Text: T0152; Liu du ji 六度集; 六度集經


Identifier T0152 [T]
Title 六度集經 [T]
Date after 247 [Zacchetti 2013]
Translator 譯 Kang Senghui, 康僧會 [T]

There may be translations for this text listed in the Bibliography of Translations from the Chinese Buddhist Canon into Western Languages. If translations are listed, this link will take you directly to them. However, if no translations are listed, the link will lead only to the head of the page.

There are resources for the study of this text in the SAT Daizōkyō Text Dabatase (Saṃgaṇikīkṛtaṃ Taiśotripiṭakaṃ).


Preferred? Source Pertains to Argument Details


[T]  T = CBETA [Chinese Buddhist Electronic Text Association]. Taishō shinshū daizōkyō 大正新脩大藏經. Edited by Takakusu Junjirō 高楠順次郎 and Watanabe Kaigyoku 渡邊海旭. Tokyo: Taishō shinshū daizōkyō kankōkai/Daizō shuppan, 1924-1932. CBReader v 5.0, 2014.

Entry author: Michael Radich



[Shyu 2008]  Shyu, Ching-mei. “A Few Good Women: A Study of the Liu du ji jing (A Scripture on the Collection of the Six Perfections) from Literary, Artistic and Gender Perspectives.” PhD dissertation, Cornell University, 2008. — 64 n. 169

Shyu notes that the last four stories of T152 (nos. 88-91) were not originally included in the collection, according to CSZJJ.

Entry author: Michael Radich



[Shyu 2008]  Shyu, Ching-mei. “A Few Good Women: A Study of the Liu du ji jing (A Scripture on the Collection of the Six Perfections) from Literary, Artistic and Gender Perspectives.” PhD dissertation, Cornell University, 2008. — 69-71, 77

Shyu notes three sets of stories in T152 sporting formal features that are unusual, against the rest of the collection.

1) Twelve stories have unusual opening phrases. Most stories begin with something like "Once, the Bodhisattva was..." By contrast, these twelves stories begin with a version of the standard sutra opening formula, "Thus have I heard..." These stories are: 15, 16, 38, 39, 40, 41, 64, 83, 87, 88, 89, 91.

2) Thirty stories have their own individual titles: 11-14, 38-41, 51-54, 64-73, 94-91.

3) Eight stories have unusual ending formulae. Most stories end with a remark that relates the story to the particular perfection governing the section of the text in which the story is placed. These eight stories, however, end with phrases commonly used to conclude independent sutra texts, e.g. the audience heard the sutra with great joy, paid homage and departed. Shyu does not list these eight stories.

Shyu suggests that these "oddities" might indicate that "Kang Senghui not only translated stories of the Liu du ji jing but also collected and added stories that were in accord with the six perfections that were available from other independent Indian texts or stories" (71).

Shyu also notes (77 n. 201) that verse is very rare, and appears only in stories 37, 41, 48 and 64.

Entry author: Michael Radich



[Zacchetti 2013]  Zacchetti, Stefano. Storie delle sei perfezioni: Racconi scelti dal Liu du ji jing a cura di Stefano Zacchetti, con testo a fronte. Venice: Marsilio, 2013. — 26 n. 20

Zacchetti notes that Kang Senghui is supposed to have arrived in Jianye in 247, and translation of T152 should therefore have taken place sometime thereafter.

Entry author: Michael Radich



[CSZJJ]  Sengyou 僧祐. Chu sanzang ji ji (CSZJJ) 出三藏記集 T2145. — T2145 (LV) 97a12-14

Sengyou's biography of Kang Senghui in CSZJJ ascribes six titles to him:

1. 阿難念彌經, cf T152(88);
2. 鏡面王[經?], cf. T152(89);
3. 察微王[經?], cf. T152(90);
4. 梵皇王經, cf. T152(91);
5. Dao pin 道品[?];
6. Liu du ji 六度集 T152.


Entry author: Michael Radich



[Kamata 1982]  Kamata Shigeo 鎌田茂雄. Chūgoku bukkyō shi, dai ikkan: Shodenki no bukkyō 中国仏教史 第一巻 初伝期末の仏教. Tokyo Daigaku Shuppankai, 1982. — 220-222

Kamata states that the biography of Kang Senghui 康僧會 in CSZJJ ascribes to Kang Senghui titles such as the Anannianmi jing 阿難念彌經 (? cf. Alinianmi jing 阿離念彌經 T152(88)), Jingmian wang jing 鏡面王經 (cf. T152(89), Caiwei wang jing 察微王經 (T152(90)), Fanhuang wang jing 梵皇王經 (cf. T152(91)), Dao pin 道品(?), and Liu du ji jing 六度集經 T152. CSZJJ states in addition that Kang Senghui wrote commentaries on the Anban shouyi jing 安般守意經 T602, the Fa jing jing 法鏡鏡 (Ugraparipṛcchā T322), and the Dao shu jing 道樹經 (“Bodhi Tree Sūtra”) (又注安般守意法鏡道樹三經, T2145 [LV] 97a15); and prefaces to other scriptures. Kamata maintains that the ascriptions given to 康僧會 in CSZJJ are dubious, and most of those six entries are later additions (220).

Dao’an, as cited in CSZJJ, ascribes only two titles to Kang Senghui, Liu du ji jing T152 and a Wu pin jing 呉品經. Kamata thinks these ascriptions are correct. Kamata claims that the so-called Anannianianmi jing 阿難念彌經, Jingmian wang jing 鏡面王經, Caiwei wang jing 察微王經, and Fanhuang wang jing 梵皇王經 ascribed to Kang Senghui in CSZJJ are actually offshoot texts excerpted from T152 and separately circulated 別行 (220-221).

Kamata states that little is known about the Wu pin 呉品 ascribed to Kang Senghui by Dao’an. CSZJJ states that it had ten chapters, but was already missing at the time of Sengyou. Kamata suspects that this Wu pin is the so-called “Smaller sūtra” 小品 that GSZ lists as a work of Kang Senghui, along with T152 and the Za piyu 雜譬喩 (cf. T206, still ascribed to Kang Senghui in T). [Apparently Kamata does not think the 雜譬喩 should be ascribed to Kang Senghui although it is listed in GSZ --- AI]. LDSBJ states that the Wu pin is a smaller Prajñāpāramitā 小品般若經, but Kamata claims that there is no evidence for that interpretation. Kamata also mentions that some scholars (such as Sakaino 1972, 237-238) suggest that the Wu pin might have been a certain part of the Aṣṭasāhasrikā prajñāpāramitā 大明度經 T225 ascribed to Zhi Qian (221).

On Kang’s commentaries, Kamata notes that the preface 自序 to the Anbao shouyi jing 安般守意經 T602 [T602 (XV) 163a6-c8] indicates that Chen Hui 陳慧 wrote a commentary on the text, helped by Kang Senghui. Kamata states it is not known who translated the “Bodhi Tree Sūtra” 道樹經 for which Kang Senghui wrote a commentary. There is a 菩薩道樹經 in Dao’an’s list of anonymous scriptures (T2145 [LV] 16c19), and the *Siṃhamati-sūtra 私呵昧經 T532 ascribed to Zhi Qian was also called 菩薩道樹經 (T2145 [LV] 6c23). Kamata states (without presenting any clear reasons) that the scripture Kang Senghui wrote the preface for might have been T532 (222).

Entry author: Atsushi Iseki



[Karashima 2013b]  Karashima Seishi 辛島静志. “Was the Aṣṭasāhasrikā Prajñāpāramitā Compiled in Gandhāra in Gāndhārī?” ARIRIAB 16 (2013): 171-188.

T152(81), which bears no title, is a version of the story of the bodhisatva Sadāprarudita 常悲菩薩, which is also found at the end of the Aṣṭasāhasrikā prajñāpāramitā.

Entry author: Michael Radich



[Chen Hong 2003]  Chen Hong 陈洪. “Liu du ji jing wenben de xingzhi yu xingtai” 《六度集经》文本的性质与形态. Xuzhou shifan daxue xuebao 徐州师範大学学报 29, no. 4 (2003): 11-17. — 12

Chen states that T152(10) is derived from Zhi Qian's 桀貪王經 T198(1).

Entry author: Michael Radich



[Chen Hong 2003]  Chen Hong 陈洪. “Liu du ji jing wenben de xingzhi yu xingtai” 《六度集经》文本的性质与形态. Xuzhou shifan daxue xuebao 徐州师範大学学报 29, no. 4 (2003): 11-17.

Chen studies the formation, history and characteristics of the Liu du ji jing T152. (To distinguish between the extant T152 and the broader range of versions of the collection, whose history Chen aims to reconstruct, we refer to the latter below as LDJJ; by contrast, “T152” refers to the extant text.) Chen characterises LDJJ as a whole as an “edited translation” 編譯 or “compilation-translation” 撰譯.

Chen notes that all external evidence ascribes LDJJ to Kang Senghui, and treats it as a “translation” 譯. However, he notes that the meaning of yi 譯 (which we usually treat as meaning “translate”) appears in fact to be rather broad, so that its precise connotation in any given case must be investigated more closely.

Chen suggests that there are in fact three different types of sources for LDJJ as a whole: Indic sūtras; other non-Buddhist Indic classics; and earlier translations (from the later Han). As an example of texts dervied from Indic sūtras, he gives T152(87), which is paralleled in MĀ 67 [and a number of other sources like the Nimi jātaka and the Makhādeva jātaka --- MR]. Further examples are T152(21) and T152(36), which Chen sees as having been “taken from” the Jātaka. Examples of texts sourced in other Indic classics are T152(43) and (46), which are paralleled in the Rāmāyana. As examples of texts from earlier Chinese translations, Chen cites T152(89) and T152(10), which he sees as deriving from Zhi Qian’s T198.

Chen divides the history of LDJJ into three phases: (1) the production of the original text; (2) a “revised version” 改編本 which he believes circulated under the Southern Dynasties; (3) a “recompilation” 新編本 produced under the Sui-Tang. He bases his argument on evidence in the historical catalogues, and excerpts from the text in later collectanea such as the Jing lü yi xiang (JLYX) and the Fa yuan zhu lin (FYZL), including notes in those collections stating the source(s) of excerpted text, and its location (by juan number) in the source. Chen argues in reverse chronological order, but we here summarise in chronological order.

(1) Kang Senghui’s original collection
Chen follows and supplements Shi Tianchang (1998) in trying to determine the shape and content of Kang Senghui’s original LDJJ. According to Chen’s summary, Ven. Tianchang had already argued that nos. 38, 39, 64, and 79-83 were not in Kang Senghui’s original collection; no. 41 may have been taken by Kang Senghui from T198; nos. 74-76 are original Kang Senghui compositions, and possibly the same is true of nos. 77-78; and the collection probably underwent a round of revision after it left Kang Senghui’s hands.

Ven. Tianchang had also pointed out that the tradition reports chapters of LDJJ that are no longer included in the extant T152. Tianchang’s example was the 忠心政行經. Chen supplements this point with discussion of two other instances.

The 桀貪王經 [cf. T198(1)] is listed in CSZJJ as from LDJJ, but in LDSBJ, KYL and JLYX as from T198. Chen takes this as evidence that, like T152(89), this text was incorporated into LDJJ by Kang Senghui from T198. A very interesting item of information pertaining to this text is that the Shi shi liu tie 釋氏六帖 of 義楚 Yichu, writing under the Five Dynasties, still lists the text as coming from LDJJ; Chen speculates that this might be because Yichu had available to him a version of LDJJ that still included it, i.e., for Chen, a version stemming from the Liang.

The 鼈喻經 is presented in CSZJJ as stemming from LDJJ, but in Fajing as by Fasui 法邃, and in KYL and JLYX as from Dharmarakṣa’s Sheng jing T154; the content in JLYX matches T154(36). Chen argues that this indicates that the Dharmarakṣa tale was incorporated into LDJJ at the stage of the “Southern Dynasties revision” (discussed further below).

Chen also discusses the problem of nos. 88, 89, 90, 91, which has been known since Tang Yongtong: In CSZJJ, the juan 2 list of Kang Senghui’s works only gives two titles, but the biography, in juan 13, gives six titles—each of these works is listed as an independent text. The possibility that T152(89) is taken from T198 was already treated by Shi Tianchang (see above).

Chen suggests that the 阿離念彌經 T152(88) is “very likely” an E. Jin translation. It is recorded in two places in CSZJJ: in the aforementioned biography of Kang Senghui, and in Dao’an’s catalogue, where an interlinear note says that it is “from the Madhyamāgama”. (Chen also takes into consideration the report of LDSBJ that the text was translated by Tanwulan under the E. Jin.) Chen interprets the entry in the Dao’an catalogue as a hint that the text came from one of the Madhyamāgamas translated in the late fourth century, and was then incorporated into LDJJ. Chen also investigates the text of the tale itself. He also compares the extant text with the text as excerpted in JLYX. He notes that the text in T152 is simpler than that in JLYX, whereas the reverse is normally true. He suggests that this indicates that the extant T152 is not identical with the pre-Liang LDJJ.

Chen also notes that JLYX says, curiously, that its excerpt of T152(88) is from the Saṃyuktāgama (SĀ) . It is important to note here that the relevant interlinear note is subject to a variant reading. [Chen’s treatment of these variant readings is somewhat confusing: he merely says that the note is carried in the Qisha version, while the Korean and Taishō versions read 出阿難念彌經 (sic; according to CBETA, they in fact read 出阿難念經). According to CBETA, the actual readings in the T apparatus are: 出阿難念經 T(K), 出雜阿含經 SYMP (to which we can add Chen’s citation of the Qisha) --- MR.] FYZL also excerpts a matching text, and likewise says it is from SĀ. Chen thus proposes that the v.l. in the interlinear note in the Northern line (through K) represents a “correction” of the older note referring to SĀ, produced by identifying the source in T152(88).

The content of the text is relatively unusual—it tells of a tree with five kinds of fruit, each suited to a different class of being—and Chen indeed finds a text with approximately matching content (but not wording) not in SĀ but in MĀ 130, T26(130) (I) 619a8-15. Chen concludes that the compilers of JLYX and later texts were correct in identifying the text as from some Āgama, but erred in identifying the source as SĀ rather than MĀ; and that it is most likely that the text was added to T152 sometime after the Sui. He also speculates that the reason that the text as excerpted in JLYX and FYZL does not match the extant MĀ T26 in details of wording is that their text was excerpted from the lost, alternative translation of MĀ.

[Later in his paper, at the conclusion of his section discussing the form of Kang Senghui’s original LDJJ (17), Chen says that the collection might have originally contained as few as 60 tales.]

(2) Southern Dynasties “revised version” 改編本
An important starting point for Chen’s arguments about the later history of LDJJ is the evidence of CSZJJ. Chen notes that Sengyou lists the LDJJ alongside a "Wu pin" in 5 fascicles, and specifies that the latter is “missing”: 吳品五卷(凡有十品今闕), T2145 (LV) 7a26. From this, he deduces that Sengyou did indeed see the LDJJ that he lists. Chen also notes that Sengyou states that the LDJJ he lists was in 9 fascicles. However, from Fajing onwards, LDJJ is recorded as comprising 8 fascicles, as does the extant T152.

Chen holds that this is one hint, among others, that the version of LDJJ that circulated under the Southern Dynasties was different from the extant T152. He holds that a similar problem is glimpsed in the location of passages excerpted in JLYX, as indicated by fascicle number. Chen reports that these fascicle locations in fact mostly match the extant T152; however, the excerpt entitled 獨母見沙門神足願後生百兒 in JLYX, T2121 (LIII) 235a23-b26, corresponding to T152(23), is said to come from juan 2, but is presently found in juan 3.

Chen also notes that JLYX generally matches T152 in wording, apart from typical paraphrasing for stylistic purposes, but again with some exceptions.

In the JLYX excerpt of T152(9), some key numbers differ from the extant T152; T2121 (LIII) 47b25-48a22.

For one story, T152(15), JLYX T2121 (LIII) 140c8-141a18 gives two different sources for the first part of the story: 出慈仁法句譬喻經第二卷,又出大乘方便經上卷; then it gives LDJJ for the second part of the story: 出度無極集經第三卷. Chen determines that the wording of the former part indeed matches T211(7) 慈仁品. JLYX takes from this text an episode in which the protagonist king, Hemo 和默, engages in a blood sacrifice in the hope of curing an illness of his mother. This episode is not found in the T152 version of the story of the same king. Chen considers two possibilities—that the episode was added into the story in JLYX, or that it was present in the version of LDJJ that Baochang (the compiler of JLYX) was using. Chen holds that this episode was in the LDJJ available to Baochang, so that the JLYX excerpt serves as additional evidence that LDJJ was revised under the Southern Dynasties. [I cannot follow his reasoning at this point: he says that if the LDJJ Baochang saw did not have these two episodes together, there would have been no need to add the 大乘方便經 as a second source for the first passage, and that Baochang was therefore trying to show that there really was a basis for this excerpt. --- MR]

As further evidence for his hypothesis, Chen notes that JLYX normally states if an excerpt comes from LDJJ, but there are several cases where content clearly matches the extant T152, but Baochang does not give LDJJ as the source. He suggests that this is because these texts were not yet included in LDJJ (implying that they only found their way into the collection in the Sui-Tang version[s]; see below). However, he also adds [seemingly confusing the issue somewhat --- MR] that some such texts may have been circulating both in independent versions, and as part of LDJJ, and cites some cases where CSZJJ indicates that this was indeed true for some parts of LDJJ.

Chen also butresses his theory of a Southern Dynasties revision of LDJJ by pointing out a notice in CSZJJ stating that the 摩調王經 (cf. T152[87]) was translated in Tai’an 3 = 302-303 CE 太安三年正月十八日出; T2145 (LV) 8b24. [Chen does not note that this date is carried only in SYM --- MR.] Chen holds that this indicates that this tale, as found in T152, is in fact a later text, which was edited into LDJJ as part of the supposed Southern Dynasties revision.

Chen also cites in support of his theory a JLYX excerpt from a 摩[+日SYM]國經 , T2121 (LIII) 142b20-143a17. He shows that the text excerpted agrees with T152(49), which is about a king Nan 難 of a country called Motianluo 摩天羅. He then cites a report in LDSBJ that Tanwulan translated a text of this title between 373 and 396; T2034 (XLIX) 69c1. Chen uses this evidence to argue, as for the prior case, that the present T152(49) was in fact a text of the late fourth century, which was subsequently incorporated into LDJJ in the course of the supposed Southern Dynasties revision.

On the basis of this evidence taken in concert, Chen holds that a distinct version of LDJJ circulated under the Southern Dynasties, which was compiled sometime between the E. Jin and the Liang. Chen also argues that there are only approximately 38 tales that we can determine with confidence were present in this supposed version of the text (listed 15 n. 2; apparently, the texts excerpted in JLYX).

[Note: So far as I was able to determine, Chen never accounts for the fact that this supposed Southern Dynasties version of LDJJ should surely have been shorter than the extant T152, since he argues that certain tales were missing from it—and yet, on the other hand, Sengyou states in CSZJJ that the text he saw was 9 fascicles, longer (if anything) than the extant T152 --- MR.]

(3) Sui-Tang “recompilation” 新編本
Chen argues for the Tang recompilation of LDJJ on the basis of evidence in catalogues, lexicons (like the Yiqiejing yin yi, YQJYY) and text-critical remarks. Fajing reports locations of various texts that circulated as individual “excerpted sūtras” 抄經, and the fascicle locations correspond to the extant T152. Chen thus holds that the sequence and arrangement of the 91-story T152 (the extant collection) basically goes back to the Sui. However, he also cites minor differences in the information given by Fajing to support the idea that the text was later recompiled. Chen also examines the location (again by fascicle number) of words glossed in YQJYY, and of passages excerpted in FYZL, stating they they also match our extant T152 exactly.

More substantially, Chen cites remarks in the treatment of LDJJ by Zhisheng in KYL, which Chen holds indicate that Zhisheng undertook text-critical work to restore parts of the collection. These remarks pertain to the Pusa wei lu wang jing 菩薩為鹿王經, the Ma wang jing 馬王經, the Jingmian wang jing 鏡面王經 and the Mingduwuji jing 明度無極經. In each case, an interlinear note states that the text was added back into the compilation(?) 新編上; a summary comment by Zhisheng states further that five texts—the Hemo wang jing 和默王經, the Xiang wang jing 象王經, the Pusa wei lu wang jing, the Ma wang jing, and the Jingmian wang jing (two not affected by the aforementioned interlinear notes) were “recorded in the various catalogues, with titles the same as these, and [I have] therefore added them back [into the collection]”(?) 此之五經雖載群錄名與此同並新編上; T2154 (LV) 653c4-654a9.

Entry author: Michael Radich



[Shi Tianchang 1998]  Shi Tianchang 釋天常. "Liu di ji yanjiu" 六度集研究. Chung-Hwa Buddhist Studies 中華佛學研究 2 (1998): 75-104.

Shi Tianchang studies the history of Kang Senghui's collection of jātaka tales, which he treats under the title Liu du ji 六度集. This article is a summary of a more extended MA thesis. The overall thrust of his study is to argue that a number of the texts included in the connection were probably not part of a presumptive underlying Indic Vorlage to the text; he holds that several may well have been added by Kang Senghui himself at the moment of composition, and that some others may have been added later. Thus, for Tianchang, the present T152 is a synthetic product that emerged over the course of a complex history.

The collection is generally heterogeneous, in a manner that may lead us to suspect a chequered history. For example, the collection seems to have a standard opening and closing formula for its tales (昔者菩薩/昔者 ... 菩薩慈惠度無極行布施如是 [e.g. for dānapāramitā, varied for other "perfections"]). However, these formulae do not always appear. Twelve texts have rather a standard sūtra nidāna; eight have standard sūtra endings [some of these phrases have peculiar patterns of distribution that may reward closer investigation, e.g. 為佛作禮而去, 諸沙門聞經 – MR]. Only thirty texts bear individual titles. Although these titles often have –jing 經 in them, there is no necessary connection between such titles and a recognisable "sūtra" format. Later catalogues record approx. 24 titles as separate, individual texts, all in one juan, and usually stating that they are anonymous works. In such cases, we must consider whether the texts originated as offshoots of T152 (e.g. as chao 抄), or whether originally independent texts were later added to the collection that became T152.

The bulk of the study comprises treatments of individual texts within the Liu du ji or T152, which are of interest from various perspectives.

The Zhong xin zheng xing jing 忠心正行經 is reported in various traditional catalogues as taken from the Liu du ji, even though it is not found in the present T152. Tianchang ultimately argues that it is not possible to determine whether or not this text was part of the original Liu du ji, because we have no evidence about its content. See separate entry for details.

Tianchang argues that the Sixing jing 四姓經 T152(16) appears originally to have formed a single whole with the following, untitled T152(17). T152(17) ends with a sūtra ending formula, and an identification of 四姓 with the Buddha. The protagonist 四姓 (which Tianchang argues at some length, 81 n. 30, designates a social role rather than a proper name, *velama) features in "both" texts, and the two together thus form a single coherent narrative. Only one text appears in catalogues that might be associated with this text (in relation to velama; there are no records that can be connected with 四姓). Tianchang lists a rather rich set of parallels in Pāli and Chinese for the story of the Brahmin velama, 81 n. 31. The text seems anomalous, in that it is included in the section of the text on the perfection of giving, but ultimately does not make that perfection supreme. It also has an irregular opening and setting, and in other formulaic features does not fit with the typical features of the Liu du ji. The text features some promotion of the virtue of filial piety, which Tianchang holds is in keeping with concerns displayed by Kang Senghui elsewhere. Tianchang speculates, on this basis, that the text (presumably meaning T152(16-17) as a unit) was not in the Indic source text(s) for the Liu du ji, but was added by Kang Senghui himself.

The Taizi mupo jing 太子墓魄經 T152(38) is the subject of conflicting reports in the catalogues. The text has the form of a full sūtra, complete with an opening nidāna and stereotypical closing section. Tianchang holds that this format does not fit with the basic design and character of the Liu du ji, and that the texts was therefore probably not included in Kang Senghui's Indic source text. In the Jing lü yi xiang and CSZJJ, no mention is made of the text being from the Liu du ji. Tianchang thus proposes that it may well have been added to the collection sometime between Sengyou and Fajing.

The title Milan jing 彌蘭經 (cf. T152(39)) is already found in Dao'an, who characterises the text as old; in CSZJJ, it is already explicitly treated as from the Liu du ji. It is excerpted in the Jing lü yi xiang. At least from the time of CSZJJ, this title is identified with the Milian jing 彌連經/彌蓮經. However, Tianchang compares the Milian jing as quoted in the Jing lü yi xiang with T152(39), and argues that the differences in wording are too great for them to represent the same text; they are, rather, two independent versions of the same material. Tianchang refers to a study by Itō Chikako of eight versions of the same narrative material; according to his summary, she treats material common to all versions, and concludes that T152(39) represents the version closest to the oldest form of the story accessible. Tianchang concludes that there are certain elements of the style that suggest the text is close to Kang Senghui, or even his work; it is at least quite old. At the same time, he holds, again, that the independent sūtra format makes it difficult to see how it could fit into the overall plan of the Liu du ji, so long as the collection is united by any coherent design whatsoever. He thus concludes that it is most likely that the text did not belong to the Indic original behind the Liu du ji, and that it should have been added into the collection before Sengyou at the latest.

The Puming wang jing 普明王經 T152(41) relates a past life of Aṅgulimāla. Dao'an lists the title among anonymous texts. Neither Dao'an nor Sengyou given any further information. Fajing treats the text as a chao from the Liu du ji. From LDSBJ, the text is ascribed to Juqu Jingsheng. This ascription was overturned by Zhisheng in KYL. The text sports the very unusual feature of four-character rhyming verse. Virtually identical verses are found in the Wunao zhiman pin 無惱指鬘品 T202(52) 426b21-c2, and the Sūtra of Humane Kings T245 (VIII) 830b5-15. In other locations paralleling the narrative content of this text, however, the verses are different in purport: Tianchang cites Pāli Jātaka 537, T205(8) 504a18-19, and MPPU T1509 (XXV) 89a27-b1. Focusing in particular on the verbatim correspondence with T202(52), Tianchang argues that these verses do not display a typical Liangzhou style, and that the direction of borrowing should therefore be T152(41) to T202(52). The Sūtra of Humane Kings is of course a famous example of a Chinese composition, and Tianchang cites Mochizuki in support of the notion that there, too, the verses are taken from T152(41). T152(41) is excerpted in the Jing lü yi xiang, but without the verse; it is only in SYM that the source is given as the Liu du ji. Tianchang judges that the style generally matches Kang Senghui and the Liu du ji as a whole. He surmises, in conclusion, that it was probably part of the (Indic) original text, but that the sūtra opening may have been added to the text later.

The Mifeng wang jing 蜜蜂王經 T152(64) is listed among Dao'an's anonymous texts, and treated by Sengyou as taken from the Liu du ji. Fajing and bibliographers after him treat the text as a chao from the Liu du ji. This title does not appear in LDSBJ. Tianchang holds that the theme of the text fits the section in which it appears, on vīryapāramitā. However, the text takes the form of an independent sūtra, and the content centres on monks exchanging verses, which is atypical for the Liu du ji. Tianchang lists quite a number of relatively rare terms and items of phrasing he also holds atypical for Kang Senghui and the Liu du ji. He notes that many of these terms can be found in Dharmarakṣa. Tianchang's terms are: 阿惟越致, 不起法忍, 陀鄰尼 [sic: actually 陀隣尼], 不退轉地, 陰蓋, 般泥曰, 解諸法本, 善權方便, 當造光明智慧之本 [a peculiar choice—a hapax! – MR], 無央數人, 發無上平等度意. He focuses particularly on 般泥曰 and 發無上平等度意, which he says are particularly rare. He proposes that the only translator in whose works both these items are found is Dharmarakṣa: 般泥曰in several texts; 發無上平等度意 only in T334 [no single Dharmarakṣa work features both items – MR]. On this basis, Tianchang proposes that T152(64) is actually by Dharmarakṣa. [Note: Tianchang's methodology here is weak, since he only asks whether the terms in question are compatible with Dharmarakṣa, not whether they are exclusive to or distinctive of him; he also only considers the occurrence of each of his terms in isolation, without considering where they might cluster with particular density -- MR.]

Tianchang also treats the Jingmian wang jing 鏡面王經 T152(89). He first surveys evidence in the catalogues. He then notes the occurence of some terms in the text that he regards as more typical of Zhi Qian (e.g. 我曹, 子曹), and on this basis concludes that the text, which is also found at T198(5), was borrowed into T152 from T198, rather than the other way around. In this proposal, he conforms with the views of a number of other scholars.

In treating the texts found in the chapter on dhyānapāramitā, Tianchang notes, like prior scholars (e.g. Link, Zürcher) that T152(74) is actually the preface to the chapter, not a tale like those comprising the typical content of the Liu du ji. However, he demurs from predecessors like Link and Zürcher in holding that T152(75) and T152(76) and not part of this same preface. Tianchang argues that T152(74) and T152(75) share a common topic, namely, the control required for entry into dhyāna and the description of states ensuing upon its attainment. He aims to demonstrate this claim by listing a series of parallels in phrasing between the two. He holds that it would not make sense for such content to be repeated at such length in a single document like a preface, and therefore, that the common passages represent the author of the preface anticipating and echoing the content of the first proper text in the chapter. He then argues that T152(76), by contrast, is closely related to T152(77) and T152(78). No. 76 describes 26 thoughts that conduce to entry into dhyāna. Nos. 77 and 78 rehearse the narrative of the Prince Siddhartha making his excursions from the palace and seeing the "sights" of age, sickness, and death—whereupon he is depicted as returning to the palace and practicing dhyāna. In light of this grouping, Tianchang believes that No. 76 belongs with 77 and 78, and it is therefore inappropriate to treat it as part of the Introduction. By this process of elimination, he concludes that only T152(74) represents Kang Senghui's preface to the chapter.

However, Tianchang also concludes that T152(75) and T152(76) do not fit the general content and structure of the Liu du ji. Tianchang believes that the general mode of presentation of meditation here matches the content of An Shigao's texts, and is compatible with Kang Senghui's authorship, since Senghui is said to have continued Shigao's practices. They feature general instructions on meditation, and do not feature the deeds of Śākyamuni when he was a bodhisattva, or present the six perfections as bodhisattva practice. For Tianchang, these texts should therefore not have been part of any Indic Vorlage for the Liu du ji.

T152(77) presents episodes from Śākyamuni's "present" life like his excursions from the palace, and the episode in which he watches the ploughing; T152(78) presents his birth, marriage, and disillusion with the world. These texts therefore appear more at home in a Buddha biography. However, they also differ from more usual Buddha biographies in some details: the episodes are presented out of chronological order, and when the prince sees the "sights", he reacts by going back to the palace and meditating. These details lead Tianchang to conclude that T152(77) and T152(78) could not have been in an Indic source text for the Liu du ji, and might be Kang Senghui's original works.

Continuing the Buddha-biography theme, T152(79) presents Śākyamuni as a bodhisattva (in his last lifetime) achieving the three superknowledges through his meditative practice, and T152(80) presents the episode in which he was oblivious to a thunderstorm because he was absorbed in meditation. Tianchang holds that these texts, too, do not fit with the usual structure of the Liu du ji, and should not have been part of its Indic original text.

Tianchang also lays out striking matches in wording in passages treating the same episodes in T152(78) and Zhi Qian's Taizi ruiying benqi jing 太子瑞應本起經 T185; and, similarly, between T152(79) and the Yichu pusa benqi jing 異出菩薩本起經 T188 ascribed to Daozhen (98-99). On this basis, he proposes that in composing these texts, Kang Senghui consulted earlier works.

T152(81) presents a version of the Sadāprarudita story, better known from its occurrence in some members of the family of Prajñāpāramitā sūtras. Tianchang notes that this version of the story deviates in some details from the Prajñāpāramitā versions of the same story. In particular, other versions of the story do not present Sadāprarudita as a prior life of Śākyamuni. Tianchang also opines that the story, which he interprets as promoting or praising the perfection of wisdom, sits oddly in a section of the Liu du ji ostensibly devoted to the perfection of dhyāna. He notes further that this is the only example in all of T152 where we encounter the idea of a Mahāyāna bodhisattva. For all these reasons, he holds that it is unlikely that the text was part of any Indic Vorlage for the Liu du ji. He claims further that there are echoes of Kang Senghui's ideas, as attested in other texts, in this version of the story, and that the style is highly consonant with that of Kang Senghui. On this basis, he argues that it is very probable that the story was added to the collection by Senghui himself.

Tianchang argues rather briefly (100-101) that T152(82) has features that fit better with the promotion of the perfection of wisdom than with the perfection of dhyāna, and proposes that the collection may have been revised, at some point after its original production, to reassign the story to the dhyāna chapter, rather than the prajñā chapter immediately following.

Entry author: Michael Radich


  • Title: Liu du ji 六度集


[Bie lu (DH mss)]  "Liu Song" Zhongjing bie lu 劉宋眾經別錄, S.2872, P.3747. Dating complex and unclear.

In the "Liu Song" Zhongjing bie lu 劉宋眾經別錄, as represented by a Dunhuang manuscript fragment, P.3747, the following title appears: 六度無極經 (title #66 in the numbering given to the Bie lu manuscript in the transcription of Tan 1991). This title obviously refers to T152. The title is followed by an interlinear note:


In CSZJJ, wording almost identical to this is used in an interlinear note for the same text, and another text called the Wu pin 吳品: 魏明帝時。天竺沙門康僧會。以吳主孫權孫亮世所譯出, T2145 (LV) 7a27-b1. The verbatim correspondence of wording between these two sources raises interesting but difficult questions about the chronological priority between the Bie lu and CSZJJ. A further difficult question is whether one of the two directly borrowed from the other, or whether they drew on a common third source. Consideration of these questions must take into consideration the fact that the Bie lu, as witnessed in two Dunhuang fragments, contains a number of notes displaying such correspondences to the wording of CSZJJ.

Entry author: Michael Radich