Collection online

painting / 繪畫

  • Object type

  • Museum number


  • Description

    Large painting showing Buddha (probably Śākyamuni) preaching in a Paradise composition. The Buddha is seated under a jewelled bodhi tree forming a canopy, surrounded by four seated bodhisattvas and six monk disciples. A small figure of a female donor is in the bottom left-hand corner; the topknot of a male donor is visible in the bottom right. An apsaras is in the top right corner, and a blank carotuche in the front centre. Ink and colours on silk.


  • Culture/period

  • Date

    • 701-750 (circa)
  • Production place

  • Findspot

  • Materials

  • Technique

  • Dimensions

    • Height: 139 centimetres
    • Width: 102 centimetres
    • Height: 163.3 centimetres (Painting in frame)
    • Width: 121.5 centimetres (Painting in frame)
    • Depth: 4 centimetres (Painting in frame; including old fixing)
  • Curator's comments

    This, one of the earliest among the paintings found in Cave 17 at Dunhuang, is also one of the best preserved and one that is clearly related to Sui and early Tang paradise or preaching scenes found on the walls of the caves themselves. Among the silk paintings, the closest parallel is the fragment of an Amitābha preaching scene from the Pelliot collection (Bannières, No.23). Several features seem to assure its early date: the composition itself; the colour scheme, both in the decoration of the canopy and throne and in the modelling of the figures; and the single surviving figure of a female donor in the lower left corner.
    The main figure can probably be identified as Sākyamuni, although Amitābha, with Avalokitesvara and Mahāsthāmaprāpta, has also been suggested. His hands are in vitarka-mudrā, or mudrā of the exposition of the Law, common to both Amitābha and Sākyamuni. Clearly, however, the entire painting is comparable to examples of “Preaching the Law” from caves of the early Tang period (e.g., Caves 103, 321, 329). In these the chief intent was to suggest the omnipresence of Buddha by constant repetition of the same scene. The type can be seen as a development from preaching groups on the walls of Sui caves at Dunhuang, where there are in general only three figures, the Buddha and two Bodhisattvas (cf. Cave 390). In the present painting, in addition to the Bodhisattvas representing the Mahāyāna, or Greater Vehicle, the presence not just of two monks representing the Hinayāna, or Lesser Vehicle, but of six in all seems to favour the identification of the central figure as Sākyamuni, the historical Buddha with whom they would be associated. Presumably there simply was not room to depict all ten of Sākyamuni’s historical disciples. In later, more developed representations of the Pure Land paradises, whether on silk or as wall paintings (e.g., Pl.8) the disciples are still fewer, with two-Ananda and Kāsyapa-characterized in appearance as young and old respectively. Here, on the other hand, there are no obvious formalized distinctions, but the figures, and especially their bald heads, are carefully studied and drawn.
    The painter’s concern with rendering the forms of the shaven heads is clearly apparent (Pls.7-2, 7-3). At the same time there is a decorative aspect to the depiction of the faces in the somewhat stylized scheme of colour shading and white highlighting that seems to be used almost independently of the outer contours in ink. We may see here, and in other paintings at Dunhuang, a fusion of the “aesthetics of light”(Bussagli, 1963, pp.31ff), which was a feature of the art of the West, with the tradition of modelling through line that was an inherent characteristic of painting and allied arts in China. We should not, however, draw too rigid a distinction here, since the shading of colours was in fact also practised in China (e. g., in the handscroll Admonitions of the Court Instructress, attributed to Gu Kaizhi, in the British Museum, especially in the garments and drapery hangings) from the period of the Northern and Southern Dynasties onwards. Nor does the stylization of the colour shading imply that the style as a whole is fixed and derivative. On the contrary, when examined closely, there are many lively details which exhibit a freshness of approach. All of the subsidiary figures are varied in pose and expression. At the lower left, one Bodhisattva is seen in three-quarter view from the back (though with the face in profile). Among the monks, two of whom are partly hidden behind the trunks of the Bodhi trees, there are many varied expressions of the eyes. Their eyebrows are drawn initially with single curved lines, over which are numerous shorter strokes in grey ink. In most cases these strokes are also curved and parallel with the initial line, but in the monk on the far left these paler strokes cross it vertically to give him bushy eyebrows. Again, one monk has his mouth open and teeth showing as if he is chanting. Here the line between cheek and mouth, normally brought close to the lips by the fullness of the cheek, is elongated and drawn back (Pl.7-3). A close parallel to this can be seen in the wall painting showing a monk reading from a sutra roll (Cave 201; Dunhuang bihua, Pl.109). As a further confirmation of the early eighth century A.D. date of this painting, the wall paintings of the tomb of Princess Yongtai (A.D. 706; Tang Yongtai gongzhu mu bihua ji, Pl.4) provides another contemporary example. By the late eighth century, the convention has changed and the line between the lips ends in a gentle swelling; later still (in the tenth century) this is elongated and terminated by a sharply angled hook.
    The concern with three-dimensional form, already noticed in the monks’ heads, extends to the composition as a whole, where the central figure is left almost free, while those surrounding him occupy relatively less of the picture area. Some are even partially obscured behind the trunks of the Bodhi trees supporting the canopy. Although all the figures have haloes, only those of the two lower Bodhisattvas are actually opaque. The transparency of the rest adds greatly to the feeling of space (cf. Cave 334). The painter has also used his discretion here with a nice sense of relative values: thus the same halo (that of the monk behind the tree trunk on the right) is transparent in order to reveal the monk who stands behind him, but opaque to the uninscribed cartouche above (Pl.7-2). The effect is to focus attention on the main figure, as well as to give the whole setting a certain measure of depth. This feature is a Chinese, rather than a Western, contribution. Among the earlier compositions recovered by Stein from Dunhuang, none is more remarkable in this respect than the great embroidery showing Sākyamuni Preaching on the Vulture Peak (Vol.3, Pl.1). There the figures of the two monks emerge as if from behind the Bodhisattvas on either side, leaving clear the rocky background of the central figure, and contributing not a little to the impressiveness of the whole composition. Later, in the tenth century particularly, the distinctions and relative importance of figures of the Buddha and his attendants seem to be lost, or rather forgotten in the need to fill the available surface evenly with colour and decoration.
    To return to the painting, several further features distinguish it from others and serve to link it both with paintings in the caves and with possible parallels that must have existed in the capital, Chang’an, hundreds of miles to the east. The surviving donor (Pl.7-5), a young lady demurely seated with a lotus flower in her hand, clearly recalls in her neat hair and high-waisted dress the funerary figurines and wall paintings of court ladies from princely tombs of the very early eighth century near Chang’an. Her face and the dark black of her coiffure are admirably set off by a side bud from the stem of the lotus throne above her, and she holds a long-stemmed red flower with pairs of opposing leaves. Although only the top of the cap of the male donor is visible, plumes of smoke curling gently upward towards the central Buddha throne show that he originally held a hand censer. In the centre foreground, the space left and ruled (although never actually used) for a dedicatory inscription takes the form of a monumental stele of the type well known from the beilin in Xi’an, supported on a tortoise and crowned by intertwined dragons. Although the dragons are here replaced by a floral design of overlapping petals, this is manifestly the explanation of the semicircular cap, and there is even a rectangular space left, which on an actual stone stele would have served for the title in formal characters (Fig.15). Similar stelae for dedicatory inscriptions are also found in some of the wall paintings (e.g., Cave 335, dated A.D. 686), as well as in Stein painting 499 in New Delhi (Stein, Thousand Buddhas, Pl.XI), and are contributory evidence of reasonably close and frequent contact between Dunhuang and the capital, such as could only have occurred in the period before the area came under Tibetan control in A.D.781.
    Among the wall paintings at Dunhuang, so far as can be judged from photographs and reproductions, preaching groups in Cave 329(Lo Archive, no.1361), Cave 103 and Cave 321 are the closest to this painting. In Cave 329, the Buddha’s right hand is in vitarka-mudrā, as in the Stein painting. The canopy above has a net fringe and cloud-like embellishments framed by the foliage of the two Bodhi trees. The lotus throne on which the Buddha sits, although not so embellished, has compressed petals similar to those seen in the Stein painting. Below it the space for the inscription may also be based on a monumental tablet with a capstone. Cave 321, high on the east wall above the entrance, has the group spread horizontally; the two monks both face in the same direction (Lo Archive, no.842). In both these groups the attendant Bodhisattvas, as well as the Buddha, have the tall horseshoe-shaped haloes, bearing some resemblance to the oval ones of the lower Bodhisattvas in the silk painting. There are similarities, too, in the shading, highlighting and delineation of the face of the preaching Buddha on the south wall of Cave 103(Lo Archive, no.1229). There we find the same convention for the transition from the cheek to the mouth, already described for the chanting monk. This cave is generally dated to the period of the Empress Wu in the late seventh century A.D., and may serve to confirm the early eighth century date here attributed to Stein painting 6(Pl.7).
    Other features also invite comparison with the wall paintings. These are the clouds above the canopy and those which decorate the petals of the Buddha’s lotus throne. The canopy clouds, ridden by apsarasas scattering blossoms (Pl.7-6), belong to a long tradition stretching back, no doubt, to the earliest wall paintings of the Liang dynasty, long since lost at Dunhuang through the collapse of that part of the cliff that had contained the earliest caves. In this preaching scene, as in other paintings of early Tang date (Cave 329), the clouds are composed of separate plumes, each a different colour and delicately shaded. This formula was to be considerably simplified in the course of the ninth and tenth centuries A.D. at Dunhuang. The cloud plumes on the petals of the lotus throne give a similarly rich effect: they can be compared not only with similar decoration in the wall paintings (for instance the decorated halo and lotus petals of an eleven-headed Avalokitesvara in Cave 334) but with actual lotus pedestals of the mid-eighth century A.D. preserved in the imperial repository of the Shōsō-in, Nara, whose individual petals are richly painted in various colours (Shōsō-in no kaiga, Pls.109ff).
    Before concluding these notes, it is appropriate to make a few comments on the construction of the painting, which is in ink and colours on fine silk. As with all the larger silk paintings, the size of the silk to a considerable extent determines the size of the finished panel. A single width of silk runs down the centre of the painting. The width of this piece is 53. 7cm. A second piece, cut down the middle into two equal halves, is sewn on, selvedge to selvedge, to the first piece. The cut edges on the sides of the painting should originally have been protected by a sewn-on border of doubled silk. In this case, however, the border has disappeared. We may infer from the fact that the half-widths are now only some 23. 7cm wide that about 3 cm on either side has also been lost. This assumption is confirmed by reference to the lotus thrones of the main Bodhisattvas, which need another centimetre or two to be complete. Nor is there anywhere any trace of stitching of the four borders. Along the top and left edges, however, about 1cm from the edge, an ink line has been ruled, and this may have marked the limit of the design when it was first laid out. It would seem that a certain amount of lateral compression was already necessary for the painter in order to fit in the many figures: note, for example, the rather vertical petals at the sides of the lotus thrones of the two main Bodhisattvas and the slight degree of asymmetry. Yet it is partly as a result of this asymmetry and the varied stances of the many figures that the painting is one of the most satisfying among those found at Dunhuang. One is left with the impression that here solutions to problems of representation were still being sought, rather than of rigid adherence to established prototypes.Zwalf 1985

    This, one of the earliest and best-preserved paintings from Cave 17, is related to Sui (AD 581- 617) and early Tang preaching scenes on the walls of the caves. Celestials scatter flowers on the assembly below where, among Bodhisattvas of the Mahāyāna, the six monks representing the Hīnayāna indicate that the central figure is probably the historical Buddha. The space left and ruled for a dedication takes the form of a monumental stele. The kneeling lady donor is in early 8th-century costume; her husband on the right is now lost except for the smoke from his incense-burner.這是敦煌第17窟中出土的繪畫中時代最早、保存狀態最好的作品之一,而且明顯與敦煌石窟壁畫中隋及初唐時期的淨土變和說法圖有密切關係。絹畫中,與它最接近的則是伯希和搜集品中的阿彌陀淨土圖的殘片 (《敦煌幡畫》圖23)。我們依據以下特徵將其定為早期作品:首先是根據它的構圖,華蓋、寶座的裝飾及表現諸畫像身體色彩的手法,以及畫面左下角女供養人像的表現形式等等。
    這幅畫畫家所要表現的顯然是對剃髮比丘頭部的圓潤(圖7-2,7-3)。與此同時,在描繪面部時使用了稍具形式化的高光,看起來似乎與黑色線描毫無關係,給人以裝飾性的感覺。此種技法在其他敦煌畫中也可見到,其在中國繪畫藝術固有技法——用線描進行立體描繪,又融入了西方美術的“光美學”(Mario Bussagli《中亞繪畫》,1963,p.31以下)。這裡沒有必要在兩者之間劃定一嚴格界限,事實上在中國,顏色勾勒是南北朝以來一直沿用的技法(例如,大英博物館藏相傳顧愷之所作的《女史箴圖》的衣褶部分)。另外,不能因說顏色勾勒方法已經是形式化了,就認為其整體風格也是固定化的,而沒有獨創性。恰恰相反,詳察其細部,可發現多處採用了新的表現方式。如,除主尊外,其他的畫像無論是姿態還是表情,都富變化。左下方的菩薩背朝外,身體稍往前傾(面向主尊)。有二個比丘的部分身體被菩提樹遮住,比丘的眼神各異。其眉毛則是先勾一條曲線,再用淡墨描出無數眉毛。雖然大部分眉毛稍有彎曲,卻基本與最初的曲線平行。最左邊的比丘淡淡的眉毛筆直,似“蚰蜒眉”。另外一個比丘張嘴露齒,神態像是正在吟誦。其他比丘臉頰和嘴間的線條緊挨嘴唇,只有細微的表現,而此比丘臉頰和嘴間的線條則拉長到下方(圖7-3)。與此極為相類的則是敦煌201窟壁畫中的誦經比丘(201窟;《敦煌壁畫》,圖109)。作為這幅畫斷為八世紀初的另一個確証,永泰公主墓壁畫(706年,《唐永泰公主墓壁畫集》,圖4)提供了同時代的例証。到了8世紀後半葉,繪畫方式有了變化,嘴唇之間線條兩端稍微變粗。再往後(10世紀)線條拉得更長,末端彎曲得很尖。


  • Bibliography

    • Zwalf 1985 312 bibliographic details
    • Whitfield & Farrer 1990 bibliographic details
    • Rawson 1984 fig. 127 bibliographic details
    • Whitfield 2004a figs.94 and 96 bibliographic details
    • Whitfield 2004b pl.232 bibliographic details
    • Whitfield 1982 pl.7 bibliographic details
    • Stein 1921b pl.X bibliographic details
    • Stein 1921a pp.1055-1056 bibliographic details
  • Location

    Not on display

  • Exhibition history


    1996 14 Jun-29 Sep, London, The British Library, The Mythical Quest
    2007 8 Feb-5 Aug, BM Gallery 91, 'Gods, Guardians and Immortals: Chinese Religious Paintings'
    2013 5 Oct – 29 Dec, London, Victoria and Albert Museum, The Making of Chinese Masterpieces: Chinese Painting from the 8th to 19th Centuries
    2016 7 May-4 Sep, Los Angeles, Getty Research Institute, Cave Temples of Dunhuang: Buddhist Art on China's Silk Road

  • Subjects

  • Associated names

  • Associated places

  • Acquisition name

  • Acquisition date


  • Acquisition notes

    For full acquisition history, see 1919,0101,0.1.

  • Department


  • Registration number


  • Additional IDs

    • Ch.liii.001 (Stein no.)
Large painting showing Buddha (probably Sakyamuni) preaching in a Paradise composition. The Buddha is seated under a jewelled bodhi tree forming a canopy, surrounded by four seated bodhisattvas and six monk disciples. A small figure of a female donor is in the bottom left-hand corner; the topknot of a male donor is visible in the bottom right. An apsaras is in the top right corner, and a blank carotuche in the front centre. One of the earliest of the paintings found in Cave 17. Ink and colours on silk.

Detail: Other

Large painting showing Buddha (probably Sakyamuni) preaching in a Paradise composition. The Buddha is seated under a jewelled bodhi tree forming a canopy, surrounded by four seated bodhisattvas and six monk disciples. A small figure of a female donor is in the bottom left-hand corner; the topknot of a male donor is visible in the bottom right. An apsaras is in the top right corner, and a blank carotuche in the front centre. One of the earliest of the paintings found in Cave 17. Ink and colours on silk.

Image description



If you’ve noticed a mistake or have any further information about this object, please email: 

View open data for this object with SPARQL endpoint

Object reference number: RFC606

British Museum collection data is also available in the W3C open data standard, RDF, allowing it to join and relate to a growing body of linked data published by organisations around the world.

View this object

Support the Museum:
donate online

The Museum makes its collection database available to be used by scholars around the world. Donations will help support curatorial, documentation and digitisation projects.

About the database

The British Museum collection database is a work in progress. New records, updates and images are added every week.

More about the database 


Work on this database is supported by a range of sponsors, donors and volunteers.

More about supporters and how you
can help