The religious art and culture of the Shang dynasty (1046BC), the second royal dynasty of China, comprised the imagery, objects, buildings and performances through which they addressed their ancestors, the nature powers and the high god Di. Almost all surviving Shang art was ritual in function, and much of it was made to be buried. It is known chiefly from the royal cemetery and settlement at Yinxu near Anyang, from the earlier centres at Zhengzhou, Yanshi and Xiaoshuangqiao, and from the oracle bone inscriptions, which supply the only contemporary written commentary on it. Because that commentary records divination rather than doctrine, and because the royal tombs were looted soon after the dynasty fell, the relationship between Shang art and the spirits named in the texts is disputed at almost every point.
The dominant motif in Shang ritual art is a frontal animal-like face, conventionally called the taotie after a term applied to it by Song dynasty antiquarians. Assembled from the horns, jaws and claws of real and imaginary creatures, it appears on ritual bronzes from the Early Shang onward, shares them with dragons, birds, cicadas and tigers, and has been read in turn as a shamanic helper, as an image of the king's transformation, as a projection of the celestial pole and as ornament without significance. No interpretation is unequivocally agreed on. Colour carried meaning of its own, recoverable in part from the vocabulary of the inscriptions and from the red pigment, cinnabar and lacquer applied to bones, coffins and stone.
Ritual was as much performed as visualised. Bells, chimestones and hide-covered drums accompanied sacrifice and dance, and the inscriptions record divinations over which instruments to use and on what day; cattle, millet ale and human victims were offered in settings ranging from open-air rammed-earth altars to the possibly cosmology-derived roofed halls and courtyards of the Yinxu precinct. Making the objects was a religious undertaking in its own right, consuming metal, fuel and skilled labour on a scale that shaped the economy around it. After the Zhou conquest the ancestral cult persisted, but the vessel repertory was remade and the mask was abandoned by the eighth centuryBC. Shang bronzes returned as objects of antiquarian study under the Song.
Ritual art Evidence Shang art was primarily ritual in function. The sinologist David Keightley argued that in a society which "even at the elite level, may largely have been illiterate", high art served as a kind of writinga medium conveying, through decorated bronzes, ornamented chariots and elaborate burials, the elite domination and authority that texts, characterised as illegible to most, could not yet convey. On this view such art functioned as the writing system later would, as an esoteric code granting privileged access to material and spiritual resources to those educated to read it.
Keightley distinguished two modes within Shang visual art. One is relatively naturalistic, seen in ornamental jades of turtles, geese, horses and kneeling figures from the Tomb of Fu Hao; these appear to have been ornaments of daily life rather than ritual insignia, and as such may not have carried any powerful symbolic or religious message, being "not involved in contacting the ancestors who had… become abstractions and were not conceived in naturalistic terms". The other, which he followed William Watson in calling hieratic, is abstract and schematic, associated with sacred things and highly stylised; on bronzes in particular "the orderly and symmetrical placement of the decorative elements was a paramount concern". He posited that these two modes were not opposed but formed a single tradition spanning a spectrum.
Shang artistic expressions bear no explicit relations to the pantheon described in oracle bone texts; rather, their art carried social and religious significance that varied by place and changed over time. Wu Hung notes that a century of work on some two hundred thousand oracle bone fragments has produced no information of the kind early scholars expected, such as proper names for the animal images, and that the bronze zoomorphs represent neither definite deities nor ancestors while still being more than merely decorative. Sarah Allan puts forward that the motifs cannot be related to the spirits named in oracle bone inscriptions. Keightley reached a comparable conclusion from the images themselves, observing that the referents of the animal masks in the natural world are often hard to discern and that Late Shang bronze decor seemed to be derived from earlier animal-mask designs.
Keightley preferred to speak of Shang "religiosity" rather than of a Shang religion. What survives is unrepresentative, and in a way that bears directly on what can be said about the art. Little was recovered from the royal tombs at Xibeigang, looted soon after the fall of Anyang, resulting in scholars still disagreeing over which tomb belongs to which king. The richest Shang burial yet excavated is not a royal one the tomb of Fu Hao, which escaped the robbers because her temple foundation sat over it, yielded some two thousand objects, including 468 bronzes and 755 jades. Kesner notes that the analysis of ritual art ideally begins from the setting in which it was seen, but that very little of the original architectural and urban context has been preserved, and that there are no known eye-witness accounts of Shang ritual.
A further absence is one of subject matter rather than of preservation. Lothar von Falkenhausen observes that, in contrast to every other early civilisation of the Old and New Worlds, the Shang and Zhou core area shows a complete absence, down to the beginning of the imperial period, of any depiction of important human beings, and that neither dynasty represented its deities in human form; he calls this an "apparent taboo". Where human faces and figures do appear in Shang art they carry the same hieratic character as the taotie, and are often absorbed into itas on the Heda fangding from Huangcai in Ningxiang, where human faces are built into the structure of the design, complete with small animal claws at the sides. He connects the prohibition to a wider feature of the tradition, the near-absence of figures shown in action, and notes that the neighbours of the Shang and Zhou were under no such constraint, as the anthropomorphic bronzes of Sanxingdui at Guanghan make plain.
Principles
Several religious principles have been identified as shaping Shang visual art: the dualistic outlook of Shang cosmology, the impersonalisation of the ancestral cult, and the concept of yi metamorphosis. Shang divination characteristically framed its charges as paired positive and negative propositions, cracked on the right and left of a single shell. Keightley took the fact that "much Shang art was also characterized by the symmetrical opposition of images" as further evidence of the value Shang elites placed on such binary oppositions.
The Shang ancestral cult had the character of an impersonalised one, in which the dead were stripped of individuality on becoming ancestors, and the art correspondingly avoids depicting them in direct human form. Keightley located the mechanism in the ritual calendar itself: each royal ancestor received cult on the day of the stem carried in his temple name, Da Jia on a jia day and Da Geng on a geng day, and it was this scheduling, he argued, that "turned the royal dead into ancestors". An art addressed to recipients so conceived had little use for portraiture. Keightley connected the same ancestral orientation to the tradition's "remarkable endurance and conservatism", reasoning that patrons who worshipped their ancestors were not disposed to alter the forms in which those ancestors were addressed, and to the value early Chinese artisans placed on abstraction and geometric ordering.
Elizabeth Childs-Johnson has argued that Shang imagery is dominated by the rendering of transformation, or yi ()in her account the power to metamorphose in order to communicate with and identify with the spirit world, and specifically the king's prerogative to bin (), "to identify with the spirit world". On this reading the principal vehicle of the idea is the semi-human animal mask, accompanied by subordinate images of metamorphosis such as the feng bird, the long dragon, the cicada, the snake and the owl. She further treats the graph yi itself as a logogram depicting a human wearing a mask, so that graph and image render the same conception.
In her fuller treatment Childs-Johnson places yi in a class of what she calls sacred, spirit-oriented verbs, alongside bin and zhu (), "to invoke spirits", whose subjects in the inscriptions are only ever the king, living or dead, and Di. The metamorphosis they name is a change of standing rather than of shape: to be empowered by a spirit through the wearing of a mask, and so to be identified with whatever the mask represents. She argues that bin, conventionally rendered "to entertain a guest", primarily means "to receive a spirit" or "to be received by one", and that because dead kings could bin more senior ancestors and Di could bin a past king, the living king who performed the rite was exercising a power otherwise proper to gods.
On her account two strands of belief run through Shang religion and its art: an older, shaman-like metamorphism which she takes to have preceded the ancestral cult, and the cult of the royal dead itself, the two fusing during the Late Shang. The mask guise, she argues, consistently belongs to a species of hunted wild animal, a claim she supports from the faunal material recovered at Shang sites.
The taotie motif
The most conspicuous motif of Shang ritual art is a frontal, animal-like face conventionally called the taotie, whose name is a later attribution and is disputed. The term first occurs in the Zuozhuan, where Tao Tie is one of four malevolent beings, and is glossed as a glutton; the Lüshi chunqiu supplies the first association of the name with a bronze motif, describing ding decorated with a taotie that "has a head but no body" and devours a man. The two were matched to the surviving motif by Song dynasty antiquarians, in particular in Lü Dalin's Kaogu tu. Wang Tao regards that association as misleading and has proposed that the field abandon taotie in favour of the neutral description "two-eyed motif". Elizabeth Childs-Johnson likewise rejects the term as anachronistic, on the ground that it is a label of Warring States and Han date applied to material at least a thousand years older, and uses "metamorphic power mask" instead.