection with the tomb at Mancheng, the slabs resemble massive wooden planks in their size and use;
so too do the roofing slabs in the tomb at Beidongshan. Wu Hung, "The Prince of Jade Revisited: The Material Symbolism of Jade
as Preserved in Mancheng Tombs," in Rosemary Scott, ed., ChineseJades, Colloquies on Art & Archaeology in Asia, No. 18, (London:
School of Oriental and African Studies, University of London, 1997), I47-69.
13 Liangguo report, figs. 64, 65.
14 For decorated Han tombs, see Kite Finsterbusch, Verzeichnis und Motivindex der Han-Darstellungen, 2 vols. (Wiesbaden: Otto
Harrassowitz, 1966 & 1971); and Jean M. James, A Guide to the Tomb and Shrine Art of the Han Dynasty 2o6B.C.-A.D. 22o (Lewiston/
Queenston/Lampeter: The Edwin Mellen Press, 1996); Xin Lixiang, "Han huaxiangshide fenqu yu fenqi yanjiu," in Yu Weichao,
ed., Kaogu leixing xuede lilun yu shiji (Beijing: Wenwu chubanshe, 1989), 234-306. For a further discussion, see Jessica Rawson,
"Eternal Mansions: The Tomb of the King of Nan Yue and those of the Imperial Princes in Eastern China," a German translation
will be published in the catalogue of an exhibition to be held at Frankfurt 1998-i999.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
structures throughout the empire were developed in the Han period.IS Indeed, the close similarities
between tombs, as with the jade suits mentioned above, illustrate the ideological consistency
achieved, it would seem, as early as the period of Wendi.
The Imperial princes took into the afterlife objects for daily use, especially drinking and eating
vessels and small tables. In their shapes many of these were derived from lacquers. Elite tombs held
sumptuous versions of these vessel categories, often made of precious materials: gold and silver and
also gilded bronze.I6 Other highly decorated bronzes have been mentioned: elaborate wine flasks, as
in the tomb of Liu Sheng, lamps, braziers, and incense burners (fig. I). Fittings, in the shapes of
animals that supported screens, were intriguing additions to the tomb furnishings, as were tents or
canopies, also seen in the tombs ofLiu Sheng (fig. 2).
Quite a number of the finest vessels were made of jade. Indeed, the rock-hewn tombs held large
numbers of jades, not just jade suits and vessels, but also many discs, weapons, and iron weapons
with jade fittings, jade ornaments, seals and pendants, even occasionally animals in jade. In all such
cases, the use of jade, rather than less valuable materials, such as lacquer or bronze, suggests that this
material, jade, had a specific significance. Again, lesser tombs held alternatives: discs were painted or
stamped or carved on stone and brick and sometimes, as at Mawangdui, they were painted on lacquer
or provided in wooden copies.7
While the terracotta warriors are the most magnificent array of figures known to date, all major
Han tombs seem to have been supplied with figurines in ceramic or sometimes wood. It is often
argued that figures were buried as substitutes for people, but such suggestions may be misleading, as
people were, in fact, interred with or alongside high ranking members of the e1ite. The First Emperor
was accompanied by his concubines and perhaps also some relatives.:8 One of the Han-period kings
of Chu, buried at Shizishan in Xuzhou, was interred with his cook, identified by the archaeologists
from his seal: no doubt the king did not wish to take any chances with his banquets and wanted a
particular cook.I9 The cook's attendants were of clay, suggesting that where a specific individual was
to accompany a king, that person was buried; but where a generic type was required, they could be
provided in clay or bronze or even jade or stone (fig. 5).
Humans and figures in wood, clay, bronze, stone and jade all belonged to a single system of
presenting the inhabitants of the afterlife. We can see the ways in which humans and models seem to
have been treated as equivalent from two examples. First, as Sofukawa Hiroshi has suggested, if the
terracotta warriors were to face the armies of the Six States massacred by the First Emperor in life, an
army of clay seems to have been viewed as a suitable defence against the armies of spirits.20 The

I5 For a survey of tomb types, including several not mentioned here, see Pu Muzhou, Muzang yu shengshi. Zhongguo gudai zongfiao zhi
shengsi (Taiwan: Lianjing chuban shiye gongsi, 1993); and Li Rusen, Handai sangzangzhidu (Jilin: Jilin daxue chubanshe, 1995).
i6 Indeed in place of ritual bronze sets, some forms of lacquer or ceramics decorated as lacquer seems to have been prevalent, see the
Mancheng report, vol. 2, pls. 79, 84-94, and pieces on loan to the Metropolitan Museum of Art, New York; Peter C. Sturman,
"Celestial Journeys: Meditations on (and in) Han dynasty Painted Pots at the Metropolitan Museum of Art," Orientations 19, no. 5
(May 1988), 54-67, fig. 4. See also Michile Pirazzoli-t'Serstevens, "The Art of Dining in the Han period: Food Vessels from Tomb
No. I at Mawangdui," Food and Foodways 4 (1991), 209-19.
I7 Tomb MI at Mawangdui had painted examples of jade discs shown on the middle inner coffin, a screen and the famous banner. The
tomb was also provided with a case of wooden bi, Changsha Mawangdui yi hao Han mu, 2 vols. (Beijing: Wenwu chubanshe, 1973),
pl. zz4, hereafter referred to as Mawangdui report.
i8 Wang Xueli, ed., Qin Shihuangling yanjiu (Shanghai: Renmin chubanshe, 1994), II4-18.
i9 Personal communication from the archaeologists.
20zo Sofukawa Hiroshi, "Lingmu zhidu he linghun guan," Wenbo 1989.2, 34-38.

IO

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
pottery figures and the dead soldiers were equivalent.2 Second, contract documents for the purchase
of the land in which Eastern Han period tombs were constructed often set down a standard clause by
which any persons who had at any previous date been buried in the said plot were deemed to have
become servants of the tomb owner, perhaps alongside other servants who had been provided in
ceramic or wood." Where ceramics or other materials were employed to make figurines, a hierarchy
seems to have been observed, with bronze or stone being more prized than ceramic. Thus stone
figures (fig. 5) stood within the inner chamber ofLiu Sheng's tomb, while ceramic ones occupied the
outer ones. It would seem that a systematic series of hierarchies operated in all aspects of the contents
of the tombs, as well as in their structures - already mentioned above.
To some extent the hierarchies were also maintained by the sizes of the figures and other objects,
but in some respects relative size seems to have been unimportant. Liu Sheng had full-sized chariots
and some parts of miniature ones; he had small attendant figures in stone and ceramic and miniature
ones in bronze and in jade. Liu Mai in the Kingdom of Liang was buried at Yongcheng with mini-
ature chariots. Much more important than scale seems to have been correct presentation of the
necessary parts to make a vehicle or servant function correctly. Thus the miniature chariots
belonging to Liu Mai each had a full set of gilt bronze fittings; the half-size bronze chariots of the
First Emperor were also complete in all their components.3 The case of the terracotta warriors is
particularly interesting, because, although of clay and thus less precious than the bronze of which the
First Emperor's chariots and drivers were made, they were fully life size. The scale of the warriors
seems to have been determined by the weapons that they carried, which were real weapons, some of
them confiscated from the Six States. Such weapons required life-sized warriors of necessity, as Lothar
Ledderose has argued.4 These weapons were, presumably, expected to make the army effective.
Dancers and story-tellers were very common among figure groups in all tombs. Scenes of feasting,
entertainment and ceremonial processions, seen on later stone slabs and bricks (fig. 6), seem to have
been their equivalent. Once again proper provision of the appropriate functions or needs seems to
have been important and the means by which they were supplied, thus either as ceramic figures or as
carved scenes on stone, less so.
Some of the figures may have been spirit mediums described in the dictionary, the Shuowen jiezi:

Spirit mediums are invocators (zhu). They are women who can perform services to the Shapeless and make the
Spirits come down by dancing. ...Among the ancients, Wu Xian was the first to act as a spirit medium.25

Instead of interpreting the figures of entertainers and servants as simply to provide comfort and
enjoyment for the tomb occupant, we might consider them and the carved and painted scenes as
making available the correct preparation of meals and ceremonies for the appropriate moments of the

z' That large numbers of soldiers were slaughtered in battle is evident not only from textual accounts but also from excavation of a
portion of the burials at the site of the battle of Changping (260o B.C.), Wenwu 1996.6, 33-40.
22zz Discussed by Terry F. Kleeman, "Land Contracts and Related Documents," in Makio Ry6kai Hakase Shdju Kinen Ronshzu: Chugoku no
shukyo, shisi to kagaku (Tokyo: Kokusho Kankokai, 1984), 1-34, who talks about the intention that all burial preparations should be
efficacious.

23 Wenwu 1983.7, 17-2I1; Wenwu 199I.I, I-13, I4-19.

24 Described in "Ten Thousand Things: Module and Mass Production in Chinese Art," The Forty-seventh Andrew W. Mellon Lec-
tures in the Fine Arts given at the National Gallery of Art, Washington, D.C., March I-April 5, 1998.
25 Xu Shen, Shuowen jiezizhu (Shanghai: Shanghai guji chubanshe, 1981), zol. Quoted after Lothar von Falkenhausen, "Reflections on
the Political Role of Spirit Mediums in Early China: The Wu Officials in the Zhou Li," Early China 2o (1995), 279-300. See also
Susan Erickson, "'Twirling Their Long Sleeves, They Dance Again and Again...' Jade Plaque Sleeve Dancers of the Western Han
Dynasty," Ars Orientalis 24 (1994), 39-63.

II

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
day or for seasonal rites and rituals.26 Some support for such a view comes from a few items in early
Han tombs. Figures of peach-wood in the tomb of the lady Dai at Mawangdui (circa 186 B.C)., for
example, seem likely to have belonged to the peach-wood ceremony, as discussed by Derk Bodde
quoting Wang Chong (A.D. 27-circa 97):

In the midst of the eastern sea there is the Tu-so (Crossing the New Year) Mountain, on which there is an
enormous peach tree, which twists and coils its way over a distance of three thousand li. Between its branches,
on the northeast, there is what is called the Gate of Demons (kuei men), in and out of which pass a myriad
demons. Above there are two divine beings, one called Shen Shu, and the other Yii Lii. They watch and control
the myriad demons, and those that are evil and harmful they seize with rush ropes and feed to tigers. This

being so, the Yellow Sovereign (Huang Ti) has prepared a ritual for their seasonal expulsion (shih ch'u), in
which large peachwood figures are set up. [Representations of] Shen Shu and Yii Lii, as well as of tigers, are
painted on gates and doors, and rush ropes are hung up, so as to ward off [evil demons].27

Patricia Berger has argued that some of the figures found in the pits near the tomb of Jingdi (reigned

156-141 B.C.) may be figures of youths and young girls to take part in ritual fights as part of the
ceremony of the Great Exorcism.28 Although it is not easy to determine the roles of the figures scenes
in the early Han tombs, it is obvious that many of the Eastern Han ones depicted ceremonies and
rites, as for example long elaborate processions carved and incised on stones from Yi'nan (fig. 6).29 In
this way the tomb owners could expect not just to follow a daily routine, but to mark out the seasons
of the year with the proper rites and ceremonies. Throughout the year, work in the fields was needed,
and in some tombs, most especially those in Sichuan, this seems to have been provided by bricks
moulded with scenes of farming activities.30
The ways in which such scenes and figures were understood and interpreted will be explained in
the next section as examples of xiang, that is, visible presentations of features of the invisible afterlife.
Such images were understood to bring to their owners and makers the effects of their counterparts in
the spirit world. In the worldly realm dancers were expected to be successful in bringing down the
spirits. What is more the figures were an effective means by which a desired effect, the presence of the
spirits, could be achieved. We can compare this phenomenon with a parallel one, the making of clay
dragons. As the sighting of dragons was linked with rain, when rain was desired, the effect, the
presence of the dragons, this time in clay, was expected to be correlated with, that is to produce,
rain.''

26 A survey of typical scenes on such tomb slabs is found in Finsterbusch. In describing some of the scenes and figure groups as
providing ceremonies and rituals, I am following Patricia Berger, "Rites and Festivities in the Art of Eastern Han China: Shandong
and Kiangsu Provinces," (unpublished Ph.D. dissertation, University of California, 1980). Wu Hung, "Beyond the 'Great
Boundary': Funerary Narrative in the Cangshan Tomb," in John Hay, ed., Boundaries in China (London: Reaktion Books, 1994),
81-Io4; and Lydia Thompson, "The Yi'nan Tomb: Narrative and Ritual in Pictorial Art of the Eastern Han (25-22o C.E.) China,"
(unpublished Ph.D. dissertation, Institute of Fine Arts, New York University, 1998), lay more stress on the depictions of elements
of the actual funeral rites.

27 Derk Bodde, Festivals in Classical China: New Year and Other Annual Observances During the Han Dynasty, 2o6 BC - AD 220
(Princeton and Hong Kong: Princeton University Press, the Chinese University of Hong Kong, 1975), 128. Bodde may have gone
too far in inserting the word "representations" as the ancient Chinese may have understood what we call representations as being
the deities themselves. For actual figures see Mawangdui report, vol. I, Ioo.
28 Patricia Berger, "Body Doubles: Sculpture for the Afterlife," Orientations 29, no. 2 (February 1998), 46-53. See also Michael Loewe,
"The Chideh-ti Games; a Re-enactment of the Battle between Ch'ih-yu and Hstian-ytian?" in Divination, Mythology and Monarchy in
Han China (Cambridge: Cambridge University Press, 1994), 236-48.
29 Yi'nan gu huaxiangshi mu fajue baogao (Nanjing: Wenhuabu Wenwu guanliju chubanshe, 1956).
30 Finsterbusch, pls. 3:10o, 5:2o.
3 Loewe, "The Cult of the Dragon and the Invocation for Rain," in Divination, I42-59, see p. I45.

12

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 The tomb builders were intent on creating more than simply daily life, they also provided a
cosmic setting.32 Most significant among the beings of the cosmos were the animals of the four
directions, the images of the sun and the moon and the creatures linked with them, the crow, the
toad and the hare, and the constellations. In later Han tombs, the Queen Mother of the West, Fuxi
and Niiwa, and the Herd Boy and the Weaver Girl added to the rich description of the universe.
Fragments surviving from one of the tombs at Yongcheng and from the tomb of the King of Nan
Yue suggest that scenes were painted on the walls of the kings' tombs. It is possible that the contents
of the paintings resembled the subjects and styles of the cosmological figures on the banner painting
from Mawangdui (fig. 7). Indeed, we might see the painting at Mawangdui as part of a sustained
attempt to reify the features of the universe for the benefit of a tomb owner.33
The generally later images on stone and brick are counterparts of such fine painting. Many Han
tombs at Luoyang, as well as a well-known example at Xi'an, have painted scenes of the heavens on
the ceilings.34 That this cosmos was provided for the benefit of the tomb occupant is evident from a
variety of tomb decorations that appear to show a figure, perhaps the tomb occupant, riding through
the skies in a chariot drawn by dragons, in the manner described in the Chuci.35 It seems that, as the
cosmos was so brought to the tomb owner, he or she did not have to leave the confines of the tomb to
partake of its realms.
In sum, the major rock-cut tombs, and to a lesser extent the smaller ones in stone and brick, were
complete settings for the afterlife. Each tomb was an entire universe centered on its occupant.36 He
or she was provided with all necessities, especially for feasting and sleeping, for excursions and for
music making, for the performance of ceremonies and rites. Such a scope had not been sought or
achieved, it would seem, in the Yellow River area before the Qin-Han period. These massive Han
tombs and their multifarious contents seem to imply that they were to be viewed from distant
regions of the universe, as though the spirits looked down upon a miniature world complete in all
parts. This sense that models were in some way, which we cannot now grasp, essential means of
controlling or even creating the universe is seen in the descriptions by Sima Qian of the models of
palaces that Qin Shihuang set up at Xianyang and the similar ones that were also put into his
tomb.37

Carefully prepared tombs indicate that there were participants in the afterlife so created. These
would, of course, have been the dead and their dead relatives, but also the deities, spirits and demons

32 Thompson, 163-84, discusses the columns within Eastern Han tombs in terms of an axis mundi. See also John Major, Heaven and
Earth in Early Han Thought (Albany: State University of New York, 1993), 47.
33 Interpretation of this painting has been a matter of considerable debate, see especially Michael Loewe, Ways to Paradise, The Chinese
Quest for Immortality (London: George Allen & Unwin, 1979); Anna Seidel, "Traces of Han Religion in Funeral Texts Found in
Tombs," in Akizuki Kan'ei, ed.,Dikyj to shukyo bunka (Tokyo: Hirakawa, 1987), 678-714; Wu Hung, "Art in a Ritual Context:
Rethinking Mawangdui," Early China 17, (1992), 11I-14.
34 See Huang Minglan and Guo Yinqiang, Luoyang Hanmu bihua (Beijing: Wenwu chubanshe, 1996); and Jiaotong daxue, Han
bihuamu (Xi'an: Xi'an Jiaotong daxue chubanshe, 1991).
35 See Lukas Nickel, "Changes in Mortuary Architecture and Painting in Northern Henan at the Time of Wang Mang," paper
presented at the School of Oriental and African Studies, London University, January 1997 (forthcoming); and Kaogu 1991.8, 713-2I,
768.
36 Compare the descriptions of the universes in David L. Hall and Roger T. Ames, Anticipating China, Thinking Through the Narratives
of Chinese and Western Culture (Albany: State University of New York Press, 1995).
37 Burton Watson, Records of the Grand Historian, Qin Dynasty (Hong Kong, New York: A Renditions-Columbia University Press
Book, 1993), 45, 63.

13

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
represented on the tomb walls, or, as we shall note, recorded in texts buried in the tombs, beings who
were, no doubt, intended to see the tomb and to interact with it and its inhabitants. We can also
infer that some of the spirits engendered fear in the builders of the tombs. This is clearly indicated by
the ceramic armies and the jade suits and weapons. Fear must also have determined the carefully
prepared system used to block up the access passages of the mountain tombs, a system that depended
on accurately cut blocks of stone assembled in groups of four or six, several deep, one behind another.
The tombs were to be cut off from the world of the living.38
The life presented in these tombs differs markedly from that on offer in earlier Yellow River
burials. What is more, many of the components, as I shall now discuss, came from areas outside the
Yellow River basin, brought together by the political unification of China.39

THE UNIVERSE AS DESCRIBED IN TEXTS

Tombs, often in southern China, have offered up textual evidence of an account of t

documents buried for the use of the tomb owners. The excavated writings on bambo
include some famous canonical texts, such as the Laozi and the Sunzi Bingfa, but mor
the current discussion are several different categories relevant to day-to-day activit
almanacs, medical texts, charms and incantations to deal with demons and baleful inf
and star charts. This evidence suggests that the tomb occupants or their assistants w
erate and were evidently committed to assistance from recipes and instructions in t
writing tools, such as ink stones, ink and brushes, were provided, indicating that th
were to take part in official duties in the afterlife.
In addition, several other types of texts were addressed to the bureaucracies of th
among them inventories of tomb contents, pleas on behalf of the tomb occupant
muwen, and the land contracts already mentioned. Such materials are typical of the
they seem to perpetuate a situation well developed already in the third century
cases, even earlier. For a document found at Fangmatan in Gansu province, an
Xueqin and by Donald Harper, gives an account of that underworld and its offices fr
one Dan, who came back to life after committing suicide.4? From this report we lear
world bureaucracy, staffed, at least in part, by individuals known to Dan's cont
records if not from personal experience. The belief system in which these texts were
different from what little we know about early Shang and Zhou religious practi
texts on omens, curses and medical recipes are often described as aspects of popu
however, clear that all levels of society subscribed to these beliefs.4'

38 A clear division between the tomb and the living is mentioned by Wu Hung in "Beyond the 'Great Boundary
39 Robert Bagley, "Changjiang Bronze and Shang Archaeology," International Colloquium on Chinese Art Histor
Antiquities, Part i, (Taipei: National Palace Museum, 199o0), 2o9-55; Jessica Rawson, "Overturning Assump
in Ancient China," Apollo I45, no. 42 (March 1997), 3-9.
40 Li Xueqin, "Fangmatan jian zhong de zhiguai gushi," Wenwu 1990.4, 43-47; Donald Harper, "Resurrectio
Popular Religion," Taoist Resources 5, no.2z (1994), 13-28.
4' The religion of the Han period has been relatively little studied, although a number of secondary w
intellectual or philosophical outlook that existed in parallel. The most recent treatment is provided by Mu-ch

14

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 This religious universe has been studied and described by a number of scholars in the west,
especially by Anna Seidel, Donald Harper and Mark Lewis.4" The salient feature of this work has
been to point out the ways in which the mystery and richness of the universe were created by
modelling all its aspects on the life of a court at the center of a great empire. Of course, the ancient
Chinese viewed this patterning the other way around. The very fact that the world had emperors and
bureaucrats, for example, they saw as expressing a pattern of relationships figured in the universe as a
whole. In other words, the living world was a microcosm of the universe.
This universe was seen to be under the direction of a celestial ruler, sometimes called the Yellow
Emperor, or Huangdi, sometimes the Yellow Spirit and sometimes the Celestial Deity, who had
forces, messengers and powers, equivalent to or surpassing those of his worldly counterparts.43 Access
to the spirit realms was from the peaks of great mountains, such as Taishan in the former state of Qi.
Near Taishan also was the region called Haoli, where entrance was gained to the underworld. Indeed,
it is evident that the mountain cults of the eastern Warring States kingdom of Qi contributed greatly
to these beliefs.44 The records by which the lives and times of death of individuals could be checked
were devised by the bureaucracy (which was parallelled by the human bureaucracy) of the Yellow
Emperor or the Celestial Deity and were transmitted to the underworld. As already mentioned, from
the Warring States period onwards, we have archaeological evidence, gained from tombs, of a con-
cern with all sorts of bureaucratic practices in the afterlife.45 Seals, characteristic of many Han tombs,
were perhaps intended to enable tomb occupants to identify themselves to the officers of the under-
world.

These views of the afterlife within the realms of the deities and controlled by a bureaucracy, were
contemporary with an interpretation of the universe described in correlative terms.46 Four and later
five gods, or Di, were thought to govern the cosmos. Such views were alleged by Sima Qian to go
back to seventh-century Qin practices and beliefs. Each of the Di came to be linked to a direction and
a color; the role of the Yellow Emperor stems in part from this source. However, in other ways
information about and worship of the Yellow Emperor came, it seems, from Qi. Although, when
mentioned in Han period texts, the Yellow Emperor is assumed to have a long lineage, he seems to

42 See Seidel, "Traces of Han Religion"; Mark Lewis, Sanctioned Violence in Early China (Albany: State University of New York Press,
199o0), Writing and Authority in Early China (Albany: State University of New York Press, 1999); and K.E. Brashier, "Han Thanato-
logy and the Division of 'Souls'," Early China 2I, (1996), I2-158. These works, in particular, have focussed attention on the role of
the celestial and underworld spirits, and have shown that suggestions of a paradise to which the dead hoped to go are, in general,
not present at this early date in the Han.
43 For evidence on the patterning of the spirit worlds on the living world, see Wu Rongzeng, "Zhenmuwen zhong suojiandaode
Dong Han daowu guanxi," Wenwu 1981.3, 56-63; Seidel, "Traces of Han Religion"; and Lewis, Sanctioned Violence, I70-212.
44 Although the subject has been little examined, it seems possible that the reverence for mountains in Qi may have come from much
further north, from the peaks of Siberia, where we know that the frozen tombs preserved the bodies of the dead buried there.
Stories of the "magically" preserved bodies might have reached northern China through long-distance rumors and reports. See also
Terry F. Kleeman, "Mountain Deities in China: The Domestication of the Mountain God and the Subjugation of the Margins,"
Journal of the American Oriental Society 114, no.z (1994), 228, where he discusses the account in the Zuo Zhuan which links the
marchmonts (the yue) or cosmological mountains to non-Chinese people. Later in the paper, I will mention other features of Qi
culture that suggest contact with the nomadic areas.
45 See Warring States tombs at Baoshan and Wangshan: Baoshan Chu mu, 2 vols. (Beijing: Wenwu chubanshe, 199I), hereafter
referred to as Baoshan report;Jiangling Wangshan Shazhong Chu mu (Beijing: Wenwu chubanshe, 1996), hereafter referred to as
Wangshan report; and Han period tombs in the same area, especially at Fenghuangshan, Kaogu xuebao I993.4, 455-513.
46 See also the discussion of Imperial cults in Michael Loewe, Chinese Ideas of Life and Death, Faith, Myth and Reason in the Han Period
(202 B.C - A.D. 220) (London: George Allen & Unwin Ltd, 1982), 127-43.

I5

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
have been added to the Chinese pantheon during the latter centuries B.C.47 To these views were
linked the philosophies of the Five Phases (wuxing) and of Yin and Yang. Omens were important
adjuncts to an understanding of these phases.48 For the heavens, the seasons, and the activities of
humans were bound together in a complex scheme or pattern.49
Indeed, the patterns, known as wen, were fundamental to the proper functioning of the universe.
All later views on this pattern, or wen, take their starting point from the 'Commentary on Attached
Verbalizations' of the YiJing (The Changes):

Looking up, we use it [The Changes] to observe the configurations (wen) of Heaven, and, looking down, we use
it to examine the patterns of the earth.5o

In this juxtaposition wen is the fundamental pattern that provides the template, so to speak for the
earth. The term was also widely used to describe "natural" patterns, such as the patterning of
animals.5I
The tombs and their contents were part of this pattern, and indeed were a way in which the
patterns of the invisible world were made visible. Within the universe they had the status of xiang.
As Willard Peterson has argued with reference to the Changes, the term xiang is fundamental to the
ways in which an understanding of the universe was structured. The term derives some of its con-
notations from its use in the "Commentary on the Attached Verbalizations" to the YiJing (just
mentioned). It has been described by Peterson as follows:

47 The Yellow Emperor is mentioned in the very beginning of the first chapter of the Shi ji. Ssu-ma Ch'ien, The Grand Scribe's Records,
Volume I, the Memoirs ofpre-Han China, ed., William H. Nienhauser (Taipei: SMC Publishing INC, 1994), 1. See also Lewis,
Sanctioned Violence, 168, 187, zoI, 308 n. 6o; and Mark Csikszentmihalyi, "Emulating the Yellow Emperor: The Theory and Practice
of Huanglao, 180-144 B.C.E.," (unpublished Ph.D. dissertation, Stanford University, 1994), Chapter 2.
48 Mu-chou Poo, 44-48; Michael Loewe, "The Oracles of the Clouds and the Winds," in Loewe, Divination, 191-213.
49 For a general outline of this philosophical development, see Denis Twitchett and Michael Loewe, ed., The Cambridge History of
China, Volume I, The Ch'in and Han Empires 221B.C.-A.D. 220 (Cambridge: Cambridge University Press, 1986), 661-65, 69o0-92.
See also, A. C. Graham, Disputers of the Tao, Philosophical Argument in Ancient China (La Salle, Illinois: Open Court, 1989), part IV.
It is evident that there was a considerable disjunction between the principles about conduct of burials and the actual practice.
Despite injunctions to create less lavish tombs, there seems to have been little move in this direction, see Jeffrey Riegl, "Do Not
Serve the Dead as You Serve the Living: the Liishi chunqiu Treatises on Moderation in Burial," Early China zo20 (1995), 301i-30. It
may be necessary, indeed, to undertake a complete reassessment of the separate strands of early Chinese intellectual life if we are to
understand fully these differences. It is also noteworthy that the Han burials do not seem to reveal changes synchronized with
changes in Imperial Cults documented by Michael Loewe, "Changes in Qin and Han China: The Religious and Intellectual
Background," Studies in Chinese History, No. 4 (October 1994), 7-45.
50 After Richard John Lynn, The Classic of Changes, A New Translation of the I Ching as Interpreted by Wang Bi (New York: Columbia
University Press, 1994), 51. The power of these lines in the Changes is seen in their quotation by Xiao Tong in the preface to the Wen
xuan:

The Changes says, "Observe the patterns (wen) of the sky,

To ascertain the seasonal changes.
Observe the patterns (wen) of man,
To transform the world."
After David Knechtges, trans., Wen Xuan or Selections of Refined Literature, volume one, Rhapsodies on Met
Xiao Tong (Princeton; Princeton University Press, 1982), 73.
5 The term wen has been remarkably little discussed in secondary literature in European languages and is als
in Chinese. For some relevant comments, see Tse-tsung Chow, "Ancient Chinese Views on Literature,
lationship," Chinese Literature: Essays, Articles and Reviews I (1979), 3-29; William G. Boltz, The Origin and E
Chinese Writing System (New Haven: American Oriental Society, 1994), 134-38; see also Lothar von Falk
Arthur Waley in discussing wen as an epithet for ancestor qualities distinct from its meaning as pattern,
the Ancient Chinese Ancestral Cult," Chinese Literature: Essays, Articles, Reviews 18 (1996), 1-22zz. I am indeb
drawing my attention to an article in press by Martin Kern, "Ritual, Music, and the Written Text: Histori
Early China," forthcoming. For the comments on animal patterns, see Lynn, 77-

16

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 Each of the sixty-four hexagrams has a name, most of which are words or terms referring to particular objects
and activities which are involved in "figuring" (hsiang) the situation revealed by the act of divination. The
word hsiang, as used in the "Commentary," is sometimes rendered into English as "image," which connotes
resemblance and implies an act of perception. Hsiang often is the object of the verb "to observe" (kuan), which
supports translating hsiang as image. However, hsiang are independent of any human observer; they are "out
there," whether or not we look. Therefore, I find that the English word "figure" comes closer to covering the
meanings of hsiang in the "Commentary." A figure is an image or likeness, but it is also a form or shape, a
design or configuration or pattern, and a written symbol; "to figure" is to represent as a symbol or image, but
also to give or to bring into shape. 52

Sima Qian, recounting the attitudes of Qin Shi Huangdi (reigned 221-10 B.C.) and of Han Wudi
(reigned 141-87 B.C.) to the universe, and above all to the spirits, gives examples of the ways in which
the term xiang was understood at that time. We know from his description of the hopes and fears of
the First Emperor and of Wudi in the Shi ji, that the peoples of the third and second centuries B.C.
expected that images or figures, xiang, of spirits would bring them, the spirits, to visit the owners of
the images.53

Shao-weng then said to the emperor, "I perceive that Your Majesty wishes to commune with the spirits. But
unless your palaces and robes are patterned (xiang) after the shapes of the spirits, they will not consent to come
to you." He fashioned five chariots symbolizing the five elements and painted them with cloud designs
(yunqi), and on the days when each element was in ascendancy, he would mount the appropriate chariot and
ride about, driving away evil demons. He also directed the emperor to build the Palace of Sweet Springs, in
which was a terrace chamber painted with pictures of Heaven, Earth, the Great Unity, and all the other gods
and spirits. Here Shao-weng set forth sacrificial vessels in an effort to summon the spirits of Heaven.54
[Words in italics added by the author.]

We see here several levels of images or xiang. The images of clouds, of pictures of Heaven, Earth and
Great Unity, were pictures in the sense in which we use the word "picture" in the West, but xiang in
the terms of Sima Qian and his times: that is, they were equivalent and equal to the clouds, Heavens,
and the Great Unity themselves. The chariot also, and the palace itself were also xiang. That is, their
forms correlated with eternal features of the universe. A third type of xiang was realized when very
fine small bronze chariots were made, decorated with cloud patterns, for burial near the tomb of the
First Emperor."55 These were half-life size, as we have already noted. Such chariots and, indeed, the
models of palaces, also mentioned, illustrate for us the ways in which the features of the universe
could be made visible at many levels and on many scales. The tombs were, therefore, further ex-
amples of xiang, partaking also of the eternal features of the universe.
This concept of xiang may help with the difficult issue of where the dead were thought to reside.
The tomb was literally a dwelling created by the people of the world, but once the dead were shut
away inside it, the tomb and the tomb owner seem to have become separated from the world. If the

52 Willard J. Peterson, "Making Connections: 'Commentary on the Attached Verbalizations' of the Book of Changes," HarvardJournal
of Asiatic Studies 42, no. I (1982), 80-81.
53 For a discussion of the use of correlation or analogy in ancient Chinese thought, see David L. Hall and Roger T. Ames, Anticipating
China, compare Graham, Disputers of the Tao.
54 Burton Watson, trans., Records of the Grand Historian of China, translatedfrom the Shi ji of Ssu-ma Ch'ien (New York: Columbia
University Press, I96I), vol. 2, 42; Shi ji (Beijing: Zhonghua shuju, 1959), I388.
55 For an account of highly decorated chariots in ritual, see Kiyohiko Munakata, ed., Sacred Mountains in Chinese Art(Urbana-
Champaign: University of Illinois Press, 1991), 23-24. The fine decoration of the bronze chariots of Qin Shi Huangdi may imply
that they too had a similar function. See note 23 for references.

17

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
tomb is understood as having the characteristics of a xiang, then the tomb was in effect a dwelling in
the world of spirits.
This subject is complicated by discussions on the nature of the soul. It is likely that the notions of
the soul varied as between members of different social classes, and between peoples in different
regions. Further, any single individual might hold different views in different contexts. Hence the
notion of where the ancestors were when receiving offerings might be distinct from their location
when buried. Some scholars have suggested that the ancient Chinese conceived of a person as having
two souls: one soul, the po, remained in the tomb, while the hun, ascended to paradise. 56 This concept
of the two souls has now been shown to have been complex and to have varied over time.57" From the
perspective of preparation for burial it seems that in this context the person was deemed to live in the
tomb, but the tomb as a xiang was, as already mentioned, as much part of the invisible spirit world as
of the living world that had first built it. The tomb occupant's journeys among the realms of deities
and spirits might be made as visits, rather than as part of an ultimate progression to a paradise.
Notions of ascent to paradise developed later.

UNIFICATION OF THE WORLD OF SPIRITS

The single condition fundamental to this discussion is that when the Qin and the H
unified and ruled their territories, they claimed to embrace the whole known universe, a
Lewis has noted, to include the spirit, demons and deities of that universe.58 Indeed as an
part of the universe, it was not surprising that the emperors sought access to and control
to the spirits and deities. Without such an enterprise it is unlikely that the accumulation o
ideas from many diverse regions would have been drawn into the center of burial practic
the homogeneity we see in the contents of Han tombs have been achieved without a shar
and a shared series of workshop practices to create objects to present this ideology. The e
tions have drawn attention to both the large rock-cut tombs and their lesser imitations a
jade suits as examples of such ideological and practical unity.
The most conspicuous evidence of the First Emperor's ambition is his choice of
Huangdi. In making his claim to a title equivalent to those of the great Di, King Zhen
in effect setting himself alongside them, not only in the human world, but also in the wh
The concerns of the First Emperor, and indeed Wudi, went much further than simply
claims. They were obsessed by spirits and by efforts to harness their powers. Sima Qian's
the life of the First Emperor and his discussion of the Feng and Shan sacrifices give
amount of information on the beliefs of the day and on the ambitions of the two emper
the spirit world.

When the First Emperor united the world, he instructed the officials in charge of sacrifices to pu
the worship of Heaven and Earth, the famous mountains, the great rivers, and the other spir

6 See Yii Ying-shih, "'O Soul, Come Back!' A Study in the Changing Conceptions of the Soul and Afterlife in Pre-
Harvard journal of Asiatic Studies 47, no. 2 (1987), 363-95; and Loewe, Ways to Paradise.
57 Brashier, "Han Thanatology."
58 Mark Lewis, "The Feng and Shan Sacrifices of Emperor Wu of the Han," in J. McDermott, ed., Court and State Ritua

I8

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 customarily been honored in the past. According to this new arrangement, there were five mountains and two
rivers east of Yao designated for sacrifice.59

Amongst the many examples of the ways in which the Qin and the Han incorporated the worship of
deities venerated in different regions into Imperial cults are the worship of the Eight Spirits, all
associated with mountains in Qi, and the offerings to Taiyi, or Great Unity, a deity whose location
was sited in the stars and whose worship may have been connected with a spirit known from the state
of Chu as powerful in combat.6o Far from being a spirit with specific powers, when described by
Sima Qian, Taiyi gained a lofty role. Sima Qian recounts the words of one Miu Ji, when instructing
Wudi on how to sacrifice to the Great Unity:

"The Great Unity," he explained, "is the most honored of the spirits of Heaven and his helpers are the Five
Emperors. In ancient times the Son of Heaven sacrificed to the Great Unity each spring and autumn in the
southeastern suburbs, offering one set of sacrificial animals each day for seven days. An altar was constructed
for the purpose which was open to the spirit roads of the eight directions."6I

The two monarchs were concerned not only with proper attention to cults and to buildings, but also
with their own mortality. The First Emperor and Han Wudi sought special advisors (fang shi) who
might help them make contact with the spirits and to find drugs that would ensure non-death. The
termfang shi to describe advisors and practitioners referred to their knowledge, especially of recipes,
for the ancient Chinese seem to have laid great stress on the correct medicines6z and potions to
extend their lives, as seen in the attempts by the First Emperor and Han Wudi to acquire elixirs that
would grant them non-death.63 In other words, the Qin and Han sought the advice of individuals
who might have gained esoteric knowledge, some of which may have been thought to derive spirits
and deities from such places as the islands of Penglai. Other remote regions may equally have been
thought to be important sources of magical items, as we shall see.

THE CHINESE WORLD

Although I shall not discuss further the contribution of the Yellow River area to
ritual preoccupations of the Qin and the Han, the practices and beliefs of this area,
center of Chinese-speaking peoples, underlay all their activities. From the Sha
heritance stemmed the belief that the afterlife resembled that of the human world an

proper preparation of the tomb was essential to the well-being of the tomb occ
descendants. Peoples in the Yellow River area developed the practice of including
vessels in their tombs. What is surprising is that by the time of the Qin conqu
vessels for offering sustenance in the form of food and wine to the ancestors were
and distinct types made of bronze may have receded in importance. Western Han tex

59 Watson, Records, vol. 2, 28; Shiji, I37I.

6o See Li Ling, "An Archaeological Study of Taiyi (Grand One) Worship," Early Medieval China z (1995
Harper, "Warring States National Philosophy and Occult Thought," in Michael Loewe and Edward Shaugh
History of Ancient China (Cambridge: Cambridge University Press, 1999), 813-884.
6 Watson, Records, vol. 2, 40; Shi ji, I386.
62 Illness was thought to be caused by evil influences especially demons. Thus, medication was closely linked to
of spirits. Mu-chou Poo, Chapters 3 and 4. See also Donald Harper, "A Chinese Demonography of the
HarvardJournal of Asiatic Studies 43 (1985) 459-98.

63 Li Ling, Zhongguo fang shu kao (Beijing: Renmin Zhongguo chubanshe, 1993).

19

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
descriptions of the contemporary rites for venerating or remembering the ancestors,64 and it seems
that during the third century B.C. new practices and beliefs from regions both to the south and to the
north of the Yellow River area had already infiltrated and greatly altered the traditional ceremonies,
perhaps changing them out of all recognition. This is a phenomenon that has passed almost without
comment or explanation in the literature.65 Some discussion of this change will be made later in the
paper.
Before I consider the contribution of three major areas, namely the south, especially Chu, the east,
namely Qi and Lu, and the northern borders in present-day Inner Mongolia and the steppes, to tombs
structures and their contents, I will briefly outline the reasons behind the Qin and Han predilection
for Chu practices. Their interest in Chu had quite simple origins. Qin's fourth- and third-century
B.C. military campaigns culminated in conquest of Chu areas in present-day Hubei and Hunan
provinces. Chu was driven eastwards and Qin assumed much of the territory; in so doing Qin took
over some Chu customs and objects. With Chu now established further east, people in parts of
present-day Anhui and Jiangsu provinces also took on many aspects of Chu culture. These eastern
Chu practices became integral to the Han-period Chu state in this same area. The founder of the Han
dynasty, Liu Bang, came from present-day Xuzhou in Jiangsu province, in what was to be the Han
kingdom of Chu.66 Hence Liu Bang brought to the court at Xi'an customs of the south, and also
many eastern practices and beliefs. Chu was influential in other ways too. Several Han monarchs
patronized Chu poetry, and poetic forms, similar to those in the Chuci, came to have a considerable
influence through later poetry known as fU.67 Chu was not the only source of southern literary and
intellectual contributions to Han culture. The great exponent offu, Sima Xiangru, himself, came
from the southwest, while the philosophical text, the Huainanzi, also provides insight into the
southern contribution to the thought and beliefs of the early Han period.
What now follows shows the ways in which the concepts that came from within the landmass we
today designate as China were given visible form as the Qin and the Han appropriated objects,
categories and motifs from several parts of East Asia. I will discuss in turn the acquisition of items
from the borders and beyond, from the state of Chu and from the east coast.

64 Such studies as there have been have revealed that much of the ritual activity took place at the tomb, in shrines adjacent to the
tumulus; see Michael Loewe, "The Imperial Tombs." The Han, of course, attempted to describe Zhou ritual in the classic ritual
texts. Moreover, later Han texts discuss the use of tablets in shrines, see Mansvelt Beck, Io5-o8.
65 Detailed studies of bronzes of course note the disappearance of traditional ritual vessels and the introduction of the specifically Qin-

type garlic-headed hu and cauldron, mou. See especially Hayashi Minao, Shunj5 Sengaku jidai seidoki no kenky5 (In Sh5 seidki suran
san). (Toky6: Yoshikawa Kobunkan, 1989). However, the possibility that a complete change in religious orientation was
responsible for such significant changes has not been fully examined. It seems that in studies of the Han period, reverence of deities
has been taken as an independent phenomenon that did not relate to anything in the earlier ritual offerings. However, it is possible
that the two phenomena, decline in ancestor rituals and an interest in such figures as the Yellow Emperor and the Queen Mother of
the West, were connected. It seems likely that ritual offerings continued to be made, but in much more every-day vessels (see fig.
Id).
66 For a general account of archaeological finds and textual references to Xuzhou in the Han period, see Wang Zhongwen, ed., Liang
Han wenhua yanjiu (Beijing: Wenhua yishu chubanshe, I996).
67 David Knechtges, "The Emperor and Literature: Emperor Wu of the Han," in Frederick P. Brandauer and Chun-Chieh Huang,
ed., Imperial Rulership and Cultural Change in Traditional China (Seattle and London: University of Washington Press, 1994), 51-76.

20

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 MOUNTAIN TOMBS AND CONTACT WITH THE BORDERLANDS

It may seem surprising to start with an account of the ways in which object t
motifs from the borders infiltrated the Han world. But the wide variety of
introduced and integrated into Chinese material culture at this date is eviden
enterprise to derive all possible gain from as many sources as possible. Some of
had had extensive contacts with the peoples of the borders long before the unifica
summarize the major changes that will be discussed, we can say that some Western
gold may have stimulated the use of gold and silver for ornaments and utensils. El
paraphernalia of mirrors, flasks, braziers, shovels, ladles and incense burners may
north or northwest, as perhaps also did tent-like elaborate canopies (such as
Sheng's tomb [fig. z]), and screens. Finally, many figures of animals, most espe
upon the lion, animals in combat, and griffins, were borrowed from quite dif
Mongolia, Siberia and Western Asia. As the foreign items were borrowed, they
and, in due course, accommodated within traditional Chinese categories, som
made in jade, that most Chinese of materials, as will be discussed in the sect
contributions to Han tombs.

Some of the early contacts between the Chinese-speaking areas and lands further west are best
traced through the use of gold. During the first thousand years of the bronze-using centuries, the
inhabitants of the major ancient Chinese centers had rarely exploited gold, concentrating on copper
and tin alloys to make ritual vessels, chariot parts and of course weapons.68 However, from the time
of the fall of the Zhou capital at Xi'an in 77I B.C., gold came to be used for weapons and horse
fittings, and for very thin decorative sheets in Qin, Jin and even further south in Zeng and Chu.
Rather surprisingly, some of the earliest use of gold is found in Shandong, at Liujiadianzi in
Yishui. 69By the sixth century B.C., warriors in the Qin state owned iron weapons with exceptionally
fine gold hilts, while the bronze casters of the Jin foundry at Houma had, at much the same time,
incorporated features of Iranian, Scythian and Siberian gold work, in the form of winged griffins, lion
heads modified as tiger or taotie heads, and tigers or other felines or griffin-like birds attacking or
devouring other creatures.70 Gold is a persistent element in sixth- to fifth-century B.C. southern
tombs, especially as sheet appliques found in Henan and Hubei.71
It seems that gold had entered the Chinese-speaking world by several routes; a direct crossing of
the borders in the north and west was probably the primary one. A route by way of the northeast also

68 Robert Bagley has drawn attention to the contrasts between the metal-working traditions of Western Asia and those of China.
Robert Bagley, Shang Ritual Bronzes in the Arthur M. Sackler Collections, Volume I (Cambridge MA: The Arthur M. Sackler Founda-
tion, and the Arthur M. Sackler Museum, 1987), 15-I7. For a view on the way in which gold was introduced to China as part of a
complex interaction between the border peoples and the ancient Chinese states, see Esther Jacobson, "Beyond the Frontier, A Re-
consideration of Cultural Interchange Between China and the Early Nomads," Early China 13 (1988), zoI-4o.
69 Noted by Jenny So, Eastern Zhou Ritual Bronzes from the Arthur M Sackler Collections, Volume III (Washington, D.C.: Arthur M.
Sackler Foundation and the Arthur M. Sackler Gallery, Smithsonian Institution, I995), I9-20.
70 See Houma zhutong yizhi, 2 vols. (Beijing: Wenwu chubanshe, I993), hereafter referred to as Houma report. The moulds show many
examples of birds of prey, griffins and feline faces, tiger-like or taotie-like in character, developed on the basis of lion faces. Many
mould pieces show tiger stripes. The bronze casting moulds derive some of their detail from work in gold. For discussion of these
two points, see Jessica Rawson, Chinesejadefrom the Neolithic to the Qing (London: British Museum Press, 1995), 60-69.
71 Use of thin gold appliques found in tombs of the Spring and Autumn period may have been transmitted by way of Siberia or
Central Asia from Iran or the Caucasus. For such appliques in China, see Xichuan Xiasi Chun Qiu Chu mu, (Beijing: Wenwu chu-
banshe, 1991), pls. 73-76; Zeng Hou Yi report, vol. 2, pl. I49; and Kaogu xuebao I988.4, 455-502, figs. 11-15.

21

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
seems likely in the light of many traits found in Shandong that suggest contact with south Siberia
and even with the Caucasus. But Western Asian tastes were, perhaps, also carried down the great
tributaries of the Yangzi River from Shaanxi through Sichuan to Hubei, and yet further south to the
kingdom of Dian in Yunnan.72
Other features that may have penetrated several regions of China along these routes include a
respect of deities or spirits, represented as crowned with antlers, mirrors, and incense or perfume
burning. In addition, we can see that the links with the metal-working traditions of the Near East
strengthened over a century or more in objects from the tombs of the rulers of the Kingdom of
Zhongshan at Pingshan in Hebei province in the late fourth century B.C.73 Among their possessions
were inlaid bronze winged beasts (fig. 8a) and a magnificent inlaid bronze casting ofa tiger attacking
deer (fig. 8b), a local version of the Western Asian animals-in-combat theme, bronze standards or
emblems and fittings for tents or screen (fig. 8c). Objects in the Zhongshan tombs and, indeed, in
many fourth- to third-century burials, included gold and gilded bronze, and pieces inlaid with
turquoise, all reminiscent of the fine metal work of Siberia, Bactria and the Caucasus.74 During the
Han, the most sumptuous bronze vessels were made of gilded and inlaid bronze, with shining sur-
faces that were much more typical of Western Asia than of the principal ancient Chinese traditions.
It is difficult to reconstruct how the peoples of the pre-Han and Han period understood the re-
gions from which these marvelous objects and creatures were borrowed. We know that access to such
regions was prized. Several scholars have suggested that Wudi's campaigns in and embassies to the
west were stimulated by a search for the lands where the treasures of these strange regions were to be
found.75 Zheng Yan has shown how non-Chinese peoples who inhabited these regions seem, from
references in poems and carved figures in stone, to have been included among the auspicious creatures
of the universe whose sighting was thought to bring good fortune.76 The most famous examples are
colossal bronze figures cast on the orders of Qin Shi Huangdi and recorded in the Shi ji. Other ex-
amples are in stone, including a sculpture from the Maoling, the tomb of the general Huo Qubing,
showing a horse trampling a man and a very unusual stone carving of the Han period, now in
Ji'nan.77 The figure has staring eyes and a long elongated face that may indicate a beard. Similar
figures occur as capitals on some brick and stone columns in Han period tombs.78 The use of stone
was another innovation that possibly entered China from the north and west.
If we turn to less exotic figures, the terracotta warriors, themselves, seem to have been thought to
include recruits from the borders, as many sport moustaches and beards and are very tall (fig. 23). To
fight the ghost enemies of the afterlife, perhaps the First Emperor sought individuals who might

72 Although the phenomenon has not been fully examined, this route should be further considered. A related example is found in the
ways in which the bronze castings of animals made by the Dian peoples in present-day Yunnan included subjects, most particu-
larly animals in combat, that came ultimately from the northwestern borderlands.
73 For all references here and later in the paper to the tombs of the Zhongshan kingdom, see Cuomu Zhanguo Zhongshanguo guowang zhi
mu, 2 vols. (Beijing: Wenwu chubanshe, 1995), hereafter referred to as Zhongshan report.
74 Zhongshan report, color pls. I7-I8.
75 Wu Hung, Monumentality, 126-42. Wu Hung treats the source of some of the features described as being derived from Buddhist
monuments in India. On the present evidence, this source seems unlikely; Siberia and Western Asia, having been the source of
many of the animal features discussed, were perhaps also the sources of the interest in stone and mountain tombs.
76 Zheng Yan, "Barbarian Images in Han Period Art," Orientations 29, no. 6 (June 1998), 50-59.
77 Watson, Qin Dynasty, 45. A fine and strange figure ofa non-Chinese individual is in the Museum of Stone Sculpture at Ji'nan, see
Zheng Yan, fig. I. Others are published in Wang Luyu, ed., Zhongguo diaosu shice. Han Jin Nan Beichao mu qian shidiao yishu, vol. 3
(Beijing: Guangbo xueyuan chubanshe, 1992).
78 Aileen Lau, ed., The Spirit of the Han, Ceramics for the After-Life (Singapore: Sun Tree Publishing Ltd., 1991), no. 7.

22

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
already have experiences of the spirit world. If the peoples of the borders had been regarded solely as
enemies it seems unlikely that they would have been included in the terracotta army, or incorporated
within tombs.79 Moreover, as Terry Kleeman has pointed out, some texts, including the Zuo zhuan,
linked yue (often translated as marchmonts) or sacred peaks with the Rong or other non-Chinese
peoples, in the sense of describing the Rong as descendants of the mountains.
The abodes of spirits were, indeed, located on the margins of the known Chinese-speaking

world.8? In the late Warring States, the Qin and the Han periods, the Queen Mother of the West was
thought to live somewhere in this north or even northeasterly direction.8 The marvelous islands of
Penglai, where dwelt immortals who had secret recipes for medicine to ensure non-death, were said
to be in the eastern sea.8 From the south came accounts of the Kunlun mountains, described as the
abodes of spirits: these mountains are mentioned in the famous southern poems of the Chuci and the
pseudo-geographical text, the Shanhaijing. 83Such attitudes described a universe seen from the center,
surrounded on all sides by liminal regions that gave access to miraculous lands. Some of the in-
habitants of these strange places are described in the Huainanzi.4 This picture is matched in the
Zhao Hun (the Summoning of the Soul) in the Chuci. 85The soul is warned to avoid all distant regions
where frightening supernatural creatures lived. Those of the south are depicted in the famous Chu
Silk Manuscript.86 Those of the north survive in the sculptures already described and in gilded and
inlaid bronze fittings and even in jade, as we shall see.
The imagined realms were to be realized in some measure in the mountains chosen for tomb sites;
it, indeed, seems likely that the great tombs hollowed out of the rock were part of an enterprise to
gain access to the spirits. I put forward this suggestion on the basis of the concept xiang, described in
the previous section on attitudes to the universe. To have made them at all, the Han period Chinese
must have thought that these tombs were appropriate in terms of their positions, forms and contents
and presumably, also, they viewed them as correlated with similar features of the universe. In
addition, a palace in a mountain might find itself literally next to the dwellings of powerful deities.
Support for this suggestion is found in a much-quoted passage in the Shi ji, indicating that the
Queen Mother of the West was believed to reside in a great cave-like palace, perhaps like the tombs
at Xuzhou or Yongcheng. The fu on The Mighty One by the court poet Sima Xiangru describes

79 See also mention of a barbarian in an inscription within a tomb at Cangshan in Shandong that describes the decoration of the stone
slabs. Wu Hung, "Beyond the 'Great Boundary'", 93.
8o Zheng Yan suggests that the lands of the barbarians were thought to be located between the known world and the spirit realms,
"Barbarian Images," 53.
8 For a discussion of early tales about Xiwangmu, see Thomas E. Smith, "Ritual and the Shaping of Narrative: The Legend of the
Han Emperor Wu," (unpublished Ph.D. dissertation, University of Michigan, 1992), 54-69. See also Riccardo Fracasso, "Holy
Mothers of Ancient China, A New Approach to the Hsi-wang-mu Problem," T'oung Pao 74 (1988), 1-46.
82 Described in the Shi ji in connection with the First Emperor's search for non-death. Watson, Qin Dynasty, 53, 56-57.
83 Munakata, 12-2o. Munakata suggests that the animal in combat scenes shown in Chu metalwork and later lacquer painting were
derived from the southern region. This, however, seems unlikely. It may be the case that motifs and objects from the steppe
entered China by way of the north and east and also from the west, reaching both Qi and Chu. The Qin and Han borrowed from
Chu the lacquer patterns that formed the basis for the scrolls of qi, also interpreted or read as mountains. The question of the
significance of the scrolls has not been definitely resolved.
84 John S. Major, Heaven and Earth in Early Han Thought, Chapters Three, Four, and Five of the Huainanzi (Albany: State University of
New York, I993), 141-216.
85 David Hawkes, Ch'u Tz'u. The Songs of the South (Harmondsworth: Penguin Books, 1985), 219-31.
86For the Chu Silk Manuscript, see Noel Barnard, ed., Early Chinese Art and its Possible Influence in the Pacific Basin, vol I, Chu and the
Silk Manuscript(New York: Intercultural Arts Press, 1972); Li Ling, Fang shu kao, Chapter 3; Jao Tsung-yi and Zeng Xiantong,
Chu bo shu (Hong Kong: Zhonghua shuju, 1985).

23

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
journeys to the spirit worlds as crossing mountains, their peaks and their valleys, to find the Queen
Mother of the West:87

"Behold!" cries the Mighty One, "the Queen Mother of the West,
With her hair of silvery white
And her burden of hairpins, living in a cave!

Fortunately she has her three-legged crow to bring her food."88

Perhaps for this reason, palaces could be built on mountains and incorporate grottoes. In thefu on the
Shanglin Park, Sima Xiangru writes:

Here the peaks have been levelled for mountain halls,

Terraces raised, storey upon storey,
And chambers built in the deep grottoes.
Peering down into the caves, one cannot spy their end;
Gazing up at the rafters, one seems to see them brush the heavens;

The rock tombs of the Liu family princes seem to be such palaces, tunneled into the mountains,
creating new grottoes in which they would live eternally, like the Queen Mother of the West. This
extraordinary enterprise of creating palaces in mountains was a complete departure from the burial
practices of the loess regions of China, however. It seems possible that the Chinese had heard stories
about tombs in Siberia, Iran and even further west.90 In Siberia, most especially in the tombs at
Pazyryk in the Altai, mummified bodies, their clothes and wooden carts have been discovered in an
extraordinary state of preservation, a consequence of the intense perma-frost below the mounds or
kurgans over the tombs. These mounds were similar to the kurgans of the Scythians in the Cauca-
sus.9x Had any such tombs ever been reported, the Chinese may have believed that the peoples of the
north had exceptional means to preserve the body from attack by baleful influences. As some scholars
have argued, mounds over tombs had been used in several parts of the Chinese landmass before the
Eastern Zhou.92 However, from the fifth century B.C. many more appeared over the great tombs in
Qi and Chu for instances. Robert Thorp argues that this tradition stems from the choice of mounds
in the northern state of Yan, borrowed perhaps from the borderlands.93
My suggestion that both stone sculptures and stone tombs were derived from Western Asia via
intermediaries in Central Asia or Siberia may seem less far-fetched if we consider certain detailed
features of them. The stone creatures are especially interesting for the large eyes and wings that can
be matched on Western Asian figures, while tombs include characteristics, such as columns and

87 In the first part of the Han, and the pre-Han period, the site of the residence of the Queen Mother of the West was not thought to
be the Kunlun mountains, but a region to the north of the main Chinese area. Indeed, interest in Kunlun was associated with Chu
and in the Queen Mother with Qi. Smith, 54-64.
88 Watson, Records, vol. 2, 335; Shi ji, 306o.

89 Watson, Records, vol. 2, 312; Shi ji, 3o26.

90 It is even possible that references to tombs being opened and found with the bodies intact within them may refer to either Siberia,
where frost preserved organic material, or present-day Central Asia where the dry heat had the same effect.
91 For a connection between mountains and the liminal areas in the writings of the period, see Kleeman, "Mountain Deities."
92 Han Guohe, "Lun Zhongguo gudai fenqiumude chansheng yu fazhan," Wenbo 1998.2, 32-41, 45.
93 Robert L. Thorp, "The Mortuary Art and Architecture of Early Imperial China," (unpublished Ph.D. dissertation, University of
Kansas 1979), 42-46.

24

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
capitals, that can equally be compared with Western Asian examples.94 For instance, in the east of
China, in Shandong and in Jiangsu, tombs show some very distinctive features: faceted columns with
several types of capitals, which occur in several tombs, may have come from Iran or the Caucasus (fig.
9).95 Occasionally such columns stand on the backs of animals, including rams (4b). This convention
is well known from Western Asia, from Persepolis for example. Some Chinese stone tombs have
stepped roofs, again resembling the structure of tombs in the Caucasus and caves in Western Asia,
especially Iran.96 Such features in late Western and Eastern Han tombs suggest that contact between
China and Western Asia was frequent and persistent. It, therefore, seems reasonable to suggest that
interchanges of this nature endured over a long of time.
Indeed, other traces of direct copying are very intriguing. A much earlier tomb in the Qi state in
eastern China, often treated as the tomb ofJing Gong (547-49o0 B.C.), is especially perplexing but
also especially suggestive. The main coffin pit lies at the center of an almost square tomb. This pit
was lined with boulders, a feature of Scythian and Siberian tombs. It was surrounded on its sides by
pits in which more than six hundred horses were neatly laid out (fig. IO), in the manner of some
famous Scythian tombs, such as that at Kostromskaja in the Caucasus (fig. II).97 While at first sight
it may seem somewhat implausible that this unusual tomb owed anything to an area as distant as the
Caucasus, other tombs in Shandong in which the central burial is surrounded by subordinate human
burials are also somewhat similar to the tomb patterns in the Caucasus.98 These particular burials are
not the ancestors of the cave-like tombs, but they appear to show that peoples from the borderlands
may have penetrated far into the heartlands of Qi. These traces provide further support for the view
that tomb shapes, in general, and rock-cut tombs, in particular, were perhaps created as a result of a
continuous trickle of information about tombs in the north and the northwest that penetrated
Shandong and later Jiangsu over several hundred years, from the Warring States period to the
Eastern Han.

The occupants of the unusual square tombs in Shandong carried distinctive pendants of crystal,
agate and amethyst, but not of jade. A persistent shape is a small tapered horn-like object, blunt and
unfinished at one end with a small branch. Such pendants are circular or roughly oval in cross section
and not flattened like the usual jade ones. It is possible that horn shapes were copied from Siberian or

94 Chuan Tianchou, Zhongguo meishu quanji, diaosu bian, 2, Qin Han diaosu (Beijing: Renmin meishu chubanshe, I985), nos. 87-88, 93.
It is possible that sculptures of sheep, especially of rams, were also derived originally from Western Asia, see ibid, no. 86. The
history of such sculptures in China is discussed in Ann Paludan, The Chinese Spirit Road, The Classical Tradition of Stone Tomb
Statuary (New Haven & London: Yale University Press, 199I). It is also likely that in China this possibly newly introduced
sculpture type was favored because of the relationship between the graphs for yang (ram) and xiang (auspicious).
95 For other categories of capital, see Wang Luyu, pls. 20o-23.
96 See Ellis H. Minns, Scythians and Greeks. A Survey of Ancient History and Archaeology on the North Coast of the Euxine from the Danube
to the Caucusus (Cambridge: Cambridge University Press, 1913), 196.
97 The Qi state tomb is very inadequately reported and the arrangement of the pits with horses not well explained, see Wenwu 1984.9,
I4-19. For tombs in the Caucasus, see Minns, zz224-27. For a discussion of the kurgans in the Caucasus, see M.I. Artamonov,
Treasures from the Scythians Tombs in the Heritage Museum, Leningrad, Tamara Talbot Rice, trans., (London: Thames and Hudson,
1969).

98 For
73-Io4three remarkable
(Langjiazhuang, burials
Linzi); and Linyi at Linzi
Fenghuangling andmuat(Ji'nan:
Dong Zhou LinyiQi Luin Shandong,
Shushe, 1987). see Wenwu 1984.9, I4-I9 (Heyaitou, Linzi); Kaogu xuebao I977.I,

25

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
Central Asian cheek pieces or harness ornaments (fig. 12).99 A common, pointed flat jade pendant
seems to be a sinicised version of these exotic crystal or agate pieces. Early examples have come from
the late Qi state in Shandong, from Shangwang at Linzi, and later ones from the tomb of the King of
Nan Yue.Io The flattened, pointed jades are often found with a category of oval ornaments that was
made by treating the thumb ring form as a flat pendant. It seems likely that the thumb rings too

were taken as ornaments from the borderlands (fig. 13).

These observations introduce an important methodological point. These particular jade pendants
have, hitherto, seldom been recognised as having exotic origins because they have been neatly cate-
gorized with ancient types of flat carved jade, such as huang arcs or bi discs. In all traditional
discussions of pendants, these two, the xi point pendant and the thumb-ring-shaped pieces, have
simply been treated as part of a long-running development of jade carving within a Chinese context.
This approach of including all items in a single category, regardless of origin, has thus obscured the
sources of such pieces and the mixing of traditions that such origins imply. When we turn next to
bronze vessels, we find something similar has happened, but this time changes have been ignored
because the category of bronze ritual vessels has been treated as exclusive.
We have to approach the problem of the apparent disappearance of the bronze ritual vessels from
this vantage point. The bronzes that I am now going to describe have rarely been accommodated
within the categories venerated for their associations with rituals for the ancestors. Thus in any ac-
count of the history of ritual bronzes, we see a consistent decline in numbers of the well-known
shapes over the fifth to third centuries B.C. But what this observation elides is evidence that, over an
even longer period, new types had also been coming into being. Many of these new types have
correctly not been accepted as being part of the ritual vessel category.I' The question of what these
bronzes were for has, however, usually been addressed piecemeal. Whether new rituals came into
being is a subject that has not been tackled, regrettably, for these interesting pieces have been
inadequately recorded or discussed."

99 See Kaogu xuebao I977I-1, 73-Io4, pl. 15. Compare also a few examples from aJin tomb at Taiyuan, in TaiyuanJinguo Zhao qing mu
(Beijing: Wenwu chubanshe, 1996), 158, fig. 85, hereafter referred to as Jinguo report. I am grateful to Roger Moorey and John
Curtis for suggesting to me that these horn-shaped hardstone pendants might descend from Iranian harness ornaments illustrated
in the sculptures at Persepolis. See P. R. S. Moorey, "The Iranian Contribution to Achaemenid Material Culture," Iran 23 (I985),
fig. 5.
i00oo See Linzi Shangwang mudi (Ji'nan: Qi Lu Shushe, 1997), 58, fig. 47:4, hereafter referred to as Shangwang report. For related jade
pendants in the tomb of the King of Nan Yue, see Nan Yue Wang report, vol. 2, pl. 116:3,4. As with the gold appliques, this
interest in horns had several different manifestations over a relatively long period. Horns appear at an early date in the Warring
States period, or even earlier as parts of sculptures that acted as guardians. Such antlered creatures are known especially in Chu, but
may have been yet earlier in the Huang state tombs in Henan, where a stand has been found. Horned beasts occur again in the bixie
of the Han, as small jade carvings and as massive stone ones. As with the lions, to be discussed below, these many imaginary
powerful creatures, later carved in stone and jade, were given horns. What is surprising about this observation is that the south, as
much as the northern periphery, seems to have been interested in the protective forces of horns. It seems possible that this interest
reached the south by way of either the east coast or down the Yangzi River from the west. In a similar manner, the south shared the
steppe area interest in gold appliques. Such gold appliques were found in the tomb of the Zeng Hou Yi, see Zeng Hou Yi report,
vol. 2, pls. 148-49. In addition, the same tomb contained some very intriguing ornaments in the shapes of horns. These are
embellished to look somewhat like later jade pendants, mentioned earlier in this note. For these ornaments, see Zeng Hou Yi
report, vol. I, fig. 252.
io0 So, Introduction, 66, notes the way in which Chu tombs show that southerners had moved away from complexes of bronzes
primarily for ritual use towards vessels that were also for more mundane functions. See also the development of lacquer in Chu and
elsewhere, Alain Thote, "Innovations techniques et diversification des commandes: l'artisanat du laque en Chine aux Ve-IVe siecles
avant J.-C.," Arts Asiatiques 45 (1990), 76-89.
z02 For a comment on changes indicated by wording in inscriptions, see Lothar von Falkenhausen, "Sources of Taoism: Reflections on
Archaeological Indicators of Religious Change in Eastern Zhou China," Taoist Resources 5 (1994), I-I2.

26

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 In fact, long before the third century B.C., and probably no later than the sixth century or even the
seventh century B.C., new vessels were routinely made and buried. Various braziers, sometimes
rectangular but also circular, often with chain handles, are among the earliest and most conspicuous
pieces and may have come from Chu or from the borderlands.o3 The contents of a fifth-century B.C.
Jin State tomb at Taiyuan include some of the other new types, among them are the braziers on
chains accompanied by shovels, small beaker-shaped vessels on a high foot, a cauldron with loop
handles and an eared cup and mirrors (fig. 14). We do not know how these items were used, or
whether or not they were employed to accompany the ritual vessels. A most unusual piece is some

sort of steamer, and it too seems to indicate a practice outside the normal run of things (fig. I5).l?4 A
late third-century B.C. tomb of the Qi state at Shangwang, Linzi, in Shandong province, which
contained a flask with a long neck, many small pouring vessels, and ear cups both in silver and in
bronze, and lamps of several kinds (fig. 16), shows how the repertory had grown.
It seems that several of these vessel types had distinct connections with the borderlands. The most
obvious of these are the small beaker-like containers on a high stem foot (fig. 14). While it is possible
that Chinese dou were the ultimate sources of the form, the fact that the vessel type is found both

within the central regions and beyond is evidence of an exchange of vessel types across the borders.I5
For similar bronzes were found on the periphery of the Chinese states and were also known in Siberia,
as at the Altai tombs of Pazyryk." Some unusual vessels were clearly brought into China. In the
sixth to fifth centuries B.C., these include various flasks, often flattened, carried by chain handles.Io7
Other pieces may also have had some northwestern sources. For a piece from Shangwang, Linzi,
appears to have a small jar with lobes sitting at the top of a slender neck (fig. 17b). This lobed surface
resembles the repousse patterns of silver bowls from Western Asia of the Achaemenid and Parthian
periods. The very unusual flask may be a possible prototype for garlic-headed hu, usually linked with

the Qin state.I?8 These were often used with, or accompanied by, a small, round-bottomed cauldrons
with two ring handles, known today as a mou. This, too, seems likely to have been based on a western
or northwestern vessel, perhaps from Sichuan. The cauldron may have sat on a ring, supported on
three legs, over a fire or some other source of heat. Examples of both categories are found in the tomb
of the King of Nan Yue, in the far south, for example.I?9 In the tomb, at Guangzhou, was also a silver
box with repouss6 lobes and a wreath-pattern around the lip; related pieces have been found as far
apart as Yunnan and the Qi state in Shandong. All these items are relatively close copies of Parthian

io3 So, 21, fig. 14, illustrates a large brazier from Xinzheng. In form, but not function, such rectangular pieces seem to have a precursor
in such bronzes as the Guo Ji Zi pan. See Ma Chengyuan, Ancient Chinese Bronzes (Hong Kong/Oxford/New York: Oxford Univer-
sity Press, 1986), 142-43.

i'4 Steamers were regular components of Han tombs, see the bronzes from Liu Sheng's tomb (fig.l).
"5 See Jenny F. So and Emma C. Bunker, Traders and Raiders on China's Northern Frontier (Seattle and London: Arthur M. Sackler
Gallery, Smithsonian Institution, University of Washington Press, 1995), lo8-o9 for a discussion of this category of vessel. The
authors place the origin of the type in China, but it may even in its earliest form have been a vessel category more typical of the
borders. Indeed, during the Western Zhou exchanges took place in both directions, many items and motifs that became common
in the steppe areas having originated in China. As the early Zhou had had strong contacts with the periphery, such exchanges may
have been fairly easy.
io6 For a discussion of cauldron types in Central Asia and Mongolia, see Miklos Erdy, "Hun and Xiong-nu Type Cauldron Finds
Throughout Eurasia," Eurasian Studies Yearbook 67 (1995), 5-94.
i07 Kaogu xuebao 1991.4, 449-95, pl. 15:5.
1o8 For Qin tombs, see Kaogu yu wenwu 1981.1, 12-35; Wenwu 1980.9, 15-24.
io See Kaogu 1992.3, 280-82, pl. 3:5; Nan Yue Wang report, vol. I, 280, fig. 194, 292, fig. 205; the latter shows the iron stands on which
such cauldrons may have been placed over fire. These tripod supports are very reminiscent of similar items in the Greek and Roman

areas. The cauldrons may have developed in Sichuan, see Wenwu 1985.5 17-22, fig. 21; 23-4o, fig.2L.

27

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
silverwork, and a jade rhyton was similarly copied from a piece in metal or hard-stone from Western
Asia.no

We should not view such items simply as exotic treasures. They seem to belong to a new set of
practices, and imply new objectives. The silver box may have held medicine.m Lobed flasks, ladles
and even boxes may possibly have been for drugs, narcotics, or special kinds of intoxicating drinks,
perhaps introduced from the peoples of the northern borders. The excavations at Pazyryk showed that
the inhabitants of the Altai used hemp seeds, perhaps scattered on hot stones in their cauldrons, to
produce intoxicating clouds of steam. Also discovered were rods which may have been covered by a
felt, under which the Altai peoples may have crouched to inhale the steam, as described by Hero-
dotus with reference to the Scyths.2 As mentioned, cauldrons appear in China at least as early as the
sixth to fifth centuries B.C. (fig. 14). In the Jin tombs, as already mentioned, and in the Zhongshan
tombs at Pingshan, circular fittings with hooks may have been part of circular canopies over
radiating ropes (fig. 18).13 Thus, peoples in both areas may have taken over some elements of the
apparatus required for inhaling incense or drugs.
Angular sections of bronze that could have supported a canopy of textile on a rectangular
framework, found in the tomb of Liu Sheng, suggest that members of the Han Imperial family may
also have sat in tents or under canopies for ceremonies.4 Screens may, similarly, have been part of
such paraphernalia. An early one (fig. 8c) from the Zhongshan tombs had the fine inlaid bronze
support in the shape of a tiger attacking a deer (fig. 8b).n5 It was accompanied by a pair of winged
felines (fig. 8a). A later screen in the distant southern tomb of the King of Nan Yue at Guangzhou
was also embellished by extraordinary, perhaps supernatural, creatures in bronze.n6 While it is
possible that these two object categories, the canopy and the screen, may have been in use during the
Zhou period, they may have been elaborated as a result of contact with the borderlands. Both were
suitable for people who moved in carts or on horseback.
During the fourth century B.C. lamps, previously unseen among Chinese metalwork, suddenly
became popular. Here too we have an object category that seems to have had strong links with the
steppe regions, Siberia and areas much further to the West. Especially fine examples have come from
the Zhongshan. One of these shows a small figure holding various sorts of snake, suggesting that the

io See Michile Pirazzoli-t'Serstevens, "Pour une archeologie des 6changes. Apports 6trangers en Chine - transmission, reception,
assimilation," Arts Asiatiques 49, 1994, 21-23, for a discussion of the box. The finds at Shangwang (Shangwang report, figs. 15-16)
show that lobed items had entered the Chinese region, perhaps from the north. They may have been linked to drinking or
medicaments.

.ii Some of the items in Liu Sheng's tomb may have been intended for medicine, see Mancheng report, vol. I, figs. 51, 81.
iiz S.I. Rudenko, Frozen Tombs of Siberia. The Pazyryk Burials of Iron-Age Horsemen, M.W. Thompson, trans., (London: J.M. Dent &
Sons Ltd, 1970), 284, gives a description of the inhaling rituals from the account in Herodotus.

ii3 See Jinguo report, 134, fig. 71, for a bronze fitting that may have been part of a tent. A similar fitting occurs in the tombs at
Zhongshan, see Zhongshan report, 282, fig. 127.
114 For the contents of the Mancheng tombs, see Mancheng report, vol. 2, color pls. 21-22, 24, pls. 17-18, 29-30, 177-78. All the Liu
family tombs and their predecessors in Shandong contained many ladles, suggesting that ladling of fine beverages, perhaps intox-
icating ones, was a major practice of court life. The very high quality of wine flasks indicates the particular importance of drinking.
Mirrors came ultimately from Siberia and were used both in central northern states, such as the state ofJin, but also were very
popular in Chu. It is likely that these mirrors were used in some rite, rather than being used solely as toilet items. For the fine
inlaid mirror from the tomb of the King of Nan Yue, see Nan Yue Wang report, vol. 2, color pl. zz22. This mirror resembles one
found in Shandong, see Wenwu 1972.5, color pl.I.
1i5 See Zhongshan report, 279, fig. 125.
n6 A lacquer screen from Mawangdui was decorated with a dragon and a bi disc threaded with cords, both auspicious or even magical
things. See Mawangdui report, vol. I, 94.

28

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
individual was in some way supernatural.I7 Wangshan in the Chu kingdom has produced another
interesting lamp, with a figure riding on a camel, a creature associated with the borderlands, but
certainly not with Chu.` The camel later occurs as part of the series of the animals of the directions
on a few Han pieces, including a hill censer from the tomb of Dou Wan.n9 When the camel
was inserted in a sequence with three animals of the directions, taking the place of the fourth, the
northern one of the tortoise and the snake, it seems as if the camel has acquired some association with
the spirit world. Thus, the earlier lamp employing a camel as a support may have been seen, not as an
exotic item, but as one with supernatural connotations. Such images, again, suggest a context in
which the spirits and the supernatural belonged to the border areas.
Spirits were also signified by some later ceramic lamps with their tree-like shapes on which
perched figures of immortals.' These ceramic lamps seem to have had rare bronze prototypes. Other
lamps are crowned by the bird of the sun and sometimes also depict the toad and hare of the moon
(fig. I9). The two characters for sun and moon together make the character ming, bright: bright light
was also associated with intercourse with the spirits.' Mirrors too were sources of light and
radiance. Here we see the beginnings of an explicit preoccupation with brightness as a necessity in
the afterlife.m While the notion of brightness, ming or shenming, had a long history in ancient China,
mirrors had been used only intermittently. The earliest mirrors seem to be based upon items from the
borderland and they would seem to have become a popular object category from the sixth century
B.C. or so, again through copies of Siberian examples. Mirrors, like gold, were distributed both
through northern central China and down the tributaries of the Yangzi River to Hubei and Hunan.
Such mirrors, most especially those with the TLV design, seem to have been seen to have a strong
link with the spirit world.I3
While the ancient Chinese seem to have adopted certain foreign object categories, such as mirrors,
lobed boxes and lamps, they accommodated them within their own context, giving them Chinese
functions and associations. We see another way in which the Chinese attributed local meanings to
foreign items when we consider widely-used animal motifs borrowed from Western Asia. The lion, a
much favored Western Asian subject, acquired several slightly different interpretations in China. We
have already seen an example of this process in the use of the Western Asian animals-in-combat
subject illustrated by a tiger devouring a deer, made as inlaid bronze fitting, possibly for a screen,
from the Zhongshan tombs (fig. 8b). Here the East Asian craftsman has replaced with a tiger the lion
that would have been much more common in Mesopotamian bronze castings, stone sculpture,
engraved seals and many other items. Lions were also common on artefacts from Iran and south
Siberia. In China, because lions were unknown, Western Asian motifs were often reworked as
tigers or as bears.

ii7 Jessica Rawson, Mysteries of Ancient China, New Discoveriesfrom the Early Dynasties (London: British Museum Press, 1996), no. 74.
**8 Wangshan report, color pl. 5.
**9 Mancheng report, vol. 2, color pl. zz.
,"o Rawson, Mysteries, no. 97; see also Smith, 91, for use of lamps in Han and later rites.
'zz See Sun Ji, Handai wuzhi wenhua ziliao tushuo (Beijing: Wenwu chubanshe, 199I), 351-57; Sun Ji, Zhongguo shenghuo. Zhongguo
guwenwu yu Dong-Xi wenhua jiaoliuzhongde ruogan wenti (Liaoning: Liaoning Jiaoyu chubanshe, 1996).
,zz E. Brashier, "Longevity Like Metal and Stone: The Role of the Mirror In Han Burials," T'oung Pao 81, (1995), zo-9. Inscriptions
on mirrors refer to brightness and to aspects of the spirit world.
123 The study of mirrors is a very broad subject. See especially Loewe, Ways to Paradise, Chapter 3.
124 Among stone sculptures, many of the tiger or winged feline beasts were based upon the Western Asian lion, see Wang Luyu, ed.,
pls. 5o-6o.

29

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 I suggest that lions may have been models for some bear motifs because a ruff figures on de-
pictions of both lions and bears. Figure 2o illustrates an Assyrian ivory carving of a lion's head with a
drawing of the bear's head on the plaques in Figure 21. Both creatures have small rounded ears, a
conspicuous ruff, and open jaws. But although the shape of the bear's head seems to have come from
beyond the borders, bears had acquired local associations, being linked with the Yellow Emperor. J5
Bears in other materials, most especially gilt bronze, occur frequently as the feet of vessels, such as
those in the tomb of Liu Sheng (fig. I). They are also depicted in many scenes of miraculous
landscapes, some of which are illustrated below in the last section of the paper (fig. 34). In these new
forms, the lion transformed not only metalworking designs, but also, unexpectedly, jade carving;
discussed in the next section (fig. 32).
Fierce felines, bear and deer appear on fine belt ornaments based upon prototypes from the steppe
area in Inner Mongolia or from South Siberia."7 The Chu king buried at Shizishan, Xuzhou, had a
particularly gorgeous pair of gold belt-clasps imitating, in some respects, the carved wooden orna-
ments of the nomadic peoples buried at Pazyryk (fig. 2I). M Here a deer-like creature (with horns
tipped by birds' heads)29 is being attacked by a wolf and a bear, already illustrated.3? The main
creature with long horns tipped with birds heads, appears in many forms in areas of Mongolia and
Siberia bordering on China. A horse frontlet from the tomb of the King of Nan Yue is a dramatic
example of a similar creature, but without the horns (fig. 22)."3 An enlarged head dominates the top
of the frontlet while elongated and twisted limbs fill the lower part. The gold belt-plaque and the
frontlet both reproduce the deep carved striations of wooden carvings. When gilded, these angular
surfaces reflected the light strongly, adding glitter to the pieces. These fine belt-plaques and harness
fittings were adopted by many of the Imperial princes and by the King of Nan Yue, a clear example
of the effects of a unified empire creating a common demand for luxury goods across the whole
region. In earlier Warring States times, such items were much more frequently restricted to one re-
gion, for example the Kingdom of Zhongshan, where exceptionally beautiful inlaid fittings with
Western Asian subjects seem to have been confined to this one area (figs 8a-c). As we shall now see,
like the belt-plaques, Chu practices seem to have been widely disseminated in the same way across a
large area as a consequence of centralized power over a vast region.

THE SOUTH AND THE STATE OF CHU

If the Han Imperial family chose mountain sites with horizontal access for their tom
contacts with the north, their practices of creating tombs with many rooms, each fo
tions, came from Chu. From at least the late fifth century B.C., if not earlier, major

I25 See Lewis, Sanctioned Violence, 190-91.

126 Compare also discs and handles with bear motifs, Mancheng report, vol. I, fig. 62:I, 4 and Nan Yue Wang rep
12.

127 See Sun Ji, "Xian Qin, Han, Jin yaodai yong jin yin daikou," Wenwu 1994.I, 5o-64.
128 Gold plaques and appliques from the area of present-day Xinjiang province provide intermediaries between the wooden carvings
from Pazyryk and the gold belt-buckles described. See Wenwu 1981.1, 18-22, pl.8.
'29 In this connection, note an antler decorated with birds found at Teng xian in Shandong province. See Kaogu xuebao I991.4, 449-95,
pl. 5:6. Compare a three-dimensional figure in gold (Wenwu I983.12, 23-24, pl. 4:I). It is possible that this interest in antlers came
from the border areas and infiltrated many areas of China, being most popular in Chu. See also footnote Ioo.
130 For other objects decorated with bears, see Nan Yue Wang report, vol. 2, pl. z52; Mancheng report, vol. 2, pl. yo:I.
I3 Related pieces, as yet unpublished, have been retrieved from the tombs at Yongcheng, see also Wenwu 1998.8, 37-43.

3o

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
south had comprised several rooms, each one dedicated to a separate function or with a different
emphasis. The famous, late fifth-century B.C. tomb of the Marquis Yi of Zeng had four rooms: one for
ceremonies; a private room for personal entertainment, with his inner coffin as his own chamber; a
room for his attendants; and a room for weapons and chariot parts. Rather smaller and more com-
pressed versions of these multi-chambered tombs for lower ranking members of Chu elite have been
found in Hubei and Hunan.32 The Han, and perhaps the Qin before them, seem to have taken from
Chu this conception of a tomb as a sequence of individual rooms with separate and distinct functions.
The officials responsible for supplying these tombs also followed Chu customs. While tombs in
the Yellow River area had been filled with bronzes and jades for ritual and war, in Chu objects for
daily life had been preferred. Lacquer eating and drinking vessels, boxes and furniture were particu-
larly prominent, and were taken over by the Qin and the Han. Many of the Chu categories of con-
tainers and furnishings were now made in a variety of materials ranging from fine silver to common
ceramics. Ceramic types were often decorated with lacquer patterns (fig. Id).'33 Several vessel types,
most especially circular vessels originally made of lacquer, known as zun and zhi, were clearly
borrowed from Chu. I134 Like the cauldrons and flasks from the west or the north, these circular vessels
lay outside the traditional ritual vessel categories. In Han tombs, they often occurred in gilt bronze
versions and occasionally in jade.
From the south, also, may have come an emphasis on music and dance in life and the afterlife.35 The
fifth-century B.C. tomb of the Marquis Yi of Zeng contained an extraordinary number of musical
instruments, including a renowned set of sixty-five bells.36 Lesser Chu tombs were also supplied
with many musical instruments. In their turn, the great Han tombs at Xuzhou held pottery models
of musicians (and perhaps also many actual instruments now lost).37 The musical instruments stored
in the tombs of Liu Sheng and the King of Nan Yue included sets of bells and stringed instruments,
or qin, decorated with gilded knobs in the shapes of mountains on which prowled wild animals. We
should not see dancing and music in isolation, for they probably accompanied the feasting and
drinking in which the vessels just described were used (fig. 6). As already mentioned, such dancing
may also have been intended to summon the spirits to join the festivities.
Essential both to the banquets and to the practices to summon the spirits were servants and
attendants of every kind. Chu seems to have contributed a conviction that inert models of figures
could become suitable servants in the afterlife, and thus, perhaps, stimulated production of tomb
models of all types.38 Well before unification, some parts of the Yellow River area had employed
models and figures in clay. For instance, the Qin had made ceramic granaries, while in Shandong

132 See, for example, the tomb at Tianxingguan, Jiangling, in Hubei. Kaogu xuebao 1982.1, 7I-II6.
'33 For later examples of ceramic imitations of lacquer, see Sturman, "Celestial Journey." Lacquer vessels found in the southern tombs,
see Wenwu 1978.8, 12-31, were often replaced by painted ceramic in more northerly tombs, as in Liu Sheng's tomb at Mancheng. See
Mancheng report, vol. I, I29-33.
'34 Discussed in Jessica Rawson, "Chu Influences on the Development of Han Bronze Vessels," Arts Asiatiques 44 (1989), 84-99;
compare also Suning Sun-Bailey, "Gained in Translation," Orientations 22, no. 7 (July 1991), 25-27.
135 Loewe, "Changes in Qin and Han China."
36 So, 19, has pointed out that this interest in music appears at an earlier date in the east, as at Liujiadianzi in Yishui, Shandong
province. See Wenwu 1984.9, 1-I3.

37 China - Verborgene Sch'tze. Grabfunde der Han-Dynastie (Leoben: Kunsthalle Leoben, 1998), 63-65.
'38 For a tomb at Changzhi, where, in the pre-Han period, wooden models and human figures were used concurrently, see Kaogu xuebao
1984.4, 503-29, fig. Io. An antler discovered in this tomb suggests that the practice may have been borrowed from Chu or from
eastern China.

31

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
small figures of musicians occur in a very few late Warring States tombs.139 But these few examples
have none of the elegance or elaborate detail of the wooden figures of the fourth- to third-century B.C.
Chu tombs at Xinyang, Baoshan and Mashan.4 The Chu figures were of wood and sometimes wore
clothes in textile, while sometimes their dress was painted. A few even had articulated, moveable
limbs, suggesting that the powers of the figures to move and so to attend to their owners' needs were
important. Figures in a single tomb were often ranked by size, according to the importance of their
roles. Their dress and ornaments, such as pendants, further defined status among them. Such ranking
seems to have been followed by the hierarchical treatment of figures by size and material under the
Han.141

This same attention to rank and function, defined by details of dress and weapons, seems to
characterize the terracotta army of the Qin and the later, smaller Han ceramic armies.42 In par-
ticular, Qin and Han figures of soldiers reproduce the correct dress and gestures appropriate for each
category of warrior (fig. 23). Those of the Chu kingdom tombs at Xuzhou, Guishan and Beidongshan,
have carefully painted clothes, and neatly differentiated hairstyles. They resemble figures in wood in
their overall form and in the care with which their dress was painted. 3 In addition, many Qin and
Han clay figures of servants have contemporary wooden counterparts, which they seem to copy.44
In the case of models of soldiers massed to defend the tomb of the First Emperor and those of later
Han emperors and princes, a further contribution from Chu must be considered. That is the matter of
fear, mentioned above in connection with Sofukawa Hiroshi's views on the role of the terracotta
warriors. We can infer that the peoples of the south feared spirits, demons and evil influences from
the strange, armed figures painted on the innermost coffin of the Marquis Yi of Zeng. These creatures
carry the kinds of halberds that warriors of the day would have used. Such images were, I presume,
intended to defend the tomb owner."45 In slightly later Chu tombs wooden carvings crowned by ant-
lers seem to have acted as guardians.146 Thus it would seem, from the south may have come notions
of creating figures, especially of supernatural creatures, as a way of enlisting them to ward off attacks
by evil, demonic forces. In the Yellow River area, ceramic armies answered some of the need for
defense. More traditional Chu practices continued in the south. On the outermost of the inner coffins

'39 Discussed more fully in Jessica Rawson, "Commanding the Spirits. Control through Bronze and Jade." Orientations 29, no. 2 (Feb-
ruary 1998), 33-45. The musician figures found in Shandong may, perhaps, have borrowed a Chu practice, as there is some evidence
that there was knowledge of the wooden tomb figures of Chu at an even earlier date. See Wenwu I993.3, 1-6, color pl.
140 See Xinyang Chu mu (Beijing: Wenwu chubanshe, 1986), figs. 39-40, 79; Baoshan report, vol. I, figs. 169-70; andfiangling Mashan
yihao Chumu (Beijing: Wenwu chubanshe, 1985), fig. 66. The role of figures in Chu tombs was discussed by Alain Thote in an
unpublished paper at a conference at the British Museum in December 1996. For concerns with function, see Jessica Rawson,
"Thinking in Pictures. Tomb Figures in the Chinese View of the Afterlife," a paper given to the Oriental Ceramic Society on April
22zz 1997, forthcoming in Transactions of the Oriental Ceramic Society 61; Patricia Berger, "Body Doubles: Sculpture for the Afterlife."
Orientations 29, no. 2 (February 1998), 46-53.
14' The Quest for Eternity, Chinese Ceramic Sculptures from the People's Republic of China (Los Angeles and London: Los Angeles County
Museum, Chronicle Books, Thames and Hudson, 1987), 15, note 9, discusses sayings attributed to Confucius. Hitherto Confucius'
references to figurines have not been fully understood, as it was not recognized that the range of figures in burial could and did
include humans. See Rawson, "Thinking in Pictures."
'42 See Rawson, "Thinking in Pictures," for a discussion of the contributions of Chu to the use of tomb figures in the Qin and Han
periods.
143 Compare the clay figures from Beidongshan, Xuzhou (Wenwu 1988.2, 2z-I8, 68, color pl.) with the wooden figures from Feng-
huangshan, Jiangling in Hubei province (Kaogu xuebao 1993.4, 455-512, figs. 30-31).
44 Compare wooden figures from Shandong (Wenwu 1980.12, 7-16, fig. 20o) with a ceramic figure from Shaanxi (Wenwu 1981.11, 24-29,
fig. 7).
,45 Zeng Hou Yi report, vol. I, 36, fig. 2I, 39, fig. 22.
'46 Sofukawa, "Lingmu zhidu."

32

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 ,, d*C ) '~G~Y ' '7 v A
-7I?

ICY

0I~0

//

20 cm

Fig. Ia. Drawings of fine bronze hu vessels, braziers and incense burners from the tomb of Liu Sheng, King of Zhongshan, buried at
Mancheng, Hebei province. Western Han period, ca. 113 B.C. After Mancheng Han mu fajue baogao, vol.I, figs. 22-23, 26, 29, 31, 32:1-2, 34-
41, 44, 46.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 10 cm

. ,, ...........

Fig. Ib. Drawings of lamps from the tomb of Liu Sheng. Western Han period, ca. 113 B.C. After Mancheng Han mu fajue baogao, vol. I, figs.
47-49-

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 i__i .
/ 20 cm

i -,.

Fig. Ic. Drawings of plain bronze hu, braziers, steamers, pouring vessels and ladles from t

ca. 113 B.C. After Mancheng Han mu fajue baogao, vol. I, figs. 32:3-4, 35, 37, 39, 42-45.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 ,\?c?c

o~

Fig. Id. Drawings of ceramic vessels decorated with lacquer pat-

terns from the tomb ofLiu Sheng. Western Han period, ca. 113 B.c.
After Mancheng report, vol. I, fig. 87. These categories of vessels
seem to have been used for offerings from at least the Han period.

//

I~I/.? / I ~
'/?/
.?

r,
~
\ / I
?Ir
~~\ //. .. //

c,
.. z? /: / //
/? ? ~'/?
.L/
/// //
5(
//I

Iff~c IPI

o a
?I~ o a
ri(~\f f
d
rS~

.44~~ ~_`2~?t~
Q B t tO OO
8"
Q 32
~9
)r,
tP' ~
-~-- ?e;LII~

B?
8

Fig. 2. Reconstruction of the tomb of Liu Sheng at Mancheng, Hebei province. After Wen Fong, ed., The Great Bronze Age of China. An
Exhibition from the People's Republic of China. (New York: The Metropolitan Museum of Art, I980), 326, fig. 112.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 NORTH ---+I---
IVx

X11

11 rJJ

blocki ng stones
VH ] V lI
23.7 m VI 23.2 m

central pillar [] well

rear chamber
XV XIV

lavatories
) stone

ante-chamber

\corridor corridor
music and dance hall

niches

4.7 m

~t~r E~3~~~ ~~I-well tomb

Swell pathomb
0 101-11 path
I I I

Fig.
Fig. 3a. Plan of rock-carved cave tomb at Beidongshan, Xuzhou, 3b. Plan of r
Jiangsu province, Western Han period. After Li Yinde,
su "The province. After
'Underground Palace' of a Chu Prince at Beidongshan," Orienta-

tions 21, no. Io (October 1990), 57, fig. I.

Fig. 3c. Plan of a tomb of the W

wangling yu qinyuan, fig. 65.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
Fig. 4a. View of the interior of a late Western Han tomb at Baicun, near Xuzhou. Photo- Fig. 4b. A column supported on the back
graph by the author. of a ram from a late Western Han tomb at

Baicun, near Xuzhou. Photograph by the

author.

Fig. 5. Stone servant figures from the inner chamber ofLiu Sheng's tomb. Western Han
period, ca. 113 B.c. Photograph by the author.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 0 10 cm

oil

WWR W6-1I,

Fig. 6. Rubbing of carved stone slabs depicting a dragon and fish Fig. 8a. Drawing of a winged beast in inlaid bronze from the tomb
parade, from Yi'nan, Shandong province. Eastern Han period, of King Cuo at Zhongshan, Pingshan, Hebei province. Late 4th
late 2nd century B.C. After Yi'nan gu huaxiangshi mufajue baogao century B.C. After Cuomu Zhanguo Zhongshanguo guowang zhi mu,
(Nanjing: Nanjing Bowuyuan, 1956), rubbing no. 34. fig. 51.

' I

':~-/I, 'F"~a~~'~,~f~i~`ctST~g _p~

?2:

J 0(0:i? ~? C
.? ?0

'O
10 c

r:

Fig. 8b. Drawing of a tig

Cuomu Zhanguo Zhongsh

......

?i"?~?;~:~

Fig. 7. Drawing of the banner found in tomb MI at Mawangdui, F

Changsha, in Hunan province. Han dynasty, second century B.c. th
After Changsha Mawangdui yihao Han mu, vol. I, 40, fig. 38. A

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 IC

Z.0

0 5

Fig. 9. Tomb of stone slabs,

column at Changli, Jiangsu
1957-12, 36.

NORTH

C-,CVR, huvv?&-4m skdotonl va fammnt

a ,,.,-- -.r evel;fi rh.orc

objects pit

WES sarcophagus
C.19. tlo -,.y
WEST -- . " -- EAST chamber
aj a urth"t""d l"" c"-(( A '
loon lljw &h4

SeE~i4:E
161IA 1I~ b ~t

Bart7 ft.
". . . .. .M- e." ... .. .. .. .."...-...
., . .

-0-0

tomb path

oI 3 4 5m
SOUTH

Fig. 11. Plan and section of the Scythian tomb at Kostromskaja in

Fig. Io. Plan of the tomb attributed to Jing Gong of the Qi
state (547-489 B.C.) at Heyaitou, Linzi, Shandong province.the Caucasus. After Sythians and Greeks, 29, fig. 128.
After

Wenwu I984.9, 14, fig. 2.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 j

AWML

\i

O
Fig. 12. Pendants in the form of horns and rings of crystal from Fig. 13. Rubbings of pendants from the tomb of the King of

Langjiazhuang, Zibo, Shandong province. 6th-5th century B.c. Nan Yue. Left: The two lower elements are derived from horn-

After Kaogu xuebao 1977.1, pl.I5. shaped items, seen first in crystal in 5th- and 4th-century B.c.
tombs in Shandong province. Length: II.2 cm. Right: A pendant
in the form of a flattened thumb ring. Length: 11.6 cm. Han

period, 2nd century B.c. After Xi Han Nan Yue Wang mu, 247,
fig. I68; 245, fig. I66:4.

~C~~?i~-:
F~a7- ~

Fig. I4b. Bronze

Fig. I4a. Drawings of a group of miscellaneous fu from Jundushan,
bronze vessels Yanqing,
and Beijing. 7th -6th
"TL
century
other items from a tomb of the Jin State at B.C. After Traders
Taiyuan, and Raiders,pro-
Shanxi fig. 22.I.

vince. 5th century B.c. After TaiyuanJinguo Zhao qing mu, fig. 65.

~I32

~i~J~- ~i
t32C I;;

~"~;J~?i~~
Fig. 15. Drawing of a steamer from a tomb of the Jin State at
Taiyuan, Shanxi province. 5th century B.c.. After Taiyuanjinguo
Zhao qing mu, fig. 65.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 O1 0

20 cm

m-o) s o

Fig. 16a. Drawings of vessels from Shangwang at

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 4i

20 cm

11-

Fig. 16b. Drawings of vessels from Shangwang at Linzi, Shandong province. 3rd century B.c. After Linzi Shangwang mudi, figs. 14-1
22-26.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 Na

a bL O 5 10 cm

Fig. 17. Fig. 18.

Comparison Drawing
of a of
gara b
King of Nan Yue,
tomb ofwith
the JinaState
fla
mouth, from a Qi
After state
Taiyuan tom
Jinguo Z

province. After Xi Han Nan

Linzi Shangwang mudi, 26, fi

Fig. 19. Ceramic mode

on a column on which
toad. While the bird i
toad refer to the moon
form that for bright,
period. Height: 27.8 cm
the Han (Singapore: L

Fig. 2oa. Drawing of a

9th-8th century B.C. A
and Empire, Treasures
British Museum Press
also shown in fig. 21,

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 ;2

fill

iiff~
\\\ ~I

~EY~~?)1

;L:od,

Fig.
Fig. 21I. Top, a drawing of the design on a gold belt plaque from a Western Han 22. Drawing
tomb at of a gilt-bronze horse
frontlet depicting an animal. From
Shizishan, Xuzhou, Jiangsu province. Length: I3.4 cm. After Zhongguo wenwu jinghua
(Beijing: Wenwu chubanshe, 1997), no. 87. The creatures shown on this belt ornament
the tomb of the King of Nan Yue. After
comprise a elongated horned beast resembling that in the carving shown below.XiItHan
is Nan Yue Wang mu, 96, fig. 64:4.
attacked by a bear and a wolf. Below, a drawing of a carved animal design in wood from
Pazyryk in South Siberia. After Frozen Tombs of Siberia. The Pazyryk Burials of Iron-Age

Horsemen, pl. I4oA. Drawings by Melanie Steiner.

Fig. 23. Drawings of two warriors from Fig. 24. Rubbing of a jade disc showing in Fig. 25. Rubbings of a jade disc decorated
pits east of the tomb of Qin Shi Huangdi openwork a feline dragon or tiger con- with a border of feline-like heads. From

at Lintong near Xi'an. Late 3rd century fronting a bird. From the tomb of the Nan Qufu, Shandong province. Late Warring
B.C. After Shaanxi Sheng Kaogu yanjiusuo, Yue Wang at Guangzhou, Guangdong States period. Diameter I9.9 cm. After
ed., Qin Shihuangling bingmayong keng, province. Han dynasty, 2nd century B.C. Qufu Luguo gucheng, fig. II6.
2 vols. (Beijing: Wenwu chubanshe, 1988), After Xi Han Nan Yue Wang mu, fig. I34.
figs. 24-25.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 ?"?
.p~.?i

c? ti: n
:I I
?n r, :~t:~ ?.?t i~ ??
LI

id ) ~nT) I /o o o

?:.", :2` ~. ,
i
?.~Fr
i bi ?
:
*. rc~ ~Cllgil
r ~i~J~~i~
...,

O'
,.4 r

"rl? ~h nrlv~r IIA~A~ nr

?~ ? i~ ~ar r.crru~ t

1 ~; '.1'~3]
a b c d
.i
?~?

I
.. ???i ?-?: Fig. 26b. D
t~
Siberia, th
LPI~
~E~, 1 r : ?i
transform
Yi3~3r~Zi~k:i u
~?~~?r

?i~ 'Q~~ii~L1~'
14?
a) lion's he
~ .~5~~ ;. Cb~
r:
1 i: x b) a carvin
.:r? T;
r: ;3i on?:? 2bt ? ;i~Ft$
-;1? on themselves.
... rl'-? nc?
f?i~ d
~5 P
/ JC?:
qyZS4.~'~~
?r' ~ n~r; vS c) carving of a head with inward turned folded ears.
i: It~'
;
d) carving of a head on which the ears have become two separate features, a long spiral
~3~?"
;? .~-?.?a~ia~-
,t ??
.:*
and a narrow blade or ear separate from the scroll.
:~ c~') t. ur-

Fig. 26a. A lion's head from Persepolis, Iran. After Frozen Tombs of Siberia. The Pazyryk Burials of Iron-Age Horsemen, 267, fig. I35, pls.
Achaemenid period, 5th century B.C. Photo- II6E, Io4F, IID.
graph by the author.

Fig. 27. Drawings of fantastic heads incised

on jades from the Di tombs at Zhongshan
and the Lu tombs at Qufu. Warring States
period, 4th-3rd century B.C.

a) plaque on which the head has been

given two views of its body as a new
way of reading the scrolls of 26 (d).
From Zhongshan. a b c
b) arc on which the same head appears
with its divided body appearing
either side, to fit the surface of the arc.

Small wing-like extensions are also

shown.

c) arc on which a similar head appears

-' rPd
with a more regular undulating body.
?/ - ,e"
d-e) sections of the borders of two bi
from Qufu, showing the ways in which
the same feline head with the two

views of its body have been extended

either side of the head, providing an
interlacing pattern.

After Cuomu Zhanguo Zhongshanguo guowang

zhi mu, 234, fig. IoI:I6, 226, fig. 98:2,

224, fig. 96:3; Qufu Luguo gucheng, 168, fig.
116, 162, fig. 109.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 Fig. 29a. Drawing of carving in stone of a feline. From
Shizishan, Xuzhou, Jiangsu province. Western Han
23.5 cm. Drawing by Melanie Steiner.

CAW CXJ'o
6)

Fig. 28. Drawing of a trapezoidal-shaped ornament with

Fig. 29b. a felineof a gold weight in the shape of a
Drawing
dragon, possibly following the shape of a scabbard
feline. ornament.
After Treasures, 3oo Best Excavated Antiques fro
From the tomb of a Chu king at Shizishan, Xuzhou, Jiangsu
(Beijing: New World Press, 1992), no. 269. Inscribed
number
province. Han period, 2nd century B.c. Length: 2zI cm.6, indicating
After Con- that it was one of a series.
Both
noisseur (Autumn 1995), 35. Drawing by Melanie felines lie on flat bases and have loops behind th
Steiner.
sharing the distinctive features of the Achaemenid w
30. Drawings by Melanie Steiner.

Fig. 30. Bronze weight in the shape of a lion in the shape of a crouching lion on a flat base. Heigh

6th-sth century B.C. From Susa. Courtesy of the Louvre, Paris.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
Fig. 31. Drawing of an ivory scabbard decorated with a lion Fig. 32. Carving in jade of a tiger. Western Han period. Width:
clutching or grappling a deer. From Tachti Sangin on the Oxus. 5 cm. Private collection.

6th-sth century B.C. Height: 27.7 cm. After Oxus. 2oooJahre

Kunst am Oxus Fluss in Mittleasien Neue Funde aus der Sowjetrepublik
Tadschikistan, no. 5.

P
\~~n~T\UIQ/nl ~iY*n\l\-PWd(7~~ D)

:9

-?

)c~Tf;?

o 3 cm

::~:,
r-?

c3 cm
3
2

?~?~~
\n

~LS

Fig. 33. Drawing of an inlaid bronze hill censor from the tomb of Fig. 34. Drawing of the designs on a tubular bronze fitting inlaid

Liu Sheng. Ca. 113 B.C. After Mancheng Han mu fajue baogao, vol. I, with gold and silver and semi-precious stones. From tomb MIzz

64, at Ding Xian, Hebei province. Height: 26.6 cm. Western Han
dynasty, Ist century B.C. After A Selection of Archaeological Finds of

the People's Republic of China (Beijing: Wenwu press, I976), no. 66.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
of the wife of the Marquis of Dai, at Mawangdui in Hunan, whole galaxies of strange creatures were
painted devouring snakes.I47
From the south, also, came depictions of benign spirits, such as the creatures that were to become
the animals of the four directions, seen first on a clothes box from the tomb of the Marquis Yi of
Zeng, and characters in myths, such as that of the Archer Yi who shot down nine suns.48 As noted
above, the descendants of such images are the painted and carved depictions in Han tombs of the
animals of the four directions and the deities, for instance, the Queen Mother of the West, and other
creatures whose sighting was thought to be a good omen. Fine birds, shown as perching on the roofs
of buildings, may have been xiangrui. 49However, as I argue below, many of these creatures should
be seen as presenting a whole cosmology. The notion of describing and illustrating the whole
universe, be it in poetry or images, seems to have been one of the most wide-ranging and
fundamental Chu contributions, although the east also had a major contribution to make. I shall
resume the subject of animal-filled landscapes at the end of the paper.

THE STATE OF QI AND THE EASTERN INTEREST IN JADE

While it is generally assumed that the relatives of the emperors established states in eastern China in
order to control a politically volatile area, the kings of the Liu family may have favored the east
because here were mountains whence one might hope to make contact with spirits and because here
too were good supplies of jade. The peoples of Liangzhu culture in Zhejiang and Jiangsu, in the
period 3500oo-z25oo B.C., had buried large numbers of discs and other jades in their tombs. It is pos-
sible that the Han stumbled upon these by accident. Had they done so they would, almost certainly,
have regarded such a discovery as a good omen, as they did the recovery of ancient bronzes from the
earth.z5o Neolithic jades were deployed in Han tombs (for example a jade cong is part of Liu Sheng's
jade suit), and show that the Han kings had, by some means as this, become owners of ancient pieces.
Many of their finest jades may even have been recarved from neolithic pieces. However, a fashion for
piling jades above and below the body had started well before the Han. Third-century B.C. Chu
tombs at Yanggong, Changfeng, in Anhui province, Qi tombs at Shangwang, Linzi, in Shandong
province, as well as the Lu state burials at Qufu, in Shandong province, all contained a profusion of
jade discs.'Y The choice of jade for parts of the tomb probably added the qualities of this precious
material - especially eternal endurance - to the picture, or pattern, of the created universe.'S5 It is
also possible that fear of demons and spirits encouraged the use of jade in some sort of protective role.
Jade was used almost obsessively near the bodies of the Liu princes. Some of their coffins, for in-
stance that of Dou Wan at Mancheng, were ornamented with discs, and lined with jade tiles.Y

I47 Mawangdui report, vol. I, fig. 17.

i48 Zeng Hou Yi report, vol. 2, pls. 121, 123.
'49 See Xuzhou Han huaxiangshi (Beijing: Zhongguo shijieyu chubanshe, 1995), lo, 16, 26, 73, 89.
'50 Noel Barnard, "Records of Discoveries of Bronze Vessels in Literary Sources and Some Pertinent Remarks on Aspects of Chinese
Historiography," Journal of the Institute of Chinese Studies of the Chinese University of Hong Kong 6, no. 2 (1973), 455-546.

I5 For tombs in which a layer or layers of jade discs were employed, see Kaoguxue jikan 2 (1982), 47-59; Qufu Luguo gucheng(Ji'nan: Ji
Lu Shushe, 1982), 129, fig. 8o; and Shangwang report, 12, fig. 7.
I52 For a brief discussion of the ways in which meaning or association accrues to objects, see Rawson, ChineseJade, 13-20.
53From the practice of using layers of jades, which in the case ofLiu Sheng at Mancheng and the Nan Yue Wang at Guangzhou were
tied with silken cords, developed two further practices, the representation of discs with stretched diagonal cords and the creation of
discs carved with openwork dragons and the depiction, as at Mawangdui, of dragons winding through discs (fig. 7).

49

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
Scattered examples of such tiles have been recovered from the looted tombs at Xuzhou. Jade suits
with densely fitting plaques were the most extraordinary evidence of the fears and concerns of the
Han. Such suits must have been costly in material and very time-consuming to make. Yet it is likely
that more than a hundred, and possibly many more, were made for members of the Imperial family.
As already mentioned this uniformity in an expensive burial garment is testimony to the highly
organized and centralized empire. The suits were tight-fitting around the body, like a garment. The
appearance and functions of such suits seem quite different from the array of pendants, plaques and
beads of jade and agate used at previous periods. The ancient Chinese seem to have thought that
disease and corruption came about because an individual was attacked by demons and other male-
volent forces. The tight-fitting plaques perhaps, therefore, implied protection, as they were of an
enduring material difficult to penetrate, even for a demon.
If the suits of plaques were for protection against evil forces, so too seem to have been the jade
halberd heads, which have been found in tombs at Xuzhou and Yongcheng.'54 All jade weapons,
jade-mounted swords as well as halberds, were equivalent to, but different from, the weapons of the
living. As Pascal Boyer has discussed, notions concerned with religion or the supernatural are often
attention-attracting by combining the real with the counterintuitive and in this manner suggesting
their powers."55 Although the jade suit has the form of a protective garment, even of armor and the
jade weapons appear to be for defense, attacks with the normal iron or bronze blades would have
shattered them. Nevertheless, they may have been perfectly fitted for battles with demons and spirits
in the afterlife. Equally impressive are the fine vessels in jade, often in alien shapes - such as the
rhyton from Western Asia, or lacquer forms from Chu. Here, too, it seems the qualities of the ma-
terial are added to the value of the pieces.
The animal motifs, which came, in part, from regions in the northwest, also had a profound effect
on jade design. A striking and rare example is an openwork disc, from the tomb of the King of Nan
Yue at Guangzhou, that shows a large feline-like dragon confronting a much smaller bird, perhaps a
phoenix (fig. 24). The feline-like dragon, or perhaps tiger, seems to be a jade carver's rendering of the
felines employed in East Asia derived ultimately from lions. Here the feline appears to take the role
of one of the directional beasts, confronting a phoenix, representing another. There is no mistaking
the mutual hostility, a hostility that normally is never seen in the creatures of the directions. This is,
indeed, a Chinese rendering of the animal-in-combat theme imported from Western Asia.s56 In a
new context, these aggressive creatures must have acquired a local meaning, being assimilated with
the animals of the directions. It is possible that such fierce encounters were seen as the cosmic
counterparts of the staged animal combats at the Warring States and Han courts, described by Mark
Lewis.57 The scrolls and flourishes of the carving were a product or a transfer of the fluent lines of
southern lacquers into jade.x58 But this is an unusual case.

'54 ChineseJade, Selected Articles from Orientations r983-i9pp96(Hong Kong: Orientations Magazine Ltd, 1997), front cover; Zhongguo yuqi
quanji (Hebei meishu chubanshe, 1993), no. 19o.
I55 Pascal Boyer, The Naturalness of Religious Ideas. A Cognitive Theory of Religion (Berkeley/Los Angeles/London: University of
California Press, I994).
I56 Related examples occur on the middle inner coffin of the wife of the Marquis of Dai buried at Changsha in Hunan province. This
example has been noted by several writers (see note I63). For Central Asian sources of the feline-dragon, see Wenwu I98I.I, 18-22,
pl. 18:3.
I57 See Lewis, Sanctioned Violence, Chapters 4 and 5.

i5 Compare also the tiger and dragon attacking each other on the top of the Mawangdui coffin, see Mawangdui report, vol. I, fig. 22.

5o

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 Western Asian designs introduced by way of south Siberia and Inner Mongolia also had a more
general impact on jade design, seen more regularly on jade discs, bordered with animal heads, in the
Liu tombs and earlier in east-coast tombs of the late Warring States period (fig. 25). In the tombs of
Liu Sheng and of the King of Nan Yue these discs were placed close to the bodies of the kings as part
of the construction of their jade suits and seem, therefore, to have been thought to be especially
potent.
The animal heads that border the discs were derived from a lion's head motif. A possible prototype
is seen in the sculptures of Persepolis, Iran (fig. 26a). This creature's head has four distinctive features
that reappear on artefacts in China as attributes of tigers' heads: large eyes ringed by folds of skin;
lines on the muzzle, a fringe or ruff around the head, and ears folded back. As early as the sixth
century B.C., these features of the lion's head appear as those of a tiger's head in some decorated
portions of model and mould fragments from the foundry site at Houma. Many of these mould
fragments carry small tiger-like stripes.59 The same head was employed in Siberia, where it appears
in various forms on wooden plaques from the burials at Pazyryk in the Altai mountains (fig. 26b).
Concurrently the same lion or tiger face was incorporated in Chinese jade carvings. The form
borrowed was a version in which the folded ears of the original lion design were greatly extended into
scrolls, as in Figure 26d. Jades from the Warring States Kingdom of Zhongshan show this head on
different categories of plaque. It appears with two views of its body on rectangular plaques (fig. 27a).
Where the face was adapted for an arc, the face overwhelms the much slighter bodies either side of
the head (fig. 27b). These are intertwined with what appear to be wings. From this time onwards, the
motifs were adapted for borders of bi discs. Early versions appear in Lu tombs at Qufu and in Qi
tombs at Shangwang, and later ones in the Liu family tombs (figs. 27d-e). The ways in which designs
from the liminal regions were appropriated and incorporated in some of the most spiritually precious
items for the afterlife indicates the value placed on these essentially "foreign" motifs.
Ornaments from Siberia or Mongolia made a further contribution. The complex plaques in gold
or gilt bronze, illustrated above (fig. 21), seem to have stimulated designs in which creatures appear
to burrow in and out of the surface of the jade, providing vivid relief effects. Such ornaments were
particularly popular as sword fittings and as parts of other weapons. An especially elaborate item is a
trapezoidal ornament from a tomb of a Chu King at Shizishan (fig. 28). Here a very fine feline-dragon
is the principal motif. It stands almost vertically and, though shown in openwork, it too weaves in
and out among abstract decorative elements. The ribbed and striated tail of this and other feline-like
dragons, seen on the disc showing two creatures in combat (fig. 24), was perhaps also borrowed by
jade craftsmen from the deep ribs on the gold or gilt-bronze plaques (fig. 21).
Small animals in jade are another category of object much influenced by the foreign lion motif.
Like other jades, their origins owe something to imports from the borderlands. The tomb of a Chu
king at Shizishan, Xuzhou, contained an intriguing stone carving of a feline, a tiger or possibly a
leopard (fig. 29a). At first sight unremarkable, the carving has two important features: a flat platform
on which it sits and a loop behind its neck. The same features also occur on an equally unusual gold
feline from Xuyi in Jiangsu (fig. 29b). Both felines seem to reproduce characteristics of well-known
Western Asian bronze weights, illustrated here by an Achaemenid example from Susa (fig. 3o). It is
quite likely that such weights travelled eastward with merchants. The Iranian or Western Asian lion
had simultaneously been adapted and adopted also in Central Asia, where we see a related creature on

'59 Houma report, vol. I, 192, fig. 97; 253, fig. 140:4.

5I

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
an ivory scabbard from a site near the Oxus (fig. 3I). This lion shares with the beast from Shizishan
bands of fine incised striations and incised loops in a tear-drop shape, reproducing the muscles shown
in undulating curves on the Achaemenid weight. In due course, the Chinese, removed the felines
from their bases (as it were) and carved them as if freely moving creatures. Such jade carvings (fig. 32)
were probably made in small sets, like the very fine group from near the tomb of Yuan Di (reigned
48-33 B.C.) in Shaanxi.I6o
Stone may have been regarded as a lesser form of jade and, as it was infinitely more available, it
could be used for massive items. Moreover, just as jade utensils might be more durable and more
potent than lacquer ones, so too stone coffins and tombs might be more potent than wood and
lacquer ones. The functions and associations of the tomb were thus set both by the way in which the
underground chamber presented a whole universe and by the materials in which that universe was
made. Han enthusiasm for jade, both in burials and in description in ritual texts, probably gave this
fine translucent material a permanent boost within Chinese cultural values.

SPIRIT LANDSCAPES AND OTHER PICTURES

So far we have considered tombs and their contents as providing literally a who
owners. Within the greater whole of the tomb might be smaller images of the
tioned the rock-carved tombs were, probably, plastered and painted, for a few such
survived at Yongcheng in Henan province and at the tomb of the King of Nan Y
We also have many examples of carved stone or brick tombs with extensive "picture
practice of presenting varied images within a tomb seems to have come from C
impression may simply be a result of the uneven preservation of organic material i

China.I6I
Some of the most famous depictions of the spirit realms of the universe appear on the coffins of the
lady Dai, buried in tomb Mr at Mawangdui in Hunan province. Such designs fall within the lineage
of southern coffin painting, also seen in the tomb of the Zeng Hou Yi. As mentioned above, the outer
layer of the Mawangdui inner coffins is black and shows strange spirit-like creatures devouring
snakes. The middle inner coffin in red displays different scenes of dragons and other creatures that
were presumably auspicious. These, as with the jade creatures, combine typically Chu features with
"foreign" ones. On the lid, tigers attack dragons and are further examples of the Chinese versions of
the animal-in-combat motifs from Siberia or Western Asia. A leaping horse on one of the sides of this
middle coffin is contorted in a manner seen particularly at Pazyryk in the Altai, and later exploited

on China's borders in gilt bronze ornaments.I6z It is closely related to the creature of the horse
frontlet from the tomb of the King of Nan Yue, mentioned above (fig. 22).I 3 A rectilinear scroll on
the other long side of the coffin is what has been identified by present-day archaeologists as a cloud

i6o Rawson, Chinesejade, 351. These pieces are of several dates.

161 This development is anticipated in the tomb of the Marquis Yi of Zeng. See Zeng Hou Yi report, vol. I, figs. 216-1I7.
i6z So and Bunker, no. 78.
63 Both Jenny So, "The Waning of the Bronze Age: The Western Han Period (2o6 B.C. - A.D. 8)," in Wen Fong, ed., The Great Bronze
Age of China, An Exhibition from the People's Republic of China (New York: Metropolitan Museum of Art, 1980), 323-27; and Tu
Cheng-sheng, "Ouya caoyuan dongwu wenshi yu Zhongguo gudai beifang minzu zhi kaocha," Zhongyang yanjiuyuan lishi yuyuan
yanjuisuo jikan 64, part 2 (1993), 231I-408, have noted the influence of forms from the steppe or Siberian area in the lacquer painting
of this coffin.

52

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
scroll, or yunqi.'64 In this angular form it stands half-way between the straight, almost criss-cross
patterns of the third century B.C. used on Chu lacquers, and the much more flowing scroll, typical of
many second- to first-century B.C. items in metalwork and ceramic as well as lacquer. Although the
motif had started off as typical of Chu, it survived and developed because it was capable of being
exploited in new forms.65 The scroll could be shaped in three-dimensions as mountains on
decorative fittings and, more spectacularly, as hill-shaped incense burners. On a very fine incense
burner from Liu Sheng's tomb the scrolling lines appear around the sides of the piece as well as being
embodied in the towering crags of the lid (fig. 33). The other element, an array of strange and won-
derful beasts, came primarily from the borderlands of the north but also from the south. On some
incense burners, including that of Dou Wan, the scenes closely resemble the "animal-style" plaques
from present-day Inner Mongolia.
The most spectacular examples of these landscapes are inlaid tubes of bronze embellished with
gold, silver, and precious stones.166 An inlaid bronze chariot fitting from an Imperial tomb at Ding
Xian in Hebei (fig. 34) has been widely discussed, most especially by Wu Hung, who has identified
the animals on it as the xiangrui, creatures whose sightings were thought to be good omens.167
However, although some of the creatures, when seen singly or brought as tribute to the emperors of
their day, may have been interpreted as xiangrui, a scene such as this, with a whole multitude of
creatures, is much closer to the cosmic quality of the poetry genre known as fu. 6 Both the Chuci and
fu describe not just local landscapes, but a universe peopled by spirits and wonderful creatures:

...a mountain which winds and twists upward,

Rearing its lofty crags on high
Covered with ragged jutting peaks
That blot out the sun and moon

And entangle them in their folds; ...

A mass of hues, glowing and shining,
Sparkling like the scales of a dragon ...
Northward rise dense forests and giant trees -
In their branches live apes, gibbons, and languers,
Phoenixes, peacocks, and pheasants,
Flying lizards and lemurs
Beneath their shade prowl white tigers and black panthers.i69

While such poems are occasionally treated as courtly rhetorical flourishes, they probably had more in
common with ritual or political recitations. It is likely, also, that these represent traces of beliefs that

'64 This fairly stereotyped design should not necessarily be identified with the cloud patterns described in omen texts. See Michael
Loewe, "Oracles." in Divination, 193-94.

,65 Large numbers of pieces have come from Yangzhou. See in particular Wenwu 1980.3, I-1o, color pl.; and Wenwu 1988.2, I9-43.
166 There are two possible precedents for these landscape scenes, neither of them very close. First incised designs of people and
creatures shown in a landscape on bronze (Kaogu xuebao 1988.2, I89-232) and rather rudimentary landscapes known in the steppe
regions (Esther Jacobson, "Mountains and Nomads: A Reconsideration of the Origins of Chinese Landscape Representation,"
Bulletin of the Museum of Far East Antiquities 57, (1985), 133-80.
167 Wu Hung, "A Sanpan Shan Chariot Ornament and the Xiangrui Design in Western Han Art," Archives of Asian Art 37, (1984),
38-59.

168 Discussed in J. P. Dieny, Aux origines de la poesie classique en Chine rtude sur la poesie lyrique a l'ipoque des Han (Leiden: E.J. Brill,
1968); and Knechtges, "The Emperor and Literature." See also Michele Pirazzoli-t'Serstevens, The Han Civilization of China
(Oxford: Phaidon Press, 1982), Ioz-o6.
169 Watson, Records, vol. 2, 302oz-o4; Shi ji, 3004.

53

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
dreams and visions, sometimes stimulated by drugs or wine, could create the spirit worlds for those
who were engaged by the poems and their visions.'70 o The Chinese of this time seem to have regarded
the events and figures of dreams, contemplation of spirits and miraculous creatures in trances, as
bringing those beings into their lives, for good or evil.
Filling the landscape on the tube are many strange entities, including a winged horse, and at least
two examples of tigers attacking another beast. Deer, bears and birds are faithfully rendered. In
addition, deliberately prominent are a dragon, a rider shooting behind himself over the rear of a
horse, an elephant, a camel, and a plumed bird, possibly a peacock. Many of the creatures either have
exotic associations or were adapted from objects that came from further west or south. The animals in
combat, deer and flying horses are found in the steppe area, south Siberia and in Iran and
Mesopotamia. In Western Asia the archer shooting behind himself is a very rare design, but is seen
on fine ivories from the Oxus region of Central Asia.7' Bears, as mentioned, were an unusual
innovation of the Han, in which the ruff of the lion's mane became a ruff of fur around the head of a
bear.r72

Hill censers (fig. 33) provide a very explicit presentation of a spirit realm located on craggy islands,
for they seem to create the islands of Penglai, which rose out of the sea, but when approached
disappeared again."73 Immortals on the islands were thought to cultivate herbs that would confer
non-death. Both the scenes on the tube and those on the hill censers are examples of the ways in
which the Han exploited southern or eastern beliefs and practices and elaborated them with
references to the fantastic creatures of the borderlands.

Like the tombs and models of palaces mentioned earlier, the landscapes depicted on Western Han
artefacts, and the fu that seem to describe them, all convey the impression of a world seen from the
point of view of a spirit or immortal journeying through the clouds, as described in such poems as the
Yuan you in the Chuci. It would seem that the unification of the Empire had brought within reach of
the court a concern with the full sweep of the universe and all that lay within it.

CONCLUSION

The great tombs and the miniature landscapes had similar purposes; namely to reali
of the universe and to make that potential manifest for the benefit of the Liu fam
associates. All aspects of this universe, the palaces of the living and the tombs of t
to a single continuum. To give these notions of the universe visual, concrete forms
craftsmen employed by their courts drew on beliefs, practices, artefacts and motifs

'70 See Donald Harper, "Wang Yen-shou's Nightmare Poem," HarvardJournal ofAsiatic Studies47, (1987), 23
171 OxUs, 2oooJahre Kunst am Oxus - Fluss in Mittelasien. Neue Funde aus der Sowjetrepublik Tadschikistan (
1989), no. 22.
I72 For typical bear-shaped feet, see Shiqiao, Xuzhou, Wenwu 1984.11, 22-40, fig.62: 3-4. DeWoskin, Doctors, Diviners, and Magicians
of Ancient China: Biographies of Fang-shih (New York: Columbia University Press, I983), I; Introduction, refers to the use of bear-
skin as a mask of a fang shi; see also Bodde, 77-78.
'73 For discussions of hill censers see Susan Erickson, "Boshanlu - Mountain Censers of the Western Han Period: A Typological and
Iconographic Analysis," Archives of Asian Art 45, (1992), 6-28; and Munakata, I991, 2o-34. Both authors, and many others, suggest
that the hill censors may represent an axis mundi, as a link between the world of humans and that of the spirits. I argue, however,
that the complex ways in which these censers are held on narrow stems by strange and miraculous creatures may indicate that the
mountains are rising up out of the sea and that they present Penglai.

54

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
that they dominated or that were adjacent to them. As a result of the consolidation of the Empire, a
vast and rich array was to hand. Indeed, without the unification of the Empire, this vivid depiction of
the universe could not have been presented, for the conquests in the south and east and the extensive
relationships with the borderlands brought new objects and motifs into metropolitan Han China.
These new objects were frequently treated as belonging to the supernatural. But although super-
natural perhaps, they belonged not to a foreign outside world or worlds, but were part of the single
great universe that the emperors sought to control and exploit.
The Qin and the Han were consolidating a process that had started many centuries earlier, but
with the strong Imperial support, new beliefs and new object categories came to occupy a central
position in Chinese attitudes to the universe. As a consequence many traditional objects and beliefs
were edged out. What is particularly striking is the uniformity of the new sets of objects employed,
presumably as new rites and rituals ousted the much more diverse range of possessions of the rulers of
the Warring States.
This change has been insufficiently noted because the ancient systems of classifications of ritual
vessels and ritual jades constructed by scholars from the Han period onwards were not sensitive to the
changes described. Scholars of the Song period, and later, did not fully take note of or adequately
describe the many objects considered here, simply because these do not belong to the major object
types regarded as part of the ancient inheritance from the Zhou. For this reason, one of the most
profound changes in ancient Chinese culture has been almost overlooked. The change was, however,
fundamental and provokes further questions. For it should lead us to ask whether fragments of Zhou
practice and beliefs could have survived such an upheaval. If this survival is in doubt, we need to
reassess also the revival of the traditional rites that preoccupied the Chinese in the later Han period.
Further, despite a return to what was thought to be ancient practices, the attitudes to the spirit
world and the afterlife described here were to remain essential elements of Chinese rites and beliefs at
all levels of society. While such beliefs are often regarded as belonging to popular religion, it seems
likely that this designation is a misnomer. The tombs of the elite from the Han to the Qing show a
continuous allegiance to this system of beliefs. Indeed, people at all levels of society seem to have
espoused two, somewhat divergent, religious systems: veneration of the ancestors and worship of a
wide-range of deities, to whose cult belonged also the complex preparations for the afterlife presented
in tombs. Of course, the emphasis on one or other of these systems varied from period to period and
place to place. They were also complicated by Buddhism, when introduced, and by the formal
elaboration of what is now known as religious Daoism. Notwithstanding many later changes, the
developments described in this paper made a fundamental contribution to the plural and parallel
systems of practices and beliefs that became an enduring feature of Chinese culture.

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
 GLOSSARY

Anhui gg Guishan & 01

Baicun - j
Guo Ji Zi bai pan @,a- - ?
Baoshan N-0 1 Guo Yinqiang rP[N
Han &
Baoshan Chu
Beidongshan J r mu
0A 7 _ Han bihuamu M j Ej
bi M Handai sangzangzhidu $ $1J
bixie ' RF
Handai
fr 4 Iwuzhi
9 wenhua ziliao tushuo M {4 J..
Cangshan
Changfeng-I A "Han huaxiangshi di fenqu yu fenqi yanjiu" } {
Changjiang &:RT
Changli f g Han Wudi AR;
Haoli 1 t
Changping :
Changsha & y
Hayashi Minao B R ;
Changsha Mawangdui yi hao Han mu i , E Henan MA]

Heyaitou
Houma tiME .
zhutongyizhi f -, '
Changzi A?:m
Ch'ih-yu A ) hsiang
Ch'in 3 Hstian-yuan $7:
Chu V hu
Chu bo shu r%6 = Huainanzi & T
Chuci VV Huangdi W
Chiieh-ti fAg Huanglao w
cong ?, Huang Minglan w7 HA r

Cuomu Zhanguo Zhongshanguo guowang zhi mu Wr Huang Ti W

R m Hubei A )
Dai # Huang
Dan YT
Di huang
hun A.
Di R") Hunan ]
Dian g Huo Qubing rjTN
ding gng Jiangling j AZ1
fianglingMashan yihao Chumu i , I ~ - M
Ding xian I ,
dou U fiangling Wangshan Shazhong Chu mu F 0A [
Dou Wan

Fangmatan ),T g Jiangsu ?_

"Fangmatan jian zhong de zhiguai gushi" )R,4gg Jiaotong daxue 2 AJ
Jin

fangshi Jti Ji'nan MA4

Fang-Shih Ij Jingdi '

Feng and Shan ,jl'-_ Jing Gong '_a

Fenghuangshan [- -- Jinguo [
Jiyuan ` ~i
fu g
Fuxi R Jundushan
Kaogu & $@[_l
Gansu -'1
Kaogu leixing xue di lilun yu shiji & EES t. I F f_
Guangdong g
Guangzhou ) l' Kaogu xuebao '

56

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
Kaoguxuejikan A fl
Kaoguyu wenwu & 2 J Penglai ATA
kuan
kuei men t r i Pingshan
po R C-
Kunlun N Pu Muzhou ~Ml\i
Langjiazhuang P?* )j
qi ,-
Qi *
Laozi T f
Qin
Liang gin n
Liang Han wenhua yanjiu T AR @R f ) Qin Shi Huangdi 9 & :
Liangzhu
Li Ling 'A.M
4 Qin Shihuangling bingmayong keng 9 R 3,,

Li Rusen $9 ,
LiXueqin $ Qin
QingShihuanglingyanjiu 9 R_
Li Yinde ~,
Qufu A_
Qufu Luguo gucheng $O % . M iA
"Lingmu zhidu he linghun guan" ~I $r gJ qi F _?. Rong A
Lintong ~?i
Sanpan
ShaanxiShan
Rf - NR_0
Linyi ,F p
Linyi Fenghuangling Dong Zhou mu g, g- g' Shandong
Shangn O_+,
Linzi @ M
Linzi Shangwang mudi 2 Av I g -M Shan glin _1
Shangwang
Liu Bang Wq9g
Shanhaijing C ,,
Shao-weng i
Liujiadianzi WiJ * -r
Liu Mai WlJ shenming F i$l
Liu Sheng glJW
Liu Zhu WIJ `if
Shen Shu iF--$,
Shiji
LuI shihch'ii -l
Shiqiao
"Lun Zhongguo gudai genqiumudi ;E"
chansheng yu

fazhan" 'F TW- gf-JO

Luoyang ?M A
. _Oj-I
Shizishan JE _
Luoyang Hanmu bihua f#Shunja
ME Sengokujidai seidoki no kenkya i 8 ,
Liishi chunqiu R shuowen jiezi R Zsqj-:

Ma Chengyuan ,g 7g Yin Sha seidoki soran san )R JJ $@ $ ,--

Maoling N F Sichuan E[]JI [

Mancheng Han mu fajue baogao

Sima Qian IR,? ) A "
Sima Xiangru I , FAV0
Mawangdui ,If
ming Be Sofukawa Hiroshi l[[.
Song i-.Ch'ien -4 I
Ssu-ma
Miu Ji a
mou Sun Ji ,,

Muzangyu shengsi: ZhongguoSunzi Bingfa ?W

gudai zong iao zhi

shengsi
Nan Yue g
_: Taishan . 2 - .
Taiyi A-@ _k
g
Nanyue Wang T:q:
Taiyuan 7.
Nilwa $803
"Ouya caoyuan dongwu wenshi yu Zhongguo Taiyuan
Tao
gudai
M Jinguo Zhao qing mu AS @; J} $1 & _
taotie W
beifang minzu zhi kaocha" B- .__@, .

57

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms
Teng xian P%,
Yi Y-
Yi fing AW
Tianxingguan ) _ & Yin k
T'oungPao
Tu $_ TF EA
Cheng-sheng Yi'nan VT A]
Tu-shuo ft ?A Yi'nan gu huaxiangshi mufajue baogao T- {

Wang Chong
Wang Bi -_6 Ji
? Yishui W7*

Wang Luyu I f, YongchengXi Han Liangguo wanglingyu qinyuan A]

Wang Mang
Wangshan _T
e I- Yuan Di ;V
WangXueli ~~3 Yuanyou M
yue WA
Wang
wen Yen-shou T-_LI

Wen Yu Lu _
yunqi ' -!
Wenbo Z# ' Yunnan SW:
Wenwu :; Z YuWeichao AmME
Wen xuan A; YiU Ying-shih H
Wu Hung ATA
Zeng "f
Wu Rongzeng A Zeng Hou Yi mu L {
Wu Xian Af J-
wuxing ~E~f ZengHuXiantong
Zhao ffih A_
xi m Zhao Hun -B A
Xi'an fi Zhao qing Lgp
"Xian Qin, Han, Jin yaodai yong jin yin daikou" 7t
? * A -4F 1 -5 V M +1 ?IR Irt Y Zhejiang
Zheng &(Itl4_
Xiang
Zheng Yan $[5"
xiang
xiang Fi Zhenmuwenzhong
"Zhenmuwen J,,. A
suojiandaode Dong Han daowu
xiangrui ff guanxi" J -
zhi
Xianyang j-,[
Xiao Tong Xi, Zhongguo diaosu shice. Han fin Nan Beichao mu g

Xichuan Xiasi Chun Qiu Chu mu ffi JII- -f4 4f( V

shidiao yishu f if f _. o (4 4) g
Zhongguo fang shu kao o WW
Xi Han Nan
Xinjiang fif Yue Wang mu fE i _ M
Zhongguo meishu quanji, diaosu bian -P N
Xin Lixiang Ifa'7fl ?

Xinyang Chu mu Mg Zhongguo shenghuo. Zhongguo guwenwu yu Dong-

Xinzheng $5 wenhua jiaoliuzhongde ruogan wenti M 9R),'

Xiongnu
Xu Shen EN:'%' ..V, Zhongguo wenwu jinghua ? Z J*
Xuyi iEfj
Zhongguo yuqi quanji ,
Xuzhou 'fl'[ Zhongshan cP`[II
Yan A Zhongyangyanjiuyuan lishi yuyan yanjiusuo jik
Yang
Zhou J
Yanggong { _3
Yangzhou JiiJ [[ zhu
Zibo igf$
Yangzi ;I -T-"
zun )
Yanqing
Yao i~_ Zuo Zhuan
Yi Z
58

This content downloaded from

204.44.106.3 on Thu, 19 Sep 2024 16:07:02 UTC
All use subject to https://about.jstor.org/terms