made by the empire: wang xizhi￢ﾀﾙs xingrangtie … · ex em pli fy ing moral rec ti tude and...

d b th p r : n X zh X n r n t nd t P r d xrt n rn

r h v f n rt, V l 6 , N b r 2, 20 , pp. ( rt l

P bl h d b n v r t f H PrD : 0. .20 6.000

F r dd t n l nf r t n b t th rt l

Access provided by Princeton University (3 May 2016 14:35 GMT)

http : .jh . d rt l 6 4

https://muse.jhu.edu/article/614911

An Iconic Ar ti fact

No Chi nese cal lig ra pher is more re vered than the “sage of cal lig ra phy” (shusheng 書聖) Wang Xizhi

王羲之 (303–ca. 361), an aris to crat whose pres ti gious Langye Wang 瑯琊王 ref u gee fam ily from Shandong helped set up the Eastern Jin dy nasty at Jiankang in 317. Among other des ig na tions, Wang Xizhi held the mil itary ti tle “Commanderoftheright” (youjun 右軍) and is known to have re tired late in life (354 or 355) from high of fice be cause of a po lit i cal ri valry with Wang Shu 王述 (303–368).1 Despite his long back ground of polit i cal in volve ment, Wang Xizhi’s tra di tional im age is grounded in his Daoist in ter ests as well as his famed Lanting ji xu 蘭亭集序 (Preface to the or chid pa vil ion col lec tion), in which he ap pears to speak in a per sonal voice free of so cial or po lit i cal am bi tion.

As noted by Eugene Wang, con nois seurs through the ages have constructed from the ap pre ci a tion of his cal lig ra phy a per sona of the cel e brated ar tis tic Wang Xizhi that is com pli cated and contradicted in mul ti ple ways by other sources, in clud ing his let ters, an ec dotes in texts such as Liu Yiqing’s 劉義慶 (403–444) Shishuo xinyu 世說新語 (A new ac count of the tales of the world), and Wang’s of fi cial bi og ra phy in the Jinshu 晉書 (History of the Jin), writ ten nearly three cen tu ries af ter his death and graced with an en co mium by Emperor Tang Taizong 唐太宗 (r. 626–649).2 As a re sult, views and eval u a tions of the his tor i cal Wang Xizhi dif fer widely in tra di tional writ ings since the fifth cen tu ry; yet in the dom i nant aes thetic ap pre ci a tion of his “Daoist” cal lig ra phy, at trib uted mostly to the late years of his

life,3 Wang ap pears to em body the ide al ized dis tance in re la tion to the im pe rial state that was highly prized among the early me di e val Chi nese in tel lec tual elite for ex em pli fy ing moral rec ti tude and per sonal in teg ri ty. This per sona is not en tirely dis sim i lar to the equally constructed one of Tao Qian 陶潛 (365–427).4 Coincidentally, the first known col lec tor of Wang Xizhi’s callig ra phy ap pears to have been Huan Xuan 桓玄 (369–404), a pa tron of other cal lig ra phers and paint ers but also of Tao.5

In the later imag i na tion, shaped by Wang’s self rep re sen ta tion as much as by the re cep tion his tory of his art,6 his aes thetic oeu vre is char ac ter ized through this par tic u lar re la tion to the im pe rial state, that is, the di a lec ti cal stance of re treat and af fir ma tion. Although Wang did pro duce for mal cal lig ra phy for pub lic pur poses,7 his sur viv ing oeu vre—aside from Lanting ji xu—was al ready greatly cel e brated for his pri vate (or ap par ently pri vate) let ters in early Tang times.8 The let ters, regarded as au then tic ex pres sions of their au thor’s emo tional self and praised purely for their cal lig ra phy, were cherished by the court as much as by the learned elite of aris to crats, schol ars, and af fil i ated of fi cials. From this per spec tive, Wang’s oeu vre was not cre ated in op po si tion to the state but still qui etly gave voice to a hu man ex is tence out side of fi cial dom. Within gen er a tions af ter Wang’s death, how ev er, this voice came to de pend largely on the im pe rial court for its pres er va tion and trans mis sion. Thus, Wang’s ar tis tic cre a tion and the per sona it rep re sents—as op posed to the his tor i cal Wang Xizhi who served for de cades in of fice—stand both within and out side the Chi nese state, a pos ture not un com mon among aris to crats of his

Made by the Em pire: Wang Xizhi’s Xingrangtie and Its Paradoxes

Martin KernShanghai Normal University and Princeton University

Abstract

The es say ex plores the his tory of one of the most cel e brated pieces of Chi nese cal lig ra phy, an early Tang trac ing copy of Ritual to Pray for Good Harvest (Xingrangtie) by Wang Xizhi (303–ca. 361). The Tang pa per slip with its fif teen char ac ters is merely 8.9 cm wide but em bed ded into a scroll of 372 cm, with seals and col o phons from three Chi nese em per ors and an il lus tri ous line of con nois seurs. The es say ex plores the nu mer ous paradoxes in the re cep tion of both the scroll and the Wang Xizhi per so na, with the scroll ex em pli fy ing the constructed na ture of Wang’s per sona in its di a lec tic re la tion ship to Chi nese im pe rial cul ture.

KEYWORDS: Chi nese cal lig ra phy, Wang Xizhi (303–ca. 361), Qianlong Emperor, Song Huizong, Dong Qichang, Xingrangtie, Ritual to Pray for Good Harvest, im pe rial cul ture, seals, col o phons, au then tic i ty, let ters, copy

118 ARCHIVES OF ASIAN ART

time; or, more pre cise ly, they oc cupy a place in which the “out side” na ture was itself appropriated by, and sublated with in, the value sys tem of court cul ture.

Under Tang Taizong, the im pe rial pal ace col lec tion re port edly in cluded 2,290 Wang Xizhi pieces—that is, orig i nals and cop ies.9 During Song Huizong’s 宋徽宗 (r. 1100–1126) reign, cen tu ries af ter the col lapse of the Tang im pe rial court, more than 3,800 works by Wang Xizhi and his son Wang Xianzhi 王獻之 (344–ca. 386) were re port edly in the im pe rial col lec tion; at least 243 by the fa ther are men tioned in Huizong’s cal lig ra phy cat a logue Xuanhe shupu 宣和書譜 (Xuanhe cal lig raphy cat a logue).10 It is un clear how many (if any) of these works were orig i nals as op posed to cop ies created in a range of dif fer ent styles and tech niques, but they vastly dom i nated the Tang and Song im pe rial collec tions.11

Today, no orig i nal by Wang Xizhi is known to have sur vived. The clos est we get to Wang’s hand writ ing is a very small num ber of trac ing cop ies pre sum ably from the sev enth or eighth cen tu ry.12 The only such copy out side col lec tions in China and Japan is held in the Princeton University Art Museum, a gift from John B. Elliott, who had pur chased it shortly be fore 1970 in Japan.13 First men tioned in the ninth cen tu ry, the scroll is known as Xingrangtie 行穰帖, a ti tle taken from its third and fourth char ac ters and con ven tion ally translated as “Ritual to Pray for Good Harvest.” The fif teen char ac ters of Wang Xizhi’s text are writ ten on a pa per slip 24.4 cm high and 8.9 cm wide; in its cur rent mounting, the slip is em bed ded into a scroll 372 cm in length, bear ing la bels, col o phons, and seals from the twelfth through the twen ti eth centuries.

Wang Xizhi’s text is not only very short; it is also very dif fi cult to de ci pher (Fig. 1). Dong Qichang’s 董其昌 (1555–1636) tran scrip tion—now part of the scroll—dif fers in four char ac ters from Zhang Yanyuan’s 張彥遠 (ninthcen tu ry) first pub li ca tion of the text:14

Zhang Yanyuan: 足下行穰久人還竟應快不大都當任

Dong Qichang: 足下行穰九人還示應決不大都當佳

Neither tran scrip tion ren ders an eas ily in tel li gi ble text, and nei ther may be cor rect with regard to the fi nal char ac ter.15 According to Zhang, the fif teen char ac ters are only the first half of a lon ger let ter of thir tytwo char ac ters, which he tran scribes in to to; the ear li est extant re pro duc tions of the sec ond half are found in thirteenthcen tury rub bings col lec tions and then again in Dong’s cat a logue Xihong tang fatie 戲鴻堂法帖 (Model calligraphies from the Hall of Playing Geese) of 1603.

Fig. 1. Wang Xizhi (China, 303–ca. 361), Ritual to Pray for Good Harvest (Xingrangtie) (de tail). Tang dy nasty trac ing copy. Handscroll, ink on yinghuang pa per, 24.4 × 8.9 cm. Complete scroll: ink on pa per and silk, 30 × 372 cm. Princeton University Art Museum, Bequest of John B. Elliott, Class of 1951. Photography: Bruce M. White.

MARTIN KERN • Made by the Em pire: Wang Xizhi’s Xingrangtie and Its Paradoxes 119

Unfortunately, the sec ond half of the let ter includes two un de ci pher able char ac ters Zhang could not tran scribe, and the two halves dif fer clearly in their cal lig ra phy, which may or may not be due to lib er ties taken in copying (Fig. 2).16 As both are cop ies, we can not de cide whether both (or ei ther) rep re sent Wang Xizhi’s hand, nor is it cer tain—de spite Zhang’s claim—that they indeed be long to geth er. Leaving the sec ond half of Zhang’s text aside, Dong Qichang’s ver sion may be parsed and trans lated as fol lows,17

足下行穰。九人還。示應決不。大都當佳。You, Sir, had a sac ri fice performed to ward off bad har vest. Nine peo ple returned [to you] to re port whether [the spir its’] re sponse was de ci sive or not. [I pre sume] al to gether things should be fi ne.

Fig. 2. The two halves of the let ter. Left: Wang Xizhi (at tribut ed), rub bing from Model calligraphies from the Hall of Playing Geese (Xihong tang fatie, 1603), part 16. Right: Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie) (de tail).

This is one plau si ble trans la tion of a pos si bly incom plete text that has sig nifi cant var i ants in its ear lier tran scrip tion. We do not know whether or not the vari ants noted above are merely dif fer ent in ter pre ta tions of the same char ac ters or the re flec tion of a copy ist’s mis takes or de lib er ate changes.18 Moreover, Zhang Yanyuan’s ver sion would not only be twice as long but al so, be cause of the con tent of the sec ond half, would be parsed and interpreted very dif fer ent ly. Here, the let ter would not con cern some sac ri fi cial rit ual but rather the ques tion of whether or not some one (who is not named) might be suit able for, and should be appointed to, an of fi cial po si tion. Nobody can de cide which ver sion to fol low—nor has this fun da men tal am bi gu ity ever dis turbed the long connoisseurial tra dition. Whatever the case, the mean ing of Xingrangtie ap pears quo tid i an: a terse com mu ni ca tion with someone anon y mous, largely de scrip tive and in clud ing—in my in ter pre ta tion above—a brief flash of per sonal concern at the end.

To sum ma rize, Xingrangtie is a triv i al, partly un intel li gi ble, tiny pa per slip of fif teen char ac ters of unknown prov e nance that is ei ther the copy of a frag ment or the frag ment of a copy. Its known his tory be gins only with Huizong’s seals from the early twelfth cen tury—some eight cen tu ries af ter Wang Xizhi’s death— followed by an other gap of five cen tu ries un til the late Ming. And yet, judg ing from the var i ous book cov ers it graces, Xingrangtie—aside from the cel e brated Lanting ji xu, which also ex ists only in copy—is of ten taken to rep re sent some of the fin est el e ments of the Chi nese calli graphic tra di tion al to geth er.19 Why, and how, would such a ti ny, ob scure, and seem ingly in sig nifi cant ar tifact, com pro mised in mul ti ple ways and yet al ways preserved and protected through the ages, rise to such iconic stat ure?

The Se quen tial Or der of the Princeton Scroll

In ad di tion to Wang Xizhi’s fif teen char ac ters, the scroll con tains three la bels (by Huizong, the Qianlong 乾隆 Emperor, and Dong), one tran scrip tion of Wang’s text (by Dong), and eight col o phons (three by Dong, one by Sun Chengze 孫承澤 [1592–1676], three by Qianlong, and one by Zhang Daqian 張大千 [1899–1983]). In ad di tion, the out side of the scroll con tains the la bel “Yuti Jin Wang youjun Xingrangtie” 御題晉王右軍行穰帖 (Imperially ti tled Xingrangtie by the Jin dy nasty Commanderoftheright, Wang), pos si bly added un der the Jiaqing 嘉慶 Emperor (r. 1796–1820) who also left a seal in side the scroll. In its cur rent mount ing, done at the New York Metropolitan Museum of Art in the

Fig. 3. Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie). Tang dy nasty trac ing copy. Handscroll, ink on pa per and silk, 30 × 372 cm. Princeton University Art Museum, Bequest of John B. Elliott, Class of 1951. Photography: Bruce M. White.


1980s, the scroll has nine sec tions from right to left (Fig. 3):20

1. A pa per slip mounted on silk, bear ing the la bel from Qianlong’s court: “Wang Xizhi Xingrangtie zhenji” 王羲之行穰帖真蹟 (Genuine traces of Wang Xizhi’s Xingrangtie), writ ten in eight large char ac ters and followed im me di ately be low by two col umns of the al to gether six smaller char ac ters shenpin 神品/neifu mibao 內府秘寶 (di vine grade/closedoff trea sure of the Palace Treasury), in di cat ing that the scroll was of the highest grade and kept in the im pe rial pal ace. Below is a small im pe rial seal.21

2. A sep a rate sheet of silk with four seals by Zhang Daqian, who ac quired the scroll in 1957 in Hong Kong and remounted it; the fifth seal is by Zhang’s wife.

3. Another sheet of silk with a sec ond nar row la bel by Dong: “Wang youjun Xingrangtie” 王右軍行穰帖 (Xingrangtie by the Commanderoftheright, Wang); to its left is a large seal by Qianlong and be low both are sev eral smaller col lec tors’ seals.

4. The cen ter piece of the scroll: Wang Xizhi’s fif teen char ac ters on yel low pa per, pre sum ably Tang yinghuang 硬黃 (hard ened yel low) pa per,22 flanked by two pieces of sim i larly dark pa per inscribed with two col o phons by Qianlong and densely cov ered with seals; the col o phon on the left is dated 1748. Barely vis i ble on the seam be tween the Tang pa per and the dark pa per to its right, Huizong’s la bel of faint gold char ac ters, writ ten on a slip of pa per “a dull sil ver in col or,”23 is spliced in at the top: “Wang Xizhi xingrangtie” 王羲之行穰帖 (Wang Xizhi’s Xingrangtie).

5. On white Song pa per, Huizong’s large pal ace seal, followed by an other Qianlong col o phon also dated 1748; there af ter, still on the same pa per, Dong’s sin gle line of tran scrip tion.

6. A nar row sheet of silk with Dong’s first (undat ed) col o phon.

7. Dong’s sec ond and third col o phons on pa per, dated 1604 and 1609.

8. Largely empty pa per with Sun Chengze’s col ophon on darker pa per at the end.

9. On pa per, two col o phons by Zhang Daqian, accom pa nied by five seals (of their thir tyfour on the scroll) by Zhang and his wife, with an other five seals by Li Jingmai 李經邁 (1876–1940) and

his fam i ly, who had sold the scroll to Zhang in 1957.24

Sections 1, 2, and 9 have no seals across their seams; Sections 3 to 8 are all connected by seals on their seams. This shows that Zhang’s (and any sub se quent) remounting af fected only the be ginning and the end of the scroll.

Colophons and Seals

The seals re veal the scroll’s for mat at var i ous stages of its his to ry. The spa tial se quence of texts does not rep re sent chro nol o gy: first, Qianlong and Dong’s la bels in Sections 1 and 3 come from later remountings, when they were moved from the out side to the in side of the scroll; sec ond, Qianlong inscribed his col o phons into the empty space cre ated by Huizong more than five cen tu ries ear li er; third, Zhang’s seals at the be gin ning and his col o phons and seals at the end rep re sent the fi nal col lec tor traces.

Seals and col o phons were im por tant for authenticating works, but they were not in vi o la ble; they could be cut off, displaced, or even moved from one work to anoth er, as Mi Fu 米芾 (1052–1107) has al ready noted for var i ous Wang Xizhi pieces. An ig no rant mounter could de prive a scroll of its his tor i cal sig nifi cance, while a cunning one could cre ate forg er ies.25 Because of its com plete lack of orig i nals, some schol ars con sider the en tire Wang Xizhi oeu vre a Tang in ven tion.26

The rou tine flex i bil ity with which the in di vid ual parts of a scroll were rearranged by remounting, copying, and re pro duc tion in wood or stone carv ings can be gleaned from the seals, la bels, and spac ing of char ac ters found in two Xingrangtie rub bings (Fig. 4): one from Wu Ting’s 吳廷 (fl. ca. 1575–1625) Yuqing zhai fatie 餘清齋法帖 (Model calligraphies from the Yuqing Studio; 1614), the other from Qianlong’s Sanxi tang fatie 三希堂法帖 (Model calligraphies from the Three Rarities Hall; 1747–1750).27 Both rearrange the seals as well as the la bels, though the rub bing from Qianlong’s cata logue di verges most dra mat i cally from the scroll—which by 1748, at the lat est, was also owned by him.28 These rub bings ad ver tise only the pres ence of cer tain seals and la bels, not their ac tual place ment on the scroll. On Xingrangtie, Qianlong’s writ ing im me di ately left and right of Wang Xizhi’s cal lig ra phy cre ated an in di vid ual bond be tween the ear lier “cal lig ra phy sage” and the lat terday Man chu em per or, whereas Qianlong’s im perial col lec tion of stone en grav ings, com plete with Dong’s la bel and Huizong’s seal yet with out Qianlong’s own traces, depersonalized the scroll, mark ing it as an ar ti fact of im pe rial and not per sonal rep re sen ta tion. The


Xingrangtie scroll and its re pro duc tion in the “Three Rarities” col lec tion of rub bings thus rep re sent “the king’s two bod ies”—the per sonal “body nat u ral” (with the scroll) and the in sti tu tional “body pol i tic” (with the rub bing)—an a lyzed in Kantorowicz’s clas sic study of me di e val Eu ro pean sov er eign ty.29

A few de cades af ter Wang Xizhi’s death, the authen tic ity of the many works at trib uted to him was al ready in doubt.30 We must con sider the cen ter piece of Xingrangtie—Wang’s fif teen char ac ters flanked by Qianlong’s col o phons, and all three texts surrounded by more than thirty seals—against these anx i eties. Each seal sig nals own er ship or ap pre ci a tion, but many do more: they au then ti cate the scroll and, placed on the seams between the dif fer ent phys i cal parts, en sure its in teg rity as a whole. The seals are the guard ians of the cul tural tra di tion

cre ated around Wang Xizhi’s words. For Xingrangtie, this stillvis i ble tra di tion—its his tory of trans mis sion through the ages—be gins only in the twelfth cen tu ry, with the earli est seals of Emperor Huizong. Whatever his tory transpired be fore that, pre sum ably be gin ning with a copy ist in the early Tang and pos si bly au then ti cated by seals from the sev enth or eighth cen tury on ward, was cut off and erased be fore the scroll en tered the Song court. Huizong’s Xingrangtie was al ready a frag ment, whether or not it was orig i nally less than half of the larger text recorded by Zhang Yanyuan.

Song Huizong’s Scroll

No later than un der Huizong, the Tang pa per with Wang Xizhi’s char ac ters was flanked with dark empty

Fig. 4. The Qianlong and Wu Ting rub bings. Left: rub bing from Model calligraphies from the Three Rarities Hall (Sanxi tang fatie), 1747–50. After Robert E. Harrist Jr., “A Letter from Wang Hsichih and the Culture of Chi nese Calligraphy,” in The Embodied Image: Chi nese Calligraphy from the John B. Elliott Collection, ed. Robert E. Harrist Jr. and Wen C. Fong (Princeton, NJ: Princeton University Art Museum, 1999), 255. Right: rub bing from Model calligraphies from the Yuqing Studio (Yuqing zhai fatie), 1614. After Yuqing zhai fatie 餘清齋法帖 (Beijing: Beijing guji chubanshe, 2003).


pa per left and right, and the en tire work, now nearly square, was mounted onto a silk back ing. This pa per could have been spe cially col ored or harvested from an older source, pos si bly also Tang. Farther left, a brighter sheet of pa per bears Huizong’s large pal ace seal. This large seal is part of the “Xuanhe pro gram” iden ti fied by Barnhart and elab o rated on by Ebrey, as is Huizong’s la bel in gold char ac ters and the round dou bledragon seal be low it (Fig. 5).

The scroll con tains six Huizong seals representing, to gether with the la bel, the nearly com plete “Xuanhe pro gram” (aside from a miss ing gourdshaped seal on the right, discussed be low) in its fixed spa tial ar rangement: the large square “Neifu tushu zhi yin” 內府圖書之印 (Seal of paint ings and calligraphies of the Palace Treasury) on the bright pa per to the left; the round doubledragon seal on the seam be tween the Tang pa per and the dark pa per to its right; and four rect an gu lar seals denoting the Zhenghe 政和 (r. 1111–1117) and Xuanhe 宣和 (r. 1119–1125) reign pe ri ods on the seams around Wang Xizhi’s cal lig ra phy and on the seam be tween the dark pa per to its left and the bright pa per far ther left, secur ing the scroll in its new ar range ment un der im pe rial au thor i ty. No fur ther Northern Song seal is seen on the

outer seam of the dark pa per on the right. The “Xuanhe pro gram” of ti tle slip and seals was not a per sonal way of bring ing the im pe rial pres ence to the scroll. It was an in sti tu tional ar range ment; it se cured Wang’s cal lig ra phy in its scroll, but al so, most im por tant, can on ized the scroll’s place in the im pe rial col lec tion of works of the highest or der, as noted ex plic itly by Huizong’s son and suc ces sor Gaozong 高宗 (r. 1127–1130).31

This re con struc tion of the Huizong scroll has some prob lems. First, the pa per bear ing Wang Xizhi’s text is cropped ex tremely close to the char ac ters, es pe cially on the right. Closer in spec tion of Huizong’s round dou bledragon seal and of his square seals on the seams re veals some remounting and pa per cut ting af ter the dragon seal had been placed; the cut runs di rectly through it (Fig. 6). Furthermore, the dark pa per to the left of Wang Xizhi’s writ ing is sig nifi cantly wider (10.3 cm) than the pa per to the right (8.5 cm), suggesting that the lat ter has been cropped on its right side. This cut may have elim i nated an other sheet of bright Song pa per sim i lar to the one on the left, the seam of which then might have car ried an other Huizong seal, the gourdshaped “yushu” 御書 (im pe ri ally writ ten) seal known from other scrolls.32

Fig. 5. Song Huizong’s la bel and seals. Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie) (de tail).


The scroll was also cropped ver ti cal ly. The three rect an gu lar reign pe riod seals in the up per and lower cor ners of the Tang pa per sheet are cut off and no longer ex tend onto the new back ing of the scroll. Since two of Qianlong’s seals cross over the up per bor der of the Tang pa per, the cuts must have been made no later than dur ing his reign, pos si bly dur ing remounting, per haps af ter the scroll had been dam aged. In the rub bing of Wu Ting’s Yuqing zhai fatie (1614), the seals seem still com plete.

Huizong did not leave a col o phon any where near Wang’s text; the dark pa per on both sides of Wang’s char ac ters was left en tirely emp ty. Moreover, ex cept for the large pal ace seal far re moved to the left, Huizong’s seals did not dom i nate the scroll, even though the round dou bledragon seal on the right slightly touches the charac ter rang 穰. Wang Xizhi’s cal lig ra phy had a pris tine pres ence all its own, ca pa ciously arranged. According to Ebrey, this re flects a change in Huizong’s style some time af ter 1107, when he de cided to min i mize his in tru sion into the space of the art work proper save for inserting a ti tle slip by his own hand (Fig. 7).33

The Qianlong Emperor’s Scroll

Consider, by con trast, Qianlong’s scroll: in ad di tion to the la bel (prob a bly placed on the out side of the scroll), three col o phons and nine teen seals ap pear closely around Wang Xizhi’s text. As is attested in nu mer ous other exam ples of Chi nese paint ing and cal lig ra phy, Qianlong was rarely shy about such im po si tions of im pe rial graffi ti.34 In the case of Xingrangtie, the em per or’s crowded col o phons and seals reframe the Wang Xizhi text as an ar ti fact that dis ap pears into the Man chu em per or’s selfrep re sen ta tion. With Qianlong, the scroll turned into an en tirely dif fer ent ob ject, mak ing it al most im pos si ble to imag ine the se rene spa cious ness it had once pos sessed.

Qianlong’s scroll had been owned and remounted by the Chi nese salt mer chant and art col lec tor of Ko rean de scent, An Qi 安岐 (1683–ca. 1744). It was still in his pos ses sion in 1744,35 four years be fore Qianlong’s col ophons of 1748. An Qi’s eleven seals range from the righthand seam of Huizong’s dark pa per to ward Section 8 of the scroll, where the pa per with Sun Chengze’s col o phon is at tached. When An Qi placed his seal onto the up per right cor ner of Huizong’s dark pa per (Fig. 8), that pa per was al ready cut to its pres ent size, and the bright Song pa per that pre sum ably had existed to its right was already re moved. Nearly the en tire scroll as we know it—per haps ex cept for Sections 1, 2, and 9—was avail able to Qianlong in 1748, when he “was just be gin ning to learn the art of con nois seur ship.”36 The em per or, howev er, lim ited his seals and col o phons to the small area that had been part of Huizong’s scroll and, in ad di tion—with his larg est seal—to the silk back ing to the im me diate right of Huizong’s dark pa per. The heavi est traces of Qianlong’s ap pro pri a tion are the mas sive seal on the up per right, three col o phons—two of which on the hith erto empty dark pa per flanking Wang Xizhi’s charac ters—and a flurry of eigh teen seals in and around Wang’s cal lig ra phy, many of them across the four ver tical seams of the Tang pa per and Huizong’s dark pa per sheets flanking it. One large rect an gu lar seal is even impressed squarely on Huizong’s faint la bel, nearly oblit erat ing it (Fig. 9).37

What are Qianlong’s col o phons about? The sin gle line to the right of Wang Xizhi’s text be gins with two large char ac ters and then, as if in an abrupt re al i za tion of the lim ited space, con tin ues with six smaller ones. Its text, 龍跳天門，虎卧鳳閣 (a dragon leap ing at Heaven’s Gate, a ti ger crouching be neath Phoenix Pavilion), quotes the Liang em peror Wu’s 梁武帝 (r. 502–549) ear lier praise of Wang Xizhi’s cal lig ra phy (Fig. 10).38 Although in Qianlong’s cal lig ra phy this line ap pears al most ca sual in the un even size of its char ac ters and

Fig. 6. Song Huizong’s dou bledragon seal. Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie) (de tail).


slight left ward slant, it is any thing but: in nearly iden tical form, dis tin guish able only through very close compar i son of in di vid ual strokes, the same line ap pears on at least four other works at trib uted to Wang Xizhi.39 The dif fer ences are so slight that Qianlong’s line ap pears to be an ex tremely care fully ex e cuted copy—as does Wang Xizhi’s orig i nal text—or is the orig i nal from which the other in stances are cop ied (Fig. 11).

In other words, the em per or’s first col o phon, cit ing the words of an other em peror more than a mil len nium

ear li er, was part of Qianlong’s for mal pro gram (in this sense sim i lar to Huizong’s “Xuanhe pro gram”) by which he iden ti fied, dis tin guished, and can on ized Wang Xizhi’s fin est works avail able to him. One may won der whether the ac tual line on the Xingrangtie scroll was in deed ex ecuted by the em peror him self or rather cop ied by an anon y mous court cal lig ra pher.

Qianlong’s col o phon, dated 1748 and placed directly to the left of Wang’s two lines, re fers to Dong Qichang as hav ing claimed that Xingrangtie is by Wang

Fig. 7. Song Huizong’s scroll reconstructed. Based on Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie) (de tail).

Fig. 8. An Qi’s seals. Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie) (de tail).


Xizhi’s own hand, “not what peo ple since the Tang could ac com plish” (非唐以後人所能到); and then, to af firm this con clu sion once again, Qianlong de clares that “this is cer tainly not what a trac ing copy can do” (要非鉤摹能辦). This ech oes nearly ver ba tim the state ment by Wang Youdun 汪由敦 (1692–1758), one of the com pilers of Qianlong’s Three Rarities Hall cat a logue, that “its bril liance flies for ward, and its an cient air is pro found and sol emn; this is not what a trac ing copy is ca pa ble of do ing” (精采飛動而古色淵穆，非鉤橅可辦).40 Dong’s 1609 col o phon, writ ten in im i ta tion of Wang Xizhi’s own dra matic style, notes (Fig. 12),

Wherever this scroll is, there should be an aus picious cloud cov er ing it. It is only that the hu man eye can not see it! Inscribed again on the 26th day of the sixth month in the year jiyou, when view ing [the scroll] to gether with Chen Jiru and Wu Ting. Written by Dong Qichang.此卷在處，當有吉祥雲覆之，但肉眼不見耳。己酉六月廿六日再題，同觀者陳繼儒、吳廷。董其昌書。

In his un dated col o phon mounted left of his transcrip tion of Wang Xizhi’s text, Dong also al ludes to a poem by Su Shi 蘇軾 (1037–1101) that praises a sim ilarly brief text by Wang as be ing worth “thirty thousand” other scrolls. Qianlong’s third col o phon (1748) is de voted to correcting Dong’s re mark (which he mis under stands), try ing to pres ent him self as the su pe rior connois seur and schol ar.41

Qianlong agrees with Dong in one cru cial point: in Dong’s col o phon dated 1604, as well as in a sep a rate col o phon (dated 1613) on a paint ing by Li Tang 李唐 (ca. 1050–ca. 1130), he de clares Xingrangtie to be Wang Xizhi’s “gen u ine traces” (zhenji 真蹟), by which he means an orig i nal from his hand.42 This is what Qianlong asserted both in his abovequoted col o phon and, more over, in his la bel 王羲之行穰帖真蹟 (Genuine traces of Wang Xizhi’s Xingrangtie) at the very be gin ning of the scroll. In vouching for the scroll’s au then tic i ty, Dong Qichang, Wang Youdong, and Qianlong all chose to ignore the judg ment of an other fa mous late Ming cal lig rapher, col lec tor, and schol ar, Zhan Jingfeng 詹景鳳, who in 1591 had pro nounced the scroll a Tang trac ing copy (a ver dict em phat i cally ech oed by Zhang Chou 張丑, writ ing in 1616). Even An Qi took it as a Tang copy.43 By con trast, the at tempts to af firm Wang Xizhi’s own hand writ ing re veal the anx i ety over au then tic ity that for cen tu ries had haunted Wang Xizhi’s works.

Letters and Per sona

To the Wang Xizhi tra di tion, per sonal let ters such as Xingrangtie are cen tral. Xingrangtie has no his tor i cal an chor be yond its at tri bu tion to Wang Xizhi. Aside from Lanting ji xu—its orig i nal purportedly bur ied with Tang Taizong—such let ters are the core of Wang’s oeu vre, and they have been cel e brated as such since at least the early Tang. Yet Xingrangtie has been dif fi cult to de ci pher not only for “out side” au di ences of later

Fig. 9. Qianlong’s seals and col o phons. Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie) (de tail).


(in clud ing mod ern) read ers; from the be gin ning it was com posed in a kind of Barthian “id i o lect”44 that presup poses much on the side of its ad dress ee. The text’s in for ma tional value is ex tremely low, re sem bling texts used in rit ual ex changes where, in the words of the linguist Wade Wheelock, “prac ti cally ev ery ut ter ance . . . is su per flu ous from the per spec tive of or di nary con

Fig. 11. Qianlong’s first col o phon in four other scrolls. From left to right: Timely sunny af ter snow (Kuaixue shiqingtie), af ter Jin Wang Xizhi moji 晉王羲之墨跡 (Ink traces of Wang Xizhi from the Jin dy nas ty) (Taipei: Guoli gugong bowuyuan, 2010), 5, 17; Seventh month and In the cap i tal, two calligraphies (Qiyue duxia ertie), af ter Jin Wang Xizhi moji, 39–40; Sightsee-ing (Youmutie), af ter Wang Xizhi shufa ji 王羲之書法集 (Wang Xizhi’s col lected cal lig ra phy) (Beijing: Beijing gongyi meishu chubanshe, 2005), 204–5; Zhong Yao’s thou sand-char ac ter es say (Zhongyao qianziwen), af ter Wang Xizhi shufa ji, 432–33.

Fig. 10. Qianlong’s first col o phon. Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie) (de tail).

ver sa tional prin ci ples.”45 It is, in other words, “commu ni ca tion with out in for ma tion.”46 Moreover, while Wang Xizhi’s let ter cal lig ra phy is cel e brated for its unre strained spon ta ne ity and its ex pres sion of the author’s in di vid u al i ty, the let ters are re plete with the clichés and for mal re quire ments pre scribed in widely cir cu lat ing epis to lary man u als.47 Such man u als be gan to emerge in Wang Xizhi’s time and mul ti plied thereaf ter; Wang him self re port edly authored one.48 Antje Richter has aptly sum ma rized the in ter pre ta tive problems of Wang’s let ters in gen er al, and her con clu sions ap ply fully to Xingrangtie:

The most ob vi ous prob lem con cerns the com mu nica tive ef fi cacy of these let ters that seem to be de fec tive in so many ways. Not only do we find empty and in com plete let ter frames, but the frames them selves con sist mainly of epis to lary con ven tions and ste reo types. The same can be said of many ac tual let ter bod ies, whose mes sages rarely go be yond the triv ial chitchat that usu al ly—and for the most part for tu nate ly—sinks into his tor i cal obliv i on. . . . Still, as many of these let ters suc ceed


in mov ing read ers even to day, they may have played an im por tant role in maintaining friend ships and fam ily con nec tions af ter all .49

Note that Richter speaks of the lack of in for ma tive func tion in let ters that are calligraphically fully in tel li gible; with Xingrangtie, even this min i mal con di tion for un der stand ing can not be taken for granted. Either way, what ever such a let ter con veys is less in di vid ual than commu nal and for mu la ic—in fact, its very pur pose is “to re fer to itself and to its own com mu ni ca tive func tion in de pendently of any prop o si tional con tent it may ex press.”50 Wang Xizhi’s unique ness is not found in his skill ful manip u la tion of existing for mu laic phrases but in stead in the un in hib ited cal lig ra phy of his let ters.51 Aside from a small num ber of more sub stan tial let ters of which we do not have cal li graphic ver sions, the Wang Xizhi per sona can be grasped solely in the flow of his brush.

Eugene Wang has discussed a trac ing copy of a Wang Xizhi let ter known as Sangluantie 喪亂帖 (Letter on be reave ment and dis or der) that has qui etly sur vived in Japan, iso lated from the remaining Wang Xizhi oeu vre, for over a mil len ni um. This text of more than a hun dred char ac ters “is an un re strained out pour ing of an guish and pa thos, and these feel ings ap pear to be ech oed in the style of the cal lig ra phy.”52 “Anguish and pa thos” stand against the tra di tion’s ba sic as sump tions about Wang’s cal lig ra phy and the art ist’s se rene and tran quil per sona reconstructed from it. In Eugene Wang’s read ing, inspired by an ear lier study by Han Yutao 韓玉濤, the letter sug gests that “the real Wang Xizhi” re mains elu sive

and hid den be hind “the Wang Xizhi tra di tion” that was first cod i fied in the early Tang.53

A prob lem re mains, how ev er: just as Lanting ji xu does not give us the real per son, nei ther does Sangluan-tie, which, like so many of Wang Xizhi’s let ters, is composed in a highly for mu laic id i om.54 In their seem ingly di rect out flow of emo tion—se rene or an guished—both are equally me di ated and rhe tor i cal, as Qianshen Bai has dem on strated in a com pel ling anal y sis that ques tions the en tire con cep tu al i za tion of Chi nese cal lig ra phy as be ing ex pres sive of the writ er’s in ner self.55 Since the Tang, this ex pres sive no tion of cal lig ra phy—ob vi ously mod elled on ear lier dis courses on mu sic and po et ry—has of ten been cited with con vic tion, but it can not be projected back onto Wang Xizhi, nor can it ex plain the widely prac ticed per for mance of cal lig ra phy, con tin ued through late im pe rial China, in so cial ex changes. Sangluantie strikes us for its seem ing match of form and con tent, just as Lanting ji xu does; but this ap par ent fu sion of text and script re mains open to in ter pre ta tion and, more over, is no where the rule in the so cial prac tice of let ter cal lig ra phy.

The no tion of cal lig ra phy as an un me di ated ex pression of its au thor’s emo tion was for mu lated no later than the early Tang, when cal lig ra pher and critic Sun Guoting 孫過庭 (ca. 648–ca. 701) ap plied it to Wang Xizhi’s com po si tions.56 A hun dred years af ter Sun, such expres siv ity was in par tic u lar as so ci ated with the seemingly untrammelled “run ning” (xingshu 行書) and “cur sive” (caoshu 草書) styles that dom i nated the body of Wang’s writ ings and were can on ized in the im pe rial col lec tions of the time.57 Moreover, Ouyang Xiu 歐陽修

Fig. 12. Dong Qichang’s 1609 col o phon. Wang Xizhi, Ritual to Pray for Good Harvest (Xingrangtie) (de tail).


(1007–1072), the most pro lific Northern Song col lec tor and con nois seur of an cient cal lig ra phy, at trib uted the ex pres sion of un re strained emo tion spe cifi cally to the calli graphic genre of short let ters.58 Exalting the in for mal and per sonal qual i ties of Wang Xianzhi’s writ ing, Ouyang re lated his own lit er ary ide als to those of the an cient cal lig ra phers. Ouyang’s selfcom piled lit er ary col lec tion con tains writ ings in sev eral genres of in for mal prose, includ ing, most nu mer ous ly, fif tyfour let ters. According to Ronald Egan, “No ear lier writer had used prose for subjec tive ex pres sion so of ten. No ear lier writer con sis tently allowed so much per sonal sen ti ment into prose.”59

Although Ouyang Xiu, writ ing just a gen er a tion before Emperor Huizong and being the lead ing “an cientstyle lit er a ture” (guwen 古文) pro po nent of his time, and Wang Xizhi, the prac ti tioner of Daoist rit u als and copyist of Daoist scrip tures, were sep a rated not only by centu ries but also phil o soph i cal ly, Ouyang’s in sis tence on writ ing as a means of per sonal ex pres sion connected him to the Jin cal li graphic mas ters. On the as sump tion that ex cel lence in cal lig ra phy should be taken to re flect moral su pe ri or i ty, Ouyang par tic u larly praised the callig ra phy of up right Tang of fi cials (fore most among them Yan Zhenqing 顏真卿 [709–785]).60 Yet by the same token, Northern Song po lit i cal and cul tural guwen phi loso phy was still com pat i ble with Jin dy nasty phil o soph i cal and aes thetic thought through their com mon em pha sis on per sonal au then tic ity and “nat u ral” style; in gen er al, “guwen schol ars em pha sized qual i ties they discerned in the cal lig ra phy that could be traced back to the per sonal ity of the cal lig ra pher.”61 This no tion of cal lig ra phy as mir ror, con cep tu al ized as an aes thetic and eth i cal ideal since the late Six Dynasties and enshrined in the im perial cal lig ra phy col lec tions ever since, had gained strong con fir ma tion in the de cades be fore Huizong’s reign.

Given the welldocumented em pha sis on ge ne al ogy in Bud dhism, Dao ism, and al so—es pe cially since Tang times—guwenCon fu cian ism,62 it is not sur pris ing that the Chi nese cal li graphic tra di tion is based on a ge ne alogy of in di vid ual mas ters63 with Wang Xizhi as, par adox i cal ly, their foun tain head and also early ze nith, and that the Tang court “appointed as court cal lig ra phers men who saw them selves in a di rect line of trans mis sion from the Wangs.”64 Yet it is not a ge ne al ogy of ac tual authors: the ar tis tic per sona of Wang Xizhi is the prod uct of his works, not their or i gin. Here the no tion of the “per sonal let ter” turns par a dox i cal—and not just because one won ders how all these “pri vate” let ters were col lected and pre served. As noted by Foucault, “a pri vate let ter may well have a signer—it does not have an author.”65 If a pri vate per son wrote the char ac ters of Xing-rangtie to a friend or group of friends, he did not write

as the au thor of the Wang Xizhi oeu vre. But if he was al ready selfaware of be ing Wang Xizhi the cel e brated art ist who expected his let ter to be cherished, pre served, and trans mit ted for its ar tis tic qual i ties, writ ings such as Xingrangtie wouldn’t be per son al, spon ta ne ous let ters at all —they would be con sciously cre ated rep re sen ta tions of the art ist and, as such, intended from their very in ception for pub lic and not pri vate con sump tion.66 There is strong con tem po ra ne ous ev i dence for such selfawareness: Wang Xianzhi once sent a let ter to the em peror and in it requested that it be pre served for its su perb callig ra phy.67 From such in stances, Qianshen Bai ob serves that a “keen aware ness that let ters were col lect ible led to a con scious ef fort to make their lit er ary style and the cal lig ra phy in which they were writ ten the ob jects of aesthetic ap pre ci a tion.”68 Bai thus sug gests that “while their texts were intended for pri vate read ers, their cal ligra phy was aimed at a pub lic au di ence”; they were “private let ters for pub lic con sump tion.”69

But is there such a thing? However seem ingly “private” the top ics raised in these let ters may be, their “pub li ca tion,” even if lim ited to a small aris to cratic circle, cat e gor i cally denies the no tion of the pri vate. The seem ingly pri vate char ac ter of these let ters is pre cisely part of their pub lic ap peal, driven by the norms and expec ta tions shared be tween the writer and his pre sumed au di ence: the “pri vate per son” is a pub licly constructed and displayed per so na—a mask and a rep re sen ta tion. When, by Tang times, let ters were “the over whelm ing ma jor i ty” of Wang Xizhi’s ex tant works,70 their au di ence of em per ors and im pe ri ally appointed schol ars rec ognized in their “pri vate” char ac ter the exalted eth i cal dispo si tion of the man re tired from of fice. On its sur face, the her metic tone of a text like Xingrangtie thus ap pears to con form to the aes thetic and po lit i cal ideal of the work of art as—in Adorno’s for mu la tion—“the noniden ti cal” (das Nichtidentische),71 es cap ing an enforced iden tity and ho mo ge ne ity with the em pir i cal re al ity of the im pe rial cos mos. Yet here, the noniden ti cal is ul timately a mi rage: as the im pe rial cal lig ra phy col lec tions prized these works of “noniden ti ty” above all oth ers, they at once reconstructed them as seam lessly iden ti cal with the norms of the im pe rial state. The char ac ters of Xingrangtie are part of the im pe rial canon not de spite but be cause of their sub lime re sis tance to the nor ma tive in tel li gi bil ity of the im pe ri ally stan dard ized script.

Performance, Style, and Copy

Meanwhile, the early his tory of the Wang Xizhi cor pus—the cen ter piece of the im pe rial col lec tions of succeeding dy nas ties—was one of as sem bling, dis pers ing, and reas


sembling dur ing the tu mul tu ous fourth through sev enth cen tu ries. By the early sixth cen tu ry, forg ery and questions of au then tic ity had be come ma jor con cerns, as early me di e val col lec tions suf fered re peated “ca lam i ties, yet new col lec tions con stantly mushroomed, some of them with a fan tas tic num ber of au to graphs.”72 This is the prob lem to which the many seals and col o phons on a scroll like Xingrangtie are the an swer: they au then ticate, they doc u ment, and they anx iously safe guard the in teg rity of the scroll. And per haps most im por tant, they turn the scroll—copy or not—into an un ques tion able and unique orig i nal in its own right: there only is one Xingrangtie bear ing the seals of Song Huizong, Qianlong, and all the oth ers around them.

While the col lec tions of Wang’s oeu vre kept changing, the idea of cal lig ra phy as the un me di ated ex pres sion of per son al ity remained sta ble. As Lothar Ledderose has not ed,

The art of cal lig ra phy is unique among the arts in the world in that the pro cess of cre a tion in all its con sec u tive phases is vis i ble in the ob ject. A proper viewer fol lows with his eyes the brush move ments through each of the char ac ters and the se quence of the lines. He thus recre ates for him self the moments of the ac tual cre a tion. The viewer senses the tech ni cal dex ter ity and sub tle ties in the move ment of the writ er’s hand, and he may feel as if he looked over the shoul der of the writer him self and observed him while he wrote. The viewer thus establishes an im me di ate and per sonal rap port with the writer of the piece. In a qua sigraph o log i cal approach he asks what the for mal qual i ties in the cal lig ra phy re veal about the writ er’s per son al i ty.73

The au then tic ity of a piece of cal lig ra phy thereby certifies the au then tic per son al ity of its writ er, and vice ver sa. While a po etic per sona like Tao Qian could be constructed out of his words, a cal li graphic per sona like Wang Xizhi is constructed through the move ment of his brush, seen as an ex ten sion of his mind.74 Wang Xizhi’s Jinshu bi og ra phy allowed later con nois seurs of Wang’s cal lig ra phy to as sign in di vid ual works to dif fer ent phases of his life, mak ing life and work il lu mi nate and ex plain each oth er.

In this func tion of cal lig ra phy as the au then tic re flection of its writ er’s mind, per haps the most im por tant as pect of the brush work was that once it was ex e cut ed, it could not be changed. Poetry—as at the fa mous Orchid Pavilion meet ing Wang pre sided over—was a perfor mance art, but any ex tem po rized poem remained open to fur ther im prove ment af ter its ini tial com po si tion and

rec i ta tion. Calligraphy, by con trast, did not: ev ery flaw in a piece of writ ing was there to stay, a wit ness to the very act of per for mance never to be erased; ev ery correc tion—for ex am ple, in Lanting ji xu—remained forever vis i ble. While fun da men tally nonprop o si tion al, cal lig ra phy was still rep re sen ta tion al: not of its con tents but of its sin gu lar mo ment of bodily en act ment. It is pre cisely for this qual i ty, and in this ex is ten tial sense, that Xingrangtie embodies Wang Xizhi’s per sona not merely au then ti cally but also truth fully re gard less of the mean ing—or even the decipherability—of its words.

But cal lig ra phy has many forms, from seal script to the reg u lar cler i cal script and, fur ther, to run ning and cur sive script. The first two, of course, de mand the writer’s sur ren der to a set of fixed, de tailed rules, a quest for per fec tion where true mas tery, to gether with the per sona of the true mas ter, is found in the era sure of in di vid ual ab er ra tion. The mas ter of seal and cler i cal script only appears from his own in vis i bil ity in the perfected char acters. This in vis i ble mas ter embodies at once the weakest and the stron gest sense of agen cy: the weakest in his surren der to rules, and the stron gest in the ab so lute con trol of their slow and me tic u lous ex e cu tion. Calligraphy in cur sive script, on the other hand, seems to re verse this par a dox: here, the writ er’s au tho rial agency is stron gest in his ex pres sion of in di vid u al, even id i o syn crat ic, choices, but—be cause of the fast mo tion of the brush—it is also the weakest in his in abil ity to fully con trol or pre dict the re sult. Every flaw will re main as tes ti mony to the writ er’s unique act of per for mance.

Wang Xizhi was con sid ered a mas ter of both the regu lar and the run ning and cur sive styles, and pieces from all three styles at trib uted to him were in cluded in the impe rial cat a logues of model calligraphies, avail able to be stud ied and cop ied.75 But the writ ings in reg u lar script formed a tiny frac tion of his ca non i cal oeu vre; in stead, he (like his son) was cel e brated pri mar ily for his run ning and es pe cially his cur sive style, that is, the forms of writing whereby the writer gained in the ap pear ance of sponta ne ous in spi ra tion what he gave up in con trol. The ideal of care fully or ches trated “nat u ral ness” and “sponta ne i ty,” merely claimed in Tao Qian or Su Shi’s po et ry,76 gained vis i ble proof in the traces of the mov ing brush. In the tra di tional Wang Xizhi nar ra tive, the ex treme man ifes ta tion of these qual i ties was found in Wang’s fa mous at tempt to recre ate his own mas ter ful writ ing of Lan-ting ji xu, the “most cel e brated piece of cal lig ra phy of all time,”77 the very next day: de spite try ing hun dreds of times, he failed to re peat his own orig i nal feat.78 Supremely iron ic, this unique re sult of a sin gle cal li graphic per for mance—ex per i men tally proven to be unachievable by sheer will and ef fort—was then cop ied, from


cop ies, through the ages. To this day, Lanting ji xu displays a “nat u ral” self, unmoored from of fice and conven tion, ra di at ing its Wang Xizhi per sona into works like Xingrangtie, and hence allowing view ers to rec ognize the writ er’s au then tic ity in the grace ful lines of a text with out par tic u lar mean ing or pur pose. But this text is a copy.

Before the Tang, the selfex pres sive modes of running and cur sive script were rarely suited for of fi cial use; in Wang’s own time, according to Wen Fong, it was a ver i ta ble “re jec tion of the statespon sored mon u mental style.”79 Even though cur sive script may have orig inated as short hand writ ing in court ad min is tra tion,80 to write in such fluid script was gen er al ly, in Wang’s time, to write in pri vate and for whim si cal pur poses, and for a small cir cle of fam ily and likeminded friends. That both the cal lig ra phy and the con tent of Xingrangtie are barely de ci pher able only con firms its ca su al, ef fort less au then tic i ty: the let ter was not writ ten for us (or for the im pe rial au di ence) in the first place; as out sid ers, we are, by defi ni tion, not sup posed to un der stand its id i om. Noniden ti cal with our pur poses, its no bil ity may be admired from an un bridge able dis tance, but it can not be appropriated.

The Sage and the State

This ap par ently un bridge able dis tance re veals Xingrang-tie’s ul ti mate par a dox. By Tang Taizong’s time, the emperor him self was not merely among the ar dent copy ists of Wang’s works; his im pe rial edicts were of ten writ ten in run ning or cur sive script as well, and were can on ized for that in Song Huizong’s Xuanhe Calligraphy Cata-logue.81 But no where is the im pe rial ap pro pri a tion of Wang Xizhi’s noniden ti cal art more di rectly performed than in the hail storm of col o phons and seals surrounding the fif teen char ac ters of Xingrangtie. If the let ter ever oc cu pied a space of pri vate qui etude out side the world of of fi cial dom, it was not left alone there. The “sage of cal lig ra phy” was a sage only be cause he was rec og nized by other sages: the mon archs whose own sagehood—like that of the an cient “plain king” (suwang 素王) Confucius—was pre mised on their su preme ca pac ity for rec og ni tion, per cep tion, and dis crim i na tion. His exalted vir tue, revealed in the traces of his brush, depended on the im pe rial state to be known and per pet u at ed. From the col lec tions of the (Liu)Song em per ors Xiaowu 孝武 (r. 452–464) and Ming 明 (r. 465–472) to those of the Liang em peror Wu, and then, fur ther, all the way from the two Sui em per ors to Tang Taizong, Song Huizong, Qianlong, and, fi nal ly, Jiaqing, Wang’s let ters sur vived in the im pe rial em brace.

But the im pe rial state was more than the se quence and sum of its em per ors. Among the Song dy nasty callig ra phers in cluded in Huizong’s Xuanhe Calligraphy Catalogue, “men who served in high court posts or had other con nec tions to the Song court are es pe cially well represented.”82 Moreover, from the sev enth through the twen ti eth cen tu ries, Xingrangtie moved back and forth be tween im pe rial and pri vate col lec tions, be ing touched by mon archs, of fi cials, and li te rati alike. Scholars and em per ors rec og nized in it the same set of shared cul tur al, eth i cal, and po lit i cal ide als. The choices made by connois seurs such as Dong Qichang, An Qi, and oth ers who left their seals and col o phons on Xingrangtie emerged from the larger tra di tion of wen 文 (cul ture as writ ing) that was con tin u ally val i dated at the im pe rial court and by those who as pired to serve it. The Wang Xizhi persona known to us was not tainted but con sti tuted by this em brace; un der Tang Taizong, his court of fi cial Chu Suiliang 褚遂良 (596–658) firmly established the Wang Xizhi tra di tion of thou sands of cop ies and no orig i nals.83 Thus, when Taizong took the pre sumed orig i nal of Lan-ting ji xu with him into his gra ve, plac ing it into the timehon ored re li gious space of the tomb and its oc cu pant’s af ter life, he exalted the work but did not erase it.84 It con tin ued to ex ist in the “gen u ine traces” of cop ies af ter cop ies, a tra di tion of cul tural per for mance and par tic ipa tion that was the tra di tion of wen, car ried along with the own er ship and re pro duc tion of its ar ti facts and authen ti cated with seals and col o phons.

Consumed and re peat edly recre ated by the im pe rial state, Wang Xizhi’s oeu vre—and Xingrangtie prom inently within it—is there fore not an ex pres sion of the mas ter cal lig ra pher’s mind but a rep re sen ta tion of the cul tural his tory of im pe rial China. As its ev erchang ing con fig u ra tions moved in to, out of, and back into the impe rial col lec tions of succeeding em per ors and dy nas ties, with the im pe rial copy at once displacing and per pet uat ing its source, it grew into the su preme icon of the empire’s con tin u ous pos ses sion, loss, and re con sti tu tion of cul ture as writ ing.85 If Wang Xizhi’s orig i nal writ ing was in for mal and with out pur pose, its pres ence in thou sands of cop ies was not: cre ated with ut most care on spe cially pre pared pa per, its very ex is tence was im pe ri ally commis sioned. Likewise, the la bels, col o phons, and eightyseven seals on Xingrangtie are not ad di tions to the real thing—they are the real thing: the em i nently in tel li gi ble text that could be con tin ued, in prin ci ple for ev er, as the dia chronic mon u ment to the im pe rial Wang Xizhi perso na, where with each ad di tion to the ev erlength en ing scroll, the “gen u ine traces” be came an ever smaller part of the whole. There is, af ter all , no Wang Xizhi other than the one whose am big u ous, lateinlife dis tance in


re la tion to the court be came di a lec ti cally sub lated and absorbed in court cul ture, whose swift, spon ta ne ous, and un con trol la ble moves of the brush were re writ ten with me tic u lous pre ci sion and pains tak ing pa tience, whose bi og ra phy emerged from the Tang im pe rial court, and whose in di vid u al ity was deciphered from cop ies of cop ies. Xingrangtie embodies a tra di tion inscribed and reinscribed over and against the to tal loss of or i gin.

Beyond In no cence

Looking again at Xingrangtie and its few sib lings of Tang trac ing cop ies scattered across East Asia, there is no return to some in no cent ad mi ra tion of the mas ter’s hand and spir it. And yet, there is also no end to the marvel and won der one may feel when con tem plat ing the fif teen char ac ters in front of us. In Robert Harrist’s de scrip tion,

The copy recre ates the buoy ant, en er getic flow of Wang’s char ac ters, which seem fully three dimensional and are enlivened by con stant changes of thick ness in the brushstrokes that re sem ble twist ing wires. The let ter also re flects Wang’s in ven tive ness in writ ing re cur ring con fig u ra tions of strokes. For ex am ple, the dots in the two char ac ters of the first col umn and the first, third, and fourth char ac ters of the sec ond col umn dem on strate the wide range of vi sual ef fects that can be achieved in even the sim plest of cal li graphic forms.86

But to rec og nize such beau ty, it does not suf fice to nar row one’s eyes and try to look past the seals and colo phons. Instead, we must re store Xingrangtie to something that per haps it never truly was: the noniden ti cal work of art be yond its func tions for the cul tural and polit i cal tra di tion, a clas sic that, in the words of Italo Calvino, “has never fin ished say ing what it has to say.”87 For this, re mov ing the words “copy” and “Wang” may well be a good start.

That said, it re mains im por tant to ac knowl edge the epis te mo log i cal lim i ta tions of the pres ent study, or of any ap proach to Xingrangtie. The var i ables in the re construc tion of the orig i nal work are im pos si ble to con trol. There is no ev i dence for reconstructing a cred i ble and de pend able re cord for the orig i nal com po si tion of the text to gether with Wang Xizhi’s psy cho log i cal dis po sition or in tent; for the cir cum stances of its copy ing presum ably at some point dur ing the Tang (or lat er, on Tang pa per?); for the gaps of cen tu ries be tween Wang Xizhi’s time and the time of the copy ing, be tween the mak ing of the copy and the time of Song Huizong’s seals, and

be tween Huizong’s time and the late Ming; for the question of whether or not Xingrangtie should be connected to the “sec ond half” first noted in the ninth cen tu ry; for de cid ing with con fi dence how to tran scribe the four disputed char ac ters within Xingrangtie, not to men tion the two un de ci pher able char ac ters in the “sec ond half”; for the pos si bly de bil i tat ing dam age to the last char ac ter of Xingrangtie; for the pe riod of time when the scroll lingered in Japan; for the cuts and remountings of the text; and, al to geth er, for the au then tic ity of the scroll that en tered Huizong’s court. More likely than not, the aboveof fered trans la tion—just like the existing Jap anese trans la tions—is in ad e quate or just wrong, but we won’t know how wrong, or in what ways wrong; and if it is ac tu ally right, we won’t know that ei ther. To take all these var i ables into ac count sends the mind spin ning at ev erin creas ing ve loc i ty. Strictly speak ing, Xingrangtie can not be read. It can only be looked at.

To write about Xingrangtie is thus an ex er cise in schol arly hu mil i ty; this, to gether with the rec og ni tion of beau ty, may be the real con clu sion. At a time when ev ery ac a demic es say sets out to “ar gue” this or that, how ever ba nal the mat ter, Xingrangtie teaches its au dience that, fun da men tal ly, hu man is tic in quiry is not as much about prov ing and scor ing points, or about re ducing com plex ity to the size of one’s own lim i ta tions and pre con cep tions, as it is about looking at the thing at hand with pa tience and ask ing ques tions wor thy of its end less com plex i ty. Viewed from this per spec tive, Xing-rangtie is sim ply an as ton ish ing gift. If Calvino is right (and I think he is) that “the only rea son one can pos si bly ad duce [for read ing the clas sics] is that to read the classics is bet ter than not to read the clas sics,” then to look at Xingrangtie is sim ply bet ter than not to look at Xin-grangtie.

Martin Kern is the Greg (’84) and Joanna (P13) Zeluck Pro-fessor in Asian Studies and Chair of the Department of East Asian Studies at Princeton University and, in ad di tion, Distin-guished Professor at the Research Center for Comparative Literature and World Literature, Shanghai Normal Univer-sity. He has published widely on early Chi nese lit er a ture, his-to ry, and tex tual cul ture, in clud ing The Stele Inscriptions of Ch’in Shihhuang: Text and Ritual in Early Chi nese Representation (Amer i can Oriental Society, 2000), var i ous edited vol umes, and some se venty ar ti cles and book chap ters, among them the open ing chap ter in the Cambridge History of Chi nese Literature (Cambridge University Press, 2010). He is cur rently com plet ing the mono graph Texts, Authors, and Performance in Early China: The Origins and Early Development of the Literary Tradition (Princeton University Press). Kern also serves as co-ed i tor of T’oung Pao.[[email protected]]


Acknowledgments

Preparation of this es say would have been im pos si ble with out the gen er ous as sis tance of Cary Y. Liu, Curator of Asian Art at the Princeton University Art Museum, who gave me ac cess to the Princeton scroll ev ery time I requested it and also to a stack of re search notes on it—most of them anon y mous—that have ac cu mu lated over time and are kept in the mu se um. I also thank es pe cially Robert E. Harrist Jr. for a de tailed cri tique and, in ad dition, Jerome Silbergeld, Paul W. Kroll, Mick Hunter, Stephen F. Teiser, Eugene Yuejin Wang, Antje Richter, and Qianshen Bai for many ad di tional com ments, cor rections, and in spi ra tions.

Notes

1. For his of fi cial bi og ra phy, see Jinshu 晉書 (History of the Jin) (Beijing: Zhonghua shuju, 1987), 80.2093–2102; for a biographical study, Ki Sh�shun [Qi Xiaochun] 祁小春, � Gishi ronkō 王羲之論考 (A study and dis cus sion of Wang Xizhi) (Osaka: T�h� shuppansha, 2001), esp. 152–249. See also Richard B. Mather, Shih-shuo Hsin-yü: A New Account of the Tales of the World, 2nd ed. (Ann Arbor: University of Michigan Center for Chi nese Studies, 2002), 530–31.

2. Eugene Y. Wang, “The Taming of the Shrew: Wang Hsichih (303–61) and Calligraphic Gentrification in the Seventh Century,” in Character & Context in Chi nese Cal-ligraphy, ed. Cary Y. Liu et al. (Princeton, NJ: Princeton University Art Museum, 1999), 145.

3. For a de tailed dis cus sion of Wang Xizhi’s com mitment to Dao ism, see Ki Sh�shun, Ō Gishi ronkō, 251–351.

4. See Alan J. Berkowitz, Patterns of Disengagement: The Practice and Portrayal of Reclusion in Early Medieval China (Stanford, CA: Stanford University Press, 2000).

5. Lothar Ledderose, Mi Fu and the Classical Tradition of Chi nese Calligraphy (Princeton, NJ: Princeton University Press, 1979), 40.

6. See, e.g., Mather, Shih-shuo Hsin-yü, 338: “Contemporaries char ac ter ized Wang Hsichih as fol lows: ‘Now drifting like a float ing cloud; now rearing up like a star tled drag on.’” This is an other par al lel to Tao Qian; see Stephen Owen, “The Self’s Perfect Mirror: Poetry as Autobiography,” in The Vitality of the Lyric Voice, ed. Shuenfu Lin and Stephen Owen (Princeton, NJ: Princeton University Press, 1986), 71–102. For the re cep tion and “mak ing” of Tao Qian, see Xiaofei Tian, Tao Yuanming & Manuscript Culture: The Record of a Dusty Table (Seattle: University of Washington Press, 2005); and es pe cially Wendy Swartz, Reading Tao Yuanming: Shifting Paradigms of Historical Reception (427–1900) (Cambridge, MA: Harvard University Asia Center, 2008). For Wang Xizhi’s re cep tion his to ry, Ledderose, Mi Fu, 43, notes that “many pieces in the im pe

rial col lec tions sim ply could not have been orig i nals. As a re sult of the se lec tions and ad di tions that were made in each col lec tion the im age of the clas si cal mas ters must constantly have changed.”

7. Including Daoist (Huangting jing 黃庭經 [Yellow court scrip ture]) and Bud dhist scrip tures as well as stele inscrip tions such as Yue Yi lun 樂毅論 (On Yue Yi) and Xiao-nü Cao E bei 孝女曹娥碑 (Stele for the fil ial daugh ter Cao E); see the ex am ples in Yang Lu 楊璐, Wang Xizhi shufa quanji 王羲之書法全集 (Complete col lec tion of Wang Xizhi’s cal lig ra phy), vol. 2 (Beijing: Zhongguo shudian, 1999); Wang Xizhi shufa ji 王羲之書法集 (Wang Xizhi’s col lected cal lig ra phy) (Beijing: Beijing gongyi meishu chubanshe, 2005); and Tōkyō National Museum et al., Shosei Ō Gishi: tokubetsuten 王羲之：特別展 (Wang Xizhi: Special Exhibition; En glish ti tle: Wang Xizhi: Master Calligrapher) (Tokyo: Mainichi Shinbunsha, 2013). On the of fi cial can on i za tion of Yue Yi lun, see Wang, “The Taming of the Shrew,” 149.

8. As noted by Qianshen Bai, “Chi nese Letters: Private Words Made Public,” in The Embodied Image: Chi nese Calligraphy from the John B. Elliott Collection, ed. Robert E. Harrist Jr. and Wen C. Fong (Princeton, NJ: Princeton University Art Museum, 1999), 383. Antje Richter, “Beyond Calligraphy: Reading Wang Xizhi’s Letters,” T’oung Pao 96, nos. 4–5 (2010): 372, notes that about seven hun dred per sonal notes by Wang are still known to day. While Richter discusses Wang’s let ters as lit er a ture (as op posed to callig ra phy), this dis tinc tion was prob a bly not made in Wang’s own time. The fact that the tra di tion has pre served so many of his per sonal let ters (most of them of rather triv ial content) is no doubt due to their orig i nal re cep tion as mas terpieces of cal lig ra phy. As Richter (p. 374, n. 8) notes, the sec ondlarg est cor pus of let ters from the pe ri od—of no more than about eighty pieces—is by Wang’s son, like wise a famed cal lig ra pher.

9. Ledderose, Mi Fu, 13, 26.10. Yang Yuanzheng 楊元錚, “Junjia lianghang shi’er

zi, qiya Ye hou san wan jian: Pulinsidun daxue yishu bowuguan cang ‘Xingrangtie’ liuchuan kaolue” 君家兩行十二字 , 氣壓鄴侯三萬籖：普林斯頓大學藝術博物館藏《行穰帖》流傳略考 (The twelve char ac ters in two lines in your col lec tion at home, Sir—/their at mo sphere counts for more than the 30,000 vol umes of the Duke of Ye: A brief ex ami na tion of the trans mis sion of Xingrangtie in the col lec tion of the Princeton University Art Museum; En glish ti tle: “The Tang Tracing Copy of Wang Xizhi’s Calligraphy, Ritual to Pray for Good Harvest: A Study of Its Provenance”), Gu-gong bowuguan yuankan 故宮博物館院刊 2008.1:8. Yang’s ar ti cle is the most com plete study of the prov e nance of Xing-rangtie. Other ex cel lent stud ies in clude Robert E. Harrist Jr., “A Letter from Wang Hsichih and the Culture of Chinese Calligraphy,” in Harrist and Fong, The Embodied Im-age, 243–59; Shen C. Y. Fu, Traces of the Brush: Studies in Chi nese Calligraphy (New Haven, CT: Yale University Press, 1977), 5–8, 241–43; and Cary Y. Liu, “Strangers in


a Strange Land: From the Qing Palaces to the Princeton University Art Museum,” in The Last Emperor’s Collection: Masterpieces of Painting and Calligraphy from the Liaoning Provincial Museum, ed. Willow Weilan Hai Chang et al. (New York: China Institute, 2008), 299–322, esp. 302–3.

11. For the sta tis tics, see Patricia Buckley Ebrey, Accu-mulating Culture: The Collections of Emperor Huizong (Seattle: University of Washington Press, 2008), 208–18. For Huizong’s cal lig ra phy cat a logue, ibid., 204–56, 421–26.

12. Yang Yuanzheng, “Junjia lianghang shi’er zi,” 6, citing Qi Gong 啓功, lists nine cop ies that are gen er ally accepted to date from the Tang. While there is gen eral (though not unan i mous) con sen sus about the Tang prov e nance of these cop ies, no ev i dence is avail able to date them more spe cifi cal ly. On No vem ber 20, 2010, an other handscroll believed to be a Tang copy was auc tioned in Beijing to an anony mous buyer for 308 mil lion yu an, at the time more than 46 mil lion US$ (http://arts.culturalchina.com/en/63Arts10054 .html; http://www.baike.com/wiki/《平安帖》). On April 17, 2015, the col lec tor Liu Yijian 劉益謙 an nounced that af ter five years of think ing, he now wishes “to return the scroll home,” name ly, to the Long Museum he had founded in Shanghai in 2012 (http://news.99ys.com/news/2015 / 0417/ 9 _191667 _1.shtml). In Jan u ary 2013, the dis cov ery of an other Tang copy, in a pri vate col lec tion in Japan, was reported by the Agence France Press (http://www.japantimes .co.jp/news/2013/01/09/national/japanfindsrarecopyof tangdynastywangxizhiswork/#.VKDph8ACA).

13. For im ages of the en tire scroll, see http://etcweb .princeton.edu/asianart/selectionsdetail.jsp?ctry = China& pd = Disunity%7CTang&id = 1091443. On John B. Elliott and the Princeton col lec tion, see Valerie C. Doran, “Mining the Seams of Language: John B. Elliott and the Collection of Chi nese Calligraphy,” Orientations 30, no. 3 (1999): 106–8; and Maxwell K. Hearn, “A Tribute to John Brailsford Elliott (1928–97),” Orientations 30, no. 3 (1999): 108–10.

14. For Zhang’s tran scrip tion, see his Fashu yaolu 法書要錄 (Essentials of cal lig ra phy) (Congshu jicheng chubian 叢書集成初編 ed.), 10.169; also Yan Kejun 嚴可均, “Quan Jin wen” 全晉文 (Complete prose of the Jin dy nas ty), 24.4b, in Yan, Quan shanggu sandai Qin Han sanguo liuchao wen 全上古三代秦漢三國六朝文 (Complete prose of High Antiquity, the Three Eras, Qin, Han, Three Kingdoms, and Six Dynasties) (Beijing: Zhonghua shuju, 1987), 1594; Yang Yuanzheng, “Junjia lianghang shi’er zi,” 8.

15. For two rather dif fer ent Jap a nese trans la tions, see Tōkyō National Museum et al., Shosei Ō Gishi, 278, and Morino Shigeo 森野繁夫 and Sat� Toshiyuki 佐藤利行, Ō Gishi zen shokan 王羲之全書翰 (Complete brush work of Wang Xizhi) (Tokyo: Hakuteisha, 1996), 326–27, #293, with the lat ter also in clud ing the sec ond half of the let ter. For a ten ta tive sum mary in En glish, again in clud ing the second half, see Harrist, “A Letter from Wang Hsichih,” 245. The fi nal char ac ter pres ents a par tic u lar prob lem. If the pres ent Xingrangtie is read on its own, it may end with the char ac ter jia 佳, which ap pears some 180 times in Wang’s

let ters (Antje Richter, per sonal com mu ni ca tion); if it connects to the “sec ond half” of the text, the read ing ren 任 is far more com pel ling. However, as noted by Fu, Traces of the Brush, 7, the char ac ter “con tains a faint ink out line between the right and left parts, in di cat ing the pres ence of a break or re pair in the orig i nal”—an ob ser va tion I have been able to ver ify when looking at the orig i nal through a strong mag ni fy ing glass. If the mid dle part of the char ac ter is thus miss ing, all bets are off—it surely will be nei ther jia nor ren. In this case, the dam age must have been there al ready in the ver sion seen by Zhang Yanyuan.

16. See Harrist, “A Letter from Wang Hsichih.” On the rel a tive free dom copy ists could al low them selves, see Ledderose, “Some Tao ist Elements in the Calligraphy of the Six Dynasties,” T’oung Pao 70, nos. 4–5 (1984): 271.

17. In part be cause I treat the fif teen char ac ters in isola tion from the pur ported sec ond half of the let ter, my parsing dif fers con sid er ably from Yan Kejun’s.

18. Bai, “Chi nese Letters,” 383. As noted by Ledderose, “Some Tao ist Elements,” 271: “The faith ful ness of cop ies var ied wide ly, and it ap pears that more of ten than not the copy ist in tro duced changes into his copy. Such changes ranged from sim ple mis takes and small cor rec tions that may have been per fectly jus ti fied and of ten were done with the best in ten tion, to em bel lish ments and rearrangements of en tire texts and to out right forg er ies. Judging from the ex tent of these fal si fi ca tions, one can imag ine that there was also con sid er able con fu sion in the early trans mis sion of the works of the Two Wangs.”

19. Books with Xingrangtie as their sole cover im age in clude Fu, Traces of the Brush; Character & Context in Chi nese Calligraphy, ed. Cary Y. Liu et al.; Tōkyō National Museum et al., Shosei Ō Gishi; and Chi nese Calligraphy Ouyang Zhongshi et al., ed. (New Haven, CT: Yale University Press, 2008). Xingrangtie also serves as the fron tis piece of Harrist and Fong, The Embodied Image.

20. See Harrist, “A Letter from Wang Xizhi,” 242–43; Tōkyō National Museum et al., Shosei Ō Gishi, 102–3.

21. The place ment of this slip is the re sult of the remounting by the Metropolitan Museum; a pho to graphic copy of the scroll (also held in the Princeton University Art Museum) shows that when the scroll was still in Japan, Qianlong’s ti tle slip was placed to the left of Zhang Da qian’s large seals in the fol low ing sec tion. The place from where it was re moved has been patched. During the Qing dynasty, the Palace Treasury was part of the Imperial Household agency.

22. The waxed pa per used for trac ing cop ies.23. Richard Barnhart, “Wang Shen and Late Northern

Sung Landscape Painting,” in Society for International Exchange of Arthistorical Studies, Ajia ni okeru sansui hyōgen ni tsuite アジアにおける山水表現について (The rep re senta tion of land scape in Asia; En glish ti tle: International Sym-posium on Arthistorical Studies 2) (Kyoto: Kokusai Kōryū bijutsushi kenkyūkai, 1983): 62. Barnhart, 61–64, discusses in de tail the stan dard ized spa tial ar range ment of Huizong’s ti tle slip and var i ous seals on paint ings and calligraphies in


the em per or’s col lec tion. Xingrangtie dis plays the nearly complete ver sion of this “Xuanhe 宣和 pro gram” ar rangement. For fur ther ex ten sive dis cus sion of Huizong’s seals, see Ebrey, Accumulating Culture, 114–21, 373–96, 409–11.

24. For the list of seals, see Fu, Traces of the Brush, 241. The Princeton University Art Museum holds handwrit ten notes iden ti fy ing all the seals.

25. Ledderose, Mi Fu, 99, 100, 106; Barnhart, “Wang Shen,” 63–64, 69.

26. See Ledderose, Mi Fu, 17, 20–24, for com ments from Ruan Yuan 阮元 (1764–1849) and Guo Moruo 郭沫若 (1892–1978), es pe cially on Lanting ji xu. On Ruan Yuan’s doubts, see also Ben ja min A. Elman, From Philoso-phy to Philology: Intellectual and Social Aspects of Change in Late Imperial China (Cambridge, MA: Harvard University, Council on East Asian Studies, 1990), 193, 197.

27. See Harrist, “A Letter from Wang Hsichih,” 252–55.28. For the rearrangement of the Xingrangtie seals and

col o phons (and also the dif fer ent spac ing of Wang Xizhi’s char ac ters) in the Sanxi tang cat a logue, see Maxwell K. Hearn, Cultivated Landscapes: Chi nese Paintings from the Collection of Marie-Hélène and Guy Weill (New York: Metropolitan Museum of Art, 2002), 139–43; also Ebrey, Accumulating Culture, 220–21.

29. Ernst H. Kantorowicz, The King’s Two Bodies: A Study in Mediaeval Political Theology (Princeton, NJ: Princeton University Press, 1957).

30. Robert E. Harrist Jr., “Replication and Deception in Calligraphy of the Six Dynasties Period,” in Chi nese Aes-thetics: The Ordering of Literature, the Arts, and the Uni-verse in the Six Dynasties, ed. Zongqi Cai (Honolulu: University of Hawai’i Press, 2004), 32, 45.

31. Ebrey, Accumulating Culture, 114–15.32. Yang Yuanzheng, “Junjia lianghang shi’er zi,” 9.

For the “yushu” seal, see Barnhart, “Wang Shen,” 61; Ebrey, Accumulating Culture, 373–74.

33. Ebrey, Accumulating Culture, 120–21.34. See, e.g., F. W. Mote, Imperial China 900–1800

(Cambridge, MA: Harvard University Press, 1999), 916; Mote, “The Intellectual Climate in EighteenthCentury China: Glimpses of Beijing, Suzhou and Yangzhou in the Qianlong Period,” Phoe bus 6, no. 1 (1988): 23.

35. Liu, “Strangers in a Strange Land,” 302.36. Kohara Hironobu, “The Qianlong Emperor’s Skill

in the Connoisseurship of Chi nese Painting,” Phoe bus 6, no. 1 (1988): 56.

37. This seal is from the Qianlong pe riod or lat er. It re fers to the “Leshou tang” 樂壽堂 (Hall of joy and lon gevi ty) that in 1772 was constructed in the im pe rial Qingyiyuan 清漪園 (later Yiheyuan 頤和園) gar den. Liu, “Strangers in a Strange Land,” 302–3, doubts the hy poth esis that the scroll left the im pe rial col lec tion dur ing the British and French lootings of 1860.

38. Lidai shufa lunwenxuan 歷代書法論文選 (Selection of stud ies on cal lig ra phy from suc ces sive eras) (Shanghai: Shanghai shuhua chubanshe, 1979), vol. 1, 81.

39. Qiyue duxia ertie 七月都下二帖 (Seventh month and In the cap i tal, two calligraphies), Kuaixue shiqingtie 快雪時晴帖 (Timely sunny af ter snow), Youmutie 游目帖 (Sightseeing), and Zhongyao qianziwen 鍾繇千字文 (Zhong Yao’s thou sandchar ac ter es say); see He Chuanxin 何傳馨 et al., Jin Wang Xizhi moji 晉王羲之墨跡 (Ink traces of Wang Xizhi from the Jin dy nas ty) (Taipei: Guoli gugong bowuyuan, 2010), 5, 17, 39–40; Wang Xizhi shufaji, 187, 204–5, 432–33.

40. Yang Yuanzheng, “Junjia lianghang shi’er zi,” 17.41. In fact, Su Shi’s praise seems to in clude both Wang

Xizhi and Wang Xianzhi; see Yang Yuanzheng, “Junjia lianghang shi’er zi,” 21–22, who ar gues that Qianlong mis under stood Dong Qichang’s com ment. For Su Shi’s po em, see Wang Wengao王文誥, Su Shi shiji 蘇軾詩集 (Collected poetry of Su Shi) (Beijing: Zhonghua shuju, 1987), 32.1685.

42. The term zhenji is of ten am big u ous, be ing var i ously ap plied ei ther to some one’s orig i nal writ ing or to a truth ful copy. However, here it is clear that Dong, Qianlong, and Wang Youdong all use it to in sist that Xingrangtie is by Wang Xizhi’s own hand. Zhang Chou un der stands the term the same way when writ ing that, on closer in spec tion, he had found that Wang’s cal lig ra phy was writ ten on Tang paper and hence “de ter mined that it is not zhenji” (ding fei zhenji 定非真跡).

43. For the en tire dis cus sion, see Yang Yuanzheng, “Junjia lianghang shi’er zi,” 9–16.

44. Ro land Barthes, Elements of Semiology, trans. Annette Lavers and Colin Smith (New York: Hill and Wang, 1964), 21–22. “An id i o lect de notes “the ‘style’ of a writ er, al though this is al ways per vaded by cer tain ver bal pat terns com ing from tra di tion, that is, from the com mu ni ty” or, alter na tive ly, “the lan guage of a lin guis tic com mu ni ty, that is, of a group of per sons who all in ter pret in the same way all lin guis tic state ments.” It is the lan guage that speaks to an au di ence of in sid ers but be comes un in tel li gi ble to a removed sec ond ary au di ence.

45. Wade T. Wheelock, “The Problem of Ritual Language: From Information to Situation,” Journal of the Amer i can Academy of Religion 50, no. 1 (1982): 50.

46. Anthony F. C. Wallace, Religion: An Anthropologi-cal View (New York: Random House, 1966), 233; see also Maurice Bloch, “Symbols, Song, Dance and Features of Articulation: Is Religion an Extreme Form of Traditional Authority?” Eu ro pean Journal of Sociology 15, no. 1 (1974): 71.

47. See Richter, “Beyond Calligraphy.”48. Wang, “The Taming of the Shrew,” 159; Bai, “Chi

nese Letters,” 389–93.49. Richter, “Beyond Calligraphy,” 396.50. Patrizia Violi, “Letters,” in Discourse and Litera-

ture, ed. Teun A. van Dijk (Amsterdam: Ben ja mins, 1985), 160; here cited from Richter, “Beyond Calligraphy,” 391.

51. Considering how com par a tively few let ters ex ist from his contemporaries, it is not pos si ble to as sess any par tic u lar dis tinc tions of Wang’s phras ing.

52. Wang, “The Taming of the Shrew,” 136.


53. See Wang, “The Taming of the Shrew”; also Stephen J. Goldberg, “Court Calligraphy of the Early T’ang Dynasty,” Artibus Asiae 49 (1988–89): 189–237; and Zhu Guantian, “An Epoch of Eminent Calligraphers: The Sui, Tang, and Five Dynasties,” in Chi nese Calligraphy, ed. Zhongshi et al., 196–97, 205–8.

54. See Richter, “Beyond Calligraphy,” 389–90.55. Bai Qianshen 白謙慎, “Cong Fu Shan he Dai

Tingshi de jiaowang lun ji Zhongguo shufa zhong de yingchou he xiuci wenti” 從傅山和戴廷栻的交往論及中國書法中的應酬和修辭問題 (Social in ter course and rhet o ric in Chi nese cal lig ra phy from the per spec tive of the ex changes be tween Fu Shan and Dai Tingshi), Gugong xueshu jikan 故宮學術季刊 16, no. 4 (1999): 95–133; and 17, no. 1 (1999): 137–56.

56. See Sun Guoting’s Shupu 書譜 (On cal lig ra phy), trans lated in Chang Ch’ungho and Hans H. Frankel, Two Chi nese Treatises on Calligraphy (New Haven, CT: Yale University Press, 1995), 1–16, esp. 10–12; also Wang, “The Taming of the Shrew,” 142; Qianshen Bai, Fu Shan’s World: The Transformation of Chi nese Calligraphy in the Seven-teenth Century (Cambridge, MA: Harvard University Asia Center, 2003), 221; and Cong Wenjun, “An Overview of Ancient Calligraphic Theories,” in Chi nese Calligraphy, ed. Zhongshi et al., 430–31.

57. Bai, Fu Shan’s World, 221–22.58. Ronald C. Egan, “Ouyang Hsiu and Su Shih on Cal

ligraphy,” Harvard Journal of Asi atic Studies 49, no. 2 (1989): 377; see also Wang, “The Taming of the Shrew,” 160.

59. Ronald C. Egan, The Literary Works of Ou-yang Hsiu (1007–72) (Cambridge: Cambridge University Press, 1984), 30–31.

60. Egan, “Ouyang Hsiu and Su Shih on Calligraphy,” 371–73; Ebrey, Accumulating Culture, 208, 240–41.

61. Ronald C. Egan, Word, Image, and Deed in the Life of Su Shi (Cambridge, MA: Council on East Asian Studies, Harvard University, 1994), 263.

62. See Thomas A. Wilson, Genealogy of the Way: The Construction and Uses of the Con fu cian Tradition in Late Im-perial China (Stanford, CA: Stanford University Press, 1995).

63. Wen C. Fong, “Prologue: Chi nese Calligraphy as Presenting the Self,” in Chi nese Calligraphy, ed. Zhongshi et al., 14.

64. Ebrey, Accumulating Culture, 208.65. Michel Foucault, “What Is an Author?,” in Textual

Strategies: Perspectives in Post-Structuralist Criticism, ed. Josué V. Harari (Ithaca, NY: Cornell University Press, 1979), 148.

66. In other words, an early let ter by Wang may well have been just a let ter, but at some point in his life, with his fame established, he knew that the ad dressee would treat the let ter dif fer ently from be ing merely an or di nary let ter.

67. Bai, “Chi nese Letters,” 381.68. Ibid., 382.69. Ibid., 382, 386.70. Ibid., 383.71. Theodor W. Adorno, Ästhetische Theorie (Aesthetic

the o ry) (Frankfurt/M.: Suhrkamp, 1990), 114.

72. Ledderose, Mi Fu, 42.73. Ibid., 29.74. See Qianshen Bai, “The Brush Sings and the Ink

Dances: Performance and Rhetoric in Chi nese Calligraphy,” pa per presented at the Annual Conference of the Association for Asian Studies, Bos ton, March 22–26, 1994.

75. Ebrey, Accumulating Culture, 221, 225, 230.76. See Zhiyi Yang, Dialectics of Spontaneity: The Aes-

thetics and Ethics of Su Shi (1037–1101) in Poetry (Leiden: Brill, 2015).

77. Ledderose, Mi Fu, 12.78. Ibid., 19.79. Fong, “Prologue,” 8. Fong may be slightly overstat

ing his case; em per ors since the Eastern Han are known to have spon sored writ ings in cur sive script. It is true, howev er, that this style was rarely used in of fi cial con texts.

80. According to the rhap sody “Fei caoshu” 非草書 (Against cur sive writ ing) at trib uted (not indisputably) to Zhao Yi 趙壹 (fl. ca. 180); see Yan Kejun, “Quan HouHan wen” 全後漢文 (Complete prose of the Latter Han), 82.9b–11a, in Yan, Quan shanggu sandai Qin Han sanguo li-uchao wen, 916–17.

81. Ebrey, Accumulating Culture, 233–35.82. Ibid., 245–55.83. As noted by Ledderose, Mi Fu, 28, Chu Suiliang

“held the mo nop oly as the ar bi ter in mat ters of Wang Hsichih and thus had a de ci sive role in shap ing the im age of Wang Hsichih as it appeared to later gen er a tions.”

84. The col lec tion of cop ies of Wang’s Lanting ji xu, Lanting moji huibian 蘭亭墨蹟彙編 (Compilation of ink traces from Lanting) (Beijing: Beijing chubanshe, 1985), con tains eight ver sions— six from the Tang and two from the Yuan dy nas ty. Mi Fu knew “more than ten cop ies” (Ledderose, Mi Fu, 76). Jiang Kui 姜夔 (ca. 1155–ca. 1221) in his Xu shupu 續書譜 (On cal lig ra phy, con tin ued) speaks of “more than sev eral hun dred cop ies” cir cu lat ing at his time; see Chang and Frankel, Two Chi nese Treatises, 25; also Ledderose, Mi Fu, 20, and Harrist, “Replication and Deception,” 46–49.

85. The history of the imperial collection often mirrored the strength and fate of dynastic rule, with power ful emperors seeking to boost their po liti cal legitimacy by amassing prestigious artifacts that, in turn, left the court in times of po liti cal weakness and disintegration; see Lothar Ledderose, “Some Observations on the Imperial Art Collection in China,” Transactions of the Oriental Ceramic Society 43 (1978–1979): 38–46. Thus, just as Xingrangtie repeatedly moved into and out of the imperial collection, other works such as Wang Xizhi’s Fengjutie 奉橘帖 (Presenting tangerines) did exactly the same.

86. In Harrist and Fong, The Embodied Image, 92 (cat a logue num ber 2).

87. Italo Calvino, “Why Read the Classics,” trans. Patrick Creagh, New York Review of Books, Oc to ber 9, 1986 (online at http://www.nybooks.com/articles/archives/1986/oct/09/whyreadtheclassics/).

made by the empire: wang xizhi￢ﾀﾙs xingrangtie … · ex em pli fy ing moral rec ti tude and...

Documents