the spirit kings in sixth century chinese buddhist sculpture

The Spirit Kings in Sixth Century Chinese Buddhist SculptureAuthor(s): Emmy C. BunkerSource: Archives of the Chinese Art Society of America, Vol. 18 (1964), pp. 26-37Published by: University of Hawai'i Press for the Asia SocietyStable URL: http://www.jstor.org/stable/20067068 .

Accessed: 14/06/2014 05:23

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

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

.

University of Hawai'i Press and Asia Society are collaborating with JSTOR to digitize, preserve and extendaccess to Archives of the Chinese Art Society of America.

http://www.jstor.org

This content downloaded from 62.122.79.78 on Sat, 14 Jun 2014 05:23:20 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=uhp

http://www.jstor.org/action/showPublisher?publisherCode=asiasoc

http://www.jstor.org/stable/20067068?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


The Spirit Kings in Sixth Century Chinese Buddhist Sculpture1

Emmy C. Bunker

Denver Art Museum, Denver, Colorado

A group of curious looking male deities, at the most ten, usually referred to as the

"Spirit Kings," appears frequently in Buddhist Sculpture of the sixth century A.D. in China, but, oddly enough, no adequate study has ever been devoted to them.

The Spirit Kings were first noticed by Chavannes in an article2on the Eastern Wei stele of 543 in the Isabella Stewart Gardner museum, where they adorn two sides and the back of the base. The deities are identified by names inscribed alongside them and their own spe cial attributes.

Fig. 1. Drawing of Lung-men, Pin-yang Cave (Cave III), Fig. 2. Drawing of Lung-men, Pin-yang Cave (Cave III), fore wall, left side, (From Mizuno and Nagahiro, fore wall, right side, (From Mizuno and Nagahiro, Lung-men, Fig. 19). Lung-m?n, Fig. 18).

26



Fig. 3. Side of the Gardner base, Boston, Isabella Stewart Gardner Museum.

1. Dragon Spirit King.3 Has a dragon head, and holds a lotus in his hand. (Fig. 3)

2. Wind Spirit King. Holds a wind bag in his hands, and has windblown hair. (Fig. 3)

3. Pearl Spirit King. Spews pearls out of his mouth into a dish held in his right hand.

(Fig. 3) 4. Fire Spirit King.4 Flames issue from his head

and he holds a flaming torch or va)ra in his

right hand. (Fig. 4) 5. Tree Spirit King. Holds a tree in his right

hand. (Fig. 4) 6. Mountain Spirit King. Sits in front of styl

ized rocks which symbolize a mountain, and holds a flaming jewel. (Fig. 4)

7. Fish Spirit King.5 Wears a fish around his neck. (Fig. 4)

8. Elephant Spirit King. Has an animal head, and holds a lotus in his right hand and a

flaming jewel in his left. (Fig. 5 ) 9. Bird Spirit King. Has a bird head, and holds

a lotus in his right hand. (Fig. 5 ) 10. Lion Spirit King. Wears a lion-head helmet,

and holds a lotus in his left hand and a

flaming jewel in his right. (Fig. 5)

The bottom two characters, Sh?n Wang, in

each cartouche make up the title, which was

translated as Spirit King, the best equivalent for the Chinese characters. Sh?n means a spirit or a god.6 Wang means a prince or king, a ruler.7

The first character in each cartouche denotes the individual Spirit King intended.

So little has ever been written about the Spirit Kings that it will be necessary to examine what ever material pertaining to them can be found before any definite conclusions can be drawn as to their origin, meaning, and position in the

Chinese Buddhist pantheon.

Fig. 5. Side of the base 543 A.D. Boston, Isabella Stewart Gardner Museum.

Fig. 4. Reverse of the stele and base, 543 A.D. Boston, Isabella Stewart Gardner Museum.

27



Fig. 6. Side of the Eastern Wei base, Philadelphia, Uni

versity Museum. Fig. 7. Side of the Eastern Wei base, Philadelphia, Uni

versity Museum.

The earliest known series of the Spirit Kings appears on the base of the carved wall surface at Lung-m?n in the Pin-yang Cave, dated by the Japanese in the years following 505 A.D.8

The deities have no inscribed titles, only their attributes to identify them.9 (Figs. 1 & 2) From left to right, they are:

1. Wind Spirit King. Holds a large bag and has windblown hair.

2. Dragon Spirit King. Looks a dragon in the eye and wears his hair in an Indian topknot.

3. Lion Spirit King.10 Wears a lion-head hel

met, and holds a flaming jewel in his right hand.

4. Tree Spirit King. Has a tree behind him and holds a flaming jewel in his right hand.

5. Fish Spirit King.11 Holds a fish in his left hand.

6. Bird Spirit King. Has a bird head. 7. Elephant Spirit King. Has an elephant head

and holds the trunk with his right hand. 8. Fire Spirit King. Flames issue from his head,

and holds a flaming object in his right hand. 9. Pearl Spirit King. Spews pearls from his

mouth into his hand.12

10. Mountain Spirit King. Holds a handful of rocks and trees which symbolize a mountain.

The Pin-yang Spirit Kings are clothed in a Chinese type of sleeved gown with a sash at the

waist. Some wear a collar caught at the shoulders

by two discs, as does the Bodhisattva in the Ma?

jusr? scene in the top register. The majority sit

in the position of 'royal ease',13 a typical Indian

position often given to Bodhisattvas and lesser

Buddhist deities, and are given no landscape or

architectural settings whatsoever.

Chavannes noted another group of Spirit Kings at the base of the carved wall surface in Cave 3 at the late Northern Wei site of Kung Hsien, dated around 531 A.D.14 Unfortunately he did not list them and only photographed three; the Fish, the Elephant, and the Bird Spirit Kings. In 1915, in connection with the Isabella Stewart

Gardner base, he recalled the Kung Hsien3 set

suggesting that both sets were identical as far as

he could remember.

The Kung Hsien costumes are still Chinese in character, but a bit different from those of the

Pin-yang deities. Each Spirit King is adorned with a scarf and a curved neckline with a pointed collar, frequently worn by Bodhisattvas. They sit cross-legged as do the musical gandharvas on

another part of the Kung Hsien wall base.15

A four legged stone base16 of the Eastern Wei

period17 in the University Museum, Philadelphia, has six Spirit Kings carved on two of its sides.

(Figs. 6 Si 7) Although there are no inscriptions, they can be identified by their attributes. The Pearl Spirit King holds a dish full of pearls in his lap, the Tree Spirit King holds a tree, the

Mountain Spirit King a handful of rocks, the

28



Wind Spirit King a bag, and the Elephant and Bird Spirit Kings have animal heads.

The type of costume worn by the deities on

both the University base and the Isabella Stewart Gardner base is identical, and consists of a scarf

and a loincloth, which leaves the belly and chest bare. Like the Lung-m?n deities, they wear cir

cular collars caught by discs at the shoulders.

The Spirit Kings on the Isabella Stewart Gardner base sit with crossed ankles, with the exception of the Mountain Spirit King, who sits with his feet out to the side in front of a pile of rocks.

The mountain deity on the University base sits in the same manner, but still holds his rocks in

his hand like his earlier Lung-m?n ancestor.

The Spirit Kings on the University base are sep- j^ arated by lotus flowers inserted between some of the deities, while those on the Isabella Stewart

Gardner base are separated only by the pendant inscriptions.

The Spirit Kings do not appear in any of the Northern Ch'i caves at T'ien-lung Shan, as far

as I know, but they do adorn the bases of several

Northern Ch'i stelae and can be found at the

late Northern Ch'i caves of Hsiang-t'ang Shan. A stele, excavated at Ch'ii-yang in Hopei pro

vince, dated Northern Ch'i by the Chinese, shows

four Spirit Kings on the bottom tier of an ex

tremely elaborate step pyramidal base. Presum

ably there are others on the opposite side of the

base. In spite of a poor photograph, the Elephant

Spirit King with his elephant head is visible to the right, and the Fish Spirit King with a fish around his neck on the left. It is almost impos sible to make out their costumes, but they ap

pear to wear loincloths and sit in the position of

'lordly ease.'

The base of a Northern Ch'i stele dated 56219 in the Cleveland Museum displays eight more

Spirit Kings on the back and two sides. (Figs. 8, 9, 10) Again with the help of their attributes,

we can identify the Mountain, Bird, Pearl, Tree,

Elephant, Fish and Wind Spirit Kings, plus one more whose identity is impossible to make out.

Another Spirit King group in the same style ap

pears on a Northern Ch'i base in the University Museum in Pennsylvania. (Fig. 11) The deities on both bases wear indentical costumes, which

consist of an upper garment with a curved

pointed collar, trousers, soft boots, and a scarf

that flares upwards like wings, except for the

Fish and Elephant deities whose scarves fall downwards. Some of the deities on the Cleve

ira

0?*%

J?L,,.

Fig. 8. Reverse of the Northern Ch'i base of 562 A.D., Cleveland, Cleveland Museum.

Fig. 9. Side of the Northern Ch'i base of 562 A.D., Cleve land, Cleveland Museum.

!Mr??

?^y^j^j&%k?

**%&<&- '*S*^ x

Fig. 10. Side of the Northern Ch'i base of 562 A.D., Cleveland, Cleveland Museum.

29



F/?. 7i. 57?/? 0/ //?? Northern Ch'i base, Philadelphia, University Museum.

land base sit crosslegged, while others sit in the

attitude of 'lordly ease' as they do on the Uni

versity base. The Fish and Elephant deities sup port their trunk and fish head respectively with the palm of their hand. The Mountain deity on the Cleveland base is now completely integrated into his rocky setting, instead of sitting merely in front of it as his counterpart does on the

Gardner base.

The back of another Northern Ch'i stele20 dis plays four Spirit Kings, each placed in a separate

rectangular space. They wear the same trousers

and soft boots seen on the Cleveland base, but without the ornate necklines. The Wind Spirit King now appears to wear a long pointed cap.

A photograph of the original base (Fig. 12) for the Northern Ch'i stele of 569 A.D.21 in Kan sas City show a group of deities who wear trous ers and soft boots similar to those worn by the

Cleveland Spirit Kings. The deity on the far left of the base holds a fish in his lap and the deity

Fig. 12. Back of the original base for the 569 A.D. stele in Kansas City, Nelson-Atkins Art Gallery.

on the right a wind bag. The rest of the attri

butes are difficult to identify, but the deities

undoubtedly represent a group of Spirit Kings.22

Mizuno identified several series of Spirit Kings in the Northern Ch'i Cave-temples of Hsiang

t'ang Shan.23 The complete set of ten Spirit

Kings, who can be identified by their attributes, appears at the base of the walls in Cave 7 in the

southern group.24 They are dressed in almost the same fashion as those on the Cleveland base. They

wear an upper garment with scarves flaring up, trousers, and soft boots. They sit cross-ankled or

in 'lordly ease', and are separated from each other

by typical Hsiang-t'ang Shan pillars crowned with lotus capitals. Six more Spirit Kings sit along the wall base under the main deities in

Cave 5,25 also in the southern group. The cos

tume, minus the scarf, remains the same as that worn by the group in Cave 7. Both Wind Spirit Kings wear long pointed caps.

Very handsome Spirit Kings adorn the bases of the Central Pillar-Stupas of the Middle and

North Caves (Figs. 14 & 15) in the northern group at Hsiang-t'ang Shan.26 Each occupies a

concave niche cut into the base rock. Flames decorate the arch of the niche. The costume here is much more elaborate than any yet discussed.

Each Spirit King wears soft boots, semi-loose

trousers, and a flying scarf, but, instead of the usual upper garment, each wears a coat of lea ther armor with circular breast plates and "a cord for fastening hide armor which binds the breast and waist tightly."27 Some wear floral headdresses with tiny flying ribbons like those that flow from the headdresses of Sasanian roy

alty. A few of the Spirit Kings are easily identi

30



F/?. i3. Reverse of the Northern Ch'i stele, Boston, Mu seum of Fine Arts, no. 50.1074.

fiable, but others seem to be without their usual

attributes, and hold only flaming pearls. Their posture is generally a type of 'lordly ease.'

One well-known set of Spirit Kings, those on

the base (Fig. 16) that now holds the Trubner stele28 in the Metropolitan Museum, will be dealt

with at this point due to a problem in dating. The base has a very short history compared with that of the stele. It was bought for the Metro

politan Museum in 1929 from the same Peking dealer who had originally sold the stele to Mr.

Trubner several years earlier.29 Since 1930, when the stele and base were erected as one monument,

they have both been dated Wei, 533-543, the dates given in the stele inscription for work done on a temple which the stele commemorates. The

style and date of the stele are irrelevant for this

study, and will be discussed in a future paper.

On the other hand, the style, costumes, and

general iconography of the Spirit Kings on the

Fig. 14. Relief at the base of the Central Pillar of the Mid dle Cave, Northern Hsiang-t'ang Shan (Mizuno,

Hsiang-t'ang-ssu, pi. LIV).

Fig. 15. Relief at the base of the Central Pillar of the North Cave. Northern Hsiang-t'ang Shan (Mi zuno, Hsiang-t'ang-ssu, pi. LX,A).

31



W?i?^BLiT'f^M ^^HR^Bh^^ >^ÎH?^B^^^K> iMB^HH^Snra^^^k?rcIsKA?ja^***? ^Hr*- 'sJimS????l?^^Mll^H^n^Bii^ûKir s?MSB^ '^NtaMfifllv ^ww ^. -^^1 I^H

^B^^^BÊ^HaMSM^âl^^^flF ^^SflpJi?^HflMfl^BBi JIll&?Î^^^Êm^^S9^^^B??B9 ^B ^v* **^B

^. <. *^^?BMB^BHI^^^^^fc?BBiS^S^^M?^5?J^^^^^BB^^^BB^^^^BB^^BBBii^B^^^B^Sw?

F/?. 76. Northern Ch'i base which holds the Trubner Stele, New York, Metropolitan Museum (A. Priest, Chinese Sculpture, Hi yhyi

' \ > r >

f--'

\\\\\\\\\\\\\\\\\\\\\\\\Wm"atrrflf ^mM fr^T -:**4lSr& i/Mw^?B&f ' - F '

&??? Îfci?r ,?-'r.J??iAKwf-5r?IK^?

^^^^Î^^^^^^SQBBBJBNLJ?flK^^M?B^!^îl^BBBSB BPS^^Bh^Bb* **

^^^^BhH^BIBHHBR?i?^

^^^^^^^^^B^^^^BBI^^^BBE^fiBH^B6HE'B?l^^^^^lBli[^^^^âiBB^B*^^3h?s

^^^BBBBi^BiHBBS3l5WiiSl?B^ H^^B^S?ft^^&fja^^^^^^BP^^Ki

BB5?e?N?8bK?I^^

Fig. 17. Details of the Northern Ch'i base, Metropolitan Museum, New York (A. Priest, Chinese Sculpture, pi I).

base are so closely related to the late Northern Ch'i sculpture found at Northern Hsiang-t'ang Shan that the Metropolitan base can be safely dated in the 570's of the Northern Ch'i period rather than to the preceding Wei eras.

Two Spirit Kings adorn each side of the Met

ropolitan base. The Wind Spirit King, wearing a long pointed cap and carrying a bag, sits to

the right of the central group on the front. On

the left is the Fire Spirit King30 with a flaming headdress holding a lotus. On one side of the base sits the Dragon and Lion Spirit King, from left to right, on the other side of the base, the Bird

and Fish Spirit Kings, wearing animal head hel mets.31 The Tree Spirit King and the Mountain

Spirit King (Fig. 17) sit on either side of a wild mountainous scene inhabited by monks on the

back of the base. The costumes, even the odd

volutes and cutouts below the waist, are almost

identical to those worn by the Spirit Kings in the Northern Hsiang-t'ang Shan Middle and

North caves. The Metropolitan deities are pro vided with blank pendant spaces intended for

inscriptions which separate them. Several hold

lotuses or have them growing behind them. The Mountain deity now rises out of the rocks and

the Tree deity has been given a hilly setting. None of the deities have animal heads, but wear

helmets in the shape of their animal attributes.

32



Six Spirit Kings appear in relief on the back

(Fig. 18) of the Freer white marble Buddha

figure, tentatively dated to the Sui period.32 They are adorned with round haloes, not worn by earl

ier Spirit Kings, and sit cross-legged on lotuses

separated by lollipop-shaped trees. Their cos

tume consists of a scarf, collar, and loincloth

rather than the trousers and armor seen in the

late Northern Ch'i period.

The octagonal base for a Sui Bodhisattva33 in

the Metropolitan Museum has a set of eight Spirit Kings set into deep niches. Their postures are

quite varied. The Mountain Spirit King even has

his back to the viewer. They are garbed in loin

cloths, and look more demonic than usual. The

Fire, Wind, Mountain, Tree, Fish and Dragon

Spirit Kings are easily identifiable, but the rest of the attributes are not easily read. The left

hand of one deity is ideally posed to hold a dish as the Pearl deity would. The remaining deity should be either the Elephant, Lion, or Bird

Spirit King, but all he has is what looks like a

vajra in his hand.

No recognizable Spirit King groups are found in Chinese Buddhist sculpture after the end of the Sui and beginning of the T'ang periods.

These deities must have been connected with

some particular Buddhist sect or belief which flourished during that short time-span. Four

figures clothed in scarves and loincloths kneel

and sit within the scooped openings of the dais

(Fig. 19) from the Chang-t? Fu gate shrine.34

They have haloes and carry flaming jewels as does the Fire Spirit King, the second from the

right, on the Freer Sui Buddha robe (Fig. 18). Four similar figures flank the incense burner at

the base of a Northern Ch'i or Sui bronze plaque in the I.S.M.E.O. Museum in Rome.35 Faint ves

tiges of the Spirit King group, one holds a fish,

Fig. 18. Detail of relief scene on the robe of the Sui Stone

Buddha, Washington, D. C, Freer Gallery of Art.

remain in these tiny unimportant Buddhist dei

ties. Apparently, the Spirit Kings did not disap pear suddenly from the Chinese Buddhist artistic repertory, but merely lost their potency after the turn of the century and were stripped of their

individual attributes, so that they became assimi

lated into the large group of nameless minor dei ties which occupy the lower registers of later

Buddhist sculptures.

It is now possible to make certain statements

about the Spirit Kings and their development during the sixth century in China. The official number is ten, but apparently it was perfectly permissable to show four, six, or eight. They are

always found at the base of the monument or

^fl^f^

f^iW ure.i6l?ii?.*,i4?

WZi? ???Mi

F/?. 79. Chang-t? Fu Gate shrine, Washington, D. C, Freer Gallery of Art.

33



cave wall they adorn, indicating a very low posi tion in the Buddhist hierarchy. Their earliest

known appearance in Chinese Buddhist art was

at Lung-m?n, then Kung-Hsien. They are fre

quently found on Eastern Wei, Northern Ch'i

and Sui monuments, but, never, as far as I can

tell, on monuments of the Western Wei or

Northern Chou periods. After the Sui period,

they are of little iconographie importance and mean almost nothing in the Buddhist art of

China.

Their attributes remain much the same

throughout the sixth century, with only minor

variations, such as the Fish deity wearing the

Fish helmet rather than holding a fish as he does on the Metropolitan Northern Ch'i base.

Their costumes do change radically though.

They first appear in typical Chinese Buddhist

clothing at their debut in the Pin-yang cave.

On the Eastern Wei monuments they are garbed in typically Indian loincloths and scarves, with

bare bellies and chests exposed, which reflect the

new Indianizing influence prevalent in Chinese

art during that period. Sometime after the mid

dle of the century in the early Northern Ch'i

period, their costume changed to baggy trousers

and soft boots which must ultimately derive

from the Near East. Then in the later North ern Ch'i, their costume was even more elaborated

into a strange type of warrior armor again with

ultimate affinities with the Near East.36 The

Spirit Kings seem to revert again to the Indian

type garments of loincloths, instead of the Near

Eastern inspired armor, in the Sui period.

At Lung-m?n and Kung-Hsien, the two

Northern Wei sites, the Spirit Kings are given no setting at all. In the Eastern Wei period, the

artists begin to separate them by placing lotuses

or inscriptions between them. They are not al

ways separated during the early Northern Ch'i, but at Hsiang-t'ang Shan their separation is com

plete in tiny niches with pillared arches. They remain separated in the Sui period. The Moun

tain Spirit King holds his rocky attribute in his hand at Lung-m?n. Toward the end of the East

ern Wei period on the Isabella Stewart Gardner

base, he sits in front of what Sullivan terms an

"Indian crag,"37 and is finally integrated into the

rocky setting during the Northern Ch'i period as seen on the Cleveland and Metropolitan bases.

The ultimate origin of the 'Spirit Kings' and who they were is not easy to determine. No

scholar of Buddhist texts and iconography has

shown any scriptural justification for them. Ori

entalists at various times called them minor na

ture gods, genii, demon-kings, or any such ap

pellation which seemed to appeal without any basis in fact or scripture.

The only names which seem to me acceptable are those translated by Chavannes from the in

scriptions beside the individual figures on the Isabella Stewart Gardner base,38 the sole monu

ment yet discovered where their names have been

recorded.39

Chavannes identified the Spirit Kings with the demon children of Hariti, on the basis of a

late copy of a Li Lung-mien painting in the

Mus?e Guimet.40 This derivation for the group is

unlikely. None of the Spirit Kings are the least bit diabolical, as Hariti's children should be, nor

are they in any way similar in attribute or

attitude.

Mizuno referred to the Spirit Kings only as

the "deities; the Pearl God, River God, Elephant God, Wind God, Tree etc.,"41 apparently based on Chavannes' translations. For some reason he

calls the Fish Spirit King the River God.

The mid-ninth century painting history Li-tai

Ming Hua Chi and early T'ang Chen-kuan Kung Ssu Hua Lu mention a painting of "Demon Dei

ties" by the Northern Wei priest, Chia-fo-t'o.

The early T'ang catalogue defines him as a west

erner, and the work as two scrolls of cartoons

(presumably done as models for the Chinese to

imitate).42 He appears to have been an Indian

and to have painted at Shao-lin-ssu, the moun

tain monastery on Sung Shan in Southern Ho

nan.43 He is also mentioned under the Liang,

intimating that he probably came to Sung Shan

by way of Nanking, the court of the Liang who

overthrew the Southern Ch'i in 502 A.D.

It is impossible to say for certain that Chia-fo

t'o's "demon deities" were the prototypes for the

Pin-yang Spirit Kings, but Shao-lin-ssu is near

Lung-m?n, where the Ten Spirit Kings first

appeared.

The Spirit Kings never appear in the earlier

Northern Wei site at Y?n-kang, which is far to

the north in Shansi province, nor can they be

found in Tun-huang44 on the northwestern

Chinese border or in the Northern Ch'i and Sui Caves at T'ien-lung Shan in Shansi.

34



On the other hand, there are certain deities in

the Chinese Buddhist pantheon which do appear in the earlier caves at Yiin-kang, and not at

Lung-m?n. The "Attack of Mara" is frequently

portrayed at Yiin-kang, and then not represented

again in Chinese art during the Six Dynasties period. The two ascetics, who hold a skull and

bird respectively and were extensively discussed

by Davidson,40 can be found at Yiin-kang and on numerous later sixth century stelae, but never

at Lung-m?n or Kung Hsien.

These regional differences in iconography would seem to be dependent on the source of in

fluence exerted on the Buddhist art of each area.

The major iconographie and stylistic innovations

at Yiin-kang were primarily Western,46 and can

be traced back through Central Asia ultimately to northwest India. Mara and his cohorts are

often represented in Gandharan art, and ascetics

of various types by the dozen,47 but no deities

comparable to the Spirit Kings.

In contrast, the sculpture at Lung-m?n dur

ing the early period was strongly Chinese in fla

vor, apparently due to "a wholesale appropria tion of artistic ideas from the Southern Ch'i," and the probable participation of newly ac

quired craftsmen from the long-important cen

ter, Shou-ch'un, a city not far north of Nan

king/8

With no Spirit Kings represented in the pri marily western inspired Buddhist art at Yiin

kang and Tun-huang, the evidence points to a

southern origin for this group of deities, as it

does for the style and composition of the donor

scenes49 carved above the Spirit Kings in the

Pin-yang cave at Lung-m?n.

Mr. Soper suggests that the renewed Indian

influences found in the art of the Southern courts of Six Dynasties China may have come

through the mediation of Southeast Asia.50 There fore certain stylistic and iconographical innova tions in Chinese Buddhist art, hitherto untrace

able in Indian and Central Asian art, may point to a Southeast Asian origin. This explanation certainly would seem to apply to the elusive

origin of our Spirit Kings.

There are no groups of deities in the early art

of India, either Buddhist or Hindu, which fit the description of the Spirit Kings. Genii, earth

spirits, known as Yaksas in Indian iconography, are legion in the folk-lore of early India. The

gates of Barhut and Sanchi abound with Yaksas

representing all sorts of earth forces; deities of

water, animal and vegetation world, most of

which were originally derived from the pre

Aryan, aboriginal traditions of India, and later

adopted by Hinduism and Buddhism. As early as the commentary on the Maha Prajnaparamita Sutra, xxxix, it is stated that "among the demon

deities there are those like the Yaksa 'Mi-chi

Chin-kang' and 'the mother of the Demon sons'

(i.e., Hariti) who have been able to find the way, and are now great Bodhisattvas."51

The origin of the Spirit Kings seems to point to an ultimate derivation from the various na ture deities found in the popular mythology of India. Mr. Priest thought that the Spirit Kings "suggest the ten Buddhist Kings of India."52 Pre

sumably he meant the ten sons of Devagabbh? from the Gbata-J?taka, who became the Ten

Kings of India until they destroyed themselves. Their names in Pali are Vasudeva, Bala-deva, Canda-deva (Moon), Suriya-deva (Sun), Aggi deva (Fire), Varuna-deva (The Heaven God),

Ajjuna (the Tree), Pajjuna (Rain-cloud), Gha ta-pandita (Ghee-sage), and Amkura (Sprout). "The story seems to contain a kernel of nature

myth",53 but this group has only a few in com mon with the Spirit King group. I do not think an identical Spirit King group can be found in Indian iconography, because I believe this group to be a collection of nature gods, who, though ultimately derived from earlier Indian types, were only combined into the 'Spirit King' group later on Chinese soil.

Direct contact with the Indian world, by the long used sea route was very rare for the South ern Ch'i, but it is clear that frequent intercourse

existed between the empire's Indonesian neigh bors, particularly Fu-nan (area of modern Cam

bodia). A king Jayavarman opened relations with the Southern Ch'i in 484, employing as

envoy "an Indian holy man."54 His successor,

Rudravarman, was represented at the Later Liang court. In other words, colonial culture in Indo nesia may have been the source for the transmis sion of Indian forms to the southern Chinese court of Nanking, which in turn came into con tact with the monestery at Sung Shan near

Lung-m?n.

Yaksas, nature deities, survived in Southeast Asian art, as evidenced by their appearance on

later Buddhist temples in Java and Cambodia,55 where we find animal-headed deities as well as

35



humanized ones with attributes such as trees,

rocks, etc. Their costume usually consists of a

loincloth, with an occasional decorative neck

lace. In Java, their function was to protect the

temple against intruders, so they were placed around the basement of the building under their

vigilance.

Unfortunately no examples of Southeast Asian

sculpture of the fifth century now remain, but

Mr. Soper, in a recent article in the Archives,56 has presented a very convincing case for its ex

istence.

The earliest Spirit Kings at Lung-m?n do not

have the physiognomy prescribed for most Yak

sas; curly hairlocks, ear ornaments, and eyes

bulging out of the head, but their distant kin

ship is nevertheless obvious in both iconography and the position they occupy, the base of the

carved wall surface.59 They too must have been

intended to function as guardians of the faith.

If Chia-fo-t'o's "demon deities" were not the

direct prototypes for the Lung-m?n Spirit Kings, it seems plausible to suggest that other deities,

with a nature myth background borrowed by Buddhism long ago in India, were transmitted

through Southeast Asia to the Southern Chinese court of Nanking either in cartoon form or in

the memory of some traveling monk. No Spirit

Kings exist on the few remaining Southern Ch'i

and Liang sculptures unfortunately to support this theory. Nevertheless, the origin of the Spirit

Kings points so strongly in this direction, that

the arrival at Sung Shan of an Indian monk, whether it be Chia-fo-t'o or some other unre

corded fellow, with cartoons or faintly remem

bered guardians with nature attributes, is a

tempting explanation for the sudden appearance of the Spirit Kings at Lung-m?n. This theory for the origin of the Spirit Kings coincides with

Mr. Soper's recent thesis "that the major devel

opments in Chinese Buddhist iconography were

initiated by the Southern Church."57

Nature deities also abound in the folk-lore and

early art of China, but no group can be found to fit the Spirit Kings. The title, Spirit King, seems to have been a title invented by the sixth

century Chinese, who apparently fashioned the

Spirit King group to use as additional guardians and protectors of the Buddhist faith and the

monuments dedicated to it, using Yaksa-types for their models. Undoubtedly, the Chinese modified the foreign prototypes. The costumes of the

Pin-yang Spirit Kings are typically Chinese, and the dragon, as he appears beside the Dragon Spirit King, is a purely Chinese mythological creature. Yet, the Dragon Spirit King's hair is dressed in an Indian topknot, which points to an

ultimate inspiration from the Indian world.

The Wind deity holding his bag, and the use of an animal head helmet on the Lion Spirit

King at Lung-m?n are neither Indian nor Chi nese in character, but presumably indirect des cendants from ancient Hellenistic Aeolus and

Hercules figures. Mr. Soper noted Persian mis sions at Lo-yang in 507, 517,58 etc., so Buddhist clerics from these missions must have reached

nearby Sung Shan monastery, which would ex

plain these two western iconographie intrusions where neither the style nor the rest of the icon

ography has any western flavor at all.

NOTES 1. This article was originally part of a Master's thesis submitted

to the Institute of Fine Arts, New York University, June 1962. I would like to express my gratitude to Dr. Alexander

Soper, my advisor, for his continued advice and encourage ment.

2. "Une Sculpture Bouddhique de l'Ann?e, 543, Ars Asi?tica, vol. II, Aug. 1915.

3. Chavannes translated the name of this deity as "The Spirit King of the Nagas". The first character, which denotes the

individual deity intended, is actually lung, so the first deity must be Lung Sh?n Wang, the Dragon Spirit King. The

N?ga is the Indian term for snake.

4. The cartouche is broken and the top character is missing, but Chavannes suggests a Fire deity because of the fire attributes.

5. The first character is broken off, but the fish attribute is un

mistakable.

6. Mathews' Chinese-English Dictionary, American Edition, Cam

bridge, Mass., p. 790, no. 5716.

7. Ibid., p. 1043, no. 7037.

8. Alexander Soper, "South Chinese Influence on the Buddhist Art of the Six Dynasties Period, Bulletin of the Museum of Far Eastern Antiquities, no. 32, I960, p. 79.

9. Mizuno and Nagahiro, A Study of the Buddhist Cave-temples at Lung-m?n, Honan, Tokyo, 1941, figs. 18 and 19. Unfor tunately these wall decorations in the Pin-yang cave are known mostly by artist sketches and a few poor photographs.

10. The headdress is difficult to see in detail, but is similar to the Isabella Stewart Gardner Lion-Spirit King's headdress. (Fig. 5).

11. The drawing is a bit inaccurate when compared to a photo graph of the same deity, see Tokiwa and Sekino, Buddhist

Monuments in China, Tokyo, 1925, vol. II, pi. 60. The fish is actually being held in the hand and does not extend around the neck.

12. The deity may originally have held a dish which was unre corded by the artist who copied the cave wall decorations.

13. E. Dale Saunders, Mudr?, New York, I960, p. 128.

14. Chavannes, Mission arch?ologique dans la Chine septentrion ale, Paris, 1905-15. Kung Hsien, Cave II, pi. CCLXXI, fig. 406.

15. Ibid., fig. 404.

16. Siren, Chinese Sculpture, London, 1925, pi. 288.

17. The author disagrees with Mr. Siren, ibid., p. 78-79, and other scholars who consider this piece to be an archaistic

36



work of some later date. The rather calligraphic style of the

carving, and the iconography of the Spirit Kings are obviously related to the same traditions which produced the Gardner base. I can see no reason for its being carved later unless it is a deliberate forgery or an exact copy of an already exist ent work. It fits easily into the early Eastern Wei period, probably having been carved in the late 530's, a few years before the Isabella Stewart Gardner base which displays a

more advanced treatment of the Mountain Spirit-King's set

ting. Mr. J. Leroy Davidson advanced a similar theory for the authenticity of the stele which the Philadelphia base

holds. (Davidson, "A Problem in Skulls," Parnassus, vol. XI, Jan. 1939, p. 34-36).

18. Wen Wu, 1958, no. 10, p. 66, and Nagahiro, T., "On the Stone figures excavated at Ch'ii-yang," (Jap.-Eng. summary), Bijutsushi, 13,1, June 1963, pis. MI.

19. J. Leroy Davidson, The Lotus Sutra in Chinese Art, New

Haven, 1954, nos. 18 & 19.

20. S. Matsubara, Chinese Buddhist Sculpture, Tokyo, 1961, p. 110.

21. I am very grateful to Mr. Laurence Sickman, who very kindly supplied me with photographs of this base, which he knew in

China but was unable to acquire for the museum.

22. It is worth noting two bases carved with Spirit Kings, who are not so readily recognizable. The deities on a Northern Ch'i base (Fig. 13) in Boston, (see Sir?n, op. cit., vol. Ill, pi. 217B), possess no attributes at all, but sit in the same manner and costume minus the scarf as the Spirit Kings do on the Cleveland Museum base. A base in Northern Ch'i

style at Cologne, Sir?n, ibid., pi. 248, is carved with four creatures which look like a parody on the Spirit King theme.

The artist seems to have misunderstood the Pearl Spirit King's attribute, and the others have none. The Boston stele appears to be just a pedestrian piece of sculpture where the artist did not add much detail in the way of iconography, but the

Cologne piece is undoubtedly either the result of recutting or of the scene being added in modern times.

23. S. Mizuno and T. Nagahiro, The Buddhist Cave-Temples of

Hsiang-t'ang Shan, Kyoto, 1937. Hsiang-t'ang Shan mountain lies across the border between Honan and Hopei provinces.

The southern group of Cave-temples lies in Hopei, the north ern group in Honan.

24. Ibid., pi. XXXIV, XXXVI, XXXVII.

25. Ibid., pi. XXIV.

26. Ibid., pi. LUI, LIV, (Middle Cave) and pi. LIX, LX (North Cave).

27. S. Mizuno, Chinese Stone Sculpture, Tokyo, 1950, p. 21.

28. A. Priest, Chinese Sculpture in the Metropolitan Museum of Art, New York, 1944, p. 30-33, and pis. XL-LII.

29. The base was given the same provenance as the stele, but was not mentioned by Omura Seigai, Shina Bijutsushi, Ch?so-hen, Tokyo, 1915, p. 262-3, when he published a rubbing of the

inscription of the stele. This whole problem of the stele's

provenance will be gone into at a later date, since it will be come obvious that the stele and base are not of the same

period. 30. This deity was mislabelled the Pearl Spirit King, (Priest,

op. cit. pi. XLIX), but its attributes point to Fire.

31. Priest, ibid., pi. LI, the figure to the right, mislabelled the

Tiger Spirit King (no such deity in the group), should be the Lion Soirit King. Plate LII, should read, from left to

right, the Bird and Fish Spirit Kings, instead of Fish and Lion. Madeleine Hallade, Arts de l Asie Ancienne, Paris, 1956, p. 31, also mislabelled the Lion Spirit King, a Tiger

genii. 32. Freer, no. S.9, C. W. Bishop Collection. Published Wen Wu,

1959, no. 2, p. 5, and dated T'ang by the Chinese. The floral motifs and incense burner carved on another part of the fig ure are far more characteristic for a late sixth century than a

T'ang date, as is the Vimalakirti scene carved on the front of the figure.

33. Priest, op. cit., pi. LVII. The Kuan-yin is a typically ornate

type of Bodhisattva figure frequently associated with the late Northern Ch'i and Sui styles.

34. Extensively published by G. Scaglia, "Central Asians on a Northern Ch'i Gate Shrine", Artihus Asiae, vol. XXI, no. 1, p. 9-28. The identification of these four figures as Spirit Kings suggests a Sui rather than Northern Ch'i date for this

piece, because their costume consists of a loincloth and scarf, as it does on other Sui examples, instead of the typical North ern Ch'i armor.

35. Matsubara, op. cit., pi. 178b. This plaque (ISMEO 670) was a gift from Ambassador Auriti to the ISMEO Museum

in Rome. It was dated T'ang by Matsubara, but Alexander

Soper, who is cataloguing the Auriti collection, dates it North ern Ch'i or Sui, due to its iconographie similarity to the

paradise frieze from Hsiang-t'ang Shan. This is enforced by the presence of at least one identifiable Spirit King, who is still represented in the late sixth century manner.

36. Miss Scaglia, op. cit., has already pointed out the presence of Near Eastern peoples in Chinese sculpture, as well as their actual existence in China itself during the second half of the sixth century.

37. Michael Sullivan, The Birth of Landscape Painting in China, Los Angeles and Berkeley, 1962, p. 134. Mr. Sullivan mis dated the Trubner stele 533-534 instead of 533-543, and

equated the mountains with those on the Isabella Stewart Gardner (A.D. 543) base. It is obvious now that the Metro

politan base is Northern Ch'i in date, so the rocks surround

ing the Mountain Spirit King are really a later development of those on the 543 A.D. base and not contemporary.

38. See Chavannes, Ars Asi?tica, op. cit., and page 1 of this article.

39. The titles given the deities by Priest, op. cit., p. 33, were

merely the Chinese equivalents for the English titles he gave them, and not derived from any written source. One, the

tiger, doesn't exist at all.

40. Chavannes, op. cit., Ars Asi?tica, IL Hariti was supposed to have 500 children, many of which do have animal and atmos

pheric attributes.

41. Mizuno, op. cit., Hsiang-t'ang Shan, p. 4, and Mizuno, Chinese

Stone, op. cit., p. 20.

42. Soper, Literary Evidence for Early Buddhist Art in China, Ascona, 1959, p. 241.

43. Ibid., p. 81, no. 29.

44. Sir?n, Chinese Painting, London, 1956, vol. Ill, pi. 31. Demons and spirits of thunder and wind do occur in the Northern Wei cave 249 (p. 101), but they are not related

to the Spirit King group.

45. L. Davidson, The Lotus Sutra in Chinese Art, New Haven, 1954, p. 45-49.

46. Soper, Bulletin of the Museum of Ear Eastern Antiquities, 1962, op. cit., p. 12. The later influence from the Southern

Ch'i in the 480's does not concern us here.

47. Harald Ingholt, Gandharan Art in Pakistan, New York, 1957.

48. Soper, Bulletin of the Museum of Ear Eastern Antiquities, op. cit., p. 77-79. Shou-ch'un fell to the Liang in 500.

49. Ibid., p. 78.

50. Soper, "Two Stelae and a Pagoda on the Central Peak, Mt.

Sung," Archives of the Chinese Art Society of America, XVI, 1962, p. 41.

51. Soper, Literary Evidence, op. cit., p. 241.

52. Priest, op. cit., p. 33.

53. E. B. Cowell, The J?taka, Vol. IV translated by W. H. D.

Rouse, M. A., p. 50-57.

54. Soper, Bulletin of the Museum of Ear Eastern Antiquities, op. cit., p. 90.

55. F. D. K. Bosch, "On some groups of Yaksa Figures in Indo nesian and Khmer Art", Artibus Asiae, XXIX, 3, 1959, fig. 10, and figs. 3 and 4.

56. Soper, Archives, op. cit.

57. Soper, ibid., p. 46. Suggested in discussing the advanced fig ure style of the Sung-yang ssu stele of 535, which, according to the Japanese description, also has some Spirit Kings on the bottom (p. 41).

58. Soper, Bulletin of the Museum of Ear Eastern Antiquities, op. cit., p. 81.

59. Six of the Spirit Kings appear on the base of a small votive bronze stele, dated 560-570, now in a private collection in

Japan. Each Spirit King has been provided with a cartouche, which, unfortunately, has never been inscribed. Further com ment on them is rendered impossible due to poor photo graphs. (J. Hackin and others, Studies in Chinese Art and Some Indian Influences, London, 1936, fig. 35.)

37



the spirit kings in sixth century chinese buddhist sculpture

Documents