Tracing the Incorporation of the Bimo shi Mulian jing into the Chinese Tripitaka and the Attribution of Its Translators: A Study Based on Buddhist Catalogs

Abstract

The reliable corpus of Buddhist sutras translated by Zhi Qian 支謙 serves as an important reference benchmark for determining the authenticity of Buddhist sutras from the Three Kingdoms 三國 period to the pre-Jin period (220–265 CE). The Bimo shi Mulian jing 弊魔試目連經 (The Sūtra of Māra Testing Maudgalyāyana) is currently included in the Taishō Tripiṭaka as an individual sutra. Since the start of block-printing of Buddhist canons, this sutra has been attributed to Zhi Qian of the Wu 吳 State in the Three Kingdoms period and included in the ruzangmu 入藏目 (“list [of texts] admitted to the canon”) of various editions of the Tripitaka. However, historical investigation reveals significant complexity and controversy surrounding its title, attributed translator, and its entries in different ancient catalogs. A systematic examination of historical Buddhist catalogs (jinglu 經錄) demonstrates that, during the times of Dao’an 道安 and Sengyou 僧祐, the sutra was given different names and recorded as a scripture with an unknown translator. During the time of Fajing 法經 in the Sui 隋 Dynasty, the sutra first appeared in the annotations of the sutra catalog under the name Bimo shi Mulian jing, and the translator was not recorded. By the time of Fei Zhangfang 費長房 in the Sui Dynasty, the sutra was first attributed to Zhi Qian, yet it was not included in the ruzangmu 入藏目. Finally, Zhisheng 智昇 in the Tang 唐 Dynasty integrated a great deal of information and attributed the sutra to Zhi Qian under the name Bimo shi Mulian jing and included it in the ruzangmu of Hinayana sutras 小乘入藏目 (List of Hinayana Sutras Admitted to the Canon). Zhisheng’s record has been followed to the present day. Furthermore, critical analysis of Fei Zhangfang’s methodology in attributing this sutra to Zhi Qian, when combined with linguistic evidence, confirms that this sutra was neither translated by Zhi Qian of the Three Kingdoms period nor produced earlier than the Western Jin 西晉 Dynasty (265-316 CE). This study’s analysis of both the canonical inclusion process and the attributed translator of the Bimo shi Mulian jing demonstrates how Buddhist catalogs—exemplified by Fei Zhangfang’s Lidai sanbao ji 曆代三寶紀 (Records of the Three Treasures Throughout the Successive Dynasties)—systematically constructed false translator attributions, while simultaneously underscoring the imperative to re-evaluate so-called “authoritative records” within the Chinese Buddhist canon through integrated multidisciplinary methodologies combining Buddhist catalog criticism with linguistic analysis.

1. Introduction

Currently, the Bimo shi Mulian jing 弊魔試目連經 (The Sūtra of Māra Testing Maudgalyāyana), an individual Āgama scripture 阿含經 included in the Taishō Tripiṭaka, narrates the story of the Buddhist elder Mulian 目連 being disturbed by Māra 魔, who is ultimately subdued by the Buddha (or Mulian himself). The scripture contains teachings related to Buddhist practice and the doctrine of karma. While the Taishō Tripiṭaka attributes this text to the translator Zhi Qian 支謙, the Chu sanzang Ji Ji 出三藏記集 (Collection of Tripitaka-Translating-Records, hereafter referred to as the CSZJJ), the earliest and most reliable source for translations from the Later Han to the Song–Qi 齊 periods, does not list it under Zhi Qian’s name. This inevitably raises the question: Was this scripture truly translated by Zhi Qian?

Zhi Qian, a renowned Buddhist scripture translator of the Three Kingdoms period, was a descendant of the Dayuezhi 大月氏 people.1 Endowed with exceptional intelligence, he was fluent in six languages and trained under master translators2. In the final years of Emperor Xian’s 漢獻帝 reign, as the Han dynasty descended into chaos, Zhi Qian fled to the Wu region with his fellow villagers. There, he gained the favor of Sun Quan 孫權, the ruler of Wu, who appointed him as a boshi (an official position in ancient China) to assist the crown prince. It was during this time that he embarked on his career in Buddhist scripture translation. From the early Huangwu 黃武 era (222 CE) to the mid-Jianxing 建興 era (252–253 CE), Zhi Qian reached the peak of his translation activities. Although the majority of his works were Mahāyāna sutras 大乘經典, he also translated Hīnayāna sutras 小乘經典 and even Buddha biographies, demonstrating an impressively broad scope of work.

The reliable corpus of Zhi Qian’s authenticated translations serves as a critical benchmark for scrutinizing the authenticity of Buddhist scriptures dated to the pre-Jin period. Consequently, systematically clarifying dubious texts attributed to Zhi Qian holds particular significance. Currently, academic circles have extensively discussed the attribution issues surrounding scriptures ascribed to Zhi Qian3, yet little attention has been paid to whether the Bimo shi Mulian jing was actually translated by him. Lü Cheng 呂澂 questioned this attribution, suggesting that the Bimo shi Mulian jing was mistakenly assigned to Zhi Qian and should instead be classified as a scripture of translator unknown according to the CSZJJ (See Lü 1980, pp. 44–45). Hayashiya Tomojirō 林屋友次郎 also argued that the attribution of the Bimo shi Mulian jing to Zhi Qian was incorrect. Through his analysis of scriptural catalogs, he concluded that the Bimo shi Mulian jing should be considered an anonymous translation from the Western Jin 西晉 dynasty (See Hayashiya 1941, pp. 1309–16). While the analyses by Lü Cheng and Hayashiya addressed certain aspects of this sutra, their investigations remained preliminary in scope. Moreover, scholarly attention to this text has remained virtually dormant for decades thereafter. Fundamental questions persist: Was Zhi Qian truly not its translator? Does its provenance indeed date to the Three Kingdoms 三國 period? What transmission trajectory did it undergo historically? And crucially, when was it incorporated into the Buddhist canon under Zhi Qian’s name? These questions concerning the Bimo shi Mulian jing remain to be addressed.

Contemporary scholarship employs two primary evidentiary categories for scrutinizing dubious early Chinese Buddhist scriptures: internal evidence (lexical and stylistic features within the text itself) and external evidence (documentary records from historical sources, notably catalogs, prefaces, colophons, commentaries, and biographical accounts). Among these, successive generations of Chinese Buddhist bibliographic catalogs constitute the most valuable resource (See Zürcher 1977; Nattier 2008; Fang and Gao 2012; Fang and Lu 2023). The rich corpus of Chinese Buddhist catalogs necessitates systematic collation and analysis of their divergent historical records—an essential preliminary step in authenticating suspect early Chinese scriptures (See Fang and Gao 2012). For instance, through textual criticism of these catalogs, Tan Shibao demonstrated that many erroneous attributions of translators in early Chinese Buddhist scriptures largely originated from Lidai sanbao ji 曆代三寶紀 (Records of the Three Treasures Throughout the Successive Dynasties, hereafter referred to as the LDSBJ), and he identified several texts within the LDSBJ that were misattributed (See Tan 1991). Building on foundational catalog research, Nattier, in a series of studies, incorporated internal evidence such as linguistic analysis to investigate suspicious scriptures attributed to several renowned translators of the Eastern Han to Northern and Southern Dynasties period—including An Shigao 安世高, Zhi Qian, and Zhu Fonian 竺佛念. Her work identified a considerable number of misattributed, and even forged, scriptures. Among her findings, by analyzing the records of scriptures attributed to An Shigao in the catalogs, Nattier argued that some scriptures which have been ascribed to An Shigao since the compilation of LDSBJ are highly likely not his works, but rather translations by translators who claimed to be his disciples (See Nattier 2008). Additionally, through the analysis of scripture catalogs and the comparison of terminology used in translated scriptures, she verified the translator attribution of Damingdu Jing 大明度經 (The Sūtra of Great Brightness) and the nature of Shizhu Duanjie Jing 十住斷結經 (The Sūtra of Ten Stages and Severing Afflictions). She pointed out that chapters 2 to 30 of Damingdu Jing were translated by Zhi Qian, while the translator of the main text of chapter 1 remains unidentifiable; the author of the annotations to chapter 1 should be the scripture translation team of the Wu Kingdom led by Kang Senghui 康僧會. Furthermore, Shizhu Duanjie Jing is highly likely not a translation by Zhu Fonian, but a scripture forged by him (See Nattier 2010a, 2010b).4 Bibliographic examination serves three critical functions: first, it facilitates preliminary determination of a text’s nature (e.g., translation vs. apocryphon, canonical vs. forged composition); second, it aids in establishing the terminus ante quem for both the translator attribution and textual emergence (See Zhu 1992); third, it illuminates the historical transmission trajectory of early contested scriptures. Consequently, this paper employs methodologically validated and academically examined Chinese Buddhist catalogs5 to investigate the Bimo shi Mulian jing through the methodology of Buddhist bibliography, examining its transmission dynamics, documentary attestation, and canonical incorporation processes to determine its probable translator and chronological emergence.

2. Tracing the Inclusion of the Bimo shi Mulian jing in the Buddhist Canon

The number of scriptures translated by Zhi Qian has been a subject of debate in various historical catalogs6. The earliest extant catalog, the CSZJJ compiled by the Liang 梁 Dynasty monk Sengyou 僧祐, records 36 texts in 48 juans 卷7. Subsequent catalogs gradually increased the number of scriptures attributed to Zhi Qian, which can generally be categorized into three types: First, those largely consistent with the Zhongjing mulu 眾經目錄 (Sui Catalog of Scriptures by Fa Jing, hereafter referred to as the FJL) of the Sui 隋 Dynasty’s Fajing 法經, with minor variations. These include the Zhongjing mulu 眾經目錄 (Sui Catalog of Scriptures by Yan Cong, hereafter referred to as the YCL) of Sui Dynasty’s Yancong 彥琮 and the Zhongjing mulu 眾經目錄 (The Catalog of Sutras and Sastras from Eastern Capital Da Jing Ai Monaster, hereafter referred to as the JTL) of Tang 唐 Dynasty’s Jingtai 靜泰. Among these, the FJL records 39 texts in 62 juans by Zhi Qian, the YCL approximately 36 texts in 57 juans8, and the JTL approximately 39 texts in 81 juans (or 78 juans)9. Second, those primarily following the LDSBJ of Sui Dynasty’s Fei Zhangfang (Changfang) 費長房 (hereafter referred to as the Fei Zhangfang), with slight differences. These include the Datang neidian lu 大唐內典錄 (Catalog of Buddhist Scriptures in the Great Tang, hereafter referred to as the NDL) of Tang Dynasty’s Daoxuan 道宣 and the Dazhou kanding zhongjing mulu 大周刊定眾經目錄 (Zhou Dynasty Corrected Catalog of Scriptures, hereafter referred to as the DZL) of Tang Dynasty’s Mingquan 明佺. The LDSBJ already increased Zhi Qian’s translations to 129 texts in 152 juans (or 130 texts in 153 juans10), representing a several-fold increase compared to previous and contemporary catalogs. The NDL and the DZL, largely following Fei Zhangfang’s account, went even further. The NDL added the Zhuanji baiyuan Jing 撰集百緣經 (Avadānaśataka, 1 text in 10 juans) to the LDSBJ’s count, totaling 130 texts in 162 juans (or 131 texts in 163 juans11). The DZL, while following the LDSBJ and NDL, added new scriptures under Zhi Qian’s name (See Z. Li 2023, pp. 42–43), totaling 136 texts in 170 juans. Third, the account in the Kaiyuan shijiao lu 開元釋教錄 (Kaiyuan Catalog of Buddhist Scriptures, hereafter referred to as the KYSJL) of Tang Dynasty’s Zhisheng 智昇. Faced with the gradually increasing numbers of Zhi Qian’s translations after the CSZJJ, particularly the dramatic increase in the LDSBJ, Zhisheng stated while compiling the KYSJL: “Now regarding the records in the LDSBJ, most are either extracted texts or duplicate entries with different titles. We will delete them accordingly as described below.”12 Through extensive research and collation, he ultimately revised the number of Zhi Qian’s translations to 88 texts in 118 juans, and compiled Zhi Qian’s works into the ruzanglu 入藏錄 (list [of texts] admitted to the canon) accordingly. Thereafter, during the woodblock printing era, the number of Zhi Qian’s translations stabilized based on the KYSJL’s records.

Currently, the Buddhist canon attributes 56 texts in 77 juans to Zhi Qian, including the Bimo shi Mulian jing. However, upon examining the earliest extant catalog, the CSZJJ, this scripture is notably absent from Zhi Qian’s corpus. If this scripture was not actually translated by Zhi Qian, then how it was incorporated into the canon and subsequently attributed to him becomes a question worthy of investigation.

In the early years of the Northern Song 北宋 dynasty, the Bimo shi Mulian jing was already included in the Kaibao zang 開寶藏 (which began engraving in the 4th year of the Kaibao 開寶 era [971 CE] of the Northern Song 北宋 dynasty (See He 2014, pp. 782, 39–42). The Kaibao zang’s scripture list was based precisely on the ruzanglu 入藏錄 in the KYSJL compiled by Zhisheng of the Tang dynasty (See Tong 1991). Upon examining juan 20, the ruzangmu of Hinayana sutras 小乘入藏錄 of the KYSJL13—compiled by the Tang dynasty’s Zhisheng and completed in the 18th year of the Kaiyuan 開元era (730 CE)—we find that the Bimo shi Mulian jing had already been recorded under Zhi Qian of Wu state 吳 during the Three Kingdoms period and included in the canon by the time of Zhisheng’s KYSJL in the Tang dynasty.14 However, investigation of pre-KYSJL catalogs shows no prior ruzanglu 入藏錄 that attributes this scripture to Zhi Qian. We can therefore conclude that the classification of Bimo shi Mulian jing as a translation by Zhi Qian of Wu state and its subsequent canonization began with Zhisheng’s KYSJL.

3. A Critical Examination of the Bimo shi Mulian jing in Buddhist Bibliographical Records

According to the information provided in juan 20 of Zhisheng’s KYSJL, we can ascertain that the Mo raoluan jing 魔嬈亂經 (Sutra on Māra Harassing and Agitating), the Mowang ru Mulianlan fu jing 魔王入目連蘭腹經 (Sutra on How Māra Entered Maudgalyāyana’s Belly), and the Bimo shi Mulian jing represent different titles of the same original text. Notably, while the Mo raoluan jing comprises eight sheets and the Bimo shi Mulian jing only five sheets—a difference of three sheets—this indicates they are not merely different titles of the same text but actually different translations of the same source. Cross-referencing this with information from juan 2 of the KYSJL, we further learn that both the Bimo shi Mulian jing and the Mo raoluan jing serve as alternative translations of juan 30 of the Madhyamāgama 中阿含經. Zhisheng specifically identifies the Bimo shi Mulian jing as an extant translation by Zhi Qian, a determination he made based on records found in Fei Zhangfang’s LDSBJ from the Sui dynasty.

3.1. Documentation in Fei Zhangfang’s LDSBJ (Sui Dynasty)

The LDSBJ was completed in the 17th year of the Kaihuang 開皇 era (597 CE) of the Sui dynasty. As discussed earlier, the number of scriptures attributed to Zhi Qian experienced a several-fold increase beginning with the LDSBJ, suddenly jumping from the thirty-some texts recorded in earlier catalogs like the CSZJJ to 129 texts in 152 juans. Regarding this dramatic expansion, Fei Zhangfang himself stated that Zhi Qian translations he compiled were gathered through his “examination and collection from various miscellaneous records 檢括眾家雜錄”15. He used the Sishi’erzhang jing 四十二章經 (The Sutra of Forty-two Chapters) as a demarcation point: this scripture and those listed below were included as “Various Sutras recorded in other documents 別記所顯雜經”, while those above it were derived from either the CSZJJ or GSZ. Notably, the LDSBJ lists 40 texts above the Sishi’erzhang jing, which contradicts both the interlinear note preceding this scripture stating16 and the actual number of Zhi Qian translations recorded in the CSZJJ. Regarding this discrepancy, Li Zhouyuan 李周淵 once suggested that “the contradictions in the LDSBJ likely resulted from later additions by subsequent compilers” (See Z. Li 2023, p. 40). Upon examining the LDSBJ, we find that the listed scripture titles numerically match its claimed total of 129 texts in 152 juans attributed to Zhi Qian. If, as Li Zhouyuan proposed, later additions caused these inconsistencies, then prior to such modifications, the LDSBJ’s statistical account of Zhi Qian’s translations would not have aligned with its actual listed titles. This situation necessitates a thorough re-examination of the textual evidence to formulate reasonable inferences.

The CSZJJ originally recorded Zhi Qian’s translation of the Siyuan jing 四願經 (Sutra on the Four Vows) in 1 juan17. However, in the current version of the LDSBJ, the Siyuan jing appears below the Sishi’erzhang jing—precisely where Fei Zhangfang claimed to have included scriptures from “Various Sutras recorded in other documents 別記所顯雜經”. More perplexingly, the LDSBJ contains an interlinear note after the Siyuan jing stating “seen in Zhu Daozu’s Wu lu and the CSZJJ 見竺道祖吳錄及出三藏記集”18, which clearly indicates the compiler was aware this scripture was documented as Zhi Qian’s translation in the CSZJJ. Nevertheless, it was still categorized among what Fei Zhangfang described in his “biography of Zhi Qian 支謙傳” as “Various Sutras recorded in other documents 別記所顯雜經”.

Regarding these internal contradictions in the LDSBJ, Michael Radich characterizes them as “Patterns Of Self-Contradiction” in Fei Zhangfang’s compilation methodology (See Radich 2019, p. 826). Radich’s analysis identifies 4 types of inconsistencies in the catalog: first, among the “Later Han translations 後漢譯經”, 17 texts marked as “translator unknown 失譯” are elsewhere in the catalog attributed to specific translators; second, in the “Eastern Jin translations 東晉譯經” section, 4 texts labeled as anonymous are elsewhere assigned to translators; third, ruzangmu 入藏目 (list [of texts] admitted to the canon) of the LDSBJ contains 64 anonymous texts that receive translator attributions elsewhere in the LDSBJ; fourth, 33 texts documented in the CSZJJ appear in the LDSBJ with conflicting information, creating contradictions both with the CSZJJ and within itself. Radich argues that such numerous inconsistencies can hardly be coincidental, and more often than not reflect Fei Zhangfang’s insufficiently rigorous approach to compilation—what might be termed Fei Zhangfang’s “unintentional carelessness” or “deliberate negligence” (See Radich 2019, pp. 827–829). While Fei Zhangfang’s lack of scholarly rigor is indeed undeniable, the contradictions here may warrant more nuanced examination. For instance, the “biography of Zhi Qian” and the listing and statistics of Zhi Qian’s translations in the LDSBJ might not have been produced simultaneously or by the same hand. The Zhi Qian section in LDSBJ’s juan 5 could well represent a composite of multiple contributors working in several stages. The first stage likely resembled the basic situation described in Fei Zhangfang’s “biography of Zhi Qian” possibly representing the original version of the Zhi Qian section—with pre–Sishi’erzhang jing texts drawn from the CSZJJ or GSZ, and that the Sishi’erzhang jing plus subsequent texts sourced from “separate catalog 別錄”. At this phase, the listing and statistics of Zhi Qian’s translations probably differed somewhat from the current version, and may not have included all Zhi Qian translations recorded in the CSZJJ—the Siyuan jing might have been omitted at this stage19. The second stage corresponds to the version described in the interlinear note preceding the Sishi’erzhang jing, where 29 texts in 47 juans (excluding the Yujia zhangzhe jing 郁迦長者經 [Ugraparipṛcchā]) were extracted from the CSZJJ20. In the third stage, compilers—perhaps overlooking both the interlinear note and Fei Zhangfang’s “biography of Zhi Qian”—made final additions and statistics to Zhi Qian’s translation list. The Siyuan jing was inserted without scrutiny below the Sishi’erzhang jing, while four Zhi Qian translations unrecorded in the CSZJJ were arbitrarily placed above it, resulting in the version we see today. Regardless of the precise process, this at minimum confirms that Fei Zhangfang (or his editorial team) lacked sufficient scholarly rigor in compiling the catalog—a conclusion that aligns perfectly with Radich’s assessment of Fei Zhangfang’s “unintentional carelessness” or “deliberate negligence”.

In the current version of the LDSBJ, the Bimo shi Mulian jing is categorized under what Fei Zhangfang termed “Various Sutras recorded in other documents”. An interlinear note following this entry states: “Bimo shi Mulian jing, 1 juan, Seen in the Jiu lu. Derived from Abhidharma. 弊魔試目連經一卷, 見舊錄. 出阿毗曇”21. The “Jiu lu 舊錄 (Old catalog)”, Fei Zhangfang admitted that he never personally examined it.22 This demonstrates that contents attributed to the Jiu lu in the LDSBJ were actually Fei Zhangfang’s secondary citations from other catalogs that referenced the Jiu lu. Cross-referencing the records of the LDSBJ, we reveal Fei Zhangfang only personally consulted four texts when compiling the Zhi Qian section: the CSZJJ, the FJL, the Zhongjing bielu 眾經別錄 (separate catalogue of the [Liu] Song canon), and the Baochang lu 寶唱錄 (Baochang’s Catalog).

3.2. Documentation in Sengyou’s CSZJJ (Liang Dynasty)

Among the currently extant catalogs that Fei Zhangfang personally consulted and which contain information related to the Bimo shi Mulian jing, only the CSZJJ and the FJL remain23. The CSZJJ, compiled by Sengyou of the Liang dynasty, continued to receive additions until the compiler’s death (518 CE). This catalog stands as a reliable source for understanding Buddhist scripture translation from the Later Han 後漢 to Song 宋–Qi 齊 periods. A thorough examination of the CSZJJ reveals no text title that completely matches the current Bimo shi Mulian jing character-for-character. However, several similar titles can be identified:

• Mowang ru Mujianlan fu jing 魔王入目犍蘭腹經 (also titled Bimo shi Mohemulian jing 弊魔試摩訶目連經, As recorded in the Jiu lu: the Mowang ru mulian fuzhong jing 魔王入目連腹中經).24
• Mo raoluan jing 魔嬈亂經 (shares broad similarities with the Mowang shi mulian 魔王試目連).25
• Mo shi mulian jing 魔試目連經 (also titled the Bimo shi Momulian jing 弊魔試摩目連經).26

The Bimo shi Mulian jing in the Taishō Tripiṭaka shows minor variations from the text titles recorded in the CSZJJ (the Mowang ru Mujianlan fu jing 魔王入目犍蘭腹經, the Bimo shi Mohemulian jing 弊魔試摩訶目連經 [Sutra on Māra Testing Mahāmaudgalyāyana], the Mowang ru mulian fuzhong jing 魔王入目連腹中經 [Sutra on How Māra Entered Maudgalyāyana’s Belly], the Mowang shi mulian 魔王試目連 [Māra Tests Maudgalyāyana], the Mo shi mulian jing 魔試目連經 [Sutra on Māra Testing Maudgalyāyana], and the Bimo shi Momulian jing 弊魔試摩目連經 [Sutra on Māra Testing Mahāmaudgalyāyana]). These differences primarily fall into two categories:

• The Mowang ru Mujianlan fu jing 魔王入目犍蘭腹經 follows a distinct naming pattern: “Name 1 + 入 [enters] + Name 2 + 腹 [abdomen] + 經 [sutra]” (hereafter referred to as “Pattern 1”).
• Second, all other titles generally conform to “Name 1 + 試 [tests] + Name 2 + 經 [sutra] (with or without “經”)” (hereafter referred to as “Pattern 2”).

Can we then conclude that scriptures sharing “Pattern 2” in the CSZJJ represent the same text under different names? The answer appears affirmative. Firstly, variations within “Pattern 2” mainly stem from translational or scribal differences in “Name1” (Mo 魔, Mowang 魔王, Bimo 弊魔) and “Name2” (Mulian 目連, Mohemulian 摩诃目连, Momulian 摩目连). Regarding “Name1”: “Mo 魔” abbreviates the Sanskrit “Māra”, and “Mowang 魔王” specifically denotes the Sixth Heaven Lord of the Desire Realm 欲界第六天之他化自在天主—the Pāpīyas 天魔中之王也. Regarding “Name2”: “Mohemujianlian 摩訶目犍連” (Mahā-maudgalyāyana) refers to one of Buddha’s ten principal disciples, renowned for supernatural powers. Its abbreviated forms include Mujianlian 目犍連 and Mulian 目連. The difference between the Mowang shi mulian and Mo shi Mulian jing (hereafter referred to as the Mo shi Mulian jing group) lies only in the characters “wang 王” and “jing 經”. Similarly, the Bimo shi mohemulian jing and the Bimo shi Momulian jing (hereafter referred to as the Bimo shi mohemulian jing group) differ by just the character “he 訶”. This suggests these parallel titles likely represent scribal omissions of the same original text. Secondly, since Sengyou identifies both the Mo shi Mulian jing and the Bimo shi Momulian jing, while the Bimo shi Mohemulian jing is noted as an alternative title for the Mowang ru Mujianlan fu jing (called the Mowang ru Mulian fuzhong jing in the Jiu lu, hereafter referred to as the Mowang ru Mujianlan fu jing group), we can reasonably conclude these three groups—the Mo shi mulian jing group, the Bimo shi Mohemulian jing group, and the Mowang ru Mujianlan fu jing group—at minimum represent the same text with different titles and translations.

Having clarified the relationships between these different text titles, we can now better understand the records in the CSZJJ. First, during Sengyou’s time, versions with “Pattern 2” titles were either lost or represented alternative titles/translations of either the Mo raoluan jing 魔嬈亂經 or scriptures with “Pattern 1” titles. In other words, Sengyou never personally examined versions with “Pattern 2” titles—he only saw versions with “Pattern 1” titles and the Mo raoluan jing, which by his time had already been included in the canon (入藏). However, neither extant versions from Sengyou’s time (the Mowang ru Mujianlan fu jing and the Mo raoluan jing) nor the then-lost Mo shi mulian jing recorded their translators in the CSZJJ. Specifically: the Mowang ru Mujianlan fu jing was documented by Sengyou based on the “Dao’an lu 道安錄 (Dao’an’s Catalog)”, indicating its translator was already unidentifiable during Dao’an’s era. Both the Mo raoluan jing and the Mo shi mulian jing were categorized by Sengyou under “Newly Compiled Continuation of the Assorted List of Anonymous Translations 新集續撰失譯雜經錄”.

In summary, we can conclude that both the Mo raoluan jing and the Mowang ru Mujianlan fu jing were texts personally examined by Sengyou, meaning these texts were certainly not forgeries created after the compilation of the CSZJJ (518 CE). Furthermore, since the Mowang ru Mujianlan fu jing was included in the “Dao’an lu”, its translation date must be even earlier—it could not have been forged after Dao’an’s era (312–385 CE). Additionally, Sengyou’s observation that the Mo raoluan jing was largely similar with minor differences to the Mo shi mulian jing indicates that the Bimo shi Mohemulian jing, the Mowang ru Mujianlan fu jing, and the Mo raoluan jing likewise shared fundamental similarities while displaying minor variations.

3.3. Documentation in the FJL (Sui Dynasty) and Subsequent Buddhist Catalogs

The Bimo shi Mulian jing, currently preserved in the Buddhist canon and attributed to Zhi Qian, bears a title most closely resembling the Bimo shi Mohemulian jing group recorded in the CSZJJ. This series includes: (1) an alternative title for the Mowang ru Mulian fuzhong jing—a text extant during Sengyou’s time; (2) an alternative title for the Mo shi mulian jing—a text already lost by Sengyou’s era. This raises two critical questions: Could the current Bimo shi Mulian jing possibly be identical to the Bimo shi Mohemulian jing documented by Sengyou? And when was it first recorded in scriptural catalogs and attributed to Zhi Qian?

In contrast to the CSZJJ’s records, texts bearing “Pattern 1” titles (i.e., the Mowang ru Mujianlan fu jing group) underwent significant transformations in subsequent catalogs after Sengyou’s documentation. Apart from Sengyou’s personal verification, these texts either appeared as lost versions27 or resurfaced under alternative titles like the Mo raoluan jing and the Bimo shi Mulian jing. This transitional phenomenon first emerged in the FJL. Completed in the 14th year of the Kaihuang era (594 CE) during the Sui dynasty, the FJL did not list the Bimo shi Mulian jing as an independent entry. Instead, it appeared as an interlinear note following the Mo raoluan jing28.

The FJL introduced several new pieces of information not found in the CSZJJ: First, the title Bimo shi Mulian jing made its initial appearance in scriptural catalogs. Second, it identified this text as an alternative title for both the Mo raoluan jing and the Mowang ru Mujianlan fu jing. Additionally, it is noted that this scripture derived from juan 30 of the Madhyamāgama as a separate recension with variant translations. This raises a crucial methodological question: How did the compilers of the FJL arrive at these conclusions?

According to the memorial Fajing submitted to the emperor, it is evident that he did not personally examine all scriptures when compiling the catalog. His work primarily relied on various catalogs accessible at that time, with the CSZJJ serving as a major reference29. However, the CSZJJ does not systematically or explicitly indicate the sources of scriptures like the Mo raoluan jing or the Bimo shi Mohemulian jing. However, Sengyou had already discerned that these scriptures were likely transcribed or abridged from the Madhyamāgama.30 Fajing, building upon Sengyou’s work and other extant catalogs of his time—possibly combined with his own textual examinations—likely determined that these scriptures derived from juan 30 of the Madhyamāgama. But what were these “other catalogs” beyond the CSZJJ that Fajing consulted? Scholarly comparison of the records concerning Zhi Qian’s translations across the CSZJJ, the FJL, and the LDSBJ suggests that the FJL shared common source materials with the Zhu Daozhu’s Wu lu 竺道祖吳錄 (Zhu Daozhu’s Catalog from the Wu Dynasty); the Zhongjing bielu 眾經別錄 from the Liu–Song 劉宋 period; and the Baochang lu 寶唱錄of the Qi–Liang 齊梁 period (See Z. Li 2019, p. 36). Thus, these catalogs most probably served as Fajing’s reference materials during compilation.

Unfortunately, among the aforementioned catalogs, only a fragmentary Dunhuang 敦煌 manuscript remains of the Liu–Song period’s Zhongjing bielu, and this remnant contains no traceable content related to the Bimo shi Mulian jing. Both the Zhu Daozhu’s Wu lu and the Baochang lu have been lost, making it impossible to verify whether they originally contained any records of the Bimo shi Mulian jing. Fortunately, Baochang’s Buddhist encyclopedia the Jinglüyixiang 經律異相 (Strange Tales of Sutras and Vinayas), compiled in the 15th year of Tianjian 天監 era (516 CE) during the Liang dynasty, survives to this day31. This work’s compilation methodology involved extracting and reorganizing passages from original scriptures, with Baochang meticulously documenting each story’s scriptural source. Notably, two stories in juan 1432 and juan 4633 specifically cite the Bimo shi Mulian jing as their source.

This evidence demonstrates that the Bimo shi Mulian jing was already extant during the time of Baochang. Given that Baochang demonstrated familiarity with this text when compiling the earlier the Jinglüyixiang, it is highly probable that his subsequently compiled Baochang lu also included this scripture. Furthermore, the Baochang lu’s records were substantially inherited by Fajing in his own compilation. Moreover, the Jinglüyixiang’s documentation allows us to reasonably conclude that the Bimo shi Mohemulian jing group recorded in Sengyou’s work most likely referred to the same text as the Bimo shi Mulian jing. The CSZJJ and the Jinglüyixiang were contemporaneous works, with the latter’s compiler Baochang having been a direct disciple of Sengyou34, the CSZJJ’s author. This close master–disciple relationship, combined with their temporal and geographical proximity, strongly suggests significant overlap in the scriptural materials they accessed. Sengyou’s recording of the title as the Bimo shi Mohemulian jing may stem from his reliance on the Dao’an lu’s records. What Sengyou termed the Bimo shi Mohemulian jing was in all likelihood identical to the Jinglüyixiang’s Bimo shi Mulian jing.

However, discrepancies exist between Baochang’s and Fajing’s records regarding the Bimo shi Mulian jing—while Baochang cites it as deriving from juan 27 of the Madhyamāgama, Fajing attributes it to juan 30. Four possible explanations may account for this divergence: (1) When compiling his catalog, Fajing primarily referenced other catalogs and likely did not consult the Jinglüyixiang. (2) Fajing treated the Bimo shi Mulian jing as an alternative title for the Mo raoluan jing, recording it only as an interlinear note without creating a separate entry. His textual attribution would thus have been based mainly on the Mo raoluan jing, which lacks corresponding documentation in the Jinglüyixiang. (3) Even if the Baochang lu included the Mo raoluan jing, it might not have attributed it to Madhyamāgama juan 27 as the Jinglüyixiang did for the Bimo shi Mulian jing. When encountering the same narrative across multiple source texts with textual variations, Baochang typically annotated them with phrases like “also appears in such–and–such scripture 出某經” or “such–and–such scripture largely agrees with such–and–such scripture 某經與某經大同”. His failure to note any relationship between the Bimo shi Mulian jing and the Mo raoluan jing suggests he might have overlooked their connection, and thus might not have carried over the Jinglüyixiang’s attribution. (4) The Madhyamāgama versions Baochang and Fajing35 consulted might have differed36. Possibly, what Baochang called “juan 27” and Fajing called “juan 30” actually referred to the same text in different editions.

Following Fajing, Fei Zhangfang—the Scripture Translation Scholar of Xing Shan Monastery 興善寺—synthesized previous catalogs and recorded the Bimo shi Mulian jing as a translation by Zhi Qian. Subsequent catalogs by Sui Dynasty’s Yancong 彥琮 and Tang Dynasty’s Jingtai 靜泰 entirely followed Fajing’s attribution, while Tang Dynasty’s Daoxuan 道宣 and Mingquan 明佺 adopted Fei Zhangfang’s account without modification. It was not until the compilation of the KYSJL by Zhisheng 智昇 during the Tang 唐 dynasty that prior historical records were systematically integrated, attributing the translation of the Bimo shi Mulian jing to Zhi Qian and identifying this scripture as a parallel text to the Mo raoluan jing and juan 30 Xiangmo jing 降魔經 (Māratajjaniya-sutta) of the Madhyamāgama’s juan 3037. Zhisheng included this entry in the ruzangmu of Hinayana sutras 小乘入藏目38. Although Zhisheng claimed his record of the Bimo shi Mulian jing derived from the LDSBJ39—our preceding analysis clearly demonstrates that Zhisheng’s account actually synthesized multiple sources rather than strictly following the LDSBJ. Subsequent catalogs through history, including those of the woodblock-printed canon era, all adopted the KYSJL as their reference. Thus, Zhisheng’s synthesized attribution became canonical, enduring through modern times—even the Taishō Tripitaka published in Japan preserves this attribution, listing the Bimo shi Mulian jing as Zhi Qian’s translation within the Madhyamāgama’s individual section.

4. Textual Criticism on the Attribution of the Bimo shi Mulian jing’s Translator

As previously discussed, the Bimo shi Mulian jing was first attributed to a translator—the renowned Three Kingdoms 三國 scripture translator Zhi Qian of the Wu state—beginning with the LDSBJ. However, Fei Zhangfang’s account significantly diverges from all aforementioned earlier catalogs, inevitably raising doubts about the accuracy of his attribution of this scripture to Zhi Qian.

The LDSBJ records the relevant scriptures mentioned in earlier catalogs as follows:

• The Mo raoluan jing is an anonymous translation (失譯經) from the Later Han dynasty (後漢, 25–220 CE).40
• The Bimo shi Mulian jing represents a retranslation by Zhi Qian of the Wu state during the Three Kingdoms period (三國, 220–280 CE), based on earlier translated works.41
• Although Fei Zhangfang did not include texts categorized under “Pattern1” from the CSZJJ in his LDSBJ, he appended novel information: (1) The Monü wen Fo shuofa de nanshen jing 魔女聞佛說法得男身經 (Sutra on the Māra’s Daughter who Heard the Buddha Preach and Obtained a Male Body, hereafter referred to as the MNJ), translated by Faju 法炬 of the Western Jin dynasty (西晉, 266–316 CE), shares textual identity with the Bimo shi Mulian jing but bears a distinct title42. (2) The same MNJ is also attributed to Zhi Qian of the Wu state43.

From this evidence, it is evident that Fei Zhangfang’s account of the Mo raoluan jing directly adopted the CSZJJ’s records. However, his documentation of the Bimo shi Mulian jing diverges conspicuously from all preceding catalogs (經錄), a discrepancy that raises significant historiographical questions.

First, Fei Zhangfang’s attribution of the Bimo shi Mulian jing to Zhi Qian as a “re–translation” (or “revised translation”) of earlier versions lacks evidentiary support and likely reflects his subjective conjecture. This conclusion stems from three key observations: (1) Fei Zhangfang provided no source for this claim, neither citing any evidence himself nor drawing from prior catalogs that might have documented such a connection. (2) Scholars have widely criticized Fei Zhangfang’s practice of arbitrarily assigning anonymous translations to known translators (Chen 2020, p. 41). (3) Fei Zhangfang’s assumption that Zhi Qian revised an earlier translation of the Bimo shi Mulian jing appears influenced by his preconceived notion of Zhi Qian as a reviser. The earliest reference to Zhi Qian revising others’ translations comes from Zhi Mindu 支敏度44. However, Zhi Mindu’s characterization itself requires verification, as does Fei Zhangfang’s parallel claim.

Secondly, Fei Zhangfang’s claim that he followed the “Jiu lu” in documenting the Bimo shi Mulian jing as deriving from “Abhidharma” (阿毗曇)45 sharply contradicts prior catalog attributions, rendering his account highly unreliable. Although Fei Zhangfang asserted he inherited this record from the “Jiu lu”, as previously discussed, he never personally examined this catalog. Moreover, Fei Zhangfang himself described it as: “Jiu lu—Resembles the scripture catalog seen by Liuxiang 劉向 of the Former Han dynasty 前漢 when compiling the imperial collection”46. Without even addressing whether this “Jiu lu” could genuinely be a Former Han document, Fei Zhangfang’s own statements contain irreconcilable contradictions. The indisputable historical fact that Zhi Qian lived during the Three Kingdoms period means scriptures he translated could never have appeared in a Former Han catalog. This fundamental chronological impossibility exposes Fei Zhangfang’s claims as internally inconsistent and patently untrustworthy. Furthermore, attributing the text to “Abhidharma” (論藏) drastically diverges from the consensus of Sengyou, Fajing, and Baochang, who all identified it with the Madhyamāgama (belonging to the Sūtra Piṭaka 經藏). Notably, the LDSBJ itself uses “sūtra” (修多羅) for this category elsewhere, making the Abhidharma attribution even more anomalous.

Finally, Fei Zhangfang’s claim that he followed the Shixing lu 始興錄 and the CSZJJ in recording the Bimo shi Mulian jing as “same text, different title 本同名異” with the MNJ (now lost) translated by Faju of the Western Jin dynasty not only reveals internal contradictions but also directly conflicts with the CSZJJ’s documentation of these scriptures. Regarding the Shixing lu, Fei admitted he never personally examined it47—all his references to it were secondhand citations from other catalogs. In contrast, the CSZJJ was among Fei Zhangfang’s firsthand sources. The MNJ was cataloged by Sengyou under “Newly Compiled Continuation of the Assorted List of Anonymous Translations” in his CSZJJ. Sengyou explicitly identifies this text as originating from the Mahāvaipulya mahāsamghāta sūtra (大方等大集經), a corpus of Mahāyāna sūtra, and asserts that it bears no textual relationship to the Bimo shi Mulian jing.48 While the LDSBJ also lists a MNJ from the Mahāsaṃnipāta 大集經 attributed to Zhi Qian, Fei Zhangfang failed to clarify its relationship to Faju’s version of the same title. The implausibility of two same–named scriptures—one allegedly from Abhidharma and another from the Mahāsaṃnipāta—further undermines credibility. As previously argued, the LDSBJ’s most charitable interpretation would be that both entries refer to the same text, with Fei Zhangfang’s “unintentional carelessness or deliberate negligence” creating this contradictory duplication.49

Furthermore, as noted in the introduction of this paper, the examination of early Chinese Buddhist sūtras of questionable authenticity may utilize both external and internal evidence. Sufficient external evidence can serve as a robust criterion for identification to a certain extent. This study investigates the Bimo shi Mulian jing through a bibliographic lens (i.e., external evidence). However, concurrent analysis employing internal evidence—such as the sutra’s lexicon and stylistic features—would enhance the reliability of conclusions drawn from external evidence.50

Textual analysis of the Bimo shi Mulian jing reveals notable lexical discrepancies when compared with translations reliably attributed to Zhi Qian51. For instance: (1) “Julouqin 拘樓秦” appears as “Juliuqin 拘留秦” in Zhi Qian’s authentic works52; (2) “Yongmengshi 勇猛士” corresponds to “Yongmengzhe 勇猛者” in Zhi Qian’s authentic works53; “E qu 惡趣” is rendered as “E dao 惡道” or “E chu 惡處” in Zhi Qian’s authentic works54.

Additionally, multiple terms in this sutra are unattested in Eastern Han or the Three Kingdoms period translations, with their earliest reliable occurrences documented in the Western Jin 西晋 translations (See

Table 1. Lexical items in the Bimo shi Mulian jing first attested in Western Jin translations55.

	Lexical Items in the Bimo shi Mulian jing	Sources of Reliably Attributed Translations from the Western Jin Period
1	Dasheng shizun 大聖世尊	T154 (2); T263 (2)
2	Wuji dasheng 無極大聖	T263 (2); T285; T292; T338; T401; T585; T627; T815
3	Shewei 阇維/蛇維	T170; T461; T513/T154; T263
4	Yongmengshi 勇猛士	T285; T342; T381; T585; T627
5	Jietuomen 解脫門	T135; T154; T221 (2);T222 (3); T263; T398; T399; T401; T813
6	E qu 惡趣	T263; T345
7	Dabang 大棒	T186; T342
8	Jiexian 節限	T154 (2); T222; T263 (3); T285; T292; T318; T342; T345; T381; T401; T403; T425 (3); T565; T585 (2); T589 (2)
9	Zhangzhe fanzhi 長者梵志	T186 (3); T263 (2); T274; T452; T461 (2); T562; T565 (5); T606; T809; T1301 (3)
10	Dadiyu/Duo dadiyu 大地獄/墮大地獄	T263 (5); T325; T342 (5); T345 (2); T378 (2); T481 (3); T606; T627; T635 (2)/T342 (3); T345 (2); T627
11	Xurusuonian 尋如所念	T263 (2); T318; T342; T461; T810
12	Julouqin 拘樓秦	T199; T263; T345; T378; T598 (2): T635
13	Qingyang 謦揚	T292; T318; T425; T433; T461; T534; T589; T810
14	Fu zhongdan 負重擔	T263; T266; T292; T425; T627
15	Zao e 遭厄	T292; T403; T425; T606; T811
16	Xincao/Ji xincao 薪草/積薪草	T263; T291/T154; T263; T291 (3); T606
17	Haochuang 好床	T186 (2); T345 (2); T425 (2)
18	Huoku 獲苦	T345; T398
19	Hengzao 橫造	T266; T403
20	Pujian 普見	T154; T170; T186 (2); T199 (6); T222 (5); T263 (9); T266 (2); T274; T285 (2); T288 (2); T291 (3); T292 (12); T315a; T315b; T324 (5); T345; T381 (2); T398 (8); T399; T403 (5); T425 (4); T433; T435; T459 (6); T460; T481; T585 (2); T589; T598; T606; T627; T636; T811
21	Si jiaodao 四徼道	T118 (2); T461 (2); T815
22	Kouyan 口演	T186; T266; T285; T291; T292 (2); T398 (2); T399; T403
23	Miaohua shan 妙華山	T349
24	Nilibang 泥犁傍	T23 (36)

), as detailed in the table below:

These lexemes bearing distinctive markers of Western Jin linguistic conventions further substantiate that the Bimo shi Mulian jing cannot be attributed to the translation activities of Zhi Qian during the Three Kingdoms period. The text’s earliest possible emergence does not predate the Western Jin era.

At this point, our textual analysis of the Bimo shi Mulian jing in the current Buddhist canon yields the following conclusions:

• The Bimo shi Mulian jing had emerged no later than the era of Dao’an (312–385 CE) during the Eastern Jin dynasty 東晉. Contemporary records, however, documented it under the title Bimo shi MoheMulian jing. The earliest definitive attestation of its abbreviated title Bimo shi Mulian jing first appears in Baochang’s Jinglüyixiang. Consequently, it is plausible that the now-lost Baochanglu 寶唱錄 likewise contained relevant documentation regarding this sutra.
• The attribution of the Bimo shi Mulian jing to juan 30 of the Madhyamāgama can be traced no earlier than the FJL of the Sui dynasty. However, its possible connection to the Madhyamāgama was already noted during Sengyou’s time in the Liang dynasty—though Sengyou did not pursue further or more detailed verification.
• The classification of the Bimo shi Mulian jing under Zhi Qian’s name originates from Fei Zhangfang’s LDSBJ of the Sui dynasty. Yet Fei Zhangfang’s account lacks substantial evidence, rendering it highly unreliable. Concurrently, linguistic analysis reveals that the Bimo shi Mulian jing employs numerous lexemes first attested in Western Jin translation corpora. Consequently, it can be inferred that this scripture neither originated during the Three Kingdoms period nor represents a translation attributable to Zhi Qian.

5. Conclusions

In summary, the Bimo shi Mulian jing was clearly not translated by Zhi Qian of the Three Kingdoms period. The earliest possible date for its translation does not predate the Western Jin dynasty, while the latest possible date does not extend beyond the era of Dao’an of the Eastern Jin dynasty. Its translator should also be someone active during this time. The primary source of the major error that has persisted for centuries, attributing this sūtra to Zhi Qian of the Wu kingdom during the Three Kingdoms period, is Fei Zhangfang, author of LDSBJ from the Sui dynasty.

The trajectory of the scripture of Bimo shi Mulian jing into the Buddhist canon and the contentious issue of its translator attribution vividly reflect the complexity and historical stratigraphy inherent in medieval Buddhist catalog compilation and sutra dissemination. Although Fei Zhangfang’s LDSBJ ascribed numerous scriptures of uncertain provenance—including the Bimo shi Mulian jing—to the translator Zhi Qian, and though subsequent catalogers widely adopted this attribution, our bibliographic investigation conclusively demonstrates the precipitous and unreliable nature of Fei’s judgment. This scripture is by no means an isolated case; it exposes a persistent and far-reaching scholarly predicament: the ‘translator attributions’ in extant Chinese Tripitakas, particularly concerning early anonymous sutras, contain a substantial proportion of retrospectively appended or misassigned ‘labels’—largely stemming from post-LDSBJ cataloging practices. This phenomenon compels critical re-examination of three fundamental issues:

First, the constructed nature and fluidity of ‘translatorial authority’. The attribution of prominent translators (e.g., Zhi Qian) to Chinese Buddhist translations often derives not from verifiable primary evidence, but rather from later catalogers’ (notably Fei Zhangfang) contextual classification, speculation, or even arbitrary assertions within specific historical milieus. This process of ‘attribution tagging’ constitutes an integral mechanism in the construction of Buddhist knowledge systems and authority, mirroring evolving conceptions of ‘translational legitimacy’ across periods.

Second, the dual nature of sutra catalogs’ functions. While Buddhist catalogs remain indispensable sources for textual genealogy, their compilation was never a purely objective endeavor. They simultaneously served as historical archives and contemporary interpretations. Despite preserving significant reliable data, Fei Zhangfang’s LDSBJ—constrained by its compositional objectives and methodological limitations—inadvertently or deliberately generated new attribution conundrums for posterity, epitomizing the adverse consequences of this duality.

Third, the evidentiary basis for early Chinese Buddhist translation studies. The case of the Bimo shi Mulian jing cautions us that research predicated uncritically on current Chinese Tripitaka attributions—across disciplines such as intellectual history, linguistics, or literary history—rests upon potentially fragile foundations. For any early scripture ascribed to a specific translator, particularly those solely documented in the LDSBJ and later catalogs, extreme scholarly caution is warranted; attributions demand rigorous verification before utilization. Fortunately, academia now recognizes this imperative, developing interdisciplinary methodologies (philological, linguistic, historical) that have successfully reappraised numerous dubious early attributions—providing mature frameworks for this scripture’s further study.

Admittedly, definitive attribution of the Scripture of the Bimo shi Mulian jing necessitates finer-grained scrutiny of internal evidence—particularly lexical and stylistic features (the present study offers only preliminary linguistic analysis). Nevertheless, our systematic bibliographic critique has successfully dislodged the spurious ‘translated by Zhi Qian’ label historically imposed upon it, while elucidating the precise historical mechanisms enabling this misattribution. This achievement removes a critical obstacle and establishes a firm foundation for subsequent rigorous interdisciplinary research on this scripture.