Two Fragments of the Old Uyghur Saṃyuktāgama from the Berlin Turfan Collection
Department of Asian and African Studies, Humboldt University of Berlin, 10117 Berlin, Germany
Religions 2025, 16(4), 473; https://doi.org/10.3390/rel16040473
Submission received: 4 March 2025
/
Revised: 1 April 2025
/
Accepted: 3 April 2025
/
Published: 7 April 2025
(This article belongs to the Special Issue Reviving Ancient Wisdom on the Northern Silk Road: Research on Old Uyghur Buddhism and Buddhist Scriptures)
Abstract
:The Āgamas hold significant importance in the context of Old Uyghur Buddhism, as evidenced by the discovery of over 100 fragments of Old Uyghur translations from Turfan and Dunhuang. Of particular significance are the two fragments housed in the Berlin Turfan Collection, U 5464 and U 5501, which contain an Old Uyghur translation of the Chinese Saṃyuktāgama, inscribed on both sides of a repurposed Manichaean manuscript. These fragments correspond to the 22nd and 23rd fascicles of the Chinese source text. This study provides a comprehensive introduction and an edition of these two fragments, accompanied by a comparative analysis with a parallel Old Uyghur translation from Manuscript N, housed in Tokyo. The analysis reveals significant variations, particularly in translation strategies, suggesting the existence of multiple Old Uyghur versions of the text.
1. Introduction
The Saṃyuktāgama, or Za ahanjing 雜阿含經 (Takakusu et al. 1924–1932, Taishō, no. 99), is one of the four Āgamas, which are considered among the oldest Buddhist texts. In addition to the Chinese version, substantial parallel texts exist in the Pāli Saṃyutta Nikāya, unearthed Sanskrit fragments, and several Tibetan Buddhist translations (Anālayo 2020). To date, several fragments of the Old Uyghur translation of the Chinese 50-fascicle version of the Saṃyuktāgama have been identified.1 This study focuses on the publication of two fragments of the Old Uyghur Saṃyuktāgama translation, U 5464 and U 5501, housed in the Berlin Turfan Collection.
The recto of both fragments was identified as early as 1975 as an Old Uyghur recitation on the small fast day, Old Uyghur kičig bačag kün, of the Manichaean tradition (Zieme 1975, pp. 23–24).2 The verso of the fragments, as well as the space between the lines of the Manichaean text on the recto, were later utilized by Old Uyghur Buddhists to write the text, which was later recognized by Peter Zieme as a translation of the Saṃyuktāgama (Kasai et al. 2017, p. 29; Özertural 2021, pp. 243–44). However, the text on the verso, which continues with the interlinear text on the recto, remains to be thoroughly investigated.
This paper first examines the formal features of the text on the verso of U 5464 and U 5501, along with the interlinear text on the recto. These features are subsequently analyzed in terms of their content. Following this, an edition of the text is presented, accompanied by philological commentary. This study then explores the parallel Old Uyghur translation, highlighting different translation patterns. Finally, this paper situates these fragments within the broader corpus of Old Uyghur translations of the Saṃyuktāgama, considering their significance within the Old Uyghur Āgama literary tradition.
2. Formal Features and the Content
A thorough description of the physical characteristics and dimensions of the two fragments, U 5464 and U 5501, which will be discussed in this paper, can be found in Wilkens (2000, pp. 373–74), and therefore will not be repeated here. The verso of U 5464 (Figure 1) contains 15 lines of text, though lines 02 and 07–12 are heavily worn, rendering parts of the text blurred or illegible. The manuscript also exhibits three perforations: two affecting lines 09–11 and one located between lines 13 and 14.
The verso of U 5501 (Figure 2) comprises 19 lines of text, exhibiting similar characteristics to U 5464. The paper is worn and torn, with five perforations present in lines 03, 06, and 15, alongside several blurred sections.
On the rectos of U 5464 and U 5501 (Figure 3 and Figure 4), two lines of the Buddhist text are inscribed between every two lines of the Manichaean text, resulting in 13 and 17 lines of interlinear Buddhist texts, respectively. It is notable that the Buddhist interlinear text is oriented in the opposite direction to the Manichaean text.
The text of the two fragments is written in a cursive type of the Old Uyghur script and incorporates Chinese characters as part of the Old Uyghur text. The inclusion of Chinese characters is a characteristic feature of Old Uyghur Āgama texts, wherein phrases or individual characters from the original Chinese are quoted and subsequently followed by their corresponding Old Uyghur translation (Shōgaito 1998, pp. 363–64; 2003, p. 213; Kasai et al. 2017, p. 21). In both fragments, dots are used after certain Chinese characters to indicate the existence of omitted Chinese characters. For instance, in U 5464 verso, line 10, the Old Uyghur phrase olar ö[trü] tuy[unmıš] oḍunmıš-lar translates the Chinese bi ze wei juewu 彼則為覺悟; however, only the initial character bi 彼 is quoted, with the subsequent two dots indicating the omission of the remainder of the Chinese sentence ze wei juewu 則為覺悟. Conversely, in U 5501 verso, line 10, the quotation xia ye shi er geng 下夜始二更 is a complete sentence, and thus no dots follow. In certain instances, dots are employed to denote the precise number of omitted Chinese characters. For example, in U 5464 verso, lines 02 and 09, two Chinese characters are followed by three dots, which indicate the omitted three Chinese characters. This phenomenon is not exclusive to the context of the Āgama texts; it has also been observed in the Old Uyghur translation of the Qianziwen 千字文 Thousand Character Essay (Shōgaito and Yakup 2001, p. 4). In certain other instances, however, dots appear in isolation, without accompanying Chinese characters. For instance, in U 5464 verso, lines 03–07, only dots are present at the beginning of the Old Uyghur translation. These dots serve to indicate the omission of the quoted Chinese original phrases, which should have been written preceding the Old Uyghur translation.
In alignment with the characteristics of other Old Uyghur Āgama texts, the content of U 5464 and U 5501 is an excerpted translation of the Chinese text as opposed to a complete rendering (Shōgaito 1998, pp. 363–64; Kasai et al. 2017, p. 21). This can be illustrated with an example from U 5501 recto, line 16: […] yer-tä urup … kü[sä]yür [siz] körgäli alku […], “[…] laying down on the ground … (do) [you] wish to see all […]”. The corresponding Chinese passage reads: 王五體投地, 供養禮拜, 即立佛塔, 尊者白王言: 大王欲見諸天見佛生時行七步處不? (Taishō, no. 99, vol. 02, p. 166c4–6), which can be translated as “The king laid the five body parts (i.e., the whole body) on the ground, offering alms and worship, and immediately erected a stupa. The venerable one said to the king: ‘Great king, do you wish to see the place where the heavens witnessed the Buddha taking seven steps at his birth?’”. In this instance, the Old Uyghur translation omits several clauses, e.g., 供養禮拜, 即立佛塔, 尊者白王言: 大王 (“offering alms and worship, and immediately erected the stupa. The venerable one said to the king: ‘Great king’”).
The selection criteria of the Old Uyghur translation of Āgama texts remains unclear (Kasai et al. 2017, pp. 24–25). However, in the fragments under discussion, it is evident that the translation predominantly focuses on the śloka (verse) portions of the Chinese original. In the U 5464 verso, only the phrase kanglı kutusın from line 09 is from the Chinese prose, while the remaining translations are all derived from the ślokas of the sūtras. Similarly, in the U 5501 verso, of the 19 lines, the majority are from ślokas, with the exception of lines 02, 03, 09, 10, and 11. Likewise, in the U 5501 recto, out of the 17 lines, only lines 05, 13, 14, 16, and 17 are derived from the prose sections of the Chinese text. The U 5464 recto presents a slightly different pattern, as lines 02–06, 12, and 13 originate from prose, accounting for approximately half of the entire text.
The analysis of the Manichaean text on the recto of the two fragments indicates that both fragments, U 5464 and U 5501, were originally part of a single manuscript, and that they constitute a continuous text (Zieme 1975, p. 23; Wilkens 2000, pp. 373–74; Clark 2013, p. 267). Consequently, fragments U 5464 and U 5501 are regarded as parts of the same text. This hypothesis is further substantiated by an examination of the Buddhist text on both fragments, namely the Old Uyghur translation of the Saṃyuktāgama.
The Old Uyghur text on the verso of the two fragments corresponds to the Chinese Saṃyuktāgama, covering sūtras 588 to 594 from the 22nd fascicle (Taishō, no. 99, vol. 02, pp. 156a29–159b03). While some of these sūtras have parallels in the Pāli Saṃyutta Nikāya, the Old Uyghur translation cannot be directly compared with the Pāli version, as it was translated from the Chinese text, and significant differences in content exist between the Chinese and Pāli versions.3 Conversely, the interlinear text on the recto corresponds to sūtra 604 from the 23rd fascicle (Taishō, no. 99, vol. 02, pp. 161b13–170c20). Notably, sūtra 604 of the Chinese Saṃyuktāgama is actually the Aśoka-Avadāna, which was not originally part of the Saṃyuktāgama. It is assumed that the Chinese Aśoka-Avadāna was later inserted to fill the gap resulting from the loss of the original 23rd fascicle (see Chung 2008, p. 24). Consequently, no parallel can be found in the Pāli Saṃyutta Nikāya.
The sequence of the Chinese sūtras indicates that the Old Uyghur content on the verso precedes that on the recto. This arrangement suggests that the Old Uyghurs initially used the blank verso side of the discarded Manichaean text to transcribe their translation of the Saṃyuktāgama, subsequently employing the recto side.
3. Edition of the Text
3.1. Transcription with Chinese Parallel Passages
U 5464 verso
| |
| [Sūtra 588] 156b8 (正向)於彼處。(…) [Sūtra 589] b18賴吒(槃提國)。(…) |
| b19(大富)足財寶。各各競求富。(…) |
| b20猶如然熾火。b21如是競(勝心)。(…) |
| b22云何當斷貪。息(世間勤求)。(…) |
| b24妻子及財寶。(…) |
| b26正智心解脫。愛(盡息方便)。(…) |
| |
| [Sūtra 590] (c12遍照商人)c13車營。(…) c14誰於覺睡眠。(…) |
| 157a10彼則為覺悟。我(於彼睡眠)。(…) |
| a11(及苦)因緣集。a12(於)此(一切苦)。(…) |
| a13(等趣)息苦處。(…) |
| a15如是覺睡眠。(…) a18善哉(覺睡眠)。(…) |
| a20諸兄弟而來。a21(緣)汝恩(力故)。(…) |
| a22(隨)道安樂(去)。(…) |
U 5501 verso
| |
| [Sūtra 591]157a28(極)毀呰欲。a29言：此(欲者)(…) |
| a30還(自己舍，恣於)五欲。(…) |
| b6於大聚會中。(…) |
| b7(自)沒於愛欲。如牛(溺深泥)。(…) |
| b9多聞明(解法)。(…) |
| b10(汝見彼)樂法。而說欲(無常)。(…) |
| b12何故樂世間。(…) |
| b13(開覺彼)優婆塞已。如b14是如是(彼優婆塞覺悟已)。(…) |
| [Sūtra 592]c14(至c15城門)下，夜始二更。(…) |
| c21(或姦狡人)恐c22怖(我耶?) (…) |
| 158a6(巨身長)大牙(…) [Sūtra 593]c2於此(祇桓林)。(…) |
| c3(諸王亦)住此。(增)我(歡喜心)。(…) |
| c4智慧為勝壽。c5以此(淨眾生)。(…) |
| c6正念常寂默。(…) |
| c7初建業良友。(…) [Sūtra 594]159a5天身(委地)。(…) |
| |
| a29受持(賢聖法)。(…) |
| b1(故)生(無熱天)。(…) |
The interlinear text on the recto of U 5501
| |
| [Sūtra 604]166a9欲得富貴(者)。(…) |
| a10(常處天上)樂。解脫(涅槃者)。(…) |
| a11(修敬今)供養。a12(未見諸)佛者。(…) |
| a19(寄在)國王，a20及我比丘僧(等)。(…) |
| a21(世雄)人中尊。正勝(妙大法)。(…) |
| a22及我比丘僧。(…) |
| a24(莊嚴諸國)界。(…) a27我興(如是福)。(…) |
| a28(珍寶及此)地。a29今已悉(捨施)。(…) |
| b3(捨身)及財命。(…) b9(我行)布施(時)。(…) |
| b10以沙施於佛。(…) |
| b15(布施諸沙)土。b16無上福田中。(…) |
| b20為(諸後世眾生)攝受(善根)。(…) |
| b26(便將b27尊者)發去。尊者隆頻林。(…) |
| b29(如來)初生處。(…) c1顧視(諸四方)。(…) |
| c4(王五體)投地。(…) c5(大王)欲見諸(天見佛生時行七步c6處不)? (…) |
| c7(指摩耶夫人所攀)樹枝，(而)告(彼樹神曰) (…) |
The interlinear text on the recto of U 5464
| c12生時青蓮華。(…) |
| c23此處菩薩現(三十二c24相)。(…) (莊嚴)其體。(…) |
| c29瞻其相(德)。(…) |
| 167a2(此處學乘馬、)乘車、弓弩。(…) |
| a10授與於車匿。(…) a12易袈裟衣。(…) |
| a13此處瓶a14沙王與菩薩半(國處)。(…) |
| a22讚歎諸菩薩。a23當(隨)古時道。(…) |
| a25我今欲見龍。(…) |
| a26證得勝妙果。(…) |
| b3(我師)無疇匹。b4(面)如淨滿月。(…) |
| b5往詣(道場)時。(…)b16天人(中特尊)。(…) |
| b20(此b21處)仙人園鹿野苑。(如來於中為)五比丘(三轉b22十二行法輪)。(…) |
| (c2及)c3立(塔廟)。(…) |
3.2. Translation
U 5464 verso (from sūtra 588 and sūtra 589):
(2) … saying, “… to (that) place”. … (a country) called(?) Lai Zha [Pan Ti] (Rāṣṭravati) … (3) … They (were) [wealthy and be] jeweled. Each one competing [to become rich] … (4) … like a mighty fire burning. Such a c[ompetitive (mind)] … (5) … How can they eliminate greed, and quiet (the world’s relentless pursuit?) … (6) … (He abandoned) his [wi]fe, his child, his property, and his jewels … (7) … If, with [upright] wisdom, [his] mind [escapes] and is freed, (that is the method (upāya) to eliminate) desire … (8) … on the road … the pure ve[hicle(?)] called āryayāna(?) ….
(from sūtra 590)
(9) … (He emitted radiant lights, illuminating the merchants’) vehicles’ field. (Who) [sleeps i]n [aw]akeness? … (10) … They then [achieved en]lightenment and awakened, I (fell asleep on that) … (11) … (I did not understand) the ga[thering] and assembly of [rea]son and cause. (On all) the[se] (sufferings) … (12) … (I did not understand either, that the right paths equally lead) toward the [pla]ce where sufferings are extinguished … (13) … In this way, [I] sleep in awakeness … (14) … The com[ings of all the younger] and elder brothers. (Due to) your goodness, (all the merchants are spared from the thieves) … (15) … (they left peacefully and) happily (following) the road …
U 5501 verso (from sūtra 591):
(2) … (They) despised this desire, saying: “This (desire) …” … (3) … [When(?)] returning (to his own house, he indulged) in the five de[sires] … (4) … When gathering together with the great assembly … (5) … (He himself) f[ell] into the desires, like a cattle sinking (into deep mud) … (6) … (The upāsakas) listen much to and have the wisdom of (the Dharma)… (7) … (You see) those who [rejoice] (in the Dharma), th[en] (you) say that the desires (are impermanent) … (8) … Why (you) love the world … (9) … when (the deity) finished (enlightening) [the upāsak]a, also like that (when the upāsaka finished attaining the enlightenment) …
(from sūtra 592)
(10) … the latter part of the night, only the second watch … (11) … (or could it be that some deceitful and treacherous person is trying) to frighten and to [horri]fy (me?)… (12) … (The elephant) has (a huge body and long and) [la]rge tusks …
(from sūtra 593)
(12) … (In) this (Jetavana) … (13) … by living here, my (heart has grown in joy) … (14) … [Wis]dom being(?) the excellent life, with this it pu[rifies] (all beings) … (15) … With the upright intelligible (mind), (Śāriputra) is [al]ways calm and peace … (16) … good fri[end] …
(from sūtra 594)
(16) … the heavenly (body laid on the ground) … (18) … obtaining and upholding the n[oble] (Dharma) … (19) … (Therefore I) was born (in the Atapa) …
U 5501 recto (from sūtra 604):
(2) … If you wish for wealth … (3) … (If you desire to always dwell) in the happiness on (the divine land), (if you seek) liberation (and nirvāṇa) … (4) … (practicing reverence now) on [wo]rship and devotion. Who (has not yet seen) the divine Buddha … (5) … (My upright Dharma is entrusted to) the kings and also [my] community of monks … (6) … (The Buddha is) the honored one among [humans]. The upright and nobl[e] (Dharma) … (7) … (He entrusted it to the great kings) and [my] community of monks … (8) … (I have decorated) the area of (all the country). I have increased (such merits) … (9) … (My treasures and this) land, (I have) now all (donated) … (10) … (abandoning his body), his property and his life … (When I) practiced generosity … (11) … By offering alms of [sa]nd to the divine Bu[ddha] … (12) … (offering various sands and) [ea]rth. [In the fi]eld of [su]preme merit … (13) … For holding (the roots of goodness for the benefit of future beings) … (14) … They [mo]ve and go. The venerable one (arrives at) the Lumbi[nīvana Forest] … (15) … (This) is the place where (the Buddha) was first born. He turned round and gazed (in all four directions) … (16) … (The king) laying down on the ground. (Great king! Do you) wish to witness (the place where) all (the heavens saw the Buddha taking seven steps at his birth?) … (17) … (The venerable one raises his hand, pointing to) the tree branch (where Māyā had grasped), says …
U 5464 recto:
(1) … When (the Buddha) was [bo]rn, (he was) like a blue lotus … (2) … At [this place], the Bodhisattva showed (the thirty-two marks and the eighty beautiful features). (He decorated) his body, and … (3) … he says … he shows his (virtues) [of si]gns … (4) … (This is the place where the Bodhisattva learned) to ride (horses and chariots), to shoot the crossbow and bow … (5) … He gave to Chandaka. He changed into [kā]ṣāya cloth … (6) … This place (is where) the King Bimbasāra (gave) half (of the country) to the Bodhisattva … (7) … (This is the place where Kālika) praised the Bodhisattva. (One should follow) the previous w[ay] … (8) … [My] current wish is to see the Nāga … (9) … (He attained) the excellent and wonderful fruit … (10) … (My teacher is) unparalleled. (He is) l[ike] the full moon … (11) … At the time when (he) went … (The most respected) among the sages … (12) … (This place is) the sages’ park Ṛṣivadana, (where the Buddha turned the Wheel of Dharma three times for) the five disciples … (13) … (The king) established and ere[cted] (stupas and temples) …
3.3. Commentary
U 5464 verso
v04 türüklüg is written here as twyrwklwk, which is not attested in any Old Turkic dictionaries. The Old Uyghur phrase [inčä kalt]ı türüklüg oot yakar täg renders the Chinese youru ran chi huo 猶如然熾火 “like a burning fire”, indicating that türüklüg functions as an attributive adjective modifying oot “fire”. I suggest that this form may be a misspelling of türklüg, meaning “powerful, mighty” (Wilkens 2021, p. 775).
v08 The Chinese source text of this line remains uncertain. yolta “on the road” may correspond to the Chinese phrase zhi zhu lu ce 止住路側 “(the deity) was residing by the roadside”. On the other hand, yolta, being the locative form of yol “way, road”, could alternatively be read as the accusative form yolug. However, as the source text does not have the verb governing a direct object, the reading remains uncertain.
The reading of aryayan(?) in the same line remains highly uncertain due to the blurred writing in the manuscript and the absence of a corresponding term in the Chinese source text. It might come from Sanskrit āryayāna “noble vehicle” (see Wilkens 2021, pp. 70–71). Another possibility is to read it as aryavan, which is derived from Sanskrit airāvaṇa via the intermediate form of Tocharian A airāvaṃ, which is the name of Indra’s elephant (see Wilkens 2021, p. 70).
U 5501 verso
v02 […] münäp bo amranmak-ıg “[…] despising this desire” corresponds to the Chinese phrase ji huizi yu 極毀呰欲 “(the upāsaka) extremely despised desires” found in the source text. Notably, the Old Uyghur bo “this” appears to have been added by the translator, as there is no direct parallel in the Taishō edition. Three possible hypotheses may account for this addition. First, the Chinese text used by the Old Uyghur translator may have differed from the Taishō edition, possibly containing the character ci 此 “this”, which was then rendered as bo in the translation. Second, the translator may have been influenced by the subsequent phrase ci yu zhe 此欲者 “as for this desire” in the Chinese text, prompting the insertion of bo for consistency. Third, it is conceivable that the translator misread huizi 毁呰 “to depise” as hui ci 毁此 “to despise this”, given the visual similarity between the characters zi 呰 and ci 此.
v03 yantuk[ta(?) …] yanturu(?) beš t[apta …] ”when(?) returning … returning(?) … on five desires …” is likely a translation of the Chinese huan ziji she, zi yu wu yu 還自己舍, 恣於五欲 (Taishō, no. 99, vol. 02, p. 157a30) “(the upāsaka) returned to his own house, and indulged in the five desires” or huan ji she yi, zi zi wuyu 還己舍已, 自恣五欲 (Taishō, no. 99, vol. 02, p. 157b4) “after (the upāsaka) returned to his own house, he himself indulged in the five desires”. Both Chinese phrases occur within this passage.
v06 bilgä bilig{ig}-lig can be translated as “of wisdom”. The spelling of the second word is pylykyk lyk. Obviously, it is mistaken for pylyk lyk. bilgä biliglig translates here the Chinese term mingjie 明解 “skilled, learned in” (DDB n.d., entry 明解).
v10 The quoted Chinese phrase xia ye shi ergeng 下夜始二更 likely results from incorrect sentence punctuation. In the Taishō edition, the original Chinese text reads zhi chengmen xia, ye shi er geng 至城門下, 夜始二更 (Taishō, no. 99, vol. 02, p. 157c14–15), meaning “when he reached the city gate, it was just the second watch (Chinese geng 更) of night”. Here, xia 下 belongs to the preceding phrase. Notably, Manuscript N, which contains parallel passages, also quotes the same mispunctuated Chinese phrase (Shōgaito 1982, p. 147; 2003, p. 296).
U 5501 recto
r05 bursang comes from Sogdian pwrsnk “monk”, Old Uyghur has two common forms, bursang and bursong (Wilkens 2023, pp. 67–71). Both variants occur in the present text (see U 5501 recto, line 05 and 07).
U 5464 recto
r03 The verb körgit “to show” is used here to render the Chinese zhan 瞻 “to look up, to look with reverence”, which does not correspond precisely in meaning. The Chinese phrase cichu fuwang yi pusa shi zhu poluomen, zhan qi xiangde 此處父王以菩薩示諸婆羅門，瞻其相德 translates as “at this place, the father king presented the Bodhisattva to the Brahmins, who looked up to his virtue of sign (lakṣaṇaguṇa)”. In this construction, the Brahmins are the subject of the second clause. However, in the Old Uyghur translation, the shift in verb usage suggests that the translator may have reinterpreted the Bodhisattva as the subject, i.e., “the Bodhisattva showed his virtue of sign (to the Brahmins)”.
r04 lukung is derived from the Chinese nu gong 弩弓 “crossbow and bow”, for an analysis, see Wilkens (2017, p. 224).
4. Parallel Old Uyghur Translations
There is partial overlap in the content between the text on the verso sides of the fragments U 5464 and U 5501 and the Old Uyghur Saṃyuktāgama manuscript (Manuscript N), housed at the Museum of Calligraphy in Tokyo. Manuscript N is a long scroll comprising 373 lines and containing the Old Uyghur translation of the sūtras 559 to 603. Manuscript N was investigated by Masahiro Shōgaito (1982, pp. 121–89; 2003, pp. 261–309). A comparison between U 5464 and U 5501 and Manuscript N reveals differences in the selection of source texts for translation. For instance, U 5501 verso lines 2–9 contain the Old Uyghur translation of sūtra 591 (Taishō, no. 99, vol. 02, 157a26–b17), which is absent in Manuscript N (Shōgaito 2003, p. 294). Furthermore, there are notable discrepancies in the translation of identical Chinese sentences. Table 1 highlights the differences in translation between U 5464, U 5501, and Manuscript N.
The observed differences between the texts can be summarized as follows. First, the translators employed different lexical choices to render the same Chinese terms. For example, in U 5464 verso line 11, the Chinese term yinyuan 因緣 “reason, cause” is translated as tıltag basutčı, whereas in Manuscript N, it appears as avant tıltag. Similarly, in U 5464 verso line 12, the Chinese verb xi 息 “to extinguish” is translated as öčür-, while Manuscript N uses amırtgur-.
Second, discrepancies in word order are evident. For instance, the Chinese phrase yun he dang duan tan 云何當斷貪 “how can (they) eliminate greed” is rendered in U 5464 verso line 05 as nätägin inčip tarkarıp azıg, whereas in Manuscript N, it appears as nätägin inčip azıg tarkarıp. The difference lies in the position of the verb tarkar- “to eliminate” and its direct object az “the greed”, suggesting that U 5464 may have been influenced by the word order of the Chinese original, which places the verb earlier in the sentence.
In general, most of these differences do not alter the overall meaning of the text, even though some variations appear quite distinct. For example, the Chinese phrase ran chi huo 然熾火 “burning fire” is translated in U 5464 verso line 04 as türüklüg oot yakar “the mighty fire burns”, while in Manuscript N, it appears as örtänü yalınayu turur oot “the fire which is burning”. However, certain discrepancies suggest a different understanding of the Chinese text by the Old Uyghur translator. For instance, the Chinese term mingjie 明解 “skilled, learned in” is accurately rendered in U 5501 verso line 06 (see the commentary in the edition section). In contrast, the translation yar[ok …] in Manuscript N appears incorrect, as the translator seems to have misinterpreted ming 明 as ”light”.
A similar pattern can be observed in U 5402 and U 5473, two additional fragments of the Old Uyghur Saṃyuktāgama. The first was published by Kudara and Zieme (1983, pp. 296–01), while the second was examined by Zieme (2000). These fragments also exhibit partial overlap with the content of Manuscript N. Masahiro Shōgaito’s comparative analysis of these texts revealed significant differences in sentence structure and phrasing, thereby suggesting the existence of variations in the Old Uyghur translations of the Saṃyuktāgama (Shōgaito 1982, p. 103). The findings of the present study further reinforce Shōgaito’s hypothesis, providing additional evidence for the presence of multiple versions of the Old Uyghur translation of the Saṃyuktāgama.
This raises the following question: why were multiple translation versions of the text produced? One possible explanation is that the translation underwent multiple revisions by Old Uyghur Buddhists. The translation of the Chinese term mingjie 明解, which, as aforementioned, is accurate in U 5501 but incorrect in Manuscript N, may serve as evidence of this process. However, another possibility is that the different versions are not directly related. The hypothesis proposed by Aydar Mirkamal, which suggests that the Old Uyghur Āgama texts were used as materials for learning the translation method from Chinese into Old Uyghur (see Mirkamal 2014, pp. 82–85), should also be considered. If these texts were indeed intended for translation practices by different monks, the significant discrepancies among different versions would not be surprising.
5. Conclusions
The two Old Uyghur fragments, U 5464 and U 5501, from the Berlin Turfan Collection provide valuable insights into the Old Uyghur Buddhist translation tradition, particularly concerning the Saṃyuktāgama corpus. The text of these two fragments is translated from the Chinese 50-fascicle version of the Saṃyuktāgama, specifically sūtras 588–594 of the 22nd fascicle and sūtra 604 of the 23rd fascicle. The juxtaposition of the Old Uyghur text with its Chinese parallel passages facilitates a clear understanding of their textual relationship, revealing a high degree of correspondence between the two.
Textual analysis also confirms that these two fragments originate from the same manuscript, supporting previous hypotheses that were initially based on observations of paper and handwriting, primarily in the context of the Manichaean text. The textual connections demonstrated in this study further strengthen these claims.
This study identifies several key characteristics shared by these fragments and other Old Uyghur Āgama translations, including the selective excerption of the Chinese source text and the incorporation of Chinese character quotations. Additionally, it highlights the use of dots as omission markers, a feature known from other texts but identified here for the first time in Old Uyghur Āgama manuscripts.
Although the corpus of Old Uyghur Āgamas is relatively extensive, the number of parallel translations is limited, leaving uncertainty regarding the existence of different versions. The fragments U 5464 and U 5501 provide new evidence for exploring this issue. A comparative analysis of these fragments with parallel translations from Manuscript N reveals significant variations in translation strategies, particularly in excerpt selections, word choices, and sentence structures. These differences suggest the existence of multiple Old Uyghur versions of the Saṃyuktāgama.
Funding
This research received no external funding.
Data Availability Statement
No new data were created or analyzed in this study. Data sharing is not applicable to this article.
Conflicts of Interest
The author declares no conflict of interest.
| 1 | The Old Uyghur Saṃyuktāgama texts have been summarized in previous scholarship, notably by Elverskog (1997, pp. 20–22) and Mirkamal (2010, pp. 1–2). A comprehensive inventory of these texts was compiled in 2017 (see Kasai et al. 2017, pp. 28–29). More recently, P. Ouïgour 215 from the Pelliot collection in Paris, previously identified by Kōgi Kudara as an Old Uyghur translation of the Saṃyuktāgama, was published by Aydar Mirkamal and Yang Fuxue, see (Mirkamal and Yang 2022, pp. 188, 191–92, 195–96). |
| 2 | A detailed summary of prior researches and a new edition of the Manichaean text can be found in Clark (2013, pp. 267–69). |
| 3 | According to the online database (SuttaCentral n.d.), the correspondences between the sūtras 588–594 of the Chinese Saṃyuktāgama (SA) and their parallels in the Pāli Saṃyutta Nikāya (SN) are as follows: SA 588 = SN 1.29, SN 2.28; SA 589 = SN 1.28; SA 592 = SN 10.8; SA 593 = SN 2.20, SN 1.48, SN 55.26. Notably, some of these Chinese sūtras also have parallels in the Chinese Bieyi za ahanjing 別譯雜阿含經 ‘Additional Translation of the Saṃyuktāgama’ (Taishō, vol. 02, No. 100). The Old Uyghur Buddhists also translated this version into Old Uyghur (see Kasai et al. 2017, pp. 29–30), however, the texts examined in this study are not related to this version from a translation perspective. |
References
- Anālayo. 2020. Āgama/Nikāya. In Brill’s Encyclopedia of Buddhism Online. Leiden: Brill. [Google Scholar] [CrossRef]
- Chung, Jin-il. 2008. A Survey of the Sanskrit Fragments Corresponding to the Chinese Saṃyuktāgama. Zōagonkyō sōtō bonbun danpen ichiran 雜阿含經相當梵文斷片一覽. Tokyo: Sankibō Busshorin 山喜房佛書林. [Google Scholar]
- Clark, Larry. 2013. Uygur Manichaean Texts. Volume II. Liturgical Texts. Corpus Fontium Manichaeorum. Series Turcica 2. Turnhout: Brepols. [Google Scholar]
- DDB. n.d. DDB = Digital Dictionary of Buddhism. Available online: http://www.buddhism-dict.net/ddb/ (accessed on 11 February 2025).
- Elverskog, Johan. 1997. Uygur Buddhist Literature. (Silk Road Studies I). Turnhout: Brepols. [Google Scholar]
- Kasai, Yukio, Simone-Christiane Raschmann, Håkan Wahlquist, and Peter Zieme, eds. 2017. The Old Uyghur Āgama Fragments Preserved in the Sven Hedin Collection, Stockholm. Silk Road Studies XV. Turnhout: Brepols. [Google Scholar]
- Kudara, Kōgi, and Peter Zieme. 1983. Uigurische Āgama-Fragmente (1). Altorientalische Forschung 10: 269–18. [Google Scholar]
- Mirkamal, Aydar. 2010. Pulinsidun Daxue Tushuguan Cang Huihuwen “Zhong ahanjing” Yanjiu 普林斯頓大學圖書館藏回鶻文《中阿含經》研究 [The Uighur Madhyama Āgama Fragment Preserved in the Gest Library Collection of Princeton University Library]. Kyoto University Linguistic Research 29: 1–26. [Google Scholar]
- Mirkamal, Aydar. 2014. Huihu Sengren Ruhe Xuexi Hanyu 回鶻僧人如何學習漢語 [How did Uyghur monks learn Chinese]. Xinjiang Shifan Daxue Xuebao (Zhexue Shehui Kexue Ban) 新疆師範大學學報 (哲學社會科學版) [Journal of the Xinjiang Normal University (Philosophy and Social Sciences Edition)] 35–1: 81–86. [Google Scholar]
- Mirkamal, Aydar 阿依達爾·米爾卡馬力, and Yang Fuxue 楊富學. 2022. Wujian Huihuwen “Ahanjing” Xieben Canpian ji Bawen Yishi 五件回鶻文《阿含經》寫本殘片及跋文譯釋 [Studies on Five Old Uyghur Fragments of Āgama]. Xiyu Wenshi 西域文史 [Literature and History of the Western Regions] 16: 187–97. [Google Scholar]
- Özertural, Zekine. 2021. Alttürkische Handschriften, Teil 4: Varia Buddhica: Buddhistische Gedichte und Kleinere Sūtra-Texte. [VOHD 13, 12]. Stuttgart: Franz Steiner Verlag. [Google Scholar]
- Shōgaito, Masahiro 庄垣内正弘. 1982. Uigurugo, Uigurugo Bunken no Kenkyū. I. “Kanonkyō ni Husawashii Sanpen no Avadāna” oyobi “Agonkyō” ni tsuiteウイグル語·ウイグル語文献の研究Ⅰ『觀音經に相応しい三篇のAvadāna』及び『阿含經』について [On the Two Buddhist Uygur Texts: With Special Reference to the Three Avadānas Suitable to Avalokiteśvara-sūtra and Āgama-sūtra]. Kobe: Kobe City University of Foreign Studies. [Google Scholar]
- Shōgaito, Masahiro. 1998. Three Fragments of Uighur Āgama. In Bahşı Ögdisi. Festschrift für Klaus Röhrborn anläßlich seines 60. Geburtstags. Edited by Jens Peter Laut and Mehmet Ölmez. Türk Dilleri Araştırmaları Dizisi 21. Istanbul: Simurg, pp. 363–78. [Google Scholar]
- Shōgaito, Masahiro 庄垣内正弘. 2003. Roshia Shozō Uigurugo Bunken no Kenkyū-Uigurumoji Hyōki Kanbun to Uigurugo Butten Tekisutoロシア所藏ウイグル語文獻の研究-ウイグル文字表記漢文とウイグル語仏典テキスト [Uighur Manuscripts in St. Petersburg. Chinese Texts in Uighur Script and Buddhist Uighur Texts]. Kyoto: Kyoto University. [Google Scholar]
- Shōgaito, Masahiro, and Abdurishid Yakup. 2001. Four Uyghur Fragments of Qianziwen ‘Thousand Character Essay’. Turkic Languages 5–1: 3–29. [Google Scholar]
- SuttaCentral. n.d. SuttaCentral = SuttaCentral. Early Buddhist Texts, Translations, and Parallels. Available online: https://suttacentral.net/sa501-600?view=dense&lang=en (accessed on 15 March 2025).
- Takakusu, Junjirō 高楠順次郎, Kaikyoku Watanabe 渡辺海旭, and Genmyō Ono 小野玄妙, eds. 1924–1932. Taishō shinshū daizōkyō 大正新脩大蔵経 [The Taishō New Edition of the Buddhist Canon]. The Chinese Buddhist Electronic Text Association (CBETA) edition. Tokyo: Taishō issaikyō kankokai. [Google Scholar]
- Wilkens, Jens. 2000. Alttürkische Handschriften. Teil 8: Manichäisch-Türkische Texte der Berliner Turfansammlung. [VOHD 13, 16]. Stuttgart: Franz Steiner Verlag. [Google Scholar]
- Wilkens, Jens. 2017. The Old Uyghur Translation of the Bieyi za ahanjing (1935.52.0012-0015). In The Old Uyghur Āgama Fragments Preserved in the Sven Hedin Collection. Turnhout: Brepols, pp. 197–280. [Google Scholar]
- Wilkens, Jens. 2021. Handwörterbuch des Altuigurischen. Altuigurisch-Deutsch-Türkisch. Göttingen: Akademie der Wissenschaften zu Göttingen. [Google Scholar]
- Wilkens, Jens. 2023. Uigurisches Wörterbuch. Sprachmaterial der vorislamischen türkischen Texte aus Zentralasien. III. Fremdelemente. Band 2: Bodivan-čigžin. Stuttgart: Franz Steiner Verlag. [Google Scholar]
- Zieme, Peter. 1975. Manichäisch-türkische Texte. Berliner Turfantexte 5. Berlin: Akademie Verlag. [Google Scholar]
- Zieme, Peter. 2000. Verse des Candrasūtra nach chinesisch-uigurischen Bilinguen. Türk Dilleri Araştırmaları 10: 65–80. [Google Scholar]
Figure 1.
U 5464 (T I D 615) verso © Depositum der BERLIN-BRANDENBURGISCHEN AKADEMIE DER WISSENSCHAFTEN in der STAATSBIBLIOTHEK ZU BERLIN-Preußicher Kurturbesitz Orientabteilung.
Figure 1.
U 5464 (T I D 615) verso © Depositum der BERLIN-BRANDENBURGISCHEN AKADEMIE DER WISSENSCHAFTEN in der STAATSBIBLIOTHEK ZU BERLIN-Preußicher Kurturbesitz Orientabteilung.
Figure 2.
U 5501 (T II 894) verso © Depositum der BERLIN-BRANDENBURGISCHEN AKADEMIE DER WISSENSCHAFTEN in der STAATSBIBLIOTHEK ZU BERLIN-Preußicher Kurturbesitz Orientabteilung.
Figure 2.
U 5501 (T II 894) verso © Depositum der BERLIN-BRANDENBURGISCHEN AKADEMIE DER WISSENSCHAFTEN in der STAATSBIBLIOTHEK ZU BERLIN-Preußicher Kurturbesitz Orientabteilung.
Figure 3.
U 5464 (T I D 615) recto © Depositum der BERLIN-BRANDENBURGISCHEN AKADEMIE DER WISSENSCHAFTEN in der STAATSBIBLIOTHEK ZU BERLIN-Preußicher Kurturbesitz Orientabteilung.
Figure 3.
U 5464 (T I D 615) recto © Depositum der BERLIN-BRANDENBURGISCHEN AKADEMIE DER WISSENSCHAFTEN in der STAATSBIBLIOTHEK ZU BERLIN-Preußicher Kurturbesitz Orientabteilung.
Figure 4.
U 5501 (T II 894) recto © Depositum der BERLIN-BRANDENBURGISCHEN AKADEMIE DER WISSENSCHAFTEN in der STAATSBIBLIOTHEK ZU BERLIN-Preußicher Kurturbesitz Orientabteilung.
Figure 4.
U 5501 (T II 894) recto © Depositum der BERLIN-BRANDENBURGISCHEN AKADEMIE DER WISSENSCHAFTEN in der STAATSBIBLIOTHEK ZU BERLIN-Preußicher Kurturbesitz Orientabteilung.
Table 1.
Variations in translation between U 5464, U 5501, and Manuscript N.
| Chinese | U 5464 and U 5501 | Manuscript N |
|---|---|---|
| 各各競求富 | U 5464 v03öngin öngin yar(ı)šmalašu b[ay barımlıg bolmakıg] | 243öngin öngin y(a)rıšmalašu tiläp bay barımlıg bolmakıg |
| 猶如然熾火 | v04[inčä kalt]ı türüklüg oot yakar täg | 244inčä kaltı örtänü yalınayu turur oot täg |
| 云何當斷貪 | v05nätägin inčip tarkarıp azıg | 246nätägin inčip azıg tarkarıp |
| 正智心解脫 | v07[köni] bilgä biligin köngü[li ozsa]r kutrulsar | 249köni bilgä biligin ozup kutrul[…] |
| 車營 | v09kanglı kutuṣın | 252kanglılıg kutuların |
| 誰於覺睡眠 | v09[… o]ḍug-lar-t[a udır] | 254kim siz oḍug-lar-nıng arasınta udır |
| 彼則為覺悟 | v10olar ö[trü] tuy[unmıš] oḍunmıš-lar | 258olar ötrü ärür oḍ[…] |
| 及苦因緣集 | [… v11tıl]tag basutčı y[ıgıltačı] teriltäči-sin | 260ulatı ämgäk-ning avant-lar-tın tıltag-lar-tın terilür-in |
| 等趣息苦處 | […] v12ämgäkin öčürgülük [oron]-ka […] | […] 262ämgäkin amırtgurguluk oron-ka barmak-ıg ymä |
| 如是覺睡眠 | v13muntag-ın oḍug-lar-ta uḍır [män] | 263muntagın odug-lar-nıng arasınta udır m(ä)n |
| 隨道安樂去 | v15yol […] mängi[lig …] | […] 266mängiligin yol eyin bargay-lar |
| 如牛溺深泥 | U 5501 v05kaltı ud čomu[k …] | 270inčä kaltı täring titig-tä čomukmıš ud täg |
| 多聞明解法 | v06üküš äšidm[äk]-lig bilgä bilig-lig […] | 271üküš äšidmäk-lig yar[ok …] |
| 而說欲無常 | v07in[čip s]özläp amranmak-[…] | 273inčip sözläyür ayıp amranmak-nıng ürlügsüzin |
| 智慧為勝壽 | [… bilgä] v14bilig yeg adrok öz yaš ärip(?) | 308bilgä biliglig yeg adrok özlüg yaš-lıg-lar-ta |
| 正念常寂默 | v15köni öglügin [uz]aṭı öčüp amrılıp | 310köni ögin uzaṭı amırtgurdı |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Sun, B. Two Fragments of the Old Uyghur Saṃyuktāgama from the Berlin Turfan Collection. Religions 2025, 16, 473. https://doi.org/10.3390/rel16040473
AMA Style
Sun B. Two Fragments of the Old Uyghur Saṃyuktāgama from the Berlin Turfan Collection. Religions. 2025; 16(4):473. https://doi.org/10.3390/rel16040473
Chicago/Turabian StyleSun, Binghan. 2025. "Two Fragments of the Old Uyghur Saṃyuktāgama from the Berlin Turfan Collection" Religions 16, no. 4: 473. https://doi.org/10.3390/rel16040473
APA StyleSun, B. (2025). Two Fragments of the Old Uyghur Saṃyuktāgama from the Berlin Turfan Collection. Religions, 16(4), 473. https://doi.org/10.3390/rel16040473
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.
