1 Introduction1
Uyghurs established the West Uyghur Kingdom around the Turfan area in the second half of the 9th century. Buddhism spread gradually there, and in the second half of the 10th century or the beginning of the 11th century, the majority of the Uyghurs had probably become Buddhists. This religious condition did not change until the end of the Mongolian period (1368).2 Scholars point out that the local Buddhist inhabitants, Chinese and Tocharian, played a significant role in the introduction of Buddhism to the Uyghurs.3 By degrees, however, Chinese Buddhism exerted a substantial impact on the Uyghurs and served as the main models for the translation of Buddhist texts into Old Uyghur.4 After the establishment of the Mongol Empire (13th/14th c.), as the Great Khans’ vassals, the Uyghurs expanded their sphere of activity to China and other territories of the Empire. Thus, they came into contact with other Buddhist communities like the Tibetans, through which Uyghur Buddhists gained further impact.
Dunhuang (
These conditions characterise the unique position of Dunhuang in terms of the Uyghur Buddhist community and its culture in Turfan during the pre-Mongolian period. The sources are, however, too fragmentary to provide the details. In addition, the majority of the written sources in Old Uyghur are sūtras, which do not give an account of daily religious practices or rituals. Thus, an essential question—whether the Uyghur Buddhist community entirely adopted the Buddhist trends and practices in Dunhuang at that time or made a choice to cultivate their own—is still debatable. Answering this question first requires clarifying the Buddhist trends and practices in both regions through examining the written sources and artistic objects. Then, the differences between the two regions can be assessed. Because of the paucity of available data in Old Uyghur, our discussion has to be developed around the Buddhist trends and practices in Dunhuang, where the most abundant materials in Central Asia have been recovered. This paper takes the cult of the Bodhisattva Avalokiteśvara as a case study since this famous bodhisattva was popular in Dunhuang from the 6th to the 11th centuries and has been worshipped in various forms, including the esoteric ones.8
2 Doctrinal Written Sources Supporting the Avalokiteśvara Cult Preserved in Old Uyghur
Before discussing the Avalokiteśvara cult in Turfan and Dunhuang, we should examine whether the relevant doctrinal written sources for that cult were in Old Uyghur in the pre-Mongolian period. One of the essential difficulties with the materials in Old Uyghur is dating. On the one hand, the sūtras were very probably translated in the pre-Mongolian or the Mongolian period, if their source language was Tocharian (pre-Mongolian period) or Tibetan (Mongolian period). On the other hand, for those translated from Chinese, the date of the extant copy can be suggested sometimes. The possibility, whether that copy was precisely the first translation or was made much later, remains debatable. Therefore, for the texts which were translated from Chinese, their first translation date has to be discussed first.9 At present, the following texts connected with the Bodhisattva Avalokiteśvara have been identified:10
Prajñāpāramitāhṛdayasūtra (T. 251–256.8)
The 25th chapter of Saddharmapuṇdarīkasūtra [Lotus Sūtra] (T. 262.9)11
Nīlakaṇṭhakasūtra (T. 1057.20/T. 1060.20)
Padmacintāmaṇisūtra (T. 1082.20)
Foding xin da tuoluoni jing
佛頂心大陀羅尼經 [Great Dhāraṇīsūtra of the Heart of the Buddha’s Crest]Cuṇḍīdevīdhāraṇī12
For the first five texts, Chinese versions probably served as the source. The origin of the last text is still under debate, although some similarities to the Chinese version of the Buddhabhāṣitasaptakoṭībuddhamātṛkacuṇḍīmahāvidyādhāraṇīsūtra (T. 1075.20) have been pointed out.13
Among these texts, the translations of the Nīlakaṇṭhakasūtra and the 25th chapter of the Saddharmapuṇdarīkasūtra [Lotus Sūtra], which became famous as an independent sūtra under the title Avalokiteśvarasūtra, date to the pre-Mongolian period. The former was translated by the famous translator Šiŋko Šäli Tutuŋ (fl. second half of 10th c./beginning of 11th c.).14 Kudara Kōgi and Klaus Röhrborn point out that his education was closely connected with Chinese Buddhist schools in Dunhuang.15 The 25th chapter of the Saddharmapuṇdarīkasūtra in Old Uyghur was identified among the fragments found in Cave 17 in Dunhuang, which was closed at the beginning of the 11th century. Thus, the translation must have been carried out before that period.16 Peter Zieme points out that illustrations in one booklet of that sūtra in Old Uyghur correspond exactly to those in a Chinese booklet found in Dunhuang.17 Some manuscripts of it in Old Uyghur, therefore, seem to have been produced in a close relationship with developments in Dunhuang.
The extant fragments of the other texts do not provide any information on their translation process. One published fragment of the Padmacintāmaṇisūtra is written in Uyghur square script, which does not indicate any date.18 On the other hand, the remaining fragments of the Prajñāpāramitāhṛdayasūtra, the Great Dhāraṇīsūtra of the Heart of the Buddha’s Crest,19 are written in cursive script or were block printed, which is one of the relevant features that dates to the Mongolian period. The manuscripts of the Cuṇḍīdevīdhāraṇī show the characteristics of the half-cursive or cursive script.20 Thus, at least indicating their popularity in the Mongolian period.
Those texts indicate that, in the pre-Mongolian period, some doctrinal texts connected with the Bodhisattva Avalokiteśvara were translated into Old Uyghur.21 It means that the Uyghurs could have access to this bodhisattva’s dogmatic background in their language.
In addition to those canonical texts, other sources that can indicate the transmission of the Avalokiteśvara cult among the Uyghurs with the doctrinal contents are some eulogies dedicated to the bodhisattva. The composition of at least two of these eulogies—one for Sahasrabhujasahasranetra and one for Avalokiteśvara—can likely be dated to the pre-Mongolian period, according to features of their extant manuscripts.22 Both eulogies are written in Uyghur square script, which does not provide any datable information. The first one is written in alliterative verse, the use of which became widespread among Uyghur Buddhists in the Mongolian period. If it had been composed in the earlier period, this verse would be counted as one of the earliest Buddhist alliterative verses.23 To the second eulogie, a colophon which is written in half-square script is added. Thus, for the second one, the possible composition in the pre-Mongolian period is supposed.24
Besides those texts, however, not many sources that show the practice of the Avalokiteśvara cult in Turfan are survived.25 This fragmentary condition of sources causes the difficulties to get a reasonable prospect: how that cult was practised there and whether it differed from that in the neighbouring oasis Dunhuang. In contrast, from Dunhuang, numerous written and artistic sources were found which provide rich information for the practice of that cult. Thus, the following section first discusses how the cult of Avalokiteśvara was practised in Dunhuang for preparing the discussion on the case in Turfan.
3 The Cult of Avalokiteśvara in Dunhuang
3.1 Practice of the Avalokiteśvara Cult in Prayer Texts
In Dunhuang, Avalokiteśvara was probably one of the most popular bodhisattvas. Besides the so-called Avalokiteśvarasūtra, the Nīlakaṇṭhakasūtra, which was translated into Old Uyghur in the pre-Mongolian period, seems to have also been widespread in Dunhuang.26 A significant number of copies have been found so that the dogmatic ground was well prepared for the Avalokiteśvara cult in that oasis.27
What kinds of benefits did the Buddhists in Dunhuang expect from Avalokiteśvara in practice? On this question, not the sūtras but the texts in which the Buddhist worship practice is reflected provide useful information: the Buddhist prayer texts.28 Each Buddhists wrote the prayer text at various events and expressed their wishes mentioning different buddhas and bodhisattvas. Compared to major buddhas like Amitābha or Maitreya, Avalokiteśvara is not frequently mentioned in the prayer texts. These buddhas’ popularity in prayer texts probably depends on their particular characteristics. If a deceased person meets with one of these buddhas, then they can be reborn in their respective heaven or reach buddhahood. An encounter with Avalokiteśvara, in contrast, does not promise immediate rebirth in any heaven because this bodhisattva does not have his heaven. Still, there are some prayer texts that mention this bodhisattva. In these texts, in which devotees pray for the healing of diseases, Avalokiteśvara often appears together with the bodhisattvas Bhaiṣajyarāja, Bhaiṣajya-samudgata, and Gadgadasvara.29 The healing of diseases is one of the well-known functions of the Bodhisattva Avalokiteśvara, which was explained in various texts and taken as the topic of the mural paintings. Most of these prayer texts do not contain any information on the date they were composed, but a few mention the governor (Chin. jiedu
His function as a saviour is seen in another prayer text, P. 2055, which was written in 958 by the devotee Zhai Fengda (ca. 881–959,
The merit from copying the scriptures itemed above is dedicated as a posthumous blessing to the departed, Mrs. Ma. We respectfully invite dragons, gods, and the eight classes of beings; Kuan-shi-yin (Skt. Avalokiteśvara: author) Bodhisattva, Ti-tsang (Skt. Kṣitigarbha: author) Bodhisattva, the four great kings of heaven, and the Eight Chin-kang to authenticate it. May she receive every bit of the field of blessings, be reborn in a happy place, and encounter good people. Offered with a single mind.31
Here, the type of suffering is not precisely defined. Avalokiteśvara, the other Buddhist guardians, and deities are invoked in hopes of benefiting the late person through the merits of the devotee’s good action in copying the sūtras.32
Furthermore, the other function of this bodhisattva appears in the manuscript that is now preserved in Paris as P. 2864. The manuscript contains the text written for the anniversary of the death of Empress Wang (?–845,
The Bodhisattva Avalokiteśvara has close ties to the Buddha Amitābha and is recognised as the latter’s successor.35 Thus, it is logical that Avalokiteśvara and Mahāsthāmaprāpta, Amitābha’s other primary attendant, guide the late person who will see Amitābha in Sukhāvatī. Those bodhisattvas are also mentioned in another text in which both buddhas, Amitābha and Maitreya, appear.36 As will be discussed later, Avalokiteśvara, as the bodhisattva ‘sentient beings’ guide’, is widespread in Dunhuang and is also a motif in paintings in the late Tang Dynasty (875–907,
These prayer texts prove that the Buddhists in Dunhuang under Guiyijun rule venerated Avalokiteśvara, because this particular bodhisattva carries out two relevant functions: a saviour who cures illness and relieves suffering, and sentient beings’ guide to Sukhāvatī.
3.2 Avalokiteśvara in Artistic Sources
Many wall paintings and paintings on silk adopted Avalokiteśvara as their primary subject or as a buddha’s attendant.38 Some of these representations provide inscriptions recording the names of the donors and the reasons for their donation; the foremost was a good rebirth.39
He is portrayed in various forms, one of which is as the bodhisattva ‘sentient beings’ guide’ mentioned above. In this form, this bodhisattva leads deceased persons to the Paradise of the Buddha Amitābha, namely, Sukhāvatī, the Western Pure Land. In the Dunhuang area, this form probably developed from the second half of the 9th century onward.40 Another form of Avalokiteśvara that was popular in Dunhuang during the same period was the Water–Moon Avalokiteśvara (Chin. Shuiyue Guanyin
Furthermore, some of the other forms attested among those paintings belong to Esoteric Buddhism.42 In this context, it is worth mentioning the popularity of Sahasrabhujasahasranetra, that is, the Thousand-armed and Thousand-eyed Avalokiteśvara.43 This form of Avalokiteśvara appears in many wall paintings, and this bodhisattva’s numerous entourage is presented in various ways. Hamada Tamami investigates this entourage in wall paintings in Dunhuang and points out different entourages from the period of the Tang to that of the Five Dynasties.44 The paintings from the Tang period show either eight great bodhisattvas or twelve celestials, both groups on lotuses and surrounding Sahasrabhujasahasranetra. In later paintings from the Five Dynasties period, the bodhisattva’s companions are depicted as deities flying on clouds. This new entourage is mentioned in the Nīlakaṇṭhakasūtra. Hamada suggests that this change was probably connected with the increasing popularity of the Dabei zhou
The textual and iconographic sources in Dunhuang under Guiyijun rule indicate that people worshipped Avalokiteśvara for the sake of healing their ailments. This bodhisattva guides beings to the Sukhāvatī and serves as a saviour for those who seek a good rebirth. Sahasrabhujasahasranetra was a favorite painting motif in Dunhuang. Changes in its iconography were probably associated with the prevalence of the Great Compassion dhāraṇī, which was recited in Buddhist assemblies from the 10th century onward.
4 Avalokiteśvara Cult in Turfan
4.1 Practice of the Avalokiteśvara Cult Traced in Written Sources
In contrast to the Chinese sources in Dunhuang, the written sources in Old Uyghur from Turfan do not show that Avalokiteśvara was worshipped as the saviour. As mentioned above, the sources that show us the Avalokiteśvara cult’s practice in Turfan are generally few. The prayer texts are found neither in Old Uyghur nor in Chinese from Turfan.46 Therefore, we have no way of knowing whether or how the Uyghur devotees expressed their aspirations to the Bodhisattva Avalokiteśvara. Colophons and inscriptions added at the end of copied sūtras or on the side of pictures on banners sometimes contain some names of buddhas, bodhisattvas, or deities to which donors’ aspirations are addressed.47 In those colophons and inscriptions, however, the Buddha (or Bodhisattva) Maitreya is mentioned most frequently. The Uyghur donors in those sources aspired to rebirth in Tuṣita heaven, to meet with Maitreya, or to get assurance of future enlightenment from that buddha.48
In this context, it is worth mentioning the fact that Buddha Amitābha does not appear at the same frequency as Maitreya. As mentioned above, in Dunhuang documents, Amitābha is closely linked with Avalokiteśvara, who is recognised as the guide to the buddha’s paradise, Sukhāvatī. This buddha was the object of worship for Uyghur donors in six colophons. Four of them are block printed or written in a cursive script, which probably date to the Mongolian period. The other two are written in square script, providing no information that enables the dating of the documents.49 The scriptural sources for the Amitābha cult were probably already available in Old Uyghur in the pre-Mongolian period, and some Uyghurs might have engaged in practicing that cult. However, the extant materials, including colophons, establish the prevalence of this cult among Uyghurs in the Mongolian period.50 The lack of material makes it difficult to determine how widespread the Amitābha cult was among the Uyghurs in the pre-Mongolian period. In Dunhuang, the Buddha Amitābha and Maitreya are often addressed side by side in prayers texts as the recipients of devotees’ devotions.51 Despite the differences between prayer texts and colophons, Amitābha’s paradise of Sukhāvatī may have been a goal for Buddhist donors or their dead relatives, exactly like Maitreya’s Tuṣita heaven. In that respect, it is no wonder that Amitābha appears beside Maitreya in the part of the colophons expressing the Uyghur donors’ aspirations. Considering Amitābha’s relative absence in the Uyghur colophons, this buddha does not seem to have gained the same popularity as Maitreya among the Uyghurs, or perhaps, he and his functions were possibly envisioned differently in Turfan than in Dunhuang.52
Remarkably, even in the texts in which the names Avalokiteśvara or Amitābha can be expected, this bodhisattva and buddha are not mentioned. In the colophon which was added to the above-mentioned eulogy dedicated to Avalokiteśvara,53 the Uyghur donors wish to meet the Buddha Maitreya and receive that buddha’s assurance of their future enlightenment.54 A similar case is also found in the inscription added to a banner III 533. Although the fragmentary condition does not allow to identify the topics of this banner, according to the inscription, Cakravarticintāmaṇicakra, a form of Avalokiteśvara, and possibly of Maitreya are depicted on it. There, the Uyghur devotees wish that: “all of us shall be joyful, and free from illness and disease, from pain and danger in this present world.”55 At last they shall:
be reborn above in the Tuṣita heaven. And let us see in the future time Maitreya Buddha by the strength of this meritorious and good deed. [Let us decorate] Maitreya Buddha’s graceful body with the Jambu-river gold (= Jāmbūnada-suvarṇa).56
The first part could possibly indicate that the Uyghur devotees were aware of Avalokiteśvara’s function as the saviour of from disease. As discussed in Section 3.1. above, this was one of Avalokiteśvara’s major functions, which motivated devotees to pray to the bodhisattva in Dunhuang. On the other hand, it is not the Buddha Amitābha but rather Maitreya who is mentioned in that banner inscription. Thus, the close link with Amitābha attested in both written and artistic sources in Dunhuang does not necessarily seem to have been recognised among the Uyghur Buddhists.
The extant written sources in Old Uyghur do not explicitly attest to Avalokiteśvara’s primary function as a saviour from disease.57 While the written sources found in Dunhuang attest to the close connection between Avalokiteśvara and Amitābha, none of the sources in Old Uyghur provide traces of this connection’s recognition by the Uyghur Buddhists. These findings are also supported by the fact that the Uyghur devotees did not address Avalokiteśvara when praying for a good rebirth.
4.2 Avalokiteśvara in Artistic Sources
The artistic sources found in Turfan are much less systematically analysed than those of Dunhuang. Still, some paintings found on walls and textiles provide clues to the iconographic trends among the Uyghur Buddhists. Bezeklik Cave 20 has drawn much attention for its wall paintings depicting Bodhisattva Avalokiteśvara. In this cave, Sahasrabhujasahasranetra has been identified as the central figure on the back wall of the cella.58 Several scenes, the topic of which is the Buddha Śākyamuni’s vow (Skt. praṇidhi) in his former lives and his getting the assurance of further enlightenment, are depicted on the walls in the ambulatory. On each side of the entrance, three monks are pictured. The monks’ clothes and associated inscriptions identify them as Chinese and Tocharians.59 This cave also contains Uyghur donor figures on the interior east (front) wall of the cella. From their clothes and ornaments, they are probably Uyghur princesses and princes.60 Thus, it is one of the caves that was established under the patronage of high-ranking Uyghur Buddhists, and in it, Sahasrabhujasahasranetra was chosen for the main iconographic programme.
Otherwise, Avalokiteśvara does not enjoy the predominant presence in the wall paintings. In contrast, other art objects provide this bodhisattva’s prevalence in Turfan: Buddhist banners. Chhaya Bhattacharya-Haesner made the catalogue of the temple banners preserved in the Berlin Turfan Collection, one of the most important collections of art objects and written sources from Turfan.61 In her catalogue, 148 entries altogether are devoted to banners depicting bodhisattvas.62 Among them, the bodhisattva is unidentified in 42 of the entries. The number of banners depicting the various forms of the Bodhisattva Avalokiteśvara amounts to 80. The number of banners depicting other bodhisattvas—Maitreya, Mahāsthāmaprāpta, Samantabhadra (Chin. Puxian
The total number in both tables indicates that the Bodhisattva Avalokiteśvara was a favourite subject for temple banners throughout the 7th and 12th centuries. The banners devoted to the Water–Moon Avalokiteśvara indicate the exchange with the Buddhist community in Dunhuang, where that form was popular.65 The same is true for the increasing number of banners devoted to Sahasrabhujasahasranetra. The Invocation of the Great Compassionate One, which was connected with that form in Dunhuang, has not, however, been identified yet in Old Uyghur.
Avalokiteśvara banners in the Berlin Turfan Collection currently dated 7th to 10th centuries
Avalokiteśvara banners in the Berlin Turfan Collection currently dated 10th to 12th centuries
Among them, only a few banners bear an inscription, so the process of making the banners remains mostly unknown. However, if we assume that locals in Turfan donated most of the banners, then the donors were probably the Buddhists, most of whom were perhaps non-Uyghur speakers, especially before the Uyghurs’ migration.66 These donors were probably Chinese rather than Tocharians—although the latter were also an essential component of the Turfan area—because of the flourishing Avalokiteśvara cult in the neighbouring Chinese Buddhist community in Dunhuang. The inscriptions on three of the banners, III 6588, III 6458, and III 6564 (cat. nos. 178, 183, and 191), are written in Chinese, and according to Bhattacharya-Haesne, were produced in that 7th–10th-centuries period. If her dating is correct, then the Chinese inscriptions also indicate the involvement of Chinese Buddhists.67
The Chinese ambassador from the Song Dynasty, Wang Yande (939–1006,
Should we then think that the Uyghurs donated most of the banners from the 10th–12th century and that these banners precisely reflect their Buddhist activities? Two of the banners, III 533 and III 8559 (cat. nos. 202 and 246), were clearly donated by Uyghur Buddhists, because their inscriptions are in Old Uyghur.70 However, the donors of most of the banners remain unidentified. The Uyghurs’ conversion to Buddhism did not mean that the Chinese Buddhist community’s activities came to a standstill, nor that they were immediately assimilated to the Uyghurs. Thus, both Uyghurs and Chinese should be considered as the donors of those banners.71 The support of the Uyghurs as the new ruling class in the Turfan area might even have been an impetus to local Buddhist activities. On banner III 7513 (cat. no. 179), which Bhattacharya-Haesner dates to the 10th–11th centuries, the cartouche is written with Chinese characters.72 If we accept her dating, the use of Chinese characters could indicate that the process of making some Avalokiteśvara banners remained in close relationship with the Chinese tradition at that time.
5 Closing Remarks
As discussed in Section 2, doctrinal written sources related to Avalokiteśvara were probably translated into Old Uyghur absorbing Chinese Buddhism in Dunhuang in the pre-Mongolian period. The other textual sources in Old Uyghur that are dealt with in Section 4.1 show different features than those from Dunhuang. In those sources, the Uyghur devotees did not choose the Bodhisattva Avalokiteśvara as the recipient of their devotions. Furthermore, neither his close connection to the Buddha Amitābha nor his role as the guide to the latter’s paradise, Sukhāvatī, appears. The complete absence of this latter role in the colophons is especially relevant. This role is strongly connected with the afterworld, such that—like Maitreya and his paradise—it is a topic that should be addressed in the colophons, because they involve donors’ devotions for themselves and their late relatives. The paintings on the banner discussed in Section 4.2 show that Avalokiteśvara was continuously one of the favourite motifs before and after Buddhism became the dominant religion of the Uyghurs when they became the ruling class in Turfan after the second half of the 9th century. Thus, during the 7th–12th centuries, those Avalokiteśvara banners were probably donated by both the Chinese and Uyghur Buddhists in Turfan. The Uyghurs were involved in the donation mainly from the second half of the 10th century onward, while the Chinese acted as donor throughout the whole period. The popularity of that bodhisattva in paintings and the bodhisattva’s absence in textual sources in Old Uyghur outside of sūtras, leads us to the position that the Avalokiteśvara cult was practised among the Chinese Buddhist community and the Uyghur Buddhists who had a close relationship with it.
In this context, it is worth reexamining the wall painting in Bezeklik Cave 20. As mentioned in Section 4.2, it was established by the donation of high-ranking Uyghurs. The iconography selected, the Sahasrabhujasahasranetra and vow (Skt. praṇidhi) scenes, reflect the presence of both Chinese and Tocharian traditions.73 At the entrance of the ambulatory, the Chinese and Tocharian monks are depicted in threes on each side. The whole iconographic programme in this cave seems to have been chosen with careful consideration for both the Chinese and the Tocharian Buddhist traditions that existed in the West Uyghur Kingdom. To give equal acknowledgement to both Buddhist traditions in their kingdom seems to have been relevant for the Uyghurs as rulers. For example, not only the Chinese but also the Tocharian monks were appointed to the high-ranking position of having the gold seal bestowed on them.74 Even in the period when the Manichaean and Buddhist communities still co-existed, both communities received official financial support.75 It seems, therefore, to have been an essential political issue for the Uyghur rulers to maintain a balance between the various religious communities in their kingdom.
Bezeklik Cave 20 was sponsored by the high-ranking Uyghurs who belonged to the ruling clans. Thus, its iconographic programme can be interpreted as a representation of the political treatment of the religious communities in the West Uyghur Kingdom. The Buddhist caves were not only a place of worship but also of political demonstration, as some of Mogao and Yulin Caves (Chin. Yulin ku
If this is the case, Sahasrabhujasahasranetra was not necessarily chosen because of the popularity of that bodhisattva among the Uyghurs at that time, or because the Uyghur donors of Cave 20 might have worshipped Avalokiteśvara privately. Instead, Avalokiteśvara was recognised as one of the essential figures worshipped in the Chinese Buddhist community. The same was true for the vow scenes and the Tocharian community. The Uyghur Buddhists absorbed Buddhist teaching and cultures from these two major Buddhist communities in the West Uyghur Kingdom. They accepted the various cults and practices, while the Chinese and Tocharian Buddhists also continuously engaged in Buddhist activities. The cult of Avalokiteśvara, which was a result of absorption from Chinese Buddhism, was one of those cults. The Uyghurs adopted it through the Chinese community in Turfan. Although the Chinese Buddhist contribution from Dunhuang to the Uyghurs was significant, the adoption of a cult or trend from that area does not necessarily mean that it became popular for all Uyghur Buddhists or was practiced exactly in the same way in Turfan as in Dunhuang. In the West Uyghur Kingdom, various Buddhist communities—commencing with those established by the Tocharians and Chinese—seem to have existed, and cultivated and retained their traditions. The Uyghurs had different relationships with those communities so that their impacts were varied.78 As mentioned above, the Chinese Buddhist elements from Dunhuang was probably the strongest ones that the majority of the Uyghurs absorbed. This, however, does not mean that all Uyghurs uniformly followed the same Chinese Buddhist practices. There were various Buddhist groups and communities that contained different language speakers and Buddhist traditions. The cult of Avalokiteśvara was one of them. The activities of those various communities are reflected in the extant textual and artistic sources in Turfan.
I would like to express my special thanks to Dr. Miki Morita (Iwakuni) and Dr. Hou Haoran (Beijing) who kindly gave me their support as specialists in art history and Tibetan Buddhism respectively. I, of course, alone am responsible for my mistakes.
For the establishment of the West Uyghur Kingdom, see, e.g., Denis Sinor et al., “The Uighurs, the Kyrgiz and the Tangut (Eight to the Thirteenth Century),” in The Age of Achievement: AD 750 to the End of the Fifteenth Century, ed. Muhammad Seyfeydinovich Asimov and Clifford Edmund Bosworth (Paris: UNESCO Publishing, 1998), 200–206; Moriyasu Takao
For the Uyghurs’ conversion in the Turfan area, see, e.g., Moriyasu Takao
The majority of the Old Uyghur Buddhist texts is translations from other languages. Johan Elverskog gives an overview of those texts. See Johan Elverskog, Uyghur Buddhist Literature (Turnhout: Brepols, 1997).
For the close relationship between these two areas, see, e.g., Takao Moriyasu, “The Sa-chou Uighurs and the West Uighur Kingdom,” Acta Asiatica 78 (2000): 28–48; Moriyasu Takao
See, e.g., Takao Moriyasu, “The Sha-chou Uighurs,” 28–48.
See Section 2 of this chapter.
See, e.g., Miyeko Murase, “Kuan-Yin as Savior of Men: Illustration of the Twenty-Fifth Chapter of the Lotus Sūtra in Chinese Paintings,” Artibus Asiae 33.1–2 (1971): 42–43. On the whole Avalokiteśvara cult, see Chün-fang Yü, Kuan-yin: The Chinese Transformation of Avalokiteśvara (New York: Columbia University Press, 2001). Further studies on this topic are discussed in Section 3 of this chapter.
At least, one text can be omitted in the following discussion: Buddhāvataṃsakasūtra (T. 279.10/T. 293.10). This text was translated by the Uyghur monk known under the name Anzang (?–1293,
Jens Wilkens mentions most of those texts in his contribution in this volume. For the previous studies on those texts, see Chapter 13 in this volume. On the first two texts, see, e.g., Elverskog, Uygur Buddhist Literature, 53–54, 58, nos. 28 and 33. Furthermore, Abudurishid Yakup (Berlin) published the complete edition of the Old Uyghur version of the Buddhāvataṃsakasūtra. See, Abudurishid Yakup, Buddhāvataṃsaka Literature in Old Uyghur (Turnhout: Brepols, 2021). The seventh text is preserved in various manuscripts. Chün-fang Yü also mentions the Śūraṅgamasūtra as a relevant text for Avalokiteśvara (T. 945.19). See Chün-fang Yü, “Guanyin: The Chinese Transformation of Avalokiteshvara,” in Latter Days of the Law: Images of Chinese Buddhism 850–1850, ed. Marsha Weidner (Honolulu: University of Hawai’i Press, 1994), 152. Up to now, this text has not been identified in Old Uyghur, although some citations from it were recognised. See, e.g., Elverskog, Uygur Buddhist Literature, 147. Dhāraṇīs or amulets connected with the bodhisattva were used for the practice of the Avalokiteśvara cult. Thus, they are also one of the relevant topics. As Peter Zieme points out, those dhāraṇīs are translated into Old Uyghur and seem to have been worn as amulets. See BT XXIII, 179–189. I prepared a detailed discussion of this topic in another paper.
In association with this text, one further text called three Avadānas to Avalokiteśvarasūtra is worth mentioning. According to Shōgaito Masahiro who first worked on this text, it was recited after the recitation of the Avalokiteśvarasūtra, namely the 25th chapter of Saddharmapuṇdarīka. The extant manuscript was copied in the Mongolian period, although its original composition date stays obscure. See, e.g., Shōgaito Masahiro
Furthermore, two versions of the Avalokiteśvarasādhana and Tārāekaviṃśatistotra can be mentioned as those which deal with the Bodhisattva Avalokiteśvara. Because they were translated from Tibetan, their production was dated in the Mongolian period.
See BT XXIII, 65–79. A part of this dhāraṇī is quoted and written in Chinese on the Buddhist temple banner III 4432, which dates to the late 10th–11th century. Both Chinese and Uyghur inscriptions are written on that banner. On the inscription, see Takao Moriyasu in collaboration with Peter Zieme, “Uighur Inscriptions on the Banners from Turfan Housed in the Museum für Indische Kunst, Berlin,” in Central Asian Temple Banners in the Turfan Collection of the Museum für Indische Kunst, Berlin, ed. Chhaya Bhattacharya-Haesner (Berlin: Dietrich Reimer Verlag, 2003), 464. For a discussion of the paintings on that banner, see Chhaya Bhattacharya-Haesner, Central Asian Temple Banners in the Turfan Collection of the Museum für Indische Kunst, Berlin (Berlin: Dietrich Reimer Verlag, 2003), 130–137, cat. no. 120.
On the date of his lifetime, see James Russell Hamilton, “Les titres Šäli et Tutung en ouïgour,” Journal Asiatique 272 (1984): 435–436; Moriyasu Takao
See, e.g., Kudara Kōgi
Oda Juten intensively examined various manuscripts of this text, and discusses their dates. See Oda Juten
See Peter Zieme, “Some Notes on Old Uigur Art and Texts,” in Torufan no bukkyō to bijutsu—Uiguru bukkyō wo chūshin ni—Shiruku rōdo no bukkyō bunka—Gandāra, Kucha, Torufan—Dainibu
See Shōgaito Masahiro
See BT XXVIII, 223–234; Georg Kara and Peter Zieme, “Die uigurische Übersetzung des apokryphen Sūtras ‘Fo ding xin da tuo luo ni’,” Altorientalische Forschungen 13 (1986): 319–322.
See BT XXIII, 65–79.
In the Mongolian period, it seems that the cult of Avalokiteśvara experienced a florescence or possibly a re-florescence. In that period, the Tibetan texts also served as sources, see fn. 12.
On those two eulogies, see BT XIII, 126–130, no. 21; BT XXVI, 229–231, colophon no. 124. Johan Elverskog classifies the eulogies under various topics. See Elverskog, Uygur Buddhist Literature, 126–129. In addition to the eulogies listed, there are some additional ones dedicated to that bodhisattva. Their extant manuscripts are either written in cursive script or block printed. Those features can only indicate that the extant manuscripts were prepared in the Mongolian period. The possibility that the originals of those eulogies were composed in an earlier period and copied in a later period, is not negated. Materials to confirm their earlier composition are lacking.
On Uyghur Buddhist alliteration, see Peter Zieme, Die Stabreimtexte der Uiguren von Turfan und Dunhuang. Studien zur alttürkischen Dichtung (Budapest: Akadémiai Kiadó, 1991), 23–25. There are a few Buddhist alliterative verses that were composed in the pre-Mongolian period. See Zieme, Die Stabreimtexte der Uiguren, 292–294; Jens Peter Laut, “Gedanken zum alttürkischen Stabreim,” in Splitter aus der Gegend von Turfan. Festschrift für Peter Zieme anläßlich seines 60. Geburtstags, ed. Mehmer Ölmez and Simone-Christiane Raschmann (Istanbul, Berlin: Şafak Matbaacılık, 2002), 129–138. Compared with those alliterative verses that give us a less polished impression, the one mentioned above was probably written by a mature Buddhist poet.
See BT XXVI, 229–230.
Those texts are discussed in Section 4.1.
Dunhuang was under Tibetan rule until the middle of the 9th century. According to Sam van Schaik, the first trace of the cult of Avalokiteśvara among the Tibetan Buddhists is already seen in the Dunhuang manuscripts in the 10th century. See Sam van Schaik, “The Tibetan Avalokiteśvara Cult in the Tenth Century: Evidence from the Dunhuang Manuscripts,” in Tibetan Buddhist Literature and Praxis. Studies in its Formative Period, 900–1400. PIATS 2003: Tibetan Studies: Proceedings of the Tenth Seminar of the International Association for Tibetan Studies, Oxford, 2003, ed. Ronald M. Davidson and Christian K. Wedemeyer (Leiden, Boston: Brill, 2006), 55–72. Thus, the possibility that the cult under Chinese Guiyijun rule was partly or completely inherited from the Tibetan cult should be taken into consideration. However, the cult of Avalokiteśvara in Tibet mainly flourished in the post-imperial period, as Matthew Kapstein points out. See Matthew Kapstein, “Remarks on the Maṇi bKa’-’bum and the Cult of Avalokiteśvara in Tibet,” in Tibetan Buddhism: Reason and Revelation, ed. Steven D. Goodman and Ronald M. Davidson (Albany: State University of New York Press, 1992), 79–93. The role of the Tibetan Avalokiteśvara cult in the same cult among the Chinese Buddhists under Guiyijun rule is, therefore, still a point of debate, such that the Tibetan contribution on the Uyghur Avalokiteśvara cult in the pre-Mongolian period also remains an open question.
Hirai Yūkei
These texts were collected and edited by Huang Zheng and Wu Wei. In the following discussion, the prayer texts included in their edition are taken into consideration. See Huang Zheng
According to Huang Zheng and Wu Wei’s edition, the following fragments mention the Bodhisattva Avalokiteśvara with the bodhisattvas mentioned above: S. 1441, S. 4081, S. 4537, S. 5561, S. 6417, P. 2058/P. 3566, P. 2237, and P. 2854. See Huang and Wu, Dunhuang yuanwenji, 34, 53, 56, 172, 308, 664, 672, 674, 694, 703, 709.
See, e.g., P. 2854 and S. 4537, in Huang and Wu, Dunhuang yuanwenji, 672, 674.
P. 2055:
Zhai Fengda also ordered a drawing Avalokiteśvara as an attendant of Mañjuśrī for the wall of Mogao Cave 220. The inscription is contained in Huang and Wu’s edition and was translated into English by Wei-cheng Lin. See Huang and Wu, Dunhuang yuanwenji, 924; Wei-cheng Lin, Building a Sacred Mountain. The Buddhist Architecture of China’s Mount Wutai (Seattle, London: University of Washington Press, 2014), 173.
P. 2864:
Ibid.:
See, e.g., Yü, Kuan-yin, 32, 36.
This text forms one of many texts contained in the manuscript that is preserved in London under the signature S. 4474. Huang and Wu give the number 10 for this text. They suppose that the manuscript was written around 908. See Huang and Wu, Dunhuang yuanwenji, 183.
See Section 3.2 below.
For an overview, see, e.g., Murase, “Kuan-Yin as Savior of Men”. See also, Yü, Kuan-yin, 224–228.
Yü, Kuan-yin, 225.
See, e.g., Yü, Kuan-yin, 225–228; Li Ling
The breakdown of 32 paintings is 27 wall paintings and 5 paintings on paper and silk. See Wang Huimin
See, e.g., Henrik H. Sørensen, “Typology and Iconography in the Esoteric Buddhist Art of Dunhuang,” Silk Road Art and Archaeology 2 (1992): 302–309; Michelle C. Wang, Maṇḍalas in the Making. The Visual Culture of Esoteric Buddhism at Dunhuang (Leiden, Boston: Brill, 2018), 138–155.
For a detailed iconographic study of this form Avalokiteśvara, see, e.g. Wang Huimin
Hamada Tamami
Maria Reis-Habito investigates the repentance ritual of the Thousand-armed Avalokiteśvara, in which the Great Compassion Dhāraṇī played the central role. She suggests that this extracted dhāraṇī text probably led to the emergence of that ritual, which was composed by the Tiantai monk Zhiyi (960–1028,
Peter Zieme recently published a manuscript which he calls “vow text” because of its contents. See, Peter Zieme, “Merit Transfer and Vow according to an Old Uyghur Buddhist Text from Qočo/Gaochang,” Sōka daigaku Kokusai bukkyōgaku kōtōkenkyūjo nenpō Reiwa ninendo
There are also block-printed sūtras whose printing was donated by the Uyghur Buddhists. As mentioned above, the block-printed texts probably date to the Mongolian period. The period which is dealt with in this paper is the pre-Mongolian period, so the block-printed sources are only used secondarily.
To my knowledge, ten colophons mention the Buddha (or Bodhisattva) Maitreya, and among them, six were probably written in the pre-Mongolian period. They are listed with the corresponding quotations in my article. See Kasai Yukiyo
There are a few colophons and inscriptions in which the Uyghur donors addressed their wishes to Avalokiteśvara. The colophon U 4707 [T III M 187] and the inscription III 7307 on the banner can be mentioned as those examples. They are, however, probably written in the Mongolian period. For a detailed description of the colophon U 4707 and information on previous studies, see BT XXVI, 56–58, colophon no. 6. For the inscription III 7307, see Moriyasu and Zieme, “Uighur Inscriptions on the Banners,” 466.
These colophons have been reedited. For the text and information on previous studies of these colophons, see BT XXVI, 49–52, 65–67, 112–115, 211–212, 239–243, and 252–253; colophon nos. 3, 12, 40, 113, 129, and 135. While colophon nos. 3 and 135 are written in Uyghur square script, colophon nos. 12, 40, and 129 are block printed. The script used in colophon no. 113 is categorised as cursive. On the classification of the various Uyghur scripts and their dating possibilities, see fn. 17.
On that topic, see also, Jens Wilkens, “Practice and Rituals in Uyghur Buddhist Texts: A Preliminary Appraisal,” Chapter 13 in this volume.
One example is quoted in Section 3.1 above. For the original text, see fn. 33. Furthermore, similar sentences appear in prayer texts that Huang and Wu edited. See Huang and Wu, Dunhuang yuanwenji, 14, 28, 139, 239, 277, 214, 765, 800.
In this context, one Buddhist temple banner, III 6242 (cat. no. 496), demands special attention. See Bhattacharya-Haesner, Central Asian Temple Banners, 352, no. 496. According to Bhattacharya-Haesner, it could date in the 10th–11th centuries. On that banner, a Uyghur lady with a buddha figure in her headpiece is depicted. Zsuzsanna Gulácsi identifies that buddha figure as Amitābha and regards this banner as a trace of the Uyghurs’ adherence to Pure Land Buddhism in that early period. Furthermore, she also claims that there are some similarities between the pictorial programmes of that banner and the Manichaean banner. See Zsuzsanna Gulácsi, “The Manichaean Roots of a Pure Land Banner from Kocho (III 4524) in the Asian Art Museum, Berlin,” in Language, Society, and Religion in the World of the Turks. Festschrift for Larry Clark at Seventy-Five, ed. Zsuzsanna Gulácsi (Turnhout: Brepols, 2018), 345–350. If her assertion is correct, banner III 6242 is one of the most relevant pieces of evidence proving the prevalence of Pure Land Buddhism among the Uyghurs in the pre-Mongolian period, and the shared pictorial programme in Manichaean and Buddhist paintings provides a powerful lens through which to view the Uyghurs’ understanding the afterworld. The identification of the buddha figure with Amitābha is, however, uncertain. Also, the inscription on that banner only mentions a buddha land without further substantiating. This, therefore, requires further research.
See, Section 2.
See BT XXVI, 229–230, lines 15–18.
The English translation follows the edition Takao Moriyas worked on in collaboration with Peter Zieme. See Moriyasu and Zieme, “Uighur Inscriptions on the Banners,” 463, lines 16–18.
The English translation follows Moriyasu and Zieme’s edition. See Moriyasu and Zieme, “Uighur Inscriptions on the Banners,” 463, lines 22–26.
This raises the further question of what kind of traditions the Uyghur Buddhists had when they prayed for healing from diseases. So far as prayer texts in Old Uyghur were not found, their tradition could differ from that in Dunhuang.
About the detailed study on that cave, see, e.g. Denise P. Leidy, “Bezeklik Temple 20 and Early Esoteric Buddhism,” Silk Road Art and Archaeology 7 (2001): 201–222; Kitsudō Kōichi
Kitsudō Kōichi summarise the previous studies on the monk’s figures and their identifications. See, Kitsudō, “Bezekuriku sekkutsu kuyō bikuzu saikō,” 525–526.
About the donor figures, see, e.g. Lilla Russell-Smith, Uyghur Patronage in Dunhuang. Regional Art Centres on the Northern Silk Road in the Tenth and Eleventh Centuries (Leiden, Boston: Brill, 2005), 23–28.
Bhattacharya-Haesner, Central Asian Temple Banners.
Those entries are from the catalogue number 157 to 312. Here, only the entries in the headline of which the word ‘bodhisattva’ or the identified bodhisattvas’ name appear, are counted. It should give the first impression of how many banners are devoted to the bodhisattvas. There are further banners on which bodhisattvas are depicted as one of the component elements of the various scenes like hell or paradises.
The entries for banners of the various forms of Avalokiteśvara are catalogue nos. 168–196, 198–199, 202–249, 260. See Bhattacharya-Haesner, Central Asian Temple Banners, 169–194, 195–199, 200–231, 240–242. Furthermore, she also indicates there is a connection between the moon motif on banner 64 and Avalokiteśvara, but Jens Wilkens refutes this. See Jens Wilkens, “Review: Central Asian Temple Banners in the Turfan Collection of the Museum für Indische Kunst, Berlin by Chhaya Bhattacharya-Haesner,” Orientalistische Literaturzeitung 100 (2005): 321. She also identifies the figures on banner 113. As she discusses, however, there are multiple possible identifications, one of which is even Zoroastrian. Because of this difference in opinion among the specialists, I do not take it into consideration in this paper. See Bhattacharya-Haesner, Central Asian Temple Banners, 35–38, 124–126. On this topic, see also, Lilla Russell-Smith, “The ‘Sogdian Deities’ Twenty Years on: A Reconsideration of a Small Painting from Dunhuang,” Chapter 5 in this volume. Banner no. 260 depicts not only Avalokiteśvara but also Kṣitigarbha and Amitābha. The banners that depict Maitreya are banners 157, 160–167; Mahāsthāmaprāpta: banner 197; Samantabhadra: banner 250; Kṣitigarbha: banners 251–262; and Mañjuśrī: banners 267–270. See Bhattacharya-Haesner, Central Asian Temple Banners, 161–169, 194–195, 231–243, 246–247. Kṣitigarbha is also depicted in the scenes of hell. See Bhattacharya-Haesner, Central Asian Temple Banners, 245, banner 265. Peter Zieme mentions an additional Avalokiteśvara banner preserved in St. Petersburg. This banner bears a short inscription in Old Uyghur, which is written in cursive script. See Zieme, “Some notes on Old Uigur art and texts,” 10.
The following tables are based on Bhattacharya-Haesner’s identification and dating. Among the entries listed in n. 53, cat. nos. 235, 236, 237, 238, and 239 belong to the same banner, so they are counted as one. The same applies for nos. 216, 229–232. Bhattacharya-Haesner only dates no. 232 in the 10th–11th centuries, but here it is treated together with the others that are dated in the 11th–12th centuries. As mentioned above, some banners are too fragmented to understand her identification. Thus, those banners are not taken into consideration here. They are catalogue nos. 168, 177, 182, 190, 194, 207, 208, 214, 221, 223, 224, 226, 227, and 228. The identification of the bodhisattva figure in the banner that depicts a buddha at the top is still debated. Bhattacharya-Haesner identifies it as Avalokiteśvara, because she interprets the buddha figure at the top of the banner as Buddha Amitābha. This interpretation, however, remains debatable. Thus, the identification of nos. 180, 181, and 183 may change through future studies. The banner III 6242 (cat. no. 496), which Zsusanna Gulácsi discusses, also belongs to this category. See fn. 52. Furthermore, Bhattacharya-Haesner points out that nos. 219, 211, 212, 213, 215–218, 220–232, 234, and 243 stylistically belong together, although they are not contiguous and could date to different periods. Because of the fragmentary condition, this assumption cannot be affirmed.
Peter Zieme also discusses the image of ‘moon in the water’ in Old Uyghur texts. See Peter Zieme, “Das Bild vom ‘Mond im Wasser (水月)’ in altuigurischen Texten,” academia.edu, last accessed March 8, 2020, 1–3. https://www.academia.edu/34518806/Das_Bild_vom_Mond_im_Wasser_
Already in the East Uyghur Kaganate period, some Uyghurs were active in that area, and among them, there were probably a few who already had contact with the local Buddhists in Turfan and became Buddhists individually. Thus, the possibility that those Uyghur speaking Buddhists donated some of those banners also has to be considered. However, the Manichaean hymn book called Maḥrnāmag was produced in the Turfan area in the Kaganate period, indicating the local Manichaean communities there. This Manichaean book is written in the Middle Iranian language, was kept unfinished in the monastery of Karashar (Chin. Yanqi
As is well known, some Uyghur Buddhists could read Chinese texts and write Chinese characters. See, e.g. Kōgi Kudara and Peter Zieme, “Uigurische ‘Āgama’-Fragmente (1),” Altorientalische Forschungen 10 (1983): 271–272; Yukiyo Kasai, “Old Uyghur Translations of Budhist Texts and Their Usage,” in Premodern Translation: Comparative Approaches to Cross-Cultural Transformations. Contact and Transmission 2, ed. Sonja Brentjes and Alexander Fidora (Turnhout: Brepols, 2021), 20–23. However, it is still debatable when they began to learn Chinese and how large that group under the Uyghurs was. Many datable pieces of evidence for using Chinese texts and characters are from the Mongolian period. Thus, the absence of Old Uyghur texts cannot immediately be interpreted as that the Uyghur Buddhists used Chinese Buddhist texts in the practice of the Avalokiteśvara cult.
For the inscriptions on banners 178 and 191, see Rong Xinjiang, “Chinese Inscriptions on the Turfan Textiles in the Museum of Indian Art, Berlin,” in Central Asian Temple Banners in the Turfan Collection of the Museum für Indische Kunst, Berlin, ed. Chhaya Bhattacharya-Haesner (Berlin: Dietrich Reimer Verlag, 2003), 475. In particular, banner 178 is the invocation Avalokiteśvara and contains a Chinese quotation from the 25th chapter of the Saddharmapuṇdarīkasūtra, namely the so-called Avalokiteśvarasūtra. The banner 245 also bears a Chinese inscription, but according to Takao Moriyasu, it was probably written in the Mongolian period. See Moriyasu and Zieme, “Uighur Inscriptions on the Banners,” 466.
This part is translated into German. See, e.g., Moriyasu, Die Geschichte des uigurischen Manichäismus an der Seidenstraße, 167–168.
Several Chinese manuscripts of the Saddharmapuṇdarīkasūtra, volume 7, which contains the 25th chapter, the relevant chapter for the Avalokiteśvara, are found from Turfan and dated in the 9th to 10th centuries. See Tulufan wenshu zongmu. It indicates the local Buddhists used and copied this text in Chinese during that period.
Those inscriptions are edited in Moriyasu and Zieme, “Uighur Inscriptions on the Banners,” 463–464, 468–469.
The involvement of other groups who used other languages living in the Turfan area, including Tocharians and Sogdians, also has to be taken into consideration. For Tocharians and Sogdians, however, their assimilation to the Uyghurs could progress quicker than the Chinese or their preferred Buddhist donations could differ from those of the Chinese and Uyghurs following the Chinese tradition. To date, any inscriptions provide no clear evidence of Tocharian and Sogdian Buddhists donating temple banners.
See Rong, “Chinese Inscriptions on the Turfan Textiles,” 476.
See, e.g., Ines Konczak, “Origin, Development and Meaning of the Praṇidhi Paintings on the Northern Silk Road,” in Torufan no bukkyō to bijutsu—Uiguru bukkyō wo chūshin ni—Shiruku rōdo no bukkyō bunka—Gandāra, Kucha, Torufan—Dainibu
Kitsudō, “Bezekuriku sekkutsu kuyō bikuzu saikō,” 527–528.
On this topic, see, e.g., Yukiyo Kasai, “Uyghur Legitimation and the Role of Buddhism,” in Buddhism in Central Asia I: Patronage, Legitimation, Sacred Space, and Pilgrimage, ed. Carmen Meinert and Henrik Sørensen (Leiden, Boston: Brill, 2020), 73–78.
See, e.g., Moriyasu Takao
Kitsudō, “Bezekuriku sekkutsu kuyō bikuzu saikō,” 527–528. In Yulin Cave 39, the numerous Uyghur male and female donor figures are also depicted. Some inscriptions inform us that the figures of the empress and the minister are included among them. See Matsui Dai
Three so-called stake inscriptions that are preserved in the Berlin Turfan Collection, for example, show one of those features. They were all donated by the high-ranking Uyghurs. While one is written in Chinese, the other two are written in Old Uyghur. See, e.g., Moriyasu Takao 森安孝夫, “Nishi uiguru ōkokushi no konpon shiryō to shiteno bōkui monjo