1 Introduction
Apart from Aurel Stein’s publications of the results of the field activities he carried out in the early 20th century,1 which still represent a mandatory starting point of any research on ancient Khotan, the most valuable contributions to the study of Khotanese painting remain Joanna Williams’s long article devoted to this topic2—for the author’s insight and praiseworthy effort to analyse the artistic phenomenon against its religious background—and the chapters dedicated to wall painting in Gerd Gropp’s monograph3—for the accurate and balanced presentation of the materials recovered in the Khotan oasis by Emil Trinkler in 1928. In the last two decades, new discoveries—at Dandān-öilïq and in the Domoko (Chin. Damagou
Along with the material found in situ through organised diggings, part of what we know of Khotanese painting consists of fragments of murals acquired by local smugglers, who, in order to get the maximum profit from their business, had the wall paintings (as well as clay sculptures and decorated wooden architecture) from ancient sites of the oasis cut into fragments to be sold separately, often to different individuals. Fragments of murals reportedly from Balawaste, Khadalik, Dandān-öilïq, and other sites ended up in a number of private collections and, eventually, museums.6 Needless to say, in these cases the loss of information is much greater than that caused by the archaeological methods Stein employed, as well as those of other explorers then working in the Tarim Basin, which were surely inadequate if compared to the modern standard. Attempts have been made to restore parts of the original compositions by piecing together the available fragments, with good results,7 but much work still waits to be done to make the most of these disiecta membra.
A reappraisal of Khotanese painting appears much desirable if we consider the amount of research carried out, during the last decades, in the fields of Buddhist archaeology and art history in the Tarim Basin, Western Central Asia, Gandhāra, and China, and the valuable results it produced. As the methodological basis of most of the Khotanese archaeological record is weak or nearly nonexistent, any new attempt of investigation of Khotanese murals is still bound to be largely based on art historical analysis; all the more reason to refine it as much as possible.
A major concern should be a correct evaluation of diversity in style, iconography, and technical quality, which is so often explained in terms of chronological diversity, whereas it may attest to diverse choices and degrees of technical skill of local workshops. In this regard, I think it appropriate to recall Williams’s keen remark that a now lost painting from Khadalik includes, and so connects, “[…] a variety of types of images which are usually preserved only in fragments”,8 several of which denote discreet stylistic traits. That is, different subjects or figures may require different styles or, at least, stylistic marks, in one and the same composition: a true challenge to any analysis based on the assumption that iconography and style dwell in two separate realms.
Another matter that warrants caution is the (mis)use of elements of the Classical (Greco-Roman) iconographic or ornamental repertoire as chronological clues in art historical analysis, a sensitive issue in the exegesis of pictorial arts in the Indo-Iranian borderland as well as in Central Asia as a whole.
Furthermore, possibly based on its specific religious background—namely, a stronger Mahāyāna orientation, which affected its thematic choices—Khotanese pictorial arts are generally viewed and analysed as a phenomenon per se, essentially unrelated to other major artistic centres in the Tarim Basin. A more careful investigation of the artistic relationship between Khotan and Kuča, first and foremost, may lead to rewarding results, and highlight, at least in part of their otherwise distinct repertoires (and doctrinal orientations), a common ground based on similar iconographic and stylistic choices.
Last but not least, and closely related to all of the desiderata listed above, a much-needed improvement is the definition of a more consistent and reliable chronological frame, which is the specific issue to which these notes are devoted.
2 Chronology of Khotanese Painting
The chronology of Khotanese painting is largely based on a few elementary, however sensible, assumptions formulated by Stein. According to Stein’s reconstruction, Khotanese painting is essentially late, covering a time span of approximately three centuries, from the 6th to the 8th centuries, with a greater concentration of the extant evidence in the 8th century.9 The most objective data for this chronological span concern its very end, that is, the eighteen Chinese documents found in a few ruined building at Dandān-öilïq (D V, D VII and D VIII), five of which date to 781–789.10 The latest date (789) serves as a terminus ante quem for the northern sector of the site. The execution of the mural paintings preserved in those ruins, as Stein suggests, cannot be earlier than the beginning of the same century. Based on an undeniable iconographic and stylistic consistency with the murals, the painted wooden panels from Dandān-öilïq and other sites of the Khotan oasis (to which we may add the findings from Endere, further east in the Southern Tarim Basin) reasonably belong to the same time span.11 Also, the architectural and pictorial evidence allows us to associate to Dandān-öilïq the wall paintings from Khadalik, Balawaste, and Domoko, which plausibly date from the 8th century as well.
As for the Farhad Beg Yailaki murals, the absence of Tang (618–907,
As is clear, even from this sketchy overview, the chronology of Khotanese painting is not yet definitely settled and there seems to be no material shedding light on the formative stages of pictorial arts in the oasis. If we rely on the bulk of the material Stein unearthed, no evidence of murals dating from a period earlier than the beginning of the time span mentioned above (6th c., according to Stein; late 7th c., according to Williams) seems to be available. Whatever the oscillations of the absolute chronology, depending on the view held by each scholar, there seems to be general agreement on the fact that Khotanese painting represents a coherent corpus dating from a relatively late period, i.e. 7th (or 6th) to 8th centuries. This is why, the wall paintings unearthed at Karadong, a site on the Keriya River, on the north-eastern outskirts of the Khotan oasis (190 km north of Yutian,
3 Karadong Murals
Karadong was first visited by Sven Hedin (1896), then by Stein (1901 and 1906),17 who carried out limited excavations, as he deemed it a site of little interest. Karadong was more extensively explored in the 1990s by a Franco-Chinese team led by Corinne Debaine-Francfort and Abduressul Idris. The investigations mainly focused on the ruins of some houses and of two Buddhist temples, named Shrine A and Shrine B, from where all the fragments of painted murals have been recovered. Shrine A was excavated in 1993, and Shrine B, one year later. In the first two reports on the Keriya Basin diggings,18 only the first building is described and commented upon; information and pictures of Shrine B are provided in the catalogue Kériya, mémoires d’un fleuve (2001). Both temples reproduce an architectural layout recorded in many sites in the south of the Tarim Basin: a central cella surrounded by one (B) or two (A) corridors for circumambulation (fig. 3.1).
Plan of Shrine A. Karadong
Debaine-Francfort and Idriss, Kériya, 83, Fig. 2
Shrine A has yielded a larger number of mural fragments, but Shrine B offers better conditions for making a credible reconstruction of the iconographic programme. The accuracy and expertise shown by the team in the hard task of handling and restoring parts of the original painted compositions is worthy of praise. According to the reconstruction proposed for Shrine B (fig. 3.2), the walls were subdivided into three registers: in the lower register there was an ornamental band; in the middle or main register (ca. 1.20 m high) there were large images of buddhas, each figure standing on a lotus flower (the lateral ones in three-quarter view and the central one in frontal view) with a pair of smaller images of seated buddhas above them, on both sides of the standing figures (fig. 3.3); in the upper register there was a row of rectangular upright panels framed by a vegetal scroll and subdivided into two halves by a horizontal band with a meander (or Greek fretwork), each panel contained a buddha seated on a lotus pedestal (fig. 3.4).
Shrine B, reconstruction of the iconographic programme on the eastern wall. Karadong
Debaine-Francfort, Kériya, 89, Fig. 15
Shrine B, northern wall, middle register. Karadong
Xinjiang Institute of Archaeology, Urumqi, after Debaine-Francfort and Idriss, Kériya, 100
Shrine B, upper register, meander. Karadong
Xinjiang Institute of Archaeology, Urumqi, After Debaine-Francfort and Idriss, Kériya, Fig. 104
The iconographic programme in Shrine A might not have been very different from the one just described. Patches of paintings still preserved in the lower part of two walls (west and south, fig. 3.5) show the feet of buddhas standing on lotus flowers, in three quarter view, alternately facing to the right and to the left, on a dark floor sprinkled with tiny flowers. In the drawing provided in the first publication (but this detail is apparently unrecorded in the description), part of a small figure of a kneeling worshipper in caftan and trousers can be seen on the left of the first buddha (from the left), on the western wall.19 From the fragments recovered among the debris, several smaller images of buddhas seated on lotus pedestals with ornamental motifs could be recomposed. The buddhas, depicted on a white background filled with white (lotus blossoms?) and dark flowers,20 were set in panels framed by vegetal scrolls and separated by horizontal ornamental bands (undulating or zigzag ribbons) (fig. 3.6). One is led to suppose that, just as in Shrine B, in this temple the main (middle) register of the walls was occupied by a series of standing buddhas, whereas the upper register was reserved for rows of small seated buddhas, a well-known scheme conventionally referred to as ‘Thousand Buddhas’.
Shrine A, drawing of the painting preserved on the northern and western wall. Karadong
Francfort, Debaine-Francfort, Idriss and Wang, “Agriculture Irriguée,” Fig. 24
Shrine A, western wall, seated buddhas from the upper register. Karadong
Xinjiang Institute of Archaeology, Urumqi, After Debaine-Francfort and Idriss, Kériya, 93
Therefore, the paintings in both temples were most likely based on similar iconographic schemes. Although, as it has been rightly remarked, a difference in workmanship is noticeable, since the figures and ornamental motifs in Shrine B are more accurate and less ‘geometric’ than those in Shrine A.
As a complete and detailed description of the murals unearthed in both temples is available in the publications mentioned above, I will focus instead on some of their most peculiar features, especially on those which I deem particularly meaningful for a reconsideration of the chronology established by the authors of the excavations, the 3rd century.
As the latest publication of the finds cursorily states, the date of Karadong paintings is based on a radiocarbon test,21 but no details are given about the sample(s) employed for the analysis; we may wonder whether it was taken from the wooden structure of a wall (as it has been elsewhere in Xinjiang sites, with misleading results) or from the mortar layers of the painting, which usually contain organic elements.
In fact, an early date (2nd–4th c.) was already assumed in the first publications about the Sino-French activities on the Keriya Basin (Karadong being considered contemporary with Miran), well before the C14 analysis was carried out.22 The reasons such an early date is proposed mainly rest on iconographic and stylistic arguments (all proposed again in the 2001 catalogue), which, in my opinion, deserve scrutiny, rather than on the finding of wuzhu (
I will start with the depictions of the buddhas in Shrine A, in particular, the seated buddhas presumably belonging to the upper register (fig. 3.6), which surely represent the most peculiar trait of the Karadong murals. What strikes most is the stylised outline of their heads, the disproportionally large necks crossed by two or more skin folds, and the straight elongated earlobes. The hair is arranged in small curls. The eyebrows are very elongated, as are the half-closed eyes. We also notice a small straight nose with a white brush stroke highlighting the nasal ridge, and a small mouth and chin.
A further stylistic feature to take notice of is, in my opinion, that the buddha heads—always in three-quarter view, facing either direction—look asymmetrical, in that the receding part of the face appears unnaturally contracted compared with the part closer to the viewer. In other words, we detect the same convention Mario Bussagli highlighted long ago in Kuča painting (6th–7th c.), which he rightly related to the principle known as kṣayavṛddhī (Skt. kṣaya, diminishing, vṛddhī, increasing)24 in ancient Indian texts on painting.25 As a matter of fact, the simplified geometric rendering of the head and neck outlines (almost round face, conical neck crossed by skinfold), associated with very sloping shoulders, substantially matches a scheme widespread in the Kızıl painting (fig. 3.7), with the difference that at Karadong, some elements of this structure—the necks, in particular—are emphasised, a feature possibly due to the specific idiosyncrasy of a local workshop.
Mural from the third last cave in the Small Gorge. Kızıl
Museum für asiatische Kunst, Berlin, after Albert von le Coq, Die Buddhistische Spätantike in Mittelasien, vol. 4. Atlas zu den Wandmalereien (Berlin: Verlag Dietrich Reimer/Ernst Vohsen, 1924), TAF. 11
The seated buddhas from the upper register are portrayed either in dhyānamudrā or with the right hand jutting out from the cloak at chest level, while the arm is concealed by the cloth (fig. 3.6). In other words, the buddhas are shown in the ‘Lateran Sophocles’ posture, which is sporadically attested in Gandhāran art (either stone or stucco sculpture) and, as Debaine-Francfort remarks, is recorded in the funerary art of Palmyra and in the Eastern Mediterranean (that is, in the 1st centuries).26 Along with this evidence, which represents the ultimate source of the iconographic element under examination, one should not forget to mention the depiction of buddhas in the posture of the ‘Lateran Sophocles’ on a painted wooden post from Balawaste (Khotan oasis, 8th c.)27 (fig. 3.8).
Drawing of a painted wooden post. Balawaste
Metropolitan Museum of Art, New York, Author’s Drawing after Williams, the Iconography of Khotanese Painting, Fig. 25
As is typical of Thousand Buddha compositions, the colour of the dress—along with the head orientation and the gesture—is one of the few elements of variety in an otherwise monotonous row of seated buddhas. In one case (fragment from Shrine B, fig. 3.9), the brown cloth is decorated with rosettes made of seven white dots. A stripe of cotton textile with identical ornamentation was found at Karadong (fig. 3.10),28 and a very similar pattern is also seen in two fragments of a mural from Balawaste, each preserving the lower part of a lokapāla (fig. 3.11).29
Fragment of cotton textile. Karadong
Xinjiang Institute of Archaeology, Urumqi, after Debaine-Francfort and Idriss, Kériya, 80, CAT. 22
Shrine B, fragment of mural from the west wall. Karadong
Xinjiang Institute of Archaeology, Urumqi, after Debaine-Francfort and Idriss, Kériya, 103, CAT. 35
The excavators rightly remark that in the Karadong murals, the Indian imprint is much stronger than the Gandhāran one,30 but they are no more specific than this about what elements of iconography and/or style denote an Indian influence. If we look for possible South Asian artistic sources, keeping in mind the date to which the authors assign the Karadong paintings (3rd c.), we find ourselves in the artistic scenario of the Kushan period. If we leave out Gandhāra, then Mathurā, the other major artistic centre in Northern India, offers no better alternative, since we hardly see in the Mathurā sculpture of the Kushan period any plausible parallel, either in iconography (to begin with the buddha image) or in style. This is not to deny any Indian influence at Karadong, where it is, on the contrary, as patent as anywhere else in the Tarim Basin, in different degrees and modes. The problem lies in the chronological mismatch, in that, apart from the cases in which a strong Gandhāran inspiration is evident (e.g. Miran), the major Indian component detectable in the painting and sculpture of the Tarim Basin ultimately bears a Gupta period stylistic imprint (early 4th to 6th c.): the rendering of the physiognomic traits of the seated buddhas in Shrine A, with their arched eyebrows, long, half-closed eyes, and head gently tilted to the side, is reminiscent of many buddha images in the Ajanta paintings (late 5th c.).
Let us now consider a few ornamental motifs, beginning with the meander (Shrine B) (fig. 3.4), for which Debaine-Francfort mentions only Western parallels dating from the 3rd century or earlier.31 The meander, however, is also witnessed in South Asian and Central Asian Buddhist painting, from the 5th to the 7th–8th centuries.32 In different variants (including the tridimensional one we see at Karadong), the meander appears in the paintings of Ajanta (late 5th c., fig. 3.11)33 and Bāgh (Cave 4, 6th or 7th c., fig. 3.12),34 and in the terracotta plaques from Harwan (Kashmir, 5th c.).35 Rather than taking it as a sign of continuity with the Romano-Hellenistic repertoire of the 1st century that was embedded in the Kushan arts, Maurizio Taddei links the later spread of the meander in Indian art during the 4th to 6th centuries to its re-introduction—along with other Classical motifs (or of ‘new ways of employing them’)—from the Near East, possibly through maritime routes.36
Detail of the ceiling of Cave 2. Ajanta (India)
© Manfred Sommer
Drawing of a meander, mural painting. Bāgh (India)
Marshall et al., the Bagh Caves, pl. 17
A direct or, more probably, an indirect influence from Indian Buddhist art of the Gupta period and cognate traditions may explain the presence of the meander in the Tarim Basin, where it is not only recorded at Karadong: it also occurs in a fragment of mural from Balawaste, as ornament in the halo of a bodhisattva;37 in an ornamental band in a painting at Endere (shrine in building E.ii, fig. 3.13),38 7th–early 8th century; and finally, in the Kuča oasis, in the painted decoration of Cave 167 ceiling (Laternendecke) (fig. 3.14),39 and in a well-known ‘2nd style’ mural at Kızıl (Cave 224) depicting the story of Ajataśatru (early 7th c.?), in which a meander decorates one of the jars in the left part of the scene (the jar behind the one in which Ajataśatru sits in despair).40 It seems clear that the meander is not good evidence for supporting a date for the Karadong murals in the 3rd century, as it fits much better in a late Khotanese context.
Mural painting in the Buddhist shrine in E.ii. Endere
Stein, Ancient Khotan, PL. 10
Painted ceiling with meander decoration in Cave 167, Kızıl (Kuča oasis)
© Ikuko Nakagawara
Mosaic in the Great Palace of Constantinople
The same is true for the ribbons separating the panels with seated buddhas in Shrine A. The undulating tridimensional ribbon, in particular, is surely a pattern of Western origin; but nothing compels us to trace its source to the Greco-Roman repertoire of the 1st century, as the motif enjoys long-lasting popularity. In a particularly fine execution, it appears in the mosaics of the Great Palace of Constantinople41 (dating from the mid-6th or early 7th c.42) (fig. 3.15), but it also occurs, in the same trompe-l’œil rendering as at Karadong, in the paintings of the Red Hall in the Varakhsha Palace, in Sogdiana (early decades of the 8th c.).43
4 Conclusion
Apart from the individual iconographic elements and ornamental motifs I discussed so far, which nonetheless provide significant chronological clues, it is the very iconographic programme, as it has been reconstructed in Karadong Shrine B and that we can reasonably attribute to Shrine A as well, which should encourage us to associate Karadong with the late Khotanese artistic and ritual horizon. In other words, the Karadong findings confirm the pictorial repertoire we are accustomed to labelling as ‘Mahayanic’, in which the primacy belongs to large buddha images (either standing or seated on lotus flowers), Thousand Buddha compositions, bodhisattvas, assemblies that include minor deities, and worship scenes, with no or scarce interest in narratives. Large standing buddha images (most often only their feet surviving) that rest on lotus blossoms against a dark floor dotted with tiny flowers, are a rather familiar picture in the ruined temples of Dandān-öilïq and other sites in the Khotan oasis; the same can be said about the many replicas of the Thousand Buddha scheme that cover the whole surface of the wall or are restricted to the upper part.
After all, the Sino-French archaeologists themselves admit, on the one hand, that the sources inspiring the Karadong murals are definitely different from those of the (supposed coeval) paintings of Miran, acknowledging, furthermore, the analogies between the Karadong murals and later Khotanese wall painting.44 Yet, they prefer an earlier date because of specific iconographic features (e.g. the ‘Lateran Sophocles’ attitude) and Classical ornamental motifs (mainly the meander); but these, as I demonstrate, are unsuitable for proving an early date.
Among the results of this research, I would also highlight the hints of an artistic connection between Khotan and Kuča, a largely unexplored topic; whether it should be explained with the activity of itinerant artists or with the circulation of sketch albums or other media, this is surely an issue deserving further investigation.
Summing up, the evidence discussed so far casts doubt on the chronology proposed for the Karadong paintings, suggesting, instead, a terminus post quem in approximately the 6th century. With its idiosyncratic style and elements of formal affinity with the art of Kuča, Karadong may represent an early stage in the flourishing of Khotanese painting, with which, however, it definitely shares the iconographic orientation.
Sir Aurel Stein, Ancient Khotan: Detailed Report of Archaeological Explorations in Chinese Turkestan, 2 vols. (Oxford: Clarendon Press, 1907); Sir Aurel Stein, Serindia: Detailed Report of Explorations in Central Asia and Westernmost China, 5 vols. (Oxford: Clarendon Press, 1921); Sir Aurel Stein, Innermost Asia: Detailed Report of Explorations in Central Asia, Kan-Su, and Eastern Īrān, 4 vols. (Oxford: Clarendon Press, 1928).
Joanna Williams, “The Iconography of Khotanese Painting,” East and West 23 (1973): 109–154.
Gerd Gropp, Archäologische Funde aus Khotan, Chinesisch-Ostturkestan: Die Trinkler- Sammlung im Übersee-Museum Bremen (Bremen: Verlag Friedrich Röver, 1974), 71–200.
See Christoph Baumer, “Sogdian or Indian Iconography and Religious Influences in Dandan-Uiliq: The Murals of Buddhist Temple D 13,” in The Art of Central Asia and the Indian Subcontinent, ed. Anupa Pande (New Delhi: Aryan Book International, 2009), 170–184, with references to previous publications on his fieldwork; for the Sino-Japanese diggings at Dandān-öilïq, see Zhongguo Xinjiang wenwu kaogu yanjiusuo
See Baumer, “Sogdian or Indian Iconography and Religious Influences in Dandan Uiliq”. For a discussion of the paintings found by Baumer, see Ciro Lo Muzio, “Skanda and the Mothers in Khotanese Buddhist Painting,” in Interaction in the Himalayas and Central Asia: Processes of Transfer, Translation and Transformation in Art, Archaeology, Religion and Polity, ed. Eva Allinger, Frantz Grenet, Christian Jahoda, Maria-Katharina Lang, and Anne Vergati (Vienna: Österreichische Akademie der Wissenschaften, 2017), 71–89; Ciro Lo Muzio, “Brahmanical Deities in Foreign Lands: The Fate of Skanda in Buddhist Central Asia,” BuddhistRoad Paper 6.1, Special Issue: Central Asian Networks. Rethinking the Interplay of Religions, Art and Politics Across the Tarim Basin (5th–10th c.), ed. Erika Forte (2019): 8–43.
For a discussion of the vicissitudes of Khotanese antiquities, see Daniel Waugh and Ursula Sims-Williams, “The Old Curiosity Shop in Khotan,” The Silk Road 8 (2010): 69–96.
See, for example, Gropp, Archäologische Funde aus Khotan, figs. 43 a–d, 46 a–d, 48 a–c, 50 a–c, 51 a–c, 52 a–c, 53 a–f, 54 a–c, 56 b; Corinne Debaine-Francfort and Idriss Abduressul, Kériya, mémoires d’un fleuve. Archéologie et civilisation des oasis du Takla Makan (Paris: Éditions Findakly, 2001), 116–117.
Williams, “The Iconography of Khotanese Painting,” 110, fig. 45.
The question is effectively summarised in Williams, “The Iconography of Khotanese Painting,” 109–112.
Stein, Ancient Khotan, 266–277, 525–533.
Williams, “The Iconography of Khotanese Painting,” 110. Roderick Whitfield proposes dating the Khotanese wooden panels to the 6th century, but not the ones from Endere, which he dates to the 7th–8th centuries, thus dissociating them from the wall paintings, which sounds a false note: murals and votive panels clearly belong to the same period, whatever that period is. Roderick Whitfield, The Art of Central Asia: The Stein Collection in the British Museum, 3. Textiles, Sculpture and Other Arts (Tokyo: Kodansha, 1985), 314–317, 311, pls. 57, 66–72.
Stein, Serindia, chap. 31, section 2.
On Endere, see Stein, Ancient Khotan, chap. 12.
Stein, Ancient Khotan, pl. 10, 430ff.
Henri-Paul Francfort and Corinne Debaine-Francfort, “Oasis irriguée et art bouddhique ancien à Karadong: premiers résultats de l’expédition franco-chinoise de la Keriya, Xinjiang, République Populaire de Chine,” Comptes rendus des séances de l’Academie des Inscriptions et Belle-Lettres 137.4 (1993): 929; Corinne Debaine-Francfort, Abduressul Idriss, and Wang Binghua, “Agriculture irriguée et art bouddhique ancien au cœur du Takla Makan (Karadong, Xinjiang, IIe–IVe siècles),” Arts Asiatiques 49 (1994): 34–52; Debaine-Francfort and Idriss, Kériya, 82–113.
On this occasion, I will make no more than a cursory mention of the mural fragments, portraying haloed amorini supporting a garland, reportedly found in the Domoko area, see Shanghai bowuguan, Silu fanxiang, 118–127. The fragments have not been properly described and analyzed yet; a date in the 3rd century is inferred from the assumption that their direct source cannot be other than Gandharan art. These fragments require adequate analysis.
Stein, Ancient Khotan, 443–452.
Francfort, Debaine-Francfort, “Oasis irriguée et art bouddhique ancien à Karadong;” Debaine-Francfort, Idriss, and Wang, “Agriculture irriguée et art bouddhique ancien,” fn. 11.
Debaine-Francfort, Idriss, and Wang “Agriculture irrigué et art bouddhique ancient,” 44–45, figs. 23, 24.
According to the authors, these are pīpal leaves, although they themselves admit, “On ne connait pour l’instant pas d’équivalent aux feuilles de pipal peintes en noir sur le fond blanc, car en général, c’est l’arbre tout entier qui est représenté.” Debaine-Francfort and Idriss, Kériya, 93.
Debaine-Francfort and Idriss, Kériya, 82: “Datés par le radiocarbone de la 1re moitié du IIIe siècle.”
An early date for the Karadong murals was accepted by Marylin M. Rhie, who first deemed the late 4th century the most plausible hypothesis (Marylin M. Rhie, Early Buddhist Art of China and Central Asia. Volume One: Later Han, Three Kingdoms and Western Chin in China and Bactria to Shan-shan in Central Asia (Leiden, Boston: Brill, 1999), 318–321, 378, 390–391, 426), whereas in a later publication, she repeatedly refers to the Karadong painting as a firm chronological landmark, preferring this time the 2nd to 4th centuries, in her analysis of Binglingsi (
Debaine-Francfort and Idriss, Kériya, 82: “Datés par le radiocarbone de la 1re moitié du IIIe siècle,” 76–77.
Mario Bussagli, Central Asian Painting. From Afghanistan to Sinkiang (Geneva: Skira, 1979), 31–32.
Isabella Nardi, The Theory of Citrasūtras in Indian Painting: A Critical Re-Evaluation of their Uses and Interpretations (Abingdon: Routledge, 2006), 88–89.
Debaine-Francfort and Idriss, Kériya, 93.
Gropp, Archäologische Funde aus Khotan, 65–68, in particular 68 [inv. A.8.2.3], fig. 24.
Sophie Desrosiers and Corinne Debaine-Francfort, “On Textile Fragments Found at Karadong, a 3rd to early 4th Century Oasis in the Taklamakan Desert (Xinjiang, China),” Textile Society of America Symposium Proceedings 958 (2016): 73–74, fig. 7.
Gropp, Archäologische Funde aus Khotan, 107–109, figs. 40–41.
“[…] le style des peintures de Karadong paraît très indianisé, et l’on y relève, bien moins que dans l’art du Gandhāra, le mélange de motifs empruntés à l’Inde ancienne et d’éléments de décor hérités de l’Occident (bandeaux de separation entre deux registres notamment).” Debaine-Francfort, Idriss, and Wang, “Agriculture irrigué et art bouddhique ancient,” 48.
Debaine-Francfort and Idriss, Kériya, 104 (cat. no. 38).
On the spread of the meander in India, see Maurizio Taddei, “Greek Fretwork in India: A Significant Presence,” in South Asian Archaeology 1995, ed. Raymond and Bridget Allchin (Cambridge, New Delhi: Ancient India and Iran Trust, Oxford and IBH Pub, 1997), 563–571.
Benoy Behl, The Ajanta Caves: Ancient Paintings of Buddhist India (New York: Thames and Hudson, 2005), 110–111 (Cave 1), 138 (Cave 2).
Sir John Marshall, et al., The Bagh Caves in the Gwalior State (London: The India Society, 1927), pl. 17.
Ram Chandra Kak, Ancient Monuments of Kashmir (London: The India Society, 1933), pls. XXVIII 13, 15, XLII 46.
Taddei, “Greek Fretwork in India,” 569.
Gropp, Archäologische Funde aus Khotan, 152, figs. 55d–e, pl. 10.
Stein, Ancient Khotan, pl. 10.
Xinjiang Weiwuer zizhiqu wenwu guanli weiyuanhui
Albert Grünwedel, Alt-Kutscha (Berlin: Otto Elsner Verlagsgesellschaft M.B.H., 1920), Taf. 42, 43.
Gerard Brett, “The Mosaic of the Great Palace in Constantinople,” Journal of the Warburg and Courtauld Institutes 5 (1942): pls. 7a, 8b, 11a.
Ken R. Dark, “Roman Architecture in the Great Palace of the Byzantine Emperors at Constantinople During the Sixth to Ninth Centuries,” Byzantion 77 (2007): 90ff.
Vasilij A. Šiškin, Varachša (Moskow: Izdatel´stvo Akademii Nauk, 1963).
Debaine-Francfort and Idriss, Kériya, 90.