Entangled Religions 14.2 (2023) er.ceres.rub.de

Manichaeism and Buddhism in Contact: The Significance of the Uyghur History and Its Literary Tradition

Yukiyo Kasai Ruhr-Universität Bochum, Germany

The exchanges between Manichaeism and Buddhism are one of the most-discussed topics in the religious study of Central Asia. Old Uyghur materials are often used in these discussions because the Uyghurs experienced the religious shift from Manichaeism to Buddhism. The sources attest that Manichaean and Buddhist communities co-existed under Uyghur rule, although the period of co-existence was limited. Thus, the texts produced in that period could show traces of exchange between these two religious communities. Previous studies, however, concentrate on the religious exchanges and do not consider the Uyghurs’ literary tradition, their historical background, and the language development in Old Uyghur. This paper re-examines the materials cited in previous studies, reconsiders exchanges between the two religious communities under Uyghur rule, and addresses the results of this survey.

Old Uyghur, West Uyghur Kingdom, Buddhism, Manichaeism, Confession Texts, Religious Contacts, Central Asia, Uyghur Literary Tradition



Manichaeism had its first contact with Buddhism in the period when its founder, Mani (216–277), was still alive.1 After his death, his disciples and followers began to missionize Central and Eastern Asia, where Manichaeans again encountered Buddhists who had established themselves in that area. As a result of that contact, Manichaeans adopted some Buddhist ideas and transformed them to Manichaean teachings. The adoption of Maitreya, the Future Buddha, is considered an example of such a Buddhist influence on Manichaeism (Sundermann 1997; Hutter 2002). The exchanges between these two religions, especially in Central Asia, have elicited special attention, mainly in religious studies.2 Texts and art objects from that area are often used for such discussions, and those produced by the Uyghurs, a Turkish-speaking tribe, constitute a part of the relevant source material. The majority of the Uyghurs converted from Manichaeism to Buddhism in the tenth to the eleventh century. It is the only case documented in history that one political unit like Uyghurs experienced a shift between these two religions.3 The Uyghur materials excavated in the Turfan area, therefore, offer comprehensive examples of the interactions between them.


The relationship between the Uyghurs and Manichaeism began in the second half of the eighth century, when the Uyghurs established their nomadic empire, the East Uyghur Kaganate (745–840), in Mongolia.4 The official introduction of Manichaeism to that kaganate is thought to have occurred under the third ruler, Bügü Kagan. He brought a few Manichaean clerics from China to his empire, and since that time the religion probably received the support of the Uyghur ruling classes.5 The close connection with and strong influence of the Iranian-speaking Sogdians played a relevant role in the introduction of Manichaeism. Using their established trade network, the Sogdians acted as traders between the West and East and also served as advisers to the Uyghur nomadic rulers.6 The kaganate collapsed in 840, and a large part of the Uyghurs immigrated into the Eastern Tianshan (天山) area, where they founded the West Uyghur Kingdom (from the second half of the ninth to the thirteenth century) (Moriyasu 2015d). In that kingdom, Manichaeism still retained substantial influence among the Uyghurs for a while. However, Buddhism increasingly exerted a dominating influence through the local Buddhist inhabitants, mainly Tocharians, who were Indo-European speakers, and the Chinese (e.g., Moriyasu 2015c; Tremblay 2012, 108–14). Finally, in the second half of the tenth or the beginning of the eleventh century, the majority of the Uyghurs seem to have been Buddhists, and the Buddhist community enjoyed the support of the Uyghur ruling classes instead of the Manichaeans (e.g., Moriyasu 2015b).


Because this religious shift was gradual, the Manichaean and Buddhist communities co-existed for a while in the West Uyghur Kingdom.7 The texts and art objects the Uyghurs produced in that period, therefore, often reflect interactions between these two religious communities. Their co-existence is an absolutely rare case, such that these materials are irreplaceably valuable for discussing interactions between Manichaeism and Buddhism. Previous discussions were, however, often concerned with aspects of religious studies, without adequately taking the historical background of the Uyghurs’ religious shift nor the development of literary traditions in Old Uyghur into consideration. This paper re-examines the phenomena that previous discussions selected as examples of such interactions, but from the perspective of Uyghur historical and literary contexts.

Similarities and Differences between Manichaean and Buddhist Texts Produced under Uyghur Rule


Although Manichaeism had already been introduced to the East Uyghur Kaganate, the majority of the Old Uyghur Manichaean texts which are now accessible to us were produced after the Uyghurs left Mongolia.8 Because of the short duration of Manichaean influence in the West Uyghur Kingdom, the number of these texts does not compare to the massive amount of Buddhist ones, which were continuously produced until the end of the Mongolian period (fourteenth century).9 The Old Uyghur texts preserved in the Berlin Turfan Collection, one of the most extensive Old Uyghur collections in the world, for example, are catalogued as part of the project “Union Catalogue of Oriental Manuscripts,” which will end in 2022.10 To date, there are nineteen published volumes of catalogues of the Old Uyghur manuscripts. Among them, only one volume is dedicated to Manichaean texts, while fifteen volumes concentrate on Buddhist ones.11


Investigating the exchange between Manichaean and Buddhist communities evident in Old Uyghur texts, this paper focuses on those which were produced in the period when both religious communities co-existed under Uyghur rule. Thus, this study primarily investigates the Buddhist texts dated to the pre-Mongolian period and only secondarily uses those from the Mongolian period. Of the Manichaean ones, those written in Old Uyghur are addressed first. Under Uyghur rule in the Turfan area, however, Manichaean texts were written in Middle Iranian, although they were in part created for the Uyghurs. These are also taken into consideration when they clearly show a connection with the Uyghurs.


Still, the number of the extant Manichaean and Buddhist texts is unbalanced, and this provides the first difficulty, and a serious one, for the comparison. They were, however, produced by the Manichaean and Buddhist communities in the West Uyghur Kingdom, such that they reflect these two communities’ interactions. It is, therefore, relevant to get a sense of extant texts.

Contents of the Preserved Texts


Besides narratives or teaching treaties, the extant Manichaean texts contain liturgical works like hymns, petitions, confessions, and church or missionary histories.12 On the subject of exchanges with Buddhism generally, traces of interactions with Buddhism have already been pointed out in some of these texts: Several Manichaean stories contain elements that indicate they were adopted from Buddhist ones. Those exchanges, however, did not necessarily take place in Turfan under Uyghur rule, so that they are not the main discussion topic of this paper (see, e.g., van Tongerloo 2005).13


Instead, it is worth mentioning that hymns constitute a demonstrative part of the Manichaean sources. Not all hymns are written in Old Uyghur. Some hymns written in Middle Iranian languages are also dedicated to the Uyghur rulers or to their kingdom, so their connection to the Uyghurs is obvious (Clark 2017, 133, No. 119–144).14 Some of these hymns, both in Old Uyghur and in Middle Iranian languages, even seem to have been used in official ceremonies, like the so-called Enthronement Hymn preserved under the signature M 919 in Berlin (von LeCoq 1922, 33–35, No. 14). This shows that the Manichaean hymns also played a relevant role in supporting the rulers’ legitimation. In the period when Manichaeism still had a strong influence among the Uyghurs, the newly founded West Uyghur Kingdom had to stabilize its rule in the Turfan area. The Uyghurs who came with their rulers to the Turfan area knew that the rulers of the East Uyghur Kaganate in Mongolia gave their official support to the Manichaean community. To show that the rulers of the newly founded kingdom had a close connection to Manichaeism was, therefore, relevant to their authority as the ruling lineage, which extended back to the kaganate. In this situation, such Manichaean hymns were effectively used to demonstrate the rulers’ and kingdom’s legitimacy in official ceremonies.15 Besides these liturgical texts, it is worth mentioning that the legend of the Uyghurs’ ancestor, Bokug Khan, contains many Manichaean elements: Bokug Khan’s birth from the tree, the essential role of the light, the person in white cloth.16 Given that it mentions the Uyghurs’ migration, this legend was probably completed in the West Uyghur Kingdom. Many Manichaean elements in it also indicate a close relationship between the Uyghur ruling house and the Manichaean community in Turfan.


In contrast, a significant portion of the Buddhist sources consists of sutras and commentaries, and most were translated from other languages, like Chinese, Tocharian, or Tibetan. The Chinese influence in particular is very strong, and Chinese is the original language from which most of the Old Uyghur texts were translated (Elverskog 1997; Zieme 1992). Similar to the Manichaean sources, there are some Buddhist confession texts, although their origin is still under discussion.17 Buddhist hymns were also produced, but most of them thematise purely religious topics. In the Mongolian period (from the thirteenth to the fourteenth century), some hymns containing praises for the Uyghur rulers or their kingdom were copied. Their production date remains unclear, however. The hymns whose contents and function can be compared with the Manichaean Enthronement Hymn are, therefore, not among the Buddhist sources which were clearly produced in the pre-Mongolian period.18 At least some Buddhist texts show that the Buddhists in the West Uyghur Kingdom, like the Manichaeans, seem to have tried to offer support for the legitimation of the Uyghurs’ ruling house. The Bokug Khan legend mentioned above, for example, is also attested in a Buddhist version dated to the pre-Mongolian period (Kasai 2004). However, this legend seems to have maintained the original form and Manichaean elements during the Mongolian period, such that the Buddhist version did not officially receive the ruler’s acknowledgement.19


Besides hymns and confessions, extant Manichaean and Buddhist texts in Old Uyghur contain teachings and narratives, although the percentage of each differs. All of these have content relevant to spreading religious teachings, so the existence of both Manichaean and Buddhist examples should not be interpreted as the result of the exchanges between these two communities. In the Buddhist sutra, Foshuo tiandi baying shenzhou jing 佛説天地八陽神咒經 [Mantrasūtra of the Eight Principles of Heaven and Earth as Spoken by Buddha, T. 2897.85], however, the reflection of Manichaean ideas and teachings is worth pointing out. The source of this sutra is Chinese apocrypha, and the first translation into Old Uyghur was made in the period when Manichaeism still maintained a strong influence among the Uyghurs. It seems that this sutra was popular because several manuscripts were found. According to Juten Oda’s investigation, the Old Uyghur translation of that sutra was later revised.20 Oda identifies a few additions and deviations in the earlier version as influenced by Manichaean teachings. In the revised version, these passages are deleted or changed to more suitable Buddhist expressions (Oda 1978).21 In his newest research, he quotes ten passages as examples which only occupy a small part of the whole text. Thus, in spite of those “Manichaean” passages, the Old Uyghur version can be still characterized as a Buddhist sutra in its entirety. Oda supposes that these “Manichaean passages were used to offer the understanding of the Chinese yin-yang dualism, which is one of the main topics in that sutra to the Uyghurs.” He also concludes that the Old Uyghur version “does not represent an Iranian or Manichaean transformation of a Chinese Buddhist apocryphal scripture” and “propounds Buddhist doctrine” (Oda 2002, 187).22

Use of Alliteration


It is worth mentioning that the use of alliterations is evident at least in the following three Manichaean hymns: those preserved in Berlin Turfan Collection under the signatures So 14411 = III 200 (T II D II 169) and U 32 (TM 419), and the hymn known under the title “Great Hymn to Mani”, to which many fragments belong.23 The exact composition of these hymns remains undated, but generally, the Manichaean texts were composed earlier than Buddhist ones.24 Among these three, the hymn preserved in the fragment So 14411 deserves special attention. This hymn, in which alliterations end rhymes are used, probably belongs to the original poetry and already shows a high degree of perfection in rhyming, as Wilkens pointed out (Wilkens 2009, 321). These hymns indicate that this literary form was probably known and used by the Uyghurs when Manichaeism had a strong influence on them. Peter Zieme, who investigated the Old Uyghur alliteration verses, suggests that this literary form originated as an Uyghur one (Zieme 1991, 29, 1992, 33). The Manichaeans probably cultivated this literary tradition and also adopted it for their literature.


Some Buddhist verses are also written in alliteration, and one such verse is preserved in a text that can be dated to the ninth or tenth century, namely, the period when Manichaean influence was still visible.25 This indicates, therefore, the possibility of exchanges between Manichaean and Buddhist literature or the Buddhists’ learning this literary form through the Manichaeans even if the Uyghurs had it in their literature tradition. However, examples of Buddhist alliterative verses from the early period are rather rare, and they were produced more often in the Mongolian period.26 Thus, this literary form had its later fluorescence among the Buddhists in this period, which was probably not the result of retaining a literary tradition which was transmitted by Manichaeans. In the Mongolian period, Buddhism became so deeply rooted in the Uyghur Buddhists that some of them had enough ability to transform the Buddhist texts in prose into alliterative verses (Zieme 1992, 33–34). In addition, the literary taste of the Mongols, who became the Uyghurs’ new sovereigns and also cultivated this literary form,27 could have had a certain impact on this fluorescence, too.

Book Form


The book forms, which were the primary ones used for Manichaean and Buddhist texts, are different.28 While Manichaean texts are mainly written in the codex, the primary form for Buddhist texts is the Indian poṭhi, which was widespread in Buddhist cultural areas. Besides, the codex or booklet was also used by Buddhists, and in Dunhuang, the neighbouring oasis of Turfan, it became dominant from the tenth century onward (Hirai Yūkei [平井宥慶] 1985, 23; Galambos 2020, 24–25, 32–36). For the Uyghur Buddhists, however, it never became a major form. Its use, therefore, does not directly indicate the Manichaean influence. For a few Buddhist texts in the codex which date to the period of Buddhism’s introduction to the Uyghurs, this choice can be interpreted as the Manichaean contribution.29 The minimal number of such texts, however, shows that the Buddhist adoption of the codex book form was neither systematically nor continuously undertaken.


In an opposite case, there is a Manichaean text which uses the Buddhist poṭhi (see Clark 1982; Moriyasu 2015c, 635, 2015b, 605).30 There are several opinions about the dating of this text, but it was probably produced in the period when Manichaeism was losing its influence and Buddhism was becoming dominant among the Uyghurs. Thus, the use of the poṭhi in this case was probably a result of the Manichaean community’s effort to regain the Uyghurs, who were increasingly interested in Buddhism.31

Characteristics Seen in Grammatical Forms and Terms


Some Manichaean and Buddhist texts share some characteristics seen in grammatical forms and terms, so that they could also be a result of exchanges between those two religions. Generally, for the changes of grammatical and phonetical features in Old Uyghur, two or three stages are recognised, called ñ-, (n-), and y-language, in chronological order.32 This change can be observed, for example, in the word meaning “bad.” This word had the form añıg in the stage of the ñ-language, and then anıg, if the n-language should be recognised as the chronological stage.33 Finally, it became ayıg in the y-language stage.


The exact date of these changes is still debated. In fragments found in the Turfan area, which the Uyghurs primarily produced after their migration, namely the second half of the ninth century, both ñ-(n-) and y-language forms are attested. Generally, Manichaean texts show the ñ-(n-)language stage. The language used in a few Buddhist texts also belongs to that stage, while in the majority of them, the y-language features appear (see Erdal 2004, 70–75). This chronological change can be observed, for example, in various manuscripts of the Buddhist text titled Maitirisimit “Meeting with Maitreya.” While the language of the earlier manuscripts belongs to the ñ-(n-)language stage, the latest one shows the features of the y-language stage. The latter was copied in 1067, at which period Old Uyghur, therefore, seems to have reached the last stage.34 The Manichaean texts generally precede the appearance of Buddhist ones (Laut 1986, 10–11; Moriyasu 2015c, 619–21), so that their presumed belonging to the ñ-(n-)language is reasonable. Moreover, this change is a genuinely linguistic process and developed utterly independently of the Uyghurs’ religious shift. Thus, it is not the result of the Buddhists’ adoption from Manichaeans that the features of the ñ-(n-)language stage appear in some Buddhist texts.


Furthermore, Old Uyghur Buddhist texts contain many Buddhist terms which were borrowed from Sogdian, although the majority were loanwords from Tocharian (Mironov [1929] 1928, 158–68; Shōgaito 1978; Moerloose 1980). In some previous studies, these Sogdian loanwords are taken as proof for a Sogdian influence, which was likely active during the introduction of Buddhism to the Uyghurs.35 These loanwords, however, had already begun to be used in Old Uyghur Manichaean texts and were continuously used in Buddhist texts by that time. Moriyasu provides three loanwords as examples:

č(a)hšap(u)t “commandment”
nom “law”
nizvani “affliction”


In his opinion, they correspond to the Manichaean Sogdian forms and not the Buddhist ones: cxš’pδ, nwm, nyzβ’ny as Manichaean forms, and škš’pt, δrm, w(y)tγ(w)y sr’yβt’m as Buddhist forms.36 However, Antje Wendtland’s recent study proves that those “Manichaean forms” can be found in Buddhist texts (Wendtland 2018). Namely, the form cxš’pt is attested at least in a Buddhist text written in Sogdian script, while nwm is also used in Christian texts, and the form nyzβ’ny is attested in Manichaean, Buddhist, and Christian texts. Thus, these words were not necessarily borrowed via Sogdian Manichaeans.37


To resolve this question, the words in the Old Uyghur Manichaean and Buddhist texts which refer to the same topics or ideas should be systematically investigated. As we see above in the section “Contents of the Preserved Texts,” however, the different contents of the preserved texts in both religious groups present difficulties. Among them, the confession texts are about the only genre which is extant in both Manichaean and Buddhist sources and composed based on the same concept. The next section, therefore, discusses the similarities and differences in terminologies and expressions used in confession texts from both religions.

Traces of Contacts: Manichaean and Buddhist Confession Texts

Confession Texts Preserved in Old Uyghur Manichaean and Buddhist Literature


Confession seems to be a religious act prevalent among the Uyghurs at that time, regardless of the religion they adhered to. Not only the clerics but also the laypeople seem to have participated in this activity because both groups composed this kind of texts. The questions of whether the texts were regularly used during a confession ritual, and if such a ritual was carried out in both Manichaean and Buddhist communities in Turfan, are still debated. As Jes P. Asmussen points out, the Manichaean confession texts were probably established under general Buddhist influence (Asmussen 1965, 254–58).38 However, it is unlikely that the Manichaeans got to know the idea only as late as the second half of the ninth century, when the Uyghurs founded their kingdom in Central Asia. Thus, by that time, neither Manichaeans nor Buddhists were likely still aware of the Buddhist contribution on the establishment of the Manichaean confessions. Rather, the opposite direction is more likely for the Uyghur case, with Manichaeans influencing Buddhist ones (see Klimkeit 1997, 193–228).39


Manichaean confessions are preserved with both clerics and laypeople as the “audience.”40 Among them, the most famous work used by the audience is called the Xuāstvānīft, which is preserved in both Sogdian and Old Uyghur.41 For the Old Uyghur version, a Sogdian intermediary is most likely, although the extant Sogdian version does not exactly correspond to the Old Uyghur one (e.g., Henning 1936, 586–88; Clark 2013, 5). This text was probably written in the period of Manichaean florescence among the Uyghurs. There is also a confession for clerics preserved as part of a Manichaean text written in poṭhi (hereafter: poṭhi book). It dates to the period when the Manichaean community was losing its influence.42 Zieme even suggests the identification of some sentences with a Buddhist original (Zieme 2015). Thus, it is supposed that these two texts show the development of the Uyghurs’ religious shift from Manichaeism to Buddhism over time.


Claudia Weber carried out a comprehensive study of Buddhist confession texts from the perspective of religious studies. She deals with the following texts:43


  • Vinaya texts
  • Suvarṇaprabhāsasottama-sūtra (fourth and fifth chapters)
  • Bodhicaryāvatāra
  • Old Uyghur version of anuttara-pūjā
  • Triskandhaka texts
  • Kšanti Kılmak Nom Bitig
  • Kšanti Kılguluk Nom Bitig
  • Maitrisimit
  • Daśakarmapathāvadānamālā


In addition, there is also a Sanskrit confession transcribed in cursive Uyghur script (Hartmann, Wille, and Zieme 1996). Most of them are, however, characterized as translation literature that is strongly constrained by the source text. The translators probably did not have a high degree of freedom to insert external ideas at their discretion. It is, therefore, not likely that adoptions from Manichaean teachings are reflected in these texts.


Among the texts mentioned above, only the Kšanti Kılmak Nom Bitig [Text to confess] shows minor changes according to the wishes and purposes of those who used it for confession. The Kšanti Kılmak Nom Bitig was probably written by Uyghur lay Buddhists, but its original source remains unidentified. Buddhist confessions were widely produced in Central Asia, including in Dunhuang. Numerous Buddhist ritual texts, including various confessions, were found in this oasis, and seem to have been recited at the rituals in practice (Wang 1998). Considering the strong Buddhist influence from Dunhuang to Turfan and the widespread Buddhist practice of regular confession, it is unlikely that this text is to be counted as the result of a purely Manichaean contribution.44 It is still possible that individual Manichaean ideas are reflected in a few sentences in copies of the Kšanti Kılmak Nom Bitig, produced in the pre-Mongolian period.


The next section is a comparative study of Manichaean Xuāstvānīft and the poṭhi book and the Buddhist Kšanti Kılmak Nom Bitig.

Comparative Study of the Manichaean and Buddhist Confession Texts



Weber also carried out the first comprehensive study that compared Manichaean and Buddhist examples. She uses the Xuāstvānīft as the Manichaean example and takes into consideration the Kšanti Kılmak Nom Bitig and the confession part of the Suvarṇaprabhāsasottama sūtra (Chin. Jinguangming zuishengwang jing 金光明最勝王經 T. 665.16) [Sutra of Golden Light] as Buddhist ones (Weber 1999, 123–52).45 The topics the Xuāstvānīft and Kšanti Kılmak Nom Bitig mention as transgressions which must be confessed are listed in various places in her book (see Weber 1999, 105, 135).46 The following table gives an overview and comparison of those transgressions in the texts mentioned above:47

Table 1: Transgressions
Xuāstvānīft Kšanti Kılmak Nom Bitig
Sins against the God of Pure Light Awareness as laypeople
Sins against the Sun and Moon Gods Transgressions as yakṣas
Sins against the fivefold God: Air, Wind, Light, Water, and Fire Transgressions which are comparable with those committed by the kings in past
Sins against the former messengers of God Reasons for our numerous transgressions
Sins against the five kinds of creatures Five ānantarya
Ten kinds of transgressions and sins in thought, word and deed Five ānantarya sabhāga
The road of the two poisons and to the gate of hell Nine gāḍha
Two principles (Light and Darkness) and the doctrine of the three times Twelve asaṃvara
Ten Commandments Ten karmapatha
Four daily prayers Transgressions with body, speech, and mind
Seven kinds of alms Transgressions as disturber
Fifty fasting days Reference to the Avīci hell
Confession on every Monday Confession in front of the Buddhas of the three times
Yimki and Bema festival Wish for decreasing the transgressions
Closing summary Wish for appearance of Maitreya
Wish for attaining the Buddhahood


In the Kšanti Kılmak Nom Bitig, it seems that the different transgressions were chosen according to the needs of the clients. Thus, none of the extant copies contain all of the topics listed in the table. The list shows that the transgressions dealt with in the Xuāstvānīft and the Kšanti Kılmak Nom Bitig explicitly differ from each other. Furthermore, both texts present them in different ways. In the Xuāstvānīft, the fifteen topics mentioned above are presented at the beginning of each section, and each section begins with the corresponding ordinal number and is addressed to one topic. The Kšanti Kılmak Nom Bitig, in contrast, often begins with the name of the clients who ordered the copy of the text or wrote it themselves, and deals with several topics together in one section. Thus, neither text seems to have adopted the transgression concepts nor the style of writing from the other.

Phrasing, Terms, and Grammatical Framework


As mentioned above, some Buddhist terms in the Old Uyghur Manichaean texts were borrowed from Sogdian and continuously used in Buddhist ones. It is, however, debatable whether the Buddhists used these terms in a way that engaged with the Manichaean concepts and teachings behind them. Generally, the Xuāstvānīft and the Kšanti Kılmak Nom Bitig do not deal with the same transgressions. So a comparison of terms that function as keywords in the texts is not possible in a strict sense.


However, there are some connections: One of the most striking similarities is the phrase at the end of each section (e.g., Weber 1999, 134–35). The ending formula is always included, and both mean “release my sins!”:


  • Xuāstvānīft: manāstār hirzā (e.g. Clark 2013, LC031)
  • Kšanti Kılmak Nom Bitig: dežit kšanti bolzun (e.g. Müller and von Gabain 1910, 77)


There is also a certain parallelism in the fact that both use a non-Uyghur phrase that is borrowed from other languages—namely, manāstār hirzā from the Parthian in the Xuāstvānīft and dežit kšanti from the Tocharian in the Kšanti Kılmak Nom Bitig. Weber, furthermore, points out some similarities in the phrasing and use of words:


  1. using the similar phrase “when you do things you shouldn’t” (kılmaguluk kılınčlarıg kıl- in the Buddhist text and išlämäsig iš išlä- in the Manichaean text)
  2. mentioning the four-footed living beings (tört butlug tınl(ı)g)
  3. using the phrase “without knowing and understanding” for the transgressions (bilmätin ukmatın)
  4. mentioning the creatures that fly (učugma) and the creatures that crawl on their stomachs (bagrın yorgma tınl(ı)g)
  5. mentioning the snake charmer (luu öntürgüči) and wizard (yadčı)
  6. mentioning the living beings on the dry and damp soil (kurug öl yer).48


In Weber’s opinion, these similarities indicate that the author of the Xuāstvānīft surely knew much of Buddhism and that the Buddhists were also familiar with Manichaean teachings. However, which text was composed earlier is still open for the debate (Weber 1999, 140). Among these similarities, (1) and (3) should not be counted because the phrasing in (1) shows an apparent difference in the words used and the sentence formed, while the phrase in (3) is too common in Old Uyghur to indicate a meaningful similarity.49 The other points could indicate some exchange between Manichaean and Buddhist communities, but they do not indicate the adoption of teaching concepts from each other.


In contrast, some features Weber does not mention provide small differences between Manichaean and Buddhist confessions. The beginning formula in the Xuāstvānīft is mostly yemä “now, also” and stands immediately after the ordinal number. The Kšanti Kılmak Nom Bitig, in contrast, puts takı yemä “and now” in front of the proper name of the clients (see, e.g., Clark 2013; Müller and von Gabain 1910, 76, line 12).50 In the explanation part of transgressions, the Xuāstvānīft often uses the phrase näčä and the verb in past tense with the personal suffix for the first-person plural followed by the subjunctive ärsär, “if in any way we have . . ..”51 In the Kšanti Kılmak Nom Bitig, the personal suffix is that for the first-person singular, even in the case that more than one person is mentioned as the clients.52 Before the ending formula, the wish for releasing the transgressions is expressed:


  • In the Xuāstvānīft: täŋrim amtı ökünür biz yazukta bošnu ötünür biz, “My God, now! We beg to be free from sins.”
  • In the Kšanti Kılmak Nom Bitig: amtı anı barča alku ötünü bilinü kšanti kılu täginürmän tsui-da yazukda boš bolayın, “So now, I want to repent of everything, recognize (my sin and transgression) and confess humbly. I want to be free from sin and transgression!”53


The basic terms like tsui / yazuk for “transgressions” or ötün- for “to request, pray” are shared by both texts, but the critical phrase for the confession, “We (or: I) would be freed from the sin,” is formed differently. In the Xuāstvānīft, the gerundive phrase yazukda bošnu with the verb bošun- “to be freed” is used together with the verb ötün- “to pray” in aorist form. In contrast, the Kšanti Kılmak Nom Bitig uses the adjective boš “free” with the verb bol- “to become” in the imperative form.


The similarities and differences observed in both texts indicate that Uyghur Buddhists sometimes borrowed terms and phrases from Manichaean ones in order to support their Uyghur readers’ understanding. However, they probably did not employ the Manichaean teaching concepts, nor were they likely influenced by them. In this context, it is worthwhile to mention here the confession of an Elect in the poṭhi book, which was produced in the later period, namely when Buddhism was becoming the Uyghurs’ primary religion while Manichaeism was losing its influence. Because of this situation, that confession text shows a Buddhist influence in its use of terms. The terms in that text are similar to the standard ones in Buddhist texts. Because of this fact, Jens Wilkens supposes that it was composed in the period when these Buddhist terms had already been established—namely, at the end of the tenth or rather the beginning of the eleventh century (Wilkens 2008, 210). The terms which appear in this text, like nizvani “affliction” or burhan kutı “Buddhahood”, are those which play an essential role in the Buddhist teachings (Clark 2013, 112, LC1005, 1007-1008, 113, LC1021). Furthermore, the use of the term tört tugum “four births” (Clark 2013, 115, LC1040) for living beings reflects the author’s active intention to adopt Buddhist terminology, because the Xuāstvānīft shows that there seems to have been a standard Manichaean expression for this concept, tört butlug tınl(ı)g “four-footed living being.”54


The adoption of the terms and phrases observed in the Kšanti Kılmak Nom Bitig and the confession of an Elect in the poṭhi book shows a different and precise level of adoption. The case of the Kšanti Kılmak Nom Bitig is comparable to the adoption in the Mantrasūtra of the Eight Principles of Heaven and Earth as Spoken by Buddha discussed above in the section “Contents of the Preserved Texts.” In the confession of an Elect, in contrast, the borrowed terms explain the fundamental Buddhist teachings, so this also included the conceptual level.

Closing Remarks


The extant texts show that Manichaean and Buddhist sources differ from each other in regard to their content. Still, a few sentences in the Mantrasūtra of the Eight Principles of Heaven and Earth as Spoken by Buddha, one of the earliest Buddhist translation literatures in Old Uyghur, indicate that the author used Manichaean terms and teaching concepts to make some foreign ideas understandable for Uyghur readers, most of whom were still familiar with Manichaean terms and teachings. The use of these terms and concepts was, however, not intended to change or assimilate the Buddhist contents of the sutra mentioned above to Manichaean ones. Several essential terms were borrowed from Sogdian and shared by Uyghur Manichaeans and Buddhists, although the majority of Buddhist terms were loanwords from Tocharian. Thus, certain adoptions from Manichaean to Buddhist texts are evident. The use of alliteration, for example, is observed in both Manichaean and Buddhist texts. The alliteration probably has its root in Uyghur literature, and Manichaeans actively adopted it. Buddhists also tried to use alliteration in the very early period, but its real florescence first developed in the Mongolian period. This gives rise to the question of whether the tradition of Old Uyghur literature was inherited by both Manichaeans and Buddhists, although Manichaeans were more actively involved in following the literary tradition.


Confession texts were produced in both Manichaean and Buddhist communities, although the specific details of their use are still debated. Regarding the contents, phrasing, and terminology, however, two almost contemporary examples, the Xuāstvānīft and the Kšanti Kılmak Nom Bitig, do not show any distinct traces of mutual influence. The result of this comparison indicates that both Manichaeans and Buddhists essentially followed their own confession tradition. In contrast, the Manichaean confession of an Elect used many fundamental Buddhist terms. It was probably composed after Buddhism gained solid footing among the Uyghurs.


These observations show that Manichaeans and Buddhists knew each other’s literature and that there was the possibility of exchange under Uyghur rule. However, Buddhists do not seem to have actively adopted Manichaean teachings, while later Manichaeans tried to assimilate their texts to Buddhist ones. These different adoptions may have resulted from the different situation of the Manichaean and Buddhist communities under the Uyghurs. As mentioned above, Manichaeism was supported by the Uyghur ruling classes as their main religion in the newly founded West Uyghur Kingdom for a while. However, this does not necessarily mean that the Uyghur rulers persecuted the local Buddhists or destroyed Buddhist temples. The Chinese ambassador from the Sung Dynasty, Wang Yande 王延徳, reports this about Kočo, one of the West Uyghur Kingdom’s capitals: “There are more than fifty Buddhist temples, all of them with nameplates on the gate given by the Tang court itself.”55


He visited this kingdom around 980, such that his account speaks to the continuous maintenance of the Buddhist temples there from the Tang period. In contrast, the Manichaeans had to face the fact that the Uyghur rulers’ favour gradually shifted to the Buddhists, and that a few rulers were even ready to order the destruction of the Manichaean monasteries to make way for new Buddhist ones.56 Although the Uyghur rulers tried to maintain balance between the two religious communities for a while, the situation was more difficult for the Manichaeans. Since Manichaeism received the support of the Uyghur ruling classes in Mongolia, the Manichaean community also developed in the Turfan area. The Manichaean hymn book Maḥrnāmag, which began to be composed around 762 in Ark in Middle Iranian, indicates the existence of Manichaean communities in this area at that time.57 The community was, however, established under the Uyghurs’ initiative and support, so that it could not continue to exist after the majority of the Uyghurs converted to Buddhism. In contrast, the Buddhists seem to have continuously maintained their tradition through the support of an established local community. In this situation, it was essential for Buddhists to obtain the Uyghur rulers’ favour, but they did not struggle for survival like the Manichaeans.


I would like to express my sincere thanks to Prof. Dr. Jens-Uwe Hartmann (Munich) for his helpful remarks, and to the editor of this volume, Prof. Dr. Kianoosh Rezania (Bochum), and two anonym reviewers for their useful suggestions and comments. Of course, I alone am responsible for any mistakes. Research and writing of this paper were facilitated through the BuddhistRoad project which has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 725519).


Asmussen, Jes P. 1965. Xᵘāstvānīft. Studies in Manichaeism. Copenhagen: Prostant Apud Munksgaard.

BeDuhn, Jason David. 2004. “The Near Eastern Connections of Manichaean Confessionary Practice.” Aram 16: 161–77.

Clark, Larry V. 1982. “The Manichaean Turkic Pothi-Book.” Altorientalische Forschungen 9: 145–218.

———. 1997. “The Turkic Manichaean Literature.” In Emerging from Darkness: Studies in the Recovery of Manichaean Sources, edited by Paul Mirecki and Jason BeDuhn, 89–141. Leiden / New York / Köln: Brill.

———. 2000. “The Conversin of Bügü Khan to Macnihaeism.” In Studia Manichaica IV. Internationaler Kongress zum Manichäismus, Berlin, 14.-18. Juli 1997, edited by Roland E. Emmerick, Werner Sundermann, and Peter Zieme, 83–123. Berlin: De Gruyter.

———. 2013. Uyghur Manichaean Texts. Texts Translations Commentary. Volume II Liturgical Texts. Corpus Fontium Manichaeorum. Series Turcica II. Turnhout: Brepols.

———. 2017. Uyghur Manichaean Texts. Texts Translations Commentary. Volume III Ecclesiastical Texts. Corpus Fontium Manichaeorum. Series Turcica III. Turnhout: Brepols.

Deeg, Max, and Iain Gardner. 2009. “Indian Influence on Mani Reconsidered. The Case of Jainism.” International Journal of Jaina Studies 5 (2): 1–30.

Doerfer, Gerhard. 1996. Formen der älteren türkischen Lyrik. Szeged: Universitas Szegediensis de Attila József Nominata.

Elverskog, Johan. 1997. Uygur Buddhist Literature. Vol. 1. Silk Road Studies. Turnhout: Brepols.

Erdal, Marcel. 2004. A Grammar of Old Turkic. Leiden / Boston: Brill.

———. 2018. “The Manuscript Variants of the Säkiz Yügmäk Yarok Nom Bitig.” In Language, Society, and Religion in the World of the Turks. Festschrift for Larry Clark at Seventy-Five, edited by Zsuzsanna Gulácsi, 301–10. Turnhout: Brepols.

Freiberger, Oliver. 2001. “Staatsreligion, Reichsreligion oder Nationalreligion? Überlegungen zur Terminologie.” In Zwischen Säkularismus und Hierokratie. Studien zum Verhältnis von Religion und Staat in Süd- und Ostasien, edited by Peter Schalk, Max Deeg, Oliver Freiberger, and Christoph Kleine, 19–36. Uppsala: Elanders Gotab.

Galambos, Imre. 2020. Dunhuang Manuscript Culture. End of the First Millennium. Berlin / Boston: De Gruyter.

Golden, Peter. 1992. An Introduction to the History of the Turkic People. Ethnogenesis and State-Formation in Medieval and Early Modern Eurasia and the Middle East. Wiesbaden: Harrassowitz.

Hartmann, Jens-Uwe, Klaus Wille, and Peter Zieme. 1996. “Indrasenas Beichte. Ein Sanskrit-Text in uigurischer Schrift aus Turfan.” Berliner Indologische Studien 9 (10): 203–16.

Henning, Walter Bruno. 1936. “Soghdische Miszellen.” Bulletin of the School of Oriental and African Studies 8: 586–8.

Hirai Yūkei [平井宥慶]. 1985. “Kongō hannyakyō [金剛般若経] Vajracchedikā-prajñāpāramitā-sūtra.” In Tonkō to Chūgoku bukkyō [敦煌と中国仏教] [Dunhuang and Chinese Buddhism], edited by Tairyō Makita and Fumimasa Fukui, 17–34. Tokyo: Daitō shuppansha.

Hutter, Manfred. 2002. “Mani Als Maitreya.” In Religionsbegegnung Und Kulturaustausch in Asien. Studien Zum Gedenken an Hans-Joachim Klimkeit, edited by Wolfgang Gantke, Karl Hoheisel, and Wassilios Klien, 111–9. Wiesbaden: Harrassowitz.

Kasai, Yukiyo. 2004. “Ein Kolophon um die Legende von Bokug Kagan.” In Nairiku Ajia gengo no kenkyū [内陸アジア言語の研究] [Studies on the Inner Asian Languages], 19:1–27.

———. 2013. “Der Ursprung des alttürkischen Maitreya-Kults.” In Die Erforschung des Tocharischen und die alttürkische Maitrisimit, edited by Yukiyo Kasai, Abudhurishid Yakup, and Desmond Durkin-Meisterernst, 67–104. Turnhout: Brepols Publishers.

———. 2020. “Uyghur Legitimation and the Role of Buddhism.” In Buddhism in Central Asia I: Patronage, Legitimation, Sacred Space, and Pilgrimage, edited by Carmen Meinert and Henrik H. Sørensen, 61–90. Leiden / Boston: Brill.

———. 2022. “Central Asian and Iranian Influence in Old Uyghur Buddhist Manuscripts–Focused on the Book Format and Donor Colophons.” In The Syntax of South, Southeast and Central Asian Colophons: A First Step Towards a Comparative and Historical Study of Manuscripts in the Poṭhi Format, edited by Giovanni Ciotti, Nalini Balbir, and Martin Delhey, 373–97. Studies in Manuscript Cultures 29. Berlin: De Gruyter. https://doi.org/10.1515/9783110795271-013.

Klimkeit, Hans-Joachim. 1985. “Buddha als Vater.” In Fernöstliche Weisheit und christlicher Glaube: Festgabe für Heinrich Dumoulin SJ zur Vollendung des 80. Lebensjahres, edited by Hans Waldenfels and Thomas Immoos, 235–59. Mainz: Matthias-Grünewald-Verlag.

———. 1986. Die Begegnung von Christentum, Gnosis und Buddhismus an der Seidenstraße. Opladen: Westdeutscher Verlag.

———. 1997. “Manichäische und buddhistische Beichtformel aus Turfan. Beobachtung zur Beziehung zwischen Gnosis und Mahāyāna.” Zeitschrift für Religions- und Geistesgeschichte 29 (3): 193–228.

Kuo, Li-ying. 1994. Confession et contrition dans le bouddhisme chinois. Paris: Publications de l’école française d’extrême-orient.

Laut, Jens Peter. 1986. Der frühe türkische Buddhismus und seine literarischen Denkmäler. Wiesbaden: Harrassowitz.

———. 2002. “Gedanken zum alttürkischen Stabreim.” In Splitter aus der Gegend von Turfan. Festschrift für Peter Zieme anläßlich seines 60. Geburtstags, edited by Mehmet Ölmez and Simone-Christiane Raschmann, 129–38. Istanbul / Berlin: Safak Matbaacilik.

LeCoq, Albert von. 1922. “Türkische Manichaica aus Chotscho III.” Abhandlungen der preußischen Akademie der Wissenschaften, phil.-hist. Klasse 2: 3–49.

Matsui, Dai. 2017. “An Old Uigur Account Book for Manichaean and Buddhist Monasteries from Tempel α in Qoco.” In Zur Lichten Heimat. Studien Zu Manichäismus, Iranistik Und Zentralasienkunde in Gedenken an Werner Sundermann, Edited by Team „Turfanforschung“, 409–20. Wiesbaden: Harrassowitz Verlag.

Mironov, Nikolaj D. (1929) 1928. “Kuchean Studies.” Rocznik Orientalistyczny 6: 89–169.

Moerloose, Eddy. 1980. “Sanskrit Loan Words in Uighur.” Journal of Turkish Studies 4: 61–78.

Moriyasu, Takao 森安孝夫. 2004. Die Geschichte des uigurischen Manichäismus an der Seidenstraße. Forschungen zu manichäischen Quellen und ihrem geschichtlichen Hintergrund. Wiesbaden: Harrasowitz Verlag.

———. 2008. “Chronology of West Uighur Buddhism—Re-Examination of the Dating of the Wall-Paintings in Grünwedel’s Cave No. 8 (New: No. 18), Bezeklik—.” Edited by Peter Zieme. Aspects of Research into Central Asian Buddhism. In Memoriam Kōgi Kudara. Silk Road Studies XVI 8 (New: No. 18): 191–227.

———. 2015a. “New Development in the History of East Uighur Manichaeism.” Open Theology 1: 316–33.

———. 2015b. “Nishi Uiguru ōkoku ni okeru manikyō no suitai to bukkyō no taitō [西ウイグル王国におけるマニ教の衰退と仏教の台頭] [The declination of Manichaeism and the rise of Buddhism in the West Uyghur Kingdom].” In Tōzai Uiguru to Chūō Yūrashia [東西ウイグルと中央ユーラシア] [Eastern and Western Uyghurs and Central Eurasia], edited by Takao 森安孝夫 Moriyasu, 590–617. Nagoya: Nagoya University Publishers.

———. 2015c. “Toruko bukkyō no genryū to ko torukogo butten no shutsugen [トルコ仏教の源流と古トルコ語仏典の出現] [L’origine du Bouddhisme chez les Turcs et l’apparition des textes bouddhiques en turc ancien].” In Tōzai Uiguru to Chūō Yūrashia [東西ウイグルと中央ユーラシア] [Eastern and Western Uyghurs and Central Eurasia], 618–44. Nagoya: Nagoya University publishers.

———. 2015d. “Uiguru no seisen ni tsuite [ウイグルの西遷について] [Nouvel examen de la migration des Ouïgours au milieu du IXᵉˢiècle].” In Tōzai uiguru to chūō yūrashia [東西ウイグルと中央ユーラシア] [Eastern and Western Uyghurs and Central Eurasia], edited by Takao 森安孝夫 Moriyasu, 276–98. Nagoya: Nagoya University publishers.

———. 2015e. “Zōho: Uiguru to Toban no hokutei sōdatsusen oyobi sonogo no saiiki jōsei ni tsuite [増補 : ウイグルと吐蕃の北庭争奪戦及びその後の西域情勢について] [The Uyghur-Tibetan Struggle for Beshbalygh and the Subsequent Situation in Central Asia: An Enlarged Edition].” In Tōzai Uiguru to Chūō Yūrashia [東西ウイグルと中央ユーラシア] [Eastern and Western Uyghurs and Central Eurasia], edited by Takao 森安孝夫 Moriyasu, 230–74. Nagoya: Nagoya daigaku shuppankai.

Müller, Friedlich Wilhelm Karl, and Annemarie von Gabain. 1910. “Uigurica II.” Abhandlungen der preußischen Akademie der Wissenschaften, phil.-hist. Klasse 3: 3–110.

Oda, Juten. 1978. “Torukogo hon Hachiyōkyō shahon no keihu to shūkyō shisōteki mondai [トルコ語本八陽經寫本の系譜と宗教思想的問題] [A Geneology of the Manuscripts of Turkish Bayangjing and Problems of Religious Thoughts].” Tōhōgaku [東方学] [Eastern Studies] 55: 104–18.

———. 2002. “On Manichaean Expressions in the Säkiz yükmäk yaruq.” In Splitter aus der Gegend von Turfan. Festschrift für Peter Zieme anläßlich seines 60. Geburtstags, edited by Mehmet Ölmez and Simone-Christiane Raschmann, 179–98. Istanbul / Berlin: Safak Matbaacilik.

———. 2015. A Study of the Buddhist Sūtra Called Säkiz Yükmäk Yaruq or Säkiz Törlügin Yarumïš Yaltrïmïš in Old Turkic. Vol. XXXIII. Berliner Turfantexte. Turnhout: Brepols.

Oki, Mie [沖美江.]. 1996. “Kyū kara jūisseiki ni okeru uiguru moji no sho tokuchō–Jidai hantei heno tegakari wo motomete [9〜11世紀におけるウイグル文字の諸特徴–時代判定への手がかりをもとめて] [A Few Characteristic Features of the Uigur Alphabet in the 9th–11th Centuries].” In Nairiku ajia gengo no kenkyū [内陸アジア言語の研究] [Studies on the Inner Asian Languages], 11:15–60.

Reck, Christiane. 2013. “Nichtbuddhistische mitteliranische Pustakablätter.” In Vom Aramäischen zum Alttürkischen. Fragen zur Übersetzung von manichäischen Texten. Vorträge des Göttinger Symposiums vom 29./30. September 2011, edited by Jens Peter Laut and Klaus Röhrborn, 167–78. Berlin: De Gruyter.

———. 2015. “Sogdian Manichaean Confessional Fragments in Sogdian Script in the Berlin Turfan Collection: The Fragments of the Xwāstwānīft.” In Mani in Dublin. Selected Papers from the Seventh International Conference of the International Association of Manichaean Studies in the Chester Beatty Library, Dublin, 8-12 September 2009, edited by Siegfried G. Richter, Charles Horton, and Klaus Ohlhafer, 289–323. Leiden / Boston: Brill.

Röhrborn, Klaus. 1983. “Zu einem dialekt-differenzierenden Lautübergang im Alttürkischen.” Materialia Turcica 7 (8): 295–305.

———. 2000. “Zum manichäischen Einfluß im alttürkischen Buddhismus.” In Studia Manichaica IV. Internationaler Kongress zum Manichäismus Berlin, 14.-18. Juli 1997, edited by Ronald E. Emmerick, Werner Sundermann, and Peter Zieme, 494–9. Berlin: De Gruyter.

Shōgaito, Masahiro 庄垣内正弘. 1978. “Kodai uigurugo ni okeru indo raigen shakuyō goi no dōnyū keiro ni tsuite [‘古代ウイグル語’ におけるインド来源借用語彙の導入経路について] [On the routes of the loan words of Indic origin in the Old Uigur language].” Ajia, Afurika gengo bunka kenkyū [アジア ‧ アフリカ言語文化研究] [Journal of Asian and African Studies] 15: 79–110.

Sundermann, Werner. 1997. “Manichaeism Meets Buddhism: The Problem of Buddhist Influence on Manichaeism.” In Bauddhavidyāsudhākaraḥ. Studies in Honour of Heinz Bechert on the Occasion of His 65th Birthday, edited by Petra Kiefer-Pülz and Jens-Uwe Hartmann, 647–56. Swisstal-Odendorf: Indica et Tibetica Verlag.

Şinasi Tekin. 1980. Maitrisimit Nom Bitig. Die Uigurische Übersetzung Eines Werkes Der Buddhistischen Vaibhāṣika-Schule. Vol. IX. Berliner Turfantexte. Berlin: Akademie Verlag.

Tongerloo, Aloïs van. 2005. “The Buddha’s First Encounter in a Manichaean Old Turkic Text.” In Il Manicheismo Nuove Prospettive Della Richerca. Dipartimento Di Studi Asiatici Università Degli Studi Di Napoli “L’Orientale”. Napoli, 2-8 Settembre 2001, edited by Aloïs van Tongerloo and Luigi Cirillo, 385–96. Turnhout: Brepols.

Tremblay, Xavier. 2012. “The Spread of Buddhism in Serindia–Buddhism Among Iranians, Tocharians and Turks Before the 13th Century.” In The Spread of Buddhism, edited by Ann Heirman and Stephan Peter Bumbacher, 75–129. Leiden, Boston: Brill.

Wang, Juan [汪娟]. 1998. Dunhuang lichanwen yanjiu [敦煌禮懺文研究] [Studies on the confession ritual texts from Dunhuang]. Taipei: Fagu wenhua.

Weber, Claudia. 1999. Buddhistische Beichten in Indien und bei den Uiguren. Wiesbaden: Harrasowitz Verlag.

Wendtland, Antje. 2018. “Zum manichäischen Ursprung sogdischer Lehnwörter in buddhistischen uigurischen Texten: Gab es ein manichäisches religiöses Vokabular im Sogdischen?” In Der östliche Manichäismus im Spiegel seiner Buch- und Schriftkultur, edited by Zekine Özertural and Gökhan Șilfeler, 141–50. Berlin / Boston: de Gruyter.

Wilkens, Jens. 2000. Alttürkische Handschriften. Teil 8: Manichäisch-türkische Texte der Berliner Turfansammlung. Stuttgart: Franz Steiner Verlag.

———. 2008. “Musings on the Manichaean ‘pothi’ book.” In Nairiku ajia gengo no kenkyū [ 内陸アジア言語の研究] [Studies on the Inner Asian Languages], 23:209–31.

———. 2009. “Ein manichäischer Alptraum?” In Literarische Stoffe und ihre Gestaltung in mitteliranischer Zeit. Kolloquium anlässlich des 70. Geburtstages von Werner Sundermann, edited by Desmond Durkin-Meisterernst, Christiane Reck, and Dieter Weber, 319–48. Wiesbaden: Dr. Ludwig Reichert Verlag.

———. Forthcoming. “Practice and Rituals in Uyghur Buddhist Texts: A Preliminary Appraisal.” In Buddhism in Central Asia II: Practice and Rituals, Visual and Material Transfer, edited by Yukiyo Kasai and Henrik H. Sørensen. Leiden / Boston: Brill.

———. Forthcoming. “Uyghur Buddhism and the Impact of Manichaeism and Native Religion: The Case of Religious Terminology.” In Buddhism in Central Asia III: Doctorine, Exchanges with Non-Buddhist Traditions, edited by Lewis Doney, Carmen Meinert, Yukiyo Kasai, and Henrik H. Sørensen. Leiden / Boston: Brill.

Yoshida, Yutaka. 2008. “Die buddhistischen sogdischen Texte in der Berliner Turfansammlung und die Herkunft des buddhistischen sogdischen Wortes bódhisattva – Zum Gedenken an Prof. Kōgi Kudaras Arbeiten an den sogdischen Texten –.” Translated by Yukiyo Kasai and Christiane Reck. Acta Orientalia Hungariae 61 (3): 325–58.

Zieme, Peter. 1975. Manichäisch-Türkische Texte. Vol. V. Berliner Turfantexte. Berlin: Akademie Verlag.

———. 1985. Buddhistische Stabreimdichtungen Der Uiguren. Vol. XIII. Berliner Turfantexte. Berlin: Akademie Verlag.

———. 1991. Die Stabreimtexte der Uiguren von Turfan und Dunhuang. Studien zur alttürkischen Dichtung. Budapest: Akadémiai Kiadó.

———. 1992. Religion und Gesellschaft im Uigurischen Königreich von Qočo. Kolophone und Stifter des alttürkischen buddhistischen Schrifttums aus Zentralasien. Opladen: Westdeutscher Verlag.

———. 2001. “Araṇemi-Jātaka und ein Sündenbekenntnistext in einer alttürkischen Sammelhandschrift.” In De Dunhuang a Istanbul. Hommage à James Russell Hamilton, edited by Louis Bazin and Peter Zieme, 401–33. Turnhout: Brepols.

———. 2015. “Eine buddhistische Vorlage zum ‘Pothi-Book’?” academia.edu August 2015: 1–3.

  1. If the influence of Indian religion on Manichaeism is discussed, Buddhism usually is the main topic. Max Deeg and Iain Gardner, however, point out that Jainism also played an important role for this exchange (see Deeg and Gardner 2009).↩︎

  2. Hans-Joachim Klimkeit was one of the scholars who worked intensively on that topic (see, e.g., Klimkeit 1986, 23–51).↩︎

  3. The detailed process of the Uyghurs’ conversion to Manichaeism and further to Buddhism is explained below.↩︎

  4. On the history of the East Uyghur Kaganate, see, e.g., Golden (1992, 155–76).↩︎

  5. For the Uyghurs’ conversion to Manichaeism, see Clark (2000); Moriyasu (2015a). Turcologists explain that Manichaeism became the state religion of the Uyghurs. The term “state religion” is used with different meanings in various contexts, as Oliver Freiberger shows (see Freiberger 2001). In the Uyghurs’ case, both the details of treatments which the Manichaean community probably received from the rulers and the practice of the population in the kaganate are still being debated. Thus, the use of this term could cause misunderstanding of the situation of the Manichaean community under the Uyghur rule. The term “state religion” is, therefore, not used in this paper. This, however, does not mean that this paper represents a new interpretation of the relationship between the Uyghurs and Manichaeism. The problem is the concept of the term and not the interpretation of the historical facts.↩︎

  6. There are several noted reasons for the introduction of Manichaeism. The previous studies on that topic are summarized in Kasai (2020, 65).↩︎

  7. They even enjoyed the simultaneous official support of the Uyghur kings, although they did not always receive it to the same degree. The account book found from Turfan documents the financial support which was given both to Manichaean and Buddhist monasteries (see Matsui 2017).↩︎

  8. Larry V. Clark lists Old Uyghur Manichaean texts preserved in the Berlin Turfan Collection and classifies them according to their contents (see Clark 1997). He and Peter Zieme published an edition of a part of those texts (see Zieme 1975; Clark 2013, 2017).↩︎

  9. For an overview of the identified Old Uyghur Buddhist texts, see Elverskog (1997). After the publication of this book, many Buddhist texts were identified and published, several in the Berliner Turfantexte series from the Brepols publishers.↩︎

  10. General information is available on the project website, accessed July 24, 2020 (https://adw-goe.de/en/research/research-projects-within-the-academies-programme/kohd/).↩︎

  11. Jens Wilkens published the volume for Manichaean texts (see Wilkens 2000). The catalogues published to date are listed on the project homepage, accessed July 24, 2020 (https://adw-goe.de/forschung/forschungsprojekte-akademienprogramm/kohd/publikations-serie/katalogbaende/11-20/). The catalogues for Old Uyghur texts are from the numbers 13,9 to 13,28.↩︎

  12. The contents of the Manichaean texts were identified in previous research (see esp. Clark 1997; Wilkens 2000).↩︎

  13. The transmission process of those stories would be clearer if one could conduct a detailed comparison between the Manichaean versions in various languages and the Buddhist ones. However, the very small number of extant Manichaean texts prevents such an investigation. Still, it is likely that some of the liturgical works can be counted as original ones. For a discussion of each text, see Clark (Clark 2013, 2017).↩︎

  14. The corpus of Manichaean hymns in Middle Iranian languages itself is much larger than that in Old Uyghur. But here, only those hymns whose connection with the Uyghurs is shown clearly are considered.↩︎

  15. For a detailed discussion of this topic, see also Kasai (2020, 71–72).↩︎

  16. For the summary of this legend, see Kasai (2004, 9).↩︎

  17. For a detailed discussion, see the section “Traces of Contacts: Manichaean and Buddhist Confession Texts.”↩︎

  18. Peter Zieme collected, published, and investigated in detail many Buddhist hymns (see Zieme 1991, 29). On the use of alliteration, see the section “Use of Alliteration.”↩︎

  19. For a detailed discussion of this topic with references to previous studies, see Kasai (2020, 72–73, 81–82).↩︎

  20. He gathered, classified, and published all accessible extant manuscripts (see Oda 2015). For the classification of the extant manuscripts, see also Erdal (2018).↩︎

  21. In reaction to Juten Oda, Klaus Röhrborn disputed whether Oda tried to see excessive Iranian influences in this Old Uyghur translation. Oda offers rebuttals to this criticism (see Röhrborn 2000; Oda 2002). Moriyasu, furthermore, assumes that this sutra was chosen for translation because the light, an important element in Manichaean teachings, is its main topic (see Moriyasu 2008, 670, 2015c, 634).↩︎

  22. Furthermore, it is assumed that the texts connected with the future Buddha Maitreya were chosen for translation in the early period because the idea of this Buddha existed in Manichaean teachings as a result of adoption from Buddhist ones (see Moriyasu 2008, 216). The cult of this Buddha was, however, introduced and spread among not only the Uyghurs in Turfan, but also other contemporary Buddhist oases and states in Central Asia, like Dunhuang. Thus, the shared idea of Maitreya between Manichaean and Buddhist teachings was probably one of the essential reasons for the wide spread of this cult among the Uyghurs, but not the only one. For the possible transmission of the Maitreya cult from Dunhuang to Turfan (see Kasai 2013).↩︎

  23. For the information on these three hymns, including their publications, see Wilkens (2000, 283–84, No. 311, 315–316, No. 354, 318–338, No. 357–381). While Zieme discusses all of them in his book on alliterations in Old Uyghur, Wilkens dealt with the former in detail (see Zieme 1991, 331–36; Wilkens 2009). For the “Great Hymn to Mani,” Clark’s newest edition is available, although he does not show the alliteration structure (see Clark 2013, 137–77).↩︎

  24. Gerhard Doerfer, who intensively worked on the Turkish verse, even dates a part of the Manichaean texts in Old Uyghur in the Mongolian period (see Doerfer 1996, esp. 69–194). His dating arguments are, however, reverified. According to the current research status, the linguistic features, based on which he dated the texts to the Mongolian period, do not prove the late production of the texts. For reverifying Doerfer’s dating criteria, see, e.g., Oki (1996). The “Great Hymn to Mani” probably belongs to the latest Manichaean texts written in Old Uyghur because it constitutes a part of a Manichaean text written in the Buddhist poṭhi book. This book form indicates that this text, including the “Great Hymn to Mani”, was written when the Buddhist influence became increasingly intense among the Uyghurs. This was, however, still the pre-Mongolian period.↩︎

  25. One is a colophon, in which the Uyghur donors explain the reason for their donation, that was added to a copy of the Maitrisimit “Meeting with the future Buddha Maitreya” (see Laut 2002). There are slight differences in dating among scholars. While Jens Peter Laut, for example, thinks it is possible that the manuscript could date to the ninth century, Moriyasu reveals that the Old Uyghur Buddhist texts began to be produced at the earliest in the second half of the ninth century rather in the tenth century. In his opinion, the manuscript in question should be dated to the tenth century (see Laut 2002, 605). At any rate, the colophon in question belongs to the earliest Old Uyghur Buddhist texts, without any doubt.↩︎

  26. Zieme published many Buddhist alliteration texts (see Zieme 1985).↩︎

  27. For the Mongolian verse tradition, see, e.g., Doerfer (1996, 69–194).↩︎

  28. For this topic, see also Kasai (2022, 375–80).↩︎

  29. These texts are, for example, listed in Moriyasu (see Moriyasu 2015c, 623). For the detailed discussion, see also Kasai (2022, 375).↩︎

  30. Among the Middle Iranian texts, both Manichaean and Christian ones in the poṭhi format are attested, although their number is very small (see, e.g., Reck 2013). Thus, the use of this format itself is not necessarily the result of religious interchanges between Manichaean and Buddhist communities, which should have been limited, but rather, its exchange took place between various book cultures in Central Asia.↩︎

  31. The content of this text is discussed below in the section “Traces of Contacts: Manichaean and Buddhist Confession Texts.”↩︎

  32. Wilhelm Radloff and Annemarie von Gabain first pointed out this change, although at first, they thought it was a dialectical difference. Now many scholars agree that the differences between the three stages are not dialectical but chronological. The previous studies on this topic are listed and summarized in some articlesm (see, e.g., Röhrborn 1983; Moriyasu 2015c, 618–20). It depends on how the stages of the n- and y-language are recognized. Recently, Marcel Erdal pointed out that the difference between the n- and y-language is not chronological but dialectal. According to him, after the ñ-language stage, some dialects showed the change to the forms in the n-language, while in others, including Old Uyghur, it turned into the forms attested in the y-language (Erdal 2004, 12–13, 70–75).↩︎

  33. According to Erdal, the word form written as ’nyk in Uyghur script should be read as añ(ı)g or a(y)ıg, not anıg, in most cases (Erdal 2004, 71–72). The exact criteria according to which the attested dialects can be classified to the n- or y-language/dialect, and the Old Uyghur manuscripts dated more precisely, are not wholly established yet. Thus, the research topic of the chronological changes in Old Uyghur and the dialectal differences in Turkic languages in a pre-modern period still request further detailed discussions.↩︎

  34. While the latest manuscript bore the exact date, the dating of the earlier manuscripts is still in debate (see, e.g., Laut 2002, 131–32, 135).↩︎

  35. Jens Peter Laut collected and investigated them (see Laut 1986, 93–114, 116–19).↩︎

  36. For a detailed discussion of this topic, see Laut (1986); Moriyasu (Moriyasu 2015c, 624–31). According to Yutaka Yoshida’s study, the Uyghur Buddhists probably used the Sogdian Buddhist texts, so the latter played a definite role for the Uyghurs (see Yoshida 2008).↩︎

  37. Besides these terms, Hans-Joachim Klimkeit points out that the expression “Buddha as father” was adopted in the Old Uyghur Buddhist texts under Manichaean influence (see Klimkeit 1985). This expression, in fact, appears neither in Sogdian nor Tocharian Buddhist texts. I appreciate my colleagues Dr. Christiane Reck (Berlin) and Prof. Dr. Hirotoshi Ogihara (Tokyo), who kindly provided the information on the above-mentioned language materials via personal communication. For this topic, see also Wilkens (forthcoming, Section 6).↩︎

  38. For further research on the Manichaean confession texts generally, see BeDuhn (BeDuhn 2004, 161–77).↩︎

  39. Klimkeit tries to show the adoption of Manichaean ideas seen in the Old Uyghur Buddhist texts by quoting an example from the Suvarṇaprabhāsasottama sūtra, but Zieme presents another interpretation and contradicts Klimkeit’s opinion (see Zieme 1991, 212). For a detailed discussion, see below.↩︎

  40. Larry V. Clark provides the newest edition of those texts (see Clark 2013, 5–134).↩︎

  41. Many scholars, commencing with Asmussen, mentioned above, worked on this interesting text. For recent studies, see, e.g., Reck (2015, 289–323); Clark (2013, 5–111).↩︎

  42. There are different opinions on the date of this text. Clark assumes that it was produced by a person who was originally Buddhist and converted to Manichaeism, and places it with other Manichaean texts in the period when Manichaeism had a strong influence among the Uyghurs. Moriyasu, in contrast, supposes that the text was produced in the period when Manichaeism was losing its influence, because many Buddhist features seen in that text were probably an attempt to make Manichaeism attractive to Buddhists (see, e.g., Clark 1982, 156–60; Moriyasu 2015b, 605). For a discussion of dating, see also the section “Phrasing, Terms, and Grammatical Framework” below.↩︎

  43. They are dealt with in her book in Chapter 3 (see Weber 1999, 78–116). On Buddhist confession texts, see also Wilkens (forthcoming).↩︎

  44. Jan Nattier points out the possible Manichaean influence on Old Uyghur confession texts. Nattier’s opinion is worth considering, as she is a well-respected Buddhologist. However, she only gave a talk on this topic, and the details of her discussion were never published. Thus, it is not possible to thoroughly consider her opinion (see, e.g., BeDuhn 2004, 161–62). The confession ritual has a long tradition in Buddhism, and many texts were compose (see Kuo 1994). The various texts are attested not only in Chinese but also in other Central Asian languages (see, e.g., Weber 1999, 153–66; Wang 1998).↩︎

  45. Hereafter, the differences are presented mainly based on the results of Weber’s research.↩︎

  46. After the publication of her study, Zieme identified an additional confession text in Old Uyghur (see Zieme 2001, 402–16). The title is [alku] kamag gaŋ ögüz i[čin]tä[ki] kum sanınča bügü biliglig burhan-lar y(a)rlık[amıš on] türlügin kılmıš ayıg kılınč-larıg ökünmäk [bilinmäk] yazok-da bošunmak nom bitig “The Sutra scripture preached by all the wise Buddhas, who are as numerous as the grains of sand in the Ganges River, on confessing the ten wicked deeds done and on being freed from sins.” This text does not mention concrete examples, but it rather explains Buddhist transgressions theoretically. Thus, the differences with the Manichaean ones are more obvious.↩︎

  47. The other Manichaean example, the poṭhi book, is too fragmentary to make such a list. There, however, the commandment seems to be the main topic, and it does not contain any long list of sins. The question of whether the differences between those two Manichaean confessions are caused by the production date or by the various users of the text remains unsolved (see Clark 2013, 112–20).↩︎

  48. See Weber (1999, 138–39). She also lists “preventing the spread of the teaching” as a sin. This sin is, however, common in many religions, so it is not necessarily confined to Buddhist and Manichaean teachings.↩︎

  49. Especially in Buddhist texts, the verb combination bil- and uk- is already used as a hendiadys, not only in confession texts but also in other texts translated in the early period (see, e.g., Şinasi Tekin 1980, vol. IX, vols. 2, 88a; Oda 2015, XXXIII:118, line 16, etc.).↩︎

  50. Most Buddhist colophons placed by the donors at the end of a copied or printed text were written following the same form, and the beginning formula is yemä (see, e.g., Zieme 1992, 48). Thus, this small difference can indicate that both Manichaeans and Buddhists did not pay much attention to each other’s form.↩︎

  51. This form is repeated many times. Some examples can be found in Clark (2013, 12, LC017, LC022, LC025–026).↩︎

  52. It is the case in the confession of Kutlug, her daughter, and her son (see, e.g., Müller and von Gabain 1910, 84, l. 8, 86, l.41).↩︎

  53. Both sentences appear with some variations (see Clark 2013, 14, LC061–063, 15, LC076–78, LC093–094, 16–17, LC121–123, et cetera; Müller and von Gabain 1910, 76, ll. 10–11, 77, ll. 19–21).↩︎

  54. See No. (2) mentioned above, which Weber also points out.↩︎

  55. This sentence is translated into German in Moriyasu (2004, 167).↩︎

  56. This fact is reported in the fragment M 112v (see, e.g., Moriyasu 2004, 175–76). The Buddhist cave, which was originally a Manichaean one, indicates the Uyghurs’ religious shift and the continuous use of such religious caves by the different communitie (see, e.g., Moriyasu 2004, 1–38). Whether this cave was destroyed by a certain Uyghur ruler’s order remains unclear.↩︎

  57. For the historical background of this hymn book, see, e.g., Moriyasu (2015e, 241–44).↩︎