The Taotie Reconsidered: Meanings and Functions of the Shang Theriomorphic Imagery

Author(s): Ladislav Kesner

Source: Artibus Asiae, Vol. 51, No. 1/2 (1991), pp. 29-53
Published by: Artibus Asiae Publishers
Stable URL: https://www.jstor.org/stable/3249675
Accessed: 24-11-2018 14:13 UTC

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 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 LADISLAV KESNER

THE TAO TIE RECONSIDERED: MEANINGS AND FUNCTIONS

OF SHANG THERIOMORPHIC IMAGERY

The question of the that

is one of the perplexities meaning
have so farand
avoidedfunction
solution despiteof theriomorphic
massive designsof on ritual objects of Bronze Age China
efforts by several generations
scholars.' Meanwhile, as knowledge of ancient Chinese society expands, the importance of this problem is
perhaps even increasing, for it directly relates to a wide range of topics within the sphere of Early China studies.
Given the ubiquity of such images in early Chinese art, these issues are pertinent to understanding the nature of
visual representation in ancient China, and thereby assume even wider art-historical implications.
In recent years we have witnessed a new surge of interest in this area, and several important studies have
emerged. Some scholars have pursued iconographical theories of Shang bronze designs which place their
meaning in a shamanistic context: K. C. Chang's "Animal in Shang and Chou Bronze Art,"2 and Elizabeth
Childs-Johnson's "The Ancestor Spirit and Animal Mask in Shang Ritual Art."3 On the other hand, Robert
Bagley, in the introductory chapters to the catalog Shang Ritual Bronzes in the Arthur M. Sackler Collections,4
has elaborated an argument denying any iconographic, symbolic meaning to the designs. These two groups of
studies represent the culmination of a long evolving polarity in the study of ancient Chinese art, a new stage in a
tug of war between iconographically versus more formalistically oriented scholarship, a distinction that may be
traced back almost half a century.5 It is my intention here first to discuss some of these current opinions. I
assume that to achieve a more thorough and coherent understanding of the problem, it is necessary to overcome
the particularist views they represent, to move beyond the dichotomy of seeing Shang designs either as iconic
absolutes or mere decoration. One way to do this is to turn attention to the functional context of this imagery,
and that will be the focus of the second part of this paper.

1 The term theriomorphic image is used in this paper to refer to monster mask and mask-and-face-like motifs and dragon motifs on
Shang and Neolithic artifacts. I reserve the term taotie for the monster-mask motif on Shang bronzes and other artifacts.
Robert Bagley, James Cahill, Lothar von Falkenhausen, Lubor Hajek, David Keightley and Nancy Price read earlier drafts of
this paper. I thank them for their critical comments and many helpful suggestions. I am especially grateful to David Keightley for
his encouragement and for allowing me to read his paper "In Clear and In Code: Pre-Classical Roots of the Great Tradition in
China," prepared for the Annual Meeting of the College Art Association (New York, February 199o).
2 Kwang-chih Chang, "Animals in Shang and Chou Bronze Art," Harvard]ournal ofAsiatic Studies (hereafter HJAS) 41.2, 1981,
pp. 527-41; reprinted in slightly different form in his Art, Myth, and Ritual, The Path to Political Authority in Ancient China,
Cambridge, Mass. 1983, chapter 4; also his "Ancient Chinese Civilization, Origins and Characteristics," in 7oo000 Years of
Chinese Civilization, Milano, 1983, pp. 27-44-
3 Elizabeth Childs-Johnson, paper delivered at International Symposium on the Yin-Shang culture of China. Anyang 1987.
Another major recent iconographic study of Shang designs is Hayashi Minao's In shu jidai seidoki moyo no kenkyu-, (Studies on
Bronze Decor in the Yin-Zhou Bronzes). Tokyo 1986, part II of Hayashi's In shu seid6ki moran. (Conspectus of Chinese
Bronzes).
4 Cambridge, Mass. 1987.
5 See Florance Waterbury, Early Chinese Symbols and Literature: Vestiges and Speculations. New York 1942; and Phyllis
Ackermann, Ritual Bronzes ofAncient China. New York 1945; versus Ludwig Bachhofer, A Short History of Chinese Art.New
York 1946, and particularly Max Loehr, Ritual Vessels of Bronze Age China, (his catalogue of an exhibition at Asia House), New
York 1968, Introduction.

29

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 K. C. Chang's theory of meaning of animal designs -
theriomorphic images as representations of shamans' helping agents

Professor Chang's theory, first published in I98I and elaborated two years later, attracted considerable
attention. Its importance lies in the fact that it not only attempted to explain the animal designs, but explicitly
established the strategic role of shamanism in Shang (and even early Zhou) society.6 I must make clear that in
what follows I limit my discussion to Shang bronzes and their designs, on the assumption that to treat Shang
and Zhou material indiscriminately with regard to this issue, means to overlook their significant differences.'
After summarizing the motifs in question, Chang, taking two passages from the Eastern Zhou texts Guoyu
and Zuozhuan as a basis, concludes that "the animal designs on Shang and Chou bronzes are iconographically
meaningful as the images of the various animals that served as the helpers of shamans and shamanesses in the
task of Heaven-Earth, dead-living intercommunication.' His key passage runs as follows:

These documentary data, recalling the passage in Kuo-yu [sic], have enabled us to understand that the
sacrificial animals were the same ones which had the power to help the shamans and the shamanesses in their
communication task, and to make animal offerings was a concrete means ofachieving communication between
Heaven and Earth, the dead and the living (my italics).'

At first glance , this may seem convincing, for the role of animal bones (many of which are of sacrificial
victims), as a means of communication through divination in Shang society is well-known. The images depicted
on bronze vessels, however, are mostly not real animals. Chang addresses this contradiction, after providing
parallels from late Zhou texts on the use of dragons in Heaven-Earth communication as well as ethnoparallels
(which, however, tell more about the general role of animals in the shamanistic context than about Shang art
motifs per se).1o He cites at length Li Ji's observation" on the evolution of such creatures to support the view
that these theriomorphs on bronze vessels were created on the basis of real animals. To close the circle, Chang
makes use of the fact that a common way in which the shaman summons his animal helpers is to make animal
sacrifices, thus releasing the spirits of these beasts to become his agents on his ascent to Heaven. So Chang
implies that the creatures depicted on ritual vessels are the spirits of sacrificial animals brought into existence
through the transformation of real animal shapes. This explanation has a further implication concerning Shang
ritual practice: that a significant portion of Shang sacrifices would have had to be directed toward summoning
such spirits.
The idea that the taotie represents a sacrificial animal has been previously challenged.12 Chang's argument
appears to rest almost solely on Li Ji's observation that "the majority of the animal patterns employed by the
decorative artists of this period, whether carving a stone, casting a bronze, inlaying a wooden article, moulding
a clay object, polishing a piece of jade, had originally an indigenous and naturalistic background."13 On this
basis Chang has concluded that "they [mythological animals] were surely not real-world animals, but
apparently they were transformedfrom naturalisticprototypes of the ox, sheep, tiger and reptiles" (my italics).14
This is, however, a rather over-simplified explanation. There are not many features in bronze designs which
would testify to this process of transfiguration of real animal components. It is true that taotie of the Anyang

6 This view has been since reiterated in several articles: "Ancient China and Its Anthropological Significance," reprinted as part of
his widely-used text book The Archaeology of Ancient China. 4th revised edition. New Haven and London 1986; "The
'Meaning' of Shang Bronze Art," Asian Art 3.2 (Spring I99o), pp. 9-I7. See also notes 59 and 87.
7 On this point see Childs-Johnson's review of R. Bagley's Shang Ritual Bronzes, in Art Bulletin 7I.I (1989), p. 153, note 26.
8 Chang, "Animals," p. 540.
9 Ibid, p. 539-40.
10 It is unfortunate that Chang fails to cite cases to document his thesis that "insofar as the shamanistic task of crossing worlds and
the role of animal helpers is concerned, they are found in oracle-bone inscriptions of the Shang." "Animals," p. 543.
11 Li Chi, "Hunting Records, Faunistic Remains, and Decorative Patterns from the Archaeological Site of Anyang," Bulletin of the
Department of Archaeology and Anthropology, National Taiwan University no.9h/Io (1957), P. 12.
12 Jordan Paper, "The Meaning of the T'ao-T'ieh," History of Religions 18.1 (1978), PP. 25, 36.
13 "Hunting Records," p. 12.
14 "Animals," p. 545.

30

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
period with their wide stylistic range and many variations do show occasional borrowing from the realm of
zoology by transplanting various parts of animals into new combinations. However, this fact alone is by no
means sufficient to explain the origin of either taotie or "dragon". Archaeological data do not provide evidence
of this transfer. What they do show are a few isolated examples of pottery sherds featuring real animal motifs
from Erlitou phase sites,"15 and an even smaller number of them from Erligang phase sites, vastly outnumbered
by bronzes with fabricated designs.16 This evidence does not even permit one to go so far as Alexander Soper's
modest statement that "all of this [i.e., some clay sculptures from Erlitou phase sites], provides, in fact the
general type of quasi-realistic original that Professor Karlgren postulated years ago for his A style."l Much less
acceptable are Li Ji's and similar explanations, still to be found in studies of early Chinese art, which see the
taotie as a development that followed a period of realistic representation, a kind of jigsaw puzzle assembled out
of real animals' parts.8 Instead, however, the discoveries of the last two decades strongly imply that from the
V-IV millenia B. C. onwards there existed two idioms in Chinese art, one which tended toward naturalistic or
quasi-naturalistic representation, the other toward rather geometric reduction; Shang styles may be seen as the
result of interaction and fusion of these two tendencies. Especially the recent finds of jade artifacts from the east
coast Neolithic sites, as will be argued below, suggest that the taotie was not only conceptually but also
stylistically related to the latter of the two above-mentioned idioms.19
Second, there is one more reason to doubt the validity of LiJi's assumptions and those derived from him. It is
hardly possible to expect that artists working within the constraints posed by activities of such great
importance, the maintenance of political authority and social and religious order, would enjoy the freedom to
render these motifs as they wished. Their "imaginative wonderlust" (in Li Ji's words) surely was not only
physically restricted by the frames of a decorative field,2 but also by the requirements of the religious and
political aims they served. This issue, a question of "patronage", becomes one of the key points in discussion of
the function of Shang designs. Of course, the absence of written documents makes this argument somewhat
elusive.21 The problem will be touched upon later; suffice it to say now that we should beware of thinking in
extremes, seeing Shang craftsmen either as totally free artistic personalities, constrained by their creative vision
only, or as mere slavish executors of their patrons' demands.22

15 For Erlitou material see Kaogu I965.5, pl.3:Io, 5:I-3. For examples of more realistic clay sculptures from the Qilipu site see
Kaogu xuebao I960. I, fig. 19:4, 6, 7, pl. 8:5, 6, 8.
16 For Erligang finds see Kaogu Xuebao 1957.1, pl. 6:7-o10; Cultural Relics Work Team of Henan Province Culture Bureau,
Zhengzhou Erligang. Beijing I9y9, pl. 7:8.
17 Alexander C.Soper, "Early, Middle and Late Shang: A Note," Artibus Asiae 28.I1 (1966), p. 24.
8 Such a view was originally probably influenced by Franz Boas's observations on the split-animal style of Northwest American
Indians. Examples are to be found in the writings of Li Ji, "Hunting Records," also "Examples of Pattern Dissolution From The
Archaeological Specimens of Anyang," Artibus Asiae 22.1/I (I959), esp. p. 142. For recent treatment, heavily indebted to LiJi
see Zhang Xiaoguang, "Yinxu qingtongqi de zhuangshiyishu," in: The Institute of Archaeology, CASS, Yinxu qingtongqi.
Beijing 1985, pp. II5-I8, fig. II.
Robert Bagley has criticized attempts to derive taotie on the basis of horn types which in some cases were adapted from real
animals. He rightly remarks that most such attempts have not gone beyond distinguishing between "feline" and "bovine" taotie
(Shang Ritual Bronzes, note 74, P- 52.)
For such attempts see Childs-Johnson, "The Ancestor Spirit," p. 8; Shanghai Museum's Research Group on Bronzes, Shang
Zhou qingtongqi wenshi. Beijing I984, pp. 4-6.
The most serious attempt to derive theriomorphic motifs from parts of real animals is Hayashi's In shu Jidai seidoki monyo no
kenkyui.
19 These jades are discussed below, pp. 36 and 45.
20 "Hunting Records," p. 1t2.
21 The problem here includes the question of the status of bronze artisans, who were certainly not slaves, as they are dubbed in
orthodox Chinese Marxist terminology (see, e.g. Ma Chengyuan, The Ancient Chinese Bronzes. Hong Kong 1986). More
probably, bronze artists and especially foundry masters enjoyed relatively high status.
22 Cf Max Loehr in Ritual Vessels of Bronze Age of China, p. 1i. More recently, Louisa G. Fitzgerald Huber touched on this
problem ("Some Anyang Royal Bronzes: Remarks on Shang Bronze Decor," in: George Kuwayama, ed., Great Bronze Age of
China. A Symposium. Los Angeles i983, p. 16). Mentioning the fact that there is virtually no Shang inscriptional evidence related
to the meaning of animal decor, she observes that these lacunae suggest relative autonomy of the artists in making the designs. I
argue, however, that the silence of oracle-bone inscriptions on these topics is inconclusive. It is most probable that the patron-
client relations between Shang elites and artists were handled in other ways, which did not enter the divinational record.

31

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 The Human-Animal Relationship

Professor Chang continues his discussion by explaining the man-beast relationship found on some Shang
bronze artifacts (fig. i). Following earlier suggestions by Douglas Fraser et al,23 Chang, in my view quite
correctly, interprets the relationship of beast and man on seven artifacts as one of closeness and not of
antagonism. Using further ethnoparallels, he takes the animal's gaping mouth as an archetypal symbol of the
separation of the worlds of dead and living, and concludes that "the combination of the shamanistic image, his
helping animals, and the gaping mouth as the source of the lifting wind on one and the same bronze vessel
depicts the communicating act or even causes that act-in its most complete form."24 Seductive as this
explanation seems to be it is again, I believe, untenable.25
Let us start with the extreme scarcity of this motif in Shang art: the author in fact cites no more than seven
examples, datable to the Shang period, to illustrate his point. Even when several pieces featuring so-called
monster-mask headgear are added, the amount is still not impressive. Such negative evidence does not of course
in itself prove anything, but it leads one to ask: why this supposed communicative act was so rarely represented,
amidst thousands of known Shang bronzes, given its alleged central importance in Shang religion and
rulership? Moreover, the question may be raised of the legitimacy of Chang's choice of his seven pieces. The
group appears to be a product of rather cursory selection. It can be easily observed that only you vessels and the
Funan zun display the fully-fledged man-beast motif, while on the handles of the Si Mu Wufangding and an axe
from the Fu Hao tomb, a head only is shown.26
Upon observing the artifacts drawn together, and especially the key pair of you from the Sumitomo and
Cernuschi collections (fig. i) one is struck by the rendering of both beast and human figure which in Shang
terms at least exhibit a fair degree of realism. Human images are relatively scarce in Shang art and there has been
little investigation into the problem of what they represent. Apart from so-called demonic images,27 there are
just human figures of jade and a few examples of more naturalistic ceramic examples from controlled
excavations, and these have been mostly regarded as slaves or retainers (fig. 2).28 It is also pertinent to recall
William Watson's suggestion that a more realistic style existed in Shang alongside the prevailing hieratic idiom,
which was used to depict sacrificial victims and prisoners.29 It is thus reasonable to ask whether a spiritual man-
beast relationship, if depicted at all, would have been rendered in such a naturalizing mode and whether the
ordered, hierarchical Shang world-view30 would have permitted indiscriminate depicting of human strata, all
the way from slaves and retainers up to shamans and kings.31
Apart from these remarks based on the art objects themselves, there are further points to be considered.
Chang supports his view of the man-beast motif as the image of the shaman and his helper by mentioning the

23 Douglas Fraser, ed., Early Chinese Art and the Pacific Basin. A Photographic Exhibition. New York 1968, p. 65.
24 Chang, "Animals," p. 549; "Ancient Chinese Civilization," p. 41.
25 For another, no more convincing interpretation of you vessels with the human-beast interaction configuration as a representa-
tion from Shanhaijing about a god on the Dushuo mountain who fed evil spirits to tigers see Ma Chengyuan, op. cit. p. 46. Sarah
Allen, "Myth and Meaning in Shang Bronze Motifs," Early China 11/12 (98 5-87), p. 286 which suggests that the configuration
is an "allusion to the passage of death , the cult of which is, after all, the central concern of Shang ritual."
26 Those pieces are profusely illustrated, see e. g. Bagley, Shang Ritual Bronzes, fig. 80 (Funan zun), 133 (Si Mu Wu Fangding), i18
( Fu Hao yue).
For a close detail of the figure on the Cernuschi/Sumitomo you vessels see Hayashi, In Sh fidai, p. 3o10, fig. I-34.
27 Doris Dohrenwend, "Jade Demonic Images in Ancient China," Ars Orientalis o10 (I975), PP. 5 5 -78 assembled many of those
types.
2 For Shang jade humans see, e.g. Institute of Archaeology, CASS, Yinxu yiqu. Beijing 1981, figs. 98- io8.
29 William Watson, Styles in the Arts of China. London 1974, PP. 28-29, 55-56.
3 For this characteristic of Shang ancestor worship see David N. Keightley, Sources of Shang History. The Oracle-Bone
Inscriptions of Bronze Age China, 2nd ed. Berkeley I985, p. 137; idem, "The Religious Commitment: Shang Theology and the
Genesis of Chinese Political Culture," History of Religions 17. 3-4 (1978), pp.211 - 25-
31 Chen Zhida, "Yinxu wangshi yuqi yuyu shirenwu diaoxiang," Wenwu 1982.I2, p. 85-6; Wu Hong, "Yi zu zaoqi de yushi
diaoke," Meishu yanjiu I979.I, pp. 64-70 suggested that some jade figures might be taken to represent kings or members of the
ruling nobility.

32

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
bin ritual.32 He suggests that by such means humans (shamans, kings) were able to go upward to the realm of the
deities. Chang is obviously willing to see Shang divination as an act of shamanism.33 However, contemporary
oracle-bone specialists seem to be far more reserved on this point. As David Keightley has pointed out:
"...Shang divination appears to have involved no shamanistic flight to other realms;" communication between
men and ancestors "required no disruption of ordinary modes of existence."34 Furthermore, in his recent study
of the bin ritual Keightley has concerned himself especially with the question of whether a living king ever
visited his dead ancestors on high, convincingly arguing that the bin was a bureaucratic ritual, during which the
king, rather than flying on high to visit dead ancestors and spirits, entertained them on the ground.35
Thus the study of Shang religion from an inscriptional perspective brings little if any support for Chang's
interpretation of the man-beast interaction motif.
To conclude, there is one fact that seems to offer at least partial explanation. Four of the seven objects
illustrated by Professor Chang, including the key pair of you vessels, are said to have come from non-
metropolitan areas in south-central China.36 All of them represent the whole human body, as opposed to heads
sandwiched between monsters on just two artifacts of presumed Anyang provenance. As Robert Bagley has
rightly observed, the notion of actual interaction between man and beast (whether protective or threatening) is
much more pronounced in those southern pieces, where the combination is treated as though it had great
significance.3' Thus it follows that the man-beast motif on these artifacts may reflect some kind of regional
tradition, varying from the prevailing Shang religious outlook, perhaps with stronger signs of shamanistic-
oriented practices.38

Dualistic nature of the Taotie

Finally, I address Professor Chang's explanation of the dualistic nature of the taotie. He suggests that both
the traditional views of the final image resulting from the splitting of a single unit, as well as from the joining of
two animal profiles, are possible, and not necessarily mutually exclusive. He seems to prefer the view that "the
artistic dualism on Shang bronzes is merely a component, a link, of dualism that permeated Shang institutions
and Shang thought." He has furthermore linked his dualistic character of bronze imagery to the concept of a
dualistic division of the Shang royal lineage.39
Such a hypothesis, however, finds little support in the evidence. First, it must be pointed out that the very
basis on which this assumption rests, i.e. the theory of division of the royal lineage into two groups which
alternated in holding the rulership, is not universally accepted.4 Second, and even more important, the
phrasing of a dualistic interpretation of the taotie itself must be revised. The long-persistent notion of an image
resulting from splitting one animal appears to be trivialized in the light of Robert Bagley's recent analysis.
Bagley is certainly right in observing that in most cases the strictly frontal, coherent nature of the image
predominates, so that the double reading of the taotie is a curiosity of limited significance, a visual pun, not

32 Chang, Art, Myth, and Ritual, p. 54-5 5.

33 Ibid., pp. 54-5 5, I10
34 Keightley, "Late Shang Divination: The Magico-Religious Legacy," Journal of the American Academy of Religions Thematic
Studies 50/2 (1984), p. 20.
35 Idem., "Royal Shamanism in the Shang: Archaic Vestige or Central Reality?" paper delivered at the Workshop on Chinese
Divination and Portent Interpretation , Berkeley 1983, esp. pp. 16-29.
36 Shi Yan, Zhongguo diaosuo shi tulu. Shanghai I983, vol.I, p. 30.
31 Shang Ritual Bronzes, p. 3 5.
38 Several authors have proposed on the basis of the differences seen on bronze artifacts found in Yangzi area, that they reflect
cultural variations. For an early formulation see Virginia C. Kane, "The Independent Bronze Industries in the South of China
Contemporary with the Shang and Western Chou Dynasties," Archives ofAsian Art 28 (i974-75), p. 85. See also Bagley, Shang
Ritual Bronzes, p. 34.
39 Chang, "Animals," p. 5 54.
4 Chang, Shang Civilization. New Haven and London i980, p. 183, note 65. Keightley, "Shang China is Coming of Age -A
Review Article," Journal of Asian Studies 41.3 (1982), pp. 552-53.

33

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
inherent in any property of the motif.41 Though this particular interpretation of K. C. Chang is obviously far-
fetched, the issue should not be easily dismissed. Dualism clearly was an important part of the Shang Zeitgeist
and it remains yet to be determined how certain dualistic principles in visual imagery may have related to other
features of Shang society.
Professor Chang's shamanistic explanation of Shang designs was earlier anticipated by Jordan Paper, who in
a I978 article concluded that the "taotie mask derived from a worn helmet signifying power and authority, or
mask derived from a worn mask symbolizing the spirit of the dead to whom sacrifices were offered."42 More
recently Elizabeth Childs-Johnson again has concentrated on the mask aspect of the taotie. In "The Ancestor
Spirit and Animal Mask in Shang Ritual Art", she puts forward a theory that the taotie is a gui mask, the artistic
symbol of the ancestor spirit in the guise of a wild animal mask, and as such a symbol of the Shang king's access
to the supernatural realm of his ancestors.43
In all these iconographic theories there appear some recurring themes that should be questioned. First,
implicit in all such treatments of the Shang monster is the notion that Shang art was basically iconographic, in a
sense that artifacts and motifs did represent certain concrete beings, concepts or even acts (the you vessel as an
illustration of an act of shaman/animal interaction). But, as will be further argued in the closing part of this
paper, we should certainly beware of automatically accepting such a conception of Shang imagery.
Second, both authors, in putting forward general theories of the meanings of bronze designs, have built on
features which were not widespread or inherent to the majority of these designs. The case of K.C. Chang's
man-beast motif has been already mentioned. In a similar vein, Elizabeth Childs-Johnson builds her theory
around the specific rendering of an animal monster mask as it appears with the addition of a cicada body." But
in fact, though more frequently encountered than the man-beast motif, the mask-cicada likewise forms only a
tiny minority among varieties of animal masks. In both cases the authors have drawn far-reaching conclusions
on the basis of untypical examples. The explanations which might have relevance for just these variations are
extrapolated to cover all taotie.
Third, and most important, the troubling point of these iconographic theories seems to be the lack of any
temporal dimensions. Very little or no attention is paid to the historical development of the bronze designs
themselves. But the artifacts and their ornaments must remain the primary base on which to build, and against
which to check any theories of meaning and function.

The Taotie as ornamental pattern -

Robert Bagley's interpretation of theriomorphic imagery

A detailed stylistic analysis of Shang bronzes, based primarily on the development of their designs, has been
recently provided by Robert Bagley in Shang Ritual Bronzes in the Arthur M. Sackler Collections. Bagley's
assumptions about taotie and "dragon" motifs are in most respects antithetical to those just reviewed. He
concentrates fully and exclusively on formal development and technique, leaving almost totally aside the
framework within which the bronze art existed. The key point for Bagley is the fact of "intimate relationship
between design and technique in Shang bronze casting."45 Theoretically supported by technically-biased art-
historians and historians of technology, Bagley is emphatic in stressing again and again through his introduc-

41 Shang Ritual Bronzes, p. 21.

42 See note 12 above.
43 Childs-Johnson, "Ancestor Spirit," passim.
44 Ibid., p. 8-9. A typical example of mask-cicada image, also illustrated by Childs-Johnson, may be found on an owl-shaped zun
from the Fu Hao tomb (Institute of Archaeology CASS, Yinxu Fu Hao mu. Beijing i980, fig. 36, p. 5 5.)
45 See esp.introductory sections, pp. 17-18.

34

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
tory essays the decisive influence of the casting process on the origin and evolution of designs, and the ultimate
contribution of artisan-casters.46 From this premise, his other arguments and assumptions evolve:
-Bronze designs were not an imitation of ornaments conceived earlier in other media.
-There is a basic distinction between taotie and "dragon" as pattern units versus images, the latter coming
into existence only with the introduction of image-ground distinction modes (the Loehr IV and V styles).
-A subordination of motifs to decorative schemes ("active role of compositional scheme" versus "passive
role of motif").
Thus for Bagley the whole question of the meaning of animal designs is wrongly formulated; characteristi-
cally, he discusses it most fully only in a footnote. Dismissing earlier attempts to interpret bronze decorations
and mentioning Ernest Gombrich's "autonomy of decorative design" Bagley asks:

"Is symbolic meaning really the only possible explanation for the decoration of Shang bronzes? By insisting
on an explanation in symbolic terms we may be posing an irrelevant question and ignoring questions to which
answers might be found... We should beware of assuming that nothing but symbolic meaning can explain the
use of decoration on ritual objects... In focusing on the invention and development of motifs and decor
schemes, the present study attempts to explain why the bronzes look as they do. Symbolic meaning does not
seem to play any necessary role in this explanation."47

Bagley must be praised for the consistency and logic with which he pursues his vision of Shang bronzes as
great art. Yet, despite his sensitive diagnoses, it is precisely his explanation of why bronzes look as they do that
leaves something to be desired. Here I refer to his rather over-simple conception of stylistic evolution. Bagley
asserts that it was the artisans who are to be credited for inventing and elaborating designs, and that the point of
view of the artists "offers the only insight into stylistic change and continuity that we can have."' Though he
avoids describing the sequence of decorative styles simply as a function of progress in piece-mold casting, the
technique, the foundry's use of production methods, clearly emerges as a decisive factor in the evolution of
decorative styles. But are the factors which accounted for the rise and development of such a complex art form
really to be sought exclusively within the bronze foundries, as Bagley would have us believe?
Technical and technological factors certainly were an important force in the development of ancient Chinese
art of bronze casting as well as jade carving or ceramic sculpture, and certainly sound art-historians should
"take seriously the processes by which the Shang artist executed his designs."49 It is, nevertheless, this kind of
technological premise implicit in Bagley's treatment at the expense of other factors that seriously distorts the
understanding of this art form. Few would nowadays agree with the notion of an art form as a sheer product of
creative imagination and technological processes, independent of the work of art and social reality, as
exemplified by Henri Focillon whom Bagley quotes.50 Most art-historians and archaeologists alike would agree
that any study of stylistic evolution has to take into account not only technique, but also content and context.51
Art styles always carry some social references, and this is undoubtedly even more so in the case of Bronze Age
ritual art. Both meaning and function of the work of art are among the constitutive elements of visual style, and
it is impossible to arrive at any proper understanding of style without taking them into account. Once it is
shown that technique was not the most important, let alone the only force behind stylistic change, other factors

46 Ibid., p. 18, Bagley states: "invented by casters, the bronze decoration was never conceived as something to be added by a
process separate from casting... The structures thus enforced by the casting method constitute the chief contribution of the
technology to the art form, for it embodies the constraints within which the bronze decor was invented and elaborated."
47 Ibid., p. 50, note 47.
48 Ibid., p. 38.
49 Ibid., p. 38.
50 Henri Focillon, The Life of Form in the Art. New Haven 1942, cited by Bagley, Shang Ritual Bronzes, p. 38.
51 Martin Powers has recently demonstrated in his study of the Wu family shrines that the style cannot be explained exclusively
through references to technique and material: "Pictorial Art and Its Public in Early Imperial China," Art History 7.2 (1984), pp.
135-63, esp. 142.

35

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
have to be considered, especially the functions of ritual bronzes and their designs.52 Bagley's rejection of
symbolic meaning in Shang bronze designs blocks the way not only to an understanding of the stylistic
development of this art form, but also to understanding the full cultural significance of these objects in Shang
society. His assertions that an art devoted to religion need not involve religious symbolism, and that the
purpose of decoration might have been merely to declare the importance of the object seem singularly ill-fitted
to the context of early Chinese art. Once we fully recognize the theocratic nature of the Shang state, the
pervasive presence of bronzes in this culture and the key role of theriomorphic imagery in bronze vessels'
designs, the answers to the questions of meaning and function become readily apparent.

Origins of the Taotie

The earliest prototheriomorphic mask and face-like designs that may be pointed out as precursors or earliest
versions of the taotie, executed in thread-relief line (Loehr I style), appear on very few vessels of archaic type,
attributable to the Erligang period (fig. 3).53 But more coherent, more clearly-defined images, both in thin relief
line and wide band follow in short chronological succession, and the Loehr notion of the absolute priority of
style I before style II is now being altered.54 The question still persists, however, of how the presence of these
Erligang taotie images can be accounted for. The only reasonable solution is to view the first Shang bronze
designs as a part of the whole spectrum of mask and face-like motifs executed in all media. The links between
Neolithic jade imagery and early Shang bronze designs have been earlier suggested by several scholars;55 some
Chinese archaeologists even routinely designate Neolithic mask and face-like designs as taotie.56
To be sure, a minority opinion does not accept Neolithic imagery as the source of Shang bronze designs.
Thus Robert Bagley, following his earlier suggestion, continues to distinguish between what he calls a "non-
representational" eye-motif (the earliest versions of the taotie on Erligang phase vessels) and representational
images with slanted canthi (such as the face on a jade baton from Erlitou).57 However, as we shall see, the links
between the designs on Neolithic jades and those on early bronzes are too substantial to be neglected.
Of all Neolithic jade-producing cultures, it is that of the third millenium B. C.E. Liangzhu culture where
mask-like imagery was most widespread, and whose links with Shang therefore deserve particular attention.
Recently published finds have substantially broadened the spectrum of this Liangzhu tradition with new, more
varied and more elaborate depictions.58 It is evidently here that the essential features of theriomorphic face-

52 A more sensitive approach to the development of Shang styles can be found in William Watson's: "Lavish official patronage and
direct service to ritual needs at Xiaotun... would account in the first place for the extraordinary unity of design and logic of
stylistic evolution that distinguishes work excavated at Xiaotun or attributed to that site..." (Art of Dynastic China, London
1981, P. 43).
s3 In addition to the you in the Asian Art Museum in San Francisco, these include a jue from Panlongcheng (illustrated in Bagley,
Shang Ritual Bronzes, fig. 22; a jue from Yangzhuang and Dengfeng (illustrated in Editorial Group of Henan chutu Shang Zhou
qingtongqi, Henan chutu Shang Zhou qingtongqi, Beijing I98I, figs. 60, 83, 84.
54 See below, p. 45.
55ss Wu Hong, "Yi zu zaoqi de;" Jessica Rawson, Ancient China: Art and Archaeology, London I980, pp. 39-40, 78-79; Louisa G.
Fitzgerald Huber, "The Tradition of Chinese Neolithic Pottery,"Bulletin of the Museum of Far Eastern Antiquities
53 (1981), note 18, p. 125.
56 E.g. An Zhimin, "Guanyu Liangzhu wenhua de ruogan wenti," Kaogu I988.3, p. 244; Li Xueqin,"Lun xinchu Dawenkou
wenhua taoqi fuhao," Wenwu I987.12, p. 79-
57 Bagley, "P'an-lung-ch'eng: A Shang City in Hupei," Artibus Asiae 39.3-4 (I977), P. 200. Criticized by Robert Thorp, "Early
Shang Civilization. New Data from Ritual Vessels," HJAS 45. I (I985), P. 39. The jade baton was published in Kaogu 1976.4, p.
262.

58 Most important recent finds of Liangzhu artifacts with mask-like designs were reported in: Nanjing Museum, "1982 nian Jiangsu
Changzhou Wujin Sidun yizhi de fajue," Kaogu 1984.2, pp. o109-I29, figs. o10- 11, pl. 2:4; a series of articles in Wenwu 1984.2,
pp. I1-22; Institute of Archaeology, Zhejiang Province," Shanghai Qiangpu Fuquanshan Liangzhu wenhua mudi, Wenwu
I986.o10, pp. I-25, figs. g19, 22, cpl. 2; I, pl. 3:5. See too Wang Zunghuo, "Liangzhu wenhua 'yuliangzang' shilue," Wenwu
I984.2, pp. 23-35 and newly reported finds from the Yaoshan and Fanshan sites. See Institute of Archaeology, Zhejiang
province, "Jiangsu Yihang Fanshan Liangzhu mudi fajue jianbao," Wenwu I988.1, pp. I-31; idem, "Yuhang Yaoshan Liangzhu
wenhua jitan yizhi fajue jianbao," pp. 32-5I1, cpl. 1-2, pls. 1-6.

36

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 Fig. I. You vessel with man-beast motif. Sumitomo collection.
After Sen-oku Hakko Kan, fig. 3.

Fig. 3. Li-he with a face-mask-like idiom in thread relief line decoration.

Asian Art Museum, San Francisco.

Fig. 2. Clay figures from Xiaotun. Collection of Academia Sinica.

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 Fig. 4. Nephrite cong tube with face mask design from M4 burial at Wujin, Sidun. After Kaogu 1984.2, pl. 2:I.

Fig. .5- Jade D-shaped plaque with theriomorphic face design. Freer Gallery of Art, Washington. After Orientations
November 1983, fig. 10o, p. 18.

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 igy:

---..;r?~Y~ ??,? . I^?ixMZ-,,

:a',g

::\A-:

lwi:.

11:

r::

MfOiiii:ti
KIM;
'R131::- ;

R",
gN:::;;?l,:::':~.:L:(_: :I;::: ;:li :
_::::i: : I:-~l:ii

.i-:iiii--::;..i l-_illii_, _ .i?4 8 "~; ;Ii~i-_ri:~,iia-.: .;::*.-;;::tj; ;::1: i~l._:i: ll:::;:::~. :?-::::?;::; j~i'j ";. ?:; ;1"4, W,-`-:

Fig. 6. Artifacts with theriomorphic face designs recently unearthed at Yuhang,

Fanshan site. After Wenwu 1988. 1, pl. 2.

Fig. 7. Face-mask design on a gui tablet from Rizhao,

Liangchengzhen. After Kaogu 1972.4, P. 57-

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 01?

.e --M, RMS

Ax
V

A .xl-
t.'?
lx

x'P'.

M E. *N.1K_1 zNk C' 5

V11 ?2:

"IM E.-NA
m%
01:11? ?N

N"O
PRIN.- 01
; r?F al
-gg
E
.. .. ... .. .

. ......... .

IN Mb&
J-p
a

4 dv

ON " MR,

Z AR,

fill
5" g!?'

X. game ...........

nywWWWW',

p,
'A "AMIN
'A
MR-'R: wo

_Aw
0

, A1.0

w ""WAN
Mum
'ON
MON ENS
"In

Asi WF. _g' ERE

fx

mom Winv
X2

Ks gem
ou
O':E"E
"MIN

R AWT
gsw

.1.0E

......................
I WE

Fig. 8. Mollusk shell dragon images from Xishuipo burial.

After China Pictorial 1988.3-

1M,
? wm?

- NO low.
;jA

NO
wm

T?W' 7'
WO
IV, WON""
?;- ?-OW,6
list? 'e? F " 4;14Er
V.- mv -
IG

"o ',4-.

? 46"PA?7?41-

&7'T-4 4e"'
'V I aN WAS-
W
own tog
4F
Ic 14

A!". XQ- 4-5-5? KAIN F, loilo'

P Im'.

IT-

t-Z

NVAIV:
XA

Fig. 9. Impression on clay of wooden sculpture from royal tomb at Xibeigang. After Chang, Shang Civ

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 Fig. Io. Human/animal mask image on cong tube (M 12:98) from Yuhang, Fanshan. After Wenwu 198 8.1, I :I.

Fig. 12. Jue vessel from Yangzhuang with an early version of face-mask
idiom. After Henan chutu Shang Zhou qingtongqi, fig. 60.

Fig. i i. Bronze inlaid plaque with animal mask from Erlitou.

After Kaogu i986.4, pl. 7:I.

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 Fig. 13. Ding vessel, formerly in Kawai collection. Fig. 14. Fanghu vessel, formerly in WernerJennings Collection.

Fig. 15. Hu from the collection of The British Museum, London.

Fig. 16. Fangyi from the col. of Museum fiir Ostasiatische Kunst, Reproduction courtesy of The British Museum.
K61n. Reproduction courtesy Museum fiir Ostasiatische Kunst.
This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 ,AM

ell

Coop

fl

. . . . . s.

Fig. 17. Si Tu Mu Fanghu (M5:807). After Yinxu

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 Fig. 18. Ding from the collection of the National Gallery, Prague.

Fig. I9. Yue axe from the collection of Museum fiir Ostasiatische Kunst, Berlin. Reproduction courtesy Museum fiir
Ostasiatische Kunst.

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
mask motifs were laid out: the conception of a frontal symmetrical face with a pair of eyes as a focus above nose
and mouth, and occasional scroll-like patterns around edges (figs. 4-6). As new discoveries continue to be
made, discussions of the meaning of these designs and the artifacts which carry them go on, with the majority of
authors placing them in a shamanistic context.59 Most probably, the Liangzhu designs represent and symbolize
supernatural forces which existed in the world view of the Lianzhu elites, nature spirits and deities with
protective and apotropaic functions.
The Liangzhu phenomenon represents, however, only one aspect-albeit the most important one-of a
widespread use of mask-face designs in Neolithic cultures, preceding the emergence of Shang bronze casting.
Theriomorphic imagery is also documented from late Neolithic Longshan cultures on the Shandong peninsula
(fig. 7).60 As compared with these eastern cultures, it was far less developed in the Central Plain area, even
though recent finds of a mollusk shell mosaic "dragon" at a IV millenium B. C.E. Yangshao burial at Xishuipo,
Henan (fig. 8) and in ceramic ware found in the late Neolithic Taosi site seem to attest to separate traditions.61
My inescapably rapid summary has served to indicate that beginning in the V-IV and through the III and the
first half of the second millenia B.C.E., there existed a widespread, regionally differentiated tradition,
predominantly in eastern cultures, of creating unreal face-mask and "dragon" imagery. In their respective
contexts, such images were used primarily to decorate ritual paraphernalia and they may be considered as
representing and/or symbolizing supernatural beings, mostly nature deities and spirits, perhaps even shamanis-
tic mediators.
The problems of the relationships between these Neolithic cultures and their contributions to the formation
of Shang civilization are complex. It is generally accepted, however, that the Shang were in important respects
culturally linked to eastern Neolithic cultures.62 There is a strong case for looking for the origin of the Shang in
the area of eastern Henan and Shandong.63 Most would agree that the Shang civilization arose as a result of an
eastern impact on the Zhongyuan tradition, as a fusion of Eastern and Central Plain elements.64
Archaeological discoveries over the past two decades have encouraged reconsideration of the development
of Shang bronze decorative styles from the Erligang period into the Anyang phase, which is of importance for
the present discussion. It is now becoming obvious that Max Loehr's original stylistic scheme should not be
viewed merely as a linear sequence, but also as a spectrum. As Robert Thorp observed:

"...rather than contrasting Styles I and II because of their mirror-image relationship, one could group
together all renderings of this repertoire of well-defined motifs...
The principal development of the period was not a technical change in relief decor but rather the full adoption
of clearly formulated motifs for decorating bronze vessels."65

59 E.g. Wang Wei, "Liangzhu wenhua yucong chuyi," Kaogu 1986.II, p. IoI 5; K.C.Chang, "Tan 'cong' ji qi zai gudai Zhongguo
shi shang de yiyi," Wenwu yu kaogu lunji 1986, pp. 2 52-60.
60 More frequent among images on jades from this area are those with anthropomorphic features (e. g. in Wu Hong, "Yi zu zaoqi
de"). See too catalogue of E. Childs-Johnson's exhibition, Ritual and Power:Jade ofAncient China, at China House, New York,
spring of 1988.
61 Puyang Archaeological Team et al., "Henan Puyang Xishuipo yizhi fajue jianbao," Wenwu I988.3, pp. I-6; "I988 nian Henan
Puyang Xishuipo yizhi fajue jianbao," Kaogu 1989.I2, pp. 10o57-66. For a "dragon" inside a bowl from Taosi, Kaogu 1983.6, pl.
4:1.
62 Chang, "Sandai Archaeology and the Formation of States in Ancient China: Processual Aspects of the Origins of Chinese
Civilization," in: Keightley, ed.,Origins of Chinese Civilization. Berkeley 1983, pp. 5io-Ii; idem, Shang Civilization, pp.
345-46.
63 Chang, Shang Civilization, pp. 345-46; Wang Yuzhe, "Shangzu de laiyuan diwang shitan," Lishiyanjiu, Beijing I984.I, pp.
61-77. For recent alternative opinions concerning the origins of Shang see Zou Heng, Xia, Shang, Zhou kaoguxue lunwenji.
Beijing 1980, pp. 253-94 ( southern Hebei); Chang Changyuan, "Shangzu qiyuan diwang fawei," Lishi yanjiu 1987.1, pp.
I36-44 (Shaanxi).
64 For an early formulation Huber, "Traditions," p. 124-2 5; also Keightley, "Archaeology and Mentality: The Making of China,"
Representations 18 (1987), pp. 116-I7.
65 Thorp, "Early Shang Civilization," pp. 41-42. Thorp has recently proposed an alternative to Loehr's linear scheme, whereby
Styles I-V are viewed as a three stage evolution (I-II, III, IV-V), with the first and second as alternative techniques ("Archaeology
of Style at Anyang: Tomb 5 in Context," Archives 41 (1988), pp. 47-69.
Further refinement of the sequence of Shang bronze decorative styles may be expected, following new finds and progress in
dating.

45

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 Further, it has become clear that the figure-and-ground- distinguishing styles (high relief or undulating
variants of Style III, Styles IV and V) were introduced earlier than previously thought. Fully-fledged monster
masks came into existence in the period just prior to the dynastic occupation at Anyang; and no later than
during the first century of rule there the use of high relief decoration became common.66 Moreover, it is
noteworthy that the evolution of theriomorphic motifs, just prior to the Anyang phase, i. e. the gradual growth
toward clearly formulated images, was coincidental with both the territorial expansion of Shang power and the
steady growth of Shang civilization toward the new type of theocratic state achieved at Anyang.67 The
importance of the adoption of the figure-ground ambiguity-eliminating modes cannot be overestimated; and
this stylistic evolution should be related to the wider socio-political changes which took place at the beginning
of the late Shang period. I assume that this major stylistic advance may be, at least partly, understood by
identifying theriomorhic imagery as a powerful symbolic device, carrying ideological and political functions.

The functions of the taotie

Every student of ancient China is familiar with the latish, apocryphal story of the nine tripods cast by
Emperor Yu which testifies to the role of bronze vessels as tokens of rule and political authority in the late
Bronze Age.68 Other bronze objects, notably bells and weapons, as well as jade and ivory regalia, were also
symbolically used by Bronze Age elites as status markers. Some symbolic and political dimensions of these art
forms have been previously remarked upon by scholars.69 Further, more in-depth discussion in this direction
seems to offer other promising means of understanding the significance of early Chinese art.
It has been repeatedly observed that symbols play a key role in the processes of establishment, maintenance,
and exercise of power in various types of societies.70 Art objects constitute a very significant part among the
symbolic forms and actions involved in power relationships. The important concept here is that of ideology
because the exercise of power is negotiated through ideologies and there is a close linkage of symbolic and
ideological forms.71
Generally, ideologies can operate in given societies in the following ways: by denying or obscuring conflicts
and contradictions; by representing the fractional interests of hegemonic groups as universal; by reifying and
objectifying the present through naturalization of existing systems of relations which appear as immutable and
part of the natural order.72 A lot of attention has been recently paid to the role played by material culture in

SThis depends on the date of Fu Hao (M 5) burial at Anyang, which is now overwhelmingly attributed to a consort of King Wu
Ding. See Chang, ed., Studies of Shang Archaeology. New Haven 1986, passim. See too review by Edward L. Shaughnessy in
Journal of the American Oriental Society 10o7.-3 (1987), pp. 500-03-.
Implications of this burial for study of bronze styles were first noticed by Nancy T. Price, "Note: Once Again on Loehr's
Bronze Styles," Early China, Berkeley, 3 (1977), pp. 96-97; Huber, "Some Anyang Royal Bronzes."
6 For a framework for such a development see Thorp's recent "network model", that seems to provide a more precise conceptual
basis for study of bronce styles than earlier models of provincial vs. metropolitan centers. Thorp observes "...The bronze
production took place at many nodes in that network but always within a tradition that shared more than it differed." ("Early
Shang Civilization," p. 55). Even this view requires one point of initial impulse to accelerate and control this tradition; in all
probability being connected with the Shang lineage and its rituals.
68 The locus classicus for the nine tripods story is the Zuo Zhuan; see The Ch'un Ts'ew with Tso Chuan, vol. V of The Chinese
Classics, trans. by James Legge. Oxford 1872 (rprt. 1931), p. 293.
69 For symbolic dimensions of the jiu ding story see Chang, Art, Myth, and Ritual, esp. pp. 96- ioo. Political functions of early
Chinese art have been envisaged also in writings of Chinese authors; treated through a Marxist paradigm (cf Ma Chengyuan,
Ancient Chinese Bronzes; Li Zehou, Meide Licheng. Beijing 1981).
70 Murray Edelman, The Symbolic Uses of Politics. Urbana I964; Abner Cohen, Two-Dimensional Man: An Essay on the
Anthropology of Power and Symbolism in Complex Society. Berkeley 1976; idem, "Political Symbolism," Annual Review of
Anthropology 8 (1979), pp. 87- 113; Jonathan Haas, The Evolution of the Prehistoric State. New York I982, p. I60.
71 Ideology may be here defined as a system of sacred or secular beliefs that serves to rationalize an economic and political order (cf
Mark Leone, "Some Opinions About Recovering Mind," American Antiquity 47.4 (1982), pp. 742-60.
For the concept of ideology and its use in current prehistoric studies see Michael Shanks and Christopher Tilley, "Ideology,
Symbolic Power and Ritual Communication: a Reinterpretation of Neolithic Mortuary Practices," in: Ian Hodder, ed.,
Symbolic and Structural Archaeology, Cambridge 1982, pp. I29-54.
72 Ibid., pp. I31-32; Hodder, Symbols in Action: Ethnoarchaeological Studies of Material Culture, Cambridge 1982, p. 209.

46

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
these representations, and to exploration of ideological practices involving its manipulation to legitimize social
dominance."3 This role has been formulated for early Bronze Age Europe, thus:

"...The contemporaneous appearance of rich individual burials in the archaeological record of many parts of
Europe in the Early Bronze Age was the result of the rise of an ideology which sought to legitimize social
differentiation, not by hiding it, but by representing it as natural and immutable through the use of material
culture in the form of prestige items and ritual symbols which constantly reiterated the message. In such
circumstances material culture as ideology could be said to be a very direct transformation of social
organization" (my italics).4

Early Bronze Age China seems both to support and significantly expand this paradigm. When used and
displayed in their social contexts (temples, tombs, shrines, and secular urban settings, military campaigns),
ceremonial artifacts, bronze vessels foremost among them, acted as a chief means of ideological representation
and symbolic coercion. What was unique in Bronze Age China was the fact that the Shang elites were able to
further articulate and emphasize these means through visual language in surface decoration of ritual parapher-
nalia. Adaptable to various media, to bronzes as well as to jade, stone, bone, and ceramic artifacts and probably
also architectural structures, and as their visual foci, powerful theriomorphic images intensified the message of
ritual artifacts, structures, and actions-statements of social hierarchy and political dominance.
This symbolical-ideological function cannot be understood apart from the dominant religious orientation of
Shang ancestor worship. Asymmetrical power relations, social hierarchy, and political dominance are our
modern analytical categories and as such they could not have been of concern to the Shang ruling class. But they
describe relations which were present in the Shang religious and sociopolitical order dominated by ancestor
worship; and Shang elites, impelled to maintain that order, consequently maintained and reproduced those
relations.

Any investigation of the functions of works of art and symbols in a given social context requires some
understanding of the framework within which they act; ideally the analysis of the viewing context precedes
such considerations. In the case of Shang ritual art, however, we are obviously hampered by scanty archaeolog-
ical evidence, for very little of original architectural and urban settings has been preserved, to say nothing of the
no less painfully felt absence of eye-witness accounts.75
It is sufficiently established that ritual vessels were primarily made for and shown in ancestral temples to
which probably only members of the nobility or gently born had access. Theriomorphic imagery, however,
appears on other bronze objects, as well as on stone, bone, and ceramic portable artifacts, the use of which was
not restricted to temples and their presumably limited audiences. Moreover all types of artifacts were routinely
put into the tombs (some being specifically designed for use in mortuary contexts) where most of them were
visible on grave ledges prior to closing. It may be assumed that the taotie mask and other images were made and
used also in monumental dimensions, and displayed in public areas as parts of urban and architectural settings.
Some archaeological evidence and parallels from other ancient cultures are indeed available (fig. 9).76

73 See contributions in Hodder, Symbolic and Structural Archaeology; also Daniel Miller and Christopher Tilley, eds., Ideology,
Power and Prehistory, Cambridge 1984.
74 Stephen Shennan, "Ideology, Change and the European Early Bronze Age," in: Hodder, ed., Symbolic and Structural
Archaeology, p. 15 6.
75 A useful overview of archaeological settings is provided in Thorp, "Origins of Chinese Architectural Style: The Earliest Plans
and Building Types," Archives 36 (1983), pp. 22-39. Also Wu Hung, "From Temple to Tomb: Ancient Chinese Art and
Religion in Transition," Early China 13 (1988), pp. 78-86.
76 For large impressions on earth of wood carvings preserving theriomorphic wood carvings in royal Anyang tombs see Liang
Siyong and Gao Quxun, Houjiazhuang: Henan Anyang Yindai mudi; diyiben, iooi hao damu. Taipei 1962, pp. 26, 56- 59. See
too Wenwu 1976.2, pl.2:2 for an animal mask carved on planks inside M2 tomb at Lijiazui, Panlongcheng.
There is ample archaeological evidence for use of imagery associated with power in theocratic societies. For Mesoamerican
analogy see Marshal Durbin, "Symbolism in Classic Maya Stelae Texts," in: Mary Foster and Stanley Brandes, eds., Symbols as
Sense. New York I980, pp. 117-20o.

47

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 All this would allow for the conclusion that a theriomorphic imagery significantly shaped the visual
environment encountered by all strata of the Shang population. It may be reasonably deduced that this
environment was rather meager in the totality of its visual stimuli; that each individual's mental model of reality
would have to be constructed from a limited store of items. In such a setting, theriomorphic images, singularly
impressive in their appearance and pervasive in their distribution, must have exercised very strong visual
impact, imposing an "inescapable tyranny of symbol.""
It is rewarding to compare this situation with other ancient societies where art played a prominent role in
ideological representations and power relations. In the Mayan theocratic state, for instance, as well as in some
Near Eastern empires and Bronze Age Greece, an elaborate iconography was employed which communicated
asymmetrical power relationships by visual puns and narrative, e. g. juxtaposing rulers and subjects, distingu-
ishing them by attributes or sizes etc." The functions were thus carried out mostly by means of representation,
even direct illustration. In Shang art, in accordance with the whole impersonal style of the culture, the
potentials of visual language were used differently and, one might argue, more efficiently. No direct informa-
tion through visual representation was provided; the social hierarchy and structures of power relationships
were not in any direct way reflected in subject matter and iconography. The taotie lent itself to carrying a
multiplicity of meanings (as will be further argued), yet at the same time it achieved a psychological power that
could hardly be matched by narrative pictorial art.

Meanings of the taotie

The old question of the meaning of the taotie has been over past decades compromised by many theories to
such a degree that many scholars today prefer to avoid addressing it at all. Far from trying to put forward
another "definitive" solution, I shall try to provide here some guidelines for reassessing the problem. First of
all, the specific character of the material under consideration should be fully acknowledged. Accordingly, our
expectations of what kind of answers we can obtain and, above all, our modes of addressing the question should
be adjusted. Interpretive method, after all, should always be born in the interaction with concrete art objects. In
the case of Shang imagery, the uncritical application of standard iconographic paradigms cannot be expected to
produce much results.'" Therefore, rather than asking "what is the meaning of the taotie?" which in all previous
essays has been tantamount to asking "what being does the taotie represent?," the central question should be
put in a different way: "How could these images have been meaningful to the Shang people?"
Let us start with the suggestion that for these motifs to be a fully meaningful part of the Shang people's
reality, did not by any means require that they be conceived as representations of supernatural beings or that
they were imbued with an univocal meaning shared by the Shang population. In the case of the Shang state,
especially in its late phase, represented by the Anyang cult center, we deal with a highly stratified society.8? As
recent research has persuasively demonstrated, perception and conceptualization of visual imagery are interest-
driven and competence-regulated mental activities.81 Accordingly, it should be recognized that there could

77 Howard Wechsler's term, Offerings of Jade and Silk. Ritual and Symbol in the Legitimation of the T'ang Dynasty. New Haven
and London 1985, p. 32.
78 Joyce Marcus, "The Iconography of Power among the Classic Maya," World Archaeology 6 (1974), PP. 83-92; Edith Porada,
"The Uses of Art to Convey Political Meanings in the Ancient Near East," in: David Castriota, ed., Artistic Strategy and the
Rhetoric of Power. Political Uses ofArtfrom Antiquity to the Present, Carbondale 1986, pp. 15 -26. For Minoan and Theran art,
Nanno Marinatos, Art and Religion in Thera: Reconstructing A Bronze Age Society, Athens 1984, esp. pp. 119-20.
79To my knowledge, in no previous iconographical theories of Shang designs has the issue of applicability of art-historical
interpretative strategies ever been mentioned.
8 I subscribe here to the division of Shang population into ruling class, i.e. members of royal and aristocratic lineages, officials,
ritual specialists; and ruled strata. For Shang social stratification see Chang, Shang Civilization, pp. 158-94; Chao Lin, The
Socio-Political System of the Shang Dynasty, Taipei 1982, esp. pp. 115-25.
81 Staale Sinding-Larsen, Iconography and Ritual. A Study of Analytical Perspective. Oslo I984, pp. 156-58. It has long been
recognized that responses to visual imagery vary according to the social position of onlookers. Cf Edwin Bewan, Holy Images,
an Inquiry into Idolatry and Image-Worship in Ancient Paganism and Christianity. London 1940, p. 29.

48

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
hardly have been at that time any universal meaning of the taotie or any other image-symbol. They must have
been perceived differently by different strata and interest groups within Shang society.
The significance of these images for members of the lowest stratum, prisoners/slaves and the non-Shang
population, may have substantially differed from that of Shang commoners, which in turn undoubtedly
differed from that of Shang elites. Moreover, the symbolic/ideological effects of this imagery would not have
depended on any precise conceptual understanding on the part of the lower strata of population. These effects
could be well achieved through intuitive understanding, based primarily on psychological responses evoked by
such images.82 Several Chinese authors have proposed that the monster mask and other therio- and zoomorphic
forms were used by the ruling class to frighten and manipulate the lower strata (referred to as "slaves" in
Chinese authors' Marxist terminology).83 Though this explanation, as rightly observed by K.C. Chang,84
would in itself be inadequate to explain the significance of the taotie, the question of the psychological effects of
the images vis-a-vis the lower strata of the Shang population is highly pertinent. In fact, here we can recognize
one of the mechanisms by which their symbolic-ideological effects were achieved. Theriomorphic images may
indeed be presumed to generate feelings of mystery and fear, thus exercising coercion on the subordinate strata.
Such effects, however, would be achieved primarily through the visual properties of these images, indepen-
dently from iconographic meanings which might have been associated with them. Whether or not they were at
the same time conceived as representations of deities or supernatural forces, the message of these images and the
objects they decorated would not have been lost on such people. Monster masks and other theriomorphic
designs would be primarily meaningful to them as visual embodiments of the Shang order.
More important, however, was the perspective of the ruling elites, the royal and noble and gently born clans
which ordered the making of bronze vessels and other artifacts. How could those ubiquitous images have been
meaningful to people at the political pinnacle of the Shang state? Once again, the specific nature of this material
must be mentioned, namely the fact that in the case of the Shang theriomorphs, it is the evolution of forms itself
which seems to provide the most telling clue to the enigma of their meaning. Only by reviewing the changing
appearances of mask- and face-like designs over the long run, from the designs on jade artifacts of east coast
Neolithic cultures through conventionalized, stabilized, "classical" taotie on late Yinxu II vessels,"85 can we
arrive at some grasp of their significance. Such review strongly defies the notion of the taotie as an icon, devised
to match some mythological or totemistic animal, unambiguously standing for some definite, preordained
concept of Shang religion. What this review reveals is the persistence of a stylistic idiom (or model), essentially
the frontal face suggested by a pair of eyes, which was continually developed over the Late Neolithic and Early
Bronze age phases of Chinese culture, and in the process was continually invested with (and then largely
denuded of) meanings.
The origins of this stylistic idiom may be traced to the above-mentioned designs on east coast Neolithic jade
artifacts. On some of them they were elaborated to a considerable degree, resulting in images with pronounced,
recognizable anatomical details. Such images undoubtedly had some conceptual significance in the world-view
of the Liangzhu people (figs. 4-6, o10).86 However, on the threshold of the Bronze Age, we find this family of

82 For an excellent analysis of "intuitive understanding," based on study of Christian imagery see Sinding-Larsen, op. cit. pp.

i69-70; David Mannings, "Panofsky and the Interpretation of Pictures," BritishJournal of Aesthetic 13 (1973), p. I5O explains
why "to react is not necessarily to understand." Cf also important remarks from the perspective of meaning analysis in
archaeology in Ian Hodder, "Theoretical Archaeology: a Reactionary View," in: Hodder, ed., Symbolic and Structural
Archaeology, p. Io.
83 Li Xueqin, The Wonder of Chinese Bronzes. Beijing I980, p. 56; Li Zehou, Meide Licheng, pp. 36- 37; Ma Chengyuan, Ancient
Chinese Bronzes.
84 Chang, "Ancient Chinese Civilization," pp. 36-37-
85 The periodization of Anyang finds into I-IV phases, based on typological sequence of Anyang ceramics is agreed upon by both
Chinese and Western scholars. Cf Yinxu qingtongqi, esp. pp. 3F-38. Thorp, "Archaeology of Style at Anyang," provides
guidelines for integrating Chinese with Western art-historical sequences.
86 The similarity between Neolithic and Bronze Age renderings of visual schemes tempts us to explain them as having common
meanings; see Chang's elaboration on his alter-ego theory ("Puyang sanqiao yu Zhongguo gudai meishu shangde ren shou
muti," Wenwu 1988. 11, pp. 36-39). However, it is well-known that forms persist while meaning accrues to them; continuity of
stylistic models or motifs does not require any continuity of meaning.

49

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
face motifs reduced again to minimalistic versions, consisting mostly of just a pair of eyes and a bar nose or
mouth, accompanied occasionally by scroll-like designs, indeterminate enough so as to discourage their
association with any definite anatomical parts, yet precise enough as a design system to allow the establishment
of stylistic continuity.8" This stage, exemplified by a few artifacts from late Neolithic and Erlitou sites (figs. 7,
I1)88, is particularly instructive: it suggests that by the time the first bronze vessels carrying this simple
decoration emerged, the meanings of the stylistic idiom had been lost, or at best only dimly perceived by the
elites and ritualists of the emerging Shang state; and in any case were not of great relevance for them.89" Had it
not been so, we would have to expect to encounter more pronounced, better defined images at least in jade
carving, lacquer, and pottery, (allowing for the fact that bronze artisans were at this stage still in the process of
learning the possibilities of the medium and thus unable to produce better articulated images).
The earliest bronze artists thus worked with hardly any well-established, clearly-defined pictorial tradition,
underlined by distinctive concepts, which would be shared by the upper echelons of society. However, once
the face-mask-like model was transformed on the walls of bronze vessels, at first only as thread lines suggestive
of eyes (figs. 3, 12), it evidently gained a new, strong impetus for development. The design evolution through
Style I toward mature Style III, i.e. the emergence of corporeal images, when the bare model was gradually
getting zoomorphic content, seems to be the result of a unique creative cooperation between bronze artisans
and their patrons." To the former, it provided a means to decorate the newly elaborated forms of bronze
vessels; to the latter it became a means of reference to ceremonial actions accompanying sacrificial rituals. I view
it as a two-way process: on the one hand, the still lingering memories of nature deities and spirits which had
inhabited the world-view of distant Shang ancestors (and perhaps also lingering memories of earlier imagery),
might have contributed to the steady rise and naturalization of both taotie and "dragon". Equally important, as
the evidence of the images suggests, was the reverse course, i. e. as images coming into existence on the walls of
vessels encouraged the religious imagination, the new forms suggesting new meanings.91 It indeed seems
probable that only as theriomorphs were acquiring more solid, determined appearances, did they assume real
significance within the Shang culture. Though zoomorphic monsters, and nature spirits do not seem to have
occupied any central place in a Shang world dominated by ancestor worship92, they may have gradually
acquired importance as a means of visual reference to the supernatural, of approximating otherwise unrepre-
sentable elements of Shang religion. In Professor Wu Hong's apt formulation: "These varying images seem to
testify to a painstaking effort to create metaphors for an intermediate state between supernatural and reality -
something that one could depict but could not portray."93

87 The reading of these designs as "a pair of eyes", is an act of interpretation beyond a purely descriptive level.
88 At least two Erlitou artifacts feature more explicit renderings of face: animal-like on a bronze plaque and a face on a jade baton
(Kaogu I976. 4, pl. 10:4). The latter piece has been widely discussed, see Bagley, Shang Ritual Bronzes, note 40, p. 48.
89 A similar conclusion has been reached by Keightley, "In Clear and In Code," pp. 30-31. Tsung-tung Chang had proposed a
view that Shang bronze imagery represented a supernatural world of zoomorphic monsters which the Shang no longer believed
in (Der Kult der Shang-Dynastie im Spiegel der Orakelinschriften. Wiesbaden 1970, p. 261.) This, however, does not
automatically imply that "In diesem Sinne ware die Kunst der Sakralbronzen bereits zur Shang-Zeit ein 'Survival' geworden, ein
unorganisches und unverstandenes Uberleben von Vorstellungen alterer Entwicklungszustande." I assume that while the Shang
neither fully understood nor believed in Neolithic zoomorphic imagery, they developed a new, organic one on its basis.
90 The stylistic development prior to the beginning of the Anyang period is well documented in Bagley, Shang Ritual Bronzes, pp.
22-23; Bagley, however, asserts that artists merely followed the course of gradual enrichment of the decorative field, enhancing
its zoomorphic content.
91 It has been observed that visual imagery may exercise a retroactive influence on religious concepts. See Robert H. Lowie,
Primitive Religion. New York 1924; also Sinding-Larsen, Iconography and Ritual, p. 113.
Essentially the same mechanism may be discerned behind the late Zhou "iconological explanations" of the taotie and other
designs, preserved in Lu Shi Chunqiu et al.
92 Reasoning here cannot avoid a circularity: our understanding of Neolithic and Bronze Age images depends on knowledge of
contemporaneous religious orientation constructed from the evidence of visual imagery. The dominance of ancestor worship in
Shang world view is well documented and commonly agreed upon.
93 Wu Hung, "From Temple to Tomb," note 18, p. io6.

50

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 Particularly instructive is the development through the so-called Transitional Period, the first century or so

of Anyang occupation, as exemplified notably by a group of vessels with early Style IV designs (figs. 13- I 5).94
Here the separation of the image from the background was achieved for the first time on bronze vessels and the
taotie came out as a powerful entity. These intensely expressive, exciting, yet still diversified images indicate a
deliberate effort to enhance the psychological power of late Style III theriomorphs, an ongoing struggle to
arrive at some visual definition of the taotie. This was finally achieved in the Fu Hao/Wu Ding period. As
exemplified by vessels datable to that time, there is still a lot of variation among individual taotie designs
(somewhat less among "dragon" motifs), but the image is stabilized and, within its unreality, more or less
defined; the diversity of early Style IV images has given way to a certain conformity (fig. 16). While some of the
monster masks of this period may be considered the most powerful and expressive ever achieved (fig. 17), they
also exhibit an increasing degree of conventionalization which will ultimately mute the expressive force of
earlier theriomorphs (fig. 18).95
Only at that moment, when the frontal face was finally transformed into a fully-developed monster mask (as
exemplified by figs. 13-I8), does the taotie seem to acquire its true significance in Shang culture as a visual
metaphor standing for basic tenets of Shang culture. Its meaning seems to be embodied and expressed first and
above all in its visual qualities. In conjunction with the forms of the artifacts it adorned, the monster mask
communicated the impersonal, ordered Shang religious experience. It was thus instrumental in defining the
Shang cultural environment, establishing a meaningful framework of existence, a sense of participation in a
common tradition for those who conceived of themselves as Shang. As such the taotie probably also had
further, more specific and more "intimate", meanings for members of the elite. Most likely it also assumed
apotropaic and auspicious connotations within the context of temple and tomb.9' But again, meaning was
communicated directly through the forms, rather than by iconography. Despite the existence of certain
typified, "model" renderings of both taotie and dragon, the flexibility of the stylistic idiom was such that on
occasion new and sometimes highly idiosyncratic versions appeared. These may be attributed to the logic of
stylistic development and artists' innovative efforts. However, some of the more specific renderings of the
monster mask, i.e. those with distinctly anthropomorphic vein may have been invested with more specific
iconographic meaning (fig. 19).97

Concluding remarks

This view of the significance of the taotie sheds additional light on the mechanism by which their ideological
effects in Shang society were achieved. It is obvious that these effects would not have been planned by elites; the

94 See also Hayashi, Conspectus, vol. 1:2, p. 217, zun no. 27 from the Nanjing Museum collection. Other related pieces are
discussed also by Huber, who puts them together mainly on the basis of similarities in leiwen. ("Some Anyang Royal Bronzes,"
pp. 19-24). Actual provenance of these is uncertain. Though probably not cast at Anyang, their association with royal foundries
is to be expected.
95 Judgements on what constitute the most expressive or typical monster masks will, of course, differ. I would typify defined, yet
very expressive taotie with the renderings on the Fu Hao Si Tu mu fanghu (M5:794, 807, Yinxu yiqi, cpl. 44, 46, fig. 26). An
example of slightly later, fully conventionalized, "classical" taotie is the animal masks on a reproduced ding in National Gallery,
Prague. Cf with Thorp's "best statement of taotie motif" ("Archaeology of Style at Anyang"), p. 57, or Huber's "most
expresssive Anyang bronzes" ("Some Anyang Royal Bronzes"), pp. 32- 33. These differences nothwithstanding, Shang imagery
reached its culmination during the later Yinxu II phase, as exemplified by vessels from the Fu Hao tomb and those presumably
dating to Wu Ding and Zu Geng.
96 Keightley, "In Clear and In Code," pp. 8-9 speaks of the "all purpose importance" of the monster mask. For multiplicity and
multivocality of symbols see, e.g., Victor Turner, "Symbolic Studies," Annual Review ofAnthropology 4 (I975), P. 15 5; Abner
Cohen, Two Dimensional Man, p. 87. Nancy Price (personal communication) suggests that in the late Anyang phase, the vessels'
inscriptions increasingly assumed the role of iconic elements. Comprehensive correlation of decor with inscriptions and
archaeological contexts might bring important revelations.
97 Another would be the monster masks adorning the upper part of lids on the set of Nezu Museumfang hu, which hover between
anthropomorph- and theriomorphism, but are clearly related to the monster mask model. See Umehara Sueji, Kanan Anyo Iho,
Kyoto 1940, pls. 44-46; for detail of mask on "left" vessel see Mizuno Seiichi, In Shu Seidoki to gyoku. Tokyo I959, pl. 44.

51

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
taotie (or any other symbol/image) was not deliberately devised as a tool of ideological coercion by the Shang
ruling class. Nor should the situation be seen merely in terms of polarities: Shang elites as hypocritical
technocrats of power, manipulating visual resources without ascribing any meaning to them, and the ruled
strata as dupes, exploited by such symbolic forms. These images though not iconic representations were still
imbued with religious significance.
Important distinctions are apparent in this respect between the effects of imagery in the early Bronze Age
(Shang) and the late Bronze Age (Eastern Zhou) societies. In the latter, the function of bronze vessels and other
artifacts as status markers was consciously ascribed to them by elites. Moreover, these symbolic roles were
reflected upon in literary texts and thereby further reiterated.98 On the other hand, while Shang elites may have
consciously associated status with access to the proper use of certain types and numbers of artifacts, their main
thrust operated on an unconscious level."
Shang art was obviously not the creation of a classical literary society, such as Han China or Renaissance
Italy, for which iconographic method was originally devised, with a body of texts underlying visual images. In
Shang art, structures of signification were not based on any one-to-one discursive-to visual equivalence basis.100
We are accustomed to see and consider Shang bronzes as separate entities, as unique works of art, or at least as a
class of mutually similar objects. However, the theriomorphic images and artifacts which they adorned formed
just a part, albeit a visually dominating part, of ritual complexes and actions. Thus, while on the one hand they
enhanced and emphasized the messages of such complexes and actions, they themselves were being effective
and meaningful only through them. In fact, the meanings of these images would be perceived and interpreted
by the contemporaneous audience above all directly through these ritual actions, rather than being translated
into the discursive or textual mode. In sum, theriomorphic images do not seem to be imbued with iconographic
meaning, in the sense that they would stand for some concepts, clearly defined outside the realm of visual arts,
but rather seem to communicate meanings directly through their forms. Nonetheless, they were still iconic
forms, having referential meaning (reference being mostly to previous forms)l10; they also had iconological
meaning in their capacity to express and embody the Shang Zeitgeist.
This fact, however, does not imply that Shang art could be equated with so-called "primitive art."102 Such
adjectives as "primitive", "mystic", or "barbaric", so often used by both Chinese and Western writers to
characterize early Bronze Age Chinese art, seem to be entirely ill-fitted. Quite to the contrary, Shang people's
ability to elaborate upon older Neolithic visual tradition and incorporate it into holistic aesthetic-symbolic
expression cementing their socio-political order was a major accomplishment of a developing civilization.
Shang art forms represent a sophisticated means of communicating these people's world view, a unique way of
visualizing religious experience.
The taotie grew with Shang, so to speak. Its story may in a somewhat idealized and simplified, not altogether
incorrect way be taken to embody the story of Shang. Its predecessors, designs on Neolithic jades, mirror the
shadowy tribal ancestors of Shang; its evolution through Erligang and transitional periods parallels the steady
growth and expansion of Shang civilization; its fullest form mirrors the powerful theocratic society at Anyang.
Thus in a sense, one can still agree with the scholar who warned that without understanding the taotie we
cannot understand Shang.103

98 For the Zhou sumptuary system see Yu Weichao and Gao Ming, "Zhou dai yong ding zhidu yanjiu," Beijing Daxue Xuebao
(Zhexue shehui kexue ban) 1978.1, pp. 84-98; I978.2, pp. 84-97, I979.I, pp. 83-96.
99 Cf Michael Baxandall, The Limewood Sculptors of Renaissance Germany. New Haven i98i, p.5 : "The distinction between
manifest and latent functions [is] in effect often a distinction between conscious motives and actual consequences... It is difficult
to be sure that either an intention or an effect is unwitting, however undeclared or secondary..."
100 See too Sarah Allen, "Myth and Meaning," pp. 285-87.
101 See, too, Keightley, "In Clear and In Code," p. 18 for a strong assertion that in late monster masks the main referents seem to be
earlier renderings of the motif.
102 Some anthropologists draw a clear distinction between the art of literate societies and primitive (ethno) art, which is
characterized as simultaneously representing, defining, and manifesting its referents, without being defined by literal tradition;
see e.g. Charlotte Otten, Anthropology and Art: Readings in Cross-cultural Aesthetics. Garden City 1974, p.xiv.
103 Keightley, Sources of Shang History, p. 137.

52

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms
 Glossary of Less Familiar Chinese Names and Terms

bei (cup)
bin (ritual) g
cong (tube)

Dawenkou
Dengfens fe.3
gI
ding (vesselI type)
Erligang CL!IM
Erlitou 9-0,
Fanshan O Ld
Feng [Si!
Fu Hao OPT

Funan zun _p
guang (vessel type)
gui (tablet)
gui (demon)
guo (coffin chamber) P
GuoyLu o-4
Hemudu * ff4a'1
Hongshan 1 Li
jiu ding (nine tripods) ft~ih
jue (vessel type)
L i angzhu 1
i -he (vessel type)
Lijiazui - [in
1 iqi
Longshan (ceremonial
wE Lr arti
Panl ongcheng meal

Ri zhao Liangchengzhe
Sanx i ngta 1t a -- _{3 E
Shanhaijing
Si Mu Wu fang ding 0-- E 5i
taotie
Taosi :
Xiaotun dft
XishuipoR 7e<fl
Yaoshan "eJ LdI
Yangshao ffU]
Yangzhuang
Yi 9 B.:?
you (vessel type) I
Zhengzhou 1311
Zhongyuan Fpr
Zuozhuan 7EiUS6

53

This content downloaded from 104.207.138.102 on Sat, 24 Nov 2018 14:13:23 UTC
All use subject to https://about.jstor.org/terms